INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CENTRAL NEURAL MECHANISMS play an essential role in the development of hypertension in Dahl salt-sensitive (S) rats on a high salt intake. Lesions in brain areas involved in cardiovascular regulation such as the anteroventral third ventricle (AV3V) (1) and the paraventricular nuclei (10) prevent or attenuate the salt-sensitive hypertension in Dahl S rats. In Dahl S rats high salt intake causes sympathetic hyperactivity associated with an increase in brain "ouabain" (16, 19) and brain angiotensin-converting enzyme mRNA and activity (29). Inhibition of either brain "ouabain" or the brain renin-angiotensin system (RAS) prevents the sympathetic hyperactivity and hypertension (19). In addition, brain amiloride-sensitive Na⁺ channels appear to contribute to the development of salt-sensitive hypertension in Dahl S rats (9). Acute or chronic increases in cerebrospinal fluid (CSF) Na⁺ by intracerebroventricular infusion of hypertonic saline cause sympathetic hyperactivity and hypertension in Wistar rats similarly as observed in Dahl S on high salt intake (2, 15, 17), and these effects of increased CSF Na⁺ can also all be prevented by blockade of brain "ouabain" or the brain RAS (15, 17). We hypothesized that high dietary sodium may evoke central hypertensive mechanisms by increasing sodium concentration in CSF (16). If this hypothesis is correct, there seem to be at least two explanations why high salt intake would lead to sympathetic hyperactivity and hypertension only in Dahl S but not in Dahl salt-resistant (R) rats. First, high salt intake may increase CSF Na⁺ more in Dahl S versus Dahl R rats. Nakamura and Cowley (21) reported that high salt intake increases sodium concentration in the CSF significantly only in Dahl S versus Dahl R rats. However, because the increases in CSF Na⁺ were observed 1-2 days after the blood pressure (BP) started to increase, the increased CSF Na⁺ may contribute only to the maintenance of the hypertension (21). Another explanation for salt-sensitive hypertension in Dahl S rats is that the same CSF Na⁺ may induce greater sympathoexcitatory and pressor responses in Dahl S versus Dahl R rats. Indeed, the pressor response to intracerebroventricular injection of hypertonic saline was found greater in anesthetized Dahl S versus Dahl R rats (11). However, this study did not assess whether intracerebroventricular hypertonic saline elicits greater sympathoexcitatory and pressor responses in conscious Dahl S versus Dahl R rats, and whether the different responses to acute intracerebroventricular Na⁺ loading are due to increased responsiveness in Dahl S versus R rats or decreased responsiveness in Dahl R versus other normotensive rats. Moreover, more importantly, it has not yet been assessed whether in Dahl S rats chronic intracerebroventricular infusion of hypertonic saline causes sympathoexcitatory and hypertensive effects, similar as caused by high dietary sodium intake.

The objectives of the present study were therefore: 1) to examine whether acute increases in CSF Na⁺ increase renal sympathetic nerve activity (RSNA), heart rate (HR), and BP to a larger extent in Dahl S versus Dahl R or Wistar rats; 2) to compare the sympathoexcitatory and pressor responses to acute central Na⁺ loading in young versus adult Dahl or Wistar rats, because salt sensitivity in Dahl S rats appears to decrease with age (22, 30); and 3) to assess whether chronic intracerebroventricular infusion of Na⁺-rich aCSF causes similar increases in CSF Na⁺ but greater increases in brain "ouabain" and more marked sympathetic hyperactivity, impairment of arterial baroreflex function, and hypertension in Dahl S versus Dahl R rats.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Male Dahl S and Dahl R rats (Harlan Sprague Dawley; Indianapolis, IN) and Wistar rats (Charles River Breeding Laboratories; Montreal, Canada) were housed on a 12:12-h light-dark cycle and were allowed a 5-day acclimatization period on normal rat chow and tap water. All procedures in the present study were carried out according to the guidelines of the University of Ottawa Animal Care Committee for the use and care of laboratory animals.

Acute Central Sodium Loading

Early in the morning, about 1 wk after brain surgery, with the rats under halothane anesthesia, polyethylene (PE) catheters were inserted into the right femoral artery and vein. With additional methohexital sodium anesthesia given to the rat (30 mg/kg iv, supplemented with 10 mg/kg as needed, Brevital, Eli Lilly Canada; Toronto, Canada), through a flank incision, a pair of silver electrodes (A-M System; Everett, WA) was placed around and fixed to the left renal nerve with silicone rubber (SilGe1 604, Wacker; Munich, Germany) for measurement of RSNA as described previously (15). The catheters and electrodes were tunnelled subcutaneously and externalized on the back of the neck.

At least 4 h after the rat recovered from the anesthesia, the intra-arterial catheter was connected to a pressure transducer, and BP and HR were recorded through a polygraph (model 7E, Grass Instrument; Quincy, MA) and Grass 7P44 tachograph. The electrodes were linked to a Grass P511 band-pass amplifier. The amplified and filtered RSNA signals were channelled to a rectifying voltage integrator (Grass model 7P10) and recorded through the polygraph. The integrated voltage signals (in mV) of RSNA, together with BP and HR, were also fed into a online computer equipped with a Grass data acquisition and analysis program (Polyview 2.0) for display, storage, and later analysis of both analog and digital data. At the end of each experiment, noise for RSNA was measured 20 min after the rat had been killed by an intravenous overdose of pentobarbital sodium, and the actual activity was determined by subtracting noise from the total activity (15).

After a 30-min rest, baseline mean arterial pressure (MAP), HR, and RSNA were recorded in resting, unrestrained animals. Subsequently, the following four solutions were intracerebroventricularly infused at 3.8 µl/min for 10 min at 30-min intervals: aCSF, and Na⁺-rich aCSF containing 0.2, 0.3, and 0.45 M Na⁺. For intracerebroventricular infusion, a 30-gauge stainless steel was inserted into the guide cannula so that its tip protruded 0.8-1.0 mm from the tip of the guide cannula and was located inside the lateral ventricle. A 500-µl volume Hamilton microsyringe was used for intracerebroventricular infusion, mounted on a Sage 355 infusion pump. Five minutes before each of the four intracerebroventricular infusions, the vasopressin V1 receptor antagonist d-(CH₂)₅-Tyr-(Me)arginine vasopressin (30 µg/kg in 0.1 ml saline; Sigma Chemical) was injected intravenouosly to exclude the vasoconstrictor and sympathoinhibitory responses to increased endogenous vasopressin by the hypertonic saline (2, 15). The vasopressin antagonist caused transient nonsignificant decreases in BP. The resting BP, RSNA, and HR did not change significantly over the 120-150 min of the experimental period.

Chronic Central Sodium Loading

Protocol 1. At the end of 14 days of infusion, with the rat under halothane anesthesia, the right femoral artery was catheterized with PE-10 fused to PE-50 tubing for measurement of the BP. The rat was then placed in a stereotaxic frame, and a 2-cm long, 22-gauge stainless steel cannula was placed into the cisterna magna and fixed on the skull with cement and screws, as described previously (17). The upper end of the cannula was connected with a short segment of PE tubing sealed with a stylet.

Protocol 2. During the 2 wk of intracerebroventricular infusion, the rats were trained on three to four different occasions to stay quietly in a small experimental cage (24 × 15 × 8 cm) in which the rat can move back and forth. Each training session lasted 1-2 h. At the end of 14 days of infusion, the rat was anesthetized under halothane, the right femoral artery and vein were catheterized, and a pair of electrodes was placed on the renal nerve as described in Acute Central Sodium Loading. The rat was then placed in the experimental cage and basal BP, HR, and RSNA were recorded at least 4 h after the rat recovered from anesthesia and 30 min after the connection of catheters and electrodes. A standardized mental stress was then provided twice at a 10-min interval by blowing the face of the rat with a jet of air (1-1.5 psi) for 30 s (16). The average of the peak changes in MAP, RSNA, and HR in response to the two applications of stress was used. After a 15-min rest, phenylephrine (5-50 µg/min) dissolved in normal saline was infused intravenously to achieve a ramp increase in MAP with a maximum of 50 mmHg over 1-2 min. After the responses had subsided and an additional 15-min stabilization period, nitroprusside (5-100 µg/min) was infused intravenously, inducing a ramp decrease in MAP with a maximum of 50 mmHg. The infusion rate was <0.08 ml/min. Twenty minutes after the intravenous infusions, intracerebroventricular injection of guanabenz at 30 and 60 µg/3-6 µl aCSF per 10 s was performed at a 20-min interval. After the responses to 60 µg of guanabenz had subsided, the animal was allowed to rest for 40 min, and ouabain at 0.5 µg/2 µl aCSF was then injected intracerebroventricularly. There were no significant changes in resting BP, RSNA, and HR observed over the experimental period.

"Ouabain" Assay

The enzyme immunoassay microplates were coated with ovalbumin-ouabain by adding to each well 200 µl of 0.8 ng/ml ovalbumin-ouabain in coating buffer (0.01 M sodium carbonate buffer, pH 9.6). Plates were stored at 4°C for 1 day before use. Unbound ovalbumin-ouabain was removed by washing each well three times with 250 µl of rinse solution [10 mmol/l, pH 7.4, phosphate-buffered saline (PBS) containing 0.05% Tween 20]. Unoccupied protein binding sites were blocked by adding 300 µl of blocking buffer (0.01 M, pH 7.4, PBS containing 2% bovine serum albumin) to each well at 37°C for 1 h. Authentic ouabain solution was diluted to 0.125, 0.25, 0.5, 1, 2, 4, 8, 16, and 32 µg/l, respectively, with PBS. The 100 µl of samples or ouabain standards were added to successive wells. In addition, 50 µl rabbit anti-ouabain antiserum (28) (produced by immunizing rabbits with bovine serum albumin-ouabain conjugate, dilution of 1:160,000) was added to each well. Plates were incubated at 37°C for 2 h with continuous shaking followed by four rinses of 250 µl of rinse solution per well. Anti-ouabain antibodies remaining bound to the ovalbumin-ouabain were reacted with 200 µl per well of a 1:1,000 dilution of goat anti-rabbit IgG-peroxidase by incubating the plates for an additional 1 h at 37°C. Unbound anti-rabbit IgG-peroxidase conjugate was washed away by rinsing. The presence of peroxidase enzyme remaining in each well was determined by the addition of 200 µl per well 3,3',5,5'-tetramethlybenzidine base (TMB) substrate solution. The substrate reaction was terminated by the addition of 50 µl of 2 M H₂SO₄. The absorbency of each well was measured at 450 nm using an ELISA Reader (Bio-Rad Microplate Reader model 3550, Bio-Rad; Hercules, CA). The concentration of "ouabain" in each sample was calculated from the absorbency according to the ouabain standard curve.

Data Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Acute Sodium Loading

In both adult and young rats, intracerebroventricular aCSF did not change resting parameters, whereas intracerebroventricular Na⁺-rich aCSF increased MAP, RSNA, and HR in a dose-related manner (Fig. 1). The parallel increases in MAP, RSNA, and HR started 1-2 min following the start of the infusion; reached their plateau in another 1-2 min; and returned to resting levels within 2 min following termination of the infusion. Responses to the Na⁺ loading at all three rates were significantly larger in young Dahl S versus young Dahl R or Wistar rats. In adult Dahl S rats, only responses to 0.3 and 0.45 M Na⁺ were significantly larger versus adult Dahl R or Wistar rats.

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Mean arterial pressure (MAP), renal sympathetic nerve activity (RSNA), and heart rate (HR) responses to intracerebroventricular (icv) infusion of artificial cerebrospinal fluid (aCSF) (0.145 M Na+) or aCSF containing 0.2, 0.3 or 0.45 M Na+ in young (A) and adult (B) Dahl salt-sensitive (S), Dahl salt-resistant (R), and Wistar rats. bpm, Beats per minute. Data are means ± SE. For n values, see Table 1. *P < 0.05 vs. Dahl R and Wistar rats at corresponding rate of Na+ infusion. **P < 0.05 vs. Dahl R at corresponding rate of Na+ infusion

MAP, RSNA, and HR responses to 0.3 or 0.45 M Na⁺ (Figs. 1 and 2) were significantly smaller in adult versus young Dahl S rats. In contrast, responses only tended to be smaller in adult versus young Dahl R or Wistar rats (Fig. 2).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   MAP, RSNA, and HR responses to icv infusion of aCSF containing 0.45 M Na+ in young versus adult Dahl S (DS), Dahl R (DR), and Wistar rats. Data are means ± SE. For n values, see Table 1. *P < 0.05 young vs. adult rats within strain. P < 0.05 vs. age-matched DR or Wistar rats.

Chronic Sodium Loading

View larger version (11K): [in this window] [in a new window]   Fig. 3.   Hypothalamus "ouabain" content in Dahl S versus Dahl R after icv infusion of aCSF or Na+-rich aCSF containing 0.8 M Na+ for 2 wk. Data are means ± SE (n = 6 rats for all groups). *P < 0.05 vs. Dahl S with aCSF.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Peak increases in MAP, RSNA, and HR in response to ouabain (0.5 µg icv) in DS versus DR rats after icv infusion of aCSF or Na+-rich aCSF containing 0.8 M Na+ for 2 wk. Data are means ± SE (n = 6 rats for all groups). *P < 0.05 vs. others. P < 0.05 vs. DR with aCSF.

Air-jet stress elicited rapid increases in MAP, RSNA, and HR. These responses were enhanced in both Dahl S and Dahl R rats with Na⁺-rich aCSF versus aCSF. The extent of enhancement was clearly larger (P < 0.05) in Dahl S versus R (Fig. 5).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Peak MAP, RSNA, and HR responses to air-jet stress in DS versus DR after icv infusion of aCSF or Na+-rich aCSF containing 0.8 M Na+ for 2 wk. Data are means ± SE (n = 6 rats for all groups) *P < 0.05 vs. others. P < 0.05 vs. DR with aCSF.

Intracerebroventricular guanabenz caused dose-related decreases in MAP, HR, and RSNA. These responses were enhanced in both Dahl S and Dahl R rats with intracerebroventricular Na⁺-rich aCSF, but the extent of enhancement was significantly larger in Dahl S versus Dahl R rats (Fig. 6).

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Peak decreases in MAP, RSNA, and HR in response to guanabenz at 30 (A) and 60 (B) µg icv in DS versus DR rats after icv infusion of aCSF or Na+-rich aCSF containing 0.8 M Na+ for 2 wk. Data are means ± SE (n = 6 rats for all groups). *P < 0.05 vs. others. P < 0.05 vs. DR with aCSF.

Changes in MAP by intravenous phenylephrine or nitroprusside elicited reflex changes in RSNA and HR in opposite directions. As shown in Fig. 7 and Table 3, arterial baroreflex control of RSNA was impaired in both Dahl S and R rats by intracerebroventricular Na⁺-rich aCSF. The impairment was more marked in Dahl S versus R rats. Reflex control of HR showed a minor (not significant) desensitization in both Dahl S and R rats by intracerebroventricular Na⁺-rich aCSF (maximal slopes for Dahl S: 2.0 ± 0.1 vs. 1.8 ± 0.1 beats · min¹ · mmHg¹; for Dahl R: 2.1 ± 0.3 vs. 1.9 ± 0.3 beats · min¹ · mmHg¹).

View larger version (32K): [in this window] [in a new window]   Fig. 7.   Arterial baroreflex control of RSNA analyzed as a logistic model in DS versus DR after icv infusion of aCSF or Na+-rich aCSF containing 0.8 M Na+ for 2 wk. Each point is mean ± SE of changes in RSNA (%resting) relative to MAP changed by intravenous phenylephrine and nitroprusside (n = 6 rats for all groups).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study provides several new findings. First, in Wistar and Dahl rats, acute intracerebroventricular infusion of Na⁺-rich aCSF increases BP, RSNA, and HR in a Na⁺ dose-related manner, and the extent of the increases is significantly larger in Dahl S rats versus that of age-matched Dahl R or Wistar rats. Second, the extent of enhancement is significantly less in adult versus young Dahl S. Third, chronic intracerebroventricular infusion of Na⁺-rich aCSF increases Na⁺ concentration in the CSF to the same extent in Dahl S and R rats but increases hypothalamus "ouabain" only significantly in Dahl S rats and attenuates sympathoexcitatory and pressor responses to intracerebroventricular exogenous ouabain to a greater extent in Dahl S versus R rats. Fourth, chronic intracerebroventricular Na⁺ loading causes more marked sympathetic hyperactivity and hypertension associated with more impairment of arterial baroreflex control of RSNA in Dahl S rats.

We hypothesized that high salt intake activates central sympathoexcitatory mechanisms by increasing CSF Na⁺ concentration (17). Increased CSF Na⁺ may depolarize relevant excitatory neurons in the brain through amiloride-sensitive Na⁺ channels (24) leading to sympathoexcitation. This hypothesis is supported by our findings that in normotensive rats increases in CSF Na⁺ by acute (2, 15) or chronic (17) intracerebroventricular administration of hypertonic saline cause similar sympathetic hyperactivity and hypertension as seen in Dahl S on high salt (16), and blockade of brain "ouabain" or brain RAS prevents these responses (15, 17). Increases in CSF Cl or osmolarity of CSF do not cause such sympathoexcitation and hypertension (2). Differences between salt-sensitive and salt-resistant strains may relate to different increases in CSF Na⁺ on high salt intake and/or different neural responsiveness to CSF Na⁺. Nakamura and Cowley (21) demonstrated that high salt intake persistently increases CSF Na⁺ in Dahl S but not Dahl R rats. In Dahl S versus R rats on regular salt intake, the blood-brain barrier is more permeable to Na⁺, and Na⁺ uptake in brain is up to five times greater after intravenous NaCl injection (25). Similarly, CSF Na⁺ concentration is significantly higher in humans with salt-sensitive hypertension versus normal subjects (12). These studies suggest that in Dahl S rats or salt-sensitive humans a genetic abnormality of the choroid plexus altering ion transport contributes to the increased CSF Na⁺ in response to high salt intake. However, in Nakamura and Cowley's study (21), BP increased 1-2 days before significant increases in CSF Na⁺ in Dahl S rats, suggesting that the increases in CSF Na⁺ may only contribute to the maintenance of salt-induced hypertension. On the other hand, sampling of CSF was only done over 24-h periods (21), and this approch may not detect increases in CSF Na⁺ during the first few days on high salt intake, because rats consume food mainly at night and increases in CSF Na⁺ during the night may initially be masked by normal Na⁺ concentration in the CSF produced during the daytime.

Alternatively, or in addition, increases in CSF Na⁺ may evoke greater sympathoexcitation and hypertension in Dahl S versus Dahl R rats. Indeed, the pressor response to intracerebroventricular injection of a single dose of hypertonic saline was three times larger in anesthetized Dahl S versus Dahl R rats (11). In contrast, Simchon et al. (25) reported recently that intracerebroventricular bolus injection of hypertonic saline containing 4.5 M Na⁺ caused large pressor responses, but to a similar extent in Dahl S versus Dahl R rats. Neither study assessed sympathetic nerve activity. Both studies were performed in rats under general anesthesia and did not use a vasopressin antagonist. General anesthesia may markedly affect the sympathetic and pressor responses (5). An increase in plasma vasopressin (2) also contributes to the pressor responses to intracerebroventricular hypertonic saline, even with very low infusion rates (14). Moreover, in the study by Simchon et al. (25), a large volume (60 µl) of hypertonic saline with a high Na⁺ concentration (4.5 M Na⁺) was used for the intracerebroventricular bolus injection. In rats, a sudden increase of CSF volume by 60 µl likely increases intracranial pressure and thereby BP. In the present study, we demonstrated that in conscious rats increases in CSF Na⁺ by aCSF, containing 0.2-0.45 M Na⁺ infused intracerebroventricularly at 3.8 µl/min for 10 min after the use of vasopressin antagonist, increase not only BP but also RSNA and HR, in a Na⁺ dose-related manner. Moreover, the magnitudes of these responses were significantly larger in Dahl S versus Dahl R and Wistar rats but were similar in Dahl R versus Wistar rats.

Responses to acute increases in CSF Na⁺ clearly decrease in adult versus young Dahl S rats and only to a minor extent in adult versus young Dahl R or Wistar rats. Although the extent of hypertension by high salt intake has been shown to decrease with age in Dahl rats (22, 30), to our knowledge, no study has so far assessed responsiveness to central sodium loading in young versus adult rats. It is not clear why the responsiveness to CSF Na⁺ decreases with maturation in Dahl S rats. The activity of 11-hydroxysteroid dehydrogenase-1 (11-HSD1), which confers the specificity for the mineralocorticoid receptors, is significantly lower in the liver of 6- but not 10-wk-old Dahl S versus age-matched Dahl R rats (6). The 11-HSD1 mRNA in the mesenteric artery of 8-wk-old Dahl S is also lower compared with Dahl R rats (26). Because mineralocorticoids may affect activity of Na⁺ channels in the brain (20), it is possible that maturation affects expression or activity of 11-HSD1 in the brain and thereby responses to intracerebroventricular sodium. In addition, maturation may affect responses of brain "ouabain" to central sodium, as well as change the profile of brain -subunits of Na⁺-K⁺-ATPase. Higher resting BP and/or sympathetic tone in adult versus young rats may also contribute to the decreased responsiveness to CSF Na⁺. It remains to be determined whether this attenuation of the enhanced responses to CSF Na⁺ in Dahl S rats explains the smaller hypertensive response to high salt intake in older Dahl S rats.

Results obtained with chronic CSF Na⁺ loading are consistent with the acute data, showing that similar chronic increases in CSF Na⁺ caused more severe hypertension associated with enhanced sympathoexcitatory and pressor responses to air-jet stress, enhanced sympathoinhibitory and depressor responses to intracerebroventricular ₂-adrenoceptor agonist guanabenz, and more severe impairment of arterial baroreflex control of RSNA in Dahl S versus Dahl R rats. Responses to air stress and intracerebroventricular guanabenz are used to assess activity of central sympathoexcitatory and sympathoinhibitory pathways (18, 27). Enhanced inhibitory responses to exogenous ₂-adrenoceptor agonist indicate an upregulation or decreased occupancy of central ₂-adrenoceptors in hypothalamic sympathoinhibitory pathways (27), consistent with a decreased activity in central sympathoinhibitory pathways and thereby increased sympathetic outflow. The present study shows that in Dahl S versus Dahl R rats, chronic CSF Na⁺ loading causes more marked sympathetic hyperactivity and hypertension associated with a marked increase in "ouabain" in the hypothalamus of Dahl S rats. Moreover, sympathoexcitatory and pressor responses to acute intracerebroventricular injection of exogenous ouabain were attenuated only to a minor extent in Dahl R rats by chronic intracerebroventricular infusion of Na⁺-aCSF and markedly in Dahl S rats. These results are consistent with a more marked increase in occupancy of brain "ouabain" receptors by increased endogenous "ouabain," and therefore decreased availability to exogenous ouabain. Taken together, the present as well as previous studies support the concept that elevated CSF Na⁺ concentration increases brain "ouabain," and the latter causes sympathetic hyperactivity and hypertension (15-17).

How an increase in CSF Na⁺ activates brain "ouabain" and central pathways leading to sympathoexcitation is not clear yet. Recent studies show that brain amiloride-sensitive Na⁺ channels may relate to sympathoexcitation (24). Through these channels, increased CSF Na⁺ may depolarize excitatory neurons in autonomic centers and thereby increase sympathetic outflow (23). It was suggested that rats with salt-sensitive hypertension exhibit a higher activity of amiloride-sensitive Na⁺ channels in the brain (23). The neuropeptide Phe-Met-Arg-Phe-NH₂ (FMRF amide) activates Na⁺ channels in the brain and makes Wistar rats salt sensitive (24). Brain mineralocorticoid receptors also appear to play an important role in the salt-sensitive hypertension in Dahl S rats (8). Because mineralocorticoids such as aldosterone increase the activity of amiloride-sensitive Na⁺ channels (20), one may speculate that a genetic alteration in mineralocorticoid metabolism and high activity of amiloride-sensitive Na⁺ channels in the brain contribute to increased responsiveness to CSF Na⁺ in Dahl S versus Dahl R rats. The relationships among CSF Na⁺, brain "ouabain," and brain Na⁺ channels have not yet been studied. It is possible that CSF Na⁺ may increase brain "ouabain" through activated brain Na⁺ channels or, alternatively, CSF Na⁺ may increase brain "ouabain" first and the latter activates brain Na⁺ channels leading to sympathoexcitation.

The present as well previous studies indicate that several differences between Dahl S and Dahl R rats may contribute to their different sodium sensitivity. First, as shown in the present study, responsiveness to CSF Na⁺ is significantly larger in Dahl S versus Dahl R rats. Therefore, even similar increases in CSF Na⁺ by high salt intake will lead to larger increases in brain "ouabain" and sympathetic hyperactivity. Second, high salt intake may actually increase CSF Na⁺ concentration more in Dahl S versus Dahl R rats (21), which will contribute to greater increases in brain "ouabain" in Dahl S versus Dahl R rats on high salt (19). Third, whereas high salt intake impairs baroreflex control of RSNA in Dahl S rats likely through central mechanisms (16), it sensitizes arterial baroreflex function in Dahl R rats (16) possibly by peripheral mechanisms (4). Impairment of baroreflex function worsens, and sensitization of baroreflex inhibits sympathetic hyperactivity. All these differences can contribute to the sodium sensitivity in terms of sympathetic hyperactivity and hypertension in Dahl S versus Dahl R rats. Although the sodium sensitivity was similar in Dahl R versus Wistar rats in the acute study, chronic sodium loading was not performed in Wistar rats in the present study. In a previous study (17), we showed that in Wistar rats, a similar degree of chronic central sodium loading caused similar patterns of increases in CSF Na⁺, brain "ouabain" content, and resting BP, and impairment of baroreflex function as observed in Dahl R rats in the present study. The pattern of responses to high salt intake is also similar in Dahl R and Wistar rats (16, 13). Thus it seems unlikely that the difference in sodium sensitivity between Dahl S and Dahl R rats is due to decreased responsiveness in Dahl R versus normotensive Wistar rats.

The pathophysiological changes induced by high salt intake and chronic CSF Na⁺ loading are identical in Dahl S rats, but not in Dahl R rats. Whereas all relevant changes were observed in Dahl S rats by either high salt intake or chronic CSF Na⁺ loading, in Dahl R rats effects of chronic CSF Na⁺ loading and high salt intake differ. In the present study, chronic CSF Na⁺ loading increased CSF Na⁺ by 9-10 mM in both Dahl S and R rats, whereas high salt intake significantly increased CSF Na⁺ only in Dahl S but not in Dahl R rats (21). Moreover, in Dahl R rats, high salt intake sensitizes baroreflex function likely by peripheral mechanisms (4, 16), whereas chronic CSF Na⁺ loading does not affect plasma sodium concentration and does not sensitize baroreflex function but desensitizes it presumably by central mechanisms. Thus it appears that in Dahl R rats high salt does not increase CSF Na⁺ sufficiently to increase brain "ouabain" high enough to overcome the effects of baroreflex sensitization and therefore does not lead to sympathetic hyperactivity or hypertension.

A possible limitation of the study was that acute experiments were performed at least 4 h after the recovery from the anesthesia. Effects of the surgery leading to high resting sympathetic drive and possible residual effects of the short-acting anesthetic may still influence the results. In freely moving rats, the RSNA signal deteriorates over 1-2 days. Moreover, fluid and food intake are often less than optimal the first one to two nights after surgery, which may also influence the baseline values. Thus the timing of these assessments is a compromise between these factors.

In summary, the present study demonstrates that acute intracerebroventricular infusion of Na⁺-rich aCSF causes significantly larger sympathoexcitatory and pressor responses in Dahl S rats versus age-matched Dahl R or Wistar rats. Chronic intracerebroventricular infusion of Na⁺-rich aCSF increases CSF Na⁺ concentration to the same extent in Dahl S and R rats, but this infusion causes clear increases in brain "ouabain" associated with more marked sympathetic hyperactivity and hypertension and more impairment of arterial baroreflex control of RSNA in Dahl S rats. These results suggest that enhanced neuronal responsiveness to CSF Na⁺ is one of the mechanisms contributing to salt-sensitive hypertension in Dahl S rats. Genetic control of mechanisms linking CSF Na⁺ with brain "ouabain" appears to be altered in Dahl S rats toward sympathetic hyperactivity and hypertension.