Ambient air temperatures (Ta) of <6°C or >29°C have been shown to induce large changes in arterial blood pressure and heart rate in homeotherms. The present study was designed to investigate whether small incremental changes in Ta, such as those found in typical laboratory settings, would have an impact on blood pressure and other cardiovascular parameters in mice and rats. We predicted that small decreases in Ta would impact the cardiovascular parameters of mice more than rats due to the increased thermogenic demands resulting from a greater surface area-to-volume ratio in mice relative to rats. Cardiovascular parameters were measured with radiotelemetry in mice and rats that were housed in temperature-controlled environments. The animals were exposed to different Ta every 72 h, beginning at 30°C and incrementally decreasing by 4°C at each time interval to 18°C and then incrementally increasing back up to 30°C. As Ta decreased, mean blood pressure, heart rate, and pulse pressure increased significantly for both mice (1.6 mmHg/°C, 14.4 beats·min−1·°C−1, and 0.8 mmHg/°C, respectively) and rats (1.2 mmHg/°C, 8.1 beats·min−1·°C−1, and 0.8 mmHg/°C, respectively). Thus small changes in Ta significantly impact the cardiovascular parameters of both rats and mice, with mice demonstrating a greater sensitivity to these Ta changes.
- blood pressure
- heart rate
- standard deviation of the interbeat interval
the external environment can have a substantial impact on the cardiovascular system. Exposure of humans (1, 7, 18, 32), rats (2, 3, 14, 20), and mice (22) to cold ambient temperature (Ta) results in elevated blood pressure and heart rate. It appears as if the tachycardia and hypertension are the indirect result of sympathetic nervous system (SNS) activation of thermoregulatory mechanisms because elevated plasma norepinephrine (NE) levels correlate with elevated blood pressure in the cold (7, 14, 17, 32). Furthermore, propranolol, a β-adrenergic blocking agent, can blunt or completely reverse the cardiovascular effects of cold exposure (21). Cold-induced activation of the SNS, in turn, appears to elevate blood pressure through activation of the renin-angiotensin system (RAS). Blockade of the RAS systemically (19), centrally (15, 16, 22), or genetically in an angiotensinogen knockout mouse model (23) blunts or prevents cold-induced hypertension, suggesting that RAS signaling pathways are necessary for Ta-induced effects on blood pressure.
Elevation of Ta beyond typical housing temperatures also impacts the cardiovascular system. Warming rats (31) and mice (30, 31) from 23°C to 28–31°C, closer to, if not within, their thermoneutral zone (TNZ) results in a drop in heart rate and blood pressure. Metabolic rate in homeotherms is at its minimum in the TNZ, which is approximately 28–31°C for rodents (4). Animals with smaller body sizes have higher surface area-to-volume ratios and thus typically exhibit warmer TNZs. At temperatures below the TNZ, nonshivering thermogenesis is stimulated by increased sympathetic activity to offset the increased rate of heat loss (5). Indeed, it appears that housing mice at 22°C, typical for most research laboratories, results in cold stress, in the form of mild hypertension, presumably due to elevated thermogenic demand (30, 31). What remains unclear, however, is whether less dramatic changes in Ta, such as those that might occur on a daily basis within an animal facility, or differences in housing temperatures between different laboratories, can significantly influence cardiovascular parameters of small rodents.
As mice are roughly 10 times smaller than rats and therefore have a much greater rate of heat exchange relative to heat production than rats, we hypothesized that the cardiovascular parameters of mice would be more sensitive to small changes in Ta than the cardiovascular parameters of rats. Our present study was designed to determine if small (4°C) changes in Ta within the typical housing and experimental temperature range of 18–30°C have a differential impact on blood pressure and other cardiovascular parameters in mice and in rats. We address this question using an implantable radiotelemetry system to monitor blood pressure in freely moving conscious rodents.
MATERIALS AND METHODS
Female NIH Swiss mice weighing 25–28 g and female Sprague-Dawley rats weighing ∼250 g were obtained from Harlan. Animals were maintained on a 12:12-h light-dark cycle (dark from 5 PM to 5 AM). Animal studies were approved by each of the local institutional animal care and use committees.
Implantation of blood pressure telemeters.
On receipt of the animals, the mice were housed at 30°C and the rats were housed at 23°C. Seven mice were anesthetized initially with 5% isoflurane in an oxygen stream and maintained on 1–2% isoflurane. Mice were kept on a heating pad (38°C) throughout implantation of the blood pressure telemeter (PAC20; Data Sciences International) in the left common carotid artery (24). Implantation of the blood pressure telemeter into the abdominal aorta of seven rats (PAC40; Data Sciences International) was performed as described previously (24). Mice and rats were maintained on a heating pad for 48 h after the surgery and then housed individually at 30°C for 1 wk to allow time for recovery.
Cardiovascular data collection.
Data from the blood pressure telemeters were recorded at 500 Hz. Between 5 PM and 4 PM on the next day, 2-min data streams were obtained every 10 min. From the pressure waveform analysis, the following cardiovascular parameters were obtained: heart rate, systolic blood pressure, diastolic blood pressure, mean blood pressure, pulse pressure, and the standard deviation of the interbeat interval (SDIBI). Activity of the animals was also monitored. Between 4 PM and 5 PM, data were not taken while the animals were cared for (watered, fed, bedding change, etc.) The dark cycle cardiovascular parameters were averaged from data collected between 5 PM and 4 AM, and the light cycle parameters were averaged from data collected between 5 AM and 4 PM. Ta was set initially at 30°C. Every 3 days, the temperature was lowered by 4°C. Data reported are the averages of two consecutive dark cycles or two consecutive light cycles, on the 2nd and 3rd days at that specific Ta. In a second set of experiments, female mice and rats of the same strain and size were implanted with telemeters as above and taken through a series of Ta changes, from 30°C to 18°C and back to 30°C. In these animals (n = 7 for both mice and rats), oxygen consumption was measured in addition to the cardiovascular parameters as described previously (2, 28).
Data are reported as means and SE. Regression analysis was performed for each measured parameter as a function of Ta. Statistical significance of these regressions was determined by Pearson's r correlation coefficient for significance levels. For comparing the cardiovascular parameters between mice and rats, the slopes from the best fit line for each cardiovascular parameter (mean blood pressure, heart rate, pulse pressure, metabolic rate, or SDIBI) relative to Ta were picked for each animal for the dark cycle, light cycle, and the 24-h period. The slopes generated from these best linear fits were averaged. A Student's t-test was used to compare the average of the slopes between rats and mice.
Effect of Ta on the cardiovascular parameters of rats.
Cardiovascular parameters and metabolic rate were measured in rats over a series of Ta, ranging from 18 to 30°C. At any given Ta, blood pressure, heart rate, oxygen consumption, and activity were elevated in the dark cycle relative to the light cycle, as has been seen previously (6, 10, 29). All of the cardiovascular and metabolic parameters varied linearly with ambient housing temperature (Fig. 1, A–E). This relationship between rat cardiovascular/metabolic parameters and Ta was observed in both the light and dark cycles. However, spontaneous cage activity was not impacted by Ta (Fig. 1F), suggesting that 1) the rats were not utilizing spontaneous cage activity as a thermoregulatory response to altered Ta, and 2) the temperature-dependent cardiovascular and metabolic parameters measured can be independent of activity in rats. Under constant-temperature conditions of 23°C, cardiovascular parameters and metabolic rate did not change in rats or mice over the same period of time (data not shown).
Effect of Ta on the cardiovascular parameters of mice.
The cardiovascular parameters and metabolic rate in mice also exhibited a circadian rhythm, with higher heart rate, blood pressure, oxygen consumption, and activity in the dark cycle, as has been observed previously in mice (9, 24, 25, 28, 31). Similar to the relationship observed in the rats, cardiovascular parameters and metabolic rate of mice also varied linearly with ambient housing temperatures (Fig. 2, A–E), during both the dark and light cycles. Importantly, spontaneous cage activity of these mice decreased with increasing Ta in the light cycle (Fig. 2F).
Comparison of rat vs. mouse.
One of the primary objectives of this study was to test the hypothesis that mouse metabolic and cardiovascular parameters would be more sensitive to changes in Ta than those parameters of rats. The mean arterial pressure-Ta relationship in the mouse exhibited a significantly greater slope of ∼1.6 mmHg/°C than that of the rat, with a slope of 1.2 mmHg/°C (P < 0.05 for rat vs. mouse, Fig. 3A). Considering that mice and rats have similar blood pressures at 22°C (approximately 100 and 102 mmHg, respectively), the difference in these slopes represents a 25% greater sensitivity in the mouse. The absolute heart rate in the mouse was also more sensitive to changes in Ta than that of the rat (14.4 vs. 8.1 beats·min−1·°C−1, respectively, P < 0.05 for rat vs. mouse, see Fig. 3B). Because mice and rats have considerably different heart rates at 22°C (645 vs. 384 beats/min, respectively), the relative change in the mouse heart rate, 2.2%/°C (relative to 22°C), was not different from that of the rat of 2.1%/°C (relative to 22°C). The slope of the SDIBI-temperature relationship for mice (0.33 ms/°C) was significantly greater (P < 0.05 for rat vs. mouse) than that of the rat (0.17 ms/°C), suggesting an increased sensitivity of the autonomic nervous system of the mouse relative to the rat (Fig. 3C). The relationship between mass-specific metabolic rate and Ta was significantly greater in mice than in rats (P < 0.05 for rat vs. mouse, Fig. 3E). The change in pulse pressure with Ta was not different between mice and rats, at 0.8 mmHg/°C (Fig. 3D).
A second set of experiments was performed where animals were taken through a “down-temperature” ramp (from 30 to 18°C) followed by an up-temperature ramp (from 18 to 30°C). This set of experiments was performed to determine 1) whether the linear relationship seen in Figs. 1–3 would hold true during warming of the animals, and 2) whether the relationship between the measured parameter and Ta would be the same during cooling and warming. A typical parameter measured (heart rate in the light cycle) is shown in Fig. 4A. The slopes of the relationship for Ta and mean arterial blood pressure, heart rate, and oxygen consumption were obtained for both the down ramp and up ramp for mice and rats. The slopes for the down ramp of each of these parameters [change in mean arterial pressure (ΔMAP): mouse = 1.7 ± 0.2 mmHg/°C, rat = 1.3 ± 0.2 mmHg/°C; change in heart rate (ΔHR): mouse = 16.4 ± 4.6 beats·min−1·°C−1, rat = 9.8 ± 1.0 beats·min−1·°C−1; change in mass-specific metabolic rate (ΔMSMR): mouse = 1.33 ± 0.15 ml O2·min−1·kg−0.75·°C−1, rat = 0.66 ± 0.03 ml O2·min−1·kg−0.75·°C−1] were not significantly different from the slopes of the up ramp for these parameters (ΔMAP: mouse = 1.6 ± 0.2 mmHg/°C, rat = 1.3 ± 0.2 mmHg/°C, ΔHR: mouse = 16.8 ± 3.8 beats·min−1·°C−1, rat = 9.7 ± 0.8 beats·min−1·°C−1; ΔMSMR: mouse = 1.28 ± 0.09 ml O2·min−1·kg−0.75·°C−1, rat = 0.66 ± 0.04 ml O2·min−1· kg−0.75·°C−1) as Fig. 4B shows.
There were two primary objectives in this study: 1) to determine whether small changes in Ta (±4°C) are sufficient to cause significant changes in the cardiovascular parameters of rats and mice, and 2) to compare the sensitivities of those cardiovascular parameters between rats and mice. Our findings demonstrate that in the normotensive rat and mouse strains studied, there is a highly sensitive and linear effect of raising or lowering Ta within the range of 18–30°C on cardiovascular function. A relationship between cold exposure and elevated blood pressure in rodents has long been recognized (3). Either acute or chronic exposure to cold (4–6°C) has repeatedly been shown to elevate blood pressure in small mammals (2, 3, 14). In accordance with these previous findings, both rats and mice in our study also exhibited elevated blood pressure at lower Ta (Figs. 1A and 2A). In fact, if one projects the linear relationship between blood pressure and Ta found in this study out to 4°C, the mean arterial pressure of the rats would be ∼125 mmHg, which agrees with the 115- to 130-mmHg range measured in rats in the 4–6°C Ta considered “cold” (2, 3, 14). Hence, not only do we observe that small incremental changes in Ta result in statistically different cardiovascular parameters, but we also observe that blood pressure and heart rate vary in a linear fashion with respect to a large range of Ta.
The mechanisms of this phenomenon of cold-induced hypertension have been well studied. The cause of cold temperature-induced hypertension appears to be dependent on thermoregulatory mechanisms, which operate via the SNS. Nonshivering thermogenesis in brown fat is activated by the SNS through the release of catecholamines. Studies have shown that SNS activation during nonshivering thermogenesis is necessary for cold tolerance in mice (8) and that elevated plasma catecholamine levels coincide with cold temperature-induced hypertension in rats (3, 14). Furthermore, rats reared at 18°C exhibit augmented development of the sympathetic ganglion relative to rats reared at 30°C (12), suggesting that temperature-induced SNS demands are capable of manifesting their effect on SNS development. Thus there is strong evidence to suggest that mammals increase SNS activity in response to thermogenic challenges. Although there is much regional specificity in activation of the SNS for thermoregulation (11), it appears as if activation of SNS in response to cooler Ta can also have a direct effect on the heart. Calculations of the SDIBI (Figs. 1D and 2D), an indicator of parasympathetic nervous system vs. SNS neurogenic output, indicate that decreasing environmental temperature below the TNZ of rats and mice leads to increased SNS and/or decreased parasympathetic nervous system input to the heart as variability in interbeat interval decreases in accordance with decreasing temperatures (25–27). The metabolic rate (Figs. 1E and 2E) of the mice and rats in these studies strongly suggests that the highest Ta used here (30°C) was near or below the lower critical temperature (the lower boundary of the TNZ). Metabolic rate, heart rate, and blood pressure all fell from 26 to 30°C (Figs. 1 and 2), suggesting that 26°C is below the TNZ. Although not tested here, it is likely that the mice had a higher TNZ than the rats based on the size of these animals. Hence, this study suggests that even modest changes in Ta can influence the autonomic nervous system in such a way as to significantly influence arterial blood pressures, heart rate, and metabolic rate.
Mice exhibited a greater responsiveness of most of the cardiovascular parameters and metabolic rate in response to altered Ta (Fig. 3), although this was not the case for pulse pressure. These data are consistent with the fact that mice likely have a greater relative thermogenic challenge at a Ta below the TNZ based on their greater surface area-to-volume ratio relative to rats. In concordance with this, we found that mouse spontaneous cage activity, and not that from the rat, is temperature dependent during the light cycle (Fig. 2F), with activity increasing as temperature decreased. This temperature dependence is not seen in the dark cycle, when rodents are much more active. This suggests that mice, and not rats, rely on increased cage activity, during a normally inactive period, as an additional source of heat production.
Several factors could influence the slope of the relationships between cardiovascular function and Ta for mice and rats. In this study, the slopes were generated based on values obtained after allowing 24 h to adjust to each temperature change. Within the range of temperatures studied, this is likely to be adequate amount of time to allow for these responses to be fully expressed. Chambers et al. (2) demonstrated that the cardiovascular responses to a large temperature change (23 to 4°C; and then 4 to 23°C) were nearly complete within 2 h in rats. Thus it seems unlikely that the slopes would be markedly different if longer periods of acclimation were incorporated into the design. However, we acknowledge that it is possible that after long periods of exposure to cool or thermoneutral conditions, the response to subsequent temperature challenges may be altered. While it might be predicted that obese animals would be less sensitive to changes in Ta, it remains clear that obese rats and mice still exhibit prompt and substantial reductions in mean arterial pressure and heart rate when Ta is increased to thermoneutrality (13, 30, 31). Even the rat model of essential hypertension, the spontaneously hypertensive rat, exhibits temperature-sensitive changes in mean arterial pressure and heart rate (Chambers JB, Williams TD, and Overton JM, unpublished results). Taken together, while it is clear that a number of actors may modulate the relationship, it seems clear that Ta serves as a highly sensitive determinant of cardiovascular function in rodents.
In summary, a Ta change of only a few degrees Celsius is enough to significantly impact mean blood pressure, heart rate, pulse pressure, heart rate variability, and metabolic rate in mice and rats. Furthermore, within the range of standard laboratory conditions, Ta exhibits a linear relationship with the blood pressure of mice and rats. Lastly, the cardiovascular system of the mouse is nearly twice as sensitive as that of the rat to a change in Ta. These findings emphasize the need for very careful control of Ta during assessment of cardiovascular function in rats and mice.
This work was supported by National Science Foundation Grant IBN-9984170 to S. J. Swoap, National Heart, Lung, and Blood Institute Grant HL-56732 to J. M. Overton, and in part by a Howard Hughes Medical Institute grant to Williams College.
We thank A. Fox and A. Christman for assistance in this work.
Present address of G. Garber: Dept. of Genetics and Development, Univ. of Connecticut Health Center, Farmington, CT 06030.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society