Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Protease-activated receptor-2 (PAR2) is a member of a family of four receptors that are activated by serine proteases (1). All members of this receptor family are activated by an extracellular amino acid sequence that is able to bind intramolecularly only after proteolytic removal of a blocking upstream sequence. As such, these receptors act as "sensors" of serine proteases (1). PAR2 is a unique member of this receptor family because it is activated by trypsin and similar enzymes rather than by thrombin, which activates PAR1, PAR3, and PAR4 (1). It is widely held that PAR2 is a receptor for the mast cell-derived serine protease tryptase and that PAR2 may therefore participate in the progression of inflammatory diseases (2). Indeed, intraplantar injection of the hexapeptide SLIGRL (single amino acid code), which mimics the "tethered ligand" region of PAR2 and causes receptor activation, causes an inflammatory response associated with an influx of polymorphonuclear leukocytes (3). However, we have suggested that in some organs, including the lung, PAR2 serves as a detector of inflammatory responses and is coupled to protective rather than destructive pathways (1, 4). Support for this hypothesis includes the observations that SLIGRL protects against coronary ischemia-reperfusion injury (5) and inhibits the development of gastric ulcers (6).

Activation of PAR2 using the peptide SLIGRL causes relaxation of isolated airway preparations (7) and protects against bronchoconstrictor challenges in vivo in guinea pigs (10) and rats (7). Whether PAR2 activation has any inflammatory effect after administration to the lungs has not been explored in depth, although Cicala and coworkers (10) showed that intravenous SLIGRL had a modest inhibitory effect on histamine-induced pulmonary microvascular leakage in guinea pigs. PAR2 activation has been shown to initiate the release of matrix metalloproteases (MMPs) (11), particularly MMP-9 and eosinophil survival factors (12) from a human alveolar epithelial cell line, suggesting a possible proinflammatory role for PAR2 over a longer time scale. In preliminary studies we have found that SLIGRL modestly inhibits lipopolysaccharide (LPS)-induced pulmonary neutrophilia (4) over a short (3 h) time frame. In the present study we have examined the possible inflammatory effect of SLIGRL delivered intranasally to the airways alone, and during a bacterial LPS-induced inflammatory response over a 72-h time course by examining cell populations obtained from bronchoalveolar lavage (BAL). We have also analyzed BAL fluid using zymographic techniques that can detect both changes in MMP-9 (13) and trypsin levels (14) to determine if either enzyme is regulated by inflammation or PAR2 activation. The results clearly demonstrate that SLIGRL does not cause pulmonary inflammation but can markedly inhibit the accumulation of neutrophils in response to an inflammatory stimulus.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Specific-pathogen free Balb/C mice (18-22 g) were obtained from the Animal Resource Centre (Perth, WA, Australia) and housed locally for less than 2 wk on a standard diet and constant light cycle. Animals received SLIGRL (50 µl, 10 mg/ml; custom synthesis of amidated peptide by Auspep, Melbourne, Vic., Australia), LPS (50 µl, 0.2 mg/ml; Escherichia coli serotype 0127:B8; Sigma, St. Louis, MO), or both in combination (25 µl SLIGRL, 20 mg/ml; 25 µl LPS, 0.4 mg/ml) intranasally under light halothane anesthesia as previously described (15). LPS and SLIGRL were dissolved in sterile phosphate-buffered saline (PBS). In preliminary studies, we found the dose of SLIGRL used had a modest anti-inflammatory effect against LPS-induced pulmonary neutrophilia, whereas lower doses had minimal effect (4). In the present study we also determined that the inactive peptide LSIGRL (16; amidated peptide a gift of Dr. P. Henry, University of Western Australia) had no effect in this model (see RESULTS). Animals were allowed to recover and were killed at 1, 3, 6, 24, 48, and 72 h and bronchoalveolar lavage performed. Because we have previously established that saline alone has no significant effect on BAL cell populations (data not shown), we did not perform time-matched saline controls, but used untreated animals as controls to establish that the baseline inflammatory status of each group was similar.

BAL

Mice were killed with an overdose of sodium pentobarbitone (80 mg/kg, intraperitoneally; Rhone Merieux, Pinkenba, QLD., Australia). The trachea was exposed and cannulated with a polyethylene tube held in place with a small artery clamp and BAL performed with three 0.5-ml aliquots of PBS. A total of 1.2-1.4 ml of BAL fluid (BALF) was consistently recovered by this technique.

BALF Cell Counts

The total number of cells retrieved in BALF was determined using a hemocytometer and differential cell counts were performed on cytospin (Shandon Scientific, Pittsburgh, PA) preparations of BALF stained with May-Grünwald and Geimsa stains (Sigma). At least 200 cells were counted in each sample.

Gelatin Zymography

Supernatants of centrifuged BALF samples were mixed 1:1 with nonreducing sample buffer sample buffer (125 mM Tris [pH 6.8], 20% glycerol, 4% sodium dodecylsulfate [SDS], and 0.002% bromophenol blue) and separated by SDS-polyacrylamide gel electrophoresis (PAGE) in 10% polyacrylamide gels (Mini-Protean 3; Bio-Rad, Hercules, CA) which had been copolymerized with gelatin (5 mg/ml; type B, porcine skin, Sigma). The gels were then washed for 30 min in 2% Triton X-100 to refold proteins and then incubated at 37°C for 18 h in 50 mM Tris buffer (pH 7.3) containing 10 mM CaCl₂, and 200 mM NaCl. In some experiments 20 mM ethylenediaminetetraacetic acid (EDTA) was added to the incubation buffer to inhibit metalloproteinases by chelating Zn²⁺ ions required for proteolytic activity (17). After incubation the gels were stained with Coomassie Blue (Brilliant Blue R250 [Sigma] in 40% methanol/10% acetic acid) and destained by boiling in distilled water for 10 min in a microwave oven (500 W). This destaining procedure produced reproducible background staining between individual gels. Areas of gelatinolytic activity appear as clear bands against a dark blue background. Images of the gels were captured with an external linear gray-scale calibration standard using a CCD camera coupled to a personal computer. The density of individual bands of gelatinolytic activity was estimated on a Macintosh computer using the public domain software NIH Image (available at http://rsb.info. nih.gov/nih-image) and is expressed in arbitrary units.

Western Blotting

Samples of BALF were mixed with sample buffer (see above) and proteins separated by SDS-PAGE on 10% polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes (Bio-Rad) using the Bio-Rad mini transblot apparatus (100 V, 1 h). The membranes were blocked for 1 h in PBS containing 1% bovine serum albumin, 0.05% Tween 20, and 5% skim milk powder at 37°C (PBST-BSA), briefly rinsed and incubated overnight at 4°C in a rabbit primary antisera raised against MMP-9 (AB19047, diluted 1:5,000; Chemicon, Temecula, CA). After washing the membranes with PBST-BSA, an anti-rabbit secondary antibody conjugated to horseradish peroxidase (diluted 1:1,000; Silenus, Melbourne, Vic., Australia) was applied for 1 h at room temperature. Enhanced chemiluminescence (NEN, Boston, MA) was used to detect labeled antigen.

Statistics

Data are expressed as the mean ± standard error (SE) of the mean. Differences between groups of mice at different time points were assessed using a one-way analysis of variance (ANOVA) with a Newman-Keuls post hoc test. A P value < 0.05 was accepted as statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

BALF Cellularity

As we observed in preliminary experiments (4), SLIGRL at a concentration of 10 mg/ml inhibited LPS-induced neutrophilia at 24 h. We now show that this effect of SLIGRL is due to activation of PAR2, because the reverse peptide LSIGRL was ineffective (Figure 1). In our extended study, SLIGRL alone did not significantly modify the number of neutrophils or macrophages present in BALF at any time point, although there were small, statistically insignificant, increases in neutrophils at 48 h and macrophages at 48 and 72 h (Figure 2). By contrast, LPS administration caused a massive influx of neutrophils, starting at 3 h, peaking at 48 h, and then declining by 72 h (Figure 2). Macrophage numbers in BAL also increased after LPS challenge, but at later (48-72 h) time points, indicating that macrophage recruitment was probably secondary to neutrophil entry (Figure 2). The combination of LPS and SLIGRL also elicited cellular recruitment into the airways (Figure 2). However, the number of neutrophils found in BAL was statistically less than that found in response to LPS alone from 6 h onwards. Co-administration of SLIGRL with LPS also resulted in a significant attenuation of the late (48 and 72 h) influx of macrophages compared with LPS alone (Figure 2).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Data from studies to test the effectiveness of the reverse peptide LSIGRL, which does not activate PAR2, in abrogating LPS-induced accumulation of neutrophils to the murine lung. BAL fluids were obtained at 24 h after administration of LPS alone or in combination with either the active PAR2-activating peptide SLIGRL (SL) or LSIGRL (LS) as described in the text. Data are expressed as mean ± SE from 4-7 animals. * indicates a significant difference compared with control (P < 0.05; ANOVA with Newman-Keuls post hoc test). Open bars, macrophages; solid bars, neutrophils.

View larger version (21K): [in this window] [in a new window]   Figure 2.   Time course of cellular influx into the airways determined by differential cell counts of cytospin preparations of BALF. The time point designated as C indicates cell counts from animals that received no treatment. # indicates significantly different (P < 0.05; ANOVA with Newman- Keuls post hoc test) than control (untreated) mice for that group. * indicates a significant difference (P < 0.05; ANOVA with Newman-Keuls post hoc test) between the LPS and LPS+SLIGRL groups at that time point. Data are expressed as mean ± SE; each group consists of 5-6 animals. Solid bars, LPS; open bars, SLIGRL; striped bars, LPS+SLIGRL.

MMP and Trypsin Activities in BALF

In all BAL samples only two bands of gelatinolytic activity were observed, one at ~ 115 kD and the other at 70 kD (Figure 3A). Neither band was observed when gels were incubated in EDTA (data not shown), suggesting that both are metalloproteinases (17). Because the 115-kD activity was modified by the experimental treatments and because it was consistently 10 kD heavier than the reported molecular weight of 105 kD (13) of murine MMP-9, we further confirmed the identity of this activity as MMP-9 by Western blot (Figure 3B). MMP-9 was markedly upregulated by LPS or the combination of LPS and SLIGRL, but was unaffected by SLIGRL alone (Figure 3C). There was no significant difference between the amount of MMP-9 activity detected in the LPS and LPS+SLIGRL groups, except at 3 h, at which time SLIGRL significantly potentiated the response to LPS (Figure 3C). The 70-kD activity was unchanged in all groups at all time points with the exception of the LPS-treated group at 3 h (Figure 3D), in which there was a variable but significant increase in activity. This activity almost certainly represents MMP-2 based on its molecular weight, the presence of a slightly heavier pro-form of the protein when higher levels were detected, and its sensitivity to EDTA. Co-administration of LPS and SLIGRL prevented the LPS-induced increase in MMP-2 activity at the 3-h time point (Figure 3D). We did not observe any gelatinolytic activity at 23 kD corresponding to trypsin itself as previously reported using this method (14), although we were able to detect such activity in lung homogenates (data not shown). We were also able to detect trypsin in lung homogenates by Western blotting (data not shown); neither technique revealed a difference in levels of trypsin between control and LPS-treated mice.

View larger version (44K): [in this window] [in a new window]   Figure 3.   Time course of activity of MMP-2 and MMP-9 in BALF. (A) An example of a gelatin zymogram showing the increased gelatinolytic activity at approximately 120 kD and the only minor changes in activity at 70 kD at various time points after LPS challenge. C indicates samples from control (untreated) mice. (B) Western blot demonstrating increased MMP-9 activity at 120 kD 24 h after LPS treatment. The symbols  and + indicate samples from control (untreated) and LPS-treated mice, respectively. (C and D) Time course of changes in estimated density of gelatinolytic bands in zymograms for the three groups of mice. # indicates significantly different (P < 0.05; ANOVA with Newman-Keuls post hoc test) than control (untreated) mice. * indicates a significant difference (P < 0.05; ANOVA with Newman-Keuls post hoc test) between the LPS and LPS+SLIGRL groups at that time point. Data are expressed as mean ± SE; each group consists of 5-6 animals. Solid bars, LPS; open bars, SLIGRL; striped bars, LPS+SLIGRL.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study clearly demonstrate that intranasal delivery of the PAR2-activating peptide SLIGRL did not cause any inflammation in the airways and, importantly, it inhibited the marked immune cell inflammatory response triggered by LPS. Previous studies have demonstrated that intranasally delivered LPS initiates a significant inflammatory response in murine lungs, including infiltration of neutrophils (13). In the present study LPS caused a similar inflammatory response with very high neutrophil numbers in BALF peaking at 48 h after challenge and then declining over the next 24 h. The significant increase the number of macrophages at 48 and 72 h presumably reflects the resolution of inflammation and phagocytosis of neutrophils by macrophages. This differential cellular response provides both a useful model to compare the effects of different inflammatory stimuli, and an assay for anti-inflammatory drugs. Unlike LPS, SLIGRL did not cause any significant inflammatory cell infiltration in this model. Our additional finding that SLIGRL abrogated the inflammatory cellular response to LPS added further weight to our suggestion that PAR2 agonists may be useful anti-inflammatory drugs in the lungs (1).

Because alterations in MMP levels have been observed in both murine BAL following LPS administration (18), and in conditioned medium from human airway epithelial cells stimulated with PAR2 agonists (11), we performed similar experiments in the present study. Our zymographic analysis of BALF from the same three groups of mice used for the cell infiltration studies revealed two consistent bands of gelatinolytic activity with approximate molecular weights of 115 and 70 kD. The 115-kD band almost certainly corresponds to MMP-9 and we confirmed this using Western blot analysis. The failure of MMP-9 to migrate to its predicted molecular weight of 105 kD (13) may be due to binding of MMP-9 to another protein. For example, human neutrophil-derived MMP-9 is usually bound in an SDS-stable complex to neutrophil gelatinase B-associated lipocalin, giving MMP-9 an apparent molecular weight of 120 kD rather than 92 kD (19). The 70-kD band detected in our studies here presumably represents MMP-2, and because the levels of this enzyme were not dramatically altered in any of the groups, except in the LPS-treated group at 3 h (see below), we did not examine the identity of this enzyme further. The marked increase in MMP-9 activity in BALF from LPS-treated mice agrees well with previous reports using this and similar models of lung injury (18, 20). Also, using cultured cell populations, Vliagoftis and colleagues (11) demonstrated that PAR2 agonists cause an increase in levels of MMP-9 in culture media. Based on their findings, we had anticipated a similar effect of SLIGRL in vivo. Our finding that SLIGRL alone did not alter gelatinolytic activities in BAL may represent a species difference, but more likely points to the more complicated in vivo environment and the differences between cells in vitro and in vivo.

Because gelatin zymography has been used to detect 23 kD trypsin in murine lung tissue (14) and trypsin levels have been shown to be elevated in other inflammatory conditions (1, 24), we were surprised that we were unable to detect any such activity in BALF from LPS-treated mice. Trypsin can be detected in lung homogenates using zymography and immunoblotting techniques in these mice, but levels appear unchanged following LPS administration (data not shown). Thus, it appears that in this model of inflammation, release of trypsin does not occur. Because PARs detect serine proteases like trypsin but not MMPs, these receptors are more likely to be involved in serine protease homeostasis, a possibility we are presently exploring using other models of pulmonary disease.

The anti-inflammatory effect of PAR2 activation in the lungs raises the question of how this effect is mediated. PAR2 is expressed by a variety of cell types present in the lung, including epithelial cells (7), endothelial cells (25), and vascular (25) and airway smooth muscle (7). Furthermore, PAR2 is expressed by neutrophils (26) and other inflammatory cells (27). However, little is known about the effects of PAR2 activation on immune cell function, and therefore we can only speculate about which of these cells, or the interactions between them, are important in the anti-inflammatory effect of SLIGRL. For example, activation of PAR2 on these inflammatory cells may directly inhibit their ability to either release chemotactic mediators or migrate through the submucosal tissues to reach the airway lumen. By contrast, it is established that activation of epithelial PAR2 in murine isolated airway preparations causes release of prostaglandin E₂ (PGE₂) by the epithelium (7, 16). Because PGE₂ inhibits a variety of pathways relevant to pulmonary inflammation (1), it is possible that the effects of SLIGRL observed here were mediated via generation of PGE₂. Notably, in a similar model of murine pulmonary inflammation, inhibitors of prostaglandin synthesis have been shown to amplify, and PGE₂ to inhibit, the cellular infiltration following LPS administration (28). Finally, SLIGRL has been reported in nonmurine tissues to activate unidentified receptors in some tissues (1). Therefore, we cannot exclude the possibility that the anti-inflammatory effects of SLIGRL were in some way unrelated to activation of PAR2. We stress, however, that regardless of possible alternative sites of action of SLIGRL, it is clear that this peptide would have activated airway epithelial PAR2 in our hands and that no proinflammatory effects were observed.

The inhibition of LPS-induced pulmonary neutrophilia by SLIGRL was delayed, suggesting either a long-term effect, or an acute effect that profoundly influences later processes. Because SLIGRL is likely to be metabolized by peptidases in the lung, the latter suggestion is appealing. Co-administration of SLIGRL and LPS had only two effects in the short term that were different from LPS alone in this study and occurred only at the 3-h time point: potentiation of MMP-9 and abolition of the smaller LPS-induced increase in MMP-2. We speculate that these small differences at the early part of the inflammatory response may, in some way, be causally related to the subsequent dampening of immune cell influx into the airways. MMP-9 is not thought to be important in LPS-induced neutrophilia because the number of neutrophils found in BAL is similar in wild-type and MMP-9 gene knockout mice (23). This finding agrees with our observation that, although SLIGRL co-administration with LPS increased the level of MMP-9 at the 3-h time point, there was no difference in the numbers of neutrophils retrieved in BALF. Hence, an elevation of the activity of this enzyme at the 3-h time point does not seem a likely cause of the subsequent inhibition of neutrophil influx. However, inhibition of MMPs has been shown to be anti-inflammatory in other models of pulmonary inflammation (28). Therefore, it is possible that the abolition of the 3-h increase in LPS-induced MMP-2 by SLIGRL may have reduced the potential for further neutrophil trafficking beyond this time point.

In conclusion, intranasal administration of SLIGRL inhibits LPS-driven recruitment of neutrophils into the airways. Although the precise mechanism underlying this effect is unknown, our findings strongly support our earlier proposal (1) that PAR2 agonists may have therapeutic potential in pulmonary inflammatory diseases.