Flightless I Homolog Negatively Modulates the TLR Pathway

Tianyi Wang, Tsung-Hsien Chuang, Tapani Ronni, Sheng Gu, Yu-Chun Du, Hong Cai, Hui-Qiao Sun, Helen L. Yin and Xian Chen
J Immunol February 1, 2006, 176 (3) 1355-1362; DOI: https://doi.org/10.4049/jimmunol.176.3.1355

Abstract

To date, much of our knowledge about the signaling networks involved in the innate immune response has come from studies using nonphysiologic model systems rather than actual immune cells. In this study, we used a dual-tagging proteomic strategy to identify the components of the MyD88 signalosome in murine macrophages stimulated with lipid A. This systems approach revealed 16 potential MyD88-interacting partners, one of which, flightless I homolog (Fliih) was verified to interact with MyD88 and was further characterized as a negative regulator of the TLR4-MyD88 pathway. Conversely, a reduction in endogenous Fliih by small-interfering RNA enhanced the activation of NF-κB, as well as cytokine production by LPS. Results from immunoprecipitation and a two-hybrid assay further indicated that Fliih directly interfered with the formation of the TLR4-MyD88 signaling complex. These results in turn suggest a new basis for the regulation of the TLR pathway by Fliih.

Toll-like receptors recognize pathogen-associated molecular patterns and initiate effective host immune responses via similar, but not identical, signaling pathways (1). MyD88, the immediate downstream adaptor protein of all TLRs except TLR3, plays an important role in the control of signaling specificity (2). Upon engagement with microbial ligands, TLRs initiate a canonical signal transduction cascade mediated by MyD88-IL-1R-associated kinase (IRAK)4-TNFR-associated factor 6 (TRAF6)-NF-κB-inducing kinase-I/NF-κB, as well as the MAPK pathway (3). In contrast, TLR3 and TLR4 can use a distinct adaptor protein, Toll/Il-1R (TIR) domain-containing adaptor inducing IFN-β (TRIF), to activate IFN-β and IFN regulatory factor-3 (4, 5, 6). Upon receptor activation, TRIF recruits signaling proteins TRAF-associated NK-κB activator-binding kinase 1 and receptor-interacting protein 1 to induce a whole different set of genes whose expression patterns are typically dependent on IFN regulatory factor families (7, 8). Besides MyD88 and TRIF (or TIR domain-containing adaptor molecule-1), two other TIR-containing adaptors, TRIF-related adaptor molecule (also known as TICAM-2) and MyD88 adaptor-like (also known as TIR domain-containing adaptor protein), were only involved in the TLR4 and TLR2/4 pathway (9, 10, 11, 12). Therefore, it is a common notion that different adaptor proteins may interact with many other unidentified proteins, leading to specific downstream signaling pathways (13).

Although studies of signal amplifiers dominate the field, several proteins are known to negatively regulate TLR signaling at the level of MyD88. For instance, membrane-bound ST2 negatively regulates the IL-1R and TLR4 by interacting with MyD88 (14). MyD88s, an alternatively spliced form of MyD88 that blocks recruitment of IRAK-4, has also been shown to act as a negative regulator of Toll and IL-1 signaling (15). A third negative regulator, IRAK-M, prevents dissociation of IRAK-1 and IRAK-4 from MyD88 and formation of IRAK-TRAF6 complexes (16). Consequently, by identifying and characterizing multiple proteins that interact with MyD88, we may begin to understand how the specificity of different TLR-mediated signaling pathways is coordinated.

Recently, systems approaches have shown great potential for the mapping of signaling networks (17, 18). We have previously shown that stable isotope-labeled amino acids can be naturally incorporated into cellular proteomes through cell culturing in a residue-specific manner (19). Subsequently, both our group and the group of Mann and colleagues (20, 21) independently applied stable isotope labeling for quantitative protein profiling. In the current study, we combined stable isotope labeling and epitope tagging to allow the sensitive and selective identification of the MyD88-signaling complex formed in actual immune cells. As a result, we identified several proteins not previously known to interact with MyD88, one of which was flightless I homolog (Fliih). We further demonstrated that Fliih suppresses TLR4-MyD88-mediated activation of NF-κB, presumably by preventing MyD88 from forming a functional signaling complex with TLR4.

Materials and Methods

Cell culture

The murine alveolar macrophage cell line AMJ2-C8 (ATCC CRL-2455; American Type Culture Collection), J774A.1 (ATCC TIB-67), and the 293T cell line (ATCC CRL-11268) were cultured in DMEM (Invitrogen Life Technologies) supplemented with 10% FBS (HyClone), 1% penicillin, and 10 μg/ml streptomycin. The AMJ-C8 cell line stably expressing Flag-calmodulin-binding protein-tagged MyD88 was made by retroviral infection and maintained in DMEM containing 1.8 μg/ml puromycin. The 293T stable cell line expressing Flag-TLR4 was made in the laboratory and maintained in DMEM containing 0.7 μg/ml puromycin. The human U373 glioblastoma/astrocytoma cell line stably expressing CD14 was described previously (22).

