|
|
||||||||
,






* Divisions of Oral Immunology,
Periodontology and Endodontology, and
Oral Microbiology, Department of Oral Biology, and
Division of Oral Diagnosis, Department of Oral Medicine and Surgery, Tohoku University Graduate School of Dentistry, Sendai, Japan;
¶ Tokyo New Drug Research Laboratories II, Pharmaceutical Division, Kowa Company Limited, Japan; and
![]()
The Renal Unit, Harvard Medical School/Massachusetts General Hospital, Boston
1Correspondence: Division of Oral Immunology, Department of Oral Biology, Tohoku University Graduate School of Dentistry, 4-1 Seiryo-machi, Aoba-ku, Sendai 980-8575, Japan. E-mail: sugawars{at}mail.tains.tohoku.ac.jp
| ABSTRACT |
|---|
|
|
|---|
and liver enzymes induced by P. acnes and LPS were abolished by anti-Gr-1 treatment, and concomitantly, liver injury (necrotic change and granuloma formation) and Gr-1+ cell infiltration into the liver were prevented by the treatment. A deficiency of PAR2 in mice significantly impaired IL-18 induction by treatment with P. acnes and LPS, and only slight pathological changes in hepatic tissues occurred in the PAR2-deficient mice treated with P. acnes and LPS. Furthermore, coadministration of exogenous murine PR3 or a synthetic PAR2 agonist (ASKH95) with LPS in the anti-Gr-1-treated mice restored the serum IL-18 levels to those in control mice treated with P. acnes and LPS. These results indicate that neutrophil recruitment and PAR2 activation by neutrophil serine proteases are critically involved in the induction of IL-18 and IL-18-dependent liver injury in vivo.
Key Words: PMN serine proteases G protein-coupled receptor cytokine inflammation
| INTRODUCTION |
|---|
|
|
|---|
(IFN-
)-inducing factor [1
] and is a multifunctional regulator of innate and acquired immune responses through its activation of T helper cell type 1 (Th1) and Th2 responses [2
3
4
5
]. IL-18 has also been suggested to be a potent proinflammatory cytokine that regulates autoimmune and inflammatory diseases [2
3
4
]. IL-18 is produced intracellularly as an inactive precursor form and secreted as an active cytokine after cleavage by caspase-1, originally designated IL-1ß-converting enzyme [2
3
4
]. Furthermore, IL-18 is identified in immune cells (activated macrophages, dendritic cells, and Kupffer cells) and nonimmune cells (keratinocytes, osteoblasts, adrenal cortex cells, epithelial cells, microglial cells, and synovial fibroblasts) [2
3
4
]. This wide range of distribution implies that IL-18 plays physiological roles and acts as a component of immune regulation. Protease-activated receptor (PAR) family members are G protein-coupled receptors, which undergo proteolytic cleavage of the N terminus, thereby exposing tethered ligands and permitting autoactivation of the receptor function so that each receptor can initiate multiple signaling cascades [6 7 8 ]. PAR1, -3, and -4 are activated mainly by thrombin, whereas PAR2 is activated by a number of proteases such as trypsin, mast cell tryptase, and coagulation factors VIIa and Xa, but not by thrombin [6 7 8 ]. PAR2 was identified by Nystedt et al. [9 ] and is expressed in a wide variety of tissues, including the gastrointestinal and respiratory tracts, pancreas, kidney, muscles, ovary, and skin, but not in platelets [9 , 10 ]. Also, recent studies suggest that PAR2 regulates several physiological processes, including growth, development, gastrointestinal functions, tissue repair, and inflammation [6 7 8 , 11 ]. Mice that lack the PAR2 gene exhibit diminished ear swelling and infiltration of inflammatory cells in a model of allergic dermatitis [12 ] and are immune to a form of adjuvant-induced arthritis [13 ].
