Role of renal nerves in regulation of vasopressin secretion and blood pressure in conscious rabbits SHIGEAKI

MATSUKAWA,

LANNY

C. KEIL,

AND

IAN

A. REID

Department of Physiology, University of California, San Francisco 94143; and Ames Research Center, Moffett Field, California 94035

MATSUKAWA, SHIGEAKI, LANNY C. KEIL, AND IAN A. REID. Role of renal nerves in regulation of vasopressin secretion and blood pressure in conscious rabbits. Am. J. Physiol. 258 (Renal

crease vasopressin release was provided by Day and Ciriello (5), who observed that afferent renal nerve stimulation in anesthetized rats caused an increase in the Fluid Electrolyte Physiol. 27): F821-F830, 1990.-The obser- activity of hypothalamic neurons, including neurosecrevation that electrical stimulation of the renal nerves increases tory vasopressin cells in the supraoptic nucleus. Subsevasopressin secretion raises the possibility that the renal nerves may participate in the control of vasopressin secretion. In the quently, this group reported that afferent renal nerve in anesthetized, sinoaortic-denervated, vapresent investigation, the effects of renal denervation on the stimulation vasopressin response to two reflex stimuli (nitroprusside infu- gotomized cats produced a large increase in plasma vasopressin concentration (4). Very recently, they reported sion and hemorrhage) and two osmotic stimuli (hypertonic saline infusion and water deprivation) were studied in con- that afferent renal nerve stimulation in conscious rats scious, chronically prepared rabbits. Nitroprusside infusion in with intact baroreceptor reflexes caused a significant 13 intact and 14 denervated rabbits caused similar decreases in increase in plasma vasopressin concentration (32). They mean arterial pressure (MAP) and the increase in plasma have also reported that intrarenal infusion of bradykinin arginine vasopressin concentration (PA& in intact (2.6 t 0.3 or capsaicin, which are known to activate the afferent to 5.8 t 0.9 pg/ml, P < 0.01) and denervated (2.8 t 0.3 to 5.7 renal nerves, caused excitation of neurosecretory cells in t 1.3 pg/ml, P < 0.01) rabbits was not significantly different. Hemorrhage (20 ml/kg) in 15 intact and 14 denervated rabbits the supraoptic nucleus (6). It has been reported that the activity in the afferent caused similar decreases in MAP. Again, the increase in PAvp receptors are from 2.7 2 0.3 to 159.0 t 37.1 pg/ml (P < 0.01) in intact and renal nerves increases when intrarenal activated by various stimuli, including a reduction of from 5.0 & 1.7 to 115.4 t 45.6 pg/ml (P < 0.01) in denervated renal perfusion pressure and perfusion of the renal pelvis rabbits was not significantly different, nor was the relationship between PAvP and MAP in the two groups. In seven intact with hypertonic solutions (1, 14, 19, 26, 27, 34). Since it rabbits, hypertonic saline infusion increased PAvP from 4.0 $- is known that reductions in systemic blood pressure or 0.9 to 10.9 t 2.8 pg/ml (P < 0.05). The change in six denervated intravenous infusion of hypertonic solutions increase rabbits was not significantly different, nor was the relationship vasopressin secretion (30, 31), these observations raise between PAvpand plasma osmolality. During water deprivation the possibility that the afferent renal nerves participate (24 h) in six intact rabbits, P AvPincreased from 4.0 t 0.7 to 6.9 in the regulation of vasopressin secretion. t 0.6 pg/ml (P < 0.05). Again, the increase in PAvp in six Indirect evidence for a role of the renal nerves in the denervated rabbits was not significantly different from that in control of vasopressin secretion has been provided by the intact rabbits. The change in MAP during water deprivation in the two groups was also not significantly different. Renal studies in nephrectomized animals. Quillen et al. (24,25) cortical norepinephrine concentration in denervated kidneys reported that the sensitivity of the osmotic control of was Cl0 rig/g wet wt. These results indicate that, in conscious vasopressin secretion, assessed by infusion of distilled rabbits, renal denervation does not impair the osmotic or reflex water or hypertonic saline, was severely reduced in neregulation of vasopressin secretion, nor does it interfere with phrectomized dogs compared with intact dogs. Similarly, blood pressure regulation during hypovolemia or hypotension. Hashemzadeh-Gargari et al. (10) observed that the increase in plasma vasopressin concentration in response renal denervation; hemorrhage; plasma osmolality; water dep- to hypotension induced by ganglionic blockade in anesrivation; nitroprusside; renal receptors; afferent renal nerves thetized rats was significantly attenuated by nephrectomy. In both cases, the authors speculated that the attenuation of the vasopressin responses in nephrectoIN A PREVIOUS STUDY in this laboratory, we observed mized animals may have resulted from elimination of the that electrical stimulation of the renal nerves in anes- renal nerves. The first goal of the present study was to thetized dogs caused a twofold increase in plasma vaso- investigate the role of the renal nerves in the osmotic pressin concentration (28, 29). This increase did not and reflex regulation of vasopressin secretion. appear to result from the accompanying increase in A second goal of this study was to investigate the role plasma renin activity or from decreased renal clearance of the renal nerves in the regulation of blood pressure. of vasopressin, and it was therefore proposed that it Increased activity of the efferent renal nerves is known resulted from stimulation of afferent renal nerves. Evi- to increase renal vascular resistance, renin secretion, and dence that stimulation of afferent renal nerves can in- sodium reabsorption (7, ll), whereas activation of the F821 Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

F822

RENAL

NERVES

AND

VASOPRESSIN

afferent renal nerves increases the activity of the sympathetic nervous system (9, 13) and, as discussed above, vasopressin secretion. It is known that the activity of the afferent and efferent renal nerves increases during hypotension and in other states (1, 26, 27, 34), and it is therefore possible that the renal nerves contribute to blood pressure regulation in these situations through the mechanisms just described. There is evidence that both the afferent and efferent renal nerves do play significant roles in several forms of experimental hypertension (9, 11,13,20). However, the role of the renal nerves in blood pressure regulation during hypotension or hypovolemia has not been studied in detail, particularly in conscious animals. In the present study the effects of renal denervation on vasopressin secretion and blood pressure regulation during sodium nitroprusside infusion, hemorrhage, hypertonic saline infusion, and water deprivation were investigated in conscious, chronically prepared rabbits. MATERIALS

AND

METHODS

The experiments were carried out in 64 male New Zealand White rabbits weighing 2.1-3.4 kg. Standard laboratory rabbit chow (Purina Rabbit Chow, St. Louis, MO) and tap water were available ad libitum, unless water intake was restricted by experimental design. Surgical Procedures

The animals were premeditated with acepromazine maleate (2 mg/kg SC) and anesthetized with pentobarbital sodium (25 mg/kg iv). Under sterile conditions, two silicone catheters (0.76 mm ID X 1.65 mm OD) connected to polyethylene tubing (0.76 mm ID X 1.22 mm OD) were inserted into the abdominal aorta and the inferior vena cava via the femoral artery and the external jugular vein, respectively. The tip of the arterial catheter was positioned below the renal arteries and that of the venous catheter was positioned near the diaphram. The cannulas were then exteriorized dorsally between the scapulae and protected by a nylon mesh jacket. Catheters were filled with heparin (1,000 U/ml) and flushed with sterile isotonic saline at least every other day. After surgery, the rabbits were treated with 0.3 ml trimethoprim-sulfadiazine (DI-TRIM, Syntex, West Des Moines, IA) for 3 days. In 36 animals, bilateral renal denervation was performed under halothane anesthesia. The kidneys were exposed via a midline abdominal incision, and all visible nerve fibers were stripped from the renal vessels, which were then painted with 10% phenol in ethanol. Three to 7 days later, catheters were placed as described above. In 11 rabbits, experiments were performed before and after renal denervation in the same animals. In these rabbits, renal denervation was performed at least 3 days after the initial experiment. No more than three experiments were performed in any one animal, and at least 3 days elapsed between experiments. In renal denervated rabbits, all experiments were performed between 3 and 14 days after denervation. After the last experiment, the kidneys were removed for the determination of renal cortical norepinephrine concentration.

SECRETION

Experimental

Protocols

The rabbits were accustomed to the laboratory environment before experiments were begun. On the day of the experiment, the rabbits were brought to the laboratory and loosely restrained in a cage. A stabilization period of at least 30 min was allowed before the start of each experiment. Throughout the control, experimental, and recovery periods, pulsatile and mean arterial pressure and heart rate were continuously recorded via the arterial cannula using Statham or Cobe pressure transducers and a Grass polygraph (Grass Instruments, Quincy, MA). In some experiments, cardiovascular data were digitized at 100 Hz, collected, and analyzed using a PDP 11/23 Plus computer system (DEC, Maynard, MA). Except in the hemorrhage study (see below), blood samples were collected via the arterial catheter and replaced with an equal volume of sterile isotonic saline. Blood samples (3.6 ml) were divided into 2.7. and 0.9-ml aliquots. The larger aliquot was placed in a tube containing 0.3 ml 0.3 M EDTA for the determination of plasma arginine vasopressin concentration and plasma renin activity, and the smaller aliquot was placed in a tube containing heparin for the determination of plasma osmolality and plasma sodium and potassium concentrations. The aliquots were placed on ice and centrifuged at 4°C within 15 min, and the plasma was frozen until assayed. Hematocrit was also measured in the hemorrhage and water deprivation experiments. The intact and renal denervated rabbits were subjected to identical procedures using the following protocols. Nitroprusside infusion. The effects of nitroprusside infusion were investigated in 13 intact and 14 denervated rabbits. After arterial blood pressure and heart rate had stabilized, a ZO-min control period was begun. At the end of this period, a control blood sample was withdrawn and then an intravenous infusion of sodium nitroprusside (Elkins-Sinn, Cherry Hill, NJ) dissolved in saline was started at 1.0 pg* kg-’ l rein-1 in a volume of 0.068 ml/ min using a Harvard infusion pump. Twenty minutes later, another blood sample was collected and then the dose of nitroprusside was increased to 3.0 fig* kg-’ min? After 20 min, another blood sample was collected and the dose of nitroprusside increased to 10.0 pg* kg-‘. min? After another 20 min, a final blood sample was collected and the nitroprusside infusion was stopped. Arterial blood pressure and heart rate were then recorded during a ZO-min recovery period. Hemorrhage. The effects of hemorrhage were investigated in 15 intact and 14 denervated rabbits. After the 30-min stabilization period, heparin sodium (1,000 U) was injected intravenously. Twenty minutes later, a control blood sample was withdrawn from the venous catheter and replaced with an equal volume of saline. Immediately after collection of the control sample, an additional 1,000 U of heparin sodium was injected and the rabbit was then bled continuously from the venous catheter at a rate of -1.0 ml. kg-‘. min-’ using a Harvard infusion-withdrawal pump. Blood samples for analysis were collected from the venous catheter after 10 and 20 ml blood/kg body wt had been withdrawn. These blood samples were considered to be part of the hemorrhage

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

l

RENAL

NERVES

AND

VASOPRESSIN

and were not replaced with saline. After the end of the hemorrhage, blood was reinfused at the same rate as blood withdrawal. Arterial blood pressure and heart rate were recorded during this period. Hypertonic saline. The effects of hypertonic saline infusion were investigated in seven intact and six denervated rabbits. Control measurements of arterial blood pressure and heart rate were made for 20 min, and a control blood sample was collected at the end of this period. Hypertonic NaCl (2 es/l) was then infused via the venous catheter at a rate of 0.25 ml/min for 40 min; i.e., a total of 20 meq NaCl was infused in a volume of 10 ml. Blood samples were taken after 20 and 40 min of the infusion period, i.e., when 10 and 20 meq NaCl had been infused. After this period, arterial blood pressure and heart rate were recorded for another 20 min. Water deprivation. The effects of water deprivation were studied in six intact and six denervated rabbits. Arterial blood pressure and heart rate were monitored during a 20-min control period, and a control blood sample was collected. The rabbits were then deprived of water for 24 h. At 12 and 24 h of this period, arterial blood pressure and heart rate were measured for 20 min and a blood sample was collected. The rabbits were then allowed to rehydrate for 24 h, after which arterial blood pressure and heart rate were again recorded for 20 min. In this experiment, body weight was also measured during the control, experimental, and recovery periods. Analytical

