Myelin

Myelin is a specialized, multilamellar plasma membrane extension synthesized by myelinating glial cells that spirals tightly around vertebrate neuron axons to form a discontinuous electrical insulator. By wrapping an axon in 10 to 160 concentric lipid bilayers with virtually all cytoplasm extruded between adjacent wraps, the myelin sheath increases effective transmembrane electrical resistance (R_m) by approximately two orders of magnitude and decreases effective transverse membrane capacitance (C_m) in inverse proportion to the number of lamellae. This passive dielectric architecture confines transmembrane ionic currents to periodic unmyelinated gaps—the nodes of Ranvier—where high densities of voltage-gated Na⁺ ion channels regenerate the action potential via saltatory conduction, boosting propagation velocities from 0.5–2 m s⁻¹ in unmyelinated fibers to 80–120 m s⁻¹ in large myelinated axons while reducing the metabolic cost of transmembrane pump operation.

In gnathostome vertebrates (jawed fishes through mammals), myelination is partitioned anatomically and ontogenetically between two distinct glial lineages. Within the central nervous system (CNS—brain, spinal cord, and optic nerve), a single neural-tube-derived oligodendrocyte extends multiple cytoplasmic processes to myelinate up to 50 distinct axonal internodes simultaneously. Within the peripheral nervous system (PNS—cranial and spinal nerves), neural-crest-derived Schwann cells establish a strict 1:1 stoichiometric relationship with a single axonal segment, enveloping a single internode of 100 to 1,500 μm in length.

Lipid-Rich Dielectric Stoichiometry and Compaction Proteins

Unlike typical eukaryotic cell membranes, which contain roughly equal dry weights of protein and lipid (50:50 ratio), compact myelin displays an inverted stoichiometry of 70–85% lipid and 15–30% protein by dry mass. This elevated lipid fraction minimizes dielectric polarizability and aqueous ionic leakage. The lipid bilayer of myelin is enriched in unesterified cholesterol (approx. 25–28% of total lipid weight), glycosphingolipids—specifically galactocerebroside (GalC) and its sulfated derivative sulfatide, which together account for ~20–25% of myelin lipids—and long-chain saturated sphingomyelin and ethanolamine plasmalogens. Together, cholesterol and glycosphingolipids assemble into tightly packed, liquid-ordered microdomains across the extracellular leaflet that resist passive water and ion permeation.

During myelination, the leading edge ("inner tongue") of the glial process advances around the axon while cytoplasmic water is excluded, fusing the intracellular cytoplasmic faces into the electron-dense major dense line (MDL) and apposing the extracellular surfaces into the intraperiod line (IPL) with a repeating lamellar periodicity of ~12–16 nm in X-ray diffraction and cryo-electron microscopy. CNS and PNS myelin employ partially distinct structural proteins to maintain this compaction:

At the cytoplasmic surface, myelin basic protein (MBP)—an intrinsically disordered, highly cationic peripheral protein (pI > 10.5)—neutralizes negatively charged phosphatidylserine headgroups on opposing inner leaflets, acting as an electrostatic molecular zipper that forms the major dense line. In the CNS, the four-transmembrane protein PLP1 stabilizes the intraperiod line, whereas in the PNS, the single-pass immunoglobulin-superfamily glycoprotein myelin protein zero (MPZ / P0) forms homophilic tetramers across the extracellular space alongside peripheral myelin protein 22 (PMP22).

Length Constant, Time Constant, and the Optimal g-Ratio

The biophysical advantage of myelination follows directly from the one-dimensional passive cable equation governing electrotonic voltage decay along a cylindrical core conductor:

λ² · (∂²V / ∂*x*²) = τ_m · (∂*V* / ∂*t*) + V

where V is the electrotonic displacement from the resting potential established by the Nernst equation, x is axial distance along the axon, t is time, λ = √(r_m / r_i) is the length constant (space constant), and τ_m = r_m · c_m is the membrane time constant (with r_m representing membrane resistance per unit length in Ω·cm, r_i axial axoplasmic resistance per unit length in Ω·cm⁻¹, and c_m capacitance per unit length in F·cm⁻¹).

When N myelin bilayers wrap an axon in series, resistors add directly (r_m,myelin ≈ N · r_m,bare) while series capacitors combine reciprocally (c_m,myelin ≈ c_m,bare / N). For a sheath of N = 100 lamellae, internodal membrane resistance increases ~100-fold and internodal capacitance drops ~100-fold. Increasing r_m expands the length constant λ by an order of magnitude (from ~0.2 mm to several millimeters), preventing transmembrane current leakage across the 1-mm internode so that axial current I_i flows directly to the next node of Ranvier with minimal attenuation. Simultaneously, the 100-fold drop in c_m drastically reduces the capacitive charge (Q = C · Δ*V*) required to depolarize the internodal membrane, allowing the passive electrotonic wave to charge the downstream nodal membrane to threshold in microseconds.

However, adding myelin wraps cannot increase conduction velocity indefinitely: for a fixed total outer fiber diameter (d_total), increasing myelin thickness constricts the internal axonal diameter (d_axon), which increases axial core resistance r_i ∝ 1 / (d_axon)². Theoretical modeling by W. A. H. Rushton (1951) and Richard Smith and Zoltán Koles (1970) demonstrated that conduction velocity is maximized when the g-ratio—defined as the ratio of inner axonal diameter to total outer myelinated fiber diameter (g = d_axon / d_total)—falls within the narrow optimum of 0.60 to 0.70 in the PNS and 0.70 to 0.80 in the CNS. Furthermore, whereas conduction velocity in unmyelinated axons scales with the square root of axonal diameter (v ∝ √*d*_axon, requiring the squid Loligo giant axon to reach 500–1,000 μm to achieve 25 m s⁻¹), conduction velocity in myelinated fibers scales linearly with outer diameter (v ∝ d_total), enabling a 20-μm mammalian α-motoneuron to conduct at 100–120 m s⁻¹.

Molecular Architecture of the Node of Ranvier

Saltatory conduction relies on the polarized molecular compartmentalization of the axonal membrane into four distinct longitudinal domains:

1. The Node of Ranvier (~1 μm long): Exposed directly to the extracellular fluid and enriched in voltage-gated Na_V1.6 channels at densities of 1,000 to 1,500 channels μm⁻² (compared to <25 μm⁻² in the internode), alongside slow K_V7.2/7.3 (KCNQ) channels. Na_V1.6 channels are anchored to the underlying βIV-spectrin/actin cytoskeleton by the scaffolding protein ankyrin-G and the cell adhesion molecule neurofascin-186 (NF186).
2. The Paranode (~3–5 μm flanking the node): Where cytoplasmic loops of each myelin lamella terminate against the axolemma, forming spiral septate-like junctions composed of axonal contactin-1 and Caspr1 (contactin-associated protein 1) bound to glial neurofascin-155 (NF155). This complex acts as a molecular fence and paracellular diffusion barrier.
3. The Juxtaparanode (~10–15 μm adjacent to the paranode): Sealed beneath compact myelin and enriched in fast delayed-rectifier K_V1.1 and K_V1.2 channels complexed with Caspr2 and TAG-1, which dampen repetitive back-firing during recovery from an action potential.
4. The Internode (100–1,500 μm): Occupied by the compact myelin sheath, across which oligodendrocytes also deliver lactate and pyruvate via monocarboxylate transporters (MCT1/MCT2) to fuel axonal mitochondrial ATP synthesis.

Disruption of the myelin sheath or paranodal junctions exposes the low-Na_V-density internodal axolemma and increases membrane capacitance C_m by up to 100-fold. Because the incoming axial current from the preceding node must now charge a massive bare membrane area, depolarization fails to reach threshold (conduction block), or action potentials slow and desynchronize across fiber tracts.

In Multiple Sclerosis (MS), an autoimmune inflammatory disorder of the CNS, autoreactive T cells (specifically CD4⁺ Th1 and Th17 subsets) and autoantibodies cross the blood-brain barrier to attack oligodendrocyte epitopes (such as MBP, PLP1, and myelin oligodendrocyte glycoprotein, MOG), recruiting macrophages and microglia that strip myelin via phagocytosis and form sclerotic plaques. In the PNS, Guillain-Barré syndrome (acute inflammatory demyelinating polyradiculoneuropathy, AIDP) arises when post-infectious molecular mimicry (frequently following Campylobacter jejuni enteritis) triggers complement system and macrophage attack against Schwann cell membranes or nodal gangliosides (GM1, GD1a). Inherited peripheral neuropathies, collectively termed Charcot-Marie-Tooth (CMT) disease, result from gene dosage or point mutations in PNS myelin genes—most commonly a 1.4-Mb duplication on chromosome 17p11.2 encompassing PMP22 (CMT1A) or mutations in MPZ (CMT1B)—causing chronic segmental demyelination, Schwann cell "onion-bulb" hypertrophy, and distal muscle atrophy.

How does myelin increase action potential velocity while reducing brain energy consumption?

Myelin decreases transverse membrane capacitance (C_m) ~100-fold and increases membrane resistance (R_m) ~100-fold across the internode. Consequently, depolarizing currents do not leak out across the internode and require minimal charge to change the internodal voltage, rapidly reaching the next node of Ranvier. Because transmembrane Na⁺ and K⁺ fluxes are restricted to the tiny surface area of the nodes (<0.5% of total axonal area), the Na⁺/K⁺-ATPase transmembrane pump expends up to 50–100 times less ATP per action potential than in an unmyelinated axon of equivalent conduction velocity.

Why is the optimal axonal g-ratio around 0.6 to 0.7?

The g-ratio (d_axon / d_total) balances two competing electrical parameters within a fixed anatomical volume. If the myelin sheath is too thin (g → 1.0), transverse capacitance is high and current leaks across the internode, slowing conduction. Conversely, if the myelin sheath is too thick (g → 0.1), the internal axonal lumen is constricted, which causes axial core resistance (r_i ∝ 1 / d_axon²) to surge and choke longitudinal current flow. Biophysical cable theory shows that conduction speed peaks when g ≈ 0.60–0.77.

Why do peripheral nerves regenerate better after injury than central nervous system tracts?

In the PNS, injured Schwann cells downregulate myelin genes, clear myelin debris alongside infiltrating macrophages, secrete neurotrophic factors (NGF, BDNF), and align within their intact basal lamina tubes to form bands of Büngner that guide regenerating axons at ~1–3 mm day⁻¹. In the CNS, oligodendrocytes lack a guiding basal lamina, clear debris slowly, and express potent axonal growth-cone inhibitors—including Nogo-A, myelin-associated glycoprotein (MAG), and oligodendrocyte myelin glycoprotein (OMgp)—which bind the axonal NgR1 receptor complex to activate RhoA/ROCK signaling and collapse regenerating growth cones.

1. Nave, K. A., & Werner, H. B. (2014). "Myelination of the nervous system: mechanisms and functions." Annual Review of Cell and Developmental Biology, 30, 503–533. https://doi.org/10.1146/annurev-cellbio-100913-013101
2. Waxman, S. G. (1980). "Determinants of conduction velocity in myelinated nerve fibers." Muscle & Nerve, 3(2), 141–150. https://doi.org/10.1002/mus.880030207
3. Salzer, J. L. (2015). "Schwann cell myelination." Cold Spring Harbor Perspectives in Biology, 7(8), a020529. https://doi.org/10.1101/cshperspect.a020529
4. Stadelmann, C., Timmler, S., Barrantes-Freer, A., & Simons, M. (2019). "Myelin in the central nervous system: structure, function, and pathology." Physiological Reviews, 99(3), 1381–1431. https://doi.org/10.1152/physrev.00031.2018