INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

FEVER IS A REGULATED INCREASE in body temperature (Tb) often described as a rise in the thermoregulatory set point (16). With few exceptions, both endothermic and ectothermic vertebrates (as well as invertebrates) develop fever in response to injections of exogenous pyrogens such as lipopolysaccharide (LPS; endotoxin), viruses, gram-positive bacteria, and yeast. LPS, which is the most purified form of a compound from the cell wall of gram-negative bacteria, usually Escherichia coli, has been extensively used to induce fever in experimental animals (for a review see Ref. 16). To our knowledge, no information exists about the effect of LPS injection in toads.

Considerable efforts have been made to identify the mechanisms of fever, and evidence suggests that exogenous pyrogens stimulate the immune cells to produce and release cytokines, such as interleukin (IL)-1, IL-6, interferons, and tumor necrosis factor (16). Classically, cytokines are known to stimulate the generation of prostaglandins in the central nervous system, particularly PGE₂, thought to act as one of the final mediators of fever (26). In addition to this conventional viewpoint, recent evidence has accumulated that distinct mechanisms may be evoked depending on the route of pyrogen administration (10) and the sort of pyrogen injected (16, 42)

In addition to the action of pyrogens increasing Tb, endogenously formed antipyretic factors also participate in the febrile response. These substances have been proposed to play an important role in preventing excessive rises in Tb (16).

The mammalian peptide arginine vasopressin (AVP) has been recognized as an important antipyretic molecule in mammals (16, 30, 39, 40). As to nonmammalian vertebrates, arginine vasotocin (AVT) is the antidiuretic and pressor substance present in the neurohypophysis (37). AVT is a nonapeptide considered to be the most primitive known vertebrate neurohypophysial peptide of the vasopressin/oxytocin hormone family and may thus be ancestral to all the other vertebrate peptide hormones (see Ref. 21). However, no information exists about the role of AVT in febrile ectotherms, which are an interesting model to study thermoregulation because they rely essentially on behavioral mechanisms for Tb control, which is usually related to changes in the thermoregulatory set point (4, 5, 38).

In view of these considerations, we tested the hypothesis that AVT is an antipyretic molecule in the toad Bufo paracnemis. We therefore measured the preferred Tb of Bufo paracnemis after injections of LPS, AVT, or a combination of both.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Toads (Bufo paracnemis Lutz) of either sex weighing 150-250 g were collected on the Campus of the University of São Paulo, Ribeirão Preto, São Paulo state, Brazil. On arrival at the laboratory, the animals were kept indoors in aquariums with free access to tap water and basking areas at ambient temperature. All animals were fed meal worms once a week until 1 wk before surgery. Each animal was used only once, and all the experiments were performed between 2:00 PM and 6:00 AM.

Surgery. The toads were anesthetized in an aqueous solution of 3-aminobenzoic acid ethyl ester (MS-222; 0.3%; Sigma). For the temperature selection measurements, a thermistor probe was secured into the cloaca with skin sutures the day before the experiments began. All animals recovered promptly from anesthesia. After suture, the animals were left undisturbed for at least 24 h at 22-25°C in containers with free access to tap water and basking areas.

Toads used for intracerebroventricular injection had the fourth cerebral ventricle cannulated as previously described (3). Briefly, during anesthesia (induced by submergence in 0.3% MS-222) the skin covering the caudal half of the skull was removed and an orifice was formed by cutting out a circular piece of connective tissue between the first vertebra and the occipital bone. This orifice provided access to the fourth ventricle. A catheter (0.7 mm OD) was stereotaxically placed inside the orifice with the tip protruding into the cerebrospinal fluid (CSF) of the fourth ventricle. The opening was sealed with bone wax followed by an application of dental acrylic (Dental Fillings do Brasil), which secured the assembly in place. Screws had been drilled in the occipital bone and first vertebra to provide a firm contact after fusion with the acrylic. A tight-fitting mandril was kept inside the guide cannula to avoid its occlusion. This preparation allowed intracerebroventricular injections. All animals recovered from surgery within 1-2 h.

Another group of toads was anesthetized (submergence in 0.3% MS-222), and an arterial catheter (PE-50) filled with heparinized Ringer solution was occlusively inserted into the femoral artery. All animals recovered promptly from anesthesia.

Measurement of preferred Tb. Preferred temperature was determined in a thermal gradient chamber (1.5-m long, 0.15-m high, and 0.2-m wide) with an aluminum floor. One end of the floor was cooled to 10°C by a copper pad connected to a refrigerated water bath (VWR Scientific, 1160A, Niles, IL). An electrical resistor heated the other end to 40°C. Petri dishes filled with tap water throughout the chamber provided access to water at all temperatures. An animal with a cloacal temperature probe was placed in the center of the temperature gradient, and the thermistor output was continuously displayed on a chart recorder using a NARCO polygraph (Narcotrace 80).

Measurement of arterial blood pressure and heart rate. Arterial blood pressure was measured 24 h after surgery by connecting the arterial catheter to a Narco pressure transducer (Austin, TX, model P-1000B). The signal from the transducer was recorded on paper (Narcotrace 80). Heart rate was determined by counting pressure pulses.

Experimental procedure. Experiments were performed on conscious, unrestrained and undisturbed toads. One toad was placed in the middle of the temperature gradient and left there for 16 h. At the 5th h (time zero), Ringer, LPS (from Escherichia coli, serotype 0111:B4, Sigma), AVT (Peninsula), or a combination of all were injected into the dorsal lymph sac of the animals. The doses used were 2, 20, or 200 µg LPS/kg body wt and 10 or 300 pmol AVT/kg body wt in a final volume of 0.4 ml. The doses of LPS were based on pilot experiments and on the fact that they represent a common range of LPS doses used in other endotherm experimental models (10, 22, 33, 36), whereas the doses of AVT were based on a previous study on toads (23). The gradient chamber was continuously flushed with humidified room air at a rate of 1.5 l/min.

In another set of experiments, toads with the fourth cerebral ventricle cannulated were kept in the thermal gradient and injected at the 5th h (time zero) with LPS into the lymph sac (200 µg/kg) followed by an intracerebroventricular injection of mock CSF or AVT (10 pmol/kg). Other animals received injections of mock CSF or AVT (10 pmol/kg) only. A 10-µl Hamilton syringe and a dental injection needle (Missy, 200 µm OD) were used for all the intracerebroventricular injections. Injection was performed over a period of 2 min, and 1 min was allowed to elapse before the injection needle was removed from the guide cannula to avoid reflux. Preferred Tb was then continuously monitored for 11 h. The ionic composition of the mock CSF solution was (in meq/l) 56.6 NaCl, 2.7 KCl, 0.9 CaCl₂, 0.45 MgSO₄, and 27.0 NaHCO₃.

Experiments in which blood pressure was measured were performed on animals maintained at 23-25°C in a chamber that was continuously flushed with humidified room air at a rate of 1.5 l/min. Arterial blood pressure was recorded for ~5 min before and 1 and 2 h after the injections of Ringer or LPS (2, 20, and 200 µg/kg) into the lymph sac.

Calculations and statistical analysis. Mean preferred Tb was determined every hour in all experiments from individual chart paper recordings by manual calculation on the basis of a previous calibration. Repeated measures ANOVA was used to compare Tbs at each hour after injection with the value at time zero (the value immediately before the injections), and the difference between means was assessed by the Tukey's test. Mean arterial blood pressure was estimated from the pressure pulse with the use of the following formula: mean blood pressure = diastolic pressure + (systolic pressure  diastolic pressure) (20). Ordinary ANOVA was used to compare mean blood pressure and heart rate among the different groups, and the difference between means was assessed by the Tukey's test. All values are reported as means ± SD. Values of P < 0.05 were considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

During the control period, mean Tb was 23 ± 1°C (a mean of the previous 5 h before injection for all protocols). Ringer injection caused a transient nonsignificant tendency to a reduction in Tb (Fig. 1).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Change in preferred body temperature over time of toads injected into lymph sac (ls) at time zero with lipopolysaccharide (LPS) at dose of 2, 20, or 200 µg/kg body wt or Ringer as control. Arrows indicate time of injection. Values are means ± SD, n = 4-7 per group. * Significant (P < 0.05) difference from 0 h value.

Effect of LPS on preferred Tb of toads. Figure 1 shows the temporal effect of LPS injection on Tb. LPS at the dose of 200 µg/kg, but not at smaller doses, caused a significant increase in Tb from the 8th to 11th h after injection (P < 0.05). Thus the dose of 200 µg/kg of LPS was chosen for further experiments. A transient nonsignificant reduction in Tb was observed after all injections of LPS (Fig. 1).

Effect of LPS on arterial blood pressure and heart rate of toads. Table 1 shows the effect of systemic LPS on arterial blood pressure and heart rate before and 1 and 2 h after injection. No dose of LPS caused a significant change in blood pressure or heart rate.

Effect of AVT on LPS-induced fever in toads. Figure 2 illustrates the effects of systemic AVT on preferred Tb and on LPS-induced behavioral fever. AVT injection only (10 or 300 pmol/kg) caused no change in preferred Tb of the toads, whereas AVT at the dose of 300 pmol/kg abolished the rise in Tb caused by systemic administration of LPS (200 µg/kg).

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Effects of arginine vasotocin (AVT) injected into lymph sac on preferred body temperature and LPS-induced behavioral fever in Bufo paracnemis. Ringer and LPS 200 µg/kg body wt curves are same as those plotted in Fig. 1. Arrows indicate time of injection. Values are means ± SD, n = 4-7 per group. * Significant (P < 0.05) difference from 0 h value.

Additionally, a small transient reduction in Tb was observed after intracerebroventricular injection of AVT (10 pmol/kg) or mock CSF (Fig. 3). However, intracerebroventricular injection of AVT (10 pmol/kg) abolished LPS-induced fever (Fig. 3).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Effects of AVT injected into fourth cerebral ventricle on preferred body temperature and LPS-induced behavioral fever in Bufo paracnemis. Arrows indicate time of injection. Values are means ± SD, n = 5 per group. * Significant (P < 0.05) difference from 0 h value. CSF, cerebrospinal fluid.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study provides evidence that LPS induces behavioral fever in the toad Bufo paracnemis, a response that is abolished by intracerebroventricular injection of AVT at a dose that does not alter fever when injected peripherally. To our knowledge, this is the first study that reported an antipyretic role of AVT by means of behavior, a fact consistent with the notion that AVT/AVP elicits antipyresis by reducing the thermoregulatory set point.

As reviewed by Kluger (16), fever results from a rise in the thermoregulatory set point. In mammals, which use autonomic and behavioral mechanisms to maintain Tb, fever is functionally expressed as increases in metabolic heat production and decreases in heat loss, in addition to behavior (6). However, thermoregulation in ectotherms is primarily behavioral and thus preferred Tb in ectotherms has been shown to be directly related to changes in the thermoregulatory set point (4, 5, 38). This characteristic certainly makes ectotherm species an interesting model for the study of thermoregulation.

In our experiments, LPS (200 µg/kg) caused a significant increase in preferred Tb (Fig. 1). This implies that toads, similar to several other species, respond to LPS by increasing Tb. However, toads seem to be less sensitive to LPS than mammals (10, 33, 36) and birds (22), because in the present study relatively higher doses were needed to cause fever in Bufo. Moreover, in our experiments, the onset of fever in toads occurred 8 h after LPS injection, a longer period than usually observed. First of all, it is important to point out that this longer latency for fever installation is not due to the inability of toads to select preferred Tb quickly when in a thermal gradient, because several studies (4, 5, 24) have observed consistent changes in preferred Tb of toads 1 h after application of the stimulus. Actually, a number of studies have observed different onset latency times for LPS fever, ranging from 15 min (19) to 2.5 h (36), depending on the species, dose, and route of administration. Usually, the higher the LPS dose the shorter the latency of fever onset after intraperitoneal LPS (19), but that may not be true when the LPS dose is further increased (33, 36). However, the LPS dose seems to be of minor importance in determining the latency of the fever onset in toads, because in the present study this pattern was not dose dependent (Fig. 1). Furthermore, intraperitoneally injected LPS is usually associated with longer fever latency compared with intravenously injected LPS (19, 36). Thus we may suggest that the long latency of LPS fever observed here in toads was probably due to a slow distribution of the drug through the lymphatic system. Finally, a low pyrogenicity of LPS in toads may account for the long latency of fever installation, whereas other pyrogens may elicit a more rapid response. Certainly, further research is needed to determine the reason for the long latency of LPS fever in toads.

The advantage of fever lies in an inherent improvement in immune system function and a reduction in pathogen (especially bacterial) growth at febrile temperatures (see Refs. 16, 22), with consequent increment in survival (17, 33, 38). Fever in ectotherms was first observed in lizards (17, 38) and has since been reported in amphibians (27), fishes, and even some invertebrates (see Ref. 18), indicating that fever has an ancient phylogenetic history. However, a few exceptions exist: rabbit leucocyte pyrogen and pyrogenic suspension of killed Aeromonas hydrophila bacteria fail to produce fever in the lizard Cordylus cataphractus (18); E. coli-endotoxin and PGE₁ also do not induce fever in the teleost fish Lepomis gibbosus (25).

The present study also demonstrates that LPS-induced fever in toads was preceded by a nonsignificant reduction in preferred Tb, which seemed to be due, at least in part, to manipulation of the animals (4, 5). It should be pointed out that a transitory hypothermia was observed after most intracerebroventricular injections when the animals were manipulated for a longer period of time (total of 3 min) compared with the systemic injections. The observed tendency to a reduction in Tb could be due to an endotoxin shock reaction (31). However, this possibility seems unlikely because LPS injection caused no significant change in arterial blood pressure (Table 1). Furthermore, combined administration of AVT (300 pmol/kg) and LPS significantly reduced Tb, although injection of each agent alone did not (Fig. 2). The reason for this phenomenon is unclear. As far as we are concerned, there is no evidence in the literature indicating an interaction between AVT and LPS in decreasing Tb. Further studies are needed to establish the mechanisms involved in the transitory reductions in Tb observed in the present study.

As to the effect of systemic AVT on Tb, previous studies have observed that AVT may decrease Tb of euthermic pigeons (12, 13), but our data indicate that the same does not occur in toads (Fig. 2). As to febrile toads, we observed that AVT abolished LPS-induced behavioral fever (Fig. 2). Altogether, these results indeed support AVT as an antipyretic molecule in the toad Bufo paracnemis, because an antipyretic is defined as a substance that can reduce the Tb of febrile animals but has no effect on normal (nonfebrile) Tb unless the dosage is excessive (1). In support of the antipyretic effect of endogenous AVT is the fact that plasma AVT levels rise during fever in Pekin ducks (11).

AVT may exert its antipyretic action by increasing arterial blood pressure in toads (23). In rats, AVP per se has been reported to cause hypothermia when administered intravenously (34, 35), a fact that has been attributed to the baroreflex suppression of thermogenesis (34). However, in the case of toads, we think this is unlikely because the effect of AVT on blood pressure is immediate (unpublished data) and persists for a short period of time (minutes; Ref. 23), whereas its antipyretic effect persists for hours (present study). More studies are needed to clarify this issue.

It is interesting to note that during dehydration plasma AVT concentration increases from 1.5 × 10¹¹ M (hydrated animal) to 4.9 × 10¹¹ M in Bufo marinus (8) and from ~5 × 10¹² M to 5 × 10¹¹ M in Rana catesbeiana (32). It would be possible to speculate that AVT is involved in the decrease of Tb when toads are exposed to dry air without access to water (24). However, this seems unlikely, because in the present study, for the 300 pmol/kg dose, a concentration of 1.5 × 10⁹ M can be estimated by assuming that AVT injected into the lymph sac is evenly distributed throughout the extracellular space and that extracellular water is 20% of total body mass (2). Thus the AVT levels of our toads after bolus injection may be ~30-fold more concentrated than the value reported for dehydrated Bufo marinus (8), and we did not observe a reduction in Tb in this condition. Nevertheless, we report an antipyretic effect of AVT. Further studies are needed to assess this effect under physiological conditions such as dehydration. Unfortunately, to our knowledge no information exists about AVT concentration in the CSF of toads, so we are unable to draw conclusions about this matter.

Although our data indicate that systemically administered AVT plays an antipyretic role in toads, we did not establish the exact site where peripherally injected AVT acts to evoke antipyresis. This site is likely to be centrally mediated. Although AVT may not cross the blood-brain barrier [in mammals AVP does not (29)], this systemically administered peptide may interact with the central nervous system (CNS) through receptors present in the vessels of the circumventricular organs of the brain, which are known to lack a blood-brain barrier (9).

In our experiments, AVT injected intracerebroventricularly at a dose (10 pmol/kg) that did not affect fever when injected peripherally, completely abolished LPS-induced fever in toads, indicating that AVT elicits antipyresis in toads by acting on the CNS. Taken together, our results indicate that central AVT acts as an antipyretic peptide in toads, a fact that may or may not account for the antipyretic effect of systemically administered AVT.

Accordingly, AVP seems to exert its antipyretic action by acting on the ventral septal area (VSA) of the CNS of mammals (for a review see Ref. 30). It has been observed since 1974 that near the end of pregnancy, ewes were unable to develop fever (see Ref. 17). Among all the hormonal pattern alterations that occur during pregnancy, Naylor et al. (28) found that AVP concentrations were consistently correlated with the suppression of fever. In agreement with this evidence, further studies gave support to the antipyretic effect of AVP: 1) AVP infusion into the VSA of the CNS attenuates or prevents febrile increases in Tb in sheep, rabbits, and rats (for a review see Refs. 16, 30), whereas in the same experiments AVP infusion into the VSA at the same dose caused no change in Tb of euthermic rats; 2) VSA is a site of AVP-containing nerve terminals and high concentrations of AVP receptors in the CNS (7); 3) AVP levels increase in plasma and CSF during fever (15); 4) hypertonic saline and hemorrhage, which are potent stimuli of AVP release, cause antipyresis in rats (14); 5) administration of specific AVP antagonists into the VSA results in greatly enhanced fever (41).

Because AVT abolished behavioral fever of the toads, whose changes in preferred Tb are related to changes in the thermoregulatory set point, our data indicate that AVT reduces the thermoregulatory set point during fever. Some studies have been conducted to determine the mechanism by which AVP attenuates fever in mammals. Wilkinson and Kasting (39) reported that AVP injected intracerebroventricularly decreases the brain temperature of febrile rats more than that of nonfebrile rats at an ambient temperature of 4, 25, and 32°C, suggesting that AVP affects the thermoregulatory set point rather than a specific thermoregulatory mechanism. Indeed, it has been reported that when AVP is injected intracerebroventricularly, the thermoregulatory strategy to reduce fever varies at different ambient temperatures and is not due to changes in a single effector mechanism (40). Altogether, these data implicate the AVP/AVT system in eliciting antipyresis by decreasing the thermoregulatory set point.

In summary, our data indicate that AVT plays an important role in the behavioral thermoregulation of the toad Bufo paracnemis, acting on the CNS as an antipyretic peptide during LPS-induced fever.

Perspectives