Skip to main content
NCBI home page
Immunology logoLink to Immunology
. 2004 Mar;111(3):298–305. doi: 10.1111/j.1365-2567.2004.01810.x

Prostaglandin E2 is a potent regulator of interleukin-12- and interleukin-18-induced natural killer cell interferon-γ synthesis

William Walker *, Dino Rotondo †
PMCID: PMC1782426  PMID: 15009430

Abstract

Synthesis of interferon (IFN)-γ by natural killer (NK) cells is an important pro-inflammatory event with interleukin (IL)-12 and IL-18 playing major inductive roles. However, other temporal events are likely to regulate such processes and as prostaglandin E2 (PGE2) is ubiquitous during inflammation this study tested the hypothesis that PGE2 was capable of directly modulating cytokine-induced NK cell IFN-γ synthesis in the absence of other immune cells. Using homogenous NK cell lines to establish direct effects, PGE2 (0·1–1 µm) was found to suppress NK cell IFN-γ synthesis and antagonized the potent synergistic IFN-γ-inducing effects of IL-12 and IL-18. The actions of PGE2 were mimicked by synthetic PGE2 analogues including misoprostol and butaprost. The selective EP2 receptor agonist butaprost, but not the EP1/EP3 agonist sulprostone, suppressed IFN-γ synthesis and exclusively competed with PGE2 for receptor binding on NK cells. Further analysis showed that PGE2 did not modulate IL-12 receptor mRNA expression and the effects of PGE2 could be mimicked by the phosphodiesterase inhibitor 3-iosobutyl-1-methylxanthine. The absence of demonstrable receptor modulation coupled with the observed suppression of IFN-γ synthesis by both EP2 receptor-selective agonists and IBMX suggest that PGE2 acts directly on NK cells via EP2 receptors with its downstream effects on cAMP metabolism. This conclusion is further supported by findings that PGE2 and its analogues consistently elevated levels of cAMP in NK cells. The ability of PGE2 to antagonize the potent inductive signal provided by the combination of IL-12 and IL-18 supports the concept that PGE2 may play an important role in limiting innate inflammatory processes in vivo through direct suppression of NK cell IFN-γ synthesis.

Introduction

Natural killer (NK) cells play a vital role in resistance to cancer and various infectious agents that include viruses, bacteria and intracellular protozoa.1–3 Upon activation NK cells synthesize interferon-γ (IFN-γ)), a major pro-inflammatory cytokine that is essential for activating macrophage microbicidal activity4 and also for promoting subsequent T helper 1 (Th1)-like immunity.5 The induction of NK cell IFN-γ synthesis occurs rapidly, primarily in response to macrophage and dendritic cell-derived interleukin-12 (IL-12)6 and this process is amplified by other pro-inflammatory cytokines especially IL-18, which acts in synergy with IL-12.7–10 The importance of these innate events are aptly demonstrated by in vivo studies showing that the absence of either IL-12 or IL-18 results in an inability to mount early IFN-γ-mediated resistance and profoundly impaired Th1-like responses to intracellular pathogens.11

In addition to regulating numerous physiological processes throughout the body, prostaglandin E2 (PGE2) is ubiquitously expressed in many tissues during inflammation and is synthesized by diverse cell types that include immune cells, fibroblasts and skin cells. In the context of innate immunity, NK cell activity involves close interaction with activated dendritic cells and macrophages12 and significantly, these cell types are major producers of PGE2.13 Given the synthesis of PGE2 by such immune cells, it is perhaps not surprising that PGE2 has been shown to modulate various aspects of the immune response. For example, PGE2 has been shown to inhibit lipoloysaccharide (LPS)-induced IL-12 production from monocytes14 decrease IL-2-driven T-cell proliferation15 and suppress IFN-γ synthesis by CD4+ or CD8+ T cells.16,17 However, conflicting reports have demonstrated PGE2-mediated enhancement of IFN-γ synthesis by antigen-stimulated Th1 cells18 and enhanced dendritic cell IL-12 production in the absence of LPS stimulation.19 These contradictory findings may be because of differences in cellular activation or on the differential expression of PGE2 receptors (EP1–EP4) as these individual G-protein-coupled receptors utilize distinct intracellular signalling pathways. For example, EP2/EP4 receptors act via Gs proteins to increase intracellular cAMP, and the three isoforms of the EP3 receptor act via Gi proteins to suppress cAMP levels.20

In contrast to T lymphocytes, the effects of PGE2 on NK cell IFN-γ synthesis has received considerably less attention. In particular, given the importance of this fundamental event to host immunity, it was pertinent to determine whether PGE2 can directly modulate the ability of IL-12-, IL-18-induced NK cell IFN-γ synthesis in the absence of other immune cells. In addressing this important question we demonstrate that physiologically relevant concentrations of PGE2 significantly suppresses NK cell IFN-γ synthesis, irrespective of cytokine stimulation. Furthermore, this suppression is a result of the direct action of PGE2 on NK cells, involving EP2 receptors and the downstream modulation of intracellular cAMP levels. This infers that concomitant production of PGE2 may be an important innate event that regulates NK cell activity in vivo.

Materials and methods

Propagation of NK cells

IL-2-dependent murine NK cell lines (5E3, 5E6) were obtained from Dr Hiroko Tsutsui, Department of Immunology & Medical Zoology, Hyogo College of Medicine, Japan and maintained as described.21 These cell lines are non-transformed and exhibit identical characteristics to primary NK cells obtained as previously described by us.10 Briefly, cells were cultured in RPMI-1640 containing 10% fetal calf serum (FCS), 2 mm l-glutamine, 100 U/ml penicillin, 10 mg/ml streptomycin, 1 mm sodium pyruvate, 10 mm HEPES and 1% non-essential amino acids. Cells were seeded at an initial density of 1 × 105 cells/ml and maintained in 500 U/ml of recombinant human IL-2 (Peprotech EC Ltd, London, UK). The IL-2 was replenished every 5–6 days when cells reached confluence. In all experiments NK cells were plated at a final density of 1 × 106/ml in either 24-well tissue culture plates (1 ml/well) or in the case of larger cultures (5 ml) in 25 cm2 tissue culture flasks.

Stimulation of NK cells

In all stimulation experiments NK cells were cultured with recombinant murine IL-12 and/or IL-18 (Peprotech EC, Ltd) at a final concentration of 1 ng/ml. Tissue culture grade PGE2 was obtained from Sigma-Aldrich Co. Ltd (Poole, UK) and used at a concentration of 1 µm unless otherwise stated. The PGE2 analogues butaprost, misoprostol (selective EP2 agonists) and sulprostone (EP1/EP3) were obtained from Cayman Chemical Co. (Ann Arbor, MI) and were used at a final concentration of 10 µm. The cAMP phosphodiesterase inhibitor 3-iosobutyl-1-methylxanthine (IBMX) was added to cultures at a final concentration of 100 µm.

Measurement of IFN-γ synthesis

IFN-γ levels in culture supernatants were measured using a standard capture cytokine enzyme-linked immunosorbent assay (ELISA) protocol. Antibody pairs were supplied by BD-PharMingen (Oxford, UK) and used as recommended. Briefly, the wells of Immulon™ microtitre plates (Dynex Technologies Inc., Chantilly, VA) were coated with detecting monoclonal antibody (R4-6A2) in phosphate-buffered saline (PBS) pH 9·0. After addition of samples, bound cytokine was detected with biotinylated anti-IFN-γ (XMG1.2) followed by streptavidin–alkaline phosphatase conjugate. Bound phosphatase was visualized with 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulphonic acid) (Sigma-Aldrich Chemical Co.). Unknown values were extrapolated from a standard curve which was constructed using recombinant mouse IFN-γ of known concentration.

Ligand-binding analysis

PGE2 receptor-binding analysis was performed as previously described.22 Briefly, washed NK cells (1 × 106) were incubated at 4° with 3 nm[3H]-PGE2 (Amersham Biosciences UK Ltd, Little Chalfont, UK) alone or in combination with 1000-fold molar excess concentrations (3 µm) of either unlabeled PGE2 or its analogues (butaprost, sulprostone) for 60 min in a total volume of 200 µl. After incubation cells were vacuum-filtered through Whatman GF/C glass-fibre filter papers, presoaked with Tyrode's buffer. Following washing, the filters were dried and radioactivity measured by liquid scintillation counting.

Cytokine receptor expression

To detect cytokine receptor expression total RNA was extracted from cultured NK cells by the guanidine isothiocyanate method and assayed for cytokine receptor mRNA content using the Riboquant™ Multi-Probe RNase Protection Assay System (BD-PharMingen). Briefly, 10 µg of RNA from each sample was hybridized in solution with the appropriate radiolabelled antisense RNA probe set (mCR-1, mCR-3) as recommended by the manufacturers. Following hybridization, free probe and remaining single stranded RNA was digested with RNases and the protected probes were purified and resolved on 5% denaturing polyacrylamide gels using Ultra Pure Sequagel™ reagents (National Diagnostics, Atlanta, GA). Protected fragments were visualized by exposing dried gels to standard X-ray imaging film. Post-development films were scanned and individual band densities analysed by NIH-Image computer software.

Measurement of intracellular cAMP levels

NK cells (1 × 106) were cultured in the presence of PGE2 or its synthetic analogues in a final volume of 250 µl RPMI-1640. Reactions were terminated after 30 min by the addition of trichloroacetic acid (TCA, 5% final concentration). Supernatants were then collected into fresh tubes and TCA removed by washing three times with 4 volumes of diethyl ether. cAMP was subsequently measured in samples by radioimmunoassay. Briefly, a constant amount of radiolabelled cyclic AMP and mouse anti-cAMP antibody was added to each sample. Samples were then incubated at 4° for 18 hr after which a second antibody (goat anti-mouse), conjugated to magnetic beads, was added followed by ice-cold polyethylene glycol 6000 (12% final concentration). Samples were then vortex-mixed and centrifuged at 13 000 g to remove the supernatant. Radioactivity in the pellet was measured by scintillation counting and the level of cAMP evaluated by comparison to the radioactivity in tubes containing standard amounts of cAMP.

Statistical analysis

All data are expressed as mean values ± 1 standard deviation (SD). Protein IFN-γ levels were compared by the two-tailed Student's t-test. Specific mRNA expression was calculated according to the formula: densitometric values for cytokine receptor bands/L32 band density. Normalized densitometric values under different stimulatory conditions were compared by Mann–Whitney U-test. P-values less than 0·05 were considered significant.

Results

PGE2 suppresses IL-12 & IL-18-induced NK cell IFN-γ synthesis

As PGE2 is known to have modulatory effects on accessory cell functions that are pertinent to NK cell IFN-γ synthesis, i.e. IL-12 production,14,19 it was important to determine direct effects on NK cells in the absence of these cell types. As contaminating accessory cells are unavoidable during generation of primary NK cell cultures from bone marrow precursors, homogeneous non-transformed NK cell lines were employed.21 In our hands these NK cells exhibited identical characteristics to primary NK cells generated by other methods.10 To mimic physiological conditions as much as possible, NK cells were stimulated with only 1 ng/ml final concentrations of cytokine, and up to 1 µm final concentrations of PGE2, a quantity previously shown to maximally inhibit macrophage IL-12 production.14 In agreement with our previous findings using primary NK cells10 IL-12 and IL-18 acted in synergy to significantly enhance production of IFN-γ when compared with stimulation by IL-12 or IL-18 alone (Fig. 1a). However, irrespective of the inducing cytokine stimulus, PGE2 suppressed the synthesis of IFN-γ. PGE2 was effective at concentrations as low as 0·01 µm with a maximal suppressive effect between 0·1 and 1 µm (Fig. 1b). Despite the potent synergism of IL-12 and IL-18, 1 µm PGE2 was able to significantly reduce IFN-γ levels under these conditions (41·5% suppression). Under weaker stimulatory conditions the suppressive effects of PGE2 were more profound with IL-12- and IL-18-stimulated IFN-γ levels reduced on average by 58 and 67%, respectively, suggesting a direct antagonism between IFN-γ-inducing stimuli and PGE2. These inhibitory effects were not caused by cell death as PGE2 did not have any significant effect on cell viability during the relatively short culture conditions (16 hr) as determined by trypan blue exclusion (not shown). We did not perform experiments with excess PGE2 (i.e. >1·0 µm) as these levels are unlikely to reflect pathophysiological conditions given the very short half-life of PGE2in vivo.23,24 These findings show that cytokine-induced NK cell IFN-γ synthesis is profoundly regulated in a negative manner by PGE2.

Figure 1.

Figure 1

Open in a new tab

PGE2 suppresses IL-12- and IL-18-induced IFN-γ synthesis by NK cells. (a) NK cells (1 × 106) were stimulated with IL-12 ± IL-18 (1 ng/ml) for 16 hr in the absence (ethanol vehicle) or presence of 1·0 µm PGE2 and IFN-γ in culture supernatants measured by standard capture ELISA. The synergistic combination of IL-12 + IL-18 in these experiments was susceptible to the suppressive effects of PGE2 with an average reduction in IFN-γ levels of 41·5%, **P < 0·05 (µ = 26·2 ± 3·1 ng/ml to 15·33 ± 3·0 ng/ml). IL-12-stimulated IFN-γ levels were reduced on average by 58·6%, *P < 0·05 (µ = 10·55 ± 2·3 ng/ml to 4·1 ± 0·9 ng/ml). IL-18-induced IFN-γ levels (1·1 ± 0·4 ng/ml) were more markedly reduced with a 67% reduction. Measurable background IFN-γ levels (µ = 0·6 ng/ml) were completely abolished in the presence of 1·0 µm PGE2. n = 3 ± SD. (b). Dose–response experiments further confirmed that PGE2 suppressed the production of IFN-γ under all stimulatory conditions with PGE2 having a maximal suppressive effect between 0·1 and 1·0 µm final concentration. Data is representative of 3 individual experiments, error bars = ± SD. The IC50 for PGE2 in the presence of IL-12/18 = 360 nm. In the presence of IL-12 or IL-18 the IC50 for PGE2 = 33 nm and 20 nm, respectively.

PGE2 mediates its effects on NK cell IFN-γ synthesis via EP2 receptors

Given the marked suppression of IFN-γ synthesis in the absence of accessory cells it was important to determine the molecular basis for these direct effects. To determine which PGE2 (EP) receptors where involved we employed synthetic analogues of PGE2 that act as selective EP receptor agonists. When NK cells were stimulated in the presence of the selective EP2 receptor agonist butaprost or the EP2/EP3/EP4-active analogue misoprostol, a significant inhibition of IFN-γ synthesis occurred in a manner analogous to that seen with natural PGE2 (Fig. 2). Butaprost suppressed IL-12 + IL-18-induced IFN-γ levels on average by 33% (from 26·3 ng/ml to 17·65 ng/ml) with misoprostol also exhibiting similar effects (reduced to 15·4 ng/ml) when compared with natural PGE2 (15·3 ± 5·1 ng/ml). Similar to PGE2, butaprost and misoprostol also suppressed more efficiently the synthesis of IL-12-stimulated or IL-18-stimulated IFN-γ production. In contrast, the selective EP1/EP3 receptor agonist sulprostone did not have any significant effect on IL-12-or IL-18-induced IFN-γ synthesis (µ = 25·7 ng/ml), suggesting an absence of EP1/EP3 receptors on activated NK cells.

Figure 2.

Figure 2

Open in a new tab

PGE2 analogues selective for EP2 receptors mediate suppressive effects on NK cell IFN-γ synthesis. NK cells (1 × 106) were stimulated with cytokines for 16 hr in the presence of either PGE2 (1 µm), butaprost (EP2 agonist, 10 µm), misoprostol (EP2/EP3/EP4 agonist, 10 µm) or sulprostone (EP1/EP3 agonist, 10 µm) and IFN-γ production measured by ELISA. In the presence of IL-12 + IL-18 butaprost reduced mean IFN-γ levels from 26·3 ± 4·1 ng/ml to 17·65 ± 2·6 ng/ml (33% reduction, **P < 0·05, n = 3 ± SD) compared with a similar reduction in the presence of PGE2 (to 15·3 ± 5·1 ng/ml, *P < 0·05). Similarly, misoprostol also inhibited IL-12 + IL-18-induced IFN-γ synthesis by NK cells (mean IFN-γ synthesis in the presence of misoprostol = 15·4 ± 3·4 ng/ml, ***P < 0·05). Sulprostone had no demonstrable effect on IFN-γ synthesis (mean IFN-γ concentration = 25·7 + 3·4 ng/ml in the presence of sulprostone, P > 0·05 compared with vehicle). The absence of EP1/EP3-mediated effects (sulprostone) combined with the activity of misoprostol and butaprost, strongly indicate the involvement of EP2 receptors in IFN-γ suppression. As shown, similar effects were discernable when these analogues were added to NK cells stimulated in the presence of IL-12 or IL-18 only. n = 3 ± SD.

Although butaprost has been well characterized as a selective agonist for the EP2 receptor, it was desirable to rule out possible indirect effects of this drug on NK cells and also to confirm that the EP2 receptor was the major ligand for PGE2 on the surface of NK cells. Therefore, radio-ligand binding analysis was performed in which PGE2 and its analogues were allowed to compete with [3H]-PGE2 for binding to NK cells (Fig. 3). Total binding of PGE2 was significantly reduced only by the presence of either unlabelled PGE2 or the EP2-selective agonist butaprost (reduced to 53·4% ± 7·3% and 48·6% ± 5·5%, respectively, P < 0·05). In contrast, sulprostone did not have any significant effect on radiolabelled PGE2 binding, confirming the absence of significant numbers of EP1/EP3 receptors on NK cells. Given that butaprost does not interact with the EP1, EP3, or EP4 receptors25–27 these findings strongly support the conclusion that PGE2 mediates its effects on NK cells predominantly through the EP2 receptor.

Figure 3.

Figure 3

Open in a new tab

NK cells predominantly express EP2 receptors. Washed NK cells (1 × 106) were incubated with 3 nanomolar [3H]-PGE2 in the presence of excess (3 µm) synthetic EP receptor agonists and bound [3H]-PGE2 determined by liquid scintillation counting. In the presence of unlabelled PGE2 binding was reduced to 53·4 ± 7·3% (*P < 0·05). In the presence of butaprost binding was reduced to 48·6 ± 5·5% (**P < 0·05) demonstrating EP2 receptor expression. In contrast, coincubation with sulprostone did not result in any alteration in [3H]-PGE2 binding (107 ± 8·9%, P > 0·05) confirming the absence of EP1 or EP3 receptors on NK cells. Total binding was 290 ± 21 dpm/106 cells (n = 4, ± SD).

PGE2 does not modulate IL-12 cytokine receptor expression on NK cells

Previous studies with T lymphocytes have indicated that PGE2 has the ability to down-regulate the IL-12 receptor β subunits (IL-12Rβ)28 and as such a mechanism could account for the observed suppression, the ability of PGE2 to modulate NK cell IL-12R expression was therefore determined. Although not involved in either IL-12 or IL-18 signaling, the effects of PGE2 on expression of the IL-2R subunits (α, β, γc) were also evaluated as an indicator of any nonspecific effects. As an additional control, the effects of PGE2 were compared with TGF-β1, a reported inhibitor of IL-12Rβ2 expression on T cells.29 As previous studies have demonstrated a direct correlation between IL-12R mRNA and cell-surface protein expression30 we measured IL-12Rβ mRNA subunit expression by RNase protection assay (RPA). Irrespective of the stimulatory conditions, PGE2 treatment did not inhibit expression of mRNA for either the IL-12 or IL-2 receptor subunits (Fig. 4). In comparison, down-regulation of both IL-12R subunits at the mRNA level was detectable after TGF-β1 treatment for 16 hr (not shown), validating the sensitivity of the mRNA assay. Therefore, PGE2 does not suppress cytokine-induced IFN-γ synthesis by a mechanism that involves down-modulation of IL-12 receptors on NK cells. Furthermore, the absence of demonstrable down-regulation of other cytokine receptors (including the IL-2R family subunits), suggests that at least under the short treatment time used here, PGE2 does not significantly modulate cytokine receptor expression on NK cells.

Figure 4.

Figure 4

Open in a new tab

PGE2 does not modulate NK cell cytokine receptor expression. NK cells (5 × 106) were stimulated with IL-12 ± IL-18 (1 ng/ml) for 16 hr in the absence (ethanol vehicle) or presence of PGE2 (1·0 µm) and cytokine receptor mRNA levels determined by RPA. Irrespective of the IFN-γ-inducing conditions, PGE2 did not modify IL-12R β1/β2 chain mRNA expression (left and middle panels). Similarly, PGE2 exposure for 16 hr did not modulate IL-2R subunit (γc) mRNA expression (right panel). IL-2R α or β chain mRNA expression was also unaffected by PGE2 exposure (not shown). Shown is the mean normalized expression (represented as the ratio of receptor band density/L32 band density) + SD, n = 3. White bars = vehicle control, black bars = PGE2-treated cultures.

IFN-γ synthesis by NK cells is suppressed in a cAMP-dependent manner

As PGE2 did not modulate expression of IL-12 receptors, more direct effects on IFN-γ gene expression could account for its suppressive effects. In particular, EP2 receptors signal via Gs proteins to increase intracellular cAMP levels20 and it was therefore pertinent to test whether PGE2 was capable of modulating cAMP levels in NK cells. To achieve this, cAMP levels in NK cells were determined by radioimmunoassay after treatment with PGE2 or its synthetic analogues. In the presence of either PGE2, misoprostol or butaprost, cAMP levels were significantly elevated compared with untreated cells (Fig. 5a). This effect was most marked at the 10 µm and 1·0 µm concentrations. Lower concentrations (0·1 µm) failed to elevate cAMP levels above background control levels (654 ± 139 fmol). In contrast, treating NK cells with identical concentrations of sulprostone did not produce any significant elevation in intracellular cAMP levels. In further experiments, NK cells were incubated with equivalent amounts (1·0 µm) of either PGE2 or butaprost plus sulprostone. Despite the presence of sulprostone in these cultures, PGE2 and butaprost were still capable of elevating NK cell cAMP levels (Fig. 5b). As sulprostone preferentially utilizes PGE2 receptors that are Gi-coupled (EP3) and inhibitory of cAMP generation, any sulprostone binding would predict an antagonism of the cAMP-elevating effects of butaprost or PGE2. As this was not observed, these experiments provide further evidence of the absence of EP3-mediated effects on NK cells. Finally, to directly test whether cAMP-dependent mechanisms could account for the suppression of IFN-γ synthesis, intracellular cAMP levels were artificially elevated in cytokine-stimulated NK cells by coincubating with the cAMP-phosphodiesterase inhibitor IBMX. As observed with natural PGE2, IBMX suppressed the synthesis of IFN-γ under all stimulatory conditions (Fig. 5c). Similar to PGE2, IBMX was also more efficient at suppressing IL-12-stimulated IFN-γ synthesis (64% reduction) than the synergistic effects of IL-12 plus IL-18 (30% reduction). IL-18-induced and background levels of IFN-γ were abolished in the presence of IBMX. PGE2 plus IBMX combined had an additive effect with a more marked suppression of IFN-γ synthesis than either of these agents on their own, although this was not significant in the IL-12 plus IL-18-stimulated cultures. These findings demonstrate that IFN-γ expression is suppressed via a cAMP-dependent mechanism and this is consistent with the data from selective agonists (Fig. 2) and radio-ligand binding analysis (Fig. 3) showing that PGE2 acts via EP2 receptors on NK cells. They also support our observations that indirect effects such as cytokine-receptor modulation (Fig. 4) are not required for suppression of NK cell IFN-γ synthesis.

Figure 5.

Figure 5

Open in a new tab

IFN-γ synthesis by NK cells is suppressed by in a cAMP-dependent manner. (a) NK cells (1 × 106) were incubated with PGE2 or its analogues and cAMP levels determined by radioimmunoassay. In the presence of 10·0 µm or 1·0 µm PGE2, butaprost or misoprostol cAMP levels were significantly elevated compared with untreated cells (e.g. 10 µm PGE2 = 6612 ± 602 fmol cAMP vs. 654 ± 139 fmol for untreated cells, *P < 0·005). In contrast, sulprostone had no effect on intracellular cAMP levels compared with vehicle-treated control cells (10 µm sulprostone = 627 ± 174 fmol, **P > 0·05). (b) When sulprostone (1·0 µm) was added simultaneously with either PGE2 (1·0 µm) or butaprost (1·0 µm) there was no significant change in the ability of PGE2 or butaprost to elevate cAMP levels (PGE2 = 3923 ± 402 fmol cAMP versus PGE2 plus sulprostone = 4112 ± 517 fmol, P > 0·05; butaprost = 2123 ± 343 versus butaprost plus sulprostone = 2003 ± 287), further confirming the absence of EP1/EP3-mediated effects. (c) In further experiments, NK cells (1 × 106) were stimulated with IL-12 ± IL-18 (1 ng/ml) for 16 hr in the absence (ethanol vehicle) or presence of either PGE2 (1 µm) or IBMX (100 µm) to raise intracellular cAMP levels. Similar to PGE2, IBMX suppressed IFN-γ synthesis irrespective of inducing stimuli. Under stimulation with IL-12 + IL-18, IBMX reduced IFN-γ levels from 59·3 ± 9·1 ng/ml to 42·2 ± 7·1 ng/ml (30% reduction, *P < 0·05). Under stimulation with IL-12 alone, IBMX reduced IFN-γ levels from 9·7 ± 4·0 ng/ml to 3·5 ± 2·1 ng/ml (64% reduction, **P < 0·05). When PGE2 and IBMX were added together there was a small additive effect in terms of suppression, although this was not significant in the IL-12 plus IL-18-stimulated cultures. For all experiments, n = 3 ± SD.

Discussion

Because prostanoids induce the symptomatic manifestations of tissue inflammation they were initially characterized as pro-inflammatory mediators. However, it has become clear that they can both promote inflammation and regulate the actions of cytokines which initiate the inflammatory cascade. This depends on many factors such as the inflammatory stimulus, prostanoid type, and receptor expression, leading to the concept that these lipid molecules are both effectors and regulators of inflammation.31 In agreement with this concept, we report here that PGE2 acts directly on NK cells to inhibit cytokine-induced IFN-γ synthesis. During early immune activation NK cell IFN-γ synthesis is driven principally by IL-12 and IL-1810 with activated macrophages and dendritic cells acting as major producers of these pro-inflammatory cytokines. However, these cell types are also major producers of PGE2, suggesting coordinated, temporal regulation of NK cell activity. The suppressive effect on NK cell IFN-γ synthesis demonstrated in the present study is in addition to the reported inhibitory effects of PGE2 on IL-12 synthesis by activated macrophages14 reflecting an additional layer of regulation. However, unlike macrophages where PGE2 has also been shown to induce suppression through indirect mechanisms that include the induction of regulatory IL-1032 the effects on NK cells appear to be direct, in that the use of pure NK cell lines allowed us to analyse effects in the absence of contaminating accessory cells. Although NK cells have been reported to secrete IL-10, this cytokine was undetectable under the pro-inflammatory conditions used in this and other studies10 and was not detectable after PGE2 treatment. Furthermore, IL-10 has been shown to have stimulatory effects on NK cell IFN-γ synthesis in the presence of IL-1833 and such a scenario is therefore an unlikely explanation for the observed suppression.

PGE2 has been reported to inhibit T cell IFN-γ production17,34 shifting the balance towards Th2 development35 and our findings are consistent with this, in that NK cell-derived IFN-γ production is recognized to be important for the promotion of Th1 development.5 Indeed, it has recently been argued that the higher expression of PGE2 receptors and therefore enhanced sensitivity to IFN-γ-suppressing effects, account for the dominant Th2 response in BALB/c mice.36 Interestingly, EP2 receptors have been demonstrated to mediate the effects of PGE2 on both T cells16 and B cells37 with more recent studies utilizing EP receptor-deficient mice confirming a major role for the EP2 receptor in the modulation of T-cell activity.38 Our findings demonstrate that PGE2 also utilizes EP2 receptors on NK cells to regulate IFN-γ production.

Findings with PGE2 analogues and cAMP-elevating agents such as IBMX present a compelling case for cAMP-mediated regulation of NK cell IFN-γ synthesis. Furthermore, given that NK cells were completely insensitive to sulprostone (EP1/EP3-selective) and given that EP1/EP3/EP4 receptors are insensitive to butaprost25–27 the conclusion that similar to T cells, PGE2 mediates effects on NK cells through the EP2 receptor is valid. However, with regard to the actual molecular basis for these effects we found important differences. Unlike previous findings with anti-CD3-stimulated human mononuclear cells28 we could not demonstrate any effect of PGE2 on NK cell IL-12 receptor subunit expression at the mRNA level. Although we did not directly measure IL-12 binding, previous studies have demonstrated a direct and strict correlation between IL-12R mRNA and cell-surface protein expression30 with PGE2 shown to have equivalent suppressive effects on both IL-12R protein and IL-12R mRNA expression28 validating our approach. The insensitivity of NK cell IL-12R expression to PGE2 contrasts with findings in T cells28 and emphasizes the differential regulation of the IL-12R between these cell types. For example, NK cells constitutively express the IL-12R and although this expression can be up-regulated by cytokines such as IL-239 NK cell IL-12R expression is generally resistant to cytokines that suppress IL-12R expression on T cells.40 Our findings with regard to IL-2R expression are interesting in that recent experiments with IL-15-stimulated human NK cells showed that PGE2 treatment for 2 days led to inhibition of the common-γ chain subunit of the IL-2R (γc).41 In contrast, in our studies where PGE2 exposure was significantly shorter (16 hr maximum) and in the absence of exogenous IL-15, we did not find any significant effect of PGE2 on expression of IL-2R mRNA transcripts. In any case, IL-12 or IL-18 receptor signalling does not involve the γc-chain and would therefore not account for the observed inhibition of IFN-γ synthesis.

The absence of cytokine receptor modulation coupled with strong evidence that PGE2 utilizes EP2-, cAMP-dependent mechanisms suggests regulation of intracellular signalling pathways that impinge on IFN-γ gene expression. In the context of cytokine-induced NK cell IFN-γ synthesis IL-12-activated Jak2/STAT4 and IL-18-activated nuclear factor (NF)-κB/AP-1 are particularly relevant as both are required for maximal IFN-γ promoter activity.42 Indeed, PGE2 has been shown to inhibit NF-AT, AP-1,15 NF-κB43 and Jak signalling44,although whether this applies to NK cells has not been determined. Interestingly, PGE2-induced cAMP has also been shown to result in phosphorylation of cAMP-response element binding protein (CREB)45 and CREB has been heavily implicated in the negative regulation of IFN-γ gene expression through competitive binding to regulatory elements of the IFN-γ promoter.46,47 Whether PGE2 modulates IFN-γ gene transcription in NK cells via CREB-induced suppression of pertinent transcription factors has not been determined and we are currently investigating these and other possibilities.

In summary, this study demonstrates an important role for PGE2 in the regulation of pro-inflammatory cytokine (IL-12 and IL-18)-induced production of IFN-γ by NK cells. Importantly, PGE2 was found to act directly and rapidly antagonize the synergistic effects of IL-12 and IL-18 without modulating cytokine receptor expression. In addition, through the use of synthetic PGE2 analogues and cAMP measurements we conclude that the suppressive effects of PGE2 are mediated via EP2 receptors and its downstream effects on cAMP metabolism. These findings demonstrate direct regulatory effects on NK cells and provide additional important information on the regulation of pro-inflammatory processes to those previously described for accessory cells such as macrophages and dendritic cells. We speculate that the direct effect of PGE2 on NK cells may be an important regulatory event in vivo, acting with other regulators to prevent the pathological consequences of IFN-γ overproduction during early immune activation.

Acknowledgments

We are grateful to Dr Hiroko Tsutsui (Institute for Advanced Medical Sciences, Hyogo College of Medicine, Nishinomiya, Japan) for providing the murine NK cell lines (LNK5E3/5E6). This work was supported in part by a grant from Tenovus Scotland to WW (S00/1).

References

  • 1.Bancroft GJ. The role of natural killer cells in innate resistance to infection. Curr Opin Immunol. 1993;5:503–10. doi: 10.1016/0952-7915(93)90030-v. [DOI] [PubMed] [Google Scholar]
  • 2.Biron CA, Nguyen KB, Pien GC, Cousens LP, Salazar-Mather TP. Natural killer cells in antiviral defense: function and regulation by innate cytokines. Annu Rev Immunol. 1999;17:189–220. doi: 10.1146/annurev.immunol.17.1.189. [DOI] [PubMed] [Google Scholar]
  • 3.Scharton-Kersten TM, Sher A. Role of natural killer cells in innate resistance to protozoan infections. Curr Opin Immunol. 1997;9:44–51. doi: 10.1016/s0952-7915(97)80157-4. [DOI] [PubMed] [Google Scholar]
  • 4.Boehm U, Klamp T, Groot M, Howard JC. Cellular responses to interferon-gamma. Annu Rev Immunol. 1997;15:749–95. doi: 10.1146/annurev.immunol.15.1.749. [DOI] [PubMed] [Google Scholar]
  • 5.Scott P, Trinchieri G. The role of natural killer cells in host–parasite interactions. Curr Opin Immunol. 1995;7:34–40. doi: 10.1016/0952-7915(95)80026-3. [DOI] [PubMed] [Google Scholar]
  • 6.Trinchieri G. Immunobiology of interleukin-12. Immunol Res. 1998;17:269–78. doi: 10.1007/BF02786451. [DOI] [PubMed] [Google Scholar]
  • 7.Hunter CA, Timans J, Pisacane P, et al. Comparison of the effects of interleukin-1α, interleukin-1β and interferon-γ -inducing factor on the production of interferon-γ by natural killer cells. Eur J Immunol. 1997;27:2787–92. doi: 10.1002/eji.1830271107. [DOI] [PubMed] [Google Scholar]
  • 8.Lauwerys BR, Renauld J-C, Houssiau FA. Synergistic proliferation and activation of natural killer cells by interleukin-12 and interleukin-18. Cytokine. 1999;11:822–30. doi: 10.1006/cyto.1999.0501. [DOI] [PubMed] [Google Scholar]
  • 9.Tomura M, Zhou X-Y, Maruo S, et al. A critical role for IL-18 in the proliferation and activation of NK1.1+CD3– cells. J Immunol. 1998;160:4738–46. [PubMed] [Google Scholar]
  • 10.Walker W, Aste-Amezaga M, Kastelein RA, Trinchieri G, Hunter CA. IL-18 and CD28 use distinct molecular mechanisms to enhance NK cell production of IL-12-induced IFN-γ. J Immunol. 1999;162:5894–901. [PubMed] [Google Scholar]
  • 11.Takeda K, Tsutsui H, Yoshimoto T, et al. Defective NK cell activity and Th1 response in IL-18 deficient mice. Immunity. 1998;8:383–90. doi: 10.1016/s1074-7613(00)80543-9. [DOI] [PubMed] [Google Scholar]
  • 12.Gerosa F, Baldani-Guerra B, Nisii C, Marchesini V, Carra G, Trinchieri G. Reciprocal activating interaction between natural killer cells and dendritic cells. J Exp Med. 2002;195:327–33. doi: 10.1084/jem.20010938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Phipps RP, Stein RH, Roper RL. A new view of prostaglandin E regulation of the immune response. Immunol Today. 1991;12:349–52. doi: 10.1016/0167-5699(91)90064-Z. [DOI] [PubMed] [Google Scholar]
  • 14.van der Pouw Kraan TC, Boeije LC, Smeenk RJ, Wijdenes J, Aarden LA. Prostaglandin E2 is a potent inhibitor of human interleukin-12 production. J Exp Med. 1995;181:775–9. doi: 10.1084/jem.181.2.775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Paliogianni F, Boumpas DT. Prostaglandin E2 inhibits the nuclear transcription of the human interleukin 2, but not the IL-4 gene in human T cells by targeting the transcription factors AP-1 and NF-AT. Cellular Immunol. 1996;171:95–101. doi: 10.1006/cimm.1996.0178. [DOI] [PubMed] [Google Scholar]
  • 16.Ganapathy V, Gurlo T, Jarstadmarken HO, von Grafenstein H. Regulation of TCR-induced IFN-γ release from islet-reactive non-obese diabetic CD8+ T cells by prostaglandin E2 receptor signaling. Int Immunol. 2000;12:851–60. doi: 10.1093/intimm/12.6.851. [DOI] [PubMed] [Google Scholar]
  • 17.Hilkens CM, Snijders A, Vermulen H, van der Meide PH, Wierenga EA, Kapsenberg ML. Accessory cell-derived IL-12 and prostaglandin E2 determine the IFN-γ level of activated human CD4+ T cells. J Immunol. 1996;156:1722–7. [PubMed] [Google Scholar]
  • 18.Bloom D, Jabrane-Ferrat N, Zeng L, et al. Prostaglandin E2 enhancement of interferon-γ production by antigen-stimulated type 1 helper T cells. Cell Immunol. 1999;194:21–7. doi: 10.1006/cimm.1999.1479. [DOI] [PubMed] [Google Scholar]
  • 19.Rieser C, Bock G, Klocker H, Bartsch G, Thurnher M. Prostaglandin E2 and tumor necrosis factor-α cooperate to activate human dendritic cells. synergistic activation of interleukin-12 production. J Exp Med. 1997;186:1603–8. doi: 10.1084/jem.186.9.1603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Sugimoto Y, Narumiya S, Ichikawa A. Distribution and function of prostanoid receptors: studies from knockout mice. Prog Lipid Res. 2000;39:289–314. doi: 10.1016/s0163-7827(00)00008-4. [DOI] [PubMed] [Google Scholar]
  • 21.Tsutsui H, Nakanishi K, Matsui K, Higashino K, Okamura H, Miyazawa Y, Kaneda K. IFN-γ-inducing factor up-regulates Fas ligand-mediated cytotoxic activity of murine natural killer cell clones. J Immunol. 1996;157:3967–73. [PubMed] [Google Scholar]
  • 22.Loh CL, Rotondo D, Dutta-Roy AK. Characterization of prostaglandin E2 receptors expressed on human monocytic leukaemia cell line, U937. Biochim Biophys Acta. 1993;1177:43–8. doi: 10.1016/0167-4889(93)90155-i. [DOI] [PubMed] [Google Scholar]
  • 23.Davidson J, Abul HT, Milton AS, Rotondo D. Cytokines and cytokine inducers stimulate prostaglandin E2 entry into the brain. Eur J Physiol. 2001;442:526–33. doi: 10.1007/s004240100572. [DOI] [PubMed] [Google Scholar]
  • 24.Samuelsson B, Granstrom E, Green K, Hamberg M, Hammerstrom S. Prostaglandins. Annu Rev Biochem. 1975;44:669–95. doi: 10.1146/annurev.bi.44.070175.003321. [DOI] [PubMed] [Google Scholar]
  • 25.Takahashi S, Takeuchi K, Okabe S. EP4 receptor mediation of prostaglandin E2-stimulated mucus secretion by rabbit gastric epithelial cells. Biochem Pharmacol. 1999;58:1997–2002. doi: 10.1016/s0006-2952(99)00286-5. [DOI] [PubMed] [Google Scholar]
  • 26.Arakawa T, Laneuville O, Miller CA, Lakkides KM, Wingerd BA, DeWitt DL, Smith WL. Prostanoid receptors of murine NIH 3T3 and RAW 264.7 cells. J Biol Chem. 1996;271:29569–75. doi: 10.1074/jbc.271.47.29569. [DOI] [PubMed] [Google Scholar]
  • 27.Nishigaki N, Negishi M, Sugimoto Y, Namba T, Narumiya S, Ichikawa A. Identification of prostaglandin E receptor ‘EP2’ cloned from mastocytoma cells as EP4 subtype. FEBS Lett. 1995;364:339–41. doi: 10.1016/0014-5793(95)00421-5. [DOI] [PubMed] [Google Scholar]
  • 28.Wu C-Y, Wang K, McDyer JF, Seder RA. Prostaglandin E2 and dexamethasone inhibit IL-12 receptor expression and IL-12 responsiveness. J Immunol. 1998;161:2723–30. [PubMed] [Google Scholar]
  • 29.Gorham JD, Guler ML, Fenoglio D, Gubler U, Murphy KM. Low dose TGF-β attenuates IL-12 responsiveness in murine Th cells. J Immunol. 1998;161:1664–70. [PubMed] [Google Scholar]
  • 30.Rogge L, Barberis-Maino L, Biffi M, Passini N, Presky D, Gubler U, Sinigaglia F. Selective expression of an interleukin-12 receptor component by human T helper 1 cells. J Exp Med. 1997;185:817–31. doi: 10.1084/jem.185.5.825. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Tilley SL, Coffman TM, Koller BH. Mixed messages. modulation of inflammation and immune responses by prostaglandins and thromboxanes. J Clin Invest. 2001;108:15–23. doi: 10.1172/JCI13416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Strassmann G, Patil-Koota V, Finkelman F, Fong M, Kambayashi T. Evidence for the involvement of interleukin 10 in differential deactivation of murine peritoneal macrophages by prostaglandin E2. J Exp Med. 1994;180:2365–70. doi: 10.1084/jem.180.6.2365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Cai G, Kastelein RA, Hunter CA. IL-10 enhances NK cell proliferation, cytotoxicity and production of IFN-γ when combined with IL-18. Eur J Immunol. 1999;29:2658–65. doi: 10.1002/(SICI)1521-4141(199909)29:09<2658::AID-IMMU2658>3.0.CO;2-G. [DOI] [PubMed] [Google Scholar]
  • 34.Katamura K, Sintaku N, Yamauchi Y, Fukui T, Ohshima Y, Mayumi M, Furusho K. Prostaglandin E2 at priming of naive CD4+ T cells inhibits acquisition of ability to produce IFN-γ and IL-2, but not IL-4 and IL-5. J Immunol. 1995;155:4604–12. [PubMed] [Google Scholar]
  • 35.Kaur K, Harris SG, Padilla J, Graf BA, Phipps RP. ProstaglandinE2 as a modulator of lymphocyte-mediated inflammatory and humoral responses. Adv Exp Med Biol. 1999;469:409–11. doi: 10.1007/978-1-4615-4793-8_59. [DOI] [PubMed] [Google Scholar]
  • 36.Kuroda E, Sugiura T, Zeki K, Yoshida Y, Yamashita U. Sensitivity difference to the suppressive effect of prostaglandin E2 among mouse strains: a possible mechanism to polarize Th2 type response in BALB/c mice. J Immunol. 2000;164:2386–95. doi: 10.4049/jimmunol.164.5.2386. [DOI] [PubMed] [Google Scholar]
  • 37.Fedyk ER, Phipps RP. Prostaglandin E2 receptors of the EP2 and EP4 subtypes regulate activation and differentiation of mouse B lymphocytes to IgE-secreting cells. Proc Natl Acad Sci USA. 1996;93:10978–83. doi: 10.1073/pnas.93.20.10978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Nataraj C, Thomas DW, Tilley SL, Nyguen M, Mannon R, Koller BH, Coffman TM. Receptors for prostaglandin E2 that regulate cellular immune responses in the mouse. J Clin Invest. 2001;108:1229–35. doi: 10.1172/JCI13640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Chakir H, Camilucci AA, Filion LG, Webb JR. Differentiation of murine NK cells into distinct subsets based on variable expression of the IL-12Rβ2 subunit. J Immunol. 2000;165:4985–93. doi: 10.4049/jimmunol.165.9.4985. [DOI] [PubMed] [Google Scholar]
  • 40.Wu C-Y, Gadina M, Wang K, O'Shea J, Seder RA. Cytokine regulation of IL-12 receptor β2 expression: differential effects on human T and NK cells. Eur J Immunol. 2000;30:1364–74. doi: 10.1002/(SICI)1521-4141(200005)30:5<1364::AID-IMMU1364>3.0.CO;2-U. [DOI] [PubMed] [Google Scholar]
  • 41.Joshi PC, Zhou X, Cuchens M, Jones Q. Prostaglandin E2 suppressed IL-15-mediated human NK cell function through down-regulation of common γ-chain. J Immunol. 2001;166:885–91. doi: 10.4049/jimmunol.166.2.885. [DOI] [PubMed] [Google Scholar]
  • 42.Murphy KM, Ouyang W, Farrar JD, Yang J, Ranganath S, Asnagli H, Afkarian M, Murphy TL. Signaling and transcription in T helper development. Annu Rev Immunol. 2000;18:451–94. doi: 10.1146/annurev.immunol.18.1.451. [DOI] [PubMed] [Google Scholar]
  • 43.D'Acquisto F, Sautebin L, Iuvone T, Di Rosa M, Carnuccio R. Prostaglandins prevent inducible nitric oxide synthase protein expression by inhibiting nuclear factor-κB activation in J774 macrophages. FEBS Lett. 1998;440:76–80. doi: 10.1016/s0014-5793(98)01407-0. [DOI] [PubMed] [Google Scholar]
  • 44.Kolenko V, Rayman P, Roy B, et al. Downregulation of JAK3 protein levels in T lymphocytes by prostaglandin E2 and other cyclic adenosine monophosphate-elevating agents: impact on interleukin-2 receptor signaling pathway. Blood. 1999;93:2308–18. [PubMed] [Google Scholar]
  • 45.Boer A-K, Drayer L, Vellenga E. Prostaglandin-E2 enhances EPO-mediated STAT5 transcriptional activity by serine phosphorylation of CREB. Blood. 2002;100:467–73. doi: 10.1182/blood.v100.2.467. [DOI] [PubMed] [Google Scholar]
  • 46.Penix LA, Sweetser MT, Weaver WM, Hoeffler JP, Kerppola TK, Wilson CB. The proximal regulatory element of the interferon-γ promoter mediates selective expression in T cells. J Biol Chem. 1996;271:31964–72. doi: 10.1074/jbc.271.50.31964. [DOI] [PubMed] [Google Scholar]
  • 47.Zhang F, Wang D, Boothby M, Penix L, Flavell RA, Aune TM. Regulation of the activity of IFN-γ promoter elements during Th cell differentiation. J Immunol. 1998;161:6105–12. [PubMed] [Google Scholar]

Articles from Immunology are provided here courtesy of British Society for Immunology

RESOURCES