| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Laboratory of Molecular Endocrinology, CHUL Research Center and Department of Anatomy and Physiology, Laval University, Sainte-Foy, Québec, Canada G1V 4G2
Address all correspondence and requests for reprints to: Dr. Serge Rivest, Laboratory of Molecular Endocrinology, CHUL Research Center, 2705 Laurier boulevard, Sainte-Foy, Québec, Canada G1V 4G2. E-mail: serge.rivest{at}crchul.ulaval.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(TNF)
(1, 2, 3). Among their central effects, the induction of fever and the
activation of the hypothalamic-pituitary-adrenal (HPA) axis have been
widely investigated because of their major impacts on the inflammatory
processes (4, 5, 6). IL-6 has been defined as one of the principal endogenous pyrogens from the observations that IL-6-deficient mice are unable to develop normal fever in response to both LPS and IL-1ß (7). Based on physiological and neuroanatomical studies, it has been proposed that IL-6 may induce fever by signaling thermoregulatory neurons of the anterior hypothalamic area through PG-dependent pathways (8, 9, 10). On the other hand, the relative contribution of IL-6 to modulating the HPA axis during endotoxemia and its exact mechanisms of action are not yet fully understood. It has been shown that IL-6 can increase glucocorticoid production by stimulating the secretion of ACTH from the anterior pituitary (11, 12) and by acting directly on the adrenal gland (13, 14, 15). Although IL-6 can also trigger the release of CRF into the hypophyseal portal system (16, 17, 18), this cytokine seems unable, in contrast to IL-1ß, to induce neuronal activation and CRF gene transcription in the hypothalamic paraventricular nucleus (PVN) (10, 19), a region that contains neuroendocrine CRF neurons. This lack of effect might, however, be explained by the fact that IL-6 receptor (IL-6R) is not expressed in this nucleus under basal conditions (1). It is plausible that the induction of IL-6R synthesis may be an essential step taking place early during inflammation to allow IL-6, when it becomes available in the blood circulation, to trigger neural responses. We, therefore, hypothesized that this event may potentiate and amplify the action of systemic IL-6 on the CRF-containing neurons that control the HPA axis.
Besides fever and CRF secretion, IL-6 may also influence a
variety of other neural functions during endotoxemia. This attractive
possibility is suggested by the observations that IL-6 is synthesized
by the circumventricular organs and secreted into cerebrospinal fluid
in response to systemic LPS administration (1, 20) and that an
intracerebroventricular injection of IL-6 induces widespread cellular
activation in the brain, particularly over the ependymal cell layer,
the ventricular walls, and the meninges (10). One possible role of IL-6
is the modulation of certain properties of the endothelial and
ependymal cells, which express the IL-6R subunits (1) and constitute
natural barriers against immunological substances present in the
circulation and cerebrospinal fluid, respectively. It has been
demonstrated that IL-6 participates in the development of reactive
gliosis affecting both microglia and astrocytes (21, 22, 23). This response
may be essential to eliminate and restrain the penetration of
circulating LPS into the nervous tissue. Furthermore, new findings
suggest that IL-6 may play neuroprotective functions, as, for example,
the repression of TNF-induced neurotoxic inflammation and the
enhancement of nerve growth factor signaling (24, 25).
In this study, we have evaluated the contribution of IL-6 in stimulating neural activity and CRF gene transcription in rodent brains during endotoxemia. Given that IL-6R has been shown to be up-regulated in response to endotoxin (1), we first characterized its expression pattern in the brain and verified whether CRF neurons of the PVN produced this receptor after different treatments with LPS and IL-6. To verify the possibility that preinduction of IL-6R synthesis may increase the influence of IL-6 on neuroendocrine functions, we further compared the effects of a systemic injection of IL-6 on the PVN when given under basal conditions and 6 h after pretreatment with low dose LPS. To confirm and complement the results obtained with this approach, we finally used IL-6-deficient mice to analyze the role of IL-6 in the activation of neural cells, a phenomenon commonly observed in the PVN and elsewhere in the brain during endotoxemia. We provide here evidence that IL-6 modulates, under LPS challenge, the expression of its own receptor, the neuroendocrine CRF, and the protooncogene c-fos, an immediate early gene used as a marker of cellular activation. In addition, we demonstrate that IL-6 is unable to stimulate the gene encoding the enzyme cyclooxygenase-2 (COX-2) normally required for PG production.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
iv injection of LPS and IL-6 in rats
A catheter containing sterile pyrogen-free heparin-saline (5.0
U/ml) was implanted into the right jugular vein of rats anesthetized
with an ip injection (300 µl) of a mixture of ketamine hydrochloride
(91 mg/ml) and xylazine (9 mg/ml). The catheter was made from a piece
of SILASTIC brand tubing (SILASTIC medical grade tubing, 0.50 mm
i.d., 0.94 mm o.d.; Dow Corning Corp., Midland, MI)
connected to Intramedic polyethylene tubing (id, 0.58 mm; od, 0.965 mm;
PE-50, Clay Adams, Parsippany, NJ). The internal SILASTIC tip of the
catheter was positioned at the atrium, and the outlet was placed at an
interscapular position and sealed. After the surgery, the animals were
kept individually for a 5-day recuperation period.
