INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THERMAL DEHYDRATION OCCURS when endotherms increase evaporative water loss to regulate body temperature during heat exposure. Rats, which have been widely used in studies of thermal dehydration, increase evaporative water loss during heat exposure by increasing saliva production and spreading (5, 13, 14). Thermal dehydration in rats leads to increased water intake when water is made available (5, 15, 25, 26). This thermal dehydration-induced thirst is primarily due to a cellular water deficit rather than to hypovolemia (1, 5), as plasma volume seems to be partially defended during thermal dehydration (12, 17). This is in contrast to dehydration due to water deprivation, in which water loss from the extracellular compartment plays an important role in stimulating water intake (31). The renin-angiotensin system is involved in the thirst that accompanies water deprivation but not that which accompanies thermal dehydration (3, 6). Although core temperature does not directly affect thirst in thermally dehydrated rats (2), it is possible that increased thermal responses to heat exposure could alter the magnitude of thermal dehydration-induced thirst. We were interested in studying thermal dehydration-induced thirst in rats with altered thermal responses to heat exposure to answer this question and as part of a more generalized question about how preexisting physiological or pathophysiological states can modify physiological responses to heat exposure.

By examining thermal dehydration-induced thirst in hypertensive rats, we were able to investigate both of these types of questions, because in rats hypertension can be associated with altered responses to heat exposure. Spontaneously hypertensive (SH) rats (Okamoto-Aoki derived) are less tolerant of heat exposure than the normotensive, parent strain, Wistar-Kyoto (WKY) rats (27, 35). SH rats have higher colonic temperatures and a higher mortality rate during heat exposure than WKY rats (9, 27, 35). Although part of the increase in colonic temperature in SH rats is due to the measurement stress itself, studies using implanted radiotelemetry temperature sensors have shown that core temperatures in heat-exposed SH rats are from 1.5 to 2.5°C above those of WKY rats (8). The increased thermal responsiveness of SH rats to heat exposure may be due to increased sympathoadrenal responses to heat (8) and a reduction in hyperthermia-induced tail blood flow (28). Evaporative water loss responses of SH rats to exposure to either 34 or 37°C have been reported to be not significantly different from those of WKY rats (9, 35). On the basis of these data and the finding that spontaneously hypertensive Wistar rats drank the same amount of water as did normotensive Wistar rats after 24 h of water deprivation (18), we hypothesized that SH rats would show the same water balance responses to thermal dehydration as WKY rats. Contrary to this hypothesis, we found that thermally dehydrated SH rats both drank and voided more water than did control rats.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

General. Thirty male WKY rats and 30 male SH rats obtained from Harlan Sprague Dawley (Indianapolis, IN) were used for these experiments. The rats were housed individually in hanging stainless steel cages in an animal room maintained at 24 ± 2°C and illuminated from 0700 to 1900. Rats were allowed access to water and Purina rat chow ad libitum except during the experiments. Rats were housed for 4 wk before being used. Systolic blood pressure was measured using an indirect method (19) at the beginning of the study when the rats were 10 wk old. Systolic blood pressures (mean ± SE) were 125 ± 2 mmHg in the WKY rats and 161 ± 2 in the SH rats, which were significantly different at P < 0.001. Experiments were carried out over the next 10 wk. The body weights over the 10-wk study period ranged from 247 to 411 g for the WKY rats and from 200 to 423 g for the SH rats. There were no significant differences in body weights among experimental groups for any individual experiment. All experiments began between 0800 and 0900 and took place in walk-in environmental chambers maintained at 25 ± 0.5 or 37.5 ± 0.5°C. The experimental protocol was approved by the Hope College Animal Care and Use Committee.

Core temperature measurements. Core temperature was measured with implanted radiotelemetry temperature sensors by the method of Berkey et al. (8). Ten WKY and 10 SH rats chosen at random had calibrated VM-FH disc transmitters (Mini-Mitter) implanted into the left lower quadrant of the abdominal cavity. For the implantation surgery, the rats were anesthetized with a mixture of 100 mg/kg ketamine hydrochloride and 1 mg/kg acepromazine maleate injected intramuscularly and 20 mg/kg pentobarbital sodium injected intraperitoneally. Two days after the last temperature experiment, the rats were reanesthetized and the transmitter was removed. Core temperature responses at 25°C (1 wk after surgery) and 37.5°C (2 wk after surgery) were recorded simultaneously on WKY and SH rats. Core temperatures were sampled and stored every 5 min using a Zenith microcomputer and a Dataquest III (Data Sciences) computer board and software.

For the temperature measurements, the rats were weighed to the nearest 0.1 g and then placed in Nalgene metabolism cages that had the lid and food and water access port covers replaced with wire screening. The cages were then placed in the 25°C chamber, and temperature measurements were begun. For the 25°C experiment, the cages were left in the 25°C chamber for 4 h. For the 37.5°C experiment, the cages were transferred from the 25°C chamber to the 37.5°C chamber after 0.5 h and left there for 3.5 h. Equipment problems led to measurements being completed on 9 WKY and 8 SH rats at 25°C and 7 WKY and 7 SH rats at 37.5°C.

Thermal dehydration-induced drinking. Eighteen WKY and 18 SH rats that had not been used for the temperature experiments were used for this experiment. At the beginning of the experiment, the bladder of each rat was emptied by gentle suprapubic pressure. Then the rat was weighed to the nearest 0.1 g and placed in a modified Nalgene metabolism cage as in expt 1. Nine WKY and 9 SH rats were exposed to 25°C and 9 WKY and 9 SH rats were exposed to 37.5°C for 3.5 h. At the end of the exposure period, each rat was removed from the cage, reweighed, and transferred to a standard Nalgene metabolism cage at 25°C. Urine and feces (wet) outputs during the exposure period were recorded, and the urine was frozen for later determination of sodium and potassium concentrations by flame photometry using a Varian AA-475 atomic absorption spectrophotometer. Evaporation of urine during collection periods was minimized by using urine collection funnels on top of the urine containers. Evaporative water loss for each rat was estimated by subtracting the urine and fecal losses during exposure from the change in body weight. Fifteen minutes after transfer, each rat was given a water bottle containing tap water at 25 ± 1°C. Cumulative water intake and urine output were measured at 0.5, 1, 2, 3, and 4 h of access to water by the weight change in the water and urine containers. The percent rehydration at each measurement period was determined by dividing the water intake by the sum of the exposure period evaporative and urine losses and the drink period urine loss and multiplying by 100 percent. At the end of the water access period, the urine was frozen for sodium and potassium analysis as previously described.

Thermal dehydration effects on blood indicators of body water status. One week after the drinking experiments, 22 WKY and 22 SH randomly chosen rats were used to determine the effects of thermal dehydration on blood indicators of body water status. Eleven rats of each type were exposed to 25°C and 11 to 37.5°C as in the drinking experiment. Fifteen minutes after the end of the exposure period, each rat was anesthetized with methoxyflurane and a 3-ml blood sample was taken by cardiac puncture. Lithium heparin (200 units) was used as an anticoagulant for each sample. Measurements on the blood samples were done in triplicate. Hematocrit was measured by the microhematocrit method. Hemoglobin concentration was determined spectrophotometrically using a hemoglobin kit from Sigma (525-A). The remaining blood was centrifuged for 20 min at 1,200 g at 4°C. The plasma was removed and stored at 80°C for later analysis. Plasma protein concentration was determined with a Sigma total protein kit (541-2 and P-6529). Plasma osmolality was determined with a Wescor vapor pressure osmometer. Plasma sodium and potassium concentrations were determined by flame photometry as previously described. The inability to obtain blood samples on some rats led to measurements being completed on 11 SH rats at 25°C and 10 SH rats at 37.5°C and 10 WKY rats at each temperature.

Water deprivation-induced drinking. Two weeks after the blood sampling experiments, 12 WKY and 12 SH rats were randomly selected for a water-deprivation experiment. The rats were weighed and placed in standard Nalgene metabolism cages in the 25°C chamber. They were provided with tap water and powdered Purina rat chow. Twenty-four hours later, the rats were reweighed, water and food intake and urine and feces (wet) output were determined, and the water was removed. The design of the metabolism cages allowed for the amounts of spilled food and water to be determined so that intake measurements could be corrected for spillage. Twenty-four hours later, the rats were reweighed and food intake and urine and feces output were determined. The rats were then provided with water but not food. Cumulative water intake and urine output were measured after 0.5, 1, 2, 3, and 4 h of access to water.

Data analysis. The software program SYSTAT was used for statistical analysis. The mean and SE for all data of interest were calculated. Statistical significance was tested with the t-test, two-way ANOVA, two-way ANOVA with repeated measures, and three-way ANOVA with repeated measures as appropriate. Significance for all statistical tests was set at the 95% confidence level.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Core temperature measurements. The core temperature responses of the WKY and SH rats to exposure to 25 or 37.5°C are shown in Fig. 1. Core temperatures of the SH rats began at higher levels than those of the WKY rats but fell quickly during the first 30 min of measurements. Core temperatures were higher in both WKY and SH rats at 37.5°C than at 25°C. Core temperatures of the SH rats were higher than those of the WKY rats at 37.5°C but not at 25°C. At 37.5°C, the WKY rats maintained steady core temperatures during the final 80 min of heat exposure whereas the SH rats began to show further increases in core temperature during the last 20 min of exposure. Three-way ANOVA with repeated measures using data from every 10 min indicated significant interactions between strain (SH vs. WKY), temperature (25°C vs. 37.5°C), and time (min) (F_23,621 = 18.6, P < 0.001) and between strain and temperature (F_1,27 = 60.1, P < 0.001) on core temperature. ANOVA also indicated significant main effects of strain (F_1,27 = 49.4, P < 0.001), temperature (F_1,27 = 241.7, P < 0.001), and time (F_23,621 = 6.1, P < 0.001) and a significant interaction between temperature and time (F_23,621 = 64.8, P < 0.001) on core temperature.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Mean ± SE core temperatures measured by radiotelemetry of 9 Wistar-Kyoto (WKY) rats at 25°C, 7 WKY rats at 37.5°C, 8 spontaneously hypertensive (SH) rats at 25°C, and 7 SH rats at 37.5°C.

Thermal dehydration-induced drinking. Exposure to 37.5°C increased evaporative water loss in both strains of rats (Fig. 2, top). At 25°C, the WKY and SH rats had exactly the same evaporative water loss, whereas at 37.5°C the SH rats had a 34% greater evaporative water loss. Two-way ANOVA indicated a significant interaction between strain and temperature (F_1,32 = 22.7, P < 0.001) on evaporative water loss as well as significant main effects of strain (F_1,32 = 22.7, P < 0.001) and temperature (F_1,32 = 664.1, P < 0.001). Urine output was increased by exposure to 37.5°C (Fig. 2, bottom). Two-way ANOVA indicated only a significant effect of temperature (F_1,32 = 11.0, P < 0.005) on urine output.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Mean evaporative water loss (top) and urine output (bottom) of WKY and SH rats during 3.5 h of exposure to either 25 or 37.5°C. Error bars show SE; n = 9 for each condition.

Exposure to 37.5°C for 3.5 h increased water intake in both WKY and SH rats (Fig. 3, top). SH rats drank more than WKY rats after exposure to either 25 or 37.5°C. Three-way ANOVA with repeated measures indicated significant interactions between strain, temperature, and time (F_4,128 = 7.8, P < 0.001) and between strain and temperature (F_1,32 = 9.8, P < 0.05) on water intake. ANOVA also indicated significant main effects of strain (F_1,32 = 60.0, P < 0.001), temperature (F_1,32 = 100.1, P < 0.001), and time (F_4,128 = 331.1, P < 0.001) and significant interactions between strain and time (F_4,128 = 44.8, P < 0.001) and temperature and time (F_4,128 = 48.3, P < 0.001) on water intake.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Mean ± SE cumulative water intake (top), urine output (middle), and percent rehydration (bottom) of WKY and SH rats (previously exposed to either 25 or 37.5°C for 3.5 h) during 4 h of access to water at ambient temperature of 25°C. Where no SE is shown, value is smaller than size of symbol; n = 9 for each condition.

Urine outputs during the water access period are shown in Fig. 3, middle. Urine output of the WKY rats that had been exposed to 25°C increased slowly during the last 3 h of access to water, whereas the output of the WKY rats that had been exposed to 37.5°C was almost zero throughout the 4 h of access to water. In contrast, the urine output of the SH rats that had been exposed to 25°C rose rapidly after the first 0.5 h and was four times that of the WKY rats by the end of the water access period. The SH rats that had been exposed to 37.5°C had lower urine outputs than the SH rats that had been exposed to 25°C during the first 3 h of access to water, but by the end of the fourth hour the urine outputs of the two groups were essentially equal. Three-way ANOVA with repeated measures indicated significant interactions between strain, temperature, and time (F_4,128 = 10.9, P < 0.001) and between strain and temperature (F_1,32 = 4.2, P < 0.05) on urine output. ANOVA also indicated significant main effects of strain (F_1,32 = 270.9, P < 0.001), temperature (F_1,32 = 37.9, P < 0.001), and time (F_4,128 = 153.6, P < 0.001) and significant interactions between strain and time (F_4,128 = 90.8, P < 0.001) and temperature and time (F_4,128 = 7.0, P < 0.001) on urine output.

The percent rehydrations during the water access period are shown in Fig. 3, bottom. Percent rehydration increased with time in both groups of WKY rats and in the SH rats that had been exposed to 37.5°C. The SH rats that had been exposed to 25°C showed a rapid increase in percent rehydration during the first hour, but as urine output increased percent rehydration decreased during the second hour and then slowly increased for the remaining 2 h. By the end of the 4 h of access to water, none of the groups had rehydrated above 75%. Three-way ANOVA with repeated measures indicated that there were significant effects of strain (F_1,32 = 7.7, P < 0.01) and time (F_4,128 = 54.4, P < 0.001) on percent rehydration but no other significant main effects or interactions.

Urinary excretion rates of sodium and potassium during the exposure and water access periods are shown in Fig. 4. Exposure to 37.5°C increased sodium excretion in both WKY and SH rats (Fig. 4, top left). SH rats had higher levels of sodium excretion than WKY rats during the exposure period at both 25 and 37.5°C. Two-way ANOVA indicated significant effects of strain (F_1,32 = 6.8, P < 0.02) and temperature (F_1,32 = 23.6, P < 0.001) but no significant interaction between strain and temperature on sodium excretion during the exposure period. Potassium excretion during the exposure period (Fig. 4, top right) was not affected by either the exposure temperature or the strain of rat.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Mean urine sodium (left) and potassium (right) excretions of WKY and SH rats during 3.5 h of exposure to either 25 or 37.5°C (top) and during 4 h of access to water at an ambient temperature of 25°C (bottom). Error bars show SE; n = 9 for each condition.

Sodium excretion during the water access period (Fig. 4, bottom left) was also higher in the SH rats than in the WKY rats. Two-way ANOVA indicated a significant effect of strain (F_1,32 = 12.4, P < 0.002) but not temperature on sodium excretion during the water access period. Although sodium excretion during the water access period was decreased by heat exposure in the WKY rats and increased by heat exposure in the SH rats, the interaction between strain and temperature did not reach a significant level (F_1,32 = 3.8, P = 0.062). Potassium excretion during the water access period (Fig. 4, bottom right) was decreased by heat exposure in the WKY rats but not affected by heat exposure in the SH rats. Two-way ANOVA indicated a significant effect of strain (F_1,32 = 16.1, P < 0.001) and a significant interaction between strain and temperature (F_1,32 = 5.6, P < 0.05) on potassium excretion during the water access period.

Thermal dehydration effects on blood indicators of body water status. The inverse indicators of blood volume (hematocrit, hemoglobin concentration, and plasma protein concentration) were all higher in SH rats than in WKY rats after exposure to either 25 or 37.5°C (Fig. 5). Two-way ANOVA indicated significant effects of strain on hematocrit (F_1,37 = 20.1, P < 0.001), hemoglobin concentration (F_1,37 = 6.0, P < 0.02), and plasma protein concentration (F_1,37 = 5.1, P < 0.05) and significant effects of temperature on hemoglobin concentration (F_1,37 = 6.9, P < 0.01) and plasma protein concentration (F_1,37 = 5.6, P < 0.02), with heat exposure increasing these parameters in both WKY and SH rats. There were no significant interactions between strain and temperature on these variables.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Mean hematocrit (top), hemoglobin concentration (middle), and plasma protein concentration (bottom) of WKY and SH rats after 3.5 h of exposure to either 25 or 37.5°C. Error bars show SE; n = 11 for SH rats at 25°C and 10 for other 3 conditions.

The blood indicators of cellular dehydration (plasma osmolality and plasma sodium concentration) were higher in SH rats than in WKY rats after exposure to either 25 or 37.5°C. Heat exposure increased plasma osmolality and plasma sodium concentration in both SH and WKY rats (Fig. 6, top and middle). Two-way ANOVA indicated significant effects of strain (F_1,37 = 4.1, P < 0.05) and temperature (F_1,37 = 8.8, P < 0.005) on plasma osmolality and significant effects of strain (F_1,37 = 5.6, P < 0.05) and temperature (F_1,37 = 6.9, P < 0.02) on plasma sodium concentration. There were no significant effects of strain or temperature on plasma potassium concentration (Fig. 6, bottom). There were no significant interactions between strain and temperature on these variables.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Mean plasma osmolality (mosmol/kg; top), sodium concentration (middle), and potassium concentration (bottom) of WKY and SH rats after 3.5 h of exposure to either 25 or 37.5°C. Error bars show SE; n = 11 for SH rats at 25°C and 10 for other 3 conditions.

Water deprivation-induced drinking. The food and water intakes and urine and feces outputs of WKY and SH rats during a 24-h period at an environmental temperature of 25°C are shown in Table 1. SH rats drank significantly (P < 0.01) more water and produced significantly (P < 0.05) more urine and feces than WKY rats while eating the same amount of food. During 24 h of water deprivation, the SH rats ate significantly (P < 0.01) more food than the WKY rats but produced similar amounts of urine and feces (Table 2). At the end of the water deprivation period, the body weight of the WKY group was 360 ± 6 g, which was not significantly different from that of the SH group (355 ± 8 g). When food was withdrawn and water provided, the rats in both groups drank large and similar amounts of water (Fig. 7, top). Two-way ANOVA with repeated measures indicated a significant effect of time (F_4,88 = 213.2, P < 0.001) on water intake but no significant effect of strain nor significant interaction between strain and time. Urine output of both groups of rats remained near zero for the first hour of access to water and then increased linearly with time at the same rate (Fig. 7, bottom). Two-way ANOVA with repeated measures indicated a significant effect of time (F_4,88 = 326.0, P < 0.001) on urine output but no significant effect of strain nor significant interaction between strain and time.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Mean ± SE cumulative water intake (top) and urine output (bottom) of 12 WKY and 12 SH rats during 4 h of access to water after 24 h of water deprivation (WD).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The primary purpose of this study was to determine whether alterations in the thermal response to heat exposure would be associated with alterations in thermal dehydration-induced thirst. We used SH rats, which are known to have an exaggerated core temperature response to heat exposure (8, 9, 27, 35). As in a previous study from our laboratory (8), we found that at 25°C SH rats initially have higher core temperatures than WKY rats but that the core temperatures of the SH rats quickly fall to levels near those of the WKY rats. This initial elevation in core temperature of the SH rats has been attributed to a greater sensitivity of the SH rats to handling stress that transiently elevates core temperature (8). Core temperatures of WKY and SH rats were similar during the final 3.5 h of exposure to 25°C, although an unexplained increase in core temperature in one WKY rat during the final 1.5 h of exposure led to the mean core temperature of that group being slightly higher than that of the SH group. During exposure to the heat, the core temperature of both SH and WKY rats increased, with the increase being 2.4°C higher after 3.5 h of exposure to 37.5°C in the SH rats compared with the WKY rats.

The increased core temperature in the SH rats exposed to 37.5°C was associated with a higher level of evaporative water loss than in the WKY rats. Other investigators have not observed an increased evaporative water loss during heat exposure in SH rats (9, 35). In those studies, however, the exposure periods were shorter than in the current study because the core temperature measurement system used led to higher core temperatures in the rats and the need to terminate exposure before lethal hyperthermia (8, 9, 35). With the longer exposure period made possible by the use of the less stressful radiotelemetry temperature sensor, differences in evaporative water loss were readily apparent. The increased evaporative water loss of the SH rats at 37.5°C allowed them to maintain a fairly constant, although elevated, core temperature until the last 20 min of heat exposure. At this time, core temperature began to increase again, perhaps indicating a decrease in saliva spreading in SH rats.

Heat exposure increased urine output and sodium excretion in both WKY and SH rats as previously shown for Sprague-Dawley strain rats (4, 5). The increase in urine output during heat exposure is due to an as-yet-unexplained diuresis that occurs during the first hour of exposure to the heat (5). Increased urinary sodium excretion also occurs during water deprivation in rats (23) and thus may be a generalized response to dehydration. The SH rats' rate of sodium excretion was higher than that of the WKY rats, which may relate to the SH rats' higher level of dehydration.

The thermal dehydration-induced thirst we observed in both WKY and SH rats was associated with losses of water from both the cellular compartment, as indicated by increases in plasma osmolality and sodium concentration, and the blood compartment, as indicated by increases in plasma hemoglobin and protein concentrations. The greater water intake of the SH rats can be explained in part by the greater water deficits, caused primarily by greater evaporative water loss in the heat and indicated by the blood data, particularly the increased plasma sodium concentration. However, the lack of a significant interaction between exposure temperature and strain of rat on these blood variables indicates that the WKY and SH rats responded in a similar fashion to heat exposure. Using these indirect measures of hydration state, the SH rats appeared to have higher levels of dehydration than the WKY rats under both control and heat exposure conditions. These data must be interpreted with caution, however, because no direct measures of body water content were performed and the indirect measures may represent strain differences that are not related to body water content. Other studies comparing the blood responses of SH and WKY rats to dehydrating conditions reveal no consistent pattern of differences between the two strains (24, 29, 32, 33).

Although the greater level of dehydration in the SH rats is no doubt an important factor in their greater water intake, other factors beyond dehydration level must be involved in controlling water intake in SH rats. Evidence for an additional factor is the finding that SH rats rehydrated at a higher percentage during the first hour of access to water and had both large water intakes and urine outputs during the last 3 h of access to water. One possibility is that SH rats have a greater responsiveness to dipsogenic stimuli than do WKY rats.

SH rats have been reported to have altered responses to the dipsogenic hormone ANG II. SH rats had greater water intakes than WKY rats after both subcutaneous ANG II (21) and intracerebroventricular ANG II (21, 30, 36, 37), although both reduced and unchanged drinking responses to intracerebroventricular ANG II also have been reported (11, 16). Although the renin-angiotensin system does not appear to play a role in thermal dehydration-induced thirst in Sprague-Dawley rats (3), it may play a role in WKY or SH rats. If so, an increased responsiveness to ANG II in the SH rats may be responsible for a portion of their increased thermal dehydration-induced water intake. Alternatively, SH rats may have an altered responsiveness to cellular dehydration. After administration of hypertonic NaCl solution, which causes cellular dehydration, SH rats drank more water than normotensive rats in some studies (18, 21, 30) but not in another study (11). However, the greater water intake of the SH rats in these studies may reflect a greater retention of sodium rather than a greater sensitivity to its dipsogenic effect (18).

The differences in water intake between thermally dehydrated SH and WKY rats are particularly striking compared with the responses to water deprivation. We found no difference in water intake between SH and WKY rats after 24 h of water deprivation. Our data are very similar to those seen previously in a comparison of water intake of SH and normotensive Wistar rats after water deprivation (18). In another study (30), SH rats had greater water intake after 18 h of water deprivation after the first 30 min of access to water.

Water intake of SH rats after dehydration caused by heat exposure or water deprivation can be compared with water intake during control conditions. We found that SH rats drank almost twice as much water as WKY rats after a 4-h control period at 25°C. This greater water intake was associated with indexes of cellular and volemic dehydration in the SH rats. In previous studies (10, 24, 29, 32), water-replete SH and WKY rats generally had similar hematocrits, plasma osmolalities, and plasma sodium and potassium concentrations, although SH rats did have a significantly higher hematocrit in two studies (32, 33). As in two previous studies (20, 21), but not another (34), we found that SH rats drank significantly more water than did WKY rats during a 24-h control period. We and others (20, 21) found that 24-hour food intake of SH and WKY rats is not significantly different. It has been suggested that SH rats show food-related hyperdipsia (20, 21).

In addition to having alterations in water intake, SH rats show significant differences in urine output compared with WKY rats. During a 24-h control period, SH rats produced more urine than did WKY rats. The increased urine production (125.6% of WKY urine production) closely paralleled the increased water intake (127.5% of WKY water intake). During the 24-h water deprivation period and during the drinking period after water deprivation, urine output of the SH rats was not different from that of the WKY rats. These findings can be contrasted to those of the thermal dehydration experiment.

After heat exposure there was a reduction in urine output in both WKY and SH rats, reflecting the dehydration and natriuresis that both groups exhibited. However, during the drink period after either control or heat exposure, the SH rats produced significantly more urine than did the WKY rats. Of particular interest was the finding that by the end of the 4-h drink period, the thermally dehydrated SH rats' urine output was 49% of their water intake. In comparison, the thermally dehydrated WKY rats' urine output was 2% of their water intake. Thus, although the thermally dehydrated SH rats drank enough water to cause 96% rehydration, their failure to retain the water led them to have rehydration levels no greater than the WKY rats (72%). In previous studies under a variety of circumstances, SH rats had urine outputs and urinary sodium excretion rates that were lower, higher, or the same as normotensive rats (7, 18, 24, 29, 34).

Differences in urine output between normotensive and hypertensive rats may be related to alterations in vasopressin concentration and effectiveness in the hypertensive rats. Under control conditions, SH rats have lower concentrations of vasopressin in the hypothalamus (32, 33), higher concentrations in the pituitary (10, 32, 33), and either higher (10, 24) or similar (29, 32) concentrations in the plasma. During water deprivation, plasma vasopressin increased more in SH rats than in WKY rats (24, 29). Plasma vasopressin responses to intraperitoneal NaCl were not different in SH and WKY rats (33). However, SH rats appear to be less sensitive to the antidiuretic effects of vasopressin (22, 24). No measurements of vasopressin were made in the current study, but the urine output data suggest that drinking causes an exaggerated decline in vasopressin levels in SH rats after thermal dehydration but not water deprivation. The mechanism responsible for this vasopressin response, if it occurs, and the failure of the SH rats to retain needed water after thermal dehydration is not known and is deserving of further study.

Perspectives

The degree to which heat exposure reduces body water and thus compromises physiological function can be modified by a preexisting physiological or pathophysiological state. The consequences of the modification will be dependent on the ability of the animal to replace body water when water becomes available to drink. SH rats show such a modification in that they have a greater thermal responsiveness to heat exposure than normotensive rats. During exposure to 37.5°C, SH rats had significantly higher core temperatures than WKY rats exposed to the same temperature, probably as a result of both increased heat production and decreased dry heat loss from the tail (8, 28). The greater core temperature of the SH rats led to significantly greater evaporative water losses in this group. However, the current study shows that, if the level of dehydration is not too severe, SH rats can compensate for increased water loss by increased water intake, thus demonstrating a continued ability to relatively maintain water homeostasis despite a failure in both blood pressure and thermal homeostatic mechanisms.

Furthermore, this study has shown that there may be an interaction between heat exposure and renal function in hypertensive rats such that the renal response to rehydration is dependent on the way in which dehydration occurs. Studies such as the one described here indicate some of the complexity of the interactions of the systems that regulate temperature regulation, blood pressure, and water balance that would not be apparent in a more reductionist approach to physiological phenomena.