Pyrogenic signaling via vagal afferents: what stimulates their receptors?

Autonomic Neuroscience: Basic and Clinical 85 (2000) 66–71 www.elsevier.com / locate / autneu

Review

Pyrogenic signaling via vagal afferents: what stimulates their receptors? Clark M. Blatteis*, Shuxin Li Department of Physiology, The University of Tennessee Health Science Center, 894 Union Avenue, Memphis, TN 38163, USA

Abstract Although there is good evidence that pyrogenic messages may be conveyed from the periphery to the brain via vagal afferents, the exact nature of the factors that activate their sensory terminals is unclear. Since IL-1b and PGE 2 have established roles in fever production and since their receptors have been identified on or near vagal nerves, they are potential candidate mediators. A difficulty, however, is that (1) IL-1b is not expressed constitutively in mononuclear phagocytes, their presumed cell source upon stimulation by exogenous pyrogens, e.g. endotoxin, and (2) similarly, the isoform of the enzyme that selectively mediates the production and release of PGE 2 by endotoxin-stimulated macrophages, COX-2, is also not constitutively expressed in these cells. Since the transcription and translation of these factors significantly lags the onset of fever induced by endotoxin administered intravenously, in particular, it is possible that a secondary, quickly-acting mediator evoked in almost immediate reaction to the presence of endotoxin excites, directly or indirectly, the sensory neurons. We have evidence that the complement component C5 contributes importantly to the initiation of the febrile response to endotoxin. This article briefly reviews the prevailing concepts of pyrogen sensing and signaling, examines their shortcomings particularly in terms of the temporal discrepancy between the very rapid onset of the febrile response to intravenously administered endotoxin and the significant delay in the elaboration of the putative mediators of fever, and presents newer data that may help to integrate the various preposed mechanisms.  2000 Elsevier Science B.V. All rights reserved. Keywords: Fever; Cytokines; Prostaglandin E2; Cyclooxygenases-1 and 2; Complement; Endotoxin; Bacterial infection

1. Introduction As described in the preceding, pertinent review articles (Berthoud, 2000; Goehler, 2000; Romanovsky, 2000), there exists compelling evidence that, notwithstanding some contradictory data (Goldbach et al., 1997; Luheshi, 1998; Hansen et al., 2000), the vagus may play an important role in the centripetal signaling of the preoptic area (POA) of the anterior hypothalamus by peripheral pyrogens for fever production. In support of this notion are the demonstrations of the constitutive presence of binding sites for interleukin (IL)-1 ligands on glomus cells located within abdominal vagus nerve-associated paraganglia (Goehler et al., 1997) and of the expression 45 min after intraperitoneal (i.p.) endotoxin [lipopolysaccharide (LPS) a constituent of the outer wall of Gram-negative bacteria] administration of IL-1b immunoreactivity in dendritic cells and macrophages present in these paraganglia (Goehler et al., 1999). Also very supportive are the findings of the enhanced expression of IL-1 receptor type I (IL-1RI) *Corresponding author. Tel.: 11-901-448-5845; fax: 11-901-4487126. E-mail address: [email protected] (C.M. Blatteis).

mRNA in the nodose ganglion and of the prompt, indomethacin-sensitive increase in vagal afferent discharge activity in response to the intravenous (iv) injection of IL-1b (Ek et al., 1998). Indeed, the focus on IL-1b as the signaling molecule in this neural pathway is well justified in view of the prevailing concept of fever production, i.e. that exogenous pyrogens that invade the body do not cause fever directly, but rather induce the production and release by systemic mononuclear phagocytes of endogenous mediators, in particular the cytokines tumor necrosis factor (TNF)a, IL-1b and IL-6, which are considered to be the principal endogenous pyrogens. The process by which their pyrogenic messages are then conveyed to the POA for action, however, is controversial. There are several possible pathways among which the vagal route is only one, and evidence for all of them has been adduced. They bear briefly reviewing again.

2. Possible mechanisms of afferent pyrogenic signaling To the extent that exogenous, especially bacterial,

1566-0702 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S1566-0702( 00 )00221-6

C.M. Blatteis, S. Li / Autonomic Neuroscience: Basic and Clinical 85 (2000) 66 – 71

pyrogens are believed ultimately to be cleared primarily by the macrophages of the liver (Kupffer cells) (Mathison and Ulevitch, 1979), the cytokines they produce may be released into the circulation and transported by the blood stream to the brain which 1. they may enter by being actively transported across the blood–brain barrier (BBB) by cytokine-specific carriers (Banks et al., 1996); or 2. they may enter it directly where the BBB is ‘leaky’, e.g., the organum vasculosum laminae terminalis (OVLT) (Blatteis et al., 1983; Komaki et al., 1992); or alternatively, and more likely, 3. they may stimulate relevant receptors on neurons that extend into or to near the perivascular space of this region by means as yet uncertain but probably involving the secondary release of the lipid mediator prostaglandin (PG)E 2 , the most proximal putative mediator of fever (Blatteis et al., 1989; Coceani and Akarsu, 1998; Lacroix and Rivest, 1998). 4. The circulating cytokines may also not penetrate the brain at all, but rather induce the generation abluminally of PGE 2 by endothelial cells of the cerebral microvasculature (Matsumura et al., 1998; Quan et al., 1998; Rivest, 1999; Li et al., 1999b) and / or by perivascular microglia and meningeal macrophages (Elmquist et al., 1996; Schiltz and Sawchenko, 1998). The secondary induction of cytokines by cerebral microvessels and / or by parenchymal microglia has also been suggested in this latter context. Although entirely plausible and, indeed, supported by evidence, these mechanisms fall short in one important respect: since all are predicated on the arrival of circulating cytokines at their sites of action (OVLT, cerebral microvasculature, etc.), their plasma levels and the febrile course should correlate. Indeed, the cytokines should appear in the blood in detectable quantities before the induction of fever provoked by exogenous pyrogens. Elevated cytokine levels have, in fact, been demonstrated in plasma in correlation with the extent of fever after LPS administration i.p. (Kluger et al., 1995), intramuscularly (i.m.; Roth et al., 1994; Jansky et al., 1995), and subcutaneously into an airpouch (Ross et al., 1999; Cartmell et al., 2000), but not consistently so after iv LPS (Givalois et al., 1994; Luheshi, 1998). Indeed, in the latter case, the appearance in plasma of the first of the pyrogenic cytokines detected, TNFa, lags that of ACTH (which is coincident with or may even precede the onset of fever) produced by iv LPS by as many as 15 min (Givalois et al., 1994). This, however, is not surprising since these cytokines are not constitutively expressed in mononuclear phagocytes, but rather are transcribed, translated, and secreted by these cells in response to a pyrogenic stimulus, a process that requires in vitro, at a minimum, 1 h (Newton, 1986). This interval, on the other hand, concords

67

well with the onset latencies of fever induced by low to moderate doses of LPS administered i.p. (|90 min). The above four mechanisms of fever initiation could logically, therefore, be operative when exogenous pyrogens are delivered into tissue rather than directly into the circulation. The onset latencies of fever after i.p. LPS, however, become reduced in a dose-related manner such that, at higher doses, they can be as short as those after i.v. LPS (Blatteis et al., 2000). Hence, it would seem unlikely that the prompt onset of fever under these conditions could be attributed to the actions of circulating cytokines if these were not yet present in the blood. Confounding the issue further is recent evidence that TNFa may not even have a part in the induction of fever (Roth et al., 1998) whereas, on the other hand, IL-6 may be essential (Chai et al., 1996; Cartmell et al., 2000). Since IL-6 is induced by both TNFa and IL-1b, its appearance in blood is even further delayed; and, as already mentioned, the presence of IL-1b in plasma after LPS by any route is itself reportedly inconstant (Rothwell, 1997; Luheshi, 1998). In sum, it would seem improbable that circulating cytokines could provide the signals for the very prompt induction of fever after i.v. LPS or after high doses of i.p. LPS. On the other hand, the concentrations of pyrogenic cytokines could rise to levels sufficient to excite appropriate local sensors in the vicinity of the cells that produce them, if they existed there, well before these mediators were detectable in the general circulation. Indeed, the rapidity of the febrile response to i.v. LPS would imply a neural rather than a blood-borne pathway between peripheral cytokines and the POA. That is why the evidence for the transmission of peripheral pyrogenic signals to the CNS via sensory nerves originating in the liver or in the abdomen (and probably from other peripheral sites as well) is so appealing.

3. Putative signal molecules

3.1. Cytokines However, there may be a fly in this ointment also, and, again, it is that the time necessary for the synthesis and secretion of these cytokines, and IL-1b in particular (since thus far only binding sites for this cytokine have been identified in the vicinity of vagal afferents) is longer than the onset latency of i.v. LPS-induced fever. If, as already discussed, TNFa does not mediate fever initiation and IL-6 occurs after TNFa and IL-1b, then IL-1b would indeed be the fever-triggering stimulus. But, to be effective so quickly, it would have to be present constitutively near the vagus nerve-associated paraganglia that express its presumptive receptors. As it happens, these paraganglia are richly invested with mast cells (Goehler et al., 1999) which express all three pyrogenic cytokines and store them preformed within their granules (Gordon et al., 1990). To

68

C.M. Blatteis, S. Li / Autonomic Neuroscience: Basic and Clinical 85 (2000) 66 – 71

test the hypothesis, therefore, that mast cells may be the source of, e.g. the IL-1b that presumptively binds to vagal receptors, we pretreated conscious guinea pigs with pyrogen-free saline or disodium cromoglycate, a substance that prevents the degranulation of mast cells, 30 min before the i.v. administration of LPS and followed the course of their febrile response. We found that the early phase of the animals’ characteristic biphasic febrile response to i.v. LPS was not affected by this treatment, but the late phase was essentially abrogated (unpublished results). These data would suggest that mast cells may indeed contribute a factor to fever production, but, contrary to what we had anticipated, it would seem to be involved in the second phase, i.e. in the continued maintenance rather than in the initiation of the febrile response. The responsible factor and the role it may play remain to be clarified, but, if it were indeed a cytokine, these results would also discount cytokines as participants in the first (early) body core temperature (T c ) rise after i.v. LPS administration; i.e. the pyrogenic stimulus to vagal afferents would seem to be a factor other than a cytokine.

3.2. PGE2 Such an alternative candidate could be PGE 2 . It is synthesized by all macrophages, including Kupffer cells and peritoneal macrophages, in response to LPS, its level rises quickly in plasma after an LPS challenge, and its receptors are widely distributed on sensory neurons (Blatteis and Sehic, 1997; Coceani and Akarsu, 1998; Rivest, 1999; Zhang and Rivest, 1999), including hepatic and abdominal vagal afferents (Ek et al., 1998). It could function, therefore, as the direct activator of these neurons. Despite this, however, LPS is actually a weak trigger of arachidonic acid (AA) release; the free AA concentration is rate-limiting in PGE 2 synthesis. Thus, the activation by LPS of group IV cytosolic phospholipase A 2 (cPLA 2 , the isoform of the enzyme that initiates the cascade of events leading to the production of PGE 2 by macrophages from membrane phospholipids) is significantly delayed in vitro (Kramer, 1994; Watson et al., 1994) as compared to the prompt elevation of both preoptic and blood PGE 2 in vivo following the systemic administration of LPS (Rotondo et al., 1988; Sehic et al., 1996). In fact, LPS stimulates the increased production of cPLA 2 after some hours by inducing post-transcriptional modifications and de novo synthesis (Dennis et al., 1991; Angel et al., 1994; Kramer, 1994). Moreover, the increased synthesis of PGE 2 by LPS-stimulated macrophages is entirely caused by the selective expression of the inducible enzyme cyclooxygenase (COX)-2, the transcription and translation of which require at least 1 h (Lacroix and Rivest, 1998). LPSinduced fevers do not develop in COX-2 null mutant mice (Li et al., 1999b; Li et al., 2000a); COX-1, the constitutive isoform of COX, is not activated by LPS. It would seem improbable, therefore, that the signals that rapidly initiate

the febrile response to an i.v. injection of LPS were provided by PGE 2 generated, be it by macrophages or by endothelial cells of the cerebral microvasculature, via the intermediation of COX-2. Indeed, we found recently that pharmacological blockade of COX-2 (with mimesulide in guinea pigs) tends to attenuate the second (late) phase of the febrile response to i.v. LPS more than the first phase (Steiner et al., 2000). Again, the time frames fit, given the demonstrated, delayed expression of COX-2 in phagocytic and / or endothelial cells after i.v. LPS challenge as compared to the onset latency of the fever that this challenge induces. By contrast, low-dose i.v. capsaicin attenuates the ´ first febrile rise while leaving the second intact (Szekely et al., 2000), further suggesting an initial neural activation of the febrile response. These data, however, do not eliminate PGE 2 as the first stimulus of peripheral vagal afferents. But, since COX-1 knockout mice develop fever normally after LPS, the data infer that the PGE 2 so quickly detected in blood and brain after i.v. LPS is derived via COX-1 catalysis from phagocytic (or non-phagocytic) cells stimulated by a factor other than LPS but elaborated in almost immediate reaction to its presence. We have postulated that this quickly-acting, readily evocable agonist may be a component of the complement (C) system.

4. Complement as a possible signal intermediator The intravascular C cascade is activated within seconds by i.v. LPS via both the classic (the lipid A moiety of LPS) and alternative (the core oligosaccharide) pathways (Vukajlovich, 1992), and macrophages, e.g. Kupffer cells, express the receptors for various C-derived components (Hinglais et al., 1989). Kupffer cells very quickly (2 min) release PGE 2 in response to the anaphylatoxic fragments, C3a and C5a, and also to the membrane attack complex ¨ (MAC, C5b-9) (Puschel et al., 1993), whereas C depletion limits this release (Fink et al., 1989). The addition in vitro of C3a and C5a also triggers the production of cytokines by macrophages (Cavaillon and Haeffner-Cavaillon, 1990). The rates of their synthesis and secretion are not any more rapid, however, than after LPS. Therefore, the neuroactive substance evoked by C is unlikely to be a cytokine. On the other hand, these anaphylatoxins also stimulate the degranulation of mast cells, resulting in the release of their stored cytokines (Erdei and Pecht, 1996), as already considered. We examined, therefore, whether C may be critically involved in the rapid onset of the febrile response to i.v. LPS. To test this hypothesis, we hypocomplemented conscious guinea pigs by pretreatment with cobra venom factor (CVF). CVF causes the continuous, uncontrolled activation of the alternative pathway of C, so that the generation of all components subsequent to C3 is dosedependently reduced because of the gradual depletion of the substrate from which they are produced (Cochrane et

C.M. Blatteis, S. Li / Autonomic Neuroscience: Basic and Clinical 85 (2000) 66 – 71

al., 1970). This leads to proportionately long-lasting (days) hypocomplementemia (Li et al., 1999a). Because the dynamics of T c rises after i.v.- and i.p.-administered LPS are different, at least in guinea pigs (Blatteis et al., 2000) and because other data suggested that the fever caused by LPS injected by routes other than i.v. may be more vulnerable to the antipyretic effect of vagotomy than that caused by i.v. LPS (Maier et al., 1998), we investigated whether the C system may be involved in the febrile response to both i.v. and i.p. LPS. We found, unexpectedly, that the magnitude and course of the febrile response to i.v. LPS were not demonstrably affected by C reduction whereas, on the other hand, the fevers caused by i.p. LPS were attenuated in direct correlation (r50.614) with the amount of C reduction (Li et al., 1999a). The reason for this differential susceptibility to depression by i.v.- and i.p.-administered LPS was not evident from the data. We speculate that it may be due to different functional and biochemical properties of peritoneal and hepatic macrophages (Fox et al., 1987). There is also evidence that the activation of macrophages by LPS for synthetic responses may proceed by several pathways (Fenton and Golenbock, 1998; Cavaillon and Haeffner-Cavaillon, 1990). These possible explanations are under study. It would appear from our data, therefore, that C may be required to activate peritoneal macrophages in response to LPS. In support, we found (Li et al., 1999a) that LPS delivered i.p. to conscious guinea pigs at 8 mg / kg caused per se a 40% reduction in serum C activity within 30 min of its administration, in conjunction with the initial rise of T c . This would indicate that C is activated and consumed early after a pyrogenic LPS load and, thus, may indeed contribute to the development of the febrile response (C per se injected i.v. causes a transient fall in T c ; it is, thus, not the pyrogenic factor that directly stimulates the vagal receptors). In further support, we found very recently (Li et al., 2000c) that congenitally C5-deficient mice fail to develop fever in response to i.p.-injected LPS, therefore implicating this C fragment as a factor in the febrile response to LPS. Interestingly, moreover, these animals, in contrast to our decomplemented guinea pigs, also did not fever following i.v. LPS. C5a, its desArg form, and MAC potently stimulate ¨ PGE 2 synthesis (Hansch et al., 1984). Pretreatment with CVF prevents the rise in plasma (Fink et al., 1989) and preoptic (Sehic et al., 1996) PGE 2 induced by LPS. Furthermore, the addition of indomethacin results in detectable IL-1b activity in supernatants of macrophage cultures stimulated with suboptimal amounts of C3a desArg (Haeffner-Cavaillon et al., 1987), whereas PGE 2 inhibits the LPS-induced synthesis of IL-1b and TNFa in Kupffer cells (Karck et al., 1988), inferring again that the cytokines may not account directly for the rapid febrile response to i.v. LPS. PGE 2 , under these conditions, could be generated indirectly through the hydrolysis of membrane associated phosphoinositide (PI, which has a high

69

arachydonoyl chain content) by PI-specific phospholiphase C (PI-PLC) (Blatteis and Sehic, 1997); indeed, AA liberation by PI-PLC is 10-fold more rapid (within seconds) than that mediated by cPLA 2 . PI-PLC is activated by C, but not by LPS or IL-1b. Moreover, the subsequent conversion of this AA to PGE 2 is catalyzed by COX-1. Hence, the initial, peripheral fever trigger could indeed be PGE 2 released by macrophages stimulated by LPS-activated C components and binding to EP3 receptors on sensory afferent nerves (EP3 knockouts do not develop fever after i.v. LPS (Ushikubi et al., 1998). Unfortunately, there is a fly in this ointment also: if COX-1-mediated PGE 2 were this signaling molecule, COX-1 knockouts should not be able to fever, and yet they do! Hence, the question arises does the PGE 2 thus liberated participate in the genesis of the febrile response to peripheral LPS at all? And, if it does not, what then so rapidly triggers the febrile T c rise? It is possible, therefore, that other, as yet unknown mediators may be secreted by macrophages or by other, also still unidentified, cell types responsive to C5.

5. Is C an intermediator for all pyrogens? A question that arises from the preceding results is whether the mediatory role of C is specifically limited to the febrile response to LPS or is found generally in fevers caused by all exogenous pyrogens. To answer this question, we replicated our LPS studies, injecting i.v. and i.p. into conscious guinea pigs muramyl dipeptide (a synthetic gram-positive bacterial cell wall analog) and polyriboinosinic–polyribocytidylic acid (a synthetic viral double-stranded RNA analog), i.e. factors that reportedly induce fever also through a cytokine- and PGE 2 -mediated process similar to that of LPS. We found that the fevers caused by both routes by both these agents were not altered by CVF-induced hypocomplementemia (Li et al., 2000b). Thus, the mediatory role of C appears to be specifically limited to the febrile response to i.p. LPS in guinea pigs and i.v. and i.p. LPS in mice.

6. Summary and conclusions The results reported are compatible with the notion that pyrogenic doses of LPS rapidly activate the C cascade and that at least one of the fragments generated, C5, contributes importantly to the initiation of the febrile response. We speculate that it causes the quick release of a mediator capable of stimulating local, primary sensory nerves that convey the pyrogenic message to the POA. The exact cell source(s) and specific nature of the mediator(s) thus released that presumably activate(s) vagal afferents present in the vicinity remain to be elucidated. Because of the time constraints of its synthesis after signaling, this mediator at the beginning is probably not IL-1b, although its receptors

70

C.M. Blatteis, S. Li / Autonomic Neuroscience: Basic and Clinical 85 (2000) 66 – 71

evidently exist within abdominal vagal paraganglia. The possibility that it may be PGE 2 seems also unlikely because its synthesis from AA via COX-2 similarly lags the latency of fever onset after i.v. LPS. The data also indicate that the signals that initiate fever differ according to the class of exogenous pyrogen (LPS, muramyl dipeptide, etc.) and, by extension, to the original microorganisms from which these compounds are derived. The pyrogen dose, its route of administration, and the species of the host also influence the response. Many other data have shown that the time of day, the ambient temperature, the age and gender of the challenged host, its nutritional and hydrational status, and many other factors also affect febrile responsiveness. Hence, it is probable that multiple, varied, and complex regulatory mechanisms are involved in the immunomodulation of fever of different etiologies. Indeed, infectious stimuli are diverse and affect the body through a variety of portals. It may be anticipated, therefore, that different profiles of mediators and, hence, also different afferent pyrogenic pathways may become activated. It is also probable that the mechanisms that initiate and those that sustain the febrile response to exogenous pyrogens are different. From our current perspective, we hypothesize that the first response is neurally and the second humorally mediated when the pyrogenic stimulus is i.v. LPS. Consequently, generalization from particular findings should be made only with great caution. Much more study is needed to resolve the many questions that remain.

Acknowledgements The authors’ studies included herein were supported, in part, by NIH grant NS 34857.

References Angel, J., Berenbaum, F., LeDenmat, C., Nevalainer, T., Masliah, J., Fournier, C., 1994. Interleukin-1-induced prostaglandin E 2 biosynthesis in human synovial cells involves the activation of cytosolic phospholipase A 2 and cyclooxygenase-2. Eur. J. Biochem. 226, 125– 131. Banks, W.A., Kastin, A.J., Broadwell, R.D., 1996. Passage of cytokines across the blood–brain barrier. Neuroimmunomodulation 2, 241–248. Berthoud, H.-R., 2000. The anatomical basis of vagal involvement in fever and thermoregulation, this volume. Blatteis, C.M., Bealer, S.L., Hunter, W.S., Llanos-Q, J., Ahokas, A.A., Mashburn, Jr. T.A., 1983. Suppression of fever after lesions of the anteroventral third ventricle of guinea pigs. Brain Res. Bull. 11, 519–526. Blatteis, C.M., Dinarello, C.A., Shibata, M., Llanos-Q, J., Quan, N., Busija, D.-W., 1989. Does interleukin-1 enter the brain. In: Mercer, J.B. (Ed.), Thermal Physiology. Elsevier, Amsterdam, pp. 385–390. Blatteis, C.M., Sehic, E., 1997. Prostaglandin E 2 : a putative fever mediator. In: Mackowiak, P.A. (Ed.), Fever: Basic Mechanisms and Management, 2nd Edition. Lippincott-Raven, Philadelphia, pp. 117– 145.

Blatteis, C.M., Sehic, E., Li, S., 2000. Complement and the pathogenesis of endotoxic fever. Int. J. Biometerol. 43, 176–183. Cartmell, T., Poole, S., Turnbull, A.V., Rothwell, N.J., Luheshi, G.N., 2000. Circulating interleukin-6 mediates the febrile response to localized inflammation in rats. J. Physiol. (Lond.) 526, 653–661. Cavaillon, J.-M., Haeffner-Cavaillon, N., 1990. Signals involved in interleukin-1 synthesis and release by lipopolysaccharide-stimulated monocytes / macrophages. Cytokine 2, 313–329. Chai, Z., Gatti, S., Toniatti, C., Poli, V., Bartfai, T., 1996. Interleukin (IL)-6 gene expression in the central nervous system is necessary for fever response to lipopolysaccharide or IL-1b: a study on IL-6deficient mice. J. Exp. Med. 183, 311–316. Coceani, F., Akarsu, E.S., 1998. Prostaglandin E 2 in the pathogenesis of fever: an update. Ann. NY Acad. Sci. 856, 76–82. ¨ Cochrane, C.G., Muller-Eberhard, H.J., Aikin, B.S., 1970. Depletion of plasma complement in vivo by a protein of cobra venom: its effect on various immunologic reactions. J. Immunol. 105, 55–69. Dennis, E.A., Rhee, D.G., Billah, M.M., Hannun, Y.F., 1991. Role of phospholipases in generating lipid second messenger in signal transduction. FASEB J. 5, 2068–2077. Ek, M., Kurosowa, M., Lundeberg, T., Ericsson, A., 1998. Activation of vagal afferents after intravenous injection of interleukin-1b: role of endogenous prostaglandins. J. Neurosci. 18, 9471–9479. Elmquist, J.K., Scammell, T.E., Jacobson, C.C., Saper, C.B., 1996. Distribution of Fos-like immunoreactivity in the rat brain following intravenous lipopolysaccharide administration. J. Comp. Neurol. 371, 85–103. Erdei, A., Pecht, I., 1996. Complement peptides and mast cell triggering. Immunol. Lett. 54, 109–112. Fenton, M.J., Golenbock, D.T., 1998. LPS-binding proteins and receptors. J. Leukoc. Biol. 64, 25–32. Fink, M.P., Rothchild, H.R., Deniz, Y.F., Cohn, S.M., 1989. Complement depletion by Naja naja cobra venom factor limits prostaglandin release and improves visceral perfusion in porcine endotoxin shock. J. Trauma 29, 1076–1085. Fox, E.S., Thomas, P., Broitman, S.A., 1987. Comparative studies of endotoxin uptake by isolated rat Kupffer and peritoneal cells. Infect. Immun. 55, 2962–2966. Givalois, L., Dornand, J., Makaouche, M., Solier, M.D., Bristow, A.F., Ixart, G., Siaud, P., Assenmacher, I., Barbanel, G., 1994. Temporal cascade of plasma level surges in ACTH, corticosterone, and cytokines in endotoxin-challenged rats. Am. J. Physiol. 267, R164–R170. Goehler, L.E., 2000. Vagal afferent signaling in fever: molecular mechanisms, this volume. Goehler, L.E., Gaykema, R.P.A., Nguyen, K.T., Lee, J.E., Tilders, F.J.H., Maier, S.F., Watkins, L.R., 1999. Interleukin-1b in immune cells of the abdominal vagus nerve: a link between the immune and nervous systems? J. Neurosci. 19, 2799–2806. Goehler, L.E., Relton, J.K., Dripps, D., Kiechle, R., Tartaglia, N., Maier, S.F., Watkins, L.R., 1997. Vagal paraganglia bind biotinylated interleukin-1 receptor antagonist: a possible mechanism for immuneto-brain communication. Brain Res. Bull. 43, 357–364. Goldbach, J.-M., Roth, J., Zeisberger, E., 1997. Fever suppression by subdiaphragmatic vagatomy in guinea pigs depends on the route of pyrogen administration. Am. J. Physiol. 272, R675–R681. Gordon, J.R., Burd, P.R., Galli, S.J., 1990. Mast cells as a source of multifunctional cytokines. Immunol. Today 11, 458–464. Haeffner-Cavaillon, N., Cavaillon, J.-M., Laude, M., Kazatchine, M.D., 1987. C3a (C3a desArg) induces production and release of interleukin-1 by cultured human monocytes. J. Immunol. 139, 794– 799. ¨ Hansch, G.M., Seitz, M., Martinotti, G., Betz, M., Rautenberg, E.W., Gemsa, D., 1984. Macrophages release arachidonic acid, prostaglandin E 2 , and thromboxane in response to late complement components. J. Immunol. 133, 2145–2150. Hansen, M.K., Daniels, S., Goehler, L.E., Gaykema, R.P.A., Maier, S.F., Watkins, L.R., 2000. Subdiaphragmatic vagotomy does not block intraperitoneal LPS-induced fever. FASEB J. 14, A85.

C.M. Blatteis, S. Li / Autonomic Neuroscience: Basic and Clinical 85 (2000) 66 – 71 Hinglais, N., Kazatchkine, M.D., Mandet, C., Appay, M.-D., Bariety, J., 1989. Human liver Kupffer cells express CR1, CR3, and CR4 complement receptor antigens. An immunohistochemical study. Lab. Invest. 61, 509–514. Jansky, L., Vybiral, S., Pospisilova, D., Roth, J., Dornand, J., Zeisberger, E., 1995. Production of systemic and hypothalamic cytokines during the early phase of endotoxin fever. Neuroendocrinology 62, 55–61. Karck, U., Peters, T., Decker, K., 1988. The release of tumor necrosis factor from endotoxin-stimulated rat Kupffer cells is regulated by prostaglandin E 2 and dexamethasone. J. Hepatol. 7, 352–361. Kluger, M.J., Kozak, W., Leon, L.R., Soszynski, D., Conn, C.A., 1995. Cytokines in fever. Neuroimmunomodulation 2, 216–223. Komaki, G., Arimura, A., Koves, K., 1992. Effect of intravenous injection of IL-1b on PGE 2 levels in several brain areas as determined by microdialysis a. Am. J. Physiol. 262, E246–E251. Kramer, R.M., 1994. Structures and function of Ca 21 -sensitive cytosolic phospholipase A 2 . In: Liscovitch, M. (Ed.), Signal-Activated Phospholipases. R.G. Landes, Austin, pp. 13–40. Lacroix, S., Rivest, S., 1998. Effect of acute systemic inflammatory response and cytokines on the transcription of the genes encoding cyclooxygenase enzymes (COX-1 and COX-2) in the rat brain. J. Neurochem. 70, 452–466. Li, S., Ballou, L.R., Morham, S.G., Blatteis, C.M., 2000a. Cyclooxygenase (COX)-2 mediates the febrile response of mice to interleukin1b. Brain Res. in press. Li, S., Sehic, E., Ungar, A.L., Blatteis, C.M., 2000b. Complement does not mediate the febrile responses of guinea pigs to muramyl dipeptide and polyriboinosinic–polyribocytidylic acid. J. Thermal. Biol. 25, 51–58. Li, S., Holers, V.M., Boackle, S.A., Blatteis, C.M., 2000c. Complement fragment 5 (C5) modulates the febrile response of mice to intraperitoneally (ip) administered endotoxin. FASEB J. 14, A86. Li, S., Sehic, E., Wang, Y., Ungar, A.L., Blatteis, C.M., 1999a. Relation between complement and the febrile response of guinea pigs to systemic endotoxin. Am. J. Physiol. 277, R1635–R1645. Li, S., Wang, Y., Matsumura, K., Ballou, L.R., Morham, S.G., Blatteis, C.M., 1999b. The febrile response to lipopolysaccharide is blocked in COX-2 2 / 2 but not in COX-1 2 / 2 mice. Brain Res. 825, 86–94. Luheshi, G.N., 1998. Cytokines and fever. Mechanisms and sites of action. Ann. NY Acad. Sci. 856, 83–89. Maier, S.F., Goehler, L.E., Fleshner, M., Watkins, L.R., 1998. The role of the vagus nerve in cytokine-to-brain communication. Ann. NY Acad. Sci. 840, 289–300. Mathison, J.C., Ulevitch, R.J., 1979. The clearance, tissue distribution, and cellular localization of intravenously injected lipopolysaccharide. J. Immunol. 123, 2133–2143. Matsumura, K., Cao, C., Watanabe, Y., 1998. Prostaglandin system in the brain: sites of biosynthesis and sites of action under normal and hyperthermic states. Prog. Brain Res. 115, 275–295. Newton, R.C., 1986. Human monocyte production of interleukin-1: parameters of the induction of interleukin-1 secretion by lipopolysaccharides. J. Leukoc. Biol. 39, 299–311. ¨ Puschel, G.P., Hespeling, U., Opperman, M., Dieter, P., 1993. Increase in

71

prostanoid formation in rat liver macrophages (Kupffer cells) by human anaphylatoxin C3a. Hepatology 18, 1516–1521. Quan, N., Whiteside, M., Herkenham, M., 1998. Cyclooxygenase-2 mRNA expression in rat brain after peripheral injection of lipopolysaccharide. Brain Res. 802, 189–197. Rivest, S., 1999. What is the cellular source of prostaglandins in the brain in response to systemic inflammation? Facts and controversies. Mol. Psychiat. 4, 501–507. Romanovsky, A.A., 2000. Multiple roles of the vagus in fever: lessons from vagotomy, this volume. ¨ B., Voigt, K., Zeisberger, E., Ross, G., Goldbach, J.-M., Roth, J., Storr, 1999. Local induction of cytokines in a subcutaneous airpouch or artificial subcutaneous chambers and fever: activation of humoral and ¨ nerval signals to the brain. Pflugers Arch. 437 (Suppl.), R133. ¨ B., Zeisberger, E., 1998. Neutralization of Roth, J., Markin, D., Storr, pyrogen-induced tumor necrosis factor by its type 1 soluble receptor in guinea pigs: effects on fever and interleukin-6 release. J. Physiol. (Lond.) 509, 267–275. Roth, J., McClellan, J.L., Kluger, M.J., Zeisberger, E., 1994. Attenuation of fever and release of cytokines after repeated injections of lipopolysaccharide in guinea-pigs. J. Physiol. (Lond.) 477, 177–185. Rothwell, N.J., 1997. Neuroimmune interactions: the role of cytokines. Br. J. Pharmacol. 121, 841–847. Rotondo, D., Abul, H.T., Milton, A.S., Davidson, J., 1988. Pyrogenic immunomodulators increase the level of prostaglandin E 2 in the blood simultaneously with the onset of fever. Eur. J. Pharmacol. 154, 145–152. Schiltz, J.C., Sawchenko, P.E., 1998. Interleukin-1 induction of cyclooxygenase-2 expression in ED2-positive perivascular microglia. Soc. Neurosci. Abstr. 24, 1615. ´ Sehic, E., Szekely, M., Ungar, A.L., Oladehin, A., Blatteis, C.M., 1996. Hypothalamic prostaglandin E 2 during lipopolysaccharide-induced fever in guinea pigs. Brain Res. Bull. 39, 391–399. Steiner, A.A., Llanos-Q, J., Blatteis, C.M., 2000. Cyclooxygenase-2 is an essential mediator of LPS fever in guinea pigs. FASEB J. 14, A86. ´ Szekely, M., 2000. The vagus nerve in thermoregulation and energy metabolism. Auton. Neurosci. 85, 26–38. Ushikubi, F., Segi, E., Sugimoto, Y., Murata, T., Matsuoka, T., Kobayashi, T., Hizaki, H., Tuboi, K., Katsuyama, M., Ichikawa, A., Tanaka, T., Yoshida, N., Narumiya, S., 1998. Impaired febrile response in mice lacking the prostaglandin E receptor subtype EP3 . Nature 395, 281–284. Vukajlovich, S.W., 1992. Interactions of LPS with serum complement. In: Ryan, J.L., Morrison, J.M. (Eds.). Bacterial Endotoxic Lipopolysaccharides, vol. II. CRC, Boca Raton, pp. 213–235. Watson, R.W.G., Redmond, H.P., Bouchier-Hayes, D., 1994. Role of endotoxin in mononuclear phagocyte-mediated inflammatory responses. J. Leukoc. Biol. 56, 95–103. Zhang, J., Rivest, S., 1999. Distribution, regulation, and colocalization of the genes encoding the EP3 - and EP4 -PGE 2 receptors in the rat brain and neuronal responses to systemic inflammation. Eur. J. Neurosci. 11, 2651–2668.