Keywords:
Journal of Physiology (2000), 526.2, pp. 417—424
0615
417
Autonomic and behavioural thermoregulation in starved rats Sotaro Sakurada*, Osamu Shido†, Naotoshi Sugimoto*, Yasuhiro Hiratsuka*, Tamae Yoda‡ and Kazuyuki Kanosue‡ *Department of Physiology, School of Medicine, Kanazawa University, 13_1 Takara_machi, Kanazawa 920_8640, †Department of Physiology, School of Medicine, Shimane Medical University, 89_1, Enya-cho, Izumo 693_8501 and ‡Department of Physiology, School of Allied Health Sciences, Osaka University Faculty of Medicine, Suita 565_0871, Japan (Received 25 January 2000; accepted after revision 11 April 2000)
1. We investigated the mechanism of starvation-induced hypothermia in rats. 2. Threshold core temperatures (T ) for tail skin vasodilatation and cold-induced thermogenesis were determined after a 3 day starvation using a chronically implanted intravenous thermode. Food deprivation significantly lowered the threshold T for heat production, but did not affect the heat loss threshold. 3. Thermogenic response to a fall in T below its threshold was enhanced by starvation. 4. Preferred ambient temperatures (T ) and T were measured before and during a 3 day starvation in a thermal gradient. The 3 day starvation significantly lowered T only in the light phase of the day. The level of hypothermia was the same throughout the fasting period, while T gradually increased during the 3 days of starvation. 5. When rats were starved at a constant ambient temperature of 25°C (no thermal gradient), their T levels were comparable with those of the rats kept in the thermal gradient. 6. The results suggest that, in rats, hypothermia caused by starvation was not due to a decrement in thermogenic capability, but was due to a decrease in the threshold for the activation of thermogenesis. cor
cor
cor
pref
cor
cor
pref
cor
During periods of reduced food availability, hypothermia was observed in various endotherms (Penas & Benito, 1986; Hohtola et al. 1991; Rashotte et al. 1991; Hoshino, 1996). A lower core temperature (T ) in starved animals is thought to conserve energy by reducing metabolic heat production due to the QÔÑ effect. However, a question arises as to whether this hypothermia is regulated as part of a general strategy to preserve energy, or simply because the thermogenic systems are unable to maintain an adequate heat production. In birds, there are reports suggesting that starvation-induced hypothermia is a regulated phenomenon. In pigeons, hypothermia was accompanied by a fall in threshold T for heat production (Graf et al. 1989). Food deprivation decreased the lower critical temperature and threshold T for shivering in penguins (Duchamp et al. 1989). Such changes in thermoregulatory function have been known to contribute to the maintenance of T at low levels (Shido et al. 1995). Interestingly, once shivering took place, the slope of metabolic heat production against ambient temperature (T ) was steeper than that of normally fed birds (Duchamp & Barre, 1989), indicating that the starved birds still had the ability to produce a sufficient amount of heat. Thus, these observations suggest that hypothermia in cor
cor
cor
cor
a
starved birds is not a result of a malfunction in heat production systems, but is a regulated energy conservation response. In starved mammals, hypothermia has also been reported (Penas & Benito, 1986; Hoshino, 1996). However, the mechanism of starvation-induced hypothermia has not yet been intensively studied. Although both birds and mammals are endotherms, their thermoregulatory mechanisms are rather different. The thermoregulatory responses elicited by changing the temperature of the anterior hypothalamus are different for birds and mammals (Schmidt, 1976; Simon et al. 1976; Simon-Oppermann et al. 1978). Pigeons also show a different febrile response to pyrogens compared with that of mammals (Nomoto, 1997). Thus, the observations on birds may not be applicable to mammals. It is therefore important to evaluate whether hypothermia in starved mammals is regulated or not. Little is known about how heat loss mechanisms are affected by food deprivation, in birds or mammals. T is thought to be regulated within a range between the threshold temperatures for heat loss and heat production by the thermoregulatory centre. Thus, both heat loss and heat cor
418
production thresholds should be examined to evaluate the mechanism of starvation-induced hypothermia. In the present study we used rats to determine their threshold T values for tail skin vasodilatation, heat loss response, and cold-induced thermogenesis in order to assess alterations in thermoregulation associated with food deprivation. In addition to autonomic effectors, endothermic animals use behavioural response to achieve and maintain a stable T . The preferred ambient temperature (T ) has been used to evaluate behavioural thermoregulation in various classes of vertebrates (Crawshaw, 1980; Gordon, 1994), e.g. when T is below a regulated temperature, as in febrile animals, they select a higher T to raise T through reducing heat loss or facilitating heat gain (Crawshaw & Stitt, 1975; Akins et al. 1991; Sugimoto et al. 1996). There are some reports suggesting that heat-seeking behaviour, one of the indicators of behavioural thermoregulation, is modified by food deprivation (Hamilton, 1959; Ostheim, 1992). However, it is not yet known how T is altered by food deprivation in rats. The present study also examined the T of starved rats to evaluate the effect of food deprivation on behavioural thermoregulation.
cor
cor
pref
cor
a
cor
pref
pref
Animals
J. Physiol. 526.2
S. Sakurada and others
METHODS
Male Wistar rats (Std: WistarÏST, Japan SLC, Shizuoka, Japan), initially weighing 280 g, were used. Rats were individually housed in wire mesh cages at a T of 24·0 ± 1·0 °C with a 12:12 h light—dark cycle (lights on at 07.00 h), and allowed access to tap water and laboratory rat chow ad libitum except during the period of fasting. After measurements, rats were killed with a large dose of anaesthetic (pentobarbital sodium i.p.). The animals used in this study were maintained in compliance with ‘Guidelines of the Care and Use of Laboratory Animals in Takara-machi Campus of Kanazawa University’ and the study was approved by the local ethical committee. a
Experiment 1: thermoeffector thresholds
Fifteen rats were used. A stainless-steel guide cannula (0·7 mm o.d.) was stereotaxically implanted into the preoptic-anterior hypothalamus under pentobarbital sodium (50 mg kg¢ i.p.) anaesthesia, using co-ordinates given in the atlas of Paxinos & Watson (1986). After the first operation, all rats underwent a minimum of six sessions of loose restraint in cylindrical wire mesh cages for 4 h per day to habituate them to the experimental conditions. Approximately 10 days after the first operation, between 15.00 and 16.00 h, the rats were again anaesthetized with the same anaesthetic. A thermocouple covered with a polyethylene tube was inserted into the hypothalamus via the guide cannula and was fixed to the skull with dental cement. The leads were passed subcutaneously and exteriorized at the nape of the neck. Additionally, an intravenous thermode, which was made of a double-lumen polyethylene tube (1·0 mm o.d., 0·4 mm i.d., 12 cm long; DP4, Natsume, Tokyo, Japan) (Sakurada et al. 1993), was inserted into the inferior vena cava via the left femoral vein. To minimize unwarranted stimulation of skin thermoreceptors, the proximal end of the thermode was exteriorized on the lower back of the rat. The tip of the thermode was protected by a metal ring affixed to the skin surface with surgical glue.
After the second operation, eight rats were denied access to chow for about 65 h. For seven control rats, rat chow was provided ad libitum during the same period. At 09.00 h on the day after the 65 h starvation period, each rat was loosely restrained in a cylindrical wire mesh cage of the same dimensions as that used to accustom the rat to the experimental conditions. A thermocouple was attached to the middle of the ventral side of the tail with surgical tape. The rat was then transferred to a temperaturecontrolled chamber (18 cm ² 18 cm ² 32 cm), wall temperature was kept at 26·0 ± 0·3 °C. Fresh air at a constant 26·0°C was continuously introduced into the chamber at a rate of 1·8 l min¢. A fraction of the air (100 ml min¢) was withdrawn from the chamber and passed through a Zirconia oxygen analyser (LC-700E, Toray, Tokyo). The rats’ oxygen consumption was calculated from the measurements of oxygen content (20189 J l¢). Metabolic heat production (M) was calculated by multiplying the oxygen consumption value by the caloric equivalent for oxygen. Hypothalamic temperature (T ), an index of T , tail skin temperature (T ), T inside the chamber, and the chamber wall temperature were monitored with thermocouples. All parameters were sampled every 5 s through an analogto-digital converter (ADC-12IB, Kanazawa Control Kiki, Kanazawa, Japan) connected to a personal computer (PC_9801VX, NEC, Tokyo). After T and T had stabilized, the intravenous thermode was used to warm and cool the rats. Warming was performed by perfusing water at 44°C through the thermode such that T was gradually increased at a nearly constant rate (•0·04 °C min¢ on average). The T values at which a sharp increase in T occurred was defined as the threshold T for tail skin vasodilatation (Fig. 1) (Sakurada et al. 1993; Shido et al. 1995; Romanovsky et al. 1996). As soon as tail skin vasodilatation was noted, body warming was stopped. Approximately 20 min later, the animals were cooled by perfusing water at 20°C through the thermode (•0·04 °C min¢ on average). The T value at which a sharp rise in M occurred was defined as the threshold T for cold-induced thermogenesis (Fig. 1) (Sakurada et al. 1993; Shido et al. 1995; Romanovsky et al. 1996). The threshold T for cold-induced thermogenesis was also determined with the method of Yeager & Ultsch (1989), since in some cases, the bending point of the M curve was not sufficiently sharp. Whole body cooling was continued for 15 min after the onset of cold-induced thermogenesis in all rats to assess the characteristic of the thermogenic response to the fall in T below its threshold. With the present method, all body core regions were warmed or cooled by manipulating the temperature of venous blood returning through the vena cava, and T was utilized to monitor the changes produced in T . hy
cor
sk
hy
a
sk
hy
hy
sk
hy
hy
hy
hy
cor
hy
cor
Experiment 2: preferred ambient temperature
A total of 26 rats were used. Each animal was anaesthetized with pentobarbital sodium (50 mg kg¢ i.p.), and a temperature transmitter (TA10TA-F40, Data Sciences International, St Paul, MN, USA) was implanted in the peritoneal cavity. After a minimum of 10 days following the operation, 18 rats were placed individually in a thermal gradient (Romanovsky et al. 1996; Sugimoto et al. 1996) at 16.00 h and were kept in the gradient for the following 7 days. Briefly, the main housing of the gradient was an aluminum box 200 cm long, 10 cm wide, and 15 cm high. A long wire mesh cage was placed inside the box. One end of the box was warmed and the other end was cooled with water perfusion devices to establish the thermal gradient (T values inside the ends of the box were 15·0 and 34·0°C). Food pellets were placed on the floor of the cage and water was supplied through four holes in the ceiling a
J. Physiol. 526.2
419
Thermoregulation of starved rats
placed at 40 cm intervals, enabling the animals to have food and water at their T . For the first 4 days, food and water were provided ad libitum for all rats. For the eight starved rats, food was removed from the cage at 16.00 h on day 5 and the animals were denied access to food for the following 3 days. The 10 control rats were allowed free access to food during the same period. The intra-abdominal temperature (T ) of the rats was measured using a biotelemetry system (Data Sciences International, St Paul, MN). The location of the animals in the box was monitored with 18 photoelectric sensors, detecting infrared beams, (PZ_41, Keyence, Osaka, Japan) located 10 cm apart along the thermal gradient. The T and the location of the rats were sampled every minute through the analog-to-digital converter connected to a personal computer (PC_980lVM, NEC, Tokyo). The measurements were made for the last 4 days in the thermal gradient (for 1 day under the ad libitum conditions and for 3 days of the fasting period). The rats’ T values were determined by converting their location to a temperature by utilizing the precalibrated table of T in the gradient as a function of location. Activity levels of the rats were quantified as the number of times the rats interrupted the infrared beams. In an additional experiment, eight rats were individually housed for 7 days in the same box as used for T measurements. However, a thermal gradient was not present, and T inside the box was kept at 25°C. The rats were starved for the last 3 days and their T was measured for the last 4 days. pref
ab
ab
pref
a
pref
a
ab
Data analyses and statistics
The initial values of the thermoregulatory parameters described in experiment 1 were obtained as averages for the 10 min period just prior to the start of body warming. The linear regression equations showing the relationship between T and M in experiment 1 were calculated by the least-squares method. The results are presented as means ± s.e.m. Statistical evaluations of values within a group were assessed by repeated measures two-way analysis of variance hy
–––––––––––––––––––––––––––––
Table 1. Initial thermoregulatory parameter levels
––––––––––––––––––––––––––––– T
T
hy
sk
hy
Experiment 1: thermoeffector thresholds
The mean body masses at the beginning of thermoeffector determinations of the control and starved groups were 324 ± 8 and 299 ± 4 g (significant difference), respectively. Table 1 summarizes the mean initial values of T , T and M just before the start of body warming. The T of the starved rats was significantly lower than that of the controls. The T values in both the control and starved rats were similar to T , indicative of vasoconstriction in the tail. The M value of the starved rats was significantly lower than that of the controls. hy
hy
sk
a
Examples showing changes in T , T and M during body warming and cooling in the control (left) and starved (right) rats. Data are plotted every minute. Open and filled arrows indicate the onset of tail skin vasodilatation and cold-induced thermogenesis, respectively. Stippled and open bars above the abscissae indicate the periods of body warming and cooling, respectively. sk
sk
RESULTS
Figure 1. Determination of thermoeffector thresholds hy
M
(°C) (°C) (W m¦Â) ––––––––––––––––––––––––––––– Control 38·01 ± 0·18 25·7 ± 0·2 55·0 ± 1·2 Starved 37·34 ± 0·08 * 26·0 ± 0·1 51·2 ± 1·2 * ––––––––––––––––––––––––––––– Values are means ± s.e.m. T , hypothalamic temperature; T , tail skin temperature; M, heat production. *Significant difference from control values. ––––––––––––––––––––––––––––– (ANOVA) followed by Student-Newman-Keul’s multiple comparison test. The differences in values between two groups were evaluated by two-way ANOVA followed by Scheffe ’s multiple comparison test. Statistical significance of the difference between the regression coefficients was performed using the F test. The level of significance was considered to be P < 0·05.
sk
420
J. Physiol. 526.2
S. Sakurada and others
Figure 2. Effect of starvation on thermoeffector thresholds Threshold T for tail skin vasodilatation (left) and cold-induced thermogenesis (right) in the control (5; n = 7) and starved ($; n = 8) rats. Values are means + s.e.m. *Significant difference between the control and the starved rats. hy
The threshold T values for tail skin vasodilatation and cold-induced thermogenesis in the control and starved rats are shown in Fig. 2. The threshold T for the tail skin vasodilatation of the starved rats did not differ from that in the controls. However, the threshold T for cold-induced thermogenesis was significantly lower in the starved rats than in the controls by ca 0·75°C. The significant difference in the threshold T values for cold-induced thermogenesis between the starved (37·05 ± 0·12 °C) and control (37·72 ± 0·10 °C) rats was also confirmed when the thresholds were obtained with the method proposed by Yeager & Ultsch (1989). Figure 3 shows the changes in M as a function of T for 15 min after cold-induced thermogenesis commenced in the control and starved rats. The starved rats responded to the fall in T with a significantly sharper rise in M than the controls, i.e. the slopes of the regression lines showing the relationship between T and M were −54·7 and −101·5 W m¦Â°C¢ in the control and starved rats, respectively. hy
hy
hy
hy
hy
hy
hy
Experiment 2: preferred ambient temperature
The mean body masses at the end of the T determinations of the starved and control groups were 308 ± 7 and 363 ± 3 g (significant difference), respectively. Figure 4 shows changes in T for 4 days before and during the 3 day starvation period in the control and starved rats. All rats displayed clear nycthemeral variations in T : T values were maintained at high levels in the dark phase and low levels in the light phase of the day. In the control rats, the pattern of diurnal changes in T was the same for the pref
ab
ab
ab
ab
4 days. However, food deprivation significantly altered the pattern of day—night variations in T . After the first 3 h starvation, T fell profoundly in the light phase. The T levels in the light phase of the starved rats were significantly lower than those of the control rats. Such marked starvation-induced hypothermia in the light phase lasted during the entire period of starvation. In contrast, the T levels in the dark phase were not affected by the 3 day starvation. Figure 5 shows changes in T before and during the 3 day starvation period in the control and starved rats. In the control rats, T showed clear nycthemeral variations: T values were maintained at low levels in the dark phase and high levels in the light phase as is well documented (Gordon, 1994; Sugimoto et al. 1996). Their T values did not differ among the 4 days regardless of the time of the day. In the starved group, the pattern of day—night variations in T before the start of fasting was the same as that in the controls. However, their T level began to increase several hours after commencing the starvation period and continued to rise to the end of the measurements (significant difference in T among days). The T values on the last day of the starvation period were, then, higher than those before the start of starvation by ca 6—10°C. When rats were kept at a constant T , the influence of food deprivation on the pattern of circadian variations of T was slightly different from that observed in the thermal gradient (Fig. 6), e.g. the T levels for the last 1—2 h in the dark phase ab
ab
ab
ab
pref
pref
pref
pref
pref
pref
pref
pref
a
ab
ab
Figure 3. Effect of starvation on thermogenic response
Changes in M as a function of T below the threshold T for M in the control (1; n = 7) and starved (0; n = 8) rats. Values are means of every minute. Vertical and horizontal bars are ± s.e.m. *Significant difference between slopes of the regression lines showing the relationship between T and M of the control and starved rats. hy
hy
hy
J. Physiol. 526.2
421
Thermoregulation of starved rats
Figure 4. Effect of starvation on nycthemeral variations in core temperature in the thermal gradient Mean changes in T before and during the 3 day fasting period in the control (left; n = 10) and starved (right; n = 8) rats. Values are means for every 90 min ± s.e.m. In the light phase of the day, T was significantly lowered by starvation, whereas T in the dark phase (filled horizontal bar) was not affected. ab
ab
ab
were significantly lowered by food deprivation. However, starvation-induced hypothermia was consistent in the light phase of the day and the minimum level of T during food deprivation was comparable with that seen in the thermal gradient. The spontaneous activities of the starved and control rats were shown in Fig. 7. In the control rats, the activity levels did not change for the 4 days in either the light or dark phases of the day. In the starved rats, also, the activity levels in the light phase were similar for the 4 days. However, the 3 day fasting significantly increased the locomotor activity during the dark phase of day. ab
DISCUSSION
especially during the resting (light) phase of the day. Hypothermia decreases metabolism (QÔÑ effect), and thus aids in decreasing energy expenditure during starvation. Indeed, we confirmed that the resting level of M in the starved rats was significantly lower than that of the control rats (Table 1). In contrast, in the dark (active) phase, the T level was not altered by the 3 day food deprivation. Also, activity was significantly increased, which could be interpreted as enhanced food-seeking behaviour (Fig. 7). Increases in voluntary activity are associated with rises in M and T , and may have prevented or masked the starvation-induced hypothermia in the dark phase of the day. The threshold T for thermogenesis of the starved rats was significantly lower than that of the controls, consistent with previous results in birds (Graf et al. 1989; Phillips et al. 1991). A new demonstration was that the threshold T for non-evaporative heat loss response was not affected by food deprivation. Therefore, the zone between the thresholds for heat loss and heat production, the interthreshold zone, was ab
cor
cor
The present study clearly showed that in rats, food deprivation for 3 days produced a profound drop in T under both restrained and freely moving conditions. In agreement with previous studies (Graf et al. 1989; Hohtola et al. 1991; Yoda et al. 2000), the hypothermia was evident
cor
cor
Figure 5. Effect of starvation on nycthemeral variations in preferred ambient temperature
Mean changes in T before and during the 3 day fasting period in the control (left; n = 10) and starved (right; n = 8) rats. Values are means for 90 min ± s.e.m. The T values of each day were significantly different from each other. pref
pref
422
J. Physiol. 526.2
S. Sakurada and others
Figure 6. Effect of starvation on nycthemeral variations in core temperature at a constant ambient temperature Mean changes in T before and during the 3 day fasting period in the starved rats (n = 8) without thermocline. Values are means of 90 min ± s.e.m. In the light phase of the day, T was significantly lowered by starvation, whereas T in the dark phase was not affected. ab
ab
ab
widened by fasting. As soon as T reaches either of these thresholds, the corresponding effector mechanism is activated and T is returned to a temperature between the two thresholds. When T is low, T will fall until the threshold for thermogenesis is reached. At such low temperatures, T will stay near the threshold for heat production, but within the interthreshold zone (Romanovsky et al. 1996). Indeed, in both starved and control rats, resting levels of T were closer to the metabolic threshold than the threshold for heat loss at a T between 26—28°C (Table 1 and Figs 2 and 3: Shido et al. 1995; Romanovsky et al. 1996). The thermoeffector thresholds have been shown to shift under various physiological or pathological conditions. However, the threshold T for heat loss and heat production can move separately as in this study. In rats undergoing endotoxin shock (Romanovsky et al. 1996) or exercise training (Sugimoto et al. 1998), the threshold T for thermogenesis significantly shifted without associated changes in the non-evaporative heat loss threshold. Bacterial endotoxins at a febrile dose (Iriki et al. 1987) or heat acclimation (Shido et al. 1995) significantly altered the threshold for heat loss, but minimally affected thermogenic thresholds. Recently, Kanosue et al. (1998) suggested that in rats, central neuronal networks controlling different autonomic thermoregulatory effectors, such as shivering and cor
cor
a
cor
cor
cor
a
cor
cor
cutaneous vasomotion, work independently of each other. Thus, food deprivation seems to affect only the neuronal network related to thermogenic effectors. The mechanism by which the fasting signals are conveyed to the central nervous system to alter threshold T for heat production remains to be investigated. When the rats were cooled below the threshold T for coldinduced thermogenesis, M increased rapidly. Consistent with the reports in pigeons (Graf et al. 1989), once the threshold was reached, the response of M to the fall in T of the starved rats was enhanced (Fig. 3). In rats, 72 h starvation has been shown to increase uncoupling protein_1 content in the brown adipose tissue mitochondria, one of the thermogenic parameters of the brown adipose tissue (Gianotti et al. 1998). Boss et al. (1997) reported that after 48 h fasting, uncoupling protein_2 mRNA expression in the skeletal muscles, another potent site of non-shivering thermogenesis (Duchamp & Barre, 1989), increased. These observations suggest that the non-shivering thermogenic capacity of food-deprived rats is well preserved. In addition, Yoda et al. (2000) reported that when rats were placed in a cold environment, heat-seeking behaviour was markedly facilitated by 3 day fasting. Taken together, in rats deprived of food for 2—3 days, both autonomic and behavioural colddefence mechanisms can be strongly activated to counteract an excessive fall in T .
Figure 7. Effect of starvation on locomotor activity
cor
cor
hy
cor
Mean spontaneous activities in the dark ($) and light (5) phases of the day before (pre) and during the 3 day fasting period (D1—D3) in the control (left; n = 10) and starved (right; n = 8) rats. *Significant differences from the value before the fasting period.
J. Physiol. 526.2
Thermoregulation of starved rats
The temperature gradient is a good method for studying long-term changes in behavioural thermoregulation, because the test animal can select an optimal T for regulating T simply by moving to a different location. In the present study, the rats were kept in a thermal gradient for 7 days, and their T values were determined without any restrictions. In agreement with previous findings (Gordon, 1994; Sugimoto et al. 1996), the T values of rats were higher in the light phase and lower in the dark phase. In the dark phase, the rats were more active and had higher metabolic rates, and it was assumed that the lower T served to dissipate the increased heat production and thus avoid an excessive increase in T (Gordon, 1994). During fasting, the day—night variations of T persisted. However, the levels of the T of the starved rats gradually increased during the 3 day fasting period in both the light and dark phases. The high T decreases the temperature gradient between the body core and the environment and hence physically reduces the amount of heat dissipation from the body. Since M progressively decreases during starvation (Graf et al. 1989; Munch et al. 1993), a stable T is attained through decreased heat loss. This stable T is, however, lower than that of the controls and thus, during the light phase, leads to the lower tissue metabolism associated with a lower temperature (QÔÑ effect). In addition, we observed that in the starved rats kept at a constant T of 25°C, the T was maintained at the same level as the T of the rats placed in the thermal gradient during the 3 day food deprivation (Figs 4 and 6). These observations suggest that heat-seeking behaviour was not indispensable for regulating T in the starved rats. Without heat-seeking behaviour, however, the starved rats may have activated the autonomic thermogenic system and consumed energy in order to match the loss of heat to maintain a stable T especially during the last 2 days of starvation. It therefore seems that the behavioural mode of thermoregulation, when available, is useful for reducing energy expenditure in starved rats. In summary, food deprivation produced hypothermia, especially in the light phase, due to a downward shift in the threshold core temperature for heat production in rats. Although the threshold for the heat loss response was not altered by fasting, it is argued that the starvation-induced hypothermia was due to a decrease in the regulated body temperature. Nevertheless, autonomic thermogenic responses to excessive decreases in core temperature of the starved rats were well preserved. In a thermal gradient, the rats selected a higher ambient temperature than the controls. However, the core temperature level of such rats was the same as that of the starved rats at a constant ambient temperature. It therefore appears that food-deprived rats possess sufficient ability to regulate core temperature autonomically, and that behavioural thermoregulation, when it is applicable, may contribute to reducing metabolic heat production and thus save energy under food deficit conditions. a
cor
pref
pref
pref
cor
pref
pref
a
cor
cor
a
ab
ab
cor
cor,
423
Akins, C., Thiessen, D. & Cocke, R. (1991).
Lipopolysaccharide increases ambient temperature preference in C57BLÏ6J adult mice. Physiology and Behavior 50, 461—463.
Boss, O., Samec, S., Dulloo, A., Seydoux, J., Muzzin, P. & Giacobino, J.-P. (1997). Tissue-dependent upregulation of rat
uncoupling protein_2 expression in response to fasting and cold. FASEB letters 412, 111—114. Crawshaw, L. I. (1980). Temperature regulation in vertebrate. Annual Review of Physiology 42, 473—491. Crawshaw, L. I. & Stitt, J. T. (1975). Behavioral and autonomic induction of prostaglandin E_1 fever in squirrel monkeys. Journal of Physiology 244, 197—206. Duchamp, C. & Barre, H. (1993). Skeletal muscle as the major site of nonshivering thermogenesis in cold-acclimated ducklings. American Journal of Physiology 265, R1076—1083. Duchamp, C., Barre, H., Delage, D., Rouanet, J. L., Cohen-Adad, F. & Minaire, Y. (1989). Nonshivering thermogenesis and
adaptation to fasting in king penguin chicks. American Journal of Physiology 257, R744—751. Gianotti, M., Clapes, J., Llado, I. & Palou, A. (1998). Effect of 12, 24 and 72 h fasting in thermogenic parameters of rat brown adipose tissue mitochondrial subpopulations. Life Sciences 62, 1889—1899. Gordon, C. J. (1994). 24-hour control of body temperature in rats? Integration of behavioral and autonomic effectors. American Journal of Physiology 267, R71—77. Graf, R., Krishna, S. & Heller, H. C. (1989). Regulated nocturnal hypothermia induced in pigeons by food deprivation. American Journal of Physiology 256, R733—738. Hamilton, C. L. (1959). Effect of food deprivation on thermal behavior of the rat. Proceedings of the Society for Experimental Biology and Medicine 100, 354—356. Hohtola, E., Hissa, R., Pyornila, A., Rintamaki, H. & Saarela, S. (1991). Nocturnal hypothermia in fasting Japanese quail: the
effect of ambient temperature. Physiology and Behavior 49, 563—567. Hoshino, K. (1996). Food deprivation and hypothermia in desynchronized sleep-deprived rats. Brazilian Journal of Medical and Biological Research 29, 41—46. Iriki, M., Hashimoto, M. & Saigusa, T. (1987). Threshold dissociation of thermoregulatory effector responses in febrile rabbits. Canadian Journal of Physiology and Pharmacology 65, 1304—1311. Kanosue, K., Hosono, T., Zhang, Y.-H. & Chen, X.-M. (1998). Neuronal networks controlling thermoregulatory effectors. Progress in Brain Research 115, 49—62. Munch, I. C., Markussen, N. H. & Orirsland, N. A. (1993). Resting oxygen consumption in rats during food restriction, starvation and refeeding. Acta Physiologica Scandinavica 148, 335—340. Nomoto, S. (1997). Pigeon endogenous pyrogen and its pyrogenic effect. In Thermal Physiology. Proceedings of The International Symposium on Thermal Physiology, ed. Nielsen Johannsen, B. & Nielsen, R., pp. 273—275. The August Krogh Institute. Ostheim, J. (1992). Coping with food-limited conditions: Feeding behavior, temperature preference, and nocturnal hypothermia in pigeons. Physiology and Behavior 51, 353—361. Paxinos, G. & Watson, C. (1986). The Rat Brain in Stereotaxic Coordinates. Academic, Australia. Penas, M. & Benito, M. (1986). Regulation of carnitine palmitoyltransferase activity in the liver and brown adipose tissue in the newborn rat: effect of starvation and hypothermia. Biochemical and Biophysical Research Communications 135, 589—596.
424
S. Sakurada and others
Phillips, D. L., Rashotte, M. E. & Henderson, R. P. (1991).
Energetic responses of pigeons during food deprivation and restricted feeding. Physiology and Behavior 50, 195—203. Rashotte, M. E., Basco, P. S. & Henderson, R. P. (1995). Daily cycle in body temperature, metabolic rate, and substrate utilization in pigeons: influence of amount and timing of food consumption. Physiology and Behavior 57, 731—746. Romanovsky, A. A., Shido, O., Sakurada, S., Sugimoto, N. & Nagasaka, T. (1996). Endotoxin shock: thermoregulatory
mechanisms. American Journal of Physiology 270, R693—703.
Sakurada, S., Shido, O., Fujikake, K. & Nagasaka, T. (1993).
Relationship between body core and peripheral temperatures at the onset of thermoregulatory responses in rats. Japanese Journal of Physiology 45, 677—685. Schmidt, I. (1976). Paradoxical changes of respiratory rate elicited by altering rostral brain stem temperature in the pigeon. Pflugers Archiv 367, 111—113. Shido, O., Sakurada, S., Sugimoto, N. & Nagasaka, T. (1995). Shifts of thermoeffector thresholds in heat-acclimated rats. Journal of Physiology 483, 491—497. Simon, E., Simon-Oppermann, C., Hammel, H. T., Kaul, R. & Maggert, J. (1976). Effects of altering rostral brain stem
temperature on temperature regulation in the Adelie penguin, Pygoscelis adeliae. Pflugers Archiv 362, 7—13.
Simon-Oppermann, C., Simon, E., Jessen, C. & Hammel, H. T. (1978). Hypothalamic thermosensitivity in conscious Pekin ducks.
American Journal of Physiology 235, R130—140.
Sugimoto, N., Shido, O., Sakurada, S., Ito, S. & Nagasaka, T. (1998). Changes in core temperature and thermoeffector thresholds
in exercise-trained rats. Japanese Journal of Physiology 48, 163—166. Sugimoto, N., Shido, O., Sakurada, S. & Nagasaka, T. (1996). Daynight variations of behavioral and autonomic thermoregulatory responses to lipopolysaccharide in rats. Japanese Journal of Physiology 46, 451—456. Yeager, D. P. & Ultsch, G. R. (1989). Physiological regulation and conformation: A BASIC program for the determination of critical points. Physiological Zoology 62, 888—907. Yoda, T., Crawshaw, L. I., Yosida, K., Su, L., Hosono, T., Shido, O., Sakurada, S., Fukuda, Y. & Kanosue, K. (2000). Effects of food
deprivation on daily changes in body temperature and behavioral thermoregulation in rats. American Journal of Physiology 278, R134—139.
Acknowledgements
The study was partly supported by Grants-in-aid for Scientific Research from the Ministry of Education, Sciences and Culture of Japan (11670059 and 11770031).
Corresponding author
O. Shido: Department of Physiology, School of Medicine, Shimane Medical University, 89_1, Enya-cho, Izumo 693_8501, Japan. Email:
[email protected]
J. Physiol. 526.2
