| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Potentiates Interleukin (IL)-6 and Tumor Necrosis Factor-
But Not IL-1ß Induced by Endotoxin in the Brain
Istituto di Ricerche Farmacologiche Mario Negri, 20157 Milan, Italy
Address all correspondence and requests for reprints to: Dr. Maria Grazia De Simoni, Istituto Mario Negri, Via Eritrea 62, 20157 Milan, Italy. E-mail: de-simoni{at}irfmn.mnegri.it
| Abstract |
|---|
|
|
|---|
(IFN
) is present in the central nervous
system during neurologic diseases associated with inflammation, its
effect on endotoxin-induced cytokines was studied. Cerebrospinal fluid
(CSF) and serum levels of interleukin (IL)-1ß, IL-6, and tumor
necrosis factor-
(TNF
), their messenger RNA expression in brain
areas (hypothalamus, hyppocampus, and striatum) and in spleen were
evaluated 2 and 8 h after endotoxin [lipopolysaccharide (LPS), 25
µg/rat icv], IFN
(2.5 µg/rat icv) or after their
coadministration in rats. CSF and serum IL-1ß levels were increased
by LPS alone and IFN
coadministration did not furtherly increase
them. IFN
potentiated LPS effect on IL-6 and TNF
levels in both
CSF and serum. LPS and IFN-
coadministration did not alter IL-1ß
messenger RNA expression induced by LPS in brain areas and in spleen,
but it potentiated that of IL-6 and TNF
. The present in
vivo data show that icv coadministration of LPS and IFN
results in a potentiation of cytokine production (IL-6 and TNF
)
which may trigger a cascade of events relevant to neurodegenerative
processes. This action is independent of IL-1ß because the production
of this cytokine is not altered by IFN
treatment. | Introduction |
|---|
|
|
|---|
The brain constitutively expresses inflammatory cytokines such as
interleukin (IL)-1ß, IL-6, and tumor necrosis factor-
TNF
, and
several in vitro and in vivo animal models show
that different stimuli activate microglia and astrocytes resulting in
overproduction of these cytokines (14, 15). Another cytokine for which
a role has been proposed in neurodegenerative diseases is
interferon-
(IFN
) (12, 16, 17, 18, 19, 20). Unlike the cytokines mentioned
above, it is generally believed that there is no endogenous cell source
of IFN
in the normal brain and this cytokine is present in the CNS
only during disease states when activated T cells or macrophages have
infiltrated the brain. In these conditions, high levels of IFN
have
been reported in brain tissue although the actual amount has not been
measured (16, 21). Little is known of IFN
actions in the brain
although it is believed to activate a local immune response. In
vitro studies have shown that IFN
can modulate gene expression
in astrocytes and microglia (16, 17, 22). Moreover, it can induce
expression of both class II MHC antigens and ICAM-1 on astrocytes and
microglia, which can contribute to the ability of these cells to act as
antigen-presenting cells in the CNS and stimulate aberrant immune
responses within this site (16).
The present study evaluates whether in vivo, in the CNS,
similarly to what observed in the periphery, IFN
may act as an
activating signal for inflammatory cytokine production. The effect of
IFN
intracerebroventricular (icv) administration was investigated on
IL-1ß, IL-6 and TNF
induced by endotoxin [lipopolysaccharide
(LPS), a potent trigger of these cytokines (23, 24, 25)]. CSF levels of
these cytokines as well as their messenger RNA (mRNA) expression in
brain areas (striatum, hypothalamus, and hippocampus) were evaluated.
Moreover, because central LPS efficiently induces inflammatory
cytokines in the periphery, serum production of the same cytokines as
well as their expression in spleen, were also investigated.
| Materials and Methods |
|---|
|
|
|---|
LPS administration
One polyethylene cannula was permanently implanted in the
lateral ventricle (23) 3 days before treatment. LPS (25 µg/6
µl·rat; phenol-extracted preparation from Escherichia
coli 055: B5, Sigma Chemical Co., St. Louis, MO), dissolved in
sterile pyrogen-free saline, rat recombinant IFN
(2.5 µg/6
µl·rat corresponding to 2.75 x 104 U; Roussel
Uclaf, Romainville, France), dissolved in 0.1% BSA, and vehicle were
administered icv. Data refer to one of three independent experiments,
yielding the same results.
