Vasoactive Intestinal Peptide and Pituitary Adenylate Cyclase-activating Polypeptide Inhibit Nuclear Factor-κB-dependent Gene Activation at Multiple Levels in the Human Monocytic Cell Line THP-1*

The neuropeptides vasoactive intestinal peptide (VIP) and pituitary adenylate cyclase-activating polypeptide (PACAP) suppress monocyte/macrophage production of proinflammatory agents. The transcription factor NF-κB regulates the transcription of most agents. VIP/PACAP inhibit NF-κB transactivation in the lipopolysaccharide-stimulated human monocytic cell line THP-1 at multiple levels. First, VIP/PACAP inhibit p65 nuclear translocation and NF-κB DNA binding by stabilizing the inhibitor IκBα. Second, VIP/PACAP induce phosphorylation of the CRE-binding protein (CREB) and its binding to the CREB-binding protein (CBP). This results in a decrease in p65·CBP complexes, which further reduces NF-κB transactivation. Third, VIP and PACAP reduce the phosphorylation of the TATA box-binding protein (TBP), resulting in a reduction in TBP binding to both p65 and the TATA box. All these effects are mediated through the specific receptor VPAC1. The cAMP/cAMP-dependent protein kinase pathway mediates the effects on CBP and TBP, whereas a cAMP-independent pathway is the major transducer for the effects on p65 nuclear translocation. Since NF-κB represents a focal point for various stimuli and induces the expression of many proinflammatory genes, its targeting by VIP and PACAP positions them as important anti-inflammatory agents. The VIP/PACAP inhibition of NF-κB at various levels and through different transduction pathways could offer a significant advantage over other anti-inflammatory agents.

Human monocytes can be induced to express a variety of genes involved in immune and inflammatory responses and cell adhesion. Following stimulation with microbial products like lipopolysaccharide (LPS), 1 monocytes secrete and/or express several inflammatory factors such as TNF␣, IL-12, IL-1, IL-6, nitric oxide, granulocyte-macrophage colony-stimulating factor, chemokines such as IL-8, RANTES (regulated on activation normal T cell expressed), macrophage chemotactic protein-1 and macrophage inflammatory protein-1␣/␤, the adhesion molecule ICAM-1, and the stimulatory complex for T cells, i.e. B7 and MHC class II molecules.
The pleiotropic transcription factor NF-B plays an important role in the transcriptional regulation of all these genes (reviewed in Ref. 1). NF-B occurs in both homo-and heterodimeric forms. The most common transcriptionally active form is a p50/p65 heterodimer (reviewed in Ref. 2). In unstimulated cells, NF-B is localized in the cytosol bound to inhibitor proteins, collectively termed IB. Stimulation results in IB phosphorylation, ubiquitination, and proteosomal degradation, followed by the rapid translocation of NF-B to the nucleus where it binds to specific B elements within promoters (1,2).
Several studies have shown that the transactivating activity of NF-B requires DNA binding and interaction with coactivators that bridge various transcriptional activators and components of the basal transcriptional machinery. The CREB-binding protein (CBP) is a ubiquitously expressed nuclear coactivator present in limiting amounts (reviewed in Ref. 3). A diverse and increasing number of transcription factors and some elements of the basal transcriptional machinery are able to form stable physical complexes with and respond to CBP (reviewed in Refs. 4 and 5). CBP functions as an integrator linking various transcription factors to the basal transcriptional apparatus, by binding to the basal transcription factor TFIIB, which in turn contacts the TATA box-binding protein (TBP) of the TFIID complex in the basal apparatus (6,7). The interaction of p65 with CBP is essential for NF-B transcriptional activity (8,9), and this interaction can be strengthen by p65 phosphorylation (6,10), or impeded by competition from other CBP-binding factors such as CREB, c-Jun, c-Fos, p53, steroid receptors, c-Myb, and Myo-D (7,(11)(12)(13).
Vasoactive intestinal peptide (VIP) and the structurally related pituitary adenylate cyclase-activating polypeptide (PACAP), two neuropeptides present in the lymphoid microenvironment, elicit a broad spectrum of biological functions, including the modulation of innate and adaptive immunity (re-viewed in Refs. 14 -17). VIP and PACAP down-regulate the innate response by inhibiting inducible nitric-oxide synthase expression and secretion of pro-inflammatory cytokines in stimulated macrophages (18 -23). VIP and PACAP affect the adaptive T cell response indirectly, by down-regulating B7.1/ B7.2 expression and the subsequent costimulatory function of activated macrophages (24) and directly by inhibiting IL-2 production and T cell proliferation (reviewed in Ref. 15). Many of the proinflammatory cytokines and costimulatory proteins affected by VIP and PACAP are known to be regulated by NF-B (1). In fact, we have previously demonstrated that VIP and PACAP inhibit NF-B nuclear translocation and DNA binding to several promoters in both murine macrophages and T cells (18,(25)(26)(27). However, the effect of VIP and PACAP on NF-B activation in human monocytes has not been investigated to date. In addition, although we showed that VIP and PACAP inhibit NF-B DNA binding, a direct effect on NF-Bdependent transcriptional activity has not yet been addressed. Furthermore, we also asked whether VIP and PACAP could regulate NF-B transcriptional activity through the regulation of coactivators. Our data show that VIP and PACAP decrease NF-B-dependent transcriptional activity in the LPS-stimulated human monocytic cell line THP-1. This effect is exerted at multiple levels. The neuropeptides inhibit NF-B nuclear translocation and DNA binding by inhibiting the IB kinase (IKK)-mediated IB phosphorylation/degradation. In addition, VIP and PACAP selectively inhibit the interaction of p65 with CBP, while increasing interactions between CBP and CREB. Furthermore, by inhibiting the LPS-induced MEKK1/MEK3/ MEK6/p38 MAPK pathway, the two neuropeptides inhibit TBP activation and its subsequent DNA binding and interaction with p65. The differential involvement of specific VIP/PACAP receptors and intracellular pathways was also addressed. The specific receptor VPAC1 mediates the effects of VIP/PACAP on p65 nuclear translocation, formation of p65⅐CBP complexes and TBP activation and formation of p65⅐TBP complexes. However, whereas a cAMP-independent pathway is primarily responsible for the effects on p65 translocation, the cAMP/PKA pathway mediates the effects on the availability and/or activation of the coactivators CBP and TBP.
Plasmids, Transfections, and Luciferase Assay-NF-B-dependent gene expression was evaluated using a luciferase reporter gene driven by four tandem copies of the enhancer (B 4 ) in a pUC vector (CLON-TECH, Palo Alto, CA). CRE-dependent gene expression was evaluated using a luciferase reporter gene (CLONTECH). The plasmid pRc/RSV-p65 containing the entire cDNA of p65 was kindly provided Drs. G. J. Nabel and J. Stein through the National Institutes of Health AIDS Research and Reference Reagent Program. The expression plasmid, pRc/RSV-mCBP.HA.RK containing the full-length mouse CBP cDNA with a hemagglutinin (HA) tag (6) was a generous gift of Dr. R. Goodman. IB degradation was assayed by transiently transfecting an enhanced green fluorescent protein (EGFP)-tagged IB signaling probe (CLONTECH) followed by flow cytometry analysis following the manufacturer's recommendations. Empty vectors pRc/RSV and pUC-18 (Invitrogen, Carlsbad, CA) were used to maintain a constant total DNA concentration in each experiment. To assess variations in transfection efficiencies, the cells were transfected with 2 g of the control plasmid pCH110 (Amersham Pharmacia Biotech) that expresses the lacZ gene. Levels of ␤-galactosidase were determined using the Galacto-Light assay system (Tropix Inc., Bedford, MA) and exhibited Ͻ15% variation between samples.
THP-1 cells were transiently transfected with a total of 10 -30 g of plasmid DNA using DEAE-dextran. Forty eight hours later, the cells were stimulated with LPS (500 ng/ml) in the absence or presence of VIP or PACAP, and 6 h later luciferase assays were carried out as recommended by the manufacturer (Promega). Luciferase activity, expressed in arbitrary light units, was corrected for protein concentration or normalized to coexpressed ␤-galactosidase levels.
RNA Extraction and Northern Blot Analysis-Northern blot analysis was performed according to standard methods. The probe for human TNF␣ was generated by RT-PCR as described previously using the following primers: TNF␣, 5Ј-GTTCCTCAGCCTCTTCTCCT-3Ј and 5Ј-ATCTATCTGGGAGGGGTCTT-3Ј (28). Signal quantitation was performed in a PhosphorImager SI (Molecular Dynamics, Sunnyvale, CA).
Immunoprecipitation Experiments and Western Blotting-For Western blot analysis, whole cell lysates, cytoplasmic fraction, or nuclear extract (see above) containing 20 -30 g of protein were subjected to reducing SDS-PAGE (12.5%). After electrophoresis and electroblotting the membranes were developed with the enhanced chemiluminescence detection system (ECL, Amersham Pharmacia Biotech).
Phosphorylated MEK3 and MEK6 were detected by immunoprecipitation with anti-MEK3 or anti-MEK6 Abs, followed by SDS-PAGE and immunoblotting with an Ab against phosphorylated MEK3/6.
Interaction of CBP or TBP with p65 and/or CREB was assessed by immunoprecipitation of cell extracts (200 g) with 1-2 g of anti-p65 or anti-CREB antibodies, followed by treatment with 25 l of protein A/G-Sepharose beads (Sigma). After extensive washing and boiling in 1ϫ SDS sample buffer, the complexes were subjected to Western blotting with anti-CBP or anti-TBP antibodies. For transient transfections with the CBP vector, we used an anti-hemagglutinin Ab (anti-HA; 12CA5 murine monoclonal antibody against influenza HA peptide; Roche Molecular Biochemicals) to detect transfected HA-tagged CBP. After detection with an appropriate secondary antibody-conjugated peroxidase, proteins were visualized by enhanced chemiluminescence.
In Vitro Kinase Assays-In vitro kinase assays were performed as described previously with some modifications (30). Endogenous IKK␣ and p38 MAPK were immunoprecipitated from cell lysates (150 -250 g/sample) by incubation with 0.5 g of anti-IKK␣ or anti-p38 MAPK antibodies, for 2 h at 4°C. The immune complexes were harvested with protein A/G-Sepharose for 45 min at 4°C. The beads were extensively washed and resuspended in 30 l of kinase buffer with 15 M ATP, 10 Ci of [␥-32 P]ATP (3000 Ci/mmol), containing 5 g of recombinant IB␣ (for IKK␣) or TBP (for p38 MAPK). The kinase reaction was performed at 30°C for 30 min and stopped by the addition of 15 l of 2ϫ SDS sample buffer. Following boiling for 5 min, the samples were subjected to SDS-PAGE (9%), electroblotting, and autoradiography.
Immunoblotting of Proteins Bound to the Proximal Region of the TNF␣ Promoter-A double-stranded oligonucleotide spanning the proximal region of the human TNF␣ promoter (Ϫ661 to Ϫ1), generated from THP-1 genomic DNA by PCR (primers used, 5Ј-TCAGAATGAAA GAA-GAGGGCC-3Ј and 5Ј-GGCTGGGTGTGCCAACAACT-3Ј) and biotinylated in our molecular biology facility, was coupled to Dynabeads M-280 streptavidin (Dynal, Lake Success, NY) according to the manufacturer's recommendations. The TNF␣ promoter-coupled matrix (250 g) was incubated with 25 l of THP-1 nuclear extract in binding buffer (50 mM NaCl, 5 mM MgCl 2 , 10 mM Tris, pH 7.5, 1 mM dithiothreitol, 1 mM EDTA, 0.25 g/ml poly(dI)⅐poly(dC), and 5% glycerol) for 20 min at room temperature while mixing every 5 min to keep the Dynabeads in suspension. The flow-through fractions were collected; the Dynabeads were washed four times with binding buffer containing 0.5 g/ml poly(dI)⅐poly(dC), and the bound proteins were solubilized in sample buffer, boiled, and subjected to SDS-PAGE. Membranes were probed with antibodies against p65, p50, CREB, c-Jun, TBP, HMG-I(Y), CBP, or NF-Y at dilutions ranging from 1:2,500 to 1:10,000 followed by chemiluminescent detection.
ELISA-TNF␣ levels in culture supernatants were determined using a human-specific sandwich ELISA (PharMingen) following the manufacturer's instructions.
RT-PCR for the Detection of VPAC1, VPAC2, and PAC1 mRNA Expression-THP-1 cells were cultured at a concentration of 2 ϫ 10 6 cells/ml in 100-mm tissue culture dishes and stimulated with LPS (0.5 g/ml) for up to 12 h. Two g of total RNA was reverse-transcribed, and cDNA was amplified with specific primers. Glyceraldehyde-3-phosphate dehydrogenase primers (Stratagene) were used as control. The primers for VPAC1, VPAC2, and PAC1 receptors have been described before (25).

VIP and PACAP Inhibit LPS-induced TNF␣ Production in
Human Monocytes-We have recently demonstrated that VIP and PACAP inhibit TNF␣ production in murine peritoneal macrophages and the macrophage cell line Raw 264.7 (19,25).
To investigate the effects of VIP/PACAP on human monocytes, THP-1 cells were stimulated with LPS in the presence or absence of VIP or PACAP, and the amounts of secreted TNF␣ were determined by ELISA. VIP and PACAP inhibit TNF␣ production in a dose-dependent manner, with a maximum effect in the concentration range of 10 Ϫ8 -10 Ϫ6 M (Fig. 1A). Both neuropeptides inhibit in a time-dependent manner TNF␣ steady-state mRNA levels (Fig. 1B). MG132, a newly described NF-B inhibitor, shows a similar inhibitory effect on TNF␣ mRNA (Fig. 1B).

VIP and PACAP Inhibit LPS-and TNF␣-induced NF-B-dependent Transcription in Monocytes-Since LPS up-regulates
TNF␣ transcription through an NF-B-dependent mechanism (31, 32), we determined whether VIP and PACAP inhibit NF-B transcriptional activity. THP-1 cells were transiently transfected with the (B) 4 -luciferase reporter plasmid, containing four copies of the NF-B consensus site. Forty eight hours later, the cells were stimulated with LPS or TNF␣ in either the presence or absence of VIP or PACAP and were assayed for NF-B-dependent transcription 5 h later. Both LPS and TNF␣ led to an ϳ18-fold increase in NF-B transcriptional activity ( Fig. 2A). Treatment with VIP or PACAP strongly inhibits LPSor TNF␣-induced NF-B activity ( Fig. 2A). The inhibitory effect is dose-dependent (Fig. 2B).
VIP and PACAP Inhibit NF-B Nuclear Translocation by Preventing LPS-induced Phosphorylation/Degradation of IB␣-To assess whether VIP and PACAP inhibit NF-B DNA binding, EMSAs were performed. Stimulation of THP-1 cells with LPS led to strong NF-B binding showing a maximum increase at 2 h after stimulation, and VIP and PACAP inhibit binding at all time points (Fig. 3A). The binding specificity was ascertained by the displacement with 50-fold excess of unlabeled homologous (NF-B) but not nonhomologous oligonucleotide (CRE) (Fig. 3A). Antibody supershift experiments indicate the presence of both p50 and p65, with no supershift for an irrelevant Ab (anti-CREB) (Fig. 3A).
The primary level of control for NF-B is mediated through its interaction with the inhibitor IB. VIP and PACAP could inhibit NF-B activity by blocking LPS-induced IB degradation and subsequent NF-B nuclear translocation. We measured the levels of p65 in cytoplasm and nucleus. As expected, p65 was predominantly localized in the cytoplasm of unstimulated cells, and LPS induced a decrease in the level of cytoplasmic p65, and an increase in nuclear p65 levels (Fig. 3B). VIP and PACAP abolished the LPS-induced change in p65 levels ( Fig. 3B), which indicates an inhibition of p65 nuclear translo- cation. Similar levels of p50 indicate equal protein loading. To determine whether VIP/PACAP interfere with the LPS-induced degradation of IB, we examined cytoplasmic IB␣ levels. As expected, we observed a time-dependent IB␣ degradation, paralleled by an increase in IB␣ phosphorylation in LPS-stimulated cells (Fig. 3C). VIP and PACAP block the phosphorylation and subsequent degradation of IB␣ (Fig. 3C). In addition, we determined the effect of VIP and PACAP on IB degradation by transiently transfecting an EGFP-tagged IB construct in THP-1 cells and assaying the percentage of fluorescent IB ϩ cells at different time points by flow cytometry. LPS led to a rapid decrease in IB ϩ cells, whereas VIP significantly increased IB half-life in LPS-treated cells (Fig. 3C,  right panel).

FIG. 2. VIP and PACAP inhibit LPSand TNF␣-induced NF-B transcriptional activity.
THP-1 cells were transfected with the (B) 4 -Luc construct (10 g) and treated 48 h later with LPS or TNF␣ (10 ng/ml), with or without VIP or PACAP (10 Ϫ8 M for A) for 6 h. Cytosolic extracts (100 g) were used in luciferase assays. Fold induction is relative to luciferase activity in unstimulated cells. Data are expressed as the mean Ϯ S.D. of three independent experiments performed in duplicate.

FIG. 3. VIP and PACAP prevent LPS-induced IB degradation and the subsequent NF-B nuclear translocation.
A, VIP and PACAP inhibit NF-B DNA binding. Nuclear extracts from cells incubated for 2 h with LPS with or without VIP or PACAP (10 Ϫ8 M) were used in EMSA. Probe, consensus NF-B. Supershift, nuclear extracts were incubated with polyclonal antibodies against p65, p50, or CREB for 20 min before the addition of the B probe. One representative experiment of three is presented. Right panel, NF-B binding at various time points expressed as arbitrary densitometric units. Data represent the mean Ϯ S.D. of three independent assays. B, VIP and PACAP inhibit p65 translocation. Cells were treated with LPS with or without VIP or PACAP (10 Ϫ8 M) for 1 h. Cytosolic and nuclear proteins were extracted, and Western blot analysis was performed for p50 and p65. One representative experiment of three is shown. C, VIP and PACAP prevent IB␣ phosphorylation and degradation. Left panels, THP-1 cells were stimulated with LPS with or without VIP or PACAP (10 Ϫ8 M). The cytosolic amounts of IB␣ and phosphorylated IB␣ at different time points were determined by Western blot. One representative experiment of three is shown. Right panels, THP-1 cells were transiently transfected with a fluorescent IB-EGFP signaling probe and treated with LPS with or without 10 Ϫ8 M VIP 24 h later. The percentage of IB ϩ cells was determined by flow cytometry at different time points. Similar results were obtained in three independent experiments. D, VIP and PACAP inhibit IKK␣ activity. THP-1 cells were stimulated with LPS with or without VIP or PACAP (10 Ϫ8 M) (10 min in upper panel). IKK␣ activity was assayed in an in vitro kinase assay. Lower panel, IKK␣ activity is expressed as arbitrary densitometric units. Data represent the mean Ϯ S.D. of three independent assays. IKK␣ protein amounts were determined by immunoblotting (upper panel; control). E, overexpression of p65 partially reverses the inhibitory effect of VIP and PACAP. THP-1 cells were transiently cotransfected with the (B) 4 -Luc construct (10 g) and increasing concentrations (0, 2.5, 5, 10, or 15 g) of pRSV-p65 vector (p65). The cells were stimulated 48 h later with LPS with or without 10 Ϫ8 M VIP or PACAP and incubated for 6 h before determining luciferase activity. Fold induction is relative to luciferase activity in unstimulated cells. Results represent the mean Ϯ S.D. of three independent experiments performed in duplicate.
Since the LPS activation of NF-B requires IKK-mediated phosphorylation of IB␣ (1, 2), we determined if VIP and PACAP inhibit IKK activity by using an in vitro kinase assay. Stimulation of THP-1 cells with LPS resulted in a time-dependent increase in IKK␣ activity, which was inhibited by VIP and PACAP (Fig. 3D). No differences in IKK␣ expression were observed (Fig. 3D).
These results demonstrate that VIP and PACAP inhibit NF-B nuclear translocation and subsequent DNA binding in LPS-stimulated cells by blocking the IKK-mediated IB phosphorylation/degradation. If the inhibitory effect of VIP and PACAP on NF-B transcriptional activity is mediated entirely through the inhibition of NF-B nuclear translocation, overexpression of p65 should reverse this effect. Therefore, THP-1 cells were transiently cotransfected with the (B) 4 -luciferase reporter plasmid and increasing concentrations of a vector expressing p65. Increasing concentrations of p65 only partially reversed the inhibitory effect of VIP and PACAP, suggesting that the neuropeptides affect more than NF-B nuclear translocation.
Neither VIP nor PACAP Affect LPS-induced p65 Phosphorylation-Several studies have demonstrated that p65 is phosphorylated during in vivo NF-B activation, leading to increased transcriptional activity (10,33). We examined the effects of VIP and PACAP on the phosphorylation of transiently transfected p65. VIP and PACAP had no effect on LPS-induced p65 phosphorylation (Fig. 4, upper panel). Similar levels of p65 were detected by Western blotting in THP-1 cells transfected with p65 in the presence or absence of LPS, VIP, or PACAP (Fig. 4,  lower panel).
VIP and PACAP Promote CREB/CBP Versus p65/CBP Interactions by Increasing CREB Phosphorylation/Activation-In addition to DNA binding, the interaction of p65 with CBP is essential for optimal NF-B transcriptional activity (9). In addition to p65, CBP interacts with other factors including CREB (6). Changes in p65 phosphorylation or competition with other factors for the limiting quantities of nuclear CBP lead to changes in p65/CBP interactions. THP-1 cells were stimulated with LPS in the absence or presence of VIP or PACAP, and total cell lysates were immunoprecipitated with antibodies to p65 or CREB and probed for the presence of CBP. LPS stimulation results in the appearance of p65⅐CBP complexes (Fig.  5A). No p65⅐CBP complexes are detected in unstimulated cells. VIP and PACAP decrease the levels of p65/CBP and increase the levels of CREB⅐CBP complexes (Fig. 5A). Moreover, VIP and PACAP induce CREB⅐CBP instead of p65⅐CBP complexes even in the presence of overexpressed CBP (Fig. 5A).
To confirm that the VIP/PACAP inhibition of NF-B transcriptional activity is related to the reduction in p65⅐CBP complexes, THP-1 cells were cotransfected with the (B) 4 -luciferase reporter system and increasing concentrations of p65 and/or CBP. Expression of increasing concentrations of CBP led a partial reversal of the inhibitory effect of VIP/PACAP on NF-B activation (Fig. 5B, upper panel). A similar conclusion was reached earlier regarding p65 (Fig. 3E). However, the coexpression of CBP (fixed concentration) and p65 (increasing concentrations) completely reversed the VIP/PACAP effect (Fig. 5B,  upper panel). This correlates with the fact that coexpression of p65 and CBP restored the p65⅐CBP complexes to levels observed in the LPS-treated cells (Fig. 5B, lower panel).
The fact that, even in the presence of excess p65, the levels of p65⅐CBP complexes and the NF-B transcriptional activity are not completely restored (Figs. 5B and 3F) indicates that VIP and PACAP affect the formation of p65⅐CBP complexes through both a reduction in nuclear p65 and a direct effect on CBP.
Since VIP receptors are mostly linked to the cAMP/PKA pathway, it is highly possible that VIP and PACAP activate CREB which then recruits CBP. Therefore, we analyzed the effects of VIP and PACAP on CREB phosphorylation. LPS increases CREB phosphorylation slightly as compared with unstimulated controls (Fig. 5C, upper panels). In contrast, VIP and PACAP strongly augment the levels of phosphorylated CREB (Fig. 5C, upper panels). Total CREB levels were not affected by either treatment. In addition, CREB levels in cytoplasmic, and nuclear extracts were assayed by Western blotting. LPS stimulation results in a slight increase in nuclear CREB, and treatment with VIP or PACAP leads to high levels of nuclear CREB (Fig. 5C, lower panels).
To determine whether VIP/PACAP-induced CREB phosphorylation correlates with increased DNA binding and CREB-dependent transcription, we performed EMSAs using a consensus CRE site and transient transfections with a CRE-luciferase reporter plasmid. LPS leads to a slight increase in CRE DNA binding, and VIP and PACAP strongly augment this binding (Fig. 5D). The binding specificity was confirmed by using homologous (CRE) and nonhomologous (NF-B) oligonucleotides as competitors (Fig. 5D). The CRE complexes are supershifted by an anti-CREB Ab but not by an anti-p65 Ab (Fig. 5D). In cells transfected with CRE-luciferase constructs, VIP and PACAP significantly increase the CRE-dependent transcriptional activity, as compared with cells treated with LPS alone (Fig. 5D). These results indicate that VIP and PACAP increase the phosphorylation/activation of CREB which then competes with p65 for limiting amounts of CBP, resulting in increased CREB⅐CBP and decreased p65⅐CBP complexes.
VIP and PACAP Reduce LPS-induced TBP DNA Binding Activity and Its Interaction with p65 by Inhibiting the MEKK1-MEK3/6-p38 MAPK Pathway-Since NF-B-driven transcription also depends on the activation of basal transcription factors, such as TFIIB and TFIID (TBP), we investigated if VIP and PACAP regulate the basal transcriptional factors. We determined first the effect of VIP and PACAP on TBP binding to the TATA box. LPS increases TBP binding, and VIP and PACAP reduce it to control levels. The specificity of TBP binding is indicated by competition of 50-fold excess of unlabeled homologous (TBP), but not nonhomologous, oligonucleotide (NF-AT). The TBP complexes are supershifted by an anti-TBP Ab but not by an irrelevant Ab (anti-CREB) (Fig. 6A).
Since the association of the carboxyl terminus of p65 with TBIIB and TBP is known to be important for the transcriptional regulation of NF-B (34, 35), we determined whether VIP and PACAP regulate p65/TBP interactions by immunoprecipitating cell lysates with anti-p65 Abs and immunoblotting for TBP. LPS increases this interaction, and VIP and PACAP inhibit the LPS-induced p65/TBP interaction (Fig. 6B, upper  panel). To determine whether the effect of VIP/PACAP is due to the previously described inhibition of p65 nuclear translocation, THP-1 cells were transiently transfected with increasing concentrations of p65. Overexpression of p65 partially reversed the effect of VIP (Fig. 6B, lower panel, compare 2nd to 5th lanes  to the 6th lane). However, the incomplete reversal suggests that VIP and PACAP might directly regulate TBP activation.
Since TBP is activated following phosphorylation by the p38 MAP kinase, we evaluated the effect of VIP and PACAP on TBP phosphorylation. Whereas no TBP phosphorylation is observed in unstimulated cells, LPS induces high levels of phosphorylated TBP (Fig. 6C). VIP and PACAP inhibit TBP phosphoryl-  4 -Luc construct (10 g) and increasing concentrations (0 g for 1st and 2nd lanes, and 2.5, 5, 10, and 15 g for 3rd to 6th lanes, and 5 g for 7th to 10th lanes) of pRSV-CBP vector (CBP) and increasing concentrations (0 g for 1st to 6th lanes and 2.5, 5, 10, or 15 g for 7th to 10th lanes) of pRSV-p65. After 48 h, cells were stimulated with LPS in the absence (1st lane) or presence of 10 Ϫ8 M VIP (2nd to 10th lanes) and incubated for an additional 5 h before determining luciferase activity. Fold induction is relative to luciferase activity in unstimulated cells. Results represent the mean Ϯ S.D. of three independent experiments performed in duplicate. Lower panels, THP-1 cells were cotransfected with increasing concentrations (2.5, 5, 10 and 15 g for 1st to 4th lanes, 5 g for 5th to 8th lanes) of pRSV-p65 vector (p65) and increasing concentrations (0 g for 1st to 4th lanes and 2. ation, similar to the p38 MAP kinase inhibitor SB 203580 (Fig.  6C). We examined the effect of VIP and PACAP on p38 MAPK activity with TBP as substrate. Treatment with LPS results in a time-dependent increase in p38 MAPK activity (Fig. 6D). VIP and PACAP inhibit the LPS-induced p38 MAPK-mediated phosphorylation of TBP, without affecting TBP and p38 MAPK protein levels (Fig. 6D).
The activation of p38 MAPK in response to LPS or proinflammatory signals involves a kinase cascade with the upstream activator MEKK1 phosphorylating and activating both MAPK kinases MEK3 and MEK6, which in turn activate p38 MAPK by phosphorylation at both threonine and tyrosine residues (36 -38). Therefore, we investigated the effect of VIP and PACAP on phosphorylation of MEK3, MEK6, and p38. LPS leads to the strong phosphorylation of p38, MEK3 and MEK6, and VIP and PACAP significantly reduce the phosphorylation of all these kinases (Fig. 6D). None of these treatments affected the expression of any of the kinases assayed (Fig. 6D). We conclude that the VIP/PACAP inhibition of TBP phosphorylation is mediated through the inhibition of the MEKK1/MEK3/ MEK6/p38 MAPK cascade.

