Neuron Glia Biology, page 1 of 10.

# Cambridge University Press, 2012

doi:10.1017/S1740925X12000038

The role of microglia at synapses in the

healthy CNS: novel insights from recent
imaging studies
marie-e‘ve tremblay

In the healthy brain, quiescent microglia continuously remodel their shape by extending and retracting highly motile pro-
cesses. Despite a seemingly random sampling of their environment, microglial processes specifically interact with subsets of
synaptic structures, as shown by recent imaging studies leading to proposed reciprocal interactions between microglia and
synapses under non-pathological conditions. These studies revealed that various modalities of microglial dynamic behavior
including their interactions with synaptic elements are regulated by manipulations of neurotransmission, neuronal activity
and sensory experience. Conversely, these observations implied an unexpected role for quiescent microglia in the elimin-
ation of synaptic structures by specialized mechanisms that include the phagocytosis of axon terminals and dendritic
spines. In light of these recent discoveries, microglia are now emerging as important effectors of neuronal circuit
reorganization.

Keywords: Motility, dendritic spine, plasticity, remodeling, elimination

INTRODUCTION promptly to any changes occurring in their environment,

and therefore experimental ex vivo and in vitro preparations
Microglia were discovered almost a century ago by Pio del inevitably result in transformation of their normally prevailing
Rio-Hortega. From his light microscopic observations of behavior. Recently, the development of in vivo two-photon
silver carbonate impregnated microglia in rabbit, cat and laser scanning microscopy (or two-photon) has for the first
human brain in situ (1919–27; see Fig. 1A), del Rio-Hortega time allowed the study of resting microglia in the intact
proposed that microglia have a mesodermal origin, colonize brain of living animals (Davalos et al., 2005; Nimmerjahn
the brain during embryonic development, become evenly dis- et al., 2005), using CX3CR1-GFP transgenic mice in which
tributed, display little morphological variation in the mature microglia are fluorescently labeled (Jung et al., 2000; see
brain, but respond to pathological insults by transforming Section 1 for details on these important mice). When
their morphology and acquiring the capacity to migrate, pro- imaged through the skull, microglia in the superficial layers
liferate and phagocytose (Del Rio-Hortega, 1932; reviewed in of somatosensory and motor cortices of adult animals dis-
Kettenmann et al., 2011). Since his formulation of these played small cell bodies with highly ramified thin processes
visionary postulates, which are still valid today, the changes extending radially and were distributed homogeneously in
in microglial morphology, gene expression and functional three dimensions (3D), as anticipated from previous in situ
behavior during infection, trauma, ischemia and various neu- observations (see Del Rio-Hortega, 1932). But quite unexpect-
ropsychiatric and neurodegenerative diseases have been edly, microglia were also extremely dynamic, continuously
extensively studied. In all these situations, where normal extending and retracting processes at an average velocity of
homeostasis is impaired, activated microglia were shown to 1.5 mm min21 (2.2 mm min21 for terminal protrusions)
perform a variety of roles essential to the immune and inflam- and maximal velocity of 4 mm min21, thereby leading to
matory response of the central nervous system (CNS), includ- comprehensive changes in cellular morphology on a time
ing the release of numerous factors and compounds, scale of minutes (Davalos et al., 2005; Nimmerjahn et al.,
presentation of antigens to T cells and phagocytosis of tissue 2005; Fig. 1B). This dynamic remodeling contrasts both to
debris, damaged cells and microbes (reviewed in Kreutzberg, neurons and other types of glial cells imaged in vivo in the cer-
1996; Hanisch and Kettenmann, 2007; Ransohoff and Perry, ebral cortex of juvenile and adult mice, which showed no com-
2009; Rivest, 2009; Graeber, 2010; Prinz et al., 2011). parable restructuring of processes. Astrocytic processes
The roles of ‘resting’ or immunologically quiescent micro- showed no motility at all over 1 h (Eom et al., 2011),
glia have remained relatively unknown (also see Tremblay neuron–glial antigen 2 (NG2)-positive glial cells remodeled
et al., 2011). This is largely due to the difficulties of studying their processes over several days (Hughes et al. (2011);
microglia in their non-activated state. Microglia respond Society for Neuroscience abstract No. 548.12), while the
most motile dendritic spines only reached an average velocity
of 0.02 mm min21 (Majewska and Sur, 2003). Following
Corresponding author:
these observations, ‘resting’ microglia have emerged as the
Marie-Ève Tremblay most structurally dynamic cells of the mature CNS discovered
Email: [email protected] so far.

1
2 marie-e‘ ve tremblay

Fig. 1. Evolving views of resting microglia, from the pioneer observation of silver carbonate impregnated microglia in situ to two-photon imaging of
microglia-synapse dynamics in vivo. (A) Drawing by Pio del Rio-Hortega of ramified microglial morphology (modified from Del Rio-Hortega, 1919). (B)
Two-photon micrograph showing microglial process extensions (green) and retractions (red) in a CX3CR1-GFP mouse over the course of 20 min (reproduced
from Nimmerjahn et al., 2005). (C) Electron micrograph showing direct contacts of an Iba1-immunostained microglial process with dendritic spines (pink),
axon terminals (blue), synaptic cleft (arrow) and perisynaptic astrocytic processes (green). Note the extracellular space pockets surrounding the microglia
(asterisks). (D) 3D reconstruction of proximal (cut in transverse) and distal (longitudinal) microglial processes uncovering simultaneous contacts with
dendritic spines varying in shape and size (pink). Extracellular space pockets are displayed in white, and a phagocytic inclusion within the proximal microglial
process is in purple. (E) Two-photon time-lapse micrographs showing a dynamic microglial process (yellow for presentation purposes) transiently interacting
with dendritic spines (green) in a CX3CR1-GFP/Thy1-YFP mouse over the course of 20 min. Each frame was captured 5 min apart. Red arrowheads indicate
non-targeted spines, and white arrowheads targeted ones. (C) to (E) reproduced from Tremblay et al. (2010b).

This unexpected behavior suggested that resting or surveil- regulation of microglial behavior by neurotransmission, neur-
lant microglia may continuously survey the brain parenchyma onal activity and sensory experience, and conversely, for an
as part of their immune function, which would justify the sub- involvement of microglia in the elimination of synaptic struc-
stantial expenditure of energy required to continuously main- tures during experience-dependent plasticity.
tain microglial dynamics in the normal brain, without
excluding the possibility of an additional, distinct contribution
to normal brain physiology. Although the field of research Characterization of microglial interactions
addressing the roles of microglia under non-pathological con-
ditions is still very young and comprises only a few studies, for
with synaptic elements
the purpose of this special issue of Neuron Glia Biology, I will Electron microscopy (EM) enables the investigator to accu-
discuss these emerging roles of microglia at synapses in the rately visualize the fine structural intricacies between all the
mammalian CNS. Within this limited scope, I will focus on cells and their various compartments, and the extracellular
non- or minimally invasive in situ and in vivo imaging matrices binding them together, thereby uncovering their
approaches, and discuss recent observations describing the immensely complex array of possible functional interactions
structural interactions between microglia and synaptic at the local level within any tissue, at the best resolution
elements. I will also examine the evidence supporting a (approximately 1 nm) so far achieved by a morphological
 role of microglia at synapses in the healthy cns 3

technique. Over the last 60 years, ultrastructural examinations Serial section EM (SSEM) with 3D reconstruction further
of the CNS have provided unprecedented insights into the revealed the geometry of various contacts between a single
functional relationships of neurons and glial cells (astrocytes, microglial process and numerous excitatory synapses
NG2-positive cells, oligodendrocytes and microglia in most (Fig. 1D) in juvenile animals (Tremblay et al., 2010b). While
regions of the mature CNS) under various conditions of microglial contacts generally occurred en passant, one of the
health and disease. For example, EM contributed to the estab- contacts displayed morphological specialization in the form
lishment of astrocytes as synaptic partners modulating neuro- of finger-like protrusions wrapping around a spine, suggesting
transmission (see Ventura and Harris, 1999; reviewed in immediate and functional interactions. Clathrin-coated pits
Theodosis et al., 2008) and to the recent discovery of synaptic were also observed at interfaces between microglial processes
interactions between NG2-positve cells and neurons and neuronal or astrocytic elements of synapses, on either side
(reviewed in Bergles et al., 2010). Some 40 years ago, a possible of the interaction, suggesting reciprocal exchange of molecular
but still controversial role of activated microglia in synaptic signals through clathrin-mediated endocytosis of membrane-
stripping also emerged from EM observations showing that bound receptors and their ligands (see Le Roy and Wrana
microglial processes structurally intervene between pre- (2005) for details on clathrin-mediated endocytosis). Even
synaptic axon terminals and post-synaptic dendrites or neur- though the nature of the exchanged signals remains to be
onal cell bodies under pathological conditions (Blinzinger and determined, these two types of morphological specializations
Kreutzberg, 1968; reviewed in Graeber, 2010; Perry and (finger-like protrusions and clathrin-coated pits) indicate
O’Connor, 2010; see also Trapp et al., 2007). that surveillant microglia can functionally interact with synap-
Microglial cell bodies under healthy, non-pathological con- tic structures under non-pathological conditions.
ditions were also described with EM just as long ago (Herndon, Inspired by the extreme dynamism of surveillant microglia
1964; Mori and Leblond, 1969), but since microglial processes in the intact CNS (Davalos et al., 2005; Nimmerjahn et al.,
are rarely contiguous to their cell bodies in single ultrathin sec- 2005), recent studies have additionally explored the dynamic
tions, and these processes have few distinctive ultrastructural nature of microglial interactions with synaptic structures
features, their discrimination from other subcellular elements using two-photon in vivo imaging. The imaging of microglia
(such as small unmyelinated axons, dendrites and other and synaptic structures required the development of trans-
small glial processes) required the development of selective genic mice in which both microglia and neurons are fluores-
labeling approaches. In particular, an enzymatic staining for cently labeled. For this purpose, Wake et al. (2009) crossed
thiamine pyrophosphatase (TPPase) allowed examination of Iba1-GFP (Hirasawa et al., 2005) with Thy1-GFP mice
microglial processes in situ for the first time 30 years ago, (Feng et al., 2000) for visualizing microglia and neuronal
revealing their direct apposition with excitatory synapses in elements including axons, dendrites, terminals and spines in
the cerebral cortex of mature rats (Murabe and Sano, 1982). the same color (green). More recently, other investigators
Since TPPase labels not only microglia, but also neurons, astro- crossed the CX3CR1-GFP mice from Jung et al. (2000),
cytes and oligodendrocytes (Kumamoto et al., 1976; Castellano which provide exceptional in vivo visualization of microglial
et al., 1989), a more specific approach was needed to validate morphology including the distal processes and fine protru-
these observations. To this end, immunocytochemical EM sions (Fig. 2), with Thy1-YFP mice (Feng et al., 2000 or
with a specific antibody directed against ionized calcium Hirrlinger et al., 2005), enabling simultaneous visualization
binding adaptor molecule 1 (Iba1), a protein only expressed of microglia and neurons in two different colors (green and
by microglia within the healthy CNS (Ito et al., 1998), was
recently performed, showing that microglial processes can
interact with both pre-synaptic axon terminals and post-
synaptic dendritic spines in the somatosensory cortex of
adult mice (Wake et al., 2009). Contrary to these interactions
with excitatory synapses, the existence of microglial inter-
actions with inhibitory synapses under normal physiological
conditions remains yet unknown.
Extending this characterization, Tremblay et al. (2010b)
revealed with a similar immunocytochemical EM approach
(see Tremblay et al. (2010a) for methodological details) that
almost all microglial processes localize to the vicinity of excit-
atory synapses in the superficial layers of juvenile mouse visual
cortex: 94% of microglial processes directly contacted term-
inals, spines, perisynaptic astrocytic processes and synaptic
clefts, in decreasing order of frequency, and 68% of these
processes contacted more than one synaptic element
(Fig. 1C). Conversely, the proportion of each of those synaptic
elements contacted by microglial processes could not be deter-
mined, because not all microglial processes were positively
identified, since only a subset were stained for Iba1 under
these stringent immunocytochemical conditions. Recently, Fig. 2. Imaging of surveillant microglia in vivo with the CX3CR1-GFP mice
microglial interactions with terminals, spines, astrocytic pro- from Jung et al. (2000). Two-photon micrograph (maximum z-projection of
20 slices captured 1 mm apart) showing the density and morphology of
cesses and synaptic clefts were also confirmed throughout microglial cell bodies, processes and terminal protrusions, with or without
adulthood and normal aging, in the superficial layers of bulbous endings (shown by arrowheads), when imaged through a thinned
mouse visual and auditory cortices (Tremblay et al., 2012). skull in somatosensory cortex.
4 marie-e‘ ve tremblay

yellow respectively, see Fuhrmann et al. (2010); Tremblay smaller spines showing the most pronounced changes.
et al. (2010b) and Dibaj et al. (2010) for examples in cerebral Surprisingly, chronic imaging over 2 days further revealed a
cortex and spinal cord). It should be noted that CX3CR1-GFP statistically significant difference in the elimination rate of
heterozygous mice generally used for imaging may be partially microglia-contacted spines: spines contacted by microglia
deficient in fractalkine signaling, a pathway that was pre- were more frequently eliminated than non-contacted spines
viously shown to modulate microglial activation and cytokine (24 versus 7%; P  0.05), and in all cases, only the small
release, neuron–microglia communication and neuronal sur- spines were seen to disappear. These observations suggest
vival in different contexts of disease and CNS regions (for that despite an apparently random sampling of the parench-
example see Harrison et al., 1998; Cardona et al., 2006; yma, microglial processes specifically target a subset of
Fuhrmann et al., 2010; Cho et al., 2011; Donnelly et al., small, structurally dynamic and transient dendritic spines.
2011) due to the replacement of one of the two alleles encod- Altogether, these immunocytochemical EM, SSEM with 3D
ing microglial CX3CR1 with a GFP reporter. Nevertheless, reconstruction and two-photon in vivo imaging observations
many aspects of microglial morphology, process motility indicate that as part of their normal surveillance, microglial
(Nimmerjahn et al., 2005; Wake et al., 2009) and interactions processes frequently interact with a subset of synaptic struc-
with synaptic elements (Wake et al., 2009; Tremblay et al., tures. This specificity raises the intriguing possibility that
2010b) that were examined in vivo were comparable microglia could perceive neuronal activity or structural
between the CX3CR1-GFP heterozygous mice and Iba1-GFP changes at individual synapses.
mice from Hirasawa et al. (2005) where CX3CR1 is not
deleted.
Using Iba1-GFP/Thy1-GFP mice and CX3CR1-GFP/ Regulation of microglial behavior by
Thy1-YFP mice, respectively, Wake et al. (2009) and neurotransmission, neuronal activity
Tremblay et al. (2010b) recently explored microglial inter- and sensory experience
actions with axon terminals and dendritic spines from
subsets of cortical layer V pyramidal neurons in vivo. With Cultured microglia were shown to express various receptors
great caution, a non-invasive thinned-skull preparation was for acetylcholine, glutamate, GABA, serotonin, dopamine,
performed to prevent brain injury and microglial activation noradrenaline, ATP, cannabinoids, opioids and neuropeptides
(see Tremblay et al. (2010b), Yang et al. (2010) and Marker among other neurotransmitters. Each of these neurotransmit-
et al. (2010) for methodological details). In this manner, ters induced changes in the electrophysiological properties of
Wake et al. (2009) first showed that microglial processes tran- microglia (reviewed in Biber et al. (2007), Pocock and
siently interact with terminals and spines, in the somatosen- Kettenmann (2007), Kettenmann et al. (2011); and see Liu
sory and visual cortices of adult animals. Microglial contacts et al. (2011) for details and receptors involved) while ATP
with terminals generally lasted 5 min and occurred at a fre- and glutamate also influenced microglial motility in vitro
quency of 1 per hour and per terminal. During the contacts, (Honda et al., 2001; Haynes et al., 2006; Liu et al., 2009; but
microglial processes displayed enlarged extremities or bulbous also see Wu and Zhuo, 2008; Chen et al., 2010). But impor-
endings (Wake et al., 2009) resembling the morphological tantly, since microglial resting state is believed to differ
specializations described with SSEM and 3D reconstruction between in vitro and in vivo environments, given the
in situ (Tremblay et al., 2010b). Additionally, Tremblay absence of in situ evidence it remains uncertain whether sur-
et al. (2010b) observed microglial contacts with terminals veillant microglia also express receptors for these neurotrans-
and spines that varied in duration between 5 and 30 min in mitters in the intact CNS (aside from ATP, see below).
the visual cortex of juvenile animals, using time-lapse Therefore, how these demanding immune cells normally
imaging every 5 min, thereby adding a new dimension to respond to the various neurotransmitters released by
the modalities of microglial interactions with synaptic neurons and glial cells remains to be seen.
elements occurring in vivo under non-pathological conditions. To determine whether surveillant microglia can detect
In the mature healthy CNS, neuronal networks are con- neuronal and synaptic activity under physiological conditions,
tinuously remodeled through the formation, modification recent studies investigated microglial response to neurotrans-
and elimination of synaptic structures (see Fortin et al. mission, neuronal activity and sensory experience in the intact
(2011) for molecular mechanisms of structural plasticity) in CNS. Specifically, microglial sampling of the parenchyma,
relation with behavioral and sensory experience. In particular, basal velocity of their processes and dynamic interactions
two-photon in vivo imaging in Thy1-YFP or Thy1-GFP mice with axon terminals and dendritic spines were examined.
revealed that spines, especially small ones, are structurally
dynamic and transient in mouse visual, somatosensory, neurotransmission
motor and frontal cortices (Trachtenberg et al., 2002; Zuo Even though it remains unknown whether changes in neuro-
et al., 2005; De Paola et al., 2006; Majewska et al., 2006; also transmission at individual synapses can influence microglia in
see Alvarez and Sabatini, 2007; Holtmaat and Svoboda, the intact CNS, two imaging studies have revealed conse-
2009). To determine a possible role of surveillant microglia quences of globally modulating glutamatergic, GABAergic
in the structural remodeling of synaptic structures under or purinergic neurotransmission on microglial morphology
normal physiological conditions, Tremblay et al. (2010b) and dynamic behavior in vivo or ex vivo (Davalos et al.,
also examined the size changes of spines and terminals 2005; Fontainhas et al., 2011).
before, during and after microglial contacts. Spines contacted The first evidence for a regulation of surveillant microglia
by microglial processes during imaging (30–120 min sessions) by neurotransmission in vivo was provided by the pioneer
were found to be smaller initially than those which remained study from Davalos et al. (2005). In this study, most of the
non-contacted. Spines, but not terminals, also underwent imaging was performed through a thinned skull, but the appli-
transient increases in size during microglial contact, with cation of pharmacological agents onto the cortical surface
 role of microglia at synapses in the healthy cns 5

required a more invasive procedure, i.e. the removal of the neuronal activity
skull and dura mater. Nonetheless, microglial morphology Two additional in vivo studies correlated the changes in spon-
and motility seemed unaffected by the procedure at the corti- taneous or evoked neuronal activity, induced by pharmacologi-
cal depths that were imaged. In this manner, cortical surface cal manipulations of neuronal excitability or neurotransmission,
application of apyrase which catalyzes the hydrolysis of extra- eye enucleation or reduction of body temperature and moni-
cellular ATP and ADP reduced the basal velocity (length tored by electrophysiological recording or calcium imaging,
changes over time) of microglial processes. In contrast, appli- with the changes in microglial sampling, process velocity and
cation of carbenoxolone (an inhibitor of connexin and dynamic interactions with synaptic elements in the intact CNS
pannexin-1) or flufenamic acid (an inhibitor of connexin) (Nimmerjahn et al., 2005; Wake et al., 2009). Contrary to the
nearly abolished their velocity within 30 min (Davalos et al., previously described modulations of neurotransmission in vivo
2005), suggesting that ATP released from astrocytes via con- or ex vivo, these manipulations influenced excitatory and
nexin channels, such as forming gap junctions, or hemichan- inhibitory neurons simultaneously within extensive networks.
nels could mediate the effects of extracellular purines on In particular, Nimmerjahn et al. (2005) examined the
microglial process velocity (see Fields and Burnstock, 2006; changes in microglial dynamics during pharmacological
Inoue et al., 2007; Kang et al., 2008; Halassa and Haydon, manipulations of spontaneous neuronal activity in vivo.
2010 for details; also see discussion below). With a methodology analogous to Davalos et al. (2005),
Building on these observations, Fontainhas et al. (2011) enhancing excitatory neuronal activity by surface application
recently discovered a modulation of microglial dynamics by of BCC increased microglial sampling of the parenchyma
glutamatergic, GABAergic and purinergic neurotransmission over 1 h, as evidenced by differences in fluorescence intensity
ex vivo in adult CX3CR1-GFP mice acute retinal explants, a between the outer (distal microglial processes) and inner rings
preparation that minimizes CNS injury because the retina is (proximal processes) surrounding individual microglial cell
imaged as a whole. In this preparation, application of gluta- bodies. However, silencing excitatory and inhibitory neuronal
mate receptor agonists (AMPA and kainate) into the record- activity with TTX produced no significant outcome over 1 h,
ing chamber increased the size (area and length), complexity suggesting that microglial sampling proceeds at a basal rate
and basal velocity of microglial processes, whereas GABA or that is unaffected by neuronal activity (Nimmerjahn et al.,
GABAA receptor antagonist bicuculline (BCC) had opposite 2005). In both cases, the changes in neuronal activity were
effects. ATP also increased the size, complexity and velocity confirmed by electrocorticogram recording using minimally
of microglial processes within a few minutes of application, invasive electrodes. Moreover, both BCC and TTX were inef-
while opposite changes were observed with apyrase, suramin fective at modifying the basal velocity of microglial processes
(a purinergic P2 receptor antagonist) or probenecid (a (Nimmerjahn et al., 2005), suggesting that neuronal activity
pannexin-1 inhibitor; Silverman et al., 2008), suggesting increases the rate of branching or turnover of individual
ATP release from pannexin-1 channels, which are expressed microglial processes without influencing their basal velocity.
by excitatory neurons, inhibitory neurons and astrocytes The discrepancy in BCC-induced changes in microglial
under non-pathological conditions (see Inoue et al., 2007; process velocity between Nimmerjahn et al. (2005) and
MacVicar and Thompson, 2010; also see Wong et al., 2011). Fontainhas et al. (2011) is unclear and could result from
Additionally, voltage-clamp recordings failed to reveal any differences between CNS regions or between in vivo and ex
inward or outward currents in microglia during local appli- vivo environments, as microglial process basal velocity
cation of AMPA or GABA, although application of ATP did was approximately two-fold higher in retina ex vivo than in
induce large inward currents (Fontainhas et al., 2011). cerebral cortex or spinal cord in vivo (Liang et al., 2009
During brain injury, purinergic signaling was shown to versus Nimmerjahn et al., 2005; Wake et al., 2009; Dibaj
regulate microglial process dynamics in vivo, with ATP et al., 2010).
released from the damage tissue and/or the surrounding Wake et al. (2009) further examined the influence of
cells inducing process extension in surveillant microglia via sensory-evoked and spontaneous neuronal activity on micro-
P2Y12 microglial receptors, and adenosine (produced from glial interactions with terminals in vivo. In this study, binocu-
the extracellular degradation of ATP into ADP, AMP and lar enucleation, injection of TTX into both eyes and reduction
adenosine notably by microglial ectonucleotidases; see of body temperature from 37 to 328C each reduced neuronal
Kettenmann et al., 2011) inducing process retraction in acti- activity over 2 h, which was confirmed by calcium imaging in
vated microglia via A2A receptors (Davalos et al., 2005; separate experiments, and subsequently induced a retraction
Haynes et al., 2006; Orr et al., 2009). Under normal physio- of microglial processes over 4–6 h (eye enucleation), while
logical conditions, since pre-synaptic release of glutamate also halving the frequency of microglial contacts with individ-
and GABA is regulated by neuronal activity, and ATP is ual terminals (TTX and body temperature). These obser-
released from connexin and pannexin-1 channels in an vations provided the first evidence for a modulation of
activity-dependent manner (see Fields and Burnstock, 2006; microglial interactions with synaptic structures by neuronal
Inoue et al., 2007; MacVicar and Thompson, 2010), ATP activity. Furthermore, microglial process basal velocity
could additionally represent an indirect mechanistic link remained unchanged during these manipulations (Wake
between glutamatergic and GABAergic neurotransmission et al., 2009), in line with the observations from Nimmerjahn
and microglial process dynamics. The particular subtypes of et al. (2005) during cortical surface application of TTX. The
microglial purinergic receptors recruited by ATP in these con- discrepancy between the other effects of TTX reported by
ditions are yet to be determined, but P2Y12 receptor is prob- these two studies, i.e. a retraction of microglial processes
ably not involved since surveillant microglia displayed a (Wake et al., 2009) versus an unchanged microglial sampling
normal basal process velocity when imaged through the (Nimmerjahn et al., 2005), could result from differences in
skull in the cerebral cortex of P2Y12 receptor-deficient adult experimental paradigms, such as the application of TTX
CX3CR1-GFP mice in vivo (Haynes et al., 2006). onto the brain surface versus into the eyes, differences in the
6 marie-e‘ ve tremblay

anesthetics used for in vivo imaging (isoflurane for continuously and differentially modulating their morphology,
Nimmerjahn et al. (2005) versus ketamine and xylazine for basal process velocity, sampling of the parenchyma and
Wake et al. (2009)), and in the durations of the experiments dynamic interactions with synaptic structures under normal
and imaging sessions during which the resting state of micro- physiological conditions.
glia could vary (2 h for Nimmerjahn et al. (2005) versus
8 h for Wake et al. (2009)).
Microglial elimination of synaptic structures
sensory experience The specificity of microglial interactions with structurally
In addition to these manipulations of neurotransmission and dynamic and frequently eliminated dendritic spines raises
neuronal activity, the effects of environmental experience on the intriguing likelihood that surveillant microglia may facili-
microglial interactions with synaptic elements were recently tate the elimination of dendritic spines, with or without synap-
examined in vivo, during binocular light deprivation tic contacts, thereby preventing the formation of new synapses
induced by housing the animals in complete darkness or promoting the elimination of existing ones. In this way,
throughout their fourth postnatal week (Tremblay et al., they could contribute to the experience-dependent remodel-
2010b). This time point coincides with the beginning to the ing of neuronal networks that is required for learning and
peak of the critical period for experience-dependent plasticity memory storage throughout the lifespan. Mechanisms by
in the visual system, where neuronal circuit remodeling in which surveillant microglia may eliminate synaptic structures
relation with visual experience is most pronounced (see were recently proposed and will be described here.
Hooks and Chen, 2007). Binocular deprivation is known to
increase spine motility and turnover in the visual cortex of
juvenile and adult mice (Majewska and Sur, 2003; Keck microglial phagocytosis of axon
et al., 2008), making this paradigm ideal for correlating micro- terminals and dendritic spines
glial behavior with the experience-dependent modification of Like macrophages, activated microglia contribute to preser-
synaptic structures under non-pathological conditions. ving normal homeostasis by their phagocytosis and clearance
In this manner, comparing differences in microglial mor- of cellular debris (see Neumann et al. (2009) and Napoli and
phology over time revealed a reduced microglial sampling of Neumann (2009) for details on the mechanisms of microglial
the parenchyma that was accompanied by a sparsening and phagocytosis in different contexts). Under normal physiologi-
thickening of microglial processes during visual deprivation, cal conditions, Nimmerjahn et al. (2005) observed a few
suggesting a relationship between microglial morphology and examples of microglial processes engulfing unidentified
surveillance behavior. Even though microglial contacts with tissue components which were subsequently transported to
spines were not significantly altered in duration and frequency the cell bodies in vivo. More recently, Tremblay et al.
during visual deprivation, microglial processes were found to (2010b) also noticed cellular inclusions with ultrastructural
change their preference of contact among the small and struc- features of axon terminals and dendritic spines, i.e. synaptic
turally dynamic spines, from transiently growing spines during vesicles and post-synaptic densities, respectively, within
normal experience, to persistently shrinking spines during microglial processes and cell bodies in juvenile mouse somato-
deprivation. All these changes reversed during reexposure to sensory cortex in situ. These putative terminals and spines
a normal 12-h light/dark cycle for 2 days (Tremblay et al., occasionally displayed synaptic contacts, and invariably pre-
2010b). These observations indicate that microglial inter- sented ultrastructural features of healthy cellular elements
actions with dendritic spines are regulated by sensory experi- such as an electron-lucent cytoplasm (see Schmechel (1999)
ence in the healthy CNS. Since microglial processes for additional features of apoptotic and necrotic cells), unlike
preferentially interact with a subset of small, structurally apoptotic cells that are phagocytosed by resting microglia
dynamic and frequently eliminated spines during normal during adult neurogenesis (Sierra et al., 2010). In vivo, micro-
sensory experience (Tremblay et al., 2010b), and small and glial processes with phagocytic specializations were also
shrinking spines, such as those contacted by microglia observed, sometimes encircling neuronal elements, but since
during visual deprivation, have been shown to gradually disap- these structures were stable throughout the 30–120 min
pear over a span of days in the somatosensory cortex of adult imaging sessions, episodes of microglial engulfment followed
mice in vivo (Holtmaat et al., 2006), these findings suggest that by a full disappearance of the synaptic elements were not com-
microglial contacts could facilitate the elimination of small and pletely visualized. Both in situ and in vivo, microglial phagocy-
structurally dynamic spines. Furthermore, since Knott et al. tic structures became more prevalent during light deprivation,
(2006) revealed that small and recently formed spines generally and still persisted after reexposure to light for 2 days (Tremblay
lack synaptic contacts in the somatosensory cortex of adult et al., 2010b), in a context of sensory experience where dendri-
mice in situ, using two-photon in vivo imaging over 1 month tic spine remodeling and elimination is also increased
combined with retrospective SSEM with 3D reconstruction, (Majewska and Sur, 2003; Keck et al., 2008).
it also remains uncertain whether the small spines contacted Additionally, Tremblay et al. (2012) recently observed pha-
by microglia receive synaptic contacts. gocytic inclusions within microglial cell bodies and processes
Altogether, by examining different CNS regions and stages throughout adulthood and normal aging, using immunocyto-
of the lifespan, using different experimental paradigms for chemical EM. This study revealed that microglial cell bodies
modulating neurotransmission, neuronal activity or sensory and processes accumulate large amounts of phagocytic
experience, and different methodologies for assessing the inclusions, which often resemble terminals and spines, with or
changes in microglial dynamics in all these conditions, these without an electron-lucent cytoplasm, but also lysosomal lipo-
studies have revealed that microglia do not behave randomly pigments, large vesicles, vacuoles and lipid droplets, in the
in the absence of pathological insults. In light of the combined visual and auditory cortices of two stains of mice normally
observations, resting microglia can now be viewed as undergoing complementary age-related loss of vision (CBA/
 role of microglia at synapses in the healthy cns 7

CaJ mice) or hearing (C57Bl/6J mice) during their second year manner, undergoing expansion during light deprivation and
of life. Surprisingly, this accumulation was exacerbated by the shrinkage during light reexposure. More recently, pockets of
age-related loss of visual or auditory function, with nearly all extracellular space juxtaposing microglial cell bodies and pro-
microglia containing phagocytic inclusions and 20% of micro- cesses were also observed in mouse auditory and visual cor-
glial cells becoming almost completely filled by inclusions, akin tices throughout adulthood and normal aging (Tremblay
to fat granule cells or gitter cells, in this context of sensory depri- et al., 2012). Interestingly, more than 40 years ago, Herndon
vation. Also, microglial processes extending between axon term- (1964) mentioned that a ‘small amount of extracellular
inals and dendritic branches were encountered, suggesting an space may be seen in a few places about the cell, particularly
early stage of phagocytosis or an evidence of synaptic stripping about its thick processes’, when describing for the first time
under non-pathological conditions. Together, these obser- microglial cell bodies and contiguous, proximal processes at
vations during adolescence, adulthood and normal aging the ultrastructural level under normal physiological con-
suggest that microglial phagocytosis is regulated in an ditions in adult rat cerebellum. Whether microglia create
experience-dependent manner in the mature CNS. these extracellular spaces remains unknown.
In line with these observations, activated microglia have Among candidate molecules for microglial involvement,
been shown to be involved in the apoptotic and phagocytic the polysialylated form of neural cell adhesion molecule
elimination of excess neurons during early postnatal develop- (PSA-NCAM) increases the size of the extracellular space
ment (see Bessis et al. (2007) and Pont-Lezica et al. (2011) for (Yang et al., 1992; see Bonfanti, 2006; Syková and
details and additional roles during development), and Nicholson, 2008) and was recently found to be expressed by
recently, Paolicelli et al. (2011) showed that microglia can microglia in vitro (Wang and Neumann, 2010). Proteases
also eliminate synaptic elements during the second and degrading extracellular matrix proteins which are expressed
third postnatal weeks in CX3CR1-GFP mouse hippocampus by activated microglia in vitro might also be involved (see
in situ. In that study, puncta immunopositive for either Nakanishi, 2003; Choi et al., 2010). Application of proteases
PSD95 (marker of dendritic spines) or SNAP25 (axon term- in vitro and in vivo has been shown to promote spine motility
inals) were found to be colocalized at nanometer resolution and pruning, as well as activity-dependent and experience-
within GFP-positive microglial processes using stimulated dependent plasticity in the visual cortex of adult mice
emission depletion microscopy (see Willig et al. (2006) for (Pizzorusso et al., 2002; Mataga et al., 2004; Oray et al.,
technical details). Microglial colocalization with PSD95 was 2004), thereby suggesting a possible link between the regu-
more prevalent in homozygous CX3CR1-GFP mice, where lation of the extracellular space and the elimination of the
fractalkine signaling is completely eliminated (Jung et al. small and structurally dynamic spines preferentially contacted
(2000); see Section 1 for details about these mice), and was by microglial processes in vivo (Tremblay et al., 2010b).
accompanied by a reduced density of microglial cells and an
increased density of PSD95 puncta with or without microglial
colocalization, suggesting that reduced microglial numbers SUMMARY AND CONCLUSION
could impact the pruning of supernumerary spines
(Paolicelli et al., 2011). Providing additional molecular Over the last century, the development of selective staining
insights, recent observations from Schafer et al. (2011; approaches and concomitant refinement of imaging tech-
Society for Neuroscience abstract No. 663.03) also revealed niques, from conventional light and electron microscopy to
that microglia can phagocytose terminals in a complement two-photon laser scanning microscopy, have provided unpre-
C3-dependent manner (Stevens et al., 2007) during early post- cedented insights into microglial morphology, dynamic be-
natal development in mouse thalamus in situ. havior and interactions with excitatory synapses in the
Altogether, these studies propose a novel physiological role intact CNS. In particular, recent imaging studies have revealed
for microglial phagocytosis, in the elimination of individual that microglia are extremely dynamic in their presumed
synaptic structures and quite likely of entire excitatory ‘resting’ state, continuously surveying their local environment
synapses, throughout the lifespan. with highly motile processes which transiently contact subsets
of synaptic structures in vivo, especially the small, dynamic
and frequently eliminated dendritic spines. Surveillant micro-
regulation of the perisynaptic glia also perceive ongoing changes in neurotransmission,
extracellular space neuronal activity and sensory experience, and respond by
Another proposed mechanism by which surveillant microglia modulating their behavior including their interactions with
could eliminate dendritic spines was suggested by the same dendritic spines, their phagocytic elimination of axon term-
study of Tremblay et al. (2010b). Immunocytochemical EM inals and dendritic spines and their remodeling of the perisy-
and SSEM with 3D reconstruction revealed that microglial naptic extracellular space. As resident immune cells, microglia
cell bodies, proximal and distal processes are distinctively sur- are not directly bound by the constraints imposed by the con-
rounded by pockets of extracellular space, which consists of tinuous flow of information processing in the brain, and may
interstitial fluid supplemented with various extracellular uniquely contribute to remodeling neuronal circuits in
matrix proteins (see Syková and Nicholson, 2008), in stark relation with behavioral and sensory experience in the
contrast to all other cellular and subcellular elements in the mature CNS. Together, these observations have opened
visual cortex of juvenile mice in situ. This tight association many functional possibilities beyond immune surveillance
indicates that microglial processes could dynamically influ- for these once presumed ‘resting’ cells.
ence the volume and geometry of the extracellular space, At this very early stage of investigation in this field, our
and thus the concentration of neurotransmitters and signaling knowledge about surveillant microglia almost exclusively
molecules in the vicinity of synapses. These pockets of extra- comes from observing the superficial layers of cerebral cortex
cellular space were regulated in an experience-dependent and cannot be confidently extrapolated to the entire CNS.
8 marie-e‘ ve tremblay

Since the recent studies have barely scratched the surface (of Bessis A., Bechade C., Bernard D. and Roumier A. (2007) Microglial
the brain in this case), the modalities of microglial interactions control of neuronal death and synaptic properties. Glia 55, 233–238.
with excitatory and inhibitory synapses throughout the CNS, Biber K., Neumann H., Inoue K. and Boddeke H.W.G.M. (2007)
much as their functional significance and particular cellular Neuronal ‘On’ and ‘Off’ signals control microglia. Trends in
and molecular mechanisms still remain undetermined. For Neurosciences 30, 596–602.
example, in which contexts do quiescent microglia directly
Bilbo S.D., Smith S.H. and Schwarz J.M. (2012) A lifespan approach to
phagocytose axon terminals and dendritic spines, use other
neuroinflammatory and cognitive disorders: a critical role for glia.
mechanisms such as proteolytic remodeling of the extracellular Journal of Neuroimmune Pharmacology 7, 24–41.
space, or refrain from intervening? How do surveillant micro-
glia recognize and respond to the various molecular signals in Blinzinger K. and Kreutzberg G. (1968) Displacement of synaptic term-
their environment, including dynamic changes in neurotrans- inals from regenerating motoneurons by microglial cells. Zeitschrift für
mission and neuronal activity at individual synapses? How do Zellforschung und mikroskopische Anatomie 85, 145–157.
these immune cells cooperate with other glial cells, as well as Bonfanti L. (2006) PSA-NCAM in mammalian structural plasticity and
peripheral myeloid cells, in maintaining or shaping neuronal neurogenesis. Progress in Neurobiology 80, 129–164.
architecture and activity? And, as in the case of microglial
Cardona A.E., Pioro E.P., Sasse M.E., Kostenko V., Cardona S.M.,
memory of past immune challenges (see Bilbo et al., 2012), Dijkstra I.M. et al. (2006) Control of microglial neurotoxicity by
do surveillant microglia somehow remember their previous the fractalkine receptor. Nature Neuroscience 9, 917–924.
behavioral states, the flux of information processing in the
brain, or the structural changes of synaptic elements in Castellano B., Gonzales B. and Palacios G. (1989) Cytochemical demon-
recent and not so recent windows of intervention? stration of TPPase in myelinated fibers in the central and peripheral
nervous system of the rat. Brain Research 492, 203–210.
Even though most essential questions remain unanswered
at present, the novel insights about microglia have established Chen T., Koga K., Li X.-Y. and Zhuo M. (2010) Spinal microglial motility
integrative in vivo and in situ imaging approaches as a valu- is independent of neuronal activity and plasticity in adult mice.
able methodological way, alongside genetics, electrophysi- Molecular Pain 6, 19.
ology and animal behavior, for studying these demanding Cho S.H., Sun B., Zhou Y., Kauppinen T.M., Halabisky B., Wes P. et al.
immune cells in the healthy CNS. Promising and increasingly (2011) CX3CR1 protein signaling modulates microglial activation and
integrative approaches for modulating or monitoring cellular protects against plaque-independent cognitive deficits in a mouse
activity in a non-invasive way are now emerging, with the model of Alzheimer disease. Journal of Biological Chemistry 286,
ongoing development of optogenetics and imaging of func- 32713–32722.
tional markers, sophistication of transgenic mice in the post- Choi M.S., Cho K.S., Shin S.M., Ko H.M., Kwon K.J., Shin C.Y.
genomic era and improvement in the spatiotemporal et al. (2010) ATP induced microglial cell migration through
resolution of microscopy techniques. Therefore, in addition non-transcriptional activation of matrix metalloproteinase-9.
to these innovative recent observations, we can optimistically Archives of Pharmacal Research 33, 257–265.
envision an explosion of discoveries regarding the roles of sur-
Davalos D., Grutzendler J., Yang G., Kim J.V., Zuo Y., Jung S. et al.
veillant microglia across the various CNS regions including
(2005) ATP mediates rapid microglial response to local brain injury
the cerebellum and spinal cord, stages of the lifespan and in vivo. Nature Neuroscience 8, 752–758.
states of normal brain physiology, in awake and normally
behaving animals. These discoveries and their underlying De Paola V., Holtmaat A., Knott G., Song S., Wilbrecht L., Caroni P.
technical approaches will facilitate a broader understanding et al. (2006) Cell type-specific structural plasticity of axonal branches
of microglial involvement in various neuropsychiatric and and boutons in the adult neocortex. Neuron 49, 861–875.
neurodegenerative diseases in the near future. Del Rio-Hortega P. (1919) El tercer elemento de los centros nerviosos I La
microglia en estado normal II Intervencı́on de la microglia en los pro-
cesos patológicos III Naturaleza probable de la microglia. Boletı́n de la
ACKNOWLEDGEMENTS Sociedad Española de Biologı́a 9, 69–120.

I am extremely grateful to Andreas Buchmann, Harris A. Del Rio-Hortega P. (1932) Microglia. In Penfield W. (ed.) Cytology and
Gelbard, Ania K. Majewska, Richard M. Ransohoff and Cellular Pathology of the Nervous System, New York: Hoeber. pp.
Cynthia D. Rittenhouse for their comments and suggestions. 482–534.
This work was supported by Fonds de la Recherche en Dibaj P., Nadrigny F., Steffens H., Scheller A., Hirrlinger J.,
Santé du Québec (FRSQ) and Canadian Institutes of Health Schomburg E.D. et al. (2010) NO mediates microglial response to
Research (CIHR) postdoctoral fellowships. acute spinal cord injury under ATP control in vivo. Glia 58, 1133–1144.

Donnelly D.J., Longbrake E.E., Shawler T.M., Kigerl K.A., Lai W.,
Tovar C.A. et al. (2011) Deficient CX3CR1 signaling promotes recov-
Statement of interest ery after mouse spinal cord injury by limiting the recruitment and acti-
vation of Ly6Clo/iNOS+ macrophages. Journal of Neuroscience 31,
None. 9910–9922.

Eom T.-Y., Stanco A., Weimer J., Stabingas K., Sibrack E., Gukassyan
REFERENCES V. et al. (2011) Direct visualization of microtubules using a genetic
tool to analyse radial progenitor-astrocyte continuum in brain.
Alvarez A.V. and Sabatini B.L. (2007) Anatomical and physiological Nature Communications 2, 446.
plasticity of dendritic spines. Annual Review of Neuroscience 30, 79–97.
Feng G., Mellor R.H., Bernstein M., Keller-Peck C., Nguyen Q.T.,
Bergles D.E., Jabs R. and Steinhäuser C. (2010) Neuron–glia synapses in Wallace M. et al. (2000) Imaging neuronal subsets in transgenic
the brain. Brain Research Reviews 63, 130–137. mice expressing multiple spectral variants of GFP. Neuron 28, 41–51.
 role of microglia at synapses in the healthy cns 9

Fields R.D. and Burnstock G. (2006) Purinergic signalling in neuron–glia Jung S., Aliberti J., Graemmel P., Sunshine M.J., Kreutzberg G.W.,
interactions. Nature Reviews. Neuroscience 7, 423–436. Sher A. et al. (2000) Analysis of fractalkine receptor CX(3)CR1 func-
tion by targeted deletion and green fluorescent protein reporter gene
Fontainhas A.M., Wang M.H., Liang K.J., Chen S., Mettu P.,
insertion. Molecular and Cellular Biology 20, 4106–4114.
Damani M. et al. (2011) Microglial morphology and dynamic
behavior is regulated by ionotropic glutamatergic and GABAergic Kang J., Kang N., Lovatt D., Torres A., Zhao Z., Lin J. et al. (2008)
neurotransmission. Plos One 6,14. Connexin 43 hemichannels are permeable to ATP. Journal of
Neuroscience 28, 4702–4711.
Fortin D.A., Srivastava T. and Soderling T.R. (2011) Structural modu-
lation of dendritic spines during synaptic plasticity. The Neuroscientist. Keck T., Mrsic-Flogel T.D., Vaz Afonso M., Eysel U.T., Bonhoeffer T.
and Hübener M. (2008) Massive restructuring of neuronal circuits
Fuhrmann M., Bittner T., Jung C.K., Burgold S., Page R.M., during functional reorganization of adult visual cortex. Nature
Mitteregger G. et al. (2010) Microglial Cx3cr1 knockout prevents Neuroscience 11, 1162–1167.
neuron loss in a mouse model of Alzheimer’s disease. Nature
Neuroscience 13, 411–413. Kettenmann H., Hanisch U.K., Noda M. and Verkhratsky A. (2011)
Physiology of microglia. Physiological Reviews 91, 461–553.
Graeber M.B. (2010) Changing face of microglia. Science 330, 783–788.
Knott G.W., Holtmaat A., Wilbrecht L., Welker E. and Svoboda K.
Halassa M.M. and Haydon P.G. (2010) Astrocytic networks modulate (2006) Spine growth precedes synapse formation in the adult neo-
neuronal activity and behavior. Annual Review of Physiology 72, cortex in vivo. Nature Neuroscience 9, 1117–1124.
335–355.
Kreutzberg G.W. (1996) Microglia: a sensor for pathological events in the
Hanisch U.K. and Kettenmann H. (2007) Microglia: active sensor and CNS. Trends in Neurosciences 19, 312–318.
versatile effector cells in the normal and pathologic brain. Nature
Neuroscience 10, 1387–1394. Kumamoto T., Nakagawa S., Yata Y. and Shimizu E. (1976) On the
histochemistry of enuronal and glial TPPase. Histochemistry 47, 101–109.
Harrison J.K., Jiang Y., Chen S., Xia Y., Maciejewski D., McNamara
R.K. et al. (1998) Role for neuronally derived fractalkine in mediating Le Roy C. and Wrana J.L. (2005) Clathrin- and non-clathrin-mediated
interactions between neurons and CX3CR1-expressing microglia. endocytic regulation of cell signalling. Nature Reviews. Molecular
Proceedings of the National Academy of Sciences of the U.S.A. 95, Cell Biology 6, 112–126.
10896–10901. Liang K.J., Lee J.E., Wang Y.D., Ma W., Fontainhas A.M., Fariss R.N.
Haynes S.E., Hollopter G., Yang G., Kurpius D., Dailey M.E., Gan et al. (2009) Regulation of dynamic behavior of retinal microglia by
W.-B. et al. (2006) The P2Y12 receptor regulates microglial activation CX3CR1 signaling. Investigative Ophthalmology & Visual Science 50,
by extracellular nucleotides. Nature Neuroscience 9, 1512–1519. 4444–4451.

Herndon R.M. (1964) The fine structure of the rat cerebellum. II. The Liu D., Wang Z., Liu S., Wang F., Zhao S. and Hao A. (2011)
stellate neurons, granule cells, and glia. Journal of Cell Biology 23, Anti-inflammatory effects of fluoxetine in lipopolysaccharide(LPS)-
277–293. stimulated microglial cells. Neuropharmacology 61, 592–599.

Hirasawa T., Ohsawa K., Imai Y., Ondo Y., Akazawa C., Uchino S. et al. Liu G.J., Nagarajah R., Banati R.B. and Bennett M.R. (2009) Glutamate
(2005) Visualization of microglia in living tissues using Iba1-EGFP induces directed chemotaxis of microglia. European Journal of
transgenic mice. Journal of Neuroscience Research 81, 357–362. Neuroscience 29, 1108–1118.

Hirrlinger P.G., Scheller A., Braun C., Quintela-Schneider M., Fuss B., MacVicar B.A. and Thompson R.J. (2010) Non-junction functions of
Hirrlinger J. et al. (2005) Expression of reef coral fluorescent proteins pannexin-1 channels. Trends in Neurosciences 33, 93–102.
in the central nervous system of transgenic mice. Molecular and Majewska A. and Sur M. (2003) Motility of dendritic spines in visual
Cellular Neuroscience 30, 291–303. cortex in vivo: changes during the critical period and effects of
Holtmaat A., Wilbrecht L., Knott G.W., Welker E. and Svoboda K. visual deprivation. Proceedings of the National Academy of Sciences
(2006) Experience-dependent and cell-type-specific spine growth in of the U.S.A. 100, 16024–16029.
the neocortex. Nature 441, 979–983. Majewska A.K., Newton J.R. and Sur M. (2006) Remodeling of synaptic
Holtmaat A. and Svoboda K. (2009) Experience-dependent structural structure in sensory cortical areas in vivo. Journal of Neuroscience 26,
synaptic plasticity in the mammalian brain. Nature Reviews. 3021–3029.
Neuroscience 10, 647–658. Marker D.F., Tremblay M.E., Lu S.M., Majewska A.K. and Gelbard
H.A. (2010) A thin- skull window technique for chronic two-photon
Honda S., Sasaki Y., Ohsawa K., Imai Y., Nakamura Y., Inoue K. et al.
in vivo imaging of murine microglia in models of neuroinflammation.
(2001) Extracellular ATP or ADP induce chemotaxis of cultured
Journal of Visualized Experiments 43, e2059.
microglia through Gi/o-coupled P2Y receptors. Journal of
Neuroscience 21, 1975–1982. Mataga N., Mizuguchi Y. and Hensch T.K. (2004) Experience-dependent
pruning of dendritic spines in visual cortex by tissue plasminogen acti-
Hooks B.M. and Chen C. (2007) Critical periods in the visual system:
vator. Neuron 44, 1031–1041.
changing views for a model of experience-dependent plasticity.
Neuron 56, 312–326. Mori S. and Leblond C.P. (1969) Indentification of microglia in light and
electron microscopy. Journal of Comparative Neurology 135, 57–80.
Hughes E.G., Fukaya M., Kang S.H. and Bergles D.E. (2011)In vivo
dynamics of NG2+ glial cells in the adult brain. Abstract Viewer/ Murabe Y. and Sano Y. (1982) Morphological studies on neuroglia. V.
Itinerary Planner. Society for Neuroscience, Program No. 548.12. Microglial cells in the cerebral cortex of the rat, with special reference
to their possible involvement in synaptic function. Cell and Tissue
Inoue K., Koizumi S. and Tsuda M. (2007) The role of nucleotides in the
Research 223, 493–506.
neuron–glia communication responsible for the brain functions.
Journal of Neurochemistry 102, 1447–1458. Nakanishi H. (2003) Microglial functions and proteases. Molecular
Neurobiology 27, 163–176.
Ito D., Imai Y., Ohsawa K., Nakajima K., Fukuuchi Y. and Kohsaka S.
(1998) Microglia-specific localisation of a novel calcium binding Napoli I. and Neumann H. (2009) Microglial clearance function in health
protein, Iba1. Molecular Brain Research 57, 1–9. and disease. Neuroscience 158, 1030–1038.
10 marie-e‘ ve tremblay

Neumann H., Kotter M.R. and Franklin R.J.M. (2009) Debris clearance Trapp B.D., Wujek J.R., Criste G.A., Jalabi W., Yin X., Kidd G.J. et al.
by microglia: an essential link between degeneration and regeneration. (2007) Evidence for synaptic stripping by cortical microglia. Glia 55,
Brain 132, 288–295. 360–368.
Nimmerjahn A., Kirchhoff F. and Helmchen F. (2005) Resting micro- Tremblay M.E., Riad M. and Majewska A. (2010a) Preparation of mouse
glial cells are highly dynamic surveillants of brain parenchyma in brain tissue for immunoelectron microscopy. Journal of Visualized
vivo. Science 308, 1314–1318. Experiments 41, e2021.
Oray S., Majewska A. and Sur M. (2004) Dendritic spine dynamics are Tremblay M.E., Lowery R.L. and Majewska A.K. (2010b) Microglial
regulated by monocular deprivation and extracellular matrix degra- interactions with synapses are modulated by visual experience. PloS
dation. Neuron 33, 1021–1030. Biology 8, e1000527.
Orr A.G., Orr A.L., Li X.-J., Gross R.E. and Traynelis S.F. (2009) Tremblay M.E., Stevens B., Sierra A., Wake H., Bessis A. and
Adenosine A2A receptor mediates microglial process retraction. Nimmerjahn A. (2011) The role of microglia in the healthy brain.
Nature Neuroscience 12, 872–880. Journal of Neuroscience 31, 16064–16069.
Paolicelli R.C., Bolasco G., Pagani F., Maggi L., Scianni M., Panzanelli Tremblay M.E., Zettel M.L., Ison J.R., Allen P.D. and Majewska A.K.
P. et al. (2011) Synaptic pruning by microglia is necessary for normal (2012) Effects of aging and sensory loss on glial cells in visual and audi-
brain development. Science 333, 1456–1458. tory cortices. Glia 60, 541–558.
Perry V.H. and O’Connor V. (2010) The role of microglia in synaptic Ventura R. and Harris K.M. (1999) Three-dimensional relationships
stripping and synaptic degeneration: a revised perspective. ASN between hipocampal synapses and astrocytes. Journal of
Neuro 2, e00047. Neuroscience 19, 6897–6906.
Pizzorusso T., Medini P., Berardi N., Chierzi S., Fawcett J.W. and Wake H., Moorhouse A.J., Jinno S., Kohsaka S. and Nabekura J. (2009)
Maffei L. (2002) Reactivation of ocular dominance plasticity in the Resting microglia directly monitor the functional state of synapses in
adult visual cortex. Science 298, 1248–1251. vivo and determine the fate of ischemic terminals. Journal of
Neuroscience 29, 3974–3980.
Pocock J.M. and Kettenmann H. (2007) Neurotransmitter receptors on
microglia. Trends in Neurosciences 30, 527–535. Wang Y. and Neumann H. (2010) Alleviation of neurotoxicity by micro-
Pont-Lezica L., Béchade C., Belarif-Cantaut Y., Pascual O. and Bessis glial human Siglec-11. Journal of Neuroscience 30, 3482–3488.
A. (2011) Physiological roles of microglia during development. Willig K.I., Kellner R.R., Medda R., Hein B., Jakobs S. and Hell S.W.
Journal of Neurochemistry 119, 901–908. (2006) Nanoscale resolution in GFP-based microscopy. Nature
Prinz M., Priller J., Sisodia S.S. and Ransohoff R.M. (2011) Methods 3, 721–723.
Heterogeneity of CNS myeloid cells and their roles in neurodegenera- Wong W.T., Wang M. and Li W. (2011) Regulation of microglia by iono-
tion. Nature Neuroscience 14, 1227–1235. tropic glutamatergic and GABAergic neurotransmission. Neuron Glia
Ransohoff R.M. and Perry V.H. (2009) Microglial physiology: unique Biology 14, 1–6.
stimuli, secialized responses. Annual Review of Immunology 27, 119– Wu L.-J. and Zhuo M. (2008) Resting microglial motility is independent
145. of synaptic plasticity in mammalian brain. Journal of Neurophysiology
Rivest S. (2009) Regulation of innate immune responses in the brain. 99, 2026–2032.
Nature Reviews. Immunology 9, 429–439. Yang P., Yin X. and Rutishauser U. (1992) Intercellular space is affected
Schafer D.P., Lehrman E.K., Kautzman A.G., Koyama R., Mardinly by the polysialic acid content of NCAM. Journal of Cell Biology 116,
A.R., Greenberg M.E. et al. (2011)Microglia shape neural circuits in 1487–1496.
the developing brain. Abstract Viewer/Itinerary Planner. Society for Yang G., Pan F., Parkhurst C.N., Grutzendler J. and Gan W.B. (2010)
Neuroscience, Program No. 663.03. Thinned-skull cranial window technique for long-term imaging of the
Schmechel D.E. (1999) Assessment of ultrastructural changes associated cortex in living mice. Nature Protocols 5, 201–208.
with apoptosis. In Hannun Y.A. and Boustany R.-M. (eds.) Zuo Y., Lin A., Chang P. and Gan W.B. (2005) Development of long-
Apoptosis in Neurobiology, Boca Raton: CRC Press. pp. 161–163. term dendritic spine stability in diverse regions of cerebral cortex.
Sierra A., Encinas J.M., Deudero J.J.P., Chancey J.H., Enikolopov G., Neuron 46, 181–189.
Overstreet-Wadiche L.S. et al. (2010) Microglia shape adult hippo-
campal neurogenesis through apoptosis-coupled phagocytosis. Cell
Stem Cell 7, 483–495.
Silverman W., Locovei S. and Dahl G. (2008) Probenecid, a gout remedy, AUTHOR’S ADDRESS
inhibits pannexin 1 channels. American Journal of Physiology. Cell
Physiology 295, C761–C767.
Department of Psychiatry, University of Wisconsin-Madison,
Stevens B., Allen N.J., Vazquez L.E., Howell G.R., Christopherson K.S., Madison, WI 53719, USA
Nouri N. et al. (2007) The classical complement cascade mediates
CNS synapse elimination. Cell 131, 1164–1178.
Correspondence should be addressed to:
Syková E. and Nicholson C. (2008) Diffusion in brain extracellular space. Marie-Ève Tremblay
Physiological Reviews 88, 1277–1340.
Department of Psychiatry
Theodosis D.T., Poulain D.A. and Oliet S.H.R. (2008) University of Wisconsin-Madison
Activity-dependent structural and functional plasticity of astrocyte- 6001 Research Park Boulevard
neuron interactions. Physiological Reviews 88, 983–1008. Madison, WI 53719
Trachtenberg J.T., Chen B.E., Knott G.W., Feng G., Sanes J.R., Welker E. USA
et al. (2002) Long-term in vivo imaging of experience-dependent phone: 585-474-8300
synaptic plasticity in adult cortex. Nature 420, 788–794. email: [email protected]