INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE CENTRAL NERVOUS SYSTEM participates in the regulation of Na⁺ and water balance through its control of hypothalamo-hypophysial neuroendocrine pathways, autonomic function, and reflex modulation of the circulation. The anterior hypothalamus is thought to be a site for perception of the extracellular fluid osmolarity in goats (1), cats (37), and rats (3, 6). Intracerebroventricular injection of hypertonic NaCl evokes a cardiovascular response that consists of increases in arterial pressure and heart rate, differential regulation of sympathetic outflow, and enhanced release of arginine vasopressin (1, 3, 6, 20, 36, 37). The effects induced by intracerebroventricular administration of hypertonic NaCl have been shown to originate in the changes in cerebrospinal Na⁺ concentration but not in cerebrospinal osmolarity (6). These findings suggest the existence in the brain of neuronal elements that are primarily sensitive to transient changes in cerebrospinal fluid Na⁺ concentration and that may be involved in the pressor mechanisms, such as activation of the sympathetic nervous system and increased secretion of arginine vasopressin, although the real identity of these neuronal elements has never been demonstrated.

On the other hand, the role of cerebrospinal fluid Na⁺ concentration in Na⁺-induced chronic hypertension is not determined. Cerebrospinal Na⁺ concentrations were reported to be increased by Na⁺ intake in human salt-sensitive essential hypertension (14) and in rat models of hypertension, such as Dahl salt-sensitive hypertensive rats or uninephrectomized spontaneously hypertensive rats (27, 42), although the extents of increase in cerebrospinal Na⁺ concentrations were much smaller that those seen at intracerebroventricular infusion of hypertonic NaCl (17). In contrast, cerebrospinal Na⁺ concentration in the deoxycorticosterone acetate (DOCA)-salt hypertensive rats is reportedly not increased by Na⁺ administration (35), and alterations in cerebrospinal Na⁺ concentration did not contribute to the increase in arterial pressure induced by a high-Na⁺ diet in Na⁺-sensitive spontaneously hypertensive rats (26). Chronic Na⁺ load elicits an increase in the secretion of vasopressin (30) and norepinephrine (43), and this increased secretion is further promoted in DOCA-salt hypertensive rats (30). These activated sympathetic activities and vasopressin release by Na⁺ administration cannot be explained by the changes in cerebrospinal Na⁺ concentrations. The previous findings that the lesion of anteroventral third ventricle, which contains osmosensitive sites, reduces hypertension in Na⁺-induced hypertensive models, such as DOCA-salt (32) or Dahl salt-sensitive hypertensive rats (13), suggest that the changes in the mechanism that perceives cerebrospinal Na⁺ concentrations rather than cerebrospinal Na⁺ concentration itself may be important in the pressor mechanism of Na⁺-induced chronic hypertension.

Non-voltage-dependent amiloride-sensitive Na⁺ channels (AMNaCh), the activity of which is blocked by amiloride and their analogs, are involved in Na⁺ and water homeostasis in organs such as the colon and kidney (4). They are also present in the brain, although the physiological role of brain AMNaCh is not clear (38). These channels have four subunits (-), and more than one form of AMNaCh exists, depending on assembly of the subunits (8, 9, 39). AMNaCh reportedly play an important role not only in transmembrane transport of Na⁺ but also in Na⁺ taste transduction in lingual epithelia (2, 16). This finding indicates the possible role of brain AMNaCh in the perception of Na⁺ concentration in cerebrospinal fluid or brain tissues.

In this study, we investigated whether intracerebroventricular preinjection of benzamil hydrochloride, a specific blocker of AMNaCh, affects the hemodynamic, autonomic nervous, and endocrinological responses induced by an acute increase in cerebrospinal Na⁺ concentration at intracerebroventricular infusion of hypertonic NaCl in rats and whether continuous intracerebroventricular infusion of benzamil attenuates Na⁺-induced hypertension by affecting sympathetic nervous activity or the secretion of arginine vasopressin. We used DOCA-salt hypertensive rats and Na⁺-loaded stroke-prone spontaneously hypertensive rats (SHRSP) as Na⁺-induced hypertensive models and renovascular hypertensive rats as a non-Na⁺-induced hypertensive model.

	METHODS

Top Abstract Introduction Methods Results Discussion References

We used 12-wk-old male Wistar rats (n = 55, 250-270 g) for the experiments of acute intracerebroventricular or intravenous injections, 5-wk-old male Wistar rats (n = 37, 150-160 g) to make DOCA-salt hypertension, and 8 wk-old male Sprague-Dawley rats (n = 34, 165-195 g) to make renovascular hypertension; the rats were purchased from Oriental Bio-Service Laboratory (Kyoto). We obtained 8 wk-old male SHRSP/Izm rats (n = 36, 195-205 g) from Disease Model Cooperative Research Association (Kyoto). Sprague-Dawley rats were used for making aortic-ligated renovascular hypertensive rats because no rats except Sprague-Dawley rats have been used in this method in original study (31) and other studies, including ours (11, 28). Rats were used for each experiment at the ages of 11 or 12 wk, as described below. Rats were housed in plastic cages at a constant temperature (22°C) with a 12:12-h dark-light cycle. During the experiment, animals had free access to water and consumed a diet of rat chow (Oriental Bio-Service Laboratory). The experimental procedure was authorized by the Committee for Animal Research of Kyoto Prefectural University of Medicine.

Acute intracerebroventricular injection in anesthetized rats. Twelve-week-old male Wistar rats were anesthetized with urethan (120 mg/100 g ip; Nakarai Tesque, Kyoto, Japan) and mounted on a Kopf stereotaxic apparatus after the implantation of a femoral arterial catheter (PE-50; Clay Adams, Parsippany, NJ) filled with heparinized (50 U/ml) saline solution. Arterial pressure was continuously recorded by connecting the catheter to a small-volume displacement pressure transducer (TP-200T; Nihon Kohden, Tokyo, Japan). Heart rate was automatically calculated by triggering the femoral arterial pulse pressure with a tachometer (AT-601G; Nihon Kohden). The trachea was intubated with a cannula (PE-240; Clay Adams), and the cannula was connected to an artificial ventilator after the skeletal muscle was paralyzed by injection of decamethonium bromide (0.2 mg/100 g iv; Sigma Chemical, St. Louis, MO) to avoid the effect of spontaneous respiration on sympathetic activity. A guide cannula (23-gauge stainless steel tubing, 20 mm long with a 30-gauge stylet) was inserted into the right lateral cerebral ventricle (stereotaxic coordinates; +5.6 mm anteroposterior, +1.6 mm lateral, +2.0 mm dorsoventral, with the upper incisor bar set at 5 mm above the interaural line), and benzamil hydrochloride (0.1 nmol/kg, n = 6; 1 nmol/kg, n = 6; 10 nmol/kg, n = 7) or vehicle (n = 6) was injected into the right lateral ventricle through a cannula connected to a microsyringe 15 min before the start of intracerebroventricular infusion of hypertonic NaCl. Each injection consisted of a volume of 10 µl delivered manually over a period of 30 s. In a preliminary study, an intracerebroventricular injection of methylene blue solution densely stained the hypothalamic area and the periventricular areas of the lateral and third ventricle but stained only faintly the midbrain and the lower brain stem. Hypertonic NaCl (1.5 M, 0.75 µmol · 0.5 µl¹ · min¹) was infused for 30 min through an injection cannula (30-gauge stainless steel tubing) connected to a 25-µl syringe by an infusion pump (type B-II; Truth, Tokyo, Japan) while continuously recording the femoral arterial pressure, heart rate, and abdominal sympathetic nerve firings. As a control of hypertonic NaCl, 0.15 M isotonic NaCl was infused for 30 min intracerebroventricularly (0.5 µl/min) 15 min after the intracerebroventricular preinjection of the vehicle (n = 6). Two milliliters of blood were collected at the end of the experiment for measurement of the plasma concentration of arginine vasopressin. To investigate the effects of benzamil on the pressor response induced by intracerebroventricular injection of the pressor agent other than hypertonic NaCl, endothelin-1, which has a potent pressor action in the brain of rats (34), was intracerebroventricularly injected (1 nmol/10 µl) 15 min after the intracerebroventricular preinjection of benzamil hydrochloride (10 nmol/kg, n = 6) or the vehicle (n = 6), and mean arterial pressure, heart rate, and abdominal sympathetic nerve firings were recorded for 20 min.

Acute intravenous injection in anesthetized rats. Catheters were implanted into both the femoral artery and vein of 12-wk-old male Wistar rats anesthetized with urethan. Both catheters were filled with heparinized saline (50 U/ml). Fifteen minutes before the start of intracerebroventricular infusion of hypertonic NaCl, benzamil hydrochloride (10 nmol/kg, n = 6) or vehicle (10 µl, n = 6) was injected into a Silastic tube connected to the femoral venous catheter by directly inserting the microsyringe, and 0.1 ml of isotonic saline solution was injected through the catheter to deliver the contents into the venous circulation.

Recording of abdominal sympathetic nervous activity. The abdominal plexus was exposed through a transverse incision in the abdominal wall, and the abdominal sympathetic nerve bundle emerging from the celiac ganglion was placed over a bipolar electrode (uninsulated tips 1 mm apart). Spike potentials, which were amplified (Biophysioamplifier; NEC-Sanei Instruments, Tokyo, Japan), were monitored on a storage oscilloscope (Nihon Kohden) and continuously recorded on a magnetic tape recorder (TEAC, Tokyo, Japan) together with arterial pressure. Integrated nerve activity is expressed as the percent change from baseline counting the number of spikes for 5 s, and these percent changes were compared between groups.

Preparation of hypertensive rats. DOCA-salt hypertension was induced by treating unilaterally nephrectomized 5-wk-old male Wistar rats with 1% NaCl drinking water and with DOCA at a concentration of 150 mg/kg, administered via a subcutaneous Silastic implant. Rats were used for experiments 6 wk after the treatment, at the age of 11 wk, when sustained hypertension had developed. The methods employed are described in detail elsewhere (33). Eight-week-old SHRSP/Izm were administered 1% NaCl drinking water and were used for experiments after 4 wk of Na⁺ administration, at the age of 12 wk. Drinking water with 1% NaCl was administered both to DOCA-salt hypertensive rats and to SHR/Izm during the experiment. Systolic blood pressure and pulse rate in rats treated with DOCA-salt or SHR/Izm were measured by the tail-cuff method (MK-1000; Muromachi Kikai, Tokyo, Japan). Renovascular hypertension was induced by ligating the abdominal aorta between the right and left renal arteries of 8-wk-old male Sprague-Dawley rats. Four weeks after the aortic ligation, at the age of 12 wk, rats were anesthetized with pentobarbital sodium (50 mg/kg ip), and an arterial catheter (PE-50) filled with heparinized saline (50 U/ml) was inserted into the ascending aorta through the right carotid artery. The other end of the catheter was pulled through a cut in the skin on the back of the neck at the level of the cervical vertebrae. Arterial pressure was recorded for 10 min one time a day by connecting the catheter tip to a small-volume displacement pressure transducer as described above. Direct measurements of arterial pressure and heart rate have been done in this renovascular hypertensive rats because it was difficult to measure arterial pressure and heart rate correctly by the tail-cuff method when the abdominal aorta was ligated. The methods employed are described in detail elsewhere (28).

Continuous benzamil infusion. Under anesthesia with pentobarbital sodium (50 mg/kg ip), rats were placed on a stereotaxic frame. The skin overlying the midline of the skull was incised, and a small hole was drilled through the appropriate portion of the skull. An L-shaped infusion cannula (26 gauge, stainless steel tubing) was inserted stereotaxically into the lateral brain ventricle as described above and was fixed to the skull with cyanoacrylate adhesive (Alon Alpha; Toa Gosei Chemical Industries, Tokyo, Japan). An osmotic minipump (model 2001; Alza, Palo Alto, CA), filled with benzamil hydrochloride or vehicle, was connected to the infusion cannula and implanted subcutaneously into the back of the body. Benzamil hydrochloride (DOCA-salt: 1 nmol · kg¹ · day¹, n = 6; 10 nmol · kg¹ · day¹, n = 7; SHRSP: 1 nmol · kg¹ · day¹, n = 6; 10 nmol · kg¹ · day¹, n = 8; aortic ligation: 1 nmol · kg¹ · day¹, n = 6; 10 nmol · kg¹ · day¹, n = 6) or vehicle (1 µl/h: DOCA-salt, n = 6; SHRSP, n = 6; aortic ligation, n = 6) was infused intracerebroventricularly for 7 days in DOCA-salt hypertensive rats, SHRSP, and aortic-ligated renovascular hypertensive rats. To rule out the effect of possible leakage of the centrally administered drug into the peripheral blood, the cannula tip of an osmotic minipump containing benzamil hydrochloride (10 nmol · kg¹ · day¹) was introduced into the right jugular vein, and another osmotic minipump containing vehicle was also implanted for intracerebroventricular infusion (DOCA-salt, n = 6; SHRSP, n = 6; aortic ligation, n = 6). Metabolic studies were performed using metabolic cages made in our laboratory. Twenty-four-hour urine samples were collected to measure the diurnal urinary excretion of Na⁺, free norepinephrine, and arginine vasopressin. At the end of the experiments, rats were anesthetized with ether inhalation and killed by decapitation, and 3 ml of blood were collected for the measurement of the plasma concentration of creatinine in DOCA-salt, SHRSP, and aortic-ligated renovascular hypertensive rats and plasma renin activity in renovascular hypertensive rats. Plasma renin activity was determined using a radioimmunoassay to measure the level of angiotensin I generated (21). Plasma creatinine and urinary Na⁺ concentration were measured with an automatic analyzer (Ektachem 700 analyzer; Eastman Kodak, Rochester, NY). Urinary concentrations of free norepinephrine and arginine vasopressin were measured by high-performance liquid chromatography with electrochemical detection or radioimmunoassay, as described previously (29).

Measurement of Na⁺ and K⁺ concentration in cerebrospinal fluid. Unilaterally nephrectomized 5-wk-old male Wistar rats were divided into two groups: DOCA-salt hypertensive group that had been administered 1% NaCl drinking water and DOCA for 6 wk (n = 6) and sham-DOCA control group that had been administered distilled water and sham-DOCA for 6 wk (n = 6). Eight-week-old SHRSP/Izm were divided into a Na⁺-loaded group administered 1% NaCl drinking water for 4 wk (n = 5) and a control group administered distilled water for 4 wk (n = 5). Eight-week-old Sprague-Dawley rats were divided into aortic-ligated rats (n = 5) and sham-operated control rats (n = 5) and were used for experiment 4 wk after the operation. Either benzamil or the vehicle was not administered to all of these rats. Under the anesthesia with pentobarbital sodium (50 mg/kg ip), rats were mounted on a Kopf stereotaxic apparatus. A small, midsagittal incision was made through the skin ~7 mm below the occipital crest, and a 27-gauge needle connected to PE-20 tubing was advanced into the cisterna magna with a micromanipulator. After the needle entered the cisterna magna, a slight negative pressure was applied by a 1-ml syringe to collect cerebrospinal fluid, and great care was taken to avoid bleeding. Approximately 100-150 µl of cerebrospinal fluid were obtained for 20-30 min. Cerebrospinal Na⁺ or K⁺ concentrations were measured with an automatic analyzer (Ektachem 700 analyzer).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Effects of benzamil on intracerebroventricular infusion of hypertonic NaCl. Intracerebroventricular infusion of hypertonic NaCl caused a sustained pressor response, accompanied by increases in heart rate, abdominal sympathetic nervous activity (Fig. 1), and plasma concentration of arginine vasopressin (Fig. 2). Intracerebroventricular pretreatment with 0.1 nmol/kg benzamil did not affect the changes in hemodynamics, sympathetic nervous discharge, or plasma concentration of arginine vasopressin induced by intracerebroventricular infusion of hypertonic NaCl (Figs. 1 and 2). Intracerebroventricular pretreatment with 1 or 10 nmol/kg benzamil abolished the increases in mean arterial pressure (5 min: F = 12.854, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 10 min: F = 32.518, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 15 min: F = 57.629, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 20 min: F = 114.900, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 25 min: F = 138.739, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 30 min: F = 122.687, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01), heart rate (5 min: F = 6.527, 10 nmol/kg, P < 0.01; 10 min: F = 18.039, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 15 min: F = 19.455, 1 nmol/kg, P < 0.01, 10 nmol/ kg, P < 0.01; 20 min: F = 17.141, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 25 min: F = 11.573, 1 nmol/kg, P < 0.05, 10 nmol/kg, P < 0.01; 30 min: F = 9.573, 1 nmol/kg, P < 0.05, 10 nmol/kg, P < 0.01), abdominal sympathetic nervous activity (3 min: F = 11.042, 10 nmol/kg, P < 0.01; 5 min: F = 21.804, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 10 min: F = 26.411, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 15 min: F = 50.837, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 20 min: F = 32.968, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 25 min: F = 38.401, 1 nmol/kg, P < 0.01, 10 nmol/kg, P < 0.01; 30 min: F = 31.731, 1 nmol/kg, P < 0.01, 10 nmol/ kg, P < 0.01), and plasma concentration of vasopressin (F = 26.745, 1 nmol/kg, P < 0.01, 10 nmol/ kg, P < 0.01) compared with vehicle (Figs. 1 and 2).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Effects of intracerebroventricular preinjection with benzamil hydrochloride or vehicle on changes in mean arterial pressure (MAP; A), heart rate (HR; B), and abdominal sympathetic nervous activity (SNA; C) from baseline levels induced during intracerebroventricular infusion of 1.5 M hypertonic NaCl solution for 30 min. To show the effects of 1.5 M hypertonic NaCl solution on MAP, HR, and abdominal SNA, isotonic NaCl solution (0.15 M NaCl) was intracerebroventricularly infused for 30 min. Intracerebroventricular pretreatment with vehicle or benzamil did not affect the baseline levels between before and 15 min after the injections (at the start of intracerebroventricular infusion of hypertonic NaCl) in mean arterial pressure (vehicle, 94 ± 2 vs. 93 ± 2 mmHg; 0.1 nmol/kg benzamil, 92 ± 2 vs. 92 ± 1 mmHg; 1 nmol/kg benzamil, 94 ± 2 vs. 93 ± 2 mmHg; 10 nmol/kg benzamil, 94 ± 2 vs. 95 ± 2 mmHg), heart rate (vehicle, 388 ± 8 vs. 385 ± 7 beats/min; 0.1 nmol/kg benzamil, 387 ± 8 vs. 394 ± 4 beats/min; 1 nmol/kg benzamil, 385 ± 7 vs. 384 ± 10 beats/min; 10 nmol/kg benzamil, 388 ± 8 vs. 386 ± 7 beats/min), or abdominal sympathetic nervous activity (vehicle, 1 ± 3%; 0.1 nmol/kg benzamil, +1 ± 3%; 1 nmol/kg benzamil, 3 ± 4%; 10 nmol/kg benzamil, 2 ± 3%).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Effects of intracerebroventricular preinjection with benzamil hydrochloride or vehicle on plasma concentrations of arginine vasopressin (AVP) after a 30-min intracerebroventricular infusion of 1.5 M hypertonic NaCl solution. Isotonic NaCl solution (0.15 M NaCl) was used to show the stimulatory effect of 1.5 M hypertonic NaCl solution on vasopressin release.

Intravenous pretreatment with 10 nmol/kg benzamil or the vehicle did not affect the baseline levels between before and 15 min after the injections (at the start of intracerebroventricular infusion of hypertonic NaCl) in mean arterial pressure (vehicle, 94 ± 3 vs. 92 ± 3 mmHg; 10 nmol/kg benzamil, 96 ± 4 vs. 97 ± 4 mmHg), heart rate (383 ± 7 vs. 388 ± 6 beats/min; 10 nmol/kg benzamil, 384 ± 8 vs. 388 ± 6 beats/min), or abdominal sympathetic nervous activity (vehicle, 3 ± 3%; 10 nmol/kg benzamil, 2 ± 3%) and had no influence on the responses induced by intracerebroventricular infusion of hypertonic NaCl for 30 min in mean arterial pressure (15 min: vehicle, +15 ± 2 mmHg, 10 nmol/kg benzamil, +16 ± 2 mmHg; 30 min: vehicle, +21 ± 3 mmHg, 10 nmol/kg benzamil, +19 ± 3 mmHg), heart rate (15 min: vehicle, +21 ± 4 beats/min, 10 nmol/kg benzamil, +20 ± 5 beats/min; 30 min: vehicle, +14 ± 10 beats/min, 10 nmol/kg benzamil, +10 ± 3 beats/min), abdominal sympathetic nervous activity (15 min: vehicle, +76 ± 14%, 10 nmol/kg benzamil, +68 ± 16%; 30 min: vehicle, +56 ± 15%, 10 nmol/kg benzamil, +46 ± 11%), and plasma concentration of arginine vasopressin at the end of intracerebroventricular infusion of hypertonic NaCl (vehicle, 117 ± 29 pg/ml; 10 nmol/kg benzamil, 121 ± 36 pg/ml).

Effects of benzamil on intracerebroventricular injection of endothelin-1. Intracerebroventricular injection of endothelin-1 (1 nmol) after intracerebroventricular preinjection of the vehicle of benzamil markedly increased both mean arterial pressure and abdominal sympathetic nervous activity for 10 min (Table 1). Intracerebroventricular preinjection of 10 nmol/kg benzamil hydrochloride did not affect the changes in mean arterial pressure, heart rate, and abdominal sympathetic activity induced by intracerebroventricular injection of 1 nmol endothelin-1 (Table 1).

Effect of continuous benzamil infusion on Na⁺-induced chronic hypertension. Blood pressure, pulse rate (heart rate), and urinary excretion of arginine vasopressin and norepinephrine decreased in all experimental rats on the 1st day of continuous intracerebroventricular infusion compared with baseline levels, probably because of the anesthesia performed during the minipump placement (Figs. 3-5). In DOCA-salt hypertensive rats and SHRSP, systolic blood pressure and pulse rate returned to baseline levels on the 2nd day, and urinary excretion of vasopressin and norepinephrine returned to baseline on the 3rd day in rats with intracerebroventricular infusion of the vehicle. In DOCA-salt hypertensive rats, intracerebroventricular infusion of either 1 or 10 nmol · kg¹ · day¹ benzamil decreased systolic blood pressure (3rd day: F = 21.209, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 4th day: F = 29.742, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 5th day: F = 32.561, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 6th day: F = 28.636, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 7th day: F = 19.097, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01), pulse rate (2nd day: F = 3.718, 10 nmol · kg¹ · day¹, P < 0.05; 4th day: F = 3.814, 10 nmol · kg¹ · day¹, P < 0.05; 5th day: F = 3.884, 10 nmol · kg¹ · day¹, P < 0.05; 6th day: F = 4.818, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.05; 7 th day: F = 3.149, 10 nmol · kg¹ · day¹, P < 0.05), urinary excretion of vasopressin (3rd day: F = 3.090, 10 nmol · kg¹ · day¹, P < 0.01; 4th day: F = 5.117, 10 nmol · kg¹ · day¹, P < 0.05; 5th day: F = 5.308, 10 nmol · kg¹ · day¹, P < 0.01; 6th day: F = 6.859, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 7th day: F = 7.049, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01), and norepinephrine (3rd day: F = 6.663, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01; 4th day: F = 4.512, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 5th day: F = 4.391, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 6th day: F = 4.391, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 7th day: F = 4.499, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01) compared with vehicle (Fig. 3).

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Responses of systolic arterial pressure (A), pulse rate (B), and urinary excretion of AVP (C) and free norepinephrine (NE; D) to a 7-day intracerebroventricular infusion of benzamil hydrochloride or vehicle and to intravenous infusion of benzamil hydrochloride in deoxycorticosterone acetate (DOCA)-salt hypertensive rats.

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Responses of systolic arterial pressure (A), pulse rate (B), and urinary excretion of AVP (C) and free NE (D) to a 7-day intracerebroventricular infusion of benzamil hydrochloride or vehicle and to intravenous infusion of benzamil hydrochloride in stroke-prone spontaneously hypertensive rats.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Responses of mean arterial pressure (A), heart rate (B), and urinary excretion of AVP (C) and free NE (D) to a 7-day intracerebroventricular infusion of benzamil hydrochloride or vehicle and to intravenous infusion of benzamil hydrochloride in aortic-ligated renovascular hypertensive rats.

In SHRSP, intracerebroventricular infusion either of 1 or 10 nmol · kg¹ · day¹ benzamil decreased systolic blood pressure (4th day: F = 6.131, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 5th day: F = 6.842, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 6th day: F = 5.988, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 7th day: F = 7.935, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01), urinary excretion of vasopressin (5th day: F = 9.385, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.05; 6th day: F = 8.783, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01; 7th day: F = 8.996, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01), and norepinephrine (3rd day: F = 4.341, 1 nmol · kg¹ · day¹, P < 0.05, 10 nmol · kg¹ · day¹, P < 0.01; 4th day: F = 9.865, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01; 5th day: F = 14.652, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01; 6th day: F = 13.376, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01; 7th day: F = 8.594, 1 nmol · kg¹ · day¹, P < 0.01, 10 nmol · kg¹ · day¹, P < 0.01) compared with vehicle, although pulse rate was not affected by intracerebroventricular infusion of benzamil (Fig. 4).

In contrast, intravenous infusion of 10 nmol · kg¹ · day¹ benzamil did not affect systolic blood pressure, pulse rate, or urinary excretion of vasopressin and norepinephrine throughout the experiment compared with intracerebroventricular infusion of vehicle in DOCA-salt hypertensive rats and SHRSP (Figs. 3 and 4). Urinary Na⁺ excretion did not differ between the groups throughout the experiment in either DOCA-salt hypertensive rats or SHRSP (Table 2), and 1% saline consumption and body weight were not different between the groups throughout the experiment in either group of hypertensive rats (data not shown). The plasma concentration of creatinine was within the normal range in all DOCA-salt-treated rats and SHRSP (0.3 ± 0.08 mg/dl).

Effect of continuous benzamil infusion on arterial pressure of aortic-ligated renovascular hypertensive rats. Continuous intracerebroventricular or intravenous infusion of benzamil affected neither mean arterial pressure, heart rate, nor urinary excretion of arginine vasopressin and norepinephrine in aortic-ligated renovascular hypertensive rats (Fig. 5). Urinary Na⁺ excretion did not differ between the groups throughout the experiment (Table 2), and water consumption and body weight were not different between the groups throughout the experiment (data not shown). There was no significant difference in plasma renin activity (vehicle, 10 ± 3 ng · ml¹ · h¹; 1 nmol · kg¹ · day¹ benzamil, 11 ± 3 ng · ml¹ · h¹; 10 nmol · kg¹ · day¹ benzamil, 9 ± 2 ng · ml¹ · h¹; 10 nmol · kg¹ · day¹ iv benzamil, 10 ± 3 ng · ml¹ · h¹). The plasma concentration of creatinine was within the normal range in all of the experimental rats (0.4 ± 0.09 mg/dl).

Cerebrospinal Na⁺ and K⁺ concentrations in hypertensive rats. The Na⁺ and K⁺ concentrations in cerebrospinal fluid were not different in rats treated with DOCA and 1% NaCl drinking water or sham-DOCA and distilled water (Na⁺: 156 ± 4 vs. 152 ± 2 meq/l; K⁺: 3.3 ± 0.3 vs. 3.2 ± 0.3 meq/l), in SHRSP administered 1% NaCl drinking water or distilled water (Na⁺: 157 ± 3 vs. 154 ± 3 meq/l; K⁺: 3.3 ± 0.3 vs. 3.3 ± 0.3 meq/l), or in aortic ligated-renovascular hypertensive rats and sham-operated control rats (Na⁺: 150 ± 3 vs. 151 ± 2 meq/l; K⁺: 3.1 ± 0.4 vs. 3.2 ± 0.3 meq/l).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The precise distribution of AMNaCh in the brain is not known. An amiloride binding site reportedly resides on the extracellular loop of the -subunit of AMNaCh (22), and gene expression of the -subunit of AMNaCh has been observed in the hypothalamus in our laboratory (unpublished data). In addition, gene expression of the -subunit is specifically high in the brain (39). Intracerebroventricular preinjection of benzamil inhibited the vasopressor response induced by intracerebroventricular injection of hypertonic NaCl but not of other central pressor agents like endothelin-1; this finding indicates that the inhibitory action of benzamil injected intracerebroventricularly on the pressor response is not nonspecific to any central pressor agents but specific to the increase in cerebrospinal Na⁺ concentration. The possible relationship between benzamil-blockable brain AMNaCh and the pressor mechanism induced by an increase in cerebrospinal Na⁺ concentration and the finding that AMNaCh play a role in Na⁺ taste transduction in lingual epithelia (2, 16) suggest that benzamil-blockable brain AMNaCh function as one of the Na⁺ receptors in the central nervous system, which perceive the changes in Na⁺ concentrations in cerebrospinal fluid or brain tissues.

The correct location of osmoreceptors, including Na⁺ receptors related to the pressor response by an increase in cerebrospinal Na⁺ concentration, is not clear, but candidate sites in the brain that have osmoreceptors are the subfornical organs and anteroventral third ventricle region, which are forming a neuronal network with paraventricular and supraoptic nuclei of the hypothalamus (15, 18). Because intracerebroventricular injection of methylene blue solution stained densely the hypothalamic area and the periventricular areas of the third ventricle, the inhibitory effects of benzamil injected intracerebrally on the cardiovascular and endocrinological responses at intracerebroventricular infusion of hypertonic NaCl or in Na⁺-induced hypertensive rats may be ascribed to the blockade of AMNaCh in the hypothalamus and the periventricular areas of the third ventricle, such as the subfornical organs or anteroventral third ventricle region.

The findings concerning the inhibitory effect of intracerebroventricular preadministration of benzamil on sympathetic discharge and vasopressin secretion induced by intracerebroventricular infusion of hypertonic NaCl indicate that benzamil-blockable brain AMNaCh are involved in the regulation of sympathetic nervous activity and vasopressin release when cerebrospinal Na⁺ concentration is increased. Benzamil-blockable AMNaCh may play a role in transmission of the signals showing the change in cerebrospinal Na⁺ concentrations to the regulatory centers of sympathetic activity or vasopressin release, although the precise mechanism how changes in cerebrospinal Na⁺ concentration are converted to an electrical signal is unknown. We need further investigation to clarify whether brain AMNaCh function as one of the Na⁺ receptors in the brain, as they may do in lingual epithelia (2, 16).

The effects of continuous intracerebroventricular administration of benzamil were completely different between Na⁺-induced hypertensive rats and renovascular hypertensive rats. Chronic blockade of brain AMNaCh inhibited hypertension as well as urinary excretion of norepinephrine and vasopressin in DOCA-salt rats and SHRSP; the results indicate that benzamil-blockable brain AMNaCh play a role in maintaining the pressor mechanisms in Na⁺-induced hypertension such as activated sympathetic nervous system or vasopressin release, although a significant increase in cerebrospinal Na⁺ concentrations by Na⁺ intake was not observed in either model of Na⁺-induced hypertension.

The precise mechanism for the involvement of brain AMNaCh in the activation of sympathetic nervous system and vasopressin release is not clear. A change in tissue Na⁺ concentration in brain sites responsible for blood pressure regulation may only require a very slight, nondetectable but chronic increase in cerebrospinal Na⁺ concentration, because small changes (<2 meq) in cerebrospinal Na⁺ concentration reportedly affect the firing rate of neurons in the paraventricular and supraoptic nuclei (19), and benzamil-blockable brain AMNaCh may be involved in the perception of subtle changes in cerebrospinal Na⁺ concentrations.

Another possible mechanism is that either the activity or density of brain AMNaCh may be increased by Na⁺ intake in Na⁺-induced hypertensive rats. The activity of AMNaCh is increased by mineralocorticoids such as aldosterone or intracellular adenosine 3',5'-cyclic monophosphate (cAMP; see Refs. 4 and 12). Aldosterone activates "cryptic" AMNaCh to functional AMNaCh by membrane methylation and increases Na⁺/K⁺ selectivity (40). Arginine vasopressin can stimulate net insertion of intracellular membrane vesicles containing activated AMNaCh with high selectivity for Na⁺/K⁺ into the cell membrane by increasing intracellular cAMP (23). DOCA has mineralocorticoid action similar to that of aldosterone, and increased secretion of arginine vasopressin in DOCA-salt hypertensive rats and SHRSP (24, 41) is expected to increase intracellular cAMP via stimulation of V₂ receptors in the brain. Administration of DOCA or increased secretion of arginine vasopressin is likely to enhance the activity, density, and selectivity for Na⁺ of AMNaCh in the brain cells of DOCA-salt hypertensive rats or SHRSP, and these activated AMNaCh may be involved in the stimulation of sympathetic nervous system or vasopressin release by allowing the influx of extracellular Na⁺ into cells more than inactivated AMNaCh, even if cerebrospinal or tissue Na⁺ concentrations are not increased. In contrast, brain AMNaCh may be inactivated, the density of the channels decreased, and the selectivity for Na⁺ low in non-Na⁺-induced hypertension, such as renovascular hypertensive rats, because substances activating AMNaCh, such as aldosterone or vasopressin, do not appear to exist in high amounts. Further studies are needed to clarify the role of brain AMNaCh in the pressor mechanism of either Na⁺-induced or non-Na⁺-induced chronic hypertension.

The antihypertensive effects by blockade of brain AMNaCh in DOCA-salt hypertensive rats or SHRSP are not derived from diuretic action because urinary Na⁺ excretion did not differ between intracerebroventricular benzamil-treated and control rats. We postulate that the main antihypertensive mechanism of intracerebroventricular infusion of benzamil is derived from a decrease in the tonus of resistant vessels by reduction in sympathetic nervous activity, which was expressed as a decrease in the urinary excretion of norepinephrine. Decreased release of vasopressin may also be involved in this vasodepressor mechanism. Although vasopressin appears not to play a major role in the regulation of normal blood pressure in rats, vasopressin is expected to be involved in the pathogenesis of DOCA-salt hypertension (25) or SHRSP (24, 41), either by acting on the vascular V₁ receptors (7), by potentiating the vasoconstrictive activity of such vasoactive peptides as angiotensin II (10), or by affecting the brain (5).

In conclusion, benzamil-blockable brain AMNaCh are likely to be involved in the pressor mechanisms of Na⁺-induced hypertension. Brain AMNaCh are expected to function as one of the Na⁺ receptors in the brain and to be activated by Na⁺ intake in Na⁺-induced hypertensive models such as the DOCA-salt hypertension or SHRSP, although no direct evidence has been obtained in this study. Brain AMNaCh may be a new factor that connects Na⁺ intake with hypertension.

Perspectives