ERK Activation Following Macrophage FcR Ligation Leads to Chromatin Modifications at the IL-10 Locus1
http://www.100md.com
免疫学杂志 2005年第13期
Abstract
We have previously demonstrated that macrophages stimulated in the presence of immune complexes produce high levels of IL-10. We now examine the mechanism of IL-10 superinduction. We report that the enhanced production of IL-10 correlates with a rapid and enhanced activation of two MAPKs, ERK and p38. The inhibition of either ERK or p38 prevented IL-10 induction, indicating that both MAPKs were required for IL-10 synthesis. By chromatin immunoprecipitation assay, we demonstrate that activation of ERK leads to the phosphorylation of serine 10 on histone H3 at the il-10 gene, making the promoter more accessible to transcription factors generated in response to p38 activation. Inhibition of ERK activation prevented histone modifications, and decreased the binding of Sp1 and STAT3 to the IL-10 promoter. We conclude that the activation of ERK following FcR ligation leads to a remodeling of the chromatin at the il-10 locus, making it more accessible to transcription factors. The rapid and transient regulation of transcription factor accessibility to the IL-10 promoter by MAPK activation represents a novel way that the production of this cytokine is regulated.
Introduction
Activated macrophages are major mediators of inflammation. These cells secrete a variety of inflammatory cytokines, lipid mediators, and potentially toxic radicals of oxygen and nitrogen (reviewed in Ref. 1). Consequently they can contribute to the pathology associated with a number of autoimmune diseases, such as rheumatoid arthritis (2), multiple sclerosis (3), and inflammatory bowel disease (4) to mention a few. We have recently shown that macrophages activated in the presence of immune complexes secreted high levels of IL-10. The IL-10 produced by these cells could reverse lethal endotoxemia (5) and prevent Th1-type adaptive immune responses (6). Thus, immune complexes give rise to a potent anti-inflammatory population of macrophages that produce large amounts of IL-10. In the present work, we examine the mechanism of IL-10 induction in these macrophages.
Several transcription factors have been implicated in IL-10 transcription. Sp1 has been shown to play an important role in IL-10 transcription, and an Sp1 responsive element in the IL-10 promoter was localized at –89 to –78 (7). The STAT3 transcription factor has also been shown to bind to an element in the human IL-10 promoter, and a dominant negative form of STAT3 has been shown to diminish IL-10 transcription (8). A role for C/EBP has been suggested (9), although it may not be required for IL-10 transcription (7). Despite the identification of these transcription factors, the regulation of IL-10 transcription in macrophages remains somewhat elusive. One of the reasons for ambiguity is that the common stimulus used to induce IL-10 in macrophages is LPS. This compound is not a particularly potent stimulator of IL-10 production, typically inducing only in the hundred-picogram range of IL-10, which correlates with modest increases in IL-10 transcription. Thus, assays to identify transcriptional elements within the IL-10 promoter are not particularly robust. In the present study, we take advantage of the fact that stimulating macrophages in the presence of immune complexes induces the production of high levels of IL-10. This creates a robust model in which to study IL-10 transcription. Importantly, immune complexes alone do not induce cytokine production from macrophages (5). However, when these complexes are combined with an inflammatory stimulus such as LPS, the two stimuli synergize to induce high levels of IL-10. The mechanism of this superinduction is the topic of the present study.
We have previously reported that activation of macrophages in the presence of immune complexes also results in an abrogation of IL-12 production (10). The synthesis of IL-12 and its regulation are more well understood than IL-10. There have been several reports describing the down-regulation of IL-12 subsequent to macrophage stimulation. The ligation of any number of macrophage receptors, including the FcRs (11), complement receptors (12), scavenger receptors (11), as well as G protein-coupled receptors (13) have all been linked to suppressed IL-12 production, regardless of the stimuli. Similarly, ligation of 2 adrenergic receptors (14) and exposure of macrophages to vitamin D3 (15), or to macrophage stimulatory protein (16), can also inhibit IL-12 production. Finally, infection of macrophages or dendritic cells with a variety of microorganisms, including measles virus (17), herpes virus (18), Leishmania (19), or mycobacteria (20) can all result in a reduction of IL-12 secretion. The molecular mechanism for this inhibition may be quite complex. Regardless of the mechanism(s), virtually all of the inhibition appears to occur at the level of IL-12 transcription, resulting in decreased mRNA production and decreased or absent protein secretion. In the present studies, the down-regulation of IL-12 in response to immune complexes is used as a control, to contrast what occurs at the il-10 gene.
Having identified the reciprocal alteration in the production of IL-10 (increase) and IL-12 (decrease) by immune complexes (6), we began to study the molecular aspects of cytokine gene expression. We reasoned that the identification of the mechanism whereby these two cytokines are regulated could lead to the development of novel therapeutic strategies that would predictably alter macrophage phenotypes, thereby influencing innate and adaptive immunity. In the present work, we show that FcR ligation of macrophages leads to ERK activation. Activation of ERK leads to transient modifications in the chromatin at the IL-10 locus. These alterations to the chromatin are necessary to allow the efficient transcription of IL-10, which occurs in macrophages that are stimulated in the presence of immune complexes.
Materials and Methods
Mice
Six- to 8-wk-old BALB/c mice were purchased from Taconic Farms. All mice were maintained in HEPA-filtered Thoren units (Thoren Caging Systems) at the University of Maryland (College Park, MD). Mice were used at 6–10 wk of age as a source of bone marrow-derived macrophages (BMM).3 All procedures were reviewed and approved by the University of Maryland Institutional Animal Care and Use Committee.
Reagents
The p38 MAPK inhibitor, SB203580, and the MEK/ERK inhibitor, PD98059, were purchased from Calbiochem. The JNK inhibitor peptide I was purchased from Alexis, USA (San Diego, CA). Washed SRBC were purchased from Lampire. Rabbit IgG Ab to SRBC (anti-SRBC IgG) was purchased from Cappel. Ultra pure LPS from Escherichia coli K12 strain was obtained from InvivoGen. Anti-phospho-H3 and acetyl-H3 Abs and chromatin immunoprecipitation (ChIP) assay kits were purchased from Upstate Biotechnology. Anti-p38 (phospho-T180/Y182), anti-STAT3, and anti-Sp1 Abs were purchased from Abcam. Anti-p38 (total), anti-ERK1/2 (total and phospho-T202/Y204), and anti-JNK (phospho-T183/Y185) were purchased from Cell Signaling Technology.
Cells
BMM were prepared as previously described (6). Briefly, bone marrow was flushed from the femurs and tibias of mice and cells were plated in petri dishes in DMEM/F12 supplemented with 10% FBS, penicillin/streptomycin, glutamine, and 10% conditioned medium from the supernatant of M-CSF secreting L929 (LCM) fibroblasts. Cells were fed on day 2, and complete medium was replaced on day 6. Cells were used at 7–10 days for experiments. The RAW264.7 macrophage cell line (American Type Culture Collection) was maintained in RPMI 1640 supplemented with 10% FBS, penicillin/streptomycin and glutamine (Invitrogen Life Technologies).
Immune complexes
IgG-opsonized erythrocytes (E-IgG) were generated by incubating SRBC with anti-SRBC IgG at nonagglutinating titers for 30 min at room temperature while rotating. Opsonized cells were washed once in HBSS (Invitrogen Life Technologies) and resuspended in complete medium. E-IgG were added to macrophages at a ratio of 10:1 E-IgG to macrophages. For some experiments IgG-OVA was used as the immune complex. IgG-OVA was prepared as previously described (6).
Cell stimulation assays
For cytokine analysis, 3 x 105 macrophages per well were plated overnight in a 48-well plate in DMEM/F12. Cells were then washed and activated with either 10 ng/ml LPS alone or in combination with a 10:1 ratio of E-IgG to macrophages. Supernatants were harvested 20 h later. Cytokines were measured by ELISA using the following Ab pairs from BD Pharmingen: IL-12p40, C15.6 and C17.8; IL-10, JES5-2A5 and JES5-16E3; TNF-, G281-2626 and MP6-XT22.
Generation of small interfering RNA (siRNA) and cell transfections
Nuclear extracts were prepared from 2 x 107 macrophages using the Nuclear Extraction kit (Panomics) following the manufacturer’s protocol. EMSAs were conducted using the EMSA "Gel-Shift" kit (Panomics). Briefly, 5 μg of nuclear extract were incubated with biotin-Sp1 or STAT3 probe with or without unlabeled probe for 30 min at 20°C, and then run on a 6% polyacrylamide gel. Oligonucleotide-protein complexes were transferred to Biodyne B membrane (Pall). Following transfer, membranes were incubated with streptavidin-HRP and protein visualized by chemiluminescence.
Luciferase assay
RAW264.7 macrophages (4 x 106) were transfected with 5 μg of pGL-IL10-luciferase reporter plasmid (7), which is a generous gift of Dr. S. Smale (Howard Hughes Medical Institute, University of California, Los Angeles, CA) with the Amaxa Nucleofector system. After transfection, 3 x 105 cells were plated per well in 48-well culture plates. After 24 h, cells were washed and stimulated, then lysed using Glo-Lysis buffer, and luciferase activity was measured using the Bright-Glo Luciferase system (Promega).
ChIP assay
ChIP assays were conducted using the ChIP Assay kit following the manufacturer’s protocol (Upstate Biotechnology). Briefly, 1 x 106 BMM were plated overnight in six-well plates. Cells were stimulated as described in figures, then fixed for 10 min at 37°C in 1% paraformaldehyde. Cells were washed on ice with ice-cold HBSS containing 1 mM PMSF, harvested and then lysed in SDS lysis buffer. DNA was sheared by ultrasonication using a High Intensity Ultrasonic Processor (Cole-Parmer) for 3 x 10 s pulses at 20% amplitude. Lysates were cleared by centrifugation and diluted in ChIP dilution buffer. Lysates were precleared using salmon sperm DNA/protein A-agarose and a sample of "input DNA" was collected at this point. Protein-DNA complexes were immunoprecipitated with 5 μg of Ab overnight at 4°C. Ab-protein-DNA complexes were then captured using salmon sperm DNA/protein A-agarose for 1 h at 4°C. After washing beads with low and high salt, LiCl, and TE buffers, the protein/DNA complexes were eluted using 1% SDS, 0.1 M NaHCO3 buffer and disrupted by heating at 65°C for 4 h. DNA was then extracted using phenol/chloroform extraction and ethanol precipitation. PCR was conducted using promoter specific primers: IL-10 promoter (Sp1 binding region, –294 to –73) sense 5'-CAGCTGTCTGCCTCAGGAAATACAA-3', antisense 5'-TATTCAGGCTCCTCCTCCCTCTTCT-3' (94°C, 15 s; 60°C, 30 s; 72°C, 1 min, 35 cycles); IL-10 promoter (STAT3 binding region, –649 to –448) sense 5'-TCATGCTGGGATCTGAGCTTCT-3', antisense 5'-CGGAAGTCACCTTAGCACTCAGT-3' (94°C, 15 s; 56°C, 30 s; 72°C, 1 min, 35 cycles); IL-12p40 promoter sense 5'-CAAATCTGGGAGGCAGGAAAC-3', antisense 5'-CAAAGCAAACCTTTCTATCAAATACACA-3' (94°C, 15 s; 56°C, 30 s; 72°C, 1 min, 35 cycles). Numerical designations are according to GenBank accession number M84340. PCR products were separated on 2% agarose gels. For relative quantitation of promoter levels, real-time PCR was performed.
Real-time PCR
Real-time PCR was conducted with the ABI Prism 7700 Sequence Detection System using SYBR Green PCR reagents (Applied Biosystems) following the manufacturer’s guidelines. Melting curve analyses were performed after PCR amplification to ensure that a single product with the expected melting curve characteristics was obtained. In addition to the primers used for IL-10 ChiP assays as mentioned, one additional pair of primers used for the control element located –1563 to –1427 of the IL-10 gene was sense 5'-CAGTCAGGAGAGAGGGCAGTGA-3' and antisense 5'-TTTCCAACAGCAGAAGC AAC-3'.
DNase I sensitivity assayed by real-time PCR
DNase I accessibility was determined as previously described (21, 22), with minor modifications. Briefly, cells grown in 100-mm tissue culture dishes were stimulated at different time intervals, and formaldehyde was added for 15 min at room temperature at a final concentration of 1%. Glycine (0.125 M) was added to neutralize formaldehyde. Cells were washed and lysed in 4 ml of ice-cold Nuclei EZ lysis buffer (Sigma-Aldrich). Cells were scrapped into conical tube, centrifuged at 500 x g for 4 min, and the nuclei were resuspended with an additional 4 ml of ice-cold Nuclei EZ lysis buffer. Washed nuclei were pooled and resuspended in ice-cold DNase I buffer (100 mM NaCl, 50 mM Tris, pH 8.0, 3 mM MgCl2, 0.15 mM spermine, and 0.5 mM spermidine) supplemented with 1 mM CaCl2. DNase I (Roche Diagnostics) was then added and incubated at 37°C for 2 min. The reaction was stopped by adding equal volume of DNase I stop buffer (containing 10 mM EDTA, 20% SDS, and 0.4 M NaCl) and incubated at 65°C for 4 h to reverse cross-links. Proteinase K (100 μg) and RNase A (10 μg) were then added at 37°C overnight. DNA was purified with phenol/chloroform extraction and ethanol precipitation. Real-time PCR was conducted as previously described (22, 23).
Results
Stimulation of macrophages in the presence of immune complexes results in the augmented activation of two MAPK, p38 and ERK
We examined the magnitude and the kinetics of MAPK activation in BMM, following stimulation in the presence or absence of immune complexes. BMM were stimulated with LPS alone or in combination with E-IgG, and cells were lysed at various intervals thereafter and total protein extracts were analyzed by Western blotting using phospho-specific Abs against p38, ERK, or JNK. LPS stimulation alone resulted in relatively modest levels of MAPK activation, which peaked at 20 min and began to decline by 40 min (Fig. 1). The combination of E-IgG and LPS resulted in a rapid and prolonged activation of ERK (Fig. 1, upper panels) and p38 (Fig. 1, middle panels). Both MAPKs were strongly phosphorylated within 5 min of stimulation. There was also an increase in the total amount of p38 and ERK phosphorylation, which persisted for 40 min (ERK) or longer (p38) (Fig. 1). The magnitude of JNK phosphorylation following LPS administration was not substantially increased by the addition of immune complexes (Fig. 1). In summary, compared with stimulation with LPS alone, stimulation of macrophages in the presence of immune complexes resulted in a more rapid and enhanced activation of ERK and p38, whereas LPS-induced JNK activation was not significantly increased by the addition of immune complexes.
The role of MAPK activation in IL-10 induction
Having observed MAPK activation following stimulation in the presence of immune complexes, we examined the effect of inhibiting MAPK on macrophage cytokine production. We previously demonstrated that the stimulation of IFN--primed macrophages in the presence of immune complexes resulted in a dramatic increase in IL-10 production (24), and a decrease in the production of IL-12 (11). We show similar data in Fig. 3, using unprimed macrophages. Stimulation of macrophages with LPS alone resulted in the production of a modest amount of IL-10, but coupling this stimulation with immune complexes caused a significant increase in IL-10 production (Fig. 3). In five separate experiments, the addition of immune complexes caused a 4.73 ± 0.45-fold increase in the production of IL-10 relative to cells stimulated with LPS alone. Stimulating macrophages with LPS alone also induced the production of the p40 subunit of IL-12 (Fig. 3, middle panel), and coupling this stimulation with E-IgG decreased IL-12 production to <200 pg/ml. Thus in Fig. 3, stimulating unprimed macrophages in the presence of immune complexes gave rise to a population of anti-inflammatory macrophages secreting 10 ng/ml IL-10 and <200 pg/ml IL-12. We have previously used several different stimuli and a variety of immune complexes, both soluble and particulate, to achieve a similar reciprocal alteration in the production of these two cytokines (5, 25).
Several recent studies have demonstrated a role for MAPKs in LPS signaling for cytokine secretion (26, 27, 28). To investigate the role of MAPKs in IL-10 induction, BMM were treated with specific pharmacological inhibitors of p38, ERK, or JNK before stimulation in the presence or absence of immune complexes (Fig. 3). Inhibition of p38 with SB203580, or ERK with the MEK inhibitor PD98059, resulted in a substantial inhibition of IL-10 secretion (Fig. 3, asterisks). Inhibition of p38 prevented the LPS-induced IL-10 production, whereas inhibition of ERK appeared to prevent the superinduction of IL-10 caused by immune complexes. Neither of these inhibitors decreased IL-12 production (Fig. 3, middle panel). In fact, IL-12 production was actually increased by the ERK inhibitor PD98059, as previously reported (29). The decrease in IL-12 production caused by the addition of immune complexes was not affected by the inhibition of either ERK or p38. Inhibition of JNK, with the JNK inhibitor 1 peptide, had no effect on IL-10 production. This inhibitor did, however, partially inhibit IL-12p40 release. For these studies, TNF was used as a control cytokine, whose production was not reproducibly affected by any of the three MAPK inhibitors (Fig. 3, lower panel).
Western blotting was performed to confirm the specificity of the MAPK inhibitors. Pretreating cells with PD98059 caused a dramatic decrease in ERK1/2 activation, but did not influence p38 activation (Fig. 4A). Similarly, treating cells with SB203580, an inhibitor that prevents phosphorylated p38 from activating downstream targets, did not inhibit ERK1/2 activation (Fig. 4). To determine the level at which IL-10 was being inhibited, real-time PCR analysis was performed to measure IL-10 mRNA in macrophages stimulated with LPS in the presence or absence of immune complexes. Macrophages stimulated in the presence of immune complexes had an increased amount of IL-10 mRNA, relative to macrophages stimulated with LPS alone (Fig. 4B). This induction of IL-10 mRNA was prevented by stimulating these cells in the presence of either ERK or p38 inhibitors (Fig. 4B). The JNK inhibitor 1 peptide had no affect on IL-10 mRNA levels (data not shown). The increased IL-10 mRNA accumulation that accompanies activation in the presence of immune complexes was not due to differences in IL-10 mRNA stability (data not shown).
To confirm a specific role for the two MAPKs in IL-10 induction, siRNA specific for p38 or ERK were transfected into primary BMM 48 h before stimulation. In all cases, gene silencing was confirmed by real-time PCR (data not shown). The primary isoform of p38 expressed in macrophages is p38. siRNA to p38 siRNA almost completely abrogated IL-10 production by macrophages stimulated with either LPS alone or LPS in combination with immune complexes (Fig. 5). Macrophages express both ERK1 and ERK2. siRNA to ERK1 almost completely prevented the augmentation of IL-10 production caused by the addition of immune complexes (Fig. 5). siRNA to ERK2 decreased IL-10 levels by 50–75% (data not shown). The combination of siRNA to both ERK1 and ERK2 completely prevented the induction of IL-10 caused by immune complexes (data not shown). Thus, both ERK1 and 2 may contribute to the induction of IL-10 caused by immune complexes, but ERK1 appears to play the dominant role in this induction. MAPK p38 appears to be required for the relatively modest levels of IL-10 that are produced in response to LPS alone.
Additional studies to examine IL-10 transcription were performed on RAW 264 macrophage-like cells, transfected with an IL-10 promoter luciferase reporter construct. By ELISA, RAW264 cells behaved similarly to primary macrophages, producing no IL-10 in response to immune complexes alone and only modest amounts when stimulated with LPS alone (Fig. 6C). Like macrophages, these cells produced much higher amounts of IL-10 when stimulated with a combination of LPS and immune complexes (Fig. 6C). The production of luciferase driven by an IL-10 promoter, however, did not reflect this superinduction. Unstimulated RAW cells expressed modest levels of luciferase activity (Fig. 6D). Stimulation of these cells with LPS resulted in a significant increase in luciferase activity (Fig. 6D), however this activity was not further increased by stimulation in the presence of immune complexes (Fig. 6D). This lack of response to immune complexes is in contrast to our previous observations with IL-12 luciferase reporter constructs, which were dramatically diminished by the addition of immune complexes (30). Thus, although immune complexes cause a dramatic increase in IL-10 secretion by macrophages, these increases were not detected by assays dependent on the regulation of extrachromosomal DNA. Neither the EMSA nor the luciferase reporter assay reflected an increase in IL-10 production following the addition of immune complexes.
Activation of MAPKs by FcR results in histone modifications at the IL-10 promoter
To determine the mechanism whereby ERK activation leads to increased IL-10 transcription, we examined histone modifications at the IL-10 locus by ChIP assays. Histone modifications, such as phosphorylation and acetylation, are thought to be important events in the regulation of gene expression (28). ERK in particular has been postulated to phosphorylate core histone proteins, including histone H3 (31). ChIP assays were performed on macrophages activated by LPS in the presence or absence of E-IgG. Macrophages treated with E-IgG had higher levels of phosphorylated H3 (serine 10) associated with the IL-10 promoter relative to resting (control) or LPS stimulated macrophages (Fig. 7A). Similar to ERK activation shown in Fig. 2, immune complexes alone were sufficient to increase IL-10 promoter-associated histone phosphorylation, and the addition of LPS to immune complexes increased this phosphorylation only slightly. We also performed ChIP assays using an Ab specific to acetylated lysines on histone H3. A similar pattern of increased acetylated histone H3 associated with the IL-10 promoter was observed following FcR ligation, although the amount of acetylation was substantially more modest (Fig. 7B). Both phosphorylation and acetylation were time-dependent events. Phosphorylation occurred rapidly and peaked at 30 min, whereas the more modest levels of histone H3 acetylation persisted for 1–2 h poststimulation (Fig. 7B). To correlate histone phosphorylation with ERK activation, similar studies were performed on macrophages that were stimulated with a soluble immune complex, IgG-OVA, in the presence or absence of PD98059 to prevent ERK activation. Similar to E-IgG, the addition of IgG-OVA caused a dramatic increase in histone phosphorylation, and this increase was completely abrogated by inhibiting ERK1/2 activation with PD98059 (Fig. 7C). ERK inhibition reduced the amount of histone phosphorylation at this locus to background levels (Fig. 7C). The inhibition of p38 with SB203580 had a more modest effect on histone H3 phosphorylation, decreasing it substantially, but not reducing it to background levels (Fig. 7C).
We examined the specificity of histone modifications following activation of macrophages in the presence of immune complexes. The IL-12(p40) promoter was used as a control. Histones associated with the IL-12 promoter were neither phosphorylated nor acetylated in response to immune complexes (Fig. 7, A and C). Therefore FcR signaling results in histone modifications that are specific to the IL-10 promoter. To determine the fine specificity of nucleosome modifications, the 12 successive nucleosomes located 5' of the IL-10 transcriptional start site were individually analyzed for modifications following stimulation. Data from three of these sites are shown in Fig. 8. The two nucleosomes comprising the Sp1 (Fig. 8A) and the STAT3 (Fig. 8B) sites underwent rapid and extensive increases in histone H3 phosphorylation following the addition of immune complexes. This phosphorylation was transient and reduced to baseline levels within 1 h poststimulation, a time at which acetylation peaks (Fig. 8). A similar analysis was performed on a control nucleosome located 1500 bp away from the transcriptional start site. There was little detectable increase in either phosphorylation or acetylation at this site (Fig. 8C). The lack of histone modifications at this site was comparable to that which occurred at the IL-12 promoter (data not shown).
Transcription factor binding to the IL-10 promoter in situ
ChIP assays were performed to examine the binding of Sp1 and STAT3 to the IL-10 promoter in situ within live cells (Fig. 9A). Control (unstimulated) cells exhibited virtually no binding of either Sp1 or STAT3 to the IL-10 promoter. Similarly, the addition of immune complexes to resting cells, a condition that does not induce IL-10 production from macrophages (see Fig. 2B), also failed to result in transcription factor binding to the IL-10 promoter (Fig. 9A). Stimulation of cells with LPS alone, a condition that induces low levels of IL-10 production, caused a modest increase in Sp1 binding to the IL-10 promoter, but no detectible STAT3 binding. However, the addition of LPS plus immune complexes induced the efficient binding of both Sp1 and STAT3 to the IL-10 promoter. Thus, the ChIP assays for Sp1 and STAT3 binding to the IL-10 promoter were accurate reflections of IL-10 transcription. Both transcription factors bound to the IL-10 promoter in situ under conditions of IL-10 superinduction. Furthermore, inhibiting ERK activation with PD98059, a condition that prevented IL-10 induction, reversed transcription factor binding to the IL-10 promoter to background levels (Fig. 9B).
Discussion
We have previously shown that activation of macrophages in the presence of immune complexes increases the production of IL-10 and reduces IL-12 production (10). This gives rise to a population of macrophages with potent anti-inflammatory properties. We have termed these cells type II activated macrophages. We have previously shown that this response to immune complexes occurs in macrophages taken from a variety of different species, including mice and human, from various anatomic locations, including the peritoneum, lung, and blood (25). The response also occurs following many different types of macrophage stimulation, including LPS, lipoteichoic acid, and CD40L (5), and in the presence of several different soluble or particulate immune complexes (6, 10). Thus, we feel that this response to FcR ligation is a universal response that is a general property of most, if not all, macrophages. In the present work we sought to determine the mechanism whereby IL-10 was induced in response to immune complexes.
Immune complexes alone were not sufficient to induce IL-10. Rather cytokine production required both the stimulation (LPS) and the addition of immune complexes. Only the combination of these two stimuli resulted in high levels of IL-10 production. Therefore, we examined signal transduction in macrophages, following the addition of each stimulus alone or in combination. LPS alone signals through TLRs to induce NF-B translocation and moderate levels of MAPK activation, as previously described (32). Although these signals were sufficient to maximally activate extrachromosomal IL-10 constructs, LPS alone induced only modest levels of IL-10 secretion by macrophages. This low level of IL-10 could be completely blocked by inhibiting p38 (Fig. 3). The coupling of LPS with immune complexes, however, resulted in a substantial increase in IL-10 production. Thus, signals generated via FcRs converge with those generated by TLRs to induce high levels of IL-10. We show that FcR ligation caused a rapid increase in ERK activation. This activation was required for IL-10 production, but not sufficient. ERK activation had to be coupled with an inflammatory stimulus to induce IL-10. The inflammatory stimuli activate the myriad transcription factors that drive cytokine and costimulatory molecule expression. In the absence of ERK activation, however, these activated transcription factors fail to effectively induce IL-10 production. In the present work, we show that activation of ERK makes the il-10 promoter accessible to these transcription factors, resulting in the production of high levels of IL-10.
Several groups have correlated cytokine production with MAPK activation (28) and some investigators have recently suggested that differential activation of the MAPKs may lead to differences in cytokine production (33). We confirm the observation of Mathur et al. (33) that p38 activation is linked to inflammatory cytokine production, and that ERK activation can lead to the production of IL-10. In the present work we show that the mechanism of IL-10 induction is the activation of ERK, which leads to chromatin modifications at the il-10 locus, to make the promoter more accessible to transcription factors that bind to it.
It has been well established that covalent modifications to chromatin, including acetylation, phosphorylation, methylation, and even ubiquitination can influence gene expression (34). In fact, some have suggested that the specific combination of histone modifications represents a code that can determine gene expression (35). In the case of differentiated lymphocytes, some of these modifications can lead to long term heritable changes in gene expression, which can define the very phenotype of the cell (36). Epigenetic changes in gene expression have been associated with T cell deviation along the Th1 or Th2 pathway. In fact, a recent study has demonstrated that IL-10 chromatin becomes altered as T cells commit to the Th2 lineage (37). The alterations described in the present work also depend on chromatin alterations, and appear to use some of the same types of histone modifications that lead replicating cells to undergo these epigenetic changes in gene expression. In the present situation, however, these changes occur quite rapidly, and their effect is transient. Alterations to chromatin are observed within the first 15 min of stimulation, and they can be reversed as quickly as 1 h later (Fig. 8). Furthermore, in end-stage cells such as macrophages, these need not be heritable alterations that can be passed on to daughter cells, and therefore their effect is transient and reversible.
Although histone phosphorylation was originally associated with chromatin condensation and gene silencing during cell division (38), several studies have correlated histone modifications and specifically phosphorylation of the serine 10 residue on histone H3 with transcriptional activation (28, 39). In fact, a human genetic disease, Coffin-Lowry syndrome, is characterized by impaired transcription of c-fos and defective histone H3 phosphorylation (40). The acetylation of histones has also been linked to transcriptional activation (41), and frequently histone acetylation occurs in association with histone phosphorylation. In yeast phosphorylation often precedes and can be a prerequisite for histone acetylation (42), whereas in Drosophila these two modifications may be independently regulated (43). Although the dramatic increase in early histone H3 phosphorylation, following exposure of macrophages to immune complexes, requires ERK activation, it is unlikely that ERK directly modifies chromatin. Rather, several histone H3 kinases have been identified that represent candidates for the observed phosphorylation events. Research is underway to identify the IL-10-associated histone kinase. Importantly, the alterations to chromatin that we observe appear to be restricted to the il-10 gene, in that no such modifications are observed at the il-12 gene. Further analyses to determine the mechanism of this modification are underway.
The increase in IL-10 production following activation in the presence of immune complexes makes these macrophages potent anti-inflammatory cells (5). In the present work we show that stimulation of cells with LPS alone leads to modest levels of ERK activation, modest binding of Sp1 to the IL-10 promoter in situ, and only low levels of IL-10 gene expression. Coupling stimulation with FcR ligation, however, leads to increased ERK activation, histone H3 modifications at the il-10 locus, and dramatic increases in Sp1 and STAT3 binding to the IL-10 promoter. We conclude that these modifications are required for the high levels of IL-10 that are produced by macrophages activated in the presence of immune complexes, and suggest that manipulating MAPK activation in macrophages can change the phenotype of the activated macrophage.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported in part by Grant AI49383 from the National Institutes of Health.
2 Address correspondence and reprint requests to Dr. David M. Mosser, Department of Cell Biology and Molecular Genetics, 1103 Microbiology Building, University of Maryland, College Park, MD 20742. E-mail address: dmosser@umd.edu
3 in this paper: BMM, bone marrow-derived macrophage; ChIP, chromatin immunoprecipitation; siRNA, small interfering RNA; E-IgG, IgG-opsonized erythrocyte.
Received for publication January 11, 2005. Accepted for publication April 13, 2005.
References
Gordon, S.. 2003. Alternative activation of macrophages. Nat. Rev. Immunol. 3: 23-35
Ridley, M. G., G. Kingsley, C. Pitzalis, G. S. Panayi. 1990. Monocyte activation in rheumatoid arthritis: evidence for in situ activation and differentiation in joints. Br. J. Rheumatol. 29: 84-88
Compston, A.. 1993. Inflammation and the brain. Mol. Chem. Neuropathol. 19: 47-64
Hausmann, M., T. Spottl, T. Andus, G. Rothe, W. Falk, J. Scholmerich, H. Herfarth, G. Rogler. 2001. Subtractive screening reveals up-regulation of NADPH oxidase expression in Crohn’s disease intestinal macrophages. Clin. Exp. Immunol. 125: 48-55
Gerber, J. S., D. M. Mosser. 2001. Reversing lipopolysaccharide toxicity by ligating the macrophage Fc receptors. J. Immunol. 166: 6861-6868(Mark Lucas, Xia Zhang, Vi)
We have previously demonstrated that macrophages stimulated in the presence of immune complexes produce high levels of IL-10. We now examine the mechanism of IL-10 superinduction. We report that the enhanced production of IL-10 correlates with a rapid and enhanced activation of two MAPKs, ERK and p38. The inhibition of either ERK or p38 prevented IL-10 induction, indicating that both MAPKs were required for IL-10 synthesis. By chromatin immunoprecipitation assay, we demonstrate that activation of ERK leads to the phosphorylation of serine 10 on histone H3 at the il-10 gene, making the promoter more accessible to transcription factors generated in response to p38 activation. Inhibition of ERK activation prevented histone modifications, and decreased the binding of Sp1 and STAT3 to the IL-10 promoter. We conclude that the activation of ERK following FcR ligation leads to a remodeling of the chromatin at the il-10 locus, making it more accessible to transcription factors. The rapid and transient regulation of transcription factor accessibility to the IL-10 promoter by MAPK activation represents a novel way that the production of this cytokine is regulated.
Introduction
Activated macrophages are major mediators of inflammation. These cells secrete a variety of inflammatory cytokines, lipid mediators, and potentially toxic radicals of oxygen and nitrogen (reviewed in Ref. 1). Consequently they can contribute to the pathology associated with a number of autoimmune diseases, such as rheumatoid arthritis (2), multiple sclerosis (3), and inflammatory bowel disease (4) to mention a few. We have recently shown that macrophages activated in the presence of immune complexes secreted high levels of IL-10. The IL-10 produced by these cells could reverse lethal endotoxemia (5) and prevent Th1-type adaptive immune responses (6). Thus, immune complexes give rise to a potent anti-inflammatory population of macrophages that produce large amounts of IL-10. In the present work, we examine the mechanism of IL-10 induction in these macrophages.
Several transcription factors have been implicated in IL-10 transcription. Sp1 has been shown to play an important role in IL-10 transcription, and an Sp1 responsive element in the IL-10 promoter was localized at –89 to –78 (7). The STAT3 transcription factor has also been shown to bind to an element in the human IL-10 promoter, and a dominant negative form of STAT3 has been shown to diminish IL-10 transcription (8). A role for C/EBP has been suggested (9), although it may not be required for IL-10 transcription (7). Despite the identification of these transcription factors, the regulation of IL-10 transcription in macrophages remains somewhat elusive. One of the reasons for ambiguity is that the common stimulus used to induce IL-10 in macrophages is LPS. This compound is not a particularly potent stimulator of IL-10 production, typically inducing only in the hundred-picogram range of IL-10, which correlates with modest increases in IL-10 transcription. Thus, assays to identify transcriptional elements within the IL-10 promoter are not particularly robust. In the present study, we take advantage of the fact that stimulating macrophages in the presence of immune complexes induces the production of high levels of IL-10. This creates a robust model in which to study IL-10 transcription. Importantly, immune complexes alone do not induce cytokine production from macrophages (5). However, when these complexes are combined with an inflammatory stimulus such as LPS, the two stimuli synergize to induce high levels of IL-10. The mechanism of this superinduction is the topic of the present study.
We have previously reported that activation of macrophages in the presence of immune complexes also results in an abrogation of IL-12 production (10). The synthesis of IL-12 and its regulation are more well understood than IL-10. There have been several reports describing the down-regulation of IL-12 subsequent to macrophage stimulation. The ligation of any number of macrophage receptors, including the FcRs (11), complement receptors (12), scavenger receptors (11), as well as G protein-coupled receptors (13) have all been linked to suppressed IL-12 production, regardless of the stimuli. Similarly, ligation of 2 adrenergic receptors (14) and exposure of macrophages to vitamin D3 (15), or to macrophage stimulatory protein (16), can also inhibit IL-12 production. Finally, infection of macrophages or dendritic cells with a variety of microorganisms, including measles virus (17), herpes virus (18), Leishmania (19), or mycobacteria (20) can all result in a reduction of IL-12 secretion. The molecular mechanism for this inhibition may be quite complex. Regardless of the mechanism(s), virtually all of the inhibition appears to occur at the level of IL-12 transcription, resulting in decreased mRNA production and decreased or absent protein secretion. In the present studies, the down-regulation of IL-12 in response to immune complexes is used as a control, to contrast what occurs at the il-10 gene.
Having identified the reciprocal alteration in the production of IL-10 (increase) and IL-12 (decrease) by immune complexes (6), we began to study the molecular aspects of cytokine gene expression. We reasoned that the identification of the mechanism whereby these two cytokines are regulated could lead to the development of novel therapeutic strategies that would predictably alter macrophage phenotypes, thereby influencing innate and adaptive immunity. In the present work, we show that FcR ligation of macrophages leads to ERK activation. Activation of ERK leads to transient modifications in the chromatin at the IL-10 locus. These alterations to the chromatin are necessary to allow the efficient transcription of IL-10, which occurs in macrophages that are stimulated in the presence of immune complexes.
Materials and Methods
Mice
Six- to 8-wk-old BALB/c mice were purchased from Taconic Farms. All mice were maintained in HEPA-filtered Thoren units (Thoren Caging Systems) at the University of Maryland (College Park, MD). Mice were used at 6–10 wk of age as a source of bone marrow-derived macrophages (BMM).3 All procedures were reviewed and approved by the University of Maryland Institutional Animal Care and Use Committee.
Reagents
The p38 MAPK inhibitor, SB203580, and the MEK/ERK inhibitor, PD98059, were purchased from Calbiochem. The JNK inhibitor peptide I was purchased from Alexis, USA (San Diego, CA). Washed SRBC were purchased from Lampire. Rabbit IgG Ab to SRBC (anti-SRBC IgG) was purchased from Cappel. Ultra pure LPS from Escherichia coli K12 strain was obtained from InvivoGen. Anti-phospho-H3 and acetyl-H3 Abs and chromatin immunoprecipitation (ChIP) assay kits were purchased from Upstate Biotechnology. Anti-p38 (phospho-T180/Y182), anti-STAT3, and anti-Sp1 Abs were purchased from Abcam. Anti-p38 (total), anti-ERK1/2 (total and phospho-T202/Y204), and anti-JNK (phospho-T183/Y185) were purchased from Cell Signaling Technology.
Cells
BMM were prepared as previously described (6). Briefly, bone marrow was flushed from the femurs and tibias of mice and cells were plated in petri dishes in DMEM/F12 supplemented with 10% FBS, penicillin/streptomycin, glutamine, and 10% conditioned medium from the supernatant of M-CSF secreting L929 (LCM) fibroblasts. Cells were fed on day 2, and complete medium was replaced on day 6. Cells were used at 7–10 days for experiments. The RAW264.7 macrophage cell line (American Type Culture Collection) was maintained in RPMI 1640 supplemented with 10% FBS, penicillin/streptomycin and glutamine (Invitrogen Life Technologies).
Immune complexes
IgG-opsonized erythrocytes (E-IgG) were generated by incubating SRBC with anti-SRBC IgG at nonagglutinating titers for 30 min at room temperature while rotating. Opsonized cells were washed once in HBSS (Invitrogen Life Technologies) and resuspended in complete medium. E-IgG were added to macrophages at a ratio of 10:1 E-IgG to macrophages. For some experiments IgG-OVA was used as the immune complex. IgG-OVA was prepared as previously described (6).
Cell stimulation assays
For cytokine analysis, 3 x 105 macrophages per well were plated overnight in a 48-well plate in DMEM/F12. Cells were then washed and activated with either 10 ng/ml LPS alone or in combination with a 10:1 ratio of E-IgG to macrophages. Supernatants were harvested 20 h later. Cytokines were measured by ELISA using the following Ab pairs from BD Pharmingen: IL-12p40, C15.6 and C17.8; IL-10, JES5-2A5 and JES5-16E3; TNF-, G281-2626 and MP6-XT22.
Generation of small interfering RNA (siRNA) and cell transfections
Nuclear extracts were prepared from 2 x 107 macrophages using the Nuclear Extraction kit (Panomics) following the manufacturer’s protocol. EMSAs were conducted using the EMSA "Gel-Shift" kit (Panomics). Briefly, 5 μg of nuclear extract were incubated with biotin-Sp1 or STAT3 probe with or without unlabeled probe for 30 min at 20°C, and then run on a 6% polyacrylamide gel. Oligonucleotide-protein complexes were transferred to Biodyne B membrane (Pall). Following transfer, membranes were incubated with streptavidin-HRP and protein visualized by chemiluminescence.
Luciferase assay
RAW264.7 macrophages (4 x 106) were transfected with 5 μg of pGL-IL10-luciferase reporter plasmid (7), which is a generous gift of Dr. S. Smale (Howard Hughes Medical Institute, University of California, Los Angeles, CA) with the Amaxa Nucleofector system. After transfection, 3 x 105 cells were plated per well in 48-well culture plates. After 24 h, cells were washed and stimulated, then lysed using Glo-Lysis buffer, and luciferase activity was measured using the Bright-Glo Luciferase system (Promega).
ChIP assay
ChIP assays were conducted using the ChIP Assay kit following the manufacturer’s protocol (Upstate Biotechnology). Briefly, 1 x 106 BMM were plated overnight in six-well plates. Cells were stimulated as described in figures, then fixed for 10 min at 37°C in 1% paraformaldehyde. Cells were washed on ice with ice-cold HBSS containing 1 mM PMSF, harvested and then lysed in SDS lysis buffer. DNA was sheared by ultrasonication using a High Intensity Ultrasonic Processor (Cole-Parmer) for 3 x 10 s pulses at 20% amplitude. Lysates were cleared by centrifugation and diluted in ChIP dilution buffer. Lysates were precleared using salmon sperm DNA/protein A-agarose and a sample of "input DNA" was collected at this point. Protein-DNA complexes were immunoprecipitated with 5 μg of Ab overnight at 4°C. Ab-protein-DNA complexes were then captured using salmon sperm DNA/protein A-agarose for 1 h at 4°C. After washing beads with low and high salt, LiCl, and TE buffers, the protein/DNA complexes were eluted using 1% SDS, 0.1 M NaHCO3 buffer and disrupted by heating at 65°C for 4 h. DNA was then extracted using phenol/chloroform extraction and ethanol precipitation. PCR was conducted using promoter specific primers: IL-10 promoter (Sp1 binding region, –294 to –73) sense 5'-CAGCTGTCTGCCTCAGGAAATACAA-3', antisense 5'-TATTCAGGCTCCTCCTCCCTCTTCT-3' (94°C, 15 s; 60°C, 30 s; 72°C, 1 min, 35 cycles); IL-10 promoter (STAT3 binding region, –649 to –448) sense 5'-TCATGCTGGGATCTGAGCTTCT-3', antisense 5'-CGGAAGTCACCTTAGCACTCAGT-3' (94°C, 15 s; 56°C, 30 s; 72°C, 1 min, 35 cycles); IL-12p40 promoter sense 5'-CAAATCTGGGAGGCAGGAAAC-3', antisense 5'-CAAAGCAAACCTTTCTATCAAATACACA-3' (94°C, 15 s; 56°C, 30 s; 72°C, 1 min, 35 cycles). Numerical designations are according to GenBank accession number M84340. PCR products were separated on 2% agarose gels. For relative quantitation of promoter levels, real-time PCR was performed.
Real-time PCR
Real-time PCR was conducted with the ABI Prism 7700 Sequence Detection System using SYBR Green PCR reagents (Applied Biosystems) following the manufacturer’s guidelines. Melting curve analyses were performed after PCR amplification to ensure that a single product with the expected melting curve characteristics was obtained. In addition to the primers used for IL-10 ChiP assays as mentioned, one additional pair of primers used for the control element located –1563 to –1427 of the IL-10 gene was sense 5'-CAGTCAGGAGAGAGGGCAGTGA-3' and antisense 5'-TTTCCAACAGCAGAAGC AAC-3'.
DNase I sensitivity assayed by real-time PCR
DNase I accessibility was determined as previously described (21, 22), with minor modifications. Briefly, cells grown in 100-mm tissue culture dishes were stimulated at different time intervals, and formaldehyde was added for 15 min at room temperature at a final concentration of 1%. Glycine (0.125 M) was added to neutralize formaldehyde. Cells were washed and lysed in 4 ml of ice-cold Nuclei EZ lysis buffer (Sigma-Aldrich). Cells were scrapped into conical tube, centrifuged at 500 x g for 4 min, and the nuclei were resuspended with an additional 4 ml of ice-cold Nuclei EZ lysis buffer. Washed nuclei were pooled and resuspended in ice-cold DNase I buffer (100 mM NaCl, 50 mM Tris, pH 8.0, 3 mM MgCl2, 0.15 mM spermine, and 0.5 mM spermidine) supplemented with 1 mM CaCl2. DNase I (Roche Diagnostics) was then added and incubated at 37°C for 2 min. The reaction was stopped by adding equal volume of DNase I stop buffer (containing 10 mM EDTA, 20% SDS, and 0.4 M NaCl) and incubated at 65°C for 4 h to reverse cross-links. Proteinase K (100 μg) and RNase A (10 μg) were then added at 37°C overnight. DNA was purified with phenol/chloroform extraction and ethanol precipitation. Real-time PCR was conducted as previously described (22, 23).
Results
Stimulation of macrophages in the presence of immune complexes results in the augmented activation of two MAPK, p38 and ERK
We examined the magnitude and the kinetics of MAPK activation in BMM, following stimulation in the presence or absence of immune complexes. BMM were stimulated with LPS alone or in combination with E-IgG, and cells were lysed at various intervals thereafter and total protein extracts were analyzed by Western blotting using phospho-specific Abs against p38, ERK, or JNK. LPS stimulation alone resulted in relatively modest levels of MAPK activation, which peaked at 20 min and began to decline by 40 min (Fig. 1). The combination of E-IgG and LPS resulted in a rapid and prolonged activation of ERK (Fig. 1, upper panels) and p38 (Fig. 1, middle panels). Both MAPKs were strongly phosphorylated within 5 min of stimulation. There was also an increase in the total amount of p38 and ERK phosphorylation, which persisted for 40 min (ERK) or longer (p38) (Fig. 1). The magnitude of JNK phosphorylation following LPS administration was not substantially increased by the addition of immune complexes (Fig. 1). In summary, compared with stimulation with LPS alone, stimulation of macrophages in the presence of immune complexes resulted in a more rapid and enhanced activation of ERK and p38, whereas LPS-induced JNK activation was not significantly increased by the addition of immune complexes.
The role of MAPK activation in IL-10 induction
Having observed MAPK activation following stimulation in the presence of immune complexes, we examined the effect of inhibiting MAPK on macrophage cytokine production. We previously demonstrated that the stimulation of IFN--primed macrophages in the presence of immune complexes resulted in a dramatic increase in IL-10 production (24), and a decrease in the production of IL-12 (11). We show similar data in Fig. 3, using unprimed macrophages. Stimulation of macrophages with LPS alone resulted in the production of a modest amount of IL-10, but coupling this stimulation with immune complexes caused a significant increase in IL-10 production (Fig. 3). In five separate experiments, the addition of immune complexes caused a 4.73 ± 0.45-fold increase in the production of IL-10 relative to cells stimulated with LPS alone. Stimulating macrophages with LPS alone also induced the production of the p40 subunit of IL-12 (Fig. 3, middle panel), and coupling this stimulation with E-IgG decreased IL-12 production to <200 pg/ml. Thus in Fig. 3, stimulating unprimed macrophages in the presence of immune complexes gave rise to a population of anti-inflammatory macrophages secreting 10 ng/ml IL-10 and <200 pg/ml IL-12. We have previously used several different stimuli and a variety of immune complexes, both soluble and particulate, to achieve a similar reciprocal alteration in the production of these two cytokines (5, 25).
Several recent studies have demonstrated a role for MAPKs in LPS signaling for cytokine secretion (26, 27, 28). To investigate the role of MAPKs in IL-10 induction, BMM were treated with specific pharmacological inhibitors of p38, ERK, or JNK before stimulation in the presence or absence of immune complexes (Fig. 3). Inhibition of p38 with SB203580, or ERK with the MEK inhibitor PD98059, resulted in a substantial inhibition of IL-10 secretion (Fig. 3, asterisks). Inhibition of p38 prevented the LPS-induced IL-10 production, whereas inhibition of ERK appeared to prevent the superinduction of IL-10 caused by immune complexes. Neither of these inhibitors decreased IL-12 production (Fig. 3, middle panel). In fact, IL-12 production was actually increased by the ERK inhibitor PD98059, as previously reported (29). The decrease in IL-12 production caused by the addition of immune complexes was not affected by the inhibition of either ERK or p38. Inhibition of JNK, with the JNK inhibitor 1 peptide, had no effect on IL-10 production. This inhibitor did, however, partially inhibit IL-12p40 release. For these studies, TNF was used as a control cytokine, whose production was not reproducibly affected by any of the three MAPK inhibitors (Fig. 3, lower panel).
Western blotting was performed to confirm the specificity of the MAPK inhibitors. Pretreating cells with PD98059 caused a dramatic decrease in ERK1/2 activation, but did not influence p38 activation (Fig. 4A). Similarly, treating cells with SB203580, an inhibitor that prevents phosphorylated p38 from activating downstream targets, did not inhibit ERK1/2 activation (Fig. 4). To determine the level at which IL-10 was being inhibited, real-time PCR analysis was performed to measure IL-10 mRNA in macrophages stimulated with LPS in the presence or absence of immune complexes. Macrophages stimulated in the presence of immune complexes had an increased amount of IL-10 mRNA, relative to macrophages stimulated with LPS alone (Fig. 4B). This induction of IL-10 mRNA was prevented by stimulating these cells in the presence of either ERK or p38 inhibitors (Fig. 4B). The JNK inhibitor 1 peptide had no affect on IL-10 mRNA levels (data not shown). The increased IL-10 mRNA accumulation that accompanies activation in the presence of immune complexes was not due to differences in IL-10 mRNA stability (data not shown).
To confirm a specific role for the two MAPKs in IL-10 induction, siRNA specific for p38 or ERK were transfected into primary BMM 48 h before stimulation. In all cases, gene silencing was confirmed by real-time PCR (data not shown). The primary isoform of p38 expressed in macrophages is p38. siRNA to p38 siRNA almost completely abrogated IL-10 production by macrophages stimulated with either LPS alone or LPS in combination with immune complexes (Fig. 5). Macrophages express both ERK1 and ERK2. siRNA to ERK1 almost completely prevented the augmentation of IL-10 production caused by the addition of immune complexes (Fig. 5). siRNA to ERK2 decreased IL-10 levels by 50–75% (data not shown). The combination of siRNA to both ERK1 and ERK2 completely prevented the induction of IL-10 caused by immune complexes (data not shown). Thus, both ERK1 and 2 may contribute to the induction of IL-10 caused by immune complexes, but ERK1 appears to play the dominant role in this induction. MAPK p38 appears to be required for the relatively modest levels of IL-10 that are produced in response to LPS alone.
Additional studies to examine IL-10 transcription were performed on RAW 264 macrophage-like cells, transfected with an IL-10 promoter luciferase reporter construct. By ELISA, RAW264 cells behaved similarly to primary macrophages, producing no IL-10 in response to immune complexes alone and only modest amounts when stimulated with LPS alone (Fig. 6C). Like macrophages, these cells produced much higher amounts of IL-10 when stimulated with a combination of LPS and immune complexes (Fig. 6C). The production of luciferase driven by an IL-10 promoter, however, did not reflect this superinduction. Unstimulated RAW cells expressed modest levels of luciferase activity (Fig. 6D). Stimulation of these cells with LPS resulted in a significant increase in luciferase activity (Fig. 6D), however this activity was not further increased by stimulation in the presence of immune complexes (Fig. 6D). This lack of response to immune complexes is in contrast to our previous observations with IL-12 luciferase reporter constructs, which were dramatically diminished by the addition of immune complexes (30). Thus, although immune complexes cause a dramatic increase in IL-10 secretion by macrophages, these increases were not detected by assays dependent on the regulation of extrachromosomal DNA. Neither the EMSA nor the luciferase reporter assay reflected an increase in IL-10 production following the addition of immune complexes.
Activation of MAPKs by FcR results in histone modifications at the IL-10 promoter
To determine the mechanism whereby ERK activation leads to increased IL-10 transcription, we examined histone modifications at the IL-10 locus by ChIP assays. Histone modifications, such as phosphorylation and acetylation, are thought to be important events in the regulation of gene expression (28). ERK in particular has been postulated to phosphorylate core histone proteins, including histone H3 (31). ChIP assays were performed on macrophages activated by LPS in the presence or absence of E-IgG. Macrophages treated with E-IgG had higher levels of phosphorylated H3 (serine 10) associated with the IL-10 promoter relative to resting (control) or LPS stimulated macrophages (Fig. 7A). Similar to ERK activation shown in Fig. 2, immune complexes alone were sufficient to increase IL-10 promoter-associated histone phosphorylation, and the addition of LPS to immune complexes increased this phosphorylation only slightly. We also performed ChIP assays using an Ab specific to acetylated lysines on histone H3. A similar pattern of increased acetylated histone H3 associated with the IL-10 promoter was observed following FcR ligation, although the amount of acetylation was substantially more modest (Fig. 7B). Both phosphorylation and acetylation were time-dependent events. Phosphorylation occurred rapidly and peaked at 30 min, whereas the more modest levels of histone H3 acetylation persisted for 1–2 h poststimulation (Fig. 7B). To correlate histone phosphorylation with ERK activation, similar studies were performed on macrophages that were stimulated with a soluble immune complex, IgG-OVA, in the presence or absence of PD98059 to prevent ERK activation. Similar to E-IgG, the addition of IgG-OVA caused a dramatic increase in histone phosphorylation, and this increase was completely abrogated by inhibiting ERK1/2 activation with PD98059 (Fig. 7C). ERK inhibition reduced the amount of histone phosphorylation at this locus to background levels (Fig. 7C). The inhibition of p38 with SB203580 had a more modest effect on histone H3 phosphorylation, decreasing it substantially, but not reducing it to background levels (Fig. 7C).
We examined the specificity of histone modifications following activation of macrophages in the presence of immune complexes. The IL-12(p40) promoter was used as a control. Histones associated with the IL-12 promoter were neither phosphorylated nor acetylated in response to immune complexes (Fig. 7, A and C). Therefore FcR signaling results in histone modifications that are specific to the IL-10 promoter. To determine the fine specificity of nucleosome modifications, the 12 successive nucleosomes located 5' of the IL-10 transcriptional start site were individually analyzed for modifications following stimulation. Data from three of these sites are shown in Fig. 8. The two nucleosomes comprising the Sp1 (Fig. 8A) and the STAT3 (Fig. 8B) sites underwent rapid and extensive increases in histone H3 phosphorylation following the addition of immune complexes. This phosphorylation was transient and reduced to baseline levels within 1 h poststimulation, a time at which acetylation peaks (Fig. 8). A similar analysis was performed on a control nucleosome located 1500 bp away from the transcriptional start site. There was little detectable increase in either phosphorylation or acetylation at this site (Fig. 8C). The lack of histone modifications at this site was comparable to that which occurred at the IL-12 promoter (data not shown).
Transcription factor binding to the IL-10 promoter in situ
ChIP assays were performed to examine the binding of Sp1 and STAT3 to the IL-10 promoter in situ within live cells (Fig. 9A). Control (unstimulated) cells exhibited virtually no binding of either Sp1 or STAT3 to the IL-10 promoter. Similarly, the addition of immune complexes to resting cells, a condition that does not induce IL-10 production from macrophages (see Fig. 2B), also failed to result in transcription factor binding to the IL-10 promoter (Fig. 9A). Stimulation of cells with LPS alone, a condition that induces low levels of IL-10 production, caused a modest increase in Sp1 binding to the IL-10 promoter, but no detectible STAT3 binding. However, the addition of LPS plus immune complexes induced the efficient binding of both Sp1 and STAT3 to the IL-10 promoter. Thus, the ChIP assays for Sp1 and STAT3 binding to the IL-10 promoter were accurate reflections of IL-10 transcription. Both transcription factors bound to the IL-10 promoter in situ under conditions of IL-10 superinduction. Furthermore, inhibiting ERK activation with PD98059, a condition that prevented IL-10 induction, reversed transcription factor binding to the IL-10 promoter to background levels (Fig. 9B).
Discussion
We have previously shown that activation of macrophages in the presence of immune complexes increases the production of IL-10 and reduces IL-12 production (10). This gives rise to a population of macrophages with potent anti-inflammatory properties. We have termed these cells type II activated macrophages. We have previously shown that this response to immune complexes occurs in macrophages taken from a variety of different species, including mice and human, from various anatomic locations, including the peritoneum, lung, and blood (25). The response also occurs following many different types of macrophage stimulation, including LPS, lipoteichoic acid, and CD40L (5), and in the presence of several different soluble or particulate immune complexes (6, 10). Thus, we feel that this response to FcR ligation is a universal response that is a general property of most, if not all, macrophages. In the present work we sought to determine the mechanism whereby IL-10 was induced in response to immune complexes.
Immune complexes alone were not sufficient to induce IL-10. Rather cytokine production required both the stimulation (LPS) and the addition of immune complexes. Only the combination of these two stimuli resulted in high levels of IL-10 production. Therefore, we examined signal transduction in macrophages, following the addition of each stimulus alone or in combination. LPS alone signals through TLRs to induce NF-B translocation and moderate levels of MAPK activation, as previously described (32). Although these signals were sufficient to maximally activate extrachromosomal IL-10 constructs, LPS alone induced only modest levels of IL-10 secretion by macrophages. This low level of IL-10 could be completely blocked by inhibiting p38 (Fig. 3). The coupling of LPS with immune complexes, however, resulted in a substantial increase in IL-10 production. Thus, signals generated via FcRs converge with those generated by TLRs to induce high levels of IL-10. We show that FcR ligation caused a rapid increase in ERK activation. This activation was required for IL-10 production, but not sufficient. ERK activation had to be coupled with an inflammatory stimulus to induce IL-10. The inflammatory stimuli activate the myriad transcription factors that drive cytokine and costimulatory molecule expression. In the absence of ERK activation, however, these activated transcription factors fail to effectively induce IL-10 production. In the present work, we show that activation of ERK makes the il-10 promoter accessible to these transcription factors, resulting in the production of high levels of IL-10.
Several groups have correlated cytokine production with MAPK activation (28) and some investigators have recently suggested that differential activation of the MAPKs may lead to differences in cytokine production (33). We confirm the observation of Mathur et al. (33) that p38 activation is linked to inflammatory cytokine production, and that ERK activation can lead to the production of IL-10. In the present work we show that the mechanism of IL-10 induction is the activation of ERK, which leads to chromatin modifications at the il-10 locus, to make the promoter more accessible to transcription factors that bind to it.
It has been well established that covalent modifications to chromatin, including acetylation, phosphorylation, methylation, and even ubiquitination can influence gene expression (34). In fact, some have suggested that the specific combination of histone modifications represents a code that can determine gene expression (35). In the case of differentiated lymphocytes, some of these modifications can lead to long term heritable changes in gene expression, which can define the very phenotype of the cell (36). Epigenetic changes in gene expression have been associated with T cell deviation along the Th1 or Th2 pathway. In fact, a recent study has demonstrated that IL-10 chromatin becomes altered as T cells commit to the Th2 lineage (37). The alterations described in the present work also depend on chromatin alterations, and appear to use some of the same types of histone modifications that lead replicating cells to undergo these epigenetic changes in gene expression. In the present situation, however, these changes occur quite rapidly, and their effect is transient. Alterations to chromatin are observed within the first 15 min of stimulation, and they can be reversed as quickly as 1 h later (Fig. 8). Furthermore, in end-stage cells such as macrophages, these need not be heritable alterations that can be passed on to daughter cells, and therefore their effect is transient and reversible.
Although histone phosphorylation was originally associated with chromatin condensation and gene silencing during cell division (38), several studies have correlated histone modifications and specifically phosphorylation of the serine 10 residue on histone H3 with transcriptional activation (28, 39). In fact, a human genetic disease, Coffin-Lowry syndrome, is characterized by impaired transcription of c-fos and defective histone H3 phosphorylation (40). The acetylation of histones has also been linked to transcriptional activation (41), and frequently histone acetylation occurs in association with histone phosphorylation. In yeast phosphorylation often precedes and can be a prerequisite for histone acetylation (42), whereas in Drosophila these two modifications may be independently regulated (43). Although the dramatic increase in early histone H3 phosphorylation, following exposure of macrophages to immune complexes, requires ERK activation, it is unlikely that ERK directly modifies chromatin. Rather, several histone H3 kinases have been identified that represent candidates for the observed phosphorylation events. Research is underway to identify the IL-10-associated histone kinase. Importantly, the alterations to chromatin that we observe appear to be restricted to the il-10 gene, in that no such modifications are observed at the il-12 gene. Further analyses to determine the mechanism of this modification are underway.
The increase in IL-10 production following activation in the presence of immune complexes makes these macrophages potent anti-inflammatory cells (5). In the present work we show that stimulation of cells with LPS alone leads to modest levels of ERK activation, modest binding of Sp1 to the IL-10 promoter in situ, and only low levels of IL-10 gene expression. Coupling stimulation with FcR ligation, however, leads to increased ERK activation, histone H3 modifications at the il-10 locus, and dramatic increases in Sp1 and STAT3 binding to the IL-10 promoter. We conclude that these modifications are required for the high levels of IL-10 that are produced by macrophages activated in the presence of immune complexes, and suggest that manipulating MAPK activation in macrophages can change the phenotype of the activated macrophage.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported in part by Grant AI49383 from the National Institutes of Health.
2 Address correspondence and reprint requests to Dr. David M. Mosser, Department of Cell Biology and Molecular Genetics, 1103 Microbiology Building, University of Maryland, College Park, MD 20742. E-mail address: dmosser@umd.edu
3 in this paper: BMM, bone marrow-derived macrophage; ChIP, chromatin immunoprecipitation; siRNA, small interfering RNA; E-IgG, IgG-opsonized erythrocyte.
Received for publication January 11, 2005. Accepted for publication April 13, 2005.
References
Gordon, S.. 2003. Alternative activation of macrophages. Nat. Rev. Immunol. 3: 23-35
Ridley, M. G., G. Kingsley, C. Pitzalis, G. S. Panayi. 1990. Monocyte activation in rheumatoid arthritis: evidence for in situ activation and differentiation in joints. Br. J. Rheumatol. 29: 84-88
Compston, A.. 1993. Inflammation and the brain. Mol. Chem. Neuropathol. 19: 47-64
Hausmann, M., T. Spottl, T. Andus, G. Rothe, W. Falk, J. Scholmerich, H. Herfarth, G. Rogler. 2001. Subtractive screening reveals up-regulation of NADPH oxidase expression in Crohn’s disease intestinal macrophages. Clin. Exp. Immunol. 125: 48-55
Gerber, J. S., D. M. Mosser. 2001. Reversing lipopolysaccharide toxicity by ligating the macrophage Fc receptors. J. Immunol. 166: 6861-6868(Mark Lucas, Xia Zhang, Vi)