Proteinase-activated receptor 2 (PAR2) is widely expressed in the respiratory tract and is an integral component of the host antimicrobial defense system. The principal aim of this study was to investigate the influence of a PAR2-activating peptide, SLIGRL, on influenza A virus (IAV)-induced pathogenesis in mice. Intranasal inoculation of BALB/c mice with influenza A/PR/8/34 virus caused time-dependent increases in the number of pulmonary leukocytes (recovered from bronchoalveolar lavage fluid), marked airway histopathology characterized by extensive epithelial cell damage, airway hyper-responsiveness to the bronchoconstrictor methacholine, and elevated levels of inflammatory chemokines (keratinocyte-derived chemokine and macrophage inflammatory protein 2) and cytokines (interferon-γ). It is noteworthy that these IAV-induced effects were dose-dependently attenuated in mice treated with a PAR2-activating peptide, SLIGRL, at the time of IAV inoculation. However, SLIGRL also inhibited IAV-induced increases in pulmonary leukocytes in PAR2-deficient mice, indicating these antiviral actions were not mediated by PAR2. The potency order obtained for a series of structural analogs of SLIGRL for anti-IAV activity (IGRL > SLIGRL > LSIGRL >2-furoyl-LIGRL) was also inconsistent with a PAR2-mediated effect. In further mechanistic studies, SLIGRL inhibited IAV-induced propagation in ex vivo perfused segments of trachea from wild-type or PAR2(−/−) mice, but did not inhibit viral attachment or replication in Madin-Darby canine kidney cells and chorioallantoic membrane cells, which are established hosts for IAV. In summary, SLIGRL protected mice from IAV infection independently of PAR2 and independently of direct inhibition of IAV attachment or replication, potentially through the activation of endogenous antiviral pathways within the mouse respiratory tract.
Annual epidemics of influenza A virus (IAV) infection cause significant morbidity and mortality, although the potentially devastating impact of these seasonal epidemics can be limited by widespread administration of antigenically matched vaccines. Humans are periodically exposed to antigenically novel transmissible strains of IAV, such as the recent human cases of avian (H5N1) and swine (H1N1) IAV, which can result in pandemics. Unfortunately, vaccines to these unpredictable new viruses can take months to develop, and in the meantime the principal countermeasures to reduce the impact of an influenza pandemic are antiviral medications (Salomon and Webster, 2009). Compounding this problem, increasing resistance is developing to many of the currently used antivirals (Hurt et al., 2009). This combination of events has heightened awareness of the urgent need to develop new antiviral medications to control pandemic influenza infections. New anti-influenza drugs could provide a first line of defense against a pandemic, allowing infection control until sufficient quantities of a suitable vaccine are generated.
Proteinase-activated receptors (PARs) are G protein-coupled receptors, existing as four subtypes (PAR1, PAR2, PAR3, and PAR4) (Macfarlane et al., 2001; Hollenberg and Compton, 2002; Steinhoff et al., 2005). PAR activation involves cleavage of a specific site within the N-terminal region of the receptor, which reveals a new tethered ligand capable of activating the receptor. For example, specific cleavage of the extracellular region of murine PAR2 by proteolytic enzymes, such as trypsin, generates a new amino terminus, …SLIGRL, that acts as a tethered ligand to autoactivate the receptor (Nystedt et al., 1994). Further studies revealed that the six-amino acid peptide SLIGRL alone could fully activate murine PAR2. Indeed, in the ensuing 15 years SLIGRL has been used routinely as a peptidic activator of PAR2. The precise role of PARs in the airways is unclear, although one hypothesis is that during inflammation newly released proteases activate PARs, which regulate the functions of innate immune cells by controlling motility, adhesion, and secretion of inflammatory mediators (Shpacovitch et al., 2007; Dale and Vergnolle, 2008). Thus, PARs can be viewed as an integral component of the host antimicrobial defense system.
Previous studies have revealed elevated levels of PAR2 expression in the airways of IAV-infected mice (Lan et al., 2004) and A549 lung epithelial cells (Khoufache et al., 2009), suggesting a role for PAR2 in IAV disease processes. However, there is considerable debate as to whether the presence of PAR2 in virally infected airways promotes disease progression or resolution. For example, the potential role of PAR2 on IAV infection in vivo has been investigated by using PAR2(−/−) mice, with starkly contrasting findings. Khoufache et al. (2009) reported PAR2(−/−) mice were more susceptible to IAV with increased levels of lung inflammation and higher death rates; that is, the presence of PAR2 played a protective role against IAV. On the other hand, Nhu et al. (2010) reported that PAR2(−/−) mice were less susceptible to IAV with lower death rates; that is, the absence of PAR2 played a protective role against IAV.
The principal aim of this study was to examine the effects of a PAR2-activating peptide SLIGRL on several characteristic features of IAV infection in PAR2(+/+) and PAR2(−/−) mice including lung leukocyte number, airway histopathology, airway hyper-responsiveness, and elevated levels of inflammatory chemokines and cytokines. Additional studies were conducted by using a novel ex vivo perfused tracheal system and traditional cell culture to investigate the cellular mechanisms through which SLIGRL inhibited IAV infection.
Materials and Methods
Mouse-adapted influenza A/PR/8/34 virus was propagated in the allantoic fluid of 9-day-old embryonated chicken eggs (Altona Hatchery, Forrestfield, Australia) at 37°C for 3 days as described previously (Williams and MacKenzie, 1977). Viral infectivity was assessed by using allantois-on-shell titration and quantitated via hemagglutination assay (Fazekas de St Groth and White, 1958). The viral infectivity of the harvested allantoic fluid was 5.62 × 107 EID50/ml. Influenza A/Mem/1/71 (H3N1) was a kind gift from Dr. Graeme Zosky (Telethon Institute for Child Health Research, Subiaco, Western Australia).
Murine IAV Infection Model.
Male BALB/c mice (specified pathogen free) at 8 to 10 weeks of age were housed at the University of Western Australia Animal Care Unit under a 12-h light/dark cycle and received food and water ad libitum. All procedures were approved by the University of Western Australia Animal Ethics Committee. Groups of mice were anesthetized (methoxyflurane) and intranasally inoculated with a 20-μl solution containing influenza A/PR/8/34 virus (470 EID50 dose) alone (virus), IAV plus SLIGRL (virus + SLIGRL), or 1:1200 dilution of allantoic fluid (vehicle). Mice were sacrificed at days 2, 4, 8, or 17 postinoculation for experimental endpoints. PAR2-deficient mice (SVJ/129 background) and wild-type controls were provided by Professor Eleanor Mackie (School of Veterinary Sciences, University of Melbourne, Melbourne, Australia) (Georgy et al., 2010).
The effect of SLIGRL on IAV-induced inflammation was determined by using differential cell counting of leukocytes recovered from bronchoalveolar lavage (BAL) fluid. At selected time points postinoculation, groups of mice were sacrificed with an overdose of sodium pentobarbitone (160 mg/kg i.p.). The trachea was cannulated, and BAL was performed by intratracheal instillation of 2.5 ml of ice-cold phosphate-buffered saline (PBS), pH 7.4, in 0.5-ml volumes. After each instillation, the BAL fluid was recovered, pooled, and centrifuged at 400g for 5 min at 4°C. The supernatant was removed and frozen at −80°C for further analysis of cytokines, and the cell pellet was resuspended in PBS + 1.0% bovine serum albumin. Total cell counts and viability were determined by using a hemocytometer and 0.4% trypan blue exclusion. Cytospin preparations of each cell sample were stained with DIFF-Quick (Thermo Fisher Scientific, Waltham, MA), and differential cell counts of macrophages, neutrophils, eosinophils, and lymphocytes were determined by counting 400 cells under a light microscope using standard morphological criteria.
Enzyme-Linked Immunosorbent Assay.
Levels of MIP-2, KC, and interferon-γ (IFN-γ) in BAL fluid were determined by using enzyme immunoassay kits purchased from R&D Systems (Minneapolis, MN) according to individual kit instructions.
At selected time points postinoculation, groups of mice were anesthetized (sodium pentobarbitone, 160 mg/kg i.p.) and subjected to gravity-fed pulmonary perfusion with heparinized PBS (15 IU/ml) followed by 2% paraformaldehyde in PBS, pH 7.4. The trachea and lungs were excised and postfixed for 48 h at 4°C. Tissue was then dehydrated and processed to paraffin wax on a standard 15-h cycle in a Leica (Wetzlar, Germany) ASP200S automated tissue processor, and 5-μm sections were stained with hematoxylin and eosin. Images are representative of at least three separate experiments with at least three mice per experiment.
Immunohistochemistry and Imaging.
The presence of immunoreactive IAV in the lungs and trachea of (±SLIGRL)-treated mice was visualized by using standard immunohistochemical procedures. In brief, 5-μm wax sections were dewaxed, rehydrated, permeabilized (Triton X-100/15 min at room temperature), blocked for 2 h (50% normal goat serum/1% bovine serum albumin in TBS, pH 7.4), subjected to avidin/biotin blocking (Avidin/Biotin Blocking Kit; Vector Laboratories, Burlingame, CA), and incubated overnight at 4°C with a 1/1000 dilution (5 μg/ml) of goat anti-influenza A polyclonal antibody (Millipore Corporation, Billerica, MA) or normal goat IgG isotype control. After extensive washing in TBS, endogenous peroxidase was quenched (3% H2O2 in TBS/15 min at room temperature), and sections were exposed to biotinylated rabbit anti-goat secondary Ig followed by an avidin-biotin-horseradish- peroxidase complex (Vector Elite ABC kit; Vector Laboratories). The bound complex was developed with diaminobenzidine (0.4 mg/ml). Sections were counterstained with Mayers hematoxylin, blued with Scotts Tap Water Substitute, washed, dehydrated through graded alcohols and xylene, then coverslipped with Depex mounting medium. Digital images were acquired on an Aperio ScanScope (Aperio Technologies, Vista, CA) and analyzed by using ImageScope software (Aperio Technologies). The number of IAV-positive cells in the horizontal sections of tracheas was expressed as a percentage of the airway epithelial basement membrane (μm).
Scanning Electron Microscopy.
