TAK1-ECSIT-TRAF6 Complex Plays a Key Role in the TLR4 Signal to Activate NF-κB*

Background: ECSIT as a multifunctional protein is involved in TLR signals. However, its regulatory function is not fully characterized. Results: ECSIT forms a high molecular endogenous complex including TAK1 and TRAF6, and that leads to activation of TAK1 and NF-κB. Conclusion: TAK1-ECSIT-TRAF6 complex plays an essential role in TLR4 signals. Significance: This study identifies a new regulatory role of ECSIT in TLR4-mediated signaling. ECSIT (evolutionarily conserved signaling intermediate in Toll pathways) is known as a multifunctional regulator in different signals, including Toll-like receptors (TLRs), TGF-β, and BMP. Here, we report a new regulatory role of ECSIT in TLR4-mediated signal. By LPS stimulation, ECSIT formed a high molecular endogenous complex including TAK1 and TRAF6, in which ECSIT interacted with each protein and regulated TAK1 activity, leading to the activation of NF-κB. ECSIT-knockdown THP-1 (ECSITKD THP-1) cells exhibited severe impairments in NF-κB activity, cytokine production, and NF-κB-dependent gene expression, whereas those were dramatically restored by reintroduction of wild type (WT) ECSIT gene. Interestingly, ECSIT mutants, which lack a specific interacting domain for either TAK1 or TRAF6, could not restore these activities. Moreover, no significant changes in both NF-κB activity and cytokine production induced by TLR4 could be seen in TAK1KD or TRAF6KD THP-1 cells transduced by WT ECSIT, strongly suggesting the essential requirement of TAK1-ECSIT-TRAF6 complex in TLR4 signaling. Taken together, our data demonstrate that the ECSIT complex, including TAK1 and TRAF6, plays a pivotal role in TLR4-mediated signals to activate NF-κB.

Innate immune responses are triggered by the engagement of Toll-like receptors (TLRs) 4 or other pattern recognition recep-tors by various pathogen components, called pathogen-associated molecular patterns (1)(2)(3). TLRs recognize a broad range of microbial pathogens, such as bacteria and viruses, triggering inflammatory and antiviral responses and dendritic cell maturation, which result in the eradication of invading pathogens (4). Individual TLRs interact with different combinations of adaptor proteins and activate various transcription factors, such as nuclear factor (NF)-B, activating protein-1, and interferon regulatory factors (IRFs) (3,4).
ECSIT (evolutionarily conserved signaling intermediate in Toll pathways) had been identified as a cytoplasmic protein interacting specifically with the multiadaptor protein and E3 ubiquitin ligase TNF receptor-associated factor 6 (TRAF6), which participates in both Drosophila and mammalian TLRs signaling pathways that regulate innate immunity (5). In addition, previous reports have shown that ECSIT plays an essential role for the bactericidal activity through mitochondrial reactive oxygen species generation in response to TLR stimulation (6) and functions in Bmp signaling in the nucleus (7). These results indicate that the intracellular localization of ECSIT might be linked with its specific roles (i.e. as a signaling adaptor protein in the cytoplasm (5), as a reactive oxygen species regulator protein in the mitochondria (6,8), and as a cofactor for Bmp signaling in the nucleus (7)).
In this study, we tried to identify the new regulatory function of ECSIT in TLR4 signaling. In order to do that, we utilized a gel filtration column chromatography assay and found that ECSIT forms a high mass signaling complex including TRAF6 and TAK1. To understand the molecular mechanism to be implicated by the complex, we performed biochemical and functional studies by using ECSIT-knockdown (KD), TAK1 KD , or TRAF6 KD THP-1 cells. Our data showed that ECSIT forms a signaling complex with TAK1 and TRAF6 through specific molecular interactions, where TAK1 activity was regulated, and thereby affects downstream cascade signaling for the activation of NF-B. Moreover, we found that the regulatory role of TAK1-ECSIT-TRAF6 complex is critically linked to the production of proinflammatory cytokines, such as IL-6 and IL-1␤, and NF-B-dependent gene expressions induced by TLR4 stimulation, strongly suggesting a pivotal role of TAK1-ECSIT-TRAF6 complex in TLR4mediated signals.
Gel Filtration Chromatography-THP-1 cells were treated with or without the addition of 100 ng/ml LPS (Sigma-Aldrich) for 45 min. Lysates from the cells were prepared and loaded onto a Superose 6 10/300 GL column as described (11). Samples were analyzed by Western blotting for the indicated antibodies.
TAK1 Kinase Assay-Control THP-1 and ECSIT KD THP-1 cells were treated with or without the addition of LPS (100 ng/ml) for different amounts of time. TAK1 kinase activity was measured by a c-TAK1 kinase assay kit (PerkinElmer Life Sci-ences, U-TRF#17), in accordance with the manufacturer's protocol.
Enzyme-linked Immunosorbent Assay-Cell culture supernatants were collected and assayed for cytokines. Cytokine production was measured by an enzyme-linked immunosorbent assay of human IL-1␤ and IL-6 (R&D Systems), according to the manufacturer's protocol.
p65 DNA Binding Assay-Control THP-1, TAK1 KD THP-1, and TRAF6 KD THP-1 cells were transiently transfected with different vectors by using Lipofectamine LRX (Invitrogen) or the Neon transfection system (Invitrogen). 36 h posttransfection, cells were treated with or without the addition of LPS (100 ng/ml) for 60 min. Nuclear proteins were prepared with a CelLytic TM NuCLEAR TM extraction kit (Sigma-Aldrich), in accordance with the manufacturer's protocol. Activities of transcription factors, p65, were determined with a TransAM NF-B transcription factor assay kit (Active Motif North America) according to the manufacturer's instructions.
Microarray Analysis-Control THP-1 and ECSIT KD THP-1 cells were treated with or without the addition of LPS (100 ng/ml)for different lengths of time (1, 3, 5, 7, and 9 h). Total RNA was extracted using TRIzol (Invitrogen) and purified using RNeasy columns (Qiagen) according to the manufacturer's protocol. Microarray analysis and processing of raw intensity data were performed as described (12,13).
Statistical Analysis-In vitro data are presented as means Ϯ S.D. from triplicate samples. Comparisons were statistically tested using Student's t test. p values of Ͻ0.05 or Ͻ0.01 were considered to be statistically significant.

RESULTS
ECSIT Forms a High Signaling Complex Including TAK1 and TRAF6 in Response to LPS Stimulation-To investigate a functional role of ECSIT in the TLR4 signaling, we attempted to isolate the ECSIT complex formed by LPS stimulation using gel filtration column chromatography (11). By LPS stimulation in human monocytic THP-1 cells, the endogenous ECSIT was specifically moved to higher molecular mass fractions (i.e. fractions 14 -16), as compared with those without stimulation (Fig.  1A, without LPS (W/O) versus with LPS (ϩLPS)). To identify ECSIT-associated proteins in the fractions, immunoblot (IB) assays were performed. Consistent with previous findings (5, 6), TRAF6 appeared in significant amounts in fractions 14 -16 ( Fig. 1B, ϩLPS). More interestingly, TAK1, which is a member of the mitogen-activated protein kinase kinase kinase family, associated with, and activated by TRAF6 in a Lys-63-linked polyubiquitin chain-dependent manner (14,15), significantly appeared ( Fig. 1C, fractions [14][15][16]ϩLPS), suggesting that the ECSIT-TRAF6-TAK1 complex might be implicated in the TLR4 signaling.
ECSIT Interacts with and Forms the TAK1-TRAF6 Molecular Complex-To investigate the functional role of ECSIT-TRAF6-TAK1 complex, we first examined the ECSIT complex results from molecular interactions with TRAF6 and TAK1 proteins. Coexpression of FLAG-ECSIT with Myc-TAK1 in HEK293T cells showed that ECSIT specifically interacted with TAK1 ( Fig.  2A, lanes 4). Truncated mutants to TAK1 and ECSIT were generated to understand the ECSIT-TAK1 interaction in detail (Fig. 2B). To identify the interaction site of ECSIT with TAK1, Myc-TAK1 was transiently expressed in HEK293T cells, along with FLAG-ECSIT wild type (WT) and FLAG-ECSIT truncated mutants, and then an immunoprecipitation (IP) assay with anti-Myc antibody was performed. As shown in Fig. 2C, only wild type FLAG-ECSIT containing the C-terminal 300 -431 region, and not other truncated mutants of FLAG-ECSIT, was co-precipitated with Myc-TAK1. Moreover, coexpression of FLAG-ECSIT with Myc-TAK1 WT and Myc-TAK1 truncated mutants in HEK293T cells showed that FLAG-ECSIT specifically co-precipitated with Myc-TAK1 WT and four truncated mutants, except for a Myc-TAK1 1-100 mutant (Fig. 2D). These results strongly suggest that the C-terminal 300 -431 region of ECSIT interacted with the N-terminal 100 -200 region of TAK1, as depicted in Fig. 2E. We further identified the interaction site of ECSIT with TRAF6. Coexpression of HA-TRAF6 with FLAG-ECSIT WT and FLAG-ECSIT truncated mutants in HEK293T cells showed that HA-TRAF6 specifically co-precipitated with FLAG-ECSIT WT and two truncated mutants, FLAG-ECSIT 1-257 and FLAG-ECSIT 1-300, but not with FLAG-ECSIT 1-100 and FLAG-ECSIT 1-200, suggesting that TRAF6 interacted with the 200 -257 region of ECSIT (Fig. 3A). A triple transfection assay in HEK293T cells revealed that FLAG-ECSIT was significantly co-precipitated with Myc-TAK1 and Myc-TRAF6 (Fig. 3B). To verify the specific interactions of TRAF6 or TAK1 to ECSIT, FLAG-ECSIT 257-431 or FLAG-ECSIT 1-257 truncated mutant was transiently expressed into HEK293T cells along with Myc-TAK1 or Myc-TRAF6, and then immunoprecipitation with anti-FLAG antibody was performed. As expected, FLAG-ECSIT 257-431 was significantly precipitated with Myc-TAK1, but not Myc-TRAF6 (Fig. 3C), whereas FLAG-ECSIT 1-257 was precipitated with Myc-TRAF6, but not Myc-TAK1 (Fig. 3D), indicating that ECSIT specifically interacts with TAK1 and TRAF6. More interestingly, endogenous IP assay in fractions 14 -16 with anti-ECSIT antibody revealed that TRAF6 and TAK1 proteins were specifically co-precipitated with ECSIT in a TLR4-dependent manner (Fig. 3C, ϩLPS). These results indicate that ECSIT interacts with and forms the TAK1-TRAF6 complex through the molecular interaction, as depicted in Fig. 3F.  DECEMBER 19, 2014 • VOLUME 289 • NUMBER 51 TAK1-ECSIT-TRAF6 Complex Is Functionally Involved in the TLR4 Signaling-Next, we examined whether the ECSIT-TAK1-TRAF6 complex is required for and functionally involved in TLR4-mediated signaling. Overexpression of WT ECSIT in the HEK293T cells resulted in the enhancement of NF-B reporter activity in a dose-dependent manner (Fig. 4A, WT ECSIT). In contrast, both the ECSIT 1-300 truncated mutant, lacking a TAK1 binding domain, and the ECSIT 257-431 truncated mutant, lacking a TRAF6 binding domain, did not enhance NF-B reporter activity by LPS stimulation (Fig.  4A, ECSIT 1-300 and ECSIT 257-431, respectively). Because TAK1 activity is regulated by TRAF6 in a Lys-63-linked polyubiquitin chain-dependent manner (14,15), we generated ECSIT KD THP-1 cells (Fig. 4B) and then examined whether ECSIT is able to regulate TAK1 activity. Consistent with the previous report (5), the phosphorylation of MEKK1 induced by LPS was significantly attenuated in ECSIT KD THP-1 cells, as compared with that of control THP-1 cells (Fig. 4C, Pho-MEKK1). Interestingly, phosphorylations of TAK1 and IKK␣␤ were decreased in ECSIT KD THP-1 cells (Fig. 4C, Pho-TAK1 and Pho-IKK␣␤). More interestingly, the TAK1 kinase activity was also markedly attenuated in ECSIT KD THP-1 cells, as compared with that of control THP-1 cells (Fig. 4D, ctrl versus ECSIT KD ). We further examined whether TAK1 and TRAF6 in the ECSIT complex are essential in the ECSIT-mediated TLR4 signaling pathway. In order to do that, we generated TAK1-or TRAF6-knockdown THP-1 cells (Fig. 5A, TAK1 KD THP-1 and TRAF6 KD THP-1) and performed NF-B reporter activity and p65 DNA binding activity in response to LPS stimulation. These activities were markedly decreased, compared with control THP-1 cells (Fig. 5, B and C, ctrl versus TAK1 KD or TRAF6 KD in LPS stimulation). These effects eventually led to a critical decrease in the production of proinflammatory cytokines, such as IL-6 and IL-1␤ (Fig. 5, D and E), confirming an essential role of both TAK1 and TRAF6 in TLR4 signaling. Upon overexpression of WT ECSIT, those activities induced by LPS stimulation were markedly enhanced in control THP-1, whereas no significant increases could be detected in TAK1 KD or TRAF6 KD THP-1 cells transfected with WT ECSIT (Fig. 5, B-E, ctrl versus TAK1 KD or TRAF6 KD in LPS stimulation). After 38 h, an IP assay with anti-FLAG antibody was performed, followed by IB with anti-Myc or anti-FLAG antibody. B, truncated mutants to ECSIT (top) or TAK1 (bottom) were generated, as described under "Experimental Procedures." C, HEK293T cells were transiently transfected with mock, Myc-TAK1, FLAG-ECSIT wild type (wt), or FLAG-ECSIT truncated mutants, as indicated. After 38 h, an IP assay with anti-Myc antibody was performed, followed by IB with anti-Myc or anti-FLAG antibody. D, HEK293T cells were transiently transfected with mock, FLAG-ECSIT, Myc-TAK1 WT, or Myc-TAK1 truncated mutants, as indicated. After 38 h, an IP assay with anti-FLAG antibody was performed, followed by IB with anti-Myc or anti-FLAG antibody. E, model of how ECSIT interacts with TAK1 at the molecular level.

ECSIT Regulates TAK1 Activity in TLR4 Signals
These results demonstrate that ECSIT forms the TAK1 and TRAF6 complex through specific molecular interactions, and this complex plays a key role in the activation of NF-B induced by TLR4 stimulation through the regulation of TAK1 activity, leading to activation of IKKs.
ECSIT Functionally Regulates NF-B-dependent Gene Expressions Induced by TLR4 Stimulation-Having shown that ECSIT plays a crucial role for the activation of NF-B through the formation of TAK1 and TRAF6 complex, we finally examined whether the regulatory mechanism is functionally linked to the expression of NF-B-dependent genes. Control THP-1 and ECSIT KD THP-1 cells were treated with or without the addition of LPS for different time periods, and then microarray analysis was performed. To see whether ECSIT is specifically involved in NF-B-dependent gene expression, 26 genes containing the consensus B binding site were selected, and their expression was compared between control and ECSIT KD THP-1 cells treated for different lengths of time in the presence or absence of LPS. Interestingly, their expressions in ECSIT KD THP-1 cells were markedly decreased, compared with control cells (Fig.  6A). To verify the expression of NF-B-dependent genes, we performed a qRT-PCR assay. Even more interestingly, the results revealed that IRF7, IL-1␤, CD44, NF-B2, IER3, IL-8, NF-B1A, and RelB were markedly decreased in the ECSIT KD THP-1 cells, compared with control THP-1 cells (Fig. 6B), strongly suggesting an essential role of ECSIT in the TLR4mediated signal leading to the expression of NF-B-dependent genes. Finally, based on these results, we examined whether the ECSIT complex including TAK1 and TRAF6 proteins is functionally involved in the NF-B-dependent gene expressions induced by TLR4 stimulation. In order to do this, control or ECSIT KD THP-1 cells were transfected with mock, FLAG-ECSIT WT, FLAG-ECSIT 1-300, or FLAG-ECSIT 257-431 vector and then treated with or without LPS (Fig. 7A). The . ECSIT forms an endogenous complex including TAK1 and TRAF6 through specific molecular interactions. A, HEK293T cells were transiently transfected with mock, HA-TRAF6, FLAG-ECSIT wild type (wt), or FLAG-ECSIT truncated mutants, as indicated. After 38 h, an IP assay with anti-HA antibody was performed, followed by IB with anti-HA or anti-FLAG antibody. B, HEK293T cells were transiently transfected with mock, FLAG-ECSIT, Myc-TAK1, or Myc-TRAF6, as indicated. After 38 h, an IP assay with anti-FLAG antibody was performed, followed by IB with anti-Myc, anti-FLAG, anti-TRAF6, or anti-TAK1 antibody. C, HEK293T cells were transiently transfected with mock, FLAG-ECSIT 257-431 truncated mutant, Myc-TAK1, or Myc-TRAF6, as indicated. After 38 h, an IP assay with anti-FLAG antibody was performed, followed by IB with anti-Myc or anti-FLAG antibody. D, HEK293T cells were transiently transfected with mock, FLAG-ECSIT 1-257 truncated mutant, Myc-TAK1, or Myc-TRAF6, as indicated. After 38 h, an IP assay with anti-FLAG antibody was performed, followed by IB with anti-Myc or anti-FLAG antibody. E, endogenous IP of ECSIT from fractions 14 -16 prepared from THP-1 cells treated with LPS ( Fig. 1), followed by IB with to anti-ECSIT, anti-TAK1, or anti-TRAF6 antibody. F, model of TAK1-ECSIT-TRAF6 complex formation at the molecular level. DECEMBER 19, 2014 • VOLUME 289 • NUMBER 51 expressions of NF-B-dependent genes, such as IRF7, IL-1␤, CD44, and NF-B2, were measured by qRT-PCR analysis. As shown in Fig. 7 (B-E), the induction of WT ECSIT in both control and ECSIT KD THP-1 cells resulted in marked increases in the expression of these genes in the presence of LPS stimulation, whereas no significant modulation was detected by the induction of ECSIT 1-300 or ECSIT 257-431 vector in both control and ECSIT KD THP-1 cells. These results strongly suggest that the formation of TAK1-ECSIT-TRAF6 complex plays a pivotal role in the regulation of NF-B-dependent gene expression induced by TLR4 stimulation.

DISCUSSION
Pattern recognition receptors, including TLRs, on innate immune cells trigger the activation of signaling cascades that activate multiple transcriptional factors, including NF-B (1-4). The NF-B activation eventually leads to regulation of NF-B-dependent gene expressions, which play essential roles in mounting immune responses against various pathogens (3,4). In TLR-mediated signals, many proteins are either directly or indirectly involved in the activation of NF-B. Among them, TRAF6 and TAK1 proteins, which are essential molecules upstream of IKKs, play a key role in the NF-B activation. Although a plausible model for how to activate TAK1 in TLRs signaling was recently proposed (16), it is still poorly characterized. Previous studies have shown that TRAF6 as E3 ubiquitin ligase interacts with TAK1 through the TAK1-associating protein TAB2 in response to various stimuli, including TLR ligands, IL-1, and RANK ligand (17)(18)(19). A recent study has Error bars, S.D. from triplicate samples. B, THP-1 cells were infected with lentiviral particles containing shRNA targeting human ECSIT or control lentiviral particles according to the manufacturer's protocol. Control THP-1 (ctrl) and ECSIT KD THP-1 cells were cultured in puromycin-containing medium (4 g/ml) for 2 weeks to select stable clones, and IB with anti-ECSIT or anti-GAPDH antibody was performed to evaluate the knockdown efficacy. C, Western blot assay in control (ctrl) THP-1 and ECSIT KD THP-1 cells with anti-ECSIT, anti-phospho-MEKK1, anti-MEKK1, anti-phospho-TAK1, anti-TAK1, anti-phospho-IKK␣␤, anti-IKK␣, or anti-IKK␤ antibody or anti-GAPDH as a loading control. Control THP-1 and ECSIT KD THP-1 cells were treated with or without the addition of 100 ng/ml LPS for the times indicated. D, control or ECSIT KD THP-1 cells were treated with or without LPS, as indicated, and then the endogenous expressions of TAK1 were analyzed by a Western blotting assay. A TAK1 kinase assay was performed, according to the manufacturer's protocol. As a positive control, active c-TAK1 (2.5 nM) was used in this study. Error bars, S.D. triplicate samples. *, p Ͻ 0.01; **, p Ͻ 0.05.
shown that TRAF6 also interacts with ECSIT, ubiquitinating ECSIT, and that it is thereby implicated in TLR signaling through the regulation of reactive oxygen species production. Because ECSIT has been identified as a novel intermediate capable of bridging TRAF6 to MEKK-1 in Toll/IL-1 signaling (5), we initially assumed in this study that ECSIT, along with TRAF6, might be implicated in the regulation of TAK1 activity in TLR-mediated signaling. Our data provide biochemical and molecular evidence that the TAK1-ECSIT-TRAF6 complex is a key component in the TLR4 signaling pathway and that it mediates NF-B activation and thereby regulates NF-B-dependent gene expression. Our data also provide genetic evidence that ECSIT KD , TAK1 KD , or TRAF6 KD THP-1 cells reveal impairment in the production of proinflammatory cytokines, p65 DNA binding activity, and NF-B-dependent gene expression.
It has been found that the conserved TRAF domain of TRAF6 interacts with ECSIT (5). Additionally, we identified in this study that the TRAF domain of TRAF6 interacted with the 200 -257 region of ECSIT. Importantly, we found that ECSIT also specifically interacted with TAK1 through the 257-431 region of ECSIT, demonstrating that TAK1 and TRAF6 interact with different regions of ECSIT. In terms of functional aspect, the association of ECSIT-TAK1-TRAF6 may be essential in TLR4 signaling. ECSIT KD , TAK1 KD , or TRAF6 KD THP-1 cells revealed severe impairment in NF-B reporter activity, cytokine production, and p65 DNA binding activity induced by TLR4 stimulation. Moreover, two truncated mutants, lacking either the TAK1 binding domain or TRAF6 binding domain of ECSIT, respectively, could not modulate these activities in ECSIT KD THP-1 cells. Additionally, we found that ECSIT KD THP-1 cells exhibited a marked decrease in TAK1 kinase activity induced by TLR4 stimulation, which correlated with impairments in the activation of IKKs and expression of NF-B-dependent genes.
In summary, as depicted in Fig. 7F, upon TLR4 stimulation, ECSIT forms the high mass signaling complex, including TRAF6 and TAK1. In the complex, TAK1 is activated. The activated TAK1 induces the activation of IKK complexes through phosphorylation, leading to the degradation of IB-␣ through phosphorylation and polyubiquitination. The dissociated p65/p50 NF-Bs from IB-␣ translocate into the nucleus, leading to the induction of p65/p50-dependent gene expression. Because several reports have also shown that ECSIT plays a key role in the signaling networks gone awry in Alzheimer disease (20 -22) and cellular oncogenesis (7), we therefore believe that our results will contribute to the understanding of ECSIT-related signals, including TLRs, Bmp, and TGF-␤ signaling, and the pathogenesis of ECSIT-related diseases, such as Alzheimer disease.