Meningeal and Perivascular Macrophages of
the Central Nervous System Play a Protective
Role During Bacterial Meningitis
This information is current as
of June 15, 2017.
Machteld M. J. Polfliet, Petra J. G. Zwijnenburg, A.
Marceline van Furth, Tom van der Poll, Ed A. Döpp, Chantal
Renardel de Lavalette, Esther M. L. van Kesteren-Hendrikx,
Nico van Rooijen, Christien D. Dijkstra and Timo K. van den
Berg
References
Subscription
Permissions
Email Alerts
This article cites 30 articles, 12 of which you can access for free at:
http://www.jimmunol.org/content/167/8/4644.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2001 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
J Immunol 2001; 167:4644-4650; ;
doi: 10.4049/jimmunol.167.8.4644
http://www.jimmunol.org/content/167/8/4644
Meningeal and Perivascular Macrophages of the Central
Nervous System Play a Protective Role During Bacterial
Meningitis
Machteld M. J. Polfliet,1* Petra J. G. Zwijnenburg,† A. Marceline van Furth,†
Tom van der Poll,‡ Ed A. Döpp,* Chantal Renardel de Lavalette,*
Esther M. L. van Kesteren-Hendrikx,* Nico van Rooijen,* Christien D. Dijkstra,* and
Timo K. van den Berg*
D
espite effective antibiotic treatment, bacterial meningitis
is still an infection with a high mortality rate, particularly for very young and elderly patients (1). Furthermore, recovered patients may suffer from neurological damage,
including cerebral infarction, recurrent seizures, hearing loss,
learning disabilities, and/or mental retardation (2, 3). When the
bacteria have entered the subarachnoidal space, endogenous cells
of the CNS and infiltrated leukocytes are triggered to produce
proinflammatory cytokines like TNF-␣, IL-1␤, and IL-6, which
lead to an increased permeability of the blood-brain barrier (BBB)2
(1). As a result of this enhanced permeability state, leukocytes,
predominantly granulocytes, are recruited to the CNS. Activated
granulocytes play a major role in bacterial clearance and also secrete a variety of inflammatory mediators (e.g., reactive oxygen
species) that may not only kill bacteria, but also cause damage to
the CNS (4 – 8). These mediators play a key role in the pathophysDepartments of *Molecular Cell Biology and †Pediatrics, Vrije Universiteit Medical
Centre; and ‡Laboratory of Experimental Internal Medicine, Academic Medical Centre, Amsterdam, The Netherlands
Received for publication May 21, 2001. Accepted for publication August 8, 2001.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
Address correspondence and reprint requests to Dr. Machteld M.J. Polfliet, Department of Molecular Cell Biology, Vrije Universiteit Medical Center, Vrije Universiteit,
Van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands. E-mail address:
[email protected]
2
Abbreviations used in this paper: BBB, blood-brain barrier; PVM, perivascular macrophage; MM meningeal macrophage; CSF, cerebrospinal fluid; BA-hm, blood agarhemin menadione; MIP, macrophage-inflammatory protein; PECAM, platelet endothelium cell adhesion molecule.
Copyright © 2001 by The American Association of Immunologists
iology of bacterial meningitis and contribute to the unfavorable
outcome of the disease due to intracranial complications such as
decreased cerebral blood flow, raised intracranial pressure, and
brain edema (6, 9).
The cerebrospinal fluid (CSF) lacks the presence of cell populations capable of initiating an effective immune response against
invading pathogens. Therefore, important questions are which (and
how) resident cell populations in the CNS initiate and regulate the
inflammatory response during meningitis. Resident CNS macrophages, and in particular, the meningeal (MM) and perivascular
(PVM) macrophages that line the meninges and surround local
blood vessels (10), are prime candidates. The MM and PVM can
be identified in the rat by the selective expression of the ED2 Ag
(11). They express scavenger receptors and are highly phagocytic
(12). Furthermore, under various inflammatory conditions the MM
and PVM can express inflammatory mediators such as TNF-␣,
IL-1␤, and cyclooxygenase-2 (13–15). Until now only one study
has examined the role of the MM during bacterial meningitis (16).
It was observed that reduction of the MM by intracisternal injection of clodronate liposomes did not affect inflammation in the
subarachnoidal space. Unfortunately, the liposome treatment only
resulted in a partial depletion of the MM, with no further information about: 1) the effect on the PVM and other CNS cell populations (e.g., microglial cells), 2) the effect on other parts of the
CNS like the spinal cord and cerebrum, and 3) possible side effects
on the periphery (e.g., Kupffer cells, splenic macrophages, and
monocytes).
We have recently described a depletion method that results in a
complete and selective depletion of the MM and PVM (17). Here
we use this method to study the role of the MM and PVM in an
0022-1767/01/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
Meningeal (MM) and perivascular macrophages (PVM) constitute major populations of resident macrophages in the CNS that can
be distinguished from microglial cells. So far, there is no direct evidence that demonstrates a possible role of MM and PVM in the
CNS during normal or pathologic conditions. To elucidate the role of the MM and PVM during CNS inflammation, we have
developed a strategy using a single intraventricular injection of mannosylated clodronate liposomes, which results in a complete
and selective depletion of the PVM and MM from the CNS. Depletion of the MM and PVM during experimental pneumococcal
meningitis resulted in increased illness, which correlated with higher bacteria counts in the cerebrospinal fluid and blood. This was
associated with a decreased influx of leukocytes into the cerebrospinal fluid, which occurred despite an elevated production of
relevant chemokines (e.g., macrophage-inflammatory protein-2) and a higher expression of vascular adhesion molecules (e.g.,
VCAM-1). In contrast, the higher bacterial counts correlated with elevated production of local and systemic inflammatory mediators (e.g., IL-6) indicating enhanced local leukocyte and systemic immune activation, and this may explain the worsening of the
clinical signs. These findings show that the PVM and MM play a protective role during bacterial meningitis and suggest that a
primary action of these macrophages is to facilitate the influx of leukocytes at the blood-brain barrier. More in general, we
demonstrate for the first time that the PVM and MM play a crucial role during inflammation in the CNS. The Journal of
Immunology, 2001, 167: 4644 – 4650.
The Journal of Immunology
experimental bacterial meningitis model. Our results provide the
first direct evidence for the role of MM and PVM in bacterial
meningitis in particular, and in inflammation in the CNS in
general.
Materials and Methods
Animals
Male Wistar rats were obtained from CPB-Harlan (Zeist, The Netherlands)
and weighed 300 –350 g at the time of the experiment. Animals were kept
under routine laboratory conditions and were allowed free access to food
and water. Microbiological status of the animals was according to Federation of European Laboratory Animal Science Associations recommendations. Approval of the Animal Ethics Committee of Vrije Universiteit was
obtained for all animal experiments.
Preparation of liposomes
Stereotaxical injection of liposomes
The rats were deeply anesthetized with an i.m. injection of 1.0 ml/kg body
weight of a mixture of 4 parts ketamine (1% solution of Ket; Aesco, Boxtel, The Netherlands) and 3 parts xylazine (2% solution of Rompun; Bayer,
Brussels, Belgium) and mounted in a stereotaxic frame. Control or clodronate liposomes (50 ␮l) were slowly (over a total time period of 25 min)
injected stereotaxically with a blunt Hamilton syringe (Hamilton Bonaduz,
Bonaduz, Switzerland) into the 4th ventricle. Coordinates (Bregma ⫹11.6
and 5.5 mm dorsoventral) were derived from the atlas of Paxinos and
Watson (21).
Bacterial meningitis model
Streptococcus pneumoniae type 6A (isolated from the CSF of a meningitis
patient) were incubated in brain heart infusion broth at 37°C. After 16 –17
h, 5 ml of this preincubation culture was mixed with 20 ml of fresh brain
heart infusion medium and further incubated for 3–5 h at 37°C until
OD620 ⫽ 1 was reached. Rats were mounted in a stereotaxic frame and
injected intracisternally with ⬃5 ⫻ 106 CFU of S. pneumoniae, 7 days post
clodronate liposome (n ⫽ 11) or PBS liposome (n ⫽ 7) injection. Twelve
hours post meningitis induction rats were clinically scored (0 ⫽ no clinical
symptoms, 1 ⫽ red nose and eyes, 2 ⫽ reduced activity/apathy/piloerection, 3 ⫽ convulsions/respiratory problems) and sacrificed with an i.p.
injection of 1 ml of Nembutal (Sanofi Sante Animale, Benelux B.V.,
Maassluis, The Netherlands). The inoculum (S. pneumoniae) was plated
(1 ⫻ 100–1 ⫻ 10⫺7) on sheep blood agar-hemin menadione (BA-hm)
plates and cultured overnight at 37°C to determine the amount of CFU
injected.
Sample collection and processing
Blood was collected via a heart puncture for blood smears, and 1 ⫻ 100–
1 ⫻ 10⫺4 eg of EDTA blood was plated on sheep BA-hm plates for identification and quantification of the bacteria. Serum was collected for measurement of TNF-␣, IL-1␤, IL-6 (National Institute for Biological
Standards and Control, South Mimms, Potter Bar, Hertfordshire, U.K.),
and macrophage-inflammatory protein (MIP)-2 (BioSource International,
Camarillo, CA) by ELISA. CSF was aspirated via a cisterna magna
puncture, and 1 ⫻ 100–1 ⫻ 10⫺7 was plated on BA-hm plates, for the
quantification of the bacteria. The leukocytes were counted using a hemocytometer, and cytocentrifuge preparations were made and stained with
May-Grünwald Giemsa dye (Merck, Darmstadt, Germany), according to
standard histological staining methods. CSF samples were then centrifuged, cells were stored in 10% DMSO/90% NBCS at ⫺80°C, and the
supernatant was stored at ⫺80°C for determination of MIP-2 by ELISA.
The cerebellum, cerebrum, spinal cord (cranial-caudal), liver, cervical
lymph nodes, and spleen were collected, snap-frozen, and stored at ⫺80°C
until histological evaluation.
Bacterial peritonitis model
Seven days post clodronate liposome (n ⫽ 6) and PBS liposome (n ⫽ 5)
injection into the 4th ventricle as described above, the rats received an i.p.
injection of 3 ⫻ 108 CFU/0.5 ml S. pneumoniae type 6A (culture conditions as described above). Four hours after this bacteria injection into the
peritoneum, the rats were sacrificed by O2/CO2 exposure. The peritoneal
cavity was washed with 20 ml of RPMI 1640 (4°C) medium (Life Technologies, Paisley, Scotland), cells were counted, and cytocentrifuge preparations were stained with May-Grünwald Giemsa dye (Merck), to determine the percentages of leukocytes.
Immunohistochemistry for light microscopy
Cryostat sections (8 ␮m) were cut serially, picked up on gelatin coated
slides, and air dried overnight in a container with silica gel. Immunohistochemistry was applied after a 10-min fixation in dehydrated acetone to
examine the presence of macrophages, neutrophils, and the expression of
cerebral endothelial adhesion molecules. The murine mAbs in this study
include the anti-rat macrophage markers ED1 and ED2. ED1 recognizes a
lysosomal membrane-related Ag on both monocytes and macrophages, and
ED2 recognizes an Ag on a subset of mature rat macrophages including the
MM and PVM in the CNS (22). HIS48 recognizes an Ag on all granulocytes (23). 1A29 (a gift of T. Tamatani, Department of Immunology, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan) directed
against ICAM-1 (24), 3A12 (a gift of W. F. Hickey, Dartmouth Medical
School, Lebanon, NH) for platelet endothelium cell adhesion molecule
(PECAM; Ref. 25), and 5F10 (a gift of R. Lobb, Biogen, Cambridge, MA),
which recognizes VCAM-1 (26) on rat cerebral endothelial cells. Abs and
conjugates were diluted in 0.1 M Tris buffer, pH 7.6, with 0.1% BSA
(Organon Technika, Oss, The Netherlands) and used at optimal final dilution of 2 ␮g/ml for ED1-biotin and ED2-biotin; 6 ␮g/ml for 5F10; 2.75
␮g/ml for 3A12; 1/5 of the 1A29 supernatant, and 1 ␮g/ml for His48.
Control slides included incubations in which the primary Ab was omitted.
All incubations were conducted horizontally at room temperature. After
incubation with the first Ab for 60 min, the slides were rinsed in Tris buffer,
pH 7.6, with 0.1% BSA, incubated with conjugate/0.1% normal rat serum
for 60 min, and washed again with Tris/BSA buffer. As conjugates we used
streptavidin-alkaline phosphatase-conjugated rabbit anti-mouse IgG
(DAKO, Glostrup, Denmark) and alkaline phosphatase-conjugated rabbit
anti-mouse IgG ⫹ IgM (DAKO). Alkaline phosphatase activity was demonstrated by incubation in naphthol AS-BI phosphate substrate (27) in 0.1
M Tris, pH 8.7, buffer for 10 min. Sections were rinsed in 0.1 M Tris, pH
7.6, light counterstained with hematoxylin, and mounted in VectaMount
(Vector Laboratories, Burlingame, CA). The sections were examined with
a microscope (Nikon Eclipse E800; Melville, NY), and the recordings were
made with a Nikon DXM1200 camera.
Quantification of leukocytes
The number of leukocytes were counted with a light microscope using
⫻400 magnification. Two hundred total white blood cells were counted per
CSF cytocentrifuge preparation and, subsequently, the percentage of neutrophils, eosinophils, macrophages, and lymphocytes of the total amount of
white blood cells was counted.
Results
Pneumococcal meningitis model in a rat
Before investigating the role of the MM and PVM during meningitis, we evaluated the kinetics of the S. pneumoniae meningitis
model in detail. We counted the number of bacteria and leukocytes
in the CSF and blood of the animals after 4, 12, 24, and 48 h. We
observed a decline in the number of bacteria in the CSF, 12 h post
meningitis induction, followed by an increase after 24 and 48 h
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
Multilamellar mannosylated liposomes were prepared as described before
(18). Briefly, 178 mg of phosphatidylcholine and 27 mg of cholesterol were
dissolved in 8 ml of chloroform, which was added to 9.25 mg of p-aminophenyl-␣-D-mannopyranoside (Sigma, St. Louis, MO) dissolved in 5 ml of
methanol in a 500-ml round-bottom flask. This was dried in vacuo on a
rotary evaporator to form a film. The molar ratio of phosphatidylcholine/
cholesterol/mannoside was 7:2:1 (according to Ref. 19). The lipid film was
dispersed in 10 ml of PBS (0.15 M NaCl in 10 mM phosphate buffer, pH
7.4) for the preparation of PBS-containing mannosylated liposomes. To
enclose the clodronate, 2.5 g (a gift of Roche Diagnostics, Mannheim,
Germany) of clodronate was dissolved in 10 ml of deionized water (adjusted to pH 7.3 with NaOH) in which the lipid film was dispersed, and the
preparations were kept for 2 h at room temperature, sonicated for 3 min,
washed, and resuspended in 10 ml of PBS. PBS-containing control liposomes and clodronate-containing liposomes were labeled with the fluorescent dye DiI (1,1⬘-dioctadecyl-3,3,3⬘,3⬘-tetramethyl-indocarbocyanine perchlorate; D282; Molecular Probes, Eugene, OR) as described by Claassen
(20), with minor alterations. DiI was dissolved in absolute ethanol at a
concentration of 2.5 mg/ml, sonicated for 5 min, and stored at 4°C. DiI was
added to 1 ml of liposome suspension in a final concentration of 62.5 ␮g/ml
and incubated for 30 min at 37°C. The suspension was washed three times
with 2 ml of PBS (24,000 ⫻ g, 15 min) to remove free DiI, and the pellet
was resuspended in 400 ␮l of PBS.
4645
4646
ROLE OF MACROPHAGES IN CNS INFLAMMATION
(Fig. 1a). There were no detectable bacteria in the blood between
12 and 24 h, and they reappeared at 48 h post S. pneumoniae
injection into the cisterna magna (Fig. 1b). To see whether the
decrease in bacteria correlated with the presence of granulocytes,
which play a major role in bacterial clearance, we quantified the
cells present in the CSF. We observed a maximal influx of granulocytes, mainly neutrophils (⬃75%), at 12 and 24 h post S. pneumoniae inoculation, followed by a decrease after 48 h (Fig. 1c).
The granulocytes were mainly situated within the subarachnoidal
and Virchow-Robin space, and only a low percentage seemed to be
located within the parenchyma of the brain (Fig. 1d). Based on
these observations, we decided to investigate the effect of MM and
PVM depletion 12 h post meningitis induction.
MM and PVM depletion enhances clinical symptoms and
bacterial numbers
We have recently developed a liposome-based depletion method
(17) that allows us to investigate the role of the MM and PVM
during inflammation in the CNS. This method is based on a single
intraventricular injection of clodronate liposomes resulting in a
complete and selective depletion of the MM and PVM throughout
the CNS. We induced the bacterial meningitis model 7 days post
FIGURE 2. Depletion of the MM and PVM from the CNS during meningitis. Staining of the MM and PVM with the mAb ED2 (pink cells). a,
Cerebellum of a control liposome-treated animal (⫻200). The small indent shows ED2⫹ PVM near a blood vessel, at a higher magnification (⫻1000). b,
Cerebellum of a clodronate liposome-treated animal (⫻200). Again, Indent represents a higher magnification (⫻1000) of a blood vessel. Note the complete
absence of PVM in b.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
FIGURE 1. S. pneumoniae meningitis model in the rat. a and b, Number of CFU in the CSF and blood. c, Number of infiltrated cells into the CSF. d,
Massive influx of granulocytes (HIS 48⫹, pink cells) in the cerebellum of the rat 12 h post meningitis induction (⫻100). The granulocytes are mainly located
in the subarachnoidal space (large arrow), perivascular space (arrow head), and only a small percentage reside in the parenchyma (small arrow). Data are
mean ⫾ SE of four rats per time point.
The Journal of Immunology
4647
duction; both the MM and PVM were still completely absent from
all CNS areas tested (Fig. 2, a and b). We evaluated the animals for
clinical symptoms 12 h post S. pneumoniae inoculation and observed that the depleted group was clearly more ill than the control
group (Fig. 3a). In correlation with the more severe disease course
( p ⫽ 0.04), we observed profoundly higher bacterial counts in
CSF (approximately ⫻600) and blood (approximately ⫻20) of the
MM- and PVM-depleted animals (Fig. 3, b and c).
MM and PVM depletion decreases leukocyte influx into the CSF
depletion of the MM and PVM. At this time point, the liver and
draining lymph nodes, which suffer from a mild and temporal macrophage depletion using this regime, are completely repopulated
with macrophages (17). First, we tested whether this method could
also be used to deplete the MM and PVM during meningitis in-
FIGURE 4. MM and PVM depletion decreases leukocyte influx into the
CSF. The amount of infiltrated leukocytes into the CSF, 12 h post meningitis induction, of control (n ⫽ 7) and depleted (n ⫽ 11) animals is shown
in a (ⴱ, p ⬍ 0.05, unpaired t test). In both groups the leukocytes are
predominantly composed of granulocytes (b).
FIGURE 5. MM and PVM depletion does not influence the intrinsic
leukocyte migratory capacity. The number of immigrated cells, 4 h after S.
pneumoniae peritonitis induction, was not significantly different between
the control (n ⫽ 5) and depleted group (n ⫽ 6) (p ⫽ 0.34, unpaired t test).
Infiltrated leukocytes were mainly granulocytes (b).
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
FIGURE 3. MM and PVM depletion enhances clinical symptoms and
bacterial numbers. a, Clinical score of control (n ⫽ 7) and depleted (n ⫽
11) animals 12 h post S. pneumoniae injection (ⴱ, p ⬍ 0.05, unpaired t test).
The numbers of bacteria (CFU) are depicted in CSF (b) 12 h post meningitis induction (ⴱⴱ, p ⬍ 0.007, Mann-Whitney U test).
Although the results described above were in agreement with a
possible role of the MM and PVM in bacterial clearance, the dominant clearing cell is probably the granulocyte (2, 6). Therefore, it
was important to investigate whether MM and PVM depletion was
affecting the influx of granulocytes and other leukocytes into the
CSF. We observed that the number of leukocytes that had extravasated was significantly decreased in comparison with the control
group (Fig. 4a). Also no signs of increased cell death were observed. The leukocytes were mainly composed of granulocytes
(⬃90%), and no difference in composition was noted between depleted animals and controls (Fig. 4b). To exclude the possibility
that the decreased influx of granulocytes in the clodronate liposome-treated animals was caused by a direct effect of the clodronate liposomes on circulating neutrophils, a control experiment
was performed. Seven days after the depletion of the MM and
PVM, S. pneumoniae peritonitis was induced, and the number of
immigrated cells in the peritoneal cavity was quantified 4 h post
peritonitis induction. There was no significant difference between
the control and depleted animals (Fig. 5a). Analysis of May-Grünwald Giemsa-stained peritoneal cell cytocentrifuge preparations
showed that 4 h after i.p. bacteria injection the major infiltrating
population is formed by the granulocytes (⬃90%); this was the
same for the control, MM-, and PVM-depleted animal groups (Fig.
5b). In other words, the intrinsic migratory capacity of leukocytes
was not affected by the liposome treatment.
4648
ROLE OF MACROPHAGES IN CNS INFLAMMATION
FIGURE 6. MM and PVM depletion enhances the MIP-2 production in the CSF (a)
and serum (b) of the depleted animals vs the
control group. c, Enhanced IL-6 production
in the serum of the depleted group vs the
control group (ⴱ, p ⬍ 0.05; ⴱⴱ, p ⬍ 0.005,
Mann-Whitney U test).
It is known that high levels of proinflammatory cytokines correlate
with the pathophysiological symptoms and the clinical outcome in
meningitis patients (reviewed in Ref. 1). Therefore, it was important to establish whether the enhanced numbers of bacteria in the
circulation of the depleted animals were accompanied by higher
levels of systemic proinflammatory cytokines. Significantly higher
levels of IL-6 in the serum of the depleted group in comparison
with the control group were observed (Fig. 6c). We could not
observe detectable levels of TNF-␣ (⬍15 pg/ml) and IL-1␤ (⬍15
pg/ml) in the serum of either group.
MM and PVM depletion enhances MIP-2 production in the CSF
during meningitis
To investigate whether the decrease in neutrophil influx after MM
and PVM depletion was the possible consequence of a lowered
chemokine expression in the CSF, we measured the expression
levels of MIP-2. MIP-2 is both essential (28, 29) as well as sufficient (P. Zwijnenburg, unpublished observation) for the infiltration of granulocytes into the CSF during rodent bacterial meningitis. Moreover, MM and PVM appear to be the only endogenous
CNS sources of MIP-2 in meningitis (28). Surprisingly, CSF
MIP-2 levels of the depleted animals were much higher (530 –2577
pg/ml) in comparison with the control meningitis group (34 – 448
pg/ml) (Fig. 6a). We could even detect elevated MIP-2 levels
(7–22 pg/ml) in the serum of depleted animals as compared with
controls, which displayed levels that were below detection (⬍10
pg/ml) (Fig. 6b). Thus, the decrease in leukocyte influx could not
simply be explained by a decrease in MIP-2 expression.
MM and PVM depletion increases the expression of vascular
adhesion molecules on the BBB during experimental meningitis
An alternative explanation for the decreased influx of granulocytes
into the CSF in MM- and PVM-depleted animals was a possible
decrease in vascular adhesion molecule expression on the BBB.
Therefore, we analyzed the vascular adhesion molecule expression
on the cerebral endothelial cells, including VCAM-1 and ICAM-1,
which are known to play a role in neutrophil extravasation (2, 30,
31) and PECAM. Although in both cases the bacterial infection
resulted in an up-regulation of VCAM in comparison with noninfected animals (Fig. 7, a vs b and c), this was clearly enhanced in
the MM- and PVM-depleted meningitis group. This enhancement
was visible at both the level of staining intensity as well as the
FIGURE 7. MM and PVM depletion enhances vascular adhesion molecule expression on the BBB during meningitis. VCAM-1 staining on cerebroendothelial cells with the mAb 5F10 (pink cells) in control cerebellum (a), cerebellum of a control liposome-treated meningitis animal (b), and cerebellum
of a clodronate liposome-treated meningitis animal (c) (magnification, ⫻100).
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
MM and PVM depletion enhances systemic proinflammatory
cytokine production in meningitis
The Journal of Immunology
number of VCAM-positive vessels. An up-regulation of ICAM-1
and PECAM was also observed, but this difference was clearly less
profound than in the case of VCAM-1 (data not shown). This
demonstrates that, despite a significantly enhanced expression of
both relevant vascular adhesion molecules and chemokines, MM
and PVM depletion suppresses leukocyte extravasation.
Discussion
(12, 17), and the increase in bacteria in CSF after depletion of
these cells is in line with a direct role in clearance. However,
considering the concomitant reduction in granulocytes, which likewise represent the major scavengers during meningitis, we cannot
presently decide what possible contribution MM and PVM have.
Taken together, these findings demonstrate that MM and PVM
play a protective role during bacterial meningitis and suggest that
this is, at least in part, due to their supportive role in the influx of
leukocytes across the BBB. This also provides the first evidence
for a role of MM and PVM during inflammation in the CNS.
References
1. van Furth, A. M., J. J. Roord, and R. van Furth. 1996. Roles of proinflammatory
and anti-inflammatory cytokines in pathophysiology of bacterial meningitis and
effect of adjunctive therapy. Infect. Immun. 64:4883.
2. Quagliarello, V., and W. M. Scheld. 1992. Bacterial meningitis: pathogenesis,
pathophysiology, and progress. N. Engl. J. Med. 327:864.
3. Roos, K. L., A. R. Tunkel, and W. M. Scheld. 1997. Acute bacterial meningitis
in children and adults. In Infections of the Central Nervous System, 2nd Ed.
W. M. Scheld, R. J. Whitley, and D. T. Durack, eds. Lippincott-Raven, Philadelphia, pp. 335– 403.
4. Boje, K. M. K. 1995. Inhibition of nitric oxide synthase partially attenuates alterations in the blood-cerebrospinal fluid barrier during experimental meningitis
in the rat. Eur. J. Pharmacol. 271:297.
5. Ormann, K. L., J. L. Shenep, and B. K. English. 1998. Pneumococci stimulate the
production of the inducible nitric oxide synthase and nitric oxide by murine
macrophages. J. Infect. Dis. 178:1649.
6. Koedel, U., and H. W. Pfister. 1999. Superoxide production by primary rat cerebral endothelial cells in response to pneumococci. J. Neuroimmunol. 96:190.
7. Leib, S. L., Y. S. Kim, L. L. Chow, R. A. Sheldon, and M. G. Täuber. 1996.
Reactive oxygen intermediates contribute to necrotic and apoptotic neural injury
in an infant rat model of bacterial meningitis due to group B streptococci. J. Clin.
Invest. 98:2632.
8. Kastenbauer, S., U. Koedel, and H. W. Pfister. 1999. Role of peroxynitrite as a
mediator of pathophysiological alterations in experimental pneumococcal meningitis. J. Infect. Dis. 180:1164.
9. Koedel, U., and H. W. Pfister. 1999. Oxidative stress in bacterial meningitis.
Brain Pathol. 9:57.
10. Angelov, D. N., M. Walther, M. Streppel, O. Guntinas-Lichius, and W. F. Neiss.
1998. The cerebral perivascular cells. Adv. Anat. Embryol. Cell Biol. 147:1.
11. Graeber, M. B., W. J. Streit, and G. W. Kreutzberg. 1989. Identity of ED2positive perivascular cells in rat brain. J. Neurosci. Res. 22:103.
12. Mato, M., S. Ookawara, A. Sakamoto, E. Aikawa, T. Ogawa, U. Mitsuhashi,
T. Masuzawa, H. Suzuki, M. Honda, Y. Yazaki, et al. 1996. Involvement of
specific macrophage-lineage cells surrounding arterioles in barrier and scavenger
function in brain cortex. Proc. Natl. Acad. Sci. USA 93:3269.
13. Angelov, D. N., M. Walther, M. Streppel, O. Guntinas-Lichius, A.-M. van Dam,
E. Stennert, and W. F. Neiss. 1998. ED2-positive perivascular phagocytes produce interleukin-1␤ during delayed neuronal loss in the facial nucleus of the rat.
J. Neurosci. Res. 54:820.
14. Bauer, J., I. Huitinga, W. Zhao, H. Lassmann, W. F. Hickey, and C. D. Dijkstra.
1995. The role of macrophages, perivascular cells, and microglial cells in the
pathogenesis of experimental autoimmune encephalomyelitis. Glia 15:437.
15. Elmquist, J. K., C. D. Breder, J. E. Sherin, T. E. Scammell, W. F. Hickey,
D. Dewitt, and C. B. Saper. 1997. Intravenous lipopolysaccharide induces cyclooxygenase 2-like immunoreactivity in rat brain perivascular microglia and
meningeal macrophages. J. Comp. Neurol. 381:119.
16. Trostdorf, F., W. Bruck, M. Schmitz-Salue, K. Stuertz, S. J. Hopkins,
N. van Rooijen, I. Huitinga, and R. Nau. 1999. Reduction of meningeal macrophages does not decrease migration of granulocytes into the CSF and brain parenchyma in experimental pneumococcal meningitis. J. Neuroimmunol. 99:205.
17. Polfliet, M. M. J., P. H. Goede, M. L. van Kesteren-Hendrikx, N. van Rooijen,
C. D. Dijkstra, and, T. K. van den Berg. 2001. A method for the selective depletion of perivascular and meningeal macrophages in the central nervous system.
J. Neuroimmunol. 116:188.
18. van Rooijen, N., and A. Sanders. 1994. Liposome mediated depletion of macrophages: mechanism of action, preparation of liposomes and application. J. Immunol. Methods 174:83.
19. Umezawa, F., and Y. Eto. 1988. Liposome targeting to mouse brain: mannose as
a recognition marker. Biochem. Biophys. Acta 153:1038.
20. Claassen, E. 1992. Post-formation fluorescent labelling of liposomal membranes:
in vivo detection, localisation and kinetics. J. Immunol. Methods 147:231.
21. Paxinos, G., and Watson, C. 1986. The Rat Brain in Stereotaxic Coordinates.
Academic Press, San Diego.
22. Dijkstra, C. D., E. A. Döpp, T. K. van den Berg, and J. G. Damoiseaux. 1994.
Monoclonal antibodies against rat macrophages. J. Immunol. Methods 174:21.
23. Van Goor, H., V. Fidler, J. J. Weening, and J. Grond. 1991. Determinants of focal
and segmental glomerulosclerosis in the rat after renal ablation. Lab. Invest. 64:
754.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017
The MM and PVM, as intrinsic components of the BBB, are considered to form a first line of defense when the endothelial integrity
is circumvented (10). To elucidate the possible role of the MM and
PVM during bacterial meningitis, we have used a previously developed strategy using a single intraventricular injection of clodronate liposomes, which results in a complete and selective depletion of the MM and PVM throughout the CNS (17).
Our present results show that depletion of MM and PVM during
pneumococcal meningitis results in an aggravation of clinical
symptoms (Fig. 3a) demonstrating for the first time that these macrophages play a protective role during bacterial meningitis. MM
and PVM elimination resulted also in elevated numbers of bacteria, as well as enhanced levels of inflammatory mediators in both
CSF (MIP-2) and blood (IL-6). Furthermore, a reduced influx of
granulocytes and other leukocytes into the subarachnoid space was
observed after MM and PVM depletion; this occurred despite elevated levels of the relevant chemokine (MIP-2) and a higher expression of vascular adhesion molecules (e.g., VCAM-1).
What does this reveal about the role of MM and PVM in the
pathogenesis of bacterial meningitis? Based on the observation
that depletion of MM and PVM, as intrinsic components of the
BBB, reduces leukocyte influx (Fig. 4), it is concluded that MM
and PVM somehow support leukocyte traffic across the BBB. We
propose that, among the parameters studied, this role in leukocyte
transmigration constitutes the primary action of MM and PVM and
that this, together with our other observations, can explain their
protective role during meningitis. Although our results indicate
that MM and PVM play a supportive role in the infiltration of
leukocytes across the BBB during bacterial meningitis, the underlying mechanism remains elusive. One obvious possibility was that
in the early stages of meningitis MM and PVM are activated to
produce mediators (cytokines/chemokines) that induce endothelial
activation and leukocyte migration. We have investigated whether
the reduced migration of leukocytes after MM and PVM depletion
was accompanied by a reduced expression of relevant vascular
adhesion molecules (e.g., VCAM-1, ICAM-1) or chemokines (e.g.,
MIP-2), and this was clearly not the case. Instead, enhanced
VCAM-1 and MIP-2 levels were found. We believe that these
findings may in fact be the consequence of a local and/or systemic
immune response. An alternative explanation for the role of PVM
and MM may be that they can affect the kinetics of the inflammatory response during meningitis. A thorough kinetic analysis may
provide further insights.
Because granulocytes in the CSF are considered to eliminate the
majority of bacteria during meningitis (2, 6, 32), a reduction in
granulocytes resulted in increased bacterial numbers in the CSF
(Fig. 3). The higher load of bacteria (and/or their degradation products) may in turn potentiate leukocyte activation in the CSF, which
will result in enhanced production of various effector molecules,
including reactive oxygen species, that cause neuronal damage (9).
These effector molecules, together with the elevated systemic cytokine response (e.g., IL-6; Fig. 6), can explain the observed worsening of the clinical signs (1).
It should be emphasized that our present results cannot demonstrate, or rule out, a possible direct role of MM and PVM in bacterial clearance. Of course, MM and PVM are actively phagocytic
4649
4650
24. Tamatani, T., and M. Miyasaka. 1990. Identification of monoclonal antibodies
reactive with the rat homolog of ICAM-1, and evidence for a differential involvement of ICAM-1 in the adherence of resting versus activated lymphocytes to high
endothelial cells. Int. Immunol. 2:165.
25. Williams, K. C., R. W. Zhao, K. Ueno, and W. F. Hickey. 1996. PECAM-1
(CD31) expression in the central nervous system and its role in experimental
allergic encephalomyelitis in the rat. J. Neurosci. Res. 45:747.
26. May, M. J., G. Entwistle, M. J. Humphries, and A. Ager. 1993. VCAM-1 is a CS1
peptide-inhibitable adhesion molecule expressed by lymph node high endothelium. J. Cell Sci. 106:109.
27. Burstone, M. S. 1958. Histochemical comparison of naphtol-phosphates for the
demonstration of phosphatases. J. Natl. Cancer Inst. 20:601.
28. Diab, A., H. Abdalla, H. L. Li, F. D. Shi, J. Zhu, B. Höjberg, L. Lindquist,
B. Wretlind, M. Bakhiet, and H. Link. 1999. Neutralization of macrophage inflammatory protein 2 (MIP-2) and MIP-1␣ attenuates neutrophil recruitment in
the central nervous system during experimental bacterial meningitis. Infect. Immun. 67:2590.
ROLE OF MACROPHAGES IN CNS INFLAMMATION
29. Seebach, J., D. Bartholdi, K. Frei, K.-S. Spanaus, E. Ferrero, U. Widmer,
S. Isenmann, R. M. Strieter, M. Schwab, H. W. Pfister, and A. Fontana. 1995.
Experimental Listeria meningoencephalitis: macrophage inflammatory protein-1␣ and -2 are produced intrathecally and mediate chemotactic activity in
cerebrospinal fluid of infected mice. J. Immunol. 155:4367.
30. Davenpeck, K. L., S. A. Sterbinsky, and B. S. Bochner. 1998. Rat neutrophils
express ␣4 and ␤1 integrins and bind to vascular cell adhesion molecule-1
(VCAM-1) and mucosal addressin cell adhesion molecule-1 (MAdCAM-1).
Blood 91:2341.
31. Freyer, D., R. Manz, A. Ziegenhorn, M. Weih, K. Angstwurm, W.-D. Döcke,
A. Meisel, R. R. Schumann, G. Schönfelder, U. Dirnagl, and J. R. Weber. 1999.
Cerebral endothelial cells release TNF-␣ after stimulation with cell walls of
Streptococcus pneumoniae and regulate inducible nitric oxide synthase and
ICAM-1 expression via autocrine loops. J. Immunol. 163:4308.
32. Kernacki, K. A., R. P. Barret, J. A. Hobden, and L. D. Hazlett. 2000. Macrophage
inflammatory protein-2 is a mediator of polymorphonuclear neutrophil influx in
ocular bacterial infection. J. Immunol. 164:1037.
Downloaded from http://www.jimmunol.org/ by guest on June 15, 2017