The effect of SLIGRL on IAV-induced epithelial damage was examined by using scanning electron microscopy. Treated mice were sacrificed as described above, and their tracheas were excised and fixed in 2.5% (w/v) glutaraldehyde in PBS. Tissue was postfixed with 1% (v/v) OsO4 aqueous solution, washed in double-distilled water, dehydrated in ethanol, and critical point dried. Specimens were subsequently mounted onto support studs, coated with a gold/paladium mix of <15-nm thickness, and examined under a Philips XL30 scanning electron microscope (SEMTech Solutions, North Billerica, MA). Photomicrographs shown are representative of at least three independent studies from three mice.
Lung Function Recording.
To determine the effect of SLIGRL on IAV-induced airway hyper-responsiveness, inoculated mice were examined on day 4 postinoculation by using in vivo lung function recording systems. In brief, mice were anesthetized (urethane, 1.5 g/kg i.p.), and cannulae were placed in the trachea (to measure airflow) and jugular vein (for drug administration). Spontaneous breathing was stopped by an intravenous injection of pancuronium bromide (0.4 and 0.2 mg/kg each 30 min thereafter), and mice were ventilated (tidal volume 8 ml/kg at 150 breaths/min; SAR-830 Ventilator, CWE, Colorado Springs, CO). Breath-to-breath measurements of airway resistance (RL) and dynamic compliance (Cdyn) were recorded (PMS800; Mumed, London, UK). Tracheal flow was measured over the tracheal cannula (Fleisch pneumotachograph; Mumed) and transpulmonary pressure was measured with a differential pressure transducer, with one end being connected to the outlet of the tracheal cannula and the other to a water-filled cannula inserted in the esophagus. A self-regulating heating pad (Fine Science Tools, Foster City, CA) was used to maintain a core temperature at 37°C. A bronchoconstrictor stimulus, methacholine (dose range 50–1600 μg/kg i.v.), was administered as a bolus dose at 5-min intervals to determine airway responsiveness. Before each methacholine challenge, lungs were hyperinflated once (by delivering twice tidal volume) to prevent and reverse atelectasis.
Isometric Tension Recording Studies.
The functionality of PAR2 in PAR2-deficient and wild-type mice was determined by using an isometric tension recording assay. Tracheal smooth muscle segments were suspended under a resting tension of 0.5 g in organ baths containing 2 ml of Krebs bicarbonate solution, maintained at 37°C and bubbled continuously with 5% CO2 in O2. The composition of the Krebs bicarbonate solution was 117 mM NaCl, 5.36 mM KCl, 25 mM NaHCO3, 1.03 mM KH2PO4, 0.57 mM MgSO4·7H2O, 2.5 mM CaCl2, and 11.1 mM d-glucose. Changes in tension were recorded via an isometric force transducer (FTO3; Grass Instruments, Quincy, MA) connected to a Powerlab system (ADInstruments Pty Ltd., Castle Hill, Australia). After a 45-min equilibration period, during which the tissues were washed every 15 min and tension was readjusted to 0.3 g, tissues were exposed to the cumulative addition of submaximal (0.2 μM) and supramaximal (10 μM) concentrations of carbachol, washed, and rested. Preparations were precontracted with carbachol (60–70% contraction maximum), and upon reaching a plateau level of contracture, a single concentration of peptide (SLIGRL, LSIGRL, IGRL, or 2-furoyl-LIGRL) was added, and the relaxation response was recorded. Preparations were washed and allowed to re-equilibrate for 15 min, and the process was repeated with another peptide.
Ex Vivo Perfused Mouse Tracheal Preparation.
Based on the concept of Pittet et al. (2010) we devised a tissue perfusion system capable of maintaining tracheal viability (both functional and structural) for at least 4 days of culture (see Fig. 4, A and B). Isolated mouse tracheas were maintained by a continuous flow (100 μl/min) with complete RPMI medium (cRPMI), consisting of RPMI 1640 medium (Sigma, St. Louis, MO) supplemented with 20 mM HEPES, 2% FBS, 2.5 μg/ml amphotericin β, 2 mM GlutaMAX, and 50 μg/ml gentamycin, supplied by (Invitrogen, Carlsbad, CA). In brief, the perfusion system was comprised of a sterile reservoir of cRPMI ultimately infusing (via Masterflex-13 tygon tubing; Cole-Parmer, Chatswood, Australia) an isolated trachea tied to a blunted 18-g needle that had been inserted through the filter cap of a 225-cm2 tissue culture flask. A peristaltic pump (Instech Laboratories, Plymouth Meeting, PA) regulated the flow of medium via two lines to both the lumen and exterior surface of the trachea. Thus, within the confines of the loosely capped flask the perfused trachea was maintained in a humid environment. Six perfusion systems were constructed and housed in a warm room set at 37°C for the duration of each experiment. Before exposure with IAV, tracheas were allowed to equilibrate for 1 h at 37°C with cRPMI. The lumen was flushed with 200 μl of sterile saline before exposure to virus (±SLIGRL, 20 or 100 μg). For delivery of IAV, SLIGRL, vehicle, and wash solution (sterile saline), the pump was turned off, and the required agent (50 μl) was administered via the needle port in the flask cap. After time of exposure and dose of virus titration experiments (unpublished data), exposure time was established at 15 min, followed by a wash with 600 μl of sterile saline and perfusion with cRPMI for 48 h. At the end of the perfusion period tracheas were removed from their supporting needles and fixed (2% paraformaldehyde and 0.2% picric acid in PBS pH 7.4) for 48 h at 4°C. Tissue was then dehydrated and processed to paraffin wax on a standard 15-h cycle in a Leica ASP200S automated tissue processor.
In Vitro Viral Replication and Pathogenicity Studies.
The effect of SLIGRL on IAV infection was studied by using chicken egg chorioallantoic cells, a well established host of influenza A/PR8/34 virus. Shells of 11-day-old embryonated eggs (chorio-allantoic membrane intact) were cut into 25-mm2 squares, washed, and placed in sterile 5-ml tubes containing 450 μl of standard medium. Serial 10-fold dilutions of influenza A/PR/8/34 virus containing saline (vehicle), SLIGRL (80, 400, or 2000 μM), or the positive control IAV replication inhibitor ribavirin (40 μM) were added to egg bits as a 50-μl aliquot and incubated for 60 h at 37°C. Media were removed and assayed for viral hemagglutination activity. In the hemagglutination assay, media were serially diluted 2-fold in a 96-well hemagglutination tray, incubated with an equal volume of 1% goose erythrocyte solution for 45 min at room temperature, and visually inspected for evidence of erythrocyte agglutination.
The influence of SLIGRL and related peptides on IAV replication was also examined in MDCK cell cultures. MDCK cells were grown in DMEM supplemented with 10% (v/v) fetal calf serum, 100 U/ml penicillin-G, 100 μg·ml−1 streptomycin, and 2 mM GlutaMAX and maintained in a humidified CO2 incubator at 37°C. Confluent cultures were washed, detached by using trypsin, and subcultured into 96-well tissue culture plates. MDCK cells were incubated in serum-free media for 3 h and exposed to aliquots of saline, SLIGRL (80, 400, or 2000 μM), or ribavirin (40 μM) together with serial 10-fold dilutions of influenza A/PR/8/34 virus for 1 h. The supernatant was removed, the cell monolayer was washed (to remove nonadhered virus), and cells were reincubated in serum-free media containing the appropriate vehicle, peptide, or ribavirin for 24 h at 37°C. At the completion of the incubation period, monolayers were washed (sterile PBS) and exposed to a 50-μl aliquot of a 1% guinea pig erythrocyte suspension in PBS for 45 min at 4°C. Monolayers were washed (to remove nonadherent erythrocytes), and the number of adherent erythrocytes per microscopic field was counted, with at least three different randomized fields counted to provide a well value.
The influence of SLIGRL and related peptides on the pathogenicity of IAV was determined by using MDCK cell cultures. MDCK cells were harvested from T-25 flasks at passages 5 to 7, resuspended in DMEM [4.5 g/liter glucose, 100 mg/liter pyruvate supplemented with 5% FCS, 2 mM GlutaMAX, 100 IU/ml penicillin, 100 μg/ml streptomycin, and 2.5 μg/ml amphotericin B (Invitrogen)], plated into 96-well tissue culture plates at 10,000 cells/well, and maintained in a humidified CO2 incubator at 37°C. Three hours before pathogenicity experiments, cells were washed and the medium was changed to serum-free DMEM. Influenza A/PR8/34 virus was serially diluted 10−1 to 10−5, and SLIGRL was diluted such that the final concentrations/well were 80, 400, and 2000 μM. Each dilution of virus was preincubated with each concentration of SLIGRL for 90 min at 37°C. After this preincubation, the IAV/SLIGRL solutions, cells, and medium were placed on ice and maintained at 4°C. Cells were then exposed (in triplicate) to each IAV/SLIGRL combination for 45 min at 4°C to allow attachment, but not replication of virus. Cells were washed then incubated in serum-free medium at 37°C for 0, 24, and 48 h before the determination of cell viability using an ATP assay (Promega Cell Titer-Glo Luminescent Cell Viability Assay; Promega, Madison, WI).
Data Analysis and Statistical Procedures.
Group data is expressed as mean ± S.E.M. Unless otherwise stated, statistical comparisons between groups were made by using one-way ANOVA, and post hoc comparisons among groups were performed by the Student-Newman-Keuls multiple comparison test using a commercial software package (SigmaStat; Systat Software, Inc., San Jose, CA). Groups were considered to be significantly different at P < 0.05.
Drugs and Chemicals.
Carboxyl-terminal amidated peptides (custom-synthesized, single-letter amino acid codes) were supplied by Auspep (Melbourne, Australia). For simplicity, SLIGRL-amide, LSIGRL-amide, IGRL-amide, and 2-furoyl-LIGRL-amide are referred to as SLIGRL, LSIGRL, IGRL, and 2-furoyl-LIGRL, respectively throughout the text. Stock solutions were prepared in sterile high-purity water with subsequent dilutions made in sterile saline. The concentration, purity, and composition of the peptides were determined by high-performance liquid chromatography, mass spectrometry, and quantitative amino acid analysis.
Influenza A/PR/8/34 Virus Infection.
BAL fluid recovered from the lungs of vehicle mice consistently contained approximately 200 × 103 cells (Fig. 1A), of which 99% were macrophages (Fig. 1B). Intranasal inoculation of influenza A/PR/8/34 virus caused a marked time-dependent influx of inflammatory cells into the lungs of mice (Fig. 1A). At day 4 postinoculation, there were 2.8 times more macrophages (Fig. 1B) and 675 times more neutrophils (Fig. 1C) in virus mice than in vehicle mice (P < 0.001; one-way ANOVA). At day 8 postinoculation, the number of macrophages and neutrophils was still significantly elevated in virus mice (2.8 and 165 times more than vehicle, respectively; P < 0.001; one-way ANOVA), and lymphocytes were also detected (40 ± 8 × 103 cells/mouse BAL). By day 17 postinoculation, macrophage (Fig. 1B) and lymphocyte (63 ± 7 × 103 cells/mouse BAL) numbers were still elevated in virus compared with vehicle mice (P < 0.001; one-way ANOVA), whereas neutrophil numbers had returned to baseline levels (Fig. 1C).
Inhibitory Effect of SLIGRL on IAV Infection.
Mice that received an intranasal dose of 20 μg of SLIGRL together with 490 EID50 doses of IAV (virus + SLIGRL) had significantly fewer macrophages, neutrophils, and lymphocytes in BAL fluid collected on day 4 postinoculation compared with virus mice (P < 0.001; one-way ANOVA; Fig. 1). Indeed, the numbers of leukocytes recovered from virus + SLIGRL mice were not significantly different from vehicle mice (Fig. 1). In additional experiments, the number of inflammatory cells recovered from mice 4 days after administration of 20 μg of SLIGRL alone (vehicle + SLIGRL mice; 90 ± 10 × 103 cells; n = 5 mice) was not statistically different from saline-treated mice (vehicle mice; 130 ± 30 × 103 cells; n = 5 mice).
Inhibitory Effect of SLIGRL on IAV Propagation.
Tracheal segments from virus mice demonstrated strong positive epithelial staining for immunoreactive IAV at day 2 postinoculation (Fig. 2B). In contrast, no immunoreactive staining for IAV was observed in tracheal sections taken from either virus + SLIGRL (Fig. 2C) or vehicle (Fig. 2A) mice.
Inhibitory Effect of SLIGRL on IAV-Induced Histopathology.
Visual inspection of scanning electron microscopic images revealed extensive damage to the tracheal epithelium in virus mice at day 4 postinoculation (Fig. 2E), particularly cilial desquamation, compared with vehicle mice (Fig. 2D). Mice inoculated with SLIGRL in conjunction with IAV exhibited markedly fewer signs of IAV-induced histopathology (Fig. 2F), although ciliated cells seemed to have slightly shorter cilia and nonciliated cells seemed more rounded in virus + SLIGRL mice (Fig. 2F) compared with vehicle mice (Fig. 2D). It is not clear whether these subtle differences in epithelial morphology between virus + SLIGRL mice (Fig. 2F) and vehicle mice (Fig. 2D) were caused by the actions of SLIGRL and/or IAV. Hematoxylin and eosin-stained tracheal and lung sections obtained from virus mice exhibited characteristic lymphocytic peribronchitis, iatrogenic atelectasis, and intraluminal cellular debris on days 4 and 8. In contrast, tracheal and lung sections from virus + SLIGRL mice were indistinguishable from vehicle mice and devoid of any signs of IAV-induced histopathology (data not shown).
Inhibitory Effect of SLIGRL on Virus-Induced Changes in BAL Cytokine and Chemokine Levels.
As expected, the levels of IFN-γ, KC, and MIP-2 in BAL fluid from virus mice were significantly and substantially greater than in vehicle mice (Fig. 2G). Consistent with the BAL cell profile obtained in Fig. 1, coadministration of SLIGRL (20 μg/mouse) to virus mice prevented the increase in IFN-γ, KC, and MIP-2 (Fig. 2G).
Inhibitory Effect of SLIGRL on IAV-Induced Airway Hyper-Responsiveness.
On day 4 postinoculation, intravenous injection of 62.5 μg/kg methacholine to virus mice produced a significantly larger increase in airway resistance than that observed in vehicle mice (P < 0.005; one-way ANOVA; Fig. 2H), indicative of virus-induced airway hyper-responsiveness. Responses to higher doses of methacholine could not be reliably obtained in virus mice, presumably because of the deleterious effects of IAV on the airways (data not shown). Virus + SLIGRL mice were not hyper-responsive to methacholine, with increases in resistance the same as those obtained from vehicle mice (Fig. 2H).
Inhibitory Effect of SLIGRL on Influenza A/Mem/1/71 Virus Infection.
The inhibitory effects of SLIGRL (20 μg/mouse intranasally) on IAV-induced pulmonary inflammation were not limited to influenza A/PR/8/34 virus but were also observed for influenza A/Mem/1/71 (Fig. 2J). Thus, coadministration of SLIGRL with IAV suppressed many characteristic features of an IAV infection, including epithelial viral propagation, epithelial cell damage, airway inflammation characterized by pulmonary neutrophilia, and airway hyper-responsiveness.
Dose-Response Relationship for SLIGRL and Structural Analogs.
The dose of SLIGRL that suppressed influenza A virus-induced increases in BAL neutrophil number by 50% was 10 μg/mouse (Fig. 3A). We were surprised that dose-response studies revealed that the partially scrambled peptide sequence LSIGRL produced the same maximal effect as SLIGRL (full suppression of virus-induced increases in BAL neutrophil number) and was only 4-fold less potent than SLIGRL (IC50 value of 40 μg/mouse; Fig. 3A). Of particular interest, 2-furoyl-LIGRL, which is typically 20- to 100-fold more potent than SLIGRL as an activator of PAR2 (Kawabata et al., 2004), was inactive up to doses of 30 μg/mouse, and IGRL, which is not known to activate PAR2, was 2.5-fold more potent than SLIGRL (IC50 value of 4 μg/mouse; Fig. 3A). These data are not consistent with the inhibitory effects of SLIGRL on IAV infection being mediated solely through the activation of PAR2 and were the trigger for further studies using PAR2(−/−) mice.
Inhibitory Effects of SLIGRL in PAR2(+/+) and PAR2(−/−) Mice.
IAV induced a similar increase in pulmonary leukocytes in PAR2(+/+) and PAR2(−/−) mice (Fig. 3B). Of particular interest, SLIGRL (20 μg/mouse intranasally) inhibited IAV-induced pulmonary inflammation in both PAR2(+/+) and PAR2(−/−) mice (Fig. 3B), indicating that SLIGRL was exerting anti-IAV effects via a PAR2-independent pathway. Confirmation of the lack of functional PAR2 in PAR2(−/−) mice was obtained from isometric tension recording studies. As shown in Fig. 3C, the well established PAR2-mediated smooth muscle relaxant actions of SLIGRL and 2-furoyl-LIGRL in precontracted mouse tracheal segments (Lan et al., 2000, 2001) were present in PAR2(+/+) mice but absent in PAR2(−/−) mice. As expected, the partially scrambled peptide sequence LSIGRL was inactive in both PAR2(+/+) and PAR2(−/−) mice. Together, these data are not consistent with the inhibitory effects of SLIGRL on IAV infection being mediated through the activation of PAR2, and additional mechanistic studies were conducted to investigate the cellular mechanisms through which SLIGRL exerts this anti-IAV activity.
Inhibitory Effect of SLIGRL in Ex Vivo Mouse Perfused Tracheal Preparations.
A novel ex vivo mouse perfused tracheal system was used to determine whether the anti-IAV activity of SLIGRL could be replicated in an in vitro system in which the tissue viability, architecture, and microenvironment are preserved (Fig. 4, A and B; modification of a recently reported explant system, Pittet et al., 2010). Ex vivo perfused tracheal segments remained viable for at least 4 days of perfusion, displaying normal epithelial and smooth muscle cell morphology and responsiveness (data not shown). A 15-min period of exposure to influenza A/PR/8/34 virus at the beginning of the perfusion period (day 0) was associated with marked proliferation of IAV within the epithelium in perfused tracheal segments obtained from PAR2(+/+) and PAR2(−/−) mice over the ensuing 48 h (Fig. 4C). Coadministration of SLIGRL with IAV during the initial exposure period significantly suppressed IAV proliferation and pathology in tracheal segments obtained from both PAR2(+/+) and PAR2(−/−) mice (Fig. 4C). The finding that the anti-IAV activity of SLIGRL was preserved in isolated tracheal segments indicates that SLIGRL acts by inhibiting the proliferation of IAV (rather than by acting as an immunomodulatory agent). Further studies were conducted to determine whether SLIGRL inhibited virus replication directly by inhibiting IAV attachment to host receptors.
Lack of Effect of SLIGRL on Influenza A Replication and Pathogenicity in Cell Culture Systems.
Several small peptides have previously been shown to produce antiviral effects by inhibiting the attachment of IAV to host cell receptors (Jones et al., 2006; Rajik et al., 2009). Thus, a series of studies were conducted to determine whether SLIGRL influenced IAV-induced effects in MDCK cells and chorioallantoic membrane cells, which are established host cells for IAV. Over the concentration range of 80 to 2000 μM SLIGRL did not inhibit the capacity of IAV to replicate in either chorioallantoic cell membranes (determined by using a hemagglutination assay; Fig. 5A) or MDCK cell cultures (determined by using a hemabsorption assay; Fig. 5B). Furthermore, SLIGRL (80–2000 μM) did not inhibit IAV-induced death of MDCK cells as determined by using an ATP cell viability assay (Fig. 5C). In contrast, the antiviral agent ribavirin (40 μM) inhibited IAV effects in these isolated cell culture systems. These findings indicate that SLIGRL does not directly inhibit IAV and raises the possibility that it acts indirectly, perhaps by releasing anti-IAV viral substances from the intact airway.
Differential Effects of SLIGRL When Given Before or After IAV In Vivo or Ex Vivo Systems.
In the experiments described above, SLIGRL was coadministered with IAV. Additional experiments sought to determine whether SLIGRL was effective when given before or after IAV. Intranasal administration of a single 20 μg/mouse dose of SLIGRL either 1 h before or 1 h after IAV inoculation did not significantly suppress IAV-induced increases in the number of macrophages or neutrophils recovered from BAL fluid collected on day 4 postinoculation (Fig. 6, A and B). In stark contrast, the sequential exposure of ex vivo mouse tracheal segments to SLIGRL (20 μg for 15 min), saline (5 min to remove SLIGRL), and then IAV (490 EID50 for 15 min) effectively suppressed the numbers of IAV-positive cells determined 48 h later (Fig. 6, E and F) compared with IAV alone (Fig. 6, C and F). As expected, the coadministration of SLIGRL and IAV suppressed infection in ex vivo mouse tracheal segments (Fig. 6, D and F).
In the current study, intranasal inoculation of mice with IAV caused a characteristic array of effects within the respiratory tract including viral replication within airway epithelial cells, extensive damage to the epithelial cell layer, recruitment of neutrophils, macrophages, and lymphocytes, and airway hyper-responsiveness to methacholine. These IAV-induced effects were inhibited by SLIGRL, a hexapeptide sequence known to activate murine PAR2. However, additional mechanistic studies indicate that SLIGRL suppresses IAV infection independently of PAR2, possibly via the generation of unidentified endogenous substances with anti-IAV activity.
The principal aim of this study was to examine the effects of administering a PAR2-activating peptide SLIGRL on the development of IAV infection in mice. Intranasal administration of SLIGRL caused dose-dependent inhibition of IAV-induced inflammation, pathology, and airway hyper-responsiveness in BALB/c mice (Figs. 1 and 2). These data concur with the findings of Khoufache et al. (2009) where SLIGRL protected C57BL/6J mice from IAV-induced pathogenesis and death. Because SLIGRL is a well established activator of murine PAR2, these findings provided prima facie evidence for PAR2 having a protective role against IAV pathogenesis. However, key pieces of evidence argue strongly against the anti-IAV effects of SLIGRL being mediated by PAR2.
First, dose-response studies using a range of structural analogs of SLIGRL that activate PAR2 (2-furoyl-LIGRL) or do not activate PAR2 (LSIGRL and IGRL) indicate that the anti-IAV activity of SLIGRL involves PAR2-independent mechanisms. In our dose-response studies, the dose of SLIGRL that suppressed IAV-induced increases in BAL neutrophil number by 50% was 10 μg/mouse. Dose-response studies revealed that LSIGRL produced the same maximal effect as SLIGRL (full suppression of IAV-induced increases in BAL neutrophil number) and was only 4-fold less potent than SLIGRL (IC50 value of 40 μg/mouse). Although LSIGRL has previously been reported to produce physiological effects in several biological systems (Al-Ani et al., 1999, 2002; Vergnolle et al., 1999; Nishikawa et al., 2005), this is the first study that has demonstrated the ability of LSIGRL to produce physiological effects of magnitude equal to that of SLIGRL, albeit with a lower potency. Khoufache et al. (2009) reported that the control peptide used in their study (RLGILS) was inactive at a dose of 30 μg/mouse, although detailed dose-response curves were not reported; thus a potential antiviral activity produced by higher doses was not revealed. Of particular interest, 2-furoyl-LIGRL, which is typically 20- to 100-fold more potent than SLIGRL as an in vivo activator of PAR2 (Kawabata et al., 2004), was inactive up to doses of 30 μg/mouse and was thus significantly less potent than SLIGRL at inhibiting IAV-induced effects. Furthermore, we observed that IGRL, which is not known to activate PAR2 but has been postulated to be the active core structure in LSIGRL-induced tear secretion in rats (Nishikawa et al., 2005), was 2.5-fold more potent than SLIGRL (IC50 value of 4 μg/mouse). Together, the potency order observed for inhibition of virus-induced neutrophilia [IGRL > SLIGRL > LSIGRL >2-furoyl-LIGRL (inactive)] was not consistent with a PAR2-mediated response [expected potency order of 2-furoyl-LIGRL > SLIGRL > LSIGRL = IGRL (inactive)].
Second, even more compelling were the findings that SLIGRL inhibited IAV-induced recruitment of pulmonary leukocytes in mice that lacked PAR2 (Fig. 3B) and inhibited the proliferation of IAV in mouse isolated tracheal segments obtained from PAR2-deficient mice (Fig. 4C). The absence of functional PAR2 in the trachea of PAR2(−/−) mice was confirmed in isometric tension recording studies, which showed that SLIGRL was inactive in tracheal smooth muscle preparations obtained from PAR2(−/−) mice, but produced dose-dependent relaxation in wild-type PAR2(+/+) mice (Lan et al., 2000, 2001). Together, the use of PAR2(−/−) mice and structure activity analyses of PAR-activating peptides (McGuire et al., 2002) powerfully demonstrate that the anti-IAV actions of SLIGRL are PAR2-independent.
In light of our unexpected findings that SLIGRL inhibits IAV infection via PAR2-independent processes, a secondary aim of the study was to investigate how SLIGRL exerts this anti-IAV activity. IAV infection within the respiratory system begins with IAV attachment to host cells, which involves the binding of the hemagglutinin subunit on the surface of IAV to sialic acid-containing receptors on the surface of (mainly) airway epithelial cells. Although several peptides have been reported to inhibit IAV attachment (Jones et al., 2006; Rajik et al., 2009), it is unlikely that SLIGRL shares this mechanism of action. First, we demonstrated that the preincubation of mouse tracheal segments with SLIGRL followed by washout, before IAV challenge, was sufficient to inhibit IAV propagation (Fig. 6E). Thus, the presence of SLIGRL at the time of IAV attachment was not required, contrary to what might be expected if SLIGRL was directly inhibiting IAV attachment. Second, SLIGRL did not inhibit ribavirin-sensitive viral replication in either chorioallantoic membranes or MDCK cells, both of which support IAV attachment (Sidwell and Smee, 2000). Finally, SLIGRL failed to inhibit the hemagglutination of red blood cells induced by IAV, which might be predicted if SLIGRL were capable of inhibiting IAV attachment.
Attachment of the virus to the cell surface is followed by the endocytosis of virus, the fusion of viral and endosomal membranes, the release, transcription, and translation of the viral genome, and the packaging and release of progeny virions (Müller et al., 2012). This process of IAV replication depends on dozens of host factors, and the majority of conventional antiviral compounds studied to date act via cellular pathways to alter the ability of the virus to propagate (McNicholl and McNicholl, 2001; Takeda et al., 2002; Magden et al., 2005; Guo et al., 2006). However, it is unlikely that SLIGRL directly inhibits IAV propagation pathways because it did not inhibit IAV replication in either chorioallantoic membranes or MDCK cells. These latter findings contrast with reports that PAR2-activating peptides decreased progeny influenza A virus titers in isolated human monocytes (Feld et al., 2008) and lung epithelial A549 cells (Khoufache et al., 2009) and indicate that in vitro antiviral effects of SLIGRL are cell type-dependent. Consistent with our findings that SLIGRL did not inhibit viral binding or replication, SLIGRL did not inhibit virus-induced death in MDCK cell cultures. The lack of in vitro efficacy of SLIGRL observed in the present study suggests that SLIGRL does not act by directly inhibiting IAV attachment or replication and raises the possibility that SLIGRL promotes host antiviral pathways.
Among the first lines of host defense against IAV is the mucociliary clearance system that seeks to prevent the virus from attaching to airway epithelial cells. Effective mucociliary clearance requires appropriate mucus production and coordinated ciliary activity, which work in a synchronized manner to remove inhaled particles from the lung (Mall, 2008). Of particular interest, the respiratory tract produces respiratory secretions containing a mixture of phospholipids and proteins that recognize and neutralize certain strains of IAV (White et al., 2008; Kesimer et al., 2009). Thus, it is possible that SLIGRL blocks IAV infection by stimulating the mucociliary system and facilitating the rapid removal of virus from the respiratory tract. Consistent with this postulate, SLIGRL exerts a stimulatory action on a wide range of exocrine glands, promoting the release of gastric secretions, saliva, and tears via PAR2-dependent and -independent pathways (Kawabata et al., 2000, 2004; Nishikawa et al., 2005). In addition, the finding that prior exposure of isolated tracheal segments to SLIGRL afforded protection against IAV administered shortly thereafter is consistent with SLIGRL having promoted the release of substances with anti-A virus activity. Related to this, Numata et al. (2012) have recently reported that palmitoyl-oleoyl-phophatidylglycerol (POPG), a minor component of pulmonary surfactant, acts as a strong antiviral agent against IAV in mice. Like SLIGRL, we found that exogenous POPG also inhibited IAV replication in ex vivo mouse perfused tracheal segments (data not shown). Although speculative, it may be that SLIGRL promotes the release from the airways of endogenous substances with anti-IAV activity.
Our novel data convincingly show that SLIGRL inhibits IAV infection independently of PAR2. So what is the target of SLIGRL? We are currently exploring the prospect that the anti-IAV actions of SLIGRL are mediated through Mas-related G protein-coupled receptor C11 (MrgprC11). This postulate is based on several lines of evidence. First, SLIGRL has recently been shown to activate MrgprC11 (Liu et al., 2011). Second, MrgprC11 is expressed specifically on sensory nerves (Lee et al., 2008), which play an important role in regulating respiratory tract function, including secretion. Third, LSIGRL and IGRL, which exhibit antiviral activity, also activate MrgprC11 (but not PARs). We are currently investigating the potential roles of MrgprC11 and POPG as mediators of the anti-IAV effects of SLIGRL in murine airways.
In summary, SLIGRL exhibited a protective role against IAV-induced pathogenesis in mice independently of PAR2. Data obtained from a series of mechanistic studies examining the influence of SLIGRL on IAV-induced responses in intact mice, ex vivo perfused mouse tracheal segments, and cell cultures indicate that SLIGRL does not directly inhibit IAV attachment or replication, but may stimulate the production by the host airway of substances with anti-IAV activity. Further studies are necessary to determine the precise molecular mechanisms through which SLIGRL exerts this novel anti-IAV activity and establish whether these pathways are operational in human airways and effective against other strains of respiratory tract viruses.
Participated in research design: Betts, Mann, and Henry.
Conducted experiments: Betts and Mann.
Contributed new reagents or analytic tools: Betts, Mann, and Henry.
Performed data analysis: Betts, Mann, and Henry.
Wrote or contributed to the writing of the manuscript: Betts, Mann, and Henry.
This research was funded by the National Health and Medical Research Council of Australia.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- influenza A virus
- analysis of variance
- bronchoalveolar lavage
- complete RPMI
- Dulbecco's modified Eagle's medium
- 50% egg infectious dose
- keratinocyte-derived chemokine
- Madin-Darby canine kidney
- macrophage inflammatory protein 2
- Mas-related G protein-coupled receptor C11
- proteinase-activated receptor
- phosphate-buffered saline
- Tris-buffered saline.
- Received May 6, 2012.
- Accepted September 11, 2012.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics