Experimental Neurology 196 (2005) 290 – 297
www.elsevier.com/locate/yexnr
Regular Article
Predominant phagocytic activity of resident microglia over hematogenous
macrophages following transient focal cerebral ischemia: An investigation
using green fluorescent protein transgenic bone marrow chimeric mice
Matthias Schilling*, Michael Besselmann, Marcus Müller, Jan K. Strecker,
E. Bernd Ringelstein, Reinhard Kiefer
Dept. of Neurology, Universitätsklinikum Münster, Albert-Schweitzer-Str. 33, D-48129 Münster, Germany
Received 1 April 2005; revised 20 July 2005; accepted 9 August 2005
Available online 8 September 2005
Abstract
Activated microglia and hematogenous macrophages are known to be involved in infarct development after cerebral ischemia.
Traditionally, hematogenic macrophages are thought to be the primary cells to remove the ischemic cell debris. However, phagocytosis is a
well known property also of activated microglia. Due to a lack of discriminating cellular markers, the cellular origin of phagocytes and the
temporal course of phagocytosis by these two cell types are largely unknown. In this study, we used green fluorescent protein (GFP)
transgenic bone marrow chimeric mice and semithin serial sections after methyl methacrylate embedding of the brains to dissect in detail the
proportion of identified activated resident microglial cells and infiltrating hematogenous macrophages in phagocytosing neuronal cell debris
after 30 min of transient focal cerebral ischemia. Already at day one after reperfusion, we found a rapid decrease of neurons in the ischemic
tissue reaching minimum numbers at day seven. Resident GFP-negative microglial cells rapidly became activated at day one and started to
phagocytose neuronal material. By contrast, hematogenous macrophages incorporating neuronal cell debris were observed in the ischemic
area not earlier than on day four. Quantitative analysis showed maximum numbers of phagocytes of local origin within 2 days and of bloodborne macrophages on day four. The majority of phagocytes in the infarct area were derived from local microglia, preceding and
predominating over phagocytes of hematogenous origin. This recruitment reveals a remarkable predominance of local defense mechanisms
for tissue clearance over immune cells arriving from the blood after ischemic damage.
D 2005 Elsevier Inc. All rights reserved.
Keywords: Microglia; Macrophages; Cerebral ischemia; Phagocytosis; Green fluorescent protein; Bone marrow transplantation
Introduction
After cerebral ischemia, complex pathophysiological
events are involved in infarct development over time and
space. Excitotoxicity, periinfarct depolarization, inflammation and programmed cell death are the predominant
mechanisms regulating neuronal survival and development
of tissue damage and final infarct size after vessel occlusion
(Dirnagl et al., 1999). The inflammatory response after focal
cerebral ischemia is characterized by an extremely rapid
* Corresponding author. Fax: +49 251 83 45090.
E-mail address: [email protected] (M. Schilling).
0014-4886/$ - see front matter D 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.expneurol.2005.08.004
activation of microglia/macrophages within hours (Morioka
et al., 1993; Kato et al., 1996; Lehrmann et al., 1997; Zhang
et al., 1994). Resident microglial cells and infiltrating
hematogenous macrophages play an important role during
the pathogenetic cascade following cerebral ischemia since
they express a plethora of growth factors, chemokines and
regulatory cytokines as well as free radicals and other
toxic mediators (Raivich et al., 1999) which are involved
in secondary infarct expansion (Del Zoppo et al., 2000;
Hallenbeck, 2002). Further, microglial cells are essential
as scavenger cells in tissue repair and are of functional
importance since insufficient removal of cell debris has been
identified as one of the major causes for regeneration failure
(Stoll et al., 2004). Phagocytosis is stimulated by specific
M. Schilling et al. / Experimental Neurology 196 (2005) 290 – 297
epitopes on phagocytic targets and requires activation of
downstream signaling cascades that lead to the rearrangement of the actin cytoskeleton and incorporation of the cell
debris (Koenigsknecht and Landreth, 2004). A large
number of actin-regulatory proteins are responsible for
the formation of multiform actin assemblies and macrophages have been reported to contain various actin-binding
proteins (Ohsawa et al., 2004). However, little is known
about factors that regulate microglial phagocytosis (Mitrasinovic et al., 2003). Traditionally, hematogenic macrophages are considered to be responsible for phagocytosis
(Perry et al., 1987). But once activated, resident microglia
and hematogenous macrophages are not distinguishable by
morphological criteria due to a lack of discriminating
cellular markers (Kreutzberg, 1996). Using GFP transgenic
bone marrow chimeric mice, we were enabled to conduct a
definitive distinction between these two cell types. Recently,
we found a rapid activation of resident microglial cells and a
remarkable delay of infiltration and unexpected small
number of hematogenous macrophages after transient focal
cerebral ischemia (Schilling et al., 2003). These findings
evoke the hypothesis that microglial cells and infiltrating
macrophages act differently in phagocytosis after ischemic
stroke. In order to answer this question and to quantify the
proportion of phagocytosis performed by hematogenous
macrophages or activated microglial cells, we examined
GFP transgenic bone marrow chimeric mice after transient
focal cerebral ischemia using the methyl methacrylate
embedding technique for immunohistochemical analysis of
multiple antigens in semithin serial sections (Mueller et al.,
2000).
Materials and methods
Production of bone marrow chimeric mice
The animal experiments were approved by the local
governmental authorities. Male C57BL/6J-mice (20 –30 g)
were obtained from Charles-River (Sulzfeld, Germany).
GFP-transgenic mice (C57BL/6J-GFP) were generously
donated by Dr. Masaru Okabe, Osaka, Japan (Okabe et
al., 1997). Bone marrow chimeric mice were created as
described previously (Mueller et al., 2001). In brief, 6 –8
weeks old male C57BL/6J-recipients (20 – 30 g) were
sublethally irradiated with 7 Gy in a cobalt source. Male
donor animals were killed under deep ether anesthesia by
cervical dislocation. Bone marrow was obtained by flushing
the femur bones with sterile phosphate buffered saline. Bone
marrow cells were suspended in the same buffer, washed
several times, counted and resuspended at 4 107 cells/ml.
The suspension (300 Al) was transplanted into each
irradiated recipient animal through injection into the tail
vein. After 3 months of recovery, all further experiments
were done. Chimerism was controlled by counting the
number of GFP-positive and GFP-negative leukocytes in
291
blood smears taken from each animal. Only animals with
greater than 90% GFP-positive leukocytes were used for
further experiments.
Transient focal cerebral ischemia
Transient focal cerebral ischemia was induced by
occlusion of the left middle cerebral artery (MCAO)
applying a modified intraluminal filament technique (Hata
et al., 1998) under inhalation anesthesia. To avoid large
infarcts of the entire MCA-territory and a consecutive
increase of postoperative mortality, a transient occlusion of
30 min was chosen. In brief, an 8 –0 nylon monofilament
(Ethilon; Ethicon, Norderstedt, Germany) coated with
silicon resin (Xantopren; Heraeus, Dormagen, Germany)
was introduced through a small incision into the left
common carotid artery and advanced approximately 9 mm
distal to the carotid bifurcation for temporary occlusion of
the MCA. Cerebral blood flow was continuously monitored
using a laser Doppler probe (Periflux 5001; Perimed,
Stockholm, Sweden) to verify ischemia and reperfusion.
Sham operation was performed by insertion of the filament
into the internal carotid artery with its tip proximal to the
carotid canal.
Tissue preparation
Eighteen bone marrow chimeric mice (n = 3 per group)
underwent transient focal cerebral ischemia. Survival times
were 1, 2, 4, 7, 10 and 14 days. Three chimeric mice 3
months after bone marrow transplantation (day 0) and
sham-operated chimeric mice (n = 2 per group at 2, 4 and 7
days) served as controls. The animals were perfused
through the left ventricle for 1 min with a 6% hydroxyethyl-starch solution (HES steril; Fresenius, Bad Homburg, Germany) followed by 4% buffered paraformaldehyde
(PFA) at pH 7.4 for 10 min under deep ether anesthesia.
Brains were rapidly removed and postfixed in 4% buffered
PFA for a further 3 h.
Methyl methacrylate embedding
Brain tissue from ischemic and sham-operated mice as
well as control mice from day 0 was embedded in methyl
methacrylate as described previously (Mueller et al., 2000).
Briefly, tissue was dehydrated in pure acetone for 24 h at
20-C. After dehydration, the tissue was placed in solution
MMA1 consisting of 6 ml of MMA (Sigma, Deisenhofen,
Germany), 3.5 ml of butyl-methacrylate (Sigma, Deisenhofen, Germany), 500 Al of methyl-benzoate (Merck, Darmstadt, Germany) and 120 Al of polyethyleneglycol 400
(Sigma, Deisenhofen, Germany) for 8 h. The tissue was then
incubated for 8 h in solution MMA2 which is MMA1
containing an additional amount of 800 mg/100 ml dry
benzoylperoxide (Sigma, Deisenhofen, Germany). Polymerization was allowed for 48 h at 20-C under vacuum in
292
M. Schilling et al. / Experimental Neurology 196 (2005) 290 – 297
solution MMA3 containing the mentioned substances above
plus 600 Al/100 ml N,N-dimethyl-p-toluidine (Merck,
Darmstadt, Germany). The blocks were cut on a ReichertJung ultracut ultramicrotome. Series of semithin sections
were transferred onto coated glass slides and dried at 35-C
for 2 h.
Immunohistochemistry
Coronal sections were deplasticized by incubation in
pure acetone for 3 20 min, pretreated with citrate buffer in
a microwave oven for 15 min and nonspecific protein
bindings were blocked by incubation for 15 min in Blocking
Reagent (Roche Diagnostics, Mannheim, Germany). Afterwards, sections were incubated with the following primary
antibodies: NeuN, diluted 1:50 (Chemicon International,
Temecula, CA, USA), recognizing vertebrate neuron-specific nuclear protein, or Iba-1, diluted 1:200 (generously
donated by Dr. Yoshinori Imai, Tokyo, Japan), directed
against murine microglia and macrophages at 4-C overnight. Secondary antibodies were applied for 45 min at
room temperature. To detect anti-Neu-N antibody, a
biotinylated goat anti-mouse antibody was used at 1:100
(Dianova, Hamburg, Germany). To detect Iba-1-antibody,
we used a biotinylated goat anti-rabbit antibody, applied at a
dilution of 1:100 (Vector, Burlingame, CA, USA). For
amplification of Iba-1 signals, the sections were pretreated
with 30% H2O2 for 10 min to block endogenous peroxidases, and subsequently incubated with horseradish peroxidase/streptavidin (DAKO, Glostrup, Denmark), diluted
1:100 for 45 min, and biotinylated tyramide, diluted 1:100
for 10 min at room temperature. Antibodies were visualized
by a conjugate of streptavidin and a fluorescent dye (Alexa
Fluori 594, Molecular Probes, Leiden, The Netherlands),
diluted 1:100. Nuclear counterstaining was done using a
fluorescence-preserving mounting medium containing 4V,6diamidino-2-phenylindole (DAPI) (Vector, Burlingame, CA,
USA).
Results
The evolution of ischemic damage after transient MCAO
for 30 min reproducibly followed a profile of unilateral
infarction within the lateral caudate putamen with little
variation between different animals. Cerebral blood flow
measurements confirmed a drop of cerebral blood flow
below 15% after placement of the intraarterial thread,
followed by adequate reperfusion following thread withdrawal. As mentioned above, only chimeric mice with
nearly complete chimerism were used containing a minimum of 90% GFP-positive leukocytes in their blood smears.
Consistent with previous results (Schilling et al., 2003),
white blood counts were similar in chimeric animals and
nonchimeric controls. Furthermore, neither the development
of ischemic changes and final infarct size nor the evolution
of the microglia/macrophage and astrocytic responses or
granulocyte infiltration (data not shown) was different
between chimeric animals or nonchimeric controls.
Neuronal damage
In the uninjured brain on day 0 as well as in shamoperated animals, we found normal numbers of neurons in
the cortex (Fig. 1A) and the caudate putamen of the
ipsilateral hemisphere (756 neurons/mm2 on day 0). Already
1 day after MCAO, an obvious reduction of cells in the
ischemic area could be seen, evoked by a marked decrease
of neurons to less than one half compared to day 0.
Assessing neuronal damage morphologically and further by
the loss of NeuN-immunostaining, already on day one,
many NeuN-positive cells had a shrunken cytoplasma and
an irregular nucleus (Fig. 1B). In the following days, the
number of neurons in the ischemic caudate putamen
decreased continuously (Fig. 2). At day seven, we found
only 6% neurons compared to day 0. At 2 weeks after
MCAO, only solitary surviving neurons could be detected in
the infarct area. In the boundary zone of the infarct and the
adherent neocortex, the neurons had a normal morphology.
Image analysis
Microglial activation
The sections were examined using a fluorescence microscope (Leica DM microscope, Bensheim, Germany) with
appropriate filter sets for Alexa Fluor 594, DAPI or GFP.
Sections were digitized with a Diagnostic Instruments SPOT
II Advanced Camera System (Visitron, München, Germany). Analysis of microglia/macrophages and neuronal
material incorporating phagocytes was possible by colocalization of immunofluorescent signals from NeuN- and
Iba-1-antibodies or GFP in adjacent sections. In each
section, neurons, microglial cells and hematogenous macrophages were counted in the infarct area in 5 random
nonoverlapping fields covering almost the entire lateral
caudate putamen. Data are presented as mean values T SD
(n = 3 animals per group except n = 6 for sham-operated
animals). Student’s t test was used for statistical analyses.
In the striatum of animals without cerebral ischemia and
in sham-operated mice, we did not find any changes in
number (65 microglial cells/mm2 on day 0) and morphology
of microglial cells (Fig. 3A). There were no signs of
microglial activation. At day one after MCAO, the earliest
postischemic time point investigated, all Iba-1-positive cells
were GFP-negative, indicating that no hematogenous
macrophages entered the ischemic area. The number of
microglial cells did not change significantly at day one, but
they lost their thin ramification towards retracted processes
and developed a more ameboid and rounded cell body (Fig.
3B). These morphological changes were intensified at day
two and strongly activated Iba-1-positive cells persisted
until 2 weeks after cerebral ischemia. Quantitative studies
M. Schilling et al. / Experimental Neurology 196 (2005) 290 – 297
293
Fig. 1. (A) Neurons in the cerebral cortex in the normal brain without cerebral ischemia (day 0). The inlay shows a magnification of intact neurons with typical
morphology. The arrows mark meningeal cells. (B) At day one after 30 min of transient middle cerebral artery occlusion, neurons with shrunken cytoplasma
and irregular nucleus were observed in the caudate putamen (arrows). Immunohistochemistry for NeuN (red) on MMA-embedded semithin sections, nuclear
counterstain with DAPI (blue). Scale bar = 70 Am (inlay = 20 Am) in panel A and 20 Am in panel B.
(Fig. 2) revealed an increased number of microglial cells
(Iba-1-positive and GFP-negative) at day 2 with a nearly 5fold increase at day 7 compared to the uninjured brain. At
day 10, we found a maximum of microglial cells in the
striatum with a 7.5-fold increase compared to the number of
microglia in the normal brain (day 0).
Macrophage infiltration
GFP-positive/Iba-1-positive hematogenous macrophages
first entered the lateral caudate putamen at day two after
reperfusion. These cells were morphologically indistinguishable from activated ameboid microglia but could be
easily identified by the green fluorescence of GFP (Figs.
3B – E). The number of hematogenous macrophages
increased to a maximum at day seven (75 cells/mm2) and
decreased until day 14. At day seven, 20% of the total
population of all Iba-1-positive cells in the ischemic area
were hematogenous macrophages. But the majority of Iba-
1-positive cells remained GFP-negative indicating their
derivation from microglia (Fig. 2).
Phagocytosis of neuronal debris by resident and infiltrating
macrophages
To analyze the phagocytic capacity of blood-borne
monocyte derived macrophages and activated local microglia, we studied the incorporation of neuronal (NeuNpositive) material into identified activated microglia (Iba-1positive/GFP-negative) and hematogenous macrophages
(Iba-1-positive/GFP-positive) by co-localizing specific signals on adjacent semithin sections and quantifying identified
phagocytic cells.
In animals without cerebral ischemia (day 0) and in
sham-operated mice, phagocytosis of neuronal debris was
not detectable. Already at day one after reperfusion, we
found the first Iba-1-positive cells which had incorporated
neuronal material (Figs. 5A – C). At day 2, the number of
Fig. 2. Quantitative assessment of neurons (black bars), resident microglial cells (grey bars) and infiltrating macrophages (white bars) over time in the infarct
area after 30 min of transient middle cerebral artery occlusion. Values are obtained from 5 ipsilateral areas per animal (n = 6 in the sham-operated group; n = 3
at day 0 and each postischemic time point; mean values T SD).
294
M. Schilling et al. / Experimental Neurology 196 (2005) 290 – 297
Fig. 3. (A) Resident microglial cell (arrowhead) with thin ramification in the striatum of a control brain (day 0). Since all stainings were performed on semithin
(0.5 Am) sections, the apparent interruptions of microglial processes (arrows) are due to their winding course leaving the plane of the section. (B and C)
Activated microglial cell in the striatum at day one after MCAO (B, arrow). The ameboid cell has retracted processes without thin ramifications. The absence of
the green fluorescence of GFP (C, arrow) identifies this cell as derived from local microglia. (D and E) Hematogenous macrophage in the striatum at day four
after MCAO (D, arrow). This cell is indistinguishable from activated microglia (3B, arrow) on morphological grounds but can easily be differentiated by the
green fluorescence of GFP (E, arrow) which identifies the hematogenous origin of this cell. Immunohistochmistry for Iba-1 (red), nuclear counterstain with
DAPI (blue), green fluorescence of GFP (green). Scale bar = 15 Am in panel A and 25 Am in panels B, C, D and E.
phagocytic cells was similar to day one (20 cells/mm2).
Quantitative analysis (Fig. 4) showed that the maximum of
phagocytosis was already reached at these early time points.
The complete absence of the GFP signal in Iba-1-positive
phagocytic cells at these time points indicated that
phagocytosis was exclusively performed by phagocytes
derived from resident microglial cells. At day four, we
observed a decrease of microglial phagocytes to nearly one
half. At this day, we found for the first time Iba-1-positive,
GFP-positive cells which had incorporated NeuN-positive
material indicating that these phagocytes were of hematogenous origin (Figs. 5D – F).
The total number of phagocytic microglial cells stayed
nearly at the same level from day four until day 10.
Hematogenous macrophages infiltrated into the ischemic
area had their maximum at day four, decreased at day seven
and 10 and were no longer found 2 weeks after reperfusion.
At day four, 15% of all phagocytes were of hematogenous
origin. At all time points, the number of phagocytes within
the total population of resident microglia or hematogenous
macrophages was very small. At day one and day two,
where we found peak numbers of phagocyting microglia,
these cells amount to only one quarter of the total number of
microglial cells at these time points.
Discussion
Traditionally, hematogenic macrophages are thought to
be the primary cell type for phagocytosing of cellular debris
Fig. 4. Quantitative analysis of phagocytic microglia (black bars) and phagocytic hematogenous macrophages (white bars) over time in the infarct area after 30
min of transient middle cerebral artery occlusion. In sham-operated animals and day 0, we did not find any phagocytes. Values are obtained from analysis of 5
ipsilateral areas per animal (n = 6 in the sham-operated group; n = 3 at day 0 and each postischemic timepoint; mean values T SD). Asterisks denote statistical
significance (at least P < 0.05; Student’s t test).
M. Schilling et al. / Experimental Neurology 196 (2005) 290 – 297
295
Fig. 5. Panels A – C and D – F are serial semithin sections of the same areas shown by identical patterns of nuclei (Nuclear counterstain with DAPI, blue). The
inlays are magnifications of the marked cells. (A, arrow) Iba-1-positive activated microglia/macrophage at day one after transient middle cerebral artery
occlusion. The absence of the green fluorescence of GFP (B, arrow) identifies the resident origin of this cell and the incorporation of NeuN-positive neuronal
material (C, arrow) shows the phagocytic property. The marked Iba-1-positive microglia/macrophage (D, arrow) at day four after transient MCAO could be
easily identified as a hematogenous macrophage by the GFP-positive signal (E, arrow). Phagocytosis is demonstrated by the incorporation of NeuN-positive
neuronal material (F, arrow). Immunohistochemistry for Iba-1 (A and D, red), green fluorescence of GFP (B and E, green), immunohistochemistry for NeuN (C
and F, red), nuclear counterstain with DAPI (blue). Scale bar = 20 Am (inlays = 10 Am) in panels A – F.
after cerebral ischemia and to promote scar formation
(Mabuchi et al., 2000; Ito et al., 2001). In this study, we
were able to dissect in detail the proportion of identified
activated resident microglial cells phagocytosing neuronal
cell debris after transient focal cerebral ischemia on the one
hand and infiltrating hematogenous macrophages on the
other. Until now, no quantitative data were available and no
distinction was made between resident microglia and
infiltrating hematogenous macrophages during phagocytosis
following cerebral ischemia.
In parallel with the rapid activation of resident microglia,
we found peak phagocytic activity in the first 2 days after
reperfusion. There was no participation of hematogenous
macrophages at these early time points. Peak phagocytic
activity of infiltrating macrophages incorporating NeuNpositive material was identified at day four after cerebral
ischemia but the absolute number of hematogenous phagocytes was surprisingly low.
According to previous experiments (Schilling et al.,
2003) and consistent with several other investigators (Kato
et al., 1996; Lehrmann et al., 1997; Zhang et al., 1994), we
found an extremely rapid activation of local microglial cells
within the infarct zone which was evident already at day one
after disease onset, the earliest time point investigated in our
experiments. Schroeter et al. (1997) summarized in a
macrophage-depletion model that the initial macrophage
response after photochemically induced ischemia in the
periinfarct region was of microglial origin while hematogenous macrophages were recruited with remarkable delay.
Furthermore, Kleinschnitz et al. (2003) suggested from
macrophage infiltration experiments by magnetic resonance
imaging that major macrophage infiltration is delayed and
occurs at a stage in which neuronal cell death is mostly
completed. It was speculated that infiltrating macrophages
play a role in tissue remodeling rather than in neuronal
injury. This assumption matches our observation that the
majority of neuronal damage, conspicuous by a continuos
decrease of neurons in the ischemic striatum, occurs before
hematogenous macrophages infiltrate the infarcted brain,
and that hematogenous macrophages participate in phagocytosis only in a small amount compared to resident
activated microglia.
Interestingly, the kinetics of microglial/macrophage
numbers and the phagocytic activity of both types of
macrophages are only partially concordant. Activated
microglia increased during the first 10 days after cerebral
ischemia while the maximum number of phagocytes from
local origin was reached already within 2 days after vessel
occlusion. Similarly, peak phagocytic activity by hematogenous macrophages precedes the time point of their highest
numbers by 3 days (day four vs. day seven). Tanaka et al.
(2003) also found hematogenous macrophages, morphologically characterized as phagocytes, infiltrating predominantly the ischemic core area at day seven in a similar GFPbone marrow chimeric mouse model of permanent cerebral
ischemia. Surprisingly, the number of activated microglial
cells in the ischemic core or transition area was very low. In
this study, phagocytosis was not investigated and no
information was given about a distinction between phagocytes of microglial or hematogenous origin. Based on the
low numbers of activated microglia in the transition area and
the ischemic core in this study of permanent cerebral
ischemia, it might be speculated that the duration of
ischemia might have a major impact on the kinetics of
296
M. Schilling et al. / Experimental Neurology 196 (2005) 290 – 297
phagocytic activity in transient cerebral ischemia of longer
duration than under our own experimental conditions and
may result in a reduction of phagocytosis by resident
microglia and hematogenous macrophages.
In our experiments, we focused on parenchymal microglia and hematogenous macrophages invading into the
injured tissue. Furthermore, a third population of macrophages exists in the brain (Graeber et al., 1989). Perivascular macrophages are able to quickly phagocytose particles
from the cerebrospinal fluid, form a population which
undergo rapid turnover with hematogenous macrophages
and might be in a position to significantly contribute to the
central nervous system immune surveillance (Bechmann et
al., 2001; Thomas, 1999). Hess et al. (2004) showed that
hematopoietic stem cells differentiate into parenchymal
microglial cells as well as perivascular cells with increasing
number of these cells after cerebral ischemia. They noted
that an increase of parenchymal microglia might be not
surprising as monocytes and macrophages are involved in
phagocytosis of injured cells and tissue debris. Nevertheless,
the kinetics of phagocytes from local or hematogenic origin
and the low number of blood-borne phagocytes in our
experiments underline that hematopoietic cells infiltrating
the brain after cerebral ischemia might be not only involved
in phagocytosis, but may have different roles after cerebral
ischemia which are at present not completely understood.
Interestingly, Rinner et al. (1995) found a similar predominance of activated microglia over hematogenous macrophages in phagocytosis of myelin debris in experimental
autoimmune encephalomyelitis using a rat radiation bone
marrow chimeric animal model.
Using the methyl methacrylate embedding technique, we
were able to assess the incorporation of neuronal material
into single microglial cells or hematogenous macrophages at
the same tissue level (Mueller et al., 2000). The Iba-1antibody was the only marker which results in sufficient
staining of microglia/macrophages on MMA-embedded
material. Ito et al. (2001) indicated in co-localisation
experiments with other molecular markers that the Iba-1antibody is suitable in the immunohistochemical detection
and recognizes all activated microglia and macrophages
after cerebral ischemia.
Analyzing the quantitative data, only a small amount of
microglial cells as well as hematogenous macrophages
contribute to the number of cells phagocytosing neuronal
material. At day one and day two, where we found peak
numbers of phagocytic microglia, less than 25% of the total
number of microglial cells incorporate neuronal (NeuNpositive) material. At later time points, the percentage of
phagocytes decreases again. The quantitative analysis of
hematogenous macrophages revealed similar data. It might
be speculated that either the process of neuronal damage or
phagocytosis of neuronal material may disturb the immunohistochemical detection of the neuronal antigen by antiNeuN-antibody. If a disturbed detection of phagocytic
neuronal material might be relevant, it would be related to
both cell types and would not perturb the relative amount of
microglia or hematogenous macrophages participating in
phagocytosis of tissue debris. More likely, only a small
proportion of microglial cells, and hematogenous macrophages at later time points, are responsible for phagocytosis,
while others do not participate and may assume different,
possibly regulatory functions. In line with this assumption,
we previously identified few ramified, nonphagocytic cells
both of resident and hematogenous origin at all time points
following ischemia (Schilling et al., 2003). Unfortunately,
our method does not allow to answer the interesting
question of how many phagocytes are required to dispose
of what amount of damaged neurons. Nevertheless, our data
indicate that microglia are more effective in tissue clearance
after ischemic damage compared to macrophages recruited
from the circulation, and outline a different role for these
cells after cerebral ischemia.
Acknowledgments
We thank Antje Stöber and Karin Wacker for excellent
technical assistance, Dr. M. Okabe, Osaka, Japan, for his
generous gift of GFP-transgenic mice, and Dr. Y. Imai,
Tokyo, Japan, for providing us with Iba-1 antibody. This
study was supported by a grant from the Interdisciplinary
Center of Clinical Research Münster (IZKF Project No. G5).
References
Bechmann, I., Priller, J., Kovac, A., Böntert, M., Wehner, T., Klett, F.F.,
Bohsung, J., Stuschke, M., Dirnagl, U., Nitsch, R., 2001. Immune
surveillance of mouse brain perivascular spaces by blood-borne macrophages. Eur. J. Neurosci. 14, 1651 – 1658.
Del Zoppo, G., Ginis, I., Hallenbeck, J.M., 2000. Inflammation and stroke:
putative role of cytokines, adhesion molecules and iNOS in brain
response to ischemia. Brain Pathol. 10, 95 – 112.
Dirnagl, U., Iadecola, C., Moskowitz, M.A., 1999. Pathobiology of
ischemic stroke: an integrated view. Trends Neurosci. 22, 391 – 397.
Graeber, M.B., Streit, W.J., Kreutzberg, G.W., 1989. Identity of ED2positive perivascular cells in the brain. J. Neurosci. Res. 22, 103 – 106.
Hallenbeck, J.M., 2002. The many faces of tumor necrosis factor in stroke.
Nat. Med. 8, 1363 – 1368.
Hata, R., Mies, G., Wiessner, C., Fritze, K., Hesselbarth, D., Brinker, G.,
Hossmann, K.A., 1998. A reproducible model of middle cerebral artery
occlusion in mice: hemodynamic, biochemical and magnetic resonance
imaging. J. Cereb. Blood Flow Metab. 18, 367 – 375.
Hess, D.C., Abe, T., Hill, W.D., Studdard, A.M., Carothers, J., Masuya,
M., Fleming, P.A., Drake, C.J., Ogawa, M., 2004. Hematopoietic
origin of microglial and perivascular cells in brain. Exp. Neurol. 186,
134 – 144.
Ito, D., Tanaka, K., Suzuki, S., Dembo, T., Fukuuchi, Y., 2001. Enhanced
expression of IBA-1, ionized calcium-binding adaptor molecule 1, after
transient focal cerebral ischemia in rat brain. Stroke 32, 1208 – 1215.
Kato, H., Kogure, K., Liu, X.H., Araki, T., Itoyama, Y., 1996. Progressive
expression of immunomolecules on activated microglia and invading
leukocytes following focal cerebral ischemia in the rat. Brain Res. 734,
203 – 212.
Kleinschnitz, C., Bendszus, M., Frank, M., Solymosi, L., Tyoka, K.V.,
Stoll, G., 2003. In vivo monitoring of macrophage infiltration in
M. Schilling et al. / Experimental Neurology 196 (2005) 290 – 297
experimental ischemic brain lesion by magnetic resonance imaging.
J. Cereb. Blood Flow Metab. 23, 1356 – 1361.
Koenigsknecht, J., Landreth, G., 2004. Microglial phagocytosis of fibrillar
h-amyloid through a h1-integrin-dependent mechanism. J. Neurosci. 24,
9838 – 9846.
Kreutzberg, G.W., 1996. Microglia: a sensor for pathological events in the
CNS. Trends Neurosci. 19, 312 – 318.
Lehrmann, E., Christensen, T., Zimmer, J., Diemer, N.H., Finsen, B., 1997.
Microglial and macrophage reactions mark progressive changes and
define the penumbra in the rat neocortex and striatum after transient
middle cerebral artery occlusion. J. Comp. Neurol. 386, 461 – 476.
Mabuchi, T., Kitagawa, K., Ohtsuki, T., Kuwabara, K., Yagita, Y.,
Yanagihara, T., Hori, M., Matsumono, M., 2000. Contribution of
microglia/macrophages to expansion of infarction and response of
oligodendrocytes after focal cerebral ischemia in rats. Stroke 31,
1735 – 1743.
Mitrasinovic, O.M., Vincent, V.A.M., Simsek, D., Murphy, G.M., 2003.
Macrophage colony stimulating factor promotes phagocytosis by
murine microglia. Neurosci. Lett. 344, 185 – 188.
Morioka, T., Kalehua, A.N., Streit, W.J., 1993. Characterization of microglial reaction after middle cerebral artery occlusion in rat brain. J. Comp.
Neurol. 327, 123 – 132.
Mueller, M., Wacker, K., Hickey, W.A., Ringelstein, E.B., Kiefer, R., 2000.
Co-localization of multiple antigens and specific DNA—a novel
method using methyl methacrylate-embedded semithin serial sections
and catalyzed reporter deposition. Am. J. Pathol. 157, 1829 – 1838.
Mueller, M., Wacker, K., Ringelstein, E.B., Hickey, W.F., Imai, Y., Kiefer,
R., 2001. Rapid response of identified resident endoneurial macrophages to nerve injury. Am. J. Pathol. 159, 2187 – 2197.
Ohsawa, K., Imai, Y., Sasaki, Y., Kohsaka, S., 2004. Microglia/macrophage-specific protein Iba-1 binds to fimbrin and enhances its actinbundling activity. J. Neurochem. 88, 844 – 856.
Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T., Nishimune, Y., 1997.
FGreen mice_ as a source of ubiquitous green cells. FEBS Lett. 407,
313 – 319.
297
Perry, V.H., Brown, M.C., Gordon, S., 1987. The macrophage response to
central and peripheral nerve injury. J. Exp. Med. 165, 1218 – 1223.
Raivich, G., Bohatschek, M., Kloss, C.U.A., Werner, A., Jones, L.L.,
Kreutzberg, G.W., 1999. Neuroglial activation repertoire in the injured
brain: graded response, molecular mechanisms and cues to physiological function. Brain Res. Rev. 30, 77 – 105.
Rinner, W.A., Bauer, J., Schmidts, M., Lassmann, H., Hickey, W.F.,
1995. Resident microglia and hematogenous macrophages in adaptively transferred experimental autoimmune encephalomyelitis: an
investigation using rat radiation bone marrow chimeras. Glia 14,
257 – 266.
Schilling, M., Besselmann, M., Leonhard, C., Mueller, M., Ringelstein,
E.B., Kiefer, R., 2003. Microglial activation precedes and predominates
over macrophage infiltration in transient focal cerebral ischemia: a study
in green fluorescent protein transgenic bone marrow chimeric mice.
Exp. Neurol. 183, 25 – 33.
Schroeter, M., Jander, S., Huitinga, I., Witte, O., Stoll, G., 1997. Phagocytic
response in photochemically induced infarction of rat cerebral cortex:
the role of resident microglia. Stroke 28, 382 – 386.
Stoll, G., Schroeter, M., Jander, S., Siebert, H., Wollrath, A., Kleinschnitz,
C., Brück, W., 2004. Lesion-associated expression of transforming
growth factor beta-2 in the rat nervous system: evidence for downregulation the phagocytic activity of microglia and macrophages. Brain
Pathol. 14, 51 – 58.
Tanaka, R., Komine-Kobayashi, M., Mochizuki, H., Yamada, M., Furuya,
T., Migita, M., Shinada, T., Mizuno, Y., Urabe, T., 2003. Migration of
enhanced green fluorescent protein expressing bone marrow-derived
microglia/macrophage into the mouse brain following permanent focal
ischemia. Neuroscience 117, 531 – 539.
Thomas, W.E., 1999. Brain macrophages: on the role of pericytes and
perivascular cells. Brain Res. Rev. 31, 42 – 57.
Zhang, R.L., Chopp, M., Chen, H., Garcia, J.H., 1994. Temporal profile of
ischemic tissue damage, neutrophil response and vascular plugging
following permanent and transient (2 h) middle cerebral artery
occlusion in the rat. J. Neurol. Sci. 125, 3 – 10.