SEARCH

SEARCH BY CITATION

Keywords:

  • atrial neurons;
  • dendrites;
  • ganglionated plexus;
  • ventricular neurons

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. SUMMARY
  7. Acknowledgements
  8. LITERATURE CITED

The gross, light, and electron microscopic anatomies of the porcine intrinsic cardiac nervous system were investigated in 26 pigs to facilitate functional studies in this model. Gross anatomy: Numerous ganglia and interconnecting nerves (ganglionated plexuses) were found to be concentrated in epicardial fat in five atrial and six ventricular regions. The five atrial ganglionated plexuses identified were (1) the ventral right atrial, (2) the right vena cava–right atrial, (3) the dorsal atrial, (4) the interatrial septal, and (5) the left superior vena cava–left atrial ones. Six ventricular ganglionated plexuses were identified in close proximity to the (1) roots of the aorta and pulmonary artery (craniomedial), extending along the left main coronary artery to the (2) ventral interventricular and (3) circumflex coronary arteries. (4) A ganglionated plexus was identified around the origin of the dorsal interventricular coronary artery, as well as the (5) right main and (6) right marginal coronary arteries. Isolated neurons were identified scattered throughout the cranial interventricular septum. Microscopic anatomy: Approximately 3,000 neuronal somata were estimated to compose this intrinsic cardiac nervous system. Some ganglia contained more than 100 neurons. Neuronal somata had dimensions of roughly 33.1 (short axis) by 46.3 (long axis) μm. Most were multipolar, a small population of unipolar neurons being identified in atrial and ventricular tissues. At the electron microscopic level, asymmetrical axodendritic synapses with small clear, round vesicles were identified, some containing large dense-cored vesicles. In summary, porcine intrinsic cardiac neurons are concentrated in 11 distinct atrial and ventricular ganglionated plexuses. These extensive plexuses, along with fewer scattered neurons, display varied neuronal morphology and synaptology that represent the anatomical substrate for complex information processing within the intrinsic cardiac component of the porcine cardiac neuronal hierarchy. These anatomical data provide a framework for physiological analyses of the porcine intrinsic cardiac nervous system. Anat Rec Part A 271A:249–258, 2003. © 2003 Wiley-Liss, Inc.

Intrinsic cardiac neurons, first identified on the human heart by Scarpa in 1794 (cited by Mitchell et al., 1952), have been described in a variety of mammalian species (Dogiel, 1899; Francillon, 1928; Kuntz, 1934; Davies et al., 1952; King and Coakley, 1958; Robb, 1965; Shvalev and Sosunov, 1985; Gagliardi et al., 1988; Yuan et al., 1994; Armour et al., 1997; Pauza et al., 1997, 2000). Intrinsic cardiac neurons are concentrated in collections of ganglia and interconnecting nerves that form identifiable ganglionated plexuses within epicardial fat (Gagliardi et al., 1988; Yuan et al., 1994; Armour et al., 1997; Pauza et al., 2000; Pauziene et al., 2000). Although the neurons are typically found in epicardial fat at the base of the atria, in the interatrial septum and on the cranial aspects of the ventricles, there is considerable interspecies variation in the gross topographical (anatomical) distributions of the somata of intrinsic cardiac neurons on the mammalian heart.

Intrinsic cardiac ganglia contain morphologically distinct neurons, including unipolar, bipolar, and multipolar types (Francillon, 1928; Kuntz, 1934; Robb, 1965; Yuan et al., 1994; Edwards et al., 1995; Armour et al., 1997; Cheng et al., 1997; Pauza et al., 1997). Such anatomical complexity is consistent with physiological studies demonstrating that intrinsic cardiac neurons are functionally diverse (Smith, 1999) and not simply cholinergic relay stations for preganglionic parasympathetic efferent neurons (Blomquist et al., 1987). Functionally, mammalian intrinsic cardiac ganglia also contain sympathetic efferent postganglionic (Gagliardi et al., 1988; Armour and Hopkins, 1990a, b; Butler et al., 1990) and afferent (Ardell et al., 1991; Cheng et al., 1997) neurons. The presence of adrenergic and cholinergic neurons within mammalian intrinsic cardiac ganglia has been demonstrated anatomically as well (Baluk and Gabella, 1990; Moravec et al., 1990; Steele et al., 1994; Horackova et al., 1999; Singh et al., 1999). In accord with that, it is known that subpopulations of intrinsic cardiac neurons, including porcine ones, receive inputs from parasympathetic and sympathetic efferent preganglionic neurons (Armour and Hopkins, 1990a, b; Smith, 1999). It has also been postulated that local circuit neurons in intrinsic cardiac ganglia interconnect afferent and efferent neurons, thereby forming functional regulatory networks within the heart (Armour, 1991; Edwards et al., 1995; Thompson et al., 2000).

Physiological studies increasingly use the porcine model to study the ontogeny of cardiovascular regulation (Gootman, 1991), functional innervation of the neonatal heart (Hopkins et al., 1984, 1997; Crick et al., 1999a, b), or the cardiodynamic effects of clinically relevant therapy (Galoyan et al., 2001). However, the topographical organization of the porcine intrinsic cardiac nervous system has yet to be elucidated in detail. The objectives of the present anatomical study were to determine the distribution, anatomic relationships, and morphology of porcine intrinsic cardiac neurons and to propose a descriptive terminology of their major cardiac locations to facilitate functional studies.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. SUMMARY
  7. Acknowledgements
  8. LITERATURE CITED

All experiments were performed in accordance with the guidelines for animal experimentation described in the “Guide for the Care and Use of Laboratory Animals” (NIH publication 85-23, revised 1996) and the Canadian Council on Animal Care “Guide to the Care and Use of Experimental Animals” (Vol. 1, 2nd ed., 1993).

Experimental Model

Twenty-six pigs (Sus Scrofa) of either gender, each weighing 18–30 kg, were anesthetized with a combination of ketamine (80 mg/kg i.v.) and sodium pentothal (20 mg/kg i.v.) for physiological studies. After the trachea was intubated, positive pressure ventilation was initiated with 0.95 FIO2 and 0.05 FICO2 by using a Bird Mark 7A ventilator (Palm Springs, CA). Anesthesia induction was performed with sodium pentothal (10–15 mg/kg i.v. every 5 min). Thereafter, the anesthesia was changed to α-chloralose (75 mg/kg, i.v. bolus, with repeat doses of 12.5 mg/kg i.v. as required). The adequacy of anesthesia was assessed at regular intervals by applying noxious stimuli to a limb to assess reflex withdrawal responses of the limb and by monitoring jaw tone. At the completion of physiological experiments, hearts and ganglia were collected for anatomical analyses. Of the 26 pigs studied, tissues were obtained from 8 animals for gross anatomical analysis, tissues from 14 pigs for light microscopic analysis, and tissues from 4 pigs for electron microscopic analysis.

Gross Anatomy

The hearts used for gross anatomical studies were removed rapidly from eight anesthetized pigs and rinsed in normal saline to remove blood. The gross anatomy of the intrinsic cardiac nervous system in all of the tissues derived from these eight pig hearts was investigated. To simplify the procedure and to ensure adequate tissue fixation, the heart was divided into sections before being placed in fixative. Specifically, the right and left atria, interatrial septum, the right sided superior and inferior vena cavae, the left superior vena cava (which is present in the pig), and the base of the ventricles were removed en block. In addition, tissues adjacent to the right and left main coronary arteries, adjacent to the origins of the ventral interventricular (VIV) coronary artery in the ventral interventricular groove (analogous to the human left anterior descending artery), the circumflex coronary artery (CCA), the right main coronary artery (RCA), the dorsal interventricular (DIV) coronary artery (analogous to the human posterior descending artery), and the right marginal (RM) and left marginal (LM) coronary arteries were isolated for subsequent analysis. All remaining tissues were also preserved for subsequent gross anatomical analysis. These tissues were rinsed again in room-temperature physiological saline. They were then placed in 1.0 mM phosphate buffered 4% paraformaldehyde at 4°C for at least 1 week.

After that, a 1% solution of methylene blue in phosphate-buffered saline was dripped directly on the fixed tissue in a dissecting dish containing phosphate buffer to facilitate the identification of nerves and ganglia in these tissues. With the aid of a Wild stereomicroscope or a Zeiss dissecting microscope, tissues were gently teased apart to identify ganglia and ganglionated plexuses. Clusters of ganglia and interconnecting nerves were identified in these atrial and ventricular tissues. The number of neurons identified in each ganglionated plexus was then counted. Ganglia varied in size from those containing 2–3 neurons to ganglia with over 100 neurons. In smaller ganglia, visualization of all cells was possible so that total neuronal numbers could be determined with reasonable precision. In contrast, adequate visualization of all the neurons in larger ganglia was not possible. Therefore, larger ganglia were estimated to contain at least 100 neurons, because that number represented the maximum number of somata that could be counted reliably in these ganglia. Small, isolated ganglia or neuronal somata were identified throughout the rest of the heart, particularly in the caudal interventricular septum. As individual neurons located in isolated regions of the ventricles could have been missed in our counting and because we could not count all neurons in the larger ganglia described above, the total number of neurons counted in the eight hearts studied in this manner represents an underestimation of the actual total. Drawings and photographs were made to demonstrate the anatomical arrangements of the ganglionated plexuses with respect to their cardiac regions and the major vessels of the heart.

Light Microscopy

Tissue was also obtained from the hearts of four additional pigs for light microscopic studies. Epicardial fat and underlying muscle were removed from the right atrium or dorsal atrial epicardium, washed in phosphate-buffered saline (PBS) and then immersed in 2.5% glutaraldehyde and 0.5% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) at 4°C. After overnight fixation, these tissues were placed in 0.1 M phosphate buffer. Subsequently, they were embedded in paraffin, sectioned at a thickness of 28 μm, and stained with toluidine. The rapid processing of intrinsic cardiac ganglia derived from the in situ heart in this manner ensured that no detectable alterations in cellular morphology occur due to tissue autolysis (Armour et al., 1997; Yuan et al., 1994).

Immunohistochemistry

Whole hearts from 10 pigs were immersed in 4% paraformaldehyde in 0.1 M phosphate buffer for overnight fixation at 4°C. Intrinsic cardiac ganglia were removed and placed in 0.05 M phosphate buffered saline. Then ganglia were dehydrated through a graded ethanol series (5 min at each concentration) and cleared in xylene for 30 min. Ganglia were then rehydrated through graded ethanols (5 min at each concentration) and incubated in 4% Triton-X 100 (Sigma, Oakville, ON, Canada) in 0.05 M PBS overnight at 4°C and then rinsed in PBS (3 × 10 min). Ganglia were immunolabeled with polyclonal, rabbit protein gene product 9.5 (PGP 9.5) antibody (Chemicon International, Inc., Temecula, CA) using a 1:400 dilution or monoclonal mouse M35 muscarinic acetylcholine receptor antibody (Chemunex, The Ivry-on-Seine Cédex, France) using 1:200 or 1:400 dilutions. Primary antibodies were diluted in 0.5% Triton X-100 in 0.05 M PBS containing 4% normal goat or donkey serum. All primary antibody incubations were for 3 days at 4°C. After incubation, ganglia were rinsed in buffer and immersed in FluoroLink Cy3 (Biological Detection Systems, Inc., Pittsburgh, PA) or fluorescein-conjugated secondary antibody (Jackson Immuno Research Laboratories, Inc., Baltimore, MD), diluted 1: 200 or 1: 400, respectively, in 0.5% Triton X-100 in 0.05 M PBS. Some ganglia were double immunostained, subsequently rinsed, and incubated in the appropriate additional secondary antibody. All secondary antibody incubations were for 18–24 h at 4°C. After rinsing in 0.05 M PBS (3 × 10 min), ganglia were mounted on glass slides with glycerol jelly and covered with no. 0 coverslips (VWR CanLab, Mississauga, ON, Canada). Ganglia were examined and photographed with Zeiss LSM 510 Laser Scanning Microscope. Images were processed by using Adobe Photoshop 5.5 software.

Electron Microscopy

Tissues obtained from four different pigs were prepared for examination by means of electron microscopy. Right atrial and dorsal atrial epicardial fat and underlying cardiac tissues were quickly removed from the beating heart. Individual ganglia (n = 13), dissected free from ganglionated plexuses and fat, were immediately immersed in 1.5% tannic acid in 0.1 M phosphate buffer solution and agitated gently overnight at 4°C on a rotary shaker. Ganglia were then rinsed briefly in 0.1 M phosphate buffer (10-min pre-rinse) and placed in 1% osmium tetroxide for 1 h. After a 10-min rinse in normal saline and three subsequent washes in distilled water, ganglia were stained in 5% aqueous uranyl acetate for 1 h at 4°C. The ganglia were then rinsed three times in distilled water and dehydrated in a graded acetone series. Tissues were infiltrated with TAAB resin 812/502 (Marivac, Halifax, NS, Canada) and polymerized at 60°C for 24 h. To estimate the size of the porcine intrinsic cardiac neurons, camera lucida tracings were made of the profiles of 146 neurons cut through the nucleolus in semi-thin sections (1.0–1.5 μm thick) derived from eight ganglia. Thin sections from eight ganglia were collected on Formvar-coated slotted grids, stained with lead citrate for 2 min, and then examined with a Zeiss EM 10A electron microscope.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. SUMMARY
  7. Acknowledgements
  8. LITERATURE CITED

Gross Anatomy

The gross topographical arrangement of the major clusters of atrial and ventricular ganglia and their interconnecting nerves that comprise the principal components of the porcine intrinsic cardiac nervous system was derived from eight porcine preparations in which the entire heart was examined by means of gross anatomic techniques (Figs. 1, 2). Interconnecting nerves, some of which had relatively large diameters (up to ∼0.5 mm), were identified coursing between the various ganglia within each ganglionated plexus, as well as between different ganglionated plexuses located at the base of the heart.

thumbnail image

Figure 1. Drawing of the ventral aspect of the porcine heart and major vessels, illustrating the locations of the major ventral atrial and ventricular ganglionated plexuses. The cross-hatched areas with gray-shade indicate locations of ganglionated plexuses within the epicardial fat, whereas unshaded areas represent epicardial fat in which neurons were not identified (shading convention is the same for Fig. 2). The diagram in the upper left (with the right atrial appendage retracted) illustrates the location of the right vena cava–right atrial ganglionated plexus (RVC-RA) in the groove between the right vena cavae and ventral right atrial free wall. The right marginal artery ganglionated plexus (RMA GP) surrounds the first marginal artery. Note that the circumflex coronary artery is not depicted because it is located along that vessel under the left atrial appendage depicted in this figure. IVC, inferior vena cava; LA App, left atrial appendage; LV, left ventricle; PA, pulmonary artery; RCA, right coronary artery ganglionated plexus; RSVC, right superior vena cava; RV, right ventricle; VRA, ventral right atrial ganglionated plexus; CA, coronary artery.

Download figure to PowerPoint

thumbnail image

Figure 2. Drawing of a dorsal view of the porcine heart, illustrating the locations of ganglionated plexuses on the dorsal surfaces of the atria and ventricles. The cranial extension of the ventral right atrial (VRA) ganglionated plexus is illustrated. In this view, the right vena cava–right atrial (RVC RA) ganglionated plexus can also be seen along the right-sided vena cavae. The dorsal atrial (Dorsal A) ganglionated plexus lies over the atrial septum. The left superior vena cava–left atrial (LSVC LA) ganglionated plexus is located medial to the origin of the left superior vena cava; the dorsal interventricular (DIV) ganglionated plexus overlies the cranial aspect of the dorsal interventricular groove. Dorsal CA, dorsal interventricular coronary artery. Other abbreviations are the same as in Figure 1.

Download figure to PowerPoint

Intrinsic cardiac ganglia were concentrated in discrete areas of the epicardial fat associated with both atria and the cranial aspects of the ventricles. These ganglia were not distributed ubiquitously throughout epicardial fat (Figs. 1, 2). Some ganglia were located deep in the fat adjacent to or among underlying cardiac muscle fascicles. The somata of isolated neurons were identified embedded in between myocardial muscle fascicles. The latter were typically found in tissues at the base of the ventricles, in the cranial interventricular septum, as well as adjacent to the middle and distal portions of the ventral interventricular coronary artery.

The number and relative size of ganglia identified in major ganglionated plexuses are summarized in Table 1 according to their locations on the porcine heart. Although the precise anatomical organization of identified ganglionated plexuses in a given cardiac region varied among animals, there was an overall consistency in the locations of the plexuses as well as the total numbers of neurons identified in each plexus. The total number of neurons per heart was estimated to be approximately 3,000; relatively equal numbers of neurons were present in atrial and ventricular ganglionated plexuses (Table 1).

Table 1. Number and relative size of ganglia identified
 Neurons per gangliaMean number of ganglia per heartMean number of neurons per heart
1–1011–5051–100>100
  1. VRA, ventral right atrial; RVC-RA, right vena cava–right atrial; Dorsal A, dorsal atrial; IAS, interatrial septal; LSVC-LA, left superior vena cava–left atrial; CM, craniomedial; VIV, ventral interventricular artery; CCA, circumflex coronary artery; DIV, dorsal interventricular; RCA, right coronary artery; RMA, right marginal artery.

Atrial ganglionated plexuses
 VRA9 ± 316 ± 72 ± 11 ± 0.328648
 RVC-RA6 ± 18 ± 11 ± 0.31 ± 0.116268
 Dorsal A1 ± 11 ± 11 ± 11 ± 1474
 IAS5 ± 36 ± 31 ± 0.61 ± 0.213296
 LSVC-LA10 ± 610 ± 51 ± 1021358
    Subtotal:821,644
Ventricular ganglionated plexuses
 CM5 ± 37 ± 41 ± 0.4013233
 VIV5 ± 17 ± 21 ± 0.51 ± 0.414337
 CCA3 ± 15 ± 21 ± 0.31 ± 0.310195
 DIV4 ± 22 ± 11 ± 0.30795
 RCA2 ± 15 ± 11 ± 0.51 ± 0.49418
 RMA2 ± 11 ± 11 ± 10416
    Subtotal:571,294
    TOTAL1392,938
Atrial ganglionated plexuses.

Approximately 82 ganglia of various sizes per heart, containing an average of over 1,600 neurons, were identified in ganglionated plexuses located in the fatty tissue associated with the atria. Five ganglionated plexuses were identified consistently in specific atrial regions (Table 1). (1) The ventral right atrial ganglionated plexus (VRA GP) was found to be embedded in fat on the ventral and lateral aspects of the right atrium (Fig. 1), cranial to the atrioventricular groove. It extended in a craniomedial direction toward the junction of the right superior vena cava with the right atrium. This ganglionated plexus contained the largest number of neurons associated with the porcine heart (Table 1). (2) The right vena cava–right atrial ganglionated plexus (RVC-RA GP) was identified in fat located on the lateral border of the right atrium, in the groove between the right-sided superior vena cava and right atrium (Fig. 1, right, lateral view). (3) The dorsal atrial ganglionated plexus (Dorsal A GP) was located in fat lying on the dorsal surfaces of the two atria, medial to the roots of the pulmonary veins. This plexus extended ventrally in the fat lying between the two atria to form the (4) interatrial septal ganglionated plexus (IAS GP). (5) The left superior vena cava–left atrial ganglionated plexus (LSVC-LA GP) was found to be a relatively long, narrow collection of ganglia and nerves located in fat on the ventrolateral aspect of the left atrium (Fig. 2). This ganglionated plexus extended from the dorsal aspect of the left atrium, medial to the origin of the left superior vena cava, medially into the fat between the pulmonary artery and aorta.

Ventricular ganglionated plexuses.

Over 57 ganglia of various sizes, estimated to contain approximately 1,300 neurons, were identified in six ganglionated plexuses located in the fatty tissue on the cranial aspect of the ventricles (Table 1). (1) The craniomedial ganglionated plexus (CM GP) was identified in fat surrounding the root of the aorta and pulmonary artery. This ganglionated plexus extended along the left main coronary artery. At the bifurcation of the left coronary artery, a relatively dense collection of neurons was identified that coursed along the first 2 cm of the ventral interventricular and circumflex coronary arteries. These neurons and their associated nerves have been designated the (2) ventral interventricular artery ganglionated plexus (VIV GP; Fig. 1) and the (3) circumflex coronary artery ganglionated plexus (CCA GP), respectively. The latter plexus included ganglia and nerves in fat overlying the caudal border of the ventral left atrium. This ganglionated plexus was found to be contiguous with ganglia and nerves associated with the origin of the left superior vena cava, the LSVC-LA ganglionated plexus (see above). (4) The dorsal interventricular ganglionated plexus (DIV GP) was identified surrounding the first centimeter of the dorsal interventricular coronary artery (Fig. 2).

With respect to the right ventricle, the (5) right coronary artery ganglionated plexus (RCA GP) surrounded the first centimeter or more of the right main coronary artery. A smaller (6) right marginal artery ganglionated plexus (RMA GP) was identified surrounding the origin of the right marginal coronary artery (Fig. 1). Most ventricular neurons were identified in the ventral interventricular and right coronary artery ganglionated plexuses (Table 1). Several isolated neuronal somata were identified scattered throughout the interventricular septum, primarily in its cranial aspect; these did not coalesce to form distinct ganglionated plexuses.

