Axonal morphological changes following impulse activity in mouse peripheral nerve in vivo: the return pathway for sodium ions.

KEY POINTS: Conduction in myelinated axons involves substantial ion movements that must be reversed to restore homeostasis. The pathway taken by sodium ions returning to their original location and the potential osmotic consequences are currently unknown. We report striking morphological changes in axons following sustained impulse conduction that appear to result from osmosis and to indicate accumulation of ions in the periaxonal space followed by their release at the paranode. We conclude that the morphological changes illustrate a hitherto unrecognized part of normal axonal physiology that may also indicate the return pathway for the sodium ions involved in impulse formation. ABSTRACT: Myelinated axons can conduct sustained trains of impulses at high frequency, but this involves substantial ion movements that must be reversed to restore homeostasis. Little attention has been paid to the potential osmotic consequences of the ion movements or to the pathway taken by sodium ions returning to their original endoneurial location, given that the axolemmal Na(+)-K(+)-ATPase extrudes these ions into the periaxonal space beneath the myelin rather than into the endoneurium. Serial confocal imaging of fluorescent axons conducting at sustained physiological frequencies in vivo has revealed surprising morphological changes that may illuminate these problems. Saphenous nerves and spinal roots of anaesthetized transgenic mice expressing axoplasmic yellow fluorescent protein were stimulated electrically or pharmacologically (veratridine). Within 2 h, the axon herniated on one or both sides of the nodal membrane, displacing the paranodal myelin and widening the nodal gap. The herniated axoplasm became directed back towards the internode, forming a 'cap' up to 30 μm long. Concurrently, the fluid in the expanded periaxonal space accumulated into droplets that appeared to travel to the paranode, where they escaped. No such alterations occurred in axons treated with sodium channel or Na(+)-K(+)-ATPase inhibitors. Remarkably, impulse conduction continued throughout, and all these changes reversed spontaneously over hours or days. The morphological changes were verified ultrastructurally, and occurred in virtually all myelinated axons. The findings appear to reveal an overlooked part of the physiological repertoire of nerve fibres, and here they are interpreted in terms of osmotic changes that may illuminate the pathway by which sodium ions return to the endoneurial space after they have entered the axon during impulse conduction.