Infection of mouse macrophages by Toxoplasma gondii renders the cells resistant to proinflammatory effects of LPS triggering. In this study, we show that cell invasion is accompanied by rapid and sustained activation of host STAT3. Activation of STAT3 did not occur with soluble T. gondii extracts or heat-killed tachyzoites, demonstrating a requirement for live parasites. Parasite-induced STAT3 phosphorylation and suppression of LPS-triggered TNF-α and IL-12 was intact in IL-10-deficient macrophages, ruling out a role for this anti-inflammatory cytokine in the suppressive effects of T. gondii. Most importantly, Toxoplasma could not effectively suppress LPS-triggered TNF-α and IL-12 synthesis in STAT3-deficient macrophages. These results demonstrate that T. gondii exploits host STAT3 to prevent LPS-triggered IL-12 and TNF-α production, revealing for the first time a molecular mechanism underlying the parasite’s suppressive effect on macrophage proinflammatory cytokine production.
The intracellular protozoan Toxoplasma gondii displays potent down-regulatory effects on IL-12, TNF-α, and NO production by infected macrophages (Mφ)3 (1, 2). Subversion of Mφ function likely reflects the need to avoid immunopathology during in vivo infection (3). Although IL-12 release eventually occurs in infected Mφ, production of TNF-α remains suppressed (4). Toxoplasma also disrupts intracellular signaling cascades including NF-κB/Rel, STAT1, and MAPK transduction pathways that are important in proinflammatory cytokine induction, although for each a molecular mechanism has yet to be defined (1, 5, 6, 7). Furthermore, a link between disabled activation of signaling cascades and suppressed cytokine production during Toxoplasma infection has not been definitively established.
In light of observations that IL-10 and Toxoplasma in parallel mediate suppression of Mφ IL-12 and TNF-α production (1, 8), we tested the hypothesis that the parasite exploits an IL-10 signaling pathway to subvert proinflammatory cytokine production. The IL-10 transduction cascade is initiated by cytokine-mediated IL-10R ligation that triggers JAK1 activation and recruitment of STAT3 (9). JAK1 phosphorylates STAT3 on Tyr705 and phosphorylated STAT3 forms dimers that translocate to the nucleus. The dimer is then activated for optimal transcriptional regulation by serine phosphorylation of each monomer (10). STAT3 is essential for the immunosuppressive activity of IL-10 (11, 12, 13). Mice lacking STAT3 display an embryonic lethal phenotype, and targeted deletion of STAT3 in Mφ and neutrophils condemns mice to chronic enterocolitis characterized by uncontrolled proinflammatory cytokine production (12).
In this study, we show that Mφ infection by T. gondii induces rapid and sustained STAT3 phosphorylation, independently of host IL-10. We further demonstrate that STAT3 is crucial for effective tachyzoite-mediated suppression of endotoxin-induced IL-12 and TNF-α responses. These results define a molecular mechanism underlying the parasite’s ability to sabotage Mφ proinflammatory cytokine production.
Materials and Methods
Female C57BL/6 mice, 6–8 wk of age, were purchased from Taconic Farms and housed under specific pathogen-free conditions in the Cornell University College of Veterinary Medicine animal facility.
The lysM/cre STAT3-deficient mice that lack neutrophil and Mφ STAT3 expression were generated as described (11, 12). Mice that lack STAT3 expression in the bone marrow were derived by breeding STAT3flox/flox mice with a transgenic strain that expresses cre recombinase in hemopoietic and endothelial precursors (TIE2/cre mice) (14, 15). Age- and gender-matched littermates containing the wild-type (WT) STAT3 allele were used as controls. Bone marrow was collected from mice at 4.0–4.5 wk of age.
Bone marrow-derived Mφ preparation
Mφ were derived from bone marrow by 5-day culture in L929-containing supernatants as previously described (7).
Tachyzoites were added to cell cultures (3:1 ratio, parasites:Mφ) unless otherwise indicated in text. Plates were briefly centrifuged to synchronize tachyzoites and Mφ contact. For endotoxin triggering studies, LPS (100 ng/ml; Salmonella minnesota, ultrapure; List Biologicals) was added 60 min after infection, and cells were collected at times indicated for analysis. In some experiments, blocking anti-IL-10R or isotype control Ab (kindly provided by D. Sacks, Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, Bethesda, MD) was included. Cytokine ELISAs were performed on supernatants collected 6 h after LPS addition.
IL-12(p40) was measured by ELISA as described (1) and TNF-α was measured using a commercial kit according to the manufacturer’s instructions (R&D Systems).
Cells (2 × 106/sample) were lysed in reducing SDS sample buffer and immunoblot analysis was performed as described (7).
Coverslips bearing infected Mφ monolayers were fixed and permeabilized with ice-cold methanol and incubated (1 h, room temperature) with anti-STAT3 Ab (Cell Signaling Technologies) and FITC-tagged anti-p30 (Argene) to detect intracellular parasites. Anti-STAT3 Ab was detected with Alexa Fluor 594 secondary Ab (Molecular Probes).
Cytoplasmic and nuclear extracts
Cytoplasmic and nuclear extracts from 107 Mφ were prepared using the NE-PER extraction kit (Pierce) according to the manufacturer’s protocol.
T. gondii infection induces rapid Mφ STAT3 phosphorylation
Bone marrow-derived Mφ were infected with RH strain tachyzoites and STAT3 activation was assessed relative to that induced by LPS. Fig. 1,A shows that Toxoplasma induces strong STAT3 Tyr705 phosphorylation as early as 2 min after addition of parasites and that phosphorylation of the transcription factor is sustained for at least 22 h. Importantly, the response was distinct from LPS-induced STAT3 phosphorylation, inasmuch as the latter occurred only after 2 h, and activation decreased after 6 h. We compared STAT3 Tyr705 phosphorylation mediated by STAg and heat-killed parasites relative to infection with live parasites. As shown in Fig. 1,B, while live parasite infection induced strong STAT3 phosphorylation, neither STAg nor heat-killed parasites activated this signaling intermediate. We also found that STAT3 underwent subsequent Ser727 phosphorylation, an event occurring after nuclear translocation that is required for full transcriptional activity (Fig. 1 C).
Nuclear and cytoplasmic extracts were prepared from infected Mφ and subjected to phospho-STAT3 immunoblotting. As shown in Fig. 2,A, T. gondii infection induced accumulation of activated STAT3 in the nuclei of the infected cell population. Immunofluorescence microscopy confirmed presence of phosphorylated STAT3 in the nuclei of parasite-containing Mφ (Fig. 2 B).
Parasite-induced STAT3 phosphorylation and inhibition of LPS-induced cytokine synthesis is not due to autocrine IL-10 or IL-6
Both IL-10 and IL-6 are potent STAT3 activators, although whereas IL-10-mediated STAT3 has an anti-inflammatory effect, IL-6 has a suppressor of cytokine signaling (SOCS)-3 dependent proinflammatory outcome (17, 18, 19). To address the role of these cytokines in Toxoplasma anti-inflammatory effects, we determined whether parasite-induced STAT3 phosphorylation and cytokine inhibition was dependent upon autocrine activity of these cytokines using Mφ from IL-10 and IL-6 knockout (KO) mice. As shown in Fig. 3,A, lack of either cytokine had no effect on tachyzoite-induced STAT3 phosphorylation. To determine whether parasite-mediated suppression of LPS-induced cytokine release was independent of endogenous IL-10 and IL-6, we preinfected gene-deleted and WT Mφ then subjected the cells to LPS stimulation. As shown in Fig. 3 B, the parasite was able to suppress both IL-12 and TNF-α production in the absence of these cytokines. The combined data show that STAT3 activation is not due to autocrine IL-10 or IL-6, and that parasite-mediated cytokine suppression occurs independently of these cytokines.
Given the strong anti-inflammatory effects of IL-10, we asked whether the parasite interacted directly with the IL-10R complex to mediate cytokine suppression. Accordingly, T. gondii-induced STAT3 activation was evaluated in the presence of blocking IL-10R Ab or an isotype control. As shown in Fig. 4, blocking the IL-10R had no effect on parasite-induced STAT3 phosphorylation in either WT or IL-10 KO cells. In contrast, IL-10-driven STAT3 activation was prevented by the blocking Ab in both WT and IL-10-deficient Mφ.
Deletion of the STAT3 gene abrogates Toxoplasma inhibition of LPS-induced cytokine production
Because STAT3 deletion yields an embryonic lethal phenotype in mice, we used bone marrow-derived Mφ from both lysM/cre and TIE2/cre generated mutant animals that exhibit myeloid lineage-specific or hemopoietic and endothelial precursor-specific STAT3 deletion, respectively. To confirm STAT3 deletion in individual mice, we infected Mφ and performed Western blot analysis for both total and phospho-STAT3. As shown in Fig. 5,A, three lysM/cre putative KOs (1.2, 1.4, and 1.7) displayed significantly less total STAT3 than their transgenic control counterparts (1.8, 2.2, and 2.3). More significantly, phosphorylation of residual STAT3 in the conditional mutants was greatly diminished. Fig. 5,A also shows Western blot analysis of two STAT3-deficient (S1 and S2) mice generated by TIE2/cre targeting and two nondeleted littermates (W1 and W2). STAT3-positive and STAT3-deleted Mφ were next infected with Toxoplasma followed by LPS stimulation, and supernatants were harvested 6 h later for cytokine ELISA. Fig. 5 B shows a dramatic alleviation in T. gondii-mediated IL-12 and TNF-α suppression in each of the gene-deleted animals. In the absence of functional STAT3, tachyzoites cannot mediate effective suppression of either TNF-α or IL-12. We conclude that Toxoplasma-triggered STAT3 activation plays a major role in the immunosuppressive effects of the parasite on LPS-induced cytokine production.
The IL-10/STAT3 signaling cascade is a major pathway involved in control of proinflammatory mediators such as IL-12 and TNF-α (12, 20). We show in this study that Toxoplasma exploits STAT3 to down-modulate IL-12 and TNF-α expression in infected Mφ. Activation of STAT3 was accomplished within minutes of infection, and did not occur using heat-killed parasites or soluble T. gondii extracts. Most importantly, the ability of Toxoplasma to suppress LPS-triggered Mφ responses was highly dependent upon STAT3 expression. Rapid phosphorylation of STAT3 argues against parasite-triggered IL-10 as mediating the response, and indeed Toxoplasma was able to activate STAT3 and suppress LPS-triggered responses in IL-10-deficient host cells.
We do not yet know how T. gondii activates STAT3. The ability of the parasite to induce STAT3 phosphorylation in the presence of blocking anti-IL-10R Ab argues against triggering through this receptor. The IL-10R complex itself induces STAT3 phosphorylation by mediating activation of Jak1 and Tyk2, and it is possible that Toxoplasma triggers activation of these intermediates. The tyrosine kinase c-src also possesses STAT3-activating capability and therefore is also a candidate for parasite-triggered STAT3 phosphorylation (21, 22). Alternatively, we do not discount the possibility that cell invasion by Toxoplasma, during which parasite apical organelles are discharged and a unique parasitophorous vacuole is formed (23, 24, 25), provides a signal for direct STAT3 phosphorylation in the absence of upstream host kinase activation.
How IL-10-triggered STAT3 activation mediates anti-inflammatory effects is currently not clear, although de novo protein synthesis is known to be required (26). The SOCS-3 gene is a target for STAT3, and it has been suggested that SOCS-3 is an essential component of anti-inflammatory signaling (17). More recent evidence demonstrates that the effects of IL-10 are independent of this suppressor molecule, and suggest instead that STAT3/SOCS-3 signaling is more important in modulation of IL-6 activity (18, 19, 27).
Other studies suggest that STAT3 may interfere with the NF-κB signaling pathway, either by stabilizing cytoplasmic IκB-α, thereby preventing NF-κB nuclear translocation (28), or by blocking the activity of IκB kinase whose enzymatic activity is required for IκB-α activation (29). We, and others, have found transient inhibition of NF-κB signaling in infected cells, but nonetheless the pathway leading to IκB-α phosphorylation, ubiquitination, and degradation is intact (1, 5). The latter argues for parasite-mediated manipulation of NF-κB nuclear import per se rather than STAT3-dependent blockade in IκB-α degradation. Interference with NF-κB nuclear translocation also appears to be a transient phenomenon (4, 7), in contrast to parasite-mediated STAT3 phosphorylation that is stable for up to 22 h postinfection (Fig. 1).
Our data show involvement of STAT3 in T. gondii-induced suppression of LPS-triggered cytokine responses. Nevertheless, low level parasite-induced IL-12 synthesis was not affected by STAT3 deletion, and the parasite failed to induce TNF-α in the presence or absence of STAT3 (data not shown). These findings suggest that parasite-induced STAT3 activation may not affect pathways of cytokine production triggered by the parasite itself.
In addition to preventing production of LPS-induced IL-12 and TNF-α, Toxoplasma has recently been reported to manipulate other host responses. These include blocks in LPS-induced MAPK signaling and IFN-γ-induced STAT1 nuclear translocation in mouse Mφ (6, 7). In mouse dendritic cells, infection inhibits proinflammatory cytokine production and expression of costimulatory molecules (30). It has also been found that T. gondii infection renders cells resistant to inducers of apoptosis (31, 32). Whether some, or all, of these effects are mediated through STAT3 has yet to be determined.
The authors have no financial conflict of interest.
We thank Dr. D. Sacks for advice and the gift of anti-IL-10R Ab, Dr. A. Sher for valuable discussion and critical review of the manuscript, and R. Rutschman for breeding the STAT3flox/flox, lysM/cre mice used in this study.
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.
This work was supported by National Institute of Allergy and Infectious Diseases Grants AI50617 (to E.Y.D.) and AI062921 (to P.J.M.), and American Heart Association, Texas Affiliate 0455143 (to S.S.W.).
Abbreviations used in this paper: Mφ, macrophage; KO, knockout; STAg, soluble tachyzoite Ag; SOCS, suppressor of cytokine signaling; WT, wild type.