Do renal nerves chronically influence and arterial pressure in spinal rats? KENDRICK A. TROSTEL AND JOHN W. OSBORN Department of Veterinary Biology and Graduate Program University of Minnesota, St. Paul, Minnesota 55108 Trostel, Kendrick A., and John W. Osborn. Do renal nerves chronically influence renal function and arterial pressure in spinal rats ? Am. J. Physiol. 263 (Regulatory Integrative Comp. Physiol. 32): R1265R1270, 1992.-Previous studies have demonstrated that renal nerve activity has acute effects on renal function in rats with cervical spinal cord transection (CST). The present study test,ed the hypothesis that renal nerves chronically influence renal and cardiovascular function in CST rats. Three groups of conscious Sprague-Dawley rats were studied: renal denervated plus CST (RDNX + CST), sham RDNX plus CST (sham + CST), and sham RDNX plus sham CST (intact). CST or sham CST surgeries were performed 8 days after RDNX or sham RDNX. Sodium and water intakes were fixed by intravenous infusion. Mean arterial pressure (MAP) and plasma renin activity (PRA) were measured before and for 9 days after CST/sham CST. In addition, urine flow, urinary sodium excretion, and urine pH were measured in the two groups of CST rats. One day after CST, MAP decreased ~25 mmHg in both RDNX + CST and sham + CST groups. PRA had fallen -50% 1 day after CST and was not different between CST groups. PRA remained depressed throughout the study. There were no differences between sham + CST and RDNX + CST rats in any of the renal or cardiovascular variables measured after CST. In summary, we found no evidence for a chronic effect of renal nerves on renal function or arterial pressure in CST rats. sympathetic excretion
nerve
activity;
plasma
renin
activity;
sodium
SYMPATHETIC NERVE ACTIVITY (SNA) is generally believed to originate in the brain stem and higher centers. However, recent studies in anesthetized rats with cervical spinal cord transection (CST) have provided evidence for spinally generated SNA. Specifically, direct nerve recordings show increased renal SNA after CST in anesthetized rats (9,13). Furthermore, the CST-induced increase in renal nerve activity acutely affected water and electrolyte handling in anesthetized rats, suggesting a functional role for spinally generated SNA (9). Is there evidence for SNA in unanesthetized spinal animals? If so, is it functionally significant? We have recently reported that spinally generated SNA exists in conscious spinal rats. This is based on the observation that acute adrenergic blockade results in three- to fourfold increases in sodium and potassium excretion in conscious rats 24 h after CST (14). Taken together, studies in both anesthetized and conscious CST rats suggest that renal sympathetic nerve discharge is present and decreases water and electrolyte excretion. Furthermore, these observations suggest that renal SNA, via regulation of fluid and electrolyte excretion, may be of functional significance in the long-term regulation of arterial pressure in spinal animals. Indeed, two previous studies in conscious spinal rats suggest that renal SNA may play a role in regulation of renal function and therefore arterial pressure. First, despite the fact that CST results in 0363-6119/92
$2.00 Copyright
renal function
in Neuroscience,
a loss of adrenergic vasomotor tone, arterial pressure is not different from control levels within 8-10 days after CST (10). Although this time course is compatible with a volume-dependent mechanism, the role of renal SNA in control of fluid balance was not investigated in that study. Further support for such a mechanism was provided by a recent study from our laboratory demonstrating that adrenergic antagonists markedly increased electrolyte excretion in conscious rats 24 h after CST (14). However, these responses were monitored for only 2 h after adrenergic blockade, and we were therefore unable to conclude whether spinally generated renal SNA is a determinant in the long-term regulation of renal function and arterial pressure. The present study was undertaken to directly test the hypothesis that renal SNA chronically influences electrolyte handling and therefore arterial pressure in conscious spinal rats. We measured renal function and arterial pressure in renal-denervated CST rats, before and for up to 9 days after spinal transection. We predicted that, if renal SNA exists in spinal rats, renal denervation would result in an enhanced excretory rate of water and electrolytes compared with CST rats with intact renal nerves. Moreover, this would be manifested as an impairment in the long-term regulation of arterial pressure in renal-denervated CST rats. METHODS General Procedures Male Sprague-Dawley rats (Sasco, Omaha, NE) were housed until the time of study in our animal housing facility with controlled temperature and lighting. Standard rat food and water were provided ad libitum. After a minimum of 2 days in this facility, rats were brought to the laboratory for renal denervation (RDNX) or sham RDNX (see below). Subsequently, they were housed in a quiet, isolated laboratory with a 12:12-h daynight cycle. All procedures were approved by the institutional animal care committee and were conducted in accordance with institutional and National Institutes of Health guidelines. Surgical Procedures The overall schedule of surgical procedures in relation to the experimental protocol is summarized in Fig. 1. Renal denervation. Rats were fasted overnight before surgery. The next morning, rats were atropinized (0.13 mg/kg ip) and anesthetized with pentobarbital sodium (50 mg/kg ip). Rats were divided into two groups for surgery: bilateral RDNX and sham RDNX. With the use of a dissecting microscope, bilateral RDNX was performed via a ventral midline incision. The area between the hilus and the origin of renal vessels at the aorta and vena cava was stripped of fat and connective tissue, then painted with 10% phenol in alcohol to ensure destruction of any remaining nerves. In sham RDNX rats, a ventral midline incision was made, and each kidney was exposed briefly. After RDNX or sham RDNX, the incision was closed and rats
0 1992 The
American
Physiological
Society
R1265
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
R1266 DAY
43
RENAL PROCEDURE
I
-4
RDNX
NUMBER
BEGIN
INFUSION
-1 0 1
3
i I I
I
5 7 9
I
CONTROL MEASUREMENTS CST OR SHAM-CST EXPERIMENT
SHAM-RDNX 13
J
/\
RDNX+CST
SHAM+CST
IN SPINAL
RATS
cages. Rats were alert and mobile the day after CST, and began to eat as early as the first day after CST. As a result of plugged arterial catheters, rats dropped out of the experiment over time. In addition, when a urine collection was unsuccessful, all data from the rat were excluded from the analysis for that day. As a result, sample sizes were not constant throughout the experiment. The number of animals studied in each group is shown in Fig. 1.
OF ANIMALS
RDNX 7
OR SHAM-RDNX
CATHETERIZE,
DENERVATION
INTACT
7 4
7 6
6 6
EXPERIMENT
6
6
5
EXPERIMENT
7
6
5
EXPERIMENT
7
5
5
EXPERIMENT
7
3
4
Fig. 1. Experimental protocol. RDNX, renal denervation; CST, cervical spinal cord transection. Numbers in each group for each day are shown.
received 10 ml of warm sterile isotonic saline intraperitoneally and 100,000 U penicillin G intramuscularly. They were subsequently housed in pairs and fed standard rat food and water ad libitum for 4 days of recovery before femoral catheter implantation. Chronic catheterization. Four days after RDNX or sham RDNX, rats were atropinized (0.13 mg/kg ip) and anesthetized with pentobarbital sodium (50 mg/kg ip) for implantation of femoral arterial and venous catheters. Catheters were tunneled subcutaneously to the top of the head, exteriorized, and passed through a metal spring that was attached to dental acrylic mounted on the rats’ heads. The rats received antibiotic after surgery (100,000 U penicillin G im) and were placed in individual stainless steel metabolic cages. The spring extended to the top of the cage, where it was attached to a hydraulic swivel, which allowed the rats to move freely. An intravenous infusion of lactated Ringer was begun (1 ml/h; 130 meq/l Na, 4 meq/l K, 28 meq/l lactate) and continued for the duration of the study. Rats were not allowed to drink and were fed sodium-free food, thereby fixing water and sodium consumption at 24 ml/day and 3.12 meq/day, respectively, throughout the experiment. This was done to ensure that changes in plasma renin activity and sodium and water excretion were not secondary to alterations in sodium and water intake. Arterial lines were drained and filled daily with 1,000 U/ml heparin, and antibiotic (20 mg ampicillin iv) was administered. Spinal transection. Four days after catheter implantation, rats were atropinized (0.26 mg/kg iv) and anesthetized with pentobarbital sodium (50 mg/kg iv). CST was performed on all RDNX rats (RDNX + CST, n = 7). Sham RDNX rats were divided into two groups. In one group, CST was performed (sham + CST, n = 7), and in the other, a sham CST surgery was performed (intact, n = 6). Preceding CST, a bladder catheter was implanted, and rats were placed in a stereotaxic apparatus for spinal transection as previously described (11). After exposing the seventh cervical and first thoracic vertebrae, the spinal cord was cut using a no. 11 scalpel blade. A blunt hypodermic 18-gauge needle, attached to a vacuum pump, was then used to complete the transection. Completeness of transection was verified by inspection. In the intact group, the bladder was exposed through a midline incision, and the muscle and skin were sutured closed. A skin incision along the neck, similar to that in CST rats, was also made and closed. After CST or sham CST, rats were returned to their home cages. Body temperature of CST rats was maintained within normal limits by circulating warm, humidified air around the
Experimental
Protocol
To examine the influence of renal nerves on cardiovascular function in CST rats, the following variables were measured before and over a g-day period after CST or sham CST: mean arterial pressure (MAP), heart rate (HR), lability of MAP, arterial oxygen tension, arterial carbon dioxide tension, arterial pH, hematocrit (Hct), plasma sodium (PNa), plasma potassium (PK), and plasma renin activity (PRA). Measurements were made before (day -1 and day 0), and 1, 3,5,7, and 9 days after CST. MAP was measured by connecting the arterial catheter to a Spectramed PlOEZ pressure transducer coupled to a Grass polygraph (model 7D). The pulsatile pressure signal was input to a second amplifier with a low-pass filter to acquire an electrical average of arterial pressure. MAP was monitored by computer (Asystant+, Asyst Software Technologies, New York, NY) for 1 h at a sampling rate of 0.5 Hz while the rats rested quietly in their home cages. A time average of MAP was then calculated from the 1,800 data points. The statistical variance was also determined as a quantitative indicator of lability of arterial pressure. HR was determined periodically by increasing the chart speed and counting peaks on the pulsatile pressure tracing. After measurement of MAP and HR, a 500-~1 arterial blood sample was collected into a chilled l-ml syringe containing 1.25 mg EDTA in a volume of 25 ~1. One hundred microliters of this sample was used to measure arterial blood gasses (pH/blood gas analyzer model 1304, Instrumentation Laboratory, Lexington, MA). On days -I, 1, 5, and 9, the remaining blood was transferred to a chilled microcentrifuge vial. This sample was centrifuged, and plasma was collected and frozen at -70°C for later radioimmunoassay (RIA) for PRA (see below). On days 0, 3, and 7, the remaining 400 ~1 of blood was used to measure PNa, PK (NOVA-l sodium-potassium analyzer, NOVA Biomedical, Waltham, MA), and Hct. Sample blood was replaced by an equal volume of whole blood from a conscious, chronically instrumented donor rat. In addition to these measurements, renal function was compared between RDNX + CST and sham + CST rats over the g-day period after CST. To compare the results of this experiment with our previous study in which adrenergic blockade acutely increased electrolyte excretion in spinal rats (14), identical methods of urine collection and analysis were used. Briefly, during the 60-min MAP recording period, urine was continuously collected into a preweighed balloon attached to the bladder catheter. Such a closed collection system was required, since the slow continuous urine flow into the funnels of the warmed metabolic cages results in large evaporative losses of urine. Urine flow through the urethra was prevented by clamping the penis with a miniature clamp (Tiemann, Plainview, NY) before the first collection period. Although the patency of bladder catheters was verified daily, clamping ensured completeness of urine collection. At the time urine collection was begun, basal HR and MAP were not different from before clamping, suggesting that use of the penis clamp did not result in reflex SNA. Because urine could not be collected in intact rats in a manner identical to that used for spinal rats, no renal function measurements were made in this group. Finally, at the end of the urine collection period, urine flow was determined gravimetrically, and urine was analyzed for sodium and potassium concentration using a NOVA- 1 sodium-potassium analyzer. Urinary sodium
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
DENERVATION
excretion and urinary potassium excretion were calculated as the product of urine flow rate and ion concentration. Urine pH was also measured immediately after the urine collection period using a pH-sensitive microelectrode (Radiometer America, Westlake, OH). Determination
of PRA
PRA was measured as previously described (14) by quantitation of endogenously generated angiotensin I (ANG I) using an RIA kit (1251-ANG I RIA kit, New England Nuclear, Boston, MA) based on an adaptation of the method of Haber et al. (5). In addition, plasma renin concentration (PRC) was measured for all rats on day 9 after CST by quantitation of endogenously generated ANG I in the presence of an excess of rat angiotensinogen. Verification
IN
SPINAL
R1267
RATS
120
I”
t 8l\ &---F+
100
E E 80
a, 4 5;
60
O-0 *.
*,+ *,+
+ +,+
*,+ *,+
*
+,+
a
of RDNX
Completeness of RDNX was confirmed by assay for renal catecholamines. After data collection on day 9, rats were anesthetized with pentobarbital sodium, and both kidneys were removed, sectioned, and inspected for signs of hydronephrosis or infection. No signs of either were observed. Sections were immediately frozen in liquid nitrogen. Kidneys were weighed, wrapped in aluminum foil, and stored at -70°C. Norepinephrine was extracted from the tissue by an established technique (2). The samples were assayed for norepinephrine content by high-pressure liquid chromatography (HPLC) with electrochemical detection. HPLC analysis was performed with a Spectra-physics SP 8700 XR liquid chromatograph; a Spectra Physics SP 4290 integrator; a Brownlee Laboratories no. 53358, RP-18, 100 x 4.6 mm, 5pm column; and a Bioanalytical Systems LC-4B electronic controller and LC-17A oxidative flow cell. The mobile phase consisted of 0.1 M KH2POd, 0.1 mM EDTA, 0.4 mM octyl sodium sulfate, and 3.5% by volume HPLC-grade methanol dissolved in Millipore-filtered deionized water. The efficiency of norepinephrine extraction was corrected by the recovery of the internal standard DHBA.
CK I
o-o e-0
250
+,+ +
+I+ *a+
+I+ +,+
+
+,+ +
+,+
‘I+
+,+
l ,+
200
o-o )
*o
+
‘I+
+,+
*,+
Statis tics All values are presented as means t SE. Groups were compared using an analysis of variance (ANOVA) for an unbalanced repeated measures design (SAS Institute, Cary, NC). A finding of significant effect of group was followed by Duncan’s test. Values of cardiovascular and renal function after CST or sham CST were compared with pretransection values using Dunnett’s test. All values were considered significant for P < 0.05. RESULTS
Effect of RDNX
and CST on Arterial
Pressure
RDNX had no statistically significant effect on cardiovascular function in CST rats. MAP, HR, and lability of MAP of RDNX + CST and sham + CST rats were not significantly different from each other (Fig. 2). In contrast, CST had significant effects on MAP, HR, and lability of MAP. In both CST groups, MAP fell -25 mmHg, to a posttransection level of 75 mmHg the first day after CST. MAP recovered from this posttransection low, increasing 10 mmHg to a level of 45 mmHg by the end of the study. By day 9, MAP was no longer significantly lower in the CST groups than in intact rats. MAP in the RDNX + CST group was also not significantly different from its own pretransection level. HR fell substantially with CST (450 beats/min for each CST group), and remained at low levels thereafter. Despite low or normal levels of arterial pressure after transection,
o!
:
-2-l
;
:
:
:
:
;
;
;
f
i
0
1
2
3
4
5
6
7
8
9
DAYS AFTER CST Fig. 2. Effect of CST and RDNX on mean arterial pressure (MAP), heart rate (HR), and variance of MAP. * P < 0.05 compared with before CST. + P < 0.05 compared with intact. # P < 0.05 compared with sham + CST.
lability of MAP increased over time in CST rats. By day 3 after CST, lability of MAP was significantly greater than pretransection levels, and by the end of the study, lability of MAP was approximately double the pretransection level. Because cardiovascular function may be affected by arterial blood gases or pH, these variables were measured in all three groups (Table 1). Although there were differences between intact and CST groups, there was little effect of RDNX on arterial blood gases or pH. Effect of RDNX
and CST on Renal Function
Contrary to the predictions of our hypothesis, RDNX had no effect on renal excretory function after CST (Fig. 3). There were no statistically significant differences
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
R1268
RENAL
Table 1. Effect of RDNX
Intact Sham RDNX
and CST on arterial
+ CST + CST
94.0tl.4 75.1+3.7”“r 84.8+4.5*-f*
Intact Sham RDNX
+ CST + CST
45.8tl.O 44.1t1.3 45.5t0.6
46.5t0.8 39.6+1.6-f39.3&2.4-f
Intact Sham RDNX
+ CST + CST
7.38t0.01 7.4O+O.Ol”f 7.39t0.01
7.40t0.01 7.42+O.Olt 7.43tO.Ol*t spinal
cord transection;
3
PO,, mmHg lOO.Ot3.7 73.3+2.4*-f 75.0+3.1*t Pcoz, mmHg 49.2tl.l 43.1k1.57 43.4+3.0-f Arterial pH 7.35t0.01 7.41+0.02-f 7.38tO.Ol-F RDNX,
between sham + CST and RDNX + CST groups for any of the renal function variables. Both groups of rats appeared to be in sodium and water balance as early as 24 h after CST, since excretory values closely approximated intake, which was fixed at 0.130 meq/h Na and 1 ml/h H20. Furthermore, excretion rates remained unchanged over the course of the study. As with renal excretory function, PRA was not affected by RDNX in CST rats (Fig. 4). However, CST resulted in significant decreases in PRA. After CST, PRA continued to decline, until by day 9 PRA was not significantly different from zero. The extremely low levels of PRA after CST (Fig. 4) may be explained by a shortage of substrate rather than low PRC. To investigate this possibility, PRC was also measured for all rats on day 9 after CST. PRC was 0.2 t 0.1 ng ANG I ml-l h-l for CST rats and 6.5 t 0.6 ng ANG I ml-l h-l for intact rats. Therefore, plasma renin levels were indeed extremely low 9 days after CST. Plasma electrolytes and Hct are shown in Table 2. Alor Hct was different after CST or sham though PNa~ pK, CST, these variables were neither affected by RDNX nor were they different between intact and CST rats. l
l
Verification
RATS
blood gases and pH
I
98.021.8 94.6t2.7 98.7t1.3
Values are means t SE. CST, cervical $ P < 0.05 vs. sham + CST.
IN SPINAL
Days
Control (pre-CST)
Group
DENERVATION
l
l
of RDNX
Renal norepinephrine was not significantly different (ANOVA followed by Duncan’s test, P > 0.05) between intact rats (128.6 t 10.9 rig/g, n = 8 kidneys) and sham + CST rats (84.8 t 17.2 rig/g, n = 8 kidneys). Renal norepinephrine was significantly lower in RDNX rats (3.3 k 1.1 rig/g, n = 12 kidneys). DISCUSSION
This experiment was designed to test the hypothesis that tonic renal sympathetic nerve activity influences electrolyte handling in conscious spinal rats. We proposed that renal nerve activity would result in cumulative retention of sodium and water and thereby chronically influence blood pressure in CST rats. This hypothesis was based on previous experiments in both anesthetized and conscious rats. In anesthetized rats, increased renal sympathetic nerve activity after CST resulted in a significant antidiuresis and antikaliuresis that could be pre-
renal denervation.
After
CST
5
7
9
97.4t2.8 77.0+2.4*t 78.6&3.0*-t
100.4t2.2 76.8*2.5*-F 81.2+2.6*-f
97.5k3.2 79.7+3.4*-t 86.3&3.2*
47.6t0.7 39.3+2.4t 42.0+1.2~$
47.2t1.3 42.9*1.7$ 45.5t2.9
48.6tl.6 46.1t0.4 41.3+1.0*t$
7.36t0.01 7.41+0.02-t 7.4oko.olt
7.36t0.01 7.4O+O.Olt 7.41&0.01j-
7.36t0.01 7.39&0.02f7.41t0.01 t
* P < 0.05 vs. group’s
own control.
t
“f P < 0.05 vs. intact.
vented by RDNX (9). A subsequent study from this laboratory (14) concluded that renal sympathetic nerve activity also affected renal excretory function in conscious CST rats. This was based on the observation that acute pharmacological blockade of adrenoceptors resulted in marked natriuresis and kaliuresis 24 h after CST. We reasoned that this adrenergically mediated reabsorption of electrolytes and water could result in the return of arterial pressure toward control levels previously observed within 8-10 days after CST in conscious rats (10). Therefore, in the present study, we predicted that RDNX would result in a cumulative loss of electrolytes and water in CST rats. As a result, arterial pressure in renal-denervated CST rats would be lower compared with CST rats with intact renal nerves. The findings of this study were not consistent with this hypothesis. RDNX had no effect on any index of renal or cardiovascular function measured in this study. One explanation for this finding is that renal nerve activity, although initially elevated after CST, may diminish over time. CST results in an immediate twofold increase in renal nerve activity in anesthetized rats (9, 13). This acutely elevated nerve activity could account for the retention of electrolytes and water in anesthetized, acute CST rats with intact renal nerves compared with those with denervated kidneys (9). Indirect evidence suggests that renal nerve activity, although present, is no longer elevated in conscious rats 24 h after CST (14). It is therefore possible that nerve activity affects renal function only immediately after CST, but not 24 h after CST. However, despite indications of reduced renal sympathetic nerve activity in conscious spinal rats, we previously reported that functionally significant renal sympathetic nerve activity is present 24 h after CST (14). This was based on the acute natriuretic response to adrenergic blockade. Nevertheless, the present study showed no differences in renal function between rats with and without renal nerves at any time after CST. A potential weakness of this study is the restriction of sampling to 1 h every other day. Ideally, electrolyte and water excretion would have been measured continuously so that the cumulative effect of RDNX on 24-h water and electrolyte balances could be calculated. Continuous urine
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
DENERVATION
IN SPINAL T
R1269
RATS
6 --
A----A O--O a---.
L.
$
5--
INTACT RDNX+CST SHAM+CST
1
0 -3
-2
-1
0
1
2
3
4
5
6
7
8
9
DAYS AFTER CST Fig. 4. Effect of CST and RDNX on plasma renin activity (PRA). * P < 0.05 compared with before CST. + P < 0.05 compared with intact.
omom
omol
0
12
3
4
3
0
7
8
#
DAYS AFTER CST Fig. 3. Effect of RDNX on urine flow, urinary sodium excretion (U,,V), urinary potassium excretion (U,V), and urine pH in CST rats.
collection was not possible, however, since heating of cages caused urine to evaporate. Despite this limitation, l-h measurements of renal function appeared to be representative of renal excretory function, as hourly sodium and water excretion closely approximated intake. Furthermore, variations in intake were minimized by constant intravenous infusion of sodium and water. In addition, a cumulative effect on renal function would be expected to affect arterial pressure, yet there were no differences in MAP between RDNX and sham rats at any time for 9 days after CST. Body weight, another indicator of cumulative sodium and water balance, was also not different in RDNX compared with sham rats. Thus all data support our conclusion that RDNX has no effect on sodium or water excretion in CST rats. How can we explain the absence of an effect of RDNX in the present study, since our own previous study in-
dicated significant effects of adrenoceptor blockade on renal excretory function in rats 24 h after CST? One explanation is that although acute pharmacological blockade of renal nerve activity in CST rats resulted in immediate loss of sodium and potassium, the chronic effects of RDNX on renal function were offset by other regulatory mechanisms. This idea is supported by studies from other laboratories that have shown that RDNX has acute effects on renal function that are not sustained over time. For example, in rats on a normal sodium diet, RDNX results in an acute loss of sodium and negative sodium balance, but within 3 days the rats were in zero sodium balance (3,4). In one study (3), chronic effects of RDNX on renal excretory function were observed only when animals were challenged with a low-sodium diet. However, another study demonstrated that renal-denervated rats were able to maintain sodium balance throughout a period of sodium restriction (4). Based on these reports, it is not surprising then that RDNX did not chronically alter renal excretory function in CST rats, since RDNX was performed 7 days before initiation of measurements, and sodium and water intakes were fixed at normal levels 3 days before the study. It is likely that other regulatory mechanisms were responsible for chronic maintenance of sodium balance after RDNX in CST rats. An alternate explanation is that changes in renal excretory function observed in CST rats during adrenergic Table 2. Effect of RDNX and CST on plasma sodium, plasma potassium, and hematocrit Group
Days
Control (pre-CST)
After
3
CST 7
PNa,meqll Intact Sham RDNX
+ CST + CST
148.4t0.7 149.3t0.3 148.0tl.O
153.4t0.7* 152.6t1.2* 153.Otl.O*
152.4tl.l" 152.0t1.3 153.5t1.2*
PK,meqll Intact Sham RDNX
+ CST + CST
4.62kO.15 4.20k0.12 4.40t0.14
+ CST + CST
37.6t0.8 38.1k0.6 37.5~11.2
Hct, Intact Sham RDNX
4.03&0.10* 4.10t0.10 3.98&O. 14
4.12t0.06* 3.84t0.14 3.68kO.11*
33.5k1.3" 32.1t1.3* 33.6tl.5
31.3&1.3* 33.5t2.2* 35.6t1.5
%
Values are means k SE. No difference 0.05). * P < 0.05 compared with group’s
between groups own control.
(ANOVA,
P >
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
R1270
RENAL
DENERVATION
blockade was the result of circulating catecholamines rather than renal sympathetic nerve activity. If so, renal denervation would not be expected to affect renal function. We feel this is unlikely, however, since both adrenal nerve activity and adrenal catecholamine secretion have been shown to decrease by 90% with CST (1). Furthermore, nonneural catecholamine release is negligible, since adrenal demedullation does not decrease plasma catecholamines in pithed rats (16). Adrenal catecholamines are therefore unlikely to be responsible for the large increases in electrolyte excretion that accompanied adrenergic blockade in our previous study (14). PRA was measured as another variable affected by renal sympathetic nerve activity. As previously reported from this laboratory (14), there was evidence for neural support of PRA in intact rats, since RDNX rats tended to have lower PRA than rats with intact renal nerves before CST. However, the difference between RDNX and sham RDNX rats disappeared after CST. This finding is in accord with our previous study, which concluded that renal nerves do not influence basal PRA in spinal rats (14). In the present study, PRA values decreased gradually after CST, reaching levels not significantly different from zero by day 9. These extraordinarily low PRA values raised the suspicion that generation of ANG I reflected reduced substrate availability rather than a concentration of plasma renin near zero. Measurement of plasma renin concentration on day 9 confirmed that PRA values accurately reflected plasma renin level rather than decreased substrate. Further investigations are needed to determine whether these low PRC levels are the result of inhibition of renin synthesis or renin release. Nevertheless, since there was no difference in PRA between CST rats with and without renal nerves, it is clear that renal nerves play no role in regulation of basal PRA in conscious CST rats. In contrast to low PRA values of spinal rats, humans with spinal cord injury have normal (12, 15) or elevated (6-8) PRA levels. It is not known to what extent PRA is dependent on sympathetic nerve activity in spinal humans, but this study and others (7, 14) suggest that basal PRA is not neurally supported after spinal transection. It is interesting to note that lability of MAP increased steadily, doubling by 9 days after CST. The cause of this increase is not known. It is possible that increased spasticity of the hindquarters is responsible for this increased variability. Further investigations are needed to elucidate the source of this variability. In summary, RDNX had no effect on any index of renal or cardiovascular function in chronic CST rats. Despite previous evidence for a role of renal nerves in acute regulation of renal function, we conclude that renal nerves are not necessary for chronic regulation of arterial pressure or renal function in CST rats. We thank George Trachte for measuring renal catecholamines and Stephen Katz forthe use of his laboratory for PRA and PRC assays. We also thank Barbara Provo for technical assistance.
IN SPINAL
RATS
This work was supported by National Heart, Lung, and Blood InstiGrant HL-39619. Address for reprint requests: J. W. Osborn, Dept. of Animal Science, Univ. of Minnesota, 1988 Fitch, Rm. 435, St. Paul, MN 55108.
tute
Received
18 December
1991; accepted
in final
form
6 May
1992.
REFERENCES 1. Araki, T., K. Ito, M. Kurosawa, and A. Sato. Responses of adrenal sympathetic nerve activity and catecholamine secretion to cutaneous stimulation in anesthetized rats. Neuroscience 12: 289299, 1984. 2. Bioanalytical Systems, Inc. Plasma Catecholamines. LCEC Application Note No. 14, 1982. G. F., and L. L. Sawin. Renal nerves in renal adap3. Dibona, tation to dietary sodium restriction. Am. J. Physiol. 245 (Renal Fluid Electrolyte Physiol. 14): F322-F328, 1983. 4. Fernandez-Repollet, E., C. R. Siva-Netto, R. E. Colindres, and C. Gottschalk. Role of renal nerves in maintaining sodium balance in unrestrained conscious rats. Am. J. PhysioZ. 249 (Renal Fluid Electrolyte Physiol. 18): F819-F826, 1985. 5. Haber, E., T. Koerner, L. B. Page, B. Kliman, and A. Purnode. Application of a radioimmunoassay for angiotensin I to the physiologic measurements of plasma renin activity in normal human subjects. J. Clin. EndocrinoZ. 29: 1349-1355, 1969. 6. Mathias, C. J., N. J. Christensen, J. L. Corbett, H. L. Frankel, T. J. Goodwin, and W. S. Peart. Plasma catecholamines, plasma renin activity and plasma aldosterone in tetraplegic man, horizontal and tilted. CZin. Sci. MOL. Med. 49: 291-299, 1975. 7. Mathias, C. J., N. J. Christensen, H. L. Frankel, and W. S. Peart. Renin release during head-up tilt occurs independently of sympathetic nervous activity in tetraplegic man. CZin. Sci. 59: 251-256, 1980. C. J., H. L. Frankel, I. B. Davies, V. H. T. James, 8. Mathias, and W. S. Peart. Renin and aldosterone release during sympathetic stimulation in tetraplegia. Clin. Sci. 60: 399-404, 1981. J. W., R. H. Livingstone, and L. P. Schramm. 9. Osborn, Elevated renal nerve activity after spinal transection: effects on renal function. Am. J. Physiol. 253 (Regulatory Integrative Comp. Physiol. 22): R619-R625, 1987. 10. Osborn, J. W., R. F. Taylor, and L. P. Schramm. Determinants of arterial pressure after chronic spinal transection in rats. Am. J. Physiol. 256 (Regulatory Integrative Comp. Physiol. 25): R666-R673, 1989. 11. Osborn, J. W., R. F. Taylor, and L. P. Schramm. Chronic cervical spinal cord injury and autonomic hyperreflexia in rats. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R169-R174, 1990. 12. Tajima, F., S. Sagawa, J. Iwamoto, K. Miki, B. J. Freund, J. R. Claybaugh, and K. Shiraki. Cardiovascular, renal, and endocrine responses in male quadriplegics during head-out water immersion. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R1424-R1430, 1990. 13. Taylor, R. F., and L. P. Schramm. Differential effects of spinal transection on sympathetic nerve activities in rats. Am. J. Physiol. 253 (Regulatory Integrative Comp. Physiol. 22): R611R618, 1987. 14. Trostel, K. A., S. A. Katz, and J. W. Osborn. Functional evidence for sympathetic nerve activity in conscious cervical spinal rats. Am, J. Physiol. 261 (Regulatory Integrative Comp. Physiol. 30): R434-R441, 1991. 15. Wall, B. M., H. H. Williams, D. N. Presley, J. T. Crofton, E. G. Schneider, L. Share, and C. R. Cooke. Altered sensitivity of osmotically stimulated vasopressin release in quadriplegic subjects. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R827-R835, 1990. 16. Yamaguchi, I., and I. J. Kopin. Plasma catecholamine and blood pressure responses to sympathetic stimulation in pithed rats. Am. J. PhysioZ. 237 (Heart Circ. Physiol. 6): H305-H310, 1979.
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
nerve
activity;
plasma
renin
activity;
sodium
SYMPATHETIC NERVE ACTIVITY (SNA) is generally believed to originate in the brain stem and higher centers. However, recent studies in anesthetized rats with cervical spinal cord transection (CST) have provided evidence for spinally generated SNA. Specifically, direct nerve recordings show increased renal SNA after CST in anesthetized rats (9,13). Furthermore, the CST-induced increase in renal nerve activity acutely affected water and electrolyte handling in anesthetized rats, suggesting a functional role for spinally generated SNA (9). Is there evidence for SNA in unanesthetized spinal animals? If so, is it functionally significant? We have recently reported that spinally generated SNA exists in conscious spinal rats. This is based on the observation that acute adrenergic blockade results in three- to fourfold increases in sodium and potassium excretion in conscious rats 24 h after CST (14). Taken together, studies in both anesthetized and conscious CST rats suggest that renal sympathetic nerve discharge is present and decreases water and electrolyte excretion. Furthermore, these observations suggest that renal SNA, via regulation of fluid and electrolyte excretion, may be of functional significance in the long-term regulation of arterial pressure in spinal animals. Indeed, two previous studies in conscious spinal rats suggest that renal SNA may play a role in regulation of renal function and therefore arterial pressure. First, despite the fact that CST results in 0363-6119/92
$2.00 Copyright
renal function
in Neuroscience,
a loss of adrenergic vasomotor tone, arterial pressure is not different from control levels within 8-10 days after CST (10). Although this time course is compatible with a volume-dependent mechanism, the role of renal SNA in control of fluid balance was not investigated in that study. Further support for such a mechanism was provided by a recent study from our laboratory demonstrating that adrenergic antagonists markedly increased electrolyte excretion in conscious rats 24 h after CST (14). However, these responses were monitored for only 2 h after adrenergic blockade, and we were therefore unable to conclude whether spinally generated renal SNA is a determinant in the long-term regulation of renal function and arterial pressure. The present study was undertaken to directly test the hypothesis that renal SNA chronically influences electrolyte handling and therefore arterial pressure in conscious spinal rats. We measured renal function and arterial pressure in renal-denervated CST rats, before and for up to 9 days after spinal transection. We predicted that, if renal SNA exists in spinal rats, renal denervation would result in an enhanced excretory rate of water and electrolytes compared with CST rats with intact renal nerves. Moreover, this would be manifested as an impairment in the long-term regulation of arterial pressure in renal-denervated CST rats. METHODS General Procedures Male Sprague-Dawley rats (Sasco, Omaha, NE) were housed until the time of study in our animal housing facility with controlled temperature and lighting. Standard rat food and water were provided ad libitum. After a minimum of 2 days in this facility, rats were brought to the laboratory for renal denervation (RDNX) or sham RDNX (see below). Subsequently, they were housed in a quiet, isolated laboratory with a 12:12-h daynight cycle. All procedures were approved by the institutional animal care committee and were conducted in accordance with institutional and National Institutes of Health guidelines. Surgical Procedures The overall schedule of surgical procedures in relation to the experimental protocol is summarized in Fig. 1. Renal denervation. Rats were fasted overnight before surgery. The next morning, rats were atropinized (0.13 mg/kg ip) and anesthetized with pentobarbital sodium (50 mg/kg ip). Rats were divided into two groups for surgery: bilateral RDNX and sham RDNX. With the use of a dissecting microscope, bilateral RDNX was performed via a ventral midline incision. The area between the hilus and the origin of renal vessels at the aorta and vena cava was stripped of fat and connective tissue, then painted with 10% phenol in alcohol to ensure destruction of any remaining nerves. In sham RDNX rats, a ventral midline incision was made, and each kidney was exposed briefly. After RDNX or sham RDNX, the incision was closed and rats
0 1992 The
American
Physiological
Society
R1265
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
R1266 DAY
43
RENAL PROCEDURE
I
-4
RDNX
NUMBER
BEGIN
INFUSION
-1 0 1
3
i I I
I
5 7 9
I
CONTROL MEASUREMENTS CST OR SHAM-CST EXPERIMENT
SHAM-RDNX 13
J
/\
RDNX+CST
SHAM+CST
IN SPINAL
RATS
cages. Rats were alert and mobile the day after CST, and began to eat as early as the first day after CST. As a result of plugged arterial catheters, rats dropped out of the experiment over time. In addition, when a urine collection was unsuccessful, all data from the rat were excluded from the analysis for that day. As a result, sample sizes were not constant throughout the experiment. The number of animals studied in each group is shown in Fig. 1.
OF ANIMALS
RDNX 7
OR SHAM-RDNX
CATHETERIZE,
DENERVATION
INTACT
7 4
7 6
6 6
EXPERIMENT
6
6
5
EXPERIMENT
7
6
5
EXPERIMENT
7
5
5
EXPERIMENT
7
3
4
Fig. 1. Experimental protocol. RDNX, renal denervation; CST, cervical spinal cord transection. Numbers in each group for each day are shown.
received 10 ml of warm sterile isotonic saline intraperitoneally and 100,000 U penicillin G intramuscularly. They were subsequently housed in pairs and fed standard rat food and water ad libitum for 4 days of recovery before femoral catheter implantation. Chronic catheterization. Four days after RDNX or sham RDNX, rats were atropinized (0.13 mg/kg ip) and anesthetized with pentobarbital sodium (50 mg/kg ip) for implantation of femoral arterial and venous catheters. Catheters were tunneled subcutaneously to the top of the head, exteriorized, and passed through a metal spring that was attached to dental acrylic mounted on the rats’ heads. The rats received antibiotic after surgery (100,000 U penicillin G im) and were placed in individual stainless steel metabolic cages. The spring extended to the top of the cage, where it was attached to a hydraulic swivel, which allowed the rats to move freely. An intravenous infusion of lactated Ringer was begun (1 ml/h; 130 meq/l Na, 4 meq/l K, 28 meq/l lactate) and continued for the duration of the study. Rats were not allowed to drink and were fed sodium-free food, thereby fixing water and sodium consumption at 24 ml/day and 3.12 meq/day, respectively, throughout the experiment. This was done to ensure that changes in plasma renin activity and sodium and water excretion were not secondary to alterations in sodium and water intake. Arterial lines were drained and filled daily with 1,000 U/ml heparin, and antibiotic (20 mg ampicillin iv) was administered. Spinal transection. Four days after catheter implantation, rats were atropinized (0.26 mg/kg iv) and anesthetized with pentobarbital sodium (50 mg/kg iv). CST was performed on all RDNX rats (RDNX + CST, n = 7). Sham RDNX rats were divided into two groups. In one group, CST was performed (sham + CST, n = 7), and in the other, a sham CST surgery was performed (intact, n = 6). Preceding CST, a bladder catheter was implanted, and rats were placed in a stereotaxic apparatus for spinal transection as previously described (11). After exposing the seventh cervical and first thoracic vertebrae, the spinal cord was cut using a no. 11 scalpel blade. A blunt hypodermic 18-gauge needle, attached to a vacuum pump, was then used to complete the transection. Completeness of transection was verified by inspection. In the intact group, the bladder was exposed through a midline incision, and the muscle and skin were sutured closed. A skin incision along the neck, similar to that in CST rats, was also made and closed. After CST or sham CST, rats were returned to their home cages. Body temperature of CST rats was maintained within normal limits by circulating warm, humidified air around the
Experimental
Protocol
To examine the influence of renal nerves on cardiovascular function in CST rats, the following variables were measured before and over a g-day period after CST or sham CST: mean arterial pressure (MAP), heart rate (HR), lability of MAP, arterial oxygen tension, arterial carbon dioxide tension, arterial pH, hematocrit (Hct), plasma sodium (PNa), plasma potassium (PK), and plasma renin activity (PRA). Measurements were made before (day -1 and day 0), and 1, 3,5,7, and 9 days after CST. MAP was measured by connecting the arterial catheter to a Spectramed PlOEZ pressure transducer coupled to a Grass polygraph (model 7D). The pulsatile pressure signal was input to a second amplifier with a low-pass filter to acquire an electrical average of arterial pressure. MAP was monitored by computer (Asystant+, Asyst Software Technologies, New York, NY) for 1 h at a sampling rate of 0.5 Hz while the rats rested quietly in their home cages. A time average of MAP was then calculated from the 1,800 data points. The statistical variance was also determined as a quantitative indicator of lability of arterial pressure. HR was determined periodically by increasing the chart speed and counting peaks on the pulsatile pressure tracing. After measurement of MAP and HR, a 500-~1 arterial blood sample was collected into a chilled l-ml syringe containing 1.25 mg EDTA in a volume of 25 ~1. One hundred microliters of this sample was used to measure arterial blood gasses (pH/blood gas analyzer model 1304, Instrumentation Laboratory, Lexington, MA). On days -I, 1, 5, and 9, the remaining blood was transferred to a chilled microcentrifuge vial. This sample was centrifuged, and plasma was collected and frozen at -70°C for later radioimmunoassay (RIA) for PRA (see below). On days 0, 3, and 7, the remaining 400 ~1 of blood was used to measure PNa, PK (NOVA-l sodium-potassium analyzer, NOVA Biomedical, Waltham, MA), and Hct. Sample blood was replaced by an equal volume of whole blood from a conscious, chronically instrumented donor rat. In addition to these measurements, renal function was compared between RDNX + CST and sham + CST rats over the g-day period after CST. To compare the results of this experiment with our previous study in which adrenergic blockade acutely increased electrolyte excretion in spinal rats (14), identical methods of urine collection and analysis were used. Briefly, during the 60-min MAP recording period, urine was continuously collected into a preweighed balloon attached to the bladder catheter. Such a closed collection system was required, since the slow continuous urine flow into the funnels of the warmed metabolic cages results in large evaporative losses of urine. Urine flow through the urethra was prevented by clamping the penis with a miniature clamp (Tiemann, Plainview, NY) before the first collection period. Although the patency of bladder catheters was verified daily, clamping ensured completeness of urine collection. At the time urine collection was begun, basal HR and MAP were not different from before clamping, suggesting that use of the penis clamp did not result in reflex SNA. Because urine could not be collected in intact rats in a manner identical to that used for spinal rats, no renal function measurements were made in this group. Finally, at the end of the urine collection period, urine flow was determined gravimetrically, and urine was analyzed for sodium and potassium concentration using a NOVA- 1 sodium-potassium analyzer. Urinary sodium
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
DENERVATION
excretion and urinary potassium excretion were calculated as the product of urine flow rate and ion concentration. Urine pH was also measured immediately after the urine collection period using a pH-sensitive microelectrode (Radiometer America, Westlake, OH). Determination
of PRA
PRA was measured as previously described (14) by quantitation of endogenously generated angiotensin I (ANG I) using an RIA kit (1251-ANG I RIA kit, New England Nuclear, Boston, MA) based on an adaptation of the method of Haber et al. (5). In addition, plasma renin concentration (PRC) was measured for all rats on day 9 after CST by quantitation of endogenously generated ANG I in the presence of an excess of rat angiotensinogen. Verification
IN
SPINAL
R1267
RATS
120
I”
t 8l\ &---F+
100
E E 80
a, 4 5;
60
O-0 *.
*,+ *,+
+ +,+
*,+ *,+
*
+,+
a
of RDNX
Completeness of RDNX was confirmed by assay for renal catecholamines. After data collection on day 9, rats were anesthetized with pentobarbital sodium, and both kidneys were removed, sectioned, and inspected for signs of hydronephrosis or infection. No signs of either were observed. Sections were immediately frozen in liquid nitrogen. Kidneys were weighed, wrapped in aluminum foil, and stored at -70°C. Norepinephrine was extracted from the tissue by an established technique (2). The samples were assayed for norepinephrine content by high-pressure liquid chromatography (HPLC) with electrochemical detection. HPLC analysis was performed with a Spectra-physics SP 8700 XR liquid chromatograph; a Spectra Physics SP 4290 integrator; a Brownlee Laboratories no. 53358, RP-18, 100 x 4.6 mm, 5pm column; and a Bioanalytical Systems LC-4B electronic controller and LC-17A oxidative flow cell. The mobile phase consisted of 0.1 M KH2POd, 0.1 mM EDTA, 0.4 mM octyl sodium sulfate, and 3.5% by volume HPLC-grade methanol dissolved in Millipore-filtered deionized water. The efficiency of norepinephrine extraction was corrected by the recovery of the internal standard DHBA.
CK I
o-o e-0
250
+,+ +
+I+ *a+
+I+ +,+
+
+,+ +
+,+
‘I+
+,+
l ,+
200
o-o )
*o
+
‘I+
+,+
*,+
Statis tics All values are presented as means t SE. Groups were compared using an analysis of variance (ANOVA) for an unbalanced repeated measures design (SAS Institute, Cary, NC). A finding of significant effect of group was followed by Duncan’s test. Values of cardiovascular and renal function after CST or sham CST were compared with pretransection values using Dunnett’s test. All values were considered significant for P < 0.05. RESULTS
Effect of RDNX
and CST on Arterial
Pressure
RDNX had no statistically significant effect on cardiovascular function in CST rats. MAP, HR, and lability of MAP of RDNX + CST and sham + CST rats were not significantly different from each other (Fig. 2). In contrast, CST had significant effects on MAP, HR, and lability of MAP. In both CST groups, MAP fell -25 mmHg, to a posttransection level of 75 mmHg the first day after CST. MAP recovered from this posttransection low, increasing 10 mmHg to a level of 45 mmHg by the end of the study. By day 9, MAP was no longer significantly lower in the CST groups than in intact rats. MAP in the RDNX + CST group was also not significantly different from its own pretransection level. HR fell substantially with CST (450 beats/min for each CST group), and remained at low levels thereafter. Despite low or normal levels of arterial pressure after transection,
o!
:
-2-l
;
:
:
:
:
;
;
;
f
i
0
1
2
3
4
5
6
7
8
9
DAYS AFTER CST Fig. 2. Effect of CST and RDNX on mean arterial pressure (MAP), heart rate (HR), and variance of MAP. * P < 0.05 compared with before CST. + P < 0.05 compared with intact. # P < 0.05 compared with sham + CST.
lability of MAP increased over time in CST rats. By day 3 after CST, lability of MAP was significantly greater than pretransection levels, and by the end of the study, lability of MAP was approximately double the pretransection level. Because cardiovascular function may be affected by arterial blood gases or pH, these variables were measured in all three groups (Table 1). Although there were differences between intact and CST groups, there was little effect of RDNX on arterial blood gases or pH. Effect of RDNX
and CST on Renal Function
Contrary to the predictions of our hypothesis, RDNX had no effect on renal excretory function after CST (Fig. 3). There were no statistically significant differences
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
R1268
RENAL
Table 1. Effect of RDNX
Intact Sham RDNX
and CST on arterial
+ CST + CST
94.0tl.4 75.1+3.7”“r 84.8+4.5*-f*
Intact Sham RDNX
+ CST + CST
45.8tl.O 44.1t1.3 45.5t0.6
46.5t0.8 39.6+1.6-f39.3&2.4-f
Intact Sham RDNX
+ CST + CST
7.38t0.01 7.4O+O.Ol”f 7.39t0.01
7.40t0.01 7.42+O.Olt 7.43tO.Ol*t spinal
cord transection;
3
PO,, mmHg lOO.Ot3.7 73.3+2.4*-f 75.0+3.1*t Pcoz, mmHg 49.2tl.l 43.1k1.57 43.4+3.0-f Arterial pH 7.35t0.01 7.41+0.02-f 7.38tO.Ol-F RDNX,
between sham + CST and RDNX + CST groups for any of the renal function variables. Both groups of rats appeared to be in sodium and water balance as early as 24 h after CST, since excretory values closely approximated intake, which was fixed at 0.130 meq/h Na and 1 ml/h H20. Furthermore, excretion rates remained unchanged over the course of the study. As with renal excretory function, PRA was not affected by RDNX in CST rats (Fig. 4). However, CST resulted in significant decreases in PRA. After CST, PRA continued to decline, until by day 9 PRA was not significantly different from zero. The extremely low levels of PRA after CST (Fig. 4) may be explained by a shortage of substrate rather than low PRC. To investigate this possibility, PRC was also measured for all rats on day 9 after CST. PRC was 0.2 t 0.1 ng ANG I ml-l h-l for CST rats and 6.5 t 0.6 ng ANG I ml-l h-l for intact rats. Therefore, plasma renin levels were indeed extremely low 9 days after CST. Plasma electrolytes and Hct are shown in Table 2. Alor Hct was different after CST or sham though PNa~ pK, CST, these variables were neither affected by RDNX nor were they different between intact and CST rats. l
l
Verification
RATS
blood gases and pH
I
98.021.8 94.6t2.7 98.7t1.3
Values are means t SE. CST, cervical $ P < 0.05 vs. sham + CST.
IN SPINAL
Days
Control (pre-CST)
Group
DENERVATION
l
l
of RDNX
Renal norepinephrine was not significantly different (ANOVA followed by Duncan’s test, P > 0.05) between intact rats (128.6 t 10.9 rig/g, n = 8 kidneys) and sham + CST rats (84.8 t 17.2 rig/g, n = 8 kidneys). Renal norepinephrine was significantly lower in RDNX rats (3.3 k 1.1 rig/g, n = 12 kidneys). DISCUSSION
This experiment was designed to test the hypothesis that tonic renal sympathetic nerve activity influences electrolyte handling in conscious spinal rats. We proposed that renal nerve activity would result in cumulative retention of sodium and water and thereby chronically influence blood pressure in CST rats. This hypothesis was based on previous experiments in both anesthetized and conscious rats. In anesthetized rats, increased renal sympathetic nerve activity after CST resulted in a significant antidiuresis and antikaliuresis that could be pre-
renal denervation.
After
CST
5
7
9
97.4t2.8 77.0+2.4*t 78.6&3.0*-t
100.4t2.2 76.8*2.5*-F 81.2+2.6*-f
97.5k3.2 79.7+3.4*-t 86.3&3.2*
47.6t0.7 39.3+2.4t 42.0+1.2~$
47.2t1.3 42.9*1.7$ 45.5t2.9
48.6tl.6 46.1t0.4 41.3+1.0*t$
7.36t0.01 7.41+0.02-t 7.4oko.olt
7.36t0.01 7.4O+O.Olt 7.41&0.01j-
7.36t0.01 7.39&0.02f7.41t0.01 t
* P < 0.05 vs. group’s
own control.
t
“f P < 0.05 vs. intact.
vented by RDNX (9). A subsequent study from this laboratory (14) concluded that renal sympathetic nerve activity also affected renal excretory function in conscious CST rats. This was based on the observation that acute pharmacological blockade of adrenoceptors resulted in marked natriuresis and kaliuresis 24 h after CST. We reasoned that this adrenergically mediated reabsorption of electrolytes and water could result in the return of arterial pressure toward control levels previously observed within 8-10 days after CST in conscious rats (10). Therefore, in the present study, we predicted that RDNX would result in a cumulative loss of electrolytes and water in CST rats. As a result, arterial pressure in renal-denervated CST rats would be lower compared with CST rats with intact renal nerves. The findings of this study were not consistent with this hypothesis. RDNX had no effect on any index of renal or cardiovascular function measured in this study. One explanation for this finding is that renal nerve activity, although initially elevated after CST, may diminish over time. CST results in an immediate twofold increase in renal nerve activity in anesthetized rats (9, 13). This acutely elevated nerve activity could account for the retention of electrolytes and water in anesthetized, acute CST rats with intact renal nerves compared with those with denervated kidneys (9). Indirect evidence suggests that renal nerve activity, although present, is no longer elevated in conscious rats 24 h after CST (14). It is therefore possible that nerve activity affects renal function only immediately after CST, but not 24 h after CST. However, despite indications of reduced renal sympathetic nerve activity in conscious spinal rats, we previously reported that functionally significant renal sympathetic nerve activity is present 24 h after CST (14). This was based on the acute natriuretic response to adrenergic blockade. Nevertheless, the present study showed no differences in renal function between rats with and without renal nerves at any time after CST. A potential weakness of this study is the restriction of sampling to 1 h every other day. Ideally, electrolyte and water excretion would have been measured continuously so that the cumulative effect of RDNX on 24-h water and electrolyte balances could be calculated. Continuous urine
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
DENERVATION
IN SPINAL T
R1269
RATS
6 --
A----A O--O a---.
L.
$
5--
INTACT RDNX+CST SHAM+CST
1
0 -3
-2
-1
0
1
2
3
4
5
6
7
8
9
DAYS AFTER CST Fig. 4. Effect of CST and RDNX on plasma renin activity (PRA). * P < 0.05 compared with before CST. + P < 0.05 compared with intact.
omom
omol
0
12
3
4
3
0
7
8
#
DAYS AFTER CST Fig. 3. Effect of RDNX on urine flow, urinary sodium excretion (U,,V), urinary potassium excretion (U,V), and urine pH in CST rats.
collection was not possible, however, since heating of cages caused urine to evaporate. Despite this limitation, l-h measurements of renal function appeared to be representative of renal excretory function, as hourly sodium and water excretion closely approximated intake. Furthermore, variations in intake were minimized by constant intravenous infusion of sodium and water. In addition, a cumulative effect on renal function would be expected to affect arterial pressure, yet there were no differences in MAP between RDNX and sham rats at any time for 9 days after CST. Body weight, another indicator of cumulative sodium and water balance, was also not different in RDNX compared with sham rats. Thus all data support our conclusion that RDNX has no effect on sodium or water excretion in CST rats. How can we explain the absence of an effect of RDNX in the present study, since our own previous study in-
dicated significant effects of adrenoceptor blockade on renal excretory function in rats 24 h after CST? One explanation is that although acute pharmacological blockade of renal nerve activity in CST rats resulted in immediate loss of sodium and potassium, the chronic effects of RDNX on renal function were offset by other regulatory mechanisms. This idea is supported by studies from other laboratories that have shown that RDNX has acute effects on renal function that are not sustained over time. For example, in rats on a normal sodium diet, RDNX results in an acute loss of sodium and negative sodium balance, but within 3 days the rats were in zero sodium balance (3,4). In one study (3), chronic effects of RDNX on renal excretory function were observed only when animals were challenged with a low-sodium diet. However, another study demonstrated that renal-denervated rats were able to maintain sodium balance throughout a period of sodium restriction (4). Based on these reports, it is not surprising then that RDNX did not chronically alter renal excretory function in CST rats, since RDNX was performed 7 days before initiation of measurements, and sodium and water intakes were fixed at normal levels 3 days before the study. It is likely that other regulatory mechanisms were responsible for chronic maintenance of sodium balance after RDNX in CST rats. An alternate explanation is that changes in renal excretory function observed in CST rats during adrenergic Table 2. Effect of RDNX and CST on plasma sodium, plasma potassium, and hematocrit Group
Days
Control (pre-CST)
After
3
CST 7
PNa,meqll Intact Sham RDNX
+ CST + CST
148.4t0.7 149.3t0.3 148.0tl.O
153.4t0.7* 152.6t1.2* 153.Otl.O*
152.4tl.l" 152.0t1.3 153.5t1.2*
PK,meqll Intact Sham RDNX
+ CST + CST
4.62kO.15 4.20k0.12 4.40t0.14
+ CST + CST
37.6t0.8 38.1k0.6 37.5~11.2
Hct, Intact Sham RDNX
4.03&0.10* 4.10t0.10 3.98&O. 14
4.12t0.06* 3.84t0.14 3.68kO.11*
33.5k1.3" 32.1t1.3* 33.6tl.5
31.3&1.3* 33.5t2.2* 35.6t1.5
%
Values are means k SE. No difference 0.05). * P < 0.05 compared with group’s
between groups own control.
(ANOVA,
P >
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
R1270
RENAL
DENERVATION
blockade was the result of circulating catecholamines rather than renal sympathetic nerve activity. If so, renal denervation would not be expected to affect renal function. We feel this is unlikely, however, since both adrenal nerve activity and adrenal catecholamine secretion have been shown to decrease by 90% with CST (1). Furthermore, nonneural catecholamine release is negligible, since adrenal demedullation does not decrease plasma catecholamines in pithed rats (16). Adrenal catecholamines are therefore unlikely to be responsible for the large increases in electrolyte excretion that accompanied adrenergic blockade in our previous study (14). PRA was measured as another variable affected by renal sympathetic nerve activity. As previously reported from this laboratory (14), there was evidence for neural support of PRA in intact rats, since RDNX rats tended to have lower PRA than rats with intact renal nerves before CST. However, the difference between RDNX and sham RDNX rats disappeared after CST. This finding is in accord with our previous study, which concluded that renal nerves do not influence basal PRA in spinal rats (14). In the present study, PRA values decreased gradually after CST, reaching levels not significantly different from zero by day 9. These extraordinarily low PRA values raised the suspicion that generation of ANG I reflected reduced substrate availability rather than a concentration of plasma renin near zero. Measurement of plasma renin concentration on day 9 confirmed that PRA values accurately reflected plasma renin level rather than decreased substrate. Further investigations are needed to determine whether these low PRC levels are the result of inhibition of renin synthesis or renin release. Nevertheless, since there was no difference in PRA between CST rats with and without renal nerves, it is clear that renal nerves play no role in regulation of basal PRA in conscious CST rats. In contrast to low PRA values of spinal rats, humans with spinal cord injury have normal (12, 15) or elevated (6-8) PRA levels. It is not known to what extent PRA is dependent on sympathetic nerve activity in spinal humans, but this study and others (7, 14) suggest that basal PRA is not neurally supported after spinal transection. It is interesting to note that lability of MAP increased steadily, doubling by 9 days after CST. The cause of this increase is not known. It is possible that increased spasticity of the hindquarters is responsible for this increased variability. Further investigations are needed to elucidate the source of this variability. In summary, RDNX had no effect on any index of renal or cardiovascular function in chronic CST rats. Despite previous evidence for a role of renal nerves in acute regulation of renal function, we conclude that renal nerves are not necessary for chronic regulation of arterial pressure or renal function in CST rats. We thank George Trachte for measuring renal catecholamines and Stephen Katz forthe use of his laboratory for PRA and PRC assays. We also thank Barbara Provo for technical assistance.
IN SPINAL
RATS
This work was supported by National Heart, Lung, and Blood InstiGrant HL-39619. Address for reprint requests: J. W. Osborn, Dept. of Animal Science, Univ. of Minnesota, 1988 Fitch, Rm. 435, St. Paul, MN 55108.
tute
Received
18 December
1991; accepted
in final
form
6 May
1992.
REFERENCES 1. Araki, T., K. Ito, M. Kurosawa, and A. Sato. Responses of adrenal sympathetic nerve activity and catecholamine secretion to cutaneous stimulation in anesthetized rats. Neuroscience 12: 289299, 1984. 2. Bioanalytical Systems, Inc. Plasma Catecholamines. LCEC Application Note No. 14, 1982. G. F., and L. L. Sawin. Renal nerves in renal adap3. Dibona, tation to dietary sodium restriction. Am. J. Physiol. 245 (Renal Fluid Electrolyte Physiol. 14): F322-F328, 1983. 4. Fernandez-Repollet, E., C. R. Siva-Netto, R. E. Colindres, and C. Gottschalk. Role of renal nerves in maintaining sodium balance in unrestrained conscious rats. Am. J. PhysioZ. 249 (Renal Fluid Electrolyte Physiol. 18): F819-F826, 1985. 5. Haber, E., T. Koerner, L. B. Page, B. Kliman, and A. Purnode. Application of a radioimmunoassay for angiotensin I to the physiologic measurements of plasma renin activity in normal human subjects. J. Clin. EndocrinoZ. 29: 1349-1355, 1969. 6. Mathias, C. J., N. J. Christensen, J. L. Corbett, H. L. Frankel, T. J. Goodwin, and W. S. Peart. Plasma catecholamines, plasma renin activity and plasma aldosterone in tetraplegic man, horizontal and tilted. CZin. Sci. MOL. Med. 49: 291-299, 1975. 7. Mathias, C. J., N. J. Christensen, H. L. Frankel, and W. S. Peart. Renin release during head-up tilt occurs independently of sympathetic nervous activity in tetraplegic man. CZin. Sci. 59: 251-256, 1980. C. J., H. L. Frankel, I. B. Davies, V. H. T. James, 8. Mathias, and W. S. Peart. Renin and aldosterone release during sympathetic stimulation in tetraplegia. Clin. Sci. 60: 399-404, 1981. J. W., R. H. Livingstone, and L. P. Schramm. 9. Osborn, Elevated renal nerve activity after spinal transection: effects on renal function. Am. J. Physiol. 253 (Regulatory Integrative Comp. Physiol. 22): R619-R625, 1987. 10. Osborn, J. W., R. F. Taylor, and L. P. Schramm. Determinants of arterial pressure after chronic spinal transection in rats. Am. J. Physiol. 256 (Regulatory Integrative Comp. Physiol. 25): R666-R673, 1989. 11. Osborn, J. W., R. F. Taylor, and L. P. Schramm. Chronic cervical spinal cord injury and autonomic hyperreflexia in rats. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R169-R174, 1990. 12. Tajima, F., S. Sagawa, J. Iwamoto, K. Miki, B. J. Freund, J. R. Claybaugh, and K. Shiraki. Cardiovascular, renal, and endocrine responses in male quadriplegics during head-out water immersion. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R1424-R1430, 1990. 13. Taylor, R. F., and L. P. Schramm. Differential effects of spinal transection on sympathetic nerve activities in rats. Am. J. Physiol. 253 (Regulatory Integrative Comp. Physiol. 22): R611R618, 1987. 14. Trostel, K. A., S. A. Katz, and J. W. Osborn. Functional evidence for sympathetic nerve activity in conscious cervical spinal rats. Am, J. Physiol. 261 (Regulatory Integrative Comp. Physiol. 30): R434-R441, 1991. 15. Wall, B. M., H. H. Williams, D. N. Presley, J. T. Crofton, E. G. Schneider, L. Share, and C. R. Cooke. Altered sensitivity of osmotically stimulated vasopressin release in quadriplegic subjects. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R827-R835, 1990. 16. Yamaguchi, I., and I. J. Kopin. Plasma catecholamine and blood pressure responses to sympathetic stimulation in pithed rats. Am. J. PhysioZ. 237 (Heart Circ. Physiol. 6): H305-H310, 1979.
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
