I

Ia

b

c

d

e

a

AA

KIVRBC

1

pala(dtdm(

Vddv

nf

1d

Respiratory Physiology & Neurobiology 175 (2011) 375–382

Contents lists available at ScienceDirect

Respiratory Physiology & Neurobiology

journa l homepage: www.e lsev ier .com/ locate / resphys io l

mpact of the vagal feedback on cardiorespiratory coupling in anesthetized rats

rina Topchiya,b,e,∗, Miodrag Radulovackic, Jonathan Waxmana,d, David W. Carleya,b

Center for Narcolepsy, Sleep and Health Research, M/C 802, University of Illinois at Chicago, 845 South Damen Ave., Chicago, IL 60612, USADepartment of Medicine, M/C 719, University of Illinois at Chicago, 840 S. Wood Street, Chicago, IL 60612, USADepartment of Pharmacology, M/C 868, University of Illinois at Chicago, 901 S. Wolcott Ave., Chicago, IL 60612, USADepartment of Electrical Engineering, University of Illinois at Chicago, Chicago, IL 60612, USAA.B. Kogan Research Institute for Neurocybernetics, Rostov State University, 194/1 Stachka Ave., Rostov-on-Don 344090, Russia

r t i c l e i n f o

rticle history:ccepted 28 December 2010

eywords:ntertrigeminal regionagotomyespirationlood pressureoherence

a b s t r a c t

Cardiorespiratory coupling can be significantly influenced by both pontine and vagal modulation ofmedullary motor and premotor areas. We investigated influences of the pontine intertrigeminal region(ITR) and peripheral vagal pathways on the coupling between systolic blood pressure (SBP) and res-piration in 9 anesthetized rats. Glutamate injection into the ITR perturbed both respiration and SBPand decreased SBP-respiratory coherence (0.95 ± 0.01 vs 0.89 ± 0.02; (p = 0.01). Intravenous infusionof serotonin (5-HT) produced apnea and hypertension and also decreased SBP-respiratory coherence(0.95 ± 0.01 vs 0.72 ± 0.06; p = 0.04). Bilateral vagotomy eliminated the cardiorespiratory coherence per-turbations induced by central (glutamate injection into the ITR: 0.89 ± 0.03 vs 0.86 ± 0.03; p = 0.63) andperipheral (5-HT infusion: 0.89 ± 0.03 vs 0.88 ± 0.02; p = 0.98) pharmacologic manipulations. Glutamate

stimulation of the ITR postvagotomy increased the relative spectral power density of SBP in the respira-tory frequency range (0.25 ± 0.08 vs 0.55 ± 0.06; p = 0.01). The data suggest that SBP-respiratory couplingis largely mediated within the central nervous system, with vagal systems acting in a way that dis-rupts coherence during transient cardiorespiratory disturbances. Although decreased cardiorespiratorycoherence may increase cardiac work during perturbations, this may be physiologically advantageous in

uilibr

restoring homeostatic eq

. Introduction

Cardiorespiratory coupling can be regulated by both central anderipheral mechanisms. Pontine structures were shown to playn essential role in coordination of respiratory and cardiovascu-ar activity. It was recently demonstrated that respiratory sinusrrhythmia, respiratory modulation of sympathetic nerve activityTraube-Hering waves in arterial pressure) and post-inspiratoryischarges from vagal efferents were eliminated after pontine

ransection (Baekey et al., 2008). Pontine nuclei modulate car-iorespiratory afferentation and its primary integration (includingodulation of baroreflexes) in the nucleus of the solitary tract (NTS)

Felder and Mifflin, 1988; Mifflin and Felder, 1990; Paton et al.,

Abbreviations: ITR, intertrigeminal region; NTS, nucleus of the solitary tract;LM, ventrolateral medulla; RVLM, rostral ventrolateral medulla; ECG, electrocar-iogram; TT, total breath duration; CVTT, coefficient of variation of total breathuration; BP, blood pressure; SBP, systolic blood pressure; CVSBP, coefficient ofariation of systolic blood pressure; 5-HT, serotonin.∗ Corresponding author at: Department of Medicine, M/C 719, University of Illi-ois at Chicago, 840 S. Wood Street, Chicago, IL 60612, USA. Tel.: +1 312 413 8461;

ax: +1 312 996 1225.E-mail address: irinatopchiy@yahoo.com (I. Topchiy).

569-9048/$ – see front matter. Published by Elsevier B.V.oi:10.1016/j.resp.2010.12.017

ium of respiration and blood pressure.Published by Elsevier B.V.

1990). The pons also may be directly involved in the regulationof cardiorespiratory efferent patterns, generated by structures ofventrolateral medulla (VLM) (Nunez-Abades et al., 1990; Morrisonet al., 1994).

The activity of central cardiorespiratory regulating systems islargely dependent on peripheral inputs, including vagal pathways(Kalia and Mesulam, 1980; Kalia and Sullivan, 1982; Dick et al.,2009). Vagal afferents arising from numerous types of receptorsinfluence both respiratory pattern and vascular tone and impact oncardiorespiratory coupling expressed as respiratory sinus arrhyth-mia, which is suppressed after vagotomy (Eckberg, 1983; Eckbergand Sleight, 1992). Vagotomy has further been suggested to disruptcentral coordination of cardiovascular and respiratory activities inthe NTS (Loewy and Spyer, 1990; Paton, 1998) and within the effer-ent circuits of VLM (Feldman and Ellenberger, 1988; Mandel andSchreihofer, 2006), including vagal efferents within the nucleusambiguus (McAllen and Spyer, 1975, 1976; Neff et al., 2003).

Recent findings indicate that the pontine intertrigeminal region

(ITR) – the group of cells in the lateral pons among the fibersbetween the motor and principal sensory trigeminal nuclei – isinvolved both in respiratory regulation (Chamberlin and Saper,1998, 2003; Radulovacki et al., 2003a,b) and cardiovascular con-trol (Topchiy et al., 2009). Anatomical connections of the ITR with

3 logy &

tSi(vse

wcasb

2

aloL

2

kxtwsmacvtpwcca

btcrtsstt

tw(O

tseawSHdb

76 I. Topchiy et al. / Respiratory Physio

he nucleus of the solitary tract and the VLM (Chamberlin andaper, 1992, 1998, 2003; Verner et al., 2008) suggest its integrat-ng role in a wide range of respiratory and cardiovascular pathwaysChamberlin and Saper, 1998). We previously showed that bilateralagotomy amplified and unmasked increases of systolic blood pres-ure as well as pulse pressure, induced by ITR stimulation (Topchiyt al., 2009).

In view of the above-mentioned studies, the aim of this workas to explore the impact of ITR perturbation on cardio-respiratory

oupling and the role of vagal pathways in these effects. Wessessed cardio-respiratory coupling before and after bilateralupranodose vagotomy with and without stimulation of the ITRy glutamate.

. Experimental procedure

Experiments were performed on 9 spontaneously breathing,nesthetized, adult, male Sprague–Dawley rats (280–300 g, Har-an, Indianapolis, IN). All procedures complied with the guidelinesf the National Institute of Health Guide for the Care and Use ofaboratory Animals (NIH Publications No. 80-23) revised 1996.

.1. Surgical preparation

Rats were anesthetized with a combination of 80 mg/kgetamine (Abbott Laboratories, North Chicago, IL) and 5 mg/kgylazine (Phoenix Scientific Inc., St. Joseph, MO) given by intraperi-oneal injection. After achieving a stable plane of anesthesia, whichas controlled by the absence of a toe-pinch reflex a midline inci-

ion was made from anterior on the mandible caudally to theanubrium (from the sternal margin of the neck to the larynx)

nd the skin was pulled laterally with retractors. The trachea wasannulated. The left sternohyoid and omohyoid muscles and sali-ary glands were gently retracted by round-tip forceps exposinghe external carotid artery and jugular vein. Blunt dissection oferivascular connective tissue was performed. The vagus nerveas identified as the white strand that lies between the common

arotid artery and the internal jugular vein. The bifurcation of thearotid artery into the external and internal branches was taken aslandmark.

Rectus capitus and longus coli were divided transversely, usinglunt dissection as far rostrally in the field as possible. The part ofhe vagus nerve in the proximity of the bifurcation of the commonarotid artery was traced to the point where it exits the poste-ior lacerated foramen of the occipital bone of the scull and formshe nodose ganglion. The nodose ganglion was gently isolated by amall round-tip forceps and a silk ligature was threaded under theupranodose vagus nerve. The same procedure was performed athe right side. The ligatures were tightened to achieve vagotomy athe appropriate point in the protocol.

For blood pressure monitoring a catheter was inserted intohe left femoral artery and secured by a suture. Another catheteras inserted into the left femoral vein for bolus infusions of 5-HT

5-hydroxytryptamine hydrochloride, MP Biomedical LLC, Aurora,H).

After the surgical preparation, rats were placed in a stereo-axic apparatus (Stoelting Co., Wood Dale, IL). The incisor bar waset at the same level as the ear bars (interaural zero).A unilat-ral osteotomy was made to allow access to rostral lateral pons,nd the dura was carefully removed. Two-barrel micropipettes

ere made using standard filament glass (1 mm × 0.25 mm, A-M

ystems, Carlsberg, WA) and a vertical puller (model no. 50-239,arvard Apparatus Ltd., Kent, England) to obtain an overall tipiameter of 10–20 �m. The micropipette was introduced into therain on a dorso-ventral axis to allow pressure microinjections into

Neurobiology 175 (2011) 375–382

the ITR (AP = −9.30 from bregma; ML = 2.4 from the midline suture;DV = −8.0 from the brain surface; Paxinos and Watson, 2004). Thesurface of the brain was used as the zero point for the dorso-ventralstereotaxic coordinates.

A millipulse pressure injector (model Picospritzer II, GeneralValve Co., Fairfield, NJ) was used to inject glutamate (l-glutamicacid monosodium salt, 10 mM, 30 nl, ICN Biomedicals, Aurora,OH) or oil red O-dye (Sigma, St. Louis, MO; solution of 7 mg in1 ml ethanol) into the ITR. The dose of glutamate was chosenaccording the previous works (Chamberlin and Saper, 1992, 1998;Radulovacki et al., 2003a,b, 2007), which showed the effectivenessof this amount to evoke the prominent apneic reactions from theITR. All drugs were dissolved in 0.2 M PBS. Injection volumes weredetermined by measuring the displacement of the pipette fluidmeniscus with a calibrated eyepiece reticle in a binocular stere-ozoom microscope (model 48920-10, Cole-Parmer, Vernon Hills,IL). In all cases, the target injection volume was 30 nl. AccordingNicholson (1985) it was suggested that within the first 30 s theeffective diffusion radius of this volume of glutamate is limited toapproximately 170–180 �m.

2.2. Recording procedure

In each experimental protocol we performed a four-channelrecording: (1) arterial blood pressure registered using a Transpac IVtransducer (Hospira, Lake Forest, IL); (2) electrocardiogram (ECG)acquired from needle electrodes placed in the left axillary and rightflank regions; (3) respiration (VelcroR Tab-Infant-Ped; SleepmateR

Technologies); (4) injection marker (logic level pulse provided bythe pressure injector). The respiration recording system compriseda 1 cm2 piezoelectric crystal attached to an elastic band that wasfixed around the animal at a substernal level. The crystal pro-vided quantitative measurements of respiratory timing and relativeuncalibrated measurements of respiratory volume.

After conventional amplification and filtering (1–50 Hz band-pass; Grass Model 12, West Warwick), the analog data weredigitized (sampling frequency 200/s) and recorded using BrainWave for Windows software (Datawave Systems, Longmont, CO).

2.3. Experimental protocol

Each recording began with a 10 min registration of the baselineactivity prior to any injections. In order to identify an ITR apneic site,the pipette was positioned into the stereotaxically defined inter-trigeminal area and advanced ventrally in 100 �m increments untilglutamate injection produced an apnea at least 2.5 s in duration,corresponding to a ≥50% increase in breath duration, on average.

In all animals at least two repeated glutamate injections weremade at the “respiratory effective” site to better document theduration and reproducibility of the response. The injections wereseparated by a 10 min interval, which was sufficient for the visu-ally assessed respiratory pattern to return to baseline. After thenext 10-min interval a bolus of 5-HT (0.05 M, 0.5 �l over 5 s) wasinfused intravenously, using a Hamilton syringe and infusion pump(model KDS210, KD Scientific Inc., Hollister, MA). Apnea followingthe 5-HT infusion indicated the integrity of the nodose ganglia andvagal afferent system. Ten minutes later vagotomy was producedby tightening the ligatures placed around the vagus nerves. After a30-min interval, sufficient for visually evident stabilization of therespiratory pattern, another bolus of 5-HT was infused to confirm

the vagal transection rostral to the nodose ganglia. To determine theeffect of vagotomy on glutamate stimulation of the ITR, two addi-tional glutamate injections were made at 10-min intervals. Finally,oil red O-dye was microinjected into the functionally identified ITRapneic site to aid in histological verification.

I. Topchiy et al. / Respiratory Physiology & Neurobiology 175 (2011) 375–382 377

Table 1The mean total breath time (TT), tidal volume (VT), systolic blood pressure (SBP) and heart rate (HR) in pre-vagotomy (PreVagX) and post-vagotomy (PostVagX) conditions.

TT (s) VT (A.U.) SBP (mm Hg) HR (beat/min)

PreVagX PostVagX PreVagX PostVagX PreVagX PostVagX PreVagX PostVagX

1.66 ± 0.17 2.23 ± 0.21 88.28 ± 13 145.86 ± 13.8 158.94 ± 10.03 169.16 ± 9.98 195.9 ± 13.39 217.5 ± 20.03

T ion of

2

abaAcgt(pb

psdqeopseotsr

Sspsosipoqpv

pmrsag

pt2a(

ts

p-Values 0.002 0.03

he table shows group means ± standard errors, *p ≤ 0.05 with respect to pre-inject

.4. Data analysis

An automated adaptive threshold algorithm was used to detectnd quantify: the duration (TT) and tidal volume (VT) of eachreath, as well as systolic pressure (SBP), diastolic pressure (DBP)nd pulse interval for each beat in the blood pressure waveform.nalysis focused on multiple 240-s intervals from each experiment:ontrol, pre-vagotomy, post-vagotomy, and immediately after eachlutamate injection or serotonin infusion. Beat by beat values werehen linearly interpolated to produce evenly sampled waveforms10 samples per second) suitable for coherence analysis. A zero-hase 64-point FIR anti-aliasing filter was then applied followedy a downsampling operation.

Coherence represents a linear measure of coupling between tworocesses, measured here as physiological signals. First, the timeeries of each signal is Fourier-transformed, creating a frequencyomain representation. Then, cross-spectral analysis allows auantification of linear coupling between the signals. In gen-ral, coherence can be determined for every individual frequencyr, more commonly, coherence may be assessed over a relevantre-determined frequency range. Here, we examined coherencepecific to the range of respiratory frequencies observed over thentire study. Additionally, we computed the spectral power densityf the SBP signal in the respiratory frequency range relative to theotal average SBP power (relative spectral power density). Relativepectral power thus represented magnitude of SBP modulation byespiration.

The relative spectral power density and the coherence betweenBP and respiration were calculated as follows. The Fourier powerpectral density was computed and recorded for each systolic bloodressure and respiratory signal divided into non-overlapping 30 segments using a 256 sample Hanning window. A linear detrendingperation was applied to each segment prior to computing powerpectral densities, and the resulting density values were normal-zed to the sampling frequency. Coherence between systolic bloodressure and respiration and the relative spectral power densityf SBP were computed for each interval within the respiratory fre-uency range. This range was separately determined for control,re-vagotomy, post-vagatomy, and post-glutamate injection inter-als using the respiratory power spectral densities.

Finally, the eight 30-s segments of each 240 s interval wereooled to produce average coherence and relative power densityeasurements. This has the advantage of increasing the statistical

eliability of the estimates. The interval length of 240 s was cho-en in view of our previous data, which showed that respiratorynd BP disturbances following infusion of 5-HT or injection of thelutamate into the ITR needs at least 200 s to stabilize.

The mean ± SE values of all parameters, the relative spectralower density of SBP within the respiratory frequency range andhe coherence between SBP and respiration were calculated over40 s intervals at baseline, just prior to and 30 min after vagotomy,

nd immediately following each injection (glutamate) or infusionserotonin).

The effects of glutamate injection, serotonin infusion and vago-omy were assessed by paired t-tests. In all cases, statisticalignificance was inferred for p < 0.05.

0.43 0.04

glutamate. A.U. – arbitrary units.

2.5. Histology

At the end of each experiment rats were deeply anesthetized andperfused intracardially. The perfusion started with a vascular rinsewith 0.9% saline until the liver was cleared (at a perfusion speedof 40 ml/min). The perfusion continued with 4% paraformaldehydesolution (300 ml at 40 ml/min, then 30 ml/min), and finally with10% sucrose solution in phosphate buffer (30 ml/min). The brainwas extracted en-bloc, cleared of meninges and blood vessels, andimmersed in 4% paraformaldehyde overnight, and then stored in30% sucrose (in 0.1 M PBS). Frozen brain tissue was cut in 40 �mthick sections in a transverse plane using a cryostat microtome(model Leica CM1900, Nussloch, Germany). To identify the loca-tion of microinjection sites the sections were stained using cresylviolet Nissl stain.

3. Results

3.1. Effects of vagotomy on cardiorespiratory responses toglutamate injection and serotonin infusion

Table 1 describes the mean ± SE values for breath by breathand beat by beat parameters of respiration and blood pressurebefore and after vagotomy. As expected, following vagotomy, TT,VT, heart rate and SBP all increased, although the change in SBPdid not achieve statistical significance (Table 1). Fig. 1 depicts typi-cal responses to glutamate injection and serotonin infusion beforeand after supra-nodose vagotomy. Microstimulation of the ITR byglutamate evoked immediate apnea of 3–10 s in duration, bothbefore (Fig. 1A) and after (Fig. 1B) vagotomy. Transient hyperten-sion appeared simultaneously with the apnea followed by briefhypotension within the first 30 s following glutamate injection(Fig. 1A). Similar transient hypertension was observed after vago-tomy, but ensuing hypotension was unusual (Fig. 1B).

The number of apneas following glutamate injection did not dif-fer in pre-vagotomy vs post-vagotomy trials (p = 0.1; Fisher ExactTest), nor did the average duration of evoked apnea (8.82 ± 1.15vs 9.16 ± 1.35; p = 0.84; Table 2). The duration of the initial hyper-tension evoked by glutamate also was unchanged by vagotomy(38.6 ± 5.82 vs 31.73 ± 5.27; p = 0.42; Table 2). These responses aretypical for glutamatergic stimulation of the ITR as we have previ-ously described (Topchiy et al., 2009).

Fig. 1C and D demonstrates the effects of serotonin infusionbefore and after supranodose vagotomy. Concomitant apnea andhypertension were observed before vagotomy (Fig. 1C). Followingvagotomy, serotonin infusion evoked brief hypertension, with orwithout ensuing transient hypotension, and apnea was blocked(Fig. 1D). In two cases post-vagotomy infusion of the serotoninevoked apneas, however, these were much shorter than the pairedapneas prior to vagotomy (Table 2).

3.2. Effects of vagotomy on cardiorespiratory coupling

Fig. 2 shows the coherence between respiration and SBP afterglutamate injections into the ITR and after intravenous serotonininfusions. Prior to vagotomy, ITR stimulation (PreVagX Glu) sig-

378 I. Topchiy et al. / Respiratory Physiology & Neurobiology 175 (2011) 375–382

Fig. 1. Examples of respiratory and blood pressure reactions evoked by glutamate injections into the ITR before the vagotomy (A, PreVagX Glu), after the vagotomy (B,PostVagX Glu) and after intravenous infusion of serotonin in pre-vagotomy (C, PreVagX 5HT) and post-vagotomy (D, PostVagX 5HT) periods.

Table 2Duration of apnea and BP reaction in pre-vagotomy and post-vagotomy conditions, following injection of glutamate into the ITR and intravenous infusion of serotonin.

Apnea length (s) BP reaction (s)

Glu 5-HT Glu 5-HT

PreVag X PostVag X PreVag X PostVag X PreVag X PostVag X PreVag X PostVag X

AVG 8.82 ± 1.15 9.16 ± 1.35 21.54 ± 7.46 4.18 ± 0.82 38.6 ± 5.82 31.73 ± 5.27 95.79 ± 24.2 59.52 ± 14.9N 9 9 6 2 9 9 6 6p-Value 0.84 0.06 0.42 0.03

p-Values were considered significant if ≤0.05.

Fig. 2. Group averages of coherence between systolic blood pressure and respiration calculated over 240 s intervals. Control (Ctrl) vs: baseline records before vagotomy(Base PreVagX) and after vagotomy (Base Post VagX). Control vs: injection of glutamate before vagotomy (PreVagX Glu) and infusion of serotonin before vagotomy (Pre-Va

agX 5HT). Base Post VagX vs: intravenous infusion of serotonin post-vagotomy (Post Vre group means ± SE. *Comparison to Ctrl, p ≤ 0.05. †Comparison to Base PreVagX, p ≤ 0.0

agX 5HT) and microinjection of glutamate post-vagotomy (PostVagX Glu). Results5.

I. Topchiy et al. / Respiratory Physiology & Neurobiology 175 (2011) 375–382 379

F requency range calculated over 240 s intervals. Control (Ctrl) vs: baseline records beforev of glutamate before vagotomy (PreVagX Glu) and infusion of serotonin before vagotomy( (PostVagX 5HT) and microinjection of glutamate post-vagotomy (PostVagX Glu). Resultsa

n0sv

(ldVt0

ipvt(p

flfvaVSspatv

3

tbt

ti

Fig. 4. (A) Histologic verification of the ITR apneic site. ITR: intertrigeminal region;

ig. 3. Group averages of relative spectral power density of SBP in the respiratory fagotomy (Base PreVagX) and after vagotomy (Base PostVagX). Control vs: injectionPreVagX 5HT). Base PostVagX vs: intravenous infusion of serotonin post-vagotomyre group means ± SE. †Comparison to Base PostVagX, p ≤ 0.05.

ificantly reduced the coupling between respiration and SBP (Ctrl:.95 ± 0.01 vs PreVagX Glu: 0.89 ± 0.02, p = 0.01). Intravenous infu-ion of 5-HT also reduced cardiorespiratory coupling prior toagotomy (Ctrl: 0.95 ± 0.01 vs PreVagX 5HT: 0.72 ± 0.06 p = 0.04).

Coherence immediately prior (240 s) to vagotomyBase PreVagX) did not differ significantly from the controlevel (Ctrl: 0.95 ± 0.01 vs Base PreVagX: 0.95 ± 0.02; p = 0.71),emonstrating the stability of the experimental preparation.agotomy per se produced a slight decrease in cardiorespira-

ory coupling (Base PreVagX vs Base PostVagX: 0.95 ± 0.02 vs.89 ± 0.03; p = 0.03) (Fig. 2).

In contrast with pre-vagotomy conditions, post-vagotomynjection of glutamate into the ITR did not change cardiores-iratory coherence (Base PostVagX vs PostVagX Glu: 0.89 ± 0.03s 0.86 ± 0.03; p = 0.63). Infusion of serotonin following vago-omy also had no significant effect on cardiorespiratory couplingBase PostVagX vs PostVagX 5HT: 0.89 ± 0.03 vs 0.88 ± 0.02;= 0.98).

Fig. 3 reveals that the relative spectral power density of SBPuctuations in the respiratory frequency range did not change

ollowing injection of glutamate into the ITR (Ctrl: 0.26 ± 0.06s PreVagX Glu: 0.27 ± 0.04, p = 0.91). Administration of serotoninlso did not produce significant changes (Ctrl: 0.26 ± 0.06 vs Pre-agX 5HT: 0.09 ± 0.04, p = 0.21) in respiratory frequency power inBP. Vagotomy per se did not alter relative spectral power den-ity (Base PreVagX vs Base PostVagX: 0.29 ± 0.07 vs 0.25 ± 0.08;= 0.7). However, glutamate injection following vagotomy initiatedsignificant increase of the relative spectral power density of BP in

he respiratory range (Base PostVagX vs PostVagX Glu: 0.25 ± 0.08s 0.55 ± 0.06; p = 0.01).

.3. Histological localization of injection sites

Histological analysis of the injection sites confirmed the loca-ion of the functionally identified apneic points within anatomical

oundaries of the ITR (9.0–9.72 mm posterior to bregma) betweenhe principal sensory and motor trigeminal nuclei.

Fig. 4A provides a photomicrographic illustration of an ITR injec-ion site with the micropipette trace. A schematic map of the ITRnjection sites is presented in Fig. 4B.

Mo5: motor trigeminal nucleus; Pr5: principal sensory trigeminal nucleus. (B)Schematic diagram of the histologically verified injections into apneic ITR sites. Thearrow indicates the injection site for the recordings, shown in Fig. 1.

In summary, stimulation of the ITR by glutamate perturbed both

respiratory and cardiovascular patterns and decreased coherencebetween SBP and respiration. Infusion of serotonin also disturbedcardiorespiratory homeostasis by decreasing coupling (decrease ofSBP-respiratory coherence). Supranodose vagotomy altered res-

3 logy &

pscIet

4

dblIwseitvlif

tcEA22swetis

htwtm2kiiRLmpGPmawtetep

sipo

80 I. Topchiy et al. / Respiratory Physio

iratory and blood pressure patterns, and produced a slight, buttatistically significant, disruption in cardiorespiratory coupling. Inontrast to pre-vagotomy conditions, glutamate stimulation of theTR following vagotomy did not change cardiorespiratory coher-nce, and in fact increased the relative spectral power density ofhe BP signal in the respiratory frequency range.

. Discussion

The present study demonstrated that under anesthetized con-itions, pharmacologic transient perturbation of respiration andlood pressure by either central (ITR) or peripheral (5-HT) manipu-

ations is associated with reduced coupling between these systems.n addition, we demonstrated a significant impact of vagal path-

ays on cardiorespiratory coupling. Although bilateral vagotomylightly reduced the coherence between respiration and SBP, it alsoliminated the larger reductions of cardiorespiratory coherencen the face of perturbations induced by pharmacologic manipula-ions of either the central or peripheral nervous system. Moreover,agotomy actually increased the magnitude of respiratory modu-ation in SBP following ITR glutamate injections, which is reflectedn the increased spectral power density of SBP in the respiratoryrequency range.

The modulatory influences of pontine structures on respira-ory pattern and rhythm generation, heart rate, blood pressure andardiorespiratory coupling are well recognized (Cohen, 1971; vonuler, 1977; Oku and Dick, 1992; Paton, 1998; Dick and Coles, 2000;lheid et al., 2004; Dutschmann and Herbert, 2006; Smith et al.,007; Baekey et al., 2008; Padley et al., 2007; Segers et al., 1985,008). It was shown that pontine transection eliminates respiratoryinus arrhythmia, respiratory modulation of BP (Traube-Heringaves) and postinspiratory discharges from vagal efferents (Baekey

t al., 2008). Thus, central pontine influences represent key modula-ors of cardiorespiratory coupling, including coherent fluctuationsn respiration and blood pressure as examined in the presenttudy.

The influence of the pontine ITR on cardiorespiratory outputsas been previously demonstrated (Topchiy et al., 2009). Respira-ory and blood pressure perturbations following ITR microinjectionith glutamate may be a direct result of activating efferent projec-

ions to the respiratory and cardiovascular sites of the ventrolateraledulla (VLM) (Chamberlin and Saper, 1992, 1998; Chamberlin,

004). The precise synaptic targets of ITR projections are notnown, but most probably include respiratory and cardiovascularntegrating and premotor neurons of the VLM. Integrated activ-ty of the VLM respiratory and cardiovascular neurons (Miura andeis, 1972; Koepchen et al., 1981; Connelly and Wurster, 1985;oewy and Spyer, 1990) may influence respiratory pattern andotor outputs via the phrenic nucleus and cardiovascular out-

uts via descending branches of the vagus (Cohen et al., 1976;ilbey et al., 1984; Bianchi and St. John, 1982; Bianchi et al., 1995;ilowsky, 1995; Alheid et al., 2004). ITR influences also may beediated via VLM projections to the mediothoracic spinal cord

rea which in turn sends terminals to the heart and vasculature,hich are an important component of vasomotor regulation. Fur-

her, ITR projects to the medullary raphe (Bernard, 1998; Verbernet al., 1999; Verner et al., 2008), a region interconnected with mul-iple respiratory and cardiovascular regulatory structures (Lindseyt al., 1998) as well as with other pontine areas (Kölliker-Fuse,arabrachial area).

Because the ITR is functionally embedded within the brain-tem homeostatic networks for respiration and blood pressure, its not surprising that glutamatergic stimulation of the ITR evokes arominent cardiorespiratory disturbance. The transient decreasef cardiorespiratory coupling after glutamate stimulation of the

Neurobiology 175 (2011) 375–382

ITR could be explained by independent activation of respiratoryand cardiovascular pathways that interferes with the ongoing cou-pling of these two systems. The notion of independently activatedpathways is consistent with our observation that the blood pres-sure reaction to ITR stimulation typically outlasted the apneicresponse.

Assuming SBP-respiratory coherence depended strictly on cen-tral pathways, vagotomy would not be expected to significantlyalter the impact of ITR stimulation on coherence. The fact thatvagotomy blocked the decrease in coherence without reducing themagnitude of the disturbance itself instead argues that the pri-mary source of SBP-respiratory coherence is central, but that vagalfeedback processes participating in the integrative response to ITRstimulation operate in a manner that reduces cardiorespiratorycoherence.

It was previously shown (Topchiy et al., 2009), that in addi-tion to the immediate effects, stimulation of the ITR by glutamateevoked a delayed increase of SBP. This delayed response is consis-tent with a possible activation of vasomotor sympathetic outputsfollowing ITR stimulation; responses that build more slowly thando vagal outputs. Thus, the delayed increase of BP could be causedby projections from the ITR to the pressor region of the VLMand subsequently activation of sympathetic premotor neurons,driving vasomotor activity (Preiss et al., 1975; Ross et al., 1984;Reis et al., 1984). The timing of the SBP response fits with thetime constant for a sympathetically mediated pressor response.For example, the average time required for the sympathetic pres-sor response to a cold stimulus is approximately 15 s, whereasthe average time for the chronotropic (cardiac vagal) parasym-pathetic response is only about 5 s (Khurana, 2007). The delayedSBP response is prolonged following vagotomy, suggesting theimportance of baroreceptor feedback from the aortic arch recep-tors in limiting the hypertension. It is also possible that delayedincreases in sympathetic activity could be driven by post-apneichypoxia in addition to or instead of by ITR projections to the VLM.In either case, the net effect of the vagus nerves on responses toITR stimulation appears to be homeostatic in limiting the bloodpressure transient responses, but disruptive to SBP-respiratorycoherence.

The negative impact of the vagus nerve on SBP-respiratorycoupling is further supported by the observation that serotonininfusion increased blood pressure and reduced coherence to a muchgreater extent than did ITR stimulation. Further, the decrease ofthe SBP-respiratory coherence following the serotonin infusion waseliminated by vagotomy, despite the fact that the blood pressureresponse was intact. Stimulation of both vagal afferent nerve end-ings and cell bodies within the nodose ganglia can be producedby intravascular infusion of serotonin (Yoshioka et al., 1992a,b).Since intravenous serotonin does not penetrate the blood–brainbarrier, the nodose ganglia are considered to be a primary targetof its action. This view is supported by the observation that theapneic response to serotonin was virtually abolished after bilat-eral supranodose vagotomy in the present and previous studies(Jacobs and Comroe, 1971; Segers et al., 1985; Yoshioka et al.,1992a; Kopczynska and Szereda-Przestaszewska, 2003).

Intravenous infusion of serotonin (prior to vagotomy) producedprolonged apnea and prominent fluctuation of blood pressure (inmost cases hypertension), which either followed or occurred simul-taneously with the apnea. A similar effect was described in dogsby Schneider and Yonkman (1954). Prior to vagotomy, infusion ofserotonin evoked a large decrease of SBP-respiratory coherence.

Following vagotomy, despite the persistence of a large blood pres-sure disturbance, serotonin infusion did not alter cardiorespiratorycoupling. Again, it suggests that vagal activity during transienthomeostatic disturbances acts to reduce cardiorespiratory coher-ence.

logy &

hinbvlarMsa

easarretpBaaRHv

saiaocst

idtismfidottptpsiimacod

rctbh

I. Topchiy et al. / Respiratory Physio

Although not the primary source of the apneic response, theypertension evoked by infusion of 5-HT may have exerted some

mpact on the overall respiratory response via baroreflex mecha-isms. Anatomically, information from baroreceptors is transferredoth via the glossopharingeal nerve (from the carotid sinus) andia the vagus nerve (from the aortic arch). However, in the rat theatter is the main source of both baroreceptor and chemoreceptorfferentation. Supranodose vagotomy removes the cardiac/aorticeflex pathway, which is carried via the vagus nerve (Krieger andarseillan, 1963; Faber and Brody, 1983) – and may therefore have

ignificantly reduced any baroreflex mediated contribution to thepnea.

However, examination of the timing of hypertension and apneavoked by the infusion of 5-HT revealed that many (but not all)pneas preceded observable hypertension and others appearedimultaneously with blood pressure changes. This fact arguesgainst hypertension as the primary cause of apnea but does notule out a contribution of baroreflex activation on the apneicesponse. In addition, pharmacologic hypertension evoked, forxample, by infusion of phenylephrine, can prolong expiratoryime, but does not commonly evoke apnea. Further, activation ofulmonary vagal stretch receptors provokes apnea via the Hering-reuer reflex (Younes et al., 1974; Carley et al., 2004) withoutctivation of baroreceptors and 5-HT can elicit dose-dependentpnea via activation of vagal afferents (Yoshioka et al., 1992a,b;adulovacki et al., 2007). Together, these facts suggest that the 5-T evoked apneic response reflects activation of non-baroreceptoragal afferents.

We thus conclude that respiration and blood pressure aretrongly coherent under baseline conditions in anesthetized ratsnd that this coherence is reduced in the face of pharmacologicallynduced disturbance in intact anesthetized animals. Conversely,s it was shown in the present study, bilateral vagotomy exertsnly minor effects on cardiorespiratory coherence in equilibriumonditions, but eliminates the loss of coherence induced by bothtimulation of the pontine ITR and of peripheral serotonin recep-ors.

In support of our data, considerable experimental evidencendicates that most peripheral and central inputs affect both car-iovascular and respiratory integration and motor outputs, ratherhan either system alone, allowing speculation as to the phys-ological importance of cardiorespiratory coupling. It has beenuggested for example, that cardiorespiratory coupling may opti-ize the performance of the cardiac pump by matching cardiac

lling to venous return, and that this may be especially importanturing sleep or anesthesia (Galletly and Larsen, 1997, 1998). Ourbservation of high respiratory-SBP coherence at baseline in anes-hetized rats is consistent with this view. However, in the face ofransient perturbations to cardiorespiratory equilibrium, it may behysiologically advantageous for the cardiovascular and respira-ory control systems to act more independently, as observed in theresent experiments. Such independent action may, for example,horten the time required for return to equilibrium. In particular,n the context of prolonged expiration, as induced by ITR glutamatenjection or serotonin infusion in this study, it may be deleterious to

aintain strong coupling between respiratory pattern generationnd cardiac vagal outputs, as this may produce profound brady-ardia (Spyer, 1991). This also is in keeping with the observationf decreased cardiorespiratory coupling in sleep related breathingisorders such as sleep apnea syndrome (Kabir et al., 2010).

Taken together, our observations suggest that coupling between

espiration and blood pressure is largely mediated within theentral nervous system, with the vagal system acting in a wayhat disrupts coherence during transient cardiorespiratory distur-ances, but that may be physiologically advantageous in restoringomeostatic equilibrium of respiration and blood pressure.

Neurobiology 175 (2011) 375–382 381

Disclosure statement

The authors: Irina Topchiy, Jonathan Waxman, Miodrag Radulo-vacki and David W Carley do not have any actual or potentialconflict of interest including any financial, personal or other rela-tionships with other people or organizations within three (3) yearsof beginning the work submitted that could inappropriately influ-ence this work.

Acknowledgements

Authors thank Milka Dokic for the experimental support.The work was supported by NIH Grants HL070870 and

AG016303.

References

Alheid, G.F., Milsom, W.K., McCrimmon, D.R., 2004. Pontine influences on breathing:an overview. Respir. Physiol. Neurobiol. 143, 105–114.

Baekey, D.M., Dick, T.E., Paton, J., 2008. Pontomedullary transection attenuatescentral respiratory modulation of sympathetic discharge, heart rate and thebaroreceptor reflex in the in situ rat preparation. Exp. Physiol. 93, 803–816.

Bernard, D.G., 1998. Cardiorespiratory responses to glutamate microinjected intothe medullary raphé. Respir. Physiol. 113, 11–21.

Bianchi, A.L., St. John, W.M., 1982. Medullary axonal projections of respiratory neu-rons of pontine pneumotaxic center. Respir. Physiol. 48, 357–373.

Bianchi, A.L., Denavit-Saubié, M., Champagnat, J., 1995. Central control of breathingin mammals: neuronal circuitry, membrane properties, and neurotransmitters.Physiol. Rev. 75, 1–45.

Carley, D.W., Pavlovic, S., Malis, M., Knezevic, N., Saponjic, J., Li, C., Radulovacki, M.,2004. C-fiber activation exacerbates sleep-disordered breathing in rats. SleepBreath. 8, 147–154.

Chamberlin, N.L., Saper, C.B., 1992. Topographic organization of cardiovascularresponses to electrical and glutamate microstimulation of the parabrachialnucleus in the rat. J. Comp. Neurol. 326, 245–262.

Chamberlin, N.L., Saper, C.B., 1998. A brainstem network mediating apneic reflexesin the rat. J. Neurosci. 18, 6048–6056.

Chamberlin, N.L., Saper, C.B., 2003. Functional anatomical analysis of respiratorycircuitry. In: Carley, D.W., Radulovacki, M. (Eds.), Sleep-related Breathing Disor-ders. Marcel Dekker, Basel, pp. 93–102.

Chamberlin, N.L., 2004. Functional organization of the parabrachial complex andintertrigeminal region in the control of breathing. Respir. Physiol. Neurobiol.143, 115–125.

Cohen, M.I., 1971. Switching of the respiratory phases and evoked phrenic responsesproduced by rostral pontine electrical stimulation. J. Physiol. 217, 133–158.

Cohen, M.I., Piercey, M.F., Gootman, P.M., Wolotsky, P., 1976. Respiratory rhythmic-ity in the cat. Fed. Proc. 35, 1967–1974.

Connelly, C.A., Wurster, R.D., 1985. Sympathetic rhythms during hyperventilation-induced apnea. Am. J. Physiol. 249, R424–R431.

Dick, T.E., Baekey, D.M., Paton, J.F., Lindsey, B.G., Morris, K.F., 2009. Cardio-respiratorycoupling depends on the pons. Respir. Physiol. Neurobiol. 168, 76–85.

Dick, T.E., Coles, S.K., 2000. Ventrolateral pons mediates short-term depression ofrespiratory frequency after brief hypoxia. Respir. Physiol. 121, 87–100.

Dutschmann, M., Herbert, H., 2006. The Kölliker-Fuse nucleus gates the postinspi-ratory phase of the respiratory cycle to control inspiratory off-switch and upperairway resistance in rat. Eur. J. Neurosci. 24, 1071–1084.

Eckberg, D.L., 1983. Human sinus arrhythmia as an index of vagal cardiac outflow. J.Appl. Physiol. 54, 961–966.

Eckberg, D.L., Sleight, P., 1992. Human Baroreflexes in Health and Disease. ClarendonPress, Oxford.

Faber, J.E., Brody, M.J., 1983. Reflex hemodynamic response to superior laryngealnerve stimulation in the rat. J. Auton. Nerv. Syst. 9, 607–622.

Felder, R.B., Mifflin, S.W., 1988. Modulation of carotid sinus afferent input to nucleustractus solitarius by parabrachial nucleus stimulation. Circ. Res. 63, 35–49.

Feldman, J.L., Ellenberger, H.H., 1988. Central coordination of respiratory and car-diovascular control in mammals. Annu. Rev. Physiol. 50, 593–606.

Galletly, D.C., Larsen, P.D., 1997. Cardioventilatory coupling during anaesthesia. Br.J. Anaesth. 79, 35–40.

Galletly, D.C., Larsen, P.D., 1998. Relationship between cardioventilatory couplingand respiratory sinus arrhythmia. Br. J. Anaesth. 80, 164–168.

Gilbey, M.P., Jordan, D., Richter, D.W., Spyer, K.M., 1984. Synaptic mechanismsinvolved in the inspiratory modulation of vagal cardio-inhibitory neurones inthe cat. J. Physiol. 356, 65–78.

Jacobs, L., Comroe Jr., J.H., 1971. Reflex apnea, bradycardia, and hypertension pro-duced by serotonin and phenylbiguanide acting on the nodose ganglia of the

cat. Circ. Res. 29, 145–155.

Kabir, M.M., Dimitri, H., Sanders, P., Antic, R., Nalivaiko, E., Abbott, D., Baumert, M.,2010. Cardiorespiratory phase-coupling is reduced in patients with obstructivesleep apnea. PLoS ONE 5, e10602.

Kalia, M., Mesulam, M.M., 1980. Brain stem projections of sensory and motorcomponents of the vagus complex in the cat. II. Laryngeal, tracheobronchial,

3 logy &

K

KK

K

K

L

L

M

M

M

M

M

M

N

N

N

O

P

P

P

82 I. Topchiy et al. / Respiratory Physio

pulmonary, cardiac, and gastrointestinal branches. J. Comp. Neurol. 193, 467–508.

alia, M., Sullivan, J.M., 1982. Brainstem projections of sensory and motor compo-nents of the vagus nerve in the rat. J. Comp. Neurol. 211, 248–265.

hurana, R.K., 2007. Cold face test: adrenergic phase. Clin. Auton. Res. 17, 211–216.oepchen, H.P., Klüssendorf, D., Sommer, D., 1981. Neurophysiological background

of central neural cardiovascular-respiratory coordination: basic remarks andexperimental approach. J. Auton. Nerv. Syst. 3, 335–368.

opczynska, B., Szereda-Przestaszewska, M., 2003. Supranodose vagotomy pre-cludes reflex respiratory responses to serotonin in cats. J. Biomed. Sci. 10,718–724.

rieger, E.M., Marseillan, R.F., 1963. Aortic depressor fibers in the rat: an electro-physiological study. Am. J. Physiol. 205, 771–774.

indsey, B.G., Arata, A., Morris, K.F., Hernandez, Y.M., Shannon, R., 1998. Medullaryraphe neurones and baroreceptor modulation of the respiratory motor patternin the cat. J. Physiol. 512, 863–882.

oewy, A.D., Spyer, K.M. (Eds.), 1990. Central Regulation of Autonomic Functions.Oxford University Press.

andel, D.A., Schreihofer, A.M., 2006. Central respiratory modulation of barosensi-tive neurones in rat caudal ventrolateral medulla. J. Physiol. 572, 881–896.

cAllen, R.M., Spyer, K.M., 1975. The origin of cardiac vagal efferent neurones in themedulla of the cat. J. Physiol. 244, 82P–83P.

cAllen, R.M., Spyer, K.M., 1976. The location of cardiac vagal preganglionicmotoneurones in the medulla of the cat. J. Physiol. 258, 187–204.

ifflin, S.W., Felder, R.B., 1990. Synaptic mechanisms regulating cardiovascularafferent inputs to solitary tract nucleus. Am. J. Physiol. Heart Circ. Physiol. 259,H653–H661.

iura, M., Reis, D.J., 1972. The role of the solitary and paramedian reticular nucleiin mediating cardiovascular reflex responses from carotid baro- and chemore-ceptors. J. Physiol. 223, 525–548.

orrison, S.F., Cravo, S.L., Wilfehrt, H.M., 1994. Pontine lesions produce apneusis inthe rat. Brain Res. 652, 83–86.

eff, R.A., Wang, J., Baxi, S., Evans, C., Mendelowitz, D., 2003. Respiratory sinusarrhythmia: endogenous activation of nicotinic receptors mediates respiratorymodulation of brainstem cardioinhibitory parasympathetic neurons. Circ. Res.93, 565–572.

icholson, C., 1985. Diffusion from an injected volume of a substance inbrain tissue with arbitrary volume fraction and tortuosity. Brain Res. 333,325–329.

unez-Abades, P.A., Portillo, F., Pasaro, R., 1990. Characterisation of afferent projec-tions to the nucleus ambiguus of the rat by means of fluorescent double labelling.J. Anat. 172, 1–15.

ku, Y., Dick, T.E., 1992. Phase resetting of the respiratory cycle before and afterunilateral pontine lesion in cat. J. Appl. Physiol. 72, 721–730.

adley, J.R., Kumar, N.N., Li, Q., Nguyen, T.B., Pilowsky, P.M., Goodchild, A.K., 2007.Central command regulation of circulatory function mediated by descendingpontine cholinergic inputs to sympathoexcitatory rostral ventrolateral medullaneurons. Circ. Res. 100, 284–291.

aton, J.F., 1998. Pattern of cardiorespiratory afferent convergence to solitary tractneurons driven by pulmonary vagal C-fiber stimulation in the mouse. J. Neuro-physiol. 79, 2365–2373.

aton, J.F., Silva-Carvalho, L., Thompson, C.S., Spyer, K.M., 1990. Nucleus tractussolitarius as mediator of evoked parabrachial cardiovascular responses in thedecerebrate rabbit. J. Physiol. 428, 693–705.

Neurobiology 175 (2011) 375–382

Paxinos, G., Watson, C., 2004. The Rat Brain in Stereotaxic Coordinates. AcademicPress, San Diego.

Pilowsky, P., 1995. Good vibrations? Respiratory rhythms in the central control ofblood pressure. Clin. Exp. Pharmacol. Physiol. 22, 594–604.

Preiss, G., Kirchner, F., Polosa, C., 1975. Patterning of sympathetic preganglionicneuron firing by the central respiratory drive. Brain Res. 87, 363–374.

Radulovacki, M., Stoiljkovic, M., Saponjic, J., Carley, D.W., 2007. Effects of inter-trigeminal region NMDA and non-NMDA receptors on respiratory responses inrats. Respir. Physiol. Neurobiol. 156, 40–46.

Radulovacki, M., Pavlovic, S., Saponjic, J., Carley, D.W., 2003a. Intertrigeminal regionattenuates reflex apnea and stabilizes respiratory pattern in rats. Brain Res. 975,66–72.

Radulovacki, M., Pavlovic, S., Saponjic, J., Carley, D.W., 2003b. Pontine intertrigeminalregion attenuates sleep apneas in rats. Sleep 27, 383–387.

Reis, D.J., Ross, C.A., Ruggiero, D.A., Granata, A.R., Joh, T.H., 1984. Role of adrenalineneurons of ventrolateral medulla (the C1 group) in the tonic and phasic controlof arterial pressure. Respir. Physiol. Neurobiol. 158, 39–44.

Ross, C.A., Ruggiero, D.A., John, T.H., Park, D.H., Reis, D.J., 1984. Rostral ventrolateralmedulla: selective projections to the thoracic autonomic cell column from theregion, containing C1 adrenaline neurons. J. Comp. Neurol. 228, 168–185.

Schneider, J.A., Yonkman, F.F., 1954. Species differences in the respiratory and car-diovascular response to serotonin (5-hydroxytryptamine). J. Pharmacol. Exp.Ther. 111, 84–98.

Segers, L.S., Shannon, R., Lindsey, B.G., 1985. Interactions between rostral pontineand ventral medullary respiratory neurons. J. Neurophysiol. 54, 318–334.

Segers, L.S., Nuding, S.C., Dick, T.E., Shannon, R., Baekey, D.M., Solomon, I.C., Morris,K.F., Lindsey, B.G., 2008. Functional connectivity in the pontomedullary respira-tory network. J. Neurophysiol. 100, 1749–1769.

Smith, J.C., Abdala, A.P., Koizumi, H., Rybak, I.A., Paton, J.F., 2007. Spatial andfunctional architecture of the mammalian brain stem respiratory network:a hierarchy of three oscillatory mechanisms. J. Neurophysiol. 98, 3370–3387.

Spyer, K.M., 1991. Functional organization of cardiorespiratory control. In: Gaultier,C., Escourrou, P., Cuni-Oascalva, L. (Eds.), Sleep and Cardiorespiratory Control.John Libbcy Euroteirt Ltd., pp. 3–8.

Topchiy, I., Radulovacki, M., Waxman, J., Carley, D.W., 2009. Cardiorespiratoryeffects of intertrigeminal area stimulation in vagotomized rats. Brain Res. 1250,120–129.

Verberne, A.J., Sartor, D.M., Berke, A., 1999. Midline medullary depressor responsesare mediated by inhibition of RVLM sympathoexcitatory neurons in rats. Am. J.Physiol. 276, R1054–R1062.

Verner, T.A., Pilowsky, P.M., Goodchild, A.K., 2008. Retrograde projections to a dis-crete apneic site in the midline medulla oblongata of the rat. Brain Res. 1208,128–136.

von Euler, C., 1977. The functional organization of the respiratory phase-switchingmechanisms. Fed. Proc. 36, 2375–2380.

Younes, M., Vaillancourt, P., Milic-Emili, J., 1974. Interaction between chemical fac-tors and duration of apnea following lung inflation. J. Appl. Physiol. 36, 190–201.

Yoshioka, M., Goda, Y., Togashi, H., Matsumoto, M., Saito, H., 1992a. Pharmacologicalcharacterization of 5-hydroxytryptamine-induced apnea in the rat. J. Pharmacol.Exp. Ther. 260, 917–924.

Yoshioka, M., Ikeda, T., Abe, M., Togashi, H., Minami, M., Saito, H., 1992b. Pharmaco-logical characterization of 5-hydroxytryptamine-induced excitation of afferentcervical vagus nerve in anaesthetized rats. Br. J. Pharmacol. 106, 544–549.