INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE PONTINE NORADRENERGIC A5 region receives extensive afferent input from major subcortical areas involved in cardiovascular/respiratory control, including the nucleus of the solitary tract, rostral and caudal ventrolateral medulla, parabrachial and Kolliker-Fuse nuclei of the pons, and paraventricular nuclei of the hypothalamus (4, 25, 32). Virtually the entire A5 cell group projects to the thoracic spinal cord (5, 27). In addition, many of the A5 cells also send axonal processes to cardiovascular nuclei of the hypothalamus, midbrain, and brain stem (4). Projections from A5 cells comprise ~38% of all descending monosynaptic inputs to the renal sympathetic preganglionic neurons (SPNs), forming along with projections from the rostral ventrolateral medulla (47%) two major descending inputs to these neurons in the rabbit (9).

Although the importance of the rostral ventrolateral medulla in circulatory control has been established, the functional significance of the A5 region is less certain. The A5 region has a little tonic influence on resting blood pressure in anesthetized animals (23, 24). Electrically stimulating the A5 cell group increased blood pressure (26, 35), and many of the A5 cells were excited by chemoreceptor activation (12, 15, 16) or baroreceptor inhibition (1, 14, 18), suggesting that they may provide an excitatory input to SPNs. However, microinjections of excitatory amino acids into the A5 region mostly decreased blood pressure, suggesting that the A5 cell group may exert sympathoinhibitory action (8, 11, 17, 34).

The most likely explanation for these variable responses is that both sympathoinhibitory and sympathoexcitatory pathways can be activated in the A5 region. The findings that stimulating the A5 region evoked the opposite vasomotor responses in regional vascular beds (8, 11, 17, 34) provide one possible mechanism, indicating that these sympathoinhibitory and sympathoexcitatory pathways may be topographically separated within the autonomic nervous system. As a result, the overall change in blood pressure after stimulating the A5 region would be determined by the balance between opposing regional responses, which, in turn, may vary because of the difference in experimental conditions and, particularly, in anesthetic used (17). An alternative possibility is that the A5 cells do not simply form an excitatory or inhibitory input to a particular subgroup of SPNs, but rather serve to modulate activity of these neurons. In this case, taking into account the extensive afferent inputs, the influence of the A5 region on sympathetic outflow may depend on their activity and may therefore be altered in response to various external and internal stimuli.

In the present study, we examined the influence of the A5 region on renal sympathetic nerve activity (RSNA) and blood pressure at resting condition and during baroreceptor or chemoreceptor stimulation. We also used simultaneous baro- and chemoreceptor stimulation to estimate whether alterations in the afferent stimuli, which are known to affect neural activity in the A5 region (1, 15), would also alter its action on the sympathetic outflow to the kidney. Because anesthesia can affect functions of the A5 region, in this study we determined the circulatory effects of inhibiting neural activity in the region with local microinjections of muscimol in conscious, chronically instrumented rabbits.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Implantation of guide cannulas. The experiments were performed in eight conscious rabbits, weighing 2.4-2.9 kg, of either sex and bred and housed at the Baker Medical Research Institute in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. At least 3 wk before the experiments, rabbits were implanted with metal guide cannulas for microinjection into the noradrenergic A5 region, which we localized earlier using a catecholamine histofluorescence technique (2).

Rabbits were premedicated with 4 mg of dexamethasone (Dexason, Troy Laboratories) to prevent inflammation around the guide cannula. Anesthesia was induced using propofol (Diprivan, 1 mg/kg iv; Zeneca) after which the rabbits were intubated and anesthesia was maintained with halothane (Fluothane, Zeneca). The animal was placed in a stereotaxic frame with lambda and bregma parallel to the horizontal stereotaxic arm. The animal's head was leveled medially/laterally by matching height measurements 4 mm lateral of lambda. The angle of the stereotaxic drill was set at 24° from the vertical with the tip pointing rostrally, and the coordinates for lambda were recorded. Holes (1.0 mm in diameter) were then drilled 5.6 mm posterior to lambda and 3.1 mm bilaterally from the central fissure. A 40-mm length of 0.2-mm diameter stainless steel wire held in a stereotaxic clamp was used to take bilateral measurements from lambda to the base of the skull via the holes to confirm that the animal was level in the medial/lateral plane. This measurement was also used to position the end of the guide cannula 6.5 mm above the base of the skull, which conforms to 4.5 mm above the intended A5 injection site. A 22-gauge, 22.5-mm-long stainless steel guide cannula was then mounted in a stereotaxic clamp and inserted down to 6.5 mm above the base of the skull. Dental cement was then placed around the guide cannula and retained in place by three cheese head screws. Guide dummies were inserted 22.5 mm into the guides to prevent material from entering the cannula.

Implantation of renal nerve electrodes. One week before the experiments, a bipolar renal nerve electrode for recording RSNA was implanted under halothane anesthesia according to the method of Dorward and colleagues (10). With the use of a dissecting microscope, the left kidney was exposed by the retroperitoneal approach and the renal nerve was identified and placed inside a coiled pair of electrodes. The nerve and recording electrode assembly was insulated from the surrounding tissue by SilGel 604 (Wacker-Chemie). The other end of the electrode was tunneled under the skin for later retrieval on the day of the experiment. The incision was sutured, and the rabbits were allowed to recover for 7 days before the first experiment.

Blood pressure and RSNA measurement. On the day of the experiment, the animal was placed in a standard rabbit box (dimensions 15 × 40 × 18 cm, width × length × height). Under local anesthesia (Lignocaine HCl 1%, Delta West), the central ear artery and marginal ear vein were catheterized and the plug of the renal nerve electrode was retrieved from under the skin and connected for measurements of RSNA. Pulsatile arterial blood pressure was measured with a Statham 23Dc pressure transducer. Sympathetic nerve activity was amplified, filtered between 50 and 5,000 Hz, and full-wave rectified and integrated using a low-pass filter with a 20-ms time constant. The integrated neurogram obtained using this low time constant allowed us to analyze oscillations of synchronized bursts of sympathetic activity and served as the input signal for all subsequent computer analysis (Fig. 1). The integrated RSNA signal and pulsatile blood pressure were continuously monitored throughout the experiment and were sampled at 1,000 Hz using an analog-to-digital data acquisition card (National Instruments). Mean arterial pressure (MAP) and heart rate (HR) were calculated online using a program written in the LabVIEW graphical programming language, which detected beat-to-beat systolic and diastolic pressures as well as R-R interval. In regard to RSNA, three parameters were calculated online from the integrated neurogram: 1) the average RSNA voltage per 2-s period, termed total RSNA, 2) amplitude of each synchronized burst of sympathetic activity, termed RSNA amplitude, and 3) the number of sympathetic bursts occurring per second, termed RSNA frequency. The bursts of sympathetic activity were detected using a dedicated program written in the LabVIEW graphical programming language. The program uses a series of fast and slow filters to detect changes in the voltage of the signal from increasing to decreasing levels. Providing these changes were above an operator-defined threshold, they were classified as bursts of RSNA. The threshold was set to 10-15% of the average maximum burst height, as discussed previously (30). Because voltage recorded from RSNA electrodes varies considerably among animals, the values were normalized to the upper plateau of the baroreflex curve measured during the first normoxia period, which was taken to equal 100 normalized units.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Original and integrated neurograms of renal sympathetic nerve activity (RSNA) from 1 conscious rabbit under normoxia (A) and hypoxia (B). Marks below integrated RSNA indicate synchronized bursts of RSNA detected online by sympathetic peak detection program. Average voltage from integrated RSNA per 2-s period was termed total RSNA. AP, arterial pressure.

Experimental protocol. The effects of muscimol (Sigma Chemicals) or vehicle Ringer solution (Baxter) on baroreceptor and/or chemoreceptor stimulation were examined in two separate experiments 2 days apart with the order of treatment randomized. At the beginning of the experiment, the rabbit box was placed inside a sealable plastic box (volume 30 liters) with an inlet and outlet valve and, after a 30-min control period, the box was sealed and air was perfused through the box at a rate of 10 l/min for a 20-min period. After a 10-min adjustment period, baroreflex curves were performed. The air was then replaced for another 20 min by the hypoxic gas mixture of 10% O₂ + 3% CO₂, and the protocol was repeated. The exposures to normoxic and hypoxic gas mixtures were performed three times, separated by two 30-min reequilibration periods. Vehicle and 175 and 875 pmol of muscimol were injected bilaterally into the A5 region 5 min before the first, second, and third exposures to gas mixtures, respectively. On the time-control day, all exposures to the gas mixtures were repeated, but vehicle was injected on all three occasions.

The injections into the A5 region were made through a 30-gauge stainless steel blunted needle connected via polyethylene PE-10 tubing to a 100-µl Hamilton syringe driven by a micromanipulator. The injection volume (300 nl) was controlled by measuring the movement of a small air bubble along a graduated scale and was delivered over a 1-min period.

To induce hypoxia, we used an isocapnic hypoxic mixture of 10% O₂ + 3% CO₂. The additional CO₂ was used to prevent hypoxia-associated hypocapnia, which has an inhibitory effect on the central chemoreceptors (6). We showed previously, using the same experimental procedure in conscious rabbits, that this hypoxic mixture decreased arterial PO₂ to 48 mmHg while maintaining arterial PCO₂ at normal levels (31).

On completion of the experiment, animals were deeply anesthetized with pentobarbital sodium and the injection sites were marked with 300 nl of 2% Pontamine sky blue solution. Brains were removed, frozen, and sectioned at 30 µm. Each tenth section from the caudal pole of the facial nucleus to the rostral pole of the superior olive was slide mounted and examined under the microscope for distribution of Pontamine sky blue and the course of the cannula track.

MAP-RSNA and MAP-HR baroreflex relationship. The baroreflex was assessed by a single slow ramp rise and fall in MAP produced by intravenous infusions of phenylephrine hydrochloride (0.5 mg/ml; Sigma) and sodium nitroprusside (1.0 mg/ml; Fluka), respectively. Injections lasted ~1 min, and the rate of change in MAP was controlled between 0.5 and 1 mmHg/s. MAP, RSNA, and HR from individual rabbits were averaged over 2-s intervals and fitted into a sigmoid logistic function to produce RSNA-MAP and HR-MAP curves. We used a nonlinear regression program using the Marquardt-Levenberg method to fit the following equation
where
defines a transition function varying smoothly between 0 and 1 and centered about the median blood pressure, and the mean curvature (_f) is given by
where P1 is lower plateau, which is a calculated minimum RSNA or HR; P2 is range between upper plateau, which is a calculated maximum activation, and lower plateau; P3 and P5 are range-independent measures of slope (parameters defining the curvature); and P4 is MAP at half the reflex range. The two curvature parameters allowed for a nonsymmetrical fit of the data. The average range-dependent gain of the curve (G), which is given by G = P2 × (P3+P5)/9.12, indicates the slope between the two inflection points of the curve.

Statistical analysis. Values are expressed as means ± SE. A two-factor repeated-measures ANOVA was used to determine treatment effects for each hemodynamic and sympathetic parameter measured. The between-animal sum of squares (SS) and the treatment SS were removed from the total SS to obtain the residual SS (33). The latter was used to calculate the average within-animal SE, indicating the variability within animals. For each parameter of the sigmoidal curve, comparisons were made by partitioning the treatment SS into orthogonal contrasts, each with one degree of freedom. The treatment SS were partitioned into the experiment day SS and time-control day SS. At each day, the treatment SS were partitioned into the normoxia SS and hypoxia SS, with further partitioning of each of them into the SS for treatment with muscimol and vehicle. Contrasts were considered significant and the null hypothesis was rejected when P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of muscimol on resting RSNA during normoxia and hypoxia. Pretreatment with muscimol into the A5 region did not affect resting sympathetic or hemodynamic parameters under normoxia (for each parameter F_1,77 < 2.0, P > 0.05; Table 1). Hypoxia gradually increased resting total RSNA (Fig. 2), starting 3-4 min after beginning and reaching a stable plateau within 8-10 min (for all sessions F_1,77 = 47.3, P < 0.001). Total RSNA was increased by 63 ± 16% compared with normoxia levels (Fig. 3). This change was mediated by an increase in RSNA amplitude (F_1,77 = 57.7, P < 0.001), but not RSNA frequency (F_1,77 = 0.1, P > 0.05). In response to hypoxia, HR tended to decrease compared with normoxia levels (F_1,77 = 3.0, P = 0.09), whereas MAP was not changed (F_1,77 = 1.3, P > 0.05). In the presence of hypoxia, pretreatment with muscimol into the A5 region increased total RSNA compared with vehicle injection (for both doses F_1,77 = 11.8, P < 0.001; Table 1). Total RSNA was increased by 37 ± 10 and 47 ± 10% for 175 and 875 pmol of muscimol, respectively (Fig. 3). The changes in total RSNA evoked by muscimol under hypoxia were mediated via increases in RSNA amplitude (F_1,77 = 5.3, P < 0.05), whereas RSNA frequency as well as MAP and HR remained unaffected (for each parameter F_1,77 < 0.2, P > 0.05).

View larger version (24K): [in this window] [in a new window]   Fig. 2.   An original recording from 1 conscious rabbit showing effect of hypoxia on mean arterial pressure (MAP), total RSNA, amplitude of RSNA, and frequency of RSNA after pretreatment with vehicle (A) or 875 pmol of muscimol (B) into A5 region. Amplitude of each synchronized burst was taken as individual RSNA burst peak height, and number of bursts occurring per second was defined as RSNA frequency. Control before hypoxia is shown before dotted line. All values are 2-s averages of each parameter.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Resting sympathetic and hemodynamic parameters after injection of 175 pmol (Mus1) or 875 pmol (Mus2) of muscimol or vehicle (Veh) into A5 region during normoxia () and hypoxia () in conscious rabbits. Left, time-control day; right, experiment day. nu, Normalized units. Error bars are average SE calculated from ANOVA, indicating variation within animals (n = 8). * P < 0.001, hypoxia vs. normoxia; # P < 0.001, muscimol (both doses) vs. vehicle. HR, heart rate in beats/min (bpm).

On the time-control day, resting sympathetic and hemodynamic parameters did not differ between three normoxic sessions (for each parameter F_1,77 < 0.8, P > 0.05). Under hypoxia, total RSNA (F_1,77 = 8.1, P < 0.01) and RSNA amplitude (F_1,77 = 7.7, P < 0.01) were increased compared with normoxia, whereas RSNA frequency, MAP, and HR remained unchanged (Fig. 3). All parameters measured were not different between hypoxic sessions (for each parameter F_1,77 < 0.6, P > 0.05).

Effects of muscimol on the baroreceptor reflex during normoxia. Pretreatment with muscimol into the A5 region increased the range of the total RSNA baroreflex compared with vehicle injection (for both doses F_1,77 = 6.2, P < 0.05; Fig. 4). The range was increased by 22 ± 6 and 33 ± 6% for 175 and 875 pmol of muscimol, respectively. By contrast, the gain of the total RSNA baroreflex was not changed after muscimol injections (for both doses F_1,77 = 0.1, P > 0.05) because of a tendency for a decrease in the curvature (Fig. 5).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Total RSNA baroreflex curves in conscious rabbits during normoxia (left) and hypoxia (right) after pretreatment with muscimol or vehicle into A5 region on experiment (A) and time-control (B) day. Circles and squares on curves represent resting values.  and dotted line, vehicle;  and dashed line, 175 pmol of muscimol (2nd vehicle injection on time-control day);  and solid line, 875 pmol of muscimol (3rd vehicle injection on time-control day). Error bars as for Fig. 1. * P < 0.05, muscimol vs. vehicle.

View larger version (42K): [in this window] [in a new window]   Fig. 5.   Gain of total RSNA baroreflex in conscious rabbits during normoxia (open bar) and hypoxia (hatched bar) after injections of 175 pmol (Mus1) or 875 pmol (Mus2) of muscimol or vehicle (Veh) into A5 region. Error bars as for Fig. 1. * P < 0.05, muscimol vs. vehicle.

The changes in the range of total RSNA baroreflex after muscimol were mediated by increases in the range of the RSNA amplitude baroreflex (F_1,77 = 4.6, P < 0.05), whereas the range of the RSNA frequency baroreflex was unaffected (F_1,77 = 0.1, P > 0.05; Fig. 6). The parameters of the HR baroreflex were not changed by muscimol injections under normoxia (for each parameter F_1,77 < 0.9, P > 0.05; Table 1).

View larger version (22K): [in this window] [in a new window]   Fig. 6.   RSNA amplitude (A) and RSNA frequency (B) baroreflex curves in conscious rabbits during normoxia (left) and hypoxia (right) after injections of muscimol or vehicle into A5 region. Circles and squares on curves represent resting values.  and dotted line, vehicle;  and dashed line, 175 pmol of muscimol;  and solid line, 875 pmol of muscimol. Error bars as for Fig. 1. * P < 0.05, muscimol vs. vehicle.

On the time-control day, the range and gain of the total RSNA baroreflex were similar between three normoxic sessions (Figs. 4 and 5). Likewise, for all other sympathetic and hemodynamic parameters measured, the range and gain were not different between normoxic sessions (for each parameter F_1,77 < 0.9, P > 0.05, data not shown).

Effects of muscimol on the baroreceptor reflex during hypoxia. Hypoxia increased the range of the total RSNA baroreflex compared with normoxia levels (for all sessions F_1,77 = 32.0, P < 0.001; Fig. 4). After vehicle pretreatment, the range was increased by 53 ± 9%. Hypoxia also increased the gain of the total RSNA baroreflex (for all sessions F_1,77 = 13.9, P < 0.001). After vehicle pretreatment, the gain was increased by 89 ± 18% compared with normoxia (Fig. 5). These changes were mediated by corresponding increases in the range (F_1,77 = 50.5, P < 0.001) and gain (F_1,77 = 6.8, P < 0.05) of the RSNA amplitude baroreflex, whereas the RSNA frequency baroreflex was not affected by hypoxia (Figs. 6 and 7). Hypoxia decreased the upper plateau (F_1,77 = 6.6, P < 0.05) and range (F_1,77 = 14.0, P < 0.001) of the HR baroreflex, but did not change the gain of this reflex (Table 1).

View larger version (43K): [in this window] [in a new window]   Fig. 7.   Gain of RSNA amplitude and RSNA frequency baroreflexes in conscious rabbits during normoxia (open bar) and hypoxia (hatched bar) after injections of 175 pmol (Mus1) or 875 pmol (Mus2) of muscimol or vehicle (Veh). Error bars and symbols as for Fig. 1. * P < 0.05, muscimol vs. vehicle.

Pretreatment with muscimol into the A5 region did not alter the range of the total RSNA baroreflex under hypoxia (for each parameter F_1,77 < 1.1, P > 0.05; Fig. 4). However, under this condition, the baroreflex gain was reduced by muscimol (for both doses F_1,77 = 5.4, P < 0.05; Fig. 5). Injections of muscimol decreased the gain of the total RSNA baroreflex by 19 ± 9 and 34 ± 9% at doses of 175 or 875 pmol, respectively. These effects of muscimol were due to decreases in the curvature (F_1,77 = 5.5, P < 0.05) of the total RSNA baroreflex (Fig. 5).

The changes in the gain and curvature of the total RSNA baroreflex after muscimol treatment were mediated by decreases in the gain (F_1,77 = 22.5, P < 0.001) and curvature (F_1,77 = 9.2, P < 0.01) of the RSNA amplitude baroreflex, whereas these parameters of the RSNA frequency baroreflex were not affected (Fig. 7). Inhibition of the A5 region with muscimol did not affect the changes in the HR baroreflex curves under hypoxia (for each parameter F_1,77 < 2.4, P > 0.05; Table 1).

On the time-control day, hypoxia significantly increased the range (F_1,77 = 34.2, P < 0.001) and gain (F_1,77 = 13.8, P < 0.001) of the total RSNA baroreflex compared with normoxia (Figs. 4 and 6). These responses were mediated by increases in the corresponding parameters of the RSNA amplitude baroreflex, whereas the RSNA frequency baroreflex was unchanged under hypoxia (data not shown). All baroreflex parameters measured were not different between three hypoxic sessions (for each parameter F_1,77 < 1.6, P > 0.05).

Histological verification of the injection sites revealed that diffusion of 300 nl of Pontamine sky blue was confined by the area surrounding the A5 region, with the center placed lateral to the superior olive and just caudal to posterior side of the facial nerve (Fig. 8). There was virtually no staining caudal to the posterior pole of the facial nucleus or rostral to the anterior side of the facial nerve band.

View larger version (40K): [in this window] [in a new window]   Fig. 8.   Schematic representation of coronal sections at 5, 6, and 7 mm rostral to obex. , Stimulation sites in left and right A5 region as revealed by microinjections of 300 nl of Pontamine sky blue in 8 rabbits. FN, facial nuclei; 7n, facial nerve; SO, superior olive.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The major finding of the present study was that inhibiting neural activity in the pontine A5 region increased renal sympathetic responses to baroreceptor unloading and chemoreceptor stimulation but produced little effect on resting RSNA. The muscimol-induced increase in the RSNA baroreflex range, which provides an estimate of the excitatory capacity of the motoneuron pool when tonic baroreceptor inhibition is reduced to low levels (10), suggests that under normoxia, the A5 region serves to limit the reflex excitation of the renal SPNs caused by baroreceptor unloading in conscious rabbits. This increase in the baroreflex range was not seen in our time-control experiments with vehicle pretreatment, indicating a specific effect of the agonist rather than a time-related effect. Because muscimol is a neuronal hyperpolarizing agent, its sympathoexcitatory effect was presumably mediated via a depressing action on an inhibitory input to the renal motoneuron pool. Taking into account that noradrenergic neurons constitute at least 90% of all spinally projecting cells present in the A5 region (5), it is possible that muscimol directly inhibited these presympathetic neurons, which may have exerted an inhibitory action directly on the renal SPNs.

By contrast, inhibiting the A5 region with muscimol did not affect the range of the RSNA baroreflex under hypoxia. It is unlikely that the excitatory capacity of the renal motoneuron pool reached the upper limit during simultaneous hypotension and hypoxia, because in previous studies from this laboratory, activating the bulbospinal sympathoexcitatory pathways or inhibiting the bulbospinal sympathoinhibitory pathways increased the RSNA baroreflex range under both normoxia and hypoxia (the same gas mixture) in conscious rabbits (3, 13). The lack of effects of muscimol in the present study indicates that the inhibitory pathway from the A5 region to the renal SPNs was not functionally active during simultaneous hypotension and hypoxia. However, under hypoxia, muscimol decreased the baroreflex gain via a range-independent mechanism, indicating that under this condition the A5 region served to facilitate transmission through the renal sympathetic baroreflex pathways. This modulatory action also depended on chemoreceptor activity, because the baroreflex sensitivity was not affected by muscimol under normoxia. Thus our results indicate that the A5 region may differentially influence the RSNA baroreflex parameters according to activity of the chemoreceptor input.

In this study, pretreatment with muscimol in the A5 region increased resting RSNA under hypoxia, indicating that the A5 region also inhibited the renal sympathoexcitatory response to chemoreceptor stimulation. This effect of muscimol is in contrast to the decrease in the pressor and splanchnic sympathetic response to hypoxia observed after microinjections of muscimol into the A5 area in vagotomized, artificially denervated anesthetized rats (24). The difference between our study and previous work may be due to a differential influence of the A5 region on splanchnic and renal sympathetic outflow (34). Furthermore, renal sympathetic responses to hypoxia were shown to be qualitatively different from those observed from the splanchnic nerve in conscious rabbits (20). Additionally, hypoxic regimens were different between studies. Although sympathetic responses to mild or moderate hypoxia are primarily mediated via stimulating the peripheral chemoreceptors in rabbits (19), secondary autonomic responses to respiratory changes or to the direct vasodilator actions of hypoxia, which are likely to be present in our study, would alter the afferent input to the A5 region. This contrasts the relatively pure stimulus of short-term (4-12 s) N₂ inhalation used in combination with extensive deafferentation as used in the previous study. Furthermore, the contribution of remaining baroreceptors to the sympathetic response to hypoxia was also likely to be different between studies, because in our experiments, blood pressure did not change during hypoxia. Finally, anesthesia used in the previous study might also affect the A5 function in the cardiovascular control (17). The difference between studies may further indicate that the A5 region does not simply exert a sympathoexcitatory or sympathoinhibitory action, but rather modulates sympathetic activity according to specific demands of the cardiovascular regulatory system.

In our study, inhibiting the A5 region with muscimol did not affect resting RSNA under normoxia. Similarly, resting splanchnic sympathetic activity and blood pressure were also unaffected by muscimol injections into the A5 region in anesthetized rats (24). Thus it is likely that the A5 region specifically modulates sympathetic reflexes but plays little role in the moment-to-moment regulation of blood pressure.

In our study, we did not distinguish between the noradrenergic and other cells in the A5 region. However, the noradrenergic A5 neurons most probably play a major role in controlling RSNA, because virtually all A5 cells that synapse on the renal SPNs also contain tyrosine hydroxylase in the rabbit (9). Notably, modulatory actions of norepinephrine in the central nervous system (CNS), and particularly in the spinal cord (see Ref. 7 for review), may provide a plausible pharmacological mechanism explaining opposing influences of the A5 region on the activity of SPNs. Nevertheless, the involvement of nonnoradrenergic pathways from the A5 region in control of RSNA cannot be excluded.

Although the involvement of the A5 region in controlling sympathetic activity has been investigated previously, the role of the region in regulating the pattern of sympathetic bursts has not been considered. There is accumulating evidence that the amplitude and frequency of the (renal) sympathetic bursts may be under separate control and may provide differential effects on hemodynamics and, particularly, on the renal excretory function (for review, see Ref. 28). The present study indicates that the changes in RSNA evoked by inhibiting the A5 region with muscimol were primarily mediated via changing the amplitude of sympathetic bursts. Because the main frequency rhythm of the sympathetic bursts is coupled with the cardiac cycle under different conditions in conscious rabbits (29), a change in HR can affect the burst frequency. However, in our study, neither HR nor RSNA frequency was altered by muscimol, indicating that, under experimental conditions, inhibiting the A5 region did not affect the generation of bursts within the CNS.

In conclusion, our results suggest that the pontine A5 region plays an important role in circulatory control by limiting the reflex renal sympathetic responses to baroreceptor unloading or chemoreceptor stimulation but has only a little tonic influence on RSNA in conscious, chronically instrumented rabbits. The different changes in the baroreflex range and gain caused by muscimol in the presence of normoxia or hypoxia indicate that functions of the A5 region depend highly on its afferent input and can be altered according to demands of the cardiovascular regulatory system.

Perspectives