Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The stimulation of human leukocytes by various agents such as calcium ionophores, soluble agonists, or phagocytic stimuli results in the biosynthesis of the bioactive arachidonic acid (AA)-derived leukotrienes (LTs). In human polymorphonuclear leukocytes (PMN), the major AA-derived metabolites synthesized are LTs A₄ and B₄ (LTA₄, LTB₄); LTB₄ is one of the most potent naturally occurring leukocyte chemoattractants (Borgeat and Naccache, 1990). Its synthesis after cell stimulation is the result of the calcium-dependent release of AA from cellular phospholipids after the activation of phospholipase(s) (PL) A₂ and the stereospecific transformation of AA to 5-hydroperoxyeicosatetraenoic acid (5-HpETE) by the 5-lipoxygenase (5-LO) (Borgeat and Naccache, 1990)_. 5-HpETE then can be reduced to 5-hydroxyeicosatetraenoic acid (5-HETE) or converted further by 5-LO to LTA₄, which then is rapidly converted to LTB₄ by LTA₄ hydrolase. LTB₄ and 5-HETE are both biologically active compounds that stimulate leukocytes by interacting with distinct cell surface receptors (Yokomizo et al., 1997).

In addition to being transformed to oxygenated metabolites, free AA generated by PL A₂ activity also is released from activated cells in vitro and also has been measured in inflammatory foci (Barr et al., 1984; Lundy et al., 1990). This extracellular AA can be utilized for LT biosynthesis by agonist-stimulated neutrophils; such transcellular metabolism of AA is believed to play an important role in the overall production of the lipid mediators of inflammation (Maclouf et al., 1989; Serhan and Sheppard, 1990). AA also can act as a multifunctional agonist because the addition of AA to PMN stimulates the mobilization of calcium, degranulation, and superoxide anion production (Smith et al., 1987).

It is now well documented that adenosine acting via its A_2a receptor is a potent suppressor of PMN functions (Cronstein et al., 1983; Newby et al., 1983; Cronstein, 1994; Krump et al., 1997). Recently, it has been shown that PMN suspensions release adenosine in quantities that can severely inhibit functional responses to various agonists (Cronstein et al., 1983; Newby et al., 1983; Krump et al., 1997). In particular, the accumulation of adenosine in the incubation medium after a 30-min preincubation period at 37°C nearly completely inhibits LTB₄ production by formyl-methionyl-leucyl-phenylalanine (fMLP)- or platelet-activating factor (PAF)- stimulated PMN (Krump et al., 1997).

In the present study, we have investigated the impact of adenosine A_2a receptor occupancy on the ability of exogenous AA to stimulate PMN for 5-LO product synthesis. These studies reveal that in the absence of the suppressive effect of adenosine, AA is a highly potent and efficient activator of LT synthesis involving an autocrine stimulatory loop that includes LTB₄ itself. These studies indicate further that adenosine exerts its potent inhibitory effect on AA-induced LTB₄ synthesis by inhibiting the ability of exogenous AA to activate cellular 5-LO. Finally, these results suggest that the pharmacological manipulation of adenosine levels at inflammatory sites may have an important impact on the transcellular biosynthesis of LT.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Fura-2-acetoxymethly ester (fura-2/AM), prostaglandin B₂ (PGB₂), 19-hydroxy-PGB₂, leupeptin, aprotinin, phenylmethanesulfonyl fluoride (PMSF), adenosine deaminase (ADA), and horseradish peroxidase-linked donkey anti-rabbit antibodies were obtained from Sigma Chemical Co. (St. Louis, MO). Methyl arachidonyl fluorophosphonate, AA, and 5,8,11,14,17-eicosapentaenoic acid (EPA) were obtained from Cayman Chemical Co. (Ann Arbor, MI). CGS21680 HCl and CGS15943 were obtained from Research Biochemicals International (Natick, MA). CP105696 was a gift from Dr. Henry Showell (Pfizer, Groton, CT), and MK0591 was a gift from Dr. Robert Young (Merck Frosst, Dorval, Canada). BN50730, LY223982, and L659989 were obtained from the Institut Henri Beaufour (Paris, France), Eli Lilly Laboratories (Indianapolis, IN), and Merck Sharp and Dohme (Rahway, NJ), respectively. SC57461A was a gift from Dr. Walter G. Smith (Searle & Co., Skokie, IL). Drug stock solutions were in dimethyl sulfoxide (DMSO) and were added directly to cell suspensions; the maximal final concentration of DMSO in the cell suspensions was 0.2%.

Isolated Cell Preparations. Venous blood was obtained from healthy donors and collected into 10-ml glass tubes (100 × 16 mm; Vacutainer, Becton Dickinson, Franklin Lakes, NJ) containing 143 USP units of heparin. PMN were isolated from peripheral blood after dextran sedimentation and centrifugation on Ficoll-Paque cushions (Pharmacia, Dorval, Canada), as described previously (Boyum, 1968). Final preparations contained 95% PMN, and viability was >95% as assessed by trypan blue exclusion.

Stimulation of Cells and Analysis of 5-LO. The cells were suspended in Hanks' balanced salt solution (HBSS) (1 × 10⁷ cells/ml) at 37°C and stimulated with the indicated concentrations of fatty acid for the indicated time periods. For experiments in which cells were pretreated with ADA, 0.4 U ADA/ml was added to the cell suspension 2 min before stimulation. In some experiments, cells were preincubated at 37°C with the indicated concentrations of receptor agonists or antagonists before stimulation. For the determination of 5-LO products, reactions were stopped at the indicated times by the addition of 1 volume of ice-cold methanol/acetonitrile (1:1, v/v) containing 12.5 ng each of PGB₂ and 19-hydroxy-PGB₂ as internal standards, and the samples were processed and analyzed by reversed-phase (RP) HPLC with the use of an on-line extraction procedure, as described previously (Borgeat et al., 1990).

Preparation of Nuclei. For the preparation of nuclei, neutrophils (2 × 10⁷ cells/2 ml) incubated under the conditions described were pelleted and resuspended in 600 µl of ice-cold Nonidet P-40 lysis buffer containing 0.1% Nonidet P-40, 10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl₂, 1 mM EDTA, 10 µg/ml leupeptin, 10 µg/ml aprotinin, and 1 mM PMSF (Pouliot et al., 1996). The cells were vortexed for 15 s, kept on ice for 5 min, and centrifuged at 300g (10 min, 4°C). The resulting supernatants (i.e., the non-nuclear fractions) and pellets (the nuclei-containing fractions) then were solubilized immediately in electrophoresis sample buffer containing 62.5 mM Tris-HCl, pH 6.8, 2% SDS, 100 mM dithiothreitol, 10% glycerol, 0.01% bromophenol blue, 10 µg/ml leupeptin, 10 µg/ml aprotinin, and 1 mM PMSF) and boiled for 5 min. Samples were analyzed by SDS-polyacrylamide gel electrophoresis, as described by Laemmli (1970) on 9% acrylamide gels. Proteins then were transferred at 0.5 A for 4 to 15 h at 4°C onto an Immobilon-P polyvinylidene difluoride blotting membrane. Transfer efficiency was visualized by Ponceau Red staining. For the determination of 5-LO, the membranes were soaked for 30 min at 25°C in Tris-buffered saline (25 mM Tris-HCl, pH 7.6, 0.2 M NaCl, 0.15% Tween 20) containing 5% dried milk (w/v), blotted with anti-5-LO, and revealed using a horseradish peroxidase-coupled monoclonal antibody and the enhanced chemiluminescence detection kit.

Measurement of Intracellular Calcium Concentration [Ca²⁺]_i. Fura-2 fluorescence was monitored, as described previously (Faucher and Naccache, 1987). Briefly, cells (1 × 10⁷/ml) were incubated for 30 min with 1 µM fura-2 AM at 37°C. The cells then were washed, resuspended at 5 × 10⁶/ml, and transferred into the thermally controlled (37°C) and magnetically stirred cuvette compartment of the spectrofluorometer (Aminco-Bowman series 2, SLM-Aminco, Urbana, IL). The excitation and emission wavelengths for Ca²⁺ measurements were 340 and 510, respectively.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

The Transformation of AA by 5-LO in Human PMN Is Inhibited by Endogenous Adenosine. A first series of experiments was designed to determine whether endogenous adenosine in isolated PMN preparations interferes with the transformation of exogenous AA into 5-LO products. It was shown previously that isolated PMN release endogenous adenosine into the incubation medium in a time- and cell concentration-dependent manner and that this release affects cell functions (Cronstein et al., 1983; Newby et al., 1983; Krump et al., 1997). As can be seen on the HPLC chromatogram in Fig. 1A, the addition of 3 µM AA to isolated PMN resulted in the synthesis of small amounts of LTB₄ and its -oxidation products 20-hydroxy-LTB₄ (20-OH-LTB₄) and 20-carboxy-LTB₄ (20-COOH-LTB₄). This is in accord with results of many previous reports that AA is a weak stimulus for LT synthesis and that the threshold concentration of AA required to induce basal LTB₄ synthesis is in the range of 3 to 10 µM (Clancy et al., 1983; McColl et al., 1989). However, when cells were pretreated with the adenosine A_2a receptor antagonist CGS15943 (Ghai et al., 1987), the capacity of the cells to synthesize LTB₄, its -oxidation products, and the nonenzymatic hydrolysis products of LTA₄, 6-trans-LTB₄ and 12-epi-6-trans-LTB₄, in response to 3 µM AA was greatly enhanced (greater than 10-fold), as shown in Fig. 1B. The cells' capacity to respond to exogenous AA was similarly enhanced when adenosine was removed enzymatically from the incubation medium by the addition of ADA (Fig. 1C). However, when the adenosine A_2a receptor agonist CGS21680 was included in these incubations containing ADA, the enhanced response to AA was reversed (Fig. 1D). In fact, the adenosine A_2a receptor agonist CGS21680 was shown to be a very potent inhibitor (IC₅₀ of ~1 nM) of the AA-induced synthesis of 5-LO products in PMN (treated with ADA) (Fig. 2).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   RP-HPLC chromatograms of eicosanoids from human PMN treated with exogenous AA. PMN (1 × 107/ml) were preincubated for 5 min at 37°C in HBSS (A), in HBSS containing 1 µM CGS15943 (B), in HBSS containing 0.4 U/ml ADA (C), or in HBSS containing adenosine deaminase and 0.1 µM CGS21680 (D). The cells then were stimulated for 5 min with 3 µM AA and the reactions stopped by the addition of organic solvents containing the internal standards PGB2 and 19-OH-PGB2 (12.5 ng each), as described in Experimental Procedures. LTB4, its -oxidation products 20-OH-LTB4 and 20-COOH-LTB4, and the nonenzymatic hydrolysis products of LTA4 (6-trans-LTB4 and 12-epi-6-trans-LTB4) were analyzed by RP-HPLC, as described in Experimental Procedures. Attenuation setting of the UV detector was 0.01 absorbance unit at full scale at 280 nm.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   The effect of CGS21680 on the biosynthesis of 5-LO products by human PMN stimulated with exogenous AA. PMN (1 × 107/ml) were preincubated at 37°C for 5 min in HBSS containing ADA (0.4 U/ml) and the indicated concentrations of CGS21680. The cells then were stimulated for 5 min with 3 µM AA and the reactions stopped by the addition of organic solvents containing the internal standards PGB2 and 19-OH-PGB2 (12.5 ng each), and 5-LO products were determined by RP-HPLC, as described in Experimental Procedures. The data show the relative amounts of LTB4 and its metabolites generated. Values are the means ± S.E. of three separate experiments, each performed in duplicate.

Kinetics of the Synthesis of 5-LO Products. The synthesis of 5-LO products normally requires both AA release from phospholipids and activation of the 5-LO enzyme. However, because substrate availability should not be limiting when exogenous AA is added to the incubation media (as in the present experimental conditions), the kinetics of the synthesis of 5-LO products under such conditions are likely only reliant on 5-LO activation. The kinetics of 5-LO product synthesis in PMN exposed to AA in the presence of ADA were evaluated; as can be seen in Fig. 3A, a lag period of approximately 40 s transpired before a measurable synthesis of 5-LO products could be detected, although the substrate AA should be available immediately for transformation. This observed lag period varied between 20 and 60 s, depending on the PMN preparation. The addition of 3 µM AA to PMN induced a synthesis of 5-LO products that was maximal within 2 min. The decrease observed in the quantities of detectable 5-HETE when incubations are extended to >3 min is because 5-HETE is efficiently acylated into cellular glycerolipids (Stenson and Parker, 1979).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   The kinetics and dose-response to exogenous AA for the biosynthesis of 5-LO products in human PMN. PMN (1 × 107/ml) were preincubated at 37°C for 2 min in HBSS containing ADA (0.4 U/ml). A, cells were stimulated with 3 µM AA and, at the indicated times reactions, were stopped by the addition of organic solvents. , LTB4; , 5-HETE. B, cells were stimulated with the indicated concentrations of AA for 7 min and reactions stopped by the addition of organic solvents. LO products were determined by RP-HPLC, as described in Experimental Procedures. , 5-LO products; , 15-HETE. Values for 5-LO products represent the sum of LTB4, its -oxidation products 20-OH-LTB4 and 20-COOH-LTB4, and the nonenzymatic hydrolysis products of LTA4 (6-trans-LTB4 and 12-epi-6-trans-LTB4). Values are the means ± range of duplicate incubations from one experiment, which is representative of four separate experiments.

LTB₄ Receptor Antagonists and an LTA₄ Hydrolase Inhibitor Block the AA-Induced Synthesis of 5-LO Products. Although no specific cell surface receptor capable of transmitting a signal has been described for fatty acids including AA, previous studies have shown that AA-induced signals can be blocked with pertussis toxin, suggesting that AA may act either directly or indirectly via G protein-coupled receptors (McColl et al., 1989). It also has been suggested that these may be LTB₄ receptors (Naccache et al., 1989). To determine whether the induction of 5-LO product synthesis by AA may involve the LTB₄ receptor, PMN were preincubated for 2 min with two different LTB₄ receptor antagonists, CP105696 and LY223982, before cell stimulation in the presence of ADA. Figure 4 shows that both LTB₄ receptor antagonists blocked the synthesis of 5-LO products induced by AA at concentrations that are consistent with their previously reported IC₅₀ for inhibiting PMN responses to LTB₄ (74 nM for LY223982 and 5 nM for CP105696) (Jackson et al., 1992; Showell et al., 1995). When AA was added along with the agonists fMLP and PAF, which stimulate PMN via their own receptors, LTB₄ receptor antagonists were without effect on 5-LO product synthesis, indicating that LTB₄ receptor antagonists were acting in a receptor-specific manner (data not shown). These results suggest that either AA itself can act as an agonist of the LTB₄ receptor or that small amounts of LTB₄ generated in response to AA can then act in an autocrine manner on its receptor and stimulate the cell for a more important biosynthesis of 5-LO products. This hypothesis of an autocrine stimulation of the PMN by endogenous LTB₄ was verified further by using the LTA₄ hydrolase inhibitor SC57461A. As can be seen in Fig. 5A, pretreatment of PMN with SC57461A inhibited in a dose-dependent manner the ability of exogenous AA to induce the synthesis of all 5-LO products (including 5-HETE and 6-trans isomers of LTB₄, whose synthesis is not dependent on LTA₄-hydrolase activity), indicating that the transformation of LTA₄ to LTB₄ is absolutely required for the AA-induced synthesis of 5-LO products in the experimental conditions used. In contrast, when SC57461A-treated PMN were stimulated with the calcium ionophore A23187, only LTB₄ synthesis was blocked, not that of LTA₄ (as evidenced by the detection of its nonenzymatic hydrolysis products 6-trans-LTB₄ and 12-epi-6-trans-LTB₄) or 5-HETE (Fig. 5B). This is in agreement with our previous report that endogenous LTB₄ does not exert a significant autocrine stimulatory effect on LT synthesis in human PMN stimulated with 1 µM A23187 (McDonald et al., 1994).

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Effect of LTB4 receptor antagonists on the synthesis of 5-LO products in PMN stimulated with exogenous AA. PMN (1 × 107/ml) were preincubated at 37°C for 5 min in HBSS containing ADA (0.4 U/ml) and the indicated concentrations of LTB4 receptor antagonists. Cells then were stimulated with 3 µM AA for 7 min and the reactions stopped by the addition of organic solvents, and 5-LO products were determined by RP-HPLC, as described in Experimental Procedures. Values represent the sum of LTB4, its -oxidation products, and the nonenzymatic hydrolysis products of LTA4. Values are the means ± S.D. of triplicate incubations from one experiment, which is representative of four separate experiments.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Effect of SC57461A on the biosynthesis of 5-LO products in PMN stimulated with exogenous AA or ionophore A23187. PMN (1 × 107/ml) were preincubated for 5 min at 37°C in HBSS containing 0.4 U/ml ADA and SC57461A at the indicated concentrations. The cells then were stimulated for 5 min with 3 µM AA (A) or 1 µM A23187 (B), and the reactions were stopped by the addition of organic solvents containing the internal standards PGB2 and 19-OH-PGB2 (12.5 ng each), as described in Experimental Procedures. LTB4 and its -oxidation products 20-OH-LTB4 and 20-COOH-LTB4 (, the nonenzymatic hydrolysis products of LTA4 (6-trans-LTB4 and 12-epi-6-trans-LTB4) (), and 5-HETE ()were analyzed by RP-HPLC, as described in Experimental Procedures. Values are the means ± S.D. of triplicate incubations from one experiment, which is representative of three separate experiments.

AA Induces the Translocation of 5-LO to the Nucleus. One of the features of 5-LO activation after stimulation of human PMN is the translocation of the enzyme from the cytosol to the nuclear envelope where the 5-LO activating protein is present (Woods et al., 1993; Pouliot et al., 1996). Therefore, the translocation of 5-LO to the nucleus after activation of PMN by AA was assessed in cells treated with ADA alone and in combination with the adenosine A_2a receptor agonist CGS21680. Figure 6 shows that the stimulation of human PMN with 5 µM AA efficiently induces the translocation of 5-LO to the nucleus. Importantly, when cells were preincubated with the 5-LO activating protein antagonist MK0591, both 5-LO product biosynthesis and 5-LO translocation were inhibited. Stimulation of the adenosine A_2a receptor with the agonist CGS21680 also completely blocked AA-induced translocation of 5-LO and 5-LO product synthesis. Similarly, the LTB₄ receptor antagonist CP105696 also blocked the translocation of 5-LO, supporting the idea that a signal transmitted through this receptor is required for the AA-induced translocation of 5-LO. This is consistent with the effects of this antagonist on LTB₄ synthesis (Fig. 4). As with the biosynthesis of 5-LO products, the PAF receptor antagonist BN50730 did not block 5-LO translocation induced by AA.

View larger version (28K): [in this window] [in a new window]   Fig. 6.   5-LO product synthesis and immunoblot analysis of 5-LO translocation to nuclei in AA-stimulated human PMN. Human PMN were pretreated for 5 min with ADA (0.4 U/ml) and the indicated compounds (0.1 µM), except for BN50730, which was at 1.0 µM. Cells then were stimulated with 3 µM AA for 5 min and centrifuged, and the supernatants were evaluated for 5-LO products. Nuclei were isolated from the cell pellets and prepared for gel electrophoresis and immunoblot analysis, as described in Experimental Procedures. Each lane was loaded with nuclei obtained from 15 × 106-cell equivalents of PMN. Results are from one experiment, which is representative of four separate experiments

AA Induces Calcium Mobilization: Effect of CGS21680, MK0591, CP105696, and SC57461A. The mechanism of the stimulation of 5-LO product biosynthesis by exogenous AA in human PMN and the inhibitory effects of adenosine on this process was investigated further in Ca²⁺ mobilization studies. As shown in Fig. 7, stimulation of PMN with 3 µM AA induced a transient elevation of [Ca²⁺]i; furthermore, as observed in the kinetics of AA-induced 5-LO product biosynthesis, a 20- to 60-s lag was observed between the time of addition of AA to the PMN suspensions and the elevation of intracellular Ca²⁺ concentrations, clearly indicating that AA does not directly trigger the Ca²⁺ mobilization, and supporting a causal relationship between these events. The treatment of cells with CGS21680 resulted in a profound inhibition of the biosynthesis of 5-LO products and 5-LO translocation induced by AA; similarly, the adenosine A_2a receptor agonist CGS21680 showed a concentration-dependent inhibitory effect on the Ca²⁺ transient induced by AA (Fig. 7A).

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Effect of CGS21680, CP105696, MK0591, and SC57461A on AA-induced calcium mobilization in human PMN. Fura-2-loaded PMN (5 × 106/ml) in HBSS were preincubated for 5 min at 37°C with ADA (0.4 U/ml) in the presence of CGS21680 at 0, 10, and 100 nM (A); CP105696 at 0, 1, and 10 nM (B); MK0591 at 0, 10, and 100 nM (C); and SC57461 at 0, 10, and 100 nM (D) before stimulation with 3 µM AA. Unstimulated cells (dotted lines) were incubated with diluent only (0.1% DMSO).  indicate the time of AA additions; the fluorescence was monitored at the excitation and emission wavelengths of 340 nm and 510 nm, respectively, as described in Experimental Procedures. The results are representative of two to five separate experiments. Full traces indicate no drug added; dashed lines, low drug concentrations; interrupted lines, high drug concentrations.

EPA Is a Weak Stimulus of Ca²⁺ Mobilization and 5-LO Product Biosynthesis. Another polyunsaturated fatty acid, EPA, which is also a substrate for 5-LO, was evaluated for its ability to induce the biosynthesis of 5-LO products in human PMN. Despite the fact that EPA is as good a substrate for 5-LO as is AA (Lee et al., 1984), the addition of 1 to 5 µM EPA to ADA-treated PMN resulted in little or no detectable synthesis of its 5-LO products (data not shown). Accordingly, EPA also was a much weaker stimulus than AA for the induction of Ca²⁺ mobilization in PMN (data not shown). Such data are entirely compatible with the hypothesis proposed above, inasmuch as LTB₅, the product of EPA transformation by 5-LO, is a much weaker (25-fold) agonist of the LTB₄ receptor (as assessed by Ca²⁺ mobilization in human PMN) (Powell et al., 1996), which likely explains the inability of EPA to trigger the formation of 5-LO products. Interestingly, the addition of small concentrations (1 nM) of exogenous LTB₄ to ADA-treated PMN exposed to 5 µM EPA resulted in an enhanced biosynthesis of 5-LO products from EPA (26 ± 2 pmol/10⁶ cells, n = 3).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The incubation of isolated human PMN in vitro results in the rapid accumulation of endogenous adenosine, which has profound inhibitory effects on many cell functions, including LT synthesis. In the present study, we show that when endogenous adenosine is eliminated enzymatically from PMN suspensions, or its actions blocked with receptor antagonists, human PMN respond very strongly to low micromolar concentrations of AA for the synthesis of 5-LO products. This observation contrasts with the widely held perception that resting human PMN respond poorly to exogenous AA for LTB₄ synthesis and that the measurable synthesis of 5-LO products in the presence of AA requires the simultaneous stimulation with another agonist. Thus, our studies establish clearly that AA can trigger, in the absence of other added PMN stimuli, an important generation of 5-LO products and that the effect of exogenous AA is highly sensitive to the inhibitory effect of adenosine A_2a receptor engagement on PMN. Therefore, these findings are in line with our previous observations that A_2a receptor agonists as well as endogenous adenosine accumulating in PMN suspensions down-regulate agonist-induced LT biosynthesis in PMN (Krump et al., 1996, 1997) and add to the potential importance of adenosine as a natural regulator of LT synthesis.

The present observation that PMN do efficiently utilize exogenous AA (when released from the inhibitory constraint of adenosine) strongly supports the concept that the transcellular metabolism of exogenous AA by PMN could be an important source of biologically active eicosanoids such as lipoxins and LTs at inflammatory sites. Indeed, PMN can be exposed to elevated concentrations of free AA on close contact with activated cells releasing AA into the extracellular milieu. Such transcellular metabolism of AA has been shown to be an important pathway for the synthesis of a number of AA metabolites. In particular, AA released by stimulated platelets can be utilized by stimulated PMN for LTA₄ and LTB₄ synthesis (Palmantier and Borgeat, 1991), whereas LTA₄ released by stimulated PMN can be converted to lipoxins by the platelet 12-LO (Serhan and Sheppard, 1990). Thus, the present data indicate that PMN likely utilize exogenous AA in a much more efficient manner than was thought previously. In fact, the quantities of LTB₄ produced by PMN in the presence of 3 µM AA (without the addition of other stimulatory agents) exceed the quantities reported previously under most conditions of PMN stimulation involving priming agents and receptor-mediated agonists that activate PL A₂ and 5-LO for LT synthesis (Dahinden et al., 1988; Surette et al., 1998). Moreover, such efficient utilization of exogenous AA may take place in pathological situations in which secreted PL(s) A₂ accumulation and associated AA release occur, such as in adult respiratory distress syndrome, in the synovial fluid of arthritic patients, and in the asthmatic lung (Nevalainen, 1993). The two potent anti-inflammatory drugs methotrexate and sulfalazine have been shown to promote adenosine accumulation at inflammatory sites (Cronstein, 1995), an effect believed to be involved in their anti-inflammatory activity. Therefore, our data support the concept that the pharmacologically mediated accumulation of adenosine in inflammatory sites constitutes a promising approach for the treatment of inflammatory diseases, inasmuch as LTB₄ might play important roles in some disease states.

AA-induced 5-LO product biosynthesis showed intriguing features that suggested a role of de novo synthesized LTB₄ in the activation of PMN. Indeed, our data demonstrate that the observed biosynthesis of 5-LO products in response to exogenous AA involves an absolute requirement for an autocrine/paracrine activation of the PMN by endogenous LTB₄, resulting in Ca²⁺ release from intracellular stores, the translocation and activation of 5-LO, and the important transformation of exogenous AA into 5-LO products (Fig. 8). Several lines of evidence support this mechanism. First, the addition of AA to PMN did not result in an immediate mobilization of Ca²⁺ or synthesis of 5-LO products, because a lag of 20 to 60 s (variable between PMN preparations) was observed consistently. This indicates that AA itself is not acting directly in a receptor-dependent manner as is observed when cells are stimulated with agonists such as fMLP, PAF, or LTB₄. Second, blocking the LTB₄ receptor with specific receptor antagonists strikingly inhibited the AA-induced delayed mobilization of Ca²⁺, 5-LO translocation, and the delayed biosynthesis of 5-LO products. This suggests that LTB₄ receptor occupation is required for AA-induced PMN stimulation leading to 5-LO translocation and activation. Third, the specific LT synthesis inhibitor MK0591 also inhibited the mobilization of calcium, clearly indicating that 5-LO activity is required for this response to exogenous AA. That LTB₄ itself is responsible for this autocrine effect is supported further by studies with the LTA₄ hydrolase inhibitor SC57461A, which prevents the transformation of LTA₄ to LTB₄, but not the synthesis of other 5-LO product such as 5-HETE. In the present studies, SC57461A inhibited AA-induced synthesis of all 5-LO products and Ca²⁺ mobilization in PMN, indicating that LTB₄ formation is specifically required for these responses to exogenous AA. Additional evidence that an initial biosynthesis of LTB₄ is required for the further biosynthesis of large amounts of 5-LO products on the addition of AA to ADA-treated cells was obtained by using EPA instead of AA in similar experiments. Indeed EPA, which is as good a substrate for 5-LO as AA (Lee et al., 1984), did not induce an important synthesis of its 5-LO products (such as LTB₅) when added to PMN. This is likely because LTB₅, which is a much weaker agonist than LTB₄ for the LTB₄ receptor, did not induce cell activation as measured herein by Ca²⁺ mobilization. Altogether, these observations point to a requirement for an initial LTB₄ synthesis in response to exogenous AA, which triggers through an autocrine/paracrine mechanism the full activation of PMN for 5-LO product synthesis.

View larger version (23K): [in this window] [in a new window]   Fig. 8.   Hypothetical mechanism of 5-LO product biosynthesis involving autocrine activation by LTB4 in PMN exposed to exogenous AA. Unstimulated PMN exposed to AA generate LTA4 and LTB4; this initial synthesis might involve a small quantity of 5-LO present on the nuclear envelope and is likely Ca2+-independent (Skorey and Gresser, 1998). The LTB4 generated initially then plays a pivotal role in activating further the PMN through engagement of the LTB4 receptors, which results in Ca2+ release from intracellular stores, stimulation of the translocation of cytosolic 5-LO, and enhanced LTA4 and LTB4 biosynthesis from exogenous AA. In these experimental conditions, this autocrine/paracrine stimulatory loop by endogenous LTB4 is required for the full activation of LT biosynthesis by exogenous AA, as demonstrated by the almost complete inhibition of 5-LO product formation by LTB4 receptor antagonists and an LTA4-hydrolase inhibitor. Engagement of the A2a receptor appears to interfere with the autocrine/paracrine activation loop by inhibiting 5-LO translocation (through a yet-unidentified mechanism; see Discussion).

This scheme of events in AA-induced 5-LO product biosynthesis (Fig. 8) implies that the initial synthesis of LTB₄ responsible for the autocrine/paracrine activation of the PMN can occur in the absence of a measurable Ca²⁺ mobilization. In support of such a mechanism, a recent report indicated that in the presence of phosphatidylcholine-rich phospholipid vesicles, the 5-LO catalyzes the conversion of AA in the absence of Ca²⁺ (Skorey and Gresser, 1998). Therefore, it seems reasonable to speculate that the small amount of 5-LO detected in/on the nuclei of unstimulated PMN (Fig. 6) (Surette et al., 1998) could generate the small amounts of LTA₄ required to initiate the activation of the PMN through engagement of LTB₄ receptors. The 20- to 60-sec lag observed in these experiments between addition of AA and the occurrence of 5-LO product biosynthesis may reflect the time required for the hydroperoxide-mediated activation of the 5-LO, i.e., the oxidation of the nonheme iron of 5-LO (Chasteen et al., 1993). A similar time lag has been observed between PMN stimulation with agonists (GM-CSF/PAF or fMLP) or ionophore A23187 and the occurrence of LT biosynthesis (Krump and Borgeat, 1994).

Finally, it is noteworthy that the present studies with this particular model (AA-induced LT biosynthesis) unraveled an additional inhibitory mechanism of A_2a receptor engagement on LT synthesis in PMN. In previous studies, it was clearly established that the A_2a receptor agonist CGS21680 is a potent inhibitor of AA release in agonist-activated human PMN (Krump et al., 1999), strongly supporting that such a mechanism was involved in the inhibitory effects of the adenosine analog on LT synthesis. Because in these and other previous studies, the inhibition of AA release also was observed with a variety of other agents known to cause an elevation of intracellular cAMP in PMN (E-type prostaglandins, the -adrenergic agent isoproterenol, and the type IV phosphodiesterase inhibitor Rolipram (Schering AG, Berlin, Germany) (Ham et al., 1983; Fonteh et al., 1993; Krump et al., 1999), it was proposed that the effect of CGS21680 on AA release was cAMP-dependent. In the present studies, it is demonstrated that CGS21680 is a potent inhibitor of LT synthesis, even under conditions of PMN stimulation by exogenous substrate (AA), clearly indicating another site of action of A_2a receptor agonists in the inhibition of LT synthesis. CGS21680 does not inhibit the release of Ca²⁺ from intracellular stores induced by LTB₄ in human PMN (data not shown). However, as shown herein, CGS21680 interferes with the activation/translocation process of the 5-LO. Hypothetically, A_2a receptor engagement might down-regulate phosphorylation events necessary for 5-LO translocation; indeed, recent studies by Lepley et al. (1996) described the inhibitory effects of various tyrosine kinase inhibitors on LT synthesis and 5-LO phosphorylation and translocation in human PMN. Studies are in progress to define further the molecular events involved in the inhibition of AA release and 5-LO activation by A_2a agonists in human neutrophils.