Renal Neurogenic Mediation of Intracerebroventricular Angiotensin II Hypertension in Rats Raised on High Sodium Chloride Diet
Abstract Chronic elevation of sodium intake may affect the sensitivity of the central nervous system to intracerebroventricular (ICV) angiotensin II (Ang II) infusion. Experiments were conducted to determine the influence of raising Sprague-Dawley rats from 2 to 3 weeks of age on low (5.0 mmol/L per kg food), normal (50 mmol/L per kg food), or high (250 mmol/L per kg food) NaCl diets on renal and cardiovascular responses to low-dose ICV Ang II infusion. At 12 weeks of age, Sprague-Dawley rats were instrumented for chronic study, including brain lateral ventricular cannulation. Artificial cerebrospinal fluid was infused (0.25 μL/min ICV) during control and recovery, whereas Ang II (20 ng/min) was infused for 5 days. During the experiment, respective sodium intakes were infused intravenously over 24 hours. In rats fed high sodium, control mean arterial pressure was 115±2 mm Hg and increased to 132±4 mm Hg by day 5 of ICV Ang II infusion. This increase in arterial pressure was associated with significant (P<.05) decreases in sodium excretion, leading to the retention of 5.4±0.6 mmol/L total sodium over the 5 days of Ang II infusion. In rats raised on low and normal sodium intakes from weaning and in 10-week-old rats exposed to a high sodium diet for only 2 weeks, arterial pressure was not increased and sodium was not retained during ICV Ang II infusion at 20 ng/min. In rats raised on the high sodium diet, bilateral renal denervation abolished the arterial hypertension and reduced the sodium retention over 5 days of ICV Ang II infusion. Thus, chronic elevation of sodium intake increases the hypertensive response to low-dose ICV Ang II infusion, which is dependent on intact renal nerves. We conclude that elevated postnatal NaCl intake enhances the pressor sensitivity of the brain to Ang II.
In 1960, Bickerton and Buckley1 first demonstrated that brain perfusion with Ang II transiently increased arterial pressure by activation of central neural structures. Numerous studies since then have demonstrated that endogenous brain Ang II may contribute to the development of certain forms of experimental hypertension. Chronic ICV Ang II infusion at high doses in normotensive rats elicits a sustained arterial hypertension.2 3 4 5 6 7 This sustained hypertension has been viewed as a function of the stimulation of sympathetic nerve activity.2 5 To achieve sustained hypertension, however, activation of renal sympathetic nerve activity should require a shift in normal renal function to an operational set point at elevated arterial pressure.8
The cardiovascular effects of centrally administered Ang II also may be related to NaCl intake. Moe9 reported that long-term changes in NaCl intake may be accompanied by changes in the density and affinity of Ang II receptors in hypothalamic areas of the brain involved in cardiovascular, fluid, and electrolyte regulation. High NaCl intake also increased arterial pressure and water ingestion after ICV Ang II infusion more in rats that were fed high NaCl chow than in rats fed chow low in NaCl.10 Bruner et al11 have demonstrated that high NaCl diets (7.5 mmol/L per day IV) exacerbate the hypertension during chronic Ang II infusion. The NaCl in this diet was 2.5 mmol/L per day greater than that used in the present study. Thus, the predominance of evidence suggests that elevation of NaCl intake is associated with exaggerated hypertension in response to CNS infusion of Ang II.
The present study was designed to further characterize the central cardiovascular effects of chronic ICV Ang II infusion in adult rats (10 to 12 weeks old) that were raised on low (5 mmol/L per kg food), normal (50 mmol/L per kg food), and high (250 mmol/L per kg food) NaCl diets from weaning (2 to 3 weeks old). This age (weanling) was chosen to begin the fixed dietary regimen because there are important developmental changes in the structure and function of the cardiovascular system, as well as in the central and peripheral nervous systems, that continue to occur beyond the postpartum and weanling age of 35 days.12 13 14 15 Furthermore, chronic activation of renal sympathetic nerve activity by Ang II may shift renal function to maintain elevated arterial pressure. Thus, lifetime exposure to elevated NaCl diets may significantly affect the regulation of blood pressure and fluid and electrolyte balance by way of changes in CNS sensitivity to Ang II. The purpose of this study was to determine whether high NaCl intake in the early postnatal period sensitizes the adult CNS to low-dose ICV Ang II (20 ng/min) infusion, resulting in sustained elevation of arterial pressure that is dependent on sympathetic outflow to the kidneys.
Sprague-Dawley rats were weaned at 21 days of age and placed on low (5 mmol/L per kg), normal (50 mmol/L per kg), or high (250 mmol/L per kg) NaCl diets (Dyets Inc). Dams were maintained on standard laboratory rat chow before gestation, during gestation, and before weaning pups. All animals were housed in Plexiglas containers in 12-h light/12-h dark cycles and temperature-controlled rooms. Rats were allowed free access to food and water. Care of the rats before and during experimental procedures was conducted in accordance with the policies of the Animal Resource Center, Medical College of Wisconsin, and the National Institutes of Health guidelines for the care and use of laboratory animals. All protocols had received prior approval by the Medical College of Wisconsin Institutional Animal Care and Use Committee. After reaching adulthood (360 to 400 g or 10 to 12 weeks of age), rats were instrumented for chronic study as detailed below and placed in individual metabolic pens.
Rats were anesthetized with ketamine HCl (60 mg/kg), xylazine (6 mg/kg), and acepromazine maleate (0.9 mg/kg) by intraperitoneal injection. All animals received an intramuscular prophylactic dose of penicillin G (20 000 U) and dihydrostreptomycin sulfate before surgery.
A small incision was made in the groin to expose the femoral artery and vein. Chronic indwelling catheters, constructed as described previously,16 were inserted into the vessels and advanced approximately 5 cm so that the tips were in the aorta and the vena cava but remained distal to the renal vessels. Catheters were tunneled subcutaneously and exteriorized near the base of the skull. The arterial catheter was filled with 0.5 mg/mL fibrinolysin solubilized in 1000 U/mL heparin to prevent clotting. The incision was closed, and rats were transferred to a stereotaxic apparatus for placement of the ICV cannula.
Heads of the rats were secured in place with the incisor bar approximately 2.5 mm above the intra-aural line. After exposure of the sutures on the skull, a guide cannula (14-cm-long, 22-gauge stainless steel tubing) was lowered into a lateral ventricle according to the following stereotaxic coordinates: anterior-posterior, 0.5 mm and lateral 1.0 mm with respect to bregma; 5.0 mm deep from the surface of the brain. The cannula was anchored in place with three gold-plated dental screws (appropriately positioned on the skull) and cranioplastic dental cement. A 26-gauge stylus was inserted into the guide cannula 0.5 mm below the tip of the guide to maintain patency.
Proper ICV cannula placement was verified at 48 hours before conducting any experiment by demonstrating short-latency, pressor, and drinking responses to a bolus injection of Ang II (100 ng). Aspiration of CSF from the guide cannula also was used to indicate correct positioning of the cannula in the lateral ventricle. At the end of each experiment, the location of the guide was confirmed postmortem after injecting 10% Alcian blue into the lateral ventricle.
A stainless steel spring connected to a 3-channel microswivel was secured to the skull to protect the guide cannula and the vascular catheters. Microswivels (Alice King Medical Arts Inc) provided free and unrestrained movement for the animal. The wound on the skull was closed with nylon suture and cleaned with hydrogen peroxide. After recovery from the anesthesia, rats were housed individually in metabolic pens, and the microswivels were attached to a holder above the pen.
In some rats, bilateral renal denervation was accomplished through a dorsal incision in the back. The muscle tissue was separated from the fascia on either side of the spine to expose the kidneys. All visible nerves leading to the pedicle of each kidney and along the adventitia of the renal artery and vein were stripped away under a stereomicroscope (magnification ×8 to ×15). The renal pedicle and blood vessels then were painted with a solution of 10% phenol in 95% ethanol. Another group of rats underwent sham denervation involving exposure of the nerves without stripping and application of saline along the vessels. The muscle and fascia were sutured together, and the skin was closed. Rats were allowed to recover after denervation or sham surgery for at least 4 to 5 days before chronic instrumentation as described above.
Immediately after instrumentation, rats were infused (30 μL/min IV) with either low (0.3 mmol/L per day), normal (1.0 mmol/L per day), or high NaCl (5.0 mmol/L per day) intakes. These sodium intakes have been determined previously to approximate closely the level of ad libitum sodium intake normally consumed by adult rats eating each of these respective laboratory chows. Sodium-free chow and water were provided ad libitum. A minimum of 3 days was allowed after surgery to ensure full recovery before the determination of sodium balance. On the fourth day after surgery, sodium balance was determined so that zero sodium balance had been achieved before the initiation of any experimental protocol.
After sodium balance was achieved, artificial CSF was infused (0.25 μL/min) during 2 control days, followed by Ang II or vehicle infusion for 5 days, at which time the Ang II was replaced with vehicle and 2 days of recovery. This general protocol was used in all rats receiving the following treatments during the 5-day experimental infusion period: (1) ICV Ang II (20 ng/min) infusion in rats raised on low, normal, or high NaCl diet from weaning, (2) ICV Ang II infusion (20 ng/min) in rats raised on a high NaCl diet with complete bilateral renal denervation or sham surgery, (3) vehicle infusion (0.25 μL/min) in the lateral ventricle of rats raised on a high NaCl diet from weaning for 8 consecutive days, (4) intravenous Ang II infusion (20 ng/min) in rats raised on a high NaCl diet, and (5) ICV Ang II infusion (20 ng/min) in adult rats raised on a normal NaCl rat chow and then fed the high NaCl diet for 2 weeks.
MAP was measured each day from the arterial catheter for 6 to 8 hours with a pressure transducer connected to an 8-channel amplifier and pressure display unit (Department of Physiology, Medical College of Wisconsin). The amplified analog signal was converted to a digitized signal (Significat model RTS-132) and analyzed on-line with a sampling frequency of 100 Hz (Significat Data Acquisition Software version 2.4). MAP was averaged over 60-second intervals throughout the experiments.
Water intake was determined daily from calibrated bottles and the 43-mL/d saline infusion. Urine was collected and the volume determined in calibrated cylinders positioned directly under the pens. Urinary sodium concentration was determined by flame photometry, and urinary sodium excretion was calculated from the product of sodium concentration and urine flow rate. Sodium intake was determined daily as the product of the volume of intravenous NaCl infused (measured daily) and the concentration of sodium (determined by flame photometry) in the infusate. Daily sodium balance was determined as the difference between sodium excretion and sodium intake.
Rats were killed with an overdose of sodium pentobarbital, and their kidneys were quickly removed and snap-frozen on dry ice. The kidneys were crushed, minced, and stored at −80°C until analysis. Tissue catecholamine content of kidneys was determined by high-performance liquid chromatography and electrochemical detection.
Data are presented as mean±SEM. Statistical significance was determined using a multiple comparison ANOVA with repeated measures. One-way ANOVA was used to analyze data that did not include more than one factor for comparison. In all cases, the Student-Newman-Keuls procedure (least-significant range) was used a posteriori to determine differences between mean values. All statistical analyses were conducted using Stats+ software (StatSoft Inc). Statistical significance was set at a value of P<.05.
Experiments were conducted to evaluate the effects of ICV Ang II infusion on renal and cardiovascular responses of rats raised from weaning on low, normal, and high NaCl diets. Preliminary experiments demonstrated that ICV Ang II infusion at 20 ng/min for 3 hours in adult rats was subthreshold for eliciting a pressor response. Therefore, this dosage of Ang II was used in all subsequent studies. Chronic ICV Ang II infusion increased MAP above control during the 5 days of infusion in rats raised on high NaCl diet from weaning (Fig 1⇓). The hypertension was sustained throughout the infusion, reaching a maximum of 132±4 mm Hg. After cessation of ICV Ang II infusion, arterial pressure returned to values (114±3 mm Hg) not different from those of control. Arterial pressure remained unchanged during low-dose ICV Ang II in rats fed either low or normal NaCl diets (Fig 1⇓).
In these same rats with fixed intravenous sodium intakes, urinary sodium excretion was determined before, during, and after ICV Ang II infusion. In rats raised on the high sodium diet, sodium excretion decreased on days 3, 4, and 5 of the ICV Ang II infusion compared with control (Fig 2⇓). The magnitude of the antinatriuresis on each day was 1.03±0.2, 1.2±0.3, and 1.04±0.3 mmol/L per day, respectively. These antinatriuretic responses to ICV Ang II provided cumulative sodium retention that totaled 5.4±0.6 mmol/L by day 5 of Ang II infusion. After cessation of ICV Ang II infusion, sodium excretion returned to values not significantly different from those of control (Fig 2⇓). Low- and normal-NaCl–fed rats did not significantly alter their sodium excretion in response to the ICV Ang II infusion, which was consistent with the observation that ICV Ang II also did not increase MAP at these lower levels of sodium intake.
Experiments also were conducted to determine whether altered responsiveness of the brain to low-dose ICV Ang II after high NaCl feeding was specific to the early postnatal exposure to high NaCl diets. Fig 3⇓ compares the pressor response of adult rats raised on a high NaCl diet from 21 days of age (weaning) for 10 weeks and adult rats 11 weeks of age and fed normal NaCl chow, at which time they were maintained on the same high NaCl diet for only 2 weeks before the study. Adult rats exposed to the high NaCl diet for 2 weeks did not show increased arterial pressure during chronic low-dose ICV Ang II infusion (Fig 3⇓).
Time-control experiments also were conducted in which comparable volumes of vehicle (artificial CSF) were infused ICV for 8 consecutive days or Ang II (20 ng/min) was administered for 5 days intravenously. The findings of these experiments are compared with blood pressure responses to ICV Ang II infusion in rats raised on the high sodium diet (Fig 4⇓). ICV vehicle infusion alone did not significantly alter arterial pressure in rats fed high NaCl diets from weaning, nor did intravenous Ang II infusion alter MAP in rats chronically fed high NaCl (Fig 4⇓).
To evaluate whether the pressor and antinatriuretic responses to ICV Ang II infusion were caused by the stimulation of brain sites controlling sympathetic outflow to the kidneys, similar experiments were conducted in adult rats fed high sodium diets from weaning with bilateral renal denervation or sham surgery. In sham-operated rats, ICV Ang II infusion increased arterial pressure from 113±3 mm Hg in control to a peak of 128±3 mm Hg on day 5 of the infusion period (Fig 5⇓). After cessation of the Ang II infusion, blood pressure returned to control values. In rats with bilateral renal denervation, ICV Ang II infusion failed to alter arterial pressure despite these animals being fed a high NaCl diet from weaning (Fig 5⇓). The cumulative sodium balance was compared with the magnitude of the hypertension in renal denervated and sham-operated rats. Arterial pressure in the sham-operated rats increased by an average of 15.1±3.3 mm Hg, and the rats retained 4.7±0.4 mmol/L of sodium during the 5 days of ICV Ang II infusion. In contrast, the arterial pressure in rats with denervated kidneys did not change significantly, and the rats retained only 2.4±0.3 mmol/L of sodium during 5 days of ICV Ang II infusion (Fig 6⇓). Renal norepinephrine content in the sham-denervated rats averaged 93.4±6.7 pg/mg tissue, and kidneys of rats after denervation averaged 7.26±2.4 pg/mg tissue, thus documenting effective bilateral renal denervation.
The present study was designed to evaluate the influence of chronic high NaCl feeding on ICV Ang II– mediated hypertension in rats. Important developmental changes in the structure and function of the cardiovascular system, as well as the central and peripheral nervous systems, continue to occur up to and beyond the postpartum age of 35 days in rats.12 13 14 15 Consequently, young animals may be more sensitive to excess dietary NaCl intake and extracellular fluid volume compared with animals raised on more modest levels of NaCl.17 Therefore, elevated NaCl diets may significantly affect the regulation of blood pressure and fluid and electrolyte balance by way of changes in the sensitivity of the brain to circulating or centrally generated humoral factors.
Infusion of Ang II at low doses increased arterial pressure to 133±3 mm Hg in rats raised from weaning on a moderately elevated NaCl diet, and this pressor response was associated with a total sodium retention of 5.4±0.6 mmol/L over 5 days. Similar pressor and antinatriuretic responses to Ang II infusion were observed in rats chronically fed a high NaCl diet and that had undergone sham renal denervation. These results were in direct contrast with those of rats raised on normal or low sodium chow, in which low-dose ICV Ang II infusion had no effect on arterial pressure or urinary sodium excretion. Chronic ingestion of this 1.5% NaCl diet did not influence the arterial pressure response to intravenous administration of this dosage of Ang II, nor was arterial pressure changed in these rats during ICV infusion of vehicle. Although others have shown low-dose Ang II–mediated pressor responses during ingestion of 4% to 8% salt diets,10 18 no studies have reported pressor Ang II activity at 20 ng/min IV during NaCl ingestion at 2% or less. Therefore, specific altered responsiveness of brain structures to Ang II must have occurred during the postnatal development of rats raised on the elevated NaCl intake.
To determine the necessity for lifetime or chronic high dietary sodium beginning with the postnatal period of life, the effects of ICV Ang II (20 ng/min) infusion in adult (10 to 12 weeks old) rats exposed to the 1.5% NaCl diet for only 2 weeks were evaluated. In these rats, low-level Ang II infused for 5 days did not significantly change either arterial pressure or urinary sodium excretion. These results document that at least in the long-term, postnatal high dietary NaCl must be required for the development of increased sensitivity of the brain to Ang II.
The hypertensive response to chronic ICV Ang II is consistent with findings of other investigators.11 19 Furthermore, previous reports document that the magnitude of the increase in blood pressure after ICV Ang II infusion is dependent on the level of NaCl intake.10 11 14 The modest level of high NaCl (1.5%) administered in this diet over a long time period and the low dosage of the Ang II infusion are unique to the present study. The dietary regimen of rats on a high sodium diet (250 mmol/L per kg) represents an approximate level of 5 mmol/L per day of sodium intake at the adult age. Other studies have shown that central Ang II infusion and high NaCl diets ranging from 4% to 8% also elicit hypertension.10 11 Other studies of ICV Ang II infusions coupled with elevated NaCl intake have often used dosages at least fivefold greater than the 20-ng/min dosage used in the present study.11 19 Therefore, the present data document that particularly modest increases in dietary NaCl administered over long time periods, coupled with only moderate increases in brain Ang II, can synergistically mediate significant hypertension for the duration of the elevated central Ang II infusion.
The mechanisms by which high sodium intake altered the brain responsiveness to chronic elevation of Ang II remain unknown. Rats raised from weaning on diets of excess NaCl may show alterations in the endogenous brain angiotensin system. Elevated NaCl alone may increase brain expression of Ang II receptor populations. Slaven20 showed that sodium depletion increased Ang I concentrations within the brain stem, whereas sodium loading decreased Ang I. This increase in angiotensin concentration might directly or indirectly alter the receptor characteristics in the brain such that high sodium intake may increase the number and/or affinity of brain Ang II receptors.10 21 22 Several investigators have shown that whole-brain Ang II receptor capacity increases with high NaCl and decreases with low NaCl diets without significant changes in receptor affinity in either group.10 21 Functionally, this increase in receptor population translates into significant increases in pressor and dipsogenic responses in high NaCl–treated rats, whereas in the low NaCl–fed rats, these same responses were reduced.10
In the present study, the pressor as well as the antinatriuretic responses to ICV Ang II were abolished by bilateral renal denervation, suggesting that these responses were directly dependent on intact renal innervation. The data provide a correlation between renal neurogenic sodium retention and the development of arterial hypertension in these rats raised on a high sodium diet, which would occur in response to activation of efferent renal sympathetic nerve activity. Other studies, however, have shown that renal sodium retention does not contribute to chronic ICV Ang II hypertension.11 21 Bruner et al11 reported that chronic ICV Ang II infusion in rats (100 ng/min) significantly increased arterial pressure without altering sodium excretion, plasma sodium, or plasma potassium concentrations. In rabbits, 10 days of high-dosage ICV Ang II infusion was associated with marked and persistent natriuresis and hypokalemia.23 It is likely that the magnitude of the hypertension in these studies provided a level of pressure natriuresis sufficient to override any neurogenic sodium-retaining influences of the ICV Ang II observed in the present study.
Other studies using normotensive rats have shown that centrally administered Ang II (10 μg) increases renal sympathetic nerve activity and intrarenal renin secretion.24 Both fourth ventricular and rostral ventrolateral medullary microinjection of Ang II increases arterial pressure and renal sympathetic nerve activity.25 26 These actions of Ang II were reversed by the injection of the Ang II antagonist saralasin. The antinatriuretic effect of ICV Ang II in the present study could be attributed to increased renal sympathetic outflow that overcomes the potential for pressure natriuresis because renal nerve activity and adrenergic mechanisms are known stimuli for sodium reabsorption.27
Another possible explanation for the differences in the renal handling of sodium in response to chronic ICV Ang II infusion between this study and others could be the length and time of exposure to the different NaCl diets. Huang and Leenen13 suggested that feeding high or low sodium diets at 4 weeks of age or earlier, when the sympathetic control of the cardiovascular system remains immature, elicits larger changes in sympathetic function than when altered sodium intakes are fed to older rats with established sympathetic nervous systems. That is, early exposure to high NaCl diet significantly attenuated baroreflex functions in blood pressure control. These alterations in the cardiovascular reflexes directly affect the CNS and efferent sympathetic outflow to the kidneys.28 29 Taken together, these observations suggest that alterations in sodium intake and body fluid balance in early life produces long-term alterations in the set point for control systems affecting sodium/volume regulation and cardiovascular homeostasis.
In summary, this study provides the first evidence that chronic high NaCl fed to rats in the early postnatal period sensitizes brain structures to ICV-administered Ang II, eliciting hypertension specifically in rats fed the high sodium diet. The hypertension was associated with a decrease in sodium excretion and an increase in cumulative sodium retention during 5 days of Ang II infusion. Bilateral renal denervation abolished the hypertension and markedly reduced the total cumulative sodium retained during the ICV Ang II. Because intravenous Ang II infusion did not increase arterial pressure or decrease sodium excretion, the effects of Ang II were mediated by actions in the brain. In conclusion, postnatal elevation of NaCl intake sensitizes central neural structures to Ang II, which mediates hypertension by mechanisms dependent on intact renal innervation. This hypertension may be related at least in part to neurogenic reductions in sodium excretion and expansion of extracellular fluid volume.
Selected Abbreviations and Acronyms
|Ang I, II||=||angiotensin I, II|
|CNS||=||central nervous system|
|MAP||=||mean arterial pressure|
This work was supported in part by a National Institutes of Health minority graduate research assistant supplement and by the National Heart, Lung, and Blood Institute (PO1-29587 and NHLBI RO1-40137). The authors thank Erez Gordin for his technical and computer assistance. We also extend our appreciation to the following individuals for their technical assistance: Terry Kurth, Harold Eick, and David Eick.
Reprint requests to Jeffrey L. Osborn, PhD, Department of Physiology, Medical College of Wisconsin, 8701 Watertown Plank Rd, PO Box 26509, Milwaukee, WI 53226-0509.
- Received August 26, 1996.
- Revision received September 20, 1996.
- Accepted December 10, 1996.
Bickerton RK, Buckley JP. Evidence for central mechanism in angiotensin induced hypertension. Proc Soc Exp Biol Med. 1961;106:834-836.
Ganten D, Hutchinson JS, Schelling P. The intrinsic brain iso-renin-angiotensin systems in the rat: its possible role in central mechanisms of blood pressure regulation. Clin Sci Mol Med. 1975;48:265s-268s.
Mizelle HL, Montani JP, Hester RL, Didlake RH, Hall JE. Role of pressure natriuresis in long-term control of renal electrolyte excretion. Hypertension. 1993;22:102-110.
Moe KE. The salt intake of rat dams influences the salt intake and brain angiotensin receptors of their adult offspring. Soc Neurosci Abstr. 1987;13:1169. Abstract.
Mann JFE, Schiffrin EL, Schiller PW, Rascher W, Boucher R, Genest J. Central actions of angiotensin II: Influence of sodium intake. Hypertension. 1980;2:437-443.
Zicha J, Kunes J, Jelinek J. Experimental hypertension in young and adult animals. Hypertension. 1986;8:1096-1104.
Fink GD, Bryan WJ, Mokler DJ. Effects of chronic intracerebroventricular infusion of angiotensin II on arterial pressure and fluid homeostasis. Hypertension. 1982;4:312-319.
Weekly LB. Angiotensin II acts centrally to alter renal sympathetic nerve activity and intrareal renin-angiotensin system. Cardiovasc Res. 1991;25:353-363.
Aki S, Dampney RAL. Toxic cardiovascular effects of angiotensin II in the ventrolateral medulla. Hypertension. 1990;15:274-285.
Osborn JL, Holdaas H, Thomas MD, DiBona GF. Renal adrenoceptor mediation of antinatriuretic and renin secretion responses to low frequency renal nerve stimulation in the dog. Circ Res. 1983;59:298-305.