(Received for publication, April 18, 1996, and in revised form, October 5, 1996)
From the Department of Immunopharmacology, SmithKline Beecham Pharmaceuticals, King of Prussia, Pennsylvania 19406
Human monocytes possess several acylhydrolase activities and are capable of producing both prostanoids (PG) and leukotriene (LT) products upon acute stimulation with calcium ionophore, A23187 or phagocytosis of zymosan particles. The cytosolic 85-kDa phospholipase (PLA) A2 co-exists with the 14-kDa PLA2 in the human monocyte, but their respective roles in LT production are not well understood. Reduction in 85-kDa PLA2 cellular protein levels by initiation site-directed antisense (SK 7111) or exposure to the 85-kDa PLA2 inhibitor, arachidonyl trifluoromethyl ketone (AACOCF3), prevented A23187 or zymosan-stimulated monocyte prostanoid formation. In contrast, neither treatment altered stimulated LTC4 production. This confirmed the important role of the 85-kDa PLA2 in prostanoid formation but suggests that it has less of a role in LT biosynthesis. Alternatively, treatment of monocytes with the selective, active site-directed 14-kDa PLA2 inhibitor, SB 203347, prior to stimulation had no effect on prostanoid formation at concentrations that totally inhibited LT formation. Addition of 20 µM exogenous arachidonic acid to monocytes exposed to SK 7111 or SB 203347 did not alter A23187-induced PGE2 or LTC4 generation, respectively, indicating that these agents had no effect on downstream arachidonic acid-metabolizing enzymes in this setting. Taken together, these results provide evidence that the 85-kDa PLA2 may play a more significant role in the formation of PG than LT. Further, utilization of SB 203347 provides intriguing data to form the hypothesis that a non-85-kDa PLA2 sn-2 acyl hydrolase, possibly the 14-kDa PLA2, may provide substrate for LT formation.
Much work has been directed toward understanding the liberation of arachidonic acid (AA)1 from human monocyte phospholipids (PL) and its subsequent metabolism to a number of cyclooxygenase (COX) and 5-lipoxygenase (5-LO) products (1, 2, 3, 4). The first rate-limiting enzyme in eicosanoid formation is phospholipase A2 (PLA2, EC 3.1.1.4), which liberates AA from the sn-2 position of cellular PL (5, 6). The two most studied mammalian forms are the type II 14-kDa PLA2, known to exist as both an extracellular (7, 8) and cell-associated form (9, 10, 11) and the cytosolic 85-kDa-PLA2 (12, 13). Although both enzymes have been extensively studied, the relative contribution of the two enzymes in stimulated eicosanoid production in a single cell system where they co-exist, as cell-associated forms, is poorly understood.
Correlative evidence exists for the participation of the 85-kDa PLA2 in growth factor or cytokine-mediated AA liberation and prostanoid formation (14, 15, 16, 17). More direct evidence for a role in prostanoid formation has been obtained through selective inhibitors of the 85-kDa PLA2 activity or modulation of enzyme levels by antisense oligonucleotides (18, 19, 20, 21). Arachidonyl trifluoromethyl ketone (AACOCF3), a slow, tight-binding inhibitor of 85-kDa PLA2, reduces stimulated AA release from platelets, U937 monocytes, and mesangial cells and subsequent platelet thromboxane (TXB2) and 12 hydroxyeicosatetraenoic acid biosynthesis (18, 19, 20). AACOCF3 is selective with respect to other PLA2 enzymes and does not have an effect on CoA-independent transacylase (18). It has, however, been reported to directly inhibit COX activity (20), and it must be used with caution in the evaluation of 85-kDa PLA2 function in prostanoid biosynthesis. Its effects on the 5-LO pathway have not been reported. We have previously demonstrated the specific action of initiation site-directed 85-kDa PLA2 antisense (SK 7111) in the reduction of endotoxin-stimulated human monocyte 85-kDa PLA2 protein and enzyme activity levels without altering the cell-associated type II 14-kDa PLA2 or COX II (21). This resulted in a concentration-dependent reduction in PGE2 formation. Correlative evidence, such as coordinate enzyme expression (22, 23, 24, 25, 26) or subcellular localization (27, 28), has been reported, implicating the 85-kDa PLA2 as a possible participant in 5-LO product formation; however, there are no reports directly linking the two.
The role of the cell-associated 14-kDa PLA2 has been studied employing structurally distinct 14-kDa PLA2 inhibitors, e.g. scalaradial (29, 30), BMS-181162 (31), WAY-125984 (32), and the active site-directed inhibitor, SB 203347 (33). Use of these agents indicates that the type II 14-kDa PLA2 does participate in cellular AA metabolism, as exemplified by their ability to inhibit stimulated neutrophil AA release and both leukotriene (LT) B4 and platelet-activating factor biosynthesis. Since neutrophils do not produce prostaglandins (PG), neither 85- nor 14-kDa PLA2 inhibitors can be assessed for their effects on prostanoid biosynthesis in this system. When studied in cells, such as monocytes, which produce both LT and PG, 14-kDa PLA2 inhibitors such as scalaradial have been shown to inhibit stimulated LTC4 production but have no effect on PGE2 production (29). Failure of 14-kDa PLA2 inhibitors to reduce prostanoid synthesis has also been reported in cell systems that produce predominantly prostanoids and little or no 5-LO products, e.g. peritoneal guinea pig macrophage PGE2 (34), human keratinocyte PGD2 (16), or endotoxin-induced human monocyte PGE2 production (21). Interestingly, exceptions to this exist. Antisense designed against the murine type II 14-kDa PLA2 reduced PGE2 production by an activated murine macrophage cell line, which may indicate species or cell line differences (35). Further, there are cell systems where both the 85-kDa and the type II 14-kDa PLA2 are implicated in prostanoid formation, e.g. stimulated mesangial cells (36, 37), mast cells (38), or endothelial cells (39). In these models, the 14-kDa PLA2 has been studied primarily as an extracellular enzyme. Application of the 14-kDa PLA2 inhibitor, CGP 43182, in the stimulated rat mesangial cell system resulted in marked attenuation of PGE2 production (36), illustrating alternative functional roles for secreted 14-kDa PLA2. This could be due, in part, to neutralization of the extracellular function of this enzyme (85-90% of the 14-kDa PLA2 is secreted in this system) and possible interference with its interaction on a cell surface receptor. This in turn could prevent activation of cellular 85-kDa PLA2, which has been hypothesized to occur in the mast cell models (40). Taken together, it appears that both the cell-associated and secreted forms of the type II 14-kDa PLA2 are important in cellular AA metabolism but may act through distinct pathways.
The bulk of the studies described above have been performed in cell systems where only prostanoids or leukotrienes are generated. As such the participation of the two distinct, cell-associated sn-2 acylhydrolases may not be fully appreciated. The monocyte/macrophage possess several acylhydrolase activities, including the 14- and 85-kDa PLA2 enzymes (11, 12, 41, 42, 43). They offer an optimal system for studying the respective roles of the two enzymes on eicosanoid synthesis, because they simultaneously produce both LT and PG products upon stimulation with soluble or receptor-mediated stimuli (4). In addition, we have reported previously that monocytes do not secrete the 14-kDa PLA2, even with endotoxin treatment (21), and therefore the cellular form can be exclusively evaluated in these acute activating systems. Here we report the utilization of 85-kDa PLA2 initiation site antisense (SK 7111), AACOCF3, the 85-kDa PLA2 inhibitor, and the selective 14-kDa PLA2 inhibitor, SB 203347 to provide data to suggest that the 85-kDa PLA2 and the 14-kDa PLA2 both provide AA substrate for stimulated human monocyte eicosanoid biosynthesis, but possibly for distinct metabolizing pathways.
Monocytes (5 × 106/ml) isolated as described previously (21) were incubated in RPMI 1640 medium (Life Technologies, Inc.) containing the treatment and/or the relevant vehicle after which the stimulus was added. The amount of stimulus was chosen from the linear portion of a concentration versus product curve usually representing 40-70% maximal stimulation as described previously (A23187, 1 µM (7-15 min) or opsonized zymosan, 5 mg/ml (2 h)) (21, 29). Both prostanoid and leukotriene production in response to the respective stimuli were submaximal, representing the linear portion of a product versus time curve.
In antisense studies, monocytes were exposed to phosphorothioate
oligonucleotides SK7111 (3-TACAGTAAATATCTAGGAATG-5
, directed against
the initiation site) or SK9030 (5
-ATGTCATTTATAGATCCTTAC-3
, sense) or
Lipofectin vehicle, alone (5 µg/ml) as described previously (21), in
serum-free conditions for 18 h prior to stimulation unless
otherwise stated. In drug studies, cells were pretreated with dimethyl
sulfoxide (Me2SO, <1%) vehicle or appropriate
concentrations of AACOCF3, purchased from Cayman Chemical
Co. (Ann Arbor, MI) or SB 203347 (2-[2-[3,5-bis(trifluoromethyl)sulfonamido]-4-trifluoromethylphenoxy]benzoic acid) (33), zileuton, or indomethacin (synthesized by Medicinal Chemistry at SmithKline Beecham Pharmaceuticals, King of Prussia, PA)
for 10 min at 27 °C prior to addition of stimulating agent. In
designated studies, AA (20 µM) was added to treated
monocytes prior to stimulation and incubated for the designated periods in order to override the need for cell-associated endogenous
deacylation and AA liberation. At the end of the incubation, cell-free
medium was collected and stored at
20 °C until analyzed.
Prostaglandin E2, PGD2, TXB2,
6-keto-PGF2
, LTB4, or LTC4 were
directly measured in cell-free medium using enzyme immunoassay kits
purchased from Cayman Chemical Co. as described previously. Data were
expressed as picograms or nanograms/ml of sample.
Mast cells were obtained from culture of bone marrow cells from BALB/C mice (Jackson Laboratories, Bar Harbor, ME) as described previously (42). Briefly, cells were cultured in RPMI 1640 medium (Life Technologies, Inc.) supplemented with 10% fetal bovine serum, 50 µM 2-mercaptoethanol, 2 mM L-glutamine, and 10-20% of the WEHI-3 cells (Collaborative Biomedical, Bedford, MA) conditioned medium. Cells were grown for 3-4 weeks in an incubator with a humidified 5% CO2 atmosphere at 37 °C. For stimulation, mast cells were passively sensitized overnight by incubation with 20 µg/ml mouse anti-dinitrophenol IgE and then stimulated with antigen bovine serum albumin-dinitrophenol (2 µg/ml) for 15 min. Following stimulation, the cells were pelleted by brief centrifugation and the supernatant fluids collected as described previously (43). LTC4 and PGD2 in the supernatants were determined by enzyme immunoassay kits (Cayman Chemical Co.) as described above.
Phospholipase A2 Activity Analysis of Monocyte Subcellular FractionsHuman recombinant 85-kDa PLA2 and type II 14-kDa PLA2 were prepared and assayed with and without inhibitors as described previously (21). Monocyte subcellular fractions were prepared from 3-4 human donors and 100,000 × g supernatant (cytosol, containing 85-kDa PLA2) and particulate (microsomes, containing 14-kDa PLA2) were used to evaluate inhibitory action of compounds on monocyte PLA2 activity as described previously (21). Cytosol was treated with dithiothreitol (30 min at 37 °C) to inactive contaminating 14-kDa PLA2 activity (44). 85-kDa PLA2 activity in cytosol fractions was assessed using the 85-kDa PLA2 preferred substrate, 1-palmitoyl-2-[C14]AA phosphatidylcholine ([C14]AA phosphatidylcholine, 52 mCi/mmol; DuPont NEN) vesicles (5, 21, 45). The 100,000 × g particulate fraction was assayed using [3H]AA-Escherichia coli (0.5 µCi/5 nmol phosphorus phospholipid (PL), DuPont NEN) substrate as described previously (11, 21). All cell fractions were maintained at equal protein concentrations (29-30 µg of protein/assay) for comparative purposes. Fractions were preincubated with various concentrations of AACOCF3, SB 203347, or Me2SO vehicle (<1%) for 10 min prior to addition of the respective substrate for 30-120 min at 37 °C. Reactions were processed as reported previously (21) and results expressed as percent free fatty acid hydrolyzed/reaction time or specific activity (picomoles of free fatty acid hydrolyzed/mg/h). Drug results were calculated by comparison with untreated Me2SO vehicle control hydrolysis and expressed as percent control.
Immunoblot AnalysisMonocyte cell fractions were prepared
as described above. Cytosol protein or rh 85-kDa PLA2
protein standard were analyzed by SDS-PAGE (10-20% gradient gels;
Integrated Separation Systems, Natick, MA) as described previously (16,
21). Immunoreactive bands, detected using the ECL Western blotting
system (Amersham) were evaluated using scanning densitometry as
described previously (21).
Quantitation of Type II 14-kDa PLA2
To quantitate 14-kDa PLA2, human monocyte 100,000 × g particulates from 4 donors were treated with 0.36 N H2SO4 for 1 h at 4 °C, then brought back to pH 7.4 by addition of 2 M Tris buffer. This sample was directly assessed for 14-kDa PLA2 mass using an enzyme-linked immunosorbent assay developed for rh type II 14-kDa PLA2 as described previously (21, 48). Data were expressed as picograms of 14-kDa PLA2/µg of particulate protein.
Measurement of Intracellular Ca2+ MobilizationA23187-induced Ca2+ mobilization was determined as described previously using the calcium fluorescent probe fura 2 (49). Isolated human monocytes were suspended at 5 × 106/ml and incubated with fura 2 (2 µM) 45 min at 37 °C. The effect of inhibitor on fluorescence of fura 2 was assessed on cells exposed to cytochalasin B (5 µg/ml) for 2.5 min. The [Ca2+]i was calculated as described previously (49).
Statistical AnalysisAll studies were performed using 2-6 human donors. Data were expressed as mean ± S.D. (n = 3 determinations) and analyzed where indicated using analysis of variance and Duncan's multiple range test (p > 0.05).
The Effect of 85-kDa PLA2 Initiation Site Antisense on Monocyte-stimulated Eicosanoid Synthesis
Western Analysis of the Effect of Antisense on Unstimulated MonocytesHuman monocytes (5 × 106/ml) from one donor were treated over 18 h with 3 µM 85-kDa PLA2 initiation site-directed antisense, SK7111, or sense oligonucleotide, SK9030 (3 µM) and cytosols were evaluated for 85-kDa PLA2 protein. Pixel values measured from scanning densitometry of the Westerns are as follows: Lipofectin control (lane 2), 87; 3 µM SK7111 (lane 3), 22; and 3 µM SK9030 (lane 4), 158, showing that in this donor a 75% reduction in the levels of cytosolic 85-kDa PLA2 protein was induced by antisense but not sense oligonucleotide treatment (Fig. 1). This agrees with previous reports in the monocyte (21) and human synovial fibroblast (50), illustrating the reduction of 85-kDa PLA2 protein by antisense but not sense oligonucleotide. In both cases this directly corresponds with reduced cytosolic sn-2 acylhydrolytic activity (21, 50).
The Effect of Antisense on Stimulated Monocyte Eicosanoid FormationAntisense SK7111 (3 µM) treatment significantly reduced (75-90% with A23187 induction or 58-77% with zymosan phagocytosis) stimulated PGE2 formation when compared to stimulated Lipofectin controls. In some cases the 3 µM SK7111-treated A23187-stimulated cells produced PGE2 levels that were equal to or below the basal PGE2 levels measured in the media of unstimulated cells (Table I, Fig. 2). Furthermore, the inhibition of PGE2 production was concentration-dependent over 0.1-3 µM (Fig. 2). A23187-induced LTC4 production was approximately equivalent to or in some cases 2-3-fold higher than PGE2 levels formed over the same time periods. Up to 3 µM antisense oligonucleotide SK7111 had no significant effect on LTC4 formation (Table I, Fig. 2). Similarly, SK7111 had no effect on zymosan-induced LT formation (Table I).
|
Antisense inhibition of prostanoids was not restricted to
PGE2 only, as prostacyclin formation, measured as
6-keto-PGF2, was also reduced in A23187-treated cells.
A23187-induced (1 µM, 7 min, 37 °C)
6-keto-PGF2
produced by monocytes of 2 individuals was
reduced 86-88% by antisense but not sense treatment (donor 1;
Lipofectin control, 0.3 ± 0.03; SK7111 (1 µM),
0.04 ± 0.01; SK9030 (1 µM), 0.5 ± 0.06 ng of
6-keto-PGF2
/5 × 106 (mean ± S.D.; n = 3) and donor 2; Lipofectin control, 0.5 ± 0.09; SK7111 (1 µM), 0.06 ± 0.003; SK9030 (1 µM), 0.5 ± 0.05 ng of
6-keto-PGF2
/5 × 106 (mean ± S.D.; n = 3)).
In a separate study, addition of AA (20 µM) to SK7111-treated cells during stimulation with A23187 (1 µM, 7 min at 37 °C) as described under "Experimental Procedures" enhanced PGE2 formation ~3-fold and completely abrogated antisense-induced inhibition of PGE2 (Lipofectin, 4.1 ± 1.3 versus SK7111, 2.1 ± 0.3 (49% inhibition) and AA + Lipofectin, 13.8 ± 0.9 versus AA + SK7111, 16.9 ± 1.6 ng of PGE2/5 × 106; mean ± S.D.; n = 3). This demonstrated the lack of effect of SK7111 on downstream AA metabolism, consistent with our previous report (21).
Assessment of Human Monocytes for Type II 14-kDa PLA2
Analysis of 100,000 × g monocyte particulates fractions from 4 donors by enzyme-linked immunosorbent assay (as described under "Experimental Procedures") demonstrated the presence of 2.1 ± 0.5 pg of 14-kDa PLA2/µg particulate protein (mean ± S.D.) and confirmed the findings of previous reports of the existence of a biochemically identical 14-kDa PLA2 in monocyte/macrophage particulate fractions (11, 21, 34).
The Effect of 85- or 14-kDa PLA2 Inhibition on Monocyte PLA2 Enzyme Activities
SB 203347 and AACOCF3 were assessed for their effects
on human monocyte subcellular fraction 85-kDa PLA2 or
14-kDa PLA2 enzyme activities as described under
"Experimental Procedures." These results were compared to the
effects of these compounds on purified rh PLA2 enzymes. SB
203347 demonstrated potent inhibition of rh type II 14-kDa
PLA2 (IC50,0.5 µM (Fig.
3, panel B) and 40-fold less inhibition of
the rh 85-kDa PLA2 (IC50, 20 µM
(panel B). In contrast, AACOCF3 selectively
inhibited rh 85-kDa PLA2 (IC50,0.1 µM) with a 300-fold greater potency than against rh type
II 14-kDa PLA2 (IC50, 31 µM (Fig.
3, panel A).
Consistent with the results observed using the recombinant enzymes, AACOCF3 inhibited monocyte cytosolic 85-kDa PLA2 in a concentration-dependent fashion (IC50, 0.17 µM; Fig. 3, panel C). Furthermore, AACOCF3 was ~350-fold more potent against the 85-kDa PLA2 than against monocyte particulate 14-kDa PLA2 activity (IC50, 64 µM). Alternatively, SB 203347 inhibited monocyte 14-kDa PLA2 activity of particulate fractions in a concentration-dependent manner (IC50, 4.5 µM), which was ~20-fold more potent then its action against the 85-kDa PLA2 activity of cytosolic fractions (IC50, 93 µM, panel D).
The Effect of 85- or 14-kDa PLA2 Inhibition on Stimulated Monocyte Eicosanoid Formation
Effect of SB 203347 or AACOCF3 on Monocyte Eicosanoid FormationTo further delineate the contribution of the respective PLA2 enzymes in monocyte AA metabolism, the effect of SB 203347 or AACOCF3 on stimulated monocyte eicosanoid formation was assessed in monocytes from 2 donors as described under "Experimental Procedures." A concentration curve was generated for AACOCF3, an 85-kDa PLA2 inhibitor, exposing human monocytes from one donor to 0.03, 0.1, 0.3, 1, or 3 µM AACOCF3 prior to activation with 1 µM A23187 (7 min, 37 °C). The concentration where PGE2 was inhibited 50% (IC50) was 0.3 µM (confidence limits, 0.25-0.41 µM) with total inhibition at 3 µM (vehicle-stimulated control, 4.0 ± 0.3 versus 3 µM AACOCF3, 0.3 ± 0.05, PGE2/5 × 106 (mean ± S.D.; n = 3), where LT formation remained unchanged (vehicle-stimulated control, 6.8 ± 0.1 versus 3 µM AACOCF3, 8.5 ± 1.2, LTC4/5 × 106 (mean ± S.D.; n = 3)).
Fig. 4 shows that SB 203347 (0.003-100
µM) treatment resulted in
concentration-dependent inhibition of A23187-induced
LTC4 formation (IC50, 0.1 µM,
donor 1; 0.4 µM, donor 2 (Fig. 4, panel A).
Alternatively, PGE2 production of monocytes from donor 1 was not affected by as much as 100 µM SB 203347, while
A23187-induced PGE2 from donor 2 (Fig. 4, panel
A) was inhibited at 100 µM SB 203347, only. This was
possibly due to the initiation of nonspecific inhibition of 85-kDa
PLA2 activity. SB 203347 exhibited a similar effect on the
eicosanoid profile of monocytes activated by phagocytosis of opsonized
zymosan (Fig. 4, panel B). LTC4 was inhibited
(IC50, 0.33 and 0.49 µM, donor 1 and 2, respectively) by SB 203347 and had no effect on PGE2
produced by cells from either donor, at concentrations as high as 100 µM.
To assess the effect of SB 203347 on other prostanoids produced, A23187-stimulated TXB2 levels were measured (7.9 ± 1.0 ng of TXB2/ml,n = 3). Treatment with as high as 30 µM SB 203347 (9.0 ± 0.7 ng of TXB2/ml,n = 3) did not alter TXB2 production. Alternatively, LT inhibition was not specific to LTC4. LTB4 measured in 2 different donors was inhibited 79 ± 2.2% or 61 ± 15%, respectively, by 10 µM SB 203347 with control levels of donor 1 being 19.4 ± 1.4 or donor 2 being 15.8 ± 8.6 ng of LTB4/5 × 106 cells (mean ± S.D.,n = 3).
To assess the possibility that the effect of SB 203347 was
cell-specific, mouse bone marrow-derived mast cells which produce both
PGD2 and LTC4 were evaluated. Mouse bone
marrow-derived mast cells were exposed to increasing concentrations
(0.3-10 µM) of SB 203347 (10 min, 25 °C) prior to
addition of IgE-complex-antigen stimulation as described under
"Experimental Procedures." Untreated control mast cells produced
707 ± 47 LTC4 ng/ml, mean ± S.D. and 78 ± 13 PGD2 ng/ml, mean ± S.D. (n = 3)
upon stimulation. SB 203347 inhibited mast cell LTC4 in a
concentration-dependent fashion (IC50 3 µM) and had no significant effect on PGD2 up
to 10 µM (
12% and +2%, respectively for two
studies).
To
assess the possibility that SB 203347 may be acting through alteration
of A23187 signal transduction, its effect on A23187-mediated Ca2+ mobilization in human monocytes was examined. A23187
at 1 µM induced Ca2+ saturation indicating
that the Ca2+ concentration was 2 µM. This
response was unaffected by SB 203347 at 3.3 µM, a
concentration 10-fold greater than its LTC4 50% inhibitory concentration (0.3 µM). Comparisons where also made using
submaximal A23187 concentrations (5 or 10 nM). In these
cases exposure to SB 203347 (0.05-10 µM) had no effect
on Ca2+ mobilization generated by vehicle alone (data not
shown). This indicates that SB 203347 inhibition of LT formation is not
through interference of stimulated intracellular Ca2+
flux.
SB 203347 and AACOCF3 were compared for their effects on
monocyte-A23187 stimulated eicosanoid formation with and without exogenous AA (20 µM) to reveal possible effects on
downstream AA-metabolizing enzymes. Concentrations that induced ~70%
or greater inhibitory effects in previous studies were chosen. As shown
previously, SB 203347 (1 µM) inhibited LTC4
(86%) and had no effect on PGE2 formation upon A23187
stimulation (Fig. 5, panel A). Addition of 20 µM AA produced a 240-fold and ~ 700-fold increase
in stimulated LTC4 and PGE2 production,
respectively (Fig. 5, panel B), clearly demonstrating that
exogenous AA was available for conversion by the respective
intracellular downstream metabolizing enzymes. In the presence of AA,
SB 203347 had no significant effect on either LTC4 or
PGE2 formation, indicating a lack of action on enzymes in
either the 5-LO or COX pathways. Alternatively, the selective 5-LO
inhibitor, zileuton, inhibited LTC4 production 75% and
60% (compared to the stimulated control), at 1 µM,
without and in the presence of AA, respectively.
Fig. 5 (panel A) shows that AACOCF3 (3 µM) inhibited PGE2 formation 73% in this donor. However, consistent with its having some inhibitory action against COX, PGE2 was again reduced when in the presence of 20 µM AA, to a lesser extent (44%, Fig. 5, panel B). This suggested PGE2 inhibition by AACOCF3, in the absence of exogenous AA, was due to its ability to inhibit both the 85-kDa PLA2 and the COX enzyme. Indomethacin (1 µM) inhibited PGE2 formation 98% in the absence of exogenous AA and inhibited the enhanced PGE2 levels 80%, in the presence of 20 µM AA. AACOCF3 had no effect on LTC4 formation when A23187-stimulated monocytes were incubated with or without 20 µM AA. This indicated that at a concentration where AACOCF3 clearly had prostanoid inhibitory effects, it had no effect on downstream 5-LO AA-metabolizing enzymes.
We have provided further support for the co-existence of both the 85-kDa PLA2 and the type II 14-kDa PLA2 in human monocyte as cell-associated enzymes. Monocytes can be induced to co-produce a number of eicosanoid classes by a variety of stimuli (4, 29) and therefore offer an ideal system for the simultaneous study of the two enzymes. Early studies indicated that cell-associated type II 14-kDa PLA2 participated in stimulated AA release and subsequent eicosanoid formation. With the discovery of 85-kDa PLA2, many laboratories turned their focus toward this enzyme, as it had the characteristics that one would expect for an enzyme responsible for cellular AA metabolism, i.e. regulation by intracellular (nanomolar) Ca2+ levels, phosphorylation, up-regulation by growth factors or inflammatory cytokines and a selectivity for AA in the sn-2 position of substrate phospholipid (6, 17). However, there are no convincing studies that would discount the type II 14-kDa PLA2 as a relevant intracellular enzyme, as it also responds to nM levels of intracellular Ca2+ (42, 35) and readily hydrolyzes AA from the sn-2 position of substrate phospholipid, such as the AA-rich phosphatidylethanolamine, despite the lack of fatty acid specificity noted in vitro (45, 51). We utilized a variety of tools to directly assess the role of the respective cell-associated PLA2 enzymes in eicosanoid formation in response to A23187 or phagocytosis of zymosan particles.
Product formation is the best measure of the COX or 5-LO activity at the whole cell level. In our studies, production of both LT and prostanoid products were well within the linear portion of a product versus time curve. The substrate requirements for the purified or recombinant cyclooxygenase or 5-LO enzymes, in solution, have been broadly reported, the substrate affinities are reasonably similar (Km, 5-50 µM) and both require 1 mol of AA to produce 1 mol of respective product (52, 53, 54). SK7111, antisense treatment, significantly reduced human monocyte 85-kDa PLA2 enzyme protein, which corresponded with comparable reductions in the ability to form prostanoids induced by either A23187 or phagocytosis of zymosan. The enzyme reduced most likely represents the constitutive form, as we did not stimulate induction of enzyme. A23187-stimulated monocytes treated with 3 µM SK7111 antisense produced 75-90% less PGE2 compared to that produced by the stimulated control monocytes. These levels were near or, in some cases below, the amount measured in the unstimulated controls, suggesting that the majority of the enzyme was depleted. Complete reversal of the PGE2 inhibition by addition of exogenous AA indicated that antisense did not alter the AA metabolism through the COX or 5-LO pathways. Under these conditions the stimulated LT formation was not altered, leading one to conclude that the 85-kDa PLA2 provides little or no AA for LT formation. The possibility that the basal levels of 85-kDa PLA2 enzyme remaining after antisense treatment could support full 5-LO metabolism is possible, but seems remote in light of the identical results with the 85-kDa PLA2 inhibitor and the data generated using a selective 14-kDa PLA2 inhibitor discussed below. Finally, zymosan-stimulated eicosanoid production responded to antisense treatment in an identical fashion, with no significant alteration in LT generation, suggests that this response is not unique to the ionophore stimulating system.
Utilization of the two PLA2 inhibitors provides further support for a lack of a role of the 85-kDa PLA2 in LT formation and preliminary evidence to hypothesize that a non-85-kDa PLA2 activity, possibly the type II 14-kDa PLA2, may also provide substrate for LT formation. Both inhibitors demonstrated the appropriate selective actions on the respective monocyte cytosolic 85-kDa PLA2 and particulate fraction 14-kDa PLA2 activities. Consistent with this, we have shown both to inhibit A23187-induced monocyte AA liberation, as assessed by mass, in a concentration-dependent fashion (data not shown). When evaluated in the whole cell system, the 85-kDa PLA2 inhibitor, AACOCF3, effectively reduced PGE2 formation (IC50, 0.3 µM), but at 10-fold greater concentrations had no effect on A23187-induced human monocyte LTC4 formation. Alternatively, concentration-dependent inhibition of 14-kDa PLA2 by SB 203347 prevented both A23187 (IC50, 0.4 µM) or phagocytosis-activated (IC50, 0.3 µM) LTC4 formation but had no effect on prostanoid generation at 10-100-fold greater concentrations. These observations were not restricted to PGE2 or LTC4, alone but were true of other LT or PG family members as well. SB 203347 inhibition of LT was not due to blockade of A23187-induced intracellular Ca2+ flux. Furthermore, overriding the need for deacylation of AA substrate by evaluation of SB 203347 in the presence of exogenous AA, prevented SB 203347 inhibition of LT formation. This supports the lack of SB 203347 direct action on AA conversion to LT via the 5-LO system. Alternatively, AACOCF3 inhibited the augmented PGE2 formation in the presence of exogenous AA, confirming reports that it also has actions on the COX pathway (20). The specific LT inhibitory actions of SB 203347 in the human monocyte corroborate its actions reported in neutrophil LT formation (33) and concur with the identical LT inhibitory actions of a variety of structurally diverse 14-kDa PLA2 inhibitors (29, 30, 31, 32). The similar inhibitory effect of SB 203347 on antigen-induced mast cell LTC4 with no change in PGD4 production by SB203347 provide preliminary data to suggest that the lack of a role of 14-kDa PLA2 in prostanoid biosynthesis is not restricted to the monocyte. Taken together, these studies provide intriguing observations to support the hypothesis that cell-associated 14-kDa PLA2 could be important in stimulated LT formation. More definitive studies are indicated to verify its participation.
In conclusion, the data provide additional evidence that the 85-kDa PLA2 primarily supports prostanoid formation. The data indicate that this is the case in both acute stimuli systems as well as the ligand-activated cell systems previously reported. Alternatively, neither 75-90% reduction in 85-kDa PLA2 by antisense nor specific inhibition of its activity with AACOCF3 altered LT formation. Inhibition of cell-associated 14-kDa PLA2 with SB 203347 produced the reverse stimulated eicosanoid profile, i.e. inhibition of LT while prostanoids were spared. The concept that two distinct enzymes might hydrolyze AA from different pools and/or supply distinct AA-metabolizing systems in a single-cell system is not new (4, 55). Taken together, the results provide a basis for the hypothesis that monocyte LT formation could be supported by substrate AA liberated by an sn-2 acylhydrolase distinct from the 85-kDa PLA2.
We acknowledge the following individuals for providing key resources without which this work could not have been conducted, i.e. Ralph Hall, Jerry Adams, Linette McMillian, Kate Kaiser, Steve Holmes, Ganesh Sathe, Alan Shatzman, Ed Appelbaum, John Breton, Chris Jones, and Jeff Kurdyla. We thank James Foley for the evaluation of SB 203347 on cellular calcium mobilization. We also thank the following individuals for helping to develop and/or support our novel findings through numerous conversations and helpful critique, i.e. Ruth Mayer, Marie Chabot-Fletcher, Mary Barnette, Mark McCord, Dotti Lavan, and Ted Torphy.