The CATERPILLER Protein Monarch-1 Is an Antagonist of Toll-like Receptor-, Tumor Necrosis Factor α-, and Mycobacterium tuberculosis-induced Pro-inflammatory Signals*

The CATERPILLER (CLR, also NOD and NLR) proteins share structural similarities with the nucleotide binding domain (NBD)-leucine-rich repeat (LRR) superfamily of plant disease-resistance (R) proteins and are emerging as important immune regulators in animals. CLR proteins contain NBD-LRR motifs and are linked to a limited number of distinct N-terminal domains including transactivation, CARD (caspase activation and recruitment), and pyrin domains (PyD). The CLR gene, Monarch-1/Pypaf7, is expressed by resting primary myeloid/monocytic cells, and its expression in these cells is reduced by Toll-like receptor (TLR) agonists tumor necrosis factor (TNF) α and Mycobacterium tuberculosis. Monarch-1 reduces NFκB activation by TLR-signaling molecules MyD88, IRAK-1 (type I interleukin-1 receptor-associated protein kinase), and TRAF6 (TNF receptor (TNFR)-associated factor) as well as TNFR signaling molecules TRAF2 and RIP1 but not the downstream NFκB subunit p65. This indicates that Monarch-1 is a negative regulator of both TLR and TNFR pathways. Reducing Monarch-1 expression with small interference RNA in myeloid/monocytic cells caused a dramatic increase in NFκB activation and cytokine expression in response to TLR2/TLR4 agonists, TNFα, or M. tuberculosis infection, suggesting that Monarch-1 is a negative regulator of inflammation. Because Monarch-1 is the first CLR protein that interferes with both TLR2 and TLR4 activation, the mechanism of this interference is significant. We find that Monarch-1 associates with IRAK-1 but not MyD88, resulting in the blockage of IRAK-1 hyperphosphorylation. Mutants containing the NBD-LRR or PyD-NBD also blocked IRAK-1 activation. This is the first example of a CLR protein that antagonizes inflammatory responses initiated by TLR agonists via interference with IRAK-1 activation.

whereas NOD2 mediates the recognition of muramyl dipeptide (19,21). These findings support the provocative idea that this family of proteins constitutes intracellular sensors of bacterial products and that mutations within these genes lead to a dysregulated inflammatory response.
In addition to the role of CLR proteins as intracellular cytoplasmic mediators, TLRs in mammals have rapidly emerged as predominant molecules by which the innate immune system senses and responds to microbial pathogens (22,23). There are 13 TLRs that recognize an array of microbial products derived from bacteria, viruses, and fungi (24 -26). TLR signal transduction is initiated by stimulation followed by the formation of an intracellular signaling complex with adapter proteins, the predominant one being MyD88 (27). An early step in TLR signaling is the recruitment of the serine/threonine kinase, IRAK-1, to activated receptor complexes. IRAK-1 activation is regulated by sequential phosphorylation events (28). Hyperphosphorylation of IRAK-1 is important for TLR signal transduction as it results in a decreased affinity for the TLR receptor complex and enables the association of IRAK-1 with the TRAF6 complex, leading to activation of NFB and its functional sequelae (29).
We have recently characterized a CLR gene designated as Monarch-1 (30), also known as Pypaf7 (6). Monarch-1 is expressed primarily by cells of the myeloid lineage, including monocytes and granulocytes. This study shows that Monarch-1 interferes with IRAK-1 function, resulting in the repression of TLR signaling, and thus, represents a novel negative regulator of inflammatory responses.
Primary Cell Isolation and Stimulation-Peripheral blood mononuclear cells (PBMC) were isolated from buffy coats (American Red Cross) using lymphocyte separation media (ICN, Costa Mesa, CA). For adherent cell purification, cells were plated at 1 ϫ 10 8 cells/150 mM plate and allowed to adhere for 2 h at 37°C. Non-adherent cells were removed by washing three times with phosphate-buffered saline and replaced with fresh media. Granulocytes were isolated from the red blood cell pellet after lymphocyte separation media purification after ammonium, chloride, potassium buffer red blood cell lysis. Granulocytes were plated at 2 ϫ 10 7 cells/150 mM plate. Cells were stimulated for 1 h with 200 ng/ml commercial LPS from E. coli (LPS 026:B6; Sigma-Aldrich) or, where noted, protein free, phenol/water-extracted E. coli LPS K235 prepared as described in McIntire et al. (31), 1 g/ml commercial peptidoglycan (Sigma), or 100 ng/ml Pam3Cys (ECM).
