15-Deoxy-Δ12,14-PGJ2 (dPGJ2) is a bioactive metabolite of the J2 series that has been identified as a ligand for peroxisome proliferator-activated receptor γ (PPARγ) and has received attention for its potential antiinflammatory effects. Because neutrophils express cell-surface receptors for PGs, the effect of dPGJ2 was tested on an inflammatory response that should not require PPARγ, the oxidative burst made by adherent human neutrophils. dPGJ2 inhibited adhesion-dependent H2O2 production with an IC50 of 1.5 μM when neutrophils were stimulated with TNF, N-formylnorleucylleucylphenylalanine, or LPS. Inhibition by dPGJ2 occurred during the lag phase, before generation of peroxide, suggesting blockade of an early signaling step. Indeed, dPGJ2 blocked adhesion of neutrophils to fibrinogen in response to TNF or LPS with an IC50 of 3–5 μM. dPGJ2 was more potent at inhibiting the adhesion-dependent oxidative burst than several other PGs tested. Further, dPGJ2 did not appear to act through either the DP receptor or receptors for PGE2. PG receptors modulate cAMP levels, and the inhibition of adhesion and oxidative burst by dPGJ2 was enhanced in the presence of 3-isobutyl-1-methylxanthine, a cAMP phosphodiesterase inhibitor. A potent PPARγ agonist (AD-5075) did not inhibit peroxide production or adhesion, nor did it change the IC50 for dPGJ2 inhibition. These studies suggest that dPGJ2 may interact with an unknown receptor on neutrophils, distinct from PPARγ, to modulate the production of reactive oxygen intermediates.

Recently, much attention has been focused on a novel biologically active PG of the J2 series, 15-deoxy-Δ12,14-PGJ2 (dPGJ2).2 dPGJ2 has been shown to be a natural ligand for peroxisome proliferator-activated receptor γ (PPARγ), a nuclear receptor implicated in adipocyte differentiation and glucose homeostasis (1, 2, 3). Besides being highly expressed in adipocytes, PPARγ is also present in hemopoietic cells including human neutrophils (4). However, little is known about PPARγ and dPGJ2 with respect to neutrophil functions. Two recent studies (5, 6) document the role of dPGJ2 and troglitazone, a potent PPARγ ligand, as antiinflammatory agents. These studies showed that agonists of PPARγ inhibited the expression of inflammatory cytokines, inducible NO synthase, gelatinase B, and scavenger receptor A in activated macrophages. These observations raise the possibility that synthetic PPARγ ligands may be of therapeutic value in inflammatory diseases such as atherosclerosis and rheumatoid arthritis.

In addition to PPARγ, which may bind and mediate responses to dPGJ2, human neutrophils also express G-protein coupled receptors that initiate responses to prostaglandins. Wheeldon and Vardey (7) have shown that the PG EP2 and DP receptors are present in stimulated human neutrophils. Ligation of either one of these receptors leads to inhibition of superoxide production in response to fMLP. The EP2 and DP receptors are linked to Gαs and act by elevating cAMP.

TNF-stimulated neutrophils exhibit well-characterized β2 integrin-mediated adhesion and produce massive amounts of reactive oxygen intermediates upon adhesion (8, 9). The adhesion-dependent production of an oxidative burst relies on the function of the β2 integrins and can be blocked by mAbs against the β2 chain (9). Pharmacological elevation of cAMP can also arrest this inflammatory response by impeding the assembly of actin filaments, cell spreading, and secretory responses in TNF-treated neutrophils (10). PGE2 and the cAMP-phosphodiesterase inhibitor, RO 20-1724, cooperatively decrease the secretory response of adherent neutrophils (11). Elevation of cAMP and consequent activation of the cAMP-dependent protein kinase A leads to phosphorylation of Rap 1A. In neutrophils, Rap 1A is a major substrate for protein kinase A, and phosphorylation of Rap 1A in the presence of elevated cAMP may down-regulate its association with the cytochrome b558 and activation of the NADPH oxidase (12). Further, Rho A activation and β2 integrin-dependent leukocyte adhesion induced by chemoattractants is also inhibited by a rise in intracellular cAMP (13). Thus, PGs may act via two distinct pathways to reduce the production of reactive oxygen intermediates by adherent neutrophils.

Here we show that dPGJ2 inhibits the adhesion-dependent oxidative burst by blocking an early step in the signaling pathway. The inhibition was apparently not mediated via PPARγ, because the inhibitory potency of dPGJ2 was not affected by AD-5075, a potent PPARγ agonist. Moreover the inhibitory potency of dPGJ2 increased significantly in the presence of a cAMP phosphodiesterase inhibitor, suggesting that the effect may be mediated through cAMP (10, 11, 12, 13). Taken together, these observations suggest that dPGJ2 may interact with a cell-surface receptor as has been observed with conventional PGs.

N-Formylnorleucylleucylphenylalanine (fNLLP), PMA, cycloheximide, and 3-isobutyl-1-methylxanthine (IBMX) were purchased from Sigma (St. Louis, MO). Recombinant human TNFα was obtained from Genzyme (Cambridge, MA). LPS Ra (R60) from Salmonella minnesota was obtained from List Biologicals (Campbell, CA). Complexes of monomeric LPS and recombinant soluble human CD14 (sCD14) were made as described earlier (14). Briefly, 5 μg/ml LPS and 50 μg/ml sCD14 were incubated overnight at 37°C. LPS forms stoichiometric complexes with monomeric sCD14 under these conditions, and the complexes effectively stimulate leukocytes. FBS was obtained from HyClone Systems (Logan, UT). Human fibrinogen was obtained from Calbiochem-Novabiochem (La Jolla, CA). AD-5075, a potent thiazolidinedione PPARγ agonist (15), and L-644,698, a potent PGD2 receptor agonist (16), were made at Merck (Rahway, NJ). All other chemicals were bought from either Sigma or Fisher Scientific (Pittsburgh, PA).

dPGJ2 and all other PGs used were purchased from Biomol Research Laboratories (Plymouth Meeting, PA). PGs were made as 50 mM stocks in DMSO and stored under nitrogen at −20°C in aliquots.

Neutrophils were prepared from human venous blood freshly drawn into heparinized syringes and separated on neutrophil isolation medium (Cardinal Associates, Santa Fe, NM) exactly as previously described (17). Contaminating erythrocytes were removed by hypotonic lysis. The neutrophils were suspended either in Dulbecco’s PBS with 0.5 mg/ml human serum albumin, 0.3 U/ml aprotinin, and 3 mM glucose or Krebs-Ringer phosphate buffer with 5.5 mM glucose.

The production of H2O2 by adherent neutrophils was measured using the previously published assay of HRP-catalyzed oxidation of scopoletin (18). Polystyrene Primaria 96-well tissue culture plates (Becton Dickinson, Lincoln Park, NJ) were coated with FBS for 30 min at 37°C. Each well was washed vigorously three times by forceful squirting of 3 ml 0.9% NaCl to prevent interference with adhesion by soluble BSA (19). To each well was added 50 μl scopoletin (39 μM in Krebs-Ringer phosphate buffer with 5.5 mM glucose), 10 μl HRP (0.5 U/ml), 15 μl NaN3 (1 mM), 10 μl of PG or vehicle, and 3 × 104 neutrophils. The reaction was started by adding agonist (TNF, fNLLP, LPS/sCD14 complexes, or PMA) at the indicated concentrations, giving a final assay volume of 125 μl. The scopoletin fluorescence was measured every 10 min (excitation, 360 nm; emission, 460 nm) in a Cytofluor 4000 fluorescence plate reader equipped with a temperature control device (PE Biosystems, Framingham, MA) to maintain the plates at 37°C throughout the assay (20). Samples were run in quadruplicate, and the data are presented as the mean ± SEM for a representative experiment of at least three performed with the same results. In some cases, the error bars are smaller than the symbols used. The nmol H2O2 produced per 30,000 neutrophils were calculated as described (18). Unless otherwise stated, percent inhibition of H2O2 production was calculated as 100 − [(nmol H2O2 from PG- and TNF-treated neutrophils − nmol H2O2 from buffer-treated neutrophils)/(nmol H2O2 from TNF-treated neutrophils − nmol H2O2 from buffer-treated neutrophils) × 100], using values taken at 60 min.

Adhesion of neutrophils to fibrinogen-coated Terasaki plates was performed as previously described (21). Fibrinogen is a known ligand for CD11b/CD18 (8, 22), and neutrophil adhesion to this substrate is therefore dependent on the β2 integrin. Briefly, 5×106/ml neutrophils were fluorescently labeled with 15 μM carboxyfluorescein diacetate succiminidyl ester (Molecular Probes, Eugene, OR) for 10 min at room temperature, and 104 cells were added to each fibrinogen-coated Terasaki well. The neutrophils were preincubated with or without dPGJ2 for 10 min at 37°C before the addition of an agonist and further incubation for 30 min at 37°C. Adhesion was quantitated by measuring the fluorescence in each well before and after washing in a Cytofluor 2300 (PerSeptive Biosystems, Framingham, MA). Percent adhesion was calculated as (fluorescence after washing/fluorescence before washing) × 100. Samples for each condition were run in triplicate, and the data are presented as the mean ± SEM. Experiments representative of at least two repetitions with similar results are shown.

To test whether dPGJ2 has antiinflammatory activity with respect to neutrophil function, we examined the effect of dPGJ2 on the β2 integrin-dependent production of reactive oxygen intermediates. Neutrophils were exposed to increasing concentrations of dPGJ2 before the addition of one of several agonists that stimulate neutrophil adhesion and a subsequent oxidative burst, TNF, fNLLP, or LPS/sCD14 complexes. In all cases, dPGJ2 blocked the production of H2O2 by adherent neutrophils in a dose-dependent manner with an IC50 between 0.6 and 1.5 μM, depending on the donor used as the source of cells (Fig. 1). In contrast, when neutrophils were stimulated with PMA, an agonist that directly stimulates adhesion-independent assembly of the oxidase complex, dPGJ2 did not block the oxidative burst up to a concentration of 3 μM. Thirty to 50% inhibition of the response to PMA was observed with 10 μM dPGJ2, suggesting that, at high concentrations, dPGJ2 may directly inhibit assembly of the oxidase. Because the adhesion-dependent production of reactive oxygen intermediates takes up to 60 min and exhibits a considerable lag phase before H2O2 release commences, it was possible to further define where in the oxidative burst cascade dPGJ2 exerted its effect.

FIGURE 1.

dPGJ2 inhibits the adhesion-dependent oxidative burst. Neutrophils were incubated at 37°C in serum-coated wells with increasing concentrations (⋄, 0.3 μM; ○, 1.0 μM; ▵, 3 μM; □, 10 μM) of dPGJ2 for 10 min before the addition of agonist, TNF (100 ng/ml), LPS/sCD14 complexes (30 ng/ml LPS), fNLLP (6 × 10−6 M), or PMA (30 ng/ml). • and ♦ are the responses with agonist alone and buffer, respectively. H2O2 production was measured over time as described in Materials and Methods.

FIGURE 1.

dPGJ2 inhibits the adhesion-dependent oxidative burst. Neutrophils were incubated at 37°C in serum-coated wells with increasing concentrations (⋄, 0.3 μM; ○, 1.0 μM; ▵, 3 μM; □, 10 μM) of dPGJ2 for 10 min before the addition of agonist, TNF (100 ng/ml), LPS/sCD14 complexes (30 ng/ml LPS), fNLLP (6 × 10−6 M), or PMA (30 ng/ml). • and ♦ are the responses with agonist alone and buffer, respectively. H2O2 production was measured over time as described in Materials and Methods.

Close modal

After the addition of agonist, there is a characteristic lag period of 10–40 min before H2O2 production, during which neutrophils adhere to the substrate, spread, and assemble their NADPH oxidase complex (8). While the length of the lag time can vary considerably between preparations of neutrophils, it is nonetheless a consistent feature of the adhesion-dependent oxidative burst. To determine at which point between the addition of agonist and the production of H2O2 the inhibition by dPGJ2 occurred, dPGJ2 (2 μM) was added at the same time as TNF (0 min), or at intervals (15, 25, and 50 min) after the addition of the agonist, and H2O2 production was measured. dPGJ2 inhibited the oxidative burst only when added before the onset of H2O2 production. Fig. 2 shows an example in which the lag period was 30 min: dPGJ2 was only effective when added up to 25 min after the agonist. By 50 min after addition of the agonist, the inhibitory effect of dPGJ2 was completely lost. These results indicate that dPGJ2 may inhibit an early stage in the signaling cascade. Moreover, dPGJ2 does not directly affect the activity of the assembled NADPH oxidase.

FIGURE 2.

dPGJ2 inhibits the oxidative burst only when added during the lag phase. dPGJ2 (2 μM) was added to neutrophils with 100 ng/ml TNF (0 min) or at intervals (indicated by arrows) up to 50 min after the addition of TNF. • and ♦ are the responses with TNF alone and buffer, respectively. H2O2 produced was measured at 10-min intervals over time.

FIGURE 2.

dPGJ2 inhibits the oxidative burst only when added during the lag phase. dPGJ2 (2 μM) was added to neutrophils with 100 ng/ml TNF (0 min) or at intervals (indicated by arrows) up to 50 min after the addition of TNF. • and ♦ are the responses with TNF alone and buffer, respectively. H2O2 produced was measured at 10-min intervals over time.

Close modal

We tested whether dPGJ2 blocked one of the earliest events in the adhesion-dependent oxidative burst, adhesion itself. Neutrophils were incubated with increasing concentrations of dPGJ2, then stimulated to adhere to fibrinogen with one of several agonists. dPGJ2 blocked adhesion in response to TNF, LPS/sCD14 complexes, or fNLLP with an IC50 of 3 μM-5 μM (Fig. 3), although complete inhibition of adhesion was not achieved at higher concentrations in the case of fNLLP. dPGJ2 did not inhibit PMA-induced adhesion at concentrations up to 10 μM, indicating that the PG was not toxic to the cells. dPGJ2 had no effect on the increase in expression of CD11b/CD18 on the surface of neutrophils in response to agonists or on the release of L-selectin, as measured by flow cytometery (data not shown). Therefore, it is likely that dPGJ2 affected not receptor expression but the increase in avidity of CD11b/CD18 for fibrinogen that is observed upon agonist stimulation of neutrophils. Thus, dPGJ2 may act to block production of reactive oxygen intermediates by adhesive cells by inhibiting neutrophil adhesion. Further studies were designed to explore the mechanism of inhibition by dPGJ2.

FIGURE 3.

dPGJ2 blocks neutrophil adhesion to fibrinogen. Increasing concentrations of dPGJ2 were added to neutrophils for 10 min at 37°C before the addition of agonists, TNF (10 ng/ml), LPS/sCD14 (10 ng/ml), fNLLP (5 × 10−8 M), or PMA (30 ng/ml). Adhesion was measured after a further 30 min at 37°C as described in Materials and Methods.

FIGURE 3.

dPGJ2 blocks neutrophil adhesion to fibrinogen. Increasing concentrations of dPGJ2 were added to neutrophils for 10 min at 37°C before the addition of agonists, TNF (10 ng/ml), LPS/sCD14 (10 ng/ml), fNLLP (5 × 10−8 M), or PMA (30 ng/ml). Adhesion was measured after a further 30 min at 37°C as described in Materials and Methods.

Close modal

To determine whether the inhibitory effects of dPGJ2 were specific, we tested a panel of different PGs for their ability to inhibit the adhesion-dependent oxidative burst. Neutrophils were preincubated with PGA1, PGB1, PGE1, PGE2, PGD2, PGF, PGJ2, or dPGJ2 before the addition of TNF and measurement of H2O2 production. Among the PGs tested, only dPGJ2 reached nearly complete inhibition at 3 μM (Table I). PGA1, PGB1, PGE1, and PGJ2 all showed modest inhibition (20–25%) at 3 μM, and there was no inhibition with PGD2 or PGF. In additional experiments using PGs at 10 μM, PGD2 and PGF still showed no inhibition, while PGJ2 inhibited the oxidative burst about 35% (data not shown). Thus, dPGJ2 was the most potent of the PGs for inhibition of the adhesion-dependent oxidative burst.

Table I.

dPGJ2 is more potent than other PGs as an inhibitor of the oxidative bursta

PG% Inhibition
PGA1 25 ± 5 
PGB1 20 ± 5 
PGE1 20 ± 5 
PGE2 5 ± 2 
PGF 
PGD2 
PGJ2 20 ± 5 
dPGJ2 95 ± 5 
PG% Inhibition
PGA1 25 ± 5 
PGB1 20 ± 5 
PGE1 20 ± 5 
PGE2 5 ± 2 
PGF 
PGD2 
PGJ2 20 ± 5 
dPGJ2 95 ± 5 
a

Neutrophils were preincubated for 10 min with 3 μM of each PG, then 100 ng/ml TNF was added. The nmol of H2O2 generated were measured at 90 min. Percent inhibition of H2O2 production was calculated as 100 − [(nmol H2O2 from PG- and TNF-treated neutrophils − nmol H2O2 from buffer-treated neutrophils)/(nmol H2O2 from TNF-treated neutrophils − nmol H2O2 from buffer-treated neutrophils) × 100].

Although PGD2 did not block the adhesion-dependent oxidative burst (Table I), recent evidence suggests that dPGJ2 binds to the PGD2 receptor (DP receptor) with a Ki of 300 nM (16). Therefore, we used L-644,698, a novel, potent, synthetic DP receptor agonist (Ki of 0.6 ± 0.2 nM) (16), to block the DP receptor and tested the inhibitory ability of dPGJ2 in the presence of the compound. L-644,698 at 5 μM neither blocked adhesion-dependent H2O2 production in response to TNF nor affected the ability of dPGJ2 to inhibit H2O2 production (Fig. 4), suggesting that dPGJ2 does not act via the DP receptor. These results also confirmed the lack of inhibition when the DP receptor was ligated by PGD2.

FIGURE 4.

dPGJ2 is unlikely to mediate its inhibitory effect through the DP receptor. Neutrophils were preincubated at 37°C with or without the DP receptor agonist L-644,698 at 5 μM in the presence or absence of dPGJ2 at 1 μM. TNF (100 ng/ml) was added for a further 60-min incubation at 37°C, and the nanomoles of H2O2 generated were measured at the end of this time.

FIGURE 4.

dPGJ2 is unlikely to mediate its inhibitory effect through the DP receptor. Neutrophils were preincubated at 37°C with or without the DP receptor agonist L-644,698 at 5 μM in the presence or absence of dPGJ2 at 1 μM. TNF (100 ng/ml) was added for a further 60-min incubation at 37°C, and the nanomoles of H2O2 generated were measured at the end of this time.

Close modal

Several additional PG receptors have been characterized and are potential candidates for mediating the effects of dPGJ2. EP1, EP2, and EP3 receptors each bind PGE2 with Kd of 3–21 nM (23). To test the potential contribution of these receptors to the effects of dPGJ2, we measured the ability of dPGJ2 to block the adhesion-dependent oxidative burst in the presence of increasing concentrations of PGE2. Inhibition of the adhesion-dependent oxidative burst by 1 μM dPGJ2 remained unchanged when neutrophils were pretreated with a 5-fold higher concentration of PGE2 (data not shown). These studies suggest that the receptor mediating the effect of dPGJ2 on the adhesion-dependent oxidative burst is distinct from the high-affinity PGE2 receptors previously characterized.

Because PGs may act to raise intracellular cAMP, we tested whether the inhibitory effect of dPGJ2 on the adhesion-dependent oxidative burst was affected by the presence of IBMX, a nonspecific cAMP phosphodiesterase inhibitor. IBMX alone at 50 μM had little effect on the production of H2O2 in response to TNF (Fig. 5). However, the IC50 for inhibition of H2O2 production by dPGJ2 dropped from about 1 μM without IBMX to 0.15 μM with IBMX (Fig. 5). The nearly 10-fold decrease in IC50 suggests that dPGJ2 may act by elevating cAMP. A similar effect was also seen on neutrophil adhesion, in which the IC50 also decreased for dPGJ2 in the presence of IBMX (Fig. 6). Thus, blockade of cAMP-phosphodiesterase synergized with dPGJ2 to inhibit the adhesion-dependent burst, suggesting that dPGJ2 may act on neutrophil functions by elevating cAMP.

FIGURE 5.

Raising cAMP enhances inhibition of the adhesion-dependent oxidative burst by dPGJ2. Increasing concentrations of dPGJ2 were added to neutrophils in the presence or absence of 50 μM IBMX for 10 min at 37°C. TNF (100 ng/ml) was then added, and H2O2 produced was measured after a further 60 min at 37°C.

FIGURE 5.

Raising cAMP enhances inhibition of the adhesion-dependent oxidative burst by dPGJ2. Increasing concentrations of dPGJ2 were added to neutrophils in the presence or absence of 50 μM IBMX for 10 min at 37°C. TNF (100 ng/ml) was then added, and H2O2 produced was measured after a further 60 min at 37°C.

Close modal
FIGURE 6.

Raising cAMP enhances the blockade of adhesion by dPGJ2. Neutrophils were preincubated for 10 min at 37°C with or without 50 μM IBMX and increasing concentrations of dPGJ2. TNF (10 ng/ml) was added, and adhesion was measured after a further 30 min at 37°C.

FIGURE 6.

Raising cAMP enhances the blockade of adhesion by dPGJ2. Neutrophils were preincubated for 10 min at 37°C with or without 50 μM IBMX and increasing concentrations of dPGJ2. TNF (10 ng/ml) was added, and adhesion was measured after a further 30 min at 37°C.

Close modal

Although the production of H2O2 by adherent neutrophils has not been demonstrated to require protein synthesis, we nonetheless sought to rule out the possibility that dPGJ2 may exert its effects on the adhesion-dependent oxidative burst through PPARγ expressed in neutrophils (4). The compound AD-5075 binds PPARγ with a Ki of 22 nM (15), inducing strong transcriptional activation, while dPGJ2 binds to PPARγ in the high μM range (1, 2). Despite strong PPARγ agonist activity, 1 μM AD-5075 had no inhibitory effect on H2O2 production in response to TNF (Fig. 7). In further studies, the inhibitory effect of dPGJ2 was measured in the presence and absence of 1 μM AD-5075, a concentration that should displace >90% of dPGJ2 binding to PPARγ, even at the highest concentration of dPGJ2 used (3 μM). In the presence of 1 μM AD-5075, there was no change in the IC50 value of dPGJ2 (1 μM) (Fig. 7). We further confirmed that inhibition of the oxidative burst by dPGJ2 was not due to the synthesis of new proteins, because cycloheximide (25 μM) had no effect on the inhibition by dPGJ2 (data not shown). These observations suggest that dPGJ2 exerts its biological effects on neutrophils through receptors other than PPARγ.

FIGURE 7.

dPGJ2 does not inhibit the oxidative burst via PPARγ. Neutrophils were preincubated with (open symbols) or without (closed symbols) 1 μM AD-5075, a potent PPARγ agonist, in the presence of increasing concentrations of dPGJ2 (○/•, 0 μM; ▿/▾, 0.5 μM; ▵/▴, 1 μM; □/▪, 3 μM) for 10 min at 37°C. • and ♦ are the responses with TNF alone and buffer, respectively. TNF (100 ng/ml) was then added, and the production of H2O2 was measured at 10-min intervals over time.

FIGURE 7.

dPGJ2 does not inhibit the oxidative burst via PPARγ. Neutrophils were preincubated with (open symbols) or without (closed symbols) 1 μM AD-5075, a potent PPARγ agonist, in the presence of increasing concentrations of dPGJ2 (○/•, 0 μM; ▿/▾, 0.5 μM; ▵/▴, 1 μM; □/▪, 3 μM) for 10 min at 37°C. • and ♦ are the responses with TNF alone and buffer, respectively. TNF (100 ng/ml) was then added, and the production of H2O2 was measured at 10-min intervals over time.

Close modal

Following the discovery that dPGJ2 is a ligand for PPARγ (1, 2), a large amount of work has focused on the potential significance of this interaction and the potential for PPARγ to be a mediator of inflammatory responses. dPGJ2 has been shown to block production of cytokines and influence the expression of surface markers on monocytes and macrophages (5, 6). These studies have employed dPGJ2 as a specific ligand for PPARγ and have interpreted the effects as signatures of PPARγ function. Here we demonstrate that dPGJ2 may have profound effects on leukocytes by means unrelated to PPARγ. Several lines of evidence indicate that dPGJ2 blocks adherence and the adhesion-dependent oxidative burst in neutrophils by mechanisms that do not depend on PPARγ. First, the blockade of the oxidative burst does not require protein synthesis, as would be expected of an effect mediated by a transcription factor. Second, the effects of dPGJ2 are not mimicked by AD-5075, a PPARγ agonist that is far more potent than dPGJ2. Third, the effects of dPGJ2 are not overcome by concentrations of AD-5075 calculated to displace nearly all of the dPGJ2 from PPARγ (Fig. 7). We conclude that the effects of dPGJ2 in neutrophils are mediated by a receptor distinct from PPARγ.

The observation that dPGJ2 may act through a receptor other than PPARγ calls into question conclusions about the function of PPARγ based on experiments using dPGJ2 (5, 6, 24, 25). A critical question in this regard is the range of expression of the putative second receptor for dPGJ2. Recent studies suggest that such an activity is present not only in neutrophils but also in monocytes and macrophages.3

The identity of the receptor mediating the effects of dPGJ2 on neutrophils is not defined in this study, but several lines of evidence are consistent with the hypothesis that dPGJ2 may interact with a receptor linked to elevation of cAMP. Conventional cyclopentane PGs act on G protein-coupled cell-surface receptors to exert their effects. It is clear that elevation of cAMP would have effects comparable to those of dPGJ2. Ottonello et al. (11) have reported that in human neutrophils, TNF-triggered superoxide production is reduced by 10 μM PGE2, and the inhibition is significantly enhanced with a phophodiesterase inhibitor. This activity could be mimicked by forskolin, a direct activator of adenylate cyclase. Nathan and Sanchez (10) have observed that cAMP-raising drugs block the adhesion-dependent oxidative burst and that, as with dPGJ2 (Fig. 2), inhibition of peroxide generation occurs only when the cAMP-elevating drugs are added before actin reorganization. In keeping with a role for cyclase in responses to dPGJ2, we observed a striking potentiation of the inhibitory effect of dPGJ2 with the phosphodiesterase inhibitor IBMX (Fig. 5).

We considered the possibility that dPGJ2 is a ligand for the human DP receptor (16). Armstrong and Talpain (26) have reported that neutrophils express receptors for PGD2, and Ney and Schror (27) have reported that PGD2 and its analogues are potent inhibitors of fMLP- and platelet-activating factor-induced superoxide anion generation, β-glucuronidase release, and Ca2+ influx. This inhibition of neutrophil function was paralleled by an increase in cellular cAMP levels, thus suggesting an inhibitory PGD2 receptor on the human neutrophil. In our experiments, freshly made PGD2 solutions had no effect on the oxidative burst in response to TNF, and the potent, selective DP receptor agonist, L-644,698, at a saturating concentration neither inhibited peroxide production by itself nor affected inhibition of the oxidative burst by dPGJ2. The above results argue that PGD2 receptors, if present on neutrophils, do not take part in the down-regulation of the adhesion-dependent oxidative burst and that the action of dPGJ2 is not mediated through them.

Several lines of evidence suggest that dPGJ2 may block the production of reactive oxygen intermediates by adherent cells by blocking integrin-dependent adhesion. dPGJ2 diminished integrin-dependent adhesion, and it has been previously shown that blockade of leukocyte integrin-ligand interactions by Abs abrogates the production of peroxide by neutrophils in response to TNF (9). It is likely that dPGJ2 was more effective in blocking the oxidative burst than adhesion in response to TNF, LPS/sCD14, or fNLLP, because the substrates used for the assays were different. Serum offers fewer ligand-binding sites for β2 integrins than does fibrinogen, and adhesion to serum would thus be easier to inhibit.

To effectively inhibit the oxidative burst, dPGJ2 had to be present during the lag period before production of H2O2, during which adhesion, cell spreading, and assembly of the oxidase components occurs. dPGJ2 may block neutrophil adhesion and spreading through cAMP-dependent phosphorylation of Rho A (13). Elevated levels of cAMP also lead to phosphorylation of Rap 1 (12), with the potential for direct down-regulation of the NADPH oxidase. Thus, while dPGJ2 inhibits adhesion, it may also act through multiple mechanisms to block the oxidative burst. Because the burst is downstream of adhesion and spreading, the effect of dPGJ2 on the burst would be more potent than its effect on adhesion alone.

We conclude that dPGJ2, a known PPARγ agonist, may have a receptor other than PPARγ and that this receptor may act through cAMP or another mechanism yet to be described. The nature of the receptor for dPGJ2 is currently under investigation.

2

Abbreviations used in this paper: dPGJ2, 15-deoxy-Δ12,14-PGJ2; PPARγ, peroxisome proliferator-activated receptor γ; fNLLP, N-formylnorleucylleucylphenylalanine; IBMX, 3-isobutyl-1-methylxanthine; sCD14, soluble CD14.

3

Thieringer, R., J. E. Fenyk-Melody, C. B. LeGrand, B. A. Shelton, P. A. Detmers, E. P. Somers, L. Carbin, D. E. Moller, S. D. Wright, and J. Berger. 1999. Activation of PPARγ does not inhibit macrophage activation in vitro or in vivo. Submitted for publication.

1
Forman, B. M., P. Tontonoz, J. Chen, R. P. Brun, B. M. Spiegelman, R. M. Evans.
1995
. 15-Deoxy-Δ12,14-prostaglandin J2 is a ligand for the adipocyte determination factor PPARγ.
Cell
83
:
803
2
Kliewer, S. A., J. M. Lenhard, T. M. Willson, I. Patel, D. C. Morris, J. M. Lehmann.
1995
. A prostaglandin J2 metabolite binds peroxisome proliferator-activated receptor γ and promotes adipocyte differentiation.
Cell
83
:
813
3
Kliewer, S. A., S. S. Sundseth, S. A. Jones, P. J. Browns, G. B. Wisely, C. S. Koble, P. Devchand, W. Wahli, T. M. Willson, J. M. Lenhard, J. M. Lehmann.
1997
. Fatty acids and eicosanoids regulate gene expression through direct interactions with peroxisome proliferator-activated receptor α and γ.
Proc. Natl. Acad. Sci. USA
94
:
4318
4
Greene, M. E., B. Blumberg, O. W. McBride, H. F. Yi, K. Kronquist, K. Kwan, L. Hsieh, G. Greene, S. D. Nimer.
1995
. Isolation of the human peroxisome proliferator activated receptor γ cDNA: expression in hematopoietic cells and chromosomal mapping.
Gene Expression
4
:
281
5
Ricote, M., A. C. Li, T. M. Willson, C. J. Kelly, C. K. Glass.
1998
. The peroxisome proliferator-activated receptor-γ is a negative regulator of macrophage activation.
Nature
391
:
79
6
Jiang, C., A. Ting, B. Seed.
1998
. PPARγ agonists inhibit production of monocyte inflammatory cytokines.
Nature
391
:
82
7
Wheeldon, A., C. J. Vardey.
1993
. Characterization of the inhibitory prostanoid receptors on human neutrophils.
Br. J. Pharmacol.
108
:
1051
8
Nathan, C. F..
1987
. Neutrophil activation on biological surfaces: massive secretion of hydrogen peroxide in response to products of macrophages and lymphocytes.
J. Clin. Invest.
80
:
1550
9
Nathan, C., S. Srimal, C. Farber, E. Sanchez, L. Kabbash, A. Asch, J. Gailit, S. D. Wright.
1989
. Cytokine-induced respiratory burst of human neutrophils: dependence on extracellular matrix proteins and CD11/CD18 integrins.
J. Cell Biol.
109
:
1341
10
Nathan, C., E. Sanchez.
1990
. Tumor necrosis factor and CD11/CD18 (β2) integrins act synergistically to lower cAMP in human neutrophils.
J. Cell Biol.
111
:
2171
11
Ottonello, L., M. P. Morone, P. Dapino, F. Dallegri.
1995
. Tumour necrosis factor α-induced oxidative burst in neutrophils adherent to fibronectin: effects of cyclic AMP-elevating agents.
Br. J. Haem.
91
:
566
12
Quilliam, L. A., H. Mueller, B. P. Bohl, V. Prossinitz, L. A. Sklar, C. H. Der, G. M. Bokoch.
1991
. Rap 1 is a substrate for cyclic AMP-dependent protein kinase in human neutrophils.
J. Immunol.
147
:
1628
13
Laudanna, C., J. J. Campbell, E. C. Butcher.
1997
. Elevation of intracellular cAMP inhibits Rho A activation and integrin-dependent leukocyte adhesion induced by chemoattractants.
J. Biol. Chem.
272
:
24141
14
Hailman, E., T. Vasselon, M. Kelly, L. A. Busse, M. C.-T. Hu, H. S. Lichenstein, P. A. Detmers, S. D. Wright.
1996
. Stimulation of macrophages and neutrophils by complexes of lipopolysaccharide and soluble CD14.
J. Immunol.
156
:
4384
15
Berger, J., P. Bailey, C. Biswas, C. A. Cullinan, T. W. Doebber, N. S. Hayes, R. Saperstein, R. G. Smith, M. D. Leibowitz.
1996
. Thiazolidinediones produce a conformational change in peroxisome-activated receptor-γ: binding and activation correlate with antidiabetic actions in db/db mice.
Endocrinology
137
:
4189
16
Wright, D. H., K. M. Metters, M. Abramovitz, A. W. Ford-Hutchinson.
1998
. Characterization of the recombinant human prostanoid DP receptor and identification of the L-644,698, a novel selective DP agonist.
Br. J. Pharmacol.
123
:
1317
17
Detmers, P. A., D. E. Powell, A. Walz, I. Clark-Lewis, M. Baggiolini, Z. A. Cohn.
1991
. Differential effects of neutrophil-activating peptide 1/IL-8 and its homologues on leukocyte adhesion and phagocytosis.
J. Immunol.
147
:
4211
18
De la Harpe, J., C. F. Nathan.
1985
. A semi-automated micro-assay for H2O2 release by human blood monocytes and mouse peritoneal macrophages.
J. Immunol. Methods
78
:
323
19
Nathan, C., Q. Xie, L. Halbwachs-Mecarelli, W. W. Jin.
1993
. Albumin inhibits neutrophil spreading and hydrogen peroxide release by blocking the shedding of CD43 (Sialophorin, Leukosialin).
J. Cell Biol.
122
:
243
20
Detmers, P. A., D. Zhou, E. Pollizi, R. Thieringer, W. A. Hanlon, S. Vaidya, V.S. Bansal.
1998
. Role of stress-activated mitogen-activated protein kinase (p38) in β2 integrin-dependent neutrophil adhesion and the adhesion dependent oxidative burst.
J. Immunol.
161
:
1921
21
Van Kessel, K. P. M., C. T. Park, S. D. Wright.
1994
. A fluoresence microassay for the quantitation of integrin-mediated adhesion of neutrophils.
J. Immunol. Method
172
:
25
22
Wright, S. D., J. I. Weitz, A. J. Huang, S. M. Levin, S. C. Silverstein, J. D. Loike.
1998
. Complement receptor type three (CR3, CD11b/CD18) of human polymorphonuclear leukocytes recognizes fibrinogen.
Proc. Natl. Acad. Sci. USA
85
:
7734
23
Negishi, M., Sugimoto Y., Ichikawa A..
1993
. Prostanoid receptors and their biological actions.
Prog. Lipid Res.
32
:
417
24
Tontonoz, P., L. Nagy, J. G. A. Alvarez, V. L. Thomazy, R. M. Evans.
1998
. PPARγ promotes monocyte/macrophage differentiation and uptake of oxidized LDL.
Cell
93
:
241
25
Nagy, L., P. Tontonoz, J. G. A. Alvarez, H. Chen, R. M. Evans.
1998
. Oxidized LDL regulates macrophage gene expression through ligand activation of PPARγ.
Cell
93
:
229
26
Armstrong, R. A., E. Talpain.
1994
. Comparison of the prostaglandin E (EP) receptor of human neutrophils and HL-60 cells differentiated with DMSO.
Prostaglandins
48
:
221
27
Ney, P., K. Schror.
1991
. PGD2 and its mimetic 2K 110.841 are potent inhibitors of receptor-mediated activation of human neutrophils.
Eicosanoids
4
:
21