Clustering of Na Channels and Formation of New Nodes of Ranvier

The results described thus far suggest that after demyeli-nation conduction may be blocked, but can be restored by just the earliest and most minimal association of Schwann cells with axons (Smith et al., 1982; Shrager, 1988; Shrager and Rubinstein, 1990). The low density of Na+ channels in the internode could, in principle, participate in the restoration of conduction, but computational models confirm that without additional restructuring, the high capacitance of the demyelinated axon limits the invading signal to levels insufficient to activate these channels. Immunocytochemical labeling has greatly improved the resolution with which ion channels can be localized, and results shed light on both recovery processes during demyelination and on normal axonal development. An antibody with excellent specificity for vertebrate Na+ channels was first made by Levinson (Ellisman and Levinson, 1982; Dugandzija-Novakovic et al., 1995). This antibody was targeted to a region of the intracellular linker between domains III and IV that is conserved in all mammalian voltage-dependent Na+ channel subtypes. Normal nodes of Ranvier are brightly labeled with indirect immunofluorescence. After demyelination by lysolecithin injection, Schwann cells proliferate and adhere to axons, and may then begin the process of myelination. At this stage (overlapping ensheathment) their protein expression pattern changes markedly, with NCAM, L1, and the p75 NGF receptor downregulated, and with increased expression of myelin proteins, notably myelin-associated glycoprotein (MAG) (Martini and Schachner, 1986). Initially, MAG appears uniformly over the Schwann cell surface and then becomes increasingly sequestered in cyto-plasmic-containing regions, including the paranode, as myelination progresses. The ability to distinguish premyeli-nating from myelinating Schwann cells was an important step in the development of a general hypothesis regarding the clustering of Na+ channels and the formation of nodes of Ranvier. At issue was the question of the controlling influence in determining the ultimate sites of nodes. Were these zones predetermined by the axon at regular intervals, or were they the result of glial influences upon the neuron? The remyelinating axon provided early clues.

At about 1 week postinjection, rat sciatic axons have very few newly adherent Schwann cells, and also very few clusters of Na+ channels within the demyelinated zone. The clusters seen are about 1 mm apart and are likely to be at the sites of the original nodes (Fig. 3A). However, just a few days later, Schwann cells appear, and associated with them are new clusters of Na+ channels (Fig 3B). In Fig. 3B-E nerves were labeled with antibodies to MAG (red) to identify Schwann cells committed to myelination and also with antibodies to pan Na+ channels (green). The uniform expression of MAG over the glial surface indicates that these cells are just beginning the process of myelination. These Schwann cells are wrapping multiple lamellae and are also growing longitudinally. As they extend processes, they appear to "push" the Na+ channel clusters along with them, because these clusters always appear just outside the tips of the MAGlabeled zone, never under it. This was confirmed at the ultrastructural level (Novakovic et al., 1996). In time, clusters associated with neighboring Schwann cells approach each other (Fig. 3C) and, ultimately, appear to fuse (Fig. 3D) to form a new node of Ranvier (Fig. 3E). The sketch in Fig. 3 illustrates the basic hypothesis. A quantitative analysis of the frequency of occurrence of each stage supported the view that this sequence of events is correct (Dugandzija-Novakovic et al., 1995). Further, the distance across remyeli-nating Schwann cells (between clusters) grows in a roughly exponential fashion over time (Fig. 4). Thus, at least in the lysolecithin model of demyelination, it is the Schwann cells, and not the axon, that determine the location of new nodes of Ranvier. Tzoumaka et al. (1995) demonstrated that at least the early stages of this process could take place even in the absence of communication with the neuronal soma. It is likely, therefore, that the Na+ channels that cluster adjacent to the glial processes are derived from the low density internodal pool described earlier. As will be discussed later, in more chronic demyelinating conditions, there is evidence for additional de novo synthesis of these channels (see also Chapter 7). Finally, although this chapter is concerned primarily with demyelination, it should be noted that a similar mechanism of node formation has been proposed for development as well (Vabnick et al., 1996).

The initial clustering of Na+ channels illustrated in Fig. 3 is likely to represent the basis for the ^-nodes seen functionally by Smith et al. (1982). Further, it is also probable that this reorganization is an essential step in the early restoration of conduction seen in Fig. 2B. If the only improvement was in passive cable properties, the intermittent one to two layers of Schwann cell ensheathment would be minimal. By concentrating Na+ channels at these sites, conduction is more readily restored. The high level of stability of nodal clusters (seen as late as 9 days after myelin disruption [Custer et al., 2003]) is almost certainly due to the link of Na+ channels to the cytoskeleton via ankyrinG (Kordeli et al., 1995; Bennett and Lambert, 1999). It is not yet clear just what the role of ankyrinG may be in the initial clustering process. AnkyrinG appears at new cluster sites virtually simultaneously with Na+ channels. On the other hand, in remyelinating axons, the early Na+ channel clusters that formed were not as stable as original adult clusters (Custer et al., 2003). It may be that there is an initial sequestration that serves to accumulate a high density of channels at the

Figure 3 Na+ channel clustering during demyelination and early remyelination in the PNS. (A) An original nodal cluster at 1 week postinjection. (B-E) New node formation, 12 to 15 days postinjection. (B) New Na+ channel clusters forming at the edges of a MAG-positive Schwann cell. (C) Na+ channel clusters associated with neighboring Schwann cells approaching each other as the Schwann cells lengthen. (D) Two clusters appear to fuse. (E) A new, broad node of Ranvier. Scale bar, 5 pm. (Bottom) Proposed mechanism of node formation.

Figure 3 Na+ channel clustering during demyelination and early remyelination in the PNS. (A) An original nodal cluster at 1 week postinjection. (B-E) New node formation, 12 to 15 days postinjection. (B) New Na+ channel clusters forming at the edges of a MAG-positive Schwann cell. (C) Na+ channel clusters associated with neighboring Schwann cells approaching each other as the Schwann cells lengthen. (D) Two clusters appear to fuse. (E) A new, broad node of Ranvier. Scale bar, 5 pm. (Bottom) Proposed mechanism of node formation.

glial edges, but then further links to ankyrinG may be forged through multiple binding sites. A hypothesis for cluster formation that incorporates the first step is presented later in this chapter.

Experimental allergic neuritis (EAN) is an autoimmune disease of the PNS that is inducible in rodents and has similarities to human neuropathies, including Guillain-Barre syndrome. When Lewis rats were immunized with purified bovine root myelin, Na+ channel immunofluorescence at PNS nodes became more diffuse, and eventually undetectable (Novakovic et al., 1998). The loss of nodal channels corresponded closely with the development of clinical dis-

Figure 4 Distance across remyelinating Schwann cells plotted vs. days postinjection. All measurements were made across a MAG-positive Schwann cell. Filled circles show distances between Na+ channel clusters. Open circles show distances between Casprl-positive sites. The solid curve is drawn to indicate trends. The dashed line is drawn at 270 pm, the distance measured after remyelination is complete. It is thus the prediction if the axon predetermined the location of new nodes of Ranvier, and clearly does not fit the experimental points. That results are virtually the same whether Na+ channels or Casprl is measured indicates that the loci of the channels are tightly linked to the Schwann cell process edges (if exclusively so the two should agree within a few mm). (Reproduced by permission from the Journal of Neuroscience [Custer et al., 2003], copyright 2003 by the Society for Neuroscience.)

Figure 4 Distance across remyelinating Schwann cells plotted vs. days postinjection. All measurements were made across a MAG-positive Schwann cell. Filled circles show distances between Na+ channel clusters. Open circles show distances between Casprl-positive sites. The solid curve is drawn to indicate trends. The dashed line is drawn at 270 pm, the distance measured after remyelination is complete. It is thus the prediction if the axon predetermined the location of new nodes of Ranvier, and clearly does not fit the experimental points. That results are virtually the same whether Na+ channels or Casprl is measured indicates that the loci of the channels are tightly linked to the Schwann cell process edges (if exclusively so the two should agree within a few mm). (Reproduced by permission from the Journal of Neuroscience [Custer et al., 2003], copyright 2003 by the Society for Neuroscience.)

ease. During recovery, new Na+ channel clusters formed at the edges of adherent Schwann cells, and the formation of new nodes paralleled that seen after lysolecithin injection. However, at many sites in EAN, paranodes were retracted but myelin stripping was absent. The nodal Na+ channel cluster then split into two parts, each remaining highly focal (Fig. 5) (Novakovic et al., 1998). It thus appeared as though these channels retain their association with an individual Schwann cell even after myelin has formed. This pattern has recently also been seen in lysolecithin-induced demyelination at early stages (Arroyo, 2003). On the other hand, when myelin is retracted by exposure to protease, Na+ channels remain as a single cluster at the original nodal site (Dugandzija-Novakovic et al., 1995). Finally, it must be pointed out that all of the discussion thus far has been focused on acute changes in Na+ channel organization in peripheral nerves (scale of days or weeks). These events may be responsible for some of the clinical remissions in demyelinating disease that can occur in such a time frame.

Figure 5 Redistribution of Na+ channels in experimental allergic neuritis (EAN). This is a retracted nodal region, 28 days after induction of EAN. Scale bar, 25 mm. (Reproduced by permission from Muscle and Nerve [Novakovic et al., 1998], copyright 1998 by Wiley Periodicals Inc.)

Quite different, but equally clinically relevant changes take place in more chronic disease, as discussed later.

et al., 1997, 2001). Only Nav1.2 is seen and the progression to Nav1.6 does not take place. These sites could signify an early stage of molecular organization in the axon, but it has been argued that they may not represent axonally determined loci of nodes of Ranvier, and the extent to which node formation in the CNS is governed by soluble factors released by glia remains unresolved (Kazarinova-Noyes and Shrager, 2002; Salzer, 2003).

Peripheral Neuropathy Natural Treatment Options

Peripheral Neuropathy Natural Treatment Options

This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.

Get My Free Ebook


Post a comment