Search

(1) Cobra venom factor (CVF)-induced hypocomplementemia dose-dependently attenuates the febrile responses of guinea pigs and mice to intraperitoneally (ip) but not to intravenously (iv) injected endotoxic bacterial lipopolysaccharide (LPS). (2) Iv but not ip LPS causes fever in complement component 3 (C3) gene-ablated mice, but neither iv nor ip LPS evokes a body core temperature (Tc) rise when WT and these mice's C5a receptors type 1 are blocked. C5 knockout mice also do not develop fever following either iv or ip LPS. C5a thus appears to be a critical mediator of LPS fever. (3) C5 knockouts develop fever in response to intracerebroventricularly (icv) injected LPS or prostaglandin (PG)E2; the site of action of C5a is therefore peripheral rather than central. (4) The initiation of the febrile responses to both iv and ip LPS is temporally correlated with the appearance of LPS in the liver's Kupffer cells (Kc). (5) PGE2 is released by liver in immediate response to the injection of CVF into the portal vein of anesthetized guinea pigs; its level rises quickly to its maximum. LPS injected similarly also evokes a quick release of PGE2 from the liver; it, however, is prevented by prior hypocomplementation. (6) Neither LPS nor IL-1 β induces PGE2 release from Kc in vitro within the first hour after treatment, but serum C and C plus LPS or IL-1 β very quickly trigger PGE2 increases of similar magnitudes, catalyzed non-differentially by cyclooxygenase (COX)-1 and COX-2. Kc would thus appear to be the principal site of action of C5a, inducing the release of PGE2. (7) PGE2 is detectable in the plasma of conscious guinea pigs in temporal correlation with the onset of the Tc rise following ip LPS; cytokines appear significantly later. (8) Taken together, these results indicate that LPS-activated C, rather than LPS or IL-1 β by itself, triggers PGE2 release by Kc. This PGE2 could be the factor that stimulates vagal afferents, thereby providing the signal to the brain that mediates the febrile response.

(1) It is generally considered that fever is modulated in the preoptic-anterior hypothalamic area (POA) in response to signaling by pyrogenic cytokines elaborated in the periphery by mononuclear phagocytes and the consequent induction of prostaglandin (PG)E 2 in the POA. The mechanism of the centripetal transmission of this pyrogenic signal, however, is controversial. One hypothesis suggests that it is conveyed via the vagus to the nucleus tractus solitarius and from there to the POA via the ventral noradrenergic bundle, causing the intraPOA release of norepinephrine (NE) which then stimulates the production of PGE 2 . (2) In this article, we review recent data from our laboratory showing that NE microdialyzed into the POA of conscious guinea pigs or injected intracerebroventricularly into conscious mice indeed evokes two distinct core temperature ( T c ) rises, viz., one α 1 - adrenoceptor (AR)-mediated, rapid in onset and PGE 2 -independent, and the other α 2 - AR - mediated , delayed and COX-2/PGE 2 -dependent. (3) We further present new data suggesting that the febrile response of conscious guinea pigs to intraperitoneally injected lipopolysaccharide (LPS) is mediated by intraPOA NE in accord with the above sequence, i.e., via α 1 - AR to initiate the first, PGE 2 -independent elevation of T c , and via α 2 - AR to induce the delayed production of COX-2-dependent PGE 2 and the continued rise of T c . (4) These results thus validate the presumptive involvement of NE in LPS fever induction in guinea pigs.

The complement (C) system and Kupffer cells (Kc) play critical roles in the febrile response of guinea pigs to i.p.-, but not i.v.-injected lipopolysaccharide (LPS). We have investigated whether C and Kc may also mediate this species’ febrile response to muramyl dipeptide (MDP) and polyriboinosinic–polyribocytidylic acid (poly I:C), two other exogenous pyrogens. Both i.v. and i.p. injections of MDP (150 μg/kg) caused initial, transient falls (0.6°C after 90 min by both routes), then rises (0.8°C at 220 and 0.5°C at 200 min, respectively) in core temperature (Tc). Cobra venom factor (CVF)-induced reductions in C variably affected these courses, but the fever indices and the reductions in serum C activity were not correlated. In addition, neither the i.v. nor the i.p. injection of MDP per se induced a significant change in the serum C activity. The i.v. injection of poly I:C (30 μg/kg) caused a 0.6°C Tc rise which started within 30 min, peaked at 60 min, and recovered in 5 h. The i.p. injection of poly I:C (50 μg/kg) caused a fever similar to that produced by the i.v. injection, but with a later onset, starting at 90 min after drug administration. Again, C reduction did not affect the overall fever indices, although some individual responses were slightly attenuated. The i.v. injection of poly I:C per se also did not affect the serum C activity. These results indicate that, in contrast to its role in i.p. LPS-induced fever, C is not involved in the febrile responses of guinea pigs to both i.v. and i.p. MDP or poly I:C.

That guinea pigs pretreated with the mast cell stabilizing agent cromolyn (Fig. 1) and mice congenitally deficient in mast cells (WBB6FI/J-kitw/kitw-v; Fig. 2) develop normal fevers to iv LPS. These latter data would therefore exclude mast cells as the contributors of C-induced degranulation products (including, e.g., 5-hydroxytryptamine, the receptors for which also exist on vagal afferents) that could account for the rapid febrile response to LPS.