Because an in vivo central action of IFN
has not been
widely investigated before and our aim was to explore a possible
pathogenetic role of cytokines, we employed relatively high doses of
LPS and IFN
. In previous in vivo studies, similar
concentrations of IFN
were used to produce inflammatory conditions
(26). High LPS doses have been used to mimic stressful conditions (27)
or to activate cytokine production and the HPA axis (25).
CSF, serum, and tissue samples
Two and eight hours after treatment, rats were anesthetized with
chloral hydrate, 350 mg/kg ip, and CSF (60100 µl) was drawn from
the cisterna magna using a glass capillary with a tip of approximately
300 µm. A careful surgery allowed to avoid blood contamination.
Immediately after, the rats were decapitated and trunk blood was
collected in sterile tubes. Serum was prepared by centrifugation at
15,000 x g for 5 min. CSF and serum were divided up
for IL-1ß enzyme-linked immunosorbent assay (ELISA), IL-6 and TNF
bioassays. Samples were stored at -20 C until assay.
Spleen and brain areas (hypothalamus, hippocampus, and striatum) were dissected. Tissues were immediately frozen on dry ice and kept at -20 C until mRNA extraction for Northern blot analysis.
IL-1ß ELISA
IL-1ß was measured in CSF and in serum by a two-site ELISA as
described by Garabedian et al. (28), except that the color
was developed using avidin-peroxidase (Sigma) and 2,
2-azino-di[3-ethyl-benzthiazoline sulfonate (6)] peroxidase
substrate system (KPL, Gaithersburg, MD) as chromogen. Absorbance was
read at 405 nm. The detection limit was 3.9 pg/ml.
IL-6 bioassay
IL-6 was measured in CSF and in serum as hybridoma growth factor
using the 7TD1 as previously described (29). Results are expressed as
units/ml in comparison with a reference curve obtained in each
experiment using recombinant human IL-6 (Immunex, Seattle, WA).
Reference curves obtained were comparable in all experiments. One unit
in the 7TD1 assay corresponded to 1 pg human recombinant IL-6. The
sensitivity of the assay with rat serum is 50 U/ml.
TNF
bioassay
CSF and serum TNF
was measured by cytotoxicity on L929 cells
in the presence of 1 µg/ml of actinomycin D as previously described
(30). TNF
levels were calculated using recombinant human TNF
(BASF/Knoll, Ludwigshafen, Germany; specific activity 107
U/mg) and expressed as ng/ml. The sensitivity of the bioassay was 5
pg/ml.
RNA extraction and Northern blot analysis
RNA was extraced from tissue samples according to the acid
guanidinium-phenol-chlorophorm procedure described previously (23).
Total RNA was separated on 1.2% agarose-formaldehyde gels and
transferred to Nylon 66 filters (Gene Screen Plus, Dupont, Raketstraat,
Belgium). Based on spectrophotometric analysis, an equal amount of
total RNA was applied to each lane (1520 µg). The membranes were
hybridized with the following probes (23): 1) ß-actin mRNA probe
corresponding to a 0.8-kb fragment from a human complementary DNA
(cDNA) clone (31); 2) IL-1ß mRNA probe corresponding to a 1.3-kb
fragment from mouse cDNA (32); 3) IL-6 mRNA probe corresponding to a
0.65-kb fragment from mouse cDNA (33); 4) TNF
mRNA probe
corresponding to a 0.45-kb from human cDNA (34). Probes were labeled
using a randomly primed DNA labeling kit from Amersham and
32P-dCTP. All labeled probes were purified through Quick
Spin Columns (Boehringer Mannheim SpA, Monza, Italy). After O/N
hybridization with the appropriate 32P-labeled cDNA probe,
blots were exposed to x-ray films at -80 C with intensifying screens,
for the time needed to obtain a signal in a linear range. The exposure
time for a given probe was the same in all experiments. Densitometric
analysis of autoradiograms was done with an IBAS 2 image analyzer
(Kontron-Zeiss, Milano, Italy), integrating the optical density with
the area of the hybridized bands (35). The signal associated with the
presence of ß-actin mRNA was used as an internal standard to
normalize IL-1ß, IL-6, and TNF
expression.
The IL-6 probe identifies two different mRNAs of 1.2- and 2.4-kb (22, 25). The 1.2-kb band is predominant in either the CNS or spleen, whereas the 2.4 kb band was not always detectable in our experiments. We therefore evaluated only the signal corresponding to the 1.2-kb transcript.
| Results |
|---|
|
|
|---|
affected LPS-induced cytokine
production, the two drugs were administered alone or simultaneously and
CSF and serum IL-1ß, IL-6 and TNF
levels were measured. Neither
saline nor IFN
administered alone induced detectable cytokine
levels. As shown in Fig. 1
did not further increase IL-1ß production in
both CSF and serum. The results in Fig. 2
coadministration potentiated LPS-induced IL-6. This
effect was observed at 8 h in the CSF, whereas it was maximal at
2 h but still present at 8 h, in serum. We next investigated
whether IFN
increased LPS-induced TNF
. As shown in Fig. 3
induced an increase in LPS-induced TNF
both in the CSF (Fig. 3A
|
|
|
affected LPS-induced cytokine
expression, the two drugs were administered alone or simultaneously and
IL-1ß, IL-6, and TNF
mRNA were measured. Neither saline nor IFN
administered alone induced a detectable signal in the brain areas
considered and in spleen (Fig. 4
were injected
simultaneously, no differences were observed in IL-1ß mRNA expression
(Fig. 5
coadministration (Fig. 6
mRNA expression was
significantly increased by LPS and IFN
coadministration in brain
areas and in spleen (Fig. 7
|
|
|
|
| Discussion |
|---|
|
|
|---|
, IL-6, and IL-1ß were clearly detectable in CSF. Differently
from TNF
, whose turnover is more rapid, high levels of IL-6 and
IL-1ß were still present 8 h after injection. Although the
cytokines considered are constitutively expressed in the brain (14),
the sensitivity of the Northern blot analysis did not allow to detect a
signal in control or IFN
-treated rats. LPS clearly induced mRNA
expression for IL-1ß, IL-6, and TNF
in the three brain areas
considered. Except for TNF
expression in hypothalamus, the induction
of mRNA was still evident 8 h after the treatment, in some cases
the induction being higher than at the shorter time. It may only be
hypothesized that the somewhat different time-course of cytokine
induction in the three brain areas is related to a specific, unexplored
role of the cytokine in that area (23).
For each cytokine, the pattern of mRNA induction and that of its
presence in CSF were similar. This indicates a close temporal relation
between expression and production suggesting that cytokines are
synthesized, rapidly secreted, and accumulated in the CSF. It should
only be mentioned that the time-course of TNF
in CSF is better
explained by the pattern of mRNA expression in hypothalamus than by the
other two brain areas considered.
The present data confirm and extend previous findings showing that
centrally administered LPS efficiently induces inflammatory cytokines
not only in the brain tissue but also in the periphery. Because LPS is
a lipophilic molecule, it may leak from the brain to the periphery.
However, the observation that icv LPS induces more marked and lasting
effects in the brain than the periphery indicated that it does not
leave the central compartment in significant amounts (36). What was
previously observed for IL-6 and IL-1ß (23) also applies to TNF
.
Icv LPS in fact caused rapid induction of the three cytokines
considered, evident on the expression and on synthesis, with a
time-course different from that in the brain. Whereas the effect in the
CNS was slower and longer lasting, it was rapid and transient in the
periphery, suggesting different regulatory mechanisms.
IFN
effect on LPS-induced cytokines
The present data show for the first time in vivo that
IFN
acts in the brain as an activating signal for brain cells,
rendering them more responsive to LPS stimuli. Because IFN
induced
no change in the cytokines considered or in their message, we assumed
that no significant contamination was present. Although IFN
did not
induce cytokine synthesis and production when administered alone, it
markedly potentiated LPS action. This was evident on IL-6 and TNF
expression and production, in the brain and in the periphery.
Unlike IL-6 and TNF
, IFN
was completely ineffective in modulating
IL-1ß expression and production in the brain and periphery,
indicating that this cytokine is regulated independently from the other
two. However, synergy with IL-1ß might be achieved at lower doses of
LPS or IFN
. Although IL-6 and TNF
induction by LPS may be
mediated by IL-1ß, the results suggest that IFN
stimulates IL-6
and TNF
synthesis independently from IL-1ß.
Much of the knowledge regarding the mechanism triggered by IFN
and
LPS comes from studies on immune cells. In monocytes and macrophages
IFN
was reported to interact directly with the CD14, the LPS
receptor (37). This does not seem to be the case in the present
experimental model because in that case LPS and IFN
coadministration
should cause similar changes in all the cytokines considered. Studies
on gene regulation in macrophages activated by IFN
and LPS have
shown that IFN
enhances LPS-initiated transcription of TNF
and
not its stability (38, 39). The present observation that increased mRNA
accumulation closely corresponds to increase in the protein synthesis
indicates that the steady-state of TNF
and IL-6 message is not
affected.
IFN
s effect on brain cells has been the subject of several
in vitro studies. IFN
has been proposed to inhibit
LPS-induced IL-1ß and IL-6 (40) or to potentiate TNF
induction
(18) in microglial cells. Other reports indicate that exposure of
astrocytes to IFN
renders them responsive to a suboptimal dose of
LPS resulting in a significant stimulation of TNF
production (17).
Actually, the cell source of cytokines produced in the present animal
model is still to be clarified.
The pathogenetic role of IFN
in neurologic diseases is not known.
High levels of IFN
are found in the brain tissue of multiple
sclerosis patients (7, 8, 9). The possible relevance of IFN
in the
pathogenesis of Alzheimers disease is proposed in studies showing
that ß-amyloid synergizes with IFN
in triggering the production of
reactive nitrogen intermediates and TNF
in microglial cells and in
mediating neuronal injury (18, 19). IFN
is present in the CNS during
other neurological diseases associated with inflammation including
viral meningitis, encephalitis, HIV dementia, and in animal models of
CNS disease such as experimental allergic encephalomyelitis (15).
Associated with these neurological diseases is an inflammatory
infiltrate in the CNS composed of activated macrophages, B and T
lymphocytes, the last cell type being the potential local source of
endogenous IFN
.
There is increasing evidence that inflammatory cytokines including
TNF
and IL-6 play a pathogenetic role in CNS conditions. For
example, anti-TNF
antibodies are protective in animal models of
cerebral malaria (41). TNF
is a potent stimulus for nitric oxide, a
mediator of ischemic brain damage (42, 43, 44). Transgenic mice
overexpressing IL-6 or TNF
in the brain show neurodegeneration,
astrocytosis, microgliosis, demyelinization, and macrophage
accumulation (45, 46). The present data, showing that IFN
administration in the brain results in a potentiation of IL-6 and
TNF
production, suggest that IFN
in the CNS may trigger a cascade
of events relevant to neurodegenerative processes.
| Acknowledgments |
|---|
Received May 9, 1997.
| References |
|---|
|
|
|---|
immunoreactivity. J Neurochem 63:18721879[Medline]
(TNF-
) increases both
in the brain and in the cerebrospinal fluid from parkinsonian patients.
Neurosci Lett 165:208210[CrossRef][Medline]
, and IL-1ß in chronic
progressive multiple sclerosis: expected lack of clinical utility.
Neurology 41:121123
- and
IFN-
-mediated signal transduction pathways: effects on glial cell
gene expression and fuction. FASEB J 9:15771584[Abstract]
production by astrocytes induction by lipopolysaccharide,
IFN-
, and IL-1ß. J Immunol 144:29993007[Abstract]
. Nature 374:647650[CrossRef][Medline]
. Brain Res 692:207214[CrossRef][Medline]
and tumor necrosis factor-
on
oligodendroglial lineage cell proliferation and differentiation
in vitro. Eur J Neurosci 8:11061116[CrossRef][Medline]
and interleukin-4 down-regutale soluble CD14 release in
human monocytes and macrophages. J Leukoc Biol 52:323330[Abstract]
and
lipopolysaccharide. J Leukoc Biol 48:412419[Abstract]
on LPS-induced interleukin 1ß production by isolated
adult rat brain microglia. Neurochem Int 29:7782[CrossRef][Medline]
. Proc Natl Acad Sci USA 92:1129411298This article has been cited by other articles:
![]() |
G. J. Paterson, Y. Ohashi, G. P. Reynolds, J. A. Pratt, and B. J. Morris Selective increases in the cytokine, TNF{alpha}, in the prefrontal cortex of PCP-treated rats and human schizophrenic subjects: influence of antipsychotic drugs J Psychopharmacol, September 1, 2006; 20(5): 636 - 642. [Abstract] [PDF] |
||||
![]() |
H. Shimizu, M. Miyoshi, K. Matsumoto, O. Goto, T. Imoto, and T. Watanabe The Effect of Central Injection of Angiotensin-Converting Enzyme Inhibitor and the Angiotensin Type 1 Receptor Antagonist on the Induction by Lipopolysaccharide of Fever and Brain Interleukin-1{beta} Response in Rats J. Pharmacol. Exp. Ther., March 1, 2004; 308(3): 865 - 873. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. F Smith and C. L Darlington Recent developments in drug therapy for multiple sclerosis Multiple Sclerosis, April 1, 1999; 5(2): 110 - 120. [Abstract] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |