| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Cardiovascular Dynamics, National Cardiovascular Center Research Institute, Suita, Osaka 565-8565; and 2The Organization for Pharmaceutical Safety and Research, Chiyoda-ku, Tokyo 100-0013, Japan
Submitted 27 December 2002 ; accepted in final form 21 April 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
renal sympathetic nerve activity; transfer function; systems analysis; rabbits; carotid sinus baroreflex
In contrast to the BJ reflex, the arterial baroreflex has been established as an important negative feedback system that stabilizes AP against exogenous pressure perturbations. The total baroreflex loop represents the AP response to pressure inputs on the carotid sinuses and aortic baroreceptors. The total baroreflex loop may be divided into the neural and peripheral arc subsystems (6, 22, 25): the neural arc representing signal transduction from baroreceptor pressure input to efferent sympathetic nerve activity (SNA) and the peripheral arc representing the regulatory pathway from SNA to AP. The dynamic characteristics of the two arcs determine the stability and quickness of AP regulation (6, 14). Because the baroreflex operates dynamically under the routine circumstances of daily activity, changes in baroreflex dynamic characteristics would critically affect AP regulation, and consequently the quality of life.
Chen (3) examined the interaction between the BJ and arterial baroreceptor reflexes. In his study, the BJ reflex reduced the steady-state responses of AP and HR to baroreceptor pressure input. However, to the best of our knowledge, the effects of the BJ reflex on the dynamic characteristics of the arterial baroreflex remain unknown. Furthermore, the effects of the BJ reflex on the arterial baroreflex have not been assessed separately with regard to the neural and peripheral arc transfer characteristics. Therefore, to test the hypothesis that the BJ reflex modulates the dynamic characteristics of the total loop, neural arc, and/or peripheral arc of the arterial baroreflex, we performed a baroreflex open-loop experiment by using a white noise method in anesthetized rabbits (6, 13, 26). The results of the present study indicate that the BJ reflex evoked by intravenous PBG administration reduced dynamic gain in the total baroreflex loop, mainly by attenuating dynamic gain in the neural arc.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals were cared for in strict accordance with the "Guiding Principles for the Care and Use of Animals in the Field of Physiological Sciences" approved by the Physiological Society of Japan.
Thirteen Japanese white rabbits weighing 2.6–3.0 kg were anesthetized
with an injection (2 ml/kg iv) composed of a mixture of urethane (250 mg/ml)
and 
-chloralose (40 mg/ml). The rabbits were ventilated artificially
with oxygen-enriched room air. To maintain the appropriate level of
anesthesia, supplemental doses of these anesthetics were administered
continuously (0.2–0.3 ml · kg–1
· h–1 iv). AP was measured using a
high-fidelity pressure transducer (Millar Instruments; Houston, TX) inserted
from the right femoral artery. A double-lumen catheter was introduced into the
right femoral vein for drug administration. We sectioned the aortic depressor
nerves after identifying their arterial pulse-synchronized activity to
eliminate the effects of the aortic baroreflex. The bilateral vagi were kept
intact to preserve the afferent pathway of the cardiopulmonary receptors. We
isolated the bilateral carotid sinuses from the systemic circulation by
ligating the external and internal carotid arteries and other small branches
originating from the carotid sinus regions. The isolated carotid sinuses were
filled with warm physiological saline through catheters inserted via the
common carotid arteries. Intracarotid sinus pressure (CSP) was controlled with
the use of a servo-controlled piston pump. We exposed the left renal
sympathetic nerve retroperitoneally and attached a pair of stainless steel
wire electrodes (Bioflex wire AS633, Cooner Wire) to record renal SNA (RSNA).
The nerve fibers distal to the electrodes were crushed by tight ligature to
eliminate afferent signals from the kidney. To insulate and fix the
electrodes, and to keep the nerve from drying, the nerve and electrodes were
soaked in addition-curing silicone gel (Semicosil 932A/B, Wacker Silicones).
The preamplified nerve signal, band-pass filtered at 150–1,000 Hz, was
then full-wave rectified and low-pass filtered with a cutoff frequency of 30
Hz to quantify nerve activity. We administered pancuronium bromide (0.3 mg/kg
iv) to prevent muscular activity contamination in RSNA recording. Animal body
temperature was kept at around 38°C with a heating pad.
Protocols
Protocol 1. After completion of surgical preparations in seven rabbits, the baroreflex negative feedback loop was closed by adjusting CSP to AP for 20 min. Mean AP (and therefore mean CSP) in the steady state was treated as the operating pressure (Pop). To assess the dynamic characteristics of the carotid sinus baroreflex, we randomly assigned CSP to either high (Pop + 20 mmHg) or low (Pop–20 mmHg) pressure every 500 ms, according to a binary white noise sequence (8–10). The power spectral density of CSP was reasonably constant up to 1 Hz. We recorded CSP, RSNA, AP, and HR for 10 min under control (CTL) conditions. We then administered PBG (100 µg · kg–1 · min–1 iv) for 20 min and recorded CSP, RSNA, AP, and HR for the last 10 min of the PBG administration (PBG condition).
Protocol 2. To confirm vagi involvement in the PBG-induced BJ reflex, we performed intravenous bolus injection of PBG to six of seven rabbits. After a 30-min recovery period from protocol 1, 50 µg iv PBG was injected under the condition of intact vagi. We then sectioned the bilateral vagi and waited until RSNA, AP, and HR reached steady state. Finally, we repeated the bolus injection of PBG under vagotomized condition.
Protocol 3. To examine the time-dependent effects of continuous anesthesia on dynamic characteristics of the arterial baroreflex, we performed an experiment similar to protocol 1 without intravenous PBG administration in six different rabbits. Data were obtained from two experimental runs of 10 min each (CTL1 and CTL2 conditions) with an intervening interval of 10 min.
Data Analysis
Data were sampled at 200 Hz using a 12-bit analog-to-digital converter and
stored on hard disk of a dedicated laboratory computer system. In
protocols 1 and 3, we estimated the total baroreflex loop
transfer function by treating CSP as the input and AP as the output. To
estimate neural arc transfer functions of the carotid sinus baroreflex, we
treated CSP as input and RSNA as output of the system. In the peripheral arc
transfer function, RSNA was the input and AP was the output. We resampled
input-output data pairs at 10 Hz and segmented them into eight sets of 50%
overlapping bins of 1,024 data points each. A linear trend was subtracted and
a Hanning window was applied for each segment. We then performed a fast
Fourier transformation to obtain frequency spectra of the input and output
signals. We ensemble averaged the input power
[Sxx(f)], output power
[Syy(f)], and cross power between input
and output [Syx(f)] over the eight
segments, where f represents frequency. Finally, we calculated the
transfer function [H(f)] from input to output by using the
following equation (17)
 | (1) |
(f)] of the transfer function using the following equations
(17)
 | (2) |
 | (3) |
 | (4) |
Statistical Analysis
In protocol 1, mean levels of CSP, RSNA, AP, and HR under CTL and PBG conditions were calculated by averaging the respective values for 10 min. Differences in mean levels of CSP, AP, and HR were examined with the use of a paired t-test (4). Because RSNA amplitude varied depending on recording conditions, such as physical contact between the nerve and electrodes, the mean level of RSNA was presented as the percent change from control value. Differences in the mean level of RSNA were examined with Wilcoxon's signed-rank test (4).
In protocol 1, the transfer function was normalized in each animal so that average gain values <0.03 Hz became unity under the CTL condition. The same normalization factor was applied to the transfer function obtained from the PBG condition. To test the difference between the CTL and PBG conditions, we obtained the gain and phase values at 0.01, 0.1, and 0.5 Hz in each animal. The group differences in these values between the CTL and PBG conditions were examined by paired t-test (4). The same analytical procedure was applied to protocol 3 between CTL1 and CTL2 conditions.
In protocol 2, maximum changes in RSNA, AP, and HR were observed within 30 s of the bolus PBG injection. We averaged CSP, RSNA, AP, and HR values for 30 s before and after the PBG injection. Changes in CSP, AP, and HR were examined by paired t-test (4). Changes in RSNA were examined by Wilcoxon signed-rank test (4) and are presented as the percent change from the value before the PBG injection. All data are expressed as means ± SE. In all the statistics, differences were considered significant at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Figure 2 summarizes mean levels of CSP, RSNA, AP, and HR averaged from the seven rabbits under CTL and PBG conditions in protocol 1. Mean levels of CSP were kept unchanged. Mean levels of RSNA, AP, and HR were significantly lower under PBG than under CTL conditions.
|
Figure 3 shows averaged
total baroreflex loop transfer functions under CTL (left) and PBG
(right) conditions in protocol 1. Gain, phase, and coherence
functions are shown. The gain value at 0.01 Hz approximated unity under the
CTL condition, owing to normalization of dynamic gain. The gain decreased as
input frequency increased under both CTL and PBG conditions, indicating the
low-pass characteristics of the total baroreflex loop. However, the gain under
the PBG condition was lower than that under the CTL condition at every
frequency. The phase approached –
radians at the lowest frequencies,
reflecting the negative feedback accomplished by the total baroreflex loop
under both CTL and PBG conditions. The coherence was 
0.4 at the lowest
frequency, increasing to 0.7 at the frequencies between 0.04 and 0.8 Hz
under the CTL condition. The coherence function in the frequencies between
0.04 and 0.8 Hz was somewhat lower under PBG than under CTL conditions.
|
Table 1 summarizes the gain and phase values of the total loop transfer functions in Fig. 3. The dynamic gain values were significantly smaller under PBG than under CTL conditions at 0.01, 0.1, and 0.5 Hz. The difference in phase delay between CTL and PBG conditions was significant only at 0.1 Hz.
|
Figure 4 shows gain, phase,
and coherence of the averaged neural arc transfer functions under CTL
(left) and PBG (right) conditions in protocol 1. As
dynamic gain was normalized, the gain value at 0.01 Hz approximated unity
under the CTL condition. The gain values increased as the frequency increased
under both CTL and PBG conditions, indicating derivative characteristics of
the neural arc. PBG caused an approximately parallel downward shift of the
gain plot compared with the CTL condition. The phase value approached
– radians at the lowest frequency, reflecting the out-of-phase
relationship between CSP and RSNA. The phase plot did not differ between CTL
and PBG conditions. The coherence was 0.4 at the lowest frequency,
increasing to 0.7 at frequencies between 0.04 and 0.8 Hz under the CTL
condition. The coherence function under the PBG condition was slightly lower
than under the CTL condition.
|
Table 2 summarizes the gain and phase values of the neural transfer functions shown in Fig. 4. The dynamic gain values were significantly lower under PBG than under CTL conditions at 0.01, 0.1, and 0.5 Hz.
|
Figure 5 shows gain, phase, and coherence of the averaged peripheral arc transfer functions under CTL (left) and PBG (right) conditions in protocol 1. As the dynamic gain was normalized, the gain value at 0.01 Hz approximated unity under the CTL condition. The dynamic gain values decreased as the input frequency increased under both CTL and PBG conditions, indicating the low-pass characteristics in the peripheral arc. PBG significantly decreased the dynamic gain at 0.1 Hz. The phase approached zero radians at the lowest frequency under both CTL and PBG conditions, reflecting the fact that a rise in RSNA increased AP. PBG significantly increased phase delay at 0.1 Hz. The coherence values under the CTL condition were from 0.6 to 0.8 at frequencies <0.4 Hz. The coherence values were slightly lower under PBG than under CTL conditions.
|
Table 3 summarizes the gain and phase values of the peripheral arc transfer functions in Fig. 5. The gain values under the PBG condition were significantly lower than under the CTL condition only at 0.1 Hz.
|
Figure 6 depicts the results obtained from protocol 2. When the vagi were kept intact, PBG significantly decreased RSNA, AP, and HR. RSNA and AP reductions did not occur after vagotomy. PBG slightly increased HR under the vagotomized condition.
|
In protocol 3, dynamic characteristics of the transfer functions in the total loop, neural arc, and peripheral arc did not differ between CTL1 and CTL2 conditions in all of the frequencies examined (Tables 1, 2, 3, protocol 3).
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reduction of Dynamic Gain in Total Baroreflex Loop by BJ Reflex
The open-loop transfer function of the total baroreflex loop shows low-pass characteristics. Intravenous PBG decreased dynamic gain of the total baroreflex loop in every frequency range under study (Fig. 3 and Table 1). The steady-state gain (i.e., dynamic gain at 0.01 Hz) was almost halved by PBG. Although there was a slight difference in phase value between CTL and PBG conditions at 0.1 Hz, the general characteristics of the phase plot were similar between the two conditions. In the following paragraphs we will discuss the transfer function of the total baroreflex loop, focusing on system stability and performance against exogenous perturbations. The absolute gain value of the total baroreflex was, however, left undetermined in the present study because of aortic denervation and possible injury to the carotid sinus nerves during the isolation procedure.
A given negative feedback system could become unstable if its open-loop
transfer function showed a gain value greater than unity at a phase value of
– radians. A phase delay of – radians was encountered at
0.2 Hz under the CTL condition in protocol 1. It must be noted
that a phase value of –2 radians corresponds to a phase delay of
– radians in Fig. 3,
because the phase value at the lowest frequency is designated as –
radians rather than 0 radians, taking into account the negative feedback
nature of the total loop. The dynamic gain at 0.2 Hz was approximately
one-fifth the steady-state gain. If we simply move the gain plot upward, the
steady-state gain of 5 would make the dynamic gain at 0.2 Hz unity,
thereby causing system instability. A previous study
(15) indicated that the
arterial baroreflex system is marginally stable, suggesting that the absolute
gain value of the total baroreflex is 5. This assumption is based on
linear analysis; the actual baroreflex system may only generate sustained
oscillation, even with greater gain values, by virtue of system nonlinearity
(14). Hosomi et al.
(5) reported that the total
baroreflex gain estimated from AP response to mild hemorrhage was as great as
7 in rabbits. Because intravenous PBG halved the dynamic gain and increased
the gain margin without affecting the phase characteristics, PBG could make
the baroreflex system more stable.
One of the important roles of the arterial baroreflex is to attenuate
exogenous disturbance on AP. Disturbance minification (m) is
calculated from the following equation
(21)
 | (5) |
 | (6) |
 | (7) |
The ratio of minification is 4/3 when G = 1 and asymptotically approaches 2 with increasing G. In other words, the exogenous disturbance affects AP approximately twofold greater under PBG than under CTL conditions. Therefore, the BJ reflex reduces the baroreflex system performance of minification, making AP more vulnerable to exogenous pressure perturbations.
Effects of PBG on Neural and Peripheral Arc Transfer Characteristics
Intravenous PBG decreased the dynamic gain of the neural arc transfer function (Fig. 4), but did not significantly affect that of the peripheral arc transfer function (Fig. 5). Previous studies indicate that the BJ reflex and the arterial baroreflex might share common central pathways, as follows. Autoradiographic studies have shown that most 5-HT3 receptors in the nucleus tractus solitarius (NTS) are found on the vagal sensory afferent fibers (18). Pires et al. (23) demonstrated that intracisternal or NTS injection of the 5-HT3 receptor antagonist granisetron significantly attenuated the hypotension and bradycardia evoked by intravenous PBG, suggesting that NTS was involved in the central pathways of the BJ reflex. Verberne et al. (31) demonstrated that barosensitive neurons in the rostral ventrolateral medulla (RVLM) were inhibited by intravenous injection of PBG in rats. It is conceivable that intravenous PBG attenuates the dynamic gain of the neural arc transfer function by affecting baroreflex signal transduction in such brainstem areas as NTS and RVLM through activation of the vagal afferent fibers.
In the present study, mean levels of RSNA were decreased by intravenous PBG. In addition, the RSNA power spectra were reduced by PBG due to decreased dynamic gain of the neural arc transfer function. Notwithstanding differences in the operating point and input power for the peripheral arc, the peripheral arc transfer function differed only at 0.1 Hz between CTL and PBG conditions (Fig. 5 and Table 3). Also, in a previous study (13), changes in input power did not significantly affect the peripheral arc transfer function. Judging from the static input-output characteristics, the SNA-AP relationship is much more linear than the CSP-SNA relationship (25). Thus the differences in operating point and input power between CTL and PBG conditions would not have been sufficiently large to alter the peripheral arc transfer function.
The total loop transfer function is determined by a product of the neural and peripheral arc transfer functions. The fast neural arc compensated for the slow peripheral arc resulting in the optimization of the dynamic AP regulation in terms of stability and quickness (6). Although dynamic gain of the neural arc was decreased by intravenous PBG compared with the CTL condition, the derivative characteristics of the neural arc were preserved (Fig. 4). In other words, the dynamic gain of the neural arc increased with increasing frequency under both CTL and PBG conditions. As a result, the decreasing slope of dynamic gain in the total loop transfer function was shallower than that in the corresponding peripheral arc transfer function (Figs. 3 vs. 5). However, the total loop gain in every frequency was smaller in PBG than in CTL conditions due to the downward shift of the neural arc transfer function (Fig. 3).
In protocol 3, none of the total loop, neural arc, and peripheral arc transfer functions differed between CTL1 and CTL2 conditions (Tables 1, 2, 3). Therefore, changes in the transfer functions between CTL and PBG conditions in protocol 1 were not attributable to the cumulative effects of continuous administration of anesthetics.
Effects of PBG on RSNA, AP, and HR
Intravenous bolus injection of PBG decreased mean levels of RSNA, AP, and HR before vagotomy, which change was erased after vagotomy (Fig. 6). HR was even increased by PBG after vagotomy, suggesting direct action of PBG on the heart. Changes in RSNA and AP during PBG administration are therefore most likely mediated by the activation of vagal afferent fibers. These results were consistent with the study by Veelken et al. (30) where C-fiber activity increased after intravenous PBG administration in rats. The direct actions of PBG on the neural arc and the peripheral arc might have been minimal in the present experimental settings.
Veelken et al. (29) reported that 15-min intravenous injection of PBG decreased RSNA, AP, and HR during the first minute of administration in conscious rats. In their study, only RSNA showed sustained decrease for the 15 min, whereas AP and HR returned to their respective baseline values. In the present study, by contrast, not only RSNA but also AP and HR decreased during PBG administration for 20 min. This apparent contradiction may be partly explained by the difference between baroreflex closed-loop versus open-loop experimental settings. Under baroreflex closed-loop conditions, a decrease in AP is sensed by arterial baroreceptors and the arterial baroreflex counteracts the effects of PBG. Under baroreflex open-loop conditions, counteraction by the arterial baroreflex does not occur, uncovering the pure effects of PBG on the circulatory system. Differences in conscious and anesthetized conditions should also be taken into account.
Clinical Implication
We used intravenous PBG administration to evoke the BJ reflex. Thus the magnitude of the BJ reflex could differ from that induced by myocardial ischemia. However, an increase in the myocardial acetylcholine level induced by intravenous PBG (80 µg/kg) was similar to that observed in the nonischemic myocardium during coronary artery occlusion in anesthetized cats (11, 12). Therefore, the extent of the BJ reflex induced by the present doses of PBG might not be far from that induced by myocardial ischemia. We speculate that the suppression of the arterial baroreflex by the BJ reflex occurs during pathological conditions associated with ischemic heart diseases. To answer the question whether ischemia-induced BJ reflex halves the dynamic gain of the arterial baroreflex, a baroreflex open-loop experiment with coronary artery occlusion should be required.
The induction of the BJ reflex causes bradycardia and hypotension, which may prevent overexertion of cardiac muscle (27). The reduction of energy consumption is considered to be beneficial in hampering ischemic insult. At the same time, however, the BJ reflex blunts the normal AP regulation by the arterial baroreflexes. Because the magnitude of sympathetic inhibition and vagal activation during the BJ reflex is not controlled in terms of systemic AP regulation, excess activation of the BJ reflex could lead to severe bradycardia and hypotension, thereby jeopardizing the patient's life.
Limitations
There are several limitations to this study. First, we performed the experiments using anesthetized rabbits. Because anesthesia affects autonomic nervous activities (28), the results might have been different had we performed the experiment without anesthesia. Second, because the vagi were kept intact, low-pressure baroreflexes from the cardiopulmonary region could interact with the arterial baroreflex, affecting estimation of carotid sinus baroreflex transfer functions. However, the transfer functions estimated in the present study were qualitatively similar to those estimated under vagotomized conditions in previous studies (8–10). We speculate that changes in RSNA, AP, and HR were mainly attributable to CSP input. Third, we filled isolated carotid sinuses with warm physiological saline. Because the ionic content affects baroreceptor sensitivity (1), it might also affect the dynamic characteristics of the neural arc. However, as we did not change the intravascular ionic content in the isolated carotid sinuses, transfer function changes most likely resulted from the PBG-induced BJ reflex.
In conclusion, intravenous PBG administration attenuated the total loop transfer function of the arterial baroreflex, mainly because of the reduction of dynamic gain in the neural arc transfer function. Excess activation of the BJ reflex during acute myocardial ischemia or infarction might exert adverse effects on AP regulation, not only through sympathetic suppression but also through attenuation of baroreflex dynamic gain.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Kawada, T. Miyamoto, K. Uemura, K. Kashihara, A. Kamiya, M. Sugimachi, and K. Sunagawa Effects of neuronal norepinephrine uptake blockade on baroreflex neural and peripheral arc transfer characteristics Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2004; 286(6): R1110 - R1120. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kawada, K. Uemura, K. Kashihara, A. Kamiya, M. Sugimachi, and K. Sunagawa A derivative-sigmoidal model reproduces operating point-dependent baroreflex neural arc transfer characteristics Am J Physiol Heart Circ Physiol, June 1, 2004; 286(6): H2272 - H2279. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
