Distribution of saxitoxin-binding sites in mammalian neural tissue

Azaspiracid-1, a potent, nonapoptotic new phycotoxin with several cell targets

This paper reports on potential cellular targets of azaspiracid-1 (AZ-1), a new phycotoxin that causes diarrhoeic and neurotoxic symptoms and whose mechanism of action is unknown. In excitable neuroblastoma cells, the systems studied were membrane potential, F-actin levels and mitochondrial membrane potential. AZ-1 does not modify mitochondrial activity but decreases F-actin concentration. These results indicate that the toxin does not have an apoptotic effect but uses actin for some of its effects. Therefore, cytoskeleton seems to be an important cellular target for AZ-1 effect. AZ-1 does not induce any modification in membrane potential, which does not support for neurotoxic effects. In human lymphocytes, cAMP, cytosolic calcium and cytosolic pH (pHi) levels were also studied. AZ-1 increases cytosolic calcium and cAMP levels and does not affect pHi (alkalinization). Cytosolic calcium increase seems to be dependent on both the release of calcium from intracellular Ca(2+) pools and the influx from extracellular media through Ni(2+)-blockable channels. AZ-1-induced Ca(2+) increase is negatively modulated by protein kinase C (PKC) activation, protein phosphatases 1 and 2A (PP1 and PP2A) inhibition and cAMP increasing agents. The effect of AZ-1 in cAMP is not extracellularly Ca(2+) dependent and insensitive to okadaic acid (OA).

A serial section electron microscope study of an identified Ia afferent collateral in the cat spinal cord

Serial section electron microscopy has been used to examine a horseradish peroxidase (HRP)-labelled group Ia afferent collateral from its entry point in the grey matter to its termination in Clarke's column of the cat spinal cord. A wide range of geometries and myelination patterns were identified along the collateral, including 1) nodes specialized to exhibit a single synaptic bouton, 2) nodes specialized to exhibit two or more synaptic boutons connected by fine, unmyelinated lengths of the collateral, 3) terminal heminodes, along which boutons were separated by unmyelinated branches, and 4) complex arrangements along which myelinated and unmyelinated branches gave rise to one or more boutons. Thirty-six synaptic boutons of varied shape and size were exhibited by this collateral. Previous studies have shown that the geometry, branching, and myelination pattern of an axon play an important role in determining the amplitude and duration of an action potential propagating along that axon. In turn, the amplitude and duration of a presynaptic action potential influence the efficacy of transmitter release. The varied axonal geometries and myelination patterns observed in the present study provide further evidence in support of our previous proposal that there may be considerable nonuniformity in the efficacy of synaptic transmission among release sites arising from the same primary afferent fiber.

A comparison of fibrillation in denervated skeletal muscle of the anaesthetized rat and guinea-pig

We have used intracellular recording in vivo to study fibrillation (spontaneous repetitive membrane activity) in extensor digitorum longus (fast twitch) and soleus (slow twitch) muscles of the anaesthetized rat and guinea-pig denervated for periods of about 10 to 60 days. The proportion of fibres fibrillating in the guinea-pig soleus was greater than 50% in most animals up to the longest period of denervation (65 days). Fibrillation was rarely found in rat soleus after three weeks of denervation. Its incidence in the extensor digitorum longus muscles of both species was intermediate. The mean frequency of fibrillation was higher in guinea-pig extensor digitorum longus (16 Hz) and soleus (8 Hz) than in the rat extensor digitorum longus (3 Hz) and soleus (2 Hz). The resting membrane potentials of the denervated muscles were less than normal and correlated inversely with the frequency of fibrillation but not with the incidence of fibrillation: in rat soleus, many fewer fibers were fibrillating at a given membrane potential than in the other three muscles. The incidence of fibrillation was compared with previously reported tensions of the four denervated muscles and was found to have the same rank order. We suggest that fibrillation may reduce atrophy (and hence tension loss) of denervated muscle, which may have implications for artificial stimulation. Fibrillation frequency was directly related to changes in twitch speed of the four muscles after denervation.

Sodium channels in the cytoplasm of Schwann cells

Immunoblotting, ultrastructural immunocytochemistry, and tritiated saxitoxin ([3H]STX) binding experiments were used to study sodium channel localization in Schwann cells. Polyclonal antibody 7493, which is directed against purified sodium channels from rat brain, specifically recognizes a 260-kDa protein corresponding to the alpha subunit of the sodium channel in immunoblots of crude glycoproteins from rat sciatic nerve. Electron microscopic localization of sodium channel immunoreactivity within adult rat sciatic nerves reveals heavy staining of the axon membrane at the node of Ranvier, in contrast to the internodal axon membrane, which does not stain. Schwann cells including perinodal processes also exhibit antibody 7493 immunoreactivity, localized within both the cytoplasm and the plasmalemma of the Schwann cell. To examine further the possibility that sodium channels are localized within Schwann cell cytoplasm, [3H]STX binding was studied in cultured rabbit Schwann cells, both intact and after homogenization. Saturable binding of STX was significantly higher in homogenized Schwann cells (410 +/- 37 fmol/mg of protein) than in intact Schwann cells (214 +/- 21 fmol/mg of protein). Moreover, the equilibrium dissociation constant was higher for homogenized preparations (1.77 +/- 0.37 nM) than for intact Schwann cells (1.06 +/- 0.29 nM). These data suggest the presence of an intracellular pool of sodium channels or channel precursors in Schwann cells.

Regulation of muscle sodium channel transcripts during development and in response to denervation

We have recently described the cloning and functional expression of a new sodium channel subtype, microI, isolated from a denervated rat skeletal muscle cDNA library. In studies described here, we have used RNase protection and Northern blot analyses to examine the expression of microI mRNA in different tissues and in neonatal, adult, and adult denervated muscle. We found that microI transcripts were not expressed in brain or heart, or in the myogenic cell line L6, even after differentiation to myotubes. Transcripts for microI were present at low levels in neonatal skeletal muscle and increased to maximum levels in adult tissue, paralleling the expression of tetrodotoxin (TTX)-sensitive sodium currents. Surprisingly, denervation of adult muscle was also followed by a rise in microI mRNA, at a time when TTX-insensitive currents reappear. These results show that expression of this channel subtype is regulated by tissue type, development, and innervation.

Sodium currents in Schwann cells from myelinated and non-myelinated nerves of neonatal and adult rabbits

1. Patch-clamp methods were used to study sodium channels in Schwann cells obtained from four different tissue sources. Primary cultures of Schwann cells were prepared from the sciatic nerve and from the vagus nerve of neonatal and of adult rabbits. In the adult, the sciatic is predominantly myelinated whereas the vagus is predominantly non-myelinated. Whole-cell currents, and single-channel currents in outside-out membrane patches, were analysed. 2. No substantial differences were noted in the passive electrical properties (input resistance, cell capacitance, resting membrane potential) of the four groups of cells. Similarly, no substantial differences were found in the average properties of sodium currents (maximum current, maximum conductance, time-to-peak current, current-voltage relation, h infinity relation) recorded from each type of cell in cultures less than 8 days old. At 10-17 days a fall in the size of the sodium currents recorded from cells in the vagal cultures was found. 3. Exposure of the cells to proteolytic enzymes or collagenase, under conditions similar to those used when the cells were put in culture initially, substantially reduced the size of the peak sodium currents recorded from the cells 24 h later. 4. The results of experiments on Schwann cells with retracted processes indicated that sodium channels are present in the processes extending from each pole of the cell soma and that the plasmalemmal density of these channels in the processes is about the same as it is at the soma. 5. Recordings from outside-out patches revealed no apparent differences in the properties of single-channel sodium currents in patches from cells obtained from the four different sources. The single-channel conductance was about 20 pS for each of the four groups. Ensemble currents from single-channel records were similar in time course to those of whole-cell currents. 6. Saxitoxin reduced the maximum sodium conductance in Schwann cells and bound to the cells with equally high affinity. The equilibrium dissociation constant was about 2 nM at 20-22 degrees C. 7. It is argued that the expression of sodium channels in myelinating Schwann cells does not differ substantially from that of non-myelinating Schwann cells.

Low density of sodium channels supports action potential conduction in axons of neonatal rat optic nerve

The density of sodium channels in premyelinated axons was estimated from measurements of the binding of [3H]saxitoxin to neonatal rat optic nerve. The maximum saturable binding capacity of the nerve was 16.2 +/- 1.2 fmol/mg of wet weight, with an equilibrium dissociation constant of 0.88 +/- 0.18 nM (mean +/- SEM). These values correspond to a high-affinity saxitoxin-binding site density of approximately 2/microns 2 within premyelinated axon membrane. Action potential propagation in neonatal rat optic nerve is completely blocked by 5 nM saxitoxin, indicating that action potential electrogenesis is mediated by channels that correspond to high-affinity saxitoxin-binding sites. These results demonstrate that action potential conduction is supported by a low density of sodium channels in this system. Since the internodal axon membrane of myelinated fibers may contain a low density of sodium channels, it is possible that restoration of conduction in some demyelinated fibers may not require additional sodium channel incorporation into the demyelinated axon membrane.

Sodium channels in astrocytes of rat optic nerve in situ: immuno-electron microscopic studies

Immuno-electron microscopic localization of sodium channels within astrocyte somata and processes of adult rat optic nerve was demonstrated with polyclonal antibody 7493. In immunoblots of crude glycoproteins from adult rat optic nerve, antisera 7493, which is directed against purified rat brain sodium channels, recognizes a 260 kDa protein. Antisera 7493 intensely immunostains axon membrane at nodes of Ranvier. Associated perinodal astrocyte processes are also stained with antisera 7493. In addition, astrocyte cell bodies and major processes exhibit immunoreactivity with antibody 7493. Immunostaining with antisera 7493 is heterogeneously distributed within astrocyte cytoplasm and also appears to be associated with some regions of astrocyte plasmalemma. Glial filaments are not immunostained with 7493 antisera. Astrocyte processes forming the glial limitans and surrounding blood vessels display reduced immunoreactivity to 7493 compared to longitudinally oriented or perinodal astrocyte processes. However, some focal regions of the glial limitans exhibit robust 7493 immunostaining. Oligodendrocytes do not display 7493 antisera immunoreactivity. Optic nerve sections incubated with preimmune sera or with 7493 antisera that had been previously adsorbed with purified sodium channel protein, exhibited no immunoreactivity. These results demonstrate localization of sodium channels within astrocytes in situ of rat optic nerve and extend previous electrophysiological and pharmacological findings of sodium channels in cultured astrocytes. Possible functional roles of sodium channels within astrocytes are discussed.

Sodium-channel turnover in rabbit cultured Schwann cells

Radiolabelled saxitoxin has been used as a chemical marker for the voltage-dependent sodium channels expressed in the plasmalemma of rabbit Schwann cells in culture. Proteolytic enzymes destroy this saxitoxin-binding capacity, which gradually reappears with an exponential time constant of about 3.1 days. Exposure of cultured Schwann cells to tunicamycin, an inhibitor of glycosylation, leads to a progressive exponential fall in saxitoxin-binding capacity, again with a time constant of about 3.1 days. The assumption that the steady-state density of Schwann cell sodium channels is maintained by a constant synthesis of channels in the face of a rate of loss from the membrane proportional to the amount of channel already present, leads to the conclusion that these channels have an average lifetime of about 3.1 days. The metabolic consequences of this rapid turnover of Schwann cell sodium channels is discussed.