Fibroblast growth factor-2 inhibits myelin production by oligodendrocytes in vivo

Fibroblast growth factor-2 (FGF-2) controls in part the timely differentiation of oligodendrocytes into the myelin-producing cells of the CNS. However, although differentiated oligodendrocytes express FGF receptors (R), the effect of FGF-2 on myelin-producing oligodendrocytes in vivo was unknown. In the present study, we show that delivery of FGF-2 into the cerebrospinal fluid of anaesthetized rat pups, aged postnatal day (P) 6 to 9, induced a severe loss of myelin in the caudal anterior medullary velum (AMV). Furthermore, we show that the caudal AMV was myelinated at the time of treatment, so the effects of FGF-2 represent a loss of myelin and not delayed differentiation. This was confirmed by injection of platelet-derived growth factor-AA (PDGF-AA), a factor known to affect the differentiation of PDGF-alphaR expressing oligodendrocyte progenitors, but which did not induce myelin loss in the caudal AMV and did not affect differentiated oligodendrocytes, which do not express PDGF-alphaR. Compared to controls treated with saline or PDGF-AA, FGF-2 induced an accumulation of PLP protein and MBP mRNA within the somata of myelin-producing oligodendrocytes. The results indicate that FGF receptor signalling disrupts myelin production in differentiated oligodendrocytes in vivo and interrupted the transport of myelin-related gene products from the oligodendrocyte cell body to their myelin sheaths.     

Fibroblast growth factor 2 induces loss of adult oligodendrocytes and myelin in vivo

Oligodendrocytes are the myelin-forming cells of the CNS and are lost in demyelinating diseases such as multiple sclerosis (MS). A role for fibroblast growth factor 2 (FGF2) has been proposed in the pathogenesis of demyelination and the failure of remyelination in experimental models of MS. However, the in vivo effects of FGF2 on oligodendrocytes and oligodendrocyte progenitors (OPCs) in the adult CNS had not previously been determined. To address this, FGF2 was delivered into the cerebrospinal fluid (CSF) of the IVth ventricle and its actions were examined on the anterior medullary velum (AMV), a thin tissue that partly roofs the IVth ventricle and is bathed by CSF. FGF2 was administered twice daily for 3 days and AMV were analysed using immunohistochemical labelling; saline was administered in controls. The results show that raised FGF2 induces severe disruption of mature oligodendrocytes and a marked loss of myelin. At the same time, FGF2 treatment resulted in the aberrant accumulation of immature oligodendrocytes with a premyelinating phenotype, together with NG2-expressing OPCs. Axons are patent within demyelinated lesions, and they are contacted but not ensheathed by surviving oligodendrocytes, newly formed premyelinating oligodendrocytes and OPCs. These results demonstrate that raised FGF2 induces demyelination in the adult CNS, and support a role for FGF2 in the pathogenesis of demyelination and regulation of remyelination in MS.     

Basic fibroblast growth factor down-regulates myelin basic protein gene expression and alters myelin compaction of mature oligodendrocytes in vitro

The effects of basic fibroblast growth factor (bFGF) on myelin basic protein (MBP) gene expression and myelin-like membrane formation were investigated in oligodendrocyte cultures containing mainly mature oligodendrocytes expressing MBP. These cultures were obtained by selective detachment of the cells of the oligodendrocyte lineage from 40-day-old mixed cultures derived from newborn rat brain. They were further purified by a 3-day pretreatment with cytosine arabinoside (ARA-C) in order to kill cycling cells. After withdrawal of ARA-C, daily treatment of the cells with bFGF for 3 days induced a drastic decrease in MBP mRNA level compared to control cultures treated only with ARA-C. Moreover, the percentage of oligodendrocytes labelled with anti-MBP antibodies decreased by 50%, as well as that of oligodendrocytes expressing myelin oligodendrocyte glycoprotein (MOG), whereas proteolipid protein (PLP) immunolabelled cells were less affected. At the ultrastructural level, myelin-like membranes were still abundant in the ARA-C-and bFGF-treated cultures, but they were conspicuously uncompacted compared to cultures only pretreated with ARA-C. These results bring the first evidence that bFGF is able to down-regulate myelin protein gene expression in mature oligodendrocytes and to alter myelin structure. They imply that if bFGF is secreted after a demyelinating lesion of the central nervous system (CNS), this plasticity of mature oligodendrocytes will allow final remyelination of axons to complete only after this factor has returned to low levels. © 1995 Wiley-Liss, Inc.     

Differential expression of fibroblast growth factor receptors in the developing murine choroid plexus

The choroid plexuses (CPs) are specialised secretory organs situated within the ventricles of the brain involved in the production of cerebrospinal fluid (CSF) and the maintenance of the blood-CSF barrier. Abnormal function of the CPs can lead to hydrocephalus and raised intracranial pressure, pathologies frequently observed in certain craniofacial syndromes caused by single point mutations in fibroblast growth factor receptors (FGFRs). At present, relatively little is known about the embryonic CPs in terms of gene or protein expression, function as the brain develops or on the potential role of FGFRs within this context. Given the limited information available on the regulation of FGFRs during development of the CPs and periventricular tissues, we have carried out a detailed analysis of the localisation of FGFR1, 2, 3 and 4 proteins in these regions of the murine embryo from the time of formation of the CP in the third ventricle at E12.5 throughout the second half of gestation, and examined the expression of different FGFR isoforms at E12.5 by RT-PCR. We show here that FGFR1 and FGFR4 are expressed in murine CPs at E12.5 but not at E15.5 or E18.5, suggesting a role for the signaling pathways transduced by these receptors at early stages of CP development. In contrast, FGFR2 expression is maintained throughout CP development, indicating that this receptor may play a role in the function of immature and mature CP. Also FGFR3 is detected at each developmental stage studied, but surprisingly its expression appears confined to the nuclei of CP cells, suggesting that FGFR3 in the CP does not respond to extracellular FGFs but may act in intracrine fashion.     

Epidermal growth factor inhibits the expression of myelin basic protein in oligodendrocytes

The major function of the oligodendrocyte is to myelinate axons in the central nervous system (CNS). Two of the components of myelin, galactocerebroside (galc) and myelin basic protein (MBP), have been used as markers of oligodendrocyte maturation in the developing CNS, and it has been found that galc+ cells arise initially, which then mature into MBP+ oligodendrocytes several days later. We have been interested in the control of expression of MBP and have followed its appearance in cultures of brain cells isolated from 4 day-old mice. In low serum (0.5% foetal bovine serum), approximately 330 MBP+ cells arise per 2 x 10(5) brain cells after 3 days incubation. We have examined the ability of several growth factors to influence the expression of MBP in these cultures, including epidermal growth factor (EGF), platelet-derived growth factor (PDGF), and the fibroblast growth factors (acidic and basic FGF). EGF was found to suppress strongly the developmental expression of MBP in these cultures, but the suppression was reversible, since the number of MBP+ cells approached control numbers 3 days after removal of EGF from the cultures. It was also found that MBP could be down-regulated in mature MBP+ oligodendrocytes. The action of EGF in these cultures could be mimicked by transforming growth factor-alpha (TGF alpha). The effects of EGF appear to be associated primarily with MBP production in oligodendrocytes since expression of galc is unaffected by EGF.     

Fibroblast growth factor is a mitogen for oligodendrocytes in vitro

The effect of fibroblast growth factor (FGF) on oligodendrocyte development has been studied using dissociated mixed brain cells, cultured in a previously described serum-free medium. A greater number of galactocerebroside-positive oligodendrocytes could be demonstrated after 7 days in the presence of FGF than in control values. Using combined immunofluorescence and autoradiography an increased [3H]thymidine incorporation by galactocerebroside-positive oligodendrocytes was demonstrated after various times of exposure to FGF.     

Glial expression of fibroblast growth factor-9 in rat central nervous system.

We examined the expression of fibroblast growth factor (FGF)-9 in the rat central nervous system (CNS) by immunohistochemistry and in situ hybridization studies. FGF-9 immunoreactivity was conspicuous in motor neurons of the spinal cord, Purkinje cells, and neurons in the hippocampus and cerebral cortex. In addition to the neuronal localization of FGF-9 immunoreactivity that we reported previously, the present double-label immunohistochemistry clearly demonstrated that the immunoreactivity was present in glial fibrillary acidic protein (GFAP)-positive astrocytes preferentially present in the white matter of spinal cord and brainstem of adult rats and in CNPase-positive oligodendrocytes that were arranged between the fasciculi of nerve fibers in cerebellar white matter and corpus callosum of both adult and young rats. There was a tendency for FGF-9 immunoreactivity in oligodendrocytes to be more pronounced in young rats than in adult rats. The variation of oligodendrocyte FGF-9 immunoreactivity in adult rats was also more pronounced than that in young rats. With in situ hybridization, FGF-9 mRNA was observed in astrocytes in the white matter of rat spinal cord and oligodendrocytes in the white matter of cerebellum and corpus callosum of adult and young rats. The expression of FGF-9 mRNA in glial cells was lower than in neurons, and not all glial cells expressed FGF-9. In the present study, we demonstrated that FGF-9 was expressed not only in neurons but also in glial cells in the CNS. FGF-9 was considered to have important functions in adult and developing CNS.     

Developmental expression of myelin proteins by oligodendrocytes in the CNS of trout.

Using immunohistochemical techniques, the pattern of cytoplasmic staining and the temporal order of expression of 5 major myelin components of oligodendrocytes were studied in the developing central nervous system of trout. The two myelin glycoproteins, IP1 and IP2, in the cytoplasm of glial cells showed a granular pattern of immunostaining, whereas the 36K protein was homogeneously distributed. Analysis of freshly dissociated cells during early stages of myelinogenesis revealed a constant chronological sequence of expression of myelin proteins by the oligodendrocytes: glycoprotein IP2 was the first protein to appear during glial development together with the galactocerebroside GalC at stage 28 followed by the 36K at stage 30 and finally IP1 at stage 32. The deposition of myelin proteins into the nascent myelin sheath occurred in the same chronological order as their expression by oligodendrocytes. Moreover myelin basic protein, which was not detectable in glial cells, on tissue sections was found to appear in parallel with IP2.     

Fibroblast growth factor signaling in the developing neuroendocrine hypothalamus

Fibroblast growth factor (FGF) signaling is pivotal to the formation of numerous central regions. Increasing evidence suggests FGF signaling also directs the development of the neuroendocrine hypothalamus, a collection of neuroendocrine neurons originating primarily within the nose and the ventricular zone of the diencephalon. This review outlines evidence for a role of FGF signaling in the prenatal and postnatal development of several hypothalamic neuroendocrine systems. The emphasis is placed on the nasally derived gonadotropin-releasing hormone neurons, which depend on neurotrophic cues from FGF signaling throughout the neurons’ lifetime. Although less is known about neuroendocrine neurons derived from the diencephalon, recent studies suggest they also exhibit variable levels of dependence on FGF signaling. Overall, FGF signaling provides a broad spectrum of cues that ranges from genesis, cell survival/death, migration, morphological changes, to hormone synthesis in the neuroendocrine hypothalamus. Abnormal FGF signaling will deleteriously impact multiple hypothalamic neuroendocrine systems, resulting in the disruption of diverse physiological functions.     

Differential expression of fibroblast growth factor-2 and fibroblast growth factor receptor 1 in a scarring and nonscarring model of CNS injury in the rat

Injury to the adult brain results in abortive axon regeneration and the deposition of a dense fibrous glial scar. Therapeutic strategies to promote postinjury axon regeneration are likely to require antiscarring strategies. In neonatal brain wounds, scar material is not laid down and axons grow across the lesion site, either by de novo growth or regeneration. To achieve the therapeutic goal of recapitulating the nonscarring neonatal response in the injured adult, an understanding of how ontogenic differences in scarring reflect developmental diversities in the trophic response to injury is required. Fibrobast growth factor-2 (FGF-2) expression is developmentally regulated and has been implicated as a regulator of the wounding response of the adult rat central nervous system. We have investigated the expression of FGF-2 and fibroblast growth factor receptor 1 (FGFR1) after penetrating lesions to the cerebral cortex of 5 days post partum (dpp) (nonscarring) and 16 dpp and adult (scarring) rats. In situ hybridization, immunohistochemistry and Western blotting showed robust and sustained increases in FGF-2 and FGFR1 mRNA and protein in reactive astrocytes around the lesion in scarring rats, a response that was attenuated substantially in the nonscarring neonate. These results demonstrate that changes in astrocyte FGF-2 and FGFR1 expression are coincident with the establishment of a mature pattern of glial scarring after injury in the maturing central nervous system, but it is premature to infer a causal relationship without further experiments.     

Behavioral control of the stressor modulates stress-induced changes in neurogenesis and fibroblast growth factor-2

The controllability of stressors modulates many of the consequences of stressor exposure. Here, we used immunohistochemistry to examine neural progenitor cell proliferation and survival and basic fibroblast growth factor-2 in the hippocampus of male rats after controllable or uncontrollable tailshock. A series of identical tailshocks were delivered to yoked pairs of rats. One rat could terminate shocks to both rats of the pair. Reductions in neural progenitor cells were observed at 1-2 days and at 28 days in rats exposed to uncontrollable shock. Controllable shock produced an increase in fibroblast growth factor-2 in the dentate gyrus and CA1 2 h after stress and in the dentate gyrus 24 h after stress. Thus, stressor controllability modulates stress-induced decreases in neurogenesis and increases in fibroblast growth factor-2.     

Fibroblast growth factor modulates HIV coreceptor CXCR4 expression by neural cells. HNRC Group

Recent studies suggest that the chemokine receptor CXCR4 may be involved in mediating the neurodegenerative process in the brains of patients with acquired immunodeficiency disease (AIDS). In this context, we hypothesize that neurotrophic factors, such as fibroblast growth factor (FGF), might protect against human immunodeficiency virus (HIV)-mediated neurotoxicity via regulating the expression of CXCR4 in neural cells. For this purpose, levels of CXCR4 were determined in neuronal and glial cell lines after FGF1 and 2 treatment. In addition, levels of CXCR4 immunoreactivity were associated with levels of FGF1 immunoreactivity in the brains of HIV-positive patients. These studies showed that neuronal CXCR4 levels decreased in a dose-dependent manner after exposure to FGF. Conversely, glial CXCR4 was increased in a dose-dependent manner after FGF2 treatment. These effects were dependent on the FGF receptor tyrosine kinase signaling pathway, because FGF-induced effects on CXCR4 were blocked by the tyrosine kinase inhibitor, 5'-deoxy-5'methylthioadenosine, or by anti-FGF receptor antibody. Stromal cell-derived factor-1, the ligand for CXCR4, and HIV gp120 neurotoxicity was attenuated by FGF1 in a dose-dependent manner in vitro, further supporting physiological relevance. In the brains of AIDS patients, the levels of neural CXCR4 immunoreactivity were inversely associated with FGF levels. Taken together, these results support the possibility that the neuroactive effects of FGF in HIV encephalitis might be mediated through regulation of the expression of CXCR4.     

Fibroblast growth factor (FGF) levels in the developing rat brain

Acidic and basic fibroblast growth factors (FGF) are polypeptides with potent multipotential trophic effects on central nervous system (CNS) glia, endothelial cells, and neurons. These factors are characterized by strong binding to heparin, and are commonly assayed by their mitogenic activity on Balb/c 3T3 cells in vitro. We found a marked (ca. 13-fold) increase in Balb/c 3T3 mitogenic activity in the developing rat brain from the embryonic stage to the third postnatal week. High levels were sustained in the mature brain. Most of the mitogenic activity from rat brain bound strongly to heparin-affinity columns, and was eluted at positions characteristic of acidic FGF (aFGF) and basic FGF (bFGF). The presence of aFGF and bFGF in eluted peaks was confirmed by immunoblotting techniques using specific anti-FGF sera. Heparin-affinity high performance liquid chromatography (HPLC) showed a proportionately greater increase in levels of aFGF than bFGF between the tenth and fortieth postnatal days. Increases in FGF levels during late embryonic and early postnatal stages of brain development may play an important role in the glial and capillary proliferation, as well as in the neuronal outgrowth and synapse formation that is occurring during this time. The differential rates of accumulation of aFGF vs bFGF suggest different physiological roles for these factors in the developing brain.     

Neuronal localization of fibroblast growth factor-9 immunoreactivity in human and rat brain

Fibroblast growth factor-9 (FGF-9) is a relatively new member of the FGF family isolated from the conditioned medium of a human glioblastoma cell line as a secreting-type factor that exhibits a growth-stimulating effect on cultured glial cells. In order to elucidate the roles of FGF-9 in the central nervous system, we investigated in detail the distribution of FGF-9 proteins in the normal human and rat brains by immunohistochemistry using two different antibodies specific to FGF-9. In both human and rat, a strong expression of FGF-9 immunoreactivity was localized mainly in neurons throughout the normal brain. Immunoreactive glial cells were rarely encountered. In the human brain, strong and uniform immunoreactivity was observed in neurons of cerebral cortex, hippocampus, substantia nigra, motor nuclei of the brainstem, and Purkinje cell layer. A detailed mapping in the rat brain showed a distribution of FGF-9 immunoreactivity in a widespread population of neurons, though the intensity varied between different locations and even among the same nucleus. The most prominent expression in rat was observed in neurons of the mitral cell layer of the olfactory bulb, red nucleus, mesencephalic trigeminal nucleus, motor trigeminal nucleus, facial nucleus, reticular nucleus and Purkinje cell layer. These findings suggest that FGF-9 plays an important role in the central nervous system and may have a potential function closely connected to neurons in the normal brain.     

Fibroblast growth factor-2 deficiency causes defects in adult hippocampal neurogenesis, which are not rescued by exogenous fibroblast growth factor-2

Neurogenesis within the adult brain is restricted to selected areas, one of which is the dentate gyrus (DG). Several growth factors have been reported to affect neurogenesis in the adult DG. However, a role of fibroblast growth factor-2 (FGF-2) in adult hippocampal neurogenesis has not been firmly established. We have analyzed neurogenesis in the DG using in vivo and in vitro approaches. FGF-2(-/-) mice revealed no alterations in the number of proliferating cells but a significant decrease in the numbers of newly generated neurons. Moreover, FGF-2 added to hippocampal slice cultures from FGF-2(-/-) mice was unable to rescue the phenotype. Although an increase in death of neurogenic cells in the FGF-2-deficient DG could not be specifically demonstrated, there was a massive increase in global cell death in FGF-2(-/-) hippocampal slice cultures compared with slices from wild-type mice. Cell death could not be prevented by addition of FGF-2. Neutralization of endogenous FGF-2 in hippocampal slices did not interfere with neurogenesis in a short-term paradigm. Together, our data suggest that FGF-2 is essentially required for maturation of new neurons in adult hippocampal neurogenesis but is likely to operate synergistically in combination with other mechanisms/growth factors.