Aspergillus fumigatus Induces Innate Immune Responses in
Alveolar Macrophages through the MAPK Pathway
Independently of TLR2 and TLR41
Marc Dubourdeau,* Rafika Athman,†Viviane Balloy,‡Michel Huerre,§Michel Chignard,‡
Dana J. Philpott,†Jean-Paul Latge ´,* and Oumaı ¨ma Ibrahim-Granet2*
Aspergillus fumigatus causes invasive aspergillosis in immunosuppressed patients. In the immunocompetent host, inhaled conidia
are cleared by alveolar macrophages. The signaling pathways of the alveolar macrophage involved in the clearance of A. fumigatus
are poorly understood. Therefore, we investigated the role of TLRs in the immune response against A. fumigatus and their
contribution to the signaling events triggered in murine alveolar macrophages upon infection with A. fumigatus conidia. Specif-
ically, we examined the MAPKs and NF-?B activation and cytokine signaling. Our investigations revealed that immunocompetent
TLR2, TLR4, and MyD88 knockout mice were not more susceptible to invasive aspergillosis as compared with wild-type mice and
that the in vitro phosphorylation of the MAPKs ERK and p38 was not affected in TLR2, TLR4, or MyD88 knockout mice following
stimulation with conidia. In vivo experiments suggest that ERK was an essential MAPK in the defense against A. fumigatus,
whereas the activation of NF-?B appeared to play only a secondary role. In conclusion, our findings demonstrate that TLR2/4
recognition and MyD88 signaling are dispensable for the clearance of A. fumigatus under immunocompetent situations. Further-
more, our data stress the important role of ERK activation in innate immunity to A. fumigatus. The Journal of Immunology, 2006,
to occur in various immunocompromised populations such as neu-
tropenic patients with hematologic malignancies, children with
chronic granulomatous diseases, and transplant recipients (1). Di-
agnosis is difficult, treatment is inefficient, and, as a consequence,
mortality is high.
One of the most striking conclusions of a literature survey on A.
fumigatus is how little is known about the pathobiological features
of this organism (2). This is especially true for the early stages of
disease development. Following inhalation of airborne A. fumiga-
tus conidia, the immunocompetent host is protected by pulmonary
innate immunity, which includes phagocytosis by alveolar macro-
phages (AMs).3Macrophages are key defenders of the lung and
play an essential role in mediating the inflammatory response.
Killing of conidia by the AMs begins 6 h after phagocytosis.
Swelling of conidia inside the AM is a prerequisite for killing. The
contribution of phagolysosomal acidification as well as NADPH
spergillus fumigatus is an opportunistic fungal pathogen
responsible for invasive aspergillosis, a severe, life-
threatening infection. Invasive aspergillosis is well known
oxidase to the conidicidal activity has been demonstrated. This
activity is inhibited by corticosteroids via impairment of the pro-
duction of reactive oxidant intermediates (3, 4). If end-killing
mechanisms are documented, the signaling pathways controlling
these mechanisms are not known.
Recognition of invading microorganisms by the innate immune
system is a first and essential step in their successful elimination.
Essential receptors for the recognition of invading microorganisms
are the TLRs and members of the NOD (nucleotide binding oli-
gomerization domain) family of proteins, including Nod1 and
Nod2. These receptors trigger the initiation of inflammatory sig-
nals. Deficiencies of specific TLRs and various adaptor proteins of
the TLR signaling transduction pathway, including MyD88, have
been shown to result in reduced pyogenic bacteria and Myco-
plasma species clearance (5, 6).
Following microbial recognition, the macrophage effector func-
tions are known to be dependent in part on the phosphorylation and
activation of MAPKs. The ERKs and p38 are activated in macro-
phages (7) and AMs by a variety of stimuli (8, 9). After phosphor-
ylation, the activated MAPKs translocate to the nucleus where they
phosphorylate several targets, including transcription factors that
then mediate the expression of a number of proinflammatory
The involvement of TLRs in the recognition of A. fumigatus by
macrophages has been investigated by different groups. Data on
this subject are in conflict, because evidence both for and against
the importance of TLRs in host defense has been reported (10–14).
Other investigations attempted to clarify the role of TLRs against
A. fumigatus by using adaptor protein MyD88 knockout mice.
These findings are not without controversy, because macrophages
from MyD88 knockout mice exposed to A. fumigatus hyphae, but
not conidia, still have inflammatory responses (15).
In this study we investigated the role of TLRs in A. fumigatus
pathogenesis. We found that TLRs are dispensable for the survival
*Unite ´ des Aspergillus,†Groupe Immunite ´ Inne ´e et Signalisation,‡Unite ´ de De ´fense
Inne ´e et Inflammation, Institut National de la Sante ´ et de la Recherche Me ´dicale
E336, and§Unite ´ de Recherche et d’Expertise Histotechnologie et Pathologie, Institut
Pasteur, Paris, France
Received for publication October 18, 2005. Accepted for publication June 7, 2006.
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.
1Institut Pasteur provided financial support for this study through the Programme
Transversal de Recherche, “A. fumigatus and the Alveolar Macrophage.” M.D. and
R.A. were supported by postdoctoral fellowships from Institut Pasteur.
2Address correspondence and reprint requests to Dr. Oumaı ¨ma Ibrahim-Granet, In-
stitut Pasteur, Unite ´ des Aspergillus, 25 Rue du Dr. Roux, 75724 Paris, Cedex 15,
France. E-mail address: email@example.com
3Abbreviations used in this paper: AM, alveolar macrophage; BAL, bronchoalveolar
lavage; MAM, murine AM; MKP, MAPK phosphatase.
The Journal of Immunology
Copyright © 2006 by The American Association of Immunologists, Inc.0022-1767/06/$02.00
of mice during infection or for phagocytosis, killing efficiency,
cytokine production, and the initial signaling events triggered upon
infection with A. fumigatus conidia. Overall, our data suggest that,
for the parameters that we examined, innate immune signaling by
A. fumigatus is independent of TLR2 and TLR4 recognition and
Materials and Methods
Reagents and antibodies
All reagents were obtained from Sigma-Aldrich unless otherwise stated.
Phospho-specific Abs to the MAPKs ERK1/2 and p38 and HRP-conju-
gated anti-mouse and anti-rabbit Ig were purchased from Cell Signaling
Technology. The mouse mAb to MAPK phosphatase (MKP) 2, specific for
the MKP carboxyl-terminal catalytic domain, was obtained from BD
Transduction Laboratories. Rabbit polyclonal NF-?B p65 and I?B-? Abs
were obtained from Santa Cruz Biotechnology. The MEK inhibitors
PD98059 and UO126 and the p38 inhibitor SB203580 were purchased
from Cell Signaling Technology.
