The demonstration of phospholipids in certain cells in neurosecretory nuclei in the rat with Baker's acid haematein after fixation in glutaraldehyde-formol-calcium

Summary With Baker's acid haematein test certain ganglion cells in the brain, their processes and, at some sites, glial cells around blood vessels stain dark blue. This article describes a study of the Baker-positive cells which occur in and around the neurosecretory nuclei. By substituting formol-calcium fixation with glutaraldehyde-formol-calcium fixation shrinkage in brain tissue is completely avoided. If such fixation is used the argument that positive staining of ganglion cells with Baker's method only indicates that these are shrunken neurons can no longer be maintained. A comparative histological study, especially of Baker's technique and controlled chromation (Elftman) showed that the Baker-positive cells contain a phospholipid, probably bound to a protein, as a labile compound, which is easily lost. We found that to immobilize and localize this labile compound in the ganglion cells the technique of fixation and the pH during chromation (which should be around 3.8) are of fundamental importance. Only under these conditions is the complex sufficiently immobilized to allow of its demonstration with acid haematein. These requirements are now completely met if Baker's acid haematein technique is used. The article stresses that only prefixed and chromated frozen sections can be used for this method, thus avoiding shrinkage and non-specific staining of proteins. The modified Baker method as used by us gives constant and reproducible staining and is described in this article. The functional significance of the Baker-positive reaction in some ganglion cells in the n. s. nuclei or glial cells around blood vessels is not dealt with in this article.     

Regulation of specific functions of glial cells in somatic hybrids. I. Control of S100 protein

Hybrids were isolated between rat glial cells and mouse fibroblasts. Micro complement (C′) fixation was used to assay S100, a highly acidic protein specific for nervous tissues. The glial cells contain large amounts of S100. Extracts of the fibroblasts contain some C′ fixing material which is detected only at very high protein concentrations and which fixes C′ only weakly. The identity of this material is not known. The hybrids contain some C′ fixing material, but the concentrations of protein necessary to reach the point of antigen-antibody equivalence is ten times greater with an extract of hybrids than with an extract of glial cells. This increase in the concentration of protein is associated with a decrease in the amount of C′ fixed. The possible significance of the C′ fixing material in the hybrids is discussed.     

Usefulness of statistic experimental designs in enzymology: example with recombinant hCYP3A4 and 1A2

Estimation of the concentration of a specific protein in a biological sample often is obtained by analysis of immunoblots. We used this technique to estimate the concentration of three proteins present in homogenates of brain: glial fibrillary acidic protein (GFAP), myelin basic protein (MBP), and synapsin I. Homogenates prepared from rat brains known to contain more than 6-fold increases in GFAP, based on a GFAP enzyme-linked immunosorbent assay (ELISA), showed only small relative increases in this protein when the same samples were subjected to immunoblot analysis with polyclonal or monoclonal anti-GFAP; quantification was based on PhosphorImager analysis of [(125)I] protein A bound to the antibodies. Estimates of GFAP in the GFAP-enriched samples approached the expected 6-fold increase when the total protein load per gel lane was reduced from 30 to 1 microgram. Pure GFAP run as standard was not affected by 10-fold increases in protein load, but spiking brain homogenates with pure GFAP "quenched" the values obtained for the standard run alone. Examination of the quenching potential of pure brain tubulin, a protein that nearly comigrates with GFAP on SDS gels, showed that it may be one component of brain homogenates that contributes to masking of immunodetection of GFAP. The effect of total brain homogenate proteins on the signal obtained for a specific protein was not limited to GFAP; similar effects were observed for MBP and synapsin I. The data indicate that estimates of the concentration of a specific protein, whether as a function of its relative amount in a given protein mixture or its relative amount in one mixture compared to another, are influenced by other homogenate proteins present in the mixture.     

Evaluation of the Relative Ureide Content of Xylem Sap as an Indicator of N(2) Fixation in Soybeans: GREENHOUSE STUDIES

The use of the relative ureide content of xylem sap [(ureide-N/total N) × 100] as an indicator of N₂ fixation in soybeans (Merr.) was examined under greenhouse conditions. Acetylene treatments to inhibit N₂ fixation were imposed upon the root systems of plants totally dependent upon N₂ fixation as their source of N and of plants dependent upon both N₂ fixation and uptake of exogenous nitrate. Significant decreases in the total N concentration of xylem sap from plants of the former type were observed, but no significant decrease was observed in the total N concentration of sap from the latter type of plants. In both types of plants, acetylene treatment caused significant decreases in the relative ureide content of xylem sap. The results provided further support for a link between the presence of ureides in the xylem and the occurrence of N₂ fixation in soybeans. The relative ureide content of xylem sap from plants totally dependent upon N₂ fixation was shown to be insensitive to changes in the exudation rate and total N concentration of xylem sap brought about by diurnal changes in environmental factors. There was little evidence of soybean cultivars or nodulating strains affecting the relative ureide content of xylem sap. `Ransom' soybeans nodulated with Rhizobium japonicum strain USDA 110 were grown under conditions to obtain plants exhibiting a wide range of dependency upon N₂ fixation. The relative ureide content of xylem sap was shown to indicate reliably the N₂ fixation of these plants during vegetative growth using a ¹⁵N method to measure N₂ fixation activity. The use of the relative ureide content of xylem sap for quantification of N₂ fixation in soybeans should be evaluated further.     

DETECTION AND DEMONSTRATION OF INHIBITORY ACTIVITIES OF BACTERIOCINS BY ISOELECTRIC FOCUSING

A technique utilizing isoelectric focusing (IEF) was developed and compared with a modified Bhunia's sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) method for directly detecting antimicrobial activities of inhibitory peptides or proteins (bacteriocins). In IEF, the gel containing separated peptide or protein bands was directly overlaid with indicator bacteria without any fixation or rinsing as done in the SDS-PAGE method. The IEF gave clear zones of inhibition surrounding inhibitory bands. Bhunia's method was modified to reduce the time for fixation and rinsing, and a 15-min fixation followed by 2-h rinsing with dd H₂O was determined to be necessary for a detection similar to the IEF. In addition, results of this study showed that denaturation of bacteriocins and, subsequently, failure to detect the presence of a bacteriocin could occur in the SDS-PAGE analysis while little denaturation was observed in the IEF assay. Therefore, the IEF assay developed in this study was more rapid and less destructive as compared with the SDS-PAGE method.     

Cortical radial glial cells in human fetuses: Depth‐correlated transformation into astrocytes

In the human brain, the transformation of radial glial cells (RGC) into astrocytes has been studied only rarely. In this work, we were interested in studying the morphologic aspects underlying this transformation during the fetal/perinatal period, particularly emphasizing the region‐specific glial fiber anatomy in the medial cortex. We have used carbocyanine dyes (DiI/DiA) to identify the RGC transitional forms and glial fiber morphology. Immunocytochemical markers such as vimentin and glial fibrillary acidic protein (GFAP) were also employed to label the radial cells of glial lineage and to reveal the early pattern of astrocyte distribution. Neuronal markers such as neuronal‐specific nuclear protein (NeuN) and microtubule‐associated protein (MAP‐2) were employed to discern whether or not these radial cells could, in fact, be neurons or neuronal precursors. The main findings concern the beginning of RGC transformation showing loss of the ventricular fixation in most cases, followed by transitional figures and the appearance of mature astrocytes. In addition, diverse fiber morphology related to depth within the cortical mantle was clearly demonstrated. We concluded that during the fetal/perinatal period the cerebral cortex is undergoing the final stages of radial neuronal migration, followed by involution of RGC ventricular processes and transformation into astrocytes. None of the transitional or other radial glia were positive for neuronal markers. Furthermore, the differential morphology of RGC fibers according to depth suggests that factors may act locally in the subplate and could have a role in the process of cortical RGC transformation and astrocyte localization. The early pattern of astrocyte distribution is bilaminar, sparing the cortical plate. Few astrocytes (GFAP+) in the upper band could be found with radial processes at anytime. This suggests that astrocytes in the marginal zone could be derived from different precursors than those that differentiate from RGCs during this period. © 2003 Wiley Periodicals, Inc. J Neurobiol 55: 288–298, 2003     

Highly Sensitive Immunoassay for Rat Brain-Type Creatine Kinase: Determination in Isolated Purkinje Cells

