GFAP-positive astrocytes are rare or absent in primary adult human brain tissue cultures

Traditionally, astrocytes are divided into fibrous and protoplasmic types based on their morphologic appearance. Here the cultures were prepared separately from the adult human cortical gray and white matter of brain biopsies. Both cultures differed only in the number of glial fibrillary acidic protein (GFAP)-positive cells. In the gray matter these were absent or rare, whereas in confluent cultures from the white matter they reached 0.1% of all cells. Three main morphologic types of GFAP-positive cells were found in this study: stellate, bipolar and large flat cells. GFAP-positive cells with two or three long processes mimic a neuron-like morphology. We did not find process-bearing cells expressing neuronal markers (MAP-2, NF, and N-CAM). The conflicting reports concerning GFAP immunostaining and the study dealing with the presence of putative neurons in adult human brain cultures are discussed with respect to these findings. The latter classification of astrocytes into type 1 and type 2 is based on immunostaining to A2B5 antigen: type 1 (GFAP+/A2B5−) and type 2 (GFAP+/A2B5+) astrocytes are proposed to be analogous to protoplasmic and fibrous astrocytes, respectively. In adult human brain cultures we found only small amount of A2B5-positive cells. Double immunofluorescence revealed that astroglial cells of similar fibrous or bipolar shape grown on one coverslip were either GFAP+/A2B5+ or GFAP+/A2B5−. On the other hand, the A2B5+/GFAP− immunophenotype was not observed. These results indicate that in general the cell phenotype from adult human brain tissue is not well established when they are in culture.     

Fibronectin is expressed by astrocytes cultured from embryonic and early postnatal rat brain

In early primary cultures from newborn rat brain, few glial fibrillary acidic protein (GFAP)-positive glial cells expressed intracytoplasmic immunoreactivity for fibronectin. After the second week in culture, however, fibronectin was expressed by a distinct population of GFAP-positive flat astrocytes, irrespective of which brain region was studied. In cerebellar cultures, these cells were more abundant than in cortical or neostriatal cultures and often formed a major population of the GFAP-positive cells. The difference in fibronectin expression between cerebellum and the other areas studied was statistically significant. When cultures were started from 9-day-old postnatal rat brain, fibronectin-positive astrocytes appeared earlier than in those from newborn animals, in all areas studied. Further, especially in the case of cerebellum, the number of fibronectin-positive astrocytes increased as a function of time in culture. In cultures started from whole brains of 12-day-old rat embryos, fibronectin was expressed within 24 h in culture by all the cells with morphology of flat astrocytes, positive for vimentin but negative for GFAP. These results indicate that astrocytes cultured from newborn and early postnatal rat brain are a heterogeneous population of cells: depending on the brain region studied and also depending on the age of brain tissue or the time in culture, less than 1-60% of the GFAP-positive flat astrocytes expressed fibronectin. This, together with the fact that fibronectin was present in early embryonic brain cells in culture, suggests that fibronectin may be a prerequisite for the development or interactions of brain cells.     

Common myofibroblastic features of newborn rat astrocytes and cirrhotic rat liver stellate cells in early cultures and in vivo.

Double-immunolabelling techniques were employed to investigate the distribution of smooth muscle alpha-actin (actin) in glial fibrillary acidic protein (GFAP)-positive cells in rat brain during early postnatal development and maturation and in glial primary culture derived from newborn rat brain. In addition the expression of desmin was studied in the glial primary cultures as a function of the differentiation of the cells. Comparison of the cultured astroglial cells at an early age with hepatic stellate cells derived from CCl4-induced cirrhotic rat liver, revealed features of the astrocytic cytoskeleton characteristic of myofibroblastic cells, i.e., strong expression of both myofibroblastic markers, actin and desmin. In astroglial cells with an initial morphology reminiscent of fibroblasts the non-filamentous perinuclear immunoreaction of GFAP increased with time at the expense of actin and, partially, desmin. GFAP filaments were spread throughout the cytoplasm of the cells which acquired stellate morphology. The alterations in the morphology of the cells and the distribution and intensity of staining for GFAP and actin during the differentiation of astrocytes in culture were similar to those observed in astrocytes during the maturation of the brain. In astrocytes from a newborn brain as well as in cirrhotic hepatic stellate cells, the area of immunoreaction of GFAP was reduced and confined mainly to the nuclear region. In contrast, the cells expressed actin throughout the cytoplasm. These findings may hint at a similar function of these regionally specialized perivascular myofibroblastic cells in a normal brain and diseased liver and at inverse organ-specific functions which the cells fulfill under non-pathological conditions in vivo.     

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. Dev. Genet. 21:187–200, 1997. © 1997 Wiley-Liss, Inc.     

Differential inducibilities of GFAP expression, cytostasis and apoptosis in primary cultures of human astrocytic tumours

Glial fibrillary acidic protein (GFAP) is an astrocytic lineage-specific intermediate filament protein, and its expression or non-expression is inversely correlated with the tumourigenecity of astrocytoma cells. To estimate the GFAP levels of astrocytes in intracranial tumour tissues, we established primary cultures from six astrocytic tumour specimens and used a double-staining flow cytometric method to detect the different levels of GFAP among these primary cultures. Although these primary cultures exhibited the same Matrigel invasiveness, their GFAP expression is inversely related to the rate of cell growth and the histologic grade of the original tumour. Phenylacetate, 12-O-tetradecanoylphorbol-13-acetate (TPA) and sodium butyrate, which are potent inducers of differentiation in various cancer cells, have been examined for their effects on these primary cultures. Cytostasis was more or less caused by these compounds in all six primary cultures, but induction of GFAP was observed only in the primary culture derived from a less malignant astrocytoma specimen having the highest intrinsic GFAP level. Interestingly, this primary culture, but not others, also exhibited increased HRG-α expression after phenylacetate or sodium butyrate treatment. Loss of the inducibility of differentiation-related gene expression could be one of the events involved in the malignant progression of astrocytomas. In addition, the chemotherapeutic agent BiCNU has a killing effect on all six primary culture cells, with LD50 less than 60nM. The underlying mechanism was through the induction of apoptosis in these primary culture cells regardless of their varying malignancies of original tumours. However, unlike colon cancer and leukaemia cells, sodium butyrate could not induce apoptosis within 4 days in these astrocytic tumour cells, indicating that the cell context of different cell types indeed determined the ability of sodium butyrate to induce apoptosis. This revised version was published online in June 2006 with corrections to the Cover Date.     

Longterm quiescent cells in the aged human subventricular neurogenic system specifically express GFAP‐δ

A main neurogenic niche in the adult human brain is the subventricular zone (SVZ). Recent data suggest that the progenitors that are born in the human SVZ migrate via the rostral migratory stream (RMS) towards the olfactory bulb (OB), similar to what has been observed in other mammals. A subpopulation of astrocytes in the SVZ specifically expresses an assembly‐compromised isoform of the intermediate filament protein glial fibrillary acidic protein (GFAP‐δ). To further define the phenotype of these GFAP‐δ expressing cells and to determine whether these cells are present throughout the human subventricular neurogenic system, we analysed SVZ, RMS and OB sections of 14 aged brain donors (ages 74‐93). GFAP‐δ was expressed in the SVZ along the ventricle, in the RMS and in the OB. The GFAP‐δ cells in the SVZ co‐expressed the neural stem cell (NSC) marker nestin and the cell proliferation markers proliferating cell nuclear antigen (PCNA) and Mcm2. Furthermore, BrdU retention was found in GFAP‐δ positive cells in the SVZ. In the RMS, GFAP‐δ was expressed in the glial net surrounding the neuroblasts. In the OB, GFAP‐δ positive cells co‐expressed PCNA. We also showed that GFAP‐δ cells are present in neurosphere cultures that were derived from SVZ precursors, isolated postmortem from four brain donors (ages 63‐91). Taken together, our findings show that GFAP‐δ is expressed in an astrocytic subpopulation in the SVZ, the RMS and the OB. Importantly, we provide the first evidence that GFAP‐δ is specifically expressed in longterm quiescent cells in the human SVZ, which are reminiscent of NSCs.     

EIF2B5 mutations compromise GFAP+ astrocyte generation in vanishing white matter leukodystrophy

Vanishing white matter disease (VWM) is a heritable leukodystrophy linked to mutations in translation initiation factor 2B (eIF2B). Although the clinical course of this disease has been relatively well described, the cellular consequences of EIF2B mutations on neural cells are unknown. Here we have established cell cultures from the brain of an individual with VWM carrying mutations in subunit 5 of eIF2B (encoded by EIF2B5). Despite the extensive demyelination apparent in this VWM patient, normal-appearing oligodendrocytes were readily generated in vitro. In contrast, few GFAP-expressing (GFAP+) astrocytes were present in primary cultures, induction of astrocytes was severely compromised, and the few astrocytes generated showed abnormal morphologies and antigenic phenotypes. Lesions in vivo also lacked GFAP+ astrocytes. RNAi targeting of EIF2B5 severely compromised the induction of GFAP+ cells from normal human glial progenitors. This raises the possibility that a deficiency in astrocyte function may contribute to the loss of white matter in VWM leukodystrophy.     

Cellular Location of Glutamine Synthetase and Lactate Dehydrogenase in Oligodendrocyte-Enriched Cultures from Rat Brain

Glial cells were isolated from 1-week-old rat brain and cultured in a serum-free medium supplemented with the hormones insulin, hydrocortisone, and triiodothyronine. After 1 week in culture the cell population consisted mainly of galactocerebroside-positive cells (GC+; oligodendrocytes), the remainder of the cells being positive for glial fibrillary acidic protein (GFAP+; astrocytes). Oligodendrocytes were selectively removed from the cultures by complement-mediated cytolysis. The activities of glutamine synthetase and of various marker enzymes were measured in the nonlysed cells remaining after complement treatment of the cultures and in the culture medium containing proteins of the lysed cells. We found that the cellular activity of glutamine synthetase decreased in parallel with the lysis of GC+ cells and that the activity of glutamine synthetase in the supernatant increased. The activity of glycerol-3-phosphate dehydrogenase, a marker enzyme for oligodendrocytes, was no longer detectable in complement-treated cultures and the activity of glutamine synthetase was markedly lowered, whereas the activity of lactate dehydrogenase was as high as in untreated cultures. The location of glutamine synthetase both in oligodendrocytes and in astrocytes was confirmed by double-label immunocytochemistry with antisera against glutamine synthetase, GC, and GFAP. We conclude that in this culture system glutamine synthetase is expressed in both types of glial cells and that the activity of lactate dehydrogenase is at least one order of magnitude higher in astrocytes than in oligodendrocytes.     

Subpopulation of nestin positive glial precursor cells occur in primary adult human brain cultures

Glial fibrillary acidic protein (GFAP) is an intermediate filament protein considered to be the best astroglial marker. However, the predominant cell population in adult human brain tissue cultures does not express GFAP; these cells have been termed “glia-like” cells. The basic question about histological origin of adult human brain cultures remains unanswered. Some authors showed that “glia-like” cells in adult human brain cultures might be of non-glial origin. We examined primary explant tissue cultures derived from 70 adult human brain biopsies. Within first 5–10 days approximately 5–10% of the small explants became attached. Outgrowing cells were mostly flat cells. These cells formed confluent layer over 3–6 weeks in culture. At confluence the cultures contained 2–5% of microglial cells, 0.1% GFAP-positive astrocytes, less than 0.01% oligodendrocytes and 95–98% GFAP-negative “glia-like” cells. This population of flat “glia-like” cells was positively stained for vimentin, fibronectin, and 20–30% of these cells stained for nestin. Our findings revealed that 1 mM dibutyryl-cAMP addition, in serum free conditions, induced a reversible stellation in 5-10% of the flat “glia-like” cells but did not induce the expression of GFAP or nestin in morphologically changed stellate cells. These results demonstrate that “glia-like” cells in primary adult human brain cultures constitute heterogeneous cell populations albeit with similar morphological features. Two distinct subpopulations have been shown: (i) the one immunostained for nestin; and (ii) the other reactive for dibutyryl-cAMP treatment.     

The relationship of glial fibrillary acidic protein to the shape,motility, and differentiation of human astrocytoma cells

Glial fibrillary acidic protein (GFAP), a protein largely limited to astrocytes, was studied in relation to the shape, motility, and differentiation and malignancy of astrocytoma cells in tissue culture by use of time-lapse photography and the immunoperoxidase method.A relationship was observed between the shape of astrocytes and the distribution of GFAP. Spindle-shaped cells showed abundant GFAP in the cell body and processes. In round or polyhedral cells without well developed processes the GFAP was largely perinuclear. As processes developed, GFAP extended out from the nucleus iri dense parallel arrays that radiated into the developing processes. Fully differentiated cells with stellate shape had abundant GFAP throughout.A relationship was also observed between the motility of astrocytes and GFAP. Stellate-shaped cells, showing paucity of locomotion and relatively rigid postures of processes, contained an abundance of GFAP which tended to form dense parallel arrays extending into the processes during their development. Spindle-shaped cells with extending and retracting processes and active migration also contained an abundance of GFAP but not organized into parallel arrays. Bulbous dilatations at the tips of processes (growth cones) contained abundant GFAP. There was also abundant GFAP in the intermittent dilatations along the processes of stellate cells. In contrast to these observations, a retraction of processes, a high degree of plasticity (undulating motion) and multidirectional locomotion were often associated with a paucity of GFAP in less differentiated cells. We hypothesize that GFAP filaments may be inhibitory to great plasticity of motion but not to extension-retraction movements.During mitosis GFAP was sparse at the spindle and in intercellular bridges. Colcemid caused GFAP to disappear from processes and peripheral parts of the cell and to become concentrated near the nucleus.In cultures derived from malignant tumors, undifferentiated and large multinucleated cells usually showed sparsity of GFAP, but occasional well differentiated stellate or spindle-shaped cells containing abundant GFAP were seen. Conversely, although cultures derived from benign tumors may have scattered less well differentiated cells, the differentiated cells with well developed processes were most densely stained and account for the high concentration of GFAP in tissue from these tumors.     

Human astrocytes can be induced to differentiate into cells with neuronal phenotype

Several recent studies have proposed that astrocytes may contribute to neurogenesis, not only as a source of trophic substances regulating it, but also as stem cells themselves. In order to better understand these mechanisms, primary astrocyte cultures were established from human fetal brain. After 3-4 weeks in culture, astrocytes (about 95% GFAP+; neurofilament, NF-; neuro-specific enolase, NSE-) were treated with a cocktail of protein kinase activators and FGF-1. After 5 h of treatment, most cells showed morphological changes that increased progressively up to 24-48 h, exhibiting a round cell body with long processes. Immunocytochemistry showed that treatment-induced NF and NSE expression in about 40% of cells. Nestin expression increased after treatment, whereas GFAP immunostaining was not significantly modified. Western blot and RT-PCR confirmed the results. No neuronal electrophysiological properties were observed after treatment, suggesting an incomplete maturation under these experimental conditions. Understanding the regenerative capability and neurogenic potential of astrocytes might be useful in devising therapeutic approaches for a variety of neurological disorders.     

Role of the Cellular Stress Response in the Biogenesis of Cysteamine-Induced Astrocytic Inclusions in Primary Culture

Abstract— Cysteamine (CSH; 2-mercaptoethylamine) stimulates the accumulation of peroxidase-positive inclusions in cultured astroglia akin to those observed in the aging periventricular brain. Because CSH induces the synthesis of a stress protein (heme oxygenase) in rat liver, we hypothesized that aspects of the cellular stress response may play a role in the biogenesis of CSH-induced astro-cyte granules. In the present study, we performed indirect immunofluorescent staining and immunoblotting for various stress proteins in rat neuroglial cultures. Exposure of astrocyte cultures to CSH enhanced immunostaining for heme oxygenase-1 (HO-1) and heat-shock proteins 27, 72, and 90, but not glucose-regulated protein 94, relative to untreated cultures. CSH-pretreated astrocytes exhibited enhanced tolerance to H₂O₂ toxicity relative to untreated cells, providing physiological evidence of an antecedent stress response in the former. In addition, exposure for 12 days to H₂O₂, a known inducer of the stress response, elicited astrocyte granulation similar to that observed with CSH. Chronic induction of HO-1 and other stress proteins may participate in the biogenesis of metal-loporphyrin-rich inclusions in CSH-treated astroglial cultures and in astrocytes of the aging periventricular brain.     

Toxicity of organotin compounds in primary cultures of rat cortical astrocytes

The neurotoxic organotin compounds trimethyl (TMT) and triethyltin (TET) are known to induce astrogliosis in vivo, which is indicated by an increased synthesis of glial fibrillary acidic protein (GFAP) in astrocytes. In contrast, tributyltin (TBT) does not induce astrogliosis. The aim of this study was to investigate whether trialkyltin derivatives can induce an increased GFAP synthesis in astrocyte cultures in the absence of neurons and whether differences between the action of TMT, TET, and TBT can be detected. Primary cultures of rat cortical astrocytes from 2-day-old rats were grown in 96-well plates until confluency and then exposed to various concentrations of TMT, TET, and TBT for 40 h. Effects on basal cell functions were measured by colorimetric determination of cell protein contents and by assessment of viability by means of the MTT assay. An indirect sandwich ELISA for 96-well plates was used for quantitative measurements of the GFAP content of the cells. All three compounds induced a concentration-dependent cytotoxicity indicated by parallel decreases of protein contents and MTT reduction. Half-maximum cytotoxic concentrations were 3 μmol/L (TBT), 30 μmol/L (TET), and 800 μmol/L (TMT). Cellular GFAP contents were reduced in parallel to cytotoxic action but no increase in GFAP expression at subcytotoxic concentrations could be observed. Thus, the astrocytes were not able to respond to TMT or TET exposure by an increased synthesis of GFAP in the absence of neuronal signals. This revised version was published online in July 2006 with corrections to the Cover Date.     

LIF and BMP signaling generate separate and discrete types of GFAP-expressing cells

Bone morphogenetic protein (BMP) and leukemia inhibitory factor (LIF) signaling both promote the differentiation of neural stem/progenitor cells into glial fibrillary acidic protein (GFAP) immunoreactive cells. This study compares the cellular and molecular characteristics, and the potentiality, of GFAP(+) cells generated by these different signaling pathways. Treatment of cultured embryonic subventricular zone (SVZ) progenitor cells with LIF generates GFAP(+) cells that have a bipolar/tripolar morphology, remain in cell cycle, contain progenitor cell markers and demonstrate self-renewal with enhanced neurogenesis - characteristics that are typical of adult SVZ and subgranular zone (SGZ) stem cells/astrocytes. By contrast, BMP-induced GFAP(+) cells are stellate, exit the cell cycle, and lack progenitor traits and self-renewal--characteristics that are typical of astrocytes in the non-neurogenic adult cortex. In vivo, transgenic overexpression of BMP4 increases the number of GFAP(+) astrocytes but depletes the GFAP(+) progenitor cell pool, whereas transgenic inhibition of BMP signaling increases the size of the GFAP(+) progenitor cell pool but reduces the overall numbers of astrocytes. We conclude that LIF and BMP signaling generate different astrocytic cell types, and propose that these cells are, respectively, adult progenitor cells and mature astrocytes.     

Iron content correlates with peroxidase activity in cysteamine-induced astroglial organelles.

A subpopulation of astrocytes in periventricular brain regions and in cysteamine-treated neuroglial cultures contains cytoplasmic granules that exhibit an affinity for Gomori stains, orange-red autofluorescence, and non-enzymatic peroxidase activity. The autofluorescence and pseudoperoxidase activity are consistent with the presence of porphyrins and heme iron, respectively. In the present study, we employed diaminobenzidine cytochemistry, transmission electron microscopy, and energy-dispersive X-ray microanalysis (electron microprobe) in an attempt to correlate fine structure with the peroxidase activity and elemental composition of the cysteamine-induced inclusions in cultured astrocytes. In osmicated preparations, these membrane-bound inclusions varied greatly in size, were round or ovoid in shape, and exhibited an intensely electron-dense granular matrix. In non-osmicated preparations, many inclusions exhibited internal membranous partitions producing complex subcompartmentalization. Diaminobenzidine reaction product, indicative of endogenous peroxidase activity, was occasionally observed distributed diffusely throughout the granule matrix. More commonly, peroxidase activity was restricted to specific intraorganellar compartments. Elemental iron was detected in the inclusions by electron microprobe analysis. The presence and concentration of iron in these organelles correlated closely with the presence and intensity of diaminobenzidine staining, suggesting that redox-active iron mediates the pseudoperoxidase reactions in these cells. Cysteamine-induced derangements of porphyrin-heme biosynthesis may be responsible for the proliferation of iron-containing gliosomes in these astrocytes.     

GFAP-Deficient Astrocytes Are Capable of Stellationin VitroWhen Cocultured with Neurons and Exhibit a Reduced Amount of Intermediate Filaments and an Increased Cell Saturation Density

Glial fibrillary acidic protein (GFAP) is an intermediate filament protein predominantly expressed in cells of astroglial origin. To allow for the study of the biological functions of GFAP we have previously generated GFAP-negative mice by gene targeting [Peknyet al.(1995)EMBO J.14, 1590–1598]. Astrocytes in culture, similar to reactive astrocytesin vivo,express three intermediate filament proteins: GFAP, vimentin, and nestin. Using primary astrocyte-enriched cultures from GFAP-negative mice, we now report on the effect of GFAP absence on (i) the synthesis of other intermediate filament proteins in astrocytes, (ii) intermediate filament formation, (iii) astrocyte process formation (stellation) in response to neurons in mixed cerebellar astrocyte/neuron cultures, and (iv) saturation cell densityin vitro.GFAP−/− astrocytes were found to produce both nestin and vimentin. At the ultrastructural level, the amount of intermediate filaments as revealed by transmission electron microscopy was reduced in GFAP−/− astrocytes compared to that in GFAP+/+ astrocytes. GFAP−/− astrocytes retained the ability to form processes in response to neurons in mixed astrocyte/neuron cultures from the cerebellum. GFAP−/− astrocyte-enriched primary cultures exhibited an increased final cell saturation density. The latter leads us to speculate that the loss of GFAP expression observed focally in a proportion of human malignant gliomas may reflect tumor progression toward a more rapidly growing and malignant phenotype.     

PD大鼠胼胝体和扣带胶质原纤维酸性蛋白的表达

目的观察不同病程帕金森病(PD)大鼠胼胝体和扣带胶质原纤维酸性蛋白(GFAP)的表达。方法大鼠右侧前脑内侧束注射六羟基多巴胺制备PD模型。大鼠分为正常对照组,2周、4周和6周模型组。以酪氨酸羟化酶(TH)和GFAP抗体免疫组化阳性分别显示黑质的多巴胺能神经元、胼胝体与扣带处的星形胶质细胞。结果 2、4、6周模型组右侧黑质TH阳性细胞数较正常对照组均显著降低,而2、4、6周模型组之间无显著差异。正常对照组和4周模型组胼胝体和扣带处星形胶质细胞呈未活化状态,两组之间的GFAP表达强度和细胞密度均无显著差异。2周和6周模型组两部位星形胶质细胞呈活化状态,其表达强度和细胞密度均显著高于正常对照组和4周模型组。结论急性完全损伤PD模型大鼠注射侧胼胝体和扣带处GFAP表达随病程呈增高、降低和增高趋势。