The A. fumigatus CBS 144.89 strain was a clinical isolate. Resting conidia
were harvested by washing a 7 day-old slant culture with PBS supple-
mented with 0.1% Tween 20 (PBST). The suspension was filtered through
a 40-?m cell strainer (Falcon) to separate conidia from the contaminating
mycelium and verified microscopically (100% resting conidia). In the kill-
ing experiments, the conidia were labeled with FITC as previously de-
scribed. Briefly, freshly harvested conidia (2 ? 108per 10 ml of 0.05 M
sodium carbonate buffer (pH 10.2)) were incubated with FITC at a final
concentration of 0.1 mg/ml at 37°C for 1 h and washed by centrifugation
three times in PBST (4).
Mouse strains and infections
Several wild-type and knockout mouse strains were used. TLR2, TLR4,
and MyD88?/?mice were obtained from S. Akira (Osaka University,
Osaka, Japan) and backcrossed eight times with C57/BL6 to ensure similar
genetic backgrounds. C57BL/6 mice were used as a control. These mice, as
well as 20- to 24-g, 6- to 8-wk-old male out-bred Swiss OF1 mice were
supplied by the Centre d’Elevage R. Janvier and used at ?8 wk of age.
Mice were fed normal mouse chow and water ad libitum and were reared
and housed under standard conditions with air filtration. Mice were cared
for in accordance with Pasteur Institute guidelines in compliance with Eu-
ropean Union animal welfare regulations.
Mice were immunosuppressed with 25 mg of cortisone acetate (Sigma-
Aldrich) injected i.p. at day 3 and immediately before intranasal inocula-
tion of conidia (day 0) for infection experiments. Before infection, mice
were anesthetized with an intracellular injection of 0.1 ml of a solution
containing 10 ?g/ml ketamine (Merial) and 2 ?g/ml xylasine (Bayer) per
mouse. Conidia in 25 ?l of PBST were inoculated intranasally using an
automatic pipetting device. Control nonimmunosuppressed mice received
the same conidial suspension. Survival of mice was followed daily over a
period of 17 days.
Lungs were collected at day 4 after intranasal infection with 1 ? 10 (7)
conidia. Organs were fixed in 3.7% neutral-buffered formaldehyde, em-
bedded in paraffin, and cut into 5-mm-thick sections. Sections were stained
with methenamine silver for fungal detection and examined microscopi-
MH-S murine AMs (MAMs) (17), derived by SV40 transformation of
MAMs, were cultured in RPMI 1640 complete medium (supplemented
with HEPES and sodium pyruvate) containing 10% FCS and maintained in
an incubator at 37°C under 5% CO2. MH-S cells were plated 24 h before
experimentation and starved for at least 5 h before stimulation.
Primary MAMs were harvested from mouse lungs with 15 ml of ice-
cold Ca2?- and Mg2?-free PBS through a 18-gauge plastic catheter in-
serted into the trachea after cervical dissection as described previously (3).
Primary MAMs were separated from the lavage fluid by centrifugation
(400 ? g for 8 min at 4°C), and the cell pellet was suspended in RPMI
1640 complete medium containing 10% FCS and plated on 12-mm-diam-
eter glass coverslips at a density of 5 ? 105or 1 ? 105cells per coverslip
in 96- well ELISA plates. Nonadherent cells were removed by washing
after a 2-h incubation at 37°C in an atmosphere of 5% CO2. The viability
of the AM preparations was ?99% as judged by trypan blue dye exclusion.
Coculture of conidia with AMs
For MAPK phosphorylation experiments, AMs were starved in RPMI 1640
medium without FCS for 5–16 h, challenged with 10 conidia per macro-
phage, and subsequently incubated at 4°C for 30 min. Unbound conidia
were removed by washing with complete cold RPMI 1640 medium, and the
cells were shifted to 37°C in an atmosphere of 5% CO2.
In some experiments, the MEK inhibitors PD98059 and UO126 and the
p38 inhibitor SB203580 were added 30 or 60 min before infection at ap-
propriate concentrations as indicated in Fig. 4.
Phagocytosis assays were performed as previously described (4). AMs
were challenged with two conidia per macrophage and were subsequently
incubated at 4°C for 30 min. Unbound conidia were removed by washing
with cold RPMI complete medium, and phagocytosis was initiated at 37°C
in an atmosphere of 5% CO2. Phagocytosis was stopped after 2 h by wash-
ing the AMs with ice-cold PBS and subjecting them to fixation with 3%
formaldehyde in PBS (10 min at room temperature) followed by three
washes with 50 mM NH4Cl in PBS. The number of A. fumigatus conidia
engulfed by macrophages was determined using a polyclonal anti-conidia
Ab and a secondary Ab conjugated to Texas Red. This procedure labels
only uningested conidia. The percentage of ingestion was estimated as the
ratio of the number of ingested conidia to the total number of conidia
bound to 100 macrophages.
MAMs or MH-S cells were stimulated for 16 h with FITC-labeled conidia
at a ratio of 1:10 and lysed with 0.2 ml of water. The lysate was suspended
by pipetting, incubated overnight at 4°C, supplemented with 0.2 ml of a
medium containing 4% glucose and 2% mycopeptone, and incubated for
6–8 h at 37°C. The killing percentage (number of nongerminated conidia
per 100 FITC labeled-conidia) was assessed by fluorescent microscopy.
Preparation of protein extracts and immunoblot analysis
After incubation with conidia for various times, cells were washed with
ice-cold PBS and lysed with ice-cold lysis buffer (150 mM NaCl, 20 mM
Tris-HCl (pH 7.4), 5 mM EDTA, 1% Nonidet P-40, 0.1% SDS, and 0.5%
sodium deoxycholate) supplemented with 1 mM PMSF, 1 mM Na3VO4,
and 50 mM NaF. Lysates were cleared by centrifugation at 13,000 rpm for
15 min at 4°C. Equal amounts of total protein extracts were suspended in
2? Laemmli buffer under reducing conditions, boiled for 5 min, and sep-
arated by SDS-PAGE. The MAPKs ERK, p38, and MKP-2 were detected
by immunoblotting according to standard procedures.
Blots were blocked for 1 h with 5% low-fat lyophilized milk in TBS, 50
mM Tris-HCl (pH 7.5), and 0.15 M NaCl and supplemented with 0.1%
Tween 20 (TBST).
Membranes were washed three times with TBST and probed overnight
with specific anti- phospho-ERK1/2 rabbit Ab (1/1000), anti-phospho p38
mAb (1/500), and anti- MKP-2 polyclonal Ab or with anti-ERK1/2 or
anti-p38 rabbit Abs (1/1000) for total proteins. Membranes were subse-
quently treated with HRP-conjugated Abs (1/1000). The blots were devel-
oped with an ECL kit. Levels of phosphorylation were quantified after
normalization to the level of total ERK and p38 using densitometry scan-
ning and the ImageQuant program.
For in vitro cytokine expression, MAMs or MH-S cells were added to
96-well plates at a density of 1 ? 105cells and then stimulated with conidia
at a conidium-to-macrophage ratio of 10:1 for 6 h. Supernatants were re-
covered and frozen at ?70°C before analysis. For other experiments,
MAPK inhibitors were added before stimulation as indicated earlier. As a
positive control, cells were treated with LPS at a concentration of 5 ?g/ml.
The expression of TNF-? in the cell supernatants was analyzed by
ELISA using the mouse TNF-? detection kit (BD Biosciences).
For in vivo experiments, cytokine expression was measured in bron-
choalveolar lavage (BAL) fluids of mice infected for 24 h with 1 ? 107
IL-6, MCP-1, and TNF-? were analyzed by cytometry using the In-
vitrogen Life Technologies mouse inflammation cytometric bead array six-
bead kit according to manufacturer’s instructions. RANTES, KC, and G-
CSF were measured by ELISA.
3995The Journal of Immunology
NF-?B nuclear translocation.
glass coverslips (25-mm diameter) and starved overnight before stimula-
tion with A. fumigatus conidia. Resting conidia were incubated with MH-S
cells at a conidium-to-macrophage ratio of 10:1. Phagocytosis was syn-
chronized by a 30-min incubation of cells on ice. Cells were rinsed and
incubated with serum-free medium at 37°C. Stimulation of MH-S cells was
stopped at 1, 4, and 6 h.
As a positive control for NF-?B translocation, cells were stimulated
with TNF-? for 1 h. After different times of simulation, cells were fixed
with 3% formaldehyde (paraformaldehyde), permeabilized with 0.2% Tri-
ton X-100, and incubated with specific Abs. NF-?B was stained with a
mouse mAb directed against the Rel-A p65 subunit, (1/200; Santa Cruz
Biotechnology). Cells were then incubated with FITC-labeled mouse sec-
ondary Ab. Nuclei were visualized using 4?,6-diamidino-3-phenylindole
(10 ?g/ml). Cells were mounted with PBS/glycerol (50%) and viewed
using fluorescence microscopy (Leica). Images acquired at the same ex-
posure time were processed by Photoshop software (Adobe).
Cells were incubated with resting conidia for 4 and
6 h, rinsed with PBS, and collected by removing cells with Laemmli buffer
(1?). Nonstimulated cells were used as controls. I?B-? was detected by
Proteins were separated on a 13% SDS-polyacrylamide gel and trans-
ferred to a nitrocellulose membrane (0.2 ?m; Schleicher and Schuell).
After incubation with the anti-I?B-? rabbit polyclonal Ab (1/1000), the
membrane was washed in 0.5% PBS-Tween 20 and probed with a HRP
secondary Ab (1/2000). The blots were developed with an ECL kit.
ELISA based NF-?B activation assay.
nuclear extracts were prepared using the nuclear extract preparation kit
(TransAM) 4 and 6 h after stimulation with resting conidia. Nonstimulated
cells and TNF-?-treated cells were used as negative and positive controls,
MH-S cells (5 ? 105) were attached to
To evaluate NF-?B activation,
respectively. Whole nuclear extracts were incubated with a double-
stranded oligonucleotide containing the NF-?B consensus site (5?-GG
GACTTTCC-3?) previously immobilized in the wells of a 96-well plate.
After washing, bound NF-?B was detected with a p65 Rel-A-specific Ab.
The reaction was developed colorimetrically with a HRP-conjugated sec-
ondary Ab and quantified spectrophotometrically at 450 nm.
Survival data were analyzed by means of log-rank comparisons of Kaplan-
Meier survival curves followed by the Wilcoxon test using JMP 5.0 soft-
ware (SAS). The Fisher test was used to determine the statistical signifi-
cance of differences in phosphorylation of MAPK, cytokine expression,
and inhibition of killing. As indicated in the legends to Figs. 1 and 3–5,
significance was defined as p ? 0.05. In vivo groups consisted of three
animals. The data reported were mean values ? SEM of triplicates from a
Role of TLR2, TLR4, and MyD88 in the innate immune response
against A. fumigatus
Immunocompetent TLR2, TLR4, and MyD88
knockout mice as well as the wild-type C57BL/6 mice were re-
sistant to infection with 1 ? 107conidia of A. fumigatus, even
though this inoculum was able to kill C57BL/6 mice within five
days after cortisone acetate immunosuppression. A Wilcoxon test
on Kaplan-Meier revealed that the survival rate of MyD88 knock-
out mice was not statistically different between knockout and pa-
rental mice (p ? ?2? 0.29) (Fig. 1A). Thin sections of lungs from
after intranasal infection with A. fumigatus conidia. A,
Percentage of survival was measured after intranasal in-
fection of immunocompetent (IC) TLR2, TLR4, and
MyD88 knockout mice and C57BL/6 wild-type mice
with 1 ? 107conidia. C57BL/6 wild-type mice were
immunosuppressed (IS) with cortisone acetate and in-
fected as a positive control for mice susceptibility to an
inoculum of 1 ? 107conidia. B, Lung histology from A.
fumigatus-infected MyD88?/?(a), C57/BL6 immuno-
competent (b), and C57/BL6 immunosuppressed mice
(c) at D4 after intranasal infection of 1 ? 107conidia
with methenamine silver stain at a magnification of
?100. C, In vivo release of TNF-? and IL-6 was mea-
sured in the BAL fluids of immunocompetent (IC) and
immunosuppressed (IS) mice infected (I) intranasally
for 24 h with A. fumigatus conidia. Infected mice were
compared with noninfected mice (NI). Data are ex-
pressed as the mean ? SEM of three independent ex-
periments done with three mice for each point; ?, p ?
Mouse survival and cytokine expression
Aspergillus fumigatus AND THE ALVEOLAR MACROPHAGE
MyD88 knockout infected mice (Fig. 1Ba) as well as lungs from
wild-type infected mice (Fig. 1Bb) displayed normal lung histol-
ogy in comparison to lung sections from immunosuppressed in-
fected mice. Analysis of the lungs from later mice showed a typical
invasive aspergillosis pattern with a massive fungal growth ob-
structing the bronchioli (Fig. 1Bc).
In vivo inflammatory response to A. fumigatus. The production
of TNF-? and IL-6 was assessed in the BAL fluids of mice chal-
lenged with conidia of A. fumigatus for 24 h. These BAL fluids
contained substantially more TNF-? (925 ? 171 vs 0 ? 0 pg/ml)
and IL-6 (64 ? 18 vs 0 ? 0 pg/ml) than samples from noninfected
control mice (p ? 0.05) (Fig. 1C).
A similar pattern of expression of these two cytokines as well as
MCP-1, RANTES, KC, and G-CSF was seen in BAL fluids of
TLR2 knockout, TLR4 knockout, and wild-type mice after 24 h of
infection. These cytokines did not show any significant difference
between TLR-deficient and wild-type mice (data not shown).
In contrast, comparison of TNF-? and IL-6 recovered from BAL
fluids of immunocompetent and immunosuppressed mice that were
obtained after stimulation with A. fumigatus revealed that the con-
centration of these two cytokines was substantially lower in the
immunosuppressed mice (Fig. 1C).
Involvement of MyD88 in phagocytosis and killing of A. fumiga-
We used immunofluorescence microscopy to mea-
sure the capacity of AMs isolated from wild-type C57BL/6 and
MyD88?/?mice to ingest and kill the conidia of A. fumigatus.
After 2 h of phagocytosis, conidia were efficiently internalized by
C57BL/6 and MyD88?/?AMs with an ingestion capacity of 81
and 88%, respectively (see Fig. 2A as an example).
AMs from wild-type C57BL/6 and MyD88?/?mice were chal-
lenged with living FITC-labeled conidia of A. fumigatus to com-
pare the efficiency of intracellular killing. The viability of conidia
was measured after 16 h of coincubation. As for the percentage of
ingestion, there was an equivalent percentage of killing of conidia
with a rate of 33% for the wild-type C57BL/6 and 32% for
MyD88?/?macrophages (see Fig. 2B as an example).
Phosphorylation of MAPKs in response to A. fumigatus
Impact of TLR2, TLR4, and MyD88.
ment of the TLR family as well as that of the adaptor protein
MyD88 in the phosphorylation of the MAPKs of AMs in response
to A. fumigatus, murine primary AMs from TLR2, TLR4, and
MyD88 knockout mice and their parental strains were stimulated
ex vivo with conidia for different periods of time. The activation
kinetic of ERK and p38 MAPKs in the MAMs was assessed by
Western blot analysis using phospho-specific Abs.
Phosphorylation of macrophage MAPKs derived from the
knockout mice strains were compared with those of murine pri-
mary AMs from control C57BL/6 wild-type mice after stimulation
with conidia. In Fig. 3A, we show that the activities of both ERK
and p38 were very low in uninfected macrophages. Upon stimu-
lation with conidia, either in wild-type AMs or AMs from TLR2,
TLR4, and MyD88 knockout mice, ERK and p38 were strongly
activated in response to the conidial stimulation with a significant
mean fold of phosphorylation of 7 ? 1.5 and 3 ? 0.25 (p ? 0.05)
for the ERK p44 and p42 subunits, respectively, and for p38 it was
7 ? 3 (p ? 0.01) after 1 h as compared with noninfected cells.
These results strongly suggest that these receptors as well as the
adaptor protein MyD88 are not required for the phosphorylation of
MAPKs in response to A. fumigatus.
Kinetics of ERK and p38 phosphorylation. Because the MAPKs
were phosphorylated similarly in primary AMs and the cell line,
the follow-up of the activation of MAPKs in AMs in response to
A. fumigatus at later time points postinfection was done using
MH-S cells, because such an assay required a high number of cells
(Fig. 3B). Incubation for up to 8 h increased the levels of phos-
phorylation even further with a significant fold of phosphorylation
of 8 ? 1.5 and 6.5 ? 0.25 for the ERK p44 and p42 subunits,
respectively, and for p38 it was 14 ? 3 (p ? 0.005) after 8 h as
compared with noninfected cells. An incubation time longer than
8 h could not be used because the high ratio (10 conidia per mac-
rophage) resulted in overgrowth of nonphagocytosed conidia.
In mammalian cells, inactivation of MAPKs is primarily con-
ducted by a group of dual-specificity MKPs (18). To examine
whether MKP-2 plays a role in the regulation of MAPKs in AMs,
we investigated the activation of MKP-2 after stimulation of MH-S
macrophages with the conidia of A. fumigatus for different times.
As shown in Fig. 3C, MKP-2 was not activated even after the
stimulation of cells with conidia for 8 h. This result is in accor-
dance with the long-term phosphorylation of MAPKs after stim-
ulation with conidia.
Activation of ERK and p38 in vivo; impact of immunosuppres-
sion. To investigate whether MAPKs play a role in the responses
of immunocompetent and immunosuppressed mice infected with
conidia of A. fumigatus, we studied the activation of ERK and p38
in AMs isolated after intranasal infection with 1 ? 107conidia.
MAMs were recovered as indicated in Materials and Methods, and
the phosphorylation of MAPKs was investigated in the cell lysates
by Western blotting. The level of phosphorylation was compared
separately between immunocompetent noninfected and infected
mice as well as between immunosuppressed noninfected and in-
fected mice. Similar to what was observed ex vivo, after 4 h of
To examine the involve-
MAMs isolated from MyD88 knockout (KO) and C57BL/6 mice. A, AMs
obtained by BAL from knockout and wild-type mice were infected with
FITC-labeled conidia at a ratio of two conidia per macrophage for 2 h.
Cells were fixed, treated with anti-conidia Ab, stained with Texas Red-
conjugated Ab (T-red) to label uningested conidia, permeabilized, and pro-
cessed for Hoechst nuclear staining. The green filter shows ingested and
uningested conidia. The red filter shows uningested conidia. DAPI, 4?,6-
diamidino-3-phenylindole; DIC, differential interference contrast. B, AMs
obtained by BAL from knockout and wild-type mice were infected with
FITC- labeled conidia at a ratio of one conidium per 10 macrophages for
16 h. After water osmotic shock, the conidia were allowed to germinate for
8 h at 37°C. The outside growth was stained with the anti-conidia Ab as
indicated above. Arrows indicate killed conidia, and arrowheads point to-
ward germinated conidia.
Examples of A. fumigatus conidia phagocytosed by primary
3997The Journal of Immunology
infection with conidia in immunocompetent mice the infection re-
sulted in a major activation of ERK with a phosphorylation fold of
2.4 and 1.9 for the p44 and p42 subunits, respectively (p ? 0.05).
In contrast, in immunosuppressed mice the level of phosphorylated
ERK was lower than that in immunocompetent mice (p ? 0.05).
In addition, the phosphorylation level of ERK remained extremely
low after infection with conidia (Fig. 3D).
Although an infection for 4 h showed a significant phosphory-
lation of ERK in vivo, the activation of p38 could not be observed
when 1 ? 107conidia were used. As shown in Fig. 3E, a similar
level of phosphorylation of p38 was observed in both noninfected
and infected immunocompetent and immunosuppressed mice.
Study of the inhibition of MAPKs; effect on cytokine release and
killing of A. fumigatus. Primary MAMs were stimulated ex vivo
with conidia for 6 h in the presence or absence of inhibitors of
MAPKs. Supernatants were removed and used for determination
of the TNF-? concentrations.
As shown in Fig. 4A, conidia induced a secretion of 137 ? 26
pg/ml TNF-? after 6 h of stimulation, which corresponds to an
8.6-fold increase in comparison with the basal level of secretion of
this cytokine in unstimulated cells (16 ? 2 pg/ml). The two in-
hibitors of MEK1/2, PD 98059 and UO126, caused a significant
decrease of TNF-? secretion between 46 and 86%, depending on
the inhibitor and the dose used (p ? 0.05). A minor but significant
inhibition (p ? 0.05) was obtained with SB203580, which is an
inhibitor of p38, at all concentrations tested. However, in contrast
to the inhibitors of ERK, SB203580 did not induce any dose-de-
pendent inhibition of TNF-? synthesis. Similar results on cytokine
production and inhibition have been obtained using the MH-S
The effect of PD98059 and SB203580 in the capacity of killing
conidia was also investigated. Because MH-S cells behave like
murine primary cells in terms of cytokine production as well as
with the conidia of A. fumigatus. A, Primary AMs isolated from MyD88,
TLR2, and TLR4 knockout (KO) mice as well as from C57BL/6 wild-type
mice were stimulated with conidia of A. fumigatus at a ratio of 10 conidia
per macrophage for the indicated time points. ERK and p38 phosphoryla-
tion was analyzed by immunofluorescence using Abs against phospho-
ERK and phospho-p38. To confirm the sample loadings, the blots were
stripped and probed with an anti-total ERK or anti-total p38 Abs. Levels of
phosphorylation were quantified after normalization to the level of total
ERK and p38 using densitometry scanning and the ImageQuant program.
The data shown are representative of three independent experiments. NI is
the noninfected control. B, MH-S cells were stimulated with conidia at a
ratio of 10:1 (conidia/macrophage) at the time points indicated. Phosphor-
ylation of ERK and p38 was analyzed by immunofluorescence as indicated
above. C, Western blot analysis of MKP-2 from B that was stripped again
and probed with the anti MKP-2 Ab. The data shown are representative of
three independent experiments. NI is the noninfected control. D and E,
AMs were harvested by BAL from immunocompetent (IC) and immuno-
suppressed mice (IS) 4 h after intranasal infection (I) with 1 ? 107conidia
or from PBS-inhaled control mice (NI). Phosphorylated and nonphospho-
rylated forms of ERK and p38 were analyzed by Western blotting as de-
scribed earlier. Levels of phosphorylation were quantified as indicated ear-
lier. Each bar represents the average of fold phosphorylation of ERK p44
and p42 subunits (D) or p38 (E) from three independent mice. Levels of
phosphorylation were compared between noninfected and infected mice
and between immunocompetent and immunosuppressed mice and analyzed
statistically (?, p ? 0.05).
ERK and p38 phosphorylation after stimulation of AMs
MAPKs. A, Primary MAMs were stimulated with conidia in the absence or
presence of the indicated concentration of ERK or p38 inhibitors. Release
of TNF-? was measured by ELISA in cell supernatants 6 h after infection.
As a positive control, cells were stimulated with LPS. Results are the
mean ? SE of triplicates in a representative experiment; ?, p ? 0.05. B,
Killing of conidia in presence or absence of MAPK inhibitors was esti-
mated as the percentage of nongerminating conidia following a 16-h in-
fection. Values represent mean ? SD from three experiments; ?, p ? 0.05.
Cytokine release and killing of A. fumigatus and the role of
Aspergillus fumigatus AND THE ALVEOLAR MACROPHAGE
MAPK phosphorylation in response to the stimulation with A. fu-
migatus, the effect of MAPK inhibitors on the killing of conidia
was studied using the MH-S cells. Fig. 4B shows that PD98059
resulted in a percentage of killing of 53 ? 4% compared with 83 ?
8% for the control cells without inhibitors (p ? 0.05). SB203580
caused a slight, nonsignificant reduction in the percentage of
Translocation of NF-?B by A. fumigatus in macrophages
Because NF-?B is associated with the regulation of immune re-
sponses in a variety of infection models, we analyzed the activa-
tion of this transcription factor following stimulation of macro-
phages with A. fumigatus. Translocation of NF-?B was analyzed
by immunofluorescence using a Rel-A specific Ab. MH-S cells
were stimulated with conidia for 1, 4, and 6 h. As a positive control
of NF-?B translocation, cells were treated with TNF-? for 1h. As
shown in Fig. 5A, TNF-? treatment resulted in complete nuclear
translocation of NF-?B after 1 h of incubation. Nonstimulated con-
trol cells as well as cells stimulated with conidia for 1 h showed no
Rel-A translocation but showed ?25% translocation after 4 h of
stimulation with conidia and reached 100% after 6 h. This result
implied that efficient Rel-A translocation was triggered only after
intracellular swelling of conidia or, indeed, by paracrine effects of
factors released following infection. To go further, we tested the
ability of the translocated Rel-A subunit to bind to its target DNA
sequence by use of an ELISA-based assay (TransAM). Nuclear
extracts were prepared from cells stimulated with conidia for 4 and
6 h. As shown in Fig. 5B, resting conidia induce an activation fold
of 1.6 after 6 h of stimulation compared with nonstimulated cells.
