Journal of Applied Physiology Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 83: 1448-1453, 1997;
8750-7587/97 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chefer, S. I.
Right arrow Articles by Engel, B. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chefer, S. I.
Right arrow Articles by Engel, B. T.

Vol. 83, Issue 5, 1448-1453, 1997

Central neural correlates of learned heart rate control during exercise: central command demystified

Svetlana I. Chefer, Mark I. Talan, and Bernard T. Engel

Laboratory of Behavioral Sciences, Gerontology Research Center, National Institute on Aging, National Institutes of Health, Baltimore, Maryland 21224

ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
FOOTNOTES
REFERENCES

ABSTRACT

Chefer, Svetlana I., Mark I. Talan, and Bernard T. Engel. Central neural correlates of learned heart rate control during exercise: central command demystified. J. Appl. Physiol. 83(5): 1448-1453, 1997.---To identify the brain areas involved in central command, four monkeys were trained to attenuate the tachycardia of exercise while different brain sites affecting heart rate (HR) were simultaneously stimulated electrically. Among 24 brain sites located mostly in the limbic structures, we have identified four types of control systems that mediate cardiovascular and motor behavior during exercise. One system increases HR equivalently during both exercise and operantly controlled HR, whereas another increases HR during both tasks and abolishes operant HR control. In the third system, the effect of brain stimulation on HR is attenuated during exercise and during exercise with operantly controlled HR. The fourth system increases HR in both tasks, but its effect is significantly attenuated during operant HR control. We believe that this last system, which includes the mediodorsal nucleus, nucleus ventralis anterior, and cingulate cortex, plays a significant role in central command.

monkeys; nonhuman primates; central nervous system; operant conditioning; tachycardia of exercise

INTRODUCTION

KROGH AND LINDHARD reported in 1913 (10) that "the rise in ventilation like the increase in heart rate [during exercise] is not produced reflexly but by irradiation of impulses from the motor cortex." Contemporary physiologists use the term central command as a substitute for "irradiation of impulses" (12). Nevertheless, there is little substantive research available to characterize the possible central neural mediators of central command. However, from a behavioral perspective, it is clear that operant learning is one of the primary mechanisms underlying the phenomena of adaptation and central command (3).

We have previously shown that nonhuman primates can be operantly conditioned to attenuate the tachycardia of exercise (4, 14); i.e., the cardiovascular adjustments to exercise can be learned behaviors. Neither sympathetic nor vagal blockade prevented the learned attenuation of tachycardia of exercise (5). This fact established that the responses of the circulation during exercise are not merely passive effects mediated by somatomotor activity or metabolic needs but that they are also a complex behavior mediated by the central nervous system in response to learned, environmental signals (6). We concluded that the mechanisms mediating cardiovascular responses during exercise in trained subjects are under direct control of the central nervous system. Metabolic and reflexive mechanisms define the limits of performance rather than its expression (5). These findings, coupled with earlier findings which showed that animals trained to slow heart rate (HR) could do so even in the presence of brain stimulation which normally caused a tachycardia (9), indicate that the cardiovascular adjustments to exercise are emitted in parallel with somatomotor responses and that somatomotor and cardiovascular responses can be dissociated under proper conditions.

The present research was directed at understanding the central neural mechanisms that mediate conditioning of cardiovascular behavior accompanying exercise. The goal of the present study was to identify the brain areas involved in learned attenuation of the tachycardia of exercise.

METHODS

Subjects. Four male monkeys (Macaca mulatta), 3-4 yr of age and weighing 4-9 kg, were used in this study. The monkeys were trained to wear protective primate jackets at all times and to be transferred daily from home cages into primate chairs (except weekends). They were adapted to remain in the chairs during experimental sessions from 9 AM to 4 PM. Animals were provided with ~200 g/day primate laboratory diet (between 5 PM and 9 AM) and water at all times except during testing. This ration was supplemented daily with fresh fruit (apples, oranges, grapes, or bananas). The animals were housed in a room with 12:12 h light-dark cycle (lights off at 2000) at an ambient temperature of 22 ± 1°C.

Animals were inspected daily to ensure their good health. Food and water intake were carefully monitored. Once every 2 wk, they were immobilized with ketamine sulfate (50 mg im) and medically evaluated.

Surgical and stimulation procedures. The first surgical procedure involved the implantation of electrodes for electrocardiographic (ECG) recording. With animals under general anesthesia (0.5-1.0% isoflurane) and aseptic conditions, three electrodes (pacemaker leads, model 1043K, Pacesetter Systems) were implanted: two into musculus iliacus on the left and right sides and one into the sternum. The electrode wires were tunneled under the skin and externalized at the back of the animal. To ensure proper healing, the monkey was kept in the primate chair during the 1st wk after surgery. At all times, the electrode leads were protected by a primate jacket.

The second surgical procedure took place 3-5 mo later after operant conditioning and collection of baseline data (see below). With animals under general anesthesia (0.5-1.0% isoflurane) and aseptic conditions, an acrylic platform was stereotaxically mounted and cemented in place on the monkey's skull. To ensure the longevity of the implant, the platform was reinforced by four stainless steel screws, which were inserted under the skull (screwheads down) through predrilled keyhole-shaped openings. The platform was 33 × 63 × 10 mm and contained 285 20-gauge stainless steel needles that served as guiding cannulas for subsequent implantation of brain electrodes. The guiding cannulas were oriented in 19 rows so that the distance between cylinders in all directions was 1.5 mm. A stereotaxic vertical zero was referenced to the upper surface of the platform. One of the guiding cannulas in the middle of the platform was referenced as anterior/posterior zero. One of the rows was situated at sagittal zero and served as a lateral reference.

Brain electrodes were implanted as needed according to the stereotaxic atlas of the monkey brain (13) but not earlier than 4 wk after implantation of the platform. Under ketamine sedation and aseptic conditions, with the monkey seated in the primate chair, the skull was drilled through one of the guiding cannulas, and a concentric electrode (60 mm × 0.5 mm with a 0.5-mm uninsulated tip; Rhodes Medical Instruments) was gradually lowered into the brain (1-3 mm/step). The precise location was selected based on the presence of a HR response to electrical stimulation. Constant-current electrical stimulation consisted of 0.5-s trains, 1 train/s, of bipolar square waves, with 45 impulses/s frequency, 2.5-ms duration of each wave, at an intensity 0.2-0.4 mA. The electrode was cemented in place. Specific sites are described in detail in RESULTS.

Apparatus and physiological measurements. While the monkey was in the primate chair, its ECG electrodes were connected to a polygraph. The ECG signal was conditioned electronically and fed into a microprocessor, which computed the R-R intervals (on a beat-to-beat basis) and controlled the experiment.

Exercise and HR slowing procedure. All animals were trained to slow their HR, then to exercise, and, finally, to slow HR while exercising (see Ref. 14 for details of training).

The data reported in this paper were obtained from experiments during which the animals either exercised only (Ex) or were doing the combined task of slowing HR and exercising (Comb). The exercise task consisted of requiring the animal to lift weights to avoid tail shock. The system was designed so that an animal had to lift the weight (12 kg) 4.5 cm and lower it before lifting it again, at least once every 6 s.

During testing, a panel with a set of three lights was placed in front of the monkey. Before each type of session, baseline HR was recorded continuously for 256 s. During baseline, no cues were made available to the animal. At the completion of baseline, the animal was cued to perform in either the Comb or Ex session as noted above. An Ex session was signaled by a white light, which was on continuously throughout the session. The animal was required to lift the weight at least once every 6 s to avoid a 0.45-s electrical shock to the tail. The intensity of electrical shocks was selected for each animal to reinforce the avoidance but never exceeded 10 mA. If the animal did not lift the weight within 2 s, clicks were delivered at the rate of 2/s for the next 4 s and then the shock occurred.

Comb sessions were signaled for exercise and for HR slowing. The exercise procedure was identical to that for the Ex sessions described above. Additionally, the HR slowing procedure was signaled by a red light that cued the animal to slow HR; this light was on continuously throughout the session. In addition to the red light, a yellow light signaled the animal about its HR performance; the criterion for correct performance was set so that the animal needed to maintain its HR at or below the average HR achieved during Ex sessions. If the animal was performing to criterion, the yellow light was on; if HR was above criterion than the light went off and remained off until the animal returned its HR to or below the criterion. During the period when the light was off, the animal was vulnerable to receive a tail shock every 8 s until the light came back on. The shock given for failing to reduce HR was identical to that given for failing to exercise properly and was given through the same set of electrodes placed on the tail of the animal. During the training procedure, the program was adjusted in such a way that if animal received too many shocks, the criterion was automatically moved to an easier level. Fully trained animals rarely received shocks.

Experimental design. After animals were fully trained to exercise and to attenuate the tachycardia of exercise, the first brain electrode was placed, and experiments with brain stimulation were started. Animals were tested in three kinds of experimental sessions: 1) 10 no-feedback sessions, during which brain stimulation occurred according to the protocol given below (animals neither exercised nor slowed HR during any of these sessions); 2) 20 exercise-only (Ex) sessions, during which animals exercised and received brain stimulation according to protocol but were not required to slow HR; and 3) 20 combined exercise and HR slowing (Comb) sessions, during which animals were required to exercise and to slow HR and received brain stimulation according to protocol. Each no-feedback, Ex, and Comb session consisted of a 256-s baseline period followed by a 1,024 s (~17.5 min) continuous experimental period. We segmented the 1,024-s period into 32-s blocks so that brain stimulation was delivered every fourth block. Thus there were seven stimulation periods per session, each preceded by 96-s of no stimulation. Each animal usually participated in five daily sessions: one no-feedback session followed by two Ex and two Comb sessions. Furthermore, the Ex and Comb sessions were counterbalanced over days. The results for each point in the brain are based on data from a total of 20 Ex and 20 Comb sessions. After a study, i.e., 40 sessions, with one brain site completed, a new brain electrode was introduced. Before the beginning of the study with a new electrode, performance in Ex and Comb sessions without electrical brain stimulation was reestablished and confirmed statistically over several days.

After the completion of all studies (no more than 9 electrodes for one animal), the animals were killed with an overdose of pentobarbital sodium and perfused through the carotid arteries with physiological saline and then with 20% formaldehyde. The brain was removed and cut at 50-µm increments in the coronal plane on the freezing microtome. Every 10th section was mounted and stained by Nissl substance to trace the electrode tracks and location of the tip. Stimulated brain sites were identified by reconstructing electrode tracks using stereotaxic atlas of the monkey brain (13). Thalamic nuclei were identified using the stereotaxic atlas of monkey thalamus (11) and sagittal cytoarchitectonic maps of Macaca mulatta thalamus (7).

Statistical methods. For each monkey, differences between average HR change from baseline attained during Ex and Comb sessions without stimulation were compared by t-tests. For this analysis, the HR change was divided by the number of pulls during each segment. In no-feedback brain stimulation sessions, the HR during the 32-s stimulation period (Stim) was compared with the 32-s nonstimulation period immediately preceding stimulation (Nonstim) using the t-test. For comparison of Ex and Comb sessions with brain stimulation the HR change from the relevant baseline period was presented to analysis of covariance (ANCOVA, P2V program of BMDP statistical package). The model for this analysis was two (Stim, Nonstim) by two (Ex, Comb). The number of weight lifts was a covariate. Results for each brain site are reported separately. Throughout this report, significance is defined as P < 0.05.

RESULTS

Before the beginning of brain stimulation experiments, each of the four experimental animals was successfully trained to attenuate the tachycardia of exercise. Table 1 presents the average changes (Delta ) in HR per weight lift and the average number of lifts during Ex and Comb sessions. For each animal, Delta HR per weight lift increased significantly less during Comb sessions than during Ex sessions, despite the fact that two animals (nos. 1 and 4) pulled less often during Comb sessions (for animals 2 and 3, no. of pulls was similar in both sessions). Delta HR for all four animals during Ex was 25.8 beats/min and during Comb was 13.8 beats/min, a difference of 12.0 beats/min.

Table  1.   Changes in heart rate per weight lift and number of lifts during exercise only and combined sessions before brain electrode implantation
Animal No. Exercise
Combined
df t (HR) t (lifts)
Heart rate, beats · min-1 · lift-1 No. of lifts Heart rate, beats · min-1 · lift-1 No. of lifts
1  -0.68 ± 1.10  18.72 ± 2.07   -2.00 ± 2.30  15.18 ± 2.83  36 2.21* 4.36*
2 2.57 ± 0.66  25.50 ± 0.36  2.14 ± 0.68  25.50 ± 0.90  38 2.04* 0.01
3 1.39 ± 0.76  18.91 ± 1.68  0.94 ± 0.75  19.11 ± 1.35  38 1.86* 0.41
4 1.26 ± 0.68  19.02 ± 1.79  0.78 ± 0.81  16.48 ± 1.42  38 2.01* 6.14*
Values are means ± SD. HR, heart rate. * Significant difference between exercise only and combined conditions, P < 0.05.

A total of 24 brain sites were tested. Most of the sites were located in the thalamus and limbic system (18 points): nucleus ventralis anterior (VA), mediodorsal nucleus (MD), ventromedial complex (VM) and intralaminar nuclei, nucleus hypothalami dorsomedialis (DM), and cingulate cortex (CCr). One electrode was located in the nucleus caudatus (CN) and five in the fiber pathways [corpus callosum (CC) and pedunculus cerebri (PC)]. Figure 1 presents a schematic view of all sites. Each site has an assigned number, which will be used throughout the text.

Fig. 1. Reconstruction of brain coronal sections showing localization of electrode tips in 4 monkeys. Each numbered point represents single stimulated site in brain (animal no. is shown in parentheses). Numerals on left indicate distance in millimeters from interaural plane. black-triangle, Type 1 responses; black-square, type 2 responses; triangle , type 3 responses; bullet , type 4 responses. Sites 1-3, 9, 10, 17, ventral anterior thalamic nucleus (VA); 4, mediodorsal thalamic nucleus (MD); 5, 18, ventromedial thalamic nucleus (VM); 6, 12, pedunculus cerebri (PC); 7, 8, paracentral thalamic nucleus (PrC); 11, cingulate cortex (CCr); 13, central median thalamic nucleus (CM); 14, 15, 24, corpus callosum (CC); 16, 23, central lateral thalamic nucleus (CL); 19, nucleus ventralis posterior inferior thalami (VPI); 20, nucleus hypothalami dorsomedialis (DM); 21, nucleus thalami parafascicularis; 22, caudate nucleus (CN).
[View Larger Version of this Image (49K GIF file)]

Table 2 presents the effects of brain stimulation of different brain sites on HR during the sessions without exercise (No Feedback) as well as Ex and Comb sessions. Table 2 also presents the results of the statistical analyses of the effects of brain stimulation without exercise (t-test) and comparisons of Ex and Comb sessions with and without brain stimulation (ANCOVA). For instance, the third row of Table 2 presents the results of the experiments for site 3 (VA). Electrical stimulation of that site during no-feedback sessions increased HR from 111.2 to 127.0 beats/min (t = 5.67). During Ex sessions, the HR increase was 18.6 beats/min without electrical stimulation and 17.7 beats/min during electrical stimulation. During Comb sessions, HR increase was 0.3 beats/min without electrical stimulation, whereas during electrical stimulation, HR decreased by 3.1 beats/min. ANCOVA revealed a significant main effect of sessions [F(1,38) = 5.94, P < 0.01], a significant main effect of stimulation [F(1,38) = 11.79, P < 0.001], and a significant interaction [F(1,38) = 55.21, P < 0.001]. ANCOVA indicated that HR change was different between Ex and Comb sessions, that HR was changed as a result of electrical stimulation, and that the effect of electrical stimulation on HR was different between Ex and Comb sessions. The high variability in resting HR in the no-feedback columns is attributed to large variations in resting HR among animals.

Table  2.   Effect of electrical stimulation of different brain sites on heart rate during sessions without exercise, exercise only, and combined sessions
Site No. No Feedback
Exercise
Combined
ANCOVA, F Ratio (1,38)
Main effects
Interaction
Nonstim HR Stim HR t(df = 9) Nonstim HR Stim HR Nonstim HR Stim HR Session Stim
1 117.9 ± 16.9  112.0 ± 12.9  3.01* 19.4 ± 3.5  17.9 ± 3.7  7.4 ± 3.8  1.5 ± 3.6  7.74* 22.25* 7.70*
2 121.2 ± 11.2  123.5 ± 8.1  1.65 14.5 ± 3.2  14.7 ± 3.1   -1.2 ± 2.5   -3.9 ± 2.3  18.90* 4.41* 6.10*
3 111.2 ± 19.9  127.0 ± 13.0  5.67* 18.6 ± 3.4  17.7 ± 3.6  0.3 ± 4.9   -3.1 ± 4.9  5.94* 11.79* 55.21*
4 87.2 ± 5.8  94.7 ± 11.9  2.67* 16.0 ± 3.5  12.6 ± 3.9   -3.9 ± 3.6   -2.9 ± 3.4  6.45* 14.48* 7.99*
5 108.8 ± 21.4  112.1 ± 20.3  2.59* 11.8 ± 3.3  21.2 ± 3.4   -6.2 ± 5.1   -0.5 ± 4.5  11.68* 67.95* 3.85
6 108.2 ± 20.4  125.2 ± 15.8  4.18* 9.6 ± 4.3  16.0 ± 3.6   -9.1 ± 3.0   -4.9 ± 3.9  14.67* 32.70* 1.34
7 138.0 ± 19.3  162.1 ± 15.9  6.88* 1.2 ± 4.5  4.4 ± 4.5   -6.2 ± 3.9  4.0 ± 4.8  0.45 6.91* 1.87
8 131.5 ± 13.2  144.8 ± 11.9  7.26* 6.7 ± 3.2  10.9 ± 3.4   -0.8 ± 4.0  3.7 ± 4.4  1.91 60.95* 0.11
9 96.7 ± 5.2  94.7 ± 5.7  2.94* 58.7 ± 3.8  59.1 ± 4.0  41.3 ± 2.8  37.2 ± 2.6  17.60* 21.00* 29.40*
10 104.5 ± 7.5  101.1 ± 7.2  6.23* 51.0 ± 4.1  53.0 ± 4.1  32.0 ± 2.2  30.2 ± 2.1  20.70* 0.03 9.95*
11 101.5 ± 13.3  107.4 ± 15.6  2.01 59.8 ± 3.9  61.0 ± 3.8  33.3 ± 4.3  31.8 ± 4.2  23.42* 0.18 11.71*
12 103.8 ± 10.4  116.4 ± 15.1  5.62* 62.8 ± 3.1  69.0 ± 3.0  43.1 ± 3.5  49.5 ± 3.7  16.98* 64.05* 0.01
13 106.9 ± 25.4  114.5 ± 31.7  3.04* 37.0 ± 3.3  39.0 ± 3.3  35.4 ± 2.7  38.1 ± 2.3  0.09 11.93* 0.31
14 105.0 ± 7.0  107.8 ± 7.7  1.56 22.8 ± 2.7  27.2 ± 3.1  19.5 ± 1.4  23.2 ± 2.3  1.26 10.85* 0.07
15 112.8 ± 13.5  119.7 ± 12.4  2.16 29.8 ± 2.4  34.3 ± 3.3  30.3 ± 2.8  33.0 ± 3.8  0.01 10.69* 0.75
16 103.9 ± 11.3  101.9 ± 11.6  4.26* 36.2 ± 4.5  36.4 ± 4.7  21.1 ± 3.6  21.0 ± 3.6  6.73* 0.02 0.07
17 166.9 ± 19.9  166.3 ± 16.2  0.29 14.2 ± 4.3  15.2 ± 4.5   -6.6 ± 3.9   -7.2 ± 3.9  13.41* 0.23 4.44*
18 160.9 ± 11.4  168.6 ± 10.2  5.13* 28.1 ± 3.9  30.5 ± 4.3  8.1 ± 3.0  1.7 ± 3.2  14.10* 15.70* 0.60
19 154.5 ± 25.8  151.2 ± 28.3  2.13 18.2 ± 2.9  14.5 ± 2.9  5.2 ± 2.8  3.2 ± 2.9  8.97* 35.30* 3.41
20 142.9 ± 11.0  156.2 ± 12.7  3.07* 18.5 ± 2.7  19.7 ± 2.7  14.5 ± 3.3  16.0 ± 3.4  6.81* 4.53* 0.36
21 154.1 ± 15.4  164.6 ± 16.5  2.22 14.5 ± 4.5  18.3 ± 5.4   -1.0 ± 4.8  10.2 ± 8.5  2.04 8.73* 2.13
22 135.6 ± 6.7  138.3 ± 7.9  2.20 18.5 ± 2.7  19.1 ± 2.7  14.5 ± 3.3  16.0 ± 3.4  0.68 4.63* 0.76
23 145.2 ± 14.8  151.6 ± 16.6  5.42* 24.3 ± 3.2  23.3 ± 3.3  14.0 ± 2.1  13.4 ± 2.0  6.81* 4.53* 0.36
24 132.1 ± 7.6  140.2 ± 9.8  4.13* 14.6 ± 2.3  14.9 ± 2.4  14.3 ± 2.4  14.8 ± 2.2  0.00 1.93 0.07
For sessions without exercise (no feedback) HR are given as means ± SD; for sessions representing exercise only and combined exercise and lowering of HR, HR are given as mean difference ± SD. F ratio is based on analysis of covariance (ANCOVA). * P < 0.05.

Electrical stimulation of most of the sites without exercise caused an increase of HR (12 points). Activation of three sites in VA (sites 1, 9, and 10) and site 16 in central lateral thalamic nucleus (CL) evoked a reduction of HR in the control conditions. Stimulation of eight sites (sites 2, 11, 14, 15, 17, 19, 21, and 22) produced biphasic effects on HR, which resulted in statistically insignificant results.

ANCOVA of the effects of brain stimulation in eight sites (7, 8, 13-15, 21, 22, and 24) revealed a nonsignificant main effect for the type of the session (Ex vs. Comb), indicating that electrical stimulation suppressed the differences in HR between Ex and Comb sessions. In the remainder of the brain sites, the main effect of the type of the session was significant; i.e., the differences between Ex and Comb sessions were preserved during electrical stimulation.

For five sites (10, 11, 16, 17, and 24), ANCOVA revealed a nonsignificant main effect of brain stimulation (Stim vs. Nonstim), indicating that, during both Ex and Comb sessions, the effect of electrical stimulation on HR was absent. For the remainder of the sites, the main effect of stimulation was significant.

In eight sites (1-4, 9-11, and 17), statistical analyses revealed a significant interaction between stimulation condition and session type, indicating that electrical stimulation affected HR differently in Ex and Comb sessions. For the remainder of the sites, there were no significant interactions; i.e., electrical stimulation affected HR similarly in both kinds of sessions.

Fig. 2. Central command in different brain sites showing 4 types of effects of electrical brain stimulation (BS) on heart rate (HR) during learned cardiovascular adjustment to exercise. Shaded columns, effect of BS on HR during no-feedback sessions (in absence of exercise or HR adjustment to exercise); filled columns, HR increase in exercise (Ex) sessions; open columns, HR increase in combined (Comb) sessions (operant HR adjustment during exercise). Type 1, effects of BS on HR are additive to effects of Ex and Comb sessions; type 2, effects of BS on HR are additive, but effects of Ex and Comb sessions are equal; type 3, effects of BS on HR are reduced in both Ex and Comb sessions; type 4, effects of BS on HR are additive in Ex sessions but are reduced in Comb sessions.
[View Larger Version of this Image (18K GIF file)]

DISCUSSION

The concept of central neural initiation of the cardiovascular adjustments to exercise (central command) has appeared in the physiological literature for at least 100 years (8). At the same time, parallel concepts characterizing experience, namely learning, as a factor in the expression of cardiovascular responses (see Ref. 1 for a review) has also flourished. However, these two conceptual models, though implicitly similar, have not merged. This study was designed to fill this lacuna. In previous research, we have shown that monkeys can be trained to attenuate HR at rest (2) and during exercise (14). Furthermore, we have shown that the ability to attenuate HR can be expressed, even in the presence of electrical stimulation of the brain, which normally elicits a tachycardia (9). The present study integrates these findings by using electrical brain stimulation and learned control of HR during exercise to identify brain regions that mediate central command.

Fig. 3. Schematic diagram of possible organization of cardiovascular participation in exercise (see details in text).
[View Larger Version of this Image (29K GIF file)]

On the basis of the statistical analyses, it is possible to establish four types of effects of brain stimulation on learned cardiovascular adjustments to exercise. These are illustrated diagrammatically in Fig. 2. Type 1 comprises sites with statistically significant main effects of the type of the session (Ex vs. Comb) and brain stimulation (Stim vs. Nonstim) and nonsignificant interactions. Functionally, the effects of brain stimulation on HR were additive to the tachycardia of exercise or operant attenuation of the tachycardia of exercise. In type 2 sites, the effects of electrical stimulation were also imposed on the tachycardia of exercise, but learned attenuation of exercise disappeared. Statistically this type was expressed by a significant effect of stimulation, but nonsignificant effect of the session type; the interaction was also nonsignificant. Type 3 effects were statistically expressed in a significant main effect of the session type, but nonsignificant main effect of stimulation and nonsignificant interaction. Functionally, the effect of electrical stimulation was HR attenuation in both Ex and Comb sessions. In type 4 sites, the main effects of session type and the interaction were always significant, whereas main effects of stimulation were significant for some sites and nonsignificant for others. Functionally, in Comb sessions, the effect of electrical stimulation that induced an increase of HR was attenuated, and the effect of electrical stimulation that induced a decrease of HR was accentuated.

Type 1 effects were observed during stimulation of the brain sites located in DM (site 20), VM (sites 5 and 18), nucleus ventralis posterior inferior thalami (VPI; site 19), and PC (sites 6 and 12). Stimulation of most of the sites caused an increase of HR. Only stimulation of the VPI (site 19) produce a decrease of HR.

Type 2 effects were observed during electrical stimulation of brain sites in intralaminar thalamic nucleus [2 points in nucleus paracentralis (PrC), sites 7 and 8; nucleus central median (CM), site 13; and nucleus parafascicularis, site 21], CN (site 22), or CC (sites 14, 15, and 24). Stimulation of all these sites, in the absence of exercise, produced an elevation of HR.

Type 3 effects were observed during electrical stimulation of brain sites in CL (sites 16 and 23). Electrical stimulation of brain sites of three areas was responsible for the type 4 effects: MD (site 4), VA (sites 1-3, 9, 10, and 17), and CCr (site 11). Stimulation of the site in MD (site 4) produced an elevation of HR in the control condition. Activation of one site in VA (site 3) produced tachycardia, but three sites in VA (sites 1, 9, and 10) evoked the reduction of HR in the control condition. Stimulation of CCr site (site 11) and two sites in VA (sites 2 and 17) produced biphasic effects in the control condition.

Figure 3 is the model we are proposing to characterize the effects seen in this study. Type 1 sites act directly on HR control centers to modulate HR changes that are initiated elsewhere. Stimulation in these sites increased HR equivalently in Ex and Comb sessions and appears to play no role in central command. Type 2 sites also act directly on HR control centers. However, they have the added effect of abolishing the ability of the animal to control HR voluntarily. These sites probably play a role in central command, but it is not clear from the present studies whether they override learned control of HR or inhibit it. Another possibility is that they interact with the punitive effects of tail shock, so that it is no longer effective as an aversive stimulus. Behaviorally, this last effect would be called extinction. Type 3 sites interact with the electrical stimulation so as to attenuate its effects similarly during Ex and Comb sessions. It is not clear whether the cardiac effects of these sites are overridden whenever the subject is behaving because the performance factors, i.e., exercise with or without learned HR attenuation, block their direct effects on the HR control centers or because the contingency (tail shock) interacts with their effects in operant conditioning integrative centers, as we have hypothesized. Thus their role in central command is uncertain. Finally, type 4 sites show different effects dependent on whether Ex or Comb is operative: during Ex, the electrical brain stimulation adds to the tachycardia of exercise; however, during Comb, stimulation is significantly attenuated. In contrast to type 2 and 3 sites, brain stimulation in type 4 sites is unlikely to interact with the tail shock contingency, since the HR response differs with the experimental condition during type 4 site stimulation but is similar during type 2 and 3 site stimulation. Therefore, according to our model, type 4 sites act on the operant conditioning integrative centers and play a role in central command.

In a normal setting, exercise is a behavioral phenomenon, initiated and maintained by environmental contingencies. Central command is a synonym for this reality. We believe that this study has provided a model for studying some of its neural mechanisms.

ACKNOWLEDGEMENTS

The authors express their gratitude to Dr. M. Mishkin, Dr. E. Murray, and T. Fobbs (Laboratory of Neuropsychology, National Institute of Mental Health) for help with histological verification of brain sites.

FOOTNOTES

   Present address of S. Chefer: Brain Imaging Sect., National Institute of Drug Abuse, 5500 Nathan Shock Dr., Baltimore, MD 21224.

Address for reprint requests: M. Talan, Lab. of Cardiovascular Sciences, Gerontology Research Center, 4940 Eastern Ave., Baltimore, MD 21224.

Received 24 February 1997; accepted in final form 18 June 1997.

REFERENCES

1. Engel, B. T., D. E. Anderson, and M. I. Talan. Assessment of cardiovascular responses as behavior. In: Methods in Neurosciences. Paradigms for the Study of Behavior, edited by P. M. Conn. Orlando, FL: Academic, 1993, vol. 14, p. 396-406.
2. Engel, B. T., and S. H. Gottlieb. Differential operant conditioning of heart rate in the restrained monkey. J. Comp. Physiol. Psychol. 73: 217-225, 1970[Medline].
3. Engel, B. T., and M. I. Talan. Operant conditioning of the cardiovascular adjustments to exercise. In: Perspectives on Research in Emotional Stress. Systems Research in Physiology, edited by P. Simonov. Moscow: Academy of Science of the USSR, 1989, vol. 3, p. 353-366.
4. Engel, B. T., and M. I. Talan. Hemodynamic and respiratory concomitants of learned heart rate control during exercise. Psychophysiology 28: 225-230, 1991[Medline].
5. Engel, B. T., and M. I. Talan. Autonomic blockade does not prevent learned heart rate attenuation during exercise. Physiol. Behav. 49: 373-382, 1991[Medline].
6. Engel, B. T., and M. I. Talan. Cardiovascular responses as behavior. Circulation 83, Suppl. II: II-9-II-15, 1991.
7. Ilinsky, L. A., and K. Kutlas-Ilinsky. Sagittal cytoarchitectonic maps of the Macaca mulatta thalamus with a revised nomenclature of the motor-related nuclei validated by observation on their connectivity. J. Comp. Neurol. 262: 331-364, 1987. [Medline]
8. Johansson, J. E. Ueber die Einwirkung der Muskelthatigkeit auf die Athmung und die Herzthatigkeit. Skand. Arch. Physiol. 5: 20-66, 1985.
9. Joseph, J. A., and B. T. Engel. Instrumental control of cardioacceleration induced by central stimulation. Science 214: 341-343, 1981[Abstract/Free Full Text].
10. Krogh, A., and J. Lindhard. The regulation of respiration and circulation during the initial stages of muscular work. J. Physiol. (Lond.) 47: 112-136, 1913.
11. Olszevski, J. The Thalamus of the Macaca mulatta. An Atlas to Use with Stereotaxic Instrument. Basel: Karger, 1952.
12. Rowell, L. B. What signals govern the cardiovascular responses to exercise? Med. Sci. Sports Exercise 12: 307-315, 1980[Medline].
13. Snider, R. S., and J. C. Lee. A Stereotaxic Atlas of the Monkey Brain (Macaca mulatta). Chicago, IL: Univ. of Chicago Press, 1961.
14. Talan, M. I., and B. T. Engel. Learned control of heart rate during dynamic exercise in nonhuman primates. J. Appl. Physiol. 61: 545-553, 1986[Abstract/Free Full Text].



This article has been cited by other articles:


Home page
 BrainHome page
H. D. Critchley, C. J. Mathias, O. Josephs, J. O'Doherty, S. Zanini, B.-K. Dewar, L. Cipolotti, T. Shallice, and R. J. Dolan
Human cingulate cortex and autonomic control: converging neuroimaging and clinical evidence
Brain, October 1, 2003; 126(10): 2139 - 2152.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chefer, S. I.
Right arrow Articles by Engel, B. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chefer, S. I.
Right arrow Articles by Engel, B. T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online