Ultrasensitive enzyme immunoassay method for the measurement of rat brain-type creatine kinase BB (CK-BB) was developed by use of purified antibodies specific to the B subunit of creatine kinase. The antibody immunoglobulin G was purified with immunoaffinity chromatography of the antiserum raised in rabbits by injecting the purified rat CK-BB. The assay system consisted of polystyrene balls with immobilized antibody F(ab')2 fragments and the same antibody Fab' fragments labeled with beta-D-galactosidase from Escherichia coli. The assay was specific to the B subunit of CK (CK-B), showing about 10% cross-reactivity with CK-MB, but it did not cross-react with CK-MM and neuron-specific gamma gamma enolase. The minimum detection limit of the assay was 0.1 pg or 1 amol CK-BB, being sufficiently sensitive for the measurement of CK-B contents in the isolated Purkinje cell bodies at the level of single cells. The average content of CK-B in a single Purkinje cell was 1.64 pg. The CK-B concentration in rat cerebellum (about 22 micrograms/mg protein) was about twofold higher than that (about 13 micrograms/mg protein) in the cerebrum. High levels (greater than 5 micrograms/mg protein) of CK-B were also found in the peripheral tissues such as gastrointestinal tract and urinary bladder, all of which are composed of smooth muscle. Immunohistochemical localization of CK-B antigens in the CNS revealed that the antigens is distributed not only in the neurons but also in the glial cells.     

Modification of the bicinchoninic acid protein assay to eliminate lipid interference in determining lipoprotein protein content.

The bicinchoninic acid (BCA) assay method for the determination of protein has been investigated for its utility in measuring the protein content of plasma lipoproteins. Although other methods, principally those based on the method of Lowry et al. (1951, J. Biol. Chem. 193, 265-275) have been extensively used for this purpose, the tolerance of the BCA method to many commonly encountered detergents and buffers offers a definite advantage over the Lowry-based methods. In this study, lipoprotein protein values obtained by the BCA method were compared to a standard modification of the Lowry et al. procedure since this assay forms the basis of much of the relevant literature. The standard BCA assay was found to overestimate the protein content of very low density lipoprotein by approximately 70% and low density lipoprotein by approximately 30%; high density lipoprotein values compared favorably. Overestimations by the BCA assay paralleled the relative phospholipid content of the lipoprotein fractions. This apparent lipid effect was eliminated by the addition of 2% sodium dodecyl sulfate to samples prior to the analysis. In the presence of this detergent, BCA assay measurements for these three lipoprotein fractions were 97, 90, and 98%, respectively, of the reference assay values.     

Use of a sensitive receptor binding assay to discriminate between full-length and truncated human recombinant tumor necrosis factor proteins

A radioreceptor assay (RRA) capable of detecting picomolar concentrations of human recombinant tumor necrosis factor (TNF) was used to compare the relative binding affinities of genetically engineered full-length and truncated TNF proteins. The specific cell-surface receptors for TNF present on the human cervical carcinoma cell line ME-180 were characterized as having a Kd of 0.2 nM and a density of 2700 sites/cell. Conditions were then defined for an RRA that maximized the specific binding of 125I-TNF to this adherent cell line. Incubation of ME-180 cells with 125I-TNF at 37 degrees C in the presence of 0.02% sodium azide resulted in a 4-fold increase in assay sensitivity and a doubling of specific counts bound, as compared to binding done at 4 degrees C with or without sodium azide. Inhibition of receptor-ligand internalization under these conditions was a likely reason for the increases. This system was utilized to compare low concentrations of the full-length TNF protein and a genetically altered TNF protein (mutein) which lacks the 10 N-terminal amino acids and contains an N-terminal methionine. Previous studies showing the truncated TNF to be 2- to 3-fold lower in cytotoxic activity on a variety of tumor cell lines were corroborated by our findings that the mutein was also three and one-half times lower in relative affinity for the TNF receptor on ME-180 cells. These results suggest a possible role for these residues in receptor binding and illustrate the use of a highly sensitive RRA for the evaluation of TNF molecules altered by recombinant DNA technology.     

The use of a fluorescent dye, Nile red, to evaluate the lipid content of single mammalian oocytes

This study aimed to investigate the use of Nile red, a fluorescent dye specific for intracellular lipid droplets, to quantify the lipid content of single mammalian oocytes. It was hypothesized that a higher amount of lipid present in lipid droplets in an oocyte would result in a higher amount of emitted fluorescent light. Following fixation and subsequent staining of denuded oocytes, the fluorescence of the whole oocyte was visualized by fluorescence microscopy and quantified with a photometer and photomultiplier connected to the microscope. The peak of fluorescence was observed in the yellow spectrum (590 nm) and the fluorescence was restricted to the lipid droplets corresponding to apolar lipids. Nile red concentrations ranging from 0.1 to 10 microg/ml yielded similar results. After fixation, a minimum of 2 h staining was necessary to reach maximal fluorescence which remained stable for several hours. The position of the microscopic focus within the oocyte had no influence on the amount of measured fluorescence. Successive measurements of the same oocyte yielded very similar results indicating the repeatability of the method. Finally, the technique was validated by comparing the lipid content of bovine, porcine and murine immature oocytes, which are known to contain different amounts of lipids. After staining, the fluorescence of murine oocytes was 2.8-fold lower than the fluorescence of bovine oocytes which in turn were 2.4 times less fluorescent than porcine oocytes. Based on this study, it can be said that this rather fast and easy technique allows for the relative quantification of the lipid content (present in the lipid droplets) of one single oocyte. The different amounts of emitted fluorescent light in bovine, porcine and murine oocytes correlated with the known lipid contents in these three species. This technique could be used to compare the lipid content of oocytes originating from different donors, from different sized follicles or cultured in various conditions.     

Effect of Adenosine Monophosphate, Adenosine Diphosphate, and Reduced Nicotinamide Adenine Dinucleotide on Adenosine Triphosphate-dependent Carbon Dioxide Fixation in the Autotroph Thiobacillus neapolitanus

The observation that adenosine triphosphate (ATP)-dependent CO(2) fixation in extracts of chemosynthetic and photosynthetic autotrophs may be regulated in part by adenosine monophosphate (AMP) was extended to the strict autotroph Thiobacillus neapolitanus (X). In addition, this report presents data which include adenosine diphosphate (ADP) in the regulatory role. When the primary CO(2) acceptor, ribose-5-phosphate, was replaced by ribulose-1,5-diphosphate, no inhibition of CO(2) fixation occurred unless the Mg(++) concentration was limiting. A molar ratio of 5:1 AMP or ADP to ATP reduced the specific activity (micromoles of CO(2) fixed per milligram of protein per minute) of the extracts from 0.22 to 0.12 and 0.11, respectively. The reported stimulation of the carboxylative phase of ATP-dependent CO(2) fixation by reduced nicotinamide adenine dinucleotide (NADH(2)) was investigated. Adding NADH(2) to the extracts did not stimulate CO(2) fixation, even at carbonate levels from 0.05 to 30 mumoles, except in the absence of ribose-5-phosphate. Slight increases in CO(2) fixation were noted when the assay system was incubated in air instead of the usual helium atmosphere.     

Spectrophotometric quantitation of anchorage-dependent cell numbers using the bicinchoninic acid protein assay reagent

A rapid and convenient method is described for the determination of the actual and relative number of adherent cells in tissue culture. The cell lines human melanoma C32, ATCC CRL 1585, mouse melanoma B16-F10, and pig epithelial LLC-PK1, suspended in Dulbecco's minimum essential medium containing no serum, were allowed to adhere to fibronectin adsorbed to wells of a 96-well microtiter plate. Nonadherent cells were removed by aspiration, wells were washed, and adherent cells were solubilized with 200 microliters of the bicinchoninic acid (4,4'-dicarboxy-2,2'-biquinoline) protein assay reagent. Plates were heated to 60 degrees C for 30 min and absorbances read at 562 nm using a microtiter plate reader. A linear correlation was observed between the number of adherent cells in the range 2-8 X 10(5)/ml cells added and the protein content of the adherent cells as measured by the BCA protein reagent. The assay procedure gave absorbance values in the range of 0.100 to 1.30 making the method highly sensitive and reproducible. Blank wells containing only coupled protein and no cells gave little or no absorbance. Cell adhesion was fibronectin specific since little or no cell attachment was observed when microtiter plates were coupled with bovine serum albumin. Similar results were obtained with other cell types such as platelets. These results indicate that measurement of total cellular protein using the BCA protein reagent can be a rapid and sensitive assay for the detection and quantitation of adherent cells.     

Nodule infection by bean yellow mosaic virus in Phaseolus vulgaris.

Infection of root nodules of beans, Phaseolus vulgaris L., by bean yellow mosaic virus (BYMV) and the effect of the disease on the specific activity of the nodule are reported. Infectivity and serological microprecipitin assays with two sources of BYMV antiserum demonstrated that nodules from bean plants whose leaves had been inoculated with BYMV contain BYMV antigen. The disease reduced the fresh weights of tops, roots, and root nodules and induced premature nodule decay and/or nodule drop. The disease also reduced leghemoglobin content, on a plant weight basis, and N2 fixation rate, on an individual plant basis, as measured by the acetylene reduction assay. The increased leghemoglobin content per gram-nodule in BYMV-infected nodules relative to healthy nodules might be associated with multiplication of the virus in the nodule and/or unknown cellular effects derived from the BYMV-Rhizobium interaction.     

Expression of manganese superoxide dismutase is not altered in transgenic mice with elevated level of copper-zinc superoxide dismutase

In evaluating the relative expression of copper-zinc and manganese superoxide dismutase (CuZnSOD and MnSOD) in vivo in states like Down syndrome in which one dismutase is present at increased levels, we measured activities of both enzymes, in tissues of control and transgenic mice constitutively expressing increased levels of CuZnSOD, during exposure to normal and elevated oxygen tensions. Using SOD gel electrophoresis assay, CuZnSOD and MnSOD activities of brain, lung, heart, kidney, and liver from mice exposed to either normal (21%) or elevated (>99% oxygen, 630 torr) oxygen tensions for 120 h were compared. Whereas CuZnSOD activity was elevated in tissues of transgenic relative to control mice under both normoxic or hyperoxic conditions, MnSOD activities in organs of transgenic mice were remarkably similar to those of controls under both conditions. To confirm the accuracy of this method in quantitating MnSOD relative to CuZnSOD expression, two other methods were utilized. In lung, which is the organ exposed to the highest oxygen tension during ambient hyperoxia, a sensitive, specific ELISA for MnSOD was used. Again, MnSOD protein was not different in transgenic relative to control mice during exposure to air or hyperoxia. In addition, lung MnSOD protein was not changed significantly by exposure to hyperoxia in either group. In kidney, a mitochondrion-rich organ, SOD assay, before and after inactivation of CuZnSOD with diethyldithiocarbamate, was used. MnSOD activity was not different in organs from air-exposed transgenic relative to control mice. The data indicated that expression of MnSOD in vivo was not affected by overexpression of the CuZnSOD and, therefore, the two enzymes are probably regulated independently.     

Relative quantification of serum proteins from pancreatic ductal adenocarcinoma patients by stable isotope dilution liquid chromatography-mass spectrometry

We report an innovative multiplexed liquidchromatography-multiple reaction monitoring/mass spectrometry (LC-MRM/MS)-based assay for rapidly measuring a large number of disease specific protein biomarkers in human serum. Furthermore, this approach uses stable isotope dilution methodology to reliably quantify candidate protein biomarkers. Human serum was diluted using a stable isotope labeled proteome (SILAP) standard prepared from the secretome of pancreatic cell lines, subjected to immunoaffinity removal of the most highly abundant proteins, trypsin digested, and analyzed by LC-MRM/MS. The method was found to be precise, linear, and specific for the relative quantification of 72 proteins when analyte response was normalized to the relevant internal standard (IS) from the SILAP. The method made it possible to determine statistically different concentrations for three proteins (cystatin M, IGF binding protein 7, and villin 2) in control and pancreatic cancer patient samples. This method proves the feasibility of using a SILAP standard in combination with stable isotope dilution LC-MRM/MS analysis of tryptic peptides to compare changes in the concentration of candidate protein biomarkers in human serum.     

Expression and localization of mitochondrial ferritin mRNA in Alzheimer's disease cerebral cortex

Mitochondrial ferritin (MtF) has been identified as a novel ferritin encoded by an intron-lacking gene with specific mitochondrial localization located on chromosome 5q23.1. MtF has been associated with neurodegenerative disorders such as Friedreich ataxia and restless leg syndrome. However, little information is available about MtF in Alzheimer's disease (AD). In this study, therefore, we investigated the expression and localization of MtF messenger RNA (mRNA) in the cerebral cortex of AD and control cases using real-time polymerase chain reaction (PCR) as well as in situ hybridization histochemistry. We also examined protein expression using western-blot assay. In addition, we used in vitro methods to further explore the effect of oxidative stress and β-amyloid peptide (Aβ) on MtF expression. To do this we examined MtF mRNA and protein expression changes in the human neuroblastoma cell line, IMR-32, after treatment with Aβ, H2O2, or both. The neuroprotective effect of MtF on oxidative stress induced by H(2)O(2) was measured by MTT assay. The in situ hybridization studies revealed that MtF mRNA was detected mainly in neurons to a lesser degree in glial cells in the cerebral cortex. The staining intensity and the number of positive cells were increased in the cerebral cortex of AD patients. Real-time PCR and western-blot confirmed that MtF expression levels in the cerebral cortex were significantly higher in AD cases than that in control cases at both the mRNA and the protein level. Cell culture experiments demonstrated that the expression of both MtF mRNA and protein were increased by treatment with H2O2 or a combination of Aβ and H2O2, but not with Aβ alone. Finally, MtF expression showed a significant neuroprotective effect against H2O2-induced oxidative stress (p<0.05). The present study suggests that MtF is involved in the pathology of AD and may play a neuroprotective role against oxidative stress.     

A Non-radioisotopic Anion-exchange Chromatographic Method to Measure the CO2/O2 Specificity Factor for Ribulose Bisphosphate Carboxylase/Oxygenase

A non-radioisotopic anion-exchange ion chromatographic method for measuring the carboxylation/ oxygenation specificity (τ) of ribulose 1, 5-bisphosphate carboxylase/oxygenase (RubisCO) is presented. The assay measures the amounts of fixation products at varying [CO₂]/[O₂] ratios to measure the relative rates of CO₂ and O₂ fixation reactions. The amount of 3-phosphoglycerate (3-PGA) and phosphoglycolate (PG) in the reaction mixture were measured with a conductivity detector and the specific factor was calculated using the following equations: ν_c = ([3-PGA] – [PG])/2 and ν_o = [PG]. By this method, specificity factors for RubisCOs were measured without using radioactive reagents.     

Rapid and sensitive quantitative immunoassay for the large simian virus 40 T antigen

A quantitative, enzyme-linked immunoadsorbent assay has been developed for the simian virus 40 large T antigen. When hamster anti-simian virus 40 tumor serum was used, this method permitted specific identification of large T antigen and its analog, the D2 hybrid protein, a molecule with the same C-terminal approximately 600 amino acids as large T antigen. The sensitivity limit of this test was 0.63 ng of protein. The slopes of the regression lines of the enzyme-linked immunosorbent assay titrations performed with highly purified D2 or simian virus 40 large T antigen and with crude extracts of simian virus 40-infected monkey and transformed human cells were identical. Thus, the curve generated with a purified protein, such as D2, can serve as a quantitative standard for the measurement of large T antigen in a wide variety of extracts. Furthermore, solutions containing high salt concentrations and buffers containing up to 0.1% Nonidet P-40 did not interfere with the assay, making it applicable to the measurement of large T antigen in a variety of chromatographic fractions. The enzyme-linked immunosorbent assay was three times more sensitive, was significantly faster to perform, and was quantitatively valid over a much broader large-T-antigen concentration range than the complement fixation test. As such, it should be useful in future studies of the structure and function of this protein.     

A solid-phase method for the quantitation of protein in the presence of sodium dodecyl sulfate and other interfering substances

We describe a simple assay for small amounts of protein that is insensitive to sodium dodecyl sulfate (SDS) or many common interfering substances including Tris and reducing sugars. For this reason, it is particularly useful in the analysis of protein content of samples prior to SDS electrophoresis. The assay consists of the following steps: (i) absorption of protein to nitrocellulose; (ii) fixation of the bound protein with methanol; (iii) staining of the bound protein with amido black; and (iv) elution and spectrophotometric measurement of the bound dye. The assay is sensitive to as little as 0.5 microgram of protein in 1 microliter of solution. Although SDS does not interfere appreciably with measurement, Nonidet-P40 does.     

Establishment of mouse oligodendrocyte/type-2 astrocyte lineage cell line by transfection with origin-defective simian virus 40 DNA.

A permanent glial cell line has been established from the neonatal mouse primary mixed glial cell cultures by transfection with replication origin-defective simian virus 40 DNA. This cell line, designated OS3, has morphological similarity to type-2 astrocyte and expresses an astrocyte-specific marker, glial fibrillary acidic protein (GFAP), when cultured in the presence of 10% calf serum (CS). OS3 cells do not express the O4 antigen, galactocerebroside (GalC) and A2B5 under this culture condition. When cultured in a medium containing 2% CS or a chemically defined medium, these cells undergo morphological transformation. Some of these cells express O4 antigen and/or GalC, and the percentage of GFAP positive cells decreases under these conditions. Thus depending on the culture conditions, the OS3 cells display either type-2 astrocyte properties or immature oligodendrocyte characteristics. Furthermore, the OS3 cells show similar responses to the various growth factors as do oligodendrocyte/type-2 astrocyte (O-2A) progenitors. Therefore, the OS3 cell line is an unique mouse bipotential permanent O-2A lineage cell line which may be useful to analyze the developmental properties of these glial cells.