|
|
||||||||
-Dependent Pathway1

,¶
* Unité de Virologie et Immunologie Moléculaires, Unité de Recherche 892, Institut National de la Recherche Agronomique (INRA), Domaine de Vilvert, Jouy-en-Josas, France;
Unité dInfectiologie Animale et Santé Publique, Unité de Recherche 1282, Infectiologie Animale et Santé Publique, INRA, Centre de Recherche de Tours, Nouzilly France;
Johnson and Johnson Pharmaceutical Research and Development, Spring House, PA 19477;
INRA, Unité Mixte de Recherche 1225, Université de Toulouse, Ecole Nationale Vétérinaire Toulouse, Toulouse, France;
¶ Université de Toulouse III Paul Sabatier, Toulouse, France;
|| Institut National de la Santé et de la Recherche Médicale, Unité 563, Centre de Physiopathologie de Toulouse Purpan, Toulouse, France; and
# Department of Pharmacology and Therapeutics, University of Calgary, Calgary, Alberta, Canada
| Abstract |
|---|
|
|
|---|
. In vivo, stimulation of PAR2 using specific agonists protected mice from IAV-induced acute lung injury and death. This effect correlated with an increased clearance of IAV in the lungs associated with increased IFN-
production and a decreased presence of neutrophils and RANTES release in bronchoalveolar fluids. More importantly, the protective effect of the PAR2 agonist was totally abrogated in IFN-
-deficient mice. Finally, compared with wild-type mice, PAR2-deficient mice were more susceptible to IAV infection and displayed more severe lung inflammation. In these mice higher neutrophil counts and increased RANTES concentration but decreased IFN-
levels were observed in the bronchoalveolar lavages. Collectively, these results showed that PAR2 plays a protective role during IAV infection through IFN-
production and decreased excessive recruitment of inflammatory cells to lung alveoli. | Introduction |
|---|
|
|
|---|
on IAV replication (21). The role of PAR2 activation in other cell types and particularly in epithelial cells, one of the primary cell types exposed to different pathogens and to IAV (22, 23), has never been studied in the context of viral infection. Furthermore, the specific role of PAR2 activation in in vivo in models of viral infection has never been addressed. In the present study, we investigated the expression of PAR2 in lung epithelial cells after influenza infection and the effects of its activation on IAV replication as well as on cytokine release by those cells. We report that PAR2 expression is increased after IAV infection and that its activation modulates cytokine release and inhibits virus replication by a mechanism dependent on the release of IFN-
. We demonstrate in vivo a major protective role for PAR2 against influenza infection also mediated by an IFN-
-dependent mechanism. These observations highlight PAR2 as a major component of the host immune response to influenza infection, contributing to a better knowledge of mediators that could be used toward the development of anti-influenza therapies. | Materials and Methods |
|---|
|
|
|---|
The human alveolar type II (A549) and the Madin-Darby canine kidney cell lines used in this study were obtained from the American Type Culture Collection and were grown as previously described (9, 24). The IAV A/PR/8/34 (H1N1), (a gift from G. F. Rimmelzwaan, Erasmus Medical Center, Rotterdam, Netherlands) was grown and produced as previously described (9). The PAR2-activating peptide (SLIGRL-NH2) and the control peptide (LRGILS-NH2) that is inactive on PAR2 were synthesized at the protein synthesis facility of the University of Calgary, Calgary, Alberta, Canada. HPLC and mass spectrometry were used to assess the purity of the peptides.
RT-PCR analysis
A549 cells were infected or not at a multiplicity of infection of 1 at different times postinfection. Total RNA were extracted and RT-PCR analysis was done as previously described (25, 26). The following primers were used for amplification: actin, 5'-ATTCCTATGTGGGCGACGAGGCCCA-3' (forward) and 5'-TTGGCGTACAGGTCTTTGCGGATGTC-3' (reverse); PAR2, 5'-AGAAGCCTTATTGGTAAGGTT-3' (forward) and 5'-AACATCATGACAGGTCGTGAT-3' (reverse); and M2, 5'-AAGACCAATCCTGTCACCTCTGA-3' (forward) and 5'-CAAAGCGTCTACGCTGCAGTCC-3' (reverse). As a positive control template for PAR2 mRNA, we used a cloned cDNA of human PAR2. Quantification of mRNA modification was done using the ImageQuant software.
Flow cytometry analysis
A549 cells were infected or not at a multiplicity of infection of 0.001 for 24 h and flow cytometry experiments were done on permeabilized cells as previously described (27, 28) using the SAM11 anti- PAR2 Ab (Santa Cruz Biotechnology). PAR2 stimulation of A549 cells was performed using 250 µM PAR2-specific activating peptide SLIGRL-NH2 and a flow cytometry experiment was performed 24 h poststimulation.
PAR2 stimulation of A549 cells
Cells were stimulated or not for 2 h with 250 µM PAR2-specific activating peptide SLIGRL-NH2 or LRGILS-NH2 before infection with the IAV A/PR8/34 strain. After the indicated times poststimulation, the amount of RANTES or IFN-
released was analyzed in the culture supernatants by ELISA (R&D Systems) (29, 30). Virus titers were analyzed by classical plaque assays. For blocking experiments, anti-IFN-
immune serum or control irrelevant isotypic immune serum (BD Biosciences) was added to A549 culture dishes at a concentration of 20 µg/ml. Preincubation of the Ab was performed 1 h before infection.
Infection and PAR2 stimulation of mice
Six-week-old C57BL/6 female mice were purchased from Charles River Laboratories and experiments were undertaken under specific pathogen-free conditions at the Institut National de la Recherche Agronomique (INRA) animal care facilities (Jouy-en-Josas, France). All animal experiments were conducted under the authority of a license issued by the Direction des Services Vétérinaires (accreditation no. 78-114). For the determination of lethal and sublethal dose of IAV infection, mice were anesthetized and inoculated intranasally with different PFU of IAV. For PAR2 stimulation experiments, mice were anesthetized every day for 3 days. The first day, anesthetized mice were infected intranasally with 50,000 PFU of A/PR/8/34 in the presence or absence of SLIGRL-NH2 or RLGILS-NH2 at different concentrations (50 µl/mouse). Intranasal treatments with SLIGRL-NH2 or the control peptide LRGILS-NH2 were also repeated at days 2 and 3 after infection. Virus loads were determined by plaque assays 24 and 48 h postinfection in the lungs of sacrificed mice.
Six-week-old IFN-
-deficient mice (background C57BL/6; n = 36) obtained from the INRA "Experimental Infectiology Platform" of Tours, France were anesthetized once and infected intranasally with 50,000 PFU of A/PR/8/34 in the presence or absence of SLIGRL-NH2. C57BL/6J/wild-type mice (n = 20) were used as control. Infected mice were then monitored daily for survival.
PAR2-deficient mice and littermates, originally obtained from Johnson and Johnson Research and Development (Spring House, PA) and bred at the University of Calgary animal care facility, were infected intranasally with 10, 30, or 60 PFU of A/PR/8/34. Infected mice were monitored daily for survival and weight.
May-Grünwald and Giemsa staining
Bronchoalveolar lavage fluid (BALF) was collected in PBS (Invitrogen) supplemented with 1 mM EDTA (Invitrogen). After cytocentrifugation, the percentage of polynuclear neutrophils was determined by counting a total of 500 cells per sample by microscopic examination of May-Grünwald- and Giemsa-stained cytocentrifuge slides.
Lung histology
Lung tissue sections were cut from whole lungs fixed in 10% formalin and embedded in paraffin. Twelve-micrometer-thick sections were taken and stained with H&E for histopathological evaluation as previously described (31).
Statistical analysis
The statistical significance of M2 and PAR2 mRNA up-regulation in IAV-infected cells was analyzed using a Kruskal-Wallis test and a t test. The Mann-Whitney U test was used for statistical significance of viral replication and ELISA experiments as well as for the quantification of neutrophils in bronchoalveolar lavage fluid. A Kaplan-Meier test was used for survival differences in mice. The statistical significance was noted when necessary and tested with a threshold of p < 0.05 (*).
| Results |
|---|
|
|
|---|
To investigate the role of PAR2 in IAV replication, we first determined the influence of A/PR/8/34 infection on the expression of PAR2 in human alveolar A549 epithelial cells infected or not with IAV. RT-PCR analysis showed that PAR2 mRNA was significantly increased in infected compared with uninfected cells by comparing the levels of PAR2 mRNA to actin mRNA used as a standard control (Figs. 1, A and B). The viral matrix M2 mRNA, which was included as a positive control of viral infection, was also increased. In addition, cytometry analysis performed on permeabilized cells indicated that A549 IAV infection was also associated with elevated PAR2 at the level of protein expression compared with uninfected cells (Fig. 1C). To highlight the increased synthesis of PAR2 after infection, we overlaid the fluorescence intensity of PAR2 binding in infected cells on top of that in the uninfected ones (Fig. 1C, right). In contrast, stimulation of A549 cells with the PAR2 agonist peptide had no effect in our conditions on PAR2 synthesis. We concluded that IAV infection increased the expression of PAR2 in vitro in lung epithelial cells.
|
Because PAR2 expression was increased in IAV-infected lung epithelial cells, which are the first targets of IAV (9), we next investigated whether PAR2 activation can modulate IAV replication in those cells. For this purpose, A549 alveolar epithelial cells were infected with IAV and stimulated with the selective PAR2 agonist SLIGRL-NH2 or a control peptide, LRGILS-NH2. When exposed to the PAR2 agonist, IAV-infected cells subsequently produced less virus (Fig. 2A, left) compared with cells that were exposed to the inactive control peptide. We concluded that PAR2 activation leads to decreased virus production in A549-infected cells.
|
were observed at the limit of detection. Infection of A549 epithelial cells (infection-treated control) induced RANTES production but had no effect on IFN-
release. However, stimulation of PAR2 in those cells (infection-treated SLIGRL) significantly inhibited RANTES from 48 to 72 h after IAV infection but induced IFN-
release (Fig. 2, B and C). Interestingly, IFN-
release peaked 8 h postinfection and was transient. Thus, an agonist of PAR2 influences cytokine release in A549-IAV-infected cells.
Because IFN-
is a powerful antiviral molecule (32), we tested whether a link could exist between the induced IFN-
release and the decreased viral production in A549-infected cells exposed to the PAR2 agonist. For this purpose, infected A549 cells stimulated with PAR2 agonists or control peptides were incubated with an anti-IFN-
neutralizing Ab or an irrelevant isotypic IgG control Ab. Results showed that the inhibition of IAV replication by PAR2 activation could be reversed by masking IFN-
with a neutralizing specific Ab but not by exposure to an isotype control Ab (Fig. 2A, right panel). This clearly demonstrates that PAR2-mediated inhibition of IAV replication occurs through an IFN-
-dependent mechanism.
Protection from IAV-induced pathogenesis and death by the PAR2 agonist
To investigate the role of PAR2 in vivo, we first characterized in infected wild-type animals the time course of IAV-induced pathogenesis and death in mice. For this purpose, mice were inoculated intranasally with IAV at different PFU numbers per mice. As expected, increasing numbers of PFU resulted in increased mortality of mice (Fig. 3A). We determined that a dose of 50,000 PFU resulted in 100% mice mortality (lethal dose), whereas a dose of 60 PFU/mouse was sublethal.
|
IAV-induced cytokine release in vivo modulated by PAR2 agonist
We then investigated whether the protective role of PAR2 activation against IAV-induced tissue damage was associated with the inhibition of an excessive inflammatory response. Neutrophil recruitment at the site of infection was evaluated in BALF from infected vs noninfected mice exposed or not to PAR2 agonists. As shown in Fig. 4A, the influx of neutrophils in the BALF from mice infected with IAV and treated with the PAR2 agonist was significantly decreased both at 24 and 48 h after infection compared with infected mice treated with the control peptide LRGILS-NH2. However, 48 h after infection the number of neutrophils in the BALF of IAV-infected mice treated with the PAR2 agonist was significantly increased compared with its level at 24 h. Thus, the PAR2 agonist inhibited but did not abolish IAV recruitment of neutrophils into the lungs.
|
in the BALF from infected vs noninfected mice exposed or not to PAR2 agonists. Naive mice that had received intranasal administration of the PAR2 agonist showed a 3-fold increase in the release of RANTES in their BALF, but no effect was observed on the release of IFN-
compared with mice that were treated with the control peptide (Fig. 4B). Infection of mice with IAV for 12, 24, and 48 h (infection-treated control) also provoked the release of RANTES, with a 3-, 20-, and 7-fold increase, respectively, of RANTES in BALF compared with uninfected mice. This increase was significantly inhibited by PAR2 agonist treatment compared with control peptide-treated mice. No significant increase in IFN-
release was observed after IAV infection in mice treated with the control peptide compared with noninfected mice (Fig. 4B). However, at 24 and 48h after infection the mice treated with the PAR2 agonist showed a significant increase in the release of IFN-
. Our results showed a concomitant increase in IFN-
and an inhibition of RANTES release in BALF fluids from infected/PAR2 stimulated mice compared with infected/control peptide stimulated mice. We concluded that the PAR2 agonist modulates IAV-induced cytokine release in vivo.
Protective role for PAR2 against IAV infection: an IFN-
-dependent mechanism
To investigate the involvement of IFN-
in the overall protection of mice against IAV-induced pathogenesis and death in vivo by the PAR2 agonist, we used IFN-
-deficient mice. First, we characterized the time course of IAV-induced pathogenesis and death in IFN-
-deficient mice (Fig. 5A). We determined that a dose of 50,000 PFU resulted in 100% mice mortality (lethal dose), which was consistent with previous reports showing no significant differences between wild-type and IFN-
-deficient mice in response to IAV infection (33). Thus, we used 50,000 PFU/mouse to investigate the involvement of IFN-
in the protective role of PAR2 against IAV infection. Results showed that, in contrast to wild-type mice that were protected by a PAR2-activating peptide treatment (not shown), treatments with the PAR2 agonist in IFN-
-deficient mice did not protect the mice from death (Fig. 5B, left). Furthermore, the PAR2-activating peptide treatment did not modify the kinetic of weight loss after IAV infection of IFN-
-deficient mice (Fig. 5B, right). We concluded that the absence of IFN-
in mice abolished PAR2-induced protection against IAV-induced pathogenesis and death.
|
To define the role of endogenous activation of PAR2, PAR2-deficient mice and littermates were infected with sublethal doses of IAV (10, 30, or 60 PFU/mouse), and the survival rate and weight loss were observed. Results showed no difference in the survival rate of mice infected with 10 PFU/mouse. However, using 30 or 60 PFU/mouse the survival rates as well as weight loss were decreased in PAR2-deficient mice compared with the wild-type littermates (Figs. 6, A and B). We concluded that a lack of PAR2 changes the susceptibility of mice after IAV infection at a threshold of 10 PFU/mouse. Lungs from PAR2-deficient mice appeared hemorrhagic and damaged compared with lungs from littermates after IAV infection (Fig. 6C). Altogether, these results further demonstrate a protective role for endogenous PAR2 during influenza infection in mice.
|
in the BALF from wild-type vs PAR2-deficient mice. Results showed a significant increase in RANTES but a decrease in IFN-
released in BALF after IAV infection with 30 or 60 PFU/mouse compared with wild-type mice but not with 10 PFU/mouse (Fig. 7B). Altogether, these results indicate that PAR2 naturally protects mice against excessive IAV-induced inflammation and death.
