Morphological and electrical properties of oligodendrocytes in the white matter of the corpus callosum and cerebellum

Abstract

In the central nervous system, electrical signals passing along nerve cells are speeded by cells called oligodendrocytes, which wrap the nerve cells with a fatty layer called myelin. This layer is important for rapid information processing, and is often lost in disease, causing mental or physical impairment in multiple sclerosis, stroke, cerebral palsy and spinal cord injury. The myelin speeds the information flow in two ways, by decreasing the capacitance of the nerve cell and by increasing its membrane resistance, but little is known about the latter aspect of myelin function. By recording electrically from oligodendrocytes and imaging their morphology we characterised the geometry and, for the first time, the resistance of myelin in the brain. This revealed differences between the properties of oligodendrocytes in two brain areas and established that the resistance of myelin is sufficiently high to prevent significant slowing of the nerve electrical signal by current leakage through the myelin.

Figure 5. Simulating the action potential in myelinated axons to investigate the effect of the oligodendrocyte conductance

A, schematic diagram of a myelinated axon. B, the electrical model which was simulated (Model C of Richardson et al. 2000). C, specimen action potentials predicted at successive nodes (nodes 4 to 14 of the 21 nodes simulated) using the morphological parameters for corpus callosum axons with a membrane conductance of 10 pS μm⁻². D, predicted dependence of conduction speed on the conductance of a single membrane of the myelin sheath for corpus callosal axons. E, predicted dependence of conduction speed on the conductance of a single membrane of the myelin sheath for cerebellar axons. Arrows denote standard conductance value (10 pS μm⁻²) assumed in many simulations, and also the conductance values derived from the measurements in this paper if we assume either that only the outer membrane contributes to the membrane conductance or that all membranes conduct (see main text).

Figure 1. Myelin basic protein labelling and oligodendrocyte morphology in corpus callosum and cerebellum

A and B, low power (A) and high power (B) views of MBP labelling in the corpus callosum at P12. Myelinated axons congregate in the white matter of corpus callosum, with diffuse myelin sheets visible in the surrounding grey matter. C and D, low power (C) and high power (D) views of MBP labelling in the cerebellum at P12. Myelinated axons enter the white matter of each lobule from where they turn into the surrounding grey matter. E–H, Lucifer yellow fills demonstrating the range of morphologies of oligodendrocytes. E and F, oligodendrocytes in the corpus callosum forming 3 (E) and 11 (F) internodes (counted by focusing up and down through the stack of images). G and H, oligodendrocytes in the cerebellum forming 5 (G) and 19 (H) internodes.

Figure 2. Morphological parameters of oligodendrocytes in the corpus callosum and cerebellum

A, percentage of oligodendrocytes in the two areas with different numbers of internodes. The modal value is larger for callosal cells (36 callosal and 52 cerebellar cells were studied). B, mean number of processes per cell. C, percentage of oligodendrocytes in the two areas with different ranges of mean internode length per cell (34 callosal and 28 cerebellar cells were studied). A significant number of cerebellar cells are found with mean process lengths longer than any seen in the corpus callosum. D, mean process length per cell. E, mean soma diameter. F and G, internode length is independent of number of internodal processes. Scatter plots (each point is one cell) of mean internodal process length per cell versus number of processes for corpus callosal (n= 33) (F) and cerebellar oligodendrocytes (n= 28) (G). Linear regressions shown are not significantly different from horizontal lines (P= 0.61 for F and 0.51 for G) indicating no dependence of process length on number of processes.

Figure 3. Typical membrane current properties of oligodendrocytes

A, specimen current responses of a corpus callosum oligodendrocyte to voltage steps in 20 mV intervals from a holding potential of –71 mV (recorded with potassium gluconate based intracellular medium). B, mean ‘steady state’ (at the end of 200 ms voltage steps) I–V relation for 8 corpus callosal oligodendrocytes. C, specimen current responses of a cerebellar oligodendrocyte to voltage steps in 20 mV intervals from a holding potential of –63 mV (recorded with KCl based intracellular medium), showing a sag in the current during very positive voltage steps (however, most cerebellar oligodendrocytes showed current responses similar to those in A). D, mean ‘steady state’I–V relation for 16 cerebellar oligodendrocytes.

Figure 4. Dependence of oligodendrocyte membrane conductance on morphology

Plots for corpus callosal (A and B) and cerebellar (C and D) oligodendrocytes of membrane conductance (each point is one cell) as a function of number of internodal processes (A and C) and total length of processes (B and D) per cell. Lines are linear regressions for which the vertical axis intercept gives the predicted soma conductance and the slope gives the conductance per internode (for A and C) or the conductance per unit length of process (for B and D).