Nitric oxide (NO) plays a role in thermogenesis but does not mediate immune-to-brain febrigenic signaling in rats. There are suggestions of a different situation in birds, but the underlying evidence is not compelling. The present study was designed to clarify this matter in 5-day-old chicks challenged with a low or high dose of bacterial LPS. The lower LPS dose (2 μg/kg im) induced fever at 3–5 h postinjection, whereas 100 μg/kg im decreased core body temperature (Tc) (at 1 h) followed by fever (at 4 or 5 h). Plasma nitrate levels increased 4 h after LPS injection, but they were not correlated with the magnitude of fever. The NO synthase inhibitor (NG-nitro-l-arginine methyl ester, l-NAME; 50 mg/kg im) attenuated the fever induced by either dose of LPS and enhanced the magnitude of the Tc reduction induced by the high dose in chicks at 31–32°C. These effects were associated with suppression of metabolic rate, at least in the case of the high LPS dose. Conversely, the effects of l-NAME on Tc disappeared in chicks maintained at 35–36°C, suggesting that febrigenic signaling was essentially unaffected. Accordingly, the LPS-induced rise in the brain level of PGE2 was not affected by l-NAME. Moreover, l-NAME augmented LPS-induced huddling, which is indicative of compensatory mechanisms to run fever in the face of attenuated thermogenesis. Therefore, as in rats, systemic inhibition of NO synthesis attenuates LPS-induced fever in chicks by affecting thermoeffector activity and not by interfering with immune-to-brain signaling. This may constitute a conserved effect of NO in endotherms.
- body temperature
- oxygen consumption
fever is considered to be one of the most common events that are part of a wider defense response of organisms to infection, trauma, or injury—the so-called sickness syndrome (32, 54). With only a few exceptions, both endothermic (birds and mammals) and ectothermic (fish, amphibians, and reptiles) vertebrates develop fever in response to exogenous pyrogens, such as Gram-positive and Gram-negative bacteria, virus, or fungi (7, 35). At least in rats, it is well known that LPS, an endotoxin from Gram-negative bacteria, induces fever by stimulating cytokine production and the synthesis of PGE2, which, in turn, affect the neural control of core body temperature (Tc) at the level of the preoptic area of the hypothalamus (POA) (36, 40, 41, 45, 47, 54–56). Even though there are some similarities in febrile responses between mammals and birds, such as the dependence of prostaglandin action on the brain (33, 38), febrile mechanisms in the latter group are much less understood (29).
In severe cases of systemic inflammation and sepsis, a reduction in Tc may occur instead of fever (1, 3, 49, 50, 52). In the rat model of LPS-induced systemic inflammation, this switch from fever to “hypothermia” seems to be a regulated thermoregulatory response (3, 49, 50, 52), which does not seem to be consequential to circulatory shock or global hypoxia, but that serves as a preemptive strategy to prevent the development of tissue hypoxia (18). It has also been shown that naturally occurring hypothermia is more advantageous than fever in the most severe cases of sepsis, as it relates to the control of endotoxemia, lung inflammation, and mortality rate (37). A similar fever-hypothermia dichotomy appears to exist in birds, as high doses of LPS induce a decrease in Tc in 4-day-old chicks (46) and 5-wk-old chickens (20).
The gaseous neurotransmitter nitric oxide (NO) has been reported to be involved in many physiological processes, including thermoregulation and fever (11). In birds, NO may play a role in the development of thermogenesis in duck embryos (23). Moreover, peripheral injection of the nonselective inhibitor of NO synthase NG-nitro-l-arginine methyl ester (l-NAME) blocks the LPS-induced fever in adult ducks (28). Another study reported an increase in nitrite concentration and inducible NOS mRNA expression in the lung and liver of 4-day-old hen chicks after injection with LPS at a dose as high as 10 mg/kg (46). Even though these data may indicate that NO is involved in the Tc responses to systemic inflammation in birds, it is not clear whether peripherally generated NO plays a role in fever by participating in immune-to-brain communication or in thermoeffector activity (e.g., thermogenesis, skin vasoconstriction, and behavioral heat gain).
In rats, it is known that NO acting on the brain is involved in the febrile response to LPS (5, 6, 61, 62, 64, 65). NO acts on the POA and the locus coeruleus as an antipyretic and pyrogenic molecule, respectively, during LPS-induced fever (61, 64). Furthermore, NO in the brain has been involved in autonomic and behavioral heat loss responses, acting as a mediator of fever suppression that occurs near parturition (5, 6), as well as a mediator of hypoxia-induced anapyrexia (opposite response to fever) in rats (10, 63) and in toads (behavioral thermoregulation utilized as a primary mechanism) (31). In contrast, NO plays no role in immune-to-brain febrigenic signaling in rats (65) but stimulates both nonshivering (brown adipose tissue) and shivering thermogenesis (11). Accordingly, intravenous l-NAME does not affect LPS-induced fever in rats exposed to a warm environment, but it does in rats exposed to a cooler environment (65). Moreover, intraperitoneal injection of l-NAME does not affect Tc, metabolic rate, and evaporative heat loss in rats kept at 30°C, but reduces Tc and inhibits the increase in metabolic rate in those animals kept at 18°C (35), reinforcing the idea that, at least in rats, peripheral NO acts on thermogenesis per se. Therefore, we asked whether the same scenario may apply to birds.
Chickens are an interesting model in which to study thermoregulation because they are precocial birds, which have clear thermogenesis and locomotor capacity immediately after hatching, are almost totally covered with down, and can respond effectively to heat and cold (19, 43, 70) and to LPS (46, 58, 59, 60). This early phase is critical to animals because any stressor at this time can affect pathophysiological processes later in life (60, 74). Thus, it is an important phase to be explored. Therefore, we investigated the effects l-NAME administered systemically on the changes in Tc induced by LPS in 5-day-old chicks. Autonomic (thermogenesis) and behavioral (huddling) thermoregulatory mechanisms were assessed, as were the levels of nitrate + nitrite and PGE2.
Day-old broiler chicks (Gallus gallus; Cobb 500 strain) were supplied by a local commercial hatchery (Globoaves, Itirapina, São Paulo, Brazil). Birds were reared as a group in temperature-controlled chambers (Premium Ecologica; Belo Horizonte, Brazil) under 32°C and a 14:10-h light-dark cycle until the day of the experiment, 5 days after hatching (body mass, 70–85 g). Standard broiler starter diet (7853, Presence, Evialis, Brazil) and water were provided ad libitum. For the oxygen consumption protocol, as individual measurements were taken, the eggs of broiler chickens (Globoaves) were incubated in standard conditions (37°C and 65% relative humidity for 21 days) on alternate days to have only a few 5-day-old chicks on the day of the experiment.
The study was conducted in agreement with the guidelines of the National Council of Control in Animal Experimentation (CONCEA-MCT-Brazil; Federal Law no. 11.794; Legislative Resolution no. 12), and with the approval of the local Animal Care and Use Committee (CEUA-FCAV-UNESP; protocol no 008184/13).
Core Body Temperature Measurements
Colonic temperature was measured by inserting a thermosensor (Yellow Springs Instrument, Yellow Springs, OH) 3 cm into the colon of the animal and connected to a tele-thermometer (Model 45TUC, Yellow Springs Instrument).
The intramuscular injections were performed in the hind limb. In pilot experiments, this route of injection revealed to be less stressful for peripheral injections than the breast muscle or coelomatic cavity. Results after injection in the hind limb (thigh) were more consistent and less variable than those after injection in the breast muscle or coelomatic cavity because 5-day-old chicks have thin breast muscles and organs are very compacted inside the cavity.
Determination of Plasma Nitrate + Nitrite Concentration
Trunk blood of chicks was collected after decapitation. The blood sample was centrifuged (1,000 g for 10 min at 4°C), and plasma was stored at −70°C. On the day of the assay, plasma samples were thawed and deproteinized with 95% ethanol at 4°C for 30 min and then centrifuged (12,000 g for 5 min at 4°C). The supernatant was used to measure nitrate + nitrite, according to the NO/ozone chemiluminescence technique (4), using a Sievers NO Analyzer (Sievers 280 NOA; Sievers, Boulder, CO). Sodium nitrate (Sigma-Aldrich Brazil, São Paulo, Brazil) was used as a standard reference.
