Bacterial sepsis induces strong activation of coagulation, complement and fibrinolytic systems that contribute to disseminated intravascular coagulation, organ damage and death. While the contact of the blood with pathogens or pathogen-associated molecular patterns (PAMPs) can trigger the activation of both systems, a bidirectional complement-coagulation crosstalk is believed to occur. Although the role of complement activation products as positive regulators of coagulation is documented, direct activation of the complement proteins by thrombin or other hemostatic proteases was alluded but not demonstrated in vivo. Here we aimed to: (i) determine if in vivo generation of thrombin and other hemostatic proteases can activate the complement proteins and (ii) discriminate between the direct effect of the pathogen/PAMPs vs. hemostatic proteases on complement activation in a clinically relevant model of sepsis. We have compared the time-course of complement activation markers (C3b, C5a and C5b-9 terminal complex) in plasma of baboons exposed to 1010 cfu/kg (LD100) E. coli vs. intravenous infusion of factor Xa/PC:PS, a potent procoagulant stimulus. In baboons challenged with LD100 E. coli, complement activation markers C3b, C5a and C5b-9 reached maximum levels after 2 hrs (see figure). Complement activation coincided with the peak of bacteremia and LPS, but not with markers of thrombin generation (TAT and fibrinogen consumption; see figure) or fibrinolysis (FDP, D-dimers), which reached peak levels after 6 hours. Differently, infusion of FXa/PC:PS (36.6 pmol/L FXa and 56.3 nmol/L PC/PS per kg body weight) induced a rapid burst of thrombin and almost full consumption of fibrinogen during the first 10 min post-infusion, with no increase of complement activation markers. Based on these data we conclude that in vivo activation of the coagulation cascade does not support complement activation as was postulated by previous in vitro studies. Therefore, we conclude that pathogens and PAMPs are the main activators of the complement during sepsis while direct activation by hemostatic proteases is minor or absent.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution