Neuroscience Vol. 46, No. Printed in Great Britain

I, pp. 91-100,

0306-4522/92

1992

$5.00 + 0.00

Pergamon Press plc IBRO

BARORECEPTOR REFLEX INHIBITION INDUCED BY THE STIMULATION OF SEROTONIN, RECEPTORS IN THE NUCLEUS TRACTUS SOLITARIUS OF THE RAT N. MERAHI, H. S. ORER, A.-M. LAPORTE, H. GOZLAN, M. HAMON and R. LAGUZZI* INSERM U288, Neurobiologie

Cellulaire et Fonctionnelle, CHU Pitit-Sal@triBre, 91, Boulevard de l’Hbpita1, 75634 Paris Cedex 13, France

Abstract-Previous studies suggested that in the nucleus tractus solitarius, cardiovascular responses to serotonin may involve the simultaneous activation of more than one receptor subtype. In the present study, the cardiovascular effects of the local application of serotonin and different serotonin, agonists and antagonists into the nucleus tractus solitarius were analysed in intact and unilaterally ganglionectomized rats. Unilateral injections of serotonin (5-15 nmol) produced a dose-dependent increase in blood pressure and partially antagonized the arterial baroreflex responses evoked by an i.v. injection of phenylephrine. Similar blood pressure responses were obtained after unilateral microinjections of phenylbiguanide (5 nmol) and 2-methyl-serotonin (5 nmol), two serotonin, receptor agonists. Bilateral microinjections of serotonin or phenylbiguanide produced more pronounced blood pressure effects and antagonized completely the baroreflex responses. Both blood pressure and baroreflex effects were antagonized by prior injections of specific serotonin, antagonists such as zacopride (100 pmol) and ondansetron (100 pmol). Concomitant autoradiographic studies performed in intact and ganglionectomized rats, using [‘Z51]iodozacopride, confirmed that serotonin, receptors in the nucleus tractus solitarius are mainly located on vagal afferent fibers. In addition, serotonin microinjections made in the nucleus tractus solitarius ipsilateral to the ganglionectomy revealed a significant reduction in cardiovascular responses compared to intact animals. These results suggest that in the nucleus tractus solitarius of the rat, serotonin is involved in the reflex regulation of blood pressure through the stimulation of serotonin, receptors presumably located on vagal afferent fibers. Since bicuculline antagonized the serotonin-mediated pressor responses, a serotoniqdependent activation of an inhibitory GABAergic system within the nucleus tractus solitarius might be involved in blood pressure regulatory mechanisms

A large amount of evidence indicates that serotonin (5-HT) plays a regulatory role in the central control of blood pressure. The cardiovascular role of S-HT probably includes the nucleus tractus solitarius (NTS), a key structure in the integration of baro- and chemoreceptor messages which is heavily innervated by centra17.32 and peripheral” serotonergic fibers. In this respect, the levels of the 5-HT metabolite, 5-hydroxyindole acetic acid, in the NTS have been proposed as a neurochemical marker for a hypertensive state2 and a relationship between the changes in 5-HT metabolism and the reflex control of the blood pressure has been demonstrated in sino-aortic denervated animals.’ Local microinjections of 5-HT into the NTS also produced well defined cardiovascular responses.4.‘3,3’,33While low doses (picomoles) of 5-HT elicited an immediate decrease in blood pressure, ‘3.31higher doses (nanomoles) produced a delayed increase.33 These results suggest that 5-HT exerts multiple actions through the activation of

several receptor types. Our previous data have shown that 5-HT, receptors are responsible for the depressor responses, 3’ but the nature of the receptor(s) that mediate the pressor responses is unknown to date. Recent autoradiographic studies have shown that most 5-HT receptor subtypes exist in the brainstem.L8,24,26 Nevertheless their functional role(s) still remain(s) to be elucidated. 5-HT, receptors in the dorsal vagal complex which includes the NTS and the area postrema are thought to intervene in vomiting responses in different animal models.’ There is a general agreement that in most species, including the rat which has no vomiting responses, the highest density of 5-HT, receptors is found within the NTS region. 26Additionally, in this nucleus, these receptors were found to be located on the vagal sensory afferentsz5 suggesting that they are capable of interfering with baro- and chemoreceptor messages. These findings may be compatible with the involvement of 5-HT, receptors in the cardiovascular responses to 5-HT. In order to explore this hypothesis we analysed the cardiovascular effects of the microinjection of selective 5-HT, receptor agonists and antagonists into the NTS of normal and previously nodose ganglionectomized rats. The fate of 5-HT, receptors in these animals was directly investigated using an autoradiographic approach with [‘*‘I]iodo-zacopride as a 5-HT, ligand.”

*To whom correspondence should be addressed. BRS, baroreceptor reflex sensitivity; DOB, 4-bromo-2,5-dimethoxy-phenylisopropylamine; DOI, 4-iodo-2,5-di-methoxy-phenylisopropylamine; HEPES, N-2-hydroxyethylpiperazine-N’-2-ethane sulphonic acid; 5-HT, serotonin; 2-methyl-5-HT, 2-methyl-serotonin; ICS 205-930, (3a tropanyl) lH-indole-3-carboxylic acid ester; NTS, nucleus tractus solitarius.

Abbreviarions:

91

92

N. MEK~HI

EXPERIMENTAL

PROCEDURES

All experiments were performed Sprague Dawley rats (250- 300 g).

on

adult

male

NO&W ganglionectom) Under pentobarbital anesthesia (40 mg/kg, i.p.) a midline Incision was made on the ventral surface of the neck and neck muscles were retracted in order to reach the left carotid area. The left vagus nerve and the nodose ganglion were carefully freed from the carotid artery and the sympathetic chain lo which they are closelv associated. The nodose ganglion was removed after cutting its distal and proximal ends. Incisions were closed up and the animals were allowed to recover for two weeks before the experiments. Hlootl pressure rec,ording.v and microiqjectionv inro the nucleus fruclus so1iruriu.r General procedures. Rats were anesthetized with penlobarbital (40mg/kg, i.p.) and polyethylene cannulae were inserted into the femoral artery and vein for blood pressure heart rate recordings and drug injections. respectively. Core body temperature was maintained at 37 ‘C by a temperature controller. The rat was placed in a stereotaxic frame with the head inclined downward (I 0 mm below the nasal zero) and the dorsal surface of the brainstem was exposed through a partial craniotomy at the level of the calamus scriptorius. Glass micropipettes (70.--90 pm external diameter) containing drug solutions or vehicle were placed in the medial part of the NTS. 0.3 mm lateral to the caudal tip of the area postrema and O.Smm beneath the dorsal surface of the medulla. Microinjections of 0. I p I of various drug solutions were made over I s with a microinfusion pump. Fifteen seconds after each injection the micropipette was removed from the NTS. The same micropipette was used for bilateral inJections of a given drug. The time interval between two symmetrical microinjections was less than I min. &onisf do.se response wwe.v. Dose-dependent eflects of the agonists were investigated by comparing the cardiovascular responses evoked by the microinjection of only one given dose of a drug to that evoked by the prior vehicle injection in the same rat. Each dose of the agonist was tested individually in six rats. In some experiments methylene blue was injected with a drug solution to verify the pipette implantation sites in freshly cut tissue. ,dgonisr ~unrugonist interucfion.\. To test the drug antagonism, the agonist was injected first and, 30min later, a second micropipette filled with the antagonist was lowered into the NTS. The injection was made and the pipette was pulled up. Then a new injection of the agonist was realized 2 IO min later into exactly the same site. Buroreceptor reflex .sensitirGr~. The baroreceptor reflex sensitivity (BRS) was determined prior and 2 30 min after the microinjection of 5-HT or 5-HT, agonists into the NTS. BRS was expressed as the ratio of the maximal reflex bradycardia over the maximal increase in mean blood pressure following an i.v. bolus injection of phenylephrine (5 fig/kg).# Stdstical unulysis. For each experimental procedure. a global statistical analysis was performed using ANOVA. Then the comparison between two groups (control and treated) was made using Student’s t-test for dependent or independent groups as appropriate. All values were expressed as means + S.E.M. Aumradiogruphic

studies

Rats were killed by decapitation, and their brains were removed and immediately frozen in isopentane at -30°C. Histological sections (20 pm thick) were cut at the level of the brainstem and mounted on microscope glass slides covered by gelatin (1%) and chromic potassium sulphate (0.05%). After storage at -20‘C (for less than two weeks) and a progressive return to room temperature. slides were

(‘I cd

pre-Incubated for 30min at 20 C 11, 5OmM HEPES buffer (pH = 7.4), then incubated for I h at 20 C in the same hufl‘el containing 0.15 nM [I” I]iodo-zacopride (specific radio activity: I 100 Ciimmol). The non-specific binding was that which persisted on adjacent sections incubated in the same medium plus I .O FM ICS 205-930. a specific 5-HT, anlagonist.” After incubation, sections were rinsed in 50 mM HEPES buffer at 4 C for 2 x 5 min. quickly dipped in distilled water and then dried under a stream of cold air They were finally apposed to tritium sensitive Hyperfilm (Amersham International) and exposed in the dark for three days at 4’C. Autoradiograms were developed in Mlcrodol (Kodak) for IOmin. al 2o’C Quantitative analyst> was made with an IMSTAR densitometer. Optical densities wcrc converted to fmoles of radioligand bound per mg ttssuc b) computer fitting of the curve drawn from “$1 standards (Amcrsham International ). />rug.t (Boehringer Mannheim). prdzoan-HC‘i (Sandoz). zacopride (Delalande), ondansetron (CR 38032F. Glaxo). 4-bromo-2.5 dimethvamphetamine‘ (DOB, generously given by Dr Casse&. Facultk de Pharmacie, ChBtenay-Malabry. France) and 2-methyl-serotonin (?-methyl-5-HT. Sandoz) were dissolved in saline. 5-HT (Sigma) and ketanserin (Janssen) were dissolved in saline containing I% ascorbic acid. Phenylbiguanide (Aldrich) and bicuculline (Sigma) were dissolved m artificial cerebrospinal fluid (NaCI: I26 mM. KU. 3.5mM, NaH,PO,: 1.2mM, MgCI,: I.3 mM, C‘aCI, 2 mM, NaHCO,: 25 mM. glucose: I I mM, gassed with 0, (95%) + CO, (SW)] and saline, respectively, and hulfcrcd with KH,PO, (pH = 7.4) Phenylephrine

(Pfizer), methysergide

RESULTS E.uperiments

in intucr

Curdioousculur

rut.5

responses

to

serotonin.

As

pre-

viously observed in rats anesthetized with chloraof 20 pmol 5-HT lose.“’ unilateral microinjections into the NTS produced a decrease in mean blood pressure ( - 30.0 + 4.1 mmHg; n = 6) and heart rate (-28.0 _t 5.5 b.p.m.; n = 6) in animals anesthetized with pentobarbital. In the latter animals, the latency between the 5-HT injection and the onset of the cardiovascular responses was l--2 s. The 5-HTinduced decrease in blood pressure lasted for 30.-35 s, but no precise duration time could be ascertained for the heart rate changes. By contrast, unilateral microinjection of higher doses of S-HT elicited dose-dependent increases in blood pressure. These responses for 5, IO and I5 nmol 5-HT were + 14.20 ) 1.83 mmHg (n = 6), + 33.40 f and (n = I I) +24.36 + 1.20 mmHg 2.42 mmHg (n = 8). respectively (P < 0.001. compared to vehicle) (see also Fig. ]A). The onset latency of the pressor responses was 20-30 s and the peak was reached approximately 2 min later. Usually, blood pressure returned to its pre-injection level within 5 ~7min after the peak effect. No consistent heart rate changes were detected in association with the blood pressure changes, although a slight increase in the heart rate could be observed in some experiments. In addition, nanomolar doses of 5-HT occasionally elicited an initial (immediate) hypotensive rcsponsc which preceded the delayed hypertensive response.

NT’S: role of serotonin, receptors in cardiova~uiar

regulation

93

Cardiovascular responses to serotonin, agonists. AS

(A)

$

1 min

5-HT

t Zocopride

+

5-Hi

lmin

Fig. 1. Inhibition by zacopride of serotonin-induced cardiovascular changes. The unilateral microinjection of S-HT (10 nmol) into the NTS elicited an increase in blood pressure (A). The same dose of 5-HT produced no response when injected 5 min after the microinjection of zacopride (100 pmol) in exactly the same site (B). Both experiments (A, B) were performed in the same rat with a 30 min interval

(see Experimental Procedures). Similar observations were made in six animals (see Table 2). MBP, mean arterial pressure; BP, blood pressure; HR, heart rate. Arrows indicate the times for the microinjections.

However, these effects were too variable to allow their systematic study. Bilateral microinje~tion of IO nmol 5-HT into the NTS produced more pronounced increases in the blood pressure (+33.30 + 7.05 mmHg, n = 6) than unilateral microinjections of the same dose. In addition to increasing the basal blood pressure, the microinje~tion of nanomolar doses of 5-HT into the NTS also affected the baroreceptor reflex response normally triggered by an i.v. administration of 5 fig/kg phenylephrine (BRS: 1.75 + 0.43 bpm/mmHg, n = 6). Thus, a significant reduction (-37%, P < 0.05, n = 5) of BRS was observed 10min after a unilateral microinjection of 10 nmol5-HT into the NTS. A much larger inhibition (-87%, P < 0.01, n = 6) could be induced by the bilateral microinjection of the same dose of the indoleamine. Such inhibitory effects of 5-HT lasted for approximately 20 min after the microinjection.

did nanomolar doses of 5-HT, unilateral microinjections of the selective 5-HT, receptor agonists 2-methyl-5-HT and phenylbiguanide into the NTS produced a delayed increase in mean blood pressure. This increase was of 14.66 f 1.46 mmHg (n = 9) and 13.01 f. 3.02 mmHg (n = 8) for 2-methyl-S-HT (5 nmol) and phenylbiguanide (5 nmol), respectively. The latency of onset and the amplitude, as well as the duration of the response, were similar to those observed with S-HT. At the same dose (5 nmol) as that used for the unilateral treatment, the bilateral microinjection of phenylbiguanide induced a much larger increase in blood pressure (Table 1). Further studies with various doses below and above 5 nmol clearly indicated that phenylbiguanide administration into the NTS produced a dose-dependent increase in blood pressure (Table I). Like that observed after the bilateral microinjection of 5-HT, the peak effect was reached approximately 2 min after the 2nd injection of phenylbiguanide. Then blood pressure returned to its preinjection level within 8-10min following the peak effect. As found with S-HT, a complete reduction in the baroreceptor reflex response to i.v. phenylephrine was also observed after bilateral microinjections of phenylbiguanide (10 nmol) into the NTS (Fig. 2A. B). The inhibitory effect of phenylbiguanide lasted for approximately 20mir1, and was clearly dose-dependent when tested against the baroreceptor reflex response to phenylephrine injected (i.v.) 10min later (Table I). Antagonist effects. Serotonin, and serotonin, antagonists. The unilateral microinjection of methysergide (100 pmol, n = 6) or ketanserin (100 pmol, n = 6) into the NTS did not

produce any significant changes in the mean blood pressure or heart rate. In addition, both antagonists failed to prevent the blood pressure increase due to the local administration of 5-HT (10 nmol, n = 6) or phenylbiguanide (10 nmol, n = 6) (Table 2). However, as previously reported,3” a lower dose of ketanserin (10 pmol) totally prevented the reduction in blood pressure and heart rate due to the local administration of 20 pmol S-HT. Serotonin, antagonists. The unilateral microinjection of zacopride (100 pmol, n = 6) or ondansetron (100 pmol, n = 6) into the NTS exerted no significant influence on blood pressure and heart rate. However, both antagonists prevented the blood pressure effects induced by phenylbiguanide (10 nmol, n = 6) (Table 2) or 5-HT (10 nmol, n = 6) (Fig. 1B, Table 2). The duration of the antagonistic effects of ondansetron or zacopride upon both agonists lasted approximately 20min. In addition, the bilateral microinjection of zacopride (IOOpmol) totally suppressed, for at least 20min, the inhibitory effect of the local microinjection of phenylbiguanide (10 nmol, bilaterally) on the baroreceptor reflex response to iv. phenylephrine (Fig. 2Cj. a,-Adrenergic receptor blockade. In order to assess

94

N.

MERAHI

(Al

p---A_

- -_-_.i_-_-_

- . ..-_..._..__---___ ~/-___,_.__ _____.___.___ .-..

CKE xoD

4

Biioteral vehicle

4

4

1 min

i.v. Phenylephnne

(8)

__&--a kg 2oo -E

!

0

OL

4

4

4 i.v. Phenybphrine

Bilateral PBF

-

1 min

(Cl

kg 2oo -

- 0[

_

the participation of the sympathetic nervous system in the pressor responses to the microinjection of phenylbiguanide into the NTS, another series of experiments (n = 4) was performed in rats pretreated with the a,-receptor antagonist prazosin (2 mg/kg. i.v.) 5 min prior to phenylbiguanide microinjection. Indeed, prazosin pretreatment totally antagonized ( -94%) the rise in blood pressure normally induced by the bilateral microinjection of phenylbiguanide (5 nmol) into the NTS. GABA,

I

0’

rt ul.

200r

so0

Of4 Biloterol

zacopride

44

Biictsrol PBG

4

i.v. Phenylephrine

_ IlTllIl

Fig. 2. Inhibition of the baroreceptor reflex by a bilateral microinjection of phenylb~~auide (10 nmol) into the NTS-antagonism by zacopride of the phenylbi~anide effects. (A) The increase in blood pressure produced by phenylephrine (5 pg/kg, i.v.) triggered a reflex decrease in heart rate in a pentobarbital-anesthetized rat that received prior miffoinj~tions of 0.1 ~1 artificial cerebrospinai fluid into both NTS (“bilateral vehicle”). (B) Prior biiateral microinjections of phenylbiguanide (PBG) (10 nmol) into the NTS inhibited the heart rate reflex response normally induced by phenylephrine (5 pg/kg, i.v.). (C) The heart rate refiex response was recovered when zacopride (100 pmoi) was microinjected bilaterally into the NTS 2--3.5 min prior to phenylbiguanide (10nmol) exactly in the same sites. Recordings A and B were obtained in the same rat, and recording C was from another rat. Similar observations were made in at feast six rats for each of the experimental conditions. Arrows indicate the times for the microinjections. Abbreviations are as in the legend to Fig. I.

receptor blockade. The bilateral

microin-

jection of bicuculline (100 pmol, n = 6) into the NTS elicited an immediate hypotensive and brady~rdia~ response (-26.50 + 6.90 mmHg and -46.25 + 15.54 b.p.m., respectively). After such a pretreatment, the local microinjection of 5-HT (10 nmol, bilaterally, n = 6), 10 min later, produced no pressor response but, instead, a slight decrease in blood pressure ( - 5.00 & 3.55 mmHg, non-significant) and heart rate ( -8.50 + 4.68 bpm, non-significant) (Fig. 3). In addition, bicuculline microinjections were also found to prevent (-90%) the inhibitory influence of the intra-NTS application of 5-HT (10nmol) on the baroreceptor reflex response to i.v. phenylephrine.

Experiments in gunglionectomized rats Cardiotwscular responses to serotonin.

No

significant differences were found in the mean blood pressure and heart rate between unilaterally ganglionectomized and control intact animals. Furthermore, the baroreceptor reflex response to i.v. phenylephrine was as pronounced in gangiionectomized rats (1.56 _t 0.32 b.p.m./mmHg, n = 7) as in controls (1.75 f 0.43 b.p.m./mmHg, n = 6). However, the rise in blood pressure due to the microinjection of 5-HT (10 nmol, n = 6) or phenylbiguanide (10 nmol, n = 6) into the NTS on the lesioned side in ganglionectomized rats was considerably less than that observed in intact animals (Table 3). In contrast, microinjections of 5-HT or phenylbiguanide into the NTS on the intact side in ganglionectomized animals elicited pressor responses which were as pronounced as those found in control animals (Table 3). Further experiments with the 5-HT, agonist DOB (0.5 pmol, n = 6) showed that its hypotensive effects were similar whether it was microinjected into the NTS on the ipsi- or contralateral side in ganglionectomized rats or on either side in intact animals (Table 3). In contrast to that found in intact rats (Fig. 2), the microinjection of 5-HT (10 nmol, n = 6) into the NTS on the lesioned side was unable to reduce significantly the baroreceptor reflex response to i.v. phenylephrine in ganglionectomiz~ rats (Table 4). However a significant effect could be elicited from the contralateral intact side. Indeed, the inhibition of phenylephrineevoked baroreceptor reflex was almost as pronounced after the injection of IO nmol S-HT into the intact NTS in lesioned

rats as after the bilateral

injection

of

NTS: role of serotonin, receptors in ~rdiova~ular

regulation

95

Table 1. Changes in mean blood pressure and baroceptor reflex inhibition induced by the bilateral microinjection of various doses of phenylbiguanide into the nucleus tractus solitarius 1.7 rim01

Phenylbiguanide AMBP (mmHg) BaroreAex inhibition (%)

2.5 nmol

5 nmol

10 nmol

0

16.0 * 3.0

26.0 & 6.6

44.5 f 4.3

16.0 + 4.6

44.0 + 4.5

85.3 * 8.3

90.7 f 8.8

Values represent the absolute increase in mean blood pressure (AMBP) and the per cent inhibition of the baroreflex responses (see Experimental Procedures) after the bilateral microinjection of 1.7-10 nmol of phenylbiguanide into the NTS. The baroreceptor reflex was tested 10 min after phenylbiguanide microinj~tions. Data are the mean f S.E.M of six independent dete~inatjons (in six rats) for each dose. Basal blood pressure range was 85-95 mmHg. The increase in mean blood pressure, at 2.5, 5 and 10 nmol of phenylbiguanide, and the inhibition of the barocep tor reflex by this agonist, at all doses tested, were significant at P < 0.01.

Table 2. Effects of various serotonin receptor antagonists on the blood pressure responses to unilateral microinj~tion of serotonin or phenylbigua~de into the nucleus tractus solitarius AMBP (mmHg) Phenylbiguanide 5-HT (10 nmol) (10 nmol)

Antagonists None (vehicle) Methysergide Ketanserin Ondansetron Zacopride

25.9 f 21.7 + 20.9 k 2.1 * 1.0 +

20.9 f 2.0 16.4 + 1.5 14.7 & 2.1 2.7 + 1.4+ 3.0 + 1.6*

2.7 1.4 1.6 0.9* 0.3*

Each antagonist was microinjected at the dose of 100 pmol exactly in the same site as 5-HT or phenylbiguanide, but 5min earlier. The increase in blood pressure (AMBP) due to each treatment is expressed in mmHg. The basal blood pressure ranged between 85 and 95 mmHg. Each value is the mean & S.E.M of at least six independent determinations (one determination per rat). *P < 0.01 when compared to mean blood pressure change in rats treated with the vehicle plus 5-HT or phenylbiguanide.

(A)

(6)

the same dose of the indoleamine in control animals {Table 4). Autoradiography. The binding sites for [12’Iliodozacopride in the coronal sections of the brainstem are shown in Fig. 4. Highest binding was found in the rostra1 and commissural (corresponding to the obex level, subjacent to the area postrema) regions of the NTS. In the most caudal region of the NTS (approx. 0.75 mm below the obex level), [‘251]iodo-zacopride binding was approx. 35% lower than in the former two regions (Table 5). No significant differences were detected between the right and left sides in intact animals. After unilateral ganglionectomy, [lz5I]iodozacopride binding at the obex and more rostra1 levels of the NTS in the contralateral side was slightly less (approx. 20%) than that found in intact animals (Table 5). In contrast, a marked reduction in [‘251]iodo-zacopride binding was observed on the ipsilateral side, at both the rostra1 (- 82.7%) and commissural (- 54.8%) levels (Fig. 4A, B). The cauda1 zone of the NTS could not be assessed in the same manner due to the lack of discrimination between the left and right sides. Within the caudal NTS, taken as a whole, the specific binding of [‘*‘I]iodo-zacopride in ganglionectomized rats was approx. 90% of that found in intact animals (Table 5). DISCUSSION

+ bicuculline

flmin 5-HT

Fig. 3. Bicuculline-induced inhibition of the pressor responses evoked by the microinjeetion of 5-HT into the NTS. (A) Changes in blood pressure and heart rate after bilateral bicuculline (lOOpmo1) mi~oinj~tion into the NTS. (B) Antagonism of the 5-HT (10 nmol)-evoked pressor response (see Fig. 1A) by bicuculline pretreatment. Bicuculline was injected bilaterally in the same sites as 5-HT, but 10 min

before. Arrows indicate the times for the microinjections. Abbreviations are as in the legend to Fig. 1.

The present data confirmed that the microinjection of nanomoiar doses of 5-HT into the NTS produced a dose-de~ndent increase in blood pressure in rats.“3 This effect could be completely abolished by the prior local administration of 5-HT, receptor antagonists such as zacopride and ondansetron. In contrast, both 5-HT, and 5-HT, receptor antagonists (methysergide and ketanserin, respectively) failed to antagonize the pressor response to 5-HT. In addition, the microinjection of phenylbiguanide or 2-methyl-5-HT, two .5-HT, receptor agonists, into the NTS, also evoked a pressor response which could be prevented by S-HT, antagonists. These results provide pharmacological evidence that the aforementioned serotonergic

N. MERAH1 et al.

96

INTACT +0.75

GANGLIONECTOMIZED (Right side)

A)

B)

c)

-0.75 mm Fig. 4. Autoradiographic localization o f [ t-~5I]iodo-zacopride binding to brain sections at the level of the NTS. Effects of unilateral ganglionectomy. Ablation of the right nodose ganglion was performed 15 days before death. Coronal brain sections (20 #m thick) were cut at the level of (A) the rostral NTS, anterior to the area postrema (AP), (B) the mid-area postrema, near the obex zone, and (C) the caudal nucleus tractus solitarius. Sections from intact and ganghonectomized rats were labeled by 0.15 nM [125I]iodozacopride, and autoradiograms were developed after three days in the dark (see Experimental Procedures). The ordinate axis indicates the antenority level with 0 corresponding to the obex. Table 3. Blood pressure responses to the microinjection of serotonin, phenylbiguanide or 4-bromo-2,5-dimethoxyphenylisoprophylamine into the nucleus tractus solitarius of intact or ganglionectomized rats Drugs Microinjection 5-HT Phenylbiguanide DOB

Intact Unilateral +24.3 + 1.3 + 15.8 + 1.2 -35.0+1.7

AMBP (mmHg) Ganglionectomized Contralateral Ipsilateral +22.8 _+ 3.6 + 15.0 _+4.2 --26.2__+14.1

+7.2 + 4.0* +4.6 + 3.0* -39.5+5.5

Ablation of the right nodose ganglion was performed 15 days before the experiments. 5-HT (10nmol), phenylbiguanide (10nmol) or DOB (0.5 pmol) were microinjected into the NTS on either side in intact rats, and on the contralateral (left) and operated sides in ganglionectomized rats. The change in blood pressure (AMBP) due to the unilateral microinjection of each drug is expressed as the mean + S.E.M. of six independent determinations (in six rats). *P < 0.001 when compared to mean blood pressure change after the unilateral microinjection of the same drug into the NTS in intact rats or on the contralateral unoperated side in ganglionectomized rats.

NTS: role of serotonin, receptors in cardiovascular regulation

91

Table 4. Effects of unilateral ablation of the nodose ganglion on the inhibition of the baroreflex responses by serotonin microinjected into the nucleus tractus solitarius AHR/AMBP (b.p.m./mmHg) Ganglionectomized 5-HT Control Contralateral Ipsilateral

Intact 5-HT Control

Unilateral

Bilateral

1.75 + 0.35

1.12*0.24*

0.20 + 0.20**

1.59 + 0.30

0.55 + 0.16** 1.26 + 0.20

Experiments were performed 15 days after ablation of the right nodose ganglion. Baroreflex responses were elicited by an i.v. administration of phenylephrine (5 pg/kg) prior to (“control”) and 10 min after the unilateral or bilateral microinjection of 10 nmol 5-HT into the NTS. Baroreflex sensitivity was determined as the ratio of the maximal change in heart rate over that in the mean blood pressure (AHRIAMBP) after phenylephrine treatment. Each value is the mean + S.E.M of six independent determinations (in six rats). *P < 0.05, **P < 0.001, when compared to respective control values.

Table 5. Effects of unilateral ablation of the nodose ganglion on the specific binding of [‘*‘I]iodo-zacopride to various zones of the nucleus tractus solitarius NTS region Rostra1 Commissural Caudal

[1Z51]Iodo-zacopride (fmol/mg tissue) Unilateral ganglionectomy Intact (control) 3.90 + 0.33 (n = 12) 3.67 k 0.33 (n = 12) 2.47 k 0.18 (n = 10)

Ipsilateral Contralateral 0.36 k 0.16* (n = 12) 2.92 + 0.34 (n = 12) 1.19 * 0.09* (n = 15) 2.99 & 0.16 (n = 14) 2.05 + 0.27 (n = 10)

Rats were killed 15 days after ganglioctomy on the right side, and coronal brain sections at the level of the NTS (see Fig. 4) were labeled with 0.15 nM [‘251]iodo-zacopride as described in Experimental Procedures. Optical densities on the corresponding autoradiograms were converted to fmol [‘*‘I]iodo-zacopride specifically bound per mg tissue using radioiodinated standards (Amersham). Each value is the mean + S.E.M. of n determinations. ‘P < 0.001 when compared to [‘251]iodo-zacopride specific binding in the same NTS region on the contralateral side in ganglionectomized rats, or on both sides in intact animals.

responses were mediated by 5-HT, receptors. In a previous study, however, we have shown that the microinjection of picomolar doses of S-HT into the NTS elicited an immediate hypotensive response that could be specifically antagonized by 5-HT, receptor blocking agents. 31Moreover, the local microinjection of picomolar doses of 5-HT, agonists (DOB and DOI) produced a dose dependent hypotension and bradycardia (N.M., unpublished observations). Taken together, these results suggest that the cardiovascular responses to 5-HT were mediated through the activation of a heterogeneous population of 5-HT receptors within the NTS. Interestingly, the 5-HT-induced increase in blood pressure was not associated with heart rate changes and could be abolished by prior i.v. administration of prazosin. Therefore, it can be proposed that the stimulation of 5-HT, receptors within the NTS involved the sympathetic nervous system rather than vagal mechanisms. However, further investigations are necessary to confirm this hypothesis, as a central effect of prazosin on vagal mechanisms cannot be totally excluded. As compared to the 5-HT, mediated responses, almost lOOO-foldgreater doses of 5-HT were required to elicit 5-HT, mediated responses. Yet, until recently, both 5-HT, and 5-HT, receptors were con-

NSC ‘W--E

sidered as low-affinity receptors for 5-HT (see Ref. 27). However, most recent studies with [3H]DOB as radioligand have clearly shown that the functional 5-HT, receptors, i.e. those coupled to G proteins in brain membranes, have a nanomolar affinity for 5_HT.‘4.‘6This may possibly explain why much lower doses of 5-HT were required to stimulate 5-HT, than 5-HT, receptors in our experimental conditions. It is now clearly established that the NTS is the relay nucleus and the integration site of the arterial baroreflex afferents.23*28Activation of the peripheral barosensitive afferent fibers by increasing arterial blood pressure evokes a reflex hypotension and bradycardia. Although the nature of the main neurotransmitter in these afferent nerve endings is still a matter of controversy, L-glutamate seems to be the best candidate to date. 21 Taking into consideration the current paradigm on the cardiovascular reflex control, it can be expected that the local application of substances that facilitate neuronal transmission by these baroreceptive fibers would produce hypotensive and bradycardiac responses. Alternatively, hypertension and/or tachycardiac responses should result from the blockade of neuronal transmission through these fibers. Except that of the inhibitory neurotransmitter GABA,” it is uncommon that the local administration of an agonist into the NTS produces

98

N. MERAHI et al.

“inhibitory-like” responses, as did 5-HT, receptor despite the high abundance of 5-HT, receptors on the agonists. Additionally, the latter drugs also exerted a presynaptic sensory fibers.’ dose-dependent inhibition of the baroreceptor reflex However, according to the present data, unilateral responses to i.v. administration of phenylephrine. 5-HT microinjections into the NTS ipsilateral to the Both 5-HT mediated hypertension and baroreflex lesion failed to alter the baroreflex responses in antagonism exhibited similar dose-response curves. ganglionectomized animals, In addition, injections These results strongly suggest that the hypertensive into the contralateral NTS showed an enhanced responses to nanomolar doses of 5-HT (and 5-HT, baroreflex blocking effect of 5-HT compared to that receptor agonists) were due to the inhibition of the evoked by the unilateral administration of the indobaroreceptor reflex mechanisms. leamine in intact animals. Such a lateralization of the Inhibition of the NTS second order baroreceptor 5-HT,-mediated responses after ganglionectomy neurons, or of presynaptic interneurons, by the actisuggests a presynaptic site of action for the 5-HT3 vation of 5-HT, receptors might explain the baroligands. The whole of these considerations seems to receptor blocking action of the 5-HT, agonists. indicate that the activation of 5-HT, receptors However, 5-HT3 receptors are known to transmit an located on vagal afferent fibers produced. via the excitatory influence of 5-HT on target cells,2y~3n~34~35 liberation of a putative neurotransmitter contained in and a direct inhibitory effect of 5-HT, receptor these fibers, the activation of a GABAergic inhibitory stimulation is unlikely (but cannot be totally excluded system. In contrast, the responses to 5-HT, agonists at present). Accordingly, a probable explanation for remained unaffected in both sides. This may infer a the inhibition of the baroreceptor reflex by 5-HT, postsynaptic location (with respect to the vagal afferagonists might be the activation of an inhibitory ent fibers) of the 5-HT2 receptors mediating the system by these drugs. GABAergic neurons were hypotensive responses. previously demonstrated to be inhibitory in the Finally, it is important to note that the depolarizNTS.“‘,17,?0 Thus, the local microinjection of a ation of the peripheral sensory nerve terminals due to GABA, receptor agonist, i.e. muscimol or GABA the activation of the 5-HT, receptors located on the itself produces hypertension3 and, conversely, the vagus nerve probably exerts an excitatory influence intra-NTS administration of GABA receptor antagon the afferent fibers. This conclusion derived from onists facilitates the responses to peripherally evoked the fact that the stimulation of 5-HT, receptors on aortic nerve stimulation.‘2 In the present study, local vagal afferent fibers elicited hypotension and bradymicroinjections of bicuculline into the NTS totally cardia as cardiovascular reflex responses.” In antagonized the pressor responses to 5-HT and precontrast, the present data suggest that the stimulation vented the baroreflex inhibition induced by S-HT, of 5-HT, receptors located on the central terminals of receptor activation. These data suggest that the vagal afferent fibers was responsible for the pressor 5-HT, effects could be mediated through the actieffects of 5-HT, agonists injected into the NTS. One vation of a GABAergic system. possible explanation of this discrepancy is that the Microinjections in intact animals did not allow us stimulation of 5-HT, receptors located on these terminals might produce a local depolarization and to assess whether the activated receptors were situconsequently decrease neuronal transmission through ated on pre- (vagal afferent fibers) or postsynaptic (inhibitory neurons) elements. The present autoradioa presynaptic inhibitory mechanism. However, another possible explanation is that different fibers graphic data confirmed that a large majority of the were involved in the cardiovascular effects induced by 5-HT, receptors was situated on the presynaptic vagal the stimulation of peripheral” versus central (in the afferent fibersZS which degenerated in the NTS ipsilateral to the nodose ganglionectomy. Indeed an NTS) 5-HT, receptors. Recently, in the ferret, it has been demonstrated 55% reduction in [‘251]iodo-zacopride approx. that bilateral vagotomy performed at the splanchnic specific binding was found at the obex level in level abolishes completely 5-HT,-ligand binding in lesioned rats, i.e. where various drugs were microinthe NTS, indicating an exclusively presynaptic lojetted. This region has been previously shown to be cation of the 5-HT, receptors on vagal afferents of the main termination site of the peripheral barorecepgastrointestinal origin.” Therefore, a functional tor fibers.6 interaction between the gastrointestinal and cardioEven though ganglionectomy has significantly vascular fibers may be possible in the rat. Further reduced the 5-HT, receptor density within the NTS, studies are necessary to investigate this possibility the possibility of a remaining postsynaptic S-HT3 and to determine the origin(s) of endogenous seroreceptor population operating directly or indirectly tonin that may activate these presynaptic 5-HT, through GABAergic neurons cannot be totally exreceptors. cluded. Recent studies on the spinal transmission of nociceptive messages Seem to provide evidence that in the dorsal horn, antinociceptive effects of 5-HT, CONCLUSION receptor agonists are predominantly mediated by This study showed, in the rat, that the stimulation receptors located postsynaptically (on GABAergic of the 5-HT, receptors presumably located on the neurons) (G. L. Wilcox, personal communication)

NTS: role of serotonin, receptors in cardiovascular regulation sensory afferents fibers within the NTS elicited an increase in blood pressure and a temporary inhibition of the baroreflex arch. Both effects could be mediated through a GABAergic system.

99

Acknowledgemenrs-This study has been possible with the financial support of INSERM (Institut National de la Sante et de la Recherche Medicale) and DRET (Direction des Recherches, Essais et Techniques, Contrat nb 90/085). We are grateful to Dr Cassels for his kind supply of DOB.

REFERENCES 1. Andrews P. L. R., Rapeport W. G. and Sanger G. J. (1988) Neuropharmacology of emesis induced by anti-cancer therapy. Trenak pharmac. Sci. 9, 334-341. 2. BhaskaranD. and Freed C. R. (1988) Changes in arterial blood pressure lead to baroreceptor-mediated changes in norepinephrine and 5-hydroxyindoleacetic acid in rat nucleus tractus solitarius. J. Pharmac. exp. Ther. 245,356363. 3. Bousquet P., Feldman J., Bloch R. and Schwartz J. (1982) Evidence for a neuromodulatory role of GABA at the first synapse of the baroreceptor reflex pathway. Effects of GABA derivatives injected into the NTS. Naunyn-Schmiedeberg’s Arch. Pharmac. 319, 168-171. 4. Carter D. A. and Lightman S. L. (1985) Cardio-respiratory actions of substance-P, TRH and 5-HT in the nucleus tractus solitarius of rats: evidence for functional interactions of neuropeptides and amine neurotransmitters. Neuropeptides 6, 425436.

5. Chemerinski E., Ramirez A. J. and Enero M. A. (1980) Sinoaortic denervation-induced changes in central serotonergic neurons. Eur. J. Pharmac. 64, 195-198. 6. Ciriello J. (1983) Brain stem projections of aortic baroreceptor fibres in the rat. Neurosci. Lelt. 36, 3742. I. Fuxe K. (1965) Evidence for the existence of monoamine neurons in the central nervous system. Acta physiol. stand. 64 (Suppl. 247), l-84. 8. Guyenet P. G., Filtz T. M. and Donaldson S. R. (1987) Role of excitatory amino acids in rat vagal and sympathetic baroreflexes. Brain Res. 407, 272-284. 9. Hamon M., Gallissot M. C., Menard F., Gozlan H., Bourgoin S. and Verge D. (1989) 5-HT, receptor binding sites are on capsaicin-sensitive fibres in the rat spinal cord. Eur. J. Pharmac. 164, 315-322. 10. Hwang B. H. and Wu J. Y. (1984) Ultrastructural studies on catecholaminergic terminals and GABAergic neurons in nucleus tractus solitarius of the rat medulla oblongata. Brain Res. 302, 57-67. 11. Koscielniak T., Penchant M., Laporte A. M., Guminski V., Verge D., Hamon M. and Gozlan H. (1990) [1251]Iodo-zacopride: un nouveau ligand pour l’etude par autoradiographe des recepteurs 5-HT, centraux. C. R. Acad. Sci. 311, 231-237.

12. Kubo T. and Kihara M. (1988) Evidence for gamma-aminobutyric acid receptor-mediated modulation of the aortic baroreceptor reflex in the nucleus tractus solitarii of the rat. Neurosci. Left. 89, 156160. 13. Laguzzi R., Reis D. J. and Talman W. T. (1984) Modulation of cardiovascular and electrocortical activity through serotonergic mechanisms in the nucleus tractus solitarius of the rat. Brain Res. 304, 321-328. 14. Leonhardt S. and Titeler M. (1989) Serotonin 5-HT, receptors: two states versus two subtypes. J. Neurochem. 53, 316318.

15 Leslie R. A., Reynolds D. J. M., Andrews P. L. R., Grahame-Smith D. G., Davis C. J. and Harvey J. M. (1990) Evidence for presynaptic 5-HT, recognition sites on vagal afferent terminals in the brainstem of the ferret. Neuroscience 38, 667673. 16 Lyon R. A., Davis K. H. and Titeler M. (1987) [‘H]DGB labels a guanyl nucleotide-sensitive state of the 5-HT, receptor. Molec. Pharmac. 31, 196199.

17 Maley B. and Newton B. W. (1985) Immunohistochemistry

of y-aminobutyric acid in the cat nucleus tractus solitarius.

Brain Res. 330, 36k368.

18 Manaker S. and Verderame H. M. (1990) Organisation of serotonin 1A and 1B receptors in the nucleus of the solitary tract. J. camp. Neurol. 301, 535-553. 19 McQueen D. (1990) Cardiovascular reflexes and 5-hydroxytryptamine. In Cardiovascular Pharmacology of 5-Hydroxy tryptamine (eds Saxena P. R., Wallis D. I., Wouters W. and Bevan P.), pp. 233-245. Kluwer Academic, Dordrecht. 20. Meeley M. P., Ruggiero D. A., Ishitsuka T. and Reis D. J. (1985) Intrinsic y-aminobutyric acid neurons in the nucleus of the solitary tract and the rostra1 ventrolateral medulla of the rat: an immunocytochemical and biochemical study. Neurosci. Lett. 58, 83-89.

21. Meeley M. P., Underwood M. D., Talman W. T. and Reis D. J. (1989) Content and in oitro release of endogenous amino acids in the area of the nucleus of the solitary tract of the rat. J. Neurochem. 53, 1807-1817. 22. Nosjean A., Compoint C., Buisseret-Delmas C., Orer H. S., Merahi N., Puizillout J. J. and Laguzzi R. (1990) Serotonergic projections from the nodose ganglia to the nucleus tractus solitarius. An immunohistochemical and double labeling study in the rat. Neurosci. Lett. 114, 22-26. 23. Palkovits M. and Zaborsky L. (1977) Neuroanatomy of central cardiovascular control. Nucleus tractus solitarii: afferent and efferent neuronal connections in relation to the baroreceptor reflex arch. Prog. Brain Res. 47, 934. 24. Pazos A., Cartes R. and Palacios J. M. (1985) Quantitative autoradiographic mapping of serotonin receptors in the rat brain. II. Serotonin-2 receptors. Brain Res. 346, 231-249. 25. Pratt G. D. and Bowery N. G. (1989) The 5-HT, receptor ligand, [3H]BRL 43694, binds to presynaptic sites in the nucleus tractus solitarius of the rat. Neuropharmacology 28, 1367-1376. 26. Pratt G. D., Bowery N. G., Kilpatrick G. J., Leslie R. A., Barnes N. M., Naylor R. J., Jones B. J., Nelson D. R., Palacios J. M., Slater P. and Reynolds D. J. M. (1990) Consensus meeting agrees distribution of 5-HT, receptors in mammalian hindbrain. Trenak pharmac. Sci. 11, 135-137. 21. Rack F., Laporte A. M., Daval G., Verge D., Gozlan H. and Hamon M. (1991) Autoradiography of serotonin receptor subtypes in the central nervous system. Neurochem. Inr. 18, l-15. 28. Reis D. J., Granata A. P., Perrone M. H. and Talman W. T. (1981) Evidence that glutamic acid is the neurotransmitter of baroreceptor afferents terminating in the nucleus tractus solitarius (NTS). J. auton. nero. Sysr. 3, 321-334. 29. Richardson B. P. and Buchkeit K. H. (1988) The pharmacology, distribution and function of 5-HT, receptors. In Neuronal Seroronin (eds Osborne N. N. and Hamon M.), pp. 465-506. John Wiley, Chichester.

100

N. MEKAHI PI ~1

30. Robertson B. and Bevan S. (1991) Properties of 5hydrotryptamine, receptor-gated currents in adult rat dorsal root ganglion neurones. Br. J. Pharmac. 102, 212-276. 3 I. Shvaloff A. and Laguzzi R. (1986) Serotonin receptors in the rat nucleus tractus solitarii and cardiovascular regulation Eur. J. Pharmac. 132, 283-288. 32. Steinbusch H. W. M. (1984) Serotonin-immunoreactive neurons and their projections in the CNS. In Handbook o/ Chemical Anatomy (eds Bjiirklund A., Hiikfelt T. and Kuhar M. J.), Vol. 3, pp. 688125. Elsevier, Amsterdam. 33. Wolf W. A., Kuhn D. M. and Lovenberg W. (1981) Blood pressure responses to local application of serotonergic agents in the nucleus tractus solitarii. Eur. J. Pharmac. 69, 291-299. 34. Yakel J. L., Shao X. M. and Jackson M. B. (1990) The selectivity of the channel coupled to the 5-HT, receptor. Brain Res. 533, 46-52. 35. Yakel J. L., Trussell L. 0. and Jackson M. B. (1988) Three serotonin responses in cultured mouse hippocampal and striatal neurons. J. Neurosci. 8, 1273 1285. (Accepted 10 June 1991)