AJP - Heart AJP: Gastrointestinal and Liver Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 295: H1855-H1866, 2008. First published August 29, 2008; doi:10.1152/ajpheart.522.2008
0363-6135/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/5/H1855    most recent
522.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
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 Google Scholar
Google Scholar
Right arrow Articles by Miyamoto, T.
Right arrow Articles by Sugimachi, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miyamoto, T.
Right arrow Articles by Sugimachi, M.

Contrasting effects of presynaptic {alpha}2-adrenergic autoinhibition and pharmacologic augmentation of presynaptic inhibition on sympathetic heart rate control

Tadayoshi Miyamoto,1,2 Toru Kawada,2 Yusuke Yanagiya,2 Tsuyoshi Akiyama,3 Atsunori Kamiya,2 Masaki Mizuno,2 Hiroshi Takaki,2 Kenji Sunagawa,4 and Masaru Sugimachi2

1Department of Physical Therapy, Faculty of Health Sciences, Morinomiya University of Medical Sciences; and 2Department of Cardiovascular Dynamics, Advanced Medical Engineering Center, and 3Department of Cardiac Physiology, National Cardiovascular Center Research Institute, Osaka; and 4Department of Cardiovascular Medicine, Graduate School of Medical Sciences, Kyusyu University, Fukuoka, Japan

Submitted 16 May 2008 ; accepted in final form 19 August 2008


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 REFERENCES
 
Presynaptic {alpha}2-adrenergic receptors are known to exert feedback inhibition on norepinephrine release from the sympathetic nerve terminals. To elucidate the dynamic characteristics of the inhibition, we stimulated the right cardiac sympathetic nerve according to a binary white noise signal while measuring heart rate (HR) in anesthetized rabbits (n = 6). We estimated the transfer function from cardiac sympathetic nerve stimulation to HR and the corresponding step response of HR, with and without the blockade of presynaptic inhibition by yohimbine (1 mg/kg followed by 0.1 mg·kg–1·h–1 iv). We also examined the effect of the {alpha}2-adrenergic receptor agonist clonidine (0.3 and 1.5 mg·kg–1·h–1 iv) in different rabbits (n = 5). Yohimbine increased the maximum step response (from 7.2 ± 0.8 to 12.2 ± 1.7 beats/min, means ± SE, P < 0.05) without significantly affecting the initial slope (0.93 ± 0.23 vs. 0.94 ± 0.22 beats·min–1·s–1). Higher dose but not lower dose clonidine significantly decreased the maximum step response (from 6.3 ± 0.8 to 6.8 ± 1.0 and 2.8 ± 0.5 beats/min, P < 0.05) and also reduced the initial slope (from 0.56 ± 0.07 to 0.51 ± 0.04 and 0.22 ± 0.06 beats·min–1·s–1, P < 0.05). Our findings indicate that presynaptic {alpha}2-adrenergic autoinhibition limits the maximum response without significantly compromising the rapidity of effector response. In contrast, pharmacologic augmentation of the presynaptic inhibition not only attenuates the maximum response but also results in a sluggish effector response.

systems analysis; transfer function; {alpha}-adrenergic blockade; rabbits

PRESYNAPTIC {alpha}2-ADRENERGIC receptors play an important role in regulating neurotransmitter release in the central and peripheral nervous systems. The concept that neurotransmitter release is modulated by presynaptic autoreceptors was proposed in the 1970s (19, 20, 25, 31–33, 37, 38). Langer (18) first demonstrated that an {alpha}-adrenergic antagonist phentolamine, at a concentration below that required to produce its negative chronotropic effect, increases the magnitude of heart rate (HR) response to sympathetic nerve stimulation. Since then, a number of in vivo and in vitro studies have been conducted to characterize the negative feedback regulation of norepinephrine (NE) release via the presynaptic {alpha}2-adrenergic receptors located on the sympathetic nerve terminals (1, 6, 9, 11, 17, 24, 26, 27a, 29, 30, 34, 35). However, the dynamic nature of the presynaptic {alpha}2-adrenergic inhibition in sympathetic HR control remains to be quantified. Because we focus on the effector response to sympathetic nerve stimulation, the term "presynaptic" may be interpreted as "prejunctional" throughout this paper to describe more specifically the NE kinetics at the neuroeffector junction.

We first schematize our hypothesis on the possible modes of operations of the presynaptic inhibition. With reference to Fig. 1, the solid and dotted lines indicate the HR responses with and without the presynaptic inhibition, respectively. Figure 1A represents a "limiter-like" operation of the presynaptic inhibition in which the steady-state response is attenuated, while the initial slope of the response is unchanged. Figure 1B represents an "attenuator-like" operation in which the steady-state response is attenuated, while the initial slope of the response is also reduced in proportion to the attenuation of the steady-state response. Since the rapid effector response is one of the important hallmarks of neural regulation compared with humoral regulation, determining which of the two operations likely occurs would contribute to the physiological understanding of the presynaptic inhibition. The words "limiter-like" and "attenuator-like" in this paper are used in the specific senses described above.


Figure 1
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 1. Schematic representations of the possible operations of the presynaptic inhibition in heart rate (HR) step response to sympathetic nerve stimulation. The solid and dashed lines indicate the HR step response with and without the presynaptic inhibition, respectively. A: the presynaptic inhibition attenuates the steady-state response without affecting the initial slope of the response (a "limiter-like" operation). B: the presynaptic inhibition attenuates the steady-state response accompanied by a decrease in the initial slope in proportion to the attenuation of the steady-state response (an "attenuator-like" operation). Rapid effector response is maintained in the former but not in the latter. C: postulated nerve stimulation rate.

 
To answer which of the two operations likely occurs in the presynaptic inhibition, we examined the HR response to dynamic sympathetic nerve stimulation, with or without blocking the {alpha}2-adrenergic receptors in anesthetized rabbits. Because the HR response is mainly mediated by the postsynaptic β1-adrenergic receptors, the administration of an {alpha}2-adrenergic receptor antagonist does not eliminate the HR response to sympathetic nerve stimulation. We also examined the effects of pharmacologic augmentation of the {alpha}2-adrenergic receptors on the HR response to dynamic sympathetic nerve stimulation. The results of the present study indicated that the presynaptic {alpha}2-adrenergic autoinhibition is a limiter-like operation. In contrast, the pharmacologic augmentation of presynaptic inhibition is an attenuator-like operation. A possible theoretical explanation for the difference in dynamic characteristics between the presynaptic {alpha}2-adrenergic autoinhibition and the pharmacologic augmentation of the presynaptic inhibition will be proposed.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 REFERENCES
 
Surgical Preparations

Animal care was in accordance with "Guiding Principles for the Care and Use of Animals in the Field of Physiological Sciences," approved by the Physiological Society of Japan. All protocols were reviewed and approved by the Animal Subject Committee of the National Cardiovascular Center. Japanese white rabbits, weighing 2.5–3.1 kg, were anesthetized by intravenous injection (2 ml/kg) of a mixture of urethane (250 mg/ml) and {alpha}-chloralose (40 mg/ml) and mechanically ventilated with oxygen-enriched room air. Tidal volume was set at 35 ml and the rate was adjusted between 35 and 40 cycles/min to be sufficient for suppressing spontaneous respiration. Supplemental doses of these anesthetics were administered by continuous intravenous infusion (1 ml·kg–1·h–1) into the marginal ear vein. Arterial pressure (AP) was monitored with a micromanometer catheter (model, Millar Instruments, Houston, TX) inserted into the right femoral artery. A catheter for drug administration was also placed in the right femoral vein. Sinoaortic denervation was performed bilaterally to minimize changes in systemic sympathetic activity via the arterial baroreflexes. The vagi were also sectioned bilaterally at the neck level to remove the vagal control on HR. The right inferior cardiac sympathetic nerve was exposed through a midline thoracotomy and sectioned. A pair of bipolar platinum electrodes was then attached to the cardiac end of the sectioned sympathetic nerve for stimulation (12, 13, 22, 23). The stimulation electrodes and nerve were secured with silicon glue (Kwik-Sil, World Precision Instruments, Sarasota, FL). Instantaneous HR was measured from the AP signal utilizing a cardiotachometer (Tachometer N4778, San-ei, Tokyo, Japan). Body temperature was maintained at 38°C with a heating pad throughout the experiment.

Experimental Procedures

Protocols. To estimate the transfer function from the sympathetic nerve stimulation to HR response, we employed a binary white noise stimulation signal with a switching interval of 5 s. The power spectrum of the sympathetic nerve stimulation rate was fairly constant up to 0.1 Hz and decreased to ~1/10 at 0.15 Hz. The upper frequency limit of the input power that covers the frequency range of physiological interest was determined based on our laboratory's previous studies (12, 23) and also preliminary experimental runs. Different sequences of binary white noise signals were used in different animals. Because HR is linearly related to cardiac output when stroke volume is unchanged, we chose HR as an output signal to understand sympathetic cardiovascular regulation. However, to rule out the possibility that the reciprocal relationship between R-R interval (RRI) and HR confounded the analytical results, we also calculated the transfer function using RRI as an output signal.

In protocol 1 (n = 6), to examine the dynamic nature of the presynaptic {alpha}2-adrenergic autoinhibition, we estimated the transfer function from dynamic sympathetic nerve stimulation to HR response from 20-min data obtained under control and {alpha}2-adrenergic blockade conditions as follows. After recording the control data, an {alpha}2-adrenergic antagonist yohimbine was administered intravenously with an initial bolus injection of 1 mg/kg, followed by continuous infusion at 0.1 mg·kg–1·h–1. The yohimbine bolus was equivalent to 10 h of infusion. The duration from the initiation of yohimbine administration until HR and AP reached new steady-state levels was ~15 min (35). We then repeated the 20-min dynamic sympathetic nerve stimulation and recorded the HR response under the {alpha}2-adrenergic blockade condition.

In protocol 2 (n = 5), to examine the effects of pharmacologic augmentation of the presynaptic {alpha}2-adrenergic inhibition on the sympathetic HR control, we estimated the transfer function from dynamic sympathetic nerve stimulation to HR response before and during the administration of an {alpha}2-adrenergic receptor agonist clonidine. Clonidine was administered intravenously at 0.3 and 1.5 mg·kg–1·h–1 in an increasing order. After 20-min baseline data collection, we started lower dose clonidine administration and waited for 15 min and then collected data for 20 min. Next, we started higher dose clonidine administration and waited for 15 min and then collected data for 20 min.

The stimulation rate of binary white noise was set at 0–1 Hz for protocol 1, and 0–5 Hz for protocol 2. Because we expected that blockade of the presynaptic {alpha}2-adrenergic inhibition would augment, whereas activation of the inhibition would attenuate, the HR response, we set a higher stimulation rate for protocol 2 than for protocol 1. The pulse width of sympathetic stimulation was set at 2 ms. The amplitude was set so that 5-Hz tonic sympathetic stimulation produced a HR increase of ~50 beats/min.

As a supplemental protocol, we performed the transfer function analysis using binary white noise signals of 0–1 Hz (Bin0-1), 0–3 Hz (Bin0-3), and 0–5 Hz (Bin0-5) in a random order (n = 5). At least a 15-min interval was allowed between the 20-min dynamic sympathetic stimulation trials. The amplitude of sympathetic stimulation was set so that 1-Hz tonic sympathetic stimulation produced a HR increase of ~50 beats/min.

Medetomidine has higher affinity to {alpha}2-adrenergic receptors over {alpha}1-adrenergic receptors compared with clonidine ({alpha}2/{alpha}1 = 1,620:1 for medetomidine, 220:1 for clonidine) (28). However, a preliminary experiment indicated that medetomidine was not as effective as clonidine to modulate the transfer function from dynamic sympathetic stimulation to HR. Accordingly, we examined the effects of clonidine or medetomidine on myocardial interstitial NE release in response to 5-Hz tonic stimulation (2.5 V, 2-ms pulse width) of the right cardiac sympathetic nerve in vagotomized rabbits. Two microdialysis probes were implanted in the myocardium of the left ventricular free wall. Ringer solution was perfused at 2 µl/min. After a 2-h equilibrium period, we collected 5-min dialysate samples to measure the dialysate NE concentration as an index of myocardial interstitial NE levels (14, 15). High-performance liquid chromatography with electrochemical detection was used to quantify the NE concentration. After the sympathetic stimulation was performed under the control condition, clonidine or medetomidine was intravenously administered (1.5 mg·kg–1·h–1). Fifteen minutes later, the sympathetic stimulation was performed under the drug administration condition. Then the drug administration was ceased. Forty-five minutes later, the sympathetic stimulation was performed under the recovery condition. We used different rabbits for clonidine and medatomidine trials. We pooled six dialysate data for statistical analysis.

Data Analysis

Data were digitized at 200 Hz utilizing a 12-bit analog-to-digital converter and stored on the hard disk of a dedicated laboratory computer system. Mean values for HR and AP during dynamic sympathetic nerve stimulation were calculated by averaging the respective data over the stimulation period.

The transfer function from dynamic sympathetic nerve stimulation to HR response was estimated by the following procedures. Twenty minutes of input data (stimulation command) and output data (HR) were resampled at 8 Hz. The resampled data were segmented into eight 50% overlapping bins consisting of 2,048 data points each. The segment length was 256 s. For each segment, the linear trend was subtracted, and a Hanning window was applied. Fast-Fourier transform was then performed to obtain the frequency spectrum of nerve stimulation rate [N(f)] and that of HR [HR(f)] (5). The power spectral density of the nerve stimulation rate [SN.N(f)], that of HR [SHR.HR(f)], as well as the cross-spectral density between these two signals [SN.HR(f)], were averaged over the eight segments. Finally, the transfer function [H(f)] from sympathetic nerve stimulation rate to HR response was calculated using the following equation (2, 21).

Formula 1(1)

Transfer function parameters were determined by fitting a second-order, low-pass filter to the estimated transfer function, according to previous studies (12, 13, 23). The second-order, low-pass filter with a pure dead time [G(f)] is expressed as

Formula 2(2)
where K is a steady-state gain, fN is natural frequency (in Hz), {zeta} is a damping ratio, L is pure dead time (in s), and j indicates an imaginary unit. A schematic explanation for these transfer function parameters is provided in the APPENDIX. To estimate the parameters, an iterative nonlinear least squares fitting was performed to minimize the following error function.

Formula 3(3)
where f0 is the fundamental frequency of the discrete Fourier transform, f0 = 1/256 = 0.004 Hz, and k is a frequency index. The n represents the upper limit of the frequency index determined from the range of sufficient input power in the sympathetic nerve stimulation; n = 40, f0 x n = 0.156 Hz.

To quantify the linear dependence of the HR response on the sympathetic nerve stimulation, the magnitude-squared coherence function [{gamma}2(f)] was calculated by the following equation (2, 21).

Formula 4(4)

The coherence value ranges from zero to unity. The unity coherence value indicates a perfect linear dependence between the input and output signals, whereas zero coherence indicates a total independence between the two signals.

To facilitate the intuitive understanding of the HR response to dynamic sympathetic nerve stimulation, we calculated the step response from the estimated transfer function. The step response was obtained from the time integral of the system impulse response derived from the inverse Fourier transform of the transfer function. The steady-state response was calculated by averaging the step response during the last 10 s of the 128-s response. To characterize the rising speed of the step response, the initial slope for the response was calculated as follows. An analysis of linear regression with a slope and an intercept was performed on the initial data points of the step response while varying the number of data points from 2 to 1,024. The maximum slope obtained was used as the initial slope of the response. The linear regression was performed, including the portion of the dead time. Although including the dead time reduced the maximum slope, the effect was small because the number of data points that yielded the maximum slope (~90 points) was much larger than that for the dead time (<10 points). The step response of RRI was also calculated from the corresponding transfer function from sympathetic nerve stimulation to RRI.

Statistics

All data are presented as means ± SE. In protocol 1, mean HR, AP, and transfer function parameters were compared before and during yohimbine administration by paired t-tests. In protocol 2, the data were compared among control, lower dose, and higher dose clonidine conditions using a repeated-measures ANOVA followed by Dunnett's test against the single control (8). In the supplemental protocol of the transfer function analysis, the data were compared among Bin0-1, Bin0-3, and Bin0-5 stimulus conditions using a repeated-measures ANOVA followed by Tukey test for all pairwise comparisons. In the supplemental protocol of the NE measurement, baseline NE levels were compared before and during drug administration using a paired t-test. The NE levels during sympathetic stimulation were compared among control, drug administration, and recovery conditions using a repeated-measures ANOVA followed by Dunnett's test against the control condition. In all of the statistical procedures, the difference was considered significant at P < 0.05.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 APPENDIX
 GRANTS
 REFERENCES
 
Figure 2A represents a typical recording obtained from protocol 1. We stimulated the cardiac sympathetic nerve according to a binary white noise signal and recorded HR response under control condition and during yohimbine administration. The presynaptic {alpha}2-adrenergic negative feedback mechanism functioned under the control condition but not during yohimbine administration. HR changed dynamically in response to the random sympathetic nerve stimulation under both conditions. Yohimbine increased the magnitude of HR variation. The augmentation of sympathetic effect was also observed in the RRI response. Although yohimbine decreased the mean level of HR in this animal, changes in mean HR varied among the animals and were not significantly different between the control and yohimbine conditions.


Figure 2
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 2. A: representative recordings of cardiac sympathetic nerve stimulation rate (Stim; top), HR response (middle), and R-R interval (RRI) response (bottom) under conditions of control (left) and yohimbine administration (right) obtained in protocol 1. Yohimbine blocks the presynaptic {alpha}2-adrenergic autoinhibition. The amplitude of HR variation and that of RRI variation become greater in the presence of yohimbine. B: transfer functions averaged over all animals in protocol 1. HR gain plots (top), phase plots (second), coherence functions ({gamma}2, third), and RRI gain plots (bottom). Yohimbine increases the dynamic gain in the frequency range between 0.004 and 0.04 Hz but not in the higher frequency range. The fine solid curve in the gain (right) duplicates the mean gain plot (left). C: step responses of HR (top) and RRI (bottom) calculated from the corresponding transfer functions. Yohimbine augments the steady-state response without affecting the initial slope of the response (fine oblique line). Bold, solid lines represent the mean, whereas dotted lines indicate means ± SE.

 
Table 1 summarizes the mean HR and AP averaged from the six animals. The {alpha}2-adrenergic blockade by yohimbine did not significantly affect the HR or AP before sympathetic nerve stimulation. Yohimbine also did not affect HR or AP significantly during the stimulation period.


View this table:
[in this window]
[in a new window]

  		Table 1. Mean heart rate and arterial pressure before and during random stimulation of the cardiac sympathetic nerve

 
Figure 2B illustrates the transfer functions averaged from the six animals in protocol 1. In the HR gain plots, the gain value was relatively constant <0.01 Hz and decreased >0.01 Hz, indicating low-pass filter characteristics of the HR response to sympathetic nerve stimulation. Yohimbine increased the HR gain from 7.1 ± 0.7 to 12.0 ± 1.7 beats·min–1·Hz–1 at the lowest frequency of 0.004 Hz (P < 0.05). In contrast, yohimbine did not affect the HR gain value at 0.1 Hz (1.8 ± 0.4 vs. 1.7 ± 0.6 beats·min–1·Hz–1). The solid fine curve in the right panel duplicates the mean gain plot in the left panel as a reference. In the phase plots, the phase value approached zero radians at the lowest frequency and lagged with increasing frequency under both conditions. In the coherence function plots, the coherence was >0.8 in the frequency range from 0.01 to 0.08 Hz, suggesting that the HR response to sympathetic nerve stimulation in this frequency range can be explained reasonably well by linear dynamics for both conditions. Changes in the RRI gain plots were similar to those in the HR gain plots. Yohimbine increased the RRI gain from 6.0 ± 0.7 to 11.3 ± 1.9 ms/Hz at the lowest frequency of 0.004 (P < 0.05) but not at 0.1 Hz (1.8 ± 0.4 vs. 1.9 ± 0.8 ms/Hz). Given the inverse relationship between RRI and HR, the RRI phase plots (not shown) quite resembled to the corresponding HR phase plots except for the rotation by {pi} radians.

Figure 2C represents the step responses of HR to sympathetic nerve stimulation calculated from the transfer functions shown in Fig. 2B. Yohimbine increased the steady-state response significantly (Table 2). The initial slope of the response, depicted by an oblique straight line, was not affected by yohimbine (Table 2). In the RRI step response, yohimbine augmented the steady-state response from –6.7 ± 0.9 to –12.6 ± 2.1 ms (P < 0.05) without affecting the initial slope (–0.71 ± 0.18 vs. –0.90 ± 0.23 ms/s).


View this table:
[in this window]
[in a new window]

  		Table 2. Parameters of the transfer functions and step responses

 
Parameters of the transfer functions and step responses estimated in protocol 1 are summarized in Table 2. The steady-state gain was significantly greater and the natural frequency was significantly lower in yohimbine condition compared with control. The damping coefficient and pure dead time did not differ significantly between the control and yohimbine conditions. Whereas the steady-state response was significantly increased by yohimbine, the initial slope of the step response was not significantly changed.

Figure 3A represents a typical recording of the sympathetic nerve stimulation and HR response obtained from protocol 2. The effects of {alpha}2-adrenergic stimulation by clonidine were tested at two doses. Lower dose clonidine did not affect the magnitude of HR variation. Although lower dose clonidine decreased the mean HR in this animal, changes in the mean HR were not significantly different among the animals (Table 3). Higher dose clonidine significantly attenuated the magnitude of HR variation and also decreased mean HR. The attenuation of sympathetic effect was also observed in the RRI response during the high-dose clonidine administration.


Figure 3
View larger version (29K):
[in this window]
[in a new window]

  		Fig. 3. A: representative recordings of cardiac sympathetic nerve stimulation rate (top), HR response (middle), and RRI response (bottom) under conditions of control (left), lower-dose clonidine (0.3 mg·kg–1·h–1; middle), and higher-dose clonidine (1.5 mg·kg–1·h–1; right) obtained in protocol 2. Clonidine activates the presynaptic {alpha}2-adrenergic inhibition independent of the amount of norepinephrine released at the sympathetic nerve terminals. The amplitude of HR variation becomes smaller, and the mean level of HR becomes lower in the presence of higher-dose clonidine. The amplitude of RRI response also became smaller under higher-dose clonidine condition. B: transfer functions averaged over all animals in protocol 2. HR gain plots (top), phase plots (second), coherence functions ({gamma}2, third), and RRI gain plots (bottom). Lower-dose clonidine does not affect the transfer function significantly. Higher-dose clonidine decreases the dynamic gain in the whole frequency range (0.004 to 0.2 Hz). The fine solid curves in the gain plots (middle and right) duplicate the mean gain plot (left). C: step responses of HR (top) and RRI (bottom) calculated from the corresponding transfer functions. Lower-dose clonidine does not affect the step response significantly. Higher-dose clonidine attenuates the steady-state response accompanied by a decrease in the initial slope of the response (fine oblique line). Bold, solid lines represent the mean, whereas dotted lines indicate means ± SE.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Mean heart rate and arterial pressure before and during random stimulation of the cardiac sympathetic nerve

 
Table 3 summarizes the mean HR and AP obtained from protocol 2. Higher dose, but not lower dose, clonidine significantly decreased the mean HR, both before and during cardiac sympathetic nerve stimulation. Clonidine did not affect mean AP significantly, before or during cardiac sympathetic nerve stimulation.

Figure 3B illustrates the transfer functions averaged from the five animals in protocol 2. Lower dose clonidine did not affect the transfer function significantly. In the HR gain plots, higher dose clonidine decreased the gain from 6.6 ± 0.9 to 2.7 ± 0.5 beats·min–1·Hz–1 at the lowest frequency of 0.004 Hz (P < 0.05) and from 1.1 ± 0.2 to 0.5 ± 0.2 beats·min–1·Hz–1 at the frequency of 0.1 Hz (P < 0.05). Higher dose clonidine did not affect the phase plot significantly. In the coherence function plots, the coherence was >0.8 in control and lower dose clonidine conditions and >0.7 in higher dose clonidine condition for the frequency range from 0.01 to 0.08 Hz, suggesting that the HR response to sympathetic nerve stimulation can be explained reasonably well by linear dynamics in all three conditions. Although relative change became smaller compared to the HR gain plots, the attenuation of transfer gain was also observed in the RRI gain plots. Higher-dose clonidine decreased the gain from 4.5 ± 0.7 to 2.8 ± 0.5 ms/Hz at the lowest frequency of 0.004 Hz (P < 0.05) and from 0.88 ± 0.19 to 0.04 ± 0.09 ms/Hz at the frequency of 0.1 Hz (P < 0.05).

Figure 3C represents the step responses of HR to sympathetic nerve stimulation calculated from the transfer functions shown in Fig. 3B. Lower dose clonidine did not affect either the steady-state response or the initial slope of the step response. In contrast, higher dose clonidine attenuated the steady-state response and also reduced the initial slope of the response. In the RRI step response, higher-dose clonidine attenuated the steady-state response from –4.9 ± 0.7 to –3.0 ± 0.6 ms (P < 0.05) with a significant reduction in the initial slope from –0.40 ± 0.07 to –0.23 ± 0.05 ms/s (P < 0.05).

Parameters of the transfer functions and step responses estimated in protocol 2 are summarized in Table 4. The steady-state gain of the transfer function and the steady-state response of the corresponding step response were decreased by higher dose but not by lower dose clonidine. The initial slope of the step response was decreased by higher dose clonidine. The ratio of the steady-state response to the initial slope was unchanged. The natural frequency and the damping ratio of the transfer function were not affected by clonidine. The pure dead time of the transfer function was increased by lower dose, but not by higher dose, clonidine.


View this table:
[in this window]
[in a new window]

  		Table 4. Transfer function parameters and step responses

 
Figure 4A represents a typical recording of the sympathetic nerve stimulation and HR response obtained from the supplemental protocol. The binary white noise signals of the same sequence but different stimulus rate were applied. Increasing the stimulus rate augmented the magnitude of HR variation and increased mean HR. The increase was not proportional to the increase in the stimulus rate, however, because of the saturation of HR response to sympathetic nerve stimulation. The increase of RRI variation was not proportional to the increase in the stimulus rate either, suggesting that the saturation effect observed in the HR response was not an artifact of reciprocal relationship between RRI and HR.


Figure 4
View larger version (28K):
[in this window]
[in a new window]

  		Fig. 4. A: representative recordings of cardiac sympathetic nerve stimulation rate (top), HR response (middle), and RRI response (bottom) obtained by differing the stimulus rate of the binary white noise signal. Bin0–1, binary white noise between 0 and 1 Hz; Bin0–3, binary white noise between 0 and 3 Hz; Bin0–5, binary white noise between 0 and 5 Hz. Increasing the stimulus rate of the binary white noise signal augments the magnitude of HR response and increased mean HR. The RRI response was also increased with increasing the stimulus rate. B: transfer functions averaged over all animals in the supplemental protocol. HR gain plots (top), phase plots (second), coherence functions ({gamma}2, third), and RRI gain plots (bottom). Increasing the stimulus rate of the binary white noise signal decreases the dynamic gain in the whole frequency range (0.004 to 0.2 Hz). The fine solid curves in the gain plots (middle and right) duplicate the mean gain plot (left). C: step responses of HR (top) and RRI (bottom) calculated from the transfer functions. Increasing the stimulus rate of the binary white noise signal attenuates the steady-state response accompanied by a decrease in the initial slope of the response (fine oblique line). Bold, solid lines represent the mean, whereas dotted lines indicate means ± SE.

 
Table 5 summarizes the mean HR and AP obtained from the supplemental protocol. There were no significant differences in mean HR and AP before cardiac sympathetic nerve stimulation. Mean HR was higher in Bin0-3 and Bin0-5 than in Bin0-1 condition. Mean AP did not differ among the three conditions.


View this table:
[in this window]
[in a new window]

  		Table 5. Mean heart rate and arterial pressure before and during random stimulation of the cardiac sympathetic nerve

 
Figure 4B illustrates the transfer function averaged from the five animals in the supplemental protocol. The contour of HR gain plots showed an approximately downward shift with increase in the stimulus rate of the binary white noise signal, indicating that the augmentation of the HR variation seen in Fig. 4A was not proportional to the increase in the stimulus rate. No significant differences were noted in the phase plot. The coherence values were slightly decreased in all frequencies with increase in the stimulus rate of the binary white noise signal, suggesting that the HR response became saturated and the linearity between the stimulation and the HR response was slightly reduced in Bin0-3 and Bin0-5 compared with that in Bin0-1 condition. The contour of RRI gain plots also showed approximately downward shift with increasing the stimulus rate of the binary white noise signal.

Figure 4C represents the step response of HR to sympathetic nerve stimulation calculated from the transfer functions shown in Fig. 4B. The increase in the stimulus rate of the binary white noise signal attenuated the steady-state response and also reduced the initial slope of the response. In the RRI step response, the steady-state response was attenuated from –27.6 ± 2.8 to –12.2 ± 0.7 (P < 0.01) and –6.7 ± 0.4 (P < 0.01) ms during Bin0–3 and Bin0–5, respectively. The initial slope was attenuated from –3.0 ± 0.3 to –1.1 ± 0.1 (P < 0.01) and –0.65 ± 0.06 (P < 0.01) ms/s during Bin0–3 and Bin0–5, respectively.

Parameters of the transfer functions and step responses estimated in the supplemental protocol are summarized in Table 6. The steady-state gain of the transfer function and the steady-state response of the corresponding step response decreased with increase in the stimulus rate of the binary white noise sequence. Although the initial slope of the step response significantly decreased with increase in the stimulus rate of the binary white noise signal, the ratio of the steady-state response to the initial slope was unchanged. The natural frequency was lower and the damping coefficient was greater in Bin0-3 and Bin0-5 than Bin0-1 condition. The pure dead time of the transfer function did not differ among the three conditions.


View this table:
[in this window]
[in a new window]

  		Table 6. Transfer function parameters and step responses

 
Figure 5 summarizes the results of the supplemental protocol of NE measurement. Baseline levels of myocardial interstitial NE did not differ before and during clonidine administration (Fig. 5A). Clonidine administration attenuated the sympathetic stimulation-induced NE release to 75.8 ± 5.4% of the control (P < 0.05) (Fig. 5B). Baseline NE levels did not differ before and during medetomidine administration (Fig. 5C). Medetomidine did not attenuate the sympathetic stimulation-induced NE release significantly (92.0 ± 6.7% of the control, not significant) (Fig. 5D).


Figure 5
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 5. Effects of clonidine (1.5 mg·kg–1·h–1 iv) or medetomidine (1.5 mg·kg–1·h–1 iv) on the myocardial interstitial norepinephrine (NE) release in response to 5-Hz tonic cardiac sympathetic nerve stimulation. Data were obtained after sectioning vagal and cardiac sympathetic nerves. Clonidine administration does not affect baseline levels of NE (A), but significantly attenuates the stimulation-induced NE release (B). C: medetomidine administration does not affect baseline levels of NE. D: it does not attenuate the stimulation-induced NE release significantly. Values are means ± SE. *P < 0.05 from control.

 
Simulation Study

To explore possible mechanisms for the observed differences between the presynaptic {alpha}2-adrenergic autoinhibition and the pharmacologic augmentation of the presynaptic inhibition via the {alpha}2-adrenergic receptors, we performed a simulation on the negative feedback regulation of the HR response to the sympathetic nerve stimulation. With reference to Fig. 6A, HFW and HFB represent the transfer functions of the forward path and the feedback path, respectively. A step input signal represents the sympathetic nerve stimulation. Both signals from presynaptic {alpha}2-adrenergic autoinhibition and pharmacologic augmentation of the presynaptic inhibition attenuate the input signal via the same {alpha}2-adrenergic receptors. Because the amount of neurotransmitter release cannot become negative, a threshold operator (Th) is added. The threshold operator is described mathematically as follows.

Formula 4


Figure 6
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 6. Possible theoretical explanation for the differential effects of presynaptic {alpha}2-adrenergic autoinhibition and pharmacologic augmentation of presynaptic {alpha}2-adrenergic inhibition on the HR response to sympathetic nerve stimulation. A: a simulation model for the HR step response to a step input in the sympathetic nerve activity. HFW, transfer function of the forward path; HFB, transfer function of the feedback path; Th, a threshold operator (see main text for details). B: simulation results under conditions of no presynaptic inhibition (dash-dot line; HFB = 0 and pharmacologic augmentation of presynaptic inhibition = 0, corresponding to the yohimbine administration condition), functional presynaptic {alpha}2-adrenergic autoinhibition (solid line; HFB = HFW and pharmacologic augmentation of presynaptic inhibition = 0, corresponding to the control condition), and functional presynaptic {alpha}2-adrenergic autoinhibition plus pharmacologic augmentation of the presynaptic inhibition (dotted line; HFB = HFW and pharmacologic augmentation of presynaptic inhibition = 0.5, corresponding to the higher-dose clonidine condition). The presynaptic {alpha}2-adrenergic autoinhibition does not attenuate the initial slope of the step response. In contrast, the pharmacologic augmentation of the presynaptic inhibition attenuates the initial slope of the step response.

 
The output from the threshold operator or the amount of neurotransmitter is then fed into HFW to yield the output or change in HR and is also fed into HFB to yield the feedback signal of presynaptic {alpha}2-adrenergic autoinhibition. Since we administered clonidine ~15 min before sympathetic nerve stimulation, the effect of clonidine should have reached the steady state at the time of sympathetic nerve stimulation. Accordingly, we treated the pharmacologic augmentation of the presynaptic inhibition as a constant input. The magnitude of pharmacologic augmentation of the presynaptic inhibition was set arbitrarily to 0.5 to mimic the results of higher dose clonidine in protocol 2. The simulation was conducted using Matlab Simulink (The Mathworks, Natick, MA).

Yohimbine administration corresponds to severing the feedback path, i.e., setting HFB = 0 in the simulation. Under this condition, the transfer function from the input to output becomes HFW. Therefore, we modeled HFW using the second-order, low-pass filter with pure dead time (Eq. 3) with the settings of fN = 0.055, {zeta} = 1.55, and L = 0.94 (Table 2, yohinbine). The gain was set at unity for simplicity. With this setting, we calculated the output response to the unit step input without the presynaptic inhibition (Fig. 6B, dash-dot line, corresponding to the yohimbine condition). The initial slope of the response, calculated from the linear regression analysis described in the method section, was 0.0763 arbitrary units (AU)/s. Next, we set HFB = HFW and performed a simulation of the condition with the presynaptic {alpha}2-adrenergic autoinhibition (Fig. 6B, solid line, corresponding to the control condition). The presynaptic {alpha}2-adrenergic autoinhibition attenuates the steady-state response without significantly affecting the initial slope of the response (0.0695 AU/s). Finally, we set the pharmacologic augmentation of the presynaptic inhibition to 0.5 on top of the functioning HFB. The simulation result (Fig. 6B, dotted line, corresponding to the higher dose clonidine condition) demonstrates that pharmacologic augmentation of the presynaptic inhibition attenuates the steady-state response accompanied by a reduction in the initial slope of the response (0.0346 AU/s).


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
APPENDIX
 GRANTS
 REFERENCES
 
We compared the blockade and activation of the presynaptic {alpha}2-adrenergic receptors and found a difference between the presynaptic {alpha}2-adrenergic autoinhibition and the pharmacologic augmentation of the presynaptic inhibition in terms of HR response to sympathetic nerve stimulation. The presynaptic {alpha}2-adrenergic autoinhibition showed a limiter-like operation that restricts the steady-state response without affecting the initial slope of the response. In contrast, the pharmacologic augmentation of presynaptic inhibition showed an attenuator-like operation that reduces both the steady-state response and the initial slope of the response.

Comparison of Blocking and Activating the Presynaptic {alpha}2-Adrenergic Receptors

Although the presynaptic {alpha}2-adrenergic negative feedback has been known to attenuate the NE release and HR response to sympathetic nerve stimulation (9, 21, 26, 27, 29, 30, 31), the dynamic nature of the negative feedback remained to be elucidated. As shown in Fig. 2C, the blockade of {alpha}2-adrenergic receptors by yohimbine increased the steady-state response without significantly affecting the initial slope of the HR step response (Table 2). That is to say, the presynaptic {alpha}2-adrenergic autoinhibition of the presynaptic inhibition attenuates the steady-state response without sacrificing the rising speed of HR response to sympathetic nerve stimulation under control condition. These characteristics of the presynaptic {alpha}2-adrenergic autoinhibition conform to the limiter-like operation shown in Fig. 1A. In contrast, pharmacologic augmentation of the presynaptic inhibition by higher dose clonidine reduced the steady-state response accompanied by a decrease in the initial slope of the HR step response (Fig. 3C). The ratio of the steady-state response to the initial slope was not changed significantly by higher dose clonidine (Table 4), suggesting that attenuation of the initial slope was proportional to that of the steady-state response. These characteristics of the pharmacologic augmentation of the presynaptic inhibition conform to the attenuator-like operation shown in Fig. 1B. Rapid effector response is one of the most important hallmarks of neural regulation compared with humoral regulation. The findings of the present study suggest that presynaptic {alpha}2-adrenergic autoinhibition, but not pharmacologic augmentation of the presynaptic {alpha}2-adrenergic inhibition, prevents excess NE outflow at the sympathetic nerve terminals without compromising the rapidity of effector response. The simulation results suggest that the initial slope of the response decreases when presynaptic inhibition occurs, independent of the negative feedback mechanism (Fig. 6B). On the other hand, the initial slope of the response does not decrease significantly when the presynaptic inhibition occurs through the negative feedback mechanism.

{alpha}2-Adrenergic receptors are classified as {alpha}2A-, {alpha}2B-, and {alpha}2C-subtypes based on gene encodings (26). Furthermore, the different ligand binding characteristics of the {alpha}2A-subtype give rise to the pharmacological subtype of {alpha}2A in humans, rabbits, and pigs and that of {alpha}2D in rats, mice, and guinea pigs (26). The {alpha}2A and {alpha}2D may be considered as "orthologous" {alpha}2-receptors, with only one being present in any given species (27). In the sympathetic nerve, {alpha}2A- and {alpha}2C-receptors operate as presynaptic inhibitory autoreceptors, whereas {alpha}2B-receptors are located on postsynaptic cells to mediate the effects of catecholamine, such as vasoconstriction (26). In tissue slices from mouse atria, {alpha}2A-receptors inhibit NE release from sympathetic nerves primarily at high-stimulation rates (1–2 Hz), whereas {alpha}2C-receptors can operate at very low stimulation rates (0.05–0.1 Hz) (10). Because {alpha}2-receptors in the rabbit heart are characterized as {alpha}2A, changes in the transfer function from sympathetic nerve stimulation to HR response observed in the present study are most likely mediated by {alpha}2A-receptors.

Clonidine administration (5 µg/kg bolus followed by 30 µg·kg–1·h–1 iv) attenuated the sympathetic outflow from the central nervous system in rabbits (35). However, lower dose clonidine at 0.3 mg·kg–1·h–1 failed to significantly affect the steady-state response or the initial slope of the HR step response in the present study (Fig. 3B, Table 4), suggesting a difference in clonidine sensitivity between the central and peripheral sympathetic nervous systems. Another factor that should be taken into account is the operating range of the HR control (i.e., mean HR during dynamic sympathetic stimulation). As an example, tonic vagal stimulation decreased mean HR during dynamic sympathetic stimulation, which increased the dynamic gain of sympathetic HR control via nonlinear sigmoidal input-output nature between autonomic activities and HR (12, 13). Therefore, the decrease in the mean HR during lower dose clonidine, although it was statistically insignificant (Table 3), should have an effect of increasing the dynamic gain of the sympathetic HR control. Such an effect might have counterbalanced the effect of reducing the dynamic gain via presynaptic inhibition during the lower dose clonidine administration. Although higher dose clonidine decreased mean HR before and during sympathetic nerve stimulation, mean AP did not decrease compared with lower dose clonidine (Table 3). The discrepancy between the changes in mean HR and AP may be due to direct vasoconstriction by higher dose clonidine through {alpha}-adrenergic stimulation.

Transfer Function Analysis vs. Step Response Analysis

In a previous study, our laboratory performed a transfer function analysis on the sympathetic HR control using a binary white noise signal (12, 23). The transfer function is a frequency-domain representation of the system dynamic characteristics over a wide frequency range and is useful for understanding the behavior of the system in response to a variety of input signals (3, 7, 21). Notwithstanding the theoretical advantages of the transfer function, the frequency-domain representation may be somewhat unfamiliar to most physiologists. Therefore, we calculated the step responses corresponding to the transfer functions. As can be seen in Figs. 2, B and C and 3, B and C, changes in transfer function in the lower frequency range reflect the steady-state response in the step response. Changes in transfer function in the higher frequency reflect the initial transient response in the step response. Because the step response and the transfer function are mathematically interchangeable, both the transfer function and the step response provide comparable information on the system dynamic characteristics.

In a previous study, our laboratory has shown that increasing mean stimulus rate of the Gaussian white noise decreased the steady-state gain of the transfer function from sympathetic nerve stimulation to HR without affecting the natural frequency or damping coefficient significantly (23). Increasing the stimulus rate of the binary white noise signal also caused an approximately parallel downward shift in the gain plot (Fig. 4B). The transfer function parameters, however, showed a decrease in the natural frequency and an increase in the damping coefficient (Table 6). The higher natural frequency in Bin0-1 than in Bin0-5 condition may account for the higher natural frequency in protocol 1 (Bin0-1 was used for sympathetic stimulation) than in protocol 2 (Bin0-5 was used for sympathetic stimulation) observed under control conditions. Notwithstanding the differences in the natural frequency and the damping coefficient, the ratio of the step response to the initial slope was not changed significantly by the difference in the stimulus rate of the binary white noise signal. Therefore, yohimbine-induced changes in the ratio of the step response to the initial slope observed in protocol 1 (Table 1) cannot be explained by changes in the magnitude of sympathetic effect on HR.

Limitations

The present study has several limitations. First, we performed the experiment under anesthetic conditions. However, because we compared the effects of yohimbine and clonidine on the sympathetic HR control under the same anesthetic condition, the interpretation of the observed changes in the transfer function may be reasonable. Second, the simulation model in Fig. 6A is not the only model that can be applied to the observed results. Although the model is convenient to explain many aspects of the observed results, other models may also be applicable to the present observation. Third, clonidine can affect HR through non-{alpha}2-adrenergic mechanisms. For instance, clonidine caused bradycardia in {alpha}2ABC-knockout mouse via direct inhibition of cardiac hyperpolarization-activated cyclic nucleotide-gated pacemaker channels (16). While we tried to use medetomidine instead of clonidine, medetomidine did not attenuate myocardial interstitial NE release in response to sympathetic nerve stimulation significantly, at least, at the same dose as clonidine (Fig. 5). Further studies using other agonists might be required to confirm our observations. Finally, we used a weak stimulus rate (0 to 1 Hz) for the yohimbine protocol. Although we had examined the effect of yohimbine using a strong stimulus rate (0–5 Hz) in a preliminary study, the steady-state gain of the transfer function did not increase much (8.4 ± 1.7 vs. 9.0 ± 1.7 beats·min–1·Hz–1, n = 3). Under such strong stimulus condition, the saturation of HR response might have masked the effect of presynaptic inhibition. Therefore, the result of protocol 1 should be carefully interpreted in view of the existence of a stimulus rate-drug interaction effect.

Conclusions

The presynaptic {alpha}2-adrenergic autoinhibition attenuates the dynamic HR response to sympathetic nerve stimulation in the low-frequency range (0.004–0.04 Hz) but not in the high-frequency range (0.05–0.15 Hz). In the time domain, the presynaptic {alpha}2-adrenergic autoinhibition attenuates the steady-state response without affecting the slope of the response in the HR step response (a limiter-like operation). In contrast, pharmacologic augmentation of presynaptic {alpha}2-adrenergic inhibition attenuates the dynamic HR response to sympathetic nerve stimulation in a frequency-independent manner. In the time domain, pharmacologic augmentation of the presynaptic inhibition attenuates not only the steady-state response but also the initial slope of the HR step response (an attenuator-like operation). Presynaptic {alpha}2-adrenergic autoinhibition would be favorable for limiting excess NE outflow at the sympathetic nerve terminals without compromising the rapidity of effector response.


   APPENDIX
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
GRANTS
 REFERENCES
 
Mathematical Modeling of the Sympathetic HR Response

To describe the estimated transfer function, we used a second-order, low-pass filter with pure dead time (L). Figure 7A shows the frequency response of a second-order, low-pass filter with L. Figure 7B shows the corresponding step response. The step response is calculated for 1-Hz sympathetic nerve stimulation. The steady-state gain (K) of the transfer function represents the value of transfer gain as the frequency approaches zero. The K corresponds to the steady-state response in the step-response representation. The natural frequency (fN) determines the upper frequency limit of the low-pass filter. For instance, if the fN were 10 times higher, the frequency axis in Fig. 7A would have to be scaled by a factor of 10, indicating that the system could respond to 10-fold higher frequency input. The phase plot in Fig. 7A indicates that, at the fN, the output is delayed by {pi}/2 radians relative to the input, in the absence of the L. The maximum phase delay of the second-order, low-pass filter is {pi} radians in the absence of L. The L is needed to account for the phase difference between the estimated transfer function and the second-order, low-pass filter. In Fig. 7B, the L corresponds to the time difference between the onset of the step input and the onset of the response. The damping coefficient ({zeta}) characterizes the system response around the fN. As an example, the gain plot shows a slight peak around fN when {zeta} = 0.5 (dotted line). Figure 7B shows that a {zeta} of 0.5 causes an initial overshoot in response to a step change in the input. A system with {zeta} < 1 is called underdamped. On the other hand, the gain plot shows more gradual decrease around fN when {zeta} = 3 (fine solid line). Figure 7B shows that the system responds sluggishly when {zeta} = 3. A system with {zeta} > 1 is called overdamped. A system with {zeta} = 1 is called critically damped (dash-dot line). The {zeta} of the estimated transfer functions ranged from 1.55 to 1.72 in the present study, indicating that the sympathetic HR control system is overdamped. The solid line represents the second-order, low-pass filter with {zeta} = 1.64 and L = 0.82 that is derived from the mean value obtained under control condition in protocol 1.


Figure 7
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 7. Schematic explanation for the frequency response of a second-order, low-pass filter with pure dead time (L; A), and the corresponding step response (B). K, dynamic gain; fN, natural frequency; {zeta}, damping ratio. See APPENDIX for details.

 

   GRANTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
REFERENCES
 
This study was supported by "Health and Labour Sciences Research Grant for Research on Advanced Medical Technology," "Health and Labour Sciences Research Grant for Research on Medical Devices for Analyzing, Supporting and Substituting the Function of Human Body," and "Health and Labour Sciences Research Grant H18-Iryo-Ippan-023" from the Ministry of Health, Labour and Welfare of Japan; "Program for Promotion of Fundamental Studies in Health Science" from the National Institute of Biomedical Innovation; and "Ground-based Research Announcement for Space Utilization" promoted by the Japan Space Forum.


   FOOTNOTES
 

Address for reprint requests and other correspondence: T. Miyamoto, Dept. of Physical Therapy, Faculty of Health Sciences, Morinomiya Univ. of Medical Sciences, Osaka 559-8611, Japan (e-mail: miyamoto{at}morinomiya-u.ac.jp)

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
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 REFERENCES
 

  1. Altman JD, Trendelenburg AU, MacMillan L, Bernstein D, Limbird L, Starke K, Kobilka BK, Hein L. Abnormal regulation of the sympathetic nervous system in {alpha}2A-adrenergic receptor knockout mice. Mol Pharmacol 56: 154–161, 1999.[Abstract/Free Full Text]
  2. Bendat JS, Piersol AG. Single-input/output relationships. In: Random Data Analysis and Measurement Procedures (3rd Ed.). New York: Wiley, 2000, p. 189–217.
  3. Berger RD, Saul JP, Cohen RJ. Transfer function analysis of autonomic regulation. I. Canine atrial rate response. Am J Physiol Heart Circ Physiol 256: H142–H152, 1989.[Abstract/Free Full Text]
  4. Brigham EO. FFT transform applications. In: The Fast Fourier Transform and Its Applications. Englewood Cliffs, NJ: Prentice-Hall, 1988, p. 167–203.
  5. Buhler FR, Bolli P, Amann WF, Erne P, Kiowski W. Sympathetic nervous system in essential hypertension and antihypertensive response to {alpha}2-adrenoceptor stimulation. J Cardiovasc Pharmacol 6: S753–S756, 1984.[Web of Science][Medline]
  6. Franklin GF, Powell JD, and Emani-Naeini A. Dynamic models and dynamic response. In: Feedback Control of Dynamic Systems (2nd Ed.). Boston, MA: Addison-Wesley, 1991, p. 17–144.
  7. Glantz SA. Primer of Biostatistics (5th Ed.). New York: McGraw-Hill, 2002.
  8. Grossman E, Chang PC, Hoffman A, Tamrat M, Goldstein DS. Evidence for functional {alpha}2-adrenoceptors on vascular sympathetic nerve endings in the human forearm. Circ Res 69: 887–897, 1991.[Abstract/Free Full Text]
  9. Hein L, Altman JD, Kobilka BK. Two functionally distinct {alpha}2-adrenergic receptors regulate sympathetic neurotransmission. Nature 402: 181–184, 1999.[CrossRef][Web of Science][Medline]
  10. Jie K, van Brummelen P, Vermey P, Timmermans PB, van Zwieten PA. Modulation of noradrenaline release by peripheral presynaptic {alpha}2-adrenoceptors in humans. J Cardiovasc Pharmacol 9: 407–413, 1987.[Web of Science][Medline]
  11. Kawada T, Ikeda Y, Sugimachi M, Shishido T, Kawaguchi O, Yamazaki T, Alexander J Jr, Sunagawa K. Bidirectional augmentation of heart rate regulation by autonomic nervous system in rabbits. Am J Physiol Heart Circ Physiol 271: H288–H295, 1996.[Abstract/Free Full Text]
  12. Kawada T, Sugimachi M, Shishido T, Miyano H, Sato T, Yoshimura R, Miyashita H, Nakahara T, Alexander J Jr, Sunagawa K. Simultaneous identification of static and dynamic vagosympathetic interactions in regulating heart rate. Am J Physiol Regul Integr Comp Physiol 276: R782–R789, 1999.[Abstract/Free Full Text]
  13. Kawada T, Yamazaki T, Akiyama T, Sato T, Shishido T, Sugimachi M, Inagaki M, Alexander J Jr, Sunagawa K. Liquid chromatographic determination of myocardial interstitial epinephrine. J Chromatogr B Biomed Sci Appl 714: 375–378, 1998.[CrossRef][Medline]
  14. Kawada T, Yamazaki T, Akiyama T, Sato T, Shishido T, Yoshimura R, Inagaki M, Tatewaki T, Sugimachi M, Sunagawa K. Local epinephrine release in the rabbit myocardial interstitium in vivo. J Auton Nerv Syst 78: 94–98, 2000.[CrossRef][Web of Science][Medline]
  15. Knaus A, Zong X, Beetz N, Jahns R, Lohse MJ, Biel M, Hein L. Direct inhibition of cardiac hyperpolarization-activated cyclic nucleotide-gated pacemaker channels by clonidine. Circulation 115: 872–880, 2007.[Abstract/Free Full Text]
  16. Langer SZ. 25 years since the discovery of presynaptic receptors, present knowledge and future perspectives. Trends Pharmacol Sci 18: 95–99, 1997.[CrossRef][Medline]
  17. Langer SZ. Presence and physiological role of presynaptic inhibitory {alpha}2-adrenoreceptors in guinea pig atria. Nature 294: 671–672, 1981.[Medline]
  18. Langer SZ, Adler-Graschinsky E, Giorgi O. Physiological significance of {alpha}-adrenoceptor-mediated negative feedback mechanism regulating noradrenaline release during nerve stimulation. Nature 265: 648–650, 1977.[CrossRef][Web of Science][Medline]
  19. Langer SZ. Presynaptic regulation of catecholamine release. Biochem Pharmacol 23: 1793–1800, 1974.[CrossRef][Web of Science][Medline]
  20. Marmarelis PZ, Marmarelis VZ. The white noise method in system identification. In: Analysis of Physiological Systems. New York: Plenum, 1978, p. 131–221.
  21. Miyamoto T, Kawada T, Yanagiya Y, Takaki H, Inagaki M, Sugimachi M, Sunagawa K. Cardiac sympathetic nerve stimulation does not attenuate dynamic vagal control of heart rate via {alpha}-adrenergic mechanism. Am J Physiol Heart Circ Physiol 287: H860–H865, 2004.[Abstract/Free Full Text]
  22. Nakahara T, Kawada T, Sugimachi M, Miyano H, Sato T, Shishido T, Yoshimura R, Miyashita H, Inagaki M, Alexander J Jr, Sunagawa K. Neuronal uptake affects dynamic characteristics of heart rate response to sympathetic stimulation. Am J Physiol Regul Integr Comp Physiol 277: R140–R146, 1999.[Abstract/Free Full Text]
  23. Niederhoffer N, Hein L, Starke K. Modulation of the baroreceptor reflex by {alpha}2A-adrenoceptors, a study in {alpha}2A knockout mice. Br J Pharmacol 141: 851–859, 2004.[CrossRef][Web of Science][Medline]
  24. Pelayo F, Dubocovich ML, Langer SZ. Regulation of noradrenaline release in the rat pineal through a negative feedback mechanism mediated by presynaptic {alpha}-adrenoceptors. Eur J Pharmacol 45: 317–318, 1977.[CrossRef][Web of Science][Medline]
  25. Philipp M, Brede M, Hein L. Physiological significance of {alpha}2-adrenergic receptor subtype diversity, one receptor is not enough. Am J Physiol Regul Integr Comp Physiol 283: R287–R295, 2002.[Abstract/Free Full Text]
  26. Rump LC, Bohmann C, Schaible U, Schöllhorn J, Limberger N. {alpha}2C-Adrenoceptor-modulated release of noradrenaline in human right atrium. Br J Pharmacol 116: 2617–2624, 1995.[Web of Science][Medline]
  27. Schwartz DD. Activation of alpha-2 adrenergic receptors inhibits norepinephrine release by a pertussis toxin-insensitive pathway independent of changes in cytosolic calcium in cultured rat sympathetic neurons. J Pharmacol Exp Ther 282: 248–255, 1997.[Abstract/Free Full Text]
  28. Sinclair MD. A review of the physiological effects of {alpha}2-agonists related to the clinical use of medetomidine in small animal practice. Can Vet J 44: 885–897, 2003.[Web of Science][Medline]
  29. Starke KM, Gothert M, Kilbinger H. Modulation of neurotransmitter release by presynaptic autoreceptors. Physiol Rev 69: 864–989, 1989.[Free Full Text]
  30. Starke KM. Presynaptic {alpha}-autoreceptors. Rev Physiol Biochem Pharmacol 107: 73–146, 1987.[Web of Science][Medline]
  31. Starke K, Langer SZ. A note on terminology for presynaptic receptors. In: Presynaptic Receptors, edited by Langer SZ, Starke K, and Dubocovich ML. Oxford, UK: Pergamon 1979, p. 1–3.
  32. Starke K, Endo T, Taube HD. Pre- and post-synaptic components in effect of drugs with {alpha}-adrenoceptor affinity. Nature 254: 440–441, 1975.[CrossRef][Web of Science][Medline]
  33. Starke K. Alpha sympathomimetic inhibition of adrenergic and cholinergic transmission in the rabbit heart. Naunyn Schmiedebergs Arch Pharmacol 274: 18–45, 1972.[CrossRef][Web of Science][Medline]
  34. Szabo B, Schramm A, Starke K. Effect of yohimbine on renal sympathetic nerve activity and renal norepinephrine spillover in anesthetized rabbits. J Pharmacol Exp Ther 260: 780–788, 1992.[Abstract/Free Full Text]
  35. Szabo B, Hedler L, Starke K. Peripheral presynaptic and central effects of clonidine, yohimbine and rauwolscine on the sympathetic nervous system in rabbits. Naunyn-Schmiedeberg's Arch Pharmacol 340: 648–657, 1989.[CrossRef][Web of Science][Medline]
  36. Vizi ES, Somogyi GT, Hadhazy P, Knoll J. Effect of duration and frequency of stimulation on the presynaptic inhibition by {alpha}-adrenoceptor stimulation of the adrenergic transmission. Naunyn-Schmiedeberg's Arch Pharmacol 280: 79–91, 1973.[CrossRef][Web of Science][Medline]
  37. Westfall TC. Local regulation of adrenergic neurotransmission. Physiol Rev 57: 659–728, 1977.[Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/5/H1855    most recent
522.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
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 Google Scholar
Google Scholar
Right arrow Articles by Miyamoto, T.
Right arrow Articles by Sugimachi, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miyamoto, T.
Right arrow Articles by Sugimachi, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.