Vasoactive Intestinal Peptide and Pituitary Adenylate Cyclase-activating Polypeptide Inhibit Interleukin-12 Transcription by Regulating Nuclear Factor κB and Ets Activation*

The vasoactive intestinal peptide (VIP) and the structurally related neuropeptide pituitary adenylate cyclase-activating polypeptide (PACAP) act as “macrophage-deactivating factors”. We showed previously that VIP and PACAP inhibit the production of macrophage-derived tumor necrosis factor-α, interleukin (IL)-6, nitric oxide, and IL-12. This study examines the molecular mechanisms involved in the VIP/PACAP inhibition of IL-12 production. VIP and PACAP inhibit IL-12 (p40) gene expression by affecting both NF-κB binding and the composition of the Ets-2 binding complex. Both neuropeptides prevent the activation-induced nuclear translocation of the NF-κB components p65 and c-Rel by inhibiting the reduction in cytoplasmic IκBα. Moreover, VIP and PACAP inhibit the synthesis of the interferon responsive factor-1. The decrease in nuclear interferon responsive factor-1 and c-Rel results in alterations of the Ets-2-binding complex. Two transduction pathways, a cAMP-dependent and a cAMP-independent pathway, are involved in the inhibition of IL-12 gene expression and appear to differentially regulate the transcriptional factors involved. Because IL-12 participates in T cell activation and cytolytic T lymphocyte activity and promotes the differentiation of T helper cells into the Th1 subset, the understanding of the mechanisms that affect IL-12 production in normal and pathological conditions could contribute to immune response-based therapies or vaccine designs.

The vasoactive intestinal peptide (VIP) and the structurally related neuropeptide pituitary adenylate cyclase-activating polypeptide (PACAP) act as "macrophage-deactivating factors". We showed previously that VIP and PACAP inhibit the production of macrophagederived tumor necrosis factor-␣, interleukin (IL)-6, nitric oxide, and IL-12. This study examines the molecular mechanisms involved in the VIP/PACAP inhibition of IL-12 production. VIP and PACAP inhibit IL-12 (p40) gene expression by affecting both NF-B binding and the composition of the Ets-2 binding complex. Both neuropeptides prevent the activation-induced nuclear translocation of the NF-B components p65 and c-Rel by inhibiting the reduction in cytoplasmic IB␣. Moreover, VIP and PACAP inhibit the synthesis of the interferon responsive factor-1. The decrease in nuclear interferon responsive factor-1 and c-Rel results in alterations of the Ets-2-binding complex. Two transduction pathways, a cAMP-dependent and a cAMP-independent pathway, are involved in the inhibition of IL-12 gene expression and appear to differentially regulate the transcriptional factors involved. Because IL-12 participates in T cell activation and cytolytic T lymphocyte activity and promotes the differentiation of T helper cells into the Th1 subset, the understanding of the mechanisms that affect IL-12 production in normal and pathological conditions could contribute to immune response-based therapies or vaccine designs.
The generation of an immune response involves the activation of effector cells such as macrophages, neutrophils, and T lymphocytes and the subsequent production of cytokines, chemokines, and reactive oxygen and nitrogen intermediates. The activated macrophages are widely recognized as cells that play an important role in inflammatory processes, as well as in the initiation, maintenance, and control of specific immune responses. In response to LPS 1 and other activating agents, macrophages secrete nitric oxide and proinflammatory cytokines, such as TNF␣, IL-1␤, IL-6, and IL-12, and immunomodulatory cytokines, such as TGF␤1 and IL-10 (1). Because the intensity and duration of an inflammatory process depends on the local balance between pro-and anti-inflammatory factors, the so-called "macrophage deactivating factors" received considerable attention lately (2)(3)(4)(5)(6).
Vasoactive intestinal peptide (VIP) and the pituitary adenylate cyclase-activating polypeptide (PACAP) are two neuropeptides that perform a broad spectrum of biological functions affecting both natural and acquired immunity (reviewed in Refs. [7][8][9], primarily as anti-inflammatory agents. VIP and PACAP have been shown to inhibit T cell proliferation and cytokine production (reviewed in Ref. 10) and to inhibit several macrophage functions, including phagocytosis, respiratory burst, and chemotaxis (reviewed in Ref. 8). In agreement with their anti-inflammatory role, VIP and PACAP were recently reported to inhibit the in vitro and in vivo production of proinflammatory cytokines such as IL-6 and TNF␣ (11)(12)(13), to reduce the expression of the inducible nitric-oxide synthase (14), to enhance the production of the anti-inflammatory cytokine IL-10 (15), to protect mice from endotoxic shock presumably through the inhibition of endogenous TNF␣ and of other inflammatory mediators (16), and to act as survival factors against tissue injury of lung and neuronal cells (17)(18)(19). Furthermore, we and others have recently demonstrated that VIP and PACAP inhibit IL-12 production in endotoxin-stimulated peritoneal macrophages (20 -22), with a subsequent inhibitory effect on IFN␥ synthesis by T cells (22). IL-12, an early proinflammatory cytokine secreted by macrophages activated by microbial products, plays a central role in the regulation of cell-mediated immunity (reviewed in Refs. [23][24][25]. IL-12 stimulates the proliferation of activated T lymphocytes and enhances IFN␥ secretion by NK cells and T lymphocytes. Consistent with this latter effect, IL-12 has a pivotal role in the induction of CD4 ϩ Th1 cell responses, acting in antagonism to IL-4, the major promoter of the Th2 response (26,27). In mice, IL-12 plays a decisive role in the protection against intracellular pathogens, including parasites and bacteria (23)(24)(25). Thus, the understanding of the mechanisms that affect IL-12 production in normal and pathological conditions could contribute to immune response-based therapies or vaccine designs. IL-12 is a unique cytokine because of its heterodimer structure. Bioactive IL-12 (p70) is composed of two disulfide-linked subunits (p35 and p40) encoded by two separate genes. When both subunits are produced within the same cell, they assemble into a biologically active heterodimer (28). However, although the expression of the p35 gene is constitutive in a wide variety of cells, the p40 gene is highly tissue-regulated, being restricted to phagocytic cells with antigen-presenting capability (29,30), and is therefore considered to function as the regulatory component for IL-12 expression (29).
