| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Laboratory of Organ and System Pathophysiology, Istituto Superiore di Sanità, Rome, Italy; and
Roche Milano Ricerche, Milan, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and,
to a lesser extent, IL-2, but were inefficient stimulators of IL-4
secretion by Th2 cells. Microglia exposed in vitro to IFN- showed
enhanced expression of MHC class II, CD40, and CD54 molecules and
became able to restimulate Th2 cells. In addition to IFN-, GM-CSF
increased the ability of microglia to activate Th1, but not Th2, cells
without up-regulating MHC class II, CD40, or CD54 molecules. These
results suggest that interaction with Th1 cells and/or Th1-secreted
soluble factors induces the functional maturation of adult mouse
microglia into an APC able to sustain CD4+ T cell
activation. Moreover, GM-CSF, a cytokine secreted by T cells as well as
reactive astrocytes, could prime microglia for Th1-stimulating
capacity, possibly by enhancing their responsiveness to Th1-derived
signals. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The sequential steps leading to reciprocal microglia-T cell activation
as well as the relative importance of contact-mediated vs
cytokine-induced effects on microglia APC function are still unclear.
IFN-, a cytokine secreted by NK and Th1 cells, induces the
expression of MHC class II molecules on microglia in vitro and in vivo
(9, 15). IFN- also up-regulates the expression of
adhesion/costimulatory molecules on cultured mouse or human microglia
(3, 6, 13, 14) and strongly enhances the microglia ability
to restimulate Th1 and Th2 cells (14) or to activate naive
T cells (11, 13), although less efficiently than dendritic
cells (DC)3
(16). Th1 cells, in turn, stimulate cultured microglia to
secrete IL-12 and PGE2, an enhancer and inhibitor
of Th1 responses, respectively, indicating a role for microglia-Th1
interactions in the amplification and regulation of T cell-mediated
neuroinflammatory processes (17). Because cultured
microglia from the neonatal mouse or adult human CNS exhibit a
phagocytic, macrophage-like phenotype (1, 9), the above
studies suggest that already activated microglia productively interact
with CD4+ T cells but do not allow us to draw
conclusions on the outcome of the interactions between resting
microglia and preactivated T cells. Recent work by Sedgwick and
collaborators (12, 18) has shown that following induction
of GVHD in the rat and infiltration of the CNS by sparse T cell blasts,
rat microglia proliferate, up-regulate surface molecules (MHC class II,
CD45, CD4), and develop into an APC capable of sustaining T cell
effector functions (e.g., IFN- and TNF secretion), but not T cell
proliferation or IL-2 secretion, in ex vivo functional assays. In the
GVHD model, activated, MHC class II+ microglia
were detected both in close proximity to T cells and throughout the CNS
parenchyma. Whether microglia activation and acquisition of APC
features are induced directly by T cells or soluble factors from the
injured CNS are also involved remains unclear.
In the present study we sought to determine to what extent T cells
directly induce microglia to mature into a competent APC. To mimic the
encounter between resting microglia and preactivated T cells, we
cocultured microglia acutely isolated from the adult BALB/c mouse CNS
with Th1 and Th2 cell lines from DO11.10 TCR transgenic mice, which is
probably a more adequate model than those previously used to dissect
the early steps involved in microglia activation. We also asked whether
signals that are known to be released in the injured CNS, such as TNF,
M-CSF, and GM-CSF, would prime resting microglia for enhanced APC
function. The results demonstrate that direct interaction with Th1, but
not Th2, cells and the Th1-derived cytokine IFN- induce typical APC
molecules on microglia and their functional maturation into an APC
capable of T cell restimulation. Moreover, we show that GM-CSF primes
microglia for enhanced Th1 stimulatory capacity, possibly by increasing
microglia responsiveness to Th1-derived signals.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DO11.10 TCR transgenic mice on BALB/c background (19) were provided by Dr. D. Y. Loh. In these transgenic mice, 95–100% of the CD4+ T cells are Vß8.1.2+ and express a TCR specific for OVA peptide 323–339 bound to I-Ad (20). Two- to three-month-old female BALB/c mice were used for the preparation of adult microglia.
