Inhibition of Interferon (IFN) (cid:1) -induced Jak-STAT1 Activation in Microglia by Vasoactive Intestinal Peptide INHIBITORY EFFECT ON CD40, IFN-INDUCED PROTEIN-10, AND INDUCIBLE NITRIC-OXIDE SYNTHASE EXPRESSION*

Interferon (IFN)- (cid:1) is one of the most important microglia stimulators in vivo participating in inflammation and Th1 activation/differentiation. IFN- (cid:1) -mediated signaling involves the activation of the Jak/STAT1 pathway. The neuropeptides vasoactive intestinal peptide (VIP) and the pituitary adenylate cyclase activating polypeptide (PACAP) are two potent microglia-deacti-vating factors that inhibit the production of proinflammatory mediators in vitro and in vivo . The present study investigated the molecular mechanisms involved in the VIP/PACAP regulation of several IFN- (cid:1) -induced micro-glia-derived factors, including IFN- (cid:1) -inducible pro-tein-10 (IP-10), inducible nitric-oxide synthase (iNOS), and CD40. The results indicate that VIP/PACAP inhibit Jak1–2 and STAT1 phosphorylation, and the binding of activated STAT1 to the IFN- (cid:1) activated site motif in the IFN regulatory factor-1 and CD40 promoter and to the IFN-stimulated response element motif of the IP-10 promoter. Through its effect in the IFN- (cid:1) -induced Jak/ STAT1 pathway, VIP and PACAP are able to control the gene expression of IP-10, CD40, and iNOS, three micro-glia-derived synthase (cid:5) tetrahydrobiopterin, dehydrogenase from of NO metabolite assay. mixed, and 550 calcu- lated a 2

Under normal conditions, brain microglia are involved in immune surveillance and host defense against infectious agents (1). However, in response to brain injury, infection, or inflammation, microglia readily become activated, in a way similar to peripheral tissue macrophages, a process that includes differentiation and probably invasion and proliferation. Activation of microglia is a histopathological hallmark of several neurodegenerative diseases, including Alzheimer and Parkinson's diseases, multiple sclerosis, and the AIDS dementia complex (1)(2)(3). Pathological microglial activation is believed to contribute to progressive damage in neurodegenerative diseases through the release of proinflammatory and/or cytotoxic factors, including TNF-␣, 1 IL-1␤, IL-6, IL-12, and nitric oxide. Hence, it is important to unravel mechanisms regulating microglia activation of inflamed brain parenchyma to provide insights into efficient therapeutic intervention.
IFN-␥ constitutes one of the most potent microglia-activating factors. Binding of IFN-␥ to its receptor induces the assembly of an active receptor complex and consequent transphosphorylation of the receptor-associated Janus tyrosine kinases Jak1 and Jak2 (4,5). The activation of these kinases induces phosphorylation of the cytoplasmic tail of the receptor itself, which lacks intrinsic kinase activity. The cytosolic protein signal transducer and activator of transcription (STAT1) is then recruited to the activated IFN-␥-receptor complex and phosphorylated (4). Upon phosphorylation, STAT1 forms homodimers and translocate to the nucleus, where they bind to the IFN-␥activated site (GAS), also termed STAT binding element (SBE), found in the promoter of many IFN-␥-induced genes including the IFN regulatory factor-1 (IRF-1) and ICAM-1 genes (6 -9). Many of the regulatory effects of IFN-␥ in microglia appear to be mediated by IRF-1 and/or STAT1, which transactivate multiple effector genes including IFN-␥-inducible protein-10 (IP-10), CD40, IL-12, and the inducible nitric-oxide synthase (iNOS) (10 -20).

Reagents-Synthetic
Cell Cultures-Microglial cell cultures were prepared as previously described (31). Briefly, cerebral cortical cells from 1-day-old BALB/c mice were dissociated after a 30-min trypsinization (0.25%) and were plated in 75-cm 2 Falcon culture flasks in Dulbecco's modified Eagle's medium high glucose formula (Invitrogen) supplemented with 10% heat-inactivated fetal calf serum (Invitrogen), containing 10 mM HEPES buffer, 1 mM pyruvate, 0.1 M nonessential amino acids, 2 mM glutamine, 50 mM 2-mercaptoethanol, 100 units/ml penicillin, and 10 g/ml streptomycin (complete medium). The medium was replenished 1 and 4 days after plating, and on day 8 of culture, plates were shaken for 20 min at a speed of 200 rpm in an orbital shaker to remove oligodendrocytes. On day 12 of culture, plates were shaken again for 2 h at a speed of 180 -200 rpm. Harvested cells were filtered through a 20-m nylon mesh, plated in a 60-mm Petri dish, and incubated for 15 min at 37°C. After extensive washing with culture medium, adherent cells (microglia) were collected with a rubber policeman and centrifuged (1000 rpm, 10 min). Purified microglial cell cultures comprised a cell population in which Ͼ98% stained positively with MAC-1 antibodies (Roche Molecular Biochemicals) and Ͻ2% stained positively with antibodies specific to the astrocyte marker glial fibrillary acid protein (Sigma).
Microglia monolayers were incubated with complete medium and stimulated with 50 units/ml IFN-␥ in the presence or absence of VIP or PACAP38 (from 10 Ϫ12 to 10 Ϫ6 M) at 37°C in a humidified incubator with 5% CO 2 . Cell-free supernatants were harvested at the designated time points and kept frozen (Ϫ20°C) until IP-10 determination.
RNA Extraction and Northern Blot Analysis-Northern blot analysis was performed according to standard methods. Murine primary microglia cells were cultured at a concentration of 2 ϫ 10 6 cells/ml in 100-mm tissue culture dishes and stimulated with IFN-␥ (50 units/ml) in the presence or absence of VIP (10 Ϫ8 M) or PACAP (10 Ϫ8 M) for different time periods at 37°C. Cells were collected, and total RNA was extracted by the acid guanidinium-phenol-chloroform method, electrophoresed on 1.2% agarose-formaldehyde gels, transferred to Nytran membranes (Schleicher & Schuell), and cross-linked to the nylon membrane using UV light. The probes for murine IRF-1, GAPDH, and IP-10 were generated by reverse transcription-PCR as previously described (32,33). Oligonucleotides were end-labeled with [␥-32 P]ATP (3000 Ci/mmol, Amersham Biosciences) by using T4 polynucleotide kinase. The RNA-containing membranes were prehybridized for 16 h at 42°C and hybridized at 42°C for 16 h with the appropriate probes. The membranes were washed twice in 2ϫ SSC containing 0.1% SDS at room temperature (20 min each time), once at 37°C for 20 min, and once in 0.1ϫ SSC containing 0.1% SDS at 50°C (20 min). The prehybridization and hybridization buffers were purchased from 5 Prime 3 3 Prime Inc. (Boulder, CO). The membranes were exposed to x-ray films (Eastman Kodak Co.). CD40 mRNA expression was determined by RNase protection assay as described previously (34).
Electrophoretic Mobility Shift Assay (EMSA)-Nuclear extracts were prepared by the mini-extraction procedure of Schreiber et al. (35) with slight modifications. Briefly, microglia cells were cultured at a density of 10 7 cells in 6-well plates, stimulated as described above, washed twice with ice-cold PBS plus 0.1% bovine serum albumin, and harvested from the dishes. Incubation times for the different factors assayed were optimized empirically. The cell pellets were homogenized with 0.4 ml of buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF, 10 g/ml aprotinin, 10 g/ml leupeptin, 10 g/ml pepstatin, and 1 mM NaN 3 ). After 15 min on ice, Nonidet P-40 was added to a final 0.5% concentration, the tubes were gently vortexed for 15 s, and nuclei were sedimented and separated from cytosol by centrifugation at 12,000 ϫ g for 40 s. Pelleted nuclei were washed once with 0.2 ml of ice-cold buffer A, and the soluble nuclear proteins were released by adding 0.1 ml of buffer C (20 mM HEPES, pH 7.9, 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 25% glycerol, 1 mM DTT, 0.5 mM PMSF, 10 g/ml aprotinin, 10 g/ml leupeptin, 10 g/ml pepstatin, and 1 mM NaN 3 ). After incubation for 30 min on ice, followed by centrifugation for 10 min at 14,000 rpm at 4°C, the supernatants containing the nuclear proteins were harvested, the protein concentration was determined by the Bradford method, and aliquots were stored at Ϫ80°C for later use in EMSAs.
Western Blot Analysis of iNOS-For detection of iNOS by Western blotting, microglia cells (10 7 cells) were lysed in cold 40 mM Tris-HCl buffer, pH 8.0, containing 0.1 mM PMSF, 5 g/ml aprotinin, 50 g/ml leupeptin, 1 g/ml chymostatin, and 5 g/ml pepstatin, and lysed by sonication. Thirty micrograms of protein extracts was separated by 7.5% SDS-PAGE gels under reducing conditions. After electrophoresis, the gel was electroblotted in Tris-glycine buffer (48 mM Tris, 39 mM glycine, pH 9.2) containing 40% methanol onto a reinforced nitrocellulose membrane (Schleicher & Schuell). The membrane was blocked with TBS-T buffer (10 mM Tris, pH 8.0, 150 mM NaCl, 0.05% Tween 20) containing 5% milk powder for 1 h at room temperature, then incubated for an additional 6 h at 4°C with a monoclonal antibody against iNOS (1:1000) in 5% nonfat milk/TBS-T solution. After washing three times in TBS-T, membranes were incubated at room temperature for 1 h with peroxidase-conjugated goat anti-mouse IgG at 1:2000 dilution. After washing three times in TBS-T for 5 min each, and once in Tris-buffered saline for 5 min, the membrane was drained briefly and subjected to the enhanced chemiluminescence detection system (ECL, Amersham Biosciences). The x-ray films were exposed for 5-20 min.
IP-10 ELISA-The amounts of IP-10 present in culture supernatants were determined using a specific sandwich ELISA. Briefly, a capture monoclonal anti-murine IP-10 antibody (clone A 102-6, 2.5 g/ml) was used for coating, and a biotinylated rabbit anti-mouse IP-10 polyclonal antibody (4 g/ml) was used for detection. The detection limit for the assay is 15 pg/ml IP-10.
Determination of NO Synthase Activity-The NO synthase activity was measured as described (39). Briefly, primary microglia cells were cultured at a density of 10 7 cells in 6-well plates, stimulated as described above, washed twice with ice-cold PBS plus 0.1% bovine serum albumin. Cells were lysed by three cycles of freeze/thaw in 40 mM Tris-HCl, pH 8.0, 0.1 mM EGTA, 0.1 mM EDTA, 12 mM 2-mercaptoethanol, and protease inhibitors (0.1 mM PMSF, 5 g/ml aprotinin, 1 g/ml leupeptin, 1 g/ml chymostatin, and 5 g/ml pepstatin), and centrifuged at 10,000 ϫ g for 60 min. For the NO synthase activity assay, microglia lysates were incubated with 2 mM L-arginine, 2 mM NADPH, 4 M flavin-adenine dinucleotide, 4 M tetrahydrobiopterin, 2 mM DTT, and 1 mM EGTA for 2 h at 37°C. The reaction was terminated by adding 15 units/ml L-lactic dehydrogenase and 83 mM sodium pyruvate and incubating for an additional 15 min. The amount of NO formed in the reaction mixture was estimated from the accumulation of the stable NO metabolite nitrite by the Griess assay. Equal volumes of culture supernatants and Griess reagents (1% sulfanilamide, 0.1% N-[naphthyl]ethylenediamine dihydrochloride in 2.5% H 3 PO 4 ) were mixed, and the absorbance was measured at 550 nm. The amount of nitrite was calculated from a NaNO 2 standard curve.
Determination of CD40 Protein Expression by Immunofluorescence Flow Cytometry-CD40 expression was determined by flow cytometry as previously described (20). Briefly, primary microglia were plated at 2 ϫ 10 5 cells/well into 12-well plates (Costar) and stimulated as described above. The cells were scraped in ice-cold RPMI complete medium and washed twice with PBS containing 0.1% sodium azide plus 2% heat-inactivated fetal calf serum (wash buffer). Cells were incubated with 10 g/ml rat IgG2 a-k anti-mouse CD40 Ab (clone 3/23) at 4°C for 45 min, then washed, and incubated with 10 g/ml biotinylated anti-rat IgG2a, for 30 min at 4°C. After washing, the cells were incubated with 10 g/ml phycoerythrin-conjugated streptavidin for 30 min at 4°C. Isotype-matched Abs were used as controls, and IgG block (Sigma) was used to block the nonspecific binding to Fc receptors. After extensive washing, the cells were fixed in 1% paraformaldehyde. Stained microglia cells, gated according to forward and side scatter characteristics, were analyzed on a FACScan flow cytometer (Becton Dickinson). Samples in which isotype-matched antibody was used instead of specific antibody were used as negative controls to determine the proper region or window setting. Fluorescence data were expressed as mean fluorescence intensity and as percentage (%) of positive cells after subtraction of background isotype-matched values.

RESULTS AND DISCUSSION
VIP and the structurally related neuropeptide PACAP have been lately identified as two potent anti-inflammatory agents, which down-regulate the activation of T cells and macrophages (reviewed in Refs. [21][22][23][24][25]. Recently, both neuropeptides were shown to have a similar anti-inflammatory action in central nervous system, modulating the in vitro and in vivo microglia secretion of proinflammatory mediators such as TNF-␣, IL-1␤, IL-6, IL-12, nitric oxide, and several chemokines (26,27). This might be of clinical relevance, because VIP and PACAP have been identified as potential therapeutic agents for several disorders where these microglia-derived cytotoxic mediators are involved, such as ischemia-reperfusion, septic shock, brain trauma, and Parkinson's disease (28 -30, 40, 41). Several recent reports have shown the intracellular mechanisms involved in the action of VIP/PACAP in microglia activated with the bacterial endotoxin LPS. VIP/PACAP regulation of LPS-induced proinflammatory cytokines and chemokines occurs at the transcriptional level, and involves several transcriptional factors, such as NFB, cAMP-responsive element-binding protein, c-Jun, and cAMP-responsive element-binding proteinbinding protein (26,27,42,43). However, the effect of VIP and PACAP on IFN-␥-stimulated microglia, and the intracellular mechanisms involved are not elucidated yet. In the present study, the molecular mechanisms for the suppression of IFN-␥-induced gene expression by VIP/PACAP were investigated, with particular emphasis on the role of the Jak1-2/STAT1/ IRF-1 signaling pathway.
VIP and PACAP Reduce IFN-␥-induced Jak1 and Jak2 Phosphorylation in Microglia-IFN-␥ has been studied extensively as a transcriptional regulator, whose action is initiated by the binding of IFN-␥ to its receptor resulting in the activation of Jak1/2 by transphosphorylation (4,44). First, whether VIP and PACAP regulate Jak1 and Jak2 activation was investigated. Cell extracts of IFN-␥-stimulated microglia, treated with or without VIP or PACAP, were immunoprecipitated with anti-Jak1 or anti-Jak2 Ab, followed by Western blotting with an anti-phosphotyrosine Ab. Both Jak1 and Jak2 were weakly phosphorylated in unstimulated controls; the Jak1/Jak2 phosphorylation level increased within 5 min following IFN-␥ stimulation (Fig. 1). Treatment with VIP or PACAP reduced Jak1 and Jak2 phosphorylation at all time points (Fig. 1).
VIP and PACAP Inhibit the IFN-␥-induced Phosphorylation of STAT1-The activation of Jak1/2 by IFN-␥ induces phosphorylation and activation of the cytoplasmic tail of the receptor itself, which recruits the cytosolic protein STAT1 and phosphorylates it (4). Tyrosine phosphorylation of STAT1 protein induces its homodimerization and translocation to the nucleus (4,44,45). To determine whether VIP and PACAP affect the activation of STAT1, microglia cells were stimulated with IFN-␥ for different times in the presence or absence of VIP or PACAP. Cell extracts were immunoprecipitated with anti-STAT1 Ab, and the levels of phosphorylated STAT1 were assessed by Western blot using a specific Ab against Tyr 701phosphorylated STAT1␣. IFN-␥ stimulated in a timedependent manner the phosphorylation of STAT1 (Fig. 2). Treatment with VIP or PACAP significantly reduced the levels of phosphorylated STAT1 in activated microglia ( Fig. 2A). In addition of phosphorylation in Tyr 701 , to act with maximal efficiency as a transcription factor, STAT1 must also be phosphorylated on Ser 727 by a kinase not clearly defined (46). As Fig. 2B shows, VIP and PACAP also decreased IFN-␥-induced STAT1 phosphorylation on Ser 727 , suggesting that both neuropeptides fully deactivate STAT1 pathway.
The majority of IFN-␥-responsive genes are induced through the interaction of phosphorylated STAT1 homodimers and an inducible palindromic enhancer termed GAS/SBE present in the promoter of many IFN-␥-induced genes, including IRF-1 and CD40 (4,6,7,19,20,44). IFN-␥ also induces transcription of a subset of genes, such as IP-10, in a GAS-independent fashion through a motif termed the ISRE (16 -18).
VIP and PACAP Inhibit STAT1 Binding to the GAS Site of the IRF-1 Promoter-Although the IRF-1 promoter contains a complex array of transactivating binding sites, GAS/SBE appears to be essential for maximal IRF-1 gene transcription following IFN-␥ stimulation (4,5). To investigate whether VIP/ PACAP affect STAT1 binding to the GAS motif, nuclear extracts from microglia stimulated with IFN-␥ in the presence or absence of VIP or PACAP were used in gel shift assays. IFN-␥ treatment led to increased binding to an oligonucleotide con-

FIG. 3. VIP and PACAP decrease IRF-1 expression in activated microglia by inhibiting STAT1 binding to the IRF-1 promoter.
Microglia cells (10 7 cells) were incubated with medium alone or activated with IFN-␥ (50 units/ml) in the presence or absence of VIP or PACAP (10 Ϫ8 M). A, after 45 min of incubation, nuclear extracts were prepared and analyzed for DNA binding activity by EMSA using radiolabeled oligonucleotides containing the GAS sequence motif of the IRF-1 promoter. Specificity was assessed by the addition of 50-fold excess of unlabeled homologous, or non-homologous (NFB) oligonucleotides (data not shown). Right panel, identification of the proteins bound to the GAS site. Nuclear extracts were incubated with polyclonal antibodies against STAT1 or NFB p65 for 20 min before the addition of the oligonucleotide probe. Arrow indicates the supershifted STAT1 band. Similar results were observed in three independent experiments. B, expression of IRF-1 mRNA was analyzed by Northern blot at different times after IFN-␥ stimulation. Cells incubated with medium alone were used as basal IRF-1 mRNA level controls. One representative experiment of three is shown. To determine the levels of STAT1 immunoprecipitated in each case, the membranes were stripped and reprobed with Abs directed to STAT1 (lower panels). Similar results were obtained in three separate experiments. taining the GAS motif, and treatment with VIP and PACAP inhibited this binding (Fig. 3A). The GAS binding was competed by an excess of unlabeled homologous oligonucleotide (GAS) but not by a non-homologous oligonucleotide (NFB) (data not shown). Antibody supershift experiments indicate that the GAS-binding complexes in IFN-␥-stimulated macrophages contain STAT1 (Fig. 3A). The VIP/PACAP inhibition of STAT1-GAS DNA binding in the IRF-1 promoter was correlated with an inhibitory effect on IFN-␥-induced IRF-1 mRNA expression (Fig. 3B).

VIP and PACAP Inhibit IFN-␥-induced iNOS Expression by
Decreasing IRF-1 Binding to Its Promoter-IFN-␥ may indirectly affect many effector genes through IRF-1. For example, although the iNOS promoter contains a complex array of transactivating binding sites, IRF-1 appears to be essential for maximal iNOS transcription after IFN-␥ stimulation (11)(12)(13). To investigate whether VIP and PACAP affect IRF-1 nuclear translocation and DNA binding in microglia stimulated with IFN-␥, IRF-1 binding to the iNOS promoter was assessed by EMSA. Stimulation of microglia cells with IFN-␥ led to an increase in IRF-1 binding, and treatment with VIP or PACAP significantly inhibited this binding (Fig. 4A). The specificity of FIG. 4. VIP and PACAP inhibit iNOS activity and protein expression in activated microglia by decreasing the binding of IRF-1 to the iNOS promoter. Microglia cells (10 7 cells) were incubated with medium alone or activated with IFN-␥ (50 units/ml) in the presence or absence of VIP or PACAP (10 Ϫ8 M). A, after 4 h of incubation, nuclear extracts were prepared and analyzed for DNA binding activity by EMSA using radiolabeled oligonucleotides containing the IRF-1 site of the iNOS promoter. Specificity was assessed by the addition of 50-fold excess of unlabeled homologous, or non-homologous (NFB) oligonucleotides (data not shown). Right panel, identification of the proteins bound to the IRF-1 site. Nuclear extracts were incubated with polyclonal antibodies against IRF-1 or NFB p65 for 20 min before the addition of the oligonucleotide probe. Arrow indicates the supershifted IRF-1-specific band. Similar results were observed in three independent experiments. B, after 24 h of incubation, cells were harvested and lysed, and the lysates were subjected to Western blots as described under "Experimental Procedures." One representative experiment of three is shown. C, after different time periods, iNOS activity was assayed in cell lysates by measuring the generation of nitrite/mg protein/min. Data are the mean Ϯ S.D. of three experiments performed in duplicate.

FIG. 5. VIP and PACAP inhibit IFN-␥-induced IP-10 production by decreasing the binding of STAT1 to the IP-10 promoter.
Microglia cells (10 7 cells) were incubated with medium alone or activated with IFN-␥ (50 units/ml) in the presence or absence of VIP or PACAP (10 Ϫ8 M). A, after 2 h of incubation, nuclear extracts were prepared and analyzed for DNA binding activity by EMSA using radiolabeled oligonucleotides containing the ISRE sequence motif of the IP-10 promoter. Specificity was assessed by the addition of 50-fold excess of unlabeled homologous, or non-homologous (NFB) oligonucleotides (data not shown). Right panel, identification of the proteins bound to the ISRE site. Nuclear extracts were incubated with polyclonal antibodies against STAT1 or STAT2 for 20 min before the addition of the oligonucleotide probe. Arrow indicates the supershifted STAT1/ ISRE-specific band. Similar results were observed in three independent experiments. B, after 8 h of incubation, expression of IP-10 mRNA was analyzed by Northern blot. One representative experiment of three is shown. C, IP-10 levels in culture supernatants were determined by ELISA at different times after IFN-␥ stimulation. Data are the mean Ϯ S.D. of three experiments performed in duplicate. the IRF-1 binding was evident by the supershifted IRF-1-specific band observed in IFN-␥-stimulated cells (Fig. 4A). VIP/ PACAP-mediated decrease in IRF-1 DNA binding to the iNOS promoter was correlated with an inhibition of iNOS protein expression and activity in IFN-␥-activated microglia (Fig. 4, B  and C). Therefore, VIP and PACAP decrease IFN-␥-induced iNOS expression by inhibiting Jak1/2-STAT1-mediated IRF-1 expression. In contrast to neurons and endothelial cells that express a constitutive NOS, microglia express an inducible type of NOS, iNOS, responsible for the prolonged, high output production of nitric oxide. In general, expression of iNOS follows a generalized or localized inflammatory response resulting from infection or tissue injury. Despite its beneficial role in host defense, sustained nitric oxide production can be deleterious for the host, especially under inflammatory conditions in the central nervous system, such as brain trauma and Parkinson's disease (1)(2)(3)(47)(48)(49). Therefore, the selective inhibition of expression of iNOS represents an important therapeutic goal. In this sense, we have previously shown that VIP down-regulated in vivo expression of microglial iNOS during brain in-FIG. 6. VIP and PACAP decrease CD40 expression in IFN-␥-activated microglia by inhibiting STAT1 binding to the CD40 promoter. Microglia cells (10 7 cells) were incubated with medium alone or activated with IFN-␥ (50 units/ml) in the presence or absence of VIP or PACAP (10 Ϫ8 M). A, after 2 h of incubation, nuclear extracts were prepared and analyzed for DNA binding activity by EMSA using radiolabeled oligonucleotides containing the GAS sequence motif of the CD40 promoter. Specificity was assessed by the addition of 50-fold excess of unlabeled homologous, or non-homologous (NFB) oligonucleotides (data not shown). Right panel, identification of the proteins bound to the GAS site. Nuclear extracts were incubated with polyclonal antibodies against STAT1 or STAT2 for 20 min before the addition of the oligonucleotide probe. Arrow indicates the supershifted STAT1/GASspecific band. Similar results were observed in three independent experiments. B, after 8 h of incubation, expression of CD40 mRNA was analyzed by RNase protection assay. One representative experiment of three is shown. C, surface expression of CD40 was assessed by flow cytometry at different times (24 h for histograms on left panels) after IFN-␥ stimulation. Upper panels represent the staining profile with isotypematched control Abs. Results are the mean Ϯ S.D. of three experiments performed in duplicate. flammation, brain trauma, and Parkinson's disease (26,27,29,30), this effect being accompanied by a neuroprotective action.

VIP and PACAP Inhibit IP-10 Expression by Down-regulating STAT1 Binding to the ISRE Motif on Its Promoter-IP-10
(CXCL10) is a member of the CXC chemokine gene family that is expressed constitutively in lymphoid organs, and strongly induced by IFN-␥ in a variety of cell types including macrophages, dendritic cells, and microglia (49 -54). ISRE has been identified as the essential element in the IP-10 promoter involved in its transcriptional activation following IFN-␥ stimulation (16 -18, 36). To investigate whether VIP and PACAP affect STAT1 binding to the ISRE motif of the IP-10 promoter, EMSAs were performed. Stimulation of microglia with IFN-␥ led to strong STAT1 DNA binding to the ISRE site of the IP-10 promoter in comparison with unstimulated cells, and VIP and PACAP significantly inhibited this binding (Fig. 5A). The binding specificity was confirmed by the displacement with 50-fold excess of unlabeled homologous (ISRE) but not non-homologous oligonucleotide (NFB) (data not shown). Antibody supershift experiments indicate the specific binding of STAT1, but not STAT2, to the ISRE motif (Fig. 5A, right panel). VIP/PACAPmediated decrease in STAT1 DNA binding to the ISRE site of the IP-10 promoter was accompanied by an inhibition of IP-10 mRNA and protein expression in IFN-␥-activated microglia (Fig. 5, B and C). In recent reports, VIP has been shown to inhibit IP-10 production by LPS-stimulated macrophages and dendritic cells (55,56); however, molecular mechanisms involved in such effect were not addressed. The biological significance of the VIP/PACAP inhibition of IP-10 production by stimulated microglia is supported by in vivo experiments. IP-10 and its receptor CXCR3 have been associated with Th1 dominant immune responses. CXCR3 is expressed primarily on T lymphocytes of the Th1 phenotype, a finding consistent with the preferential chemoattraction of activated Th1 cells by IP-10 (57,58). As expected, IP-10 is expressed in high levels in several Th1-type diseases such as rheumatoid arthritis, psoriasis, multiple sclerosis, inflammatory bowel disease, and type I diabetes (59 -63). Thus, the inhibition of directed migration of Th1 cells is of particular significance for multiple sclerosis, where IP-10 neutralization prevented central nervous system recruitment of effector Th1 cells in a model of experimental autoimmune encephalomyelitis (EAE) (64). Therefore, VIP/ PACAP impairment of IP-10 production by activated microglia in an inflamed brain parenchyma could be beneficial for the treatment of multiple sclerosis avoiding the preferential recruitment of activated autoreactive Th1 effector cells. min (for Jak1), the phosphorylation of STAT1 and Jak1 was assayed by immunoprecipitation and immunoblotting as described in Figs. 1 and 2, respectively. After 45 min of incubation, nuclear extracts were prepared and analyzed for DNA binding activity by EMSA using radiolabeled oligonucleotides containing the GAS sequence motif of the IRF-1 promoter. Arrow indicates the specific STAT1-GAS complex. After 90 min, IRF-1 mRNA expression was determined by Northern blot. After 24 h, expression of iNOS protein was determined by Western blot, CD40 expression was assessed by flow cytometry, and IP-10 production was determined by ELISA. Similar results were obtained in three independent experiments.

VIP and PACAP Inhibit IFN-␥-induced CD40 Expression by
Down-regulating STAT1-GAS Binding-CD40, a member of the tumor necrosis factor receptor family, is expressed on antigen-presenting cells such as dendritic cells, monocytes, B cells and microglia, where after its ligation with CD40 ligand, expressed transiently on activated T cells, produces bidirectional signaling essential for both costimulation of T cell and CD40bearing cell activation (reviewed in Ref. 65). IFN-␥ is the most potent inducer of CD40 expression by microglia (19,20,66). IFN-␥-induced CD40 expression is mainly mediated by the binding of STAT1 homodimers to the GAS/SBE site of its promoter (19,20). Therefore, I next investigated whether VIP and PACAP could affect Jak/STAT1-mediated CD40 expression. EMSA experiments showed that treatment of microglia cells with IFN-␥ led to strong binding of nuclear extracts to the CD40/GAS-containing probe compared with unstimulated cells, and treatment with VIP or PACAP decreased this binding (Fig. 6A). Antibody supershift experiments were performed to determine the composition of the CD40/GAS-binding factors. Addition of anti-STAT1, but not anti-STAT2 Abs, to the binding reaction resulted in a marked reduction in the intensity of the GAS band and led to the appearance of a slow migrating band, indicating that the GAS-binding complex of the CD40 promoter in IFN-␥-activated microglia is composed primarily of STAT1 homodimers (Fig. 6A). Inhibition of STAT1 binding to the GAS motif of the CD40 promoter by VIP/PACAP was correlated by a decreased IFN-␥-induced CD40 expression, at both mRNA and protein levels (Fig. 6, B and C). A recent report showed that VIP and PACAP inhibit CD40 expression in LPSstimulated rat microglia (67), an effect indirectly mediated through the production of the anti-inflammatory cytokine IL-10. Although IL-10 involvement cannot be totally excluded in our study, because of the different timing in the expression of both CD40 and IL-10 following microglia activation, a direct effect of VIP/PACAP on IFN-␥-induced signal transduction seems to be the most plausible mechanism mediating the inhibition of CD40 expression by both neuropeptides. Ligation of microglial CD40 plays a crucial role in the development of immune and inflammatory responses in the central nervous system. This is particularly important in vivo under pathological conditions. For example, ligation of microglial CD40 is essential for the development of EAE (68,69), and its levels in the central nervous system correlate with the severity of the disease (70,71), governing the pathogenesis of EAE locally by controlling the recruitment/retention of autorreactive T cells (72). In addition, microglial CD40 ligation induces microglia maturation and the production of proinflammatory neurotoxic cytokines (73)(74)(75). Therefore, negative regulation of IFN-␥-induced CD40 in microglia by VIP/PACAP, is physiologically important, because it may be one of the mechanisms by which these neuropeptides antagonizes the development of Th1 cells and dampens inflammatory responses.
Involvement of VPAC1 and cAMP in the Inhibitory Effect of VIP on IFN-␥-induced Jak1-STAT1 Activation in Macrophages-Next we investigated whether the inhibitory effect of VIP/PACAP could be related to occupancy of specific receptors. VIP and PACAP act through a family of receptors consisting of VPAC1, VPAC2, and PAC1 (76). Reverse transcription-PCR analysis indicates that murine primary microglia express both PAC1 and VPAC1 mRNA (26). In contrast, VPAC2 mRNA was not expressed even following activation. A similar pattern of VIP/PACAP receptor expression was observed in rat primary microglia (77) and in the two murine microglia cell lines, EOC13 and BV2 (26).
The majority of the effects of VIP and PACAP in LPS-activated microglia are exerted through the VIP/PACAP-receptor VPAC1, and the subsequent activation of the cAMP/PKA pathway (26,27). Therefore, it was next investigated whether the To determine the levels of STAT1 immunoprecipitated in each case, the membranes were stripped and reprobed with Abs directed to STAT1 (lower panel). Similar results were obtained in three separate experiments. inhibitory effect of VIP on the Jak/STAT/IRF signaling pathway could be related to occupancy of this specific receptor and to activation of PKA. The VPAC1 is coupled primarily to the adenylate cyclase system (76), and IFN-␥-induced Jak1-STAT1 activation and production of some of the factors studied (i.e. IP-10, CD40, and iNOS) is indeed inhibited in activated microglia by agents that increase intracellular cAMP levels (33, 39, 78 -84). We investigated the ability of a specific VPAC1 antagonist (85) and of H89, a potent and selective PKA inhibitor, to reverse the effects of VIP. Both the VPAC1 antagonist and the PKA inhibitor reversed the inhibitory effects of VIP on IP-10, CD40, iNOS, and IRF-1 expression (Fig. 7), on the binding of STAT1 to GAS/IRF-1 (Fig. 7), and on STAT1 and Jak1 phosphorylation (Fig. 7). In contrast, PACAP 6 -38 , an antagonist specific for the VIP/PACAP receptors PAC1 and VPAC2 (86), or calphostin C, a PKC inhibitor, did not affect the inhibitory effects of VIP (data not shown). These results suggest that the inhibition of Jak/STAT/IRF signaling pathway by VIP is mediated through VPAC1 and the cAMP/PKA pathway. This is supported by the fact that, whereas forskolin (a cAMP-inducing agent) and a VPAC1 agonist (87) exert an effect similar to VIP, the VPAC2 agonists Ro-25-1553 and helodermin (76) and the VIP fragment 10 -28 did not show any significant effect on IFN-␥-induced STAT1 phosphorylation, IRF-1 expression, and IP-10 production (Fig. 7). The link between the cAMP pathway and the Jak1/2 activation has not been elucidated. One of the possibilities is the induction of some of the members of the family of the suppressors of cytokine signaling (SOCS) by VIP/ PACAP. The members of the SOCS family act as inhibitors of cytokine signaling (88), and SOCS1 and SOCS3 inhibit STAT1 activation by IFN-␥ (89). However, VIP and PACAP did not induce the expression of SOCS1 and 3 in unstimulated microglia, and did not alter SOCS1 and/or SOCS3 expression in cells treated with IFN-␥ (data not shown). Another possibility that should be addressed is that VIP/PACAP activate tyrosine phosphatases that dephosphorylate Jaks and STAT1. Results showed in Fig. 8 raise against this possibility because the tyrosine phosphatase inhibitor sodium orthovanadate did not reverse the VIP-dependent suppression of STAT1 Tyr phosphorylation and GAS binding. In summary, the present work demonstrates that, as in macrophages (37), VIP and PACAP sequentially inhibit IFN-␥-induced Jak1/Jak2 activation, and the subsequent activation of the STAT1, in terms of phosphorylation, nuclear translocation, and binding to the GAS motif in the IRF-1 and CD40 promoters and to the ISRE motif in the IP-10 promoter (Fig. 9). Because IFN-␥ is one of the most important microglia stimulators in vivo participating in inflammation and Th1 activation/differentiation, and through its effect in the IFN-␥-induced Jak/STAT1 pathway, VIP and PACAP are able to control the gene expression of IP-10, CD40, and iNOS, three microglia-derived mediators that plays an essential role in several pathologies, i.e. inflammation and autoimmune disorders, VIP and PACAP emerge as two attractive therapeutic factors of such disorders and could represent a significant element in the regulation of the inflammatory response in the central nervous system by endogenous neuropeptides.