Thermoregulatory responses in exercising rats: methodological aspects and relevance to human physiology

Site / method of temperature measurement

The most used theoretical model to explain human thermoregulation consists of a 2-compartment model, which estimates an average body temperature by separating the body temperature into a "core" compartment temperature (which comprises the abdominal, thoracic and cranial cavity temperatures) and a "shell" compartment temperature (which comprises the skin, subcutaneous tissue and muscle temperatures).⁶⁷ Therefore, the mean body temperature of humans is calculated using an equation including weighted T_CORE and T_SKIN values, with the weight attributed to each temperature index being dependent on the environmental conditions.⁶⁷ However, the use of a 2-compartment model is not universally accepted; for example, Jay et al.⁶⁸ proposed the use of a 3-compartment thermometry model that includes the thermal influences of muscles and that appears to yield a more precise estimate of the average body temperature compared with the estimate using a 2-compartment model.

In the case of rats, the average body temperature is not commonly calculated; experiments using this species usually analyze the T_CORE and skin temperatures separately. The T_CORE (or deep body temperature) is measured in the inner tissues of the body, the temperatures of which are not changed by circulatory adjustments that allow heat dissipation to the environment and that affect the thermal shell of the body. ⁶⁹ The brain is generally accepted as a section of the thermal core⁶⁹ even though the brain temperature (T_BRAIN) in several mammalian species may deviate to some extent from the temperature of the remaining thermal core region due to selective brain cooling.^70,71 In contrast, the shell temperature corresponds to the skin and subcutaneous tissue temperature (and possibly the muscle temperature). The skin and mucosal surface temperature may deviate from the T_CORE due to heat exchange with the environment and to changes in the circulatory convection of heat from the core to heat-exchange surfaces.⁶⁹

Although the temperatures of the inner body tissues are not directly changed by circulatory adjustments that allow heat loss to the environment, these temperatures are not homogeneous and do not respond in a similar manner (particularly regarding their time course) to several arousing stimuli.⁷² Baseline T_CORE values measured using rectal / colonic probes are approximately 0.85 to 1.20°C higher than the abdominal temperature (T_ABD) values measured by telemetry.⁷³ Despite this difference in baseline values, both the telemetry and rectal / colonic probe methods are able to reflect the hyperthermic or hypothermic effects induced by different drugs, and high positive correlations were observed between the 2 methods, suggesting that either method could be used in most studies of temperature regulation.⁷³

The exercise-induced changes in the T_COL [or rectal temperature (T_REC)] and the T_ABD have never been simultaneously recorded in an exercising rat. The relationship between these 2 temperatures is relevant to a better understanding of the outcomes of different studies because the first exercise thermoregulation experiments, which were performed in the 1960s, 1970s and 1980s, used the T_COL as an index of the T_CORE. The use of telemetry for measuring the T_ABD as an index of the T_CORE, which started in the 1990s, is certainly more appropriate in studies consisting of long-term measurements, for which the animals must remain undisturbed.⁷⁴

We recently compared the kinetics of exercise-induced increases in the T_BRAIN and T_ABD in rats subjected to treadmill running. At the beginning of constant-speed running at 23-25°C, the T_BRAIN increased more rapidly than did the T_ABD. This observation was true irrespective of whether the T_BRAIN was measured in the cortex (Drummond LR et al., unpublished data), thalamus⁷⁵ (Fig. 2A) or hypothalamus.⁷⁶ This early difference in the rate of increase of the T_CORE indexes likely resulted from intra-brain heat production rather than from the delivery of warm peripheral blood, as indicated by the increase in the brain-abdominal temperature differential. In addition, exercise promotes increased sympathetic outflow to the visceral vascular beds, which induces vasoconstriction and therefore allows a blood flow redistribution that is essential for the organism's ability to meet the increased demands of the active muscles, lungs and brain for oxygen and energetic substrates.⁷⁷ Underperfusion of the abdominal viscera (which is not highly active during exercise) may limit the delivery of the heat produced by the active organs and tissues, thus slowing the rate of increase in the T_ABD and contributing to an increase in the brain-abdomen temperature differential at the beginning of exercise.

Open in a separate window

Figure 2.

