| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
NEUROHUMORAL CONTROL OF CIRCULATION AND HYPERTENSION
1Departments of Pediatric, Kosair Children's Hospital Research Institute and Departments of 2Physiology and Biophysics and 4Pharmacology and Toxicology, University of Louisville School of Medicine, Louisville, Kentucky 40202; and 3Department of Physiology, Stritch School of Medicine, Loyola University Chicago, Maywood, Illinois 60153
Submitted 21 March 2003 ; accepted in final form 17 November 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
brain stem; parasympathetic; anterograde tracing; heart; baroreflex
Vagal efferent preganglionic axons arise mainly from neurons in two brain stem nuclei [the nucleus ambiguus (NA) and the dorsal motor nucleus of the vagus (DmnX)], whereas a small number of vagal efferent axons arise from neurons in the intermediate zone between DmnX and NA. The functional roles of cardiac neurons in these nuclei are not well understood (6, 8, 11, 13, 16, 25). Furthermore, neurons in different cardiac ganglia control heart rate, arteriovenous (A-V) conduction, and myocardial contractility (17, 18, 20-22). In contrast to the conventional concept that cardiac ganglia are simple relay stations for central nervous system input, these structures are more likely operating as complex integration centers (19, 23). They consist of sensory neurons, motor neurons, and interneurons that employ complex chemical coding (12, 26) and are involved in local, as well as central, cardiac reflexes (1, 19, 23, 27). Thus a major question is how cardiac ganglia are innervated and controlled by neurons in the NA and DmnX in the brain stem.
Traditionally, it has been assumed that the NA is the major vagal motor neuron pool controlling the heart, whereas neurons in the DmnX play only a minor role in cardiac control (13, 16). Using anterograde tracing and confocal microscopic techniques, however, we recently demonstrated that neurons in both NA and DmnX project to each cardiac ganglion where their axons form extensive basket endings around ganglionic principal neurons (6, 8). Therefore, this anatomic evidence suggests that both nuclei could be important for cardiac function. Consistent with this morphological evidence, physiological data also support the functional significance of the two nuclei: 1) electrical stimulation of vagal efferent B fibers (presumably from NA) and vagal C fibers (presumably from DmnX) selectively evokes bradycardia, A-V block, and reduction of cardiac contractility (9, 14, 15, 29); 2) stimulation of the DmnX elicits bradycardia and reduces myocardial contractility (4, 10, 24); 3) activation of the NA induces negative chronotropic, dromotropic, and inotropic effects (2, 3, 17, 28); and 4) both DmnX and NA neurons are barosensitive (30). However, whether the two vagal motor nuclei perform different functional roles remains unclear.
To study the difference between the NA and DmnX, we first compared cardiac projections of the two nuclei (6, 8). DmnX and NA motor neuron projections to the heart differ significantly: 1) efferent axons originating from NA neurons project to the heart and diverge and innervate principal neurons three times as much as the efferent fibers arising from DmnX neurons, 2) axons arising from NA neurons are larger in caliber than those arising from DmnX neurons, and 3) fibers from DmnX neurons project to small intensely fluorescent cells (SIF; presumably interneurons) as well as to cardiac principal neurons, whereas axons from NA neurons innervate only principal neurons. Furthermore, we used domoic acid, an excitatory neurotoxin that acts on glutamate receptors, to selectively lesion the NA and DmnX. Lesions of the NA almost completely abolished the baroreflex control of the heart rate, whereas lesions of the DmnX did not, suggesting that the NA plays a more important role in baroreflex control of the heart rate than the DmnX (5, 31).
Morphological and physiological differences between the NA and DmnX raise the question of whether the NA and DmnX control the heart through different pathways and whether axons arising from neurons in these two nuclei target different populations of intrinsic cardiac principal neurons. We hypothesized that vagal fibers originating from neurons in the NA and DmnX project separately to cardiac ganglia, so that they innervate different populations of cardiac neurons and control different aspects of cardiac function. Alternatively, axons arising from NA and DmnX neurons might converge to innervate the same cardiac neurons because recruitment of C fibers from the DmnX does not increase the degree of cardiac slowing produced by stimulation of B fibers arising from NA (cf., Ref. 13).
To elucidate these issues, we used a double-labeling strategy and confocal microscopy to examine dual projections of the NA and DmnX neurons to the heart. Our results show that neurons in the NA and DmnX project to neurons throughout individual cardiac ganglia. Rather than innervating the same principal neurons, however, these two vagal nuclei project to separate subpopulations of principal neurons within each cardiac ganglion.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In the first tracing group, each animal received a series of unilateral DmnX injections of the tracer 1,1'-dioleyl-3,3,3',3'-tetramethylindocarbocyanine methanesulfonate, rhodamine red (DiI; catalog 3886; Molecular Probes, Eugene, OR) paired with injections of 4-[4-(di-hexadecylamino)-styryl]-N-methylpyridinium iodide, fluorescein yellow-green (DiA; catalog 3883; Molecular Probes) in the contralateral NA (left DmnX: right NA, n = 4; right DmnX: left NA; n = 4). In the second tracing group, each rat received a series of unilateral DmnX injections of DiI paired with injections of DiA in the ipsilateral NA (left DmnX: n = 4; right DmnX: n = 4). In the third tracing group, the tracers DiI and DiA were switched at each injection site. Each rat received a series of unilateral DmnX injections of the tracer DiA paired with injections of DiI in the ipsilateral (n = 4) or contralateral (n = 4) NA to test the tracer effect. In the fourth group, each rat (n = 2) was injected with DiI in the left DmnX, and the left cervical vagal trunk was transected immediately after the DiI injection to assess the extent of DiI spread in the contralateral DmnX. Animals in these groups were also injected with Fluoro-Gold (FG: ip, 2 ml of 2 mg/ml) to counterstain cardiac ganglia. In a reference group, rats (n = 10) were injected with FG (3 mg/ml, 5 µl) in the cervical vagal trunk to identify DmnX and NA motor neurons. This group was used to ascertain the precision of DiI injections in DmnX and of DiA injections in the NA, i.e., the registration of DiI in the brain stem with FG-labeled DmnX motor neurons and the registration of DiA with NA motor neurons.
Reference animals from the fourth group were killed 5 days after injections to confirm tracer injection sites within the brain stem. In contrast, the other three experimental groups were killed 3 wk after injections, and the brain stem tissue, nodose ganglia, and atrial specimens of these animals were harvested for analysis of NA and DmnX fibers and endings in cardiac tissues. In the event that tracer injection sites were off center from the target nuclei, data were discarded from further analysis.
Tracer injection in the DmnX and NA. Each animal in the tracing groups was anesthetized with pentobarbital sodium (60 mg/kg ip), treated with atropine (1 mg/kg sc), and placed in a stereotaxic instrument equipped with a head holder adapted to permit the neck to be flexed sharply. A dorsal incision was made over the neck muscles, which were retracted to expose the atlanto-occipital membrane. The membrane was opened with an incision, exposing the cisterna magna and the dorsal medulla. The occipital bone was trimmed with the bit of a dental drill until the caudal cerebellum was visible. The caudal end of the area postrema was used as a reference for stereotaxic coordinates. A glass micropipette, filled with DiI or DiA and connected to a picospritzer, was then advanced to the DmnX or NA. DiI or DiA was injected in small aliquots (2.5-12.5 nl each) either at seven different sites for DmnX (-1,200 to +1,200 µm; total volume 17.5-87.5 nl) or at nine different sites (-1,600 to +1,600 µm; total volume 22.5-112.5 nl) for NA, separated 400 µm longitudinally. All analyzed animals had well-placed injections, as confirmed by postmortem examination; medullary injection sites tended to fuse longitudinally. Injections were considered to be well placed only if 1) the injections, and their central cores, were centered in the DmnX and NA at all frontal levels, 2) the injection spheres extended minimally both ventral to the DmnX and dorsal to the NA (see injection sites in Fig. 1; also Fig. 1 in Ref. 8 and Fig. 1 of Ref. 6). Diameters of DiI and DiA injections in NA are 
300 and 150 µm, respectively, and 3) the series of injections primarily covered the bulk of the longitudinal extent of the DmnX and the NA.
|
Tissue preparation. Three weeks were allowed for the tracers to be transported to the heart. Animals were anesthetized with an overdose of pentobarbital sodium (100 mg/kg) and perfused through the heart with 0.9% saline (300 ml) and 10% phosphate-buffered (pH = 7.4, 600 ml) formalin. Nodose ganglia and brain stems were removed. Each brain stem containing the entire DmnX and NA was stored in 10% sucrose formalin overnight and sectioned transversely at 100 µm on the second day. Thoracic vagal trunks, with the cardiopulmonary branches attached from the level of recurrent laryngeal nerves to a level 3 mm caudal to the junction of the inferior vena cava and the right atrium, were also dissected. Atria were then separated and cut open as previously described (7). Briefly, the left atrium tissue block included the region of the junction with pulmonary veins; the right atrium had the superior vena cava, inferior vena cava, and the left precava vein attached. The interatrial septum was separated from atria. Tissue surrounding the heart was gently removed, using extreme care to avoid peeling off the ganglionated plexuses (8) on the dorsal surface of the atria. Tissue was then dehydrated through graded concentrations (70%, 90%, and 2 x 100%) of glycerin for 4 h. Finally, the tissue was mounted and covered with a cover slip in 100% glycerin and n-propyl gallate (5%) to prevent fading.
Examination of DiI and DiA injection sites. Brain stem slices were initially examined at x100 magnification using an epifluorescence microscope equipped with filter cubes appropriate for DiI, DiA, and FG and were later scanned using a confocal microscope (Zeiss LSM 510) at x250 magnification. To present the registrations of DiI and DiA injection sites in vagal motor nuclei in a complete fashion, a montage of individual confocal projection images was assembled using Adobe Photoshop 5.5. DiI injection sites were evident by labeling with orange-red to red colors; DiA injection sites were in the yellow-green or green part of the spectrum; and FG-labeled neurons appeared blue in color.
Data acquisition and analysis of dual vagal projections in intrinsic cardiac ganglia. A detailed description of the data acquisition strategy has been published previously (6, 8). Briefly, cardiac and nodose ganglia specimens were screened with a conventional epifluorescence microscope equipped with filter cubes appropriate for DiI (rhodamine), DiA (FITC), and FG (UV). When DiI and DiA nerve fibers and endings were both found in the same region of cardiac ganglia at x200 magnification, their locations were recorded for subsequent detailed confocal microscopic analysis at x400 magnification. The digitized confocal images were stored on compact disks.
Consistent with earlier observations (6, 8), three major ganglionated plexuses labeled by FG were identified in the epicardium of the atria under UV light. Within each plexus, multiple ganglia were found (8). Using stacks of optical sections collected with the confocal microscope through filter cubes appropriate to DiI and DiA, we examined the dual innervation of DmnX and NA cardiac axons and their terminal endings in each of the three ganglionated plexuses. Dual vagal axonal innervation of the three plexuses was examined systematically and completely. Stacks of confocal optical sections that contained both DiI and DiA fibers and endings were analyzed sequentially, and each cell apposed by labeled varicosities or fiber swellings was considered as being innervated. The presence and close proximity of varicosities were the criteria for judging a labeled fiber to have contact with a cell (6, 8).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
DmnX and NA efferent fibers innervate different populations of intrinsic cardiac ganglionic neurons. Both DmnX and NA efferent fibers entered the epicardium of the atria and innervated the ganglia in all identified cardiac ganglionated plexuses (8). Frequently, DmnX and NA fibers and endings were found in the same cardiac ganglion with axons forming basket endings around cardiac principal neurons. As previously observed, DmnX neurons innervated SIF cells as well. In contrast, NA neurons did not. Figure 2 illustrates three examples showing that the DmnX and NA project to nonoverlapping, different subpopulations of neurons within the same ganglion. We have carefully examined such doubly labeled DiI and DiA terminals in 605 frames of confocal images randomly sampled from three ganglionated plexuses (which included 3,025 DiI and 1,815 DiA basket endings) and found that DmnX and NA neurons project to completely distinct populations of cardiac principal neurons in these plexuses.
|
|
|
Contralateral DmnX and NA efferent fibers innervate neurons in separate subregions of intrinsic cardiac ganglionic plexuses. Axons arising from neurons in contralateral DmnX and NA entered a cardiac ganglionated plexus via separate routes or connectives at different entry points (a connective is a mixture of extrinsic and intrinsic nerves connecting two separated cardiac ganglia in a ganglionated plexus). Each plexus contains a number of ganglia. Figure 4 includes three montages that show the right NA and left DmnX project to a cardiac ganglionated plexus (also see Fig. 5 of Ref. 8 for an FG-labeled cardiac ganglionated plexus and different connectives at low magnification). Figure 4A shows that a bundle of DiA-labeled NA fibers enters the plexus from a connective at one entry region, tending to innervate principal neurons situated near the entry region. Figure 4B shows that a bundle of DiI-labeled DmnX axons enters the same plexus through another entry point, to innervate principal neurons at the immediate entry region. Between these two opposite entry points, the DiA- and DiI-labeled basket endings intermingle and innervate principal neurons interdigitately, as shown in Fig. 4C. Therefore, neurons in the left and right sides of the brain stem appear to innervate topographically separated subdivisions of neurons in cardiac ganglionated plexuses.
Projection of DmnX and NA efferent fibers to cardiac ganglia: counterbalancing DiA and DiI. To examine whether our results were influenced by the differential fluorescence properties of DiI and DiA, we switched the tracers used to inject the DmnX and NA. In eight animals, we injected DiA unilaterally in the DmnX and DiI in the NA (either ipsilateral or contralateral). Comparable to previous labeling using alternate tracers in the brain stem nuclei, DiA-labeled baskets are among numerous DiI baskets. Individually, DiA- and DiI-labeled basket formations were evident. Labeled axons arising from the DmnX and NA innervated different populations of principal neurons in the same cardiac ganglion (Fig. 4, D-E). Therefore, our results did not depend on tracer fluorescence properties.
In contrast to the dense DiI-labeled axons and terminals projecting from the DmnX in Fig. 3, the DiA-labeled axons and basket endings projecting from the DmnX in Fig. 4D were much more sparse than the DiI-labeled axons and terminals originating from the NA, indicating that DiA is a weaker tracer.
Specificity of DiI and DiA labeling. We injected tracers to the DmnX and NA to label the cardiac motor axons and terminals in the heart. Three types of potential secondary labeling could occur. First, a tracer injection in the DmnX could also label ipsilateral NA neurons and vice versa. Second, a tracer injection in the DmnX or NA could also label the contralateral DmnX or NA. Third, a tracer injection in the DmnX or NA could label the nucleus of the solitary tract (NTS) region, and the vagal afferent terminals might pick up the tracer and transport it to the heart. For example, DiI injection in the DmnX may label NA, contralateral DmnX, and NTS; and DiA injection in the NA may label contralateral NA, DmnX, and NTS. Obviously, such potential problems could affect the interpretation of our findings.
According to our previous studies and after thorough examination of the brain stem, nodose ganglia, and atrial tissues used in the present study, we did not observe any evidence suggestive of such technical problems. Indeed, previously we have shown that DiI injections in the DmnX or NA did not label the other nucleus (6, 8). The present study confirms these findings. In addition, after examination of the atrial tissue whole mounts, no flower spray sensory terminals were found. Because flower-spray terminals are the major type of afferent terminals (7), we conclude that tracer injections did not label any vagal afferents in the heart. Examination of brain stem serial sections did not show any labeled contralateral NA neurons. In contrast, as shown in Fig. 1, tracer injections in the DmnX could label the contralateral DmnX, especially in the caudomedial portion of the nucleus, since the left and right DmnX are close to the midline and to each other at this level. Such contralateral labeling may account for the occasional DiI axons in the DiA bundle. However, this relatively limited technical issue concerning the two sides of the caudal DmnX has no bearing on our major observation that the DmnX and NA project to different populations of cardiac neurons.
It should be pointed out that it was important to make a relatively large injection of DiI in the DmnX, as shown in Fig. 1, because a major aim of the present study was to test the hypothesis that the NA and DmnX project to the same cardiac ganglia but different populations of principal neurons therein. Therefore, we aimed to label as many DmnX and NA cardiac motor neurons as possible such that more DiI- and DiA-labeled terminals colocalizing at the same region within a cardiac ganglion would be available for analysis.
Such strategy, however, could lead to a potential leak of the tracer to the contralateral DmnX, as shown in Fig. 1, and might affect our finding that neurons in the right and left brain stem topographically project to the different subdivisions of the plexus. To eliminate such possibility, we quantitatively estimated the maximal number of fibers being labeled from the contralateral DmnX and the innervation pattern of these axons. DiI was injected in the left DmnX. The left cervical vagal trunk was transected in two rats to evaluate the extent of spread in the contralateral DmnX. In these animals, 5 to 5-7 axons and 20-28 basket endings were found, corresponding to 7.4% of the total axons and 3.9% of the total baskets, which were estimated previously as originating from the left DmnX projection (8). Thus only a minority of fibers on the contralateral DmnX was labeled by injections we employed in our experiments. In contrast to the well-labeled DiI axons from the ipsilateral DmnX that entered the heart in large bundles (Fig. 4, A and B; also see Refs. 6 and 8), these sparse secondarily labeled axons were weakly labeled and randomly distributed in the cardiac ganglionated plexuses. Because in the majority of other animals the large DiI and DiA bundles were separated and entered ganglionated plexuses through different pathways and because the secondarily labeled fibers from the contralateral DmnX appeared weakly labeled, and sparsely and randomly distributed, the secondary labeling of the contralateral DmnX neurons should not detract from our findings.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Topographical representation of left and right vagus in cardiac ganglionated plexuses. As mentioned above, the left and right vagal cardiac axons enter the cardiac ganglionated plexus through different routes. This observation raises the possibility that the left and right brain stem may have a well-defined topographical representation of the heart. Furthermore, this may indicate that cardiac ganglionic plexuses are not homogeneous structures and that they may be organized hierarchically to receive functionally related inputs from the vagal brain stem nuclei in a topographical pattern. Whether our findings provide the anatomic basis for certain physiological observations indicative of left and right vagus lateralization is worthy of future investigation.
Regarding the topographic projections of the vagal motor nuclei to cardiac ganglionated plexuses, an additional intriguing issue is whether the different regions of the DmnX and NA project to different areas of cardiac ganglionated plexuses. In dogs and cats, Massari and colleagues (3, 17) found that different locations of the NA projected to different cardiac ganglia in fat pads and had differential control over the heart rate, A-V conduction, and left ventricle contractility. Such data in the rat are not currently available. Our present dual-labeling strategy, i.e., injection of different areas of the NA or DmnX with two tracers simultaneously and subsequent examination of cardiac ganglionated plexuses with the confocal microscope, may provide a novel method for future investigation of this issue.
NA and DmnX cardiac motor neurons: functional signifi-cance. Conventionally, the NA was considered the major vagal motor nucleus that controls the heart. Recently, we demonstrated that neurons in the DmnX also project substantially to the heart, implying that the DmnX might control cardiac function as well. The present finding that NA and DmnX project to the same cardiac ganglia suggests that they could both modulate the heart rate, A-V conduction, and myocardial contractility (6) because the adjacent cardiac principal neurons are thought to have similar cardiac functions. From these data, we reasoned that NA and DmnX might be both involved in baroreflex control of heart rate.
To study the functional role of the NA and DmnX, we injected domoic acid to disrupt the integrity of the NA and DmnX, respectively, and studied their respective roles in baroreflex effect on the heart rate. To our surprise, lesions of the DmnX did not affect baroreflex sensitivity (5). In contrast, lesions of the NA almost completely eliminated baroreflex control of heart rate, indicating that the NA is critical for baroreflex control (31), whereas the DmnX is not.
Whether the DmnX may control the heart through some other cardiac reflexes is an interesting issue. Because the NA and DmnX innervate nonoverlapping populations of cardiac principal neurons and play different roles in baroreflex, we hypothesize that the NA and DmnX may play distinct roles in different cardiopulmonary reflexes. Actually, it has been postulated that the DmnX may mediate pulmonary C-fiber-evoked bradycardia (15, 28).
In summary, neurons in two spatially separated medullary nuclei, the NA and DmnX, project axons to separate nonoverlapping populations of interspersed cardiac neurons. These findings provide the anatomic framework to account for the possibility that the NA and DmnX may have different functional roles in controlling the heart. The exact nature of such potential functional differences of the two nuclei and the property of the two populations of cardiac principal neurons await further studies.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
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 |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Kong, J.-H. Liu, A. G. Ramage, and Y. Wang Cardiovascular Control: Cardiac vagal preganglionic neurones in the intermediate zone of the brainstem in anaesthetized cats Exp Physiol, November 1, 2007; 92(6): 1023 - 1028. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Gu, M. Lin, J. Liu, D. Gozal, K. E. Scrogin, R. Wurster, M. W. Chapleau, X. Ma, and Z. Cheng Selective impairment of central mediation of baroreflex in anesthetized young adult Fischer 344 rats after chronic intermittent hypoxia Am J Physiol Heart Circ Physiol, November 1, 2007; 293(5): H2809 - H2818. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Lin, R. Liu, D. Gozal, W. B. Wead, M. W. Chapleau, R. Wurster, and Z. Cheng Chronic intermittent hypoxia impairs baroreflex control of heart rate but enhances heart rate responses to vagal efferent stimulation in anesthetized mice Am J Physiol Heart Circ Physiol, August 1, 2007; 293(2): H997 - H1006. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. K. Soukhova-O'Hare, Z. Cheng, A. M. Roberts, and D. Gozal Postnatal intermittent hypoxia alters baroreflex function in adult rats Am J Physiol Heart Circ Physiol, March 1, 2006; 290(3): H1157 - H1164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Liu and M. T. T. Wong-Riley Postnatal developmental expressions of neurotransmitters and receptors in various brain stem nuclei of rats J Appl Physiol, April 1, 2005; 98(4): 1442 - 1457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. M. Verberne Differential cardiac parasympathetic innervation--what is the functional significance? Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2004; 287(2): R485 - R486. [Full Text] [PDF]
|
|
|
|
|
|
P. B. Persson and J. A. Armour Dual vagal cardiac efferent pathways Am J Physiol Regulatory Integrative Comp Physiol, April 1, 2004; 286(4): R624 - R624. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
