S100 beta stimulates inducible nitric oxide synthase activity and mRNA levels in rat cortical astrocytes.

The glia-derived, neurotrophic protein S100β has been implicated in development and maintenance of the nervous system. However, S100β has also been postulated to play a role in mechanisms of neuropathology, because of its specific localization and selective overexpression in Alzheimer's disease. To begin to address the question of whether S100β can induce potentially toxic signaling pathways, we examined the effects of the protein on nitric oxide synthase (NOS) activity in cultures of rat cortical astrocytes. S100β treatment of astrocytes induced a time- and dose-dependent increase in accumulation of the NO metabolite, nitrite, in the conditioned medium. The S100β- stimulated nitrite production was blocked by cycloheximide and by the NOS inhibitor N-nitro-L-arginine methylester, but not by the inactive D-isomer of the inhibitor. Direct measurement of NOS enzymatic activity in cell extracts and analysis of NOS mRNA levels showed that the NOS activated by S100β addition is the calcium-independent, inducible isoform. Furthermore, the specificity of the effects of S100β on activation of NOS was demonstrated by the inability of S100α and calmodulin to induce an increase in nitrite levels. Our data indicate that S100β can induce a potent activation of inducible NOS in astrocytes, an observation that might have relevance to the role of S100β in neuropathology.

The glia-derived, neurotrophic protein S100␤ has been implicated in development and maintenance of the nervous system. However, S100␤ has also been postulated to play a role in mechanisms of neuropathology, because of its specific localization and selective overexpression in Alzheimer's disease. To begin to address the question of whether S100␤ can induce potentially toxic signaling pathways, we examined the effects of the protein on nitric oxide synthase (NOS) activity in cultures of rat cortical astrocytes. S100␤ treatment of astrocytes induced a time-and dose-dependent increase in accumulation of the NO metabolite, nitrite, in the conditioned medium. The S100␤stimulated nitrite production was blocked by cycloheximide and by the NOS inhibitor N-nitro-L-arginine methylester, but not by the inactive D-isomer of the inhibitor. Direct measurement of NOS enzymatic activity in cell extracts and analysis of NOS mRNA levels showed that the NOS activated by S100␤ addition is the calcium-independent, inducible isoform. Furthermore, the specificity of the effects of S100␤ on activation of NOS was demonstrated by the inability of S100␣ and calmodulin to induce an increase in nitrite levels. Our data indicate that S100␤ can induce a potent activation of inducible NOS in astrocytes, an observation that might have relevance to the role of S100␤ in neuropathology.
The normal development and maintenance of the brain involves the temporal and spatial coordination and proper functioning of a number of intracellular and cell-cell signaling events, and the contribution of glial cells to these signaling processes is becoming more widely appreciated. The classical concept of the role of glia in brain function is rapidly changing with newer evidence of the crucial nature of these cells in controlling neurotransmitter levels, maintaining calcium homeostasis, and synthesizing and releasing neurotrophic and growth factors (for review, see Ref. 1). One such glia-derived factor is S100␤, a protein that promotes neuritic outgrowth of specific neuronal populations (e.g. cortical (2,3), dorsal root ganglia (4), serotonergic (5,6), and motoneurons (7)) and enhances survival of neurons during development (7,8) and after insult (9). S100␤ is also a glial mitogen, inducing phosphoinositide hydrolysis, increases in intracellular calcium, and protooncogene expression (10,11). These trophic functions require nanomolar concentrations of a disulfide-linked S100␤ dimer (see Ref. 12). Thus, S100␤ may be beneficial during development of the nervous system, and increased S100␤ expression and secretion following acute glial activation in response to central nervous system injury may be one mechanism the brain uses in attempts to repair injured neurons.
However, S100␤ may also reach concentrations that are deleterious, e.g. in neurodegenerative diseases like Alzheimer's disease and Down syndrome where chronic glial activation occurs (13). It has been found that S100␤ levels in severely affected brain regions of Alzheimer's disease patients are severalfold higher than in age-matched control samples (13,14), that S100␤ is increased selectively in regions that exhibit the most neuropathological involvement (15), and that S100␤ overexpression is correlated with the prevalence of neuritic plaques (16). These data raise the possibility that high concentrations of S100␤ may be detrimental, an idea supported by the recent demonstration that an S100 isoform can induce apoptosis in PC12 cells through sustained increases in intracellular calcium (17, 18).
As a first approach to addressing the question of whether S100␤ can induce potentially toxic signaling pathways, we examined the effect of S100␤ on nitric oxide (NO) generation. NO is synthesized by the enzyme nitric oxide synthase (NOS) 1 through the conversion of L-arginine to L-citrulline. In the nervous system, NO has been implicated in the regulation of cerebral blood flow, synaptic plasticity, and cell growth (see Ref. 19). A large amount of data also supports the concept that NO production in the central nervous system may be involved in the neuropathology associated with ischemia, traumatic insults, and neurodegeneration (see Refs. 19 -21). We report here that treatment of rat cortical astrocytes with S100␤ results in a stimulation of NOS activity and generation of NO.

EXPERIMENTAL PROCEDURES
Purification of Proteins-Calmodulin and S100␣ were purified as described previously (22,23). Bovine S100␤ was expressed from a synthetic gene in Escherichia coli (24). Purification was achieved by DEAE anion exchange chromatography followed by calcium-dependent phenyl-Sepharose chromatography as described previously (11). Elution of S100␤ from phenyl-Sepharose was done with buffer containing 20 mM Tris-HCl, 500 mM NaCl, 1 mM EGTA, pH 7.4 (elution buffer). For experiments with tissue-isolated protein, S100 was isolated from bovine brain (Pel-Freez, Rogers, AR) by a similar procedure. In addition, E. coli lysates (lacking the S100 expression vector) were taken through the same purification protocol. The purity of the S100␤ was analyzed by electrophoresis on 15% acrylamide-SDS minigels (Idea Scientific, Minneapolis MN) in the absence of reducing agents. S100␤ was stored at Ϫ20°C in storage buffer (elution buffer plus 4 mM CaCl 2 ). Protein concentrations were determined by the method of Lowry et al. (25) using bovine serum albumin as standard or by amino acid analysis as * These studies were supported in part by National Institutes of Health Grants AG10208 and AG11138. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ¶ To whom correspondence and reprint requests should be addressed: Dept. of Cell and Molecular Biology, Northwestern University Medical School, 303 E. Chicago Ave., Chicago, IL 60611-3008. Tel.: 312-503-0697; Fax: 312-503-0007; E-mail: vaneldik@nwu.edu. described previously (26).
Cell Culture-Primary astrocytes were prepared from neonatal (1day-old) Sprague-Dawley rat pups as described by . Briefly, the cerebral cortex from one rat pup was dissected out and trypsinized, and cells were passed through a 136-m and then a 40-m nylon mesh. Cells were then seeded into two 100-mm tissue culture plates at a density of ϳ2 ϫ 10 6 cells/plate in ␣-MEM (Life Technologies, Inc.) containing 10% fetal bovine serum (HyClone Laboratories, Logan, UT) and antibiotics (100 units/ml penicillin and 100 g/ml streptomycin), and grown at 37°C in a humidified 5% CO 2containing atmosphere. After 11 days in culture, cells were trypsinized from the two plates and re-plated into 20 plates and grown until confluence.
Treatment of Cells for NO Assays-Confluent secondary astrocytes were trypsinized and re-plated into 24-well plates at 1 ϫ 10 5 cells/well (for the nitrite assay) or into 100 mm plates at 1 ϫ 10 6 cells/plate (for the citrulline assay and Northern blot analysis). Cells were allowed to attach for 4 h, at which time (time 0) S100␤ (1-50 g/ml) was added to the culture medium and cells were grown for various time periods. Some cultures were treated with lipopolysaccharide (LPS; 10 g/ml) or storage buffer instead of S100␤, and these cultures served as positive and negative controls, respectively. When cycloheximide or the NOS inhibitor, N-nitro-L-arginine methylester (L-NAME) were used, these agents were added to the cultures at the same time as the S100␤.
Measurement of NO by Nitrite Assay-NO production was assessed by measurement of nitrite (a stable oxidation product of NO) in the conditioned medium, based on the Griess reaction (28). Sodium nitrite dissolved in culture medium was used as the standard. Nitrite levels were determined by mixing 100-l aliquots of conditioned medium with 50 l of 1% sulfanilamide in water plus 50 l of 0.1% N-1-naphthylethylenediamine dihydrochloride in 5% phosphoric acid, and incubating for 10 min at room temperature. The absorbance at 540 nm was then measured by using a Titertek Multiskan MC plate reader (ICN/Flow Biochemicals, Huntsville AL). (29). Briefly, cells treated with various agents as described above were washed with ice-cold phosphate-buffered saline (PBS) and then scraped into homogenization buffer (20 mM HEPES, pH 7.4, 0.5 mM EGTA, 1 mM dithiothreitol, 0.32 M sucrose). Cells were homogenized by brief sonication, and then centrifuged at 12,000 ϫ g for 10 min in a Sorvall MC-12V microcentrifuge. The resulting supernatant was passed through a Dowex AG50W-X8 (Na ϩ form) column (Bio-Rad) to remove endogenous arginine. Aliquots of the unbound material (300-l reaction volumes containing 100 g protein) were incubated at 37°C for 45 min in homogenization buffer containing 200 M NADPH, 50 M tetrahydrobiopterin, 10 M L-arginine, and 1 Ci/ml [ 3 H]L-arginine in the absence and presence of 0.5 mM CaCl 2 (1 M calculated free calcium) or 1 mM L-NAME. Reactions were terminated by adding 2 ml of 20 mM HEPES, 2 mM EDTA, pH 5.5. One ml of each reaction sample was passed over a Dowex AG50W-X8 (Na ϩ form) column to separate [ 3 H]L-citrulline from [ 3 H]L-arginine. The level of [ 3 H]L-citrulline was determined by liquid scintillation counting and was expressed as the radioactivity (cpm/g protein) obtained after correcting for nonspecific radioactivity in blank reactions containing all components except cell extract.

Measurement of NOS by Citrulline Assay-NOS activity was determined by measurement of the conversion of [ 3 H]L-arginine (Amersham Corp.) to [ 3 H]L-citrulline according the the method of Bredt and Snyder
GFAP and OX-42 Immunohistochemistry-Secondary astrocytes were re-plated onto glass coverslips, fixed for 5 min in ice-cold methanol, and incubated for 1 h at room temperature with rabbit anti-GFAP (1:500 dilution; Dako, Carpinteria, CA) or mouse OX-42 antibody (1:500 dilution; gift of Dr. Sue Griffin, University of Arkansas for Medical Sciences). After washing with PBS, coverslips were incubated with fluorescein-conjugated goat anti-rabbit or anti-mouse IgG (1:20 dilution; Kirkegaard & Perry, Gaithersburg, MD), washed and mounted. The number of cells in 15 fields (magnification, ϫ63) were counted under phase-contrast microscopy, and the percentage of GFAP-or OX-42-positive cells in those fields was determined by fluorescence microscopy on a Leitz Diaplan microscope.
RNA Isolation and Northern Blot Analysis-RNA was isolated from treated astrocyte cultures as described by Ausubel et al. (30). Briefly, cells were washed twice with PBS and then lysed in a solution of 4 M guanidine thiocyanate, 25 mM sodium acetate, pH 6.0, 0.1 M 2-mercaptoethanol. The lysed cells were centrifuged through a 5.7 M cesium chloride step gradient at 140,000 ϫ g for 16 h at 20°C. The RNA in the pellet was dissolved, ethanol-precipitated, and quantitated spectrophotometrically at A 260 .
Data Analysis-The significance of differences was determined with Student's t test. Statistical significance was established at a level of p Ͻ 0.05.

RESULTS
The ability of S100␤ to stimulate NO production was determined initially by measuring the accumulation of nitrite (a stable NO metabolite) in the conditioned medium of astrocytes following exposure to S100␤. Tertiary cultures of neonatal rat cortical astrocytes were prepared as described under "Experimental Procedures." These cultures contained approximately 98% astrocytes, as assessed by positive staining for the astrocyte intermediate filament protein, GFAP (data not shown). This result was consistent with the observation that only 2-3% of the cells showed positive staining for OX-42 (data not shown), a marker for microglia (33). Fig. 1 shows the time course of nitrite accumulation following exposure of these cells to S100␤ (40 g/ml S100␤ for 6 -72 h). S100␤-treated astrocytes showed increased nitrite accumulation in their conditioned medium compared to control cultures treated with storage buffer alone. The increased nitrite levels were evident after a 12 h exposure of cells to S100␤ and reached a plateau at about 48 h. The maximal level of nitrite accumulation in response to S100␤ was 3-4-fold higher than that generated by exposure of cells to buffer alone.
The response to S100␤ was dose-dependent in the range of 1-40 g/ml S100␤, assayed after 48 h treatment of cells (Fig.  2). The concentration of S100␤ required to elicit 50% of the maximal nitrite increase was 20 -25 g/ml, and the maximal response was achieved by 35-40 g/ml S100␤.
To test whether the S100␤ -induced nitrite accumulation was dependent on NOS activity, we examined the effect of S100␤ in the presence of the NOS inhibitor, L-NAME. Incubation of cells for 48 h with S100␤ in the presence of L-NAME almost completely abolished the nitrite accumulation, whereas the inactive inhibitor isoform, D-NAME, was ineffective (Fig. 3). The basal level of nitrite accumulated during 48 h was also affected by L-NAME but not D-NAME.
Incubation of cells with S100␤ in the presence of the protein synthesis inhibitor cycloheximide prevented the S100␤ stimulation of nitrite accumulation (Table I). Moreover, the specificity of the response to S100␤ was examined by comparing the ability of two other related calcium binding proteins, S100␣ and calmodulin, to enhance nitrite accumulation. Table I shows that neither S100␣ nor calmodulin stimulated nitrite accumulation, under conditions where the LPS positive control induced nitrite accumulation to a similar extent as S100␤. To exclude the possibility that the S100␤ stimulation of nitrite accumulation was a result of contaminating LPS in the recombinant S100␤ preparation, we isolated S100␤ from bovine brain tissue and we took an E. coli lysate lacking the S100 expression vector through the same purification protocol as S100. We found that bovine brain S100␤ induced nitrite accumulation to a similar extent as recombinant S100␤, and that the E. coli lysate showed no induction above the buffer control (data not shown).
The ability of S100␤ to enhance nitrite accumulation that was inhibited by NOS inhibitors suggested an activation of NOS enzyme activity. To test this directly, we examined NOS enzyme activity in cytosolic extracts of astrocytes treated with S100␤ for 24 h, by measuring the conversion of L-arginine to L-citrulline. As shown in Fig. 4, S100␤ treatment of astrocytes resulted in a ϳ4-fold stimulation of NOS activity. The S100␤ -evoked NOS activity was independent of the presence of calcium in the enzyme assay, suggesting that the NOS stimulated by S100␤ is the inducible enzyme (iNOS). This was further confirmed by the demonstration that S100␤ -treated astrocytes showed increased levels of iNOS mRNA, as measured by Northern blot analysis (Fig. 5). DISCUSSION We have demonstrated that treatment of astrocytes with S100␤ enhances NOS activity and release of NO, as demonstrated by the increase in accumulation of the stable NO metabolite nitrite in the conditioned medium. Measurement of   13.1 Ϯ 3.6 b 7 S100␤ (40 g/ml) ϩ cycloheximide (10 M) a Tertiary cultures of rat cortical astrocytes were treated with S100␤, S100␣, calmodulin, LPS, or control storage buffer for 48 h, and then nitrite concentration in the conditioned medium was measured by the Griess reaction as described under "Experimental Procedures." For experiments with cycloheximide, the drug was added at the same time as S100␤. Data shown are the mean Ϯ S.E. from n number of independent experiments each done in triplicate. b p Ͻ 0.05 versus control values.
FIG. 4. S100␤ stimulation of inducible iNOS activity in rat astrocyte cultures. Cells were incubated for 24 h with S100␤ (40 g/ml) or control storage buffer, and NOS activity in cytosolic extracts determined. Enzyme assays were done in the presence or absence of calcium as described under "Experimental Procedures." Data shown are from one of four independent experiments, each done in triplicate. Values shown are the mean Ϯ S.E. from the triplicate determinations. S100␤-stimulated NOS activity was significantly different from control (p Ͻ 0.05).
NOS enzyme activity and mRNA levels confirms that the NOS stimulated by S100␤ addition is the calcium-independent inducible iNOS isoform. The specific NOS inhibitor L-NAME, but not the inactive D-NAME, suppressed the S100␤-induced iNOS activity. Moreover, the specificity of the effects of S100␤ on iNOS activation was demonstrated by the inability of S100␣ or calmodulin to induce an increase in NO. These data demonstrate that S100␤ induces a stimulation of astrocytic iNOS activity and generation of NO.
The NOS enzyme responsible for synthesizing NO from arginine exists in three major forms: two constitutive cNOS isoforms (eNOS and nNOS) that are calcium-calmodulin dependent and present in a variety of cells, including endothelial cells, neurons, platelets, and astrocytes; and an inducible form (iNOS) that is calcium-independent and expressed after gene induction in a variety of cells, including macrophages, endothelial cells, and astrocytes (see Ref. 34). Induction of iNOS in astrocytes has been well documented previously by exposure of cells in vitro to bacterial endotoxin (LPS) or combinations of cytokines, such as interleukin-1␤, interferon-␥ or tumor necrosis factor-␣ (1,35). Our data represent the first report that S100␤ is another potent inducer of astrocytic iNOS activity. The S100␤-stimulated NO production exhibits similar characteristics to those reported for induction by LPS or other cytokines in terms of long (hour) time frame and narrow doseresponse curve (31,35). The response occurs over several hours of continuous application of S100␤, and nitrite concentrations continue to increase up to 48 h after S100␤ exposure. Similar to IL-1␤ (35), the effect of S100␤ on iNOS activity was dose-dependent through a small concentration range. The magnitude of the nitrite accumulation in response to S100␤ also varied somewhat among cultures and among S100␤ preparations, suggesting that endogenous coordinators or specific S100␤ conformational states might be required for maximal S100␤ induction of iNOS.
We considered the possibility that our results might reflect an activation of NOS activity in microglia, rather than astrocytes, because microglia express high levels of iNOS activity (36). However, this possibility seems unlikely because our tertiary cultures are ϳ98% astrocytes, as determined by GFAP and OX-42 immunoreactivity, and because the intensity of the iNOS mRNA on Northern blots and the iNOS enzyme activity from cytosolic extracts are too high to reflect a signal from only 2% of the cells in the culture. Thus, our data support an S100␤-induced activation of iNOS activity in astrocytes.
The mechanisms involved in activation of iNOS by S100␤ addition are unknown at present. Our observation that S100␤ stimulates iNOS mRNA levels suggests that regulation may be at the level of transcription of the iNOS gene, as has been shown previously for iNOS stimulation by LPS and interferon-␥ (34,37,38). However, iNOS has also been found to be regulated at the level of mRNA and protein stability (39,40). Definition of the mechanisms by which S100␤ induces iNOS activity, and whether S100␤ interacts with other cytokines to modulate iNOS activity await further investigation.
The consequences of S100␤-induced NO release from astrocytes are not known. There is a wealth of sometimes conflicting evidence in the literature that NO can be beneficial or detrimental to nervous system function (see Refs. 19 and 21). For example, it has been found that NO can act as a diffusible messenger to mediate cell-cell signaling pathways involved in synaptic plasticity and regulation of cerebral blood flow. However, there is also a large amount of biochemical and pharmacological data to suggest that NO is involved in the neuropathology associated with traumatic or ischemic insults, autoimmune diseases, and neurodegenerative disorders. Relevant to our studies, it has been reported (41) that stimulation of astrocytic iNOS activity and release of NO leads to enhanced NMDA receptor-mediated neurotoxicity. We are currently pursuing studies with astrocytic/neuronal co-cultures to evaluate the effects of S100␤-stimulated NO release on the neuronal cell. In this regard, it is interesting to note that in some experiments, the S100␤-treated astrocytes exhibited morphological changes consistent with toxicity, such as release of lactate dehydrogenase and cell rounding and detachment from the substrate (data not shown). The reason for this observation has not been defined yet, but appeared to correlate with the levels of nitrite accumulated in response to S100␤. It is probable that the toxic potential of NO in vivo depends, at least in part, on the concentration released from cells. However, the biology of NO is complex and the susceptibility of a cell to NO-mediated toxicity likely depends on a number of variables, such as acute versus chronic exposure to NO, the array of redox forms in which NO can exist, availability of reactive oxygen species and anti-oxidant defenses, changes in ion homeostasis, presence of other cytoprotective or cytotoxic agents, and the immune status of the organism.
Our data demonstrate that treatment of astrocytes with S100␤ results in a potent induction of iNOS activity and NO production. Stimulation of iNOS required micromolar concentrations of S100␤, unlike the nanomolar concentrations required for neurotrophic and mitogenic effects. The requirement of higher concentrations of S100␤ for activation of iNOS suggests that S100␤ might possess dual roles in regulation of cell function, being beneficial to cells at low doses and detrimental at higher doses. If S100␤ stimulates astrocytic iNOS activity in vivo, this might be relevant to the role of S100␤ in neurodegenerative disorders like Alzheimer's disease and Down syndrome, where S100␤ levels are increased severalfold in temporal lobe samples compared to age-matched control samples (13). It has been found that S100␤ levels are elevated in specific brain regions from Alzheimer patients (15), that the overexpression of S100␤ correlates with the pattern of regional neuropathology and neuritic plaque involvement (15,16), and that S100␤ is localized primarily in activated astrocytes surrounding neuritic plaques (15). It should also be noted that the local concentration of S100␤ in astrocytes surrounding neuritic plaques in Alzheimer's disease has not been determined, although the tissue levels are in the micromolar range (15). The relationship between increased S100␤ levels and neuropathology is not known. It may be that the up-regulation of S100␤ after acute central nervous system injury or perhaps in neurodegenerative disease is part of a compensatory response the brain uses in attempts to repair injured neurons, through an action of S100␤ in its neurotrophic and neuroprotective roles. FIG. 5. S100␤ stimulation of iNOS mRNA. Cells were incubated for 24 h with S100␤ (40 g/ml) or control storage buffer, and iNOS mRNA levels were determined by Northern blot analysis of 10 g/lane total RNA, as described under "Experimental Procedures." The blot was re-probed with cyclophilin cDNA, and the iNOS mRNA levels were expressed as the ratio of density of iNOS mRNA versus cyclophilin mRNA. Inset shows the iNOS mRNA bands from control (C) versus S100␤-treated (S) astrocytes. However, the chronic overexpression of S100␤ or expression of S100␤ above some threshold level as in Alzheimer's disease may have deleterious consequences leading to neuronal dysfunction and eventual death, perhaps through altered calcium homeostasis or inappropriate neuritogenesis. Our data that S100␤ can stimulate astrocytes to produce NO provide another possible mechanistic scenario by which the high levels of S100␤ seen in Alzheimer's disease and other neurodegenerative disorders might contribute to neuropathology.