Galactocerebroside and sulfatide are two major glycolipids in myelin; however, their independent functions are not fully understood. The absence of these glycolipids causes disruption of paranodal junctions, which separate voltage-gated Na(+) and Shaker-type K(+) channels in the node and juxtaparanode, respectively. In contrast to glial cells in the central nervous system (CNS), myelinating Schwann cells in the peripheral nervous system (PNS) possess characteristic structures, including microvilli and Schmidt-Lanterman incisures, in addition to paranodal loops. All of these regions are involved in axo-glial interactions. In the present study, we examined cerebroside sulfotransferase-deficient mice to determine whether sulfatide is essential for axo-glial interactions in these PNS regions. Interestingly, marked axonal protrusions were observed in some of the nodal segments, which often contained abnormally enlarged vesicles, like degenerated mitochondria. Moreover, many transversely cut ends of microvilli surrounded the mutant nodes, suggesting that alignments of the microvilli were disordered. The mutant PNS showed mild elongation of nodal Na(+) channel clusters. Even though Caspr and NF155 were completely absent in half of the paranodes, short clusters of these molecules remained in the rest of the paranodal regions. Ultrastructural analysis indicated the presence of transverse bands in some paranodal regions and detachment of the outermost several loops. Furthermore, the numbers of incisures were remarkably increased in the mutant internode. Therefore, these results indicate that sulfatide may play an important role in the PNS, especially in the regions where myelin-axon interactions occur.

In myelinated axons, voltage-gated sodium channels specifically cluster at the nodes of Ranvier, while voltage-gated potassium channels are located at the juxtaparanodes. These characteristic localizations are influenced by myelination. During development, Nav1.2 first appears in the predicted nodes during myelination, and Nav1.6 replaces it in the mature nodes. Such replacements may be important physiologically. We examined the influence of the paranodal junction on switching of sodium channel subunits using the sulfatide-deficient mouse. This mutant displayed disruption of paranodal axoglial junctions and altered nodal lengths and channel distributions. The initial switching of Nav1.2 to Nav1.6 occurred in the mutant optic nerves; however, the number of Nav1.2-positive clusters was significantly higher than in wild-type mice. Although no signs of demyelination were observed at least up to 36 weeks of age, sodium channel clusters decreased markedly with age. Interestingly, Nav1.2 stayed in some of the nodal regions, especially where the nodal lengths were elongated, while Nav1.6 tended to remain in the normal-length nodes. The results in the mutant optic nerves suggested that paranodal junction formation may be necessary for complete replacement of nodal Nav1.2 to Nav1.6 during development as well as maintenance of Nav1.6 clusters at the nodes. Such subtype abnormality was not observed in the sciatic nerve, where paranodal disruption was observed. Thus, the paranodal junction significantly influences the retention of Nav1.6 in the node, which is followed by disorganization of nodal structures. However, its importance may differ between the central and peripheral nervous system.

Glycosphingolipids (GSLs) occur in all mammalian plasma membranes. They are most abundant in neuronal cells and have essential roles in brain development. Glucosylceramide (GlcCer) synthase, which is encoded by the Ugcg gene, is the key enzyme driving the synthesis of most neuronal GSLs. Experiments using conditional Nestin-Cre Ugcg knockout mice have shown that GSL synthesis in vivo is essential, especially for brain maturation. However, the roles of GSL synthesis in mature neurons remain elusive, since Nestin-Cre is expressed in neural precursors as well as in postmitotic neurons. To address this problem, we generated Purkinje cell-specific Ugcg knockout mice using mice that express Cre under the control of the L7 promoter. In these mice, Purkinje cells survived for at least 10-18 weeks after Ugcg deletion. We observed apparent axonal degeneration characterized by the accumulation of axonal transport cargos and aberrant membrane structures. Dendrites, however, were not affected. In addition, loss of GSLs disrupted myelin sheaths, which were characterized by detached paranodal loops. Notably, we observed doubly myelinated axons enveloped by an additional concentric myelin sheath around the original sheath. Our data show that axonal GlcCer-based GSLs are essential for axonal homeostasis and correct myelin sheath formation.

Patients with peripheral neuropathy frequently suffer from positive sensory (pain and paresthesias) and motor (muscle cramping) symptoms even in the recovery phase of the disease. To investigate the pathophysiology of increased axonal excitability in peripheral nerve regeneration, we assessed the temporal and spatial expression of voltage-gated Na(+) channels as well as nodal persistent Na(+) currents in a mouse model of Wallerian degeneration. Crushed sciatic nerves of 8-week-old C57/BL6J male mice underwent complete Wallerian degeneration at 1 week. Two weeks after crush, there was a prominent increase in the number of Na(+) channel clusters per unit area, and binary or broad Na(+) channel clusters were frequently found. Excess Na(+) channel clusters were retained up to 20 weeks post-injury. Excitability testing using latent addition suggested that nodal persistent Na(+) currents markedly increased beginning at week 3, and remained through week 10. These results suggest that axonal regeneration is associated with persistently increased axonal excitability resulting from increases in the number and conductance of Na(+) channels.