Microinjection of angiotensin II into the nucleus tractus solitarii attenuates the baroreceptor reflex-mediated bradycardia by inhibiting both vagal and cardiac sympathetic components. However, it is not known whether the baroreflex modulation of other sympathetic outputs (i.e., noncardiac) also are inhibited by exogenous angiotensin II (ANG II) in nucleus tractus solitarii (NTS). In this study, we determined whether there was a difference in the baroreflex sensitivity of sympathetic outflows at the thoracic and lumbar levels of the sympathetic chain following exogenous delivery of ANG II into the NTS. Experiments were performed in two types of in situ arterially perfused decerebrate rat preparations. Sympathetic nerve activity was recorded from the inferior cardiac nerve, the midthoracic sympathetic chain, or the lower thoracic-lumbar sympathetic chain. Increases in perfusion pressure produced a reflex bradycardia and sympathoinhibition. Microinjection of ANG II (500 fmol) into the NTS attenuated the reflex bradycardia (57% attenuation, P < 0.01) and sympathoinhibition of both the inferior cardiac nerve (26% attenuation, P < 0.05) and midthoracic sympathetic chain (37% attenuation, P < 0.05) but not the lower thoracic-lumbar chain (P = 0.56). We conclude that ANG II in the nucleus tractus solitarii selectively inhibits baroreflex responses in specific sympathetic outflows, possibly dependent on the target organ innervated.
- baroreceptor reflex
- blood pressure
in the rat, numerous peptides have been shown to modulate the baroreflex via actions within the nucleus tractus solitarii (NTS), which is the primary site of termination for baroreceptor afferent fibers (5, 23). It was first shown over 20 years ago that microinjection of angiotensin II (ANG II) into the NTS of rat caused an inhibition of the heart rate (HR) component of the baroreflex (4). This observation has been confirmed in other laboratories and is now well established (9, 11, 14, 18, 19). The cellular and subcellular pathways by which ANG II in the NTS depresses the baroreflex are now being elucidated (17, 33). Surprisingly, however, few studies have examined the effects of NTS application of ANG II on the sympathoinhibitory component of the baroreflex. Our laboratory (2) has demonstrated that baroreflex modulation of cardiac sympathetic nerve activity (SNA) is also attenuated by exogenous ANG II in the NTS. However, considering the high degree of functional specificity within the sympathetic nervous system (7, 12, 13), the same cannot be assumed for other sympathetic outputs. In the present study, therefore, we compared the effects of microinjection of ANG II into the NTS on the baroreflex-mediated sympathoinhibition of the cardiac nerve with midthoracic and lower thoracic-lumbar levels of the sympathetic chain in the rat. Some of these data have been presented as a poster communication (25).
MATERIALS AND METHODS
All procedures were performed in compliance with the United Kingdom's Animals (Scientific Procedures) Act 1986 and were approved by the University of Bristol ethics committee. Experiments were performed on Wistar rats (males, 70–125 g, University of Bristol colony), using two variations of a decerebrated, in situ artificially perfused preparation. The first of these was the “working heart brain stem preparation” (WHBP; Ref. 15). Briefly, following deep anesthesia with halothane, the rat was bisected subdiaphragmatically and decerebrated at the level of the superior colliculus. A double-lumen cannula was inserted retrogradely into the descending aorta, and the preparation was perfused with modified Ringer solution (32°C), using a peristaltic pump. The second variation, termed the “decerebrate artificially perfused rat” preparation (DAPR; Ref. 22) utilizes the whole rat and has the advantage of maintaining an entirely intact spinal cord, which allows the recording of lower thoracic-lumbar sympathetic chain activity. Under halothane anesthesia, the majority of the intestine, peritoneum, and stomach were ligated and resected. The heart was exposed via a midline thoracotomy, and the rat was decerebrated as described above. In this preparation, the perfusion with modified Ringer solution was performed transcardially via a double-lumen cannula inserted through the left ventricle into the ascending aorta. In both preparations, arterial perfusion pressure was monitored using the second lumen of the cannula, connected to a pressure transducer. Phrenic nerve activity and electrocardiograms (ECG) were recorded and used to monitor respiratory rate and HR, respectively. Arginine vasopressin (200–400 pM) was added to the perfusate to increase perfusion pressure in the DAPR preparation (22).
Sympathetic nerve recording.
Recordings of SNA were made from either the inferior cardiac nerve or from one of two levels of the sympathetic chain (T5–T8 or T11–L1) using a monopolar suction electrode. The inferior cardiac nerve was identified anatomically, exiting from the inferior portion of the stellate ganglion and traveling in close proximity to the azygous vein toward the heart. In some experiments, the identity of the cardiac nerve was confirmed by the generation of a tachycardia following stimulation (20–30 Hz, 0.2- to 1-ms pulse duration, 1–15 V) of the peripheral cut end of the nerve delivered via a glass suction electrode. SNA was measured from the cardiac nerve in the WHBP (n = 12) and from the lower thoracic-lumbar sympathetic chain (T11–L1) in the DAPR preparation (n = 9). Both preparations (WHBP, n = 6; DAPR, n = 3) were used to measure midthoracic chain (T5–T8) SNA, and the responses were found to be equivalent. Nerve activity was amplified, filtered (100–2,000 Hz), and acquired to a computer using a CED 1401 analog-to-digital interface (CED, Cambridge, UK). The signals were rectified and integrated (time constant 100–500 ms) digitally using Spike 2 (CED). At the conclusion of the experiment, the noise level was determined by either the application of lignocaine (2% wt/vol) to the nerve or addition of hexamethonium dichloride to the perfusate (50–100 μM).
To stimulate the baroreceptors, we increased the pump flow rate for 2–3 s to generate pressure ramps (similar to Ref. 19) that were highly reproducible over the course of an experiment. Because SNA shows respiratory rhythmicity and the magnitude of the baroreflex responses are dependent on the phase of the respiratory cycle, the pressor ramp was timed to correspond with the commencement of a burst in phrenic nerve activity such that its peak coincided with late inspiration (19, 21). Baroreflex gain was measured before and after microinjection of ANG II (50 nl, 500 fmol) into the NTS. To confirm that baroreflex responses could be inhibited at these NTS sites, in some experiments we also measured baroreflex gain before and after microinjection of the GABAA receptor agonist isoguvacine hydrochloride (50 nl, 50–500 pmol) into the same sites. The doses of the drugs were based on those of previous experiments and have been shown to be near maximal (ANG II; Ref. 19) or sufficient to block reflexly evoked changes in HR (isoguvacine; Refs. 3, 21). To perform the injections, we visualized the dorsal surface of the medulla after removal of the cerebellum. This allowed high precision and consistent placement of the microelectrodes across preparations. A glass micropipette containing one of the drugs was positioned in the NTS relative to the calamus sciptorius. The drugs were injected into two sites, bilaterally, over a period of 1–2 min. The locations of these sites were the same as those described previously (2, 19). The first site was at the level of the calamus sciptorius, 300 μm lateral to the midline and 500 μm deep. The second site was ∼500 μm rostral and 200 μm lateral to the first injection, at a depth of 500 μm. Injections were made by applying pressure manually. The volume injected was measured by observing the displacement of the meniscus using a binocular microscope fitted with a calibrated eyepiece.
Data were analyzed using Spike2 software with custom-written scripts. The baseline values for MAP and HR were calculated as the average values over the 10-s period immediately preceding a pressor ramp. Baroreflex gain for HR was calculated as the maximum change in HR during a pressor ramp divided by the change in perfusion pressure (i.e., beats·min−1·mmHg−1). SNA exhibited respiratory modulation such that burst discharges occurred during the late inspiratory and early expiratory phases. Baseline values were determined during these periods over two phrenic cycles immediately preceding a pressor ramp (for more details see Fig. 1 in Ref. 21). Baroreflex gain for SNA was then calculated as the average integrated SNA over the same period in a single respiratory cycle during a pressor ramp (which was timed to coincide with the respiratory-related burst), expressed as a percentage of the change in SNA from baseline SNA divided by the change in perfusion pressure (i.e., %ΔSNA/mmHg). This method of analysis allows for the determination of the maximum degree of baroreflex-mediated sympathoinhibition. It has been shown previously (22) in the in situ preparations that the pressure threshold for initiation of a baroreflex bradycardia is ∼80 mmHg and that the linear part of the baroreflex function curve for HR lies in the range of ∼5–20 mmHg above threshold (19, 22, 30). Therefore, to determine baroreflex gain on the linear part of the curve, we ensured that the pressor ramps were made over this pressure range. This was confirmed in five preparations by constructing baroreflex function curves using a series of pressor ramps (19, 22, 30).
The rate of recovery of SNA following a baroreceptor stimulus was also determined, because this has been shown to demonstrate modulation in the central baroreflex pathway, even when gains are not affected (26, 28). For this purpose, the raw SNA signal was resampled at 100 Hz, integrated (100-ms time constant), and plotted against time. Using Prism software (version 4.0; GraphPad, San Diego, CA), nonlinear regression was used to fit the data to the first-order exponential recovery equation Y= Ymax[1 − exp(−KX)], where Y is SNA, Ymax is baseline SNA, X is time, and K is the time constant for the rate of recovery. Therefore, the recovery time to 50% of baseline SNA is 0.69/K.
All data are means ± SE, and n is the number of preparations. Data were tested for normality using a Shapiro-Wilk's test. Comparisons of the changes in baroreflex gain evoked by microinjections of ANG II or isoguvacine hydrochloride into the NTS were made using ANOVA or a two-tailed Student's t-test as appropriate. Differences were considered significant at P < 0.05.
The composition of the Ringer solution was (mM) 125 NaCl, 24 NaHCO3, 5 KCl, 2.5 CaCl2, 1.25 MgSO4, 1.25 KH2PO4, and 10 dextrose, pH 7.3. Ficoll-70 (1.25%) was added to the solution as an oncotic agent, and the solution was gassed with carbogen (95% O2-5% CO2). The following drugs were used: arginine vasopressin (250 pM, dissolved in 0.9% NaCl), ANG II (10 μM, dissolved in 0.9% NaCl), and isoguvacine hydrochloride (10 mM, dissolved in 0.9% NaCl). All drugs and chemicals were purchased from Sigma UK.
Comparison of cardiovascular variables in the WHBP and DAPR preparations.
Our laboratory reported previously (22) that the perfusion pressure in the DAPR is lower than in the WHBP, reflecting the larger vascular tree and subsequent lower total peripheral resistance in this model. Perfusion pressure was therefore raised in these preparations using vasopressin. Under these conditions, resting perfusion pressure for the WHBP and DAPR preparation were comparable (WHBP: 72.9 ± 3.9 mmHg, n = 14; DAPR: 72.0 ± 2.0 mmHg, n = 13; P = 0.84). Likewise, resting HR (WHBP: 317 ± 14 beats/min; DAPR: 307 ± 5 beats/min; P = 0.52) and phrenic nerve activity cycle length (WHBP: 2.7 ± 0.3 s; DAPR: 3.0 ± 0.4 s; P = 0.46) were not different between the two preparations.
In the two groups, ramps of increasing perfusion pressure of similar magnitude were used to compare baseline cardiac baroreflex responses. In the WHBP (n = 14), a pressure increase of 24.4 ± 1.7 mmHg evoked a bradycardia of 49.6 ± 5.3 beats/min, and in the DAPR (n = 13), a pressure increase of 26.4 ± 2.4 mmHg evoked a comparable bradycardia of 46.7 ± 6.1 beats/min. Hence, the corresponding cardiac baroreflex gains were similar (WHBP: 2.1 ± 0.2 beats·min−1·mmHg−1; DAPR: 1.8 ± 0.2 beats·min−1·mmHg−1; P = 0.42). These results are highly consistent with those published in earlier studies by our laboratory (2, 19, 21).
Sympathetic nerve activity.
Consistent with previous reports, all three nerves showed rhythmic activity, with bursts of increased activity coincident with mid inspiratory-early expiratory periods relative to phrenic nerve discharge (2, 21). Increasing perfusion pressure caused a marked diminution of SNA in all nerves. Overall, the baroreflex sympathoinhibition in all three nerves was ∼70–75%. The sympathetic baroreflex gains were not significantly different between the three nerves, although the gain of the lower thoracic-lumbar chain SNA tended to be lower.
Effects of ANG II on baroreflex sensitivity.
Microinjection of ANG II (500 fmol) into the NTS did not produce a consistent change in resting perfusion pressure (P = 0.85) or HR (P = 0.60) in either in situ preparation. There was a small reduction in respiratory period (∼15%, P < 0.05) that was the same in both preparations. The baroreflex gains for HR, cardiac SNA, and midthoracic chain SNA were all attenuated by ANG II (Figs. 1 and 2). The attenuation in HR baroreflex gain was similar in the two preparations (gain after ANG II: WHBP, 0.8 ± 0.2 beats·min−1·mmHg−1 or 57.7 ± 5.9% reduction, n = 14; DAPR, 0.8 ± 0.2 beats·min−1·mmHg−1 or 52.5 ± 6.4% reduction, n = 12; P = 0.81 between preparations) and is similar to that described previously (2, 19). The gain of the inferior cardiac nerve sympathoinhibition was attenuated by 32%. The baroreflex sympathoinhibition in the midthoracic chain was attenuated by 37% in the WHBP (Fig. 2) and by 42% in the DAPR (control gain: 2.7 ± 0.47 %ΔSNA/mmHg; gain after ANG II: 1.6 ± 0.30 %ΔSNA/mmHg; n = 3). There was no significant difference in the ANG II-induced attenuation of midthoracic SNA between the two preparations (P = 0.78). In stark contrast to cardiac and midthoracic chain SNA, the gain of the baroreflex control of lower thoracic-lumbar chain SNA was unaffected by ANG II (Figs. 1 and 2). After a waiting period of 15–30 min following ANG II injection into the NTS, the gains of the baroreflex control of HR and of SNA recorded from all three sympathetic outflows had returned to values not significantly different from their respective control values (Fig. 2). Finally, the rate of recovery from baroreflex sympathoinhibition in the cardiac and midthoracic chain, but not the lower thoracic-lumbar chain, after a ramp increase in pressure was increased by microinjection of ANG II into the NTS (Fig. 3).
Effects of isoguvacine hydrochloride on baroreflex gains.
One interpretation of the absence of ANG II effect in NTS on the reflex lower thoracic lumbar nerve responses to baroreceptor stimulation is that the region of the NTS where the injections were made does not mediate this component of the reflex. To test this possibility, we microinjected a GABAA agonist (isoguvacine hydrochloride) into the same region of the NTS (i.e., the coordinates of the isoguvacine injections were the same as for the ANG II injections). Isoguvacine (500 pmol) had no effect on resting perfusion pressure (before: 68.4 ± 3.6 mmHg; after: 74.6 ± 6.3 mmHg; n = 10, P = 0.41) or HR (before: 318 ± 16.1 beats/min; after: 328 ± 16 beats/min, P = 0.64) but had a dramatic effect on all measured outputs of baroreflex sensitivity including lower thoracic-lumbar chain sympathoinhibition (Figs. 4 and 5). The baroreflex HR gain was reduced by over 90% (P < 0.01), cardiac SNA gain by >75% (n = 5, P < 0.01), and lower thoracic-lumber chain SNA gain by 70% (n = 5, P < 0.01; Fig. 5). The effects of isoguvacine on the sensitivity of the midthoracic component of the baroreflex gain was tested in one preparation, and in this case the gain was reduced by 67.5%. Finally, in three preparations, a lower dose of isoguvacine (50 pmol) attenuated HR gain by 84 ± 4% and lower thoracic-lumbar chain sympathoinhibition by 50 ± 26%. In all cases, these baroreflex gains returned to the control levels within ∼15 min.
It is well established that the HR component of the baroreflex is attenuated by exogenous ANG II in the NTS (4, 9, 14, 19). Although this is mediated largely via inhibition of vagal activation (19), it has recently been shown that the baroreflex sympathoinhibition of cardiac SNA is also attenuated (2). The present study confirmed these results and, in addition, revealed that the ANG II-mediated depression of baroreflex sensitivity is not uniform across all sympathetic motor outputs. We found that NTS microinjection of ANG II had no effect on the baroreflex-evoked sympathoinhibitory response recorded from the lower thoracic-lumbar sympathetic chain, yet it did attenuate the response recorded at the middle level of the thoracic sympathetic chain (as well as the inferior cardiac nerve). These results indicate that ANG II does not act globally within the NTS on all components of the baroreceptor reflex pathway but confers a differential degree of inhibition within networks regulating sympathetic outflow.
We used two in situ artificially perfused preparations for this study, since they offer a number of advantages over conventional in vivo anesthetized animal experiments, as discussed previously (2, 15, 22). In addition to the WHBP, we chose to use the DAPR preparation for experiments in which lower thoracic-lumbar chain SNA was recorded, because in the WHBP the spinal cord is transected just below the level of the diaphragm, and this could affect lower thoracic-lumbar chain SNA per se and/or alter the baroreflex responses in this sympathetic outflow. It is possible that the effects of ANG II on the baroreflex responses of the two preparations may be different. Although we cannot discount this possibility, it seems unlikely because the physiological characteristics of the two preparations are very similar: both exhibit a eupneic pattern of phrenic nerve activity, and numerous somatic and visceral reflex responses, including the baroreflex, appear to be the same (21, 22, 30). We found that SNA exhibited similar properties in the two preparations, showing a clear bursting pattern with the same phase relationship to phrenic nerve activity, which is in agreement with previous studies (2, 7, 21, 22). Furthermore, baroreceptor stimulation inhibited SNA significantly in all three nerves, regardless of the preparation, whereas the cardiac baroreflex gains were almost identical in the two preparations and were modulated to a similar degree by parenchymal injections of ANG II into the NTS. Most importantly, the effect of NTS application of ANG II on the baroreflex response of midthoracic chain SNA was tested in both preparations and found to attenuate the gain to a similar degree. Finally, the GABAA agonist isoguvacine applied to the same NTS region blocked baroreflex inhibition of all three sympathetic nerves in both preparations.
A major aim of the present study was to examine the effects of direct application of ANG II in the NTS on baroreflex sensitivity of sympathetic nerves supplying targets other than the heart. An important observation was that exogenous ANG II in the NTS exerted differential effects on baroreflex modulation of distinct sympathetic outflows. The concentration of ANG II used in the present study has been used previously to attenuate baroreflex-mediated inhibition of HR and cardiac SNA. At this dose, the effects of ANG II are reversible and specific, acting selectively on ANG II type 1 (AT1) receptors, because blockade of AT1 receptors with losartan prevents the ANG II inhibition of the reflex (2, 19). Consistent with the present study, Matsumura et al. (11) reported that baroreflex gain of renal SNA was increased following blockade of AT1 receptors in the NTS of the anesthetized rat, suggesting that endogenous ANG II may act tonically in the NTS to inhibit the baroreflex control of renal SNA. Interestingly, these investigators found that the increase in baroreflex gain in the renal nerve response (37%) was much lower than in the HR response (108%) following AT1 receptor blockade in the NTS. This provided the first clue that ANG II at the level of the NTS can cause a differential degree of modulation between baroreflex control of HR and renal SNA. More recently, differential regulation between renal and lumbar SNA has been demonstrated (27). In that study, intravenous administration of the nitric oxide synthase inhibitor nitro-l-arginine methyl ester (l-NAME) produced a significantly greater decrease in renal than lumbar SNA. The decrease in SNA is believed to be baroreflex mediated in response to the increase in blood pressure, because if blood pressure is prevented from rising, then intravenous l-NAME causes no change in SNA (8, 27). By contrast, after blood pressure is increased using phenylephrine, the reductions in SNA in the renal nerve and lumbar sympathetic chain are of similar magnitude (27). A possible explanation for this is that nitric oxide may act tonically to inhibit the baroreflex control of renal but not lumbar chain SNA. Interestingly, recent evidence suggests that the inhibition of baroreflex control by ANG II in the NTS is mediated via an increase in nitric oxide synthesis (17, 18). Therefore, although l-NAME was applied intravenously in that study (27), when considered in the context of the present results and those of Matsumura et al. (11), the results are consistent with the idea that nitric oxide at the level of the NTS inhibits the baroreflex control of renal SNA to a greater degree than that of lumbar chain SNA. The actions of nitric oxide in the NTS are complex and have been reported to increase both excitatory (glutamatergic) and inhibitory (GABAergic) neurotransmission (32). Further studies examining the effect of NTS application of nitric oxide and/or ANG II on baroreflex-mediated sympathoinhibition in simultaneous recordings of the renal nerve and lumbar sympathetic chain are required to confirm this hypothesis.
The size of the ANG II-induced depression of the sympathoinhibitory response in the midthoracic chain and inferior cardiac nerve were similar in magnitude. A possible explanation for this is that the recordings of midthoracic chain SNA (T5–T8) included a substantial cardiac component, because sympathetic preganglionic neurons supplying the heart arise from the T1–T8 thoracic spinal segments (31). At the same time, we cannot rule out the possibility that the baroreflex control of noncardiac sympathetic outflows in the midthoracic chain was inhibited by ANG II within the NTS. Therefore, although our data clearly show that ANG II in the NTS inhibits the baroreflex control of cardiac and midthoracic chain SNA but not that of the lower thoracic-lumbar chain SNA, we cannot say whether this is due to differential effects on cardiac and noncardiac components of the sympathetic outflow or, alternatively, on sympathetic outflows arising from different rostrocaudal spinal levels.
In addition to determining the effect of NTS application of ANG II on the size of the baroreflex-evoked sympathoinhibition (i.e., the gain), we also examined the effect that ANG II had on the rate of recovery of SNA following baroreflex inhibition. Previously, it has been shown that microinjection of agonists to the cannabinoid CB1 receptor into the NTS causes facilitation of the baroreflex that is expressed as a delay in the rate of recovery of the sympathoinhibition to a baroreceptor stimulus, rather than affecting the gain of the reflex (26, 28). The mechanism by which the cannabinoid agonist specifically altered the rate of recovery without affecting the gain is unclear. One possibility is that changes in recovery rate are mediated via a separate component of the baroreceptor reflex pathway, possibly involving type II rather than type I baroreceptor afferent fibers (26). We thought that the determination of recovery rate might provide a more sensitive method of measuring ANG II modulation of the baroreflex and that this might demonstrate inhibition in the lower thoracic-lumbar chain. However, the results were consistent with the effects we observed on baroreflex gain. We found that ANG II in the NTS increased the rate of recovery from baroreflex-mediated sympathoinhibition in the inferior cardiac nerve and midthoracic chain but not in the lower thoracic-lumbar chain, proving further support for a differential action of ANG II on different sympathetic outflows.
The question arises, what are the means by which ANG II in the NTS produces differential baroreflex regulation of sympathetic outflows? Clearly, our data suggest that there are discrete conduits within the central baroreceptor reflex pathway that allow for specific modulation of different sympathetic outflows by ANG II. It is possible that there is a topographical distribution of second-order baroreflex neurons in the NTS such that neurons that regulate lower thoracic-lumbar chain sympathetic outflow are located more rostral to those controlling midthoracic chain sympathetic outflows. If this were the case, it could be argued that our injections of ANG II were made too caudal to act on the putative population of neurons that control the baroreflex inhibition of lower thoracic-lumbar chain SNA. Although we cannot discount it, this possibility is unlikely for several reasons. First, the available data do not support the idea of topographical organization of second-order neuronal populations within the NTS. For example, intracellular recording and labeling studies have shown considerable overlap of these neuron populations, with little apparent viscerotopy (16), whereas barosensitive NTS neurons have widespread dendritic arborizations, with their dendritic trees typically spreading for over 1 mm (6). This makes it highly unlikely that our microinjection protocol could have targeted specific subpopulations of barosensitive NTS neurons based on site selectivity. Second, the density of ANG II receptors in the NTS is much higher at levels corresponding to the middle and more caudal levels of the area postrema (where we injected) than at the rostral limit of the area postrema (1). Our injections, therefore, were centered on this region of the NTS with the highest density of ANG II receptors. Finally, injections of isoguvacine into the same site were capable of blocking baroreflex inhibition of lower thoracic-lumbar chain SNA, suggesting that the ANG II injections were capable of reaching the neuronal population regulating sympathetic outflow to the lower thoracic-lumbar chain. We therefore believe that a more likely explanation is that there is selective expression of ANG II receptors on different populations of barosensitive NTS neurons. Supporting this concept, preliminary results from our laboratory (24) have shown that not all barosensitive NTS neurons respond to ANG II; however, further data demonstrating differential expression of ANG II receptors on NTS barosensitive neurons are required to confirm this hypothesis.
Several studies have already shown that there is divergence of the baroreflex pathway into its cardiac vagal and sympathetic components at the level of the NTS. For example, the vagal limb of the baroreceptor reflex is dependent on glutamate receptors (both ionotropic and metabotropic) in the NTS, which is not the case for the sympathetic limb (10, 29). In addition, our group (21) has shown previously that substance P in the NTS inhibits the cardiac vagal, but not the cardiac sympathetic, component of the baroreflex and that this inhibition is dependent on the activation of GABAA receptors. This mechanism appears similar to the baroreflex inhibition by ANG II in the NTS, which is also dependent on GABAA receptors within the NTS (17, 24). Our group (21) proposed initially that there were two distinct populations of GABAergic interneurons within the NTS that could differentially control the cardiac vagal and sympathetic components of the baroreflex pathway. The present work suggests that there also may be divergence within the sympathetic component of the baroreflex at the level of the NTS. Because baroreflex inhibition of the lower thoracic-lumbar chain SNA was also blocked by isoguvacine in the NTS, this component of the baroreflex pathway also appears to possess the anatomical substrate within the NTS for GABAergic modulation, even though it is not inhibited by ANG II. Therefore, we suggest that GABAergic interneurons that modulate the baroreflex within the NTS form a heterogeneous population in terms of their differential control over different sympathetic outflows. Whether such differential regulation actually reflects an underlying viscerotopic organization of the baroreflex pathway within the NTS such that the regulation of baroreflex responses to different organs (for example, heart, kidney, and muscle) are controlled by different subgroups of neurons within the NTS remains hypothetical.
Perspectives and Significance
It is now well known that the sympathetic nervous system is not activated in an “all or none” manner but that it maintains highly specific control of the various effector organs. This has been demonstrated convincingly at the post- and preganglionic sympathetic neurons (7) and also more recently at higher levels of the neuraxis, such as within the cardiovascular presympathetic neurons in the rostral ventrolateral medulla (12, 13). The present data provide further evidence of such heterogeneity, demonstrating that the baroreflex regulation of different sympathetic outflows is modified differentially by ANG II within the NTS. The renin-angiotensin system plays an important homeostatic function during periods of low blood volume and/or blood pressure. One could argue that in these circumstances, it is beneficial for baroreflex control of certain target organs, such as the heart, to be attenuated, so as to maintain a higher cardiac output. It is possible, therefore, that ANG II within the NTS plays an important role in maintaining blood pressure in times of cardiovascular stress. Conversely, inappropriately high levels of ANG II within the NTS, and the subsequent changes to baroreflex sensitivity, may be a factor in the ontogeny or progression of cardiovascular diseases such as hypertension and heart failure.
We are grateful to the British Heart Foundation, the Leverhulme Trust of the University of Bristol, and the National Heart, Lung, and Blood Institute (HL033610-18) for supporting this research. J. F. R. Paton was the recipient of a Royal Society Wolfson Research Merit Award.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society