Regulation of TLR4-mediated signaling by IBP/Def6, a novel activator of Rho GTPases

Abstract

TLRs play a fundamental role in innate immune responses. Although Rho GTPases have been implicated in TLR-mediated signaling pathways, the molecules that control their activation in response to TLR engagement are largely unknown. IFN regulatory factor-4-binding protein (IBP; which is encoded by the gene Def6) is a unique type of activator for Rac that plays a crucial role in TCR-mediated signaling and adaptive immune responses. Here, we demonstrate that IBP/Def6 also controls innate immune responses by modulating TLR-induced signaling events. Mice deficient in IBP/Def6 are protected from LPS-induced septic shock. This protection is associated with a decrease in the production of proinflammatory cytokines and is accompanied by diminished activation of MAPKs and NF-κB. Our results thus identify IBP/Def6 as a novel component of the TLR4-induced signaling cascade that controls the production of proinflammatory cytokines.

Activation of TLRs is fundamental for the initiation of innate immune responses [1, 2]. The ability of different TLRs to recognize distinct sets of microbial products is translated rapidly into the activation of several innate cell types including macrophages. For instance, recognition of LPS by TLR4 plays a key role in the host’s defenses against Gram-negative bacteria [3]. Excessive production of proinflammatory cytokines upon TLR4 triggering in systemic infections can, however, carry untoward effects and is a critical pathogenic event in septic shock [4]. The signaling cascade initiated by TLR4 engagement has been investigated extensively [5, 6]. Two major pathways emanate from TLR4, a MyD88-dependent pathway controlling the rapid activation of NF-κB and MAPKs, leading to the production of proinflammatory cytokines such as TNF-α and IL-1, and a MyD88-independent pathway, leading to the late activation of NF-κB as well as to the activation of IFN regulatory factor-3 (IRF-3) and the expression of IFN-β and IFN-inducible genes.

Given the profound pathophysiological consequences of excessive TLR signaling, it is not surprising that this signaling cascade is controlled in a complex manner and that each step requires an elaborate interplay between positive and negative regulators [7]. In addition to adapters and kinases, members of the Rho family of GTPases, which includes Rac1 and Rac2, have been shown to participate in the transmission of TLR-mediated signals [8]. Rho GTPases have been shown to be recruited to the cytosolic domain of TLR2 and the closely related IL-1R and to regulate the production of proinflammatory cytokines [9, 10]. Activation of Rho GTPases by guanine nucleotide exchange factors (GEFs) is required for their ability to exert their downstream effects [11]. The identity of the GEFs that regulate the activation of Rho GTPases in response to TLR engagement is, however, unknown. Our laboratory has cloned a novel type of activator for Rho GTPases termed IBP (IRF-4 binding protein, encoded by the gene Def6, also known as SLAT), which exhibits significant homology with SWAP-70 [12, 13]. We have shown previously that IBP/Def6 plays a unique role in TCR-mediated signaling pathways and in the control of adaptive immune responses [14, 15]. Here, we report that disrupting the expression of IBP/Def6 also leads to defects in TLR4 signaling and in innate immunity.

The 5′ portion of the murine IBP cDNA is identical to Def6, a gene whose expression is down-regulated upon differentiation of FDCP-Mix 4a cells toward the myeloid lineage [16]. To investigate whether IBP/Def6 is expressed selectively in the lymphoid but not in the myeloid compartment, the expression of IBP/Def6 in murine myeloid cells was investigated by Western blotting. Surprisingly, murine bone marrow-derived macrophages (BMDM) and thioglycolate-elicited peritoneal macrophages all expressed IBP/Def6, albeit at levels lower than T cells (Fig. 1, and data not shown). Expression of IBP/Def6 did not change significantly upon stimulation with LPS (Fig. 1). IBP/Def6 could also be detected in human myeloid cell lines as well as in human monocytes and macrophages (data not shown). Thus, the expression of IBP/Def6 is not restricted to the lymphoid compartment and can be detected readily in distinct subsets of myeloid cells.

