THE PROTEIN C PATHWAY AND PATHOLOGICAL PROCESSES

Summary

Alterations in expression of Protein C (PC) pathway components have been identified in patients with active inflammatory disease states. While the PC pathway plays a pivotal role in regulating coagulation and fibrinolysis, activated Protein C (aPC) also exhibits cytoprotective properties. For example, PC-deficient mice challenged in septic/endotoxemic models exhibit phenotypes that include hypotension, disseminated intravascular coagulation, elevated inflammatory mediators, neutrophil adhesion to the microvascular endothelium, and loss of protective endothelial and epithelial cell barriers. Further, IBD has been correlated with diminished EPCR and TM levels in the intestinal mucosa. Downregulated expression of cofactor Protein S (PS), as well as PC, is also associated with ischemic stroke. Studies to further elucidate structural elements that differentiate the various functions of PC serve to identify novel therapeutic approaches towards regulating these and other diseases.

Introduction

The PC pathway plays a pivotal role in regulating the extent of coagulation through inactivation of two key co-factors, FV and FVIII [1,2]. However, recent studies have identified other functions of PC that are associated with cytoprotection, i.e., maintainence of vascular barrier integrity, inhibition of apoptosis, and control of inflammation, which may contribute to the regulation of a number of pathological processes [3–7]. These functions have been shown to be mediated through aPC interaction with its endothelial cell protein C receptor (EPCR) and activation of protease activated receptor-1 (PAR-1) [7]. While thrombin can effectively activate PAR-1, downstream effects of this activation process are in direct opposition to that of aPC [8]. These differential responses are believed to be due to aPC/PAR1 preferential cross-activation of sphingosine 1-phosphate receptor-1 (S1P1) while thrombin/PAR1 effects are through cross-activation of sphingosine 1-phosphate receptor-3 [9,10]. A number of studies have demonstrated that alterations in expression of components of the PC pathway occur in active disease states. Studies to elucidate specific functional properties of PC that control these diseases would serve to identify mechanisms that regulate the genesis and progression of associated pathologies.

The Protein C Pathway and its Anticoagulant Function

Human PC is synthesized predominantly in the liver, but recent evidence suggests that in mice, the epididymus, kidney, brain, lung, and cells from male reproductive tissues, are also sites of its synthesis [11]. PC is synthesized as a single polypeptide chain which undergoes coand post-translational modifications. These include β-hydroxylation, γ-carboxylation, glycosylation, and endoproteolytic cleavage, with release of a dipeptide (residues K¹⁵⁶-R¹⁵⁷; human PC numbering). Mature two-chain PC consists of a 262-amino acid residue heavy chain and a 155-amino acid residue light chain. These two chains are linked by a single disulfide bond [12]. PC is multimodular and contains structural elements characteristic of other vitamin K-dependent coagulation proteins, e.g., FVII, FIX, and FX. These modules include, sequentially, from the amino terminus, a Gla-rich motif, two epidermal growth factor-like modules, an activation peptide region, and a serine protease domain. Activation of PC to aPC is catalyzed by the thrombin-TM complex [13], or, less efficiently, by thrombin alone [14]. Activation occurs consequent to cleavage of the Arg¹⁶⁹-Leu¹⁷⁰ peptide bond, with release of a dodecapeptide (residues 158–169) from the heavy chain of the zymogen. The heavy chain is a serine protease with a catalytic triad of H²¹¹, D²⁵⁷, S³⁶⁰ (H⁵⁷, D¹⁰², S¹⁹⁵ in chymotrypsin numbering) [15]. In its anticoagulant role, aPC catalyzes limited proteolytic inactivation of coagulation FV/FVa [1] and FVIII/FVIIIa [2] in the presence of Ca²⁺, phospholipids (PL), and the cofactor, Protein S (PS), thus attenuating production of thrombin. EPCR enhances the activation of PC by binding the protein on cell surfaces and presenting it to the thrombin/TM complex. aPC also plays an indirect profibrinolytic role through its property of limiting thrombin formation, and, consequently, attenuating the subsequent thrombin-catalyzed activation of a fibrinolytic inhibitor, TAFI [16]. Further, aPC enhances fibrinolysis by inactivation of another fibrinolytic inhibitor, PAI-1 [17].

