INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ARTERIAL BARORECEPTOR AFFERENTS are known to elicit a baroreflex through which arterial pressure (AP) is rapidly controlled via a negative feedback. On the basis of data from several animal species (3, 27), baroreceptors are clearly important in stabilizing AP but appear to have considerably less importance with regard to long-term mean AP (MAP) level. Here, we consider the contribution of baroreceptors to the changes in MAP that accompany physical activity.

It is known (14, 24, 26) that MAP increases immediately after exercise is started. This reflective pressure increase is referred to as an exercise pressor reflex or response (17) and is well known to be mediated by two mechanisms. One mechanism is a centrally generated motor command signal called "central command" (13, 16) that promotes sympathoexcitation and an increase in MAP and is speculated to be generated at locomotor centers probably located in the midbrain and a superior central region (2, 30). The other mechanism, called the "muscle chemoreflex" and "muscle mechanoreflex," is the afferent signals that originate from working muscles, enhancing sympathetic outflow to increase MAP (14, 24, 26). Therefore, muscular exercises induce a pressor response due to an increase in sympathetic outflow (18, 20, 33) via the central command and/or muscle chemoreflex (14, 24, 26). The pressor response is thought to play an important role in increasing the perfusion pressure to working muscles during exercise. On the other hand, the baroreflex system is considered to stabilize these exercise pressor responses (6, 28, 29). If baroafferent information supports only the stabilization of MAP via the above baroreflexive negative feedback mechanism, then a pressor response will become greater in sinoaortic-denervated (SAD) animals because of a lack of baroreflex. However, several studies (5, 6, 15, 34) have showed that the exercise pressor response is abolished by complete deafferentation of arterial baroreceptors. This suggests that baroafferent signals might not only serve as a negative feedback control but also actively support the exercise pressor response to control blood pressure appropriately.

Dynamic exercises such as running on a treadmill or pedaling on an ergometer require intense muscular activity. This causes consumption of large amounts of oxygen and energy, production of a large quantity of lactate and other acids, and the loss of body fluids through sweating. These reactions complicate the mechanisms of the pressor response, which includes, among others, the exercise pressor response, a chemoreceptor reflex, the energy-supplying responses, and the body fluid-controlling responses. Static exercises such as handgrip or isometric muscle contraction do not require such intense physical activity and activate only local muscles for a short period. This type of muscular activity may be too brief and localized to allow analysis of the 24-h blood pressure control system. However, routine physical activity, including body posture changes, eating, drinking, urination, defecation, walking, and grooming, is neither as intense as dynamic exercise nor as brief and localized as handgrip exercise. We therefore used routine physical activity as a physiological perturbation to the 24-h blood pressure-regulating system.

On the basis of the conventional concept that the baroreflex system just stabilizes the exercise pressor response, it may be conjectured that SAD shifts a MAP histogram toward a higher pressure level during physical activity (Fig. 1). To challenge the new concept of arterial baroafferent function in the 24-h blood pressure-regulating system rather than in negative feedback control, we hypothesized that SAD would abolish such a shift of a MAP histogram during daily physical activity. To test this hypothesis, we analyzed the data obtained by 24-h recording of MAP, heart rate (HR), and cardiac output from acclimatized free-moving rabbits as variables. The following three phases were set up based on observed MAP, HR, and cardiac output values, taking observed electromyogram (EMG) activities and reference hydrostatic pressures (RPs) into account: 1) a quiet rest phase (rest phase), 2) an active movement phase (movement phase), and 3) the "other phase" (other phase). The variables were then expressed as histograms for each phase.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Diagram showing conjectural changes in a 24-h mean arterial pressure (MAP) histogram in baroafferent-intact (intact) and sinoaortic-denervated (SAD) animals.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Twenty-eight Japanese White rabbits weighing 2.5-3 kg, of either sex, were obtained from Inoue Laboratory Animal Center (Kumamoto, Japan), housed in individual cages, and allowed free access to food and drinking water. All rabbits were chronically prepared with sterile surgery and used in a conscious state for experiments. All experimental procedures were conducted in accordance with the guidelines of the National Defense Medical College Committee on Animal Care and Use and the Physiological Society of Japan.

Surgery. Acclimated rabbits were anesthetized with subcutaneous injection of the anesthetic mixture [50 mg/kg ketamine (Sankyo Inc, Tokyo, Japan), 5 mg/kg chlorpromazine hydrochloride (Wako Pure Chemical Industries, Osaka, Japan), and 5 mg/kg xylazine hydrochloride (Sigma Aldrich, Tokyo, Japan)], intubated, and mechanically ventilated with room air. Sinoaortic denervation was performed according to a previously described method (19, 27). Briefly, through a midcervical incision, the aortic nerve, carotid sinus nerve, and cervical sympathetic nerve were identified and cut bilaterally with the aid of a surgical microscope. For sham operation of SAD to "intact" rabbits, the aortic and carotid sinus nerves were identified but not cut. SAD was judged to be complete when HR did not show any change in response to the MAP changes induced by intravenous injections of phenylephrine and nitroglycerine in the conscious state 2 days after catheterization. After a 2-wk recovery period, the rabbits were anesthetized again, and Silastic catheters were implanted in the abdominal aorta via the right femoral artery for measurement of arterial blood pressure and in the right external jugular vein for infusion of drugs. A plastic catheter was placed on the back region between the scapulae for recording RP to indicate the height of the rabbit's back. The tip of this catheter was positioned 1 cm away from the skin. Electrodes were placed on the nuchal muscles for recording the EMG of the posterior cervical muscles. These three catheters and electrodes were tunneled subcutaneously and exteriorized in the occipital region between the ears. All vascular catheters were heparinized every 2 days until the start of the experiments. Thirteen rabbits underwent left thoracotomy through the second intercostal space 4 wk before SAD or sham operation for placement of a blood flow probe to measure the rate of aortic blood flow (AoBF). An AoBF probe was placed on the ascending aorta, and the cable from this probe was tunneled and exteriorized as described above. Ampicillin (5 mg/kg) was administered for 2 days after each surgery, and pentazocine (1 mg/kg) was administered just after catheterization and thoracotomy.

Recording procedures. The rabbits prepared chronically were housed in an individual cage (35 cm wide × 50 cm deep × 33 cm tall) in an isolated room with a 12:12-h light-dark cycle (lights on at 7 AM). All catheters and electrodes and the flow probe cable were extended and covered with a lightweight coiled spring to prevent dislodging and damaging. The cover spring was suspended from the ceiling of the rabbit cage so as to allow 360° rotation. The rabbits could move freely inside the cage. The venous catheter was extended, plugged, and left free until an injection experiment started. The arterial catheter was extended and connected to a cannula swivel fixed at the end of the cover spring. An extension tube from the cannula swivel was connected to a pressure transducer (DX-360, Nihon Koden, Tokyo, Japan). AP, MAP, and HR were measured using a polygraph system (Nihon Koden). The RP catheter was extended and connected to the second cannula swivel, and an extension tube from this second cannula swivel was connected to a second pressure transducer (DX-360, Nihon Koden). MAP and mean RP (mRP) values were obtained using a filter with a time constant of 2 s. Both pressure transducers were fixed on the wall at the heart level of a prone rabbit. Raw EMG signals were amplified at a bandwidth of 50-3,000 Hz using an amplifier (AB-651-J, Nihon Koden), rectified, and integrated by an analog device. The mean EMG (mEMG) value was obtained from the signal, rectified, and integrated using a filter with a time constant of 2 s. AoBF was measured using a transit time flowmeter (T101, Transonic Systems). The mean AoBF (mAoBF) value was obtained using a filter with a time constant of 2 s. The pulsatile AP, MAP, beat-to-beat HR, mRP, raw EMG, mEMG, and mAoBF were recorded on an analog thermal recorder (Nihon Koden; Fig. 2). Outputs from the analog recorder were digitized by the analog-digital converter (MacLab, AD Instruments, Castle Hill, Australia) at a sampling rate of 2 samples/s. These digitized values were displayed and recorded using a Macintosh personal computer (see Figs. 3 and 4).

View larger version (49K): [in this window] [in a new window]   Fig. 2.   Analog recordings of AP, MAP, heart rate (HR), mean aortic blood flow (mAoBF), mean reference hydrostatic pressure (mRP), mean electromyogram (mEMG) of the nuchal muscles and raw EMG in an intact rabbit over a 13-min period. bpm, Beats/min.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Digital chart data obtained from an intact rabbit. Note that EMG, RP, MAP, and HR were low during the rest phase, whereas MAP and HR were apparently kept high with active EMG and high RP during the movement phase. AoBF was not synchronized with either the EMG activity or the RP level.

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Digital chart data obtained from a SAD rabbit. Note that MAP fluctuated and was not synchronized with either the EMG activity or the RP level. AoBF remained relatively high during the movement phase and low during the rest phase.

Experimental protocols. Three days were allowed for rabbits to recover from the last surgery and to acclimatized to a tether system before starting experimental recordings. In all experiments, recordings were started at noon and continued for several days. All recorders and computers were placed in the room next door to the rabbit room. Rabbits were housed, as usual, in a quiet isolated room during a recording period. Each morning, a designated person supplied water and food and cleaned the drawer under the floor of each rabbit cage. Twisted electric lines were untangled every 6-8 h. The arterial catheter line was flushed by continuous infusion of heparinized saline (300 U/ml) at a rate of 4 ml/day using the syringe pump (STC-523, Terumo, Tokyo, Japan). The data obtained by 24-h recording were downloaded to another personal computer for analysis at about noon each day.

Data analysis. Data digitized for 24 h were displayed on a computer monitor as multichannel chart recordings (see Figs. 3 and 4). A total of 172,800 sampling points was obtained for the data from one channel per day. Ten digitized values each for MAP, HR, AoBF, mEMG, and mRP were transferred to a spreadsheet (17,280 points/channel). Measured MAP values included not only the real MAP but also the hydrostatic pressure when a rabbit stood up, because the pressure transducer was fixed at the heart level of a prone rabbit as described above. The value of real MAP was thus calculated by subtracting RP from measured MAP. TPR was calculated by dividing the calculated MAP by mAoBF. The mEMG value was expressed as the percentage of 24-h EMG data. The values of calculated MAP, HR, AoBF, and TPR for 24 h were individually converted into frequency histograms from which the mode of each variable in 24-h data was obtained. The average and standard deviation of 24-h data were calculated for each variable. The results obtained by the analysis of 24-h data were considered to be for "all phases."

Statistical analysis. Summed data for the mode, average, or standard deviation of each variable from 13 or 15 rabbits were expressed as means ± SE. All statistical analyses were performed using a statistical package for Macintosh computers (StatView-J, version 4.11, and SuperANOVA, version 1.11, Abacus Concepts, Berkeley, CA). Statistical significance was analyzed using two-way ANOVA followed by contrasts (4). P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Rest and movement phases. The rest or movement phase was selected from the 24-h data based on EMG and RP features and the definitions explained in METHODS. Summed data for mEMG or mRP from 15 intact or 13 SAD rabbits in the rest and movement phases are shown in Table 1. No significant difference was observed between intact and SAD rabbits in mEMG or mRP in the rest, movement, and other phases.

MAP and HR. In intact rabbits, MAP and HR apparently increased in the movement phase and were stabilized in the rest phase (Fig. 3). In SAD rabbits, MAP fluctuated and was not synchronized with EMG activities or RP levels (Fig. 4). HR was relatively constant and not synchronized with EMG activities or RP levels. The histograms prepared from 24-h data from one intact rabbit and one SAD rabbit are depicted in Figs. 5 and 6, respectively. The MAP histogram in the movement phase shifted toward a higher pressure level in intact rabbits compared with the histogram in the rest phase, whereas no such shift from the rest phase was observed in SAD rabbits. HR histograms in the movement phase shifted to a higher pressure in both intact and SAD rabbits.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Histograms of 24-h MAP (in mmHg; A), HR (in bpm; B), AoBF (in ml/min; C), and total peripheral resistance (TPR; in mmHg · min · ml1; D) in an intact rabbit, calculated from 24-h data for all phases (All), the rest phase (Rest), and the movement phase (Move). Values are means ± SD of 24-h data; n refers to the no. of data point samples. Physical activity shifted the MAP, HR, and TPR histograms but did not affect the AoBF histogram.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Histograms of 24-h MAP (in mmHg; A), HR (in bpm; B), AoBF (in ml/min; C), and TPR (in mmHg · min · ml1; D) in a SAD rabbit. Physical activity did not shift the MAP or TPR histograms but did shift the AoBF and HR histograms.

AoBF and TPR. To determine whether cardiac output or TPR was responsible for movement-induced increase in MAP in intact rabbits, AoBF was measured over 24 h concomitantly with MAP and HR in an additional seven intact and six SAD rabbits. As depicted in Figs. 3 and 4, mAoBF was not synchronized with EMG activities or RP levels in intact rabbits (Fig. 3) and appeared to be slightly synchronized with the EMG activities or RP levels in SAD rabbits (Fig. 4). Physical activity did not shift the AoBF histogram but did shift the TPR histogram to a higher pressure in intact rabbits (Fig. 5). In SAD rabbits, physical activity slightly shifted the AoBF histogram to a higher pressure level but did not shift the TPR histogram (Fig. 6).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major findings of this study are as follows: 1) daily physical activity increased the mode and the average value of MAP by 15.4 ± 0.6 and 15.9 ± 1.4 mmHg, respectively, through an increase in TPR in arterial baroafferent-intact rabbits; 2) SAD augmented the standard deviation of 24-h MAP and abolished the resetting of pressure; and 3) physical activity slightly decreased TPR and slightly increased cardiac output in SAD rabbits and increased TPR but did not change cardiac output in baroafferent-intact animals. These findings support our hypothesis that SAD may not augment the shift of a MAP histogram to a higher-pressure level but rather abolish the shift of a MAP histogram during daily physical activity. The data from other research groups, our analyses of the observations, and the possible mechanisms of the above phenomena are discussed below.

SAD did not augment the movement-induced pressor response but abolished it. A similar phenomenon has also been observed in exercise experiments using dogs (15) and rabbits (6).

Daily physical activity increased TPR and MAP but did not affect cardiac output in baroreceptor-intact (Intact) rabbits (Table 2). This indicates that the increase in MAP during daily physical activity may be caused by an increase in TPR but not by an increase in cardiac output. The mechanism for these responses has not been clarified. However, the following discussion might suggest a part of the mechanism.

MAP histograms in SAD rabbits (Table 2) show that daily physical activity slightly increased the cardiac output and slightly decreased TPR, resulting in little change in MAP. These data suggest that movement entailing weak but wide-ranged muscle activity may induce local vasodilatation in activated muscles and a consequent decrease in TPR. The decrease of TPR may increase venous return; hence, cardiac output may increase. The combination of a decrease in TPR and an increase in cardiac output may result in no apparent change in MAP. In baroreceptor-intact rabbits, on the other hand, MAP histograms showed that physical activity increased TPR and MAP but did not change cardiac output. On the basis of these findings, it seems likely that the pressor responses induced by the central command and/or muscle chemoreflex can be realized only when the baroreceptor is intact. The postulated mechanisms are discussed below.

Once central command and/or muscle chemoreflex is realized, activated sympathetic outflow evokes neural vasoconstriction in the peripheral vascular beds that are not in active muscles, resulting in an increase in TPR. The increased TPR then promotes resetting of MAP to a higher-pressure level. When MAP increases, the cardiac afterload is increased and cardiac output is increased only slightly. Because of this series of hemodynamic changes, blood pressure is increased concomitantly with vasodilatation specifically in activated muscles in baroafferent intact animals and then the circulating blood is redistributed to support muscle activities. In SAD rabbits, the lack of baroafferent function might prevent expression of the central command and/or muscle chemoreflex and therefore efferent nerve activity, TPR, and MAP do not increase. This results in a lack of blood redistribution for working muscles, which may not be a suitable state for physical activity.

A number of studies (10, 11, 23, 24) have shown that central command signals from the locomotor center and/or muscle afferent signals originating from active muscles reset the baroreflex function. Although the electrophysiological mechanism through which the baroreflex is reset to a higher-pressure level remains unknown, there may be two possible mechanisms. First, the central command and/or muscle chemoafferent signals are transmitted to the baroreceptors and acutely reset the transduction function of the baroreceptors. Second, baroreceptor afferent signals interact with central command signals and/or muscle chemoafferent signals in the baroreflex center, and then the medullary cardiovascular center is facilitated to activate the sympathetic system. Evidence and reports in favor of these two possible mechanisms are discussed below.

It has been demonstrated histologically that baroreceptors are innervated with adrenergic neurons (25) and electrophysiologically that electrical stimulation of the sympathetic nerves innervating the carotid sinus excites the activity of baroafferent nerves (32). Felder et al. (7) have demonstrated that suppression of the sympathetic nerves innervating the carotid sinus decreases the activity of baroreceptor afferent nerves, consequently increasing AP. These data lead to the possibility that the central command and/or muscle chemoreflex directly modulates the transducer function of baroreceptors via a sympathetic pathway. Heesch and Barron (10) have demonstrated that a central mechanism does not mainly contribute to acute resetting of baroreceptors and discussed that resetting of the operating pressure range to a higher pressure may be due mainly to receptor adaptation, whereas the sensitivity (or gain) of baroreflex is possibly due to a central mechanism. The above results and discussion strongly suggest that neural signals to baroreceptors modulate the receptor operating range toward a higher-pressure level, resulting in acute resetting of the baroreflex. To our knowledge, however, there has been no study published about the neural control of the receptor operating range or the baroreceptor sensitivity during extensive exercise or daily physical activity.

Direct electrical stimulation of the mesencephalic locomotor center promotes the increases of MAP and HR even when the skeletal muscle is paralyzed (2). Microinjection of a GABA antagonist into the posterior hypothalamic locomotor center blunts baroreflex resetting (1). Even when the muscle chemoreflex is eliminated, exercises promote a rise in blood pressure (9, 12, 13). These findings suggest that a central command from the locomotor center may also reset the baroreflex independently of a muscle chemoreflex. On the other hand, Potts and Mitchell (23) have demonstrated that activation of skeletal muscle afferent fibers by electrically induced muscle contractions increases a threshold pressure for carotid baroreceptors and consequently induces an exercise pressor response. This finding means that muscle afferent signals reset the baroreflex in the absence of central command motor signals. Both central commands and muscle afferent signals can promote the baroreflex resetting. However, a precise electrophysiological mechanism remains unclear.

If the baroafferent signal supports only a negative feedback control to stabilize blood pressure against any perturbation, then SAD would enhance physical activity-induced pressor responses because of the lack of a stabilizing mechanism. However, SAD abolishes the pressor responses induced by exercises (5, 6, 15, 34). This phenomenon was confirmed to occur not only in the case of physical exercises but also in daily physical activities such as those observed over 24 h in the current study. Even though it is unclear whether a central or receptor mechanism is operating, baroafferent signals are essential for the induction of pressure resetting. The phenomenon that SAD eliminates some physiological function other than the baroreflex was observed not only in exercise-induced pressure resetting but also in some neurohumoral interactions. For example, circulating vasopressin interacts with the area postrema and consequently suppresses the sympathetic outflow, and this function is abolished by SAD (19).

One limitation of the present study is that the SAD procedure damages not only baroreceptor afferents but also chemoreceptor (carotid body) afferents. Our study does not show arterial blood oxygen tension or differentiate whether baroafferents or chemoafferents are more important for the activation of the pressor response. We initially believed that the effect of deafferentiation of peripheral chemoreceptors might be minimal because central chemoreceptors are intact in SAD rabbits and also because the activity of peripheral chemoreceptor afferents is low in rabbits with normoxia and normocapnia (31). Furthermore, we did not use dynamic exercise such as pedaling on an ergometer but instead chose daily physical activity for perturbation to the blood pressure stabilization, which means that the stimulation of chemoreceptors may be not as strong. However, the deafferentation of carotid chemoreceptors in rats produces chronic hypoxemia and hypercapnia and affects the blood pressure control (8). Therefore, there is a possibility that peripheral chemoreceptor afferents may also contribute to the exercise pressor response.

Perspectives