Negative Feedback Between Prostaglandin
and α- and β-Chemokine Synthesis in Human
Microglial Cells and Astrocytes
This information is current as
of June 17, 2017.
Permissions
Email Alerts
J Immunol 1999; 162:1701-1706; ;
http://www.jimmunol.org/content/162/3/1701
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 1999 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
Subscription
Nazila Janabi, Isabelle Hau and Marc Tardieu
Negative Feedback Between Prostaglandin and a- and
b-Chemokine Synthesis in Human Microglial Cells and
Astrocytes1
Nazila Janabi,2,3 Isabelle Hau, and Marc Tardieu
I
nflammatory processes resulting from an intricate network of
mediators play a central role in several central nervous system (CNS)4 diseases. The proinflammatory cytokines induce
astrocytes and microglial cells to secrete several inflammatory mediators such as chemokines, lipid mediators, and free radicals (1–
6). Chemokines are involved in the activation and recruitment of
monocytes and lymphocytes into the brain parenchyma (7). Products of arachidonic acid metabolism include prostanoids (derived
from the cyclooxygenase pathway) that are involved in the control
of body temperature, neuroendocrine function, sleep, cerebral ischemia, brain edema, and inflammation (8 –12). Prostanoids are also
implicated in chemotaxis as well as in the modulation of blood
flow and vascular permeability (12, 13). Although these inflammatory mediators are all constitutively expressed in brain parenchyma for regulation of normal CNS activities, their production
has to be under strict control to prevent inflammatory tissue damage. Thus, cascade production of cytokines, positive and negative
feedback, and synergistic mechanisms are parameters that characterize the action of these mediators in the inflammatory process.
The role of chemokines and PGs in several biological effects
induced by proinflammatory cytokines, as well as the presence of
common events in the pathways of synthesis and signal transduction of these two classes of mediators such as phospholipase A2
activation and intracellular calcium mobilization (14, 15),
prompted us to study a possible interaction between chemokine
and PG pathways of synthesis. In this report, we demonstrate that
human microglia and astrocytes produce a- and b-chemokines
(IL-8, growth related proteina (GROa), RANTES, microphage inflammatory protein (MIP)-1a, and MIP-1b) and PGs under proinflammatory conditions. The production of b-chemokines in these
cells is negatively regulated by their PG secretion, whereas PG
production itself is limited by a-chemokine secretion.
Materials and Methods
Primary cultures of human CNS cells
Laboratory of Virus, Neuron and Immunity, Unité de Formation et de Recherche,
Kremlin Bicêtre, University of Paris-South, Paris, France
Received for publication July 21, 1998. Accepted for publication October 16, 1998.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This work was supported by Institut National de la Santé et de la Recherche Médicale Research Contract 96-12, by grants from the Agence Nationale de Recherche sur
le Sida, Sidaction and Université Paris XI, and by fellowships (to N.J. and I.H.) from
the Fondation pour la Recherche Médicale.
2
Address correspondence and reprint requests to Dr. Nazila Janabi, Laboratoire “Virus, Neurone et Immunité,” Faculté de Médecine, 63 rue Gabriel Péri, 94276 Le
Kremlin Bicêtre Cedex, France.
3
Present address: Laboratory of Molecular Medicine and Neuroscience, National
Institutes of Health, 9000 Rockville Pike, Bethesda, MD 20892.
Eight- to 10-wk-old human embryos were obtained after elective abortion
in compliance with the recommendations of the French National Ethics
Committee and following approval by the local ethics committee. Primary
cultures of spinal cord and prosencephalon were prepared as described (16,
17). Briefly, tissues were dissected, trypsinized, and resuspended in MCDB
153 medium (Polylabo, Paris, France) completed by 5% FCS, 2 mM glutamine, 105 U/liter penicillin, and 0.1 g/liter streptomycin. The cell suspension was seeded at a density of 5 3 105 cells per well (24-well plates,
Nunc, Roskilde, Denmark) coated with collagen (100 mg/ml). Cultures
were kept at 37°C in a 10% CO2 atmosphere. The medium was changed
completely after 48 h and then by half every 3 days. These cultures consisted of neurofilament-positive neuronal clusters lying on a monocellular
layer containing 40% glial fibrillary acidic protein (GFAP)-positive astrocytes and 40% CD68-positive microglial cells associated with ,10% fibronectin-positive fibroblasts (16 –18).
Purified cultures of human microglial cells and astrocytes
4
Abbreviations used in this paper: CNS, central nervous system; MIP, microphage
inflammatory protein; GRO, growth related protein; GFAP, glial fibrillary acidic protein; MCP-1, monocyte chemotactic protein-1.
Copyright © 1999 by The American Association of Immunologists
To obtain purified microglial and astrocyte cultures, CNS cells were seeded
at a density of 2 3 106 cells per well in 6-well plates. After plating (10 –15
0022-1767/99/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
The understanding of immune surveillance and inflammation regulation in cerebral tissue is essential in the therapy of neuroimmunological disorders. We demonstrate here that primary human glial cells were able to produce a- and b-chemokines (IL-8 >
growth related protein a (GROa) .. RANTES > microphage inflammatory protein (MIP)-1a and MIP-1b) in parallel to PGs
(PGE2 and PGF2a) after proinflammatory cytokine stimulation: TNF-a 1 IL-1b induced all except RANTES, which was induced
by TNF-a 1 IFN-g. Purified cultures of astrocytes and microglia were also induced by the same combination of cytokines, to
produce all these mediators except MIP-1a and MIP-1b, which were produced predominantly by astrocytes. The inhibition of PG
production by indomethacin led to a 37– 60% increase in RANTES, MIP-1a, and MIP-1b but not in GROa and IL-8 secretion.
In contrast, inhibition of IL-8 and GRO activities using neutralizing Abs resulted in a specific 6-fold increase in PGE2 but not in
PGF2a production by stimulated microglial cells and astrocytes, whereas Abs to b-chemokines had no effect. Thus, the production
of PGs in human glial cells down-regulates their b-chemokine secretion, whereas a-chemokine production in these cells controls
PG secretion level. These data suggest that under inflammatory conditions, the intraparenchymal production of PGs could control
chemotactic gradient of b-chemokines for an appropriate effector cell recruitment or activation. Conversely, the elevated intracerebral a-chemokine levels could reduce PG secretion, preventing the exacerbation of inflammation and neurotoxicity. The
Journal of Immunology, 1999, 162: 1701–1706.
1702
CHEMOKINE AND PROSTAGLANDIN RETROREGULATIONS IN HUMAN GLIA
FIGURE 1. a- and b-chemokine and PG
production by human mixed glial cultures. A,
IL-8 and GROa production tested at day 3
poststimulation with IFN-g, IL-1b, and TNF-a
at 200 U/ml, or a combination of all of them. B
and C, Accumulation of RANTES, MIP-1a,
and MIP-1b in supernatants of human glial
cells, tested at day 3 poststimulation with
proinflammatory cytokines. D, PGF2a and
PGE2 accumulated during the 3 days of stimulation of glial cells with the same combinations of cytokines. Mean 6 SD of three to five
independent experiments performed in triplicate, using cells from three to six different
brain specimens.
Cell stimulations and reagents
To stimulate cultures, cells were refed with serum-free MCDB 153 medium containing the tested inducers, inhibitors, or Abs. IL-1b (200 U/ml)
was purchased from Genzyme (Cambridge, MA). TNF-a (200 U/ml) and
IFN-g (200 U/ml) were purchased from Boehringer Mannheim (Maylan,
France). All are recombinant human cytokines. Human rIL-8, GROa (500
ng/ml), and neutralizing Abs against human RANTES (polyclonal, 200
mg/ml), MIP-1a and MIP-1b (polyclonal, 100 mg/ml) (19 –22), IL-8, and
GRO (monoclonal, 100 mg/ml) were purchased from R&D Systems (Minneapolis, MN). The specific inhibitor for cyclooxygenase (indomethacin,
5–15 mM) was purchased from Sigma (St. Louis, MO). PGF2a and PGE2
were purchased from Cayman Chemicals (Ann Arbor, MI; PGF2a and
PGE2, 16 ng/ml) (23).
Chemokine and PG titrations
The concentrations of all tested mediators were measured by enzyme immunoassay (PGF2a and PGE2, Cayman Chemicals; a- and b-chemokines,
R&D Systems). All assays were performed using standards and instructions supplied by the manufacturers.
Statistical analysis
Each measurement was performed at least in triplicate. Student’s t test was
used for determination of statistical significance of two to five independent
experiments. Results were considered significant at p # 0.05.
Results
Induction of a- and b-chemokine production by proinflammatory
cytokines in human glial cultures
Spontaneous IL-8 and GROa production was detected in the supernatant of mixed glial cultures. The effect of proinflammatory
cytokines on this production was tested in the absence of serum:
IL-1b and, to a lesser extent, TNF-a, but not IFN-g alone led to an
increase in IL-8 and GROa production by glial cells at day 3
poststimulation (Fig. 1A). However, a combination of TNF-a 1
IL-1b was most efficient at inducing a-chemokine production
(IL-8: 290,435 6 33,483 vs 2, 704 6 1,590 pg/ml; GROa: 64,
978 6 10, 884 vs 1, 516 6 977 pg/ml, (TNF-a 1 IL-1b)-stimulated vs control, n 5 5, p , 0.0001, Fig. 1A). RANTES secretion
by mixed glial cultures was also induced in the presence of proinflammatory cytokines although the induction profile of the cytokine combinations differed (IFN-g 1 TNF-a . IFN-g 1 IL-1b .
TNF-a 1 IL-1b, Fig. 1B). IL-1b, TNF-a, or IFN-g alone had no
significant effect on RANTES production by these cells. The level
of RANTES production by glial cells was at least 10-fold lower
than that of a-chemokines (4,958 6 1, 190 vs 7 6 0.3 pg/ml,
(IFN-g 1 TNF-a)-stimulated vs control, day 3, n 5 3, p 5 0.001).
The same cytokines induced production of MIP-1a and MIP-1b in
mixed glial cultures (TNF-a 1 IL-1b . IFN-g 1 IL-1b 5 IFN-g
1 TNF-a, Fig. 1C), which nonetheless remained low compared
with RANTES (MIP-1a: 881 6 88 vs 11 6 1 pg/ml; MIP-1b: 1,
345 6 159 vs 36 6 7 pg/ml, (TNF-a 1 IL-1b)-stimulated vs
control, day 3, n 5 3, p , 0.0001). Kinetics of IL-8, GROa,
RANTES, MIP-1a, and MIP-1b production by cytokine-stimulated mixed glial cells demonstrated a similar profile of chemokine
induction starting between 6 and 24 h poststimulation and accumulating over the 3 days tested (Fig. 2). All purified cultures of
both microglia and astrocytes tested secreted high levels of IL-8,
GROa, and RANTES, respectively, after TNF-a 1 IL-1b and
IFN-g 1 TNF-a stimulation (Fig. 3, A and B). In contrast, only
two of six purified cultures of microglia could be induced by
TNF-a 1 IL-1b to produce MIP-1a, and three of six produced
MIP-1b, whereas four of five purified cultures of astrocytes were
induced to secrete both MIP-1a and MIP-1b under the same condition of stimulation (Fig. 3C). Accordingly, MIP-1a and MIP-1b
secretion was induced more efficiently in purified cultures of astrocytes compared with microglia. Nevertheless, the level of
MIP-1a and MIP-1b secretion in purified cultures of astrocytes
remained low in comparison with mixed glial cultures.
Regulation of b-chemokine production by PGs and of PG
production by a-chemokines in human glial cells
To investigate the possible involvement of PGs in the regulation of
chemokine production by glial cells, we first tested the induction of
PGF2a and PGE2 in the supernatant of mixed glial cultures after
proinflammatory cytokine stimulation (PGE2: 19, 814 6 4, 315 vs
421 6 230 pg/ml; PGF2a: 23,596 6 3,739 vs 735 6 268 pg/ml,
(TNF-a 1 IL-1b)-stimulated vs control, day 3, n 5 5, p , 0.0001,
Fig. 1D). The induction of PG production was inhibited in the
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
days), the microglial cells were released by circular shaking, followed by
a 20-min incubation at 37°C to select adherent cells, as described (17).
These cells were subsequently grown in the same medium as the primary
mixed cultures for 3 wk to 2 mo to obtain a monolayer containing .95%
CD68/KiM-7 and EBM11-positive cells (17). The adherent cells remaining
after release of microglial cells were trypsinized, plated, and passaged two
to four times to obtain purified cultures of astrocytes (.95% GFAP-positive cells) (17). Under the usual experimental conditions, 5 3 105 astrocytes or microglial cells were seeded in 24-well plates and stimulations
were performed at confluence.
The Journal of Immunology
1703
FIGURE 2. Kinetics of a- and b-chemokine
production in human mixed glial cultures during the
3 days poststimulation with proinflammatory cytokines (E, control; F, TNF-a 1 IL-1b; Œ, IFN-g 1
TNF-a). The increase in GROa, IL-8, MIP-1a,
MIP-1b, and RANTES production started between
6 – 8 h after stimulation of glial cells with a combination of TNF-a 1 IL-1b and IFN-g 1 TNF-a,
respectively, and accumulated until day 3 poststimulation. Results are the mean 6 SD of two independent experiments using cells from two different brain specimens.
PGF2a secretion remained unchanged (Fig. 6). The enhancement
of PGE2 levels in mixed glial cultures, in the presence of anti-IL-8
and anti-GRO Abs, was due to an increase in PGE2 production in
both microglia and astrocytes (Fig. 6). The treatment of unstimulated glial cells by purified IL-8 and GROa did not significantly
affect their spontaneous production of PGF2a or PGE2 (data not
shown).
Discussion
The understanding of the different interactions between inflammatory mediators leading to a restricted local inflammatory reaction is
important in the development of novel therapies to treat inflammatory CNS diseases. Our results describe the parallel production
of a- and b-chemokines in human microglial cells and astrocytes.
This is also the first demonstration of a negative retroregulatory
loop between a- and b-chemokine and PG production in human
glial cells.
Studies on chemokine production by CNS resident cells of either human or rodent origin have focused mainly on a single class
of chemokine in either microglial cells or astrocytes (5, 6, 24 –28).
Here we describe the production of high amounts of IL-8 and
GROa by microglia and astrocytes, either in close interaction or
isolated, after stimulation with TNF-a in combination with IL-1b.
Mixed glial cells were also induced by the same cytokines to secrete MIP-1a and MIP-1b but to a lesser extent compared with
a-chemokines. This production was probably due to astrocytes
rather than microglia because 80% of the astrocyte cultures tested
could be induced to secrete MIP-1a and MIP-1b, compared with
20 –30% of microglial cultures. However, the observation of a
higher level of MIP-1 production in mixed glial cultures than in
astrocytes suggests that cell interactions between microglia and
astrocytes facilitate MIP-1a and MIP-1b production. The difference in the ability of astrocytes and microglia to secrete MIP-1a
and MIP-1b is reminiscent of previous data describing the inability
of rodent microglial cells to produce monocyte chemotactic protein-1 (MCP-1) protein, whereas astrocytes from the same origin
secreted MCP-1 under the same condition of stimulation (25).
Moreover, it has been shown that LPS-stimulated murine alveolar
and peritoneal macrophages expressed MIP-1a mRNA, whereas
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
presence of 5 mM indomethacin, a specific cyclooxygenase inhibitor (PGE2: 12,773 6 2,098 vs 147 6 24 pg/ml; PGF2a: 13, 983 6
2, 144 vs 287 6 25 pg/ml, (TNF-a 1 IL-1b)-stimulated vs
(TNF-a 1 IL-1b 1 indomethacin)-treated). Both microglial cells
and astrocytes were responsible for PG production as demonstrated previously (data not shown; see also Ref. 17). The effect of
indomethacin was subsequently tested on the induced production
of chemokines. Treatment of mixed glial cultures with 5–15 mM
indomethacin at the time of stimulation led to a 37– 60% increase
in the production of RANTES, MIP-1a, and MIP-1b induced, respectively, by IFN-g 1 TNF-a and TNF-a 1 IL-1b (Fig. 4, A and
B, RANTES: 4,142 6 597 vs 6, 100 6 674; MIP-1a: 420 6 56 vs
661 6 57; MIP-1b: 602 6 45 vs 894 6 52 pg/ml, cytokine-treated
vs (cytokine 1 5 mM indomethacin)-treated, day 3, n $ 4, p #
0.03). The effect of indomethacin was dose-independent in the
range of the tested concentrations. This suggested a limitation of
b-chemokine secretion by human glial cells in the presence of
PGs. Addition of exogenous PGF2a and PGE2 to the culture supernatants abolished the effect of indomethacin on this chemokine
production (Fig. 4, A and B), indicating a direct effect of PGs. In
contrast, inhibition of PG secretion by indomethacin had no effect
on GROa and IL-8 production in (TNF-a 1 IL-1b)-stimulated
glial cells (Fig. 4C). To investigate the role of microglia and astrocytes in the regulation of b-chemokines by PGs, purified cultures of microglial cells and astrocytes were analyzed under the
same conditions. Although all purified microglial or astrocyte cultures were induced with IFN-g 1 TNF-a to produce RANTES,
this production was not significantly modified after inhibition of
PG secretion by indomethacin (not shown). In contrast, indomethacin treatment of astrocytes led to a significant increase in MIP-1a
and MIP-1b production by these cells after TNF-a 1 IL-1b stimulation (Fig. 5).
To determine whether a- or b-chemokines could also regulate
PG secretion by glial cells, we tested the effect of neutralization of
chemokine activities on the induced production of PGs. Neutralization of GROa and/or IL-8, but not of b-chemokines, in the
presence of neutralizing Abs led to a strong increase in the production of PGE2 by mixed glial cultures (Fig. 6A, 23,969 6 7,500
vs 147,218 6 14,320 pg/ml, cytokine-treated vs (cytokines 1 antiIL-8 1 anti-GRO)-treated, day 3, n 5 4, p , 0.0001), whereas
1704
CHEMOKINE AND PROSTAGLANDIN RETROREGULATIONS IN HUMAN GLIA
only alveolar macrophages could produce MIP-1a protein in their
supernatant after stimulation (29). The disparity in MIP-1a secretion
between these two populations of macrophages was attributed to an
impairment in MIP-1a protein translation and secretion in peritoneal
macrophages (29). Another explanation for the differentially expressed MIP-1a and MIP-1b by astrocytes and microglia could be a
difference in transcription factors between the two cell types such as
activating transcription factor/cAMP response element binding protein (ATF/CREB)-, activating protein-1 (AP-1)-, NF-kB-, CCAAT/
enhancer binding protein (C/EBP), cellular-E 26-specific (C-ET)-related proteins, and/or MIP-1a nuclear protein (30 –32). However, the
possibility remains that other stimuli are required for an efficient induction of these b-chemokines in microglial cells. Finally, the production of RANTES, which was less than that of a-chemokines but
more than that of MIP-1a and MIP-1b, was increased in mixed or
isolated glial cultures after stimulation by TNF-a in combination with
IFN-g rather than IL-1b. The fact that the induction of different chemokines did not require the same combination of cytokines suggests
different mechanisms of induction. The inducing effect of inflamma-
tory cytokines might involve the induction of different transcription
factors such as STAT1, NF-kB, or AP-1, as shown in certain cell
types (33–36).
FIGURE 5. Effect of PG secretion blockade by indomethacin on proinflammatory cytokine-induced production of MIP-1 in purified cultures of
human astrocytes. MIP-1a- and MIP-1b-induced production was increased
by $40% in the presence of indomethacin. Mean 6 SD of four independent experiments using four different isolated astrocyte cultures.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
FIGURE 3. a- and b-chemokine production in purified cultures of human microglial cells and astrocytes. Both microglial cells and astrocytes
were able to produce GROa, IL-8 (A) MIP-1a, MIP-1b (C), and RANTES
(B) at day 3 poststimulation with TNF-a 1 IL-1b and IFN-g 1 TNF-a,
respectively. Five independent experiments were performed in triplicate.
Results shown are the mean 6 SD of five independently isolated glial
cultures (for RANTES), of three and two of six isolated microglial cultures
(for MIP-1a and MIP-1b, respectively) and of four out of five isolated
astrocyte cultures (for MIP-1a and MIP-1b).
FIGURE 4. Effect of PG secretion blockade by indomethacin on proinflammatory cytokine-induced production of a- and b-chemokines in human mixed glial cultures. A, IFN-g 1 TNF-a-induced RANTES production was increased by 37% in the presence of 15 mM indomethacin. This
increase was reversed after exogenous addition of PGF2a and PGE2 (16
ng/ml). B, The 47% increase in (TNF-a 1 IL-1b)-induced MIP-1a and
MIP-1b production in the presence of 15 mM indomethacin was also reversed after addition of PGF2a and PGE2. C, Inhibition of PG production
in (TNF-a 1 IL-1b)-stimulated glial cells had no effect on their GROa and
IL-8 production. Mean 6 SD of three to five independent experiments
performed in triplicate, using cells from four to five different brain
specimens.
The Journal of Immunology
1705
FIGURE 6. Effect of a- and b-chemokine neutralization on (TNF-a 1
IL-1b)-induced production of PGs in human mixed glial cultures, purified
microglia and astrocytes. Neutralizing Ab against RANTES was used at
200 mg/ml and neutralizing Abs against MIP-1a, MIP-1b, IL-8, and GRO
at 100 mg/ml. Mean 6 SD of two to four independent experiments performed in triplicate, using cells from 2 to 10 different brain specimens.
Numerous in vitro studies have demonstrated the role of a- and
b-chemokines in inducing the migration of specific populations of
leukocytes (6, 25, 27, 37). However, in vivo studies have demonstrated that the physiological functions of these chemokines are
highly dependent on their level of production. In fact, transgenic
mice overexpressing IL-8 exhibited a marked circulating neutrophilia and a decreased neutrophilic exudation into body cavities in
response to acute inflammatory stimulants (38). Similarly, transgenic mice overexpressing MCP-1 displayed no monocyte infiltration into tissues and showed increased sensitivity to intracellular
pathogens (39). In contrast, mice in which murine GROa or MCP-1
were expressed at lower levels and in anatomically restricted areas
displayed an appropriate infiltration of neutrophils or monocytes to
the site of chemokine production (40, 41). Thus, overproduction of
chemokines in vivo might lead to a loss of their specific biological
effect through either the inactivation of leukocytes by receptor desensitization or the abolition of the chemotactic gradient.
We investigated whether chemokine production could be under
the control of other inflammatory mediators such as PGs, secreted
in parallel in proinflammatory cytokine-stimulated glial cells. Inhibition of PG production in glial cells by the cyclooxygenase
inhibitor, indomethacin, resulted in an increase in glial secretion of
RANTES, MIP-1a, and MIP-1b, but not of IL-8 and GROa. The
reversion of this effect in the presence of exogenous PGs demonstrated that PGs are directly implicated in the specific limitation of
b-chemokines. Again, differences were observed between isolated
glial cultures and mixed glial cells. The blockade of PG secretion
did not significantly modify RANTES production by microglia or
astrocytes cultured in isolation. However, it did result in an increase in MIP-1 secretion (similar to that observed with mixed
glial cells) in astrocytes, which are the main cells producing
MIP-1a and MIP-1b (Fig. 5). This suggested that the feedback
regulation of RANTES by PGs might be dependent on interactions
between microglia and astrocytes. Thus, PGs may control local
physiological production of b-chemokines, leading to an appropriate activation of monocytes and lymphocytes, and/or an appropriate recruitment of these cells to the inflammatory foci (Fig. 7).
Surprisingly, inhibition of IL-8 and GRO activities in the supernatant of cytokine-stimulated glial cells by neutralizing Abs led
to a specific sharp increase in microglial and astrocyte production
of PGE2 but not of PGF2a. The addition of Abs to neutralize
RANTES, MIP-1a, and MIP-1b had no effect on the induced production of PGs by glial cells. Thus, the high levels of a-chemokines produced under proinflammatory conditions might be responsible, in part, for maintaining intracerebral PG secretion at a
low level. This observation suggests a role for a-chemokines in the
protection of cerebral cells from PG toxicity during the inflammatory response (Fig. 7).
The immune surveillance in CNS inflammation has to be tightly
controlled so as to mount an efficient defense against pathogens.
An imbalance between immunoregulatory factors during infectious or autoimmune diseases may have drastic effects on the integrity of the brain parenchyma. In this respect, studies of antiinflammatory treatments in different in vivo experimental systems
are of great interest. Although in such models, anti-inflammatory
drugs generally lead to the inhibition of vasoactive product secretion by effector cells (42– 44), the inconsistency of their effect on
the neurological lesions reflects subtle and time-dependent retroregulations, which determine the final biological effect on neuron
function and survival.
Acknowledgments
We thank Dr. J. C. Magnier for his precious support in tissue collection and
Dr. Claire Wolfrom for valuable discussions.
References
1. Lipton, S. A., M. Yeh, and E. B. Dreyer. 1994. Update on current models of
HIV-related neuronal injury: platelet-activating factor, arachidonic acid and nitric
oxide. Adv. Neuroimmunol. 4:181.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
FIGURE 7. Proposed model of the different mechanisms of CNS injury
during the inflammatory process. The recruitment of activated leukocytes
following the enhancement of adhesion molecules on their surface and the
release of proinflammatory cytokines such as IL-1, TNF, and IFN-g lead to
the establishment of an inflammatory mediator network due to leukocytes
and glial cells. This may induce the development of the deleterious effects
of inflammation on brain parenchyma. Elevated production of PGs may
lead to neurotoxicity through alteration of calcium homeostasis. Chemokines may also induce a secondary recruitment of leukocytes and in this
manner lead to further intraparenchymatous injuries. Retroregulation between PGs and chemokines may maintain their level of production so as to
avoid an exacerbation of inflammatory cell recruitment and neurotoxicity.
1706
CHEMOKINE AND PROSTAGLANDIN RETROREGULATIONS IN HUMAN GLIA
26. Murphy, Jr., G. M., X. C. Jia, Y. Song, E. Ong, R. Shrivastava, V. Bocchini,
Y. L. Lee, and L. F. Eng. 1995. Macrophage inflammatory protein 1-a mRNA
expression in an immortalized microglial cell line and cortical astrocyte cultures.
J. Neurosci. Res. 40:755.
27. Calvo, C. F., T. Yoshimura, M. Gelman, and M. Mallat. 1996. Production of
monocyte chemotactic protein-1 by rat brain macrophages. Eur. J. Neurosci.
8:1725.
28. Ehrlich, L. C., S. Hu, W. S. Sheng, R. L. Sutton, G. L. Rockswold, P. K. Peterson,
and C. C. Chao. 1998. Cytokine regulation of human microglial cell IL-8 production. J. Immunol. 160:1944.
29. VanOtteren, G. M., T. J. Standiford, S. L. Kunkel, J. M. Danforth, M. D. Burdick,
L. V. Abruzzo, and M. Strieter. 1994. Expression and regulation of macrophage
inflammatory protein-1a by murine alveolar and peritoneal macrophages.
Am. J. Respir. Cell. Mol. Biol. 10:8.
30. Proffitt, J., G. Crabtree, M. Grove, P. Daubersies, B. Bailleul, E. Wright, and
M. Plumb. 1995. An ATF/CREB-binding site is essential for cell-specific and
inducible transcription of murine MIP-1b cytokine gene. Gene 152:173.
31. Grove, M., and M. Plumb. 1993. C/EBP, NF-k B, and c-ETs family members and
transcriptional regulation of cell-specific and inducible macrophage inflammatory
protein 1 a immediate-early gene. Mol. Cell. Biol. 13:5276.
32. Ritter, L. M., M. Bryans, O. Abdo, V. Sharma, and N. M. Wilkie. 1995. MIP1a
nuclear protein (MNP), a novel transcription factor expressed in hematopoietic
cells that is crucial for transcription of the human MIP-1a gene. Mol. Cell. Biol.
15:3110.
33. Marfaing-Koka, A., O. Devergne, G. Gorgone, A. Portier, T. J. Schall,
P. Galanaud, and D. Emilie. 1995. Regulation of the production of the RANTES
chemokine by endothelial cells. Synergistic induction by IFN-g plus TNF-a and
inhibition by IL-4 and IL-13. J. Immunol. 154:1870.
34. Marfaing-Koka, A., M. Maravic, M. Humbert, P. Galanaud, and D. Emilie. 1996.
Contrasting effects of IL-4, IL-10, and corticosteroids on RANTES production by
human monocytes. Int. Immunol. 8:1587.
35. Nelson, P. J., H. T. Kim, W. C. Manning, T. J. Goralski, and A. M. Krensky.
1993. Genomic organization and transcriptional regulation of the RANTES chemokine gene. J. Immunol. 151:2601.
36. Yasumoto, K., S. I. Okamoto, N. Mukida, S. Murakami, M. Mai, and
K. Matsushima. 1992. Tumor necrosis factor-a and interferon-g synergistically
induce interleukin 8 production in a human gastric cancer cell line through acting
concurrently on AP-1 and NF-kB-like binding sites of the interleukin 8 gene.
J. Biol. Chem. 267:22506.
37. Rollins, B. J. 1997. Chemokines. Blood 90:909.
38. Simonet, W. S., T. M. Hughes, H. Q. Nguyen, L. D. Trebasky, D. M. Danilenki,
and E. S. Medlock. 1994. Long-term impaired neutrophil migration in mice overexpressing human interleukin-8. J. Clin. Invest. 94:1310.
39. Rutledge, B. J., H. Rayburn, R. Rosenberg, R. J. North, R. P. Gladue,
C. L. Corless, and B. J. Rollins. 1995. High level monocyte chemoattractant
protein-1 expression in transgenic mice increases their susceptibility to intracellular pathogens. J. Immunol. 155:4838.
40. Lira, S. A., P. Zalamea, J. N. Heinrich, M. E. Funtes, D. Carrasco, A. C. Lewin,
D. S. Barton, S. Durham, and R. Bravo. 1994. Expression of the chemokine
N51/KC in the thymus and epidermis of transgenic mice results in marked infiltration of a single class of inflammatory cells. J. Exp. Med. 180:2039.
41. Fuentes, M. E., S. K. Durham, M. R. Swerdel, A. C. Lewin, D. S. Barton,
J. R. Megill, R. Bravo, and S. A. Lira. 1995. Controlled recruitment of monocytes
and macrophages to specific organs through transgenic expression of monocyte
chemoattractant protein-1. J. Immunol. 155:5769.
42. Limoges, J., Y. Perdisky, P. Bock, and H. E. Gendelman. 1997. Dexamethasone
therapy worsens the neuropathology of human immunodeficiency virus type 1
encephalitis in SCID mice. J. Infect. Dis. 175:1368.
43. Zysk, G., W. Brück, J. Gerber, Y. Brück, H. W. Prange, and R. Nau. 1996.
Anti-inflammatory treatment influences neuronal apoptotic cell death in the dentate gyrus in experimental pneumococcal meningitis. J. Neuropathol. Exp. Neurol. 55:722.
44. Morimoto, K., D. C. Hooper, A. Bornhorst, S. Corisdeo, M. Bette, Z. F. Fu,
M. K.-H. Schäfer, H. Koprowski, and E. Weihe. 1996. Intrinsic responses to
Borna disease virus infection of the central nervous system. Proc. Natl. Acad. Sci.
USA 93:13345.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
2. Chao, C. C., S. Hu, and P. K. Peterson. 1995. Modulation of human microglial
cell superoxide production by cytokines. J. Leukoyte Biol. 58:65.
3. Ding, M., B. A. St. Pierre, J. F. Parkinson, P. Medberry, J. L. Wong,
N. E. Rogers, L. J. Ignarro, and J. E. Merrill. 1997. Inducible nitric-oxide synthase and nitric oxide production in human fetal astrocytes and microglia. J. Biol.
Chem. 272:11327.
4. Liu, J., M. L. Zhao, C. F. Brosnan, and S. C. Lee. 1996. Expression of type II
nitric oxide synthase in primary human astrocytes and microglia: role of IL-1b
and IL-1 receptor antagonist. J. Immunol. 157:3569.
5. McManus, C. M., C. F. Brosnan, and J. W. Berman. 1998. Cytokine induction of
MIP-1a and MIP-1b in human fetal microglia. J. Immunol. 160:1449.
6. Peterson, P. K., S. Hu, J. Salak-Johnson, T. W. Molitor, and C. C. Chao. 1997.
Differential production of and migratory response to b chemokines by human
microglia and astrocytes. J. Infect. Dis. 175:478.
7. Eng, L. F., R. S. Ghirnikar, and Y. L. Lee. 1996. Inflammation in EAE: role of
chemokine/cytokine expression by resident and infiltrating cells. Neurochem.
Res. 21:511.
8. Chen, S. T., C. Y. Hsu, E. L. Hogan, P. V. Halushka, O. I. Linet, and F. M. Yatsu.
1986. Thromboxane, prostacyclin, and leukotrienes in cerebral ischemia. Neurology 36:466.
9. Dinarello, C. A., J. G. Cannon, and S. M. Wolff. 1988. New concepts on the
pathogenesis of fever. Rev. Infect. Dis. 10:168.
10. Dore-Duffy, P., S. Y. Ho, and C. Donovan. 1991. Cerebrospinal fluid eicosanoid
levels: endogenous PGD2 and LTC4 synthesis by antigen-presenting cells that
migrate to the central nervous system. Neurology 41:322.
11. Hocking, D. C., P. G. Philips, T. J. Ferro, and A. Johnson. 1990. Mechanisms of
pulmonary edema induced by tumor necrosis factor-a. Circ. Res. 67:68.
12. Rubin, R. M., and J. T. Rosenbaum. 1988. A platelet-activating factor antagonist
inhibits interleukin 1-induced inflammation. Biochem. Biophys. Res. Commun.
154:429.
13. Powell, W. S., D. Chung, and S. Gravel. 1995. 5-oxo-6,8,11,14-eicosatetraenoic
acid is a potent stimulator of human eosinophil migration. J. Immunol. 154:4123.
14. Bokoch, G. 1995. Chemoattractant signaling and leukocyte activation. Blood
86:1649.
15. Inagaki, N., and H. Wada. 1994. Histamine and prostanoid receptors on glial
cells. Glia 11:102.
16. Peudenier, S., C. Héry, L. Montagnier, and M. Tardieu. 1991. Human microglial
cells: characterization in cerebral tissue and in primary culture, and study of their
susceptibility to HIV-1 infection. Ann. Neurol. 29:152.
17. Janabi, N., S. Chabrier, and M. Tardieu. 1996. Endogenous nitric oxide activates
prostaglandin F2a production in human microglial cells but not in astrocytes: a
study of interactions between eicosanoids, nitric oxide, and superoxide anion
(O22) regulatory pathways. J. Immunol. 157:2129.
18. Peudenier, S., C. Héry, K. Heng Ng, and M. Tardieu. 1991. HIV receptors within
the brain: a study of CD4 and MHC-II on human neurons, astrocytes and microglial cells. Res. Virol. 142:145.
19. Mackewicz, C. E., E. Barker, and J. A. Levy. 1996. Role of b-chemokines in
suppressing HIV replication. Science 274:1393.
20. Paliard, X., A. Y. Lee, and C. M. Walker. 1996. RANTES, MIP-1a, and MIP-1b
are not involved in the inhibition of HIV-1SF33 replication mediated by CD81 T
cell clones. AIDS 10:1317.
21. Kimata, H., A. Yoshida, C. Ishioka, M. Fujimoto, I. Lindley, and K. Furusho.
1996. RANTES and macrophage inflammatory protein 1a selectively enhance
immunoglobulin (IgE) and IgG4 production by human B cells. J. Exp. Med.
183:2397.
22. Taub, D. D., S. M. Turcovski-Corrales, M. L. Key, D. L. Longo, and
W. J. Murphy. 1996. Chemokines and T lymphocyte activation: b chemokines
costimulate human T lymphocyte activation in vitro. J. Immunol. 156:2095.
23. Pozzoli, G., A. Costa, M. Grimaldi, G. Schettini, P. Preziosi, A. Grossman, and
P. Navarra. 1994. Lipopolysaccharide modulation of eiosanoid and corticotrophin-releasing hormone release from rat hypothalamic explants and astrocyte
cultures in vitro: evidence for the involvement of prostaglandin E2 but not prostaglandin F2a and lack of effect of nerve growth factor. J. Endocrinol. 140:103.
24. Aloisi, F., A. Carè, G. Borsellino, P. Gallo, S. Rosa, A. Bassani, A. Cabibbo,
U. Testa, G. Levi, and C. Peschle. 1992. Production of hemolymphopoietic cytokines (IL-6, IL-8, colony-stimulating factors) by normal human astrocytes in
response to IL-1b and tumor necrosis factor-a. J. Immunol. 149:2358.
25. Hayashi, M., Y. Luo, J. Laning, R. M. Strieter, and M. E. Dorf. 1995. Production
and function of monocyte chemoattractant protein-1 and other b-chemokines in
murine glial cells. J. Neuroimmunol. 60:143.