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







* Institut Pasteur, Unité de Défense Innée et Inflammation, Paris, France;
Institut National de la Santé et de la Recherche Médicale Unité 874, Paris, France;
Institut Pasteur, Unité Postulante des Hépacivirus, Paris, France; and
Department of Tumor Cell Biology, The Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Although recently challenged by some authors (3, 4), the viral replicative intermediate dsRNA is considered critical for the outcome of influenza infection (reviewed in Refs. 5, 6). Thus, RT-PCR experiments and binding of anti-helical dsRNA Abs to viroplasm from whole cell extracts suggest that true dsRNA accumulates within IAV-infected cells. Moreover, synthetic dsRNA and dsRNA isolated from IAV-infected lungs are each able to induce both the local and systemic cytotoxic effects typical of this viral infection (7, 8). Consistently, we recently used an in vitro approach to establish a role for the host dsRNA-recognition receptor TLR3 in the immune response of lung epithelial cells to IAV (9), and we demonstrated that, in vivo, IAV-TLR3 interaction critically contributes to viral pathology (10).
Nonetheless, additional cellular nucleic acid-recognition proteins have lately been implicated as key sensors of viral infection. These include two caspase recruiting domain (CARD) containing, DExD/H family RNA helicases, i.e., the retinoic acid-inducible gene-I (RIG-I) protein and the melanoma differentiation-associated gene (MDA)-5 protein (also known as helicard) (11, 12). Thus, there are at least two receptor systems in place to detect a viral presence and mount an immune response. These receptors localize to different compartments within a cell and recognize distinct ligands. Indeed, whereas it is established that MDA-5, like TLR3, acts as a dsRNA sensor (13, 14), RIG-I was recently shown to be activated by single-stranded viral genomic RNA bearing a 5'triphosphate end (3, 15). Once triggered by their respective agonist, TLR3, RIG-I, and MDA-5 activate intracellular signaling pathways that may all culminate in the induction of antiviral cytokines such as type I IFNs as well as proinflammatory mediators (11, 12, 16).
These apparent similarities raise a pivotal question as to whether TLR3 and RNA helicases have redundant function and whether their activation leads to similar biological consequences or whether different cellular responses are induced depending on the class of viral nucleic acid receptor triggered. To delineate the respective role of TLR3 vs RIG-I/MDA-5 in the respiratory mucosa infected by IAV, we generated a set of experimental cell systems that allowed us to establish that the sensing of IAV by these receptors differs and has only a partial redundant outcome. Indeed, TLR3 activation critically regulates the induction of a proinflammatory response, whereas RIG-I (but not MDA-5) activation mediates both a type I IFN-dependent antiviral signaling and a proinflammatory response.
| Materials and Methods |
|---|
|
|
|---|
The human bronchial epithelial cell line BEAS-2B was cultured as described previously (9).
Virus preparation
Influenza A/Scotland/20/74 (H3N2) virus was prepared as indicated in Ref. 9 .
Cytokine ELISA
Human IL-8, IL-6, RANTES, and IFN-
concentrations were determined using DuoSet ELISA kits obtained from R&D Systems.
RT-PCR
Total RNA was extracted using a RNeasy kit (Qiagen). Reverse transcriptase was performed using 1 µg of total RNA. PCR was performed using specific primers (Proligo) for human 2'5' OAS (sense, 5'-ACA GCT GAA AGC CTT TTG GA-3'; antisense, 5'-AGA CCC CTT TGG CTT GAG TT-3'), Mx1 (sense, 5'-GTG CAT TGC AGA AGG TCA GA-3'; antisense, 5'-CGG CTA ACG GAT AAG CAG AG-3'), Mx2 (sense, 5'-AAG CAG TAT CGA GGC AAG GA-3'; antisense, 5'-TCG TGC TCT GAA CAG TTT GG-3'), RIG-I (sense, 5'-AGG AAA ACT GGC CCA AAA CT-3'; antisense, 5'-TTT CCC CTT TTG TCC TTG TG-3'), MDA-5 (sense, 5'-GTG CAT GGA GGA GGA ACT GT-3'; antisense, 5'-GTT ATT CTC CAT GCC CCA GA-3'), and mitochondrial antiviral signaling protein (MAVS; also called Cardif, Visa, or IPS-1 (11, 12, 16)) (sense, 5'-GCA GCA GAA ATG AGG AGA CC-3'; antisense, 5'-AAA GGT GCC CTC GGA CTT AT-3'). As an internal control, we used primers for human
-actin (sense, 5'-AAG GAG AAG CTG TGC TAC GTC GC-3'; antisense, 5'-AGA CAG CAC TGT GTT GGC GTA CA-3'). Amplifications were performed in a Peltier thermal cycler apparatus (MJ Research).
Transfection of pulmonary epithelial cells
Stably transfected pZero-hTLR3 or control BEAS-2B cells were generated using either 500 ng of a vector expressing TLR3, from which the Toll/IL-1R domain is deleted and thus encoding a nonfunctional TLR3 molecule (purchased from InvivoGen), or a pcDNA3 vector (Invitrogen Life Technologies), respectively. The procedure used the FuGENE 6 transfection reagent (Roche Molecular Diagnostics), according to the manufacturers instructions. Concerning the reporter gene studies, BEAS-2B cells or pZero-hTLR3 BEAS-2B were transiently transfected with 150 ng of a NF-
B- (provided by Dr. A. Israel, Pasteur Institute, Paris, France), IFN-
-, or an ISG56-luciferase-reporter plasmid (provided by Dr. J. Hiscott, McGill University, Montreal, Canada) and 50 ng of pRSV-
-Gal to control DNA uptake. Cotransfection experiments with plasmids expressing a dominant-negative form of MAVS (a gift from Dr. Z. Chen, University of Texas Southwestern Medical Center, Dallas, TX) or vectors overexpressing a functional form of either MDA-5 (a gift from Dr. S. Goodbourn, University of London, U.K.) or RIG-I were performed by including 200 ng of each plasmid to the previous transfection mixture. After 24 h, cells were left untreated or stimulated with IAV (5 x 104 PFU; multiplicity of infection (MOI) = 1) for 18 h at 37°C. Luciferase activity was measured in cell lysates as described previously (9).
Statistical analysis
The statistical significance of differences between groups was tested using the unpaired Students t test with a threshold of p < 0.05.
| Results and Discussion |
|---|
|
|
|---|
B, but not IFN regulatory factor (IRF)-3-dependent gene expression in human bronchial epithelial cells infected by IAV
Effective antiviral immunity essentially relies on the production of type I IFNs such as IFN-
and IFN-
. The expression of type I IFNs is strictly regulated by the activation of latent transcription factors, including NF-
B and IRF-3 (17). Type I IFNs subsequently activate in an autocrine and paracrine manner the expression of IFN-stimulated genes (ISGs), which further collectively inhibit viral replication and assembly and elicit an antiviral state in the host (17). In contrast, beyond its role in the immune response, NF-
B has a central position in promoting inflammation by stimulating the expression of genes that contribute to the pathogenesis of inflammatory products, including a variety of cytokines (18). As a result, NF-
B represents an important and very attractive therapeutic target for drugs to treat many inflammatory pathologies, including lung diseases (19).
To better understand the specific contribution of the viral nucleic acid sensor TLR3 vs RIG-I/MDA-5 in the pathogenesis as well as the immune response associated to IAV infection, we generated human bronchial epithelial BEAS-2B cells that were stably transfected either with a control plasmid (control cells) or with a vector encoding a dominant-negative, nonfunctional, form of TLR3 (pZero-hTLR3 cells). These cells were infected by 5 x 104 PFU of IAV, a virus amount previously shown to potently activate pulmonary epithelial cells (9), and were further analyzed for the induction of antiviral and inflammatory responses.
Fig. 1 shows that NF-
B-driven luciferase activity was barely detectable in IAV-infected pZero-hTLR3 cells in comparison to a strong reporter signal in control BEAS-2B cells (80 ± 4 and 686 ± 91 relative light units, respectively). The activity of the IFN-
promoter was strongly stimulated in IAV-infected control cells and stimulated to a lower, yet significant (p < 0.004) extent in pZero-hTLR3 cells. By contrast, IAV infection resulted in a similarly high stimulation of the IRF-3-dependent, IFN-independent, ISG56 promoter in both pZero-hTLR3 cells and control BEAS-2B cells (p > 0.05). To support this latter finding, the expression of three known IFN-stimulated antiviral genes (2'5' oligoadenylate synthetase (2'5'OAS) and myxovirus resistance protein (Mx)1 and 2) was analyzed by RT-PCR in the two cell types. The three genes were hardly detectable in noninfected BEAS-2B cells (Fig. 1D) and pZero-hTLR3 cells (data not shown) but were clearly up-regulated after IAV infection, to a comparable expression level. Thus, densitometric analysis of the PCR gels indicated the following gene:
-actin ratio: Mx1 = 0.35 ± 0.10 vs 0.39 ± 0.29, Mx2 = 0.55 ± 0.05 vs 0.48 ± 0.02, and 2'5'OAS = 1.32 ± 0.08 vs 1.08 ± 0.05, in control and pZero-hTLR3 cells, respectively. Altogether, these results show that inactivation of TLR3 in human bronchial epithelial cells strongly impairs the NF-
B-dependent signaling pathway while leaving the IRF-3-dependent signaling pathway rather unaffected. This result is particularly noteworthy in regard to the fact that in myeloid cells, TLR3 strongly regulates both NF-
B as well as IRF-3-dependent signal transduction (11, 20).
|
The foregoing findings are important in regard to the numerous reports suggesting a major link between IAV-induced inflammatory cytokines and chemokines and injury of the lung tissue and disease severity (10, 21, 22). We measured, in the supernatants of pZero-hTLR3 and control cells, the secretion of an inflammatory cytokine whose maximal induction is essentially dependent on NF-
B, i.e., IL-8 (23) and an antiviral cytokine that is regulated mostly by IRF-3 (although additional transcriptional factors may play a role, including NF-
B), i.e., RANTES (24). Fig. 2, A and B, clearly shows that whereas IL-8 release is severely impaired in cells expressing the altered form of TLR3 (1.4 ± 0.1 and 12.3 ± 1.2 ng/ml in pZero-hTLR3 and control cells, respectively, at 24 h after IAV infection; p < 0.001), the secretion of RANTES is only partially inhibited (4.1 ± 0.4 and 7.6 ± 0.2 ng/ml in pZero-hTLR3 and control cells, respectively, at 24 h after IAV infection; p < 0.005). Similar results were obtained when considering the IAV-induced release of IL-6 and IFN-
, two cytokines that are regulated rather like IL-8 and RANTES, respectively (Fig. 2, C and D) (25, 26). The pivotal inflammatory role of TLR3 was confirmed by a still ongoing microarray study that is aimed to evaluate the overall impact of TLR3 on the transcriptome of IAV-infected pulmonary epithelial cells. Indeed, we found that among the genes that are specifically modulated by TLR3, many belong to a group of NF-
B-regulated genes, including GRO
, GRO
, GRO
, and IL-8, but not IRF-3-driven genes such as IP-10, MIG, IFN-
, and I-TAC (data not illustrated).
|
Although TLR3 has emerged as a key sensor of viral dsRNA, RIG-I and MDA-5 may also function as alternative pattern-recognition receptors of viral motifs (11, 12, 16). Guo et al. (27) recently indicated that RIG-I is important for induction of IFN by IAV. Siren et al. (28) reported that IAV does not activate HEK cells transfected by TLR3. This latter finding is markedly contradictory with our previous studies using both complementary in vitro and in vivo techniques that clearly demonstrated the major contribution of TLR3 in IAV sensing (9, 10). The same authors (28) reported that RIG-I and MDA-5 both mediate IAV-induced IFN synthesis. This finding also differs from other works that suggest a selective role for MDA-5 in the antiviral response to picornaviruses (29, 30).
In an attempt to clarify these discordant data, we aimed at dissecting the direct involvement of RIG-I and/or MDA-5 via MAVS in IAV detection. First, expression of those molecules was examined by RT-PCR in unstimulated or 5 x 104 PFU IAV-infected BEAS-2B cells for 24 h. Fig. 3A clearly shows that RIG-I, MDA-5, and MAVS are constitutively expressed in resting bronchial epithelial cells and that IAV challenge strikingly up-regulates the expression of RIG-I and MDA-5 but not that of MAVS. IAV up-regulates RNA helicases expression likely through an IFN feedback loop, as suggested by an experiment using BEAS-2B cells preincubated with either an isotype control Ab or a blocking anti-human type I IFN receptor before infection by IAV (MOI = 1; data not illustrated). The obtained data are consistent with previous works showing that RIG-I is overexpressed by inflammatory or viral stimuli including IFNs or IFN-inducing stimuli such as dsRNA, Sendai, or measles viruses and lipopolysaccharides (31, 32, 33). Next, we evaluated whether MAVS was involved in NF-
B and/or IRF-3 signaling pathways activated by IAV, using NF-
B, IFN-
, and ISG56 luciferase reporter plasmids as well as a vector encoding a dominant-negative form of MAVS (MAVS
CARD) or a control plasmid. In comparison to BEAS-2B cells transfected with this latter vector, the foregoing signal transduction pathways were strongly reduced in cells transfected with MAVS
CARD and infected by IAV (Fig. 3B). We next verified the upstream role of RIG-I or MDA-5 per se in the sensing and cell activation triggered by IAV using cells overexpressing a functional form of either receptor. Interestingly, we found that only RIG-I effectively activates NF-
B and IRF-3-dependent signaling pathways induced by IAV (Fig. 3C). It is of interest to notice in that regard that these results are in fair agreement with a previous study focusing on the mechanisms by which dendritic cells (DCs) are activated by viral stimuli. It showed key differences in the relative importance of signaling mediated by RIG-I vs TLRs in these leukocytes. Thus, the ability to induce antiviral response of conventional DCs was severely impaired in RIG-I/ mice but not MyD88/TRIF/ mice (which lack all TLR signaling). Remarkably, opposite results were observed in plasmacytoid DCs (34).
|
|
| Acknowledgments |
|---|
CARD and MDA-5 plasmids, respectively. We also thank V. Lorin, Dr. N. Escriou, and Prof. S. van der Werf (Unité de Génétique Moléculaire des Virus Respiratoires, Institut Pasteur, Paris, France) for providing the IAV. | Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was partially funded by the Institut Pasteur through a Programme Transversal de Recherche (Grant 186). Part of the work was also generously supported by the Société de Pneumologie de Langue Française. R.L.G. is a recipient of a Bourse Roux awarded by the Institut Pasteur. J.P. and D.V. were financially supported by the French association Vaincre la Mucoviscidose and by the Agence Nationale de la Recherche sur le Sida, respectively. ![]()
2 Address correspondence and reprint requests to Dr. Mustapha Si-Tahar, Unité de Défense Innée et Inflammation, Institut National de la Santé et de la Recherche Médicale Unité 874, Institut Pasteur, 25 rue du Dr. Roux, 75015 Paris, France. E-mail address: sitahar{at}pasteur.fr ![]()
3 Abbreviations used in this paper: IAV, influenza A virus; CARD, caspase recruiting domain; MAVS, mitochondrial antiviral signaling protein; MOI, multiplicity of infection; IRF, IFN regulatory factor; DC, dendritic cell. ![]()
Received for publication November 28, 2006. Accepted for publication January 16, 2007.
| References |
|---|
|
|
|---|
B by Toll-like receptor 3. Nature 413: 732-738. [Medline]
B regulation in the immune system. Nat. Rev. Immunol. 2: 725-734. [Medline]
B is a promising therapeutic target in inflammatory lung disease. Curr. Drug Targets 7: 661-668. [Medline]
B and activator protein-2. J. Immunol. 167: 366-374.
-interferon gene expression. Semin. Cancer Biol. 1: 89-95. [Medline]
enhances influenza A virus-induced expression of antiviral cytokines by activating RIG-I gene expression. J. Virol. 80: 3515-3522. This article has been cited by other articles:
![]() |
V. Balloy, J.-M. Sallenave, Y. Wu, L. Touqui, J.-P. Latge, M. Si-Tahar, and M. Chignard Aspergillus fumigatus-induced Interleukin-8 Synthesis by Respiratory Epithelial Cells Is Controlled by the Phosphatidylinositol 3-Kinase, p38 MAPK, and ERK1/2 Pathways and Not by the Toll-like Receptor-MyD88 Pathway J. Biol. Chem., November 7, 2008; 283(45): 30513 - 30521. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. N. Kalali, G. Kollisch, J. Mages, T. Muller, S. Bauer, H. Wagner, J. Ring, R. Lang, M. Mempel, and M. Ollert Double-Stranded RNA Induces an Antiviral Defense Status in Epidermal Keratinocytes through TLR3-, PKR-, and MDA5/RIG-I-Mediated Differential Signaling J. Immunol., August 15, 2008; 181(4): 2694 - 2704. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Liao, X. Bao, T. Liu, S. Lai, K. Li, R. P. Garofalo, and A. Casola Role of retinoic acid inducible gene-I in human metapneumovirus-induced cellular signalling J. Gen. Virol., August 1, 2008; 89(8): 1978 - 1986. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Pothlichet, M. Chignard, and M. Si-Tahar Cutting Edge: Innate Immune Response Triggered by Influenza A Virus Is Negatively Regulated by SOCS1 and SOCS3 through a RIG-I/IFNAR1-Dependent Pathway J. Immunol., February 15, 2008; 180(4): 2034 - 2038. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Mizgerd Acute Lower Respiratory Tract Infection N. Engl. J. Med., February 14, 2008; 358(7): 716 - 727. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |