Protease-Activated Receptor- Activation

【摘要】  We have investigated the specific contribution of protease-activated receptor-2 (PAR2) to host defense during Porphyromonas gingivalis infection. Culture supernatants from P. gingivalis strains 33277 and W50 provoked Ca2+ mobilization in cells transfected with PAR2 (PAR2-KNRK) and desensitized the subsequent responses to PAR2-selective agonist. In addition, culture supernatants of P. gingivalis E8 (RgpA/RgpB double knockout) did not cause calcium response in PAR2-KNRK cells, evidencing the involvement of the arginine-specific cysteine proteases RgpA and RgpB in PAR2 activation by P. gingivalis. Injection of P. gingivalis into mouse subcutaneous chambers provoked an increased proteolytic activity, which was inhibited by serine protease inhibitors. Fluids collected from chambers of P. gingivalis-injected mice were able to activate PAR2 and this activation was inhibited by serine protease inhibitors. P. gingivalis inoculation into subcutaneous chambers of wild-type mice induced an inflammatory response that was inhibited by a serine protease inhibitor and was significantly reduced in PAR2-deficient mice. Finally, mice orally challenged with P. gingivalis developed alveolar bone loss, which was significantly reduced in PAR2-deficient mice at 42 and 60 days after P. gingivalis infection. We conclude that PAR2 is activated on P. gingivalis infection, in which it plays an important role in the host inflammatory response.
--------------------------------------------------------------------------------
The protease-activated receptors (PARs) belong to a family of G-protein-coupled, seven-transmembrane-domain receptors.1,2 Activation of PARs occurs through proteolytic cleavage of the extracellular domain, resulting in generation of a new N-terminal tethered ligand.3 To date, four PARs have been identified: PAR1, PAR2, PAR3, and PAR4.4,5 For PAR1, PAR2, and PAR4, synthetic peptide agonists corresponding to the newly created N-terminus are able to activate the receptor in the absence of receptor cleavage.6,7
Although these receptors have a similar mechanism of activation,8 they may have different biological functions and tissue distribution, and they can be activated by different proteases. PAR1, PAR3, and PAR4 are activated by thrombin and are implicated in platelet aggregation.8 Trypsin, mast cell tryptase, neutrophil protease 3, tissue factor/factor VIIa/factor Xa, membrane-tethered serine protease-1, and gingipains have been identified as activators of PAR2.9-11 Some studies have shown that PAR2 participates in inflammatory processes in vivo.12,13 Although PAR2 activation causes edema and granulocyte recruitment in the rat paw,13 the role of PAR2 on mucosal surfaces is not as clearly defined. PAR2 agonists exert protective effects in airways against lipopolysaccharide challenge14 and in the colon in a model of chronic colitis induced by trinitrobenzene sulfonic acid,15 but they are proinflammatory when acutely administered in the colon16 or in the airways17 of mice.
Recently, some studies have suggested a role for PAR2 in periodontitis, a chronic oral inflammation leading to bone and tooth loss, because it was found to be expressed by osteoblasts, oral epithelial cells, and human gingival fibroblasts.9,11,18 Importantly, Lourbakos and colleagues9 reported that bacterial cysteine proteases such as Arg-gingipain (Rgp), which are produced by Porphyromonas gingivalis (a major causative agent of chronic periodontitis), was able to activate PAR2 in oral epithelial cells and to induce the secretion of the proinflammatory cytokine interleukin (IL)-6, which is a potent stimulator of osteoclast differentiation and bone resorption. The production of potent proinflammatory mediators was also shown by Uehara and colleagues11 who demonstrated that a synthetic PAR2 agonist peptide activated the production of IL-8 in human gingival fibroblasts. In a previous study,19 we have evaluated the in vivo effects of PAR2 activation by a selective agonist (SLIGRL-NH2) on periodontal disease in rats. The PAR2 agonist caused periodontitis (inflammation and alveolar bone loss) through a mechanism involving prostaglandin release and matrix metalloproteinase activation. Taken together, these studies strongly suggest a role for PAR2 activation in inducing inflammation and bone resorption during periodontitis. However, a study by Smith and colleagues20 in 2004 suggested an opposite role for PAR2 activation: they hypothesized that PAR2 activation may inhibit bone resorption in the context of periodontal diseases. They showed that PAR2 agonists inhibited osteoclast differentiation, therefore being possibly protective against bone loss signals. Interestingly, Chung and colleagues21 have identified PAR2 as a receptor used by Rgp in induction of human ß-defensin-2 (hBD-2) in oral epithelial cells and have also demonstrated hBD-2 induction by PAR2 peptide agonist. Notably, antimicrobial peptides of the hBD family are part of the innate immune responses that play a role in mucosal defense. No definitive answer on the role of PAR2 in oral mucosal inflammation or in periodontal diseases (pro- or anti-inflammatory effect, pro- or anti-bone loss signal) has so far emerged. Therefore, our strategy was to use a genetic approach, using PAR2-deficient mice, to study the specific contribution of PAR2 activation and proteolytic activity during infection by the proteolytic periodontal pathogen P. gingivalis.

【关键词】  protease-activated receptor- activation

Materials and Methods

Animals

PAR2-deficient (PAR2C/C) and wild-type (WT) littermate (PAR2+/+) mice, both of the C57BL6 background22 originally obtained from Johnson and Johnson Pharmaceutical Research Institute (Spring House, PA), were bred at the University of Calgary animal care facility. Six- to nine-week-old mice were used in the present study. All animals were housed in a temperature-controlled room; food and water were provided ad libitum. The Animal Care and Ethics Committees of the University of Calgary approved all experimental protocols, which followed the guidelines of the Canadian Council on Animal Care.

Bacteria

P. gingivalis strains 33277 , W50 (WT), and E8 (RgpA/RgpB double mutant, obtained from Dr. J. Aduse-Opoku, Queen Mary??s School of Medicine and Dentistry, London, UK) were grown on anaerobic blood agar plates in an anaerobic chamber with 85% N2, 5% H2, and 10% CO2. After incubation at 37??C for 7 days, the bacteria were collected and suspended in Schaedler broth (Difco Laboratories, Detroit, MI) to a final optical density of 1.2 (109 CFU/ml) at 660 nm. Treponema denticola strain ATCC 35405 was obtained from the culture collection at the University of Toronto. It was grown in modified new oral spirochete medium23 for 3 days (mid-exponential phase) to a concentration of 2.8 x 109 cells/ml, as determined by microscopic count.

Subcutaneous Chamber

Coil-shaped chambers were prepared from 0.5-mm stainless-steel wire and were surgically implanted in the subcutaneous tissue of the dorso-lumbar region of each mouse, as previously described.24 Ten days after implantation, chambers were inoculated with 0.1 ml of P. gingivalis 33277, P. gingivalis W50, or P. gingivalis E8, suspended in sterile saline (109 cells/ml). Control mice were inoculated with vehicle only. Mice were sacrificed by cervical dislocation at 1 day after bacterial inoculation. Another group of mice was inoculated with 0.1 ml of a solution containing P. gingivalis (109 cells/ml) and 0.3 mg/ml of SBTI (soybean trypsin inhibitor; Sigma, St. Louis, MO) and sacrificed 1 day after.

Subcutaneous Chamber Fluid Analysis

A sample of chamber fluid (100 µl) was aseptically collected from each animal at 1 day after P. gingivalis or vehicle challenge to assess the host inflammatory response. A 10-µl aliquot of each chamber fluid was diluted 10-fold in Turk??s staining solution to determine the total number of inflammatory cells by light microscopy, using a Neubauer counting chamber (Hauser Scientific, Horsham, PA). The rest of the samples were aliquoted and kept for evaluation of proteolytic activity as well as cytokine and prostaglandin release.

Cell Culture

PAR2-expressing Kirsten sarcoma-transformed rat kidney epithelial cells (KNRK: ATCC, Manassas, VA) were propagated in geneticin (0.6 mg/ml)-containing medium (Dulbecco??s modified Eagle??s medium, 10% fetal bovine serum, 1% Penstrep) to maintain vector-selective pressure. Nontransfected KNRK cells were grown in geneticin-free medium. Cells at 90% confluence in 80-cm2 flasks (Life Technologies, Inc., Grand Island, NY) were rinsed with phosphate-buffered saline, lifted with nonenzymatic cell dissociation fluid, and pelleted before resuspension in 1 ml of Hanks?? balanced salt solution (pH 7.4), 2.5 µl of sulfinpyrazone (100 mmol/L), 1 µl of a 20% Pluronic F-127 solution, and 10 µl of 2.5 mg/ml Fluo-3 acetoxymethylester (Molecular Probes, Inc., Eugene, OR). The final solution was incubated at room temperature while shaking gently for 25 minutes. Cells were then washed three times and resuspended in calcium assay buffer (150 mmol/L NaCl, 3 mmol/L KCl, 1.5 mmol/L CaCl2, 10 mmol/L glucose, 20 mmol/L HEPES, 0.25 sulfinpyrazone, pH 7.4).

Calcium Signaling Assay

Determination of intracellular calcium mobilization was performed as previously described.25 Briefly, fluorescence measurements were performed on a fluorescence spectrometer 650-10S (Perkin-Elmer, Norwalk, CT). Suspensions of cells (5 x 105 cells/ml) loaded with the calcium fluorochrome fluo-3 (Molecular Probes, Inc.) were placed in 4-ml cuvettes, stirred with a magnetic flea bar, and maintained at 24??C.

The signal produced by PAR2-expressing KNRK or nontransfected KNRK cells was measured after the addition of culture supernatants (40 µl) of P. gingivalis 33277, W50, or E8, and T. denticola 35405. In addition, we also evaluated the signal produced by PAR2-expressing KNRK with the addition of supernatants from chamber fluid samples, previously incubated with either 10 nmol/L SBTI (trypsin inhibitor), 1 mmol/L PMSF (phenylmethyl sulfonyl fluoride) (serine proteinase inhibitor, Sigma), 1 mmol/L TLCK (N-p-tosyl-L-lysine chloromethyl ketone, cysteine proteinase inhibitor; Sigma), or 1 mmol/L leupeptin (Rgp cysteine proteinase inhibitor, Sigma) for 10 minutes. The calcium mobilization after the addition of culture supernatants of P. gingivalis 33277, W50, or E8 preincubated with either 1 mmol/L TLCK or 1 mmol/L leupeptin for 10 minutes was evaluated. The calcium responses to known PAR2 agonists were also evaluated. The results are expressed as percentages (% A23187) of the fluorescence emission (E530) caused by 2-µmol/L concentrations of the calcium ionophore A23187. The PAR2 agonist peptide was synthesized by the Peptide Synthesis Facility (University of Calgary, Calgary, AB, Canada). All experiments were repeated three to five times, for at least eight samples.

Proteolytic Activity

After centrifugation of the samples from the fluid chambers, supernatants were collected and assayed for proteolytic activity using a Fluoroskan Ascent fluorimeter (Thermo Electron, Franklin, MA). Besides testing the proteolytic activity of the supernatants of chamber fluid samples, we also evaluated the effect of preincubation (10 minutes) of the samples with either 10 nmol/L SBTI (trypsin inhibitor), 1 mmol/L PMSF (serine proteinase inhibitor, Sigma), 1 mmol/L TLCK (serine and cysteine proteinase inhibitor, Sigma), or 1 mmol/L leupeptin (serine and Rgp cysteine proteinase inhibitor, Sigma). Assays were performed at 25??C, with a 355-nm excitation wavelength filter and a 460-nm emission wavelength filter. Fluorescence from wells on the microplate was measured throughout 20 minutes. The hydrolysis of the substrate (5 µl) added to sample solution (75 µmol/L Boc-Gln-Ala-Arg-AMC in 50 mmol/L Tris/20 mmol/L CaCl2 buffer, pH 7.4) was calibrated with the rate of AMC hydrolysis product in a standard reaction mixture using a serial dilution of trypsin (0.05 to 2 U/ml). All assays were performed in triplicate; the range of values observed was always less than 10% of the mean. The results are expressed in units of trypsin per milliliter (U/ml).

Measurement of Prostaglandin-E2, Interferon (IFN)-, IL-1, IL-6, and IL-10 Levels

Prostaglandin-E2, IFN-, IL-1, IL-6, and IL-10 levels in the supernatants of chamber fluid samples collected from mice that received inoculation of P. gingivalis 33277 or vehicle were determined by using commercially available enzyme-linked immunosorbent assay kits according to manufacturer??s instructions (R&D Systems, Minneapolis, MN). The concentration of the inflammatory mediators was determined using the Softmax data analysis program (Molecular Devices, Menlo Park, CA).

Oral Infection

A total of 0.2 ml of P. gingivalis 33277 suspended in sterile saline (109 cells) was given to each mouse via a gavage needle every other day for a total of 4 days, in part by gavage, and by local application in the oral cavity. Mice were then allowed free access to standard mouse chow and water. The mice were sacrificed by cervical dislocation 42 and 60 days after the last bacterial administration. Ten sham-infected (sterile saline-injected) and 10 P. gingivalis-infected mice were used at each time point in each animal group (PAR2C/C or WT mice). Mandibles were removed, hemisected, exposed to NaOH (2N), and then mechanically defleshed.

Alveolar Bone Loss

Horizontal bone loss around the mandibullary molars was assessed by measuring the distance between the cementoenamel junction and alveolar bone crest under a dissecting microscope (x40). Measurements of bone level were done at seven sites on the lingual side of the left and right mandibula molars, and a total of 14 measurements per mouse were done three times in a random, blinded protocol by one evaluator.

Statistical Analysis

One-way analysis of variance was used to compare means among groups. In case of significant differences among the groups, posthoc two-group comparisons were assessed with Tukey-Kramer test. A P value <0.05 was considered statistically significant. Data are expressed as mean ?? SE.

Results

Proteases Released by P. gingivalis Activate PAR2

To evaluate the role of proteolytic activity produced by the bacteria itself, we examined the effects of P. gingivalis 33277, P. gingivalis W50, and P. gingivalis E8 culture supernatants on Ca2+ mobilization in PAR2-KNRK cells (Figure 1) . We first confirmed that in the KNRK cell line, no calcium mobilization was observed after addition of any of the P. gingivalis culture supernatants or trypsin (data not shown). Then, we showed that P. gingivalis culture supernatants (40 µl) from strains 33277 and W50 provoked Ca2+ mobilization in PAR2-KNRK cells. However, the calcium response induced by culture supernatants of P. gingivalis (33277 and W50) was significantly lower than the response to 2 µmol/L of the selective agonist SLIGRL-NH2 (Figure 1A) . Preincubation of PAR2-KNRK cells with culture supernatants of P. gingivalis (33277 and W50) significantly reduced the subsequent response of those cells to SLIGRL-NH2 (Figure 1A) . We demonstrated that after desensitization of PAR2 receptor by two sequential treatments with trypsin for 5 minutes, the addition of P. gingivalis 33277 culture supernatant did not cause any calcium response in PAR2-KNRK cells (Figure 1B, XI) . Taken together, these results show that proteases released by P. gingivalis 33277 and P. gingivalis W50 activate PAR2 in PAR2-transfected cells.

Figure 1. Calcium response in PAR2-expressing KNRK cells to the addition of culture supernatant samples from P. gingivalis (P.g.) 33277, P. gingivalis W50, P. gingivalis E8, or T. denticola (T.d.). A: Effects of the PAR2 agonist SLIGRL-NH2 (2 µmol/L) or of culture supernatants from P. gingivalis 33277 or P. gingivalis W50 on calcium mobilization in PAR2-transfected KNRL cells. Some cells were preincubated with the culture supernatants of P. gingivalis 33277 or P. gingivalis W50 (shown at the top of the bars) in the presence or not of 1 mmol/L TLCK or 1 mmol/L leupeptin. *Significant difference versus calcium response generated by SLIGRL-NH2. Significant difference versus calcium response generated by the same sample without protease inhibitor treatment. Increases in intracellular calcium were monitored as a percentage (% A23187) of the fluorescence emission (E530) caused by 2-µmol/L concentrations of the calcium ionophore A23187. Means ?? SEM from eight separate experiments. B: Schematic figures showing the calcium response in PAR2-expressing cells (KNRK) to the addition of PAR2 agonists SLIGRL-NH2 (I), trypsin (VI), or culture supernatant samples from P. gingivalis 33277 (II, VII), P. gingivalis W50 (III, VIII), P. gingivalis E8 (IV, IX), or T.d. (V, X) and subsequent calcium response to SLIGRL-NH2 (2 µmol/L) or trypsin (2 nmol/L) 5 minutes later. XI represents the desensitization of PAR2 receptor by two sequential treatments with trypsin for 5 minutes. Representative data from three experiments.

The involvement of the cysteine proteases Arg-gingipain in PAR2 activation was delineated by two different approaches. First we used the culture supernatants from strains P. gingivalis W50 (WT) and E8 (RgpA/RgpB double knockout). We found that P. gingivalis W50 culture supernatants (40 µl) provoked Ca2+ mobilization in PAR2-KNRK cells and decreased the subsequent responses to SLIGRL-NH2 or trypsin (Figure 1B, III and VIII) . No significant difference was found between the two strains of P. gingivalis (33277 and W50, Figure 1A ). However, P. gingivalis E8 culture supernatant (40 µl) did not cause calcium response in PAR2-KNRK cells and did not alter the subsequent responses to trypsin or SLIGRL-NH2 (Figure 1B, IV and IX ; not shown for SLIGRL-NH2). Then, we found that preincubation of P. gingivalis 33277 and P. gingivalis W50 culture supernatant with two serine/cysteine protease inhibitors, TLCK and leupeptin, led to a significant increase in the subsequent response to SLIGRL-NH2 compared to untreated supernatants (Figure 1A) . The P. gingivalis 33277 and P. gingivalis W50 culture supernatant-induced PAR2 desensitization was inhibited by serine/cysteine protease inhibitors (leupeptin and TLCK). Taken together, these results implicate the arginine-specific cysteine proteases RgpA and RgpB in PAR2 activation.

The ability to activate PAR2 exerted by P. gingivalis was compared with the effects exerted by T. denticola, a periodontal pathogen that expresses a serine protease in its outer sheath and several endo-acting peptidases in its culture supernatants. The T. denticola culture supernatant did not cause calcium signal in PAR2-transfected cells (Figure 1B, V) . However, the trypsin response of PAR2-transfected cells pre-exposed to T. denticola culture supernatant was diminished (Figure 1B, X) , whereas the response to the peptide agonist SLIGRL-NH2 was similar to that of cells only exposed to the peptide (Figure 1B, V) . This suggests that proteases released in the supernatant of T. denticola cultures do not activate PAR2 but may disarm the receptor for subsequent activation by trypsin.

P. gingivalis Infection Resulted in Release of Proteolytic Activity

We hypothesized that PAR2 can be activated upon P. gingivalis infection, thereby participating in the host response to infection. To this end, we first evaluated whether P. gingivalis infection provoked the release of proteolytic activity. Trypsin-like activity was tested in fluids collected from skin chambers after inoculation of P. gingivalis 33277, P. gingivalis W50, P. gingivalis E8, or saline. In Figure 2 , WT mice showed a significant (P < 0.001) increase with regards to the proteolytic activity of chamber fluids after inoculation with P. gingivalis 33277, P. gingivalis W50, or P. gingivalis E8 when compared to chamber fluid from mice injected with saline (control). Yet, the proteolytic activity from samples of P. gingivalis E8-treated mice was significantly decreased when compared to the proteolytic activity found in chamber fluids after inoculation with P. gingivalis 33277 or P. gingivalis W50. Preincubation with the serine protease inhibitors SBTI or PMSF led to a significant reduction of the proteolytic activity on P. gingivalis 33277, P. gingivalis W50, or P. gingivalis E8 infection, whereas preincubation with TLCK or leupeptin showed a significant decrease only in the proteolytic activity of samples from P. gingivalis 33277-infected mice. In addition, systemic treatment of mice with the serine protease inhibitor SBTI significantly (P < 0.05) decreased the proteolytic activity because of P. gingivalis 33277 inoculation (mean ?? SEM, 0.44 ?? 0.23 versus 1.56 ?? 0.21 for P. gingivalis 33277 + SBTI and P. gingivalis 33277 alone, respectively, data not shown in the figure).

Figure 2. Proteolytic activity in chamber fluid samples from WT mice inoculated with P. gingivalis 33277, P. gingivalis W50, or P. gingivalis E8 preincubated or not with 10 nmol/L SBTI, 1 mmol/L PMSF, 1 mmol/L TLCK, or 1 mmol/L leupeptin. Means ?? SEM, n = 8 (group). *Significant difference versus proteolytic activity generated by control treatment. Significant difference versus proteolytic activity generated by P. gingivalis 33277 treatment in the same conditions. Significant difference versus proteolytic activity generated by P. gingivalis W50 treatment in the same conditions. Significant difference versus proteolytic activity generated by the same sample without protease inhibitor treatment.

PAR2 Is Activated by P. gingivalis Infection-Induced Proteolytic Activity

To determine whether or not proteolytic activity released on P. gingivalis infection could cleave PAR2, we examined the effects of P. gingivalis-infected chamber fluid samples on subsequent intracellular calcium i mobilization by SLIGRL-NH2 in cultured cells transfected with PAR2 (PAR2-KNRK). We observed that chamber fluid samples (40 µl) collected from WT mice after subcutaneous challenge with P. gingivalis 33277 or P. gingivalis W50 provoked calcium mobilization in PAR2-KNRK but not in nontransfected (KNRK) cells (data not shown). Fluids collected from P. gingivalis 33277- or P. gingivalis W50-infected mice but not from P. gingivalis E8-injected mice were able to significantly (P < 0.05) reduce calcium mobilization in PAR2-transfected KNRK cells generated by SLIGRL-NH2 (mean ?? SEM, 12.38 ?? 1.35, 12.35 ?? 1.55, 22.10 ?? 2.12 for P. gingivalis 33277, P. gingivalis W50, and P. gingivalis E8 treatment, respectively; Figure 3 ), as compared to the effects of saline-injected chambers (mean ?? SEM, 29.13 ?? 0.30). Preincubation of the P. gingivalis 33277-infected or P. gingivalis W50-infected chamber samples with SBTI or PMSF significantly increased the subsequent calcium response to SLIGRL-NH2 (Figure 3) , thereby inhibiting the desensitization to SLIGRL-NH2 response. Preincubation of chamber fluid samples with TLCK or leupeptin also provoked a significant increase in the subsequent calcium response to SLIGRL-NH2, but only in samples from P. gingivalis 33277-infected mice.

Figure 3. Effect of preincubation of chamber fluids collected from P. gingivalis 33277-, P. gingivalis W50-, or P. gingivalis E8-infected mice with 10 nmol/L SBTI, 1 mmol/L PMSF, 1 mmol/L TLCK, or 1 mmol/L leupeptin on the subsequent calcium response generated by the PAR2 agonist SLIGRL-NH2 (2 µmol/L). Means ?? SEM from eight separate experiments. *Significant difference (P < 0.05) versus calcium response generated by the same sample without protease inhibitor treatment. Increases in intracellular calcium were monitored as a percentage (% A23187) of the fluorescence emission (E530) caused by 2-µmol/L concentrations of the calcium ionophore A23187.

Fluid samples collected from mice treated with both P. gingivalis 33277 and SBTI did not provoke a significant reduction in the subsequent calcium signal response to SLIGRL-NH2 (mean ?? SEM, 21.56 ?? 1.72; data not shown), similar to the samples from P. gingivalis E8-treated and control-treated animals. Collectively, these results demonstrated the presence of PAR2-activating proteases upon infection by P. gingivalis.

PAR2 Activation Plays a Pivotal Role in the Host Response to P. gingivalis Infection

Next, we evaluated whether PAR2 activation affected the host response to P. gingivalis infection, by comparing the inflammatory effects of subcutaneous chamber infection with P. gingivalis 33277 in WT and PAR2C/C mice. Figure 4A shows that subcutaneous challenge with P. gingivalis led to a significant (P < 0.001) increase in inflammatory cell infiltration when compared to saline treatment (day 1 after saline or P. gingivalis injection) in WT mice. P. gingivalis-induced increased inflammatory cell counts were significantly diminished in PAR2C/C mice compared to WT. The inflammatory cell count was not significantly increased in P. gingivalis-injected PAR2C/C mice compared with saline-injected PAR2C/C mice (Figure 4A) .

Figure 4. Inflammatory effects of subcutaneous chamber infection with P. gingivalis 33277 in WT and PAR2C/C mice. Means ?? SEM, n = 10 (group). A: Inflammatory cell counts; B: PGE2 levels; C: IFN- levels; D: IL-1ß levels; E: IL-6 levels; and F: IL-10 levels.

Subcutaneous challenge of WT mice with P. gingivalis led to significant (P < 0.05) increases in PGE2, IFN-, IL-1ß, and IL-6 levels when compared with saline treatment (Figure 4, BCE) . No significant difference was observed between WT mice and PAR2C/C mice after saline or P. gingivalis injection for PGE2 (Figure 4B) or IL-6 levels (Figure 4E) . However, although a significant increase in PGE2 was observed on P. gingivalis infection in WT mice, this increase was not present in PAR2C/C mice. Further, IFN- and IL-1ß levels were significantly reduced in PAR2C/C mice treated with P. gingivalis compared with WT mice. For IFN-, the cytokine level in PAR2C/C mice was similar to noninfected mice (saline injected) (Figure 4C) . No significant difference was observed between saline- or P. gingivalis-treated WT or PAR2C/C mice for the levels of IL-10 release (Figure 4F) . Evidently, PAR2 activation contributes to the host response to P. gingivalis infection in the anaerobic chamber model in mice, causing the release of IL-1 and IFN- and stimulating to the recruitment of inflammatory cells.

PAR2 Activation Plays a Pivotal Role in P. gingivalis-Induced Periodontitis

Finally, we evaluated whether PAR2 activation impacts P. gingivalis-induced periodontitis in mice. Satisfactory outcome of the experimental periodontitis model was confirmed by an increased bone loss after 42 (Figure 5, ACC) and 60 days (Figure 5, DCF) after P. gingivalis administration. Both in WT and PAR2C/C mice, P. gingivalis infection caused significant alveolar bone loss. However, PAR2C/C mice showed significantly less alveolar bone loss at both time points compared with WT mice after oral infection.

Figure 5. PAR2 activation role in P. gingivalis-induced periodontitis. A: Alveolar bone loss at 42 days after infection. Means ?? SEM, n = 10 (group). Mandibular lingual aspect of alveolar bone loss in P. gingivalis-infected PAR2C/C (B) and WT (C) mice at 42 days after infection. D: Alveolar bone loss at 60 days after infection. Means ?? SEM, n = 10 (group). Mandibular lingual aspect of alveolar bone loss in P. gingivalis infected PAR2C/C (E) and WT (F) mice at 60 days after infection. The black arrows indicate the areas of greater loss of alveolar bone.

Discussion

Trypsin-like proteolytic activity is able to cleave the N-terminal peptide of PAR2, providing a newly exposed N-terminal end that functions as a tethered ligand, activating the receptor and resulting in intracellular signaling by Gq proteins.3 Pancreatic and extrapancreatic (endothelial and epithelial) trypsins, coagulation factors (FVIIa-FXa), mast cell tryptase, leukocyte proteases (neutrophil proteinase 3), and membrane-tethered serine protease-1 are some of the endogenous proteases that cleave and activate PAR2.9,12,13 In addition to endogenous host enzymes, the arginine-specific proteinases, gingipains-R (HRgpA and RgpB) derived from P. gingivalis, have also been identified as potential activators of PAR2. In vitro studies have reported that gingipains-R can activate PAR2 in human neutrophils,26 human oral epithelial cells,9 and in primary rat calvarial osteoblast-like cells18 through the induction of a dose-dependent increase in i. In fact, gingipains are considered major virulence determinants of the bacterium,27-30 including their activity in periodontal breakdown, disruption of host defense mechanisms, and loss of viability of human fibroblasts and endothelial cells.

In the present study, we used KNRK cells transfected or not with PAR2 combined with desensitization experiments using previous exposure to PAR2 agonist to evaluate the role of bacterial proteases from P. gingivalis on PAR2 activation. We show that P. gingivalis culture supernatants provoke Ca2+ mobilization in PAR2-KNRK cells and decrease the subsequent responses to the two PAR2 agonists SLIGRL-NH2 or trypsin. In addition, we demonstrate that after desensitization of the receptor by trypsin, the addition of P. gingivalis culture supernatant did not cause calcium response in PAR2-KNRK cells, excluding the possible concurrent activation of other receptors. Because trypsin signal to PAR2-transfected KNRK cells is exclusively mediated by PAR2,31 our results demonstrate unequivocally that P. gingivalis supernatants and trypsin act on the same receptor, namely PAR2.

P. gingivalis produces at least eight different endopeptidases and a number of exopeptidases that belong to the cysteine-, serine-, and metallo-classes of peptidases.29 It is well documented that among these enzymes, the cysteine proteinases, arginine-gingipain (Rgp) and lysine-gingipain (Kgp), which specifically cleave after arginine and lysine, respectively, have trypsin-like activity and are responsible for most of P. gingivalis-associated proteolytic activity.32 The gingipains are associated with both the cell surface and the outer membrane vesicles of the bacterium and can be produced in large amounts.33 In the present study, Arg-gingipain-specific involvement in PAR2 activation was delineated by using the culture supernatants of P. gingivalis W50 (WT) and E8 (RgpA/RgpB double knockout) strains. We showed the lack of PAR2 activation by the Arg-gingipain knockout strain and inhibition of PAR2 activation by incubating WT P. gingivalis culture supernatants with serine/cysteine inhibitors. These results, which evidence the capacity of P. gingivalis-released RgpA and RgpB in activating PAR2, are in agreement with previous studies in the literature that show that the purified gingipains-R (HRgpA and RgpB) are able to activate PAR2.9,18,21,26

