IntroductionAs well as several other effects [1–3] nitric oxide (NO),the molecule generated from L-arginine (L-Arg) by theaction of nitric oxide synthase (NOS), has been impli-cated in the regulation of local and systemic vascular resis-tance, in the distribution of blood flow and oxygendelivery, in sodium balance and, therefore, in arterialpressure control [4]. NO has also been postulated as aneuronal modulator, acting in the brain as a signalingmolecule that disregards spatial constraints, opening a newdimension in the classic concept of neural communica-tion [5]. The brain contains the highest activity of NOSof any tissue so far examined and the enzyme has a wide-

spread distribution [6,7]: it is present in almost all sitesinvolved in the integration of arterial baroreflex, such thenucleus tractus solitarii (NTS), the dorsal motor nucleusof the vagus, nucleus ambiguus and ventrolateral medulla.NO formation is stimulated by the excitatory amino-acidL-glutamate (the main neurotransmitter involved in thecentral baroreflex pathways) administered into the NTS[8] as well as NO increases excitatory amino-acid releasein the NTS [9] and NOS inhibition blocks the effect ofseveral excitatory amino-acid receptor agonists [10]. It hasbeen shown that inhibition of NO formation in the NTSincreases significantly sympathetic tone and pressure inrabbits [11]. These findings suggest that NO is acting in

111123456789101112345678920111234567893011123456789401112345678950111234567111

Original article 1993

0263-6352 © 1998 Lippincott Williams & Wilkins

Role of endogenous nitric oxide in the nucleus tratus solitarii on baroreflex control of heart rate in spontaneously hypertensive ratsVera Pontieria, Maria Kelly Venezuelaa, Cristóforo Scavoneb and Lisete C. Michelinia

Objective Toinvestigate the modulatory effect ofendogenous nitric oxide (NO) in the nucleus tractussolitarii (NTS) on the baroreceptor reflex control of heartrate in conscious spontaneously hypertensive (SHR) andnormotensive (WKY) rats.

Design and methods Male age- and weight-matchedSHR and WKY chronically instrumented with cannulas inthe NTS, artery and vein were used. Basal pressure (AP),heart rate (HR) and reflex HR responses duringloading/unloading of baroreceptors (phenylephrine/sodium nitroprusside, iv) were recorded during vehicle(3 nl/min) NG-monomethyl-L-arginine (L-NMMA) and L-arginine (L-Arg) infusions into the NTS. Constitutive NOsynthase (NOS) activity was inferred by 3H-citrullineformation in the dorsal brain stem of other SHR and WKYgroups.

Results In SHR a small dose of L-NMMA (30 ng/kg/min)restricted to the NTS did not change AP and HR(185 ± 4 mmHg, 373 ± 12 beats/min, respectively), butdecreased the HR range (57 ± 7 beats/min, a 34%reduction, P < 0.05) without changing further the impairedgain of baroreceptor reflex control of HR. In the WKYgroup similar results (significant 32% reduction in HRrange, gain unchanged) were only attained with a dose10 times higher (L-NMMANTS = 300 ng/kg/min), no effectbeing observed with the small dose (HRrange = 163 ± 12 beats/min). In SHR, L-ArgNTS

(900 ng/kg/min) did not improve baroreflex control ofHR, but restored the depression of HR range when givenafter L-NMMANTS. Basal NOS activity in the dorsal brainstem was reduced in SHR (P < 0.05) when compared toWKY group.

Conclusions NO modulates, at the NTS level, thebaroreceptor reflex control of HR in both SHR and WKYnot by altering the gain, but by increasing HR rangeduring afferent stimulation. In SHR the depressed NOmodulation is in accordance with the smaller NOS activityin the dorsal brain stem. J Hypertens 1998, 16:1993–1999© 1998 Lippincott Williams & Wilkins.

Journal of Hypertension 1998, 16:1993–1999

Keywords: nitric oxide, NG- monomethyl-L-arginine (L-NMMA), L-arginine,nucleus tractus solitarii, baroreflex, conscious, SHR, WKY

aDepartment of Physiology and Biophysics and bDepartment of Pharmacology,ICB, University of São Paulo, SP, Brazil.

Sponsorship: Supported by Fundação de Amparo à Pesquisa do Estado deSão Paulo (FAPESP, 96/0043-4 and 94/0463-8). CS and LCM are researchfellows from CNPq.

Correspondence and requests for reprints to Dr Lisete C. Michelini,Department of Physiology and Biophysics, Institute of Biomedical Sciences,USP, Av. Prof. Lineu Prestes 1524, 05508-900 São Paulo, SP, Brazil. Tel: +55 11 8187213; Fax: +55 11 8187285; E-mail: lisete@fisio.icb1.usp.br

central areas involved in the regulation of cardiovascularcontrol, namely the NTS.

Previous works addressing the study of NO modulatoryeffect on baroreceptor reflex control of heart rate andsympathetic nerve activity showed discrepant results innormotensive animals: no change in heart rate × pressureand/or sympathetic nerve activity × pressure relationships[11,12], increased gain with enhancement of the brady-cardic component [13,14], or, decreased sensitivity tonitroprusside-induced changes in pressure [15]. In spon-taneous hypertension NOS inhibitor increased bothpressure and gain of sympathetic nerve activity × pressurerelationship [16]. It should be noted that chronic inhibi-tion of NO synthesis caused sustained hypertension[15,17].

Therefore, the objectives of this study were to determineand compare the modulatory effect of endogenous NOblockade into the NTS on baroreceptor reflex control of heart rate in normotensive (WKY) and spontan-eously hypertensive rats (SHR). Since changes in baro-receptor reflex after blockade of NO formation may be related to the accompanying changes in blood pressurewe employed a dose of NG-monomethyl-L-arginine (L-NMMA, an inhibitor of NOS) that did not change base-line levels of pressure when infused into the NTS ofconscious freely moving rats.

MethodsAnimals/surgical preparationsMale WKY and SHR rats which have been inbred in theUniversity of São Paulo since 1991 were used. The ratswere aged 3–4 months and weighed 250–330 g. They werekept in 12 : 12 light/dark cycle with standard laboratorychow and water provided ad libitum. All surgical proce-dures and protocols used are in accordance with Guide-lines for Ethical Care of Experimental Animals and wereapproved by the Institutional Animals Care and UseCommittee.

The rats were submitted to chronic cannulation of thebrain stem, according to a technique developed by us [18].Briefly the rats were anesthetized with Nembutal(40 mg/kg, ip) and placed in a stereotaxic apparatus (Kopf,Tujunga, CA, USA) with their head in a horizontal posi-tion. The skull was widely exposed, one screw was fixedat the parietal bone, and a small window was openedcaudally to the lambda to allow the introduction of a stain-less steel guide cannula (17 mm length, 0.6 mm outsidediameter) at an angle of 24°. The stereotaxic coordinateswere as follows: 1 mm caudal to interaural line, 0.4 mmlateral (right or left) to the midline and 8.92 mm belowthe skull surface, in such a way that the tip of the chroniccannula did not touch the NTS but lay in the ventralcerebellum, avoiding any cicatricial retraction of thetissue, the function of which would be studied. The

cannula plus screw were fixed on the skull with fast poly-merizing methacrylate, and the cannula was closed by anoccluder. The rats were given 60 000 units of penicillin(Pentabiotico Veterinario, Fontoura Wyeth, São Paulo,Brazil) and kept in individual Plexiglas cages. After arecovery period of 3–4 days, catheters (PVC tube,Critchley, Silverwater, NSW, Australia) were inserted intoa femoral artery and vein (for cardiovascular monitoringand drugs administration) under ether anesthesia. Thecatheters were tunneled subcutaneously to the back ofthe neck where they were fixed with sutures. The ratsrecovery for another 1-day period.

Experimental session/experimental sequencesBoth arterial pressure (AP = systolic, diastolic and mean)and heart rate (HR, Biotach, triggered by AP pulses) wererecorded continuously (P23XL transducer, connected toRS3400 recorder, Gould, Cleveland, OH, USA) in freelymoving rats. At the beginning of the experimental sessionthe occluder was removed from the brain stem guidecannula and 20–30 min were allowed for stabilization ofcardiovascular parameters. After recording the basal valuesof AP and HR, a microinjection needle (33 gauge, 18 mmlength) was introduced into the guide cannula and NTS infusion with vehicle (VEH = saline 3 nl/min,Syringe Pump, Orion, Boston, MA, USA) started. Controlmeasurement of AP and HR were followed by repeatedstimulation of baroreceptors with 0.1 ml bolus injectionsof graded doses of phenylephrine (0.1 up to 6.4 µg/kg)and sodium nitroprusside (0.2–12.8 µg/kg) given into thefemoral vein. Phenylephrine and nitroprusside injectionswere given in a random order and subsequent injectionswere not made until the recorded parameters had returnedto pre-injection levels. VEH infusion (~30–40 min) wasdiscontinued and replaced, after 20–30 min interval, byL-NMMA infusion (30 ng/kg/min, 3 nl/min). Basal valuesof AP and HR were measured and control and peak MAPand HR changes were determined again by loading/unloading of baroreceptors with phenylephrine/nitroprus-side, respectively.

To check the effects of NO availability in the SHR group,L-arginine (L-Arg = 900 ng/kg/min) was also infused into the NTS. Two experimental sequences were usedVEH → L-NMMA → L-ARG and VEH → L-ARG →L-NMMA with measurement of basal AP and HR andassessment of baroreflex function during all infusions. Thedose of L-NMMA (30 ng/kg/min) used in SHR and WKYgroups was selected in preliminary experiments (n = 4) asthe largest dose that did not change basal levels of APand HR when infused into the NTS of SHR. In threeout of 10 WKY it was also administered into the NTS adose of L-NMMA 10 times higher (300 ng/kg/min).

After completion of experimental sequence the rat wasanesthetized (ether) and Evans blue was infused into theNTS. It was followed by transcardiac perfusion of saline

111123456789101112345678920111234567893011123456789401112345678950111234567111

1994 Journal of Hypertension 1998, Vol 16 No 12 (part 2)

(30–40 ml) and 10% buffered formalin (50–80 ml). Thebrain was removed and stored in 10% formalin plus 10%sucrose. The exact location of injection site and its exten-sion were assessed a posteriori by histological examinationof 40 µm serial coronal sections of brain stem obtainedwith a cryostat (Sartorius Werke, Göttingen, Germany).

Nitric oxide synthase (NOS) assay In other SHR and WKY groups, NOS activity wasmeasured in high-speed supernatants from dorsal brainstem, as described by McKee et al. [19] with some modi-fications. The rat was decapitated, the brain immediatelyremoved from the skull and frozen in dry ice. The obexwas visualized and two slices (caudal and rostral to theobex, 1 mm thick), were carefully obtained. Threepunches (one from caudal and two from rostral slice, corre-sponding to the medial portion of dorsal brain stem) wereremoved and homogenized in ice-cold 0.32 Msucrose/20 mM HEPES buffer (pH 7.4) containing 1 mMdithiothreitol in an ice bath for 1 min, using a Teflonhomogenizer. Each homogenate was centrifuged at10 000g for 30 min at 4°C. Supernatants were removedand protein concentration were determined in triplicateby the assay method of Bradford [20]. The NOS assaywas performed by incubation (37°C for 60 min) of 100 µgof protein (20 µl) to a final volume of 200 µl of assaymixture containing 100 µM L-Arg/L-[3H]Arg (5 µCi/ml)/20 nM HEPES buffer (pH 7.4)/0.425 mM EDTA/0.45 mM CaCl2/1 mM NADPH/80 units of calmodulin/0.1 µM TET/4 µM FAD, 4 µM FMN. For Ca2+/calmod-ulin-free activity, EDTA was replaced by EGTA andCaCl2, and calmodulin was omitted. The reaction wasstopped with ice-cold stop buffer (20 mM HEPES, 2 mML-Arg, 5 mM EDTA, pH 5.5). Each reaction mixture (intriplicate) was added to 2 ml of 20 mM HEPES pH 5.5,applied to a 2.5 ml cation-exchange resin column (Dowex,Na+ form) pre-equilibrated with 20 mM HEPES 5.5, andthen eluted with 2 ml of distilled water. The aliquots werecollected in vials and scintillation liquid (8 ml) was addedand the radioactivity quantified. Samples of rat cerebellumwere analyzed simultaneously as a positive control. TheCa2+-independent (iNOS) form of NOS represents <5%of the total NOS activity in the dorsal brain stem.

Data analysis and statisticsResults are expressed as mean ± SEM. Baroreceptor reflex control of HR was estimated by two different ways:(1) the traditional linear regression equation (DHR =aDMAP + b) correlating changes in HR with changes inMAP during loading (phenylephrine) and unloading(nitroprusside) of baroreceptors; (2) the sigmoidal logisticequation [21,22] correlating HR and MAP values duringinjections of both phenylephrine and sodium nitroprus-side:

where P1 = lower HR plateau, P2 = HR range, P3 = acurvature coefficient which is independent of range andP4 = BP50, i.e. the MAP at half the HR range. The averagegain (G) or slope of the curve between the two inflectionpoints is given by G = −P2 × P3/4 and the upperplateau = P1 + HR range (P2).

Basal MAP and HR, maximal pressor and depressorresponses and the parameters of both linear fit andsigmoidal fitting baroreceptor curves of SHR and WKYinfused with VEH and L-NMMA were analyzed usingtwo-way analysis of variance with repeated measures. Inthe SHR group the effects of L-NMMA and L-Arg vsVEH were analysed by one-way analysis of variance withrepeated measures. Student–Newman–Keuls was used asa post hoc test. Student’s t-test was used to analyze thevalues of NOS activity. Significance was accepted at avalue of P < 0.05.

ResultsBasal values and pressure changesBasal values of MAP and WKY (n = 10) and HR in SHR(n = 9) groups were 108 ± 2 mmHg and 353 ± 15 beats/min;185 ± 4 mmHg and 373 ± 12 beats/min, respectively at thebeginning of the experiments. These values were notchanged by either VEH and L-NMMA (30 ng/kg/min)administrations into the NTS (Figure 1). In both groupsintravenous bolus injections of graded doses of phenyle-phrine and sodium nitroprusside caused dose-relatedchanges in MAP: increases of +6 ± 2 up to +56 ± 3 mmHg(0.1 up to 6.4 µg/kg) and decreases of −2 ± 0.3 down to −25 ± 2 mmHg (0.2 up to 12.8 µg/kg), respectively wereobserved in WKYVEH rats. The S-shaped pressor responsewas also observed in SHRVEH but the maxi-mal response (+36 ± 3 mmHg) was significantly reduced(vs WKYVEH, P < 0.05, Figure 1). Maximal depressor response of SHRVEH (−28 ± 4 mmHg) was unaffected(Figure 1). During continuous infusion of L-NMMA(30 ng/kg/min) in the NTS no significant changes wereobserved in pressor and depressor responses of WKY andSHR, although there was a tendency toward reduction ofmaximal pressor response in the SHRL-NMMA (+25 ± 5 vs+36 ± 3 mmHg in SHRVEH, Figure 1).

Baroreceptor reflex control of HRThe average sigmoidal-fitting and sigmoidal-fitting para-meters of baroreceptor curves in WKY and SHR aredepicted in Figure 2 and Table 1, respectively. SHRVEH

curve was shifted to the right of WKYVEH curve, closelyfollowing the resting MAP. Attenuation of both HR range(87 ± 9 beats/min, with lower plateau displaced towardhigher HR value) and reflex gain (−2.90 ± 0.57 beats/min/mmHg) were observed in SHRVEH when comparedto WKYVEH (Table 1). During continuous infusion of L-NMMA (30 ng/kg/min) into the NTS loading/unloadingof baroreceptors produced in the SHR group a small butsignificant displacement of baroreceptor curve to the right.

heart rate 5 P1 1 P2

[1 1 eP3(MAP2P4)]

111123456789101112345678920111234567893011123456789401112345678950111234567111

NO in the NTS and baroreflex modulation Pontieri et al. 1995

Throughout the baroreflex function test, MAP was higher(SHRL-NMMA = 188 ± 6 vs 180 ± 7 mmHg in SHRVEH) anda further significant reduction (−34%, determined by aslight elevation of lower plateau and a significant reduc-tion in the upper plateau, Table 1) was observed in theHR range. L-NMMA infusion into the NTS did notchange SHR reflex gain (−2.59 ± 0.33 vs −2.90 ± 0.57beats/min/mmHg in SHRVEH). Baroreceptor reflex func-tion assessed by linear regression equations (values aregiven in Table 1) yielded similar results: reflex brady-cardia and reflex tachycardia of SHRVEH are significantlydepressed when compared to WKYVEH. During L-NMMAinfusion SHR rats showed further depression of bothbradycardic and tachycardic responses, in accordance tothe marked depression in the HR range, revealed by thelogistic function curve (Table 1). On the other hand nochanges (measured by sigmoidal fitting parameters or by

linear regression analysis) were observed during L-NMMA (30 ng/kg/min) administration into the NTS ofWKY rats.

The effects of L-NMMA on HR reflex control in SHRrats are due to the blockade of NO release into the NTSbecause the reduced HR range (71 ± 4 beats/min, seeTable 2, sequence 1) was completely reversed when L-NMMA treatment was replaced by L-Arg (900 ng/kg/min)administration into the NTS of the same rats. The oppo-site sequence of treatments in other SHR rats (sequence2 in Table 2) showed that increased NO formation intothe NTS did not improve the baroreflex control: allmeasured parameters were unaffected by L-Arg treat-ment. L-NMMA administration after L-ARG infusionagain depressed specifically and markedly the HR range(47 ± 15 beats/min, a 33% reduction, Table 2).

In three other WKY rats in which NTS was sequentiallytreated with VEH, L-NMMA 30 ng/kg/min and L-NMMA300 ng/kg/min (Figure 3) there was no change in basalvalues of MAP and HR. Baroreceptor reflex control of HRwas not changed by the smaller dose, but significantchanges were observed with the larger dose: MAPthroughout the baroreflex test (points in Figure 3) wasdisplaced from 114 ± 2 to 122 ± 4 mmHg (P < 0.05) andHR range (113 ± 2 beats/min during L-NMMA 300 ng/kg/min, P < 0.05) was on average 31% smaller than the rangesmeasured during VEH and L-NMMA (30 ng/kg/min)infusions (166 ± 21 and 163 ± 12 beats/min, respectively).There was a tendency to increase gain during the higherL-NMMA dose infusion (−6.58 ± 3.34 vs −3.97 ± 1.87beats/min/mmHg during small dose infusion, P > 0.05).

111123456789101112345678920111234567893011123456789401112345678950111234567111

1996 Journal of Hypertension 1998, Vol 16 No 12 (part 2)

Average values of mean arterial pressure (MAP, upper panel) andmaximal pressor and depressor responses (DMAP, lower panel) tointravenous phenylephrine and sodium nitroprusside, respectively inWKY (n = 10) and SHR (n = 9) groups. Measurements were madebefore (basal) and during vehicle (VEH) and L-NMMA (30 ng/kg/min)infusions into the nucleus tractus solitarii of conscious rats.Significance *(P < 0.05) vs WKY.

Fig. 1

Average logistic function curves expressing the relation betweenmean arterial pressure (MAP) and heart rate (HR) in conscious WKY(n = 10) and SHR rats (n = 9) treated with vehicle (VEH) or L-NMMA(30 ng/kg/min) into the nucleus tractus solitarii. Points on the curvesrepresent average basal MAP and HR throughout the baroreflexfunction test. Significances in Table 1.

Fig. 2

Dorsal brain stem: areas injected and NOS activityMicroscopic examination of the brain stem in SHR andWKY rats included in the functional studies revealed thatin both groups the microinjections were directed mainlyto the commissural NTS. The dye apparently followedthe structure of the NTS by spreading predominantly ina rostro-caudal direction that extended 509 ± 39 µm inSHR and 483 ± 51 µm in WKY group, with the mean pointof injection located around the obex in both SHR andWKY rats. Additionally to the NTS, the dye stained aportion of the dorsal motor nucleus of the vagus in twoSHR and three WKY rats.

In other SHR (n = 5) and WKY groups (n = 8), measure-ment of L-[3H]citrulline formation in dorsal brain stemareas, corresponding mainly to the NTS and dorsal motornucleus of the vagus indicated that basal constitutive NOSactivity was significantly depressed in SHR whencompared to WKY group (0.96 ± 0.04 vs 1.57 ± 0.04 pmol/min/µg protein, respectively, P < 0.05, Student’s t-test).

DiscussionIn the present study we investigated the effects ofendogenous NO in the nucleus tractus solitarii on thebaroreflex control of HR in conscious WKY and SHR withthe use of the NOS inhibitor L-NMMA. There are threenew findings: (1) inhibition of NO synthesis reduced therange without changing the gain of heart rate–pressurerelationship in both WKY and SHR; (2) constitutive NOSactivity was reduced in the dorsal brain stem of SHR and,accordingly, attenuation of HR range was attained with asmaller dose of NOS inhibitor; (3) the impaired barore-ceptor reflex control of HR was not improved by an excessof L-arg infused in the NTS over a period of 30–40 min.

The involvement of NO in the control of blood pressurehas been proved by several experiments. Chronic NOblockade caused sustained hypertension [3,13,15,17]; NO served as an important intercellular mediator in thevasculature, controlling systemic vascular resistance andregional distribution of flow [1,2,4]; NO is involved in the

111123456789101112345678920111234567893011123456789401112345678950111234567111

NO in the NTS and baroreflex modulation Pontieri et al. 1997

Table 1 Parameters of logistic function curve (sigmoidal fitting) and linear regression analysis of baro-receptor reflex control of heart rate in conscious WKY and SHR treated with vehicle (VEH) or L-NMMA into the nucleus tractus solitarii

WKY (n = 10) SHR (n = 9)

L-NMMA L-NMMA VEH (30 ng/kg/min) VEH (30 ng/kg/min)

Logistic function curveLower plateau (beats/min) 304 ± 10 291 ± 13 344 ± 15* 356 ± 11*Upper plateau (beats/min) 444 ± 13 433 ± 15 431 ± 17 414 ± 14†

HR range (beats/min) 140 ± 12 142 ± 9 87 ± 9* 57 ± 7*,†

BP50 (mmHg) 109 ± 3 114 ± 2 170 ± 6* 180 ± 7*,†

Gain (beats/min/mmHg) −4.92 ± 0.83 −4.99 ± 0.92 −2.90 ± 0.57* −2.59 ± 0.33*

Linear regressionReflex bradycardiaSlope (beats/min/mmHg) −1.28 ± 0.15 −1.26 ± 0.16 −0.56 ± 0.12* −0.21 ± 0.17*,†

Intercept (beats/min) 2.87 ± 2.53 1.84 ± 3.15 −3.82 ± 1.59* −5.83 ± 1.99*Reflex tachycardiaSlope (beats/min/mmHg) −3.13 ± 0.38 −4.14 ± 0.86 −2.20 ± 0.18* −1.03 ± 0.28*,†

Intercept (beats/min) 24.19 ± 3.57 11.64 ± 7.73 −1.56 ± 4.59* 14.03 ± 3.14*,†

Values are mean ± SEM. BP50 = mean arterial pressure at midrange.

Significances (P < 0.05): *vs WKY; †vs VEH.

Table 2 Basal values of mean arterial pressure (MAP) and heart rate (HR) and parameters of logistic function curve of baroreceptorreflex control of HR in conscious SHR rats treated with L-NMMA and L-arginine (L-Arg) into the nucleus tractus solitarii with twodifferent sequences

Sequence 1 (n = 5) Sequence 2 (n = 4)

L-NMMA → L-arg L-arg → L-NMMAVEH → (30 ng/kg/min) (900 ng/kg/min) VEH → (900 ng/kg/min) (30 ng/kg/min)

Basal valuesMAP (mmHg) 181 ± 10 188 ± 8 184 ± 12 179 ± 10 192 ± 8 188 ± 10HR (beats/min) 376 ± 17 383 ± 12 397 ± 12 349 ± 22 354 ± 21 355 ± 20

Logistic function curveLower plateau (beats/min) 355 ± 21 361 ± 15 376 ± 15 332 ± 22 340 ± 18 351 ± 19Upper plateau (beats/min) 450 ± 15 432 ± 19 456 ± 12 412 ± 30 409 ± 29 398 ± 28HR range (beats/min) 94 ± 7 71 ± 4* ,† 80 ± 14 80 ± 18 70 ± 18 47 ± 15*,†

BP50 (mmHg) 169 ± 9 182 ± 10 177 ± 12 170 ± 11 179 ± 10 176 ± 12Gain (beats/min/mmHg) −2.60 ± 0.75 −2.12 ± 0.42 −3.05 ± 0.75 −3.20 ± 0.95 −3.04 ± 1.15 −3.18 ± 0.39

Values are mean ± SEM. BP50 = MAP at midrange.

Significances (P < 0.05): *vs VEH; †vs L-arg.

reflex control of HR and sympathetic nerve activity[11,13,14,16,23]. An elegant experiment [12] investigatingthe effects of intravenous infusion of NO donor and NOSinhibitor on the reflex arc (with measurement of aorticnerve activity, pre- and post-ganglionar sympathetic nerveactivities, HR and blood pressure) showed that NO-modu-lated sympathetic tone not by altering the afferent orefferent limbs but by acting on central pathways of thebaroreflex. Actually neurons immunoreactive to NOSwere localized in many bulbar areas involved in theprimary integration of the reflex [6,7]. Specifically NOS-immunoreactive neurons were found in the NTS (firstsynaptic relay of peripheral afferents), showing a distrib-ution complementary to tyrosine hydroxylase immunore-active neurons [6]. Also, administration of L-glutamate,the main neurotransmitter of baroreceptor neurons intothe NTS, elicited (via NMDA receptors) the release of aNO-like factor, causing reduction of both pressure andheart rate [8]. These evidences pointed out to the NTSas a site where NO modulates the baroreflex control andstressed the validity of the present findings.

Stimulation of baroreceptor afferents determines releaseof L-glutamate in the NTS [24]. L-Glutamate acts onNMDA receptors to activate NOS, that catalyzes theformation of NO and L-citrulline from L-Arg [25]. NO isan intercellular messenger molecule readily diffusible thatacts to generate a cascade effect by activating guanylatecyclase in several cell types and thereby augmentinglevels of second messenger cyclic GMP [10]. This is anefficient and widespread transduction mechanism thatleads to the opening of ion channels. What we showednow is that this mechanism is acting in the NTS of both

normotensive and hypertensive animals to maintainsynaptic efficiency in a larger HR interval during afferentstimulation. In addition we showed that NOS activity isreduced in the NTS of SHR: NO formation (as indicatedby L-citrulline formation) is reduced and HR range issignificantly shortened by L-NMMA treatment in a dosethat does not alter HR control in WKY rats. On the otherhand, the same effect in the WKY rats is only attainedby the administration into the NTS of a L-NMMA dose10 times higher. Since L-Arg alone into the NTS (exper-imental sequence 2) does not alter the baroreceptor reflex control of HR (it only reverses the changes inducedby previous administration of L-NMMA – experimentalsequence 1), the reduction of the substrate and/or NOavaibility in the NTS of SHR are not the cause of theimpaired baroreceptor function, that characterizes thechronic phase of hypertension.

The use of a small dose of L-NMMA (it was 103–105 timessmaller than those used intravenously [12–14,16] and 23times smaller than the NTS dose administered in rabbits[11]) was a great advantage in the present study. In theabsence of pressure increase, the net modulatory effectof NO on baroreflex function was uncovered and sepa-rated from that elicited by the pressure change itself. Asnoted previously [16,26], changes in baroreceptor reflexgain are related to both the induced changes in pressurelevels and the proper effects of the administered agonist/antagonist.

It is outstanding that L-NMMA and L-Arg into the NTSonly changed the range of HR–pressure relationship(decrease in the range and correction of the previouschange, respectively), without changing baroreflex gain.Head and McCarty [22] analyzing the vagal and sympa-thetic components of the HR range and gain of the barore-ceptor heart rate reflex in conscious rats showed that thevagus makes a greater contribution to the HR range thanthe sympathetic, the converse being true for the gain.Also Liu et al. [14] described that NO played a role onthe baroreflex control of HR (with important vagal compo-nent) but not on the control of renal sympathetic nerveactivity. Although the present experiments were notdesigned to differentiate between vagal and sympatheticcomponents during loading/unloading of the reflex in thepresence of L-NMMA and L-Arg into the NTS, one mayspeculate (based on the specific changes observed in theHR range and on previous studies [14, 22]) that NOmodulation affects primarily vagal component of the HRreflex. This hypothesis remains to be confirmed.

The data of the present study show clearly that NO modu-lates centrally the activity of mechanoreceptor afferentsstimulated by the transient pressure changes. If theactivity of low pressure receptors (and other receptors) isalso modulated by NTS NO release, it remains to bedetermined.

111123456789101112345678920111234567893011123456789401112345678950111234567111

1998 Journal of Hypertension 1998, Vol 16 No 12 (part 2)

Average logistic function curves expressing the relation betweenmean arterial pressure (MAP) and heart rate in three conscious WKYtreated with vehicle (VEH) and two different doses of L-NMMA (30 and 300 ng/kg/min) into the nucleus tractus solitarii. Points onthe curves represent average basal MAP and HR values throughoutthe baroreflex function test. Significant differences are given in thetext.

Fig. 3

It should be stressed that NTS is not the only bulbar areawhere NO formation modulates the baroreceptor func-tion. NOS immunoreactivity was present in dorsal motornucleus of the vagus, nucleus ambiguus and ventrolateralmedulla as well [6,7]. In fact NO is shown to be an inter-cellular messenger released after NMDA stimulation inthe neurons of dorsal motor nucleus of the vagus [27].NO has also proved to influence vasomotor tone: NOdonor attenuated renal sympathetic nerve activity afteradministration into the rostral ventrolateral medulla andpotentiated this effect following administration into thecaudal ventrolateral medulla [28].

In summary, our data show that NO modulation of thebaroreflex is active in both WKY and SHR. It keeps thesynaptic efficiency in a larger interval of pressure changesby determining appropriate HR responses. Although NOSactivity and therefore NO modulation are reduced in theNTS of SHR the baroreceptor reflex control of HR isvery sensitive to small changes in NO formation, beingdepressed even with low dose of NOS inhibitor.

AcknowledgmentThe authors thank Mauro Sucupira for expert technicalassistance.

References1 Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology,

pathophysiology, and pharmacology. Pharmacol Rev 1991; 43:109–142.2 Dawson TM, Dawson VL. Nitric oxide: actions and pathological roles. The

Neuroscientist 1995; 1:7–18.3 Gross SS, Wolin MS. Nitric oxide: pathophysiological mechanisms. Annu

Rev Physiol 1995; 57:737–769.4 Umans JG, Levi R. Nitric oxide in the regulation of blood flow and arterial

pressure. Annu Rev Physiol 1995; 57:771–790.5 Garthwaite J, Boulton CL. Nitric oxide signaling in the central nervous

system. Annu Rev Physiol 1995; 57:683–706.6 Ohta A, Takagi H, Matsui T, Hamai Y, Iida S, Esumi H. Localization of nitric

oxide synthase–immunoreactive neurons in the solitary nucleus andventrolateral medulla oblongata of rat: their relation to catecholaminergicneurons. Neurosci Lett 1993; 158:33–35.

7 Rodrigo J, Springall DR, Uttenthal O, Bentura ML, Abadia-Molina F,Riveros-Moreno V, et al. Localization of nitric oxide synthase in the adult ratbrain. Phil Trans R Soc Lond B 1994; 345:175–221.

8 Di Paola ED, Vidal MJ, Nisticó G. L-glutamate evokes the release of anendothelium-derived relaxing factor-like substance from the rat nucleustractus solitarius. J Cardiovasc Pharmacol 1991; 17(suppl 3):S269–S272.

9 Lawrence AJ, Jarrott B. Nitric oxide increases interstitial excitatory aminoacid release in the rat dorsomedial medulla oblongata. Neurosci Lett 1993;151:126–129.

10 Wood PL, Emmett MR, Rao TS, Cler J, Mick S, Iyengar S. Inhibition if nitricoxide synthase blocks N-methyl-D-aspartate-, quisqualate-, kainate-,harmaline-, and pentylenetetrazole-dependent increases in cerebellar cyclicGMP in vivo. J Neurochem 1990; 55:346–348.

11 Harada S, Tokunaga S, Momohara M, Masaki H, Tagawa T, Imaizumi T, et al. Inhibition of nitric oxide in the nucleus tractus solitarius increasesrenal sympathetic nerve activity in rabbits. Circ Res 1993; 72:511–516.

12 Jimbo M, Suzuki H, Ichikawa M, Kumagai K, Nishizawa M, Saruta T. Role ofnitric oxide in regulation of baroreceptor reflex. J Auton Nerv System 1994;50:209–219.

13 Vasquez EC, Cunha RS, Cabral AM. Baroreceptor reflex function in ratssubmitted to chronic inhibition of nitric oxide synthesis. Brazilian J MedBiol Res 1994; 27:767–774.

14 Liu J-L,Murakami H, Zucker I. Effects of NO on baroreflex control of heartrate and renal nerve activity in conscious rabbits. Am J Physiol 1996; 270:R1361–R1370.

15 Manning Jr RD, Hu L, Mizelle L, Montani J-P, Norton MW. Cardiovascularresponses to long-term blockade of nitric oxide synthesis. Hypertension1993; 22:40–48.

16 Kumagai H, Averill DB, Khosla MC,Ferrario CM. Role of nitric oxide andangiotensin II in the regulation of sympathetic nerve activity inspontaneously hypertensive rats. Hypertension 1993; 21:476–484.

17 Ribeiro MO, Antunes E, Nucci G, Lovisolo SM, Zatz R: Chronic inhibitionof nitric oxide synthesis: a new model of arterial hypertension. Hypertension1992; 20:298–303.

18 Michelini LC, Bonagamba LGH. Baroreceptor reflex modulation byvasopressin microinjected into the nucleus tractus solitarii of consciousrats. Hypertension 1988; 11(suppl 1):I-75–I-79.

19 McKee M, Scavone C, Nathanson JA. Nitric oxide, cGMP, and hormoneregulation of active sodium transport. Proc Natl Acad Sci USA 1994;91:12 056–12 060.

20 Bradford MM. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principles of protein-dyebinding. Anal Biochem 1976; 72:248.

21 Kent BB, Drane JW, Blumenstein B, Manning JW. A mathematical model to assess changes in the baroreceptor reflex. Cardiology 1972;57:295–310.

22 Head GA, McCarty R. Vagal and sympathetic components of the heart raterange and gain of the baroreceptor-heart rate reflex in conscious rats. J Auton Nerv System 1987; 21:203–213.

23 Kumagai K, Suzuki H, Ichikawa M, Jimbo M, Murakami M, Ryuzaki M, et al.Nitric oxide increases renal blood flow by interacting with the sympatheticnervous system. Hypertension 1994, 24:220–226.

24 Ohta H, Li X, Talman WT. Release of glumate in the nucleus tractussolitarii in response to baroreflex activation in rats. Neuroscience 1996;74:29–37.

25 Knowles RG, Palacios M, Palmer RM, Moncada S. Formation of nitric oxidefrom L-arginine in the central nervous system: a transduction mechanism forstimulation of the soluble guanylate cyclase. Proc Natl Acad Sci USA1989; 86:5159–5162.

26 Santos CM, Pontieri V, Leomil Neto M, Michelini LC. Losartan improvesbaroreflex control of heart rate of coarcted hypertensive rats. Am J Physiol(Heart Circulatory Physiol. 38) 1995; 269: H812–H818.

27 Travagli RA, Gillis RA. Nitric oxide-mediated excitatory effect on neurons ofdorsal motor nucleus of vagus. Am J Physiol (Gastrointest. Liver Physiol.29) 1994; 266: G154–G160.

28 Shapoval LN, Sagach VF, Pobegailo LS. Nitric oxide influencesventrolateral medullary mechanisms of vasomotor control in the cat.Neurosci Lett 1991; 132:47–50.

111123456789101112345678920111234567893011123456789401112345678950111234567111

NO in the NTS and baroreflex modulation Pontieri et al. 1999