Prostaglandin E2 Up-regulates HIV-1 Long Terminal Repeat-driven Gene Activity in T Cells via NF-κB-dependent and -Independent Signaling Pathways*

Replication of human immunodeficiency virus type-1 (HIV-1) is highly dependent on the state of activation of the infected cells and is modulated by interactions between viral and host cellular factors. Prostaglandin E2(PGE2), a pleiotropic immunomodulatory molecule, is observed at elevated levels during HIV-1 infection as well as during the course of other pathogenic infections. In 1G5, a Jurkat-derived T cell line stably transfected with a luciferase gene driven by HIV-1 long terminal repeat (LTR), we found that PGE2 markedly enhanced HIV-1 LTR-mediated reporter gene activity. Experiments have been conducted to identify second messengers involved in this PGE2-dependent up-regulating effect on the regulatory element of HIV-1. In this study, we present evidence indicating that signal transduction pathways induced by PGE2 necessitate the participation of cyclic AMP, protein kinase A, and Ca2+. Experiments conducted with different HIV-1 LTR-based vectors suggested that PGE2-mediated activation effect on HIV-1 transcription was transduced via both NF-κB-dependent and -independent signaling pathways. The involvement of NF-κB in the PGE2-dependent activating effect on HIV-1 transcription was further confirmed using a κB-regulated luciferase encoding vector and by electrophoretic mobility shift assays. Results from Northern blot and flow cytometric analyses, as well as the use of a selective antagonist indicated that PGE2 modulation of HIV-1 LTR-driven reporter gene activity in studied T lymphoid cells is transduced via the EP4receptor subtype. These results suggest that secretion of PGE2 by macrophages in response to infection or inflammatory activators could induce signaling events resulting in activation of proviral DNA present into T cells latently infected with HIV-1.


Replication of human immunodeficiency virus type-1 (HIV-1) is highly dependent on the state of activation of the infected cells and is modulated by interactions between viral and host cellular factors. Prostaglandin E 2 (PGE 2 ), a pleiotropic immunomodulatory molecule, is observed at elevated levels during HIV-1 infection as well as during the course of other pathogenic infections. In 1G5, a Jurkat-derived T cell line stably transfected with a luciferase gene driven by HIV-1 long terminal repeat (LTR), we found that PGE 2 markedly enhanced HIV-1 LTR-mediated reporter gene activity. Experiments have been conducted to identify second messengers involved in this PGE 2 -dependent up-regulating effect on the regulatory element of HIV-1. In this study, we present evidence indicating that signal transduction pathways induced by PGE 2 necessitate the participation of cyclic AMP, protein kinase A, and Ca 2؉ . Experiments conducted with different HIV-1 LTR-based vectors suggested that PGE 2 -mediated activation effect on HIV-1 transcription was transduced via both NF-B-dependent and -independent signaling pathways. The involvement of NF-B in the PGE 2 -dependent activating effect on HIV-1 transcription was further confirmed using a B-regulated luciferase encoding vector and by electrophoretic mobility shift assays. Results from Northern blot and flow cytometric analyses, as well as the use of a selective antagonist indicated that PGE 2 modulation of HIV-1 LTR-driven reporter gene activity in studied T lymphoid cells is transduced via the EP 4 receptor subtype. These results suggest that secretion of PGE 2 by macrophages in response to infection or inflammatory activators could induce signaling events resulting in activation of proviral DNA present into T cells latently infected with HIV-1.
Infection with human immunodeficiency virus type-1 (HIV-1), 1 the etiologic agent of AIDS (1), leads to a progressive decline of CD4-expressing T cells resulting in impaired cellular immune functions. This infection is influenced by a complex interplay between viral and host factors, as well as by microbial agents termed cofactors. It has been postulated that such cofactors may be important in disease progression by enhancing cell-to-cell transmission or through up-regulation of HIV-1 expression in latently infected cells (2). Mycoplasma (3), mycobacteria (4), viruses (5,6), and the protozoan parasite Leishmania (7,8) may act as cofactors for the pathogenesis of HIV-1 infection either by directly modulating virus replication or by inducing a more profound immunosuppressive state. During coinfections, the inability of the host to develop an effective immune response may involve the participation of the immunosuppressive molecule PGE 2 , an oxygenated polyunsaturated fatty acid that contain a cyclopentane ring structure. PGE 2 are molecules that have been shown to modulate the immune response both in vitro and in vivo (9,10). Macrophages, follicular dendritic cells, fibroblasts, and vascular endothelial cells synthesize PGE 2 , while lymphocytes do not secrete this major product of arachidonic acid metabolism (11)(12)(13)(14). A marked increase in PGE 2 production is generated in response to a variety of immunological stimuli including interleukin (IL)-1, tumor necrosis factor-␣ (TNF-␣), antigen-antibody complexes, and lipopolysaccharide (15). Moreover, production of PGE 2 has been shown to be induced by infection with several pathogens such as Leishmania donovani (16), Leishmania major (17), Entamoeba histolytica (18,19), Pseudomonas aeruginosa (20,21), Staphylococcus epidermidis (22), Mycobacterium avium (23), herpes simplex virus type 1 (24), coxsackie virus (25), respiratory syncytial virus (26), and HIV-1 (27)(28)(29)(30). PGE 2 has been implicated in decreasing T-cell proliferation, IL-2 production, and IL-2 receptor expression (15,(31)(32)(33)(34)(35). PGE 2 shifts the balance of the cellular immune response away from T-helper type 1 (Th1) favoring a Th2 response which drives humoral responses toward the production of IgE (36). However, more recent findings have depicted PGE 2 as a pleiotropic molecule that can act both negatively or positively on the immune system (15).
