Specialized paranodal and interparanodal glial-axonal junctions in the peripheral and central nervous system: a freeze-etching study

KV1 channels identified in rodent myelinated axons, linked to Cx29 in innermost myelin: support for electrically active myelin in mammalian saltatory conduction

Saltatory conduction in mammalian myelinated axons was thought to be well understood before recent discoveries revealed unexpected subcellular distributions and molecular identities of the K(+)-conductance pathways that provide for rapid axonal repolarization. In this study, we visualize, identify, localize, quantify, and ultrastructurally characterize axonal KV1.1/KV1.2 channels in sciatic nerves of rodents. With the use of light microscopic immunocytochemistry and freeze-fracture replica immunogold labeling electron microscopy, KV1.1/KV1.2 channels are localized to three anatomically and compositionally distinct domains in the internodal axolemmas of large myelinated axons, where they form densely packed "rosettes" of 9-nm intramembrane particles. These axolemmal KV1.1/KV1.2 rosettes are precisely aligned with and ultrastructurally coupled to connexin29 (Cx29) channels, also in matching rosettes, in the surrounding juxtaparanodal myelin collars and along the inner mesaxon. As >98% of transmembrane proteins large enough to represent ion channels in these specialized domains, ∼500,000 KV1.1/KV1.2 channels define the paired juxtaparanodal regions as exclusive membrane domains for the voltage-gated K(+)conductance that underlies rapid axonal repolarization in mammals. The 1:1 molecular linkage of KV1 channels to Cx29 channels in the apposed juxtaparanodal collars, plus their linkage to an additional 250,000-400,000 Cx29 channels along each inner mesaxon in every large-diameter myelinated axon examined, supports previously proposed K(+)conductance directly from juxtaparanodal axoplasm into juxtaparanodal myeloplasm in mammalian axons. With neither Cx29 protein nor myelin rosettes detectable in frog myelinated axons, these data showing axon-to-myelin linkage by abundant KV1/Cx29 channels in rodent axons support renewed consideration of an electrically active role for myelin in increasing both saltatory conduction velocity and maximum propagation frequency in mammalian myelinated axons.

Structural bases for central nervous system malfunction in the quaking mouse: dysmyelination in a potential model of schizophrenia

The dysmyelinating mouse mutant quaking (qk) is thought to be a model of schizophrenia based on diminution of CNS myelin (Andreone et al., 2007) and downregulation of the Qk gene (Haroutunian et al., 2006) in the brains of schizophrenic patients. The purpose of this study was to identify specific structural defects in the qk mouse CNS that could compromise physiologic function and that in humans might account for some of the cognitive defects characteristic of schizophrenia. Ultrastructural analysis of qk mouse CNS myelinated fibers shows abnormalities in nodal, internodal, and paranodal regions, including marked variation in myelin thickness among neighboring fibers, spotty disruption of paranodal junctions, abnormal distribution of nodal and paranodal ion channel complexes, generalized thinning and incompactness of myelin, and on many axonal profiles complete absence of myelin. These structural defects are likely to cause abnormalities in conduction velocity, synchrony of activation, temporal ordering of signals, and other physiological parameters. We conclude that the structural abnormalities described are likely to be responsible for significant functional impairment both in the qk mouse CNS and in the human CNS with comparable myelin pathology.

Electron tomography of paranodal septate-like junctions and the associated axonal and glial cytoskeletons in the central nervous system

The polarized domains of myelinated axons are specifically organized to maximize the efficiency of saltatory conduction. The paranodal region is directly adjacent to the node of Ranvier and contains specialized septate-like junctions that provide adhesion between axons and glial cells and that constitute a lateral diffusion barrier for nodal components. To complement and extend earlier studies on the peripheral nervous system, electron tomography was used to image paranodal regions from the central nervous system (CNS). Our three-dimensional reconstructions revealed short filamentous linkers running directly from the septate-like junctions to neurofilaments, microfilaments, and organelles within the axon. The intercellular spacing between axons and glia was measured to be 7.4 ± 0.6 nm, over twice the value previously reported in the literature (2.5-3.0 nm). Averaging of individual junctions revealed a bifurcated structure in the intercellular space that is consistent with a dimeric complex of cell adhesion molecules composing the septate-like junction. Taken together, these findings provide new insight into the structural organization of CNS paranodes and suggest that, in addition to providing axo-glial adhesion, cytoskeletal linkage to the septate-like junctions may be required to maintain axonal domains and to regulate organelle transport in myelinated axons.

Development of a model for microphysiological simulations: small nodes of ranvier from peripheral nerves of mice reconstructed by electron tomography

The node of Ranvier is a complex structure found along myelinated nerves of vertebrate animals. Specific membrane, cytoskeletal, junctional, extracellular matrix proteins and organelles interact to maintain and regulate associated ion movements between spaces in the nodal complex, potentially influencing response variation during repetitive activations or metabolic stress. Understanding and building high resolution three dimensional (3D) structures of the node of Ranvier, including localization of specific macromolecules, is crucial to a better understanding of the relationship between its structure and function and the macromolecular basis for impaired conduction in disease. Using serial section electron tomographic methods, we have constructed accurate 3D models of the nodal complex from mouse spinal roots with resolution better than 7.5 nm. These reconstructed volumes contain 75-80% of the thickness of the nodal region. We also directly imaged the glial axonal junctions that serve to anchor the terminal loops of the myelin lamellae to the axolemma. We created a model of an intact node of Ranvier by truncating the volume at its midpoint in Z, duplicating the remaining volume and then merging the new half volume with mirror symmetry about the Z-axis. We added to this model the distribution and number of Na+ channels on this reconstruction using tools associated with the MCell simulation program environment. The model created provides accurate structural descriptions of the membrane compartments, external spaces, and formed structures enabling more realistic simulations of the role of the node in modulation of impulse propagation than have been conducted on myelinated nerve previously.

The paranodal axo-glial junction in the central nervous system studied with thin sections and freeze-fracture

The lateral belts of the myelin sheath wind helically around the paranodal region of the axon. The lateral belt coil leaves an imprint on the axon and thus confers a conspicuous, indented configuration to the freeze-fracture faces of the axolemma. The contact area between the axolemma and the lateral belt membrane is the site of an extensive and unusual cell junction (axo-glial junction). In thin sections the junctional membranes are undulated, the peaks in one membrane mirroring the peaks in the other. The transverse bands (intercellular septa) are in register with the undulations. The intercellular space measures about 30 A. In freeze-fracture replicas, the undulations are evident as alternating ridges and grooves which run strictly parallel and are oriented at an angle with respect to the helical path of the lateral belt. Both junctional membranes contain parallel rows of intramembrane particles which coincide with the ridges and grooves and, therefore, with the intercellular septa. The center-to-center distance between septa or, equivalently, between adjacent rows of particles measures approximately 250 A. Although the axo-glial junction possesses structurally symmetrical features, there exist important differences between the two junctional membranes. The intramembrane particles of the glial and the axonal membrane differ in cleaving properties. Furthermore, in some of the fibres the E face of the junctional axolemma displays a crystalline array which is not present in the fracture faces of the glial membrane. The axo-glial junction is limited to the paranodal region, although the inner belt of the myelin sheath may form occasional junctional spots with the internodal region proper of the axolemma. The classification and the presumptive functions of the paranodal axo-glial junction are briefly discussed.

Gap junctions in the adult cerebral cortex: regional differences in their distribution and cellular expression of connexins

Gap junctions are membrane channels that mediate electrical and metabolic coupling between adjacent cells. Immunocytochemical analysis by using a panel of anti-connexin antibodies, as well as electron microscopy of thin sections and freeze-fracture replicas, has shown that gap junctions and their constituent proteins are abundant in the cerebral cortex of the adult rat. Their frequency and distribution vary in different cortical regions, which may reflect differences in the cellular and functional organization of these areas of the cortex. Gap junctions were identified between glial cells and, less frequently, between neuronal elements. Heterologous junctions were also identified between astrocytes and oligodendrocytes and between neurons and glia; the latter category included abundant junctions between astrocytic processes and neurons. Double-antibody labelling experiments in tissue sections and in acutely dissociated cells showed that connexin 32 was expressed in neurons and oligodendrocytes, whereas connexin 43, widely believed to be expressed only in astrocytes, was also localized in a population of cortical neurons. These results show that gap junctions can provide a major nonsynaptic means of communication between cortical cell types.

An orderly development of paranodal axoglial junctions and bracelets of Nageotte in the rat sural nerve

The present study was designed to assess the normal development of the paranodal apparatus with particular emphasis on axoglial junctions (AGJs) which constitute the paranodal barrier system. The sural nerve was examined in 10- and 31-day-old rats. During the early phase of myelination AGJ attachment of terminal myelin loops to the axolemma proceeded from the node to the internode. The frequency of terminal loops with AGJ attachment increased with fiber growth. As myelination advanced internodal-most loops became almost 100% attached to the axolemma by AGJs, whereas at the same time an increasing number of nodal-most loops were unattached, suggesting a lack of AGJ formation at this site. The formation of bracelets of Nageotte increased with the progressive addition of myelin loops. They formed most frequently at the juxtanodal interface between unattached and attached loops, probably reflecting crowding of terminal loops along the unchanged length of the paranodal axolemma. The findings suggest a complex but orderly age- and fiber size-dependent maturation process of the paranode and its structural barrier system. The present data will serve as a basis for the evaluation of this anatomical region in regenerating and remyelinating fibers in various neuropathies.

Development of nodes of Ranvier in feline nerves: an ultrastructural presentation

The ultrastructure of developing nodes of Ranvier and adjacent paranodes of future large myelinated fibers in feline lumbar spinal roots is described. The development starts before birth concurrent with myelination and is finished at the end of the first postnatal month when the nodal regions of future large fibers, now 4-5 microns of diameter, for the first time appear like miniatures of those of their 4 times thicker and fully mature counterparts. At this stage the fibers also begin to show mature functional properties. The latent maturation process is denoted "nodalization" and includes two major events: (1) the formation of a narrow node gap bordered by compact myelin segments and filled with Schwann cell microvilli that interconnect an undercoated nodal axolemma with rapidly increasing accumulations of mitochondria lodging in the longitudinal cords of Schwann cell cytoplasm that is distributed outside a more and more crenated paranodal myelin sheath; (2) the setting of a fixed number of nodes along the axons; an event that includes segmental axonal and myelin sheath degeneration and is concluded by the elimination of supernumerary Schwann cells.

Clusters of axonal Na+ channels adjacent to remyelinating Schwann cells

Rat sciatic nerve fibres were demyelinated by injection of lysolecithin and examined at several stages as Schwann cells proliferated, adhered, and initiated remyelination. Immunoperoxidase EM has been used to follow the clustering of Na+ channels that represents an early step in the formation of new nodes of Ranvier. At the peak of demyelination, 1 week post-injection, only isolated sites, suggestive of the original nodes, were labelled. As Schwann cells adhered and extended processes along the axons, regions of axonal Na+ channel immunoreactivity were often found just beyond their leading edges. These channel aggregates were associated only with the axolemma and Na+ channels were not detected on glial membranes. Sites with more than one cluster in close proximity and broadly labelled aggregates between Schwann cells suggested that new nodes of Ranvier formed as neighbouring Na+ channel groups merged. Schwann cells thus seem to play a major role in ion channel distributions in the axolemma. In all of these stages Na+ channel label was found primarily just outside the region of close contact between axon and Schwann cell. This suggests that Schwann cell adherence acts in part to exclude Na+ channels, or that diffusible substances are involved and can act some distance from regions of direct contact.

Myelin formation by mouse glia in myelin-deficient rats treated with cyclosporine

Previous attempts to generate myelin in the myelin-deficient rat spinal cord by transplanting mouse glia were not successful. In order to determine whether this result was due to graft rejection or to interspecies mismatch of cellular or molecular components at the axoglial junction, we have repeated the experiment in cyclosporine-treated rats. Our results show that in the immunosuppressed hosts, foetal glial xenografts form an abundance of myelin within the dorsal columns at or near the injection site about two weeks after the operation. In some cases, myelination extends virtually across the entire width of the dorsal columns. Ultrastructurally, the myelin sheaths are normal in all respects, including the presence of the 'radial component'. The lateral edges of the myelin lamellae form typical paranodal axoglial junctions, some displaying periodic 'transverse bands'. We infer that previous mouse to rat xenograft failures reflect host immune response rather than mismatch of heterologous junctional components. We also compared foetal, early post-natal and adult xenografts. Foetal donor cells, containing an abundance of precursors but virtually no mature oligodendrocytes, are more effective than neonatal donor cells in forming myelin, and after adult grafts, we found no myelin formation. Thus, in xenografts, as in allografts, foetal precursor cells are far more suitable than glia from mature donors in generating significant amounts of myelin.

GAP-43 is expressed by nonmyelin-forming Schwann cells of the peripheral nervous system

Recently it has been demonstrated that the growth-associated protein GAP-43 is not confined to neurons but is also expressed by certain central nervous system glial cells in tissue culture and in vivo. This study has extended these observations to the major class of glial cells in the peripheral nervous system, Schwann cells. Using immunohistochemical techniques, we show that GAP-43 immunoreactivity is present in Schwann cell precursors and in mature non-myelin-forming Schwann cells both in vitro and in vivo. This immunoreactivity is shown by Western blotting to be a membrane-associated protein that comigrates with purified central nervous system GAP-43. Furthermore, metabolic labeling experiments demonstrate definitively that Schwann cells in culture can synthesize GAP-43. Mature myelin-forming Schwann cells do not express GAP-43 but when Schwann cells are removed from axonal contact in vivo by nerve transection GAP-43 expression is upregulated in nearly all Schwann cells of the distal stump by 4 wk after denervation. In contrast, in cultured Schwann cells GAP-43 is not rapidly upregulated in cells that have been making myelin in vivo. Thus the regulation of GAP-43 appears to be complex and different from that of other proteins associated with nonmyelin-forming Schwann cells such as N-CAM, glial fibrillary acidic protein, A5E3, and nerve growth factor receptor, which are rapidly upregulated in myelin-forming cells after loss of axonal contact. These observations suggest that GAP-43 may play a more general role in the nervous system than previously supposed.

Three-dimensional fine structure of cytoskeletal-membrane interactions at nodes of Ranvier

Cytoskeleton-membrane-extracellular matrix interactions at the node of Ranvier were examined in both central and peripheral axons by combining three different methods for tissue preparation with three different electron microscopic techniques for imaging supramolecular structure. Conventional and three-dimensional high voltage electron microscopy of thin and semithick sections of tissues stained en bloc with ferric chloride revealed the presence of transcellular structures across the nodal gap traversing the paranodal glial-axonal junction. These structures penetrate both axonal and glial membranes and are further traced to the cortical axoplasm. This observation was verified by an examination of similar regions in rapidly-frozen freeze-substituted fresh axons. The filamentous nature of these structures, their focal attachment to the external true surface of the nodal and paranodal axolemma and their association with membrane particles were visualized in deep etch rotary-shadow replicas. At the node, both extracellular gap-crossing filaments and membrane-cytoskeletal linkers in the nodal axoplasm are joined to one of the prominent membrane particles of the nodal axolemma. At the paranodal axo-glial junction, the anchoring site of these membrane-cytoskeleton linkers are found on the linear arrays of 16 nm particles. Thus, cytoplasmic filaments and extracellular filaments or bridge structures are involved in the membrane-cytoskeletal interaction at the node and paranode. Some of these membrane particles are known to play a role in ionic conductances known to occur at this site. An additional role in cell adhesion or maintenance of the membrane specialization of this functionally important site of axolemma is now indicated.

Axolemmal abnormalities in myelin mutants

Evidence is reviewed that the paranodal axoglial junction plays important roles in the differentiation and function of myelinated axons. In myelin-deficient axons, ion flux across the axolemma is greater than that in myelinated fibers because a larger proportion of the axolemma is active during continuous, as opposed to saltatory, conduction. In addition, older myelin-deficient rats that have developed spontaneous seizures display small foci of node-like E-face particle accumulations in CNS axons as well as more diffuse regions of increased particle density and number. Assuming that the E-face particles represent sodium channels, such regions could underlie high sodium current density during activity, low threshold for excitation, and increased extracellular potassium accumulation. Depending on the degree of spontaneous channel opening, they could also represent sites of spontaneous generation of activity. The appearance of seizures and their gradual increase in frequency and severity could represent an increase in the number of such regions. In addition, diminution in the dimensions of the extracellular space during maturation would result in increased extracellular resistance, which, together with increasing axonal diameter, would tend to increase the likelihood of ephaptic interaction among neighboring axons as well as the likelihood of extracellular potassium rises to levels that could cause spontaneous activity.

Macromolecular structure of axon membrane and action potential conduction in myelin deficient and myelin deficient heterozygote rat optic nerves

The macromolecular structure of the axon membrane in optic nerves from 25-day-old male littermate control and myelin deficient (md) rats and 16-month-old md heterozygotic rats was examined with quantitative freeze-fracture electron microscopy. The axon membrane of control optic nerves displayed an asymmetrical partitioning of intramembranous particles (IMPs); P-fracture faces of myelinated internodal axon membrane were more particulate than those of pre-myelinated axons (approximately 1600 v 1100 microns-2, respectively), while relatively few IMPs (approximately 150 microns-2) were present on external faces (E-faces) of internodal or pre-myelinated axon membrane. Amyelinated axons of md optic nerves also exhibited an asymmetrical partitioning of IMPs; protoplasmic membrane face (P-face) IMP densities, taken as a group, exhibited a wide range (approximately 600-2300 microns-2) and, in most regions, E-faces displayed a relatively low IMP density (approximately 175 microns-2). Axons of greater than 0.4 microns diameter exhibited significantly greater mean P-face IMP density than axons less than 0.4 microns diameter. Aggregations of E-face IMPs (approximately 350 microns-2) were occasionally observed along amyelinated axon membrane from md optic nerves. Optic nerves from md heterozygote rats exhibit myelin mosaicism, permitting examination of myelinated and amyelinated axon membrane along the same tract. The axon membrane exhibits different ultrastructure in these two domains. Myelinated internodal axon membrane from md heterozygote optic nerves exhibits similar P- and E-face IMP densities to those of control internodal axolemma (approximately 1800 and 140 microns-2, respectively). Amyelinated axons in the heterozygote exhibit a membrane structure similar to amyelinated axons in md optic nerve. P-face IMP density of large diameter (greater than 0.4 microns) amyelinated axons from md heterozygote optic nerves is significantly greater than that of small calibre (less than 0.4 microns) axons. In most regions, amyelinated axon membrane exhibits a relatively low E-face IMP density (approximately 200 microns-2); however, focal aggregations (approximately 400 microns-2) of E-face particles are present. Electrophysiological recordings demonstrate that amyelinated axons in md optic nerves support the conduction of action potentials. Compound action potentials in md optic nerves exhibit a monophasic configuration, even at 20-days postnatal, similar to that of pre-myelinated optic nerve of 7-day-old normal rats. Moreover, conduction velocities in the amyelinated 20-day-old md optic nerve are similar to those displayed by pre-myelinated axons from 7-day-old optic nerves.(ABSTRACT TRUNCATED AT 400 WORDS)

Intercellular septate-like junction of neoplastic cells in myxopapillary ependymoma of the filum terminale

Septate junction, a common intercellular feature of invertebrate epithelium, is absent in most vertebrate tissues. In an ultrastructural study of three cases of myxopapillary ependymoma of the filum terminale, structures similar to septate junction were observed in two cases. They were circumferential bands around the finger-like processes of neoplastic cells in which slightly widened 30-40-nm intercellular spaces (as compared to 15-20 nm in non-junctional region) were transversed at regular intervals by parallel septa resulting in characteristic ladder-like appearance. The electron-dense septa, 30-40 nm in cross-length and 20-30 nm in width, were arranged in a periodicity of 40-50 nm. The septa connected to the outer leaflets of the apposing cytoplasmic membranes which appear undisrupted. Most junctions were short; the longest one contained 15 septa in a length of 1.4 microns. Hemiseptate-like junctions with septa of the same measurements were noted between the processes and the investing basement membrane. Junctional complexes such as zonula adherens and gap junction were present in the vicinity. They may represent a specific intercellular feature of myxopapillary ependymoma, and function as cellular adhesions and a mechanical support of the neoplastic cells.

Ultrastructural changes of the peripheral nerve induced by vibration: an experimental study

To investigate the effects of vibration on the peripheral nerves, rabbits were exposed to vibration of 60 cycles/s frequency with 0.35 mm amplitude (acceleration: 51 m/s2) for two hours daily. After 150, 250, 450, and 600 hours vibration, thin sections of the saphenous and median nerves were examined under the electron microscope. Vibration was found to induce the following changes: (1) disruption of the myelin sheath and constriction of the axon, (2) accumulation of vacuoles in the nodal gap and paranodal region, (3) disorganisation of the paranodal end loops and detachment of the paranodal end loops from the axolemma, (4) dilatation of the Schmidt-Lanterman incisures (SLI) and increased density of SLI, and (5) disappearance of neurotubules and neurofilaments in axons. The diameters of myelin sheaths disrupted by vibration varied from 2 to 12 microns. The extent of the myelin disruption is proportional to the vibration dose.

Freeze-fracture studies on unmyelinated axolemma of rat cervical sympathetic trunk: correlation with saxitoxin binding

The density and diameter distributions of intramembranous particles (IMPs) within unmyelinated axolemma from rat cervical sympathetic trunk were examined with freeze-fracture electron microscopy. The axolemma displays a highly asymmetrical partitioning of IMPs with ca. 1200 IMPs microns-2 on P-faces and ca. 100 IMPs microns-2 on E-faces. Particle sizes (diameters) are unimodally distributed on both fracture faces, with a range from 2.4 nm to 15.6 nm. Approximately 16% of the particles on P-faces and 28% of particles on E-faces are of a large (greater than 9.6 nm) diameter. On both fracture faces, the IMPs appear to be randomly distributed; no aggregations of particles were observed. The results indicate that there are ca. 230 large IMPs microns-2 of unmyelinated axolemma from rat cervical sympathetic trunk. The density of these IMPs is similar to the density of saxitoxin binding sites on unmyelinated axolemma from rat cervical sympathetic trunk (Pellegrino et al. 1984 (Brain Res. 305, 357-360)), which suggests that many of the large diameter particles may be the morphological correlate of voltage-sensitive Na+ channels.

Role of glial cells in the differentiation and function of myelinated axons

Myelinated axons are highly differentiated in the vicinity of the node of Ranvier, both structurally and with respect to ion channel distribution. Evidence is reviewed showing that axonal differentiation depends upon two distinct types of interaction between glial cells and the axolemma, one at the node itself, with astrocyte processes, and the second, more extensive one, in the paranodal region, with oligodendrocyte processes. In the peripheral nervous system, Schwann cells fulfill both roles. Glial or Schwann cell abnormalities, due to genetic deficiencies, diseases or experimental procedures, result in corresponding abnormalities in the axolemma and can have devastating effects on nerve fiber function. An example, the myelin-deficient mutant rat, is presented, and the defects underlying the profound and ultimately lethal neurological abnormalities seen in this mutant are discussed in relation to abnormalities in its axoglial interactions.

Freeze-fracture studies of reactive myelinated nerve fibres after diffuse axonal injury

We have studied the axonal and myelin sheath response in diffuse axonal injury after angular acceleration using the freeze-fracture and thin section techniques. It was found that the glial-axonal junction was intact until 1 h after injury. But upon loss of the nodal axolemma specialisations, after 3 to 4 h, the dimeric particles of the glial-axonal junction (GAJ) were lost and, by 6 h, the myelin lamellae became separated from the axonal remnant. There was a correlated loss of glial membrane specialisations of the GAJ during this separation. In the internodal region a suggestion of membrane damage occurred after 20 min but discrete myelin dislocations (particle-free areas) were not found until 1-h survival and were extensive by 6 h. Areas of loosely organised myelin occurred between intact axons at 7-28 days after injury. No evidence for growth cone formation was obtained.

Development of axonal-oligodendroglial relationships and junctions during myelination of the optic nerve

The early stages of myelination were examined in optic nerves of rats aged 12-15 days. The initial association between oligodendroglial processes and bare axons involves no junctional specialization, as the axoglial extracellular space remains unaltered. Following ensheathment by a collar of glial cytoplasm, at least one full rotation of mesaxon was evident before compact myelin formed. Furthermore, myelin was generally evident before a second rotation was completed. In longitudinal sections, an axoglial junction was always observed beginning on the first paranodal loop, continuing through to the last (or outermost) loop. Thus, the formation of myelin and elaboration of a junctional complex in the paranodal region follow a promyelination phase and appear to be synchronous (and possibly related) events. Although the paranodal plasmalemma and axolemma are in close apposition, there is a material in the extracellular space that precipitates phosphotungstic acid, a characteristic that appears to be featured in a number of different types of cell junctions.

Distribution of filipin-sterol complexes in the unmyelinated nerve fibre

The filipin-sterol technique was employed, together with freeze-fracture, to investigate the fine structure of the unmyelinated nerve fibre in the peripheral nerve. No heterogeneity was observed in the distribution of filipin labelling either in the Schwann cell plasma membrane or along the axolemma. The distribution of labelling is contrasted with that in the myelinated nerve fibre and related to the relative morphology and electrophysiology of the two types of fibre.

Node of Ranvier formation along fibres regenerating through silicone tube implants: a freeze-fracture and thin-section electron microscopic study

Thin-section and freeze-fracture electron microscopy have been used to examine the morphogenesis of the node of Ranvier in peripheral nerves regenerating through silicone tubes. A major question posed by this study is whether node formation in fibres regenerating across a gap recapitulates that occurring in normal development. Node formation occurs concurrently with myelination and follows a similar spatial gradient of progression from a proximal to distal direction along the regenerated nerve. Presumptive nodal sites appear prior to myelin formation and are identified as a prominent subaxolemmal density in thin sections and axonal particle patches in freeze-fracture. Following the appearance of presumptive nodes in regenerating fibres, dimeric particles are inserted into the axolemma adjacent to the node. These particles are in close apposition to the overlying Schwann cell terminal processes and with maturity adopt the same circumferential orientation seen in adult nodes. The nodal axolemma of regenerating fibres shows a characteristic increase in the prominence of its subaxolemmal densification and number of heterogeneously sized particles. Mature regenerated nodes demonstrate a complete annulus of nodal particles indistinguishable from control nodes. The results of the present study show that the nodal architecture of regenerating fibres is a faithful reconstruction of normal mature nodes, thus indicating that the morphological correlates associated with saltatory conduction at the node are present in regenerated nodes.

Freeze-fracture studies on the giant axon and ensheathing Schwann cells of the squid

The giant axons and encompassing sheaths from the stellar nerves of the squids Sepioteuthis sepioidea and Loligo forbesi have been analysed by freeze-fracture. The axolemma exhibits many intramembranous particles (IMPs) that fracture onto the cytoplasmic membrane half-leaflet (P-face); the larger IMPs may be aggregated into clusters. Axoplasmic subsurface cisternae are found beneath this membrane. Clustered or aligned arrays of P-face IMPs are also found on the membranes of the Schwann cells that intimately encapsulate the giant axons as well as 'capitate' projections of Schwann cells into the axons. When adjacent Schwann cells abut directly against one another, aligned E-face IMPs are found along the fracture plane of the upturning membranes. These E-face alignments of IMPs have complementary furrows on the Schwann cell membranes which exhibit no complementary structure on the axolemma as they represent the clefts between adjacent glial cells. The other Schwann cell membranes exhibit P-face dimples and E-face (extracellular membrane half-leaflet) protuberances which may reflect endo- or exocytotic activity; alternatively they may represent caveolae. Comparable structures are occasionally observed at axo-glial interfaces. However, those in the Schwann cell membrane could be part of the transverse tubular lattice system which also exists in adaxonal glia. Beyond the Schwann cells, layers of endoneurial cells (fibrocytes) are interleaved by collagen-filled spaces. These cells exhibit extensive cross-fractured intracellular invaginations as well as inpushings of the extracellular matrix material. Their membranes exhibit a large number of IMPs.

Effects of delayed myelination by oligodendrocytes and Schwann cells on the macromolecular structure of axonal membrane in rat spinal cord

The macromolecular structure of axonal membrane from dorsal funiculi of control and irradiated spinal cord of 45-day-old rats was examined with freeze-fracture electron microscopy. In control spinal cords, virtually all myelination is mediated by oligodendrocytes, and the internodal axonal membrane of these fibres displays highly asymmetrical partitioning of intramembranous particles (IMPs). The internodal P-face particle density is approximately 2350IMPs per micron 2, whereas the E-face IMP density is approximately 150 per micron 2. In control dorsal spinal roots, myelination is mediated by Schwann cells, and the ultrastructure of the internodal axolemma of the myelinated fibres is similar to that displayed by myelinated fibres of dorsal funiculi. On the internodal P-face of Schwann cell-myelinated fibres the IMP density is approximately 2350 per micron 2, whereas on the E-face the density is approximately 175 per micron 2. Irradiation of the lumbosacral spinal cord at 3 days of age results in a glial cell-deficient region within the spinal cord such that myelination in irradiated dorsal funiculi is delayed and subsequent myelination is mediated by both oligodendrocytes and Schwann cells. By 45 days of age, dorsal funiculi of irradiated spinal cords are well populated with fibres myelinated by oligodendrocytes and Schwann cells. However, fibres myelinated by oligodendrocytes display very thin myelin sheaths whereas Schwann cell-myelinated fibres exhibit myelin sheaths with normal thicknesses. Internodal membrane of fibres myelinated by Schwann cells and oligodendrocytes exhibit similar macromolecular structure, with approximately 2400 IMPs per micron 2 on P-faces and approximately 150 IMPs per micron 2 on E-faces. Occasional large (greater than 1.5 micron diameter) axons without glial-Schwann cell ensheathment are observed. These axons display a high density of P-face particles (approximately 2000 per micron 2) and a moderate density (approximately 350 per micron 2) of E-face IMPs on their fracture faces. These results demonstrate that CNS fibers exhibit similar axonal membrane ultrastructure irrespective of whether they are myelinated by Schwann cells or oligodendrocytes, or whether myelination is delayed. Moreover, when myelination does not occur, the axolemmal E-face IMP density, which may be related to the density of voltage-sensitive sodium channels, is not reduced.

Nodal and paranodal structural changes in frog optic nerve during early Wallerian degeneration

Ultrastructural changes in the nodal and paranodal regions of myelinated nerve fibres of frog optic nerves were studied during early stages of Wallerian degeneration. The earliest changes seen include retraction of paranodal loops of myelin from the axolemma and disconnection of paranodal myelin loops from myelin lamellae. These paranodal changes are asymmetric around the node and may be more advanced on either the proximal or distal side. Axoplasmic changes, including segregation of microtubules from neurofilaments, disorientation of microtubules and accumulation of abnormal organelles at nodes, appear shortly. In some axons the 'undercoating' along the widened nodal surfaces becomes patchy, and blebs appear in the nodal axolemma. In freeze-fracture replicas a mixture of particle clusters and particle-free areas appears in both E- and P-faces of the nodal axolemma. Blebs remain particle free. Initially, E-face particles remain segregated to the node and are present only at much lower concentrations in the demyelinated paranodal axolemma, suggesting that they are not freely mobile at this stage. Nodal E-face particles begin to decrease on day 5 associated with an increase in particles at the adjacent demyelinated paranode, and by day 11 the particle distribution is uniformly low over the entire extent of the nodal and demyelinated paranodal axolemma. If nodal E-face particles represent sodium channels, as has been proposed, the sequence of changes in Wallerian degeneration would be compatible with a gradual redistribution of nodal sodium channels into the demyelinated paranode.

Freeze-fracture characterization of proteolipid protein and basic protein of central nervous system myelin incorporated in liposomes

Proteolipid protein (PLP) and basic protein (BP) of central nervous system myelin were purified from calf brain white matter and incorporated in liposomes of L-dimyristoyl-alpha-phosphatidylcholine (DML) or in liposomes formed with an extract of natural lipids from myelin. Freeze-fracture replicas of the liposomes were prepared to study the number and size of intramembrane protein particles (IMP) in the fracture faces of the lipid bilayer. Globular and elongated IMP were observed in the freeze-fracture liposome membranes after incorporation of proteolipid protein. Globular IMP were the most frequently found (91-96% of the total IMP), and some of them showed a tiny black spot or pit on the top, suggesting the presence of hydrophilic channels in these particles. Globular and elongated IMP were also observed in the fractured membranes when basic protein was incorporated in liposomes. Again, globular IMP were the most frequent (92-95%) but no spots were present on the top. In addition, both globular and elongated IMP generated by basic protein were significantly larger than IMP generated by PLP. The proportion, size and form of globular and elongated particles generated by PLP and BP were unaffected by the amount of protein incorporated in liposomes (0.13-0.75 protein/lipid, w/w) nor by the type of lipid matrix used (DML or myelin natural lipid mixture). Intramembrane particles were absent from membranes of liposomes of pure lipid.

Relation between myelin sheath thickness, internode geometry, and sheath resistance

The thickness of the myelin sheath of nerve fibers was traditionally assessed solely as a function of axon caliber. Studies concerning the additional effect of variation in internode length are of relatively recent date. Carefully calibrated measurements of sheath thickness and internode geometry were used in this study to define an equation to predict the approximate number of lamellae from axon caliber and internode length, for normal and regenerated peripheral nerve fibers, and for fibers from hypomyelinated murine mutants. The definition of sheath thickness thus obtained was compared with different assumptions on the biophysical nature of myelin sheath resistance. The observed relations between sheath thickness and internode geometry were not compatible with an effective adjustment of sheath thickness to a radial flow of current across the sheath. Conversely, sheath thickness was found to vary in such a way that the resistance of the spiral path between the lamellae was matched precisely to axonal current density. The calculated resistance of the spiral leakage path, furthermore, was equal to measured sheath resistance. This new concept reconciles low sheath resistance with a high resistance of the myelin leaflet, yielding, at the same time, a fine tuning of sheath resistance to variations of internode geometry.

Ultrastructure of the squid axon membrane as revealed by freeze-fracture electron microscopy

The structure of the axolemma of the squid giant axon was studied by freeze-fracture electron microscopy. Three types of preparations were examined: intact axons, axons with their Schwann cell sheaths stripped off prior to freezing, and axons with their Schwann cell sheaths chemically detached but not mechanically removed. Because of a problem of cross-fracturing, the first two types of preparations revealed very few membrane faces of the axolemma. This cross-fracturing problem, however, was eliminated when we used a complementary replication method to fracture the third type of preparation. We found that the E-face of the axon membrane was smooth relative to the P-face, which showed many prominent intramembrane particles (IMP). The diameters of the typical IMP range from 6 to 15 nm. The P-face of the adjacent Schwann cells also showed many large IMP. The sizes and heights of the Schwann-cell IMP, however, appear to be more homogeneous than the P-face axolemma.

Congenital hypomyelinating neuropathy

Two patients with congenital hypomyelinating neuropathy are reported with details of sural nerve pathology. The resemblance of this condition to the hypomyelinating neuropathy of Trembler mice is discussed and the pertinent medical literature reviewed.

Ultracytochemical localization of ouabain-sensitive K+-dependent, p-nitrophenyl phosphatase in myelin

Ouabain-sensitive, K+-dependent p-nitrophenyl phosphatase (K-NPPase) activity was demonstrated ultracytochemically in the myelin of nerve fibers in peripheral and central white matter. Enzyme activity was more prominent in paranodal than compact myelin, and it was absent from nodal and interparanodal axolemma. Since K-NPPase is part of the Na-KATPase complex, we consider myelin as an important site of the sodium pump and believe that myelin participates in cationic regulation of the nervous tissue.

Membrane ultrastructure of developing axons in glial cell deficient rat spinal cord

In order to investigate axolemmal development in a glial cell deficient environment, normal and irradiated dorsal funiculus in rat lumbosacral spinal cord was examined by freeze-fracture electron microscopy. At 3 days of age, normal fibres are all unmyelinated and of small (less than 0.5 micron) diameter. The unmyelinated axons have a moderate density (approximately 850 microns-2) of intramembranous particles (IMPs) on P-fracture faces and a low IMP density (approximately 300 microns-2) on E-faces. IMPs are homogeneously distributed along both fracture faces. By 19 days of age, the normal dorsal funiculus is well populated with myelinated axons and glial cells, as well as a sizable population of unmyelinated fibres. Nearly all of the myelinated fibres have a large (greater than 1.0 micron) diameter; whereas, most unmyelinated axons are of small (less than 0.5 micron) calibre. The axolemma of unmyelinated axons is relatively undifferentiated, with an asymmetrical distribution of IMPs (P-face: approximately 1100 microns-2; E-face: approximately 450 microns-2). Myelinated fibres show nodal and paranodal regions with P-face and E-face ultrastructure similar to previous descriptions. Internodal axolemma appears relatively homogeneous, with P-faces being highly particulate (approximately 2100 microns-2) and a low IMP density (approximately 200 microns-2) on E-faces. Following irradiation of the lumbosacral spinal cord at 3 days of age, there is a severe reduction in the number of glial cells and myelinated fibres in this region when the tissue is examined at 19 days of age. Despite the deficiency of glial cells in this tissue, axonal and axolemmal development continue. Numerous large (greater than 1.0 micron) diameter axons are present in this irradiated tissue. Large diameter axons show a high (approximately 2000 microns-2) density of IMPs on P-faces; E-face IMP density remains at approximately 440 micron-2. Small calibre axons also have an asymmetrical distribution of particles (P-face: approximately 1100 microns-2; E-face: 280 microns-2). The axolemmal E-faces of some glial cell deprived fibres exhibit regions with greater than normal (approximately 750 microns-2) density of IMPs. These results demonstrate that some aspects of axonal and axolemmal development continue in a glial cell deficient environment, and it is suggested that axolemmal ultrastructure is, at least in part, independent of glial cell association.

Nodes of Ranvier and Schmidt-Lanterman incisures: an in vivo lanthanum tracer study

The permeability of the tight junctional system of myelin, at the juxtanodal myelin terminal loops and Schmidt-Lanterman incisures, was investigated using the ionic tracer lanthanum (a) in vivo followed by fixation, (b) concurrently with fixation, (c) following fixation. Employing the same methods the juxtanodal membrane complex formed between myelin loops and axolemma was also tested. The results of this study demonstrate that the periaxonal space (between axon and Schwann cell) is apparently accessible to lanthanum via the myelin loop-axolemmal junction, irrespective of the mode of exposure of myelinated fibres to the tracer. Similarly, the tight junctions between adjacent myelin terminal loops apparently do not prevent lanthanum penetration either in living or in fixed nerves. By contrast the tracer obtained access to the extracellular space within incisures only in vivo. The results are interpreted in terms of the permeability of nodes and incisures in vivo to physiologically important ions and related to current concepts of the electrophysiology of the myelinated nerve fibre.

Structural specializations in cat of chronically demyelinated spinal cord axons as seen in freeze-fracture replicas

Axons in chronically demyelinated spinal cord lesions, induced by ethidium bromide injection, display patches of intramembranous particles and indentations resembling nodal and paranodal axolemmal specializations respectively. Both occur in the marginal region of the lesions where the demyelinated axons are intimately associated with astrocytic and oligodendrocytic processes, and probably correspond to the aberrant node-like and paranodal junctional complexes seen in thin sections of this region. Demyelinated axons in the center of the lesion, which are not in contact with glial processes, do not display these membrane specializations.

Membrane specialization and axo-glial association in the rat retinal nerve fibre layer: freeze-fracture observations

The ultrastructure of non-myelinated ganglion cell axolemma within the retinal nerve fibre layer of adult rats was examined by thin section and freeze-fracture electron microscopy. Most of the axolemma within the nerve fibre layer does not exhibit any membrane specializations; intramembranous particles are partitioned with a density of approximately 1750 microns-2 on the P-fracture face and approximately 225 microns-2 on the E-face of the non-specialized axolemma. The nerve fibres also exhibit specialized foci of axolemma, at which the axons are abutted by the tips of blunt, radially oriented processes from Müller cells. At such sites of axo-glial association, an electron-dense undercoating is present beneath the axon membrane. Freeze-fracture analysis revealed a substantial increase in the density of E-face particles (greater than 500 microns-2) at sites of association between the tips of blunt glial processes and the axon. These findings demonstrate that non-myelinated axolemma of the retinal nerve fibre layer can exhibit spatial heterogeneity, with patches of node-like membrane at regions of specialized association with glial cell processes. On the basis of their morphological similarity to nodes of Ranvier, we suggest that these specialized axon regions represent foci of inward ionic current.

Axolemmal differentiation in myelinated fibers of rat peripheral nerves

In developing rat peripheral fibers, nodal specialization appears early, prior to myelin compaction, and is first detected as a junction between the axon and the overhanging Schwann cell process characterized by a uniformly wide (approximately 18 nm) intercellular gap containing a patchy dense substance and a cytoplasmic undercoating subjacent to the axolemma. The gap width is rather consistent but the axolemmal undercoating is more variable and lower in density than that found at more mature nodes of Ranvier, and it is also highly variable in length, ranging from 0.5 to 3 micron. The outermost Schwann cell layer is usually prominent with a large volume of cytoplasm and many organelles. In freeze-fracture replicas, modal specializations are characterized by accumulations of large (approximately 10 nm) particles in the axolemma, especially the E face, but immature nodes generally have a lower particle concentration than mature nodes. No node-like particle aggregates have been found in axons not intimately associated with Schwann cells. Mature paranodal axon-Schwann cell junctions are usually formed first by the loops closest to the node and are characterized by a 2-3 nm gap between the apposed membranes, periodic intercellular densities (transverse bands) in the gap and cisternae flattened against the junctional Schwann cell membrane. The loops further removed from the node display a wider gap containing irregularly spaced or diffuse intercellular densities, or none. Mature junctions appear relatively late in the rat, and it is not unusual to find developing nodes with several Schwann cell loops present that do not indent the axolemma significantly and are not associated with the paracrystalline pattern characteristic of the mature junctional axolemma. In such instances, the nodal particle aggregates do not have sharply circumscribed boundaries. The majority of the developing nodes are asymmetric with one paranodal segment more mature than the other.

Electron microscopic serial section analysis of nodes of Ranvier in lumbar spinal roots of the cat: a morphometric study of nodal compartments in fibres of different sizes

Serially sectioned nodes of Ranvier from nerve fibres 2-20 micron in diameter of feline ventral and dorsal spinal roots were examined electron microscopically, reconstructed to scale and analysed morphometrically. The assumed 'fresh-state' value of several structural variables, considered to be of functional significance, were calculated by the use of compensation factors. The compensated data were plotted against fibre and axon diameters. It was calculated that the membranous area of the 'fresh-state' nodal axon segment increased more or less exponentially from less than 5 micron2 to 30 micron2 with increasing fibre diameter (D). Most variables associated with the nodal gap and the Schwann cell initially increased rapidly with D and then levelled out or even decreased in fibres with a D value greater than 8-12 micron. The area open for communication between the nodal axolemma and the endoneurial space was 30-100 times smaller than the membrane area of the nodal axolemma. The volume of the extracellular space in the nodal gap, outside the nodal axolemma, increased linearly from less than 0.1 micron3 to about 0.6 micron3 with increasing fibre size. The Schwann cell membrane area facing the nodal gap outnumbered the membrane area of the nodal axon by 10-15 times in nerve fibres with a D value between 5 and 15 microns. Some functional implications of the 'fresh-state' nodal model are discussed.

Electron microscopic serial section analysis of nodes of Ranvier in lumbosacral spinal roots of the cat: ultrastructural organization of nodal compartments in fibres of different sizes

The general ultrastructural organization of nodes of Ranvier in peripheral nerve fibres from 2 to 20 microns in diameter (D) was investigated in the adult cat using serially sectioned ventral and dorsal spinal roots. The study was performed in order to collect and systematize information considered necessary for a morphometric analysis of the node of Ranvier. In all cases a node of Ranvier could be divided into a central nodal axon segment and a surrounding nodal Schwann cell compartment. The latter included a nodal gap matrix substance, more or less overlapping nodal Schwann cell collars and, as a rule, also a Schwann cell brush-border emanating from the nodal Schwann cell collars and occupying the nodal gap. The relative size and the organization level of the nodal Schwann cell compartment increased with increasing fibre size up to a fibre diameter of 8-10 microns. At this fibre size the nodal gap was of a fairly even height (1 micron) all around the nodal axon and contained a thick brush-border of densely packed, more or less radially arranged Schwann cell microvilli. In very small fibres (D less than 3 microns) the nodal gap was low (less than 0.1 microns) and contained no or few microvilli. In fibres greater than 10 microns in diameter the relative size and the degree of structural order of the nodal Schwann cell compartment decreased with increasing fibre size. Drastic sectorial variations in nodal gap height and local thinning-out of the brush-border became prominent features in the largest fibres. The possible in vivo organization of the nodal Schwann cell compartment is discussed. Preliminary calculations indicate that the extracellular space directly surrounding the nodal axon might be quite small and that the area open for free communication between this extracellular space and the endoneurial space might be very much restricted, measuring as little as 2% of the area of the nodal axolemma. Algorithms for calculating various nodal structural parameters are discussed.

Distortions of the nodes of Ranvier from axonal distension by filamentous masses in hexacarbon intoxication

A study has been made of the structural changes of nodal and paranodal regions of the nodes of Ranvier of peripheral nerves of rats in which marked accumulations of neurofilaments have occurred within axons under the influence of 2,5-hexanediol over 10 weeks. The neurofilamentous masses caused distension of the axon at two points of apparent weakness as they attempted to slide through the axonal constriction at the nodes. Principally, a spiral axonal protrusion pushed into the zone of unattached myelin loops in the proximal paranodal spinous bracelet of Nageotte. This led to a conical widening of the paranodal constriction and considerable attenuation of the overlying myelin. No degeneration of the myelin occurred however. Alternatively, or additionally, a protrusion occurred of the axon at the nodal region which increased the nodal gap width and occasionally compressed and displaced the adjacent distal paranodal constriction which could have led to some obstruction of axoplasmic flow. Swelling of distal paranodal regions occurred later and was usually associated with proximal swelling. It was also accompanied by evidence suggesting transnodal passage of filamentous material. Sometimes, however, striking nodal constriction occurred in association with symmetrical paranodal swelling. These observations suggest that the spiral glial-axonal relationships at nodes of Ranvier are capable of marked deformation that might allow the intra-axonal neurofilamentous masses to move distally. These findings are discussed in relation to the structural features of the paranodal constrictions.

Demyelination: a failure of cell communication?

A hypothesis is proposed that demyelination in both the CNS and PNS involves a failure of cell communication between the axon and oligodendrocyte/Schwann cell, as a primary event. The site of communication is assumed to be the paranodal myelin loop-axolemma membrane complex. It is postulated that "cross-talk" between the two cell types can be interrupted, and hence demyelination initiated, by pathophysiological changes in either the axon or myelinating cell. Experimental evidence in support of the hypothesis is cited in so far as it exists.

Rat optic nerve: freeze-fracture studies during development of myelinated axons

This freeze-fracture study examines the development of myelinated fibers in the rat optic nerve. Axolemma of optic nerve fibers were studied before, during, and after myelination. At birth, the optic nerve is composed entirely of non-myelinated (premyelinated) axons, while in the adult, virtually all fibers acquire compact myelin. Myelination begins at 6-8 days postparturition and proceeds rapidly, such that by 28 days of age approximately 85% of the axons are myelinated. The axolemma of premyelinated fibers from 2-day-old animals exhibits an asymmetrical partitioning of intramembranous particles (IMPs) between E- and P-fracture faces; the E-face had approximately 125 particles/micron2 and the P-face approximately 550 particles/micron2. Particle densities for premyelinated axolemma from 8, 12, 14, 16, and 28-day-old nerves were similar to those observed at 2 days. Beginning at 8-12 days postnatal, definitive association between oligodendroglial processes and axons (termed 'ensheathed' fibers) was observed. At the time of glial ensheathment, there was a 50-100% increase in the number of P-face particles; in contrast, the E-face did not display an overall increase in particle density. In certain regions, however, localized aggregations of E-face particles were observed. IMPs on P-faces of ensheathed axons had a greater mean particle size and higher percentage of 'large' (greater than 9.6 nm) particles than did IMPs on the corresponding fracture face of premyelinated fibers. Myelinated axons from 14-16 day optic nerves displayed several differences from adult myelinated fibers. The P-face of the internodal axolemma had approximately 45% fewer particles than that of adult internodal membrane, and the percentage of large IMPs on the P-face of the younger internodal membrane was approximately 50% of the value for adult internodal axolemma. E-faces of internodal axolemma from 14-16-day-old and adult animals had equivalent IMP densities and size distributions. The nodal region of myelinated axons from 14-16-day-old rats had fewer large particles on both E- and P-faces than did adult fibers, though particle densities on both fracture faces were similar for the two age groups. These studies demonstrate a clear reorganization of axon membrane structure concomitant with axo-glial ensheathment, followed by continued gradual axolemmal changes as myelination progresses.

Differentiation of the nodal and internodal axolemma in the optic nerves of neonatal rats

Axon plasma membranes (axolemma) were studied by freeze-fracture electron microscopy at stages prior to and during myelination in the optic nerves of neonatal rats. In unensheathed axons, intramembranous particles associated with the internal (P) and external (E) leaflets of the axolemma increased in number before reaching a plateau (approximately 600/micron2 in both leaflets) at about 9 days postnatally. In newly myelinated fibres, by contrast, the distribution of particles was asymmetrical; fewer particles (approximately 200/micron2) were found on the E-face and greater numbers (approximately 1400/micron2) were present on the P-face, distributions similar to those observed in mature myelinated fibres. Node-like aggregations of particles were not found in unensheathed pre-myelinated axons nor were they present in axons presumed to be ensheathed by glial cytoplasm but not yet myelinated, although nodal specializations could be easily identified in fibres with only a few turns of compact myelin. These observations show first that there is a redistribution of particles in the P- and E-faces of the internodal axolemma coincident with the onset of myelination and secondly, that nodal specializations (represented by the increased densities of E-face particles) appear after ensheathment but before the formation of compact myelin in fibres of the rat optic nerve.

Repetitive propagation of action potentials destabilizes the structure of the myelin sheath. A dynamic x-ray diffraction study

Time courses of myelin lattice swelling in toad sciatic nerves preexposed to different treatments were determined by x-ray diffraction using a one-dimensional position-sensitive detector. In the nerves supramaximally stimulated for 1 h at 200 Hz, the subsequent process of myelin swelling occurred 45.0 +/- 7.3 min (n = 24) sooner than in resting controls. Sciatic nerves incubated for 1 h in a Ringer's solution deprived of divalent cations (Ca++ and Mg++) exhibited a kinetics of swelling similar to that shown by the stimulated nerves, that is, 52.5 +/- 14.2 min (n = 6) sooner than controls preincubated for the same time in normal Ringer's solution (with divalent cations). The fact that both pretreatments supramaximal stimulation and removal of divalent cations from the perfusion solution produced a similar effect; namely, a decrease of the myelin lattice stability against swelling in distilled water, suggests that the repetitive propagation of action potentials could modify the ionic composition at either the intraperiod channel or the paranodal axoglial junction complexes.