Transfection, Immunoprecipitation, and Western Analysis-HEK293T cells were transfected with the indicated plasmid cDNA using FuGENE 6. Cells were lysed in buffer containing 1% Triton X-100, 150 mM NaCl, 50 mM Tris, pH 8, 50 mM NaF, 2 mM EDTA, plus a protease inhibitor mixture (Roche Applied Science). Protein concentrations were determined by Bradford assay (Bio-Rad), and equivalent amounts of cellular extract were used in subsequent immunoprecipitations. Immunoprecipitates were washed four times in lysis buffer and eluted by boiling in Laemmli sample buffer. Samples were fractionated by SDS-PAGE and transferred to nitrocellulose.
Generation of Antibodies Specific for Human Monarch-1 and Expression Analysis-Polyclonal antisera against human Monarch-1 was generated by immunization of rabbits with keyhole limpet hemocyaninconjugated Monarch-1 peptide (RGQREDLVRDTPPGC). The IgG from the polyclonal antiserum was purified with protein G-Sepharose affinity chromatography utilizing the manufacturer's protocols. A mouse monoclonal Monarch-1 antibody was generated from mice immunized with hexahistidine-tagged Monarch-1 (amino acids 1-593) expressed in E. coli and purified on nickel nitrilotriacetic-agarose (Qiagen) using denaturing conditions described in the manufacturer's protocol. THP-1 cells stimulated with purified LPS for the indicated time points were lysed in lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Nonidet P-40) supplemented with Complete protease inhibitor mixture (Roche Applied Science). Protein concentrations were determined by the Bradford assay (Bio-Rad), and equivalent amounts of cellular extract were used in subsequent immunoprecipitations. Immunoprecipitation of Monarch-1 was accomplished utilizing the rabbit anti-Monarch-1 peptide IgG. To control for immunoprecipitation efficiency and recovery, a rabbit polyclonal antibody for actin (I19, Santa Cruz) was included in the immunoprecipitation reaction. Monarch-1 and actin immune complexes were precipitated with protein A-agarose beads. Immunoprecipitates were washed 2 times in lysis buffer and eluted by boiling in 60 l of Laemmli sample buffer. Samples were fractionated by SDS-PAGE and transferred to nitrocellulose. Immunoblots were probed with mouse monoclonal anti-Monarch-1 and HRPconjugated anti-actin antibody (C11, Santa Cruz) and visualized by enhanced chemiluminescence (Pierce).
Monarch-1-specific Small Interference RNA (siRNA) Construction-Wild-type and mutant human Monarch-1 short hairpin RNAs were stably expressed in the human THP-1 monocyte cell line by infection of the HSPG retrovirus containing short hairpin RNA transcription cassettes driven by the H1 RNA promoter and a separate cassette containing green fluorescent protein driven by the phosphoglycerate kinase promoter. The targeted sequences are GTCCATGCTGGCACA-CAAG, and the mutant sequence is GTCCATGCTAACACACAAG (siRNA#1), and GGACATCAACTGTGAGAGG, and the mutant sequence is GGACATCGGCTGTGAGAGG (siRNA#2).
Promoter-Reporter Retroviral Plasmids-pHermes-Bluc was derived from pHRSpuro-GUS (a gift of H. Blau, Stanford University), a retroviral vector bearing a self-inactivating 3Ј long terminal repeat (LTR) that lacks promoter and enhancer elements (37). A fragment of pGL3basic (Promega) bearing the transcription silencer, multicloning site, firefly luciferase-coding sequence was inserted into pHRSpuro-GUS. This yielded pHermes-luc. pHermes-Bluc was constructed by inserting the 5ϫ NFB response element and TATA box from pNFBluc (Stratagene) into pHermes-luc. Transduction with pHermes-Bluc retrovirus results in the integration of a synthetic promoter-reporter gene into the host genome that has no retroviral LTRdriven transcription, because the self-inactivating 3Ј LTR, which lacks promoter and enhancer elements, determines the sequences of both the 5Ј and 3Ј LTRs of the integrated retroviral genome. Reporter luciferase retroviral supernatants were generated as described and used for infection of THP-1 cells. Cells were stimulated for 6 h with LPS followed by luciferase assay (Promega) as per the manufacturer's recommendations.
ELISA-THP-1 Monarch-1 siRNA cell populations were stimulated with purified K235 LPS at 200 ng/ml for 24 h followed by analysis of IL-6 protein expression. The IL-6 cytokine levels in cell supernatants were analyzed by sandwich ELISA using antibody pairs and protocols recommended by R&D Systems. The sensitivity of the assays was 16 pg/ml IL-6.
