Vous êtes sur la page 1sur 8

J Appl Physiol 96: 22652272, 2004. First published February 20, 2004; 10.1152/japplphysiol.00620.2003.

Parasympathetic control of the heart. I. An interventriculo-septal ganglion is the major source of the vagal intracardiac innervation of the ventricles
Tannis A. Johnson,1 Alrich L. Gray,1 Jean-Marie Lauenstein,1 Stephen S. Newton,1 and V. John Massari1,2
Department of Pharmacology and 2Specialized Neuroscience Research Program, Howard University College of Medicine, Washington, District of Columbia 20059
Submitted 16 June 2003; accepted in nal form 25 September 2003
1

Johnson, Tannis A., Alrich L. Gray, Jean-Marie Lauenstein, Stephen S. Newton, and V. John Massari. Parasympathetic control of the heart. I. An interventriculo-septal ganglion is the major source of the vagal intracardiac innervation of the ventricles. J Appl Physiol 96: 22652272, 2004. First published February 20, 2004; 10.1152/ japplphysiol.00620.2003.The locations, projections, and functions of the intracardiac ganglia are incompletely understood. Immunocytochemical labeling with the general neuronal marker protein gene product 9.5 (PGP 9.5) was used to determine the distribution of intracardiac neurons throughout the cat atria and ventricles. Fluorescence microscopy was used to determine the number of neurons within these ganglia. There are eight regions of the cat heart that contain intracardiac ganglia. The numbers of neurons found within these intracardiac ganglia vary dramatically. The total number of neurons found in the heart (6,274 1,061) is almost evenly divided between the atria and the ventricles. The largest ganglion is found in the interventricular septum (IVS). Retrogradely labeled uorescent tracer studies indicated that the vagal intracardiac innervation of the anterior surface of the right ventricle originates predominantly in the IVS ganglion. A cranioventricular (CV) ganglion was retrogradely labeled from the anterior surface of the left ventricle but not from the anterior surface of the right ventricle. These new neuroanatomic data support the prior physiological hypothesis that the CV ganglion in the cat exerts a negative inotropic effect on the left ventricle. A total of three separate intracardiac ganglia innervate the left ventricle, i.e., the CV, IVS, and a second left ventricular (LV2) ganglion. However, the IVS ganglion provides the major source of innervation to both the left and right ventricles. This dual innervation pattern may help to coordinate or segregate vagal effects on left and right ventricular performance. autonomic control; cardioinhibitory; intracardiac ganglia; parasympathetic nervous system; vagus nerve

cardiovascular research, we have reexamined the distribution and counted the number of neurons contained within intracardiac ganglia throughout the cat heart. Activation of the vagus nerve can cause substantial effects on ventricular functions. Thus, for example, increased vagal tone can 1) decrease ventricular contractility (24), 2) prolong ventricular refractoriness (38), and 3) terminate some types of ventricular arrhythmias in humans (15). The origins of the vagal postganglionic neurons that innervate the ventricles, however, are poorly understood. Previous physiological data from this laboratory have suggested that a left ventricular cranioventricular (CV) ganglion selectively mediated negative inotropic effects of vagal stimulation on the left ventricle without inuencing negative chronotropic or negative dromotropic actions (17, 20). These data imply that the axons of the CV ganglion should selectively project to the walls of the left ventricle. Furthermore, other ventricular ganglia should exist that innervate the walls of the right ventricle. We have tested these hypotheses by using a uorescent retrograde tracing method to determine the source(s) of the vagal innervation of the right and left ventricular walls.
MATERIALS AND METHODS

ALTHOUGH THERE HAS BEEN

substantial interest in the neuroanatomic innervation of the heart for a considerable period of time (25), the locations, projections, and functions of the intracardiac ganglia are incompletely understood. The number of ganglia that have been characterized in the heart varies across species and ranges from as low as 4 in the rat (9, 27) to as many as 10 in humans (5, 32). Although many intracardiac ganglia are found in the atria, ganglia can also be found on the ventricular surface of the heart in diverse species, including humans (5, 28); dogs (4); cows, sheep, deer, and porpoises (14); rabbits and monkeys (25); and pigs (6). In the cat, it was suggested that intracardiac ganglia are found exclusively within the atria (10); however, in a recent study (20), it has been shown that at least one ventricular ganglion is found in the cat heart. Because the cat is an important animal model in

All experiments were reviewed and approved by the Institutional Animal Care and Use Committee of Howard University in accordance with guidelines established by the National Institutes of Health for the humane treatment of animals. Four adult cats (24 kg) were used to determine the topography of intracardiac ganglia. The animals were deeply anesthetized with pentobarbital sodium (55 mg/kg iv) and perfused sequentially via a cannula inserted retrogradely in the abdominal aorta with 1 liter of 0.1 M PBS (pH 7.4) containing heparin sulfate (2,500 U) and 3 liters of 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). There were no readily apparent differences in the gross anatomy of the cat hearts that were examined. The hearts were removed, postxed in the same xative for 1 h, and then transferred to 30% sucrose in phosphate buffer for 48 h. The hearts were subsequently frozen by immersion in isopentane chilled to a slurry over liquid nitrogen, embedded in optimum-cutting temperature (OCT) compound (Pella, Redding, CA), and cut in a cryostat into 50- m sections. One series of transverse sections was mounted onto gelatinized slides and stained with hematoxylin and eosin. A second series of sections from each heart was processed for the immunocytochemical visualization of the general neuronal marker protein gene product 9.5 (PGP 9.5) using a Vectastain Elite ABC kit and rabbit anti-PGP 9.5 antibody (Biogenesis) diluted 1:2,000. A third series was unstained. Tissues were dehydrated with alcohols, cleared in xylene, coverslipped with Permount, and examined in a Nikon FXA photomicroscope equipped with brightThe 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. 2265

Address for reprint requests and other correspondence: V. J. Massari, Dept. of Pharmacology, Howard Univ. College of Medicine, 520 W St., N.W., Washington, DC 20059 (E-mail: vmassari@howard.edu). http://www.jap.org

8750-7587/04 $5.00 Copyright 2004 the American Physiological Society

2266

PARASYMPATHETIC CONTROL OF THE LEFT AND RIGHT VENTRICLES

eld, dark-eld, and Nomarski differential interference contrast optics. The locations of intracardiac ganglia were plotted onto drawings of transverse sections taken through eight representative levels of the heart (Fig. 1). Experiments to determine the origin(s) of ganglia innervating the ventricles were performed on nine cats of either sex. Cats were sedated with a mixture of ketamine (22 mg/kg im) and acepromazine (0.22 mg/kg im) and given atropine (0.05 mg/kg im) to reduce respiratory secretions. Anesthesia was induced and maintained with isourane by inhalation. Subsequently, the animal was intubated with a cuffed endotracheal tube, and a thoracotomy was performed. Positive end-expiratory pressure articial respiration was maintained thereafter. Body temperature was maintained at 37C with warming pads. The pericardium was incised and carefully retracted. In each of eight animals, a total of 10 l of the retrograde tracer diamidino yellow (1.0% in ethylene glycol containing 2% DMSO) was injected into four sites on the anterior surface of either the left or right ventricle using a Hamilton syringe (Fig. 2). In a control animal, 10 l of diamidino yellow was injected into the pericardial space but not into the myocardium. The pericardium was then closed, and the muscles and skin were sutured in layers. Before the end of surgery, all animals received torbugesic (0.2 mg/kg im) for pain. Postoperatively, all cats received the analgesic drug ketoprofen (1 mg/kg sc) and the antibi-

otic procaine penicillin G (30,000 U/kg sc) daily. Animals were killed under deep barbiturate anesthesia (pentobarbital sodium, 55 mg/kg) by intravascular perfusion with xatives after a 3-day survival time as described above. Every third 50- m transverse frozen section of the heart taken from the aortic arch to within 3 mm of the apex of the ventricle was mounted on gelatinized slides, coverslipped with 50% aqueous glycerol, and examined using FITC uorescence optics. Only neurons with an identiable nucleus ( 200 magnication) were counted. Cell counts were multiplied by three to compensate for examination of every third section. Retrogradely labeled cells were identied by the presence of a yellow nucleus and a clear cytoplasm. For each ganglion, we determined the total number of neurons and the number of retrogradely labeled neurons. Statistical analyses of the data were performed using a one-way ANOVA. Homogeneity of variance was evaluated using the Levene test (23), and multiple comparisons were evaluated using Tukeys multiplerange test when group variances were equal or Dunnetts T3 post hoc test when group variances were unequal (22, 33).
RESULTS

Microscopic analysis of serial sections through the heart immunocytochemically stained with an antibody to the general

Fig. 1. Schematic illustrations of the locations (AH) of intracardiac ganglia in the cat. L1L8, 8 representative levels of the heart. A, posterior atrial ganglion; B, sinoatrial ganglion; C, atrioventricular ganglion; D, cranioventricular ganglion; E, right ventricular ganglion; F, interatrio-septal ganglion; G, interventriculo-septal ganglion; H, left ventricular ganglion 2; I, sporadic neurons were sometimes observed here; aa, aortic arch; gcv, great coronary vein; ivc, inferior vena cava; la, left atrium; lau, left auricle; lca, left anterior descending coronary artery; lv, left ventricle; paa, pulmonary arteries; ra, right atrium; rau, right auricle; rv, right ventricle; svc, superior vena cava. *Locations of neurons within the intracardiac ganglia. J Appl Physiol VOL
96 JUNE 2004

www.jap.org

PARASYMPATHETIC CONTROL OF THE LEFT AND RIGHT VENTRICLES

2267

Fig. 2. Schematic illustrations of the locations (*) of retrograde tracer injections into the right ventricle (A) or left ventricle (B) of the cat heart. Abbreviations as in legend to Fig. 1.

neuronal marker PGP 9.5 (36) revealed eight separate regions that consistently contained multiple closely associated clusters of neurons that we have referred to as ganglia (Figs. 1 and 3). Two ganglia were found on the right atrium [the posterior atrial (PA) and sinoatrial (SA) ganglia], one ganglion was found on the left atrium [the atrioventricular (AV) ganglion], one ganglion was found within the interatrial septum [the interatrioseptal (IAS) ganglion], one ganglion was found on the right ventricle [the right ventricular (RV) ganglion], two ganglia were found on the left ventricle (the CV ganglion and the LV2 ganglion), and one ganglion was found within the interventricular septum [the interventriculo-septal (IVS) ganglion]. Most of the ganglia were found within fatty and connective tissues between the surface of the heart and cardiac muscle, or surrounding the great vessels, and were in close proximity to nerve bundles. In some cases, however, particularly within the IVS ganglion, substantial numbers of neurons were found embedded in areas of myocardium that were completely surrounded by cardiac muscle cells and had little or no surrounding fatty matrix.
J Appl Physiol VOL

The PA ganglion is found in a fat pad located between the superior vena cava and the aorta on the posterior medial aspect of the right atrium. It is one of the largest ganglia on the heart. The PA ganglion is also the most rostrally located intrinsic cardiac ganglion. It begins just superior to the common pulmonary artery and ends at the superior aspect of the left atrium. In cross section, the PA ganglion is found ventral to the pulmonary artery and dorsal to the aorta and right atrium. The PA ganglion is comprised of large teardrop-shaped or small linear clusters of neurons widely separated and completely surrounded by fatty connective tissue. Clusters may be composed of as many as 60 neurons. The SA and AV ganglia are located within fat pads on the surfaces of the atria. The SA ganglion is found on the right atrium located at the junction of the right pulmonary veins and the superior vena cava. The cells in this ganglion lie adjacent to the atrial myocardium and are oriented in a columnar or teardrop-shape fashion along the external border of the atrial myocardium. Occasionally, small ganglia can be found within the atrial myocardium. The outer border of the ganglion is
www.jap.org

96 JUNE 2004

2268

PARASYMPATHETIC CONTROL OF THE LEFT AND RIGHT VENTRICLES

Fig. 3. Protein gene product 9.5 (PGP 9.5) immunocytochemical staining of intracardiac ganglia. AH are photomicrographs that correspond, respectively, to the schematic drawing of ganglia labeled A through H in Fig. 1. Arrows point to examples of ganglion cells. f, Fat cells; g, ganglion; lam, left atrial myocardium; n, intracardiac nerve; ram, right atrial myocardium. Nomarski optics. Final magnication: 220 in AF; 110 in G and H.

comprised of fatty tissue (21). The AV ganglion is located at the junction of the inferior vena cava and the inferior left atrium. The cells in this ganglion are found in several oval clusters dispersed in connective and fatty tissue (21). The CV ganglion is found in a fat pad located at the cranioventricular margin of the left ventricle. The rostral aspect of the ganglion originates at the ventral left lateral aspect of the conus trunci. Caudally, this ganglion begins to migrate dorsally toward the dorsal left lateral aspect of the right ventricle. The CV ganglion is found in a large mass of fatty tissue containing scattered clusters of ganglia comprised of a few cells (18, 20). The RV ganglion is a relatively small ventricular ganglion located on the rostral right lateral aspect of the conus trunci and is found in small groups of four to ve cell clusters embedded in fatty connective tissue. The IVS ganglion is the largest ventricular ganglion. It is found deeply embedded in the ventricular myocardium at the posterior aspect of the interventricular septum. This ganglion is composed of large round clusters of cells surrounded by fatty connective tissue. Individual clusters typically contain 4060 cells. This ganglion begins at the rostral aspect of the ventricular myocardium and ends just below the aortic outlet. The LV2 ganglion is a smaller ventricular ganglion found on the epicardial surface of the left
J Appl Physiol VOL

ventricle in small elongated and somewhat scattered clusters embedded in fat tissue and often in close association with epicardial blood vessels. Although PGP 9.5 immunoreactivity permitted the initial mapping of the locations of intracardiac ganglia, actual counts of both labeled and unlabeled neurons within these ganglia (Fig. 4) were made at 200 magnication using uorescence microscopy with a FITC excitation and barrier lter set (Nikon) because this optical system permitted the ready differentiation of neuronal cytoplasm, nuclei, and axons from other tissue elements within the heart. Nuclear labeling of retrogradely labeled neurons with diamidino yellow was also readily observed with this optical system but was conrmed by using a UV optical lter that visualizes diamidino yellow (Nikon). By contrast, it was often difcult to estimate cell counts with PGP 9.5 because the nuclei of PGP 9.5immunoreactive neurons as well as the cytoplasm are both strongly labeled. Therefore, the presence or absence of nuclear boundaries within a cell could not be used to help differentiate neurons from fragments. Other cardiac landmarks were also poorly distinguished in these tissues as well. While examining adjacent control sections, however, we observed that by using FITC optics we could much more clearly distinguish and count neurons within intracardiac ganglia on the basis of their morwww.jap.org

96 JUNE 2004

PARASYMPATHETIC CONTROL OF THE LEFT AND RIGHT VENTRICLES

2269

Fig. 4. Nos. of neurons (means SE) found within individual intracardiac ganglia of the cat. PA, posterior atrial ganglion; SA, sinoatrial ganglion; AV, atrioventricular ganglion; IAS, interatrio-septal ganglion; RV, right ventricular ganglion; CV, cranioventricular ganglion; IVS, interventriculo-septal ganglion; LV2, left ventricular ganglion 2.

phology and autouorescence rather than their content of PGP 9.5. Autouorescence was not a problem because it was readily distinguished from diamidino yellow by rapidly rotating different lter sets into the optical path. Furthermore, when on the rare occasion in these relatively thick sections that the same neuron could tentatively be identied in two adjacent sections, neurons that were identied morphologically with FITC optics were PGP 9.5 immunoreactive. The FITC optical system also permitted the easy visualization and identication of fat cells, epithelial cells, cardiac muscle, blood vessels, valves, and other cardiac landmarks that further simplied the not inconsequential problem of relating the microscopic anatomy to the gross anatomy of the heart. Finally, the regional pattern of distribution of intracardiac ganglia that was observed using PGP 9.5 immunoreactivity was identical to that seen using FITC optics or in tissues stained with hematoxylin and eosin. In 3 of a total of 14 animals, a few sporadic clusters of intracardiac ganglia were seen in epicardial fat within another intraventricular site (Fig. 1, I). Because these neurons were so infrequently noted, however, they are not discussed further, although they may represent some standard variation in the neuroanatomy of the cat heart. The cat heart was observed to contain a mean of 6,274 1,061 neurons. From this total, 51% (3,210 587) were found within the ventricles. The number of neurons observed within individual intracardiac ganglia varied signicantly [F(7,39) 6.50, P 0.001]. The two largest ganglia in the heart were the IVS and PA ganglia (Fig. 4). Injections of diamidino yellow into the myocardium resulted in elongated but erratically shaped areas of intensely uorescent labeling 35 mm in length, which traversed the area from the epicardial surface to well within the cardiac muscles but did not extend into the ventricular lumens. Subsequent to multiple microinjections of diamidino yellow into the anterior apical surface of the right ventricle (Fig. 2), there was a statistically signicant difference in the percentage of the total number of neurons that were retrogradely labeled within individual intracardiac ganglia [F(6,23) 15.77, P 0.001; Fig. 5]. In fact, more than two-thirds of the labeled neurons in the heart were found in the IVS ganglion (Fig. 5). Percent labeling in the IVS ganglion was statistically larger than that of any other ganglion in the heart (mean 66.5% 6.7; P 0.05).
J Appl Physiol VOL

Percent labeling in all other ganglia was not signicantly different from one another (P 0.16) and averaged 6.51 2.54%. Note particularly that very few retrogradely labeled neurons were found in the CV ganglion (Figs. 5 and 6). Subsequent to multiple microinjections of diamidino yellow into the anterior apical surface of the left ventricle (Fig. 2), there was a statistically signicant difference in the percentage of the total number of neurons that were retrogradely labeled within individual intracardiac ganglia [F(7,22) 16.76, P 0.001; Fig. 7]. In this case, more than three-fourths of the retrogradely labeled neurons found in the heart were located in three ventricular ganglia, i.e., the CV, IVS, and LV2 ganglia. Percent labeling in the IVS ganglion was signicantly larger than that observed in any other intracardiac ganglion (mean 42.8 7.6%; P 0.05). There was no signicant difference in retrograde labeling observed between the CV and LV2 ganglia (mean 18.4 3.2%; P 0.05); however, labeling in the CV ganglion was signicantly greater than that observed in the SA, AV, IAS, or RV ganglia (P 0.05). Percent labeling in all ganglia other than the CV, IVS, and LV2 ganglia averaged 4.8 0.93%. When diamidino yellow was injected into the pericardium in the control animal, 49 0.54% of all neurons within each intracardiac ganglion were labeled.
DISCUSSION

Considerable uncertainty exists concerning the anatomic location and precise physiological functions of the intracardiac ganglia of multiple species, including the cat. Because the sympathetic input to the heart largely arises from neurons located in the intrathoracic ganglia (3), intracardiac ganglia are believed to be predominantly parasympathetic, although a few neurons containing catecholamines have been found on atrial ganglia (19). Some evidence also indicates that intracardiac ganglia may contain sensory neurons and local circuit neurons (2). Although it has been long known that vagal efferents to the heart richly innervated parasympathetic ganglia located on the atria, it was generally believed that vagal parasympathetic control either did not extend to the ventricles or had minimal physiological relevance (29). In the cat, Calaresu and St. Louis (10) reported that intracardiac ganglia were found surrounding

Fig. 5. Percentage of total number of labeled neurons in individual intracardiac ganglia subsequent to injections of a retrograde tracer into the right ventricle. Abbreviations as in legend to Fig. 4. www.jap.org

96 JUNE 2004

2270

PARASYMPATHETIC CONTROL OF THE LEFT AND RIGHT VENTRICLES

Fig. 6. Injections of a retrograde tracer into the right ventricle do not cause labeling in the CV ganglion (A) but result in substantial labeling in the IVS ganglion (B). Arrows in B indicate examples of ganglion cells with labeled nuclei. Note the absence of labeled nuclei within neurons in the CV ganglion in A. Fluorescence microscopy. Original magnication, 220.

the great vessels and on the atria, or within the interatrial septum. However, ganglion cells were never seen in the kitten ventricles (Ref. 10; p. 61). This conclusion was based, however, on an examination of the ventricles in only one kitten. In analogy with the data in the cat, early data suggested that the rat and dog ventricles also contained few or no intracardiac ganglia (8, 11). Recent evidence, however, indicates that at least one intraventricular ganglion is present in both the cat (20) and the rat heart (34, 35), and multiple intraventricular ganglia are present in the dog heart (39). In the present report, therefore, we have examined sections throughout the atria and ventricles of the cat heart for the presence of intracardiac ganglia. Our data indicate that there are eight regions of the cat heart that consistently contain ganglia. Within the atria, we have previously dened the location of two intracardiac gan-

Fig. 7. Percentage of total number of labeled neurons in individual intracardiac ganglia subsequent to injections of a retrograde tracer into the left ventricle. Abbreviations as in legend to Fig. 4. J Appl Physiol VOL

glia, i.e., the SA and AV ganglia (21). We have now determined the distribution of two additional atrial ganglia, i.e., the PA ganglion and the IAS ganglion. Furthermore, we have detected a total of four intraventricular ganglia, one on the right ventricle, two on the left ventricle, and one within the interventricular septum. Although there is some homology between species in the distribution of intracardiac ganglia, there are also substantial interspecies differences in the number of atrial or ventricular ganglia, their distribution, and in estimates of the total number of neurons found in the heart (5, 6, 10, 27). The distribution of parasympathetic postganglionic nerves to many regions of the heart, including both ventricles, has largely been inferred from physiological studies (1, 8, 12, 29), not neuroanatomic studies. We have previously determined the intracardiac location and physiological effects of the CV ganglion on cardiac functions. It was reported that microinjections of a ganglionic blocking drug into the CV ganglion attenuated the negative inotropic effects of vagal stimulation on the left ventricle, without inuencing either vagally mediated negative chronotropic or negative dromotropic effects (18, 20). These physiological data predict that, at a minimum, neurons in the CV ganglion should project to the left ventricle and not to other regions of the heart. In the present study, a retrograde tracer was placed into the anterior apical surface of the left ventricle because some data indicate that the autonomic innervation of the ventricles is not homogeneous. Thus, for example, the sympathetic postganglionic innervation of the anterior apical surface of the left ventricle arises from the left stellate ganglion, while the corresponding innervation of the posterobasal wall arises from the right stellate ganglion (30). We observed that the CV ganglion was robustly labeled subsequent to microinjections of tracer into the left ventricle but was not signicantly labeled subsequent to tracer injections into the right ventricle. These neuroanatomic data are therefore consistent with our previous physiological hypothesis that the CV
www.jap.org

96 JUNE 2004

PARASYMPATHETIC CONTROL OF THE LEFT AND RIGHT VENTRICLES

2271

ganglion selectively mediates a negative inotropic effect on the left ventricle because the CV ganglion does not effect chronotropic or dromotropic cardiac functions (18, 20) and does not project to the right ventricle to potentially inuence the physiological function of this chamber. The present data further indicate, however, that although the physiological effects mediated by the CV ganglion are selective, it is not the exclusive source of vagal efferents to the left ventricle. Rather, two other intraventricular ganglia also signicantly mediate vagal control of left ventricular functions. These include the IVS and LV2 ganglia. Percent labeling in the IVS ganglion was signicantly larger than that found in any other intracardiac ganglion, and more than three-fourths of the retrogradely labeled neurons found in the heart were located in these three ventricular ganglia. By contrast, the percent labeling in each of the other ganglia averaged 4%. Further studies will be required to determine the precise physiological roles of the IVS and LV2 ganglia in the control of left ventricular functions and the potential extent of their inuence on left ventricular contractility. To control for potential labeling due to leakage of tracer from the injection sites, tracer was injected into the pericardial space. This resulted in a uniform 49 0.54% degree of neuronal labeling within each intracardiac ganglion. By contrast, injections of tracers into either the left or right ventricular wall resulted in a highly statistically signicant difference (P 0.001) in percent labeling across the different intracardiac ganglia. We conclude from these data that leakage of the tracer from its injection sites was minimal. We have also considered the possibility that the degree of retrograde ganglionic labeling seen with diamidino yellow might potentially be proportional to the distance between labeled ganglia and the sites of injection of this tracer. In this way, relatively close ganglia would be labeled, whereas distant ganglia would not. Several lines of evidence indicate that failure to nd retrogradely labeled neurons within various intracardiac ganglia subsequent to injections of the tracer into either ventricle was not due to relative distance from the injection site or inadequate survival time. Thus, for example, the present data indicate that injections of tracer into the right ventricle failed to label neurons in a closely adjacent ganglion actually contained within the right ventricle (the RV ganglion) while signicantly labeling neurons approximately equidistant from the injection sites in the interventricular septum (the IVS ganglion). Furthermore, the 4-day survival time that was used in the present report has also been demonstrated to provide ample staining of cardioinhibitory vagal preganglionic neurons in the nucleus ambiguus of the brain stem of the cat after microinjections into the SA or AV ganglia (7), a distance that is substantially greater than that separating all intracardiac ganglia from injection sites in either the left or right ventricle. The present data indicate that the vagal postganglionic control of right ventricular function is mediated primarily from the IVS ganglion, which contained more than two-thirds of the retrogradely labeled neurons found in the heart. This was statistically signicantly greater than that of any other intracardiac ganglion. By comparison, average percent labeling in all other ganglia was not signicantly different from one another and averaged 6% of the total labeling in the heart. The IVS ganglion, therefore, is the largest source of neurons innervating the anterior surfaces of both the right ventricle and
J Appl Physiol VOL

the left ventricle. It is interesting to speculate that this dualinnervation pattern may help to coordinate or segregate vagal effects on ventricular performance. It would be interesting indeed to inject two separate retrograde tracers into the left and right ventricles to see whether the same neurons in the IVS ganglion project to both targets or whether separate populations of neurons within this ganglion project to individual ventricles. Greater understanding of the pathways and mechanisms whereby the vagus nerve exerts its effects on the ventricles may be helpful in understanding autonomic responses during acute myocardial ischemia. Autonomic responses during acute myocardial ischemia play a major role in predicting the probability of a fatal outcome. If activation of the sympathetic nervous system occurs, malignant ventricular brillation and sudden death are likely (31). If, on the other hand, acute myocardial ischemia is followed by activation of cardiac vagal efferents, a protective antibrillatory effect is seen and the probability of survival is enhanced (16, 37). These results may be due to the fact that enhanced vagal tone increases ventricular brillation threshold and reduces the arrhythmogenic effects of sympathetic activation (37). Human studies indicate that vagal activity is reduced for several months after myocardial infarction, whereas sympathetic tone is enhanced (32, 34, 48). It is believed that this altered autonomic balance may in part mediate the increased incidence of sudden cardiac death after myocardial infarction (26). The autonomic nervous system also has a profound effect on the pathogenesis of numerous arrhythmias, including supraventricular tachycardias, atrioventricular block, and ventricular arrhythmias associated with myocardial ischemia and sudden death (13, 26). These data emphasize the clinical importance of understanding how the parasympathetic nervous system regulates ventricular functions. This knowledge is essential to the future rational design of experimental interventions in heart disease.
GRANTS This research was supported in part by grants from the National Heart, Lung, and Blood Institute (RO1-HL-51917) and American Heart Association to V. J. Massari. Additional funding was provided by the Gustavus and Louise Pfeiffer Research Foundation to A. L. Gray and the Specialized Neuroscience Research Program (1U54-NS-39407, M. A. Haxhiu, Principal Investigator). REFERENCES 1. Ardell JL. Structure and function of mammalian intrinsic cardiac neurons. In: Neurocardiology, edited by Armour JA and Ardell JL. New York: Oxford Univ. Press, 1994, p. 95114. 2. Armour JA. Myocardial ischaemia and the cardiac nervous system. Cardiovasc Res 41: 4154, 1999. 3. Armour JA and Hopkins DA. Anatomy of extrinsic efferent autonomic nerves and ganglia innervating the mammalian heart. In: Nervous Control of Cardiovascular Function, edited by Randall WC. New York: Oxford Univ. Press, 1984, p. 2045. 4. Armour JA and Hopkins DA. Activity of in vivo canine ventricular neurons. Am J Physiol Heart Circ Physiol 258: H326H336, 1990. 5. Armour JA, Murphy DA, Yuan BX, Macdonald S, and Hopkins DA. Gross and microscopic anatomy of the human intrinsic cardiac nervous system. Anat Rec 247: 289298, 1997. 6. Arora RC, Waldmann M, Hopkins DA, and Armour JA. Porcine intrinsic cardiac ganglia. Anat Rec 271A: 249258, 2003. 7. Blinder KJ, Gatti PJ, Johnson TA, Lauenstein JM, Coleman WP, Gray AL, and Massari VJ. Ultrastructural circuitry of cardiorespiratory reexes: there is a monosynaptic path between the nucleus of the solitary tract and vagal preganglionic motoneurons controlling atrioventricular conduction in the cat. Brain Res 785: 143157, 1998. www.jap.org

96 JUNE 2004

2272

PARASYMPATHETIC CONTROL OF THE LEFT AND RIGHT VENTRICLES 24. Levy MN and Pappano AJ. Vagal control of myocardial contractility. In: Vagal Control of the Heart: Experimental Basis and Clinical Implications, edited by Levy MN and Schwartz PJ. Armonk, NY: Futura, 1994, p. 241259. 25. Mitchell GAG, Brown R, and Cookson FB. Ventricular nerve cells in mammals. Nature 172: 812, 1953. 26. Mitrani RD and Zipes DP. Clinical neurocardiology: arrhythmias. In: Neurocardiology, edited by Armour JA and Ardell JL. New York: Oxford Univ. Press, 1994, p. 365396. 27. Pardini BJ, Patel KP, Schmid PG, and Lund DD. Location, distribution and projections of intracardiac ganglion cells in the rat. J Auton Nerv Syst 20: 91101, 1987. 28. Pauza DH, Skripka V, Pauziene N, and Stropus R. Morphology, distribution, and variability of the epicardiac neural ganglionated subplexuses in the human heart. Anat Rec 259: 353382, 2000. 29. Randall WC, Randall DC, and Ardell JL. Autonomic regulation of myocardial contractility. In: Reex Control of the Circulation, edited by Zucker IH and Gilmore JP. Boston, MA: CRC, 1991, p. 3965. 30. Schlack W and Thamer V. Unilateral changes of sympathetic tone to the heart impair left ventricular function. Acta Anaesthesiol Scand 40: 262 271, 1996. 31. Schwartz PJ. Manipulation of the autonomic nervous system in the prevention of sudden cardiac death. In: Cardiac Arrhythmias: Where To Go From Here? edited by Brugada P and Wellens HJJ. Mt. Kisko, NY: Futura, 1987, p. 741765. 32. Singh S, Johnson PI, Lee RE, Orfei E, Lonchyna VA, Sullivan HJ, Montoya A, Tran H, Wehrmacher WH, and Wurster RD. Topography of cardiac ganglia in the adult human heart. J Thorac Cardiovasc Surg 112: 943953, 1996. 33. SPSS. SPSS Base 10.0 Applications Guide. Chicago, IL: SPSS, 1999. 34. Standish A, Enquist LW, and Schwaber JS. Innervation of the heart and its central medullary origin dened by viral tracing. Science 263: 232234, 1994. 35. Ter Horst GJ, Van den Brink A, Homminga SA, Hautvast RW, Rakhorst G, Mettenleiter TC, De Jongste MJ, Lie KI, and Korf J. Transneuronal viral labelling of rat heart left ventricle controlling pathways. Neuroreport 4: 13071310, 1993. 36. Thompson RJ, Doran JF, Jackson P, Dhillon AP, and Rode J. PGP 9.5a new marker for vertebrate neurons and neuroendocrine cells. Brain Res 278: 224228, 1983. 37. Vanoli E, De Ferrari GM, Stramba-Badiale M, Hull SS Jr, Foreman RD, and Schwartz PJ. Vagal stimulation and prevention of sudden death in conscious dogs with a healed myocardial infarction. Circ Res 68: 14711481, 1991. 38. Warner MR and Zipes DP. Vagal control of myocardial refractoriness. In: Vagal Control of the Heart: Experimental Basis and Clinical Implications, edited by Levy MN and Schwartz PJ. Armonk, NY: Futura, 1994, p. 261276. 39. 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: 7587, 1994.

8. Blomquist TM, Priola DV, and Romero AM. Source of intrinsic innervation of canine ventricles: a functional study. Am J Physiol Heart Circ Physiol 252: H638H644, 1987. 9. Burkholder T, Chambers M, Hotmire K, Wurster RD, Moody S, and Randall WC. Gross and microscopic anatomy of the vagal innervation of the rat heart. Anat Rec 232: 444452, 1992. 10. Calaresu FR and St. Louis AJ. Topography and numerical distributions of intracardiac ganglion cells in the cat. J Comp Neurol 131: 5566, 1967. 11. Chen RF and Chiou RJCTY. Distribution and number of intracardiac ganglion cells in the rat. Acta Zool Taiwanica 8: 4754, 1997. 12. Chiou CW and Zipes DP. Selective vagal denervation of the atria eliminates heart rate variability and baroreex sensitivity while preserving ventricular innervation. Circulation 98: 360368, 1998. 13. Corr PB, Yamada KA, and Witkowski FX. Mechanisms controlling cardiac autonomic function and their relation to arrhythmogenesis. In: The Heart and Cardiovascular System, edited by Fozzard HA. New York: Raven, 1986, p. 13431403. 14. Davies F, Francis ETB, and King TS. Neurological studies of the cardiac ventricles of mammals. J Anat 86: 130143, 1952. 15. De Ferrari GM, Vanoli E, and Schwartz PJ. Vagal activity and ventricular brillation. In: Vagal Control of the Heart: Experimental Basis and Clinical Implications, edited by Levy MN and Schwartz PJ. Armonk, NY: Futura, 1994, p. 613636. 16. De Ferrari GM, Vanoli E, Stramba-Badiale M, Hull SS Jr, Foreman RD, and Schwartz PJ. Vagal reexes and survival during acute myocardial ischemia in conscious dogs with healed myocardial infarction. Am J Physiol Heart Circ Physiol 261: H63H69, 1991. 17. Dickerson LW, Honey KW, Fleming TJ, Panico WH, Gatti PJ, Massari VJ, and Gillis RA. Negative inotropic effects produced by L-glutamate activation of regions in the ventral lateral nucleus ambiguus of the dog. Soc Neurosci Abstr 24: 1029, 1998. 18. Dickerson LW, Rodak DJ, Fleming TJ, Gatti PJ, Massari VJ, McKenzie JC, and Gillis RA. Parasympathetic neurons in the cranial medial ventricular fat pad on the dog heart selectively decrease ventricular contractility without effect on sinus rate or AV conduction. J Auton Nerv Syst 70: 129141, 1998. 19. Ehinger B, Falck B, Persson H, and Sporrong B. Adrenergic and cholinesterase-containing neurons of the heart. Histochemie 16: 197205, 1968. 20. Gatti PJ, Johnson TA, McKenzie JC, Lauenstein JM, Gray AL, and Massari VJ. Vagal control of left ventricular contractility is selectively mediated by a cranioventricular intracardiac ganglion in the cat. J Auton Nerv Syst 66: 138144, 1997. 21. Gatti PJ, Johnson TA, Phan P, Jordan IK, Coleman W, and Massari VJ. The physiological and anatomical demonstration of functionally selective parasympathetic ganglia located in discrete fat pads on the feline myocardium. J Auton Nerv Syst 51: 255259, 1995. 22. Keppel G. Design and Analysis: A Researchers Handbook. Englewood Cliffs, NJ: Prentice-Hall, 1973. 23. Levene H. Robust tests for equality of variances. In: Contributions to Probability and Statistics, edited by Oloff CM. Stanford, CA: Stanford Univ. Press, 1960, p. 278292.

J Appl Physiol VOL

96 JUNE 2004

www.jap.org