Neutrophil-derived serine proteases, human leukocyte elastase (HLE; EC 3.4.21.37), cathepsin G (Cat G; EC 3.4.21.20), and proteinase 3 (PR3; EC 3.4.21.76), are stored in the azurophilic granules of neutrophils as active enzymes. The major physiological function of the proteases is commonly thought to be the intralysosomal degradation of engulfed cell debris or microorganisms [14
]. It has also become evident that HLE, Cat G, and PR3 play crucial roles in extracellular, proteolytic processes at sites of inflammation. Recently, we found that human oral epithelial cells constitutively express a precursor form of IL-18 and that PR3 induces the secretion of bioactive IL-18 from these epithelial cells in combination with lipopolysaccharide (LPS) after priming with IFN-
[15
]. Subsequently, we showed that PR3 activates cells through the PAR2 pathway [16
]. Human epithelial cells express a secretory leukocyte protease inhibitor (SLPI), which inhibits neutrophil serine proteases, including HLE and Cat G, but not PR3, and we also showed that HLE and Cat G as well as PR3 activate SLPI lacking nonepithelial cells such as human gingival fibroblasts via the PAR2 pathway [17
]. These observations suggest that neutrophil serine proteases have an equal ability to activate nonepithelial cells through PAR2.
IL-18 was cloned from a murine liver cell cDNA library generated from animals primed with heat-killed Propionibacterium acnes and subsequently challenged with LPS [1 ], and sequential treatment with heat-killed P. acnes and LPS was found to induce IL-18-dependent acute liver injury in mice [1 , 18 19 20 ]. Neutrophils play a role in acute inflammation, and activated neutrophils release serine proteases [14 ], which are possible PAR2-activating proteases. As it is unclear that the observations obtained from in vitro studies [15 16 17 ] actually occur in vivo, we extended our investigation of whether neutrophils and PAR2 are involved in the induction of serum IL-18 and IL-18-dependent liver injury using mice treated with heat-killed P. acnes and LPS. Our results show that neutrophil recruitment and activation of PAR2 are critical for the induction of IL-18 in these mice.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Bacteria and reagents
P. acnes was grown in brain-heart infusion medium (Difco Laboratories, Detroit, MI) with L-cysteine and Tween-80, as described previously [21
]. The harvested bacteria were washed with sterile, distilled water, killed by heating at 60°C for 1 h, and then lyophilized. The lyophilized bacteria were next suspended in phosphate-buffered saline (PBS; 5 mg/ml) and used to prime the mice. LPS from Escherichia coli O55:B5 and clodronate (dichloromethylene bisphosphonate) were obtained from Sigma-Aldrich (St. Louis, MO). A novel, synthetic murine PAR2 agonist (ASKH95) was provided by Kowa Company (Tokyo, Japan) [13
]. Nafamostat mesilate (FUT-175), a serine protease inhibitor, was kindly provided by Torii Pharmaceutical Company (Tokyo, Japan). Mouse PR3 cDNA was expressed in insect cells, and the recombinant (r)PR3 was purified. The specificity of rPR3 was confirmed by two specific antibodies, which react with the internal regions of murine PR3. The functional status of rPR3 was demonstrated by proteolytic activity toward a specific substrate and by
1-antitrypsin activity inhibition, demonstrating a mature and functional protein. All other reagents were obtained from Sigma-Aldrich, unless otherwise indicated.
Treatment of mice
The mice were injected intraperitoneally (i.p.) with heat-killed bacteria (1 mg dry weight/mouse), and 7 days later, they were challenged intravenously (i.v.) with various concentrations of LPS. Blood was collected directly into test tubes following their decapitation, and serum was recovered by centrifugation at 2000 g at 4°C, after which it was stored at 80°C until use. The Ethical Board for nonhuman species of the Tohoku University Graduate School of Medicine approved the experimental procedure followed in this study.
Depletion of neutrophils
Hybridoma RB6-8C5, which secretes anti-Gr-1 monoclonal antibody (mAb), was kindly provided by Fujiro Sendo (Yamagata University, Japan) [22
]. Anti-Gr-1 mAb was purified from the ascites fluid of the hybridoma. To deplete neutrophils, 0.25 mg mAb was administered i.p. to the mice. Treatment with this dose of mAb resulted in severe neutropenia for up to 5 days, as assessed by a blood smear test [22
].
Depletion of macrophages
A macrophage "suicide" technique, using liposomes encapsulating clodronate, has been shown to be specific for phagocytotic cells of the mononuclear phagocyte system, and within 1 or 2 days of i.v. injection of such liposomes into mice or rats, phagocytic macrophages have been shown to be depleted completely [23
24
25
]. A suspension of liposomes encapsulating clodronate (clodronate-liposomes) was prepared by a method described previously [23
, 26
]. Briefly, 75 mg phosphatidylcholine and 11 mg cholesterol were dissolved in chloroform in a round-bottomed flask. The thin film that formed on the walls of the flask after rotary evaporation at 37°C was dispersed by gentle shaking for 10 min in 10 ml clodronate solution (200 mg/ml), in 10 mM sodium phosphate buffer, pH 7.4. This suspension was kept for 2 h at room temperature, then sonicated for 3 min (50 Hz), and kept for another 2 h. To the resulting suspension of liposomes, PBS was added to give a final volume of 50 ml and mixed gently. Then, the suspension was centrifuged at 2000 g for 5 min. The precipitated liposomes were finally suspended in 4 ml PBS. We have confirmed that within 24 h of a single i.v. injection of 0.2 ml of this suspension, a complete depletion of F4/80-positive cells (mature macrophages) occurs in the liver and splenic red pulp, although these cells are not affected significantly in the lung, as described previously [26
].
Measurement of cytokines
The levels of IL-18 and tumor necrosis factor
(TNF-
) in the sera were determined using a mouse IL-18 enzyme-linked immunosorbent assay (ELISA) kit (Medical and Biological Laboratories, Okayama, Japan) and a mouse TNF-
OptEIA ELISA kit (BD PharMingen, San Diego, CA), respectively. The concentrations of the cytokines were determined using the Softmax data analysis program (Molecular Devices, Menlo Park, CA).
Histological analysis
Immunohistochemistry was conducted as follows: Tissues were fixed in periodate-lysine 4% paraformaldehyde for 6 h at 4°C. After washing in PBS containing sucrose, fixed tissues were embedded in optical cutting temperature compound (Sakura, Tokyo, Japan) and immediately frozen. Frozen tissue sections (6 µm) were subsequently stained with hematoxylin and eosin (H&E). For immunostaining, the sections were incubated with anti-Gr-1 mAb overnight at 4°C. After that, sections were treated with peroxidase-blocking reagent (Dako Cytomation, Kyoto, Japan) for 20 min and secondary antibodies such as the goat anti-rat simple stain mouse MAX-PO (Nichirei, Tokyo, Japan). The chromogen used was 3', 3-diaminobenzidine tetrahydrochloride (Dako Cytomation). The sections were counterstained with hematoxylin. As a negative control, rat isotype-matched, control immunoglobulin G2b (BD Biosciences, San Jose, CA) was used.
For histopathological analysis, formalin-fixed liver samples were embedded in paraffin and stained with H&E to assess liver inflammation and necrosis.
Determination of liver enzymes
Blood was collected directly from the neck into test tubes after decapitation, and serum asparate aminotransferase (AST) and alanine aminotransferase (ALT) activities were measured photometrically using commercial kits (Wako Pure Chemical Industries, Osaka, Japan).
Statistical analysis
Experimental values were expressed as mean ± SD, and the statistical significance of differences between two means was evaluated by one-way ANOVA using the Bonferroni or Dunnett method, for which values of P < 0.05 were considered to be statistically significant.
| RESULTS |
|---|
|
|
|---|
|
|
Prevention of liver injury by anti-Gr-1 mAb treatment in mice
As treatment with P. acnes and LPS induces IL-18-dependent, acute liver injury in mice through the induction of hepatotoxic factors such as TNF-
and Fas ligand [18
19
20
], we next examined whether neutrophils are involved in the induction of serum TNF-
and liver injury. Treatment with P. acnes and LPS resulted in a marked increase in serum TNF-
levels, and the induction of serum TNF-
was almost completely prevented by the anti-Gr-1 mAb treatment (Fig. 3A
). Levels of the serum liver enzymes AST and ALT were also increased markedly by treatment with P. acnes and LPS, and their levels were decreased significantly to control levels by treatment with anti-Gr-1 mAb (Fig. 3B)
. Consistent with this, histological analysis showed that treatment with P. acnes and LPS induced severe liver injury (necrotic change and granuloma formation; Fig. 4
, left panels), and concomitantly, a marked infiltration of Gr-1+ cells was observed in the liver (Fig. 4
, right panels). P. acnes alone induced granuloma formation and Gr-1+ cell infiltration in the liver compared with untreated liver. When P. acnes-primed mice were treated with anti-Gr-1 mAb, no obvious liver injury or cell infiltration was observed without LPS challenge or even after the LPS challenge. These results indicate that Gr-1+ neutrophils are critically involved in IL-18-dependent liver injury in P. acnes-primed mice.
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
-inducing factor in the serum and liver of mice, which had been sequentially administered P. acnes and LPS [1
]. Neutrophils play a role in acute inflammation, and activated neutrophils release serine proteases [14
]. We previously showed that human epithelial cells constitutively express a precursor form of IL-18 and that costimulation with PR3 and LPS induces the secretion of bioactive IL-18 from these epithelial cells after priming with IFN-
[15
]. We then demonstrated that neutrophil serine proteases (PR3, HLE, and Cat G) activate cells through the PAR2 pathway [16
, 17
]. These studies raised the question of whether the in vitro findings really occur in vivo, and the present study showed that neutrophil recruitment and activation of PAR2 are required for the induction of serum IL-18 and IL-18-dependent liver injury in mice treated with P. acnes and LPS. Anti-Gr-1 mAb reacts with a common epitope on Ly-6C and Ly-6G and is generally considered to be a marker for neutrophils [29 ]. This mAb was administered to mice to deplete neutrophils in vivo as a useful technique in numerous previous studies. However, recent studies revealed that Gr-1 is expressed on a subset of monocyte/macrophage lineages as well [27 , 28 ]. Gr-1+ CD11b+ monocytes are a short-lived, inflammatory subset, which homes to inflamed tissue, where it can trigger immune responses [27 ]. Gr-1+ CD11b+ macrophages produce IL-12 p40 and generate reactive nitrogen intermediates during acute infection [28 ]. To differentiate between the contributions of neutrophils and phagocytic macrophages to the IL-18 induction in mice, clodronate-liposomes were used in this study. Van Rooijen and co-workers [23 24 25 ] showed that liposomes have a high selectivity for macrophages in vivo and when encapsulating clodronate, selectively eliminate phagocytic macrophages but not neutrophils. The present study showed that administration of anti-Gr-1 mAb significantly prevented the induction of serum IL-18 in the mice treated with P. acnes and LPS but that clodronate-liposomes had no obvious effect on IL-18 (Fig. 2) . This finding suggests that Gr-1+ neutrophils but not Gr-1+ macrophages are critically involved in the induction of IL-18 in these mice.
FUT-175 is a synthetic, low molecular-weight inhibitor of trypsin-like serine proteases such as trypsin, plasmin, pancreatic and plasma kallikrein, and thrombin with inhibitory concentration 50% (IC50) values ranging from 2.7 x 108 M to 5.0 x 107 M [30
, 31
] and is clinically used for the therapy of disseminated, intravascular coagulation and acute pancreatitis. FUT-175 also inhibits complement activation [32
] and suppresses nuclear factor (NF)-
B activation and niric oxide overproduction in LPS-treated macrophages [33
]. However, FUT-175 showed weak inhibitory activity against neutrophil serine proteases such as HLE with IC50 values greater than 103 M [34
], and we confirmed that the IC50 of FUT-175 against murine rPR3 was 5.0 x 103 M (data not shown). However, it is notable that administration of FUT-175 to pigs showed a reduction in neutrophil activation as well as reduced liver and gut injury on histology. Furthermore, this effect was achieved, despite starting the administration of FUT-175 after the shock period [35
], and a serine protease inhibitor, antithrombin III, inhibits LPS-mediated NF-
B activation by Toll-like receptor 4 [36
]. As FUT-175 has a short half-life in circulation and body fluids, FUT-175 was administered simultaneously with LPS in this study, and coadministration of FUT-175 with LPS into P. acnes-primed mice significantly decreased their serum IL-18 levels (Fig. 2A)
. A possible explanation of these results is that FUT-175 inhibits neutrophil activity, which in turn inhibits the production of neutrophil serine proteases or inhibits LPS-mediated signaling in addition to inhibiting protease activity in vivo. However, a PAR2 deficiency prevented the IL-18 induction in mice (Fig. 5A)
, and exogenous murine PR3, the PAR2 agonist, restored the protective effect of the anti-Gr-1 mAb (Fig. 6)
, although protease inhibitors exist in serum. Therefore, these results also point to the possibility that activation of PAR2 by serine proteases such as PR3 from Gr-1+ neutrophils is involved in the induction of IL-18 in vivo.
IL-18 is produced by immune cells such as activated macrophages and by nonimmune cells, including keratinocytes, epithelial cells, the bone-cartilage system, the endocrine system, the reproductive system, and the nerve system [2
3
4
, 15
], indicating that serum IL-18, detected in this study, is derived, not only from IL-18-expressing immune cells but also IL-18-expressing, nonimmune cells in vivo. IL-18 is a regulator of innate and acquired immune responses, but overproduction of IL-18 is harmful, as IL-18 is a powerful, proinflammatory cytokine with broad activity [2
3
4
]. As PAR2 is also expressed in a wide variety of tissues [6
7
8
], IL-18 and PAR2 are expressed together in many tissues. In fact, overexpression of IL-18 in the skin exacerbates and prolongs allergic and nonallergic, inflammatory, cutaneous reactions in mice [37
], and PAR2 deficiency diminishes allergic dermatitis [12
]. IL-18 enhances collagen-induced arthritis by recruiting neutrophils via TNF-
and leukotriene B4 [38
], and PAR2 deficiency protects against a form of adjuvant-induced arthritis [13
]. In the cytokine cascade, IL-18, in synergy with IL-12, induces IFN-
, and IL-18 induces, directly or indirectly via IFN-
, the production of effector cytokines such as TNF-
[2
3
4
, 39
, 40
]. The present study showed that a PAR2 deficiency impairs LPS-induced serum IL-18 in P. acnes-primed mice (Fig. 5B) and that anti-Gr-1 mAb treatment markedly reduces LPS-induced serum IL-18 and TNF-
(Figs. 2
and 3A)
. Therefore, this study suggests a novel mechanism by which the activation of PAR2 by serine proteases from activated neutrophils leads to the production of IL-18 and by which the induced IL-18 regulates the cytokine cascade, resulting in modulation of various diseases in vivo.
Treatment with P. acnes and LPS also induces IL-18-dependent liver injury in mice [18
19
20
]. IL-18 induction and the associated liver injury consist of two phases: a priming phase induced by P. acnes treatment and a subsequent effector phase induced by LPS challenge [41
]. During the priming phase, P. acnes treatment increases sensitivity to a subsequent challenge of LPS in mice, and IFN-
and IL-12 are required for priming [41
]. Three administrations of anti-Gr-1 mAb before LPS challenge (1 µg/mouse) almost completely abrogated LPS-induced serum IL-18 levels (Fig. 2B)
and concomitantly protected mice from LPS-induced liver injury (Fig. 4)
. Furthermore, hepatic tissues in PAR2/ mice showed only slight pathological changes after P. acnes priming alone and after sequential treatment with P. acnes and LPS (1 µg/mouse) as compared with wild-type mice (Fig. 5) . These observations suggest that neutrophil recruitment and PAR2 activation critically contribute to the priming phase.
To test the role of IL-18 in mice, in this study and other studies, P. acnes is always used for priming [41
]. The P. acnes priming raises the question of whether serum IL-18 is induced without P. acnes. We examined this point using the ovalbumin (OVA)-induced allergic inflammation model in mice. OVA (50 µg) and alum (3 mg) were injected i.p. on days 0 and 10 in BALB/c mice, and then OVA (50 µg) was used i.v. challenge on day 20. IL-18 levels in sera, obtained 2 h after the antigen challenge, were elevated significantly (
400 pg/ml), and this elevation was impaired markedly by anti-Gr-1 treatments 1 day before and 9 and 19 days after the initiation of the experiment (data not shown). The observation indicates that neutrophil recruitment is important for the induction of serum IL-18 in allergic inflammation, although further study is required to clarify the involvement of PAR2 in this system. It is also reported that serum IL-18 levels were elevated markedly in mice with acute graft-versus-host disease [42
].
Priming of mice with the Actinomyces species is also effective in inducting serum IL-18 after challenge with LPS, although the magnitude of the priming effect is lower than that with P. acnes (data not shown). Actinomyces species are dominant, commensal oral bacteria, which colonize the dental and mucosal surfaces of various animal hosts and have been implicated in caries [43 ], periodontitis [44 ], and root canal infection [45 ] as well as actinomycosis and septic infection [46 ]. This indicates that infection with oral bacteria in periodontal tissues augments the sensitivity to LPS of periodontopathic Gram-negative bacteria and serine proteases from infiltrating neutrophils, resulting in induction of IL-18 from oral epithelial cells through the PAR2 pathway, as described previously [15 ]. An arginine-specific cysteine protease from periodontopathic Porphyromonas gingivalis was reported to activate PAR2 on human oral epithelial cells [47 ]. IL-18 also exerts a protective role in host defense against microbial infection [2 3 4 ], and therefore, it is conceivable that IL-18 plays a role in the clearance of bacteria. Other Gram-positive bacteria may exhibit the same priming effect as P. acnes and the Actinomyces species.
Neutrophils express IL-18 receptor, and IL-18 induces cytokine, chemokine, and granule release from neutrophils and enhances the respiratory burst following exposure to formyl-methionyl-leucyl-phenylalanine [48 ]. Therefore, it is also possible that induced IL-18 in turn activates neutrophils, thereby promoting inflammatory responses.
In conclusion, the present study indicates that neutrophil recruitment and PAR2 activation by neutrophil serine proteases are critically involved in the induction of IL-18 and IL-18-dependent liver injury in P. acnes-primed mice. The observations obtained from the mouse model in this study may also apply to general inflammation. IL-18 could be induced from IL-18-expressing immune and nonimmune cells through PAR2 activation at the inflamed site and may play a role in the regulation of inflammation through the development of Th1 or Th2 responses.
| ACKNOWLEDGEMENTS |
|---|
Received March 15, 2005; revised July 6, 2005; accepted July 20, 2005.
| REFERENCES |
|---|
|
|
|---|
production by T cells Nature 378,88-91[CrossRef][Medline]
- and Fas ligand-mediated hepatotoxic pathways in endotoxin-induced liver injury in mice J. Immunol. 159,3961-3967[Abstract]

+ and TCR
CD2+CD3 double negative lymphocytes by bacterial stimulation Int. Immunol. 2,775-785
B activation and NO overproduction in LPS-treated macrophages Int. Immunopharmacol. 3,1335-1344[CrossRef][Medline]
B activation by TLR-4 FEBS Lett. 508,313-317[CrossRef][Medline]
and leukotriene B4 J. Immunol. 171,1009-1015
-deficient mice by a concomitant reduction of TNF-
, IL-12, and IL-18 production J. Immunol. 162,1049-1055
-inducing factor) induces IL-8 and IL-1ß via TNF
production from non-CD14+ human blood mononuclear cells J. Clin. Invest. 101,711-721[Medline]
This article has been cited by other articles:
![]() |
T. Nishioka, T. Kuroishi, Y. Sugawara, Z. Yu, T. Sasano, Y. Endo, and S. Sugawara Induction of serum IL-18 with Propionibacterium acnes and lipopolysaccharide in phagocytic macrophage-inactivated mice J. Leukoc. Biol., August 1, 2007; 82(2): 327 - 334. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |