Inhibition of C5 or absence of C6 protects from sepsis mortality

Abstract

Inhibiting complement anaphlytoxin C5a during sepsis may prevent sepsis mortality. Although human anti-C5 antibodies exist, their therapeutic use in microbial sepsis has been avoided because of the hypothesis that inhibiting C5b will prevent formation of the bactericidal membrane attack complex (MAC) and worsen clinical outcome. We wished to test the hypothesis that inhibition of C5 would improve outcomes in sepsis. Sepsis was induced in rats by laparotomy and cecal ligation and puncture (CLP) by an IACUC-approved protocol. Sham animals underwent laparotomy without CLP. Following CLP rats were randomized to receive a single IV dose of purified IgG ant-C5 antibody (Ab) or control IgG Ab. Anti-C5 Ab treated rats $(n=20)$ had significantly lower mortality vs. controls $(n=21)$, 20% vs. 52% ($P=0.019$, log-rank). Analysis of bacterial load by culture of spleen and liver homogenates showed a reduction in colony forming units in anti-C5 Ab treated rats vs. control IgG ($P=0.003$ and 0.009, respectively). Anti-C5 treatment reduced lung injury as measured by total MPO content of lung tissue $(P=0.024)$. Finally, rats genetically deficient in C6 production, unable to form MAC but capable of producing C5a and C5b, were protected from CLP-induced sepsis mortality. Our results show that an anti-C5 antibody therapy prevents CLP sepsis-induced mortality and improves lung injury. Inhibition of the complement MAC does not increase bacterial load or mortality, therefore, the use of anti-C5 therapy may be beneficial rather than detrimental in sepsis.

Introduction

Sepsis remains a significant cause of morbidity and mortality affecting grater than 500,000 people annually (Hotchkiss and Karl, 2003). Despite multiple years of research, there are few proven therapies available for sepsis. Recently, complement inhibition has been identified as a potential sepsis therapy (Ward, 2004; Gerard, 2003). Experimental studies have shown that dysregulated overproduction of the anaphylatoxin C5a may be detrimental in sepsis (reviewed in Ward, 2004). Limited human study suggests that elevated C5a serum levels may correlate with mortality in sepsis (Nakae et al., 1996). Direct inhibition of C5a via blocking antibodies or via C5a receptor (C5aR) has reduced mortality from sepsis (Czermak et al., 1999a; Strachan et al., 2000). The mechanism of anti-C5a therapy in sepsis is mediated in part through improved neutrophil phagocyte function and enhanced NADPH oxidase assembly (Huber-Lang et al., 2002).

Inhibition of C5a rather than C5 is the suggested point of complement inhibition during sepsis as this strategy allows production of the C5 split product C5b (Gerard, 2003). C5b is required for optimal formation of the multimeric protein C5b-9 terminal membrane attack complex (MAC). The MAC is assumed to have a protective role in sepsis through its potential ability to directly lyse bacteria in vivo (Bloch et al., 1997). There is also some evidence supporting a role for C5b in clearance of Pseudomonas aeruginosa during experimental lung infection (Younger et al., 2003). However, direct evaluation of MAC function during sepsis has not been fully described. There is some evidence suggesting that C5a and C5b-9 are not always required for host responses in sepsis. Experiments employing genetically C5-deficient B.10 mice demonstrated prolonged survival time and reduced lung injury following sepsis-induced cecal ligation and puncture (CLP) (Olson et al., 1985). Also, C5 and its split products were not required for pulmonary clearance Bordetella bronchoseptia in vivo (Pishko et al., 2004). Furthermore, genetically C5-deficient mice are less susceptible to mortality and organ damage induced by zymosan (Nieuwenhuijzen et al., 1995; Mahesh et al., 1999). These studies suggest that C5 may not be essential to all host responses during inflammation and sepsis. However, these studies are limited as the genetically deficient C5 animal model has the potential for unrecognized adaptive compensatory responses. Furthermore, study of C5 deficiency does not discriminate between the specific contributions of C5a and C5b to the host sepsis response. Acute inhibition of C5 or isolated inhibition of MAC in the presence of functional C5a during sepsis has not been reported. The first objective of this study was to evaluate whether acute inhibition of C5 represents a potential target in treatment of polymicrobial sepsis. A second study objective was to determine whether C5 inhibition altered bacterial clearance as a consequence of reducing C5a and C5b/MAC levels. The final study objective was to determine the relative contribution of MAC to sepsis outcome in the presence of functional C5a.