|
| Discussion |
|---|
|
|
|---|
The endogenous activators of PAR2 in the airways are not well characterized. PAR2 expressed by respiratory epithelial cells is susceptible to cleavage by various proteinases of diverse classes and from different sources. The first proteases that maybe involved at an early stage during IAV infection could be proteases derived from epithelial cells such as matriptase, human airway trypsin-like protease, and extra pancreatic tryptic enzymes (19). Then, during the course of IAV infection, immune cells are recruited to and activated at the site of infection. Among those inflammatory cells, mast cells are able to massively release tryptase, which is also a good candidate for PAR2 activation (35). In contrast, other immune cells such as neutrophils can release proteases (cathepsin G, elastase, or proteinase 3) that can disarm PAR2 (19, 36). Thus, the course of IAV infection could also be modulated by endogenous proteases that inactivate PAR2. Finally, at a late stage during IAV infection, lungs can become hemorrhagic and, thus, proteases that come from the blood such as kallikreins, FVIIa, and FXa are also potential endogenous activators of PAR2 (37, 38).
Although the effects of PAR2 in an innate immune response to bacterial infection have been well documented (13, 39, 40), this study is the first to show a protective role for PAR2 in the context of viral infection in vivo. However, the overall role of PAR2 in airway immune responses is not well characterized. In some studies PAR2 activation promotes inflammation (14, 41), whereas Moffat et al. reported that PAR2 activation inhibits airway eosinophilia and hyperresponsiveness (42). Although the experimental approaches were similar, the concentrations of the activating peptide used were different:
2.5 nmol/mouse for Ebeling et al. (41), 5 nmol/mouse for Schmidlin et al. (14), and 800 nmol/mouse for Moffatt and coworkers (42). Possibly, a protective role for PAR2 is only observed at effective concentrations of the PAR2 agonist when PAR2 is strongly activated. High concentrations of the activating peptide may maintain its presence in the cell environment and thus cause sustained activation of the cells. In contrast, acute lung inflammation may be associated with weak or brief activation of PAR2 and inflammation in the airways. In support of this, previous studies in mice have shown that the amount/concentration of the peptide may determine whether PAR2 activation improves or exacerbates inflammation in the airways (43). Our results support these studies and show a protective role for PAR2 in the airways only at higher doses of the peptide agonist (50 nmol/mouse significantly protected the mice compared with 5 nmol/mouse). Thus, PAR2 may turn from destructive to protective in inappropriate inflammatory lung diseases.
Importantly, our results showed that the absence of IFN-
in mice abolished PAR2-induced protection against IAV. Thus, the protection conferred by PAR2 appears to be IFN-
dependent. This is consistent with a recent report showing that the PAR2 agonist enhances IFN-
-protective effects against IAV replication in monocytes (21). This is also consistent with several studies showing that IFN-
plays an important role in recovery from IAV infection by helping to clear the virus (44, 45, 46, 47). The antiviral effect of IFN-
has been unclear since, using IFN-
-deficient mice, Graham et al. showed that IFN-
is not necessary for recovery from IAV infection (33). It is possible that tight regulation of IFN-
release is necessary to efficiently fight IAV infection and that this regulation is among the most limiting factors for antiviral host response. From our results it appears that PAR2 activation could regulate this subtile effect of IFN-
. In this study we have shown the following: 1) PAR2 activation increased IFN-
release in the BALF of mice infected with IAV; and 2) IFN-
release was decreased in the BALF of infected PAR2-deficient mice compared with wild-type mice. From our results, PAR2 seems to trigger a signal that counteracts the IAV-mediated inhibitory signal for IFN-
release. In line with this hypothesis, it was recently shown that IAV abrogates the IFN-
response to evade its antiviral activity (48). Thus, a balance between inhibitory and activating signals could determine the release of IFN-
after IAV infection, and full activation of PAR2 would overcome IAV-induced-inhibition of IFN-
release. This is also in agreement with the potential of A549 cells to produce IFN-
after some but not all pathogen infections (32, 49).
