INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE PARABRACHIAL NUCLEUS (PBN) is located in the dorsal lateral pons and has been suggested to be involved in the descending modulation of the arterial baroreflex. Parabrachial involvement in baroreflex function was first demonstrated in the early 1980s when it was shown that heart rate responses to electrical stimulation of the carotid sinus nerve were attenuated by simultaneous electrical stimulation of the PBN in the anesthetized cat (20). More recently, it has been demonstrated in rats that either electrical stimulation or electrolytic lesion of the PBN can significantly alter heart rate responses to increases in arterial pressure (13, 15). Because electrical stimulation of the PBN is primarily sympathoexcitatory, the PBN may modulate baroreflex control of heart rate by increasing central sympathetic drive to the heart. However, most stimulation studies have demonstrated only a modest increase in heart rate during activation of the PBN alone, suggesting that the PBN may modulate baroreflex function centrally. In fact, there is evidence in cats that electrical stimulation of the PBN can directly inhibit the responses of neurons in the nucleus of the solitary tract (NTS) to baroreceptor stimulation (6), suggesting that PBN stimulation may block vagally mediated slowing of heart rate at some site in the hindbrain baroreflex pathway.

There are several important limitations to these previous studies. First, they have focused almost exclusively on the role of the PBN in baroreflex regulation of heart rate; the influence of the PBN on baroreflex control of sympathetic nerve activity has not been investigated. Second, they have relied on electrical stimulation as the primary method to activate this nucleus. Because electrical stimulation activates both cell bodies and fibers of passage, the observed effects might have resulted from activation of fibers of passage rather than PBN neurons. Finally, the previous studies have primarily focused on establishing the effect of the medial PBN on baroreflex function. More recent work suggests an important role for the lateral PBN (LPBN) in cardiovascular autonomic control (4, 8, 19, 22, 23, 29). Thus the present study was undertaken to test the hypothesis that neurons in the LPBN have a descending modulatory influence on baroreflex control of sympathetic nerve activity. In the first group of studies, we investigated the effect of chemical versus electrical activation of the LPBN on baroreflex control of arterial pressure and sympathetic nerve activity. In the second group of studies we examined the effect of chemical blockade of the LPBN on baroreflex control of arterial pressure and sympathetic nerve activity to determine whether the LPBN tonically influences baroreflex control of sympathetic nerve activity. Preliminary data from these experiments have been presented in abstract form (7).

	METHODS

Top Abstract Introduction Methods Results Discussion References

General preparation. All experimental procedures were approved by the Animal Care and Use Committee at the University of Iowa. Experiments were performed on urethan (1.2-1.5 g/kg ip)-anesthetized male Sprague-Dawley rats (325-430 g). Animals were instrumented with arterial and venous femoral catheters for the monitoring of arterial pressure (P10EZ transducer connected to a Gould TA240S chart recorder) and the administration of supplemental anesthesia (0.1-0.2 g/kg urethan iv) or fluids, respectively. A tracheotomy was performed, and the animals were mechanically ventilated and paralyzed (Flaxedil, 4 mg/kg iv). Blood gases were monitored and kept within normal ranges by changing ventilation rate and/or volume. The renal nerve was exposed retroperitoneally and isolated. The animals were placed in the prone position in a stereotaxic head holder, and a small hole was drilled in the skull near the interaural line, just lateral to the midline. The left aortic depressor nerve (ADN) was visualized from a lateral approach, identified by its junction with the superior laryngeal nerve, and isolated from the vagus nerve. The ADN was placed on a bipolar silver wire stimulating electrode and covered with a mixture of mineral oil and silicone to prevent drying of the nerve.

Recording and stimulation techniques. Renal sympathetic nerve activity (RSNA) was recorded from an intact branch of the renal nerve placed on a bipolar recording electrode. The nerve and electrode were covered in a pool of mineral oil. RSNA was amplified 5,000-10,000 times and band-pass filtered (between 0.1 and 1 kHz) with a Grass P511 AC amplifier with a high-impedance probe. The amplified signal was fed to a window discriminator (WPI, model 121), and the output of the window discriminator was displayed on an oscilloscope to monitor the signal. The window discriminator issued a 5-V pulse every time the nerve signal exceeded a selected voltage level set above the signal noise. In most experiments, the noise level was determined when sympathetic activity was reduced to zero by a phenylephrine-induced increase in arterial pressure (>40 mmHg). In the remaining experiments, the noise level was set above the baseline noise level observed in between sympathetic bursts. In a small number of experiments, we verified the accuracy of the window discriminator measurement technique by comparing it with the results of integrated waveform analysis of the same signal. In these studies (n = 4), the amplified RSNA signal was split and fed simultaneously into both the window discriminator and a rectifying, integrating filter (BAK Electronics, time constant 20 ms). In these experiments both the rectified and integrated signal and the output from the window discriminator were fed to the computer (sampling rate 300-500 Hz) and sampled simultaneously.

Stimulation of the isolated ADN was performed with a programmable stimulator (Master8, AMPI) connected to a stimulus isolator (DS2, Digitimer). Stimulus intensity was set at 8-10 V, 0.2-ms pulse duration, with frequency varied from 2 to 15 Hz.

For central activation or blockade of the LPBN, a micropipette/stimulating electrode combination was constructed. A glass microinjection pipette (WPI, 1 barrel, 1.2-mm OD) was heated and bent at an ~20° angle and then attached to a monopolar stimulating electrode (100 Kohms, Frederick Haer) with the tips of the micropipette and the stimulating electrode aligned. The micropipette and stimulating electrode were held together with a dental acrylic. The microinjection pipette was then filled with either 10 or 20 mM D,L-homocysteic acid (DLH) for chemical activation of LPBN area neurons, 10 mM kainic acid for chemical blockade of LPBN neurons, or artificial cerebrospinal fluid (aCSF) as a control solution. All drugs were diluted in aCSF containing (in mM) 122 NaCl, 3 KCl, 25.7 NaHC0₃, and 1 CaCl₂, with the pH adjusted 7.4. The microinjection pipette was attached to a pressure injection system (BH-2, Medical Systems). The stimulating electrode was connected to an stimulus isolator (DS2, Digitimer) in series with the programmable stimulator (Master 8, AMPI). The micropipette/stimulating electrode combination was secured to a micromanipulator (Kopf Instruments) and stereotaxically positioned in the region of the left lateral LPBN (9.4-9.6 mm caudal to bregma, 1.9-2.2 mm lateral to the midline, and 5.6-6.0 mm ventral to the surface of the brain). The volume of solution (75-150 nl) that was pressure injected from the micropipette was determined by monitoring the movement of the meniscus in the microinjection pipette through a microscope equipped with a calibrated eyepiece. The current passed through the stimulating electrode was monitored on a differential oscilloscope (Nicolet) as the voltage drop across a 1-k resistor.

Experimental protocols. In all experiments, the micropipette/stimulating electrode was stereotaxically lowered into the region just above the left LPBN. The micropipette/stimulating electrode was then advanced to the optimal LPBN stimulation site, at which electrical stimulation (30-40 µA, 30-40 Hz, 0.2 ms) produced the greatest increase in arterial pressure and RSNA. The effects of left ADN stimulation on mean arterial pressure (MAP) and RSNA were then tested under the following experimental conditions: 1) electrical stimulation of the left LPBN, 2) chemical activation of the left LPBN with DLH, 3) chemical blockade of the left LPBN with kainic acid, 4) injection of aCSF into the LPBN, and 5) constant infusion of phenylephrine (a control for the pressor effect of LPBN stimulation). Data were acquired over a 1-min interval, consisting of 20 s of baseline recording, 15 s of recording during ADN stimulation at a single frequency (2, 5, 10, or 15 Hz), and 25 s of recording immediately after ADN stimulation (see Fig. 1). All data (arterial pressure, RSNA, and stimulus artifact) were fed to a Cambridge Electronics Design (CED) 1401 laboratory interface coupled to a Gateway 486 PC. Data were digitized and recorded with SPIKE2 (CED) software for storage and later analysis.

In those experiments in which the effect of electrical stimulation of the LPBN on the baroreflex control of MAP and RSNA was tested, responses to ADN stimulation were recorded before and during either threshold or two times threshold electrical activation of the LPBN. Threshold electrical stimulation of the LPBN was defined as the minimum frequency required (7-10 Hz, 30-40 µA, 0.2-ms duration) to produce a small change in RSNA but no change in baseline MAP (i.e., LPBN activation threshold of sympathetic activity). Within trials, electrical activation of the LPBN was started 10-15 s before the ADN stimulation protocol to allow time to achieve a new steady state in baseline values. LPBN stimulation was continued during ADN stimulation and then was terminated 5-10 s after the offset of ADN stimulation (see Fig. 1).

In those experiments in which DLH was used to selectively activate the LPBN neurons, the baroreflex response to electrical stimulation of the ADN at a single frequency was recorded before and 30-40 s after microinjection of 75-150 nl of DLH (10-20 mM) into the left LPBN. Thirty to forty seconds were allowed after the central injection of DLH to achieve new steady-state values in baseline MAP and RSNA. Five minutes of recovery were allowed before baroreflex responses were retested. Chemical activation of the LPBN was repeated before each ADN stimulation protocol, with 15-20 min between central DLH injections.

In those experiments in which the effect of chemical blockade of the LPBN on baroreflex regulation of arterial pressure and RSNA was tested, depolarizing chemical blockade of the LPBN was achieved with 10 mM kainic acid (28). First, control baroreflex responses were recorded (ADN stimulation 2-15 Hz) and the response to electrical stimulation of the left LPBN alone was recorded (30-40 Hz, 0.2 ms, 30-40 µA). Next, 100-150 nl of 10 mM kainic acid were microinjected into the left LPBN. One minute after kainic acid microinjection, electrical stimulation of the left LPBN was repeated to verify chemical blockade (i.e., expressed as an absence of an arterial pressure and RSNA response to electrical stimulation of the LPBN with the same parameters used just before blockade). Baroreflex responses to electrical stimulation of the ADN were then recorded at 2 and 6 min after the onset of LPBN blockade.

Finally, two sets of control experiments were performed. First, to control for the effect of microinjection alone, aCSF (the vehicle for both the kainic acid and DLH studies) was microinjected into the left LPBN. In these studies, baroreflex responses were recorded before and 30-60 s after microinjection of aCSF (100 nl) in the LPBN. In a second set of experiments, the effect of increased peripheral vasoconstriction on baroreflex control of MAP and RSNA was tested. An increase in peripheral vasoconstriction was induced by a constant infusion of intravenous dilute phenylephrine (2-10 µg at a rate of 1 ml/h). In these experiments, baroreflex responses were tested before and during a 15- to 30-mmHg increase in baseline arterial pressure (an increase similar to that observed during two times threshold electrical stimulation and during chemical activation of the LPBN).

A small amount of fluorescent latex beads (LumaFluor) was added to all microinjection solutions to facilitate identification of the microinjection sites. Electrical stimulation sites were marked by electrolytic lesions (1 mA for 2 s). After termination of the experiments, the animals were euthanized, and the brain was removed and placed in a 10% Formalin solution for 24-72 h. The brains were then removed from the Formalin and frozen to 12 to 14°C, and the rostral pons was sliced into 40-µm sections with a freezing microtome. The microinjection sites were recovered by imaging the brain slices with a microscope equipped with epifluorescence. The microscope was also equipped with a drawing side arm so the region of the LPBN containing the fluorescent beads and the electrolytic lesions could be recorded.

Data analysis. All data were analyzed offline using SPIKE2 software (CED). Peak changes in MAP during ADN stimulation were calculated as the difference between the baseline MAP (a 5-s average measured immediately preceding each ADN stimulation) and the peak drop in MAP during ADN stimulation (typically a 5-s average during the last 5 s of ADN stimulation). Changes in baseline MAP were calculated by taking the difference between the baseline value measured immediately before LPBN stimulation or microinjection and the new, steady state-value achieved during LPBN stimulation.

In all experiments, measurements of RSNA were quantified using a window discriminator. The window discriminator detected the number of RSNA voltage spikes passing over a preset voltage window (which was set above the noise level). For the present set of experiments the method of quantifying RSNA with a window discriminator was chosen over other methods of nerve analysis, such as full-wave rectification and averaging, because use of the window discriminator afforded elimination of multiple stimulus artifacts. This was particularly important because several of the experimental protocols elicited stimulus artifacts from two different sources (central monopolar LPBN electrode and a bipolar ADN electrode) and it was often difficult to simultaneously ground out both stimulus artifacts. Baroreflex-induced changes in RSNA were calculated as the difference between the baseline RSNA (a 5-s average measured immediately preceding each ADN stimulation, spikes/s) and the peak drop in RSNA during ADN stimulation (a 5-s average typically taken during the first 5 s of ADN stimulation).

Because there is evidence that the window discriminator may not be the most accurate method for measuring changes in nerve activity under some conditions (12), in four additional experiments raw RSNA was fed through both a window discriminator and a rectifying, integrating filter and recorded simultaneously. Baseline and peak RSNA during DLH activation of the LPBN were then averaged over 5-s time periods. The mean increase in RSNA from baseline during LPBN-induced sympathoexcitation was then quantified using both methods of data collection. The mean increase in RSNA during LPBN activation was 21 ± 9% when calculated from the window discriminator data versus 21 ± 6% when calculated from the rectified integrated nerve activity (P > 0.68). Because no significant difference was detected in the method of quantifying sympathetic activity when averaging data over 5-s time windows, all the RSNA data represented in RESULTS reflect data collected from the window discriminator.

Significant changes in baroreflex responses after manipulation of the LPBN or control manipulations were analyzed with a two-factor ANOVA with repeated measures (one factor was ADN stimulation frequency, and the second factor was the effect of LPBN or control manipulations). Significant changes in baseline levels of MAP or RSNA after LPBN stimulation/microinjection were analyzed using paired Student's t-test (comparing 5-s averages before and after LPBN stimulation/microinjection). Changes were considered significant when P < 0.05. All mean data are reported as means ± SE.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Baroreflex control of blood pressure and RSNA during electrical activation of the LPBN. Figure 1 illustrates two typical examples of the effect of unilateral electrical stimulation of the LPBN on baroreflex control of blood pressure and RSNA. In Fig. 1, A and C, the effect of threshold stimulation on baroreflex control of MAP and RSNA is shown. In this experiment, electrical stimulation of the left ADN alone (10 Hz, 0.2 ms, 8 V; Fig. 1A) produced a mean decrease in MAP and RSNA of 46 mmHg and 91 spikes/s, respectively (5-s averages). During electrical activation of the ipsilateral LPBN at a threshold intensity (10 Hz, 30 µA, 0.2 ms; Fig. 1C) baroreflex control of MAP and RSNA was attenuated (10-Hz ADN stimulation resulted in a 28-mmHg decline in MAP and a 58 spikes/s decline in RSNA; decreases were calculated relative to the new baselines evoked during 10-Hz LPBN stimulation). In this example, threshold stimulation of the LPBN resulted in little change in baseline MAP (i.e., from 127 to 130 mmHg) and a small increase in baseline RSNA (from 182 to 197 spikes/s). Figure 1, B and D, illustrates the effect of higher frequency LPBN activation on baroreflex control. In this animal, electrical stimulation of the ADN alone at 10 Hz produced a 42-mmHg decrease in MAP and a 93 spikes/s drop in RSNA (5-s averages; Fig. 1B). During electrical activation of the ipsilateral LPBN at two times the frequency for threshold activation (20 Hz, 40 µA; Fig. 1D), the baroreflex responses were markedly attenuated. In this experiment, ADN stimulation during simultaneous two times threshold activation of the LPBN evoked only a 6-mmHg decline in MAP and a 13 spikes/s decrease in RSNA. Baseline MAP and RSNA increased 7 mmHg and 57 spikes/s, respectively, during two times threshold LPBN stimulation. In these studies, in the urethan-anesthetized preparation, electrical stimulation of the ADN with a pulse width of 0.2 ms typically produced little observable change in heart rate, either before or during LPBN stimulation. Therefore changes in heart rate were not analyzed.

View larger version (38K): [in this window] [in a new window]   Fig. 1.   Effect of electrical stimulation of the LPBN on baroreflex control of arterial pressure (AP) and renal sympathetic nerve activity (RSNA). Records from 2 different paralyzed, ventilated, urethan-anesthetized rats are shown. A and C: baroreflex-mediated changes in AP and RSNA elicited by electrical stimulation of the left aortic depressor nerve (ADN; 8 V, 0.2 ms) before (A) and during (C) simultaneous electrical stimulation of the left lateral parabrachial nucleus (LPBN) at threshold intensity (10 Hz, 30 µA), which elicits a small increase in baseline RSNA. B and D: attenuation of baroreflex control of AP and RSNA was greater when LPBN stimulation frequency was set at 2× threshold for LPBN activation of sympathetic drive (20 Hz, 40 µA).

The averaged results of six experiments demonstrated that threshold electrical activation of the LPBN (1× LPBN threshold) produced a significant blunting of both arterial pressure and RSNA baroreflex responses (Fig. 2A). During threshold stimulation of the LPBN, there was no change in baseline MAP compared with control, but baseline RSNA did increase (Table 1). In five experiments (Fig. 2B), two times threshold LPBN stimulation resulted in an even greater attenuation of baroreflex control of MAP and RSNA than threshold LPBN stimulation. During two times threshold LPBN stimulation, there was a significant increase in baseline for both MAP and RSNA compared with control (Table 1).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Effect of electrical activation of the LPBN at 2 different frequencies on baroreflex control of mean AP (MAP) and RSNA. A: baroreflex-evoked (ADN stimulation, 8 V, 0.2 ms) changes () in MAP (top) and RSNA (bottom) before (solid bars) and during (open bars) threshold stimulation of the LPBN (n = 6). B: baroreflex-evoked (ADN stimulation, 8 V, 0.2 ms) changes in MAP (top) and RSNA (bottom) before (filled bars) and during 2× threshold stimulation (open bars) of the LPBN (n = 5). Electrical activation of LPBN significantly attenuated baroreflex control of MAP and RSNA. * Significantly different (P < 0.05) from control.

Baroreflex control of blood pressure and RSNA during chemical activation of the LPBN. In five experiments, the effect of chemical activation of the LPBN on baroreflex control was tested. Chemical activation of the left LPBN with DLH produced a significant rise in both baseline MAP and RSNA (Table 1), which typically peaked 20-30 s after a unilateral microinjection of DLH into left LPBN. During chemical activation of the LPBN, baroreflex control of MAP and RSNA were both modestly, but significantly (P < 0.004), attenuated (Fig. 3A). Baroreflex responses returned to control within 5 min after microinjection of DLH in the left LPBN (Fig. 3B).

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Effect of chemical activation of the LPBN on baroreflex control of MAP and RSNA. A: baroreflex-evoked (ADN stimulation, 8 V, 0.2 ms) changes in MAP (top) and RSNA (bottom) before (filled bars) and 1 min after (open bars) chemical activation of the left LPBN with D,L-homocysteic acid (DLH; 10 mM, 75-100 nl, n = 5). B: baroreflex-evoked (ADN stimulation, 8 V, 0.2 ms) changes in MAP (top) and RSNA (bottom) before (filled bars) and 5 min after (open bars) chemical activation of the LPBN (recovery). Chemical activation significantly attenuated baroreflex control of both MAP and RSNA. * Significantly different (P < 0.05) from control.

Baroreflex control of blood pressure and RSNA during chemical inactivation of the LPBN. Figure 4 illustrates the effect of depolarizing blockade of the left LPBN on baroreflex control of blood pressure and RSNA from a single animal. In this example before blockade, electrical stimulation of the left ADN at 5 and 10 Hz (0.2 ms, 8 V) produced a 28- and a 31-mmHg decrease in MAP and a decrease in RSNA of 98 and 129 spikes/s, respectively (Fig. 4A, top, middle and right). Before depolarizing blockade of the LPBN, electrical activation of the left LPBN produced a rapid rise in arterial pressure and RSNA (40 Hz, 40 µA, 0.2 ms; Fig. 4A, top left). One minute after chemical blockade, electrical stimulation of the LPBN no longer produced an increase in arterial pressure or RSNA and baroreflex control of MAP was slightly enhanced (i.e., after blockade 5 and 10 Hz ADN stimulation produced a 38- and a 41-mmHg decrease in arterial pressure). Baroreflex control of RSNA, however, was only modestly changed (i.e., after blockade, 5- and 10-Hz ADN stimulation produced a decrease in RSNA of 104 and 160 spikes/s, respectively).

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Effect of chemical blockade of the LPBN on baroreflex control of AP and RSNA in an anesthetized, paralyzed, ventilated rat. A: changes in AP and RSNA evoked during electrical stimulation of the left LPBN (left; 40 Hz, 40 µA, 0.2 ms) and electrical stimulation of the left ADN at 5 Hz (middle) and 10 Hz (right) (10 V, 0.2 ms). B: 1 min after chemical blockade of the left LPBN (10 mM kainic acid, 100 nl), electrical stimulation of the left LPBN no longer produces an increase in AP and RSNA (left). Three to five minutes after LPBN blockade, baroreflex control of AP during 5- and 10-Hz stimulation of the left ADN were slightly enhanced (middle and right), but baroreflex control of RSNA appears unchanged.

Data from seven experiments demonstrated that depolarizing blockade of the LPBN produced a small but statistically significant enhancement of baroreflex control of MAP (Fig. 5, A and B, top). Both immediately and at 6 min after LPBN blockade (see Fig. 5, A and B, top), the absolute decrease in MAP during ADN stimulation was significantly greater compared with before blockade. Baroreflex control of RSNA, however, was not altered after LPBN blockade (Fig. 5, A and B, bottom). At 1 and 6 min after chemical blockade of the LPBN, there was no significant change in baseline MAP (see Table 1). In contrast, immediately after LPBN blockade there was an increase in baseline RSNA, which reached significance at 6 min (Table 1).

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Effects of chemical inactivation of the LPBN on baroreflex control of MAP and RSNA. A: baroreflex-evoked (ADN stimulation; 8 V, 0.2 ms) changes in MAP (top) and RSNA (bottom) before (filled bars) and 1 min after (open bars) chemical inactivation of the left LPBN with 10 mM kainic acid (100-150 nl; n = 7). B: baroreflex-evoked (ADN stimulation, 8 V, 0.2 ms) changes in MAP (top) and RSNA (bottom) before (filled) and 6 min after (open bars) chemical inactivation of the left LPBN with 10 mM kainic acid. Chemical inactivation of the LPBN produced significant changes in baroreflex control of MAP but not RSNA. * Significantly (P < 0.05) different from control.

Control baroreflex curves. As a control for the microinjection studies, in five experiments baroreflex control of MAP and RSNA was compared before and after microinjection of aCSF into the left LPBN. Similar to the sites for DLH and kainic acid injection, sites for aCSF injections were first identified by the response to electrical stimulation of the dorsal lateral pons. Figure 6A shows the results of aCSF microinjection into the LPBN on baroreflex control of MAP and RSNA. Microinjection of aCSF into the LPBN had no significant effect on baroreflex control of MAP or RSNA. Microinjection of aCSF into the LPBN also had no effect on baseline MAP (130 ± 7 vs. 129 ± 7 mmHg, control vs. after aCSF microinjection) or RSNA (158 ± 5 spikes/s before vs. 152 ± 9 spikes/s after aCSF microinjection).

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Effect of both microinjection of artificial cerebrospinal fluid (aCSF) in the LPBN and phenylephrine-induced peripheral vasoconstriction on baroreflex control of MAP and RSNA. A: ADN-evoked changes in MAP (top) and RSNA (bottom) were not significantly changed after microinjection of 100 nl aCSF in the left LPBN (filled bars, before microinjection; open bars, after microinjection; n = 5). B: ADN-evoked changes in MAP (top) and RSNA (bottom) were not significantly changed during peripheral vasoconstriction induced by continuous intravenous infusion of phenylephrine (2-5 µg/min; filled bars, before vasoconstriction; open bars, during vasoconstriction; n = 4). Mean increase in MAP during phenylephrine infusion was 27 ± 8 mmHg.

A second set of control experiments was performed to determine whether baroreflex control of MAP was significantly altered by a change in the level of peripheral vasoconstriction that might be associated with chemical or electrical activation of the LPBN. In four animals, baroreflex responses were characterized before and during an intravenous infusion of phenylephrine sufficient to raise MAP 27 ± 8 mmHg (2-10 µg/min of phenylephrine). Continuous peripheral vasoconstriction with phenylephrine did not significantly alter baseline RSNA (111 ± 20 spikes/s before phenylephrine vs. 106 ± 19 spikes/s during phenylephrine) or baroreflex control of MAP or RSNA (Fig. 6B).

Histology. The histologically identified microinjection and electrical stimulation sites recovered for the experiments presented above are illustrated in Fig. 7. The majority of the effective sites for producing modulation of baroreflex control of MAP and RSNA were located within the boundaries of the LPBN (both electrical and chemical). Microinjection of DLH in the regions of the medial LPBN or just dorsal to the LPBN (Fig. 7, C and D) was not effective in modulating baroreflex function.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Histologically identified stimulation and microinjection sites in the PBN. A and B: electrical stimulation sites; shaded circles represent threshold stimulation sites (n = 6), and filled triangles represent 2× threshold stimulation sites (n = 5). C and D: chemical stimulation sites; filled circles represent sites where DLH microinjection increased AP and RSNA and significantly reduced baroreflex responses (n = 5), and shaded squares represent sites where DLH did change baroreflex responses (n = 5). E and F: chemical blockade sites and control injection sites; filled squares represent chemical blockade sites (n = 7), and shaded circles represent aCSF injection sites (n = 5). Inset shows position of the location of PBN in the dorsolateral pons. A, C, and E represent sections approximately 8.9 caudal to bregma, and B, D, and F are representative sections ~9.3 caudal to bregma according to Paxinos and Watson (24). MPBN, medial parabrachial nucleus; LC, locus ceruleus; scp, superior cerebellar peduncle; IC, inferior colliculus.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Previous studies have demonstrated that electrical activation of the PBN attenuates baroreflex control of heart rate (15, 20). Using threshold stimulation intensities, we have shown that electrical stimulation of the LPBN also significantly attenuates baroreflex control of MAP and RSNA. Furthermore, we have shown that chemical stimulation of the LPBN with DLH, which activates cell bodies and not fibers of passage, also significantly attenuates baroreflex control of sympathetic activity. Finally, using chemical blockade of the LPBN with kainic acid, we have demonstrated that there is little tonic effect of LPBN neurons on baroreflex control of MAP and RSNA in the anesthetized rat.

Our results suggest that, when activated, LPBN neurons can significantly reduce the capacity of the baroreflex to regulate sympathetically mediated increases in arterial pressure. Attenuation of baroreflex function is an important feature of centrally elicited pressor responses, such as the defense reaction, because increases in blood pressure activate baroreceptors and would normally trigger reflex-mediated decreases in sympathetic drive and heart rate (27). Thus a coordinated suppression of baroreflex function contributes to the animal's ability to respond to a variety of environmental stressors by permitting sustained increases in sympathetic drive and arterial pressure. Sympathetically mediated pressor responses can be elicited from several central sites, including the spinal trigeminal nucleus caudalis and the periaqueductal gray, components of the central nocioceptive pathways, and the hypothalamus. These nuclei have prominent connections to specific regions of the LPBN (3, 5, 9, 18, 25, 26), and lesioning studies have demonstrated that destruction of the LPBN markedly attenuates the pressor responses elicited by these same central nuclei (2, 21, 30). Our data would suggest that activation of the LPBN by these central pressor pathways provides a descending inhibition of medullary baroreflex pathways, facilitating the pressor response.

There are two likely points along the central baroreflex pathway at which LPBN neurons may modulate baroreflex function: the caudal NTS and the rostral ventrolateral medulla (RVLM). The caudal NTS is the first central termination site for baroreceptor afferents. Descending projections from the external LPBN and Kolliker-Fuse nucleus to the caudal NTS have been anatomically identified (11). One mechanism by which activation of the LPBN could modulate baroreflex function would be through gating baroreceptor afferent inputs at the level of the NTS. In this regard, Felder and Mifflin (6) have demonstrated that electrical activation of the PBN can directly inhibit the responses of single NTS neurons to baroreceptor inputs in the cat. Anatomic interconnections between the LPBN and RVLM have also been described, suggesting that LPBN modulation of baroreflex function may take place on the output side of central baroreflex pathway (10, 11). The RVLM contains neurons that innervate sympathetic preganglionic neurons in the spinal cord and has been described as the final integration site in the central baroreflex pathway (16). The activity of these bulbospinal RVLM neurons is subject to modulation by a variety of inputs. They are typically inhibited during periods of increased baroreceptor input and excited by the withdrawal of baroreceptor input. Stimulation of the LPBN, which generates an increase in sympathetic drive, excites RVLM neurons (1). Hypothetically, during a centrally driven pressor response involving activation of the LPBN, excitatory inputs from the LPBN and inhibitory baroreceptor-mediated inputs may converge on the same population of sympathoexcitatory RVLM neurons. A predominance of LPBN excitatory input might account for a blunting of the baroreflex at this level, although this hypothesis has not yet been tested.

In contrast to the effects on baroreflex function we observed during LPBN activation, we found that inactivation of LPBN neurons had little effect on baroreflex control. To our knowledge this is the first study to utilize chemical blockade rather than electrolytic lesions to examine the contribution of the LPBN to tonic baroreflex regulation of sympathetic activity. Our results demonstrated that a unilateral depolarizing blockade of LPBN neurons produced only a small enhancement of baroreflex regulation of blood pressure and had no significant influence on baroreflex regulation of RSNA. The failure of LPBN blockade to modulate baroreflex control of sympathetic nerve activity is in agreement with recent work by Koshiya and Guyenet (14) in which electrolytic lesion of almost the entire PBN (medial and lateral regions) was reported to have no effect on baroreflex control of splanchnic sympathetic nerve activity. The lack of a tonic influence over baroreflex control of sympathetic nerve activity is in contrast to the results of several studies that have suggested that destruction of the PBN does influence baroreflex control of heart rate (13, 21). These results suggest that separate populations of LPBN neurons may provide descending modulation of the baroreflex control of vagal and sympathetic outflows. It should be noted, however, that in the present study a greater tonic influence of LPBN neurons on baroreflex control of sympathetic nerve activity may have been masked by the presence other baroreceptor inputs (i.e., in our preparation the right ADN and carotid sinus nerves were left intact) and an intact contralateral LPBN. Thus the role of the LPBN in tonic regulation of sympathetic nerve activity needs to be investigated further in the presence of bilateral LPBN blockade and in a more completely barodenervated animal.

Although little tonic influence on baroreflex function was observed in our study or the study by Koshiya and Guyenet (14), in both studies baseline sympathetic nerve activity was reported to increase after PBN blockade. Koshiya and Guyenet suggested the changes in baseline sympathetic nerve activity after destruction of the PBN may have been due to changes in the level of anesthesia over time. In the present paradigm, however, changes in RSNA consistently occurred within only 6 min after blockade of the nucleus and thus could not solely be attributed to changes in anesthesia. In addition, Hubbard et al. (13) also saw evidence for an increase in sympathetic drive following electrolytic destruction of the PBN in conscious rats. We have now demonstrated that the loss of LPBN neuronal activity, not just of fibers passing through the region, can increase sympathetic nerve activity. Thus the results of several studies suggest that removing the influence of the LPBN results in an increase in baseline sympathetic drive. This finding is somewhat paradoxical because it is well established that activation of the PBN is sympathoexcitatory. An hypothesis to explain these observations might be that the LPBN contains two groups of neurons: one group that is tonically active and suppresses sympathetic tone (either directly or indirectly via actions on another central control system such as respiration) and a second that when activated are sympathoexcitatory. Our results would suggest that the sympathoexcitatory neurons are responsible for descending modulation of baroreflex function. Recent work by both Chamberlin and Saper (4) and Miura and Takayama (17) supports the suggestion that two groups of neurons with opposing influences on blood pressure are indeed present within the LPBN. However, it remains to be seen whether selective destruction of either of these two different groups of neurons has differential effects on sympathetic tone. In the present study, the size of the chemical blockade was relatively large so it would be difficult to attribute the rise in sympathetic activity to selective destruction of one group of neurons.

In summary, the results of the present study are the first to demonstrate that selective activation of neurons within the lateral portion of the PBN significantly attenuates baroreflex control of arterial pressure and RSNA in the urethan-anesthetized rat. Chemical blockade of the LPBN did not produce a significant modulation of baroreflex control of sympathetic activity, suggesting that the LPBN provides little tonic baroreflex regulation in the anesthetized preparation. These results are consistent with data from other studies demonstrating that activation of the LPBN is an essential component of several central pressor pathways and suggest that one important role of the LPBN might be to provide descending inhibition of baroreflex control of sympathetic activity, thereby permitting the maintenance of a sustained increase in sympathetic activity.