CD45 Inhibits CD40L-induced Microglial Activation via Negative Regulation of the Src/p44/42 MAPK Pathway*

It has been reported that ligation of CD40 with CD40 ligand (CD40L) results in microglial activation as evidenced by p44/42 mitogen-activated protein kinase (MAPK) dependent tumor necrosis factor α (TNF-α) production. Previous studies have shown that CD45, a functional transmembrane protein-tyrosine phosphatase, is constitutively expressed at moderate levels on microglial cells and this expression is greatly elevated on activated microglia. To investigate the possibility that CD45 might modulate CD40L-induced microglial activation, we treated primary cultured microglial cells with CD40L and anti-CD45 antibody. Data show that cross-linking of CD45 markedly inhibits CD40L-induced activity of the Src family kinases Lck and Lyn. Further, co-treatment of microglia with CD40L and anti-CD45 antibody results in significant inhibition of microglial TNF-α production through inhibition of p44/42 MAPK activity, a downstream signaling event resulting from Src activation. Accordingly, primary cultured microglial cells from mice deficient in CD45 demonstrate hyper-responsiveness to ligation of CD40, as evidenced by increased p44/42 MAPK activation and TNF-α production. Taken together, these results show that CD45 plays a novel role in suppressing CD40L-induced microglial activation via negative regulation of the Src/p44/42 MAPK cascade.

It has been reported that ligation of CD40 with CD40 ligand (CD40L) results in microglial activation as evidenced by p44/42 mitogen-activated protein kinase (MAPK) dependent tumor necrosis factor ␣ (TNF-␣) production. Previous studies have shown that CD45, a functional transmembrane protein-tyrosine phosphatase, is constitutively expressed at moderate levels on microglial cells and this expression is greatly elevated on activated microglia. To investigate the possibility that CD45 might modulate CD40L-induced microglial activation, we treated primary cultured microglial cells with CD40L and anti-CD45 antibody. Data show that crosslinking of CD45 markedly inhibits CD40L-induced activity of the Src family kinases Lck and Lyn. Further, cotreatment of microglia with CD40L and anti-CD45 antibody results in significant inhibition of microglial TNF-␣ production through inhibition of p44/42 MAPK activity, a downstream signaling event resulting from Src activation. Accordingly, primary cultured microglial cells from mice deficient in CD45 demonstrate hyper-responsiveness to ligation of CD40, as evidenced by increased p44/42 MAPK activation and TNF-␣ production. Taken together, these results show that CD45 plays a novel role in suppressing CD40L-induced microglial activation via negative regulation of the Src/p44/42 MAPK cascade.
Microglial activation, which is characterized by transformation of microglia from a ramified to a reactive phenotype exhibiting neurotoxic properties, has been implicated as pathological in a variety of neurodegenerative diseases, including Alzheimer's disease (AD), 1 Creutzfeld-Jacob disease, and multiple sclerosis (MS) (1). As central nervous system-resident professional macrophages, activated microglia produce and secrete potentially neurotoxic pro-inflammatory cytokines including interleukin 1␤ and tumor necrosis factor ␣ (TNF-␣) (2), both of which have been shown to promote neuronal injury (3)(4)(5). Microglial activation is also associated with an increased expression of cell surface molecules, including CD45, major histocompatibility complex class II antigens, protein complement receptors such as CR4 and membrane attack complex 1, and the immunoglobulin receptors Fc␥RI and Fc␥RII (2,6,7). Additionally, we have recently shown that microglial activation resulting from stimulation with Alzheimer's ␤-amyloid peptides and CD40 ligand (CD40L) results in increased CD40 expression on microglia with resultant TNF-␣ secretion by these cells (8).
Intracellularly, microglial activation induced by a variety of stimuli including CD40L, lipopolysaccharide (LPS), ␤-amyloid peptides, and prion, has been shown to involve activation of the mitogen-activated protein kinase (MAPK) module, ultimately leading to production of neurotoxic products by these cells (9,10). Additionally, it has been shown that members of the Src family, including the tyrosine kinase Lyn (10), regulate activation of MAPK in these cells. Similar regulation of MAPK by Src occurs in T cells following mitogenic stimulation with interleukin-18 and anti-CD3 antibody, where the activated Src family member Lck has been shown to associate with and promote activation of MAPK (11). Yet, in microglial cells, the role of cell surface receptors in regulation of this intracellular Src/MAPK cascade has been largely unexplored.
CD45 is a membrane-bound protein-tyrosine phosphatase (PTP), which is expressed on a variety of immune cells, including T and B lymphocytes, where it has been shown to play a critical role in negative regulation of cellular activation (12). In addition, CD45 is expressed on microglia at low to moderate levels, and is markedly increased following activation of these cells (13,14). It is generally thought that CD45 couples to Src family kinases, functioning to maintain Src in a dephosphorylated, and hence inactive, state (12). This is supported by studies in T and B lymphocytes, where CD45-deficient cell lines demonstrate increased Src activity (15)(16)(17)(18). Yet, the mechanism of CD45 modulation of Src activity is complex, and it is thought that CD45 might function as both a positive and negative regulator of Src in a site-specific manner (19).
CD40 is a 45-50-kDa receptor, which is a member of the TNF receptor superfamily and is expressed on a wide range of both immune and non-immune cell types, including dendritic cells, monocytes, macrophages, fibroblasts, endothelial cells, and smooth muscle cells (20,21). The CD40 pathway was initially shown to play a critical role in the humoral and cellular immune response, as ligation of B cell CD40 induces B cell proliferation and differentiation into antibody-secreting plasma cells (20), and the action of T helper1 cells in priming of cytotoxic T lymphocytes is mediated by CD40-CD40L interactions (22). Recently, we and others have shown that CD40 is constitutively expressed at low levels on microglia (N9 cells and murine primary culture; Refs. 5, 14, 23, and 24), and ligation of microglial CD40 by CD40L leads to marked TNF-␣ secretion by these cells, which is neurotoxic at such levels (5). CD40 signal-ing in T cells has been shown to be dependent on interaction between CD40 and Src family kinases, in particular Lck (26,27), and we have recently shown that the CD40-CD40L interaction on microglia leads to activation of p44/42 MAPK in these cells (9). Based on the idea that stimulation of CD45 might oppose the effects of CD40 ligation (28), we wished to evaluate the effects of cross-linking CD45 in the presence of CD40L on microglial activation. Specifically, we wished to determine the possible involvement of the Src/MAPK cascade as an early signaling event in mediating this effect. We were particularly interested in searching for putative negative regulators of CD40-mediated microglial activation, as we have previously shown both in vitro and in vivo in a mouse model of AD that stimulation of this pathway results in exacerbation of microglial-mediated AD-like pathology (8). Therefore, the identification of a molecule that could oppose this effect may provide a molecular target for the treatment of neurodegenerative diseases with a reactive microglial component, such as AD.
In this study, we show that cross-linking of CD45 markedly inhibits p44/42 MAPK-dependent TNF-␣ production induced by CD40 ligation in murine primary culture microglia. Furthermore, we also provide evidence that cross-linking of CD45 opposes these effects through inhibiting CD40L-induced activation of Src family kinases, particularly Lck and Lyn. Finally, we demonstrate that primary culture microglia, which are deficient for CD45, are hyperresponsive to CD40 ligation, leading to marked p44/42 MAPK activation and TNF-␣ secretion. Taken together, our data show that CD45 plays a novel role in mitigating CD40L-induced microglial activation via negatively regulating the Src/p44/42 MAPK cascade, suggesting that CD45 might be a potential therapeutic target for the suppression of microglial activation associated with neurodegenerative diseases such as AD and MS.

EXPERIMENTAL PROCEDURES
Reagents-Monoclonal antibodies (purified rat anti-mouse CD45 and purified rat IgG 2b control antibodies; fluorescein isothiocyanate-conjugated rat anti-mouse CD45 and fluorescein isothiocyanate-conjugated rat IgG 2b control antibodies) were purchased from PharMingen (San Diego, CA). Antibodies for phospho-p44/42 MAPK (Thr 202 /Tyr 204 ), and total p44/42 MAPK were obtained from New England Biolabs (Beverly, MA). TNF-␣ antibody for Western blotting was obtained from R&D Systems (Minneapolis, MN). Human soluble recombinant CD40L protein was obtained from Alexis Biochemicals (San Diego, CA). The CD45 phosphatase activity assay kit was purchased from Biomol (Plymouth Meeting, PA). The anti-mouse alkaline phosphatase-conjugated IgG secondary antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Immun-Blot polyvinylidene difluoride membranes and the Immun-Star chemiluminescence substrate were purchased from Bio-Rad.
Murine Primary Cell Culture-Breeding pairs of BALB/c, CD45deficient (29) and CD40-deficient (30) mice were purchased from Jackson Laboratory (Bar Harbor, ME) and housed in the animal facility at the University of South Florida Health Science Center. Murine primary culture microglia were isolated from mouse cerebral cortices and were grown in RPMI medium supplemented with 5% fetal calf serum, 2 mM glutamine, 100 units/ml penicillin, 0.1 g/ml streptomycin, and 0.05 mM 2-mercaptoethanol according to previously described methods (5,9). Briefly, cerebral cortices from newborn mice (1-2 days old) were isolated under sterile conditions and were kept at 4°C prior to mechanical dissociation. Cells were plated in 75-cm 2 flasks, and complete medium was added. Primary cultures were kept for 14 days so that only glial cells remained, and microglia were isolated by shaking flasks at 200 rpm in a Lab-Line Incubator-Shaker. More than 98% of these glial cells stained positive for membrane attack complex-1 (CD11b; Roche Molecular Biochemicals). Additionally, between 85% and 95% of microglial cells stained positive for CD45 by fluorescence-activated cell sorter (FACS) analysis as described previously (5), irrespective of CD40L and/or anti-CD45 antibody treatment (data not shown). To verify CD45 deficiency status, CD45 expression on microglia isolated from CD45deficient mice was also measured by FACS analysis, and CD45 was undetectable on these cells (data not shown). To verify CD40 deficiency status in microglia isolated from CD40 receptor-deficient mice, CD40 expression was measured by FACS analysis, and CD40 was undetectable on these cells, either before or after interferon-␥ stimulation (data not shown).
TNF-␣ ELISA-Primary cultured microglial cells were plated in 24-well tissue culture plates (Nunclon, Nalge Nunc International, Roskilde, Denmark) at 5 ϫ 10 4 cells/well and stimulated for 24 h with CD40L protein (1 g/ml) in the presence or absence of anti-CD45 mAb (1:200) or appropriate controls. In some experiments, microglial cells were pre-treated with PD 98059 (5 M, Calbiochem, La Jolla, CA) for 1 h and then incubated with CD40L protein for 24 h. Cell-free supernatants were collected and assayed for TNF-␣ by the DuoSet TNF-␣ ELISA kit (R&D Systems, Minneapolis, MN) in strict accordance with the manufacturer's instruction. The Bio-Rad protein assay was performed to measure total cellular protein from each of the cell groups under consideration just prior to quantification of cytokine release by ELISA.
Western Immunoblotting-Murine primary culture microglia were plated in six-well tissue culture plates (Nunclon) at a density of 8 ϫ 10 5 cells/well. Cells were then incubated for 30 min (for examining p44/p42 MAPK) or 24 h (for detecting TNF-␣ protein and CD40 expression) with or without CD40L protein (1 g/ml) in the presence or absence of anti-CD45 mAb, control antibodies (1:200 dilution for each), or Src inhibitors (damnacanthal, 1000 nM; PP1, 1000 nM; obtained from Calbiochem, San Diego, CA) or appropriate controls. Immediately following culturing, microglia were washed in ice-cold phosphate-buffered saline (PBS) three times, scraped into ice-cold PBS, and lysed in an ice-cold lysis buffer containing 20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerol phosphate, 1 mM Na 3 VO 4 , 1 g/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride. After incubating for 30 min on ice, samples were centrifuged at high speed for 15 min, and supernatants were collected. Total protein content was estimated using the Bio-Rad protein assay. An aliquot corresponding to 50 g of total protein of each sample was separated by SDS-polyacrylamide gel electrophoresis and transferred electrophoretically to Immun-Blot polyvinylidene difluoride membranes. Nonspecific antibody binding was blocked with 5% nonfat dry milk in Tris-buffered saline (20 mM Tris, 500 mM NaCl, pH 7.5) for 1 h at room temperature. Membranes where first hybridized with a phosphospecific p44/42 MAPK antibody or rat anti-mouse TNF-␣ monoclonal antibody, stripped with ␤-mercaptoethanol stripping solution (62.5 mM Tris-HCl, pH 6.8, 2% SDS, and 100 mM ␤-mercaptoethanol), and then re-probed with an antibody that recognizes total p44/42 MAPK (or actin, in the case of TNF-␣ Western immunoblots). Alternatively, membranes with identical samples were probed with either with a phosphospecific p44/42 MAPK antibody or with an antibody that recognizes total p44/42 MAPK. Immunoblotting was carried out with a primary antibody followed by an anti-mouse horseradish peroxidaseconjugated IgG secondary antibody as a tracer. The Immun-Star chemiluminescence substrate was used to develop the blots. Densitometric analysis was preformed for all blots using the Fluor-S MultiImager with Quantity One software (Bio-Rad).
Flow Cytometric Analysis-CD40 expression was assessed by FACS analysis. Primary cultured microglial cells were plated in six-well tissue culture plates (Nunclon) at 2 ϫ 10 5 cells/well and incubated with CD40L protein in the presence or absence of anti-CD45 mAb. Twentyfour hours after incubation, microglial cells (approximately 1 ϫ 10 6 cells) were re-suspended in 250 l of ice-cold PBS for FACS analysis, according to methods described previously (5). A minimum of 10,000 cells were accepted for FACS analysis. Cells were gated based on morphological characteristics such that apoptotic and necrotic cells were not accepted for FACS analysis. Percentages of positive cells (CD40-expressing) were calculated as follows; for each treatment, the mean fluorescence value for the isotype-matched control antibody was subtracted from the mean fluorescence value for the CD40-specific antibody.
Immune Complex Kinase Assay-Primary culture microglial cells were seeded in six-well tissue culture plates (Nunclon TM ) at 8 ϫ 10 5 cells/well. Thirty minutes after co-treatment with CD40L protein (1 g/ml) in the presence or absence of anti-CD45 mAb or appropriate controls, microglial cells were lysed in ice-cold lysis buffer (as described above). Immunoprecipitation was performed for the Elk1 fusion protein as described below. Total immunoprecipitates were quantified by the Bio-Rad protein assay, and an aliquot of 50 g of protein for each treatment condition was separated by SDS-polyacrylamide gel electrophoresis. Activity of p44/42 MAPK was determined using the p44/42 MAP kinase assay kit (New England Biolabs) in strict accordance with the manufacturer's instruction. The phosphorylated form of the Elk1 fusion protein was visualized by Western immunoblotting (as described above) using a specific antibody for phosphorylated Elk1 supplied with the kit.
Immunoprecipitation and Src Kinase Assay-Primary culture microglial cells were seeded at 10 ϫ 10 5 cells/dish in 100-mm cell culture dishes and incubated overnight to 80% confluence. The following day, cells were treated in the presence or absence of CD40L or anti-CD45 mAb for 30 min. Cells were then lysed in 200 l of cell lysis buffer as described above, and cell lysates were immunoprecipitated overnight at 4°C with either Lyn-or Lck-specific antibodies (1:50 dilution, polyclonal rabbit anti-Lyn or anti-Lck antibodies, PharMingen). Immunoprecipitates were then immobilized with 10 l of 50% protein A-Sepharose beads diluted in PBS (Protein A on Sepharose CL-4B, Sigma) for 3 h at 4°C. The resulting immobilized immunoprecipitates were pelleted and washed 2 ϫ in ice-cold cell lysis buffer, followed by an additional two washes in ice-cold kinase buffer (containing 25 mM Tris, pH 7.5, 5 mM ␤-glycerol phosphate, 2 mM dithiothreitol, 0.1 mM Na 3 VO 4 , and 10 mM MgCl 2 ), and pellets were re-suspended in 50 l of Src kinase reaction buffer (containing 100 mM Tris-HCl, pH 7.2, 125 mM MgCl 2 , 25 mM MnCl 2 , 2 mM EGTA, 0.25 mM Na 3 VO 4 , and 2 mM dithiothreitol). The Src kinase assay kit (Upstate Biotechnology, Inc., Lake Placid, NY) was used in accordance with the manufacturer's instruction for radioactive quantitation of immunoprecipitated Src activity based on incorporation of [␥-32 P]ATP into Src kinase substrate peptide (31). Radioactivity was measured using a 1209 Rackbeta liquid scintillation counter (LKB Wallac, Inc., Gaithersburg, MD), and data are reported as picomoles of PO 4 /min/mg of total cellular protein.
Statistical Analysis-Data were analyzed using analysis of variance (ANOVA) followed by post hoc comparisons of means by Bonferroni's or Dunnett's T3 method, where Levene's test for homogeneity of variances was used to determine the appropriate method of post hoc comparison. ␣ levels were set at 0.05 for each analysis. All analyses were performed using SPSS for Windows release 9.0.

Cross-linking of CD45 Results in Reduction of CD40L-induced Microglial TNF-␣ Production-
We have recently shown that ligation of CD40 by CD40L induces p44/42 MAPK-dependent TNF-␣ production in microglia (9). It has previously been shown that stimulation of the CD40 pathway results in T cell activation that is mediated by Src and MAPK activation (32). CD45 is a prototypical membrane-associated PTP, which maintains Src in a dephosphorylated state resulting in its decreased kinase activity (12). We wished to evaluate the possibility that stimulation of CD45 might mitigate microglial TNF-␣ production by decreasing Src and downstream MAPK activity induced by CD40 ligation. In order to evaluate whether cross-linking of microglial CD45 results in stimulation of this PTP, we measured free inorganic phosphate (P i ) in microglial cell lysates treated in the presence or absence of anti-CD45 mAb or isotype-matched control antibody, and find significantly higher levels of P i in anti-CD45 mAb-treated microglia compared with appropriate controls (data not shown). To investigate the possible functional significance of CD45 stimulation in the presence of CD40L, we co-treated primary culture microglia with monoclonal anti-CD45 mAb and CD40L for 24 h. Results show that secretion of TNF-␣ protein is markedly increased following treatment with CD40L, and these levels are dramatically reduced after co-treatment of these cells with anti-CD45 mAb (Fig. 1). Similar results were obtained by Western Blot (data not shown).
Cross-linking of CD45 in the Presence of CD40L Does Not Affect CD40 Expression-A previous study that focused on inhibition of CD40-mediated monocyte activation found that such effects could be accounted for, at least in part, by decreased CD40 receptor expression levels (33). Thus, we wished to determine whether our observed effect of inhibition of CD40Linduced microglial activation after cross-linking CD45 was dependent upon decreased CD40 expression. To rule out this possibility, we examined CD40 expression within 24 h after co-treatment with anti-CD45 mAb and CD40L. Data show that treatment of microglia with anti-CD45 mAb in the presence of CD40L does not affect CD40 expression compared with appropriate controls as measured by Western immunoblotting (data not shown) and FACS analysis (Fig. 2). These data also show that the observed effect of anti-CD45 mAb treatment on microglial activation does not involve modulation of CD40 expression levels across the 24-h time course examined. Interestingly, we find that treatment of microglia with CD40L alone results in a significant increase in CD40 receptor levels on microglia, supporting the idea that CD40L can positively regulate its receptor on microglia.
CD40L-induced Increased Activation of p44/42 MAPK Is Specific to the CD40-CD40L Interaction-We have previously shown that CD40L is able to stimulate microglial p44/42 MAPK in a time-dependent fashion, from 30 min to 240 min, with peak activation at 60 min (9). When taken together with the present data showing ϳ4% CD40 receptor expression on microglia, we sought to reconcile how such a low expression level of CD40 could mediate marked effects on increasing p44/42 MAPK phosphorylation and activity following CD40 ligation. Thus, we sought to address the possibility that interaction between CD40 ligand and a receptor other than CD40 may bring about these effects. To examine this possibility, we employed murine primary culture CD40 knockout microglia, and treated them with CD40L. Data show that CD40L is unable to elicit p44/24 MAPK phosphorylation (Fig. 3A) or activity (Fig. 3B) in these cells following stimulation with CD40L, showing that the microglial CD40-CD40L interaction markedly elicits p44/42 MAPK activity.
Microglial CD40L-induced p44/42 MAPK and TNF-␣ Production Are Dependent on Src Activation-It has previously been reported that cross-linking of CD40 by anti-CD40 mAb induces phosphorylation and activation of the Src family kinase Lyn in B cells (34). In addition, we and others have shown that ligation of CD40 results in TNF-␣ secretion that is brought about by activation of p44/42 MAPK in monocytes and microglial cells (9,35). These data led us to investigate the possibility that ligation of CD40 might result in activation of Src family kinases and consequent downstream activation of p44/42 MAPK, ultimately resulting in TNF-␣ secretion by microglia. Thus, we co-incubated microglial cells with CD40L and either a general inhibitor of Src family kinases, PP1 (1000 nM), or the Lck-specific inhibitor, damnacanthal (1000 nM), for 30 min. In order to confirm that these agents inhibited Src kinase activity FIG. 1. CD45 cross-linking results in decreased CD40L-induced microglial TNF-␣ production. Graph represents a summary of TNF-␣ release ELISA results (mean TNF-␣ pg/mg of total protein Ϯ 1 S.E.) with n ϭ 3 for each condition presented. ANOVA revealed significant main effects of CD40L (p Ͻ 0.001) and anti-CD45 (p Ͻ 0.01), and an interaction between them (p Ͻ 0.01). One-way ANOVA revealed significant between-groups differences (p Ͻ 0.001), and post hoc testing showed significant differences between control and CD40L (p Ͻ 0.001) and between CD40L/anti-CD45 and CD40L/control antibody (p Ͻ 0.01).
in our system, we first assayed activity of the Src family kinases Lck and Lyn after co-treatment of microglia with CD40L and either PP1 or damnacanthal. Results show that both Src inhibitors markedly reduce CD40L-induced Src kinase activity (data not shown). Activity of p44/42 MAPK was examined by Western blot and immune complex kinase assay using antibodies that specifically recognize phosphorylated p44/42 MAPK or the phosphorylated form of the Elk1 fusion protein, respectively. Data as shown in Fig. 4 (A and B) indicate that cotreatment of microglia with CD40L and either Src family kinase inhibitor results in marked reduction of p44/42 MAPK activity, suggesting that CD40L-induced activation of p44/42 MAPK is dependent on activity of Src family kinases. We then assessed whether or not PP1 and damnacanthal inhibition of CD40L-induced p44/42 MAPK phosphorylation and activity might be dose-dependent. Data indicate that this is the case, with p44/42 MAPK phosphorylation (Fig. 4C) and activity (Fig.  4D) decreasing with increasing doses of these inhibitors (from 200 nM to 1000 nM). Furthermore, a significant reduction of TNF-␣ was observed after co-treatment of microglia with CD40L and Src kinase inhibitors for 24 h, supporting the idea that CD40L-induced microglial activation is dependent upon activation of Src and downstream p44/42 MAPK (Fig. 4E).
Cross-linking of CD45 Inhibits Microglial CD40L-induced Lck and Lyn Kinase Activity-It is well known that CD45 is involved in negative regulation of activity of Src family kinases, particularly Lck and Lyn (12). Having shown that treatment of microglia with CD40L results in increased Src kinase activity, we wished to evaluate the possibility that CD45 could oppose this effect by decreasing Src kinase activity. To investigate this possibility, we co-treated microglia with CD40L and/or anti-CD45 mAb or appropriate controls for 30 min. Phosphotransferase activity of Lck and Lyn kinases was measured as described under "Experimental Procedures." Results indicate that cross-linking of CD45 markedly inhibits Lck (Fig. 5A) and Lyn (Fig. 5B) kinase activity induced by CD40 ligation, suggesting that microglial CD40 and CD45 signaling pathways cross-modulate each other at the level of membrane-associated Src family kinases.
Cross-linking of CD45 Suppresses CD40L-induced p44/42 MAPK Activity-It has been reported that Src kinases are involved in regulation of MAPK activation (10,11,36). We and others have shown that activation of MAPK, in particular p44/42 MAPK, is involved in TNF-␣ production in macrophages, monocytes, and microglia following activation of these cells with a variety of stimuli, including LPS and CD40 ligand (9,37,38). Having shown that cross-linking CD45 inhibits CD40L-induced activity of the Src family kinases Lck and Lyn in microglial cells, we wished to examine whether this reduced Src kinase activity could lead to down-regulation of p44/42 MAPK activity. To investigate this possibility, microglial cells were co-incubated with anti-CD45 mAb and CD40L. Cell lysates were then analyzed for phosphorylated forms of p44/42 MAPK by Western immunoblotting. Results show that crosslinking of CD45 significantly inhibits CD40L-induced activation (phosphorylation) of p44/42 MAPK (Fig. 6A). To determine if this effect could result in decreased MAPK activity, a direct method, immune complex kinase assay, was performed. Results show that cross-linking of CD45 markedly reduces p44/42 MAPK activity in CD40L-treated microglia (Fig. 6B), demonstrating the functionality of CD45 cross-linking on p44/42 MAPK activity.
Ligation of CD40 Results in Marked p44/42 MAPK Activity   FIG. 2. Microglial CD40 expression is not affected by CD45 cross-linking. Graph represents a summary of FACS analysis results for CD40 expression on microglia (mean percentage of CD40-expressing cells Ϯ 1 S.E.) with n ϭ 3 for each condition presented. ANOVA revealed a significant main effect of CD40L (p Ͻ 0.001), but not for anti-CD45 (p Ͼ 0.05), and no significant interaction was noted between them (p Ͼ 0.05). One-way ANOVA revealed significant between-groups differences (p Ͻ 0.001), and post hoc testing showed significant differences between control and CD40L (p Ͻ 0.05). However, no significant differences were noted between control and anti-CD45 (p Ͼ 0.05) or between CD40L/anti-CD45 and CD40L/control antibody (p Ͼ 0.05). summarizing band densities (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. For A and B, one-way ANOVA revealed a significant difference between wild-type microglia before and after CD40L treatment (p Ͻ 0.001), but did not show a significant difference between CD40-deficient microglia before and after CD40L treatment (p Ͼ 0.05), indicating that CD40L mediates its effect on p44/42 MAPK specifically through the CD40-CD40L interaction. def., deficient.
and TNF-␣ Production in CD45-deficient Microglial Cells-To further substantiate the role of CD45 in negatively regulating CD40L-induced microglial activation, microglia were obtained from CD45-deficient or wild type mice and incubated with or without CD40L. Activity of p44/42 MAPK was then evaluated in cell lysates from these conditions 30 min after treatment. Data show that p44/42 MAPK activation (Fig. 7A) and activity (Fig. 7B) are markedly enhanced in CD40L-challenged micro- showing phosphorylated p44/42 MAPK in microglia, and graph (bottom) summarizing band density ratios (phospho-p42 MAPK/total p42 MAPK) (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. B, immune complex kinase assay (top) showing phosphorylation of the MAPK fusion protein, Elk1, and graph (bottom) summarizing band densities (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. For C and D, microglia were co-treated with CD40L (1 g/ml) and PP1 at the doses indicated. C, Western blot (top) showing phosphorylated p44/42 MAPK in microglia, and graph (bottom) summarizing band density ratios (phospho-p42 MAPK/total p42 MAPK) (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. D, immune complex kinase assay (top) showing phosphorylation of the MAPK fusion protein, Elk1, and graph (bottom) summarizing band densities (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. Similar results were observed when microglia were co-treated with CD40L (1 g/ml) and damnacanthal (dose range from 200 to 1000 nM). E, summary of TNF-␣ release ELISA results (mean TNF-␣ pg/mg of total protein Ϯ 1 S.E.) with n ϭ 3 for each condition presented. For A, B, and E, ANOVA revealed a significant main effect of CD40L (p Ͻ 0.001), and significant interactive terms between CD40L and either damnacanthal (p Ͻ 0.001) or PP1 (p Ͻ 0.001). One-way ANOVA revealed significant between-groups differences (p Ͻ 0.001), and post hoc testing showed significant differences between control and CD40L (p Ͻ 0.001) as well as between CD40L and either CD40L/damnacanthal (p Ͻ 0.05) or CD40L/PP1 (p Ͻ 0.001). For C and D, ANOVA revealed a significant main effect (p Ͻ 0.001) of Src kinase inhibitor dose, indicating dose-dependent inhibition of p44/42 MAPK. glia that are deficient for CD45. As we have previously shown that TNF-␣ release induced by CD40 ligation is dependent on p44/42 MAPK, we went on to measure TNF-␣ production by CD45-deficient microglia treated with CD40L for 24 h. Results shown in Fig. 7C indicate much greater activation of CD45deficient microglia compared with wild-type microglia following stimulation with CD40L, supporting that CD45 is a negative regulator of CD40-mediated microglial activation. Moreover, in order to evaluate whether CD45 could be a central regulator of the p44/42 MAPK pathway, we pre-treated CD45deficient microglial cells for 1 h with PD 98059 (an inhibitor of MEK1/2, the upstream activator of p44/42 MAPK) and then incubated them with CD40L for 24 h. Microglial activation was subsequently evidenced by TNF-␣ production. Data show that PD 98059 notably decreases CD40L-induced TNF-␣ production by CD45-deficient microglia (Fig. 7C), further suggesting that CD45 plays a major role in negative regulation of the p44/42 MAPK pathway. Yet, as PD 98059 does not completely block CD40L-induced TNF-␣ secretion by CD45-deficient microglia, it seems likely that, although CD45 is not an obligatory regulator of the CD40 pathway, it does control the flux of signals emanating from CD40. DISCUSSION It has previously been shown that microglia express CD45, and this expression level is markedly enhanced following activation of these cells (13,14). CD45 is well known to couple to Src family kinases, including Lyn and Lck, where it modulates Src activity via dephosphorylation of tyrosine residues (15,19). Yet, the role of CD45 in microglial activation is currently speculative. We and others have shown that CD40 is also constitutively expressed on microglia at low levels, and markedly increases after activation of these cells (5). Ligation of microglial CD40 results in p44/42 MAPK-dependent TNF-␣ production (9), and it has been shown that stimulation of Src family kinases results in activation of the MAPK module (11). Thus, we wished to investigate whether CD45 might modulate FIG. 5. CD45 cross-linking inhibits CD40L-induced Lck and Lyn kinase activity in microglia. Lck (A) and Lyn (B) kinase activity (pmol ATP/min/mg total protein) reported as the mean Ϯ 1 S.E. with n ϭ 3 for each condition presented. For A and B, ANOVA revealed a significant main effect of CD40L (p Ͻ 0.001), and significant interactive terms between CD40L and anti-CD45 (p Ͻ 0.001), but not between CD40L and control antibody (p Ͼ 0.05). One-way ANOVA revealed significant between-groups differences (p Ͻ 0.001), and post hoc testing showed significant differences between control and either CD40L (p Ͻ 0.001) or anti-CD45 (p Ͻ 0.001), as well as between CD40L/anti-CD45 and CD40L/control antibody (p Ͻ 0.001).
FIG. 6. CD45 cross-linking suppresses CD40L-induced p44/42 MAPK activity in microglia. A, Western blot (top) showing phosphorylated p44/42 MAPK in microglia, and graph (bottom) summarizing band density ratios (phospho-p42 MAPK/total p42 MAPK) (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. B, immune complex kinase assay (top) showing phosphorylation of the MAPK fusion protein, Elk1, and graph (bottom) summarizing band densities (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. For A and B, ANOVA revealed a significant main effect of CD40L (p Ͻ 0.001), and a significant interactive term between CD40L and anti-CD45 (p Ͻ 0.001), but not between CD40L and control antibody (p Ͼ 0.05). One-way ANOVA revealed significant between-groups differences (p Ͻ 0.001), and post hoc testing showed significant differences between control and either CD40L (p Ͻ 0.001) or anti-CD45 (p ϭ 0.001), as well as between CD40L/anti-CD45 and CD40L/control antibody (p Ͻ 0.05).
CD40L-induced microglial activation through regulation of the Src/p44/42 MAPK pathway. Our results show that cross-linking CD45 potently inhibits microglial activation induced by CD40 ligation, as evidenced by TNF-␣ production. Furthermore, CD40 ligation results in marked activation of the Src family kinase members Lyn and Lck, with consequent downstream p44/42 MAPK activation in activated microglia. Cotreatment of microglia with CD40L and anti-CD45 mAb results in reduced Lck and Lyn kinase as well as p44/42 MAPK activity, showing that CD45 is a negative regulator of CD40Linduced microglial activation and suggesting a mechanism whereby CD45 brings about this effect by inhibiting Src kinase activity, a known function of CD45 (12).
As we had shown that cross-linking CD45 reduces CD40Linduced microglial TNF-␣ production, the possibility arose that this effect may be due, at least in part, to reduced CD40 receptor expression on the microglial cell surface. This idea was highlighted in a previous report, where it was shown that pharmacological inhibition of CD40-mediated monocyte activation was partially attributable to reduced gene expression of CD40 (33). To rule out this possibility in our system, we treated microglia with anti-CD45 mAb in the presence or absence of CD40L and measured CD40 expression levels on these cells compared with appropriate controls. We did not observe a significant effect of anti-CD45 mAb on CD40 protein expression alone or in combination with CD40L (Fig. 2). However, treatment of microglia with CD40L does result in increased CD40 receptor expression (Fig. 2), an effect that is most likely mediated by NF-B activation, as the CD40-CD40L interaction has previously been shown to activate functional NF-B (39,40). These data suggested to us that stimulation of CD45, unlike CD40, does not effect transcription factor-mediated gene expression of CD40, and led us to investigate the initial intracellular mediators of CD45-mediated negative regulation of CD40L-induced microglial activation.
CD45 is a membrane-bound PTP that is well known to couple to and directly regulate the activity of Src family tyrosine kinases. However, CD45-mediated dephosphorylation of Src family kinases is a complex and not well understood phenomenon (15,19,41). For example, CD45 can either activate or inactivate Src, depending on whether CD45 dephosphorylates inhibitory or activating sites within the SH1 kinase domain (19). It is thought that the receptor occupation and activation status of the immune cell under consideration (i.e. resting or antigen-associated receptor-ligated) may be a critical determinant of which sites CD45 dephosphorylates on Src (19). Thus, we considered CD45 modulation of Src activity against a background of ligation of the CD40 receptor, which is well known to participate in both immune cell activation and antigen-receptor signaling. Data show that cross-linking of microglial CD45 in the presence of CD40L results in reduced activity of Lck and Lyn, showing negative regulation of CD40L-induced Src activity by CD45. Interestingly, we find that cross-linking of microglial CD45 alone results in increased Src activity (Fig. 5), supporting the hypothesis that in non-activated, resting microglia, stimulation of CD45 results in dephosphorylation of inhibiting regions of the Src SH1 domain. This idea is in line with the dualistic nature of CD45-mediated Src kinase modulation proposed by Ashwell and D'Oro (15), who concluded that CD45 can act not only as a simple "on" switch, but also as an "off" summarizing band densities (mean Ϯ 1 S.D.) for above with n ϭ 3 for each condition presented. C, summary of TNF-␣ release ELISA results (mean TNF-␣ pg/mg of total protein Ϯ 1 S.E.) with n ϭ 3 for each condition presented. For A and B, ANOVA revealed significant main effects of CD40L (p Ͻ 0.001) and CD45 deficiency (p Ͻ 0.001), and a significant interaction between them (p Ͻ 0.05). One-way ANOVA revealed significant between-groups differences (p Ͻ 0.001), and hoc testing showed a significant difference between control microglia/ CD40L and CD45-deficient microglia/CD40L (p Ͻ 0.001). For C, oneway ANOVA revealed significant between-groups differences (p Ͻ 0.001), and post hoc testing showed a significant difference between CD45-deficient microglia/CD40L and CD45-deficient microglia/CD40L/ PD98059 (p Ͻ 0.001). def., deficient. switch depending on the activation status of the immune cell under consideration.
CD40L treatment has been shown to result in Src family kinase activation, particularly Lck and Lyn in B and T cells (26,27,42,43), and it has further been shown that, in B cells deficient for Lyn, ligation of CD40 results in a decreased proliferative response induced by interleukin-4 or B cell receptor stimulation (44,45). These data suggest that CD40 may be a positive regulator of Src, and we evaluted this possibility in microglia challenged with CD40L. Our data show that ligation of microglial CD40 results in increased activity of the Src family kinases Lck and Lyn (Fig. 5) as well as TNF-␣ secretion by these cells (Fig. 1). We have previously shown that CD40Linduced microglial TNF-␣ production is dependent on p44/42 MAPK (9), and we wished to evaluate the possibility that Src activation might bridge stimulation of microglial CD40 and consequent p44/42 MAPK activation. Thus, we co-treated microglia with CD40L and the Src family kinase inhibitors PP1 or damnacanthal, and find marked reduction in both p44/42 MAPK activation and TNF-␣ secretion by these cells (Fig. 4), suggesting that activation of Src is required to transduce p44/42 MAPK-dependent TNF-␣ production following CD40 ligation. This is particularly interesting when considered together with stimulation of microglial CD45, where co-treatment with CD40L and anti-CD45 mAb results in dramatic reduction of Src kinase and downstream p44/42 MAPK activities as well as TNF-␣ secretion. This suggests an antagonistic system that regulates microglial activation, whereby CD40 ligation leads to activation of these cells, while co-stimulation with CD40L and CD45 opposes it.
Having shown that cross-linking of CD45 opposes CD40Linduced microglial activation, we asked the question whether stimulation of CD45 with anti-CD45 mAb could also mitigate against microglial activation induced by other pro-inflammatory stimuli, such as LPS. To examine this possibility, we stimulated microglia with LPS in the presence of anti-CD45 mAb, and found marked reduction in microglial p44/42 MAPK activation and TNF-␣ secretion (data not shown). It has previously been reported that LPS transduces microglial activation via activation of the MAPK module (38,46). Additionally, LPSinduced macrophage activation has been shown to involve one or more Src family kinases (25,47), suggesting that LPS, like CD40L, stimulates the intracellular Src/MAPK pathway in microglia. It is suggested, then, that stimulation of CD45 is effective at blocking microglial activation induced by a variety of stimuli by virtue of its ability to oppose Src/MAPK pathway activation. Thus, in vivo stimulation of CD45 might be a viable therapeutic target in the treatment of neurodegenerative diseases which involve pathological microglial activation, such as AD and MS.