The Protein C Pathway and Cytoprotection

In addition to the anticoagulant activity of aPC, which is manifest when the Gla-domain of aPC binds to cell surfaces [e.g., via anionic PLs and glycosphingolipids, as well as via its cofactors FV(a) and FVIII(a)], aPC also possesses cytoprotective activity in collaboration with EPCR, thus implicating aPC in humoral aspects of the innate immune response. The cytoprotective effects of aPC are based on alterations of gene expression (e.g., antiapoptotic genes), antiinflammation, and preservation of cellular barriers. Protease activated receptors (PAR), especially PAR1, a G-protein-coupled receptor, are involved in the cytoprotective functions of aPC

Arguing against any oversimplification of aPC function is the fact that, in vivo, the loss or gain of aPC activity is inversely related to the levels of thrombin and to the downstream activities (e.g., a proinflammatory agent) of thrombin. For example, PAR1 is also a major receptor of thrombin, most widely studied in vascular cells, and, like aPC/EPCR, also signals through PAR1. Although both aPC and thrombin function via PAR1 as a signaling or co-signaling receptor, in many cases opposite signaling effects are observed with thrombin and aPC, e.g., gene expression in cytokine-stimulated EC [8]. Further, in vitro assays demonstrate that aPC and thrombin counterbalance vascular endothelial barrier protection, in that aPC inhibits thrombin-mediated vascular permeability [9,18,19]. The aPC effects on EC barrier protection depend on raft localization of aPC and EPCR [20], and on cross-activation of S1P1 [9]. On the other hand, barrier-disruptive thrombin-PAR1 signaling appears to involve cross-activation of S1P3 and requires sphk1 in DCs [10]. While much of this work has been accomplished in vitro, a recent study has shown that the aPC-EPCR-PAR1-S1P1 and thrombin-PAR1-S1P3 pathways also function in vivo in sepsis [21], a study that was based on work that identified the involvement of the balance of S1P1/S1P3 ratios in DC-based signaling as a link between coagulation and inflammation in sepsis [10]. In any in vivo inflammatory system in which PC pathway components are the investigative targets, the challenge thus becomes to distinguish between direct and indirect effects of altering PC and aPC levels in terms of altering hemostatic balance through aPC/thrombin ratios and signaling via S1P1/S1P3 ratios.

The Protein C Pathway and Sepsis (Old PPG)

Sepsis results when a bacterial or viral infectious agent induces an inflammatory response that in-turn causes a number of systemic changes, viz., fever or hypothermia, abnormal leukocyte counts, heart rate abnormalities. Microbial mediators, e.g., endotoxin (LPS), are responsible for many pathophysiological events occurring during infections including pyrogenicity, neutropenia, lymphopenia, leukocyte and endothelial cell (EC) activations, induction of an innate immune response, cytokine/chemokine and iNOS production, expression of adhesion proteins on the endothelium and on WBCs resulting in leukocyte trafficking, complement activation, platelet aggregation, coagulopathy, and an increase in capillary permeability.

A serious complication of sepsis is the development of DIC. Acquired DIC begins with the production of a procoagulant state by enhanced TF expression from activated monocytes [22–24]/macrophages [25–27], activated ECs (28–30], and/or activated neutrophils [26,31]. ECs also release soluble TM [28,32], which interferes with anticoagulation. TF expression is mediated by proinflammatory cytokines, e.g., TNFα [33–36], IL1 [33,37,38], IL6 [39], and MIP-2 [39], as well as prostaglandins [40,41]. In addition, suppression of the PC anticoagulant pathway [38], and depletion/consumption of inhibitors, such as ATIII [42], also contributes to disseminated fibrin deposition within the microvasculature. While fibrinolysis is activated early in sepsis-induced DIC by release of tPA from damaged ECs [43], the subsequent enhanced levels of PAI-1 [43,44], and the activation of Thrombin Activatable Fibrinolysis Inhibitor (TAFI) by increased thrombin levels [45,46], results in inhibition of Pg activation and Pm biological activity, respectively. This ultimately causes imbalance of the hemostasis system toward coagulation, a situation found to be important in the outcome of patients with sepsis [47,48]. Further, as a result of depletion of clotting factors, consumptive bleeding can also occur in these patients.

The generation of proteases in DIC, including FXa and thrombin, results in the potential for interactions with pathways that stimulate systemic inflammatory response syndrome (SIRS). In addition to occlusive microvascular thromboses and resulting tissue and organ hypoxia, neutrophil activation, and endothelium damage, SIRS often leads to organ dysfunction in individuals exhibiting DIC. Due to the multiplicity of systems and components that are involved in sepsis, therapeutic approaches have focused on its different stages, including inhibiting coagulation [49–51], enhancing natural anticoagulation [52–54], attenuating proinflammatory cytokine release [55,56], antagonizing cytokine receptors [57], administrating competitive soluble cytokine receptors, inhibiting bradykinin [58], inhibiting iNOS [59], providing anti-inflammatory cytokines [60–62], stimulating fibrinolysis [54,63], use of prostacyclin analogues [64], and designing peptides that bind to endotoxin [25]. Because of the anticoagulant, anti-inflammatory, and profibrinolytic properties of aPC, its effectiveness as a drug for severe sepsis has been studied in animal models of sepsis and in human clinical trials [65]. In this latter case, treatment with aPC (drotrecogin alfa) of 1690 patients with severe sepsis led to a decrease in the relative risk of death at 28 days by 19.4%, although there was a trend toward more serious bleeding associated with its use.

Studies in mice expressing low levels of PC in an LPS-induced endotoxin model demonstrated that an endogenous low level of PC results in early onset DIC, hypotension, thrombocytopenia, organ damage, and diminished survival relative to similarly challenged WT mice [66]. In addition, the inflammatory response was enhanced relative to WT mice. Administration of PC to these low PC mice, during challenge, improved blood pressure and enhanced survival. A more recent study utilized pharmacological interventions and genetic mouse models to demonstrate that diminishing aPC activity in LPS-challenged mice enhanced thrombin/PAR-1 signaling leading to increased inflammation and vascular permeability [21]. Additionally, utilizing an aPC variant (aPC5A) that retains signaling function but has diminished anticoagulant activity rescued the pathology observed in genetically defective PC pathway mice. Further, effects of a defective aPC pathway on inflammation and permeability were reversed in a S1P3 deficiency or through the addition of agonists of S1P1 indicating that a balance in the S1P1/S1P3 signaling is mediated by select protease activation of PAR-1. The importance of the cytoprotective function of aPC in LPS-challenge models was further confirmed through in vivo studies utilizing [E149A]PC which retains anticoagulant function but not signaling function [67]. Another property of aPC is its ability to bind to beta 1 and beta 3 integrins through RGD sequence in the catalytic domain of aPC [68]. Integrins α3β1, α5β1, and αvβ3 regulate neutrophil chemotaxis and aPC has been shown to bind to all three of these integrins. Further studies demonstrated that aPC can simultaneously bind to EPCR and integrin on cell surfaces. In order to confirm that integrin/aPC interactions, which regulate neutrophil recruitment, play a role in regulating the outcome of sepsis, LPS was administered to mice at a LD₉₀ dose in the presence of the RGD peptide. Results from these studies demonstrated a reduction of mortality to 50%. This would indicate that aPC interaction with integrins involved in neutrophil migration also plays a protective role in sepsis.

The Protein C Pathway and Inflammatory Bowel Disease

The inflammatory bowel diseases (IBD), viz., Crohn's disease (CD) and ulcerative colitis (UC), comprise idiopathic chronic conditions that predispose patients to gastrointestinal and colorectal cancers. The early molecular/cellular pathogenesis of this disease includes loss of mucosal epithelial cell (EpC) barrier protection, invasion by pathogenic agents, and activation of mucosal microvascular endothelial cells (EC), which leads to altered hemostasis and abnormal inflammation. Recent studies have linked the PC pathway to the regulation of Inflammatory Bowel Disease (IBD). Clinically, TM and EPCR are diminished in the colonic mucosal microvasculature of IBD patients [69,70], but enhanced in their sera, suggesting increased shedding of TM and EPCR from cells. This appears to correlate with disease activity [70]. Accordingly, restoring the function of the PC pathway has anti-inflammatory effects on cultured MEC and in animal models of colitis, suggesting a pathogenic role of this system in IBD [69,71,72]. Studies have also identified an increase in PAR1 expression in biopsies from IBD patients [73]. Other studies have shown the presence of EPCR on DCs, colocalized with complement receptor CD21, although it is unknown whether these proteins directly interact or function together [71]. EPCR is structurally similar to MHC class 1 proteins, which are antigen-presenting proteins and participants in immunity and inflammation [74]. EPCR, in colonic DCs, is associated with both the membrane and cytoplasm, in contrast to ECs, where it is only associated with the membrane. This potentially implicates a role for EPCR in other functions [75]. Based on indirect evidence, it has been speculated that EPCR functions in pathways leading to antigen processing and presentation. EPCR has been shown to be expressed in DCs from the gut mucosa, and downregulated by exposure to bacteria [75]. These results suggest that EPCR is involved in immune responses to intestinal antigens. Inflammatory cytokines also downregulate TM and EPCR by inhibiting transcription in MECs [69]. In vitro, addition of aPC to human intestinal MECs (HIMECs) inhibited TNFα-mediated increases in cell adhesion molecules and secretion of MCP-1 [69]. Additionally, this study demonstrated that aPC diminished T cell adhesion to HIMECs. In mice treated with DSS, a reduced ability to convert PC to aPC was observed [69], and administration of aPC to DSS-challenged mice reduced the colitis disease activity index (DAI) and improved the colitis histology score [69]. However, the importance of TM and EPCR, in vivo, in these processes is unknown and further investigations on the impact of altered expression of EPCR and functional TM will facilitate in delineating the role of aPC activity in IBD.

The Protein C Pathway and Ischemic Stroke

Ischemic stroke is the result of a focal brain infarction with downstream deleterious effects on neurological functions. Common causes of ischemic stroke are embolisms or arterial thrombosis, resulting in diminished cerebral blood flow. Pathologies of ischemic injury include edema, thrombosis of microvessels, apoptosis, and cell necrosis. Clinically, arterial ischemic stroke has been associated with PS deficiency [76] and plasma PC level is an inverse risk factor for stroke [77].

Studies in mice using a middle cerebral artery occlusion/reperfusion model of ischemic stroke demonstrated that survival is enhanced in mice administered aPC either before or shortly after the onset of stroke. In the aPC treated mice, brain infarct size, edema, and brain infiltrated neutrophils were reduced. Additionally, the extent of fibrin deposition and ICAM-1 expression in cerebral vessels was diminished [78]. In vitro studies using hypoxic human brain endothelial cells, demonstrated that aPC, but not the [S360A]PC variant, prevented apoptosis through inhibition of p53, normalization of Bax/BCl-2, and diminishing caspase 3, events which were dependent on both EPCR and PAR-1 [79]. Studies in murine cortical neurons demonstrated that aPC can protect these cells from apoptosis induced by either NMDA or staurosporine. In vivo, intracerebral infusion of aPC diminished NMDA excitoxicity in mice in a dose-dependent manner and both in vitro and in vivo neuronal protection required PAR-1 and -3 [80]. Additionally, aPC has been shown to diminish tPA-mediated neurotoxicity, in vitro, and cerebral ischemic injury, in vivo [81]. Further, the safety and efficacy of late administration of aPC in mice following the onset of ischemia was studied. aPC administered as a single dose or multiple times 6–72 hr or 72–144 hr after the onset of ischemia, demonstrated significant improvement in cerebral perfusion in areas of ischemia and efficiently blocked blood-brain barrier leakage. Both EC replication and neurogenic progenitor cell proliferation were enhanced. Further, migration of neuroblasts towards ischemic areas was also enhanced. These neovascular and neurogenesis effects required PAR-1. Results from these studies indicate that aPC has a wide therapeutic window for effectively controlling downstream effects of ischemic brain injury [82]. Of further relevance, aPC and aPC variants with diminished anticoagulant activity, but sustained cytoprotective function, can cross the blood-brain barrier and this process is more efficient than with zymogen PC, a process that requires EPCR but not PAR-1 [83]. Results from these studies support a therapeutic role for aPC in regulating pathologies associated with ischemic stroke that is not necessarily dependent on its anticoagulant function.

Conclusions

While much is known of the role of the PC pathway in regulating coagulation, it is only recently that its function as a cytoprotective mediator has been revealed. Future studies using recombinant variants of components of the PC pathway, genetically modified mice, and experimental models of diseases will further assist in elucidating mechanisms by which PC regulates cell functions during disease states.