Increase of large intramembranous particles in the presynaptic active zone after administration of 4-aminopyridine

The History of the Synapse

Why did I choose this particular topic for my lecture rather than the history of neuroscience or the history of the neuron? Simply because I believe that every disciple has the obligation to pay homage to their mentors once in their lifetime. My formation as a neuroscientist involved three such mentors spanned across three countries. The first was Spain, where I was born, completed my medical studies, and had my first glimpse of neuroscience at the Cajal Institute with Fernando de Castro. It was him who, in 1961, advised me to spend some time abroad, and to that purpose he obtained me a scholarship from the French government, that allowed me to settle in Paris. Once in France I had the good fortune to meet Prof. René Couteaux, another generous mentor, who took care of my stay in the country. Two years later, he made me a proposition to which I could only answer in the affirmative by offering me a research position in France. I got married (the best thing that happened in my life), and spent the next 57 years working on the cerebellum. The third person I want to honor and remember in this presentation is Sanford Louis Palay who was my postdoc professor during the 2 years I worked at Harvard Medical School in Boston. And as it turns out, all three of my mentors have made positive contributions to the history of the synapse. So, without further delay, let us dive in. Anat Rec, 303:1252-1279, 2020. © 2020 American Association for Anatomy.

Differential distribution of release-related proteins in the hippocampal CA3 area as revealed by freeze-fracture replica labeling

Synaptic vesicle release occurs at a specialized membrane domain known as the presynaptic active zone (AZ). Several membrane proteins are involved in the vesicle release processes such as docking, priming, and exocytotic fusion. Cytomatrix at the active zone (CAZ) proteins are structural components of the AZ and are highly concentrated in it. Localization of other release-related proteins including target soluble N-ethylmaleimide-sensitive-factor attachment protein receptor (t-SNARE) proteins, however, has not been well demonstrated in the AZ. Here, we used sodium dodecyl sulfate-digested freeze-fracture replica labeling (SDS-FRL) to analyze quantitatively the distribution of CAZ and t-SNARE proteins in the hippocampal CA3 area. The AZ in replicated membrane was identified by immunolabeling for CAZ proteins (CAZ-associated structural protein [CAST] and Bassoon). Clusters of immunogold particles for these proteins were found on the P-face of presynaptic terminals of the mossy fiber and associational/commissural (A/C) fiber. Co-labeling with CAST revealed distribution of the t-SNARE proteins syntaxin and synaptosomal-associated protein of 25 kDa (SNAP-25) in the AZ as well as in the extrasynaptic membrane surrounding the AZ (SZ). Quantitative analysis demonstrated that the density of immunoparticles for CAST in the AZ was more than 100 times higher than in the SZ, whereas that for syntaxin and SNAP-25 was not significantly different between the AZ and SZ in both the A/C and mossy fiber terminals. These results support the involvement of the t-SNARE proteins in exocytotic fusion in the AZ and the role of CAST in specialization of the membrane domain for the AZ.

Morphological changes related to reconstituted acetylcholine release in a release-deficient cell line

The membrane changes accompanying Ca(2+)-dependent acetylcholine release were investigated by comparing release-competent and release-incompetent clones of mouse neuroblastoma N18TG-2 cells. No release could be elicited in native N18 cells or in a N18-choline acetyltransferase clone in which acetylcholine synthesis was induced by transfection with the gene for rat choline acetyltransferase. However, acetylcholine release was operative in a To/9 clone which was co-transfected with complementary DNAs from rat choline acetyltransferase and Torpedo mediatophore 16,000 mol. wt subunit. In thin sections, the aspect of resting N18 and To/9 cells was identical: a very dense cytoplasm with practically no vesicle-like organelles. Cells were chemically fixed at different times during a stimulation using A-23187 and Ca2+, and examined following both freeze-fracture and thin section. Stimulation of To/9 cells induced a marked change affecting the intramembrane particles. The number of medium-sized particles (9.9-12.38 nm) increased, while that of the small particles decreased. This change was not observed in control, release-incompetent cell lines. In the To/9 clone (but not in control clones), this was followed by occurrence of a large new population of pits which initially had a large diameter, but subsequently became smaller as their number decreased. Coated depressions and invaginations became abundant after stimulation, suggesting an endocytosis process. By considering the succession of events and by comparison with data from experiments performed on synapses in situ, it is proposed that a particle alteration was the counterpart of acetylcholine release in co-transfected To/9 cells; this was followed by a massive endocytosis.

Geometry, kinetics and plasticity of release and clearance of ATP and noradrenaline as sympathetic cotransmitters: roles for the neurogenic contraction

The paper compares the microphysiology of sympathetic neuromuscular transmission in three model preparations: the guinea-pig and mouse vas deferens and rat tail artery. The first section describes the quantal release of ATP and noradrenaline from individual sites. The data are proposed to support a string model in which: (i) most sites (> or = 99%) ignore the nerve impulse and a few (< or = 1%) release a single quantum of ATP and noradrenaline; (ii) the probability of monoquantal release is extremely non-uniform; (iii) high probability varicosities form 'active' strings; and (iv) an impulse train causes repeated quantal release from these sites. Analogy with molecular mechanisms regulating transmitter exocytosis in other systems is proposed to imply that coincidence of at least two factors at the active zone, Ca2+ and specific cytosolic protein(s), may be required to remove a 'fusion clamp', form a 'fusion complex' and trigger exocytosis of a sympathetic transmitter quantum, and that the availability of these proteins may regulate the release probability. The second section shows that clearance of noradrenaline in rat tail artery is basically > or = 30-fold slower than of co-released ATP, and that saturation of local reuptake and binding to local buffering sites maintain the noradrenaline concentration at the receptors, in spite of a profound decline in per pulse release during high frequency trains. The third section describes differences in the strategies by which mouse vas deferens and rat tail artery use ATP and noradrenaline to trigger and maintain the neurogenic contraction.

Spatiotemporal pattern of quantal release of ATP and noradrenaline from sympathetic nerves: consequences for neuromuscular transmission

The recent explosive development in research concerning the fundamental mechanisms of synaptic transmission helps put the present paper in context. It is now evident that not all transmitter vesicles in a nerve terminal, not even all those docked at its active zones, are immediately available for release (36). We watch, fascinated, the unraveling of the amazingly complex cellular mechanisms and molecular machinery that determine whether or not a vesicle is "exocytosis-competent" (77,78,39,79). Studies on quantal release in different systems show that neurons are fundamentally similar in one respect: that transmitter release from individual active zones is monoquantal (2). But they also show that active zones in different neurons differ drastically in the probability of monoquantal release and in the number of quanta immediately available for release (3). This implies that one should not extrapolate directly from transmitter release in one set of presynaptic terminals (e.g., in neuromuscular endplate or squid giant synapse) to that in other nerve terminals, especially if they have a very different morphology. As shown here, one should not even extrapolate from transmitter release in sympathetic nerves in one tissue (e.g., rat tail artery) to that in other tissues or species (e.g., mouse vas deferens). It is noteworthy that most studies of quantal release are based on electrophysiological analysis and therefore deal with release of fast, ionotropic transmitters from small synaptic vesicles at the active zones, especially in neurons in which these events may be examined with high resolution (49,48,46,33,32). Such data are useful as general models of the release of both fast and slow transmitters from small synaptic vesicles at active zones in other systems, provided that these transmitters are released in parallel, as are apparently ATP and NA in sympathetic nerves. They tell us little or nothing, however, about the release of transmitters (e.g., neuropeptides) from the large vesicles, nor about the spatiotemporal pattern of monoquantal release from small synaptic vesicles in the many neurons that have boutons-en-passent terminals. They show that the time course of effector responses to fast, rapidly inactivated transmitters such as ACh or ATP is necessarily release related. But they do not even address the possibility that the effector responses to slow transmitters such as NA, co-released from the same terminals, may obey completely different rules and perhaps rather be clearance related (7).(ABSTRACT TRUNCATED AT 400 WORDS)

K+ and Ca2+ channel blockers may enhance or depress sympathetic transmitter release via a Ca(2+)-dependent mechanism "upstream" of the release site

Extracellular recording of the pre- and postjunctional electrical activity in guinea-pig or mouse vas deferens or rat tail artery was employed to study the mechanisms by which the K+ channel blockers, tetraethylammonium and 4-aminopyridine and the Ca2+ channel blockers, Cd2+, Mn2+ or nifedipine influence the nerve stimulation-induced release of adenosine 5'-triphosphate as a sympathetic co-transmitter. The K+ and Ca2+ channel blocking agents examined had no effect on the spontaneous quantal release of adenosine 5'-triphosphate. However, addition of tetraethylammonium and 4-aminopyridine inside the recording electrode broadened the nerve terminal action potential and caused it to become more resistant to local application of tetrodotoxin, and dramatically increased the magnitude and tetrodotoxin resistance of adenosine 5'-triphosphate release within the patch. Surprisingly, tetraethylammonium and 4-aminopyridine were equally effective when added outside the recording electrode; now they did not increase the duration of the nerve terminal action potential inside the patch but increased its resistance to locally applied tetrodotoxin and dramatically increased the magnitude as well as the tetrodotoxin resistance of adenosine 5'-triphosphate release from sites inside the patch. Both tetraethylammonium and 4-aminopyridine contributed to these effects, with a strong potentiating interaction. Nifedipine was without effect, but application of 1-100 microM Cd2+ or 1-5 mM Mn2+ either inside or outside the recording electrode blocked adenosine 5'-triphosphate release inside the patch. The results indicate: (i) that the nerve terminal action potential is generated by activation of voltage-gated, regenerative Na+ channels but also has a small component carried by influx of Ca2+ and that it is "normally" terminated by activation of voltage- as well as Ca(2+)-dependent K+ channels; (ii) that the release probability is tonically depressed by the resting K+ efflux, and promoted by the resting Ca2+ influx, "upstream" of the release sites; and (iii) that the upstream control of the release probability may involve both changes in properties of ionic channels in the nerve terminal membrane, and effects on the cytoskeleton leading to changes in the availability of releasable quanta in varicosities within the patch.

4-Aminopyridine induces the release of neuropeptide Y (NPY) to produce an atropine- and tetrodotoxin-resistant contraction in rabbit isolated jejunum

4-Aminopyridine (4-AP) induced an atropine- and tetrodotoxin (TTX)-insensitive contraction (resistant contraction), in a concentration-dependent manner, in the isolated jejunum of rabbits. The failure of specific antagonists of histamine, serotonin and substance P to affect this resistant contraction ruled out the participation of histamine, serotonin and/or substance P. Antiserum against neuropeptide Y (NPY) reduced this resistant contraction in a concentration-dependent manner and inhibited the action of 4-AP totally at a high concentration (1.25% dilution) whereas normal serum lacked this ability. This suggested that the release of NPY was involved in this 4-AP-induced resistant contraction. Radioimmunoassay of NPY-like immunoreactivity in isolated synaptosomal preparations indicated that 4-AP possessed the ability to induce the release of NPY. However, guanethidine did not affect the actions of 4-AP, indicating that NPY is released mainly from non-adrenergic nerves. Our results indicate that 4-AP induces the release of NPY from non-adrenergic nerves to produce an atropine- and TTX-resistant contraction in the isolated jejunum of rabbits.

Small vesicle bouton synapses on the distal half of the lateral dendrite of the goldfish Mauthner cell: freeze-fracture and thin section study

To understand principles of synaptic integration, it is necessary to define the types of synapses on a particular neuron and their distribution. Thin sectioning and double replica freeze-fracture techniques were employed to characterize the small vesicle bouton (SVB) synapses on the distal half of the Mauthner (M) cell lateral dendrite, which probably mediate a remote dendritic inhibition. Three morphologically distinct SVB synapses, types A, B, and C, were found. These three SVB synapses form roughly 90% of the synapses on the distal half of the lateral dendrite, with types A and B being most common. The SVB A synapse is characterized by mostly oval and round synaptic vesicles, a discrete presynaptic active zone with a highly variable shape, and a postsynaptic active zone with no apparent particle aggregate in either the E or P face. At the SVB B synapse, most of the synaptic vesicles are flat. A very high particle density is present throughout the presynaptic P face, and vesicle attachment sites are dispersed over much of the presynaptic membrane. Postsynaptic P face particle aggregates are subjacent to the presynaptic vesicle attachment sites, and are often large and anastomosing. The SVB C synapse is characterized by synaptic vesicle profiles that vary from flattened to round. The SVB C cytoplasm was unclouded by the flocculent material that characterized SVBs A and B. The presynaptic active zones at the SVB C synapse are discrete, and macular or oblong. No particle aggregates are apparent in the postsynaptic active zone. Small, macular particle aggregates were found in nonactive zone regions of the postsynaptic E face of all three types of SVBs. Small subsurface cisterns were also observed underlying the M cell membrane at all three types of SVB synapses. Neither the postsynaptic E face aggregates nor the subsurface cisterns were ever observed directly subjacent to presynaptic active zones, but were often seen adjacent to active zones. Short, straight rows of particles and short cylinders were often seen in both pre- and postsynaptic surrounding zone regions of SVB A and C synapses. These structures are thought to represent tight junctions.

Presynaptic effects of 4-aminopyridine and changes following a single giant impulse at the Torpedo nerve-electroplaque junction

The presynaptic changes caused by 4-aminopyridine were studied in the electric organ of Torpedo marmorata, in the resting state and during the period following transmission of a single giant discharge. Incubation with 4-aminopyridine provoked a 30-40% decrease in the density of synaptic vesicles in nerve terminals, and a similar decrease in the content of vesicular and free acetylcholine. These changes were not observed when 4-aminopyridine was applied in a low-calcium, high-magnesium solution. In the standard medium, 4-aminopyridine treated junctions were able to generate a giant electrical discharge of long duration in response to a single stimulus. During the seconds and minutes following the giant discharge, the number of synaptic vesicles was not found to be significantly altered in the whole population of nerve terminals. However, new membranous structures--looking like sacs with double membranes encircling a part of cytoplasm--were seen in approximately 25% of nerve endings; in those terminals, the number of synaptic vesicles was significantly decreased. At this stage, the junctions had not recovered their capability to generate a second giant discharge of full size and the yield of acetylcholine, adenosine 5'-triphosphate (ATP) and creatine phosphate was diminished. Thirty minutes after the single discharge, the functional recovery was achieved and the membranous sacs had disappeared; but the levels of acetylcholine, ATP and creatine phosphate were still not restored.

Brief occurrence of a population of presynaptic intramembrane particles coincides with transmission of a nerve impulse

Small pieces of Torpedo electric organ were cryofixed at 1-ms time intervals in a liquid medium at -190 degrees C before, during, and after the passage of a single nerve impulse. In contrast to studies using this or other preparations, these experiments were done without 4-aminopyridine or other drugs that potentiate transmitter release. Freeze-fracture replicas were made from the most superficial layers of the tissue, where the rate of cooling was rapid enough to retain ultrastructure in the absence of chemical fixation. We found that the transmission of an impulse was accompanied by the momentary appearance of a population of large intramembrane particles in both the protoplasmic (P) and the external (E) leaflets of the presynaptic plasma membrane. The change was very brief, appearing soon after the stimulus artifact. It lasted for 2-3 ms. Large pits denoting vesicle openings at the presynaptic membrane were found in a small proportion of nerve terminals; their number did not increase during transmission of the nerve impulse. Reducing the temperature from 16 to 5 degrees C slowed the time course of both the electrophysiological response and the change in intramembrane particles. The number of large particles did not increase when stimulation was applied in a low-Ca medium, a condition where the nerve terminals were still depolarized by the action potential but did not release the neurotransmitter. From these and other observations, we conclude that this transient change of intramembrane particles is closely linked to the mechanism of acetylcholine release at the nerve-electroplaque junction.

Increase in the number of presynaptic large intramembrane particles during synaptic transmission at the Torpedo nerve-electroplaque junction

Small pieces of Torpedo electric organ were treated with 4-aminopyridine, a drug which greatly increases the duration of transmitter release in a single nerve impulse, transforming the normally brief electroplaque potential to a giant discharge. Specimens of tissue were cryofixed by rapid freezing using liquid coolants at precise time intervals during transmission of a single giant discharge, and then examined by freeze fracture. In each experiment, we monitored the electrical response of one specimen during the freezing run to check the physiological responsiveness of the tissue and to determine the precise time of contact with the cryogenic liquid. The general appearance of nerve terminals after cryofixation was similar to that of terminals from chemically fixed and cryoprotected tissue. The major morphological change observed during the time course of the giant discharge was a marked increase in the density of intramembrane particles larger than 10 nm on both the protoplasmic and external faces of the presynaptic membrane. This change appeared in specimens frozen within the first few milliseconds after the stimulus, that is, at a time corresponding to the onset of the rising phase of the potential (3 ms). At the end of the giant discharge, the particle density returned to control values with the same time course as the potential trace. Pits of 20 nm or larger, probably due to vesicle-membrane interaction, were found in a small proportion of nerve terminals. Their occurrence increased only at 120-150 ms after the stimulus, that is, a long time after the beginning of the giant potential and of the change in intramembrane particles. The size distribution of particles was also determined in the membrane of synaptic vesicles exposed by cross fracture of terminal boutons; it was found to be similar to that of the unstimulated presynaptic membrane and it did not change during the giant discharge. Stimulation experiments were also carried out in a modified solution containing no added calcium, 20 mM magnesium and 4-aminopyridine. The propagation of impulses along the nerves to the electric organ was not inhibited in the modified solution but acetylcholine release was prevented and no increase in particle density was found on the presynaptic membrane. These and previous biochemical experiments on this tissue suggest that the release of the neuro-transmitter acetylcholine is associated with a transient occurrence of large intramembrane particles on the two fracture faces of the presynaptic membrane.(ABSTRACT TRUNCATED AT 400 WORDS)

Freeze-fracture analysis of synaptogenesis in glomeruli of mouse olfactory bulb

The relationship between intramembranous particle (IMP) aggregates appearing on the extracellular leaflet (E-face) of the postsynaptic membrane and postsynaptic densities was examined by electron microscopy during mouse olfactory bulb development. During prenatal development the IMP aggregates first increased in size and then decreased in size to the adult level, while the length of the postsynaptic densities tended to increase to a plateau. Concomitant with the size change, the shape of the IMP aggregates appeared to change during development from small, round clusters to large, anastomotic aggregations. Some of the IMP aggregates appeared to have a particle-free area in their centre. As development proceeded, the overall IMP density increased. The density of particles measuring 7-11 nm remained unchanged throughout prenatal life and decreased in the adult. These particles may be involved in stabilization of initial contacts and maintenance of mature synapses, rather than representing receptors or ion channels which would be expected to increase during development. The density of particles smaller than 7 nm increased prenatally, decreased at birth, then increased in the adult. These particles may represent two or more different macromolecules, one important in synaptogenesis, the other important in adult synapses.

Reconstitution of a functional synaptosomal membrane possessing the protein constituents involved in acetylcholine translocation

Reconstitution of a functional presynaptic membrane possessing calcium-dependent acetylcholine release properties has been achieved. The proteoliposomal membrane obtained gains its acetylcholine-releasing capabilities from presynaptic membrane proteins. At the peak of acetylcholine release, intramembrane particles became more numerous in one of the proteoliposomal membrane faces. This phenomenon resembles the intramembrane particle rearrangements found in stimulated synaptosomes. No visible structures capable of releasing acetylcholine as a result of the calcium influx were found inside the proteoliposomes. This supports the view that the release of free cytosolic acetylcholine from stimulated nerve terminals can be directly attributed to presynaptic membrane proteins. These proteins were extracted in a functional form from the synaptosomal membrane.

An ultrastructural study of presynaptic membrane specializations in sympathetic ganglia of 4-aminopyridine treated guinea pigs and rats

The presynaptic membranes of synapses in sympathetic ganglia of 4-aminopyridine treated guinea pigs and rats were investigated in either freeze-fractured or thin-sectioned material. After freeze-fracture many presynaptic membranes showed dimples in the P face and corresponding crater-like protrusions in the E face. In about 45% of the thin-sectioned synapses small clear omega-shaped invaginations were present at the active zone of the presynaptic membrane. Larger omega profiles with dense cores were sometimes also seen. These membrane features were very rarely observed in non-treated material.

Neuropathological alterations in the neocortex of rats subjected to focal aminopyridine seizures

Acute focal seizures were produced in anaesthetized albino rats by topical application of 3- and 4-aminopyridine on the exposed fronto-parietal neocortical areas. After 15 min of seizure activity neuronal and glial changes were studied by light (toluidine blue and Golgi staining) and electron microscopy. Shrinkage and increased electron density of some pyramidal cells, astrocytic swelling and depletion of 40-60 nm synaptic vesicles from the nerve terminals in layers I, II and III were found. The possible significance of the alterations in the seizure is discussed.

4-aminoquinoline-induced 'giant' miniature endplate potentials at mammalian neuromuscular junctions

4-Aminoquinoline (4-AQ) in concentrations around 200 micrometers induces, within minutes of its application to isolated mouse or rat neuromuscular junctions, the appearance of a population of miniature endplate potentials (m.e.p.ps) with a larger than normal amplitude, so-called giant m.e.p.ps (g.m.e.p.ps). With amplitudes 2-12 times the modal value of m.e.p.p. amplitude, the population of g.m.e.p.ps varied between 15 and 45% of the total population of m.e.p.ps. There was no increase in the frequency of m.e.p.ps but a positive correlation between the frequency of g.m.e.p.ps and the total frequency of m.e.p.ps. In many instances the rise time and decay time of g.m.e.p.ps were prolonged compared to normal. Elevated extracellular calcium concentrations increased the frequency of m.e.p.ps but had no effect on g.m.e.p.p. frequency. High extracellular potassium concentrations markedly increased m.e.p.p. frequency but failed to influence g.m.e.p.p. frequency. Similar observations were made with ethanol 0.1 M, ouabain 200 micrometers or black widow spider venom. Botulinum toxin type A markedly reduced total m.e.p.p. frequency but 4-AQ still induced g.m.e.p.ps. Nerve stimulation failed to release quanta corresponding to the g.m.e.p.ps. G.m.e.p.ps seemed to originate from quantal acetylcholine release from the nerve terminal since they were abolished by surgical denervation and by the addition of d-tubocurarine to the medium. Blockade of voltage-sensitive calcium or sodium channels by, respectively, manganese ions or tetrodotoxin failed to affect the appearance and the frequency of g.m.e.p.ps. The electrophysiological findings and a statistical analysis of the characteristics of the m.e.p.ps indicate that they belong to two populations. One population is accelerated by the depolarization-release coupling mechanism responsible for evoked transmitter release and is characterized by an amplitude distribution and a process in time that indicate that they correspond to releases occurring at 'active zones' in the nerve terminal. The second population of m.e.p.ps is uninfluenced by nerve terminal depolarization and transmembrane calcium fluxes. This population apparently originates from sites dispersed in the nerve terminal membrane and outside the 'active zones'. 4-AQ increases the frequency of this second m.e.p.p. population without affecting the first population.

The effect of 4-aminopyridine on the spinal locomotor rhythm induced by L-DOPA

In high spinal curarized cats rhythmic motor output similar to locomotion ('fictive locomotion') of all 4 limbs was obtained after intravenous application of the noradrenergic precursor, L-DOPA, and nialamide combined with 4-aminopyridine (4-AP). The activity was recorded from muscle nerves. In the presence of 4-AP, which enhances transmission at various excitatory and inhibitory synapses, reduced amounts of DOPA were sufficient to evoke fictive locomotion. 4-AP alone did not elicit locomotion. The burst rate increased up to 6 Hz with the amount of 4-AP given (0.5-50 mg/kg). The cyclic frequency of high spinal cats exhibiting either fictive locomotion or walking on a treadmill was accelerated by 4-AP. After a supplementary transection of the spinal cord at the upper lumbar level both fore- and hindlimbs generally continued to show fictive locomotion with similar frequencies. In the presence of high doses of clonidine (alpha-receptor-activator, greater than 4 mg/kg), the locomotor pattern was replaced by regular (2 Hz) synchronous discharges in all flexors and extensors.

Freeze-fracture study of the postsynaptic membrane of the cerebellar mossy fiber synapse in the frog

We have examined the postsynaptic membrane of the synaptic junctions of frog cerebellar mossy fibers by electron microscopy of freeze-fracture replicas and thin sections. The intramembranous particles (imps) in the E fracture face of the postsynaptic membrane are approximately 10 nm in size and form conspicuous aggregates which we classified as macular, annular, or anastomotic in form, according to the occurrence and placement of imp-free "windows" within the aggregate. The size and shape of the aggregates appear related in that the area of macular aggregates is consistently smaller than the area of annular or anastomotic aggregates. Measurements of aggregate area range from 0.06 to 0.75 micrometer2. The variable size and shape of the imp aggregate in the postsynaptic membrane sets it apart from other excitatory synapses in the central nervous system, where macular aggregates are usually described. Examination of serial thin sections suggests that the shape of the postsynaptic density is equivalent to that of the imp aggregate observed in the postsynaptic membrane by freeze-fracture. This supports the notion that the region of postsynaptic membrane associated with the postsynaptic density in thin sections corresponds to the particle-rich regions of E face membrane observed in freeze-fracture replicas.

Cell types and synaptic organization of the medullary electromotor nucleus in a constant frequency weakly electric fish, Sternarchus albifrons

The medullary electromotor nucleus (EMN) of Sternarchus albifrons was studied at the light and electron microscopic levels. The EMN consists of a dense meshwork of myelinated axons and glial elements with interposed large neurons; it is provided with an abundant supply of capillaries. Two types of essentially adrendritic nerve cells were distinguished on the basis of size: giant neurons (approx. 70 micrometers in diameter) and large neurons (approx. 30 micrometers in diameter). Their population ratio is 1:4. Only giant cells are labelled following the injection of retrograde tracer into the spinal cord; they are therefore identified with the so-called "relay cells" of other gymnotids. Tracer experiments further suggest that the descending axons of these relay cells give off collateral branches throughout the elongated spinal electromotor nucleus. In contrast, the large cells remain unlabelled and therefore lack spinal projections; they most likely correspond to "pacemaker cells." The perikaryal surface, including axon hillock and proximal part of initial segment of both types of EMN cells, is contacted by clusters of synaptic terminals and astrocytic processes. Two main varieties of synaptic terminals occur: (1) large endings and (2) ordinary end feet with standard size (S-type) and variable size (Sv-type) clear, spherical vesicles. The junction between large endings and EMN cells is characterized by the combination of gap junctions and surrounding intermediate junctions whose freeze-fracture characteristics were morphometrically analyzed. The large endings were formed by nodes of Ranvier as well as by fiber terminations, and synchronization within the EMN may be achieved by presynaptic fibers. Some of the contacts occur directly on the initial segment, which could allow activity to bypass the soma. It is concluded that the elctromotor system of Sternarchus is comprised of a rapid conduction pathway where medullary pacemaker and relay cells as well as spinal electromotor neurons are coupled by synapses with gap junctions. In contrast to the spinal electromotor neurons, the medullary EMN cells receive synapses with morphological characteristics of chemical transmission, and the S-type and SV-type terminals may possibly correspond to Gray's Type I and Type II synapses, respectively. These synapses may be involved in modulation of the electric organ discharge frequency.