Enterococcus faecalis stimulates collagenolysis by macrophages via plasminogen activation: a role in anastomotic leak

Author(s):
Richard Jacobson; John Alverdy; Olga Zaborina

Background:

We have previously shown that Enterococcus faecalis can cause anastomotic leak (AL) in rodents via its production of collagenase that promotes macrophage matrix metalloprotease 9 (MMP9) activity, resulting in collagen degradation and impaired healing. Despite recent work confirming its role in bacterial virulence, the involvement of the fibrinolytic protease plasminogen (PLG) in the pathogenesis of E. faecalis-mediated AL is unknown. Gut macrophages are known to bind and activate PLG, and its activation is readily inhibited by pharmaceuticals already in use in elective surgery. Therefore, the aim of this study is to define the relative contribution of E. faecalis-induced PLG activation by macrophages on collagen degradation as it applies to AL.

Hypothesis:

The interaction of E. faecalis with macrophages leads to overactivation of PLG and contributes to the accelerated collagen degradation seen in AL.

Methods:

RAW 264.7 immortalized murine macrophages were incubated with a collagenolytic E. faecalis strain (E44), previously shown to cause AL in rodents. E44 was plated at various multiplicities of infection (MOI). Activation of PLG was assessed kinetically with a PLG-specific fluorogenic substrate. Fluorescein-labeled gelatin was used to assess collagenolytic activity. Given that the activation of PLG by macrophages is through apical surface alpha-enolase (ENO1), and that active PLG can activate MMP9, tranexamic acid (TXA) and anti-mouse enolase antibodies were used to inhibit PLG activation.

Results:

When macrophages were plated with live E44, a direct correlation was observed between PLG activity and MOI (no E44 36.9±2.9 AU(PLG activity); MOI2 40.6±3.5; MOI10 80.9±3.6, p<0.01). Subsequently, collagenolytic activity was increased in a MOI-dependent fashion, particularly in the presence of pro-MMP9 (no E44 28.8±1.2 AU(collagenase); MOI 10 35.8±1.4 p<.01). TXA significantly decreased PLG activation in a concentration-dependent manner and decreased collagen degradation both in the presence and absence of pro-MMP9 (MOI10/no TXA 21.0±1.0 AU(collagenase); MOI10/5mM TXA 16.5±1.0; MOI10/10mM TXA 14.0±1.6, p<.01). Anti-mouse enolase antibodies diminished PLG activation in E44-stimulated macrophages (No Ab 22.3±0.3 AU(PLG activity); Control Ab 20.5±0.3; Anti-ENOL 15.6±0.4 p<.01).

Conclusions:

E. faecalis-mediated PLG activation occurs through a surface-exposed enolase on murine macrophages and significantly contributes to collagen degradation. Titrated use of the FDA approved agent TXA can potentially inhibit excessive collagen degradation at its most proximate point of activation and may play a preventative role in AL.