Endogenous protein C inhibits activation of coagulation and transiently lowers bacterial outgrowth in murine Escherichia coli peritonitis


Marcel Schouten, Center for Experimental and Molecular Medicine (CEMM), Center for Infection and Immunity Amsterdam (CINIMA), Academic Medical Center, University of Amsterdam, Meibergdreef 9, Room G2-130, 1105 AZ Amsterdam, the Netherlands.
Tel.: +31 20 566 5910; fax: +31 20 697 7192.
E-mail: m.schouten@amc.uva.nl

Sepsis is the leading cause of death in critically ill patients. Peritonitis is the second most common cause of sepsis, with Escherichia coli being one of the major pathogens involved [1]. Severe bacterial infection results in systemic activation of coagulation accompanied by downregulation of anticoagulant mechanisms and fibrinolysis [2]. In murine peritonitis, coagulation is activated intraperitoneally, resulting in intra-abdominal fibrin formation, which, on the one hand, limits bacterial spreading, but on the other hand hampers bacterial killing [3]. Indeed, administration of the coagulation inhibitor activated protein C (APC) not only decreased coagulation activation, but also improved bacterial clearance and survival in a murine model of peritonitis [4]. The endogenous protein C (PC) system is important for the regulation of coagulation and inflammation, as illustrated by studies in mice with very low PC levels because of genetic deficiencies, which display severe procoagulant and inflammatory responses to lipopolysaccharide [5]. We investigate here the role of the endogenous PC system in local and systemic activation of coagulation and fibrinolysis, bacterial outgrowth and inflammation during murine E. coli peritonitis.

Ten-week-old female C57BL/6 mice were used. All experiments were approved by the Institutional Animal Care and Use Committee of the Academic Medical Center. Peritonitis was induced by intraperitoneal administration of 104 colony-forming units of E. coli O18:K1 [6,7]. Thirty minutes prior to induction of peritonitis, mice received an intraperitoneal injection of 200 μL of isotonic saline containing 200 μg of either MPC1609, a rat mAb directed against murine PC (anti-PC), or MCO1716, a cross-matched mAb targeted against the mouse keyhole limpet hemocyanin protein (control antibody) [8]. Mice were killed after 6, 14 or 20 h. Samples were harvested and processed as previously described [9]. PC plasma activity levels were measured with an amidolytic assay [10]. Thrombin–antithrombin (TAT) complexes (Behringwerke AG, Marburg, Germany) and plasminogen activator inhibitor-1 (PAI-1) [11] were measured by ELISA. Plasminogen activator activity (PAA) was determined with an amidolytic assay [12]. Tumor necrosis factor (TNF)-α, interleukin (IL)-6, monocyte chemoattractant protein (MCP)-1, interferon (IFN)-γ and IL-10 were measured with a cytometric bead array multiplex assay (BD Biosciences, San Jose, CA, USA).

Anti-PC treatment strongly reduced PC levels in peritoneal lavage fluid (PLF) and plasma (Fig. 1A,B). TAT complex levels in PLF were significantly higher in anti-PC-treated animals than in mice treated with control antibody at 14 h (P < 0.05) and 20 h (P < 0.001) after infection (Fig. 1C). Plasma TAT complex levels were already higher in anti-PC-treated animals at 6 h after infection, and remained higher during the course of peritonitis (P < 0.01 vs. mice treated with control antibody at all time points) (Fig. 1D). These findings are in line with those of previous studies in which anti-PC enhanced plasma TAT complex levels after intravenous administration of lipopolysaccharide in mice [7] and triggered fibrinogen consumption after intravenous infusion of E. coli in baboons [13]. Notably, however, heterozygous PC-deficient mice showed unaltered systemic activation of coagulation during polymicrobial peritonitis induced by cecal ligation and puncture; this study did not report data on local activation of coagulation and inflammation or bacterial outgrowth [14]. Evidence derived from in vitro investigations suggests that APC may stimulate fibrinolysis by inhibiting the main inhibitor of this system, PAI-1 [2]. Anti-PC greatly increased PAI-1 levels in PLF (P < 0.05 to P < 0.001 vs. mice treated with control antibody) (Fig. 1E), but this was not paralleled by a decrease in PAA in PLF (Fig. 1G), suggesting that, in the peritoneal cavity, inhibition of PAA does not solely depend on PAI-1 and/or that the increased PAI-1 release was accompanied by a concurrent increase in the release of tissue-type or urokinase-type plasminogen activator. Anti-PC treatment also increased plasma PAI-1 levels at 6 and 14 h of infection (P < 0.05 vs. mice treated with control antibody) (Fig. 1F), and this was accompanied by strongly reduced PAA at all time points (P < 0.01 to P < 0.001) (Fig. 1H). In vitro data suggest that APC reduces PAI-1 levels [2]. Our study is the first to show the inhibitory effect of endogenous PC on PAI-1 levels in vivo. Higher PAI-1 levels, resulting in lower systemic fibrinolytic activity, probably further contribute to enhanced fibrin formation in anti-PC-treated animals.

Figure 1.

 Treatment with anti-protein C (PC) antibodies enhances activation of coagulation, increases plasminogen activator inhibitor-1 (PAI-1) levels, inhibits systemic fibrinolytic activity, enhances systemic cytokine production and enhances bacterial outgrowth in murine Escherichia coli peritonitis. (A–J) Peritoneal lavage fluid (PLF) and plasma levels of PC) activity (A, B), thrombin–antithrombin (TAT) complexes (C, D), PAI-1 (E, F), and plasminogen activator activity (PAA) (G, H). (I, J) Plasma levels of interleukin (IL)-6 (I) and monocyte chemoattractant protein (MCP)-1 (J). (K, L) Bacterial outgrowth in PLF (K) and blood (L) 6, 14 and 20 h after induction of Ecoli peritonitis in control antibody-treated (white) and anti-PC-treated (gray) mice. Data are expressed as box-and-whisker diagrams depicting the smallest observation, lower quartile, median, upper quartile, and largest observation (eight mice per group at each time point). Statistical significance as compared with control antibody: *P < 0.05, **P < 0.01, and ***P < 0.001, respectively, Mann–Whitney U-test). CFU, colony-forming unit; ND, not determined.

Anti-PC treatment did not impact on cell counts or differentials or on cytokine concentrations in PLF at any time point after infection (data not shown). Whereas no differences in cytokine levels were detected at 6 h after infection, at 14 h plasma levels of IL-6, MCP-1 and IL-10 were substantially higher in anti-PC-treated animals (P < 0.01 vs. mice treated with control antibody; shown for IL-6 and MCP-1 in Fig. 1I,J); after 20 h, these differences had subsided. There were no differences in TNF-α and IFN-γ levels between groups at any time point (not shown). IL12p70 levels were below the limit of detection at all time points.

It is of interest that, at 14 h post-infection, anti-PC treatment enhanced bacterial loads approximately 10-fold in PLF (Fig. 1K), blood (Fig. 1L), liver and lung (not shown) (P < 0.05 vs. mice treated with control antibody in all compartments tested); after 20 h of infection, bacterial loads were very high in all mice, and differences between groups had subsided. The inhibitory effect of endogenous PC on bacterial growth and dissemination is remarkable, considering that (A)PC is not known to impact on antibacterial mechanisms per se, and considering that anti-PC did not influence the inflammatory response to E. coli in a way that might have impaired antibacterial defense. It is conceivable that the lower bacterial loads in PLF of mice not treated with anti-PC result from less activation of coagulation, which could deprive bacteria of an intraperitoneal niche provided by fibrin, as was suggested earlier [3]. This is in agreement with studies from our group that linked less activation of coagulation or enhanced fibrinolysis to lower bacterial counts in murine E. coli peritonitis [6,7]. The differences in blood, liver and lung bacterial outgrowth in our study are probably secondary to different outgrowth at the primary site of infection.

In conclusion, we show that endogenous PC decreases local and systemic activation of coagulation and enhances systemic fibrinolytic activity in murine E. coli peritonitis, thereby limiting the procoagulant trigger elicited by this rapidly disseminating infection. Moreover, we demonstrate that endogenous PC transiently lowers bacterial outgrowth, possibly by preventing bacteria from using fibrin clots to escape from bacterial killing. These data reveal endogenous PC as an important regulator of an adequate antibacterial host response during abdominal sepsis.


The authors thank J. Daan de Boer, who assisted in measuring inflammatory markers. M. Schouten is sponsored by a grant of the Dutch Thrombosis Foundation (grant number TSN 2005-1).

Disclosure of Conflict of Interest

The authors state that they have no conflict of interest.