Neuron

Article

Dual Roles for Perivascular Macrophages

in Immune-to-Brain Signaling
Jordi Serrats,1,3 Jennifer C. Schiltz,1,3,4 Borja Garcı́a-Bueno,1,5 Nico van Rooijen,2 Teresa M. Reyes,1,6
and Paul E. Sawchenko1,*
1Laboratory of Neuronal Structure and Function, The Salk Institute for Biological Studies and The Clayton Medical Research Foundation,

La Jolla, CA 92037, USA

2Department of Molecular Cell Biology, Vrije Universiteit Medical Center, 1007 MB Amsterdam, The Netherlands
3These authors contributed equally to this work
4Present address: Department of Anatomy, Physiology, and Genetics, Uniformed Services University of the Health Sciences,

Bethesda, MD 20814, USA

5Present address: Department of Pharmacology, School of Medicine, Universidad Complutense, 28040 Madrid, Spain
6Present address: Department of Pharmacology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA

*Correspondence: sawchenko@salk.edu
DOI 10.1016/j.neuron.2009.11.032

SUMMARY notably microglia, to precipitate or exacerbate a host of neuro-

degenerative disorders (Choi et al., 2009; Phillis et al., 2006).
Cytokines produced during infection/inflammation Clarifying the cellular-molecular mechanisms of immune-to-
activate adaptive central nervous system (CNS) brain communication thus has implications not only for under-
responses, including acute stress responses medi- standing basic central processes involved in coping with acute
ated by the hypothalamo-pituitary-adrenal (HPA) illness but also for identifying targets for intervention in neurolog-
axis. The mechanisms by which cytokines engage ical disease.
Here we focus on one key acute phase response system, the
HPA control circuitry remain unclear, though stimu-
hypothalamo-pituitary-adrenal (HPA) axis, an integral part of the
lated release of prostanoids from neighboring
brain’s stress response machinery (Turnbull and Rivier, 1999;
vascular cells has been implicated in this regard. van der Meer et al., 1996). Glucocorticoid mediators of HPA
How specific vascular cell types, endothelial cells function exert catabolic effects that mobilize energy reserves
(ECs) versus perivascular cells (PVCs; a subset of to facilitate coping with inflammatory insults and powerfully
brain-resident macrophages), participate in this suppress immune-inflammatory reactions. This latter effect
response remains unsettled. We exploited the plays a critical regulatory role in preventing excess cytokine
phagocytic activity of PVCs to deplete them in rats production and immune cell proliferation (Webster et al., 2002).
by central injection of a liposome-encapsulated proa- Dysfunction of the central arm of this feedback loop is implicated
poptotic drug. This manipulation abrogated CNS and in the genesis of autoimmune disorders in susceptible animal
hormonal indices of HPA activation under immune models (Harbuz et al., 1997) and in humans (Wick et al., 1993).
The mechanisms by which immune stimuli impact the brain to
challenge conditions (interleukin-1) that activated
engage the HPA axis remain unsettled. Multiple routes of access
prostanoid synthesis only in PVCs, while enhancing have been supported, whose involvement may vary with the
these responses to stimuli (lipopolysaccharide) that strength and nature of the insult (Dantzer and Kelley, 2007;
engaged prostanoid production by ECs as well. Quan, 2008). For stimuli involving intravenous administration of
Thus, PVCs provide both prostanoid-mediated drive individual proinflammatory cytokines (interleukin-1 [IL-1]) or
to the HPA axis and an anti-inflammatory action that pathogen analogs (bacterial lipopolysaccharide [LPS]), which
constrains endothelial and overall CNS responses model systemic infection, substantial evidence indicates that
to inflammatory insults. circulating cytokines can be monitored by non-neuronal cells
of the cerebral vasculature, which appear capable of engaging
proximate afferent projections to relevant effector neurons by
INTRODUCTION releasing local signaling molecules, notably prostaglandin E2
(PGE2; Elmquist et al., 1997; Schiltz and Sawchenko, 2003). In
Episodes of systemic infection or inflammation engage the the case of HPA control circuitry, evidence supports a role for
innate immune system to release proinflammatory cytokines PGE2 acting on brainstem catecholaminergic neurons that
that act on the brain to initiate specific central nervous system project to corticotropin-releasing factor (CRF)-expressing hypo-
(CNS) responses. These include a constellation of acute phase thalamic neurosecretory cells in initiating IL-1- or LPS-stimulated
reactions, including somnolence, fever, lethargy, anorexia, and drive on the axis (Ericsson et al., 1994, 1997; Schiltz and Saw-
metabolic effects (Hart, 1988; Konsman et al., 2002), which facil- chenko, 2007; van der Meer et al., 1996).
itate adaptation to the challenge at hand. Such insults can also Questions remain as to the manner and extent to which
impact the brain’s intrinsic immune effector mechanisms, inducible prostaglandin-dependent mechanisms within the brain

94 Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc.

Neuron
Vascular Mediation of Inflammatory Effects on CNS

contribute to HPA responses and the identity of the vascular cell a similar disposition of DiI/ED2 labeling, which was overlain by
type(s) involved in transducing immune signals and mounting induced COX-2 expression. Vascular and meningeal COX-2
prostanoid responses. Endothelial cells (ECs) of the cerebral immunoreactivity was strictly confined to ED2-labeled cells. In
vasculature are optimally positioned to record circulating contrast, Clod-Lip-treated animals challenged with IL-1 showed
immune signals, but their threshold to inducible cyclooxygenase in most cases a complete absence of vascular and meningeal
(COX-2) expression is high (Schiltz and Sawchenko, 2002). Peri- labeling for DiI, ED2, and COX-2. In these cases, reliable COX-2
vascular cells (PVCs), a subset of brain-resident macrophages, labeling was seen only in neurons (mainly in cortical regions) that
are more sensitive to COX-2 induction (Schiltz and Sawchenko, express the enzyme constitutively.
2002), but their position in the perivascular space between the Ancillary analyses were carried out to assess the broader
EC basement membrane and the glia limitans (Thomas, 1999; impact of central liposome treatment. Electron microscopy
Williams et al., 2001) makes them unlikely to be accessed (conventional and of ED2-immunolabeled material) of brain
directly by blood-borne cytokines. tissue from control and Clod-Lip-treated rats revealed no major
To address these issues, we used a method that exploits the disruption of vascular structure in the latter, save for instances of
phagocytic capabilities of PVCs to selectively deplete them by vacant areas and/or apparent cellular debris in the perivascular
central administration of a liposome-encapsulated drug, clodro- spaces, presumably reflecting the loss of ED2+ PVCs (Figure 1).
nate (Polfliet et al., 2001b). The results support a prominent and Basal laminae and junctions (zonulae occludens) between
selective involvement of brain macrophages and their capacity adjoining ECs appeared unaffected, and there was no evidence
to mount prostanoid responses in immune challenge-induced of infiltration by circulating lymphocytes. The integrity of the
HPA activation. They also identify a novel, anti-inflammatory blood-brain barrier was assessed by injecting experimental
influence of this cell type in restraining endothelial involvement and control animals i.v. with horseradish peroxidase. Light level
in CNS host-defense responses to inflammatory insults. analysis of histochemical preparations revealed no detectable
penetration of enzyme into the brain parenchyma in either group,
aside from that expected at circumventricular organs. Finally, as
RESULTS central injection of Clod-Lips is reported to variably and tran-
siently attenuate peripheral macrophage populations (Polfliet
Liposome-Mediated Targeting of Brain Macrophages et al., 2001b), we assessed the most sensitive of these, hepatic
We sought first to validate a method (Polfliet et al., 2001b) that Kupffer cells, during the time window (5–7 days postinjection)
exploits the phagocytic activity of PVCs and meningeal macro- in which immune challenges were administered (Figure S1,
phages (MMs) to selectively deplete them by i.c.v. injection of available online). Comparisons of the density of ED2-labeled
liposome-encapsulated clodronate. Clodronate is a bisphospho- cells in comparable series through the liver of control and
nate drug used to treat osteolytic disease, which can cause Clod-Lip-treated rats (n = 3) revealed no significant difference
irreversible metabolic damage and apoptosis at high intracellular (p > 0.1).
concentrations (Van Rooijen and Sanders, 1994). Liposomes These observations support the efficacy and selectivity of
were labeled with the fluorophore DiI to allow visualization of i.c.v. injection of Clod-Lips in depleting brain-resident vascular
cells that have incorporated them (Polfliet et al., 2001b). Injection and MMs.
of control liposomes encapsulating PBS (PBS-Lips) resulted
in selective DiI labeling of meningeal and vascular cells that Vascular Responses to Immune Insults
expressed the ED2 macrophage differentiation antigen, a defini- Control rats sacrificed 3 hr after IL-1 injection (2 mg/kg, i.v.) dis-
tive PVC/MM marker (Dijkstra et al., 1985; Figure 1). Using clod- played the expected induction of COX-2 in the cerebral vascula-
ronate liposomes (Clod-Lips), ED2-positive cells were frankly ture and meninges; pretreatment with i.c.v. injection of Clod-Lips
reduced in rats sacrificed at 2–3 days and virtually absent in 5 days earlier completely eliminated detectable IL-1-induced
animals killed 5–7 days after injection. COX-2 expression at these loci (Figure 2). LPS injections (2 mg/kg,
The effectiveness of liposome-mediated targeting of brain i.v.) in PBS-Lip rats provoked the expected COX-2 induction in
macrophages was evaluated after each experiment by costain- both multipolar and round vascular profiles shown previously
ing series of brain sections for ED2, challenge-induced COX-2 by dual staining to conform to PVCs (ED2+) and ECs (RECA-
expression, and/or the DiI fluorophore. Rats killed 5–7 days after 1+), respectively (Schiltz and Sawchenko, 2002). Predictably,
i.c.v. injection of PBS-Lips and i.v. injection of vehicle displayed LPS challenges of Clod-Lip pretreated rats elicited COX-2
DiI labeling strictly limited to ED2-stained cells associated with responses only in ECs, but this effect was remarkably potentiated
the meninges and cerebral vasculature. The efficiency of labeling in that COX-2-ir ECs were far greater in number (2.4-fold
was high, with >95% of all ED2-positive cells colabeled for DiI increase) and staining intensity in macrophage-depleted than
in samplings from multiple brain regions (see Experimental control rats. Because ECs far outnumber PVCs, the overall
Procedures). COX-2 was not detected in DiI- and/or ED2-stained density of COX-2-labeled vascular cells in Clod-Lip rats chal-
cells of control animals. In separate series of sections prepared lenged with LPS more than doubled that of PBS-Lip-pretreated
for concurrent localization of DiI and Iba1, a marker that labels controls (Figure 2).
microglia, as well as PVCs/MMs, we failed to detect DiI labeling Analysis of more specific indices of PGE2 production as func-
of parenchymal microglia in any experiment (data not shown). tion of treatment status yielded generally compatible findings.
Rats that received i.c.v. injection of control liposomes and Thus, vascular expression of microsomal prostaglandin E syn-
were challenged 7 days later with IL-1 (2 mg/kg, i.v.) displayed thase-1 (mPGES-1), a terminal enzyme in PGE2 synthesis, which

Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc. 95

 Neuron
Vascular Mediation of Inflammatory Effects on CNS

Figure 1. Liposome Targeting of Brain Macrophages

(Top) Fluorescence images showing triple labeling for DiI (left panels, red), ED2 (middle left panels, green), and COX-2 (middle right panel, blue) in the meninges
and a blood vessel of a rat given DiI-labeled PBS-liposomes into a lateral ventricle 7 days prior to an i.v. injection of IL-1 (2 mg/kg). Merged images (right panels)
show that virtually all DiI-labeled profiles colocalize with ED2- and COX-2-irs, indicating that the liposomes were selectively taken up by ED2-positive PVCs and
MMs and that inducible COX-2 expression is discretely localized to this cell type.
(Second row) Images showing the same markers, but from a rat that received i.c.v. injection of clodronate-filled liposomes (Clod-Lips) prior to an IL-1 challenge.
Clod-Lip pretreatment results in a loss of detectable DiI, ED2, and COX-2 labeling in the meninges and blood vessels. Scale bar: 100 mm.
(Third row) Immunoelectron micrographs of arterioles from rats injected i.c.v. with control or Clod-Lips 7 days earlier. Surface labeling for the ED2 antigen (arrow-
heads) is seen in cellular elements displaying macrophage-like features (e.g., numerous lysosomes or multivesicular bodies) in the perivascular spaces (pvs) of
PBS-, but not Clod-Lip-, treated animals. Other aspects of vascular structure, including the morphology of endothelial (EC) and smooth muscle (SM) cells and
basal laminae (bl), are ostensibly unaffected by Clod-Lip treatment. Scale bar: 1 mm.
(Bottom row) Light micrographs through comparable regions of temporal cortex from control and Clod-Lip pretreated rats, showing vascular labeling for HRP
60 min after i.v. injection. No evidence of leakage of enzyme into brain parenchyma is evident. Scale bar: 100 mm.

is commonly coupled to COX-2 activity, was not detectable Clod-Lip pretreated animals displayed massively potentiated
under basal conditions, but was mildly upregulated in response upregulation of mPGES-1 mRNA in presumed ECs following an
to IL-1 and more strongly in response to LPS in control (PBS-Lip) LPS challenge, consistent with the COX-2 findings in this cell
rats (Figure S2). Concurrent dual localization of mPGES-1 mRNA type described above, but this was also seen in IL-1-injected
and ED2-ir indicated that enzyme expression included PVCs rats, whose endothelia failed to mount a detectable COX-2
in both challenge paradigms, though these were sparse. response.

96 Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc.

Neuron
Vascular Mediation of Inflammatory Effects on CNS

To determine whether altered expression of prostaglandin CRF-rich hypophysiotropic zone of the PVH, and this response
biosynthetic enzymes in Clod-Lip pretreated rats is reflected was reduced (by 80%) in Clod-Lip pretreated animals to levels
in altered release of PGE2, we attempted time course analyses that did not differ from those of vehicle-injected control animals
in which prostanoid levels were measured by enzyme immuno- (p > 0.1). LPS provoked a similarly focused, but more robust,
assay in extracts of hypothalamus or medulla. This was not activation of PVH neurons in PBS-Lip rats. Clod-Lip pretreat-
feasible in IL-1-challenged animals, as basal and stimulated ment resulted in a reliable increase in the number of responsive
PGE2 were below the detection limits of the assay. With LPS, PVH neurons, attributable to recruitment of cells in the magno-
however, Clod-Lip pretreated rats displayed a more rapid, cellular division of the nucleus, which have also been implicated
pronounced, and prolonged LPS-induced rise in brain tissue as contributing to central drive on the HPA axis (Holmes et al.,
levels of PGE2 than controls (Figure 2). To overcome the lack 1986). This analysis was carried out in material prepared under
of sensitivity in the IL-1 model, we established a protocol for im- staining conditions that optimize sensitivity of immunolocaliza-
munolocalization of PGE2 itself (see Experimental Procedures) tion, and the robustness of LPS-induced Fos in the parvocellular
and found a clear and discrete upregulation of PGE2-ir in identi- neurosecretory zone of the PVH might impose a ceiling effect.
fied PVCs of intact rats injected with IL-1; similarly challenged Staining of additional series of sections from control and Clod-
Clod-Lip pretreated animals failed to display detectable prosta- Lip animals challenged with LPS using higher dilutions of Fos
noid signals in brain (Figures 2 and S3). LPS challenge in control antiserum revealed a reliable (31%) increase in the number of
rats also gave rise to detectable PGE2 immunostaining, which labeled cells in the parvocellular PVH, supporting the contention
was punctate in appearance and codistributed in part with an that macrophage depletion results in enhanced drive to both
endothelial marker. This staining was enhanced in macro- major functional compartments of the nucleus. This finding
phage-depleted rats and, in addition, extended into the paren- was further supported by comparisons of relative levels of CRF
chyma, with some taking the form of microglial-like cellular mRNA in the PVH, which revealed a reliably greater increase in
profiles, a localization confirmed by costaining with a microglial the Clod-Lip (2.2-fold) than PBS-Lip group (1.6-fold) in response
marker (Iba-1; data not shown). to LPS (p < 0.05).
Overall, the enhanced inducible PGE2 responses to LPS in the Together, these findings indicate that brain macrophages are
brain macrophage-depleted model indicates that PVCs exert required for IL-1 engagement of HPA control circuitry, which at
a potent restraining influence on EC activity in transducing circu- both medullary and hypothalamic levels responds to brain macro-
lating immune signals. The starkly differential effects of liposome phage depletion in a manner indicative of a positive relationship
treatment on multiple indices of PGE2 production after IL-1 with vascular COX-2/PGE2 production. This extends to the
versus LPS challenges define models for evaluating the role of exaggerated HPA-regulatory responses seen in the LPS model.
vascular prostanoid synthesis in acute phase responses.
Stress Hormone Secretion
Effects on Central Stress-Related Circuitry To determine whether altered responsiveness of HPA control
Differences in IL-1- and LPS-induced vascular COX-2 induction circuitry as a function of brain macrophage status is mirrored
in Clod-Lip animals were mirrored in challenge effects on HPA by changes in hormonal output, separate groups of rats (n = 5)
control circuitry, assessed using Fos immunostaining as a pretreated with control or Clod-Lips were implanted with jugular
generic marker of cellular activation (Figure 3). Projections to catheters and challenged 2 days later with IL-1 or LPS as
the hypothalamus arising from catecholamine-containing (adren- above, and repeated blood samples were collected for radioim-
ergic and noradrenergic) neurons in the caudal brainstem munoassay of plasma adrenocorticotropic hormone (ACTH)
(nucleus of the solitary tract and ventrolateral medulla) have and corticosterone. Basal titers of neither hormone varied
been specifically implicated in mediating IL-1 and LPS effects reliably as a function of brain macrophage status (Ps > 0.1;
on HPA output (Ericsson et al., 1994; Schiltz and Sawchenko, Figure 4). An IL-1 challenge elicited rapid and reliable increases
2007; van der Meer et al., 1996). Here, we confirmed that in plasma levels of both hormones that peaked at similar time
both cell groups display markedly increased Fos expression points. However, both the ACTH and corticosterone secretory
in response to standard doses of IL-1 or LPS (2 mg/kg, i.v.). responses of Clod-Lip-treated rats were of significantly lesser
Costaining for Fos and the catecholamine biosynthetic enzymes magnitude and duration than those of control animals. Total inte-
dopamine-b-hydroxylase and phenylethanolamine-N-methyl- grated responses, as assessed by calculating areas under the
transferase identified the overwhelming majority of responsive curve (AUC), were reduced to 49% (ACTH) and 37% (corticoste-
neurons as displaying the adrenergic or noradrenergic pheno- rone) of control values (Ps < 0.001).
type (Figure S4). Whereas Clod-Lip pretreatment exerted no LPS also elicited significant increases in ACTH and corticoste-
discernible effect on Fos expression in saline-injected controls, rone secretion in both groups that peaked at similar time points,
it resulted in markedly reduced responses to IL-1, to levels that and in the case of corticosterone achieved similar maxima
did not differ reliably from those of controls. In contrast, brain (Figure 4). But the magnitude of peak ACTH titers and the dura-
macrophage depletion yielded significant increases in the tion of both hormonal responses were substantially greater in
number of NTS and VLM neurons displaying activational Clod-Lip-treated rats, as were AUC measures (2.4- and 1.5-fold
responses to LPS, over and above that seen after similar injection increases for ACTH and corticosterone, respectively; Ps < 0.001).
in PBS-Lip pretreated animals (p < 0.01). Thus, the differential effects of brain macrophage depletion on
At the level of hypothalamic effectors that control HPA output, IL-1- and LPS-driven engagement of HPA control circuitry are
IL-1-induced Fos expression in control rats was focused in the predictive of alterations in stress hormone secretion.

Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc. 97

 Neuron
Vascular Mediation of Inflammatory Effects on CNS

Figure 2. Liposome Effects on Indices of Vascular Prostanoid Production

(Top) Differential effects of liposome treatment on induced vascular COX-2 induction. IL-1 treatment in control (PBS-Lip) animals induces COX-2 in multipolar
cells (shown previously to correspond to PVCs), while LPS activates the enzyme in both PVCs (arrows) and in round profiles that conform to nuclear/perinuclear
regions of endothelia. While Clod-Lip treatment eliminates IL-1-induced COX-2 expression by ablating PVCs, it enhances LPS effects on the number and
staining intensity of COX-2-ir in endothelia, indicating that PVCs restrain EC responsiveness. Histogram shows quantitative comparison of the density (number
of cells/mm2 vascular area) of COX-2-ir PVCs, endothelia, and total vascular cells as a function of treatment status. n = 6 per group. *p < 0.001 versus PBS-
liposome-treated controls.
(Middle) LPS effects on tissue PGE2 levels in control and Clod-Lip pretreated rats. LPS-induced elevations in mean (±SEM) PGE2 concentration in medulla are
significantly greater over 1–3 hr after injection in brain macrophage-deficient rats.
(Bottom) Effects of brain macrophage depletion on immune challenge-induced PGE2-ir in vascular cell types. Merged confocal images from rats treated with IL-1
(top row) or LPS (bottom row) co-stained for PGE2-ir (green) markers (red) for perivascular (ED2; top) or EC (RECA-1; bottom) in untreated (left) and IL-1- or LPS-
challenged rats pretreated with control (middle) or clodronate liposomes (right). Doubly stained elements appear as yellow. Under control conditions, PGE2-ir is

98 Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc.

Neuron
Vascular Mediation of Inflammatory Effects on CNS

Other Acute Phase Responses which acts in a paracrine manner to engage HPA control circuitry
Groups of rats (n = 9) pretreated centrally with clodronate or and, consequently, its effector arm in the endocrine hypothal-
control liposomes were implanted with telemetry transducers amus. Accordingly, elimination of PVCs abrogates vascular
to allow continuous remote monitoring of body temperature PGE2 production and the HPA response. Moderate to high doses
and locomotor activity. Two days later, they were challenged of LPS engage PVCs in a similar manner and are able to
with IL-1 or LPS 1 hr prior to the onset of the nocturnal phase partially overcome the PVC-imposed brake, leading to endothe-
of the lighting cycle, when animals are most active. lial PGE2 synthesis and a greater total vascular prostanoid
In rats pretreated with control liposomes, IL-1 elicited response. PVC ablation removes the brake, leading to further
a biphasic core temperature response, with an initial mild hypo- enhancement of EC and overall PGE2 production, and its effects
thermia preceding a more substantial fever that peaked at 2 hr are manifest particularly during the later stages of HPA and
after injection (Figure 5). Clod-Lip treated animals failed to exhibit febrile responses.
a significant decline in core temperature, but the peak magnitude The immune challenge parameters used here were chosen
of their febrile responses did not differ reliably from those of based on their ability to elicit vascular COX-2 responses exclu-
controls. LPS provoked in both groups an early rise and decline sively from PVCs (IL-1) or from both these and ECs (LPS; Schiltz
in core temperature, whose timing and magnitude were similar. and Sawchenko, 2002). Bolus administration of IL-1 is an admit-
Thereafter, controls exhibited a biphasic rise in temperature tedly artificial situation, but presents a discrete stimulus that
with peaks at roughly 2.75 and 6 hr after injection, whereas brain mimics many CNS effects of systemic infection (Dinarello,
macrophage-depleted rats displayed a monophasic fever that 1991). LPS is a component of the cell wall of gram-negative
peaked at 6 hr and whose magnitude was significantly greater bacteria and is widely used to model infection or sepsis (Ulevitch
over the 3.5–4.5 hr time interval (Ps < 0.05). and Tobias, 1995). It stimulates the release of several proinflam-
In contrast, activity responses were unaffected by Clod-Lip matory cytokines, prominently including IL-1 (Turnbull and Riv-
pretreatment. Both IL-1 and LPS treatments (Figure 5) gave ier, 1999), and can also act directly by binding a toll-like receptor
rise to rapid and significant (Ps < 0.01) decrements in activity 4/MD2/CD14 complex (Laflamme and Rivest, 2001). Impor-
beginning at 0.5–1.5 hr after injection, and their magnitude did tantly, like IL-1, low doses of LPS stimulate COX-2 only in
not differ reliably as a function of brain macrophage status at PVCs (Schiltz and Sawchenko, 2002) and we would predict
any time point through 6 hr after injection (Ps > 0.10; Figure 5). that results obtained in the IL-1 model would generalize to mild
Repetition of this study in separate groups challenged with LPS challenges.
IL-1 or LPS during the subjective morning hours yielded compat- These findings clarify the cellular mechanisms by which chal-
ible results (Figure S5). lenges that model systemic infection access the brain to engage
specific CNS response systems, notably HPA axis activation and
DISCUSSION fever, that facilitate coping with insult. Glucocorticoid mediators
of HPA function serve to mobilize energy reserves and constrain
Differential effects of brain macrophage ablation on indices of systemic immune responses, while fever may create a subop-
cerebrovascular PGE2 production induced by IL-1 (reduced) timal thermal environment for pathogen proliferation and can
versus LPS (exaggerated) were paralleled by altered responses directly enhance certain host defense mechanisms (Mackowiak,
of the HPA axis and its CNS control circuitry. Other acute phase 1998).
responses were affected less profoundly (fever) or not at all (leth- Understanding the transit of immune signals across the blood-
argy). The results support a dependence of HPA responses to brain barrier also has implications for the many neurodegenera-
proinflammatory insults on the integrity of PVCs and their tive diseases in which central inflammatory mechanisms are
capacity to mount prostanoid responses and define a novel believed to play a contributing role. Systemic LPS and/or IL-1
anti-inflammatory interaction between perivascular cells and treatment is commonly reported to exacerbate neuropathology
ECs in transducing circulating cytokine signals and sculpting and cognitive/behavioral symptoms in animal models of
specific CNS responses to them. Alzheimer’s, Parkinson’s and Prion diseases and ALS (Byler
et al., 2009; Cunningham et al., 2005; Kitazawa et al., 2005;
Model and Implications Letiembre et al., 2009). A single (high) dose of LPS, alone, can
A model to summarize these findings is provided in Figure 6. In cause persistent CNS inflammation and neuronal loss (Qin
response to IL-1, the circulating cytokine activates PVCs most et al., 2007). Moreover, animals challenged neonatally with IL-1
likely indirectly, after first being bound by endothelial type 1 or LPS display altered central regulation of the HPA axis that
IL-1 receptors (Ericsson et al., 1995). ECs are responsive to persists into adulthood (Shanks et al., 1995) and may predispose
this stimulus and signaling through them is required for upstream individuals to an even broader array of stress-related patholo-
CNS effects (Ching et al., 2007; Gosselin and Rivest, 2008), but gies. In this light, the identification of a potent anti-inflammatory
they do not display detectable indices of PGE2 production, due mechanism at the blood-brain interface should provide leverage
at least partly to a restraining influence exerted on them by in identifying targets for intervention in a range of pathologies.
PVCs. PVC activation leads to production and release of PGE2, The liposome-targeting approach used here has been shown

not detectable in either vascular cell type. In intact rats, IL-1 provokes robust cytoplasmic PGE2 staining localized discretely to PVCs, while cytokine-induced
labeling is absent in PVC-depleted animals. LPS induces more widespread punctate PGE2 staining, a portion of which localizes to ECs; this labeling is enhanced
in Clod-Lip-pretreated animals. Scale bars: 100 mm.

Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc. 99

 Neuron
Vascular Mediation of Inflammatory Effects on CNS

Figure 3. Liposome Effects on Immune Challenge-Induced Activation of HPA Control Circuitry

Brightfield images of IL-1- and LPS-induced Fos expression in the ventrolateral medulla (top) and PVH (middle) as a function of brain macrophage status. Histo-
grams show mean ± SEM number of cells in the A1 and C1 regions of the ventrolateral medulla and PVH in each condition (n = 5–7 per group). Liposome treatment
did not significantly affect basal levels of Fos expression in either region, and data from PBS- and Clod-Lip rats that received i.v. saline injections are pooled for
presentation. Clod-Lip treatment reduced IL-1 stimulated Fos expression in medullary neurons whose projections are required for HPA activation and their hypo-
thalamic targets in the parvocellular part of PVH (mp; outlined in blue) that governs HPA output. In contrast, brain macrophage depletion enhanced activational
responses induced by LPS at both levels of the HPA control circuit. This includes recruitment of additional cell types in the magnocellular division of the nucleus
(pm; outlined in red).
(Bottom) Darkfield photomicrographs showing CRF mRNA under basal (saline-injected) and 3 hr after LPS challenge in control and Clod-Lip-injected rats.
Mean ± SEM relative levels of this transcript in these treatment groups (n = 5–8) are shown. Clod-Lip pretreated rats display significantly enhanced upregulation
of CRF mRNA, relative to similarly challenged rats that received control liposome injections. *p < 0.05; **p < 0.01 versus saline-injected controls. Scale bars:
100 mm.

100 Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc.
Neuron
Vascular Mediation of Inflammatory Effects on CNS

Figure 4. Liposome Effects on Challenge-

Induced Stress Hormone Secretion
Mean ± SEM plasma ACTH (top) and corticoste-
rone (CORT; bottom) levels in control rats (black)
and animals pretreated with Clod-Lips (red) and
subsequently challenged with i.v. injection of
3 mg/kg IL-1 (left) or 2 mg/kg LPS (right). In IL-1-
challenged rats, macrophage ablation did not
affect the timing of peak ACTH and CORT
responses, but attenuated their magnitude, and
markedly reduced the time over which significant
elevations were observed. Conversely, and in
line with the COX-2 and Fos data, the principal
effect of Clod-Lip pretreatment in LPS-challenged
rats was to significantly enhance the longevity of
hormonal responses. n = 5–6 per group. *, differs
significantly from control liposome group,
p < 0.05; **p < 0.01; ***p < 0.001.

Watanabe et al., 1990). The cellular sour-

ces of central prostaglandin production
have remained a major point of conten-
tion, with previous reports being divided
as to whether PVCs (Elmquist et al.,
1997; Schiltz and Sawchenko, 2002) or
ECs (Cao et al., 1996; Matsumura et al.,
to modulate disease progression in animal models of multiple 1998; Quan et al., 1998) are the dominant seats of LPS- and/or
sclerosis, meningitis, and Alzheimer’s disease (Hawkes and IL-1-driven vascular COX-2 induction. We offered a potential
McLaurin, 2009; Polfliet et al., 2001a, 2002) and the feasibility reconciliation by identifying PVCs as the sole site of COX-2
of using the stem cells that give rise to them as vehicles for expression following IL-1 or low doses of LPS, with endothelia
gene therapy has been established (Hahn et al., 1998; Priller recruited at higher endotoxin doses (Schiltz and Sawchenko,
et al., 2001). 2002).
Multiple indices of IL-1-driven HPA activity were suppressed
What Are Perivascular Cells? in brain macrophage-depleted rats, supporting a prominent
Along with MMs, PVCs are derived from bone marrow precur- role for PVCs in the activation of central stress circuitry and
sors that populate the brain in early postnatal life and turn over hormone secretion under this condition. As IL-1 is the most
slowly throughout adulthood (Bechmann et al., 2001a; Vallieres potent of the proinflammatory cytokines induced by LPS in stim-
and Sawchenko, 2003). They are distinct from pericytes and ulating HPA output (Turnbull and Rivier, 1999), we conclude that
parenchymal microglia, though they may share a common PVCs provide a generalized, low-threshold impetus for the
lineage with them (Gehrmann et al., 1995; Thomas, 1999). As engagement of this system by infection/inflammation. Hormonal
shown in the present study, PVCs are constitutively phagocytic, responses to IL-1 were not completely eliminated by central
with their definitive ED2 marker having been identified as a macrophage depletion, leaving open possible contributions of
macrophage scavenger receptor, CD163 (Fabriek et al., 2007). adjunct central or peripheral mechanisms. Direct IL-1 actions
PVCs can serve as antigen-presenting cells (Hickey and Kimura, on the pituitary and adrenal glands to stimulate ACTH and corti-
1988), participate in the immune surveillance of the nervous costerone secretion, respectively, have been described (Bernton
system (Bechmann et al., 2001b; Hickey, 2001), and have been et al., 1987; Engstrom et al., 2008).
implicated in a host of pathological processes, including the In addition to indicating a requirement for PVCs in IL-1-induced
entry of HIV into the nervous system (Stoll and Jander, 1999; HPA activation, our findings are also consistent with COX-depen-
Thomas, 1999). dent PGE2 production in mediating this influence. PGE2 produc-
tion from the COX-dependent intermediate PGH2 may be
Prostaglandin Involvement in CNS Host Defense catalyzed by several terminal PGE2 synthases, with mPGES-1
PGE2 levels in brain are increased following systemic IL-1 or LPS being commonly coupled to COX-2 activity (Ivanov and Roma-
injection (Dinarello, 1991; Sehic et al., 1996), and a wealth of novsky, 2004). While we found that vascular mPGES-1 expres-
evidence indicates that central or peripheral blockade of prosta- sion co-varied directly with COX-2, PGE2, and HPA-related
noid synthesis using COX inhibitors (NSAIDs) can disrupt the endpoints across most treatment conditions, it was expressed
HPA activation normally elicited by these challenges (Ericsson only weakly in PVCs of IL-1-stimulated rats. This raises the ques-
et al., 1997; Katsuura et al., 1988; Lacroix and Rivest, 1998; tion of how PGE2 is generated in this cell type with the alacrity

Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc. 101
 Neuron
Vascular Mediation of Inflammatory Effects on CNS

Figure 5. Liposome Effects on IL-1- and

LPS-Induced Fever and Lethargy
(Left) Mean ± SEM change in core body tempera-
ture (Tc; data points in 3 min bins) of rats pretreated
with control (black) or Clod-Lips (red) and chal-
lenged (arrowhead) with i.v. injection of IL-1 (top)
or LPS (bottom) 1 hr before lights out. Brain macro-
phage ablation eliminated an early hypothermic
response to IL-1, but did not significantly affect
the magnitude or duration of subsequent fever. In
contrast, LPS-induced fever was enhanced at
3.5–4.5 hr after injection.
(Right) Locomotor activity data (averaged into
30 min bins) from the same animals. Clod-Lip treat-
ment did not affect the hypoactivity elicited by
either treatment. n = 9 per group. *, differs signifi-
cantly from control liposome group, p < 0.05.

Involvement of PGE2, itself, in CNS

responses to inflammatory stimuli is
commonly inferred on the basis of COX-
2 expression. Here, we provide bio-
chemical/histochemical data supporting
discrete PGE2 induction in PVCs after
IL-1 treatment, more broadly in response
to LPS, and for regulation in step with
HPA endpoints in macrophage-depleted
required of acute phase mechanisms. There are recent data to rats. This evidence is correlative, and it must be noted that
implicate the constitutively expressed cyclooxygenase isoform PGE2 is one of several prostanoids whose levels increase in brain
COX-1 in the initial phase of LPS-induced HPA activation following insults of the kind we used (Choi et al., 2008; de Vries
(Elander et al., 2009; Garcia-Bueno et al., 2009; Zhang et al., et al., 1995). Supporting specific involvement of PGE2 are the
2003). New findings that COX-1 is expressed under basal findings that IL-1 activates medullary catecholaminergic neurons
conditions by PVCs and microglia, and is LPS-inducible in that express EP3 (and inducible EP4) PGE2 receptors (Ek et al.,
endothelia (Garcia-Bueno et al., 2009), are consistent with a 2000) and that microinjection of PGE2 into these cell groups
role in the present context. It remains to be determined whether discretely activates them and their hypothalamic targets that
other PGE2 synthases, notably COX-1-associated cytosolic govern HPA output in a manner closely mimicking the response
PGES, are positioned to support COX-1 involvement. to systemic immune challenge (Ericsson et al., 1997). In addition,

Figure 6. Cellular Mechanisms for

Engaging HPA Control Circuitry by IL-1 or
LPS Challenges
(Top) Reprise of the core circuitry required for acti-
vation of the HPA axis, highlighting projections
from medullary adrenergic and noradrenergic
(blue and red arrows) cell groups to CRF-express-
ing neurosecretory neurons in the PVH.
(Bottom) Vascular mechanisms for transducing
circulating cytokine signals, involving initial moni-
toring by ECs with consequent activation of
PGE2 synthesis/release (green dots) by PVCs
under IL-1 treatment (left) and by both perivascular
and ECs under the more complex stimulus
presented by LPS (right). Factor(s) produced by
PVCs restrain (red arrow) endothelial responsive-
ness under basal and challenge conditions.
Paracrine mediators of bidirectional interactions
between the two cell types remain to be identified.

102 Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc.
Neuron
Vascular Mediation of Inflammatory Effects on CNS

genetic deletion of PGE2 receptor subtypes (EP1 and EP3) atten- latter stages of HPA and febrile responses to LPS are enhanced.
uates HPA responses to LPS (Matsuoka et al., 2003). Across conditions, HPA and febrile responses co-varied closely
with overall vascular COX-2/PGE2 expression, supporting a
Response Specificity of PVC Involvement dynamic, prostanoid-based transduction mechanism at the
The disruptive effect of brain macrophage depletion on IL-1- brain-vascular interface. Further elaboration of this interplay is
induced HPA activation did not generalize to other acute phase expected to clarify central mechanisms of adaptation to acute
responses. The body temperature profile was not affected, save sickness and to provide leverage in the many CNS disease
for an attenuation of an early hypothermic response. In contrast, states in which inflammatory mechanisms are involved.
later phases of LPS-induced fever (stages 2–3 in the terminology
of Romanovsky et al. [2005]) were enhanced. The timing of this EXPERIMENTAL PROCEDURES
effect correlates well with observed elevations in endothelial
COX-2, mPGES-1 expression, and PGE2 levels in Clod-Lip- Animals
Adult male Sprague-Dawley albino rats (260–340 g) were housed under stan-
treated rats, again reflecting PVC-imposed restraint of endothe-
dard vivarium conditions with food and water freely available and were adapted
lial PGE2 production. Our failure to detect any effect of brain to handling for at least 5 days prior to manipulation. All protocols were approved
macrophage depletion on IL-1- or LPS-induced reductions in by the Institutional Animal Care and Use Committee of the Salk Institute.
locomotor activity is consistent with reports that COX inhibitors
affect this response subtly, if at all (Otterness et al., 1991; Wiec- Liposome Preparation and Delivery
zorek and Dunn, 2006). Overall, the results indicate differential Liposomes were prepared as previously described (Van Rooijen and Sanders,
involvement of the mechanisms we describe in acute phase 1994). These are polylamellar phosphatidylcholine-cholesterol membranes
that encapsulate clodronate (used at a concentration of 250 mg/ml), are man-
responses.
nosylated to facilitate receptor-mediated uptake, and are labeled with carbo-
cyanine dye DiI (D282; Molecular Probes) to enable detection of cells that have
Perivascular-Endothelial Cell Interactions incorporated them. For i.c.v. injection, rats were anesthetized with ketamine-
The shift in the EC COX-2/PGE2 production that we describe xylazine-acepromazine (25:5:1 mg/kg, s.c.) and mounted in a stereotaxic
following ablation of an adjacent cell type (PVCs) has an frame. Control (encapsulating PBS) or clodronate-liposomes were equilibrated
intriguing parallel in recent work on cardiac myocytes and fibro- to room temperature, gently shaken to resuspend them, and injected in
blasts (Wang et al., 2009). In the present case, the throttling a volume of 50 ml over 10 min into a lateral ventricle using a 26 gauge needle
mounted onto a stereotaxic arm and attached via PE tubing to a 1000 ml
influence of PVCs is also evident using more generic markers
gastight syringe (Bee Stinger, BAS). Rats were allowed 5–7 days to recover
of EC activation, including the vascular early response gene prior to testing, the point at which depletion of brain macrophages is maximal,
Verge and components of the nuclear factor-kB signaling and prior to repopulation by bone marrow-derived progenitors (Polfliet et al.,
pathway (J.S. and P.E.S., unpublished data). In view of the 2001b; Van Rooijen and Sanders, 1994).
potential for exploiting this mechanism to intervene in inflamma-
tory CNS disease, its molecular underpinnings are of particular Intravenous Administration of IL-1b and LPS
interest. One candidate for fulfilling such a role is nitric oxide; Indwelling jugular catheters (PE 50) containing sterile, heparin-saline (50 U/ml)
were implanted under isoflurane anesthesia (Ericsson et al., 1994). After 2 days
its expression can be induced by LPS or IL-1 in a number of
recovery, awake and freely moving rats were injected with 2 mg/kg of recombi-
tissue macrophage populations and cerebrovascular elements nant rat IL-1b (provided by R. Hart, Rutgers University) or its vehicle (1 ml/kg,
(Wong et al., 1996a, 1996b) and it can inhibit cytokine-stimulated 0.01% BSA, 0.01% ascorbic acid, 10 mM Tris-HCl, and 36 mM sodium phos-
endothelial COX-2 expression (Blais and Rivest, 2001), as well as phate buffer [pH 7.4]) and returned to their home cages. In similar experiments,
HPA activation (Turnbull and Rivier, 1999). Another potential groups of rats were injected with LPS from E. coli at 2.0 mg/kg (Sigma-Aldrich;
mediator is 15-deoxy D12,14 prostaglandin J2 (15d-PGJ2), an serotype 055:B5) or sterile saline (1 ml/kg).
alternate COX-dependent prostanoid product. 15d-PGJ2 has
Perfusion and Histology
been shown capable of suppressing LPS-induced COX-2
At appropriate time points (2–4 hr after injection), rats were anesthetized and
synthesis by serving as an endogenous ligand for the peroxi- perfused via the ascending aorta with 4% paraformaldehyde in borate buffer
some proliferator-activated receptor-g (PPARg) and/or by direct (pH 8.0) at 4 C and the brains were removed, postfixed for 3 hr, and cryopro-
interference with nuclear factor-kB signaling pathway, both of tected overnight. Regularly spaced series of coronal sections (30 mm thick)
which are present and active in the vascular endothelium (Bian- were collected in cryoprotectant solution and stored at 20 C until processing.
chi et al., 2005; Inoue et al., 2000; Simonin et al., 2002). Inhibitory
effects of 15d-PGJ2 on LPS-induced fever have been described Immunohistochemistry
Immunolocalization was achieved using conventional avidin-biotin immuno-
(Mouihate et al., 2004), though it remains unclear whether this or
peroxidase (Vectastain Elite kit; Vector Laboratories) and dual indirect immu-
other PPARg ligands are produced in meaningful concentrations nofluorescence methods (Sawchenko et al., 1990), the latter employing fluo-
in the cellular context of interest here. rescein-, Cy5-, and/or Alexa 488-conjugated secondary antisera (Molecular
Probes). Combined immunoperoxidase staining with dual fluorescence or
Concluding Remarks with in situ hybridization was carried out using protocols described previously
The results indicate dual, time-dependent roles of brain resident (Chan et al., 1993; Ericsson et al., 1994).
Descriptions of all but one of the antisera used are provided as Supple-
macrophages in CNS responses to inflammatory insults. These
mental Experimental Procedures. To localize PGE2-ir, a rabbit polyclonal anti-
cells are required for full manifestation of HPA responses to
serum raised against PGE2 (Oxford Biomedical Research) was used. This
IL-1 challenge, and this requirement exhibits acute phase serum recognizes PGE2 and associated pathway-specific arachidonate deriv-
response specificity. This same cell type serves normally to atives, including PGE1, PGA2, and PGB2, but does not cross-react with other
restrain endothelial prostanoid production; in its absence, the prostanoids. After screening a range of fixation protocols in sections from rat

Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc. 103
 Neuron
Vascular Mediation of Inflammatory Effects on CNS

tissues that express PGE2 constitutively (kidney and gut) and the brains of IL-1- Investigator of the Clayton Medical Research Foundation. Fellowship support
or LPS-treated animals, the following yielded the most sensitive and discrete was provided by the National Institutes of Health (NS10695 to J.C.S. and
localization in the cerebral vasculature. Rats were rapidly anesthetized and DK064086 to T.M.R.), IBRO (J.S.), and the Spanish Ministry of Education
perfused briefly via the ascending aorta with saline (100 ml over 2 min). and Science (J.S. and B.G.B.) and CIBERsam (B.G.B.).
Brains were then extracted, immersed in ice-cold saline for 1 min, and trans-
ected in the coronal plane into forebrain and brainstem blocks, and immersion Accepted: November 20, 2009
fixed overnight in 4% paraformaldehyde in borate buffer (pH 9.5) with 15% Published: January 13, 2010
sucrose added. Adsorption tests were carried out by incubating antiserum
overnight in the presence of graded concentrations of PGE2 (half order of REFERENCES
magnitude increments between 2.8 mM and 2.8 mM) prior to application to
brain sections from IL-1- or LPS-challenged rats. Concentrations R280 mM Bechmann, I., Kwidzinski, E., Kovac, A.D., Simburger, E., Horvath, T., Gimsa,
abolished all detectable labeling. U., Dirnagl, U., Priller, J., and Nitsch, R. (2001a). Turnover of rat brain perivas-
cular cells. Exp. Neurol. 168, 242–249.
Hybridization Histochemistry Bechmann, I., Priller, J., Kovac, A., Bontert, M., Wehner, T., Klett, F.F.,
Techniques for probe synthesis, hybridization, and autoradiographic localiza- Bohsung, J., Stuschke, M., Dirnagl, U., and Nitsch, R. (2001b). Immune surveil-
tion of mRNA signal were adapted from Simmons et al. (1989) and are detailed lance of mouse brain perivascular spaces by blood-borne macrophages. Eur.
as Supplemental Experimental Procedures. J. Neurosci. 14, 1651–1658.
Bernton, E.W., Beach, J.E., Holaday, J.W., Smallridge, R.C., and Fein, H.G.
Electron Microscopy
(1987). Release of multiple hormones by a direct action of interleukin-1 on pitu-
These procedures are described as Supplemental Experimental Procedures.
itary cells. Science 238, 519–521.
Intravenous HRP Injections Bianchi, A., Moulin, D., Sebillaud, S., Koufany, M., Galteau, M.M., Netter, P.,
This procedure is described as Supplemental Experimental Procedures. Terlain, B., and Jouzeau, J.Y. (2005). Contrasting effects of peroxisome-prolif-
erator-activated receptor (PPAR)gamma agonists on membrane-associated
Stress Hormone Assays prostaglandin E2 synthase-1 in IL-1beta-stimulated rat chondrocytes:
Separate groups of animals were implanted with indwelling jugular catheters evidence for PPARgamma-independent inhibition by 15-deoxy-Delta12,
5 days after i.c.v. injection of clodronate or control liposomes and 2 days before 14prostaglandin J2. Arthritis Res. Ther. 7, R1325–R1337.
stress exposure. Blood samples (300 ml) were taken prior to i.v. injection of IL-1 Blais, V., and Rivest, S. (2001). Inhibitory action of nitric oxide on circulating
or LPS at 30 min and then hourly through 6 hr after injection. They were collected tumor necrosis factor-induced NF-kappaB activity and COX-2 transcription
into chilled EDTA-containing tubes and centrifuged; plasma was stored at in the endothelium of the brain capillaries. J. Neuropathol. Exp. Neurol. 60,
20 C until assay. ACTH was measured using a two-site immunoradiometric 893–905.
assay obtained in kit form (DiaSorin), with intra- and inter-assay coefficients Byler, S.L., Boehm, G.W., Karp, J.D., Kohman, R.A., Tarr, A.J., Schallert, T.,
of variation of 3% and 9%, respectively, and a sensitivity of 5 pg/ml. Plasma and Barth, T.M. (2009). Systemic lipopolysaccharide plus MPTP as a model
corticosterone was measured without extraction, using an antiserum raised of dopamine loss and gait instability in C57Bl/6J mice. Behav. Brain Res.
in rabbits against a corticosterone-BSA conjugate and 125I-corticosterone- 198, 434–439.
BSA as tracer (MP Biomedicals). The sensitivity of the assay was 0.8 mg/dl;
Cao, C., Matsumura, K., Yamagata, K., and Watanabe, Y. (1996). Endothelial
intra- and inter-assay coefficients of variation were 5% and 10%, respectively.
cells of the rat brain vasculature express cyclooxygenase-2 mRNA in response
to systemic interleukin-1 beta: a possible site of prostaglandin synthesis
PGE2 Assays
responsible for fever. Brain Res. 733, 263–272.
Tissue PGE2 levels were measured by enzyme immunoassay using reagents
in kit form (Prostaglandin E2 enzyme immunoassay kit-monoclonal; Cayman Chan, R.K.W., Brown, E.R., Ericsson, A., Kovács, K.J., and Sawchenko, P.E.
Chemical Company). Details are given as Supplemental Experimental (1993). A comparison of two immediate-early genes, c-fos and NGFI-B, as
Procedures. markers for functional activation in stress-related neuroendocrine circuitry.
J. Neurosci. 13, 5126–5138.
Telemetry Ching, S., Zhang, H., Belevych, N., He, L., Lai, W., Pu, X.A., Jaeger, L.B., Chen,
Separate groups of rats received i.c.v. injections of control or Clod-Lips and Q., and Quan, N. (2007). Endothelial-specific knockdown of interleukin-1 (IL-1)
5 days later were implanted with jugular catheters for i.v. injection of IL-1, type 1 receptor differentially alters CNS responses to IL-1 depending on its
LPS, or their respective vehicles, as described above. Under the same anes- route of administration. J. Neurosci. 27, 10476–10486.
thetic dose, telemetry transmitters (Mini-Mitter) were implanted intra-abdom- Choi, S.H., Langenbach, R., and Bosetti, F. (2008). Genetic deletion or phar-
inally. Two days later, core body temperature and activity levels were recorded macological inhibition of cyclooxygenase-1 attenuate lipopolysaccharide-
continuously for 24 hr before and after i.v. injection. Transmitter output was induced inflammatory response and brain injury. FASEB J. 22, 1491–1501.
monitored by a receiver board beneath the animals’ cages, which recorded Choi, S.H., Aid, S., and Bosetti, F. (2009). The distinct roles of cyclooxygenase-
core temperature (to ±0.1 C) and episodes (counts) of gross horizontal 1 and 2 in neuroinflammation: implications for translational research. Trends
movement. Pharmacol. Sci. 30, 174–181.
Cunningham, C., Wilcockson, D.C., Campion, S., Lunnon, K., and Perry, V.H.
SUPPLEMENTAL INFORMATION (2005). Central and systemic endotoxin challenges exacerbate the local
inflammatory response and increase neuronal death during chronic neurode-
Supplemental Information includes five figures and Supplemental Experi- generation. J. Neurosci. 25, 9275–9284.
mental Procedures and can be found with this article online at doi:10.1016/
Dantzer, R., and Kelley, K.W. (2007). Twenty years of research on cytokine-
j.neuron.2009.11.032.
induced sickness behavior. Brain Behav. Immun. 21, 153–160.
ACKNOWLEDGMENTS de Vries, H.E., Hoogendoorn, K.H., van Dijk, J., Zijlstra, F.J., van Dam, A.M.,
Breimer, D.D., van Berkel, T.J., de Boer, A.G., and Kuiper, J. (1995). Eicosa-
The authors thank Carlos Arias, Kris Trulock, and Casey Peto for excellent noid production by rat cerebral endothelial cells: stimulation by lipopolysac-
assistance with surgery/histology, imaging/graphics, and electron micros- charide, interleukin-1 and interleukin-6. J. Neuroimmunol. 59, 1–8.
copy, respectively. This research was supported by NS-21182 and was con- Dijkstra, C.D., Dopp, E.A., Joling, P., and Kraal, G. (1985). The heterogeneity
ducted in part by the Clayton Medical Research Foundation. P.E.S. is a Senior of mononuclear phagocytes in lymphoid organs: distinct macrophage

104 Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc.
Neuron
Vascular Mediation of Inflammatory Effects on CNS

subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and Inoue, H., Tanabe, T., and Umesono, K. (2000). Feedback control of
ED3. Immunology 54, 589–599. cyclooxygenase-2 expression through PPARgamma. J. Biol. Chem. 275,
Dinarello, C.A. (1991). Interleukin-1 and interleukin-1 antagonism. Blood 77, 28028–28032.
1627–1652. Ivanov, A.I., and Romanovsky, A.A. (2004). Prostaglandin E2 as a mediator of
Ek, M., Arias, C., Sawchenko, P.E., and Ericsson-Dahlstrand, A. (2000). Distri- fever: synthesis and catabolism. Front. Biosci. 9, 1977–1993.
bution of the EP3 prostaglandin E2 receptor subtype in the rat brain: relation- Katsuura, G., Gottschall, P.E., Dahl, R.R., and Arimura, A. (1988). Adrenocor-
ship to sites of interleukin-1-induced cellular responsiveness. J. Comp. Neurol. ticotropin release induced by intracerebroventricular injection of recombinant
428, 5–20. human interleukin-1 in rats: possible involvement of prostaglandin. Endocri-
Elander, L., Engstrom, L., Ruud, J., Mackerlova, L., Jakobsson, P.J., Engblom, nology 122, 1773–1779.
D., Nilsberth, C., and Blomqvist, A. (2009). Inducible prostaglandin E2 Kitazawa, M., Oddo, S., Yamasaki, T.R., Green, K.N., and LaFerla, F.M. (2005).
synthesis interacts in a temporally supplementary sequence with constitutive Lipopolysaccharide-induced inflammation exacerbates tau pathology by a
prostaglandin-synthesizing enzymes in creating the hypothalamic-pituitary- cyclin-dependent kinase 5-mediated pathway in a transgenic model of
adrenal axis response to immune challenge. J. Neurosci. 29, 1404–1413. Alzheimer’s disease. J. Neurosci. 25, 8843–8853.
Elmquist, J.K., Scammell, T.E., and Saper, C.B. (1997). Mechanisms of CNS Konsman, J.P., Parnet, P., and Dantzer, R. (2002). Cytokine-induced sickness
response to systemic immune challenge: the febrile response. Trends Neuro- behaviour: mechanisms and implications. Trends Neurosci. 25, 154–159.
sci. 20, 565–570. Lacroix, S., and Rivest, S. (1998). Effect of acute systemic inflammatory
Engstrom, L., Rosen, K., Angel, A., Fyrberg, A., Mackerlova, L., Konsman, J.P., response and cytokines on the transcription of the genes encoding cyclooxy-
Engblom, D., and Blomqvist, A. (2008). Systemic immune challenge activates genase enzymes (COX-1 and COX-2) in the rat brain. J. Neurochem. 70,
an intrinsically regulated local inflammatory circuit in the adrenal gland. Endo- 452–466.
crinology 149, 1436–1450. Laflamme, N., and Rivest, S. (2001). Toll-like receptor 4: the missing link of the
Ericsson, A., Kovács, K.J., and Sawchenko, P.E. (1994). A functional anatom- cerebral innate immune response triggered by circulating gram-negative
ical analysis of central pathways subserving the effects of interleukin-1 on bacterial cell wall components. FASEB J. 15, 155–163.
stress-related neuroendocrine neurons. J. Neurosci. 14, 897–913. Letiembre, M., Liu, Y., Walter, S., Hao, W., Pfander, T., Wrede, A., Schulz-
Ericsson, A., Liu, C., Hart, R.P., and Sawchenko, P.E. (1995). Type 1 inter- Schaeffer, W., and Fassbender, K. (2009). Screening of innate immune recep-
leukin-1 receptor in the rat brain: distribution, regulation, and relationship to tors in neurodegenerative diseases: a similar pattern. Neurobiol. Aging 30,
sites of IL-1-induced cellular activation. J. Comp. Neurol. 361, 681–698. 759–768.
Ericsson, A., Arias, C., and Sawchenko, P.E. (1997). Evidence for an intrame- Mackowiak, P.A. (1998). Concepts of fever. Arch. Intern. Med. 158,
dullary prostaglandin-dependent mechanism in the activation of stress- 1870–1881.
related neuroendocrine circuitry by intravenous interleukin-1. J. Neurosci. Matsumura, K., Cao, C., Ozaki, M., Morii, H., Nakadate, K., and Watanabe, Y.
17, 7166–7179. (1998). Brain endothelial cells express cyclooxygenase-2 during lipopolysac-
Fabriek, B.O., Polfliet, M.M., Vloet, R.P., van der Schors, R.C., Ligtenberg, charide-induced fever: light and electron microscopic immunocytochemical
A.J., Weaver, L.K., Geest, C., Matsuno, K., Moestrup, S.K., Dijkstra, C.D., studies. J. Neurosci. 18, 6279–6289.
and van den Berg, T.K. (2007). The macrophage CD163 surface glycoprotein Matsuoka, Y., Furuyashiki, T., Bito, H., Ushikubi, F., Tanaka, Y., Kobayashi, T.,
is an erythroblast adhesion receptor. Blood 109, 5223–5229. Muro, S., Satoh, N., Kayahara, T., Higashi, M., et al. (2003). Impaired adreno-
Garcia-Bueno, B., Serrats, J., and Sawchenko, P.E. (2009). Cerebrovascular corticotropic hormone response to bacterial endotoxin in mice deficient in
cyclooxygenase-1 expression, regulation, and role in hypothalamic-pituitary- prostaglandin E receptor EP1 and EP3 subtypes. Proc. Natl. Acad. Sci. USA
adrenal axis activation by inflammatory stimuli. J. Neurosci. 29, 12970–12981. 100, 4132–4137.

Gehrmann, J., Matsumoto, Y., and Kreutzberg, G.W. (1995). Microglia: intrinsic Mouihate, A., Boisse, L., and Pittman, Q.J. (2004). A novel antipyretic action of
immuneffector cell of the brain. Brain Res. Brain Res. Rev. 20, 269–287. 15-deoxy-Delta12,14-prostaglandin J2 in the rat brain. J. Neurosci. 24, 1312–
1318.
Gosselin, D., and Rivest, S. (2008). MyD88 signaling in brain endothelial cells is
essential for the neuronal activity and glucocorticoid release during systemic Otterness, I.G., Golden, H.W., Seymour, P.A., Eskra, J.D., and Daumy, G.O.
inflammation. Mol. Psychiatry 13, 480–497. (1991). Role of prostaglandins in the behavioral changes induced by murine
interleukin 1 alpha in the rat. Cytokine 3, 333–338.
Hahn, C.N., del Pilar Martin, M., Zhou, X.Y., Mann, L.W., and d’Azzo, A. (1998).
Correction of murine galactosialidosis by bone marrow-derived macrophages Phillis, J.W., Horrocks, L.A., and Farooqui, A.A. (2006). Cyclooxygenases,
overexpressing human protective protein/cathepsin A under control of the lipoxygenases, and epoxygenases in CNS: their role and involvement in neuro-
colony-stimulating factor-1 receptor promoter. Proc. Natl. Acad. Sci. USA logical disorders. Brain Res. Rev. 52, 201–243.
95, 14880–14885. Polfliet, M.M., Zwijnenburg, P.J., van Furth, A.M., van der Poll, T., Dopp, E.A.,
Renardel de Lavalette, C., van Kesteren-Hendrikx, E.M., van Rooijen, N.,
Harbuz, M.S., Conde, G.L., Marti, O., Lightman, S.L., and Jessop, D.S. (1997).
Dijkstra, C.D., and van den Berg, T.K. (2001a). Meningeal and perivascular
The hypothalamic-pituitary-adrenal axis in autoimmunity. Ann. N Y Acad. Sci.
macrophages of the central nervous system play a protective role during
823, 214–224.
bacterial meningitis. J. Immunol. 167, 4644–4650.
Hart, B.L. (1988). Biological basis of the behavior of sick animals. Neurosci.
Polfliet, M.M.J., Goede, P.H., van Kesteren-Hendrikx, E.M.L., van Rooijen, N.,
Biobehav. Rev. 12, 123–137.
Dijkstra, C.D., and van den Berg, T.K. (2001b). A method for the selective
Hawkes, C.A., and McLaurin, J. (2009). Selective targeting of perivascular depletion of perivascular and meningeal macrophages in the central nervous
macrophages for clearance of beta-amyloid in cerebral amyloid angiopathy. system. J. Neuroimmunol. 116, 188–195.
Proc. Natl. Acad. Sci. USA 106, 1261–1266.
Polfliet, M.M., van de Veerdonk, F., Dopp, E.A., van Kesteren-Hendrikx, E.M.,
Hickey, W.F. (2001). Basic principles of immunological surveillance of the van Rooijen, N., Dijkstra, C.D., and van den Berg, T.K. (2002). The role of peri-
normal central nervous system. Glia 36, 118–124. vascular and meningeal macrophages in experimental allergic encephalomy-
Hickey, W.F., and Kimura, H. (1988). Perivascular microglial cells of the CNS elitis. J. Neuroimmunol. 122, 1–8.
are bone marrow-derived and present antigen in vivo. Science 239, 290–292. Priller, J., Flugel, A., Wehner, T., Boentert, M., Haas, C.A., Prinz, M., Fernan-
Holmes, M.C., Antoni, F.A., Aguilera, G., and Catt, K.J. (1986). Magnocellular dez-Klett, F., Prass, K., Bechmann, I., de Boer, B.A., et al. (2001). Targeting
axons in passage through the median eminence release vasopressin. Nature gene-modified hematopoietic cells to the central nervous system: use of green
319, 326–329. fluorescent protein uncovers microglial engraftment. Nat. Med. 7, 1356–1361.

Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc. 105
 Neuron
Vascular Mediation of Inflammatory Effects on CNS

Qin, L., Wu, X., Block, M.L., Liu, Y., Breese, G.R., Hong, J.S., Knapp, D.J., and Turnbull, A.V., and Rivier, C.L. (1999). Regulation of the hypothalamic-pitui-
Crews, F.T. (2007). Systemic LPS causes chronic neuroinflammation and tary-adrenal axis by cytokines: actions and mechanisms of action. Physiol.
progressive neurodegeneration. Glia 55, 453–462. Rev. 79, 1–71.
Quan, N. (2008). Immune-to-brain signaling: how important are the blood-brain Ulevitch, R.J., and Tobias, P.S. (1995). Receptor-dependent mechanisms of
barrier-independent pathways? Mol. Neurobiol. 37, 142–152. cell stimulation by bacterial endotoxin. Annu. Rev. Immunol. 13, 437–457.

Quan, N., Whiteside, M., and Herkenham, M. (1998). Cyclooxygenase 2 mRNA Vallieres, L., and Sawchenko, P.E. (2003). Bone marrow-derived cells that
expression in rat brain after peripheral injection of lipopolysaccharide. Brain populate the adult mouse brain preserve their hematopoietic identity. J. Neu-
Res. 802, 189–197. rosci. 23, 5197–5207.
van der Meer, M.J., Sweep, C.G., Rijnkels, C.E., Pesman, G.J., Tilders, F.J.,
Romanovsky, A.A., Almeida, M.C., Aronoff, D.M., Ivanov, A.I., Konsman, J.P.,
Kloppenborg, P.W., and Hermus, A.R. (1996). Acute stimulation of the hypo-
Steiner, A.A., and Turek, V.F. (2005). Fever and hypothermia in systemic
thalamic-pituitary-adrenal axis by IL-1 beta, TNF alpha and IL-6: a dose
inflammation: recent discoveries and revisions. Front. Biosci. 10, 2193–2216.
response study. J. Endocrinol. Invest. 19, 175–182.
Sawchenko, P.E., Cunningham, E.T., Jr., Mortrud, M.T., Pfeiffer, S.W., and
Van Rooijen, N., and Sanders, A. (1994). Liposome mediated depletion of
Gerfen, C.R. (1990). Phaseolus vulgaris-leucoagglutinin (PHA-L) anterograde
macrophages: mechanism of action, preparation of liposomes and applica-
axonal transport technique. In Methods in Neurosciences, Volume 3, Quanti-
tions. J. Immunol. Methods 174, 83–93.
tative and Qualitative Microscopy, P.M. Conn, ed. (San Diego: Academic
Wang, D., Patel, V.V., Ricciotti, E., Zhou, R., Levin, M.D., Gao, E., Yu, Z.,
Press), pp. 247–260.
Ferrari, V.A., Lu, M.M., Xu, J., et al. (2009). Cardiomyocyte cyclooxygenase-2
Schiltz, J.C., and Sawchenko, P.E. (2002). Distinct brain vascular cell types influences cardiac rhythm and function. Proc. Natl. Acad. Sci. USA 106,
manifest inducible cyclooxygenase expression as a function of the strength 7548–7552.
and natrue of immune insults. J. Neurosci. 22, 5606–5618.
Watanabe, T., Morimoto, A., Sakata, Y., and Murakami, N. (1990). ACTH
Schiltz, J.C., and Sawchenko, P.E. (2003). Signaling the brain in systemic response induced by interleukin-1 is mediated by CRF secretion stimulated
inflammation: the role of perivascular cells. Front. Biosci. 8, S1321–S1329. by hypothalamic PGE. Experientia 46, 481–484.
Schiltz, J.C., and Sawchenko, P.E. (2007). Specificity and generality of the Webster, J.I., Tonelli, L., and Sternberg, E.M. (2002). Neuroendocrine regula-
involvement of catecholaminergic afferents in hypothalamic responses to tion of immunity. Annu. Rev. Immunol. 20, 125–163.
immune insults. J. Comp. Neurol. 502, 455–467. Wick, G., Hu, Y., Schwarz, S., and Kroemer, G. (1993). Immunoendocrine
Sehic, E., Szekely, M., Ungar, A.L., Oladehin, A., and Blatteis, C.M. (1996). communication via the hypothalamo-pituitary-adrenal axis in autoimmune
Hypothalamic prostaglandin E2 during lipopolysaccharide-induced fever in diseases. Endocr. Rev. 14, 539–563.
guinea pigs. Brain Res. Bull. 39, 391–399. Wieczorek, M., and Dunn, A.J. (2006). Relationships among the behavioral,
noradrenergic, and pituitary-adrenal responses to interleukin-1 and the effects
Shanks, N., Larocque, S., and Meaney, M.J. (1995). Neonatal endotoxin expo-
of indomethacin. Brain Behav. Immun. 20, 477–487.
sure alters the development of the hypothalamic-pituitary-adrenal axis: early
illness and later responsivity to stress. J. Neurosci. 15, 376–384. Williams, K., Alvarez, X., and Lackner, A.A. (2001). Central nervous system
perivascular cells are immunoregulatory cells that connect the CNS with the
Simmons, D.M., Arriza, J.L., and Swanson, L.W. (1989). A complete protocol
peripheral immune system. Glia 36, 156–164.
for in situ hybridization of messenger RNAs in brain and other tissues with
radiolabeled single-stranded RNA probes. J. Histotechnol. 12, 169–181. Wong, M.L., Bongiorno, P.B., al-Shekhlee, A., Esposito, A., Khatri, P., and Lic-
inio, J. (1996a). IL-1 beta, IL-1 receptor type I and iNOS gene expression in rat
Simonin, M.A., Bordji, K., Boyault, S., Bianchi, A., Gouze, E., Becuwe, P., brain vasculature and perivascular areas. Neuroreport 7, 2445–2448.
Dauca, M., Netter, P., and Terlain, B. (2002). PPAR-gamma ligands modulate
Wong, M.L., Rettori, V., al-Shekhlee, A., Bongiorno, P.B., Canteros, G.,
effects of LPS in stimulated rat synovial fibroblasts. Am. J. Physiol. Cell
McCann, S.M., Gold, P.W., and Licinio, J. (1996b). Inducible nitric oxide syn-
Physiol. 282, C125–C133.
thase gene expression in the brain during systemic inflammation. Nat. Med.
Stoll, G., and Jander, S. (1999). The role of microglia and macrophages in the 2, 581–584.
pathophysiology of the CNS. Prog. Neurobiol. 58, 233–247.
Zhang, Y.H., Lu, J., Elmquist, J.K., and Saper, C.B. (2003). Specific roles of
Thomas, W.E. (1999). Brain macrophages: on the role of pericytes and perivas- cyclooxygenase-1 and cyclooxygenase-2 in lipopolysaccharide-induced
cular cells. Brain Res. Brain Res. Rev. 31, 42–57. fever and Fos expression in rat brain. J. Comp. Neurol. 463, 3–12.

106 Neuron 65, 94–106, January 14, 2010 ª2010 Elsevier Inc.