Fas/Apo-1 is a member of the TNF receptor superfamily that signals apoptotic cell death in susceptible target cells. Fas or Fas ligand (FasL)-deficient mice are relatively resistant to the induction of experimental allergic encephalomyelitis, implying the involvement of Fas/FasL in this disease process. We have examined the regulation and function of Fas expression in glial cells (astrocytes and microglia). Fas is constitutively expressed by primary murine microglia at a low level and significantly up-regulated by TNF-α or IFN-γ stimulation. Primary astrocytes express high constitutive levels of Fas, which are not further affected by cytokine treatment. In microglia, Fas expression is regulated at the level of mRNA expression; TNF-α and IFN-γ induced Fas mRNA by ∼20-fold. STAT-1α and NF-κB activation are involved in IFN-γ- or TNF-α-mediated Fas up-regulation in microglia, respectively. The cytokine TGF-β inhibits basal expression of Fas as well as cytokine-mediated Fas expression by microglia. Upon incubation of microglial cells with FasL-expressing cells, ∼20% of cells underwent Fas-mediated cell death, which increased to ∼60% when cells were pretreated with either TNF-α or IFN-γ. TGF-β treatment inhibited Fas-mediated cell death of TNF-α- or IFN-γ-stimulated microglial cells. In contrast, astrocytes are resistant to Fas-mediated cell death, however, ligation of Fas induces expression of the chemokines macrophage inflammatory protein-1β (MIP-1β), MIP-1α, and MIP-2. These data demonstrate that Fas transmits different signals in the two glial cell populations: a cytotoxic signal in microglia and an inflammatory signal in the astrocyte.

Fas (CD95) is a type I transmembrane glycoprotein belonging to the TNF receptor superfamily (1). It is expressed on activated T and B cells, thymocytes, malignant T and B cells, and in a variety of tissues outside the immune system, including liver and lung (2). Upon ligation with agonistic Ab or the natural Fas ligand (FasL),3 Fas trimerizes and recruits a number of proteins sharing a death domain that leads to the formation of a specific death-inducing signaling complex (DISC) at the intracellular region of the Fas receptor (3). The recruitment of caspase-8 to DISC results in proteolytic activation of the enzyme, which, in turn, activates a series of other caspase members (4). FasL is a type II transmembrane glycoprotein that induces apoptosis in target cells in both the membrane-bound form and the soluble form (sFasL) (5, 6). In the immune system, cell death through the Fas/FasL system is involved in B and T cell homeostasis and is also critical for CTL-mediated cytotoxicity (7, 8, 9, 10). Recently, it was reported that the Fas/FasL system is responsible for immune down-regulation in previously known immune-privileged sites such as cornea and testis (11, 12). Mouse strains carrying mutations in the Fas (lpr) or FasL (gld) genes exhibit lymphoadenopathy and autoimmunity reminiscent of systemic lupus erythematosus (13).

In experimental allergic encephalomyelitis (EAE), an autoimmune inflammatory disease of the CNS, the involvement of Fas/FasL was clearly demonstrated by observations that lpr or gld mice are relatively resistant to the clinical development of EAE (14, 15). These data suggested that the Fas/FasL pathway plays a significant role in the development of EAE, possibly by mediating apoptosis of potential target cells. However, in two separate studies, oligodendrocytes, the myelin-producing cells in the CNS that are considered a major target in the disease of EAE, were resistant to FasL-mediated apoptosis and did not undergo apoptosis during the pathogenesis of EAE (16, 17). To the contrary, infiltrating CD4+ T cells and parenchymal microglia were killed by apoptosis, suggesting that the effector cells, rather than the target cells, are eliminated by apoptosis in the EAE disease process (16). More recently, the involvement of Fas/FasL in EAE progression has been dissected using the adoptive transfer model of EAE (18, 19). According to these studies, Fas expressed in the recipient animal as well as FasL expressed on donor cells are both important for the development of EAE. These data suggest that Fas expression on CNS-resident glial cells contributes to EAE induction.

In an attempt to elucidate the involvement of Fas/FasL in EAE as well as MS pathogenesis, Fas expression and function on glial cells has been investigated in several in vitro studies. In a study using human oligodendrocytes, Fas ligation with anti-Fas Ab induced rapid lysis of oligodendrocytes, supporting the idea of Fas-dependent oligodendrocyte elimination in MS (20). However, this Fas-dependent cell lysis did not exhibit characteristics of typical apoptosis. High constitutive expression of Fas on mouse astrocytes has been reported as well, but astrocyte susceptibility to FasL-mediated apoptosis is still controversial (21, 22). In a study using astrocyte cultures from fetal CNS, although Fas was constitutively expressed on astrocytes, Fas ligation failed to induce astrocyte cell death (23). In another report, Fas was constitutively expressed by human fetal astrocytes as well as adult astrocytes and was up-regulated by several proinflammatory cytokines such as IL-1, IL-6, IFN-γ, and TNF-α (24). In addition, the fetal astrocytes underwent apoptosis when treated with agonistic anti-Fas Ab. It has been suggested that the susceptibility of human astrocytes to Fas-mediated apoptosis is dependent on cell passage and other microenvironmental factors; only late passage astrocytes are sensitive to Fas-induced apoptosis and the presence of IFN-γ enhances apoptosis (25). Apoptosis in microglia has not been well studied compared with that in astrocytes. In several studies using immunohistochemistry in EAE brain, microglial cells were positively stained by the TUNEL assay more frequently than astrocytes (16). Recently, Fas-mediated apoptosis was reported in cultured microglia upon TNF-α or IFN-γ stimulation (26).

In this study we have investigated the regulation and function of Fas expression using mouse astrocytes and microglia. We examined the influence of several pro- and anti-inflammatory cytokines (IFN-γ, TNF-α, TGF-β) on Fas expression in glial cells as well as the functional outcome of Fas ligation on these cells. Herein, we report that Fas expression is differentially modulated by cytokines on astrocytes and microglia, and Fas ligation transmits distinct signals in these two glial cell types.

Primary glial cell cultures were established from neonatal CD1 as well as C57BL/6/lpr mouse cerebra as previously described (27). Cells were cultured in DMEM, high glucose formula supplemented with glucose to a final concentration of 6 g/L, 2 mM glutamine, 0.1 mM nonessential amino acid mixture, 0.1% gentamicin, and 10% FBS (HyClone, Logan, UT). After 2 wk in primary culture, oligodendrocytes and microglia were separated from astrocytes by mechanical dislodgment. Microglia-enriched cultures were obtained by incubating the detached cells in six-well plates and changing the medium after 1 h to remove nonadherent cells. Microglia were also prepared from CD1 STAT-1α-deficient mice (provided by Dr. David E. Levy, New York University School of Medicine, New York, NY) (28). Astrocyte cultures were routinely >97% positive for glial fibrillary acidic protein, a specific marker for astrocytes, and primary microglial cultures were >95% positive as assessed by immunostaining for the Mac-1 surface Ag. The microglial cell line EOC13 was derived from C3H/HeJ CH-2k mice using a nonviral immortalization procedure as previously described (29). This CSF-1-dependent cell line is B7.1+, Mac-1+, CD45+, and class I MHC+ as well as phagocytic. The EOC13 cell line was maintained in DMEM complete medium (2 mM glutamine, 10% heat-inactivated FBS, and 20% LADMAC-conditioned medium, which contains CSF-1) as previously described (27).

Recombinant murine IFN-γ was purchased from Genzyme (Boston, MA), human TGF-β1 was obtained from R&D Systems (Minneapolis, MN), and rat TNF-α was purchased from BioSource International (Camarillo, CA). Hamster anti-mouse Fas Ab (Jo2), PE-conjugated hamster anti-mouse Fas Ab (Jo2), hamster anti-mouse Fas ligand Ab (MFL3), hamster anti-mouse ICAM-1 Ab (3E2), and hamster anti-TNP IgG were purchased from PharMingen (San Diego, CA). The NF-κB inhibitor SN50, the p38 MAP kinase inhibitor SB202190, and olomoucine were purchased from Calbiochem (San Diego, CA). The MAP/ERK kinase (MEK) inhibitor U0126 was purchased from Promega (Madison, WI). Murine Fas cDNA in the pcDNA 1.1 plasmid was generated by PCR as previously described (30). The plasmid was cut with XbaI, and the 500-bp fragment containing the C-terminal half of mouse Fas cDNA was subcloned into pcDNA3 at an XbaI restriction enzyme site. After linearization with HindIII, this construct (pcDNA3-mouse Fas) was used for in vitro transcription to generate a 420-bp antisense RNA probe. A pGEM-4Z vector containing a fragment of mouse GAPDH cDNA (corresponding to bp 223–434) inserted at the polylinker sites EcoRI/KpnI was linearized with EcoRI. In vitro transcription of this plasmid with T7 RNA polymerase generated a 270-bp antisense RNA probe.

Primary astrocytes were plated at 5 × 105 cells/well into six-well plates (Costar, Cambridge, MA) and treated with IFN-γ or TNF-α in the absence or the presence of TGF-β1 for various time periods. The cells were trypsinized, washed with PBS, incubated with 20 μg/ml of PE-conjugated anti-Fas Ab (Jo2) for 1 h at 4°C, washed, then fixed in a final volume of 100 μl of 1% paraformaldehyde. The cells were then analyzed on the FACScan (Becton Dickinson, Mountain View, CA). Total fluorescence intensity was calculated as the mean fluorescence intensity × percentage of positive cells. Negative controls were incubated with isotype-matched Ab. For the analysis of primary microglia or EOC13 cells, cells were treated as described above, then scraped to detach from the well and incubated with 50 μl of 2.4G2 supernatant (which contains rat anti-mouse FcγR Ab) supplemented with 10% normal mouse serum for 30 min at 4°C before addition of PE-conjugated anti-Fas Ab.

Total cellular RNA was isolated from confluent monolayers of EOC13 cells or astrocytes that were incubated with cytokines or anti-Fas Ab, respectively, as previously described (31). Briefly, cells were washed once in PBS and lysed directly in the culture dish. RNA was extracted with guanidinium isothiocyanate and phenol, and precipitated with ethanol. Ten micrograms of total cellular RNA was analyzed by RPA using an RPA kit (Ambion, Austin, TX) as previously described (31). Total RNA was hybridized with both mouse Fas and GAPDH riboprobes (2.5 × 105 cpm) at 42°C overnight in 20 μl of 40 mM PIPES (pH 6.4), 80% deionized formamide, 400 mM NaOAc, and 1 mM EDTA. The hybridized mixture was treated with RNase A/T1 (1/100 dilution in 200 μl of RNase digestion buffer) at 37°C for 30 min, and then analyzed by 5% denaturing (8 M urea) PAGE. The protected fragments of the Fas and GAPDH riboprobes are 374 and 212 nucleotides in length, respectively. Quantification of protected RNA fragments was performed by scanning with the PhosphorImager (Molecular Dynamics, Sunnyvale, CA), and values for Fas mRNA were normalized to GAPDH mRNA levels for each experimental condition. The linearized mouse chemokine multiprobe set mCK-5 (catalogue no. 45026P) was purchased from PharMingen. The mCK-5 was in vitro transcribed with T7 RNA polymerase to produce antisense chemokine RNA probes as previously described (32).

Fas-mediated microglial cell lysis was measured by the 51Cr release assay. FasL-expressing effector cells were generated by transfection of FasL cDNA-containing adenovirus vector into macrophages from Fas-deficient lpr mice as previously described (33). EOC13 cells were incubated with medium alone or different cytokines for 40 h and radiolabeled by incubation with 20 μCi of [51Cr]sodium chromate in 200 μl of DMEM containing 10% FBS at 37°C for 1 h. After washing the cells three times, 51Cr-labeled EOC cells (1 × 105) were incubated with an equivalent number of effector cells. The release of 51Cr in the supernatant was assessed 8 h later using a gamma counter. For each condition, FasL-negative macrophages were used as a negative control, 51Cr release was measured, and values were used as spontaneous release. Maximum 51Cr release was measured from each positive control in which cells are lysed in 1% Triton X-100. The percentage of specific cell death rate was calculated as (experimental 51Cr release − spontaneous release)/(maximum 51Cr release − spontaneous release).

We initially analyzed cell surface Fas expression on primary mouse astrocytes and microglia by FACS analysis. Both cell types were incubated in medium alone or with TNF-α or IFN-γ for 40 h before staining. On astrocytes, Fas is constitutively expressed and is not significantly modulated by treatment with TNF-α or IFN-γ (Fig. 1). In primary mouse microglia, Fas is expressed at a low level compared with that of astrocytes, and expression is up-regulated upon TNF-α or IFN-γ treatment (∼5-fold induction; Fig. 1).

FIGURE 1.

Fas expression on primary mouse astrocytes and microglia. Primary mouse astrocytes and microglia were stimulated with medium alone (Control), rat recombinant TNF-α (50 ng/ml), or mouse recombinant IFN-γ (100 U/ml). After 40 h of stimulation, cells were either trypsinized (astrocytes) or scraped (microglia) for the analysis of surface Fas expression by immunofluorescence flow cytometry (shown in black). Cells stained with isotype-matched control Ab are shown in white. Results are representative of three independent experiments.

FIGURE 1.

Fas expression on primary mouse astrocytes and microglia. Primary mouse astrocytes and microglia were stimulated with medium alone (Control), rat recombinant TNF-α (50 ng/ml), or mouse recombinant IFN-γ (100 U/ml). After 40 h of stimulation, cells were either trypsinized (astrocytes) or scraped (microglia) for the analysis of surface Fas expression by immunofluorescence flow cytometry (shown in black). Cells stained with isotype-matched control Ab are shown in white. Results are representative of three independent experiments.

Close modal

Because Fas expression was regulated on microglia by the cytokines TNF-α and IFN-γ, we further investigated cytokine regulation of Fas expression using the microglial cell line, EOC13. EOC13 is a transformed microglial line that exhibits numerous characteristics of primary microglia (27, 29). Surface Fas expression on unstimulated or cytokine-stimulated EOC13 cells was regulated in a comparable manner to that of primary microglia (Fig. 2,A). Kinetic analysis demonstrated that IFN-γ up-regulated Fas expression with slightly different kinetics compared with those of TNF-α (Fig. 2,B). TNF-α up-regulated Fas expression by ∼6–12 h after stimulation, with a peak at 36 h, whereas IFN-γ most actively induced Fas within a 24- to 36-h period, with expression still increasing at 40 h (Fig. 2 B).

FIGURE 2.

Fas expression on EOC13 cells. A, EOC13 cells were treated either with TNF-α (50 ng/ml) or IFN-γ (100 U/ml) for 40 h. Surface expression of Fas was analyzed by immunofluorescence flow cytometry (shown in black). Cells stained with isotype-matched control Ab are shown in white. Results are representative of four independent experiments. B, EOC13 cells plated in six-well plates were stimulated with either TNF-α or IFN-γ for various time periods, then surface Fas expression was analyzed. The data are represented as fold induction, which was calculated by dividing the total fluorescence intensity of each cytokine stimulated sample by that of the unstimulated sample. Results are representative of three independent experiments.

FIGURE 2.

Fas expression on EOC13 cells. A, EOC13 cells were treated either with TNF-α (50 ng/ml) or IFN-γ (100 U/ml) for 40 h. Surface expression of Fas was analyzed by immunofluorescence flow cytometry (shown in black). Cells stained with isotype-matched control Ab are shown in white. Results are representative of four independent experiments. B, EOC13 cells plated in six-well plates were stimulated with either TNF-α or IFN-γ for various time periods, then surface Fas expression was analyzed. The data are represented as fold induction, which was calculated by dividing the total fluorescence intensity of each cytokine stimulated sample by that of the unstimulated sample. Results are representative of three independent experiments.

Close modal

Next, Fas mRNA expression was examined in EOC13 cells upon cytokine stimulation. Constitutive Fas mRNA was barely detectable (Fig. 3, lanes 1 and 7), but was induced by TNF-α or IFN-γ in a time-dependent manner (lanes 2–6 and 8–12). TNF-α induced Fas mRNA expression peaked at 6 h, while IFN-γ-induced Fas mRNA expression was optimal at 12 h of stimulation. These data demonstrate that Fas is regulated by the cytokines TNF-α and IFN-γ at the mRNA level.

FIGURE 3.

Cytokine-mediated Fas up-regulation on EOC13 cells is regulated at the mRNA level. Cells were stimulated with medium alone (lanes 1 and 7), IFN-γ (100 U/ml; lanes 2–6), or TNF-α (50 ng/ml; lanes 8–12) for various time periods (0–18 h). Total RNA was prepared from each sample and used for RPA. The bands of Fas mRNA and GAPDH mRNA were quantitated in the linear range using the PhosphorImager, and the radioactivity of each Fas band was normalized to that of GAPDH. Fold induction was calculated by dividing the Fas/GAPDH value for each cytokine-stimulated sample by the value for unstimulated cells. Results are representative of three independent experiments.

FIGURE 3.

Cytokine-mediated Fas up-regulation on EOC13 cells is regulated at the mRNA level. Cells were stimulated with medium alone (lanes 1 and 7), IFN-γ (100 U/ml; lanes 2–6), or TNF-α (50 ng/ml; lanes 8–12) for various time periods (0–18 h). Total RNA was prepared from each sample and used for RPA. The bands of Fas mRNA and GAPDH mRNA were quantitated in the linear range using the PhosphorImager, and the radioactivity of each Fas band was normalized to that of GAPDH. Fold induction was calculated by dividing the Fas/GAPDH value for each cytokine-stimulated sample by the value for unstimulated cells. Results are representative of three independent experiments.

Close modal

TNF-α- or IFN-γ-mediated Fas up-regulation has been reported in several other cell types (34). However, the transcriptional regulatory mechanisms and/or the transcription factors involved in Fas expression have been only recently reported (35). It is well known that the IFN-γ receptor transmits signals through the JAK-STAT pathway to induce transcription of various target genes (for review, see Ref. 36). Therefore, we investigated the involvement of STAT-1α in IFN-γ-mediated Fas expression using microglia from STAT-1α-deficient mice. As shown in Fig. 4,A, IFN-γ-induced Fas expression was severely impaired in STAT-1α-deficient mouse microglia, while TNF-α mediated Fas up-regulation was intact, demonstrating that STAT-1α is critical for IFN-γ mediated Fas induction. TNF-α binding to its receptor rapidly activates the transcription factor NF-κB in many cell types (for review, see Ref. 37). To further test the involvement of NF-κB activation in TNF-α-induced Fas expression, the activation of NF-κB was suppressed using the specific inhibitor, SN50. This oligopeptide contains the nuclear localizing signal found in NF-κB and specifically blocks nuclear translocation of NF-κB (38). SN50 inhibited TNF-α-induced Fas mRNA expression by ∼60%, while having no effect on IFN-γ-mediated Fas expression (Fig. 4 B). These data clearly demonstrate that NF-κB activation is required for TNF-α-mediated Fas expression in microglial cells.

FIGURE 4.

STAT-1α and NF-κB are involved in IFN-γ- and TNF-α-induced Fas expression, respectively. A, Microglia from STAT-1α-deficient mice were stimulated with medium alone, TNF-α (50 ng/ml), or IFN-γ (100 U/ml) for 40 h, then surface expression of Fas was analyzed by immunofluorescence flow cytometry. B, EOC13 cells were stimulated with medium alone (lane 1), TNF-α (50 ng/ml; lanes 2 and 3) for 6 h, or IFN-γ (100 U/ml; lanes 4 and 5) for 10 h. SN50 (100 μg/ml) was added 15 min before TNF-α (lane 3) or IFN-γ (lane 5) stimulation. Total RNA was prepared from each sample and used for RPA. Results are representative of three experiments.

FIGURE 4.

STAT-1α and NF-κB are involved in IFN-γ- and TNF-α-induced Fas expression, respectively. A, Microglia from STAT-1α-deficient mice were stimulated with medium alone, TNF-α (50 ng/ml), or IFN-γ (100 U/ml) for 40 h, then surface expression of Fas was analyzed by immunofluorescence flow cytometry. B, EOC13 cells were stimulated with medium alone (lane 1), TNF-α (50 ng/ml; lanes 2 and 3) for 6 h, or IFN-γ (100 U/ml; lanes 4 and 5) for 10 h. SN50 (100 μg/ml) was added 15 min before TNF-α (lane 3) or IFN-γ (lane 5) stimulation. Total RNA was prepared from each sample and used for RPA. Results are representative of three experiments.

Close modal

Previous studies from our laboratory have shown that TGF-β exerts a strong anti-inflammatory effect on microglia by antagonizing the effect of proinflammatory cytokines such as TNF-α and IFN-γ. Hence, we tested the effect of TGF-β on cytokine-induced Fas expression on EOC13 cells. FACS analysis demonstrated that TGF-β slightly inhibited the constitutive expression of Fas. In addition, TGF-β inhibited TNF-α-mediated Fas expression by ∼60% and IFN-γ-mediated Fas expression by ∼70% in EOC13 cells (Fig. 5). We also tested the inhibitory effect of TGF-β on Fas expression in primary mouse microglia and found comparable inhibitory effects (data not shown). Regulation of Fas mRNA expression was also tested upon TGF-β stimulation (Fig. 6). TGF-β alone had no effect on Fas mRNA expression (lane 4), whereas it inhibited TNF-α-induced Fas mRNA expression by ∼50% (compare lanes 2 and 5) and IFN-γ-induced Fas expression by ∼45% (compare lanes 3 and 6). These data demonstrate that the inhibitory effect of TGF-β on Fas expression is mediated at the mRNA level.

FIGURE 5.

Influence of TGF-β on Fas expression by EOC13 cells. EOC13 cells were stimulated with medium alone (Control) or with TNF-α or IFN-γ in the presence (dotted line) or the absence (solid line) of TGF-β1 (10 ng/ml) for 40 h. Cells were analyzed for surface Fas expression using FACS. Results are representative histograms of three independent experiments.

FIGURE 5.

Influence of TGF-β on Fas expression by EOC13 cells. EOC13 cells were stimulated with medium alone (Control) or with TNF-α or IFN-γ in the presence (dotted line) or the absence (solid line) of TGF-β1 (10 ng/ml) for 40 h. Cells were analyzed for surface Fas expression using FACS. Results are representative histograms of three independent experiments.

Close modal
FIGURE 6.

TGF-β inhibits TNF-α- and IFN-γ-induced Fas mRNA expression. Total RNA was prepared from EOC13 cells stimulated with medium alone (lane 1), TNF-α (50 ng/ml; lanes 2 and 5), or IFN-γ (100 U/ml; lanes 3 and 6) in the absence (lanes 1–3) or the presence of TGF-β1 (10 ng/ml; lanes 4–6) for 10 h and used for RPA. Quantification of the data is shown below as a bar graph. Results are representative of three experiments.

FIGURE 6.

TGF-β inhibits TNF-α- and IFN-γ-induced Fas mRNA expression. Total RNA was prepared from EOC13 cells stimulated with medium alone (lane 1), TNF-α (50 ng/ml; lanes 2 and 5), or IFN-γ (100 U/ml; lanes 3 and 6) in the absence (lanes 1–3) or the presence of TGF-β1 (10 ng/ml; lanes 4–6) for 10 h and used for RPA. Quantification of the data is shown below as a bar graph. Results are representative of three experiments.

Close modal

Fas transmits apoptotic signals into susceptible target cells (39). We tested whether the Fas receptor on the surface of microglia could induce cell death upon ligation with FasL. FasL-expressing macrophage cells from the lpr mouse that were transiently transfected with FasL cDNA were used. This cell line expresses FasL on the cell surface and induces apoptosis in Fas-bearing target cells (33). EOC13 cells were labeled with 51Cr and incubated with either FasL+lpr macrophages or FasL-untransfected control cells, then 51Cr released in the medium was measured. Without any stimulation, ∼20% of EOC13 cells underwent Fas-mediated cell death upon Fas ligation. This cell death rate increased to ∼60% when cells were stimulated with either TNF-α or IFN-γ (Fig. 7,A), indicating that TNF-α and IFN-γ potentiate Fas-mediated cell death in EOC13 cells. This cytokine-mediated potentiation of microglial cell death was almost completely inhibited when cells were incubated with TGF-β (Fig. 7,A). In addition, incubation with TGF-β alone inhibited the basal level of cell death of EOC13 cells (Fig. 7,A). The specificity of cell death was demonstrated by addition of Abs specific for Fas or FasL in the coculture. Abs against either Fas or FasL completely blocked the potentiation of cell death by TNF-α stimulation (Fig. 7,B). Likewise, the IFN-γ-induced cell death was completely blocked by Abs against Fas or FasL (data not shown). However, neither anti-ICAM-1 nor isotype control Ab (hamster IgG) inhibited Fas-induced cell death (Fig. 7 B).

FIGURE 7.

TGF-β inhibits Fas-mediated cell death of EOC13 cells. A, EOC13 cells were stimulated with medium alone (Control) or with different combination of cytokines for 40 h, then radiolabeled with 51Cr. Equal numbers of FasL+ effector cells were added to radiolabeled EOC13 cells (1 × 105), and the 51Cr release in medium was measured to calculate the specific cell death rate. The spontaneous release of 51Cr using this assay is routinely ∼20% of the maximum release in control and IFN-γ-stimulated samples, whereas TNF-α stimulation increased the spontaneous release to ∼35%. However, this effect is normalized upon calculation of the specific cell death rate (see Materials and Methods). The data shown are the mean ± SEM of five independent experiments. B, EOC13 cells stimulated with medium or TNF-α (50 ng/ml) for 40 h were used for the 51Cr release assay. To some TNF-α-stimulated samples, Abs against Fas (α-Fas), FasL (α-FasL), hamster IgG (hIgG), or ICAM-1 (α-ICAM-1) were added at a concentration of 20 μg/ml before the addition of FasL+ effector cells. Data are the mean ± SEM of two independent experiments.

FIGURE 7.

TGF-β inhibits Fas-mediated cell death of EOC13 cells. A, EOC13 cells were stimulated with medium alone (Control) or with different combination of cytokines for 40 h, then radiolabeled with 51Cr. Equal numbers of FasL+ effector cells were added to radiolabeled EOC13 cells (1 × 105), and the 51Cr release in medium was measured to calculate the specific cell death rate. The spontaneous release of 51Cr using this assay is routinely ∼20% of the maximum release in control and IFN-γ-stimulated samples, whereas TNF-α stimulation increased the spontaneous release to ∼35%. However, this effect is normalized upon calculation of the specific cell death rate (see Materials and Methods). The data shown are the mean ± SEM of five independent experiments. B, EOC13 cells stimulated with medium or TNF-α (50 ng/ml) for 40 h were used for the 51Cr release assay. To some TNF-α-stimulated samples, Abs against Fas (α-Fas), FasL (α-FasL), hamster IgG (hIgG), or ICAM-1 (α-ICAM-1) were added at a concentration of 20 μg/ml before the addition of FasL+ effector cells. Data are the mean ± SEM of two independent experiments.

Close modal

Recently, several studies have reported that Fas can transmit inflammatory signals depending on the cell type (25, 40). Because primary mouse astrocytes were completely resistant to Fas-mediated cell death in the 51Cr release assay (data not shown), we examined whether Fas on astrocytes has an alternative function, i.e., transduction of inflammatory signals. Specifically, we examined whether Fas ligation on astrocytes affected chemokine expression (Fig. 8). Astrocytes incubated with anti-mouse Fas Ab (Jo2) for 1 h up-regulated the mRNA expression of MIP-1β by ∼3-fold, MIP-1α by ∼2-fold, and MIP-2 by ∼2-fold (Fig. 8,A, lane 2) compared with that in the control sample (lane 1) in which cells were incubated with isotype-matched Ab. The addition of anti-hamster IgG Ab (as a secondary Ab) with anti-Fas Ab did not further enhance anti-Fas Ab-induced chemokine expression (lane 4), suggesting that anti-Fas Ab alone is sufficient to transduce inflammatory signals. As well, the addition of secondary Ab alone did not affect chemokine expression (lane 3). Considering the abundant expression of the proinflammatory cytokine TNF-α in the CNS during EAE pathogenesis (41), it is likely that Fas transmits signals in the presence of TNF-α during the disease process. To test possible concerted effects of Fas-mediated signals and TNF-α, we stimulated cells with anti-Fas Ab in the absence or the presence of TNF-α (Fig. 8,B). TNF-α alone induced mRNA expression of MIP-1α and -β by about 2-fold (lane 3). Furthermore, in the presence of TNF-α, cross-linking of Fas resulted in a synergistic effect on MIP-1β expression (∼7-fold induction) and an additive effect on MIP-1α expression (∼4-fold induction; lane 4). Possible additive/synergistic effects between Fas cross-linking and TNF-α on MIP-2 expression were not obvious, because TNF-α-induced MIP-2 expression was so strong (Fig. 8,B, lane 3). To test the specificity of Fas signaling, we used primary astrocytes isolated from lpr mice that are deficient in functional Fas expression (Fig. 8,C). Although TNF-α-mediated chemokine induction was intact (lane3), anti-Fas Ab-induced chemokine expression was completely abolished in lpr astrocytes (compare lanes 1 and 2). As well, addition of anti-Fas Ab plus TNF-α showed neither a synergistic nor an additive effect on chemokine expression in lpr astrocytes compared with TNF-α alone (lanes 3 and 4). These data clearly demonstrate that anti-Fas Ab generates an inflammatory signal through binding to the Fas molecule on astrocytes. To test whether this Fas-mediated chemokine induction is astrocyte-specific, we incubated microglial cells with anti-mouse Fas Ab (Jo2) for 1 h (Fig. 8,D). Unlike astrocytes, EOC13 cells constitutively express MIP-1α and MIP-1β mRNA at high levels, and Fas ligation did not further enhance expression (compare lanes 1 and 2). Because microglia constitutively express low levels of Fas, we stimulated EOC13 cells with TNF-α for 40 h to enhance Fas expression, then subjected these cells to Fas ligation. As shown in Fig. 8,D, TNF-α stimulation enhances the expression of MIP-1β, MIP-1α, and MIP-2 (compare lanes 1 and 3); however, Fas ligation did not affect expression of MIP-1β, MIP-1α, or MIP-2 mRNA expression (compare lanes 3 and 4). These data show that the Fas-mediated inflammatory signal occurs in astrocytes, but not in microglial cells. Finally, we investigated the possible signal transduction pathways involved in this Fas-mediated inflammatory signal using pharmacological inhibitors (Fig. 8 E). Among several kinase inhibitors that we tested, U0126, a specific inhibitor of MEK (42), most potently inhibited anti-Fas Ab-induced chemokine mRNA expression; a near-complete inhibitory effect was observed at 10 μM U0126 (lanes 3–5). Olomoucine, a cyclin-dependent kinase inhibitor as well as a weak inhibitor of extracellular signal-related kinase-1 (ERK-1), and SB202190, a specific p38 MAP kinase inhibitor, showed only moderate inhibitory effects at concentrations of 50 μM (lanes 6–11). MEK is a serine/threonine kinase that phosphorylates the MAP kinases ERK-1 and -2 (43). Taken together, these data suggest that cross-linking of Fas on astrocytes signals through the MAPK pathway, probably by ERK activation.

FIGURE 8.

Ligation of Fas on astrocytes induces chemokine expression. A, Primary mouse astrocytes were incubated with hamster anti-TNP Ab (1 μg/ml; lane 1), anti-Fas Ab (1 μg/ml; lane 2), anti-hamster IgG Ab (0.5 μg/ml; lane 3), or anti-Fas plus anti-hamster IgG Ab (lane 4) for 1 h. Total RNA was prepared from each sample and used for multiprobe RPA with the chemokine probe set mCK-5. The name of each protected mRNA signal is denoted in the middle. Results are representative of five experiments. B, Astrocytes were stimulated with either hamster anti-TNP Ab (1 μg/ml; lane 1), anti-Fas Ab (1 μg/ml; lane 2), TNF-α (50 ng/ml; lane 3), or anti-Fas Ab plus TNF-α (lane 4) for 1 h. Total RNA was used for chemokine multiprobe RPA. Results are representative of three experiments. C, Astrocytes from lpr mice were stimulated as described and used for chemokine multiprobe RPA. Results are representative of two experiments. D, EOC13 cells were stimulated with medium alone (lanes 1 and 2) or TNF-α (50 ng/ml; lanes 3 and 4) for 40 h, then incubated with either anti-Fas Ab (1 μg/ml; lanes 2 and 4) or isotype control Ab (1 μg/ml; lanes 1 and 3) for 1 h and used for chemokine multiprobe RPA. Results are representative of three experiments. E, Primary mouse astrocytes were incubated with varying concentrations of U0126 (lanes3–5), olomoucine (OL; lanes 6–8), or SB202190 (SB; lanes 9–11) for 30 min, then stimulated with either hamster anti-TNP Ab (1 μg/ml; lane 1) or anti-Fas Ab (1 μg/ml; lanes 2–11) for 1 h. Total RNA was prepared from each sample and used for chemokine multiprobe RPA. Results are representative of two experiments.

FIGURE 8.

Ligation of Fas on astrocytes induces chemokine expression. A, Primary mouse astrocytes were incubated with hamster anti-TNP Ab (1 μg/ml; lane 1), anti-Fas Ab (1 μg/ml; lane 2), anti-hamster IgG Ab (0.5 μg/ml; lane 3), or anti-Fas plus anti-hamster IgG Ab (lane 4) for 1 h. Total RNA was prepared from each sample and used for multiprobe RPA with the chemokine probe set mCK-5. The name of each protected mRNA signal is denoted in the middle. Results are representative of five experiments. B, Astrocytes were stimulated with either hamster anti-TNP Ab (1 μg/ml; lane 1), anti-Fas Ab (1 μg/ml; lane 2), TNF-α (50 ng/ml; lane 3), or anti-Fas Ab plus TNF-α (lane 4) for 1 h. Total RNA was used for chemokine multiprobe RPA. Results are representative of three experiments. C, Astrocytes from lpr mice were stimulated as described and used for chemokine multiprobe RPA. Results are representative of two experiments. D, EOC13 cells were stimulated with medium alone (lanes 1 and 2) or TNF-α (50 ng/ml; lanes 3 and 4) for 40 h, then incubated with either anti-Fas Ab (1 μg/ml; lanes 2 and 4) or isotype control Ab (1 μg/ml; lanes 1 and 3) for 1 h and used for chemokine multiprobe RPA. Results are representative of three experiments. E, Primary mouse astrocytes were incubated with varying concentrations of U0126 (lanes3–5), olomoucine (OL; lanes 6–8), or SB202190 (SB; lanes 9–11) for 30 min, then stimulated with either hamster anti-TNP Ab (1 μg/ml; lane 1) or anti-Fas Ab (1 μg/ml; lanes 2–11) for 1 h. Total RNA was prepared from each sample and used for chemokine multiprobe RPA. Results are representative of two experiments.

Close modal

We initiated this study by characterizing cell surface Fas expression in primary astrocytes and microglia using FACS analysis. Our data show that Fas is constitutively expressed at a high level on astrocytes and is not further up-regulated by the cytokines TNF-α and IFN-γ. Fas is also highly expressed on human fetal astrocytes and is not modulated by IFN-γ stimulation (23). However, our data differ from the report by Choi et al. (24) in which they could up-regulate Fas expression on human fetal astrocytes by TNF-α stimulation and less so by IL-1β or IL-6. We did not detect any increase in Fas expression on murine astrocytes by TNF-α; this suggests a species-specific differential sensitivity of astrocytes to TNF-α stimulation of Fas expression.

In distinction to astrocytes, microglia express low basal levels of Fas, which are further up-regulated by TNF-α or IFN-γ. The RPA data demonstrate that Fas expression is regulated by cytokines at the mRNA level. Interestingly, we observed that cytokine induction of Fas mRNA expression is higher than that of Fas protein expression, suggesting the existence of possible regulatory mechanisms at the post-transcriptional and/or translational level. Studies using STAT-1α-deficient microglia and the specific NF-κB inhibitor SN50 prove that STAT-1α and NF-κB activation are required for IFN-γ- and TNF-α-mediated Fas induction in microglia, respectively. In HeLa cells, CCAAT/enhancer-binding protein β (C/EBPβ) is responsible for influenza virus-infection mediated Fas up-regulation (44). In addition, NF-κB and C/EBP consensus elements are suggested to be involved in rat Fas expression (45). Recently, STAT-1α involvement in IFN-γ-induced Fas expression in the human colon carcinoma cell line HT29 was reported (35). To date, the transcriptional regulatory mechanism of Fas expression by TNF-α has not been reported. Within a 1-kb region of the mouse Fas promoter, five NF-κB consensus elements have been found (46). Further investigation of which NF-κB element(s) is critical for TNF-α induction of Fas and which isoforms of NF-κB are induced to bind to the elements is warranted. IFN-γ activation of STAT-1α leads to STAT-1α binding to IFN-γ activation sequence (GAS) elements in the promoter regions of IFN-γ inducible genes (for review, see Ref. 36). Interestingly, a GAS element is not present in the 1-kb promoter region of the mouse Fas gene (46). Considering the relatively late induction of Fas mRNA by IFN-γ, it is possible that additional transcription factors are induced/activated by STAT-1α to ultimately transcribe Fas. As well, a GAS element may be present in other regions of the Fas promoter.

In our study of the function of Fas expression on microglial cells, we observed that incubation of microglia with FasL-bearing cells induced cell death. This Fas-mediated microglial cell death was further increased by TNF-α or IFN-γ stimulation, consistent with a previous report (26). Interestingly, microglial cell death was not observed when cells were incubated with anti-Fas Ab alone or with anti-hamster IgG Ab as a secondary Ab (data not shown). These results suggest that engagement of Fas by membrane-bound FasL is necessary for subsequent cell death. Although the 51Cr release assay used in our study does not differentiate cell death through apoptosis vs necrosis, in conjunction with data from previous reports (16, 26), our data argue that activated microglia undergo Fas-mediated apoptosis by FasL-bearing cells.

Furthermore, we have found that TGF-β exerts a strong antiapoptotic effect by inhibiting surface Fas expression on microglia. TGF-β has differential inhibitory mechanisms on the apoptotic process depending on the cell type under study. TGF-β protects dendritic cell precursors from apoptosis by reducing Fas expression (47). In contrast, TGF-β decreases apoptosis of human T cells by inhibiting FasL expression, with no apparent effect on Fas expression (48). These findings in conjunction with the results from this study indicate that TGF-β uses different pathways to inhibit the apoptotic process. It has been reported that TNF-α or IFN-γ potentiate microglial cell death partly by down-regulating antiapoptotic molecules, such as Bcl-2 and Bcl-xL (26). In this regard, we tested the effect of TGF-β on mRNA expression of Bcl family members, including antiapoptotic members (Bcl-w, Bcl-xL, and Bcl-2) as well as proapoptotic members (Bak, Bax, and Bad). In these experiments TGF-β slightly down-regulated mRNA expression of all of the above-mentioned Bcl family members (data not shown), indicating that TGF-β-mediated inhibition of microglial cell death does not involve down-regulation of proapoptotic molecules or up-regulation of antiapoptotic molecules. Rather, TGF-β inhibits microglial cell death by inhibiting surface Fas expression.

TGF-β is a strong immunosuppressive cytokine, and its expression is elevated in various diseases, including MS (49). In the CNS, astrocytes, microglia, and oligodendrocytes can be activated to express TGF-β (50, 51, 52). In EAE, TGF-β expression is prominent during the recovery phase (53) and has been proposed to contribute to disease recovery (54, 55). The immunosuppressive function of TGF-β on microglia has been reported in our laboratory as well as others. TGF-β inhibits IFN-γ-induced class II MHC and CD40 expression by microglia (27, 56, 57). As well, TGF-β inhibits TNF-α, IL-1, IL-6, and IL-12 production by microglia (56, 58). These results in conjunction with our findings that TGF-β inhibits cell death of microglia suggest that TGF-β expression at the recovery phase of EAE can protect microglia from cell death as well as inhibit their ability to function as an APC within the CNS. Given that microglia can exert protective effects in the CNS, such as neurotropin production (59, 60) and regeneration of neurons (61), preservation of these cells may be viewed as beneficial at certain stages of disease.

Compared with microglia, the function of Fas molecules on astrocytes has been controversial. In one report mouse astrocytes were shown to be susceptible to Fas-mediated apoptosis in vitro (17). Apoptotic astrocytes have been detected in the white matter of Theiler’s virus-infected mice as well (62). However, no apoptotic astrocytes have been detected in brains from EAE-induced mice (16). In our study we could not detect astrocyte cell death based on the 51Cr release assay (data not shown). It is not clear why astrocytes, which express high constitutive levels of Fas, are resistant to Fas-mediated cell death. Recently, it was proposed that susceptibility to Fas-mediated apoptosis of human astrocytes varies depending on the cell passage state and also on microenvironmental factors, such as the presence of cytokines (25). In that study the resistance of early passage fetal astrocytes to Fas-mediated apoptosis correlated with the expression of Fas-associated phosphatase-1 (FAP-1), a tyrosine phosphatase interacting with the C-terminal region of Fas (63). FAP-1 inhibits Fas-mediated apoptosis upon overexpression and is highly expressed in tissues and cell lines that are resistant to Fas-mediated apoptosis (64). Interestingly, we detected mRNA expression of FAP, a mouse homologue of human FAP-1, in mouse astrocytes but not in microglia (data not shown), which may explain the relative resistance of astrocytes to Fas-mediated cell death compared with microglia.

Based on our in vitro data as well as the in vivo data reported by Bonetti et al. (16), we argue that mouse astrocytes do not undergo Fas-mediated apoptosis during EAE pathogenesis. To the contrary, we propose that Fas expressed on astrocytes can facilitate inflammatory responses by inducing the expression of chemokines such as MIP-1α, MIP-1β, and MIP-2. MIP-1α and -β are members of the C-C chemokine family that function as chemoattractants primarily for monocytes and T lymphocytes (for review, see Ref. 65). MIP-2 is a functional mouse homologue of human IL-8, a potent chemoattractant for neutrophils (for review, see Ref. 65). The importance of chemokine expression in EAE induction has been previously reported (for review, see Ref. 65). In the mouse model of EAE, production of MIP-1α in the CNS correlated with development of severe clinical disease, and administration of anti-MIP-1α Ab inhibited EAE induction (66). Similarly, MIP-1β expression was detected in rat brain upon EAE induction (67). It has been noted that CNS infiltration of activated immune cells is ameliorated in EAE-induced lpr mice (14). Considering the critical role of chemokines in recruitment of activated immune cells to the CNS, it is quite possible that the lack of Fas-mediated induction of chemokines by astrocytes is partly responsible for the resistance of lpr mice to EAE induction. Furthermore, we found that Fas-mediated signaling exerts a synergistic effect with TNF-α for MIP-1β mRNA expression. This suggests that during EAE pathogenesis, when TNF-α expression in the CNS is prominent, signaling through Fas on astrocytes can enhance MIP-1β expression. In our study we could not detect MIP-1β expression at the protein level due to the lack of a sensitive mouse MIP-1β detection system. However, in another study we have found that Ab cross-linking of Fas on human astrocytes can induce inflammatory chemokine and cytokine expression at both the mRNA and protein levels (C. Choi, X. Xu, J. W. Oh, S. J. Lee, and E. N. Benveniste, manuscript in preparation). To date, CNS-infiltrating activated T cells, microglia, and astrocytes have been reported to express FasL (16, 68, 69). However, it is not clear which cell type(s) contributes to the ongoing inflammatory response by engaging Fas on astrocytes in vivo. A preliminary analysis of the signaling cascades activated by Fas ligation on astrocytes suggests that the MAPK pathway is used, based on the fact that U0126, a specific inhibitor of MEK, abrogates Fas-induced chemokine expression. It has been reported that Daxx is involved in Fas-induced c-Jun N-terminal kinase activation (70). Activation of MAP kinase kinase kinase-1 by caspase-3 upon Fas ligation was documented (71). However, Fas-induced MEK activation has not been reported. In this regard, it will be interesting to test whether Daxx and MAP kinase kinase kinase-1 are involved in Fas-mediated MEK activation in astrocytes. Future studies on the signaling pathways activated through Fas on astrocytes may enable us to interrupt Fas-facilitated inflammatory responses in the CNS.

We thank Evelyn Rogers for excellent secretarial assistance, Dr. David E. Levy (New York University, NY) for providing STAT-1α deficient mice, and Dr. William Walker (St. Jude Children’s Research Hospital, Memphis, TN) for the EOC13 cell line.

1

This work was supported in part by National Institutes of Health Grants MH55795, NS36765, and NS29719 (to E.N.B.). We acknowledge the support of the University of Alabama at Birmingham Flow Cytometry Core Facility (Grant AM20614).

3

Abbreviations used in this paper: FasL, Fas ligand; DISC, death-inducing signaling complex; EAE, experimental allergic encephalomyelitis; MIP, macrophage inflammatory protein; MS, multiple sclerosis; MAPK, mitogen-activated protein kinase; RPA, RNase protection assay; ERK, extracellular signal-related kinase; MEK, MAP/ERK kinase; FAP-1, Fas-associated phosphatase-1.

1
Itoh, N., S. Yonehara, A. Ishii, M. Yonehara, S. Mizushima, M. Sameshima, A. Hase, Y. Seto, S. Nagata.
1991
. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis.
Cell
66
:
233
2
Watanabe-Fukunaga, R., C. I. Brannan, N. Itoh, S. Yonehara, N. G. Copeland, N. A. Jenkins, S. Nagata.
1992
. The cDNA structure, expression, and chromosomal assignment of the mouse Fas antigen.
J. Immunol.
148
:
1274
3
Kischkel, F. C., S. Hellbardt, I. Behrmann, M. Germer, M. Pawlita, P. H. Krammer, M. E. Peter.
1995
. Cytotoxicity-dependent APO-1 (Fas/CD95)-associated proteins form a death-inducing signaling complex (DISC) with the receptor.
EMBO J.
14
:
5579
4
Medema, J. P., C. Scaffidi, F. C. Kischkel, A. Shevchenko, M. Mann, P. H. Krammer, M. E. Peter.
1997
. FLICE is activated by association with the CD95 death-inducing signaling complex (DISC).
EMBO J.
16
:
2794
5
Suda, T., T. Takahashi, P. Golstein, S. Nagata.
1993
. Molecular cloning and expression of the Fas ligand, a novel member of the tumor necrosis factor family.
Cell
75
:
1169
6
Dheln, J., H. Walczak, C. Bäumier, K.-M. Debatin, P. H. Krammer.
1995
. Autocrine T-cell suicide mediated by APO-1/(Fas/CD95).
Nature
373
:
438
7
Crispe, I. N..
1994
. Fatal interactions: Fas-induced apoptosis of mature T cells.
Immunity
1
:
347
8
Singer, G. G., A. K. Abbas.
1994
. The Fas antigen is involved in peripheral but not thymic deletion of T lymphocytes in T cell receptor transgenic mice.
Immunity
1
:
365
9
Daniel, P. T., P. H. Krammer.
1994
. Activation induces sensitivity toward APO-1 (CD95)-mediated apoptosis in human B cells.
J. Immunol.
152
:
5624
10
Ju, S.-T., D. J. Panka, H. Cui, R. Ettinger, M. El-Khatib, D. H. Sherr, B. Z. Stanger, A. Marshak-Rothstein.
1995
. Fas(CS95)/FasL interactions required for programmed cell death after T-cell activation.
Nature
373
:
444
11
Griffith, T. S., T. Brunner, S. M. Fletcher, D. R. Green, T. A. Ferguson.
1995
. Fas ligand-induced apoptosis as a mechanism of immune privilege.
Science
270
:
1189
12
Bellgrau, D., D. Gold, H. Selawry, J. Moore, A. Franzusoff, R. C. Duke.
1995
. A role for CD95 ligand in preventing graft rejection.
Nature
377
:
630
13
Cohen, P. L., R. A. Eisenberg.
1991
.
lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol.
9
:
243
14
Waldner, H., R. A. Sobel, E. Howard, V. K. Kuchroo.
1997
. Fas- and FasL-deficient mice are resistant to induction of autoimmune encephalomyelitis.
J. Immunol.
159
:
3100
15
Sabelko, K. A., K. A. Kelly, M. H. Nahm, A. H. Cross, J. H. Russell.
1997
. Fas and fas ligand enhance the pathogenesis of experimental allergic encephalomyelitis, but are not essential for immune privilege in the central nervous system.
J. Immunol.
159
:
3096
16
Bonetti, B., J. Pohl, Y. L. Gao, C. S. Raine.
1997
. Cell death during autoimmune demyelination: effector but not target cells are eliminated by apoptosis.
J. Immunol.
159
:
5733
17
Malipiero, U., K. Frei, K.-S. Spanaus, C. Agresti, H. Lassmann, M. Hahne, J. Tschopp, H.-P. Eugster, A. Fontana.
1997
. Myelin oligodendrocyte glycoprotein-induced autoimmune encephalomyelitis is chronic/relapsing in perforin knockout mice, but monophasic in Fas- and Fas ligand-deficient lpr and gld mice.
Eur. J. Immunol.
27
:
3151
18
Sabelko-Downes, K. A., A. H. Cross, J. H. Russell.
1999
. Dual role for Fas ligand in the initiation of and recovery from experimental allergic encephalomyelitis.
J. Exp. Med.
189
:
1195
19
Dittel, B. N., R. M. Merchant, C. A. J. Janeway.
1999
. Evidence for Fas-dependent and Fas-independent mechanisms in the pathogenesis of experimental autoimmune encephalomyelitis.
J. Immunol.
162
:
6392
20
D’Souza, S. D., B. Bonetti, V. Balasingam, N. R. Cashman, P. A. Barker, A. B. Troutt, C. S. Raine, J. P. Antel.
1996
. Multiple sclerosis: Fas signaling in oligodendrocyte cell death.
J. Exp. Med.
184
:
2361
21
Sun, D., J. N. Whitaker, L. Cao, Q. Han, S. Sun, C. Coleclough, J. Mountz, T. Zhou.
1998
. Cell death mediated by Fas-FasL interaction between glial cells and MBP-reactive T cells.
J. Neurosci. Res.
52
:
458
22
Thilenius, A. R., K. A. Sabelko-Downes, J. H. Russell.
1999
. The role of the antigen-presenting cell in Fas-mediated direct and bystander killing: potential in vivo function of Fas in experimental allergic encephalomyelitis.
J. Immunol.
162
:
643
23
Becher, B., S. D. D’Souza, A. B. Troutt, J. P. Antel.
1998
. FAS expression on human fetal astrocytes without susceptibility to FAS-mediated cytotoxicity.
Neuroscience
84
:
627
24
Choi, C., J.-Y. Park, J. Lee, J.-H. Lim, E.-C. Shin, Y. S. Ahn, C.-H. Kim, S.-J. Kim, J. D. Kim, I. S. Choi, et al
1999
. Fas ligand and Fas are expressed constitutively in human astrocytes and the expression increases with IL-1, IL-6, TNF-α or IFN-γ.
J. Immunol.
162
:
1889
25
Saas, P., J. Boucraut, A.-L. Quiquerez, V. Schnuriger, G. Perrin, S. Desplat-Jego, D. Bernard, P. R. Walker, P.-Y. Dietrich.
1999
. CD95 (Fas/Apo-1) as a receptor governing astrocyte apoptotic or inflammatory responses: a key role in brain inflammation?.
J. Immunol.
162
:
2326
26
Spanaus, K. S., R. Schlapbach, A. Fontana.
1998
. TNF-α and IFN-γ render microglia sensitive to Fas ligand-induced apoptosis by induction of Fas expression and down-regulation of Bcl-2 and Bcl-xL.
Eur. J. Immunol.
28
:
4398
27
Nguyen, V., W. S. Walker, E. N. Benveniste.
1998
. Post-transcriptional inhibition of CD40 gene expression in microglia by transforming growth factor-β.
Eur. J. Immunol.
28
:
2537
28
Durbin, J. E., R. Hackenmiller, M. C. Simon, D. E. Levy.
1996
. Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease.
Cell
84
:
443
29
Walker, W. S., J. Gatewood, E. Olivas, D. Askew, C. E. G. Havenith.
1995
. Mouse microglial cell lines differing in constitutive and interferon-γ-inducible antigen-presenting activities for naive and memory CD4+ and CD8+ T cells.
J. Neuroimmunol.
63
:
163
30
Wu, J., T. Zhou, J. Zhang, J. He, W. C. Gause, J. D. Mountz.
1994
. Correction of accelerated autoimmune disease by early replacement of the mutated lpr gene with the normal Fas apoptosis gene in the T cells of transgenic MRL-lpr/lpr mice.
Proc. Natl. Acad. Sci. USA
91
:
2344
31
Lee, S. J., J. Y. Park, J. Hou, E. N. Benveniste.
1999
. Transcriptional regulation of the intercellular adhesion molecule-1 gene by proinflammatory cytokines in human astrocytes.
GLIA
25
:
21
32
Oh, J.-W., L. M. Schwiebert, E. N. Benveniste.
1999
. Cytokine regulation of CC and CXC chemokine expression by human astrocytes.
J. Neurovirol.
5
:
82
33
Zhang, H.-G., D. Liu, Y. Heike, P. Yang, Z. Wang, X. Wang, D. T. Curiel, T. Zhou, J. D. Mountz.
1998
. Induction of specific T-cells tolerance by adenovirus-transfected, Fas ligand-producing antigen presenting cells.
Nat. Biotechnol.
16
:
1045
34
Luttmann, W., A. Opfer, E. Dauer, M. Foerster, H. Matthys, H. Eibel, K. Schulze-Osthoff, C. Kroegel, J. C. Virchow.
1998
. Differential regulation of CD95 (Fas/APO-1) expression in human blood eosinophils.
Eur. J. Immunol.
28
:
2057
35
Xu, X., X.-Y. Fu, J. Plate, A. S.-F. Chong.
1998
. IFN-γ induces cell growth inhibition by Fas-mediated apoptosis: requirement of STAT1 protein for up-regulation of Fas and FasL expression.
Cancer Res.
58
:
2832
36
Darnell, J. E., Jr.
1997
. STATs and gene regulation.
Science
277
:
1630
37
Baldwin, Jr. A. S. 1996. The NF-κB and IκB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649.
38
Lin, Y. Z., S. Y. Yao, R. A. Veach, T. R. Torgerson, J. Hawiger.
1995
. Inhibition of nuclear translocation of transcription factor NF-κB by a synthetic peptide containing a cell membrane-permeable motif and nuclear localization sequence.
J. Biol. Chem.
270
:
14255
39
Nagata, S., P. Golstein.
1995
. The Fas death factor.
Science
267
:
1449
40
Chen, J.-J., Y. Sun, G. J. Nabel.
1998
. Regulation of the proinflammatory effects of Fas ligand (CD95L).
Science
282
:
1714
41
Baker, D., J. K. O’Neill, J. L. Turk.
1991
. Cytokines in the central nervous system of mice during chronic relapsing experimental allergic encephalomyelitis.
Cell. Immunol.
134
:
505
42
Favata, M. F., K. Y. Horiuchi, E. J. Manos, A. J. Daulerio, D. A. Stradley, W. S. Feeser, D. E. Van Dyk, W. J. Pitts, R. A. Earl, F. Hobbs, et al
1998
. Identification of a novel inhibitor of mitogen-activated protein kinase kinase.
J. Biol. Chem.
273
:
18623
43
Crews, C. M., A. Alessandrini, R. L. Erikson.
1992
. The primary structure of MEK, a protein kinase that phosphorylates the ERK gene product.
Science
258
:
478
44
Wada, N., M. Matsumura, Y. Ohba, N. Kobayashi, T. Takizawa, Y. Nakanishi.
1995
. Transcription stimulation of the Fas-encoding gene by nuclear factor for interleukin-6 expression upon influenza virus infection.
J. Biol. Chem.
270
:
18007
45
Kimura, K., K. Asami, M. Yamamoto.
1997
. Structure of the promoter for the rat Fas antigen gene.
Biochim. Biophys. Acta
1352
:
238
46
Pang, H., K. Miranda, A. Fine.
1998
. Sp3 regulates Fas expression in lung epithelial cells.
Biochem. J.
333
:
209
47
Riedl, E., H. Strobl, O. Majdic, W. Knapp.
1997
. TGF-β1 promotes in vitro generation of dendritic cells by protecting progenitor cells from apoptosis.
J. Immunol.
158
:
1591
48
Genestier, L., S. Kasibhatla, T. Brunner, D. R. Green.
1999
. Transforming growth factor β1 inhibits Fas ligand expression and subsequent activation-induced cell death in T cells via downregulation of c-Myc.
J. Exp. Med.
189
:
231
49
Beck, J., P. Rondot, P. Jullien, J. Wietzerbin, D. A. Lawrence.
1991
. TGF-β-like activity produced during regression of exacerbations in multiple sclerosis.
Acta Neurol. Scand.
84
:
452
50
Morganti-Kossmann, M. C., T. Kossmann, M. E. Brandes, S. E. Mergenhagen, S. M. Wahl.
1992
. Autocrine and paracrine regulation of astrocyte function by transforming growth factor-β.
J. Neuroimmunol.
39
:
163
51
da Cunha, A., L. Vitkovic.
1992
. Transforming growth factor-β (TGF-β1) expression and regulation in rat cortical astrocytes.
J. Neuroimmunol.
36
:
157
52
da Cunha, A., J. A. Jefferson, R. W. Jackson, L. Vitkovic.
1993
. Glial cell-specific mechanisms of TGF-β1 induction by IL-1 in cerebral cortex.
J. Neuroimmunol.
42
:
71
53
Kiefer, R., T. Schweitzer, S. Jung, K. V. Toyka, H.-P. Hartung.
1998
. Sequential expression of transforming growth factor-β1 by T-cells, macrophages, and microglia in rat spinal cord during autoimmune inflammation.
J. Neuropathol. Exp. Neurol.
57
:
385
54
Racke, M. K., B. Cannella, P. Albert, M. Sporn, C. S. Raine, D. E. McFarlin.
1992
. Evidence of endogenous regulatory function of transforming growth factor-β1 in experimental allergic encephalomyelitis.
Int. Immunol.
4
:
615
55
Kuruvilla, A. P., R. Shah, G. M. Hochwald, H. D. Liggitt, M. A. Palladino, G. J. Thorbecke.
1991
. Protective effect of transforming growth factor β1 on experimental autoimmune diseases in mice.
Proc. Natl. Acad. Sci. USA
88
:
2918
56
Suzumura, A., M. Sawada, H. Yamamoto, T. Marunouchi.
1993
. Transforming growth factor-β suppresses activation and proliferation of microglia in vitro.
J. Immunol.
151
:
2150
57
O’Keefe, G. M., V. T. Nguyen, E. N. Benveniste.
1999
. Class II transactivator and class II MHC gene expression in microglia: modulation by the cytokines TGF-β, IL-4, IL-13, and IL-10.
Eur. J. Immunol.
29
:
1275
58
Lodge, P. A., S. Sriram.
1996
. Regulation of microglial activation by TGF-β, IL-10, and CSF-1.
J. Leukocyte Biol.
60
:
502
59
Elkabes, S., E. M. DiCicco-Bloom, I. B. Black.
1996
. Brain microglia/macrophages express neurotrophins that selectively regulate microglial proliferation and function.
J. Neurosci.
16
:
2508
60
Elkabes, S., L. Peng, I. R. Black.
1998
. Lipopolysaccharide differentially regulates microglial Trk receptors and neurotrophin expression.
J. Neurosci. Res.
54
:
117
61
Lazarov-Spiegler, O., A. S. Solomon, A. B. Zeev-Brann, D. L. Hirschberg, V. Lavie, M. Schwartz.
1996
. Transplantation of activated macrophages overcomes central nervous system regrowth failure.
FASEB J.
10
:
1296
62
Palma, J. P., R. L. Yauch, S. Lang, B. S. Kim.
1999
. Potential role of CD4+ T cell-mediated apoptosis of activated astrocytes in Theiler’s virus-induced demyelination.
J. Immunol.
162
:
6543
63
Sato, T., S. Irie, S. Kitada, J. C. Reed.
1995
. FAP-1: a protein tyrosine phosphatase that associates with Fas.
Science
268
:
411
64
Yanagisawa, J., M. Takahashi, H. Kanki, H. Yano-Yanagisawa, T. Tazunoki, E. Sawa, T. Nishitoba, M. Kamishohara, E. Kobayashi, S. Kataoka, et al
1997
. The molecular interaction of Fas and FAP-1: a tripeptide blocker of human Fas interaction with FAP-1 promotes Fas-induced apoptosis.
J. Biol. Chem.
272
:
8539
65
Ransohoff, R. M..
1997
. Chemokines in neurological disease models: correlation between chemokine expression patterns and inflammatory pathology.
J. Leukocyte Biol.
62
:
645
66
Karpus, W. J., N. W. Lukacs, B. L. McRae, R. M. Strieter, S. L. Kunkel, S. D. Miller.
1995
. An important role for the chemokine macrophage inflammatory protein-1α in the pathogenesis of the T cell-mediated autoimmune disease, experimental autoimmune encephalomyelitis.
J. Immunol.
155
:
5003
67
Miyagishi, R., S. Kikuchi, C. Takayama, Y. Inoue, K. Tashiro.
1997
. Identification of cell types producing RANTES, MIP-1α and MIP-1β in rat experimental autoimmune encephalomyelitis by in situ hybridization.
J. Neuroimmunol.
77
:
17
68
White, C. A., P. A. McCombe, M. P. Pender.
1998
. The roles of Fas, Fas ligand and Bcl-2 in T cell apoptosis in the central nervous system in experimental autoimmune encephalomyelitis.
J. Neuroimmunol.
82
:
47
69
Bechman, I., G. Mor, J. Nilsen, M. Eliza, R. Nitsch, F. Naftolin.
1999
. FasL (CD95L, Apo1L) is expressed in the normal rat and human brain: evidence for the existence of an immunological brain barrier.
GLIA
27
:
62
70
Yang, X., R. Khosravi-Far, H. Y. Chang, D. Baltimore.
1997
. Daxx, a novel Fas-binding protein that activates JNK and apoptosis.
Cell
98
:
1067
71
Deak, J. C., J. V. Cross, M. Lewis, Y. Qian, L. A. Parrott, C. W. Distelhorst, D. J. Templeton.
1998
. Fas-induced proteolytic activation and intracellular redistribution of the stress-signaling kinase MEKK1.
Proc. Natl. Acad. Sci. USA
95
:
5595