Oxidative Stress

NITRIC OXIDE: Biology and Chemistry Vol. 5, No. 2, pp. 137–149 (2001) doi:10.1006/niox.2001.0335, available online at http://www.idealibrary.com on

Protection of Primary Glial Cells by Grape Seed Proanthocyanidin Extract against Nitrosative/ Oxidative Stress Sanjoy Roychowdhury,* ,1 Gerald Wolf,* Gerburg Keilhoff,* Debasis Bagchi,† and Thomas Horn* *Otto-von-Guericke University, Institute for Medical Neurobiology, Leipziger Strasse 44, D-39120 Magdeburg, Germany; and †Department of Pharmacy Sciences, Creighton University School of Pharmacy & Allied Health Professions, Omaha, Nebraska 68178

Received October 10, 2000, and in revised form December 21, 2000; published online March 21, 2001

Previous studies showed that proanthocyanidins provide potent protection against oxidative stress. Here we investigate the effects of grape seed proanthocyanidin extract (GSPE) as a novel natural antioxidant on the generation and fate of nitric oxide (NO) in rat primary glial cell cultures. GSPE treatment (50 mg/L) increased NO production (measured by NO 2ⴚ assay) by stimulation of the inducible isoform of NOS. However, GSPE failed to affect the LPS/IFN-␥-induced NO production or iNOS expression. Similar responses were found in the murine macrophage cell line RAW264.7. GSPE did not show any effect on dihydrodichlorofluorescein fluorescence (ROS marker with high sensitivity toward peroxynitrite) either in control or in LPS/IFN-␥induced glial cultures even in the presence of a superoxide generator (PMA). GSPE treatment alone had no effect on the basal glutathione (GSH) status in glial cultures. Whereas the microglial GSH level declined sharply after LPS/IFN-␥ treatment, the endogenous GSH pool was protected when such cultures were treated additionally with GSPE, although NO levels did not change. Glial cultures pretreated with GSPE showed higher tolerance toward application of hydrogen peroxide (H 2O 2) and tert-butylhydroperoxide. Furthermore, GSPEpretreated glial cultures showed improved viability after H 2O 2-induced oxidative stress demonstrated 1

To whom correspondence should be addressed.

1089-8603/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

by reduction in lactate dehydrogenase release or propidium iodide staining. We showed that, in addition to its antioxidative property, GSPE enhances low-level production of intracellular NO in primary rat astroglial cultures. Furthermore, GSPE pretreatment protects the microglial GSH pool during high output NO production and results in an elevation of the H 2O 2 tolerance in astroglial cells. © 2001 Academic Press

Key Words: iNOS; hydrogen peroxide; astrocytes; microglia; glutathione; DCF.

Reactive oxygen species (ROS) such as superoxide, hydrogen peroxide, and hydroxyl radicals as well as nitric oxide and their metabolites are the subject of intense research because of their active role in cellular pathology and because they also affect the CNS under various neuropathological conditions (1–3). Therefore the quest for antioxidative drugs, including those that are derived from natural sources, is of general interest. Proanthocyanidins are potent natural antioxidants composed of various polyphenolic components, including phenoldienones, epicatechin, epigallocatechin, epigallocatechin gallate, ferulic acid, caffeic acid, p-coumaric acid, kaempferol, quercetin, and myricetin (4 –7). These compounds possess a broad spectrum of antioxidative properties that protects against free radicals and oxidative stress, both in 137

138

ROYCHOWDHURY ET AL.

vitro and in vivo (8, 9). Previous studies showed that proanthocyanidins provide significant protection against free-radical-induced lipid peroxidation and DNA fragmentation in liver and brain tissue (9) and provided better protection than vitamin C, vitamin E, and ␤-carotene. Although various attempts have been made to monitor the antioxidative effects of proanthocyanidins, the mechanisms by which these compounds affect the generation of nitric oxide and other free radicals in situ are still unclear. Our objective was to test the effect of grape seed proanthocyanidin extract (GSPE) on the generation and fate of oxygen and nitrogen radicals in a glial cell culture model. Grape seeds have already been reported to be a good source of proanthocyanidins (10). We chose the model of mixed primary glial cultures for our study that contains both iNOSexpressing (microglia) and -nonexpressing (astroglia) cells (11, 12). Microglia cells play a dual role in neuropathological processes: they are a major source of NO and other radicals including their metabolites, and at the same time they produce GSH, which is delivered to neighboring cells to protect them against oxidative stress (13). The role of NO in neuropathology is still a matter of controversy: it was reported to have both neuroprotective (1, 14, 15) and neuropathological effects in neurodegenerative diseases (14, 15). NOmediated pathophysiology was presumed to be caused mainly by its reaction product with superoxide, forming the highly cytotoxic peroxynitrite (16, 17). Upregulation of iNOS in microglial cells or invading macrophages is considered to be the source for high NO output (18). To study the effect of GSPE on intracellular production of nitric oxide, rat primary glial cultures and murine macrophage-derived RAW264.7 cells were treated with GSPE in the presence or absence of LPS/IFN-␥ and the iNOS inhibitor N-iminoethyl lysine (L-NIL), and the NO oxidation product nitrite released into the medium was assayed. Intracellular GSH, a major endogenous antioxidant (13), was labeled with monochlorobimane in iNOS-induced glial cultures in the presence or absence of GSPE. Furthermore, to examine the effect of GSPE on peroxynitrite formation, changes in dihydrodichlorofluorescein (DCF) fluorescence (ROS marker with high sensitivity toward peroxynitrite) intensity were

monitored in cultures treated with GSPE in the presence or absence of LPS/IFN-␥ and L-NIL. Moreover, to test the effect of GSPE on hydrogen peroxide (H 2O 2) tolerance of glial cells, DCF fluorescence was monitored in GSPE-pretreated and untreated control cultures during H 2O 2 application. In addition, to examine the effect of GSPE on cell viability after H 2O 2 exposure, propidium iodide staining and lactate dehydrogenase (LDH) assay were employed in glial cultures treated with H 2O 2 (19). MATERIALS AND METHODS

Reagents 2⬘,7⬘-Dichlorodihydrofluorescein diacetate, (H2DCFDA), monochlorobimane (mBCl), and propidium iodide were purchased from Molecular Probes, Inc. (Eugene, OR). Phorbol 12-myristoyl 13-acetate (PMA) was purchased from ALEXIS Deutschland (Gru¨nberg, Germany). A commercially available powdered IH636 grape seed proanthocyanidin extract (GSPE, Batch 609016) was obtained from the Interhealth Nutraceuticals Incorporated (Benicia, CA, U.S.A.). Lipopolysaccharide (LPS, Escherichia coli) and interferon-␥ (IFN-␥) were purchased from Sigma (St. Louis, MO) and Gibco (Berlin, Germany), respectively. N-Iminoethyl lysine was purchased from Calbiochem (Bad Soden, Germany). Rabbit anti-iNOS antiserum was obtained from Upstate Biotechnology, Inc. (New York). Microgliaspecific monoclonal anti-Ox-42 was purchased from Becton Dickinson GmbH (Pharmingen, Europe), and the corresponding secondary antibodies used were CY-3-labeled goat anti-rabbit IgG (Rockland, Germany) and Alexa-488-labeled goat anti-mouse IgG (Molecular Probes), respectively. Cell Cultures Primary glial cultures were prepared from the cortex of newborn rats (Wistar) by the method of Schousboe (20). In brief, the cerebral cortex was removed, cleaned of meningeals, and placed in Dulbecco’s modified Eagle medium (DMEM), supplemented with 10% fetal bovine serum (FBS). After mechanical dispersion, aliquots of cell suspension were plated on poly-D-lysine-coated coverslips

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

PROTECTION OF PRIMARY GLIAL CELLS BY GSPE

(25-mm diameter) in 35-mm-diameter petri dishes (for confocal studies) or directly on 35-mm petri dishes (for NO measurement, GSH-mBCl study and cell viability experiments) at a final density of 2 ⫻ 10 6 cells per dish. Cultures were maintained at 37°C in a humidified 5% CO 2 atmosphere for 10 days in vitro (DIV). Cultures maintained under these conditions were almost free of neurons and contained mainly astroglial cells (up to 90%) and microglial cells (10 –20%). About 2% of the cells were oligodendrocytes (determined by immunohistochemistry; data not shown). Mouse macrophage-like RAW 264.7 cells were grown in RPMI medium supplemented with 10% BS, 2 mM glutamine, and ampicilin (5000 units/mL) for 2 DIV, prior to the experiments. Experiments GSPE (150 g/L) was first dissolved in 70% alcohol and then diluted to 5 g/L in Hepes buffer (pH 7.4). The diluted GSPE solution was passed through a 0.22-␮m sterile Millipore filter. The rat primary glial cultures and RAW 264.7 cell cultures were incubated with the diluted solution of GSPE (5 g/L) in the culture medium (final concentration in the medium, 50 mg/L) for 18 h. To upregulate the inducible nitric oxide synthase isoform (iNOS), LPS from E. coli (2 ␮g/mL culture medium) and IFN-␥ (100 units/mL culture medium) were added to the cell cultures as described earlier by Noack et al. (19). L-NIL (0.5 mmol/L) was used to inhibit the iNOS-mediated NO formation during iNOS induction. After 18 h, the cell culture supernatants were assayed for nitrite levels. Cultures treated with the vehicle alone are considered to be the control. Nitrite Analysis Nitric oxide levels in culture supernatants were determined using a Sievers NO Analyser in conjunction with the computerized data analysis program NOAWIN as described before (21). In brief, nitrite in the biological samples was reduced to NO by potassium iodide in presence of acetic acid. The generated NO was carried from the reaction vessel to the analysis chamber by a steady flow of N 2. Chemilumines-

139

cence that resulted from the reaction of ozone with NO was measured via a photomultiplier. The instrument was calibrated by injection of different NaNO 2 concentrations with a fixed sample volume. H 2DCF-DA Fluorescent Measurements H 2DCF-DA fluorescence in glial cultures was measured according to earlier reports from our laboratory (22). Glial cultures on coverslips were incubated with 50 ␮M H 2DCF-DA in Locke’s solution (154 mM NaCl, 5.6 mM KCl, 2.3 mM CaCl 2, 1 mM MgCl 2, 3.6 mM NaHCO 3, 15 mM Hepes, 10 mM glucose, pH 7.3) for 30 min, washed with Locke’s solution, mounted in a steel chamber (Attoflour), and covered with 1 mL of Locke’s solution. Experiments were conducted using a Zeiss laser scanning microscope (LSM 410 Axiovert, Zeiss 40 ⫻ oil lens; excitation, 488-nm argon laser; emission, 515-nm long pass filter). Laser attenuation, pinhole diameter, photomultiplier sensitivity, and offset were kept constant for every set of experiments. The obtained data were quantitatively analyzed using the Zeiss LSM software. For continuous monitoring of DCF fluorescence, a time series of images was started and H 2O 2 or tertbutylhydroperoxide (t-BH) was added to the cells after 3 min of baseline fluorescence collection. The fluorescence intensity was monitored at an interval of 1 min for a total period of 10 min. Fluorescence values were normalized to the H 2O 2 or t-BH pretreatment levels for quantitative analysis. Immunocytochemistry Cultures grown on poly-D-lysine-coated coverslips were washed in phosphate buffered saline (PBS) and fixed for 10 min with 4% paraformaldehyde (PFA) and 0.2% glutaraldehyde in PBS. The cultures were treated for 30 min with blocking buffer (PBS, 0.05 M sodium phosphate, 0.15 M NaCl, pH 7.4, containing 0.3% Triton X-100 and 3% normal goat serum) followed by an overnight incubation with the primary antibodies (rabbit anti-iNOS, 1:1000, and microglia specific mouse anti-OX-42, 1:800, at 4°C), diluted in blocking buffer. The coverslips were washed with PBS, incubated overnight at 4°C with the appropriate fluorochrome-conjugated secondary antibodies

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

140

ROYCHOWDHURY ET AL.

FIG. 1. GSPE (50 mg/L) increased basal NO production after 18 h of incubation in both rat primary glial cultures (10 DIV, A) and RAW264.7 cell cultures (2 DIV, B). Nitrite production in the cultures was stimulated by LPS (2 ␮g/mL)/IFN-␥ (100 units/mL) treatment (18 h). GSPE in combination with LPS/IFN-␥ showed a similar increase in nitrite level in the culture medium. The NOS inhibitor L-NIL (0.5 mmol/L) inhibited this enhanced nitrite accumulation in the medium. Data represent the mean concentration ⫾SEM of 4 sets of experiments of 3 culture dishes for each treatment (n ⫽ 12). Significance levels were calculated by using ANOVA single factor analysis, and P ⬍ 0.05 is designated by an asterisk (*).

(CY-3-labeled goat anti-rabbit 1:250 IgG and Alexa 488-labeled goat anti-mouse IgG 1:500), which were diluted in blocking buffer, and then mounted in Immunomount (Shandon, UK) to prevent fading. Fluorescence images were collected using a Zeiss laser scanning microscope with a 63 ⫻ oil lens, using dual excitation at 488 and 543 nm, emission at 515–

565 nm bandpass, and a long pass filter of 570 nm, respectively. Confocal fluorescence images were obtained as described above. Images of the doublelabeled cultures (anti-iNOS and anti-OX-42) were analyzed by the integrated Zeiss software for overall fluorescence intensities and their frequency distribution. Pixel co-localization was depicted by plotting

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

PROTECTION OF PRIMARY GLIAL CELLS BY GSPE

141

FIG. 2. iNOS immunoreactivity was not affected by GSPE. GSPE (50 mg/L) treatment alone had no effect on basal or induced (LPS, 2 ␮g/mL; IFN-␥, 100 units/mL) iNOS immunoreactivity. The iNOS inhibitor L-NIL (0.5 mmol/L) or GSPE (50 mg/L) added in combination with LPS/IFN-␥ did not show any alteration in LPS/IFN-␥-induced iNOS immunoreactivity. The fluorescence measurement was performed in not less than 2 culture dishes of 3 different culture dates (total number of cells, n ⫽ 20). Significance levels were calculated by using ANOVA single factor analysis, and P ⬍ 0.05 is designated by an asterisk (*).

the frequency distribution of the two fluorescence labels against each other. Measurement of Intracellular Glutathione Level with Monochlorobimane

imaging program Kappa Image Base. Fluorescence intensity was quantified by the Scion Image program using digitized images of four culture dishes for each treatment. Cell Viability Assay

Intracellular GSH was measured by monitoring the fluorescence of mBCl– glutathione adduct in situ according to Chatterjee et al. (23). Glial cultures were incubated at 37°C with 60 ␮M mBCl in PBS for 30 min. Cultures were then washed with PBS and fixed with 4% paraformaldehyde and 0.2% glutaraldehyde in 0.1 M PBS for 10 min, followed by three PBS washes and then coverslipped with Immunomount (Shandon UK). Images were taken using a fluorescence microscope (Axiophot, excitation, 365– 400 nm; emission, 450 – 490 nm) equipped with a CCD camera (Kappa CF 8/1 DXC) controlled by the

Cell viability assay with propidium iodide and the determination of lactate dehydrogenase activity were performed as described previously (19). Fluorescence images obtained from propidium iodide staining were quantified by the Scion Image program using digitized images of two culture dishes for each treatment. Data Analysis In all experiments, fluorescence values were measured within squares drawn inside the individual

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

PROTECTION OF PRIMARY GLIAL CELLS BY GSPE

whole cells. For the continuous fluorescence measurement experiments, data were presented for each test condition as the change in normalized fluorescence (mean ⫾ SEM). Each condition was tested on at least two coverslips from no less than two different culture dates, typically yielding a total cell of 20 –30 cells per condition (see figures for details). Statistical significance was determined by ANOVA single factor analysis, and values with P ⬍ 0.05 were considered significant. RESULTS

Effect of GSPE on NO Production in Rat Primary Glial Cultures and RAW264.7 Cell Line To test the effect of GSPE on basal and induced NO levels, cultures were pretreated with GSPE in the presence or absence of LPS/IFN-␥ and L-NIL followed by the measurement of medium nitrite accumulation after 18 h. Our experiments showed that the GSPE treatment significantly increased the nitrite production in primary rat glial cultures (to 147.7% of control; P ⬍ 0.001, Fig. 1A). LPS/IFN-␥ treatment in comparison to GSPE alone, however, was much more potent in increasing the production of NO (to 4200% of control). These nitrite levels were similar to those received GSPE in combination with LPS/IFN-␥ (4761% of control). The presence of L-NIL prevented the increase in nitrite level in all cases, indicating that the increase in nitrite level was due to elevated iNOS activity (Fig. 1A). The same protocol was repeated with the macrophage cell line RAW264.7 to test the effect of GSPE on other cells of the immune system that express

143

iNOS in a similar pattern. Basal nitrite levels in control RAW264.7 cultures were higher (639%) than those in rat glial cultures (P ⬍ 0.001). GSPE treatment increased the nitrite level (222% of control) also in the RAW264.7 cell line, but the extent of this increase was higher than that in glial cultures (to 147.7% of control). When RAW264.7 cells were stimulated with LPS/IFN-␥, the nitrite level was increased (800% of control). Interestingly, the rise in nitrite level was similar to that after LPS/IFN-␥ in combination with GSPE treatment (853% of control). The presence of L-NIL prevented all GSPEinduced increases in nitrite levels (Fig. 1B).

iNOS Immunoreactivity To investigate the effect of GSPE on iNOS expression, cell cultures were treated with different combinations of GSPE, LPS/IFN-␥, and L-NIL followed by immunostaining for iNOS. The fluorescence signal of the iNOS-IR was unaffected by GSPE alone, while it was strongly increased after LPS/IFN-␥ treatment (P ⬍ 0.01, Fig. 2). GSPE treatment in combination with LPS/IFN-␥ did not alter the iNOS-IR in comparison to LPS/IFN-␥ alone. The presence of L-NIL did not alter the LPS/IFN-␥stimulated expression of iNOS. To map the cell-type specificity of iNOS induction in the glial cultures, iNOS was induced by LPS/IFN-␥ treatment in the rat primary glial cultures and thereafter co-stained for iNOS- and OX-42-IR. As depicted in Fig. 3, iNOS was found to be induced exclusively in the microglial cells, as shown previously (19, 22).

FIG. 3. Induction of iNOS was restricted to microglial cells in LPS/IFN-␥-induced primary glial cultures. The upper panel shows the single channel images for the individual immunoreactivity (A, iNOS; and B, OX-42), and the lower panel depicts the overlay of both channels (yellow indicates colocalization; C). Graph D shows the pixel intensity frequency for both channels. The frequency distribution of iNOS and the OX-42-IR fluorescence signal intensities are drawn on the x and y axes, respectively, to visualize colocalization. FIG. 4. GSPE showed no significant effect on peroxynitrite levels. Oxidation of H 2DCF-DA to the fluorescent DCF (peroxynitritesensitive indicator) in LPS (2 ␮g/mL)/IFN-␥ (100 units/mL) stimulated or in control cultures in the presence or absence of GSPE (50 mg/L). GSPE alone did not show any significant difference from the control, while LPS/IFN-␥ treatment induced the DCF fluorescence signal, which was diminished in the presence of the iNOS inhibitor L-NIL. Cultures were stimulated for 18 h by LPS/IFN-␥ with the simultaneous application of GSPE and/or L-NIL (0.5 mmol/L). Cells were loaded with H 2DCF-DA and mounted on the microscope followed by PMA (3 ␮M for 5 min) application to induce superoxide generation. The upper panel represents the confocal fluorescence images of the cultures with different treatments, as shown below the images. Fluorescence intensity was measured using not less than 2 culture dishes of 3 different culture dates (n ⫽ 20). Significance levels were calculated by using ANOVA single factor analysis, and P ⬍ 0.05 is designated by an asterisk (*).

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

144

ROYCHOWDHURY ET AL.

FIG. 5. Significant protection of mBCl-GSH fluorescence was observed in GSPE-pretreated cultures 18 h after LPS/IFN-␥ treatment. Fluorescence images of mBCl-GSH adduct in rat primary glial cultures with different treatments are shown in the images. GSPE (50 mg/L) caused preservation of mBCl-GSH fluorescence after LPS (2 ␮g/mL)/IFN-␥ (100 units/mL) treatment (18 h), although it did not bring any significant change alone. The upper panel shows the microscopic images, while the lower panel represents the fluorescence intensities in microglia (A) and astrocytes (B) separately. Significance levels were calculated using ANOVA single factor analysis, and P ⬍ 0.05 is designated by an asterisk (*).

Effect of GSPE on Peroxynitrite Generation (H 2DCF-DA Fluorescence) We compared the change in DCF fluorescence intensity in glial cultures following LPS/IFN-␥ treatment to control and GSPE-treated cultures in the presence of superoxide generated by PMA. Cultures treated with GSPE alone in the presence of PMA did not show any significant change compared to controls (Fig. 4). Cultures pretreated with LPS/IFN-␥ showed a 780% higher fluorescence intensity in the presence of PMA, compared to controls, while this increase was diminished by prior treatment with L-NIL. DCF fluorescence levels of LPS/IFN-␥treated cultures were similar to those cultures which received GSPE in addition to LPS/IFN-␥. In the absence of PMA, cultures treated with LPS/ IFN-␥ showed a much lower increase in DCF fluorescence compared to controls (data not shown), as reported previously from our laboratory (22).

Intracellular GSH and the Role of GSPE Fluorescence microscopy using mBCl as the GSH conjugating fluorochrome was employed to study the effect of GSPE on intracellular GSH content, both in control and in LPS/IFN-␥-induced cell cultures (Fig. 5). Our fluorescence images showed that the round microglia cells (identified by OX-42 immunoreactivity) possess much higher GSH contents than the neighboring astrocytes. GSPE alone did not affect the intracellular mBCl-GSH fluorescence (Fig. 5A). However, the fluorescence intensity of the mBClGSH adduct in LPS/IFN-␥-treated microglia cells was 39% less than the control. However, when cultures were treated with LPS/IFN-␥ in combination with GSPE, the intracellular mBCl-GSH fluorescence signal was preserved significantly (P ⬍ 0.01), although the nitrite concentrations were similar in these cultures. Interestingly, there was no detect-

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

PROTECTION OF PRIMARY GLIAL CELLS BY GSPE

able change in mBCl-GSH fluorescence in astrocytes under any treatment mentioned above (Fig. 5B). Effect of GSPE on H 2O 2-Induced Toxicity Glial cell cultures were loaded with H 2DCF-DA followed by H 2O 2 (0 –200 ␮M) treatment. Our experiments using a confocal laser scanning microscope showed that H 2O 2 increased the DCF signal in the concentration range 50 –200 ␮M (Fig. 6A). Comparing the end-point fluorescence levels, the increase was minimal at 50 ␮M (600% over baseline) and maximal at 200 ␮M (1226% over baseline). To assess the antioxidative potential of GSPE in this model of oxidative stress, cells pretreated with GSPE were exposed to H 2O 2 (100 ␮M) or t-BH (100 ␮M), and the change in DCF fluorescence signal was measured (Fig. 6B). Cell cultures pretreated with GSPE showed a lower (51% at the end point) rise in fluorescence signal compared to the controls during 100 ␮M H 2O 2 treatment. In cell cultures pretreated with GSPE, a similar trend was seen during t-BH application, but the extent of protection was lower (24%) than that of the H 2O 2 application (51%, Fig. 6C). To monitor the effect of GSPE on H 2O 2-induced oxidative stress in glial cell cultures, cell viability was examined with propidium iodide staining, which stains only dead cells (Fig. 7). We also measured the relative release of LDH into the medium, an indicator of cell viability following H 2O 2 treatment (400 ␮M H 2O 2 for 6 h, Fig. 7). Cultures treated with H 2O 2 showed a higher propidium iodide fluorescence when compared to controls (3154% of control, Fig. 7). Surprisingly, GSPE pretreatment abolishes that H 2O 2-induced increase in propidium iodide staining in such cells. These experiments also showed that the GSPE-pretreated cultures had 30% (P ⬍ 0.05) less release of LDH than the controls under H 2O 2 treatment alone. DISCUSSION

Although proanthocyanidins are shown to possess high antioxidative potential, the mechanism of their cytoprotection against oxidative stress is yet to be unraveled. Our results show that GSPE treatment increases the nitric oxide production in glial cells as well as in RAW264.7 cells by activating iNOS. A

145

similar trend was observed in RAW264.7 cell cultures, which points toward a generalized increase in iNOS activity in the presence of GSPE in both the cell types, although there was no significant difference in iNOS-IR in glial cultures. However, LPS/ IFN-␥-induced high-output NO production was not altered by GSPE. Moreover, our immunohistochemical study confirmed that the induction of iNOS protein in the presence of LPS/IFN-␥ was specifically localized in microglia cells, consistent with our previous results (19, 22). Recent studies showed that NO can be regulatory, protective, or toxic depending on its rate of cellular production (24). Low-output NO production can act protectively against ROS-associated injury (25), whereas a higher rate of NO production has been reported to cause cytotoxicity (26). Our study showed that, although GSPE alone causes low output NO production and did not increase either propidium iodide staining or LDH release (data not shown), it proves not to be cytotoxic at the concentration used. Earlier studies showed that the pathological effects of NO are partly mediated by its reaction with superoxide radicals, yielding the strong oxidant peroxynitrite (27). We therefore used H 2DCF-DA, a ROS-sensitive fluorochrome, that has recently been reported to be a very sensitive tool for the detection of peroxynitrite in situ (22) to monitor its formation with or without GSPE treatment. We found that GSPE alone did not increase the DCF fluorescence even in the presence of PMA-induced superoxide generation. LPS/IFN-␥-treated cultures showed a significant increase in DCF fluorescence in the presence of PMA, indicating enhanced peroxynitrite formation, which was inhibited by the iNOS blocker L-NIL. These findings are consistent with our previous reports (22). GSPE alone increased the baseline nitrite level only by 47.7%, which could be below a certain threshold level that is necessary for the detection of peroxynitrite. Increased production of NO following LPS/IFN-␥ treatment has already been reported to cause depletion of GSH and resulted in enhanced formation of S-nitrosoglutathione (GSNO) in glial cultures (12, 28). Since GSH is one of the major cellular antioxidants, conservation of the GSH level under oxidative stress should be beneficial to the cells. Our data

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

146

ROYCHOWDHURY ET AL.

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

PROTECTION OF PRIMARY GLIAL CELLS BY GSPE

147

FIG. 7. GSPE-pretreated (50 mg/L) glial cultures were more tolerant to the H 2O 2-induced oxidative stress compared to controls. The upper panel represents the propidium iodide fluorescence in control and GSPE-pretreated (50 mg/L, 18 h) cultures exposed to H 2O 2 (200 ␮M for 1 h). The lower panel represents the LDH release into the medium after H 2O 2 application (400 ␮M for 6 h) to the basal LDH level (as 100%) in control or GSPE-pretreated glial cultures. Significance levels were calculated by using ANOVA single factor analysis, and P ⬍ 0.05 is designated by an asterisk (*).

demonstrate that the decline in GSH levels after LPS/IFN-␥ treatment was significantly diminished by GSPE pretreatment. Furthermore, there was no significant change in the basal GSH level in glial

cells with GSPE treatment alone. The present study shows that GSPE treatment increased the NO production in glial cells but did not impose any nitrosative stress; rather it protects the cells from LPS/

FIG. 6. (A) Concentration-dependent increase in the DCF signal in rat primary glial cultures was observed with different doses of H 2O 2 (0 –200 ␮M). After 3 min of baseline fluorescence collection, glial cultures were treated with the indicated concentrations of H 2O 2, and fluorescence values were collected for a further period of 7 min at intervals of 1 min. Data represent the mean ⫾ SEM. (B) GSPEpretreated (50 mg/L) glial cultures showed higher tolerance toward H 2O 2 (100 ␮M) than the controls. H 2O 2 was applied to control and GSPE (50 mg/L, 18 h) pretreated cultures. After 3 min of baseline fluorescence collection, glial cultures were treated with the indicated concentrations of H 2O 2, and fluorescence values were collected for a further period of 7 min at intervals of 1 min. Data represent the mean ⫾ SEM. (C) GSPE-pretreated (50 mg/L) glial cultures showed higher tolerance toward t-BH (100 ␮M) application than the controls. t-BH (100 ␮M) was applied to control and GSPE (50 mg/L, 18 h) cultures. After 3 min of baseline fluorescence collection, glial cultures were treated with the indicated concentration of t-BH, and fluorescence values were collected for a further period of 7 min at intervals of 1 min. Data represent the mean ⫾ SEM. Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

148

ROYCHOWDHURY ET AL.

IFN-␥-induced nitrosative stress by protecting the GSH pool, although the exact mechanism is yet unclear. Proanthocyanidins are reported to have a strong H 2O 2-detoxifying ability in macrophage J774A-1 cells and neuroactive adrenal pheochromocytoma PC12 cells (29). Therefore we monitored the effect of GSPE on peroxide-induced cytotoxicity (H 2O 2 and t-BH) in a rat glial culture model by measuring the ROS-sensitive DCF fluorescence during H 2O 2 and t-BH treatment in both control and GSPEpretreated cultures. GSPE-pretreated glial cultures were more tolerant toward H 2O 2 exposure than the controls were. We observed that GSPE also provides protection against t-BH, but the extent of protection was lower than in the case of H 2O 2. GSPE showed a generalized anti-peroxidative effect, which is effective against H 2O 2 as well as t-BH, with a varying capacity. H 2DCF-DA was reported to have varying sensitivity toward different ROS, e.g., Hoyt et al. (30) found no detectable increase in DCF fluorescence, when neuronal cell cultures were exposed to less than 3 mM H 2O 2. In our experiments, DCF showed a significant increase in the fluorescence signal even at a concentration of 0.05 mM, proving it to be a sensitive tool for monitoring low concentrations of H 2O 2 within the cells. H 2O 2 is well known to cause DNA damage and nuclear fragmentation in neuronal cell cultures (30). We therefore examined the effect of GSPE on cellular viability under H 2O 2-induced cytotoxicity in glial cultures. Our data show that GSPE-pretreated cell cultures were more tolerant to H 2O 2 than the untreated cell cultures determined by both LDH release into the culture medium and propidium iodide staining. Our data indicate that GSPE exerts its cytoprotective effect in rat glial cultures by preserving the cellular GSH pool during exposure to high output NO generation. Furthermore, GSPE pretreatment also shows a cytoprotective effect against H 2O 2induced oxidative stress. Since ROS overproduction is shown to be the cause of various neuropathological conditions, the potent antioxidative effect of GSPE could be utilized as a therapeutic tool under such pathological situations.

ACKNOWLEDGMENTS The technical assistance of L. Bu¨ck and A. Rudloff is gratefully appreciated. This work was supported by DFG Grant WO 474/ 10-3, 11-4, and grant of “Land Sachsen Anhalt” LSA 2997A/ 0088H.

REFERENCES 1. Bolanos, J. P., Almeida, A., Stewart, V., Peuchen, S., Land, J. M., Clark, J. B., and Heals, S. J. R. (1997). Nitric oxidemediated mitochondrial damage in the brain: Mechanisms and implications for neurodegenerative diseases. J. Neurochem. 68, 2227–2240. 2. Sastry, P. S., and Subba Rao, K. (2000). Apoptosis and the nervous system. J. Neurochem. 74, 1–20. 3. Beal, M. F. (1995). Aging, energy and oxidative stress in neurodegenerative diseases. Ann. Neurol. 38, 357–366. 4. Gonzalez, J. G., delle-Monache, G., delle-Monache, F., and Marini-Bettolo, G. B. (1982). Chuchuhuasha-a drug used in folk medicine in the amzonian and andean areas: A chemical study of Maytenus leavis. J. Ethnopharmacol. 5, 73–77. 5. Hanefeld, M., and Herrmann, K. (1976). On the occurence of proanthocyanidins, leucoanthocyanidins and catechins in vegetables. Z. Lebensm. Unters. Forsch. 161, 243–248. 6. Masquelier, J., Michaud, J., Laparra, J., and Dumon, M. C. (1979). Flavonoids and pycnogenols. Int. J. Vitam. Nutr. Res. 49, 307–311. 7. Rice Evans, C. A., Miller, N. J., and Paganda, G. (1996). Structure antioxidant activity relationships of flavonoids and phenolic acids. Free Radical Biol. Med. 20, 933–956. 8. Bagchi, D., Garg, A., Krohn, R. L., Bagchi, M., Tran, M. X., and Stohs, S. J. (1997). Oxygen free radical scavenging abilities of vitamin C and E and a grape seed proanthocyanidin extract in vitro. Res. Commun. Mol. Pathol. Pharmacol. 95, 179 –189. 9. Bagchi, D., Garg, A., Krohn, R. L., Bagchi, M., Bagchi, D. J., Balmoori, J., and Stohs, S. J. (1998). Protective effects of grape seed proanthocyanidins and selected antioxidants against TPA induced hepatic and brain lipid peroxidation and DNA fragmentation, and peritoneal macrophage activation in mice. Gen. Pharmacol. 30, 771–776. 10. Kovac, V., Alonso, E., and Revilla E. (1995). The effect of adding supplementary quantities of seeds during fermentation on the phenolic composition of wines. Am. J. Enol. Vitic. 46, 363–367. 11. Possel, H., Noack, H., Putzke, J., Wolf, G., and Sies, H. (2000). Selective upregulation of inducible nitric oxide synthase (iNOS) by lipopolysaccharide (LPS) and cytokines in microglia: in vitro and in vivo studies. Glia 32, 51–59. 12. Chatterjee, S., Noack, H., Possel, H., and Wolf, G. (2000). Induction of nitric oxide synthesis lowers intracellular glutathione in primary glial cultures. Glia 29, 98 –101.

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

PROTECTION OF PRIMARY GLIAL CELLS BY GSPE

149

13. Dringen, R. (2000). Metabolism and functions of glutathione in brain. Prog. Neurobiol. 62, 649 – 671.

cent marker for peroxynitrite formation. FEBS Lett. 416, 175–178.

14. Dalkara, T., Yoshida, T., Irikura, K., and Moskowitz, M. A. (1994). Dual role of nitric oxide in focal cerebral ischemia. Neuropharmacology 33, 1447–1452.

23. Chatterjee, S., Noack, H., Possel, H., Keilhoff, G., and Wolf, G. (1999). Glutathione levels in primary glial cultures: Monochlorobimane provides evidence of cell type specific distribution. Glia 27, 152–161.

15. Szabo, C. (1996). Physiological and pathophysiological roles of nitric oxide in the central nervous system. Brain Res. Bull. 41, 131–141. 16. Crow, J. P., and Beckman, J. S. (1995). The role of peroxynitrite in nitric oxide mediated toxicity. Curr. Top. Microbiol. Immunol. 196, 61–73. 17. Torreilles, F., Salman-Tabchech, S., Guerin, M. C., and Torreilles, J. (1999). Neurodegenrative disorders: The role of peroxynitrite. Brain Res. Rev. 30, 153–163. 18. Licinio, J., Prolo, P., McCann, S. M., and Wng, M. L. (1999). Brain iNOS: Current understanding and clinical implications. Mol. Med. Today 5, 225–232. 19. Noack, H., Possel, H., Rethfeldt, C., Keilhoff, G., and Wolf, G. (1999). Peroxynitrite mediated damage and lowered superoxide tolerance in primary cortical glial cultures after induction of the inducible isoform of NOS. Glia 28, 13–24. 20. Schousboe, A. (1980). Primary cultures of astrocytes from mammalian brain as a tool in neurochemical research. Cell Mol. Biol. 26, 505–513. 21. Kelly, E. E., Wagner, B. A., Buettner, G. R., and Burns, C. P. (1999). Nitric oxide inhibits iron induced lipid peroxidation in HL-60 cells. Arch. Biochem. Biophys. 370, 97–104. 22. Possel, H., Noack, H., Augustin, W., Keilhoff, H., and Wolf, G. (1997). 2,7-Dihydrodichlorofluorescein diacetate as a fluores-

24. Wink, D., Vodovotz, Y., Grisham, M. B., DeGraff, W., Cook, J. C., Pacelli, R., Krishna, M., and Mitchell, J. B. (1999). Antioxidant effects of nitric oxide. Methods Enzymol. 301, 413– 424. 25. Wink, D. A., Hanbauer, I., and Laval, F. (1994). Nitric oxide protects against the cytotoxic effects of reactive oxygen species. Ann. N.Y. Acad. Sci. 738, 265–278. 26. Brune, B., Knethen, A. V., and Sandau, K. B. (1999). Nitric oxide (NO): An effector of apoptosis. Cell Death Differ. 6, 969 –975. 27. Ducrocq, C., Blanchard, B., Pignatelli, B., and Ohshima, H. (1999). Peroxynitrite: An endogenous oxidizing and nitrating agent. Cell. Mol. Life. Sci. 55, 1068 –1077. 28. Chieuh, C. C., and Rauhala, P. (1999). The redox pathway of S-nitrosoglutathione, glutathione to nitric oxide in cell to neuron communications. Free Radical Res. 31, 641– 650. 29. Bagchi, D., Kuszynski, C., Balmoori, J., Bagchi, M., and Stohs, S. J. (1998). Hydrogen peroxide-induced modulation of intracellular oxidized states in cultured macrophage J774A.1 and neuroactive PC-12 cells, and protection by a novel grape seed proanthocyanidin extract. Phytother. Res. 12, 568 –571. 30. Hoyt, K. R., Gallagher, A. J., Hastings, T. G., and Reynolds, I. J. (1997). Characterization of hydrogen peroxide toxicity in cultured rat forebrain neurons. Neurochem. Res. 22, 333–340.

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.