AJP - Regu Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Regul Integr Comp Physiol 281: R1474-R1482, 2001;
0363-6119/01 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 ISI 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 HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Smith, F. M.
Right arrow Articles by Ardell, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Smith, F. M.
Right arrow Articles by Ardell, J. L.
Vol. 281, Issue 5, R1474-R1482, November 2001

Effects of chronic cardiac decentralization on functional properties of canine intracardiac neurons in vitro

F. M. Smith1, A. S. McGuirt2, J. Leger1, J. A. Armour3, and J. L. Ardell4

Departments of 1 Anatomy and Neurobiology and 3 Physiology and Biophysics, Dalhousie University, Halifax, Nova Scotia B3H 4H7, Canada; 2 Department of Physiology, University of South Alabama, Mobile, Alabama 36688; and 4 Department of Pharmacology, James H. Quillen College of Medicine, East Tennessee State University, Johnson City, Tennessee 37614 - 0577


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Although intrinsic cardiac neurons display ongoing activity after chronic interruption of extrinsic autonomic inputs to the heart, the effects of decentralization on individual neurons remain unknown. The objective of this study was to determine the effects of chronic (3-4 wk) surgical decentralization on intracellular properties of, and neurotransmission among, neurons contained within the canine intrinsic right atrial ganglionated plexus in vitro. Properties of neurons from decentralized hearts were compared with those of neurons from sham-operated hearts (controls). Two populations of neurons were identified by their firing behavior in response to intracellular current injection. Fifty-nine percent of control neurons and 72% of decentralized neurons were phasic (discharged one action potential on excitation). Forty-one percent of control neurons and 27% of decentralized neurons were accommodating (multiple discharge with decrementing frequency). After chronic decentralization, input resistance of phasic neurons increased, whereas the duration of afterhyperpolarization of accommodating neurons decreased. Postsynaptic responses to interganglionic nerve stimulation were evoked in 89% of control neurons and 83% of decentralized neurons; the majority of these responses involved nicotinic receptors. These results show that, after chronic decentralization, intrinsic cardiac neurons 1) undergo changes in membrane properties that may lead to increased excitability while 2) maintaining synaptic neurotransmission within the intrinsic cardiac ganglionated plexus.

neurocardiology; autonomic nervous system; nicotinic receptor; intracellular recording; right atrial ganglionated plexus


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

CARDIAC FUNCTION IN MAMMALS is regulated by autonomic neurons located in the central nervous system, the intrathoracic ganglia, and the intracardiac nervous system. These neurons are organized into a hierarchical control system, with the neurons in the heart exerting local, beat-by-beat control over myocardial function (6). Afferent, local-circuit, and efferent intracardiac neurons are in turn organized into a complex intracardiac neuronal network extending throughout the atria and ventricles (3, 5, 6). These neurons display ongoing activity in the beating heart (8, 10-12, 16, 19, 20), and this activity can be modulated by inputs from extracardiac autonomic efferent neurons in the intrathoracic extracardiac ganglia (9, 13) and the central nervous system (16). Neurons in the intrinsic cardiac nervous system also receive inputs from sensory neurites in the atria and ventricles (4, 8, 10, 16, 18, 21). These afferent inputs drive cardio-cardiac reflexes localized within the intrinsic cardiac nervous system. Such short-loop intracardiac reflexes act in concert with longer-loop reflexes, relaying through neurons of extracardiac autonomic ganglia and in the central nervous system to regulate cardiac function (3, 6). The classical view of the role of intrinsic cardiac neurons as simple relays between vagal efferent preganglionic neurons and the myocardium has therefore been modified in the light of recent evidence indicating that intracardiac neurons are capable of operating with some degree of independence from extrinsic neuronal inputs (9).

Intrinsic cardiac neurons with intact extrinsic innervation display ongoing activity that may be correlated with, or independent of, cardiac and respiratory events (4, 8, 10-12, 16, 18-20). Electrical stimulation of neurons within the intrinsic cardiac nervous system can evoke powerful inhibitory or augmentatory effects on cardiac chronotropy and isotropy (11-13, 19-21). Acute cardiac decentralization, performed by surgically interrupting extrinsic nerve pathways to and from the heart, can depress but does not eliminate ongoing activity generated by intrinsic cardiac neurons in situ (8, 10, 12, 13, 16, 18). In addition, the effects on regional cardiac function evoked by activating intracardiac efferent neurons with cholinergic or peptidergic agonists are profoundly obtunded in this state (11, 20). However, after chronic (3-4 wk) decentralization, the level of activity generated by intrinsic cardiac neurons recovers to resemble that in intact hearts, and such activity is continuously modulated by inputs of intracardiac sensory origin (4). Moreover, Priola and Spurgeon (24) showed that the ability of decentralized intrinsic cardiac efferent neurons to modulate cardiac function is restored differentially to separate regions of the heart with time after the decentralization procedure. In their study, local neural control of right atrial cardiodynamics was enhanced relative to that in the other cardiac chambers after chronic decentralization. These findings thus suggest that the intrinsic cardiac nervous system undergoes remodeling after decentralization. The nature of this remodeling is not known, but changes in either intrinsic membrane properties or patterns of synaptic connectivity, or both, within the intracardiac nervous system could contribute to this process. However, whereas data on the effects of decentralization on passive and active electrical properties of peripheral autonomic neurons in general are equivocal (14, 22), evidence is accumulating that synaptic reorganization may occur within autonomic ganglia after interrupting their central neuronal inputs (see Ref. 27 for review).

The aims of this study were therefore, first, to determine the effects of chronic decentralization on the passive and active membrane properties of intrinsic cardiac neurons and, second, to determine whether decentralization altered the number of neurons responding to synaptic inputs or the characteristics of these responses. For this purpose, intracellular recordings were made from neurons of the canine right atrial ganglionated plexus (RAGP) in vitro 1 mo after all extracardiac afferent and efferent inputs were surgically interrupted; these results were compared with control recordings made from neurons of sham-operated hearts.


    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Experiments were done on 21 mongrel dogs of both sexes, weighing 17-27 kg. All experimental procedures in this study conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Washington, DC: National Academy, 1996) and were approved by the Animal Care and Use Committees of East Tennessee State University and the University of South Alabama.

Chronic decentralization of the intracardiac nervous system. Twelve adult mongrel dogs of either sex were pretreated with the antibiotic cephazolin sodium (1 g im) and then anesthetized with pentobarbital sodium (30 mg/kg iv), supplemental doses being provided as needed to maintain a surgical level of anesthesia. The animals were provided positive pressure ventilation and a left T4-T5 thoracotomy was performed under aseptic conditions. Decentralization of the intrinsic cardiac nervous system was achieved by dissection around the major intrapericardial vessels, as described previously (4). This dissection entailed severing all mediastinal cardiac nerves intrapericardially by stripping the adventitia from the common pulmonary artery and by dissecting between this artery and the ascending aorta. In addition, all adventitial tissue around the right pulmonary artery, to the level of its first bifurcation, was removed. The ventrolateral cardiac nerve was transected as it crossed the lateral edge of the left atrium. Tissue surrounding the left superior vein was also removed, and the superior vena cava was cleared of tissue where this vessel penetrated the pericardium. The azygous vein was doubly ligated and sectioned between the ligatures. All fatty tissue lying in the intrapericardial space between the superior vena cava and the ascending aorta superior to the right atrium was removed. To confirm that cardiac decentralization was complete, supramaximal stimulation of the right and left vagosympathetic complexes (5-ms duration, 10-V rectangular pulses at 20 Hz) and ansae subclavia (10 Hz) was performed before and after dissection. If changes in heart rate or atrioventricular conduction persisted during stimulation of these autonomic efferent nerves, further dissections were performed around the major vessels until all cardiac responses to vagal and sympathetic nerve stimulation were eliminated. Previous experience with this decentralization technique has demonstrated that the intrinsic cardiac ganglia under study [RAGP (32)] are not damaged during the procedure, that these ganglia retain the capability for generation of spontaneous activity, and that this activity can be reflexly modified by stimulating intracardiac sensory neurites (4). After decentralization was completed, the thoracic cavity was closed and residual air withdrawn. Analgesic therapy (Buprenex, 0.2 mg im) was given postoperatively at 8-h intervals for 24 h and as needed thereafter. Antibiotic (cephalexin, 500 mg, 2× daily) was administered orally for 7 days after the surgery. The animals were allowed to recover for 3-4 wk. Nine additional animals underwent the same surgical procedure with the exception that no intrapericardial dissections were performed, leaving the extrinsic innervation of the heart intact; these animals constituted the sham-operated control group. Postoperative care for control animals was identical to that of decentralized animals.

Experimental protocols. Three to four weeks after surgery, all animals in this study were reanesthetized with thiopental sodium (15 mg/kg iv), supplemental doses (5 mg/kg iv) being provided every 5-10 min throughout the surgery. The thorax was reopened, and the ansae subclavia and cervical vagi were exposed and stimulated with the same parameters used previously. In decentralized animals, this stimulation confirmed that reinnervation of the heart had not occurred; in the control group stimulation confirmed that extrinsic innervation of the heart remained intact. The pericardial sac was then opened, and the ventral RAGP and its associated epicardial fat deposit were removed, placed into a dissecting dish, and superfused with modified Krebs solution (in mM): 120 NaCl, 25 NaHCO3, 1 NaH2PO4, 5 KCl, 2 MgCl2, 2.5 CaCl2, 11 D-glucose, pH 7.4, equilibrated with a gas mixture containing 95% O2 and 5% CO2 at room temperature. Using a dissecting microscope, we trimmed away most of the fat and transferred the remaining tissue to a recording chamber (5 ml vol) where it was pinned to the silicone rubber chamber floor and superfused with modified Krebs solution (temperature 36°C, flow rate 5-10 ml/min). Total time between removal of tissue from the heart and placement in the recording chamber was ~2 min. The epicardial sheath was then removed, and underlying tissue was dissected to expose ganglia and interganglionic nerves. For intracellular recording, a selected ganglion was freed of most of its associated connective tissue and was mechanically stabilized with a small (0.2 × 0.5 mm) metal platform mounted on a micropositioner.

Pipette electrodes made from standard borosilicate glass capillary tubing were drawn to fine tips on a micropipette puller (model P87, Sutter Instruments, Novato, CA). Electrodes had resistances of 50-80 MOmega when filled with 3 M KCl. To locate and impale individual neurons for recording, electrodes were advanced through the ganglion sheath with a mechanical three-axis micromanipulator. Transmembrane potentials were recorded in current-clamp mode with an intracellular amplifier (model 1600, A-M Systems, Everett, WA). Before a ganglion was penetrated, microelectrode resistance was nulled with the amplifier's bridge-balancing circuitry, and the amplifier offset and electrode tip potential were also nulled to establish the zero volt level relative to the bath reference electrode. The reference electrode consisted of a pipette containing 1% agar dissolved in 3 M KCl with its tip immersed in the bath solution, and it was connected to the amplifier by a silver wire coated with AgCl. Transmembrane potential was taken as the difference between the bath reference potential and the intracellular electrode potential. At the end of trials on each neuron, the electrode was withdrawn into the bath and the zero volt level was confirmed.

Neurons were activated intracellularly by direct injection of current through the recording electrode via voltage-to-current conversion circuitry in the amplifier, driven by rectangular pulses from a stimulator (model S-88, Grass Instrument, Quincy, MA). Plexus nerves connecting to ganglia under study were stimulated by two separate pairs of bipolar wire electrodes connected to both channels of a second stimulator operating through constant-current photoisolation units (Grass PSIU6). Nerve stimulus currents ranged from 10 µA to 5 mA. Current and voltage waveforms were monitored on an oscilloscope during the experiments and were recorded in digital format on videotape (model 3000A, maximum analog frequency response 20 kHz; A. R. Vetter, Rebersberg, PA) for later analysis.

Data analysis. Selected portions of the recorded data were played back from the tape into a personal computer through an analog-to-digital converter (Digidata 1200, sampling frequency 333 kHz; Axon Instruments, Foster City, CA). These data were then analyzed with Axon Instruments pCLAMP6 software. Numerical data are presented as means ± SE. Pairwise comparisons between means were done using Student's two-tailed t-test, with P <=  0.05 for all comparisons.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Intracellular recordings were made from a total of 97 intrinsic cardiac neurons. Thirty-seven neurons were sampled from nine hearts with intact extrinsic cardiac innervation; data obtained from these neurons represent control data for this study. Sixty neurons were sampled from 12 hearts after chronic decentralization of the intrinsic cardiac nervous system. Decentralization was confirmed by the lack of cardiac responses to stimulation of vagus nerves and ansae subclavia before removal of atrial tissue for recording (data not shown).

Membrane properties and responses to intracellular stimulation. The membrane properties and active neuronal responses of control and decentralized neurons to stimuli delivered through the recording electrode are summarized in Table 1. Neurons were divided into two groups, based on their action potential (AP) firing behavior in response to prolonged intracellular depolarizing currents delivered through the recording electrode (Fig. 1). Fifty-nine percent of neurons in the control group discharged only one AP in response to depolarizing stimuli (Table 1; Fig. 1Ai). This type of firing behavior was designated as "phasic." The remainder of the neurons (41%) in the control group, when stimulated, discharged APs at a high initial rate that decremented with time (Table 1; Fig. 1Aii). This type of firing behavior was designated as "accommodating." Mean resting membrane potentials in these two populations of neurons were similar (Table 1), whereas mean whole cell resistance for accommodating neurons was twice the value for phasic neurons (Table 1). Yet, despite the difference in input resistance, the mean threshold voltage for AP generation was not significantly different in the two types of neurons (Table 1). Within the control group, mean AP duration and peak AP amplitude of accommodating neurons were significantly greater than the corresponding values of these variables in phasic neurons. In accommodating neurons, the duration of the afterhyperpolarizations (AHP) was nearly double that of phasic neurons (Table 1), but AHP amplitude was similar in both neuron types.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Membrane properties and characteristics of intracellularly evoked AP and AHP of intrinsic cardiac neurons



View larger version (11K):
[in this window]
[in a new window]
  		Fig. 1.   Responses of membrane potential (top main traces in each panel) of intrinsic cardiac neurons to prolonged depolarizing current delivered intracellularly through the recording electrode (bottom traces). A: neurons sampled from hearts with intact extrinsic innervation (control) displayed two types of firing behavior: phasic neuron (i), discharged one action potential (AP) at start of depolarizing pulse, and accommodating neuron (ii), discharged multiple APs that displayed spike frequency accommodation. Inset trace in each panel shows time course of a single AP (evoked by a 5-ms intracellular depolarizing pulse, stimulus not shown) and afterhyperpolarization (AHP) after the AP; in these traces, the level of the resting potential is indicated by dotted line. AHP duration of phasic neuron was shorter than that of accommodating neuron. B: neurons sampled from decentralized hearts also displayed phasic (i) and accommodating (ii) firing behavior. Inset traces show that in the experimental group, as in the control group, AHP duration of phasic neurons was shorter than that of accommodating neurons. Resting membrane potential for each neuron is indicated at left of each main trace. Vertical bar: 20 mV and 1 nA for main traces; 40 mV for insets. Horizontal bar: 100 ms for main traces and insets.

Both phasic and accommodating neurons were also found in ganglia obtained from decentralized preparations (Table 1, Fig. 1, Bi and Bii). The majority (72%) of neurons sampled from decentralized hearts was phasic; this neuron type displayed a significantly smaller mean AP amplitude and shorter mean duration than the corresponding values in accommodating neurons (Table 1). However, mean whole cell input resistance and mean AHP duration were similar in phasic and accommodating neurons from decentralized hearts (Table 1) in contradistinction to the control condition. Thus the effects of decentralization on intrinsic properties of intracardiac neurons varied depending upon neuron type. With respect to phasic neurons, chronic decentralization of the intrinsic cardiac nervous system resulted in increased mean whole cell resistance and AHP amplitude, whereas resting membrane potential became more negative (Table 1). In accommodating neurons, mean AHP duration was reduced relative to that in control neurons (Table 1).

Responses to stimulation of interganglionic connecting nerves. Synaptic inputs were detected in 89% of neurons sampled from control hearts and in 83% of neurons from decentralized hearts, indicating that there was no significant deficit in the proportion of neurons receiving synaptic inputs after decentralization. All postsynaptic responses to nerve stimulation observed in this study were depolarizing in nature; no hyperpolarizing responses were recorded. Examples of orthodromic responses to nerve stimulation are illustrated in Figs. 2-4. At the threshold intensity of nerve stimulation, these neurons typically displayed a small excitatory postsynaptic potential (EPSP; Fig. 2, second trace from top); graded increases in stimulus intensity then produced higher-amplitude EPSPs with multiple components, indicating summation of multiple synaptic inputs. In nine neurons from control hearts and 12 neurons from decentralized hearts, maximal nerve stimulation evoked EPSPs that did not reach threshold for AP generation (example shown in Fig. 4). Failure to generate synaptically evoked APs in these neurons was not, however, due to a general failure of regenerative responses in these cells, as intracellular stimulation evoked full APs. In the remaining neurons, the amplitude of the composite EPSP exceeded threshold for a regenerative response, and neurons discharged APs. Figure 3A shows that orthodromic responses of neurons derived from decentralized hearts were similar to those for neurons from control hearts (Fig. 2). That such responses were synaptically mediated was confirmed in five cells each from decentralized and control hearts by the elimination of stimulus-evoked EPSPs and APs upon switching to a modified perfusate containing 0 Ca2+ and 10 mM Mg2+ (bottom traces, Figs. 2 and 3A).


View larger version (9K):
[in this window]
[in a new window]
  		Fig. 2.   Excitatory postsynaptic responses of neuron from control heart to single-pulse stimulation of interganglionic connecting nerve (S, arrow; rectangular pulse, 0.5-ms duration, stimulus intensity indicated in µA to left of traces). Graded increases in stimulus intensity (from top down) produced postsynaptic depolarizations of graded amplitude; an AP was generated when depolarization exceeded firing threshold (5th trace from top). Resting membrane potential is indicated at left of this trace. The bottom 2 traces show that hexamethonium (HEX, 100 µM in perfusate for 5 min) and a modified perfusate containing 0 mM Ca2+ and 10 mM Mg2+ (low Ca, high Mg) eliminated responses to nerve stimulation. In all traces, the initial rapid deflection from baseline is a stimulus artifact. Vertical bar represents 10 mV; horizontal bar represents 5 ms.



View larger version (12K):
[in this window]
[in a new window]
  		Fig. 3.   Responses of two neurons from decentralized hearts to single-pulse stimulation of interganglionic connecting nerve (S, arrows; 5-ms duration). A: format as for Fig. 2. Graded increases in stimulus intensity evoked graded increases in depolarization, with an AP resulting when membrane potential exceeded threshold. HEX (100 µM in perfusate for 5 min) or altering the perfusate to contain low Ca, high Mg eliminated responses to nerve stimulation. B: response of intrinsic cardiac neuron to antidromic stimulation. No subthreshold depolarizations were evoked by incremental increases in stimulus amplitude (top and middle traces), and perfusate containing low Ca2+, high Mg2+ did not affect AP generation (bottom trace). Vertical bar represents 10 mV in A and 20 mV in B; horizontal bar represents 5 ms in A and 10 ms in B.



View larger version (7K):
[in this window]
[in a new window]
  		Fig. 4.   Postsynaptic responses of an intrinsic neuron from decentralized heart to single-pulse stimulation (S, arrows) of two different interganglionic connecting nerves (nerve 1, nerve 2). A: separate stimulation of each nerve (1, 800 µA; 2, 3.5 mA; 0.5-ms pulse duration for both) evoked excitatory postsynaptic potentials but no APs (stimulus intensity was set to evoke maximal responses). B: HEX (200 µM in perfusate, 5 min) blocked postsynaptic response to stimulation of nerve 2 but only obtunded the depolarizing response to stimulation of nerve 1. Resting membrane potential is indicated to left of each trace and was unaffected by HEX. Vertical bar: 10 mV; horizontal bar: 20 ms.

In addition to receiving multiple synaptic inputs from axons in a single interganglionic nerve, some neurons from both control and decentralized hearts could be activated when axons coursing in at least two interganglionic nerves were stimulated. In some instances, maximal stimulation of either nerve alone produced subthreshold EPSPs (Fig. 4A, from decentralized heart), and concurrent stimulation of both nerves was still insufficient to evoke an AP. In other cases, concurrent stimulation of two nerves was sufficient to evoke APs (data not shown). It is likely that some intracardiac neurons received inputs from more than two interganglionic nerves, but this was not directly evaluated in this study.

Antidromically mediated responses were also recorded in neurons from both control and decentralized hearts. These responses were characterized by the abrupt appearance of an AP without a preceding EPSP (Fig. 3B, example from decentralized heart). Furthermore, these responses were not blocked in 0-Ca2+, high-Mg2+ perfusate (Fig. 3B). Antidromic responses were also capable of following high-frequency nerve stimulation, producing APs at stimulus frequencies of >100 Hz (data not shown).

Effects of hexamethonium on synaptic responses. Most postsynaptic responses to nerve stimulation (28 of 31 neurons tested in control and decentralized hearts) were eliminated when exposed to the nicotinic antagonist hexamethonium (100 µM in perfusate for 5 min; Figs. 2 and 3A). However, in three neurons (1 from control heart, 2 from decentralized hearts) stimulus-induced depolarizations were blunted but not completely eliminated by hexamethonium (example from decentralized heart shown in Fig. 4B). These responses persisted in concentrations of hexamethonium as high as 500 µM for up to 15 min (data not shown). The doses of hexamethonium used in this study did not affect resting membrane potentials, responses to direct intracellular stimulation, or antidromically evoked responses.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

The results of this study demonstrate that intrinsic cardiac neurons remain active after chronic removal of inputs from extrinsic cardiac neurons in the thorax and the central nervous system. Synaptic communication between neurons in the intracardiac nervous system was also maintained after chronic decentralization, such interconnections being mediated by neurotransmission involving both nicotinic and nonnicotinic receptors. Because no evidence for "pacemaker-like" activity has been reported for neurons within mammalian intrinsic cardiac ganglia, ongoing activity within this nervous system in vivo after decentralization is thus likely to be dependent on synaptic interactions within the ganglionated plexus, generated by afferent feedback of local intracardiac origin (3, 4, 6, 9).

Changes in membrane properties of intrinsic cardiac neurons after decentralization. Intrinsic cardiac neurons can be grouped according to their firing behavior in response to depolarizing current pulses. Potentially, such firing patterns may identify specific subpopulations of intrinsic cardiac neurons that have distinct functional roles in cardiac control, e.g., local circuit, afferent, or efferent neurons (3, 7). Neurons from intact and decentralized hearts were categorized into two broad categories according to their responses to prolonged intracellular current injection: 1) those that discharge phasically and 2) those that display spike frequency accommodation (Fig. 1, Table 1). The relative proportion of phasic to accommodating neurons was greater in both control and decentralized hearts (Table 1). Although this difference in proportion could reflect sampling bias due to the limited number of neurons evaluated, this seems unlikely because neurons were sampled from the same sites in each group, and the mean number of neurons sampled per preparation was similar in both groups (control, 4.1 neurons per preparation; decentralized, 5 neurons per preparation). A possible explanation for the shift in firing pattern may be that chronic decentralization results in a differential decrease in the number of accommodating neurons secondary to loss of extracardiac neuronal inputs. Alternatively, neurons that previously displayed accommodating firing behavior may have been converted to fire phasically after chronic decentralization of the intrinsic cardiac nervous system. In either case, this change could reflect alterations in trophic processes and/or reorganization of synaptic interconnections within the intrinsic cardiac ganglionated plexus. Neurons with phasic and accommodating firing behavior have been previously reported not only in the mammalian intrinsic cardiac nervous system, but also in other parts of the peripheral autonomic nervous system (see Ref. 1 for summary). Xi and coworkers (28, 29) described "S-cells" in the canine heart with a discharge pattern resembling that of phasic neurons in the present study, as well as "tonic" or "R-cells" displaying spike-frequency accommodation. These different types of neurons have also been identified in pig and guinea pig hearts (2, 25, 26). The functional significance of these firing patterns relative to overall cardiac control remains to be determined.

The mean values of membrane properties, regenerative responses, and AHP properties of intrinsic cardiac neurons from control hearts in the present study fell within the ranges of values previously reported for neurons in intact canine (12, 28, 29), porcine (25, 26), and rodent (15, 17, 30) hearts, as well as being similar to values for other autonomic neurons (1). In the present study, chronic decentralization exerted differential effects on the electrical properties of intrinsic cardiac neurons. Comparing control and decentralized groups, we found no difference in mean values of resting membrane potential in accommodating neurons. Yet the mean resting potential of phasic neurons sampled from decentralized hearts was significantly more negative than that of phasic neurons from control hearts. Although this difference was relatively small (4 mV), it may represent an important difference when translated to the whole population of phasic neurons in decentralized hearts. For instance, a hyperpolarization of 4 mV, if consistent across this neuronal population, could selectively depress the excitability of these neurons by increasing the amount of depolarization necessary to bring membrane potential to firing threshold. This in turn could help maintain stability within this neuronal population as it functions without benefit of extracardiac inputs.

In the present study, mean values for whole cell resistance of neurons from control hearts fell within the range of values reported for normal mammalian intrinsic cardiac neurons (12, 15, 17, 25, 26, 29, 30). Accommodating neurons from control hearts had a mean whole cell resistance value that was twice that of phasic neurons in this group. Whole cell resistance is the inverse of overall membrane ionic conductance. Therefore, a difference in resistance between these groups of neurons implies a corresponding difference in conductance in the opposite direction. Differences in whole cell resistance between the two neuron types would thus suggest that different transmembrane current flows are required to produce comparable displacements of membrane potential. This measure may therefore serve as an approximate index of neuronal excitability. Accordingly, cells exhibiting higher resistances would require less current flow to reach threshold for AP generation and may potentially be more excitable than cells with lower resistances. This is consistent with the finding that accommodating neurons in neurally intact hearts have a higher resistance than phasic neurons. After chronic decentralization, mean resistance of phasic neurons was significantly greater than the resistance of this type of neuron in control hearts. In fact, after decentralization, the resistance of phasic cells increased to nearly the same value as that of accommodating neurons. On the other hand, decentralization exerted no significant effects on the resistance exhibited by accommodating neurons. These results suggest that chronic decentralization of the intrinsic cardiac nervous system induces differential effects on membrane conductances of different populations of neurons by decreasing overall ion conductance of phasic neurons while minimally affecting conductance of accommodating neurons.

The mean amplitude and duration of APs in accommodating neurons obtained from control and decentralized hearts were significantly greater than the corresponding values of phasic neurons in both groups. The importance of these differences to the function of the intrinsic cardiac nervous system is not clear. However, regarding the characteristics of AHP potentials, decentralization produced an increase in AHP amplitude in phasic neurons compared with that of control neurons. As suggested above for whole cell resistance, decentralization may affect membrane conductances of phasic intrinsic cardiac neurons, some of which may be responsible for the time course of the AHP. The mean duration of AHPs in accommodating neurons was significantly greater than that of phasic neurons derived from control hearts. In contradistinction to its effect on AHP amplitude, decentralization reduced AHP duration in accommodating neurons to values similar to those of phasic neurons. The observation that AHP duration was greater in neurons with higher intrinsic firing frequencies extends and confirms observations derived from porcine (25) and guinea pig (15) intrinsic cardiac neurons. The physiological significance of this is not apparent, but these findings together with the results of the present study suggest that there may be no direct relationship between AHP duration and firing behavior in intrinsic cardiac neurons. This is contrary to the contention of Adams and Harper (1) that firing rate in autonomic neurons may be set at least partly by AHP duration. Data from the present study indicate that some membrane electrical properties of intrinsic cardiac neurons can change after chronic decentralization. These changes include: 1) increases in resistance of phasic neurons, possibly leading to an increase in excitability of this neuronal type; 2) a reduction in AHP duration in accommodating neurons; and 3) a hyperpolarizing shift in resting membrane potential in phasic neurons, an effect that could modify the efficacy of intraganglionic neuromodulatory mechanisms.

Synaptic integration within the intrinsic cardiac nervous system. Based on the classical view that intrinsic cardiac ganglia function solely as relay stations for vagal preganglionic commands to the myocardium, it follows that chronic decentralization of the intrinsic cardiac nervous system should eliminate all synaptic inputs to intrinsic cardiac neurons. This proved not to be the case. The proportion of neurons displaying postsynaptic depolarizing responses to interganglionic nerve stimulation in control hearts was 89%; this proportion was unchanged after chronic decentralization (83%). These values are higher than the proportions of intrinsic cardiac neurons reported to receive synaptic inputs in previous studies [canine heart, 70% (31); porcine heart, 47% (26)]. In the present study, the actual number of synaptic inputs to individual neurons is likely to have been underestimated because many sampled neurons received inputs from axons in multiple interganglionic nerves; in our experiments, we were able to evaluate at most two input pathways to individual neurons. The finding that decentralized neurons did not show a significant deficit in the number of functional synaptic inputs suggests that 1) a large fraction of the inputs to neurons in control hearts must have originated from sources within the heart that were not affected by decentralization, or 2) new inputs arose as a consequence of axonal sprouting from other intrinsic cardiac neurons, or 3) both of these may obtain.

Unitary synaptic inputs (single fast EPSPs suprathreshold for AP generation) elicited by local nerve stimulation have been observed in guinea pig intrinsic cardiac neurons (15). Such unitary inputs were not identified in the present study or in other studies of canine (29) or porcine (25, 26) intrinsic cardiac neurons. Neurons from both control and decentralized hearts received multiple subthreshold synaptic inputs from axons in one (Figs. 2 and 3) or more (Fig. 4) interganglionic nerves, as discussed above, and summation of fast EPSPs from several axon terminals was required to exceed threshold for AP generation. This finding suggests that integration of multiple inputs at the level of single neurons is an important component of ganglionic transmission in both normally innervated and decentralized hearts. That these inputs were orthodromically mediated was confirmed by blockade of synaptic activity by the application of a perfusate containing no calcium and 10 mM magnesium.

Most synaptic inputs to neurons from control and decentralized hearts were mediated by fast nicotinic cholinergic neurotransmission, as indicated by the effect of hexamethonium blockade on postsynaptic responses elicited by electrical stimulation of interganglionic nerves. For animals with intact extrinsic cardiac innervation, these data confirm previous results (29, 31). One neuron from a control heart and two neurons from decentralized hearts had postsynaptic depolarizing responses that were not eliminated by hexamethonium, even at doses up to five times greater than that required to block all other responses to nerve stimulation in the ganglia under study. These neuronal responses were presumably mediated via nonnicotinic postsynaptic receptors. This supports the finding in the in situ canine intracardiac nervous system that, although hexamethonium may decrease the overall level of neuronal activity, this agent does not abolish the reflex modulation of such activity evoked by stimulation of intrinsic cardiac afferent neurons (20). Nonnicotinic neurotransmission has also been reported in studies of pig intracardiac neurons (25, 26). However, the precise roles of nicotinic and nonnicotinic synaptic mechanisms in regulation of overall activity within the intrinsic cardiac nervous system remain to be determined.

Perspectives

Taken together, these data demonstrate that intracardiac neurons can survive the chronic removal of all their extracardiac neuronal inputs. Some electrophysiological properties displayed by intrinsic cardiac neurons become modified after chronic decentralization. Furthermore, intrinsic cardiac neurons retain their capacity for synaptic interactions when chronically disconnected from other intrathoracic and central neurons. In conjunction with previous results indicating the maintenance of functional intracardiac reflex loops after chronic cardiac decentralization (4), the present study indicates that neural substrates for local reflex control of cardiodynamics can operate independently of inputs from the central nervous system. One consequence of the existence of a higher proportion of phasic than accommodating neurons in both neurally intact and decentralized hearts could be that the maximum frequency of transmission of APs within the intracardiac nervous system is limited. That is, the average firing rate of the whole population of intrinsic cardiac neurons will be relatively low no matter how strong the incoming presynaptic drive, whether of extra- or intracardiac origin. This is supported by the finding in in vivo experiments that even very high frequency stimulation of extrinsic cardiac nerves evokes few APs from intrinsic cardiac neurons (16). In contrast to the relatively high proportion of phasic intracardiac neurons reported here, Mendelowitz (23) reported that the majority of vagal preganglionic neurons projecting to the heart from the medulla display high-frequency discharge rates when stimulated intracellularly. Moreover, Smith (25) reported that, in the pig heart, the majority of neurons receiving synaptic inputs from the vagus nerve were phasic. Taken together, these data suggest that phasic neurons receiving vagal preganglionic inputs may act as low-pass filters in hearts with intact innervation. The intrinsic cardiac nervous system would therefore exert a smoothing and limiting function to mitigate potential imbalances in neural control of the heart arising from excessive vagal drive or in response to acute pathophysiological events such as myocardial ischemia (3, 7). We have shown here that chronic decentralization of the intrinsic cardiac nervous system apparently increased the proportion of phasic to tonic neurons, and such remodeling could thus impose an even lower limit on the maximum frequency of neurotransmission within the intrinsic cardiac nervous system. Therefore, in situations where autonomic inputs to the heart become disrupted, the potential for alterations in neuronal properties shown in the present study may have important consequences for adaptive information processing within the intrinsic cardiac nervous system in the coordination of regional cardiac function.


    ACKNOWLEDGEMENTS

This work was supported by National Heart, Lung, and Blood Institute Grant HL-58140 (to J. L. Ardell) and by grants from the Medical Research Council of Canada (to F. M. Smith and J. A. Armour) and the New Brunswick and Nova Scotia Heart and Stroke Foundations (to F. M. Smith and J. A. Armour). F. M. Smith was a Research Scholar of the Heart and Stroke Foundation of Canada for a portion of this study.


    FOOTNOTES

Address for reprint requests and other correspondence: F. M. Smith, Dept. of Anatomy and Neurobiology, Dalhousie Univ., Halifax, NS B3H 4H7 Canada (E-mail: fsmith{at}tupdean2.med.dal.ca).

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.

Received 11 August 2000; accepted in final form 11 July 2001.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Adams, DJ, and Harper AA. Electrophysiological properties of autonomic ganglion neurons. In: The Autonomic Nervous System. Autonomic Ganglia, edited by McLachlan EM.. Reading, UK: Harwood Academic, 1995, p. 153-212.

2.   Allen, TGJ, Hassall CJS, and Burnstock G. Mammalian intrinsic cardiac neurones in cell culture. In: Neurocardiology, edited by Armour JA, and Ardell JL.. New York: Oxford University Press, 1994, p. 115-138.

3.   Ardell, JL. Neurohumoral control of cardiac function. In: Heart Physiology and Pathophysiology, edited by Sperelakis N.. San Diego: Academic, 2001, p. 45-59.

4.   Ardell, JL, Butler CK, Smith FM, Hopkins DA, and Armour JA. Activity of in vivo atrial and ventricular neurons in chronically decentralized canine hearts. Am J Physiol Heart Circ Physiol 260: H713-H721, 1991[Abstract/Free Full Text].

5.   Armour, JA. Intrinsic cardiac neurons. J Cardiovasc Electrophysiol 2: 331-341, 1991.

6.   Armour, JA. Peripheral autonomic neuronal interactions in cardiac regulation. In: Neurocardiology, edited by Armour JA, and Ardell JL.. New York: Oxford University Press, 1994, p. 219-244.

7.   Armour, JA. Myocardial ischaemia and the cardiac nervous system. Cardiovasc Res 41: 41-54, 1999[Abstract/Free Full Text].

8.   Armour, JA, and Hopkins DA. Activity of in vivo canine ventricular neurons. Am J Physiol Heart Circ Physiol 258: H326-H336, 1990[Abstract/Free Full Text].

9.   Armour, JA, Collier K, Kember G, and Ardell JL. Differential selectivity of cardiac neurons in separate intrathoracic autonomic ganglia. Am J Physiol Regulatory Integrative Comp Physiol 274: R939-R949, 1998[Abstract/Free Full Text].

10.   Armour, JA, and Hopkins DA. Activity of canine in situ left atrial ganglion neurons. Am J Physiol Heart Circ Physiol 259: H1207-H1215, 1990[Abstract/Free Full Text].

11.   Armour, JA, Huang MH, and Smith FM. Peptidergic modulation of in situ canine intrinsic cardiac neurons. Peptides 14: 191-202, 1993[ISI][Medline].

12.   Armour, JA, Smith FM, Losier AM, Ellenberger HH, and Hopkins DA. Modulation of intrinsic cardiac neuronal activity by nitric oxide donors induces cardiodynamic changes. Am J Physiol Regulatory Integrative Comp Physiol 268: R403-R413, 1995[Abstract/Free Full Text].

13.   Butler, CK, Smith FM, Cardinal R, Murphy DA, Hopkins DA, and Armour JA. Cardiac responses to electrical stimulation of discrete loci in canine atrial and ventricular ganglionated plexi. Am J Physiol Heart Circ Physiol 259: H1365-H1373, 1990[Abstract/Free Full Text].

14.   Dun, N, Nishi S, and Karczmar AG. Electrical properties of the membrane of denervated mammalian sympathetic ganglion cells. Neuropharmacology 15: 219-223, 1976[ISI][Medline].

15.   Edwards, FR, Hirst GDS, Klemm MF, and Steele PA. Different types of ganglion cell in the cardiac plexus of guinea-pigs. J Physiol (Lond) 486: 453-471, 1995[ISI][Medline].

16.   Gagliardi, M, Randall WC, Bieger D, Wurster RD, Hopkins DA, and Armour JA. Activity of in vivo canine cardiac plexus neurons. Am J Physiol Heart Circ Physiol 255: H789-H800, 1988[Abstract/Free Full Text].

17.   Hardwick, JC, Mawe GM, and Parsons RL. Evidence for afferent fiber innervation of parasympathetic neurons of the guinea-pig cardiac ganglion. J Auton Nerv Syst 53: 166-174, 1995[ISI][Medline].

18.   Huang, MH, Ardell JL, Hanna BD, Wolf SG, and Armour JA. Effects of transient coronary artery occlusion on canine intrinsic cardiac neuronal activity. Integr Physiol Behav Sci 28: 5-21, 1993[Medline].

19.   Huang, MH, Smith FM, and Armour JA. Amino acids modify activity of canine intrinsic cardiac neurons involved in cardiac regulation. Am J Physiol Heart Circ Physiol 264: H1275-H1282, 1993[Abstract/Free Full Text].

20.   Huang, MH, Smith FM, and Armour JA. Modulation of in situ canine intrinsic cardiac neuronal activity by nicotinic, muscarinic, and beta -adrenergic agonists. Am J Physiol Regulatory Integrative Comp Physiol 265: R659-R669, 1993[Abstract/Free Full Text].

21.   Huang, MH, Wolf SG, and Armour JA. Ventricular arrhythmias induced by chemically modified intrinsic cardiac neurones. Cardiovasc Res 28: 636-642, 1994[Abstract/Free Full Text].

22.   McLachlan, EM. The formation of synapses in mammalian sympathetic ganglia reinnervated with preganglionic or somatic nerves. J Physiol (Lond) 237: 217-242, 1974[Abstract/Free Full Text].

23.   Mendelowitz, D. Firing properties of identified parasympathetic cardiac neurons in nucleus ambiguus. Am J Physiol Heart Circ Physiol 271: H2609-H2614, 1996[Abstract/Free Full Text].

24.   Priola, DV, and Spurgeon HA. Cholinergic sensitivity of the denervated canine heart. Circ Res 41: 600-606, 1977[Free Full Text].

25.   Smith, FM. Extrinsic inputs to intrinsic neurons in the porcine heart in vitro. Am J Physiol Regulatory Integrative Comp Physiol 276: R455-R467, 1999[Abstract/Free Full Text].

26.   Smith, FM, Hopkins DA, and Armour JA. Electrophysiological properties of in vitro intrinsic cardiac neurons in the pig (Sus scrofa). Brain Res Bull 28: 715-725, 1992[ISI][Medline].

27.   Taxi, J, and Eugène D. Effects of axotomy, deafferentation, and reinnervation on sympathetic ganglionic synapses: a comparative study. Int Rev Cytol 159: 195-263, 1995[ISI][Medline].

28.   Xi, X, Randall WC, and Wurster RD. Electrophysiological properties of canine cardiac ganglion cell types. J Auton Nerv Syst 47: 69-74, 1994[ISI][Medline].

29.   Xi, X, Thomas JX, Jr, Randall WC, and Wurster RD. Intracellular recordings from canine intracardiac ganglion cells. J Auton Nerv Syst 32: 177-182, 1991[ISI][Medline].

30.   Xi-Moy, SX, and Dun NJ. Potassium currents in adult rat intracardiac neurones. J Physiol (Lond) 486: 15-31, 1995[ISI][Medline].

31.   Xi-Moy, SX, Randall WC, and Wurster RD. The nicotinic and muscarinic synaptic transmission in canine intracardiac ganglion cells innervating the sinoatrial node. J Auton Nerv Syst 42: 201-214, 1993[ISI][Medline].

32.   Yuan, BX, Ardell JL, Hopkins DA, Losier AM, and Armour JA. Gross and microscopic anatomy of the canine intrinsic cardiac nervous system. Anat Rec 239: 75-87, 1994[Medline].

Am J Physiol Regul Integr Comp Physiol 281(5):R1474-R1482
0363-6119/01 $5.00 Copyright © 2001 the American Physiological Society


This article has been cited by other articles:


Home page
 Am. J. Physiol. Regul. Integr. Comp. Physiol.Home page
H. M. Stauss
Heart rate variability
Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2003; 285(5): R927 - R931.
[Full Text] [PDF]

Home page
 Am. J. Physiol. Regul. Integr. Comp. Physiol.Home page
R. C. Arora, R. Cardinal, F. M. Smith, J. L. Ardell, L. J. Dell'Italia, and J. A. Armour
Intrinsic cardiac nervous system in tachycardia induced heart failure
Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2003; 285(5): R1212 - R1223.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 ISI 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 HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Smith, F. M.
Right arrow Articles by Ardell, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Smith, F. M.
Right arrow Articles by Ardell, J. L.

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