Fig. 1.

Expression of IBP/Def6 in murine myeloid cells. IBP/Def6 protein expression in BMDM and thioglycolate (thio)-elicited peritoneal macrophages derived from wild-type (Wt) or Def6^trap/trap mice, as indicated. Samples of BMDM stimulated with LPS (1 μg/ml) for 30 min were also assayed. Extracts from RAW264.7 (a murine macrophage cell line) and M12 (a murine B cell line) cells were also tested. Total cell lysates (20 μg) from different myeloid cell subsets were analyzed by Western blotting with an IBP/Def6 antibody reactive against the C terminus of IBP/Def6 (upper panel). Reprobing with a p38 antibody is shown as a loading control (lower panel).

The expression of IBP/Def6 in myeloid cells suggested that in addition to regulating adaptive immune responses [14], IBP/Def6 may participate in the regulation of innate immunity. To start evaluating the role of IBP/Def6 in the myeloid compartment, we used mice deficient in IBP/Def6. These mice have been termed IBP^trap/trap mice previously, as they were generated by Lexicon Pharmaceuticals (Omnibank) using a gene-trapping strategy, but will henceforth be termed Def6^trap/trap to conform with the official gene name. These mice, which have been described previously [14], were backcrossed for more than six generations into C57/BL6 mice and were kept under specific pathogen-free conditions. Lack of IBP/Def6 was found not to affect the development of macrophages. Indeed when single-cell suspensions were obtained from BM of Def6^+/+ and Def6^trap/trap mice, cultured in RPMI 1640/10% FCS with 20% L929 cell-conditioned medium at a density of 1 × 10⁶ cells/ml and macrophages collected at Day 7, the yield and phenotype of the macrophages obtained were similar (Fig. 2A, and data not shown). Furthermore, when BMDM from Wt and Def6^trap/trap mice were seeded at 0.5 × 10⁶ cells/ml (in 24-well plates) and stimulated with LPS (1 μg/ml) or LPS (1 μg/ml) + IFN-γ (60 U/ml), no significant differences in the up-regulation of the costimulatory molecule CD86 (Fig. 2A) were observed. When thioglycolate-elicited peritoneal macrophages were prepared as described previously [17], similar numbers of macrophages could also be elicited from Def6^+/+ and Def6^trap/trap mice (data not shown). Furthermore, FACS analysis confirmed that the populations of macrophages elicited from Def6^+/+ and Def6^trap/trap mice displayed similar levels of CD11b, F4/80, MHC II, and CD86 (Fig. 2B). Importantly, thioglycolate-elicited macrophages from Def6^+/+ and Def6^trap/trap mice showed similar levels of TLR4 surface expression (Fig. 2B). The morphology of the Def6^+/+ and Def6^trap/trap macrophage populations was indistinguishable, and no differences in the ability of Def6^+/+ and Def6^trap/trap macrophages to ingest IgG-opsonized particles could be detected (data not shown). Taken all together, these results thus indicate that the absence of IBP/Def6 does not significantly affect the differentiation of macrophages.

Fig. 2.

Lack of IBP/Def6 does not significantly affect the differentiation of BMDM or thioglycolate-elicited peritoneal macrophages. (A) BMDM obtained from Def6^+/+ and Def6^trap/trap mice were cultured in the presence of LPS (1 μg/ml), IFN-γ (60 U/ml), or LPS (1 μg/ml) and IFN-γ (60 U/ml). After 24 h, cells were harvested and stained with antibodies against CD11b, Class II MHC (MHC II), and CD86. Cell staining was analyzed by FACS. Results are representative of five different experiments. Unstimulated cells (Medium) were also evaluated as control. (B) Phenotype of thioglycolate-elicited peritoneal macrophages obtained from Def6^+/+ and Def6^trap/trap mice. Cells were stained with antibodies against CD11b, F4/80, MHC II, CD86, and TLR4. Cell staining was analyzed by FACS.

An analysis of the functional capabilities of peritoneal macrophages from Def6^+/+ and Def6^trap/trap mice was undertaken next. Upon stimulation with LPS (1 μg/ml), thioglycolate-elicited peritoneal macrophages from Def6^trap/trap mice displayed a marked decrease in the production of the proinflammatory cytokines TNF-α, IL-1β, and IL-6, as assessed by ELISA (eBioscience, San Diego, CA, USA; Fig. 3A). Interestingly, the lack of IBP/Def6 did not affect the ability of macrophages to produce proinflammatory cytokines upon stimulation with other TLR ligands such as Poly I:C (100 μg/ml), CpG (1μM), or Zymosan A (1.5×10⁶bioparticles/ml; Fig. 3A). Furthermore, the absence of IBP/Def6 did not alter the ability of macrophages to produce IFN-β upon LPS stimulation, as evaluated by real-time PCR (Fig. 3B). Defects in TNF-α production could also be observed when Def6^trap/trap mice were injected with LPS i.p. (25 μg/g), killed 90 min after the LPS injection, blood obtained, and serum TNF-α levels measured by ELISA (eBioscience; Fig. 3C). Engagement of TLR4 leads to the activation of a well-described series of signaling events. In particular, the activation of MAPKs and NF-κB is critical to the ability of TLR4 to elicit the production of proinflammatory mediators [2, 5]. As Rho GTPases play an important role in the activation of MAPKs and NF-κB [18, 19], we explored the possibility that the defects in macrophage function observed in the absence of IBP/Def6 were linked to abnormalities in the activation of these crucial signaling mediators. As shown in Figure 3D, the LPS-mediated activation of ERK1/2 and even more strikingly of JNK1/2 was impaired in the absence of IBP/Def6, as assessed by Western blotting of whole cell lysates from thioglycolate-elicited macrophages, which were prepared and analyzed as described previously [14]. Activation of NF-κB, as gauged by the degradation of its inhibitor IκBα, was also diminished in the absence of IBP/Def6. These results thus indicate that IBP/Def6 is critical for the optimal propagation of TLR4-mediated signals.

Fig. 3.

The absence of IBP/Def6 leads to defective production of proinflammatory cytokines by peritoneal macrophages. (A) Thioglycolate-elicited peritoneal macrophages, obtained from Def6^+/+ and Def6^trap/trap mice, were cultured in the presence or absence of LPS (1 μg/ml), polyinosinic:polycytidylic acid (Poly I:C; 100 μg/ml), CpG (1 μM), or Zymosan A (1.5×10⁶ bioparticles/ml). After 24 h, supernatants were collected and TNF-α, IL-1β, and IL-6 production measured by ELISA. Results are representative of three different experiments. (B) Thioglycolate-elicited peritoneal macrophages obtained from Def6^+/+ and Def6^trap/trap mice were cultured in the presence or absence of LPS (1 μg/ml) for the times indicated. IFN-β production was measured by real-time PCR. Results are representative of three different experiments. (C) Def6^+/+ and Def6^trap/trap mice (n=4 in each group) were injected with LPS i.p. (25 μg/g). Serum TNF-α levels before the injection and 90 min after the LPS injection were measured by ELISA. Each symbol represents serum TNF-α levels from an individual mouse. (D) Thioglycolate-elicited peritoneal macrophages derived from Def6^+/+ or Def6^trap/trap mice were stimulated with LPS (1 μg/ml) for the indicated times. Whole cell lysates were prepared, and active ERK1/2 was detected by Western blotting using an anti-phospho-ERK antibody (pERK; top panel). Active JNK1/2 was detected using an anti-pJNK antibody (second panel from top), and activation of NF-κB was assessed by following the degradation of IκBα using an anti-IκBα antibody (third panel from top). Total ERK1/2 levels are shown in the bottom panel as loading control.

The magnitude of the inflammatory responses mediated by innate immune cells must be controlled carefully, as excessive responses, such as those observed in septic shock, can lead to multiple organ failure and ultimately, to the death of an individual [4]. The pathophysiology of septic shock in mice can be mimicked by systemic administration of LPS. To evaluate whether IBP/Def6 plays a physiologic role in the regulation of innate inflammatory responses, groups of six male C57BL/6 mice or six male Def6^trap/trap mice (8–12 weeks old) were injected i.p. with two different doses (37 μg/g or 25 μg/g) of LPS (Escherichia coli serotype 055:B5, Sigma Chemical Co., St. Louis, MO, USA) in 200 μl PBS. Clinical status was monitored every 6 h for 4 days (Fig. 4, A and B). At the higher dose of LPS (37 μg/g), most Wt mice died within 30 h after the injection. In contrast, ∼70% of the Def6^trap/trap mice were alive by 30 h, and 50% had survived by the end of the experiment. The resistance of the Def6^trap/trap mice to endotoxin shock was even more striking when mice were injected with a lower dose of LPS (25 μg/g). Although only ∼15% of the Def6^+/+ mice survived through the 96-h period, ∼80% of the Def6^trap/trap were alive by the end of the experiment (Fig. 4B). Thus, in agreement with the diminished ability of Def6^trap/trap macrophages to produce proinflammatory cytokines, lack of IBP/Def6 leads to a significant protection against the exaggerated innate inflammatory responses that normally accompany systemic exposure to pathogenic bacterial components.

Fig. 4.

Def6^trap/trap mice are resistant to LPS challenge. (A) Survival curves of Def6^+/+ and Def6^trap/trap mice after challenge with LPS (37 μg/g) injected i.p. Kaplan-Meier analysis demonstrates a significant difference in survival between Def6^+/+ and Def6^trap/trap mice (P<0.05; n=6 in each group). (B) Survival curves of Def6^+/+ and Def6^trap/trap mice after challenge with LPS (25 μg/g) injected i.p. Kaplan-Meier analysis demonstrates a significant difference in survival between Def6^+/+ and Def6^trap/trap mice (P<0.05; n=6 in each group). Similar results were obtained on mice backcrossed onto the C57/BL6 background for more than 10 generations.

In this study, we demonstrate that IBP/Def6, a novel type of activator for Rho GTPases, is expressed in different subsets of myeloid cells and that disrupting the expression of IBP/Def6 leads to selective defects in TLR4 signaling and in innate immune responses. Mice that lack IBP/Def6 are indeed resistant to endotoxin-induced shock. This protection is a result of a decrease in the production of proinflammatory cytokines, which in turn, is accompanied by impairments in the activation of MAPKs and NF-κB. The precise mechanism by which IBP/Def6 controls this process is presently under investigation. We suspect that these effects are primarily a result of the ability of IBP/Def6 to function as a GEF for Rac, as Rac activation has been shown previously to be involved in the activation of MAPKs and NF-κB [18, 19]. Consistent with this notion, preliminary studies suggest that IBP/Def6 is tyrosine-phosphorylated upon LPS stimulation, and we have shown previously that tyrosine phosphorylation of IBP/Def6 is critical for its ability to function as a GEF [15]. Although IBP/Def6 can also interact with IRF-4, an IRF family member that can act as a negative regulator of TLR signaling [20, 21], it is less likely that this function of IBP/Def6 participates in the regulation of TLR4 signaling, as the interaction of IBP/Def6 with IRF-4 occurs primarily in the nucleus, and the inhibitory effects of IRF-4 on TLR signaling have been shown to be related to the ability of IRF-4 to interfere with the recruitment of IRF-5 to MyD88 in the cytoplasm [21]. Experiments are now in progress to test this possibility formally. Taken all together, these data indicate that IBP/Def6 can play a broad role in the immune system, via its ability to control innate and adaptive immune responses.