On the day of the experiment (0900 h), the outlet of the catheter was
fastened to a truncated 27-gauge needle attached to PE-50 tubing. This
connector was then fixed to a 1-ml syringe, and the rats were placed
individually in a quiet room for at least 2 h before
experimentation. This procedure was used to avoid disturbing the
animals during the injections. Rats received first an iv injection of
either LPS (5 µg dissolved in 300 µl 0.9% sterile pyrogen-free
saline; from Escherichia coli, serotype 055:B5, catalog no.
L2880, Sigma Chemical Co., Oakville, Canada) or vehicle
solution. The animals were conscious and freely moving during the
procedure. After a 6-h waiting period, they received a second iv
injection of either recombinant rat IL-6 (1.5 µg dissolved in 300
µl sterile pyrogen-free distilled water; bioactivity, 
1 x
107 U/mg; endotoxin, <0.1 ng/µl; catalog no. PRC0065,
lot H032506, BioSource Technologies, Inc., Camarillo, CA)
or vehicle solution. The animals were killed 1 h after the second
injection with an iv injection of ketamine-xylazine. Tissue processing
was performed as described below.
ip injection of LPS in mice
In the morning (at 
0830 h), IL-6 knockout and wild-type mice
received an ip injection of LPS (25 µg dissolved in 100 µl 0.9%
sterile pyrogen-free saline). All animals were conscious and freely
moving during the procedure and were kept individually in their cages.
The mutant mice seemed less affected by LPS, showing higher locomotor
activity than wild-type controls. The animals were killed 1, 4, and
8 h after LPS administration with an ip injection of
ketamine-xylazine. Additional mice were not injected with LPS and were
used as controls (time zero). Tissue processing was performed as
described below.
Tissue processing
All animals were rapidly perfused transcardially with 0.9%
saline followed by ice-cold 4% paraformaldehyde in 100 mM
borax buffer (pH 9.5). Their brains were removed, postfixed with fresh
4% paraformaldehyde for 2–5 days at 4 C, and placed in the same
fixative containing 10% sucrose overnight at 4 C. They were mounted
onto a microtome (Reichert-Jung, Cambridge Instrument Co., Deerfield,
IL), frozen with dry ice, and cut into 30-µm (rats) or 20-µm (mice)
coronal sections from the olfactory bulb to the caudal medulla. The
slices were collected in a cold cryoprotectant solution [50
mM sodium phosphate buffer (pH 7.3), 30% ethylene glycol,
and 20% glycerol] and stored at -20 C until histochemical
analysis.
Riboprobe synthesis and preparation
Details of the plasmids and enzymes used for complementary RNA
(cRNA) probe synthesis are presented in Table 1
. Rat c-fos complementary DNA
(cDNA) was obtained from Dr. I. Verma (The Salk Institute, La Jolla,
CA), rat CRF cDNA from Dr. K. Mayo (Northwestern University, Evanston,
IL), rat intronic CRF cDNA from Dr. S. Watson (University of Michigan,
Ann Arbor, MI), rat COX-2 cDNA from Dr. K. Peri (Sainte-Justine
Hospital Research Center, Montréal, Canada), and rat IL-6 cDNA
from American Type Culture Collection (Manassas, VA). All
protocols were adapted from Simmons et al. (27). Sense and
antisense radioactive cRNA copies were synthesized by incubating 250 ng
linearized plasmid in 6 mM MgCl2, 40
mM Tris (pH 7.9), 2 mM spermidine, 10
mM NaCl, 10 mM dithiothreitol, 0.2
mM ATP/GTP/CTP, [-35S]UTP, 40 U RNAsin
(Promega Corp., Madison, WI), and 20 U of the appropriate
RNA polymerase (Table 1) for 60 min at 37 C. The unincorporated
nucleotides were removed by adding 100 µl deoxyribonuclease solution
(1 µl deoxyribonuclease, 5 µl 5 mg/ml transfer RNA, and 94 µl 10
mM Tris-10 mM MgCl2) for 10 min,
followed by phenol-chloroform extraction. The probes were precipitated
with 80 µl 5 M ammonium acetate and 500 µl 95% ethanol
for 20 min on dry ice. After centrifugation, the pellets were washed
with 500 µl 70% ethanol, dried, and resuspended in 100 µl 10
mM Tris/1 mM EDTA. The probes (107
cpm/ml) were mixed into an appropriate volume of hybridization solution
[822 µl solution 1 (500 µl formamide; 60 µl 5 M
NaCl; 10 µl 1 M Tris, pH 8.0; 2 µl 0.5 M
EDTA, pH 8.0; 20 µl 50 x Denhart’s solution; 200 µl 50%
dextran sulfate; and 30 µl diethylpyrocarbonate (depc) water), 50
µl 10 mg/ml transfer RNA, and 10 µl 1 M dithiothreitol
(118 µl depc water - volume of probe used)] and heated for 5
min at 65 C before being spotted on the slides (100 µl). Radioactive
sense cRNA copies (controls) were also prepared to verify the
specificity of each probe.
|
Dual labeling
Immunohistochemistry and in situ hybridization were
combined to determine whether CRF neurons of the PVN expressed IL-6R
mRNA. Hypothalamic sections were stained for CRF using the
avidin-biotin amplification bridge method with peroxidase as a
substrate. The slices were washed in sterile depc-treated KPBS at room
temperature and incubated for 2 h with rabbit antihuman/rat CRF
serum (diluted 1:20,000; code PBL rC70; provided by Dr. W. Vale,
Peptide Biology Laboratory, The Salk Institute, La Jolla, CA) mixed in
KPBS containing 0.4% Triton X-100 and 0.25% heparin sodium salt USP.
The sections were then washed in KPBS and incubated for 1 h with
biotinylated goat antirabbit IgG (diluted 1:1500; Vector Laboratories, Inc. Burlingame, CA) mixed in KPBS containing
0.4% Triton X-100 and 0.25% heparin sodium salt USP. The slices were
then rinsed with KPBS and incubated for 1 h with an
avidin-biotin-peroxidase complex (Vectastain ABC elite
kit, Vector Laboratories, Inc., Burlingame, CA). After
several washes, the brain slices were reacted in KPBS containing 0.05%
3,3'-diaminobenzidine tetrahydrochloride and 0.003% hydrogen peroxide
(H2O2). Thereafter, the slices were rinsed with
KPBS, mounted onto gelatin- and poly-L-lysine-coated
slides, and vacuum-dried. In situ hybridization was
performed as described above to localize IL-6R mRNA, except that the
steps with alcohol were shortened (10 dips) to avoid decoloration of
immunostained cells.
Blood sampling and corticosterone measurement in rats
Blood was collected quickly by cardiac puncture before
perfusion, transferred in EDTA tubes (Becton Dickinson and Co., Franklin Lakes, NJ), and centrifuged for 10 min at 3000 rpm
(4 C). Plasma was collected and stored at -20 C until the analysis.
Corticosterone levels were measured by RIA (Immuchem corticosterone RIA
kit for rats, ICN Biomedicals, Inc., Costa Mesa, CA)
according to the manufacturer’s protocol. The intra- and interassay
coefficients of variation were both 7%.
OD analysis
Semiquantitative analysis of hybridization signals for
c-fos, CRF, hnCRF, and IL-6R mRNAs was carried out on x-ray
films over at least two hypothalamic paraventricular nuclei (bilateral)
for each animal. Transmittance values (referred here as OD) of the
hybridization signals were measured on a Northern Light Desktop
Illuminator (Imaging Research, Inc.) using a Sony Camera
Video System fastened to a MicroNikkor 55-mm Vivitar extension tube
set for Nikon lens and coupled to a Macintosh computer
(Power Macintosh 7100/66) and the NIH Image software version 1.59/ppc
(written by W. Rasband at the NIH and available from the internet by
anonymous ftp from zippy.nih.gov). The OD values for each pixel were
calculated using a known standard of intensity and distance
measurements from a logarithmic specter adapted from Bioimage Visage
110s (Millipore Corp., Ann Arbor, MI). The wedge was
calibrated before correcting for film saturation, which was determined
by sampling darkest PVNs and adjusting the light source and exposure
time. All samples displaying clear positive signals were evaluated
within the linear OD curve to avoid pixel saturation and
underestimation. Sections from experimental and control animals were
digitized and subjected to densitometric analysis, yielding
measurements of mean density per area. The OD of each side of the PVN
(bilateral) was then corrected for the average background signal by
subtracting the OD of areas without positive signal located immediately
outside the digitized nucleus. Because of the lack of basal expression
of c-fos mRNA and CRF hnRNA in the PVN, the stained slides
were used to delineate the nucleus on the film. Data are expressed as
the mean ± SEM and were analyzed by two-way ANOVA
followed by the Bonferroni/Dunn test used for post-hoc
comparisons when appropriate.
Qualitative analysis
Anatomical identification of brain structures was based on the
Swanson’s atlas (28). The relative intensity of c-fos mRNA
signals throughout mouse brains was assessed on x-ray film images and
graded according to the scale of undetectable (-), low (+), moderate
(2+), strong (2++), or very strong (2+2+).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, Veh+Veh). This transcript, however,
was undetectable or very slightly expressed in the PVN of
vehicle-treated animals. The IL-6R mRNA levels increased 7 h after
LPS administration in the OVLT, CC, BNST, PVN, and CeA (Fig. 1, LPS+Veh). Interestingly, IL-6 was also capable of activating the gene
encoding its receptor. This effect was restricted to blood vessels
after a single injection of IL-6 (Fig. 1, Veh+IL-6, and Fig. 2), but extended to the above-mentioned
structures when the cytokine was injected 6 h after LPS
administration (Fig. 1, LPS+IL-6). Indeed, IL-6 injection induced a
more pronounced expression of its receptor in several regions of the
parenchyma of rats pretreated with the endotoxin. In particular, the OD
analysis of hybridization signals revealed that the IL-6R mRNA levels
in the PVN were significantly higher (P < 0.05) after
treatment with LPS and IL-6 together than after LPS or IL-6 alone (Fig. 3). Adjacent sections hybridized with a
sense probe did not exhibit positive signal in any of the regions
expressing the transcript with the antisense probe.
|
|
|
) led us to believe that
CRF-containing cells might express this receptor. The dual labeling
experiment revealed that few CRF-labeled perikarya were positive for
the IL-6R transcript in response to LPS and IL-6 (Fig. 4C). However,
the IL-6R transcript was very sensitive to the immunohistochemical
procedures, making evaluation of the exact proportion of double labeled
cells difficult. The possibility that other cell types, such as
microglia, express this receptor cannot be ruled out.
|
). Moderate
levels of c-fos mRNA were still observed in the PVN 7 h
after LPS injection, whereas CRF hnRNA was barely detectable.
Confirming our hypothesis, both transcripts were significantly
(P < 0.05) higher in the PVN of rats treated with both
LPS and IL-6 than in the animals treated with either LPS or IL-6 alone
(Fig. 5, A and B). Plasma corticosterone levels were also significantly
higher in animals presensitized and receiving a single iv bolus of
recombinant IL-6 than vehicle- and LPS-injected rats (Fig. 5C).
However, the Bonferroni post-hoc analysis did not reveal a
significant difference between the circulating corticosterone
concentrations of animals that were injected only with IL-6 (Veh+IL-6)
and those of vehicle-treated rats (Veh+Veh).
|
IL-6 is necessary for the prolonged stimulation of neural activity
during endotoxemia
It has been commonly observed that LPS induces widespread neural
activation in both rat and mouse brains (30). To verify the
contribution of IL-6 to this response and to confirm its involvement in
stimulating the PVN during endotoxemia, we compared the effects of LPS
on c-fos and CRF expression in the brains of IL-6-deficient
and wild-type mice. Table 2 describes the
qualitative analysis of hybridization signals for c-fos mRNA
in mice killed before or 1, 4, and 8 h after LPS administration.
As shown in Fig. 6A, the
distribution and the intensity of c-fos mRNA hybridization
signals were similar in wild-type and knockout mice under basal
conditions (0 h). One hour after LPS administration, increased levels
of c-fos mRNA were observed in several nuclei without a
difference between knockout mice and their controls, namely the CC,
BNST, medial preoptic area (MPO)/OVLT, SFO, PVN, supraoptic nucleus
(SON), ME, CeA, dorsomedial hypothalamic nucleus, locus coeruleus (LC),
AP, nucleus of the solitary tract, blood vessels, choroid plexus,
leptomeninges, and parabrachial nucleus (Fig. 6A). Interestingly, we
found that c-fos mRNA expression was markedly reduced in
knockout mice 4 h after LPS injection, but remained elevated in
several brain regions of wild-type animals (Fig. 6B). These regions
were the ependymal cell layer, which did not express c-fos
mRNA at any time in IL-6-deficient mice, the PVN, CC, LC, MPO/OVLT, ME,
leptomeninges, SFO, submeningual zone (sm), and SON (Table 2). In the
PVN, the OD analysis indicated that the mRNA encoding c-fos
was indeed significantly higher (P < 0.05) in
wild-type mice than in knockout animals at 4 h post-LPS injection
(Fig. 7A). In wild-type mice,
c-fos mRNA levels decreased and reached levels observed in
knockout animals at 8 h postinjection, except in the locus
coeruleus, where the message remained elevated (Fig. 6B).
|
|
|
).
Statistical analysis of OD data showed a significant difference
(P = 0.0365) between the IL-6-deficient and control
mice. Indeed, the CRF mRNA levels in the PVN were generally higher in
wild-type than in knockout mice. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
IL-6 signaling in rat brain
Our previous work (1) together with the present results suggest
that neural cells become more sensitive to IL-6 during systemic immune
challenges by increasing the number of receptor molecules on their
surface. Although this modification seems to be essential, additional
physiological changes might be required to allow such a permissive
effect. The shedding of soluble IL-6R, the alteration of the
blood-brain barrier (BBB), and the activation of transduction
pathways cooperating with those solicited by IL-6 are examples of
mechanisms that may amplify, widen, and prolong the IL-6 activities. On
the one hand, these changes may be initiated by the direct action of
LPS on endothelial and parenchymal cells after binding to its specific
receptor CD14. Indeed, we have recently demonstrated that CD14 is
distributed in brain regions accessible from the blood circulation
under basal conditions and that scattered microglial cells produce this
receptor within a few hours after LPS administration (31). Although the
endothelial cells do not express CD14, they can respond to circulating
endotoxin in a soluble CD14-dependent manner (32, 33, 34). On the other
hand, the enhancement of IL-6 signaling may be induced by
proinflammatory cytokines released by macrophages and glial cells in
response to LPS. It has been shown that systemically injected IL-1ß
is capable of stimulating IL-6R synthesis in cerebral blood vessels (1)
and that TNF can modulate the expression of both IL-6R and gp130
mRNAs in cultured neurons (35). Moreover, as observed in the present
study, IL-6 can increase the levels of its receptor within perivascular
and parenchymal elements of the brain, an effect consistent with the
observation that IL-6 can up-regulate its receptor in the liver
(36).
IL-6 and the HPA axis
The presence of IL-6 is essential for coordinating a variety
of physiological responses during infections, but overexposition of the
nervous tissue to IL-6 induces a pronounced reactive gliosis and leads
to neurodegeneration associated with impaired learning capacities (22, 37, 38). Appropriate modulation of IL-6 signaling is thus not only
important for enhancing the effects of IL-6, but also to avoid its
deleterious actions. It is well known that IL-6 stimulates the HPA
axis, which provides negative feedback signals acting to reduce the
immune response in general and the secretion of IL-6 and other
proinflammatory cytokines in particular. IL-6 is thought to be a long
term regulator of plasma ACTH and glucocorticoid levels by acting
directly on the anterior pituitary (11, 12) and the adrenal gland
(13, 14, 15), respectively. Moreover, IL-6 can exert an acute stimulatory
effect on the HPA axis by acting on the hypothalamus to induce
infundibular CRF secretion (16, 17, 18). However, the importance of IL-6 in
this process was recently challenged by the demonstration that iv
injected IL-6 is unable by itself to activate PVN neurons and to
stimulate CRF gene transcription (10, 19), contrasting with the
capacity of IL-1ß (1) and TNF (Nadeau, S., and S. Rivest,
unpublished results) to induce these changes. We prove here that IL-6
can, in fact, activate PVN neurons, but the cytokine seems to require
prior activation of the immune response and the synthesis of its
receptor. The present study also demonstrated that the increase in
plasma glucocorticoid levels at 1 h after iv IL-6 injection is
significant only in animals that were pretreated 6 h before with a
very low dose of LPS. It is interesting to note that coadministration
of IL-6 and IL-1ß results in synergistic activation of ACTH and
corticosterone release (39, 40, 41, 42). Although these effects may be caused
by the interaction of both cytokines in the peripheral glands and
despite IL-1ß is unable to induce IL-6R expression in the PVN (1), it
cannot be excluded that IL-1ß and IL-6 may cooperate in the brain to
modulate the HPA axis or other nervous functions. It will be exciting
in future work to characterize clearly the nature of these
interactions, but the fact that circulating IL-1ß is capable of
inducing IL-6R in blood vessels of the brain may be a crucial event
leading to such permissive action of the proinflammatory cytokine
(1).
Possible mechanisms of action
The exact mechanisms by which IL-6 stimulates PVN neurons
remain to be elucidate. The presence of IL-6R in some CRF neurons
offers the possibility that IL-6 may directly target these cells.
Alternatively, IL-6 may act on neurons devoid of IL-6R, but expressing
the signal transducer gp130, after binding its soluble receptor. At
first sight, these possibilities seem to be conflicting with the
current view that circulating cytokines cannot cross the BBB, but a
number of mechanisms may explain this discrepancy. First, PVN neurons
project to the median eminence and to the neurohypophysis, thereby
exposing their axon terminals to molecules circulating into the
bloodstream or secreted locally from cells of myeloid lineage. It has
been proposed that IL-6 may act within the median eminence to trigger
the release of CRF into the hypophyseal portal system (43). This is
supported by our recent observations that median eminence expresses the
IL-6R subunits and that iv injected IL-6 induces c-fos
expression in this region (1, 10). Second, excessive nitric oxide and
IL-6 production caused by endotoxin can alter cerebral endothelium
functions and disrupt the BBB (44). The resulting increase of
permeability may then allow high weight molecules, including cytokines,
to reach sites behind the BBB, but such a mechanism has been challenged
by numerous studies. Third, it has been demonstrated that a small
quantity of IL-6 can penetrate across the intact BBB via a transport
system distinct from those for IL-1ß and TNF (45). Some brain
areas may, therefore, be equipped with an as yet uncharacterized active
transport for IL-6, which might be activated under stressful
circumstances. Forth, the choroid plexus is a cerebrospinal
fluid-synthesizing structure that has been pointed out as one of the
principal sources of central IL-6 during endotoxemia (1). After being
released into the ventricular system, centrally produced IL-6 may
circulate throughout the brain and reach PVN neurons through passive
diffusion.
In addition to the above mechanisms and given that many CRF neurons
seem not to express IL-6R, it is conceivable that IL-6 may stimulate
intermediate cells, such as endothelial or microglial cells, to release
paracrine factors that, in turn, could influence neuronal activity. As
potential candidates, PGs and nitric oxide are key mediators in
neuroimmune communication, and their production can be controlled by
IL-1ß and TNF (46, 47, 48, 49). Unfortunately, depending on the
experimental procedure and the tissue studied, there are as many
studies that prove their involvement in mediating the central effects
of IL-6 as there are studies that are in disagreement with it. For
examples, it has been demonstrated that PGs mediate IL-6-induced fever
(8, 9, 50) and HPA axis stimulation (16, 17), but IL-6 is unable to
activate PG formation in cerebral microvessels (51) and to induce COX-2
mRNA synthesis in rat brains (29) or in cultured microglial cells (52).
It is clear, however, that IL-6 does not stimulate the production of
PGs in peripheral organs, and that, conversely, its own synthesis is
induced by them (53, 54). In agreement with the latter, it has recently
been shown that IL-6 expression in astrocytes can be induced in
vitro by PGE2 (55). Taking together, these
observations suggest that IL-6 does not stimulate COX-2 gene
transcription, but the possibility that IL-6 may influence PG synthesis
at posttranscriptional levels or may cooperate with them to activate
CRF neurons cannot be ruled out. On the other hand, much less
information is available regarding the effect of IL-6 on nitric oxide
production. It has been shown that IL-6 stimulates nitrite formation in
cultured hippocampal slices (56). However, IL-6-induced glial cell
growth (57) and neuroblastoma cell differentiation (49) are not
mediated by nitric oxide, and transgenic mice overexpressing IL-6 in
the brain do not manifest any increase in inducible nitric oxide
synthase gene expression (22). In light of the above observations, the
roles of PGs and nitric oxide in mediating the effects of IL-1ß and
TNF- cannot be generalized to IL-6, and more detailed studies will
be needed to clarify the mechanisms by which this cytokine may affect
the neuroendocrine response.
Relative importance of IL-6 during endotoxemia
It is generally believed that proinflammatory cytokines released
by macrophages during endotoxemia are mainly responsible for the
stimulation of the HPA axis. The relative contribution of IL-6 in this
process is currently the subject of intense debates and investigations.
Accumulating data are, however, in disagreement with this view and
rather suggest that LPS may directly act within the hypothalamus to
activate PVN neurons without any help of systemic cytokines during the
early phases. Indeed, the levels of plasma ACTH and corticosterone are
increased in response to LPS before the elevation of circulating
IL-1ß, TNF, and IL-6 (58). It is therefore not surprising that
pretreatment with IL-1ß antagonist or anti-TNF antibody does not
abolish the LPS-induced HPA axis and that IL-6-deficient mice produce
normal corticosterone levels during the first hour after LPS injection
(41, 59, 60, 61). Thereafter, the increase in circulating corticosterone
levels has recently been found to be lower in IL-6-/-
than IL-6+/+ mice in response to ip LPS injection, but not
during restraint stress (62). This suggests that the involvement of
IL-6 in the control of the HPA axis is quite specific to the immune
stimuli and not to neurogenic stresses. The participation of IL-6 is
further suggested by the fact that pretreatment with anti-IL-6 antibody
abrogates ACTH secretion both 2 and 4 h after LPS administration,
but not at 1 h (41). This cytokine may be considered as a crucial
modulator of the HPA axis, as anti-IL-6 antibody also abolishes the
IL-1-induced increase in plasma ACTH (40). Interestingly, this
determining role has been confirmed in models using inflammatory agents
lacking the intrinsic capacity to stimulate the HPA axis, but able to
induce cytokine production. It was concluded that IL-6 is an obligate
factor to increase glucocorticoid production during cytomegalovirus
infection or after the injection of a synthetic analog of viral nucleic
acid (62).
The present study brings some light on the mechanisms by which IL-6 contributes to the regulation of the neuroendocrine response by demonstrating that IL-6 is required during the later phases of endotoxemia to sustain the activated state of CRF neurons controlling the HPA axis. It is interesting to mention that IL-6 can also induce the secretion of arginine vasopressin (AVP) (63, 64), an important ACTH and corticosteroid secretagogue, and that chronic cerebral expression of IL-6 modulates the stress-induced increase in plasma corticosterone via a mechanism involving AVP (65). Given that the activation of neural cells located in both parvocellular and magnocellular PVN was observed in the present study after cotreatment with LPS and IL-6, it is plausible that IL-6 stimulates both CRF and AVP neuroendocrine systems during endotoxemia. An important point to address in future investigations will be determination of the exact proportions of CRF-, AVP- and CRF/AVP-activated neurons during the acute phase response in both wild-type and IL-6-deficient mice.
Besides controlling the HPA axis, IL-6 induces a series of other brain-mediated responses and participates in the modulation of local inflammatory events. Over the past years, we used the protooncogene c-fos to identify within the brain the sites of action of the cytokine. We found that CVOs, the meninges, the ependymal cell layer covering the ventricular spaces, and the portion of parenchyma surrounding the ventricles are profoundly activated after an intracerebroventricular injection of IL-6 (10). More recently, we reported that IL-6 is not synthesized in the brain under basal conditions, but is rapidly expressed by the choroid plexus and other CVOs in response to systemically injected endotoxin. Furthermore, the present study shows that LPS-induced c-fos expression is markedly reduced in the brain of knockout animals, especially in structures that were responding to the intracerebroventricular IL-6 treatment. Taken together, these observations suggest that IL-6 may be secreted by the choroid plexus into cerebrospinal fluid, which diffuses across the brain through the ventricles. This may allow the cytokine to stimulate various neural cells, principally those associated with or close to the ventricular system. The presence of IL-6R in parvocellular neurons of the PVN is of great interest, as it suggests that IL-6 of central origin may bind to its cognate receptor directly within the endocrine hypothalamus. Such a mechanism obviously does not take place in the brain of IL-6-deficient mice, which may contribute to the reduced activity of PVN neurons during endotoxemia.
Conclusion
In summary, the following sequence of events presumably takes
place in the central nervous system during endotoxemia. Bacterial
endotoxin reaches the brain through the circulation and acts directly
on nonparenchymal and parenchymal structures accessible from the blood,
which leads to the activation of vascular-associated cells and
development of the inflammatory response. PVN neurons are also
activated in presence of LPS and respond by increasing the
transcription of the gene encoding CRF, the neurosecretagogue that
controls the HPA axis. A variety of neural cells become gradually more
sensitive to IL-6, mainly by expressing more receptor molecules on
their surface. When it becomes available, IL-6 maintains the activation
state of the hypothalamic CRF neurons and participates in specific
cellular responses necessary for eliminating the endotoxin. Centrally
produced IL-6 may also play local roles in regulating inflammatory
processes and can serve as a neuroprotective signal after being
released into cerebrospinal fluid. As a consequence, the inflammation
can be properly controlled by the inhibitory effects of
glucocorticoids on cytokine expression.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Supported by Ph.D. studentships from the Natural Sciences and
Engineering Research Council of Canada and the Fonds pour la Formation
des Chercheurs et l’Aide à la Recherche du Québec.
Presently a postdoctoral fellow at The Salk Institute (La Jolla,
CA).
3 Medical Research Council Scientist.
Received December 1, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, -1ß, 2 and 6 on ACTH
synthesis and release in the mouse pituitary tumour cell line AtT-20. J
Endocrinol 122:33–39, IL-1ß or IL-6
on hypothalamo-pituitary-adrenal axis activation. J Endocrinol 133:349–355 (TNF-).
Brain Res 706:71–79[CrossRef][Medline]
(TNF-),
interferon-
, and interleukin-6 but not TNF-ß induce
differentiation of neuroblastoma cells: the role of nitric oxide.
J Neurochem 62:1330–1336[Medline]
in the neurochemical and neuroendocrine responses
to endotoxin. Brain Res Bull 29:807–812[CrossRef][Medline]
in the rat. Neuroimmunomodulation 2:137–140[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  K. Kageyama, K. Hanada, Y. Iwasaki, and T. Suda Regulation and role of suppressor of cytokine signaling-3 in hypothalamic 4B cells J. Endocrinol., June 1, 2009; 201(3): 369 - 376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Nilsberth, L. Elander, N. Hamzic, M. Norell, J. Lonn, L. Engstrom, and A. Blomqvist The Role of Interleukin-6 in Lipopolysaccharide-Induced Fever by Mechanisms Independent of Prostaglandin E2 Endocrinology, April 1, 2009; 150(4): 1850 - 1860. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kageyama, K. Hanada, Y. Iwasaki, S. Sakihara, T. Nigawara, J. Kasckow, and T. Suda Pituitary adenylate cyclase-activating polypeptide stimulates corticotropin-releasing factor, vasopressin and interleukin-6 gene transcription in hypothalamic 4B cells J. Endocrinol., November 1, 2007; 195(2): 199 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Dominguez-Punaro, M. Segura, M.-M. Plante, S. Lacouture, S. Rivest, and M. Gottschalk Streptococcus suis Serotype 2, an Important Swine and Human Pathogen, Induces Strong Systemic and Cerebral Inflammatory Responses in a Mouse Model of Infection J. Immunol., August 1, 2007; 179(3): 1842 - 1854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. G. Helwig, T. I. Musch, R. A. Craig, and M. J. Kenney Increased interleukin-6 receptor expression in the paraventricular nucleus of rats with heart failure Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2007; 292(3): R1165 - R1173. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Rummel, C. Sachot, S. Poole, and G. N. Luheshi Circulating interleukin-6 induces fever through a STAT3-linked activation of COX-2 in the brain Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2006; 291(5): R1316 - R1326. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. L. Sparkman, J. B. Buchanan, J. R. R. Heyen, J. Chen, J. L. Beverly, and R. W. Johnson Interleukin-6 facilitates lipopolysaccharide-induced disruption in working memory and expression of other proinflammatory cytokines in hippocampal neuronal cell layers. J. Neurosci., October 18, 2006; 26(42): 10709 - 10716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Wernstedt, A. Edgley, A. Berndtsson, J. Faldt, G. Bergstrom, V. Wallenius, and J.-O. Jansson Reduced stress- and cold-induced increase in energy expenditure in interleukin-6-deficient mice Am J Physiol Regulatory Integrative Comp Physiol, September 1, 2006; 291(3): R551 - R557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Komori, S. Miyahara, M. Yamamoto, T. Matsumoto, K. Zhang, M. Nakagawa, S. Nomura, E. Motomura, T. Shiroyama, and Y. Okazaki Effects of Odorants on the Hypothalamic-Pituitary-Adrenal Axis and Interleukin-6 (IL-6) and IL-6 Receptor mRNA Expression in Rat Hypothalamus after Restraint Stress Chem Senses, November 1, 2003; 28(9): 767 - 771. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Beishuizen and L. G. Thijs Review: Endotoxin and the hypothalamo-pituitary-adrenal (HPA) axis Innate Immunity, February 1, 2003; 9(1): 3 - 24. [Abstract] [PDF]
|
|
|
|
|
|
S. Rivest Editorial: Does Circulating Leptin Have the Ability to Cross the Blood-Brain Barrier and Target Neurons Directly? Endocrinology, September 1, 2002; 143(9): 3211 - 3213. [Full Text] [PDF]
|
|
|
|
|
|
V. Chesnokova and S. Melmed Minireview: Neuro-Immuno-Endocrine Modulation of the Hypothalamic-Pituitary-Adrenal (HPA) Axis by gp130 Signaling Molecules Endocrinology, May 1, 2002; 143(5): 1571 - 1574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M. Johnson, S. K. Vogt, M. W. Burney, and L. J. Muglia COX-2 inhibition attenuates anorexia during systemic inflammation without impairing cytokine production Am J Physiol Endocrinol Metab, March 1, 2002; 282(3): E650 - E656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Vallieres, I. L. Campbell, F. H. Gage, and P. E. Sawchenko Reduced Hippocampal Neurogenesis in Adult Transgenic Mice with Chronic Astrocytic Production of Interleukin-6 J. Neurosci., January 15, 2002; 22(2): 486 - 492. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Marks, N. Ling, and R. D. Cone Role of the Central Melanocortin System in Cachexia Cancer Res., February 1, 2001; 61(4): 1432 - 1438. [Abstract] [Full Text]
|
|
|
|
|
|
E. LeBel, L. Vallieres, and S. Rivest Selective Involvement of Interleukin-6 in the Transcriptional Activation of the Suppressor of Cytokine Signaling-3 in the Brain during Systemic Immune Challenges Endocrinology, October 1, 2000; 141(10): 3749 - 3763. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Rivest, S. Lacroix, L. Vallières, S. Nadeau, J. Zhang, and N. Laflamme How the Blood Talks to the Brain Parenchyma and the Paraventricular Nucleus of the Hypothalamus During Systemic Inflammatory and Infectious Stimuli Experimental Biology and Medicine, January 1, 2000; 223(1): 22 - 38. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
