Compositional analysis of growing axons from rat sympathetic neurons.

We describe culture systems for neurons of an adrenergic autonomic ganglion which: (a) permit cultivation of neurons without supporting cells, (b) permit separate harvest of somal and axonal material, and (c) permit direct access to the neuronal surface. The antimetabolites used to suppress supporting cell growth did not have demonstrable effects on neuronal polypeptide synthesis. Rapid neurite outgrowth, which characterized these cultures, was prevented by colchicine or cycloheximide and resumed promptly after their withdrawal. Axons separated from cell bodies showed no incorporation of label from leucine or fucose, but did exhibit incorporation of glucosamine. The major polypeptides present in this neuron, as demonstrated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis, are described. No major differences in polypeptide content were observed when soma and axons were compared. Likewise, there were no differences detected in polypeptides synthesized by neurons in suspension or neurons actively extending processes. Analysis of the polypeptides within the neurites after labeling with amino acids indicated transport at a number of different rates; certain of these polypeptides corresponded in size and transport characteristics to polypeptides observed in the rabbit optic nerve after labeling of retinal ganglion cells. Tubulin and actin have been definitively identified in this cell type (18); we found proteins similar in size and proportionate amounts to be among the rapidly transported soluble polypeptides. The prominent polypeptides observed after several methods of surface labeling are described.

Evidence that contact with connective tissue matrix is required for normal interaction between Schwann cells and nerve fibers.

Explants of fetal rat sensory ganglia, cultured under conditions allowing axon and Schwann cell outgrowth in the absence of fibroblasts, occasionally develop nerve fascicles that are partially suspended in culture medium above the collagen substrate. In these suspended regions, fascicles are abnormal in that Schwann cells are decreased in number, are confined to occasional clusters along the fascicle, provide ensheathment for only a few axons at the fascicle periphery, and do not form myelin. When these fascicles are presented with a substrate of reconstituted rat-tail collagen, Schwann cell numbers increase, ensheathment of small nerve fibers occurs normally, and larger axons are myelinated. We conclude that, for normal development, Schwann cells require contact with extracellular matrix as well as axons. The Schwann cell abnormalities in suspended fascicles are similar to those observed in nerve roots of dystrophic mice.

Factors influencing the release of proteins by cultured Schwann cells.

Cultured rat schwann cells grown in association with sensory neurons when labeled with [(3)H]leucinem, [(3)H]glucosamine, or [(35)S]methionine release labeled polypeptides into the culture medium. Analysis by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) of the culture medium reveals a reproducible pattern of more than 20 polypeptides with molecular weights ranging from 15,000 to more than 250,000. Five major polypeptides (apparent molecular weights 225,000, 210,000, 90,000, 66,000, 50,000, and 40,000) account for approximately 40 percent of the leucine or methionine radioactivity in medium polypeptide. Schwann cells grown in a serum-free defined medium, in which schwann cells do not relate normally to axons, release approximately four times less labeled medium polypeptides tha cultures grown in medium supplemented with serum and chick embryo extract. In addition, there is a qualitative difference in the pattern of medium polypeptides resolved by SDS-PAGE, so that a single polypeptide (mol wt 40,000) accounts for nearly all of the label in medium polypeptides. Switching of cultures grown in defined medium to supplemented medium for 2 d results in a fourfold increase in the amount of labeled polypeptides appearing in the culture medium, and a return to the normal pattern of medium polypeptides appearing in the culture medium, and a return to the normal pattern of medium polypeptides as resolved by SDS-PAGE. This change in the pattern of polypeptides release by schwann cells is accompanied by changes in the association between schwann cells and axons. An early step in the establishment of normal axon-schwann cell relations appears to be an inward migration of schwann cells into axonal bundles and spreading of schwann cells along neurites. These changes are evident within 48 h after medium shift. Our results thus suggest that the release of proteins by schwann cells may be important for the development of normal axonal ensheathment.

A neuronal cell surface heparan sulfate proteoglycan is required for dorsal root ganglion neuron stimulation of Schwann cell proliferation.

Axons of dorsal root ganglion neurons express on their surfaces one or more proteins which are mitogenic for Schwann cells (Salzer, J., R. P. Bunge, and L. Glaser, 1980, J. Cell Biol., 84:767-778). Incubation of co-cultures of dorsal root ganglion neurons and Schwann cells with 4-methylumbelliferyl-beta-D-xyloside, an inhibitor of proteoglycan biosynthesis, decreases the mitogenic response of the Schwann cell by over 95%. The effect of the beta-D-xyloside has been localized to the neurons; pretreatment of neurons but not of Schwann cells with the inhibitor causes a marked reduction of the mitogenic response. In addition, Schwann cells treated with beta-D-xyloside are still mitogenically responsive to soluble Schwann cell mitogens (cholera toxin and glial growth factor). Neurons treated with heparitinase and membrane vesicles prepared from heparitinase-treated neurons show diminished mitogenicity for Schwann cells, while other proteoglycan lyases have no effect. We conclude that a cell surface heparan sulfate proteoglycan is a component of the Schwann cell mitogen present on the surface of dorsal root ganglion neurons.

PC12 cells as a source of neurite-derived cell surface mitogen, which stimulates Schwann cell division.

Primary cultures of rat dorsal root ganglion Schwann cells were used to assay the efficacy of PC12 cells in stimulating Schwann cell proliferation. Co-cultures of PC12 cells and Schwann cells assayed by [3H]thymidine labeling followed by autoradiography showed proliferation of Schwann cells only where contact occurred between PC12 neurites and Schwann cells. Membranes derived from PC12 cells were shown to have many characteristics similar to membranes derived from sensory neurons; both could mimic whole cells in stimulating Schwann cell division; both were inactivated by mild heat treatment and by trypsinization, and both elevated intracellular cyclic AMP concentrations in Schwann cells 16 h after addition of membranes. We conclude that PC12 cells will be a valuable source for the isolation of the neuronal cell surface component which controls proliferation of Schwann cells during development of the peripheral nervous system.

Movements of the Schwann cell nucleus implicate progression of the inner (axon-related) Schwann cell process during myelination.

Although it has been known for several decades that peripheral myelin is formed from an extended, spiraled, and compacted sheet of Schwann cell (SC) plasma membrane, the mechanism by which this unique spiraling is accomplished remains unknown. We have studied the movements of SC nuclei before, during, and subsequent to myelin formation (over periods of 24-72 h) to determine if this nuclear motion (noted in earlier reports) would provide useful insights into the mechanism of myelinogenesis. We used rodent sensory neuron and SC cultures in which initiation of myelinogenesis is relatively synchronized and bright field conditions that allowed resolution of the axon, compact myelin, and position of the SC nucleus. Observed areas were subsequently examined by electron microscopy (EM); eight myelinating SCs with known nuclear movement history were subjected to detailed EM analysis. We observed that, prefatory to myelination, SCs extended along the length of larger axons, apparently competing with adjacent SCs for axonal surface contact. This lengthening preceded the deposition of compact myelin. SC nuclear circumnavigation of the axon was found to attend early myelin sheath formation. This movement was rarely greater than 0.25 turns per 3 h; on the average, more nuclear motion was seen in relation to internodes that formed during observation (0.8 +/- 0.1 turns/24 h) than in relation to those that had begun to form before observation (0.3 +/- 0.1 turns/24 h). Nuclear circumnavigation generally proceeded in one direction, could be in similar or opposite direction in neighboring myelinating SCs on the same axon, and was not proportional to the number of major dense lines within the myelin sheath. A critical finding was that, in all eight cases examined, the overall direction of nuclear movement was the same as that of the inner end of the spiraling SC process, and thus opposite the direction of the outer end of the spiral. We conclude that the correspondence of the direction of nuclear rotation and inner end of the spiraling cytoplasmic lip implicates active progression of the inner lip over the axonal surface to form the membranous spiral of myelin, the nuclear motion resulting from towing by the advancing adaxonal lip. This interpretation fits with finding basal lamina and macular adhering junctions associated with the external lip of SC cytoplasm; these attributes would imply anchorage rather than movement of this region of the SC.

Freeze-fracturing of nerve growth cones and young fibers. A study of developing plasma membrane.

Neural and non-neural cellular processes have been studied in organotypic cultures of spinal cord and olfactory bulb by means of the freeze-fracturing technique. Identification of specific cellular elements in replicas has been achieved by comparison with thin-sectioned material in which differences in shape and contents are evident. Freeze-fracturing reveals that neural growth cones may be distinguished from glial pseudopodia by the low number of intramembranous particles within their plasma membrane; the counts of particles within the growth cone membrane average 85/microm(2) (for the inner leaflet) as opposed to hundreds per square micrometer in glial pseudopodia. Whereas the intramembranous particle number in glial pseudopodia is only slightly lower than in their perikaryal plasmalemma, the number of particles in outgrowing axons increases about eightfold from the periphery towards the perikaryon. Furthermore, with prolonged time of growth in culture, the particle density in the young nerve fibers increases by about the same factor. The same phenomenon, i.e. a low intramembranous particle level at earlier stages and an increase in numbers as the nerve fiber matures, is observed in fetal nerve tissue in vivo. These findings suggest that the plasmalemma of the outgrowing nerve, and especially of the growth cone, is immature and that maturation is accompanied by the insertion of intramembranous particles. Furthermore, these data indicate that the chemistry of the growth cone membrane is distinct from that of the neuron soma which may be significant for the mechanisms of guidance and recognition in the growing nerve tip.

Effects of calcium ion concentration on the degeneration of amputated axons in tissue culture.

Light and electron microscope studies were conducted on the nature of the degenerative changes in amputated nerve fibers of cultured rat sensory ganglia and on the effects of media with differing calcium concentrations upon these changes. With glucose-enriched Eagle's media (MEM) containing 1.6 mM calcium, the amputated myelinated and unmyelinated axons undergo a progressive granular disintegration of their axoplasm with collapse and fragmentation of myelin sheaths between 6 and 24 h after transection. With MEM containing only 25-50 microM calcium, the granular axoplasmic degeneration does not occur in transected fibers and they retain their longitudinal continuity and segmental myelin ensheathment for at least 48 h. Addition of 6 mM EGTA to MEM (reducing the estimated Ca(++) below 0.3 microM) results in the structural preservation of both microtubules and neurofilaments within transected axons. A transient focal swelling of amputated axons occurs, however, in cultures with normal and reduced calcium. These observations suggest that an alteration in the permeability of the axolemma is a crucial initiating event leading to axonal degenerative changes distal to nerve transection. The loss of microtubules and neurofilaments and the associated granular alterations of the axoplasm in transected fibers appears to result from the influx of calcium into the axoplasm.

Radioautographic studies of choline incorporation into peripheral nerve myelin.

This radioautographic study was designed to localize the cytological sites involved in the incorporation of a lipid precursor into the myelin and the myelin-related cell of the peripheral nervous system. Both myelinating and fully myelinated cultures of rat dorsal root ganglia were exposed to a 30-min pulse of tritiated choline and either fixed immediately or allowed 6 or 48 hr of chase incubation before fixation. After Epon embedding, light and electron microscopic radioautograms were prepared with Ilford L-4 emulsion. Analysis of the pattern of choline incorporation into myelinating cultures indicated that radioactivity appeared all along the length of the internode, without there being a preferential site of initial incorporation. Light microscopic radioautograms of cultures at varying states of maturity were compared in order to determine the relative degree of myelin labeling. This analysis indicated that the myelin-Schwann cell unit in the fully myelinated cultures incorporated choline as actively as did this unit in the myelinating cultures. Because of technical difficulties, it was not possible to determine the precise localization of the incorporated radioactivity within the compact myelin. These data are related to recent biochemical studies indicating that the mature myelin of the central nervous system does incorporate a significant amount of lipid precursor under the appropriate experimental conditions. These observations support the concept that a significant amount of myelin-related metabolic activity occurs in mature tissue; this activity is considered part of an essential and continuous process of myelin maintenance and repair.

Reversible alterations in the Golgi complex of cultured neurons treated with an inhibitor of active Na and K transport.

Highly differentiated cultures of rat and mouse sensory ganglia were treated for varying periods (up to 6 hr) with selected doses (1 x 10(-3)M to 5 x 10(-5)M) of the cardiac glycoside, ouabain (Strophanthin G). This inhibitor of active Na and K transport produced a selective and progressive swelling of elements of the Golgi complex in many, but not all, neurons. After 6 hr of treatment, virtually all Golgi complexes in affected neurons were either altered or absent; large cytoplasmic vacuoles limited by agranular membrane were prominent in these neurons. This response was not observed in satellite cells and Schwann cells. Within a few hours of ouabain withdrawal, the neuronal vacuoles disappeared and normal Golgi areas were again observed. These observations suggest that there is a site for active transport of Na and K on the Golgi membrane of these neurons. In discussing the possible significance of this observation, it is suggested that if this site were directed so that cation was actively pumped from Golgi cisternae into cytoplasm (and if there were differential water and ion permeabilities in various parts of the endoplasmic reticulum-Golgi system), then such a pumping mechanism could provide an explanation for the concentrating function of the Golgi apparatus.