Isolation of microglia
Microglia were isolated by modification of a previously published, nonenzymatic procedure (21). Briefly, mice were sacrificed with an overdose of anaesthetic and perfused with ice-cold glucose-containing buffered saline, GKN (8 g/L NaCl, 0.4 g/L KCl, 3.56 g/L Na2HPO412H2O, 0.78 g/L NaH2PO42H2O, and 2 g/L D(+)-glucose, pH 7.4) (22). Pooled CNS tissues were mechanically dissociated with GKN containing 0.02% BSA by passing them through a wire mesh. After centrifugation at 200 x g for 10 min, dissociated CNS tissue was resuspended in 70% isotonic Percoll (Pharmacia Biotech, Uppsala, Sweden) and centrifuged for 20 min at 500 x g on a 30:37:70% Percoll gradient. Microglia were obtained from the 37 (1.048 g/ml):70% (1.086 g/ml) interface, passed through a 30-µm pore size filter, washed, and resuspended in RPMI supplemented with 50 µM 2-ME, 2 mM L-glutamine, 50 µg/ml gentamicin, and 10% FCS. Cells were plated at the desired density in 96-well flat-bottom tissue culture plates. Anti-CD11b (Mac-1)-coated magnetic microbeads (Miltenyi-Biotec, Bergish, Germany) were used for positive selection of microglia.
DC preparation
DC were enriched from inguinal lymph nodes of 2- to 3-mo-old BALB/c mice immunized with CFA into the hind footpads 6 days earlier, as previously described (23). Briefly, lymph node cells were depleted of T cells by cytotoxic elimination with HO-13-4 anti-Thy 1.2 Ab (TIB 99) followed by rabbit complement. Thy 1.2-negative cells were separated by Percoll gradient in low buoyant and high buoyant density cells. DC were purified from low buoyant density cells using anti-CD11c (N418)-coated microbeads (Miltenyi-Biotec). After two rounds of elution on MiniMACS separation columns, a highly enriched population of DC (85–95% N418+ cells) was obtained. As previously shown (24), the Ag-presenting capacity of DC was not affected by positive selection with N418 mAb.
T cell lines
CD4+ cells were positively selected from
inguinal and mesenteric lymph nodes of naive DO11.10 TCR transgenic
mice by anti-CD4-coated magnetic microbeads (Miltenyi-Biotec).
CD4+ T cells (2 x 105
cells/well) were cultured with OVA peptide 323–339 (0.3 µM), and
mitomycin C-treated BALB/c splenocytes (5 x
106 cells/well) as APC in a total volume of 2 ml
in 24-well plates in the presence of either 0.1 ng/ml recombinant mouse
IL-12 (Hoffmann-La Roche, Nutley, NJ) and 10 µg/ml anti-mouse
IL-4 (11B11, American Type Culture Collection, Manassas, VA), or 20
ng/ml mouse rIL-4 (Hoffmann-La Roche, Basel, Switzerland) and 10
µg/ml anti-mouse IL-12 mAb (10F6, Hoffmann-La Roche, Nutley, NJ)
to obtain Th1 or Th2 cell lines, respectively. Cells were cultured in
RPMI supplemented with 50 µM 2-ME, 2 mM L-glutamine, 50
µg/ml gentamicin, and 10% FCS (Sigma, St. Louis, MO). After 3 days
in vitro, T cells were expanded and grown in complete medium containing
10 ng/ml of recombinant human IL-2 (Hoffmann-La Roche, Basel,
Switzerland). The cell lines obtained exhibited a clear Th1 or Th2
profile, as detected by intracellular staining for IFN- and IL-4
production (14).
T cell stimulation assay
Graded numbers of mouse microglia or DC were seeded in 96-well
flat-bottom tissue culture plates and tested for the ability to
restimulate T cells immediately after seeding. Microglia Ag-presenting
function was also tested after 24–120 h of culturing without or with
mouse recombinant IFN- (10 U/ml), TNF (10 ng/ml), M-CSF (20 and 50
ng/ml; all from Genzyme, Cambridge, MA), GM-CSF (100 U/ml; R & D
Systems, Abingdon, U.K.), or LPS from Escherichia coli,
serotype 026:B6 (100 ng/ml; Sigma). Before addition of T cells,
microglia were gently washed three times with complete medium to remove
the stimulating agents. OVA TCR transgenic Th1 or Th2 cells were added
(5 x 104) to wells containing microglia or
DC in the presence or the absence of optimal concentrations of
OVA323–339 (0.3 µM) or native OVA (10 µM;
grade V, from Sigma) (14). For inhibition of Ag
presentation, anti-MHC class II (I-Ab,d) mAb
(B21.22, American Type Culture Collection; 10 µg/ml) was added to the
cultures 30 min before addition of T cells. In some experiments,
neutralizing rat anti-mouse IFN- mAb (XMG1.2) and rat IgG1
isotypic control Ab (10 µg/ml; both from PharMingen, San Diego, CA)
were added at culture initiation. For analysis of T cell-derived
cytokines, supernatants from duplicate or triplicate cultures were
harvested after 24 or 48 h, centrifuged at 1200 rpm, and stored at
-20°C until used for cytokine determination.
IL-2, IFN-, and IL-4 were quantified by two-site sandwich ELISA
(Genzyme). Detection limits for all cytokines were in the range of
5–15 pg/ml.
Immunocytochemical staining and flow cytometric analysis
Immunocytochemical stainings were performed on freshly isolated microglia, microglia cultured in the absence or the presence of stimulating agents, and microglia-T cell cocultures. Cells were washed with FACS buffer (HBSS containing 0.1% sodium azide and 2% FCS) and preincubated with anti-mouse CD16/CD32 mAb (2.4G2, from PharMingen) for 10 min at 4°C to block nonspecific binding to Fc receptors. For single or double stainings, the following fluorochrome-conjugated mAbs (all from PharMingen) were used at optimal dilutions: rat anti-mouse Mac-1/CD11b-PE or -FITC (M1/70, IgG2b), rat anti-mouse CD45-PE (30-F11, IgG2b), rat anti-mouse CD86-FITC (GL1, IgG2a), hamster anti-mouse CD40-FITC (HM40–3, IgM), hamster anti-mouse CD54-FITC (3E2, IgG), mouse anti-mouse MHC class II (I-Ad)-FITC (39-10-8, IgG3), hamster anti-mouse CD11c-FITC (HL3, IgG), and rat anti-mouse CD4-Cy-Chrome (H129.19, IgG2a). The background fluorescence was evaluated by staining the cells with isotype control Abs: rat IgG2b-PE or-FITC, rat IgG2a-FITC, mouse IgG3-FITC, hamster IgG- and IgM-FITC, and mouse IgG2a-Cy-Chrome (all from PharMingen). At the end of incubations, cells were washed three times with FACS buffer. Two-color analysis was performed on a FACScan using LYSYS II software (Becton Dickinson, Mountain View, CA); 3000–5000 events were reported.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Intraparenchymal microglia were enriched by Percoll gradient from
the perfused CNS of adult BALB/c mice, immunostained immediately after
isolation, and analyzed by flow cytometry. In agreement with previous
studies (22, 25), microglia were identified as a
homogeneous population of double-labeled
CD11b+/CD45+ cells,
comprising 88–93% of the freshly isolated cells (Fig. 1
, A and B). The
intensity of CD45 staining on microglia was much lower than that on
splenic cells (Fig. 1D), and this has been reported to
distinguish microglia from CNS perivascular and other tissue
macrophages (22). Among cells isolated from the CNS, only
a few (
5%) expressed high levels of CD45 but no CD11b (Fig. 1B), and <2% were CD4+ T cells (not
shown). No cells expressed CD11c, a marker for DC in lymphoid organs
(not shown).
|
C), but nearly all the positively
selected cells died within a few hours after seeding, suggesting that
the selection procedure negatively affects microglial survival.
Therefore, all subsequent experiments were performed with freshly
isolated microglia enriched by Percoll gradient without further
purification. After seeding at a relatively high density (3 or 9 x 105 cells/well in 96-well plates) in FCS-containing medium, most microglial cells loosely adhered to the plastic without extending cytoplasmic processes. About 50% of microglial cells (range, 42–56 in two different experiments) survived after 24 h in vitro, as assessed by trypan blue exclusion of resuspended cells. After 3 days in vitro, microglia were still loosely adherent to the plastic surfaces and few extended thin processes. At this time no further decrease in the number of living cells was detected, indicating adaptation of microglia to culture conditions.
Interaction of microglia with Th1, but not Th2, cells induces expression of MHC class II and adhesion/costimulatory molecules
Ex vivo analyzed CD11b+ microglia were
largely MHC class II-,
CD40-, and CD86-, and
only a small proportion (10–15%) expressed CD54 (Fig. 2). To test whether expression of these
molecules is up-regulated by interaction with T cells,
microglia-enriched populations were cultured either alone or together
with polarized Th1 or Th2 cell lines from DO11.10 TCR transgenic BALB/c
mice in the absence or the presence of
OVA323–339. After about 40 h of culture,
the cells were double labeled with anti-CD11b and anti-MHC
class II, -CD40, -CD54, or -CD86 mAbs and analyzed by flow cytometry.
