Published online before print December 10, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
BDivision of Microbiology, National Institute of Health Sciences, Tokyo, Japan
1Correspondence: Division of Microbiology, National Institute of Health Sciences, 1-18-1 Kamiyoga, Setagaya, Tokyo 158-8501, Japan. E-mail: tanamoto{at}nihs.go.jp
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B in a TNFR-associated factor 6 (TRAF6)-dependent manner. We found in this study, however, that a dominant-negative mutant of TRAF6, lacking the N-terminal RING and zinc-finger domain, did not inhibit IRAK-1-induced activation of NF-B in human embryonic kidney 293 cells, although the TRAF6 mutant strongly suppressed the MyD88-induced activation. The dominant-negative mutant of TRAF6 did not affect the IRAK-1-induced activation, regardless of the expression level of IRAK-1. In contrast, small interfering RNA silencing of TRAF6 expression inhibited MyD88-induced and IRAK-1-induced activation, and supplementation with the TRAF6 dominant-negative mutant did not restore the IRAK-1-induced activation. Expression of IRAK-1, but not MyD88, induced the oligomerization of TRAF6, and IRAK-1 and the TRAF6 dominant-negative mutant were associated with TRAF6. These results indicate that TRAF6 is involved but with different mechanisms in MyD88-induced and IRAK-induced activation of NF-B and suggest that TRAF6 uses a distinctive mechanism to activate NF-B depending on signals.
Key Words: Toll-like receptor • IL-1 receptor • lipopolysaccharide
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B [1
].
MyD88 is known as a universal adaptor molecule that interacts with IL-1R and most of TLRs. MyD88 consists of an N-terminal death domain separated by a short internal linker from a C-terminal TIR domain, which is necessary for the interaction with the TIR domain of TLR/IL-1R. The death domain and the internal linker domain have been implicated in the interaction with IRAK-1 and IRAK-4, respectively [5
]. IRAK-1 consists of an N-terminal death domain, which is involved in the binding of MyD88 [6
], and a central serine/threonine kinase domain. The C-terminal region of IRAK-1 contains three potential TRAF6-binding sites, and mutation of the amino acids (Glu544, Glu587, Glu706) in these sites to alanine greatly reduces activation of NF-B [7
]. The death domain and the internal domain between the death domain and the kinase domain of IRAK-1 are also involved in binding TRAF6. The N-terminal region (death domain and internal domain) and the first half of the C-terminal region are sufficient for IL-1-induced activation of NF-B [8
].
It is known that all of MyD88, IRAK-1, and TRAF6 are involved in TLR/IL-R signaling to activate NF-B. However, it is still enigmatic how these molecules lead to the activation of NF-B [9
]. Polyubiquitination of TRAF6 is reportedly important for TLR/IL-1R signaling [10
]. TRAF6 itself functions, in conjunction with the ubiquitin-conjugating enzyme complex Ubc13-Uev1A, as a ubiquitin ligase that catalyzes the formation of unique Lys63-linked polyubiquitin chains [11
, 12
]. TRAF6 catalyzes Lys63-linked polyubiquitination on TRAF6 itself, and the polyubiquitinated TRAF6 activates NF-B signaling proteins by a proteasome-independent mechanism [11
, 12
]. On the other hand, it has also been reported that oligomerization of TRAF6 induces activation of NF-B [12
, 13
]. However, the relationship between the polyubiquitination and the oligomerization is unknown, and the role of MyD88 and IRAK-1 in these events is still ambiguous. We report here for the first time in our knowledge that TRAF6 distinctively mediates MyD88-induced and IRAK-1-induced activation of NF-B and that only IRAK-1 leads to oligomerization of TRAF6.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasmids
The coding regions of human MyD88 and IB kinase β (IKKβ) were amplified by RT-PCR from total RNA prepared from human spleen (OriGene Technologies, Rockville, MD, USA) and THP-1 cells, respectively. The coding region of human TRAF6 was amplified from a human spleen cDNA library (Clontech, Palo Alto, CA, USA). A plasmid containing human IRAK-1 cDNA was obtained from the Mammalian Gene Collection (http://mgc.nci.nih.gov/). Deletions found in the IRAK-1 plasmid were corrected by PCR-mediated mutagenesis. The coding regions of all of these constructs were subcloned into mammalian expression vectors containing the N-terminal EIAV-tag and FLAG-tag epitope sequences. NF-B-dependent luciferase reporter plasmid pELAM-L was described previously [16
]. All mutant plasmids were created by PCR-mediated mutagenesis, and mutations were confirmed by DNA sequencing.
NF-B reporter assay, RNA interference, immunoprecipitation, and immunoblotting
The NF-B-dependent luciferase reporter assay was performed as described elsewhere [17
]. Briefly, HEK293 cells (2–5x105 cells) were plated in six-well plates and transfected the following day by the calcium phosphate precipitation method with the indicated plasmids plus 0.2 µg pELAM-L and 5 ng phRL-TK (Promega, Madison, WI, USA) for normalization. At 24–32 h after transfection, cellular extracts were prepared by adding a lysis buffer [10 mM HEPES-KOH, pH 7.9, 10 mM KCl, 5 mM EDTA, 40 mM β-glycerophosphate, 0.5% Nonidet P-40 (NP-40), 30 mM NaF, 1 mM Na3VO4, 100 nM okadaic acid] containing a protease inhibitor cocktail (Roche Diagnostics, Mannheim, Germany). Reporter gene activity was measured with a portion of the cellular extract, according to the manufacturer’s (Promega) instruction. To another portion of the cellular extract, anti-FLAG M2-agarose (Sigma-Aldrich) was added, and the mixture was incubated at 4°C for 1 h. The agarose was washed three times with PBS containing 0.5% NP-40, and bound proteins were subsequently eluted from the agarose by incubating with 1% SDS. The resulting supernatant was subjected to SDS-PAGE. Proteins were transferred to a polyvinylidene difluoride membrane (Immobilon-P, Millipore, Bedford, MA, USA) and subjected to immunoblotting with the indicated antibodies. The signals were visualized by using an enhanced chemiluminescence system (GE Healthcare Bio-sciences, Piscataway, NJ, USA). For RNA interference, HEK293 cells (1–3x105 cells) were plated in six-well plates and transfected the following day by the calcium phosphate precipitation method with the indicated amounts of a siRNA oligo. On the following day after the first transfection, reporter plasmids, indicated expression plasmids, and the siRNA oligo were transfected further as described above. The transfected amount of siRNA oligo was normalized by supplementing an unrelated oligo. At 24–32 h after the second transfection, cellular extracts were prepared, and reporter activities were determined as above.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
BB, we examined the effects of a dominant-negative mutant of TRAF6. It is well known that the deletion of the N-terminal RING and zinc-finger domain (aa 1–288) of TRAF6 abolishes the ability of TRAF6 to mediate IL-1- and LPS-induced activation of NF-B [18
] and that the N-terminal deletion mutant acts as a dominant-negative mutant [4
]. Thus, this N-terminal deletion mutant of TRAF6 was expressed with MyD88 or IRAK-1 and measured NF-B-dependent reporter activity in HEK293 cells (Fig. 1
). As expected, the expression of MyD88 activated NF-B, and the coexpression of the TRAF6 deletion mutant inhibited this activation in a dose-dependent manner. However, IRAK-1-induced activation of NF-B was surprisingly unaffected by coexpression of the deletion mutant. On the other hand, MyD88- and IRAK-1-induced activation of NF-B were inhibited by a kinase-dead mutant (K44A) of IKKβ, indicating that the activation induced by MyD88 and IRAK-1 is IKK-dependent. The expression levels of MyD88 and IRAK-1 were not affected by coexpression of the TRAF6 deletion mutant (Fig. 1
, lower panels).
 View larger version (34K): [in a new window] |
Figure 1. A TRAF6 dominant-negative mutant inhibits MyD88-induced but not IRAK-1-induced activation of NF-B. HEK293 cells were transiently transfected with a NF-B-dependent luciferase reporter plasmid and an expression plasmid (0.1 µg) for MyD88 (left panel) or IRAK-1 (right panel) together with a kinase-dead (KD) mutant of IKKβ (K44A) or an increasing amount of a dominant-negative mutant plasmid for TRAF6 (TRAF6-DN; aa 289–522). After 30 h, cellular extracts were subjected to luciferase activity measurements and SDS-PAGE followed by immunoblotting. Values are means ± SEM from three independent experiments.
|
B, the effect of the deletion mutant was examined further by changing the expression levels of MyD88 and IRAK-1 (Fig. 2
). The TRAF6 deletion mutant inhibited MyD88-induced activation of NF-B (Fig. 2A
, right), irrespective of the MyD88 expression level (Fig. 2B)
; however, IRAK-1-induced activation was not affected (Fig. 2A
, left) at any of the IRAK-1 expression levels (Fig. 2B)
.
 View larger version (34K): [in a new window] |
Figure 2. A TRAF6 dominant-negative mutant did not affect IRAK-1-induced activation at any of the IRAK-1 expression levels. HEK293 cells were transiently transfected with a NF-B-dependent luciferase reporter plasmid and an increasing amount of MyD88 (A) or IRAK-1 (B) expression plasmid in the absence ( ) or presence (•) of a dominant-negative mutant plasmid (0.5 µg) for TRAF6 (TRAF6-DN; aa 289–522). After 30 h, cellular extracts were subjected to luciferase activity measurements and SDS-PAGE followed by immunoblotting. Values are means ± SEM from three independent experiments.
|
B require TRAF6B. To confirm this finding, the effects of a TRAF6 siRNA were examined. An increasing amount of a TRAF6 siRNA oligo was transfected into HEK293 cells with MyD88, IRAK-1, or IKKβ, and NF-B-dependent reporter activity was measured. Unexpectedly, IRAK-1-induced (Fig. 3B
) as well as MyD88-induced (Fig. 3A)
activation of NF-B was inhibited by the transfection of the siRNA oligo in a dose-dependent manner. IKKβ-induced activation of NF-B was not significantly affected by the siRNA oligo (Fig. 3C)
, indicating that the inhibition was not nonspecific. The expression levels of MyD88, IRAK-1, and IKKβ were not affected by the TRAF6 siRNA oligo (Fig. 3
, upper panels). Furthermore, another TRAF6 siRNA oligo that targets different regions of TRAF6 mRNA also inhibited MyD88 and IRAK-1-induced activation of NF-B (data not shown). These results indicate that although IRAK-1-induced activation was not inhibited by a TRAF6 dominant-negative mutant, MyD88- and IRAK-1-induced activation of NF-B require TRAF6.
 View larger version (34K): [in a new window] |
Figure 3. A TRAF6 siRNA oligo inhibits MyD88-induced and IRAK-1-induced activation of NF-B. HEK293 cells were transiently transfected with a NF-B-dependent luciferase reporter plasmid and an expression plasmid (0.1 µg) for MyD88 (A), IRAK-1 (B), or IKKβ (C), together with an increasing amount of a TRAF6 siRNA oligo. After 30 h, cellular extracts were subjected to luciferase activity measurements and SDS-PAGE followed by immunoblotting. Values are means ± SEM from six independent experiments.
|
B was not affected by an N-terminal deletion mutant of TRAF6 but required TRAF6, the possibility that the C-terminal portion of TRAF6 is involved in IRAK-1-induced activation of NF-B still remains. Thus, the effect of the TRAF6 N-terminal deletion mutant was evaluated after siRNA silencing of endogenous TRAF6 (Fig. 4
). The transfection of a TRAF6 siRNA oligo did not affect basal NF-B-dependent reporter activity. Coexpression of TRAF6, but not expression of the N-terminal deletion mutant, increased reporter activity (Fig. 4
, left panel). The significant increase in reporter activity observed upon expression of MyD88 was inhibited by the transfection of the TRAF6 siRNA oligo as shown above. Under this condition, coexpression of TRAF6, but not the TRAF6 N-terminal deletion mutant, overcame the inhibition induced by TRAF6 siRNA (Fig. 4
, middle panel). IRAK-1-induced activation of NF-B was also inhibited by TRAF6 siRNA, and coexpression of TRAF6 again overcame the inhibition. However, coexpression of the N-terminal deletion mutant was not able to overcome the inhibition induced by TRAF6 siRNA (Fig. 4
, right panel). TRAF6, the N-terminal deletion mutant of TRAF6, IRAK-1, and MyD88, were properly expressed (Fig. 4
, upper panels). Thus, it is unlikely that the C-terminal portion of TRAF6 is capable of transmitting IRAK-1-induced activation of NF-B.
 View larger version (48K): [in a new window] |
Figure 4. C-terminal portion of TRAF6 is not involved in IRAK-1-induced activation of NF-B. HEK293 cells were transiently transfected with a NF-B-dependent luciferase reporter plasmid and a control vector (left panel), an expression plasmid (0.1 µg) for MyD88 (middle panel), or IRAK-1 (right panel) together with wild-type or a dominant-negative mutant plasmid (0.5 µg) for TRAF6 (TRAF6-DN; aa 289–522) in the absence (left bar) or presence (right three bars) of a TRAF6 siRNA oligo (0.5 pmol). After 30 h, cellular extracts were subjected to luciferase activity measurements and SDS-PAGE followed by immunoblotting. Values are means ± SEM from seven independent experiments.
|
B. The fact that TRAF6 is required for both types of activation suggests that TRAF6 is differentially involved in the activation induced by these molecules. As it has been reported that TRAF6 oligomerization induces activation of NF-B [12
, 13
], the oligomerization of TRAF6 was examined in response to the expression of MyD88 and IRAK-1. HEK293 cells stably expressing FLAG-tagged TRAF6, and EIAV-tagged TRAF6 were transiently transfected with an expression plasmid for EIAV-tagged MyD88 or EIAV-tagged IRAK-1. After preparing cell extracts, FLAG-tagged TRAF6 was immunoprecipitated (Fig. 5
, second panel from the top) with anti-FLAG M2 affinity gel, and coprecipitated EIAV-tagged proteins were detected by immunoblotting (Fig. 5
, top panel). Upon expression of MyD88, a trace amount of MyD88, but no EIAV-tagged TRAF6, was coprecipitated with FLAG-tagged TRAF6. In contrast, EIAV-tagged TRAF6 as well as IRAK-1 were coprecipitated with FLAG-tagged TRAF6 when IRAK-1 was expressed. MyD88, IRAK-1, EIAV-tagged TRAF6 (Fig. 5
, second panel from the bottom), and FLAG-tagged TRAF6 (Fig. 5
, bottom panel) were properly expressed. Therefore, expression of IRAK-1 but not MyD88 induces oligomerization of TRAF6.
 View larger version (61K): [in a new window] |
Figure 5. IRAK-1 but not MyD88 induces oligomerization of TRAF6. HEK293 cells stably expressing FLAG-tagged TRAF6 and EIAV-tagged TRAF6 were transiently transfected with an expression plasmid for EIAV-tagged MyD88 or EIAV-tagged IRAK-1. After 30 h, cellular extracts were prepared, and FLAG-tagged TRAF6 was immunoprecipitated (IP). Precipitated, FLAG-tagged TRAF6 (second panel from top) and coprecipitated EIAV-tagged proteins (top panel) were detected by immunoblotting (IB). Part of each cell extract prepared above was subjected to the detection of EIAV-tagged proteins (second panel from bottom) and FLAG-tagged TRAF6 (bottom panel) by immunoblotting.
|
B and that IRAK-1 but not MyD88 induces TRAF6 oligomerization.
 View larger version (44K): [in a new window] |
Figure 6. IRAK-1 and a TRAF6 dominant-negative mutant were coprecipitated with TRAF6. HEK293 cells stably expressing FLAG-tagged TRAF6 and EIAV-tagged TRAF6 were transiently transfected with an increasing amount of a dominant-negative mutant plasmid for EIAV-tagged TRAF6 (TRAF6-DN; aa 289–522) together with an expression plasmid for EIAV-tagged MyD88 or EIAV-tagged IRAK-1. After 30 h, cellular extracts were prepared, and FLAG-tagged TRAF6 was immunoprecipitated. Precipitated FLAG-tagged TRAF6 (right half of top panel) and coprecipitated EIAV-tagged proteins (right half of bottom panel) were detected by immunoblotting. Part of each cell extract (Cell Ext.) prepared above was subjected to the detection of EIAV-tagged proteins (left half of bottom panel) and FLAG-tagged TRAF6 (left half of top panel) by immunoblotting.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B. MyD88 and IRAK-1 act as adaptor molecules in TLR/IL-1R signaling, and overexpression of each molecule leads to activation of NF-B via their downstream signaling molecule TRAF6 (see ref. [1
]). These observations were confirmed in the experiment in which TRAF6 siRNAs inhibited the activation of MyD88- and IRAK-1-induced NF-B activation (Fig. 3)
. However, we found that a dominant-negative mutant of TRAF6 (N-terminal deletion of aa 1–288) inhibits only MyD88-induced activation (Figs. 1
and 2)
. It is unlikely that IRAK-1 activates NF-B by using the C-terminal portion of TRAF6, as IRAK-1 failed to activate NF-B when endogenous, wild-type TRAF6 was silenced by TRAF6 siRNA, and the N-terminal deletion mutant of TRAF6 was overexpressed (Fig. 4)
. We also found that expression of IRAK-1 but not MyD88 leads to oligomerization of TRAF6 (Fig. 5)
. It has been reported that oligomerization of TRAF6 induces activation of NF-B [12
, 13
]. Thus, oligomerization of TRAF6 is probably involved in the IRAK-1-induced activation of NF-B. The N-terminal deletion mutant of TRAF6 did not inhibit IRAK-1-induced TRAF6 oligomerization. Instead, the mutant formed a complex with the TRAF6 oligomer (Fig. 6)
, indicating that the TRAF6 oligomer consists of more than two molecules of TRAF6. This finding explains why the N-terminal deletion mutant of TRAF6 was not able to inhibit IRAK-1-induced activation of NF-B.
Not only does IRAK-1 induce TRAF6 oligomerization, it is also associated with the TRAF6 oligomer (Fig. 5)
. We found that an IRAK-1 mutant (E544A/E587A/E706A), in which three putative TRAF6-binding sites were mutated, the mutation known to greatly impair the ability to activate NF-B [7
], did not induce oligomerization of TRAF6 (data not shown), suggesting that TRAF6 molecules form a complex through the binding to IRAK-1. Overexpression of IRAK-1 in HEK293 cells appears mainly as two forms on SDS-PAGE (see Fig. 1
), with the slower migrating form recognized as the hyperphosphorylated form of IRAK-1 (see ref. [1
]). Interestingly, the slower migrating form of IRAK-1 was predominantly coprecipitated with the TRAF6 oligomer (see Fig. 6
). It has been reported that IL-1 stimulation leads to hyperphosphorylation of IRAK-1 by autophosphorylation and to association between phosphorylated IRAK-1 and TRAF6 [4
]. Thus, it is likely that autophosphorylated IRAK-1 promotes TRAF6 oligomerization by binding to TRAF6.
Overexpression of MyD88 did not induce detectable TRAF6 oligomerization (Fig. 5)
, although expression led to a strong activation of NF-B (Fig. 1)
. TRAF6 oligomerization in response to TLR/IL-1R stimulation has not been reported. There was also no detectable TRAF6 oligomerization in cells stably expressing FLAG-tagged TRAF6 and EIAV-tagged TRAF6 in response to IL-1, LPS, or Pam3CSK4 stimulation, when these cells were transiently expressed with the IL-1RI/IL-1R accessory protein, CD14/TLR4/myeloid differentiation protein-2 or TLR1/TLR2, respectively, although these stimulations induced a strong activation of NF-B (data not shown). Thus, it is unlikely that TRAF6 oligomerization is required for the activation of NF-B in response to TLR/IL-1R stimulation. TRAF6 reportedly functions, in conjunction with ubiquitin-conjugating enzyme complex Ubc13-Uev1A, as a ubiquitin ligase, and this ubiquitin ligase activity is involved in the activation of NF-B [11
, 12
]. We found that transfection of a Ubc13 siRNA oligo into HEK293 cells inhibited MyD88-induced activation of NF-B (data not shown). Therefore, the ubiquitin ligase activity of TRAF6 may be involved in the MyD88-induced activation. Fukushima et al. [19
] reported that LPS-induced degradation of IB
was severely impaired in macrophages and splenocytes isolated from heterozygous Ubc13+/– mice. However, Yamamoto et al. [20
] reported that Ubc13-deficient B cells, bone marrow macrophages, and embryonic fibroblasts showed almost normal NF-B activation in response to LPS, IL-1β, or a bacterial lipoprotein. Thus, further studies are needed to clarify the role of the ubiquitin ligase activity of TRAF6 in TLR/IL-1R signaling.
It is considered that IRAK-1 lies downstream of MyD88 in the TLR/IL-1R signaling processes. Thus, our result that a dominant-negative mutant of TRAF6 inhibited MyD88-induced, but not IRAK-1-induced, activation of NF-B was surprising. There are two possible explanations for this finding. One is that IRAK-1 is not necessary for TLR/IL-1R signaling. The other is that the activation of NF-B in response to IRAK-1 overexpression is qualitatively different from the activation induced physiologically in response to TLR/IL-1R stimulation. We are not able to exclude the second possibility. However, it has been reported that macrophages from IRAK-1 knockout mice showed only partial impairment of cytokine production and NF-B activation in response to TLR4 stimulation [21
]. In addition, Kawagoe et al. [22
] recently found that the IRAK-1/IRAK-4 double-knockout did not affect macrophage activator lipoprotein peptide-2-induced activation of NF-B and proposed the existence of a TLR-mediated, IRAK-1/IRAK-4-independent signaling pathway. It is possible that another IRAK member, such as IRAK-2, compensates for the lack of IRAK-1/IRAK-4 in this knockout mouse. However, we also found that IRAK-2-induced activation of NF-B was not inhibited by a dominant-negative mutant of TRAF6 (data not shown). Therefore, IRAK-1 may not be involved in TLR/IL-1R signaling. This possibility remains to be studied.
Received September 13, 2007; revised November 9, 2007; accepted November 12, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B and c-Jun N-terminal kinase (JNK) activation diverge at IL-1 receptor-associated kinase (IRAK) Proc. Natl. Acad. Sci. USA 98,4461-4465B Genes Dev. 18,2195-2224B pathway Nat. Cell Biol. 7,758-765[CrossRef][Medline]B kinase complex by TRAF6 requires a dimeric ubiquitin-conjugating enzyme complex and a unique polyubiquitin chain Cell 103,351-361[CrossRef][Medline]B through human Toll-like receptor 4 Infect. Immun. 70,6043-6047This article has been cited by other articles:
 |
|
 |
|
  M. Torroella-Kouri, R. Silvera, D. Rodriguez, R. Caso, A. Shatry, S. Opiela, D. Ilkovitch, R. A. Schwendener, V. Iragavarapu-Charyulu, Y. Cardentey, et al. Identification of a Subpopulation of Macrophages in Mammary Tumor-Bearing Mice That Are Neither M1 nor M2 and Are Less Differentiated Cancer Res., June 1, 2009; 69(11): 4800 - 4809. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