The plasma and brain samples were assayed for PGE2 by UPLC-MS/MS. Each plasma sample (500 μl) was acidified to pH 3 with HCl and subjected to solid-phase extraction in a C18 column (Supelco, Bellefonte, PA) via sequential washings with water (2 ml), 10% methanol in water (2 ml), and ethyl acetate containing 0.005% of BHT (3 ml). The ethyl acetate fraction was dried in a SpeedDry rotational vacuum concentrator (Martin Christ; Osterode am Harz, Germany), and the residue was reconstituted in 50 μl of methanol (37). The brain was homogenized in precooled (4°C) methanol containing 50 μM of indomethacin and 0.005% of BHT using an ultrasonic processor (Sonics and Materials, Newton, CT). The resulting homogenate was centrifuged at 3,220 g for 10 min at 4°C, and the supernatant was subjected to liquid-liquid extraction in four steps: 1) hexane was added to the supernatant at a ratio of 2:1, vortex-mixed, and then centrifuged for phase separation; 2) the lower (methanol) phase was transferred to another tube and acidified with formic acid to pH 3.0; 3) ethyl acetate containing 0.005% BHT was added to the acidified phase at a 2:1 ratio, vortex-mixed, and then centrifuged for phase separation; and 4) the lower (ethyl acetate) phase was transferred to another tube and dried under a stream of nitrogen. The residue was then reconstituted in 100 μl of methanol and passed through a 0.22-μm filter (Millipore; Darmstadt, Germany).
Each processed plasma or brain sample (5 μl) was introduced into an ACQUITY TQD UPLC-DAD-MS Waters system (Waters, Milford, MA). Separation was performed on a 2.1 × 100 mm Kynetex C18 column packed with 1.7-μm particles (Phenomenex; Torrance, CA). The mobile phase was eluted at a rate of 0.2 ml/min with acetonitrile and water containing 0.1% formic acid, according to a gradient program: the acetonitrile-water ratio was 0.35 at 0–6 min, 0.90 at 6–8 min, and 0.35 at 8–10 min. Mass spectrometry detection was performed using a TD detector equipped with an electrospray ionization in negative mode: source temperature of 150°C; desolvation temperature of 400°C; capillary voltage of 3.5 kV; cone voltage of 20 kV and argon gas collision pressure and flow rate of 3.02 × 10−3 and 0.15 l/h, respectively. PGE2 was detected by multiple reaction monitoring (MRM) using the 351.3 → 271.3 m/z transition. Quantification was linear (r2 = 0.9982) for concentrations ranging from 0.5 to 50 ng/ml. All analyses were performed using MassLynx V 4.1 Software (Waters, Milford, MA).
Measurement of Oxygen Consumption
Oxygen consumption was determined using open-system respirometry. A flowmeter (MFS; Sable Systems International, Las Vegas, NV) continuously pulled air from the respirometer (3,000 ml) at a constant flow (1,000 ml/min), and the concentrations of the O2 and CO2 input and output were monitored by gas analyzers (Sable Systems International) connected to a computer for recording and data storage. V̇o2 and V̇co2 were determined from the rate of flow and concentration differences between the gases at the inlet and outlet of the respirometer. Values were determined in standard conditions of temperature, pressure, and dry air (STPD). The air entering the analyzers was dried using Drierite (Sigma-Aldrich Brazil).
In all protocols, experiments were performed with nonanesthetized, 5-day-old chicks between 8:00 AM and 3:00 PM. From the second to the fourth day, colonic temperature was measured three times, separated by intervals of 1 h, for habituation of the animals, thus avoiding handling influences on Tc on the day of the experiment. Ambient temperature was maintained at ∼32°C—except where specified—which can be considered thermally “comfortable” because the animals exhibit normal exploratory behavior (looking for food and water), and they do not show huddling behavior or panting (personal observations).
Protocol 1: effect of intramuscular injection of LPS on Tc in chicks.
Body temperature was measured before, and 1, 2, 3, 4, and 5 h after intramuscular injection of saline or LPS (from E. coli, 0127:B8; Sigma-Aldrich Brazil) at the doses of 2, 10, 50, and 100 μg/kg. In addition to the saline group, Tc was also measured in another control group of animals that received no injection.
Protocol 2: effect of intramuscular injection of LPS on plasma nitrate + nitrite concentration in chicks.
The possible activation of a peripheral NO pathway during LPS challenge was tested by measuring the plasma levels of the NO metabolites, nitrate + nitrite (16, 27), after LPS injection. Body temperature was measured, and blood samples were collected for determination of plasma nitrate + nitrite concentrations before, and 1 or 4 h after intramuscular injection of saline or 2 and 100 μg/kg LPS. The times for blood collection were those that coincided with the reduction in Tc (1 h) and the fever (4 h) responses (Fig. 1).
Protocol 3: effect of intramuscular injection of l-NAME on Tc in LPS-treated chicks.
Peripheral injection of the nonselective NO synthase (NOS) inhibitor l-NAME (50 mg/kg im; Sigma-Aldrich Brazil) was used to determine the role of systemic NO in LPS-induced fever and the fall in Tc. Body temperature was measured before, and 1, 2, 3, 4, and 5 h after intramuscular injection of saline or LPS at the doses of 2 and 100 μg/kg. l-NAME was injected intramuscularly 50 min before saline or LPS. The dose of l-NAME is the highest one that does not change Tc per se in chicks (16) and reduces plasma nitrite + nitrate levels by ∼52% 4 h after injection of LPS at 100 μg/kg (pilot experiments). In some animals, Tc was measured up to 10 h after 100 μg/kg LPS injection, preceded by l-NAME or saline, to verify long-term effect of these treatments on Tc.
Protocol 4: effect of intramuscular injection of l-NAME on oxygen consumption in LPS-treated chicks.
To investigate the effect of NO inhibition on thermogenesis (an autonomic thermoeffector) during LPS challenge, oxygen consumption (an indirect measure of metabolic rate) was determined. Each animal was placed in the respirometry chamber for an acclimation time of ∼30 min. After this time, l-NAME (50 mg/kg im) was injected, followed by saline or LPS (2 and 100 μg/kg) 50 min later. Then, the lid of the respirometer was sealed, and oxygen consumption was determined at intervals of 40 min from 0.5 to 4 h after saline or LPS injection. Body temperature was measured just before l-NAME injection (Ti) and 4 h after LPS treatment (Tf).
Protocol 5: effect of intramuscular injection of l-NAME on Tc in LPS-treated chicks in the warm condition.
This protocol is based on the fact that in a warm environment, thermogenesis is not required for the development of fever, which in this case is brought about by activation of heat conservation effectors. Therefore, if the effects of l-NAME on the Tc response to LPS happen to persist in the warm environment, it would imply that they are not linked to a specific thermoeffector and more likely reflect altered immune-to-brain signaling. To this end, the temperature inside the experimental chamber was raised from 31–32°C to 35–36°C at least 12 h before the experiment. Body temperature of 5-day-old chicks was measured the following morning, before, and 1, 2, 3, 4, and 5 h after intramuscular injection of saline or LPS at the doses of 2 and 100 μg/kg. l-NAME (50 mg/kg im) was injected 50 min before saline or LPS.
Protocol 6: effect of intramuscular injection of l-NAME on huddling behavior in LPS-treated chicks.
Huddling is a heat conservation mechanism used by groups of neonate mammals (42, 48) and birds (26, 71) in cold conditions. As fever is known to be brought about by autonomic and behavioral mechanisms of heat gain (7), huddling is a predictable behavior during fever development. In this case, if l-NAME inhibits thermogenesis only, and not a febrigenic signal, it is expected that a more prominent compensatory behavioral heat conservation response (huddling) will be activated to increase Tc during fever development. To test this hypothesis, animals were divided into groups of five individuals in separate climatic chambers, all maintained at ∼30°C, on the day before the experiment. On the next day, they were injected with l-NAME (50 mg/kg im) or saline 50 min before LPS (100 μg/kg im) injection. Animals were monitored by a webcam (LifeCam HD-3000, Microsoft) that was programed to take pictures every 2 min until 4 h after LPS injection. Tc was measured just before l-NAME injection (Ti) and 4 h after LPS treatment (Tf).
Protocol 7: effect of intramuscular injection of l-NAME on plasma and brain levels of PGE2 in LPS-treated chicks.
PGE2 levels were determined as an index of febrigenic inflammatory signaling. In the present protocol, l-NAME or saline was injected 50 min before LPS (100 μg/kg im); control chicks received saline. Trunk blood and brain were collected 4 h after LPS injection, and PGE2 was assayed by ultra-performance liquid chromatography-tandem mass spectrometry, as described above.
Data are presented as means ± SE. To show the effects of LPS and l-NAME on Tc, thermal indexes (TI) were calculated using the area under curve during the first 2–3 h (LPS-induced reduction in Tc) and the last 3 h (LPS-induced fever) after LPS injection (°C/h). Thermal indexes were compared using one-way analyses of variance (ANOVA) followed by the Tukey test. For the long-term effect of l-NAME or saline on LPS-induced fever, TI were calculated during the febrile response (from 4 to 10 h after 100 μg/kg LPS injection) and compared between the two treatments using unpaired t-test. Plasma levels of nitrate + nitrite were compared by two-way ANOVA, followed by Bonferroni test. Plasma and brain PGE2 concentrations were compared by one-way ANOVA, followed by the Tukey test. For the behavioral experiments, the number of single chicks and the total surface area occupied by the five chicks were determined. Repeated-measures two-way ANOVA was used to analyze the effect of LPS and l-NAME + LPS on behavioral parameters, as well as on oxygen consumption. In these cases, differences among means were identified by the Holm-Sidak post hoc test. Values of P < 0.05 were considered to be statistically significant.
Effect of Intramuscular Injection of LPS on Tc in Chicks
The initial Tc values for chicks in noninjected (40.3 ± 0.1°C), saline (40.3 ± 0.1°C), LPS 2 (40.2 ± 0.1°C), 10 (40.2 ± 0.1°C), 50 (40.1 ± 0.1°C), and 100 mg/kg (40.2 ± 0.1°C) groups were not statistically different (P > 0.05; one-way ANOVA and Tukey post hoc test). The same was observed for the initial Tc values in the other protocols.
Figure 1 depicts the effect of intramuscular injection of different doses of LPS on Tc in chicks. LPS at a dosage of 50 and 100 μg/kg, but not 2 and 10 μg/kg, caused a decrease in Tc during the first hour after injection (Fig. 1, A and B). All of the doses of LPS induced increases in Tc 3 to 5 h after injection (Fig. 1, A and C). On the basis of the results of this protocol, where 2 μg/kg of LPS induced only fever and 100 μg/kg LPS induced a decrease in Tc followed by fever (Fig. 1), these doses were chosen for the next protocols.
Effect of Intramuscular Injection of LPS on Plasma Nitrate + Nitrite Concentration in Chicks
Plasma nitrate + nitrite concentration in intact chicks (noninjected) was 16.45 ± 0.82 μM. As seen in Fig. 2A, saline did not change this variable at 1 and 4 h after injection. Plasma nitrate levels were significantly higher at 4 h, but not at 1 h, after LPS treatment regardless of the dose (2 or 100 μg/kg). The higher dose of LPS induced a more prominent response than the lower dose, though (Fig. 2A).
In Fig. 2B, individual values are shown for Tc and plasma concentration of nitrite + nitrate 1 h and 4 h after LPS treatment (both doses). Correlation between the two parameters was found neither at 1 h (r = 0.026) nor 4 h (r = 0.039) after LPS injections. At 1 h, there were changes in Tc without proportional changes in plasma nitrate + nitrite levels, while at 4 h, the alterations in Tc were not comparable with the big changes in plasma nitrate + nitrite levels (Fig. 2B).
Effect of Intramuscular Injection of l-NAME on Tc in LPS-Treated Chicks
An intramuscular injection of 50 mg/kg of l-NAME did not change Tc in chicks (Fig. 3), but an increase in the magnitude of the reduction in Tc, induced by LPS 100 μg/kg (P < 0.05; Fig. 3, A and B), was observed, as was an attenuation of the febrile responses to both LPS at 2 and 100 μg/kg (P < 0.05; Fig. 3, A and C). Fig. 3D depicts the long-term febrile response to LPS 100 μg/kg (from 4 to 10 h after injection) when chicks were pretreated with l-NAME or saline. It can be seen that Tc continued to rise to a peak at ∼7–8 h after LPS injection and then tended to return to baseline at time 10 h. l-NAME inhibited the whole LPS-febrile response, as can be observed by the lower thermal index compared with the group pretreated with saline (P < 0.01; unpaired t-test).
Effect of Intramuscular Injection of l-NAME on Oxygen Consumption in LPS-Treated Chicks
Figure 4 depicts the time course for the effect of LPS and l-NAME on oxygen consumption in chicks. Neither saline injections nor l-NAME + saline treatment changed oxygen consumption (Fig. 4, A and B). This parameter was reduced 70 min after intramuscular injection of 100 μg/kg LPS (effect of time: P < 0.05; Fig. 4, A and D). Oxygen consumption was lower in the animals injected with l-NAME + LPS100 than in those injected with saline + LPS100 for the duration of the experiment (interaction treatment time: P < 0.05; Fig. 4D). In contrast, no significant difference in oxygen consumption was observed between the saline + LPS2 and the l-NAME + LPS2 groups (Fig. 4C).
Effect of Intramuscular Injection of l-NAME on Tc in LPS-Treated Chicks in Warm Conditions
Figure 5 shows that by exposing chicks to a warmer environment, with temperatures around 35–36°C, saline and l-NAME had no effect on Tc. Furthermore, the lowest dose of LPS induced an increase in Tc 4–5 h after injection and was not altered by pretreatment with l-NAME (Fig. 5, A and C). The higher dose of LPS caused an initial decrease followed by an increase in Tc 1 and 4–5 h after injection, respectively. These responses to LPS were not affected by l-NAME (Fig. 5, B and C).
Treatment with saline or l-NAME altered neither the number of ungrouped chicks (Fig. 6, A1 and A2) nor the surface area occupied by the chicks (Fig. 6, B1 and B2). The injection of LPS at 100 μg/kg induced a reduction in the number of single chicks (Fig. 6A1). Animals showed clear sickness behavior and huddling; i.e., they stopped moving and moved closer to each other and to the wall of the chamber, responses that started about 2 h after LPS injection. These behaviors were totally different from those of untreated chicks, which kept moving and did not huddle in the same ambient condition. The treatment with l-NAME before LPS injection caused a further reduction in the number of single chicks and in the surface area occupied by the animals (Fig. 6, A3 and B3), i.e., they huddled much closer to each other.
Effect of Intramuscular Injection of l-NAME on Plasma and Brain PGE2 Concentrations in LPS-Treated Chicks
100 μg/kg of LPS increased brain, but not plasma, levels of PGE2 4 h after injection (Fig. 7). The pretreatment with l-NAME did not change PGE2 concentrations in both plasma and brain of LPS-treated chicks.
The results of the present study corroborate the hypothesis that NO is not involved in the immune-to-brain signaling to induce thermoregulatory responses to LPS in chicks. Its contribution to fever seems to be related to a permissive action on thermogenesis.
Peripheral LPS administration triggers widespread, inducible NO synthase expression, resulting in abundant production of NO in chickens (9) and mice (53). In our chicks, plasma nitrate + nitrite concentration did not change 1 h after LPS injection, but increased significantly after 4 h (Fig. 2A). Comparable results were reported in broilers of 3.8 kg body wt (14) and in adult rats (27), in which plasma nitrate did not change before 2 h of LPS injection.
We found no correlation between the magnitude of fever and the plasma levels of NO metabolites during systemic inflammation. Plasma nitrite + nitrate levels may be considered as a qualitative rather than quantitative indication of total NO production because of the dynamic NO transformations and reactions with proteins, including hemoglobin, as well as the fast elimination of NO as exhaled gas and as nitrate excreted in the urine (2, 12, 25, 31, 67, 70). Despite these facts, the much higher increase in plasma levels of nitrite + nitrate after administration of 100 μg/kg of LPS without a higher increase in Tc raises a question about the relevance of NO to LPS-induced febrigenic signaling in chicks. Accordingly, the present study also shows that l-NAME did not affect the LPS-induced rise in the brain levels of PGE2 in chicks. To our knowledge, this has never been assessed in any species before. PGE2 is considered the final mediator of fever, acting directly in the brain to induce fever in mammals (36, 40, 41, 45, 47, 54, 55, 56). Therefore, the brain levels of PGE2 are generally considered as an overall reflection of febrigenic inflammatory signaling. The critical role of PGE2 in febrigenesis appears to be applicable to birds, as intracerebroventricular injection of PGE2 increases Tc in chickens (38) and inhibition of prostaglandin synthesis attenuates fever in these animals (33, 38), as well as in ducks (28). Failure of previous studies to detect enhanced brain PGE2 levels during LPS fever in chickens (24) and ducks (39) may have resulted from the different techniques used for the assays, since we used a state-of-the-art quantification method (UPLC-MS/MS), which is more accurate and specific for PGE2 than the enzyme immunoassays employed in the studies by Fraifeld et al. (24) and Marais et al. (39). Regarding plasma PGE2, we did not observe any effect of LPS on this variable, contrasting with the increased levels reported in 5-wk-old Ross chickens injected with 2.5 mg/kg LPS (21). Because the latter study used a similar method for PGE2 measurement as we did, differences among species, strains, LPS doses, and age should also be considered.
The question then arises as to why l-NAME induced a more pronounced Tc fall early on and an attenuated fever at a later stage of the inflammatory response (Fig. 3). One possibility is that l-NAME attenuates thermogenesis. This was indeed observed in our LPS-injected chicks, which reduced oxygen consumption after NOS blockade in a neutral-subneutral ambient temperature (Fig. 4D). The finding that LPS fever was unimpaired by l-NAME in a warm environment strengthens this argument. The contribution of thermogenesis to fever in a warm environment is usually negligible, and activation of heat conservation mechanisms seems to be dominant for febrigenesis under this condition. Corroborating the present finding, studies in rats also show an inhibitory effect of l-NAME on the febrile response at 24°C, but the opposite effect, i.e., an increase in LPS-induced fever, at an ambient temperature (Ta) of 31°C (65). There is enough evidence of the further importance of vasoconstriction for fever development in mammals in the warm condition (15, 66, 68, 72). Regarding birds, although conclusive data of the role of heat conservation mechanisms during fever are lacking, there is no reason to believe it would be different, as heat gain (production + conservation) activation is a general response to pyrogens in several species of vertebrates (7). The role of vasoconstriction in febrile chicks is actually under investigation in our laboratory.
At least in rats, the attenuation of fever by systemic l-NAME seems to be the result of direct inhibition of thermogenesis in brown adipose tissue (BAT), since some studies show that peripheral inhibition of NO synthesis induces vasoconstriction in this tissue and blocks its activation by adrenaline (11, 35, 44, 62). Unlike mammals, though, birds do not have BAT. They produce heat by shivering or nonshivering mechanisms in skeletal muscle. The latter process is still not well understood, but there is evidence implicating Ca2+-dependent mechanisms (8) and the activity of the proton uncoupling proteins avUCP (17) and hmUCP (8), which are analogous to UCP-1 in the mammalian BAT (13). Further studies are needed to identify the role of NO in specific thermogenic mechanisms at work in birds. Moreover, one could argue that the NO interaction with hemoglobin (Hb) might account for its metabolic effect; however, this may not be the case because, even though there are no data about NO-Hb, the interaction of NO with other proteins, such as albumin, or with glutathione is demonstrated to have no thermogenic effects in rats (65).
Additional evidence indicating that thermogenesis rather than febrigenic signaling is impaired by l-NAME came from the huddling experiment performed in the present study. In that experiment, pretreatment with l-NAME made the chicks huddle more at 2–4 h after a LPS dose of 100 μg/kg, as if the brain were making use of Tc-increasing effectors other than thermogenesis to compensate for the inability of the l-NAME-pretreated animals to develop fully blown fevers at Ta of 30–31°C. Such compensation can be confirmed by comparing changes in Tc in single (∼0.3°C decrease; Fig. 4) and grouped chicks (∼0.4°C increase; Figs. 3 and 6) 4 h after LPS injection, both pretreated with l-NAME, at the same Ta. Nevertheless, the effectiveness of this strategy seems to be partial, since the febrile response of the l-NAME-pretreated animals was still lower than that of the animals pretreated with saline up to 10 h after LPS injection.
In conclusion, the results of the present study are consistent with the notion that NO is not involved in the immune-to-brain febrigenic signaling induced by LPS and that the thermal effects of systemic l-NAME are more likely to be related to inhibition of thermogenesis in 5-day-old chicks. Our findings might explain the inhibitory effects of systemic l-NAME on the LPS fever of adult ducks (28), as well as the Tc-reducing effects of l-NAME in chicks (16). Last, but not least, the similarities between the present findings and those of previous studies in rats (11, 35, 62, 65) are also suggestive of a preserved mechanism in endothermic animals.
Perspectives and Significance
Although birds and mammals have evolved endothermy independently and possess different mechanisms for thermogenesis (7), the neurochemical mediators and pathways involved in thermoregulation may present similarities. In this context, the investigation of the role of NO in the shivering and nonshivering thermogenesis in birds, both occurring in the skeletal muscle (8, 17), will be important for elucidating the thermogenic role of NO in endotherms.
Furthermore, the dichotomous (hypothermia vs. fever) nature of the Tc response to LPS in rats (3, 51) has now been described in chicks. To our knowledge, this is the first study to show reduction in thermogenesis during decrease in Tc induced by LPS in birds, which is again similar to what is shown in rats (22, 49). As the LPS-induced Tc reduction is suggested to be a regulated process in the latter animals (3, 49, 50, 52), avoiding tissue hypoxia (4, 18), it would be reasonable to suggest the same occurs in birds. In the present study, huddling was inhibited until almost 2 h after LPS injection (coinciding with the Tc decrease), indicating behavioral responses of heat conservation might be inhibited at this time; however, a complete study demonstrating the regulated or nonregulated nature of this response in birds is still lacking and is currently under investigation in our laboratory.
The present study was supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (2013/13386-4 to K. C. Bícego, 2012/03831-8 to A. A. Steiner, 2012/19966-0 to L. H. Gargaglioni, 2014/50265-3 to N. P. Lopes, and 2014/13586-6 to E. C. Carnio) and was part of the activities developed by V. Dantonio to obtain a master's degree at the Joint Graduate Program in Physiological Sciences (PIPGCF) from UFSCar/UNESP. V. Dantonio was the recipient of a Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) Fellowship.
No conflicts of interest, financial or otherwise, are declared by the authors.
Author contributions: V.D., A.A.S., and K.C.B. conception and design of research; V.D., M.E.B., M.H.F., E.N.K., G.A.B., and K.C.B. performed experiments; V.D., M.E.B., M.H.F., E.N.K., G.A.B., N.P.L., A.A.S., and K.C.B. analyzed data; V.D., E.C.C., A.A.S., and K.C.B. interpreted results of experiments; V.D. and K.C.B. prepared figures; V.D., L.H.G., E.C.C., A.A.S., and K.C.B. drafted manuscript; V.D., A.A.S., and K.C.B. edited and revised manuscript; V.D., M.E.B., M.H.F., E.N.K., G.A.B., N.P.L., L.H.G., E.C.C., A.A.S., and K.C.B. approved final version of manuscript.
We thank Bruno Mangili P. Rodrigues for technical assistance in some protocols.
- Copyright © 2016 the American Physiological Society