Bacterial lipopolysaccharide (LPS) induces fever that is mediated by pyrogenic cytokines such as interleukin (IL)-1β. We hypothesized that the anti-inflammatory cytokine IL-10 modulates the febrile response to LPS by suppressing the production of pyrogenic cytokines. In rats, intravenous but not intracerebroventricular infusion of IL-10 was found to attenuate fever induced by peripheral administration of LPS (10 μg/kg iv). IL-10 also suppressed LPS-induced IL-1β production in peripheral tissues and in the brain stem. In contrast, central administration of IL-10 attenuated the febrile response to central LPS (60 ng/rat icv) and decreased IL-1β production in the hypothalamus and brain stem but not in peripheral tissues and plasma. Furthermore, intravenous LPS upregulated expression of IL-10 receptor (IL-10R1) mRNA in the liver, whereas intracerebroventricular LPS enhanced IL-10R1 mRNA in the hypothalamus. We conclude that IL-10 modulates the febrile response by acting in the periphery or in the brain dependent on the primary site of inflammation and that its mechanism of action most likely involves inhibition of local IL-1β production.
during systemic inflammation, various brain-mediated responses occur as part of the acute phase reaction, including fever, sickness behavior, and activation of the hypothalamus-pituitary-adrenal (HPA) axis. Fever is a regulated rise in body temperature that is due to a change in the thermoregulatory setpoint (25). It is widely accepted that fever is controlled by thermosensitive neurons in the preoptic area of the anterior hypothalamus (POA). The febrile response to systemic inflammation involves the activation of blood monocytes and/or hepatic macrophages (Kupffer cells) by exogenous pyrogens, e.g., bacteria-derived lipopolysaccharide (LPS) or other endotoxins, which results in the production and release of cytokines such as interleukin-1β (IL-1β), IL-6, and tumor necrosis factor-α (TNF-α). These cytokines then act as endogenous pyrogens and signal, directly or indirectly, thermoregulatory neurons in the brain to initiate fever (reviewed in Ref. 25).
Cytokines produced outside the central nervous system (CNS) signal the brain by multiple routes, including humoral and neural pathways (53). Blood-borne cytokines may activate meningeal macrophages, cerebral endothelial cells, and perivascular microglial cells, thereby eliciting the local production of secondary mediators such as prostaglandin E2 and IL-6, which are implicated in fever (11, 17, 57). Circulating cytokines may also act on cells in circumventricular organs (CVOs) such as the organum vasculosum of the lamina terminalis and the area postrema (AP), which lack a functional blood-brain barrier (4, 48). The neurons located close to these CVOs innervate regions in the hypothalamus involved in thermoregulation. In addition, a role for the vagus nerve has been proposed in transmitting inflammatory signals from the periphery to the brain (21, 22, 42, 49, 61). Peripherally produced cytokines may activate vagal afferent nerves that innervate the nucleus of the solitary tract in the brain stem, from which catecholaminergic projections lead to the hypothalamus (5, 18,50).
Because central administration of pyrogenic cytokines elicits fever in doses lower than those required after peripheral administration (12, 30, 31, 45), they are suggested to act within the brain. Indeed, irrespective of the exact signaling route, pyrogenic cytokines are produced in the brain following systemic LPS exposure. For instance, IL-1β, TNF-α, and IL-6 mRNA and protein are expressed in microglial cells and/or neurons in the brain after peripheral administration of LPS (8, 9, 20, 28, 53, 54, 56). Furthermore, blocking the action of brain-derived IL-1β or IL-6 by central administration of neutralizing antibodies to these cytokines (24, 46) or the IL-1 receptor antagonist (33) attenuates the febrile response to peripheral LPS, suggesting that these brain-derived cytokines play an instrumental role in the febrile response.
Production of pyrogenic cytokines by both peripheral macrophages and glial cells can be inhibited by IL-10 (29, 37, 58). Recent studies further show that IL-10 inhibits fever induced by systemic LPS administration in rats and mice (32, 36). Moreover, IL-10 knockout mice show exacerbated and prolonged febrile responses to LPS (32) and elevated TNF and IL-6 levels in the brain (1). These observations led us to hypothesize that IL-10, by acting on its receptor, may play a role in the control of fever by modulating cytokine production in peripheral organs and/or the brain.
The present study was designed to examine the role and site of action of IL-10 in the febrile response in rats. To elucidate target sites of IL-10, we compared the effects of central (intracerebroventricular) or peripheral (intravenous) administration of recombinant rat IL-10 (rrIL-10) on the febrile response to peripheral injection of LPS. In addition, we evaluated the effect of central administration of rrIL-10 on fever induced by central administration of LPS. To test the hypothesis that the effects of IL-10 on the febrile response involve modulation of the production of pyrogenic cytokines, we assessed the concentrations of IL-1β, TNF-α, and IL-6 in plasma, peripheral tissues (liver, spleen, pituitary), and in the brain (hypothalamus and brain stem). To investigate site-specific expression and LPS-induced regulation of IL-10 receptors, we used quantitative RT-PCR to study mRNA expression of IL-10 receptor (IL-10R1; the ligand-binding chain) in the liver and hypothalamus. The present results show that IL-10 is a site-specific modulator of fever and can act either in the periphery and the brain by inhibiting local cytokine production.
MATERIALS AND METHODS
Male Wistar rats (Harlan CPB, Horst, The Netherlands) were housed two per cage under controlled temperature (20–22°C) and light-dark conditions (lights on 7 AM, lights off 7 PM). Food and water were available ad libitum. At the time of the experiments, the animals had body weights of 250–300 g. To minimize stress-induced hyperthermia, rats were habituated to experimental procedures by noninvasive handling twice daily for 3 consecutive days before the experiments. The experimental protocol had been approved by the Institutional Committee for Animal Health and Care.
LPS (Escherichia coli serotype 0128-B12; Sigma, St. Louis, MO) was dissolved in sterile water, divided in aliquots, and stored at −20°C. Before administration, appropiate dilutions were made in sterile, pyrogen-free saline (Braun, Melsungen, Germany). rrIL-10 (National Institute for Biological Standards and Control, Potters Bar, UK) (2) was diluted in a vehicle of sterile, pyrogen-free saline containing 0.1% (wt/vol) BSA (fatty acid-free, low endotoxin; Sigma).
Rats were anesthesized with ketamine (50 mg/kg im; Kombivet, Etten-Leur, The Netherlands) and xylazine (Rompun, 5 mg/kg ip; Bayer, Leverkusen, Germany), and a permanent cannula was implanted into the right jugular vein according to Steffens (52) for continuous intravenous infusion and/or repeated blood sampling. In other groups of animals, a permanent 22-gauge guide cannula with an indwelling dummy cannula (Plastics One, Roanoke, VA) was implanted into the right lateral ventricle of the brain at stereotaxic coordinates: 1.5 mm lateral to the midline, 0.6 mm posterior to bregma, and 3.7 mm under the brain surface. Simultaneously, a battery-operated, temperature-sensitive radiotransmitter (Data Sciences, St. Paul, MN) was implanted into the peritoneal cavity of all rats. After surgery, rats were individually housed in macrolon cages (20 × 30 × 40 cm) and were allowed to recover for at least 6 days.
In a pilot experiment in which serial blood samples were taken from a chronically implanted jugular vein cannula in rats that were given rrIL-10 intravenously, plasma levels of IL-10 showed a first-order decline with a half-life of plasma IL-10 concentrations of ∼6 min. Therefore, we decided to deliver IL-10 by continuous infusion to reach steady-state IL-10 concentrations rather than by bolus injection.
All experiments started between 8 and 10 AM. In rats with a cannula in the lateral ventricle, the dummy cannnula was replaced by a 28-gauge stainless steel injector (Plastics One) just before the experiment. Because pilot experiments revealed no differences in temperature responses between vehicle/saline-treated rats and IL-10/saline-treated rats (data not shown), the latter were used as control groups in further experiments.
In experiment 1, groups of rats (total n = 8–9) were given vehicle or rrIL-10 (4 μg · h−1 per rat) via continuous intravenous infusion to freely moving rats at a flow rate of 5 μl/min using a microsyringe infusion pump (Harvard Apparatus, South Natick, MA). Thirty minutes after start of the infusion, LPS (10 μg/kg, 500 μl/rat) was administered to the rats by intravenous injection via a lateral tail vein. The control group (n = 3) was given the same dose of IL-10, followed by saline (500 μl/rat iv). Animals from each LPS-treated group (n = 5–6) and the entire control group were killed by decapitation 5 h after the administration of LPS to determine cytokine levels in plasma and tissues. Trunk blood was collected in cold heparin-coated tubes (Sarstedt, Etten-Leur, The Netherlands) and subsequently centrifuged (4,000 g, 15 min, 4°C). Aliquots of plasma were stored at −20°C until assayed. Parts of liver and spleen, the pituitary, the hypothalamus region (containing the POA), and the dorsal part of the brain stem (containing the AP) were dissected immediately, frozen in liquid nitrogen, and stored at −80°C until processing.
In experiment 2, groups of rats (n = 4) were given vehicle or rrIL-10 (100, 300, or 600 ng · h−1 per rat) via continuous infusion to the lateral ventricle (flow rate 5 μl/h) in freely moving rats. LPS (10 μg/kg) or saline (500 μl/rat) was injected intravenously in rats via a lateral tail vein 15 min after start of the intracerebroventricular infusion with IL-10 or vehicle.
In experiment 3, groups of rats (n = 10–11) were given LPS (60 ng/rat) or saline (5 μl/rat during 10 min) into the lateral ventricle in freely moving rats, followed by continuous intracerebroventricular infusion of vehicle or rrIL-10 (600 ng · h−1 per rat) at a rate of 5 μl/h. Animals from each LPS-treated group (n = 3–4) were killed by decapitation 3.5 or 6 h after the administration of LPS to determine cytokine levels in plasma and tissues. Trunk blood and tissues were collected as described (experiment 1).
In experiments 1–3, the body temperature was monitored by telemetry every 15 min for 6–8 h following LPS injection. The output signals from each of the intraperitoneally implanted transmitters (frequency in Hz) were monitored by a receiver (Data Sciences) placed under each animal's cage, channelled into a consolidation matrix (BCM 100) connected to a PC, and converted to degrees Celsius (°C) by the processor.
In experiment 4, the expression of endogenous IL-10R1 mRNA was studied. Groups of rats (n = 3–4) were untreated (control group) or injected with LPS (100 μg/kg iv) via a lateral tail vein and killed by decapitation after 3, 8, and 24 h. Other groups of animals (n = 5) were injected with LPS (600 ng/rat icv) or saline (5 μl/rat) and killed by decapitation after 3 and 6 h. Liver and hypothalamus were dissected immediately, frozen in liquid nitrogen, and stored at −80°C until processing.
For cytokine assays, dissected brain regions, pituitary gland, and fragments of liver and spleen were homogenized for 5 min in 200–500 μl of Tris-buffered saline (pH 7.5) containing a protease inhibitor cocktail (Sigma) using a rotor-stator homogenizer (Heidolph Instruments, Schwabach, Germany). Homogenates were centrifuged (16,000 g, 15 min, 4°C), and the supernatants were used for the determination of cytokine levels. Before assay, all homogenates were diluted to a protein concentration of ∼1 mg/ml as determined by means of a Bradford assay (7).
For quantitative RT-PCR, tissue fragments were homogenized in 250 μl RNA lysis buffer containing 2% β-mercaptoethanol (Promega, Madison, WI) and used for isolation of total RNA.
IL-1β, IL-6, and TNF-α concentrations were measured in plasma and tissue homogenates using ELISAs specific for rat IL-1β (47), rat IL-6 (40), and rat TNF-α (41), respectively. Detection limits of the assays in plasma were 10, 16, and 40 pg/ml for IL-1β, IL-6, and TNF-α, respectively. Cytokine levels in homogenates are expressed as picograms per milligram protein (detection limits: 1.5 pg/mg protein for IL-1β and 16 pg/mg protein for IL-6 and TNF-α).
RNA Isolation and cDNA Synthesis
Total RNA was isolated from tissue homogenates using the SV Total RNA Isolation System (Promega) as described by the manufacturer. Concentration and purity of the RNA were determined by measuring the absorbance at 260 and 280 nm in a microtiter plate reader (ICN Biomedicals, Zoetermeer, The Netherlands). One microgram of RNA was reverse transcribed into cDNA using the Reverse Transcription System (Promega) with oligo-dT primers and AMV enzyme, according to the manufacturer's instructions. The RT reaction was carried out at 42°C for 45 min, followed by deactivation of the enzyme for 5 min at 99°C and 5 min at 4°C.
Real-Time Quantitative PCR
For quantitative PCR, the SYBR Green PCR Core reagents kit (Applied Biosystems, Foster City, CA) was used. Amplification of cDNA was performed in MicroAmp Optical 96-well Reaction Plates (Applied Biosystems) on an ABI PRISM 7700 Sequence Detection System (Applied Biosystems). The reaction mixture (20 μl) was composed of 1× SYBR Green buffer, 3 mM MgCl2, 875 μM dNTP mixed with dUTP, 0.3 U AmpliTaq Gold, 0.12 U Amperase UNG, 15 pmol of each primer, 12.5 ng cDNA and nuclease-free H2O. The primers for rat IL-10R1 were 5′-CTGGTCACCCTGCCATTGAT-3′ (forward) and 5′-AGGCATGGCTAAAATACAAAGAAAC-3′ (reverse) (60) and for rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) 5′-GAACATCATCCCTGCATCCA-3′ (forward) and 5′-CCAGTGAGCTTCCCG-TTCA-3′ (reverse). The reaction conditions were an initial 2 min at 50°C, followed by 10 min at 95°C and 40 cycles of 15 s at 95°C and 1 min at 59°C. The levels of IL-10R1 mRNA expression were quantified relatively to the level of the housekeeping gene GAPDH using the following calculation: 2−(thresholdcycleoftargetmRNA−thresholdcycleofGAPDHmRNA)× 100.
Body temperature data were analyzed by ANOVA for repeated measurements with treatment as the between-subject factor and time as the within-subject factor. Post hoc time-matched comparisons between groups were carried out by Fisher's least significant difference (LSD) multiple comparison test. ELISA and quantitative PCR data were analyzed by one- or two-way ANOVA, followed by Fisher's LSD test. If necessary, data were log-transformed before analysis. Parameters that could not be transformed to meet criteria of ANOVA were tested nonparametrically by the Mann-Whitney U-test. Correlations were analyzed using the Pearson correlation test. P values <0.05 were considered to indicate a significant difference. The statistical evaluation was carried out by using the NCSS 2000 statistical program.
Effects of intravenous IL-10 on intravenous LPS-induced fever.
Figure 1 shows the changes in body (core) temperature (Tc) in rats chronically infused with vehicle or rrIL-10 (4 μg · h−1 per rat iv) after intravenous injection with LPS or saline. Before LPS injection (−30 and −15 min), the body temperature was not different between the groups (basal value ∼37.4°C). The transient increase in Tc observed 15–30 min after LPS injection represents stress-induced hyperthermia induced by the intravenous administration of LPS. Subsequently, Tc of IL-10/saline-injected rats returned to basal values and remained ∼37°C. Tc from vehicle/LPS-injected rats tended to drop 60–90 min after LPS injection and subsequently shows a typical biphasic febrile response, a first phase peaking ∼2.5–3 h after LPS injection (rise in Tc ∼0.6°C compared with basal values) and a second phase peaking after 5–7 h (rise in Tc ∼1.3°C). Chronic infusion with IL-10 significantly inhibited the second hyperthermic phase but not the first hyperthermic phase. In addition, IL-10 resulted in higher Tc 75–105 min after LPS, compared with the vehicle/LPS group.
Effects of intravenous IL-10 on intravenous LPS-induced cytokine levels in peripheral and brain tissues.
At 5 h after injection of LPS, when Tc is rising, IL-1β protein was detected in plasma and various tissue homogenates of rats (Fig. 2). In plasma and tissues of IL-10/saline-treated rats, IL-1β was undetectable or detected at very low levels, except for liver and spleen. LPS markedly increased IL-1β protein levels in plasma compared with IL-10/saline-treated rats (∼63-fold). Chronic intravenous infusion with IL-10 significantly reduced plasma IL-1β levels 5 h after LPS administration. In the liver, spleen, and pituitary gland, LPS elevated IL-1β levels ∼6-, 4-, and 12-fold, respectively, and infusion with IL-10 reduced IL-1β levels in the liver and pituitary gland but not in the spleen.
In the brain, IL-1β levels were markedly increased by LPS in the brain stem (∼26-fold) but not in the hypothalamus. IL-10 infusion blocked the rise in brain stem IL-1β levels but had no effect on IL-1β levels in the hypothalamus.
IL-6 plasma levels in rats of the vehicle/LPS group were 76 ± 29 pg/ml, whereas levels in rats of the IL-10/saline and IL-10/LPS groups were below detection limit of the assay (15 pg/ml). TNF-α was only detectable in plasma of four of six rats in the vehicle/LPS group (108 ± 29 pg/ml), and levels in rats of the other groups were below detection limit of the assay (40 pg/ml). Thus IL-10 reduced both IL-6 and TNF-α levels in plasma (Mann-Whitney U-test,P < 0.01 and P < 0.05, respectively). In all tissue homogenates, TNF-α and IL-6 levels were below detection limits (16 pg/mg protein).
As illustrated in Fig. 6 A, cytokine concentrations in plasma and hypothalamus of individual rats were plotted against their Tc at time of death. Within both the vehicle/LPS and IL-10/LPS groups, positive correlations were found between IL-1β plasma levels and Tc 5 h after LPS injection (Pearson correlation coefficients 0.76 and 0.85, respectively). Pooled data from both groups showed a significant positive correlation (coefficient 0.86, P < 0.01). In addition, positive but not significant correlations were observed between Tc and plasma IL-6 and TNF-α levels in the vehicle/LPS group (coefficients 0.68 and 0.72, respectively). Thus, after peripheral administration of LPS, levels of IL-1β and possibly other cytokines in plasma, but not in the hypothalamus, are determinants of the hyperthermic response.
Effects of intracerebroventricular IL-10 on intravenous LPS-induced fever.
The febrile response to peripheral (intravenous) administration of LPS showed similar kinetics as described for experiment 1. Central administration of IL-10 by continuous intracerebroventricular infusion at doses of 100, 300, and 600 ng/rat throughout the experiment did not affect the febrile response induced by intravenous LPS injection. Data of experiments with the IL-10 doses of 100 and 300 ng/rat are not shown. Data of the experiment with the dose of 600 ng/rat are illustrated in Fig. 3.
Effects of intracerebroventricular IL-10 on intracerebroventricular LPS-induced fever.
Figure 4 shows the changes in Tc in rats given vehicle or rrIL-10 (600 ng/rat) by intracerebroventricular infusion throughout the experiment, after intracerebroventricular injection with LPS or saline. The febrile response to intracerebroventricular LPS consisted of a typical single hyperthermic phase (rise in Tc ∼1.5°C), with onset at ∼60 min and peaking around 3–4 h after LPS injection, whereas IL-10/saline-infused animals did not show any hyperthermia. The febrile response was significantly attenuated in the IL-10/LPS-treated group compared with the vehicle/LPS-treated group.
Effects of intracerebroventricular IL-10 on intracerebroventricular LPS-induced cytokine levels in peripheral and brain tissues.
IL-1β levels were detected in various tissue homogenates of rats killed 3.5 and 6 h after administration of LPS, at the peak and at the end of the febrile response, respectively (Fig.5). Plasma levels of IL-1β remained below detection limit of the assay (10 pg/ml). In the liver, spleen, and pituitary gland, no marked differences in IL-1β concentrations were observed between vehicle/LPS- and IL-10/LPS-infused rats. In the hypothalamus, IL-10 infusion significantly reduced IL-1β levels at 3.5 and 6 h after LPS administration by more than twofold. In the brain stem, IL-1β levels in IL-10/LPS-infused rats were markedly lower (∼6-fold) than those in vehicle/LPS-infused rats at 6 h after LPS injection.
No significant differences were observed in IL-6 and TNF-α levels in plasma and tissue homogenates of vehicle/LPS-treated rats compared with those in IL-10/LPS-treated rats (data not shown).
Figure 6 B shows the correlations between IL-1β concentrations in plasma and hypothalamus of individual rats with their body temperature at 3.5 h after central LPS administration. Within the vehicle/LPS and IL-10/LPS groups, positive correlations were found between IL-1β levels in the hypothalamus and Tc (correlation coefficients 0.95 and 0.77, respectively). Pooled data from both groups showed a significant positive correlation (coefficient 0.91, P < 0.05), demonstrating that expression levels of IL-1β in the hypothalamus, but not in plasma, are determinants of the hyperthermic response to central administration of LPS.
IL-10R1 mRNA expression in liver and hypothalamus.
To investigate LPS-induced changes in expression levels of IL-10R1 mRNA, rats were given LPS (100 μg/kg iv) for 3, 8, and 24 h, and expression was studied in liver and hypothalamus by quantitative RT-PCR. As illustrated in Fig.7 A, IL-10R1 mRNA expression in the liver was upregulated 3 h after peripheral administration of LPS and still elevated after 24 h, compared with control rats. In contrast, IL-10R1 mRNA levels in the hypothalamus were not affected by LPS treatment at any of the time intervals studied.
After central administration of LPS (600 ng/rat icv), IL-10R1 mRNA expression in the hypothalamus was significantly enhanced at 3 and 6 h after LPS, compared with control (saline injected) rats (Fig.7 B). IL-10R1 mRNA levels in the liver were not affected by central LPS administration.
In the present study, we demonstrate that IL-10 can attenuate the febrile response to LPS both by acting in the CNS and in peripheral targets, depending on the site of the primary inflammatory response. Thus peripheral administration of IL-10 markedly attenuated the febrile response to peripheral LPS and inhibited LPS-induced production of the endogenous pyrogen IL-1β in peripheral tissues and the brain stem but not in the hypothalamus. Conversely, central administration of IL-10 attenuated the hyperthermic response to central, but not to peripheral, LPS and inhibited IL-1β production in the hypothalamus and brain stem but not in peripheral tissues.
Our findings that intravenous infusion of IL-10 inhibits the febrile response of rats to intravenously given LPS are consistent with previous data of mice given LPS by an intraperitoneal route (32). The present data show that IL-10 prevents the tendency of the body temperature to drop shortly after intravenous LPS administration, which may be reminiscent of the hypothermic phase reported after higher doses of LPS (13, 42). Because the LPS-induced hypothermia depends on macrophage-derived factors, including cytokines and prostaglandins (13, 59), we speculate that IL-10 may prevent the initial tendency to hypothermia by suppressing cytokine levels, as seen in mice (32). The data further show that the second phase of the hyperthermic response is selectively attenuated by circulating IL-10. This cryogenic effect of IL-10 is correlated with decreased levels of circulating IL-1β, TNF-α, and IL-6 in plasma and supports the concept that suppression of circulating levels of IL-1β and other pyrogenic cytokines attenuates the late phase of the febrile response (43,44). In contrast, the first hyperthermic phase is not affected by IL-10 infusion and is therefore probably controlled by mechanisms different from those involved in the late phase.
Interestingly, the suppressing effect of IL-10 on fever is also associated with decreased concentrations of IL-1β in the liver and the pituitary gland. Presumably, IL-1β production by macrophages in these structures, in particular Kupffer cells in the liver (14), contributes to circulating IL-1β levels. Thus the IL-10-mediated decrease in IL-1β levels in plasma may reflect impaired production by blood monocytes and/or by tissue macrophages and may lead to reduced humoral signaling of IL-1β to the brain. By suppressing IL-1β concentrations in the liver, IL-10 may also affect fever by interfering with the neural route of signaling of IL-1β. If hepatic IL-1β is indeed involved in vagal afferent signaling, reduced production of intrahepatic IL-1β may lead to a decreased activation of neurons in the lower brain stem and, consequently, to reduced actions of these neurons on thermoregulatory mechanisms in the hypothalamus (5, 50). In the pituitary gland, IL-1β is produced by various endocrine and nonendocrine cells with activation by peripheral LPS (3, 55). Its local actions involve modulation of the secretion of a variety of hormones that are primarily under control of the hypothalamus (3). IL-10-induced downregulation of intrapituitary IL-1β production may therefore have neuroendocrine implications, e.g., regulation of the HPA axis (51).
The present results further show that the brain stem represents a target for circulating IL-10. The effects of circulating IL-10 are most likely due to actions in the AP, a CVO that lacks a blood-brain barrier and is thereby easily accessible for endotoxins but also for cytokines in the bloodstream. Indeed, IL-1β-immunoreactive cells have been found in the AP of rats 5 h after intravenous injection of this low dose of LPS (unpublished observations) that may represent microglial cells (55). Because IL-1 receptor type I mRNA is present in the AP (19), this brain structure not only produces IL-1β but likely also expresses the receptors that are necessary to mediate its effects. Therefore, we postulate that the local inhibitory action of IL-10 on IL-1β production in the AP may contribute to a diminished neural input to thermoregulatory centers in the hypothalamus and thereby inhibit the febrile response.
In the hypothalamus, IL-1β concentrations were not affected by LPS, which is somewhat surprising as IL-1β mRNA has been demonstrated in CVOs after administration of LPS (38). It probably relates to the low dose of LPS (10 μg/kg) used in the present study and/or the time interval studied. This may also explain the absence of detectable levels of TNF-α and IL-6 in tissue homogenates under these conditions.
Taken together, our findings suggest that peripherally administered IL-10 attenuates LPS-induced fever by inhibiting the production of pyrogenic cytokines in peripheral tissues and the brain stem and, consequently, interferes with humoral or neural signaling routes of these cytokines to the brain.
Strong IL-1β production in the hypothalamus and brain stem was seen after central administration of LPS, i.e., when the primary inflammatory response is generated within the CNS. This is in accordance with the induction of IL-1β mRNA and bioactivity in the CNS after intracerebroventricular LPS (15, 23, 39). Correlation analysis revealed that IL-1β production in the hypothalamus is associated with a febrile response, which shows kinetics different from that observed after peripheral LPS (10, 35, present study). Interestingly, both the febrile response following central LPS and IL-1β levels in the hypothalamus, but not in peripheral organs, were attenuated by central administration of IL-10, which is in agreement with findings in mice (16). In addition to the antipyretic effect, central administration of IL-10 also antagonized the effects of intracerebroventricular LPS on social exploratory behavior and body weight (6). Taken together, our data support the concept that brain-derived IL-1β is involved in fever (33) and possibly other sickness symptoms induced by central administration of LPS.
From the preceding observations, we conclude that a local action of IL-10 may be required to have an effect on body temperature under the conditions used. It is thus conceivable that central administration of IL-10, at doses up to 600 ng/rat, does not affect the febrile response to peripherally injected LPS, as central IL-10 does not affect IL-1β production in peripheral tissues, most likely because it does not reach such peripheral targets in sufficient concentrations.
The observed effects of exogenous IL-10 on fever implicate the presence of functional IL-10 receptors. Indeed, IL-10R1 mRNA is constitutively expressed both in peripheral organs (e.g., the liver) and in the hypothalamus. To test the hypothesis that IL-10 receptor expression is regulated at specific sites depending on the route of LPS administration, relatively high doses of LPS, but relevant for fever (33, 42), were administered peripherally or centrally. Interestingly, peripheral administration of LPS enhances IL-10R1 mRNA levels in the liver but not in the hypothalamus. Conversely, central LPS administration upregulated IL-10R1 mRNA expression in the hypothalamus but not in the liver. Although we cannot exclude the possibility that lower LPS doses as used in the fever studies regulate IL-10R1 expression to a lesser extent, the results indicate that LPS-induced IL-10 receptor modulation is site specific. This local (transient) upregulation of IL-10 receptors may further facilitate and/or amplify the inhibitory actions of IL-10. The presumed cellular sources of IL-10 receptors in the liver are Kupffer cells (27), in which IL-10 has been shown to inhibit IL-6 production and its own mRNA expression (26, 27). In the hypothalamus, IL-10 acts most likely on glial cells that are known to exhibit IL-10 receptors (34, unpublished observations), and IL-10 has been shown to be capable of inhibiting IL-1β production in glial cells in vitro (29).
In conclusion, our data show that IL-10 can modulate brain-mediated fever in response to low doses of LPS. Whether it indeed does so seems dependent on IL-10 reaching its receptors in the compartment where the primary inflammatory response is initiated. The mechanisms mediating the site-specific, antipyretic effects of IL-10 relate to suppression of the production of pyrogenic cytokines, notably IL-1β. Although our data show that IL-10 in the brain can attenuate the production of IL-1β by brain cells, we propose that this mechanism plays a limited role in peripheral inflammation but may become crucial in conditions of infections or inflammation of the CNS, such as meningitis, encephalitis, or trauma.
The authors thank Dr. A. B. Smit (Research Institute Neurosciences, Department of Molecular and Cellular Neurobiology, Faculty of Biology, Free University Amsterdam) for allowing the use of the ABI PRISM 7700 Sequence Detection System and Dr. A. A. Romanovsky (Trauma Research, St. Joseph's Hospital and Medical Center, Phoenix, AZ) for valuable comments. rrIL-10 and reagents for measurement of rat IL-1β, TNF-α, and IL-6 were kindly provided by Dr. S. Poole and Dr. A. F. Bristow (National Institute for Biological Standards and Control, Division of Endocrinology, Potters Bar, UK).
This study was supported by the Janssen Research Foundation and the Biomed II program (PL-96–2492) (to A. Ledeboer) and Netherlands Organization for Scientific Research Grant 903–50–240 (to A-M. Van Dam).
Address for reprint requests and other correspondence: A-M. Van Dam, Research Institute Neurosciences Free Univ., Dept. of Medical Pharmacology, VU Medical Center, Van der Boechorststraat 7, 1081 BT, Amsterdam, The Netherlands (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published February 7, 2002;10.1152/ajpregu.00766.2001
- Copyright © 2002 the American Physiological Society