MK-801 affects cardiovascular responses produced by GABAA agents in area postrema

Brain Research Bulletin. Vol. 35, No. I, pp. 63-67, 1994 Copyright 0 1994 Elsevier Science Ltd Printed in the USA. All rights reserved

Pergamon

0361-9230/94 $6.00 + 00

0361-9230(94)EOO87-G

MK-80 1 Affects Cardiovascular Responses Produced by GABA, Agents in Area Postrema BAOHONG

TIAN

AND

DIANE

K. HARTLE’

Department of Pharmacology and Toxicology, College of Pharmacy, University of Georgia, Athens, GA 30602 Received

1 July 1993; Accepted

6 April 1994

TIAN, B. AND D. K. HARTLE. MK-801 affects cardiovascular responses produced by GABA, agents in area postrema. BRAIN RES BULL 35( 1) 63-67, 1994.-The cardiovascular effects of nanoliter microinfusions of muscimol, bicuculline, and MK-801 (dizocilpine) into area postrema (AP) were tested in urethane-anesthetized male Sprague-Dawley rats. Microinfusion of muscimol (10 nl/min/5min, 1 nglnl) into the AP produced significant hypertension (+29 % 8 mmHg), but no significant change in heart rate (HR). Microinfusion of bicuculline (10 nl/min/5 min, 1 ng/nl) into the AP produced significant hypotension (-27 ? 4 mmHg) and bradycardia (-53 ? 10 bpm). Treatment of the AP with MK-801 (20 nl/min/S min, 6 nglnl) produced no changes in mean arterial pressure or HR by itself, but it completely blocked the hypotension and bradycardia produced by bicuculline infusion, without affecting the hypertension produced by muscimol microinfusion into AP. Following pretreatment of AP with MK-801, microinfusion of muscimol into the AP produced significant tachycardia (+5 1 -C 16 bpm). The data suggest that a) GABAergic neurotransmission in the AP affects its cardiovascular functions; b) functional interactions between NMDA receptor-mediated neurotransmission and GABAA receptor-mediated neurotransmission within the AP modulate blood pressure and HR regulation. Mean arterial pressure N-methyl-o-aspartate

Heart rate NMDA

Circumventricular

organ

Gamma-aminobutyrate

GABA

Medulla

during pressor events elicited by intravenous injection of phenylephrine was attenuated when MK-801 had been microinfused into the area postrema (AP) (29). Thus, it appeared that MK-801 could produce many cardiovascular effects via actions at the AP and the adjacent mNTS. The AP is innervated by several cardiovascular regulatory sites including the dorsolateral paraventricular nucleus, the lateral parabrachial nucleus (PBL), the NTS, and vagal afferents (21,26). The presence of both GABA and glutamate has been demonstrated in bovine AP using thin layer chromatography technique (18). Numerous nerve terminals in the AP are known to have either GABA or glutamate immunoreactivity, indicating possible local neurotransmitter roles for these amino acids (8,30). The AP is a chemosensory region in the brain stem for many blood-borne peptides, hormones, drugs, and chemical substances (3). Individual AP neurons demonstrate responsiveness to multiple hormones and neuromodulators with characteristic long latency, low discharge, and long duration effects (5). It is likely that the response of neural substrates within AP is highly influenced by neuromodulatory substances released either by intrinsic AP intemeurons or by incoming projection fibers. Functional studies on the roles of either GABA or NMDA receptors in cardiovascular regulation mediated via specific AP efferents have not been previously reported. While L-glutamate injections (27) and NMDA infusions (29) into AP have been shown to produce both hypotension and bradycardia, there have been no functional neuropharmacological studies concerning the cardiovascular ef-

GABAERGIC mechanisms are already known to be involved in regulation of neural control of the circulation at several different sites. These include the dorsal medial hypothalamus (25), posterior hypothalamus (lo), paraventricular hypothalamus (20), dorsal raphe (33). rostra1 ventrolateral medulla (RVLM) (3 1,32), caudal ventrolateral medulla (CVLM) (31), nucleus tractus solitarius (NTS) (6,15,16,28), and the spinal cord (11). The actions of GABAergic agonists at NTS (16) and CVLM (32) increase blood pressure while their actions in the dorsomedial (25), posterior (10) and paraventricular hypothalamus (20), the dorsal raphe (33), RVLM (32), and spinal cord (11) decrease blood pressure. MK-801 (dizocilpine) is a selective, noncompetitive Nmethyl-D-aspartate (NMDA) receptor-linked ion channel blocker. It is a potent anticonvulsant, neuroprotective drug with central sympathomimetic properties (1,9,23). The lipid solubility of MK-801 allows it to cross the blood-brain barrier in concentrations sufficient to reduce NMDA receptor-mediated neurotoxicity (4,7,14). Intravenous MK-801 injections (25-250 mg/kg dose range) cause dose-dependent increases in mean arterial pressure and heart rate (HR) in conscious rats that appear to be centrally mediated and partially buffered by arterial baroreflexes (19). We have previously investigated two potential sites in the dorsal vagal complex for roles in the centrally mediated cardiovascular effects of MK-801. We found that when it was microinfused bilaterally into the mNTS, MK-801 increased mean arterial blood pressure and HR (29). Baroreflex-mediated bradycardia

’ To whom requests for reprints should be addressed. 63

64

TIAN

fects of either GABA agonists or antagonists or glutamate antagonists in the AP. The purpose of this study was to test the blood pressure and heart rate effects of the GABAA agonist, muscimol, and the GABAA antagonist, bicuculline, by microinfusing them directly into the AP using urethane-anesthetized male Sprague-Dawley rats. We also tested for an interaction between NMDA receptormediated neuromodulation and the cardiovascular responses produced by muscimol and bicuculline microinfusion in AP by comparing the mean arterial pressure and HR effects of these agents, with and without prior blockade of NMDA receptor-linked ion channels with MK-801. METHOD

AND HARTLE

in AP was tested at the beginning of each experiment. These tests were uniformly found to be without effect on any of the recorded variables. Source of Drugs

(+) MK-801 hydrogen maleate (Dizocilpine maleate, (+)-5methyl-IO,1 I-dihydro-5H-dibenzo-(a,d) cyclohepten-5,10-imine maleate) was purchased from Research Biochemicals Incorporated (Natick, MA). Muscimol and bicuculline methiodide were purchased from Sigma Chemical Company (St. Louis, MO). The vehicle for all drug solutions was a sterile plasma electrolyte solution at pH 7.4 (Plasma-Lyte, Baxter Healthcare Corporation, Deerfield, IL). The pH of all drug solutions was adjusted to 7.4 before microinfusion.

Animals and Housing Conditions Adult male Sprague-Dawley (SD) rats (Charles River, Raleigh, NC), weighing 300-380 g, were used in all experiments. Rats were housed in a temperatureand humidity-controlled room on a 12L: 12D cycle and allowed free access to standard laboratory Purina rat chow and tap water. All protocols were approved by the University of Georgia’ s Animal Care and Use Committee.

Stutistical Analyses All data are presented as mean 2 SEM. The paired Student’s t-test was used to determine the statistical significance of differences for paired comparisons between mean arterial pressure and HR before and after a single treatment. One-way and two-way ANOVA and LSD post hoc were used to determine the statistical significance of differences among treatments. The criterion for significance was p < 0.05.

Surgical Preparations All experiments were performed using urethane (1 .O- 1.2 g/kg, IP) anesthesia. The left femoral artery of each rat was catheterized with PE-50 tubing and then used to monitor blood pressure. Body temperature during anesthesia was maintained at 37.0 2 1.O”C using an automatic electric heating pad. The arterial line was then attached to a Statham P23Dc small volume displacement pressure transducer, and blood pressure was continuously recorded on a Grass Model 79 C polygraph. HR was determined from the systolic peak-to-peak time intervals of the direct arterial blood pressure signal. Stereotaxic Brain Surgery and Drug MicroinfisionLs Each animal was placed in a stereotaxic frame with the head flexed anteriorly downward to an angle of 30” from the horizontal plane between lambda and bregma. After a dorsal cervical midline incision, the neck muscles were retracted, exposing the foramen magnum. The caudal portion of the 4th ventricle, the AP, and the dorsal medullary surface were exposed by removal of atlantooccipital membrane, and when necessary, removal of a piece of cranial bone overlying the cerebellum. The total time required for all preparation surgery was less then 30 min from induction of surgical anesthesia. Thirty minutes were allowed after this surgery for stabilization of hemodynamic parameters before initiation of experimental treatments. Experimental

RESULTS

Effects of Muscimol and MK-801 Microinfusions MAP and HR

Into AP on

Microinfusion of muscimol into the AP of urethane-anesthetized Sprague-Dawley male rats (n = 8) at a dosage of 10 ngl IO nl/min for 5 min produced significant increases in MAP. The mean maximal increase, 29 t- 8 mmHg, was attained by IO min and was maintained for another IO min (Fig. I). Full recovery to baseline blood pressure required approximately 1 h after the cessation of the treatment. Microinfusion of MK-801 (20 nl/min/5 min, 6 ng/nl) into AP produced no changes in MAP (actual data not presented here, but were insignificantly different from the data presented here for the same MK-801 treatment in another experiment, see Fig. 3). Prior infusion of MK-801 into AP also affected neither the hypertensive response nor the duration of that response during subsequent microinfusion of muscimol into the AP (Fig. 1).

o-o a- -.

before MK--801 cflerMK-801

Procedure

Microinfusion of either vehicle solution or drugs into AP was performed using a glass micropipet (tip o.d. = 30 ,um) attached to a nanopump (World Precision Instruments, Model A 1400). The micropipet was backfilled with either a vehicle or a drug, and positioned over the middle AP. The micropipette tip was then positioned on the midline and inserted 100 pm below the surface of the AP and midway between its rostra1 and caudal extent. Blood pressure and heart rate responses were recorded continuously. Peak responses during microinfusion of either MK-801, muscimol, or bicuculline were determined. A maximum of three penetrations occurred per rat; one of these was the initial vehicle control infusion below. The effect of isovolumic vehicle infusion

0

10

20

30

._

40 TIME

_-

Xl

^^

OU

(min)

FIG. 1, Effects of microinfusion of MK-801 (20 nUminI5 min, 6 ng/nl) into the area postrema (AP) on the change of mean arterial pressure (MAP) induced by infusion of muscimol (IO nVminI5 min, I ng/nl) into the AP. n = 8. mean 2 SEM.

MK-801 AND GABAA I~RA~ION~

65

AT AP

Neither mic~i~sion of muscimol (10 ng/lO nl/min/S min) into the AP nor a separate micmin~sion of MK-801 (20 nl/min/ 5 min, 6 ng/nl) produced significant changes in HR. If, however, the AP was pretreated with MK-801, significant increases in HR (+-51 ?z 16 bpm) occurred in response to a subsequent microinfusion of muscimol. This tachycardic elect lasted a&out 10 min (Fig. 2).

O

8

T it

-20

&?

Ejfkts of Bicuculline and MK-801 ~i~roi~~sion MAP and HR

Into AP on

Mic~in~sion of bicuculline into the AP (10 n&l0 nlimin for 5 min) produced significant decreases in MAP (-27 + 4 mmHg) and HR (-53 ~?r10 bpm). The maximal depressor responses were obtained within 3-5 min and recovery to baseline occurred within 10 min. As above, microinfusion of MK-801 (20 nl/min/ 5 min, 6 ng/nl) into the AP produced no changes in MAP and HR by itself. However, pretreatment of AP with MK-801 completely blocked both the hypotension and bradycardia effect of bicuculline (Fig. 3). DISCUSSION

GABA, as an inhibitor neurotr~smitter, and glut~ate~ as an excitatory neuro~~smitter, are known to modulate cardiovascular regulation in opposite directions at various brain loci. For example, the GABAA agonist, muscimol, when injected into the CVLM, produces increases in blood pressure ( 17), while the glutamate agonist, NMDA, produces decreases in blood pressure when injected into the same site (17). In the RVLM, the GABAA agonist, muscimol, produces decreases in blood pressure (32) while NMDA application produces increases in blood pressure (22). The data presented here support the concept that functionally opposing actions and perhaps interactions between excitatory and inhibitor amino acid neurom~ulation of AP functions modulating blood pressure and HR regulation. Our previous studies indicate that NMDA infusion into AP causes both hy~tension and brady~rdia (29). These effects are totally blocked by ~e~~trnent of the AP with MK-801 using the same dosage regimen as was used in these studies (29). The results of the present experiments provide evidence that mimicking (muscimol adminis~ation) or interfering (bicucul~ine administration) with endogenous GABAA receptor-mediated functions in AP also evokes changes in cardiovascular function. Microinfusion of muscimol, a GABA,, agonist, into the AP produced an increase in arterial blood pressure with no accompanying change __ o----o before MK-801

TIME

(mh)

-10

FIG. 2. Effects of micr~infusion of MK-801 (20 ntlminl5 min, 6 nglnl) into the AP on the change of heart rate (HR) induced by infusion of muscimol (10 al/mini5 min. In g/nI) into the AP, n = 8, mean 2 SEM. *p < 0.05, compared with baseline values and within groups.

-60

5i

BtCUCUWNEafbr UK-801

3. Effects of mi~infus~on of MK-8OI (20 n~mi~5 min, 6 ngtnl) into the AP on the alteration of MAP and HR evoked by microinfusion of bicuculline (I 0 nVmin/5 min, I ng/nl) into the AP. n = 6, mean 2 SEM. *p < 0.05, compared with baseline values.

FIG

in HR, but mic~injection of bicuculline, a GABA,, ~tagonist, into the AP produced decreases in both arterial pressure and HR. Altogether our present studies indicate functional cardiovascular roles for either endogensus GA13AA ~ceptor-rn~iat~ neurotransmission or stimulation of noninne~at~ GABAA receptors in AP. Similarly, our previous studies indicate functional cardiovascular roles for either innervated or noninnervated NMDA receptors in rat AP (31). It is inte~sting that the maxima1 depressor and bradycardic responses produced by bicucuiline were quantitatively and qualitatively similar to the maximal responses produced by direct NMDA microinfusion into the AP (29). Because both the effects of NMDA and bicuculline are completely blocked by MK-801 in the AP, we interpret these data to mean that both depressor and cardiac decelerator efferent pathways originate in AP and both can be activated by NMDA or by complete withdrawal of tonic GABAA receptor-mediated inhibition. The results appear to indicate that NMDA receptor-m~iated events are required for the blood pressure and heart rate effects of GABAA antagonism mediated at AP. Although the hypertensive effect of GABAA agonism by muscimol microinfusion was not affected by MK-801 p~~ea~ent, it should be noted that HR responses were significantly altered. Thus, although withdrawal of NMDA agonist activity by MK801 in AP does not produce an integrated change in blood pressure or HR by itself, the cardiovascularresponses to both GABAA agonism and antagonism are significantly altered during MK-801 blockade. A~~ough N~A receptor-rn~iat~ ne~o~~s~ssion in AP was not necessary for the pressor effect produced by GABA* agonism in AP with muscimol, NMDA receptor-mediated neuro~ansmission appeared to function~ly buffer the HR effects of GA13AA agonism at AP. This may suggest differences in the neurochemical mechanism transducing the heart rate and blood pressure responses to muscimol, or it may suggest that one system was active and the either inactive under the physiological conditions of the experiment. MK-801 pre~eatment unmasked a tachycardic response to muscimol at AP. Because GABA is an inhibitory amino acid, GABA receptormediated neurotransmission in AP, mimicked by muscimol in these experiments, probably increases blood pressure by inhibiting an inhibitor efferent system. One attractive hypothesis is that indirect (e.g., via NTS) or direct neuro~atomical connections exist between the AP and the CVLM that account for this effect. CVLM is known to inhibit bulbospinal neurons in the RVLM that control sympa~etic vasomotor outflow (2,31). Elec-

66

TIAN AND HARTLE

trical simulation of the AP causes a 62% increase in meta~lism in the CVLM (12), which indicates ne~oanatomical connections do, indeed, exist that can functionahy activate a GABAergic depressor area in the GVLM. Chemical stimulation in AP with L-glutamate lowers blood pressure via a mechanism dependent upon release of GABA in the RVLM (27). If muscimol inhibits such an efferent system, this could raise blood pressure by relieving CVLM inhibition of RVLM. Finally, with the infusion of any drug into a discrete brain locus, there is always the possibility of diffusion of the drug to a neighboring active site. In the case of the AP, the adjacent NTS has been shown to be an active site for GABAergic modulation of cardiovascular regulation. In the NTS, both GABAA and GABAB receptor subtypes may contribute to the regulation of arterial blood pressure. Stimulation of GABAA receptors in the medial nucfeus tractus sohtarius (mNTS) by muscimol, a GABAA agonist, increases arterial pressure and heart rate, afthough GABAA antagonism with bicu~ulline in mNTS decreases both parameters (16). Thus, it appears that GABAergic transmission is very important in adjacent NTS reflex pathways. If muscimol or bicuculline reached the NTS in pharmacologically active concentrations, it is probabie that some of the effects we observed when we infused these substances into AP were actually due to actions in NTS. Although possible, we would like to argue against such an explanation and in favor of the presence of GABAergic ~~diovascular regulation in both AP and NTS. In these experiments, we minimized the diffusional area of the infusates by using a nanoliter microinfusion technique. We estimate that infusion at the rates of 20 nllmin

for MK-801 and 10 nl/min for b~c~cul~~ne or muscjrno~ represent a maximum of 2% and 107, respectively, of the plasma flow rate in AR. These estimations were based on data and calculations of plasma how rate in rat AP by others (I 3). With our procedure, the total volume of injectate was similar to the total volumes that have been used by others studying the NTS with microinjection techniques (24) but was delivered over a period of minutes vs. a period of seconds. Although similar volume dye infusions cannot indicate the pharma~oio~ically active radius of a given drug infusion, the approximate extent of diffusion can be visualized. We reported that microinfusion of Evan’s blue dye into AP at the same rate used in these experiments resulted in staining limited to the AP (29). Microinfusion of 20 nI/min for 5 min in a highly perfused region like AP results in 11’75the volume of injectate or total dose of drug/unit time being delivered to a given site than would be delivered in a 50 nl injection in 2 s. We argue, therefore, that 50- f 00 bolus microinjections in NTS are far more likely to treat AP than the 20 nllmin microinfusions in AP are to treat NTS because our technique minimizes both the pressure and concentration gradients that promote interstitial spread of the in.jectate compared with bolus microinjection technique. In summ~y, the present studies provide evidence that GABAergic neurom~uiation in the AP can play an important role in neural control of AP cardiovascular functions. Furthermore, there appear to be functional interactions between NMDA

1. Auer, R. N.; Rod, M. R.; Whishaw, I. Q. MK-801 (dizocilpine) efficacy and limitations in cerebral ischemia. In: Meldmm, B. S.; Moroni, R.; Simon, R, P.; Woods, J. H., eds. Excitatory amino acids, New York: Raven press; 1991. 2. Blessing, W. W. Depressor neurons in rabbit caudal medulla act via GABA receptors in rostra1 medulla. Am. J. Pbysiol. 254:H686H692; 1988. 3. Borison, II. L. Ama postrema: Chemoreceptor ~irc~rnven~cul~ organ of the medulla obionga~. Prog. Neurobiol. 32:351-3w, 1989. 4. Buchan, A. M.; Slivka, A.: Xue, D. The effect of the NMDA receptor antagonist MK-801 on cerebral blood flow and infarct volume in ~x~~rnenta1 focal stroke. Brain Res. 574: 17I- 177; 1992. 5. Carpenter, D. 0.; Briggs, D. B.; Knox, A.P. and Strominger, N. Excitation of area posttema neurons by transmitters, peptides, and cyclic nucleotides. J. Neurophysiol. 59:358-369; 1988. 6. Catelli, J, M. ; Giakus, W. J.; Sved, A. F. GABAergic mechanisms in nucleustractus solitarius alter blood pressure and vasopressin release. Brain Res. 403279-289; 1987. 7. Corbett. D.: Evans. S.: Thomas. C.: Wang, D.; Jonas, R. A. MK-801 reduced cerebral ischemic injury by inducing hernia. Brain Res. 514:300-3Ort; 1990. 8. D’Amelio, F. E.; Mehler, W. R.; Gibbs, M. A.; Eng, L. F.; Wu, J.Y. Irnrn~~yt~herni~~ i~ali~tion of glutamic acid decarboxylase (GAD) and glu~ne syn~e~e (GS) in the area postrema of the cat. Light and electron microscopy. Brain Res. 410:232-244; 1987, 9. DeSarro, G. B.; DeSarro, A. Amiconvulsant properties of non competitive Zionism of the ~-M~hyl-~~~~ receptor in genetically epileps~prone rats: Comparison with CPpene. Neu~ph~macology 3251-58; 1993. 10. DiMicco, J. A.; Abshire, V. M. Evidence for GABAergic inhibition of a hypothalamic sympathoexcitatory mechanism in anesthetized rats. Brain Res. 402:1- 10, 1987. 11. Gordon, F, J. Spinal GABA receptors and central cardiovascular control. Brain Res. 328:165-169; 1985.

12. Gross. P. M.; Wainman, D. S.; Shaver, S. W.; Wall, K, M.; Ferguson, A. V. Metabolic activation of efferent pathways from the rat area postrema. Am. J. Physiol. 258:R788-R797; 1990. I?. Gross, P. M.; Wall, K. M.; Wainman, D. S.; Shaver, S. W. Subregional to~graphy of capil~a~es in the dorsal vagal complex of rats: II. physiological properties. J. Comp. Neurol. 306~83-94; 1991. 14. Hall, R.; Murdock, J. Brain protection: Physiological and pharmacological ~onsi~ration of capillaries in the dorsal vagal complex of rats: II Physiological properties. J. Comp. Neural. 306:83-94; 1991. 15. Klausmair, A.; Phillipu, A. Carotid occlusion increases &berelease of endogenous GABA in the nucleus of the solitary tract. Naunyn Schmiede~rgs Arch. Fha~a~ol. 340:764-766; 1989. 16 Kubo, T.; 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. 74:33 I -336; 1987. 17. Kubo, T.; Kihara, M.; Misu, Y. Ipsilateral but not contralateral blockade of excitatory amino acid receptors in the caudal ventrolateral medulla inhibits aortic baroreceptor reflex in rats. Naunyn Sehmiedebergs Arch. Pharmacol. 343:46-51; 1991. 18. Leslie, R. A.; Osborne, N. N. Amines and other transmitter-like compounds in the bovine area postrema. Brain Res. Bull. 13:357-362: 1984. 19. Lewis, S. J.; Banes, C.; Jacob, H. J.; Ohta, H.; Brady, M. J. Cardiovasdar effects of the ~-me~yl-D-~~~te receptor ~tagonist MK-801 in conscious rats. Hypertension 13:759-765; 1989. 20. Martin, D. S.; Segura, ‘I’.;Haywood, 3. R. Cardiovascular responses to bi~uculline in the p~aven~cul~ nucleus of the rat. ~y~~ension 1848-55; 1991. 21. Miselis, R. R.; Shapiro, R. E.; Hyde, T. M. The area postrema. In: Gross, P. M., ed. Circumventricular organs. vol. 2. Florida: CRC Press; 1987: 186-204. 22. Mitra, J.; Prabbaker, N. R.; Overholt, J. L.; Chemiack, N.S. Respiratory and vasomotor effects of excitatory amino acid on ventral medulhny surface. Brain Res. Bull. 1868 I -684; 1987.

receptor-mediated and CABA receptor-mediated modulation of cardiovascular functions transduced by neural circuitry of the rat AP.

MK-801 AND GABA,

67

INTERACTIONS AT AP

23. Perkins, W. J.; Lanier, W. L.; Karlsson, B. R.; Milde, J. H.; Michenfelder, J. D. The effect of the excitatory amino acid receptor antagonist dizocilpine maleate (MK-801) on hemispheric cerebral blood flow and metabolism in dogs: Modification by prior complete cerebral &hernia. Brain Res. 498:34-44; 1989. 24. Rettig, R.; Healy, D.; Printz, M. Cardiovascular effects of microinjection of angiotensin II into the nucleus tractus solitarii. Brain Res. 364:233-240; 1986. 25. Soltis, R. P.; DiMicco, J. A. Interaction of hypothalamic GABAA and excitatory amino acid receptors controlling heart rate in rats. Am. J. Physiol. 26l:R427-R433; 1991. 26. Strain, S. M.; Gwyn, D. G.; Rutherford, J. G.; Losier, B. J. Direct vagal input to neurons in the area postrema which project to the parabrachial nucleus: An electron microscopic-HRP study in the cat. Brain Res. Bull. 24:457-463; 1990. 27. Sun, M. K.; Spyer, K. M. GABA-mediated inhibition of medullary vasomotor neurons by area postrema stimulation in rats. J. Physiol. 436:669-684; 1991.

28.

29.

30. 31.

32.

33.

Sved, A. F.; Tsukamoto, K. Tonic stimulation of GABA receptors in the nucleus tractus solitarius modulates the baroreceptor reflex. Brain Res. 592:37-43; 1992. Tian, B.; Hartle, D. K. Cardiovascular effects of NMDA and MK801 infusion at area postrema and NT’S in rat. Phannacol. B&hem. Behav. (in press). Walberg, F.; Ottersen, 0. P. Neuroactive amino acids in the area postrema. Anat. Embryol. 185:529-545; 1992. Willette, R. N.; Barcas, P. P.; Krieger, A. J.; Sapru, H. N. Endogenous GABAergic mechanisms in the medulla and the regulation of blood pressure. J. Pharmacol. Exp. Ther. 230:34-39; 1984. Willette, R. N.; Krieger, A. J.; Barcas, P. P.; Sapru, H. N. Medullary gamma-aminobutyric acid (GABA) receptors and the regulation of blood pressure in the rat. J. Pharmacol. Exp. Ther. 226:893-899; 1983. Yang, C. P.; Liu, H. J.; Chou, L. S.; Lin, M. T. Effects of intra-raphe injection of amino acids on cardiovascular function in rats. Chin, J. Physiol. 35:143-149; 1992.