Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Prostaglandin (PG) endoperoxide synthase is the rate-limiting enzyme for the production of PGs and thromboxanes from arachidonic acid (AA). The enzyme is bifunctional, with both fatty-acid cyclooxygenase (COX) activity (producing PGG₂ from AA) and PG hydroperoxidase activity (converting PGG₂ to H₂) (1). It is now evident that in many cells there are two forms of COX: a constitutive enzyme, designated COX-1, which is important in physiologic functions, and an inducible enzyme, COX-2, which is upregulated in inflammatory cells such as monocytes, macrophages, vascular endothelial cells, and fibroblasts (2). Therefore, COX-2 is believed to play an important role in the production of PGs at sites of inflammation. For example, COX-2 induction in alveolar macrophages (AM) is thought to be potentially very important in relation to inflammation and pathophysiologic states mediated by PGs, including septic shock and the adult respiratory distress syndrome (ARDS) (9).

15-Lipoxygenase converts AA to 15-(S)-hydroxyeicosatetraenoic acid (15-[S]-HETE) and lipoxin A₄, which have been proposed as anti-inflammatory molecules because of their capacities to suppress leukocyte chemotaxis, adherence, and activation, and to antagonize proinflammatory leukotrienes (10).

CD4⁺ T cells are classified into two subsets, Th1 and Th2 cells, according to the pattern of cytokines they produce. Typically, Th1 cells produce interleukin (IL)-2 and interferon- (IFN-), whereas Th2 cells secrete IL-4, IL-5, IL-6, IL-10, and IL-13. These Th1- and Th2-cell-associated cytokines have different functions not only in immune but also in inflammatory responses. IFN- primes human monocyte-macrophages to secondary stimuli (e.g., lipopolysaccharide [LPS]) to produce proinflammatory cytokines (IL-1 and tumor necrosis factor [TNF]), enhancing the inflammatory response (14, 15). In contrast, IL-4, IL-10, and IL-13 suppress the production of proinflammatory cytokines such as IL-1, IL-1, IL-6, IL-8, TNF-, granulocyte colony-stimulating factor (G-CSF), and macrophage inflammatory protein 1 (MIP-1), and increase the production of IL-1 receptor antagonist (IL-1ra) by human monocytes and macrophages (16). There are recent reports that IL-4, IL-10, and IL-13 downregulate COX-2 expression and PG production in human monocytes (24), and that IL-4 and IL-13 upregulate 15-lipoxygenase expression in human monocytes (27). These findings suggest that IL-4, IL-10, and IL-13 are anti-inflammatory agents.

We have reported that IL-4, IL-10, and IL-13 downregulate COX-2 expression and PG production, whereas IFN- does not affect them in human monocytes (26). Since AM are thought to play an important role in several pathophysiologic events in the alveolar space by producing PG, we thought it necessary to clarify the regulatory mechanism of PG production in AM. Little is known about the regulatory effects of Th1- and Th2-cell-associated cytokines on the expression of COX and on PG production in AM. In addition, there is little information on differences in the AA metabolic pathways in AM and monocytes. Therefore, in this study we compared the regulatory effects of Th2-cell-associated cytokines on COX-2 and 15-lipoxygenase in AM and monocytes derived from the same individuals.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents

Fetal calf serum (FCS) and RPMI 1640 were purchased from GIBCO BRL (Gaithersburg, MD) and Nissui Chemical Co. (Tokyo, Japan), respectively. RPMI 1640 supplemented with FCS (10%), glutamine (1 mM), and gentamicin (50 µg/ml) was used in all experiments. Recombinant human IL-4 (specific activity: 10⁶ U/mg of protein) and recombinant human IFN- (specific activity: 5.36 × 10⁶ U/ mg of protein) were kindly provided by Ono Pharmaceutical Co. (Osaka, Japan) and Nippon Roche (Tokyo, Japan), respectively. Recombinant human IL-10 (specific activity: 10⁷ U/mg of protein) and recombinant human IL-13 (specific activity: 1.62 ×10⁶ U/mg of protein) were kind gifts from DNAX (Palo Alto, CA). None of these materials contained endotoxins, as evaluated with the Limulus amoebocyte assay (Seikagaku Kogyo Co., Tokyo, Japan). LPS derived from the Escherichia coli 055:B5 strain was purchased from Difco Laboratories (Detroit, MI). (5,6,8,11, 12,14,15-³H[N])PGE₂ (5.698 TBq/mmol), and unlabeled PGE₂ used in the radioimmunoassay (RIA) were purchased from DuPont/NEN Research Products (Boston, MA) and Funakoshi Co. (Tokyo, Japan), respectively. Anti-PGE₂ antibody, prepared as described previously (28), was kindly provided by Dr. Shozo Yamamoto, of the Department of Biochemistry of our institution and was used in the RIA. Rabbit polyclonal antibody against human COX-2 and human 15-lipoxygenase complementary DNA (cDNA) probe were purchased from Cayman (Ann Arbor, MI). Mouse monoclonal antibody against human COX-1 (hPES01), prepared as described previously (29), was also kindly provided by Dr. Shozo Yamamoto. NS-398 was donated by Taisho Pharmaceutical Co. (Saitama, Japan).

Subjects

AM were obtained from healthy volunteers (21 to 35 yr of age). These subjects had no evidence by history or physical examination of lung disease, were nonsmokers, were not taking medication, and had normal chest roentgenograms. All gave written informed consent to participate in the experiments.

Isolation of AM and In Vitro Stimulation

Bronchoalveolar lavage (BAL) was performed as described in detail elsewhere (30). Briefly, the oral cavity and the upper airway were anesthetized with lidocaine spray, and the tip of an Olympus fiberoptic bronchoscope (Model 1T20; Olympus Co., Tokyo, Japan) was wedged into a segment of the right lung. The lung was washed with 50 ml of sterilized saline (0.9% NaCl) prewarmed to 37°C, and the fluid was gently sucked out with a 50-ml syringe. This process was repeated three times. Subsquently, the same operation was also performed in the left lung. A total of 300 ml of saline was instilled, of which about 65% was recovered. The total number of AM was approximately 2.5 × 10⁷ cells (> 92% viable as determined by trypan blue dye exclusion). More than 89% of these cells were AM as judged by morphology and staining for nonspecific esterase. The other cells were either small mononuclear cells or neutrophils, which were eliminated during subsequent washing. The cells were suspended in RPMI 1640 medium and were plated in 24-well plastic plates (AM density: 2 × 10⁵ cells/well) or 60-mm plastic dishes (AM density: 2 × 10⁶ cells/plate). Nonadherent cells (< 10%) were removed 1 h after plating by gentle washing. The adherent cells (AM) were then incubated in RPMI 1640 containing 10% FCS supplemented with test compounds.

Isolation of Monocytes and In Vitro Stimulation

Mononuclear cells were separated from leukocyte concentrates obtained from healthy volunteers by density gradient centrifugation in lymphocyte separation medium. Monocytes were then isolated by counterflow centrifugal elutriation with a Beckman JE-5.0 system (Beckman Instruments, Fullerton, CA) as described previously (31). Usually, the purity of the monocyte fraction was over 95%, as judged by morphologic examination and nonspecific esterase staining, and cell viability was over 98% as assessed by the trypan blue dye exclusion test. Monocytes were suspended in RPMI 1640 containing 10% FCS, and were incubated in 24-well plastic plates (monocyte density: 2 × 10⁵ cells/well) or 60-mm plastic dishes (monocyte density: 5 × 10⁶ cells/ plate) in medium containing test compounds.

Determination of PGE₂ Synthesis

AM or monocytes were incubated in 24-well plastic plates in medium with a compound to be tested for 24 h. Culture media were then harvested, transferred to polypropylene tubes, and centrifuged at 400 × g for 10 min. The cell-free supernatants were removed and stored at 30°C until determination of PGE₂ content by RIA. RIA of PGE₂ was performed as described previously (32). The sensitivity of the assay for PGE₂ was 70 pg/ml. As previously described (28), the cross-reactivities of anti-PGE₂ antibodies were as follows: PGE₁ (7.0%), PGF₂ (4.3%), and 6-keto-PGF₁ (5.4%), and other arachidonate metabolites (less than 1%).

