Retinoic acid treated P19 embryonal carcinoma cells differentiate into oligodendrocytes capable of myelination

Retinoic acid treatment of P19 embryonal carcinoma cells induces their differentiation into cultures containing neurons and astrocytes. We present two lines of experimentation indicating that oligodendrocytes also develop from retinoic acid-treated P19 cells. We isolated an immortal cell line from retinoic acid-treated P19 cell cultures whose proliferation is dependent upon epidermal growth factor. Upon removal of the growth factor these cells differentiate into both astrocytes and oligodendrocytes as determined by immunostaining with antibodies to the astrocyte marker glial fibrillar acidic protein and the oligodendrocyte markers, myelin associated glycoprotein and 2', 3'-cyclic nucleotide 3'-phosphodiesterase. This cell line appears to be a bi-potential glial precursor. We also found that oligodendrocytes developed directly from P19 cells when retinoic acid-treated cells were transplanted into the brains of neonatal rat pups. Cells that developed into oligodendrocytes migrated into fiber bundles up to several millimeters from the site of the graft. These P19-derived oligodendrocytes appeared to myelinate axons from host neurons. Thus, retinoic acid-treated P19 cells differentiate into neurons, astrocytes and oligodendrocytes, the three cell types that normally develop from embryonic neuroectoderm, indicating that these cell cultures differentiate in a fashion closely resembling that of embryonic neuroectoderm.     

Immunocytochemical and ultrastructural characterization of type 1 astrocytes and 0-2A lineage cells in long-term co-cultures

We examined cultures of purified type 1 astrocytes and mixed glial co-cultures containing type 1 astrocytes and 0-2A lineage cells in media containing fetal calf serum at 5 days in vitro (DIV), 12 DIV, and 30 DIV, using cell-specific immunocytochemical markers and electron microscopy. At all three time points and in both culture systems, the polygonal-shaped type 1 astrocytes were A2B5-, GFAP+, and GalC-(specific markers for 0-2A lineage cells, and mature astrocytes and oligodendrocytes, respectively). From 5 to 30 DIV, the type 1 astrocytes increased markedly in size and the appearance of the cytoskeleton changed dramatically, with the amount of glial filaments increasing and microtubules decreasing. At 5, 12, and 30 DIV, the 0-2A lineage cells were multipolar, A2B5 +, HNK-1 +, GFAP-, and GalC-. The 0-2 lineage cells could not be distinguished as either astrocytes or oligodendrocytes on the basis of immunocytochemical or ultrastructural characteristics. These cells had dense cytoplasm, very few intermediate filaments, and a large number of vacuoles and dense bodies. The general characteristics of the cultured astrocytes at 12 DIV and 30 DIV were similar to mature and aged astrocytes in vivo, respectively. These findings suggest that the culture environment in this study accelerated aging of type 1 astrocytes. 0-2A lineage cells, on the other hand, appeared unable to differentiate into either type 2 astrocytes or oligodendrocytes when cultured in the presence of both type 1 astrocytes and fetal calf serum.     

Antigenic and differentiative heterogeneity among human glioblastomas

Increasing evidence suggests that in mammals, astrocytes are a heterogenous family of cells all of which share certain properties, but differ in lineage, biochemical and functional aspects. It seems likely that glioblastomas, arising from glial precursors, may also represent a family of related but distinct cell types. We have examined the antigenic characteristics and differentiative potential of 7 different human glioblastoma cell lines in vitro. All the cell lines were labeled with a monoclonal antibody 7B11 which labels all classes of astrocytes and their precursors in the rat CNS. U138MG and Tm3 cells expressed antigens on their surfaces recognized by the monoclonal antibodies A2B5 and HNK-1. When grown in serum-free medium in the presence of cAMP and theophylline, U138MG cells assumed a process-bearing morphology and some cells expressed the Gal-C antigen specific for oligodendrocytes. Under identical conditions, Tm3 cells converted to process-bearing cells, some of which expressed glial fibrillary acidic protein (GFAP) specific for astrocytes. Other cell lines with similar antigenic characteristics did not respond similarly to cAMP and theophylline. Finally, A2781 cells were GFAP immunoreactive and unlabeled by either A2B5 or HNK-1 antibodies. These observations suggest that individual glioblastoma cell lines may be derived from distinct glial precursor cells in the vertebrate CNS.     

Identification of the dopamine D3 receptor in oligodendrocyte precursors: potential role in regulating differentiation and myelin formation

Expression of the dopamine D3 receptor (D3r) was found in primary mixed glial cultures from newborn brain and in the corpus callosum in vivo during the peak of myelination. Expression of the D3r mRNA, but not D2r mRNA, was detected as early as 5 d in vitro (DIV) by RT-PCR. Immunoblot studies revealed D3r protein was also expressed in the cultures. Double immunofluorescence analysis for the D3r and for surface markers of specific stages of oligodendrocyte development indicated that D3r expression occurred in precursors and in immature oligodendrocytes but not in mature oligodendrocytes (i.e. , A2B5(+) 007(-) 01(-) and A2B5(+) 007(+) 01(-) cells but not A2B5(-) 007(+) 01(+) cells). Confocal microscopic analysis indicated that D3r was associated with cell bodies and cell membranes but not with the processes emanating from cell somas. Immunohistochemistry of brain sections revealed the presence of D3r in some oligodendrocytes located mainly within the genu and radiato of the corpus callosum during the active period of myelination. Treatment of cultures with 20 microM quinpirole led to decreased numbers of O1(+) oligodendrocytes possessing myelin-like membranes as well as an increase in the number of precursors in 14 DIV cultures. This effect was prevented by the dopamine antagonist haloperidol. These results show that the D3r expression is not restricted to neurons but it is also expressed in differentiating oligodendrocytes before terminal maturation. It also suggests that dopamine or some other D3r ligand may play a role in oligodendrocyte differentiation and/or the formation of myelin by mature oligodendrocytes.     

Spinal cord neuronal precursors generate multiple neuronal phenotypes in culture

Neuronal restricted precursors (NRPs) () can generate multiple neurotransmitter phenotypes during maturation in culture. Undifferentiated E-NCAM+ (embryonic neural cell adhesion molecule) immunoreactive NRPs are mitotically active and electrically immature, and they express only a subset of neuronal markers. Fully mature cells are postmitotic, process-bearing cells that are neurofilament-M and synaptophysin immunoreactive, and they synthesize and respond to different subsets of neurotransmitter molecules. Mature neurons that synthesize and respond to glycine, glutamate, GABA, dopamine, and acetylcholine can be identified by immunocytochemistry, RT-PCR, and calcium imaging in mass cultures. Individual NRPs also generate heterogeneous progeny as assessed by neurotransmitter response and synthesis, demonstrating the multipotent nature of the precursor cells. Differentiation can be modulated by sonic hedgehog (Shh) and bone morphogenetic protein (BMP)-2/4 molecules. Shh acts as a mitogen and inhibits differentiation (including cholinergic differentiation). BMP-2 and BMP-4, in contrast, inhibit cell division and promote differentiation (including cholinergic differentiation). Thus, a single neuronal precursor cell can differentiate into multiple classes of neurons, and this differentiation can be modulated by environmental signals.     

Differential expression of gap junctions in neurons and astrocytes derived from P19 embryonal carcinoma cells

The P19 embryonal carcinoma cell line represents a pluripotential stem cell that can differentiate along the neural or muscle cell lineage when exposed to different environments. Exposure to retinoic acid induces P19 cells to differentiate into neurons and astrocytes that express similar developmental markers as their embryonic counterparts. We examined the expression of gap junction genes during differentiation of these stem cells into neurons and astrocytes. Untreated P19 cells express at least two gap junction proteins, connexins 26 and 43. Connexin32 could not be detected in these cells. Treatment for 96 hr with 0.3 mM retinoic acid induced the P19 cells to differentiate first into neurons followed by astrocytes. Retinoic acid produced a decrease in connexin43 mRNA, protein, and functional gap junctions. Connexin26 message was not affected by retinoic acid treatment. The neurons that developed consisted of small round cell bodies extending two to three neurites and expressed MAP2. Connexin26 was detected at sites of cell-cell and cell-neurite contact within 3 days following differentiation with retinoic acid. The astrocytes were examined for production of their intermediate filament marker, glial fibrillary acidic protein (GFAP). GFAP was first detected at 8 days by Western blotting. In culture, astrocytes co-expressed GFAP and connexin43 similar to primary cultures of mouse brain astrocytes. These results suggest that differentiation of neurons and glial cells involves specific connexin expression in each cell type. The P19 cell line will provide a valuable model with which to examine the role gap junctions play during differentiation events of developing neurons and astrocytes.     

Monoclonal antibody 14F7, which recognizes a stage-specific immature oligodendrocyte surface molecule, inhibits oligodendrocyte differentiation mediated in co-culture with astrocytes

Cells at an intermediate stage of oligodendrocyte lineage are not only well characterized by biochemical studies but also are likely to relate to the outcome of physiological events. To elucidate the molecular events leading to the development of oligodendrocyte lineage cells, we have raised monoclonal antibodies against stage-specific immature oligodendrocytes, which have previously been isolated by a novel oligodendrocyte-lineage cell culture technique (Sakurai et al.: J Neurosci Res 52:17-26, 1998). We have isolated a mouse monoclonal antibody termed 14F7 which predominantly labels stage-specific immature oligodendrocytes and have found that the expression of 14F7 immunoreactivity in the developing neonatal rat forebrain is closely associated with cells expressing the oligodendrocyte progenitor marker A2B5 and to immature oligodendrocyte expressing O4 antigen. 14F7+ cells were distributed in the ventricular and subventricular zone and the nearby forming corpus callosum as non-myelinating cells. In contrast to cell culture observations, 14F7+ cells were seen only in oligodendrocyte lineage cells. For instance, dissociated cell culture studies indicated that 14F7 labels a cell surface molecule, and its cellular distribution is coincident with all of O4+ cells and A2B5+ cells, and even A2B5- cells. By contrast, 14F7-positive cells did not label astrocytes and, furthermore, did not label myelin basic protein (MBP)-positive oligodendrocytes. 14F7 recognized a 48-kDa protein on sodium dodecyl sulfate polyacrylamide gel electrophoresis. 14F7 immunoreactivity was detectable in rat brain as early as embryonic day 18. Furthermore, in these cells, the total time for differentiation was extended, and on maturation, these cells subsequently expressed an array of myelin-specific proteins, which normally occurs by direct contact with type-1 astrocytes. However, in the presence of 14F7, stage-specific oligodendrocytes co-cultured with astrocytes completely failed to express MBP. These data suggest that the 14F7 antigen is a novel cell surface molecule that is expressed in the intermediate stage of oligodendrocyte-lineage cells, and it is expected that it regulates the differentiation of oligodendrocyte throughout development.     

Iron-deficiency anemia and infant development: effects of extended oral iron therapy

Ceramide is a lipid second messenger implicated in the mechanism of apoptotic cell death. The effect of the cell-permeable ceramide analogue C2 has been tested on primary cortical cultures of neurons, astrocytes, and oligodendrocytes as well as on the bipotential glial precursor cell line CG-4. After 24 hr of treatment, C2 ceramide induced a dose-dependent cell death in primary oligodendrocytes and precursor cells, with a maximum effect at 10 microM. Commitment of oligodendrocytes to cell death occurred within the first 6 hr of treatment. Ultramicroscopic analysis of primary oligodendrocytes exposed to C2 ceramide for 3.5 hr revealed extensive membrane blebbing in the absence of nuclear condensation. In contrast, similar treatment of primary neuronal or astrocytic cortical cultures had no effect on cell survival. Neurons and astrocytes were resistant to 10 microM C2 ceramide. Furthermore, bipotential progenitors that were differentiated toward astrocytes also became resistant to ceramide treatment as they acquired a mature astrocytic phenotype. These experiments suggest that cell type specific factors are required for ceramide-mediated cell death in the nervous system.     

Accumulation of the cyclin-dependent kinase inhibitor p27/Kip1 and the timing of oligodendrocyte differentiation

Many types of vertebrate precursor cells divide a limited number of times before they stop and terminally differentiate. In no case is it known what causes them to stop dividing. We have been studying this problem in the proliferating precursor cells that give rise to postmitotic oligodendrocytes, the cells that make myelin in the central nervous system. We show here that two components of the cell cycle control system, cyclin D1 and the Cdc2 kinase, are present in the proliferating precursor cells but not in differentiated oligodendrocytes, suggesting that the control system is dismantled in the oligodendrocytes. More importantly, we show that the cyclin-dependent kinase (Cdk) inhibitor p27 progressively accumulates in the precursor cells as they proliferate and is present at high levels in oligodendrocytes. Our findings are consistent with the possibility that the accumulation of p27 is part of both the intrinsic counting mechanism that determines when precursor cell proliferation stops and differentiation begins and the effector mechanism that arrests the cell cycle when the counting mechanism indicates it is time. The recent findings of others that p27-deficient mice have an increased number of cells in all of the organs examined suggest that this function of p27 is not restricted to the oligodendrocyte cell lineage.     

In vitro and in vivo analysis of a rat bipotential O-2A progenitor cell line containing the temperature-sensitive mutant gene of the SV40 large T antigen

Primary cultures from neonatal optic nerve contain pluripotential O-2A progenitor cells that are capable of differentiating into oligodendrocytes, type-2 astrocytes or adult O-2A progenitors (O-2Aadult). Since primary optic nerve cultures contain a mixture of glial cell types of which only a small number are O-2A progenitors, experiments on cell lineage and differentiation carried out using these cultures are both intrinsically limited and difficult to interpret. Ideally, cells from a clonal cell population would provide the optimal starting material for biological studies. In this paper we describe the creation of an O-2A progenitor cell line using a retrovirus carrying a temperature-sensitive mutant SV40 large T antigen gene. This cell line has provided sufficient numbers of cells to allow analysis of their in vitro properties and their behaviour following transplantation into an in vivo environment. At the non-permissive temperature (39 degrees C), these cells differentiate into oligodendrocytes and type-2 astrocytes in a similar fashion to O-2A progenitor cells from primary cultures (O-2Aprim). When grown in media containing platelet-derived growth factor and basic fibroblast growth factor, the cell numbers can be expanded in culture without differentiating, consistent with the behaviour of O-2Aprim progenitor cells. By exploiting this property, it has been possible to culture large numbers of O-2A progenitors for in vivo analysis. In this study we have shown that transplantation of this O-2A cell line into glia-free areas in adult rat spinal cord results in differentiation of a proportion of cells into oligodendrocytes which are capable of myelinating axons. Furthermore, differentiation of O-2A cells into astrocytes was also observed, indicating that the bipotentiality of these cells in vitro can also be demonstrated in vivo.     

Expression of two neuronal markers, growth-associated protein 43 and neuron-specific enolase, in rat glial cells

Recent studies have revealed that proteins such as growth-associated protein 43 (GAP-43) and neuron-specific enolase (NSE), believed for many years to be expressed exclusively in neurons, are also present in glial cells under some circumstances. Here we present an overview of these observations. GAP-43 is expressed both in vitro and in vivo transiently in immature rat oligodendroglial cells of the central nervous system, in Schwann cell precursors, and in non-myelin-forming Schwann cells of the peripheral nervous system. GAP-43 mRNA is also present in oligodendroglial cells and Schwann cells, indicating that GAP-43 is synthesized in these cells. GAP-43 is also expressed in type 2 astrocytes (stellate-shaped astrocytes) and in some reactive astrocytes but not in type 1 astrocytes (flat protoplasmic astrocytes). These results suggest that GAP-43 plays a more general role in neural plasticity during development of the central and peripheral nervous systems. NSE enzymatic activity and protein and mRNA have been detected in rat cultured oligodendrocytes at levels comparable to those of cultured neurons. NSE expression increases during the differentiation of oligodendrocyte precursors into oligodendrocytes. In vivo, NSE protein is expressed in differentiating oligodendrocytes and is repressed in fully mature adult cells. The upregulation of NSE in differentiating oligodendrocytes coincides with the formation of large amounts of membrane structures and of protoplasmic processes. Similarly, NSE becomes detectable in glial neoplasms and reactive glial cells at the time when these cells undergo morphological changes. The expression of the glycolytic isozyme NSE in these cells, which do not normally contain it, could reflect a response to higher energy demands. This expression may also be related to the neurotrophic and neuroprotective properties demonstrated for this enolase isoform. NSE activity and protein and mRNA have also been found in cultured rat type 1-like astrocytes but at much lower levels than in neurons and oligodendrocytes. Thus GAP-43 and NSE should be used with caution as neuron-specific markers in studies of normal and pathological neural development.     

Both oligodendrocytes and astrocytes develop from progenitors in the subventricular zone of postnatal rat forebrain

The developmental fates of subventricular zone (SVZ) cells of the postnatal rat forebrain were determined by retroviral-mediated gene transfer and immunolabeling for glial antigens. A beta-galactosidase-containing retrovirus injected stereotactically into the SVZ infected small, immature cells. By 28 days post-injection labeled cells had appeared in both gray and white matter of the ipsilateral hemisphere. White matter contained labeled oligodendrocytes, but few astrocytes, while neocortex and striatum contained both glial types, often appearing in tightly knit clusters. An analysis after simultaneously injecting alkaline phosphatase- and beta-galactosidase-containing retroviruses showed that cells in each cortical cluster were related. Most clusters contained a single cell type, but approximately 15% contained both astrocytes and oligodendrocytes. These observations strongly suggest that a single SVZ cell can differentiate into both glial types.     

Glial cells are increased proportionally in transgenic optic nerves with increased numbers of axons

To study how an increase in axon number influences the number of glial cells in the mammalian optic nerve, we have analyzed a previously described transgenic mouse that expresses the human bcl-2 gene from a neuron-specific enolase promoter. In these mice, the normal postnatal loss of retinal ganglion cell axons is greatly decreased and, as a consequence, the number of axons in the optic nerve is increased by approximately 80% compared with wild-type mice. Remarkably, the numbers of oligodendrocytes, astrocytes, and microglial cells are all increased proportionally in the transgenic optic nerve. The increase in oligodendrocytes apparently results from both a decrease in normal oligodendrocyte death and an increase in oligodendrocyte precursor cell proliferation, whereas the increase in astrocytes apparently results from an increase in the proliferation of astrocyte lineage cells. Unexpectedly, the transgene is expressed in oligodendrocytes and astrocytes, but this does not seem to be responsible for the increased numbers of these cells. These findings indicate that developing neurons and glial cells can interact to adjust glial cell numbers appropriately when neuronal numbers are increased. We also show that the expression of the bcl-2 transgene in retinal ganglion cells protects the cell body from programmed cell death when the axon is cut, but it does not protect the isolated axon from Wallerian degeneration, even though the transgene-encoded protein is present in the axon.     

Differential expression of gangliosides and galactolipids in fetal human oligodendrocytes and astrocytes in culture

The phenotypic expression of gangliosides and galactolipids was investigated using primary cultures of fetal human oligodendrocytes and astrocytes. These glial cells were isolated from fetal human brains of 12-18 weeks' gestation. Expression of gangliosides and galactolipids in oligodendrocytes and astrocytes was investigated by double labeling immunocytochemistry using rabbit antibodies specific for galactocerebroside (GalC, a cell type-specific marker for oligodendrocyte) and glial fibrillary acidic protein (GFAP, a cell type-specific marker for astrocyte) in combination with a panel of mouse monoclonal antibodies which react with specific gangliosides or galactolipids. A considerable number of GalC+ oligodendrocytes expressed intense immunoreactivities specific for GM3 (19%) and GM2 (45%) gangliosides. Approximately 11% of GalC+ oligodendrocytes expressed GM4 immunoreactivity, and smaller numbers of GalC+ oligodendrocytes expressed GD3 (4%), GD2 (1%), GT1b (5%) and A2B5 (3%) immunoreactivities. However, GalC+ oligodendrocytes did not express GM1, GD1a, GT1b or GQ1c. Major populations of GalC+ oligodendrocytes immunolabeled by rabbit anti-GalC antibody reacted with anti-GalC mAb (Ranscht mAb, 81%) or by anti-sulfatide mAb (O4 mAb, 91%). A considerable number of GFAP+ astrocytes expressed intense GM2 (26%) and GD2 (15%) immunoreactivities, while a smaller population expressed intense GM3 (3%), GD3 (6%) and GM4 (4%) immunoreactivities. Weak immunoreactions specific for GD1b, A2B5 and sulfatide were found in less than 1% each of GFAP+ astrocytes, while GFAP+ astrocytes did not express GM1, GD1a, GT1a, GT1b or GQ1b. These results indicate that GM3, GM2 and sulfatide are expressed in a major population of GalC+ oligodendrocytes, while GM3, GM2, GD3, GD2, and GM4 are expressed in a small but distinctive population of GFAP+ astrocytes. Our results suggest that GM4, GM1 and GD3, which are utilized as markers for adult human oligodendrocytes and myelin, are not the major ganglioside constituents in cultured fetal human oligodendrocytes.     

Quinolinate immunoreactivity in experimental rat brain tumors is present in macrophages but not in astrocytes

Experimental tumors of the central nervous system were investigated with antibodies to quinolinate to assess the cellular distribution of this endogenous neurotoxin. In advanced F98 and RG-2 glioblastomas and E367 neuroblastomas in the striatum of rats, variable numbers of quinolinate immunoreactive cells were observed in and around the tumors, with the majority being present within tumors, rather than brain parenchyma. The stained cells were morphologically variable, including round, complex, rod-shaped, and sparsely dendritic cells. Neuroblastoma and glioma cells were unstained, as were neurons, astrocytes, oligodendrocytes, ependymal cells, endothelial cells, and cells of the choroid plexus and leptomeninges. Glial fibrillary acidic protein immunoreactivity was strongly elevated in astrocytes surrounding the tumors. Dual labeling immunohistochemistry with antibodies to quinolinate and glial fibrillary acidic protein demonstrated that astrocytes and the cells containing quinolinate immunoreactivity were morphologically disparate and preferentially distributed external and internal to the tumors, respectively, and no dual labeled cells were observed. Lectin histochemistry with Griffonia simplicifolia B4 isolectin and Lycopersicon esculentum lectin demonstrated numerous phagocytic macrophages and reactive microglia in and around the tumors whose distribution was similar to that of quinolinate immunoreactive cells, albeit much more numerous. Dual labeling studies with antibodies to quinolinate and the lectins demonstrated partial codistribution of these markers, with most double-labeled cells having the morphology of phagocytes. The present findings suggest the possibility that quinolinate may serve a functional role in a select population of inflammatory cell infiltrates during the immune response to brain neoplasms.     

Conditionally immortalized neural progenitor cell lines integrate and differentiate after grafting to the adult rat striatum. A combined autoradiographic and electron microscopic study

Neural progenitor cell lines, generated by conditional immortalization from the embryonic CNS, have previously been shown to survive and integrate after transplantation to the adult brain. The present study was designed to investigate the in vivo differentiation and morphological features of grafted neural progenitors using combined autoradiography and transmission electron microscopy of two temperature-sensitive neural progenitor cell lines, HiB5 and ST14A, labeled with 3H-thymidine prior to grafting. Two weeks after transplantation to the striatum the cells were found dispersed over an area extending about 1.5 mm from the injection site. Labeled cells located within the myelinated fiber bundles of the internal capsule were closely associated with myelinated axons and presented profiles similar to oligodendrocytes, while most of the grafted cells in the grey matter had morphological features of astroglia. Some labeled cells occurred also in close association with small blood vessels, morphologically resembling host pericytes. The results show that the immortalized neural progenitors can differentiate into mature glial cells, including astrocytes, oligodendrocytes and pericytes, after implantation into the adult striatum. The ability of the cells to become fully integrated with the resident glial population suggests that they will be highly useful as vehicles for intracerebral transgene expression in ex vivo gene transfer.     

The adult rat hippocampus contains primordial neural stem cells

Adult-derived hippocampal progenitors generate neurons, astrocytes, and oligodendrocytes in vitro and following grafting into the adult brain. Although these progenitors have a considerable capacity for in vitro self renewal, it is not known if each lineage is generated by separate committed precursors or by multipotent stem cells. By genetic marking, we have followed individual cells through the process of proliferative expansion, commitment, and differentiation. All three lineages are generated by single marked cells and the relative proportions of each lineage can be strongly influenced by environmental cues. Differentiation is accompanied by a characteristic progression of lineage-specific markers and can be potentiated by retinoic acid, elevated cyclic AMP, or neurotrophic factors. The ability to genetically mark and clone normal diploid hippocampal progenitors provides the first definitive evidence that multipotent neural stem cells exist outside of the adult striatal subventricular zone and supports the hypothesis that FGF-2-responsive neural stem cells may be broadly distributed in the adult brain.     

Cytokines affecting survival and differentiation of an astrocyte progenitor cell line

The effects of various cytokines on survival and differentiation of an astrocyte progenitor cell line (AP-16) were examined. Epidermal growth factor (EGF) deprivation caused death of AP-16 cells by apoptosis. Transforming growth factor-alpha (TGF-alpha) and basic fibroblast growth factor (bFGF) prevented the apoptosis occurring in the absence of EGF. Leukemia inhibitory factor (LIF) and ciliary neurotrophic factor (CNTF) induced glial fibrillary acidic protein (GFAP) and decreased A2B5 antigen in AP-16 cells, indicating that these cytokines induced AP-16 cells to differentiate into astrocytes.