37

Brain Research, 592 (1992) 37-43 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

BRES 18116

Tonic stimulation of G A B A B receptors in the nucleus tractus solitarius modulates the baroreceptor reflex A l a n F. S e e d and Kazuyoshi T s u k a m o t o Department of Behavioral Neuroscience, Universityof Pittsburgh, Pittsburgh, PA 15260 (USA) (Accepted 5 May 1992)

Key words: y-Aminobutyric acid; Blood pressure; Baclofen; Hydroxysaclofen;Muscimol;Aortic depressor nerve

Previous studies have indicated that tonic stimulation of GABA B receptors in the nucleus tractus solitarius (NTS) contributes to the regulation of arterial blood pressure (AP). The present studies examined the hypotheses that (1) tonic stimulation of GABAB receptors in the NTS provides a tonic attenuation of the baroreceptor reflex and (2) enhanced stimulation of these GABA u receptors markedly attenuates the baroreceptor reflex resulting in an increase in AP. In chloralose-anesthetized rats electrical stimulation of the aortic depressor nerve elicited frequency-dependent decreases in AP and heart rate (HR). These responses were markedly attenuated, but not eliminated, by injection of the GABA u receptor agonist baclofen into the ipsilateral NTS. In contrast, the GABAA receptor agonist muscimol completely inhibited aortic depressor nerve-evoked responses. Blockade of GABA n receptors in the NTS by local injection of CGP-35348 elicited a dose-dependent decrease in AP, and a dose-dependent blockade of the presser response elicited by injection of baelofen into the NTS. These results support the hypothesis that GABA acts tonically on GABA n receptors in the NTS to attenuate the baroreceptor reflex, thereby contributing to the regulation of AP.

INTRODUCTION The baroreceptor reflex plays a critical role in the normal regulation of arterial blood pressure (AP) 34. The dorsld medial region of the nucleus traetus solitarius (NTS) is the site of termination of baroreceptor afferent fibers t~'t7 and is therefore a critical relay site in the baroreceptor reflex pathway. Administration of neuroinhibitory agents into this region of the NTS can block the baroreceptor reflex and increase AP s'22'32, similar to the acute effects of arterial baroreceptor deafferentation2t. The baroreceptive region of the NTS contains a high density of GABA-containing nerve terminals 4'ts'23-2~'2~'3° and a high density of both GABAA and GABA n receptors ~33'33. Previous studies in this laboratory 3~'3s and others "~'t2'28 have demonstrated that stimulation of GABA B receptors in this region of the NTS elicits a marked pressor response. Doses of the GABAn-receptor agonist baclofen that elicit a maximal increase in AP also block the baroreceptor reflexmediated bradycardia evoked by increasing AP t2'2s.

Furthermore, the GABAB receptors involved in these responses appear to be tonically stimulated by endogenous GABA, since local administration of indirectly acting GABA agonists elicits a presser response that can be completely reversed by the specific GABAu receptor antagonist phaclofen3~'. The present studies further examine the role of GABA B receptors in the NTS in the regulation of AP. These studies address the hypotheses that (1) tonic stimulation of GABA B receptors in the NTS provides a tonic attenuation of the baroreceptor reflex and (2) enhanced stimulation of these GABA B receptors markedly attenuates the baroreceptor reflex resulting in an increase in AP. MATERIALS AND METHODS These experiments were conducted on adult male Sprague-Dawley rats (Zivic-Miller, Allison Park, PA) weighing 300-480 g. Rats were housed singly in wire-mesh cages in a temperature-controlled room on a 12 h light/dark cycle with food and tap water available ad libitum for at least 1 week prior to use in experiments. Rats were anesthetized with halothane (2% in 100% 0 2 adminis-

Correspondence: A. Sved, Department of Behavioral Neuroscience, 446 Crawford Hall, University of Pittsburgh, Pittsburgh, PA 15260, USA. Fax: (I) (412) 624-9198.

38 tered through a cone placed over the nose). A cannula (PE-50 tubing filled with heparinized saline) was inserted into the right femoral artery for recording of AP and heart rate (HR) (Grass model 7 Physiograph). As cannula was placed in the right femoral vein for administering drugs. The trachea was cannulated and rats were artificially ventilated with 2% halothane in 100% 02 (Harvard Small Animal Respirator) followed by the administration of a muscle relaxant (d-tubocurarine, 0.5 m r / k s i.v., supplemented hourly with 0.2 m r / k s i.v.), in experiments requiring electrical stimulation of the aortic depressor nerve (ADN), the right or left ADN was identified at the level of the carotid bifurcation and approximately 5 tam of the nerve was isolated. Two Teflon-coated stainless steel wires (seven stranded, 230 pm o.d., with approximately 2 mm of the tip exposed and only two strands of wire remaining at the tip; A-M Systems, Everett, WA) were wrapped around the ADN approximately 1-2 mm apart. These wires were then anchored in place using a vinyl polysiloxane cement (light bodied Vy. P, Kent Dental, Aston, PAl. Rats were placed in a stereotaxic instrument (Kopf Instruments) with the incisor bar positioned ! ! mm below the interaural line. The dorsal surface of the medulla was exposed by limited craniotomy and, with the aid of a surgical microscope, the area postrema visualized, a-Chloralose was administered (60 m r / k s i.v. then supplemented hourly with 20 mg/kg i.v.) and the halothane terminated. Rats were ventilated with 100% O 2 throughout the remainder of the experiment. Animals were left to stabilize for at least 20 rain before the start of the experiment. Injections of drugs were made into the NTS using single or double barrel glass micropipettes pulled to an outer diameter of 40-50/~m, and beveled to approximately 45°, The tip of the micropipette was positioned 0.5 mm rostral to the caudal tip of the area postrcma, 0,5 mm lateral from the midline, and 0.5 mm deep to the surface of the brainstem for injection into the NTS. All drugs were dissolved in artificial cerebrospinal fluid (aCSF) 36, All drugs were injected in a 100 nl volume over a period of several seconds using a PicoPump (WPI, New Haven, CT). The volume of drug injected was carefully monitored by w, tchin~ the movement of the fluid meniscus in the calibrated mteropipette, Some animals received unilateral electrolytic lesions of one NTS

at least 20 rain prior to injection of drugs into the contralateral, intact NTS, Lesions were made using n Teflon.insulated tungsten

electrode (150 /tm o,d.) with approximately 150 p,m of the tip exposed. The coordinates for creating the lesions were the same as those used for microinjections. Lesions were made by passing anodal current of I mA (Grass LMS, Quincy, MA) for 10 s through the electrode. A clip attached to the neck muscles served as the cathode. Effectiveness of the lesion was confirmed by the presence of a presser response of at least 30 mm Hg to nipecotic acid (10 nmol) injected into the intact NTS '~s. In animals in which the ADN was electrically stimulated, the ADN electrodes were connected to a stimulator (Grass $88, Quincy, MA) equipped with a stimulus isolation unit (Grass PSUI6), The stimulus parameters used were: 10.s train, 0.5 ms pulses, 100-150 ~tA, 2.5-25 Hz, in the experiments examining the effect of injection of aCSF or 40 pmol bactofen on ADN.evoked responses, frequency response curves of 2.5 Hz, 5 Hz, and 10 Hz were generated in each rat before injection into NTS. Stimulus trains were administered 1-2 rain apart. Beginning 1.5 min after drug injection the stimulus response curves were repeated. Frequency-response curves were then tested at 15-rain intervals until responses had returned to pre-drug values, in experiments examining the effect of 200 pmol baclofen a 25 Hz stimulation frequency was also tested. At the conclusion of the study, 100 nl of !% Fast green was injected into the NTS using the same micropipette that was previously used for drug injections. The animal was then decapitated and the brainstem rapidly removed and frozen in isopentane on dry ice. Brainstems were subsequently cut into 40 tom sections using a cryostat and sections were mounted on glass microscope slides. Sections were stained with Cresyl violet or thionin. Only animals in which microinjection sites and lesions were centered in the medial subnurleus of the NTS at the rostral-caudal level of the area postrema were included in the data analysis. The following drugs were used in these studies: +baclofen

A

160

[,: 'e

'r

160

v

e

2,6 Hz

i

6.0 Hz

i

10 I-Is

Fig. 1. Response to aortic depressor nerve stimulation. Typical AP, MAP, and HR recordings from a chloralose-anesthetized, paralyzed, ventilated rat during stimulation of the left ADN. Stimulus parameters were: 10-s train (indicated by the bar), 2.5-10 Hz, 0.5 ms pulses, 100 /~A. These recordings were obtained from a single rat, with stimulus trains being delivered 1-2 min apart. These results are typical of the control responses included in Fig. 2. (Ciba-Geigy, Summit NJ), CGP-35348 (Ciba-Geigy, Basel, Switzer. land), muscimol (RBI, Wayland, MA) and 2-hydroxysaclofen (RBI, Wayland MA), All other chemicals were purchased from standard commercial suppliers. Data are expressed as means+S.E.M, and were analyzed by analysis of variance followed by the Tukey-Kramer test (SYSTAT, Systat, Evanston, ILl. The specific type of analysis of variance (e.g. 2 factor, repeated measure) varied depending on the design of the specific experiment.

RESULTS Electrical stimulation of the ADN produced a frequency dependent decrease in AP and HR (Figs. 1 and 2). Unilateral microinjection of +baclofen (40 pmol) into the NTS markedly attenuated the depressor and bradycardic responses to electrical stimulation of the ipsilateral ADN (Fig. 2). This dose of baclofen injected unilaterally into the NTS had no effect on baseline AP or HR. At stimulus frequencies of 2,5 and 5 Hz, baclofen completely blocked ADN-evoked cardiovascular responses, At a higher stimulation frequency (10 Hz) the depressor and bradycardic responses were attenuated but not eliminated, A higher dose of baclofen (200 pmol) was also not able to completely block the responses evoked by stimulation of the ADN at 10 Hz (Fig. 3), At a supramaximal stimulation frequency (25 Hz) the responses to ADN stimulation were not significantly different from control (Fig. 3). This high dose of baclofen elicited a marked increase in AP (reaching a maximum of +35 + 11 mm Hg 5 rain following injection and remaining at this elevated level for at least 20 rain), presumably due to spread of the drug to the contralateral HTS 8. Administration of baclofen (40 pmol) into the NTS contralateral to the

stimulated ADN had no effect ADN-evoked responses (n = 4, data not shown). Injection into the NTS of the GABA A agonist muscimol (10 pmol) blocked ADNevoked depressor and bradycardic responses, and this blockade was complete even at high stimulus frequencies (Table I). If stimulation of GABA e receptors by endogenous GABA acts to tonically inhibit the baroreceptor reflex, then blockade of GABA n receptors in the NTS should elicit a response similar to that produced by stimulation of baroreceptor afferents, i.e., a decrease in AP and HR. Phaclofen, a specific but weak GABA B receptor antagonist tg, has previously been shown to elicit a small decrease in AP 36. To further examine this issue, two drugs reported to be specific and potent antagonists of GABA B receptors, 2-hydrox3,saclofen 9'18

O[

:#:1'

39 ~t

$ Bacloren

(200 pmol)

-I0

0 Control

-20

.aor

-50 |

I

I

i

2.5 Hz 5 Hz 10 Hz 25 Hz Frequency of ADN stimulation

O

mt

*t

• Baclofen (200 pmol)

-)O

O Control

-20 O

~'1"

~_t

-lO

• Badofen (40 pmol)

-30

O Control

-40

410

.50

.30

.60

I

i

I

I

2,5 Hz

5 Hz

10 Hz

25 Hz

Frequency of A D N stimulation

.60 -60

I

I

t

2.5 Hs 5 Ha 10 Hz FrequencyofADNstimulation • Baelof~n (40 pmol)

0

0 Control

-tO

.30 -40

.~0 "6C

I

2,S Hz

I

I

5 Hz

10 Hz

Frequency of ADN stimulation Fig. 2. Effect of baclofen injected into the NTS on AP and HR responses elicited by ADN stimulation. The ADN was stimulated at 2,5, 5, and 10 Hz before (control) and 2-5 min after injection of baclofen (40 pmol) into the ipsilateral NTS (n = 5). Data are expressed as the maximal change in MAP and HR occurring daring stimulation of the ADN. Prior to baclofen injection, all stimulus frequencies produced significant depressor and bradycardic responses ( P < 0.01). * indicates responses that are significantly smaller than the control response ( P < 0.01) and * indicates responses that are not significantly different from zero. Baseline AP and HR were not altered by injection of baclofen.

Fig. 3, Effect of a supramaximal dose of baclofen injected into the NTS on AP and HR responses elicited by ADN stimulation, The ADN was stimulated at 2,5, 5, 10, and 25 Hz before (control) and 2-7 min after injection of baclofen (200 pmol) izzto the ipsilateral NTS (n = 8). Data are expressed as the maximal change in MAP and HR occurring during stimulation of the ADN. Prior to baclofen injection, all ~timulas frequencies produced significant depres~r and bradycardic responses (P