|
|
||||||||
CUTTING EDGE |


* Division of Biological Sciences and University of California San Diego Cancer Center, University of California San Diego, La Jolla, CA 92093; and
Laboratory of Cell Signaling, Graduate School of Pharmaceutical Sciences, University of Tokyo, Tokyo, Japan
| Abstract |
|---|
|
|
|---|
, IFN-
, and a subset of IFN-stimulated genes (ISGs) as a result of viral infection. In addition, bacterial cell wall components such as LPS activate IRF3 in a p38-dependent manner. In this study we show that IRF3-mediated ISG induction by LPS requires the production of reactive oxygen species (ROS) by the NADPH-dependent oxidase NOX4. Furthermore, we present evidence that LPS-mediated ROS production leads to activation of apoptosis-regulating-signal kinase (ASK) 1, a MAPK kinase kinase family member capable of activating the MAP kinase 6/p38 axis. ASK1 kinase activity proved essential for IRF3-mediated ISG induction by LPS. Thus, our results presented here suggest a novel role for ROS and ASK1 in the innate immune response as signaling intermediates in the IRF3 activation pathway. | Introduction |
|---|
|
|
|---|
Significant progress had been made in the identification of the signaling molecules involved in TLR-induced gene expression; however, much of the work focused on the MyD88-dependent activation of NF-
B.
The ubiquitously expressed transcription factor known as IFN regulatory factor (IRF3)4 3 emerged as an important response factor to viral infection. Infection with CMV, Newcastle disease virus, or Sendai virus causes nuclear translocation of IRF3 and cooperative DNA binding with the transcriptional coactivator CREB-binding protein/p300 (5, 6, 7, 8). In addition, we have previously shown that IRF3 activation by LPS occurs in a p38-dependent manner (9). This activation is followed by the induction of a distinct subset of IFN-stimulated regulatory element (ISRE)-containing genes. In the last few years, several components of LPS-mediated IRF3 activation have been identified. IRF3 activation by LPS occurs in a MyD88-independent manner; however, cells deficient in the adaptor molecule Toll/IL-1R domain-containing adaptor inducing IFN-
(TRIF) or the TRIF-related adaptor molecule lack IRF3 activation after TLR4 engagement (10, 11, 12, 13). Two I
B kinase (IKK)-related proteins, IKK
and TANK-binding kinase (TBK) 1, were identified as kinases that phosphorylate IRF3 in response to viral infection or TLR3 stimulation (12, 14). Overexpression of IKK
or TBK1 induces the phosphorylation of IRF3 and activates IFN-inducible genes. Furthermore, analysis of TBK1/ murine embryonic fibroblasts (MEFs) confirmed the role of TBK1 in IRF3-dependent gene expression.
Reactive oxygen species (ROS) include H2O2 and radicals such as superoxide anion (O ![]()
![]()
Apoptosis-regulating signal kinase (ASK) 1 is a proapoptotic MAPK kinase kinase that actives the MAPK kinase (MKK) 3/MKK6-p38 and MKK4/MKK7-JNK kinase pathways in response to oxidative stress, anticancer drugs, growth factor deprivation, or TNF-
. In resting cells ASK1 is bound to thioredoxin (TRX), preventing its activation (18). Under oxidizing conditions, TRX forms a disulfide bridge and dissociates from ASK1, allowing ASK1 to autophosphorylate and become activated. ASK1 regulation by TRX denotes that it is regulated by ROS.
This association, in addition to being an upstream activator of p38, led us to hypothesize that LPS-stimulated ROS production could activate ASK1, leading to the activation of p38, IRF3, and subsequent IFN-stimulated gene (ISG) induction.
| Materials and Methods |
|---|
|
|
|---|
Human U373 astrocytoma cells (U-373 MG), stably transfected with human CD14, were generously provided by Dr. P. S. Tobias (The Scripps Research Institute, La Jolla, CA). Peritoneal macrophages were collected 5 days after i.p. administration of thioglycolate.
Reagents
LPS and allopurinol were from Sigma-Aldrich. N-acetyl-L-cysteine (L-NAC), MK886, rotenone, antimycin A3, and diphenylene iodonium chloride (DPI) were obtained from Calbiochem. IFN-
was a generous gift from Biogen.
Mice
Inducible NO synthase (iNOS), endothelial NO synthase (eNOS), and gp91phox-deficient mice were purchased from The Jackson Laboratory. IRF3/ mice were generously provided by Dr. T. Taniguchi (University of Tokyo, Tokyo, Japan).
RNase protection assay (RPA)
Total RNA was isolated using TRIzol Reagent (Invitrogen Life Technologies). Ten micrograms of RNA and 32P-labeled riboprobes for ISG54, RANTES, and GAPDH were incubated in buffer (4:1 formamide and 5x stock (200 mM PIPES (pH 6.4), 2 M NaCl, and 5 mM EDTA)) overnight at 56°C before digestion with T1 RNase (Invitrogen Life Technologies). Protected fragments were solubilized in 98% formamide plus 10 mM EDTA and subjected to electrophoresis on a 4.5% polyacrylamide/urea gel.
EMSA
EMSAs were performed using a 32P-labeled probe corresponding to the ISG15-ISRE (5'-GATCCATGCCTCGGGAAAGGGAAACCGAAACTGAAGCC-3'). Extracts were incubated with poly(dI-dC) and labeled probes in buffer (40 mM KCl, 20 mM HEPES (pH 7.0), 1 mM MgCl2, 0.1 mM EGTA, 0.5 mM DTT, 4% Ficoll, and 0.02% Nonidet P-40) and resolved by electrophoresis on a 6% Tris-borate-EDTA-polyacrylamide gel.
Western blotting
Cells were lysed in a buffer containing 20 mM HEPES (pH 7.4), 1% Triton X-100, 100 mM NaCl, 50 mM NaF, 10 mM
-glycerophosphate, 1 mM sodium vanadate, and 1 mM PMSF. Cell lysates were resolved by SDS-PAGE, and blots were immunoblotted with the indicated Abs. All blots were developed with HRP-conjugated secondary Abs and ECL (Amersham Biosciences).
Luciferase assay
Cells were transfected with 5x ISRE luciferase and Renilla luciferase under the control of the thymidine kinase promoter using Superfect (Qiagen). Sixteen hours after transfection, cells were stimulated with 1 µg/ml LPS for 24 h. Luciferase activity was measured using a Dual luciferase assay kit (Promega) according to the manufacturers instructions.
Immunofluorescence
Slides were washed with PBS, fixed with 4% paraformaldehyde, permeabilized with PBS plus 0.2% Triton X-100, and blocked with 10% goat serum. After rinsing, and incubation with anti-IRF3 serum in PBS, 0.05% Tween 20, and 3% BSA, cells were incubated with goat anti-rabbit IgG-tetramethylrhodamine isothiocyanate Ab conjugate (Jackson ImmunoResearch Laboratories) and overlaid with Vectashield (Vector Laboratories) after rinsing.
Real-time PCR
ISG54 expression was assayed by real-time PCR using Omniscript (Qiagen), SYBR Green (Applied Biosystems), and the primers for ISG54 (5'- TCTGATTCTGAGGCCTTGCA-3' and 5'-CTTGCTGACC-TCCTCCATTCTC3') and
-actin (5'-ACGGCCAGGTCATCACTATTG-3' and 5'-CAAGAAGGAAGGCTGGAAAAGAG-3').
Small interfering RNA (siRNA) transfection
siRNA against murine NOX4 was purchased from Dharmacon (catalog no. M-058509-00). Cells were transfected with 20 µM siRNA using Lipofectamine 2000 (Invitrogen Life Technologies) and stimulated with LPS 48 h later.
| Results and Discussion |
|---|
|
|
|---|

occurs strictly in a STAT1/STAT2/IRF9-dependent manner, whereas its activation as a consequence of viral infection or LPS stimulation requires the activation of IRF3.
In our earlier studies (9) we noticed that LPS-induced, IRF3-mediated ISG54 induction is substantially delayed when compared with the rapid response elicited by the IFN-
-stimulated Jak/STAT pathway (Fig. 1A). This delayed activation kinetics is reminiscent of the slow activation of hypoxia-inducible factor 1, a transcription factor that is activated by ROS (19). Intriguingly, most stimuli (virus infection, bacterial cell surface contact, or environmental stresses such as DNA damage) that activate IRF3 also induce the production of ROS (20). Thus, we hypothesized that ROS might play a role in the TLR4 pathway leading to IRF3 activation.
|
. As shown in Fig. 1B, pretreatment of cells with L-NAC completely inhibited the LPS-mediated induction of ISG54. To exclude the possibility that L-NAC was acting in a nonspecific, potentially cytotoxic manner, we also examined its effect on ISG54 mRNA levels upon IFN
stimulation. As anticipated, L-NAC failed to exert an inhibitory effect upon the IFN-induced, Jak/STAT-mediated expression of ISG54 (Fig. 1B, lanes 5 and 6), demonstrating its specificity toward LPS-induced IRF3 activation. Similar results were seen when other antioxidants, such as butylated hydroxyanisole or nordihydroguaiaretic acid, were used. Thus, our results support the notion of a critical role for ROS in the LPS-induced activation of IRF3 and, subsequently, ISGs. Numerous enzymatic processes lead to the production of ROS. To define the source of ROS generated after LPS stimulation, we tested a wide array of specific inhibitors directed against different ROS sources such as xanthine oxidase, lipoxygenase, cyclo-oxygenase, and mitochondrial respiratory chain complex I and II for their potential effect on LPS-induced ISG54 expression. However, none of these compounds were able to prevent LPS-mediated ISG induction (Fig. 2, A and B; not shown).
|
The seven members of the NADPH oxidase family (Nox/Duox) share significant homology with gp91phox (22). DPI has been widely used to inhibit NADPH oxidase-dependent, inducible ROS production in phagocytes (23, 24). As shown in Fig. 3A, pretreatment of cells with DPI before stimulation with LPS completely abrogated the induction of ISG54. Importantly, DPI had no effect on the Jak/STAT-dependent activation of ISG54 by IFN
(Fig. 3A).
|
B activation. To test whether this enzyme was required for IRF3-mediated responses, we used siRNA against NOX4 as described in (25) and tested its effect on LPS-mediated induction of ISG54. Indeed, significant inhibition of ISG54 transcription (Fig. 3E) and IRF3 nuclear translocation (Fig. 3F) in response to LPS resulted from the abrogation of NOX4 expression, demonstrating that NOX4 is responsible for the generation of the ROS required for LPS-induced, IRF3-mediated transcription. Next, we decided to explore the connection between free ROS and the phosphorylation events that lead to IRF3-mediated ISG induction. LPS-induced activation of p38 was reported to require ROS production (26) as well as ASK1 (27). We had previously found that IRF3 activation by LPS requires the activity of MKK6 and p38 (9, 28) and therefore hypothesized that ASK1 might be the link between LPS-induced ROS production and activation of the MKK6/p38/IRF3 cascade. Indeed, ASK1/ was recently shown to be significantly less susceptible to LPS-induced septic shock (29).
To examine a potential role of ASK1 in LPS-induced, IRF3-mediated transcription, we first tested whether ectopic expression of the ASK1 inhibitor TRX would attenuate the LPS-induced activation of an ISRE luciferase reporter. Indeed, ectopic expression of TRX caused a dose-dependent decrease in the induction of ISRE luciferase by LPS (Fig. 4A); however, no such inhibition was seen when the constitutively active IRF3-5D mutant was used to activate the reporter construct (data not shown). Although these results were indicative of an ASK1 requirement in LPS-induced ISG induction, we also decided to test whether ASK1 was sufficient to induce the ISRE luciferase. Ectopic expression of WT ASK1 resulted in a dose-dependent increase in luciferase production after LPS stimulation, whereas the kinase-dead ASK1(K709M) mutant failed to promote such a response (Fig. 4B). These findings further indicated a role of ASK1 in the LPS signaling pathway leading to the induction of ISGs.
|
treatment (Fig. 5A).
|
IRF3 activation is a crucial part of the innate immune response as evidenced by the fact that IRF3/ mice fail to produce type I IFNs, as well as other proinflammatory cytokines and chemokines, in response to TLR ligation (31, 32). Recent reports have demonstrated a role for TBK1 and IKK
in IRF3 activation in response to viral infection or dsRNA (12, 14), which requires the phosphorylation of IRF3 on S396 (33). The notion that phosphorylation of this residue occurs in response to LPS is somewhat controversial, offering the possibility that a distinct pathway mediates IRF3 activation via TLR2 and TLR4 ligands. In the present study we show that ISG induction by LPS, which occurs in an IRF3-dependent manner, requires the generation of ROS by the NADPH-dependent oxidase NOX4. Subsequent activation of ASK1 links LPS-induced ROS production to the activation of MKK6 and p38, two kinases that we had previously identified as components of the LPS-induced IRF3 activation cascade. It remains to be determined whether ASK1 is also a contributor to the activation of IRF3 in response to other stimuli such as viral infection or DNA damaging agents.
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported in part by National Institutes of Health Grant R01 AI47182(to M.D.). ![]()
2 E.C. and O.D. contributed equally to this work ![]()
3 Address correspondence and reprint requests to Dr. Michael David, Department of Biology, University of California San Diego, Bonner Hall 3138, 9500 Gilman Drive, La Jolla, CA 92093-0322. E-mail address: midavid{at}ucsd.edu ![]()
4 Abbreviations used in this paper: IRF, IFN regulatory factor; ASK1, apoptosis-regulating signal kinase 1; DPI, diphenylene iodonium chloride; IKK, I
B kinase; ISG, IFN-stimulated gene; eNOS, endothelial NO synthase; iNOS, inducible NO synthase; ISRE, IFN-stimulated regulatory element; L-NAC, N-acetyl-L-cysteine; MEF, murine embryonic fibroblast; MKK, MAPK kinase; ROS, reactive oxygen species; RPA, RNase protection assay; siRNA, small interfering RNA; TBK1, TANK-binding kinase 1; TRIF, Toll/IL-1R domain-containing adaptor inducing IFN-
; TRX, thioredoxin; WT, wild type. ![]()
Received for publication July 1, 2005. Accepted for publication March 17, 2006.
| References |
|---|
|
|
|---|
enhancer in vivo. Mol. Cell 1: 507-518. [Medline]
B involves the toll adapters TRAM and TRIF. J. Exp. Med. 198: 1043-1055.
and TBK1 are essential components of the IRF3 signaling pathway. Nat. Immunol. 4: 491-496. [Medline]
and vascular endothelial growth factor through phosphatidylinositol 3-kinase/Akt pathway and reactive oxygen species. J. Biol. Chem. 277: 31963-31971.
B. J. Immunol. 173: 3589-3593.
/
gene induction. Immunity 13: 539-548. [Medline]
gene expression and endotoxin shock. Biochem. Biophys. Res. Commun. 306: 860-866. [Medline]This article has been cited by other articles:
![]() |
J. P. Luyendyk, J. D. Piper, M. Tencati, K. V. Reddy, T. Holscher, R. Zhang, J. Luchoomun, X. Chen, W. Min, C. Kunsch, et al. A Novel Class of Antioxidants Inhibit LPS Induction of Tissue Factor by Selective Inhibition of the Activation of ASK1 and MAP Kinases Arterioscler. Thromb. Vasc. Biol., August 1, 2007; 27(8): 1857 - 1863. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |