F I Consequences of Demyelination

The concept of saltatory conduction, first defined more than 50 years ago (Tasaki and Takeuchi, 1942; Huxley and Stampfli, 1949), firmly established the fact that in normal myelinated axons, inward current through sodium (Na+) channels occurs uniquely at nodes of Ranvier. Correspondingly, voltage clamp experiments, first on amphibian axons and later on mammalian fibers, demonstrated directly that there is a high density of Na+ channels at these sites (Dodge and Frankenhaeuser, 1959; Chiu et al., 1979). At issue, however, was the possible expression of these channels within internodal regions as well. Na+ channels under myelin are not likely to be activated during action potential propagation, or under voltage clamp, because they would see only a fraction of the resulting depolarization. Thus, other approaches were required to establish with certainty the distribution of these channels. This information is important both to define the steps that neurons must follow during development and because it has functional consequences in demyelinating disease. Some early experiments provided important clues. Nodal regions were found to have biochemically distinct cyto-plasmic surfaces, suggesting unique cytoskeletal components (Quick and Waxman, 1977). The more recent demonstration of a specific adapter protein (ankyrinG) and a spectrin isoform (PIV) localized to nodes and initial segments confirms this idea (Kordeli et al., 1995; Berghs et al., 2000; Komada and Soriano, 2002). Freeze-fracture replicas demonstrated high densities of large intramembranous particles in both the node and juxtaparanode, and much lower densities in the axoglial junction region of the paranode and in the remainder of the internode (Rosenbluth, 1976, 1981). The densities of these particles in the nodal gap (~1,300/pm2) agree with biophysical estimates of Na+ channel density (1,000-1,500/pm2; reviewed in Hille, (2001). This work suggested a specific clustering of Na+ channels at nodes and initial segments, but left open the possibility of a lower density within internodes.

Direct electrical measurements provide the most sensitive method of detecting voltage-dependent channels. Two groups studied gap-voltage-clamped internodes acutely exposed to lysolecithin to disrupt myelin. Voltage-dependent Na+ and potassium (K+) currents could be recorded concomitant with an increase in membrane capacitance. Grissmer (1986) found the internodal Na+ current density to be only 0.2% of the nodal level. Chiu and Schwarz (1987) measured this ratio to be about 3%, but considered the possibility that the recorded Na+ currents originated from Schwann cell membranes fused to the axolemma by lysolecithin. Schwann cells in vitro express voltage-dependent Na+ channels (Chiu et al., 1984; Shrager et al., 1985). However, it was later shown that, in vivo, these channels are restricted to nonmyelinating Schwann cells (Wilson and Chiu, 1990; Chiu, 1993).

Studies on axons demyelinated in vivo provided more details. Hall and Gregson (1971) introduced a method for focal demyelination that reproduces many aspects of inflammatory demyelinating disease and can be used in amphibian and mammalian species. A small amount of lysolecithin (1 pl, 1% in adult sciatic nerve) is injected surgically directly into a nerve trunk and the animal is allowed to recover. The drug vesiculates the outmost layers of myelin, which initiates an inflammatory response, with macrophages entering the lesion from the blood and removing myelin debris by phagocytosis. Affected internodes are completely stripped of myelin, a process that is completed by 1 week postinjection. At this time, if the nerve is dissected and teased, axons can be found that are devoid of all glial membranes and are surrounded only by a disrupted basal lamina. If the animal is allowed to recover for longer periods, Schwann cells proliferate and begin the process of remyelination (Shrager and Rubinstein, 1990). In the rat sciatic nerve, the earliest signs of repair are seen at about 9 days postinjection, and by 14 days many fibers have thin sheaths of new myelin. In the mouse, all events are speeded by 1 to 2 days. Working with both Xenopus and rat sciatic axons, Shrager recorded Na+ currents with the loose patch clamp (Fig. 1) and found an internodal density about 4% of the nodal value (Shrager, 1987, 1988, 1989). Measurements could be made as early as 1 day postinjection, by applying suction to allow the patch pipette to advance through the myelin debris and seal to the axolemma. The internodal density was constant during 2 months postinjection, suggesting that these channels are not introduced as a result of the demyelination.

Figure 1 Ionic currents recorded from a rat sciatic demyelinated internode with a loose patch clamp pipette. The membrane potential was held 30 mV negative to the resting value and was depolarized by pulses of 40, 60, 70, 80, 90, and 110 mV This nerve was 3 days postinjection. (Reprinted from Brain Research [Shrager, 1989], copyright 1989, with permission from Elsevier.)

Figure 1 Ionic currents recorded from a rat sciatic demyelinated internode with a loose patch clamp pipette. The membrane potential was held 30 mV negative to the resting value and was depolarized by pulses of 40, 60, 70, 80, 90, and 110 mV This nerve was 3 days postinjection. (Reprinted from Brain Research [Shrager, 1989], copyright 1989, with permission from Elsevier.)

The measurement also agreed well with that of the acute experiments of Chiu and Schwarz (1987). Since the nodal density of Na+ channels is about 1,000-1,500/pm2, the internodal density is 40-60/pm2. This latter figure is significant for two reasons. First, it is close to the value expected for nonmyelinated axons of similar caliber. Thus, in principle, it could support conduction. Second, although it represents only a few percent of the nodal density, since the internodal surface area is about 1,000 times that of nodes, it suggests that more than 95% of the axonal Na+ channels are internodal. Therefore, they constitute a large pool of channels that may be used in repair or replacement. When axons (both peripheral [PNS] and central nervous systems [CNS]) remyelinate they typically form several short internodes within a single previous internodal region. The gaps between these short internodes must function as nodes if saltatory conduction is to be successful through this zone, and they must therefore obtain a high density of Na+ channels from some source. These channels may be synthesized de novo in the soma and transported down the fiber, or they may be recruited from the internodal pool.

II. Reorganization of Axonal Na+ Channels after Demyelination

A series of reports from Sears and his colleagues established several important points regarding modes of conduction and localization of functional Na+ channels in demyelinated fibers. Rasminsky and Sears (1972) developed a method in which rat ventral roots were demyelinated by injection of diphtheria toxin and longitudinal currents recorded. By selective stimulation with a micro-electrode, it was possible to record propagating signals from single fibers. These authors found axons in which conduction was delayed, but remained saltatory. Bostock and Sears (1978) improved on this technique, increasing spatial resolution, and demonstrated that in addition to saltatory conduction, some fibers exhibited broad stretches of inward current characteristic of continuous conduction through a demyelinated internode. Smith et al. (1982) increased the spatial resolution of this technique even further, and also used lysolecithin to initiate demyelination instead of diphtheria toxin. In contrast with the earlier study, they found that conduction in demyelinated axons proceeded via new foci of inward current at spacings of a few hundred microns. These foci, called ^-nodes, formed before remyelination, and the authors concluded that they were likely to represent aggregates of Na+ channels. In the light of later experiments, detailed previously, showing a significant density of Na+ channels in the internodal axolemma, one may now conclude that (1) under at least some circumstances the internodal channels can be activated and support conduction and (2) these channels may be reorganized after demyelination.

One disadvantage of the approach used in the previously described work is that the recorded fiber is not seen visually and the extent of demyelination, or association with Schwann cells, is thus not known. Studies on teased sciatic axons circumvent this difficulty and allow a more definitive interpretation. Some early physiological experiments were carried out on Xenopus fibers, using the lysolecithin injection system described earlier (Shrager, 1988). At the edges of the injection zone (3 to 5 mm long) one finds heminodes separating normal myelin from the demyelinated region. Applying loose patch clamp pipettes to sites adjacent to these heminodes at about 1 week postinjection resulted in outward currents with a voltage dependence that was very sharp and was in some cases "all-or-none." These currents originated from zones of high Na+ channel density just outside the patch (inward through the channels and outward through the patch) and suggested that Na+ channels at the original nodes of Ranvier were very stable, remaining clustered even 1 week after myelin disruption. It was also possible to record propagating action potentials with the loose patch clamp because only the external potential is controlled by the pipette. At 2 weeks postinjection, a time when in Xenopus new myelin is not yet seen, but Schwann cells have begun to adhere, action potentials invaded the demyelinated zone far more than would have been possible by passive spread alone. This result was confirmed and extended by optical recording utilizing potential-sensitive dyes (Shrager and Rubinstein, 1990). It was possible to record propagating signals from single nodes of Ranvier and from small patches of membrane 10 |im long in demyelinated zones. Before Schwann cell adherence, action potentials were blocked at the heminode border of the demyelinated region (Fig. 2A). However, as soon as Schwann cells associated with the axon, but before remyelination, signals were able to traverse a full demyelinated internode, though with much lower velocity than in the myelinated regions of the same fiber (Fig 2B). As would be expected, action potentials resumed normal velocity when they emerged from the lesion into the distal myelinated zone. The block at the heminode is due to the high capacitance of the demyelinated zone, which creates an "impedance mismatch" (i.e., the last node cannot supply sufficient current to depolarize the Na+ channels at the heminode to the level required for activation) (see Chapter 6). Calculations show that the internodal density of Na+ channels is adequate for conduction (but just barely so) if the demyelinated segment can be stimulated. The newly adherent Schwann cells in Fig. 2B improve the passive cable properties, but as will be seen next, this is not the only

-3 X 10-s 3 ms

Figure 2 Optical recording of propagating action potentials. Xenopus axons were stained with the dye RH155, and changes in the amplitude of transmitted light at 705 ± 25 nm were recorded from a 10 x 10 |m region. (A) An action potential reached the last intact node before the demyelinated zone, but was blocked at the proximal heminode. The light micrograph shows the heminode. This nerve was 8 days postinjection (dpi), and macrophages (arrow) but no Schwann cells were adherent. (Reproduced from the Journal of General Physiology [Shrager and Rubinstein, 1990] by copyright permission of The Rockefeller University Press.) (B) At 10 dpi an action potential successfully propagates through a demyelinated internode. Schwann cells are adhering to the demyelinated region, but myelin is not yet seen at the ultrastructural level at this stage. The light micrograph shows the distal heminode. A few macrophages (dark cells) are also visible.

consequence of this early stage of repair that is important for restoring function.

Brain Blaster

Brain Blaster

Have you ever been envious of people who seem to have no end of clever ideas, who are able to think quickly in any situation, or who seem to have flawless memories? Could it be that they're just born smarter or quicker than the rest of us? Or are there some secrets that they might know that we don't?

Get My Free Ebook

Post a comment