In agreement with a previous study (25), only CD86 was
spontaneously up-regulated on most (50–75%) short term cultured adult
microglia, whereas MHC class II, CD40, and CD54 were expressed on a
minority of the cells (4 ± 1, 12 ± 3, and 12 ± 4%,
respectively; mean ± SEM of three experiments). A representative
experiment is shown in Fig. 2. Following interaction of microglia with
Th1 cells in the absence of Ag, the percentages of
CD11b+ cells coexpressing MHC class II and CD54
molecules rose to 25–35 and 30–65%, respectively (results from two
experiments), indicating a certain stimulating effect of microglia-T
cell interactions even in the absence of Th1 activation and cytokine
secretion (see below). However, it was only in the presence of
OVA323–339 that the microglia-Th1 interaction
induced the highest percentages of CD11b+
microglia coexpressing MHC class II (55 ± 13%), CD40 (34 ±
3%), and CD54 (75 ± 14%; mean ± SEM of three experiments;
see a representative experiment in Fig. 2). During Ag presentation by
microglia, up-regulation of CD54 and de novo expression of CD40 were
also observed on a substantial proportion (
60 and 80%,
respectively) of the CD11b- cells (mostly Th1
cells), suggesting that the CD40-CD154 and CD54-CD11a/CD18 pathways may
be involved in T-T cell interactions.
|
). Coculture of microglia with Th2 cells in the
presence of OVA323–339 induced only a very
modest up-regulation in the percentage of CD11b+
cells coexpressing MHC class II and CD54 (11 ± 0.5 and 22 ±
1%, respectively; mean ± SEM from three experiments), but not
CD40 or CD86 molecules. Collectively, the above data indicate that with the exception of CD86, which is spontaneously expressed in vitro, only interaction with Th1 strongly up-regulates the expression of APC-related surface molecules on microglia.
Freshly isolated microglia induce Th1, but little Th2, activation
We next asked whether adult mouse microglia were able to stimulate
cytokine secretion from Th1 and Th2 cells. Coculture of graded numbers
of freshly isolated microglia with Th1 cells for 24–48 h in the
presence of OVA323–339 led to a progressive
increase in IFN- and IL-2 secretion (Fig. 3). While IFN- secretion was readily
induced within 24 h, prominent induction of IL-2 was evident only
after 48 h. No IL-4 was detectable (not shown). Th1 cytokine
secretion was not observed in the absence of Ag and was completely
blocked by neutralizing anti-MHC class II mAb (Fig. 3), indicating
that T cell activation was Ag dependent and MHC class II restricted.
The finding that in the presence of native OVA (10 µM) Th1 activation
was lower than that observed in the presence of
OVA323–339 peptide (0.3 µM) suggests that ex
vivo tested adult microglia do not possess a very efficient Ag uptake
and processing machinery (Fig. 3). This is in contrast with what was
reported for phagocytically active microglia obtained from the neonatal
mouse brain (14).
|
secretion similar to that triggered by adult mouse
microglia. However, DC were up to 100-fold more efficient than
microglia in inducing IL-2 secretion. At low APC density
(<104/well) only DC were able to induce
substantial Th1 activation.
When microglia were cocultured with Th2 cells in the presence of either
OVA323–339 or native OVA, the secretion of IL-4
was induced only with the highest dose (9 x
104/well) of adult mouse microglia. Th2 cells
secreted neither IL-2 nor IFN-. The modest Th2 activation induced by
microglia was Ag and MHC class II dependent. Compared with DC, adult
mouse microglia were definitely inefficient Th2 stimulators (Fig. 3).
It is interesting to note that Th2 activation resulted in only a modest
up-regulation of MHC class II expression on microglia (Fig. 2), thus
explaining their poor Th2-restimulating capacity.
IFN- is the major Th1-derived cytokine inducing microglia APC
function
Subsequent experiments were aimed at identifying the signals
involved in stimulation of the APC function of adult mouse microglia.
We first examined the role of IFN-, a well-known stimulator of APC
function in cultured neonatal mouse microglia (3, 9, 13, 14). To assess the extent of IFN- involvement in the
Th1-induced maturation of microglia APC capacity, we added neutralizing
anti-mouse IFN- mAb to microglia-Th1 cocultures and evaluated
OVA323–339-dependent microglia and T cell
activation. Addition of anti-IFN- mAb, but not of an isotypic
control, prevented the up-regulation of MHC class II expression on
microglia and reduced by about 50% the levels of IL-2 secreted by Th1
cells (Fig. 4). These results demonstrate
that IFN- is the major Th1-derived factor inducing APC features in
microglia.
|
and then tested for the ability to activate Th1 and Th2 cells
in the presence of OVA323–339 (Fig. 5, left panels). Compared with
freshly isolated microglia, 24-h-cultured, unstimulated microglia were
much less efficient APC for Th1 activation (compare Figs. 3 and 5). The
decreased T cell stimulatory capacity of short term cultured microglia
could be due to their relatively poor survival in vitro (as mentioned
above, 50% of microglia died after 24 h in culture). Similarly
to what was observed with neonatal mouse microglia (14),
24-h pretreatment of adult microglia with IFN- (10 U/ml) activated
these cells to become efficient APC for the restimulation of both Th1
and Th2 cells.
|
induced the expression of MHC class II and
CD54 molecules on most (65–85%) adult mouse microglia. Conversely,
IFN- induced little CD40 expression. The percentage of
CD40+/CD11b+ cells in
IFN--treated cultures only doubled compared with that in
unstimulated cultures (23 ± 3 vs 13 ± 3%; mean ± SEM
of three experiments). A representative experiment is shown in Fig. 6. Despite a small increase in the mean
fluorescent intensity, the constitutive expression of CD86 was not
significantly enhanced by IFN-.
|
plus LPS did
not enhance the capacity of adult mouse microglia to stimulate Th1 and
Th2 cytokine secretion above control levels (data not shown).
IFN--pretreated microglia and microglia exposed to supernatants (at
1/5 dilution) from cultures of Th1, but not Th2, cells, activated by
plate-bound anti-TCR mAb, expressed similar levels of MHC class II
and adhesion/costimulatory molecules and induced similar Th1 and Th2
activation (data not shown). This confirms that IFN-, rather than
other soluble mediators present in Th1 supernatants, is the dominant
cytokine inducing microglia APC function. GM-CSF enhances microglia Th1-restimulating capacity
GM-CSF and M-CSF have been proposed as stimulators of microglia
proliferation and activation (1), but contrasting data
exist on the capacity of GM-CSF to modulate the APC function of long
term cultured rodent microglia (27, 28). To clarify this
issue, we examined the influence of these two mediators on the T cell
stimulatory activity of adult mouse microglia. Long (72–120 h), but
not short (24 h), pretreatments of microglia with GM-CSF enhanced the
capacity of microglia to stimulate cytokine secretion from Th1 cells
(Fig. 5). The slow kinetics of the GM-CSF effect could be due to
enhanced microglial responsiveness to GM-CSF with time in vitro,
possibly due to increased expression of GM-CSF receptors. Conversely,
no major differences were detected between unstimulated and
GM-CSF-pretreated microglia in the stimulation of IL-4 secretion from
Th2 cells. The Th1 stimulatory capacity of GM-CSF-pretreated microglia
did not differ significantly from that observed after a 24-h
pretreatment with IFN- alone. A similar Th1 induction was observed
when microglia were pretreated simultaneously with GM-CSF and IFN-,
suggesting that there was no synergism between the two cytokines (Fig. 5). Both short and prolonged pretreatments with M-CSF (20 or 50 ng/ml)
failed to enhance microglia Ag-presenting capacity (data not
shown).
GM-CSF did not induce the proliferation of CD11b+ microglia (which in basal conditions fail to divide, as assessed by bromodeoxyuridine incorporation) or cell survival, as assessed by counting trypan blue-excluding cells (data not shown), suggesting that the enhanced APC function was not a direct consequence of increased microglia numbers. As revealed by flow cytometric analysis, GM-CSF failed to increase the expression of MHC class II, CD40, and CD54 molecules on microglia, even following prolonged in vitro treatment (data not shown from two different experiments). These observations suggest that GM-CSF does not directly induce APC features in microglia, but may render microglia more responsive to Th1-derived signals.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. These results suggest that CNS infiltration by
peripherally primed Th1-type cells may induce resident microglia to
mature into competent APC able to sustain CD4+ T
cell responses within the CNS.
Our results indicate that APC-related molecules are induced on
microglia even in the absence of detectable MHC class II expression
before interaction with Th1 cells. Although maximally stimulated during
Ag-specific interactions, the expression of MHC class II and CD54
molecules was already up-regulated in the absence of Ag, suggesting
that preactivated T cells express activation-induced surface molecules
implicated in MHC class II and CD54 molecule induction. Conversely,
CD40 induction required microglia-Th1 interaction via an Ag bridge,
which indicates the involvement of additional signals. CD40 is a key
molecule for APC activation, since ligation of CD40 by CD154 expressed
on T cells up-regulates MHC and costimulatory (CD80/CD86) molecules and
induces cytokine secretion in DC and macrophages (29). We
have recently shown that CD40-CD154 interaction and IFN- mediate
induction of IL-12 secretion by microglia during MHC class
II-restricted, Ag-dependent activation of Th1 cells (17).
The observations that CD86 was undetectable on ex vivo microglia and
was spontaneously up-regulated in vitro suggest either that expression
of this molecule is kept under tight control by the CNS
microenvironment or that culture conditions include CD86-inducing
factors.
IFN- secreted during Th1 activation appears to be the most important
cytokine inducing microglia APC features. Because activation of
differentiated T cells is very rapid and occurs within a few hours
(30), it is likely that Ag bound to yet undetectable MHC
class II molecules expressed on ex vivo microglia induces rapid IFN-
secretion, which amplifies the bidirectional signaling cascade between
microglia and Th1 cells. In vitro exposure of adult mouse microglia to
IFN- also induced MHC class II and adhesion/costimulatory molecules,
resulting in enhanced APC potency for both Th1 and Th2 restimulation.
This is the first report in which IFN- alone is shown to act on
adult mouse microglia ex vivo, with most cells responding to the
cytokine. This suggests that functional IFN- receptors are already
expressed on resting microglia. In previous reports adult mouse
microglia exposed to IFN- plus LPS were shown to exhibit very poor T
cell stimulatory function (25) and to preferentially
promote a Th1-type phenotype (21). Because LPS is known to
inhibit IFN--induced MHC class II expression in several cell types,
including macrophages (Ref. 31 and references therein) and
because we have shown that IFN-/LPS treatments fail to stimulate the
APC function of neonatal (14) and adult (this study) mouse
microglia, the use of LPS may be a confounding factor in the analysis
of microglia-T cell interactions.
Compared with DC isolated from lymph nodes of CFA-immunized mice, ex
vivo tested adult mouse microglia were less efficient APC in inducing
IFN- and IL-2 secretion from Th1 cells. This was expected, because
mature DC migrating to lymph nodes already display on their surface
elevated levels of MHC class II, CD40, CD54, and CD86 molecules.
Induction of IL-2 secretion differed profoundly between the two cell
types, with microglia stimulating lower levels of IL-2 and with slower
kinetics. Because IL-2 secretion by effector T cells was shown to be
more dependent than IFN- secretion on CD54 and CD80/CD86 expression
by APC (30), this observation is consistent with their
progressive up-regulation on adult microglia during contact with Th1
cells. Ford et al. (12) have recently shown that microglia
isolated from the CNS of rats undergoing GVHD fail to induce IL-2
secretion from T cell lines in a 24-h assay and induce a certain degree
of T cell apoptosis. Discrepancies between this study and the present
data could be ascribed to different levels of adhesion/costimulatory
molecules expressed on rat and mouse microglia, signals delivered by
TCR transgenic vs nontransgenic T cells, as well as the timing of IL-2
detection. Both studies, however, support the concept that adult
microglia, compared with professional APC, are poor stimulators of IL-2
secretion by T cells. This has been proposed as a regulatory mechanism
to prevent and/or limit T cell reactivity and its deleterious
consequences within the CNS parenchyma (12). It is,
however, possible that, at least in certain chronic inflammatory
processes, such as multiple sclerosis (2, 5, 6, 7, 8), the
elevated expression of adhesion/costimulatory molecules (CD40, CD54,
CD80) on MHC class II-expressing microglia could promote IL-2
secretion, thus leading to a more sustained T cell response.
We have recently reported that IFN--pretreated neonatal mouse
microglia behave as efficiently as DC for Th1 and Th2 restimulation,
although they are less potent inducers of naive T cell proliferation
and Th1 differentiation (16). Even following treatment
with IFN-, adult microglia appear to be less efficient than their
neonatal counterparts in T cell restimulation (14, 16),
which could be due to impaired survival in vitro. The Ag-presenting
capacity of neonatal microglia could indeed differ due to their
activated, phagocytic phenotype and/or the astrocyte-rich milieu in
which they are expanded in vitro.
Th2 cells are inefficient in delivering stimulating signals to
microglia, consistent with the observation that Th2 cells secrete
macrophage/microglia-deactivating factors such as IL-4 and IL-10
(32). Although both preactivated Th1 and Th2 cells express
on their surface CD154, which binds to CD40 leading to APC activation
(33), it is clear that Th2 cells lack additional
microglia-inducing stimuli, such as IFN-. The dynamics of Th1 and
Th2 responses within the CNS have been examined in various animal
models of CNS infection and autoimmunity. Although Th1-type cytokines
predominate in the early and acute phases of CNS inflammation, during
both the progression and remission phases the presence of Th2-type
cytokines was also detected (34, 35, 36). This suggests that
at any time during CNS inflammation both T cell subsets may be
activated, and their balance could contribute to the extent and
duration of CNS inflammation. Our data indicate that only in a Th1
cytokine-rich environment are microglia induced to display their full
stimulatory potential for Th2 activation, thus highlighting the
complexity of the intracerebral Th1/Th2 balance.
To date, the nature of the signals triggering microglia activation
remains poorly defined, although products released by damaged neurons
and/or glial cells are likely to be involved (1). Compared
with resting microglia, microglia activated during CNS injury or
infection could respond more promptly to Th1-derived signals and behave
as a more potent APC for T cell restimulation. Among the factors
synthesized by reactive CNS cells, particularly microglia and
astrocytes, M-CSF and GM-CSF have been proposed as major stimulators of
microglia proliferation and activation (1), and increased
expression of GM-CSF and M-CSF receptors has been observed in the
injured CNS (37). We have shown that the potency of adult
mouse microglia in restimulating Th1 cells was increased after exposure
to GM-CSF, but not M-CSF. These data support a previous study showing
that GM-CSF increases the capacity of long term cultured neonatal mouse
microglia to induce Ag-dependent T cell proliferation
(27), but are at variance with another study showing an
inhibitory effect of GM-CSF on microglia APC function
(28). Because GM-CSF fails to stimulate the expression of
MHC class II and adhesion/costimulatory molecules in adult mouse
microglia, its mode of action does not appear to involve induction of
typical APC features. The ability of GM-CSF-primed microglia to sustain
Th1, but not Th2, activation suggests that GM-CSF could increase the
responsiveness of microglia to Th1-derived signals. Because our initial
experiments indicate that, at variance with what was reported for
monocytes (38), GM-CSF does not enhance the expression of
IFN- receptors on microglia (F. Aloisi and R. De Simone, unpublished
observations), future studies will explore the possibility that GM-CSF
affects some steps in the IFN- signal transduction pathway or acts
on other microglia-activating pathways.
In conclusion, the present data suggest that Th1 cells recognizing viral or self Ags expressed in the CNS may activate resting as well as GM-CSF-primed microglia, resulting in the generation of an effective APC capable of contributing to the local Th1/Th2 balance. Because mediators released by Th1-activated microglia (e.g., IL-12 and PGE2) as well as costimulatory molecules expressed on their membranes are involved in the regulation of T cell effector function and survival, it is likely that mutual interactions between microglia and Th1 cells might be determinants of the outcome of CNS inflammation.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Francesca Aloisi, Neurophysiology Unit, Laboratory of Organ and System Pathophysiology, Istituto Superiore di Sanità, Viale Regina Elena 299, 00161 Rome, Italy.
3 Abbreviations used in this paper: DC, dendritic cell; GVHD, graft-vs-host disease.
Received for publication August 9, 1999. Accepted for publication December 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-treated microglial cells. Eur. J. Immunol. 17:1271.[Medline]
are required for IL-12 but not prostaglandin E2 secretion by microglia during antigen presentation to Th1 cells. J. Immunol. 162:1384.-independent antigen presentation function. J. Neuroimmunol. 42:87.[Medline]
-induced MHC class II expression by lipopolysaccharides. J. Immunol. 155:5826.[Abstract]
receptors but suppression of IFN--induced expression of the gene IP-10. J. Immunol. 150:2383.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  E. D. Ponomarev, L. P. Shriver, K. Maresz, J. Pedras-Vasconcelos, D. Verthelyi, and B. N. Dittel GM-CSF Production by Autoreactive T Cells Is Required for the Activation of Microglial Cells and the Onset of Experimental Autoimmune Encephalomyelitis J. Immunol., January 1, 2007; 178(1): 39 - 48. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. D. Ponomarev, L. P. Shriver, and B. N. Dittel CD40 Expression by Microglial Cells Is Required for Their Completion of a Two-Step Activation Process during Central Nervous System Autoimmune Inflammation J. Immunol., February 1, 2006; 176(3): 1402 - 1410. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G.-X. Zhang, S. Yu, B. Gran, and A. Rostami Glucosamine Abrogates the Acute Phase of Experimental Autoimmune Encephalomyelitis by Induction of Th2 Response J. Immunol., December 1, 2005; 175(11): 7202 - 7208. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Schmidt, J. Zhao, J. Chain, D. Hepburn, B. Gitter, G. Sandusky, S. Chintalacharuvu, A. Glasebrook, and S. Na Resistance to Myelin Oligodendrocyte Glycoprotein-Induced Experimental Autoimmune Encephalomyelitis by Death Receptor 6-Deficient Mice J. Immunol., August 15, 2005; 175(4): 2286 - 2292. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Gregerson, T. N. Sam, and S. W. McPherson The Antigen-Presenting Activity of Fresh, Adult Parenchymal Microglia and Perivascular Cells from Retina J. Immunol., June 1, 2004; 172(11): 6587 - 6597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Guillemin and B. J. Brew Microglia, macrophages, perivascular macrophages, and pericytes: a review of function and identification J. Leukoc. Biol., March 1, 2004; 75(3): 388 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G.-X. Zhang, S. Yu, B. Gran, J. Li, I. Siglienti, X. Chen, D. Calida, E. Ventura, M. Kamoun, and A. Rostami Role of IL-12 Receptor {beta}1 in Regulation of T Cell Response by APC in Experimental Autoimmune Encephalomyelitis J. Immunol., November 1, 2003; 171(9): 4485 - 4492. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Granucci, F. Petralia, M. Urbano, S. Citterio, F. Di Tota, L. Santambrogio, and P. Ricciardi-Castagnoli The scavenger receptor MARCO mediates cytoskeleton rearrangements in dendritic cells and microglia Blood, October 15, 2003; 102(8): 2940 - 2947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Monsonego, J. Imitola, V. Zota, T. Oida, and H. L. Weiner Microglia-Mediated Nitric Oxide Cytotoxicity of T Cells Following Amyloid {beta}-Peptide Presentation to Th1 Cells J. Immunol., September 1, 2003; 171(5): 2216 - 2224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Gregerson and J. Yang CD45-Positive Cells of the Retina and Their Responsiveness to In Vivo and In Vitro Treatment with IFN-{gamma} or Anti-CD40 Invest. Ophthalmol. Vis. Sci., July 1, 2003; 44(7): 3083 - 3093. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Gran, G.-X. Zhang, S. Yu, J. Li, X.-H. Chen, E. S. Ventura, M. Kamoun, and A. Rostami IL-12p35-Deficient Mice Are Susceptible to Experimental Autoimmune Encephalomyelitis: Evidence for Redundancy in the IL-12 System in the Induction of Central Nervous System Autoimmune Demyelination J. Immunol., December 15, 2002; 169(12): 7104 - 7110. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Re, S. L. Belyanskaya, R. J. Riese, B. Cipriani, F. R. Fischer, F. Granucci, P. Ricciardi-Castagnoli, C. Brosnan, L. J. Stern, J. L. Strominger, et al. Granulocyte-Macrophage Colony-Stimulating Factor Induces an Expression Program in Neonatal Microglia That Primes Them for Antigen Presentation J. Immunol., September 1, 2002; 169(5): 2264 - 2273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Magnus, A. Chan, O. Grauer, K. V. Toyka, and R. Gold Microglial Phagocytosis of Apoptotic Inflammatory T Cells Leads to Down-Regulation of Microglial Immune Activation J. Immunol., November 1, 2001; 167(9): 5004 - 5010. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. McQualter, R. Darwiche, C. Ewing, M. Onuki, T. W. Kay, J. A. Hamilton, H. H. Reid, and C. C.A. Bernard Granulocyte Macrophage Colony-stimulating Factor: A New Putative Therapeutic Target in Multiple Sclerosis J. Exp. Med., September 24, 2001; 194(7): 873 - 882. [Abstract] [Full Text] [PDF]
|
|