VIP and PACAP Change the LPS-induced Composition of Nuclear Factors Bound to TNF␣ Promoter in Human Monocytic
Cells-We next investigated the effects of VIP and PACAP on the LPS-induced transcriptional activators binding the proximal regulatory region in the human TNF␣ promoter, which contains two essential transactivating binding sites, i.e. the NF-B and CRE elements (31,32). Depending on the activation state, the CRE site may bind either CREB or c-Jun, with CREB preferentially bound in unstimulated cells and c-Jun in LPSstimulated cells (25,32). To determine whether the activators present in nuclear extracts would bind to the TNF␣ promoter, a biotinylated affinity matrix spanning the proximal regulatory region of human TNF␣ promoter was generated and coupled to streptavidin-coated magnetic beads. This biotinylated probe was incubated with nuclear extracts from unstimulated or LPS-stimulated THP-1 cells treated with or without VIP or PACAP. Transcription factor complexes were released from the magnetic beads by boiling in SDS sample buffer and detected by immunoblotting.
p65 is present in LPS-treated samples, but not in unstimulated or stimulated cells treated with VIP or PACAP (Fig. 7,  p65, input). The p65 present in the LPS-treated cells binds to the TNF␣ promoter region (Fig. 7, p65, bound). In contrast, p50 is constitutively expressed in THP-1 cells and binds partially to the TNF␣ promoter (Fig. 7, p50, input, bound, and flow-thru). The p50 binding is not affected by LPS, VIP, or PACAP (Fig. 7,  p50). CREB is slightly induced by LPS and highly induced by VIP and PACAP (Fig. 7, CREB, input). All induced CREB binds to the TNF␣ promoter region (Fig. 7, CREB, bound, and flowthru). Both TBP and CBP are constitutively present in the nucleus, and neither LPS not VIP/PACAP affect their levels (Fig. 7, CBP and TBP, input). However, whereas CBP from unstimulated, LPS-stimulated, and VIP/PACAP-treated cells is all bound to the TNF␣ promoter region, binding of TBP is induced by LPS and inhibited by VIP and PACAP (Fig. 7, CBP  and TBP, bound). Similar to TBP, c-Jun is present in the nucleus and binds to the TNF␣ promoter region when the cells are stimulated with LPS, and this binding is inhibited by VIP and PACAP (Fig. 7, c-Jun). As a control, we used the nuclear factor-Y (NF-Y), a transcription factor present in the nucleus which binds constitutively to various promoters, including TNF␣. None of the treatments affected NF-Y binding (Fig. 7,  NF-Y).
Finally, this experimental design allowed us to investigate an additional possible regulatory element of the NF-B trans- Left panels, at different time points (15 min for blots), p38 MAPK activity was assayed in an in vitro kinase assay, with TBP as substrate. p38 MAPK activity is expressed as arbitrary densitometric units. Data represent the mean Ϯ S.D. of three independent assays. As control, the amounts of TBP were determined by immunoblotting. Right panels, the levels of phosphorylated MEK3, phosphorylated MEK6, and phosphorylated p38 MAPK were determined by Western blotting (15 min). The amounts of p38 MAPK, MEK3, MEK6, and MEKK1 were determined by immunoblotting. One representative experiment of three is shown. activation, i.e. the nonhistone chromosomal proteins of the high mobility group (HMG)-I(Y) family, two chromatin architectural proteins that play a role in the transcriptional regulation of certain mammalian genes (39,40). HMG-I(Y) was shown to enhance the DNA binding of several transcription factors, including NF-B (41,42). HMG-I(Y) was present in the nucleus from unstimulated, LPS-stimulated, and VIP/PACAP-treated THP-1 cells, and none of the treatments affected its binding to the TNF␣ promoter, suggesting that HMG-I(Y) is an unlikely element in the regulation of NF-B activation by VIP and PACAP.
Involvement of VPAC1 and cAMP/PKA in the Effects of VIP and PACAP on NF-B-mediated Gene Activation-Although THP-1 cells were previously shown to express VIP/PACAPbinding sites, primarily coupled to cAMP production (43), the nature of these binding sites was not elucidated. We investigated the expression of VPAC1, VPAC2, and PAC1 by RT-PCR in unstimulated and LPS-stimulated THP-1 cells. Both VPAC1-and PAC1-specific fragments were amplified from unstimulated and stimulated monocytes, whereas VPAC2 fragments were only detected in stimulated cells (Fig. 8A).
Our previous studies identified VPAC1 and cAMP as the major mediators of VIP/PACAP effects on macrophage-derived cytokines (18 -21, 22-27). However, VIP and PACAP inhibit NF-B nuclear translocation in murine macrophages through a cAMP-independent pathway (18,(25)(26)(27). To determine the receptor and the transduction pathways involved, we used specific VPAC antagonists and a specific protein kinase A inhibitor (H89). The VIP inhibition of NF-B transcriptional activity is completely reversed by the VPAC1 antagonist and only slightly reversed by increasing concentrations of H89. In addition, forskolin (a cAMP-inducing agent) mimics only partially the effect of VIP (Fig. 8B). These findings suggest that whereas the effect of VIP on NF-B activation is entirely VPAC1-dependent, it is mediated by both a cAMP-dependent and a cAMP-independent pathway. When NF-B DNA binding and IB phosphorylation were analyzed, we found again that the VPAC1 antagonist completely reversed the inhibitory effect of VIP and that H89, even at the highest concentrations used, only minimally reversed this effect. This correlates with the fact that forskolin inhibits only weakly the NF-B DNA binding and IB phos-phorylation (Fig. 8C). Therefore, the major pathway for the inhibition of NF-B nuclear translocation by VIP is non-cAMP mediated.
In contrast, the VPAC1 antagonist and the PKA inhibitor completely reversed the effect of VIP on CREB phosphorylation, and forskolin entirely mimicked the effect of VIP (Fig. 9A). A similar conclusion was reached for the effect of VIP on the preferential induction of CREB⅐CBP versus p65⅐CBP complexes (Fig. 9A). Finally, the VPAC1 antagonist and H89 also reversed the effects of VIP on the phosphorylation of TBP and the activation of the MEKK1/MEK3/p38 MAPK pathway, with forskolin mimicking the effects of VIP (Fig. 9B). These results suggest that the regulatory activities of VIP on CBP and TBP are mediated entirely through the cAMP/PKA pathway. DISCUSSION VIP and PACAP control inflammatory processes by suppressing the monocyte/macrophage production of several proinflammatory factors known to be transcriptionally controlled by NF-B (21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32). Although VIP and PACAP have been found to inhibit the translocation of NF-B and its DNA binding in mouse macrophages (18,(25)(26)(27), the effects of these neuropeptides on NF-B-dependent transcriptional activity and the detailed molecular mechanisms governing this process have still to be elucidated. The present study shows that VIP and PACAP specifically inhibit the LPS-induced NF-B transcriptional activity in the human monocytic cell line THP-1.
VIP and PACAP operate at three different levels to inhibit the NF-B transcriptional activity. First, VIP and PACAP inhibit p65 nuclear translocation and subsequent DNA binding. This process is mediated through the VPAC1 receptor and a non-cAMP transduction pathway. Second, VIP and PACAP induce CREB phosphorylation, and the phosphorylated CREB competes with p65 for the coactivator CBP. Third, VIP and PACAP inhibit the MEKK1/MEK3/6/p38 pathway ultimately affecting the phosphorylation of TBP and its binding to both p65 and the TNF␣ promoter. The effects on CBP and TBP are both mediated through VPAC1 and the cAMP/PKA transduction pathway.
The inhibition of p65 translocation by VIP/PACAP is mediated through the stabilization of IB␣. IB␣, a member of the FIG. 7. VIP and PACAP change the composition of nuclear factors bound to the TNF␣ promoter. Nuclear extracts prepared from THP-1 cells incubated for 2 h with LPS with or without VIP or PACAP (10 Ϫ8 M) were incubated with a biotinylated oligonucleotide spanning the proximal region of the human TNF␣ promoter. Bound proteins were identified by Western blotting using the indicated antibodies. One representative experiment of three is shown.
IB family, is the major player in the response to inflammatory stimuli. Upon phosphorylation at specific serine residues by the kinases IKK␣ and -␤, IB␣ is ubiquitinated and degraded by the 26 S proteosome (reviewed in Ref. 2). As previously demonstrated for murine macrophages and T cells (27), 2 VIP and PACAP inhibit IB␣ phosphorylation and its subsequent degradation. This is accomplished through an inhibitory effect on IKK␣. Similar results were obtained in this study. The inhibitory effect of VIP and PACAP on p65 nuclear translocation and NF-B DNA binding has functional consequences, since overexpression of p65 partially reverses this inhibitory effect. The fact that the reversal is incomplete suggests additional mechanisms for the inhibition of NF-B transcriptional activity. To investigate this hypothesis several possibilities were considered.
It has been reported that the catalytic subunit of PKA associates with the cytosolic NF-B⅐IB complex phosphorylating p65 and that p65-mediated transcription is strongly dependent on its phosphorylation (10). However, our results clearly demonstrate that neither VIP nor PACAP affect the in vivo LPSinduced phosphorylation of p65.
An additional regulatory element in the NF-B transcriptional activity is the coactivator CBP. CBP performs an important role in the integration of diverse signaling pathways by 2 M. Delgado and D. Ganea, submitted for publication.

FIG. 8. Involvement of VPAC1 and cAMP/PKA in NF-B gene activation and IKK␣ activity.
A, expression of VPAC1, VPAC2, and PAC1 mRNA in THP-1 cells. Total RNA extracted from unstimulated and LPS-stimulated (12 h) THP-1 cells was subjected to RT-PCR with specific primers for VPAC1, VPAC2, and PAC1. One representative experiment of two is shown. B and C, THP-1 cells were activated with LPS in the absence (1st lanes) or presence of VIP (10 Ϫ8 M, 2nd lanes), or forskolin (10 Ϫ6 M, 5th lanes). VPAC1 antagonist (10 Ϫ7 M, 3rd lanes) or H89 (100 ng/ml, 4th lanes) were added simultaneously with VIP (10 Ϫ8 M). B, NF-B gene activation was analyzed as described in Fig. 2. Results represent the mean Ϯ S.D. of three independent experiments performed in duplicate. C, upper panels, NF-B binding was analyzed 1 h after stimulation as described in Fig. 3. Lower panels, IKK␣ activity (20 min) was analyzed with IB␣ as substrate by using an in vitro kinase assay. For blots, one representative experiment of three is shown. Dotted lines in graphs B and C are control values from samples treated with LPS alone. Results shown in graphs are the mean Ϯ S.D. of three independent experiments performed in duplicate. linking p65 with components of the basal transcriptional machinery, such as TFIIB, TBP, and histone acetyltransferases (44). The present report demonstrates that VIP and PACAP indeed inhibit the formation of p65⅐CBP complexes and that this event is directly related to the inhibition of NF-B transcriptional activity. The fact that overexpression of p65 did not completely reverse the VIP/PACAP inhibition of p65⅐CBP complex formation suggests that VIP/PACAP might directly affect CBP. Since CBP is in limiting amounts in the nucleus and is capable to interact with several transcriptional factors (3, 6 -13), competition for CBP provides another mechanism for transcriptional regulation (8,45,46). CBP binds to phosphorylated CREB, and formation of CREB⅐CBP complexes reduces the CBP available for complexing with p65 (8,9). VIP/PACAP were shown to induce CREB DNA binding in activated murine macrophages (25,26). This study shows that VIP and PACAP increase CBP binding to CREB, replacing p65⅐CBP with CREB⅐CBP complexes in LPS-stimulated THP-1 cells. This is due to VIP/PACAP-induced increases in CREB phosphorylation/activation. The fact that cotransfections with p65 and CBP completely reverse the inhibitory effect of VIP/PACAP on NF-B transcriptional activity, whereas p65 and CBP sepa-rately result in only a partial reversal, suggests that VIP/ PACAP operate by inhibiting both p65 nuclear translocation and CBP availability.
The observation that VIP and PACAP induce high levels of nuclear CREB was unexpected. In most systems, CREB is localized in the nucleus in both stimulated and unstimulated cells and becomes transcriptionally active upon phosphorylation (reviewed in Ref. 47). However, in our system, the unstimulated THP-1 cells express mostly cytoplasmic CREB, and VIP/PACAP induce significant levels of nuclear CREB. This might be a characteristic of this particular cell line, with VIP/ PACAP contributing to the retention of phosphorylated CREB in the nucleus.
In addition, p65 was shown to interact with TBP and TFIIB of the basal transcriptional complex, and these interactions appear essential for optimal NF-B-driven transcription (48). The interaction of p65 with TBP and TFIIB is presumably mediated through CBP (reviewed in Ref. 4). Our study indicates that VIP and PACAP inhibit LPS-induced TBP binding to the TATA box and p65/TBP interaction. Similar to CBP, TBP is found in limiting amounts in the nucleus but is activated following LPS stimulation by the MEKK1/MEK3/MEK6/p38 MAPK pathway (36 -38, 50). Several studies demonstrated the involvement of p38 MAPK in the activation of NF-B (50 -52). VIP and PACAP inhibit the p38 MAPK pathway and the subsequent TBP phosphorylation/activation. Finally, we investigated another possible regulatory element in this system. HMG-I and HMG-Y are two architectural proteins that facilitate the assembly of functional nucleoprotein complexes by modifying DNA conformation and recruiting nuclear proteins to the promoter. HMG-I(Y) also enhance the DNA binding of NF-B (41). VIP and PACAP did not affect either the nuclear expression or the DNA binding of HMG-I(Y) in LPS-stimulated THP-1 cells.
VIP and PACAP act through three specific receptors, i.e. VPAC1, VPAC2, and PAC1 (53). Human monocytes, including THP-1 cells, have been shown to express VIP/PACAP-binding sites (43), and our data demonstrate that, similar to mouse macrophages (21,25), THP-1 cells express VPAC1 and PAC1 constitutively, and VPAC2 following LPS stimulation. Although stimulated cells express all three receptors, our studies using specific antagonists show that VPAC1 is the major receptor involved in the VIP/PACAP regulation of all the previously discussed aspects of NF-B-driven transcription.
Intracellular cAMP is the major secondary mediator induced by VPAC1 (53). However, in most cases, the effects of VIP/ PACAP on cytokine production in both macrophages and freshly activated T cells are mediated through both a cAMPdependent and -independent pathway (16 -18, 25, 27, 54). Similarly, in the present study we have found that VIP and PACAP regulate NF-B transcriptional activity through both cAMP-dependent and -independent pathways. VIP/PACAP regulation of both the MEKK1/MEK3/p38 MAPK-mediated TBP activation and CREB/CBP interactions is entirely cAMP-dependent. CREB phosphorylation was previously shown to be mediated by the cAMP/PKA pathway (reviewed in Ref. 47). On the other hand, the neuropeptide-mediated inhibition of IB phosphorylation and p65 nuclear translocation is mainly cAMP-independent. This is in agreement with previous reports showing that the NF-B nuclear translocation and DNA binding in human monocytes and endothelial cells and in murine macrophages is cAMP-independent (8,18,25,27).
A proposed model representing the VIP/PACAP regulation on LPS-induced NF-B transcriptional activity is provided in Fig. 10. LPS stimulation leads to IKK-mediated IB phosphorylation/degradation and the subsequent nuclear translocation of p65/p50 heterodimers that bind to different B sites in the promoter. In parallel, LPS, presumably through MEKK1, induces both c-Jun phosphorylation and its binding to the CRE site, and TBP activation and its binding to the TATA box. The binding of HMG-I(Y) to multiple sites increases the binding affinity of NF-B and bends DNA to facilitate the formation of a higher order transcriptional complex. These events place multiple transcriptional activators in a favorable position to compete for the coactivator CBP present in limiting amounts. CBP acts as an efficient integrator, bridging transactivators to the components of the basal machinery TFIIB and TBP. Since RNA polymerase II is constitutively associated with CBP, binding of the coactivator to the promoter facilitates the recruitment of the polymerase. VIP and PACAP dramatically change this optimal transcriptional conformation. Binding of VIP or PACAP to VPAC1 initiates two transduction pathways. The cAMP-dependent pathway leads to the phosphorylation/activation of CREB, which binds to the CRE site in the promoter and competes with p65 for limiting amounts of CBP. The cAMP-dependent pathway also inhibits MEKK1 activity, resulting in the inhibition of both c-Jun and TBP phosphorylation/activation. On the other hand, the cAMP-independent pathway in-hibits the IKK-mediated IB phosphorylation and subsequent NF-B nuclear translocation and DNA binding. A link between the two pathways could be established through the regulation of IKK activity by MEKK1 (49,55). As a result of the activation of the two pathways, VIP and PACAP block NF-B transcriptional activity.
Since NF-B positively regulates the transcription of different monocyte/macrophage-derived proinflammatory genes, which are commonly associated with inflammatory and autoimmune disorders, the inhibition of the NF-B transcriptional activity by VIP and PACAP could have significant therapeutic potential. The fact that VIP and PACAP regulate NF-B activation at multiple levels, and through different transduction pathways, could offer a significant advantage over other antiinflammatory agents.