Light Microscopy

Confocal microscopy of whole ganglia and nerves stained immunohistochemically with the generalized neuronal marker PGP 9.5 or the M35 muscarinic acetylcholine receptor antibody were used to reveal the three-dimensional morphology and relationships of small and large ganglia to nerves (Fig. 3A,B). Ganglia were frequently found at the branch points of nerves, whereas others were located along the course of or embedded in intrinsic cardiac nerves of the ganglionated plexuses. Unipolar neurons were occasionally identified in small clusters (Fig. 3C). Additionally, confocal and brightfield microscopy demonstrated numerous small satellite cells associated with neurons in the intrinsic cardiac ganglia (Fig. 3B).

thumbnail image

Figure 3. Confocal laser scanning microscopic photomicrographs of porcine intrinsic cardiac nerves, ganglia, and neurons. A: Ganglion at the juncture of nerves stained with rabbit protein gene product 9.5 (PGP 9.5). B: Unipolar neurons stained with M35 muscarinic receptor antibody. C: Ganglion composed of approximately 100 neuronal somata stained with PGP 9.5. Arrows point to unstained satellite cells. Scale bars = 50 μm in A, 20 μm in B,C.

Download figure to PowerPoint

Serial confocal sections through individual intrinsic cardiac ganglia, as well as routine histology, revealed that neuronal somata in many of the medium to large ganglia were often distributed in the periphery of these ganglia adjacent to the capsule (Fig. 4A). The interior of these ganglia was comprised of neuropil, with numerous dendrites of the peripherally located neurons (Fig. 4B,C) or small nerves. The somata of intrinsic cardiac neurons (n = 146) measured 33.1 ± 8.5 μm (short axis) × 46.3 ± 10.9 μm (long axis).

thumbnail image

Figure 4. A–C: Confocal laser scanning microscope photomicrographs showing three confocal slices through an intrinsic cardiac ganglion stained with rabbit protein gene product 9.5 (PGP 9.5). Note that the interior of this ganglion (B,C) is devoid of somata but contains dendritic processes (arrow in C). Scale bar = 20 μm in C (applies to A–C).

Download figure to PowerPoint

Electron Microscopy

Intrinsic cardiac ganglia had a well-defined capsule formed by fibroblasts and collagen fibers. The intrinsic cardiac neurons displayed the characteristic features of autonomic neurons with substantial cytoplasm and a large, eccentrically located nucleus with one or occasionally two nucleoli (Fig. 5A). Individual neurons were usually completely invested with a thin glial sheath (Fig. 5A), except where adjacent neurons were so closely apposed that their plasmalemmas touched without any intervening structure or tissue. Supporting glial or satellite cell somata were closely apposed to neuronal somata, sometimes indenting the neuronal profile (Fig. 5A) or proximal dendrite. These corresponded to the glial cells seen with confocal and brightfield microscopy (see Fig. 3C). Two types of satellite cells could be distinguished morphologically. Many satellite cells were characterized by an irregularly shaped, dark nucleus with condensed chromatin and little cytoplasm (Fig. 5A). A second type of satellite cell had a round, lightly stained nucleus with condensed chromatin at the periphery. These latter cells were present at the origins of neuronal dendrites (Fig. 5B,C) and in association with axons (Fig. 6A,B). Profiles of these cells exhibited a greater amount of cytoplasm, some ribbons of rough endoplasmic reticulum and granular vesicles (Figs. 5C, 6A,B).

thumbnail image

Figure 5. Electron photomicrographs of intrinsic cardiac neurons, satellite cells, dendritic processes, and axon terminals in porcine intrinsic cardiac ganglia. A: An intrinsic cardiac neuron showing a nucleus with two nucleolar profiles. The neuronal cytoplasm is rich in Golgi complexes and mitochondria. B: Satellite cell adjacent to soma and proximal dendrites. C: High magnification of boxed area in B. Scale bars = 5 μm in A,B, 2 μm in C.

Download figure to PowerPoint

thumbnail image

Figure 6. Electron photomicrographs of axon terminals, dendrites, and synapses in porcine intrinsic cardiac ganglia. A: Two axon terminals and two dendritic profiles in close association with satellite cells, one of which contains rough endoplasmic reticulum and granular vesicles. B: High magnification of the boxed area in A. Small, round clear vesicles cluster adjacent to the postsynaptic density (indicating an active zone). Note the profile containing numerous large dense-cored vesicles. C: An axodendritic synapse with an extensive synaptic contact. D: An axon terminal profile containing numerous large dense-cored vesicles intermixed with many small clear vesicles. Scale bars = 10 μm in A, 5 μm in B, 1 μm in C, 0.5 μm in D.

Download figure to PowerPoint

In the neuropil of ganglia, nerve fascicles containing myelinated and small unmyelinated axons wrapped in glial processes were surrounded by a collagen matrix. Similarly, dendrites, axons, and axon terminals were also wrapped by glial processes (Fig. 6A,B). Axon terminals usually contained clear, small, round vesicles and formed asymmetrical synapses with dendritic processes (Fig. 6B–D). Some axon terminals or boutons contained a few to many dense-cored vesicles (Fig. 6B,C). Axosomatic synapses were not identified in the material examined.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. SUMMARY
  7. Acknowledgements
  8. LITERATURE CITED

The results of the present study demonstrate that porcine intrinsic cardiac neurons are localized primarily in discrete areas within atrial and ventricular epicardial fatty tissue, as well as the interatrial septal fat. As with other mammals (King and Coakley, 1958; Gagliardi et al., 1988; Ardell et al., 1991; Yuan et al., 1994; Pauza et al., 2000), the porcine intrinsic cardiac nervous system contains significant numbers of neurons and interconnecting nerves that form ganglionated plexuses in specific regions of the heart. In contrast to the canine model (Yuan et al., 1994), the numbers of neurons identified in the pig heart were divided relatively evenly among the atrial and ventricular ganglionated plexuses (Table 1). The largest collection of neurons identified was that found on the ventral surface of the right atrium (Table 1; VRA GP). As in the dog (Yuan et al., 1994), fewer atrial neurons are located in the dorsal atrium and interatrial septum. In contrast to the canine model, the pig has a relatively small diameter left superior vena cava connected with the left portion of the coronary sinus on the dorsal aspect of the left atrium. Associated with the caudal-most region of this vein was a sizable collection of ganglia and interconnecting nerves, the left superior vena cava–left atrial ganglionated plexus (Fig. 2).

Four major ventricular ganglionated plexuses (the CM GP, VIV GP, CCA GP, and RCA GP) were identified in this study (Table 1). The one containing the largest number of ventricular neurons was associated with the right main coronary artery (RCA GP) as it coursed in the fat of the atrioventricular groove (Fig. 1). The next largest collection of ventricular neurons surrounded the origin of the ventral interventricular coronary artery (VIV GP). This area is readily accessible to functional studies such as have been conducted in other mammalian models (Armour and Hopkins, 1990b). The population of neurons surrounding the aorta (CM GP) was also considerable, as was that surrounding the origin of the CCA. That the porcine heart contains similar numbers of neurons associated with atrial and ventricular tissues may have implications with respect to neuronal control of local cardiac myocyte function. For instance, this anatomical arrangement may facilitate the neural control of atrioventricular nodal function and regional ventricular contractile behavior.

Relatively large mediastinal nerves arising from the vagosympathetic complexes and from the more cranially located mediastinal ganglia course between the root of the aorta and main pulmonary artery (Hopkins et al., 1997). These medial mediastinal nerves project relatively large branches caudally to the adjacent dorsal atrial ganglionated plexus, with smaller branches projecting ventrally to the ventral cranial medial ventricular ganglionated plexus. From there, small branches project to the ventricular interventricular ganglionated plexus. Medial mediastinal cardiopulmonary nerves also course dorsally to join the dorsal atrial ganglionated plexus, with smaller branches derived from these nerves connecting with the dorsal aspect of the right vena cava–right atrial ganglionated plexus. Small branches arising from the right thoracic vagosympathetic complex at the level of the right atrium course directly to the cranial portion of the ventral right atrial ganglionated plexus and the right vena caval–right atrial ganglionated plexus. No isolated left-sided nerve was identified that connected left-sided middle cervical ganglia with the left atrial ganglionated plexus, as is found in canines (Yuan et al., 1994). Thus, the major nerves arising from the right- and left-sided vagosympathetic complexes and more cranially located mediastinal ganglia interconnect with intrinsic cardiac ganglionated plexuses at the base of the porcine heart, primarily but not exclusively dorsal to the aortic root.

Porcine intrinsic cardiac ganglia of different sizes ranged from a few neurons to more than 100 neurons. Neurons were located peripherally in medium to large ganglia, with dendritic processes directed toward the interior of their ganglia where numerous axodendritic synapses were observed (Fig. 4). With this ganglion architecture, it was not possible to accurately ascertain from external appearances the actual number of somata in larger ganglia. Consequently, the estimation of the total numbers of neurons listed in Table 1 is based on more precise counts of neurons in smaller ganglia, with an estimate of a maximum of 100 neurons derived from the surface layers of larger ganglia. As a result, these estimates likely represent an underestimation of the actual total numbers of neurons in the porcine intrinsic cardiac nervous system.

Most ganglia were situated along the course of a nerve, at nerve junctions or as clusters of neurons surrounding a nerve (Fig. 3A). Confocal microscopic analysis confirmed rosette-like formations of neuronal somata in larger ganglia, with their dendritic processes being directed centrally to the interior of the ganglion (Fig. 4). A similar arrangement has been observed in canine (Yuan et al., 1994) and human (Armour et al., 1997) intrinsic cardiac ganglia. Additionally, unipolar neurons were also identified in porcine intrinsic cardiac ganglia. This finding is in agreement with findings in other species (Dogiel, 1899; Davies et al., 1952; Hoover et al., 1994; Yuan et al., 1994; Pauza et al., 1997). Somata in these ganglia could be immunostained with the nonspecific M35 muscarinic receptor antibody (Fig. 3C). Collectively, these anatomical findings are consistent with the complex physiological and neurochemical observations obtained in functional studies of porcine intrinsic cardiac ganglia (Crick et al., 1999a, b; Smith, 1999; Galoyan et al., 2001).

Electron microscopy of the porcine intrinsic cardiac nervous system demonstrated cytological features typical of autonomic neurons. The neuronal organization identified within porcine intrinsic cardiac ganglia was similar to that found in guinea pig, dog, and human intrinsic cardiac ganglia (King and Coakley, 1958; Robb, 1965; Yuan et al., 1994; Armour et al., 1997). Specifically, axodendritic synapses were common, especially in the middle of the ganglia where the dendrites ramified. A similar predominance of axodendritic synapses and paucity of axosomatic synapses has been found in canine (Yuan et al., 1994) and monkey (Ellison and Hibbs, 1976) intrinsic cardiac ganglia. In contrast, intrinsic cardiac ganglia of rats and rabbits are reported to have many axosomatic synapses (Ellison and Hibbs, 1976; Papka, 1976), whereas those in cats and guinea pigs have both axosomatic and axodendritic synapses (Ellison and Hibbs, 1976).

Two main types of axon terminals were identified in porcine intrinsic cardiac ganglia. First, terminals containing many small clear vesicles and a few small, dense-cored vesicles (Fig. 6A–C) were identified. The presence of axon terminals associated with such anatomy has been considered by others to be cholinergic in nature (Ellison and Hibbs, 1976; Shvalev and Sosunov, 1985). Second, axon terminals that contain numerous large, dense-cored vesicles and small translucent vesicles (Fig. 6D) were identified. A few small cells containing granular vesicles were also identified, as reported in guinea pig, cat, and rabbit intrinsic cardiac ganglia (Ellison and Hibbs, 1976; Papka, 1976). Complex arrangements of membrane specializations between dendrites were also observed (Fig. 6C), as has been found in human intrinsic cardiac ganglia (Armour et al., 1997). This varied synaptology supports the concept that complex interneuronal relationships exist within the porcine intrinsic cardiac nervous system.

SUMMARY

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. SUMMARY
  7. Acknowledgements
  8. LITERATURE CITED

This study identified 11 major atrial and ventricular ganglionated plexuses associated with the porcine heart. Porcine ventricular ganglionated plexuses possess a larger proportion of neurons than canine ventricular ganglionated plexuses. As observed in other species, porcine intrinsic cardiac ganglia contain neurons with varied morphologies, as well as complex synaptic and dendritic relationships. These data indicate that the porcine model is suitable for the study of the functional interactions that occur within the mammalian intrinsic cardiac nervous system, particularly with respect to its ventricular components.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. SUMMARY
  7. Acknowledgements
  8. LITERATURE CITED

The authors thank Richard Livingston, Sara E. MacDonald, and Stephen Whitefield for their technical assistance.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. SUMMARY
  7. Acknowledgements
  8. LITERATURE CITED