Mol Neurobiol (2008) 38:231–241

DOI 10.1007/s12035-008-8043-y

Targeted Activation of Astrocytes: A Potential

Neuroprotective Strategy
Carole Escartin & Gilles Bonvento

Received: 20 June 2008 / Accepted: 26 September 2008 / Published online: 18 November 2008
# Humana Press Inc. 2008

Abstract Astrocytes are involved in many key physiolog- GFAP Glial fibrillary acidic protein
ical processes in the brain, including glutamatergic trans- FGF Fibroblast growth factor
mission, energy metabolism, and blood flow control. They KO Knockout
become reactive in response to pathological situations, a HD Huntington’s disease
response that involves well-described morphological alter- IGF-1 Insulin-like growth factor 1
ations and less characterized functional changes. The IL Interleukin
functional consequences of astrocyte reactivity seem to iNOS Inducible nitric oxide synthase
depend on the molecular pathway involved and may result LPS Lipopolysaccharide
in the enhancement of several neuroprotective and neuro- MCAO Middle cerebral artery occlusion
trophic functions. We propose that a selective and con- MRI Magnetic resonance imaging
trolled activation of astrocytes may switch these highly NGF Nerve growth factor
pleiotropic cells into therapeutic agents to promote neuron NFκB Nuclear factor-κ B
survival and recovery. This may represent a potent NMDA N-Methyl-D-aspartate
therapeutic strategy for many brain diseases in which PET Positron emission tomography
neurons would benefit from an increased support from STAT3 Signal transducer and activator of transcription 3
activated astrocytes. SOCS3 Suppressor of cytokine signaling 3
SOD Superoxide dismutase
Keywords Reactive astrocytes . Activation of astrocytes . SCI Spinal cord injury
Brain diseases . Therapeutic target . Cytokines TGFβ Transforming growth factor β
TNFα Tumor necrosis factor α
Abbreviations
AD Alzheimer’s disease
ALS Amyotrophic lateral sclerosis
BrdU Bromodeoxyuridine Astrocytes and Brain Diseases
CNTF Ciliary neurotrophic factor
COX2 Cyclooxygenase 2 Over the last 20 years, a new field of research has emerged
focusing on astrocytes, the most abundant glial cell type in
the brain. Extensive study has led to many surprising
C. Escartin (*) : G. Bonvento
discoveries regarding the role of these cells and their
CEA, IB2M, MIRCen, CNRS URA2210,
4, place du General Leclerc, importance for brain function. Some of the functions of
91401 Orsay, France astrocytes were first proposed by Ramon y Cajal, more than
e-mail: [email protected] a century ago, when he described the unique anatomical
properties of these cells [1]. Astrocytes—with their endfeet
C. Escartin
Department of Neurology, UCSF and VA Medical Center, extending toward blood vessels on one side and enclosing
San Francisco, CA, USA synapses on the other—are ideally situated to control
232 Mol Neurobiol (2008) 38:231–241

metabolic supply, blood flow, ionic homeostasis, and General Features of Reactive Astrocytes
neurotransmitter levels [2]. More recently, studies have
suggested that in addition to their supportive role toward Astrocyte reactivity is a general term encompassing
neurons, astrocytes play an active role in neuronal changes in both the morphology and function of astrocytes
transmission and synaptic plasticity [3]. in response to any pathological condition. The pathological
It is now thought that astrocytes are organized into stimulus may be an acute injury, such as mechanical lesions
nonoverlapping spatial domains [4–6]. In the rodent brain, a of brain parenchyma, brain trauma, ischemia, infection, or a
single astrocyte domain may encompass more than 100,000 chronic deleterious situation such as neurodegenerative
synapses that may be regulated in a coordinated fashion by diseases or aging [14–16]. Astrocyte reactivity is generally
gliotransmitters released from this cell [5, 6]. Another key associated with microglial reactivity and, in some cases,
anatomical feature of astrocytes is their interconnection by leukocyte recruitment. Microglia, the resident immune cells
gap junctions (two apposed hemichannels of connexins) of the brain, may in fact be responsible for the subsequent
that forms a syncytium facilitating transfer of information activation of astrocytes and may coordinate complex
and metabolites over distance [7]. Because of these peculiar processes of neuroinflammation [17]. This review will
anatomical properties, any alteration in astrocyte function is focus on astrocyte activation and will not specifically
susceptible to have profound consequences for neuronal address the potential dual effects of reactive microglia and
activity [3]. immune cell infiltration for neuronal survival and brain
Astrocytes have recently been shown to play a causal or recovery. Several comprehensive reviews have already been
exacerbating role in several brain diseases previously thought published on this interesting topic [18–20].
to be purely neuronal in origin, including status epilepticus, The reactive state of astrocytes was initially defined on
schizophrenia, amyotrophic lateral sclerosis (ALS), and the basis of morphological criteria. Reactive astrocytes have
Huntington’s disease (HD; for comprehensive reviews, see both hypertrophic processes and soma and overexpress
[3, 8, 9]). For example, mutant huntingtin, the protein intermediate filaments (glial fibrillary acidic protein
responsible for HD, is expressed in the astrocytes of patients (GFAP), vimentin, and nestin [21, 22]; see Fig. 1). This
and has been shown to reduce glutamate uptake by cultured definition was recently refined by studies using dye-filling
astrocytes [10]. The mutant form of superoxide dismutase 1 [23] or diolistic labeling [24] of reactive astrocytes and
(SOD1) produced in the astrocytes of some familial ALS three-dimensional reconstruction of confocal images. Re-
patients is toxic to primary and embryonic stem-cell-derived active astrocytes appear to have thicker main processes than
motor neurons [11]. Conversely, selective invalidation of the resting astrocytes, without any significant alteration in their
gene encoding the mutant SOD1 in astrocytes slows disease overall domain volume [23]. Their processes also tend to
progression [12]. It has also been suggested that the amyloid overlap more at the domain boundaries after epileptic
deposits observed in Alzheimer’s disease (AD) may result seizures, but not during chronic neurodegenerative diseases
from the dysfunctional degradation of amyloid fibrils by [24].
astrocytes [13]. Therefore, astrocyte dysfunction may play an After a local disruption of the brain parenchyma, reactive
unsuspected role in the occurrence and exacerbation of many astrocytes extend their processes in a single plane forming a
diseases. These cells may therefore be a promising new glial scar that demarcates the lesioned area from healthy
therapeutic target for many of these pathological conditions, parenchyma. This type of astrocyte reactivity, called
particularly given the unique ability of astrocytes to become anisomorphic gliosis, is irreversible [16]. By contrast, with
reactive, which may be accompanied by marked functional less focal injuries such as chemical lesions or during
alterations of potential benefit to damaged neurons. chronic diseases, reactive astrocytes are more evenly
This review will address the following issues: What are distributed and have randomly orientated processes, a
the main morphological and functional features of reactive reversible phenomenon called isomorphic gliosis. Thus,
astrocytes? Are these features unique and universal or even if reactive astrocytes display common features
dependent on the stimulus involved? Do reactive astrocytes (hypertrophy and upregulation of intermediate filaments),
promote, exacerbate, or combat pathogenic processes? other characteristics may depend on the stimulus involved
Could novel therapeutic strategies for brain diseases be (see below).
based on a targeted activation of astrocytes? The classic definition of astrocyte reactivity involves
In the following paragraphs, we will use the term cellular proliferation; however, the intensity of astrocyte
“reactive astrocytes” or “astrogliosis” as a state of astro- proliferation appears quite variable depending on the model
cytes that is induced by any pathological situation by means studied and the techniques used. Early reports using
of endogenous mechanisms, while “activated astrocytes” bromodeoxyuridine (BrdU) labeling or radioactive thymi-
will refer to a state that is induced experimentally by dine incorporation suggested that limited astrocyte prolif-
pharmacological or genetic manipulations. eration occurred after brain injury (less than 5% of total
Mol Neurobiol (2008) 38:231–241 233

