Springer Science and Business Media LLC

Schwann cells are an important component of neurofibromas, one of the primary lesions encountered in neurofibromatosis type 1 in man. A central question in studies of neurofibromatosis type 1 has been whether the Schwann cells present in these tumours are intrinsically abnormal or exhibit abnormal phenotypes in response to stimuli from other cell types in these tumours. Damselfish neurofibromatosis is a naturally occurring disease in a species of marine fish, the bicolour damselfish, that is being developed as an animal model of neurofibromatosis type 1. Affected fish exhibit multiple neurofibromas and neurofibrosarcomas (malignant schwannomas). The present study compares the morphology, antigen expression and proliferative capacityin vitro of Schwann cells derived from peripheral nerves of normal, healthy fish with cells isolated from both spontaneously occurring and experimentally induced neurofibromas. Schwann cells from normal nerves expressed S100 antigens but not fibronectin or glial fibrillary acidic protein antigens and were similar in morphology and proliferative capacity to Schwann cells isolated from mammalian peripheral nerves. Tumour-derived cultures contained variable proportions (27–79%) of S100-positive cells that were identified as Schwann cells based on this feature. These tumour-derived Schwann cells exhibited a different morphology than normal Schwann cells, usually exhibited an increased reactivity to anti-S100 antibodies and were able to proliferate in vitro without added mitogens. Repeated subculturing of tumour-derived cultures led to the production of six cell lines all of which were composed exclusively of Schwann cells as indicated by S100 expression. These findings show that Schwann cells are an important component of tumours in Damselfish neurofibromatosis and that these cells are morphologically and physiologically altered in this disease. Observations of cell lines also suggest that tumour-derived Schwann cells are intrinsically abnormal and that this phenotype is not a result of stimuli from other cell types in the tumours.

The HNK-1 carbohydrate, an unusual 3′-sulfated glucuronic acid epitope characteristic of many neural recognition molecules, serves as a ligand in neural cell interactions and is differentially expressed in the quadriceps and saphenous branches of the femoral nerve in the PNS of adult mice. Based on these observations, we investigated the possibility that the HNK-1 carbohydrate may be differentially distributed in neurons and fiber tracts also in the CNS thereby contributing to different targeting and guidance mechanisms. We have used antibodies with different HNK-1 epitope specificities to probe for subtle differences in expression patterns. In the adult mouse cerebellum the HNK-1 carbohydrate is detectable in stripe-like compartments in the molecular and Purkinje cell layers, whereas N-CAM and its associated α2,8 polysialic acid does not show this compartmentation. In the adult hippocampus, the HNK-1 carbohydrate localizes to perineuronal nets of inhibitory interneurons and marks the inner third of the molecular layer of the dentate gyrus. In the adult spinal cord, HNK-1 labeling is most pronounced in gray matter areas. White matter enriched regions show differential labeling with regard to fiber tracts and antibody specificity. Whereas the different antibodies do not show differences in staining in the cerebellum and the hippocampus, they show differences in staining pattern of fiber tracts and motoneurons in the spinal cord. The HNK-1 expression pattern also differed in the adult spinal cord from that observed at embryonic day 14 and postnatal day 14. Our observations suggest a functional role in the specification of functionally discrete compartments in different areas of the CNS and during development.

The buccal ganglia ofNavanax inermis were studied by thin section, lanthanum infiltration and freeze fracture. Freeze fracture clearly demonstrates small gap junctions between neuronal processes in the neuropil, many of which are known to be electrotonically coupled. Junctional particles cleave with the P-face. In thin section, gap junctions appeared as small blurred contacts, presumably because of the small size of the junctions. Lanthanum infiltration was poor and failed to aid in identifying gap junctions. However, it did reveal septate-like junctions whose septa were not osmiophilic. Corresponding E-face grooves and ridges were seen in freeze fracture, sometimes adjacent to gap junctions. The septate-like junctions have parallel membranes and may have been mistaken for gap junctions in several other thin section studies of invertebrate neurons.

Light- and dark-adaptation leads to changes in rhabdom morphology and photopigment distribution in the octopus retina. Molecular chaperones, including heat shock proteins (Hsps), may be involved in specific signaling pathways that cause changes in photoreceptor actin- and tubulin-based cytoskeletons and movement of the photopigments, rhodopsin and retinochrome. In this study, we used immunoblotting, in situ RT-PCR, immunofluorescence and confocal microscopy to localize the inducible form of Hsp70 and the larger Hsp90 in light- and dark-adapted and dorsal and ventral halves of adult octopus retinas. The Hsps showed differences in distribution between the light and dark and in dorsal vs. ventral position in the retina. Double labeling confocal microscopy co-localized Hsp70 with actin and tubulin, and Hsp90 with the photopigment, retinochrome. Our results demonstrate the presence of Hsp70 and Hsp90 in otherwise non-stressed light- and dark-adapted octopus retinas. These Hsps may help stabilize the cytoskeleton, important for rhabdom structure, and are perhaps involved in the redistribution of retinochrome in conditions of light and dark.

The cell bodies of the layer II/III pyramidal cells in rat visual cortex receive three morphologically distinct types of axon terminals. These axon terminals all form symmetric synapses and have been termed large, medium-sized, and dense axon terminals. The present study shows that each of these different kinds of axon terminals contains gamma-aminobutyric acid (GABA) which suggests that they are inhibitory. From an analysis of the profiles of 50 cell bodies it is calculated that the average layer II/III pyramidal cell has 65 axosomatic synapses, of which 43 are formed by medium-sized terminals, 10 by large terminals/and 12 by dense terminals. Comparison of these different kinds of axon terminals with labelled axon terminals of known origin suggests that the medium-sized terminals are derived from smooth multipolar cells with unmyelinated axons, and that at least some of the dense terminals originate from bipolar cells that contain vasoactive intestinal polypeptides. The source of the large axon terminals is not known, but it is suggested that they originate from multipolar non-pyramidal cells with myelinated axons. Since the initial axon segments of these same neurons receive GABAergic axon terminals from chandelier cells, at least four different types of neurons provide inhibition to the cell bodies and axons of layer II/III pyramidal cells. This serves as an illustration; of the complexity of the neuronal circuits in which pyramidal cells are involved.

This study examined the fate of axon terminals of one of the major sources of hypoglossal afferents, the spinal V nucleus, after XIIth nerve resection in adult Sprague-Dawley rats. In order to anterogradely label trigemino-hypoglossal projections, small quantities of horse radish peroxidase were pressure-injected into the ipsilateral dorsal (mandibular) portion of the spinal V nucleus two days before the animals were killed. Survival periods ranged from 5 to 33 days after nerve injury (dpo). Axonal injury produced relative changes in the association of labelled axon terminals to structures in the hypoglossal nucleus on the injured side. The proportion of horse radish peroxidase-labelled spinal V nucleus terminals with spherical vesicles (S-terminals) that were unapposed to hypoglossal somata or dendrites increased rapidly and reached maximal levels by 11 dpo. By contrast, the isolation of labelled terminals with pleomorphic/flattened vesicles (P/F-terminals) from postsynaptic structures began later, advanced at a slower rate and did not attain maximal levels until 20 dpo. S-terminals not apposed to neuronal cell parts increased at a rate of 2.2 times greater than unapposed P/F-terminals. In addition, at peak levels, the proportion of labelled S-terminals that were detached from somata and dendrites was significantly greater than unapposed, labelled P/F-terminals. Axotomy did not alter the caliber of the labelled axon terminals. However, by 29 days after axotomy, the average diameter of dendrites remaining in contact with SPVN terminals was 1/3 the diameter of dendrites of uninjured neurons apposed to labelled axon terminals. These findings provide the morphological correlate for physiological and pharmacological evidence that the effectiveness of excitatory and inhibitory synapses are down-regulated in a coordinated manner after hypoglossal nerve injury.

Striatal dopamine deficiency in weaver mutant mice is associated with loss of mesencephalic dopamine neurons. The maximum dopamine concentration in the striatum of weaver mutants is found on postnatal day 20, when it represents 50% of the control value. By day 180, it declines to 25% of the control value. Correspondingly, the number of nigral dopamine neurons is 58% of the normal number on day 20 and becomes 31% of the normal value by day 90. The aim of the present study was to examine whether dopamine axon terminals in the weaver striatum establish synaptic connections with postsynaptic neurons at the time when striatal dopamine concentration is at its peak value (i.e. on postnatal day 20), and if so, to compare the profile of synaptic connectivity of dopamine axon terminals found in the striatum of normal mice with that of heterozygous and homozygous weaver mutants. To that end, 20-day-old weaver homozygotes, along with age-matched weaver heterozygotes and wild-type mice were studied by electron microscopy after immunocytochemical labelling for tyrosine hydroxylase. A single micrograph of each of 1543 dopamine axon terminals was examined in total in the three genotypes; quantitative analyses of the relations of tyrosine hydroxylase immunoreactive nerve terminals were carried out in the dorsolateral striatum, which receives the dopamine projection from the substantia nigra proper. In all three genotypes, junctional contacts formed by tyrosine hydroxylase immunoreactive nerve terminals in the striatum were predominantly of the symmetrical type. In wild-type and heterozygous mice, the majority of contacts (92% and 91% respectively) were formed with dendrites and spines. In weaver mutant mice, the majority of contacts (87%) were also with dendrites and spines, but the proportion of axosomatic contacts was double that found in normal animals. The proportions of contacts that displayed junctional membrane specializations in single sections were 27% in wild-type mice, 29% in weaver heterozygotes, and 17% in homozygous weaver mutants. Taking into consideration that the plane of the section might not always have included the synaptic specialization, a stereological formula was applied. It was estimated that 85–89% of the contacts may be truly junctional in the striatum of normal and heterozygous mice, whereas only 53% may be junctional in the striatum of weaver homozygotes. The reduced incidence of junctional synapses in weaver homozygotes may suggest either inadequate synaptogenesis, or an early loss of synapses after their formation, or both. Further, the increased incidence of axosomatic contacts may be indicative of synaptic immaturity, as such contacts are commonly seen in early developmental stages. Our results support the developmental nature of the nigrostriatal deficit in weaver mutants, since the synaptic investment of striatal neuronal elements by dopamine afferents appears to be immature at the time when nigrostriatal synaptogenesis is normally complete.

Hyaline cells are non-sensory epithelial cells of the vibrating part of the basilar membrane of chicks; they receive an extensive efferent innervation. Although these anatomical features suggest roles in auditory transduction, very little is known about the function of these cells. One possible way to understand function is by lesion experiments. We used synapsin-specific antibodies to study changes that occur in the pattern of efferent innervation in hyaline cells after lesion of the sensory epithelium induced by acoustic overstimulation. We found only small changes in hyaline cells after such trauma. These included a small increase in size and a small decrease in density of nerve terminals on hyaline cells. This suggests that hyaline cells and their nerve terminals are less susceptible to acoustic trauma than hair cells. Using neurofilament-specific antibodies we found little or no trauma-induced change in the density of nerve fibres that cross the basilar papilla and reach the hyaline cell region. This finding suggested that trauma to the hair cells does not necessarily lead to changes in the efferent fibres that cross the papilla and extend into the hyaline cell region. Using the trauma and the morphological parameters studied here, it appears that a moderate lesion in the hair cell region in the avian inner ear does not influence the hyaline cells or their innervation.