Protein Kinase C Facilitation of Acetylcholine Release at the Rat Neuromuscular Junction

Protein kinase C (PKC) is a Ca²⁺-dependent enzyme involved in synaptic transmission, which can be experimentally activated by the phorbol ester, phorbol 12-myristate-13-acetate (TPA). We studied the effects of TPA application on acetylcholine (ACh) release at the rat neuromuscular junction by means of the focal recording technique; possible effects of TPA at the postsynaptic site had been ruled out in preliminary studies. In extracellular solutions containing 2 mM Ca²⁺ and at the stimulation frequency of 0.1 Hz, TPA increased endplate current (EPC) amplitude. In non-stimulated preparations spontaneous current frequency was increased at a similar rate. The similar time course of TPA action on evoked and spontaneous currents suggests that an increased presynaptic Ca²⁺ efficacy can be considered to be the probable mechanism of action. The interactions of PKC with ACh release were further investigated. In 0.1 mM Ca²⁺ extracellular solutions, TPA enhanced evoked currents only at stimulation frequencies (e.g. 40 Hz) that were themselves capable of inducing facilitation. This facilitation is classically associated with presynaptic Ca²⁺ accumulation, indicating that PKC interacts synergistically with Ca²⁺ to facilitate ACh release. In particular, since mean quantum size and release probability remained almost unchanged during TPA facilitation, it was concluded that PKC acted by enlarging the immediately available store. Interestingly, TPA also increased the presynaptic currents that were observed to be largely brought about by Ca²⁺-dependent K⁺ currents: evidence was obtained to suggest that increases in these currents provide negative feedback against excess release activation rather than being an expression of enhanced Ca²⁺ influx.     

Skin Sympathetic Response In Subclinical Diabetic Neuropathy

Hyperglycemia and its associated Na⁺/K⁺ pump activity has been implicated in the development of diabetic neuropathy. We recently reported that high glucose in the presence of ouabain induced a progressive increase in the delayed K⁺ current which was suppressed by a blocker of Ca²⁺-activated K⁺ channels and blockers of Ca²⁺ channels in rat single myelinated nerve fibers, suggesting an increase of cytosolic free Ca²⁺ concentration ([Ca²⁺]_i). However, the influences of high glucose with ouabain on [Ca²⁺]_i in sensory neurons remain to be elucidated. The present study was undertaken to examine the modulation of depolarization-induced Ca²⁺ transients by high glucose and ouabain in isolated adult rat dorsal root ganglion (DRG) neurons using the fluorescent Ca²⁺ indicator fura-2. Bath application of KCl (50 mM) evoked a rapid increase in [Ca²⁺]_i through voltage dependent Ca²⁺ channels ([Ca²⁺]_i: 154.2 ± 22.5 nM). This increase was enhanced under high glucose (30 mM D-glucose) in the presence of ouabain (100 M) ([Ca²⁺]_i: 764.8 ± 210.1 nM). We conclude that a combination of high glucose and decreased Na⁺/K⁺ pump activity leads to an increase in [Ca²⁺]_i in rat DRG neurons, thereby resulting in nerve dysfunction.     

Calcium-Dependent Regulation of Genetically Determined Spike and Waves by the Reticular Thalamic Nucleus of Rats

Summary: The pathophysiologic role of the reticular thalamic nucleus (RT) in rat generalized nonconvulsive epilepsy was investigated in the selected strain GAERS (genetic absence epilepsy rats from Strasbourg). After the RT was lesioned by the excitotoxic agent ibotenic acid stereotaxically injected in previously callosotomized rats, a disruption of ipsilateral spike and wave discharges (SWD) was observed in freely moving animals. In a second group of animals Cd²⁺ (0.5–1.5 μl, 1 m M ), which is known to block Ca²⁺ and Ca²⁺-dependent K⁺ conductances (gK⁺ (Ca²⁺), was injected into the thalamus. Cd²⁺reversibly suppressed ipsilateral SWD when injected in RT, whereas it slightly reduced SWD expression when injected in the ventrobasal (VB) complex. The difference was highly significant. We conclude that Ca²⁺-dependent oscillatory properties of the RT are critical for expression of genetically determined SWD in GAERS.     

