Modulation of baroreflex function by angiotensin II endogenous to the caudal ventrolateral medulla

BRAIN RESEARCH ELSEVIER

Brain Research 671 (1995) 38-44

Research report

Modulation of baroreflex function by angiotensin II endogenous to the caudal ventrolateral medulla Shogo Sesoko *, Hiromi Muratani, Shuichi Takishita, Hiroshi Teruya, Nobuyuki Kawazoe, Koshiro Fukiyama Third Department of Internal Medicine, University of The Ryukyus Medical School, 207 Uehara, Nishihara-cho, Okinawa 903-01, Japan

Accepted 25 October 1994

Abstract

Neurons in the ventrolateral medulla (VLM) mainly determine the tonic sympathetic activity. The caudal VLM (CVLM) relays baroreflex signals to the rostral VLM. We have reported that endogenous angiotensin II (ANG II) contributes to the ongoing activity of the VLM neurons. In the present study, we examined if ANG II endogenous to the CVLM modulates the baroreflex function in anesthetized normotensive Sprague-Dawley rats. Changes in renal sympathetic nerve activity (RSNA) in response to changes in mean arterial pressure (MAP) induced by i.v. infusion of phenylephrine and nitroglycerin were recorded before and after bilateral microinjection of [Sarl, Thr8]-ANG II, an ANG II antagonist, into the CVLM. The ANG II antagonist injection into the CVLM significantly increased MAP and RSNA by 17.6 + 8.0 mmHg (mean + S.D.) and 36.3 + 18.1%, respectively. It also significantly increased the baroreflex sensitivity (BS) from -0.49 + 0.38 to -0.74 + 0.37%/mmHg during nitroglycerin infusion. In contrast, the BS examined by phenylephrine infusion was not altered by the pretreatment with ANG II antagonist. Injection of artificial CSF affected neither the baseline values of MAP and RSNA nor the BS. These results suggest that ANG II endogenous to the CVLM exert a modulating role in baroreflex control of RSNA. Keywords: Angiotensin II; Angiotensin II antagonist; Caudal ventrolateral medulla; Baroreflex control; Blood pressure; Renal

sympathetic nerve activity

1. Introduction

The ventrolateral medulla (VLM) regulates the intrinsic activity of spinal preganglionic sympathetic neurons which innervate the peripheral vascular beds. Neurons in the rostral ventrolateral medulla (RVLM) send excitatory input directly to the spinal sympathetic preganglionic neurons [17] while neurons in the caudal ventrolateral medulla (CVLM) tonically inhibit the activity of the cardiovascular neurons in the R V L M [10]. Also, the C V L M is a nucleus that transmits baroreflex input from the nucleus of the solitary tract (NTS) to the R V L M [1,9,19]. Recently, Cravo et al. [7,11] demonstrated that there are at least two groups of

* Corresponding author. Fax: (81) (98) 895-2702. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0006-8993(94)01315 - 2

cardiovascular neurons in the CVLM; neurons that relay baroreflex input from the NTS to the R V L M neurons and neurons that send tonic inhibitory input to the R V L M neurons but are insensitive to the baroreflex input. Recent studies by us and other investigators have shown that there is an important interaction between the tonic activity of V L M neurons and angiotensin II ( A N G II). A N G II topically applied to the V L M [4] or microinjected into the R V L M and the C V L M [3,5,12,15] acts as an excitatory agent on the cardiovascular neurons in these sites. Further, A N G II endogenous to the R V L M and CVLM modulates the tonic activity of cardiovascular neurons in rabbit [15] and hypertensive and normotensive rats [13]. In the present study, we have investigated whether A N G II endogenous to the CVLM modulates the baroreflex control of the renal sympathetic nerve activity (RSNA).

S. Sesoko et aL / Brain Research 671 (1995) 38-44 2. M a t e r i a l s

and breathed room air spontaneously. The right femoral artery and bilateral veins were cannulated for measurement of arterial pressure and administration of drugs. Body temperature was kept within 37 + I°C using a heating pad. Arterial blood gases and pH measured at the end of the experiments were within normal limit. Anesthetized rats were fixed on a stereotaxic frame (Narishige Scientific Instruments, Japan) in a supine position. The trachea and

and methods

2. I. Animal preparation 16 male Sprague-Dawley rats (7-8 weeks of age, 230-320 g; Charles River Japan) were anesthetized with urethane (1.0 g/kg i.p.)

2ool-"

I

....

39

-~

AP

(mmHo) 0

200 E'100" ~ ~ .......... r

MAP

( mmHg)

•/

0 . ...

i

.

.

.

.

HR (bl~) i

..

~-,t~::......................! . . . . . . . . . . .

"l:-il I :t:!."

t

1 .i..

R S N A 2so

(~tw/n)

i I •

- ;~f~

I i i

H

IB

L@utamate Phmyto~rim 2 nrnol

AP ( mmHg )

f~

I I 30 sec

i.v. infusion

15 rain

Nitroglycerin i.v. infusion

100 0

MAP

( mmHg)

1~

-

0

HR (bm) .

.-

~ . ., ~

.

~. o ~ ,. 1

.

"

.

.

.i .

,:.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. _ .

.

.

.

(- ~'~.~-~,~,~.t,/-~---

¢.A.T

no. os

.

oEmi| u

.

m

ANG II mnlmgofl~ It. CVI,.M rt. CVI.M 100 pmoi 100 pmoi

H

PtwnylepNkw I.v. inlu~on

~ 30uc

15 min

m

nm

Nllmglycmln i.v. InfuJon

Hexamahonium 40 mgdtgi.v.

Fig. 1. Effect of [Sarl, ThrS]-ANG II (ANG II antagonist) microinjected into CVLM. Injection site was functionally verified by L-glutamate response. Administration of phenylephrine (i.v.) increased arterial pressure, MAP and decreased RSNA. In contrast, i.v. administration of nitroglycerin decreased arterial pressure, MAP and increased RSNA (upper panel). Microinjection of ANG II antagonist (100 pmol) into bilateral CVLM increased MAP and RSNA. Then, phenylephrine or nitroglycerin infusion was repeated. After ANG II antagonist injection, increases in RSNA were augmented in response to depression of arterial pressure with nitroglycerin (lower panel).

S. Sesoko et aL / Brain Research 671 (1995) 38-44

40

2.4. Test for baroreflex sensitivity (BS)

esophagus were transected in the lower neck and reflected rostrally. The distal trachea was cannulated to facilitate ventilation. After retraction of the bilateral longus capitis muscles, the inferior occipital bone was removed and the surface of the ventral medulla oblongata was exposed. Great care was taken not to damage the aortic depressor and the carotid sinus nerves. The left kidney was exposed via a transperitoneal approach. A branch of renal nerves was identified around the renal artery and vein distal to the left adrenal vein, was separated from surrounding tissues and was placed on a bipolar silver wire electrode (7855; A-M Systems, USA). When an optimal neurogram was obtained, the nerve and the electrode were embedded in silicone gel (Silige1604, Wacker, Germany) and allowed to harden. For measurement of RSNA, original renal nerve signals were amplified and filtered between 30 and 1000 Hz (DPA-100E; Dia Medical System, Japan). The amplified nerve pulses were counted with a spike-counter (DSE-325A; Dia Medical System).

We tested the BS of RSNA by changing the arterial pressure before and after the microinjection of ANG II antagonist (n = 10) or artificial CSF (n =6) into the CVLM bilaterally. In 10/10 rats, phenylephrine was infused slowly by using an infusion pump (model 975; Harvard Apparatus, USA) to raise MAP by ~ 50 mmHg in 2 min. In 9/10 rats, nitroglycerin was infused to lower MAP by ~ 40 mmHg in 20 s. The order of administration of phenylephrine or nitroglycerin was randomized. As an index of the BS, we used the slope of the regression line between changes in MAP and % changes in RSNA. After MAP returned to the control level, the ANG II antagonist or artificial CSF was injected into the CVLM bilaterally. The BS was then determined again. At the end of each experiment, hexamethonium (40 mg/kg) was injected i.v. to evaluate the background noise of RSNA.

2.5. Histological analysis 2.2. Microinjection procedure At the completion of each experiment, rats were perfused transcardially with 0.9% NaC1 followed by phosphate-buffered 10% formalin. The brainstem was removed, stored overnight in 10% phosphate-buffered formalin and then transferred to fixative containing 30% sucrose. Frozen brain tissue was sectioned in the coronal plane (50 /zm) and stained with neutral red. Microinjection sites were identified by deposition of Alcian blue dye and referred to standard anatomical structures of the brainstem according to Paxinos and Watson [14].

After the renal nerve and the electrode were embedded in silicone gel, the dura was incised and retracted to expose the ventral surface of the medulla, which was kept moist by endogenous cerebrospinal fluid (CSF). Multibarrel micropipettes with tip diameters of 20-50/zm, which were made from calibrated microbore capillary glass tubing (Accu-Full 90; Clay Adams, USA), were used for microinjections. Tips were drawn on a glass micropipette puller (type PE-2; Narishige Scientific Instruments). The inner surface was coated with silicon (Sigmacote; Sigma Chemical, USA). The injections (50 hi) were made over a 30-s period with a hand-held syringe. The injected volume was measured by observing the movement of the fluid meniscus along a reticule in a microscope. The CVLM was identified by injection of 2 nmol of L-glutamate based on the criteria of our previous studies [12,13] and corresponded to the injection sites located between the second and third rootlet of the hypoglossal nerve, 1.9-2.1 mm lateral to the midline and 0.7-0.9 mm below the ventral surface [12].

2.6. Statistical analysis Data were expressed as mean + S.D. Paired t tests were used to compare the effects of ANG II antagonist with those of artificial CSF on baseline values of MAP, heart rate (HR) and RSNA. Linear regression between changes in MAP and % changes in RSNA was calculated by the least-squares method. Wilcoxon's signed-rank tests were used to compare the changes in BS obtained before and after the microinjection of ANG II antagonist or artificial CSF. A P value of < 0.05 was considered to be statistically significant.

2.3. Drugs

3. Results

[Sarl, Thr8]-ANG II, an ANG II antagonist (100 pmol), and L-glutamate were dissolved in artificial CSF (in mM: 133.3 NaCI, 3.4 KCI, 1.3 CaCI2, 1.2 MgCI2, 0.6 NaHzPO4, 32.0 NaHCO 3 and 3.4 glucose). 10 nl of an emulsion of Alcian blue dye was used to mark the site of injection.

Baseline values of MAP

AHR

&MAP (bpm)

( mmHg )

and HR

in t h e r a t s f o r

i n j e c t i o n o f A N G II a n t a g o n i s t ( n = 10) w e r e 99.5 + 6.5

ARSNA

(%)

60]

801

40

60

40

2O ]

ilk

0

ANG II

antagonist

artificial

CSF

40 20 -

0

g.$.

o

I

n.s. J

-20

- -J n.s.

ANG II artificial antagonist CSF

0 ANG II

antagonist

artificial CSF

Fig. 2. Changes in baseline values after microinjection of [Sarl, Thr8]-ANG II (ANG II antagonist, n = 10) or artificial artificial CSF (n = 6) into CVLM. MAP (left), HR (middle) and renal sympathetic nerve activity (right) were increased by injection of ANG II antagonist. * P < 0.01 ANG II antagonist or artificial CSF vs. control. * * P < 0.005 ANG II antagonist or artificial CSF vs. control.

S. Sesoko et aL / Brain Research 671 (1995) 38-44 CSFinjection . . . . Control

ANG II antagonist-

- -

- -

injection

. . . . Control

41

+60

artificial

+60 -+40

-+.o

< +20 z rr <3 0

p=0.02 [

,

.

,

-50

.

-40

,

.

-30

,

.

,

-20

-20

.

n.s.1-~ ~ . ~

•

t0 . . . . . . . . . . . -50 -40 -30 -20 -10

-10

z ~o n-

+20

AMAP (mmHg)

0

20

&MAP (mmHg)

ANGII antagonist-injection

artificial CSF-injection

Fig. 3. Changes in BS during decreases in blood pressure with nitroglycerin. Left: BS was increased after microinjection of [Sarl, Thr8]-ANG II ( A N G II antagonist, n = 9). Right: BS did not significantly change after artificial CSF (n = 6). P = 0.02 A N G II antagonist vs. artificial CSF.

mmHg and 406 + 48 bpm, respectively. In the rats for the treatment with artificial CSF (n = 6), baseline values of MAP and HR were 100.9-t-11.5 mmHg and 425 + 60 bpm, both of which were similar to those in the rats receiving the ANG II antagonist injection. Bilateral injection of ANG II antagonist into the CVLM increased MAP, HR and RSNA (Fig. la). Mean increases in MAP, HR and RSNA were 17.6 + 8.0 mmHg, 10 + 9.4 bpm and 36.3 + 18.1%, respectively. All these changes evoked by the injection of ANG II antagonist into the CVLM were statistically significant (Fig. 2). On the other hand, the injection of the artificial CSF into the CVLlVl caused no significant changes in baseline values of these parameters. Sensitivity of the baroreflex control of the RSNA was examined before and after the bilateral injection of the ANG II antagonist or the artificial CSF into the CVLM by elevating and reducing blood pressure (Fig. 1). As shown in Fig. 3, the ANG II antagonist injected into the CVLM significantly increased the BS during decreases in blood pressure with i.v. infusion of nitro-

glycerin from - 0 . 4 9 + 0.38% to - 0.74 + 0.37%/mmHg (n = 9, P = 0.02). In contrast, during increases of blood pressure with i.v. infusion of phenylephrine, ANG II antagonist did not elicit any significant change in the BS (n = 10, Fig. 4). The injection of artificial CSF had no significant effect on the BS (n = 6, Figs. 3,4). Values of each BS are shown in Table 1. Fig. 5 is a composite picture of the locations where ANG II antagonist or artificial CSF was injected. Depressor responses were produced by injections of ANG II antagonist restricted to a region ventral to the nucleus ambiguus and dorsal to the lateral reticular nucleus. According to Paxinos and Watson [14], these regions are in medullary sections extending from 4.3 to 5.3 mm caudal to interaural line.

4.Discussion In the present study, ANG II antagonist injected into the CVLM elevated arterial pressure, HR and &MAP (mmHg)

&MAP (mmHg) 0

0

-20- ~

, 10

.

,

~

20

~

.

3,0

.

40

.

5,0

.

.< -40. Z cO fie -60. <3

•

10 ,

.

20 ,

.

30 i

.

40 L

.

50 i

-20

-~

~ -40 Z

n.s. co ~ n- -6o-

-nn.s.

-80-

-80-100.

0

ANG II antagonistinjection . . . . Control

- -

ANGII antagonist-injection

-100-

--

artificial CSFinjection . . . . Control

artificial

CSF-injection

Fig. 4. Changes in BS during increases in blood pressure with phenylephrine. There were no significantly changes in BS either after microinjection of [Sarl, ThrS]-ANG II ( A N G II antagonist, n = 10) or after artificial CSF (n = 6).

S. Sesoko et aL / Brain Research 671 (1995) 38-44

42

Table 1 Changes in BS after microinjection of A N G II antagonist or artificial CSF A N G II antagonist

Nitroglycerin Phenylephrine

Artificial CSF

Before

After

Before

After

- 0 . 4 9 ± 0.38 (9) - 1.37 ± 0.42 (10)

- 0 . 7 4 ± 0.37 * (9) - 1.42 ± 0.50 (10)

- 0 . 5 6 ± 0.30 (6) - 1.55 ± 0.49 (6)

- 0 . 4 7 ± 0.25 (6) - 1.53 ± 0.50 (6)

Values are m e a n + S.D. (in % z I R S N A / A m m H g ) ; n u m b e r of rats studied under different conditions is given in parentheses. *P = 0.02 vs. Before A N G II antagonist.

RSNA and it also increased BS of RSNA tested by decreases in arterial pressure. These results suggest that A N G II present in the CVLM under physiological conditions acts as a tonically excitatory substance in this region and it also modulates the baroreflex control of RSNA. We have already demonstrated that A N G II endogenous to the VLM contributes to maintain the ongoing activity of cardiovascular neurons in the RVLM and CVLM of spontaneously hypertensive and normotensive Wistar-Kyoto rats [13]. Sasaki and Dampney [15] also reported the contribution of endogenous A N G II in the VLM to the tonic level of hemodynamic parameters and RSNA in rabbit. The results that arterial pressure, H R and RSNA increased after the A N G II antagonist injection into the CVLM are consistent with the results presented in previous papers [12,13,15]. The modulation of the baroreflex control of RSNA by local A N G II in the CVLM is a new finding. There has been a controversy whether the CVLM relays

baroreflex input from the NTS to the RVLM. Blessing and Willoughby [6] did not observe such baroreflextransmitting function by CVLM neurons. A number of investigators, however, have reported evidence for the involvement of CVLM neurons in the baroreflex function. The depressor response to stimulation of the aortic depressor nerve or the NTS was abolished [9] or converted to a pressor response [1] after suppression of CVLM neurons. In addition, a direct projection from barosensitive neurons in the CVLM to the RVLM neurons has been clearly demonstrated in rats [1] and rabbits [18]. Recently, Cravo et al. [7,11] have shown that two groups of sympathoinhibitory neurons reside in the CVLM: barosensitive neurons which transmit the baroreceptor input from the NTS to the RVLM and those which tonically inhibit the sympathoexcitatory neurons in the RVLM independent of the baroreflex. Our results suggest that A N G II endogenous to the CVLM acts at least on the first group of neurons in the CVLM to influence the baroreflex control of

artificial CSF

ANG II antagonist 1 mm

4.3 mm Amb

4.68 mm

Amb 4.8 mm

5.08 mm

~

Amb

]

)J

5.3mm Fig. 5. Injection sites of [Sarl, Thr8]-ANG II ( A N G II antagonist, solid circles) or artificial CSF (solid triangles) in VLM. Amb, nucleus ambiguus; IO, inferior olivary nucleus; LRt, lateral reticular nucleus; nTS nucleus of tractus sol±tar±us; 12, hypoglossal nucleus; 12n, root of hypoglossal nerve (according to Paxinos and Watson [14]).

S. Sesoko et al. / Brain Research 671 (1995) 38-44

RSNA. Histological analysis indicates that injection sites of ANG II antagonist or artificial CSF were located between the nucleus ambiguus and the lateral reticular nucleus at level from 4.3 to 5.3 mm caudal to interaural line. These sites correspond to the CVLM region where barosensitive neurons reside [7,11]. Selective blockade of N-methyl-D-aspartic acid (NMDA) receptors in the CVLM has been shown to abolish the baroreceptor reflex elicited by aortic nerve stimulation [9,16]. These previous reports suggest that the main neurotransmitter concerned with baroreflextransmitting function in the CVLM is an excitatory amino acid acting on NMDA receptors. Furthermore, the presence of ANG II receptors in the CVLM has been confirmed [2] whereas no previous report has demonstrated ANG II-containing fibers from the NTS to the VLM. Taken together, it may be reasonable to consider that ANG II endogenous to the CVLM may acts as a neuromodulator. However, the precise mechanisms by which ANG II modulates the baroreflex function and stimulates the depressor neurons in the CVLM still remain to be determined. The elevation of baseline blood pressure after injection of ANG I1 antagonist might influence BS. Dorward et al. [8] showed that the gain of baroreflex control of RSNA decreased during resetting associated with phenylephrine-induced rises in resting MAP. In contrast, in our present study, ANG II antagonist microinjected into the CVLM facilitated the BS during decreases in blood pressure in the presence of an elevation of the baseline blood pressure. This fact suggests that the facilitation of baroreflex function by ANG II antagonist injected into the CVLM was not due to the change in the baseline blood pressure. The ANG II antagonist did not alter the BS when tested by elevating arterial pressure with infusion of phenylephrine. As mentioned above, the elevated baseline blood pressure might have suppressed BS facilitated by inhibition of ANG II endogenous to the CVLM. In addition, a direct sympathoexcitatory pathway from the NTS to the RVLM [20], which is unmasked when neuronal activity of the CVLM is depressed, might have attenuated the depression of RSNA induced by elevation of blood pressure. A limitation of our present study is that we could not elucidate which type of the ANG II receptor, AT1 or AT2, modulates the baroreceptor reflex sensitivity in the CVLM. Several kinds of non-peptide antagonists of ANG II receptor are now available. Especially, losartan is widely used for blocking AT1 receptor. However, it was reported that this substance had nonspecific activity to inhibit both actions of L-glutamate and ANG II in the VLM while [Sarl, Thr8]-ANG II has specific inhibitory effects on ANG II receptors [19]. Therefore, we decided to use this peptide antagonist in our experiments.

43

ANG II antagonist injected into the CVLM of Sprague-Dawley rats elevated the baseline level of MAP, HR and RSNA and it facilitated the baroreceptor reflex sensitivity of RSNA during decreases in blood pressure. These findings provide evidence that endogenous ANG II in the CVLM exerts a tonic excitatory action on depressor neurons and it also modulates the baroreflex control of RSNA in normotensive rats.

Acknowledgments We thank Ms. R. Matayoshi for excellent technical assistance. Parts of these results were presented in the 66th Scientific Session of American Heart Association.

References [1] Agarwal, S.K., Gelsema, J. and Calaresu, F.R., Inhibition of rostral VLM by baroreceptor activation is relayed through caudal VLM, Am. J. Physiol., 258 (1990) R1271-RI278. [2] Aldred, G.P., Chai, S.Y., Song, K., Zhuo, J., Macgergor, D.P. and Mendelsohn, F.A.O., Distribution of angiotensin II receptor subtypes in the rabbit brain, Regul. Pept., 44 (1993) 119-130. [3] Allen, A.M., Sasaki, S., Dampney, R.A.L., Mendeisohn, F.A.O. and Blessing, W.W., Action of angiotensin II in the ventrolateral medulla oblongata. In G. Kunos and J. Ciriello (Eds.), Central Neural Mechanisms in Cardiovascular Regulation, Springer, New York, NY, 1991, pp. 95-103. [4] Andreatta, S.H., Averill, D.B. and Santos, R.A.S., The ventrolateral medulla: a new site of action of renin-angiotensin system, Hypertension, 11 (1988) 1163-1166. [5] Averill, D.B., Muratani, H., Madalin, K.J. and Ferrario, C.M., Cardiovascular action of angiotensin II in the ventrolateral medulla. In G. Kunos and J. Ciriello (Eds.), Central Neural Mechanisms in Cardiovascular Regulation, Springer, New York, NY, 1991, pp. 104-121. [6] Blessing, W.W. and Willongby, J.O., Depressor neurons in rabbit caudal ventrolateral medulla do not transmit the baroreceptor-vasomotor reflex, Am. J. Physiol., 253 (1987) H777-H786. [7] Cravo, S.L., Morrison, S.F. and Reis, D.J., Differentiation of two cardiovascular regions within caudal ventrolateral medulla, Am. J. Physiol., 261 (1991) R985-R994. [8] Dorward, P.K., Riedel, W., Burke, S.L., Gipps, J. and Korner, P.I., The renal sympathetic baroreflex in the rabbit: arterial and cardiac baroreceptor influences, resetting, and effect of anestesia, Circ. Res., 57 (1985) 618-633. [9] Gordon, F.J., Aortic baroreceptor reflexes are mediated by NMDA receptor in caudal ventrolateral medulla, Am. J. Physiol., 252 (1987) R628-R633. [10] Granata, A.N., Numao, Y., Kumada, M. and Reis, D.J., Al Noradrenergic neurons tonically inhibit sympathoexcitatory neurons of C1 area in rat brainstem, Brain Res., 377 (1985) 127-146. [11] Eske, I., Morrison, S.F., Cravo, S.L. and Reis, D.J., Caudal ventrolateral medulla (CVLM): source of tonic and baroreceptor-mediated sympathoinhibition, Am. J. Physiol., 264 (1993) R169-R178. [12] Muratani, H., Averill, D.B. and Ferrario, C.M., Effect of angiotensin II in ventrolateral medulla of spontaneously hypertensive rats, Am. J. Physiol., 260 (1991) R977-R984.

44

S. Sesoko et aL /Brain Research 671 (1995) 38-44

[13] Muratani, H., Ferrario, C.M. and Averill, D.B., Ventrolateral medulla in spontaneousry hypertension rats: role of angiotensin II, Am. J. Physiol., 264 (1993) R388-R395. [14] Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, 2nd Ed., Academic, New York, NY, 1986 [15] Sasaki, S. and Dampney, R.A.L., Tonic cardiovascular effects of angiotensin II in the ventrolateral medulla, Hypertension, 15 (1990) 274-283. [16] Somogyi, P., Minson, J.B., Morilak, D., Llewellyn-Smith, I.J., Mcilhinney, J.R.A. and Chalmers, J.P., Evidence for an excitatory amino acid pathway in brainstem and for its involvement in cardiovascular control, Brain Res., 496 (1989) 401-407. [17] Sun, M., Young, B.S., Hackett, J.T. and Guyenet, P.G., Reticulospinal pacemaker neurons of the rat rostral ventrolateral medulla with putative sympathoexcitatory function: an intracellular study in vitro, Brain Res., 442 (1988) 229-239.

[18] Terui, N., Masuda, N., Saeki, Y. and Kumada, M., Activity of barosensitive neurons in caudal ventrolateral medulla that send axonal projections to the rostral ventrolateral medulla in rabitts, Neurosci. Lett., 118 (1990) 211-214. [19] Tsuchihashi, T., Khosla, M.C., Ferrario, C.M. and Averill, D.B., DUP753, AT1 angiotensin II (ANGII) antagonist, attenuates pressor and sympathoexcitatory responses evoked by ANGII and L-glutamate (L-GIu) in the rostral ventrolateral medulla (RVLM) (Abstract), FASEB J., 2 (1992) Al165. [20] Urbanski, R.W. and Sapru, H.N., Evidence for a sympathoexcitatory pathway from the nucleus tractus solitarii to the ventrolateral medullary pressor area, J. Auton. Nerv. Syst., 23 (1988) 161-174.