Thalamic and abdominal temperatures (A) and the thalamic-abdominal temperature differential of rats (B) subjected to constant-speed treadmill running until fatigue. * denotes a significant difference (p< 0.05) compared with the abdominal temperature. Panel C shows the brain-abdominal temperature differential of running rats at volitional fatigue. The brain temperature was measured at different sites (i.e., the brain cortex, thalamus and hypothalamus). All of these experiments were conducted in temperate environments (ambient temperatures ranging from 24 to 25°C). Please note that these temperature differentials were obtained from experiments with distinct exercise intensities and protocols. The data measured were taken from the experiments reported in the following manuscripts: (i) thalamus - Damasceno WC, Pires W, Lima MR, Lima NR, Wanner SP. The dynamics of physical exercise-induced increases in thalamic and abdominal temperatures are modified by central cholinergic stimulation. Neurosci Lett 2015; 590:193-8. © 2015 John Wiley and Sons. Reproduced by permission of John Wiley and Sons. Permission to reuse must be obtained from the rightsholder; (ii) hypothalamus - Fonseca CG, Pires W, Lima MR, Guimarães JB, Lima NR, Wanner SP. Hypothalamic temperature of rats subjected to treadmill running in a cold environment. PLoS One 2014; 9:e111501. Open-access manuscript; (iii) brain cortex - Drummond LR et al., unpublished data. The rats were subjected to an incremental-speed exercise at an ambient temperature of 25°C. The brain cortex temperature was measured using a thermistor inserted at the following stereotaxic coordinates: anteroposterior: 3 mm anterior to the bregma; mediolateral: 3 mm right to the midline; and dorsoventral: 1.8 mm dorsal to the skull. The abdominal temperature was measured by telemetry. This experiment was approved by the Ethics Commission for Animal Use of the Universidade Federal de Viçosa (Brazil; protocol number 58/2012).

The temperature inside the brain is not homogenous, and a dorsoventral gradient has been reported in resting rats, with higher temperature values observed in ventral areas than in dorsal areas of the brain.^38,78,79 During exercise, the hypothalamic temperature was higher than the T_ABD throughout the exercise period, including a 0.3°C-higher value at volitional fatigue.⁷⁶ Similarly, the thalamic temperature was never lower than the T_ABD during a fatiguing exercise (Fig. 2A). At the beginning of the exercise, the thalamic temperature was 0.7°C higher than the T_ABD, and the largest difference between these T_CORE indexes was 1.1°C, which was recorded as early as the 6th min of exercise. Subsequently, this thalamic-abdomen temperature differential decreased, approaching 0 after 30 min of treadmill running⁷⁵ (Fig. 2B). Interestingly, the T_BRAIN measured in the brain cortex was lower than the T_ABD at the final moments of exercise and at volitional fatigue (Drummond LR et al., unpublished data), which is a quite different response compared with the measurements made in the thalamus or hypothalamus (Fig. 2C). Hasegawa et al. previously demonstrated that the brain cortex temperature was lower than the T_ABD at volitional fatigue.¹⁶ Taken together, these findings suggest that the T_BRAIN value and the brain-abdomen temperature gradient are dependent on the site/depth of T_BRAIN measurement. Moreover, the selective brain cooling phenomenon might occur in specific brain areas rather than the whole brain. Further studies with measurements of the temperature at multiple brain sites in the same exercising rat are warranted.

In our recent studies,^75,76 the T_ABD and T_BRAIN were simultaneously measured using telemetric devices and thermistors, respectively. This fact raises an important methodological issue because telemetric devices have high thermal inertia and may not accurately measure rapid changes in the T_ABD.⁸⁰ Therefore, the methodology employed here may have overestimated the faster increase in the T_BRAIN that occurred in response to exercise initiation. However, our observations support the findings from studies that measured the abdominal aorta and brain temperatures with thermocouples in rats subjected to several arousing stimuli; these rats exhibited faster increases in the T_BRAIN than in the temperature of the abdominal aorta.⁷⁸

Another relevant question regards the site used for measuring the skin temperature. The tail surface is the most common site for measuring the skin temperature, and many investigators associate this temperature with tail-skin blood flow.^47,81 The evaporative heat loss from tail skin is negligible, and, therefore, water evaporation does not influence the T_TAIL value in a running rat. Thus, the T_TAIL measurement is a valid measurement for determining the dry heat loss from the tail skin to the environment. This measurement becomes even more precise when the heat loss index (HLI) is calculated to minimize the effects of the T_AMB or T_CORE on changes in the tail-skin temperature. The heat loss index consists of a ratio between 2 thermal gradients and is calculated by dividing the tail skin-ambient temperature differential by the core-ambient temperature differential. This index varies from 0 (when the T_TAIL is equal to the T_AMB, indicating cutaneous vasoconstriction) to 1 (when the T_TAIL is equal to the T_CORE, indicating vasodilation).^82,83 However, this upper limit is theoretical, and HLI values above 0.7 have not been recorded in a running rat.

Thermocouples are often placed on the lateral surface of the tail to measure the T_SKIN.^15,66,84,85 This position was selected based on observations of the considerable amount of blood flow in the 2 major lateral veins when the tail is heated.⁴⁸ However, measurement of the T_SKIN using thermocouples attached to the dorsal surface of the tail is also typical^17,34,36,53 and provides T_SKIN results similar to those of experiments measuring the temperature at the lateral surface of the tail. In fact, a more recent study histologically re-examined rat tail vascularization and showed the presence of subcutaneous veins in the dorsal aspect of the tail.⁸⁶ The number and size of these veins are not uniform along the entire length of the tail; in the proximal segment, where the T_TAIL is usually measured, a small vein is present on each side close to the tail midline.⁸⁶

Influence of exercise intensity, duration and protocol

The exercise intensity, duration and protocol are among the factors that interfere with the exercise-induced changes in body temperature. In experiments performed in the treadmill setup, the exercise intensity is dependent on the treadmill speed or incline, whereas the exercise duration is determined by the running time, and the exercise protocol refers to the possible combinations of intensity and duration (e.g., constant-speed exercise, incremental-speed exercise and intermittent exercise).

The finding that the magnitude of hyperthermia is dependent on the exercise intensity seems obvious. To guarantee an adequate energy supply for active muscles during physical exertion, the body metabolism is accelerated based on the absolute exercise intensity. The absolute exercise intensity refers to an exercise intensity that requires a given amount of energy expenditure, expressed as kcal/min or mLO₂/min (e.g., a rat running at a constant speed of 15 m/min). As a byproduct of the augmented metabolic rate, heat is generated in the contracting muscles and then dissipated to the other body tissues by conduction or through circulating blood. In humans, the T_CORE increases in proportion to the work intensity over a wide range of T_AMB values.⁸⁷

In running rodents, the magnitude of exercise-induced increases in the T_CORE is also positively associated with the exercise intensity. This positive association indicates that a higher exercise intensity is related to a higher increase in the T_CORE. However, this is a not a universal finding; some researchers⁵⁷ have failed to reliably demonstrate the proportional relationship between the steady-state T_CORE and workload that is so clearly found in humans.^87-89 In fact, intensity-dependent hyperthermia is clearly observed during a single exercise bout consisting of multi-stage, incremental-speed treadmill running.^90,91 However, the effects of the exercise duration and of accumulated physical exertion, particularly at higher speeds, on the magnitude of hyperthermia cannot be ruled out in this exercise protocol.

Gollnick and Ianuzzo⁹⁰ performed one of the first experiments that measured the T_CORE in running rats. These researchers showed that the T_COL values detected during exercise were higher than the pre-exercise values and that the T_COL increased as a function of the exercise intensity during incremental-speed treadmill running. Moreover, an individual recording presented in their manuscript clearly showed that the T_COL plateaued while the rat ran at 1.0 mph and 1.33 mph. In contrast, a sharp increase in temperature was observed at a running speed of 2.0 mph, with the animals becoming fatigued within a short period of time.⁹⁰

The above-mentioned findings were reproduced in subsequent studies. Shellock and Rubin⁵⁷ observed increases in both the T_COL and rate of oxygen consumption (VO₂) during successive incremental exercise work rates at cooler ambient temperatures, with a highly linear relationship between the T_COL and VO₂ (pooled correlation coefficient of 0.91). More recently, Hasegawa et al.⁹¹ subjected rats to 1 h of incremental-speed treadmill running at 23°C. The treadmill speed was increased every 20 min (10, 20 and 26 m/min). Relative to the resting T_CORE, the T_CORE immediately increased with exercise at the lowest exercise intensity (10 m/min). In addition, treadmill running increased the T_CORE in a linear manner at the second (20 m/min) and final (26 m/min) stages of exercise such that the T_CORE at the end of each exercise stage was significantly increased from that at the previous stage. Similar findings were observed with the VO₂ measurements.⁹¹

However, when rats were subjected to a single-stage exercise in the study of Shellock and Rubin,⁵⁷ no statistically significant increases in the T_COL related to the progressively higher treadmill speeds or levels of VO₂ were observed. At the 16th minute of exercise, the rats subjected to running at a constant speed of 18 m/min and no incline had a T_COL similar to that of rats running at 25 m/min and 5% incline.⁵⁷ Therefore, the researchers concluded that rats do not reliably exhibit the proportional relationship between the steady-state T_CORE and work. However,⁵⁷ some research groups have identified an association between hyperthermia and exercise intensity during single-stage treadmill running.^37,38,92 For example, a trend toward higher T_COL with increasing speed was evident in the study conducted by Wilson et al.³⁷ In addition, the increase in T_BRAIN clearly differed among the 3 running speeds (18, 21 and 24 m/min), as evaluated by Kunstetter et al.³⁸ (Fig. 3). At the speed of 18 m/min, equilibrium was attained between the rates of heat production and heat loss, and a plateau in the T_BRAIN was therefore observed. In contrast, a temperature plateau was never observed at 21 and 24 m/min.³⁸ These high intensities likely provoked elevated rates of heat production that overcame the ability of rats to dissipate heat, causing the T_BRAIN to constantly increase during exercise.

Open in a separate window

Figure 3.

Cortical brain temperature of rats that were subjected to treadmill running until fatigue at 3 different speeds (18, 21, and 24 m/min). The ambient temperature was controlled at 25.2 ± 0.2°C by the use of air conditioning. * denotes a significant difference (p< 0.05) compared with the 21 m/min and 18 m/min trials. + denotes a significant difference (p < 0.05) compared with the 18 m/min trial. The data used to plot this graph were taken from the control experiments reported in the following manuscript: Kunstetter AC, Wanner SP, Madeira LG, Wilke CF, Rodrigues LO, Lima NR. Association between the increase in brain temperature and physical performance at different exercise intensities and protocols in a temperate environment. Braz J Med Biol Res 2014; 47:679-88. Open-access manuscript.

The initial increase observed in the T_CORE of rats in response to physical exercise appears to not be specific to the exercise protocol and intensity (Fig. 3). When measuring the T_BRAIN of rats subjected to 3 different running speeds, we observed similar rates of T_BRAIN increases during the first 8 minutes of exercise.³⁸ During the initial phase of exercise, i.e., up to the 8^th minute, the rats exhibited the highest rates of T_BRAIN increases, which were independent of the treadmill speed. This marked T_BRAIN increase was likely caused by animal handling, a stressful procedure that includes inserting the brain thermistor prior to physical exertion.

Because the magnitude of hyperthermia in rats depends on the exercise intensity, as demonstrated by most studies,^37,38,92 the next important issue to address is whether this hyperthermia magnitude is determined by the absolute or relative exercise intensity. The explanation for the absolute exercise intensity was provided earlier. The relative exercise intensity refers to an exercise intensity relativized by the maximum rate of oxygen consumption (%VO_2MAX) or the maximum treadmill speed attained by an individual or animal.

In a seminal study performed with humans, Saltin and Hermansen⁹³ described that at a given room temperature, the changes in the T_CORE are related to the relative exercise intensity, which is expressed as a percentage of the VO_2MAX. The findings of Saltin and Hermansen⁹³ were re-examined almost 50 y later. Gant et al.⁹⁴ confirmed that the magnitude of hyperthermia in humans subjected to treadmill running was associated with the relative rather than the absolute exercise intensity. When comparing the physiological responses of 2 groups (i.e., one group with moderate VO_2MAX and another with higher VO_2MAX) during a 60-min period of running at 60% of the VO_2MAX, no differences were observed in the exercise-induced increases in the T_CORE and heart rate.⁹⁴ However, when these 2 groups were subjected to the same running speed (similar absolute exercise intensity), the group with moderate VO_2MAX exhibited higher increases in the T_CORE and heart rate than did the group with high VO_2MAX.⁹⁴

Tanaka et al.⁵³ performed a very well-designed study to characterize the thermoregulatory responses of running rats relative to the work intensity, which was expressed as a percentage of the VO_2MAX. At a T_AMB of 24°C, the T_REC at the beginning of tail-skin vasodilation increased in proportion to the exercise intensity at a range from 45% to 75% of the VO_2MAX; at intensities higher than 75% of the VO_2MAX, this linear, positive association levelled off. After tail vasodilation, the T_REC remained steady and was also proportional to the work intensity (in the same range as described above) at the end of the 30-min period of exercise. Thus, the T_CORE of running rats increases in proportion to the relative exercise intensity.

Exercise thermoregulation is also influenced by the duration of treadmill running. The changes in the T_CORE, heat production and heat loss that occur during the initial minutes of exercise have been previously discussed in detail; therefore, we will now highlight the effects of the duration of exercises lasting more than 30 min. Even when performed under conditions of compensable heat stress, the T_CORE begins to increase again after a plateau is reached. For example, Kunstetter et al.³⁸ reported that rats subjected to a constant running speed of 18 m/min exhibited a plateau in the T_BRAIN that lasted approximately 40 min and was followed by a second, clear increase in the T_BRAIN. This second increase in the T_CORE may result from either a gradual reduction in the running economy (i.e., the ratio of the power output to VO₂), which causes a subsequent increase in the heat production rate, or to augmented exposure to electrical stimulation, which causes a subsequent increase in the sympathetic outflow that promotes heat conservation and thermogenesis.

Another important factor that influences exercise thermoregulation is the protocol employed during treadmill running. In a comparison of the thermal responses between incremental and constant-speed running, the T_BRAIN from the 24th minute to the end of the incremental running protocol was lower than the T_BRAIN during constant exercise at 24 m/min.³⁸ The influence of the running protocol on the T_BRAIN increase may result from differences in the evolution of exercise intensity, which is an inherent characteristic of each running protocol. During the initial stages of the incremental exercise protocol, the power output by the animals and, consequently, the rate of heat production were low (e.g., it took 42 min for rats to begin running at 24 m/min). However, even when the animals achieved high speeds during the final stages of the incremental exercise protocol, their T_BRAIN values remained lower.

Collectively, the data indicate the influence of exercise intensity, duration and protocol on the changes in the body temperature of rats subjected to treadmill running. Therefore, a researcher should be aware of the outcomes of his/her choices before selecting the most suitable exercise. If a marked increase in T_CORE is desirable to investigate the association between thermoregulation and physical performance, incremental-speed treadmill running is not the most suitable experimental protocol.

Influence of the environmental conditions

Despite the importance of the 3 factors mentioned in the previous section, environmental conditions, particularly the T_AMB, are the most important modulators of changes in the body temperatures of exercising rats.

The dry ambient temperature is one of the parameters that is evaluated to determine the environmental conditions. Aside from temperature, the relative humidity, wind speed and radiation represent alternative parameters that modulate the heat exchange between a body and the ambient and thus contribute to the overall environmental thermal stress. In experiments with resting rats, the environmental conditions are usually classified as thermoneutral, subneutral or supraneutral.⁸² According to the more recent glossary of terms for thermal physiology, the term thermoneutral zone (TNZ) refers to the range of ambient temperatures at which temperature regulation is achieved only by the control of sensible heat loss, without regulatory changes in metabolic heat production or evaporative heat loss.⁶⁹ In this context, the use of terms such as cool, cold, moderate, warm and hot is not adequate for describing an environmental condition; these terms are suitable for describing a thermal sensation.

Several research groups have attempted to identify the range of ambient temperatures associated with the TNZ. Nevertheless, because a neutral T_AMB measured in a given experimental setup cannot be used as a standard for experiments conducted in other experimental setups,⁸² the results reported by different laboratories are contradictory. Several research groups^95,96 determined the rat TNZ to be between 28 and 34°C, but other groups reported lower temperature ranges from 22 to 27°C⁹⁷ or from 18 to 28°C.⁹⁸ Accounting for the more recent concept of the TNZ,⁶⁹ Romanovsky et al.⁸² suggested the use of skin thermometry, which is a definition-based, simple, and inexpensive technique, to determine whether the experimental conditions are neutral, subneutral, or supraneutral for a given animal. Using skin thermometry and under the conditions tested (rats placed in confiners, without bedding or filter tops and without the possibility of group thermoregulation), the ambient temperature range of 29.5 to 30.5°C satisfied the TNZ criteria in Wistar rats.⁸²

Using the method proposed by Romanovsky et al.,⁸² we observed HLI values ranging from 0.20 to 0.30 in a resting rat maintained inside the chamber that contained the treadmill belt (with the electrical fan off) at a local T_AMB of 24-25°C.³⁶ Similar findings (average HLI values of 0.23) were reported by Lima et al.,⁹⁹ who maintained the rats inside the same chamber, but with the electrical fan on and at a local temperature of 26°C.³⁶ Collectively, these data suggest that ambient temperatures ranging from 24 to 26°C correspond to the lower extremity of the TNZ of rats in the treadmill setup.

The concept of thermoneutrality does not apply to ectotherms (i.e., bradymetabolic species) and is not suitable for exercising endotherms because the increased metabolic rate caused by muscular contractions is an inherent characteristic of physical exercise. It would be counterintuitive to state that a rat runs under thermoneutral conditions while displaying a T_CORE value of 39°C and a metabolic rate that is five-fold higher than the resting metabolic rate. Considering the latter statement, authors usually do not describe the environment as thermoneutral during exercise experiments and use alternative descriptions, including temperate environments³⁸ or normal T_AMB.^100,101

The increase in the T_CORE of exercising humans is largely dependent on the power output but is largely independent of a wide range of ambient temperatures.^87,88,102 The term prescriptive zone is used to describe the range of ambient temperatures at which humans can achieve a steady-state response for the T_CORE, with the level of steady-state T_CORE dependent on the metabolic rate (either absolute or relative exercise intensity) and independent of the environment within a range of compensable heat stress conditions. Of note, the linear relationship between the metabolic rate and the T_CORE is valid for a given person (controlling for factors such as heat acclimation and hydration status) but does not always hold for comparisons between different people.

The existence of a prescriptive zone in exercising rats is unlikely and has never been reported. Even when varying the ambient temperature by a few °C in our previous experiments, we could not observe the existence of a prescriptive zone in the response of the T_CORE of rats to treadmill running. As illustrated in Figure 4, small alterations in the T_AMB (a reduction from 15°C to 8°C) have altered the profile of the T_CORE response in rats exercising at the same absolute intensity, with the increase in the T_CORE being replaced by a decrease.¹⁰³ Indeed, the data collected from different studies^76,103,104 clearly show that the magnitude of the T_CORE change induced by fatiguing treadmill running is strongly determined by the T_AMB (Fig. 5). This relationship was observed at running speeds ranging from 20 to 21 m/min (r² = 0.97; p < 0.001). The fact that the exercise-induced thermoregulatory responses in rats are more sensitive to the T_AMB than the responses in humans is explained by differences in their body size. Small animals exhibit a high body surface area-to-mass ratio,^105,106 which facilitates heat exchange with the environment.

Open in a separate window

Figure 4.

Abdominal temperature of rats subjected to treadmill running until fatigue at a constant speed of 20 m/min at 3 ambient temperatures (i.e., 8, 12 and 15°C). * denotes a significant difference (p< 0.05) compared with the 12°C trial. + denotes a significant difference (p < 0.05) compared with the 15°C trial. The graph was modified from: Guimarães JB, Wanner SP, Machado SC, Lima MR, Cordeiro LM, Pires W, La Guardia RB, Silami-Garcia E, Rodrigues LO, Lima NR. Fatigue is mediated by cholinoceptors within the ventromedial hypothalamus independent of changes in core temperature. Scand J Med Sci Sports 2013; 23:46-56. Copyright © 2013 John Wiley and Sons. Used with permission.

Open in a separate window

Figure 5.

Correlation between the ambient temperature and the change in the abdominal temperature of rats subjected to fatiguing treadmill running. The treadmill speed was set to 20-21 m/min, and the inclination was set to 0-5%. The criterion to determine fatigue was the same in these 3 studies, i.e., the inability to keep the pace with the treadmill, as indicated by exposure to electrical stimulation for 10 consecutive seconds. The data used to plot this graph were taken from experiments reported in the following manuscripts: (i) Rodrigues LO, Oliveira A, Lima NR, Machado-Moreira CA. Heat storage rate and acute fatigue in rats. Braz J Med Biol Res 2003; 36:131-5. Open-access manuscript. (ii) Guimarães JB, Wanner SP, Machado SC, Lima MR, Cordeiro LM, Pires W, La Guardia RB, Silami-Garcia E, Rodrigues LO, Lima NR. Fatigue is mediated by cholinoceptors within the ventromedial hypothalamus independent of changes in core temperature. Scand J Med Sci Sports 2013; 23:46-56. Copyright © 2013 John Wiley and Sons. Used with permission; (iii) Fonseca CG, Pires W, Lima MR, Guimarães JB, Lima NR, Wanner SP. Hypothalamic temperature of rats subjected to treadmill running in a cold environment. PLoS One 2014; 9:e111501. Open-access manuscript.

Another relevant aspect regarding the environmental conditions is the wind speed. Some setups are designed with an electrical fan in front of the treadmill belt, whereas others do not have this electrical fan. The airflow generated by the electrical fan is expected to facilitate cutaneous heat loss through convection.¹¹ Until a systematic study on the effects of the wind speed on the thermoregulation of running rats is performed, investigators should report whether their exercise setups include an electrical fan and, if so, they should also report the wind speed.

Finally, the effect of different relative humidities on the thermoregulatory responses of exercising rats has never been systemically investigated. The lack of studies on this topic may result from the limited evaporation capacity of rats; Gordon¹⁰⁷ reported that the evaporation capacity of rats per unit of body mass is half of that of humans. In fact, modern humans possess an extraordinary number of eccrine glands that produce a large amount of watery sweat.¹⁰⁸ This improved sweating and the resulting higher evaporative capacity are among the anatomo-physiological features that make humans better adapted for long-distance locomotion rather than rapid locomotion and for dissipating heat rather than retaining heat.¹⁰⁹ Therefore, a higher relative humidity impairs the prolonged physical performance of humans subjected to cycling in hot environments compared with their performance at a lower relative humidity.¹¹⁰

Evaporation losses in rodents are dependent on the association between an autonomic (saliva production) and a behavioral thermoeffector (saliva spread onto body surfaces). Saliva evaporation can effectively maintain the T_CORE of a rat exposed to heat,⁵¹ but this process is not effective in an exercising rat because the animal cannot stop running to spread saliva onto its body surface.^37,57 Nevertheless, despite the limited evaporative capacity of exercising rats, the evaporative water losses (likely water evaporated from the respiratory tract) increase in proportion to the work intensity during 30 min of treadmill running.⁵³ This finding warrants future investigations designed to evaluate the impact of different relative humidities on the exercise thermoregulation of running rats.

The influence of pre-exercise exposure to the treadmill setup

Guaranteeing that rats are not stressed before exercise initiation and determining the resting body temperature values of rats are some of the concerns faced by investigators that study exercise physiology. For example, undisturbed T_CORE values of a rat at exercise initiation are essential to obtain recordings showing clear exercise hyperthermia. These resting values are even more interesting if they are recorded while the rats are resting in the treadmill setup. Nevertheless, obtaining such ideal resting values can be challenging and may confound the experimental outcomes.

After being handled by an experimenter for the insertion of a thermistor that allows T_BRAIN measurement, rats display marked but transient hyperthermia: the T_BRAIN increases at a rapid rate, peaking at ∼20 min at a value 1.0°C above the initial values, and then decreases slowly toward the initial values (Fig. 6). When rats are placed on the treadmill setup before exercise initiation, a consistent and marked increase in the T_CORE is also observed (Fig. 7). This increase in the T_CORE may be a consequence of emotional hyperthermia or may be a conditioned or anticipatory response. The hyperthermia that occurs prior to exercise was first described by Gollnick and Ianuzzo;⁹⁰ when the rats were placed on stationary treadmills, their resting T_COL values increased to 39.4°C before the treadmill was turned on. The authors suggested that the pre-exercise hyperthermia represented an anticipatory response that would be elicited by the exercise itself or by the use of electrical stimulation early in the training program. Gollnick and Ianuzzo also suggested that the hyperthermic response would be ablated as the rats become acclimated to the exercise conditions.⁹⁰ This pre-exercise hyperthermia was consistently observed in our experiments even though our rats were familiarized with treadmill running over a period of 5 consecutive days and were minimally exposed to electrical stimulation during the last familiarization sessions.

Open in a separate window

Figure 6.

Change in brain temperature induced by manual handling and the insertion of the thermistor for recording the brain cortex temperature (A). The rats were kept in their home cages, and the ambient temperature was controlled at 25.2 ± 0.2°C by the use of air conditioning. The data used to plot this graph were taken from the control experiments reported in the following manuscript: Kunstetter AC, Wanner SP, Madeira LG, Wilke CF, Rodrigues LO, Lima NR. Association between the increase in brain temperature and physical performance at different exercise intensities and protocols in a temperate environment. Braz J Med Biol Res 2014; 47:679-88. Open-access manuscript.

Open in a separate window

Figure 7.