This fold activation is statistically identical with that after stimu-
lation with TNF-?.
Another way to evaluate NF-?B activation is to examine I?B-?
degradation. MH-S macrophages were stimulated either with
TNF-? as a control or with resting conidia for 4 and 6 h. As shown
in Fig. 5C, the addition of conidia resulted in an efficient degra-
dation of I?B-? only after 4 and 6 h of stimulation, similarly as the
TNF-? stimulation control.
A. fumigatus is a human pathogen that is able to cause invasive
aspergillosis in immunosuppressed patients. In the immunocom-
petent patient, inhaled conidia are easily cleared by the immune
system. Macrophages play an essential role in this process. Their
activation results in the acquisition of new molecular and cellular
capacities that are determined largely by new gene expression. The
up-regulation of specific genes depends mainly on the mode of
activation and frequently results in production of the proinflam-
matory cytokines including TNF-?, IL1-?, and IL-6. Indeed, one
of the major cytokines involved in immune defense against A.
fumigatus is TNF-?. A high level of TNF-? synthesis is directly
accompanied by an increased resistance against A. fumigatus in-
Secretion of cytokines for immune defense often results from
the stimulation of TLRs. Regarding A. fumigatus, the current lit-
erature on this topic is somewhat puzzling, because although it is
extensively studied, conflicting results are presented (20). In mice,
both conidia and hyphae stimulate cytokine production via TLR2
(11), whereas conidia signal solely through TLR4 to stimulate mu-
rine macrophages (12). In contrast to those findings, the results of
Wang et al. (21) implicate TLR4 but not TLR2 for the stimulation
of human monocytes by A. fumigatus hyphae. Furthermore, Mam-
bula et al. (11) reported that TNF-? production in response to A.
fumigatus is MyD88 dependent. Contrary to that finding, responses
to A. fumigatus have been reported by Marr et al. (15) to occur in
a MyD88-independent manner.
Our results support the idea that innate immune signaling by A.
fumigatus in immunocompetent MAMs is largely TLR2 and TLR4
independent. First, we observed a late translocation of NF-?B in
response to the conidia, suggesting that direct stimulation of the
TLR- NF-?B axis does not occur following A. fumigatus stimula-
tion of AMs. Indeed, stimulation of TLR by direct and specific
binding of conidia would likely result in rapid translocation of
NF-?B, which was not the case in our studies. Second, we ana-
lyzed phosphorylation of ERK and p38 after stimulation with
conidia from A. fumigatus in TLR2-, TLR4-, and MyD88-deficient
AMs. Neither these receptors nor the adaptor protein MyD88 ap-
peared to play a role in the phosphorylation of ERK and p38 in
vitro. Third, analysis of the cytokine and chemokine profiles 24 h
after infection of the TLR-deficient mice compared with wild-type
mice gave the indication that these receptors are not involved in
the innate immune response against A. fumigatus, because no sig-
nificant differences were seen between the two groups. Further-
more, TLR2, TLR4, and MyD88 knockout immunocompetent
stimulation with the conidia of A. fumigatus. A, MH-S cells were chal-
lenged with the conidia at the ratio of 10 conidia per macrophage for
different times or with TNF-? for 1 h as a positive control. NF-?B trans-
location was visualized using a Rel-A-specific polyclonal Ab (upper green
panel). 4?,6-Diamidino-3-phenylindole nuclear staining is shown in the
lower panel. Stained cells were observed using conventional fluorescence
microscopy. Arrows indicate nuclei with prominent concentrations of
translocated Rel-A. NI is the noninfected control. B, Time course of NF-?B
activation and I?B-? degradation in MH-S cells challenged with the
conidia of A. fumigatus. MH-S cells were challenged as indicated in A.
Nuclear extracts were assayed for NF-?B DNA binding activity by the
TransAM method; ?, p ? 0.05. C, Western blot analysis of cytoplasmic
levels of I?B-? in cells infected with A. fumigatus conidia (A.f) or treated
Translocation and activation of NF-?B in MH-S cells after
3999 The Journal of Immunology
mice challenged with 1 ? 107conidia did not develop aspergil-
losis similar to that seen in wild-type strain, which is in agree-
ment with others stating that TLR2?/?and TLR4?/?immuno-
competent mice do not display an increased susceptibility to
aspergillosis (22). As well, the phagocytosis and intracellular
killing of the conidia in MyD88?/?mice was similar to that of
wild-type mice (15).
We also investigated the involvement of different MAPKs in the
response toward A. fumigatus conidia, because these kinases are
important also for cytokine production (23, 24). However, their
exact functions in the response to fungal infection are poorly stud-
ied (25). Even less is known about the involvement of MAPKs in
signaling and the killing of conidia of A. fumigatus by AMs.
Through the use of phospho-specific Abs, we found an activa-
tion of ERK and p38 after in vitro stimulation with conidia. The
phosphorylation occurred at early time points postinfection and
was sustained for at least 8 h, which correlated with the absence of
MKP-2 induction. The sustained late activation of ERK and p38
may suggest the involvement of the swelling state of conidia in the
innate response against A. fumigatus. We were the first to show
that 6 h after phagocytosis, conidia significantly increase their size
but are inhibited in germinating within the phagolysosome. We
demonstrated that this swelling state is an absolute requirement for
production of the reactive oxygen intermediates that are responsi-
ble for conidial killing (3). Follow-up studies have shown that
conidial swelling is sufficient to induce inflammatory cytokine/
chemokine production by dendritic cells, AMs, and human CD14?
macrophages after stimulation either with conidia for 18 h or with
swollen conidia (26, 27). Although ERK and p38 were activated
after stimulation of conidia in vitro, activation of p38 in vivo was
only observed when 2 ? 108conidia were applied, whereas a
strong response of ERK was observed by the use of at least 10
times fewer conidia. This finding suggested that p38 might only be
of minor importance in recognition and defense against A.
MAPKs and NF-?B are known to be involved in the control of
cytokine gene expression, including that of TNF-?, IL-1?, and
IL-6. In the immune response toward A. fumigatus, TNF-? has
been shown to play a critical role in innate immunity in both im-
munocompromised and immunocompetent hosts. Pretreatment
with TNF-? agonist peptides significantly enhance resistance to A.
fumigatus in neutropenic animals (19). Moreover, the reduction in
the level of cytokines released by macrophages after treatment
with corticosteroids leads to insufficient clearance of conidia (28).
In our study, conidia induced a significant increase in TNF-? pro-
duction by AMs after 6 h of stimulation in vitro. When mice were
challenged in vivo for 24 h, TNF-? production and IL-6 increased
substantially. Our results and those of others (26, 27, 29) on the
modulation of the expression of TNF-? in macrophages after ex-
posure to live but not heat-killed conidia for 6 h as well as to
swollen conidia suggest that TNF- ? and IL-6 are essential cyto-
kines in the host defense against exposure to A. fumigatus conidia.
Investigations with MAPK inhibitors were performed to deter-
mine the relationship between the production of TNF-? and the
phosphorylation of ERK and p38. The use of SB203580, the clas-
sically known inhibitor of p38, had a minor effect on TNF-? pro-
duction independently of the dose used. Additionally, we were not
able to observe any significant variation in the killing of conidia
under in vitro conditions using SB203580, which underlines the
minor role of this MAPK. In contrast to p38, we revealed that, after
the application of conidia, blocking of the ERK pathway by the use
of UO126 and PD98059 inhibited TNF-? synthesis in a dose-de-
pendent manner. In addition, treatment with PD98059 resulted in
a reduction of 30% in the killing of conidia as compared with
untreated control cells. This demonstrates an important role of the
ERK pathway in the defense against conidia.
We were further interested in the role of corticosteroids in the
inflammatory response against A. fumigatus. Corticosteroids are
potent anti-inflammatory drugs. They diffuse into target cells
where they bind to their cytoplasmic receptor, which, in turn,
translocates to the nucleus and blocks transcription of several cy-
tokine genes (30). Glucocorticoids inhibit members of the MAPK
family and p38 by sustaining the expression of MAPK phospha-
tase (31). When bronchoalveolar macrophages are treated with
dexamethasone and stimulated with living conidia of A. fumigatus,
the production of IL1-? and TNF-? is much less pronounced (32).
In agreement with these data, we found that immunosuppression in
vivo resulted in a significant decrease in the phosphorylation of
ERK and, as a consequence, a strong decrease in TNF-? and IL-6
In addition to the role of MAPKs in the production of cytokines,
the activation of NF-?B was studied because this transcription fac-
tor controls the expression of many genes involved in the inflam-
matory response (24). As discussed earlier, we have shown that
resting conidia do not induce the translocation of NF-?B at early
time points. However, we observed a translocation of NF-?B after
a 6-h incubation of conidia with macrophages. Our data are in
agreement with results published during the review of our manu-
script of showing that swollen conidia stimulated NF-?B to a
greater degree than resting conidia (26)
In summary, we have demonstrated that the MAPK ERK plays
an important role in the inflammatory host response against A.
fumigatus. Although our data strongly suggest that TLR2 and
TLR4 are not involved in this response, we cannot exclude the
possibility that other TLRs and/or MyD88 may be of importance
during different stages of infection. In the situation of neutrophil
depletion, the lack of TLR2 apparently renders macrophages less
responsive to A. fumigatus with a reduced level of TNF-? (14).
The residual level of TNF-? that was observed in our system may
be derived from an activation of these receptors and may be of
importance under these conditions. Nevertheless, in the immuno-
competent situation the contribution of TLRs appears to be less
relevant. Further studies will be required to elucidate the recep-
tor(s) that specifically senses A. fumigatus conidia. It is likely that
multiple receptors are involved that recognize different compo-
nents of the conidial cell wall to drive the innate immune response
against A. fumigatus. In this regard, the combined contributions of
TLRs as well as those of the scavenger family of receptors must be
We thank Dr. Matthias Brock for reading the manuscript and Dr. V.
Quesniaux for providing MyD88 knockout mice.
The authors have no financial conflict of interest.
1. Denning, D. W. 1991. Epidemiology and pathogenesis of systemic fungal infec-
tions in the immunocompromised host. J. Antimicrob. Chemother. 28(Suppl. B):
2. Latge, J. P. 1999. Aspergillus fumigatus and aspergillosis. Clin. Microbiol. Rev.
3. Philippe, B., O. Ibrahim-Granet, M. C. Prevost, M. A. Gougerot-Pocidalo,
M. Sanchez Perez, A. Van der Meeren, and J. P. Latge. 2003. Killing of As-
pergillus fumigatus by alveolar macrophages is mediated by reactive oxidant
intermediates. Infect. Immun. 71: 3034–3042.
4. Ibrahim-Granet, O., B. Philippe, H. Boleti, E. Boisvieux-Ulrich, D. Grenet,
M. Stern, and J. P. Latge. 2003. Phagocytosis and intracellular fate of Aspergillus
fumigatus conidia in alveolar macrophages. Infect. Immun. 71: 891–903.
Aspergillus fumigatus AND THE ALVEOLAR MACROPHAGE
5. Picard, C., A. Puel, M. Bonnet, C. L. Ku, J. Bustamante, K. Yang, C. Soudais, Download full-text
S. Dupuis, J. Feinberg, C. Fieschi, et al. 2003. Pyogenic bacterial infections in
humans with IRAK-4 deficiency. Science 299: 2076–2079.
6. Takeuchi, O., A. Kaufmann, K. Grote, T. Kawai, K. Hoshino, M. Morr,
P. F. Muhlradt, and S. Akira. 2000. Cutting edge: preferentially the R-stereoiso-
mer of the mycoplasmal lipopeptide macrophage-activating lipopeptide-2 acti-
vates immune cells through a Toll-like receptor 2- and MyD88-dependent sig-
naling pathway. J. Immunol. 164: 554–557.
7. Rao, K. M. 2001. MAP kinase activation in macrophages. J. Leukocyte Biol. 69:
8. Monick, M. M., A. B. Carter, G. Gudmundsson, R. Mallampalli, L. S. Powers,
and G. W. Hunninghake. 1999. A phosphatidylcholine-specific phospholipase C
regulates activation of p42/44 mitogen-activated protein kinases in lipopolysac-
charide-stimulated human alveolar macrophages. J. Immunol. 162: 3005–3012.
9. Carter, A. B., M. M. Monick, and G. W. Hunninghake. 1999. Both Erk and p38
kinases are necessary for cytokine gene transcription. Am. J. Respir. Cell Mol.
Biol. 20: 751–758.
10. Meier, A., C. J. Kirschning, T. Nikolaus, H. Wagner, J. Heesemann, and F. Ebel.
2003. Toll-like receptor (TLR) 2 and TLR4 are essential for Aspergillus-induced
activation of murine macrophages. Cell. Microbiol. 5: 561–570.
11. Mambula, S. S., K. Sau, P. Henneke, D. T. Golenbock, and S. M. Levitz. 2002.
Toll-like receptor (TLR) signaling in response to Aspergillus fumigatus. J. Biol.
Chem. 277: 39320–39326.
12. Netea, M. G., A. Warris, J. W. Van der Meer, M. J. Fenton, T. J. Verver-Janssen,
L. E. Jacobs, T. Andresen, P. E. Verweij, and B. J. Kullberg. 2003. Aspergillus
fumigatus evades immune recognition during germination through loss of Toll-
like receptor-4-mediated signal transduction. J. Infect. Dis. 188: 320–326.
13. Braedel, S., M. Radsak, H. Einsele, J. P. Latge, A. Michan, J. Loeffler, Z. Haddad,
U. Grigoleit, H. Schild, and H. Hebart. 2004. Aspergillus fumigatus antigens
activate innate immune cells via toll-like receptors 2 and 4. Br. J. Haematol. 125:
14. Balloy, V., M. Si-Tahar, O. Takeuchi, B. Philippe, M. A. Nahori, M. Tanguy,
M. Huerre, S. Akira, J. P. Latge, and M. Chignard. 2005. Involvement of Toll-like
receptor 2 in experimental invasive pulmonary aspergillosis. Infect. Immun. 73:
15. Marr, K. A., S. A. Balajee, T. R. Hawn, A. Ozinsky, U. Pham, S. Akira,
A. Aderem, and W. C. Liles. 2003. Differential role of MyD88 in macrophage-
mediated responses to opportunistic fungal pathogens. Infect. Immun. 71:
16. Sinha, B. K., D. P. Monga, and S. Prasad. 1988. A combination of Gomori-
Grocott methenamine silver nitrate and hematoxylene and eosin staining tech-
nique for the demonstration of Candida albicans in tissue. Quad. Sclavo Diagn.
17. Matsunaga, K., T. W. Klein, H. Friedman, and Y. Yamamoto. 2001. Alveolar
macrophage cell Line MH-S is valuable as an in vitro model for Legionella
pneumophila infection. Am. J. Respir. Cell. Mol. Biol. 24: 326–331.
18. Keyse, S. M. 2000. Protein phosphatases and the regulation of mitogen-activated
protein kinase signalling. Curr. Opin. Cell Biol. 12: 186–192.
19. Mehrad, B., R. M. Strieter, and T. J. Standiford. 1999. Role of TNF-? in pul-
monary host defense in murine invasive aspergillosis. J. Immunol. 162:
20. Levitz, S. M. 2004. Interactions of Toll-like receptors with fungi. Microbes In-
fect. 6: 1351–1355.
21. Wang, J. E., A. Warris, E. A. Ellingsen, P. F. Jorgensen, T. H. Flo, T. Espevik,
R. Solberg, P. E. Verweij, and A. O. Aasen. 2001. Involvement of CD14 and
Toll-like receptors in activation of human monocytes by Aspergillus fumigatus
hyphae. Infect. Immun. 69: 2402–2406.
22. Bellocchio, S., S. Moretti, K. Perruccio, F. Fallarino, S. Bozza, C. Montagnoli,
P. Mosci, G. B. Lipford, L. Pitzurra, and L. Romani. 2004. TLRs govern neu-
trophil activity in aspergillosis. J. Immunol. 173: 7406–7415.
23. Saklatvala, J. 2004. The p38 MAP kinase pathway as a therapeutic target in
inflammatory disease. Curr. Opin. Pharmacol. 4: 372–377.
24. Li, Y., R. F. Schwabe, T. DeVries-Seimon, P. M. Yao, M.-C. Gerbod-Giannone,
A. R. Tall, R. J. Davis, R. Flavel, D. A. Brenner, and I. Tabas. 2005. Free
cholesterol-loaded macrophages are an abundant source of tumour necrosis fac-
tor-? and interleukin-6. J. Biol. Chem. 280: 21763–21772.
25. Ibata-Ombetta, S., T. Jouault, P. A. Trinel, and D. Poulain. 2001. Role of extra-
cellular signal-regulated protein kinase cascade in macrophage killing of Candida
albicans. J. Leukocyte Biol. 70: 149–154.
26. Gersuk, G. M., D. M. Underhill, L. Zhu, and K. A. Marr. 2006. Dectin-1 and
TLRs permit macrophages to distinguish between different Aspergillus fumigatus
cellular states. J. Immunol. 176: 3717–3724.
27. Hohl, T. M., H. L. Van Epps, A. Rivera, L. A. Morgan, P. L. Chen,
M. Feldmesser, and E. G. Pamer. 2005. Aspergillus fumigatus triggers inflam-
matory responses by stage-specific ?-glucan display. PLoS Pathog. 1: e30.
28. Balloy, V., M. Huerre, J. P. Latge, and M. Chignard. 2005. Differences in patterns
of infection and inflammation for corticosteroid treatment and chemotherapy in
experimental invasive pulmonary aspergillosis. Infect. Immun. 73: 494–503.
29. Pylkkanen, L., H. Gullsten, M. L. Majuri, U. Andersson, E. Vanhala, J. Maatta,
T. Meklin, M. R. Hirvonen, H. Alenius, and K. Savolainen. 2004. Exposure to
Aspergillus fumigatus spores induces chemokine expression in mouse macro-
phages. Toxicology 200: 255–263.
30. Munck, A., D. B. Mendel, L. I. Smith, and E. Orti. 1990. Glucocorticoid receptors
and actions. Am. Rev. Respir. Dis. 141: S2–S10.
31. Lasa, M., S. M. Abraham, C. Boucheron, J. Saklatvala, and A. R. Clark. 2002.
Dexamethasone causes sustained expression of mitogen-activated protein kinase
(MAPK) phosphatase 1 and phosphatase-mediated inhibition of MAPK p38. Mol.
Cell. Biol. 22: 7802–7811.
32. Kamberi, M., E. Brummer, and D. A. Stevens. 2002. Regulation of bronchoal-
veolar macrophage proinflammatory cytokine production by dexamethasone and
granulocyte-macrophage colony-stimulating factor after stimulation by Aspergil-
lus conidia or lipopolysaccharide. Cytokine 19: 14–20.
4001 The Journal of Immunology