Search

Background/Aims:Enteroadherent Escherichia colithat produce Shiga-like toxins are important causes of human disease, including enterohemorrhagic E. coli-induced colitis (EHEC). The role of Shiga-like toxins in these illnesses is unclear. The aim of this study was to establish an animal model for human EHEC and to determine the role of Shiga-like toxin I (SLT-I) in this model. Methods: E. colistrain RDEC-1 is an enteroadherent rabbit diarrheal pathogen. An isogenic variant of RDEC-1 (termed RDEC-H19A) producing high levels of SLT-I was obtained by infecting RDEC-1 with an SLT-I-converting bacteriophage. The effects of in vivo enteric infection produced in rabbits by RDEC-H19A were compared with those in uninfected and RDEC-1-in-fected animals. Results:SLT-I-producing RDEC-H19A induced a severe, noninvasive, enteroadherent infection in rabbits. Clinically, infection with RDEC-H19A was more severe than infection with RDEC-1 and caused more serious histological lesions including vascular changes, edema, and more severe inflammation. Interleukin 1 and platelet-activating factor appear to be important inflammatory mediators to this infection. Conclusions:The illness induced by RDEC-H19A in rabbits resembled enterohemorrhagic E. coli-induced colitis of humans. SLT-I is an important virulence factor in the pathogenesis of EHEC.

The role of defective fibrinolysis caused by elevated activity of plasminogen activator inhibitor-1 (PAI-1) in promoting fibrin deposition in vivo has not been well established. The present study compared the efficacy of thrombin or ancrod, a venom-derived enzyme that clots fibrinogen, to induce fibrin formation in rabbits with elevated PAI-1 levels. One set of male New Zealand rabbits received intravenous endotoxin to increase endogenous PAI-1 activity followed by a 1-hour infusion of ancrod or thrombin; another set of normal rabbits received intravenous human recombinant PAI-1 (rPAI-1) during an infusion of ancrod or thrombin. Thirty minutes after the end of the infusion, renal fibrin deposition was assessed by histopathology. Animals receiving endotoxin, rPAI-1, ancrod, or thrombin alone did not develop renal thrombi. All endotoxin-treated rabbits de-developed fibrin deposition when infused with ancrod (n = 4) or thrombin (n = 6). Fibrin deposition occurred in 7 of 7 rabbits receiving both rPAI-1 and ancrod and in only 1 of 6 receiving rPAI-1 and thrombin (P< .01). In vitro, thrombin but not ancrod was inactivated by normal rabbit plasma and by purified antithrombin III or thrombomodulin. The data indicate that elevated levels of PAI-1 promote fibrin deposition in rabbits infused with ancrod but not with thrombin. In endotoxin-treated rabbits, fibrin deposition that occurs with thrombin infusion may be caused by decreased inhibition of procoagulant activity and not increased PAI-1 activity.