Porocytosis: a new approach to synaptic function.

We propose a new approach to address the question of how a single quantum of neurotransmitter is secreted from a presynaptic terminal whose clustered secretory vesicles are locally bathed in high levels of calcium ions [Proceedings of the Symposium on Bioelectrogenesis (1961) 297-309; The Physiology of Synapses (1964) Chapters 1, 4, 5, 6; How the Self Controls its Brain (1994) Chapters 1, 4, 5, 6; Science 256 (1992) 677-679]. This hypothesis, which we term 'porocytosis', posits that the post-synaptic quantal response results from transmitter secreted through an array of docked vesicle/secretory pore complexes. The transient increase in calcium ions, which results from the voltage activated calcium channels, stimulates the array of secretory pores to simultaneously flicker open to pulse transmitter. Porocytosis is consistent with the quantal nature of presynaptic secretion and transmission, and with available biochemical, morphological and physiological evidence. It explains the frequency dependency of quantal size as a function of the secretion process. It permits a signature amount of transmitter release for different frequencies allowing a given synapse to be employed in different behavioral responses. The porocytosis hypothesis permits fidelity of secretion and the seemingly apposed characteristic of synaptic plasticity. The dynamics inherent in an array insure a constant quantal size as a function of the number of units within the array. In this hypothesis, plasticity is a consequence of concurrent pre- and post-synaptic changes due to a change in array size. Changes in the number of docked vesicle-secretory pore complexes composing the array can explain facilitation, depletion, graded excitation-secretion and long term plasticity.

KCl stimulation and ultrastructural responses of tannic acid-stained cholinergic synaptic terminals.

The quantal-vesicular hypothesis equates miniature end-plate potentials (MEPPs) with fusions of synaptic vesicles. MEPP production thus predicts vesicle losses, increases in vesicle fusions and increases in terminal plasma membrane. MEPP production and these ultrastructural parameters have been evaluated in the cholinergic presynaptic terminals of skate electric organ following tannic acid saline incubation, known to promote capture and selective staining of dense-core granule fusions, and KCl stimulation, known to elevate MEPP production dramatically in these cholinergic terminals. After pretreatment in tannic acid-elasmobranch saline, KCl stimulation produced MEPPs at 40/s/microm(2)of terminal surface for several minutes with gradual reduction to spontaneous levels by 25-30 min. No loss of vesicles, no vesicle fusions, no expansions of plasma membrane and no tannic acid enhanced staining of vesicles or vacuoles accompanied the generation of 800 MEPPs/microm(3)of terminals having densities of 567 vesicles/microm(3). No ultrastructural footprints were found to support the notion that unnaturally high rates of vesicular exocytosis had occurred.

Transmitter quantal size in Torpedo electrocytes is determined by frequency of release.

Miniature end-plate potentials (MEPPs) were focally recorded from the cytoplasmic surface of electrocytes in isolated columns of the Torpedo electric organ. Double electrode studies showed that the junctional area was restricted to 12 micron2. MEPP frequencies ranging from 1/min to 400/s were controlled with electrode advancement against the cytoplasmic surface. Stable membrane potentials and noise levels indicated constant intracellular, focal recording conditions. Focal MEPPs are only 1-3 mV and MEPP amplitudes smoothly decreased with an increase in MEPP frequency which demonstrates a process that meters quantal size at moment of release. Thus, release if not from a prepackaged store. MEPP interval analyses showed that events are weakly interactive at low frequencies and periodic at higher frequencies. The interdependency of MEPP amplitudes and intervals indicates that the mechanism of release controls both rate and quantal size. We propose that the amplitude and frequency dependencies of MEPPs at the Torpedo nerve-electrocyte junction are best described by a membrane channel (e.g., mediatophore, Israël and Dunant, Neurochem. Int. 28 (1996) 1-9) that meters transmitter from a presynaptic store.

Dynamic responses of presynaptic terminal membrane pools following KCl and sucrose stimulation.

The cholinergic presynaptic terminals of Torpedo electric organ have been examined morphometrically following stimulation by KCI and sucrose. The objective was to confirm correlations predicted by the vesicle hypothesis between miniature end-plate potentials (MEPPs) and morphometric changes in terminal ultrastructure. Both secretegogues generated high frequencies of MEPPs and also distinctive though differing ultrastructural changes. The synaptic vesicles show classes of 68 and 90 nm diameters and both store acetylcholine (ACh). KCl stimulation depleted the 90 nm class first whereas sucrose reversed the order of depletion. Very few instances of actual vesicle fusion were seen. Dose-response correlations between vesicle density and secretegogue strength (mM) and duration were higher with sucrose. Both secretegogues produced declines in vesicle numbers and densities and yielded multimodal distributions of large vesicles with an average 160 nm mean diameter. No meaningful correlations were detected between numbers of MEPPs and vesicles and little evidence was found to indicate that vesicles were fusing to terminal plasma membrane in numbers approximating MEPP release. Linear regression analysis was used to quantitatively examine relationships between the vesicle membrane pool and other pools of the putative exo/endocytotic pathway. Correlation coefficients between vesicle and terminal plasma membrane pools were non-significant and of positive sign, indicating independent, similar responses. Non-significant, negative coefficients were obtained when vacuole and 160 nm vesicle membrane values were included. These tests further argue against claims that vesicles are actively fusing with the plasma membrane. These conflicting findings for both secretegogues preclude meaningful correlations between vesicle changes and numbers of MEPPs generated and again emphasize the difficulty of validating the vesicle hypothesis by ultrastructural means. On the other hand, the study shows that vesicular, vacuolar and terminal membrane pools are dynamically changing during transmitter release, presumably interacting with cytosolic membrane constituents. A dynamical release process therefore has been proposed to account for the two classes of MEPPs, the rapid changes in class ratio and the mutable characteristics of the bell-MEPP that presently challenge the quantal-vesicular claims of prepackaged, immutable, exocytotically released packets of transmitter. This model features a state for each MEPP class with class and size determined at moment of release. For example, a single flicker of a channel would generate the sub-MEPP (defined subunit of an MEPP) and 7-20 flickering channels would generate the bell-MEPP.

A morphometric analysis of exocytosis in KCl-stimulated bovine chromaffin cells.

Transmission electron microscopy has been used to morphometrically evaluate exocytosis in bovine adrenal medulla chromaffin cells as the mechanism of catecholamine release. Purified cell suspensions were stimulated with KCl at varying strengths and durations and then conventionally processed for ultrastructural analysis. Quantitation of exocytotic images of dense cored chromaffin granules was a major objective and such images were found in all preparations, attesting to the efficacy of chemical fixation to preserve this event. However, because hundreds of cell profiles had to be screened to find a single granule in the process of release this low frequency precluded any meaningful correlations with estimates of granular involvement based on catecholamine release. Neither KCl molarity nor duration altered this finding nor did these variables significantly affect other parameters linked to exocytotic activity. For example, cell size and numbers of "empty' granules and vesicles remained constant and attempts to label "any' organelle with 30-nm colloidal gold or lanthanum precipitate proved unsuccessful. In short, if exocytosis is responsible for release, it would appear to function without leaving a morphological trace. An alternative hypothesis, therefore, is outlined which better accommodates existing data.

Structural hierarchy in the clustering of HLA class I molecules in the plasma membrane of human lymphoblastoid cells.

Major histocompatibility complex (MHC) class I antigens in the plasma membranes of human T (HUT-102B2) and B (JY) lymphoma cells were probed by immunochemical reagents using fluorescence, transmission electron, and scanning force microscopies. Fluorescent labels were attached to monoclonal antibodies W6/32 or KE-2 directed against the heavy chain of HLA class I (A, B, C) and L368 or HB28 against the beta 2-microglobulin light chain. The topological distribution in the nanometer range was studied by photobleaching fluorescence resonance energy transfer (pbFRET) on single cells. A nonrandom codistribution pattern of MHC class I molecules was observed over distances of 2-10 nm. A second, nonrandom, and larger-scale topological organization of the MHC class I antigens was detected by indirect immunogold labeling and imaging by transmission electron microscopy (TEM) and scanning force microscopy (SFM). Although some differences in antigen distribution between the B- and T-cell lines were detected by pbFRET, both cell lines exhibited similar clustering patterns by TEM and SFM. Such defined molecular distributions on the surfaces of cells of the immune system may reflect an underlying specialization of membrane lipid domains and fulfill important functional roles in cell-cell contacts and signal transduction.

Dynamic responses of presynaptic terminal membrane pools to electrical stimulation.

The anatomical tenets of the quantal-vesicular hypothesis of neurotransmission are a 1:1 ratio between numbers of releasable quanta and vesicles, a reciprocal response between vesicle and terminal membrane pools and constancy of the total membrane pool. We have used electrical stimulation and morphometry to study these relationships in the cholinergic presynaptic terminals of Torpedo electric organ. Our results show that during neurotransmission changes in vesicle numbers do not correlate with quantal release, vesicle and terminal membranes do not change in reciprocal fashion and total nerve terminal membrane does not remain constant. We conclude that these vesicular tenets of quantal release are not verifiable at the Torpedo electric organ junction.

A morphometric analysis of isolated Torpedo electric organ synaptic vesicles following stimulation.

The electric organ of Torpedo has been stimulated with 1800 pulses at 0.1 Hz to produce biochemical and morphological heterogeneity of its synaptic vesicle population. This was verified by biochemical and morphometric analyses of the synaptic vesicle population isolated by sucrose density gradient zonal separation following stimulation. Biochemical or metabolic heterogeneity was verified using 2 established criteria: the appearance of a second peak of acetylcholine (ACh) in denser fractions of the zonal gradient and a corresponding overlapping peak of incorporated radiolabelled ACh. Morphologic heterogeneity was deduced by the presence in this second peak of a subclass of synaptic vesicles having a mean diameter of 68 nm i.e., a diameter 20-25% smaller than the 90 nm subclass that represents the most prominent subclass of the intact terminal population. Despite having satisfied these 3 criteria, functionally relevant heterogeneity cannot be assumed. One reason is due to our failure to recover the 90 nm subclass of vesicle which provides the physical basis to explain the 2 ACh peaks along the gradient. Because of this, the point is raised whether the stimulation-induced ACh peak is not merely an artifact due to inadequate sampling. On the other hand, radioactive labelling of the ACh pool provides a more convincing demonstration of the existence of 2 metabolically different subclasses. We conclude that morphological heterogeneity of the ACh vesicle population has never been established and that metabolic heterogeneity, as it has been studied to date, pertains to a single-sized subclass population of vesicles measuring 68 nm in diameter.

Morphological, physiological and biochemical observations on skate electric organ.

The electric organs of two species of skate have been examined morphologically, physiologically and biochemically. They can be easily dissociated into innervated or denervated component electrocytes by a Torpedo Ringer's solution containing 1% collagenase. Collagenase treatment did not, however, separate the Schwann cell cover capping the synaptosomes. Isolated electrocytes generate normal MEPP frequencies and show evoked responses for two days in Torpedo Ringer's. The nerve terminals retain excitability and transmitter release properties up to the time of separation. Since isolated terminals and denervated electrocytes show normal ultrastructural characteristics for up to 12 h, the skate electric organ provides several preparations which are not attainable with Torpedo tissue. Acetylcholine (ACh) content of supernatant fractions containing the synaptosomes was comparable to that found in Torpedo (sps.). Collagenase specifically eliminates the basal lamina associated with the synaptic junctional region. Neuronal cell death and synaptic terminal degeneration were also noted in the adult organs of both species. The skate electric organ is ideally suited for the study of cholinergic development and transmission.

A morphometric analysis of Torpedo synaptic vesicles isolated by iso-osmotic sucrose gradient separation.

The presynaptic terminal vesicle population of Torpedo electric organ is heterogeneous in size, consisting of two prominent subpopulations that comprise 80% of the total. The use of standard iso-osmotic sucrose gradients with zonal centrifugation to isolate vesicle fractions that co-localize with the acetylcholine (ACh) peak results in the recovery of: (1) 10% of the total estimated vesicle population; and (2) a single 68-nm diameter vesicle size class. The whereabouts of the major 90-nm subclass, which accounts for 60% of the total terminal population and which has long been considered to represent the resident ACh population, has been investigated. Assuming this subclass to have undergone severe osmotic stress, the effects of hypo- and hyper-osmotic salines, buffers and fixatives were examined and found to produce only negligible changes on vesicle size. Isolation of vesicles by hypo-osmotic shocking of synaptosomes purified on a Ficoll gradient, however, resulted in a reasonable approximation of the in situ distribution. As the iso-osmotic sucrose gradient procedure utilizes frozen blocks of electric tissue, this step is suspected of being involved in the loss, perhaps because of the slow freezing rates employed. These findings indicate that the 90 nm subclass is lost rather than transformed during isolation by sucrose gradient separation and that dimensionally, the cholinergic vesicle is a constant-sized and relatively stable structure.

A morphometric analysis of synaptic vesicle distributions.

Synaptic vesicle populations have been morphometrically analyzed for size and density. Populations composed of a single size class of vesicles are represented by normal (Gaussian) or positive (log-normal) skew histograms. Populations with multiple size classes generate negative (left) skew distributions. Fixatives containing aldehydes differentially affect these distribution patterns but vesicles are able to withstand tonic effects over a wide range. Reader bias' contribute the most error in the data-collecting process. But despite this, the sizing of vesicle populations can be accomplished with great accuracy. Vesicle density computations, on the other hand, vary over a wide range and are of less value for comparative purposes.

Development of the electromotor system in Torpedo marmorata: cationic staining of the electric organ.

The electric organs of embryonic Torpedo marmorata have been reacted with three cationic stains to evaluate the appearance and distribution of anionic sites. Ruthenium red, alcian blue and lysozyme were used at different pHs and found to react in a time-related manner to anionic components within the interelectrocyte space. The basal lamina covering the ventral electrocyte surface possesses the greatest number of anionic sites whereas growth cone, presynaptic terminal and glial membranes displayed almost no staining. Since this lamina serves as the exclusive substrate for ingrowing neurites during synaptogenesis, the results are consistent with the idea that charge distribution on the membrane surface may provide a necessary cue for neurite motility, extension and eventual synaptogenesis.

Development of the electromotor system of Torpedo marmorata: distribution of extracellular matrix and cytoskeletal components during acetylcholine receptor focalization.

A combination of direct fluorescence and indirect immunofluorescence microscopy has been used to compare the distribution of the acetylcholine receptor with the distribution of major cytoskeletal and extracellular matrix components during electrocyte differentiation in the electric organs of Torpedo marmorata. Laminin, fibronectin and extracellular matrix proteoglycan are always more extensively distributed around the differentiating cell than the acetylcholine receptor-rich patch that forms on the ventral surface of the cell. The distribution of acetylcholinesterase within the ventral surface of the differentiating electrocyte closely resembles the distribution of the acetylcholine receptor. Areas of apparently high acetylcholine receptor density within the ventrally forming acetylcholine receptor-rich patch are always areas of apparently high extracellular matrix proteoglycan density but are not always areas of high laminin or fibronectin density. Desmin levels appear to increase at the onset of differentiation and desmin initially accumulates in the ventral pole of each myotube as it begins to form an electrocyte. During differentiation F-actin-positive filament bundles are observed that extend from the nuclei down to the ventrally forming acetylcholine receptor-rich patch. Most filament bundles terminate in the acetylcholine receptor-rich region of the cell membrane. Electron-microscopic autoradiography suggests that the filament bundles attach to the membrane at sites where small acetylcholine receptor clusters are found. The results of this study suggest that, out of the four extracellular matrix components studied, only the distribution of acetylcholinesterase (which may be both matrix- and membrane-bound at this stage) closely parallels that of the acetylcholine receptor, and that F-actin filament bundles terminate in a region of the cell that is becoming an area of high acetylcholine receptor density.

Torpedo electromotor system development: neuronal cell death and electric organ development in the fourth branchial arch.

The fourth branchial arch of Torpedo marmorata has been examined at the light and electron microscopic level during development. Of interest was the determination of the extent of electric organ tissue reported to be present in this arch and its possible relationship to electromotoneuron cell death in the electric lobes. The main electric organ of the torpedo is derived from the hyoid and first three branchial arches and is innervated by four major electromotor nerves. Extensive electromotoneuron cell death occurs in the electric lobes and most notably in the posterior poles. This feature could be due to a tendency for these neurons to innervate the fourth branchial arch where little or no electric tissue is formed. Our findings support this conclusion but are not entirely consistent with the idea that a population mismatch has occurred. This is because cell death precedes the genesis of the target cells. The presence of innervated differentiated electric tissue in this arch is also reported, leading to the conclusion that Torpedo marmorata possesses an accessory electric organ.

Torpedo electromotor system development: developmentally regulated neuronotrophic activities of electric organ tissue.

Explant cultures of electric lobe from 45-60 mm stage Torpedo embryos and both ganglionic and dissociated cell cultures prepared from 8-day chick ciliary ganglia have been used to determine whether the electric organs of Torpedo marmorata contain developmentally regulated neuronotrophic activity. Electric lobe explants were evaluated by measuring their neurone density, choline acetyltransferase (CAT0, and low salt, Triton X-100-soluble protein contents. Addition of soluble extracts prepared from the electric organs of late stage embryos (85-105 mm) to standard medium results in the maintenance of nearly theoretical neurone densities in electric lobe explants during a 7-day culture period. Soluble electric organ extracts from early embryonic stages (42-59 mm) do not increase neurone density relative to control cultures but cause an elevation in the CAT content of the explants over control values. On the basis of this analysis it is concluded (1) that late embryonic stage and adult electric organs contain neuronotrophic activity that allows electromotor neurones to survive in vitro and (2) that activity increases rapidly in the electric organs between the 59 nd 72 mm stages of development at a time when rapid increases in postsynaptic membrane markers in the electric organs occur and when peripheral synaptogenesis begins. The activity of late stage embryonic electric organs is heat stable and lost on dialysis. Using ciliary ganglion explants and evaluating both the initial fibre outgrowth and the CAT content after 4 days in vitro, trophic activity is found to be maximal at early embryonic stages (45-55 mm) and to decline thereafter. It is shown that the decline in activity is not due to an increase in toxicity. Using established dissociated ganglionic cell survival assays the specific activity of neuronotrophic factors allowing survival is constant between the 45 and 73 mm stages in the electric organs and then rapidly declines, but activity per electric organ increases rapidly between the 45 and 73 mm stages and then remains at a constant level. The use of poly-dl-ornithine substrates coated with heart-conditioned medium for the cell survival assay results in up to tenfold increase in the trophic titre of the electric organ extracts. The neuronotrophic activity supporting survival of ciliary motorneurones present in embryonic electric organs is heat labile and retained on dialysis. It is concluded that developing electric organs contain at least two neuronotrophic factors that have different properties and are differently regulated. Both factors may contribute during development to bringing naturally occurring electromotor neurone cell death to an end.

The localization and rate of disappearance of a synaptic vesicle antigen following denervation.

A proteoglycan-specific antiserum has been used to monitor the effects of denervation in the electric organ of Torpedo marmorata. The antiserum was produced by injecting a highly purified synaptic vesicle fraction prepared from the electric organs of Torpedo marmorata. Following absorption the serum appears to be specific towards synaptic vesicles. The ultrastructural localization of the antigen determined by immuno-electron microscopy confirmed the specificity of the antiserum and showed that it did not cross-react with the proteoglycans of the basal lamina. The rate of disappearance of the vesicle proteoglycans following denervation was evaluated by means of the antiserum and was compared to the rate of disappearance of other vesicular and nerve terminal-associated markers. The results suggest that degeneration affects the vesicular constituents at varying rates resulting in a progressive disappearance of the entire functional capacity of the synaptic vesicles.

Investigations into a bioelectric component of synaptogenesis.

Synaptogenesis has been investigated in the electric organ of Torpedo marmorata with the objective of determining whether a bioelectric effect could be demonstrated. Answers to 3 questions were sought. (1) Are currents and/or fields present within the organ? (2) Can they be localized? (3) Are they involved with the synaptogenic process? Voltage measurements across pieces of electric organ revealed the presence of a dorsal positive potential in the low millivolt range. Injection of DC current against this dorsal positive dipole had the effect of reducing the percent of neuritic coverage on the ventral surface as measured by quantitative electron microscopy. These results indicate the presence of a field potential, dorsal positive which, when reversed, causes a retardation in the synaptogenic rate. They are consistent with published reports of neurites growing preferentially towards cathodal sources and implicate that bioelectric forces may be one component of the synaptogenesis process.

Torpedo electromotor system development: a quantitative analysis of synaptogenesis.

Synaptogenesis in the electric organ of Torpedo marmorato has been studied quantitatively at the ultrastructural level of observation. In addition to establishing the normal developmental time course for this event we were interested in determining whether a gradient of synaptogenesis might be present because the electric organ produces several morphologically recognizable spatiotemporal gradients during its early ontogeny. These gradients genesis of electrocyte columns, both gradients of which are operative for periods of weeks. No gradient of synaptogenesis was found, indicating this to be a synchronous process. The idea is advanced that synaptogenesis in the electric organ is modulated by extrinsic influences, many of which may originate from the target electrocytes which, by this time, have become synchronized in their development.

The developmental morphology of Torpedo marmorata: the electric lobes.

The development of the electric lobes of Torpedo marmorata has been investigated using light and electron microscopical techniques. The lobe Anlagen become visible in the rhombencephalon along the floor of the 4th ventricle at the 10-mm stage. Many of the neuroepithelial cells in the Anlagen differentiate, Becoming postmitotic and axonic by the 24 mm stage. Proliferative zones of neuroepithelial cells disappear from the electric lobes by the 30-mm stage. After their initial, early differentiation the electromotor neurons remain monopolar until the 40-mm stage when dendrite formation begins. The differentiation of the electromotor neuron from a mono- to an immature multi polar form occurs between the 40- and 55-mm stages and involves, in addition to dendrite formation, a change from a pear-shaped to a spherical cell body, a dramatic increase in cytoplasmic volume, a centralization of the nucleus, an enlargement of the nucleolus and its migration away from the nuclear membrane, and differentiation of the axon hillock. The electric lobes are invaded by sinusoids at the 24-mm stage but formation of the capillary network by sprouting cords of endothelial cells begins later at the 40-mm stage. Neuronal cell death (26-74-mm stages) appears to be mainly an autolytic process and the debris is removed by immature glial cells. Afferent fiber growth cones are first recognized in the lobes at the 60-mm stage but synapses are not observed until the 78-mm stage. Myelination begins in the electric lobes concomitantly with the onset of synaptogenesis. A twofold increase in dendrite length occurs over the period when synapses begin to form in the lobes but dendritic maturation is not complete until the neonatal (120-mm) stage. The results are discussed in relation to the development of the electric organs.