| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Pulmonary Center, School of Medicine, Boston University, Boston, MA 02118;
Department of Microbiology and Immunology, School of Medicine, University of Maryland, Baltimore, MD 21205; and
Department of Microbiology and Genetics, Darmstadt University of Technology, Darmstadt, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production and phosphorylation of
the transcription factor STAT1. TLR4, but not TLR2, agonists can induce
IFN- expression via a mechanism that requires the adapter protein
Toll-IL-1R domain-containing adapter protein (TIRAP)/myeloid
differentiation protein 88 (MyD88) adapter-like (Mal), but not
the adapter protein MyD88. Thus, the failure of TLR2 agonists to induce
STAT1-dependent genes results, in part, from their failure to induce
the expression of IFN-. In this study, we show that IL-6 expression
is also preferentially induced by activation of TLR4. TLR4-dependent
induction of IL-6 expression did require Toll-IL-1R domain-containing
adapter protein (TIRAP)/MyD88 adapter-like (Mal), but unlike iNOS and
IP-10, it did not require the expression of IFN-. Although exogenous
IFN- and IFN-
could synergize with TLR2 agonists to restore high
levels of iNOS expression and NO production, these IFNs could not
synergize with TLR2 agonists to induce high levels of IL-6. Similarly,
neutralizing anti-IFN Abs could block iNOS gene expression in
LPS-stimulated murine macrophages, whereas these Abs had little effect
on IL-6 gene expression in these cells. Together, these studies
demonstrate that IL-6, like iNOS and IP-10, is differentially expressed
in macrophages stimulated via TLR2 vs TLR4, although these differences
appear to arise from distinct signaling
mechanisms. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Activation of TLR2 or TLR4 results in recruitment of one or more
adapter proteins capable of interacting with the TIR domain of the
receptor. The myeloid differentiation protein 88 (MyD88), an adapter
protein previously known to be essential for IL-1 and IL-18 signaling
(7), was the first molecule identified as being involved
in TLR signal transduction and NF-
B activation (8). The
finding that a TLR4 agonist could induce NF-B activation in
MyD88-deficient mice (9), however, suggested that other
adapter molecules must exist that can mediate signaling independently
of MyD88. A novel adapter protein, TIR domain-containing adapter
protein (TIRAP) (10), also termed MyD88 adapter-like (Mal)
(11), was found to control activation of a
MyD88-independent signaling pathway activated by TLR4. Unlike MyD88,
TIRAP/Mal does not participate in IL-1 signaling. Furthermore, the
dsRNA-dependent protein kinase (PKR) was also reported to be a
downstream target of TIRAP/Mal (10).
Previous studies have shown that cellular responses induced by TLR2
agonists, such as Porphyromonas gingivalis LPS or the
synthetic lipopeptide Pam3Cys, differ from those
induced by TLR4 agonists, such as Escherichia coli LPS
(12, 13). TLR2-dependent expression of inflammatory genes
is much more restricted, which led to the suggestion that some
signaling pathways elicited by activation of TLR4 are absent in TLR2
signaling. Previously, our laboratories showed that IFN--induced
STAT1 phosphorylation was one of the signals involved in the activation
of a subset of TLR4-inducible proinflammatory genes (e.g., inducible NO
synthase (iNOS), inducible protein (IP)-10, monocyte chemoattractant
protein (MCP)-5) that are poorly or not at all induced via TLR2
(13). We found that TLR4 agonists induce IFN-
expression via a mechanism that requires TIRAP/Mal, but not MyD88.
Thus, the inability of TLR2 agonists to activate these genes appears to
result, in part, from the inability of these agonists to induce
TIRAP/Mal-dependent IFN- secretion and STAT1 phosphorylation.
In the current study, we show that expression of the proinflammatory
cytokine IL-6, like iNOS and IP-10, is also preferentially induced via
TLR4 signaling by a mechanism dependent on both MyD88 and TIRAP/Mal. In
contrast to iNOS and IP-10, TLR4-dependent IL-6 expression does not
depend on IFN- secretion and new protein synthesis. Together, these
findings support the existence of a novel TLR4- and TIRAP/Mal-dependent
signaling pathway that is independent of IFN-.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The RAW 264.7 murine macrophage cell line (TIB-71) was obtained from American Type Culture Collection (Manassas, VA), and cultured as we have previously described (14). C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME). RNA from PKR knockout mice (backcrossed >12 generations onto C57BL/6 background) were provided by Dr. T. Hamilton (Cleveland Clinic Foundation, Cleveland, OH), and were previously described (15). All of the animal use protocols were approved by the Institutional Animal Care and Use Committee. Thioglycollate-elicited, peritoneal exudate macrophages (PECs) were prepared and cultured as previously described (1). Bone marrow-derived macrophages (BMDM) were prepared and cultivated as described in Schilling et al. (16, 17) with slight modifications. Briefly, bone marrow stem cells were isolated from the femurs of 6- to 8-wk-old mice and washed twice in basic medium (RPMI 1640 medium; Life Technologies, Frederick, MD) supplemented with 10% (v/v) FBS (HyClone Laboratories, Logan, UT), 10 mM HEPES, 1 mM sodium pyruvate, 1% nonessential amino acids, 0.05 mM 2-ME, 100 U/ml penicillin/streptomycin). The stem cells were cultured at a density of 5 x 106 cells/30 ml in autoclaved 5 x 30-cm Teflon bags (bioFOLIE; Sartorius, Goettingen, Germany) in basic medium plus 7.5% FBS, 5% horse serum, 1.5 mM glucose, and 30% L929 cell-conditioned medium as a source of M-CSF (18). To avoid adherence of the cells, the bags were prepared with the hydrophobic surface facing inward (19). The bags were incubated at 37°C under a humidified atmosphere with 5% CO2 for 7 days. At the end of this incubation period, at least 98% of the cells were macrophages. These resulting BMDM were cultivated in basic medium and incubated 1 day in tissue culture plates before stimulation.
Reagents and Abs
LPS (purified from E. coli K-235) was purchased from
Sigma-Aldrich (St. Louis, MO), phenol/water was extracted by the method
of Hirschfeld et al. (20), and stored in small aliquots as
a 0.5 mg/ml stock solution in sterile water at -20°C. The synthetic
lipopeptide Pam3Cys
(S-[2,3-bis-(palmitoyloxy)-(2-RS)-propyl]-N-palmitoyl-(R)-Cys-(S)-Ser-Lys4-OH,
trihydrochloride) was obtained from EMC Microcollections (Tuebingen,
Germany) and was stored as a 1 mg/ml stock solution in sterile LPS-free
water at -20°C. A partially purified Mycobacterium
tuberculosis H37Ra-conditioned culture broth, termed soluble
tuberculosis factor (STF), was prepared under LPS-free conditions as
previously described (21). Recombinant IFN- and IFN-
were purchased from R&D Systems (Minneapolis, MN). Cycloheximide was
purchased from Sigma-Aldrich and stored as -80°C. Sheep polyclonal
anti-mouse IFN-
Abs were obtained from BioSource
International (Camarillo, CA), and sheep IgG was obtained from
Sigma-Aldrich. The TIRAP blocking peptide (TIRAP-BP;
NH2-RQIKIWFQNRRMKWKKLQLRDAAPGGAIVS-OH) consisted
of the Drosophila antennapedia protein leader
sequence positioned at the NH2-terminal end of a
mouse TIRAP peptide (10). The control peptide consisted of
the antennapedia sequence positioned at the
NH2-terminal of the reversed TIRAP peptide
sequence. Both peptides were synthesized by Biosynthesis (Lewisville,
TX), reconstituted to 10 mM in sterile DMSO and stored at -20°C.
Phosphorothioate CpG DNA immunostimulatory oligonucleotides (5'-TCC ATG
ACG TTC CTG ACG TT-3') were synthesized by Oligos Etc. (Wilsonville,
OR) and were stored as a 1 mg/ml solution in sterile LPS-free water at
-80°C.
Measurements of cytokine and NO levels
Murine macrophages were cultures in 24-well plates at a density
of 8 x 105 cells per well, incubated
overnight and stimulated for 24 h. After the stimulation period
the IL-6 and TNF- protein levels were determined in culture
supernatants by using a specific ELISA (OptEIA set; BD PharMingen San
Diego, CA), as recommended by the manufacturer. Levels of NO catabolite
nitrite in the culture supernatants were measured using Griess reagent
assay as previously described (22). All assays were
performed in triplicate, and data are expressed as mean values
± SD.
Measurement of IL-6, iNOS, IP-10, and -actin mRNA gene
expression by RT-PCR
Murine macrophages were cultured in 6-well plates at a density of 2 x 106 cells per well, incubated overnight, and then stimulated for different time periods (2–6 h). Total RNA was extracted from the cells using RNeasy (Qiagen, Valencia, CA) according to the manufacturer’s protocol. Reverse transcriptase reactions were performed using 2 µg of total RNA in a 20-µl mixture of 1 mM dNTPs (each), 0.5 ng/µl oligo (dT)15 primer, 62.5 U of rRNasin ribonuclease inhibitor and 32.5 U of AMV RT (Promega, Madison, WI) for 1 h at 42°C, followed by 5 min at 92°C. The cDNA obtained was stored at -20°C.
Semiquantitative PCR were performed using 2 µl of cDNA, 0.12 µM
gene specific oligonucleotide primer (each), 2 mM
MgCl2, 0.2 mM dNTPs, and 1 U of Taq
DNA polymerase (Promega), in a final reaction volume of 25 µl. PCR
was conducted for 29 cycles (except for detection of IP-10 gene
expression, where 25 cycles were used) with the following parameters:
denaturation at 94°C for 1 min, annealing at 55°C for 1 min,
extension at 72°C for 1 min, and a final extension at 72°C for 10
min. All PCR primers were purchased from Life Technologies. Primers for
the measurement of iNOS and -actin gene expression are described in
Jones et al. (21). PCR primers used for the detection of
IL-6 and IP-10 gene expression are: sense strand IL-6 primer, 5'-CAT
GTT CTC TGG GAA ATC GTG G-3'; antisense strand IL-6 primer, 5'-AAC GCA
CTA GGT TTG CCG AGT A-3'; sense strand IP-10 primer, 5'-GTG TTG AGA TCA
TTG CCA CGA-3'; and antisense strand IP-10 primer, 5'-GCT TAC AGT ACA
GAG CTA GG-3'. As a control for contaminating genomic DNA, parallel PCR
were performed in which the template nucleic acids were not
reverse-transcribed. After amplification, the PCR products were
electrophoresed on 2% agarose gels containing 0.4 mg/ml ethidium
bromide. All procedures for the detection of specific mRNA for PKR,
IL-6, and hypoxanthine phosphoribosyltransferase (HPRT) by
semiquantitative RT-PCR and Southern blot analysis were described in
detail elsewhere (13, 21, 23, 24).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We previously showed that a variety of proinflammatory genes
(e.g., iNOS, IP-10, MCP-5) were preferentially induced in macrophages
by the TLR4 agonist E. coli LPS, whereas all TLR2 agonists
tested failed to induce these genes (Ref. 13 and
references therein). In contrast, both TLR2 and TLR4 agonists strongly
elicited expression of the proinflammatory cytokines IL-1 and
TNF-. We subsequently sought to determine whether IL-6 expression
was differentially induced by TLR2 vs TLR4. IL-6 production was
measured in the supernatants of RAW 264.7 murine macrophages stimulated
for 24 h with either E. coli LPS or different
concentrations of the synthetic lipopeptide
Pam3Cys or STF. As shown in Fig. 1
A, the TLR4 agonist E.
coli LPS induced a high level of IL-6 secretion. In contrast, both
TLR2 agonists (Pam3Cys and STF) induced little
IL-6 secretion, even at concentrations that were comparable to E.
coli LPS in their capacities to induce TNF- secretion (Fig. 1B). Together, these studies show that IL-6 production is
selectively and maximally activated in macrophages stimulated by TLR4
agonists.
|
and IFN- do not strongly synergize with TLR2
agonists to induce IL-6 production
Our previous data demonstrated that TLR2 agonists fail to induce
macrophage expression of STAT-dependent genes such as iNOS or IP-10 as
a consequence of their failure to induce IFN-, whereas
TLR4-dependent induction of IFN- promotes their expression via STAT1
activation (13). We also demonstrated that exogenous
IFN- could restore iNOS mRNA gene expression in macrophages
stimulated via TLR2. In this study, we found that the addition of
exogenous IFN- to TLR2 agonist-stimulated RAW 264.7 macrophages also
restored NO secretion. As shown in Fig. 2A, the TLR2 agonists
Pam3Cys and STF alone poorly induced NO secretion
by murine macrophages, whereas addition of these TLR2 agonists plus
exogenous IFN- or IFN- led to the same levels of NO production
that were induced by the TLR4 agonist E. coli LPS (100
ng/ml) alone. Neither IFN- nor IFN- alone could activate the
macrophages to secrete NO. IL-6 levels were subsequently measured in
these various culture supernatants (Fig. 2B). In contrast to
NO production, IL-6 secretion was not restored to the levels induced by
E. coli LPS alone by stimulating the cells with exogenous
IFNs plus the TLR2 agonists. Thus, exogenous IFNs can synergize with
TLR2 agonists to induce maximal levels of NO secretion, but not maximal
levels of IL-6 secretion.
|
, C and D,
there was little detectable NO or IL-6 in culture supernatants of the
cells stimulated with Pam3Cys alone (10 or 100
ng/ml), or in combination with 0.1 or 1 pg/ml E. coli LPS.
But at a concentration of 100 pg/ml LPS, there was a significant
increase in TLR2-induced IL-6 secretion (Fig. 2D), and an
even stronger increase in NO secretion (Fig. 2C). The
combination of 1 ng/ml E. coli LPS and
Pam3Cys induced macrophages to secrete levels of
IL-6 and NO that were similar to levels induced by 100 ng/ml E.
coli LPS alone. These data demonstrate that as little as 100 pg/ml
LPS can strongly synergize with TLR2 agonists to augment the secretion
of NO and IL-6 compared with cells stimulated with each agonist
alone.
iNOS, but not IL-6, gene expression is synergistically induced by
TLR2 agonists plus exogenous IFN- and IFN-
We also investigated the effect of exogenous IFNs on TLR2- and
TLR4-induced iNOS and IL-6 gene expression in the murine RAW 264.7
macrophages. As shown in Fig. 3, the TLR4
agonist E. coli LPS activated iNOS and IL-6 mRNA expression
in macrophages, whereas the TLR2 agonists Pam3Cys
and STF alone did not induce detectable amounts of iNOS mRNA or IL-6
mRNA. Stimulation of cells with the TLR2 agonists in the presence of
IFN- or IFN- significantly increased iNOS gene expression.
Conversely, there was only a poor increase in IL-6 gene expression
detected in response to TLR2 agonists plus IFN- or IFN-. These
results were consistent with results obtained for NO and IL-6 secretion
by macrophages stimulated with TLR2 agonists in the presence and
absence of IFNs (Fig. 2, A and B).
|
and IFN- also
synergize poorly with TLR2 agonists to induce IL-6
We subsequently sought to confirm these findings using primary
peritoneal and BMDM. BMDM from C57BL/6 mice were stimulated with LPS,
Pam3Cys, or STF in the absence or presence of
IFN- or IFN-, and NO and IL-6 levels in the cell culture
supernatants were measured 24 h later. As shown in Fig. 4A, NO secretion was strongly
induced in cells incubated with a TLR2 agonist in the presence of
exogenous IFN- or IFN-. In contrast, the TLR2 agonists alone, and
the IFNs alone, induced little or no NO secretion compared with TLR4
agonist E. coli LPS (Fig. 4A). Moreover, IFN-
and IFN- did not strongly synergize with the TLR2 agonists to induce
IL-6 secretion, and maximal IL-6 production was only observed in BMDM
stimulated with E. coli LPS (Fig. 4B). Similar
results were obtained using PECs (data not shown).
|
secretion
We previously demonstrated that TLR2 agonists failed to induce
iNOS expression and STAT1 phosphorylation (13).
Furthermore, TLR4-dependent activation of STAT1 phosphorylation was
largely attributable to the action of LPS-induced IFN-. As shown in
Fig. 5, E. coli LPS-induced
iNOS gene expression was significantly decreased in macrophages
incubated with neutralizing anti-IFN- Abs. In contrast, no
inhibitory effect was observed for E. coli LPS-induced IL-6
gene expression when using the same Abs. PCR performed using fewer
cycles to minimize the possibility of nonlinear amplification also
revealed that the neutralizing anti-IFN- Abs had no effect on
LPS-induced IL-6 gene expression (data not shown). These data
demonstrate that, unlike iNOS gene expression, IFN- secretion does
not mediate the induction of IL-6 gene expression via TLR4.
|
The failure of neutralizing anti-IFN- Abs to block IL-6
mRNA induced by LPS excluded IFN- as a mediator of maximal
TLR4-induced IL-6 production. To determine whether there was any
requirement for de novo protein synthesis, murine macrophages were
stimulated with E. coli LPS for 2 h in the absence or
presence of the protein synthesis inhibitor cycloheximide. As shown in
Fig. 6, iNOS gene expression was
significantly decreased with increasing concentrations of
cycloheximide. In contrast, the protein synthesis inhibitor had no
effect on LPS-induced IL-6 gene expression. Thus, de novo protein
synthesis does not appear to be required for TLR4-induced IL-6 gene
expression, and activation of the IL-6 promoter is a direct consequence
of TLR4 signaling.
|
As we reported previously, TLR4-induced IFN- gene expression
was dependent on TIRAP/Mal, but not MyD88 (13).
Nevertheless, TIRAP/Mal is not sufficient for most TLR4-dependent
cellular responses, as shown by the inability of MyD88-deficient
macrophages to secrete NO and IL-6 following stimulation with E.
coli LPS (Ref. 9 and unpublished observations). These
findings led us to investigate whether TLR4-induced IL-6 gene
expression in macrophages was also dependent on TIRAP/Mal. The data
shown in Fig. 7 demonstrate that in the
presence of a TIRAP-BP, the induction of endogenous IL-6 gene
expression by E. coli LPS was inhibited. As a control for
the efficacy of the TIRAP-BP, the peptide was shown to completely
inhibit LPS-induced IP-10 gene expression in the same experiment.
Specificity of the blocking peptide was demonstrated by the finding
that the peptide did not inhibit IP-10 gene expression induced by CpG
DNA. Activation of TLR9 signaling by CpG DNA was previously shown to be
TIRAP-independent (10). Taken together with published data
(9), these findings demonstrate that induction of IL-6
gene expression via TLR4 is both MyD88- and TIRAP/Mal-dependent.
|
It has been reported that PKR is a downstream mediator of TIRAP
signaling (10), although molecular targets for this kinase
in TIRAP-dependent signaling have not yet been identified. We
previously showed that E. coli LPS-induced IFN- and iNOS
gene expression are TIRAP-dependent, but are PKR-independent
(13). We sought to determine whether LPS-induced IL-6
expression was dependent on PKR in macrophages. Fig. 8 shows that, like IFN- and iNOS gene
expression, IL-6 mRNA was also comparably inducible by E.
coli LPS in PKR+/+ and
PKR-/- macrophages. Thus, TLR4-induced IL-6
gene expression is both MyD88- and TIRAP/Mal-dependent, but is
PKR-independent.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production and
phosphorylation of the transcription factor STAT1. TLR2 agonists failed
to induce macrophage expression of STAT1-dependent genes as a
consequence of their failure to activate IFN- gene expression
(13). In this study, we investigated TLR-dependent
production of the proinflammatory cytokine IL-6. The TLR4 agonist,
E. coli LPS, stimulated macrophages to secrete high levels
of the cytokine IL-6, whereas IL-6 production was only poorly activated
in RAW264.7 macrophages by all TLR2 agonists tested (Figs. 1 and 3).
These data confirm results of a previous study using the TLR2 agonist
P. gingivalis LPS in which induction of IL-6 mRNA was
substantially lower than message levels induced via TLR4
(12). In this study, we have also shown that differential
IL-6 production induced by TLR2 and TLR4 is a characteristic of
different macrophage populations. Similar results were obtained using
RAW264.7 cells, BMDM, and PECs. All three macrophage populations showed
diminished production of IL-6 in response to the TLR2 agonists
Pam3Cys and STF compared with the TLR4 agonist
E. coli LPS.
We report in this study that the TLR2 agonists
Pam3Cys and STF alone did not induce substantial
IL-6 secretion in the RAW264.7 macrophage cell line. Using primary
murine bone marrow macrophages (Fig. 4) and peritoneal macrophages
(data not shown), Pam3Cys stimulation was
sufficient to induce IL-6 secretion, but always to levels substantially
lower than those induced by E. coli LPS. Moreover, several
investigators have previously shown that TLR2 agonists can induce IL-6
secretion by human and murine macrophages (4, 25, 26).
Most recently, DNA microarray analysis was used to demonstrate that LPS
induced higher levels of steady-state IL-6 mRNA in elutriated human
monocytes, compared with monocytes stimulated by the TLR2 agonists
lipoteichoic acid and muramyl dipeptide (27). Our findings
extend these studies by directly comparing TLR2 and TLR4 agonists, and
demonstrate that TLR4-dependent signaling via TIRAP/Mal is necessary
for maximal IL-6 secretion.
Our results appear to contradict an earlier study in which the TLR2
agonist Staphylococcus aureus peptidoglycan (PGN) was shown
to induce substantial secretion of IL-6 and NO by peritoneal
macrophages (3). Takeuchi et al. (3)
clearly showed that TLR2-deficient macrophages could not be activated
by PGN, whereas wild-type and TLR4-deficient macrophages responded
similarly. Although the reason for these differences remains unclear,
one possible explanation could be that PGN activates a distinct
TLR2-containing receptor complex, compared with
Pam3Cys and STF. As discussed below, TLR2 has
been reported to function in association with other TLR proteins, such
as TLR1 and TLR6 (28). Thus, it is possible that the
TLR2-containing receptor complexes engaged by
Pam3Cys and STF do not activate signaling
molecules necessary for IL-6 and NO secretion, whereas the complex
engaged by PGN recruits additional signaling molecules that can mediate
IL-6 and NO secretion. It also important to note that there is a
strong synergistic response when cells were stimulated via TLR2 in the
presence of picogram quantities of E. coli LPS (Figs. 2, C and D). It is possible that many purified TLR2
agonists contain small quantities of LPS, and that these low levels of
contaminating LPS would fail to induce IL-6 and NO secretion in the
absence of the TLR2 agonist. When present together, these TLR agonists
can induce cellular responses of a magnitude similar to those induced
by high concentrations of LPS. Our studies imply that care must be
taken to exclude LPS contamination when using commercially available
and purified natural TLR2 agonists.
By using dominant-negative forms of either TLR1, 2, and 6, Ozinsky et al. (28) showed that TLR2 is capable of forming a functional signaling complex with either TLR1 or TLR6. The capacity of TLR2 to discriminate between pathogen-associated molecular patterns may depend on the heterodimerizing receptor partners. The fact that TLR2 does not appear to function as a homooligomeric complex (e.g., a homodimer), but can confer responsiveness to bacterial lipopeptides in the presence of TLR1 and TLR6, led these investigators to suggest that there may be additional partners for TLR2. It remains to be determined whether receptor complexes containing TLR2 plus TLR1 and TLR2 plus TLR6 activate distinct signal transduction pathways. In vitro studies showed that dominant-negative mutants of TLR2 or TLR6 inhibit the cellular activation by STF in TLR2-transfected human dermal endothelial cells (29), implying that STF signaling occurs via receptor complexes containing both TLR2 and TLR6. Similarly, Pam3Cys has recently been shown to activate receptor complexes containing both TLR2 and TLR1 (30). Our findings suggest that IL-6 production is only poorly activated via both the TLR2/1 and TLR2/6 signaling pathways. Lastly, we cannot exclude the possibility that TLR2 could form complexes with TLR proteins other than TLR1 and 6, or with as yet unidentified receptors, that can induce IL-6 production.
As reported by Toshchakov et al. (13), the inability of
TLR2 agonists to induce iNOS gene expression arose from their inability
to activate IFN- production. In contrast, IFN- did not appear to
play an intermediary role in IL-6 gene expression. Furthermore,
exogenous IFN- failed to synergize with TLR2 agonists to induce
maximal levels of IL-6 (Figs. 2 and 4). In addition, studies using
neutralizing Abs revealed that LPS-induced IL-6 gene expression was not
dependent on TLR4-mediated IFN- induction (Fig. 5). In contrast to
iNOS gene expression, de novo protein synthesis was not required for
TLR4-dependent induction of IL-6 mRNA expression (Fig. 6). Taken
collectively, these data suggest that IL-6 production is a direct
consequence of TLR signaling.
All TLR proteins appear to use the adapter protein MyD88 for signaling
(7, 8). However, TLR4 can also signal via a
MyD88-independent pathway. Kawai et al. (9) previously
published that IL-6 gene expression and secretion occurs via a
TLR4-signaling pathway that is MyD88-dependent, as evidenced by
impaired IL-6 production in LPS-stimulated
MyD88-/- macrophages. By using a cell-permeable
blocking peptide, we showed in this study that IL-6 production in
LPS-stimulated macrophages is also dependent on the adapter molecule
TIRAP/Mal. This is consistent with a previous study that reported
TIRAP/Mal mediated IL-6 production in LPS-stimulated dendritic cells
(10). Because TIRAP/Mal specifically mediates activation
of cells via TLR4, and not TLR2 (L. O’Neill, unpublished
observation), it is likely that TIRAP/Mal signaling leads to the
activation of the putative pre-existing protein described above. Taken
together, our findings imply that both MyD88 and TIRAP/Mal mediate
TLR4-specific signaling, which, in combination with NF-B, mediate
LPS-induced IL-6 production. A proposed model for this is presented in
Fig. 9.
|
B, AP-1, CREB, and C/EBP (33). We have
previously shown that TLR2 and TLR4 agonists are equally capable of
inducing NF-B and AP-1 activation in RAW264.7 macrophages
(21). Thus, it is likely that activation of NF-B and
AP-1 by TLR2 agonists is sufficient to induce a modest level of IL-6
gene expression. In contrast, TLR4 engagement by E. coli
LPS, which can activate all four of these transcription factors,
consequently leads to maximal IL-6 transcription. Unlike the iNOS
promoter, the IL-6 promoter does not appear to contain regulatory
elements that are activated by IFN signaling (e.g., IFN-stimulated
response and IFN- activation site elements). This is
consistent with our finding that exogenous IFN- and IFN- failed
to induce any significant production of IL-6 by RAW 264.7 cells and
primary murine macrophages (Figs. 2 and 4). This lack of IFN-responsive
elements within the IL-6 promoter is a likely explanation for why
exogenous IFNs failed to strongly synergize with TLR2 agonists to
induce IL-6 secretion, but could strongly synergize to induce NO
production (Fig. 2).
Lastly, a proposed downstream target of TIRAP/Mal is PKR
(10). Although both LPS-inducible IFN- and iNOS gene
expression are TIRAP-dependent, we could not identify a role for PKR in
these responses (13). In this study, we have shown that
LPS-activated macrophages from wild-type and PKR-deficient mice
produced equivalent levels of IL-6 mRNA (Fig. 8). Thus, like IFN-
gene expression, induction of IL-6 gene expression is also
TIRAP/Mal-dependent and PKR-independent. As we proposed previously for
IFN- gene expression, the PKR independence of IL-6 expression could
arise from bifurcation of the signaling pathway distal to TIRAP/Mal,
which may separate the activation of PKR from both IFN- and IL-6
gene expression (Fig. 9). Overall, several questions remain unanswered
by our studies. It is unclear whether induction of IFN- and IL-6
gene expression relies on the same TLR4- and TIRAP/Mal-dependent
mechanisms. Also, the biological rationale for why IL-6 production is
preferentially induced by TLR4 engagement, in contrast to the other
proinflammatory cytokines TNF- and IL-1, remains to be
determined. These questions will be the focus of future studies.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Matthew J. Fenton, Pulmonary Center, R-220, School of Medicine, Boston University, Boston, MA 02118-2394. E-mail address: mfenton{at}bu.edu
3 Abbreviations used in this paper: TLR, Toll-like receptor; TIR, Toll-IL-1R; MyD88, myeloid differentiation protein 88; TIRAP, TIR domain-containing adapter protein; Mal; MyD88 adapter-like; iNOS, inducible NO synthase; IP, inducible protein; MCP, monocyte chemoattractant protein; PEC, peritoneal exudate macrophage; BMDM, bone marrow-derived macrophage; TIRAP-BP, TIRAP blocking peptide; STF, soluble tuberculosis factor; PKR, dsRNA-dependent protein kinase; PGN, peptidoglycan.
Received for publication July 11, 2002. Accepted for publication September 16, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced STAT1/-dependent gene expression in macrophages. Nat. Immunol. 3:392.[Medline]
production in macrophages stimulated with lipopolysaccharide by the protein kinase inhibitor staurosporine. J. Endotoxin Res. 4:251. production at the level of transcription in macrophages stimulated with LPS. Cytokine 16:51.[Medline]
and interleukin-6 by human monocytes. Infect. Immun. 62:2715.This article has been cited by other articles:
|
|
 |
|
  S. M. Dann, M. E. Spehlmann, D. C. Hammond, M. Iimura, K. Hase, L. J. Choi, E. Hanson, and L. Eckmann IL-6-Dependent Mucosal Protection Prevents Establishment of a Microbial Niche for Attaching/Effacing Lesion-Forming Enteric Bacterial Pathogens J. Immunol., May 15, 2008; 180(10): 6816 - 6826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Meier, G. Alter, N. Frahm, H. Sidhu, B. Li, A. Bagchi, N. Teigen, H. Streeck, H.-J. Stellbrink, J. Hellman, et al. MyD88-Dependent Immune Activation Mediated by Human Immunodeficiency Virus Type 1-Encoded Toll-Like Receptor Ligands J. Virol., August 1, 2007; 81(15): 8180 - 8191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Liang, M. Wang, R. I. Tapping, V. Stepensky, H. F. Nawar, M. Triantafilou, K. Triantafilou, T. D. Connell, and G. Hajishengallis Ganglioside GD1a Is an Essential Coreceptor for Toll-like Receptor 2 Signaling in Response to the B subunit of Type IIb Enterotoxin J. Biol. Chem., March 9, 2007; 282(10): 7532 - 7542. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Winkler, D. Ghadimi, J. Schrezenmeir, and J.-P. Kraehenbuhl Molecular and Cellular Basis of Microflora-Host Interactions J. Nutr., March 1, 2007; 137(3): 756S - 772S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pevsner-Fischer, V. Morad, M. Cohen-Sfady, L. Rousso-Noori, A. Zanin-Zhorov, S. Cohen, I. R. Cohen, and D. Zipori Toll-like receptors and their ligands control mesenchymal stem cell functions Blood, February 15, 2007; 109(4): 1422 - 1432. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Hise, K. Daehnel, I. Gillette-Ferguson, E. Cho, H. F. McGarry, M. J. Taylor, D. T. Golenbock, K. A. Fitzgerald, J. W. Kazura, and E. Pearlman Innate Immune Responses to Endosymbiotic Wolbachia Bacteria in Brugia malayi and Onchocerca volvulus Are Dependent on TLR2, TLR6, MyD88, and Mal, but Not TLR4, TRIF, or TRAM J. Immunol., January 15, 2007; 178(2): 1068 - 1076. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bagchi, E. A. Herrup, H. S. Warren, J. Trigilio, H.-S. Shin, C. Valentine, and J. Hellman MyD88-Dependent and MyD88-Independent Pathways in Synergy, Priming, and Tolerance between TLR Agonists J. Immunol., January 15, 2007; 178(2): 1164 - 1171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Morris, L. C. Parker, J. R. Ward, E. C. Jones, M. K. B. Whyte, C. E. Brightling, P. Bradding, S. K. Dower, and I. Sabroe Cooperative molecular and cellular networks regulate Toll-like receptor-dependent inflammatory responses FASEB J, October 1, 2006; 20(12): 2153 - 2155. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Shoenfelt and M. J. Fenton TLR2- and TLR4-dependent activation of STAT1 serine phosphorylation in murine macrophages is protein kinase C-{delta}-independent Innate Immunity, August 1, 2006; 12(4): 231 - 240. [Abstract] [PDF]
|
|
|
|
 |
|
 R. A. Frost, G. J. Nystrom, and C. H. Lang Multiple Toll-like receptor ligands induce an IL-6 transcriptional response in skeletal myocytes Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2006; 290(3): R773 - R784. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Machida, K. T. H. Cheng, V. M.-H. Sung, A. M. Levine, S. Foung, and M. M. C. Lai Hepatitis C Virus Induces Toll-Like Receptor 4 Expression, Leading to Enhanced Production of Beta Interferon and Interleukin-6 J. Virol., January 15, 2006; 80(2): 866 - 874. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Guo, S. Garg, K. M. Hill, L. Jayashankar, M. R. Mooney, M. Hoelscher, J. M. Katz, J. M. Boss, and S. Sambhara A Distal Regulatory Region Is Required for Constitutive and IFN-{beta}-Induced Expression of Murine TLR9 Gene J. Immunol., December 1, 2005; 175(11): 7407 - 7418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Jenkins, J. P. Hewitson, S. Ferret-Bernard, and A. P. Mountford Schistosome larvae stimulate macrophage cytokine production through TLR4-dependent and -independent pathways Int. Immunol., November 1, 2005; 17(11): 1409 - 1418. [Abstract] [Full Text] [PDF]
|
|
|
