- Email: [email protected]
GABA receptor subtypes involved in the neuronal mechanisms of baroreceptor reflex in the nucleus tractus solitarii of rabbits
Journal of the Autonomic Nervous System, 43 (1993) 27-36
27
© 1993 Elsevier Science Publishers B.V. All rights reserved 0165-1838/93/$06.00 JANS 01374
GABA receptor subtypes involved in the neuronal mechanisms of baroreceptor reflex in the nucleus tractus solitarii of rabbits Masahiko Suzuki, Tomoya Kuramochi and Toshio Suga Department of Pharmacology, Saitama Medical School, Moroyama, Iruma-gun, Saitama 350-04, Japan (Received 3 August 1992) (Revision received 20 October 1992) (Accepted 21 October 1992)
Key words: Baroreceptor reflex; GABAergic system; Microiontophoresis; Nucleus tractus solitarii; Single unit activity Abstract The role of GABA A and GABA B receptors in neuronal mechanisms of the baroreceptor reflex in the nucleus tractus solitarii (NTS) of the anesthetized and immobilized rabbits were investigated using a microiontophoretic technique. Baroreceptive neurons (BRNs) activated or depressed by baroreceptor stimulation (phenylephrine, 10 p.g/kg, i.v.) were identified in the NTS (activated BRN (A-BRN) and depressed BRN (D-BRN), respectively). The GABA A antagonist bicuculline (40-80 hA) increased spontaneous activities of these neurons, but the GABA B antagonist phaclofen (80-160 nA) did not. Evoked responses of A-BRNs were potentiated by bicuculline and phaclofen, while the responses of D-BRNs were not clearly affected by these drugs. These results suggest that most of A- and D-BRNs are tonically inhibited by endogenous GABA acting on GABA A receptors, but not on GABA B receptors, and that GABAergic mechanisms suppressively modulate the baroreceptor reflex acting on GABA A and GABA B receptors of A-BRNs, but not of D-BRNs.
Introduction
The nucleus tractus solitarii (NTS) is the site of the first synapse in the circulatory, respiratory and other visceral reflex pathways [13]. This region contains a large amount of y-aminobutyric acid (GABA) and specific GABA receptors [2,11,17]. Local injections of GABA and the GABA A agonist muscimol into the NTS produce
Correspondence to: M. Suzuki, Department of Pharmacology, Saitama Medical School, Moroyama, Iruma-gun, Saitama 35004, Japan.
hypertension, and these effects were antagonized by the GABA A antagonist bicuculline [3,15,24]. Local injections of muscimol also attenuated the refex hypotension and bradycardia produced by the electrical stimulation of the aortic depressor nerve. In contrast, bicuculline potentiated those reflex effects [16]. In addition, the local injection of a GABA B agonist into the NTS also increased the arterial blood pressure (AP) as in the case of GABA A agonists [3,9,21,24]. Thus it is considered that the GABAergic system modulates the neuronal activity of the baroreflex pathway in the NTS. A recent study indicates that the pressor response evoked by the local injection of the GABA
28 uptake inhibitor nipecotic acid into the NTS is attenuated by the selective GABA B antagonist phaclofen, but not by the selective GABA A antagonist bicuculline [23]. These results suggest that GABA tonically released in the NTS acts only on the GABA a receptor to increase AP. It has been revealed in electrophysiological studies that stimulation of the hypothalamic defense area in cats suppresses the baroreceptor reflex [6,20]. This stimulation also produces inhibition of tonic and responsive activities of the NTS neurons receiving a baroreceptor input, and the transmitter for this inhibition is estimated to be GABA, because bicuculline antagonizes this inhibitory response elicited by the hypothalamic defense area stimulation [12]. Furthermore, it has been demonstrated that bicuculline increases the evoked activities of the NTS neurons by electrical afferent stimulation of the cardiac or pulmonary vagal branch, the aortic nerve [1], or the carotid sinus nerve [18]. In the NTS, two types of neurons that respond to baroreceptive input have been identified with extracellular recording [8,22]: one is activated and another is depressed by baroreceptor stimulation (activated baroreceptive neuron (A-BRN) and depressed baroreceptive neuron (D-BRN), respectively). However, it has not been clarified yet how GABAergic mechanism functions on the ABRN or D-BRN. The aim of the present study was to elucidate the participation of GABAergic mechanisms in the baroreflex modulation in the NTS neurons that were identified as baroreceptive in nature (i.e., A- and D-BRN), using the microiontophoretic technique, and, in addition, to clarify the role of GABA receptor subtypes, i.e., GABA A and GABA a receptors in the regulatory mechanisms of the BRNs.
Materials and Methods
Preparation of animals Male albino rabbits (Japanese White) weighing 2.8-3.3 kg were anesthetized with urethane-chloralose (500 m g / k g and 50 mg/kg, i.v., respectively). An additional dose of urethane was given
when necessary. Cannulae were inserted into the trachea, the femoral artery and the femoral vein. After fixing the rabbits to a stereotaxic instrument (Narishige, SN-2) in a prone position, the animals were immobilized with gallamine triethiodide (initially 10 mg/kg, i.v., thereafter 5 m g / k g every 1-1.5 hr, Sigma) and artificially ventilated (Natsume, KN-56). The end-tidal CO 2 was constantly monitored (Nihondenki San-ei, Respina IH26) and was maintained at 4.0-4.5% by changing volume of the respirator. The atlanto-occipital membrane was exposed, cut at the midline and then retracted to expose the dorsal surface of the medulla and the surface was covered with warm mineral oil. To reduce respiratory movement of the brain stem, pneumothorax was induced in all experiments. Bilateral vagal nerves were cut in the neck. Rectal temperature was maintained at 37-38°C by an isothermal pad (Braintree, Deltaphase). AP was recorded from the abdominal aorta cannulated through the femoral artery and connected to a pressure transducer (Nihon Kohden, MPU-0.5). Heart rate (HR) was computed from AP pulses by a tachometer unit (Nihondenki Sanei, Type 1236). Baroreceptor stimulation was carried out by the injection of phenylephrine (10 ~zg/kg) through the femoral vein.
Recording of single unit activity and microiontophoretic application The multibarrel electrode, which was made by gluing together a single glass micropipette with angled tip and a glass 5-barreled micropipette, was used. The single pipette was filled with 0.5 M sodium acetate and 2% Pontamine Sky Blue (Brilliant Blue 6B, pH 7.7, Tokyokasei) for extracellular recording of the single neuronal activity and marking of the recording sites (tip diameter 1 ~m, 3-8 Mg2). Four barrels of the 5-barreled pipette (tip diameter 3-6 /~m, 5-15 MS2) were filled with the following drugs for microiontophoretic application: GABA (1 M, pH 4.0, Sigma), sodium glutamate (0.5 M, pH 8.0, Wako), L-bicuculline methiodide (20 mM, pH 4.0, Sigma), phaclofen (13.3 mM, pH 4.0, Tocris Neuramin) and De-baclofen (20 mM, pH 4.0, Sigma). These
29
drugs were applied at up to 160 nA using a microiontophoretic equipment (Dia Medical, DPI-30). Automatic current balancing was held through the remaining barrel of the 5-barreled micropipette filled with 3 M NaCI. Electrode penetration was performed stereotaxically in 0.5-1.0 mm caudal to the rostral border of the area postrema, in 1-2.0 mm lateral to the midline and within 2 mm deep to the dorsal surface of the medulla. Cells displaying respiratory rhythmic activity were excluded from the analysis [8]. Drug actions to cells were defined as an increase or decrease if they displayed changes of at least 30% in control activities. The single unit activity was amplified (Nihon Kohden, 50-3000 Hz, MEZ-7200 and AVH-10) and displayed on an ocilloscope (Nihon Kohden, VC-10). The discharge frequency was counted by a spike counter with 1 s bin (Dia Medical, DSE325P) and displayed on a polygraph (Nihondenki San-ei, Rectigraph-8S) together with AP and HR. Some recording sites were checked in histological sections (stained with Neutral Red) of the brain stem after the experiment.
Results
Forty-nine neurons in the NTS which responded to transient elevations of AP were examined. Their unit activities were confirmed to be derived from the soma of neurons by activation of their activities to the microiontophoretically applied sodium glutamate (5-40 nA) [10]. These neurons exhibited randomized discharge patterns which bore no obvious relationship to spontaneous fluctuations in relation to HR, and had basal firing frequency between 5 and 40 spikes/s. The neuron was classified as A- or D-BRN, if it displayed an increase or a decrease of at least 30% in baseline spike activity in response to transient elevation of AP produced by the intravenous injection of phenylephrine (10 /zg/kg) (Fig. 1). Twenty-six out of 49 neurons were identified as A-BRN and 23 neurons were identified as D-BRN.
Effects of GABA on the spontaneous activities of BRNs To determine if the spontaneous activities of the two classes of BRNs show different sensitivities to GABA, GABA was applied microiontophoretically to these neurons (2-80 nA). GABA decreased the spontaneous activities dose-dependently in all 26 A-BRNs and 22 of 23 D-BRNs tested. In each 48 GABA-sensitive neurons, the inhibitory effects of GABA were tested against the selective GABA A antagonist bicuculline and selective GABA B antagonist phaclofen [14,25]. Inhibitory effects of the microiontophoretically applied GABA on the BRNs were markedly prevented by simultaneously applied bicuculline (1020 nA) and partially antagonized by phaclofen (20-40 nA) (Fig. 2). These results indicate that A- and D-BRNs were affected by both GABA A and GABA B receptors which were thus involved in mediating the inhibitory action. Tonic effects of GABAergic system on BRNs activities To determine if the GABAergic system in the NTS tonically affects A- and D-BRNs activities, bicuculline and phaclofen were applied microiontophoretically to those neurons. Low doses of bicuculline (10-20 nA) had no effect on these neurons, but high doses (40-80 nA) increased the
A-BRN
D-BRN
o
AP
2oo mmHg 5O 3O s
Fig. 1. Effects of baroreceptor stimulation on the discharge rate of neurons in the NTS. The left and right traces show the baroreceptive neurons whose spontaneous activities were activated and depressed by the baroreceptor stimulation, respectively (activated baroreceptive neuron (A-BRN) and depressed baroreceptive neuron (D-BRN), respectively). Upper traces indicate unit activities/s (Unit). Lower traces indicate arterial blood pressure (AP). Arrows indicate intravenous injections of phenylephrine (10/xg/kg).
30
w GABA
20
~ GABA
40
m GABA
80 m
GABA 10
m
m
Bic 20
Bic 40 spikes/s 20
GAB-'~80
spikes/s
GAB-'~ 80
t0
Bic 10 "G-ABA80 Bic 20 ~'ABA 80
Phac 20 ~'ABA 80 Phac ~ A
80
3-6% Fig. 2. Blockade of neuronal responses to G A B A by bicuculline and phaclofen ( G A B A A and G A B A B antagonists, respectively). In 48 out of 49 BRNs examined, G A B A decreased the spontaneous activity. This figure shows the typical effects of GABA, bicuculline and phaclofen on the spontaneous activity of the same D-BRN. The traces indicate unit activities/s (Unit) and solid bars indicate periods of application of drugs. G A B A (20-80 nA) dose-dependently decreased the spontaneous activity (top traces). The neuronal response of G A B A (80 nA) was dose-dependently antagonized by bicuculline (Bic, 10-20 nA) and phaclofen (Phac, 20-40 hA) (middle and bottom traces, respectively).
spontaneous activities in 11 of 14 A-BRNs and 9 of 11 D-BRNs. However, phaclofen (up to 160 nA) had no effect on them in 14 of 15 A-BRNs and 12 of 13 D-BRNs, although their spontaneous activities were suppressed by the G A B A ~ agonist baclofen (40-80 nA). These results suggest that most of A- and D-BRNs were tonically inhibited by endogenous G A B A acting on G A B A A receptors, but not on G A B A n receptors. The typical effects on the same neuron were shown in Fig. 3. In the study on the same neurons, the neurons whose spontaneous activities were increased by bicuculline and unaffected by phaclofen were 10 of 13 A-BRNs and 7 of 9 D-BRNs. The remaining 3 neurons of A-BRNs and 2 of D-BRNs were not affected by either drug.
Effects of bicuculline and phaclofen on the evoked responses of BRNs Effects of bicuculline and phaclofen on the evoked response of A- and D-BRNs elicited by the baroreceptor stimulation were examined. Evoked responses of A-BRNs (10 of 12 neurons)
Bac 80
Phac g0
Phae 160 - 30 s Fig. 3. Effects of bicuculline and phaclofen on the spontaneous activity of BRN. In 20 out of 25 BRNs, high doses of bicucu0ine (Bic, 20-40 nA) increased the discharge rate of BRNs, while phaclofen (Phac, 80-160 nA) had no effect on 26 out of 28 B R N s (see Fig. 6). This figure is displayed in the same m a n n e r as Fig. 2 and shows the typical effects of the two drugs on the spontaneous activity of the same D-BRN. Following G A B A (10 nA), bicuculline was applied at 20 and 40 nA. Low dose (20 nA) bicuculline had no effect on the spontaneous activity, but a high dose (40 nA) increased it (top traces). Following baclofen (Bac, 80 nA), phaclofen was applied at 80 and 160 nA. Baclofen decreased the spontaneous activity, while phaclofen had no effect on the spontaneous activity up to 160 nA (bottom traces).
were potentiated by bicuculline (10-40 nA), while the responses of D-BRNs (8 of 10 neurons) were not clearly affected by it (Fig. 4). Although the A-BRN Unit
AP
1' Unit
1'
20
D-BRN t 20 spikes/s J 0
AP
150 mmHg 50 1.
1.
20
30 s Fig. 4. Effects of bicuculline on the evoked responses of Aand D-BRNs. Evoked responses of A - B R N s elicited by baroreceptor stimulation were potentiated by bicuculline (Bic, 10-40 nA), while the responses of D-BRNs were not clearly affected by it (see Fig. 6). Arrows and solid bars indicate intravenous injections of phenylephrine ( 1 0 / ~ g / k g ) and periods of application of bicuculline, respectively. Unit: unit activities/s, AP: arterial blood pressure.
31
Unit
A-BRN
spontaneous firing of D-BRN was increased by bicuculline, in the case shown in Fig. 4, the inhibitory response produced by the baroreceptor stimulation was hardly changed under the application of bicuculline compared to the control. Similarly, evoked responses of A-BRNs (7 of 9 neurons) were potentiated by phaclofen (40-160 nA), while the responses of D-BRNs (9 of 10 neurons) were not clearly affected by it (Fig. 5). These results suggest that GABAergic mechanisms suppressively modulated the baroreceptor reflex acting on GABA A and GABA a receptors of A-BRNs, but not of D-BRNs. In the study on the same neurons, the evoked responses in 5 of 8 A-BRNs were potentiated by both bicuculline and phaclofen, and the responses in 4 of 5 D-BRNs were unaffected by either drug. The remaining neurons of A- and D-BRNs were variously affected by those drugs, i.e., in A-BRNs (3 neurons), one neuron was
t
AP
t
t
Unit
a0
D-BRN spikes/s 0 mmNg 50
t
so
_ _
30
s
Fig. 5. Effects of phaclofen on the evoked responses of A- and D-BRNs. Evoked responses of A-BRNs elicited by baroreceptor stimulation were potentiated by phaclofen (Phac, 40-160 nA), while the responses of D-BRNs were not clearly affected by it (see Fig. 6). This figure is displayed in the same manner as Fig. 4.
