Oligodendroglial cells in culture effectively dispose of exogenous hydrogen peroxide: comparison with cultured neurones, astroglial and microglial cells


Address correspondence and reprint requests to Dr Ralf Dringen, Physiologisch-chemisches Institut der Universität, Hoppe-Seyler-Str. 4, D-72076 Tübingen, Germany. E-mail: ralf.dringen@uni -tuebingen.de


To investigate the antioxidative capacities of oligodendrocytes, rat brain cultures enriched for oligodendroglial cells were prepared and characterized. These cultures contained predominantly oligodendroglial cells as determined by immunocytochemical staining for the markers galactocerebroside and myelin basic protein. If oligodendroglial cultures were exposed to exogenous hydrogen peroxide (100 µm), the peroxide disappeared from the incubation medium following first order kinetics with a half-time of approximately 18 min. Normalization of the disposal rate to the protein content of the cultures by calculation of the specific hydrogen peroxide detoxification rate constant revealed that the cells in oligodendroglial cultures have a 60% to 120% higher specific capacity to dispose of hydrogen peroxide than cultures enriched for astroglial cells, microglial cells or neurones. Oligodendroglial cultures contained specific activities of 133.5 ± 30.4 nmol × min−1 × mg protein−1 and 27.5 ± 5.4 nmol × min−1 ×  mg protein−1 of glutathione peroxidase and glutathione reductase, respectively. The specific rate constant of catalase in these cultures was 1.61 ± 0.54 min−1 × mg protein−1. Comparison with data obtained by identical methods for cultures of astroglial cells, microglial cells and neurones revealed that all three of the enzymes which are involved in hydrogen peroxide disposal were present in oligodendroglial cultures in the highest specific activities. These results demonstrate that oligodendroglial cells in culture have a prominent machinery for the disposal of hydrogen peroxide, which is likely to support the protection of these cells in brain against peroxides when produced by these or by surrounding brain cells.

Abbreviations used



Dulbecco's modified Eagle's medium


fetal calf serum


fluorescein isothiocyanate




glial fibrillary acidic protein


glutathione peroxidase(s)


glutathione reductase




glutathione disulphide


total glutathione = amount of GSH plus twice the amount of GSSG


myelin basic protein


phosphate-buffered saline


penicillin and streptomycin


reactive oxygen species


trimethylrhodamine isothiocyanate.

In mammalian cells, reactive oxygen species (ROS) like peroxides and radicals are continuously generated during aerobic metabolism and, consequently, have to be detoxified continuously. These processes appear to be especially important for the brain, since oxidative stress has been connected with neurodegenerative diseases, i.e. Parkinson's disease and Alzheimer's disease (Bains and Shaw 1997; Schulz et al. 2000). Several studies demonstrate that the tripeptide glutathione (GSH) plays an important role in the detoxification of reactive oxygen species in brain and that neuronal damage induced by insults which have been discussed to act by generation of ROS is substantially increased if brain glutathione levels are reduced (for an overview see Bains and Shaw 1997; Cooper 1997; Dringen 2000; Schulz et al. 2000).

During detoxification of radicals and peroxides, GSH is involved in two types of reactions (Halliwell and Gutteridge 1999; Dringen 2000): (i) GSH reacts nonenzymatically with radicals such as the superoxide radical anion, nitric oxide or the hydroxyl radical; and (ii) GSH is the electron donor for the reduction of peroxides in the reactions catalyzed by glutathione peroxidases (GPx). The final product of the oxidation of GSH is glutathione disulphide (GSSG). Within cells GSH is regenerated from GSSG by the reaction catalyzed by the NADPH-dependent glutathione reductase (GR). Besides the glutathione system, the diffusion-controlled catalase also participates in the detoxification of hydrogen peroxide, especially if this peroxide is present in high concentrations (Aebi 1984).

Oligodendrocytes are the myelin-forming cells of the central nervous system (Baumann and Pham-Dinh 2001). Oligodendroglial damage by oxidative stress in vivo has been connected to inflammatory demyelinating disorders such as multiple sclerosis (Smith et al. 1999; Cassacia-Bonnefil 2000). In vitro, oligodendroglial cells and their precursors have been reported to be highly susceptible to ROS-induced damage (Kim and Kim 1991; Hussain and Juurlink 1995; Back et al. 1998; Richter-Landsberg and Vollgraf 1998; Hollensworth et al. 2000). Low antioxidative capacities and high concentrations of putative radical-generating iron ions have been discussed as reasons for this finding (Juurlink 1997). Treatments which reduce cellular GSH levels strongly compromise the survival of oligodendroglial cells in culture (Oka et al. 1993; Yonezawa et al. 1996; Back et al. 1998). This cell death can be prevented in the presence of free radical scavengers (Yonezawa et al. 1996; Back et al. 1998). Glutathione deprivation is especially fatal for oligodendrocyte precursors, while mature oligodendroglial cells in culture are more resistant against loss of glutathione (Back et al. 1998). These data indicate that the capacity to detoxify ROS via an intact glutathione system is essential for the survival of oligodendroglial cells in culture. Besides glutathione depletion, increased oxidative stress by application of H2O2 also leads to cell death of cultured oligodendroglial cells (Noble et al. 1994; Richter-Landsberg and Vollgraf 1998; Laszkiewicz et al. 1999). This process is mediated by activation of the transcription factors AP-1 and NF-κB (Vollgraf et al. 1999) and can be partially abolished by application of N-acetylcysteine (Richter-Landsberg and Vollgraf 1998).

During recent years glutathione metabolism and peroxide disposal of brain cells have been predominantly studied on cultures enriched for one brain cell type. From experiments performed on such cultures ample information is available regarding glutathione and peroxide metabolism of astroglial cells and neurones (for an overview see Dringen 2000). In contrast, little is known about the GSH metabolism and the peroxide disposal of oligodendroglial cells. In order to learn about the antioxidative properties of oligodendrocytes, oligodendroglia-rich cultures were prepared and characterized. Glutathione content, the activities of enzymes involved in H2O2 detoxification as well as the disposal of exogenous H2O2 were determined in these cultures and compared with values obtained under identical conditions for cultured astroglial cells, microglial cells and neurones.

Materials and methods


Dulbecco's modified Eagle's medium (DMEM) and horse serum were obtained from Life Technologies (Eggenstein, Germany). Fetal calf serum (FCS), GSH, GSSG, glutathione reductase from yeast, and insulin were purchased from Roche Diagnostics (Mannheim, Germany). NADPH and NADH were from Applichem (Darmstadt, Germany). Bovine serum albumin, cytosine arabinoside, 4′,6-diamidino-2-phenylindole (DAPI), 5,5′-dithio-bis(2-nitrobenzoic acid), poly-d-lysine, progesterone, putrescine, transferrin, 5-sulphosalicylic acid and xylenole orange were obtained from Sigma (Deisenhofen, Germany). Sodium selenite was purchased from Fluka (Neu-Ulm, Germany). Streptomycin sulphate, penicillin G, and Triton X-100 were from Serva (Heidelberg, Germany). The polyclonal antibody against glial fibrillary acidic protein (GFAP) was from Dako (Glostrup, Denmark). The monoclonal antibody against galactocerebroside (anti-Gal-C) (Ranscht et al. 1982) and the polyclonal antibody against myelin basic protein (MBP) werekind gifts from Dr B. Ranscht (La Jolla, CA, USA) and DrK. Hempel (Würzburg, Germany), respectively. Fluorescein isothiocyanate (FITC)-labelled anti-mouse immunoglobulin and the tetramethyl rhodamine isothiocyanate (TRITC)-labelled anti-rabbit immunoglobulin were from Dianova (Hamburg, Germany). Immuno-mount was from Shandon (Pittsburg, PA, USA). All other chemicals, of the highest purity available, were obtained from E.Merck (Darmstadt, Germany). Cell culture dishes (50 mm in diameter), 24-well dishes, cell culture flasks (175 cm2), and 96-well microtitre plates were from Nunc (Wiesbaden, Germany) or Greiner (Frickenhausen, Germany).

Cell cultures

Mature oligodendroglia-rich cultures were prepared by a modification of the method described by Richter-Landsberg and Vollgraf (1998). Oligodendroglial precursor cells were harvested from 175 cm2 flasks containing astroglia-rich primary cultures at a culture age between 15 days and 21 days. The flasks were closed with gas tight caps and rotated for 2 h [190 rotations per minute (r.p.m)] on a shaker (Unimax 1010, Heidolph, Kelheim, Germany) at 37°C. The medium was discarded, the cells washed with 15 mL of culture medium (90% DMEM/10% FCS containing penicillin and streptomycin [PS; 20 units/mL of penicillin G and 20 µg/mL of streptomycin sulphate)] and the cells were incubated in 50 mL of this medium. Subsequently the cultures were rotated for 17 h with 220 r.p.m. at 37°C. The oligodendroglia-containing medium was harvested (McCarthy and de Vellis 1980) and the cells pelleted by centrifugation (10 min, 500 g, 4°C). The supernatant was removed and the cells resuspended in DMEM supplemented with insulin (5 µg/mL), transferrin (5 µg/mL), and sodium selenite (25 ng/mL) (maturation medium). One million viable cells in 5 mL medium were seeded on poly-d-lysine-coated dishes (50 mm in diameter) or 200 000 viable cells in 1 mL medium in wells of poly-d-lysine-coated 24-well dishes. After 3 days in culture half of the maturation medium was renewed. The cultures were used at an age of 6 days.

Astroglia-rich primary cultures derived from the brains of newborn Wistar rats were prepared and maintained as described (Hamprecht and Löffler 1985). The cells were seeded in wells of 24-well dishes (300 000 viable cells in 2 mL) or in cell culture flasks (175 cm2; 30 million viable cells in 50 mL) and incubated in DMEM/10% FCS/PS. The results were obtained with 14- to 21-day-old cultures. These cultures contain about 90% GFAP-positive astroglial cells, up to 5% of both oligodendroglial and microglial cells as well as a few ependymal cells (Hamprecht and Löffler 1985; Gutterer et al. 1999) but no neurones (Löffler et al. 1986).

Neurone-rich primary cultures were prepared from the brains of embryonic Wistar rats as described (Löffler et al. 1986; Dringen et al. 1999b). Five hundred thousand viable cells were seeded into poly-d-lysine-coated wells of 24-well dishes in 2 mL of 90% DMEM/10% horse serum containing PS and were maintained as described previously. The cultures were used at an age of 6 days. The neurone-rich primary cultures contain 5% astroglial cells (Dringen et al. 1999a) but no oligodendroglial or ependymal cells (Löffler et al. 1986).

Microglia-rich cultures were prepared from astroglia-rich primary cultures by a modification (Hirrlinger et al. 2000) of the method described by Giulian and Baker (1986). Three hundred thousand viable cells in 2 mL medium were seeded in wells of 24-well dishes. The cultures were used at an age of 3 days and contain approximately 90% microglial cells and minor amounts of astroglial and oligodendroglial cells (Hirrlinger et al. 2000).

Immunocytochemical staining

The cells were stained according the procedure described in Gutterer et al. (1999). All staining steps were performed at room temperature. For the staining with anti-Gal-C oligodendroglial cultures on coverslips were washed twice in DMEM/25 mm HEPES, pH 7.4, followed by an incubation for 1 h with anti-Gal-C (1 : 10 diluted in DMEM/HEPES) before the fixation with paraformaldehyde took place. The cells on coverslips were fixed with 3.5% paraformaldehyde in phosphate-buffered saline (PBS; 10 mm potassium phosphate buffer, 150 mm NaCl, pH 7.4) for 10 min. After two washing periods of 10 min each in PBS and a 5-min incubation with 0.1% (w/v) glycine in PBS the cells were permeabilized by incubation in 0.3% (w/v) Triton X-100/0.1% (w/v) glycine in PBS for 10 min. After each incubation with an antibody the cells were washed twice by incubation for 5 min in PBS. With the exception of anti-Gal-C, all antibodies were diluted in PBS with 10% goat serum. The fixed and permeabilized cells were incubated with a polyclonal antibody (dilutions: anti-GFAP, 1 : 200; anti-MBP, 1 : 100) for 2 h, followed by an incubation for 30 min in the dark with FITC-labelled anti-mouse-IgG (1 : 100) and TRITC-labelled anti-rabbit-IgG (1 : 100). For quantification of the total cell number, cells were incubated for 5 min with DAPI (1 µg/mL) in PBS. The labelled cells were embedded in Immuno-mount and the fluorescence documented using fluorescence microscopes (Zeiss IM35; Olympus BH2).

Experimental incubation with H2O2

After removal of the culture medium the cells were washed with 2 mL of incubation buffer (20 mm HEPES, 145 mm NaCl, 1.8 mm CaCl2, 5.4 mm KCl, 1 mm MgCl2, 0.8 mm Na2HPO4, 5 mm glucose, pH 7.4) and incubated with 500 µL of incubation buffer (containing 50 µm or 100 µm H2O2) on an aluminium grid touching the surface of the water of a 37°C water bath for the time periods indicated. During the incubation the extracellular concentration of H2O2 was monitored by the colourimetric method described previously (Dringen et al. 1998b). Briefly, 10 µL of peroxide-containing incubation buffer, collected after gentle swirling of the dish at the time points indicated, were added to 170 µL of 25 mm H2SO4 in a well of a microtitre plate. Forty-five minutes after the further addition of 180 µL reaction mixture (0.5 mm (NH4)2Fe(SO4)2, 200 mm sorbitol, and 200 µm xylenol orange in 25 mm H2SO4), the absorbance at 550 nm was determined using a microtitreplate reader (MRX TC Revelation, Dynex Technologies, Denkendorf, Germany) and compared with the absorbance read at known standard concentrations of the peroxide investigated. The increase in absorbance of the complex generated is proportional to the peroxide content in the range of 0–2.5 nmol peroxide per well of the microtitre plate (Dringen et al. 1998b).

Determination of glutathione and protein content

The cultures were washed with 2 mL of ice-cold PBS and lysed with 0.2 mL sulphosalicylic acid [1% (w/v)] on ice. The lysates were scraped off the dish and centrifuged (1 min, 15 000 g) before aliquot parts of the supernatants were taken for measuring the content of total glutathione (GSx; amount of GSH plus two times the amount of GSSG) and GSSG in a microtitre plate assay as described previously (Dringen and Hamprecht 1996) using a modification of the photometric method originally described by Tietze (1969).

The content of glutathione per dish of a cell culture, the activities of the enzymes, and the concentrations of extracellular H2O2 were normalized to the protein content determined for replica plates of the same culture. For this purpose the cells were dissolved in 0.5 m NaOH (Hirrlinger et al. 1999) and aliquot parts of the solutions were analysed for its protein content according to the method described by Lowry et al. (1951). Bovine serum albumin was used as a standard.

Determination of enzyme activities

The activities of glutathione peroxidase (GPx), glutathione reductase (GR) and catalase were determined as previously described in detail (Hirrlinger et al. 2000). Briefly, the cells were hypotonically lyzed by incubation with 0.4 mL of 20 mm potassium phosphate buffer pH 7.0 for 10 min on ice. After centrifugation (10 min, 15 000 g, 4°C) aliquot parts of the supernatant were used for determination of GR and GPx activities and the pellet was resuspended for the catalase assay. Activities of GPx and GR were assayed by monitoring the decrease in absorbance due to the oxidation of NADPH at 340 nm in a total volume of 1 mL (Shimadzu UV-120–02 photometer, Shimadzu, Kyoto, Japan) at 30°C. GR activity was determined as described previously (Gutterer et al. 1999). The final concentrations in the reaction mixture were 100 mm potassium phosphate buffer, 1 mm EDTA, 1 mm GSSG, and 0.2 mm NADPH, pH 7.0. GPx activity was determined according to the assay described by Flohé and Günzler (1984). The final concentrations were 50 mm potassium phosphate buffer, 0.5 mm EDTA, 1 mm GSH, 0.2 mm NADPH, 1 mm sodium azide, 0.5 U GR, and 150 µm H2O2, pH 7.0. For determining catalase activity the pellet of the centrifugation was resuspended in 100 µL 1% (w/v) Triton X-100 in 20 mm potassium phosphate buffer, pH 7.0. Catalase activity was measured according to Aebi (1984) by monitoring the decomposition of H2O2 at 240 nm. As recommended, the specific activity of catalase was expressed as first-order rate constant k per mg of protein.

Presentation of data

If not stated otherwise the results shown in the figures and the tables are mean values ± SD of data obtained from n dishes derived from at least two independently prepared neural cell cultures. Statistical analysis was performed using anova followed by Bonferroni's post-hoc test. p < 0.05 was considered as statistically significant.


Immunocytochemical characterization of oligodendroglial cultures

Oligodendroglia-rich secondary cultures were prepared from astroglia-rich primary cultures derived from brains of newborn rats. After 6 days in cultures the majority of cells showed the expected (McCarthy and de Vellis 1980; Richter-Landsberg and Vollgraf 1998) morphology of mature oligodendroglial cells in culture (Fig. 1). The majority of the cells in these cultures were positively stained with the antibody against the oligodendroglial marker Gal-C (Figs 1a and d). In addition, all cells positive for Gal-C stained also for MBP (Figs 1b and e). Some cells positive for MBP (Fig. 1b) were not or only weakly stained for Gal-C (Fig. 1a), indicating that Gal-C was a more stringent marker of oligodendroglial cells in the cultures used. Counting of Gal-C-positive cells in several independently prepared cultures revealed that about 90% of the cells in oligodendroglia-rich secondary cultures represented oligodendroglial cells (Table 1). Cells positive for the astroglial marker protein GFAP contributed to only about 10% of the total number of cells in oligodendroglial cultures (Table 1). Consequently, the cultures prepared were considered suitable to study antioxidative capacities of cultured oligodendroglial cells.

Figure 1.

Immunocytochemical labelling of cells in oligodendroglia-rich cultures (c and f, phase contrast views) with antibodies against the oligodendroglial markers galactocerebroside (a and d) and myelin basic protein (b and e). The arrows in (a) and (b) point to cells positive for myelin basic protein but not for galactocerebroside. The bars in (c) and (f) correspond to 100 µm and apply to the frames (a–c) and (d–f), respectively.

Table 1.  Characterization of oligodendroglia-rich secondary cultures
Immunoreactivity forPositive cells (%)Number of cultures (n)
  1. Oligodendroglial cultures were characterized using the protocols described in Material and methods. The data represent the percentage of cells which were stained positive for the indicated glial markers, given as mean values ± SD of n independently prepared cultures. A total number of 5601 cells were counted to obtain the data presented.

Galactocerebroside87.2 ± 7.17
Glial fibrillary acidic protein 8.9 ± 1.65

Disposal of H2O2 by neural cell cultures

In order to test for the ability of oligodendroglial cells to dispose of H2O2, oligodendroglial cultures as well as cultures enriched for astroglial cells, neurones and microglial cells were incubated with the peroxide. The ability of these cultures to dispose of exogenous H2O2 (100 µm = 50 nmol/0.5 mL) was investigated by monitoring the time course of the concentration of the peroxide in the incubation buffer. In the absence of cells but under otherwise identical conditions, the peroxide is stable for at least 1 h (Dringen et al. 1999a). In contrast, in the presence of each type of brain cell culture H2O2 disappeared from the incubation buffer (Fig. 2a). There were no obvious changes in cell morphologies and no decline in cell viability during the experiments (data not shown). The clearance rate of H2O2 differed strongly between the four culture types used (Fig. 2a). Since the disposal of H2O2 followed first order kinetics (Fig. 2b), half-times for the peroxide in the presence of the different brain cell cultures could be calculated. With a half-time for H2O2 (initial concentration 100 µm) of 18.0 ± 3.4 min the clearance of the peroxide in oligodendroglial cultures was slower than in astroglial cultures (4.8 ± 0.9 min), identical to that in cultured neurones (17.5 ± 1.9 min), and much faster than that in microglial cultures (87.9 ± 33.2 min) (Table 2). If H2O2 was applied to the cultures at an initial concentration of 50 µm, half-times were around 15% lower than those for H2O2 at 100 µm initial concentration (Table 2).

Figure 2.

Disposal of exogenous H 2O2 by oligodendroglia-rich (▾), astroglia-rich (●), microglia-rich (▿) and neurone-rich (○) cultures. The cells were cultured in wells of 24-well dishes and incubated with 500 µL of incubation buffer containing 100 µm H2O2. (a) Time course of the concentration of H 2O2 in the incubation buffer of the cultures. (b) Semilogarithmic representation of the data obtained. (c) Concentration of exogenous H 2O2 normalized to the protein content per well of the respective cultures [ y = 100(lnc0 − lnc)/(pln2) = − 100Dt (c0, initial concentration of H2O2 (100 µm); c, concentration of H2O2 at the incubation time indicated; p, protein content per well)]. The content of cellular protein was determined for each culture. The mean value ± SD of protein per well was 22 ± 2 µg, 190 ± 9 µg, 8 ± 1 µg and 65 ± 9 µg for oligodendroglial, astroglial, microglial and neurone-rich cultures, respectively.

Table 2.  Protein contents, half-times of H 2O2, and specific H2O2 detoxification rate constants (D) of oligodendroglial, astroglial, microglial and neurone-rich cultures
  50 µm H2O2100 µm H2O2
Protein content
[1/(min × mg)]
[1/(min × mg)]
Culture type
  1. The cultures were incubated in incubation buffer containing H2O2 (50 or 100 µm) and the decline in extracellular concentration of the peroxide was monitored. Half-times of the peroxide were calculated from the semilogarithmic representations of the data. The specific detoxification rate constant D is defined as 1/(half-time in min ×initial protein content in mg). The data represent mean values ±SD of (n) wells of the respective cultures. The significance of the differences of the D-values obtained for oligodendroglial cultures to those of the other cultures was calculated by anova followed by Bonferroni's post-hoc test (***p < 0.001).

Oligodendroglia-rich 25 ± 11 (19)15.3 ± 2.7 (24)2.64 ± 0.45 (24)***18.0 ± 3.4 (24)2.28 ± 0.53 (24)***
Astroglia-rich168 ± 25 (33) 4.1 ± 0.8 (33)1.50 ± 0.28 (33) 4.8 ± 0.9 (33)1.29 ± 0.26 (33)
Microglia-rich 10 ± 3 (15)77.3 ± 37.3 (18)1.64 ± 0.49 (18)87.9 ± 33.2 (18)1.35 ± 0.30 (18)
Neurone-rich 60 ± 8 (12)14.1 ± 1.4 (12)1.20 ± 0.18 (12)17.5 ± 1.9 (12)0.98 ± 0.16 (12)

Specific H2O2 detoxification rate constant D

The half-time of H2O2 reflects the ability of all cells in one well of a culture to dispose of the peroxide (Dringen et al. 1999a). However, the number of cells per well differed strongly between the four culture types used as indicated by the strong differences of the protein content per well of the different cultures (Table 2). Therefore, in order to allow comparison of the peroxide detoxification of the different cultures, the peroxide concentrations present at each time point of the experiments (Figs 2a and b) were normalized to the protein content per well of the respective culture. The plot obtained due to such a calculation (Fig. 2c) demonstrated that oligodendroglial cultures disposed of exogenous H2O2 quicker than the other brain cell cultures investigated. The differences in protein content per well of the neural cell cultures are also taken into consideration by calculating the specific H2O2 detoxification rate constant (D = p−1 t1/2 − 1; p, protein content in mg; t1/2, half-time in min). Assuming that the rate constant k for the first order peroxide disposal is proportional to the amount p of cellular protein per dish, it can be derived from the known relationship in first order kinetics between k and the half-time t1/2 that p is inversely proportional to t1/2. In the equation t1/2 − 1 = D p the proportionality constant D represents the specific detoxification rate constant, which is characteristic for a certain experimental cell culture paradigm (Dringen et al. 1999a). D is proportional to the capacity of a cell type to detoxify a peroxide. The higher the D-value the better the ability of the cultured cells to dispose of a peroxide. Since all four culture types were investigated under identical conditions, the D-value is a suitable value for comparing the ability of the cells in the four culture types to dispose of H2O2. No significant differences (p > 0.05) in the D-values were found between astroglia-rich and neurone-rich cultures (Table 2). The D-values for microglial cultures were not different from those obtained for astroglial cultures (p > 0.05) but were significantly higher than those calculated for neurone-rich cultures (p < 0.05). In contrast, the D-values of oligodendroglial cultures for both initial H2O2 concentrations applied (50 µm and 100 µm) were 75%, 65%, and 125% higher (p < 0.001) than those calculated for astroglial, microglial and neurone-rich cultures, respectively (Table 2). Consequently, the cells in oligodendroglial cultures disposed of H2O2 more efficiently than the cells of the other brain cell culture types.

Specific glutathione contents and specific activities of antioxidative enzymes in neural cell culture

In order to investigate components of the oligodendroglial machinery to dispose of hydrogen peroxide, the glutathione content and activities of enzymes involved in the detoxification of H2O2 were determined for oligodendroglial cultures and compared with values previously obtained (Dringen et al. 1999a,b; Hirrlinger et al. 2000) under identical conditions for astroglia-rich and neurone-rich primary cultures as well as for microglia-rich secondary cultures (Table 3). Oligodendroglial cultures contained 33.0 ± 8.2 nmol of total glutathione (GSx) per mg of protein, approximately the same specific glutathione content as cultured astroglial cells (Table 3). The specific glutathione content of oligodendroglial cultures was significantly (p < 0.001) higher than that of neurones but lower than that of cultured microglial cells (Table 3). As in other neural cultures, GSSG accounted only for a minor part of the total glutathione content of cultured oligodendroglial cells (Table 3). In homogenates of oligodendroglial cultures the specific activities of GPx, GR, and catalase were significantly higher than those determined in homogenates of the other three types of neural cultures. The specific activities of GPx, GR, and catalase in oligodendroglial cultures were 233%, 309%, and 212% that of astroglial cultures, 347%, 229%, and 194% that of neurone-rich cultures, and 194%, 181%, 575% that of microglial cultures, respectively (Table 3).

Table 3.  Glutathione content and specific activities of glutathione peroxidase, glutathione reductase and catalase in homogenates of neural cell cultures
 Culture type
  1. GSx, total glutathione = amount of GSH plus twice the amount of glutathione disulphide. The data for the specific glutathione content and of specific enzyme activities of astroglia-rich and neurone-rich primary cultures and of microglia-rich cultures have recently been published: *(Dringen et al. 1999a), †(Hirrlinger et al. 2000), ‡(Dringen et al. 1999b) and §(Dringen and Hamprecht 1997). The activities of the enzymes were normalized to the total protein content per dish. The data were determined for (n) dishes derived from four to 14 independently prepared cultures. The activity of catalase is given as first order rate constant k. The significance of the differences of the specific enzyme activities of oligodendroglial cultures to those obtained for the other cultures was calculated by anova followed by Bonferroni's post-hoc test (***p < 0.001).

GSx content (nmol/mg protein) 33.0 ± 8.2 (33)32.1 ± 5.4 (20)*41.2 ± 11.2 (40)23.7 ± 6.0 (27)
Glutathione disulphide content (% of GSx)  3.7 ± 3.7 (22)< 2§< 2 (40)< 2.5
Specific activities of
 Glutathione peroxidase [nmol/(min × mg)]133.5 ± 30.4 (21)***57.2 ± 3.9 (14)68.7 ± 23.5 (21)38.5 ± 8.0 (14)
 Glutathione reductase [nmol/(min × mg)] 27.5 ± 5.4 (20)*** 8.9 ± 2.0 (14)15.2 ± 3.2 (11)12.0 ± 2.2 (14)
 Catalase [k; 1/(min × mg)] 1.61 ± 0.54 (14)***0.76 ± 0.07 (14)0.28 ± 0.11 (20)0.83 ± 0.14 (14)


In order to investigate antioxidative defence mechanisms of oligodendrocytes, oligodendroglia-rich secondary cultures were prepared and characterized. These cultures contain about 90% oligodendroglial cells as demonstrated by immunocytochemical staining for glial markers. The predominant morphology of the Gal-C-positive and MBP-positive cells in the oligodendroglial cultures was that of differentiated mature oligodendroglial cells (McCarthy and de Vellis 1980; Richter-Landsberg and Vollgraf 1998). Therefore, these cultures can be considered as a useful model system for obtaining information on the peroxide metabolism of mature oligodendroglial cells.

In order to allow comparison of antioxidative capacities of different neural cell types, the clearance of H2O2 was studied in one experimental paradigm using cell cultures enriched for oligodendroglial cells, astroglial cells, microglial cells or neurones. Kinetics for peroxide disposal by cells in oligodendroglia-rich and microglia-rich cultures have so far not been reported. However, as has been shown previously for cultures of astroglial cells and neurones (Dringen and Hamprecht 1997; Dringen et al. 1998a; Dringen et al. 1999a; Kussmaul et al. 1999), also cultures enriched for oligodendroglial and microglial cells cleared extracellular H2O2 following first-order kinetics.

Apparently, astroglial cultures were highly potent in H2O2 clearance while the other culture types appeared to be less efficient. However, it has to be considered that cultures of oligodendroglial cells, neurones and microglial cells are not confluent and contain a much lower number of peroxide-detoxifying cells per well than confluent astroglial cultures. In order to compare cultures of different cell densities regarding their capacity for peroxide detoxification, the H2O2 concentration was normalized to the protein content of the respective cultures. In addition, the specific detoxification rate constant D was calculated which equals to the inversed product of the half-time of a peroxide (as indicator for the detoxification by the cells present on one dish) with the protein content (as indicator for the amount of cells responsible for peroxide detoxification). D is proportional to the capacity of a cell type to detoxify a peroxide (Dringen et al. 1999a). Both types of analysis revealed that the high ability of astroglial cultures to clear exogenous H2O2 was most likely due to a higher cell density. Compared with the other three culture types, the D-values for oligodendroglial cultures were significantly higher, indicating that oligodendroglial cells are able to dispose of exogenously applied H2O2 better than the other brain cell types. The reason for this finding is most likely the presence of GPx, GR and catalase in higher specific activities in oligodendroglial cultures. However, we cannot exclude that small molecular weight antioxidants could also participate in the high capacity of oligodendroglial cells to detoxify H2O2.

The D-values calculated for astroglial cultures and cultured neurones varied at best marginally compared with the half-times for H2O2 and the protein content of these cultures. This observation confirms previous results which were obtained using a different experimental paradigm (Dringen et al. 1999a). Although microglial cultures contain only little catalase activity, the D-values of this culture type differed not significantly to those obtained for astroglial cultures and are even higher than those of neurones. That suggests that, compared with astroglial and neurone cultures, the lower specific catalase activity in microglial cultures is compensated for by the stronger glutathione system (Hirrlinger et al. 2000).

The oligodendroglial cultures used in the present study contained about 30 nmol glutathione per mg of protein. This value is higher than data previously published for cultured oligodendroglial cells (Back et al. 1998; Juurlink et al. 1998; Hollensworth et al. 2000). In our hands, the glutathione content of oligodendroglial cultures is not significantly different to that of astroglia-rich cultures, whereas previously oligodendroglial cultures have been reported to contain less (Juurlink et al. 1998) or much more (Hollensworth et al. 2000) glutathione than astroglial cultures. Such discrepancies of glutathione levels reported for cultures of one brain cell type are not unusual. Also for cultures of neurones and astroglial cells a large range of specific glutathione contents have been reported in the literature (for an overview see Dringen 2000). Most likely different culture conditions as well as different assay systems are responsible for such inconsistent findings.

Previously, it has been reported that activities of GPx and catalase in oligodendroglial cultures are almost identical to those of astroglial cultures (Hollensworth et al. 2000) and that activities of GPx and GR are much lower than those of astroglial cultures (Juurlink et al. 1998). However, in the present study the comparison of oligodendroglia-rich cultures to astroglial, microglial, and neuronal cultures (Dringen and Hamprecht 1997; Dringen et al. 1999a,b; Hirrlinger et al. 2000) revealed that amongst these cultures oligodendroglia-rich cultures contain significantly higher specific activities of GPx, GR and catalase. The high specific activity of GR in oligodendroglia-rich cultures is in accord with the observation that in astroglia-rich primary cultures especially oligodendroglial cells stain intensively for GR (Gutterer et al. 1999). In addition, the higher specific activity of catalase in oligodendroglial cultures compared with mixed glial cultures is also in accord with a previous report (Singh et al. 2000) as well as with the prominent immunocytochemical staining of catalase in oligodendroglial cells in brain slices (Moreno et al. 1995), reaggregation cultures (Aspberg and Tottmar 1994) and astroglia-rich cultures (Sokolova et al. 2001). These higher activities in oligodendroglia-rich secondary cultures of enzymes involved in H2O2 disposal are most likely the reason why these cultures detoxify exogenous H2O2 more efficiently than the other types of brain cell cultures investigated here.

If different types of brain cell cultures are to becomparedregarding their resistance against peroxide stress, the differences in cell density and numbers of peroxide detoxifying cells per well have to be considered. If a given initial concentration of a peroxide is applied to a low number of cells, the concentration of the peroxide in the respective well will decrease slower than in a well with high cell density, even if the intracellular antioxidative machinery per cell is identical for high and low density cultures. Consequently, the cells incubated in low density are exposed for a longer time period to the peroxide and therefore the toxic effects of the peroxide might become more prominent (Dringen et al. 1998b) and cells incubated at higher cell density appear to be more resistant against peroxide damage than those of low cell density. Therefore, a single application of peroxides may not be a good experimental paradigm to compare cell cultures of different cell densities regarding their resistance against exogenously applied oxidative stress.

The prominent ability of cultured oligodendroglial cells to clear H2O2 might reflect the properties of oligodendrocytes in brain. These cells contain and produce substantial amounts of unsaturated lipids which are targets of lipid peroxidation, have a strong oxidative metabolism (Juurlink 1997), which favours the generation of ROS and contain, amongst the brain cells, the highest amount of iron (Connor and Menzies 1996) which, in a redox-active form, would catalyse the generation of hydroxyl radicals by Fenton chemistry. Consequently, oligodendrocytes are especially in need of a good defence system against radicals and peroxides. A strong glutathione system as indicated by a substantial GSH content and the high specific activities of GPx and GR in oligodendroglial cells is perfectly suited for the defence against reactive compounds, since glutathione reacts directly with the radicals in nonenzymic reactions and is also electron donor in the GPx-catalyzed reduction of peroxides (Halliwell and Gutteridge 1999; Dringen 2000). In addition, the NADPH required for rapid glutathione recycling is also efficiently regenerated in oligodendroglial cells because these cells strongly express the NADP+-reducing enzymes glucose-6-phosphate dehydrogenase (Kugler 1994) and cytosolic malic enzyme (Kurz et al. 1993), as well as NADP+-dependent isocitrate dehydrogenase (T. Minich and R. Dringen, unpublished results).

However, even the strong capacity of oligodendroglial cells to dispose of peroxides might be insufficient to prevent oligodendroglial damage due to oxidative stress. These cells have been reported to be highly susceptible towards oxidative stress, even more so than other brain cell types (Kim and Kim 1991; Hussain and Juurlink 1995; Back et al. 1998; Richter-Landsberg and Vollgraf 1998; Hollensworth et al. 2000). Our findings are not contrary to these findings. They might even reflect an adaptation to compensate at least partially for the disadvantages of oligodendroglial cells regarding ROS generation and ROS-mediated damage.

The vulnerability of oligodendroglial cells to oxidative stress in cultures has been investigated using cultures enriched with oligodendroglial cells (see above). These cultures do not contain substantial numbers of the other brain cell types which are the physiological neighbours of oligodendrocytes in brain. Therefore, in oligodendroglial cultures the support of oligodendroglial cells by other brain cell types against exogenously applied peroxides is almost completely prevented. In brain, particularly astrocytes are considered to have important functions in the antioxidative defence and in the protection of oligodendrocytes and neurones against peroxides and radicals (Wilson 1997; Cooper 1998; Dringen 2000). This view is strongly supported by the findings that astroglial cells protect cocultured oligodendroglial cells (Noble et al. 1994) and neurones (Langeveld et al. 1995; Desagher et al. 1996) against oxidative stress. In brain, besides the astrocytes, oligodendrocytes could also participate in the protection of neighbouring neurones against oxidative stress.

In conclusion, mature oligodendroglial cells in culture have a high capacity for the disposal of H2O2 which is more prominent than that of cultured astroglial cells, microglial cells and neurones. These results are likely to reflect the higher demand of protection against peroxides which are produced by oligodendrocytes themselves and their neighbouring cells.


The authors thank the Deutsche Forschungsgemeinschaft and the Strukturfond der Universität Tübingen for financial support.