Effects of ozone exposure on lipid metabolism in human alveolar macrophages.

Alveolar macrophages (AM) store arachidonic acid (AA), which is esterified in cellular phospholipids until liberated by phospholipase A2 or C after exposure to inflammatory stimuli. After release, there can be subsequent metabolism of AA into various potent, biologically active mediators including prostaglandins and platelet-activating factor (PAF). To examine the possibility that these mediators may account for some of the pathophysiologic alterations seen in the lung after ozone (O3) exposure, human AM were collected by bronchoalveolar lavage of normal subjects, plated into tissue culture dishes, and the adherent cells were incubated with [3H]AA or [3H]lysoPAF. Human AM exposed to 1.0 ppm O3 for 2 hr released 65 +/- 12% more tritium, derived from [3H]AA, than paired, air-exposed controls into media supernatants. In other studies using a similar O3 exposure protocol, there was also a significant increase in human AM prostaglandin E2 production (2.0 +/- 0.5-fold increase above air-exposure values, p less than 0.01, n = 17). In additional studies, using a similar O3 exposure protocol (1.0 ppm for 1 hr), there was also a significant increase in human AM PAF content (1.7 +/- 0.2-fold increase above air-exposure values, p less than 0.02, n = 5). These potent lipid mediators, originally derived from human AM, may play an important role in the mechanisms of O3 lung toxicity.


Introduction
Ozone (03) is a major photochemical air pollutant that causes deleterious health effects on inhalation (1). This oxidant gas, formed by complex reactions ofnitrous oxides with oxygen, can be found at concentrations exceeding 0.3 ppm in several U.S. areas, with ambient levels generally around 0.01-0.04 ppm. The lung is the main site of 03-induced toxicity, although extrapulmonary effects have been reported. Altered lung function, bronchial hyperresponsiveness, and a lung inflammatory response (measured as the presence ofneutrophils and biochemical mediators of inflammation, e.g., AA metabolites, in bronchoalveolar lavage) have been reported in humans after 03 exposure (2)(3)(4).
An important resident cell of the lung that may be affected by ozone is the alveolar macrophage. These cells are the predominant cell type within the alveolus and serve as the resident mononuclear phagocytes ofthe lung. Alveolar macrophages play an important role in immune and inflammatory processes be- cause of the numerous mediators they secrete in response to phagocytic or inflammatory stimuli (5,6). Therefore, alveolar macrophages serve as the first line ofhost defense against inhaled organisms and soluble and particulate molecules. Ozone exposure has been shown to adversely affect host defense mechanism(s) of animals against certain infectious agents. Inhalation of03 has been reported to result in a significant increase in mortality ofmice exposed to aerosols ofinhaled microorganisms (7). It has been suggested that this effect may be attributable, at least in part, to perturbations in pulmonary alveolar macrophage immune defense mechanisms. Decreased phagocytosis, decreased superoxide anion production, decreased lysosomal enzyme release, and decreased interferon synthesis have been reported for alveolar macrophages after either in vivo or in vitro 03 exposure (8)(9)(10)(11)(12)(13). This observed toxicity of 03 to alveolar macrophage functions may be due in part to the location of alveolar macrophages in the airway lumen, allowing greater exposure to 03 compared to other lung cell types.
Metabolites of the membrane lipid constituents of alveolar macrophages represent an important class of macrophage mediators. Alveolar macrophages can store and metabolize the important lipid mediator arachidonic acid (AA). AA, a 20carbon fatty acid, is usually esterified in cellular phospholipids in most cell Mts until liberated by phospholipase A2 or C activity from membrane phospholipids. Once liberated, AA can serve as a source for the production ofpotent lipid mediators including cyclooxygenase and lipoxygenase products and platelet-activating factor (PAF).
Metabolism ofAA through cyclooxygenase initially generates the oxygenated product prostaglandin endoperoxidase G2. This endoperoxide is enzymatically converted to another cyclic endoperoxide, PGH2, by either cyclooxgenase activity or other peroxidases. PGH2 can be further metabolized to prostacyclin (PGI2) via prostacyclin synthetase activity; thromboxane A2(TxA2) by thrmboxane synthetase catalysis; or other primary prostaglandins (PG) through enzymatic and nonenzymatic means (PGE2, PGF2,). Cyclooxygenase activity can be stimulated by low concentrations ofeither hydroperoxides and/or lipid peroxides (generally 1-10 ytM depending on the cell type), whereas high concentrations (usually 10-100 AM) are inhibitory to activity. In addition, cyclooxygenase is a "suicide" enzyme, i.e., the enzyme is inactivated after conversion of a sufficient amount of substrate into product most probably through formation of an active oxygen species.
Conversion of free AA by the action of lipoxygenases results in the formation ofunstable hydroperoxyeicosatetraenoic acids (HPETEs), generally the 5-, 12-, or 15-isomers in most cell types. These compounds are then metabolized either to their hydroxyor dihydroxy-acid forms, e.g.,15-HETE, or to leukotrienes when 5-HPETE is converted to the unstable intermediate leukotriene A4 (LTA4). The 5-lipoxygenase pathway is a major pathway ofAA metabolism in alveolar macrophages. LTA4 is converted either to LTB4 by enzymatic hydrolysis or to the sulfidopeptide leukotrienes (L1C4, LTD4, LTE4) via the addition and sequential cleavage of glutathione via the action of glutathione-S-transferase, y-glutamyl transpeptidase, and dipeptidase. The leukotrienes C4, D4, and E4, collectively known as "slow-reacting substance of anaphylaxis" (SRS-A), cause bronchoconstriction via smooth muscle contraction and stimulate mucus secretion either individually or in the SRS-A complex.
LIC4 and LTD4 produce coughing and chest tightness in both normal and asthmatic subjects. They also increase vascular permeability. LTB4 is a potent chemotactic agent for PMNs and eosinophils and may play a role in the recruitment of PMNs to inflammatory foci in the lung. It is unclear at present whether lipoxygenases are self-inactivated after metabolism of AA. Unlike cyclooxygenase, lipoxygenases are not known to possess the peroxidase activity for converting their hydroperoxy acid metabolites to the alcohol form.
Thus, there are several potential mechanisms for the release of AA in alveolar macrophages. Depending on the type of AA release mechanism(s), a number of potent and important lipid mediators may be formed, which can play an important role in the inflammatory and physiologic response ofthe lung. We now report that exposure ofhuman alveolar macrophages in vitro to 03 results in significant changes in alveolar macrophage lipid metabolism including release of AA, increased synthesis of PGE2, and increased content of PAF.

Ozone Exposure System
Cells cultures were exposed to predetermined concentrations ofozone or air in 35or 60-mm tissue culture polystyrene petri dishes using an in vitro ozone chamber system (Fig. 1) and consisting of two plexiglass and stainless-steel chambers (15 L in capacity) through which filtered room air at 37C, with or without 03, was passed at a flow rate of7.5 L/min. Separate input lines carried humidified air containing CO2 into the chambers to maintain a 5 % CO2 atmosphere as regulated by 2 CH/P CO2 Analyzers (Forma Scientific, Marietta, OH). The chambers were mounted on rocking platforms (Bellco Glass, Vineland, NJ), which permitted rocking of the petri dishes during exposure. Chambers and platforms were enclosed in 37C incu- FIGURE 1. In vitro ozone exposure system. A, temperature-controlled incubator; B, zero-grade air; C, humidifier; D, ozone generator; E, plexiglass and steel chamber; F, rocker platform: G, condenser; H, ozone analyzer.
OZONE ALTERS ALVEOLAR MACROPHAGE ARACHINDONATE METABOLISM bators (Forma Scientific, Marietta, OH). Ozone was generated by passing the humidified air over a UV lamp; the brightness was regulated by partially covering the lamp with a steel tube until the desired 03 concentration was reached. Ozone concentrations were determined by sampling the chamber atmosphere, dehumidifying it with a condenser, and passing the air into a Model 1003-AH Ozone Analyzer (Dasibi, Glendale, CA), which was periodically calibrated by instrumentation traceable to the National Bureau ofStandards. All lines and fittings in the system were Teflon, glass, or stainless steel. The pH of phosphatebuffered saline (PBS) or Hank's buffered salt solution was found to be unchanged after 2 hr of exposure to 1.0 ppm 03.

Human Alveolar Macrophages
Healthy, nonsmoking male volunteers 18-35 years of age underwent fiberoptic bronchoscopy and bronchoalveolar lavage to obtain normal human alveolar macrophages for use in these experiments. Each subject was remunerated and signed a statement ofconsent after being informed ofthe purpose, procedure, and risks of the bronchoalveolar lavage protocol. The procedure was approved by the Committee on the Rights ofHuman Subjects of the University of North Carolina School of Medicine.
Before bronchoscopy, all subjects were premedicated with 0.5 mg of intravenous atropine. The posterior pharynx was anesthetized by gargling with a 4% lidocaine solution in saline and a lubricating jelly containing 2% lidocaine was placed in the nostril through which the bronchoscope (BF-1T1O, Olympus, New Hyde Park, NY) was passed. In addition, some subjects were administered 25-50 mg of Demorol intravenously. Lidocaine was injected through the bronchoscope at the level of the vocal cords, the carina, and more distal airway bifurcations to decrease coughing. A maximum of20 mL of 2% lidocaine was used. For the lavage, the bronchoscope was wedged in a segmental or subsegmental bronchus ofthe lingula, and 50 mL ofsterile saline at room temperature was slowly injected through the bronchoscope. The saline was immediately aspirated into a 50 mL sterile polypropylene tube, which was then placed on ice. A total ofsix washes were performed for a total of 300 mL of saline instilled. This procedure was then repeated in the right middle lobe with an additional 300 mL of saline. Approximately 30 x 106 total viable alveolar macrophages are recovered from a typical subject, with a recovery of the injected saline of approximately 75 %. The viability ofthe recovered cells exceeded 85 % by trypan blue exclusion.
Immediately after the procedure, the lavage fluid containing the cells was centrifuged at 350g for 10 min at 4°C Cells were then pooled and washed twice in RPMI-1640 supplemented with 0.025 % gentamycin. Between the two washes, cells were counted in a hemocytometer and adjusted to 1 x 106 viable macrophages per milliliter of RPMI 1640. Cell differentials were done on cytocentrifuged slides prepared at 700 rpm for 5 min and stained with a modified Wrightstain (Leukostat Solution, Fisher Scientific, Fairlawn, NJ). A typical differential count yielded 88% alveolar macrophages, 11% lymphocytes, and 1% polymorphonuclear leukocytes.

Radiolabeling of Cells
Alveolar macrophages, in culture flasks, were radiolabeled with 1 yi Ci/mL ofeither [3H]LysoPAF or [3H]AA for 30 min-5 hr in RPMI with 2 % fetal calf serum. After labeling, the cells were washed three times with PBS and then 1.5 mL PBS with calcium, magnesium, and glucose (1 mg/mL) was added and the cells were exposed to either air, 03, or the calcium ionophore A23187 (10 sM). Approximately 70% of [3H]lysoPAF in the medium was taken up by the cells by 30 min of incubation and 50% of the [3H]AA was taken up after 4 hr.

Extraction of Cellular Lipids
Reactions and cell exposures were termiinated by scraping the cells from the dishes with a rubber scraper and transferring them to a polypropylene tube containing chloroform, rinsing the dish with methanol, and pooling the rinse with the chloroform/cell mix. The proportions of solvents were always 0.8:1.0:2.0 aqueous:methanol:chloroform, as described by Folch et al. for the extraction of lipids from tissues (23). In some cases, the chloroform contained a few hundred disintegrations per minute of [14C]PAF as an internal stndard. After mixing, the tubes were centrifuged at 500g for 2 min to separate the phases and the chloroform phase was removed with a pasteur pipette and placed it in a siliconized disposable glass tube. A second extraction was performed on the aqueous/methanol phase with an additional two volumes of chloroform, and the chloroform phases were pooled and evaporated in a 370C heating block under a stream of N2. The resulting lipid residue was redissolved in a small volume of chloroform for HPLC or TLC.
Fgh-Performance Liquid Chromatography iWo different HPLC methods were employed, one specifically designed for the separation ofphospholipids and another for the separation ofAA metabolites. The instrumentation used was the same for all methods and consisted of two Waters 510 HPLC Pumps, a U6K manual injector, a System Interface Module, a 490E Multiwavelength Detector, a NEC APC IV computer, and the 820 Maxima software program (Witers Millipore, Millford, MA). Detection of radioactivity in the HPLC effluent was monitored in a CR Flow ONE Beta Radioactive Flow Detector with a DIX Data 1000 computer (Radiomatic Instnuments, Tampa, FL). Scintiverse LC liquid scintillation cocktail was used at a mixing ratio of 3:1.

Phospholipid Separation
The separation method is described in greater detail in another publication (14). The method uses two 25 cm x 4.8 mm cyanopropyl columns in series, connected with a 10-cm length of0.010 in of stainless-steel tubing and protected by an LC-CN guard column cartridge. The mobile phase, solvents A and B, consisted of 100% acetonitrile and 80% acetonitrile/20% 5 mM sodium acetate (pH 5.0), respectively, and were filtered and degassed under vacuum before use. Solvent B was prepared daily to prevent precipitation ofsalts. The mobile phase and columns were kept at 60°C via a circulating water bath to facilitate the separation of the acidic phospholipids and reduce the back pressure in the system. A gradient was used at a flow rate of 2 mL/min, as follows: from 0 to 5.0 min 10% B, from 5.1 to 12.0 min increasing to 75 % B, from 12.1 to 30.0 min maintaining 75 % B, from 30.1 to 30.2 min decreasing to 10% B, and from 30.3 to 45.0 re-equilibrating at 10% B. The columns were regenerated regularly with 50 mL of 30% acetonitrile, followed by 200 mL of 100 % acetonitrile, and the guard column cartridges were replaced periodically. Peaks of interest isolated with this HPLC system were evaporated under N2, redissolved in chloroform, methanol and water, and extracted. The chloroform phases were then dried under N2 and redissolved in the desired solvent.

HPLC Analysis of [3lHArachidonic Acid Metabolites
Media were analyzed using a previously published HPLC procedure (15). Just before analysis, samples were brought to room temperature, mixed (30 sec), and centrifuged at 2000g for 6 min.
An aliquot (750 1sL) ofthe supernatant was then injected into the HPLC system employing an an Alltech Ultrasphere ODS 5 ,m reverse-phase column (Rainin Instrument Co., Woburn,MA) preceded by a precolumn filter (0.5 jAm frit; Upchurch Scientific, Oak Harbor, WA). Pump A delivered a water/methanol/acetic acid (9/1/0.01 v/v/v) solution, pH 5.05, prepared by titration with concentrated NH40H: pump B delivered 100% methanol to the system. Witha flow rate of 1.1 mL/minthroughout the run, agradient elution was utilized with linear changes in eluant composition at the following time points: 0 min, 47% pump A; 27 min, 40% pump A; 52 min, 27% pump A; 74 min, 0% pump A; 101 min, re-equilibration to initial conditions for 15 min. Metabolite peaks were identified as eicosanoids based on the retention time compared to externally applied authentic standards. Radioactivity associated with peaks were corrected for baseline radioactivity.

Thin Layer Chromatography
TLC was used to isolate, quantify, and identify PAF according to the method described by Chilton et al. (16). The chromatography was carried out on silica gel 60 plates that were heat activated at 120°C for at least 1 hr before use. Samples were spotted under a stream of warm air from a blow drier 2 cm from the bottom ofthe plate in 20-50 AL ofchloroform with a microliter syringe. The plate was reheated at 120°C for 3 min, cooled to room temperature, and developed to within 1 cm from the top. The development tank was lined with 3MM Whatman paper and equilibrated with a solvent mix consisting of 100 mL of chloroform, 50 mL of methanol, 16 mL ofglacial acetic acid, and 8 mL of water. After development, the plate was allowed to air dry under a fume hood and the lipid bands were visualized in a tank containing sublimed I2 crystals. For quantification ofradioactive species, the bands were scraped into scintillation vials, 1 mL of methanol:water 1:1 and 3 mL of Scintiverse LC were added, and the vials were counted for 5 min in a TRI-CARB 1500 Liquid Scintillation analyzer (Packard, Sterling, VA), with a quenchcorrecting program capable of discriminating between 3H and "4C disintegrations per minute.
Data were analyzed using t-tests for paired or unpaired variates. All data are expressed as means ± standard error of the mean.

Results and Discussion
Human alveolar macrophages were plated into tissue culture dishes and the adherent cells ( > 95 % plating efficiency) were incubated with 3HJAA for 1-24 hr. A 4-hr period is sufficient to equilibrate [ H]AA into cellular lipid pools within alveolar macrophages as assessed by HPLC (Thble 1). There were no major changes in the tritium associated with each cellular pool after 2 hr of [3H]AA labeling, with the possible exception of the radioactivity associated with triglycerides, which decreased from containing 29% ofthe tritium at 1 hr to 15% at 24 hr and an increase in the combined PE + LPE pool from 5 to 28 %. Phosphatidylcholine was the majar reservoir of [3H]AA at 1-16 hr (> 24% at all time points sampled).
After incorporation of [3HJAA into cellular pools, human alveolar macrophages were washed to remove unincorporated label, 0.5 mLphosphate-buffered saline with glucose (1 mg/mL; PBS-glucose) was added, and then alveolar macrophages were exposed to air or 1.0 ppm 03 for 2 hr. After exposure, media was centrifuged and an aliquot counted for radioactivity. Human alveolar macrophages exposed to 03 released 65 ± 12% more tritium, derived from [3H]AA, than paired air-exposed controls into media supernatants ( Table 2). Human peripheral blood monocytes were prelabeled with [3H]AA and exposed to 03 in a similar manner. This cell type had a similar response as the alveolar macrophage (Table 2). Both cell types remained viable throughout the exposures with no difference in viabilities ( > 90% trypan blue exclusion) between 03-exposed and airexposed cells. Approximately 10% ofthe cells detached during either air or 03 exposure. In a separate study, there appeared to be a continuous increase in release of tritiuni over the 2-hr exposure period in the ozone group (1.42 ± 0.39% at 15,min; 2.03 ± 0.32% at 30 min; 3.14 ± 1.21% at 160 min; and 3.58 ± 1.13% 20 min) compared to the air-exposed cells. The values at the 30and 120-time points were significantly higher than those obtained in paired cultures exposed to air alone for similar time periods (p< 0.05;n=4-5).  *Significant difference from air-exposed value, p < 0.05. HPLC analysis of media supematants from human O3-exposed alveolar macrophages revealed that the released radioactivity was associated with HETEs, free AA, and the highest radioactivity was associated with a polar peak that eluted with a similar retention time as 03-exposed AA. This lack ofchange in eicosanoid content compared to our previously published data in rat macrophages, using similar exposure and separation methods (17), may have been due to an insufflcient number of alveolar macrophages to synthesize detectable AA metabolites (< 5 x 10 human macrophages were used per exposure group compared to > 5 x 10 rat macrophages used previously), degradation of formed AA metabolite, or AA itself into other derivatives, and/or methodological reasons. Therefore, a more sensitive technique, radioimmunoassay, was used to determine eicosanoid concentrations in media supernatants of alveolar macrophages exposed to air or 03. This technique can detect eicosanoids at approximately 100 pg/mL concentrations. Human alveolar macrophages were cultured as previously described, washed to remove serum, 0.5 mL PBS-glucose was added, and the cells were exposed to air or 03(O.1-1.O ppm for 2 hr). After exposure, medium was collected, centrifuged, and medium supernatants analyzed by radioimmunoassay (RIA) for specific metabolites. The results are shown in Table 3. Alveolar macrophages exposed to 03 produced more PGE2 than paired, air controls in a concentration-dependent manner, with a significant 2.03-fold increase observed in the 1.0-ppm exposed cultures (p < 0.05). No differences were observed in the formation of TxB2 or LTB4 between the air-exposed and O3-exposed alveolar macrophages.
To further determine whether alterations in AA metabolism were present after 03 exposure, immediately after exposure to air or 03 (0.1-1.0) for 2 hr, the cells were washed and then incubated (at 37°C with 5 % C02) with 1.0 mL serum-free RPMI containing 10 zM calcium ionophore A23187 (in 0.13% DMSO) for 30 min. The medium was collected, centrifuged, and supernatants analyzed by RIA for eicosanoid content. Results are shown in Table 4. These data show that 03-exposed ¶lble 3. Human alveolar macrophage eicosanoid production in response to 03 exposure (2 hr). 'Data are shown as means ± SEM; n in parentheses. *Significant difference from paired, air control value, p < 0.05. alveolar macrophages still could produce eicosanoids, similar to air-exposed cultures in response to calcium ionophore incubation. These data suggest that the observed 03-dependent increase in alveolar macrophage PGE2 synthesis was not via a calciumdependent mechanism.
Ozone exposure has been shown to alter AA metabolism in the lung. A 10-fold increase in AA was found in bronchoalveolar lavage fluid of rats exposed to 1.1 ppm 03 (24). Seltzer and associates found increased levels of prostaglandins E2 and F20, and thromboxane B2 (a product of thromboxane A2) in bronchoalveolar lavage fluid of subjects exposed to 0.4 ppm 03 for 2 hr (2). A 2-fold increase in prostaglandin E2 in the lavage fluid of subjects exposed to 0.4 ppm for 2 hr was reported by Koren et al. (3). Further evidence ofthe significance ofeicosanoids in 03 toxicity was provided by studies reporting protection against 03-induced decreases in functional volume capacity and forced expiratory volume in 1 sec in subjects administered the cyclooxygenase inhibitor indomethacin before exposure to 0.35 ppm 03 for 1 hr (18). In vitro studies have also shown AA and eicosanoid release from lung cells exposed to 03. Bovine endothelial cells exposed to 0.3 ppm 03 for 2 hr were shown to release AA (24), whereas bovine epithelial cells exposed to 0.1 ppm 03 for 2 hr released increased amounts of prostaglandin F2, (19).
The effect ofin vitro 03 exposure on eicosanoid release from alveolar macrophages has also been studied. Rabbit alveolar macrophages released prostaglandin E2 and other cyclooxygenase products when exposed to 0.3 and 1.2 ppm 03 for 2 hr (20). A 3.3-fold increase in AA release was found in rat alveolar macrophages exposed to 1.0 ppm 03 for 2 hr. These authors also found increases in thromboxane B2 and leukotrienes B4, C4, and D4 (17). These studies indicate that 03 can alter the amounts of AA metabolites biosynthesized as well as increase the release of the parent compound, AA. Reports indicating 03-induced alterations of AA product formation suggest that these alterations are due to modifications in enzymatic activities (17).
As reviewed, macrophages can respond to a variety ofstimuli by production of another lipid metabolite, PAF. The effects of a 2-hr in vitro exposure to human alveolar macrophages, prelabeled with precursor [3H]lysoPAF, are shown in Table 5.
No changes in viability were seen in cells exposed to 1.0 ppm 03 compared to air-exposed cells for 60 min. As shown, human alveolar macrophages obtained from healthy subjects, radiolabeled and exposed in culture to 1.0 ppm 0 for 60 min, had a 1.7-fold significant increase in the level of [ H]PAF compared to human alveolar macrophages exposed to air alone (p < 0.002, n = 5). In contrast, human alveolar macrophages incubated with 10 ,uM A23187 for 60 min had nonsignificant in-   A2 to form 1-0-alkyl-2-lyso-GPC, which is in turn acetylated by an acetyltransferase to produce PAF (1-0-alkyl-2-acetyl-GPC).
There is evidence that the PAF acetyltransferase enzyme can be activated in some stimulated cells through a biochemical mechanism involving protein phosphorylation by serine/threonine kinases and suggest a role of protein kinase C, a Ca -dependent and phospholipid-dependent protein kinase, which is a pivotal regulatory element in signal transduction. PAF is degraded by PAF-acetylhydrolase, an enzyme that is highly selective for phospholipids with short acyl groups at the sn-2 position and thus catalyzes the inactivation of PAF by removal of the acetate moiety from PAF to form alkyllyso-GPC, which is biologically inactive. Interestingly, it has recently been shown that oxidized phospholipids, such as those that could occur in cells after 03 exposure, are degraded by the PAF acetylhydrolase (25).
The metabolism and biologic effects of these two classes of lipid mediators, eicosanoids and PAF, may also be interrelated. The release of AA from the sn-2 postion of phospholipids is postulated to be the first step in the production ofcylcooxygenase and lipoxygenase products as well as PAF. Alveolar macrophages are rich in esterified AA, and the release appears to be calcium mediated, including phospholipase A2-mediated deacylation, phospholipase C action, and enzymatic processes governing formation of diacyl glycerol, cardiolipin, and phosphocholine/ ethanolamine from phosphatidylcholine (PC) and phosphatidylethanolamine (PE). However, there is a complex pattern of AA turnover in phospholipids of alveolar as shown in Figure 2. In unstimulated cells, AA is predominantly incorporated into PI and the diacyl species of PE and PC and then mobilized to the ether-linked species of PE and PC. In stimulated cells, several phospholipid species (primarily phosphoinositides and etherand ester-linked species ofPC) serve as sources ofAA for oxygenation into various bioactive mediators. Thus, in stimulated cells both phospholipase A2 and phospholipase C are activated. Of the total label lost from PC in response to inflammatory stimuli such as lipopolysaccharide or opsonized zymosan, half is derived from the I-alkyl-linked species and the rest from 1acyl-linked species. The phospholipase A2 enzyme can hydrolyze alkyl-linked PC, providing a substrate for PAF production. The phospholipase C can activate phosphoinositide biphosphate to produce diacylglycerol and inositol phosphates. These reactions may involve specific receptors that link G-proteins, protein kinases, phospholipase A2 and phospholipase C, phosphatidylinositol turnover, cytosolic Ca 2, Ca -dependent effectors such as calmodulin, diacylglycerol, and monoglycerol lipase, to the production ofbioactive AA metabolites and PAR Thus, there are several potential mechanisms for the release and subsequent metabolism of AA in alveolar macrophages in response to 03.
In summary, our results demonstrate that 03 exposure results in a concentration-dependent release of AA from cell membranes of human alveolar macrophages with subsequent increased synthesis ofseveral important lipid mediators including PGE2 and PAF. These lipid mediators formed in response to 03 may play an important role in the inflammatory and physiologic response of the lung to 03 exposure. Based on the pathways depicted in Figure 2, these data suggest that 03 may produce cellular responses through one or more enzymatic pathways that may be receptor mediated. Future work directed at the cellular processes regulating the biosynthesis and secretion ofpotent lipid mediators will further the understanding of the response of the lung to inhaled toxicants such as 03. This work was supported by U.S. Environmental Protection Agency Cooperative Agreement CR812738 and National Institutes of Health grants ES04951 and ES07126. The research described in this article has been reviewed by the Health Effects Research Laboratory, U.S. Environmental Protection Agency and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies ofthe Agency nor does mention oftrade names or commercial products constitute endorsement or recommendation for use.