Recognizing the polymicrobial etiology of periodontitis, we compared the ability exerted by P. gingivalis to activate PAR2 with the activity of T. denticola, an important periodontal pathogen that co-habits and forms biofilms with P. gingivalis and that also expresses proteinases with trypsin-like activity.29,34 Unlike P. gingivalis, T. denticola culture supernatant did not activate PAR2, as it did not cause calcium signaling and did not desensitize PAR2 activation by SLIGRL-NH2. Notably, both T. denticola outer membranes and its major outer sheath protein are known to cause calcium mobilization in fibroblasts, because of influx through the plasma membrane, but also to suppress subsequent agonist-induced calcium release from internal stores.35,36 In this study, the T. denticola culture supernatant neither stimulated a calcium signal nor desensitized PAR2 activation by SLIGRL-NH2. Yet, T. denticola seemed to disarm the receptor because subsequent activation by trypsin was inhibited. This result suggests a possible inhibitory effect of T. denticola proteases or peptidases on proteolysis-mediated PAR2 activation. This disarming effect could be due to the subtilisin family serine protease dentilisin (PrtP), which is known to be shed with outer sheath vesicles into the culture medium in aging cultures. Similar observations have been made for Pseudomonas aeruginosa elastase, which disarms PAR2 for subsequent activation by trypsin but not by the synthetic receptor activating peptide SLIGKV-NH2.37 It can be hypothesized that some pathogens may silence PAR2 functions and avoid the organization of an effective host inflammatory response.

In addition to finding that culture supernatants of P. gingivalis were able to signal to PAR2, we also show that increased proteolytic activity was observed in subcutaneous chambers on infection by P. gingivalis (Figure 2) . Our results suggest that not only can Arg-gingipains be responsible for the proteolytic activity observed on infection but serine proteases might also be involved. First, infection by the gingipain RgpA/RgpB-defective mutant E8 also provoked an increased proteolytic activity in infected chambers, although this activity was significantly lower than what was observed after WT strain infection. Second, serine protease inhibition by SBTI, a known reversible inhibitor of trypsin, chymotrypsin, and elastase, and by PMSF, an irreversible serine inhibitor that does not affect gingipains, resulted in a significant decrease in the proteolytic activity of P. gingivalis-infected subcutaneous chambers. Mixed serine/cysteine protease inhibitors (TLCK and leupeptin) were not more effective at reducing P. gingivalis infection-induced proteolytic activity than serine protease-only inhibitors such as PMSF. Similar results were obtained in terms of PAR2 activation, in which serine protease inhibitors (SBTI and PMSF) were as effective or more effective as serine/cysteine protease inhibitors (leupeptin or TLCK) to inhibit PAR2 desensitization by P. gingivalis-infected chamber fluids. At this stage, it is not clear whether the serine proteases present in chamber fluids of P. gingivalis-infected mice were generated by the host or by the bacteria itself. P. gingivalis infection could trigger the host to generate PAR2-activating proteases as it was recently demonstrated for Citrobacter rodentium infection in mouse colon.38 It is also possible that the proteolytic activity observed in chamber fluids from P. gingivalis-infected mice is due solely to the host inflammatory reaction and is independent of the cause that triggers inflammation. It is interesting to note that infection by the RgpA/RgpB double-mutant E8 caused significantly less proteolytic activity in chamber fluids than WT strains and that E8-derived chamber fluids were not able to desensitize PAR2 activation. This suggests that the presence of PAR2-activating proteases in infected tissues might also depend on the virulence of the infection (E8 being less virulent than the other two strains, W50 and 33277) and potentially the release of Arg-gingipains. It is possible that Arg-gingipains in the subcutaneous chambers activate proteases of the coagulation cascade, which then could account for the increased proteolytic activity in the chambers of W50 and 33277 compared to Arg-gingipain mutant E8. As a matter of fact, studies have shown that P. gingivalis Arg-gingipains are able to activate prothrombin and coagulation factor X.39,40 Interestingly, PAR2 has been described as the endogenous receptor mediating the effects of factor Xa in endothelial cells.41 Therefore, P. gingivalis Arg-gingipains could account for PAR2 activation both directly (by cleaving the receptor) and indirectly (by inducing the release of PAR2-activating proteases).

A number of studies have suggested a dual role for PAR2 in inflammatory processes. Although PAR2 activation seems to be involved in leukocyte migration, inflammation of joints, skin, and kidney, and allergic inflammation of airways,3 it has also been linked to some protective anti-inflammatory activities via the epithelium and vascular endothelium in the airways,14 in the mucosal tissues of the gastrointestinal tract,15 or in response to cardiovascular injury.42 Conversely, the results from the present study clearly demonstrate that upon infection, PAR2 plays a proinflammatory role in gingival tissues. Our data demonstrate that in the presence of P. gingivalis, PAR2 activation leads to a significant increase in inflammatory cell infiltration, which was inhibited by SBTI and significantly reduced in PAR2-deficient (PAR2C/C) mice. In addition, increased levels of prostaglandin E2, IFN-, IL-6, and IL-1ß, important mediators of alveolar bone loss, were found in chambers of P. gingivalis-treated mice compared with control saline-treated mice. All these inflammatory mediators, with the exception of IL-6, were either not increased or significantly inhibited in PAR2C/C mice infected by P. gingivalis. Interestingly, although IL-6 production was significantly increased in subcutaneous chambers on P. gingivalis infection, no difference was found with regards to IL-6 levels released on P. gingivalis infection in PAR2C/C mice compared to WT mice. This result suggests that IL-6 production is probably not mediated by PAR2 activation. Conversely, the study by Loubarkos and colleagues9 has shown in an oral epithelial cell line that expresses both PAR1 and PAR2 that treatment with selective agonists of PAR1 and PAR2, or Arg-gingipain, induced secretion of IL-6. It could be hypothesized that in vivo gingipain might act preferentially on PAR1 rather than on PAR2 to induce IL-6 production. Additionally, P. gingivalis may induce IL-6 production in the host by a mechanism that does not involve proteolytic activity in vivo.

The present study also provides evidence for an important role of PAR2 in alveolar bone loss in mice orally challenged with P. gingivalis because a significant prevention of alveolar bone loss was verified in PAR2C/C mice at 42 and 60 days after infection. These results clearly show that PAR2 is activated upon P. gingivalis infection, in which it contributes to the host inflammatory response. PAR2-mediated activation of cells during infection is an intriguing new mechanism in bacterial pathogenicity that remains to be studied in depth. Because PAR2 is a one-shot receptor being desensitized after proteolysis, it can be considered as an emergency or sensor mechanism of the environment, which is activated on infection. However, its possible role in the maintenance of inflammation cannot be overlooked because PAR2 activation mediates the release of inflammatory mediators that, in theory, can up-regulate PAR2 expression,43 therefore resulting in an ultimate increased severity of the disease.

Interestingly, a recent paper by Tancharoen and colleagues,44 reported that the cysteine proteinase RgpB from P. gingivalis elicits the release of neuropeptides by human dental pulp cells through PAR2 activation, therefore suggesting its role also in pulp inflammation. These data also suggest that PAR2 activation in response to P. gingivalis infection could organize an analogous neurogenic inflammatory response.

High levels of proteolytic activity have been found in gingival crevicular fluid from periodontal pockets, where a mixture of endogenous host enzymes and bacterial proteases combine to mediate degradation of connective tissue. Among these enzymes, neutrophil serine proteinase 3, mast cell tryptase, and gingipain have been isolated from the periodontal environment and are known to activate PAR2.28 Our present study further suggests a role for mixed endogenous and bacterial proteases in the pathophysiology of periodontitis. It identifies the activation of PAR2 as one of the significant target of periodontitis-associated proteases and as a major factor in pathogenesis. Therapies focusing on the inhibition of proteinases or, more specifically, the use of PAR2 antagonists, may constitute an important approach for the modulation of an infectious pathology such as periodontal inflammatory disease.

Acknowledgements

We thank Dr. Joe Aduse-Opoku for providing the strains W50 and E8 of P. gingivalis; Drs. Steeve Houle, Kristina K. Hansen, and Morley Hollenberg for expert assistance with cell culture and calcium signaling analysis; Dr. Thomas J. Louie??s laboratory personnel for expert assistance with P. gingivalis culture; Dr. John L. Wallace and Webb McKnight for their valuable assistance in preparing the macroscopic figures; and Kevin Chapman and Laurie Cellars for technical assistance in general.

【参考文献】
  Nystedt S, Emilsson K, Wahlestedt C, Sundelin J: Molecular cloning of a potential proteinase activated receptor. Proc Natl Acad Sci USA 1994, 91:9208-9212

Vu TK, Hung DT, Wheaton VI, Coughlin SR: Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell 1991, 64:1057-1068

Ossovskaya VS, Bunnet NW: Protease-activated receptors: contribution to physiology and disease. Physiol Rev 2004, 84:579-621

Coughlin SR, Vu TK, Hung DT, Wheaton VI: Characterization of a functional thrombin receptor: issues and opportunities. J Clin Invest 1992, 89:351-355

Coughlin SR: Thrombin signaling and protease-activated receptors. Nature 2000, 407:258-264

Cocks TM, Moffatt JD: Proteinase-activated receptors: sentries for inflammation? Trends Pharmacol Sci 2000, 21:103-108

Corvera CU, Dery O, McConalogue K, Bohm SK, Khitin LM, Caughey GH, Payan DG, Bunnett NW: Mast cell tryptase regulates rat colonic myocytes through proteinase-activated receptor 2. J Clin Invest 1997, 100:1383-1393

Coughlin SR, Camerer E: PARticipation in inflammation. J Clin Invest 2003, 111:25-27

Lourbakos A, Potemba J, Travis J, D??Andrea MR, Andrade-Gordon P, Santulli R, Mackie EJ, Pike RN: Arginine-specific proteinase from Porphyromonas gingivalis activates proteinase-activated receptors on human oral epithelial cells and induces interlukin-6 secretion. Infect Immun 2001, 69:5121-5130

Vergnolle N: Modulation of visceral pain and inflammation by protease-activated receptors. Br J Pharmacol 2004, 141:1264-1274

Uehara A, Muramoto K, Takada H, Sugawara S: Neutrophil serine proteinases activate human nonepithelial cells to produce inflammatory cytokines through proteinase-activated receptor 2. J Immunol 2003, 170:5690-5696

Vergnolle N, Wallace JL, Bunnett NW, Hollemberg MD: Proteinase-activated receptors in inflammation, neuronal signaling and pain. Trends Pharmacol Sci 2001, 22:146-152

Vergnolle N, Hollenberg MD, Sharkey KA, Wallace JL: Characterization of the inflammatory response to proteinase-activated receptor-2 (PAR-2)-activating peptides in the rat paw. Br J Pharmacol 1999, 127:1083-1090

Cocks TM, Fong B, Chow JM, Anderson GP, Frauman AG, Goldie RG, Henry PJ, Carr MJ, Hamilton JR, Moffatt JD: A protective role for protease-activated receptors in the airways. Nature 1999, 398:156-160

Fiorucci S, Mencarelli A, Palazzetti B, Distrutti E, Vergnolle N, Hollenberg MD, Wallace JL, Morelli A, Cirino G: Proteinase-activated receptor 2 is an anti-inflammatory signal for colonic lamina propria lymphocytes in a mouse model of colitis. Proc Natl Acad Sci USA 2001, 98:13936-13941

Cenac N, Coelho AM, Nguyen C, Compton S, Andrade-Gordon P, MacNaughton WK, Wallace JL, Hollenberg MD, Bunnett NW, Garcia-Villar R, Bueno L, Vergnolle N: Induction of intestinal inflammation in mouse by activation of proteinase-activated receptor-2. Am J Pathol 2002, 161:1903-1915

Schmidlin F, Amadesi S, Dabbagh K, Lewis DE, Knott P, Bunnett NW, Gater PR, Geppetti P, Bertrand C, Stevens ME: Protease-activated receptor 2 mediates eosinophil infiltration and hyperreactivity in allergic inflammation of the airway. J Immunol 2002, 169:5315-5321

Abraham LA, Chinni C, Jenkins AL, Lourbakos A, Ally N, Pike RN, Mackie EJ: Expression of protease-activated receptor-2 by osteoblasts. Bone 2000, 26:7-14

Holzhausen M, Spolidorio LC, Vergnolle N: Proteinase-activated receptor-2 (PAR2) agonist causes periodontitis in rats. J Dent Res 2005, 84:154-159

Smith R, Ransjo M, Tatarczuch L, Song SJ, Pagel C, Morrison JR, Pike RN, Mackie EJ: Activation of protease-activated receptor-2 leads to inhibition of osteoclast differentiation. J Bone Miner Res 2004, 19:507-516

Chung WO, Hansen SR, Rao D, Dale BA: Protease-activated receptor signaling increases epithelial antimicrobial peptide expression. J Immunol 2004, 173:5165-5170

Damiano BP, Cheung WM, Santulli RJ, Fung-Leung WP, Ngo K, Ye RD, Darrow AL, Derian CK, de Garavilla L, Andrade-Gordon P: Cardiovascular responses mediated by protease-activated receptor-2 (PAR-2) and thrombin receptor (PAR-1) are distinguished in mice deficient in PAR-2 or PAR-1. J Pharmacol Exp Ther 1999, 288:671-678

Dawson JR, Ellen RP: Tip-oriented adherence of Treponema denticola to fibronectin. Infect Immun 1990, 58:3924-3928

Gyurko R, Boustany G, Huang PL, Kantarci A, Van Dyke TE, Genco CA, Gibson FC: Mice lacking inducible nitric oxide synthase demonstrate impaired killing of Porphyromonas gingivalis. Infect Immun 2003, 71:4917-4924

Compton SJ, Cairns JA, Palmer KJ, Al-Ani B, Hollenberg MD, Walls AF: A polymorphic protease-activated receptor 2 (PAR2) displaying reduced sensitivity to trypsin and differential responses to PAR agonists. J Biol Chem 2000, 275:39207-39312

Lourbakos AC, Chinni P, Thompson J, Potempa J, Travis J, Mackie EJ, Pike RN: Cleavage and activation of proteinase-activated receptor-2 on human neutrophils by gingipain-R from Porphyromonas gingivalis. Fed Eur Bone Soc Lett 1998, 435:45-48

Imamura T: The role of gingipains in the pathogenesis of periodontal disease. J Periodontol 2003, 74:111-118

Kadowaki T, Yoneda M, Okamoto K, Maeda K, Yamamoto K: Purification and characterization of a novel arginine-specific cysteine proteinase (argingipain) involved in the pathogenesis of periodontal disease from the culture supernatant of Porphyromonas gingivalis. J Biol Chem 1994, 269:21371-21378

Potemba J, Banbula A, Travis J: Role of bacterial proteinases in matrix destruction and modulation of host responses. Periodontol 2000 2000, 24:153-192

Travis J, Potempa J: Bacterial proteinases as targets for the development of second-generation antibiotics. Biochim Biophys Acta 2000, 1477:35-50

Al-Ani B, Saifeddine M, Kawabata A, Renaux B, Mokashi S, Hollenberg MD: Proteinase-activated receptor 2 (PAR(2)): development of a ligand-binding assay correlating with activation of PAR(2) by PAR(1)- and PAR(2)-derived peptide ligands. J Pharmacol Exp Ther 1999, 290:753-760

Potempa J, Pike R, Travis J: Titration and mapping of the active site of cysteine proteinases from Porphyromonas gingivalis (gingipains) using peptidyl chloromethanes. Biol Chem 1997, 378:223-230

Potempa J, Pike R, Travis J: The multiple forms of trypsin-like activity present in various strains of Porphyromonas gingivalis are due to the presence of either Arg-gingipain or Lys-gingipain. Infect Immun 1995, 63:1176-1182

Ellen RP, Galimanas VB: Spirochetes at the forefront of periodontal infections. Periodontol 2000 2005, 38:13-32

Ko KS-C, Glogauer M, McCulloch CAG, Ellen RP: Treponema denticola outer membrane inhibits calcium flux in gingival fibroblasts. Infect Immun 1998, 66:703-709

Wang Q, Ko KS, Kapus A, McCulloch CAG, Ellen RP: A spirochete surface protein uncouples store-operated calcium channels in fibroblasts. J Biol Chem 2001, 276:23056-23064

Dulon S, Leduc D, Cottrell GS, D??Alayer J, Hansen KK, Bunnett NW, Hollenberg MD, Pidard D, Chignard M: Pseudomonas aeruginosa elastase disables proteinase-activated receptor 2 in respiratory epithelial cells. Am J Respir Cell Mol Biol 2005, 32:411-419

Hansen KK, Sherman PM, Cellars L, Andrade-Gordon P, Pan Z, Baruch A, Wallace JL, Hollenberg MD, Vergnolle N: A major role for proteolytic activity and proteinase-activated receptor-2 in the pathogenesis of infectious colitis. Proc Natl Acad Sci USA 2005, 102:8363-8368

Inamura T, Potempa J, Tanase S, Travis J: Activation of blood coagulation factor X by arginine-specific cysteine proteinases (gingipain-Rs) from Porphyromonas gingivalis. J Biol Chem 1997, 272:16062-16067

Inamura T, Bambula A, Pereira PJ, Travis J, Potempa J: Activation of human pro-thrombin nu arginin-specific cysteine proteinases (gingipains R) from Porphyromonas gingivalis. J Biol Chem 2001, 276:18984-18989

Camerer E, Kataoka H, Kahn M, Lease K, Coughlin SR: Genetic evidence that protease-activated receptors mediate factor Xa signaling in endothelial cells. J Biol Chem 2002, 277:16081-16087

Napoli C, Cicala C, Wallace JL, de Nigris F, Santagada V, Caliendo G, Franconi F, Ignarro LJ, Cirino G: Protease-activated receptor-2 modulates myocardial ischemia-reperfusion injury in the rat heart. Proc Natl Acad Sci USA 2000, 97:3678-3683

Nystedt S, Ramakrishnan V, Sundelin J: The proteinase-activated receptor 2 is induced by inflammatory mediators in human endothelial cells. Comparison with the thrombin receptor. J Biol Chem 1996, 271:14910-14915

Tancharoen S, Sarker KP, Imamura T, Biswas KK, Matsushita K, Tatsuyama S, Travis J, Potempa J, Torii M, Maruyama I: Neuropeptide release from dental pulp cells by RgpB via proteinase-activated receptor-2 signaling. J Immunol 2005, 174:5796-5804

作者单位：From the Department of Pharmacology and Therapeutics,* Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada; Canadian Institute for Health Research Group in Matrix Dynamics and Dental Research Institute, University of Toronto, Toronto, Canada; the Department of Periodontology and Ora