| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2003;42:1124.)
© 2003 American Heart Association, Inc.
Scientific Contributions |
From the Department of Physiology, the University of Texas Health Science Center at San Antonio, San Antonio, Tex.
Correspondence to Lila P. LaGrange, PhD, Department of Physiology7756, The University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, TX 78229-3900. E-mail lagrange{at}uthscsa.edu
| Abstract |
|---|
|
|
|---|
-aminobutyric acid)-ergic function in the PVN is reduced by peripheral infusion of Ang II. To accomplish this, rats received either intravenous Ang II (12 ng/kg per minute) or vehicle (D5W) for 7 days, and renal sympathetic nerve activity (SNA), mean arterial pressure (MAP), and heart rate (HR) responses were recorded after unilateral PVN microinjection of the GABA-A receptor antagonist bicuculline methiodide (BMI, 0.1 nmol). Results indicate that in contrast to a significant increase in renal SNA, MAP, and HR observed in vehicle-infused rats (P<0.05), BMI injection into the PVN of Ang IIinfused animals was without effect on all recorded variables. In a separate groups of animals, ganglionic blockade produced a significantly greater fall in MAP (P<0.01) in Ang IIinfused rats than in vehicle-infused control rats, indicating that the contribution of SNA to the maintenance of blood pressure was elevated in the Ang IIinfused group. Overall, these data indicate that cardiovascular and sympathoexcitatory responses to acute GABA-A receptor antagonism in the PVN are significantly blunted in rats after 7 days of intravenous infusion of Ang II. We conclude that an Ang IIinduced reduction in GABAergic inhibition within the PVN may contribute to elevated SNA observed in this study.
Key Words: angiotensin II hypothalamus sympathetic nervous system hypertension, arterial
| Introduction |
|---|
|
|
|---|
-aminobutyric acid (GABA)1 plays an important role in regulating cardiovascular function.2,3 In brain regions such as the nucleus tractus solitarius2 and rostral ventrolateral medulla2,3 that control sympathetic nerve activity (SNA), GABA tonically suppresses neuronal activity and excitability. In the hypothalamic paraventricular nucleus (PVN), neuronal activity is also regulated by GABA46 and by a number of excitatory neurotransmitters as well.7,8 Among the latter is the peptide angiotensin II (Ang II). Ang IIcontaining fibers arise from Ang IIsensitive regions of the forebrain known as circumventricular organs (CVO), which lack a complete blood-brain barrier.9,10 Neurons in two forebrain CVOs, the subfornical organ and organum vasculosum of the lamina terminalis, express Ang II AT1 receptors in high density11 and appear to sense and respond to circulating Ang II.9,1215 Although Ang II increases SNA through these and other central actions,16,17 the mechanism by which circulating Ang II enhances PVN neuronal excitability is not fully understood. On the basis of literature evidence, it is clear that GABA46 and Ang II1315 can each act individually within the PVN to influence cardiovascular function. More recently, it has been shown that local GABAAng II interactions may be key to regulating PVN neuronal excitability and sympathetic outflow.4 Although mechanisms of GABAAng II interaction within the PVN have not been fully defined, in vitro electrophysiological studies indicate that GABA tonically opposes excitatory actions of Ang II on neurosecretory neurons,18 and excitation of spinally projecting neurons involves presynaptic actions of Ang II to reduce GABA release.19 In addition, functional studies in rats with sodium-dependent hypertension20 and heart failure6 have suggested that Ang IIinduced sympathoexcitation involves reduced GABA-mediated inhibition within the PVN. This conclusion is consistent with evidence that both plasma Ang II and SNA are elevated whereas the sympathetic response to blockade of PVN GABA-A receptors is attenuated in these disease models.6,20
The aim of the present study was to determine more directly the effect of elevated plasma Ang II on cardiovascular and renal SNA responses to acute GABA-A receptor blockade in the PVN. This was accomplished by infusing rats intravenously with Ang II at a rate selected to leave plasma volume undisturbed. Results indicate that responses to blockade of GABA-A receptors in the PVN are blunted in Ang IIinfused rats compared with vehicle-infused control rats and are consistent with the hypothesis that in heart failure6 and certain forms of hypertension,20 elevated peripheral Ang II could act to reduce GABAergic function within the PVN. Some of these results have been presented as an abstract.21
| Methods |
|---|
|
|
|---|
Twelve-Day Infusion Protocol
Rats were anesthetized (pentobarbital, 50 mg/kg IP), and a telemetry probe was placed in the aorta to record blood pressure. The vena cava was catheterized to infuse Ang II or vehicle.17 After regaining consciousness, a solution of D5W and ampicillin (28 mg/24 h) was infused (5.7 mL/24 h) continuously for 5 days. Rats were then separated into two groups; one received fresh Ang II (12 ng/kg per minute) in the infusate daily and the other (control group) was maintained on the vehicle infusion for the final 7 days.
In a separate group of instrumented animals, the contribution of sympathetic activity to the maintenance of blood pressure was determined 3 days before and after completing the 7-day Ang IIinfusion or vehicle-infusion period. While animals were conscious, the maximal fall in mean arterial pressure (MAP) produced by the ganglionic blocker chlorisondamine (2.5 mg/kg IV) was determined. Because of long-lasting drug effects, animals treated with chlorisondamine were not used in subsequent microinjection studies.
Experimental Procedures
On completing the infusion protocol, rats were anesthetized (
-chloralose, 80 mg/kg; urethane, 800 mg/kg IP) and a femoral vein and artery were catheterized to administer drugs and measure arterial pressure, respectively. Standard methods were used to record renal SNA.4,17 Renal SNA responses to intravenous sodium nitroprusside and phenylephrine were used to assess the signal-to-noise ratio of each recording. Hematocrit was measured from venous blood. Animals were paralyzed (gallamine triethiodide, 25 mg/kg IV) and artificially ventilated. End-tidal PCO2 and body temperature were maintained within normal limits.
Microinjection of Drugs
A 30-gauge injector was lowered stereotaxically into the PVN (coordinates: 1.7 to 1.9 mm caudal to bregma, 0.5 mm lateral to midline, and 7.4 to 7.6 mm ventral to the skull surface). The GABA-A receptor antagonist bicuculline methiodide (BMI) (or saline vehicle) was injected (0.1 nmol in 100 nL) unilaterally over a period of 60 seconds. In some experiments, BMI was purposely delivered outside the PVN (1.8 mm lateral to midline) to serve as an anatomical control.
Histology
After each experiment, Evans blue dye (2% in saline) was injected (100 nL) into the PVN (or adjacent control site). Brains were removed, placed in paraformaldehyde (4% for 24 hour), sectioned (40 µm), and viewed through a microscope to identify injections sites.
Analysis
During the infusion period, 24-hour averages of MAP and HR were calculated. Changes in MAP and HR were assessed by means of a 2-way repeated-measures ANOVA. A t test was used to compare hematocrit and responses to nitroprusside in vehicle-infused and Ang IIinfused rats. Responses to BMI or vehicle microinjection were determined as the difference between a 60-second baseline average and a 60-second average of data centered on the maximum increase. BMI and vehicle effects were compared by means of a 2-way ANOVA. A Newman-Keuls test was used for pairwise comparisons. When group variances were unequal, data were transformed through the use of a logarithmic function. Significance was defined as P<0.05. Voltage caused by noise was subtracted from all SNA data before statistical analysis.4,17
| Results |
|---|
|
|
|---|
|
As shown in Figure 2, 7 days of Ang II infusion significantly increased SNA, as indexed by the magnitude of the maximal depressor response to acute ganglionic blockade. Intravenous administration of chlorisondamine caused a significantly greater decrease (P<0.01) in MAP in Ang IItreated animals (117±3 to 42±1 mm Hg, n=5) than in vehicle-infused control rats (87±4 to 40±6 mm Hg, n=5). Likewise, there was a tendency for a greater fall in HR during ganglionic blockade in Ang IIinfused animals, but differences did not reach statistical significance (Figure 2). Overall, these data indicate that the contribution of SNA in maintaining arterial pressure was significantly increased in animals infused with Ang II. Notably, this effect does not appear to result from differences in extracellular fluid volume, since hematocrit was not significantly different in animals infused with vehicle (43.2±2, n=5) or Ang II (45±1, n=5).
|
Figure 3 shows the response to PVN microinjection of BMI (0.1 nmol) in a vehicle-infused (left) and an Ang IIinfused rat (right). Note that in the vehicle-infused animal, renal SNA, ABP, and HR increased promptly. Conversely, in the Ang IIinfused animal, PVN injection of BMI had little effect. Summary data in Figure 4 show that in vehicle-infused rats (n=6), PVN injection of BMI significantly increased (P<0.01) renal SNA by an average of 57±4%. In contrast, PVN injection of BMI had no effect on renal SNA (1±5.6%) in Ang IIinfused rats (n=6). Similarly, MAP (P<0.01) and HR (P<0.05) were significantly increased after PVN injection of BMI in vehicle-infused rats (n=7), but again, no significant effect was observed in Ang IIinfused animals (n=6) (Figure 4). Overall, these data indicate that Ang II infusion effectively attenuates renal sympathetic and cardiovascular responses to acute blockade of GABA-A receptors in the PVN.
|
|
It should be noted that differences in renal SNA, MAP, and HR responses to BMI in vehicle-infused and Ang IIinfused animals did not depend on the level of ongoing MAP or HR, since values just before BMI injection were not statistically different between groups (Figure 4). Furthermore, intravenous administration of nitroprusside increased renal SNA similarly in Ang IIinfused (169.7±3.2% baseline, n=6) and vehicle-infused (174±2.6% baseline, n=6) animals, indicating that Ang II treatment did not elevate basal SNA to a maximal level.