Mycobacterium tuberculosis Infection-THP-1 cells were infected with M. tuberculosis strain H37Rv. M. tuberculosis was grown to early log phase in Middlebrook 7H9 with oleic acid, albumin, dextrose, catalase media. Colony-forming units/ml were determined by plating on Middlebrook 7H10 with oleic acid, albumin, dextrose, catalase media from actively growing cultures A ϭ 2.5 ϫ 10 8 . Bacilli were washed in 0.5% Tween 80, phosphate-buffered saline. Briefly, 2 ϫ 10 6 THP-1 cells were seeded in 6-well, flat-bottom cell culture plates 1 day before infection and incubated at 37°C. Bacteria were added at a multiplicity of infection of 10 for THP-1 cells. Infection was evaluated by measuring colony-forming units. THP-1 cells were collected at the time points indicated; RNA was made using Trizol reagent (Invitrogen) and evaluated for Monarch-1 or IL-6 expression by real-time PCR.
Cytokine Bead Array-THP-1 Monarch-1 siRNA#1 cell populations were stimulated with purified E. coli K235 LPS at 200 ng/ml, with Pam3Cys at 100 ng/ml or with TNF␣ at 20 ng/ml for 24 h followed by analysis of cytokine protein expression. The cytokine levels in cell supernatants were analyzed by Human Inflammation Cytokine Bead Array using antibody pairs and protocols recommended by R&D Systems.

Expression of Monarch-1 Is Decreased by TLR Signaling in PBMC-
Monarch-1 is found primarily in cells of the myeloid and monocytic lineage (6, 30) that represent a first line of defense against invading microorganisms. These cell types can discriminate among different pathogens via the capacity of their TLRs to sense pathogen-associated molecular patterns (39). This causes a very rapid defensive response including the up-regulation of inflammatory and antimicrobial genes. Because of the highly restricted expression of Monarch-1 to cells of the innate immune system (30), we hypothesized that activation of cells via TLRs may influence the expression level of Monarch-1. To test this, human primary adherent PBMC or granulocytes were isolated and exposed to TLR agonists. After 1 h of treatment with LPS from E. coli, which activates TLR4, or peptidoglycan, which activates TLR2, Monarch-1 expression was significantly reduced in both cell types (Fig. 1, A  and B). Because of the concern of contaminants in commercial preparations of TLR agonists, highly purified LPS was prepared by phenol purification (32). Pam3Cys, the pure synthetic agonist for TLR2, was used in concert to specifically activate TLR2. Analysis of Monarch-1 expression in granulocytes stimulated with phenol-purified LPS and synthetic Pam3Cys for 1 h confirmed that Monarch-1 expression is down-regulated after TLR activation (Fig. 1C). Monarch-1 expression is down-regulated upon exposure of PBMC to TNF␣ (30). To compare the level of down-regulation to that observed for LPS, human PBMC were stimulated for 1 and 6 h with phenol LPS or TNF␣. A similar reduction of Monarch-1 mRNA was observed at 1 h for all treatments (Fig. 1D). At 6 h post-stimulation Monarch-1 expression was further down-regulated by LPS and TNF␣ stimulation. This suggests that Monarch-1 is reduced to similar levels after stimulation of human PBMC with LPS or TNF␣.
In the human monocytic cell line, THP-1, Monarch-1 expression was also down-regulated by exposure to purified LPS for 1 h, and a further reduction was observed at 3 h (Fig. 1E). Additionally, Monarch-1 protein level, measured by immunoblot analysis with antibodies raised against recombinant Monarch-1, was decreased in THP-1 cells after exposure to Pam3Cys over a 6-h time period with respect to untreated cells (Fig. 1F). It was necessary to immunoprecipitate Monarch-1 protein followed by immunoblotting, as endogenous Monarch-1 has been difficult to detect in cell lysates. Taken together these data indicate that Monarch-1 expression is down-regulated after exposure of cells to TLR agonists. This suggests that down-regulation of Monarch-1 may be required for the normal defensive response to microorganisms.

Monarch-1 Down-regulates TLR-and TNF-induced NFB Activation-
The decrease in Monarch-1 expression after TLR and TNF␣ stimulation suggested that the down-regulation of Monarch-1 mRNA expression may be necessary for an antimicrobial response to occur. If this were the case, Monarch-1 would be predicted to exert inhibitory effects on steady-state cells and to initially dampen potentially overzealous pro-inflammatory activities that are activated by TLR agonists or by microbial infection. To test this hypothesis, we co-transfected the HEK293T epithelial cell line with expression vectors for TLR4, MD-2, and CD14 to reconstitute the TLR4 receptor complex. The ELAMluciferase reporter, which can be activated by LPS-induced NFB translocation, was co-transfected together with varying amounts of the Monarch-1 expression vector. Cells were treated with LPS, and ELAMluciferase reporter activity was measured. Transfection of increasing amounts of Monarch-1 reduced LPS-stimulated ELAM reporter activation in a dose-dependent manner ( Fig. 2A). The inhibitory effect of Monarch-1 was observed at both low and high concentrations of LPS but may be suboptimal because of the strong signal provided by overexpression of exogenous TLR4, MD-2, and CD14. To bypass the requirement for overexpression of all three components to reconstitute an LPS-responsive receptor complex, we examined the effect of Monarch-1 on the ability of the TLR signaling proteins MyD88, IRAK-1, and TRAF6 as well as the NFB subunit p65 to transactivate the NFB luciferase reporter. This reporter consists of three copies of the NFB binding site linked to a luciferase gene. MyD88, IRAK-1, TRAF6, and p65 overexpression caused a dramatic induction of NFB activation (Fig. 2B). Co-transfection of a Monarch-1 expression plasmid resulted in decreased NFB activity induced by MyD88, IRAK-1, and TRAF6 but not by p65. This indicates Monarch-1 functions upstream of p65. In the experiments below we show that Monarch-1 interferes with IRAK-1 function. Another report 3 shows that Monarch-1 also interferes with the downstream NFB-inducing kinase, which explains the inhibition of TRAF6-activated NFB by Monarch-1. This is consistent with the data in Fig. 2A, which measured the affect on an LPS-responsive receptor complex. These data indicate that the observed capacity of Monarch-1 to diminish TLR signaling is mediated through its effect on MyD88 and/or IRAK-1 and TRAF6 but not at the level of p65, resulting in reduced NFB activity.  DECEMBER 2, 2005 • VOLUME 280 • NUMBER 48

Monarch-1 Down-regulates TLR and TNFR Responses
To determine whether the negative regulatory effect of Monarch-1 is restricted to the TLR pathway, we next sought to investigate if Monarch-1 could regulate the TNFR pathway. HEK293T cells were transfected with different amounts of Monarch-1 expression vector and either stimulated with TNF␣ or co-transfected with expression vectors for TNFR signaling molecules, TRAF2 or RIP1 (Fig. 2C). Overexpression of Monarch-1 resulted in decreased NFB activity induced by TNF␣, TRAF2, or RIP1. Monarch-1 did not suppress the activation of an HLA-DR luciferase reporter, suggesting that the effects are specific. These data suggest that Monarch-1 down-regulates both TLR and TNFR pathways.
Monarch-1 Associates with IRAK-1 and Down-regulates IRAK-1 Activation-One mechanism by which Monarch-1 could reduce TLR signaling protein-induced NFB activity may be through its association with the signaling complex. It is currently believed that once MyD88 is recruited to the TLR4 complex, and it in turn recruits IRAK-4 to a region between the MyD88 TIR and death domain and IRAK-1 to its death domain (40). IRAK-4 is believed to phosphorylate IRAK-1, leading to autophosphorylation and hyperphosphorylation of IRAK-1 (41). To investigate the mechanism of reduced TLR signaling by Monarch-1, HEK293T cells were co-transfected with expression vectors for Monarch-1 and the TLR signaling proteins MyD88 or IRAK-1 followed by immunoprecipitation. Immunoprecipitation of FLAG-tagged IRAK-1 was performed by using anti-FLAG antibody followed by immunoblotting for HA-tagged Monarch-1 using an anti-HA antibody. Monarch-1 was found to co-precipitate with IRAK-1, indicating that these proteins interact in situ (Fig. 3A). In contrast, no association was detected between Monarch-1 and MyD88 in experiments in which HA-tagged MyD88 was immunoprecipitated using anti-HA antibody, and immunoblots were probed with anti-FLAG antibody to detect FLAG-tagged Monarch-1 (Fig. 3B). Upon IRAK-1 overexpression, both native and phosphorylated forms of IRAK-1 could be visualized by Western analysis, providing a convenient surrogate for IRAK-1 activation in the absence of extracellular stimuli (Fig. 3C, lane 3) (28). Because IRAK-1 activation is regulated by sequential phosphorylation events (28), it is possible to distinguish between the species that is first phosphorylated by IRAK4 (resulting in the appearance of a single band that migrates more slowly than native IRAK-1 during SDS-PAGE) (40,41) and the hyperphosphorylated species that is the consequence of autophosphorylation (28). The appearance of hyperphosphorylated IRAK-1 was nearly eliminated by the presence of Monarch-1, suggesting inhibition of IRAK-1 autophosphorylation (Fig. 3C, lane 2, band III). In contrast, the faster migrating protein bands (Fig. 3C, bands I and II, respectively) corresponding to both native and IRAK4-phosphorylated IRAK-1 were not affected, demonstrating specificity for the inhibition of hyperphosphorylation.
Presence of the NBD Domain Is Associated with Reduced IRAK-1 Activation-Common to all CLR proteins, Monarch-1 consists of three distinct domains. To determine the domain(s) responsible for IRAK-1 association, truncation mutants were constructed that consisted of the N-terminal pyrin domain and the C-terminal LRR domain alone or in combination with the NBD. The analysis of NBD alone was not possible because this domain appeared to be very unstable in the absence of either PyD or LRR (not shown). These truncation mutants (shown in Fig. 4A) were transfected along with recombinant IRAK-1 into HEK293T cells. Protein complexes were then immunoprecipitated with anti-IRAK-1-specific antibodies. These experiments revealed that IRAK-1 association is dependent upon the presence of the NBD region in the presence of either PyD or LRR. Both PyD-NBD and NBD-LRR interacted with IRAK-1 (Fig. 4C). Neither the pyrin domain (Fig. 4B) nor the LRR region (Fig. 4C) alone formed molecular complexes with overexpressed IRAK-1. Incongruent with this, the analysis of IRAK-1 in cellular lysate demonstrated that the reduction of IRAK-1 hyperphosphorylation (Form III) was most dramatic in the presence of the NBD-LRR mutant followed by the PyD-NBD mutants (Fig. 4D).

Monarch-1 Associates with IRAK-1 and Prevents Its Activation in THP-1 Cells-The previous experiment utilized HEK293T cells.
To determine whether Monarch-1 negatively regulates IRAK-1 activation in a more physiologically relevant system, human monocytic THP-1 cells that stably expressed HA-tagged Monarch-1 (THP-Mon) were produced (Fig. 5). Endogenous IRAK-1 containing protein complexes were immunoprecipitated with an IRAK-1-specific antibody, and Western blots were probed with anti-HA to detect co-precipitating Monarch-1. Complex formation between IRAK-1 and Monarch-1 was induced after Pam3Cys treatment (Fig. 5B, top panel, lanes 2-4). In contrast, no Monarch-1 was detected in IRAK-1 immunoprecipitates derived from resting cells, indicating a dependence upon TLR signaling (Fig. 5B, top panel, lane 1). Notably, when these Western blots were re-probed to detect immunoprecipitated IRAK-1, sharply reduced levels of hyperphosphorylated IRAK-1 were detected in THP-Mon cells as compared with their empty vector controls (compare the second panel, Fig. 5, A and B). This supports the data derived from transient transfection experiments in HEK293T cells and confirms a role for Monarch-1 in reducing the accumulation of hyperphosphorylated IRAK-1 in monocytes. These results demonstrate that Monarch-1 associates with IRAK-1 upon TLR stimulation and inhibits the accumulation of hyperphosphorylated IRAK-1. Together these data provide a mechanism by which Monarch-1 inhibits TLR induced NFB activation.

Endogenous Monarch-1 Functions as a Negative Regulator in Myeloid
Cells-The observation that Monarch-1 expression is rapidly decreased after TLR stimulation and that Monarch-1 is a negative regulator of inflammatory gene activity suggest that Monarch-1 may be involved in reducing the inflammatory response to pathogens. Specifically, it may be necessary for innate immune cells to reduce Monarch-1 expression in response to pathogens to allow for an increase in inflammatory gene induction. If this hypothesis were correct, the ablation of Monarch-1 expression in myeloid cells would be predicted to result in increased TLR-induced activation signals. To investigate this, Monarch-1-specific siRNAs were generated to reduce endogenous Monarch-1 expression in the THP-1 myeloid cell line. Because of the difficulty in transfecting human myeloid/monocytic cell lines in general and THP-1 cells specifically, a retroviral based vector was used to express Monarch-1-specific siRNAs (Fig. 6A). Two sets of siRNAs specific for distinct regions of the Monarch-1 transcript (designated as siRNA#1 and siRNA#2) were utilized to control for potential nonspecific affects of siRNA. As further controls, corresponding mutant siRNA with two mutated nucleotides (designated as mutMon#1 and mutMon#2) were generated to ensure that the observed effects of siRNA are specific. Monarch-1 short hairpin RNAs were inserted into the pHSPG retroviral vector. Transcription of siRNAs were driven by the H1 RNA gene promoter situated immediately upstream of the insertion site in the 3Ј LTR. A phosphoglycerate kinase promoter that drives transcription of the enhanced green fluorescent protein permitted transduced cells to be separated by fluores-   DECEMBER 2, 2005 • VOLUME 280 • NUMBER 48 cence-activated cell sorting. Therefore, the THP-1 monocytic cell line was transduced with the siRNA-containing constructs and their corresponding controls. Sorted enhanced green fluorescent protein-positive cells were maintained in bulk culture and used to investigate endogenous Monarch-1 function. Expression of Monarch-1 in the four cell samples were assessed by real-time PCR (Fig. 6B). Compared with their controls, both siRNA#1 and siRNA#2 achieved greater than 70% reduction in endogenous Monarch-1 expression. To assess the level of NFB activity in the Monarch-1 siRNA cells, the siRNA cells and their mutant controls were transduced with a retrovirus that contained the luciferase reporter driven by the NFB-responsive promoter element 5ϫ NFB response element and TATA box from pNFBluc (Stratagene). Cells were stimulated with phenol-purified LPS derived from E. coli K235 for 6 h followed by analysis of NFB luciferase activity. If Monarch-1 were a negative regulator of pro-inflammatory signaling, ablation of the negative regulatory activity of Monarch-1 would result in an increase in NFB activity. Indeed, reduced Monarch-1 expression achieved by the siRNA resulted in a Ͼ1000-fold increase in NFB luciferase reporter Western blots were probed with anti-HA to detect co-precipitating HA-Monarch-1 (top panels). These immunoblots (IB) were then stripped and probed with anti-IRAK-1 antibodies (second panels). The hyperphosphorylated form of IRAK-1 was greatly reduced by the presence of Monarch-1. Interestingly, analysis of cellular lysates indicated that the disappearance of the 80-kDa nonphosphorylated form of IRAK-1 occurs more slowly in the presence of Monarch-1. Actin immunoblots were performed to ensure equivalent levels of protein were used among the samples. p-, phosphorylated. activity compared with the control mutant siRNA (Fig. 6C). Basal levels of NFB reporter activity were higher in the Monarch-1 siRNA cells than in their mutant control cells, suggesting that endogenous Monarch-1 functions to suppress NFB activity in the absence of stimulation. Monarch-1 siRNA#2 cells stimulated with LPS had increased NFB activity, although the levels were less dramatic (data not shown). These data suggest that endogenous Monarch-1 functions to dampen NFB activation.

Monarch-1 Down-regulates TLR and TNFR Responses
Monarch-1 Is a Negative Regulator of IL-6 Cytokine Expression in Response to TLR Agonists-The experiment shown above supports the previous hypothesis that Monarch-1 functions as a negative regulator of TLR4-induced NFB activity as assessed by an NFB reporter assay. To further investigate the function of Monarch-1 in mediating the inflammatory response to pathogens, we tested whether endogenous Monarch-1 could regulate the IL-6 gene induced by TLR activation. Both siRNA#1 and siRNA#2 THP-1 cells were stimulated with E. coli K235 LPS for 6 h followed by analysis of IL-6 mRNA expression by real-time PCR. Stimulation of both siRNA#1 and siRNA#2 (Mon-1) THP-1 cells resulted in a 4-fold increase in IL-6 mRNA expression compared with mutant controls (mutMon-1) (Fig. 7A). To investigate if the increased IL-6 expression were also observed at the protein level, supernatants from LPS-stimulated Monarch-1 siRNA#1 and siRNA#2 THP-1 cells were analyzed for IL-6 cytokine expression by ELISA. Consistent with the mRNA data, both Monarch-1 siRNA-containing lines exhibited increased IL-6 protein expression compared with their mutant controls (Fig. 7B). These data suggest that endogenous Monarch-1 functions as a negative regulator of IL-6 cytokine expression in response to TLR stimulation.
Monarch-1 Is a Negative Regulator of IL-6 Cytokine Expression in Response to M. tuberculosis Infection-The above observation was obtained with a pure TLR agonist. To assess if this observation could be extended to a clinically relevant live human pathogen, cells were infected with M. tuberculosis. M. tuberculosis, the causative agent of human tuberculosis, is the most prevalent and deadly bacterial infectious disease worldwide and affects one-third of the world population (42). Incomplete understanding of the molecular nature of protective immune responses has hampered the development of more effective vaccines and therapies. M. tuberculosis leads to the induction of inflammatory cytokines through the activation of TLRs (43). To determine whether exposure of cells to M. tuberculosis could regulate Monarch-1 expression, THP-1 cells were infected with the virulent H37Rv strain of M. tuberculosis for the indicated time points (Fig. 8A). Monarch-1 expression was reduced after exposure of cells to live M. tuberculosis for 2 h, and the expression was nearly abolished after 4 h of exposure. This further suggests that down-regulation of Monarch-1 may be required for the normal defensive response to microorganisms. Because IL-6 is an important response of the innate immune cells to infection by M. tuberculosis (43), we determined if exposure of Monarch-1 siRNA cells to live bacilli would also result in increased IL-6 cytokine expression. Monarch-1 siRNA THP-1 cells were infected with the virulent M. tuberculosis followed by analysis of IL-6 cytokine expression by real-time PCR (Fig. 8B). The bacilli caused a dramatic enhancement of IL-6 production, but cells containing siRNA#1 or siRNA#2 had much higher expression of IL-6 cytokine compared with both wild-type THP-1 cells and the mutant control siRNA.
Monarch-1 Is a Negative Regulator of TLR2-, TLR4-and TNFR-mediated Cytokine Secretion-To investigate more broadly the function of Monarch-1 in mediating the inflammatory response to TLR agonists as well as TNF␣, we investigated whether endogenous Monarch-1 could regulate pro-inflammatory genes in addition to IL-6. We also investigated if Monarch-1 could mediate inflammatory responses by other TLR agonists. Taking into consideration that Monarch-1 inhibits IRAK-1 function, a downstream signaling protein common to both the TLR2 and TLR4 pathway, we stimulated THP-1 Monarch-1 siRNA cells with E. coli K235 LPS (TLR4) and with the synthetic TLR2 agonist, Pam3Cys, in three separate experiments. To investigate the role of endogenous Monarch-1 in regulating the TNFR pathway, we also stimulated these cells with TNF␣. Stimulation by Pam3Cys in general resulted in higher cytokine levels than that observed after LPS stimulation. Supernatant from stimulated cells was analyzed using the Human Inflammation Cytokine Bead Array (Bio-Rad). In Monarch-1 siRNA cells we observed increased levels of IL-6, IL-1␤, TNF␣ and a slight increase in IL-8 when compared with mutMon-1 control cells (Fig. 9A) upon stimulation with Pam3Cys. Among LPS-treated Monarch-1 siRNA containing cells, an increase in IL-6 and IL-1␤ was observed. TNF␣ and IL-8 were expressed at similar levels in Monarch-1 siRNA cells compared with the mutant control (Fig. 9B). However, TNF␣ production in LPS-treated THP-1 cells was low, and the data should not be over-interpreted. Finally, in response to TNF␣, Monarch-1 siRNA cells produced higher levels of IL-1␤ with a modest increase in IL-8 compared with control cells (Fig. 9C). These data support the observation that Monarch-1 broadly down-regulates TLR signaling by a TLR2 agonist. It also caused a more specific down-regulation of IL-6 and IL-1␤ by a TLR4 agonist. These findings also support the observation that Monarch-1 is a negative regulator of the TNFR receptor pathway. In a separate set of experiments using microarray analysis we determined that other genes were not affected by Monarch-1 including lymphotoxin ␣, lymphotoxin ␤, TGF␤1, CCL1 (data not shown). Taken together, these date imply that Monarch-1 is a broad inhibitor of both TLR and TNFR signal transduction pathways

DISCUSSION
The inflammatory response to infection represents a two-edged sword, whereas a proper pro-inflammatory response is necessary to contain infectious microorganisms and foreign antigens, an overzealous response is detrimental to the host (44). The impact of the latter is clearly demonstrated by the 660,000 cases of sepsis per year with a 20% mortality rate (45). Although an array of positive regulators of inflammation has been discovered, relatively few negative feedback modulators have been identified. The SOCS (suppressor of cytokine signaling) molecules represent a major class of negative regulators that are generally induced by stimulants, including TLR agonists (44). Like SOCS, IRAK-M is also induced by TLR agonists. Studies using the inactive kinase IRAK-M-deficient mice reveal that IRAK-M is a negative regulator of TLR signaling and acts by inhibiting the dissociation of IRAK-1/IRAK-4 from the TLR4/MyD88 complex, resulting in blocking of downstream NFB activation (46). An alternatively spliced form of MyD88, called MyD88s, binds to TLRs but cannot interact with IRAK-4. As a result, phosphorylation of IRAK-1 does not occur, and NFB activation is abrogated (40,47). The cytokine-inducible zinc finger protein called A20 and A20-like proteins including ZNF216 and the zinc finger protein TIZ have been found to inhibit IL-1/TLR4-mediated activation of NFB (48). Recently, several CLR proteins are shown to negatively affect cell activation pathways (49 -51). In this report we show that Monarch-1 is also a negative regulator of pro-inflammatory responses. However, it is the first to exhibit an inhibitory effect on the TLR signaling pathway induced by MyD88/IRAK-1/TRAF6 and specifically on IRAK-1 activation.
The conclusion that Monarch-1 is a CLR protein with an anti-inflammatory function is most convincingly demonstrated by the use of siRNA in myeloid/monocytic cells. Monarch-1 is a negative regulator of both the TLR and TNFR pathways. Because Monarch-1 is the first CLR protein that interferes with both TLR2 and TLR4 activation, we explored the molecular mechanism by which Monarch-1 interferes with this activation. We show that Monarch-1 blocks TLR signaling by association with IRAK-1 and inhibition of IRAK-1 hyperphosphorylation. IRAK-1 activation is regulated by sequential phosphorylation events involving the initial phosphorylation by IRAK4 and subsequent hyperphosphorylation by IRAK-1 itself. The hyperphosphorylation of IRAK-1 is important for TLR signal transduction because it serves to decrease affinity for the TLR receptor complex, which is required for TRAF6 binding and subsequent downstream signal transduction (28). Our results show that Monarch-1 clearly interferes with the hyperphosphorylation of IRAK-1. Consistent with these findings, mitigated IRAK-1 hyperphosphorylation has been observed in endotoxin-tolerant cells, where LPS exposure no longer stimulates NFB activation (52,53). This is thought to be secondary to a failure to recruit MyD88 to the TLR4 receptor complex (53) and the subsequent failure to recruit IRAK-1 appropriately to MyD88 (52). In addition, an inhibitory splice variant of MyD88 has been shown to block phosphorylation of IRAK-1, leading to the inhibition of LPS-induced NFB activation (40). Thus, phosphorylation of IRAK-1 is emerging as a critical regulatory step in the control of TLR signaling and Monarch-1 is an inhibitor of this important pathway.
An emerging theme among a subgroup of CLRs is that they exhibit negative regulatory function. Among these, CLR16.2 is unique in its down-regulation of T cell receptor-mediated cellular activation of NFB, AP-1, and NFAT (49). PAN1 down-regulates NFB and the expression of NFB regulated genes in a monocytic cell line (50). PYPAF3 down-regulates LPS-stimulated IL-1␤ production (51). Both the PAN1 study and the present work relied on the reduction of endogenous CLR by siRNA. Thus, these two reports provide powerful evi- dence that a subgroup of CLRs represents negative regulators of immune and inflammatory activation.
In conclusion, this work supports the hypothesis that Monarch-1 normally functions as a negative regulator of inflammatory activity in resting monocytic/myeloid cells, but its expression is down-regulated by TLR agonists, TNF␣, or whole pathogens. This down-regulation is necessary for proper inflammatory and anti-microbial responses to proceed. The down-regulation of Monarch-1 expression is not specific to M. tuberculosis as we have also observed a decrease of Monarch-1 expression in cells infected with other pathogens (data not shown). Monarch-1 down-regulates LPS-induced NFB reporter activity in cells co-transfected with TLR4 and in cells co-transfected with the TLR/IL-1 receptor signaling mediators MyD88, IRAK-1, and TRAF6. This inhibition is not specific to the TLR pathway as Monarch-1 also inhibits TNFR signaling mediators such as TRAF2 and RIP1. Monarch-1 is likely to inhibit multiple pathways, although a focus on the TLR pathway in this report shows that it interferes with IRAK-1 hyperphosphorylation. Inhibition of Monarch-1 by siRNA in cell types that naturally express it significantly altered NFB activation and proinflammatory cytokine synthesis in response to TLR agonists, TNF␣, and virulent M. tuberculosis. A greater understanding of the ability of Monarch-1 to modulate the immune response to pathogens including M. tuberculosis will enhance the development of more effective vaccines and therapies. Taken together these data provide compelling evidence that Monarch-1 is a novel negative regulator of TLR-and pathogen-mediated pro-inflammatory gene induction through its interaction with IRAK-1 and its inhibition of IRAK-1 hyperphosphorylation