elderly humans consume adequate fluids on a daily basis (7). Yet dehydration is a major health risk for the elderly (5), and the elderly are susceptible to dehydration because of increased fluid losses (41). Aged kidneys have a decreased ability to concentrate urine (17, 27) and to conserve sodium (8), along with a relative resistance to vasopressin (25), and release less renin when challenged (6). These changes with age predispose the elderly to hypovolemia and dehydration because of the reduced ability to conserve water and sodium during times of relative water and sodium loss. The elderly have decreased thirst sensation (24, 26, 32). As a result, they are slower than young people to restore fluid balance after exercise-induced dehydration (19) and water deprivation (26). It thus appears that the elderly are more vulnerable to the consequences of physiological challenges to body fluid homeostasis and that diminished thirst in the elderly contributes to their problems of fluid balance (15).

The capacities of the aging kidney have been extensively investigated in rat models of aging. As in elderly humans, kidneys of aging rats have diminished capacity to concentrate urine (4, 16, 21,23) and to release renin when challenged (1-3,28). However, there are only a few studies in aging rats investigating the capacities of behavioral systems, namely, thirst and salt appetite, to supply sufficient water and sodium to maintain adequate hydration. Investigators variously report that old rats have diminished (14, 23, 33), increased (4, 14, 16,22), or unchanged (35, 36) daily intakes of water compared with young cohorts. Compared with young rats, old rats reportedly drink equivalent (30) or smaller (33) amounts of water in response to water deprivation and exogenously administered ANG II. Old rats of some strains drink less water than their younger cohorts in response to isoproterenol treatment (30) but drink as much as young rats after osmotic loads (30). Old rats have diminished salt appetite responses after sodium depletion and converting enzyme inhibition (28,30). There are insufficient data to derive a consensus concerning the conditions under which old rats may express diminished thirst and salt appetite responses.

The Brown Norway (BN) rat is often used in aging studies as an alternative to commonly employed Fischer 344 (F344) and Sprague-Dawley (SD) strains (18). The BN rat has been shown to have some advantages over these strains. The BN rat has a lower incidence of many kinds of organ lesions than the F344 rat and survives relatively disease free until very old age (18). The survival curve of the BN rat is similar in shape to that of the human, an important feature for an animal model of aging (12). BN rats have smaller age-related increases in adiposity than SD or F344 rats (13, 42) and, therefore, may be a better model of healthy aging. The present experiments tested the thirst and salt appetite responses of young and old BN rats to a series of standard body fluid challenges.

METHODS

Animals.

Male BN rats aged 4 mo (young, n = 10) and 20 mo (old,n = 10) were obtained from Harlan (Indianapolis, IN) through services provided by the National Institute on Aging. They were housed singly in hanging stainless steel cages in a temperature-controlled room (23°C) on a 12:12-h light-dark cycle. They were fed standard Purina rat chow and had ad libitum access to water and 0.3 M NaCl unless indicated otherwise. Intakes were recorded daily from 100-ml graduated cylinders with attached stainless steel spouts fastened to the fronts of the cages. All procedures conformed to the guidelines of the American Physiological Society.

Drugs.

Furosemide (Abbott Laboratories, N. Chicago, IL) was administered at 10 mg/kg body wt sc. Captopril (SQ-14225, Bristol-Meyers-Squibb Pharmaceutical Research Institute, Princeton, NJ) was dissolved in tap water to achieve a dose of 0.1 mg/ml.

Experimental protocols.

The rats received a series of experimental challenges to body fluid homeostasis over ∼3 mo. The experiments were performed in the order presented, with 6–12 days separating challenges within an experiment and between experiments.

Experiment 1: extracellular fluid depletion.

Rats were weighed in the morning and placed in standard metabolism cages with stainless steel funnels. At 1 PM, furosemide was injected (10 ml/kg body wt sc) to induce natriuresis and diuresis. After 1 h, access to water was provided in 100-ml graduated cylinders attached to the fronts of the cages. Food was not present. On the next morning, 20 h later, overnight water intakes were recorded. Water and 0.3 M NaCl were then provided from 0.1-ml graduated glass burettes with sipper spouts, and intakes were recorded every 30 min for 4 h. The rats were then returned to their home cages, and intakes were recorded over the next 20 h from graduated cylinders. Rats received three such tests separated by 8–12 days. In tests 1 and3, urine was collected into Nalgene tubes (0.1-ml resolution) for the 1st h after furosemide injection. Urine for the remainder of the overnight period was collected into preweighed glass beakers, and urine volume was recorded in the morning simultaneously with overnight water intakes. This urine volume was calculated as 1 g = 1 ml. For test 2, urine for the entire 20-h depletion period was collected into preweighed glass beakers. Samples were refrigerated for later analysis of sodium and potassium content. We noted that old rats did not excrete as much urine as young rats during tests 1 and 2; therefore, in test 3, old rats received two injections of furosemide (i.e., a second injection 2 h later) in an attempt to equalize the water losses between young and old rats.

Experiment 2: intracellular fluid depletion.

There were two tests separated by 1 wk. On test days, rats were weighed and placed in standard metabolism cages, as described above. They were injected with hypertonic saline (2 ml/kg body wt sc, 1.0 M NaCl fortest 1 and 2.0 M NaCl for test 2). Water was provided immediately from glass burettes placed at the fronts of the cages, and intakes were recorded every 30 min for 3 h. Urine was collected in Nalgene tubes (0.1-ml resolution) placed under the funnels. Urine volume was measured at the end of 3 h, and samples were refrigerated for later analysis of sodium and potassium content. To minimize potential discomfort from the subcutaneous injections of hypertonic saline, the solutions were made with 0.2% lidocaine. The animals showed no signs of discomfort.

Experiment 3: overnight food and fluid restriction.

At 10 AM on test days, the rats were weighed and placed in standard metabolism cages with stainless steel funnels. Neither food, nor water, nor 0.3 M NaCl was available throughout the day or overnight. Urine was collected in preweighed glass beakers placed beneath the funnels. On the next morning, 23 h later, urine volume was measured, water and 0.3 M NaCl were provided from glass burettes, and intakes were recorded every 30 min for 3 h. Samples of overnight urine were refrigerated for later analysis of sodium and potassium content.

Experiment 4: captopril adulteration of drinking water.

Rats increase their daily intakes of concentrated saline solutions or water when low concentrations of captopril are added to the drinking fluids (11, 39). In this experiment, daily intakes of water and 0.3 M NaCl were recorded when captopril (0.1 mg/ml) was added to the drinking water. Fluids were provided from 100-ml graduated cylinders. In the first part of the experiment, daily intakes of water and 0.3 M NaCl were recorded for 3 days without captopril, for 4 days with captopril in the drinking water, and for 3 days after removal of captopril from the drinking water. In the second part of the experiment, the saline tubes were removed, and only water was available for drinking. Water intakes were recorded for 3 days without captopril, for 4 days with captopril in the drinking water, and for 3 days after removal of captopril from the drinking water.

Urine analysis.

Urine was measured for urine volume (UV). Urinary concentration of sodium (U_Na) and potassium was determined by ion-specific electrodes (NOVA Biomedical, Waltham, MA) and used for calculation of urinary excretion of sodium (U_NaV) and potassium. Relative water balances were calculated by subtracting the total UV collected from the total amount of fluid (i.e., water or water + saline) ingested over the course of an experiment. Relative sodium balances were calculated by subtracting the total U_NaV collected from the total amount of sodium ingested in the form of 0.3 M NaCl over the course of an experiment. We use the term “relative” for the balance measures, inasmuch as respiratory and fecal losses of water and sodium were not considered.

Statistical analysis.

Data were analyzed by ANOVA appropriate to the experimental designs. Planned comparisons were made with Fisher's least significant difference tests when the global F ratio was significant. Values are significant at P < 0.05.

RESULTS

All testing was performed in the same animals over the course of ∼3 mo. One old rat died within the first 3 wk of testing (from undetermined cause), and its data were discarded. The rest of the old rats were vigorous throughout the study.

Body weights.

Old rats weighed significantly more than young rats throughout testing (F_1,17 = 71.75, P < 0.05; Fig. 1). Body weight of old rats averaged ∼100–120 g more than that of young rats at the time of each experiment. The groups showed near-parallel gains of ∼70 g during the course of testing. Because of the significant difference in body weight between the age groups, the data are analyzed and presented as absolute measures and as measures adjusted for body weight.

Experiment 1: extracellular fluid depletion.

Intakes were analyzed as rates, i.e., milliliters per 30 min. Old rats drank significantly less water and 0.3 M NaCl than young rats on each test, in terms of absolute and body weight-adjusted intakes (age main effects: all F_1,17 ≥ 5.71,P < 0.05). Furthermore, young rats increased intakes of 0.3 M NaCl on repeated testing, in terms of absolute and body weight-corrected values, while old rats did not (interactions: bothF_14,238 ≥ 3.92, P < 0.05). Young rats increased ingestion of saline over tests independent of weight gain, inasmuch as their body weight-adjusted measures also increased with testing. The cumulative intakes of water and 0.3 M NaCl, adjusted for body weight, are presented in Fig.2. When returned to their home cages, old rats continued to consume significantly less saline on a body weight basis than young rats over the next 20 h. Thus the total amount of 0.3 M NaCl consumed in 24 h of saline access after sodium depletion was also significantly less in old rats than in young rats (means of the 3 tests = 0.4, 0.3, and 0.6 vs. 1.3, 2.0, and 2.4 ml/100 g body wt, F_2,34 = 4.42,P < 0.05).

In the 20-h period after furosemide injection and before the drinking tests, old rats tended to excrete less urine and drink less water than young rats. UV was less in old rats than in young rats in tests 1 and 2 (Table 1). Intest 3, old rats received two injections of furosemide, and their total UV values were equivalent to those of young rats but were still significantly reduced when adjusted for body weight (interaction: both F_2,34 ≥ 3.77, P < 0.05). The generally reduced UV values of old rats were accompanied by smaller overnight water intakes (main effects: bothF_1,17 ≥ 16.06, P < 0.05), so that water balances immediately before tests 1 and2 were equivalent between groups, in terms of absolute and body weight-adjusted measures, and were significantly reduced in old rats before test 3 (interaction: bothF_2,34 ≥ 4.42, P < 0.05). The total (i.e., water + saline) amounts of fluid ingested during the 4-h salt appetite tests were significantly reduced in old rats (interaction: both F_2,34 ≥ 5.51,P < 0.05) as were the relative water balances after the drinking tests (interaction: both F_2,34≥ 9.61, P < 0.05). Only young rats increased the total amount of fluid ingested and the relative water balance with repeated testing. Young rats drank enough fluid during the 4-h test to more than replace the amount lost during the 20 h of depletion, whereas old rats did not.

In tests 1 and 3, urine was collected for the 1st h after furosemide injection and again in the morning 19 h later. In the 1st h, UV, U_Na, and U_NaV were equivalent between young and old rats (all F_1,17 ≤ 2.92, not significant; Table 2). U_Na and U_NaV were significantly greater intest 3 than in test 1 (test main effect: bothF_1,17 ≥ 6.60, P < 0.05). When adjusted for body weight, UV and U_NaV were significantly reduced in old rats (age main effect: bothF_1,17 ≥ 6.61, P < 0.05), and there were no main effects of test. U_Na over the entire 20-h depletion period was consistently higher in old rats (F_1,17 = 25.51, P < 0.05; Table 3). It appears that young rats reduced U_Na at some point after the 1st h of depletion, presumably when the effects of furosemide began to decline, whereas old rats did not. U_NaV was significantly increased in old rats (F_1,17 = 8.64, P < 0.05), although body weight-adjusted U_NaV was equivalent between ages (F_1,17 = 1.43, not significant). U_NaV increased over tests for both ages (main effect:F_2,17 = 14.87, P < 0.05), but body weight-adjusted U_NaV did not. Old rats ingested less sodium than young rats on all tests and did not increase sodium ingestion over tests (interaction: F_2,34 ≥ 5.81, P < 0.05). Relative sodium balance was significantly reduced in old rats compared with young rats on all tests and did not increase over tests in old rats as it did in young rats (interaction: F_2,34 ≥ 5.14,P < 0.05). Thus old rats drank significantly smaller amounts of sodium during the 4-h drinking tests than young rats, despite the fact that old rats were in equivalent negative sodium balance (as per body weight-adjusted 20-h U_NaV) or in significantly greater negative sodium balance (as per absolute 20-h U_NaV). On repeated testing, young rats increased ingestion of sodium in amounts that far exceeded additional urinary losses of sodium on repeated testing. Old rats did not. We did not collect urine during the 4-h drinking tests, so we cannot address the cumulative water and sodium balances at the end of testing, but it is apparent that old rats were refractory in their ability to ingest sufficient water and saline to repair deficits accrued during the 20 h preceding the drinking tests.

Experiment 2: intracellular fluid depletion.

Both ages showed dose-dependent water intakes in response to subcutaneous injections of NaCl (i.e., 2 ml/kg body wt of 1.0 or 2.0 M NaCl; dose main effect: F_1,17 = 92.04,P < 0.05). Absolute water intakes did not differ between groups. However, when intakes were adjusted for the significant differences in body weight (F_1,17 = 61.53,P < 0.05), old rats drank significantly less than young rats; i.e., only half as much on both tests (F_1,17 = 13.06, P < 0.05). Cumulative water intakes adjusted for body weight are presented in Fig.3. The groups drank nearly equivalent amounts of water in the first 30 min, and then old rats drank at a significantly reduced rate compared with young rats for the next 30 min. Drinking by both groups was essentially completed by 1 h.

Old rats received significantly greater sodium loads than young rats (F_1,17 = 46.10, P < 0.05); the loads were equivalent when adjusted for body weight, inasmuch as the loads were delivered on a body weight basis (Table4). Old rats excreted significantly more urine than young rats during the 3-h tests, even when UV values were adjusted for body weight (main effects:F_1,17 ≥ 6.99, P < 0.05). U_Na was significantly lower in old rats (F_1,15 = 7.15, P < 0.05). Absolute U_NaV was increased in old rats compared with young rats (F_1,17 = 8.05, P < 0.05) in both tests, although U_NaV was equivalent on a body weight basis. Cumulative sodium balances were equivalent between ages and significantly higher after 2.0 M NaCl than after 1.0 M NaCl (dose main effects: F_1,17 ≥ 94.25,P < 0.05). The tendency for old rats to excrete more urine and ingest less water during testing resulted in significantly reduced water balances at the end of testing in old rats compared with young rats, in terms of absolute measures and on a body weight basis (age main effects: F_1,17 ≥ 13.65,P < 0.05; Table 5). Water balances were significantly greater for both groups on the second test (dose main effects:F_1,17 ≥ 28.58, P < 0.05).

The “estimated required water intake” in Table5 is the amount of water we estimated the animals must drink to dilute the sodium loads to isotonicity without taking excretion into account. The formula of Wolf (43) was used for the calculations. Old rats required significantly more water to dilute their sodium loads, because the loads were determined on the basis of body weight and old rats were significantly heavier. The amount of water required to dilute the loads decreases after urinary excretion is factored in, as can be seen by “postexcretion estimated required water intake.” For this measure, we modified Wolf's formula to take into account the amount of the sodium load and water excreted in the urine during the 3 h of testing. Urinary excretion of part of the load benefits the animals by reducing the amount of water required through ingestion to dilute the remaining load to isotonicity. However, for the smaller sodium load, excretion benefited old rats less. Urinary excretion reduced the required water intake by 17% in young rats and by only 3% in old rats (P < 0.05). As seen in Table 4, old rats excreted more dilute urine and lost more water through excretion than young rats; hence, they derived less benefit in terms of the amount of water that must be ingested. Absolute water intakes did not differ between groups. However, old rats ingested only 50% of the estimated water required to dilute body fluids to isotonicity, after U_NaV and UV are considered. The difference in calculated required water intakes and actual intakes was significantly different between groups. On a body weight basis, the estimated amount of water required to dilute the sodium load to isotonicity was equivalent between ages, inasmuch as the load was administered on a body weight basis. After excretion was factored in, old rats required more water than young rats at the smaller sodium load. This measure again shows that urinary excretion benefited young animals more than old animals at the lower dose. Old rats drank a significantly smaller fraction of the estimated required water intakes on both tests than young rats. Old rats drank roughly half of the amount they needed to drink to achieve isotonicity.

Experiment 3: overnight food and fluid restriction.

Water intakes in 3 h did not differ between young and old rats as absolute measures or as measures adjusted for body weight. Intakes of 0.3 M NaCl were significantly reduced in old rats by either measure (interaction effects: F_5,85 ≥ 5.20,P < 0.05). Cumulative intakes of water and 0.3 M NaCl adjusted for body weight are presented in Fig.4.

The UV values in the 24 h preceding the drinking tests were significantly increased in old rats (age main effects:F_1,17 ≥ 12.72, P < 0.05; Table 6), indicating a significantly greater negative water balance in old rats before the drinking test. U_Na was significantly reduced in old rats (age main effect: F_1,17= 10.46, P < 0.05), but 24-h sodium excretion was equivalent between ages as absolute or body weight-adjusted measures (both F_1,17 ≤ 1.28, not significant). Therefore, the groups had equivalently negative sodium balances before the drinking test. The total (i.e., water + saline) amount of fluid ingested in 3 h was equivalent between groups but was significantly lower in old rats when adjusted for body weight (age main effect: F_1,17 = 5.56, P < 0.05). The relative water balance after ingestion was significantly reduced in old rats compared with young rats as absolute or body weight-adjusted water balance (age main effects:F_1,17 ≥ 16.16, P < 0.05). The total amount of sodium ingested in 3 h was significantly reduced in old rats on a body weight basis (age main effect:F_1,17 = 7.37, P < 0.05). The relative sodium balances were not different between the age groups (both F_1,17 ≤ 3.63, not significant).

Experiment 4: captopril adulteration of drinking water.

In the first part of the experiment, water and 0.3 M NaCl were available for drinking. Old rats weighed significantly more than young rats (480 ± 16 vs. 375 ± 8 g,F_1,17 = 39.31, P < 0.05) and drank significantly more water daily than young rats (main effect:F_1,17 = 27.33, P < 0.05). However, young rats had significantly greater daily intakes of saline, and young rats increased their daily saline intakes on captopril adulteration of the drinking water (interaction:F_9,153 = 8.65, P < 0.05). When adjusted for body weight, there were no age effects on water drinking, but a significant effect of days (F_9,153 = 6.49, P < 0.05) showed that both ages increased water intake on the 1st day of captopril adulteration compared with the first 3 days of baseline intakes (Fig. 5). On a body weight basis, old rats drank significantly less saline than young rats on 1 of the 3 days preceding captopril treatment and on 2 of 3 days after captopril treatment (interaction: F_9,153 = 9.17,P < 0.05). Young rats significantly increased daily saline intakes in response to captopril treatment, whereas old rats did not.

In the second part of the experiment, only water was available for drinking. Old rats weighed significantly more than young rats (481 ± 16 vs. 381 ± 9 g, F_1,17 = 33.11,P < 0.05). Old rats drank significantly greater amounts of water on a daily basis than young rats but were completely without response to the addition of captopril to the drinking water (interaction: F_9,153 = 5.09,P < 0.05). Young rats significantly increased daily water intakes on captopril adulteration. When daily intakes were adjusted for body weight, young and old rats drank nearly identical amounts of water on baseline days before and after captopril (Fig. 5). Captopril adulteration of the drinking water caused a significant increase in water drinking only in young rats (interaction:F_9,153 = 5.71, P < 0.05).

Daily water and 0.3 M NaCl intakes throughout experimental testing.

The body weight-adjusted daily intakes of water and 0.3 M NaCl fordays 3–66 are presented in Fig.6 [intakes recorded during the last 20 days of experimental testing (days 67–86) are presented in Fig. 5]. For purposes of statistical analysis, intakes were averaged over blocks of 3 days before each experimental challenge, (e.g., days 4–6 and 13–15), yielding seven data points per group through day 66. The intakes were adjusted for body weight using the body weight determined on the day of each challenge. There were main effects of age and days and significant age-by-days interactions for daily water intakes as absolute and body weight-adjusted measures. It appears that old rats had unstable water intakes initially. Old rats had significantly reduced water intakes for the first block of days (i.e., days 4–6) but greater daily water intakes overall [26.6 ± 1.7 vs. 21.8 ± 0.5 (SE) ml]. When adjusted for body weight, old rats also drank significantly less than young rats in the first block of days and drank essentially the same amount of water on a daily basis overall [6.0 ± 0.3 vs. 6.5 ± 0.2 (SE) ml]. There were no age differences in daily intakes of 0.3 M NaCl through day 66. Both ages drank very small amounts of saline on a daily basis. Absolute daily intakes of saline averaged 0.6 ± 0.1 ml for old rats and 0.7 ± 0.1 ml for young rats. There were significant main effects of days for daily saline intakes as absolute and body weight-adjusted measures. The groups drank significantly more saline on blocks 6 and 7 than on block 2 (i.e., 0.9 and 1.0 vs. 0.3 ml), corresponding to days 53–55 and64–66 compared with days 13–15. This was also the case for body weight-adjusted daily saline intakes. Thus there was a significant tendency to drink more saline on a daily basis as testing progressed.

DISCUSSION

All studies showed behavioral differences between young and old rats. Experiment 1 showed that old rats consistently drank less water and saline than young rats after periods of sodium depletion and, unlike young rats, did not increase saline ingestion on repeated depletions. Experiment 2 showed that old rats drank less water in response to osmotic challenge (i.e., subcutaneous hypertonic saline) than young rats. Although young and old rats demonstrated dose-dependent water drinking in response to subcutaneous hypertonic saline, old rats drank only ∼50% of the estimated water required to dilute the loads to isotonicity. In experiment 3, old and young rats drank roughly equivalent amounts of water after food and fluid restriction, but old rats drank significantly less saline and significantly less fluid overall (i.e., water + saline). Finally, we noted in old rats an almost total absence of increased thirst and salt appetite in response to adulteration of the drinking water with captopril, a procedure that readily increases intakes of these substances in young rats. Thus old BN rats show diminished thirst and salt appetite responses to several experimental challenges to body fluid homeostasis. There may be possible order effects of the treatments, inasmuch as all rats received the same sequence of tests.

One commonly employed model to stimulate water and sodium ingestion involves administration of a diuretic combined with sodium restriction (30, 31, 40). Injections of furosemide are given to cause diuresis and natriuresis, and then animals are sodium restricted overnight, having access only to water to drink or to sodium-deficient chow and water. On the following morning, access to sodium solution is provided. The sodium-depleted animals ingest substantial amounts of saline solution along with smaller amounts of water. Some investigators find that, with repeated testing in this manner, rats drink significantly more sodium solution on subsequent tests (31,40). In experiment 1, young rats ingested more saline and water after overnight sodium depletion than old rats. The saline intakes of these BN rats were rather small and more similar to the intakes of F344 (20, 29) than SD (40) or Long-Evans (37) rats after sodium depletion. The young rats significantly increased sodium ingestion on repeated testing, whereas old rats did not.

The decreased behavioral responses of the old rats are not likely to be due to significantly reduced water and sodium need compared with young rats after depletion. The calculated water balances immediately before the salt appetite tests were equivalent between young and old rats. The amounts of sodium lost by old rats during the depletion periods were significantly greater than those of young rats or equivalent when adjusted for body weight differences between the ages. Therefore, old and young rats appeared to have an equivalent need for water and sodium at the start of the salt appetite test, yet old rats drank much less of both fluids than young rats. In test 3, old rats received two injections of furosemide in an attempt to increase their overnight urinary output to match that of young rats. This treatment resulted in the highest urinary loss of water and sodium in old rats and the greatest amount of water ingestion on the night before access to sodium solution of the three tests. Yet, despite the urinary loss of ∼2 mmol of sodium, old rats barely ingested saline when given access to it. Even after 24 h of saline access, old rats failed to consume sufficient amounts of saline to replace the sodium lost by excretion during the depletion period. We assume that old rats must eventually return to sodium balance through the sodium ingested in the chow, because they do not appear to do so by ingesting the available concentrated saline solution.

In their study using similar sodium depletion procedures, Rowland et al. (30) found that 20-mo-old male SD and F344 rats had reduced sodium intakes compared with their younger (5-mo-old) cohorts. Their old rats repaired through ingestion <30% of the accrued sodium deficit. We found that old BN rats replaced only 2% of the sodium lost during the previous 20 h of depletion in test 1 and only 16% of the sodium lost from depletion in test 3. In contrast, young BN rats replaced half of the sodium deficit intest 1 and fully compensated by test 3 by ingesting more sodium than was lost from depletion. In their test, Rowland et al. offered hypotonic sodium solution (0.05 M NaCl) for drinking and speculated that the deficits in sodium ingestion by old rats might have been greater if hypertonic sodium solutions were offered. Our present results suggest that they were correct.

The ingestion of sodium after furosemide-induced depletion is mediated largely by increased renin secretion and the resultant increased levels of circulating ANG II (10, 38). Old rats are generally impaired in secreting renin (1, 2). Therefore, old rats may have deficient salt appetite responses after sodium depletion, because they do not produce as much ANG II as young rats. In addition, old rats have increased levels of circulating atrial natriuretic peptide (1), which not only inhibits renin secretion but also inhibits thirst and salt appetite responses directly.

We found that old rats had deficient water drinking responses to osmotic loads after subcutaneous injection of two doses of hypertonic saline (i.e., 2 ml/kg body wt of 1.0 and 2.0 M NaCl). These results differ from other work showing no reductions in water intake by old rats after subcutaneous hypertonic saline (30). In the present study, young and old rats drank equivalent amounts of water in response to sodium loads. However, the old rats were injected with more sodium, because the loads were administered on a body weight basis. This means that old rats did not ingest proportionately as much water as young rats relative to their greater sodium load. Indeed, calculations of the amount of water required to dilute the sodium load to isotonicity indicate that old rats drank ∼50% of the water required to dilute their loads. This fractional intake of the estimated required amount of water by old rats is significantly less than that of young rats who drank nearly the entire amount of required water. The thirst deficit of old rats is apparent in their significantly reduced body weight-adjusted intakes of water, in which the body weight-adjusted sodium load is, obviously, equivalent between young and old rats. Urinary handling of the sodium load was also different between young and old rats. Old rats excreted a greater fraction of the NaCl load than young rats at both doses, a finding that is consistent with an earlier report (30). However, UV was also greater in old rats, both as absolute and body weight-adjusted measures, and the urine from old rats was more dilute. This is consistent with the reduced concentrating ability of the aged kidney that has been observed by others (4, 16, 21, 23).

In experiment 3, rats were deprived of food, water, and saline overnight and then given access to water and saline in the morning. Old rats drank significantly less saline solution and less fluid overall (i.e., water + saline) than young rats. In this experiment, old rats excreted greater amounts of urine during the depletion period than young rats, and this urine was significantly more dilute. Thus old rats were in relatively greater need of water than young rats before fluid access. Overnight sodium excretions were equivalent between ages, yet old rats ingested less sodium than young rats when given the opportunity. Thus, in this experiment, as inexperiment 1, old rats had greater or equivalent need of water and sodium than young rats after periods of deprivation yet drank less than young rats.

In experiment 4, captopril was added to the drinking water to stimulate additional daily intakes of sodium and water. We found that young rats increased saline drinking during captopril treatment when a choice of water or saline was provided and increased water drinking during captopril treatment when only water was available for drinking. These results are typical (11, 28, 30, 39). Old rats failed to increase water or saline intake under identical conditions. Rowland et al. (28, 30) also observed that old rats are deficient responders to converting enzyme inhibition. During this manner of captopril administration, circulating levels of renin and ANG I are greatly increased, because the drug partially prevents formation of ANG II in the circulation, thus releasing renin secretion from negative-feedback control. The increased levels of circulating ANG I probably encounter unblocked converting enzyme in circumventricular organs of the brain and undergo conversion, locally, into ANG II, thereby stimulating increased water and sodium ingestion (30,39). As noted above, in general, renin secretion is impaired in old rats, including renin secretion in response to converting enzyme inhibition (3, 28). Thus old rats may not increase ingestion of sodium or water during converting enzyme inhibition, because they do not secrete as much renin and, subsequently, form as much ANG II in circumventricular organs as young rats. However, Rowland et al. (30) found reduced salt appetite in old rats compared with young rats using a similar paradigm (converting enzyme inhibitor mixed in the chow), despite finding similarly increased plasma renin levels between the ages. So reduced renin secretion may not entirely explain the deficits in salt appetite by old rats after chronic converting enzyme inhibition. Evidence regarding centrally mediated responses to ANG II in old rats is mixed. One report (30) found that old rats drink as much water as young rats in response to ANG II administered subcutaneously. Because water drinking in response to peripherally administered ANG II is centrally mediated, this result suggests that the brains of old rats respond normally to the ANG II stimulus when it is available. However, another report (33) found that old rats drink less in response to subcutaneous ANG II. On the basis of limited evidence, it seems possible that reduced renin secretion or increased atrial natriuretic peptide levels are candidate mechanisms for the diminished intakes by old rats.

We found that old BN rats had consistently elevated daily water intakes compared with the young rats, but the daily intakes were almost exactly the same when adjusted for the age differences in body weight. Our old rats initially had highly variable daily intakes. This may be an artifact reflecting stress reactions by the old rats to the novel laboratory situation. Others have noted highly variable responses in old rats before significant handling (34). Differences in ad libitum intakes between young and old rats seem to depend on strain and whether intakes are adjusted for body weight. There are reports of increased daily water intakes in old Wistar (16), F344 (4, 30, 34), and SD rats (14, 22, 30) compared with the young animals of these strains. Others report equivalent daily water intakes between young and old Lewis (35) or Wistar rats (36) and decreased daily water intakes in old F344 × BN rats (33).

Young and old BN rats drank very little 0.3 M NaCl on a daily basis. Episodes of sodium depletion have been shown to increase subsequent ad libitum consumption of concentrated saline solutions (9,31). We found only small support for this. Although young and old BN rats had statistically significant increases in ad libitum saline intake ∼30–40 days after the sodium depletion experiments, these increased daily intakes were still small.

The overall conclusion is that aging BN rats drink similar amounts of water and saline under ad libitum conditions compared with their young cohorts but have difficulty responding to homeostatic challenges. In general, we found that old rats did not drink sufficient water or saline to make up the deficits in water and sodium accrued during the challenges. Renal responses of old rats were generally consistent with previous observations that old animals are less able to conserve water and sodium. In agreement with Rowland et al. (28, 30), old rats almost completely lack salt appetite responses to sodium depletion and low-dose converting enzyme inhibition. For some challenges, these behavioral deficits are persistent. Animal models are increasingly employed to study mechanisms of aging. The BN rat is considered an acceptable model of aging and arguably has several advantages compared with some other strains. The apparently normal ad libitum intakes of old compared with young animals of this strain and generally diminished thirst responses favor the BN rat as a useful model of thirst deficits in elderly humans.

FOOTNOTES

• Address for reprint requests and other correspondence: R. L. Thunhorst, Dept. of Psychology, University of Iowa, 11 Seashore Hall E, Iowa City, IA 52242-1407 (E-mail:[email protected]weeg.uiowa.edu).

• 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.

• First published October 10, 2002;10.1152/ajpregu.00368.2002