Myelinated axon physiology and regulation of neural circuit function

Abstract

The study of structural and functional plasticity in the central nervous system (CNS) to date has focused primarily on that of neurons and synapses. However, more recent studies implicate glial cells as key regulators of neural circuit function. Among these, the myelinating glia of the CNS, oligodendrocytes, have been shown to be responsive to extrinsic signals including neuronal activity, and in turn, tune neurophysiological function. Due to the fact that myelin fundamentally alters the conduction properties of axons, much attention has focused on how dynamic regulation of myelination might represent a form of functional plasticity. Here, we highlight recent research that indicates that it is not only myelin, but essentially all the function-regulating components of the myelinated axon that are responsive to neuronal activity. For example, the axon initial segment, nodes of Ranvier, heminodes, axonal termini, and the morphology of the axon itself all exhibit the potential to respond to neuronal activity, and in so doing might underpin specific functional outputs. We also highlight emerging evidence that the myelin sheath itself has a rich physiology capable of influencing axonal physiology. We suggest that to fully understand nervous system plasticity we need to consider the fact that myelinated axon is an integrated functional unit and adaptations that influence the entire functional unit are likely to underpin modifications to neural circuit function.

Keywords: activity; axon sub-domains; ion channels; myelin; myelinated axon plasticity.

Figure 1

Overview of a neuron and its axonal subdomains, their possible modifications and predicted effects on axonal physiology (table). (1) Axon initial segment has been observed to change its position along the axon (a), alter its length (b), and ion channel density (c). (2) Myelin sheaths have been reported to change their length (a) and thickness, by increasing the number of wraps around the axon (b). (3) Nodes of Ranvier could increase their length (a), ion channel concentration (b) and diameter (c). (4) Within the nodal‐paranodal region the length and ion channel density of the node of Ranvier may change upon changes to myelin (a), or via an inside‐out mechanism, whereby the axonal cytoskeleton fine tunes the morphology and composition of the region (b). (5) The axon itself can vary in diameter (a, b) as well as alter ion channel density in unmyelinated segments. (6) In the axon terminus regions, the concentration of ion channels can be increased (a) or decreased (b) at the heminode

Figure 2

Ion channel/transporter expression contributing to axonal homeostasis. Following an action potential axonal K⁺ ions enter the periaxonal space via voltage‐gated potassium channels (Kv). In order to prevent elevated axonal excitability and prolonged excitotoxicity K⁺ is expected to be rapidly cleared from underneath the myelin sheath by myelinating oligodendrocytes via K_IR, and possibly also Na⁺─K⁺─Cl⁻ cotransporter‐1 and/or the Na⁺/K⁺ ATPase. Excess of K⁺ in the oligodendrocyte is then passed on to astrocytes via gap junctions. At the nodes of Ranvier, K⁺ efflux is predicted to be buffered by axonal Na⁺/K⁺ ATPases and astrocytic K_IR channels. Paranodal proteins Neurofascin, contactin and Contactin‐associated protein (Caspr) separate nodal and juxtaparanodal regions and form a barrier preventing protein diffusion. Oligodendrocytes also provide metabolic support to axons via monocarboxylate transporters (MCTs) at the interface of the axon and myelin sheath