Caffeic Acid Phenethyl Ester Inhibits T-Cell Activation by Targeting Both Nuclear Factor of Activated T-Cells and NF-κB Transcription Factors

Caffeic acid phenethyl ester (CAPE) is an active phenolic compound present in propolis, which is the generic name of a resinous product derived from the bark of conifer trees and carried by honeybees to their hives. The biological activities of propolis and CAPE have been extensively studied, and it has been shown that CAPE has antitumoral (Chiao et al., 1995; Huang et al., 1996) and anti-inflammatory properties (Mirzoeva and Calder, 1996; Michaluart et al., 1999; Fitzpatrick et al., 2001). However, CAPE inhibits the transcriptional activity of the COX-2 gene in epithelial cells (Michaluart et al., 1999), inducible nitric-oxide synthase gene expression, and nitric oxide production in macrophage cell lines (Song et al., 2002; Nagaoka et al., 2003). In addition, CAPE may suppress the eicosanoid synthesis (Mirzoeva and Calder, 1996) and the release of arachidonic acid from cell membranes (Michaluart et al., 1999). It has been recently shown that this phenolic compound is a potent inhibitor of mitogen-induced T-cell proliferation, lymphokine production (Ansorge et al., 2003), and NF-κB activation as well (Natarajan et al., 1996). However, the detailed inhibitory mechanisms at the molecular level of CAPE on T-cell activation are still unknown.

The signal transduction pathways triggered by the activation of the TCR/CD3 complex in T-cells lead to the immediate activation of transcription factors that regulate a variety of activation-associated genes. Many of them are cytokines and surface receptors that play an important role in coordinating the immune response (Crabtree and Clipstone, 1994). The signal transduction pathways involved in T-cell activation are initiated by the activation of phospholipase C-γ by specific tyrosine kinases at the lipid rafts resulting in the hydrolysis of the phosphatidylinositol 4,5 bisphosphate and the generation of inositol (1,4,5) triphosphate and diacylglycerol. Although inositol (1,4,5) triphosphate mobilizes Ca²⁺ from intracellular stores, diacylglycerol mediates activation of protein kinase C family members (Baier, 2003). As a consequence of an increase of intracellular Ca²⁺ levels, several signaling pathways are activated (Lewis, 2001). For instance, calcineurin, a Ca²⁺-calmodulin dependent protein phosphatase, is activated and subsequently dephosphorylates the nuclear factor of activated T-cells (NFAT), allowing its nuclear shuttling (Rao et al., 1997). This transcription factor was first described as an inducible regulatory complex critical for transcriptional induction of IL-2 gene in activated T-cells (Shaw et al., 1988) but was subsequently shown to regulate the transcription of many other genes, including cytokines (IL-4, γ-interferon, tumor necrosis factor-α, and granulocyte/macrophage-colony stimulating factor), cell surface receptors such as FasL and CD40L (Kiani et al., 2000), and regulatory enzymes such as COX-2 (de Gregorio et al., 2001). In the nucleus, NFAT binds to the DNA either alone or in conjunction with AP-1 proteins (Macian et al., 2001). Nevertheless, the coordinate induction and activation of the transcription factors NFAT, NF-κB, and AP-1 is required to regulate cytokine gene expression (Crabtree and Clipstone, 1994).

Stimulation via TCR-CD3 complex alone is sufficient for NFAT activation (Weiss and Littman, 1994), but in the case of NF-κB and AP-1 activation, a costimulatory signal, provided by CD28 receptors, is also required in antigen-stimulated T-cells (Edmead et al., 1996). The transcription factor NF-κB is one of the key regulators of genes involved in the immune/inflammatory response as well as in survival from apoptosis. NF-κB is an inducible transcription factor made up of homo- and heterodimers of p50, p65 (RelA), p52, RelB, and c-rel subunits that interacts with a family of inhibitory IκB proteins, of which IκBα is the best characterized (Karin and Ben-Neriah, 2000). In most cell types, these proteins sequester NF-κB in the cytoplasm by masking its nuclear localization sequence. Antigen stimulation in T-cells triggers a signaling pathway that results in the phosphorylation, ubiquitination, and subsequent degradation of IκB proteins, resulting in the translocation of NF-κB from the cytoplasm to the nucleus (Karin and Ben-Neriah, 2000). The prototypical, inducible NF-κB complex is a heterodimer containing p50 and p65, and in addition to the control of NF-κB activity exerted at the nuclear translocation level, there is increasing evidence for another complex level of regulation that is mediated by post-translational modifications of both subunits (Garcia-Piñeres et al., 2001; Nishi et al., 2002; Vermeulen et al., 2002).

In this paper, we studied the effect of CAPE on early and late T-cell activation events, and we demonstrate that CAPE inhibits antigen-induced proliferation and IL-2 production in human peripheral T-cells. Moreover, we show here, for the first time, that in addition to NF-κB, CAPE also targets the NFAT signaling pathway that is known to play a critical role in the immune response.

Materials and Methods

Cell Lines and Reagents. Jurkat cells (American Type Culture Collection, Manassas, VA) were maintained in exponential growth in RPMI 1640 medium supplemented with 10% heat-inactivated fetal calf serum, 2 mM l-glutamine, 1 mM Hepes, and the antibiotics penicillin (100 U/ml) and streptomycin (100 μg/ml) (Invitrogen, Barcelona, Spain). The Hela-Tet-On-Luc cell line was constructed by stable cotransfection of HeLa cells (American Type Culture Collection) following Tet-On system kit instructions (BD Biosciences Clontech, Basingstoke, UK). Briefly, this cell line contains two plasmids: the pTET-ON codifies constitutively for rtTA protein that, in response to doxycycline, gets active and binds to the pTRE2hyg-Luc starting the synthesis of luciferase gene. HeLa-Tet-On-Luc cell line was maintained in Dulbecco's modified Eagle's medium (Invitrogen) in the presence of 100 μg/ml of Hygromycin (Invitrogen) and 100 μg/ml of G418. Cells were maintained in a humid chamber at 37°C under 5% CO₂. The mAb anti-CD25 antibody (clone ACT-1, PE-labeled) was from DAKO (Glostrap, Denmark) and the mAbs anti-CD69 (clone FN50, fluorescein isothiocyanate-labeled) and anti-ICAM-1 (clone HA58, PE-labeled) were from BD Biosciences PharMingen (San Diego, CA). Dual-color reagent mouse IgG1/fluorescein isothiocyanate + mouse IgG1/PE from DAKO (clone DAK-GO1 directed toward Aspergillus niger glucose oxidase) was used as negative control. The anti-IκBα mAb was a gift from R. T. Hay (St. Andrews, Scotland), and the rabbit polyclonal anti-NFAT1 was a gift from J. M. Redondo (CBM, Madrid, Spain). [γ-³²P]ATP (3.000 μCi/mmol) was purchased from ICN Pharmaceuticals (Costa Mesa, CA). All other reagents were from Sigma-Aldrich (Barcelona, Spain).

Plasmids. The NFAT-Luc plasmid contains three copies of the NFAT binding site of the IL-2 promoter fused to the luciferase gene (Durand et al., 1988). The KBF-Luc contains three copies of the major histocompatibility complex enhancer κB site upstream of the conalbumin promoter followed by the luciferase gene (Yano et al., 1987). The IL-2-Luc (–326 to +45 of the IL-2 promoter) plasmid was previously described (Durand et al., 1988). The plasmids pRSVGal4-DBD, pGal4-NFAT1 (1–415), and pGal4-NFAT1 (1–171) have already been described (Luo et al., 1996). The Gal4-p65 plasmid containing the C-terminal region of the human p65 (amino acids 286–551) fused to the Gal4 binding domain (Schmitz et al., 1995) was obtained from M. L. Schmitz (University of Bern, Switzerland). The Gal4-Luc reporter plasmid includes five Gal4 DNA-binding sites fused to the luciferase gene (Schmitz et al., 1995).

Isolation of Human Peripheral Mononuclear Cells and T-Cell Proliferation Assays. Human peripheral blood mononuclear cells (PBMC) from healthy adult volunteer donors were isolated by centrifugation of venous blood on Ficoll-Hypaque density gradients (Amersham Biosciences Inc., Piscataway, NJ). Cells (10⁵) were cultured in triplicate in 96-well round-bottom microtiter plates (NUNC A/S, Roskilde, Denmark) in 200 μl of complete medium and stimulated with staphylococcal enterotoxin B (SEB) (1 μg/ml) or PHA (1 μg/ml) in the presence or absence of increasing concentrations of CAPE. A SEB-activation model was used, since it mimics T-cell activation induced by TCR and costimulators. The cultures were carried out for 3 days and pulsed with 0.5 μCi [³H]TdR/well (ICN Pharmaceuticals) for the last 12 h of culture. Radioactivity incorporated into DNA was measured by liquid scintillation counting.

Measurement of IL-2 Synthesis. PBMC (10⁶/ml) were preincubated with CAPE for 30 min in complete medium. Thereafter, cells were treated with SEB (1 μg/ml) for 18 h. After culture, supernatants were harvested and centrifuged for 10 min at 10,000g, and the levels of IL-2 in the supernatant were measured by enzyme-linked immunosorbent assay (R&D Systems, Wiesbaden-Norderstedt, Germany) according to the instructions of the manufacturer. Experiments were carried out in triplicate.

Cytofluorimetric Analyses of Cell Surface Antigen and Cell Cycle. For cell cycle analyses and measurement of CD25, CD69, and ICAM-1 expression, PBMC (10⁶/ml) were stimulated with SEB (1 μg/ml) in 24-well plates in a total volume of 2 ml of complete medium for 48 h in the presence or absence of CAPE (10 μM). Cell surface expression of CD25, CD69, and ICAM-1 antigens was measured by direct fluorescence using specific mAbs and analyzed by flow cytometry in an EPIC XL flow cytometer (Beckman Coulter, Fullerton, CA). For DNA profile analyses, cells were washed in PBS, fixed in ethanol (70% for 24 h at 4°C) followed by RNA digestion (Rnase-A; 50 U/ml) and propidium iodide (20 μg/ml) staining, and analyzed by cytofluorimetry. Ten thousand gated events were collected per sample, and the percentage of cells in every phase of the cell cycle was determined. The frequency of cells having undergone chromatinolysis was calculated by determining the sub G₀/G₁ fraction.

Isolation of Nuclear Extracts and Mobility Shift Assays. Jurkat cells (10⁶/ml) were treated with the agonists in complete medium as indicated. Cells were then washed twice with cold PBS, and proteins from total cell extracts (for NF-κB, AP-1, and Sp-1 binding) or nuclear extracts (for NFAT binding) were isolated as previously described (Sancho et al., 2003). Protein concentration was determined by the Bradford method (Bio-Rad, Hercules, CA). For the electrophoretic mobility shift assay (EMSA), double-stranded oligonucleotides containing the consensus sites for NF-κB, AP-1, Sp-1 (Promega, Madison, WI) and NFAT 5′-GATCGGAGGAAAAACTGTTTCATACAGA AGGCGT-3′ (distal NFAT site of human IL-2 promoter) were endlabeled with [γ-³²P]ATP. The binding reaction mixture contained 3 μg of nuclear extract (or 15 μg of total extracts), 0.5 μg poly(dI-dC) (Amersham Biosciences Inc., Piscataway, NJ), 20 mM Hepes (pH 7), 70 mM NaCl, 2 mM DTT, 0.01% NP40, 100 μg/ml bovine serum albumin, 4% Ficoll, and 100,000 cpm of endlabeled DNA fragments in a total volume of 20 μl. When indicated, 0.5 μl of rabbit anti-NFAT1 or preimmune serum was added to the standard reaction before the addition of the radiolabeled probe. For cold competition, a 100-fold excess of the double-stranded oligonucleotide competitor was added to the binding reaction. After 30 min incubation at 4°C (room temperature in the case of NFAT), the mixture was electrophoresed through a native 6% polyacrylamide (4% in the case of NFAT) gel containing 89 mM Tris-borate, 89 mM boric acid, and 1 mM EDTA. Gels were pre-electrophoresed for 30 min at 225 V and then for 2 h after loading the samples. These gels were dried and exposed to X-ray film at –80°C.

Transient Transfections and Luciferase Activity. Jurkat cells (10⁶) were transfected with the indicated plasmids in Opti-MEM (Invitrogen) by using Lipofectin Reagent (Invitrogen) following the instructions of the manufacturer. Twenty-four hours after transfection, cells were preincubated with different concentrations of CAPE for half an hour and then stimulated as indicated for 6 h. Then, the cells were lysed in 25 mM Tris-phosphate (pH 7.8), 8 mM MgCl₂, 1 mM DTT, 1% Triton X-100, and 7% glycerol. Luciferase activity was measured using an Autolumat LB 9501 (Berthold Technologies, Bad Wildbad, Germany) following the instructions of the luciferase assay kit (Promega), and protein concentration was measured by the Bradford method. The background obtained with the lysis buffer was subtracted in each experimental value, and the specific transactivation was expressed as total RLU induction. All of the experiments were repeated at least three times.

Western Blots. Jurkat cells (10⁶ cells/ml) were stimulated as indicated in the presence or absence of CAPE for the indicated periods of time. Cells were then washed with PBS and resuspended in lysis buffer [20 mM Hepes (pH 8.0), 0.35 M NaCl, 0.1 mM EGTA, 0.5 mM EDTA, 1 mM MgCl₂, 20% Glycerol, 1 mM DTT, 1 μg/ml leupeptin, 0.5 μg/ml pepstatin, 0.5 μg/ml apronitin, and 1 mM phenylmethylsulfonyl fluoride] containing 0.5% NP40. Cells were incubated for 15 min in ice, and cellular proteins were obtained by centrifugation. Protein concentration was determined by Bradford assay (Bio-Rad), and 30 μg of proteins were boiled in Laemmli buffer and electrophoresed in 10% sodium dodecyl sulfate-polyacrylamide gel (IκBα) or in 6% sodium dodecyl sulfate-polyacrylamide gel (NFAT). Separated proteins were transferred to nitrocellulose membranes (0.5 A at 100 V; 4°C) for 1 h. Blots were blocked in Tris-buffered saline solution containing 0.1% Tween 20 and 5% nonfat dry milk overnight at 4°C, and immunodetection of specific proteins was carried out with primary antibodies using an enhanced chemiluminescence system (Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK).

Results

CAPE Inhibits Antigen-Specific T-Cell Proliferation and Cell Cycle Progression. It has been recently shown that CAPE inhibits mitogen-induced T-cell proliferation (Ansorge et al., 2003). Thus, we studied the effects of CAPE on several T-cell activation events induced by stimulation with the SEB. This enterotoxin is a T-cell superantigen that mimics T-cell activation induced by TCR and costimulators (Marrack and Kappler, 1990). In Fig. 1A, it is shown that DNA synthesis measured by [³H]TdR uptake in both PHA and SEB-stimulated T-cells was markedly inhibited in a concentration-dependent manner by CAPE. As a consequence of cell activation, primary T-cells progress to the S-phase and G₂/M phases of the cell cycle, and in Fig. 1B, it is shown that unstimulated PBMC remained largely in the G₀/G₁ phase of the cell cycle. Three days following activation with SEB, T-cells were full cycling and progressed through the S, G₂, and M phases of the cell cycle (37% of the cells), whereas pretreatment with CAPE (10 μM) almost completely prevented the entry of the cells in the S-phase of the cell cycle (Fig. 1B). Interestingly, no significant differences were found in the percentage of hypodiploid cells (sub G₀/G₁) between SEB and SEB plus CAPE-stimulated cells. These results indicate that, at the doses used, CAPE did not induce cytotoxicity or apoptosis in primary T-cells. T-cell activation involves the induction of several surface molecules, such as CD69, CD25, or ICAM-1, whose gene transcriptional regulation is highly dependent on NF-κB. Thus, the effect of CAPE on the cell surface expression of these activation markers was studied in SEB-stimulated primary T-cells. Figure 2 demonstrated that CAPE at 10 μM greatly inhibited not only the percentage of cells expressing the CD25, CD69, and ICAM-1 markers at the cell surface but also the relative intensity of fluorescence in the positive cells (relative intensity of fluorescence measured in gated positive cells).

Effects of CAPE on IL-2 Synthesis and Promoter Activity. IL-2 represents one of the major growth factors for the clonal expansion of activated T-cells. Thus, we studied the effects of CAPE on SEB-induced IL-2 production in primary T-cells, and in Fig. 3, it is shown that this compound was able to inhibit the release of IL-2 in a concentration-dependent manner (with an IC₅₀ of approximately 1 μM). Since IL-2 gene expression is regulated mainly at the transcriptional level, we investigated the regulation of IL-2 promoter in Jurkat cells transiently transfected with the luciferase reporter plasmid IL-2-Luc. After transfection, cells were preincubated with CAPE for 30 min, activated with PMA (20 ng/ml) plus ionomycin (1 μM) for 6 h, and tested for luciferase activity. CAPE efficiently inhibited PMA plus ionomycin-induced luciferase expression driven by the IL-2 promoter in a dose-dependent manner (Fig. 4A). The inhibitory effects of CAPE were not due to an interference with the transcriptional machinery or with the in vitro activity of the luciferase enzyme, since the inducible expression of luciferase mediated by doxycycline in Hela-Tet-On-Luc cells was not affected by CAPE at any of the concentrations tested (Fig. 4B).

CAPE Inhibits NF-κB Transcriptional Activity. The transcriptional activity of many lymphokines, including IL-2, depends on the coordinated activation of several transcription factors, such as NFAT, NF-κB, and AP-1. First, we evaluated the effects of CAPE on the DNA binding of NF-κB, AP-1, and Sp-1 transcription factors, and we found that NF-κB was the only transcription factor inhibited by CAPE in a concentration-dependent manner (Fig. 5). The DNA-binding specificity was studied by supershift experiments with specific anti-p50 and anti-p65 (RelA) antibodies and by cold competition experiments with unlabeled competitors, and the heterodimer p50/p65 was identified as the main complex (data not shown; Sancho et al., 2003). Next, the effect of CAPE on the NF-κB transcriptional activity was evaluated by using the luciferase reporter construct KBF-Luc (Yano et al., 1987). Activation by PMA clearly increased (37-fold inductions) the luciferase gene expression driven by the NF-κB dependent promoter in Jurkat cells, and we found that, once more, CAPE effectively inhibited this activity in a dose-dependent manner (Fig. 6A). The inhibitory effect of CAPE on NF-κB inhibition was reversible, and CAPE removed from the cells after 2 h incubation did not affect the NF-κB-dependent luciferase activity in PMA-treated cells (data not shown). To investigate the level at which CAPE exerted its inhibitory effect on NF-κB activation, we stimulated Jurkat cells with PMA for different times in the presence or absence of CAPE (10 μM), and proteins from total cell extracts were analyzed for NF-κB-DNA binding activity by EMSA and for studying the steady-state levels of IκBα by Western blot. The kinetic experiments revealed a clear increase in NF-κB–DNA binding after 15 min of stimulation that slightly decreased through the time of stimulation; the increased DNA binding paralleled with degradation of IκBα that was more evident after 15 min of PMA stimulation and recovered after 60 min of stimulation. In Fig. 6B, it is shown that CAPE was able to prevent the NF-κB binding to DNA in PMA-stimulated Jurkat cells. However, under the same conditions, CAPE did not prevent IκBα degradation; even the recovery of IκBα protein to the basal levels, which also depend on NF-κB activation (Chiao et al., 1994), was delayed in the presence of CAPE, which did not affect the steady state levels of α-tubulin. These results are fully coincidental with those reported by Natarajan et al. (1996) in U937 cells and indicate that the NF-κB inhibitory effects of CAPE take place at a level downstream of IκBα degradation. Since the NF-κB heterodimer identified in PMA-stimulated Jurkat cells is composed of p50 and p65 subunits, and taking into account that p50 serves mainly as a DNA-binding subunit and p65 is the transcriptional active member of the complex (Karin and Ben-Neriah, 2000), we further analyzed whether CAPE directly inhibits p65-transcriptional activity. Thus, we performed cotransfection experiments using Gal4-p65, a fusion protein containing the transactivation domain of p65 (amino acids 286–551) and the DNA binding domain of the yeast Gal4 transactivator together with a reporter plasmid in which the luciferase gene is under the control of a Gal4-responsive element (Gal4-Luc). This system has the advantage that the Gal4 transactivator fusion protein is exclusively nuclear and regulated independently of IκBs, thus, it can be used to study the basal levels of p65 transcriptional activity (Schmitz et al., 1995). The results presented in Fig. 6C revealed that transcriptional activity of Gal4-p65 was inhibited by the presence of CAPE in a concentration-dependent manner. However the inhibitory activity of CAPE was less evident using this heterologous Gal4-p65 system when compared with the endogenous NF-κB (Fig. 6A). Since the Gal4-p65 fusion protein only contains amino acids 286 to 551, it is likely that CAPE exerts its effect acting not only at the transactivation domain but also in other p65 domains.

CAPE Inhibits NFAT Dephosphorylation and Transcriptional Activity. Transcriptional activation of NFAT requires its translocation to the nucleus, where it binds to specific consensus sites in the promoter region of IL-2 gene (Maggirwar et al., 1997). TCR signaling that activates NFAT can be mimicked by a combination of PMA plus the calcium ionophore, ionomycin. To study whether CAPE inhibits NFAT activation, we first performed electrophoretic mobility shift assays with nuclear extracts of Jurkat cells stimulated with PMA plus ionomycin in the presence or absence of increasing concentrations of CAPE. Using the distal NFAT site of the IL-2 promoter, we found a major complex that was retarded in PMA plus ionomycin-treated cells, and the binding to DNA of this complex was clearly inhibited in the presence of increasing concentrations of CAPE. This complex was characterized as NFAT1 by supershift experiments with an anti-NFAT1 antiserum and by cold competition experiments (Fig. 7A). To dissect the mechanism responsible for NFAT inhibition by CAPE, we studied the dephosphorylation of NFAT1 by Western blot using a specific antiserum against NFAT1. In Fig. 7B, it is shown that, upon PMA plus ionomycin treatment, NFAT1 was dephosphorylated in Jurkat cells and CAPE inhibited this dephosphorylation in a concentration-dependent manner. To further demonstrate the inhibitory effects of this plant-derived phenolic compound in NFAT transactivation activity, Jurkat cells were transfected with a luciferase reporter construct under the control of minimal promoter containing three NFAT binding sites. Activation by PMA plus ionomycin increased the luciferase gene expression driven by this promoter in Jurkat cells, and we found that CAPE effectively inhibited the inducible transcriptional activity of the NFAT promoter in a dose-dependent manner (Fig. 8A). Next, to explore the inhibitory mechanisms of CAPE on NFAT activation, a Gal4-derived reporter system was employed. Jurkat cells were cotransfected with the chimeric vector pGal4-NFAT1 (1–415), encoding the Gal4 DBD fused to amino acids 1 through 415 of human NFAT1, or the parental vector pRSV-Gal4-DBD along with the reporter plasmid Gal4-Luc. The fusion protein pGal4-NFAT1 (1–415) contains both the calcineurin-binding regulatory and transactivation domains. As shown in Fig. 8B, CAPE prevented the transactivation function of NFAT1 induced by PMA plus ionophore in a concentration-dependent manner. Strikingly, CAPE did not affect the transcriptional activity of the construct pGal4-NFAT1 (1–171) that contains only the transactivation domain of NFAT and is not regulated by the calcium-dependent calcineurin pathway that dephosphorylates NFAT1 (Luo et al., 1996). Altogether, our results strongly suggest that CAPE inhibits NFAT by targeting a component of the signaling pathways leading to NFAT dephosphorylation and not by interfering with the NFAT binding to DNA.

Discussion

Since activated T-cells play a crucial role in the onset of several inflammatory diseases, the inhibition of transcription factors NF-κB and NFAT represents a rationale for the development of novel and safe anti-inflammatory agents. CAPE has been shown to be a pharmacologically safe compound with known anti-inflammatory, immunomodulatory, and anticarcinogenic properties (Huang et al., 1996; Michaluart et al., 1999; Ansorge et al., 2003), and we show here that CAPE is an effective in vitro inhibitor of both NFAT and NF-κB transcription factors. As a consequence, CAPE inhibits IL-2 gene transcription, IL-2R (CD25) expression, and proliferation in antigen-stimulated human T-cells.

Increasing evidence is accumulating to demonstrate that post-translational modifications of the NF-κB subunits are crucial for its transcriptional activity. Thus, modifications of critical sulfhydryl groups in p50 (Nishi et al., 2002) and in p65 (Garcia-Piñeres et al., 2001) may account for the NF-κB inhibitory activity of CAPE. It has been previously shown that curcumin and other compounds that are structurally similar to CAPE may covalently modify sulfhydryl groups by oxidation and alkylation reactions, thus affecting selectively cysteine residues of targeted proteins that control the transcription of inducible genes (Dinkova-Kostova et al., 2001). This biological activity of CAPE could explain the finding that this phenolic compound prevented the binding to DNA of the p50/p65 NF-κB in vitro and in vivo ((Natarajan et al., 1996 and our results). However, CAPE also inhibits the transcriptional activity of the chimeric protein Gal4-p65 (amino acids 286–551) that contains the transactivation domain (Schmitz et al., 1995). Since this Gal4-p65 fusion protein is not able to interact with endogenous nuclear p50 and does not contain cysteine residues susceptible of modification, it is possible that other yet-to-be-identified mechanisms may account for the inhibitory effect of CAPE on NF-κB. It has been recently shown that IKKα translocates to the nucleus, where it phosphorylates histone H3, which is required for optimal NF-κB-dependent gene transcription (Yamamoto et al., 2003). Whether or not CAPE could interfere with this pathway at the nuclear level is an interesting possibility that warrants further research.

We show here, for the first time, that CAPE is also a potent inhibitor of the NFAT pathway, and our results suggest that the calcineurin phosphatase can represent one of the major targets for CAPE, since this compound inhibits NFAT dephosphorylation and nuclear binding to DNA. Moreover, CAPE did not affect the transcriptional activity of the fusion protein Gal4-NFAT (1–171), which does not require calcineurin activation to induce NFAT-dependent transcription. The catalytic subunit of calcineurin (calcineurin A) contains a phosphoesterase motif, which accommodates an active dinuclear metal center (Goldberg et al., 1995). Interestingly, in cell lysates from Jurkat cells, the calcineurin activity can be affected by both oxidation and reduction processes (Reiter et al., 1999; Sommer et al., 2000). Then, it is possible that phenolic compounds like CAPE and curcumin could interact with specific cysteine residues that seem to be critical for the calcium-dependent changes of the structural conformation of this phosphatase, impairing its activity (Tan et al., 1996).

The inhibitory effects of CAPE on other genes regulated by both NF-κB and NFAT transcription factors may be extended to nonlymphoid cells, such as human endothelial cells. Thus, angiogenesis, the formation of new blood vessels from the existing vasculature, occurs in many pathologies, including rheumatoid arthritis, atherosclerosis, and tumor growth (Sullivan and Bicknell, 2003). It has been reported that COX-2 plays an important role in such conditions, as overexpression of COX-2 in transformed cells (Michaluart et al., 1999) and in the joints of rheumatoid arthritis patients has been described (Woods et al., 2003). Angiogenesis regulation involves a complex signaling network in which inducible expression of COX-2 and subsequent prostaglandins synthesis plays a central role (Woods et al., 2003). It has been described that both NF-κB and NFAT participate in COX-2 gene transcriptional regulation, the calcium/calcineurin pathway being essential for COX-2 transcription not only in endothelial (Hernandez et al., 2001) but also in T-cells (Iñiguez et al., 2000). Accordingly, COX-2 gene transcription has been reported to be inhibited by CAPE through an NF-κB-independent pathway in oral epithelial cells (Michaluart et al., 1999), suggesting that additional elements at the COX-2 promoter are targeted by CAPE. A detailed analysis of COX-2 promoter in T-cells has identified two NFAT binding sites within this region, which are essential for COX-2 expression (Iñiguez et al., 2000). Since CAPE has been described as a potent NF-κB inhibitor, and we have reported here that NFAT is also inhibited by CAPE, it is likely that COX-2 gene inhibition mediated by CAPE is the consequence of the combined inhibitory effects of both NF-κB and NFAT transcription factors. In this sense, this COX-2 inhibitory activity of CAPE could explain the potential use of this natural compound in the treatment of disease conditions such as tumor growth, rheumatoid arthritis, and atherosclerosis, in which angiogenesis plays a key role.