Given that microinjection of BMI outside the PVN and saline within the PVN were each without effect (P>0.1) on renal SNA, MAP, or HR in both vehicle-infused (n=6 per group) and Ang IIinfused (n=6 per group) animals, it appears that actions of BMI were specific to the PVN. On the basis of dye diffusion patterns, microinjection sites appeared to be confined to superior portions of the PVN near the dorsal cap region without penetrating the ependyma of the third cerebral ventricle (Figure 5). Anatomic control injections were verified at
1.8 mm lateral to midline and did not significantly invade lateral portions of the PVN.
|
| Discussion |
|---|
|
|
|---|
What is becoming increasingly clear is that GABAergic control of SNA within the PVN is subject to modulation in diseases accompanied by sympathoexcitation. Indeed, it has been reported that GABA-mediated control of SNA within the PVN is reduced in rats with salt-sensitive hypertension20 and heart failure.6 One interpretation of these data is that elevated plasma Ang II present in both disease models may lead to reduced GABAergic function within the PVN. Results from the present study suggest that GABA-A receptormediated control of PVN neuronal excitability may indeed be reduced by direct intravenous infusion of Ang II.
It should be noted that these results appear to conflict with those of a recent study by DiBona and Jones,23 in which it was reported that the renal SNA response to BMI injected into the PVN is enhanced in rats fed a low versus normal sodium diet. Given that anesthesia has been reported to alter cardiovascular responses to PVN stimulation,24 differences could involve the type of anesthetic used. Additionally, it may be that the difference in methods used to increase circulating Ang II in their study23 and ours make the results not directly comparable. For example, different levels of sodium intake and associated differences in extracellular fluid volume could underlie these disparate responses. This could be the case because volume-sensitive inputs to the PVN have been reported to inhibit discharge of PVN autonomic neurons25 and chemical lesions of PVN parvocellular cells have been shown to reduce the renal SNA response to acute volume expansion.26 Given that hematocrit was unchanged by the Ang II infusion in our study, it does not appear that blood volume was altered and therefore effects of Ang II infusion do not appear to have involved a change in volume-sensitive input to the PVN. Moreover, effects of aldosterone do not appear to contribute significantly to the findings of the current study, since values of plasma aldosterone return toward baseline within 48 hours during continuous infusion of Ang II.27
Acute and long-term Ang II treatments are well known to differentially modulate SNA and arterial pressure through both peripheral28 and central17,29 actions. In the current study, 7-day peripheral Ang II infusion produced a significant and sustained increase in MAP. Furthermore, this Ang II infusion protocol consistently increased SNA, as indexed by the depressor response to ganglionic blockade. As a result, experiments were performed after at least 7 days of continuous infusion. Although MAP was unchanged throughout the infusion protocol in vehicle-treated control rats, it increased after urethane/
-chloralose anesthesia, as reported previously.17 Nevertheless, responses to BMI injected in the PVN in vehicle-infused animals were similar to those reported in previous studies.46 It is important to note that anesthesia did not affect the elevated MAP produced by continuous Ang II infusion. Overall, the present findings suggest that GABAergic function within the PVN may be reduced in response to Ang II infusion.
In addition to GABA, the PVN receives input from Ang IIsensitive regions of the forebrain,9,13,15 which supply the PVN with Ang II immunoreactive fibers.10,15 Furthermore, electrophysiological studies demonstrate that intravenous Ang II can increase the discharge of PVN neurons13,30 through activation of locally expressed AT1 receptors.11 Collectively, these data indicate that the PVN is an important central site where information concerning the level of circulating Ang II is processed. Although numerous studies describe the individual importance of PVN GABAergic and Ang II systems in the regulation of cardiovascular function, fewer investigations4 have examined interactions between GABA and Ang II in the control of SNA.
One possible explanation for the observed ability of Ang II infusion to effectively abolish the renal SNA response to PVN GABA-A receptor blockade is that basal SNA could have been increased to a level that did not allow for a further increase to occur. One argument against this is that renal SNA did increase in Ang IIinfused animals when blood pressure was lowered with nitroprusside. Nevertheless, it remains possible that Ang II infusion could have maximally activated PVN neurons and thereby precluded any further increase in SNA during GABA-A receptor blockade. To address this possibility, additional studies will be needed to test whether PVN neuronal activity can be increased similarly in both treatment groups on exposure to an excitatory stimulus such as glutamate. Another possibility is that blunted responses to GABA-A receptor blockade in Ang IIinfused rats could reflect an overall reduction in GABA-A receptor tone. This could result either from removal of excitatory input to local GABAergic neurons or it could be due to a reduction in the number and/or affinity of local GABA-A receptors. The latter appears to be the case in the spontaneously hypertensive rat (SHR), where GABA-A receptor expression31 and affinity32 in the hypothalamus are reduced compared with Wistar-Kyoto control rats, as are typical cardiovascular responses to hypothalamic injection of BMI.46 This does not appear to be a general mechanism, however, since rats with salt-sensitive hypertension showed no change in either GABA-A receptor expression or binding affinity compared with control rats.33 It is important to note that this discrepancy suggests that hypertension per se is not likely to explain or be explained by changes in GABA-A receptor function in the hypothalamus.
Despite differences noted in the aforementioned models, evidence indicates that Ang II can inhibit GABA release by acting presynaptically and can alter responses to GABA-A receptor activation by acting postsynaptically. That Ang II and GABA are colocalized in axon terminals in the subfornical organ34 supports both possibilities. Regarding presynaptic actions, Ang II has been reported to modulate GABA release in rat hippocampal neurons.35 Moreover, a recent in vitro patch clamp study by Li et al19 demonstrated that Ang II stimulates spinally projecting PVN neurons by acting presynaptically to reduce the frequency of inhibitory postsynaptic currents. In the same study, it was reported that AT1 receptors are expressed on putative synaptic terminals within the PVN. If such a process were activated under conditions of the present study, an Ang IIstimulated reduction in GABA release could explain the blunted renal SNA response observed during acute GABA-A receptor blockade.
In addition to presynaptic modulation of GABA release described above, Ang II can also act postsynaptically to modulate GABA-A receptormediated inhibition. For example, Cl- conductance through GABA-A receptor channels is reduced after phosphorylation by protein kinase C,36,37 a major signaling intermediate activated by stimulation of postsynaptic AT1 receptors.38 Furthermore, an in vitro patch-clamp study by Latchford and Ferguson18 recently reported that Ang IIinduced excitation of PVN neurosecretory neurons is enhanced by blockade of GABA-A receptors, suggesting that GABAergic feedback restrains activation of these neurons by Ang II. Moreover, data from their study18 indicate that Ang II increases GABAergic input that buffers the Ang IIinduced membrane depolarization. These findings describe a mechanism of postsynaptic interaction between Ang II and GABA in which Ang II input supplies at least a portion of the GABAergic tone present among nonautonomic neurons of the PVN. A recent in vivo study by Chen et al4 suggests that a similar process could occur among autonomic neurons, since prior blockade of AT1 receptors in the PVN nearly prevented the renal SNA response to subsequent blockade of PVN GABA-A receptors. It should be emphasized, however, that for feedback activation of local GABAergic neurons to contribute to the findings of the present study, it would seem necessary for this feedback restraint to be diminished after 7 days of Ang II infusion. Under this condition, GABA-mediated inhibition in the PVN could be reduced and therefore could result in attenuated cardiovascular and sympathetic responses to acute GABA-A receptor blockade. Regardless of the underlying mechanism, recent studies in both nonautonomic and sympathetic regulatory neurons of the PVN highlight the potential importance of local Ang IIGABA interactions in the control of cardiovascular and sympathetic function.
In conclusion, results from this study indicate that 7 days of intravenous infusion of Ang II significantly attenuates the increase in renal SNA, MAP, and HR induced by acute GABA-A receptor blockade in the PVN. These data as well as literature evidence6,20 raise the possibility that Ang II may act centrally to reduce GABAergic function in the PVN, thereby providing a potential mechanism by which sympathetic activity could become elevated in cardiovascular diseases such as congestive heart failure6 and arterial hypertension.20
| Acknowledgments |
|---|
Received July 31, 2003; first decision August 19, 2003; accepted October 8, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Maliszewska-Scislo, H. Chen, R. A. Augustyniak, D. Seth, and N. F. Rossi Subfornical organ differentially modulates baroreflex function in normotensive and two-kidney, one-clip hypertensive rats Am J Physiol Regulatory Integrative Comp Physiol, September 1, 2008; 295(3): R741 - R750. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. McMullan, A. K. Goodchild, and P. M. Pilowsky Circulating angiotensin II attenuates the sympathetic baroreflex by reducing the barosensitivity of medullary cardiovascular neurones in the rat J. Physiol., July 15, 2007; 582(2): 711 - 722. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. B. Park, S. Skalska, S. Son, and J. E. Stern Dual GABAA receptor-mediated inhibition in rat presympathetic paraventricular nucleus neurons J. Physiol., July 15, 2007; 582(2): 539 - 551. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-L. Pan Brain Angiotensin II and Synaptic Transmission Neuroscientist, October 1, 2004; 10(5): 422 - 431. [Abstract] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2003 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |