| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Cardiology, Technical University RWTH, Aachen, Germany; Departments of 2Physiology and Biophysics and 3Physics, Dalhousie University, Halifax, Nova Scotia, Canada; 4Department of Pharmacology, James H. Quillen College of Medicine, East Tennessee State University, Johnson City, Tennessee; and 5Department of Pharmacology, Faculty of Medicine, University of Montréal, Montréal, Québec, Canada
Submitted 21 October 2005 ; accepted in final form 15 February 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
atrial neuron; intrinsic cardiac nervous system; myocardial ischemia; stochastic control; ventricular neuron
Neurons in intrinsic cardiac vs. intrathoracic extracardiac ganglia are known to display noncoherent behavior (5). As a result, if a pathological process compromises one level within this neuronal hierarchy, other elements can compensate to ensure adequate regional cardiac control (2, 5). It is also known that neurons in a single, intrinsic cardiac ganglionated plexus display interdependent behavior over short time scales (32). Their local interactions likely contribute to coordination of efferent outputs to specific cardiac indexes, as with the integrated neuromodulation of the sinoatrial (SA) nodal complex (6, 10, 25).
There is functional and anatomical evidence to support the concept of local circuit neuron-dependent interactions occurring among various intrinsic cardiac ganglionated plexuses. For example, Gray et al. (15) demonstrated the existence of anatomical interconnections among neurons within the right atrial and the posterior atrial ganglionated plexuses, both of which have been associated with reflex control of SA nodal function (3, 24, 27, 28). Data likewise indicate that coordination of SA and atrioventricular nodal function may reflect interconnections within the intrinsic cardiac nervous system and/or common shared inputs from the extracardiac sources (21, 26). It remains to be established how neurons in atrial and ventricular ganglionated plexuses interact on an ongoing basis in the short-term control of disparate cardiac regions.
Given the fact that, as yet, we do not understand how atrial and ventricular neurons interact on a short-term basis, the coherence of their activities was determined to elucidate how they concurrently transduce physiological as well as pathological states. This is particularly relevant when considering targeting select intrinsic cardiac neurons therapeutically to stabilize control over regional cardiac electrical or mechanical indexes in the presence of cardiac pathology (7, 19, 31).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Canines were tranquilized with pentothal sodium (15–20 mg/kg iv) and then anesthetized with pentothal sodium (5 mg/kg iv to effect every 5–10 min for the duration of the surgical procedures). Thereafter, the animals were intubated, and positive-pressure ventilation was maintained with a Bird Mark 7A ventilator at a respiratory rate of about 30 times/min, using a gas mixture of 95% O2 and 5% CO2. After all the surgery had been completed, anesthesia was changed to 
-chloralose that was first administered as a bolus (25–50 mg/kg iv) and then as repeat doses (25 mg/kg iv) every hour or less throughout the experiments, as required. Noxious stimuli were applied to a paw periodically throughout the experiments to ascertain (limb withdrawal, heart rate changes) adequacy of the anesthesia.
Experimental procedures. A bilateral thoracotomy was performed in the fourth intercostal space. Umbilical tape was placed around the inferior vena cava and the descending thoracic aorta so that each of these vessels could be transiently occluded later in the experiments. A 3-0 silk ligature was placed around the ventral descending coronary artery 1 cm from its origin and lead through polyethylene tubing so that this vessel could be transiently occluded later in the experiments. A midline incision was made in the neck to expose the caudal cervical vagosympathetic trunks, and silk threads were placed around them so that they could be severed later in the experiments. Silk ligatures were also placed around the right and left subclavian ansae so that the middle cervical, mediastinal, and intrinsic cardiac ganglia could be decentralized from the central nervous system.
A lead II ECG was recorded. Left atrial and left ventricular chamber pressures, as well as aortic pressure, were monitored using Bentley Trantec model 800 transducers connected, respectively, to a PE-50 catheter placed in the left atrial cavity via its appendix, a Cordis no. 7 pig-tail catheter inserted into the left ventricular chamber via the right femoral artery and a Cordis no. 6 catheter inserted into the ascending aorta via the left femoral artery. Miniature solid-state pressure transducers (Konigsberg Instruments, Pasadena, CA, model P190; 5-mm diameter, 1.5-mm thick) were inserted in the right ventricular conus and into the left ventricular ventral wall to record regional intramyocardial pressures in both chambers. These sensing devices were employed because ventricular chamber pressure by itself is inadequate for detecting regional ventricular inotropic alterations induced by activating select populations of cardiac efferent neurons. All data, including intrinsic cardiac neuronal activity (see below), were recorded on an Astro-Med, model MT 9500, eight-channel rectilinear recorder. Data were stored on VHS tape (T120 Scotch, 3M Canada, London, ON, Canada) using a VCR recorder (A. R. Vetter, model 820, Rebersburg, PA) for later analysis.
Recording neuronal activity.
The activity generated by neurons in a locus of the right atrial ganglionated plexus (atrial neurons) was recorded concurrently with the activity generated by neurons in a locus of the cranial medial ventricular ganglionated plexus (ventricular neurons). The ventral pericardium was incised and retracted laterally to expose fat on the ventral surface of the right atrium that contains the ventral component of the right atrial ganglionated plexus and the fat overlying the ventral interventricular grove that contains the ventral septal component of the cranial medial ventricular ganglionated plexus (35). Separate circular rings of heavy-gauge wire were placed on epicardial fat located on the ventral surface of the right atrium and cranial interventricular groove to minimize epicardial motion. Fatty tissues in these two regions were explored via two separate tungsten microelectrodes mounted on micromanipulators. These recording microelectrodes had 250-µm shank diameters, exposed tips of 10 µm, and impedances of 9–11 M
at 1,000 Hz. Indifferent electrodes for each were attached to the mediastinum. Epicardial fatty tissues were examined from their surfaces to the underlying cardiac musculature.
Signals generated by atrial and ventricular neurons were differentially amplified via separate Princeton Applied Research model 113 amplifiers that had band-pass filters set at 300 Hz to 10 kHz and amplification ranges of x100–500. The outputs of these devices were further amplified (x50–200) and filtered (bandwidth 100 Hz to 2 kHz) by two optically isolated amplifiers (Applied Microelectronics Institute, Halifax, NS, Canada). The activity generated by individual neurons, as identified by the amplitude and shape of recorded action potentials with signal-to-noise ratios greater than 3:1, were recorded. Using these techniques and criteria, action potentials generated by cell bodies and/or dendrites rather than axons of passage can be identified (32).
Interventions.
Loci on the epicardium of the left atrium, the right ventricular conus or sinus, as well as the ventral or lateral surfaces of the cranial left ventricle, are known to contain sensory inputs to the intrinsic cardiac nervous system (5). These regions were touched sequentially with a saline-soaked cotton swab. Right atrial tissues were not disturbed in order not to disturb the recording electrode in right atrial fat. Thereafter, chemicals known to activate cardiac sensory neurites (5, 32) were applied (60–100 s) individually to ventricular epicardial loci that responded to mechanical stimuli via 1-cm x 1-cm gauze squares soaked with 0.5 ml of a chemical. After removing a square containing a chemical, each site was washed with normal saline (
2 ml/s) for 30 s; each response terminated on average within 1 min of chemical removal. Gauze squares soaked with room-temperature normal saline were also applied to identified ventricular epicardial sensory fields to determine whether neuronal responses elicited by epicardial chemical application were due to vehicle effects or the mechanical effects elicited by gauze squares. The following chemicals were applied individually to ventricular epicardial sensory fields: adenosine (1 µM); angiotensin II (1 µM); bradykinin (1 µM); the -adrenoceptor agonist phenylephrine (1 µM); the 2-adrenoceptor agonist clonidine (1 µM); the 
1-adrenoceptor agonist dobutamine (1 µM); the 2-adrenoceptor agonist terbutaline (1 µM); the nitric oxide donor S-nitroso-N-acetyl-penicillamine (SNAP; 10 µM); and the sodium channel modifier veratridine (5 x 10–6 g). The order of chemical application varied among experimental animals. Active chemicals were reapplied at least twice to the same epicardial locus to verify response reproducibility.
To discretely activate one neuronal population while recording activity change generated by both, nicotine was administered to right atrial neurons via their local arterial blood supply. Nicotine is known to activate intrinsic cardiac neurons in a consistent manner when administered into their local arterial blood supply (23). Nicotine (0.1 ml bolus of a 100 µg/ml solution) was administered locally via a 24-French catheter placed in the right coronary artery. That cannula was threaded retrograde to flow so that its tip lay 1 cm proximal to the arterial branch that supplied blood to the right atrial ganglionated plexus without obstructing blood flow in either artery. The cannula was fixed in place with 0.5 ml of super-glue adhesive. PE-15 tubing was inserted into the hub of this catheter with a stopcock in order that nicotine could be administered repeatedly into the local arterial blood supply of ventral, right atrial neurons. Monitored cardiac indexes were unaffected by cannula placement. Postmortem examination of appropriate catheter placement was confirmed by injecting methylene blue dye through this catheter. By this method, nicotine could be delivered into the regional arterial blood supply of right atrial neurons and other adjacent tissues while leaving the flow of blood in the right coronary artery patent. The arteries supplying blood to the cranial medial ventricular ganglionated plexus arise along the proximal left anterior descending coronary artery (22); thus neurons in this ganglionated plexus were not in the direct perfusion path of such right-sided intracoronary administered nicotine. To control for potential systemic effects elicited by local nicotine administration or epicardial chemical application, each agent was administered into descending aorta blood in the same doses.
To determine whether altered cardiovascular mechanical status affected the activity generated by either population of neurons, the inferior vena cava and then the descending thoracic aorta were partially occluded for 3–5 s. Thereafter, the ventral descending coronary artery was occluded for 30 s. Once all of the interventions had been completed, the cervical vagosympathetic complexes were severed, as were all connections between the stellate ganglia and spinal cord. Following decentralizing intrathoracic autonomic ganglia from the central nervous system, interventions that had previously induced responses were repeated. At least 5 min were allowed to elapse between interventions to ensure preparation stabilization.
Data analysis. Spontaneous cardiodynamic fluctuations were minimal during control periods: heart rate varying less than 5 beats/min, and systolic pressure fluctuating less than 5 mmHg. Thresholds for classifying induced cardiovascular changes were chosen to be greater than these ranges. Action potentials recorded simultaneously from the right atrial and ventral ventricular ganglionated plexuses were counted for 30-s periods to establish average activity immediately before and during maximal responses elicited by each intervention. Fluctuations in the amplitude of action potentials generated by a unit varied by <50 µV over several minutes, with action potentials retaining the same configurations over time. Action potentials recorded from each locus with the same configuration and amplitude (±50 µV) were considered to be generated by a single unit. Recorded action potentials with signal-to-noise ratios greater than 3:1 were analyzed. Changes in neuronal activity and monitored cardiac indexes induced by each intervention were evaluated by comparing data generated immediately before each intervention with data obtained at the point of maximum change during that intervention. Data were expressed as means ± SD. One-way ANOVA and paired t-test with Bonferroni correction for multiple tests were used for statistical comparisons. A significance value of P < 0.05 was used for these determinations.
The coupling of activities generated by atrial and ventricular neurons was determined by continuous cross-covariance analysis (32). To accomplish this, the activity generated by atrial and ventricular neurons was digitized offline at sampling frequencies of 2,000 Hz. This was done by window discriminating the action potentials recorded at each site, i.e., values below a threshold were zeroed, and those above the assigned window value were truncated to have unit value. In that manner, the activity recorded from neurons in the two ganglionated plexuses could be computed simultaneously so that the activities generated by identified atrial and ventricular neurons could be cross-correlated over time. This was done by using a sliding window of data lasting 10 s (32). This approach permitted the determination of the ongoing cross-covariance function between identified atrial and ventricular neurons in basal states and in response to chemical or ischemic activation of local cardiac sensory neurites. Continuous analysis of heart rate and cardiac inotropic function was performed over the same time periods to compare alterations in neuronal activity with concomitant changes in the monitored cardiac indexes.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
80% of active sites) or the cranial medial surface of the left ventricle (20% of active sites).
|
|
|
All multimodal neurons responded to more than one of the chemicals applied to ventricular epicardial loci. Both populations were activated when angiotensin II, clonidine, dobutamine, or the nitric oxide donor SNAP were applied individually to the right ventricular outflow tract (Table 2). Fewer responded when chemicals where applied to the ventral cranial, left ventricular epicardium. Right atrial neurons were also activated consistently by ventricular application of adenosine, bradykinin, terbutaline, and veratridine; phenylephrine suppressed neuronal activity. These responses took time to develop (Fig. 2) and lasted for 1 min after chemical removal and epicardial rinsing. Reapplication of each chemical to previously identified epicardial sites induced similar neuronal responses. Cardiovascular variables were unaffected by epicardial chemical application, presumably due to the small quantities of chemicals tested. Neuronal activity and monitored cardiac variables were unaffected by epicardial application of gauze squares soaked with room-temperature normal saline. Systemic administration of these chemicals in the doses studied did not modify recorded variables.
|
Neuronal activity increased in a transient fashion immediately following acute decentralization of the intrathoracic nervous system (atrial neurons: 7.5 ± 8.7 to 31.5 ± 45.7 impulses/min; ventricular neurons: 11.2 ± 10.4 to 24.5 ± 21.8 impulses/min; P < 0.01). Within 5 min of acute decentralization, the activities of both populations returned to baseline values. Thereafter, reapplication of the same chemicals to previously responsive epicardial sites affected fewer neurons, such that neuronal activity did not change overall. As an example, local epicardial application of SNAP increased the activity generated by both populations in the intact, but not acutely decentralized, state (Table 2).
Regional ventricular ischemia. In the intact state, transient occlusion (30 s) of the left ventral descending coronary artery distal to the site of origin of the small arteries supplying blood to the ventral ventricular ganglionated plexus enhanced the activity generated by both neuronal populations (Fig. 3; Table 1). Recorded cardiovascular variables were unaffected overall by these brief occlusions. Following acute decentralization of the intrinsic cardiac nervous system, coronary artery occlusion no longer generated significant activity changes.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
It is now recognized that the intrinsic cardiac nervous system does not act as a simple monosynaptic relay station for central efferent neuronal control of regional cardiac function (5, 30). As demonstrated in this study, this target organ nervous system processes sensory information arising from various regions of the heart (Tables 1 and 2). As a consequence of their multiple multimodal inputs, the activity generated by most atrial and ventricular neurons is irregular with respect to time. Given their capacity to transduce multiple chemicals, including those known to be liberated by the ischemic myocardium, such as adenosine (29) and bradykinin (17), most intrinsic cardiac neurons respond to regional ventricular ischemia. Transduction of myocardial ischemia involves mechanosensory and sensory inputs, with chemosensitive being the predominant signal (5a). Previous data indicate that the majority of neurons whose activity can be recorded by the techniques employed in this study represent local circuit ones (2, 4). That is because of their predominance numerically as well as the relatively large size of their somata with respect to other intrinsic cardiac neuronal populations (2, 4). Presumably, most neurons identified in this study belonged to that category.
Limited subpopulations of intrinsic cardiac neurons display cardiac phase-related activity, reflective of primary mechanosensory inputs arising from restricted cardiac regions (5, 32). Presumably, such inputs accounted for the fact that a small (3%) population of solely mechanotransducing neurons was identified in both ganglionated plexuses studied. Because of the similar nature of their inputs, they displayed tightly coupled behavior (Fig. 1). Thus commonality in mechanosensitive sensory inputs may impose a degree of coordination among disparate intrinsic cardiac neuronal populations.
The inherent stochastic nature of cardiac chemosensory inputs minimizes the potential for short-term coordination between disparate atrial and ventricular populations during myocardial ischemia or during the subsequent reperfusion. Yet, because of the underlying neuronal substrate of local circuit-dependent interganglionic interactions, separate elements of the intrinsic cardiac nervous system do have the capacity to exert longer term coordination of disparate efferent outflows.
Coordination among and between different populations of intrinsic cardiac neurons amplifies control over select cardiac indexes (8, 9, 12, 14). For example, control of SA nodal function resides primarily with neurons in the right atrial and posterior atrial ganglionated plexuses (13, 24, 28). Gray et al. (15) have recently demonstrated the anatomical substrate for interganglionic coordination between these two ganglionated plexuses. The present study enlarges upon such a concept. Specifically, administering nicotine to one population (right atrial neurons) activated not only that population but another (ventricular) as well (Table 1). As no vascular connections exist between these populations, these data support the concept of functional interconnectivity existing among atrial and ventricular neurons. Future studies should be directed at identifying the anatomical substrate for such atrial/ventricular neural interactions.
The common, shared inputs from higher centers can also impose coordination of activities among spatially discrete populations of intrinsic cardiac neurons. In fact, both independent and shared preganglionic inputs to separate intrinsic cardiac ganglionic plexuses (right atrial and posterior atrial ones) originate from neurons in the ventral lateral nucleus ambiguous (16). In agreement with that concept, in the present study, the activity generated by both right atrial and cranial medial ventricular neurons transiently increased and then adapted at lower activity levels following their acute decentralization. Moreover, acute decentralization also obtunded the responsiveness of intrinsic cardiac neurons to cardiac milieu alterations (Table 2).
Limitations. There are multiple factors that may impact on the data generated in the present study. 1) Anesthesia and the surgery necessitated to record the activity generated by these neurons may impact on their basal activity, as well as reflexes generated within the cardiac nervous system. 2) Coordination among peripheral neuronal populations may be influenced by the neuronal subtype identified. For this study, the predominant neuronal subtype identified was likely local circuit neurons. 3) Coordination among neuron population may be influenced by multiple functional factors. For example, neurons involved in SA nodal function are tightly linked (15, 27, 28), whereas those involved in control of global electrical or mechanical function may not be, as indicated in this study. Finally, conclusions based on acute decentralization of the intrinsic cardiac nervous system may underestimate the potential for interganglionic interconnections due to the profound suppressor effects that acute decentralization exerts on autonomic ganglia.
Perspectives. In the context of previous studies, these data imply the existence of 1) overlapping cardiac sensory inputs to atrial and ventricular neurons, and 2) functional connectivity among such populations that subserve longer term interactions. Moreover, these data indicate that cardiac phase-related activity generated by sympathetic and parasympathetic preganglionic inputs to the intrinsic cardiac nervous system is not reflected in similar activity profiles of its local circuit neurons. That most atrial and ventricular neurons display no short-term interactive behavior appears to be predicated upon their stochastic chemosensory inputs. As a consequence, they display relatively independent beat-to-beat reflex control of regional cardiac indexes, responding rapidly and discretely to localized cardiac stress. Their functional interdependence, manifest by interganglionic interconnections and varied descending inputs, presumably acts to maintain overall balance of efferent neuronal control over longer time scales. Such an anatomical function arrangement may provide a substrate that compensates for altered function of select populations. Much more research is required to elucidate the varied linkages within the intrinsic cardiac nervous system to determine how its redundancy affects cardiac electrical and mechanical control in the presence of emergent pathology.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
-hydroxylase in tissue culture. Brain Res 422: 74–82, 1987.[CrossRef][Web of Science][Medline]-adrenergic agonists. Am J Physiol Regul Integr Comp Physiol 265: R659–R669, 1993.1-Adrenergic receptor blockade attenuates angiotensin II-mediated catecholamine release into the cardiac interstitium in mitral regurgitation. Circ Res 108: 225–230, 2003.This article has been cited by other articles:
|
|
 |
|
  L.-P. Richer, A. Vinet, T. Kus, R. Cardinal, J. L. Ardell, and J. A. Armour {alpha}-Adrenoceptor blockade modifies neurally induced atrial arrhythmias Am J Physiol Regulatory Integrative Comp Physiol, October 1, 2008; 295(4): R1175 - R1180. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Armour Potential clinical relevance of the 'little brain' on the mammalian heart Exp Physiol, February 1, 2008; 93(2): 165 - 176. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
