Published online before print March 19, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Dermatology and Ludwig Boltzmann Institute for Cell Biology of the Skin, University of Münster, Münster, Germany; and
Inflammation Research Network, Department of Pharmacology and Therapeutics, University of Calgary, Calgary, Alberta, Canada
1Correspondence: Dept. of Dermatology, Laboratory of Cell Biology, University of Münster von-Esmarch-Str. 58, 48149 Münster, Germany. E-mail: shpacovi{at}ukmuenster.de
 TOP ABSTRACT INTRODUCTIONPARs AND THE FUNCTION...PAR-MEDIATED EFFECTS ON...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
Key Words: proteolytic enzymes • leukocytes • allergy • infectious diseases
|
TOPABSTRACTINTRODUCTIONPARs AND THE FUNCTION...PAR-MEDIATED EFFECTS ON...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
The fact that 
1200 genes (4.5% of all human genes) encode proteases in the human genome reflects the importance of proteases in the human body under physiological and pathophysiological conditions. However, the role of protease signaling in innate and adaptive immunity is just at the beginning of intensive investigation. Although, there are a number of ways that proteases can trigger cell signaling, a recently described family of protease-activated receptors (PARs) can account for a significant proportion of signaling generated by proteolytic enzymes (reviewed in refs. [5
, 6
]). This review will focus on PARs and their importance for the function of immune cells in vitro as well as for immune responses in vivo.
The unique mechanism whereby serine proteases signal via the PARs involves the cleavage of the receptor N-terminal part to expose a new, previously cryptic sequence. The exposed sequence remains tethered to the receptor and acts further as a receptor-activating ligand, named "tethered ligand" (reviewed in refs. [5 , 6 ]). Some proteases could cleave PARs downstream of the tethered ligand sequence, making further proteolytic activation of PARs impossible (receptor inactivation; see Table 1 ). Thus, proteases can regulate PAR signaling by activation or inactivation. Four PARs have been cloned and characterized [5 , 6 ]. PARs 1, 3, and 4 were the first identified targets for thrombin but can also be activated by trypsin or cathepsin-G (CG). In contrast, PAR2 is resistant to thrombin but can be activated by trypsin, mast cell tryptase, leukocyte proteinase-3 (PR3), and bacteria-derived enzymes (see Table 1 ) [5 6 7 ]. PARs also can be activated without the need for proteolysis by synthetic peptides (so-called PAR-APs). PAR-APs have sequences based on those of the revealed tethered ligands [5 , 6 ]. Specific PAR-APs are important probes for investigating the role of PAR activation, as serine proteases are known to cause PAR-dependent as well as independent responses in various cells. Although PARs 1, 2, and 4 signaling could be triggered upon activation by proteases or receptor-specific PAR-APs, the role of PAR3, which cannot signal on its own, remains a bit of an enigma. Currently, PAR3 is viewed as an accessory receptor for PAR1 or PAR4 [8 9 10 ].
|
View this table: [in a new window] |
Table 1. PARs and the Proteases Acting via PARs
|
|
TOPABSTRACTINTRODUCTIONPARs AND THE FUNCTION...PAR-MEDIATED EFFECTS ON...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
 View larger version (30K): [in a new window] |
Figure 1. Scheme of the activation of PAR1 and PAR1-triggered responses in human immune cells. (A) The inactive status of PAR1 is associated with a hidden N-terminal-tethered ligand sequence. The tethered ligand part cannot interact with the second extracellular loop of the receptor and does not activate it. (B) The accessible serine protease cuts the N-terminal part of the receptor and unmasks a tethered ligand sequence of the receptor. Downstream cleavage of the receptor results in the receptor inactivation. Further, the tethered ligand interacts with the second extracellular loop of the same receptor and thus, triggers signaling events. Synthetic activating peptides (PAR1-APs) activate PAR1 without proteolytic cleavage. PAR1-AP directly interacts with the second extracellular loop of the receptor, inducing identical signaling pathways as the natural protease. (C) PAR1 activation on human immune cells leads to various responses mentioned on the scheme. ZAP-70,  -associated protein 70; SLP-76, Src homology 2 domain-containing lymphocytic protein of 76 kDa; MIF-1 , macrophage migration inhibitory factor-1.
|
 View larger version (33K): [in a new window] |
Figure 2. Scheme of the activation of PAR2 and PAR2-induced effects in human immune cells. (A) Inactive PAR2 has an N-terminal-tethered ligand sequence, which remains cryptic. This tethered ligand does not interact with the second extracellular loop of the receptor and does not trigger downstream signaling events. (B) Endogenous or exogenouse serine protease could cut the N-terminal part of the receptor, unmasking a previously cryptic tethered ligand sequence of the receptor. Further, the tethered ligand interacts with the second extracellular loop of the same receptor triggering signaling events. Synthetic activating peptides (PAR2-APs) activate PAR2 without proteolytic cleavage. PAR2-AP directly interacts with the second extracellular loop of the receptor, inducing identical signaling pathways as the natural protease. (C) PAR2 activation on human immune cells results in various immune cell responses. Mac-1, Membrane-activated complex 1; VLA-4, very late activation Ag-4.
|
The expression of functional PAR2 by human neutrophils was first described by Howells and colleagues [11 ], who showed that the PAR2-AP, SLIGKV, increased [Ca2+]I levels and induced neutrophil shape changes. The authors also demonstrated that costimulation of isolated human neutrophils with the PAR2-AP, along with the well-known neutrophil activator fMLP, leads to a more pronounceable up-regulation of Mac-1 (CD11b/CD18) expression than does fMLP alone. In further studies, Shpacovitch and colleagues [13 ] demonstrated that stimulation of human neutrophils with PAR2 agonists (tc-LIGRLO-NH2 and trypsin) led to a significantly enhanced motility of these cells in three-dimensional (3-D) collagen lattices, suggesting a role of PAR2 in the regulation of neutrophil migration through the extracellular matrix (ECM) toward the site of inflammation and/or infection. Moreover, PAR2 agonists enhance the shedding of L-selectin from the cell surface of human neutrophils. The effect of PAR2 agonists at neutrophil motility in 3-D collagen lattices was significantly abolished by the L-selectin shedding inhibitor KD-IX-73-4 [13 ]. This finding indicates a link between PAR2 agonist-induced shedding of L-selectin and the enhanced motility of neutrophils through the ECM. Further, it was reported that simultaneous stimulation of human endothelial cells and neutrophils with proteases (trypsin, tryptase) or PAR2-trans-cinnamoyl-LIGRLO-NH2 reduces transendothelial migration of granulocytes and also prolongs neutrophil survival in vitro [17 ]. Additionally, PAR2 agonists have been reported to enhance the expression of Mac-1 and VLA-4 on neutrophils [13 ] and stimulate the release of IL-1β, IL-8, and IL-6 from these cells [13 ]. Recently, tryptase, trypsin, and PAR2-APs (SLIGKV-NH2 or tc-LIGRLO-NH2) were demonstrated to induce lactoferrin secretion by human neutrophils [18 ].
It is also important to note that neutrophil granule serine proteases such as human leukocyte elastase, CG, and PR3 are able to activate or disarm PARs (reviewed in ref. [7 ]).
In contrast to human cells, the expression of PAR4 has been detected immunohistochemically on the cell surface of rat neutrophils [19 ]. In further studies, it was revealed that local injection of PAR4-AP (AYPGKF-NH2) induced neutrophil recruitment and edema formation in a rodent paw inflammation model [20 ]. Additionally, application of a PAR4 antagonist (P4pal-10) attenuated neutrophilic, inflammatory responses in a mouse model of disseminated intravascular coagulation [21 ].
The expression pattern of PARs on neutrophils is different in rodents (mice, rats) and in humans. PAR4 activation appears to affect the migratory ability of rodent neutrophils. However, PAR2 seems to be functionally the most important member of the PAR family expressed by human granulocytes. These facts limit the use of mouse models for the investigation of the PAR-mediated responses on neutrophil functions. Despite well-documented effects of PAR2 agonists at cytokine production, motility, and adhesion molecule expression by human neutrophils, effects of PAR2 activation at main functions of human neutrophils—phagocytosis and killing of bacteria—remain unknown. Moreover, further investigation of the PAR-associated effects of neutrophil granule proteases at functions of epithelial, endothelial, and immune cells also looks prospective. Such studies might reveal new approaches that allow direct interfering of the inflammatory and immune responses.
Eosinophils and basophils
Human basophils do not possess any member of PAR’s family [22
]. There is no report indicating that rodent basophils possess any PARs. However, human eosinophils reportedly express PAR2 mRNA and protein [14
, 15
]. Receptor expression has been detected on the surface of eosinophils as well as intracellularly [14
, 15
]. Moreover, the number of PAR2-positive eosinophils increases during seasonal allergic rhinitis in the nasal mucosa [23
]. The ability of human eosinophils to express functional PAR1 remains unclear. Miike and colleagues [14
] failed to observe PAR1 mRNA expression and functional responses to PAR1 agonists in isolated human eosinophils. Nevertheless, Bolton and colleagues [15] found that PAR1 is expressed in isolated human eosinophils, but this group did not find significant functional responses in eosinophils after PAR1 agonist stimulation. However, more recent studies have been able to demonstrate effects of thrombin and a specific PAR1-AP on eosinophil migration [24
]. The different outcomes for these studies could be explained by the variations in experimental procedures and a possible pre-activation of eosinophils during preparation followed by receptor internalization or inactivation. Thus, the presence and functionality of PAR1 on human eosinophils merit a further in-depth evaluation.
In contrast with the information summarized above, PAR2, despite its low and variable eosinophil surface expression amongst donors, seems to be the predominant functional member of the PAR family on these cells in humans. Indeed, trypsin and a PAR2-AP (SLIGKV-NH2) can induce superoxide anion production and degranulation of human eosinophils via PAR2 [14 , 15 ]. This effect of trypsin requires its proteolytic activity, supporting the idea that the responses are associated with PAR2 activation on human eosinophils [14 ].
The data concerning the influence of PAR2 deficiency on OVA-induced airway eosinophilia in mice remain contradictive. Takizawa and colleagues [25 ] reported that the number of eosinophils is reduced significantly in bronchoalveolar lavage fluid (BALF) of PAR2 knockout mice as compared with wild-type animals after OVA challenge. In contrast, De Campo and Henry [26] demonstrated that intranasal administration of PAR2-AP at the time of OVA challenge also causes a reduction in the numbers of BALF eosinophils.
Findings concerning the role of PAR2 in eosinophil activation are important, as serine and cysteine proteases produced by allergenic organisms such as fungi, mites, and pollens might act in part by triggering PAR2 signaling. Additionally, during the allergic response and subsequent mast cell degranulation, the released tryptase and chymase may also stimulate PAR2 on human eosinophils. Several mite-derived proteases (Der p 3, Der p 9, and Der p 1) are capable of regulating PARs, including PAR2 [27 , 28 ]. However, whether mite-, fungus-, or pollen-derived proteinases activate PAR2 expressed on eosinophils remains unknown. Moreover, mite-derived proteases are reported to mediate their effects also via a PAR2-independent mechanism [29 ]. The biological and pathophysiological impact of these activation mechanisms thus need to be clarified in the future.
Mast cell tryptase is thought to induce some of its effects via activation of PAR2 [30 32 33 ]. However, the ability of this enzyme to activate PAR2 expressed on eosinophils needs further investigation. Temkin and colleagues [34] demonstrated that stimulation of isolated human eosinophils with human recombinant as well as with a human mast cell line 1 (HMC-1)-derived tryptase results in enhanced release of IL-6 and IL-8 by eosinophils [34 ]. These tryptase-induced effects on cytokine release by human eosinophils were reduced by adding an anti-PAR2 antibody that can block receptor cleavage/activation. On the basis of these findings, the authors suggested that the effects of tryptase on eosinophil functions may be at least partially mediated via PAR2 activation [34 ]. However, Vliagoftis and colleagues [35 ] showed that induction by mast-cell tryptase of eosinophil peroxidase (EPO) and β-hexosaminidase (β-hex) release from human eosinophils does not depend on PAR2 activation. The authors based their conclusion on the inability of the PAR2-AP (SLIGRL-NH2) to induce the release of EPO and β-hex from eosinophils. Thus, some tryptase effects on eosinophils may be PAR-independent.
In summary, the ability of allergen-derived proteases (house dust mite, cockroach, and others) to mediate their effects on functions of human eosinophils via PARs remains enigmatic and needs further investigation. New findings in this field might be useful, not only for a better understanding of the way of protease signaling to eosinophils but also could serve for designing novel, anti-allergic treatment strategies in humans.
Monocytes, macrophages, and dendritic cells (DC)
Monocytes
According to the work of Colognato and colleagues [36
], isolated human monocytes express PAR1, PAR2, and PAR3 mRNA and at protein levels PAR1 and PAR3. A receptor-selective PAR1-AP as well as thrombin were demonstrated to stimulate an increase of [Ca2+]I levels in human monocytes. Any role for PAR3 in monocytes remains uncertain, as this receptor does not signal on its own. That said, evidence for the ability of undifferentiated human monocytes to express functional PAR2 appears to be conflicting. Colognato and colleagues [36
] reported that human monocytes do not express PAR2 on the cell surface, whereas a report by Johansson et al. [37
] demonstrated that human monocytes display functional PAR2. Colognato’s group [36
] used monocytes separated mainly from buffy coats by a surface adherence protocol. They used an affinity bead (anti-CD14 MACS microbeads; positive selection method) protocol only for monocyte isolation in RT-PCR experiments. Johansson and colleagues [37
] used human monocytes isolated from healthy adult volunteers via blood collection in a heparinized tube, followed by negative magnetic cell sorting (MACS method). Isolation of human monocytes by adhesion might affect the monocyte phenotype [38
], and thus, the differences between the apparently conflicting results related to the presence of PAR2 might be a result of differences in the methods of cell isolation [37
]. Moreover, the intracellular stores of PAR2 were found in human monocytes. These stores could be mobilized rapidly to the monocyte cell surface [37
]. Thus, the ability of human monocytes to express functional PAR2 may also depend on the state of cell activation.
Stimulation of isolated monocytes with thrombin as well as with a receptor-selective PAR1-AP (TFLLRNPNDK) leads to the up-regulation of MCP-1 expression and its release from human monocytes. This effect of thrombin was attenuated by the PAR1-blocking antibody WEDE15, which inhibits proteolytic activation of PAR1 [36 ]. The anti-apoptotic effect of plasminogen treatment demonstrated on isolated human monocytes and U937 cells (a human monocyte cell line) appears to depend on PAR1 activation, as anti-PAR1-blocking antibodies diminished this effect [39 ]. Thrombin as well as a PAR1-AP can modulate the production and release by monocytes of IL-6, platelet basic protein, and platelet factor 4 and can also stimulate the proliferation of U937 cells [40 41 42 43 44 ]. Additionally, treatment of human monocytes with thrombin or the nonselective PAR1/PAR2-AP (SFLLRNPNDKYEPF; named SFLLRN-14) enhances the ability of these cells to kill bacteria such as Salmonella enterica serovar typhimurium or Listeria monocytogenes [45 ]. Unfortunately, in all of these studies [40 41 42 43 44 45 ] SFLLRN-14, SFLLR-NH2, or SFFLR-NH2 was used as a PAR agonist; however, they can activate PARs 1 and 2 [46 , 47 ]. Thus, although PAR1 would appear to be involved, based on the results with thrombin, which does not readily activate PAR2, a cooperative role for PAR2 along with PAR1 in the monocytes cannot yet be ruled out. Further work with the PAR1- and PAR2-selective agonists TFLLR-NH2 and SLIGKV-NH2 should be able to resolve this issue.
Monocyte-derived and residential macrophages
Human monocyte-derived macrophages express PAR1, PAR2, and PAR3 at the mRNA and protein levels [36
]. Stimulation of monocyte-derived macrophages with PAR1- or PAR2-selective APs (TFLLRNPNDK or SLIGKV-NH2, respectively) induces an increase in [Ca2+]I level, confirming that both receptors are functional [36
]. Prolonged treatment with IL-4 significantly reduces the expression of PAR1, PAR2, and PAR3 by monocyte-derived macrophages [36
]. PARs are reported to be expressed by different types of human tissue macrophages. Human alveolar macrophages express PAR1 and PAR2. The level of PAR1 protein expression has been found to be higher in smokers than in healthy nonsmokers, but the expression levels of both receptors in asthmatic patients were comparable with those in healthy humans [48
]. Immunohistochemical staining has demonstrated the expression of PAR2 on vascular macrophage-derived foam cells, indicating a potential role of this receptor in the development of atherosclerotic lesions [49
]. PAR4 expression could not be verified in monocyte-derived macrophages [36
]. However, studies in the human liver (normal, cirrhotic, or hepatocellular carcinomas) revealed PAR4 expression to be restricted to macrophage-like cells (Kupfer cells), B lymphocytes, and nerves [50
].
Stimulation of human monocyte-derived macrophages with thrombin or specific PAR1-AP (TFLLRNPNDK) results in up-regulation of the expression and release of MCP-1 by these cells. PAR2-AP failed to induce those effects in human monocyte-derived macrophages [36 ].
It is also interesting to notice, that murine brain macrophages (microglia cells) express mRNA for all PARs at different levels (the lowest for PAR4) [51 ]. Activation of mouse microglial PAR1 induces [Ca2+]I increase and transient activation of p38 as well as p44/42 MAPKs [52 ]. Moreover, PAR1 appears to contribute directly to thrombin-induced microglial proliferation in mice [52 ]. On the other hand, the proteolytic activity of thrombin was unlikely involved in the protease-mediated effects on chemokine and cytokine production in murine microglia cells [53 ]. When tested, the ability of thrombin to activate rat microglial cells, as assessed by NO release, was found not to be mimicked by a nonselective PAR1/PAR2-AP and therefore, appeared to be PAR1-independent [54 ]. However, a role for thrombin-activated PAR4 in that study was not explored.
DC
Human monocyte-derived DC, generated after stimulation of monocytes with GM-CSF and IL-4, express PAR1, PAR2, and PAR3 mRNA but do not express the PARs at the protein level [36
]. The lack of PAR2 expression was also reported for human blood DC derived from CD34-positive stem cells [37
]. However, thrombin is able to induce the release of MCP-1, IL-10, and IL-12 from plasmocytoid and myeloid DC isolated from human blood. This effect of enzyme might be mediated via PAR1, as PAR1 is expressed by plasmocytoid and myeloid DC [55
].
Activation of PAR2 with a receptor-selective AP (SLIGRL-NH2) can stimulate mouse DC development [56
]. Additionally, bone marrow progenitor cells from PAR2 knockout mice failed to generate DC under standard culture conditions (GM-CSF and IL-4 stimulation) but required additional stimulation by TNF- [56
]. Further, Csernok and colleagues [57
] demonstrated that PR3, a well-known "Wegener autoantigen", also induces differentiation of DC via a PAR2-dependent pathway.
DC isolated from the mouse spleen and bone marrow progenitor cells express the identical pattern of PARs [56 ]. The level of PAR1 and PAR2 expression significantly decreases in bone marrow progenitor cells during cell culture and also remains low in mature spleen DC [56 ]. The work of Colognato and colleagues [36 ] sheds light on the possible mechanism of such regulation in vitro, wherein prolonged treatment of cultured monocytes with IL-4 resulted in a significant reduction of PAR expression during DC generation.
Despite that the role of PARs in functional responses of brain macrophages (microglial cells) is well documented, the impact of PAR activation at functions of other macrophages, especially in disease models, is still poorly investigated in rodents. Human monocytes, macrophages, and DC possess at least one functional member of the PAR family, namely PAR1. However, the role of protease signaling mediated via PAR1 in main functions of these cells, such as antigen uptake and processing, phagocytosis, and the production of inflammatory mediators, needs further investigation.
Mast cells
Human Mast Cells (hMC) are well recognized as key players in the initiation of allergic diseases and also as participants in the host response to certain types of parasitic infections. However, little is known about the impact of PARs and their proteolytic activators on mast cell function. Mast cells located in human tonsils, skin, and colon express PAR2 [58
59
60
]. Trypsin as well as PAR2-AP (SLIGKV) stimulation results in a concentration-dependent release of histamine from hMC in the skin and tonsils [59
, 60
]. Additionally, skin hMC have been found to express mRNA for PAR1, PAR3, and PAR4 [60
]. HMCs from different tissues display PAR2- and PAR1-positive intracellular staining on tryptase-containing granules [61
]. Leukemia-derived HMC-1 possesses functional PAR2 and PAR4 and also expresses mRNA for PAR1 and PAR3 [60
, 62
]. Stimulation of HMC-1 with PAR2- or PAR4-APs enhanced TNF- secretion by these cells [62
].
Murine mast cells respond to PAR1 activation with enhanced IL-6 secretion [63
]. Moreover, thrombin and a receptor-selective PAR1-AP have been found to induce the adhesion of murine bone marrow-derived mast cells to fibronectin and laminin. This process was found to involve 4 and 5 integrins along with PI-3K, MAPK, and protein kinase-3 signaling pathways [64
]. Stimulation with IL-12 down-regulates the expression of PAR2 and enhances the expression of PAR4 on murine mast cell line P815 cells [65
]. Rat peritoneal mast cells (PMCs) express PAR1 and PAR2 mRNA [66
, 67
]. However, the functional role for these receptors on PMCs still remains unclear. On the one hand, stimulation with PAR1- or PAR2-AP of PMCs does not appear to trigger degranulation [66
]. On the other hand, the nonselective PAR1/PAR2-AP (SFFLRN) is capable of activating NO [68
] as well as β-hex and histamine release from PMCs [69
]. Further studies are, however, demanded to clarify the precise role of PARs in mast cell function and in mast cell-mediated hypersensitivity.
Mast cell tryptase has been proposed as an important regulator of PAR2 (reviewed in refs. [7 , 70 ]). This enzyme is reported to activate PAR2 on different cell types such as lung fibroblasts [71 ], smooth muscle cells [72 ], myenteric neurons [31 ], keratinocytes [32 , 73 ], endothelial cells [74 , 75 ], testicular peritubular cells [76 ], and primary HMCs [60 ]. However, the ability of mast cell tryptase to activate PAR2 in vivo under physiological and pathophysiological conditions needs further investigation. Finally, future studies should verify a potential role of PAR antagonists for the treatment of inflammatory, parasitic, and allergic diseases in which mast cells are involved.
B and T lymphocytes
B cells
Cell surface expression of PAR4 has been detected on B cells in the human liver [50
], but the role of this receptor in B cell function remains unknown. In mice, activation of PAR2 stimulates B lymphocyte adhesion [77
].
T cells
Human T cell lines and human T cells in tissues appear to express PAR1, PAR2, and PAR3 [78
79
80
81
]. Stimulation of peripheral blood lymphocytes with PHA and PMA enhances the level of PAR2 mRNA expression in these cells [78
]. Mouse T lymphocytes also express functional PAR2, although its biology remains to be clarified [82
].
Stimulation of human Jurkat T cells with thrombin, trypsin, or PAR1- or PAR2-APs has been found to elevate [Ca2+]I [78
]. Further, Bar-Shavit and colleagues [79
] found that the activation of PARs (PAR1, PAR2) in Jurkat T cells induces tyrosine phosphorylation of Vav-1 and also leads to tyrosine phosphorylation of ZAP-70 and SLP-76, which are known to play a crucial role in TCR signaling. Unfortunately, these investigators used a PAR1/PAR2 nonselective agonist (SFLLRNPNDK). Therefore, the observed effects might have been a result of a simultaneous activation of PAR1 and PAR2. Although a functional role for PAR3 on T cells was suggested in that study, further work revealed that the PAR3-AP (TFRGAPPNSF) does not activate PAR3 in Jurkat T cells but rather stimulates PAR1 or PAR2 [80
]. In human primary T cells, treatment with thrombin, trypsin, tryptase, and PAR-APs resulted in an increased IL-6 secretion [81
]. However, the analysis of the underlying signaling cascades and their role under physiological and pathophysiological conditions need further investigation. Recently, an important role for PAR2 signaling has been found for the cytokine production by mouse CD4+ T cells [82
]. A reduction of IL-4 production by splenic CD4+ T cells was found during OVA-induced airway inflammation in PAR2-deficient mice, as compared with wild-type animals. After antigen stimulation, IFN-
production was enhanced in PAR2 null animals as compared with wild-type mice. This PAR2-mediated regulation of T cell cytokine production appears to be associated with JNK1 phosphorylation [82
].
The role of PAR2 in cytokine production and other responses of human and murine T cells has been shown. However, there is a lack of detailed information concerning the involvement of PAR1 in the regulation of T cell function and the role of PARs in T cell-mediated diseases.
|
TOPABSTRACTINTRODUCTIONPARs AND THE FUNCTION...PAR-MEDIATED EFFECTS ON...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
The effect of intranasal administration of proteases and PAR-APs on leukocyte recruitment into the airways has been analyzed in a murine model. Thrombin-induced recruitment of polymorphonuclear cells appeared to be associated with PAR4 activation, as the inhalation of a nonselective PAR1/PAR2-AP failed to reproduce thrombin-induced effects, and thrombin itself did not effectively activate PAR2 [86 ]. Additionally, a thrombin-induced macrophage adhesion in vivo has been demonstrated to be PAR1-independent [87 ]. In summary, PAR2as well as PAR4can be considered as the main members of the PAR family affecting leukocyte motility in vivo.
Role of PARs in inflammatory disease models
A significant role of PAR activation in the progression of inflammatory diseases has been demonstrated in studies that used different animal models, including transgenic and knockout mice [88
89
90
91
92
93
]. However, only a few of these studies analyzed the involvement of PAR agonist-stimulated immune cells in disease progression. Generally, it is important to notice that the disruption of PAR1 and PAR2 genes might benefit or worsen the prognosis for the host depending on the type and stage of the inflammatory disease [91
, 94
95
96
]. The effect of PAR4 deficiency on the development of inflammatory diseases is still poorly investigated. Additionally, the downstream signaling cascades triggered after PAR activation during inflammation in vivo are just at the beginning of investigation [93
]. The effects of PAR deficiency in animal models of different inflammatory pathologies are presented in Table 2
. The following review section provides detailed information concerning important types of inflammatory animal models, which were investigated.
|
View this table: [in a new window] |
Table 2. The Crucial Effects of PAR Deficiency or PAR Agonist Stimulation in Animal Models of Inflammation
|
Pawlinski and colleagues [110 ] demonstrated that the deficiency of PAR1 or PAR2 alone did not affect mouse survival in a model of endotoxemia. The lack of any significant effect of the PAR2 deficiency at the pathophysiological changes induced by LPS in the mouse was also confirmed in the work of Kazerani and coauthors [111 ]. However, an application of hirudin (a thrombin inhibitor) in PAR2 knockout mice protected these animals better than PAR2 wild-type mice against LPS-induced lethality [110 ]. This fact allowed the authors to suggest that the combination of PAR2 deficiency and inhibition of thrombin might be beneficial for mouse survival during LPS-induced endotoxemia.
This hypothesis was checked later in the study of Camerer and colleagues [112 ] about PAR single-deficient mice (PAR1, PAR2, and PAR4 knockouts) and on double-knockouts such as PAR1–/–:PAR2–/– and PAR2–/–:PAR4–/–. These authors also used i.p. injection of LPS to induce endotoxemia. The results of the study clearly indicated that neither single PAR deficiency nor combined PAR deficiency (such as PAR1–/–:PAR2–/– or PAR2–/–:PAR4–/–) played any significant role in mouse survival during LPS-induced endotoxemia [112 ]. Altogether, the results from a mouse model of endotoxemia do not support any role for PAR-mediated protease signaling in murine LPS-induced lethality.
However, it is still important to notice that a mouse model of LPS-induced endotoxemia might just hardly mimic the changes observed during human sepsis [113 ]. Thus, it cannot be considered as a really adequate model of human sepsis but rather as a model of endotoxic shock. Moreover, the circulating LPS levels in septic human patients are also reported to be low [113 , 114 ]. According to several criteria, the cecal ligation and puncture model seems to be a more appropriate model, reflecting more precisely the dynamics of sepsis occurring in humans [113 ]. The role of PAR deficiency for pathophysiological changes, which occur in this model, is very interesting for investigation.
PARs and inflammatory gastrointestinal diseases
PAR activation appears to play a significant role in gastrointestinal inflammation. Inflammatory bowel diseases (IBD) such as Crohn’s disease, ulcerative colitis, and infectious colitis are frequent and potentially life-threatening inflammatory diseases. In the normal gut, PARs are constitutively expressed in the colon and small intestine by epithelial cells, colon mast cells, nerves, and smooth muscle cells [58
, 115
116
117
118
119
]. A role of PARs in IBD was confirmed in several in vitro and in vivo studies [120
121
122
], some of which showed the involvement of particular immune cell types [91
, 94
].
PAR2 activation reportedly played a protective role in a mouse model of TNBS-induced colitis (intrarectal administration of TNBS) [91
]. T cell-derived cytokines (IL-2, IL-12, IFN-, TNF-) were shown to be involved in the development of TNBS-induced colitis [123
124
125
], which suggests that anti-inflammatory effects of PAR2 activation may be associated with the modulation of T lymphocyte function in this tissue. Moreover, PAR2-AP prevented the induction of Th1 and Th2 proinflammatory cytokines in the colon that was caused by TNBS injection. PAR2 agonist stimulation in vivo affected T lymphocyte proliferation in the lamina propria after exposure to anti-CD3/CD28 [91
]. These findings indicate a protective role of PAR2 activation in a mouse model of a chronic, chemically induced colitis. However, PAR2 activation seems to contribute in the development of enteritis caused by toxin A from C. difficile, as mice lacking PAR2 showed a reduced intestinal, inflammatory response to toxin A [99
].
Remarkably, PAR1 appears to play anti-inflammatory [95 ] as well as proinflammatory [94 ] roles in different models of inflammatory bowel diseases. In patients with IBD, PAR1 is overexpressed in the colon. T and B lymphocytes are involved in the development of PAR1-induced bowel inflammation [94 ]. This fact was confirmed by the failure of PAR1-AP to induce colonic inflammation in SCID–/– and RAG1–/– mice [94 ]. These types of mice are known to lack functional and mature T and B lymphocytes [126 ]. Moreover, PAR1 activation exacerbated and prolonged TNBS-induced colonic inflammation, whereas PAR1 antagonism decreased it [94 ]. However, an anti-inflammatory role of PAR1 has been described in Th2-mediated colitis [95 ]. It is also interesting to notice that the microspheres with an encapsulated PAR1-AP accelerated the healing process of experimental ulcers in rats [127 ].
These studies clearly demonstrate an important involvement of the adaptive immune system in the PAR-mediated control of colitis. Remarkably, the observed anti-inflammatory or proinflammatory responses stimulated by PAR family members during colonic inflammation appear to be associated with specific effects of these receptors on lymphocyte function or dysfunction, respectively.
The role of PARs in inflammatory processes in the CNS
All four PARs are expressed in the CNS (reviewed in refs. [128
, 129
]), and their activating proteases could be produced within the brain (trypsinogen IV, tryptase) or infiltrate the brain via the "leaky" blood-brain barrier under inflammatory conditions.
The effects of PAR1 activation on specific cells in the brain have been summarized recently in a review [130 ]. However, it is interesting to mention the facts concerning a potential role of PAR1 activation during CNS inflammation. Encephalitis is a type of brain inflammation caused by viruses or other microbial pathogens. This pathology is also known to be induced by HIV-1. mRNA and protein levels of PAR1 significantly increase in astrocytes during HIV-induced encephalitis [131 ]. Moreover, implication of PAR1-APs in the mouse striatum leads to the activation of astrocytes and microglia cells, indicating an enhanced local inflammatory response [131 ]. These observations allowed the authors to conclude that activation of astrocyte PAR1 during HIV-1 infection might contribute to brain inflammation and progress of the infection.
Recent studies performed on animals and also with sections of human brain tissues demonstrated a crucial role of PAR2 activation during brain inflammation and progression in neurodegenerative diseases. Remarkably, it seems that enhanced PAR2 expression on neurons of the CNS is neuroprotective in animal models of brain ischemia and inflammatory HIV-associated dementia. On the other hand, increased PAR2 expression on astrocytes and microglia appears to be associated with neurodegenerative processes in a mouse model of experimentally induced autoimmune encephalomyelitis (EAE) and multiple sclerosis (MS; reviewed in ref. [129 ]). Striatal implantation of PAR2-APs significantly inhibited the neurotoxicity induced by the HIV-1-transactivating protein Tat in PAR2 wild-type mice [101 ]. Moreover, the severity of neuroinflammation and neuronal damage was higher in PAR2 knockout animals after Tat implication [101 ]. On the other hand, enhanced PAR2 expression on astrocytes and infiltrating macrophages (no changes on neurons were indicated) in EAE and MS mouse models contributes to demyelination and thus, to neurodegenerative processes. In such demyelination, T cells appear to play an important role, and the number of T cells in the spinal cord of PAR2 wild-type animals was significantly enhanced as compared with knockout mice during EAE [100 ]. Thus, a role of PAR2 activation in the process of neuroinflammation in the CNS appears to be dual depending on the disease model and the modification of the expression patterns in the various cells (neuron, astrocyte, glia).
The role of PARs in joint inflammation
A key role of PAR2 in the development of chronic arthritis was demonstrated recently in vivo. In the adjuvant monoarthritis model, joint swelling was inhibited in PAR2 knockout mice, and the injection of PAR2 synthetic agonists (2-fuoryl-LIGKV-OH or SLIGRL-NH2) resulted in joint swelling and hyperemia [88
]. It is also important to notice that serine protease-induced joint swelling appears to be mediated via PAR2 activation, as this response to exogenous application of trypsin and tryptase was absent in PAR2 knockout mice [103
]. Also, the inhibition of PAR2 up-regulation in synovium (using siRNA technology) reduced joint inflammation in mice [103
]. The data indicating a proinflammatory role of PAR2 during the development of rheumatoid arthritis (RA) in humans were received after the investigation of biopsies of the patients with RA and osteoarthritis (OA). Of note, PAR2 expression was enhanced in the RA synovium as compared with controls (synovial tissue from patients with OA) [132
, 133
]. Busso and colleagues [132] studied the effect of PAR2 deficiency in different types of arthritis animal models, and a significant beneficial effect of PAR2 gene disruption was found only in an antigen-induced arthritis model. However, the crucial immune cell types mediating the proinflammatory effects triggered by PAR2 activation during joint inflammation remain unknown.
Yang and colleagues [98 ] also shed a new light on the role of PAR1 in a mouse model of antigen [methylated BSA (mBSA)]-induced arthritis. Here, arthritis severity was reduced significantly in PAR1 null mice as compared with wild-type animals [98 ]. The notable reduction of total serum anti-mBSA IgG levels in PAR1 null mice suggested that B cells might be involved in the effects triggered by PAR1 during joint inflammation. Thus, PAR1 activation could play an important role in the development of antigen-induced arthritis. Moreover, the effects of the receptor activation might be mediated via modulation of B cell function. The role of PARs in the regulation of B cell function, however, is still poorly understood.
The role of PARs in the progression of skin inflammation and allergy
PAR1 and PAR2 are widely expressed in human skin, including keratinocytes, endothelial cells, and sensory nerves [32
, 33
, 74
, 90
, 134
]. A role of PARs in acute and chronic inflammatory skin diseases such as contact dermatitis (CD) or atopic dermatitis was demonstrated in several works [89
, 90
, 135
].
In a mouse model of experimentally induced CD, PAR2 plays a crucial role in the development of allergic CD [135 ]. Indeed, chemically induced ear-swelling responses were significantly suppressed in PAR2 null mice as compared with wild-type animals. Moreover, the infiltration of immune cells such as neutrophils, T lymphocytes, macrophages, and eosinophils was inhibited significantly in PAR2-deficient mice [135 ]. Seeliger and colleagues [89] further explored the role of PAR2 during cutaneous inflammation. They used a model of experimentally induced allergic as well as toxic CD to demonstrate that responses such as ear swelling, plasma extravasation, and leukocyte adherence were significantly attenuated in PAR2 null mice as compared with wild-type mice. Remarkably, selectins such as E-selecin seem to be up-regulated after PAR2 activation in humans. PAR2 activation was also found to be involved in CD-induced leukocyte rolling and recruitment into the skin in vivo [89 ]. PAR2 has also been shown to mediate itch. In patients with atopic dermatitis, a chronic inflammatory skin disease mediated by keratinocytes, T cells, endothelial cells, as well as mast cells, PAR2 mediates pruritus and the "triple response of Lewis" (edema, wheal, flare). Thus, serine proteases (tryptase, keratinocyte-derived kallikreins, cathepsins, endothelial-derived trypsin IV) and PAR2 may be important components in the amplification cycle of inflammation and pruritus in atopic dermatitis [90 , 136 , 137 ].
Taken together, the intriguing findings from these studies strongly support the idea that PAR2 activation plays a proinflammatory role during cutaneous inflammation and pruritus and affects leukocyte recruitment to inflammatory sites within the skin. However, clinical studies using protease inhibitors and PAR2 antagonists are still necessary to clarify the role of PAR2 expressed on immune cells and proteases derived from immune cells in the development of atopic dermatitis.
The role of PARs in airway inflammation and allergy
PARs have emerged recently as important effector receptors in the pathophysiology of airway diseases such as asthma, lung edema, and hypersensitivity [70
, 138
, 139
]. There is extensive evidence that PARs are expressed in the airways in a variety of cell types, such as endothelial cells, epithelial cells, smooth muscle cells, neutrophils, and macrophages [70
, 140
]. The effects of PARs on immune cells in the airways of different species have also been verified.
PAR1 appears to play a role in the development of lung injury. Intratracheal instillation of bleomycin in a mouse model reflects conditions of acute and chronic fibrotic phases of lung injury. In this model, PAR1-deficient mice showed a significantly attenuated recruitment of inflammatory cells in the BALF as compared with wild-type mice [141 ].
A potential role of PAR2 in lung injury was investigated recently [92 ]. In this study, PAR2-AP instilled into airspaces caused a dramatic increase in lung endothelial and epithelial cell permeability to proteins. PAR2 activation also led to increased leukocyte recruitment in BALF. These findings suggest that PAR2 activation promotes lung inflammation [92 ].
Some allergens (house dust mite, cockroach) are demonstrated to activate PAR2. In some animal models, the importance of the protease activity of particular allergens for their allergic potential has been shown [142 , 143 ]. Ebeling and colleagues [144 ] reported that PAR2 activation in the airways at the same time as exposure to inhaled antigens induced allergic sensitization; however, exposure to the same antigen alone induced tolerance. Moreover, this PAR2-mediated, allergic sensitization in mice appears to be TNF-dependent [144 ]. A concept that PAR2 contributes to allergic inflammation of the airways was also supported in the study performed by Schmidlin and colleagues [145 ]. On the other hand, a protective role of PAR2 activation was demonstrated recently in an experimentally induced allergic asthma model [146 ]. Here, the pretreatment with PAR2-AP significantly inhibited bronchoconstriction as well as airway hyper-responsiveness and modulated the immune response induced by allergic challenge in sensitized rabbits [146 ]. Additionally, a concept of PAR2 as a cytoprotective receptor involved in prostanoid-dependent cytoprotection in the airways has been supported [147 ]. Intranasal administration of PAR2-AP inhibited the development of airway eosinophilia and hyper-responsiveness in allergic mice via a COX-2-dependent generation of PGE2 [26 ].
Further investigation of cellular elements of the immune system involved in PAR-associated effects during airway inflammation as well as the identification of novel, natural allergens with proteolytic potential acting via PARs will be helpful for development of novel, anti-allergic therapies.
|
TOPABSTRACTINTRODUCTIONPARs AND THE FUNCTION...PAR-MEDIATED EFFECTS ON...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
For instance, although a significant role for PAR activation in the development of an acute inflammatory response has been established in several animal models, there are still important questions concerning the participation of PARs expressed on immune cells during the progression of chronic inflammation. This issue involves what can be termed "the fourth dimension" of inflammation, wherein the role(s) of PARs at one point in time (e.g., the acute inflammatory response) may differ substantially from the role(s) at later times when the resolution of inflammation (or its lack) results in healing or in a chronic inflammatory process. Indeed, although the activation of PAR2 can trigger acute inflammation in a number of settings, its activation at times later than the initial insult can be "anti-inflammatory". Thus, it will be of a key importance to study in-depth the potential roles of PARs 1, 2, and 4 on specific types of immune cells that can be involved at the early and delayed time-points in progression of an inflammatory process. This kind of information will be essential for the development of therapeutic approaches involving the PARs.
The identification of novel endogenous and exogenous (pathogen-derived) proteases, which could signal via PARs under inflammatory conditions in vivo, also represents a prospective research field. Clearly, PAR signaling does not occur in isolation, and the network of signals, conveyed by proteases and inflammatory mediators, must be taken into account for understanding the innate and adaptive response to injury and infection. In this context, it is especially important to extend already accumulated information [93 ] concerning the transcription factor network of downstream PARs, which could modulate the development of inflammation in vivo. This knowledge will allow the interfering of protease–PAR interactions during inflammatory diseases at different stages, making possible a wide spectrum of therapeutic approaches.
Received January 1, 2008; revised February 12, 2008; accepted February 13, 2008.
|
TOPABSTRACTINTRODUCTIONPARs AND THE FUNCTION...PAR-MEDIATED EFFECTS ON...CONCLUSIONS AND FUTURE...REFERENCES |
|---|
-induced apoptosis in monocytes Blood 107,4383-4390-thrombin stimulates proliferation of interferon- differentiated, growth-arrested U937 cells, overcoming differentiation-related changes in expression of p21CIP1/WAF1 and cyclin D1 J. Cell. Physiol. 191,290-297[CrossRef][Medline]
B J. Biol. Chem. 275,29955-29959 secretion by mouse mast cells: threshold-level thrombin receptor and very low level Fc
RI signaling synergistically enhance IL-6 secretion Cell. Immunol. 205,128-135[CrossRef][Medline]B in human dermal microvascular endothelial cells J. Invest. Dermatol. 118,380-385[CrossRef][Medline] after antigen challenge in mice Allergol. Int. 55,271-278[CrossRef][Medline] B J. Invest. Dermatol. 124,38-45[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