Plasmids and reagents

The full-length cDNA fragment of murine MyD88 was cloned into a retroviral vector (pMIR-DFT) containing a double Flag tag and a calmodulin-binding peptide at the N terminus. The pMIR-DFT was a gift from Dr. S. Smale (University of California, Los Angeles, CA (UCLA)). The expression vector for HA-tagged human Fliih was constructed in the laboratory of H. L. Yin at University of Texas Southwestern Medical Center. The ELAM NF-κB LUC reporter was a gift from Drs. P. Godowski (Genentech, South San Francisco, CA) and R. Modlin (UCLA). The IFN NF-κB LUC plasmid was obtained from Drs. J. Pomerantz and M. Boldin (Caltech, Pasadena, CA). The expression plasmids for MyD88-N, MyD88-C, and IRAK-1 were obtained from Dr. Z. Cao (Tularik, South San Francisco, CA). pcDNA3-Fliih and plasmids expressing Flag-tagged gelsolin and leucine-rich repeat (LRR) fragments of the human Fliih were gifts from Dr. M. Stallcup (University of Southern California, Los Angeles, CA; Ref.23). The IL-8 promoter-driven luciferase reporter and the IFN-stimulated regulatory element-Luc were made in Dr. R. J. Ulevitch’s laboratory (The Scripps Research Institute, La Jolla, CA) previously (24). The following reagents were purchased from the companies indicated: polyclonal anti-MyD88, LPS (serotype R515), and lipid A (Alexis Biochemical); human (h) TNF-α and hIL-1β (PeproTech); pFlag-CMV-1, Abs against Flag, hemagglutinin (HA) and Myc epitopes, and anti-Flag M2 resin (Sigma-Aldrich); pNF-κB-Luc (Stratagene); leucine-d3 (Cambridge Isotope Laboratories); mouse mAb against Fliih (Santa Cruz Biotechnology); and Pam3cys, poly(I:C), and CpG oligos (Invivogen).

Amino acid-coded mass tagging and purification of MyD88-interacting proteins, liquid chromatograph (LC)-mass spectrometry (MS)/MS analysis, and database searching and protein identification

The detailed protocol has been described previously (25, 26).

Immunoprecipitation and Western blot

Unless indicated otherwise, we typically transfected 6–10 million 293T cells with the plasmids indicated, using the calcium phosphate method. Forty-eight hours after transfection, cells were lysed using a lysis buffer described elsewhere (27). A quantity of 50–100 μg of total soluble extract was incubated with an appropriate amount of beads conjugated with specified Ab for 2 h at 4°C. After washing, the bound proteins were eluted and then were separated using SDS-PAGE. The proteins were transferred to a nitrocellulose membrane and incubated with appropriate Abs for Western blot.

RNA interference (RNAi)

Small interfering (si) RNA experiments were performed as described previously (28). ssRNAs were synthesized by Integrated DNA Technologies and annealed using the commercial annealing buffer. The siRNA sequence used in this study has been published previously (23). siRNA duplexes were transfected into 293T cells using Lipofectamine 2000 (Invitrogen Life Technologies) according to the manufacturer’s protocol. The siRNA sequences used (listed 5′ to 3′) were as follows: si-FliA, GCUGGAACACUUGUCUGUGdTdT and CACAGACAAGUGUUCCAGCdTdT; si-FliB, CAACCUGACCACGCUUCAUdTdT and AUGAAGCGUGGUCAGGUUGdTdT. U373/CD14 cells were plated in 6-well plates and transfected with 2 μg of si-FliA or control (scramble) siRNA duplex using GeneSilencer siRNA transfection reagent (Gene Therapy Systems). The sequences of control (scramble) siRNA were UUCUCCGAACGUGUCACdTdT and ACGUGACACGUUCGGAGAAdTdT (purchased from Qiagen). The retroviral vector used for RNAi was a gift from Dr. C. Li from Dr. S. Smale’s laboratory at UCLA.

Reporter assay

For the luciferase reporter assay, 0.25 million 293T cells/well were seeded on 6-well plates. The next day, cells were transfected with Lipofectamine 2000 (Invitrogen Life Technologies) according to the manufacturer’s instructions. A prespecified amount of DNA was added into each transfection and the pFLAG-CMV-1 plasmid was used as filler DNA. When indicated, cells were treated with various stimuli for the specified time. Cells were harvested between 24 and 48 h after transfection and luciferase activity was measured using the Luciferase Assay System (Promega). The CD14-transfected U373 cells were grown in DMEM plus 10% FBS. Cells were plated in 6-well plates and transfected with 0.1 μg of IL-8 promoter luciferase reporter gene plus 0.2 μg of β-galactosidase plasmid for normalization with Lipofectamine 2000 (Invitrogen Life Technologies). Twenty-four hours later, the cells were treated with 0.2 μg/ml LPS, 10 ng/ml IL-1, and 10 ng/ml TNF-α for 6 h. The cells were lysed and luciferase activities were determined using reagent from Promega. The data presented are the mean ± SE (n = 3).

ELISA

Thirty-six hours after transfection of the U373 cells stably expressing CD14 with siRNA, the cells were stimulated with 0.2 μg/ml LPS or 10 ng/ml IL-1. Ten hours later, supernatants were removed and assayed for TNF-α and IL-8 production by ELISA using capture Abs and biotinylated detection Abs followed by avidin-HRP for detection following the manufacturer’s protocol (eBioscience).

Mammalian two-hybrid assay

The reporter plasmid pG5GFP-hygromycin driven by five copies of Gal4-binding sites was obtained from Dr. T. Shioda (Massachusetts General Hospital Cancer Center, Charlestown, MA) (29). Bait and prey proteins in the mammalian two-hybrid assay were cloned into BamHI and SalI sites of the BD Clontech pM and pVP16 cloning plasmids, which result in the expression of fusion proteins carrying the Gal4 DNA-binding domain (bait) and the VP16 activation domain (prey) separately. The TRIF clone was originally obtained from Dr. B. Beutler (The Scripps Research Institute, La Jolla, CA). A positive interaction pair pM-p53 and pVP16-SV40-T and the negative control vector (pVP16-CP) came from the mammalian Matchmaker two hybrid assay kit (BD Clontech). Typically, 2 μg of pG5GFP-hyg, bait vector, and prey vector were transfected into 293T cells growing in a 60-mm plate at ∼20% confluence. On the third day, cells were analyzed on a FACSCalibur (BD Biosciences) with CellQuest software for GFP expression.

Results

General experimental approach

Fig. 1⇓ illustrates the “dual-tagging” (single epitope tag and isotope mass tag) quantitative proteomic approach for the identification and characterization of MyD88-interacting proteins (for more detail, consult Materials and Methods and Ref.26). In this study, a macrophage cell line stably expressing Flag-MyD88 was generated for pull-down of the MyD88 signaling complex, while stable isotope-labeled amino acids served as “in-spectra” quantitative markers to distinguish genuine MyD88-interacting components from a nonspecific background. This approach minimizes the number of affinity purification steps necessary, thereby improving detection sensitivity. A total of 16 proteins were highly enriched in the sample containing the Flag-MyD88 (Table I⇓) (26). The current report focuses on a single-interacting protein of interest, Fliih.

FIGURE 1.
FIGURE 1.

Overall experimental approach for identification and characterization of MyD88-interacting proteins. Parental macrophage cells were grown in regular medium (green plate), while macrophages stably expressing Flag-MyD88 were maintained in the leucine-d3-containing medium (red plate). After stimulation and lysis, a single affinity purification was performed to pull-down proteins that interact with Flag-MyD88 and then the eluates from both cell pools were mixed at a 1:1 ratio and separated on a one-dimensional SDS-PAGE before tryptic digestion followed by μLC-MS/MS analysis. On the mass spectra, green peaks represent background signals generated from peptides derived from unlabeled cells, which do not express Flag-MyD88, while red peaks represent peptides derived from cells labeled with leucine-d3 that stably express Flag-MyD88. In principle, background-binding proteins can be distinguished based on the 1:1 ratio between labeled and unlabeled states if a pair of mass speaks with expected mass split can be found (peaks 1 and 3). Proteins that interact with Flag-MyD88 will be distinguished by increased intensity of the labeled peaks as compared with the unlabeled peaks (peaks 2, 4, and 5). The specific interaction identified by this approach can be validated by coimmunoprecipitation or two-hybrid assay. Reporter assay and RNAi will further characterize the function of these new MyD88-interacting molecules.

View this table:
Table I.

Inhibition of direct interactions between TIR/TLR4 and MyD88 and TRIF by Fliiha

Identification of the Fliih as a MyD88-interacting protein

Using the dual-tagging method, Fliih was identified as a specific MyD88-interacting protein with high confidence: 17 of the peptides containing the deuterated leucine (Leu-d3) derived from Fliih were simultaneously identified and quantified. On average, this protein is enriched by 4-fold (SD = 0.38), compared with the 2.5-fold enrichment of IRAK. A representative mass spectrum is given in Fig. 2⇓. Fliih was first identified as a protein involved in gastrulation and muscle degeneration in flies (30). The N-terminal region of Fliih contains 16 copies of the LRR motif, while the central and C-terminal regions are homologous to the gelsolin family of actin-binding proteins. LRR domains have been implicated in protein-protein interactions (31). Subsequent studies in human and mouse cells suggested that Fliih links signal transduction pathways involving Ras-related GTPases to cytoskeletal control mechanisms (31, 32). According to a basic local alignment search tool comparison, the LRR regions of Fliih and TLR4 share a 29% sequence identity and a 42% similarity.

FIGURE 2.
FIGURE 2.

Representative mass spectra of peptides derived from actin (A) and Fliih (B). The labeled peak from Fliih peptide was enriched 4-fold while the signal intensity of the unlabeled actin peptide appears similar to that of the labeled one.

Fliih negatively regulates MyD88-mediated pathways

We used reporter assays to examine the effects of Fliih expression on the MyD88-dependent pathway in various cell lines. Specifically, we used 293T cells stably expressing TLR4 and transiently transfected with MD-2, which are responsive to LPS and lipid A (Fig. 3⇓A), and macrophage cell line J774, which is responsive to LPS and lipid A (Fig. 3⇓B). In both cases, the overexpression of Fliih exerted a significant inhibitory effect on LPS or lipid A-induced NF-κB activation. The inhibitory effect of Fliih on a specific cytokine promoter was further examined in CD14 stably transfected U373 cells (22). Transfection of these cells with HA-Fliih blocked IL-1- and LPS-induced IL-8 promoter activity but had no significant effect on TNF-α-induced IL-8 promoter activity (Fig. 3⇓C). This observation is consistent with the observation that TNF-α signaling is not MyD88-dependent.

FIGURE 3.
FIGURE 3.

Fliih negatively regulates MyD88-dependent NF-κB activation in multiple cell types. Cells were transfected with an NF-κB reporter gene plus expression vectors and treated with various stimuli, then the NF-κB activity in each sample was determined with luciferase reporter assay. A, 293T cells stably expressing TLR4 were transfected with the NF-κB reporter gene, MD2, and HA-Fliih. B, J774 cells transfected with the NF-κB reporter gene plus HA-Fliih. Control, Empty pcDNA3 vector. C, Transfection of CD14 stably transfected U373 cells with 0.4 μg of HA-Fliih blocked IL-1 and LPS-induced IL-8 promoter activity, but had no effect on the activity of TNF-α. A result representative of four independent experiments is shown here.

We next investigated how depletion of endogenous Fliih affected TLR activation. When we transfected 293T cells stably expressing TLR4 with the siRNA duplex (si-FliB) using an MD-2 expression plasmid, we observed an enhanced cellular response to LPS (Fig. 4⇓A). When the same cell lysate was analyzed by Western blot, expression of both an endogenous and an HA-tagged Fliih was significantly suppressed by the siRNA duplex. Another siRNA duplex (si-FliA), which targets a different part of Fliih, also inhibited the expression of endogenous Fliih in U373 cells (Fig. 4⇓B). Transfection of this duplex into U373 cells enhanced the production of TNF-α and IL-8 in response to IL-1 and LPS treatment. Similarly, a retroviral vector-based RNAi against Fliih enhanced the LPS-induced NF-κB activity in J774 cells (Fig. 4⇓C). Taken together, these siRNA studies confirm that the endogenous Fliih indeed plays a role in regulating the MyD88-mediated pathway. Considering the fact that in the current study the inhibition of Fliih expression by siRNA at the time when the cells were subjected to reporter assay is ∼2- to 3-fold, the actual effect of complete depletion of Fliih should be much more dramatic. Genetic knockout of Fliih will provide further validation of the role of Fliih in vivo.

FIGURE 4.
FIGURE 4.

Depletion of endogenous Fliih enhanced cellular response to LPS and IL-1. A, 293T cells stably expressing TLR4 were transfected with si-FliB, HA-Fliih, the NF-κB reporter gene, and MD2. Cells were stimulated with LPS (1 μg/ml) for 16 h before reporter assay (upper panel). Lysates from A were separated by SDS-PAGE and Fliih was detected by Western blot (lower panel). 15′ represents an exposure time of 15 min; 30“represents 30 s exposure time. B, CD14 stably transfected U373 cells were transfected with 2 μg of si-FliB for 36 h and then stimulated with LPS or IL-1 for another 8–12 h. IL-8 and TNF-α secreted into the medium were determined by ELISA (upper panel). The effect of si-FliA on endogenous Fliih was verified by Western blot (lower panel). C, A 1 or 3 μg retroviral vector expressing RNAi against mouse Fliih was transfected into J774A.1 cells alone with the NF-κB reporter gene. The cells were then stimulated by LPS for 12 h before luciferase activity was determined. Data are from one of three independent experiments with similar results.

Association of Fliih with MyD88

To further learn the biochemical properties of the Fliih-MyD88 interaction, we first examined the kinetics of LPS-induced Fliih recruitment to MyD88 in macrophages. Because neither Abs against endogenous MyD88 nor Fliih worked in the immunoprecipitation studies (data not shown), we decided to use a macrophage cell line that stably expresses Flag-tagged MyD88. When these cells were stimulated with LPS, endogenous Fliih coprecipitated with MyD88 5 min after the stimulation and then quickly disappeared (Fig. 5⇓A). To determine the domains of interaction between Fliih and MyD88, the gelsolin and LRR fragments of Fliih fused with the Flag epitope were transiently coexpressed with either the N-terminal or the C-terminal half of MyD88 in 293T cells. As a result, the gelsolin domain of Fliih coprecipitated with the C-terminal TIR domain of MyD88 (Fig. 5⇓B). Subsequently, we tested whether Fliih interacts with TRIF, another TIR domain-containing adaptor protein. We used a sensitive mammalian two-hybrid assay in which the interaction of two proteins results in the production of GFP that can be measured directly by flow cytometric analysis. As indicated in Fig. 6⇓A, this assay detected interactions between TRIF and Fliih, and between MyD88 and Fliih. Interestingly, we also detected an interaction between the TIR domain common to TLR4 and Fliih, suggesting Fliih may generally interact with TIR-containing proteins. Consistent with these observations, overexpression of Fliih inhibited TLR2- and TLR9-mediated NF-κB activation (MyD88-dependent) as well as the TLR3-mediated IFN-stimulated regulatory element activation (TRIF-dependent) in the reporter assays (Fig. 6⇓B).

FIGURE 5.
FIGURE 5.

Association of Fliih with MyD88. A, Kinetics of LPS-induced MyD88-Fliih association. C8 macrophages stably expressing Flag-MyD88 (100 million) were stimulated with 1 μg/ml lipid A for the indicated time. MyD88 complexes from cell lysates were immunoprecipitated with anti-Flag M2 affinity resin and immunoblotted with anti-Fliih Ab (top) and anti-Flag M2 Ab (bottom). The first lane was sample derived from parental C8 cells. B, Mapping the interaction domains between Fliih and MyD88. 293T cells (6 × 106) in a 10-cm dish were transfected with the indicated combinations of expression plasmids for Flag-Gelsolin (aa 495-1268), Flag-LRR (aa 1–494), Myc-tagged MyD88-N (aa 1–151), and MyD88-C (aa 152–296) (2 μg of each construct/transfection). Twenty-four hours after transfection, lysates were prepared and immunoprecipitations were performed with anti-Flag M2 resin. Coprecipitating Myc-tagged MyD88 was detected with anti-Myc Ab (top). Immunoblotting the same membrane with anti-Flag mAb indicates that similar amounts of the Gelsolin and LRR were immunoprecipitated (middle). Lysates of the transfected cells were immmunoblotted with anti-Myc to monitor the expression of MyD88-N and C (bottom).

FIGURE 6.
FIGURE 6.

Effect of Fliih on other TLRs. A, Combinations of plasmids expressing Fliih, MyD88, TRIF, and TIR/TLR4 (aa 627–835 of mouse TLR4) were transfected into 293T cells in a 60-mm plate with 2 μg of the reporter construct pG5GFP-Hyg. Twenty-four hours after transfection, GFP expression was determined by flow cytometry. p53 and SV40-T Ag served as a positive-interacting pair. The percentage of positive cells and the mean fluorescence value of these cells are indicated at the upper right corner of each dot plot. The thin dotted line indicates an arbitrary cutoff point for positive cells. B, Fliih was overexpressed in HEK293 cells stably expressing TLR2, TLR3, and TLR9. The specified amount of agonists were added into the medium for 12 h before luciferase activity was measured. Data are from one of three independent experiments with similar results.

The biochemical evidence presented above suggests that Fliih interferes with the TLR signaling via its interaction with the TIR domain-containing proteins. To determine the hierarchy of this inhibition, we performed a reporter assay in 293T cells. We found that overexpression of Fliih completely abolished the ability of MyD88 to activate NF-κB (Fig. 7⇓A); likewise, it also abolished the ability of CD4/TLR4 to activate NF-κB (Fig. 7⇓B). However, IRAK was still able to activate NF-κB (Fig. 7⇓C) in the presence of overexpressed Fliih. Therefore, we conclude that Fliih works at the level of MyD88 and above to inhibit the TLR4-MyD88-mediated NF-κB activation.

FIGURE 7.
FIGURE 7.

Fliih inhibits the TLR4-MyD88 but not IRAK-induced NF-κB activation. A, Fliih inhibited MyD88-induced activation of NF-κB. The means and SD of the means (error bars) are from three independent experiments. B, Fliih inhibited CD4/TLR4-caused activation of NF-κB. The means and SD of the means (error bars) are from four independent experiments. C, Activation of NF-κB by IRAK was not affected by overexpression of Fliih. The means and SD of the means (error bars) are from four independent experiments. The ratio of SD/mean in B is at ∼15–20%.

Fliih interferes with the formation of the TLR4-MyD88 signaling complex

Binding of the gelsolin domain of Fliih to the TIR domain of MyD88 suggests that Fliih may compete with TLRs for binding with MyD88. To examine this possibility, we transiently expressed Fliih and MD2 in 293T cells stably expressing Flag-TLR4, then stimulated the cells with LPS. Immunoprecipitation with an anti-Flag Ab coprecipitated the tagged TLR4 and the endogenous MyD88 in mock-transfected cells upon stimulation but not in Fliih-transfected cells, indicating that overexpression of Fliih abolished the formation of the TLR4-MyD88 signaling complex (Fig. 8⇓). To further test the hypothesis, the mammalian two-hybrid assay was conducted to examine whether the presence of Fliih can directly disrupt the interaction between MyD88 and TLR4. As indicated in Table I⇑, coexpression of Fliih significantly suppressed the interaction between MyD88 and TLR4 as well as that between TRIF and TLR4, both by ∼80%, yet showed minimal effect (∼15%) on the interaction between the control pair of p53 and SV40-T Ag.

FIGURE 8.
FIGURE 8.

Overexpression of Fliih inhibited the formation of the TLR4-MyD88 complex. 293T cells (200 million) stably expressing TLR4 were transfected with MD2 or MD2 plus Fliih. After stimulation with LPS (1 μg/ml), lysates were immunoprecipitated with anti-Flag M2 resin. Coprecipitated endogenous MyD88 was detected using an anti-MyD88 polyclonal Ab (top). The same membrane was immunoblotted with anti-Flag to indicate where similar amounts of the TLR4 were precipitated (middle).

Discussion

TLR signal transduction is regulated at many levels, from protein and mRNA expression to protein ubiquitination and degradation (16, 33, 34, 35). The activation of TLR signaling has been the main focus of the research in the field; however, it is equally important to understand how and when the signaling is terminated. MyD88, the critical adaptor protein for most of the TLRs, plays a major role in this complicated regulatory network. Previously, it has been suggested that ST2 and IRAK-M both inhibit TLR signaling by interacting with MyD88 (14, 16). SIGIRR, another inhibitor of TLR-IL-1R signaling, interacted with IL-1R, TLR4, and TRAF6 (36). By using a novel quantitative proteomic approach, we were able to identify many more proteins, including the previously known partner that interacts with MyD88 in macrophages upon stimulation with lipid A. To our knowledge, this is the first time such a study was successfully conducted in actual immune cells. Our data indicate that Fliih exerts an inhibitory effect on the TLR4-MyD88 signaling pathway by interfering with the formation of the TLR4-MyD88 signaling complex. Unlike ST2 and Triad3A, whose effects are restricted to certain TLRs (14, 28), Fliih seems to inhibit TLR pathways in general. This is not surprising because Fliih was found to interact with MyD88, TRIF, and the TIR domain derived from TLR4 in the mammalian two-hybrid assay. Collectively, these observations have significantly broadened our understanding of the termination mechanism of TLR signaling.

In Drosophila melanogaster, homozygous Fliih mutant embryos develop normally in the absence of the maternal product until cellularization, when defects first appear, followed by the failure to complete gastrulation, leading to the arrest of embryonic development (37). Fliih expression is also essential for early embryonic development in mice (38). It has been suggested that Fliih links signaling proteins to cytoskeletons (31). Its expression was mainly observed in the nucleus, with translocation into the cytoplasm in the presence of serum (39). A recent report indicates that Fliih interacts with an arginine-specific histone methyltransferase CARM1 and serves as a nuclear receptor coactivator (23). In our hands, we found it is predominantly expressed in cytoplasm, while a small portion of the protein can be found in nuclear extract (data not shown). Overexpression of Fliih did not change MyD88 expression at the protein level, nor was the protein expression of Fliih induced upon stimulation (data not shown). We are puzzled as to how Fliih serves as an inhibitor without being induced upon stimulation. Examination of the binding kinetics between Fliih and TLR4 will help to clarify this issue. Fliih may only interact with TLR4 upon cell stimulation.

One possibility is that Fliih specifically interacts with the TIR domains of MyD88 and TLR4. Such an interaction prevents the formation of the TLR4-MyD88 signaling complex (Fig. 9⇓). Similarly, Fliih may interact with TRIF, thereby blocking the TRIF-dependent pathway. Results from both the reporter assay and the mammalian two-hybrid assay strongly support such a model. However, the interactions between Fliih and TRIF and other TLRs should be carefully examined by alternative approaches for validation because the TIR domains derived from different TLRs apparently can interact with different proteins. One cannot assume a protein will bind all TIR domain-containing proteins if it binds the TIR derived from TLR4. For example, Triad3A binds TLR3, 4, 5, and 9, but not TLR2 (28). Furthermore, the mammalian two-hybrid assay we used is highly sensitive yet artificial. This system forces the expression of bait and prey proteins in the same nucleus where the reporter gene is transcribed. Therefore, while we observed the nuclear expression of Fliih, known physiological interactions of TLR4, MyD88, and TRIF normally occur in the cytosol, in close proximity to the cell membrane. Nevertheless, we could speculate that Fliih binds to the TIR domains of TLR4 and MyD88, thereby blocking the interaction sites on both proteins that are required for TLR4-MyD88 complex formation. This interaction could either block the recruitment of MyD88 to TLR4 or speed up the dissociation process of MyD88 from the receptor complex. The second possibility seems to be more in line with what we observed. In the present study, Fliih quickly interacted with MyD88 upon stimulation and then dissociated. Therefore, it is less likely that Fliih would constantly block the interaction of MyD88 and TLR4. In fact, neither the gelsolin nor the LRR domain of Fliih could efficiently block the MyD88-induced NF-κB activation in a reporter assay (data not shown), suggesting the inhibition requires not only an interaction between Fliih and MyD88 (through gelsolin-TIR) but also an unknown mechanism acquired via the LRR domain of Fliih. Perhaps Fliih undergoes some kind of posttranslational modification or becomes released from some kind of inhibitor(s) upon stimulation, which then interacts with MyD88 and TLR4 to terminate the signaling. It has been reported that Fliih interacts with Ras protein via the LRR domain (40, 41). Ras participates in CpG oligodeoxynucleotide signaling through association with TLR9 and promotion of IRAK/TRAF6 complex formation in macrophages (42). It would be interesting to examine whether Ras plays any role in the Fliih-mediated inhibitory effect on the TLR pathway. Further characterization of Fliih upon cellular stimulation may solve the mystery.

FIGURE 9.
FIGURE 9.

Model of the inhibitory effect of Fliih on TLR4-MyD88 pathways.

Another important question is whether Fliih plays a similar role in terminating other TLR-mediated pathways, both in vitro and in vivo. Does it play any role in endotoxin tolerance? In our reporter assays, overexpression of Fliih was able to block the signaling mediated by TLR2, TLR3, and TLR9. This is not surprising because TLR2, TLR3, and TLR9 use either MyD88 or TRIF as downstream adaptor. Genetic knockout of Fliih in mice will provide a more definite role of Fliih in regulating different TLR-mediated pathways in vivo.

In summary, we used a novel proteomic approach to identify the MyD88 signalosome formed in macrophages upon stimulation. Such an approach can be readily applied in other signaling studies. Our results also demonstrated inhibition of TLR4-MyD88 pathways by Fliih. Fliih interacts with the TIR domain of TLR4 and MyD88, as well as TRIF. This interaction interferes with the formation of the TLR4-MyD88 signaling complex upon stimulation. Subsequent experiments with siRNA to deplete endogenous Fliih suggest that it plays a prominent role in the down-regulation of IL-1/TLR4 signaling. While the detailed mechanism is under study, the inhibitory effect of Fliih on the TLR pathway provides a molecular basis for regulation of the pathway, a topic of potential interest for future studies.

Acknowledgments

We thank Dr. Monica Eiland for editorial assistance. We are grateful to Drs. Z. Cao, S. Smale, P. Godowski, Geng-Hong Cheng, J. Pomerantz, M. Boldin, M. Stallcup, T. Shioda, and R. Modlin for providing plasmids. We also thank Jianhong Zhou for her assistance in preparation of the reagents.

Disclosures

The authors have no financial conflict of interest.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • ↵1 This work was supported by Department of Energy Grants ERW9840 and ERW9923 and Los Alamos National Laboratory Directed Research Development Grants 20030508ER (to X.C.), and University of Texas Southwestern Burn Center Grant P50GM0216681 and the Welch Foundation (to H.L.Y.). X.C. is a recipient of a Presidential Early Career Award for Scientists and Engineers.

  • ↵2 T.W., T.-H.C., T.R., and S.G. contributed equally to this work.

  • ↵3 Address correspondence and reprint requests to Dr. Xian Chen, Bioscience Division, Los Alamos National Laboratory, Mail Stop M888, Los Alamos, NM 87545. E-mail address: chen_xian{at}lanl.gov

  • ↵4 Abbreviations used in this paper: IRAK, IL-1R-associated kinase; TRAF, TNFR-associated factor; TIR, Toll/IL-1R; TRIF, TIR domain-containing adaptor-inducing IFN-β; Fliih, flightless I homolog; HA, hemagglutinin; LRR, leucine-rich repeat; LC, liquid chromatography; MS, mass spectrometry; RNAi, RNA interference; si, small interfering.

  • Received September 13, 2005.
  • Accepted November 4, 2005.