Change in the abdominal temperature of a representative rat that was placed on a treadmill (blue) and 2 h later was subjected to constant-speed treadmill running at 18 m/min (red). A thermocouple remained attached to the rat's tail, and a microinjection needle was placed inside a brain guide cannula throughout the experiment. The ambient temperature was controlled at 24.1 ± 0.4°C by the use of air conditioning. The data used to plot this graph were taken from the experiments reported in the following manuscript: Wanner SP, Leite LH, Guimarães JB, Coimbra CC. Increased brain L-arginine availability facilitates cutaneous heat loss induced by running exercise. Clin Exp Pharmacol Physiol 2015; 42:609-16. © John Wiley and Sons. Reproduced by permission of John Wiley and Sons. Permission to reuse must be obtained from the rightsholder.

In experiments in which rats are placed on a treadmill setup prior to exercise initiation, the animals exhibit average pre-exercise T_ABD values higher than 37°C^17,111 and, in some cases, close to 38°C.³⁶ In contrast, when the rats are transferred from their home cages directly to the treadmill, their T_ABD values are usually lower than 37°C.⁷⁵ The impact of these initial T_CORE differences on physical performance, the cutaneous heat loss threshold, and the T_CORE at volitional fatigue, among other physiological parameters, has not been investigated.

Interestingly, in some experiments, the increase in the T_CORE observed prior to treadmill running is similar to or even higher than the running-induced increase in the T_CORE (Fig. 7; the maximal T_CORE increase during rest and during treadmill exercise was 1.46 and 1.56°C, respectively). This pre-exercise hyperthermia may be physiologically relevant because a state of cognitive fatigue prior to exertion impairs physical performance.^112,113 In some cases, the T_ABD of a rat will not return to the baseline values during a 3-h period of exposure to the treadmill, particularly when a thermocouple is attached to the tail surface of a rat to measure the T_TAIL. When performing the experiments for a recently published manuscript,³⁶ we aborted 35.5% of the experimental trials because the rats exhibited constantly high T_ABD values during the treadmill exposure period that preceded running. Taking our experience into account, we recommend that other authors do not expose rats to the treadmill setup in the moments prior to exercise in experiments designed for body temperature measurements even though handling prior to exercise may confound the thermoregulatory response. Future studies should be conducted to further investigate which experimental protocol yields less "noise" when measuring the body temperatures of a running rat.

Influence of the time of day

The circadian oscillations of the neuroendocrine and autonomic nervous systems directly affect the T_CORE.¹¹⁴ In animals with nocturnal habits, including laboratory rats, T_CORE peaks are observed during the early hours of environmental darkness; in contrast, the lowest T_CORE values are observed during the light phase of the day.¹¹⁵ The circadian rhythm of the T_CORE is closely associated with the circadian rhythm of spontaneous locomotor activity¹¹⁶ such that metabolic heat production and, consequently, body heat storage increase during the dark phase partially due to an increased metabolic rate in the contracting muscles. In addition, changes in autonomic activity throughout the day modulate the T_CORE independently of the locomotor activity rhythm¹¹⁷ as the temperature rhythm also reflects the combined effects of the body clock, sleep, feeding, and mental activity.^{118 119} Because the circadian rhythm significantly affects thermoregulation, it would be expected that the exercise-induced changes in body temperature would be dependent on the time of the day at which a laboratory rodent is subjected to physical exertion.

Machado et al.¹²⁰ recently conducted experiments with rats, which were subjected to incremental treadmill running during the beginning of the light or dark phase of the day. As previously observed by other authors,¹²¹ the pre-exercise T_CORE values were lower during the light phase than during the dark phase.¹²⁰ Interestingly, no differences were observed in the T_CORE at volitional fatigue between the exercises performed at the 2 phases. This finding suggests that rats show improved heat loss mechanisms and/or running economy in the dark phase. Furthermore, Tanaka et al.¹²¹ observed higher or unaltered oxygen consumption depending on the exercise intensity but a short latency time for the T_TAIL increase during the dark phase; these findings rule out the hypothesis of improved running economy (would be associated with reduced VO₂) and support the notion that accelerated heat loss may have contributed to the slower increase in the T_CORE during exercise at night. An unexpected outcome of the study conducted by Machado et al.¹²⁰ was that physical performance, as measured by the running time to fatigue, was higher during the light phase, which is the phase of the day when rats are less active. An interesting hypothesis to explain the accelerated fatigue in the dark phase is that high pre-exercise T_CORE values are associated with reduced prolonged (aerobic) physical performance in both rats and humans,^80,122 particularly in warm environments.