COX Assay

AM were incubated in 60-mm culture dishes in medium with a compound to be tested for 24 h. The cells were then scraped off the dishes, washed three times with PBS, suspended in 20 mM Tris-HCl (pH 7.4) containing 5 mM tryptophan, and sonicated twice at 20 kHz for 5 s each. The COX reaction was done by incubation of the sonicated cells (65 µl) at 24°C for 2 min in the standard mixture (100 µl) containing 10 µM [1-¹⁴C]AA (Amersham International, Bucks, UK; 2.1 GBq/mmol), 100 mM Tris-HCl (pH 8.0), 2 µM hematin, and 5 mM tryptophan. Ether extracts of the products were spotted onto precoated silica-gel-60 F254 glass plates for thin layer chromatography (Merck, Darmstadt, Germany), and were separated with a solvent system of ethyl acetate/trimethylpentane/acetic acid/H₂O (110: 50:20:100, vol/vol/vol). The conversion rate of [1-¹⁴C]AA to COX products was determined with a Fujix bioimaging analyzer BAS 2000 (Tokyo, Japan).

Western Blotting

AM were stimulated as for COX assay, and were sonicated as described previously. The cell lysate was mixed with an equal volume of loading buffer (125 mM Tris-HCl, pH 6.8; 4% sodium dodecyl sulfate [SDS]; 2 mg/ml methyl green; 10% glycerol), and subjected to 10% polyacrylamide gel electrophoresis (PAGE) in SDS running buffer (25 mM Tris, pH 8.3; 192 mM glycine; 0.1% SDS). The protein bands were then transferred to an Immobilon-P membrane (Millipore, Bedford, MA) in buffer consisting of 25 mM Tris, pH 8.3; 192 mM glycine; 0.1% SDS; and 20% methanol. The blots were pretreated with 5% goat serum, and were then incubated with rabbit polyclonal antibody raised against human COX-2 as primary antibody, and with biotinylated goat antirabbit IgG as secondary antibody. For Western blot analysis of COX-1 protein, the blots were pretreated with 5% horse serum and then incubated with mouse monoclonal antibody raised against human COX-1 as primary antibody and with biotinylated horse antimouse IgG as secondary antibody. The enzyme protein was visualized by use of the Vectastain avidin-biotin conjugate (ABC) kit (Vector, Burlingame, CA) according to the manufacturer's instructions. PBS containing 0.05% Tween 20 was used as washing buffer throughout the experiment.

Northern Blotting

AM or monocytes were stimulated for 24 h with a compound to be tested as described previously. The cells were then scraped off the dishes and homogenized in a mixture of guanidinium isothiocyanate and phenol (ISOGEN; (Nippon Gene, Tokyo, Japan). Total RNA was extracted with chloroform and precipitated with isopropanol. The denatured RNA was subjected to electrophoresis on 1% agarose formaldehyde gel, transferred to a Hybond-N+ nylon membrane (Amersham International) in 10× saline-sodium phosphate-EDTA (SSPE) transfer buffer (1× SSPE: 150 mM NaCl, 10 mM sodium phosphate, 1 mM ethylenediamine tetraacetic acid [EDTA]), and fixed by shaking the membrane in 0.05 N NaOH for 5 min. The membrane was prehybridized at 42°C for 4 h in buffer containing 50% formamide, 0.5% SDS, 5× SSPE, 5× Denhardt's mixture (1× Denhardt's mixture: 0.02% bovine serum albumin [BSA], 0.02% Ficoll, 0.02% polyvinylpyrollidone), and 200 µg/ml salmon sperm DNA. The blots were hybridized at 42°C for 12 h with ³²P-labeled probes (10⁶ cpm/ml) in buffer containing 50% formamide, 0.5% SDS, 5× SSPE, 5× Denhardt's mixture, and 200 µg/ml salmon sperm DNA. The blot was washed twice with 0.1× standard saline citrate (SSC)-0.1% SDS-1 mM EDTA (1× SSC: 150 mM NaCl, 15 mM sodium citrate) for 5 min at room temperature and then with the same solution for 5 min at 65°C, and were autoradiographed with a Fujix bioimaging analyzer BAS 2000. Stripping for rehybridization was done by washing the membrane in 0.5% SDS for 5 min at 95°C. The probes used were cDNA fragments of human COX-1 and COX-2, which were prepared as described previously (8), and -actin and cDNA (Wako, Osaka, Japan). These probes were labeled with [-³²P] deoxycytosine triphosphate ([-³²P]dCTP) (Amersham International; 110 TBq/mmol) using a random primer labeling kit (Takara Shuzo, Kyoto, Japan).

Statistical Analysis

Results are given as means ± SD. Significant differences were calculated through one-way analysis of variance (ANOVA), with post hoc comparisons according to Bonferroni's method. A value of P < 0.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Th1- and Th2-cell-associated Cytokines on PGE₂ Release

The effects of Th1- and Th2-cell-associated cytokines on PGE₂ release by AM into the culture medium were examined. As shown in Figure 1, LPS-stimulated PGE₂ production was significantly inhibited by IL-4 (P < 0.05), but was not affected significantly by IL-10, IL-13, or IFN-. The PGE₂ content of the culture medium of cells incubated in medium alone or with each cytokine alone was below the detection limit. AM were preincubated with IL-10 or IL-13 for 2 h and then washed and incubated with LPS for 24 h. Again, PGE₂ production was not significantly affected by IL-10 or IL-13 (data not shown).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Effects of IL-4, IL-10, IL-13, and IFN- on PGE2 synthesis by AM. AM were incubated with 100 U/ml IL-4, 100 U/ml IL-10, 100 ng/ml IL-13, or 100 U/ml IFN- for 24 h in the presence or absence of 1 µg/ml LPS, and the amount of PGE2 released into the culture medium was determined by RIA. Values are means ± SD of data for cells from three volunteers.

Effects of Th1- and Th2-cell-associated Cytokines on COX Activity

After determination of PGE₂ release from LPS-stimulated or unstimulated AM incubated with each cytokine, the COX activities of the cell lysates were measured with [1-¹⁴C]AA as substrate. Previous experiments showed a time-dependent increase of COX activity in AM stimulated with 1 µg/ ml LPS (data not shown). To determine whether the COX activity observed in the lysate of LPS-stimulated AM was derived from COX-2, we examined the effect of NS-398, a selective inhibitor of COX-2 (33). NS-398 inhibited the LPS-stimulated COX activity dose-dependently, suggesting that most of the COX activity in the LPS-stimulated AM was derived from COX-2 (Figure 2). As expected, indomethacin, a nonspecific COX inhibitor, also inhibited the COX activity dose-dependently. We also investigated the effects of Th1- and Th2-cell-associated cytokines on COX activity. IL-4 significantly inhibited the LPS-induced increase in COX activity (P < 0.05) (Figure 3), suggesting that it inhibited LPS-dependent COX-2 induction in AM. The activity was not significantly affected by IL-10, IL-13, or IFN- (Figure 3).

View larger version (15K): [in this window] [in a new window]   Figure 2.   Effects of NS-398, a specific inhibitor of COX-2, and indomethacin, a nonspecific COX inhibitor, on LPS-stimulated COX activity. AM were incubated for 24 h with LPS (1 µg/ml). The sonicated cells were preincubated with the indicated concentrations of NS-398 or indomethacin for 2 min at 24°C in the standard reaction mixture. The reaction was started by the addition of [1-14C]AA. Values are means ± SD of data for cells from three volunteers. The COX activity in the absence of NS-398 and indomethacin is taken as 100%.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Effects of IL-4, IL-10, IL-13, and IFN- on COX activity of AM. AM were incubated with 100 U/ml IL-4, 100 U/ml IL-10, 100 ng/ml IL-13, or 100 U/ml IFN- for 24 h in the presence or absence of 1 µg/ml LPS. COX activity was determined after sonication of the cells, using [1-14C]AA as substrate, and was expressed as nanomoles of converted [1-14C]AA per milligram of protein in 2 min. Values are means ± SD of data for cells from three volunteers.

