Neuroanatomy of Demyelination

1. The Oligodendrocyte — the Cell at the Centre of MS

The oligodendrocyte is a CNS glial cell whose principal function is to wrap myelin around axons. A single mature oligodendrocyte myelinates segments of 20–60 different axons, in striking contrast to the peripheral Schwann cell, which myelinates a single internode of a single axon.

• Lineage — oligodendrocytes derive from oligodendrocyte progenitor cells (OPCs, also called NG2 cells), which persist throughout adult CNS as ~5% of all CNS cells.
• Differentiation — OPCs express PDGFRα and NG2 (CSPG4); mature oligodendrocytes express MBP, PLP1, MOG, MAG, CNP and CC1 (APC).
• Vulnerability — oligodendrocytes have a remarkable metabolic load (a single cell synthesises >100× its dry weight in myelin lipid daily) and are exquisitely sensitive to oxidative stress, glutamate excitotoxicity (via AMPA/kainate receptors), and immune attack.
• Remyelination capacity — OPCs can be recruited to demyelinated lesions and produce new myelin, but this efficiency declines with age and after repeated demyelinating events — a major target for therapy (Part VIII).

2. The Myelin Sheath

CNS myelin is a multilamellar membrane wrapped concentrically around an axon, compacted by elimination of cytoplasm at major dense lines (cytoplasmic apposition) and intraperiod lines (extracellular apposition). Its composition is unusual:

• ~70% lipid, ~30% protein by dry weight (CNS myelin); peripheral myelin is somewhat less lipid-rich.
• Dominant lipids — cholesterol, galactosylceramide, sulphatide, plasmalogens; very-long-chain fatty acids.
• Proteins — PLP1 (proteolipid protein, ~50% of CNS myelin protein), MBP (myelin basic protein, ~30%), CNPase, MAG, and the surface autoantigen MOG (myelin oligodendrocyte glycoprotein), which is the immunogen used in EAE and the target of MOGAD antibodies.
• g-ratio — the ratio of axon diameter to total fibre diameter is ~0.6–0.8; a higher (thinner-myelin) g-ratio is the standard signature of remyelinated lesions.

3. Saltatory Conduction — Why Myelin Matters

The myelin sheath is a high-resistance, low-capacitance insulator that confines voltage-gated Na⁺channel clustering to the node of Ranvier, between two internodes. The action potential effectively jumps from node to node:

The cable equation describes the membrane potential along the axon: \( \tau \partial_t V = \lambda^2 \partial_x^2 V - V \), where \( \lambda = \sqrt{r_m/r_i} \) is the length constant. Myelin raises \( r_m \) by ~5000-fold, raising \( \lambda \) from ~0.1 mm to ~5 mm. Demyelination collapses \( \lambda \) and produces:

• Conduction slowing — clinically detectable as prolonged P100 latency on visual evoked potentials.
• Conduction block — complete failure to propagate, expressed as the negative symptoms of MS (vision loss, weakness, sensory loss).
• Heat sensitivity (Uhthoff’s phenomenon) — small temperature rises further reduce safety margin in demyelinated fibres; the «hot bath test» is historical.
• Ectopic discharge — bare axons develop new Na⁺ channels that can fire spontaneously, producing positive symptoms (paroxysmal tonic spasms, trigeminal neuralgia, paraesthesiae).

4. The Internodal Architecture — Node, Paranode, Juxtaparanode

The molecular architecture of the myelinated fibre is finely regionalised:

• Node of Ranvier — ~1 µm gap; densely packed Na_v1.6 (Na_v1.2 in immature axons), KCNQ2/3 K⁺ channels, neurofascin-186, NrCAM and ankyrin-G anchors.
• Paranode — the septate-like junction between the axolemma and the lateral loops of myelin; defined by Caspr-contactin-NF155, the molecular «diffusion barrier» that segregates nodal Na⁺ channels from juxtaparanodal K⁺ channels.
• Juxtaparanode — high density of K_v1.1/1.2 channels, sequestered under intact myelin; their exposure after demyelination contributes to conduction failure (and is the target of dalfampridine, 4-aminopyridine).
• Internode — ~1 mm long under healthy myelin; ~30–50% shorter after remyelination.

Demyelination disrupts this architecture: paranodes detach, K⁺ channels become exposed, Na⁺channel density redistributes (with up-regulation of Na_v1.2 and Na_v1.6 along denuded axons), and chronic axonal Na⁺/Ca²⁺ overload produces secondary axonal degeneration — the substrate of progressive disability.

5. The Classical MS Plaque

The pathological hallmark of MS is the plaque: a focal, well-demarcated lesion of demyelination. Plaques are classified by activity:

Lucchinetti et al. (Ann Neurol 2000) classified active lesions into four immunopathological patterns (I–IV). Patterns I/II are T-cell/macrophage and antibody-complement mediated (the majority of MS); pattern III shows preferential MAG loss with apoptotic oligodendrocytes (a «dying-back» oligodendrogliopathy reminiscent of hypoxia); pattern IV is rare. Whether patterns are stable disease subtypes or temporal phases of one process remains debated.

6. Dawson’s Fingers

In 1916 the Edinburgh pathologist James Dawson described, in microtome serial sections of MS brain, demyelinated lesions that radiate outward from the lateral ventricles along the deep medullary veins (venae caudate, v. thalamostriatae), perpendicular to the corpus callosum. These Dawson’s fingers are the most pathognomonic MRI sign of MS: ovoid, periventricular, perpendicular to the ventricular ependyma, best appreciated on sagittal FLAIR.

Their topography reflects the perivenular nature of MS inflammation: the dense plexus of subependymal venules draining toward the ventricles is the route of immune-cell entry into white matter.

7. Lesion Topography — the Predilection Sites

MRI lesion distribution is non-random. The McDonald 2017 criteria require T2 hyperintense lesions in ≥2 of 4 typical locations to satisfy dissemination in space:

8. Cortical & Deep Grey-Matter Pathology

Cortical demyelination — underestimated for a century because of poor histological visualisation — is now recognised as a major substrate of cognitive decline and progression. Bo, Kidd, Geurts and Calabrese described three cortical lesion types:

• Type I (leukocortical) — lesions span the grey-white junction; share features with juxtacortical white-matter lesions.
• Type II (intracortical) — small, perivascular lesions confined within cortex.
• Type III (subpial) — ribbons of cortical demyelination extending from pia inward; the most MS-specific lesion type, associated with overlying meningeal B-cell follicle-like structures (Magliozzi et al., Brain 2007).

Subpial cortical demyelination correlates strongly with progressive disease, cortical atrophy, and the presence of meningeal lymphoid follicles — the rationale for B-cell-targeted and meningeal-penetrant (BTK-inhibitor) therapy.

Deep grey-matter structures — thalamus, basal ganglia, hippocampus — show selective atrophy detectable on volumetric MRI from early disease stages; thalamic atrophy in particular is a sensitive marker of progression and is used as a secondary endpoint in modern trials.

9. The Chronic Plaque, Smouldering MS & Brain Atrophy

In long-standing disease, the brain at autopsy shows widespread cortical thinning, ventricular enlargement, corpus callosum thinning, optic-nerve atrophy, and spinal-cord atrophy. Brain volume loss in untreated MS proceeds at ~0.5–1.0% per year (vs ~0.1–0.3% in age-matched controls); modern DMTs reduce this to control levels in best-responders.

Chronic active (smouldering) lesions — identified on susceptibility-weighted MRI as paramagnetic rim lesions (PRLs) — harbour iron-laden, activated microglia and continue to expand outward over years (Absinta et al., Nature 2021, JAMA Neurol 2019). Patients with >4 PRLs at diagnosis have substantially worse 10-year outcomes. PRLs are now proposed biomarkers for the «compartmentalised inflammation» that classical, peripherally-acting DMTs do not adequately treat — the biological rationale for BTK inhibitors (covered in Part VII).

The transition from inflammatory relapsing disease to neurodegenerative progressive disease — secondary progressive MS — is anatomically heralded by axonal loss outpacing inflammation, extensive subpial cortical demyelination, and meningeal B-cell follicle-like aggregates.