An overproduction of PGE 2 (as high as 10 Ϫ4 M) is seen in a number of disorders (e.g. allergy, hyper-IgE syndrome, Hodgkin lymphoma, trauma, sepsis, and transplantation), most of which are characterized by elevated Th2 and IgE re-sponses (15, 36 -38). As specified above, elevated levels of PGE 2 have also been reported in individuals infected with HIV-1 (27)(28)(29)(30) and it has been postulated that this may contribute to the immunosuppressive state seen in such virally-infected patients (39). The mechanism(s) responsible for the enhanced prostaglandin formation is still undefined. The initial contact between the virus particle and its target cell might represent the crucial step leading to the production of PGE 2 by macrophages. This concept is supported by the finding that a significant production of endogenous PGE 2 is induced (20-to 40-fold increase) following incubation of primary human monocytes with the HIV-1 external envelope glycoprotein gp120 (40). However, in sharp contrast with this report, a previous study has demonstrated that interaction between gp120 and THP-1, a human monocytoid cell line, does not increase exogenous production of PGE 2 (39). It is important to specify that, unlike monocyte/macrophages, promonocytoid THP-1 cells are not at a terminal stage of differentiation. In addition, a monomer form of gp120 was used in this study which might not parallel physiological conditions where gp120 is under a multimeric form (41). Depending on the cell type, binding of PGE 2 to one of its six different described receptors (EP 1 , EP 2 , EP 3I , EP 3II , EP 3III , and EP 4 ) can lead to activation of phospholipase C, phosphatidylinositol turnover, activation of adenylate cyclase via cholera toxin-sensitive G ␣S proteins and mobilization of intracellular Ca 2ϩ (42). Given that HIV-1 is highly dependent on intracellular signaling machinery for its life cycle, it is therefore possible that interaction of PGE 2 with its surface receptor(s) can modulate virus replication. A previous cellular study supports this postulate since PGE 2 was found to enhance HIV-1 replication in acutely infected T lymphoid cells (43).
The primary goal of the present work was thus to investigate the putative modulatory role of PGE 2 on the regulatory elements of HIV-1 (LTR) at both biochemical and molecular levels. For this purpose, we treated human T lymphoid cells stably and transiently transfected with different HIV-1 LTR-driven luciferase reporter gene vectors with concentrations of PGE 2 known to be found under physiological conditions. We then explored the intracellular second messengers participating in PGE 2 -mediated signaling transduction pathway(s) and investigated DNA-binding transcriptional factor(s) and cell surface receptor(s) implicated in the PGE 2 -dependent effect on HIV-1 transcription.

EXPERIMENTAL PROCEDURES
Reagents-PGE 2 , phorbol 12-myristate 13-acetate (PMA), phytohemagglutinin (PHA), indomethacin and dibutyryl-cAMP were purchased from Sigma. H7 was purchased from Seikagaku America Inc. (Tampa, FL). BAPTA/AM, IBMX, MDL-12,330A, and HA-1004 were purchased from BioMol (Plymouth Meeting, PA). The calcium inhibitor CAI was a generous gift from Dr. E. C. Kohn (National Institutes of Health, Bethesda, MD). AH 23848B was kindly provided by Dr. S. G. Lister (Glaxo Wellcome, United Kingdom). Trizol came from Life Technologies, Inc. (Grand Island, NY). Stock solutions of PGE 2 (1 mM) were prepared by dissolving the lyophilized product into absolute ethanol and were stored at Ϫ20°C until needed.
Cells and Culture Conditions-The parental lymphoid T cell line, Jurkat E6.1, was obtained from the American Type Culture Collection (ATCC, Rockville, MD), while 1G5 was supplied by the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, National Institute of Health (Rockville, MD). 1G5 is a clonal cell line derived from Jurkat E6.1 cells which has been stably transfected with a luciferase gene driven by the HIV-1 LTR (44). The cells were grown in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum (Hyclone Laboratories, Logan, UT), 2 mM glutamine, 100 units/ml penicillin, 100 g/ml streptomycin, 0.22% NaHCO 3 , and were maintained at 37°C in a 5% CO 2 humidified atmosphere.
Plasmids and Antibodies-In our studies, we have used pLTR-LUC and pmBLTR-LUC that have been kindly provided by Dr. K. Calame (Columbia University, NY). These molecular constructs contain the luciferase reporter gene under the control of wild-type (GGGACTTTCC) or NF-B-mutated (CTCACTTTCC) HIV-1 HXB2 LTR (Ϫ453 to ϩ80) (45). We have also used the pB-TATA-LUC vector which contains the minimal HIV-1 B region and a TATA box placed upstream of the luciferase reporter gene (46). This plasmid is a generous gift from Dr. W. C. Greene (The J. Gladstone Institutes, San Francisco, CA). The molecular construct pNF-B-LUC contains five (5) consensus sequences of NF-B-binding sites placed in front of the luciferase reporter gene (Stratagene, La Jolla, CA). NFAT-LUC contains the IL-2 minimal promoter with three tandem copies of NFAT-binding site placed upstream of the luciferase reporter gene (kindly provided by Dr. G. R. Crabtree, Howard Hughes Medical Institute, CA). Northern blot analyses were performed using human EP 1 , EP 2 , EP 3 , and EP 4 subtypes cDNA fragments. The cDNAs for human prostanoid receptor EP 1 (1.3 kilobases), EP 3 (1.8 kilobases), and EP 4 (1.5 kilobases) were kind gifts from Dr. M. Abramovitz (Merck Frost, Qué, Canada). The cDNA for hEP 2 (1.1 kilobases) was generously provided by Dr. K. M. Kedzie (Allergan, Irvine, CA). The polyclonal antibody specific for EP 4 was obtained from Cayman Chemical (Ann Arbor, MI). This rabbit serum is directed against a synthetic peptide from the human EP 4 receptor.
Modulation of HIV-1 LTR Activity by PGE 2 -In order to assess whether PGE 2 could modulate HIV-1 LTR activity, 1G5 cells (5 ϫ 10 5 ) were either left untreated or treated with PHA (3 g/ml), anti-CD3 antibody (clone OKT3 at 1 g/ml), PMA (20 ng/ml), and TNF-␣ (2 ng/ml: R&D systems, Minneapolis, MN) in a final volume of 200 l for 1 h at 37°C. Next, the cells were incubated in the absence or presence of 100 nM PGE 2 for 24 h at 37°C. Dose-response experiments were done using a similar number of cells, washed once in phosphate-buffered saline (PBS, pH 7.4), and resuspended in 1 ml of fresh complete culture medium before incubation for 24 h at 37°C with PGE 2 at final concentrations of 1, 10, 100, and 1000 nM. Kinetic experiments were done by incubating 1G5 cells with 100 nM PGE 2 for 2, 6, 8, and 24 h. In some experiments, 1G5 cells were pretreated for 1 h at 37°C with second messenger inhibitors such as H7, HA-1004, BAPTA/AM, indomethacin, MDL-12,330A, and IBMX at subcytotoxic and subcytostatic concentrations prior to treatment with 100 nM PGE 2 for 8 h at 37°C. The inhibitor CAI requires a pretreatment of at least 8 h for optimum inhibition and shows no acute interference with the growth properties of the cells (47). All experiments were performed three times and luciferase activity was evaluated for each quadruplicate samples by a modified version of a previously published procedure (48). Briefly, following the incubation period, 100 l of cell-free supernatant were withdrawn from each well and 25 l of cell culture lysis buffer (25 mM Tris phosphate, pH 7.8, 2 mM dithiothreitol, 1% Triton X-100, and 10% glycerol) were added before incubation at room temperature for 30 min. Transient Transfection by DEAE-dextran-Jurkat E6.1 (5 to 10 ϫ 10 6 ) were first washed once in a TS buffer (25 mM Tris-HCl, pH 7.4, 5 mM KCl, 0.6 mM NaHPO 4 , 0.5 mM MgCl 2 , and 0.7 mM CaCl 2 ) and resuspended in 0.5-1 ml of TS containing 10 -20 g of the indicated plasmids and 500 g/ml DEAE-dextran (final concentration). The cells/ TS/plasmid/DEAE-dextran mixture was incubated for 25 min at room temperature. Thereafter, cells were diluted at a concentration of 1 ϫ 10 6 per ml using complete culture medium supplemented with 100 M chloroquine (Sigma). After 45 min of incubation at 37°C, cells were centrifuged, washed once, resuspended in complete culture medium, and incubated at 37°C for 24 h. To minimize variations in plasmid transfection efficiencies, transfected cells were pooled 24 h after transfection and were next separated into various treatment groups as follow. Transiently transfected cells were seeded at a density of 10 5 cells per well (100 l) in 96-well flat-bottom plates. Cells were left untreated or were treated with TNF-␣ (2 ng/ml), PHA/PMA (3 g/ml and 20 ng/ml, respectively), and 100 nM PGE 2 in a final volume of 200 l for a period of 8 h at 37°C. Cells were then lysed and luciferase activity was assessed as described above.
Electrophoretic Mobility Shift Assay-Nuclear extracts were prepared as described previously (49). In brief, 1G5 cells (10 6 ) were either left untreated or were treated with PGE 2 (100 nM) or TNF-␣ (2 ng/ml) for 30 min at 37°C. The incubation of cells with the stimulating agents was terminated by the addition of ice-cold PBS and nuclear extracts were prepared according to the microscale preparation protocol (50). Sedimented cells were resuspended in 400 l of cold buffer A (10 nM HEPES, pH 7.9, 1.5 mM MgCl 2 , 10 nM KCl, 0.5 mM dithiothreitol, and 0.2 mM phenylmethylsulfonyl fluoride). After 10 min on ice, the lysate was vortexed for 10 s and the samples were centrifuged for 10 s at 12,000 ϫ g. The supernatant fraction was discarded and the pellet was resuspended in 100 l of cold buffer B (20 mM HEPES-KOH, pH 7.9, 25% glycerol, 420 mM NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5 mM dithiothreitol, and 0.2 mM phenylmethylsulfonyl fluoride) and incubated on ice for 20 min. Cellular debris were removed by centrifugation at 12,000 ϫ g for 2 min at 4°C and the supernatant fractions were stored at Ϫ70°C until used. Ten micrograms of nuclear extracts were used to perform electrophoretic mobility shift assay. Protein content was determined by the commercial BCA Protein Assay Reagent (Pierce, Rockfold, IL). Nuclear extracts were incubated for 30 min at room temperature in 15 l of buffer of a binding solution (100 mM HEPES, pH 7.9, 40% glycerol, 10% Ficoll, 250 mM KCl, 10 mM dithiothreitol, 5 mM EDTA, 250 mM NaCl, 2 g of poly(dI-dC), 10 g of nuclease-free bovine serum albumin fraction V) containing 1 ng of 32 P-5Ј-end-labeled doublestranded (dsDNA) oligonucleotide. Double-stranded DNA (100 ng) was labeled with [␥-32 P]ATP and T4 polynucleotide kinase in a kinase buffer (New England Biolabs, Beverly, MA). This mixture was incubated for 30 min at 37°C and the reaction was stopped with 5 l of 0.2 M EDTA. The labeled oligonucleotide was extracted with phenol/chloroform and passed trough a G-50 spin column. The double-stranded DNA oligonucleotide, which was used as a probe, contained the consensus NF-Bbinding site corresponding to the sequence 5Ј-ATGTGAGGGGACTTTC-CCAGGC-3Ј and was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, Ca). DNA⅐NF-B complexes were resolved from free labeled DNA by electrophoresis in native 4% (w/v) polyacrylamide gel containing 50 mM Tris-HCl (pH 8.5), 200 nM glycine, and 1 mM EDTA. The gel was subsequently dried and autoradiographed.
Flow Cytometry-Cell surface expression of hEP 4 receptor was evaluated by flow cytometry as follow. Cell lines 1G5 and Jurkat E6.1 (5 ϫ 10 5 ) were washed once in PBS containing 2% fetal bovine serum (PBS pH 7.4 ϩ 2% fetal bovine serum (PBSA)). Cells were then resuspended in 100 l of PBSA to which was added 0.5 g of polyclonal rabbit anti-hEP 4 antibody, vortexed gently, and incubated for 30 min on ice. Cells were subsequently washed with PBSA and resuspended in 100 l of PBSA containing fluorescein isothiocyanate-labeled chicken antirabbit IgG antibody (0.5 g total) and further incubated for 30 min on ice. Cells were finally centrifuged and resuspended in 1% paraformaldehyde in PBS before being analyzed by flow cytometry (EPICS XL, Coulter Corp., Miami, FL).
Experiments with the EP 4 Receptor-specific Antagonist AH 23848B-Assays with AH 23848B were performed by incubating 1G5 cells with 30 M AH 23848B for 1 h at 37°C (53). After this pretreatment, cells were incubated with 100 nM PGE 2 for 8 h at 37°C. Experiments were performed three times and luciferase activity was monitored as described above.

Modulation of HIV-1 LTR-driven Gene Expression in T Cells
by PGE 2 -In order to assess whether PGE 2 could affect the regulatory elements of HIV-1, we initially set out a dose-response experiment which was performed with PGE 2 in a 24-h incubation period. The results showed a steady increase of HIV-1 LTR activity in 1G5 cells starting at as little as 1 nM concentration of PGE 2 (Fig. 1A). In these experiments, PGE 2 was reconstituted in absolute ethanol to get a stock solution of 1 mM which was serially diluted in complete culture medium to get our working dilutions (1, 10, 100, and 1000 nM). Incubation of 1G5 cells with the concentration of ethanol corresponding to the one used with 1000 nM PGE 2 resulted in a 1.5-fold increase of HIV-1 LTR dependent activity, while no effect was seen with the equivalent ethanol concentration for 100 nM PGE 2 (data not shown). Therefore, subsequent experiments were carried out using 100 nM PGE 2 to avoid any putative influence of ethanol on PGE 2 -mediated enhancement of HIV-1 LTR activity. Kinetics analyses were next performed to determine the optimal incubation time for this PGE 2 -mediated HIV-1 LTR-driven activation. The maximal positive effect of PGE 2 was seen 8 h after the initiation of treatment (fold enhancement of 7.4) (Fig. 1B). Although this type of kinetic might be reminiscent of degradation of PGE 2 , it is unlikely since similar kinetic of time-dependent HIV-1 LTR-driven luciferase activity were measured with activators such as TNF-␣, PHA, and PMA (data not shown). Moreover, preincubation of PGE 2 in complete culture medium for 24 h at 37°C resulted in equal fold-induction of HIV-1 LTR-driven luciferase activity in 1G5 cells as compared with incubation with fresh PGE 2 (data not shown), which demonstrates that PGE 2 is fairly stable under these experimental conditions. Taken together, these data indicate that activation of HIV-1 regulatory elements by PGE 2 was rapid and transient, thereby suggesting that the effect was direct and was resulting from PGE 2 -mediated signal transduction events.
Given that PGE 2 is generally seen as a down-modulator of T-cell activation, we were next interested in determining the action of PGE 2 on typical pathways known to lead to HIV-1 LTR activation in T cells. 1G5 cells were hence stimulated with various HIV-1 LTR activators in the absence or the presence of PGE 2 for 8 h. These stimuli were shown, as expected, to act as potent inducers of HIV-1 LTR activity (fold increase over un-  (Fig. 2). Again, a marked up-regulation of HIV-1 LTR-dependent luciferase activity was also seen when 1G5 cells were incubated with PGE 2 alone (10.9-fold increase over untreated 1G5 cells). A PGE 2 -mediated activating effect on HIV-1 LTR was also present with all stimuli used in this set of experiments (fold increase over 1G5 cells treated with each stimuli in the absence of PGE 2 : PHA, 1.8; OKT3, 2.3; PMA, 1.2; and TNF-␣, 4.3). It was thus apparent that PGE 2 specifically synergized with TNF-␣ in activating HIV-1 LTR activity. These results clearly indicated that PGE 2 could further increase the overall positive effect mediated by various HIV-1 LTR-activating agents and thus confirmed that PGE 2 could be considered by itself as a potent inducer of HIV-1 LTR transcription in T cells.
The biosynthesis of prostaglandins is known to be regulated at two different levels. The arachidonic acid, the precursor form, is stored in membrane phospholipids prior to its release into cells by phospholipase A 2 . Free arachidonic acid is then metabolized by cyclooxygenase to an intermediate that leads to the formation of prostaglandins (54). To exclude any effect by endogenous PGE 2 in our studies, 1G5 cells were pretreated with concentrations of indomethacin sufficient to inhibit cyclooxygenase-1 (0.4, 0.8, 2.0, and 10.0 M) and cyclooxygenase-2 (150 M) (55) before stimulation with PGE 2 for 8 h. No changes in PGE 2 -dependent increase in HIV-LTR driven luciferase activity could be detected with indomethacin suggesting that activation of HIV-1 LTR-driven gene expression was only due to exogeneous PGE 2 (data not shown).
Signaling Events Involved in PGE 2 -mediated Enhancement of HIV-1 LTR Activity-Our results so far suggested that binding of PGE 2 to its cell surface receptors (EPs) triggers signal transduction that positively affect HIV-1 LTR expression. Several specific inhibitors were used to identify cellular element(s) participating to the signaling events involved in the PGE 2induced effect on HIV-1 transcription. 1G5 cells were first pretreated with H7, a selective serine/threonine kinase inhibitor that can inhibit protein kinase A (PKA) (K i ϭ 3.0 M), PKC (K i ϭ 6.0 M), as well as PKG (K i ϭ 5.8 M) (56). A dose-dependent inhibition of PGE 2 -mediated HIV-1 LTR activation was seen when 1G5 cells were pretreated with H7 at concentrations sufficient to inhibit all of these enzymes (Fig. 3A). To more deeply scrutinize signaling molecule(s) implicated in PGE 2 -induced positive effect on HIV-1 LTR activity, cells were next pretreated with HA-1004, a serine/threonine kinase inhibitor that preferentially inhibits PKA (K i ϭ 2.3 M) and PKG (K i ϭ 1.3 M) over PKC (K i ϭ 40.0 M) (56). The PGE 2 -mediated activation of HIV-1 transcription was almost completely abrogated by a pretreatment with HA-1004 at concentrations sufficient to inhibit both PKA and PKG, but not PKC (1 and 5 M) (Fig. 3B). To clearly discriminate between PKA and PKG in the PGE 2 -induced signaling cascade, 1G5 cells were pretreated with a specific inhibitor of adenylate cyclase (MDL-12, 330A) (57)(58)(59). Activation of HIV-1 LTR-mediated reporter gene expression by PGE 2 was totally abolished by concentrations of MDL-12,330A sufficient to completely inhibit cAMP activity (Fig. 3C). Results from these experiments hence demonstrated that PKA was an essential intracellular second messenger participating in the PGE 2 -dependent up-regulating effect on HIV-1 LTR activity in T cells.
Adenylate cyclase transforms adenosine triphosphate (ATP) in cyclic adenosine monophosphate (cAMP), which is necessary for PKA activity. Knowing that PKA activation is negatively modulated by phosphodiesterase due to transformation of cAMP to 5Ј-AMP, cells were pretreated with IBMX, an inhibitor of phosphodiesterase activity (60). It is hence presumed that PGE 2 -induced activation of PKA should be sustained for longer periods of time in cells treated with IBMX thus resulting in a greater stimulation of HIV-1 LTR-dependent gene activity. The increase in intracellular cAMP levels caused by the presence of IBMX indeed led to a dose-dependent enhancement of HIV-1 LTR-driven luciferase activity in cells treated with PGE 2 (Fig. 3D). This PKA/cAMP-dependent activation of HIV-1 transcription is in agreement with previous studies (61,62). It should be noted that treatment of 1G5 cells with IBMX alone (1, 5, and 10 M) had no effect on HIV-1 LTR-dependent reporter gene activity (data not shown).
The implication of Ca 2ϩ in this process was next investigated by pretreating 1G5 cells with increasing concentrations of BAPTA/AM (1, 5, and 10 M), an intracellular Ca 2ϩ chelator (63). The capacity of PGE 2 to influence HIV-LTR activity was monitored as described above. Data from this experiment suggested that Ca 2ϩ was partly involved in this process as the maximal subcytotoxic concentration of BAPTA/AM used (10 M) could not totally eliminate PGE 2 -mediated activating effect on HIV-1 transcription (Fig. 3E). Finally a newly described inhibitor of calcium mobilization, carboxyamidotriazole (CAI) (47), was used to reinforce the implication of Ca 2ϩ in the PGE 2 -induced activation of HIV-1 LTR. Data obtained from this set of experiments confirmed that Ca 2ϩ is indeed an important component of the PGE 2 -initiated signaling cascade which culminates in activation of HIV-1 LTR-dependent gene expression (Fig. 3F). Altogether, the use of specific inhibitors allowed us to demonstrate that PKA, cAMP, and Ca 2ϩ are all involved to some degree to PGE 2 -dependent positive effect on the regulatory elements of HIV-1.
To mimic PGE 2 -induced HIV-1 LTR activation in T cells, 1G5 cells were next treated for 8 h with dibutyryl-cAMP, a cAMP analog, along with the calcium ionophore, ionomycin. Results showed that treatment of 1G5 cells with both chemical compounds was not sufficient to trigger HIV-1 LTR activity (data not shown). However, a dose-dependent significant increase in HIV-1 LTR-driven luciferase activity was seen upon the addition of PMA, ionomycin, and increasing concentrations of dibutyryl-cAMP (25, 50, 100, and 200 M) (Fig. 4). It should be specified that PMA alone was strongly activating HIV-1 LTRdriven gene activity because this agent is recognized as one of the most potent activator of NF-B (64), a pleiotropic transcription factor complex known as a good inducer of HIV-1 expression (65). Therefore, these data suggested that up-regulation of HIV-1 LTR activity requires the activation of the transcription factor NF-B in addition to cAMP-and calcium-dependent signaling pathways. both NF-B-dependent and -independent signaling pathways (data not shown). As shown in Fig. 5A, we observed a 13.7-, 83.3-, and 12.1-fold increase in luciferase activity for the fulllength HIV-1 LTR (pLTR-LUC) in the presence of TNF-␣, PHA/ PMA, and PGE 2 , respectively. With the molecular construct pmBLTR-LUC, as expected, no increase in HIV-1 LTR-driven gene activity was detected with TNF-␣, while PHA/PMA was still inducing HIV-1 LTR-dependent luciferase activity (8.5-fold increase). Interestingly, the luciferase-encoding vector mutated at both NF-B-binding sites of the LTR was still responding to PGE 2 , although at a slightly lower level compared with the wild-type HIV-1 LTR construct (3.8-versus 12.1-fold). This experiment was repeated several times and gave consistent results. Therefore, data from experiments conducted with pLTR-LUC and pmBLTR-LUC were indicating that PGE 2mediated positive effect on HIV-1 LTR-driven activity required both NF-B-dependent and -independent signal transduction pathways.

NF-B-dependent and -independent Signaling Pathways Are Involved in Activation of HIV-1 LTR by PGE 2 -To
To confirm that nuclear translocation and activation of NF-B was indeed induced by PGE 2 , cells were transiently transfected with pB-TATA-LUC. It should be specified that this vector allows monitoring of HIV-1 activation almost exclusively via NF-B-dependent mechanism. Indeed, this vector is made of the minimal HIV-1 NF-B-binding domains and a TATA box driving the luciferase reporter gene (46). In this case, a 15.4-fold increase in HIV-1 LTR-dependent reporter gene activity was seen in cells treated with PGE 2 , thereby confirming that NF-B is directly involved in the PGE 2 -dependent activating effect on the regulatory elements of HIV-1. To further substantiate the participation of NF-B in PGE 2 -mediated up-regulation of HIV-1 LTR transcription, Jurkat E6.1 were transiently transfected with pNF-B-LUC, a vector made of five consensus binding sites for NF-B, prior to incubation with PGE 2 . The implication of NF-B in the PGE 2 -mediated activating effect on HIV-1 transcription was again clearly shown using this B-driven reporter gene construct (17.5-fold increase) (Fig.  5B). The involvement of NF-B was also examined by mobility shift assays. Results shown in Fig. 6 revealed the presence of a band specific for NF-B that is induced following the treatment for 1 h with either PGE 2 or TNF-␣ (lanes 3 and 4, respectively). The specific band for NF-B was eliminated by competition experiment with unlabeled probe for NF-B. These results were thus reinforcing the notion that PGE 2 is up-regulating HIV-1 LTR dependent activity also through a NF-B-dependent mechanism.
It has recently been reported that the nuclear factor of activated T cells (NFAT) can synergize with NF-B and the viral transactivating protein Tat in transcriptional activation of HIV-1 following its binding to the NF-B binding sequences (67). We therefore evaluated the putative implication of NFAT in the PGE 2 -induced up-regulation of HIV-1 LTR activity. Using a molecular construct made of the luciferase reporter gene placed under the control of the minimal IL-2 promoter containing three tandem copies of the NFAT-binding site, we observed that transiently transfected Jurkat E6.1 cells showed no induction of luciferase activity by the addition of PGE 2 . In contrast, a marked increase in reporter gene activity was observed in cells treated with a combination of PMA/PHA (data not shown). We could conclude from these data that the transcription factor NFAT was not playing a role in the PGE 2 -induced activation of HIV-1 LTR transcription.
Surface Expression of the Human PGE 2 Receptor EP 4 Subtype in the Studied T Lymphoid Lineages-PGE 2 is known to bind to specific protein receptors on a large array of target cells.
Previous cDNA cloning and pharmacologic experiments have identified six different PGE 2 receptors (EPs) (68). Northern blot analyses were then performed to evaluate PGE 2 receptor(s) expression on the T lymphoid cell lines used in the present study. The cDNAs for hEP 1 , hEP 2 , hEP 3 , and hEP 4 were hybridized with total RNA from 1G5 and Jurkat E6.1 cells. Results indicated that the hEP 4 gene was expressed on both 1G5 and Jurkat E6.1 cell lines, while EP 1 , EP 2 , and EP 3 subtype receptors were not expressed (data not shown). Flow cytometry analysis was also carried out with a polyclonal antibody specific for EP 4 receptor and confirmed its presence on the surface of Jurkat E6.1 (Fig. 7A) and 1G5 (Fig. 7B) cells. The identity of the prostaglandin receptor on T lymphoid 1G5 cells was directly addressed using a subtype selective pharmacologic antagonist. For this purpose, 1G5 cells were pretreated with AH 23848B (30 M), a selective antagonist of human EP 4 receptor, prior to the addition of PGE 2 . AH 23848B was found to abrogate PGE 2 -mediated up-regulation of HIV-1 LTR activity in 1G5 cells (Fig. 8). Altogether these results indicate that the hEP 4 receptor is involved in the PGE 2 -mediated activating effect on HIV-1 transcription in T lymphoid cells.

DISCUSSION
Immune and inflammatory responses are triggered by microorganisms such as bacteria, viruses, and protozoan, all known to be potential opportunistic pathogens in HIV-1-positive patients. The formation and production of elevated levels of inflammatory mediators such as PGE 2 is a hallmark of the HIV-1 infection (28 -30). Prostaglandins play a role in disease exacerbation by directly altering T-cell functions or macro-phage activation. Although it was thought that PGE 2 is primarily an immunosuppressive molecule that acts as a downregulator of many aspects of B-and T-cell function and proliferation, recent findings support a role for PGE 2 as a potentiator of immunoglobulin class switching and cytokine and cytokine receptor synthesis (15). This PGE 2 -dependent positive effect on the immune response and the observation that higher levels of PGE 2 are detected in HIV-1-infected individuals (2-5-fold increase) have been the compelling force for our investigation. Knowing that PGE 2 is a good inducer of cAMP and that a 4-fold increase in intracellular levels of cAMP is seen in asymptomatic HIV-1-seropositive subjects as compared with uninfected controls (69), it was thus of prime importance to study the putative effect of PGE 2 on the regulatory elements of HIV-1.
We therefore asked whether the proinflammatory PGE 2 molecule had the ability to modulate HIV-1 transcription in T cells. This particular cell type was chosen since T cells are considered to be a major cellular reservoir for HIV-1 in the human peripheral blood (70). In this report, we present evidence indicating that PGE 2 up-regulates HIV-1 LTR-driven reporter gene expression in human T cells (Fig. 1). Our results are indicative of an optimal signal after 6 to 8 h of treatment. This is very similar to time kinetics of HIV-1 LTR activation by PMA, PHA, and TNF-␣ agents which directly act on the HIV-1 promoter. Although we cannot refute a possible indirect mechanism for HIV-1 LTR activation by PGE 2 which would involve production of cytokines, time course experiments suggested that a more direct process might be at the basis of the effect of PGE 2 on HIV-1 LTR. Interestingly, a specific synergistic HIV-1 LTR activation was observed using both PGE 2 and TNF-␣ (Fig. 2). This might be accounted by the fact that the effect exerted by TNF-␣ is exclusively via NF-B, while PGE 2 acts also on region(s) other than NF-B in the HIV-1 LTR (see below). Previously described synergistic activation by NF-B and other factors have been reported (66,67). The induction of HIV-1 LTR activity by PGE 2 thus support a previous cellular study showing a 2.5-fold increase in virus production following the addition of exogeneous PGE 2 to MT-4 cells acutely infected with HIV-1 (43).
In the present study, the involvement of specific intracellular second messengers in PGE 2 -mediated up-regulation of HIV-1 LTR activity has been dissected using several signal transduction inhibitors. Experiments with indomethacin, a potent inhibitor of the cyclooxygenase pathway and thus of PGE 2 pro-  duction, suggest that only exogeneous PGE 2 plays a role in the activation of HIV-1 LTR-driven gene expression. These results were expected based on studies that T cells had a limited capacity to metabolize arachidonic acid to prostaglandins (71)(72)(73). An earlier report indicated that interaction between PGE 2 and an adenylate cyclase-coupled stimulatory receptor leads to activation of adenylate cyclase, hydrolysis of ATP, enhanced turnover of intracellular cAMP, and binding to PKA (74). Our findings are clearly supportive of this signaling cascade since we found that PGE 2 -induced enhancement of HIV-1 LTR dependent activity requires the participation of adenylate cyclase, cAMP, and protein kinase A (Fig. 3, A-D). Using MT-4 cells, another group has shown that elevation of cAMP levels resulted in HIV-1 replication (62). It is also well known that cAMP-dependent pathways regulate the immune effector functions of lymphocytes and macrophages. For example, during immune response, cAMP exhibits positive regulatory effects at low concentrations whereas inhibitory effects are seen at high concentrations (75). Many of the earlier studies have shown that PGE 2 interaction with T cells in vitro resulted in an elevation of the cAMP level (35) and that such elevated intracellular cAMP levels were responsible for the proliferative disturbances in T cells (76 -78). Data from our experiment with the calcium chelator BAPTA/AM and the calcium inhibitor CAI are suggestive of the importance of Ca 2ϩ in the PGE 2 -induced activation of HIV-1 transcription (Fig. 3, E and F). However, given that there is no published report indicating Ca 2ϩ influx through the EP 4 receptor, our results with BAPTA/AM and CAI, two inhibitors of intracellular calcium mobilization, lead us to postulate that PGE 2 could generate calcium release from intracellular storage organelles. All these results were supported by data shown in Fig. 4 indicating that up-regulation of HIV-1 LTR requires the implication of cAMP and calcium, as well as the participation of the NF-B transcription factor.
Several agents known as potent activators of HIV-1 transcription (e.g. PMA, PHA, TNF-␣, and anti-CD3 antibody) are all acting through a common mechanism, namely via the nuclear translocation of the transcription factor NF-B which binds to the enhancer region of the HIV-1 LTR (79). This transcription factor is sequestered in the cytoplasm due to its physical association with the inhibitor named IB. NF-B is a pleiotropic transcription factor that controls the expression of a wide variety of genes, including cytokines such as IL-1, IL-2, IL-6, IL-8, interferon-␤, and TNF-␣, as well as known genes for some cell adhesion molecules including ICAM-1 and VCAM-1. Its importance in the regulation of HIV-1 gene expression has been stated in numerous studies (80). Results from mobility shift assays suggest that the PGE 2 -mediated effect on HIV-1 LTR activity is due to activation of the transcription factor NF-B by PGE 2 . This is in agreement with the previous demonstration that PGE 2 activates NF-B in the macrophage-like cell line J774 (81). The fact that we have noticed that both NF-B and Ca 2ϩ are key elements in the PGE 2 effect on HIV-1 transcription is of interest considering that calcineurin, a Ca 2ϩ / calmodulin-dependent serine/threonine protein phosphatase, has been reported to activate NF-B through the inactivation of IB (82). Moreover, researchers had earlier found that cAMPmediated enhancement of PKA might be involved in the dissociation of IB from NF-B (79). Recent studies have revealed that NF-B is regulated through phosphorylation of the p65 subunit by PKA which is directly regulated by intracellular levels of cAMP (83). Our experiments hence support the notion that PGE 2 might be activating the transcription factor NF-B via cAMP/PKA and calcium signaling pathways in human T lymphoid cells. However, our experiments were performed with B-driven reporter gene constructs (pB-TATA-LUC and pNF-B-LUC) and HIV-1 LTR-based vectors (pLTR-LUC and pmBLTR-LUC), furthermore, suggest that NF-B-binding regions and another element(s) in the HIV-1 LTR are involved in the activation of HIV-1 LTR-dependent transcription induced by PGE 2 .
NFAT is an immediate-early activation factor that plays a crucial role in T-cell activation and commitment processes through its control of IL-2 gene activation (84). Based on the demonstrated synergistic effect between NFAT and NF-B on the activation of HIV-1 transcription (67) and the proposed PGE 2 -induced NF-B-independent pathway, we looked at the putative role of NFAT in the effect of PGE 2 . We found that NFAT was not involved in PGE 2 -dependent activation of HIV-1 LTR-driven luciferase activity. These data were expected considering that PGE 2 has been reported to inhibit NFAT activity (85). PGE 2 could have the capacity to modulate several signal transduction pathways through its effect on transcription factors regulated by cAMP such as the cAMP response-element binding factor, the activating protein-1 (38), and Sp1 (86). The involvement of these three transcription factors in the observed NF-B-independent activation of HIV-1 LTR mediated by PGE 2 is currently under investigation.
Finally, by Northern blot assays, flow cytometric analyses, and pharmacological studies, we demonstrated that studied T lymphoid cell lines (Jurkat E6.1 and 1G5) express the EP 4 receptor subtype on their surfaces and that EP 1 , EP 2 , and EP 3 receptors seem not to be expressed (Figs. 7 and 8 and data not shown). This finding is in accord with previous studies that have found by Northern blot analysis that the EP 4 gene is expressed on T lymphoid cells such as Molt-4, KM-3, and Jurkat E6.1 (87)(88)(89). It has been demonstrated that EP 4 receptors are coupled to adenylate cyclase via a stimulatory G protein (G ␣s ) and that such activation results in an enhancement of intracellular cAMP levels (53,68). Interestingly, PGE 2 has been shown to lead to an increase in intracellular cAMP levels partly via the EP 4 receptor (90), a finding which lend credence to the potential implication of the EP 4 receptor in the PGE 2induced up-regulation of HIV-1 LTR activity.
Because of their intrinsic intracellular obligatory parasitic form of life, viruses depend heavily on cell metabolic machinery for their replication. Thus, changes in cellular metabolism might influence the viral life cycle. The experiments reported here highlight the positive action of PGE 2 , a powerful cAMPinducing agent, on the regulatory elements of HIV-1. The presented data suggest that elevated levels of PGE 2 detected in HIV-1-infected persons or induced by opportunistic pathogens might actively participate to immunological disturbances associated with AIDS and modify the pathogenesis of this retroviral disease by inducing a higher viral load. Finally, high concentrations of PGE 2 (up to 100 M) found in seminal fluids of HIV-1-infected persons might directly enhance virus replication and facilitate viral transmission during sexual activities (40).