The Effect of Barium on [3H]Noradrenalin Release from the Human Neuroblastoma SH-SY5Y

Replacement of Ca²⁺ with Ba²⁺ in HEPES-buffered saline stimulated [³H]noradrenalin release in the human neuroblastoma clone SH-SY5Y by up to 20% of the cell content in the absence of other secretory stimuli. The Ba²⁺-evoked release was inhibited by 85% by 3 μM tetrodotoxin and 95% by 5 μM nifedipine. Ba²⁺ also increased the potency of K⁺-evoked release of [³H]noradrenalin, as maximal release was observed with 60 mM K⁺ compared with the 100 mM K⁺ necessary to achieve maximal release in the presence of Ca²⁺. In contrast, replacing Ca²⁺ with Ba²⁺ had little effect on carbachol- and bradykinin-evoked release of [³H]noradrenalin. No evidence was obtained from studies on changes in [Ca²⁺]_i (in response to 100 pM carbachol) using fura-2 that Ba²⁺ could enter intracellular stores in SH-SY5Y cells. Whole-cell patch-clamp studies showed that Ba²⁺ depolarizes SH-SY5Y cells as well as enhancing inward Ca²⁺ channel currents and shifting their voltage dependence to more negative values. These results are discussed in terms of the hypothesis that Ba²⁺ blocks K⁺ channels, leading to depolarization followed by opening of voltage-sensitive Na⁺ channels. This in turn opens voltage-sensitive L-type Ca²⁺ channels, which are coupled to the release of [³H]noradrenalin in SH-SY5Y cells.     

Sodium, Calcium and Late Potassium Currents are Reduced in Cerebellar Granule Cells Cultured in the Presence of a Protein Complex Conferring Resistance to Excitatory Amino Acids

Whole-cell, patch-clamp recordings were used to study voltage-gated currents generated by cerebellar granule cells that were cultured in medium containing either 10% fetal calf serum (hereafter termed S + granules) or neurite outgrowth and adhesion complex (NOAC, hereafter called NOAC granules). NOAC is a protein complex found in rabbit serum that renders granules resistant to the excitotoxic action of excitatory amino acids. During depolarizing commands both S+ and NOAC granules generated Na⁺ and Ca²⁺ inward currents and an early and a late K⁺ outward currents. However, Na⁺ and Ca²⁺ Inward currents and late outward K⁺ currents recorded in NOAC granules were smaller than those seen in S+ granules. Furthermore, although of similar amplitude, early K⁺ currents displayed different kinetics in the two types of neurons. Thus, these data demonstrate that the electrophysiological properties of cerebellar granules, and probably of other neuronal populations, depend upon serum components and raise the possibility that an analogous modulation might be operative in vivo, and play a role in development, synaptic plasticity or neuropathological processes.     

Ca2+ Channel Expression in the Oligodendrocyte Lineage

The development of oligodendrocytes from their precursor cells through different developmental stages can be studied in vitro. These stages can be distinguished by specific monoclonal antibodies and by a characteristic K⁺ channel profile. In this study we demonstrate that the occurrence of Ca²⁺ currents also undergoes marked changes during the development of mouse oligodendrocytes. Immature precursor cells which can develop into astrocytes or oligodendrocytes expressed two different types of voltage-activated Ca²⁺ channels. The expression of Ca²⁺ channels in precursor cells was strongly correlated with the expression of Na⁺ channels. When cells started to express the O1 antigen and were committed to the oligodendrocyte lineage, Ca²⁺ and Na⁺ currents could no longer be detected. Large Ca²⁺ currents were, however, recorded later in the development of the oligodendrocytes, correlated with the expression of the O10 antigen. The Ca²⁺ channels were classified as high and low voltage-activated Ca²⁺ channels according to their range of activation, and are further described by their kinetic and pharmacological properties.     

Cholinergic Modulation of the Action Potential in Rat Hippocampal Neurons

The cholinergic input to the hippocampus from the medial septum is important for modulating hippocampal activity and functions, including theta rhythm and spatial learning. Neuromodulation by transmitters in central nervous system neurons usually affects cell excitability by modifying the membrane potential, discharge pattern and spike frequency. Here we describe another type of neuromodulation: changing the action potential waveform. During intracellular recordings from CA1 pyramidal cells in hippocampal slices from rats, the cholinergic agonist carbachol caused several reversible changes in the action potential: low doses (2 μM) caused an increase in spike duration; high doses (10–40 μM) or long-lasting applications also reduced the spike amplitude and rate of rise, and raised the spike threshold. These effects are similar to those of metabotropic glutamate receptor agonists or phorbol esters, both of which activate protein kinase C. The effects were blocked by the muscarinic antagonist atropine, and were prevented by Ca²⁺-free medium and by Ca²⁺-channel blockers. However, the cholinergic spike modulation was not occluded or mimicked by blocking the Ca²⁺-dependent K⁺ currents I_c or I_AHP, suggesting that these K⁺ currents are not involved in the modulation.     

Switching Between Two Types of Bursting Activity of Single Ca+2-Activated K+ Channels in Dissociated Neurons

Single channel recordings of Ca²⁺-activated K⁺ currents were made from dissociated cockroach neurons by means of the gigaohm-seal patch-clamp technique. Bursts of single channel openings were composed of two distinct classes: the 'long-open burst' contained groups of long, rectangular, pulse-like openings with durations of 3.5 to 1.2 ms (depending on membrane potential), whereas the 'flickering burst' consisted of clusters of brief openings with an average duration of 0.4 ms (voltage-independent) separated by short closings with a duration of about 1.0 ms. The long-open burst and the flickering burst appeared to reflect distinct states of a single Ca²⁺-activated K⁺ channel because direct transitions between these two types of burst were often detected. We present a kinetic scheme for the gating activation pathway of a neuronal Ca²⁺-activated K⁺ channel, based on these findings.     

Methoxyverapamil Reduction of Nicotine-induced Catecholamine Release Involves Inhibition of Nicotinic Acetylcholine Receptor Currents

The mechanism by which the putative Ca²⁺ channel blocker methoxyverapamil (D600) inhibits nicotine-induced catecholamine release was investigated in bovine adrenal chromaffin cells and in neurons from paravertebral sympathetic ganglia of chick embryos. We found D600 to prevent catecholamine release evoked by 30 s applications of nicotine with a significantly higher potency than the release induced either by 30 s K⁺ depolarizations or by electrical field stimulation of sympathetic neurons. Like the use-dependent action of D600 upon Ca²⁺ channels, the magnitude of inhibition of the K⁺-evoked secretion depended on the duration of stimulation (10 s to 5 min). Data on catecholamine release were supplemented by patch-clamp recordings. We found whole-cell currents in chromaffin cells evoked by (extrapolated) 0.5 s applications of nicotine to be significantly more sensitive to D600 than Ca²⁺ currents induced by a 0.5 s depolarization from -80 to 0 mV. In both instances, the potency of D600 depended on the duration of the (nicotinic and depolarizing) stimuli. Our data suggest that D600 inhibits nicotine-induced catecholamine release by reducing nicotinic acetylcholine receptor currents rather than voltage-gated Ca²⁺ currents. Hence, in chromaffin cells as well as in sympathetic neuronal preparations, D600 does not appear to be a suitable tool to investigate the part voltage-activated Ca²⁺ currents play in cellular events induced by nicotine.     

Electrophysiological Interactions Between 5-Hydroxytryptamine and Thyrotropin Releasing Hormone on Rat Hippocampal CA1 Neurons

Intracellular recording from CA1 neurons of the rat hippocampal slice preparation was used to examine the possibility of functional interactions between 5-hydroxytryptamine (5-HT) and thyrotropin releasing hormone (TRH), which act as cotransmitters in other areas of the central nervous system. 5-HT (30 μM) elicited complex effects consisting of biphasic changes in membrane potential and a strong depression of the afterhyperpolarization (AHP) following a spike burst. TRH (10 μM) did not alter membrane potential or input conductance but it produced a partial block of the AHP. Under single-electrode voltage clamp, 5-HT and TRH both reduced the amplitude of voltage-activated total K⁺ currents. When the two substances were co-applied, their actions were occluded. The voltage-activated K⁺ current remaining in Ca²⁺-free solution lost its sensitivity to 5-HT and TRH, suggesting that the K⁺ current modulated by TRH and 5-HT was Ca²⁺-dependent, although TRH itself did not depress high-threshold voltage-activated Ca²⁺ currents. When a relatively small concentration (5 μM) of 5-HT was co-applied with an equimolar amount of TRH, the degree of block of the spike AHP was the sum of the two individual effects of these drugs. It is suggested that in hippocampal pyramidal cells 5-HT and TRH influenced neuronal excitability by depressing a Ca²⁺-dependent K⁺ current, a phenomenon perhaps mediated through a common intracellular second messenger pathway.     

Ca2+ Channels and Ca2+-Activated K+ Channels in Adult Rat Gonadotrophin-Releasing Hormone Neurones

Gonadotrophin-releasing hormone (GnRH) neurones represent the final output neurones in the neuroendocrine system for the control of reproduction. To understand the reproductive neuroendocrine system, an investigation of the intrinsic and extrinsic properties of GnRH neurones is essential. In this review, we focus on the intrinsic properties and summarise our recent findings of ion channels expressed in rat GnRH neurones. Rat GnRH neurones express all four types of high voltage-activated Ca²⁺channel (L, N, P/Q, R) and the low voltage-activated Ca²⁺ channel (T). GnRH neurones also express two types of Ca²⁺-activated K⁺ [K(Ca)] channel: the small conductance Ca²⁺-activated K⁺ (SK) channel and the large conductance Ca²⁺- and voltage-activated K⁺ (BK) channel. The activities of these Ca²⁺ and K(Ca) channels regulate cell excitability and cellular calcium homeostasis.     

Analysis of the Ion Channel Complement of the Rat Oligodendrocyte Progenitor in a Commonly Studied In vitro Preparation

We have analysed the ion channel complement of the oligodendrocyte-type 2 astrocyte (O-2A) glial cell progenitor obtained from the commonly studied neonatal rat mixed brain preparation. Ionic currents, in O-2A progenitors identified on both morphological and immunological grounds, were recorded using the whole-cell variant of the patch-clamp technique. The cells had an average resting membrane potential close to -50 mV and fired single action potentials in response to suprathreshold current injections. Using voltage-clamp methods we were able to identify and characterize a voltage-activated TTX-sensitive Na⁺ current, two classes of voltage-activated outward K⁺ currents, an inactivating inwardly rectifying K⁺ current, a voltage-activated Cl^- current and at least three classes of Ca²⁺ current.     

Effect of Thiol Reagents on Phosphoinositide Hydrolysis in Rat Brain Synaptoneurosomes

Some divalent ions, such as Cd²⁺ and Zn²⁺, are able to stimulate phosphoinositide (PI) breakdown and to inhibit receptor-mediated PI metabolism. These ions are also known to react with the free – SH groups of proteins. This prompted us to investigate the effects of more potent sulphhydryl reagents, Hg²⁺ and p -chloromercuric benzosulphonic acid (PCMBS), on the inositol phosphate (IP) accumulation triggered by the neuroactive substances: glutamate, carbachol and K⁺, using synaptoneurosomes from 8-day-old rat forebrains. Hg²⁺ and PCMBS, depending on their concentration, had two distinct effects on IP accumulation: at low doses, Hg²⁺ (from 1 to 10 μM) and PCMBS (0.1 mM) by themselves stimulated PI breakdown, inhibited glutamate-elicited IP accumulation and had additive effects with respect to carbachol-induced IP stimulation. At higher doses, Hg²⁺ (from 0.01 to 1 mM) inhibited both basal and neuroactive substance-stimulated IP accumulation. PCMBS (1 mM), provoked only an inhibition of the agonist-stimulated IP formation. Monitoring membrane potential and intracellular Ca²⁺ with the fluorescent dyes diSC2(5) and fura2, respectively, indicated that these mercurials could strongly depolarize the synaptoneurosomal membrane and produce a Ca²⁺ influx dependent on extracellular Ca²⁺. The stimulatory effects of low concentrations of mercurials on PI turnover could be linked to the depolarization they provoke and the subsequent Ca²⁺ rise, which in turn is known to stimulate some phospholipase C enzymes. The inhibitory effects observed at high concentrations might be due to a loss of activity of proteins involved in PI breakdown, as all receptor-mediated IP accumulations were inhibited.     

Exocytosis and Selective Neurite Calcium Responses in Rat Cerebellar Granule Cells During Field Stimulation

The free calcium concentration, [Ca²⁺]_c, in fura-2-loaded rat cerebellar granule cells was investigated by digital imaging during trains of uniform field stimuli in order to compare the ability of calcium channels in somata and neurites to respond to brief, physiologically relevant depolarizations. Very few somata responded to 20 Hz trains of 1 ms pulses, while virtually all neurites showed an extensive increase which was rapidly reversed when stimulation was terminated. In contrast, both somata and neurites responded when cells were depolarized with 50 mM KCl. The field stimuli evoked a tetrodotoxin-sensitive increase in Na⁺ concentration in both somata and neurites. When 4-aminopyridine, which inhibits delayed K⁺ currents in these cells, was present during the field stimulus both somata and neurites increased their [Ca²⁺]_c, suggesting that prolongation of the duration of depolarization is required for somatic Ca²⁺ channel activation. The neurite response did not depend on the orientation of the neurite relative to the applied field. The neurite response was insensitive to nifedipine (1 μM) and ω-agatoxin-IVA (30 nM) but was uniformly inhibited by ω-conotoxin-GVIA (30% inhibition at 1 μM) and ω-conotoxin-MVIIC (44% inhibition at 5 μM). The two inhibitors were not additive. The neurite [Ca²⁺]_c response was insensitive to the combination of ionotropic glutamate receptor antagonists. Field stimulation caused the exocytosis of the fluorescent probe FM1-43 previously loaded during KCl depolarization, suggesting that presynaptic Ca²⁺ channels contribute to the field-evoked neurite response.     

Unique clustering of A-type potassium channels on different cell types of the main olfactory bulb

Theoretical and functional studies predicted a highly non-uniform distribution of voltage-gated ion channels on the neuronal surface. This was confirmed by recent immunolocalization experiments for Na⁺, Ca²⁺, hyperpolarization activated mixed cation and K⁺ channels. These experiments also indicated that some K⁺ channels were clustered in synaptic or non-synaptic membrane specializations. Here we analysed the subcellular distribution of Kv4.2 and Kv4.3 subunits in the rat main olfactory bulb at high resolution to address whether clustering characterizes their distribution, and whether they are concentrated in synaptic or non-synaptic junctions. The cell surface distribution of the Kv4.2 and Kv4.3 subunits is highly non-uniform. Strong Kv4.2 subunit-immunopositive clusters were detected in intercellular junctions made by mitral, external tufted and granule cells (GCs). We also found Kv4.3 subunit-immunopositive clusters in periglomerular (PGC), deep short-axon and GCs. In the juxtaglomerular region some calretinin-immunopositive glial cells enwrap neighboring PGC somata in a cap-like manner. Kv4.3 subunit clusters are present in the cap membrane that directly contacts the PGC, but not the one that faces the neuropil. In membrane specializations established by members of the same cell type, K⁺ channels are enriched in both membranes, whereas specializations between different cell types contain a high density of channels asymmetrically. None of the K⁺ channel-rich junctions showed any of the ultrastructural features of known chemical synapses. Our study provides evidence for highly non-uniform subcellular distributions of A-type K⁺ channels and predicts their involvements in novel forms of intercellular communication in the olfactory pathway.     

Voltage-dependent Currents in Microvillar Receptor Cells of the Frog Vomeronasal Organ

Vomeronasal receptor cells are differentiated bipolar neurons with a long dendrite bearing numerous microvilli. Isolated cells (with a mean dendritic length of 65 μm) and cells in mucosal slices were studied using whole-cell and Nystatin-perforated patch-clamp recordings. At rest, the membrane potential was −61 ± 13 mV (mean ± SD; n = 61). Sixty-four per cent of the cells had a resting potential in the range of –60 to –86 mV, with almost no spontaneous action potential. The input resistance was in the GΩ range and overshooting repetitive action potentials were elicited by injecting depolarizing current pulses in the range of 2 – 10 pA. Voltage-dependent currents were characterized under voltage-clamp conditions. A transient fast inward current activating near –45 mV was blocked by tetrodotoxin. In isolated cells, it was half-deactivated at a membrane potential near –75 mV. An outward K⁺ current was blocked by internal Cs⁺ ions or by external tetraethylammonium or Ba²⁺ ions. A calcium-activated voltage-dependent potassium current was blocked by external Cd²⁺ ions. A voltage-dependent Ca²⁺ current was observed in an iso-osmotic BaCl₂ solution. Finally, a hyperpolarization-activated inward current was recorded. Voltage-dependent currents in these microvillar olfactory receptor neurons appear qualitatively similar to those already described in ciliated olfactory receptor cells located in the principal olfactory epithelium.     

Pronase Acutely Modifies High Voltage-activated Calcium Currents and Cell Properties of Lymnaea Neurons

Pronase E ('pronase') is one of the proteolytic enzymes that are used in preparative procedures such as cell isolation and to soften the sheath of invertebrate ganglia. Although several effects of proteolytic enzymes on the physiology of non-neuronal tissues have been described, the effects of these enzymes on central neurons have received little attention. We examined the effects of bath-applied pronase on neurons in the Lymnaea central nervous system and in vitro . Pronase caused action potential broadening in neurons that exhibit a shoulder on the repolarization phase of their action potentials. This effect of pronase was accompanied by, although unrelated to, a depolarization and decrease in action potential interval. Some, but not all, effects of pronase in the central nervous system were reversible. For example, the decreases in membrane potential and action potential interval were both reversed after ∼1 h of washing with saline. However, the effect of pronase on the action potential duration was not reversed after a period of 90 min. The modulation of action potential width prompted us to examine Ca²⁺ currents. Exposure to pronase resulted in an increase in both peak and late high voltage-activated Ca²⁺ currents in isolated neurons. Pronase neither changed the inactivation rate nor caused a shift in the current-voltage relationship of the current. The changes in action potential duration could be prevented by application of 0.1 mM Cd²⁺, indicating that the action potential broadening caused by pronase depends on Ca²⁺ influx. This is the first systematic study of the acute and direct actions of pronase on Ca²⁺ currents and cell properties both in the CNS and in vitro .     

Cholinergic Responses in Developing Outer Hair Cells of the Rat Cochlea

Acetylcholine-evoked currents were investigated using the conventional whole-cell patch-clamp recording technique in developing outer hair cells (OHCs). The cells were isolated from the rat cochlea at different stages of postnatal development ranging from day 4 (P4) to P30. Acetylcholine-evoked currents could be recorded at P6 and P8. At this developmental stage, the majority of OHCs displayed inward nicotinic-like currents near the resting membrane potential. These cholinergic currents zeroed near 0 mV, as expected for a non-selective cation current, and could be reversibly blocked by d-tubocurarine. At P12 and adult stage, the cholinergic response of OHCs switched to an outward current reversing near E _K and displaying a bell shape peaking between -40 and -30 mV. This change in polarity of the acetylcholine response during postnatal development might be explained by progressive functional coupling between acetylcholine ionotropic receptors permeable to Ca²⁺ and nearby Ca²⁺-activated K⁺ channels at the synaptic pole of OHCs.     

Locatization of Voltaae-sensitive Ca2+ Fluxes and Neuropeptide Y Immunoreactivity to Varicosities in SH-SY5Y Human Neuroblastoma Cells Differentiated by Treatment with the Protein Kinase Inhibitor Staurosporine

The distribution of voltage-sensitive elevations of the level of Ca²⁺ in untreated SH-SY5Y cells and cells that had been induced to differentiate with staurosporine was investigated by monitoring fura-2 fluorescence in cell suspensions, and by using microfluorometry and quantitative fluorescence imaging on cell bodies and on cellular processes. Cell bodies of both types of cells displayed small Ca²⁺ elevations, which were composed of transient and sustained components. Elevations were partially sensitive to the L- and N-channel blockers nifedipine (1 μM) and ω-conotoxin GVIA (100 nM) respectively. Up to ten times higher Ca²⁺ elevations were observed in varicosities of treated cells than in cell bodies of treated and untreated cells. These elevations were insensitive to compounds known to release Ca²⁺ from intracellular stores. Elevations of Ca²⁺ were sustained, and they were insensitive to 5 pM nifedipine, 100 nM ω-agatoxin IVA and 100 nM ω-conotoxin GVIA, and partially sensitive to 2 pM ω-conotoxin GVIA, indicating predominance of non-L-type, non-N-type, non-P-type channel activity. The intracellular localization of neuropeptide Y, a marker of differentiation in these cells, was also investigated by fluorescence immunocytochemistry. Varicosities of treated cells displayed marked fluorescence when viewed in a confocal microscope. These findings show that the varicosities of staurosporine-treated cells exhibit some of the functional properties of nerve terminals. The varicosities resemble boutons en passant nerve endings and they seem to express Ca²⁺ channels different from those in the cell body.