ABSTRACT
Mycobacterium tuberculosis survives within macrophages and employs immune evasion mechanisms to persist in the host. Protective T helper type 1 (Th1) responses are induced, and the immune response in most individuals is sufficient to restrict M. tuberculosis to latent infection, but most infections are not completely resolved. As T cells and macrophages respond, a balance is established between protective Th1-associated and other proinflammatory cytokines, such as interleukin-12 (IL-12), interferon gamma (IFN-γ), and tumor necrosis factor alpha, and anti-inflammatory cytokines, such as IL-10. The mechanisms by which M. tuberculosis modulates host responses to promote its survival remain unclear. In these studies, we demonstrate that M. tuberculosis induction of IL-10, suppression of IL-12, and inhibition of class II major histocompatibility complex (MHC-II) molecules in infected macrophages are all driven by Toll-like receptor 2 (TLR2)-dependent activation of the extracellular signal-regulated kinases (ERK). Elimination of ERK signaling downstream of TLR2 by pharmacologic inhibition with U0126 or genetic deletion of Tpl2 blocks IL-10 secretion and enhances IL-12 p70 secretion. We demonstrate that M. tuberculosis regulation of these pathways in macrophages affects T cell responses to infected macrophages. Thus, genetic blockade of the ERK pathway in Tpl2 −/− macrophages enhances Th1 polarization and IFN-γ production by antigen-specific CD4+ T cells responding to M. tuberculosis infection. These data indicate that M. tuberculosis and its potent TLR2 ligands activate ERK signaling in macrophages to promote anti-inflammatory macrophage responses and blunt Th1 responses against the pathogen.
INTRODUCTION
Tuberculosis, caused by infection with Mycobacterium tuberculosis, remains a major disease worldwide, causing an estimated 2 million deaths per year. M. tuberculosis infection is spread by aerosol, and initial infection mainly occurs in the lungs (1), where M. tuberculosis persists as an intracellular pathogen harbored by macrophages. Infection of alveolar and tissue-resident macrophages leads to engagement of innate immune receptors by pathogen-derived molecules and activates macrophage responses that help contain the infection (2, 3) but fail to eradicate it. T helper type 1 (Th1) responses and the production of interferon gamma (IFN-γ) are particularly important to the containment of M. tuberculosis infection (4–6), but T cells exhibit delayed responses in the lung and do not provide sterilizing immunity (7–10). Effector T cells may exhibit plasticity in their Th1 polarization due to effects of the lung microenvironment (11–13).
M. tuberculosis possesses mechanisms to interfere with host immunity and establish latent infection, enabling it to persist primarily within macrophages in lung granulomas (14). Some immune evasion mechanisms affect macrophage functions; examples are interference with macrophage microbicidal responses, such as reactive oxygen and reactive nitrogen intermediates (15, 16); suppression of class II major histocompatibility complex (MHC-II) expression and, hence, presentation of antigens to CD4+ T cells (17–21); and regulation of cytokines expressed by macrophages, e.g., the induction of interleukin-10 (IL-10), which has immune-suppressive functions (22–24). Regulation of some macrophage functions, such as cytokine and MHC expression, may influence the polarization and functions of T cells responding to M. tuberculosis-infected macrophages. We propose that M. tuberculosis regulation of macrophages affects the responses of effector T cells in the lung, blunting Th1 responses and T cell production of IFN-γ.
The plasticity of macrophages and the heterogeneity of their responses to infections and inflammatory stimuli are increasingly appreciated. Classically activated (M1) macrophages, activated by IFN-γ and lipopolysaccharide (LPS), are potent antigen-presenting cells (APCs) and secrete proinflammatory cytokines, while alternatively activated (M2) macrophages, induced by IL-4, are phagocytes that eliminate cellular debris and secrete anti-inflammatory cytokines, such as IL-10 (25–27). However, this model is based on a stereotypic in vitro system that does not encompass the complexity of host-pathogen relationships in a chronic infection, such as tuberculosis. While M. tuberculosis induces macrophage markers of M2 polarization, such as IL-10 and arginase 1 (28), it does so in an environment largely devoid of IL-4 (29, 30). Furthermore, despite the association of host resistance with IFN-γ-secreting T cells (31–33), M. tuberculosis downregulates typical IFN-γ-induced M1 polarization markers, such as MHC-II antigen-processing and presentation molecules (17–21).
M. tuberculosis engages numerous receptors on macrophages, including Toll-like receptor 2 (TLR2), TLR9, and C-type lectin receptors (34), resulting in the activation of multiple signaling pathways. TLR signaling leads to activation of the IκB kinase complex, which then triggers NF-κB liberation from cytoplasmic sequestration, nuclear translocation of NF-κB, and transcription of proinflammatory genes (35). In parallel, IκB kinase complex activation leads to degradation of p105 and liberation of active Tpl2, a mitogen-activated protein kinase kinase kinase (MAP3K) (36), which in turn specifically activates the MAPK/extracellular signal-regulated kinase (ERK) kinase (MEK), which phosphorylates and activates ERK1 and ERK2 (37). ERK phosphorylates numerous downstream targets, including transcription factors, other protein kinases, and other proteins (38, 39). Of particular importance to the macrophage immune response, ERK can activate the AP-1, Sp1, and C/EBPβ/δ transcription factors, leading to increased transcription of IL-10 and arginase 1 (28, 40–42) and transcriptional repression of the class II transactivator (CIITA) (19, 20). In addition, IL-10 is known to negatively cross-regulate IL-12 expression, and vice versa (43–45); similarly, arginase 1 dampens microbicidal nitric oxide (NO) produced by inducible nitric oxide synthase (iNOS), the Nos2 gene product, as both arginase 1 and iNOS utilize arginine for their biochemical reactions (46). In the setting of M. tuberculosis infection, we propose that TLR2 signaling may establish a balance of opposing downstream pathways that include NF-κB activation and proinflammatory cytokine production versus ERK activation, IL-10 production, and MHC-II downregulation. The outcome of these opposing mechanisms may determine clearance versus persistence, or containment versus progression, of M. tuberculosis infection. The role of TLR2 in the setting of M. tuberculosis infection may be to limit immunopathology and facilitate host survival during chronic infection, albeit permitting prolonged M. tuberculosis survival as a consequence (34).
In this paper, we dissect a specific set of signaling pathways utilized by M. tuberculosis to regulate macrophage functions, including mechanisms by which macrophages influence T cell responses. We model these aspects of M. tuberculosis infection in vitro using the attenuated strain H37Ra, since other studies have revealed similar expression of lipoprotein TLR2 agonists by H37Ra and the virulent M. tuberculosis strain H37Rv (47), and other studies of Tpl2-ERK roles with H37Rv (48) and our current studies with H37Ra together indicate that these two strains similarly induce Tpl2-ERK signaling and similarly regulate IL-10 in macrophages, validating the use of H37Ra to study these mechanisms in macrophages in vitro. Our data further demonstrate that M. tuberculosis infection of macrophages activates the ERK pathway in a TLR2-dependent manner. In addition, inhibition of ERK phosphorylation (using the highly specific MEK inhibitor U0126 [49] or genetic deletion of Tpl2) can prevent IL-10 induction or MHC-II downregulation by M. tuberculosis. In contrast, ERK inhibition enhances IL-12 secretion by infected macrophages. Interestingly, Tlr2−/− macrophages had decreased expression of both IL-10 and IL-12. This finding demonstrates that TLR2 has the capability to drive both IL-10 and IL-12 expression through different downstream pathways and that the ERK pathway regulates the balance of these cytokines (inducing IL-10 and suppressing IL-12). Furthermore, we present novel results demonstrating that ERK pathway-deficient Tpl2−/− macrophages function as better Th1-polarizing APCs, inducing more IFN-γ production by M. tuberculosis-specific CD4+ T cells. Thus, we demonstrate that M. tuberculosis regulates macrophages and shapes the CD4+ T cell response through macrophage-intrinsic TLR2 and ERK signaling.
MATERIALS AND METHODS
Murine cells.All experiments using animals were reviewed and approved by the Institutional Animal Care and Use Committee of Case Western Reserve University. Wild-type C57BL/6J mice were obtained from Jackson Laboratories (Bar Harbor, ME). Tlr2−/− and Myd88−/− mice were provided by Shizuo Akira (Osaka University, Osaka, Japan). P25 T cell receptor (TCR)-transgenic mice, expressing a TCR specific for the peptide comprised of residues 240 to 254 from M. tuberculosis antigen 85B presented by I-Ab, were a gift from Kiyoshi Takatsu (University of Tokyo, Tokyo, Japan). Tpl2−/− (also known as Map3k8−/−) mice were generated as described previously (37). All knockout mouse strains were backcrossed onto the C57BL/6 genetic background.
Throughout this paper, “macrophages” refers to murine bone marrow-derived macrophages unless murine lung macrophages or human macrophages are specified. Cells were cultured in a humidified incubator at 37°C with 5% CO2 in Dulbecco's modified Eagle's medium (HyClone, Logan, UT) containing 10% heat-inactivated fetal bovine serum (Gibco, Carlsbad, CA), 50 μM 2-mercaptoethanol (Bio-Rad, Hercules, CA), 1 mM sodium pyruvate (HyClone), 10 mM HEPES (HyClone), 100 units/ml penicillin, 100 μg/ml streptomycin (HyClone) (complete medium, referred to here as D10F). To obtain macrophages, bone marrow was harvested from mouse femurs and tibias, and red blood cells were lysed using ACK lysing buffer (Lonza, Walkersville, MD). The cells were cultured in D10F with 25% LADMAC (50) cell-conditioned medium (changed once on day 5). Macrophages were used on day 7. Macrophages were incubated with M. tuberculosis in D10F lacking penicillin and streptomycin.
To obtain lung macrophages, mice were euthanized with tribromoethanol, exsanguinated by severing the left renal artery, and perfused with cold phosphate-buffered saline (PBS). Lung tissue was isolated into RPMI 1640 medium, homogenized using a GentleMACS Dissociator (Miltenyi Biotec, Auburn, CA), and digested using 5,000 U collagenase and 30 U benzonase (Sigma, St. Louis, MO). Lung cells were filtered through a 70-μm screen, and lung tissue-resident and alveolar macrophages were purified by magnetic affinity sorting (CD11b and CD11c MicroBeads; Miltenyi Biotec) using a dual-bead incubation protocol modified from the manufacturer's instructions, replacing an amount of buffer with an equal amount of additional beads to keep the recommended antibody-conjugated bead dilutions. Lung macrophages were cultured in D10F for M. tuberculosis infections.
To obtain T cells, splenocytes were harvested from P25 TCR-transgenic mice by homogenizing spleens in ACK lysing buffer, and naive splenic CD4+ T lymphocytes were isolated using a magnetic affinity bead selection kit (CD4+ CD62L+ T cell Isolation Kit II; Miltenyi Biotec). Primed Th1 effector P25 T cells were generated by anti-CD3/anti-CD28 stimulation of 106 naive P25 T cells per ml in RPMI 1640 medium supplemented with 5% heat-inactivated fetal bovine serum, 100 units/ml penicillin, 100 μg/ml streptomycin, and 50 μM 2-mercaptoethanol. Prior to addition of T cells, the plates were incubated with 3 μg/ml anti-CD3ε (clone 145-2C11) for 3 h at 37°C in PBS and then washed 3 times in PBS. P25 T cells were then added in the presence of 3 μg/ml anti-CD28 (clone 37.51), 10 μg/ml anti-IL-4 (clone 11B11), 5 ng/ml recombinant mouse IL-2, and 10 ng/ml recombinant mouse IL-12 (all from Biolegend, San Diego, CA). The T cells were cultured for 2 days in 1 ml of medium. An additional 2 ml of medium was then added, and the culture was continued for 2 days. Th1 effector cells were harvested and used on day 4 (approximately 50% of the Th1 effector cells produced by this method expressed IFN-γ by intracellular flow cytometry [data not shown]).
Reagents.Where used, macrophages were pretreated for 1 h with U0126, SB203580, or SP600125 (all used at 10 μM final concentration; Calbiochem, Billerica, MA) dissolved in dimethyl sulfoxide (DMSO) (Sigma) and then cultured in the continuous presence of inhibitor or were treated similarly with an equivalent volume of DMSO as a control. In certain experiments, macrophages were treated with IFN-γ or IL-4 (R&D Systems, Minneapolis, MN), recombinant M. tuberculosis lipoprotein LprG (expressed in Mycobacterium smegmatis and prepared as described previously [51]), ultrapure LPS from Escherichia coli O111:B4, or the synthetic lipopeptide Pam3CSK4 (Invivogen, San Diego, CA).
Human cells.THP-1 monocyte-like cells were obtained from the ATCC (Manassas, VA), cultured in RPMI 1640 medium supplemented with 50 μM 2-mercaptoethanol and 10% heat-inactivated fetal bovine serum, and passaged every 2 to 3 days at a density of 5 × 105 to 8 × 105 cells/ml. To differentiate THP-1 cells into macrophages, THP-1 cells (2 × 106 per condition) were treated for 2 days with 20 ng/ml phorbol myristate acetate (Sigma; 1 mg/ml stock in DMSO) in medium with a 2% final concentration of DMSO. The Institutional Review Board of University Hospitals Case Medical Center approved all studies using primary human cells. Heparinized whole blood was collected from healthy volunteers, and peripheral blood mononuclear cells were prepared by Ficoll density gradient centrifugation. Monocytes were enriched from the peripheral blood mononuclear cells by CD14 positive selection (Miltenyi Biotec). Monocytes (106 per condition) were cultured and differentiated into macrophages for 7 days in RPMI 1640 medium supplemented with l-glutamine and 10% human male blood type AB serum (Sigma).
Bacteria.M. tuberculosis H37Ra (ATCC 25177) was grown to mid-log phase in Middlebrook 7H9 broth supplemented with albumin-dextrose-catalase (ADC) enrichment (BBL, Sparks, MD), pelleted for 15 min at 3,000 × g, and resuspended in medium containing 25% glycerol. Single-bacterium suspensions were prepared by passing the bacteria through syringe-fitted needles of progressively increasing gauges, ending with 10 passages through a 26-gauge needle. This bacterial suspension was centrifuged for 2 min at 200 × g, and aliquots of bacteria in the supernatant were snap-frozen in dry ice-ethanol and stored at −80°C. Aliquots were thawed, passed again 10 times through a 26-gauge needle, serially diluted, and cultured for 2 weeks on Middlebrook 7H9-ADC agar to enumerate CFU/ml. This value was used to calculate the multiplicity of infection (MOI) prior to infecting macrophages, and all infections were done from the same lot.
