Myelin intrusions in beaded nerve fibers

Neuroscience Vol. 36, No. 2, pp. 553-567, 1990

0306-4522/90$3.00+ 0.00 PergamonPress plc ~'~ 1990IBRO

Printed in Great Britain

M Y E L I N I N T R U S I O N S IN B E A D E D N E R V E FIBERS S. OCHS and R. A. JERSILDJR Departments of Physiology/Biophysicsand Anatomy, Indiana University School of Medicine, Indianapolis, IN 46223, U.S,A. Abstract--Small intrusions form in the internodes in or near the constrictions of beaded fibers prepared by fast-freezing and freeze-substituting mildly stretched nerves in the cat and the rat. They appear as inwardly directed folds of the inner lamellae of the myelin sheath, or regularly formed spheres composed of lamellae with major dense and interperiod lines like those of the myelin sheath. A splitting of the lamellae and separation of the major dense lines may occur with an accumulation of Schwann cell cytoplasm between them, the result of an influx of cytoplasmic fluid from nearby constrictions. Longitudinally oriented microtubules have been observed in the intrusions, in the adaxonal Schwann cell cytoplasm, and in the innermost lamellae of the myelin sheath. The paranodes contain a number of larger intrusions in the form of spurs and globules along with shelve-like folds of the myelin sheath oriented in the longitudinal direction. Axoplasmic fluid driven from the constrictions during beading can enter the paranodes to smooth out their folds leaving the globular and spur-shaped myelin intrusions in isolation. Their wall thickness, measured from the central opening to the surface of the intrusion, is the same as that of the myelin sheath or, in some cases, double, the result of the folding of a spur-like intrusion upon itself. Intrusions unconnected to the sheath are seen in unbeaded fibers with regular, compact lamellae surrounded by axolemma. Others lack a covering axolemma and consist of variably disorganized and irregularly shaped lamellae suggesting that they are undergoing fragmentation and dissolution within the axon. The hypothesis is advanced that the intrusions in the internodes arise from an excess of lipid and other myelin components when the diameter of the sheath is reduced in the beading constrictions. In the paranodes, excess myelin components moved into these regions form the shelf-like folds which may fuse to form intrusions. These, separated from the myelin sheath, undergo fragmentation and dissolution and are carried by retrograde transport to the cell bodies where their constituent components can be reutilized.

Myelinated nerve fibers are closer to a cylinder in cross-section as a crenation or cogwheeling of the shape in nerves fixed by fast-freezing and freeze-sub- fiber wall. 5'15'54'59'61Along with these shelf-like folds, stitution than when fixed with glutaraldehyde, t7'4~ various globular and spur-shaped intrusions are also Unlike routine glutaraldehyde fixation, freeze-substi- present. These stain like myelin and have been pictution also holds in place the labile shape of fibers tured from the earliest electron microscopic descriptermed beading which is elicited by a mild stretch. 39'42 tions of nerve fibers either without comment, ~s,z6 or Beading is seen as a series of unduloid enlargements referred to as "dark bodies". 46Intrusions described as and constrictions at intervals along the internodes of ovoids or tongue-like projections were described in the nerve fibers. A striking phenomenon connected normal nerve by Webster and Spiro 59 and as multiple with beading is the presence of small myelin-like myelin sheaths in the form of small rings or ovoids intrusions in and near the constrictions.42 These either contained within the myelinated fiber or, more intrusions arise quickly, within seconds between the often, lying independently within the Schwann cell imposition of the stretch to produce beading and cytoplasm by Dixon. H He found it difficult to account fast-freezing the nerve. The present study was ini- for the presence of such isolated myelin structures on tiated to better characterize these myelin intrusions in the basis of the mechanism of myelin sheath formaorder to understand the process by which they are tion described by Geren) 9 Myelin extrusions into the formed. These are to be differentiated from larger Schwann cell termed "droplets" were taken by intrusions present throughout the axon but more Berthold and Skoglun& to represent pinched off prominent in the paranodes. We have additionally portions of myelin sheath loops within the Schwann examined these larger myelin intrusions in the paran- cell, these subsequently undergoing disintegration. odal region using either glutaraldehyde or freeze-subA number of questions arise with regard to the stitution to fix the nerve fibers. relation of the shelf-like folds of the myelin wall in the The paranodal region adjoining the node of paranodes to the various myelin intrusions seen in Ranvier is often slightly bulged in shape and extends the axon. Also of interest and a challenge to interpresome 50 ~m from the node in the larger fibers. 15'61The tation is the lability of these structures. Some of the paranode differs from the rest of the internode6'16'43'6° variations seen come from the way the nerve is in having deep folds or shelves of the myelin wall handled and the method of fixation used. The in situ extending in the longitudinal direction and seen in observations of living fibers by Williams and Hall6° 553

554

S. Oc~Is and R. A. JERSILDJR

suggest that the paranodal structures do not change much over long periods of time when the nerves are not handled. Lubinska 35 found the myelin folds and intrusions in the paranodal region of single nerve fibers in vitro to increase with time following their isolation. These structures also increase following nerve compression/5 Intrusions are more often seen in the nerves of older animals. 3° They increase in fibers central to a distal lesion, 3° are prominent in Wallerian degeneration, ~2"27 in demyelinating diseases 9 and in various neuropathies and neuro-

toxicities 9,34,37,38,50 We were able to study the intrusions in the paranodes in isolation from the folds of the sheath using our earlier observation that beading constrictions do not occur in the paranodes. 42 When the nerve fibers become beaded, the influx of axoplasmic fluid expressed from the constrictions in the internode can enter into the paranodes to open out their folds leaving the intrusions in isolation which then can be held in place by freeze-substitution. EXPERIMENTAL PROCEDURES

The bulk of these studies was carried out on cat dorsal and ventral roots and peripheral nerves taken close to the L7 dorsal root ganglia. Rat sciatic nerves were also studied. These were prepared as described in a preceding paper. 42 In brief, segments of nerve or roots were removed from the animal for freeze-substitution, placed on a thin strip of brass shim stock and rapidly immersed in stirred Freon-12 cooled to a temperature close to -160cC with liquid N 2. The frozen segments were then placed in 2% osmium tetroxide in acetone at a temperature of - 60 to - 70°C for one to two weeks. They were then warmed and embedded in paraffin for light microscopy or in epon for electron microscopy. To observe the effect of beading, nerves or roots were maintained under a light stretch of 2 5 g and freeze-substituted as described above/2 Additionally, portions of nerve or roots were stretched and quick-frozen by the slam-freezing technique of Van Harreveld and Crowel157 using the Boyne7 modification of the slam-freeze apparatus. Glutaraldehyde fixation in situ was carried out by perfusion of glutaraldehyde via a micropipette into the cat L7 dorsal root ganglionfl either in vivo after exposure of the ganglia, or a few minutes after bleeding the animal. A desheathed nerve preparation was also used for immersion fixation. For this, the perineurium of the peroneal branch of cat sciatic nerve was removed 8 and pieces of nerve approximately 20 30 mm in length were ligated at both ends and pinned straight to a wax-bottomed trough A 3% glutaraldehyde solution in 0.1 N sodium cacodylate buffer, pH 7.4, replaced the Ringer solution in the trough and fixation carried out for 3 h. Small segments then were cut for a further fixation period of 1 h. Post-fixation was carried out using I% osmium tetroxide in the same buffer. All tissues prepared for electron microscopy were embedded in epon. Thin sections were stained with uranyl acetate and lead citrate. Serial paraffin sections were cut at 4-6 #m for light microscopy. RESULTS Intrusions within the interneuron related to beading constrictions

The beaded form of nerve fibers is shown by alternating expanded and constricted regions in the

internodes in longitudinal light microscopic sections of stretched fast-frozen and freeze-substituted nerve (Fig. 1). The unstretched nerve does not show beading (Fig. 2). In and near the constrictions a number of small intrusions can be detected by electron microscopy with varying shapes, many appearing as spherical bodies (Fig. 3). Their lamellae are arranged in concentric rings, rather than spiral wrappings, with the same alteration of major dense and interperiod lines as that of the myelin sheath. They are partially surrounded by the axolemma (Fig. 4). The spherical intrusions may appear disconnected from the myelin sheath or display a few of their outermost lamellae in continuity with it. A central core is generally present, often containing profiles indicating an origin from Schwann cell cytoplasm. Some intrusions appear to be forming from an inward folding into the axon of the innermost lamellae of the myelin sheath (Figs 5, 6). Splitting of the lamellae with separation of the membranes of the major dense and interperiod lines can be observed at the base of the inward folds of myelin intrusions (Figs 6, 7). Splitting is most commonly observed in the major dense lines with an accumulation of Schwann cell cytoplasm between the membranes (Fig. 7). Such separations of lamellae have been described earlier as leafing, This results from an influx of lamellar fluid expressed from nearby constricted regions. 42 Leafing and intrusions are most c o m m o n in an intermediate region of partial constriction between the most constricted region and the expanded areas, and are recognizable by the degree of microtubule and neurofilament compaction produced by beading. In electron micrographs of cross-sections of beaded fibers, the axon in regions of leafing with associated intrusions shows only a moderate increase in the density of microtubules and neurofilaments compared with an extreme compaction of the organelles in the more constricted regions/2 Electron-lucid regions frequently occur to the side of the intrusions (Fig. 7) which represent large accumulations of fluid as a result of the influx from nearby beading constrictions. The amount of lamellar myelin contributed by the sheath to the intrusion is variable with some intrusions consisting of only a few split and widely separated lamellae (Fig. 8). Other intrusions appear as finger-like profiles containing only a pair of unfused membranes covered by axolemma (Fig. 9). The membranes within these intrusions consist of the two normally unfused adaxonal Schwann cell membranes that surround the axon. Finger-like intrusions may extend for relatively long distances into the axon and show a complex branching pattern. Spencer and Thomas pictured a similar intrusive structure in axons having the form of sheet-like invaginations they considered could fold to entrap organelles/° The finger-like intrusions on occasion were more complex being associated with a spherical intrusion (Fig. 4B). The adaxonal membrane ends blindly

Fig. 1. Light micrograph of a longitudinal section of a stretched dorsal root, fast-frozen and substitutionfixed with osmium-acetone. Beading is seen as alternate enlargements and constrictions. Intrusions are present in the internodal (arrowhead) and paranodal regions (arrows). Cat L7 dorsal root. x 1400.

Fig. 2. Root similarly prepared but not stretched before fast-freezing. Beading is absent. Paranodal intrusions are present (arrows). The clear areas speckled throughout are due to ice crystals. These do not affect the overall form of the fibers. Cat L7 dorsal root. x 1400. 555

556

S. OCHS and R. A. JERSILD JR

Fig. 3. Electron micrograph of a beaded fiber near a constriction. The section contains three spherical intrusions each joined by only a few lamellae to the inner wall of the myelin sheath, and a grazing slice through a fourth. All the intrusions are outside the axon proper as indicated by their axolemmal covering. This and Figs 4-10 are slam-freeze preparations of cat L7 dorsal root. × 23,400.

Fig. 4. Spherical intrusions with a large central core of Schwann cell cytoplasm. Lamellae o f the intrusions and myelin sheath have the same periodicity. In B, a finger-like offshoot containing the adaxonal Schwann cell membranes arises from the outer part of the intrusion, x 66,000.

¢

7

Fig. 5. Intrusion consisting of a folded portion of the inner myelin lamellae protruding into the axon. The inner mesaxon (arrow) is separated from the sheath by an accumulation of fluid, x 42,000.

Fig. 6. Inward folding of the inner myelin sheath lamellae is accompanied by separation or leafing of the lamellae (arrowheads). The spherical intrusion at left contains a microtubule within the Schwann cell cytoplasm between the two innermost adaxonal membranes (left arrow). Its size matches that of axonal microtubules (right arrow). × 45,000. 557

558

S. OCHSand R. A. JERSILDJR

Fig. 7. Intrusion with folding and leafing of the lamellae. The major dense lines at the base shows splitting and the membranes are widely separated. Dense granular material fills the space between the membranes as expected of an influx of Schwann cell cytoplasm. Electron-lucid regions at the periphery of the intrusion represent an influx of fluid from a nearby constriction. The intrusion is covered with axolemma, x 81,000. within some intrusions, suggesting that they originate from the inner mesaxon (Figs 5, 8, 9).

Microtubule profiles in the myelin sheath Profiles with the appearance and dimension of longitudinally oriented microtubules were seen in the Schwann cell cytoplasm on the axonal surface of the myelin sheath and within intrusions. In intrusions, they occurred within the blind loop of the inner mesaxon (Figs 8, 9), in the cytoplasm between the inner, unfused adaxonal Schwann cell membranes and in the cytoplasm between the split major dense lines of the inner myelin sheath (Figs 6, 8, 9). They also have been observed in the relatively intact inner layers of the myelin sheath where leafing is minimal and intrusions are absent (Fig. 10).

Intrusions in the paranodes The redundant myelin structures within the para-

nodes of myelinated fibers of unstretched nerve or spinal roots prepared by freeze-substitution are readily apparent in longitudinal sections as dark, osmiumstaining, elongated spurs and folds on either side of the nodes (Fig. 2). In cross-sections of unstretched nerves or roots, the paranodal region of fibers can be identified by their crenated form (Fig. 11). The relative infrequency of crenated fibers is due to the comparatively short length of the paranodes with respect to that of the internodes. The paranodal region does not undergo the constrictions that occur in the rest of the fiber when it is beaded. 42 Where beading constrictions in the internodes are close enough to the paranodes, axoplasmic fluid can be driven into them causing their diameter to become enlarged. In serial cross-sections passing through the paranodes and nodes of fibers, the nodal region is identified by the loss of myelin as the sections pass through this region (Fig. 12). The

Myelin intrusions

559

Fig. 8. Dome-shaped intrusion containing three structures representing splitting of the major dense lines. A fourth structure, the inner mesaxon, ends blindly. Microtubular profiles present in three of the structures (arrows) are similar to those within the axon. × 66,000. swollen paranodes nearby show a larger g ratio of 0.8 or more, the diameter of the axon to that of the fiber: 8'51"52 compared with the normal g ratio of approximately 0.6. The larger g values found for these paranodes with the lack of the usual crenated form is consistent with an influx of axoplasmic fluid smoothing out the folds and increasing its perimeter without changing the thickness of the sheath. Intrusions which take the form of spurs arise from the myelin sheath in the paranodes and then extend axially in the internodes within the axoplasm unattached to the sheath for varying distances up to 50/am or more. These intrusions generally are directed away from the nodes, though on occasion some point toward the node and even enter into the nodal region. More than one intrusion may be observed in a paranode. These intrusions are not the result of handling of the nerve or root in the process of removing them for stretching to induce beading. Similar globular or spur-like intrusions within enlarged paranodes were present within fibers of nerves prepared by fast-freezing in situ while the limbs were held in extension to place their nerves under stretch.

Some preliminary observations were made with respect to the node. The node is usually reported to be constricted, to as little as 20-30% of the diameter of the internode. 6'~6'31:3 When the paranodes are swollen in the beaded fibers, the node may also appear to increase in diameter, to approximately 70-80% of the internode (Fig. 12). This observation requires a more extensive electron microscopic investigation. The intrusions in the paranodes often show a clear central opening. The wall thickness of the intrusion measured from its central opening to its outer surface is usually similar to that of the myelin sheath [Fig. 12 (frame 3B)]. In some cases the wall thickness appears to be twice that of the myelin sheath, apparently the result of a folding of the intrusion which first appears as an inward closed pouch of the myelin wall [Figs 13 (frames 2C-3B) 14 (frames 2C-3A)]. The intrusions are composed of lamellae that are circumferentially organized and continuous with those of the myelin sheath. The outer surface is covered by the axolemma where it projects into the axon (Figs 15, 16). When present in the plane of section, the central opening in the intrusion shows a small amount of Schwann cell cytoplasm (Fig. 16).

560

S. Ocns and R. A. JERSILDJR

Fig. 9. Finger-like intrusions in continuity with the adaxonal membranes of the parent myelin sheath and the mesaxon. Cross-sectioned profiles of microtubules, similar to those in the axon, and microfilaments are present (arrows). The intrusion is covered by the axolemma, x 73,000. In general, the lamellae of the intrusions are compact and similar in organization to that of the myelin sheath (Fig. 15). In a few cases where the myelin sheath is loosely structured, the intrusion also shows a similar lack of compact structure (Fig. 16). Other intrusions, however, are radically different and show a marked disorganization in contrast to a compact and regularly lamellated myelin sheath (Fig. 17). Some of these intrusions show a more disorganized

periphery and depart from a globular shape (Fig. 18). These irregular intrusions are not covered with an axolemma. DISCUSSION The intrusions of the myelin sheath seen in and near beading constrictions differ from the larger globular intrusions which may be seen in the

Fig. 10. (A) A single microtubule (arrow) in the adaxonal cytoplasm of the myelin sheath in an area free of intrusions. (B) A single microtubule (arrow) in the area of a split major dense line. × 84,000.

Fig. 11. Cross-section of a fascicle from an unbeaded dorsal root. Crenated fibers (arrows) are seen interspersed between approximately circular myelinated fibers. Freeze-substitution. This and Figs 12 14 are freeze-substitution preparations of cat L7 dorsal root. × 1900.

A

B

C

D

E

2

Fig. 12. Serial cross-section through the paranodal regions on each side of a node from a beaded nerve fiber. Two intrusions are present on the one side (frames IB 2C) of the node. The node is indicated by the loss of myelin which is almost complete in frame 2D. An intrusion is also present on the opposite side of the node (frames 3A42). A central opening is seen in this intrusion (frame 3B). The thickness of the intrusion wall, measured from the central opening to the outside of the intrusion, is similar to the myelin wall thickness, x 770.

A

B

C

D

E

F

1

3

4

5 Fig. 13. Serial cross-section of a beaded nerve. The origin of the intrusion is revealed in frames 2 D - F as a fold in the inner myelin wall doubling back on itself. This intrusion does not show any tubular extension in the longitudinal direction. The node is near frames 4A and B. The paranode on the opposite side shows some furrowing, x 770.

A

B

C

D

1

2

3 J

4 Fig. 14. Serial cross-sections through a nodal region. In sections proximal to its origin (frames IB-D), the intrusion is seen as a ring with a central core of cytoplasm. Note the extra thickness of the myelin intrusion in frames 1D-2C in comparison with that of the myelin sheath. Frame 3C represents the section taken through the node which has a portion of the intrusion in it. The paranodal region on the distal side shows furrowing (frames 4B D), This could be the result of constrictions on the one side moving more fluid into the paranode than those on the other side. x 770.

~ • ~i!~i~:¸

Fig. 15. Electron micrograph o f a cross-section o f an unstretched nerve fiber showing the usual departure from circularity characteristic o f glutaraldehyde fixed fibers) 7 The intrusion is broadly attached at its base to the myelin sheath. This and Figs 1(~18 are cat nerves close to the L7 dorsal root ganglion prepared by glutaraldehyde perfusion, x 15,000.

~ '¸F~ ,

Fig. 16. Cross-section of an intrusion showing lamellae passing directly into the sheath. The lamellae in this case shows a mixture of compact and loose regions in both the sheath and the intrusion. The axolemma is reflected over the surface o f the intrusion. A central core of Schwann cell cytoplasm is present, x 26,000.

C

t

Fig. 17. Intrusion unattached to the myelin sheath displaying an irregular myelin pattern. An axolemma around the intrusion is not evident. × 11,400.

Fig. 18. Intrusion with a more disorganized appearance of the myelin at its periphery compared with that in Fig. 17. x 11,400. 564

Myelin intrusions internodes and from those in in vitro preparations where they were reported to develop over several hours of observation. In time-lapse microscopic observations, Gitlin and Singer2~ observed the Schmidt-Lanterman clefts to undergo an "opening" and "closing" with an inward bulging of part of the myelin sheath into the axon, at times taking the form of spur-like intrusions such as we have seen in our studies of the paranodes. These changes, however, develop slowly compared with the rapid elicitation, within seconds, of the intrusions associated with the constrictions, the time taken to stretch nerves to produce the beading and to fast-freeze them. The beading intrusions also are not related to the incisures. The small intrusions, visible as spheres or folds of part of the myelin sheath in or near the constrictions of the beaded nerve fibers, appear to be produced as a result of the rapid decrease in the diameter of the myelin sheath in the beading constrictions. The reduction of the fiber diameter in the constrictions, as previously described,42 occurs without a change in the overall thickness of the sheath or the periodicity of the lamellae. This requires a reduction of the total content of lipid and other myelin membrane components within the constricted regions of the myelin sheath. For this to occur, the lipid and other lamellar components must move from the constricted regions in the longitudinal direction to the non-constricted regions of the beaded fibers within the plane of their lamellar membranes, a movement accounted for by the fluidity of the myelin in the lamellae.42 Additionally, the temporary surplus of myelin components in the constrictions, which is greatest in the inner layers of the myelin sheath where the circumference is smaller, acts as a force which is radially directed inwardly to produce the intrusions seen as folds and spherical profiles of the myelin lamellae. Their concentric lamellar organization is consistent with their origin as intrusions from the myelin sheath. Parts of the inwardly folded intrusions contain lamellae that are widely separated (Fig. 7), the separation resulting from an influx of Schwann cell cytoplasmic fluid from nearby constrictions. As much as 30--50% of the myelin sheath consists of w a t e r 47 distributed within the cytoplasmic and extracellular spaces of the lamellae29 and this too must be expressed from the constrictions. The fluid moving in from the constrictions splits the major dense lines of the lamellae into their respective membranes to give rise to a picture typical of leafing,a-" Some of the fluid also forms large electron-lucid areas outside the axolemma (Figs 5, 7, 8). The spherical intrusions could arise from myelin folds by a process of membrane fusion similar to that occurring in exocytosis or endocytosis. In one model of fusion, non-bilayer regions of lipid are proposed to exist in the membrane, these consisting of phospholipid species forming hexagonal phase II lipids able to give rise to inverted micelles. '°'58 Where their membranes meet, the micelles participate in their fusion.

565

This process, carried successively through the layers of membranes at the base of the folds, could produce the lamellated spherical intrusions. This process could be facilitated by the splitting of the myelin lamellae. Alternatively, the lipid in the spherical intrusion may become reorganized into multilayered lamellae in a process similar to the formation of multilaminar liposomes? ,36 The intrusions which appear as a narrow sheet or finger-like projection (Figs 8, 9) contain only the innermost adaxonal membranes of the Schwann cell which normally are not fused. This may reflect some special molecular composition of those membranes which prevents their fusion. Microtubules are readily found in the Schwann cell cytoplasm outside the myelin sheath where they generally run in the longitudinal direction. It is also known that microtubules are present in the myelin sheath wound spirally within the paranodal loops. 44 Their presence there was suggested to serve as "tracks" to transport myelin basic proteins during myelination in the radial direction. 53 Additionally, a helical disposition of microtubules within the Schmidt-Lantermann incisures of the internodes has been described. 24 Our observations of longitudinally directed microtubules within the intrusions and in the adaxonal cytoplasm of Schwann cells has not been described previously, though pictured without comment in one other known source? s The longitudinally directed microtubules could serve for the transport of myelin components in the axial direction within the sheath of the Schwann cell just as microtubules subserve axoplasmic transport within the axon# ° This possibility must be tempered by noting the relative spareness of the microtubules in these sites. Considering the intrusions in the internodes to result from a temporary augmentation of myelin and a local reorganization, a similar explanation may account for the origin of the globular and spurshaped intrusions in the paranodes. A surplus of myelin in the paranodes may come about by metabolic processes. While the myelin of the adult had previously been considered to be relatively inert, 3A3'23'33 m o r e recent studies with labeled myelin components have shown myelin to have a greater level of metabolic activity.22'25'4~ Studies using [3H]glycerol to label myelin phospholipids and other lipid precursors suggest a transport of these components in the axon entering successively the incisures of Schmidt-Lanterman, the inner myelin leaflets of the myelin sheath and then with time passing longitudinally within the myelin sheath into the paranodes. 13"~4'33'55Other reports have supported the concept of an axonal transport supply of components to Schwann cell myelin. ',2°,32. However, the extra lipid forming the paranodal intrusions is derived, whether from the Schwann cell or in some part from the axon, a surplus of myelin components in the paranodes suggests the following sequence of events as an hypothesis: excess of myelin

566

S. OcHs and R. A. JERSILDJR

in the paranodes leads to the inward folding of the myelin sheath. By the merging of the folds through membrane fusion or reorganization of their lamellae, globular and spur-like intrusions are formed which at first are extra-axonal. These intrusions become pinched off within the axons, lose their axolemma to undergo fragmentation and dissolution in the axon where they are carried back toward the cell bodies by retrograde transport for utilization of their components. They may contribute in part to the multilamellated bodies observed to be retrogradely transported in the a x o n f l '56 Such an hypothesis has interesting possibilities with respect to interactions

known to occur between the axon and the Schwann cell warranting further investigation along those lines.

Acknowledgements--We wish to thank Kathy WelbornCraig and Patricia Summerlin for their technical assistance in these experiments, Vera McAdoo for assistance with the light microscopic and electron microscopic preparations, Clint Myers and Carol Chueden for help with the illustrations, and Dr James McAteer for the use of the Boynetype slam-freeze apparatus. These studies were supported by NIH PHS ROI NS 8706, NSF BNS 82-17727 and the Biomedical Research Committee of the Indiana School of Medicine.

REFERENCES

1. Alberghina M., Viola M. and Giuffrida A. M. (1982) Transfer of axonally transported phospholipids into myelin isolated from the rabbit optic pathway. Neurochem. Res. 7, 139-149. 2. Bangham A. D., Hill M. W. and Miller N. G. A. (1974) Preparation and use of liposomes as models of biological membranes. In Methods in Membrane Biology (ed. Korn, E. D.), Vol. l, pp. 1-68. Plenum Press, New York. 3. Benjamin J. A. and Smith M. E. (1984) Metabolism of myelin. In Myelin, 2nd edn (ed. Morell P.), pp. 225-258. Plenum Press, New York. 4. Berthold C.-H. and Skoglund S. (1968) Postnatal development of feline paranodal myelin-sheath segments. II Electron microscopy. Acta Soc. Med. upsal. 73, 127 144. 5. Berthold C.-H. (1978) Morphology of normal peripheral axons. In Physiology and Pathology of Axons (ed. Waxman S. G.), pp. 3-63. Raven Press, New York. 6. Berthold C.-H. (1982) Some aspects of the ultrastructural organization of peripheral myelinated axons in the cat. In Axoplasmic Transport (ed. Weiss D. G.), pp. 39 54. Springer, New York. 7. Boyne A. F. (1979) A gentle, bounce-frec assembly for quick-freezing tissues for electron microscopy: application to isolated torpedine ray electrocyte stacks. J. Neurosci. Meth. 1, 353-364. 8. Chan S. Y., Ochs S. and Worth R. M. (1980) The requirement for calcium ions and the effect of other ions on axoplasmic transport in mammalian nerve. J. Physiol., Long. 301, 477 504. 9. Collins G. H., Webster H. de F. and Victor M. (1964) The ultrastructure of myelin and axonal alterations in sciatic nerves of thiamine deficient and chronically starved rats. Acta Neuropath. 3, 511-521. 10. Cullis P. R. and De Kruijff B. (1979) Lipid polymorphism and the functional roles of lipids in biological membranes. Biochem. Biophys. Acta 559, 399-420. 11. Dixon A. D. (1962) Multiple myelin sheaths in single Schwann cells. Nature 193, 1004-1005. 12. Donat J. R. and Wisniewski H. M. (1973) The spatio-temporal pattern of Wallerian degeneration in mammalian peripheral nerves. Brain Res. 53, 41-53. 13. Droz B., Brunetti M., Giamberardino L. D., Koenig H. L. and Porcellati G. (1985) Selective distribution of axonally transported phospholipids to nerve endings and/or myelin: the case of ethanolamine glycerophospholipids. In Phospholipids in the Nervous System, (ed. Horrocks L. A.), Vol. 2, pp. 315-327. Raven Press, New York. 14. Droz B., Giamberardino L. and Koenig H. L. (1981) Contribution of axonal transport to the renewal of myelin phospholipids in peripheral nerves. I. Quantitative radioautographic study. Brain Res. 219, 57 71. 15. Dullmann J. and Wulfhekel U. (1970) ~ber Schmidt-Lantermannsehe Einkerbungen und Paranodale Bulbi des N. ischiadicus. Z. Anat. Entw. Gesch. 132, 350 358. 16. Ellisman M. H., Wiley C. A., Lindsey J. D. and Wurtz C. C. (1984) Structure and function of the cytoskeleton and endomembrane systems at the node of Ranvier. In The Node of Ranvier (eds Zagoren J. C. and Fedoroff S.), pp. 153-181. Academic Press, New York. 17. Friede R. L. and Bischausen R. (1982) How are sheath dimensions affected by axon caliber and internode length? Brain Res. 235, 335-350. 18. Gasser H. S. (1952) Discussion of Frankenhaeuser's paper. Cold Spring Harbor Symp. quant. Biol. 17, 32-36. 19. Geren B. B. (1954) The formation from the Schwann cell surface of myelin in the peripheral nerves of chick embryos. Expl Cell Res. 7, 558-562. 20. Giorgi P. P. and DuBois H. (1981) Labelling by axonal transport of myelin-associated proteins in the rabbit visual pathway. Biochem. J. 196, 537-545. 21. Gitlin G. and Singer M. (1974) Myelin movements in mature mammalian peripheral nerve fibers. J. Morph. 143, 167 186. 22. Gould R. M, and Dawson R. M. C. (1976) Incorporation of newly formed lecithin into peripheral nerve myelin. J. Cell Biol. 68, 480-496. 23. Gould R. M., Spivack W. D., Sinatra R. S., Lindquist T. D. and Ingoglia N. A. (1982) Axonal transport of choline lipids in normal and regenerating rat sciatic nerve. J. Neurochem. 39, 1569-1578. 24. Hall S. M. and Williams P. L. (1970) Studies on the "incisures" of Schmidt and Lanterman. J. Cell Sci. 6, 767-791. 25. Hendelman W. J. and Bunge R. P. (1969) Radioautographic studies of choline incorporation into peripheral nerve myelin. J. Cell Biol. 40, 190. 26. Hess A. and Young J. Z. (1952) The nodes of Ranvier. Proc. R. Soc. B. 140, 301-320. 27. Honjin R. and Takahashi A. (1962) Electron microscopy of nerve fibers VI. On the myelin changes in the peripheral myelinated nerve fibers during Wallerian degeneration. J. Electron Microsc. 11, 139-156. 28. Kelly D. E., Wood R. L. and Enders A. C. (1984) Bailey's Textbook of Microscopic Anatomy, 18th edn. Williams and Wilkins, Baltimore.

Myelin intrusions

567

29. Kirschner D. A., Ganser A. L. and Caspar D. L.D. (1984) Diffraction Studies of Molecular Organization and Membrane Interactions in Myelin, 2nd edn (ed. Myelin P. M.), pp. 51-95. Plenum Press, New York. 30. Krinke G., Froehlich E., Herrmann M., Schnider K., Da Silva F., Suter J. and Traber K. (1988) Adjustment of the myelin sheath to axonal atrophy in the rat spinal root by the formation of infolded myelin loops. Acta Anat. 131, 182 187. 31. Landon D. N. and Hall S. (1976) The myelinated nerve fiber. In The Peripheral Nerve (ed. Landon D. N.), pp. 1-105. John Wiley, New York. 32. Ledeen R. W. and Haley J. E. (1983) Axon-myelin transfer of glycerol-labeled lipids and inorganic phosphate during axonal transport. Brain Res. 269, 267 275. 33. Ledeen R. W., Kunishita T., Wu P. S., Haley J. E. and Novak G. P. (1985) Phospholipid synthesis in myelin: putative role of the axon. In Phospholipids in the Nervous System (ed. Horrocks L. A.), pp. 329-340. Raven Press, New York. 34. Love S. and Cruz-Hofling M. A. (1986) Acute swelling of nodes of Ranvier caused by venoms which slow inactivation of sodium channels. Acta Neuropath. Berlin 70, 1-9. 35. Lubifiska L. (1961) Sedentary and migratory states of Schwann cells. Expl Cell Res. Suppl. 8, 74-90. 36. Lucy J. A. (1975) The fusion of cell membranes. In Cell Membranes--Biochemistry. Cell Biology and Pathology (eds Weissmann G. and Claiborne R.), pp. 75-83. HP Publishing, New York. 37. Moore S. A., Peterson R. G., Felten D. L. and O'Conner B. L. (1982) Ultrastructural axonal pathology in experimentally diabetic and aging control rats. Brain Res. 81, 317-323. 38. Mufioz-Martinez E. J., Massieu D. and Ochs S. (1984) Depression of fast axonal transport produced by Tullidora. J. Neurobiol. 15, 375-392. 39. Ochs S. (1965) Beading of myelinated nerve fibers. Expl Neurol. 2, 84 95. 40. Ochs S. (1982) Axoplasmic Transport and its Relation to Other Nerve Functions, pp. 1-462. John Wiley-lnterscience, New York. 41. Ochs S., Erdman J., Jersild R. A. Jr and McAdoo V. (1978) Routing of transported materials in the dorsal root and nerve fiber branches of the dorsal root ganglion. J. Neurobiol. 9, 465 481. 42. Ochs S. and Jersild R. A. Jr (1987) Cytoskeletal organelles and myelin structure of beaded nerve fibers. Neuroscience 22, 1041-1056. 43. Okamura Y. and Tsukita S. (1986) Morphology of freeze-substituted myelinated axon in mouse peripheral nerves. Brain Res. 383, 146-158. 44. Peters A., Palay S. L. and Webster H. de F. (1976) The Fine Structure o f the Nervous System: The Neurons and Supporting Cells, 2nd edn, Vol. 2, pp. 406-440. Saunders, Philadelphia. 45. Rawlins F. A. (1973) A time-sequence autoradiographic study of the in vivo incorporation of [l,2-3H]cholesterol into peripheral nerve myelin. J. Cell Biol. 58, 42 53. 46. Robertson J. D. (1960) The molecular structure and contact relationships of cell membranes. Prog. Biophys. 10, 344-418. 47. Schmitt F. O., Bear R. S. and Palmer K. J. (1941) X-ray diffraction studies on the structure of the nerve myelin sheath. J. Cell comp. Physiol. 18, 31-42. 48. Schroder J. M. (1972) Altered relation between axon diameter and myelin sheath thickness in regenerated nerve fibers. Brain Res. 45, 49-65. 49. Smith R. S. (1980) The short term accumulation of axonally transported organelles in the region of localized lesions of single myelinated axons. J. Neurocytol. 9, 39-65. 50. Spencer P. S. and Thomas P. K. (1974) Ultrastructural studies of the dying back process. II. The sequestration and removal by Schwann cells and oligodendrocytes of organelles from normal and diseased axons. J. Neurocytol. 3, 763-783. 51. Sunderland S. (1978) Peripheral nerve fibers. In Nerves and Nerve Injuries, 2nd edn (ed. Sunderland S.) Churchhill Livingstone, Edinburgh. 52. Sunderland S. and Roche A. F. (1958) Axon-myelin relationships in peripheral nerve fibers. Acta Anat., Basel 33, 1-37. 53. Tabira T. and Webster H. de F. (1979) E-PTA stains oligodendroglial surface membranes and microtubules in optic nerves during myelination. J. Neurol. Sci. 42, 215-227. 54. Thomas P. K. and Ochoa J. (1984) Microscopic anatomy of peripheral nerve fibers. In Peripheral Neuropathy, 2nd edn (eds Dyck P. J., Thomas P. K., Lambert E. H. and Bunge R.), pp. 39-96. Saunders, Philadelphia. 55. Toews A. D. and Morell P. (1985) Axonal transport of phospholipids. In Phospholipids in the Nervous Sw~tem (ed. Horrocks L. A.), Vol. 2, pp. 299-314. Raven Press, New York. 56. Tsukita S. and Ishikawa H. (1980) The movement of membranous organelles in axons. Electron microscopic identification of anterogradely and retrogradely transported organelles. J. Cell Biol. 84, 513-530. 57. Van Harreveld A. and Crowell J. (1964) Electron microscopy after rapid freezing on a metal surface and substitution fixation. Anat. Rec. 149, 381-385. 58. Verkleij A. J., Leunissen-Bijvelt J., De Kruijff B., Hope M. and Cullis P. R. (1984) Non-bilayer structures in membrane fusion. In Cell Fusion (Ciba Foundation Symposium 103) (eds Evered D. and Whelan J.), pp. 45-59. Pitman, London. 59. Webster H. de F. and Spiro D. (1960) Phase and electron microscopic studies of experimental demyelination. J. Neuropath. exp. Neurol. 19, 42-69. 60. Williams P. L. and Hall S. M. (1970) In vivo observations on mature myelinated nerve fibers of the mouse. J. Anat. 107, 31 38. 61. Williams P. L. and Landon D. N. (1963) Paranodal apparatus of peripheral nerve fibers of mammals. Nature, Lond. 198, 670 673. (Accepted 27 November 1989)