INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

FEVER IS A REGULATED INCREASE in body temperature (T_b) often described as a rise in the thermoregulatory set point. The rise in T_b caused by fever is the actively inducting heat-gain effectors, such as an increase in metabolic heat production (shivering and nonshivering thermogenesis) and a decrease in heat loss and heat-seeking behavior (3). It is generally accepted that products of the immune system (cytokines) mediate fever (19) and that animals injected with lipopolysaccharide (LPS) produce several cytokines.

A revolution in the understanding of blood pressure control started after the description of the endothelium-derived relaxing factor (14), which was later identified as nitric oxide (NO) (24). Recently, it has been reported that NO also plays an important role in thermoregulation. It has been shown that NO is required for the production of fever (27) and that the L-arginine-NO pathway in the central nervous system (CNS) is necessary to produce hypoxia-induced hypothermia (4). The family of NO synthases (NOS), the enzymes that produce NO in vivo, consists of two different classes, i.e., the inducible and constitutive forms (5) that can be blocked by arginine analogs such as N-nitro-L-arginine (L-NNA) (21).

Most animals respond to a shortage of oxygen by lowering their T_b. This response should be adaptive because it decreases O₂ demand when availability of oxygen is limited (22), promotes a leftward shift of the oxyhemoglobin dissociation curve, and blunts the energetically costly response to hypoxia, e.g., increased cardiac output and ventilation (32). Moreover, hypoxia leads to activation of the NO-guanosine 3',5'-cyclic monophosphate pathway in the CNS (4), and central NO is known to cause a reduction in T_b (4, 8).

Previous studies have established that, during fever produced by endotoxin injection, the increase in T_b is markedly reduced by hypoxia (9, 12, 26). Thus the aim of the present study was to test the hypothesis that hypoxia inhibition of fever would be mediated by NO.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Experiments were performed in adult male Wistar rats, weighing 200-250 g, that were housed at controlled temperature (24 ± 1°C) on a daily 12:12-h light-dark cycle. The animals were allowed free access to water and food. All measurements were made between 8:00 AM and 1:00 PM.

Determination of the effect of LPS on T_b. Rats received LPS (from Escherichia coli, serotype 0111:B4; Sigma Chemical) by intraperitoneal injection of 2, 4, or 20 µg/kg body wt. LPS was dissolved in pyrogen-free saline, and the volume of each injection was 0.5 ml. Control animals received an intraperitoneal injection of saline (0.5 ml). T_b was determined at 30-min intervals by insertion of a thermocouple probe into the colon. It should be pointed out that, before the experiment, the animals were habituated to handling by means of temperature measurements performed twice a day for 2 consecutive days.

Determination of the effect of hypoxia on T_b. Rats were housed in a plastic glass chamber ventilated with room air for at least 2 h before control T_b was determined. Subsequently, a hypoxic gas mixture of 10, 7, or 5% O₂ (AGA) was flushed through the chamber for 30 min, and T_b was measured immediately after exposure to hypoxia.

Determination of the combined effects of LPS and hypoxia on T_b. The same animal chamber (now flushed with room air) was used. After the animals were habituated to the experimental condition (~2 h), control T_b was determined, and then experimental rats received LPS by intraperitoneal injection (2, 4, or 20 µg/kg) 3 h before hypoxic exposure (30 min, 7% inspired O₂). T_b was measured at 30-min intervals. This period of time was chosen on the basis of a previous study that assessed the effect of LPS on T_b (27).

Determination of the effect of NOS blocker on T_b. Rats were treated with L-NNA (Sigma Chemical) by intraperitoneal injection at a dose of 25 mg/kg body wt twice a day for 4 consecutive days. The volume of each injection was 0.5 ml. Control animals received intraperitoneal injections of saline (0.5 ml). T_b was measured daily during treatment. Doses, method of administration, and period of time after injections when T_b was determined were chosen on the basis of previous studies (1, 6, 11, 29). Traystman et al. (29) showed that nearly complete inhibition of NOS activity was observed at doses >10 mg/kg. Moreover, Dwyer et al. (11) have shown that repeated L-NNA administration for 7 days produces no greater inhibition of NOS activity than that for 4 days of treatment. Accordingly, we observed no further drop in T_b in rats injected with L-NNA for 5 days compared with 4 days of treatment (data not shown).

Determination of the combined effect of LPS, hypoxia, and L-NNA on T_b. After treatment with L-NNA (25 mg/kg body wt twice a day for 4 consecutive days), rats housed in an animal chamber flushed with room air received LPS by intraperitoneal injection (20 µg/kg) 3 h before hypoxic exposure, when the chamber was kept ventilated with room air. After this period, a hypoxic gas mixture (7 or 10% inspired O₂) was applied for 30 min. T_b was measured at 30-min intervals.

Statistical analysis. All values are reported as means ± SD. Changes in T_b were evaluated by repeated-measures one-way ANOVA, and the difference between means was assessed by the Dunnett multiple-comparisons test. In the analysis of the last experiment (Table 1), we compared the mean change in T_b in the test groups after hypoxic exposure with the control group using Student's t-test. To evaluate the effect of LPS on the L-NNA-treated group, a cumulative 3-h thermal index (TI) was calculated as areas under fever curves, as previously described (7). TIs were analyzed by using Student's t-test. Values of P < 0.05 were considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In all five experimental protocols, T_b was 36.6 ± 0.3°C during the control period. No group baseline differed significantly from the saline group. During the experiments, the mean chamber temperature was 25.4 ± 0.8°C, and the room temperature was 24.0 ± 1.0°C.

Effect of LPS on T_b. Figure 1 shows the effect of LPS on T_b. Injection of saline was followed by the typical small rise in T_b, which returned to baseline after 1 h. LPS injections produced a characteristic biphasic fever. Injection of 2 and 4 µg/kg produced the first peak (0.5 ± 0.3°C), which occurred at ~2 h, and the second (0.6 ± 0.2°C) at ~4 h after LPS injection (P < 0.05). Injection of 20 µg/kg caused a first peak (0.9 ± 0.3°C), which occurred at ~2.5 h, and a second (1.6 ± 0.3°C) at ~4 h postinjection (P < 0.01).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Effect of intraperitoneal injection of lipopolysaccharide (LPS) on body temperature (Tb). LPS produced significant increase (P < 0.05) in Tb in all experimental groups. Values are reported as means ± SD (n = 8 in each group).

Effect of hypoxia on T_b. Figure 2 shows the effect of hypoxia on T_b. When inspired O₂ was reduced from 21 to 10, 7, or 5%, a significant (P < 0.05) decrease in T_b was observed.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Effect of hypoxia on Tb. Values are expressed as means ± SD (n = 5 in each group). * Significant difference in mean Tb before vs. after hypoxia.

Combined effect of LPS and hypoxia on T_b. Figure 3 shows the combined effect of LPS and hypoxia (7% inspired O₂) on T_b. Animals injected with saline showed a significant decrease in T_b (1.6°C) after hypoxic exposure. When animals received LPS at a dose of 2, 4, or 20 µg/kg, a significant 1.8, 1.8, or 2.1°C drop occurred in T_b when inspired O₂ was reduced to 7% (P < 0.01).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Combined effects of hypoxia (7% inspired O2) and LPS-induced fever on Tb. Hypoxia abolished LPS-induced fever. There was significant decrease in Tb after exposure to hypoxia in all groups. Values are reported as means ± SD (n = 8 in each group).

Effect of the NOS blocker on T_b. Injections of saline had no effect on T_b during treatment (Fig. 4). L-NNA injections caused a significant (P < 0.05) drop in T_b (0.5, 0.7, and 0.9°C, respectively, on the second, third, and fourth day of treatment).

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Effect of chronic treatment with N-nitro-L-arginine (L-NNA) on Tb. Values are given as means ± SD (n = 10 in each group). * Significant decrease in Tb after treatment with L-NNA.

Combined effect of LPS, hypoxia, and L-NNA treatment on T_b. Two levels of hypoxia were used. Seven percent hypoxia caused a drop of 1.6 ± 0.1°C in animals injected with saline (control) and a drop of 2.2 ± 0.2°C in animals injected with LPS (Fig. 5A). In rats chronically treated with L-NNA (25 mg/kg body wt twice a day for 4 consecutive days), 7% hypoxia caused a drop of 1.0 ± 0.4°C in rats injected with saline and a drop of 1.2 ± 0.2°C in rats injected with LPS (Fig. 5C). The magnitude of this decrease in T_b was significantly lower (P < 0.05) when compared with the control animals (Table 1). Ten percent hypoxia caused a drop of 0.8 ± 0.2°C in animals injected with saline (control) and a drop of 0.7 ± 0.3°C in animals injected with LPS (Fig. 5B). In rats chronically treated with L-NNA, hypoxic exposure (10% inspired O₂) caused no significant change in T_b in the rats injected with saline and LPS (Fig. 5D), which was lower (P < 0.05) than that observed in the control group (Table 1).

View larger version (38K): [in this window] [in a new window]   Fig. 5.   Combined effects of LPS (20 µg/kg) and hypoxia [7 (A and C) or 10% inspired O2 (B and D)] on Tb in control rats (A and B) and in rats chronically treated with L-NNA (C and D). , Data from saline injection; , data from LPS injection (statistics in Table 1). Thermal index (TI), cumulative 3-h thermoregulatory response to LPS in control rats or in rats chronically treated with L-NNA. Values are given as means ± SD (n = 8 in each group). * P < 0.05 vs. control.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study provides evidence that NO participates in hypoxia inhibition of fever because chronic treatment with a NOS blocker reduced the drop in T_b caused by hypoxia.

A comparison between the present study and that of Feng et al. (13) deserves comment. Feng et al. examined the effects of central injection of PGE₂ on T_b under conditions of light and darkness and observed that the peak T_b achieved after injection of the agent depends on the dose, but not on initial T_b, time of injection, or light-dark cycle. However, the change in T_b was reported to be strongly dependent on initial T_b. In the present study, rats treated with L-NNA showed lower baseline temperatures before administration of saline or LPS (Fig. 5). This reduced baseline core temperature may have blunted the hypothermic effects of hypoxia in rats treated with L-NNA. The peak T_b during LPS fever was ~37.4°C in control rats and ~36.4°C in L-NNA-treated rats. This suggests that the controls have a greater operating range to decrease core temperature when exposed to hypoxia. Thus our results may be the consequence of a change in the regulated temperature, but not a regulated change in temperature, according to Feng et al.

In newborns and adults of a number of species, including humans, T_b decreases in response to a lack of oxygen (32). The importance of this response is emphasized by reports that show an increase in the survival of several species of animals if they are allowed to become hypothermic during hypoxia exposure (10, 15, 17, 22). Hypoxic hypothermia, sometimes aptly referred to as anapyrexia (a regulated decrease in T_b; Ref. 15), can be a beneficial response in hypoxic animals because it reduces oxygen consumption, produces a leftward shift of the oxyhemoglobin dissociation curve with the resulting improvement of oxygen loading in the lungs, and decreases the ventilation with a consequent blunting of the energetically costly responses to hypoxia (32). In the present study, the rise in T_b produced in conscious rats by LPS administration was prevented when the animals breathed 10 or 7% O₂, in agreement with previous reports (9, 12, 26). Moreover, hypoxia is known to stimulate the NO pathway in the CNS (4).

Recent studies have shown that NO accounts for a large part of the biological actions of endothelium-derived relaxing factor (21). The importance of NO can be demonstrated by inhibition of the effects of NO (25) with the use of L-arginine analogs such as L-NNA. In the present study, we have chosen L-NNA because it is a nonselective inhibitor of NOS, which acts on both the constitutive and inducible isoforms of the enzyme. These enzyme subtypes seem to be distributed throughout the body, including the CNS. The inhibition of NOS by L-NNA appears to be irreversible, suggesting that L-NNAs form a covalent linkage with the enzyme (11).

The sites of action of L-NNAs include the CNS, where they are responsible for thermoregulation (8); brown adipose tissue, where they are responsible for heat production (23); and vascular smooth muscle, where they are responsible for heat conservation (28). It is interesting to note that treatment with L-NNA caused progressive falls in T_b on successive days of treatment (Fig. 4), despite the fact that inhibition of NOS should decrease cutaneous heat loss because it causes vasoconstriction in both large and small arteries (2). Most likely, L-NNA elicits hypothermia by reducing the firing rate of sympathetic nerves innervating intercaspular brown adipose tissue (8), as well as brown fat blood flow (23). Besides these peripheral effects, the NOS blocker may influence the central thermoregulatory mechanism. However, intracerebroventricular injection of NO blocker was reported to have little effect on T_b (4, 18) or even to enhance T_b and oxygen consumption (8). Such opposite effects on T_b of NOS inhibitors given intracerebroventricularly or systemically may be confusing. To clarify this matter, the effect of a NOS blocker given intracerebroventricularly should be considered. Brain NO is known to mediate a drop in T_b caused by both hypoxia (4) and LPS (16).

Peripheral L-NNA injection of >10 mg/kg markedly decreases not only peripheral but also central NOS activity (29), and, as previously shown, the CNS plays an important role in the regulation of T_b (11). The combination of hypoxia inhibition of fever and chronic treatment with L-NNA reduced the change in T_b observed after hypoxic exposure, indicating that the NO pathway plays a role in hypoxia inhibition of fever (Fig. 5).

Scammell et al. (27) have demonstrated that NO is required for the production of fever, because fever was observed when LPS was injected alone, whereas when LPS was injected together with the NOS inhibitor nitro-L-arginine methyl ester, a marked hypothermia was observed. In the present study, according to the calculated TIs, there was a reduction in LPS-induced fever in the animals chronically treated with the NOS inhibitor L-NNA (Fig. 5). Moreover, we observed a delayed onset in T_b increase, which suggests that NO is required not only for the maintenance of fever, but also to trigger it. The complexities of NO action that underlie these responses have yet to be determined, but our results (obtained after chronic treatment), together with previous observations on acutely treated animals (27), caution against the assumption of a generalized view of NO action. The use of different protocols may complicate the interpretation of data obtained by application of NOS inhibitors. It seems clear, then, that the role of NO in thermoregulation is complex and may vary according to drug administration protocol. Indeed, the differences between acute and chronic inhibition of NOS have been assessed recently, with quite different results (30, 31). Cardiovascular parameters that are altered acutely by inhibition of tonic NO release can be normalized by compensatory mechanisms during chronic inhibition of NO production (31). For instance, acute inhibition of the NO pathway in the CNS leads to increased arterial blood pressure, whereas its chronic inhibition does not cause hypertension (30). Although the exact mechanisms remain to be determined, NO clearly plays an important role in thermoregulation, including hypoxia inhibition of fever.