Bicuculline Spontaneous o activity
5
10
15
I ....
Phaclofen 20
I ....
0
I . . . . . .
5
10
I ....
15
I ....
20
I ....
25
I ,,
,,
(N)
I,,
Increase ~ Decrease i ~ No effect
] I
Evoked response Potentiation
5 . . . .
i
10 . . . .
i
15 . . . .
I
0
5
10
15
(N)
i
......
i ......
'
Inhibition No
effect
I
D : A-BRN []
: D-BRN
Fig. 6. Summary of responses of A- and D-BRNs to bicuculline and phaclofen. This graph shows the number of neurons whose spontaneous activities and evoked responses produced by phenylephrine (10/,~g/kg, i.v.) were changed by these antagonists at least 30% vs. the control level, thus indicating GABA A and GABA a receptor sensitivities in each neuron group. Upper and lower graphs indicate effects of drugs on spontaneous activities and evoked responses of BRNs, respectively. Left and right sides indicate effects of bicucuUine and phaclofen, respectively. Open and hatched columns indicate A- and D-BRNs, respectively. The difference between the numbers of BRNs whose spontaneous activities were increased by bicuculline and phaclofen was significant (Fisher's exact test, P < 0.01, two-tailed). The difference between the numbers of A- and D-BRNs whose evoked responses were potentiated by bicuculline and phaclofen was significant (Fisher's exact test, P < 0.01, two-tailed, in each drug).
32 unaffected by either drug, another was depressed and unaffected, and the third was unaffected and potentiated, by bicuculline and phaclofen, respectively. In D-BRN (one neuron), it was potentiated and unaffected by bicuculline and phaclofen, respectively.
Summary of responses of A- and D-BRNs to bicuculline and phaclofen In Fig. 6, the effects of bicuculline and phaclofen on the spontaneous activities and evoked responses of A- and D-BRNs (upper and lower traces, respectively) were summarized. The spontaneous activities in 20 of 25 BRNs were increased by bicuculline and the others were unaffected, whereas phaclofen increased the spontaneous activities in only 2 of 28 BRNs and did not affect the others. This difference of the effects of bicuculline and phaclofen was significant (Fisher's exact test, P < 0.01, two-tailed). Both bicuculline and phaclofen potentiated most of the evoked responses of A-BRNs, but did not affect those of D-BRNs. Precisely, bicuculline potentiated 10 of 12 A-BRNs and phaclofen 7 of 9 A-BRNs, whereas 8 of 10 D-BRNs were not affected by bicuculline and 9 of 10 D-BRNs were also not affected by phaclofen. The difference in the effects of the antagonists on the evoked responses between A- and D-BRNs was significant (Fisher's exact test, P < 0.01, two-tailed). In this study, both A- and D-BRNs were found in the dorsal and medial regions of NTS, but there were no significant differences in localization of these neurons.
Discussion
We demonstrated the following differences of GABAergic modulation in the baroreceptor reflex in the NTS due to their receptor subtypes. First, the spontaneous activities of A- and DBRNs were increased by the GABA A antagonist bicuculline, but not by the GABA B antagonist phaclofen. These results indicate that GABAergic system in the NTS tonically suppresses A- and D-BRNs by acting on GABA A receptor, but not on G A B A u receptor. Second, the evoked re-
sponses of A-BRNs were potentiated by bicuculline and phaclofen, but those of D-BRNs were not affected by both drugs. These results indicate that the GABAergic system exerts the inhibitory modulation to the baroreceptor reflex by acting on GABA A and G A B A B receptors of A-BRNs, but not of D-BRNs. In the hitherto reported pharmacological study, it has been demonstrated that GABA A and GABA B receptors are involved in cardiovascular regulation in the NTS. Namely, the microinjection of GABA and muscimol, a GABA A agonist, into the NTS produces hypertension and tachycardia, and these effects are antagonized by bicuculline [3,15]. Baclofen, a GABA u agonist, also increases AP and H R and these effects are not affected by bicuculline [3,21,24], but antagonized by phaclofen and 2-OH-saclofen, GABA B antagonists [9]. In addition, the reflex bradycardia elicited by AP elevation and the reflex hypotension and bradycardia by electrical stimulation of the aortic depressor nerve are inhibited by muscimol and baclofen [9,16], and the effects of baclofen are blocked by phaclofen and 2-OH-saclofen [9]. Here we demonstrate the inhibitory effects of GABA on the spontaneous activities of A- and D-BRNs and that these effects are antagonized by bicuculline and phaclofen. The inhibitory effect of baclofen on the spontaneous activities is also demonstrated (see Fig. 3). Additionally, the high dose of bicuculline augmented the spontaneous activities of these neurons. On the contrary, phaclofen has no effect on these spontaneous activities, indicating that the involvement of the GABA u receptor is excluded from the tonically inhibitory modulation with GABAergic system in the NTS. Indeed, phaclofen microinjected into the NTS scarcely changed AP and H R in spite of blocking the cardiovascular effects of baclofen [9,23]. Sved and Sved [23] have reported that the cardiovascular effects of the microinjected nipecoptic acid, a GABA uptake inhibitor, into the NTS of rats are antagonized by phaclofen but not by bicuculline. Thus, they have proposed that tonically released GABA in the NTS acts specifically on GABA u receptors, but not on GABA A
33 receptors, to increase AP and HR. However, in this study in rabbits, bicuculline increased the spontaneous activities of A- and D-BRNs, but phaclofen did not. In addition, both bicuculline and phaclofen potentiated the evoked response of A-BRNs. These results indicate that both of GABA A and GABA B receptors are involved in the cardiovascular regulation in the NTS. Although this discrepancy might be based on the species difference used, it has been reported that the microinjection of bicuculline into the NTS in rats [15] and cats [3] decreased AP and HR, and the evoked responses in the NTS in cats by baroreceptor afferent stimulations were potentiated by the microiontophoretically applied bicuculline [12,18]. These results seem to be more comparable with our data than those of Sved and Sved. We, moreover, observed that the microiontophoretically applied nipecotic acid decreased the spontaneous activities of A- and D-BRNs in rabbits, and these effects were antagonized by both bicuculline and phaclofen in preliminary studies (unpublished observations). After all, we cannot explain these differences clearly yet. Surprisingly, phaclofen potentiated the evoked responses of A-BRNs, while it had no effect on the spontaneous activities of A- and D-BRNs. There are at least three possibilities for these phenomena. First, phaclofen could not have substantially penetrated into the synaptic cleft enough to block the inhibitory effects of the tonically released GABA completely, while in the evoked responses of A-BRNs, one small part of phaclofen applied might have partially exerted the potentiating effects. Second, as the binding affinity of phaclofen with GABA B receptor in the brain membranes is low [4], phaclofen might not have blocked the effect of the tonically released GABA completely. Third, there might be two types of GABAergic systems in the NTS: one is tonically inhibiting A- and D-BRNs through the GABA A receptor and, moreover, is activated from the other nucleus, e.g., hypothalamus defense area [12,20], and the other is activated by the baroreceptor inputs and affects both GABA A and GABA B receptors of A-BRNs. According to the latter mechanism, the presence of the phasic inhibition induced by baroreceptor afferent would
