INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MODELS OF NA⁺ ingestion based on selective depletion of the extracellular fluid compartment have revealed several potential mechanisms that control Na⁺ ingestion. These models typically employ combinations of Na⁺ deprivation and depletion to induce hypovolemia. In the rat, Na⁺ ingestion after Na⁺ loss in these models is associated with increased levels of circulating ANG II and aldosterone (9), unloading of baroreceptors (20), and inhibition of oxytocin secretion by dilution of body fluids after the Na⁺-depleted animals have ingested water (42). Water deprivation is another procedure associated with an increased preference for ingesting NaCl solutions (48). During water deprivation, the obligatory excretion of water coupled with increased natriuresis results in dehydration of both the intracellular and extracellular fluid compartments (21, 25, 35). Like selective extracellular depletion models of Na⁺ ingestion, water deprivation reduces blood volume and increases plasma renin activity (PRA) and levels of circulating ANG II (8, 29, 30, 35). However, in contrast to the selective extracellular dehydration models, water deprivation is accompanied by reductions, or no changes, in plasma levels of aldosterone (29, 34) and by increases in plasma Na⁺ concentration ([Na⁺]) and osmolality (29, 34, 35). Increases in plasma [Na⁺] and osmolality inhibit Na⁺ intake (see Ref. 42 for review). Thus it is reasonable to consider whether water deprivation-induced Na⁺ intake is expressed when plasma [Na⁺] and osmolality levels are reduced after dehydrated animals drink water.

Water-deprived animals clearly prefer to drink water, but there is also considerable ingestion of hypertonic saline solutions when both are available (33, 48). A protocol has recently been developed to separate the ingestion of water (i.e., the operational definition of thirst) from the ingestion of hypertonic saline solution (i.e., the operational definition of sodium appetite) after water deprivation (33). Rats deprived of water for 24 h are first given access only to water for 2 h, and then they are given access to both water and hypertonic saline solution for an additional 2 h. During the second 2 h, the animals show a marked preference for drinking hypertonic saline compared with water (33). ANG II is likely to be an important mediator of this behavior, because PRA and circulating ANG II are elevated after 12 h of water deprivation (8, 23, 29), and the subsequent Na⁺ intake is attenuated by systemic administration of converting-enzyme inhibitor and by central injections of nonpeptide ANG II receptor antagonists (33).

Physiological changes that occur during water deprivation, such as hypovolemia and increased PRA, have not been examined in detail in relation to the appearance of sodium appetite after water deprivation. In addition, the neural basis of Na⁺ intake in the water-deprivation model of sodium appetite has not been studied.

The method of using the expression of immediate early genes in the brain [e.g., Fos immunoreactivity (Fos-ir)] has been successfully employed in many studies to identify brain areas that subserve the control of body-fluid balance (24, 26, 40, 44, 49, 50). Therefore, Fos-ir methods may aid in identifying brain areas related to water deprivation-induced Na⁺ intake. The present work investigated neural, humoral, and cardiovascular responses to water deprivation and subsequent rehydration that immediately precede Na⁺ intake after water deprivation. The results show that Na⁺ intake after water deprivation occurs when 1) PRA is elevated and blood volume is low, and 2) plasma osmolality, [Na⁺], and arterial pressure are comparable to those of nondeprived control rats. The Fos-ir in several forebrain areas was activated by water deprivation and then subsequently reduced by rehydration with water preceding Na⁺ intake. However, an exception to this observation is that Fos-ir was sustained in the subfornical organ (SFO).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Experiment 1: Water Deprivation-Induced Na⁺ Intake

Rats ingest hypertonic NaCl solutions when thirsty (32). A protocol that distinguishes between water deprivation-induced sodium appetite and thirst has been used previously (33) and will be referred to here as the water deprivation-water repletion (WD-WR) protocol. The drinking tubes containing water and 1.8% NaCl were removed from the home cage. After 24 h, food was removed, the cages were rinsed with water, and a burette of water was placed on the front of the cage. Water intake was recorded for 2 h. A burette of 1.8% NaCl was then placed on the front of the cage, and intake measurements of both water and 1.8% NaCl were determined for an additional period of 2 h (at 15-, 30-, 60-, and 120-min time points). The same protocol was used in each experiment described except when otherwise indicated.

Variations in the WD-WR protocol included one group of 6 rats that were deprived of water for 24 h and allowed only 1 h of rehydration, and another group of 6 rats deprived of water for 30 h and allowed 2 h of rehydration.

Experiment 2: Immunocytochemistry for Fos Expression

The rats were anesthetized with pentobarbital sodium (50 mg/kg) and perfused by gravity through the aorta with 100 ml of 0.1 M phosphate buffered saline (PBS) followed by 200 ml of 4% paraformaldehyde (PFA) in PBS. The brains were removed and immersed in PFA overnight and subsequently immersed in 20% sucrose-PBS for 24 h. The brains were then cut coronally in 40-µm sections on a freezing microtome.

The sections were stained for Fos-ir by the avidin-biotin-peroxidase technique. The primary antibody was raised in rabbits against the first 16 amino acids specific to rat Fos protein (supplied by Drs. G. I. Evan and D. C. Hancock, ICRF, UK). Tissue sections were soaked twice for 5 min in PBS and then incubated overnight in primary antibody (1:4,000 dilution in antibody buffer with 0.3% Triton X-100) at room temperature. The next day, the sections were washed twice with PBS, then incubated for 1 h in goat anti-rabbit serum (1:200 dilution), washed, and finally processed using the Vectastain ABC kit (Vector Labs, Burlingame, CA). The sections were treated for ~3 min in 1 mg/ml diaminobenzidine tetrahydrochloride dissolved in PBS with 0.02% hydrogen peroxide, mounted on gelatinized slides, dried overnight, dehydrated in alcohol, and covered with coverslips.

Fos-positive nuclei were counted in brain areas related to the control of Na⁺ and water balance with a computerized system (Macintosh-based NIH Image program written by Wayne Rasband), and the results were expressed as the number of positively stained cell nuclei per 10² square millimeters. Fos-positive nuclei were counted in two sections passing through the maximal anterior and posterior cross-sectional areas of the organum vasculosum of the lamina terminalis (OVLT), ventral and dorsal median preoptic nucleus (MnPO), SFO, supraoptic nuclei (SON), lateral parabrachial nuclei (LPBN), and area postrema.

Experiment 3: Blood Volume, Na⁺, and K⁺ Balance

Urine samples were collected in 0.1-ml graduated tubes from each group during the 24 h of water deprivation. Food intake in replete and dehydrated animals and 1.8% NaCl intake in replete animals were also measured during this period. A sample of ground food pellets was placed in distilled water (3 g of food in 9 ml of water) to extract the Na⁺ and K⁺ contents from the food. The sample was gently shaken during 24 h and then centrifuged. The [Na⁺] and [K⁺] of the clear phase were measured with a flame photometer (IL 343, Instrumentation Laboratories, Lexington, MA). From these values we calculated the Na⁺ and K⁺ contents of the food by assuming that all salts were in the water phase.

Experiment 4: PRA, Osmolality, [Na+], and Arterial Pressure

1

1

Arterial blood pressure was recorded directly through arterial catheters in 14 other animals. Each catheter was made from polyethylene tubing (PE-50) ~15 cm in length that was heat-welded to a shorter piece of PE-10 tubing. The PE-10 tubing was inserted into the femoral artery and advanced 4 cm. The catheters were tunneled under the skin and secured between the scapula with the free end exiting at the base of the neck. When not in use, the catheters were filled with heparinized saline (50 U/ml) and plugged with 23-gauge obturators. The rats were allowed at least 1 day to recover from surgery before testing. On the next day, the femoral catheter was connected to a Narco (model P-1000B) pressure transducer coupled to a multichannel recorder (NarcoTrace 40, Narco Bio-Systems) and a Narco biotachometer. Direct mean arterial pressure (MAP) and heart rate (HR) were recorded from the abdominal aorta in unanesthetized, unrestrained, normovolemic rats. The animals were divided into two groups. In one group (n = 8), arterial pressure was recorded for 30 min immediately before the removal of water (24 h), after 24 h of water deprivation (0 h), and after 2 h of water access (2 h). The recording intervals were the same for the second group (n = 6), which had free access to water at all times. MAP was averaged over the last 10 min of recording at each interval.

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiment 1: Water-Deprivation-Induced Na+ Intake

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Cumulative 1.8% NaCl intake measurements for rats after water deprivation for 24 or 30 h (WD 24 and WD 30, respectively) and either 1 or 2 h of rehydration (WR 1 and WR 2, respectively) with water. Cont, control (1.8% NaCl intake of rats after being deprived of 1.8% NaCl but not water for 24 h). *P < 0.05 vs. control; n, no. of rats/group is shown in parenthesis.

Experiment 2: Effect of Water-Deprivation-Induced Protocol on Fos Expression

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Number of Fos-positive immunostained cells in the organum vasculosum of the lamina terminalis (OVLT), median preoptic nucleus (MnPO), subfornical organ (SFO), and supraoptic nucleus (SON) of nondeprived replete controls, 24-h water-deprived (Deh, dehydrated) and rehydrated (Reh) rats. *P < 0.05 vs. Cont; +P < 0.05 vs. Deh; n, no. of rats/group is shown in parenthesis.

Rehydration for 2 h reduced Fos-ir by 40, 64, and 100%, respectively, in the OVLT, MnPO, and SON (n = 6; P < 0.05), but did not alter Fos expression in the SFO.

Representative photomicrographs of the Fos-positive cells in the OVLT and dorsal MnPO (top), and in the SON and SFO (bottom) are depicted in Fig. 3. In the OVLT, dehydration induced the most labeling in the periphery, and this labeling was partially reversed by rehydration. Fos labeling seen in the dorsal MnPO in dehydrated animals was partially reversed by rehydration. A similar result was obtained in the ventral MnPO (not shown). Examination of the labeling in rehydrated OVLT and MnPO suggests that the labeling is also less intense than in dehydrated animals. The strong Fos labeling in the SON during dehydration disappeared after rehydration. Labeling was scattered in the SFO during dehydration and tended to concentrate in the ventral part of the anterior third of the organ in one animal after rehydration. This trend was not observed in other rehydrated animals, although the counting of positive cells remained similar.

View larger version (120K): [in this window] [in a new window]   Fig. 3.   Photomicrographs (×40) of coronal sections through the OVLT and MnPO (A) and the SON and SFO (B) of control (Con), Deh, and Reh rats. Arrows indicate the upper limits of labeled cells.

Fos expression in water-deprived or rehydrated animals was not different from that seen in control rats (0.1 positive nuclei/10² mm²) in the LPBN, the area postrema, or the medial part of the nucleus of the solitary tract adjacent to the area postrema.

Experiment 3: Blood Volume and Na+ and K⁺ Balance in the WD-WR Protocol

Blood volume was reduced by 6.9 ± 0.7% after 22 h of water deprivation. This decrease in blood volume was only partially corrected within 2 h of rehydration (Fig. 4). Food intake was reduced by water deprivation, and the animals went into negative Na⁺ and K⁺ balance (Table 1). The 1.8% NaCl ingested (~1 mmol) in the dehydrated rats that had access to water first and then to NaCl was enough to replace blood volume and to correct slightly more than 50% of the negative Na⁺ balance that resulted from deprivation.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Changes in blood volume over a 26-h period. Samples for determination of blood volume were collected at 24, 2, 0, and 2 h. Water and 1.8% NaCl were removed from the Deh-Reh group at 24 h but not from the replete (Rep) group. Water and 1.8% NaCl were reoffered at 2 and 0 h, respectively. *P < 0.05 vs. Rep; n = 6 /group.

Experiment 4: PRA, Osmolality, Na+ Concentration, and Arterial Pressure in the WD-WR Protocol

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of these experiments confirm previous findings that sodium appetite is induced after water deprivation followed by rehydration with water (i.e., the WD-WR protocol; Ref. 33). The Na⁺ intake produced by the WD-WR protocol is not dependent on a rigid 24-h schedule of water deprivation, because a similar amount of 1.8% NaCl was ingested after 30 h of water restriction. The 2-h rehydration period is also not critical; animals ingested the same amount of 1.8% NaCl after 1 h of rehydration. Na⁺ intake was expressed after rehydration when plasma Na⁺ and osmolality were restored to normal, but PRA and blood volume were not. Water deprivation increased Fos-ir in forebrain structures related to water and Na⁺ intake and vasopressin secretion (i.e., OVLT, MnPO, SFO, and SON). Rehydration either completely or partially reversed Fos expression in the OVLT, MnPO, and SON, but not in the SFO.

Much of the work conducted to understand the mechanisms of Na⁺ intake in the rat has relied on extracellular dehydration. For example, previous studies have employed dialysis, diuretic treatment, or administration of hyperoncotic colloids (10, 11, 17, 36). Such protocols have the advantage of producing Na⁺ depletion and need-induced Na⁺ intake under clearly defined initial conditions. These initial conditions invariably involve hypovolemia and hyponatremia and subsequent hormonal responses, in particular activation of the renin-angiotensin-aldosterone system and inhibition of oxytocin release (9, 11, 42). In these protocols, sodium appetite typically occurs only after a delay of several hours subsequent to an earlier development of thirst.

Three different mechanisms have been proposed to explain sodium appetite associated with extracellular dehydration: 1) synergy between ANG II and adrenocortical hormones; 2) removal of an inhibitory factor or factors (e.g., oxytocin or reduction of [Na⁺] in the cerebrospinal fluid) after dilution of extracellular fluid; and 3) unloading of cardiovascular baroreceptors (see Refs. 9, 20, and 42 for review). The present results may provide some insights into how these factors contribute to sodium appetite induced by water deprivation.

Over the course of water deprivation, increasing tonicity of extracellular fluid appears to inhibit the production of sodium appetite (42). Osmolality and [Na⁺] of the cerebrospinal fluid may also increase during water deprivation (5), and this might contribute to the inhibition of sodium appetite (4, 5, 7, 46). The inhibition arising from hyperosmolality and hypernatremia is then reduced when rats ingest water during the rehydration phase. The amount of water ingested is not enough to restore blood volume (Experiment 3), which was expected (36). However, by diluting body fluids, the water ingestion changes the hydromineral conditions to one of extracellular dehydration only (i.e., similar to that observed in selective Na⁺ depletion models). It is worth noting that sodium appetite may appear under similar conditions in humans (37). The reduction in inhibition occurs in the presence of one or more major facilitory components arising from elevated levels of renin-angiotensin and unloading of the baroreceptors.

ANG II is a likely mediator of the Na⁺ intake produced by water deprivation. Hypovolemia is a stimulus for the release of renal renin (2). Both hypovolemia and elevated levels of PRA persisted after rehydration with water. Previous work showed that antagonism of the renin-angiotensin system inhibited Na⁺ intake in the WD-WR model (33). Aldosterone is unlikely to contribute to the sodium appetite observed in the WD-WR model because hyperosmolality and hypernatremia inhibit aldosterone secretion (29, 34). However, at this point we cannot completely rule out a synergistic participation of aldosterone or another corticoid with ANG II (22) to activate water-deprivation-induced Na⁺ intake.

An inhibitory circuit in the hindbrain has been associated with water-deprivation-induced Na⁺ intake (28). Antagonism of serotonin receptors in the LPBN increases Na⁺ intake in both furosemide-induced Na⁺ depletion and water-deprivation models of Na⁺ intake (28). The remaining hypovolemia and correction of osmolality and natremia might have contributed to Na⁺ intake by removing inhibitory signals (28, 42) in the WD-WR animals.

Arterial pressure was not different in dehydrated rats compared with controls. Therefore, potential signals arising from arterial baroreceptors are not likely to be important for facilitation of Na⁺ intake in the present experiments. However, the persistent (although partially corrected) hypovolemia could unload other systemic receptors, such as the cardiopulmonary receptors, on the low-pressure side of the circulation (20). Low-pressure volume receptors have been postulated to act in synergy with ANG II in the activation of sodium appetite (20) and the present conditions are appropriate for the expression of this synergy because elevated PRA was accompanied by persistent hypovolemia after water intake.

The increased Fos-ir found in several areas of the brain known to mediate the actions of ANG II is consistent with a role for this peptide in stimulating sodium appetite after water deprivation (33). Fos activation in the OVLT, SFO, and MnPO obtained here resembles that produced by centrally and peripherally administered ANG II and by Na⁺ depletion (24, 31, 44, 49, 50). Fos activation in the same areas occurs during water deprivation (26, 41, 49). These areas have been implicated by ablation and pharmacological studies to be involved in ANG II- and Na⁺-depletion-induced Na⁺ intake (3, 6, 12, 13, 15, 38, 39, 43, 47). The integrity of preoptic periventricular areas such as the OVLT and MnPO, which send projections to the SON, is also necessary for the expression of Fos-ir in the SON in the presence of hyperosmolality or ANG II (49). This suggests that activation of SON cells in the present work might have occurred at least in part through indirect activation of one or more of the lamina terminalis associated structures (i.e., MnPO, OVLT, and/or SFO).

One brain region that we examined but found no indication of increased Fos-ir was the paraventricular hypothalamic nucleus (PVN). This is somewhat surprising given the recognized role of the PVN in the control of body fluid and cardiovascular homeostasis. We are uncertain why there was a lack of PVN Fos expression. However, this observation may have been related to the duration of water deprivation relative to the time of death for collection of brain tissue and the time course of Fos expression or perhaps the Fos antibody.

Rehydration with water was associated with reductions in Fos-ir levels in the OVLT, MnPO, and SON compared with levels observed in these areas after dehydration. These results suggest that cells in these nuclei may be activated by hyperosmolality and hypovolemia and that this activation can be fully or partially reversed by the ingestion of water. This suggestion is consistent with the role of these areas in the control of thirst and vasopressin secretion (18, 19, 27). Hydration also inhibits ANG II-induced Fos-ir in the SON (50), but hardly induces any change in the other areas (16). Thus the reversal of Fos-ir, either partially in the OVLT and MnPO or totally in the SON, possibly depends on the reversal of more than one signal or on interactions between signals.

The persistence of Fos-ir expression in the OVLT, MnPO, and SFO after rehydration suggests that these three areas may still be processing signals that activate Na⁺ intake. Those are the same areas where Fos-ir is expressed in response to exogenous ANG II (24, 27, 31, 50). The parallel between persistence of Fos-ir expression in the brain, low blood volume, and elevated PRA after rehydration (present results) and the inhibition of water-deprivation-induced Na⁺ intake by antagonism of the renin-angiotensin system (33) suggests that ANG II, or ANG II combined with another index of hypovolemia (e.g., from cardiopulmonary baroreceptors), is a key signal.

In conclusion, the present results show that the activation of Fos-ir in brain structures subserving hydromineral balance, elevated PRA, and hypovolemia precede Na⁺ intake induced by water deprivation.

Perspectives