Central Command Increases Sympathetic Nerve Activity During ...

66

Central Command Increases SympatheticNerve Activity During Spontaneous

Locomotion in CatsGeorge Hajduczok, Jon S. Hade, Allyn L. Mark, John L. Williams, and Robert B. Felder

A controversial issue in exercise physiology is the relative contribution of central commandversus afferent input from contracting muscles and baroreceptors in the regulation ofsympathetic nerve activity (SNA) during exercise. Recent studies of exercising humans havesuggested that central command increases cutaneous sympathetic sudomotor nerve activity buthave challenged the concept that central command contributes importantly to increases insympathetic vasoconstrictor nerve activity to skin and skeletal muscle. The purpose of thisstudy was to examine the influence of central command on renal SNA and lumbar SNA duringspontaneous locomotion in decorticate cats. Unanesthetized decorticate cats that developedlocomotion spontaneously or during electrical stimulation of the subthalamic locomotor regionwere studied in the presence and absence of input from skeletal muscle and baroreceptorafferents. Spontaneous rhythmic locomotion in the unparalyzed state was associated withsignificant increases in mean arterial pressure (MAP) from 106±10 to 133±11 mm Hg(p<0.05) and increases in renal SNA of 301±100% (p<O.OS). During spontaneous fictiverhythmic locomotion in paralyzed cats, there were also significant (p<0.05) increases in MAP(43+6%), renal SNA (183+32%), and lumbar SNA (223+±83%). Baroreceptor denervation didnot attenuate increases in MAP, renal SNA, and lumbar SNA during locomotion. Duringelectrical stimulation of the subthalamic locomotor region in paralyzed cats, MAP increased by43±17% (p<0.05), and renal SNA increased by 175+±47% (p<0.05). These findings indicatethat central command is capable of increasing sympathetic neural drive in unanesthetizeddecorticate cats. This increase in sympathetic drive occurs even in the absence of feedback fromcontracting muscles or from arterial and cardiopulmonary baroreceptors. (Circulation Research1991;69:66-75)

I ncreases in heart rate, myocardial contractility,arterial pressure, and ventilation occur with theonset of exercise.1-3 Peripheral feedback and

central command are mechanisms that have tradi-tionally been proposed to account for the autonomiccardiovascular and respiratory changes associatedwith muscular exercise.4The concept of central command invokes a feed-

forward mechanism by which activation of cardiovas-cular and respiratory centers is accomplished by

From the Department of Internal Medicine and the Cardiovas-cular Center, University of Iowa College of Medicine and VAMedical Center, Iowa City, Iowa.

Supported by the American Heart Association, Iowa Affiliate(G.H.), National Institutes of Health grants HL-36224 (A.L.M.),HL-25488 (A.L.M.), HL-29302 (R.B.F.), and HL-14388 (R.B.F.),and research funds from the VA Medical Center (A.L.M.). R.B.F.was an Established Investigator of the American Heart Associa-tion at the time these studies were performed.Address for correspondence: Robert B. Felder, MD, Cardiovas-

cular Division, Department of Internal Medicine, University ofIowa College of Medicine, Iowa City, IA 52242.

descending signals from the suprapontine regionsthat initiate somatomotor activity.5-7 Chemical orelectrical stimulation of the subthalamic locomotorregion (STLR) in the hypothalamus in animals pro-duces both locomotion and increases in cardiopul-monary function,8- 1 which simulate the responses tovoluntary exercise in humans. Similar findings havebeen reported during spontaneous locomotion inunanesthetized decorticate cats.8-10Although these studies5-11 have provided support

for a role of central command in the cardiorespira-tory adjustments to exercise, an area of recent con-troversy has been the role of central command in thesympathetic nerve responses to exercise. This contro-versy originated several years ago from studies inhumans involving direct intraneural recordings ofsympathetic nerve activity during exercise.'2 Theseand subsequent studies13-15 demonstrated that cen-tral command increases heart rate but elicits no oronly modest increases in muscle sympathetic nerveactivity in humans. A recent study'6 suggested that

by guest on February 14, 2018http://circres.ahajournals.org/

Dow

nloaded from

http://circres.ahajournals.org/

Hajduczok et al Central Command Increases Sympathetic Nerve Activity 67

central command increases skin sympathetic activityin humans, but this increase consisted of sympatheticsudomotor activity and not vasoconstrictor activity.Thus, studies in humans in the past six years havefailed to demonstrate that central command plays amajor role in regulation of sympathetic neural vaso-constrictor activity in humans.

Sympathetic nerve activity has been recorded re-cently in response to posterior hypothalamic stimu-lation in cats17 and rats.18 Microinjection of they-aminobutyric acid (GABA) antagonist picrotoxininto the posterior hypothalamus produced a markedincrease in cervical17 and splanchnic'8 sympatheticnerve activity along with increases in arterial pressureand heart rate. The site of injection of picrotoxin inthe study performed on cats17 was the same as thatdescribed for the location of the STLR,8-10 butlocomotor activity was not documented. It cannot beconcluded from this study that the sympathetic nerveresponses emanated from activation of the locomo-tion neurons in the STLR as opposed to otherneurons in the posterior hypothalamus. In addition,it should be noted that there have been no reports ofsympathetic nerve responses to spontaneous fictivelocomotion in animals. Thus, the role of centralcommand in regulation of sympathetic nerve activitywith animals has not been definitively addressed.

Since the relative contribution of central commandand peripheral feedback to the cardiovascular re-sponse to exercise remains controversial, we soughtto determine if central command alone can increasesympathetic nerve activity during spontaneous loco-motion and fictive locomotion in decorticate cats.Cats that developed locomotion spontaneously orduring electrical stimulation of the STLR were stud-ied in the presence and absence of input fromskeletal muscle and baroreceptor afferents.

Materials and MethodsGeneral Preparation

Experiments were performed on cats weighingbetween 2.2 and 5.8 kg. Anesthesia was induced withhalothane (4%). The trachea was then cannulated,and the cat was connected to a respirator (model 611,Harvard Apparatus, South Natick, Mass.) for con-trolled ventilation. Anesthesia was maintained with2.5% halothane. The left femoral vein and externalcarotid artery were cannulated for injection of drugsand for measurement of arterial pressure, respec-tively. The carotid sinus nerves and vagosympathetictrunks were isolated bilaterally and surrounded withligatures for later sectioning. Body temperature wasmaintained at 37-38°C with a heating pad and heatlamp. Arterial blood gases were maintained withinthe normal range by adjusting the ventilator or byadministering sodium bicarbonate.The cats were placed in a stereotaxic device (David

Kopf Instruments, Tujunga, Calif.) with the headflexed ventrally. The spinal processes of the lowercervical (C-7 through T-2) and lumbar (L-4 and L-5)

regions were exposed and clamped, and the cats weresuspended above the table to allow the limbs to movefreely during locomotion. After ligation of the re-maining external carotid artery, the cranium wasremoved, and decortication was performed accordingto the method described by Eldridge and col-leagues.10 Briefly, the brain was sectioned trans-versely -10-12 mm anterior to the superior colliculi.The brain rostral to this section was removed bysuction, and bleeding was controlled with Gelfoam(The Upjohn Co., Kalamazoo, Mich.). After decorti-cation, no additional anesthesia was required. Thecats were disconnected from the ventilator when theybegan to breath spontaneously.Nerve recordings. Efferent sympathetic nerve activ-

ity was recorded from renal and/or lumbar sympa-thetic nerves (L-2 or L-3) that were exposed andisolated via a left lateral flank incision. After cuttingthe nerve, the sheath was removed from the proximalend, and the nerve was placed on bipolar platinumelectrodes and covered with a mixture of paraffin oiland vacuum grease to avoid desiccation. In theunparalyzed cats, the nerves were secured to theelectrodes with dental epoxy (Reprosil, DentsplyInternational, Milford, Del.) to prevent nerve dam-age and minimize motion artifact during locomotion.The multiple fiber nerve signals were passed througha high-impedance probe (model HIP 511E, GrassInstrument Co., Quincy, Mass.) and band-pass am-plifier (model P511K, Grass Instrument), and nerveactivity was monitored on a storage oscilloscope(model 5115, Tektronix Inc., Beaverton, Ore.) and anaudioamplifier (model AM8, Grass Instrument).Hind limb motor nerve activity was monitored in

the paralyzed cats to permit recognition of spontane-ous central locomotor drive. The biceps femorisnerves were exposed bilaterally in the upper thighand placed on bipolar recording electrodes. Thetechniques used for recording and quantitating mo-tor nerve activity were similar to those describedabove for sympathetic nerve recording. The motornerves were identified before the recording sessionby electrically evoking contractions of the bicepsfemoris muscle during single-pulse electrical stimu-lation. The absence of sympathetic nerve activity inthe motor nerve recordings was determined in somecats (n =5) by the failure to reflexly activate or inhibitnerve activity in response to bolus injections ofnitroglycerin or phenylephrine, respectively.

In five studies, phrenic nerve activity was recordedas an index of central respiratory drive in paralyzedcats. Phrenic nerve recordings were obtained fromthe C-5 root of the phrenic nerve via a right lateralcervical approach as previously described.'9

Electromyographic recordings. Spontaneous loco-motion in the unparalyzed condition was monitoredby recording electromyographic (EMG) activity fromthe hind limb biceps femoris muscles. Hook elec-trodes were placed in the exposed muscles, and thesignals were amplified and recorded in a mannersimilar to that described above.


Dow

nloaded from


68 Circulation Research Vol 69, No 1 July 1991

ProtocolsCats decorticated rostral to the STLR began to

locomote spontaneously and intermittently 30-60minutes after cessation of anesthesia. When the catswere hemodynamically stable and spontaneously lo-comoting, arterial pressure, heart rate, renal and/orlumbar sympathetic nerve activity, and, in some cats,phrenic nerve activity were recorded under the fol-lowing experimental conditions.

Protocol 1: Spontaneous locomotion with peripheralfeedback intact. In six cats, the cardiovascular andsympathetic nerve responses to spontaneously occur-ring rhythmic, coordinated movement of all fourlimbs were examined. This activity was presumed tooriginate from the STLR, since mesencephalic cats(i.e., those with STLR removed) do not developspontaneous locomotion.10 Quadriceps muscle EMGactivity served as a marker for locomotion when thelimbs were freely moving.

Protocol 2: Spontaneous locomotion with baroreceptorinput eliminated. Five cats from protocol 1 were alsostudied after baroreceptor and chemoreceptor dener-vation, which was accomplished by severing the carotidsinus nerves and the vagosympathetic trunks bilaterally.In all five cats, complete baroreceptor denervation wasconfirmed by demonstrating that inhibition of renalnerve activity in response to a phenylephrine-inducedpressor response was eliminated.

Protocol 3: Spontaneous fictive locomotion with mus-cle afferent input eliminated. Twenty cats that exhib-ited spontaneous locomotor behavior were paralyzedwith gallamine triethiodide (3 mg/kg initially andsupplemented with 3 mg/kg/hr) to eliminate skeletalmuscle activity and, thus, afferent input from exercis-ing muscle. Ventilation was controlled with a respi-rator. The occurrence of central locomotor drive wasmonitored in this "fictive" locomotor state by record-ing hind limb motor nerve activity.

Protocol 4: Spontaneous fictive locomotion with bothmuscle afferent and baroreceptor input eliminated. For17 cats, the responses to spontaneous fictive locomo-tion were also examined after baroreceptor andchemoreceptor denervation.

Protocol 5: Electrical activation of STLR. In fourcats, locomotion was induced by electrical stimula-tion in the STLR. The STLR was identified using acombination of stereotaxic coordinates and charac-teristic responses to electrical stimulation. This ap-proach has been followed previously using the sameanimal model8-11,20 and is briefly described below. Abipolar stimulating electrode (100 kfl, FrederickHaer and Co., Brunswick, Me.) was inserted into theSTLR using the following stereotaxic coordinates:A9, L1-L2, and H-3.8 Stimulus trains were deliveredby passing current ranging from 25 to 300 ,uA with a200-,usec pulse duration at a frequency of 50 Hz. Ifcoordinated limb movement did not occur with stim-ulation at this site, the electrode was repositioneddorsoventrally within a tract or moved to anotherposition within a 1-mm radius of this location until a

responsive site was found. Sympathetic and hind limbnerves were then placed on recording electrodes, andthe cat was ventilated with a respirator and thenparalyzed with gallamine triethiodode to eliminatefeedback from the exercising muscle. Locomotion inthese baroreceptor intact cats was monitored underthese conditions by recording hind limb motor nerveactivities.

Data AnalysisArterial pressure, renal sympathetic nerve activity

(RSNA), lumbar sympathetic nerve activity (LSNA),phrenic nerve activity, and EMG or motor nerve activ-ity were recorded and stored on FM (model G Re-corder, A.R. Vetter Co., Rebersburg, Pa.) or VHS(model 4000 PCM digital processor, A.R. Vetter) tapefor off-line analysis. The output from the tape wasplayed back to an electrostatic recorder (modelES1000, Gould Inc., Cleveland, Ohio) for data analysisand figures. Control values for the above variables wereobtained over 10-30 seconds during steady-state con-ditions before the onset of locomotion. The data frommultiple (two to four) bouts of locomotion for each catwere averaged to obtain a single value for each cat, andentries represent the average response for each cat.RSNA and LSNA were integrated after full-wave rec-tification of the nerve signal and placed into 1-secondbins. Control values were averaged over 10-30 secondsand compared with the peak mean arterial pressure(MAP), RSNA, and LSNA responses during locomo-tion. The voltage from phrenic nerve activity was inte-grated using 100-msec time bins, and minute activitywas computed.10 Phrenic nerve minute activity wascalculated as the product of tidal activity (peak phrenicactivity) and respiratory frequency.The timing of the activation of RSNA with respect

to the onset of rhythmic locomotor activity wasanalyzed in cats in which bilateral hind limb EMG ormotor nerve activity was recorded. Recordings wereexamined visually for an obvious burst of sympatheticnerve activity correlated with the onset of locomo-tion. Two investigators independently inspected eachrecording to determine the relation between onset ofEMG or muscle nerve activity and sympathetic nerveactivity. Time zero was taken at the beginning of thelocomotor burst from the hind limb that first exhib-ited locomotion. Negative values indicate that RSNApreceded locomotor activity; positive values reflectactivation of RSNA after the onset of locomotoractivity.

Statistical analysis was performed using analysis ofvariance and Fisher's protected least significant dif-ference post hoc test for multiple comparisons.21Values of p<0.05 were considered significant. Re-sults were expressed as mean + SEM.

All experiments performed in this study compliedwith the guiding principles of the American Physio-logical Society on animal experimentation.


Dow

nloaded from



Arterial 1501Pressure m(mmHg) 50

LeftEMGRight _,EMG

40,IntegratedRSNA(sV) 0TX

RSNA

iMI.MMM.IUMmIm.

-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l! .-i ,1 -1 1 If ll I1 1MpifiJAIIIik-ik

-

FIGURE 1. Experimental recordings showing theincrease in renal sympathetic nerve activity (RSNA)in a decorticate cat with baroreceptors intact duringspontaneous rhythmic locomotion. With the onset ofbilateral locomotion, as seen in the altemating rhyth-mic hind limb biceps femoris electromyograms(EMGs), RSNA increased immediately with a grad-ual rise in arterial pressure. Integrated RSNA (thirdtracing) was obtained afterfull-wave rectification ofthe raw renal voltage neurogram and was reset everysecond.

a

5 sec

ResultsPatterns of Locomotor Activity

Several patterns of spontaneously occurring loco-motor activity were observed. Some cats (n= 14)exhibited coordinated, rhythmically alternating limbmotion (rate, 0.5-2.5 steps/sec) in the unparalyzedstate and rhythmic alternating discharge in the hindlimb motor nerves during paralysis. Other cats (n =7)demonstrated coordinated rhythmically synchronouslimb motion (rate, 0.5-2.8 steps/sec) in the unpara-lyzed state and rhythmic hind limb motor nerveactivity in the paralyzed state. Cats displaying eitherof these two patterns of activity were assigned to the"rhythmic" category of spontaneous locomotion. Asecond group of cats (n=9) displayed spontaneousepisodic nonrhythmic limb motion or hind limb mo-tor nerve discharge. Motor nerve activity in thisgroup of cats was characterized by the sudden onsetof continuous discharge with prolonged bursts ofactivity. These cats were assigned to the "static"category of locomotion. Animals with static patternsof limb muscle activity also occasionally exhibitedcoordinated rhythmic locomotor activity as definedabove. Thus, data from some cats (n=2) appear inboth the rhythmic and static categories.The results are reported from 34 cats. In all cats

analyzed, there was coordinated four-legged walkingin the unparalyzed condition, with a walking fre-quency of -2 steps/sec/limb. The duration of thelocomotor episodes ranged from 5 seconds to morethan 30 minutes.

Responses During Spontaneous Rhythmic LocomotionFigure 1 illustrates a typical response to the onset

of spontaneous locomotion in a cat with intactbaroreceptors and with functioning skeletal muscle.The onset of exercise was associated with an imme-diate increase in RSNA and a gradual increase inarterial pressure. Figure 2 shows the grouped datafor six cats. With intact baroreceptors, spontaneousrhythmic locomotion increased MAP by 29±12%from a baseline of 106± 10 mm Hg and increasedRSNA by 301±100%. After barodenervation, MAPincreased by 28±11%, and RSNA increased by

249+67% from a baseline of 93±12 mm Hg duringlocomotion. Baseline MAP was not significantly dif-ferent before or after barodenervation.

Responses During Spontaneous RhythmicFictive LocomotionThe responses during spontaneous rhythmic fictive

locomotion are shown in Figures 3 and 4. With theonset of fictive locomotion (Figure 3), evidenced bythe increased bilateral motor nerve activity, therewere abrupt increases in RSNA and LSNA, followedby a slower rise in arterial pressure. With the cessa-tion of fictive locomotion, sympathetic nerve activityreturned promptly to or below control levels with aslower return of arterial pressure to control.With spontaneous rhythmic fictive locomotion be-

fore barodenervation (Figure 4), MAP increased by43±6%, RSNA increased by 183+32%, and LSNAincreased by 223+83%. During fictive locomotionafter barodenervation, MAP, RSNA, and LSNAincreased significantly by 56±14%, 226+42%, and169±25%, respectively (Figures 3 and 4). Minutephrenic nerve activity increased during spontaneousfictive locomotion before (446%, n=2) and after

MAP(%M)

RSNA(%)

intat Denervated

FIGURE 2. Bar graphs showing increases of mean arterialpressure (MAP, left panel) and renal sympathetic nerveactivity (RSNA, right panel) during spontaneous rhythmiclocomotion with baroreceptors intact (open bars) and afterbaroreceptor denervation (shaded bars). Baroreceptor dener-vation did not alter the magnitude of the MAP and RSNAincreases. Values are presented as percent change from prelo-comotion control (mean-+SEM). Numbers within the barsrepresent the number of cats. *p<0.05 compared with prelo-comotion control values.


Dow

nloaded from



Intact DenervatedArterial 2001Pressure .ii

W1

(mmHg) 100.1

Right _-__ _-_

Motor Nerve mmLeft L_&___

Motor Nerve v r r

RSNA

..I.I~~~~~~~~~~~-EUU..-

,, -lr--vr----l- ir Fsl-. Xr ---- 1-r I -r vry ,.,.,1

5 sw

FIGURE 3. Experimental recordings showing the increases in renal sympathetic nerve activity (RSNA) and lumbar sympatheticnerve activity (LSNA) with baroreceptors intact (left recordings) and after baroreceptor denervation (right recordings) in the samecat with the onset of spontaneous fictive locomotion. RSNA and LSNA increased abruptly with the onset of bilateral hind limbmotor nerve activity while arterialpressure increased gradually. The responses of mean arterial pressure, RSNA, and LSNA duringlocomotion were similar after baroreceptor denervation (right recordings).

(541%, n=2) barodenervation. No significant differ-ences were observed in the MAP, RSNA, or LSNAresponses during fictive rhythmic locomotion beforeversus after barodenervation.

Responses During Static Fictive Locomotor ActivityIn cats that responded with a continuous burst of

motor nerve discharge, there were similar significantincreases in MAP (50±5% versus 57±6%) andRSNA (145±17% versus 190±47%) before and afterbarodenervation, respectively (Figure 5). There were

no differences in these responses during fictive loco-motion before versus after barodenervation. In twocats, LSNA increased by 495% and 461%, andminute phrenic nerve activity increased by 87% and138% before and after barodenervation, respectively.

Responses to Electrical Stimulation of STLRDuring stimulation of the STLR, MAP increased

significantly by 43±17%, and RSNA increased by175±47% (Figures 6 and 7). The responses were

similar to those seen during spontaneous locomotion.

MAP RSNA LSNA(9A) (%A) (%A)80- * 300- * 300-

60-

:0. 0 04-Intact Denervated Intact Denervated I

Timing of Locomotor and Sympathetic Changes

We found no significant differences in the time ofonset of hind limb locomotor activity and RSNA.During spontaneous rhythmic locomotion, values forthis relation were - 170+550 msec in intact (n =6) and- 10±210 msec in denervated (n=5) cats. No signifi-cant differences between RSNA and motor nerve

activity were observed during spontaneous rhythmicfictive locomotion in intact (-100±260 msec, n=5)and denervated (-240±410 msec, n=5) cats.

DiscussionIn recent years, the role of central command in

regulation of sympathetic nerve activity has beencontroversial. From previous studies4-7,22 using mea-

surements of heart rate and arterial pressure, it hasbeen assumed that central command plays a majorrole in augmenting sympathetic neural drive. Forinstance, Freund et a122 demonstrated that cardiacoutput or arterial pressure responses to muscle con-

traction were not attenuated when the sensory input

FIGURE 4. Bar graphs showing that spontane-ous fictive rhythmic locomotion produced in-creases of mean arterial pressure (MAP, leftpanel), renal sympathetic nerve activity (RSNA,center panel), and lumbar sympathetic nerve

activity (LSNA, right panel) in cats with barore-ceptors intact (open bars) and after baroreceptordenervation (shaded bars). Control values forMAP were 105±6 mm Hg in the intact and96±10 mm Hg in the denervated state. Barore-ceptor denervation did not alter the magnitude ofthe MAP, RSNA, or LSNA increases. Values are

Dernted presented as percent change from prelocomotion

control (mean ±SEM). Numbers within the barsrepresent the number of cats. *p<0.05 comparedwith prelocomotion control values.

LSNA

bL,., j UM

AW-UJ--- Lii-il-

PT -1-rI-rTT"~wnv., 1

1AAA -. .. - 1 .11 L

711-11-iii. 11. T77,---TI . 'Tr lg'fr 11------ 1- 1

1,

AmbiliALAL A-LA --

000 -


Dow

nloaded from



MAP(%A)

RSNA *

200,

100-

Intact Intact Denenrated

FIGURE 5. Bar graphs showing that spontaneous fictivestatic locomotion produced increases ofmean arterialpressure(MAP, left panel) and renal sympathetic nerve activity(RSNA, right panel) in cats with baroreceptors intact (openbars) and after baroreceptor denervation (shaded bars). Con-trol values for MAP were 94±13 mm Hg in the intact and84+10 mm Hg in the denervated state. Baroreceptor denerva-tion did not alter the magnitude of the MAP or RSNAincreases. Values are presented as percent change from prelo-comotion control (mean±SEM). Numbers within the barsrepresent the number of cats. *p<0.05 compared with prelo-comotion control values.

from exercising muscle was blocked by epiduralanesthesia in humans. With attempted handgrip dur-ing neuromuscular blockade, increases in heart rate,arterial pressure, and muscle sympathetic nerve ac-tivity have also been shown.14,23 However, recentstudies1214 in humans involving direct recording ofsympathetic nerve activity have suggested that cen-tral command does not play a major role in increasesin sympathetic neural vasoconstrictor activity to mus-cle or skin. Mark et a112 concluded that centralcommand does not increase and may actually inhibitslightly sympathetic activity to nonexercising muscleduring static handgrip in humans. In a study in whichexercise was attempted during partial neuromuscularblockade, Victor et a114 concluded that central com-mand produces modest increases in sympatheticnerve activity during static handgrip. It has recentlybeen shown that central command may play animportant role in increases in sympathetic neuralsudomotor activity in skin but not in sympatheticneural vasoconstrictor activity to muscle.16 Thus,recent studies in humans have renewed interest inthe role of central command in the regulation ofsympathetic nerve activity. In light of these recentobservations, we sought to determine whether cen-

*

4 4

MAP RSNAFIGURE 7. Bar graph showing the effect of electrical stimu-lation ofthe subthalamic locomotor region on the responses ofmean arterial blood pressure (MAP) and renal sympatheticnerve activity (RSNA) in baroreceptor intact cats. Values arepresented as percent change from prelocomotion control(mean +SEM). Numbers within the bars represent the numberofcats. *p <0.05 compared withprelocomotion control values.

tral command alone can increase sympathetic nerveactivity during spontaneous fictive locomotion in adecorticate cat model in which confounding sensoryinputs from muscle afferents and cardiovascular re-ceptors could be effectively excluded.The major finding of the present study is that

sympathetic nerve activity increased substantiallyduring actual and fictive spontaneous locomotion andduring electrical stimulation of the STLR in unanes-thetized decorticate cats. This increase in sympa-thetic drive occurred in the absence of feedback fromcontracting muscles, arterial and cardiopulmonarybaroreceptors, and arterial chemoreceptors. We ob-served both rhythmic and static patterns of motoractivity during fictive locomotion. The responses ofarterial pressure, RSNA, and LSNA were similarregardless of the pattern of motor activity. In thefollowing discussion, these results are considered inthe context of the potential contribution of "centralcommand" to the autonomic responses to exercise.

Central Command and the Decorticate Cat Modelof ExerciseThe onset of exercise is associated with character-

istic changes in autonomic outflow, which result inincreases in arterial pressure and heart rate and incentral respiratory drive.1-4,10 In the decorticate cat

200.ArterialPressure j FIGURE 6. Experimental recordings showing the

0] responses to electrical stimulation (solid bar) of theRight L I subthalamic locomotor region in a paralyzed, decor-

Motor Nerve ticate cat with baroreceptors intact. Stimulationresulted in rhythmic motor nerve activity, increasedrenal sympathetic nerve activity (RSNA), and a rise

vY _7 -1 F MI-FrIT. of arterial pressure.

50 tA 5 sec

RSNA

n .

m


Dow

nloaded from



model, similar changes have been observed duringactivation of the STLR, whether occurring spontane-ously or in response to electrical stimulation.91020Thus, spontaneous or electrical activation of theSTLR has been used as a model to investigate thepotential role of central command in the exerciseresponse.9,10,20 In our study, pronounced increases inarterial pressure and the discharge in renal andlumbar sympathetic nerves as well as an increase inphrenic motor nerve activity were associated with theonset of either rhythmic or static locomotion in thismodel.The responses to spontaneous locomotion in the

unanesthetized decorticate cat may not mimic thetotal contribution of central command during volun-tary exercise in intact humans. Cortical sites areinvolved in the voluntary decision to exercise. In thedecorticate cat model for locomotor studies, the con-nections with cortical neurons have been severed. Thedecorticate animal is subject to episodes of spontane-ous locomotion, often cyclical in pattern, withoutregulation by higher centers. It is well established thatthis locomotor behavior originates from neurons ofthe subthalamic locomotor center when the tonicinhibitory influence from higher centers is removed. Itis conceivable that, under intact conditions, signalsoriginating at the cortical level during exercise mightimportantly modulate the contribution of STLR neu-rons and central command. Thus, it is possible that therole of central command in this model might differfrom that in intact animals or humans. In fact, theinfluences from higher centers may partially accountfor the differences in the response of sympatheticnerve activity observed between our study and those inhumans (see below). Despite this caveat, insightsconcerning the role of central command have emergedfrom the decorticate cat model.

Central Command Increases Sympathetic NerveActivity Independent of Feedback FromContracting Muscle

Previous studies24-28 have attributed the auto-nomic responses during exercise to activation ofmuscle afferents. Muscle mechanoreceptors, withtype III afferents in particular, have been reported toincrease renal sympathetic nerve activity.29 Me-taboreceptors responding to cellular metabolitesfrom contracting muscle are also capable of activat-ing sympathetic nerve activity.30 Moreover, Waldropand Stremel31 have recently demonstrated that neu-rons in the STLR can be activated by stimulation ofmuscle afferents, suggesting that there may be directfeedback from exercising muscle to this locomotorcenter.We observed striking increases in RSNA and

LSNA as well as arterial pressure during spontane-ous locomotion and fictive locomotion after paralysis.In the fictive state, respiratory drive, as indicated byphrenic motor nerve activity, also increased, consis-tent with earlier studies.9'10 Similarly Eldridge et al10have noted that increases in respiratory and arterial

pressure did not differ during spontaneous locomo-tion versus fictive locomotion in this model. Weobserved that sympathetic nerve activity responsesdid not differ during spontaneous locomotion andfictive locomotion. The observation that muscle af-ferents do not contribute importantly to the sympa-thetic and pressure responses in this preparation isnot surprising since the locomotion occurred withoutlimb resistance. In the absence of a substantialworkload, the stimulus to muscle metaboreceptors orto muscle mechanoreceptors would be minimal.Thus, our results do not totally exclude a role formuscle afferents in the exercise response. However,the results do indicate that signals originating fromthe STLR, when operating in isolation, are capableof engendering a prominent sympathoexcitatory re-sponse. This response might be construed to berepresentative of events occurring at the initiation ofexercise before the development of substantial met-abolic or mechanical stresses in exercising muscle.

Central Command Increases Sympathetic NerveActivity Independent of Feedback From Baroreceptors

The precise role of the arterial and cardiopul-monary baroreceptors in the cardiovascular adjust-ments to exercise is controversial and undoubtedlycomplex. Earlier studies3233 suggested that exercisewas accompanied by inhibition of baroreceptor re-flexes. Ludbrook and Graham34 observed that thebaroreflex control of systemic vascular resistance issuppressed at the onset of exercise but that, asexercise continues, the arterial baroreceptors regaintheir capacity to control the circulation. Moreover,evidence indicates that baroreceptor influences re-main operative but are reset to a higher pressureduring steady-state exercise.3536 In theory, this up-ward resetting of the baroreflex might account in partfor the sympathetic response to exercise, since up-ward resetting would result in an increase in sympa-thetic drive for a given level of arterial pressure.

In the decorticate cat, we found that denervation ofsinoaortic and cardiopulmonary baroreceptors had noeffect on the magnitude of the peak increases insympathetic nerve activity and arterial pressure. Thisobservation indicates that central command can in-crease sympathetic nerve activity independent ofbaroreceptor reflexes. Our study was not designedprimarily to determine the interaction of baroreceptorreflexes and central command during exercise, and wecannot exclude a role of baroreceptors in modulatingthe responses of central command during steady-stateexercise. Nevertheless, our studies indicate that thebaroreflexes are not essential for the sympathetic nerveactivity response to central command.During the denervation procedure, peripheral

chemoreceptor and pulmonary receptor inputs alsowere eliminated. Although these receptor regionscould potentially modulate the responses to exercise,their elimination had no effect on the magnitude ofthe peak increases in sympathetic activity in ourpreparation. In fact, the increased ventilation and


Dow

nloaded from


Hajduczok et al Central Command Increases Sympathetic Nerve Activity

slight hypocapnia that may occur during exercise inthe spontaneously breathing unparalyzed cats wouldtend to minimize rather than enhance the increase insympathetic nerve activity. In addition, in the para-lyzed cats, ventilation was controlled, and any stimulito the chemoreceptors would be further minimized.These data suggest that neither baroreceptors,chemoreceptors, nor pulmonary receptors accountfor the sympathetic nerve activity response to theonset of exercise in our study.

Timing of Locomotor and Sympathetic ChangesThe onset of spontaneous rhythmic locomotion

(actual and fictive) was associated with a virtuallysimultaneous increase in RSNA. Occasionally, sym-pathetic nerve activity preceded locomotion by amatter of milliseconds, but this was not a consistentfinding. Since motor nerve or EMG activity wasrecorded only from the hind limbs, it is still conceiv-able that motor activity to forelimb muscles mighthave preceded any change in sympathetic discharge.Increases of arterial pressure, on the other hand,always followed the onset of locomotion and sympa-thetic nerve activity. Our results suggest that theonset of locomotion is associated with a parallelactivation of sympathetic drive.

In the recording shown in Figure 6, electricalstimulation of the STLR appears to activate RSNAbefore the onset of locomotion. Earlier activation ofsympathetic nerve activity in this protocol would beconsistent with the idea10 that mild electrical stimu-lation might first activate systems that are tonicallyactive and above threshold (cardiovascular neurons)and then activate locomotor neurons that are nor-mally below threshold. However, this finding is notconclusive, because only unilateral hind limb motornerve activities were obtained in this series of studies.

Eldridge et al10 noted that respiratory and circula-tory changes often slightly preceded the developmentof actual or fictive locomotion. However, thesechanges were not quantitated, and only one exampleeach of actual and fictive locomotion was provided insupport of their conclusion so that a direct compar-ison with our findings is difficult. Electrical stimula-tion of the STLR in this same study10 also appearedto activate respiratory activity and arterial pressurebefore the onset of locomotion, with a variable timelag. Despite the slight differences between our studyand that of Eldridge et al, both studies support theconclusion that signals originating from the hypothal-amus are responsible for locomotor, respiratory, andcardiovascular adjustments at the onset of exercise.

Physiological ImplicationsThe augmentation of sympathetic drive in renal

and lumbar nerves during locomotion may haveimportant functional implications. An increase inRSNA by central command would be expected toincrease renal vascular resistance and arterial pres-sure and thereby promote perfusion of contractingmuscle. In studies of regional blood flow responses

during electrical stimulation of the STLR, Waldropet a120 observed increases in vascular resistance to therenal and gastrointestinal regions, with increases inblood flow to heart, diaphragm, and skeletal muscle.The increases in renal vascular resistance accompa-nying the pressor response to STLR stimulationcould be caused by autoregulation or by sympatheticstimulation of the renal vasculature. Our findingssuggest that the increased renal vascular resistancecan be attributed to a pronounced increase in sym-pathetic drive to the renal vascular bed.LSNA also increased during spontaneous fictive

locomotion in our study. In this study, we measuredLSNA at the L-2 and L-3 sympathetic chain region.At this level, lumbar sympathetic outflow suppliesinnervation to the hind limb skin and skeletal muscleas well as to abdominal visceral organs. Sympatheticoutflow to skin regulates vasoconstriction and pilo-motor and sudomotor activity.37 Sympathetic outflowto muscle regulates both cholinergic vasodilator andadrenergic constrictor influences.37 Our results mightbe consistent with the view that activity increases tovasoconstrict inactive regions and to vasodilate activemuscles during exercise. However, Vissing and Vic-tor16 have recently reported that, in humans, centralcommand increases sympathetic sudomotor but notvasoconstrictor activity to skin. Other findings38 inhumans indicate that central command increasessympathetic cholinergic vasodilator activity to musclebut does not contribute importantly to increases insympathetic vasoconstrictor activity to muscle. Thepresent study does not address the exact nature ofthe lumbar sympathetic pathways activated duringspontaneous locomotion.Our results indicating a relation between engage-

ment of central command and striking increases insympathetic nerve discharge with exercise should bediscussed in comparison with recent studies in hu-mans. First, a direct comparison between human andanimal studies is rendered difficult by the fact thatvisceral nerves are not accessible for study in humansand also because selective sympathetic outflow tomuscle beds has not been examined in an animalmodel.

Second, there is a lack of modulatory influences ofhigher brain centers on the STLR in the decorticatecat. The volitional component of exercise in humansis absent in this model. Although the potential influ-ences of cortical sites on the cardiovascular responsesto exercise have not been explored, it is clear thatcortical sites can effect indexes of cardiovascularfunction. For instance, Oppenheimer and Cechetto39noted that a chronotropic organization exists withinthe insular cortex. Electrical stimulation of the ros-tral posterior or caudal posterior insular regionsresults in a tachycardia or bradycardia, respectively,in rats. These changes occur in the absence of anychanges in blood pressure or respiration.39 Oppen-heimer et a140 hypothesized that the insular cortex,which has extensive connections with the limbicsystem, the hypothalamus, and other areas involved

73


Dow

nloaded from



in autonomic control, integrates emotional and auto-nomic states. Thus, activation of cortical neuronsduring exercise in humans could account for a selec-tive tachycardia with no appreciable change in mus-cle sympathetic nerve activity.

Third, the nature of the "exercise" stimulus mightcontribute to the different responses to exerciseobserved in humans and in the decorticate cat model.In the decorticate cat, the onset of locomotion ap-pears to be associated with a diffuse activation of thesympathetic nervous system, as evidenced by in-creases in RSNA and LSNA in this study. That viewis supported by the observation that cardiovascularand sympathetic responses can be elicited by micro-injection of GABA antagonists in posterior hypothal-amus. The association between enhanced sympa-thetic activity and locomotion was not examined inthese latter studies.17'18 In human studies measuringsympathetic nerve activity, the exercise has usuallybeen limited to a relatively small muscle mass, andsympathetic nerve activity is commonly recordedfrom muscle regions that are voluntarily maintainedin an inactive state. Under these conditions, thehypothalamic neurons effecting changes in auto-nomic activity may be more selectively engaged.

ConclusionThis study indicates that central command originat-

ing in the STLR in the unanesthetized decorticate catmodel increases renal and lumbar sympathetic nerveactivity. This increase in sympathetic drive can occur inthe absence of feedback from contracting muscle andarterial and cardiopulmonary baroreceptors.

AcknowledgmentsWe would like to thank Harold D. Brinegar for

research assistance, Carolyn Wagner and Deb Schiekfor the preparation of the figures, and Nancy Stampfor secretarial assistance.

References1. Mitchell JH, Kaufman MP, Iwamoto GA: The exercise pressor

reflex: Its cardiovascular effects, afferent mechanisms, andcentral pathways. Annu Rev Physiol 1983;43:229-242

2. Dejours P: Control of respiration in muscular exercise, inFenn WO, Rahn H (eds): Handbook of Physiology, Section 3:Respiration, Volume I. Washington, DC, American Physiolog-ical Society, 1964, pp 631-648

3. Millhorn DE, Eldridge FL, Waldrop TG, Kiley JP: Dience-phalic regulation of respiration and arterial pressure duringactual and fictive locomotion in cat. Circ Res 1987;61(supplI):I-53-I-59

4. Mitchell JH: Cardiovascular control during exercise: Centraland reflex neural mechanisms.Am J Cardiol 1985;55:34D-41D

5. Johansson JE: Uber die einwirkung der muskelta tigkeit aufdie atmung und die herzta tigkeit. Skand Arch Physiol 1893;5:20-66

6. Krogh A, Lindhard J: The regulation of respiration andcirculation during the initial stages of muscular work. J Physiol(Lond) 1913;47:112-136

7. Hobbs SF: Central command during exercise: Parallel activa-tion of the cardiovascular and motor systems by descendingcommand signals, in Smith OA, Galosy RA, Weiss SM (eds):Circulation, Neurobiology, and Behavior. New York, ElsevierScience Publishing Co, Inc, 1982, pp 217-232

8. Orlovskii GN: Spontaneous and induced locomotion of thethalamic cat. Biofizika 1969;6:1095-1102

9. Eldridge FL, Millhorn DE, Waldrop TG: Exercise hyperpneaand locomotion: Parallel activation from the hypothalamus.Science 1981;211:844-846

10. Eldridge FL, Millhorn DE, Kiley JP, Waldrop TG: Stimula-tion by central command of locomotion, respiration andcirculation during exercise. Respir Physiol 1985;59:313-337

11. Waldrop TG, Bauer RM, Iwamoto GA: Microinjection ofGABA antagonists into the posterior hypothalamus elicitslocomotor activity and a cardiorespiratory activation. Brain Res1988;444:84-94

12. Mark AL, Victor RG, Nerhed C, Wallin BG: Microneuro-graphic studies of the mechanisms of sympathetic nerveresponses to static exercise in humans. Circ Res 1985;57:461-469

13. Victor RG, Seals DR, Mark AL: Differential control of heartrate and sympathetic nerve activity during dynamic exercise. JClin Invest 1987;79:508-516

14. Victor RG, Pryor SL, Secher NH, Mitchell JH: Effects ofpartial neuromuscular blockade on sympathetic nerveresponses to static exercise in humans. Circ Res 1989;65:468-476

15. Pryor SL, Lewis SF, Haller RG, Bertocci LA, Victor RG:Impairment of sympathetic activation during static exercise inpatients with muscle phosphorylase deficiency (McArdle'sdisease). J Clin Invest 1990;85:1444-1449

16. Vissing SF, Victor RG: Central command activates sympa-thetic outflow to skin during static exercise (abstract). ActaPhysiol Scand 1989;136(suppl 584):44

17. Waldrop TG, Bauer RM: Modulation of sympathetic dis-charge by a hypothalamic GABAergic mechanism. Neurophar-macol 1989;28:263-269

18. Wible JH Jr, Luft FC, DiMicco JA: Hypothalamic GABAsuppresses sympathetic outflow to the cardiovascular system.Am J Physiol 1988;254:R680-R687

19. Hade JS, Mifflin SW, Donta TS, Felder RB: Stimulation ofparabrachial neurons elicits a sympathetically mediated pres-sor response in cats. Am J Physiol 1988;255:H1349-H1358

20. Waldrop TG, Henderson MC, Iwamoto GA, Mitchell JH:Regional blood flow responses to stimulation of the subtha-lamic locomotor region. Respir Physiol 1986;64:93-102

21. Snedecor GW, Cochran WG: Statistical Methods, ed. 7. Ames,Iowa, Iowa State University Press, 1980, pp 215-333

22. Freund PR, Rowell LB, Murphy TM, Hobbs SF, Butler SH:Blockade of the pressor response to muscle ischemia bysensory nerve block in man. Am JPhysiol 1979;237:H433-H439

23. Lassen A, Mitchell JH, Reeves DR Jr, Rogers HB, SecherNH: Cardiovascular responses to brief static contractions inman with topical nervous blockade. JPhysiol (Lond) 1989;409:333-341

24. Alam M, Smirk FH: Observations in man upon a bloodpressure raising reflex arising from the voluntary muscles. JPhysiol (Lond) 1937;89:372-383

25. Coote JH, Hilton SM, Perez-Gonzalez JF: The reflex nature ofthe pressor response to muscular exercise. J Physiol (Lond)1971;215:789-804

26. McCloskey DI, Mitchell JH: Reflex cardiovascular and respi-ratory responses originating in exercising muscle. J Physiol(Lond) 1972;224:173-186

27. Mitchell JH, Schmidt RF: Cardiovascular reflex control byafferent fibers from skeletal muscle receptors, in Shepherd JT,Abboud FM (eds): Handbook of Physiology, Section 2: TheCardiovascular System, Volume III, Part 2. Bethesda, Md,American Physiological Society, 1983, pp 623-658

28. Mitchell JH, Reardon WC, McCloskey DF: Reflex effects oncirculation and respiration from contracting skeletal muscle.Am J Physiol 1977;233:H374-H378

29. Victor RG, Rotto DM, Pryor SL, Kaufman MP: Stimulation ofrenal sympathetic activity by static contraction: Evidence formechanoreceptor-induced reflexes from skeletal muscle. CircRes 1989;64:592-599


Dow

nloaded from



30. Victor RG, Bertocci LA, Pryor SL, Nunnally RL: Sympatheticnerve discharge is coupled to muscle cell pH during exercise inhumans. J Clin Invest 1988;82:1301-1305

31. Waldrop TG, Stremel RW: Muscular contraction stimulatesposterior hypothalamic neurons. Am J Physiol 1989;256:R348-R356

32. Bristow JD, Brown EB Jr, Cunningham DJC, Howson MG,Petersen ES, Pickering TG, Sleight P: Effect of bicycling on

the baroreflex regulation of pulse interval. Circ Res 1971;28:582-592

33. Cunningham DJC, Petersen ES, Peto R, Pickering TG, SleightP: Comparison of the effect of different types of exercise on

the baroreflex regulation of heart rate. Acta Physiol Scand1972;86:444-455

34. Ludbrook J, Graham WF: Circulatory responses to onset ofexercise: Role of arterial and cardiac baroreflexes. Am JPhysiol 1985;248:H457-H467

35. Melcher A, Donald DE: Maintained ability of carotid barore-flex to regulate arterial pressure during exercise. Am J Physiol1981;241:H838-H849

36. Walgenbach SC, Shepherd JT: Role of arterial and cardiopul-monary mechanoreceptors in the regulation of arterial pres-

sure during rest and exercise in conscious dogs. Mayo Clin Proc1984;59:467-475

37. Janig W: Organization of the lumbar sympathetic outflow toskeletal muscle and skin of the cat hindlimb and tail. RevPhysiol Biochem Pharmacol 1985;102:119-213

38. Sanders JS, Mark AL, Ferguson DW: Evidence for cholinergi-cally mediated vasodilation at the beginning of isometricexercise in humans. Circulation 1989;79:815-824

39. Oppenheimer SM, Cechetto DF: Cardiac chronotropic orga-nization of the rat insular cortex. Brain Res 1990;533:66-72

40. Oppenheimer SM, Cechetto DF, Hachinski VC: Cerebrogeniccardiac arrhythmias: Cerebral electrocardiographic influencesand their role in sudden death. Arch Neurol 1990;47:513-519

KEY WORDS * exercise reflex * peripheral feedbackbaroreflex * subthalamic locomotor region


Dow

nloaded from


G Hajduczok, J S Hade, A L Mark, J L Williams and R B Feldercats.

Central command increases sympathetic nerve activity during spontaneous locomotion in

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1991 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.69.1.661991;69:66-75Circ Res.

http://circres.ahajournals.org/content/69/1/66World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


Dow

nloaded from

http://circres.ahajournals.org/content/69/1/66

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


Central Command Increases Sympathetic Nerve Activity During ...

Documents

Transcript of Central Command Increases Sympathetic Nerve Activity During ...