RESULTS

Nitroprusside

Infusion

The effects of sodium nitroprusside infusion on mean arterial pressure, heart rate, plasma renin activity, and plasma vasopressin concentration in intact and denervated rabbits are summarized in Fig. 1. In the intact rabbits, the three doses of sodium nitroprusside caused step reductions in mean arterial pressure, the mean value decreasing from 75 t 2 to 58 t 3 mmHg at the highest dose (P < 0.01). These decreases were accompanied by increases in heart rate from 207 t 5 to 292 t 8 beats/ min at the highest dose (P < 0.01).Twenty minutes after cessation of the sodium nitroprusside infusion, mean arterial pressure increased to 76 t 3 mmHg and heart rate decreased to 234 t 7 beats/min. These hemodynamic changes were accompanied by increases in plasma vasopressin concentration from 2.6 t 0.3 to 5.8 t 0.9 pg/ml

** L

8 I

6-

Plasma Vasopressin

4-

Cont. (pg/ml) 25

Methods

Renal cortical norepinephrine concentration in the intact and denervated kidneys was measured using the method of Peuler and Johnson (22).Plasma vasopressin concentration was measured bY radioimmunoassay as described previously (12). Plasma renin activity was measured by radioimmunoassay using a slight modification of the method of Menard and Catt (17) and was expressed as nanograms of angiotensin I generated Per milliliter of plasma during a 2-h incubation at 37°C and pH 6.5. Plasma osmolality was determined by freezingpoint depression, and plasma sodium and potassium concentrations were determined by flame photometry. Hematocrit was measured by a microcapillary technique.

F823

SECRETION

20

1

0

Intact Donmmtod

t Plasma

Renin Activity (ng/ml/2h)

*

300 t Heart

Rate (bpm)

250

*

200

90

Statistical Analysis

80 I

All data are expressed as means t SE. Statistical evaluation of the data was performed using one-way and two-way analysis of variance for repeated measures and the Neuman-Keuls multiple range test (36). Where appropriate, the unpaired t test was used to identify significant differences between groups. Linear regression analysis was performed using the least-squares method. Vasopressin data were transformed logarithmically before analysis of the relationship between plasma vasopressin concentration and mean arterial pressure in the hemorrhage study. A P value of ~0.05 was considered to be statistically significant.

Mean

AT

70

*

*

u

60 Ptessure (mmHg)

50

Arterial

40 ~

I

u

Ill

0 Dose

: of

Nitroprussi

3

10

Recovery

e (ug/kg/min)

FIG. 1. Effect of nitroprusside infusion on mean arterial pressure, heart rate, plasma renin activity, and plasma vasopressin concentration in intact (n = 13) and denervated (n = 14) rabbits. * P < 0.05 compared with control; t P < 0.05 compared with intact group. Mean arterial pressure, heart rate, and plasma vasopressin concentration did not differ between the 2 groups, but increase in plasma renin activity was significantly reduced in denervated animals.

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

F824

RENAL

NERVES

AND

VASOPRESSIN

(P < 0.01.) and plasma renin activity from 5.2 t 0.9 to 19.9 t 3.6 ng- ml-’ .2 h-l (P < 0.01) at the highest dose. In the denervated rabbits, the changes in mean arterial pressure and heart rate produced by sodium nitroprusside were not significantly different from the changes in the intact rabbits. At the highest dose, mean arterial pressure decreased from 72 t 3 to 59 t 3 mmHg (P < 0.01) and heart rate increased from 221 t 8 to 306 t 9 beats/min (P < 0.01). Plasma vasopressin concentration increased from 2.8 f 0.3 to 5.7 t 1.3 pg/ml at the highest dose (P < O.Ol), and the vasopressin response to each dose of sodium nitroprusside was not significantly different from the response in the intact rabbits. In contrast, the increases in plasma renin activity were significantly attenuated in the denervated animals (4.0 t 0.7 vs. 8.7 t 1.7 ngoml-’ 2 h-l at 3 pg. kg-’ *mine’, P < 0.01; and 10.8 t 2.2 VS. 19.9 t 3.6 ngoml-102 h-l at 10 pg=kg-‘* min-l, P C 0.05). nitroprusside on. .plasma osmo. .The . . effects q . of sodium . . . lality and plasma electrolyte concentrations are summarized in Table 1. Plasma osmolality and plasma sodium concentration increased slightly during sodium nitroprusside infusion, but only the increase in plasma osmolality at the highest dose in the intact animals was statistically significant. The increase in plasma osmolality may have contributed to the increase in plasma vasopressin concentration. Plasma potassium concentration tended to decrease during sodium nitroprusside infusion, but this was significant only in the denervated rabbits at the highest dose of sodium nitroprusside.

SECRETION

cl

Intact Denervatsd

200 160 120 Plasma Vasopressin 80iiiF;r1 40 t 060 1

l

1. Effect of sodium nitroprusside infusion on plasma osmolality and plasma sodium and potassium conc&trations TABLE

Nitroprusside pg. kg-l

Control

1.0

meq/l

PK, meq/l

260 -

Heart Rate

(bpm)

260 240 -

80

Intact Denervated Intact Denervated Intact Denervated

Values are means sodium concentration; no differences between value.

293t3 294t3 143&l 143&l

297&l 295t2 145k2 144&l

l

Infusion, min-l

3.0

29823 297k2 147t4 145tl

10.0

300&2* 300t3 147k2 146tl

4.5t0.2

4.5k0.3

4.4t0.2

4.2k0.3

4.6kO.l

4.4tO.l

4.3kO.l

4.1&0.1*

t SE. Posmol, plasma osmolality; Pk, plasma potassium concentration. the 2 groups. * P < 0.05 compared

PNa, plasma There were with control

(mmHg)

60 50 40 I

* * L.!L rh

70

Pressure

The effects of hemorrhage in intact and denervated rabbits are compared in Fig. 2. After removal of 10 ml/ kg of blood (10 min) in the intact rabbits, mean arterial pressure did not change (79 t 3 to 76 t 3 mmHg), but heart rate increased from 220 t 7 to 266 t 9 beats/min (P c 0.01). After removal of 20 ml/kg of blood (20 min), mean arterial pressure decreased to 49 t 3 mmHg (P < 0.01) but there was no further increase in heart rate (266 t 9 beats/min). After reinfusion of the blood, mean arterial pressure returned to the control value and heart rate decreased to 235 t 6 beats/min. Plasma vasopressin concentration did not change significantly at 10 min (2.7

PNa,

300 -

Mean Arterial

Hemorrhage

P osmol9 mosmol/kg

320 -

~ 0

10

Blood

20

Loss

Recovery

(ml/kg)

FIG. 2. Effect of hemorrhage on mean arterial pressure, heart rate, plasma renin activity, and plasma vasopressin concentration in intact (n = 15) and denervated (n = 14) rabbits. * P < 0.05 compared with control. There were no significant differences between intact and denervated rabbits.

-+ 0.3 to 4.0 t 1.0 pg/ml) but increased markedly to 159.0 t 37.1 pg/ml at 20 min (P < 0.01). Plasma renin activity increased progressively during hemorrhage from 5.9 t 0.7 to 46.8 t 9.2 ng*ml-’ l 2 h-l at 20 min (P < 0.01). The cardiovascular effects of hemorrhage in the denervated rabbits were similar to those in the intact animals. Although there was a small decrease in mean arterial pressure from 74 t 2 to 67 t 3 mmHg at 10 min (P < 0.05) and the pressure at 20 min was lower than in the intact rabbits (45 t 3 vs. 49 t 3 mmHg), the overall change in mean arterial pressure was not different between the two groups. Heart rate increased from 221 t 6 to 281 t 6 beats/min at 10 min (P < 0.01) and to 288 t 14 beats/min at’20 min (P < O.Oi). After reinfusion of the blood, mean arterial pressure increased to 74 t 2 mmHg and heart rate decreased to 244 t 11 beats/min. Plasma vasopressin concentration did not change after removal of 10 ml/kg (5.0 t 1.7 to 3.9 t 1.2 pg/ml) but increased markedly to 115.4 t 45.6 pg/ml after removal

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

RENAL

NERVES

AND

VASOPRESSIN

of 20 ml/kg (P < 0.01). The increase in plasma vasopressin concentration in the denervated animals tended to be less than the increase in the intact rabbits, but there was no statistically significant difference. The increase in plasma renin activity also tended to be attenuated at each level of the hemorrhage in the denervated rabbits, (10.9 t 3.2 vs. 14.3 t 1.17 ngoml-‘*2 h-l at 10 min and 39.3 t 8.0 vs. 46.8 t 9.2 ng*ml-’ 2 h-l at 20 min), but again, there was no significant difference. The relationship between plasma vasopressin concentration and mean arterial pressure is shown in Fig. 3. There was little or no change in plasma vasopressin concentration until blood pressure decreased below -65 mmHg. Further decreases in blood pressure resulted in marked increases in plasma vasopressin concentration. The relationship between plasma vasopressin concentration (PAvP) and mean arterial pressure (MAP) could be expressed by the following regression equations

F825

SECRETION

electrolyte concentrations did not change significantly except for a small increase in plasma osmolality in the intact rabbits at 20 min. The increase in plasma osmolality may have made a small contribution to the increase in plasma vasopressin concentration. Hematocrit decreased progressively during the hemorrhage in both groups.

l

Hypertonic

The effects of hypertonic saline infusion on mean arterial pressure, plasma osmolality, and plasma vasopressin concentration are summarized in Fig. 4. In the intact rabbits, mean arterial pressure did not change significantly during the hypertonic saline infusion but heart rate increased from 231 t 11 to 266 t 14 beats/ min at 40 min (20 meq) (P C 0.01). Plasma osmolality increased progressively from 296 t 1 to 312 t 1 mosmol/ kg at 20 min (10 meq) (P c 0.01) and to 321 t 3 mosmol/ kg at 40 min (P c 0.01). Plasma vasopressin concentration did not change significantly at 20 min (4.0 t 0.9 to 7.3 + - 1.3 pg/ml) but increased to 10.9 t 2.8 pg/ml at 40 min (P < 0.05). Plasma renin activity decreased from 5.8 + - 1.4 to 2.8 t 0.5 ng*ml-’ l 2 h-l at 20 min (P < 0.05) and to 2.7 t 0.7 ng*ml-’ 02 h-l at 40 min (P < 0.05). In the denervated rabbits, resting mean arterial pressure was similar to the value in the intact rabbits and did not change during the hypertonic saline infusion. Heart rate again increased from 239 t 11 to 262 t 6

intact: log

PAVP

-0.03

=

x

MAP + 3.0 (r = -0.68,

P < 0.01)

X

MAP + 2.8 (r = -0.68, P < 0.01)

denervated: 1%

PAVP

=

-0.03

Saline Infusion

The slopes of the regression lines were not different. The effects of hemorrhage on plasma osmolality, plasma electrolyte concentrations, and hematocrit are summarized in Table 2. Plasma osmolality and plasma

+ +

+

t 0

+ 0 +O +u" t

t

0

t

+ 0

Ol!!l 00

0

0

+ 0

om?t”o+QoM~o

a+@+*

t

t

110.0

Mean

Arterial

Pressure

(mmffg)

FIG. 3. Relationship between plasma vasopressin concentration and mean arterial pressure during hemorrhage in intact (+, n = 15) and denervated (0, n = 14) rabbits. Least-squares regression analysis was performed after logarithmic transformation of vasopressin data. A significant correlation was observed in both intact (r = -0.68, P < 0.01) and denervated rabbits (r = -0.68, P < O.Ol), but slopes were not different between the 2 groups (intact, -0.03; denervated,

-0.03). Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

F826

RENAL

NERVES

AND

VASOPRESSIN

TABLE 2. Effect of hemorrhage on plasma OS‘molality, plasma sodium and potassium concentration,

P

Intact Denervated Intact Denervated Intact Denervated Intact Denervated

osmol,

mosmol/kg PNa, meq/l PK,

Hct,

meq/l

%

295t2 292&l 142&l 143tl 4.2k0.2 4.3t0.2 34kl.l 3421.2

Values are means t SE. Abbreviations crit. There were no differences between compared with control value.

297k3 293t2 144tl 144kl 4.2t0.2 4.2tO.l 31t1.3* 32&1.2*

300t3* 297t3 145&l 146tl 4.4k0.2 4.6t0.2 30*0.9* 28tl.l*

as in Table 1. Hct, hematothe 2 groups. * P c 0.05

cl

Intact Denervatsd

it

SECRETION

0.6 ngoml-102 h-l, but in contrast to the change in the intact rabbits, this decrease was not statistically significant. The relationship between plasma vasopressin concentration and plasma osmolality is shown in Fig. 5. The regression equations relating plasma vasopressin concentration (PA& to plasma osmolality (PoSmOJwere as follows intact: P AVP

=

0.22

X

Posmol- 61.1 (r = 0.49, P < 0.05);

0.26

X

Posmol- 73.2 (r = 0.68, P < 0.01);

denervated: PAVP

=

the slopes of the regression lines were not significantly different. The changes in plasma electrolyte concentrations during hypertonic saline infusion are summarized in Table 3. There were progressive increases in plasma sodium concentration and decreases in plasma potassium concentration in both the intact and the denervated rabbits. Water Deprivation

Plasma

320

Plasma Osmolality

310

t

m

300

(mOsm/kg) 290

80

t

The effects of water deprivation on mean arterial pressure, plasma osmolality, and plasma vasopressin concentration are shown in Fig. 6. In the intact rabbits, mean arterial pressure decreased from 82 t 3 to 72 t 3 mmHg at 24 h (P < 0.05) and then returned to the control level at the end of the 24-h recovery period. Heart rate did not change significantly (210 t 13 to 205 t 12 beats/min). Plasma osmolality increased progressively from 291 t 2 to 300 t 4 mosmol/kg at 12 h (P c 0.05) and to 304 t 3 mosmol/kg at 24 h (P < 0.01). The increase in plasma osmolality was accompanied by an increase in plasma vasopressin concentration from 4.0 t 0.7 to 6.8 t 0.8 pg/ml at 12 h (P < 0.05) and to 6.9 t 0.6 pg/ml at 24 h (P < 0.05). Plasma renin activity tended to increase during water deprivation, but the increase from 5.9 t 1.3 to’ 7.0 t 2.0 ng*ml-’ 2 h-l was not statistically significant. The effects of water deprivation in the denervated rabbits were similar to those in the intact animals. Mean arterial pressure decreased from 73 t 2 to 62 t 3 at 24 h (P < 0.05) and returned to the control value at the end of the recovery period. Heart rate did not change significantly (210 t 13 to 205 t 12 beats/min). Plasma osmolality increased from- 295 t 2 to 304 t 4 mosmol/kg at 12 h (P C 0.05) and to 306 t 3 mosmol/kg at 24 h (P < 0.05). The control value for plasma vasopressin concentration in the denervated rabbits (2.2 t 0.6 pg/ml) tended to be less than the value in the intact animals (4.0 t 0.7 pg/ml), but this difference was not statistically significant. During water deprivation, plasma vasopressin concentration increased to 5.4 t 0.9 pg/ml at 12 h (P < 0.01) and to 5.9 t 0.6 pg/ml at 24 h (P < 0.01). Plasma renin activity tended to increase, but the change from 5.8 t 2.0 to 7.9 t 3.0 ng*ml-’ l 2 h-l was not statistically significant. In both the intact and denervated rabbits, the decrease in arterial pressure during water deprivation may have contributed to the increase l

Mean

70 -

Arterial

60 w

Pressure (mmHg)

50 -

Dose

of

NaClhEq)

FIG. 4. Effect of hypertonic saline infusion on mean arterial pressure, plasma osmolality, and plasma vasopressin concentration in intact (n = 7) and denervated (n = 6) rabbits. * P < 0.05 compared with control. There were no significant differences between the 2 groups.

beats/min at 40 min (P < O.Ol), and plasma osmolality increased from 292 t 2 to 319 t 2 mosmol/kg (P < 0.01). The control value for plasma vasopressin concentration tended to be less than the value in the intact rabbits, but this difference was not statistically significant. Plasma vasopressin concentration increased progressively from 2.7 -+ 0.2 to 9.6 t 2.6 pg/ml at 40 min (P c 0.05), and the increase was the same as that in the intact animals. Plasma renin activity decreased from 5.2 t 2.0 to 3.9 t

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

RENAL

NERVES

AND

VASOPRESSIN

F827

SECRETION

25.0

t 0

t +-

0

t

0 if

+

9 0

4-t

t 0

t

0

280.0

0

t

0 +

0

t 0

t 0 t

290.0

300.0

Plasma

310.0

Osmolality

320.0

330.0

(mOsm/kg)

FIG. 5. Relationship between plasma vasopressin concentration and plasma osmolality in intact (+, n = 7) and denervated (0, n = 6) rabbits. There was a significant correlation between plasma vasopressin concentration and mean arterial pressure in intact (r = 0.49, P < 0.05) and denervated (r = 0.68, P < 0.01) rabbits, but slopes were not different between the 2 groups (intact, 0.22; denervated, 0.26).

neys from nine intact rabbits averaged 103.8t 14.6 ng/ g wet wt.

3. Effect of hypertonic saline infusion on plasma sodium and potassium concentrations

TABLE

Control

Hypertonic Saline, meq 10

20

Intact 142t2 151t2* 154t1* Denervated 14322 152t2* 156t2* Intact 4.6tO.l 4.2tO.l* 4.0t0.1* PK, meq/l Denervated 4.6kO.l 4.3t0.2* 4.0t0.1* Values are means t SE. Abbreviations as in Table 1. There were no differences between the 2 groups. * P C 0.05 compared with control value. PNa, m&l

in plasma vasopressin concentration. The effects of water deprivation on plasma electrolyte concentrations, hematocrit, and body weight are shown in Table 4. Plasma sodium and potassium concentration increased in both groups of animals, but these changes were not statistically significant. Hematocrit did not change significantly. Body weight decreased during water deprivation in both the intact and the denervated animals, and the changes were not different between the two groups. At the end of the recovery period, body weight returned to the control value in both groups of animals. Renal Cortical Norepinephrine

Concentration

Renal cortical norepinephrine concentration in denervated kidneys, measured within 2 wk of renal denervation, was C10.0 rig/g wet wt in all kidneys. By comparison, the norepinephrine concentration in innervated kid-

DISCUSSION

These experiments indicate that chronic bilateral renal denervation does not attenuate the response of vasopressin secretion to osmotic and reflex stimuli in conscious rabbits. In addition, renal denervation does not impair blood pressure regulation during sodium nitroprusside infusion, hemorrhage, or water deprivation, nor does it impair osmoregulation during hypertonic saline infusion or water deprivation. Therefore it appears that the renal nerves play no major role in the osmotic and reflex regulation of vasopressin secretion or in blood pressure regulation, at least under the conditions of the present study. These findings will be discussed further in the context of other studies. Reflex Stimuli

and the Renal Nerves

In the present study, infusion of nitroprusside decreased blood pressure and increased plasma vasopressin concentration in both the intact and denervated animals. The change in mean arterial pressure and heart rate during nitroprusside infusion was similar in both groups, and the increase of plasma vasopressin concentration was not significantly different between the two groups. In the hemorrhage study, although the overall cardiovascular changes were similar, the change in mean arterial pressure showed some difference between the two groups. In the intact rabbits, blood pressure did not decrease

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

F828

RENAL

NERVES

cl

AND

VASOPRESSIN

Intact Denmmted

* Plasma Vasopressin

4

Cont. bg/mlI

2

*

*

*

~

0 310

i 300 -

Plasma Osmolality

(mOsm/kg)

-

290

280 ’ 90 80 I 70 -

Mean Arterial Pressure

60.

(mml-lg)

50 40 L-

0

12

Recovery

24

Time (h) 6. Effect of water deprivation on mean arterial pressure, plasma osmolality, and plasma vasopressin concentration in intact (n = 6) and denervated rabbits (n = 6). * P < 0.05 compared with control. There were no significant differences between the 2 groups. FIG.

4. Effect of water deprivation on plasma sodium and potassium concentrations, hematocrit, and body weight

TABLE

Control

Water 12

PN~, meq/l PK, meq/l

Hct,

%

BW,

kg

Intact Denervated Intact Denervated Intact Denervated Intact Denervated

144tl 143tl 4.2k0.2 4.3t0.2 38kO.4 35t0.9 2.71t0.15 2.63t0.15

Values are means t SE. Abbreviations body weight. There were no differences 0.05 compared with control value.

144&l 144tl 4.2t0.2 4.3t0.2 35t1.3 34tl.l 2.66t0.14 2.63k0.14

Deprivation,

h 24

14622 147t2 4.4t0.2 4.6t0.2 35t1.1 3421.1 2.5&0.13* 2.54*0.13*

as in Tables 1 and 2. BW, between the 2 groups. * P c

significantly after removal of 10 ml/kg blood but decreased markedly after removal of 20 ml/kg. These changes are in good agreement with results obtained in rabbits by others (1.8). In the denerva ted rabbits, on the other hand, mean arteri .a1 pressure decreased significantly even after removal of 10 ml/kg blood, and although the difference was not statistically significant, mean arterial pressure tended to be lower in the denervated rabbits after removal of 20 ml/kg. Despite this

SECRETION

lower blood pressure, plasma vasopressin concentration tended to be lower after hemorrhage in the denervated animals, suggesting that renal denervation attenuated the vasopressin response to hemorrhage. However, the slopes of the regression for log plasma vasopressin concentration vs. mean arterial pressure were not different between the intact and the denervated animals. Therefore these results suggest that the renal nerves do not participate in the regulation of vasopressin secretion during hypotension induced by nitroprusside infusion or hemorrhage. However, two points should be considered before dismissing a possible role of the renal nerves in the regulation of vasopressin secretion. First, the decrease in blood pressure induced in the present study may not have reached the threshold required to activate intrarenal receptors and increase afferent renal nerve activity. For example, Recordati et al. (27) reported that the renal chemoreceptors responsive to renal ischemia (RI-chemoreceptors) in anesthetized rats are activated only during markedly impaired renal blood flow corresponding to a systemic blood pressure of ~40 mmHg. On the other hand, Barber et al. (2) reported that the other class of renal chemoreceptors (Rn-chemoreceptors) can be activated by a reduction in renal perfusion pressure to 80 mmHg. The largest dose of nitroprusside employed in the present study reduced mean arterial pressure to -60 mmHg in both the intact and denervated animals, and this may have been sufficient to activate the R2-chemoreceptors. Hemorrhage reduced blood pressure to ~40 mmHg in both intact ( 6 of 15) and denervated (5 of 14) rabbits. Therefore both R1- and Rz-chemoreceptors should have been activated in these rabbits. However, even at these low pressures, the increase in plasma vasopressin concentration was not attenuated in the denervated rabbits (Fig. 3). The second point that should be considered is that the major reflex control of vasopressin secretion is mediated by the arterial and atria1 baroreceptors (3, 23, 30, 31, 35), and it is possible that any contribution of the renal nerves was masked by this control mechanism. Thus studies in baroreceptor-denervated animals are required to further investigate whether or not the renal afferent nerves contribute to the reflex regulation of vasopressin secretion. The activation of the renin-angiotensin system also seems to play no major role in the vasopressin response to hypotension induced by nitroprusside infusion, because the increase in plasma renin activity was significantly attenuated in the denervated rabbits but the increase in plasma vasopressin concentration was the same in the two groups. In the hemorrhage study, however, the increase in plasma renin activity was the same in the two groups. Renal sympathetic nerve activity has been reported to increase during nitroprusside-induced hypotension but to decrease during hypotensive hemorrhage (8, 18). These divergent changes in renal nerve activity may explain the differing contribution of the renal nerves to the responses in plasma renin activity to nitroprusside infusion and hemorrhage. Osmotic Stimuli

and the Renal Nerves

The possible role of renal nerves in the osmotic regulation of vasopressin secretion was also investigated in

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

RENAL

NERVES

AND

VASOPRESSIN

the present study. Quillen et al. (24, 25) reported that the sensitivity of the osmotic control of vasopressin is severely reduced in conscious nephrectomized dogs. They speculated that this reduction may result from the loss of the afferent renal nerves based on the study of Recordati et al. (26), who demonstrated the presence of chemoreceptors in the renal pelvis. One class of receptor, which they termed R2 chemoreceptors, elicits an excitation in the afferent renal nerves when the receptors are exposed to high concentrations of sodium, potassium, or mannitol (19, 26). Furthermore, in anesthetized antidiuretic rats, the afferent renal nerves exhibit a high resting discharge that declines progressively during volume expansion (26). From these findings, Recordati et al. postulated that Ra-chemoreceptors act as peripheral osmoreceptors, sensitive to extracellular dehydration. Subsequent studies have provided evidence that these sensory receptors are important in the regulation of afferent renal nerve activity and in the initiation of renorenal and renosystemic reflexes (14-16, 19). If the receptors also play a physiological role in the control of vasopressin secretion, the vasopressin responses to osmotic stimuli should be attenuated by renal denervation. However, this was not the case, suggesting that R2-chemoreceptors do not play an important role in the vasopressin response to osmotic stimuli, at least under the conditions of the present study. Nevertheless, the possibility that these stimuli were not sufficient to activate R2-chemoreceptors or that the central osmoreceptor control mechanism masked a role of the renal nerves in the present study cannot be excluded. Experiments with selective intrarenal infusions of hypertonic solutions may help to clarify this issue. Blood Pressure Regulation

and the Renal Nerves

Activation of both the afferent and efferent renal nerves occurs during hypotension and hypovolemia, and this might help to defend blood pressure in these states. The observation by Tanigawa et al. (33) that acute renal denervation increased the decrease in blood pressure during hemorrhage in anesthetized dogs is consistent with this possibility. On the other hand, Pate1 and Kline (21) demonstrated that the hypotensive response to nitroprusside infusion in conscious rats was not altered by renal denervation. In the present study, blood pressure decreased significantly after removal of 10 ml/kg of blood in denervated rabbits but not in intact rabbits. However, the overall change in blood pressure during hemorrhage was not different between the intact and denervated rabbits. Furthermore, renal denervation did not alter the decreases in blood pressure produced by nitroprusside infusion or water deprivation. Therefore the results of the present study, together with the results of Pate1 et al. (21), indicate that the renal nerves do not play an important role in the regulation of blood pressure during hypotension or hypovolemia. However, again, the possibility that a role of the renal nerves in blood pressure regulation was masked by arterial and atria1 baroreceptor reflexes should be considered. In conclusion. the present results demonstrate that

F829

SECRETION

renal denervation does not attenuate the increase in plasma vasopressin concentration in response to two reflex and two osmotic stimuli nor impair blood pressure regulation during nitroprusside infusion, hemorrhage, or water deprivation. It therefore appears that the renal nerves do not contribute significantly to the reflex or osmotic regulation of vasopressin secretion or to blood pressure regulation during hypotension or hypovolemia, at least under the conditions of the present investigation. The expert Angie DeCarlo This work Institute Grant Address for sity of California, Received

technical assistance of Lance Chou, Dina San Juan, and is gratefully acknowledged. was supported by National Heart, Lung, and Blood HL-29714. correspondence: I. A. Reid, Dept. of Physiology, UniverSan Francisco, CA 94143-0444.

30 May

1989;

accepted

in final

form

11 October

1989.

REFERENCES 1. ASTROM, A., AND J. CRAFOORD. Afferent and efferent activity in the renal nerves of cats. Acta Physiol. Stand. 74: 69-78, 1968. 2. BARBER, J. D., A. SCOLTOCK, AND N. G. Moss. R2 chemoreceptor activation during renal hypoperfusion is prostaglandin (PG) dependent (Abstract). Kidney Int. 29: 329, 1986. 3. BROOKS, V. L., L. C. KEIL, AND I. A. REID. Role of the reninangiotensin system in the control of vasopressin secretion in conscious dogs. Circ. Res. 58: 829-838, 1986. 4. CAVERSON, M. M., AND J. CIRIELLO. Effect of stimulation of afferent renal nerves on plasma levels of vasopressin. Am. J. Physiol. 252 (Regulatory Integrative Comp. Physiol. 21): R801R807,1987. 5. DAY, T. A., AND J. CIRIELLO. Afferent renal nerve stimulation excites supraoptic vasopressin neurons. Am. J. Physiol. 249 (Regulatory Integrative Comp. Physiol. 18): R368-R371, 1985. 6. DAY, T. A., AND J. CIRIELLO. Effects of renal receptor activation on neurosecretory vasopressin cells. Am. J. Physiol. 253 (Regulatory Integrative Comp. Physiol. 22): R234-R241, 1987. G. F. The functions of the renal nerves. Rev. Physiol. 7. DIBONA, Biochem. PharmacoZ. 94: 75-181,1982. P. K., W. RIEDEL, S. L. BURKE, J. GIPPS, AND P. I. 8. DORWARD, KORNER. The renal sympathetic baroreflex in the rabbit: arterial and cardiac baroreceptor influences, resetting, and effect of anesthesia. Circ. Res. 57: 618-633, 1985. 9. FABER, J. E., AND M. J. BRODY. Afferent renal nerve-dependent hypertension following acute renal artery stenosis in the conscious rat. Circ. Res. .57: 676-688, 1985. H., A. J. BAERTSCHI, AND P. G. GUYE10. HASHEMZADEH-GARGARI, NET. Baroreceptor-independent medullary mechanism for release of vasopressin during hypotension in rats. J. Endocrinol. 118: lOl111,1988. 11. KATHOLI, R. E. Renal nerves in the pathogenesis of hypertension in experimental animals and humans. Am. J. Physiol. 245 (Renal Fluid Electrolyte Physiol. 14): Fl-F14, 1983. 12. KEIL, L. C., AND W. B. SEVERS. Reduction in plasma vasopressin levels of dehydrated rats following acute stress. Endocrinology 100: 30-38,1977. 13. KLINE, R. L., K. P. PATEL, J. CIRIELLO, AND P. F. MERCER. Effect of renal denervation on arterial pressure in rats with aortic nerve transection. Hypertension Dallas 5: 468-475, 1983. 14. KOPP, U. C. Renorenal reflexes in normotension and hypertension. Miner. Electrolyte Metab. 15: 66-73, 1989. 15. KOPP, U. C., L. A. OLSON, AND G. F. DIBONA. Renorenal reflex responses to mechanoand chemoreceptor stimulation in the dog and rat. Am. J. Physiol. 246 (Renal Fluid Electrolyte Physiol. 15): F67-F77,1984. 16. KOPP, U. C., L. A. SMITH, AND G. F. DIBONA. Renorenal reflexes: neural components of ipsilateral and contralateral renal responses. Am. J. Physiol. 249 (Renal Fluid Electrolyte Physiol. 18): F507F517,1985. 17. MENARD, J., AND K. J. CATT. Measurement of renin activity, concentration and substrate in rat plasma by radioimmunoassay

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

F830

RENAL

NERVES

AND VASOPRESSIN

of angiotensin I. Endocrinology 90: 422-430, 1972. H., Y. NISHIDA, H. MOTOCHIGAWA, N. UEMURA, H. AND S. F. VATNER. Opiate receptor-mediated decrease in renal nerve activity during hypotensive hemorrhage in conscious rabbits. Circ. Res. 63: 165-172, 1988. 19. MOSS, N. G. Electrophysiological characteristics of renal sensory receptors and afferent renal nerves. Miner. Electrolyte Metub. 15:

18. MORITA, HOSOMI,

59-65,1989. 20. OPARIL,

S., W. SRIPAIROJTHIKOON, AND J. M. WYSS. The renal afferent nerves in the pathogenesis of hypertension. Can. J. Phys-

iol. Pharmacol. 21. PATEL, K.

65: 1548-1558,

Physiol. 247 (Regulatory R620,1984. 22. PEULER, J. D., AND G.

Integrative

Comp.

Physiol.

16):

25.

JR., AND A. W. COWLEY, JR. Influence of volume changes on osmolality-vasopressin relationships in conscious dogs. 244 (Heart

Circ.

Physiol.

13): H73-H79,

1983.

W., JR., M. M. SKELTON, J. RUBIN, AND A. W. Osmotic control of vasopressin with chronically altered volume states in anephric dogs. Am. J. Physiol. 247 (Endo-

crinol. Metab. 10): E355-E361, 1984. 26. RECORDATI, G. M., N. G. Moss, S. GENOVESI, AND P. R. RoGENES. Renal receptors in the rat sensitive to chemical alterations

gist 21: 97, 1978. SCHRIER, R. W.,

nonosmotic 31.

H1120-H1126,1987. QUILLEN, E. W.,

Am. J. Physiol. QUILLEN, E. COWLEY, JR.

30.

R615-

A. JOHNSON. Simultaneous single isotope radioenzymatic assay of plasma norepinephrine, epinephrine, and dopamine. Life Sci. 21: 625-636, 1977. 23. QUAIL, A. W., R. L. WOODS, AND P. I. KORNER. Cardiac and arterial baroreceptor influences in release of vasopressin and renin during hemorrhage. Am. J. Physiol. 252 (Heart Circ. Physiol. 21): 24.

of their environment. Circ. Res. 46: 395-405, 1980. G. M., N. G. Moss, AND L. WASELKOV. Renal chemoreceptors in the rat. Circ. Res. 43: 534-543, 1978. 28. REID, I. A., R. GOLIN, L. C. GREGORY, P. L. NOLAN, E. W. QUILLEN, JR., AND L. C. KEIL. Vasopressin, the renal nerves, and renin secretion. In: Vasopressin: Cellular and Integrative Functions, edited by A. W. Cowley, Jr., J. F. Liard, and D. A. Ausiello. New York: Raven, 1988, p. 447-454. 29. REID, I. A., P. L. NOLAN, AND L. C. KEIL. Renal nerve stimulation increases plasma vasopressin concentration (Abstract). Physiolo27. RECORDATI,

1987.

P., AND R. L. KLINE. Influence of renal nerves on noradrenergic responses to changes in arterial pressure. Am. J.

SECRETION

32.

T. BERL, AND R. J. ANDERSON. Osmotic and control of vasopressin release. Am. J. Physiol. 236

(Renal Fluid Electrolyte Physiol. 5): F321-F332, 1979. SHARE, L. Role of vasopressin in cardiovascular regulation. Physiol. Rev. 68: 1248-1284,1988. SIMON, J. K., N. W. KASTING, AND J. CIRIELLO. Afferent renal

nerve effects on plasma vasopressin and oxytocin in conscious rats. Am. J. Physiol. R1240-R1244,1989. 33. TANIGAWA, H.,

256

(Regulatory

Integrative

Comp.

Physiol.

25):

S. L. DUA, AND T. A. ASSAYKEEN. Effect of renal and adrenal denervation on the renin response to slow haemorrhage in dogs. Clin. Exp. Pharmacol. Physiol. 1: 325-332, 1974. 34. UCHIDA, Y., K. KAMISAKA, AND H. UEDA. TWO types of renal mechanoreceptors. Jpn. Heart J. 12: 233-241,197l. 35. WANG, B. C., W. D. SUNDET, M. 0. K. HAKUMWKI, AND K. L. GOETZ. Vasopressin and renin responses to hemorrhage in conscious, cardiac-denervated dogs. Am. J. Physiol. 245 (Heart Circ. Physiol. 36. WINER,

14): H399-H405,

1983.

B. J. Statistical Principles York: McGraw-Hill, 1971.

in Experimental

Downloaded from www.physiology.org/journal/ajprenal at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.

Design.

New