Western blots.Antibodies for Western blotting were obtained from Cell Signaling Technology (Danvers, MA) and included anti-phospho-ERK1/2 (clone D13.14.4E), anti-ERK1/2 (clone 137F5), horseradish peroxidase-linked goat anti-rabbit IgG (catalog number 7074), and horseradish peroxidase-linked horse anti-mouse IgG (catalog number 7076). Anti-GAPDH (glyceraldehyde-3-phosphate dehydrogenase) (clone GA1R) and anti-IκBα (clone T.937.7) antibodies were from Thermo Scientific/Pierce Biotechnology (Rockford, IL). Macrophages (106 per condition) were plated overnight and then treated with M. tuberculosis or TLR agonists, washed in cold PBS, and lysed using RIPA buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) supplemented with a protease and protein phosphatase inhibitor cocktail (Pierce). The lysates were quantified using a bicinchoninic acid (BCA) protein assay (Pierce); samples containing equal amounts of total protein were boiled in reducing sample buffer (Bio-Rad), loaded on SDS-PAGE gels (Bio-Rad), transferred to polyvinylidene difluoride membranes (Millipore, Billerica, MA), and blocked in 5% nonfat dry milk in PBS-0.1% Tween 20. Primary antibodies were added overnight at 4°C; the membranes were then washed in PBS-0.1% Tween 20, treated with secondary antibodies in blocking buffer, washed again, and treated with enhanced-chemiluminescence reagent (Pierce). The membranes were exposed to autoradiography film (Amersham GE Healthcare, Pittsburgh, PA), and the film was developed using an automated film processor (Konica Minolta, Wayne, NJ). Images of bands were scanned and processed minimally—cropped and aligned but otherwise unaltered—using Photoshop software (Adobe Systems, San Jose, CA).
Flow cytometry.Macrophages (2 × 106 per condition) were replated on day 7 of culture, infected with M. tuberculosis at an MOI of 3 for 4 h, washed with fresh medium, and cultured for 48 h in D10F medium supplemented with 2 ng/ml IFN-γ (R&D Systems). The macrophages were detached and collected in PBS with 10 mM EDTA, incubated with 5 μg/ml Fc block (anti-CD16/CD32 [clone 2.4G2]; BD Biosciences, San Jose, CA) for 15 min on ice, stained for 30 min on ice with 5 μg/ml Alexa Fluor 647-conjugated anti-CD11b (clone M1/70; Biolegend) and 2 μg/ml phycoerythrin-conjugated anti-I-A/I-E (clone M5/114.15.2; BD Biosciences), washed twice in cold PBS, and then stained for 20 min on ice with LIVE/DEAD Green cell viability reagent (Molecular Probes, Eugene, OR). The cells were washed in PBS, fixed in 1% formaldehyde (Pierce), and analyzed on an Accuri C6 flow cytometer (BD Biosciences). Compensation was calculated using singly stained compensation control beads (BD Biosciences and Molecular Probes). Macrophages were identified using forward and side scatter properties and CD11b positivity, and live cells were identified based on dim LIVE/DEAD Green fluorescence. MHC-II expression was analyzed relative to nonspecific background staining with phycoerythrin-conjugated rat IgG2b, κ isotype (clone A95-1; BD Biosciences).
Antigen presentation assays and macrophage-CD4+ T lymphocyte coculture.Macrophages (2 × 106 per condition) were replated on day 7 of culture, infected with M. tuberculosis at an MOI of 3 for 4 h, and washed three times in D10F at 37°C. P25 CD4+ T cells were added to the infected macrophages in a 1:1 ratio for 72 h. To recover activated T cells, nonadherent cells were harvested, and the macrophages were depleted using CD11b positive selection by magnetic affinity sorting (Miltenyi Biotec). The T cells were processed for RNA extraction. In parallel experiments, macrophages and P25 T cells (105 of each cell type per condition, naive or Th1 effector as indicated) were incubated with M. tuberculosis or synthetic P25 peptide (NH2-FQDAYNAAGGHNAVF-COOH; Bio-Synthesis, Lewisville, TX) for 48 or 72 h, respectively.
Enzyme-linked immunosorbent assay (ELISA).Macrophages (105 per condition) were infected with M. tuberculosis for 0 to 72 h, and then the plates were centrifuged at 1,500 × g. Cell-free supernatants were collected and frozen at −80°C for later analysis. Macrophage-T cell cocultures were treated as indicated above and centrifuged at 1,500 × g, and the supernatants were frozen at −80°C. Kits were obtained from R&D Systems, and assays were performed according to the manufacturer's instructions. ELISA plates were read using a model 680 microplate reader (Bio-Rad). Cytokine concentrations were determined from seven-point standard curves from the same plates.
Quantitative reverse transcriptase PCR (qRT-PCR).Macrophages (106 per condition) were plated overnight, treated with U0126, and infected with M. tuberculosis for the indicated times. RNA was prepared from these macrophage cultures or from T cells isolated from T cell-macrophage cocultures using the RNeasy Plus minikit according to the manufacturer's instructions (Qiagen, Valencia, CA). Total RNA was reverse transcribed into cDNA using the QuantiTect reverse transcription kit (Qiagen). Equal quantities of cDNA for each experimental condition were amplified by real-time PCR using iQ SYBR green Supermix (Bio-Rad). The primers are described in Table S1 in the supplemental material. Samples were amplified using a hot start at 95°C for 3 min, followed by 50 cycles of 10 s at 95°C, 10 s at 59°C, and 30 s at 72°C and a postamplification melting curve ramping from 65°C to 95°C in increments of 0.5°C per 5 s. The abundances of each transcript were calculated relative to GAPDH gene expression in each sample, using the formula 2−[CT(target gene) − CT(GAPDH gene)] (52), normalizing the cycle threshold (CT) for each sample using the CT for the GAPDH gene amplified in parallel.
Statistical tests.Data were analyzed using Prism 5 (GraphPad, La Jolla, CA). Two-way analysis of variance (ANOVA) was used to determine significance among multiple groups and over time courses, and an unpaired Student t test was used to assess significance for individual comparisons.
RESULTS
M. tuberculosis induces TLR2-dependent ERK signaling in macrophages, with a rapid peak followed by a lower, sustained phase.M. tuberculosis is recognized by numerous receptors on macrophages and triggers many downstream signaling cascades. The roles of individual receptors and specific downstream signaling pathways in innate immune responses to M. tuberculosis remain unclear. We sought to explore the roles of TLR2 and certain signaling pathways downstream of TLR2 in regulating macrophages during M. tuberculosis infection. Phosphorylation of ERK (both ERK1 and ERK2) and induction of NF-κB signaling (assessed by IκBα degradation) were prominent at 30 min of M. tuberculosis infection and depended on TLR2 expression (Fig. 1A). ERK phosphorylation and IκBα degradation were also induced by M. tuberculosis lipoprotein LprG (TLR2 dependent) and LPS (TLR2 independent), although M. tuberculosis drove stronger signaling responses (Fig. 1A). A kinetic series using purified M. tuberculosis lipoprotein LprG, a potent TLR2 agonist, revealed that TLR2-driven ERK activation peaked at 15 min and was sustained at an intermediate level to at least 4 h, while NF-κB signaling peaked at 15 min and returned to baseline by 45 min (Fig. 1B). Similar results were observed in macrophages infected with M. tuberculosis for various times, from minutes to several hours, which showed sustained ERK phosphorylation but rapid recovery of IκBα (see Fig. S1 in the supplemental material). There are two implications of these findings. First, our data demonstrate that ERK phosphorylation persists at low levels following an early peak, with persistent stimulation with a defined TLR2 agonist. Second, M. tuberculosis may provide such persistent TLR2 stimulation during infection in vivo, although that remains to be studied. Although M. tuberculosis signals through many receptors, these data establish TLR2 as the primary driver of both ERK and NF-κB signaling in M. tuberculosis-infected macrophages and, furthermore, demonstrate that ERK signaling is a component of the macrophage response to M. tuberculosis infection at later time points.
M. tuberculosis signaling through TLR2 produces rapid and strong activation of both ERK and NF-κB that is followed by sustained lower-level activation of ERK. (A) Macrophages were infected with M. tuberculosis (Mtb) (MOI = 3) or treated with purified TLR ligands for 30 min. The cells were then lysed and analyzed by Western blotting for ERK phosphorylation and IκBα degradation (a readout of NF-κB activation). Recombinant M. tuberculosis LprG (30 nM) was used as a pure TLR2 stimulus; E. coli LPS (10 ng/ml) is a TLR4 agonist used as a control. p-ERK, phosphorylated ERK. (B) A series of LprG treatments for the indicated times revealed that TLR2 triggered maximal ERK and NF-κB signaling by 15 min. ERK signaling persisted at an intermediate level to at least 4 h, whereas NF-κB signaling declined rapidly, as demonstrated by normalization of IκBα expression. The results are representative of those of three independent experiments.
Sustained TLR2-dependent ERK signaling by M. tuberculosis provides prolonged expression of IL-10 and regulates other genes in macrophages.Since M. tuberculosis may persist in macrophages for long periods, it is important to consider signaling driven by M. tuberculosis at both early and late time points. The sustained ERK signaling that we observed in macrophages (Fig. 1) may influence macrophage gene expression over an extended period of M. tuberculosis infection. This hypothesis was tested by genetic and pharmacologic interventions to assess the roles of components of the TLR2-MyD88-ERK signaling pathway in regulation of selected genes during infection of macrophages with M. tuberculosis for 2 to 24 h. We focused on expression of Il10 and Il12b (encoding the p40 subunit of IL-12); these genes encode cytokines that are generally anti-inflammatory or proinflammatory, respectively. They are both expressed by M. tuberculosis-infected macrophages and are known to significantly influence the course of M. tuberculosis infection (43, 53, 54). Moreover, differential expression of Il10, Il12b, and other genes targeted in our studies (e.g., Arg1 and Nos2) may characterize distinct states of macrophage inflammatory function (27, 28, 55).
M. tuberculosis induced two phases of Il10 mRNA expression, a high peak within 2 h and a sustained lower level that was significantly above baseline and persisted for at least 24 h (Fig. 2A and C). Il10 mRNA expression at both early and late time points was dependent on TLR2 and MyD88 (Fig. 2A) and was inhibited by U0126, an inhibitor of MEK that prevents ERK activation (49) (Fig. 2B and C), indicating dependence on the ERK pathway downstream of TLR2. In addition, Il10 mRNA expression was diminished by genetic ablation of Tpl2, which mediates TLR-dependent ERK activation (Fig. 2D). Thus, M. tuberculosis drove Il10 expression at early and late time points following infection of macrophages, and TLR2-MyD88-ERK signaling was necessary for both phases of Il10 expression.
M. tuberculosis induces TLR2- and ERK pathway-dependent gene regulation that appears within 2 h and persists for at least 24 h. Macrophages were pretreated for 1 h with U0126 (10 μM) or DMSO vehicle control, as indicated, and then incubated with or without M. tuberculosis (MOI = 3) for 24 h or the indicated times. RNA was prepared, and gene expression was analyzed by qRT-PCR in triplicate. (A to D) Regulation of Il10. (A and C) Macrophages expressed two phases of Il10, an early peak within 2 h of M. tuberculosis infection and a late elevation at 24 h. Il10 expression in both phases was inhibited by deletion of Tlr2 (A and B), Myd88 (A), or Tpl2 (D) or inhibition of ERK signaling with U0126 (B and C). (E to H) Regulation of Il12b. Expression of Il12b was induced maximally by 8 h of M. tuberculosis infection but declined to near baseline by 24 h. Il12b expression was inhibited by deletion of Tlr2 (E and F) or Myd88 (E). In contrast to Il10, Il12b expression was enhanced when the ERK pathway was blocked by inhibition with U0126 (F and G) or deletion of Tpl2 (H). The data are shown as means ± standard errors of the mean (SEM) and are representative of the results of at least two independent experiments. Statistical significance was determined as described in Materials and Methods. White bars, wild type; grey bars, U0126 treated; black bars, Tlr2−/−; hatched bars, Tpl2−/−. *, wild-type versus Tlr2−/−; +, wild-type versus Myd88−/−; #, DMSO versus U0126; 3 symbols, P < 0.001; 2 symbols, P < 0.01; 1 symbol, P < 0.05.
In contrast, Il12b mRNA expression was induced more slowly, with a peak at approximately 8 h of M. tuberculosis infection (Fig. 2E). Il12b mRNA expression was reduced in Tlr2−/− macrophages and more completely abrogated in Myd88−/− macrophages (Fig. 2E), indicating that TLR2 was a significant contributor to Il12b mRNA induction, along with other MyD88-dependent receptors. U0126-treated wild-type macrophages expressed more Il12b than untreated macrophages (Fig. 2F and G), indicating that TLR2 activation of ERK leads to suppression of maximal Il12b expression. In addition, Il12b mRNA expression was enhanced in Tpl2−/− macrophages (Fig. 2H), confirming negative regulation of Il12b by the ERK pathway, as opposed to the positive regulatory role of the pathway for Il10.
To investigate the specificity of the role of ERK pathway signaling in regulation of IL-10 and IL-12 expression, we employed inhibitors of other MAPKs, p38 MAPK (inhibited by SB203580) and JNK (inhibited by SP600125); these inhibitors have been employed successfully in prior studies by our group on CIITA regulation in macrophages (20). Macrophages were treated with inhibitors or vehicle and infected with M. tuberculosis for 24 h. Cytokine protein expression was measured by ELISA. Inhibition of p38 suppressed IL-10 production, but inhibition of JNK did not suppress IL-10 (see Fig. S2A in the supplemental material). On the other hand, both p38 and JNK inhibitors failed to significantly affect IL-12 p70 secretion, while ERK inhibition resulted in strong induction of IL-12 p70 (see Fig. S2B in the supplemental material). We conclude that the balance of IL-10 and IL-12 expression in M. tuberculosis-infected macrophages is highly dependent on ERK, which both enhances IL-10 expression and suppresses IL-12 expression, whereas p38 only enhances IL-10 expression, and JNK does not regulate either of these cytokines. Since IL-12 is known to be important for the induction of Th1 responses against M. tuberculosis, these results suggested a potential role for macrophage-intrinsic ERK in suppressing protective Th1 responses, a hypothesis that is explored below. Additionally, our data using Tpl2−/− macrophages (Fig. 2), which should not have defects in p38 or JNK signaling (56), confirm a specific and critical role for the ERK pathway in the regulation of these cytokines.
We also sought to address the specificity of this pathway for TLR2 signaling. Many TLRs activate MyD88 and therefore potentially activate Tpl2 and ERK. We treated macrophages for 24 h with purified LPS and Pam3CSK4 to test TLR4 and TLR2 signaling, respectively, and measured IL-10 and IL-12 expression by ELISA. TLR4-induced IL-10 was only partially Tpl2 dependent (IL-10 production by Tpl2−/− macrophages was 42% of that by wild-type macrophages), whereas TLR2-induced IL-10 was dependent to a greater degree on Tpl2 (IL-10 production by Tpl2−/− macrophages was only 11% of that by wild-type macrophages) (see Fig. S3A in the supplemental material). Furthermore, LPS failed to induce IL-12 p70, in contrast to Pam3CSK4 (see Fig. S3B in the supplemental material). These data using Pam3CSK4 also indicate that the cytokine expression changes seen with M. tuberculosis infection are recapitulated by a TLR2 agonist, indicating that they do not require other mechanisms that might involve active bacterial processes dependent on bacterial viability. Furthermore, the data we show for Tpl2 dependence of IL-10 induction by M. tuberculosis are consistent with a role for TLR2, since induction of IL-10 by both M. tuberculosis and TLR2 agonists was strongly Tpl2 dependent, whereas induction of IL-10 by TLR4 agonists was less Tpl2 dependent. In addition, M. tuberculosis does not express known TLR4 agonists that have been specifically identified, although TLR4 activity in M. tuberculosis-derived preparations has been reported (57). We conclude that the balance of these cytokines induced by M. tuberculosis infection depends on TLR2 signaling.
To establish the relevance of these results to pulmonary immune responses and extend the mechanisms in a human system, we performed similar studies with murine lung macrophages, human monocyte-derived macrophages, and human macrophage-like THP-1 cells. Il10 expression was induced in murine lung macrophages infected with M. tuberculosis for 24 h and was inhibited by U0126 (Fig. 3A), confirming a role for ERK signaling in Il10 induction. Lung macrophages did not significantly upregulate Il12b at 24 h of M. tuberculosis infection, but U0126-treated, M. tuberculosis-infected lung macrophages did express increased levels of Il12b (Fig. 3B). M. tuberculosis infection of human THP-1 cells for 24 h induced both IL10 and IL12B expression; IL10 expression was suppressed, and IL12B expression was enhanced by U0126 (Fig. 3C and D). IL10 expression followed a similar regulatory pattern in primary human monocyte-derived macrophages; M. tuberculosis infection induced IL10 expression, and U0126 treatment suppressed IL10 (Fig. 3E). IL12B was not detected in monocyte-derived macrophages; prior work had established that these cells require IFN-γ activation to express IL12B (58). These results confirm that M. tuberculosis drives ERK signaling to induce IL-10 expression and inhibit IL-12 expression in murine pulmonary macrophages and also corroborate similar mechanisms in human macrophages. Further work is needed to determine the relationship of our in vitro studies of cytokine expression in pulmonary macrophages to in vivo pulmonary immune responses to M. tuberculosis.
M. tuberculosis regulates cytokine gene transcription in an ERK-dependent manner in murine lung macrophages and human macrophages. Lung macrophages from wild-type mice (A and B), human THP-1 cells (C and D), or human monocyte-derived macrophages (E) were infected with M. tuberculosis (MOI = 3) for 24 h in the continuous presence of U0126 or vehicle control. Total RNA was prepared and analyzed in triplicate by qRT-PCR. The data represent the results of two independent experiments and are graphed as means ± SEM. White bars, DMSO treated; grey bars, U0126 treated. ***, P < 0.001; **, P < 0.01; NS, P > 0.05.
M. tuberculosis regulates levels of IL-10 and IL-12 secreted by infected macrophages via TLR2 and ERK signaling.We next explored the role of ERK signaling in regulating the production of functional IL-10 and IL-12 by M. tuberculosis-infected macrophages. Wild-type, U0126-treated, Tpl2−/−, and Tlr2−/− macrophages were infected with M. tuberculosis, and the concentrations of cytokines were measured in the supernatants by ELISA. Wild-type macrophages secreted IL-10 in large amounts in response to M. tuberculosis infection (Fig. 4A) but secreted no detectable amounts of IL-12 p70 and minimal IL-12 p40 (Fig. 4B and C). In contrast, when ERK signaling was blocked by U0126 treatment or Tpl2 deletion, M. tuberculosis-infected macrophages secreted less IL-10 (Fig. 4A), and only under these conditions did macrophages produce detectable levels of the functional Th1-polarizing cytokine IL-12 p70 (IL-12 p40 was also increased under these conditions) (Fig. 4B and C). Tlr2−/− macrophages did not produce IL-10 or IL-12, further demonstrating the requirement for TLR2 to induce both of the cytokines (Fig. 4A to C). In summary, while TLR2 signaling in M. tuberculosis-infected macrophages has the potential to induce both IL-10 and IL-12, downstream ERK signaling affects the balance of the cytokines, inducing IL-10 secretion and inhibiting secretion of the Th1-stimulatory cytokine IL-12 p70.
M. tuberculosis-induced ERK signaling promotes IL-10 production and inhibits IL-12 production. Macrophages were incubated with or without U0126 (as for Fig. 2) and infected with M. tuberculosis (MOI = 3) for the indicated times. The supernatants were harvested, and cytokine levels were determined by ELISA. The data shown are representative of the results of three independent experiments and reflect the means ± SEM of biological triplicates handled in parallel throughout the experiment. Statistical comparisons were wild type to Tpl2−/− (*), wild type to Tlr2−/− (+), and wild type untreated to wild type U0126 treated (#); 3 symbols, P < 0.001; 2 symbols, P < 0.01; NS, P > 0.05.
M. tuberculosis-infected macrophages express anti-inflammatory markers in a pattern that overlaps with yet is distinct from the M2 macrophage polarization state.As differential expression of IL-10 and IL-12 has been associated with polarization of macrophages along the M1-M2 spectrum (25–27), we sought to determine whether M. tuberculosis infection and concomitant upregulation of IL-10 and suppression of IL-12 are accompanied by other hallmarks of M2 differentiation. We studied the expression of various M2-associated genes by qRT-PCR. Wild-type, Tlr2−/−, Myd88−/−, Tpl2−/−, and U0126-treated wild-type macrophages were infected with M. tuberculosis as described in the legend to Fig. 2. Arg1 expression was strongly suppressed by deletion of Tlr2 or Myd88 or inhibition of ERK by U0126 (Fig. 5A to C), indicating a requirement for this pathway to induce Arg1 (deletion of Tpl2 showed significant but partial suppression of Arg1 expression) (Fig. 5E). In contrast, Nos2 was upregulated in Tpl2−/− or U0126-treated macrophages but was unchanged in Tlr2−/− macrophages infected with M. tuberculosis, indicating ERK-dependent suppression of Nos2 expression (Fig. 5D and F). The arginase 1 and iNOS enzymes compete for the substrate arginine (46), and therefore, the condition of high Arg1 and low Nos2 expression in wild-type, M. tuberculosis-infected macrophages would be predicted to diminish microbicidal NO production. While the pattern of Arg1, Nos2, IL-10, and IL-12 regulation by M. tuberculosis was consistent with M2 differentiation, other M2-associated genes were regulated by M. tuberculosis in a manner that did not fit typical M2 differentiation. We did not observe upregulation of Mrc1, Retnla, or Chi3l3 during M. tuberculosis infection, but all three were induced by IL-4 treatment of macrophages (see Fig. S4 in the supplemental material). We conclude that the stereotypic M2 gene regulation program results from IL-4 signaling and that M. tuberculosis-induced TLR2-ERK signaling produces an overlapping gene expression pattern but does not fully recapitulate the established M2 gene regulation pattern. Thus, M. tuberculosis-triggered TLR2-ERK signaling produces an anti-inflammatory macrophage state that is different from M2 polarization.
M. tuberculosis-induced TLR2-Tpl2-ERK signaling regulates the balance of macrophage proinflammatory/anti-inflammatory functions, such as Nos2 and arginase 1 expression. Macrophages were treated with U0126 or control treatment, infected with M. tuberculosis for the indicated times (A and B) or for 24 h (C to F), and analyzed for expression of Arg1 and Nos2. (A) Induction of mRNA for markers of anti-inflammatory macrophage states (e.g., Arg1) was dependent on TLR2 and MyD88. Induction of Arg1 was diminished by U0126 or genetic deletion of Tpl2 (B, C, and E), whereas expression of Nos2 (associated with proinflammatory states) was increased by these conditions (D and F). The data are represented as means ± SEM and are representative of at least two independent experiments. Statistical significance was determined as described in Materials and Methods. White bars, wild type; grey bars, U0126 treated; black bars, Tlr2−/−; hatched bars, Tpl2−/−. *, wild type versus Tlr2−/−; +, wild type versus Myd88−/−; #, DMSO versus U0126; 3 symbols, P < 0.001; 2 symbols, P < 0.01; 1 symbol, P < 0.05; NS, P > 0.05.
M. tuberculosis regulates MHC-II expression and antigen presentation in a TLR2-Tpl2-ERK-dependent manner.Along with cytokine production and intrinsic microbicidal functions, a key function of macrophages is MHC-II antigen processing and presentation to effector T cells. It has been appreciated that the lipoprotein agonists of TLR2 produced by M. tuberculosis inhibit expression of MHC-II and some other IFN-γ-stimulated genes in macrophages (20, 59). We sought to explore the roles of Tpl2 and ERK signaling in regulation of MHC-II-associated genes in M. tuberculosis-infected macrophages. The master transcriptional regulator CIITA controls the expression of these genes, including those for MHC-II molecules and H2-DM. As expected, M. tuberculosis downregulated the expression of IFN-γ-stimulated Ciita and its target gene, H2-DMb, following 24 h of infection (Fig. 6A and B). Furthermore, deletion of Tlr2 in macrophages, inhibition of ERK using U0126, or deletion of Tpl2 reduced M. tuberculosis-induced downregulation of Ciita and H2-DMb (Fig. 6A to D). U0126 also induced higher expression of H2-Ab1 (I-Ab β chain) in M. tuberculosis-infected macrophages (data not shown). A lesser degree of M. tuberculosis-induced Ciita and H2-DMb downregulation persisted with deletion of Tlr2 in macrophages, inhibition of ERK using U0126, or deletion of Tpl2, perhaps reflecting some contribution of receptors other than TLR2 (e.g., TLR9) and signaling branches other than ERK (e.g., p38 [20]) to the inhibition of MHC-II-associated genes in macrophages. These data indicate that M. tuberculosis inhibits CIITA and downstream target genes through TLR2 and ERK signaling, confirming and extending our earlier results (20).
M. tuberculosis inhibits MHC-II antigen presentation through TLR2-Tpl2-ERK signaling. (A to D) Macrophages were infected with M. tuberculosis for 4 h and washed. IFN-γ (2 ng/ml) was added for an additional 20 h, and RNA was prepared for qRT-PCR analysis. (E) Macrophages were infected for 4 h to allow processing of M. tuberculosis antigen, and naive P25 T cells were then added (1:1 ratio) for 72 h. The T cells were recovered, and Il2 mRNA expression was analyzed by qRT-PCR. (F) Macrophages were infected with M. tuberculosis and cultured with naive P25 T cells for 48 h, and then the supernatants were collected for ELISA measurement of IL-2. (G) Macrophages were cocultured with naive P25 CD4+ T cells and synthetic cognate peptide for 72 h, and the supernatants were analyzed by ELISA for IL-2 secretion. (H) Macrophages were infected with M. tuberculosis for 4 h, washed, treated with IFN-γ for an additional 44 h, stained, and analyzed by flow cytometry to measure the specific median fluorescence intensity (MFI) of MHC-II. The data are means ± SEM and represent the results of at least three independent experiments. White bars, wild type; grey bars, U0126 treated; black bars, Tlr2−/−; hatched bars, Tpl2−/−. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; NS, P > 0.05.
To assess the impact of TLR2-ERK-dependent regulation of macrophage antigen presentation on M. tuberculosis-specific T cell responses, we compared Il2 gene expression by M. tuberculosis-specific primary P25 T cells after 3 days of activation by M. tuberculosis-infected macrophages from wild-type, Tpl2−/−, and Tlr2−/− mice. After M. tuberculosis infection, Tpl2- or TLR2-deficient macrophages stimulated more Il2 expression by P25 T cells (Fig. 6E), consistent with the refractoriness of these macrophages to inhibition of MHC-II genes by M. tuberculosis. The IL-2 protein level measured by ELISA was also significantly increased following stimulation of naive P25 T cells for 2 days by Tpl2−/− or Tlr2−/− macrophages (Fig. 6F); previously primed Th1 effector P25 T cells did not express IL-2 to a high degree and could not be restimulated by M. tuberculosis-infected macrophages to secrete IL-2 (data not shown). Uninfected macrophages from all three strains of mice were able to present synthetic antigen peptide to stimulate equal IL-2 responses by P25 T cells, indicating no intrinsic differences in baseline antigen presentation (Fig. 6G). Finally, MHC-II expression by M. tuberculosis-infected macrophages was assessed by flow cytometry. Tpl2−/− and Tlr2−/− macrophages were less susceptible to M. tuberculosis-driven MHC-II downregulation than wild-type macrophages following 48 h of infection (Fig. 6H). This difference was more pronounced with Tlr2−/− macrophages, suggesting that TLR2 inhibition of MHC-II expression is partially dependent on Tpl2. These data indicate that Tpl2 and TLR2 play important roles in M. tuberculosis-induced inhibition of MHC-II antigen presentation and activation of M. tuberculosis-specific T cell responses.
ERK pathway signaling within M. tuberculosis-infected macrophages suppresses Th1 responses during antigen-specific T cell activation.M. tuberculosis may also regulate T cell responses by mechanisms other than diminution of antigen presentation, especially since this mechanism is incomplete and delayed (substantial MHC-II inhibition occurs after 48 to 72 h of M. tuberculosis infection) (17, 59, 60). We propose that T cell responses to pulmonary M. tuberculosis infection may be influenced by other M. tuberculosis-induced immune evasion mechanisms. In particular, we noted that Tpl2-ERK signaling enhanced macrophage IL-10 expression and suppressed macrophage IL-12 p70 expression (Fig. 4). IL-12 p70 induces Th1 polarization, which is associated with protection in M. tuberculosis infection. Thus, we hypothesized that macrophage-intrinsic ERK signaling influences responding antigen-specific T cells by reducing Th1 polarization and associated Th1 functions, e.g., production of IFN-γ.
To test whether Th1 polarization and function are influenced by macrophage-intrinsic ERK signaling, we incubated primary M. tuberculosis-specific P25 CD4+ T cells with M. tuberculosis-infected macrophages for 3 days, separated the T cells from the macrophages, and isolated RNA from the T cells for qRT-PCR analyses. Relative to wild-type macrophages, Tpl2−/− macrophages promoted higher levels of Th1-associated gene expression, as measured by increased Tbx21 (T-bet transcription factor) and Ifng expression by responding T cells (Fig. 7A and B). Importantly, Tlr2−/− macrophages were not significantly different from wild-type macrophages in their ability to induce Tbx21 expression by responding T cells (Fig. 7A), despite the observed relief of MHC-II inhibition in Tlr2−/− macrophages (Fig. 6), indicating a special role of the Tpl2-ERK pathway among the various pathways downstream of TLR2 in macrophages for suppression of Tbx21 expression in responding T cells. Moreover, relative to wild-type macrophages, Tlr2−/− macrophages induced lower levels of T cell Ifng expression (Fig. 7B), in contrast to the increase in T cell Il2 expression induced by Tlr2−/− macrophages (Fig. 6E and F). Thus, despite higher MHC-II antigen presentation and T cell IL-2 induction by Tlr2−/− macrophages (Fig. 6), the macrophages were not more efficient at inducing Th1 polarization. Previously primed Th1 effector P25 T cells also exhibited increased IFN-γ expression when restimulated with M. tuberculosis-infected Tpl2−/− macrophages compared to wild-type or Tlr2−/− macrophages (Fig. 7C). This establishes that the role of the ERK pathway in suppressing Th1 responses is not limited to a general loss of T cell activation due to reduced antigen presentation but, rather, involves specific inhibition of Th1 responses relative to other CD4+ T cell responses (e.g., IL-2 production). Furthermore, deletion of Tpl2 in macrophages did not alter T cell expression of Il4, Il10, Il17a, or Il22 in a substantial fashion (Fig. 7D), indicating that macrophage ERK signaling results in inhibition of Th1 responses and not all T cell responses.
ERK signaling in macrophages inhibits Th1 polarization in responding T cells. Macrophages were infected with M. tuberculosis and cocultured with P25 T cells. The T cell responses were analyzed as for Fig. 6. (A and B) Expression of the Th1 markers T-bet and IFN-γ is increased following stimulation of naive P25 T cells for 72 h when antigen-presenting macrophages are Tpl2 deficient. (C) Tpl2-deficient macrophages induced higher levels of IFN-γ protein expression by previously primed Th1 effector P25 T cells following 48 h of restimulation. (D) The expression of other T cell-expressed cytokine genes, such as Il4, Il10, Il17a, or Il22, was not substantially altered by knockout of Tpl2 signaling in macrophages following 72 h stimulation of naive P25 T cells. The data represent the results of two independent experiments and show means ± SEM. White bars, wild type; black bars, Tlr2−/−; hatched bars, Tpl2−/−. **, P < 0.01; *, P < 0.05; NS, P > 0.05.
In summary, these data indicate that suppression of Th1 polarization and function is more specifically associated with macrophage ERK signaling than the broader set of TLR2-activated pathways in macrophages. Thus, ERK signaling in M. tuberculosis-infected macrophages influences responding CD4+ T cells to mount a less potent anti-M. tuberculosis Th1 response (Fig. 7). This may result, in part, from the role of the ERK pathway in suppressing IL-12 (and inducing IL-10), but other mechanisms may also contribute. While M. tuberculosis is known to induce Th1 responses, this model indicates that Tpl2-ERK signaling in M. tuberculosis-infected macrophages suppresses maximal Th1 polarization and IFN-γ production upon restimulation in responding CD4+ T cells, suggesting the possibility that Th1 responses could be made more effective by manipulating these mechanisms.
DISCUSSION
Our results reveal that ERK signaling in M. tuberculosis-infected macrophages inhibits Th1 polarization in responding M. tuberculosis-specific T cells; this observation has significant implications for understanding M. tuberculosis immune control. Infection of macrophages with M. tuberculosis reduced their ability to induce a Th1 phenotype, as measured by reduced induction of Tbx21 and Ifng expression by T cells (Fig. 7). Induction of Tbx21 and Ifng was enhanced in T cells stimulated by Tpl2−/− macrophages, which are deficient in ERK activation upon M. tuberculosis infection, indicating that the ERK signaling pathway in macrophages contributes significantly to M. tuberculosis-driven inhibition of Th1 responses. Importantly, Tlr2−/− macrophages demonstrated no difference in stimulating Th1 polarization, which has several implications for understanding the observations with Tpl2−/− macrophages. First, the increased induction of Th1 differentiation by Tpl2−/− macrophages is not simply due to relief of MHC-II suppression in Tpl2−/− macrophages, since this is more notable in Tlr2−/− macrophages (Fig. 6) that do not promote the Th1 phenotype. This implies that macrophage-intrinsic Tpl2 signaling produces a selective inhibition of the Th1 phenotype rather than a generalized inhibition of T cell responses; this conclusion is supported by the lack of a similar effect on cytokine markers of other T cell differentiation states (Fig. 7D). Second, we conclude that the pathway for inhibiting Th1 differentiation is linked more closely with Tpl2 than TLR2; although TLR2 is upstream of Tpl2, it is also upstream of other signaling branches that may oppose or modify the effects of Tpl2 on Th1 differentiation. Finally, we observe that Tpl2 in macrophages suppresses IFN-γ responses by Th1 effector cells, as well as naive T cells (Fig. 7). Th1 effector cells are present in the lung and interact with infected macrophages during M. tuberculosis infection, suggesting potential importance of this pathway to pulmonary pathogenesis of M. tuberculosis infection. In summary, our findings indicate that M. tuberculosis-induced TLR2-Tpl2-ERK signaling in macrophages blunts Th1 IFN-γ responses that are critical for controlling the infection, delineating an important role for the pathway in immune evasion by M. tuberculosis.
Our studies have addressed whether the role of TLR2 is to stimulate immune responses or to orient the resulting phenotype or differentiation of immune responses. Despite the existence of other receptors that recognize M. tuberculosis and may activate other signaling pathways, the activation of ERK and NF-κB by M. tuberculosis was dependent on TLR2 in the systems we studied (Fig. 1A). Thus, TLR2 plays an important role in triggering both of these innate immune signaling pathways, which are known to be important for macrophage functions and control of M. tuberculosis infections. We cannot exclude indirect effects, such as TLR2-dependent regulation of the expression or function of other receptors or signaling molecules, though such a situation would still indicate the importance of TLR2 in controlling these mechanisms. TLR2 activates multiple signaling pathways and downstream branches, and some of these pathways and branches may modify or oppose the outcomes of others. Thus, TLR2 induces a balance of sometimes opposing effector mechanisms, including the induction of both anti-inflammatory cytokines (e.g., IL-10; ERK dependent) and proinflammatory cytokines (e.g., IL-12; NF-κB dependent). While TLR2 provides the drive for these responses, they are steered by the balance of signaling pathways/branches downstream of TLR2, as exemplified by the role of Tpl2-ERK signaling in promoting IL-10 expression and inhibiting IL-12 expression in macrophages.
The higher induction of IL-12 in Tpl2−/− macrophages indicates a role for ERK pathway signaling in suppressing IL-12. This may be due to intrinsic ERK-dependent regulation of IL-12 genes or may result from ERK-dependent induction of IL-10 and IL-10-triggered negative feedback on IL-12 expression (61). Il12b is induced approximately 6-fold in U0126-treated or Tpl2−/− macrophages (Fig. 2F to H) but only 2-fold in Il10−/− macrophages (data not shown), suggesting that Tpl2 may induce both IL-10 and other mechanisms that further suppress IL-10 expression, possibly by direct ERK-dependent regulation of IL-12 genes.
Previously, it was observed that macrophages expressed low levels of IL-12 in response to M. tuberculosis, whereas dendritic cells (DCs) were capable of secreting large amounts of IL-12; this difference was attributed to intrinsic differences in receptor utilization by macrophages and DCs, with macrophages responding to M. tuberculosis primarily through TLR2 (62), consistent with our results. We found that only U0126-treated or Tpl2−/− macrophages secreted IL-12 p70 at detectable levels in response to M. tuberculosis (Fig. 4C), implying that the deficiency of IL-12 p70 production by macrophages (relative to DCs) is due to suppression of IL-12 p70 production by Tpl2-ERK signaling in M. tuberculosis-infected macrophages. These data provide further evidence that the ERK pathway promotes anti-inflammatory responses in macrophages and consequently diminishes Th1 responses induced by M. tuberculosis-infected macrophages.
We also report that similar regulation of IL-10 and IL-12 occurs in other relevant models of M. tuberculosis infection. Lung macrophages, the host cells harboring most M. tuberculosis bacteria during pulmonary infections, also upregulated Il10 and suppressed Il12b expression in an ERK-dependent manner. Human macrophages (THP-1 cells or primary monocyte-derived macrophages) also upregulated IL10 via an ERK-dependent mechanism, and THP-1 cells demonstrated blunted IL12B expression that was relieved by ERK inhibition. We note that Randhawa et al. have published results that show that humans expressing TLR1 and TLR6 alleles with defective signaling function produce less IL-10 and more IFN-γ in response to Mycobacterium bovis BCG vaccination, consistent with our model (63) (TLR1 and TLR6 heterodimerize with TLR2 to form signaling-competent receptors). Together, these data support the relevance of our findings to pulmonary and human macrophages, suggesting the possibility of their clinical application to host-directed immunotherapeutics for tuberculosis.
Other macrophage functions regulated by M. tuberculosis through ERK signaling include MHC-II antigen presentation and the balance of mechanisms that may promote or dampen host defense (such as iNOS or arginase 1, respectively). M. tuberculosis infection reduced expression of Ciita, H2-DMb, and MHC-II, and these effects were reversed in Tlr2−/− macrophages, Tpl2−/− macrophages, or macrophages treated with U0126 (Fig. 6). M. tuberculosis infection induced Arg1 expression, but this effect was reversed in Tlr2−/− macrophages, Tpl2−/− macrophages, or macrophages treated with U0126 (Fig. 5). M. tuberculosis infection induced only low levels of Nos2, but Nos2 induction was enhanced in Tpl2−/− or U0126-treated macrophages; Tlr2−/− macrophages displayed no difference in Nos2 induction (Fig. 5). These results indicate that Tpl2-ERK signaling regulates macrophages to produce a phenotype that is generally more proinflammatory. M. tuberculosis-infected macrophages did not express typical genes of M2 polarization other than Arg1, such as Mrc1, Retnla, or Chi3l3 (see Fig. S4 in the supplemental material), indicating that the regulatory state resulting from M. tuberculosis-triggered Tpl2-ERK signaling is distinct from the IL-4-induced M2 program.
This report offers additional clarity on the role of Tpl2 in M. tuberculosis infection and covers new ground that has not previously been studied by other groups, such as McNab et al., who also demonstrated Tpl2-dependent induction of IL-10 and suppression of IL-12 in macrophages by M. tuberculosis (48). Among the multiple receptors that recognize M. tuberculosis, we demonstrate that TLR2 is responsible for activation of Tpl2 and ERK in M. tuberculosis-infected macrophages (Fig. 1 and 2). Furthermore, the role of M. tuberculosis-induced ERK signaling in T cell responses has not been studied. The data presented here demonstrate that ERK strongly suppresses IL-12 p70 expression by M. tuberculosis-infected macrophages (Fig. 4). Consequently, M. tuberculosis-specific T cells responding to infected macrophages achieve less Th1 polarization and express less IFN-γ (Fig. 7). Prior research has not connected Tpl2-ERK signaling in macrophages with regulation of T cell responses. Indeed, one of the major issues in understanding the role of Tpl2 is that in vitro, Tpl2 deletion facilitates a more proinflammatory macrophage response (this work and reference 48), but in vivo, Tpl2 is protective against M. tuberculosis (48). These in vivo results may be confounded by a combination of different roles for Tpl2 in different cell types, making the results of systemic Tpl2 knockout difficult to interpret mechanistically. For example, Tpl2−/− T cells demonstrate poorer IL-12 responsiveness and IFN-γ production (64), yet our in vitro studies clearly demonstrate that IFN-γ production by Tpl2-expressing CD4+ T cells is enhanced when M. tuberculosis antigens are presented by Tpl2-deficient macrophages. Further research should elucidate the distinct roles of Tpl2 in various immune cell types, and especially the regulation of Tpl2-ERK and downstream target genes in alveolar macrophages and macrophages recruited to the sites of M. tuberculosis infection in the lung. In summary, our study provides novel information regarding macrophage-T cell interactions in M. tuberculosis infection and the regulation of cytokines expressed by macrophages that have important consequences for modulating T cell responses.
Our data collectively suggest a model of the M. tuberculosis-regulated macrophage (Fig. 8) in which TLR2 signaling activates multiple pathways and outcomes, including the NF-κB pathway (inducing microbicidal functions, such as NO production and expression of proinflammatory cytokines like IL-12) and the ERK pathway (via Tpl2, leading to anti-inflammatory IL-10 and arginase 1 synthesis). The ERK pathway may contribute to immune evasion and the persistence of latent infection by M. tuberculosis via several mechanisms, including inhibition of macrophage antigen processing and presentation, altered cytokine balance (increased IL-10 and decreased IL-12), and inhibition of Th1 effector function. In the future, a more complete understanding of macrophage-intrinsic ERK signaling and its outcomes may allow the design of host-directed therapies by interfering with this pathway and related immune evasion mechanisms to enhance protective immune functions in M. tuberculosis infection.
Model of ERK pathway signaling and outcomes in the M. tuberculosis-regulated macrophage. TLR2 signaling represents a major macrophage-intrinsic signaling pathway in response to M. tuberculosis infection, and downstream ERK activation shapes the macrophage response to outcomes that favor establishment of latent M. tuberculosis infection. TLR2-ERK signaling drives macrophages to enhance IL-10 production and diminish IL-12 production. ERK signaling also promotes arginase 1 expression and inhibits expression of MHC-II and iNOS. These phenomena collectively result in reduced presentation of M. tuberculosis antigen to T cells, reduced stimulation of T cells into a Th1 polarization state, and reduced innate macrophage microbicidal mechanisms, such as NO elaboration. If ERK signaling is selectively blocked, M. tuberculosis infection of macrophages results in less expression of IL-10 or arginase 1; higher levels of IL-12, iNOS, and MHC-II; and enhanced Th1 polarization.
ACKNOWLEDGMENTS
We thank Nancy Nagy and Logan Hubbard for technical assistance.
Research funding was provided by the following National Institutes of Health grants: T32 GM007250 (to E.T.R. and D.R.S.) and T32 AI089474 (to E.T.R.), R21 AI103443 (to P.A.W.), R01 AI027243 (to W.H.B.), and R01 AI069085 and R01 AI034343 (to C.V.H.). Portions of this research were made possible with the resources of the Case Western Reserve University/University Hospitals Center for AIDS Research (NIH P30 AI036219).
FOOTNOTES
- Received 3 February 2015.
- Returned for modification 23 February 2015.
- Accepted 12 March 2015.
- Accepted manuscript posted online 16 March 2015.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/IAI.00135-15.
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