The key role of IL-12 in the immune response and in inflammation and the importance of this cytokine in anti-tumor resistance have raised considerable interest in the mechanisms of IL-12 gene transcription. Two functional promoter regions of the p40 gene that confer LPS inducibility and IFN␥ augmentation have been identified (31)(32)(33)(34). The region spanning from position Ϫ132 to Ϫ122 contains a novel NF-B site (31), whereas the region from Ϫ211 to Ϫ207 contains an Ets-2 element that binds a complex formed by the protein transactivators Ets-2, GLp109, IRF-1, and c-Rel (32)(33)(34). The aim of this study was to understand the molecular mechanisms through which VIP and PACAP inhibit IL-12 production in macrophages stimulated with endotoxin and IFN␥.
Cell Cultures-Mouse peritoneal macrophages were elicited by intraperitoneal injection of 2 ml of 4% Brewer's thioglycollate medium (Difco, Detroit, MI) into male Balb/c mice (age 6 -10 weeks). Peritoneal exudate cells were obtained 72 h after injection by peritoneal lavage with icecold RPMI 1640 medium. Peritoneal exudate cells containing lymphocytes and macrophages were washed twice and resuspended in ice-cold RPMI 1640 medium supplemented with 2% heat-inactivated fetal calf serum (Life Technologies, Inc.), 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 (RPMI 1640 complete medium). Cells were seeded in flat bottom 96-well microtiter plates (Corning Glass, Corning, NY) at 8 ϫ 10 4 cells/well in a final volume of 200 l. The cells were incubated at 37°C for 2 h to allow adherence to plastic, and nonadherent cells were removed by repeated washing with RPMI 1640 medium. At least 96% of the adherent cells were macrophages as judged by fluorescence-activated cell sorter analysis.
Raw 264.7 mouse macrophage cells (ATCC, Manasses, VA) were cultured in Dulbecco's modified Eagle's medium supplemented with 2 mM L-glutamine, 100 units/ml penicillin, 10 g/ml streptomycin, and 10% fetal calf serum (CM). The cells (8 ϫ 10 4 ) were plated in flat bottom 96-well microtiter plates in 200 l of CM for 24 h. Nonadherent cells were removed by aspiration and two washings with Dulbecco's modified Eagle's medium.
Macrophage monolayers (murine peritoneal macrophages or Raw 264.7 cells) were incubated in CM and stimulated with 200 units/ml IFN␥ before (8 or 12 h) or simultaneously (0 h) with treatment with LPS (0.5 g/ml). VIP or PACAP38 (10 Ϫ12 -10 Ϫ7 M) were added at the same time with IFN␥ or LPS. Control cultures were treated with medium alone. Cell-free supernatants were harvested at the designated time points and kept frozen (Ϫ20°C) until IL-12 determination by ELISA.
RNA Extraction and Northern Blot Analysis-Northern blot analysis was performed according to standard methods. Macrophage monolayers (2 ϫ 10 6 cells/ml) were stimulated with LPS (0.5 g/ml) and IFN␥ (200 units/ml), in the absence or presence of VIP and PACAP (10 Ϫ8 M) for different time periods at 37°C. 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.
Electrophoretic Mobility Shift Assay-Nuclear extracts were prepared by the mini-extraction procedure of Schreiber et al. (40) with slight modifications. Raw 264.7 cells were plated at a density of 10 7 cells/well in 6-well plates, stimulated, washed twice with ice-cold phosphate-buffered saline/0.1% bovine serum albumin, and scraped off the dishes. 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 dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 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 concentration of 0.5%, 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 dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 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.
Oligonucleotides corresponding to the NF-B half site (Ϫ134/Ϫ110) (31) and to the Ets-2 motif (Ϫ292/Ϫ196) (32, 33) of the IL-12 p40 promoter were synthesized and annealed. Aliquots of 50 ng of the double-stranded oligonucleotides were end-labeled with [␥-32 P]ATP by using T4 polynucleotide kinase. For EMSAs with macrophage nuclear extracts, 20,000 -50,000 cpm of double-stranded oligonucleotides, corresponding to approximately 0.5 ng, were used for each reaction. The binding reaction mixtures (15 l) contain 0.5-1 ng of DNA probe, 5 g of nuclear extract, 2 g of poly(dI-dC)⅐poly(dI-dC), and binding buffer (50 mM NaCl, 0.2 mM EDTA, 0.5 mM dithiothreitol, 5% glycerol, and 10 mM Tris-HCl, pH 7.5). The mixtures were incubated on ice for 15 min before adding the probe, followed by another 20-min incubation at room temperature. Samples were loaded onto 4% nondenaturing polyacrylamide gels and electrophoresed in TGE buffer (50 mM Tris-HCl, pH 7.5, 0.38 M glycine, and 2 mM EDTA) at 100 V, followed by transfer to Whatman paper, drying under vacuum at 80°C, and autoradiography. In competition and antibody supershift experiments, the nuclear extracts were incubated for 15 min at room temperature with the specific antibody (1 g) or competing cold oligonucleotide (50-fold excess) before the addition of the labeled probe.
Western Blot-Lysates, cytoplasmic fractions, or nuclear extracts (see above) containing 20 -30 g of protein were subjected to reducing SDS-polyacrylamide gel electrophoresis (12.5%). After electrophoresis, the gel was electroblotted in Tris-glycine buffer containing 40% methanol onto a nitrocellulose membrane (Trans-blot, Bio-Rad). 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 and then incubated with primary antibodies (rabbit antimouse IgG) against IB (1:250), phosphorylated IB␣ (1:500), NF-B p50 (1:1000), NF-B p65 (1:1000), or IRF-1 (1:500) in TBS-T containing 1% milk powder for 2 h at room temperature. The membrane was washed with TBS-T and incubated with the secondary antibody (goat anti-rabbit IgG conjugated to horseradish peroxidase) at 1:5000 dilution for 1 h at room temperature. After washing three times in TBS-T for 5 min each and once in TBS for 5 min, the membrane was drained briefly and subjected to the enhanced chemiluminiscence detection system (ECL, Amersham Pharmacia Biotech). The x-ray films were exposed for 5-20 min.

RESULTS
VIP and PACAP Inhibit LPS-induced IL-12 Production by IFN␥-primed Macrophages-To investigate the effects of VIP and PACAP on IL-12 production, peritoneal macrophages were stimulated with IFN␥ and/or LPS in the absence or presence of various doses of VIP or PACAP, and the amounts of IL-12 p40 and p70 released in the culture supernatants were assayed by ELISA. IFN␥ alone did not induce significant IL-12 p40 and p70 production, whereas LPS alone stimulated a modest increase over that of unstimulated macrophages (data not shown). Macrophages treated simultaneously with IFN␥ and LPS produced marginally more IL-12 than those stimulated with either IFN␥ or LPS alone (Fig. 1A). However, pretreatment with IFN␥ followed 12 h later by LPS stimulation resulted in a dramatic increase in both IL-12 p40 and p70 production (Fig. 1A). VIP and PACAP inhibited IL-12 p40 and p70 release by IFN␥/LPS-stimulated macrophages (Fig. 1A). The addition of VIP or PACAP at the time of IFN␥ priming resulted in a slightly higher inhibitory effect than when neuropeptides were added at the same time with LPS (data not shown).
The murine macrophage cell line Raw 264.7 primed with IFN␥ and stimulated with LPS shows a similar pattern of regulation. VIP and PACAP inhibit IL-12 p40 and p70 in a dose-and time-dependent manner (Fig. 1, B and C). The dose-FIG. 1. VIP and PACAP inhibit IL-12 p40 and p70 production in LPS/IFN␥-stimulated macrophages. A, VIP and PACAP inhibit IL-12 p40 and p70 in peritoneal macrophages stimulated with IFN␥ and/or LPS. Peritoneal macrophages (4 ϫ 10 5 cells/ml) were cultured in the presence of IFN␥ (200 units/ml) and LPS (0.5 g/ml) (left panels) or primed with IFN␥ for 12 h followed by the addition of LPS (right panels). VIP or PACAP (10 Ϫ8 M) were added together with IFN␥ and LPS (left panels) or at the time of IFN␥ priming (right panels). Control cultures were incubated with medium alone. Supernatants were collected 24 h later, and IL-12 p40 and p70 release was determined by ELISA. Each result is the mean Ϯ S.D. of six experiments performed in duplicate. B and C, VIP and PACAP inhibit IL-12 in IFN␥-primed Raw 264.7 cells. B, time course for the inhibitory effect of VIP/PACAP on IL-12 p40 and p70 production. Raw 264.7 cells (4 ϫ 10 5 cells/ml) were pretreated with IFN␥ (200 units/ml) for 12 h before LPS stimulation (0.5 g/ml). VIP or PACAP (10 Ϫ8 M) was added at the time of IFN␥ priming. Control cultures were incubated with medium alone. Supernatants collected at different times were assayed for IL-12 p40 and p70 production by ELISA. C, dose-response curve for the inhibitory effect of VIP and PACAP on IL-12 p40 and p70 production. Raw 264.7 cells (4 ϫ 10 5 cells/ml) were pretreated with IFN␥ (200 units/ml) for 12 h before LPS stimulation (0.5 g/ml). Different concentrations of VIP or PACAP were added at the time of IFN␥ priming. Supernatants were collected 18 h after LPS stimulation and IL-12 p40 and p70 release was determined by ELISA. For A-C, cells cultured in the absence of IFN␥ and LPS with VIP or PACAP did not produce detectable levels of IL-12 p40 and p70 (Ͻ30 pg/ml). Each result is the mean Ϯ S.D. of six experiments performed in duplicate.
response curves were similar for VIP and PACAP, showing maximal effects at 10 Ϫ8 M (Fig. 1, B and C).
VIP and PACAP Inhibit IL-12 Production at a Transcriptional Level-To determine whether VIP/PACAP affect IL-12 transcription, Raw 264.7 cells were primed with IFN␥ or treated simultaneously with LPS and IFN␥ in the presence or absence of 10 Ϫ8 M VIP or PACAP for 4, 8, and 16 h, and total RNA was prepared and subjected to IL-12 p40 and p35 North-ern blot analysis. IL-12 p40 mRNA is absent in unstimulated cells, marginally induced upon simultaneous treatment with IFN␥ and LPS, and induced strongly upon pretreatment with IFN␥ followed by LPS stimulation (Fig. 2A). In contrast, IL-12 p35 mRNA was constitutively expressed in unstimulated macrophages and was not affected by IFN␥ priming and/or LPS stimulation ( Fig. 2A). VIP and PACAP inhibited IFN␥/LPSinduced IL-12 p40 mRNA, without affecting IL-12 p35 mRNA FIG. 2. VIP and PACAP inhibit IL-12 p40 transcription. Raw 264.7 cells (2 ϫ 10 6 cells/ml) were cultured in the presence of IFN␥ (200 units/ml) and LPS (0.5 g/ml) or primed with IFN␥ followed by LPS stimulation 8 h later. VIP or PACAP (10 Ϫ8 M) were added at the time of IFN␥/LPS stimulation, of IFN␥ priming, or of LPS stimulation. Cells incubated with medium alone were used as basal IL-12 p35 and p40 mRNA level controls. Total RNA was extracted, and the expression of IL-12 p35, IL-12 p40 and ␤-actin mRNA was analyzed by Northern blot analysis at the indicated time points. Results are expressed in densitometric units normalized for the expression of ␤-actin. One representative experiment of three is shown. expression (Fig. 2). These results indicate that both neuropeptides down-regulate steady-state IL-12 p40 mRNA levels.
VIP and PACAP Inhibit NF-B Binding to the IL-12 p40 Promoter-Activation and nuclear translocation of members of the NF-B/c-Rel family constitutes the hallmark of macrophage stimulation by proinflammatory cytokines and bacterial products (41). Although the IL-12 p40 promoter contains a complex array of transactivating binding sites, NF-B is essential for maximal IL-12 p40 transcription after IFN␥/LPS stimulation (31,32). To investigate whether VIP/PACAP affects NF-B binding, we used electrophoretic mobility shift assays. Treatment of Raw 264.7 cells with IFN␥ followed by LPS stimulation led to NF-B binding, and VIP and PACAP inhibited the binding (Fig. 3A). The NF-B binding was competed by an excess of unlabeled homologous oligonucleotide (NF-B) but not of nonhomologous oligonucleotide (CRE) (Fig. 3A). Antibody supershift experiments indicate that the NF-B-binding complexes in IFN␥/LPS stimulated macrophages contain p50, p65, and c-Rel (Fig. 3B). Because VIP and PACAP almost completely blocked NF-B binding, to identify the NF-B-binding factors, we used a 10-fold excess of nuclear extract in the VIP treated samples. The NF-B binding complexes contain p50, p65, and c-Rel (Fig. 3B), suggesting that although VIP significantly reduces NF-B binding, it does not change the composition of the NF-B binding complexes.

VIP and PACAP Modulate the Composition of the Ets-2-
binding Complexes-Recently, it has been established that the Ets-2 element situated at position Ϫ211/Ϫ207 in the IL-12 p40 promoter is required for optimal transcription of the IL-12 p40 gene in macrophages (32)(33)(34). This region interacts with the nuclear complex F1, induced in monocytic cells primed with IFN␥ and stimulated with LPS. The F1 complex is composed of Ets-2, the nuclear factor GLp109, and the additional components IRF-1 and c-Rel (33). Stimulation of Raw 264.7 cells with IFN␥/LPS led to an increase in Ets-2 binding compared with unstimulated cells, and treatment with VIP or PACAP did not significantly affect the binding (Fig. 4A). The specificity of the Ets-2-binding activity was confirmed with homologous (Ets-2) or nonhomologous (CRE) oligonucleotides as competitors (Fig.  4A). Antibody supershift experiments were performed to determine the composition of the Ets-2-binding complexes. In IFN␥/ LPS-stimulated cells, the majority of the complexes were supershifted by anti-Ets-2, anti-IRF-1, and anti-c-Rel Abs but not by anti-CREB Abs (an irrelevant Ab) (Fig. 4B). In contrast, in IFN␥/LPS-stimulated cells treated with VIP and PACAP, the complex was supershifted only by the anti-Ets-2 Ab, with no supershift with anti-IRF-1, anti-c-Rel, or anti-CREB Abs (Fig.  4B). This indicates that in VIP/PACAP-treated cells, the Ets-2-binding complexes contain Ets-2 and minor amounts, if any, of IRF-1 and c-Rel.
Involvement of VPAC1 and cAMP in the Effects of VIP on B and Ets-2 Binding-Next we investigated whether the inhibi- tory effect of VIP/PACAP on the IL-12 production could be related to occupancy of specific receptors. The immunological actions of VIP and PACAP are exerted through a family of VIP/PACAP receptors that were recently reclassified (42): VPAC1 and VPAC2, which exhibit similar affinities for the two neuropeptides and activate primarily the adenylate cyclase system, and PAC1, which exhibits a 300 -1000-fold higher affinity for PACAP than for VIP and activates both the adenylate cyclase and phospholipase C systems (reviewed in Ref. 42). Previously we showed that the inhibition of IL-12 production in peritoneal macrophages by VIP/PACAP is mediated primarily through VPAC1 and involves both cAMP-dependent and -independent transduction pathways (22). We reached similar conclusions regarding the inhibition of IL-12 production in Raw 264.7 cells. A VPAC1 agonist (43) inhibited IL-12 transcription and release from IFN␥/LPS-induced Raw 264.7 cells to a similar degree as VIP/PACAP (Fig. 5A). Also, a specific VPAC1 antagonist (44) reversed the inhibitory effect of VIP/PACAP (Fig. 5A). In contrast, PACAP 6 -38 , an antagonist specific for PAC1 and to a lesser degree for VPAC2 (45), did not reverse the effects of VIP and PACAP (Fig. 5A). The role of second messengers was investigated by using calphostin C (a protein kinase C inhibitor), H89 (a cAMP-dependent protein kinase A inhibitor), and forskolin and PGE 2 (two strict cAMP-inducing agents). Similar to peritoneal macrophages, forskolin and PGE 2 inhibited IL-12 p40 release in IFN␥/LPS-stimulated Raw 264.7 cells, although they showed less of an effect at lower concentrations (10 and 100 nM) as compared with VIP and PACAP (Fig. 5B). In addition, the involvement of cAMP is supported by the results obtained with the two protein kinase inhibitors. The VIP/ PACAP inhibition of IL-12 transcription and release from IFN␥/LPS-stimulated Raw 264.7 cells was not affected by calphostin C but was partially reversed by H89 (Fig. 5B).
Because the inhibitory effect of VIP on IL-12 production is mediated primarily through VPAC1 and cAMP represents at least one of the second messengers involved, we determined the effect of the VPAC1 antagonist and of the cAMP-dependent protein kinase A inhibitor H89 on the changes induced by VIP in B and Ets-2-binding complexes. The VPAC1 antagonist, but not PACAP 6 -38 , reversed the inhibitory activity of VIP on NF-B binding and on the changes in the composition of the Ets-2-binding complexes (Fig. 6). This is in accordance with the fact that the VPAC1 agonist showed a similar effect than that of VIP on NF-B-and Ets-2-binding complexes (Fig. 6). In contrast, H89 had a more limited effect. H89 did not reverse the inhibitory effect on NF-B binding; however, H89 affected the composition of the Ets-2-binding complexes, i.e. it reversed the VIP-induced supershift pattern obtained with the anti-IRF-1 Ab but not with the anti-c-Rel Ab (Fig. 6). These results suggest that both the inhibition of NF-B and the change in the composition of the Ets-2-binding complexes by VIP are mediated through VPAC1, but only the reduction in IRF-1 in the Ets-2binding complexes is entirely cAMP-dependent. This conclusion is supported by the fact that forskolin (a cAMP inducer) did not affect NF-B binding and reduced the presence of IRF-1, but not c-Rel, in the Ets-2-binding complexes (Fig. 6).

VIP and PACAP Inhibit IFN␥/LPS-dependent Reduction of IB␣ Levels and Nuclear Translocation of the p65 and c-Rel
Subunits of NF-B-Nuclear translocation of NF-B is preceded by the phosphorylation and proteolytic degradation of IB␣ (41). To determine whether the VIP/PACAP inhibition of NF-B binding was due to an effect on IB␣ phosphorylation/ degradation, we examined the levels of cytoplasmic IB␣ by Western blot. Significant lower levels of IB␣ were observed in IFN␥/LPS-stimulated cells, compared with unstimulated cells (Fig. 7A). VIP and PACAP completely blocked the IFN␥/LPSinduced reduction in IB␣ (Fig. 7A). Next, to investigate whether VIP and PACAP block IFN␥/LPS-induced reduction in IB levels by increasing IB␣ mRNA expression and/or by inhibiting IB␣ phosphorylation and further degradation, we examined the levels of IB␣ mRNA and cytoplamic phosphorylated IB␣ protein by Northern blot and Western blot analysis, respectively, in IFN␥/LPS-stimulated macrophages in the presence or absence of VIP or PACAP. As Fig. 7B shows, neither VIP nor PACAP affected IB mRNA steady-state levels. However, both neuropeptides significantly inhibited IFN␥/ LPS-induced phosphorylation of IB␣ protein (Fig. 7C).
Because NF-B activation requires the nuclear translocation of p65 or c-Rel, we measured the levels of p65 and c-Rel proteins in cytoplasm and nucleus. As expected, upon IFN␥/LPS treatment, the levels of p65 and c-Rel declined in the cytoplasm and concurrently increased in the nucleus (Fig. 7A). Treatment with VIP and PACAP abolished the IFN␥/LPS-dependent changes in nuclear and cytoplasmic p65/c-Rel levels (Fig. 7A). These results indicate that VIP and PACAP inhibit the IFN␥/ LPS-induced nuclear translocation of p65 and c-Rel, which is consistent with the inhibition of reduction in cytoplasmic IB␣ levels. In contrast to p65, c-Rel and IB, p50 levels were not affected by IFN␥/LPS, with or without VIP/PACAP (Fig. 7A).
Ets-2 Protein Complexes Are Not Affected by VIP and PACAP-To investigate the effect of VIP and PACAP on the expression of the Ets-2 protein, nuclear extracts from unstimulated and IFN␥/LPS-stimulated Raw 264.7 cells in the absence or presence of VIP or PACAP were analyzed by Western blot. As described previously (33), a single band of approximately 54 kDa was present in unstimulated cells, whereas several additional bands, GLp58, GLp109, Lp119, and Lp165, were detected in IFN␥/LPS-stimulated cells (Fig. 7D). Treatment with VIP or PACAP did not significantly modify this protein pattern (Fig. 7D). These results suggest that the two neuropeptides do not inhibit the synthesis of the Ets-2 components.
VIP and PACAP Inhibit IRF-1 Expression-Next we investigated the effect of VIP and PACAP on the expression and synthesis of IRF-1, another component of the Ets-2-binding complex. Northern blot analysis indicated that although IRF-1 mRNA was not detectable in unstimulated cells, it was strongly induced in the IFN␥/LPS-stimulated Raw 264.7 cells (Fig. 7E). VIP and PACAP significantly reduced the levels of specific IRF-1 mRNA (Fig. 7E). In addition, Western blot analysis of nuclear extracts confirmed the inhibitory effect of VIP and PACAP on IRF-1 expression (Fig. 7E, lower panel).
Involvement of VPAC1 and cAMP in the Effect of VIP on IB␣ Degradation, the Nuclear Translocation of p65 and c-Rel, and IRF-1 Expression-To correlate the involvement of VPAC1 and cAMP with changes in NF-B and Ets-2 complexes at the protein level, we investigated the effect of the VPAC1 antagonist and H89 on the changes induced by VIP/PACAP in B and Ets-2-binding complexes. The VIP-induced decrease in nuclear p65, c-Rel and IRF-1 and the blockage in reduction of cytoplasmic IB␣ levels were completely reversed by the VPAC1 antagonist (Fig. 8A). However, H89 reversed only the inhibitory effect of VIP on IRF-1 expression (Fig. 8) without affecting p65, c-Rel, and IB␣ changes (Fig. 8A). In contrast, calphostin C did not affect VIP effect on IRF-1 expression nor on NF-B complex changes (Fig. 8A). None of the treatments affected the nuclear p50 levels (Fig. 8A). These results indicate that the VIP-induced inhibition of IB␣ reduction, nuclear translocation of p65 and c-Rel, and IRF-1 expression are mediated through VPAC1, but only the inhibition of IRF-1 expression is entirely cAMPdependent. This is also supported by the fact that forskolin did not affect neither nuclear levels of p65 and c-Rel or the cyto-plasmic IB␣ levels but, similar to VIP/PACAP, inhibited IRF-1 expression (Fig. 8). DISCUSSION Macrophages are widely recognized as cells that play a central role in the regulation of immune and inflammatory activities, as well as tissue remodeling (1). In response to antigens such as LPS, macrophages secrete proinflammatory cytokines and oxidants such as TNF␣, IL-6, IL-1␤, IL-12, and nitric oxide (1). VIP and PACAP are potent anti-inflammatory agents that down-regulate the activation of T cells and macrophages (7)(8)(9)(10). VIP and PACAP were shown to modulate the macrophage secretion of proinflammatory mediators such as TNF␣, IL-6, nitric oxide (11)(12)(13)(14), and recently IL-12 (20 -22). Here we extend these studies to the molecular mechanisms involved in the inhibitory effect of VIP/PACAP on IL-12 production. We focused on the regulation of the 40-kDa IL-12 subunit that is induced by bacterial stimulation in phagocytic cells. Production of IL-12 is maximal when macrophages are primed with IFN␥ for 12-18 h prior to LPS stimulation (32,33); this was confirmed in the present study. Our results indicate that VIP/ PACAP inhibit IL-12 (p40 and p70) production in murine peritoneal and Raw264.7 macrophages. The inhibitory effect is dose-dependent within a wide range of neuropeptide concen- Peritoneal macrophages have been previously shown to express VPAC1 and PAC1 mRNA and both high and low affinity VIP/PACAP binding sites (46,47). Recently we showed that VPAC1 and PAC1 mRNA are expressed constitutively, and VPAC2 expression is induced following LPS stimulation in both peritoneal and Raw 264.7 macrophages (22,48). Agonist studies indicated that although both VPAC1 and VPAC2 mediate the inhibitory effect on IL-12, the VPAC1 agonist is significantly more efficient (75% inhibition, as compared with 25-35% for the VPAC2 agonist) (22). However, the apparent major role of VPAC1 could reflect the balance between VPAC1 and VPAC2 expression. Because VPAC2 is expressed relatively late during macrophage activation (12 and 24 h), VPAC1 is probably the major receptor type present during the early culture period. The role of VPAC1 as a major player in mediating the effect of VIP/PACAP on IL-12 is supported by the fact that a VPAC1 antagonist reverses the inhibitory effect and blocks the effect of VIP/PACAP on both NF-B and Ets-2 complex binding and that a VPAC1 agonist mimics the effect of VIP/PACAP. An understanding of the events mediating IL-12 secretion is complicated by the fact that the biologically active IL-12 is a p35/p40 heterodimer (28,29). To secrete the biologically active heterodimer, mRNAs for both subunits must be expressed, and both proteins must be translated and assembled within the same cell (28). The p40 gene is transcribed only in IL-12producing cells, such as macrophages/monocytes, dendritic cells, human peripheral blood mononuclear cells, and to a lesser degree B cells. In contrast, the expression of p35 mRNA is ubiquitous and constitutive, although free p35 polypeptide chains do not appear to be secreted (29,30). Cells producing the active IL-12 heterodimer also secrete high levels of free p40 polypeptide chains, although their biological role, particularly in humans, is still undefined (reviewed in Refs. [23][24][25]. These findings raise the question of how the expression of these subunits is cooperatively modulated by either positive or negative regulatory factors. Previous experiments regarding VIP modulation of cytokine expression indicated different molecular mechanisms, i.e. transcriptional regulation for IL-2, IL-6, IL-10, and TNF␣ versus post-transcriptional regulation for IL-4 (10,11,15,48,49). The present study indicates that the inhibitory effect of VIP and Control cultures were incubated with medium alone. Cytosolic and nuclear proteins were extracted 4 h after LPS stimulation. A, Western blot analysis was performed for IB␣ in the cytoplasmic fraction and for p50, p65, and c-Rel in the cytoplasmic as well as nuclear extracts. B, total RNA was extracted 1 h after LPS stimulation, and the expression of IB␣ mRNA was analyzed by Northern blot analysis. One representative experiment of three is shown. Lower panel, results are expressed in densitometric units normalized for the expression of ␤-actin. Each result is the mean Ϯ S.D. of three experiments performed in duplicate. C, cytosolic proteins were extracted 40 min after LPS stimulation, and Western blot analysis was performed for specific phosphorylated IB␣. One representative experiment of three is shown. D, Western blot analysis for Ets-2 was performed on nuclear extracts. E, total RNA was extracted 3 h after LPS stimulation, and the expression of IRF-1 and ␤-actin mRNA was analyzed by Northern blot analysis. Western blot analysis was performed for IRF-1 on nuclear extracts. One representative experiment of three is shown. PACAP on IL-12 production occurs at a transcriptional level, reducing the p40 mRNA levels, with no apparent effect on p35 gene expression.
The regulation of the IL-12 p40 gene transcription is complex and involves multiple cis-acting elements. Transcriptional regulation by LPS and IFN␥ of the IL-12 p40 gene has been shown to involve a "NF-B half-site" and transcriptional factors from the Rel family (31). In mammalian cells the Rel family includes NF-B1 (p50), RelA (p65), c-Rel, RelB, and NF-B2 (p50B, p52) (41). NF-B consists mostly of p50/p65 heterodimers, which are complexed to the inhibitor IB in the cytoplasm of unstimulated cells; stimuli such as LPS and proinflammatory cytokines induce the phosphorylation and degradation of IB, followed by the release and subsequent nuclear translocation of the p50/ p65 heterodimers, which bind to regulatory sequences in a variety of target genes (41). The present study indicates that VIP and PACAP inhibit NF-B binding to the IL-12 p40 promoter in IFN␥/LPS-stimulated Raw 264.7 cells. Similar to other studies (31), the NF-B complex induced by LPS/IFN␥ in macrophages was supershifted by anti-p50, anti-p65, or anti-c-Rel Abs, suggesting that the NF-B complexes consist of p50/ c-Rel and p50/p65 and that VIP and PACAP inhibit their nuclear translocation. We described a similar inhibitory effect of VIP and PACAP on NF-B binding activity for macrophage TNF␣ and inducible nitric-oxide synthase (14,48). In cells treated with VIP/PACAP, the reduced NF-B binding correlates with increased cytoplasmic IB, p65, and c-Rel levels and with decreased nuclear p65 and c-Rel levels. It has been previously described that the inhibition of NF-B nuclear translocation by other anti-inflammatory agents, such as IL-11, IL-10, TGF-␤1, glucocorticoids, and antioxidants, results from an increase in IB protein levels, a decrease in IB degradation, and/or phosphorylation (3, 50 -54). In the present study, we demonstrate that VIP and PACAP block IFN␥/LPS-induced reduction of IB levels by inhibiting phosphorylation of IB␣ subunit, without affecting mRNA IB␣ expression. It remains to be determined whether both neuropeptides mediate their effects also through inhibition of IB proteolytic degradation.
In addition to the B site, the ets element TTTCCT was identified as a major response region in the p40 promoter in Raw 264.7 cells (32)(33)(34). This element interacts with the large nuclear complex F1 that binds to the region between Ϫ196 and Ϫ292 in a complex way, requiring substantial flanking "anchoring" space (32,33). The induction of the F1 complex appears to correlate closely with the expression of the IL-12 p40 gene in various cell lines and human monocytes (33,34). F1 consists of multiple proteins including Ets-2, IRF-1, c-Rel, and a novel 109-kDa protein (GLp109) that is induced by either LPS or IFN␥ (33). VIP and PACAP do not change Ets-2 binding. However, supershift experiments indicate that both neuropeptides reduce IRF-1 and c-Rel, without affecting the Ets-2 protein. This suggests that the inhibitory effect of VIP/PACAP on IL-12 p40 gene expression is mediated, at least partially, through a change in the composition of the F1-binding complex. Indeed, the absence of either IRF-1 or c-Rel was reported to dramatically decrease the transcriptional activation of the p40 gene (34).
The VIP/PACAP-induced lack of c-Rel in the F1 complex could be related to the inhibition of c-Rel nuclear translocation as discussed above. In contrast, the reduced presence of IRF-1 in the F1 complex is probably due to a direct inhibitory effect of VIP/PACAP on IRF-1 gene expression. Unlike NF-B, IRF-1 is synthesized de novo following exposure to IFN␥ (55). In macrophages, the IRF-1 gene responds to IFN␥ through binding of the GAF complex generated by the Jak1/2-STAT1 pathway (reviewed in Ref. 56). Indeed, we have recently demonstrated that VIP and PACAP inhibit IRF-1 synthesis in macrophages by inhibiting IFN␥-induced Jak1/2 activation and the subsequent STAT1 phosphorylation through a mechanism that implies an increase in the intracellular cAMP levels. 2 In a previous study (22) we demonstrated that similar to the effect of VIP/PACAP on TNF␣ and inducible nitric-oxide synthase expression (14,48), the VPAC1-mediated inhibition of IL-12 secretion in peritoneal macrophages involves two transduction pathways, a cAMP-dependent and a cAMP-independent pathway. In the present study we sought to correlate the two pathways with the effects on the IL-12 transcriptional factors. The VPAC1 and VPAC2 are coupled primarily to the adenylate cyclase system (42), and IL-12 production is indeed inhibited by agents that increase intracellular cAMP levels (57)(58)(59). In the present study, forskolin and PGE 2 , two cAMPinducing agents, inhibited IL-12 production. In addition H89, a cAMP-dependent protein kinase A inhibitor, partially reversed the inhibitory effect of VIP/PACAP on IL-12 secretion and reversed the VIP effect on IRF-1 associated Ets-2 binding and IRF-1 expression at both protein and mRNA levels. In contrast, H89 did not reverse the inhibitory effect of VIP/PACAP on NF-B binding or on the nuclear translocation of c-Rel and p65. Also, H89 did not reverse the inhibitory effect of VIP/PACAP on reduction of cytoplasmic IB␣ levels. These results suggest that the cAMP-dependent pathway mediates IRF-1 expression, whereas the cAMP-independent pathway is responsible for the reduction in NF-B binding, presumably by stabilizing IB␣. This conclusion is supported by the effects of forskolin, a cAMP inducer. Forskolin inhibits both IRF-1 associated Ets-2 binding and IRF-1 expression but does not affect NF-B binding nor protein levels of c-Rel, p65, and IB. Similar observations were 2 M. Delgado and D. Ganea, submitted for publication. , or calphostin C (100 nM). Nuclear and cytoplasmic extracts obtained 4 h after LPS stimulation were analyzed by Western blotting with antibodies against p50, p65, c-Rel, and IRF-1 (nuclear extracts) or IB␣ (cytosolic fraction). One representative experiment of three is shown. made for TNF␣ and inducible nitric-oxide synthase expression in macrophages, where forskolin did not affect NF-B binding but changed the composition of the CRE-binding complexes and reduced IRF-1 binding, respectively (14,48). The effects of cAMP on NF-B are still debatable. For example, in some cell types, particularly thymocytes and T cells, cAMP-elevating agents reduced NF-B binding through stabilization of IB␣ and subsequent impairment of p65 nuclear translocation (60 -62). In contrast, other studies reported that the inhibition of the NF-B transcriptional activity by elevated cAMP or by cAMP-dependent protein kinase A overexpression does not result from an impaired nuclear translocation of the p50/p65 subunits but from the competition between the cAMP-induced CREB and NF-B for limited amounts of the coactivator CREBbinding protein (63,64). VIP and PACAP have been reported to increase CREB phosphorylation and CREB-regulated transcription in several cell types (65)(66)(67). Therefore, an additional mechanism in the VIP/PACAP inhibition of B-mediated transactivation of the IL-12 gene may involve the competition between NF-B and CREB for CREB-binding protein.
In conclusion, we have shown that the binding of VIP and PACAP to VPAC1 inhibits IL-12 production at a transcriptional level in LPS/IFN␥-stimulated Raw 264.7 macrophages through two intracellular pathways. The cAMP-dependent pathway preferentially inhibits IRF-1 mRNA expression, therefore affecting a component of the Ets-2 transcriptional complex. The cAMP-independent pathway inhibits the nuclear translocation of p65, a component of NF-B, and of c-Rel, a component of both NF-B and Ets-2 transcriptional complexes, by inhibiting the phosphorylation of IB␣ subunit.
The biological significance of the anti-inflammatory actions of VIP/PACAP relates to several areas of the immune response. Antigens and microbial products in particular are potent macrophage activators that induce the sequential release of early proinflammatory cytokines such as TNF␣, IL-1, IL-6, and IL-12, followed later by anti-inflammatory cytokines such as IL-10. TH cells, activated subsequent to macrophage stimulation, secrete a range of pro-and anti-inflammatory cytokines such as IL-2, IFN␥, IL-4, and IL-13. In normal physiological conditions the proinflammatory cytokine cascade is down-regulated in a timely manner by anti-inflammatory factors, such as various cytokines, hormones, and possibly neuropeptides such as VIP and PACAP. However, in pathological conditions such as septic shock, the activation of the proinflammatory cytokine network becomes excessive and leads ultimately to serious tissue damage, and possibly death. The central importance of TNF␣, IL-12, and IFN␥ in the pathogenesis of the endotoxic shock is indicated by the fact that pretreatment with corresponding neutralizing antibodies protects against lethality (68 -70). In this respect, the ability of the neuropeptides VIP/PACAP to inhibit TNF␣, IL-12, and subsequently IFN␥ and to stimulate IL-10 production, which in turn down-regulates TNF␣ and IL-12, may provide a new therapeutical tool for the down-regulation of the proinflammatory cytokine network.
Because IL-12 participates in T cell activation and CTL activity and promotes the differentiation of TH cells into the TH1 subset (reviewed in Refs. [23][24][25], VIP/PACAP might play a significant role in the down-regulation of cell-mediated immunity in vivo. Overall our results provide an understanding by which IL-12 is regulated by these neuropeptides, providing the means to manipulate TH1 and TH2 responses and thereby alleviate TH1-and TH2-associated diseases.