Glial influence on nerve fiber formation from rat ventral mesencephalic organotypic tissue cultures

Rat fetal ventral mesencephalic organotypic cultures have demonstrated two morphologically different dopamine nerve fiber growth patterns, in which the initial nerve fibers are formed in the absence of astrocytes and the second wave is guided by astrocytes. In this study, the presence of subpopulations of dopamine neurons, other neuronal populations, and glial cells was determined. We used "roller-drum" organotypic cultures, and the results revealed that beta-tubulin-positive/tyrosine hydroxylase (TH)-negative nerve fibers were present as early as 1 day in vitro (DIV). A similar growth pattern produced by TH-positive neurons was present from 2 DIV. These neurites grew to reach distances over 4 mm and over time appeared to be degenerating. Thin, vimentin-positive processes were found among these nerve fibers. As the first growth was retracted, a second outgrowth was initiated and formed on migrating astrocytes. TH- and aldehyde dehydrogenase-1 (ALDH1)-positive nerve fibers formed both the nonglia-associated and the glia-associated outgrowth. In cultures with membrane inserts, only the glia-associated outgrowth was found. Vimentin-positive cells preceded migration of NG2-positive oligodendrocytes and Iba-1-positive microglia. Oligodendrocytes appeared not to be involved in guiding neuritic growth, but microglia was absent over areas dense with TH-positive neurons. In conclusion, in "roller-drum" cultures, nerve fibers are generally formed in two sequences. The early-formed nerve fibers grow in the presence of thin, vimentin-positive processes. The second nerve fiber outgrowth is formed on astroglia, with no correlation to the presence of oligodendrocytes or microglia. ALDH1-positive nerve fibers, presumably derived from A9 dopamine neurons, participate in formation of both sequences of outgrowth.     

Inhibition of proteoglycan synthesis affects neuronal outgrowth and astrocytic migration in organotypic cultures of fetal ventral mesencephalon

Grafting fetal ventral mesencephalon has been utilized to alleviate the symptoms of Parkinson's disease. One obstacle in using this approach is the limited outgrowth from the transplanted dopamine neurons. Thus, it is important to evaluate factors that promote outgrowth from fetal dopamine neurons. Proteoglycans (PGs) are extracellular matrix molecules that modulate neuritic growth. This study was performed to evaluate the role of PGs in dopamine nerve fiber formation in organotypic slice cultures of fetal ventral mesencephalon. Cultures were treated with the PG synthesis inhibitor methyl-umbelliferyl-beta-D-xyloside (beta-xyloside) and analyzed using antibodies against tyrosine hydroxylase (TH) to visualize dopamine neurons, S100beta to visualize astrocytes, and neurocan to detect PGs. Two growth patterns of TH-positive outgrowth were observed: nerve fibers formed in the presence of astrocytes and nerve fibers formed in the absence of astrocytes. Treatment with beta-xyloside significantly reduced the distance of glial-associated TH-positive nerve fiber outgrowth but did not affect the length of the non-glial-associated nerve fibers. The addition of beta-xyloside shifted the nerve fiber growth pattern from being mostly glial-guided to being non-glial-associated, whereas the total amount of TH protein was not affected. Further, astrocytic migration and proliferation were impaired after beta-xyloside treatment, and levels of non-intact PG increased. beta-Xyloside treatment changed the distribution of neurocan in astrocytes, from being localized in vesicles to being diffusely immunoreactive in the processes. To conclude, inhibition of PG synthesis affects glial-associated TH-positive nerve fiber formation in ventral mesencephalic cultures, which might be an indirect effect of impaired astrocytic migration.     

Inhibition of astrocytes promotes long-distance growing nerve fibers in ventral mesencephalic cultures

Tyrosine hydroxylase-positive nerve fiber formation occurs in two diverse morphological patterns in rat fetal ventral mesencephalic slice cultures; one is non-glial-associated and the other is glial-associated. The aim of this study was to characterize the non-glial-associated nerve fibers and its relation to migration of astrocytes. Organotypic slice cultures were prepared from embryonic days 12, 14, and 18 rat fetuses and maintained for 5, 7 or 14 days in vitro. Inhibition of cell proliferation using cytosine β-d-arabinofuranoside was conducted in embryonic day 14 ventral mesencephalic cultures. The treatment impaired astrocytic migration at 7 and 14 days in vitro. The reduced migration of astrocytes exerted a negative effect on the glial-associated tyrosine hydroxylase-positive nerve fibers, reducing the outgrowth from the tissue slice. The non-glial-associated outgrowth was, however, positively affected by reduced astrocytic migration, reaching distances around 3 mm in 2 weeks, and remained for longer time in culture. Co-cultures of fetal ventral mesencephalon and frontal cortex revealed the cortex as a target for the non-glial-associated tyrosine hydroxylase-positive outgrowth. The age of the fetal tissue at plating affected the astrocytes such that older tissue increased the length of astrocytc migration. Younger tissue at plating promoted the presence of non-glial-asscociated outgrowth and long radial-glia-like processes, while older tissue promoted migration of neurons instead of formation of nerve fiber network. In conclusion, inhibition of astrocytic proliferation promotes the persistence of long-distance growing tyrosine hydroxylase-positive nerve fibers in ventral mesencephalic slices cultures. Furthermore, the long-distance growing nerve fibers target the frontal cortex and are absent in cultures derived from older tissue.     

L1 substrate enhances outgrowth of tyrosine hydroxylase-immunoreactive neurites in mesencephalic cell culture.

We have evaluated neurite outgrowth from mesencephalic tyrosine hydroxylase-positive neurons grown in vitro on different substrates. Cultures of ventral mesencephalon from rat embryos (E13) were plated on plastic dishes coated with the following substrates: L1, L2/HNK-1 "residual" (mainly J1/160 but also tenascin), MAG antigens from mouse brains, laminin, fibronectin, poly-L-lysine, RGD peptide, and plastic alone. After 3, 4, and 6 days in vitro, the cultures were stained using an antibody against tyrosine hydroxylase (TH), and the length of TH-positive neurites was measured by computer-assisted image analysis in a double-blind fashion. L1 antigen had a significant positive effect on neurite outgrowth compared to the other substrates studied. Laminin and fibronectin were also favorable substrates. In cultures treated with cytosine arabinoside to prevent mitoses and glial proliferation, the positive effect of L1 was abolished, but laminin still had a stimulatory effect. These data indicate that L1 may be indirectly involved in differentiation or axonal elongation of substantia nigra dopaminergic neurons and suggest a complex effect involving both neurons and glia on dopaminergic neurite development.     

Protection from 1-methyl-4-phenylpyridinium (MPP) toxicity and stimulation of regrowth of MPP-damaged dopaminergic fibers by treatment of mesencephalic cultures with EGF and basic FGF

Several peptide growth factors can maintain survival or promote recovery of injured central neurons. In the present study, the effects of epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) on the toxicity produced by the dopaminergic neurotoxin, 1-methyl-4-phenylpyridinium (MPP+), were investigated in rat mesencephalic dopaminergic neurons in culture. High affinity [3H]DA uptake and morphometric analyses of tyrosine hydroxylase immunostained neurons were used to assess the extent of MPP+ toxicity, dopaminergic neuronal survival and growth of neurites. Consistent with previous reports, EGF and bFGF treatments stimulated neuritic outgrowth in dopaminergic neurons, increased DA uptake and enhanced their long-term survival in vitro. These growth factors also stimulated proliferation of astrocytes. The time course of EGF and bFGF effects on dopaminergic neurons coincided with the increase in glial cell density, suggesting that proliferation of glia mediates their trophic effects. Several findings from our study support this possibility. When MPP+ was applied to cultures at 4 days in vitro, before glial cells had proliferated, the damage to dopaminergic neurons was not affected by EGF or bFGF pretreatments. However, when cultures maintained in the presence of the growth factors for 10 days were exposed to MPP+, after they had become confluent with dividing glial cells, the MPP(+)-induced decreases in DA uptake and cell survival were significantly attenuated. Furthermore, when glial cell proliferation was inhibited by 5-fluoro-2'-deoxyuridine, the protective effects of EGF and bFGF against MPP+ toxicity were abolished. Continuous treatment of MPP(+)-exposed cultures with EGF or bFGF resulted in the stimulation of process regrowth of damaged dopaminergic neurons with concomitant recovery of DA uptake, suggesting that the injured neurons are able to respond to the trophic effects of EGF and bFGF. In summary, our study shows that the trophic effects of EGF and bFGF on mesencephalic dopaminergic neurons include protection from the toxicity produced by MPP+ and promotion of recovery of MPP(+)-damaged neurons. Stimulation of glial cell proliferation is necessary for these effects.     

Interleukin-1beta mediates GDNF up-regulation upon dopaminergic injury in ventral midbrain cell cultures

We recently proposed the involvement of diffusible modulators in signalling astrocytes to increase glial cell line-derived neurotrophic factor (GDNF) expression after selective dopaminergic injury by H2O2 or L-DOPA. Here we report that interleukin-1beta (IL-1beta) is involved in this crosstalk between injured neurons and astrocytes. IL-1beta was detected only in the media from challenged neuron-glia cultures. Exogenous IL-1beta did not change GDNF protein levels in astrocyte cultures, and diminished GDNF levels in neuron-glia cultures. This decrease was not due to cell loss, as assessed by the MTT assay and immunocytochemistry. Neither H2O2 nor L-DOPA induced microglia proliferation or appeared to change its activation state. The IL-1 receptor antagonist (IL-1ra) prevented GDNF up-regulation in challenged cultures, showing that IL-1beta is involved in the signalling between injured neurons and astrocytes. Since IL-1ra decreased the number of dopaminergic neurons in H2O2-treated cultures, we propose that IL-1 has a neuroprotective role in this system involving GDNF up-regulation.     

Expression of S-100 protein is related to neuronal damage in MPTP-treated mice

S-100beta is a calcium-binding protein expressed at high levels in brain and is known as a marker of brain damage. However, little is known about the role of S-100beta protein during neuronal damage caused by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). To determine whether S-100beta protein is induced in glial cells after MPTP treatment, we investigated the expression of S-100 protein immunohistochemically, using MPTP-treated mice. We also examined the change of neurons and glial cells in mice after MPTP treatment. The present study shows that tyrosine hydroxylase (TH) immunoreactivity decreased gradually in the striatum and substantia nigra from 1 day after MPTP treatment. Thereafter, TH-immunopositive cells and fibers decreased in the striatum and substantia nigra at 3 days after MPTP treatment. In contrast, S-100-immunopositive cells and glial fibrillary acidic protein (GFAP)-immunopositive cells increased markedly in the striatum and substantia nigra at 3 days after MPTP treatment. Seven days after MPTP treatment, S-100-immunopositive cells decreased in the striatum and substantia nigra. However, the number of GFAP-immunopositive cells increased in these regions. In double-labeled immunostaining with anti-S-100 and anti-GFAP antibodies, S-100 immunoreactivity was observed only in the GFAP-positive astrocytes. These results provide evidence that astrocytic activation may play a role in the pathogenesis of MPTP-induced degeneration of dopaminergic neurons. Furthermore, the present study demonstrates that S-100 protein is expressed selectively by astrocytes, but not by microglia, after MPTP treatment. These results provide valuable information for the pathogenesis of the acute stage of Parkinson's disease.     

Enhanced survival of apparent presynaptic elements on polylysine-coated beads by inhibition of non-neuronal cell proliferation

Increased survival of presynaptic-like neuronal profiles was found in cell cultures of rat cerebellum when the non-neuronal cell numbers were reduced with an antimitotic drug. In both treated and untreated cell cultures, neurites grew onto the polylysine-coated surface of sepharose beads and formed a swelling. The neuronal swelling contained an accumulation of synaptic vesicles and a membrane density at the site of contact with the bead and was called an apparent presynaptic element. The apparent presynaptic elements in untreated cultures increased in number from the time the beads were added to the culture to 7 days incubation and then showed a decrease to one half the 7-day value at 14 days incubation. A 75% reduction in cell division of non-neuronal cells was seen in cultures exposed to a 5 X 10(-6)M cytosine arabinoside (Ara-C) for 2 days. Adding polylysine-coated beads to cultures treated with Ara-C showed at 14 days incubation a 7-fold increase in the number of apparent presynaptic elements as compared to untreated cultures. Additional experiments examined the numbers of neurites on the beads and found only small differences between treated and untreated cultures. A decrease, however, was shown in the number of glial fibrillary acidic protein staining astrocytes on the surface of the beads in treated cultures. The reduction of astrocytes by Ara-C appeared to enhance the survival of apparent presynaptic elements but did not enhance the growth of neurites. These results suggest that proliferating non-neuronal cells at a site of injury in the central nervous system may inhibit the formation of synaptic contacts and the growth of neurites through the site of injury.     

A novel way of removing quiescent astrocytes in a culture of subcortical neurons grown in a chemically defined medium

A new method has been described for removing a very small number of contaminating astrocytes in neuronal cultures (derived from the septal-diagonal band region of 17-day-old embryonic rat brain) grown in a chemically defined medium. The proportion of these glial fibrillary acidic protein (GFAP)-positive cells was usually less than 1.5% up to 10 days, but thereafter their number increased rapidly reaching 10-15% by 22 days in vitro. A prolonged exposure to normally used concentration of cytosine arabinoside (Ara-C; 10 microM) was toxic to both astroglial and neuronal cells, while a brief treatment (48 h) with a low level (4 microM) of Ara-C failed to eliminate these astrocytes, as judged by glutamine synthetase activity and GFAP-positive cell count. However, these quiescent astroglial cells could be easily eliminated if they were induced to proliferate by epidermal growth factor before exposure to Ara-C. The combined treatment with these agents had no effect on the number of acetylcholinesterase-positive cells, and on the development of cholinergic and GABA-ergic neurons, as measured in terms of choline acetyltransferase and glutamate decarboxylase activity, respectively.     

Differential regulation of gliogenesis in the context of adult hippocampal neurogenesis in mice

In adult hippocampal neurogenesis, new neurons appear to originate from a cell with astrocytic properties expressing glial fibrillary acidic protein (GFAP). Also, new astrocytes are generated in the adult dentate gyrus. Whereas the putative astrocyte-like progenitor cells are consistently S-100beta-negative, many new astrocytes are S-100beta-positive. Thus, it is unclear whether the GFAP-positive progenitor cells are astrocytes in a general sense or rather neural progenitor cells with certain astrocytic characteristics. We therefore investigated the development of GFAP-expressing cells in the context of adult hippocampal neurogenesis. Proliferating cells could be either GFAP-positive or doublecortin-positive (DCX), but never both, indicating two independent populations of dividing cells in the glial and neuronal lineages. Two distinct populations of cells with astroglial properties were detected-one expressing GFAP, the other co-expressing GFAP and S-100beta. We never found S-100beta-cells to be in S-phase. No overlap between neuronal and glial markers was seen at any time point. Thus, astrogenesis occurred in parallel and to some degree independent of adult neurogenesis. The uninterrupted GFAP expression in this lineage, and neuronal markers in the other lineage, argue against a late common precursor for neurogenesis and gliogenesis in the adult hippocampus. Very few newly generated microglia and no new oligodendrocytes were detected. Environmental enrichment and voluntary wheel running-two experimental paradigms with robust stimulatory effects on adult hippocampal neurogenesis-affected hippocampal astrogenesis differentially: Running, but not enrichment, strongly induced net astrogenesis (GFAP/S-100beta), but also GFAP-positive S-100beta-negative cells, which thus appear to be a transiently amplifiable intermediate population within the glial lineage.     

Stimulation of the regrowth of MPP+-damaged dopaminergic fibers by the treatment of mesencephalic cultures with basigin

Summary. Basigin (Bsg) is a transmembrane glycoprotein belonging to the immunoglobulin superfamily and widely expressed in the central nervous system. To elucidate functional role of Bsg in the central nervous system, the effects of its glutathione-S-transferase (GST) fusion protein on the number and neurite outgrowth of cultured rat mesencephalic dopaminergic neurons were measured. The fusion protein was not able to promote the survival and neurite outgrowth of tyrosine hydroxylase (TH)-positive neurons under serum-free condition. However, the treatment of 1-methyl-4-phenylpyridinium (MPP⁺)-exposed cultures with the fusion protein resulted in stimulation of the regrowth of damaged TH-positive fibers. Basic fibroblast growth factor (bFGF) also stimulated the regrowth of neurites in damaged neurons. These results indicate that Bsg may play an important role in the regrowth of damaged dopaminergic fibers. Received May 9, 2001; accepted July 2, 2001     

Astrocytic reactions in spinal gray matter following sciatic axotomy

Astrocytic responses following unilateral sciatic nerve axotomy were examined in the spinal gray matter. Using an antiserum to glial fibrillary acidic protein (GFAP), immunoreactive astrocytes were studied in both dorsal and ventral gray matter at intervals from 2 days through 34 days post-axotomy. In all axotomized animals, increased numbers of strongly immunoreactive astrocytes were present in the gray matter ipsilateral to the surgery. Such astrocytes were absent from the contralateral intact side and from gray matter bilaterally in adjacent spinal segments not involved in formation of the sciatic nerve. These GFAP-positive astrocytes occurred not only in association with large motor neurons in the ventral gray matter but also in association with central processes of dorsal root ganglion neurons in the dorsal gray matter. The response was quite rapid, being discernible both dorsally and ventrally as early as the second post-operative day. This increased GFAP immunoreactivity persisted throughout the entire observation period, with the perikarya of large ventral motor neurons appearing to become surrounded or encapsulated by the immunoreactive processes. A further alteration noted at the longest post-operative intervals was the presence in the ventral gray matter of astrocytes appearing to be binucleate. The data obtained indicate that the astrocytic response is not related solely to reactions in motor neurons and, furthermore, the rapidity with which it develops in the dorsal gray matter suggests that its induction is not dependent upon transganglionic degeneration, which others have reported to occur weeks after peripheral nerve injury.     

Vimentin-positive astrocytes in canine distemper: a target for canine distemper virus especially in chronic demyelinating lesions?

In canine distemper demyelinating leukoencephalitis (DL), caused by canine distemper virus (CDV), astrocytes represent the main virus target. In these cells, glial fibrillary acidic protein (GFAP) is the main intermediate filament, whereas vimentin occurs early in the astrocytic lineage and is replaced gradually by GFAP. To further characterize the role of astrocytic infection in dogs with DL, an animal model for multiple sclerosis, formalin-fixed paraffin-embedded cerebella were investigated immunohistochemically and by immunofluorescence. The expression and morphological alterations of these intermediate filaments were also determined by immunofluorescence studies of CDV-infected canine mixed brain cell cultures. In acute distemper lesions, the astrocytic response was mainly composed of GFAP- and CDV-positive cells. In contrast, vimentin-positive astrocyte-like cells were present in advanced lesions, which represented the main cell type harboring the pathogen, indicating a change in cell tropism and/or susceptibility of glial cells during lesion progression in CDV encephalomyelitis. Canine cell cultures were composed of GFAP-positive astrocytes, vimentin-positive cells and other glial cells. Following infection with the CDV-R252 strain, GFAP-positive astrocytes, especially multinucleated syncytial giant cells, displayed a disrupted cytoskeleton, whereas vimentin-positive cells though more frequently infected did not show any alteration in the filament network. This indicates increased vulnerability of mature GFAP-positive astrocytes compared to immature, vimentin-positive astrocytes. The latter, however, exhibited increased susceptibility to CDV. To conclude, the present findings indicate a change in cell tropism of CDV and/or the occurrence of less differentiated astrocytes representing a permanent source for virus infection and spread in advanced lesions of DL.     

Do Rosenthal fibers contain glial fibrillary acid protein?

Summary Sixteen cases of pilocytic astrocytomas with excessive Rosenthal fiber (RF) formation were examined by the indirect immunoperoxidase method for the localization of glial fibrillary acid protein (GFAP). RF nerve contained GFAP but they were often enclosed in plump and thickened GFAP-positive astrocytic processes. The border between the negative RF and the surrounding positive rim of cytoplasm was always sharp and without gradual transitions. The antigenic difference between RFs and glial filaments imply that glial filaments undergo a profound change in their chemical composition during their transformation into RFs. The possibility that RFs are not degradation products of glial filaments but consist of some chemically unknown substance produced by metabolically activated astrocytes cannot be excluded.     

Neurotrophic effects of basic and acidic fibroblast growth factors are not mediated through glial cells

Basic and acidic fibroblast growth factors (bFGF, aFGF) increase the survival of fetal hippocampal pyramidal neurons in serum-free cultures. bFGF is also a mitogen for astrocytes either in highly purified glial cultures or as a contaminant in neuronal cultures. The possibility that bFGF enhances neuronal survival indirectly through stimulating glial proliferation is unlikely. In the presence of 1 ng/ml bFGF, the total number of contaminating astrocytes (as defined by immunohistochemical staining for glial fibrillary acidic protein (GFAP] was increased to 4.3% vs 0.9% in control hippocampal cultures. aFGF did not significantly increase astrocyte number while supporting neuronal survival. Two other agents which stimulated equal or greater astrocytic proliferation, epidermal growth factor (EGF) and 10% serum, did not support neurons, and bFGF still significantly increased neuronal survival in their presence. When glial proliferation was inhibited by aphidicolin, contamination decreased to 0.1% in controls and 1.0% with 1 ng/ml bFGF, yet the neurons remained responsive to FGF. Cultures lacking any detectable GFAP-positive cells were identified, and even in the absence of glial cells, aFGF and bFGF increased neuronal survival. Because there is no significant correlation between the neuronal response and astrocyte number, it appears that bFGF and aFGF can directly support neuronal survival.     

Ciliary neurotrophic factor activates spinal cord astrocytes, stimulating their production and release of fibroblast growth factor-2, to increase motor neuron survival.

At focal CNS injury sites, several cytokines accumulate, including ciliary neurotrophic factor (CNTF) and interleukin-1beta (IL-1beta). Additionally, the CNTF alpha receptor is induced on astrocytes, establishing an autocrine/paracrine loop. How astrocyte function is altered as a result of CNTF stimulation remains incompletely characterized. Here, we demonstrate that direct injection of CNTF into the spinal cord increases GFAP expression and astroglial size and that primary cultures of spinal cord astrocytes treated with CNTF, IL-1beta, or leukemia inhibitory factor exhibit nuclear hypertrophy comparable to that observed in vivo. Using a coculture bioassay, we further demonstrate that CNTF treatment of astrocytes increases their ability to support ChAT(+) ventral spinal cord neurons (presumably motor neurons) more than twofold compared with untreated astrocytes. Also, the complexity of neurites was significantly increased in neurons cultured with CNTF-treated astrocytes compared with untreated astrocytes. RT-PCR analysis demonstrated that CNTF increased levels of FGF-2 and nerve growth factor (NGF) mRNA and that IL-1beta increased NGF and hepatocyte growth factor mRNA levels. Furthermore, both CNTF and IL-1beta stimulated the release of FGF-2 from cultured spinal cord astrocytes. These findings demonstrate that cytokine-activated astrocytes better support CNS neuron survival via the production of neurotrophic molecules. We also show that CNTF synergizes with FGF-2, but not epidermal growth factor, to promote DNA synthesis in spinal cord astrocyte cultures. The significance of these findings is discussed by presenting a new model depicting the sequential activation of astrocytes by cytokines and growth factors in the context of CNS injury and repair.     

Preconditioning injury-induced neurite outgrowth of adult rat sensory neurons on fibronectin is mediated by mobilisation of axonal alpha5 integrin

A preconditioning sciatic nerve crush promotes the capacity of adult sensory neurons to regenerate following a subsequent injury to their axons. The increase in regeneration is detected in cultures of dissociated neurons, as an earlier and enhanced rate of neurite elongation. We compare neurotrophin-stimulated neurite outgrowth from sensory neurons on laminin and fibronectin. There is a poor response of sensory neurons to fibronectin in comparison to laminin, but this is enhanced by a preconditioning lesion to the sciatic nerve 7 days prior to culture. By using specific integrin-binding fibronectin fragments and function-blocking antibodies, we demonstrate that the enhanced preconditioned neurite outgrowth on fibronectin is largely mediated by alpha5beta1 integrin. Preconditioning injury alter the subcellular localisation of alpha5 integrin in preconditioned neurites. We show that alpha5 integrin localises to adhesion complexes in the growth cone and neurites of preconditioned neurons, but not control neurons.     

Retinal neurite growth on astrocytes is not modified by extracellular matrix, anti-L1 antibody, or oligodendrocytes.

Two factors that may influence the course of axonal regeneration in the central nervous system (CNS) are extracellular matrix (ECM) and cell surface molecules that may enhance or inhibit neurite outgrowth. Whereas cultured astrocytes have been reported to be a good substratum for neurite outgrowth, there is recent evidence that cultured oligodendrocytes are inhibitory. To test the influences of 1) ECM components, 2) the L1 adhesion molecule, and 3) the inhibitory potential of mature oligodendrocytes in the astrocytic environment, we have utilized a culture system in which neurites from embryonic rat retina grow vigorously on astrocyte monolayers. The major ECM components were assembled in neonatal rat cortical astrocyte-retina co-cultures only when the medium contained serum. In electron microscopic studies of serum containing cultures, retinal neurites were seen to be related to astrocyte surfaces but rarely were found in contact with ECM; in serum-free medium the association between neurites and astrocytes was similar. In addition, the growth of neurites was vigorous whether ECM was present or absent. Presence of antibodies against the cell surface adhesion molecule L1 did not inhibit retinal neurite elongation on glial fibrillary acidic protein-positive astrocytes. When oligodendrocytes from adult rat spinal cord were combined with the astrocytes, retinal neurites grew as well on the mixed glial population as on astrocytes alone. Immunostaining for galactocerebroside showed many oligodendrocyte processes to be aligned in the direction of neurite growth, suggesting association between the two cell types. This association was verified by electron microscopy. Furthermore, retinal explants extended neurites among myelin basic protein-positive oligodendrocytes cultured without astrocytes. Thus, the astrocyte surface is a strong promoter of neurite growth from embryonic rat retina. This growth did not depend upon either ECM or the L1 adhesion molecule. Because neurites grew on astrocytes in the presence of mature oligodendrocytes or among oligodendrocytes alone, we conclude that oligodendrocytes do not inhibit neurite growth under certain conditions.     

The neurotrophic effects of fibroblast growth factors on dopaminergic neurons in vitro are mediated by mesencephalic glia

Neurotrophic support is generally believed to result from a direct action of growth factors on developing neurons. However, there is increasing evidence that growth factors can indirectly affect neuronal development by glial-mediated processes. To investigate a possible role of glia in mediating neurotrophic effects on dopaminergic neurons, four purified growth factors were screened for dual effects on the survival and differentiation of dopaminergic neurons and on the proliferation of mesencephalic glial cells in vitro. Dissociated embryonic day 14.5 rat mesencephalon was grown at low cell density without serum, conditions under which both glial growth and neuronal survival are not optimal. Treatment of these cultures with acidic fibroblast growth factor (aFGF) or basic fibroblast growth factor (bFGF) increased the number of surviving tyrosine hydroxylase-immunoreactive (TH-IR) neurons by 90-110% [corrected] at 8 d in vitro in a dose-dependent manner. The effects of these factors were not additive. High-affinity dopamine uptake was increased by bFGF, but not by aFGF. Length of TH-IR neurites was not affected by either aFGF or bFGF. Both growth factors induced proliferation of mesencephalic astrocytes as demonstrated by autoradiographic labeling with 3H-thymidine combined with immunocytochemistry for glial fibrillary acidic protein (GFAP). In contrast, platelet-derived growth factor (PDGF) and interleukin-1 had no effect on the survival or differentiation of dopaminergic neurons or the proliferation of mesencephalic astrocytes. Inhibition of glial proliferation abolished the neurotrophic effects exerted by aFGF or bFGF on dopaminergic neurons. Moreover, conditioned medium derived from mesencephalic glial cultures replated in the virtual absence of neurons also contained neurotrophic activity.(ABSTRACT TRUNCATED AT 250 WORDS)