Signaling the brain in systemic inflammation: the role of complement

The complement (C) cascade is activated in almost immediate reaction to the appearance in the body of pathogenic microorganims and their products, e.g., bacterial endotoxic lipopolysaccharide (LPS), resulting in the generation of a series of potent bioactive fragments that have critical roles in the innate immune response of the afflicted host, including, potentially, the production of the fever that so characteristically marks bacterial infections. For instance, its derivatives C3a, C3b, iC3b, C5a, and C5b-9 independently induce the production by myeloid and non-myeloid cells of the cytokines interleukin (IL)-1(, IL-6 and tumor necrosis factor-(, and of prostaglandin (PG)E2, all putative mediators of fever. Therefore, any one of these C components could be involved, centrally or peripherally, in the induction of the febrile response to LPS. Indeed, we have shown that hypocomplementation by cobra venom factor (CVF) dose-dependently attenuates LPS-induced fever in guinea pigs and wild-type (WT) mice, and that C5 gene-ablated mice are unable to develop fever after LPS. In further studies, we found that a specific antagonist to the C5a receptor, C5aR1a, prevents the LPS-induced febrile rise of WT and C3 null mutant mice, implicating C5a as the responsible factor. Various lines of evidence from our laboratory suggest that the macrophages of the liver (Kupffer cells [Kc]) may be the specific target cells of C5a and that the product they release may be PGE2. PGE2, in turn, may be the substance that binds to vagal afferents in the liver that convey the pyrogenic message to the brain. Other studies by our group (not included in this review) have separately traced the neural pathway by which this message may be transmitted from the liver to the brain and processed there for action. The purpose of this article is to review the studies that have led us to conclude that C5a, Kc and Kc-generated PGE2 may be integrally involved in the pathogenesis of LPS fever. If further verified, these results will be important for better understanding how infectious stimuli may trigger the multivariate acute-phase responses generally, and fever particularly, that promptly spring into action to defend the continued well-being of the afflicted host.