be postulated, because in in vitro [5] and in vivo [19] studies, the neurons that evoked IPSP or EPSP-IPSP by electrical stimulation of the afferents were demonstrated. In the present study, the inhibitory responses of D-BRNs evoked by baroreceptor stimulation could not be affected by bicuculline and phaclofen. This indicates that GABA may not be included in this inhibitory response of D-BRNs by baroreceptor stimulation. Mifflin et al. [19] demonstrated the three types of baroreceptive neurons referred to the synaptic responses with a stable intracellular recording. Namely, these neurons produced EPSP, EPSPIPSP or IPSP by electrical stimulation of the carotid sinus nerve. The IPSP was reversed to a depolarizing potential either by applying hyperpolarizing DC current or an intracellular injection of CI-, suggesting that it was mediated by a Cl--dependent process. It is well known that GABA, acting on GABA A receptor, exerts an inhibitory response by a C1--dependent process, thus GABA is the most likely candidate of this inhibitory transmitter. D-BRNs in this study accordingly could be comparable to the neuron which produces IPSP for its inhibitory response, but we failed to demonstrate the antagonistic effect of bicuculline on the evoked inhibitory response of D-BRNs. Although we used a dose of bicuculline that almost completely antagonized the effects of the exogenously applied GABA on the BRNs, it may not penetrate in the synaptic cleft sufficiently and, therefore, may not be able to antagonize the inhibitory action of the endogenously released GABA. This possibility could not be excluded, but the inhibitory response of the D-BRNs was not affected by the dose of bicuculline which somewhat augmented their spontaneous activities (see Fig. 4). To clarify the transmitter that induces the evoked inhibitory responses of D-BRNs, further studies will be needed. Although the evoked responses of A-BRNs were potentiated by bicuculline and phaclofen, the sites of actions of these drugs are not clarified. The following possibilities can be considered according to previous reports [4,7,9]. Firstly, both GABA A and GABA B receptors are located on the postsynaptic membrane of A-BRNs; and sec-
34
ondly, one of the two (it may be G A B A B receptor) locates on the presynaptic membrane of the excitatory input to A - B R N s with the other receptor (it may be G A B A A receptor) on the postsynaptic membrane [4]. Florentino et al. [9] have reported that the cardiovascular effects of stimulation of G A B A B receptors in the NTS are due, at least in part, to the presynaptic inhibition of the baroreceptor afferents. On the other hand, it has been observed that phaclofen selectively inhibits responses mediated by postsynaptic but not presynaptic G A B A B receptor in CA1 hippocampal pyramidal neurons in vitro [7]. Therefore, both G A B A receptors might be located on the postsynaptic membrane in the NTS, but further examination is necessary to determine the locations of G A B A A and G A B A B receptors. In summary, the present findings provide pharmacological evidence of the neuronal modulation of G A B A by acting on G A B A A and G A B A B receptors in the NTS. These results suggest that most of A- and D - B R N s are tonically inhibited by e n d o g e n o u s G A B A acting on G A B A A receptors, but not on G A B A B receptors, and that GABAergic mechanisms suppressively modulate baroreceptor reflex acting on G A B A A and G A B A a receptors of A-BRNs, but not of D-BRNs.
6
7
8
9
10
II
12
13
14
15
References 16 1 Bennett, J.A., McWilliam, P.N. and Shepheard, S.L., A y-aminobutyric-acid-mediated inhibition of neurones in the nucleus tractus solitarius of the cat, J. Physiol. (Lond.), 392 (1987) 417-430. 2 Blessing, W.W., Oertel, W.H. and Willoughby, J.O., Glutamic acid decarboxylase immunoreactivity is present in perikarya of neurons in nucleus tractus solitarius of rat, Brain Res., 322 (1984) 346-350. 3 Bousquet, P., Feldman, J., Bloch, R. and Schwartz, J., Evidence for a neuromodulatory role of GABA at the first synapse of the baroreceptor reflex pathway. Effects of GABA derivatives injected into the NTS, NaunynSchmiedeberg's Arch. Pharmacol., 319 (1982) 168-171. 4 Bowery, N., GABA B receptors and their significance in mammalian pharmacology, Trends Pharmacol. Sci., 10 (1989) 401-407. 5 Champagnat, J., Denavit-Saubi6, M., Grant, K. and Shen, K.F., Organization of synaptic transmission in the mam-
17
18
19
20
malian solitary complex, studied in vitro, J. Physiol. (Lond.), 381 (1986) 551-573. Coote, J.H., Hilton, S.M. and Perez-Gonzalez, J.F., Inhibition of the baroreceptor reflex on stimulation in the brain stem defense centre, J. Physiol. (Lond.), 288 (1979) 549560. Dutar, P. and Nicoll, R.A., Pre- and postsynaptic GABA B receptors in the hypocampus have different pharmacological properties, Neuron, 1 (1988) 585-591. Feldman, P.D. and Moises, H.C., Adrenergic responses of baroreceptive cells in the nucleus tractus solitarii of the rat: a microiontophoretic study, Brain Res., 420 (1987) 351-361. Florentino, A., Varga, K. and Kunos, G., Mechanism of the cardiovascular effects of GABA B receptor activation in the nucleus tractus solitarii of the rat, Brain Res., 535 (1990) 264-270. Fries, W. and Zieglg~insberger, W., A method to discriminate axonal from cellbody activity and to analyse "silent cells", Exp. Brain Res., 21 (1974) 441-445. Gale, K., Hamilton, B.L., Brown, S.C., Norman, W.P., Souza, J.D. and Gills, R.A., GABA and specific GABA binding sites in brain nuclei associated with vagal outflow, Brain Res. Bull., 5, Suppl. 2 (1980) 325-328. Jordan, D., Mifflin, S.W. and Spyer, K.M., Hypothalamic inhibition of neurones in the nucleus tractus solitarius of the cat is GABA mediated, J. Physiol. (Lond.), 399 (1988) 389-404. Jordan, D. and Spyer, K.M., Brainstem integration of cardiovascular and pulmonary afferent activity, Prog. Brain Res., 67 (1986) 295-314. Kerr, D.I.B., Ong, J., Prager, R.H., Gynther, B.D. and Curtis, D.R., Phaclofen: a peripheral and central baclofen antagonist, Brain Res., 405 (1987) 150-154. Kubo, T. and Kihara, M., Evidence for the presence of GABAergic and glycine-like systems responsible for cardiovascular control in the nucleus tractus solitarii of the rat, Neurosci. Lett., 74 (1987) 331-336. Kubo, T. and Kihara, M., Evidence for y-aminobutyric acid receptor-mediated modulation of the aortic baroreceptor reflex in the nucleus tractus solitarii of the rat, Neurosci. Lett., 89 (1988) 156-160. Maley, B. and Newton, B.W., Immunohistochemistry of y-amino butyric acid in the cat nucleus tractus solitarius, Brain Res., 330 (1985) 364-368. McWilliam, P.N. and Shepheard, S.L., A GABA-mediated inhibition of neurones in the nucleus tractus solitarius of the cat that respond to electrical stimulation of the carotid sinus nerve, Neurosci. Lett., 94 (1988) 321-326. Mifflin, S.W., Spyer, K.M. and Withington-Wray, D.J., Baroreceptor inputs to the nucleus tractus solitarius in the cat: postsynaptic actions and the influence of respiration, J. Physiol. (Lond.), 399 (1988) 349-367. Mifflin, S.W., Spyer, K.M. and Withington-Wray, D.J., Baroreceptor inputs to the nucleus tractus solitarius in the cat: modulation by the hypothalamus, J. Physiol. (Lond.), 399 (1988) 369-387.
35 21 Persson, B., A hypertensive response to baclofen in the nucleus tractus solitarii in rats, J. Pharm. Pharmacol., 33 (1981) 226-231. 22 Suzuki, M., Investigation on the neuronal mechanisms of baro- and chemoreceptor reflexes in the nucleus tractus solitarii: participation of adrenergic and opioid peptidergic system, J. Saitama Med. School, 17 (1990) 31-43. 23 Sved, A.F. and Sved, J.C., Endogenous GABA acts on GABA n receptors in nucleus tractus solitarius to increase blood pressure, Brain Res., 526 (1990) 235-240.
24 Sved, J.C. and Sved, A.F., Cardiovascular responses elicited by y-aminobutyric acid in the nucleus tractus solitarius: evidence for action at the GABA a receptor, Neuropharmacology, 28 (1989) 515-520. 25 Wiillner, U., Klockgether, T. and Sontag, K.-H., Phaclofen antagonizes the depressant effect of baclofen on spinal reflex transmission in rats, Brain Res., 496 (1989) 341-344.
27
© 1993 Elsevier Science Publishers B.V. All rights reserved 0165-1838/93/$06.00 JANS 01374
GABA receptor subtypes involved in the neuronal mechanisms of baroreceptor reflex in the nucleus tractus solitarii of rabbits Masahiko Suzuki, Tomoya Kuramochi and Toshio Suga Department of Pharmacology, Saitama Medical School, Moroyama, Iruma-gun, Saitama 350-04, Japan (Received 3 August 1992) (Revision received 20 October 1992) (Accepted 21 October 1992)
Key words: Baroreceptor reflex; GABAergic system; Microiontophoresis; Nucleus tractus solitarii; Single unit activity Abstract The role of GABA A and GABA B receptors in neuronal mechanisms of the baroreceptor reflex in the nucleus tractus solitarii (NTS) of the anesthetized and immobilized rabbits were investigated using a microiontophoretic technique. Baroreceptive neurons (BRNs) activated or depressed by baroreceptor stimulation (phenylephrine, 10 p.g/kg, i.v.) were identified in the NTS (activated BRN (A-BRN) and depressed BRN (D-BRN), respectively). The GABA A antagonist bicuculline (40-80 hA) increased spontaneous activities of these neurons, but the GABA B antagonist phaclofen (80-160 nA) did not. Evoked responses of A-BRNs were potentiated by bicuculline and phaclofen, while the responses of D-BRNs were not clearly affected by these drugs. These results suggest that most of A- and D-BRNs are tonically inhibited by endogenous GABA acting on GABA A receptors, but not on GABA B receptors, and that GABAergic mechanisms suppressively modulate the baroreceptor reflex acting on GABA A and GABA B receptors of A-BRNs, but not of D-BRNs.
Introduction
The nucleus tractus solitarii (NTS) is the site of the first synapse in the circulatory, respiratory and other visceral reflex pathways [13]. This region contains a large amount of y-aminobutyric acid (GABA) and specific GABA receptors [2,11,17]. Local injections of GABA and the GABA A agonist muscimol into the NTS produce
Correspondence to: M. Suzuki, Department of Pharmacology, Saitama Medical School, Moroyama, Iruma-gun, Saitama 35004, Japan.
hypertension, and these effects were antagonized by the GABA A antagonist bicuculline [3,15,24]. Local injections of muscimol also attenuated the refex hypotension and bradycardia produced by the electrical stimulation of the aortic depressor nerve. In contrast, bicuculline potentiated those reflex effects [16]. In addition, the local injection of a GABA B agonist into the NTS also increased the arterial blood pressure (AP) as in the case of GABA A agonists [3,9,21,24]. Thus it is considered that the GABAergic system modulates the neuronal activity of the baroreflex pathway in the NTS. A recent study indicates that the pressor response evoked by the local injection of the GABA
28 uptake inhibitor nipecotic acid into the NTS is attenuated by the selective GABA B antagonist phaclofen, but not by the selective GABA A antagonist bicuculline [23]. These results suggest that GABA tonically released in the NTS acts only on the GABA a receptor to increase AP. It has been revealed in electrophysiological studies that stimulation of the hypothalamic defense area in cats suppresses the baroreceptor reflex [6,20]. This stimulation also produces inhibition of tonic and responsive activities of the NTS neurons receiving a baroreceptor input, and the transmitter for this inhibition is estimated to be GABA, because bicuculline antagonizes this inhibitory response elicited by the hypothalamic defense area stimulation [12]. Furthermore, it has been demonstrated that bicuculline increases the evoked activities of the NTS neurons by electrical afferent stimulation of the cardiac or pulmonary vagal branch, the aortic nerve [1], or the carotid sinus nerve [18]. In the NTS, two types of neurons that respond to baroreceptive input have been identified with extracellular recording [8,22]: one is activated and another is depressed by baroreceptor stimulation (activated baroreceptive neuron (A-BRN) and depressed baroreceptive neuron (D-BRN), respectively). However, it has not been clarified yet how GABAergic mechanism functions on the ABRN or D-BRN. The aim of the present study was to elucidate the participation of GABAergic mechanisms in the baroreflex modulation in the NTS neurons that were identified as baroreceptive in nature (i.e., A- and D-BRN), using the microiontophoretic technique, and, in addition, to clarify the role of GABA receptor subtypes, i.e., GABA A and GABA a receptors in the regulatory mechanisms of the BRNs.
Materials and Methods
Preparation of animals Male albino rabbits (Japanese White) weighing 2.8-3.3 kg were anesthetized with urethane-chloralose (500 m g / k g and 50 mg/kg, i.v., respectively). An additional dose of urethane was given
when necessary. Cannulae were inserted into the trachea, the femoral artery and the femoral vein. After fixing the rabbits to a stereotaxic instrument (Narishige, SN-2) in a prone position, the animals were immobilized with gallamine triethiodide (initially 10 mg/kg, i.v., thereafter 5 m g / k g every 1-1.5 hr, Sigma) and artificially ventilated (Natsume, KN-56). The end-tidal CO 2 was constantly monitored (Nihondenki San-ei, Respina IH26) and was maintained at 4.0-4.5% by changing volume of the respirator. The atlanto-occipital membrane was exposed, cut at the midline and then retracted to expose the dorsal surface of the medulla and the surface was covered with warm mineral oil. To reduce respiratory movement of the brain stem, pneumothorax was induced in all experiments. Bilateral vagal nerves were cut in the neck. Rectal temperature was maintained at 37-38°C by an isothermal pad (Braintree, Deltaphase). AP was recorded from the abdominal aorta cannulated through the femoral artery and connected to a pressure transducer (Nihon Kohden, MPU-0.5). Heart rate (HR) was computed from AP pulses by a tachometer unit (Nihondenki Sanei, Type 1236). Baroreceptor stimulation was carried out by the injection of phenylephrine (10 ~zg/kg) through the femoral vein.
Recording of single unit activity and microiontophoretic application The multibarrel electrode, which was made by gluing together a single glass micropipette with angled tip and a glass 5-barreled micropipette, was used. The single pipette was filled with 0.5 M sodium acetate and 2% Pontamine Sky Blue (Brilliant Blue 6B, pH 7.7, Tokyokasei) for extracellular recording of the single neuronal activity and marking of the recording sites (tip diameter 1 ~m, 3-8 Mg2). Four barrels of the 5-barreled pipette (tip diameter 3-6 /~m, 5-15 MS2) were filled with the following drugs for microiontophoretic application: GABA (1 M, pH 4.0, Sigma), sodium glutamate (0.5 M, pH 8.0, Wako), L-bicuculline methiodide (20 mM, pH 4.0, Sigma), phaclofen (13.3 mM, pH 4.0, Tocris Neuramin) and De-baclofen (20 mM, pH 4.0, Sigma). These
29
drugs were applied at up to 160 nA using a microiontophoretic equipment (Dia Medical, DPI-30). Automatic current balancing was held through the remaining barrel of the 5-barreled micropipette filled with 3 M NaCI. Electrode penetration was performed stereotaxically in 0.5-1.0 mm caudal to the rostral border of the area postrema, in 1-2.0 mm lateral to the midline and within 2 mm deep to the dorsal surface of the medulla. Cells displaying respiratory rhythmic activity were excluded from the analysis [8]. Drug actions to cells were defined as an increase or decrease if they displayed changes of at least 30% in control activities. The single unit activity was amplified (Nihon Kohden, 50-3000 Hz, MEZ-7200 and AVH-10) and displayed on an ocilloscope (Nihon Kohden, VC-10). The discharge frequency was counted by a spike counter with 1 s bin (Dia Medical, DSE325P) and displayed on a polygraph (Nihondenki San-ei, Rectigraph-8S) together with AP and HR. Some recording sites were checked in histological sections (stained with Neutral Red) of the brain stem after the experiment.
Results
Forty-nine neurons in the NTS which responded to transient elevations of AP were examined. Their unit activities were confirmed to be derived from the soma of neurons by activation of their activities to the microiontophoretically applied sodium glutamate (5-40 nA) [10]. These neurons exhibited randomized discharge patterns which bore no obvious relationship to spontaneous fluctuations in relation to HR, and had basal firing frequency between 5 and 40 spikes/s. The neuron was classified as A- or D-BRN, if it displayed an increase or a decrease of at least 30% in baseline spike activity in response to transient elevation of AP produced by the intravenous injection of phenylephrine (10 /zg/kg) (Fig. 1). Twenty-six out of 49 neurons were identified as A-BRN and 23 neurons were identified as D-BRN.
Effects of GABA on the spontaneous activities of BRNs To determine if the spontaneous activities of the two classes of BRNs show different sensitivities to GABA, GABA was applied microiontophoretically to these neurons (2-80 nA). GABA decreased the spontaneous activities dose-dependently in all 26 A-BRNs and 22 of 23 D-BRNs tested. In each 48 GABA-sensitive neurons, the inhibitory effects of GABA were tested against the selective GABA A antagonist bicuculline and selective GABA B antagonist phaclofen [14,25]. Inhibitory effects of the microiontophoretically applied GABA on the BRNs were markedly prevented by simultaneously applied bicuculline (1020 nA) and partially antagonized by phaclofen (20-40 nA) (Fig. 2). These results indicate that A- and D-BRNs were affected by both GABA A and GABA B receptors which were thus involved in mediating the inhibitory action. Tonic effects of GABAergic system on BRNs activities To determine if the GABAergic system in the NTS tonically affects A- and D-BRNs activities, bicuculline and phaclofen were applied microiontophoretically to those neurons. Low doses of bicuculline (10-20 nA) had no effect on these neurons, but high doses (40-80 nA) increased the
A-BRN
D-BRN
o
AP
2oo mmHg 5O 3O s
Fig. 1. Effects of baroreceptor stimulation on the discharge rate of neurons in the NTS. The left and right traces show the baroreceptive neurons whose spontaneous activities were activated and depressed by the baroreceptor stimulation, respectively (activated baroreceptive neuron (A-BRN) and depressed baroreceptive neuron (D-BRN), respectively). Upper traces indicate unit activities/s (Unit). Lower traces indicate arterial blood pressure (AP). Arrows indicate intravenous injections of phenylephrine (10/xg/kg).
30
w GABA
20
~ GABA
40
m GABA
80 m
GABA 10
m
m
Bic 20
Bic 40 spikes/s 20
GAB-'~80
spikes/s
GAB-'~ 80
t0
Bic 10 "G-ABA80 Bic 20 ~'ABA 80
Phac 20 ~'ABA 80 Phac ~ A
80
3-6% Fig. 2. Blockade of neuronal responses to G A B A by bicuculline and phaclofen ( G A B A A and G A B A B antagonists, respectively). In 48 out of 49 BRNs examined, G A B A decreased the spontaneous activity. This figure shows the typical effects of GABA, bicuculline and phaclofen on the spontaneous activity of the same D-BRN. The traces indicate unit activities/s (Unit) and solid bars indicate periods of application of drugs. G A B A (20-80 nA) dose-dependently decreased the spontaneous activity (top traces). The neuronal response of G A B A (80 nA) was dose-dependently antagonized by bicuculline (Bic, 10-20 nA) and phaclofen (Phac, 20-40 hA) (middle and bottom traces, respectively).
spontaneous activities in 11 of 14 A-BRNs and 9 of 11 D-BRNs. However, phaclofen (up to 160 nA) had no effect on them in 14 of 15 A-BRNs and 12 of 13 D-BRNs, although their spontaneous activities were suppressed by the G A B A ~ agonist baclofen (40-80 nA). These results suggest that most of A- and D-BRNs were tonically inhibited by endogenous G A B A acting on G A B A A receptors, but not on G A B A n receptors. The typical effects on the same neuron were shown in Fig. 3. In the study on the same neurons, the neurons whose spontaneous activities were increased by bicuculline and unaffected by phaclofen were 10 of 13 A-BRNs and 7 of 9 D-BRNs. The remaining 3 neurons of A-BRNs and 2 of D-BRNs were not affected by either drug.
Effects of bicuculline and phaclofen on the evoked responses of BRNs Effects of bicuculline and phaclofen on the evoked response of A- and D-BRNs elicited by the baroreceptor stimulation were examined. Evoked responses of A-BRNs (10 of 12 neurons)
Bac 80
Phac g0
Phae 160 - 30 s Fig. 3. Effects of bicuculline and phaclofen on the spontaneous activity of BRN. In 20 out of 25 BRNs, high doses of bicucu0ine (Bic, 20-40 nA) increased the discharge rate of BRNs, while phaclofen (Phac, 80-160 nA) had no effect on 26 out of 28 B R N s (see Fig. 6). This figure is displayed in the same m a n n e r as Fig. 2 and shows the typical effects of the two drugs on the spontaneous activity of the same D-BRN. Following G A B A (10 nA), bicuculline was applied at 20 and 40 nA. Low dose (20 nA) bicuculline had no effect on the spontaneous activity, but a high dose (40 nA) increased it (top traces). Following baclofen (Bac, 80 nA), phaclofen was applied at 80 and 160 nA. Baclofen decreased the spontaneous activity, while phaclofen had no effect on the spontaneous activity up to 160 nA (bottom traces).
were potentiated by bicuculline (10-40 nA), while the responses of D-BRNs (8 of 10 neurons) were not clearly affected by it (Fig. 4). Although the A-BRN Unit
AP
1' Unit
1'
20
D-BRN t 20 spikes/s J 0
AP
150 mmHg 50 1.
1.
20
30 s Fig. 4. Effects of bicuculline on the evoked responses of Aand D-BRNs. Evoked responses of A - B R N s elicited by baroreceptor stimulation were potentiated by bicuculline (Bic, 10-40 nA), while the responses of D-BRNs were not clearly affected by it (see Fig. 6). Arrows and solid bars indicate intravenous injections of phenylephrine ( 1 0 / ~ g / k g ) and periods of application of bicuculline, respectively. Unit: unit activities/s, AP: arterial blood pressure.
31
Unit
A-BRN
spontaneous firing of D-BRN was increased by bicuculline, in the case shown in Fig. 4, the inhibitory response produced by the baroreceptor stimulation was hardly changed under the application of bicuculline compared to the control. Similarly, evoked responses of A-BRNs (7 of 9 neurons) were potentiated by phaclofen (40-160 nA), while the responses of D-BRNs (9 of 10 neurons) were not clearly affected by it (Fig. 5). These results suggest that GABAergic mechanisms suppressively modulated the baroreceptor reflex acting on GABA A and GABA a receptors of A-BRNs, but not of D-BRNs. In the study on the same neurons, the evoked responses in 5 of 8 A-BRNs were potentiated by both bicuculline and phaclofen, and the responses in 4 of 5 D-BRNs were unaffected by either drug. The remaining neurons of A- and D-BRNs were variously affected by those drugs, i.e., in A-BRNs (3 neurons), one neuron was
t
AP
t
t
Unit
a0
D-BRN spikes/s 0 mmNg 50
t
so
_ _
30
s
Fig. 5. Effects of phaclofen on the evoked responses of A- and D-BRNs. Evoked responses of A-BRNs elicited by baroreceptor stimulation were potentiated by phaclofen (Phac, 40-160 nA), while the responses of D-BRNs were not clearly affected by it (see Fig. 6). This figure is displayed in the same manner as Fig. 4.
Bicuculline Spontaneous o activity
5
10
15
I ....
Phaclofen 20
I ....
0
I . . . . . .
5
10
I ....
15
I ....
20
I ....
25
I ,,
,,
(N)
I,,
Increase ~ Decrease i ~ No effect
] I
Evoked response Potentiation
5 . . . .
i
10 . . . .
i
15 . . . .
I
0
5
10
15
(N)
i
......
i ......
'
Inhibition No
effect
I
D : A-BRN []
: D-BRN
Fig. 6. Summary of responses of A- and D-BRNs to bicuculline and phaclofen. This graph shows the number of neurons whose spontaneous activities and evoked responses produced by phenylephrine (10/,~g/kg, i.v.) were changed by these antagonists at least 30% vs. the control level, thus indicating GABA A and GABA a receptor sensitivities in each neuron group. Upper and lower graphs indicate effects of drugs on spontaneous activities and evoked responses of BRNs, respectively. Left and right sides indicate effects of bicucuUine and phaclofen, respectively. Open and hatched columns indicate A- and D-BRNs, respectively. The difference between the numbers of BRNs whose spontaneous activities were increased by bicuculline and phaclofen was significant (Fisher's exact test, P < 0.01, two-tailed). The difference between the numbers of A- and D-BRNs whose evoked responses were potentiated by bicuculline and phaclofen was significant (Fisher's exact test, P < 0.01, two-tailed, in each drug).
32 unaffected by either drug, another was depressed and unaffected, and the third was unaffected and potentiated, by bicuculline and phaclofen, respectively. In D-BRN (one neuron), it was potentiated and unaffected by bicuculline and phaclofen, respectively.
Summary of responses of A- and D-BRNs to bicuculline and phaclofen In Fig. 6, the effects of bicuculline and phaclofen on the spontaneous activities and evoked responses of A- and D-BRNs (upper and lower traces, respectively) were summarized. The spontaneous activities in 20 of 25 BRNs were increased by bicuculline and the others were unaffected, whereas phaclofen increased the spontaneous activities in only 2 of 28 BRNs and did not affect the others. This difference of the effects of bicuculline and phaclofen was significant (Fisher's exact test, P < 0.01, two-tailed). Both bicuculline and phaclofen potentiated most of the evoked responses of A-BRNs, but did not affect those of D-BRNs. Precisely, bicuculline potentiated 10 of 12 A-BRNs and phaclofen 7 of 9 A-BRNs, whereas 8 of 10 D-BRNs were not affected by bicuculline and 9 of 10 D-BRNs were also not affected by phaclofen. The difference in the effects of the antagonists on the evoked responses between A- and D-BRNs was significant (Fisher's exact test, P < 0.01, two-tailed). In this study, both A- and D-BRNs were found in the dorsal and medial regions of NTS, but there were no significant differences in localization of these neurons.
Discussion
We demonstrated the following differences of GABAergic modulation in the baroreceptor reflex in the NTS due to their receptor subtypes. First, the spontaneous activities of A- and DBRNs were increased by the GABA A antagonist bicuculline, but not by the GABA B antagonist phaclofen. These results indicate that GABAergic system in the NTS tonically suppresses A- and D-BRNs by acting on GABA A receptor, but not on G A B A u receptor. Second, the evoked re-
sponses of A-BRNs were potentiated by bicuculline and phaclofen, but those of D-BRNs were not affected by both drugs. These results indicate that the GABAergic system exerts the inhibitory modulation to the baroreceptor reflex by acting on GABA A and G A B A B receptors of A-BRNs, but not of D-BRNs. In the hitherto reported pharmacological study, it has been demonstrated that GABA A and GABA B receptors are involved in cardiovascular regulation in the NTS. Namely, the microinjection of GABA and muscimol, a GABA A agonist, into the NTS produces hypertension and tachycardia, and these effects are antagonized by bicuculline [3,15]. Baclofen, a GABA u agonist, also increases AP and H R and these effects are not affected by bicuculline [3,21,24], but antagonized by phaclofen and 2-OH-saclofen, GABA B antagonists [9]. In addition, the reflex bradycardia elicited by AP elevation and the reflex hypotension and bradycardia by electrical stimulation of the aortic depressor nerve are inhibited by muscimol and baclofen [9,16], and the effects of baclofen are blocked by phaclofen and 2-OH-saclofen [9]. Here we demonstrate the inhibitory effects of GABA on the spontaneous activities of A- and D-BRNs and that these effects are antagonized by bicuculline and phaclofen. The inhibitory effect of baclofen on the spontaneous activities is also demonstrated (see Fig. 3). Additionally, the high dose of bicuculline augmented the spontaneous activities of these neurons. On the contrary, phaclofen has no effect on these spontaneous activities, indicating that the involvement of the GABA u receptor is excluded from the tonically inhibitory modulation with GABAergic system in the NTS. Indeed, phaclofen microinjected into the NTS scarcely changed AP and H R in spite of blocking the cardiovascular effects of baclofen [9,23]. Sved and Sved [23] have reported that the cardiovascular effects of the microinjected nipecoptic acid, a GABA uptake inhibitor, into the NTS of rats are antagonized by phaclofen but not by bicuculline. Thus, they have proposed that tonically released GABA in the NTS acts specifically on GABA u receptors, but not on GABA A
33 receptors, to increase AP and HR. However, in this study in rabbits, bicuculline increased the spontaneous activities of A- and D-BRNs, but phaclofen did not. In addition, both bicuculline and phaclofen potentiated the evoked response of A-BRNs. These results indicate that both of GABA A and GABA B receptors are involved in the cardiovascular regulation in the NTS. Although this discrepancy might be based on the species difference used, it has been reported that the microinjection of bicuculline into the NTS in rats [15] and cats [3] decreased AP and HR, and the evoked responses in the NTS in cats by baroreceptor afferent stimulations were potentiated by the microiontophoretically applied bicuculline [12,18]. These results seem to be more comparable with our data than those of Sved and Sved. We, moreover, observed that the microiontophoretically applied nipecotic acid decreased the spontaneous activities of A- and D-BRNs in rabbits, and these effects were antagonized by both bicuculline and phaclofen in preliminary studies (unpublished observations). After all, we cannot explain these differences clearly yet. Surprisingly, phaclofen potentiated the evoked responses of A-BRNs, while it had no effect on the spontaneous activities of A- and D-BRNs. There are at least three possibilities for these phenomena. First, phaclofen could not have substantially penetrated into the synaptic cleft enough to block the inhibitory effects of the tonically released GABA completely, while in the evoked responses of A-BRNs, one small part of phaclofen applied might have partially exerted the potentiating effects. Second, as the binding affinity of phaclofen with GABA B receptor in the brain membranes is low [4], phaclofen might not have blocked the effect of the tonically released GABA completely. Third, there might be two types of GABAergic systems in the NTS: one is tonically inhibiting A- and D-BRNs through the GABA A receptor and, moreover, is activated from the other nucleus, e.g., hypothalamus defense area [12,20], and the other is activated by the baroreceptor inputs and affects both GABA A and GABA B receptors of A-BRNs. According to the latter mechanism, the presence of the phasic inhibition induced by baroreceptor afferent would
be postulated, because in in vitro [5] and in vivo [19] studies, the neurons that evoked IPSP or EPSP-IPSP by electrical stimulation of the afferents were demonstrated. In the present study, the inhibitory responses of D-BRNs evoked by baroreceptor stimulation could not be affected by bicuculline and phaclofen. This indicates that GABA may not be included in this inhibitory response of D-BRNs by baroreceptor stimulation. Mifflin et al. [19] demonstrated the three types of baroreceptive neurons referred to the synaptic responses with a stable intracellular recording. Namely, these neurons produced EPSP, EPSPIPSP or IPSP by electrical stimulation of the carotid sinus nerve. The IPSP was reversed to a depolarizing potential either by applying hyperpolarizing DC current or an intracellular injection of CI-, suggesting that it was mediated by a Cl--dependent process. It is well known that GABA, acting on GABA A receptor, exerts an inhibitory response by a C1--dependent process, thus GABA is the most likely candidate of this inhibitory transmitter. D-BRNs in this study accordingly could be comparable to the neuron which produces IPSP for its inhibitory response, but we failed to demonstrate the antagonistic effect of bicuculline on the evoked inhibitory response of D-BRNs. Although we used a dose of bicuculline that almost completely antagonized the effects of the exogenously applied GABA on the BRNs, it may not penetrate in the synaptic cleft sufficiently and, therefore, may not be able to antagonize the inhibitory action of the endogenously released GABA. This possibility could not be excluded, but the inhibitory response of the D-BRNs was not affected by the dose of bicuculline which somewhat augmented their spontaneous activities (see Fig. 4). To clarify the transmitter that induces the evoked inhibitory responses of D-BRNs, further studies will be needed. Although the evoked responses of A-BRNs were potentiated by bicuculline and phaclofen, the sites of actions of these drugs are not clarified. The following possibilities can be considered according to previous reports [4,7,9]. Firstly, both GABA A and GABA B receptors are located on the postsynaptic membrane of A-BRNs; and sec-
34
ondly, one of the two (it may be G A B A B receptor) locates on the presynaptic membrane of the excitatory input to A - B R N s with the other receptor (it may be G A B A A receptor) on the postsynaptic membrane [4]. Florentino et al. [9] have reported that the cardiovascular effects of stimulation of G A B A B receptors in the NTS are due, at least in part, to the presynaptic inhibition of the baroreceptor afferents. On the other hand, it has been observed that phaclofen selectively inhibits responses mediated by postsynaptic but not presynaptic G A B A B receptor in CA1 hippocampal pyramidal neurons in vitro [7]. Therefore, both G A B A receptors might be located on the postsynaptic membrane in the NTS, but further examination is necessary to determine the locations of G A B A A and G A B A B receptors. In summary, the present findings provide pharmacological evidence of the neuronal modulation of G A B A by acting on G A B A A and G A B A B receptors in the NTS. These results suggest that most of A- and D - B R N s are tonically inhibited by e n d o g e n o u s G A B A acting on G A B A A receptors, but not on G A B A B receptors, and that GABAergic mechanisms suppressively modulate baroreceptor reflex acting on G A B A A and G A B A a receptors of A-BRNs, but not of D-BRNs.
6
7
8
9
10
II
12
13
14
15
References 16 1 Bennett, J.A., McWilliam, P.N. and Shepheard, S.L., A y-aminobutyric-acid-mediated inhibition of neurones in the nucleus tractus solitarius of the cat, J. Physiol. (Lond.), 392 (1987) 417-430. 2 Blessing, W.W., Oertel, W.H. and Willoughby, J.O., Glutamic acid decarboxylase immunoreactivity is present in perikarya of neurons in nucleus tractus solitarius of rat, Brain Res., 322 (1984) 346-350. 3 Bousquet, P., Feldman, J., Bloch, R. and Schwartz, J., Evidence for a neuromodulatory role of GABA at the first synapse of the baroreceptor reflex pathway. Effects of GABA derivatives injected into the NTS, NaunynSchmiedeberg's Arch. Pharmacol., 319 (1982) 168-171. 4 Bowery, N., GABA B receptors and their significance in mammalian pharmacology, Trends Pharmacol. Sci., 10 (1989) 401-407. 5 Champagnat, J., Denavit-Saubi6, M., Grant, K. and Shen, K.F., Organization of synaptic transmission in the mam-
17
18
19
20
malian solitary complex, studied in vitro, J. Physiol. (Lond.), 381 (1986) 551-573. Coote, J.H., Hilton, S.M. and Perez-Gonzalez, J.F., Inhibition of the baroreceptor reflex on stimulation in the brain stem defense centre, J. Physiol. (Lond.), 288 (1979) 549560. Dutar, P. and Nicoll, R.A., Pre- and postsynaptic GABA B receptors in the hypocampus have different pharmacological properties, Neuron, 1 (1988) 585-591. Feldman, P.D. and Moises, H.C., Adrenergic responses of baroreceptive cells in the nucleus tractus solitarii of the rat: a microiontophoretic study, Brain Res., 420 (1987) 351-361. Florentino, A., Varga, K. and Kunos, G., Mechanism of the cardiovascular effects of GABA B receptor activation in the nucleus tractus solitarii of the rat, Brain Res., 535 (1990) 264-270. Fries, W. and Zieglg~insberger, W., A method to discriminate axonal from cellbody activity and to analyse "silent cells", Exp. Brain Res., 21 (1974) 441-445. Gale, K., Hamilton, B.L., Brown, S.C., Norman, W.P., Souza, J.D. and Gills, R.A., GABA and specific GABA binding sites in brain nuclei associated with vagal outflow, Brain Res. Bull., 5, Suppl. 2 (1980) 325-328. Jordan, D., Mifflin, S.W. and Spyer, K.M., Hypothalamic inhibition of neurones in the nucleus tractus solitarius of the cat is GABA mediated, J. Physiol. (Lond.), 399 (1988) 389-404. Jordan, D. and Spyer, K.M., Brainstem integration of cardiovascular and pulmonary afferent activity, Prog. Brain Res., 67 (1986) 295-314. Kerr, D.I.B., Ong, J., Prager, R.H., Gynther, B.D. and Curtis, D.R., Phaclofen: a peripheral and central baclofen antagonist, Brain Res., 405 (1987) 150-154. Kubo, T. and Kihara, M., Evidence for the presence of GABAergic and glycine-like systems responsible for cardiovascular control in the nucleus tractus solitarii of the rat, Neurosci. Lett., 74 (1987) 331-336. Kubo, T. and Kihara, M., Evidence for y-aminobutyric acid receptor-mediated modulation of the aortic baroreceptor reflex in the nucleus tractus solitarii of the rat, Neurosci. Lett., 89 (1988) 156-160. Maley, B. and Newton, B.W., Immunohistochemistry of y-amino butyric acid in the cat nucleus tractus solitarius, Brain Res., 330 (1985) 364-368. McWilliam, P.N. and Shepheard, S.L., A GABA-mediated inhibition of neurones in the nucleus tractus solitarius of the cat that respond to electrical stimulation of the carotid sinus nerve, Neurosci. Lett., 94 (1988) 321-326. Mifflin, S.W., Spyer, K.M. and Withington-Wray, D.J., Baroreceptor inputs to the nucleus tractus solitarius in the cat: postsynaptic actions and the influence of respiration, J. Physiol. (Lond.), 399 (1988) 349-367. Mifflin, S.W., Spyer, K.M. and Withington-Wray, D.J., Baroreceptor inputs to the nucleus tractus solitarius in the cat: modulation by the hypothalamus, J. Physiol. (Lond.), 399 (1988) 369-387.
35 21 Persson, B., A hypertensive response to baclofen in the nucleus tractus solitarii in rats, J. Pharm. Pharmacol., 33 (1981) 226-231. 22 Suzuki, M., Investigation on the neuronal mechanisms of baro- and chemoreceptor reflexes in the nucleus tractus solitarii: participation of adrenergic and opioid peptidergic system, J. Saitama Med. School, 17 (1990) 31-43. 23 Sved, A.F. and Sved, J.C., Endogenous GABA acts on GABA n receptors in nucleus tractus solitarius to increase blood pressure, Brain Res., 526 (1990) 235-240.
24 Sved, J.C. and Sved, A.F., Cardiovascular responses elicited by y-aminobutyric acid in the nucleus tractus solitarius: evidence for action at the GABA a receptor, Neuropharmacology, 28 (1989) 515-520. 25 Wiillner, U., Klockgether, T. and Sontag, K.-H., Phaclofen antagonizes the depressant effect of baclofen on spinal reflex transmission in rats, Brain Res., 496 (1989) 341-344.