Bacterial Lipopolysaccharide Priming of P388D1 Macrophage-like Cells for Enhanced Arachidonic Acid Metabolism PLATELET-ACTIVATING FACTOR RECEPTOR

P388D1 cells are stimulated by platelet-activating factor (PAF) to release arachidonic acid metabolites (Lister, M. D., Glaser, K. B., Ulevitch, R. J., and Dennis, E. A. (1989) J. Biol. Chem. 264, 8520-8528). While the release of prostaglandin E2 (PGE2) in response to PAF is only two to three times the constitutive PGE2 production, bacterial lipopolysaccharides (LPS) are able to prime P388D1 cells for enhanced arachidonic metabolism, increasing PAF-stimulating PGE2 production to 9-12 times the constitutive PGE2 production. The extent and rate of [3H]arachidonic acid release from prelabeled P388D1 cells are also increased in primed cells relative to unprimed cells in response to PAF-stimulation. LPS from either Salmonella Re595 or Escherichia coli 0111:B4 prime P388D1 cells in a concentration-dependent manner but have themselves no ability to stimulate arachidonic acid metabolism. LPS priming is sensitive to inhibition by actinomycin D, while primed PAF-stimulation of PGE2 production is blocked by cyclohexamide which implicates a protein which is rapidly turning over. Primed PAF stimulation is also inhibited by the phospholipase A2 inhibitor manoalogue and the tyrosine-specific protein kinase inhibitor genistein, but not by the kinase inhibitor H-7. These results suggest that priming amplifies signal transduction pathways for PAF, which results in increased arachidonate availability. The multiple levels at which primed PAF-stimulated PGE2 production appears to be regulated are discussed.

While the release of prostaglandin EB (PGE2) in response to PAF is only two to three times the constitutive PGEz production, bacterial lipopolysaccharides (LPS) are able to prime P388D1 cells for enhanced arachidonic metabolism, increasing PAF-stimulating PGEz production to 9-12 times the constitutive PGEz production.
The The multiple levels at which primed PAF-stimulated PGE, production appears to be regulated are discussed.
The release of arachidonic acid and the production of eicosanoids is an early event in the activation of macrophages by many types of inflammatory stimuli. This has been extensively studied in murine resident peritoneal macrophages (l-5) and several murine macrophage-like cell lines (6-11). The release of arachidonic acid from the sn-2 position of membrane phospholipids is thought to be the rate-limiting event in the biosynthesis of the eicosanoids. This rate-limiting event is most likely controlled by a phospholipase type enzyme (12), the most direct mechanism being the action of a phospholi- pase AP (PLA*)' which would release arachidonate directly from the m-2 position of membrane phospholipids (13). Characterization of the different types of phospholipases in these inflammatory cells has been an important first step in the elucidation of the mechanisms which regulate arachidonic acid release (14, 15).
The characterization of the phospholipase activities has revealed multiple forms present in both resident peritoneal macrophages (16, 1'7) and macrophage-like cell lines (14, 18,19). The most detailed characterization is available on a membrane-associated, Ca'+-dependent PLAz from the P388D1 cell line (14). This enzyme has been purified (20), kinetically characterized (21), and evaluated with potential phospholipase AZ inhibitors (7) as has a soluble lysophospholipase from the same cell (22). Nonetheless, the mechanisms by which these enzymes are activated or regulated in the intact cell are still poorly understood.
In peritoneal macrophages, distinct differences in the activation mechanisms for receptor-mediated uersus soluble stimulants have been observed (5). There appears to be a Na' requiring event early in phagocytic/receptor-mediated activation of arachidonic acid release but not for activation by soluble stimulants such as PMA or Ca*+ ionophore A23187. Protein synthesis is also required for both PGEz production (23) and arachidonic acid release in peritoneal macrophages (5). The model proposed by Aderem et al. (5) for receptormediated arachidonic acid release consists of a sequential series of signals involving Na+ influx, protein synthesis, and finally an elevation in intracellular Ca'+. Therefore, in the macrophage, there are mechanisms which appear to control the ability of these cells to generate eicosanoids which are complex and stimulus dependent which ultimately lead to activation of PLA*. A divergence in this receptor-mediated sequence of signals for arachidonic acid release is demonstrated by bacterial lipopolysaccharides (LPS) which are poor triggers for the release of arachidonic acid in macrophages (24), but are able to prime macrophages for enhanced arachidonic acid metabolism in response to various stimuli (24,25). The mechanism of LPS priming of macrophages is poorly understood but affects many of the functional capacities of these cells (26,27). Many of the actions of LPS in endotoxin shock may be mediated by macrophages (26) which produce potent mediators of shock including the eicosanoids and platelet-activating factor and tumor necrosis factor (28)(29)(30) may also participate in a priming effect which allows for enhanced production of these mediators, exacerbating the acute onset in shock.
In this report, we demonstrate the priming of a murine macrophage-like cell P388D1 by bacterial lipopolysaccharide for enhanced arachidonic acid metabolism. The priming of the P388D1 cell is qualitatively different than that observed in resident murine peritoneal macrophages and appears to require RNA synthesis as priming is sensitive to inhibition by actinomycin D but not cyclohexamide. The priming of P388D, cells enhances PGEl production in response to Ca*+ ionophore A23187 and PAF but not to the protein kinase C activator oleoyl-acetyl-sn-glycerol (OAG

LPS Priming-Stimulus
Response-The effect of LPS on the "priming" of P388D1 macrophage-like cells for enhanced PGE, production in response to various stimuli is shown in Fig. 1. LPS-primed P388D1 cells produce three to five times as much PGE, as unprimed (control) cells in response to Ca2+ ionophore A23187 and PAF stimulation.
Priming is effective at enhancing PGEz production for stimulants such as A23187 and PAF but does not enhance PGE, production in response to stimuli which do not stimulate unprimed cells, such as OAG (7). In non-stimulated cells, LPS priming results in a 1.6-fold increase in PGEz production as compared with unprimed cells. In contrast, priming increases PGEz production No  by 3-and 3.6-fold in response to A23187 and PAF, respectively, as compared with unprimed stimulated cells. OAG treatment of primed P388D, cells showed only a l.l-fold increase in PGEz production which is not significantly different than the increase observed in non-stimulated cells. Priming results in increased levels of PGE2 as well as the other arachidonic acid metabolites as compared with control cells, but no redistribution of metabolites among the prostaglandins is observed (data not shown).

PAF Dose Response in LPS-primed
Cells-In P388Di cells stimulated with PAF (10 nM to 1 pM) only a 2-3-fold increase in PGE, production over constitutive PGE2 production (7) could be observed which made it difficult to determine a doseresponse relationship for PAF. However, in LPS-primed P388D1 cells, PAF (10 nM to 1 pM) produced a 9-12-fold increase in PGE2 production over constitutive production as shown in Figs. 1 and 2. In primed cells, a PAF dose-response was determined between 0.01 and 10 nM (Fig. 2). Concentrations of PAF between 10 nM and 1 pM gave no further increase in PGE2 production. The apparent EDbO for PAF-stimulated PGE, production in primed P388D1 cells is approximately 0.2 nM. This EDsa value is consistent with a Kd determined for PAF binding to P388D1 cells of 0.08 nM (32). Enantio-PAF (D-PAF) is approximately 500-fold less potent than PAF (L-PAF) at stimulating PGE? production in P388D, cells. This demonstrates stereospecificity of the PAF receptor in P388Di cells for PGE, production in response to PAF. The PAF antagonist L-659,989 (33) inhibited PGE, production in LPS primed cells by 75 and 91% at 10 and 100 nM, respectively, when stimulated with 10 nM PAF.

PHlArachidonic
Acid Release in Primed Cells-LPSprimed P388D1 cells, upon stimulation with PAF, release greater amounts of ["Hlarachidonic acid from prelabeled cells and the rate of release appears to be increased (as per the 30min time point) relative to release from unprimed cells ( sentative of the maximum release in both primed and unprimed cells. LPS-primed cells release approximately two to three times more [3H]arachidonic acid in response to PAF. This result is somewhat less than the 3-5-fold increase in PGE2 production observed with LPS priming and may reflect the dispartate labeling of phospholipid pools with exogenously applied [3H]arachidonic acid (7, 34).
Effect of LPS Priming on Cyclooxygenase and Phospholipase A, Activities-The level of cyclooxygenase activity, whether measured in a P388D1 cellular homogenate or by application of exogenous arachidonic acid to intact cells, showed no enhanced activity as compared with unprimed (resting) cells. This was also true of the phospholipase A2 activity present in the P388D1 cellular homogenate when measured with vesicles of dipalmitoylphosphatidylcholine at pH 9.0 with 5 mM Ca*+ (data not shown). Therefore, it appears that the activities of neither of the key enzymes involved in arachidonic acid metabolism in P388D, cells are altered appreciably by LPS priming.
LPS Priming of P388D1 Cells-LPS priming of P388D1 cells is concentration dependent from 10 to 1000 rig/ml of rough LPS Re 595 from Salmonella as well as smooth LPS from Escherichia coli Olll:B4 with respect to enhanced PGEz production in response to PAF (Fig. 4A). Some effects of LPS are also observed at concentrations as low as 1 rig/ml. The effect of the exposure time of P388Di cells to LPS (100 ng/ ml) was evaluated to determine if prolonged exposure of P388D1 cells to LPS down-regulated the primed response. The time course of LPS priming demonstrated that enhanced production of PGE2 in response to PAF can be observed at 30 min of exposure to LPS and is maximal at l-2 h of exposure (Fig. 4B). Between 2-4 h of LPS exposure, the P388D1 cells become less responsive to the subsequent stimulation with PAF. After 4 h of exposure to LPS, PAF stimulation results in a response which is 30-35% of the maximal response observed after 1 h of LPS exposure. Therefore, prolonged exposure of P388D1 cells to LPS (longer than 2 h) appears to down-regulate the primed state with regard to subsequent stimulation by PAF.
The stability of the LPS-primed state was determined by priming the cells for 1 h with 100 rig/ml LPS and then washing the cells free of LPS and allowing them to incubate for different time periods in serum-free medium before stimulation with PAF. As shown in Fig. 4C, the primed state of the P388D1 cells was unstable and decreased to 20% of maximal after a 4-h incubation in serum-free medium. The time at which the primed state was reduced to 50% of its maximum was 1.75 h (see inset). Therefore, it appears that the LPSprimed state in P388D1 cells is transient and deactivates with an apparent half-life of approximately 1.75 h in the absence of LPS or stimulus.

Effects of Manoalogue and BW755c on LPS Priming and
Primed Stimulation-The phospholipase inhibitor manoalogue and the dual cyclooxygenase/lipoxygenase inhibitor BW755c when present during LPS exposure had little effect on the ability of LPS to prime P388D1 cells for enhanced PGE, production (data not shown). However, these compounds are effective inhibitors of primed PAF stimulation of PGEz production. Manoalogue has an apparent I&o of 0.4 pM in LPS-primed cells (Fig. 5), as compared to an apparent ICSO of 1.0 pM for A23187 stimulation of unprimed cells (7) and BW755c at 10 pM resulted in 95% inhibition of PGEz production (data not shown). These results suggest that production of cyclooxygenase/lipoxygenase products may not be necessary for LPS priming of P388D1 cells. This is also consistent with previous findings that LPS alone at concentrations up  mycin D) and translational (cyclohexamide) inhibitors on the priming of P388Di cells with low doses of LPS.
As shown in Fig. 6A, actinomycin D reduced LPS priming to 58 and 16% of control at 1 and 3 pM, respectively. The temporal effects of actinomycin D action on LPS priming are shown in Fig. 6B. The maximal effect of actinomycin D (3 pM) is observed when present 30 min prior to LPS exposure or during the first 30 min of LPS exposure. However, a LPS Priming of P388D1 Cells significant (P < 0.05) decrease in the effect of actinomycin D is observed when present during the last 30 min of LPS exposure (26% of control when present during the first 30 min to 56% of control when present during the last 30 min). These data suggest an accumulation of mRNA during the initial exposure of P388Di cells to LPS is essential for the development of the primed state. Cyclohexamide (10 PM) partially prevented (70% of control) the LPS priming if present during the entire LPS exposure period as shown in Fig. 6A. Cyclohexamide present during the first 30 min of LPS exposure had no inhibitory effect on LPS priming and appears to slightly enhance (130% of control) the primed state in P388D, cells (data not shown). This observation is similar to that observed with y-interferon priming of peritoneal macrophages (35). These results suggest the possibility of a partial dependence of LPS priming in P388D1 cells on protein synthesis but a much less dramatic effect than that seen for transcription.
Effects of Actinomycin D and Cyclohexamide on Primed PAF Stimulation-Cyclohexamide (10 pM) inhibits primed PAF stimulation of PGE, production by 88% (Fig. 6A) and reduces PAF-stimulated PGE2 production in primed cells to the level observed in non-primed cells. These data are consistent with the involvement of protein synthesis (a rapid turnover protein) in the stimulation of arachidonic acid metabolism by a number of various stimuli (5, 23) which now include PAF. Actinomycin D had only a partial inhibitory effect on primed PAF stimulation, 63% of control at 3 PM actinomycin D, suggesting the possibility of a partial involvement of transcription in primed PAF stimulation but a much greater role of translation in PAF stimulation of PGEz production.
Effects of H-7, PMA, and Genistein on Primed PAF Stimulation-The role of protein kinase C in LPS priming of neutrophils is apparently unclear (36,37), and LPS does not result in translocation or activation of this enzyme for priming of the respiratory burst. In the P388Di cells, LPS priming for enhanced arachidonic acid metabolism is not inhibited by the protein kinase C inhibitor H-7 (lo-50 FM) nor does H-7 inhibit primed PAF stimulation (data not shown). PMA, a phorbol ester protein kinase C activator, (l-10 PM) reduced LPS priming to 67% of control but did not affect primed PAF stimulation (data not shown). These results suggest a minimal contribution of protein kinase C to the priming of macrophages for enhanced PGEz production and that protein kinase C activation is apparently not required for PAF stimulation of PGE, production in P388D1 cells. Genistein (38) is reported to be a selective tyrosine-specific protein kinase inhibitor with partial inhibitory effects on serine and threonine kinase activity (CAMP-dependent kinases). The dose-response for genistein effects on both LPS [ priming and primed PAF stimulation are shown in Fig. 7. Genistein had a small inhibitory effect on LPS priming in P388D, cells (about 58% of control was observed at 30 PM genistein). However, genistein had a potent inhibitory activity against primed PAF stimulation with an apparent I&o of 7 PM and only 4% of control primed PAF stimulation was observed at 30 PM genistein. These results implicate a protein kinase activity in the PAF-stimulated arachidonic acid release in LPS-primed P388D1 cells.

DISCUSSION
The P388Di cell produces arachidonic acid metabolites, predominantly cyclooxygenase products, in response to various stimuli (7). In the attempts to correlate the effects of various phospholipase inhibitors on the P388Di Ca*+-dependent membrane associated PLA2 in vitro with inhibition of PGE2 and arachidonic acid release in intact P388Di cells, the Ca*+ ionophore A23187 was used as a stimulus (7). However, a key point in studying the regulation of arachidonic acid release as the rate-limiting event for the biosynthesis of eicosanoids is the use of physiologically and pharmacologically relevant stimulants, i.e. receptor-mediated events. P388D1 cells respond to PAF to produce arachidonic acid metabolites (7) and possess high affinity PAF receptors (32); however, in resting cells (unprimed) only a 2-3-fold increase over constitutive PGE, production is observed in response to PAF (7). In contrast, as shown in this report, LPS priming of P388Di cells resulted in a 9-12-fold increase over constitutive PGEz production in response to PAF, thus allowing more quantitative interpretations of inhibitor studies. With LPS priming, it then became possible to evaluate the mechanisms governing both LPS priming and PAF-stimulated arachidonic acid metabolism in P388Di cells.
LPS Priming and PAF-The mechanism by which bacterial lipopolysaccharides prime both neutrophils and macrophages for enhanced metabolic responses have been studied in detail, but still remain unclear. In general, the LPS priming of neutrophils for enhanced superoxide production does not result from the augmentation of the terminal enzymes in this system, e.g. NADPH oxidase, but rather appears to be a modulation (up-regulation) of the signal transduction mechanism (39-41). In the murine resident peritoneal macrophage, LPS priming also results in enhanced arachidonic metabolism in response to various stimuli (24). Herein, we report a response to LPS priming in a murine macrophage-like cell line, P388Dl. P388D1 cells exposed to low doses of LPS become primed for enhanced arachidonic acid metabolism. This enhanced metabolism is apparently stimulus-dependent, as responses to OAG or PMA do not become primed upon LPS exposure. This implies that responses which are coupled to selective transduction pathways for arachidonic acid release are those which are able to be primed by LPS exposure.
The enzymes which would most likely be involved in enhancing arachidonic acid metabolism in P388DI cells or macrophages would be PLAz and/or the cellular cyclooxygenase. In P388D1 cells, LPS priming results in increased VH]arachidonic acid release and increased PGE, (immunoreactive) production.
However, neither the total cellular cyclooxygenase activity nor the Ca*+-dependent PLA2, pH 9.0 optimum, activity was found to be different in primed cells. This is consistent with a change in the signal transduction mechanism(s) leading to a greater availability of substrate (arachidonic acid). This effect (priming) may be the result of a greater efficiency in the translation mechanism(s) or receptorligand interactions which result in arachidonic acid release, rather than increased enzyme levels to account for the increased product observed after LPS priming. PAF-stimulated PGE, production in LPS-primed cells was 9-12 times the constitutive PGE, production. The dose-response curve for primed PAF-stimulated PGE, production gave an EDs0 value of 0.2 nM PAF which is in good agreement with the reported Kd of 0.08 nM for PAF binding (32). The PAF response was also stereospecific, D-PAF being 500-fold less potent, and the selective PAF antagonist L-659,989 (33) effectively inhibited PAF-stimulated PGEz production. These results demonstrate that the PGE, production in P388D1 cells upon PAF stimulation is a receptor-mediated event. Therefore, evaluation of different inhibitors with PAF stimulation appears to be more physiologically relevant as PAF is an important agonist for macrophages and the concentrations of PAF being studied are in a more physiologically relevant range (i.e. a10 nM).
Regulation of LPS Priming-The mechanism of LPS priming still remains unclear. Early studies demonstrated that LPS priming of neutrophils was not sensitive to inhibition by cyclohexamide (42) and therefore a distinct event from induction of cellular protein synthesis. It was then suggested that an increase in resting Ca" levels may account for the priming effect (43,44). In macrophages, LPS priming results in the myristolation of a set of cellular proteins which are also substrates for protein kinase C (45). A sequential set of events starting with LPS-induced myristolation of cellular proteins, membrane association of the myristolated proteins, and phosphorylation by protein kinase C has been proposed to account for the increased arachidonic acid release by LPS priming (46).
Although these mechanisms may be involved in LPS priming, we have observed dependence of LPS priming on transcriptional events (mRNA production) which are sensitive to actinomycin D inhibition. This is consistent with the observed increases in mRNA levels for oncogenes and competence genes during the first 30 min of LPS exposure (35). These results also demonstrate a lesser dependence of priming on protein synthesis which is also observed in human neutrophils (42 23). This is in contrast to non-physiological stimulation (e.g. Ca*+ ionophore A23187) of arachidonic acid release which does not need protein synthesis (5).
Herein we demonstrate that primed PAF stimulation of P388D, cells is also dependent on protein synthesis (sensitive to inhibition by cyclohexamide). The rapid release of arachidonic acid in response to PAF stimulation suggests that this cyclohexamide-sensitive event may depend on a rapid turnover protein which is induced upon receptor activation. This rapidly produced protein may be a product which is related to the LPS-induced transcriptional events occurring during LPS priming. This relationship is currently under investigation. Protein Kinase Regulation of PAF Stimulation-Genistein, a selective tyrosine-specific protein kinase inhibitor (38), appears to be a potent inhibitor of primed PAF stimulation of arachidonic acid metabolism in P388D1 cells. The release of arachidonic acid when P388D1 cells are primed and subsequently stimulated with PAF is most likely mediated by a phospholipase AZ-like enzyme (7). The regulation of cellular PLA2 may be complex considering the ubiquity of this type of enzyme and its general role in cellular homeostasis; therefore, regulation of PLA2 activity by a protein kinase would be an attractive mechanism.
The effects of genistein and the lack of effect of H-7 suggest that a tyrosine-specific protein kinase may be this regulating kinase activity, but this does not exclude the possibility of other kinases, other than protein kinase C, or other nucleotide-dependent enzymes also being involved in this regulation.
This would be the first correlation of tyrosine-kinase inhibition with inhibition of arachidonic acid metabolism and possibly the PLA2 liberating the arachidonic acid. The mechanism by which the kinase regulates the release of arachidonic acid would depend on identification of the kinase substrate which regulates the PLA, activity be it the enzyme (PLA1) itself or another part of the stimulusresponse coupling mechanism. Regulation of phospholipid metabolism by a tyrosine-specific protein kinase has been suggested for phospholipase C activity and growth factor responses (47). The tyrosine-specific protein kinase is also known for its involvement in insulin action. Therefore, the regulation of arachidonic acid release via regulation of the PLA2 or its activation mechanism seems a particularly attractive mechanism to regulate the biosynthesis of eicosanoids during cell activation in inflammatory and other responses.
Proposed Signal Transduction Mechanisms in P3880, Cells--We have described herein inhibitor studies designed to better understand both the LPS priming and PAF-mediated arachidonic acid metabolism in P388D1 cells. As shown in Fig. 8 produces a signal within the cell which results in the generation of a primed state. This signal involves transcriptional and probably translational events. This mechanism for priming has not been proposed previously since in earlier studies cyclohexamide demonstrated little effect on priming in neutrophils (42). However, the involvement of transcriptional events in other early responses to LPS has been clearly demonstrated (35) and would therefore be likely in the mechanism controlling priming. Primed PAF stimulation of P388D1 cells is clearly dependent on translational events and apparently on a protein kinase activity. The results herein are suggestive of a kinase other than protein kinase C and possibly a tyrosine-specific protein kinase. These translational events occurring after PAF stimulation may be direct products of the LPS transcriptional events, but this would have to be more rigorously evaluated in further experiments. The protein kinase is an attractive mechanism for regulation of arachidonic acid release, possibly of the cellular phospholipase AZ, but its substrate would have to be determined to understand how the release of arachidonic acid is being regulated.
There are undoubtedly other pathways of phospholipid metabolism being activated in these cells (e.g. phospholipase C hydrolysis of phosphatidylinositides, production of diacylglycerol and inositol phosphates, or possibly phospholipase D activation): however, their relevance to arachidonic acid release awaits further purification and characterization of the enzymes and the evaluation of specific inhibitors. The effects observed with manoalogue on both labeled arachidonate release and PGEz production in both unprimed (7) and primed cells (this study) with different stimuli suggest that a phospholipase A2 (20,21) is the most likely phospholipase candidate responsible for the majority of arachidonic acid release.