In vivo generated thrombin and plasmin do not activate the complement system in baboons

Ravi S. Keshari, Robert Silasi, Cristina Lupu, Fletcher B. Taylor Jr. and Florea Lupu

Key points

  • In vivo generated thrombin and plasmin do not contribute to complement activation in non-human primates.

  • Bacteria and LPS are the main drivers of in vivo complement activation in E. coli sepsis in baboons.


Sepsis concurrently activate both coagulation and complement systems. While complement activation by bacteria is well-documented, work in mice and in vitro suggests that coagulation proteases can directly cleave complement proteins. We aimed to determine if generation of coagulation proteases in vivo can activate the complement cascade in two highly coagulopathic models. We compared temporal changes in activation biomarkers of coagulation (thrombin-antithrombin, TAT), fibrinolysis (plasmin-antiplasmin, PAP) and complement (C3b, C5a, C5b-9) in baboons infused with factor Xa and phospholipids (FXa/PCPS) vs. LD100 Escherichia coli (E. coli). We found that, albeit different timing, both FXa/PCPS and E. coli infusion led to robust thrombin and plasmin generation. Conversely, only E. coli challenge activated the complement system reaching a maximum at 2 hours post-challenge, during the peaks of LPS and bacteremia but not of TAT and PAP. Despite inducing a strong burst of thrombin and plasmin, FXa/PCPS infusion did not produce measurable levels of complement activation in vivo. Similarly, ex vivo incubation of baboon serum with thrombin, plasmin or FXa did not show noticeable complement cleavage unless supra-physiologic amounts of enzymes were used. Our results suggest that in vivo generated thrombin and plasmin do not directly activate the complement in non-human primates.

  • Submitted June 14, 2017.
  • Accepted October 5, 2017.