Gliofibroma. A peculiar neoplasia of collagen forming glia-like cells.

A tumor of highly unusual composition was found in the lower medulla oblongata of a 3 year, 9 months old girl. Light microscopy of the lesion showed an intimate intermingling of astroglial processes with collagen fibers, both projecting congruent to the pattern of preexisting fiber tracts. Electronmicroscopy disclosed a mosaic formed by the intimate apposition of collagen bundles, of glial processes filled with glial fibrils, and of scattered myelinated fibers, all abutting upon each other, often without intervening basement membranes. This well differentiated mass within the medulla adjoined a dedifferentiated endophytic extension of the tumor into the fourth ventricle which had seeded into the cerebellar cortex and the cauda equina. The dedifferentiated portion of the neoplasm also displayed intimate apposition of collagenous and glial elements. The tumor was tentatively identified as a gliofibroma or a desmo-plastic glioma.

Suprasellar neoplasm with a granular cell component.

A classification of granular cell lesions of neuropathologic interest is presented along with an unique case in which a suprasellar neoplasm having light and electron microscopic features of a meningioma contained multifocal nests of granular cells which apparently persisted in this tumor over a period of nearly five years. A spectrum of cell differentiation suggested the development of granular cells from small, electron-dense mesenchymal elements having numerous mitochondria. No evidence for a Schwann cell origin was found. It is proposed that the pituitary-hypothalamic axis, where intracranial granular cell nests and granular cell tumors are most common may also give origin to rare mixed neoplasms with a granular cell component.

Fine structure of medulloepithelioma.

A survey of published cases of medulloepitheliomas indicates that this tumor is highly malignant (average course 6 months), occurs in early childhood (average age 4,4 years) and has no sex predilection. Although it may originate in any part of the central nervous system, it is most common in the cerebral hemispheres. The first case studied with the electron microscope is reported here. The fine structure of the tumor is similar to that of the fetal neural tube in showing a primitive epithelium having uniform nuclei rich in chromatin but little cytoplasm and sparse organelles; it rests on a distinct basement membrane. The epithelial surface lacks cilia or microvilli and is covered with an irregular amorphous coat which is probably the cause of the light-microscopic misinterpretation of an inner limiting membrane. The lateral cell surfaces lack interdigitations but show extensive primitive forms of cell junctions. The case presented is also unusual in being the second case on record originating in the cerebellum and the first showing massive seeding along the spinal canal.

Dystopic myelination with hypertrophy of pyramidal tract.

In a case of hypertrophy of the pyramidal tract the cross section of the hypertrophic pyramids was 174% of controls with an increase (148%) in the total population of myelinated fibers. However, there was no commensurate increase in the density of Betz cells. Confined to the hypertrophic tract there was a peculiar anomaly consisting of tubes of thick myelin sheaths that encompassed columns of glial nuclei instead of axons. This type of change, along with the clinical data, may indicate that the lesion originated in the perinatal period when myelin formation is in progress and is susceptible to derangement.

Fine structure of arachnoid cysts.

Recent studies of the fine structure of the cranial meninges of laboratory animals and man have shown that there is no subdural space. The latter is formed artificially by the tendency of meningeal tissues to cleave along a collagen-free zone, the dura-arachnoid interface layer. This layer is composed of an outer zone of dural border cells and an inner arachnoid barrier layer. The fine structure of nine arachnoid cysts was studied to determine the derivation of the cyst's wall from the various components of normal human meninges. A cleaved dura-arachnoid interface layer covered only the dome of the cyst where the latter had abutted the dura mater. The interface layer did not partake in forming the cyst's wall. The dominant phenomenon of the cyst's wall was an absence of the normal trabeculation of the subarachnoid space, the trabecules being replaced by tightly packed collagen fibrils and a few scattered cells in between. Some cells were layered discontinuously at the inner face of the cyst wall, but there was no organized inner lining. No evidence was found for either a tight sealing of the extracellular spaces in the cyst's wall, nor for the existence of an active transcellular fluid movement.

The role of endoneurial fibroblasts in myelin degradation.

Counts of nuclei in electron micrographs of rat sciatic nerve show that approximately one third of the nuclear population is fibroblasts. Comparison of these figures with counts in teased fibers indicates that the majority of fibroblasts adheres tightly to myelinated nerve fibers. The early reactions and cellular transformations of the endoneurial fibroblast population upon intraneurial injections of india ink or of a purified preparation of rat sciatic myelin were studied for intervals of from 6 hours to 3 days after injection. There was phagocytosis of carbon or myelin by endoneurial fibroblasts as early as 6 hours after injection. A subsequent massive increase in phagocytic endoneurial cells appeared to develop from, and at the expense of, the normal endoneurial fibroblast population. The extent to which hematogenous monocytes also contributed to the macrophage population could not be assessed accurately by morphological criteria. Schwann cells never ingested carbon or myelin preparation during the period of investigation. Early invasion of the Schwann tube by macrophages from the endoneurial spaces could be demonstrated however. These macrophages were labeled by either carbon or myelin preparation which they had taken up from the endoneurial spaces before they entered the fibers through small gaps of the basal lamina, preferentially at the nodes of Ranvier. Carbon- or myelin-labeled hematogenous granulocytes entered nerve fibers in the same way, but the invasion of fibers by granulocytes was a transient early response limited to the first two days after injection. The data demonstrate the phagocytic capacities of endoneurial fibroblasts, their capacity for myelin degradation, and an invasion of Schwann tubes by elements of the endoneurial macrophage population during very early phases of fiber injury.

The cytogenetic basis for classifying ependymomas.

The phylogeny of ependymal cells and astrocytes can be traced to a single primitive progenitor the ependymoglia or the tanycyte, respectively. Ependymoglia cells have ependymal perikarya having astrocyte-like processes that terminate subpially in primitive glial footplates. Such cells prevail in primitive nervous systems, but they also persist regionally in the mature mammalian brain. Their fine structure has been studied in many species. An electronmicroscopic study of 8 ependymomas reveals that the neoplastic cells possess features characteristic of primitive ependymoglia; in particular they possess cell processes filled with glial filaments, terminating submesenchymally in a primitive, piston-shaped footplate. The perivascular pseudorosettes of ependymomas are the equivalents of these cell poles. The dominant phenomenon of ependymoma structure appears to be a reversion of cellular organization to the stage of primitive ependymoglia cells. On reviewing 43 ependymomas and 71 astrocytomas 11 neoplasms were found having a tissue structure reminiscent of the evolution of piloid astrocytes from ependymoglia or tanycytes, respectively. These features correspond to transitional stages seen in normal primitive brains. Tumors of this type may be characterized as a tanycytic variant of ependymomas. They appear to be relatively common in the spinal cord and present a source of confusion with piloid astrocytomas.

A new approach toward analyzing peripheral nerve fiber populations. I. Variance in sheath thickness corresponds to different geometric proportions of the internodes.

The thickness of the myelin sheath is known to increase with axon caliber, but there is also a superimposed, slight variation in sheath thickness depending on whether a fiber of a given caliber has very long or very short internodes. This relationship between myelin sheath thickness and the geometric proportion of the internode has been shown in subserial sections of isolated nerve fibers. It allows a prediction of sheath thickness from the quotient internode length/axon caliber, or conversely, a prediction of internode foreshortening from sheath thickness. We applied this new approach to the analysis of sciatic fiber populations of frogs, mice, rats and cats. The geometric proportions of these fibers were defined by the quotient internode length/fiber caliber. This quotient was compared with minor variation in sheath thickness as determined with a computer-assisted technique measuring large numbers of fibers in low-power electron micrographs. The method also calculated fiber shrinkage and recalculated all data for circular fiber profiles. The data obtained confirmed previous electron microscopic measurements showing that there is a slight reduction in sheath thickness when a fiber of a given caliber has relatively short internodes, and vice versa. A population of very thin, thinly myelinated fibers was also revealed. Sheath thickness and the geometric proportions of internodes in frogs differed markedly from those in mammals.

A new approach toward analyzing peripheral nerve fiber populations. II. Foreshortening of regenerated internodes corresponds to reduced sheath thickness.

The new approach used in this study is based on the concept that axon caliber is not the only factor affecting the thickness of the myelin sheath. It is necessary to consider the entire geometric proportions of the internode, since sheath thickness corresponds to the relationship between axon caliber and the length of the internode. This type of analysis was applied to the regenerated internodes in rat sciatic nerves. Survival periods of 4, 9, 18 and 36 weeks were studied after lesions had been placed in young adult rats. The data show significantly thinner sheaths for regenerated fibers as compared with normal nerves, consistent with previous observations. This reduction in sheath thickness, however, corresponded quantitatively to the degree of foreshortening of internodes in the regenerated nerves. An average reduction of 10 in the quotient internode length/fiber caliber corresponded to a reduction of about 0.015 in the relative thickness of the sheath (quotient axon diameter/fiber diameter). This means that regenerated myelin sheaths are not truly hypoplastic; rather, they are adapted to the reduced internode length, and have the same relationship found for normal fibers. In partially damaged nerves there was a clear distinction in terms of sheath thickness between regenerated fibers and undamaged fibers. Demonstration of this phenomenon by scatter diagrams opens new possibilities for the quantitative assessment of neuropathies.

A morphometric study of nerve fiber atrophy in rat spinal roots.

A morphometric study of atrophy of nerve fibers was made in the ventral and dorsal roots of rats, four to 48 weeks (wk) after transection of their sciatic nerves and with regeneration prevented. The pathophysiological events of fiber atrophy may be summarized as follows: wasting of the axon caliber coincides with a loss of neurofilaments with relative preservation of microtubules. This leads to non-circularity of the fiber, evident from four wk on. The caliber of the circular profile is also reduced. Adaptive changes in myelin sheath structure follow. Thinner fibers with relatively thick sheaths are first detectable after eight wk and prominent after 24 to 48 wk. This change may indicate passive slippage of the sheath, but sheath remodelling in adjustment to a changed internodal geometry appears more likely.

Myelin phagocytosis by peritoneal macrophages in organ cultures of mouse peripheral nerve. A new model for studying myelin phagocytosis in vitro.

Organ cultures of degenerating nerve fascicles were exposed to cultured macrophages obtained by peritoneal lavage. Invasion of the nerve fascicle by phagocytes was shown by prelabeling with carbon and with electron microscopy. There was massive active phagocytosis of degenerating myelin sheaths. The invading phagocytic cell population was identified as Fc receptor-positive, Mac-1-positive macrophages by immunocytochemistry. The Schwann cell population persisted without significant myelin phagocytosis. The vitality of the Schwann cell population was shown by subsequent reimplantation of the organ cultures into host animals. The reimplants had retained their ability to remyelinate regenerating axon sprouts. These observations were made in cultures exposed to cytostatic agents. If cytostatic agents were omitted, there was proliferation of endogenous phagocytes in the nerve fascicles without added peritoneal cells. These endogenous phagocytes were identified as proliferating resident monocytes and were positive for the Fc receptor and Mac-1 markers. This model allows studies on how monocytes recognize and digest degenerating myelin apart from surviving Schwann cells.

Macrophage properties during peripheral nervous tissue rejection in vitro.

The present study introduces an in vitro model of xenogeneic peripheral nerve rejection to analyze the role of macrophages in this complex immunological situation. Nerve-sensitized mouse peritoneal exudate cells were co-cultured with rat peripheral nerve segments. The cultured rat nerve segments were fulminantly rejected in vitro by the co-cultured mouse peritoneal cell population. The massive tissue destruction included Schwann cell damage and was quite distinct from basic myelin phagocytosis observed during Wallerian degeneration in earlier experiments. The nerve-sensitized peritoneal exudate consisted of macrophages and T-cells. Antibody depletion experiments were performed to analyze T-lymphocyte effects in this model. Fulminant tissue rejection depended on the presence of T-lymphocytes in the culture medium. Their presence at the immediate site of tissue rejection, however, was not required. Further experiments were aimed at defining the role of T-cell-derived mediators during in vitro rejection. Depletion experiments using a panel of antibodies to cytokines revealed a critical involvement of IL-2, IL-3, IL-4, IL-6 and interferon-gamma in the induction of tissue rejection in vitro. Fulminant tissue rejection in vitro depended on the interaction of these cytokines with macrophages. The participation of macrophage surface receptors was studied in another series of experiments. The macrophage complement receptor type 3 was shown to be critically involved in the phagocytic attack during rejection. Antibodies to MHC class II antigens also abolished fulminant in vitro rejection, indicating that continuous antigen presentation is required in this process.

Midline hyperplasia with malformation of the fornical system.

A physically and intellectually normally developed boy had a resection of a lumbar myelomeningocele shortly after birth and a shunt was inserted for hydrocephalus. He died at age 12 years after a traffic accident. Autopsy disclosed a hyperplasia of the cerebellar vermis. An abnormal median-sagittal triangular fiber tract ran on the dorsal surface of the corpus callosum and contained more than 3 million fibers, most of them very thin and myelinated. The tract was identified as a hyperplastic fornix dorsalis. The complex malformation was attributed to hyperplasia of midline structures.

Central spinal myelinolysis.

A midline lesion of the funiculus gracilis was found in four patients with histories of alcoholism, nutritional disturbance, and repeated episodes of electrolyte imbalance. Histologic and anatomic abnormalities were similar to those of central pontine myelinolysis. Sodium imbalance may play a role in the pathogenesis of both conditions.

Activation of macrophages by recombinant interferon-gamma has no effect on myelin phagocytosis but hinders invasion of nerves in organ culture.

Myelin phagocytosis in nerves undergoing Wallerian degeneration was shown to depend on their invasion by non-resident, hematogenous macrophages. This process can be studied in vitro using organ cultures of peripheral nerves exposed to cultured peritoneal macrophages. The present report concerns the effect of recombinant interferon-gamma (rIFN-gamma) on luminol-dependent chemiluminescence, macrophage migration and myelin phagocytosis in organ cultures. Chemiluminescence was activated by rIFN-gamma compared to untreated cells. The macrophage population was capable of activation at any phase of exposure to organ cultures. The engagement of macrophages in myelin phagocytosis, however, varied with the timing of the application of rIFN-gamma. Application from the start of the experiment led to activation of chemiluminescence and also to a complete inhibition of macrophage invasion of the organ culture, thus preventing myelin removal. Application of rIFN-gamma at a later phase of the experiment had no effect on cell invasion and also no detectable effect on the efficiency of myelin phagocytosis. There was no indication that myelin phagocytosis by itself activated chemiluminescence in untreated cultures. Phagocytosis of myelin appears to be a function of macrophages independent of activation causing production of oxygen radicals.

L-fucosidase treatment blocks myelin phagocytosis by macrophages in vitro.

Myelin phagocytosis in nerves undergoing Wallerian degeneration has been shown to depend on their invasion by non-resident, hematogenous macrophages. This process can be studied in vitro using organ cultures of peripheral nerves exposed to cultured peritoneal macrophages. The present report concerns the role of cell surface carbohydrates in the invasion of degenerating nerves and in the recognition and ingestion of myelin by the phagocytic cells. Additional experiments explored the effect of pH, calcium and cytochalasin D on myelin phagocytosis. Organ cultures with peritoneal macrophages were treated with 14 simple or complex sugars or with eight sugar-splitting enzymes. Macrophage invasion was diminished by many simple or complex sugars, but exposure to sugars had no effect on the recognition or ingestion of myelin by the invading macrophages. Macrophage invasion was abolished upon treatment with beta-mannosidase. Exposure to L-fucosidase abolished the myelin phagocytic capacity of invading macrophages completely without affecting their capacity to ingest carbon or latex particles. The results indicate that the phagocytosis of myelin by macrophages is an L-fucosidase-sensitive process, probably by interaction with their complement receptor type C3.

TNF-alpha suppresses CR3-mediated myelin removal by macrophages.

Mononuclear cells of the monocyte/macrophage system play an important role in myelin ingestion during Wallerian degeneration. The present in vitro study clarifies the role in this process of two macrophage-secreted cytokines, TNF-alpha and interleukin-1. Treatment with TNF-alpha massively reduced the amount of myelin ingested by macrophages via their complement receptor type 3 (CR3). Anti-TNF-alpha antibodies reversed the effect. Immunofluorescence of macrophages indicated that TNF-alpha caused a reduced expression of the CR3 by phagocytic cells. Further experiments revealed an interaction of TNF-alpha with its receptor on the macrophage cell membrane. Interleukin-1 had no effect on myelin ingestion in the in vitro system used in these experiments.

Myelin phagocytosis in Wallerian degeneration of peripheral nerves depends on silica-sensitive, bg/bg-negative and Fc-positive monocytes.

Previous experiments with nerves enclosed in millipore diffusion chambers had shown that myelin degradation during Wallerian degeneration depends on invasion by non-resident cells. The present study was aimed at a more precise identification of the invading cell population. Monoclonal antibody studies of degenerating nerves showed many cells with the Fc marker; cells having the Lyt-1, Lyt-2, Ia or Mac-1 markers were sparse or absent. Nerves transplanted into mice of the Chediak-Higashi bg/bg strain were invaded by cells lacking the bg/bg marker (giant lysosomes), while cotransplanted muscle tissue was invaded by cells with the bg/bg marker. Blocking monocytes with silica reduced both cell invasion and myelin degradation in degenerating nerves. These observations show that Wallerian degeneration of peripheral nerve fibers involves a subset of monocytes which are silica-sensitive and have Fc receptors but no bg/bg giant lysosomes.

Conduction velocity varies with osmotically induced changes of the area of the axon's profile.

The conduction velocity of frog ischiadic nerves incubated in vitro in osmolarities between 220 and 1000 mOsm decreased with the degree of fiber shrinkage. The latter (non-circularity factor) was determined from computer-assisted measurements in freeze-substituted or in chemically fixed fibers. Freeze-substituted normal nerves had a non-circularity factor of 0.91 for fibers of all calibers, which likely reflects the in vivo state of the fiber population. Chemically fixed nerves had a non-circularity factor near 0.68, consistent with previous data. Non-circularity factors decreased with increasing osmolarities of the media, regardless of the type of tissue preparation. Conduction velocity decreased with decreasing non-circularity. Restoration of the nerves to normotonic media increased conduction velocity. The rates of change were accelerated in nerves chemically desheathed with Triton. The decrease in the conduction velocity in osmotically shrunken nerves did not correspond to changes in the absolute refractory period for the propagation of the impulse, used as a sensitive index of non-specific damage. These experimental observations corroborate data from computer simulation of relative sensitivities of nodal and internodal parameters.

How are sheath dimensions affected by axon caliber and internode length?

The significance of internode length for sheath thickness was analyzed by electron microscopic morphometry in isolated internodes from the human roots C3 and S1. These populations differ in length but have similar caliber. The amount of myelin per internode was in linear relation with the product of axon circumference and the length of the ensheathed axon segment. Neither one of these two vectors was in a statistically significant relationship with sheath thickness. The ratio between the axolemmal area covered by the Schwann cell and the area of the myelin leaflet averaged 1:163 for human root fibers. It was 1:177 for previous data from canine sciatic nerve. The proportions of an internode were defined by an 1/d-quotient, expressing its length as multiples of axon diameter. Relative sheath thickness (g-ratio: diameter axon/diameter fiber) relates inversely with the 1/d-quotient. For a given axon caliber, the g-ratio (sheath thickness) decreases by 0.006 for every 10.0 increase in 1/d-quotient (relative internode length). Thus, internodes relatively long for axon caliber possess slightly thicker sheaths than internodes short for axon caliber. Axon caliber and relative internode length emerge as the two key factors determining the amount of myelin in a sheath.