Effects of Th1- and Th2-cell-associated Cytokines on COX-2 Protein Levels

To determine whether the suppression of COX activity by IL-4 was due to a reduction in the amount of COX-2 protein, we performed Western blot analysis using a primary antibody against COX-2. As shown in Figure 4, a 71-kD protein band corresponding to COX-2 was evident in the cell lysate from LPS-stimulated AM. COX-2 was not detectable in the cell lysate of control AM or AM treated with IL-4, IL-10, IL-13, or IFN- alone. The LPS-induced increase in COX-2 protein was inhibited by the addition of IL-4 but not affected appreciably by IL-10, IL-13, or IFN-. A 58-kD protein band was also detected, and appeared to represent nonspecifically bound antibody because it was detectable in all AM lysates. In contrast, no protein band corresponding to COX-1 was detectable in the cell lysates (13 µg of protein) of AM incubated with or without LPS or in the presence or absence of IL-4, IL-10, IL-13, or IFN- (data not shown). However, the specificity of hPES01 for COX-1 protein was verified by detecting a 71-kD protein band corresponding to COX-1 in a microsome fraction obtained from human platelets and in IMR-90 cells (human embryonic lung fibroblasts) (Figure 5A). In addition, COX-1 protein was detected in very large amounts (100 µg) of the total protein obtained from resting AM and LPS-stimulated AM (Figure 5A). With primary antibody against COX-2, COX-2 protein was detectable in the cell lysate (100 µg of protein) of resting AM, and was markedly increased upon stimulation with LPS (Figure 5B).

View larger version (31K): [in this window] [in a new window]   Figure 4.   Effects of IL-4, IL-10, IL-13, and IFN- on COX-2 enzyme mass in AM. AM were incubated with 100 U/ml IL-4, 100 U/ml IL-10, 100 ng/ml IL-13, or 100 U/ml IFN- for 24 h in the presence or absence of 1 µg/ml LPS. AM lysates (13 µg of protein) were subjected to SDS-PAGE. Immunoblotting was performed with polyclonal antibody against human COX-2. Positions of molecular-mass marker proteins are indicated by arrows.

View larger version (38K): [in this window] [in a new window]   Figure 5.   COX-1 and -2 protein levels in resting AM, in LPS-stimulated AM, in human platelets, and in IMR-90 cells. AM were incubated for 24 h with or without 1 µg/ml LPS. AM lysate (100 µg of protein ), human platelet microsome, and IMR-90 cell lysate were subjected to SDS-PAGE. Immunoblots were performed with monoclonal antibody against human COX-1 (hPES01) (A) or with polyclonal antibody against human COX-2 (B). Positions of molecular-mass marker proteins are indicated by arrows.

Effects of Th1- and Th2-cell-associated Cytokines on COX-1 and -2 mRNA Levels

We evaluated the effects of Th1- and Th2-cell-associated cytokines on the mRNA levels of the two COX isozymes. Total RNA (3 µg) was extracted from AM incubated with or without LPS in the presence or absence of IL-4, IL-10, IL-13, or IFN-, and was subjected to Northern blot analysis (Figure 6). COX-2 mRNA was undetectable in resting cells, but was induced dramatically on addition of LPS. This induction was suppressed by IL-4, but scarcely affected by IL-10, IL-13, or IFN-. In contrast, with the cDNA for COX-1 as a probe, no band was detected in 3 µg of total RNA from any of the AM preparations. However, when a larger amount (30 µg) of total RNA obtained from resting AM, LPS-stimulated AM, and IMR-90 cells was subjected to Northern blot analysis for COX-1, a very weak signal of a 3.0-kb band was detected in RNA obtained from resting and LPS-stimulated AM, whereas a larger amount of COX-1 mRNA was detected in RNA from IMR-90 cells (Figure 7). In addition, a very weak signal of COX-2 mRNA was detected in RNA obtained from resting AM, and was markedly increased after stimulation by LPS (Figure 7).

View larger version (43K): [in this window] [in a new window]   Figure 6.   Effects of IL-4, IL-10, IL-13, and IFN- on COX-1 and -2 mRNA levels in AM. AM were incubated for 24 h with or without LPS (1 µg/ml) in the presence of 100 U/ml IL-4 (A), 100 U/ml IL-10 (B), 100 ng/ml IL-13 (C), or 100 U/ml IFN- (D). Total RNA (3 µg) was isolated and subjected to Northern blot analysis with cDNA probes complementary to mRNAs of COX-1, COX-2, and -actin for detection of bands. The positions of 28S and 18S ribosomal RNAs (rRNAs) are indicated by arrows.

View larger version (26K): [in this window] [in a new window]   Figure 7.   COX-1 and -2 mRNA levels in resting AM, in LPS-stimulated AM, and in IMR-90 cells. Total RNA (30 µg) was isolated from AM that were incubated for 24 h with or without LPS (1 µg/ml), and from IMR-90 cells. Northern blot analysis was performed for COX-1, COX-2, and -actin. The positions of 28S and 18S rRNAs are indicated by arrows.

Comparison of PGE₂ Production by AM and Monocytes

We found previously that IL-4, IL-10, and IL-13 inhibited LPS-induced PGE₂ production by human monocytes (26). Therefore, we next compared the PGE₂ production in AM and monocytes, and their regulation by Th2-cell-associated cytokines, using AM and monocytes from the same individuals. As shown in Figure 8, the PGE₂ content of the culture medium was below the limit of detection when AM or monocytes were incubated in medium alone, but was siginificantly increased by LPS. LPS-induced PGE₂ production by AM was about 6-fold greater than that by monocytes. Furthermore, IL-4, IL-10, and IL-13 significantly suppressed LPS-induced PGE₂ production by monocytes (P < 0.05), but only IL-4 significantly suppressed that by AM (P < 0.05).

View larger version (14K): [in this window] [in a new window]   Figure 8.   Effects of IL-4, IL-10, and IL-13 on PGE2 synthesis by AM and monocytes derived from the same individuals. AM and monocytes were incubated with 100 U/ml IL-4, 100 U/ml IL-10, or 100 ng/ml IL-13 for 24 h in the presence or absence of 1 µg/ml LPS, and the amount of PGE2 released into the culture medium was determined by RIA. Value are means ± SD of data for cells from three volunteers.

COX-2 and 15-Lipoxygenase mRNA Levels in AM and Monocytes from the Same Individuals

We investigated the COX-2 mRNA levels in the same AM and monocytes as used for evaluation of PGE₂ production. In parallel with PGE₂ production, LPS induced the synthesis of a large amount of COX-2 mRNA in AM and a lesser amount in monocytes, and IL-4, IL-10, and IL-13 significantly suppressed LPS-dependent COX-2 mRNA induction in monocytes, whereas only IL-4 significantly suppressed it in AM (Figure 9). 15-Lipoxygenase catalyzes the C-5 hydroperoxidation of fatty acids, resulting in the generation of anti-inflamatory products. Expression of its mRNA was detected only in monocytes incubated with LPS plus IL-4 (Figure 9).

View larger version (58K): [in this window] [in a new window]   Figure 9.   Effects of IL-4, IL-10, and IL-13 on COX-2 and 15-lipoxygenase mRNA levels in AM and monocytes derived from the same individuals. AM and monocytes were incubated with 100 U/ ml IL-4, 100 U/ml IL-10, or 100 ng/ml IL-13 for 24 h in the presence or absence of 1 µg/ml LPS. Total RNA (10 µg) was isolated and subjected to Northern blot analysis with cDNA probes complementary to mRNAs of COX-2, 15-lipoxygenase (15-LOX), and -actin for detection of bands. The positions of 28S and 18S rRNA are indicated by arrows.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

AM can produce various COX products, such as thromboxane, PGE₂, and PGD₂, among which thromboxane is the major product (34, 35). However, the increase in PGE₂ production has been shown to correlate, in mass, with newly synthesized COX in LPS-stimulated human AM (35). Therefore, in this study, we measured PGE₂ production in order to evaluate the regulation of COX-2 activity in LPS-stimulated human AM.

We have reported that Th2-cell-associated cytokines (IL-4, IL-10, and IL-13) suppress PG production by downregulating COX-2 expression in human monocytes (26). Therefore, we examined whether these cytokines had similar effects on AM. Unexpectedly, of these cytokines, only IL-4 significantly inhibited PG production, COX activity, and COX-2 induction in AM. There are several possible explanations for the difference between the regulations in AM and monocytes. One is a difference in affinity or number of Th2-cell-associated cytokine receptors, because we found that suppressive effects of Th2-cell-associated cytokines on IL-1 and TNF- production in monocytes were significantly greater than those in AM (data not shown). Another explanation is a difference in the signal-transduction pathway for COX-2 induction. Contrary responses are reported for AM and monocytes concerning IFN-- and C-reactive-protein-dependent IL-1ra production (36, 37). Further studies are needed to clarify the mechanisms involved.

Although there are reports that COX-1 mRNA is detectable in AM, the discrepancy between previous results and ours may be due to the sensitivity of the experimental methods or differences in the animal species used. Namely, Lee and colleagues and Chanmugam and associates investigated the expression of COX-1 mRNA using the reverse transcription-polymerase chain reaction (RT-PCR) and ribonuclease (RNase) protection assay, respectively (38, 39), and found relatively large amounts of COX-1 (40).

There are reports that PGE₂ may be a cytoprotective and anti-inflammatory prostanoid (41, 42); however, LPS-stimulated human AM have been shown to produce various prostanoids other than PGE₂ upon induction of COX-2 (35). At a certain stage of inflammation, these COX-2 products may act as proinflammatory factors, because administration of NS-398 can reduce the inflammatory response in vivo (43). 15-(S)-HETE inhibits leukotriene B₄ (LTB₄) generation by leukocytes (44), antagonizes neutrophil chemotaxis by LTB₄ (45), and inhibits leukocyte activation (44, 46). Lipoxin A₄ attenuates LTB₄-induced chemotaxis and decreases leukocyte-endothelial-cell adhesion (10, 13). Because these compounds are produced by 15-lipoxygenation of AA, the inducers of this enzyme are potent anti-inflammatory agents. IL-4 is a leading candidate for such an agent, since it not only downregulates COX-2 but also upregulates 15-lipoxygenase in monocytes stimulated with LPS. Interestingly, Levy and coworkers reported that IL-4 induces 15-lipoxygenase mRNA in AM under spontaneous conditions (47). Furthermore, Nassar and associates reported that IL-4 and IL-13 induce 15-lipoxygenase mRNA individually in monocytes (27).

Yano and coworkers reported that IL-4 inhibits LPS-induced expression of COX-2 in human AM (48). In the present study, we examined the regulatory effects of IL-10, IL-13, and IL-4 on the expression of COX-2 and PG production in AM for the first time, and showed that IL-10 and IL-13 had hardly any affect, whereas IL-4 caused significant suppression of COX-2 and PG production, in accord with Yano and coworkers' observation. In addition, we clearly demonstrated the differential regulation of COX-2 and 15-lipoxygenase in AM and monocytes. We also found that IL-4, IL-10, and IL-13 regulated monocyte chemoattractant protein-1 (MCP-1) production by human peripheral monocytes and AM in different ways (49). Peripheral blood monocytes, believed to be the primary source of AM, migrate from the blood compartment to become residents of the lung air space (50). Unlike blood monocytes, AM are always exposed to inhaled environmental agents and must defend the human lung from invasion by these harmful agents. Because an inflammatory response is a defense of the human body, it may be beneficial that inflammatory COX-2 products are more difficult to downregulate, whereas anti-inflammatory 15-lipoxygenase products are more difficult to upregulate in AM than in monocytes.

IL-4 and IL-10 have been reported to exert protective effects in an animal model of acute lung injury in association with a reduction in the neutrophil content of the alveolar space (53). Our results suggest that these cytokines, and especially IL-4, may suppress the accumulation of neutrophils, consistent with a downregulation of COX-2 products increasing vascular permeability and an upregulation of 15-lipoxygenase products inhibiting neutrophil chemotaxis.

Further study is warranted to clarify the significance of regulation of the AA metabolic pathway in cells of monocyte-macrophage lineage under physiologic or pathologic conditions.