From our studies, treatment of infected mice with the PAR2 agonist modulated cytokine release and virus replication as well as the inflammatory response as manifested by neutrophil infiltration into the lungs. Particularly, it is interesting to note that the PAR2 agonist inhibited but did not abolish the recruitment of neutrophils into the lungs. Thus, our results suggest that the recruitment of neutrophils is delayed by the PAR2 agonist treatment. The recruitment of neutrophils into the BALF is necessary for the clearance of IAV infection, and the depletion of neutrophils before sublethal infections with IAV has been shown to result in uncontrolled virus growth and mortality in mice (50). Although PAR2 activation reduced neutrophil infiltration, this inhibition is not detrimental to the clearance of the virus. In contrast, it seems that PAR2 activation regulates the recruitment of inflammatory cells to maintain it at a level that is sufficient to fight against infection but that avoids the acute injury induced by excessive inflammatory cell recruitment.
Pharmacological activation of PAR2 seems to warrant the induction of a proportionate and efficient inflammatory response to influenza infection. This suggests that PAR2 agonists may be of interest for therapeutic effects against IAV pathogenesis in humans. However, in our experiments the number of repeated stimulations of PAR2 was proportional to the protective effect of PAR2 against IAV pathogenesis (data not shown). Thus, repeated treatment would be necessary for optimal protection against IAV. Because the peptide candidates are small peptides (six amino acids), they should be well tolerated and their intranasal inoculation easy to use. Recent outbreaks of highly pathogenic H5N1 IAV subtypes have raised health-related concerns that a new influenza pandemic will occur in the future. Whether PAR2 agonists also protect against H5N1 is thus of particular interest for new approaches in case of pandemic influenza.
Taken together, these results definitively demonstrate a crucial role for PAR2 activation in immune response to viral infection by IAV and provide new therapeutic potential against IAV pathogenesis.
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by the Canadian Institute of Health Research (to N.V.), the Foundation Bettencourt-Schueller, and the Institut National de la Santé et de la Recherche Médicale-Avenir Program (to N.V.), as well as by the Projet Transversal de Recherche INRA-Institut Pasteur (to B.R.) and the Agence Nationale de la Recherche (ANR) (to N.V. and B.R.). K.K. and E.M. are recipients of fellowships from the ANR and the Region Ile de France, respectively. ![]()
2 Address correspondence and reprint requests to Dr. Béatrice Riteau, Unité de Virologie et Immunologie Moléculaires, Unité de Recherche 892, Institut National de la Recherche Agronomique, Domaine de Vilvert, 78352 Jouy-en-Josas, France. E-mail address: beatrice.riteau{at}jouy.inra.fr ![]()
3 Abbreviations used in this paper: IAV, influenza virus type A; BALF, bronchoalveolar lavage fluid; PAR, protease-activated receptor. ![]()
Received for publication November 7, 2008. Accepted for publication April 6, 2009.
| References |
|---|
|
|
|---|
-inducible effects on human monocytes: role in influenza A infection. J. Immunol. 180: 6903-6910.
receptor: a paradigm for cytokine receptor signaling. Annu. Rev. Immunol. 15: 563-5691. [CrossRef][Medline]
gene. J. Exp. Med. 178: 1725-1732.
-dependent clearance of influenza A virus and protection from consolidating pneumonitis in nitric oxide synthase 2-deficient mice. J. Exp. Med. 188: 1541-1546.
by influenza hemagglutinin-specific CD8 effector T cells influences the development of pulmonary immunopathology. Am. J. Pathol. 158: 119-130.
response in respiratory epithelial cells by disruption of the Jak/Stat pathway. Eur. J. Immunol. 38: 1559-1573. [CrossRef][Medline]
in response to Mycobacterium tuberculosis infection. Immunol. Cell Biol. 85: 229-237. [Medline]Related articles in The JI:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |