Light- and electron-microscopic study of gamma-aminobutyric-acid-like immunoreactivity in the guinea pig organ of Corti

With light and electron microscopy, gamma-aminobutyric-acid (GABA)-like immunoreactivity was examined in the guinea pig organ of Corti. In whole-mount preparations, although GABA-like immunoreactivity was present within efferent components in all turns of the cochlea, it was more intense apically. At the ultrastructural level, GABA-like immunoreactivity was clearly restricted to the efferent system, appearing in axons of the inner and tunnel spiral bundles, in axons crossing the tunnel of Corti, and in terminals filled with numerous labeled vesicles synapsing on outer hair cells.

Ultrastructural characterization and GAD co-localization of somatostatin-like immunoreactive neurons in CA1 of rabbit hippocampus

Immunocytochemical techniques have been used to identify a striking interneuronal population which is immunoreactive for the peptide, somatostatin. The cell population, which is seen most densely in stratum oriens and at the oriens/alveus border of the CA1 region of rabbit hippocampus, was characterized in light and electron microscopic observations. The cells have dendrites which extend parallel to and into the alveus, with occasional processes ascending through stratum pyramidale toward the hippocampal fissure. The dendrites receive numerous synaptic contacts directly onto aspinous dendritic shafts. Axon collaterals ramify profusely within the pyramidale region, and among the proximal apical and basal pyramidal cell dendrites in areas of stratum radiatum and stratum oriens. Somatostatin-like immunoreactive terminals make synaptic contact, primarily of the symmetric type, with the somata and proximal dendrites of pyramidal neurons. Somatostatin-like neurons are found at approximately equal density in the hippocampus of immature (8 days postnatal) and mature (30 days postnatal) rabbit. Double-labelling techniques, to identify both somatostatin-like and glutamic acid decarboxylase (GAD) immunoreactive neurons, demonstrated that a large proportion of the somatostatin neurons were also GABAergic.

Ultrastructural organization of regenerated serotonin axons in the dorsomedial hypothalamus of the adult rat

The ultrastructural organization of regenerated serotonin (5-HT) axons was examined in the dorsomedial hypothalamus (DMH) of the adult rat using high-resolution radioautography after intraventricular infusion of [3H]5-HT. An analysis of the microenvironment of the [3H]5-HT-labelled terminals in the DMH was made 30 and 50 days after unilateral injection of 5,7-dihydroxytryptamine (5,7-DHT) or vehicle solution into the dorsolateral hypothalamus. In sham-treated animals [3H]5-HT-labelled axons were small, contained many small clear vesicles, one or more large granular vesicles, and showed only rare synaptic specializations. In 5,7-DHT-treated animals the internal organization of [3H]5-HT-labelled profiles resembled that of sham-treated animals. A tendency toward increased synaptic frequency was found for [3H]5-HT-labelled terminals in the 5,7-DHT-treated group 50 days post-lesion, and an increase in the number of [3H]5-HT-labelled terminals abutting unlabelled perikarya was found in both 30- and 50-day post-lesion groups as compared to sham-treated groups. No other differences in ultrastructural environment were found between sham- and 5,7-DHT-treated animals at either 30 or 50 days post-lesion. These results suggest that 5-HT fibres in the hypothalamus regenerate with a great deal of cellular specificity.

Functional-anatomical studies on the terminal nerve projection to the retina of bony fishes

We have explored the structure and actions of terminal nerve (TN) fibers in the teleostean retina, the most accessible of TN projections. Using immunocytochemistry we have shown that the goldfish TN contains neuropeptides related to the molluscan cardioexcitatory peptide (FMRFamide) as well as luteinizing hormone-releasing hormone (LHRH). Retinal TN terminals were found upon major dendrites in the distal inner plexiform layer and neuronal cell bodies in the amacrine cell layer. Electron-microscopic double-labeling revealed TN terminals applied to the surface of [3H]-dopamine-, glycine-, and gamma-aminobutyric acid (GABA)-accumulating cells. Synthetic LHRH and FMRFamide at less than 1 microM modified spontaneous and light-evoked activity of ganglion cells in isolated superfused goldfish retina, especially during the active breeding season. Salmon(I)-LHRH was 10-30 times as potent as mammalian LHRH and caused rapid, prolonged desensitization. We conclude that LHRH- and FMRFamide-like peptides may be released by retinal TN endings, probably in concert with reproductive activity, and that they act independently through horizontal and/or amacrine cell pathways to modify visual information processing in the retina.

An ultrastructural study of 5-hydroxytryptamine-, thyrotropin-releasing hormone- and substance P-immunoreactive axonal boutons in the motor nucleus of spinal cord segments L7-S1 in the adult cat

The distribution and fine structure of 5-hydroxytryptamine-, thyrotropin-releasing hormone- and substance P-immunoreactive synaptic boutons and varicosities were studied in the motor nucleus of the spinal cord segments L7-S1 in the cat, using the peroxidase-antiperoxidase immunohistochemical technique and analysis of ultrathin serial sections. The 5-hydroxytryptamine-, thyrotropin-releasing hormone- and substance P-immunoreactive boutons had a similar ultrastructural appearance as judged from serial section analysis. The boutons could be classified into two types on the basis of their vesicular content, with one type containing a large number of small agranular vesicles together with only a few, if any large granular vesicles, while the other type contained a large number of large granular vesicles in addition to small agranular vesicles. The vesicles were spherical or spherical-to-pleomorphic. Postsynaptic dense bodies (Taxi bodies) were occasionally observed in relation to all three types of immunoreactive boutons, which almost invariably formed synaptic junctions with dendrites. Judged by the calibre of the postsynaptic dendrites, the boutons were preferentially distributed to the proximal dendritic domains of motoneurons. In one case, a substance P-immunoreactive bouton formed an axosomatic synaptic contact. In addition to synaptic boutons, 5-hydroxytryptamine-, thyrotropin-releasing hormone- and substance P-immunoreactive axonal varicosities containing a large number of large granular and small agranular vesicles but lacking any form of conventional synaptic contact were observed. Such varicosities were either directly apposing surrounding neuronal elements or separated from the neurons by thin glial processes. The origin of the immunoreactive boutons was not traced, but it was thought likely that the main source of the boutons was neurons with their cell bodies located in the medullary raphe nuclei.

The development of a population of spinal cord neurons and their axonal projections revealed by GABA immunocytochemistry in frog embryos

The development of a population of cerebrospinal-fluid-contacting neurons in the spinal cord of the Xenopus embryo ('Kolmer-Agduhr' cells) has been followed by using an immunocytochemical procedure that identifies GABA in fixed nervous tissue. Stained Kolmer-Agduhr cells containing GABA first appeared at stage 25 and their numbers increased steadily with the developmental age of the embryo. The Kolmer-Agduhr neurons had ascending ipsilateral axons that often terminated in growth cones. These axons and growth cones could be stained by the GABA antiserum from the earliest stages of outgrowth from the Kolmer-Agduhr cell body. We measured the angle of the earliest axons' outgrowth relative to the rostrocaudal axis of the spinal cord. The initial outgrowth of axons was always rostral over a narrow range of angles. This observation is inconsistent with the hypothesis of random initial outgrowth followed by later selection of the correct orientation, which would predict that axons would initially grow out over a wide range of angles. Instead, it suggests that, even from the earliest moments, axon outgrowth from the Kolmer-Agduhr cells is directed rostrally in a specific stereotyped manner.

Experimentally induced alteration in the polarity of developing neurons

Despite the great diversity of shapes exhibited by different classes of nerve cells, nearly all neurons share one feature in that they have a single axon and several dendrites. The two types of processes differ in their morphology, in their rate of growth, in the macromolecular composition of their cytoskeletons and surface membranes, and in their synaptic polarity. When hippocampal neurons are dissociated from the embryonic brain and cultured, they reproducibly establish this basic form with a single axon and several dendrites, despite the absence of any spatially organized environmental cues, and without the need for cell to cell contact. We have cut the axons of young hippocampal neurons within a day of their development: in some cases the initial axon regenerated, but more frequently one of the other processes, which if undisturbed would have become a dendrite, instead became the axon. Frequently the stump of the original axon persisted following the transection and subsequently became a dendrite. Evidently the neuronal processes that first develop in culture have the capacity to form either axons or dendrites. The acquisition of axonal characteristics by one neuronal process apparently inhibits the others from becoming axons, so they subsequently become dendrites.

Localization of an epitope of a microtubule-associated protein 1x in outgrowing axons of the developing rat central nervous system

We have shown that monoclonal antibody, G10, labels developing rat cerebellum in a manner consistent with the presence of microtubule-associated protein 1x in growing axons. In this paper we show that all growing axons investigated in post-natal and pre-natal developing rat brain contain microtubule-associated protein 1x and that neither dendrites, cell bodies nor glia are labelled by G10. The distribution of the G10 epitope as shown by immuno-electron microscopy is consistent with its localization to axonal microtubules. On Western blots there is only a small decrease in microtubule-associated protein 1x from first-natal day 7 to adulthood, in contrast to the dramatic decrease in immunofluorescence during axonal maturation. The G10 epitope is therefore probably masked in adult brain. The spatial and chronological coincidence of the G10 epitope with axonal elongation offers the possibility of using this monoclonal antibody to probe the function of the cytoskeleton during neuronal growth.

The rat striatum: a target nucleus for ascending axon collaterals of the entopedunculo-habenular pathway

Direct projections from the entopeduncular nucleus (Ep) to the striatum were examined in the rat using retrograde tracing techniques. After injecting horseradish peroxidase (HRP) into the caudoputamen (CPU), labeled fibers could be traced medially to go through and/or terminate in both the globus pallidus (GP) and Ep ipsilateral to the injection. Interestingly, HRP-positive neuronal perikarya were observed in the Ep as well as in the GP after the CPU injections. These labeled Ep neurons were medium in size and restricted to the rostral 1/3 of the nucleus, in a distribution which overlapped the terminal fields from the CPU. A fluorescent tracer injection into the CPU resulted in retrograde labeling of Ep cells in the same fashion as after HRP injections. A second experiment was designed to determine whether the Ep cells projecting to the CPU have axon collaterals to any of the other known terminal fields of the Ep. Following True blue (TB) injections into the CPU and Diamidino yellow (DY) injections into the lateral habenular nucleus (lHb), all the TB-positive neurons in the rostral 1/3 of the Ep were unequivocally double-labeled with DY. On the other hand, the TB-labeled Ep neurons were never double labeled with DY injected into the centre median-parafascicular complex (CM-Pf) or ventroanterior-ventrolateral complex (VA-VL) of the thalamus or nucleus tegmenti pedunculopontinus pars compacta (TPC). The present study shows that single Ep neurons in the rostral 1/3 of the nucleus innervate both the striatum and lHb, and raises the possibility that these neurons may have a critical role in integrating motor and limbic functions in the basal ganglia.

Colocalization of GABA and tyrosine hydroxylase immunoreactivities in the axons innervating the neurointermediate lobe of the rat pituitary: an ultrastructural immunogold study

Distribution of the gamma-aminobutyric acid (GABA)ergic and dopaminergic innervations was studied in the rat neurointermediate lobe using antibodies against GABA and tyrosine hydroxylase. In light microscopy, immunoperoxidase staining revealed similar distribution patterns of the axons reacting with both antibodies. Diffusely scattered in both lobes, they were more concentrated along the marginal zone of the neural lobe. Application of a double, recto-verso, immunogold labelling method in electron microscopy revealed systematic colocalization of GABA and tyrosine hydroxylase (TH) immunoreactivities in the axons innervating the intermediate lobe; in the neural lobe, almost all GABA-immunoreactive axons were also labelled for TH. Thus, GABA and dopamine, hitherto reported to occur in distinct axons, in fact colocalize in the axonal systems which innervate the pituitary neurointermediate lobe. These observations suggest possible interactions (pre- or postsynaptic) of both transmitters in the functional regulation of the intermediate and neural lobes.

A Golgi study on the neuronal organization of the neostriatum in the mouse

The neuronal organization of the neostriatum in mice was studied, using the rapid Golgi method. Based on the size of the somata, the neostriatal neurons were divided into groups of large, medium-sized and small cells, and the neurons of each group were further divided into 2-5 types, according to the shape of the somata and dendritic morphology. Three types of large neurons were recognized. Large type I neurons were triangular, piriform or fusiform cells with a few thick dendrites, whereas large type II and type III neurons were round or polygonal cells with numerous slender dendrites. The dendrites of the large type II neurons were far longer than those of large type III. Medium-sized neurons were grouped into 5 types. Medium type I neurons were round with spiny dendrites and were found mainly in the caudal portion of the neostriatum. Medium type II neurons had numerous thin dendrites and were predominant in the rostral portion of the neostriatum. Some medium type II neurons were arranged in cell chains extending perpendicular to Wilson's pencils. The cell bodies of medium type III neurons were triangular, and generated long spiny dendrites. Medium type IV neurons were polygonal, and dendrites with numerous short branchlets were evidenced. Medium type V neurons had poorly branched and sparsely spinous dendrites. The small neurons were of two types: small type I had piriform cell bodies, which gave rise to very thin dendrites, while small type II had dendrites with irregular contours and filiform appendages. Of these, the large type I and type II, the medium type I-V, and the small type I neurons seemed to be the projection neurons, whereas the large type III and small type II neurons were merely internuncials. Thus, the neostriatum in the mouse was shown to be composed of a wide variety of projection neurons and only two types of interneurons.

Synaptic organization of GABAergic neurons in the mouse SmI cortex

Immunocytochemical methods were used to examine GABAergic neurons in the barrel region of the mouse primary somatosensory cortex. GABAergic neurons occur in all layers of the barrel cortex but are more concentrated in the upper portion of layers II/III and in layers IV and VI. Nine cells in layer IV were examined with the electron microscope, and portions of their dendrites were reconstructed from serial thin sections. These cells are of the nonspiny, multipolar or bitufted varieties, and some of them have beaded dendrites. The labeled cell bodies and their reconstructed dendrites were postsynaptic at asymmetrical synapses with thalamocortical axon terminals labeled by lesion-induced degeneration and with unlabeled axon terminals. Each cell also received symmetrical synapses from GABAergic axon terminals and from unlabeled axon terminals. Our results indicate that GABAergic cell bodies and processes receive synapses from thalamocortical axon terminals but that different cells display marked differences in the proportion of thalamocortical and other synapses they receive. These results indicate that GABAergic cells form a heterogeneous population with respect to their morphologies and patterns of synaptic inputs. The synaptic sequences revealed here for GABAergic neurons represent an anatomical substrate for various inhibitory processes known to occur within the cerebral cortex.

Characterization of posttranslational processing of the mammalian high-molecular-weight neurofilament protein in vivo

Sensory neurons in the dorsal root ganglion (DRG) were used in an in vivo pulse-chase labeling paradigm to examine the time course and nature of posttranslational processing of neurofilament (NF) proteins. Ganglia of adult rats were labeled with 35S-methionine and harvested 1-72 hr later. Samples containing the cell bodies and short initial axonal segments of DRG neurons were analyzed by 1- and 2-dimensional PAGE/fluorography. For comparison, axonally transported NF proteins (200, 145, and 68 kDa) were harvested from the sciatic nerve 21 d after labeling the fifth lumbar (L5) DRG. Analysis of the pulse-chase experiments revealed that the mature 200 kDa protein (NF200) was not identifiable in gels of DRG samples until 24-48 hr after labeling. Immunoblotting/fluorography of 2-dimensional gels with monoclonal antibodies to phosphorylated and unphosphorylated NF proteins identified the high-molecular-weight NF subunit in various stages of processing in the DRG between 1 and 48 hr after labeling. The precursor to NF200 migrated on 2-dimensional PAGE as a 160 kDa protein with a pI of about 7.2. During the next 48 hr, the migration of this protein progressively changed to the mature pattern of 200 kDa and a pI of about 5.2. The 145 and 68 kDa NF proteins exhibited very little change in migration on gels during this same interval. Dephosphorylation of mature NF proteins with E. coli alkaline phosphatase regenerated the 160 kDa precursor, confirming that phosphorylation was the main posttranslational mechanism involved in the maturation of newly synthesized high-molecular-weight NF protein. Detergent extraction of labeled DRGs suggested that the 160 kDa NF protein was present in assembled neurofilaments. Immunohistochemical experiments with monoclonal antibodies were performed to explore the intracellular location of phosphorylated and unphosphorylated high-molecular-weight NF protein. Analysis revealed that neuronal cell bodies, as well as short initial segments of DRG axons located within the ganglion, contained unphosphorylated NF protein, while axons in the distal nerve contained mature, phosphorylated NF200. These findings provide support for a model in which posttranslational processing of the 160 kDa precursor occurs in the initial axonal region of DRG cells after the assembled NFs have left the cell body and begun axonal transport.

Biochemical and immunocytochemical characterization and distribution of phosphorylated and nonphosphorylated subunits of neurofilaments in squid giant axon and stellate ganglion

Monoclonal antibodies to squid neurofilament (aNFP) and intermediate filament (aIFA) proteins were used as probes for the biochemical and immunocytochemical analyses of neurofilament structure and distribution in the squid giant axon and stellate ganglion. On Western blots the aNFP antibody stained exclusively the 220 kDa and high-molecular-weight (HMW) components of neurofilaments in the giant axon, whereas the aIFA antibody primarily labeled the 60 kDa protein in the giant axon and the 60 and 65 kDa proteins in the stellate ganglion. Dephosphorylation of axoplasmic proteins by alkaline phosphatase resulted in a decrease in the molecular weights of both the 220 kDa and HMW neurofilament proteins and a concomitant loss of reactivity with the aNFP antibody on Western blots. This indicated that the aNFP antibody is specific for a phosphorylated epitope in the neurofilament. Increased dephosphorylation of the 220 kDa protein led to an enhanced immunostaining of the resultant 190 kDa polypeptide by the aIFA antibody, suggesting that the phosphorylated epitope may mask the conserved epitope recognized by aIFA. Light and electron microscopic immunocytochemical studies show intense labeling by the aNFP antibody in the giant axon. In contrast, the aIFA antibody labeled the glial cells around the giant axon intensely, while labeling of the giant axon itself was considerably less than that with the aNFP antibody. Since the 60 kDa protein in axoplasm is intensely stained by the aIFA antibody on Western blots, the relatively low amounts of labeling seen on semithin and thin sections of the giant axon by this antibody may be due to the masking of the 60 kDa protein by in situ fixed axoplasmic proteins. However, the aIFA antibody intensely labeled glial cells within the stellate ganglion and "islands" of filaments and nuclear membranes within ganglion cells. No reactivity for either antibody was seen in synapses. The aNFP antibody specifically labeled "beadlike" portions and cross-bridges on the axonal neurofilaments, suggesting that these components consist of the 220 kDa and HMW proteins. In contrast, the aIFA antibody labeled relatively smooth filaments in ganglion and glial cells. These data suggest that the 65 kDa protein represents the squid glial filament protein and that the 60 kDa protein found in axoplasm represents the low-molecular-weight subunit in the axonal neurofilament. The latter appears to be formed and/or organized in "islands" of filaments within ganglion cells.(ABSTRACT TRUNCATED AT 400 WORDS)

Axon pathway boundaries in the developing brain. I. Cellular and molecular determinants that separate the optic and olfactory projections

When optic fibers first approach the chiasmatic region of the diencephalon in the chick embryo on days 3 and 4 (E3-4), they rarely grow rostrally into the olfactory region of the telencephalon. Conversely, olfactory tract axons grow as far as, but never cross the diencephalic/telencephalic (D/T) boundary to enter the optic chiasm. In this study, a region of specialized neuroepithelium, originally named the "knot" in mouse by Silver (1984), has been identified at the D/T border of chick embryos. At pre-axonal stages, the presumptive knot region undergoes a cataclysmic cell death, with concomitant phagocytosis of necrotic debris by the remaining cells. When fibers subsequently appear in the chiasm and olfactory tracts, the knot consists of a very dense, interwoven cluster of non-neuronal cells that lack marginal radial processes, and whose cell bodies directly abut the glial limiting membrane. Thus, the morphology of the knot is in sharp contrast to the cell body-free marginal zone and endfoot regions along which axons tend to grow. In addition, we found that the neural cell adhesion molecule (N-CAM), which is expressed on neuroepithelial cell processes within the central optic and olfactory pathways, is not present on cells in the knot region during periods of axon growth. These results suggest that the knot, through its elimination of the marginal zone processes, absence of large extracellular spaces, and relative absence of adhesion molecules, functions as an axon-refractory barrier that effectively separates the optic and olfactory projections.

Two immunohistochemically identified populations of calcitonin gene-related peptide (CGRP)-immunoreactive axons in human skin

Double labelling immunofluorescence has shown two populations of unmyelinated sensory axons in human skin that contain immunoreactivity (IR) to calcitonin gene-related peptide (CGRP). One population also contains IR to substance P (SP), whilst the other also contains IR to somatostatin (SOM). Axons containing both CGRP-IR and SOM-IR comprised more than 75% of CGRP-IR axons associated with the epidermis; the rest of the CGRP-IR axons contained SP-IR. No axons contained both SP-IR and SOM-IR. Some dermal blood vessels were surrounded by axons containing both CGRP-IR and SP-IR, but most CGRP-IR perivascular axons contained SOM-IR without SP-IR. Sweat glands were well supplied with sensory axons containing CGRP-IR and weak SOM-IR but not SP-IR. Therefore, CGRP is a histochemical marker for a larger number of unmyelinated cutaneous afferents in human skin than is SP. CGRP itself may have a role in the mediation of responses to stimulation of at least two populations of sensory axons.

Heterogeneity of cold-stable and cold-labile tubulin in axon- and soma-enriched portions of the adult mouse brain

Microtubules from the optic nerve (axonal tubulin) and lateral geniculate nucleus (cell tubulin) were separated by cold treatment and the cold-soluble fraction was purified in the presence of Taxol. Isoforms of cold-stable and cold-soluble tubulin were resolved by the use of high-resolution isoelectric focusing. The cold-soluble fraction of axonal tubulin has only 14 of the 20 isotypes seen in the same fraction of cell tubulin. The two cold-stable fractions have 20 isotypes but axonal tubulin has a specific pattern of isotypes 1, 2 and 5. Cold-stable fractions of both axonal and cell tubulin display the existence of an intensely stained alpha-isotype, isotype 7, which seems associated with the property of cold stability. Our results highly favor the hypothesis of a physiological role of the heterogeneity of tubulin in neuronal microtubules.

Posttranslational modifications of alpha-tubulin: acetylated and detyrosinated forms in axons of rat cerebellum

The distribution of acetylated alpha-tubulin in rat cerebellum was examined and compared with that of total alpha-tubulin and tyrosinated alpha-tubulin. From immunoperoxidase-stained vibratome sections of rat cerebellum it was found that acetylated alpha-tubulin, detectable with monoclonal 6-11B-1, was preferentially enriched in axons compared with dendrites. Parallel fiber axons, in particular, were labeled with 6-11B-1 yet unstained by an antibody recognizing tyrosinated alpha-tubulin, indicating that parallel fibers contain alpha-tubulin that is acetylated and detyrosinated. Axonal microtubules are known to be highly stable and the distribution of acetylated alpha-tubulin in other classes of stable microtubules suggests that acetylation and possibly detyrosination may play a role in the maintenance of stable populations of microtubules.

Colocalization of microtubule-associated protein 1A and microtubule-associated protein 2 on neuronal microtubules in situ revealed with double-label immunoelectron microscopy

Microtubule-associated protein 1A (MAP1A) and microtubule-associated protein 2 (MAP2) were shown to be colocalized on the same microtubules (MTs) within neuronal cytoskeletons by double-label immunoelectron microscopy. To investigate the electron microscopic disposition of MAP1A and MAP2 and their relationship to MTs in vivo, and to determine whether there are different subsets of MTs which specifically bind either MAP1 or MAP2, we employed a double-label immunogold procedure on rat cerebella using mouse monoclonal antibody against rat brain MAP1A and affinity-purified rabbit polyclonal antibody against rat brain MAP2. MAP1A and MAP2 were identified with secondary antibodies coupled to 10- and 5-nm gold particles, respectively. In Purkinje cell dendrites, both 10- and 5-nm gold particles were observed to be studded on the fuzzy structures attached to the same MTs. Many such structures connected MTs to each other. There was no particular MT which bound either MAP1A or MAP2 alone. Furthermore, there seemed to be no specific regions on MTs where either MAP1A or MAP2 was specifically attached. Hence, we conclude that MAP1A and MAP2 are colocalized on MTs in dendrites and assume that MAP1A and MAP2 have some interrelationship in vivo and that their interactions are responsible for forming the network of cross-bridges between MTs and MTs in neuronal cytoskeletons.

N-acetylaspartylglutamate identified in the rat retinal ganglion cells and their projections in the brain

N-Acetylaspartylglutamate like immunoreactivity (NAAG-L) was identified in retinal ganglion cell bodies and their axons. The presence of the dipeptide in ganglion cell projection areas, the lateral geniculate nucleus (LGN) and superior colliculus (SC), was confirmed following NAAG purification from these tissues by a high-performance liquid chromatographic method. NAAG-L was identified in the optic tract as well as within fibers and puncta in the LGN and SC. The hypothesis that NAAG is present within ganglion cell axons in the brain was tested by unilateral enucleation which resulted in loss of NAAG and NAAG-L within the contralateral LGN and SC.

A group of related surface glycoproteins distinguish sets and subsets of sensory afferents in the leech nervous system

The distribution of 4 surface glycoproteins on axons of peripheral neurons was studied in the leech Hirudo medicinalis through monoclonal antibodies. All 4 glycoproteins have a similar molecular weight of 130 kDa. Immunohistochemical localization of these glycoproteins on tissue sections of nerves and neuropil reveals tracts of afferent axons organized as nested sets. Their distribution suggests a possible role for these molecules in mediating axon fasciculation.

Human amnion membrane as a substratum for cultured peripheral and central nervous system neurons

We report here on the use of human amnion membrane as a substratum for the culture of neuronal cells. Pieces of amnion membrane were bound to nitrocellulose paper as a supporting material, seeded with neurons, and cultured for 1-4 days. Neurons and neurites were visualized after fixation by immunoperoxidase staining using an anti-neurofilament monoclonal antibody. Neurons from embryonic chick ciliary and dorsal root ganglia and fetal rat hippocampus were cultured on either the basement membrane or stromal surface of the amnion membrane. Neurons adhered to both surfaces but extended neurites only on the basement membrane surface. Neurons survived and continued to grow neurites in serum-free medium for at least 4 days. When cultured for 4 days on the basement membrane surface in the presence of 10% fetal calf serum, the ciliary ganglion neurons survived but neurite growth was markedly inhibited, while dorsal root ganglion neurons survived, hypertrophied and grew an extensive network of neurites. Other experiments addressed the question whether the basement membrane surface had an ability to guide growing neurites. Amnion membranes were folded, frozen, cross-sectioned using a cryostat, and placed on the nitrocellulose to give irregular patterns of basement membrane juxtaposed to the collagenous stroma. Ciliary ganglion neurons after 4 days in culture had initiated and extended neurites in patterns which corresponded and were limited to those areas visualized by indirect immunofluorescence staining using anti-laminin antibodies. Thus, in vivo assembled human amnion basement membranes appear to contain signals that both promote and guide neuritic growth from previously axotomized embryonic peripheral and central nervous system neurons. The amnion membrane represents a novel tool for the culture of neuronal cells in vitro and potentially could be used as a neurite-promoting bridging material in vivo for regeneration studies.

Light and electron microscopic study of somatostatin-like immunoreactive neurons in rat hippocampus

The distribution and fine structure of somatostatin-like immunoreactive neurons in rat hippocampus and gyrus dentatus were investigated by light and electron microscopic immunocytochemistry. Somatostatin-like immunoreactive neuronal perikarya and fibers could be visualized by using modified PAP immunocytochemistry. Immunoreactive neurons were selectively localized in the stratum orience of the hippocampus and hilar region of the gyrus dentatus, however, immunoreactive neurons were also sparsely observed throughout hippocampal formation. Somatostatin-like immunoreactive varicosities were abundantly distributed in the stratum pyramidale and some of them were considered to terminate on the pyramidal cells. Somatostatin-like immunoreactive neurons in the hippocampal formation generally contained many mitochondria, well-developed rough surfaced endoplasmic reticulum (rER), polysomes and some dense granules. Immunoreactivity was observed especially in the dense granules and membrane of rER. Pre- and post-synaptic elements of somatostatin-like immunoreactive neurons were detected throughout the hippocampal formation.

Cytochemical characteristics of the axon membrane at nodes of Ranvier in resting, tetrodotoxin-blocked, and electrically stimulated peripheral nerves of the rat

To test whether intraaxonal ferric ion-ferrocyanide staining at nodes of Ranvier is influenced by functional state of the nodal membrane, the tibial nerves of Sprague-Dawley rats were subjected to one of the following experimental procedures prior to fixation and staining: General anesthesia to induce nerves at rest, tetrodotoxin blocking of nerve activity, and high-frequency (100 Hz) electrical stimulation. In all cases most but not all nodes were stained. No statistically significant differences were found either in the percentage of total stained nodes or in the percentage of stained nodes from large- and small-diameter fibers among the three conditions tested. An explanation is offered to account for the apparent discrepancy between these results and those from other studies involving related cytochemical markers of the nodal apparatus.