Frog Cardiac Ganglion Research Articles

Origin and morphogenesis of neurons in the frog cardiac ganglion.

This paper reviews the cellular events underlying the formation of the cardiac ganglion in the frog. The first neurons become postmitotic at the end of embryogenesis and begin differentiating in a functioning heart. Neuronal precursors in the heart continue dividing and differenting at least through the beginning of metamorphosis. Since cell death is not a characteristic of early cardiac ganglion development, proliferation and differentiation alone regulate the addition of neurons to the heart. As new neurons are added, their position is unique and reflects their time of origin. The developmental addition of new neurons to the ganglion is neither time nor stage dependent, but matches the size of the growing heart, even into adult life. During this period, the shape and polarity of cardiac neurons is established through elimination of supernumerary axons that initially extend throughout the heart. The temporal and spatial accumulation of cardiac SIF cells occurs in parallel with cardiac neurons and may indicate a common mechanism underlying the origin of these neural crest derivatives. These studies contribute to our understanding of the cellular processes underlying the development of this important autonomic control mechanism and provide the background for current studies on the regulation of cardiac neurogenesis.

Effects of denervation on acetylcholine receptor clusters on frog cardiac ganglion neurons as revealed by quantitative laser scanning confocal microscopy

We examined the effects of denervation on clusters of nicotinic acetylcholine receptors (AChRs) on autonomic neurons in the frog heart using immunofluorescence techniques and laser scanning confocal microscopy. We showed previously that normally innervated neurons have both large, brightly stained AChR clusters and small, dim AChR clusters. A majority (80%) of the large/bright AChR clusters are located at synaptic sites, whereas the small/dim clusters are distributed widely over the cell surface. Here, we use image analysis to identify these two classes of clusters on images generated from stacks of optical sections through neuronal cell bodies and to examine the effects of denervation on their number, size, and brightness (pixel intensity). Denervation reduces the number of large/ bright AChR clusters per cell to < 10% of sham-operated values and increases the number of small/dim clusters per cell by two- to threefold. These changes occur at 4 d of denervation, the earliest time examined, and are sustained for 6 weeks. The size of large/bright AChR clusters is decreased compared with sham-operated controls, and their brightness is unchanged. The size of small/dim AChR clusters is unchanged by denervation, but their brightness is increased. Denervation results in a shift in the contribution of each AChR cluster class to the total measurable AChR pool-from one dominated by large/bright clusters to one dominated by small/dim clusters. These results show that the nerve terminals on cardiac ganglion neurons appear to exert a continual and reversible influence on the organization of the postsynaptic membrane.

Synaptic and extrasynaptic distribution of two distinct populations of nicotinic acetylcholine receptor clusters in the frog cardiac ganglion.

We examined the distribution of neuronal nicotinic acetylcholine receptor clusters in relation to synaptic sites on autonomic neurons in the frog heart using immunofluorescence techniques and laser scanning confocal microscopy. Acetylcholine receptor clusters were visualized using the rat anti-Electrophorus acetylcholine receptor monoclonal antibody no. 22 and cyanine 3.18-labelled goat anti-rat secondary antibody. Synaptic boutons were labelled with the mouse anti-synaptic vesicle protein SV2, monoclonal antibody no. 10h and cyanine 5.18-labelled goat anti-mouse secondary antibody. Acetylcholine receptor clusters on the neuronal surface exist in two populations that vary in size, staining intensity, and surface distribution. The more prominent population consists of large, brightly stained clusters numbering 30 +/- 15 per cell, while the second class is smaller and less brightly stained and numbers over 100 per cell. The large clusters tend to be organized into groups of 2-6 members. This arrangement results from the fact that 80% of the large clusters colocalize at synaptic boutons and that single boutons can have several associated clusters. The remaining 20% of large/bright acetylcholine receptor clusters are extrasynaptic, but they, too, are clustered and are found in close proximity to synaptic boutons. The small/dim acetylcholine receptor clusters are randomly distributed over the cell surface. The large/bright synaptic acetylcholine receptor clusters presumably underlie fast excitatory synaptic transmission. The small/dim clusters and the large/bright extrasynaptic clusters may represent intermediates in the metabolism of large/bright synaptic clusters.

The characterization of α-bungarotoxin receptors on the surface of parasympathetic neurons in the frog heart

Receptors for α-bungarotoxin are found on the surface of parasympathetic neurons in the frog cardiac ganglion by light microscopic autoradiography. Competition studies suggest that these receptors are cholinergic and indicate that they are also recognized by neuronal bungarotoxin (κ-bungarotoxin). These receptors are outnumbered by those recognized exclusively by neuronal bungarotoxin. Unlike neuronal bungarotoxin receptors, α-bungarotoxin receptors are not concentrated at synaptic sites. Fluorescence techniques fail to find evidence for clusters of α-bungarotoxin receptors anywhere on the neuronal surface. The possible function of these receptors, which apparently do not play a role in fast synaptic transmission, is discussed.

Denervation does not alter the number of neuronal bungarotoxin binding sites on autonomic neurons in the frog cardiac ganglion

The binding of neuronal bungarotoxin (n-BuTX; also known as bungarotoxin 3.1, kappa-bungarotoxin, and toxin F) was analyzed in normal and denervated parasympathetic cardiac ganglia of the frog Rana pipiens, n-BuTX blocks both EPSPs and ACh potentials at 5-20 nM, as determined by intracellular recording techniques. Scatchard analysis on homogenates indicates that cardiac ganglia have two classes of binding sites for 125I-n-BuTX: a high-affinity site with an apparent dissociation constant (Kd,app) of 1.7 nM and a Bmax (number of binding sites) of 3.8 fmol/ganglion and a low-affinity site with a Kd,app of 12 microM and a Bmax of 14 pmol/ganglion. alpha-Bungarotoxin does not appear to interfere with the binding of 125I-n-BuTX to either site. The high-affinity binding site is likely to be the functional nicotinic ACh receptor (AChR), given the similarity between its affinity for 125I-n-BuTX and the concentration of n-BuTX required to block AChR function. Light microscopic autoradiographic analysis of 125I-n-BuTX binding to the ganglion cell surface reveals that toxin binding is concentrated at synaptic sites, which were identified using a synaptic vesicle-specific antibody. Scatchard analysis of autoradiographic data reveals that 125I-n-BuTX binding to the neuronal surface is saturable and has a Kd,app similar to that of the high-affinity binding site characterized in homogenates. Surface binding of 125I-n-BuTX is blocked by nicotine, carbachol, and d-tubocurarine (IC50 less than 20 microM), but not by atropine (IC50 greater than 10 mM). Denervation of the heart increases the ACh sensitivity of cardiac ganglion cells but has no effect upon the number of high-affinity binding sites for 125I-n-BuTX in tissue homogenates. Moreover, autoradiographic analysis indicates that denervation does not alter the number of 125I-n-BuTX binding sites on the ganglion cell surface. n-BuTX is as effective in reducing ganglion cell responses to ACh in denervated ganglia as it is in normally innervated ganglia. These results suggest that denervation alters neither the total number of nicotinic AChRs in the cardiac ganglion nor the number found on the surface of ganglion cells. These autonomic neurons thus respond differently to denervation than do skeletal myofibers. The increase in ACh sensitivity displayed by cardiac ganglion cells upon denervation cannot be explained by changes in AChR number.

Distribution of Synaptic Specializations in the Frog Cardiac Ganglion.

1. Szamier, R. B., and H. Ripps. 1983. J. Comp. Neurol. 215: 51-62. 2. Bruun, A., et 41. 1984. Bruin Rex 295: 233-248. 3. Brunken, W. J., et al. 1986. J. Comp. Neurol. 243: I-12. 4. Schlemermeyer, E., and R. L. Chappell. 1991. Invest. Ophthalmol. and Visual Sci. 32: 1264. 5. Weiler, R., and M. Schiitte. 1985. Cell. Tiss. Rex 241: 373-382. 6. Ritchie, T. C., and R. B. Leonard. 1983. Bruin Res. 267: 352-356. I. Villar, M. J., et al. 1987. Neuroscience 22: 681-686. 8. Schiitte, M., and R. Weiler. 1988. Neurosci. 91: 289-294. Let. 9. Schtitte, M., and P. Witkovsky. 1990. J. Neurocyfol. 19: 504-5 18.

Acetylcholine receptor-like molecules are found in both synaptic and extrasynaptic clusters on the surface of neurons in the frog cardiac ganglion

Sixty monoclonal antibodies (mAbs) made against nicotinic acetylcholine receptors (AChRs) from electric organ were tested for their ability to cross-react at synaptic sites in the frog cardiac ganglion, where transmission is mediated by ACh via nicotinic receptors. Forty-one of the mAbs tested were known to bind to AChRs in frog skeletal muscle (Sargent et al., 1984). As determined by double-label immunofluorescence microscopy, 8 of the 60 mAbs bound to small, punctate sites that lay within areas of synaptic contact, marked by an anti-synaptic vesicle antibody. One of the 8 cross-reacting antibodies that produced particularly intense staining (mAb 22) was chosen for further study and was found to bind to the postsynaptic membrane beneath active zones, as determined by peroxidase immunocytochemistry and electron microscopy. This suggests that mAb 22 crossreacts with AChRs on the ganglion cell surface. While most mAb 22 immunoreactivity was located in the postsynaptic membrane, about 20% of the peroxidase-stained patches were extrasynaptic (327 patches analyzed from 8 ganglia). Virtually no peroxidase-stained patches were observed when an isotype-matched control mAb was used in place of mAb 22. In frog skeletal muscle peroxidase-stained patches obtained with mAb 22 were found exclusively at synaptic sites (179 patches examined from 6 muscles). These results suggest that extrasynaptic patches of AChRs are present in innervated autonomic neurons but not in innervated skeletal muscle.

Denervation alters the size, number, and distribution of clusters of acetylcholine receptorlike molecules on frog cardiac ganglion neurons

Acetylcholine receptor (AChR)-like molecules are found in clusters on the surface of parasympathetic neurons in the frog cardiac ganglion. Electron microscopy of immunoperoxidase-stained tissue reveals that in normally innervated ganglia most of these clusters are located at synaptic sites. Denervation for 2–3 weeks results in a 64% reduction in the total surface area occupied by AChR-like clusters; this change is brought about by the combined effects of a 4-fold decrease in cluster size and a 30% increase in cluster number. Denervation also changes the distribution of AChR-like clusters: clusters, normally restricted to portions of the cell surface, are more widely distributed following denervation. Denervation of amphibian skeletal muscle for a comparable period of time has no effect on the size or the number of synaptic clusters of AChRs. These results suggest that AChRs in nerve and in muscle are regulated differently by innervation.

The influence of normal innervation upon abnormal synaptic connections between frog parasympathetic neurons

Synaptic connections form between parasympathetic neurons in the frog cardiac ganglion when their normal preganglionic innervation is removed. The synapses between the ganglion cells are inappropriate in the sense that they are absent in normally innervated ganglia. These novel synapses form slowly; they are readily detected by intracellular recording only after 1–2 months of preganglionic denervation. After 3 months of extrinsic denervation two-thirds of the ganglion cells receive physiologically detectable postganglionic inputs, yet the efficacy of these synapses remains substantially less than that of native, preganglionic synapses. This reduction in efficacy is caused at least in part by the location of the postganglionic synapses; electron microscopy reveals them to be situated on the axons of ganglion cells, whereas preganglionic axons terminate primarilly on the cell bodies. The intrinsic synapses that form in the absence of extrinsic innervation are no longer detectable physiologically after ganglion cells are reinnervated by preganglionic axons. This suppression of connections between ganglion cells is nearly complete (90%) and is a consequence of competition with preganglionic axons: it does not occur in the absence of preganglionic reinnervation. The data indicate that the presence of preganglionic inputs to the ganglion cells discourages the formation and maintenance of synaptic connections between them.

Binding properties of a neurotoxin from the venom of the green mamba, Dendroaspis viridis.

A toxin, alpha-mambatoxin, was purified from the venom of the green mamba Dendroaspis viridis using the procedures of Shipolini et al. (Shipolini, R. A., Bailey, G. S., Edwardson, J. A., and Banks, E. C. (1973) Eur. J. Biochem. 40, 337-344). The purified toxin blocks agonist-induced activation of acetylcholine receptors on muscle cells but, like alpha-bungarotoxin, it does not affect agonist-induced activation of receptors on a clonal sympathetic nerve cell line (PC12), an endothelial cell line, cultured chick ciliary ganglion neurons, or frog cardiac ganglion neurons. The toxin does block binding of alpha-bungarotoxin to cultures of muscle and nerve, and iodinated alpha-mambatoxin binds to cultures of muscle and nerve. The alpha-mambatoxin binding component on muscle was identified as acetylcholine receptor on the basis of sedimentation, immunoprecipitation, and rate of degradation. The alpha-mambatoxin binding component on PC12 cells, like the alpha-bungarotoxin binding component on these cells, is not recognized by anti-acetylcholine receptor antisera which do recognize acetylcholine receptor on these cells. The number of alpha-mambatoxin binding sites on both nerve and muscle when assayed in situ is twice that of alpha-bungarotoxin binding sites. However, when muscle cells are solubilized in nonionic detergents and then labeled with toxins, the number of alpha-mambatoxin binding sites is reduced and the two toxins bind in equal molar amounts. Finally, unlike alpha-bungarotoxin, which dissociates from complexes formed with nerve, alpha-mambatoxin forms complexes with nerve which dissociate only very slowly if at all.

Loss of extrasynaptic acetylcholine sensitivity upon reinnervation of parasympathetic ganglion cells.

1. The extrasynaptic acetylcholine sensitivity of frog cardiac ganglion cells was measured both after denervation and during the early stages of reinnervation by preganglionic axons. Sensitivity was measured by ionophoretic application of acetylcholine (ACh) to randomly chosen sites on ganglion cell bodies. 2. Extrasynaptic sensitivity rose gradually following denervation and after 3 weeks reached a mean value of approximately 1000 mV/nC. 3. Reinnervation of the cardiac ganglion began about 3 weeks after nerve crush. The ACh sensitivity of ganglion cells fell markedly during the 23--31 day period, to a mean of 184 mV/nC. None of forty-three neurones studied during that period received synaptic inputs sufficient to generate action potentials. 4. Twenty-nine of the forty-three neurones examined 23--31 days after nerve crush had not yet received detectable synaptic inputs, yet even these cells had markedly reduced ACh sensitivity. 5. When reinnervation of cardiac ganglia was delayed by resecting the preganglionic nerves, ACh sensitivity was reduced slightly (43%) between 14--21 and 23--31 days after surgery. Thus, most of the sixfold reduction in sensitivity that occurs during this time after nerve crush is a specific effect of reinnervation. 6. We conclude that loss of extrasynaptic receptors coincides with, or may even precede, the earliest physiological signs of synapse formation. Restoration of action potential activity in the ganglion cells is not essential to initiate this loss.

Multiple innervation of normal and re‐innervated parasympathetic neurones in the frog cardiac ganglion.

1. Multiple innervation of parasympathetic neurones was examined in normal and re-innervated frog cardiac ganglia. The number of synaptic inputs impinging upon individual ganglion cells was determined by recording intracellularly and stimulating the vagosympathetic nerves. 2. In unoperated cardiac ganglia most neurones (93%) received a large, suprathreshold synaptic input. Some ganglion cells received additional, small synaptic inputs. Roughly equal numbers of cells encountered were singly and doubly innervated, and only 8% received more than two inputs. 3. Re-innervation of cardiac ganglion cells began three weeks after bilateral crush of the vagosympathetic nerves. By 7 weeks more than 90% of the ganglion cells were re-innervated. At this stage the pattern of multiple innervation was significantly different than normal: doubly innervated neurones outnumbered singly innervated ones, and 31% of the cells encountered received more than two inputs. This pattern was stable for at least a year. 4. These results indicate that polyneuronal innervation of cardiac ganglion cells is more widespread after re-innervation than it is normally and, furthermore, that synapse elimination does not occur during re-innervation of these cells.

Sprouting and regeneration of synaptic terminals in the frog cardiac ganglion

SYNAPTIC connections are capable of a remarkable proliferation after nervous tissue has been damaged and some synapses have been destroyed. This proliferation, or ‘sprouting’, of remaining, intact, synapses within partially denervated tissue has been found to occur in the peripheral, autonomic and central nervous systems1–5. As yet, however, the functional ability of these sprouted synaptic contacts has not been extensively tested and the persistence of the sprouted synapses over long periods has not been studied. We have been investigating these problems in the para-sympathetic cardiac ganglion in the frog. Earlier studies have shown that when the nerve supply to the ganglion is partially destroyed by crushing one vagus nerve, axons in the other vagus nerve sprout and synapse with the denervated neurones1. Here I describe experiments in which the persistence of the sprouted synapses in the cardiac ganglion was tested by allowing the damaged axons to regenerate into the tissue. The results indicate that regenerating vagal fibres are able to re-establish some, but not all, of their synaptic connections and that innervation from the sprouted nerve decreases.

Sprouting of synapses after partial denervation of frog cardiac ganglion

SYNAPTIC connections in the adult nervous system of several species can be quite flexible and this characteristic has been termed neuronal plasticity. For example, when lesions are made in the axonal pathways of the brains of mature rats, surviving nerve terminals can proliferate and innervate the synaptic sites vacated by the degenerating axons1,2. This “sprouting” occurs in the central nervous system2–4, autonomic nervous system5 and peripheral nervous system6, and seems to be a characteristic response of nervous tissue to partial denervation. In only a few studies has sprouting been reported not to occur7–9. Sprouting of synapses has important implications for the maintenance of neuronal connections and the ability of the nervous system to recover from focal lesions. But several questions need to be answered. For example, are sprouted synapses functional, and if so, do they transmit normally? Is there any specificity in or control over the proliferation of new connections? What initiates and terminates the sprouting? We can now begin to answer these questions. Our studies on the para-sympathetic cardiac ganglion of the frog have shown that synaptic sprouting occurs very rapidly after partial denervation. Furthermore, the number of boutons per cell remains constant throughout sprouting and is the same as the number per cell in control ganglia.