Fig. 1 CNTF-activated astro-

cytes in the mouse hippocam-
pus. Mice were injected into the
hippocampus with a lentiviral
vector containing the cDNA for
the cytokine CNTF (CNTF) or
the cDNA for β-galactosidase
(Control). Two months later,
astrocytes were stained for
GFAP. Activated astrocytes dis-
play classical hypertrophied
GFAP-positive processes. Scale
bar=20 μm (a, c) and 10 μm (b,
d). Unpublished data from G.
Bonvento and C.W. Shuttle-
worth

reactive astrocytes) and that the apparent increase in the early genes such as c-fos and heat shock protein genes [15].
number of astrocytes in pathological conditions might be They then overexpress intermediate filaments and undergo
linked to the production of larger amounts of GFAP (see morphological changes within a few hours [16]. In the case
references in [16]). However, a recent study based on of isomorphic gliosis, the overexpression of some of these
genetic mapping with inducible Cre-mediated recombina- proteins is transient and disappears after several weeks (see
tion in reactive astrocytes and lentivirus-tagging suggested references in [16]). The time course of astrocyte reactivity is
that adult astrocytes are a significant source of proliferating, quite complex and depends on the type of injury; the
reactive astrocytes in response to stab wound injury in mice sequence of events being even less clear in chronic
[25]. They found that up to 40% of tagged adult astrocytes pathological states such as neurodegenerative diseases [26].
incorporated BrdU after injury. The use of this elegant The reader is referred to several comprehensive reviews that
technique in other models of astrogliosis should provide a describe the molecular profile and time course of astrocyte
more general assessment of the degree of proliferation. reactivity in different pathological conditions such as
More functional markers of astrocyte reactivity include ischemia [27, 28], traumatic brain injury [29], mechanical
an increase in the levels of certain cytokines, such as lesions [30], or neurodegenerative diseases [18, 31].
interleukin-6 (IL-6) and ciliary neurotrophic factor (CNTF), In conclusion, while the morphological alterations of
adhesion/recognition molecules, as well as other proteins reactive astrocytes are universal and quite easily detected,
generally associated with detrimental effects such as functional changes appear much more subtle and depend on
calcium-binding protein S100β, inducible NO synthase the stimulus involved. This complex picture has contributed
(iNOS), and cyclooxygenase 2 (COX2; reviewed in [14, to the difficulty in evaluating the functional role of reactive
15]). Following acute injuries, astrocytes express immediate astrocytes in pathological conditions.
234 Mol Neurobiol (2008) 38:231–241

Reactive Astrocytes: Deleterious or Protective? KO mice 2 weeks after transection of the entorhinal cortex
axons [33]. Kinouchi et al. showed that astrocytes from
Astrocytes are involved in many important brain functions. these double-KO mice were more permissive for stem
It is therefore of key importance to evaluate whether these cell migration and neurite extension after retinal trans-
functions are enhanced, lost, or unchanged when astrocytes plantation [34]. However, at earlier time points after
are in a reactive state. This dictates whether manipulation of entorhinal axon transection (4 days), double-KO mice
astrocyte reactivity is a viable therapeutic strategy for displayed a lower number of synapses [33]. Indeed, a more
improving neuronal survival and restoring neuronal func- recent study on these mice has demonstrated that reactive
tion. Two main approaches have been used to answer this astrocytes may rather be beneficial for neuronal survival. Li
question: prevention of endogenous astrocyte reactivity and et al. found that exposing GFAP/vimentin double-KO mice
activation of astrocytes by pharmacological or genetic to middle cerebral artery occlusion (MCAO) resulted in an
manipulation. infarct size three times larger than in littermate controls or
single-KO mice [36]. This increase in infarct size was
Assessing the Role of Endogenous Reactive Astrocytes accompanied by a decrease in uptake of the excitotoxic
neurotransmitter glutamate and levels of plasminogen
One approach to interfere with astrocyte reactivity (or at activator inhibitor 1, a key protective protein in ischemia
least to disrupt the glial scar) involves the knockout (KO) [36].
of intermediate filament genes (encoding GFAP and/or Other studies have used elaborate genetic tools to more
vimentin; Table 1). Knockout mice showed no develop- directly address the role of reactive astrocytes. Sofroniew et al.
mental or breeding abnormalities (see references in [21]). developed a transgenic mouse expressing the herpes simplex
However, GFAP and vimentin double-KO mice displayed virus type 1-thymidine kinase “suicide gene” under control of
lower levels of astroglial reactivity (as seen with S100β and the mouse GFAP promoter (Table 1). The enzyme encoded
nestin immunolabeling) and greater sprouting of supra- by this gene phosphorylates gangiclovir, preventing further
spinal axons compared with control mice, 3 days, 1 and elongation of the newly synthesized DNA and triggering the
5 weeks after spinal cord hemisection. These features were death of dividing cells. This results in the specific ablation of
associated with locomotor functional recovery 2 to 4 weeks reactive astrocytes that have strong GFAP promoter activity
following injury [32]. A similar improvement in axonal and are undergoing proliferation. Sofroniew et al. used this
regeneration was observed in the hippocampus of double- model to investigate the role of reactive, dividing astrocytes

Table 1 Main genetic approaches to investigate the role of reactive astrocytes in vivo

