European Journal of Neuroscience, Vol. 16, pp. 1275±1283, 2002
ã Federation of European Neuroscience Societies
Astrocytes enhance lipopolysaccharide-induced nitric oxide production by microglial cells Carme SolaÁ,1 Carme Casal,1 Josep M. Tusell2 and Joan Serratosa1 1
Department of Pharmacology and Toxicology, Institut d'Investigacions BiomeÁdiques de Barcelona-CSIC, IDIBAPS, Barcelona, Spain 2 Department of Neurochemistry, Institut d'Investigacions BiomeÁdiques de Barcelona-CSIC, IDIBAPS, Barcelona, Spain Keywords: glial interaction, immunocytochemistry, inducible nitric oxide synthase, rat glial cell cultures
Abstract Several stimuli result in glial activation and induce nitric oxide (NO) production in microglial and astroglial cells. The bacterial endotoxin lipopolysaccharide (LPS) has been widely used to achieve glial activation in vitro, and several studies show that both microglial and, to a lesser extent, astroglial cell cultures produce NO after LPS treatment. However, NO production in endotoxintreated astrocyte cultures is controversial. We characterized NO production in microglial, astroglial and mixed glial cell cultures treated with lipopolysaccharide, measured as nitrite accumulation in the culture media. We also identi®ed the NO-producing cells by immunocytochemistry, using speci®c markers for the inducible NO synthase (iNOS) isoform, microglial and astroglial cells. Only microglial cells showed iNOS immunoreactivity. Thus, contaminating microglial cells were responsible for NO production in the secondary astrocyte cultures. We then analysed the effect of astrocytes on NO production by microglial cells using microglial±astroglial cocultures, and we observed that this production was clearly enhanced in the presence of astroglial cells. Soluble factors released by astrocytes did not appear to be directly responsible for such an effect, whereas nonsoluble factors present in the cell membrane of LPS-treated astrocytes could account, at least in part, for this enhancement.
Introduction Glial cells respond to neuronal damage with morphological and functional changes, which is known as glial activation. Reactive astroglial and microglial cells have been detected in experimental models of neuronal injury such as ischaemia (Rischke & Krieglstein, 1991; Jorgensen et al., 1993), axotomy (Graeber & Kreutzberg, 1986; Graeber et al., 1988; Barron et al., 1990), electrical stimulation (LeeHall et al., 1989) and neurotoxic insult (Jorgensen et al., 1993; van den Berg & Gramsbergen, 1993; Akiyama et al., 1994; AcarõÂn et al., 1996). These cells have also been observed in the human brain in neurodegenerative diseases (Haga et al., 1989; Delacourte, 1990; Graeber & Streit, 1990). Reactive glial cells can produce a variety of growth factors, cytokines, nitric oxide (NO) and neuropeptides, that are either involved in neuroprotection or in neurodegeneration (Chao et al., 1995; Raivich et al., 1999; Streit et al., 1999; Vitkovic et al., 2000). Given the complexity of glial activation studies in vivo, this phenomenon has been frequently characterized using in vitro approaches. The bacterial endotoxin lipopolysaccharide (LPS) has been widely used to establish an experimental model of glial activation in vitro. Enriched rat microglial cell cultures respond to LPS with morphological changes and an increased production of several cytokines and NO as a result of the upregulation of the inducible nitric oxide synthase isoform (iNOS; Zielasek et al., 1992; Fiebich et al., 1998; Casal et al., 2001). Lipopolysaccharide has also been used to activate enriched rat astroglial cell cultures. However, the response of these cells to LPS differs from that of microglial cells: both produce cytokines, but there is a controversy regarding NO
production by cultured astrocytes. Several authors report the production of NO in rodent astrocyte cultures stimulated with LPS, mostly through the detection of its decomposition product nitrite in the culture supernatant (Galea et al., 1992; Simmons & Murphy, 1992; Park & Murphy, 1994; Kong et al., 1996; Pahan et al., 1997; (Chen et al., 1998; GarcõÂa-Nogales et al., 1999; Suk et al., 2001), whereas other authors do not detect nitrite production in LPS-treated astrocyte cultures (Boje & Arora, 1992; Chao et al., 1992; Hewett et al., 1993; Yang et al., 1998). However, only a few studies have determined the cellular localization of NO production by means of iNOS immunocytochemistry or NADPH diaphorase staining, revealing that either microglial cells and some astrocytes (Galea et al., 1992; Kong et al., 1996) or only microglial cells (Yang et al., 1998) are responsible for NO production in LPS-treated astrocyte cultures. In addition, some authors claim that only microglial cells are responsible for NO production in LPS-treated mixed glia or astroglia±microglia cocultures (Vincent et al., 1996, 1997; Yang et al., 1998; Possel et al., 2000). Here, we study the NO production and identify the cell type responsible for the expression of iNOS in LPS-stimulated astroglial, microglial and mixed glial cell cultures. We also determine the effect of astrocytes on NO production by microglia, using astroglia± microglia cocultures where microglial cells are seeded in the presence of different densities of astroglial cells.
Correspondence: Dr Carme SolaÁ, as above. Email:
[email protected]
Materials
Received 13 May 2002, revised 5 July 2002, accepted 30 July 2002 doi:10.1046/j.1460-9568.2002.02199.x
Materials and Methods Reagents were from Sigma Chemicals Co. (St Louis, MO, USA) unless otherwise stated.
1276 C. SolaÁ et al. Glial cell cultures Animal care procedures followed the Spanish Legislation on Protection of Animals Used for Experimental and Other Scienti®c Purposes, in agreement with EC regulations (OJ. of EC No. L 358/1, 18/12/1986). Primary mixed glial cell cultures were prepared from brains of newborn Wistar rats (Iffa Credo, Lyon, France) on postnatal day 1 or day 2 as described by Giulian & Baker (1986) and modi®ed by Vincent et al. (1996). Brie¯y, animals were killed by decapitation, the cerebral cortex was removed aseptically and the meninges were carefully removed. The tissue was dissociated in 0.25% trypsin diluted 1 : 10 in phosphate-buffered saline (PBS). Brain cells were seeded at a density of 1 3 106 cells/mL in 75 cm2 culture ¯asks (TRP, Switzerland) or at 50 3 103 cells/mL in four-well Permanox Laboratory-Tek chamber slides (Nalge Nunc International, Naperville, IL, USA) in Dulbecco's modi®ed Eagle's medium with Ham's nutrient mixtures F12 (1 : 1) (DMEM/F12, Gibco-BRL, Life Technologies, Paisley, UK) supplemented with 2 mM L-glutamine, 33 mM glucose, 3 mM NaHCO3, 10 mM Hepes, penicillin±streptomycin (10 000 IU/mL±10 000 UG/mL; Gibco-BRL) and 10% heatinactivated foetal bovine serum (FBS) (Gibco-BRL). The medium was changed 1, 6 and 11 days after plating. Microglial and astrocyte cultures were prepared from the mixed glial cell cultures as follows. To obtain microglial cell cultures, the culture ¯asks were shaken at 200 r.p.m. for 18 h at 37 °C on day 12 of the mixed glia cultures. The medium containing detached microglial cells was collected and centrifuged at 200 g for 10 min. The cells were resuspended, counted and plated at various densities in 24-well plates or in four-well Permanox Laboratory-Tek chamber slides. Secondary astroglial cell cultures were obtained by removing the cells attached in the culture ¯asks with 0.25% trypsin (GibcoBRL)/0.02% EDTA. The cells were centrifuged at 200 g for 5 min, resuspended in PBS, centrifuged again at 200 g for 5 min, resuspended with the culture medium and plated at various densities in 24-well plates or in four-well Permanox Laboratory-Tek chamber slides. Microglial±astroglial cell cocultures were obtained by seeding the same amount of microglial cells (50 3 103/well) in the presence of various densities of astroglial cells (from 5 3 103 to 150 3 103 cells/ well) in 24-well culture plates. Twenty-four hours after seeding, microglial and astroglial cell cultures and astroglia/microglia cocultures were treated with 10 mg/mL LPS (Escherichia coli 026:B6, reconstituted in PBS and kept at ±20 °C) for 48 h. Fourteen-day-old mixed glial cell cultures in chamber slides were also treated with 10 mg/mL LPS for 48 h. Nitrite assay Nitric oxide production in the glial cell cultures was assessed by the Griess reaction, a colorimetric assay that detects a stable reaction product of NO and molecular oxygen, nitrite (NO2±), in culture supernatants. Brie¯y, after stimulation with LPS for 48 h, 100 mL aliquots of culture supernatants were incubated with 100 mL Griess reagent (1% sulphanilamide, 0.1% N-(1-naphthyl)ethylenediamine dihydrochloride and 5% phosphoric acid) at room temperature for 10 min. Optical density at 540 nm was then determined using a microplate reader (iEMS Reader MF, Labsystems, Finland). The nitrite concentration was determined from a sodium nitrite standard curve. Immunocytochemistry Glial cells were ®xed with chilled 100% methanol for 7 min at 48 h after treatment with LPS. When single immunocytochemistry was
performed, cells were incubated overnight at 4 °C with a rabbit antiglial ®brillary acidic protein (GFAP) polyclonal antibody (1 : 500; DAKO, Glostrup, Denmark) or with the monoclonal anti-CD11b (MRC OX-42) antibody (1 : 50; Serotec Ltd. Oxford, UK) to identify astroglial and microglial cells, respectively. Immunolabelling of iNOS was performed either with an anti-iNOS monoclonal antibody (1 : 1000; Transduction Laboratories, Lexington, KY, USA) or with an anti-iNOS polyclonal antibody (1 : 200; Chemicon International, Inc., Temecula, CA, USA). When double immunocytochemistry was performed, cells were coincubated with rabbit anti-GFAP polyclonal antibody (1 : 800) and anti-iNOS monoclonal antibody (1 : 1000), or OX-42 monoclonal antibody (1 : 50) and anti-iNOS polyclonal antibody (1 : 200). After rinsing in PBS, the ¯uorescent secondary antibody goat anti-rabbit tetramethylrhodamine isothiocyanate (TRITC) 1 : 400 in PBS was used for the polyclonal antibodies and the ¯uorescent secondary antibody sheep anti-mouse ¯uorescein isothiocyanate (FITC) 1 : 50 in PBS was used for the monoclonal antibodies, for 1 h at room temperature. Immunoreactivity was visualized under a ¯uorescence microscope (Zeiss Axioplan, Germany). Preparation of astrocyte-conditioned medium To determine the effect of the diffusible factors released by astrocytes on NO production by microglial cells, the medium obtained from the secondary astrocyte culture in the 24-well plates (150 3 103 cells/ well), either from cells treated with LPS for 48 h or from their corresponding controls, was used as astrocyte-conditioned medium. This was stored at ±20 °C and centrifuged at 200 g for 5 min and then added to the microglial cell cultures. Preparation of culture plates with ®xed astrocytes or with astrocytic extracellular matrix To determine the effect of nondiffusible factors released by astrocytes on NO production by microglial cells, the microglial cells were seeded on top of a con¯uent layer of ®xed astrocytes or astrocytic extracellular matrix as follows (adapted from Tanaka & Maeda, 1996; Tanaka et al., 1998). Astrocytes were seeded in 24-well plates at a density of 150 3 103 cells/well and treated 24 h later with LPS for 48 h. To prepare ®xed astrocytes, the cells were ®xed with 4% paraformaldehyde solution in phosphate buffer (PB) for 20 min at room temperature, rinsed with PBS and stored at 4 °C for 3 days in PBS. They were incubated in serum-free DMEM/F12 at 4 °C for 16 h before microglial cells were seeded. In another set of experiments, astrocytes were ®xed with chilled 100% methanol for 7 min, rinsed with PBS and stored at 4 °C for 3 days in PBS. They were incubated in serum-free DMEM/F12 at 37 °C for 2 h before microglial cells were seeded. To prepare astrocytic extracellular matrix-coated 24well plates, astrocyte layers were lysed with sterile water for 15 min at room temperature, the water was eliminated by gently pipetting and the plates were stored at ±80 °C for 4 days. They were incubated in serum-free DMEM/F12 at 37 °C for 2 h before microglial cells were seeded. Statistical analysis Results are expressed as the mean 6 SEM from 3±7 independent cell cultures. Statistical analysis of nitrite production in LPS-treated cells vs. their respective controls was performed using a Student's t-test. When nitrite production was compared between more than two groups, we applied two-way analysis of variance (ANOVA) followed by Duncan's test.
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
Astrocytes enhance microglial activation 1277
Results Nitrite production in astroglial, microglial and mixed glial cell cultures after LPS treatment In preliminary experiments, astroglial cells cultured in four-well Permanox Laboratory-Tek chamber slides at a density 200 3 105 cells/well were treated with various concentrations of LPS (0.1± 10 mg/mL) for 24 h. We only detected a certain increase in NO production, measured as nitrite accumulation in the culture medium, when the astrocytes were treated with 10 mg/mL LPS (data not shown). For that reason, subsequent experiments were performed using 10 mg/mL LPS and increasing the exposure time to 48 h. The concentrations of LPS tested did not result in relevant cell death as observed using propidium iodide staining (data not shown). Astroglial cell cultures treated with 10 mg/mL LPS for 48 h showed an increase in nitrite accumulation in the culture medium, which was proportional to the number of cells (Fig. 1A). The minimal number of astroglial cells required to detect NO formation and release into the supernatant was 100 3 105 cells/well. In comparison, microglial cells also treated with 10 mg/mL LPS for 48 h already showed an increase in nitrite accumulation in the culture medium at 25 3 105 cells/well (Fig. 1B). For a given number of cells, nitrite production in astroglial cell cultures was much lower than that observed in microglial cell cultures. Mixed glial cell cultures treated with LPS for 48 h showed a clear increase in NO production (Fig. 1C). iNOS immunoreactive cells in astroglial, microglial and mixed glial cell cultures treated with LPS In the astroglial cell cultures, GFAP immunostaining was detected in cells with the characteristic astroglial cell morphology (Fig. 2A). No iNOS-positive cells were observed in the control astrocyte cultures and only a few cells showed iNOS immunoreactivity 48 h after the addition of LPS. When double immunocytochemistry was performed, iNOS immunoreactivity did not colocalize with the staining of the GFAP antibody (Fig. 2B), indicating that astrocytes were not the cellular source of iNOS protein in these cultures. Most iNOS positive cells were round. When OX-42 antibody, which recognizes a macrophage/microglia-speci®c antigen that is not expressed by astrocytes or oligodendrocytes, was used in these astroglial cell cultures, some immunoreactive cells, corresponding to contaminant microglial cells, were detected (Fig. 2C). In addition, most OX-42 immunoreactive cells presented iNOS immunoreactivity (Fig. 2D). Although some OX-42-positive/iNOS-negative cells were detected after LPS treatment, all iNOS-positive cells were also OX-42-positive. Most of the cells that were double labelled were round, but OX-42- and iNOS-positive cells with rami®cations were also observed. Microglial cells in pure culture, labelled with OX-42 antibody, differentiated into rami®ed cells with time in culture (Fig. 2E). No iNOS immunoreactivity was detected in the control microglial cell cultures, pointing to the lack of NO production in microglial cells in the absence of stimulation. However, most of the cells were round and iNOS-immunopositive 48 h after LPS treatment (Fig. 2F). In mixed glial cell cultures, microglial cells showed larger cell bodies and fewer rami®cations than in microglial cell cultures (Fig. 3A). In general, they acquired a rounded morphology when treated with LPS (Fig. 3B). The morphology of astroglial cells was similar in mixed glial cell cultures and in astroglial cell cultures. iNOS immunoreactivity was not detected in GFAP-positive cells (Fig. 3C) but in OX-42-positive cells (Fig. 3D).
FIG. 1. Nitrite production in glial cell cultures after LPS treatment. Comparison of LPS-induced nitrite production by various amounts of astroglial (A) or microglial (B) cells. (C) LPS-induced nitrite production in mixed glial cell cultures. Bars represent the means 6 SEM of seven, three and three independent astroglial, microglial and mixed glial cell cultures, respectively. Nitrite production in LPS-treated cells differed from that of the respective controls at *P < 0.05, **P < 0.01, Student's t-test.
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
1278 C. SolaÁ et al.
FIG. 3. Identi®cation of NO-producing cells in mixed glial cell cultures. GFAP (red) and OX-42 (green) immunolabelling in a control culture (A) and LPS-treated culture (B). After LPS-treatment (C), iNOS (green) immunoreactivity is not detected in GFAP (red)-positive cells (D) but iNOS (red) colocalizes with OX-42 (green)-positive cells. Scale bar, 50 mm.
FIG. 2. Identi®cation of NO-producing cells in secondary astrocyte and primary microglial cell cultures treated with LPS. (A) GFAP immunolabelling in a control secondary astrocyte culture. (B) GFAP (red) and iNOS (green) double immunocytochemistry in a LPS-treated astroglial cell culture. The two immunolabelling signals do not colocalize, showing that iNOS-positive cells do not correspond to astrocytes. (C) OX-42 immunoreactive cells in a control astrocyte culture, showing the presence of a few contaminating microglial cells. (D) OX-42 (green) and iNOS (red) double immunocytochemistry in a LPS-treated astroglial cell culture, where the two antibodies are colocalized, showing that microglial cells are responsible for iNOS induction in the secondary astroglial cell cultures. (E) OX-42 immunolabelling in a control microglial cell culture, showing rami®ed microglial cells in a resting state. (F) OX-42 (green) and iNOS (red) double immunocytochemistry in a LPS-treated microglial cell culture. Scale bar, 50 mm.
Stimulation of LPS-induced NO production in microglia in the presence of astrocytes A concentration of 50 3 103 microglial cells/well was established as that required to examine nitrite production in astroglia/microglia cocultures in subsequent experiments. When these microglial cells were cocultured with increasing densities of astroglial cells (5± 150 3 103 cells/well), nitrite production after 48 h of LPS addition increased with the number of astroglial cells, and the nitrite concentration in the cocultures (Fig. 4; measured nitrite) was higher than the sum of nitrite produced by independently cultured microglial cells and astrocytes (Fig. 4; estimated nitrite). However, only the microglial cells in these cocultures showed iNOS immunoreactivity (Fig. 5). In¯uence of astrocyte diffusible and nondiffusible factors on nitrite production by LPS-treated microglia Microglial cells were cultured in astrocyte-conditioned medium to determine the effect of soluble factors released by astrocytes on LPSinduced nitrite production in microglial cells. However, nitrite production in microglial cells cultured in the conditioned medium from either control or LPS-treated astrocytes was equivalent to that
FIG. 4. Stimulation of nitrite production in microglia±astrocyte cocultures. Comparison between LPS-induced nitrite production determined in microglial cells cocultured with various amounts of astrocytes (measured nitrite) and the sum of nitrite produced by independently cultured microglial cells and various amounts of astrocytes after LPS treatment (estimated nitrite). Bars represent the means 6 SEM of three independent cultures. Measured nitrite was signi®cantly different from estimated nitrite at *P < 0.05, **P < 0.01, ***P < 0.001; Student's t-test.
obtained when the cells were cultured in DMEM/F12 medium (Fig. 6). To test whether nondiffusible factors present in the astrocyte cell membrane enhanced nitrite production in LPS-treated microglial cells, these were seeded on top of a con¯uent layer of ®xed astrocytes. In these conditions, LPS-induced nitrite production in microglial cells cultured on top of ®xed control astrocytes was the same as that from the same number of microglial cells cultured alone (Fig. 7). Thus, the presence of astrocytes per se did not appear to modify nitrite production by microglial cells. However, when
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
Astrocytes enhance microglial activation 1279
FIG. 5. Identi®cation of NO-producing cells in microglial±astroglial cell cocultures. GFAP (A and B) and OX-42 (C and D) immunocytochemistry in control (A and C) and LPS-treated (B and D) cocultures. In LPS-treated cocultures (E) iNOS (green) does not colocalize with GFAP (red) immunolabelling but (F) iNOS (red) colocalizes with OX-42 (green) immunolabelling. These images correspond to cocultures of 50 3 103 microglial cells with 150 3 103 astroglial cells/well. Scale bar, 50 mm.
FIG. 7. Effect of factors present in astrocyte cellular membrane on LPSinduced nitrite production by microglial cells. Nitrite production by microglial cells cultured alone (NM) or seeded onto a con¯uent layer of ®xed control (FCA) or ®xed LPS-treated astrocytes (FTA). Astrocytes were ®xed with paraformaldehyde (A) or methanol (B). Bars represent the means 6 SEM of seven (paraformaldehyde-®xed astrocytes) or four (methanol-®xed astrocytes) independent cultures. Nitrite production in LPStreated microglial cells seeded on top of ®xed astrocytes differed signi®cantly from nitrite production in LPS-treated microglial cells cultured alone at *P < 0.05, **P < 0.01 (ANOVA, Duncan's test). Nitrite production in LPS-treated cells vs. the respective controls increased signi®cantly in all experimental groups (P < 0.05, Student's t-test).
FIG. 6. Effect of astrocyte-conditioned media on nitrite production in LPStreated microglial cell cultures. Nitrite production by microglial cells cultured with normal medium (DMEM/F12) (NM), or with astrocyteconditioned media obtained from control (CMCA) or LPS-treated astrocyte cultures (CMTA). Bars represent means 6 SEM of four independent cultures. No signi®cant differences were detected between nitrite production by LPS-microglial cells in the three groups (ANOVA, Duncan's test). Nitrite production in LPS-treated cells vs. the respective controls signi®cantly increased in the NM and the CMTA groups (P < 0.05, Student's t-test).
microglial cells were cultured on top of ®xed LPS-treated astrocytes, LPS-induced nitrite production was signi®cantly higher than that found when the same number of microglial cells was cultured alone. The same effect was observed when astrocytes were ®xed with paraformaldehyde (Fig. 7A) or with methanol (Fig. 7B). Therefore, nitrite production in LPS-treated microglial cells was enhanced by some factor present in the cell membrane of LPS-treated astrocytes. Nevertheless, the amount of NO produced was lower than that observed when microglial cells were cocultured with live astrocytes. To test whether nondiffusible factors present in the astrocyte extracellular matrix enhance nitrite production in LPS-treated
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
1280 C. SolaÁ et al.
FIG. 8. Effect of factors present in astrocyte extracellular matrix on LPSinduced nitrite production by microglial cells. Nitrite production by microglial cells cultured alone (NM), or seeded onto extracellular matrix of control (EMCA) or LPS-treated astrocytes (EMTA). Bars represent means 6 SEM of ®ve independent cultures. Nitrite production by LPSmicroglial cells in the three groups did not differ signi®cantly (ANOVA, Duncan's test). Nitrite production in LPS-treated cells vs. the respective controls increased signi®cantly in all experimental groups (P < 0.05, Student's t-test).
microglial cells, microglial cells were seeded on top of a layer of astrocyte extracellular matrix. The LPS-induced nitrite production by microglial cells cultured on top of extracellular matrix of control astrocytes did not differ from nitrite production by the same number of microglial cells cultured alone (Fig. 8). The same result was obtained when microglial cells were cultured on top of extracellular matrix of LPS-treated astrocytes (Fig. 8). Therefore, the astrocytic extracellular matrix was not involved in nitrite production by microglial cells.
Discussion The present results show that NO production, measured as nitrite accumulation in the culture medium, is induced in secondary astrocyte cell cultures treated with LPS. However, microglial cells that are usually identi®ed as contaminants in these cultures are responsible for such a production, as determined by iNOS immunocytochemistry. In addition, NO production by these contaminant microglial cells is higher than expected because it is strongly enhanced in the presence of astrocytes. We suggest that direct contact between microglia and astroglial cells is necessary for such potentiation and that a nonsoluble factor present in the astrocyte cellular membrane is involved. There is some controversy regarding NO production in LPSstimulated astrocyte cell cultures. When NO production in LPStreated astrocyte cell cultures is determined by the presence of nitrite in the culture supernatant, the contaminant microglial cells present in these cultures, which are dif®cult to eliminate, could contribute partially to this production (either directly or indirectly) through the release of some activator of NO production in astrocytes (Galea et al., 1992; Simmons & Murphy, 1992; Feinstein et al., 1994; Park & Murphy, 1994; Suk et al., 2001). However, the number of contaminant microglial cells in astroglial cell cultures is too low to
account for the extent of nitrite production detected (Simmons & Murphy, 1992). Nevertheless, our results suggest that relatively few microglial cells are responsible for a higher production of NO than expected if they are in contact with astroglial cells. Several studies show that the only cells with iNOS expression in rodent glial cell cultures after LPS-stimulation are microglial cells (Boje & Arora, 1992; Vincent et al., 1996, 1997; Yang et al., 1998; Possel et al., 2000), whereas some authors have detected GFAP-positive cells with NADPH diaphorase staining or iNOS immunolabelling after the addition of LPS to rat (Galea et al., 1992) or mouse (Kong et al., 1996) astrocyte cell cultures, respectively. The NO-producing astrocytes that have been identi®ed represent a small population of astrocytes. Galea et al. (1992) detected nitrite accumulation in the medium of primary astrocyte cultures after LPS treatment, but none was observed in the medium of LPS-treated secondary astrocyte cultures. They suggest that factors produced by microglial cells, which are diminished after the passaging procedure normally resulting in depletion of the microglia population, are responsible for nitrite production in the astrocytes. However, we did not detect iNOS immunoreactivity in astrocytes either in primary glial cultures or in secondary astroglial cultures, showing that astroglial cells do not produce NO after LPS treatment even in the presence of factors produced by microglial cells. Differences in culture preparation, stimulus protocol and the presence of inhibitory factors may explain the discrepancies observed. A similar controversy has arisen with reference to oligodendrocyte cultures (Merrill et al., 1993, 1997; Hewett et al., 1999). The morphology of microglial cells in culture varies if the cells are cultured alone or in contact with astrocytes (Tanaka & Maeda, 1996; Tanaka et al., 1999). Thus, microglial cells could also suffer functional changes when they are in contact with astrocytes, modifying the expression of certain molecules. Different observations have been published as regards the possible in¯uence of astrocytes on NO production in LPS-treated microglial cells. On the one hand, Vincent et al. (1996, 1997) observed that 4-day-old noncon¯uent mixed glial cell cultures showed more iNOS-positive microglial cells than 6-day-old noncon¯uent or 8-day-old con¯uent mixed glial cell cultures after adding LPS, suggesting that in this mixed glial cell culture LPS-induced NO production by microglial cells is inhibited when the number of astrocytes increases. However, nitrite production in the LPS-stimulated cultures was higher in the 6- and 8-day-old cultures than in the 4-day-old culture, indicating that although the number of iNOS-positive cells decreases, NO production increases. Our microglial and astrocyte cell cultures were obtained from con¯uent 12-day-old mixed glial cell cultures. After addition of LPS to mixed glial, microglial or astroglial cell cultures, we observed that all iNOS immunoreactive cells corresponded to microglial OX-42positive cells, although not all OX-42-positive cells were iNOSpositive, as in the astroglia±microglia cocultures. Thus, NO production in microglial cells in the presence of astrocytes was probably enhanced because of an increase in NO production by these cells rather than an increase in the number of iNOS-positive microglial cells. On the other hand, Yang et al. (1998) suggest that the serine and glycine released to the culture medium by astrocytes are responsible for nitrite production in microglial cells cocultured with astrocytes and treated with LPS. These authors cultured the cells in serum-free medium at low concentrations of serine and glycine, whereas we used DMEM/F12 medium, which contains a high concentration of serine and glycine (250 mM each). However, we did not detect NO production in microglial cells in our control cultures. In addition, conditioned medium from astrocytes did not enhance NO
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
Astrocytes enhance microglial activation 1281 production by LPS-treated microglial cells. Factors other than serine and glycine are therefore responsible for the stimulation observed. Our results indicate that the soluble factors released by astrocytes do not enhance NO production in LPS-treated microglial cells. Nitrite production in LPS-treated microglial cells was similar when these cells were cultured either in DMEM/F12 medium, conditioned medium from control astrocytes or conditioned medium from LPStreated astrocytes. The unusually high levels of nitrite detected in control microglial cells cultured in the conditioned medium of LPStreated astrocytes were probably a result of the LPS present in this medium. The reason why nitrite levels were not even higher in LPStreated microglial cells cultured in the conditioned medium of LPStreated astrocytes, where additional LPS was also present, is probably that 10 mg/mL LPS was inducing maximal nitrite production in the microglial cells. In fact, we carried out the dose-dependent curve of LPS effect on NO production in microglial cells and we observed that a similar increase of nitrite production is observed from 0.5 mg/mL to 10 mg/mL LPS (data not shown). However, we also observed that the maximal nitrite production induced by LPS in microglial cells could be further increased when LPS was added to the cultures together with IFN-g (data not shown). An enhanced NO production in microglial cells treated with the combination of LPS and different compounds (mostly cytokines) has been reported repeatedly in the bibliography (Meda et al., 1995; Kong et al., 1996; Possel et al., 2000). Thus, the maximal nitrite production induced in microglial cells depends on the stimulus considered, and our results also show that it can be further increased in the presence of astrocytes. Consequently, if soluble factors present in the conditioned medium from LPS-treated astrocytes were responsible for the additional increase in NO production in LPS-treated microglial cells we observed when astrocytes were present, we would expect an increase in the nitrite production in LPS-treated microglial cells when culturing these cells in the presence of these soluble factors. As this was not the case, we conclude that astrocytic soluble factors are not involved in the effect here described. Nevertheless, the participation of a soluble factor altered by the freeze±thaw process cannot be discarded. On the contrary, we partially reproduced the effect when we cultured microglial cells on top of ®xed LPS-treated astrocytes. Tanaka & Maeda (1996) show that isolated amoeboid microglial cells ramify in the presence of either living astrocytes or ®xed astrocyte monolayers, an effect rarely observed when they culture the microglial cells alone. The same effect is observed in the presence of astrocytic extracellular matrix (Tanaka & Maeda, 1996; Tanaka et al., 1999). These results show that nondiffusible factors derived from astrocytes are essential for microglial rami®cation. In addition, astroglial±microglial cell interaction could induce changes in the response of glial cells to different stimuli. In fact, our results suggest that a nonsoluble factor present in the cell membrane of astrocytes is responsible, at least in part, for an increased NO production in LPS-treated microglial cells. The extent of the effect observed when microglial cells were cocultured with ®xed astrocytes was lower than that detected using live astrocytes. This could be because of changes occurring in the astrocytes during the process of ®xation, or the fact that factors present in or released by microglial cells also contribute to the changes occurring in LPS-treated astrocytes. We observed morphological changes in microglial cells placed in contact with ®xed astrocytes (data not shown). However, this effect was not suf®cient to explain the increase in NO production in LPS-treated microglia, because we only observed the effect when microglial cells were seeded on top of ®xed LPS-treated astrocytes and then treated with LPS. A possible effect of remaining LPS in the ®xed astrocytes is discarded because ®xed astrocytes were
extensively washed before microglial cells seeding and no increase in NO production was observed in microglial cells seeded on top of ®xed LPS-treated astrocytes if microglial cells were not treated further with LPS. Thus, the changes occurring in surface molecules of astrocytes as a consequence of LPS treatment probably play a key role in NO production by microglial cells. Lipopolysaccharide increases the expression of several cell adhesion molecules, e.g. the neural cell adhesion molecule and the vascular cellular adhesion molecule (VCAM), in cultured astrocytes (Mahler et al., 1997; Pang et al., 2001). In addition, the proin¯amatory cytokines tumour necrosis factor-a, interleukin-1a and IFN-g enhance the expression of the intercellular adhesion molecule (ICAM) and VCAM in cultured astrocytes (Shrikant et al., 1994; Aloisi et al., 1995; HeÂry et al., 1995; Rosenman et al., 1995; Kyrkanides et al., 1999). As these in¯ammatory cytokines are induced in microglial and astroglial cells in culture by LPS (Chung & Benveniste, 1990; Lieberman et al., 1989; Lee et al., 1993; Romero et al., 1996; De Simone et al., 1998; Ledeboer et al., 2000), LPS treatment could also result in an increase in the expression of ICAM and VCAM in these cells. By contrast, astrocytic factors induce rami®cation and downregulate the expression of ICAM and other macrophage surface antigens in microglial cells (Eder et al., 1999; Hailer et al., 2001). Moreover, Merrill et al. (1993) show that treatment of microglial cells with an anti-ICAM antibody increases NO production. There are data to suggest that cell adhesion molecules can behave as signal transducers and that the same molecule can induce quite opposite signals depending on the cell types (Lee & Benveniste, 1999). However, the role of cell adhesion molecules in brain cells and the interaction between alterations in cell adhesion molecules in these cells and the regulation of NO production remains to be elucidated. All these data suggest that cell adhesion molecules are probably altered in our experimental model and that they could play a role in the potentiation of NO production in LPS-treated microglial cells in the presence of astrocytes. Further studies are currently being performed to determine the implication of cell adhesion molecules in this effect.
Acknowledgements The authors thank Sonia AlcaÂzar for technical assistance. C.C. is the recipient of a fellowship from the IDIBAPS (Institut d'Investigacions BiomeÁdiques August Pi i Sunyer). This study was supported by grants SAF 98±0067 and SAF 2001±2240 from CICYT (ComisioÂn de InvestigacioÂn Cientõ®ca y TeÂcnica).
Abbreviations DMEM/F12, Dulbecco's modi®ed Eagle's medium with Ham's nutrient mixtures F12 (1 : 1); FBS, foetal bovine serum; GFAP, glial ®brillary acidic protein; ICAM, intercellular adhesion molecule; iNOS, inducible nitric oxide synthase; LPS, lipopolysaccharide; NO, nitric oxide; PB, phosphate buffer; PBS, phosphate buffered saline; VCAM, vascular cellular adhesion molecule.
References AcarõÂn, L., GonzaÂlez, B., Castellano, B. & Castro, A.J. (1996) Microglial response to N-methyl-D-aspartate-mediated excitotoxicity in the immature rat brain. J. Comp. Neurol., 367, 361±374. Akiyama, H., Tooyama, I., Kondo, H., Ikeda, K., Kimura, H., McGeer, E.G. & McGeer, P.L. (1994) Early response of brain resident microglia to kainic acid-induced hippocampal lesions. Brain Res., 635, 257±268. Aloisi, F., Borsellino, G., CareÂ, A., Testa, U., Gallo, P., Russo, G., Peschle, C. & Levi, G. (1995) Cytokine regulation of astrocyte function: in vitro studies using cells from the human brain. Int. J. Dev. Neurosci., 13, 265±274.
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
1282 C. SolaÁ et al. Barron, K.D., Marciano, F.F., Amundson, R. & Mankes, R. (1990) Perineuronal glial responses after axotomy of central and peripheral axons. A comparison. Brain Res., 523, 219±229. van den Berg, K.J. & Gramsbergen, J.B.P. (1993) Long-term changes in glial ®brillary acidic protein and calcium levels in rat hippocampus after a single systemic dose of kainic acid. Ann. NY Acad. Sci., 679, 394±401. Boje, K.M. & Arora, P.K. (1992) Microglial-produced nitric oxide and reactive nitrogen oxides mediate neuronal cell death. Brain Res., 587, 250± 256. Casal, C., Tusell, J.M. & Serratosa, J. (2001) Role of calmodulin in the differentiation/activation of microglial cells. Brain Res., 902, 101±107. Chao, C.C., Hu, S., Molitor, T.W., Shaskann, E.G. & Peterson, P.K. (1992) Activated microglia mediate neuronal cell injury via a nitric oxide mechanism. J. Immunol., 149, 2736±2741. Chao, C.C., Hu, S. & Peterson, P.K. (1995) Glia, cytokines and neurotoxicity. Crit. Rev. Neurobiol., 9, 189±205. Chen, C.C., Wang, J.K., Chen, W.C. & Lin, S.B. (1998) Protein kinase C h mediates lipopolysaccharide-induced nitric-oxide synthase expression in primary astrocytes. J. Biol. Chem., 272, 19424±19430. Chung, I.Y. & Benveniste, E.N. (1990) Tumor necrosis factor a production by astrocytes. Induction by lipopolysaccharide, IFN-g, and IL-1b. J. Immunol., 144, 2999±3007. De Simone, R., Levi, G. & Aloisi, F. (1998) Interferon gamma gene expression in rat central nervous system glial cells. Cytokine, 10, 418±422. Delacourte, A. (1990) General and dramatic glial reaction in Alzheimer brains. Neurology, 40, 33±37. Eder, C., Schilling, T., Heineman, U., Haas, D., Hailer, N. & Nitsch, R. (1999) Morphological, immunophenotypical and electrophysiological properties of resting microglia in vitro. Eur. J. Neurosci., 11, 4251±4261. Feinstein, D.L., Galea, E., Roberts, S., Berquist, H.W. & Reis, D.J. (1994) Induction of nitric oxide synthase in rat C6 gliana cells. J. Neurochem., 62, 315±321. Fiebich, B.L., Butcher, R.D. & Gebicke-Haerter, P.J. (1998) Protein kinase Cmediated regulation of inducible nitric oxide synthase expression in cultured microglial cells. J. Immunol., 92, 170±178. Galea, E., Feinstein, D.L. & Reis, D.J. (1992) Induction of calciumindependent nitric oxide synthase activity in primary rat glial cultures. Proc. Natl. Acad. Sci. USA, 89, 10945±10949. GarcõÂa-Nogales, P., Almeida, A., FernaÂndez, E., Medina, J.M. & BolanÄos, J. (1999) Induction of glucose-6-phosphate dehydrogenase by lipopolysaccharide contributes to preventing nitric oxide-mediated glutathione depletion in cultures rat astrocytes. J. Neurochem., 72, 1750±1758(1999). Giulian, D. & Baker, T.J. (1986) Characterization of ameboid microglia isolated from developing mammalian brain. J. Neurosci., 6, 2163±2178. Graeber, M.B. & Kreutzberg, G.W. (1986) Astrocytes increase in glial ®brillary acidic protein during retrograde changes of facial motor neurons. J. Neurocytol., 15, 363±373. Graeber, M.B. & Streit, W.J. (1990) Microglia: immune network in the CNS. Brain Pathol., 1, 2±5. Graeber, M.B., Tetzloff, W., Streit, W.J. & Kreutzberg, G.W. (1988) Microglial cells but not astrocytes undergo mitosis following rat facial nerve axotomy. Neurosci. Lett., 85, 317±321. Haga, S., Akai, K. & Ishii, T. (1989) Demonstration of microglial cells in and around senile (neuritic) plaques in the Alzheimer brain. Acta Neuropathol., 77, 569±575. Hailer, N.P., Wirjatijasa, F., Roser, N., Hischebeth, G.T., Korf, H.W. & Dehghani, F. (2001) Astrocytic factors protect neuronal integrity and reduce microglial activation in an in vitro model of N-methyl-D-aspartate-induced excitotoxic injury in organotypic hippocampal slice cultures. Eur. J. Neurosci., 14, 315±326. HeÂry, C., SeÂbire, G., Peudenier, S. & Tardieu, M. (1995) Adhesion to human neurons and astrocytes of monocytes: the role of the interaction of CR3 and ICAM-1 and modulation by cytokines. J. Neuroimmunol., 57, 101±109. Hewett, S.J., Corbett, J.A., McDaniel, M.L. & Choi, D.W. (1993) Interferon-g and interleukin-1b induce nitric oxide formation from primary mouse astrocytes. Neurosci. Lett., 164, 229±232. Hewett, J.A., Hewett, S.J., Winkler, S. & Pfeiffer, S.E. (1999) Inducible nitric oxide synthase expression in cultures enriched for mature oligodendrocytes is due to microglia. J. Neurosci. Res., 56, 189±198. Jorgensen, M.B., Finsen, B.R., Jensen, M.B., Castellano, B., Diemer, N.H. & Zimmer, J. (1993) Microglial and astroglial reactions to ischemic and kainic acid-induced lesions to the adult rat hippocampus. Exp. Neurol., 120, 70±88. Kong, L.Y., McMillian, M.K., Maronpot, R. & Hong, J.S. (1996) Protein tyrosine kinase inhibitors suppress the production of nitric oxide in mixed
glia, microglia-enriched or astrocyte-enriched cultures. Brain Res., 729, 102±109. Kyrkanides, S., Olschowka, J.A., Williams, J.P., Hansen, J.T. & O'Banion, M.K. (1999) TNF a and IL-1b mediate intercellular adhesion molecule-1 induction via microglia±astrocyte interaction in CNS radiation injury. J. Immunol., 95, 95±106. Ledeboer, A., BreveÂ, J.J.P., Poole, S., Tilders, F.J.H. & Van Dam, A.-M. (2000) Interleukin-10, interleukin-4, and transforming growth factor-b differentially regulate lipopolysaccharide-induced production of proin¯ammatory cytokines and nitric oxide in co-cultures of rat astroglial and microglial cells. Glia, 30, 134±142. Lee, S.J. & Benveniste, E.N. (1999) Adhesion molecule expression and regulation on cells of the central nervous system. J. Neuroimmunol., 98, 77± 88. Lee, S.C., Dickson, D.W., Brosnan, C.F. & Berman, J.W. (1993) Cytokine production by human fetal microglia and astrocytes. Differential induction by lipopolysaccharide and IL-1 beta. J. Immunol., 150, 2659±2667. Lee-Hall, L., Borke, R.C. & Anders, J.J. (1989) Transection or electrical stimulation of the hypoglossal nerve increases glial ®brillary acidic protein immunoreactivity in the hypoglossal nucleus. Brain Res., 490, 157±161. Lieberman, A.P., Pitha, P.M., Shin, H.S. & Shin, M.L. (1989) Production of tumor necrosis factor and other cytokines by astrocytes stimulated with lipopolysaccharide or a neurotropic virus. Proc. Natl. Acad. Sci. USA, 86, 6348±6352. Mahler, M., Ben-Ari, Y. & Represa, A. (1997) Differential expression of ®bronectin, tenascin-C and NCAMs in cultured hippocampal astrocytes activated by kainate, bacterial lipopolysaccharide or basic ®broblast growth factor. Brain Res., 775, 63±73. Meda, L., Cassatella, M.A., Szendrei, G.I., Otvos, L. Jr, Baron, P., Villalba, M., Ferrari, D. & Rossi, F. (1995) Activation of microglial cells by bamyloid protein and interferon-g. Nature, 374, 647±650. Merrill, J.E., Ignaro, L.J., Sherman, M.P., Melinek, J. & Lane, T.E. (1993) Microglial cell cytotoxicity of oligodendrocytes is mediated through nitric oxide. J. Immunol., 151, 2132±2141. Merrill, J.E., Murphy, S.P., Mitrovic, B., Mackenzie-Graham, A., Dopp, J.C., Ding, M., Griscavage, J., Ignarro, L.J. & Lowenstein, C.J. (1997) Inducible nitric oxide synthase and nitric oxide production by oligodendrocytes. J. Neurosci. Res., 48, 372±384. Pahan, K., Namboodiri, A.M.S., Sheikh, F.K., Smith, B.T. & Singh, I. (1997) Increasing cAMP attenuates induction of inducible nitric-oxide synthase in rat primary astrocytes. J. Biol. Chem., 272, 7786±7791. Pang, Y., Cai, Z. & Rhodes, P.G. (2001) Analysis of genes differentially expressed in astrocytes stimulated with lipopolysaccharide using cDNA arrays. Brain Res., 914, 15±22. Park, S.K. & Murphy, S. (1994) Duration of expression of inducible nitric oxide synthase in glial cells. J. Neurosci. Res., 39, 405±411. Possel, H., Noack, H., Putzke, J., Wolf, G. & 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. Raivich, G., Bohatschek, M., Kloss, C.U.A., Werner, A., Jones, L.L. & Kreutzberg, G.W. (1999) Neuroglial activation repertoire in the injured brain: graded response, molecular mechanisms and cues to physiological function. Brain Res. Rev., 30, 77±105. Rischke, R. & Krieglstein, J. (1991) Postischemic neuronal damage causes astroglial activation and increase in local cerebral glucose utilization of rat hippocampus. J. Cereb. Blood Flow Metab., 11, 106±113. Romero, L.I., Tatro, J.B., Field, J.A. & Reichlin, S. (1996) Roles of IL-1 and TNF-a in endotoxin-induced activation of nitric oxide synthase in cultured rat brain cells. Am. J. Physiol., 270, R326±R332. Rosenman, S.J., Shrikant, P., Dubb, L., Benveniste, E.N. & Ransohoff, R.M. (1995) Cytokine-induced expression of vascular cell adhesion molecule-1 (VCAM-1) by astrocytes and astrocytoma cell lines. J. Immunol., 154, 1888±1899. Shrikant, P., Chung, I.L., Ballestas, M.E. & Benveniste, E.N. (1994) Regulation of intercellular adhesion molecule-1 gene expression by tumor necrosis factor a, interleukin-1b, and interferon-g in astrocytes. J. Neuroimmunol., 51, 209±220. Simmons, M.L. & Murphy, S. (1992) Induction of nitric oxide synthase in glial cells. J. Neurochem., 59, 897±905. Streit, W.J., Walter, S.A. & Pennell, N.A. (1999) Reactive microgliosis. Prog. Neurobiol., 57, 563±581. Suk, K., Lee, J., Hur, J., Kim, Y.S., Lee, M.S., Cha, S.H., Kim, S.Y. & Kim, H.
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
Astrocytes enhance microglial activation 1283 (2001) Activation-induced cell death of rat astrocytes. Brain Res., 900, 342± 347. Tanaka, J. & Maeda, N. (1996) Microglial rami®cation requires nondiffusible factors derived from astrocytes. Exp. Neurol., 137, 367±375. Tanaka, J., Toku, K., Matsuda, S., Sudo, S., Fujita, H., Sakanaka, M. & Maeda, N. (1998) Induction of resting microglia in culture medium devoid of glycine and serine. Glia, 24, 198±215. Tanaka, J., Toku, K., Sakanaka, M. & Maeda, N. (1999) Morphological differentiation of microglial cells in culture: involvement of insoluble factors derived from astrocytes. Neurosci. Res., 34, 207±215. Vincent, V.A.M., Tilders, F.J.H. & Van Dam, A.-M. (1997) Inhibition of endotoxin-induced nitric oxide synthase production in microglial cells by the presence of astroglial cells: a role for transforming growth factor b. Glia, 19, 190±198.
Vincent, V.A.M., Van Dam, A.-M., Persoons, J.H.A., Schotanus, K., Steinbusch, H.W.N., Schoffelmeer, A.N.M. & Berkenbosch, F. (1996) Gradual inhibition of inducible nitric oxide synthase but not interleukin-1b production in rat microglial cells of endotoxin-treated mixed glial cell cultures. Glia, 17, 94±102. Vitkovic, L., Bockaert, J. & Jacque, C. (2000) `In¯ammatory' cytokines: neuromodulators in normal brain? J. Neurochem., 74, 457±471. Yang, L., Tanaka, J., Zhang, B., Sakanaka, M. & Maeda, N. (1998) Astrocytes modulate nitric oxide production by microglial cells through secretion of serine and glycine. Biochem. Biophys. Res. Commun., 251, 277±282. Zielasek, J., Tausch, M., Toyka, K.V. & Hartung, H.P. (1992) Production of nitrite by neonatal rat microglial cells/brain macrophages. Cell. Immunol., 141, 111±120.
ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 16, 1275±1283
