INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NORMALLY FUNCTIONING kidneys exhibit the phenomenon of pressure natriuresis, which allows the kidneys to respond to elevations in arterial pressure by increasing sodium and water excretion. During periods of elevated blood pressure this mechanism increases urine output and eliminates extraneous volume until the arterial pressure is normalized. If arterial pressure falls, urine output is reduced until the arterial pressure is again normalized (12). A number of observations suggest that the renal efferent sympathetic nerves play a pivotal role in long-term blood pressure regulation by modulating renal hemodynamic and excretory function and consequently the pressure-natriuresis relationship. This modulation by the renal nerves occurs primarily via increased renal vascular resistance, increased renal sodium and water reabsorption, and increased renin release (5, 6). Results from measurements of renal tissue norepinephrine concentration or renal vein norepinephrine spillover, renal nerve stimulation and recording experiments, as well as renal denervation studies in various hypertensive models all suggest that altered renal sympathetic nerve activity contributes to the derangements in renal function that lead to the development of hypertension (5, 14, 24).

Despite the wealth of evidence implicating the renal nerves in the development of hypertension, few, if any, studies of the electrophysiological behavior of efferent sympathetic neurons have been performed. This is mainly due to technical difficulties associated with isolation of renal sympathetic neurons because the nerve cell bodies are located in the paravertebral ganglia and celiac ganglion complex and make up only a small proportion of the neurons of these ganglia (23, 27). The advent of fluorescent retrograde tracers has enabled isolation and identification of single neuron cell bodies for the investigation of the biophysical properties of these neurons using the patch-clamp technique (15). Two potential problems exist that could seriously compromise electrophysiological studies of identified renal sympathetic neurons. First, the dye-labeling technique employs application of the fluorescent tracer to the nerve terminals by microinjection into the kidney and could adversely affect renal hemodynamic and excretory function, thereby compromising the regulation of extracellular fluid volume and mean arterial pressure in the animals studied. Second, the dye or UV illumination required to identify the labeled neurons could alter their properties, compromising the subsequent electrophysiological studies.

The present study was performed to determine whether renal efferent sympathetic neurons of normotensive rats can be identified and isolated by injecting the retrograde tracer fast blue directly into the kidney and whether this dye-label injection procedure adversely affects renal hemodynamic and excretory function. In addition, we determined whether the presence of the tracer and the short period of UV illumination required to identify dispersed renal sympathetic neurons altered G protein modulation of Ca²⁺ currents in these neurons.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Male Sprague-Dawley rats (Harlan, Indianapolis, IN) weighing between 300 and 400 g were used in the present study. The rats were maintained on a 12:12-h light-dark cycle and were fed standard laboratory rat chow (Purina, Richmond, IN) and had free access to water.

Identification of renal sympathetic neurons. Renal efferent sympathetic neurons were labeled using the fluorescent retrograde tracer fast blue, after aseptic surgical isolation of the left kidney (23, 27). Briefly, the rats were anesthetized with pentobarbital sodium (50 mg/kg ip), and the left kidney was exteriorized via a flank incision. A total of 10 µl of 2% fast blue in distilled water was injected at multiple sites (~10) into the kidney cortex via a 30-gauge needle connected to a solenoid-controlled pressure-injection system. After injection of the fluorescent tracer the abdominal musculature was sutured and the skin was closed with wound clips. The animals received an injection of penicillin-streptomycin to prevent infection and were placed on a heating pad to recover from the effects of the anesthesia. Renal function studies were performed 1 wk later according to the following protocol.

Renal function studies. Ten rats previously injected with the retrograde tracer were prepared for renal function experiments in the following manner. The animals were anesthetized with Inactin (100 mg/kg ip) and placed on a thermostatically controlled heated surgical table. After a tracheostomy, a polyethylene catheter (PE-50) was inserted into a carotid artery for measurement of mean arterial pressure (MAP) and collection of arterial blood samples. Both jugular veins were catheterized to allow continuous infusion of solutions and additional anesthetic as required. The animals received an intravenous bolus injection of an isotonic saline solution containing 10% inulin and 1.28% p-aminohippurate (PAH), administered at 200 µl/100 g body wt. This injection was followed by a continuous infusion of isotonic saline containing 5% inulin and 0.64% PAH at a rate of 500 µl · 100 g body wt¹ · h¹. After a laparotomy, the left and right ureters were catheterized (PE-10) to obtain free-flow urine collections from the left (dye injected) and right (control) kidneys. A surgical stabilization period of at least 30 min was observed, and then coincident with at least 60 min of inulin-PAH infusion the first of two consecutive 30-min baseline renal clearance periods was begun. A 200-µl arterial blood sample was obtained at the beginning, midpoint, and end of the 60-min control (Pre) period. These two baseline renal clearance periods (Pre) were obtained in all 10 rats. The rats were then randomly divided into two groups. Group 1 consisted of five rats that were subjected to a rapid hemorrhage maneuver totaling 5% of calculated blood volume (~8% of body wt). After a 30-min equilibration period a third clearance period (Post) of 30 min was then obtained. The five rats in group 2 were infused intravenously with 0.9% saline to achieve a volume expansion of 2.5% body wt in 30 min. The saline infusion was then matched to urinary output, and a 30-min (Post) clearance period was then obtained in these rats. In both groups, arterial blood samples (200 µl) were obtained at the beginning, midpoint, and end of the experimental (Post) clearance period. After completion of the experimental (Post) period in both groups, the animals were prepared for cell isolation as subsequently described. Coincident with the cell isolation procedure the kidneys were inspected for visible damage, removed from the animal, blotted dry, and weighed.

Isolation of renal sympathetic neurons. After the renal function protocol, the rats were decapitated while they were still anesthetized, and paravertebral ganglion (PVG) neurons were isolated using the enzymatic dispersion technique described previously (22). Briefly, the peritoneal cavity was opened, the viscera were removed, and the peritoneal cavity was irrigated with ice-cold Hanks' balanced salt solution (HBSS). The left paravertebral ganglia (T₁₂-L₁) were dissected free, and the connective tissue capsule was removed. The ganglia were then minced and transferred to 5 ml of modified Earle's balanced salt solution (EBSS) containing 1 mg/ml collagenase D and 0.5 mg/ml trypsin (Boehringer Mannheim Biochemicals, Indianapolis, IN) and 0.1 mg/ml DNase type I (Sigma, St. Louis, MO). The EBSS was modified by the addition of 3.6 g/l glucose and 10 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) and adjusted to pH 7.4 with NaOH before the addition of NaHCO₃. The enzyme solution containing the ganglion fragments was then incubated at 34°C under 5% CO₂-95% O₂ in a shaking water bath. After 1 h, the flask was shaken vigorously by hand to release the cell somata from the ganglion fragments. The resulting cell suspension was brought up to 10 ml by the addition of modified HBSS containing 10% fetal calf serum (GIBCO, Grand Island, NY), 10 mM CaCl₂, and 5 mM HEPES and then centrifuged at 50 g for 5 min. Finally, the resulting pellet was gently resuspended in the above solution and aliquoted into 35-mm poly-L-lysine-coated tissue culture dishes (1 ganglion/dish).

Isolation and patch clamp of dye-labeled sympathetic neurons. In order to obtain a large yield of fast blue-labeled sympathetic neurons for the patch-clamp studies, superior cervical ganglion (SCG) neurons from adult male Wistar rats (250-450 g) were labeled by a single intraperitoneal injection of fast blue (0.2 ml of 0.1% wt/vol, in normal saline) (Sigma Chemical) at least 1 wk before recording.

Whole cell voltage clamp. Isolated single (fast blue labeled or unlabeled) neurons were studied with the whole cell variant of the patch-clamp technique (13) using an Axoclamp 1-C patch-clamp amplifier (Axon Instruments). Patch electrodes were fabricated from N51A borosilicate capillary tubing (Garner Glass) using a model P 80/PC flaming brown micropipette puller (Shutter Instrument). Before use, pipettes were coated with Sylgard (Dow Corning) and fire-polished on a microforge (Narashige Scientific Instrument Laboratory) and had resistances of 1-2 M when filled with internal solution.

Data analysis. Renal function data were analyzed with a two-way analysis of variance with repeated measures on one factor, and Ca²⁺ current data were analyzed with a one-way analysis of variance. The least significant difference test was applied post hoc to locate significant differences within and between appropriate groups at the P  0.05 level. Data are reported as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Phase-contrast and fluorescence photomicrographs of an intact ganglion and a field of acutely isolated neurons are presented in Fig. 1. The fluorescence photomicrograph of a whole mount of the paravertebral ganglion T₁₂ isolated 1 wk after the injection of 10 µl of fast blue into the ipsilateral kidney (Fig. 1B) shows intensely fluorescent cells scattered throughout the ganglion. Figure 1C shows a phase-contrast photomicrograph of a field of acutely isolated neurons. The arrows indicate fluorescent efferent sympathetic neurons, which can be identified by epifluorescence optics in Fig. 1D.

View larger version (94K): [in this window] [in a new window]   Fig. 1.   Phase-contrast and fluorescence photomicrographs of acutely isolated renal efferent sympathetic neurons (RESN). A: phase-contrast photomicrograph of a whole paravertebral ganglion (PVG) isolated 1 wk after injection of fast blue into the left kidney. B: fluorescence photomicrograph of the same PVG shows intensely fluorescent cells scattered throughout the ganglion. C: phase-contrast photomicrograph of a field of acutely isolated PVG neurons. Fast blue-labeled neurons are indicated by arrows. D: fluorescence photomicrograph of the same field of neurons shows fluorescent fast blue-labeled RESNs.

The baseline MAP and heart rates in the hemorrhage group averaged 126 ± 4 mmHg and 369 ± 13 beats/min, respectively, values not different from those in the volume expansion group (124 ± 3 mmHg and 365 ± 12 beats/min). Similarly, there were no differences between the body weights (375 ± 15 vs. 388 ± 10 g), left kidney weights (1.34 ± 0.09 vs. 1.31 ± 0.09 g), or right kidney weights (1.33 ± 0.10 vs. 1.40 ± 0.10 g) of the hemorrhage and volume expansion groups, respectively. Hematocrit and plasma protein concentration were also similar in the two groups and averaged 50.0 ± 0.7% and 5.0 ± 0.0 g/dl in the hemorrhage group and 50.0 ± 0.9% and 4.8 ± 0.2 g/dl, respectively, in the volume expansion group.

Renal function measurements obtained from the rats before and after either 5% blood volume hemorrhage or 2.5% body weight volume expansion are presented in Table 1. Baseline renal function was not significantly different between the dye-injected (left) kidney and the control (right) kidney within either group. Indeed, there were no differences between the GFR, estimated renal plasma flow (C_PAH), urine flow, or urinary sodium excretion of the injected and noninjected kidneys in either group (Table 1). GFR ranged from 0.90 to 1.12 ml · min¹ · g¹ and C_PAH ranged from 3.68 to 3.91 ml · min¹ · g¹. These measured baseline renal excretory and hemodynamic variables are within the normal range for rats under similar experimental conditions (1, 25).

After 5% blood volume hemorrhage, MAP decreased from 126 ± 4 to 103 ± 5 mmHg (P < 0.05). This decrease was accompanied by slight reductions in renal hemodynamics and renal excretory function (Table 1) in both the dye-injected (left) and control kidneys (right); however, these changes did not reach statistical significance. MAP was unchanged after 2.5% body wt saline volume expansion (124 ± 3 vs. 121 ± 6 mmHg, NS). Similarly, both GFR and C_PAH were slightly elevated after volume expansion in the injected and control kidneys, but these changes did not reach statistical significance. However, volume expansion elicited marked increases in renal excretory function. Urine flow increased from 3.7 ± 0.9 to 13.0 ± 1.7 µl/min (P < 0.05) in the dye-injected kidney and from 4.8 ± 1.0 to 19.1 ± 2.7 µl/min (P < 0.05) in the contralateral control kidney. The magnitude of the volume expansion-induced increase in urine flow in the dye-injected kidney (9.3 ± 1.2 µl/min) was slightly but significantly smaller (P < 0.05) than in the contralateral control kidney (14.3 ± 2.0 µl/min). Urinary sodium excretion increased in both kidneys approximately ninefold from 0.33 ± 0.21 to 3.01 ± 0.51 µeq/min in the dye-injected kidney and from 0.43 ± 0.20 to 4.26 ± 0.82 µeq/min in control kidneys. Urinary potassium excretion tended to be elevated in both kidneys after volume expansion, but these changes did not reach statistical significance.

Ca²⁺ current modulation of labeled and unlabeled sympathetic neurons. For patch-clamp studies of Ca²⁺ currents, four groups of neurons were used. SCG neurons isolated from uninjected control rats were prepared for patch-clamp recording and either received no UV illumination (group 1) or were illuminated with UV for 1 min (group 2). SCG neurons isolated from rats that were labeled by a single intraperitoneal injection of fast blue 1 wk before the experiment were illuminated by UV for only 5 s, enough to position an identified neuron for recording, and then either received no further UV (group 3) or were illuminated for a further 1 min (group 4). Cell capacitance and series resistances estimated from the transient cell charging currents were not different between the four groups of neurons as follows: group 1 42.8 ± 2.0 pF, 3.1 ± 0.1 M, n = 41; group 2 44.0 ± 2.6 pF, 3.1 ± 0.1 M, n = 26; group 3 45.8 ± 3.5 pF, 3.4 ± 0.3 M, n = 18; and group 4 42.1 ± 1.9 pF, 3.1 ± 0.2 M, n = 28.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Ca2+ current amplitudes of labeled and unlabeled superior cervical ganglion (SCG) neurons before and after 1 min exposure to UV illumination. A: Ca2+ currents of a fast blue-labeled neuron before (left) and after (right) 1 min exposure to UV. Ca2+ currents were elicited by 10-ms depolarizing steps to the indicated potential from a holding potential of 80 mV, using an interpulse interval of 10 s. B: leak-subtracted current-voltage relationship before () and after () 1-min UV exposure. C: mean peak current amplitudes before (open bars) and after (filled bars) a 1-min exposure or time control period for the 4 groups of neurons. Nos. in parentheses, no. of neurons tested; vertical bars, SEs. See RESULTS for further description of 4 neuron groups.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Dose-response parameters for norepinephrine (NE)-induced Ca2+ current (ICa) inhibition in labeled and unlabeled SCG neurons after either 1-min UV exposure or a time control period. A: mean dose-response curves of NE-induced inhibition of the Ca2+ current in nonlabeled unexposed neurons () and labeled neurons exposed to UV for 1 min (). B: maximal NE-induced Ca2+ current inhibition for the 4 groups of neurons after 1 min exposure to UV or a 1-min time control period. Nos. in parentheses, no. of neurons tested; bars, SEs. C: half-maximal inhibiting concentration (Kd) of norepinephrine for the 4 groups derived from a nonlinear fit of the data to a single site binding curve. Nos. in parenthesis, no. of neurons tested; vertical bars, SD of the fit to all the data points from each group.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Ca2+ current facilitation and reinhibition in labeled and unlabeled SCG neurons after 1-min exposure to UV or a 1-min time control period. A: superimposed current traces and voltage protocol used to determine the reinhibition time constant that reflects G protein activity. B: percent Ca2+ current facilitation induced by a 25-ms pulse to 80 mV in labeled and unlabeled SCG neurons after 1-min UV exposure or a 1-min time control period. C: time constant of reinhibition derived from a single-exponential fit of the loss of facilitation with increasing hyperpolarizing intervals that precede the test pulse to 10 mV shown in A. B and C: nos. in parentheses, no. of neurons tested; bars, SEs.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although retrograde tracer studies have previously been used to identify cells of origin of the sympathetic renal innervation (23, 27), the electrophysiological properties of these cell bodies have yet to be characterized. To evaluate the role of putative electrophysiological derangements in renal efferent cell bodies during the development of hypertension, it is critical that the identification process does not alter renal function, which may in turn alter the temporal pattern and/or severity of the hypertension (5, 14) or the function of the innervating neurons that would be expected to interrupt sympathetic control of the kidney. There are several agents available that can be used to label neurons. In pilot studies we compared blue and green microspheres, fluorogold, and fast blue for their ability to label PVG neurons. In these experiments, only fast blue produced intense labeling of PVG neurons. Rhodamine proved to be the agent of choice over fluorogold, microspheres, and 3,3-diethyloxadicarbocyanine iodide for the identification of cardiac motoneurons (17). It is possible that rhodamine may be less damaging to neurons than fast blue in that UV is not required to excite the dye. However, the biophysical properties of labeled and nonlabeled neurons were not compared in that study (17). We therefore chose to use fast blue in this study because it has previously been shown to be highly effective for the identification of renal efferent sympathetic neurons (24).

The preferred method of labeling large numbers of renal sympathetic neurons is to cut the renal nerves close to their point of entry into the kidney and soak the cut ends in the tracer (27). This procedure is obviated in studies of neuronal function because the axotomy would induce profound alterations of electrophysiological behavior (18-20). An alternative labeling method that preserves axon integrity was used in this study. This method, however, employs multiple injections into the renal cortex to allow uptake of the dye by the sympathetic nerve terminals and transport to the cell bodies in the paravertebral ganglia (23). It is possible that these multiple microinjections into the kidney cortex could cause renal derangements, which might compromise physiological studies. However, the lack of visible damage of dye-injected kidneys suggests that this labeling procedure does not produce profound renal damage. More importantly, the similarity of baseline renal hemodynamics and excretory function in the dye-injected and control kidneys, which are within the normal range for anesthetized rats, provides further direct evidence that renal function is not compromised by the dye-labeling procedure.

A second possible technical complication of this procedure could involve dye-induced damage to the renal nerves. In an effort to address this issue, we evaluated the renal function responses to 5% blood volume hemorrhage. This maneuver has been shown to result in substantial changes in renal hemodynamics and excretory function, which are mediated in large part by activation of renal efferent sympathetic fibers (5). In the present study renal hemodynamics and excretory function of the dye-injected and contralateral control kidneys after 5% blood volume hemorrhage displayed a similar trend to decreased values, which argues against substantial renal nerve damage at the time the renal function experiments were undertaken (1 wk after dye injection). However, the hemorrhage-induced changes did not reach statistical significance in either control or dye-injected kidneys, and therefore altered neuronal function cannot be ruled out by these experiments. Although it is not clear why hemorrhage failed to elicit marked reductions in urine flow and sodium excretion, it is possible that the renal sympathetic nerves were maximally activated and exerting maximal antidiuretic and antinatriuretic effects under control conditions, and thus hemorrhage failed to elicit a further increase in efferent sympathetic nerve activity. In this regard, it should be recognized that the present experiments were performed in anesthetized surgically stressed rats, which have been shown to have an increased activity of the sympathetic nervous system compared with unanesthetized rats. Furthermore, the basal hematocrits indicate that the rats were hydropenic, which would further activate the sympathetic nervous system. Thus it is possible that the effects of hemorrhage to reduce urine flow and sodium excretion were masked by the already low basal levels of urine flow and sodium excretion resulting from the elevated level of activity of the sympathetic nervous system induced by anesthesia, surgical stress, and altered volume status. Regardless, the finding that basal levels of urine flow and sodium excretion in the dye-injected kidneys were equally as low as those in the control kidneys suggests that the dye-injected kidneys did not have an impaired ability to retain salt and water under conditions in which the level of activity of the sympathetic nervous system was presumably markedly increased. In essence, the present findings suggest that the renal nerves were still capable of eliciting salt and water retention by the dye-injected kidney. Because the patch-clamp studies of SCG and renal efferent sympathetic neurons displayed no differences in the biophysical properties (4) or the neuromodulation of Ca²⁺ currents between fast blue-labeled and unlabeled neurons, it is unlikely that the dye-labeling procedure produces significant damage to the renal innervation. The possibility remains, however, that renal nerve or kidney function may be transiently altered sometime during the 1-wk recovery period between the dye injection and the time of the experiments. Further studies are required to address this issue.

In an effort to evaluate whether the dye-labeling procedure compromised the ability of the renal nerves to decrease activity and therefore results in an inappropriate retention of salt and water, renal function was examined after a 2.5% body wt saline volume expansion. In the present study, volume expansion increased urine flow by 2.5- to 3-fold and sodium excretion by 8- to 9-fold in both the dye-injected and control kidneys, without influencing renal hemodynamic function. However, the absolute magnitude of the volume expansion-induced increment in urine flow was slightly but significantly smaller in the dye-injected kidneys, indicating that these kidneys exhibited an impaired ability to respond appropriately to acute volume expansion. Nevertheless, the finding that volume expansion elicited marked increases in both urine flow and sodium excretion in the dye-injected kidneys clearly indicates that the dye injection procedure does not substantively interfere with the renal response to volume expansion.

The present observation that the membrane capacitance and series resistance values of labeled and unlabeled neurons with and without UV exposure were not different suggests that it is unlikely that the neuronal geometry or size was influenced by the dye label or UV exposure during identification. Because Ca²⁺ current amplitudes of SCG neurons or PVG neurons were not affected by the presence of the dye label and/or the UV exposure required to identify and position a neuron for recording, it appears that the procedure has no deleterious effect on neuronal function. A more rigorous test of the effects of the dye label or the UV exposure was provided by comparing neuromodulation of Ca²⁺ currents in labeled and unlabeled neurons. Intact norepinephrine-induced neuromodulation of Ca²⁺ currents requires functional receptors and channels and preservation of the G protein signaling pathway. The observation that the sensitivity (K_d) for norepinephrine and maximal inhibition were not different between the four groups strongly suggests that the norepinephrine receptors were not affected by the presence of the dye label or UV exposure. Moreover, because the percent facilitation and time constant of recovery of inhibition recorded in 10 µM norepinephrine were similar between the four groups, it seems unlikely that the dye or UV exposure compromised G protein signaling in these neurons. We chose to examine Ca²⁺ currents and their neuromodulation because Ca²⁺ currents play a vital part in the neuronal control of renal function, by virtue of their role in transmitter release from sympathetic nerve terminals. At the terminals, norepinephrine release can be inhibited by modulation of the Ca²⁺ current via ₂-adrenoceptor activation. In addition, the effects of norepinephrine can be monitored as current inhibition and by depolarization-induced facilitation, the latter being a specific indicator of N-channel inhibition. Thus we expected assessment of Ca²⁺ currents and their modulation would be a sensitive indicator of neuronal function. Although there were no differences detected in any of the Ca²⁺ current parameters tested, the possibility remains that other conductance systems could be affected by this dye-labeling procedure that might alter neuronal function. Further experiments would be required to address this issue.

Perspectives