Genetic construct Effect Model of Main outcomes Reference

astrogliosis Injury

GFAP-vimentin KO Disruption of cytoskeleton Spinal cord Increased axonal sprouting functional recovery − [32]
in reactive astrocytes hemisection
Entorhinal cortex Improved axonal regeneration at 2 weeks +/− [33]
axon transection Decrease in synapse number at 4 days
Cell transplant in Increased stem cell migration and neurite − [34]
the retina extension
Retinal detachment Improved photoreceptor survival − [35]
MCAO Increase in infarct size + [36]
PromGFAP-TK Ablation of dividing Forebrain stab Increase in neurite outgrowth lesion +/− [37]
(+oral ganciclovir) reactive astrocytes wound exacerbated
Spinal cord injury Lesion and motor symptoms exacerbated + [38]
Traumatic cortical Lesion exacerbated when mild injury, no effect + [39]
injury on severe injury
PromNestin-Cre X Inhibition of astrocyte Spinal cord injury Lesion aggravated and more severe symptoms + [40]
STAT3loxP reactivity
PromNestin-Cre X Enhancement of astrocyte Spinal cord injury Reduced lesion and improved motor + [40]
SOCS3loxP reactivity symptoms

This table reports only the main studies on the functional role of reactive astrocytes in vivo that are based on genetic targeting of astrocytes
through the use of cell type-specific promoters. Many other studies with a nonselective KO or overexpression (of cytokines for example) have
been carried out but are not mentioned here. See [41] for review on the outcomes of such experiments. Signs “+”, “−”, and “+/−” refer,
respectively, to beneficial, detrimental, or mixed effects of reactive astrocytes on the injury outcome
Prom promoter
Mol Neurobiol (2008) 38:231–241 235

in various acute pathological conditions, including forebrain pathological processes and assessing their functional prop-
stab wound [37], traumatic cortical brain injury [39], and erties. This approach is based on the overexpression of known
stab or crush spinal cord injury (SCI) [38]. The absence of inducers of astrocyte reactivity (see “Strategies to Promote a
dividing reactive astrocytes limited glial scar formation, Neuroprotective Phenotype in Activated Astrocytes”), and
allowing greater neurite outgrowth in the forebrain stab the goal is to specifically target astrocytes, leaving other
wound model [37]. However, the ablation of reactive cell types as unaffected as possible. We found that
astrocytes greatly increased the severity of all these lesions: astrocytes could be activated selectively in vivo through
Blood brain barrier repair was impaired, immune cell lentiviral gene transfer of the cytokine CNTF or in vitro by
infiltration levels were higher, and neuronal death was exposing primary mixed cultures to recombinant CNTF
significantly enhanced. In the forebrain stab wound model, [43, 44]. CNTF synthesis and release by infected cells of
memantine (a N-methyl-D-aspartate (NMDA) glutamate the striatum (mainly neurons) induced robust and stable
receptor antagonist) partially prevented neuronal death in activation of astrocytes that displayed classic features of
the CA1 region of the hippocampus of reactive astrocyte- reactive astrocytes (hypertrophy, overexpression of inter-
ablated mice [37], highlighting the key role of reactive mediate filaments) (see Fig. 1) [43]. On the contrary, the
astrocytes in promoting glutamate homeostasis. In the SCI number of CD11b- and ED1-positive microglia/macro-
model, significant levels of oligodendrocyte death accompa- phages, the expression of several neuronal markers, and
nied by severe demyelination were also observed in reactive the spontaneous glutamatergic activity of striatal neurons
astrocyte-ablated mice, together with less favorable motor remained unaffected by CNTF [43]. CNTF-activated
outcomes [38]. Myer et al. used controlled cortical impacts astrocytes displayed significant alterations in two key brain
of different intensities to produce either moderate or severe functions: glutamate homeostasis and regulation of energy
cortical lesions leading to 18% and 88% loss of cortical metabolism. CNTF-activated astrocytes expressed the glu-
tissue, respectively, and found that the ablation of dividing tamate transporters excitatory amino acid transporter 1 and
reactive astrocytes affected only moderate injuries [39]. This 2 (EAAT1 and EAAT 2) with two previously undescribed
last observation suggests that the capacity of reactive posttranscriptional modifications (hyperglycosylation and
astrocytes to promote brain recovery is limited, and recruitment to functional raft domains at the membrane),
therefore, stimulating this endogenous response to brain and these features were associated with significant
insult may constitute an efficient therapeutic strategy. improvements in glutamate handling in vivo [43]. CNTF-
Okada and colleagues provided more direct evidence for activated astrocytes also expressed a new set of metabolic
beneficial roles of reactive astrocytes and for the aggravating enzymes and transporters involved in the ketone body
effects of preventing astrocyte reactivity in the SCI model pathway and fatty acid β-oxidation and displayed a higher
(Table 1). They used a conditional KO of signal transducer rate of oxidation of these alternative energy substrates to
and activator of transcription 3 (STAT3) or suppressor of glucose [44]. This metabolic plasticity increased the
cytokine signaling 3 (SOCS3) based on the Cre-mediated resistance of CNTF-activated astrocytes to metabolic
recombination of these genes in reactive astrocytes [40]. injuries (glycolysis inhibition and prolonged exposure to
STAT3 is part of the intracellular signaling pathway of many palmitate) and eventually improved neuronal survival in
cytokines (see “Strategies to Promote a Neuroprotective vitro [44]. A recent study of primary cultures of mouse
Phenotype in Activated Astrocytes”) and seems to be astrocytes also reported that astrocyte activation by cyto-
involved in astrocyte reactivity. SOCS3 is induced by STAT3 kines (IL-1β and/or tumor necrosis factor α (TNFα))
and prevents STAT3 activation, providing a negative altered the metabolic profile of these cells (enhanced
feedback on this pathway [42]. Okada et al. showed that glycolysis, pentose phosphate pathway and Krebs cycle
impairing astrocyte reactivity by invalidating STAT3 reduced activities and lower intracellular glycogen levels [45]).
the migration of reactive astrocytes toward the site of injury, These cytokines induced additive metabolic changes (i.e.,
increased demyelination and neuronal disruption, and wors- enhancement of glycolysis and decreased glycogen levels)
ened clinical outcomes. Invalidating SOCS3 improved all when used in combination [45]. However, the increase in
these outcomes, providing another proof-of-principle that the activity of the pentose phosphate pathway and the decreased
endogenous astrocyte response may be insufficient and that metabolic response to glutamate were only observed in
its stimulation may be beneficial [40]. presence of the two cytokines. This is another demonstration
that the signaling pathways responsible for astrocyte activa-
Characterizing the Function of Experimentally Activated tion determine the functional outcome.
Astrocytes More direct neuroprotective effects of activated astro-
cytes have also been reported. Using CNTF or IL-1β to
The role of reactive astrocytes can also be investigated by activate spinal cord astrocytes, Albrecht et al. showed
selectively activating astrocytes independently of any enhanced production of fibroblast growth factor 2 (FGF-2)
236 Mol Neurobiol (2008) 38:231–241

by CNTF-activated astrocytes, which was associated with Thus, reactive astrocytes and experimentally activated
greater survival and neurite outgrowth of cocultured motor astrocytes display marked functional changes, which may
neurons [46]. Similarly, FGF-1-activated astrocytes over- either induce direct neurotrophic effects (increased release
express antioxidant enzymes and release larger amounts of of neurotrophic factors, enhanced energetic supply, etc.) or
nerve growth factor (NGF) [47]. Damaged neurons express trigger more subtle, indirect changes leading to a general
the proapoptotic p75 form of the NGF receptor and are improvement in brain cell function (enhanced glutamate
eliminated by this release of NGF, whereas “healthy” uptake, reorganization of metabolic pathways, modulation
neurons may benefit from this enhanced trophic support of synaptic transmission etc., Fig. 2) [62, 63]. Even the glial
[47]. Activation of astrocytes by both IL-1β and TNFα in scar that hinders axonal regrowth has some beneficial
vitro also results in significant increases in release of consequences for the brain, demarcating the injured area
glutathione [45], which is then available to neurons for their and preventing propagation of damage. Astrocyte reactivity
own antioxidant defense [48]. appears then to be a potent endogenous defense mechanism

Fig. 2 Several beneficial functions may be enhanced in reactive or may modify synapse coverage and thus, neurotransmitter removal and
experimentally activated astrocytes. Activation of astrocytes by synapse modulation [3], Finally, changes in blood flow directly affect
pharmacological agents or induction of astrocyte reactivity in nutrient availability and metabolite supply. Numbers refer to key
pathological conditions may enhance the regular functions of reviews or articles reporting an improvement in astrocyte function (in
astrocytes. Many of them involve subtle interactions with neurons bold) following the experimental activation of astrocytes in vitro or in
and control their survival and correct function. Some of these vivo (glutamate removal [43, 50, 51]; metabolic support [44];
functional changes are interrelated. For example, changes in astrocyte antioxidant defense [45, 52–54]; trophic support [46, 55, 56]; K+/
connectivity may have various effects on metabolite supply, synaptic water homeostasis [57, 58]). For some more complex functions
transmission, and glutamate uptake, because metabolites such as (italics), the outcome of activation is still unclear (change in
glucose and neurotransmitters can pass through gap junctions [49]. connectivity [49, 59]; synaptic modulation [60]; blood flow regulation
Similarly, changes in the three-dimensional morphology of astrocytes [61]). The background image was supplied by Servier Medical Art
Mol Neurobiol (2008) 38:231–241 237

that could be manipulated to promote neuronal survival and more efficient to regulate astrocyte activation with a more
recovery [20]. selective target. The insulin-like growth factor (IGF-1)-
calcineurin pathway has recently been identified as a
potential pathway controlling the switch between the pro-
Strategies to Promote a Neuroprotective Phenotype and antineuroprotective properties of reactive astrocytes
in Activated Astrocytes [71]. In situ, calcineurin is expressed by reactive astrocytes
surrounding amyloid plaques in a transgenic mouse model
In order to be a valid therapeutic approach, the activation of of AD and in the hippocampus of aged mice [72]. The
astrocytes needs to be selective and controlled. Astrocyte overexpression of a constitutively active form of calci-
activation is associated with many different functional neurin through adenoviral gene transfer in hippocampal
outcomes that may have variable degrees of effectiveness neuron-astrocyte cultures activates astrocytes, which dis-
on neuronal survival. It is thus necessary to characterize play cellular hypertrophy and express several “classic”
precisely the molecular cascades resulting in these different genes of reactive astrocytes (S100β, vimentin, antioxidant
activation states. Cytokines are key inducers of astrogliosis enzymes…), as seen with microarray analysis [72]. Con-
(see above) [26, 64]; they activate various intracellular versely, in vitro calcineurin gene knockdown decreases the
pathways, including the p38/MAPK pathway [65], the production of iNOS and COX2 by reactive astrocytes after
Janus kinase/STAT pathway [40], and the nuclear factor-κ exposure to lipopolysaccharide (LPS) [71]. But intriguingly,
B (NFκB) pathway [66], all of which are potential targets the overexpression of a constitutive form of calcineurin in
for promoting tight control of activation state. Some astrocytes reduces their production of deleterious molecules
cytokines are typically described as detrimental, inducing such as COX2, iNOS, and proinflammatory cytokines after
excessive inflammatory processes and cell death in the exposure to TNFα or LPS in vitro and in response to
brain, while others, used at an appropriate concentration, penetrating cortical injury or LPS injection in vivo [71].
may have more physiological and even neuroprotective This suggests that calcineurin does not simply activate
effects [17, 67, 68]. astrocytes but rather modulates their phenotype. Indeed, the
One way to make activated astrocytes beneficial would authors made the interesting observation that in rat cultured
be to modulate their cytokine “repertoire” by favoring the astrocytes, a delayed activation of calcineurin by IGF-1, 3
production of beneficial cytokines. or 16 h after treatment with TNFα or LPS respectively,
Munoz et al. showed that a specific inhibitor of p38/ could counteract their deleterious effects on neuronal
MAPK decreases the production of two proinflammatory survival and astrocyte ROS production. Therefore, the
cytokines (IL-1β and TNFα) and improves behavioral authors propose that a primary activation of calcineurin by
outcomes following amyloid β injection into the mouse proinflammatory stimuli such as TNFα or LPS triggers
hippocampus [69]. However, the functional changes in- deleterious proinflammatory cascades and that a secondary
duced by this treatment remain unclear. Brambrilla and or sustained activation of astrocyte (by IGF-1 or by genetic
colleagues have developed a transgenic mouse overexpress- manipulation) modulates astrocyte reactivity, shifting them
ing a dominant form of the NFκB inhibitor IκB under the toward a more protective phenotype [71].
GFAP promoter [70]. These mice showed no obvious motor There are some successful reports of neuroprotection
or behavioral deficits, no gross anatomical alterations, or using cytokines known to activate astrocytes in relevant in
neuronal cell loss in the spinal cord [70]. The nearly vivo models of brain diseases. Lentiviral gene transfer of
complete abrogation of NFkB activation in astrocytes IL-6 in the rat striatum activates astrocytes (as seen with
following contusive SCI did not prevent the appearance of their increased GFAP expression after 2 weeks) and
GFAP-positive astrocytes after 8 weeks nor did it change protects neurons against excitotoxic lesions [73]. Intra-
the production of some proinflammatory cytokines (TNFα, cerebroventricular injections of recombinant IL-6 signifi-
RANTES) after 1 day. However, it decreased the injury- cantly decreased infarct size when injected 30 min before
induced production of other cytokines/chemokines (trans- and 15 min after MCAO [74]. In mouse, 1-h pre-injection
forming growth factor β2 (TGFβ2), MCP-1, and CXCL10) of recombinant TGFβ1 or 1-week pre-infection with an
while increasing the release of IL-6. This change in the adenovirus encoding for the TGFβ1 gene protects CA3
pattern of cytokine production was associated with a greater mouse hippocampal neurons against kainate toxicity [75].
sparing of white matter tracks and better functional In several primate models of HD, CNTF treatment has been
recovery 8 weeks after SCI [70]. This study confirms that shown to be neuroprotective [76, 77]. In particular, in the
the NFκB pathway is a potential target for modulating the progressive model of striatal degeneration induced by the
cytokine repertoire and astrocyte activation. However, mitochondrial toxin 3-nitropropionic acid, implantation of
NFκB is a key transcription factor controlling many other capsules of cells genetically modified to secrete CNTF at
prosurvival and neurotrophic genes [66], and it may be the onset of symptoms results in a significant protection of
238 Mol Neurobiol (2008) 38:231–241

striatal and cortical neurons and reversal of existing motor techniques allow to evaluate changes in activated astrocytes
and cognitive deficits [76]. Thus, even a delayed activation at the single cell level or to quantify astrocyte activation at
of astrocytes is effective, suggesting that the targeted the brain-structure level. Specific monitoring of activated
activation of astrocytes may be a valid therapeutic strategy astrocytes is particularly important in the context of a cell-
for ongoing diseases in which astrocytes may already be based therapeutic approach such as the targeted activation
reactive. of astrocytes. Using two-photon imaging of astrocytes
However, these experiments only provide a necessary labeled with sulforhodamine 101 in the mouse cortex
proof-of-principle that activation of astrocytes can be [82], Nimmerjahn et al. studied the response of astrocytes
associated with neuroprotective effects. The demonstration to “micro-stroke” and showed that these cells are much less
that cytokine-activated astrocytes really mediate these motile than microglia [83]. Similar anatomical analyses of
protective effects is still lacking. Transgenic mice, in which astrocyte activation can be performed with fluorescent
astrocyte reactivity can be prevented, such as the condi- reporter genes under the control of the GFAP promoter
tional STAT3 KO mice [40], may allow to test this (or that of another intermediate filament gene induced in
hypothesis. reactive astrocytes [84]). Fluorescent proteins have been
used to analyze the morphological features of normal [6,
85] or reactive astrocytes [86] in brain slices and in vivo.
New Molecular Tools to Target and Monitor Activated To obtain more global but quantitative information on
Astrocytes astrocyte activation, luciferase may be used as an alterna-
tive reporter gene and can be monitored and quantified by
Advances in molecular biology and gene transfer technol- biophotonic/bioluminescence imaging [87].
ogies have made it possible to target astrocytes more Positron emission tomography (PET) is another nonin-
specifically and thus both to address long-standing ques- vasive and quantitative imaging system, which can be used
tions regarding the role of reactive astrocytes and to use to study the whole brain simultaneously, as opposed to
these cells as therapeutic agents. analysis of superficial structures performed with two-
Conditional KO mice that express genes selectively in photon microscopy. PET ligands available for following
astrocytes or even in reactive astrocytes are now available. neuroinflammation bind mostly to reactive microglia, but
The expression of the Cre recombinase gene is placed may also label reactive astrocytes to some extent [88]. The
under the control of an astrocyte-specific promoter (such as development of more selective ligands for reactive astro-
the GLAST, connexin 30, aquaporin 4, or apolipoprotein E cytes is required to follow astrocyte activation specifically
promoters) [78] or a reactive astrocyte-specific promoter in rodents and primates, using PET in preclinical para-
(nestin) [40]. In addition, the activity of a modified form of digms. Magnetic resonance imaging (MRI) is an additional
the recombinase (CreERT2) can be controlled by tamoxifen standard brain imaging technique that can be used in
injections [79], facilitating the temporal control of the KO rodents, primates, and humans. This technique has been
process in reactive astrocytes. These are powerful models to recently used to follow astrocyte reactivity 2 h following
investigate the involvement of specific proteins in astrocyte two types of acute brain injury in rats. These studies
reactivity. demonstrated that both low-flow ischemia induced by
Viral vectors are useful alternative tools, and recently, endothelin-1 injection and NMDA-mediated excitotoxic
we developed lentiviral vectors, which result in high levels injury induced a detectable MRI signal (increase in T1
of transgene expression in astrocytes, while transgene relaxation) [89]. Both PET and MRI may therefore be
expression is repressed in neurons (Colin et al., submitted). invaluable tools in both preclinical and clinical studies to
These vectors will facilitate the targeted activation of provide noninvasive measurements of astrocyte activation
astrocytes without interfering with other cell types. Lenti- over time throughout the brain.
viral vectors are very versatile experimental tools and may
carry conditional sequences to control the pattern of
transgene expression over time. Further, they can be Conclusions
injected into spatially restricted structures in adult animals
from various species [80, 81]. They also represent a unique Reactive astrocytes were once thought to be detrimental
way of selectively delivering therapeutic molecules to brain cells responsible for neuronal demise, but recent studies
cells while inducing minimal peripheral side effects [80, have changed this view and they are now considered as
81]. potential endogenous repair agents [62, 63, 90, 91]. The
The development of potent imaging techniques, partic- intrinsic capacity of astrocytes to react to any type of brain
ularly for live imaging, has also opened new opportunities injury and to develop a broad range of defense mechanisms
to investigate the role of reactive astrocytes [3]. Such makes them a very interesting therapeutic target. This is
Mol Neurobiol (2008) 38:231–241 239

particularly true, because many neurological diseases share 16. Kalman M (2004) Glial reaction and reactive glia. In: Non
neuronal cells of the nervous system: function and dysfunction.
common pathological mechanisms such as oxidative stress,
Hertz L (ed) Elsevier, Amsterdam, pp 787–835
excitotoxicity, and metabolic impairment that are linked to 17. Allan SM, Rothwell NJ (2003) Inflammation in central nervous
functions regulated by astrocytes. The pleiotropic nature of system injury. Philos Trans R Soc Lond B Biol Sci 358:1669–
astrocytes makes these cells ideally suited for combating 1677
18. Wyss-Coray T, Mucke L (2002) Inflammation in neurodegener-
multifactorial brain diseases. It is now necessary to develop
ative disease—a double-edged sword. Neuron 35:419–432
better molecular tools to direct and control the status of 19. Hohlfeld R, Kerschensteiner M, Meinl E (2007) Dual role of
activated astrocytes and to switch them into supportive cells inflammation in CNS disease. Neurology 68:S58–S63, discussion
for neurons exposed to various detrimental conditions. S91–56
20. Popovich PG, Longbrake EE (2008) Can the immune system be
harnessed to repair the CNS? Nat Rev Neurosci 9:481–493
Acknowledgment We thank Dr. Angela Brennan for her careful 21. Pekny M, Wilhelmsson U, Bogestal YR, Pekna M (2007) The role
reading of the manuscript. of astrocytes and complement system in neural plasticity. Int Rev
Neurobiol 82:95–111
22. Pekny M, Nilsson M (2005) Astrocyte activation and reactive
gliosis. Glia 50:427–434
References 23. Wilhelmsson U, Bushong EA, Price DL, Smarr BL, Phung V,
Terada M, Ellisman MH, Pekny M (2006) Redefining the concept
1. Somjen GG (1988) Nervenkitt: notes on the history of the concept of reactive astrocytes as cells that remain within their unique
of neuroglia. Glia 1:2–9 domains upon reaction to injury. Proc Natl Acad Sci USA
2. Ransom B, Behar T, Nedergaard M (2003) New roles for 103:17513–17518
astrocytes (stars at last). Trends Neurosci 26:520–522 24. Oberheim NA, Tian GF, Han X, Peng W, Takano T, Ransom B,
3. Halassa MM, Fellin T, Haydon PG (2007) The tripartite synapse: Nedergaard M (2008) Loss of astrocytic domain organization in
roles for gliotransmission in health and disease. Trends Mol Med the epileptic brain. J Neurosci 28:3264–3276
13:54–63 25. Buffo A, Rite I, Tripathi P, Lepier A, Colak D, Horn AP, Mori T,
4. Bushong EA, Martone ME, Ellisman MH (2004) Maturation of Gotz M (2008) Origin and progeny of reactive gliosis: a source of
astrocyte morphology and the establishment of astrocyte domains multipotent cells in the injured brain. Proc Natl Acad Sci USA
during postnatal hippocampal development. Int J Dev Neurosci 105:3581–3586
22:73–86 26. Raivich G, Bohatschek M, Kloss CU, Werner A, Jones LL,
5. Bushong EA, Martone ME, Jones YZ, Ellisman MH (2002) Kreutzberg GW (1999) Neuroglial activation repertoire in the
Protoplasmic astrocytes in CA1 stratum radiatum occupy separate injured brain: graded response, molecular mechanisms and cues to
anatomical domains. J Neurosci 22:183–192 physiological function. Brain Res Brain Res Rev 30:77–105
6. Halassa MM, Fellin T, Takano H, Dong JH, Haydon PG (2007) 27. Swanson RA, Ying W, Kauppinen TM (2004) Astrocyte influen-
Synaptic islands defined by the territory of a single astrocyte. J ces on ischemic neuronal death. Curr Mol Med 4:193–205
Neurosci 27:6473–6477 28. Trendelenburg G, Dirnagl U (2005) Neuroprotective role of
7. Theis M, Sohl G, Eiberger J, Willecke K (2005) Emerging astrocytes in cerebral ischemia: focus on ischemic precondition-
complexities in identity and function of glial connexins. Trends ing. Glia 50:307–320
Neurosci 28:188–195 29. Floyd CL, Lyeth BG (2007) Astroglia: important mediators of
8. Lobsiger CS, Cleveland DW (2007) Glial cells as intrinsic traumatic brain injury. Prog Brain Res 161:61–79
components of non-cell-autonomous neurodegenerative disease. 30. Silver J, Miller JH (2004) Regeneration beyond the glial scar. Nat
Nat Neurosci 10:1355–1360 Rev Neurosci 5:146–156
9. Seifert G, Schilling K, Steinhauser C (2006) Astrocyte dysfunc- 31. Minagar A, Shapshak P, Fujimura R, Ownby R, Heyes M,
tion in neurological disorders: a molecular perspective. Nat Rev Eisdorfer C (2002) The role of macrophage/microglia and
Neurosci 7:194–206 astrocytes in the pathogenesis of three neurologic disorders:
10. Shin JY, Fang ZH, Yu ZX, Wang CE, Li SH, Li XJ (2005) HIV-associated dementia, Alzheimer disease, and multiple sclero-
Expression of mutant huntingtin in glial cells contributes to sis. J Neurol Sci 202:13–23
neuronal excitotoxicity. J Cell Biol 171:1001–1012 32. Menet V, Prieto M, Privat A, Gimenez y Ribotta M (2003) Axonal
11. Nagai M, Re DB, Nagata T, Chalazonitis A, Jessell TM, Wichterle plasticity and functional recovery after spinal cord injury in mice
H, Przedborski S (2007) Astrocytes expressing ALS-linked deficient in both glial fibrillary acidic protein and vimentin genes.
mutated SOD1 release factors selectively toxic to motor neurons. Proc Natl Acad Sci USA 100:8999–9004
Nat Neurosci 10:615–622 33. Wilhelmsson U, Li L, Pekna M, Berthold CH, Blom S, Eliasson
12. Yamanaka K, Chun SJ, Boillee S, Fujimori-Tonou N, Yamashita C, Renner O, Bushong E, Ellisman M, Morgan TE, Pekny M
H, Gutmann DH, Takahashi R, Misawa H, Cleveland DW (2008) (2004) Absence of glial fibrillary acidic protein and vimentin
Astrocytes as determinants of disease progression in inherited prevents hypertrophy of astrocytic processes and improves post-
amyotrophic lateral sclerosis. Nat Neurosci 11:251–253 traumatic regeneration. J Neurosci 24:5016–5021
13. Wyss-Coray T, Loike JD, Brionne TC, Lu E, Anankov R, Yan F, 34. Kinouchi R, Takeda M, Yang L, Wilhelmsson U, Lundkvist A,
Silverstein SC, Husemann J (2003) Adult mouse astrocytes Pekny M, Chen DF (2003) Robust neural integration from retinal
degrade amyloid-beta in vitro and in situ. Nat Med 9:453–457 transplants in mice deficient in GFAP and vimentin. Nat Neurosci
14. Ridet JL, Malhotra SK, Privat A, Gage FH (1997) Reactive 6:863–868
astrocytes: cellular and molecular cues to biological function. 35. Nakazawa T, Takeda M, Lewis GP, Cho KS, Jiao J, Wilhelmsson
Trends Neurosci 20:570–577 U, Fisher SK, Pekny M, Chen DF, Miller JW (2007) Attenuated
15. Eddleston M, Mucke L (1993) Molecular profile of reactive glial reactions and photoreceptor degeneration after retinal
astrocytes—implications for their role in neurologic disease. detachment in mice deficient in glial fibrillary acidic protein and
Neuroscience 54:15–36 vimentin. Invest Ophthalmol Vis Sci 48:2760–2768
240 Mol Neurobiol (2008) 38:231–241

36. Li L, Lundkvist A, Andersson D, Wilhelmsson U, Nagai N, Pardo evidence for delayed up-regulation and redistribution of GLT-1.
AC, Nodin C, Stahlberg A, Aprico K, Larsson K, Yabe T, Moons Neurochem Int 48:604–610
L, Fotheringham A, Davies I, Carmeliet P, Schwartz JP, Pekna M, 52. Siushansian R, Tao L, Dixon SJ, Wilson JX (1997) Cerebral
Kubista M, Blomstrand F, Maragakis N, Nilsson M, Pekny M astrocytes transport ascorbic acid and dehydroascorbic acid
(2008) Protective role of reactive astrocytes in brain ischemia. J through distinct mechanisms regulated by cyclic AMP. J Neuro-
Cereb Blood Flow Metab 28(3):468–481 chem 68:2378–2385
37. Bush TG, Puvanachandra N, Horner CH, Polito A, Ostenfeld T, 53. Lindenau J, Noack H, Asayama K, Wolf G (1998) Enhanced
Svendsen CN, Mucke L, Johnson MH, Sofroniew MV (1999) cellular glutathione peroxidase immunoreactivity in activated
Leukocyte infiltration, neuronal degeneration, and neurite out- astrocytes and in microglia during excitotoxin induced neuro-
growth after ablation of scar-forming, reactive astrocytes in adult degeneration. Glia 24:252–256
transgenic mice. Neuron 23:297–308 54. Noack H, Lindenau J, Rothe F, Asayama K, Wolf G (1998)
38. Faulkner JR, Herrmann JE, Woo MJ, Tansey KE, Doan NB, Differential expression of superoxide dismutase isoforms in
Sofroniew MV (2004) Reactive astrocytes protect tissue and neuronal and glial compartments in the course of excitotoxically
preserve function after spinal cord injury. J Neurosci 24:2143– mediated neurodegeneration: relation to oxidative and nitrergic
2155 stress. Glia 23:285–297
39. Myer DJ, Gurkoff GG, Lee SM, Hovda DA, Sofroniew MV 55. Nakagawa T, Yabe T, Schwartz JP (2005) Gene expression
(2006) Essential protective roles of reactive astrocytes in traumatic profiles of reactive astrocytes cultured from dopamine-depleted
brain injury. Brain 129:2761–2772 striatum. Neurobiol Dis 20:275–282
40. Okada S, Nakamura M, Katoh H, Miyao T, Shimazaki T, Ishii K, 56. Rudge JS, Pasnikowski EM, Holst P, Lindsay RM (1995) Changes
Yamane J, Yoshimura A, Iwamoto Y, Toyama Y, Okano H (2006) in neurotrophic factor expression and receptor activation follow-
Conditional ablation of Stat3 or Socs3 discloses a dual role for ing exposure of hippocampal neuron/astrocyte cocultures to kainic
reactive astrocytes after spinal cord injury. Nat Med 12:829–834 acid. J Neurosci 15:6856–6867
41. Stoll G, Jander S, Schroeter M (2000) Cytokines in CNS 57. Perillan PR, Chen M, Potts EA, Simard JM (2002) Transforming
disorders: neurotoxicity versus neuroprotection. J Neural Transm growth factor-beta 1 regulates Kir2.3 inward rectifier K+ channels
Suppl 59:81–89 via phospholipase C and protein kinase C-delta in reactive
42. Auernhammer CJ, Melmed S (2001) The central role of SOCS-3 astrocytes from adult rat brain. J Biol Chem 277:1974–1980
in integrating the neuro-immunoendocrine interface. J Clin Invest 58. Leis JA, Bekar LK, Walz W (2005) Potassium homeostasis in the
108:1735–1740 ischemic brain. Glia 50:407–416
43. Escartin C, Brouillet E, Gubellini P, Trioulier Y, Jacquard C, 59. Kielian T (2008) Glial connexins and gap junctions in CNS
Smadja C, Knott GW, Kerkerian-Le Goff L, Deglon N, Hantraye inflammation and disease. J Neurochem 106:1000–1016
P, Bonvento G (2006) Ciliary neurotrophic factor activates 60. Bezzi P, Domercq M, Brambilla L, Galli R, Schols D, De Clercq
astrocytes, redistributes their glutamate transporters GLAST and E, Vescovi A, Bagetta G, Kollias G, Meldolesi J, Volterra A
GLT-1 to raft microdomains, and improves glutamate handling in (2001) CXCR4-activated astrocyte glutamate release via TNFal-
vivo. J Neurosci 26:5978–5989 pha: amplification by microglia triggers neurotoxicity. Nat Neuro-
44. Escartin C, Pierre K, Colin A, Brouillet E, Delzescaux T, sci 4:702–710
Guillermier M, Dhenain M, Deglon N, Hantraye P, Pellerin L, 61. Iadecola C, Nedergaard M (2007) Glial regulation of the cerebral
Bonvento G (2007) Activation of astrocytes by CNTF induces microvasculature. Nat Neurosci 10:1369–1376
metabolic plasticity and increases resistance to metabolic insults. J 62. Liberto CM, Albrecht PJ, Herx LM, Yong VW, Levison SW
Neurosci 27:7094–7104 (2004) Pro-regenerative properties of cytokine-activated astro-
45. Gavillet M, Allaman I, Magistretti PJ (2008) Modulation of cytes. J Neurochem 89:1092–1100
astrocytic metabolic phenotype by proinflammatory cytokines. 63. Sofroniew MV (2005) Reactive astrocytes in neural repair and
Glia 56:975–989 protection. Neuroscientist 11:400–407
46. Albrecht PJ, Dahl JP, Stoltzfus OK, Levenson R, Levison SW 64. John GR, Lee SC, Brosnan CF (2003) Cytokines: powerful
(2002) Ciliary neurotrophic factor activates spinal cord astrocytes, regulators of glial cell activation. Neuroscientist 9:10–22
stimulating their production and release of fibroblast growth 65. Schieven GL (2005) The biology of p38 kinase: a central role in
factor-2, to increase motor neuron survival. Exp Neurol 173:46– inflammation. Curr Top Med Chem 5:921–928
62 66. O’Neill LA, Kaltschmidt C (1997) NF-kappa B: a crucial
47. Pehar M, Vargas MR, Cassina P, Barbeito AG, Beckman JS, transcription factor for glial and neuronal cell function. Trends
Barbeito L (2005) Complexity of astrocyte-motor neuron inter- Neurosci 20:252–258
actions in amyotrophic lateral sclerosis. Neurodegener Dis 2:139– 67. Allan SM, Rothwell NJ (2001) Cytokines and acute neuro-
146 degeneration. Nat Rev Neurosci 2:734–744
48. Dringen R (2000) Metabolism and functions of glutathione in 68. Lucas SM, Rothwell NJ, Gibson RM (2006) The role of
brain. Prog Neurobiol 62:649–671 inflammation in CNS injury and disease. Br J Pharmacol 147
49. Rouach N, Avignone E, Meme W, Koulakoff A, Venance L, (Suppl 1):S232–S240
Blomstrand F, Giaume C (2002) Gap junctions and connexin 69. Munoz L, Ranaivo HR, Roy SM, Hu W, Craft JM, McNamara
expression in the normal and pathological central nervous system. LK, Chico LW, Van Eldik LJ, Watterson DM (2007) A novel p38
Biol Cell 94:457–475 alpha MAPK inhibitor suppresses brain proinflammatory cytokine
50. Vermeiren C, Najimi M, Vanhoutte N, Tilleux S, de Hemptinne I, up-regulation and attenuates synaptic dysfunction and behavioral
Maloteaux JM, Hermans E (2005) Acute up-regulation of deficits in an Alzheimer’s disease mouse model. J Neuroinflam-
glutamate uptake mediated by mGluR5a in reactive astrocytes. J mation 4:21
Neurochem 94:405–416 70. Brambilla R, Bracchi-Ricard V, Hu WH, Frydel B, Bramwell A,
51. O’Shea RD, Lau CL, Farso MC, Diwakarla S, Zagami CJ, Karmally S, Green EJ, Bethea JR (2005) Inhibition of astroglial
Svendsen BB, Feeney SJ, Callaway JK, Jones NM, Pow DV, nuclear factor kappaB reduces inflammation and improves
Danbolt NC, Jarrott B, Beart PM (2006) Effects of lipopolysac- functional recovery after spinal cord injury. J Exp Med
charide on glial phenotype and activity of glutamate transporters: 202:145–156
Mol Neurobiol (2008) 38:231–241 241

71. Fernandez AM, Fernandez S, Carrero P, Garcia-Garcia M, Torres- 81. Wong LF, Goodhead L, Prat C, Mitrophanous KA, Kingsman SM,
Aleman I (2007) Calcineurin in reactive astrocytes plays a key Mazarakis ND (2006) Lentivirus-mediated gene transfer to the
role in the interplay between proinflammatory and anti-inflamma- central nervous system: therapeutic and research applications.
tory signals. J Neurosci 27:8745–8756 Hum Gene Ther 17:1–9
72. Norris CM, Kadish I, Blalock EM, Chen KC, Thibault V, Porter 82. Nimmerjahn A, Kirchhoff F, Kerr JN, Helmchen F (2004)
NM, Landfield PW, Kraner SD (2005) Calcineurin triggers Sulforhodamine 101 as a specific marker of astroglia in the
reactive/inflammatory processes in astrocytes and is upregulated neocortex in vivo. Nat Methods 1:31–37
in aging and Alzheimer’s models. J Neurosci 25:4649–4658 83. Nimmerjahn A, Kirchhoff F, Helmchen F (2005) Resting micro-
73. Bensadoun JC, de Almeida LP, Dreano M, Aebischer P, Deglon N glial cells are highly dynamic surveillants of brain parenchyma in
(2001) Neuroprotective effect of interleukin-6 and IL6/IL6R vivo. Science 308:1314–1318
chimera in the quinolinic acid rat model of Huntington’s 84. Correa-Cerro LS, Mandell JW (2007) Molecular mechanisms of
syndrome. Eur J Neurosci 14:1753–1761 astrogliosis: new approaches with mouse genetics. J Neuropathol
74. Loddick SA, Turnbull AV, Rothwell NJ (1998) Cerebral interleu- Exp Neurol 66:169–176
kin-6 is neuroprotective during permanent focal cerebral ischemia 85. Benediktsson AM, Schachtele SJ, Green SH, Dailey ME (2005)
in the rat. J Cereb Blood Flow Metab 18:176–179 Ballistic labeling and dynamic imaging of astrocytes in organo-
75. Boche D, Cunningham C, Gauldie J, Perry VH (2003) Trans- typic hippocampal slice cultures. J Neurosci Methods 141:41–53
forming growth factor-beta 1-mediated neuroprotection against 86. Shannon C, Salter M, Fern R (2007) GFP imaging of live
excitotoxic injury in vivo. J Cereb Blood Flow Metab 23:1174– astrocytes: regional differences in the effects of ischaemia upon
1182 astrocytes. J Anat 210:684–692
76. Mittoux V, Joseph JM, Conde F, Palfi S, Dautry C, Poyot T, Bloch 87. Cordeau P Jr, Lalancette-Hebert M, Weng YC, Kriz J (2008) Live
J, Deglon N, Ouary S, Nimchinsky EA, Brouillet E, Hof PR, imaging of neuroinflammation reveals sex and estrogen effects on
Peschanski M, Aebischer P, Hantraye P (2000) Restoration of astrocyte response to ischemic injury. Stroke 39:935–942
cognitive and motor functions by ciliary neurotrophic factor in a 88. Rojas S, Martin A, Arranz MJ, Pareto D, Purroy J, Verdaguer E,
primate model of Huntington’s disease. Hum Gene Ther 11:1177– Llop J, Gomez V, Gispert JD, Millan O, Chamorro A, Planas AM
1187 (2007) Imaging brain inflammation with [(11)C]PK11195 by PET
77. Emerich DF, Winn SR, Hantraye PM, Peschanski M, Chen EY, and induction of the peripheral-type benzodiazepine receptor after
Chu Y, McDermott P, Baetge EE, Kordower JH (1997) Protective transient focal ischemia in rats. J Cereb Blood Flow Metab
effect of encapsulated cells producing neurotrophic factor CNTF 27:1975–1986
in a monkey model of Huntington’s disease. Nature 386:395–399 89. Sibson NR, Lowe JP, Blamire AM, Martin MJ, Obrenovitch TP,
78. Slezak M, Goritz C, Niemiec A, Frisen J, Chambon P, Metzger D, Anthony DC (2008) Acute astrocyte activation in brain detected
Pfrieger FW (2007) Transgenic mice for conditional gene by MRI: new insights into T(1) hypointensity. J Cereb Blood
manipulation in astroglial cells. Glia 55:1565–1576 Flow Metab 28:621–632
79. Mori T, Tanaka K, Buffo A, Wurst W, Kuhn R, Gotz M (2006) 90. Chen Y, Swanson RA (2003) Astrocytes and brain injury. J Cereb
Inducible gene deletion in astroglia and radial glia—a valuable Blood Flow Metab 23:137–149
tool for functional and lineage analysis. Glia 54:21–34 91. Dirnagl U, Simon RP, Hallenbeck JM (2003) Ischemic
80. Déglon N, Hantraye P (2005) Viral vectors as tools to model and tolerance and endogenous neuroprotection. Trends Neurosci
treat neurodegenerative disorders. J Gene Med 7:530–539 26:248–254