References

  1. ↵
    Akira, S.. 2003. Mammalian Toll-like receptors. Curr. Opin. Immunol. 15: 5-11.
    OpenUrlCrossRefPubMed
  2. ↵
    Medzhitov, R., P. Preston-Hurlburt, E. Kopp, A. Stadlen, C. Chen, S. Ghosh, C. A. Janeway, Jr. 1998. MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol. Cell. 2: 253-258.
    OpenUrlCrossRefPubMed
  3. ↵
    Barton, G. M., R. Medzhitov. 2003. Toll-like receptor signaling pathways. Science 300: 1524-1525.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Doyle, S., S. Vaidya, R. O’Connell, H. Dadgostar, P. Dempsey, T. Wu, G. Rao, R. Sun, M. Haberland, R. Modlin, G. Cheng. 2002. IRF3 mediates a TLR3/TLR4-specific antiviral gene program. Immunity 17: 251-263.
    OpenUrlCrossRefPubMed
  5. ↵
    Hoebe, K., X. Du, P. Georgel, E. Janssen, K. Tabeta, S. O. Kim, J. Goode, P. Lin, N. Mann, S. Mudd, et al 2003. Identification of Lps2 as a key transducer of MyD88-independent TIR signalling. Nature 424: 743-748.
    OpenUrlCrossRefPubMed
  6. ↵
    Yamamoto, M., S. Sato, H. Hemmi, K. Hoshino, T. Kaisho, H. Sanjo, O. Takeuchi, M. Sugiyama, M. Okabe, K. Takeda, S. Akira. 2003. Role of adaptor TRIF in the MyD88-independent Toll-like receptor signaling pathway. Science 301: 640-643.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Sato, S., M. Sugiyama, M. Yamamoto, Y. Watanabe, T. Kawai, K. Takeda, S. Akira. 2003. Toll/IL-1 receptor domain-containing adaptor inducing IFN-β (TRIF) associates with TNF receptor-associated factor 6 and TANK-binding kinase 1, and activates two distinct transcription factors, NF-κB and IFN-regulatory factor-3, in the Toll-like receptor signaling. J. Immunol. 171: 4304-4310.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Meylan, E., K. Burns, K. Hofmann, V. Blancheteau, F. Martinon, M. Kelliher, J. Tschopp. 2004. RIP1 is an essential mediator of Toll-like receptor 3-induced NF-κB activation. Nat. Immunol. 5: 503-507.
    OpenUrlCrossRefPubMed
  9. ↵
    Oshiumi, H., M. Sasai, K. Shida, T. Fujita, M. Matsumoto, T. Seya. 2003. TIR-containing adapter molecule (TICAM)-2, a bridging adapter recruiting to Toll-like receptor 4 TICAM-1 that induces interferon-β. J. Biol. Chem. 278: 49751-49762.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Fitzgerald, K. A., D. C. Rowe, B. J. Barnes, D. R. Caffrey, A. Visintin, E. Latz, B. Monks, P. M. Pitha, D. T. Golenbock. 2003. LPS-TLR4 signaling to IRF-3/7 and NF-κB involves the Toll adapters TRAM and TRIF. J. Exp. Med. 198: 1043-1055.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Fitzgerald, K. A., E. M. Palsson-McDermott, A. G. Bowie, C. A. Jefferies, A. S. Mansell, G. Brady, E. Brint, A. Dunne, P. Gray, M. T. Harte, et al 2001. Mal (MyD88-adapter-like) is required for Toll-like receptor-4 signal transduction. Nature 413: 78-83.
    OpenUrlCrossRefPubMed
  12. ↵
    Horng, T., G. M. Barton, R. Medzhitov. 2001. TIRAP: an adapter molecule in the Toll signaling pathway. Nat. Immunol. 2: 835-841.
    OpenUrlCrossRefPubMed
  13. ↵
    O’Neill, L. A., K. A. Fitzgerald, A. G. Bowie. 2003. The Toll-IL-1 receptor adaptor family grows to five members. Trends Immunol. 24: 286-290.
    OpenUrlCrossRefPubMed
  14. ↵
    Brint, E. K., D. Xu, H. Liu, A. Dunne, A. N. McKenzie, L. A. O’Neill, F. Y. Liew. 2004. ST2 is an inhibitor of interleukin 1 receptor and Toll-like receptor 4 signaling and maintains endotoxin tolerance. Nat. Immunol. 5: 373-379.
    OpenUrlCrossRefPubMed
  15. ↵
    Janssens, S., K. Burns, J. Tschopp, R. Beyaert. 2002. Regulation of interleukin-1- and lipopolysaccharide-induced NF-κB activation by alternative splicing of MyD88. Curr. Biol. 12: 467-471.
    OpenUrlCrossRefPubMed
  16. ↵
    Kobayashi, K., L. D. Hernandez, J. E. Galan, C. A. Janeway, Jr, R. Medzhitov, R. A. Flavell. 2002. IRAK-M is a negative regulator of Toll-like receptor signaling. Cell 110: 191-202.
    OpenUrlCrossRefPubMed
  17. ↵
    Blagoev, B., I. Kratchmarova, S. E. Ong, M. Nielsen, L. J. Foster, M. Mann. 2003. A proteomics strategy to elucidate functional protein-protein interactions applied to EGF signaling. Nat. Biotechnol. 21: 315-318.
    OpenUrlCrossRefPubMed
  18. ↵
    Bouwmeester, T., A. Bauch, H. Ruffner, P. O. Angrand, G. Bergamini, K. Croughton, C. Cruciat, D. Eberhard, J. Gagneur, S. Ghidelli, et al 2004. A physical and functional map of the human TNF-α/NF-κB signal transduction pathway. Nat. Cell. Biol. 6: 97-105.
    OpenUrlCrossRefPubMed
  19. ↵
    Chen, X., L. M. Smith, E. M. Bradbury. 2000. Site-specific mass tagging with stable isotopes in proteins for accurate and efficient protein identification. Anal. Chem. 72: 1134-1143.
    OpenUrlCrossRefPubMed
  20. ↵
    Zhu, H., S. Pan, S. Gu, E. M. Bradbury, X. Chen. 2002. Amino acid residue specific stable isotope labeling for quantitative proteomics. Rapid Commun. Mass Spectrom. 16: 2115-2123.
    OpenUrlCrossRefPubMed
  21. ↵
    Ong, S. E., B. Blagoev, I. Kratchmarova, D. B. Kristensen, H. Steen, A. Pandey, M. Mann. 2002. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol. Cell. Proteomics 1: 376-386.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Orr, S. L., P. Tobias. 2000. LPS and LAM activation of the U373 astrocytoma cell line: differential requirement for CD14. J. Endotoxin Res. 6: 215-222.
    OpenUrlCrossRef
  23. ↵
    Lee, Y. H., H. D. Campbell, M. R. Stallcup. 2004. Developmentally essential protein flightless I is a nuclear receptor coactivator with actin binding activity. Mol. Cell. Biol. 24: 2103-2117.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    da Silva Correia, J., R. J. Ulevitch. 2002. MD-2 and TLR4 N-linked glycosylations are important for a functional lipopolysaccharide receptor. J. Biol. Chem. 277: 1845-1854.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Gu, S., J. Chen, K. M. Dobos, E. M. Bradbury, J. T. Belisle, X. Chen. 2003. Comprehensive proteomic profiling of the membrane constituents of a Mycobacterium tuberculosis strain. Mol. Cell. Proteomics 2: 1284-1296.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Wang, T., S. Gu, T. Ronni, Y. C. Du, X. Chen. 2005. In vivo dual-tagging proteomic approach in studying signaling pathways in immune response. J. Proteome Res. 4: 941-949.
    OpenUrlCrossRefPubMed
  27. ↵
    Cao, Z., W. J. Henzel, X. Gao. 1996. IRAK: a kinase associated with the interleukin-1 receptor. Science 271: 1128-1131.
    OpenUrlAbstract
  28. ↵
    Chuang, T. H., R. J. Ulevitch. 2004. Triad3A, an E3 ubiquitin-protein ligase regulating Toll-like receptors. Nat. Immunol. 5: 495-502.
    OpenUrlCrossRefPubMed
  29. ↵
    Shioda, T., S. Andriole, T. Yahata, K. J. Isselbacher. 2000. A green fluorescent protein-reporter mammalian two-hybrid system with extrachromosomal maintenance of a prey expression plasmid: application to interaction screening. Proc. Natl. Acad. Sci. USA 97: 5220-5224.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Campbell, H. D., T. Schimansky, C. Claudianos, N. Ozsarac, A. B. Kasprzak, J. N. Cotsell, I. G. Young, H. G. de Couet, G. L. Miklos. 1993. The Drosophila melanogaster flightless-I gene involved in gastrulation and muscle degeneration encodes gelsolin-like and leucine-rich repeat domains and is conserved in Caenorhabditis elegans and humans. Proc. Natl. Acad. Sci. USA 90: 11386-11390.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    Liu, Y. T., H. L. Yin. 1998. Identification of the binding partners for flightless I, a novel protein bridging the leucine-rich repeat and the gelsolin superfamilies. J. Biol. Chem. 273: 7920-7927.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Davy, D. A., E. E. Ball, K. I. Matthaei, H. D. Campbell, M. F. Crouch. 2000. The flightless I protein localizes to actin-based structures during embryonic development. Immunol. Cell. Biol. 78: 423-429.
    OpenUrlCrossRefPubMed
  33. ↵
    Yamin, T. T., D. K. Miller. 1997. The interleukin-1 receptor-associated kinase is degraded by proteasomes following its phosphorylation. J. Biol. Chem. 272: 21540-21547.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Aderem, A., R. J. Ulevitch. 2000. Toll-like receptors in the induction of the innate immune response. Nature 406: 782-787.
    OpenUrlCrossRefPubMed
  35. ↵
    Bosisio, D., N. Polentarutti, M. Sironi, S. Bernasconi, K. Miyake, G. R. Webb, M. U. Martin, A. Mantovani, M. Muzio. 2002. Stimulation of Toll-like receptor 4 expression in human mononuclear phagocytes by interferon-γ: a molecular basis for priming and synergism with bacterial lipopolysaccharide. Blood 99: 3427-3431.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Wald, D., J. Qin, Z. Zhao, Y. Qian, M. Naramura, L. Tian, J. Towne, J. E. Sims, G. R. Stark, X. Li. 2003. SIGIRR, a negative regulator of Toll-like receptor-interleukin 1 receptor signaling. Nat. Immunol. 4: 920-927.
    OpenUrlCrossRefPubMed
  37. ↵
    de Couet, H. G., K. S. Fong, A. G. Weeds, P. J. McLaughlin, G. L. Miklos. 1995. Molecular and mutational analysis of a gelsolin-family member encoded by the flightless I gene of Drosophila melanogaster. Genetics 141: 1049-1059.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Campbell, H. D., S. Fountain, I. S. McLennan, L. A. Berven, M. F. Crouch, D. A. Davy, J. A. Hooper, K. Waterford, K. S. Chen, J. R. Lupski, et al 2002. Fliih, a gelsolin-related cytoskeletal regulator essential for early mammalian embryonic development. Mol. Cell. Biol. 22: 3518-3526.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Davy, D. A., H. D. Campbell, S. Fountain, D. de Jong, M. F. Crouch. 2001. The flightless I protein colocalizes with actin- and microtubule-based structures in motile Swiss 3T3 fibroblasts: evidence for the involvement of PI 3-kinase and Ras-related small GTPases. J. Cell. Sci. 114: 549-562.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    Claudianos, C., H. D. Campbell. 1995. The novel flightless-I gene brings together two gene families, actin-binding proteins related to gelsolin and leucine-rich-repeat proteins involved in Ras signal transduction. Mol. Biol. Evol. 12: 405-414.
    OpenUrlAbstract
  41. ↵
    Goshima, M., K. Kariya, Y. Yamawaki-Kataoka, T. Okada, M. Shibatohge, F. Shima, E. Fujimoto, T. Kataoka. 1999. Characterization of a novel Ras-binding protein Ce-FLI-1 comprising leucine-rich repeats and gelsolin-like domains. Biochem. Biophys. Res. Commun. 257: 111-116.
    OpenUrlCrossRefPubMed
  42. ↵
    Xu, H., H. An, Y. Yu, M. Zhang, R. Qi, X. Cao. 2003. Ras participates in CpG oligodeoxynucleotide signaling through association with Toll-like receptor 9 and promotion of interleukin-1 receptor-associated kinase/tumor necrosis factor receptor-associated factor 6 complex formation in macrophages. J. Biol. Chem. 278: 36334-36340.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section