INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

TUMOR NECROSIS FACTOR (TNF)- and interleukin (IL)-1, pleiotropic proinflammatory cytokines, are both 17-kDa polypeptides that specifically bind to distinct receptors found on most mammalian cells (9). These two cytokines share many similar biological activities such as induction of hemodynamic shock, acute phase protein synthesis, production of other cytokines including IL-6 and IL-8, and activation of immune cells (9, 22, 31). Because of the coexistence of TNF- and IL-1 in a variety of pathophysiological conditions and the similarity of their biological properties, both cytokines have been suggested to be closely associated with immune- and injury-mediated changes in brain (31) and cardiovascular function (9, 22, 31). However, the precise cellular mechanisms by which TNF- and IL-1 mediate the same cellular responses remain unclear. These two cytokines have been shown to activate common multiple protein kinases in the early events of their signal transduction pathways, including mitogen-activated protein kinase kinase (11, 30). Both TNF- and IL-1 have also been shown to induce nitric oxide synthase (7, 33) and the expression of a variety of genes including phospholipase A₂ (PLA₂) (20, 28, 34) and to stimulate PLA₂ activity and/or secretion (20, 28). Among these pathways, production of arachidonic acid through the activation of PLA₂, thereby releasing eicosanoids, is one of the commonly observed results during TNF- and IL-1 stimulation. The resultant arachidonic acid metabolites and lysophospholipids are suggested to be associated with cytokine-induced inflammation and cell injury (9, 18, 20).

Different types of PLA₂ identified in a variety of cell types include low-molecular-weight secretory PLA₂ (sPLA₂), high-molecular-weight cytosolic PLA₂ (cPLA₂), and Ca²⁺-independent intracellular PLA₂ (iPLA₂) (2, 29). Methyl arachidonyl fluorophosphonate (MAFP), an inhibitor of cPLA₂ and macrophage iPLA₂, and bromoenol lactone (BEL), an irreversible inhibitor of plasmalogen-selective iPLA₂, have been commonly used to identify isoforms of PLA₂ in a variety of cells. TNF- and IL-1 have been reported to activate gene expression and enzyme activity of sPLA₂ (28) and cPLA₂ (34). Recently, iPLA₂ in rat cardiac ventricular myocytes has also been shown to be activated during exposure to IL-1 (26). Despite many indistinguishable cardiac effects of TNF- and IL-1, the effect of TNF- on PLA₂ in cardiac muscle cells has not been characterized.

In this study, we demonstrate different isoforms of PLA₂ existing in subcellular compartments of adult rat ventricular myocytes and have examined the modulation of PLA₂ activity induced by TNF-. Our data show that, in the presence of 4 mM EGTA, TNF- activates intracellular PLA₂ in a concentration-dependent manner when plasmenylcholine or alkylacyl glycerophosphorylcholine but not phosphatidylcholine is used as substrate. Because IL-1 activates only membrane-associated, plasmalogen-selective iPLA₂ in these preparations (26), we used BEL and MAFP to distinguish whether TNF- and IL-1 activate different classes of PLA₂. Our results suggest that TNF- and IL-1 synergistically stimulate arachidonic acid release via activation of distinct PLA₂ isoforms in adult rat ventricular myocytes.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Materials. The stock solutions of human recombinant IL-1 and TNF- were purchased from R&D Systems (Minneapolis, MN). [³H]oleic acid was purchased from NEN (Boston, MA). Bovine heart lecithin was purchased from Avanti Polar Lipids (Birmingham, AL). BEL was a generous gift from Hoffmann-La Roche (Nutley, NJ). MAFP was purchased from Cayman Chemical (Ann Arbor, MI). All other reagents were purchased from Sigma Chemical (St. Louis, MO). Rabbit iPLA₂ polyclonal antiserum was purchased from Cayman Chemical, and mouse monoclonal antibody against cPLA₂ was purchased from Santa Cruz (Santa Cruz, CA). Protein molecular markers were from Bio-Rad (Richmond, CA) or GIBCO (Grand Island, NY). Horseradish peroxidase (HRP)-linked anti-mouse and anti-rabbit antibodies were from Amersham (Arlington Heights, IL). The enhanced chemiluminescence kit (SuperSignal Ultra) was from Pierce (Rockford, IL).

Myocyte isolation. Single ventricular myocytes were isolated from the hearts of adult male Sprague-Dawley rats (250-300 g) using protocols described previously (24). Isolated ventricular myocytes were collected in normal Tyrode solution for the enzyme assay. Measurements of total arachidonic acid release were performed using myocytes incubated in serum- and antibiotic-free culture medium 199 (60%; GIBCO) with 40% Earle's balanced salt solution composed of (in mM) 116 NaCl, 4.7 KCl, 0.9 NaH₂PO₄, 0.8 MgSO₄, 26 NaHCO₃, and 5.6 glucose (pH 7.4 in 5% CO₂-95% air at 37°C) overnight in a CO₂ incubator at 37°C.

Preparation of cytosolic and membrane fractions for Western blot analysis. Myocytes were rinsed twice with ice-cold phosphate-buffered solution. The cell pellets were collected after centrifugation at 800 g at 4°C for 1 min and resuspended in 1 ml ice-cold lysis buffer containing 20 mM HEPES (pH 7.6), 250 mM sucrose, 2 mM dithiothreitol, 2 mM EDTA, 2 mM EGTA, 10 mM -glycerophosphate, 1 mM sodium orthovanadate, 2 mM phenylmethylsulfonyl fluoride, 20 µg/ml leupeptin, 10 µg/ml aprotinin, and 5 µg/ml pepstatin A. After a 15-min incubation, cells were sonicated on ice with three bursts of 15 s each and centrifuged at 14,000 g at 4°C for 10 min to remove remaining cellular debris and nuclei. An aliquot of the supernatant (total protein) was saved and stored at 80°C until use. The rest of the supernatant was centrifuged at 100,000 g at 4°C for 1 h to separate cytosolic (supernatant) and membrane or particulate fractions (pellet), which were resuspended in 0.5 ml lysis buffer containing 1% Triton X-100 and stored at 80°C until use. Protein concentrations of total protein samples and cytosolic and membrane fractions were determined with the Bio-Rad Bradford assay before Western blot analysis using BSA as standard.

Western blot analysis of PLA₂. Protein (2-40 µg) from each sample (total protein, cytosolic and particulate fractions) of ventricular myocytes obtained from each heart and 5 µg protein containing molecular weight markers were mixed with an equal volume of SDS sample buffer, boiled at 95°C for 5 min, and loaded onto two 10% polyacrylamide gels. Protein was separated by standard SDS-PAGE at a constant 200 V for 35 min and electrophoretically transferred to nitrocellulose membranes (Bio-Rad) at a constant 100 V for 2 h. Blots on nitrocellulose membranes were then reversibly stained with 0.1% Ponceau S solution to visualize protein bands, confirm consistent protein loading among wells, and determine the efficiency of protein transfer to membranes. After membranes were destained with MilliQ water, nonspecific sites were blocked with Tris-buffer solution containing 0.05% (vol/vol) Tween 20 (TBST) and 5% (wt/vol) nonfat milk for 1 h at room temperature. Membranes were incubated with primary antibodies against cPLA₂ or iPLA₂ (each with 1:1,000 dilution in the blocking solution) for 1 h at room temperature. After membranes were washed in TBST solution three times for 10 min each, they were incubated for 1 h at room temperature with an HRP-linked secondary antibody (1:50,000 dilution in the blocking solution). Membranes were then washed in TBST six times for 5 min each, detected with the enhanced chemiluminescence method (SuperSignal kit), and exposed to a film (Hyperfilm, Amersham). After detection of specific PLA₂ signal, some of membranes were stripped in a solution containing 2% SDS, 100 mM -mercaptoethanol, and 62.6 mM Tris · HCl (pH 6.7) for 30 min at 50°C and blocked as described above. Membranes with or without stripping were then reprobed with another primary anti-PLA₂ antibody (i.e., iPLA₂ or cPLA₂). After detection of specific signals, followed by stripping, some membranes were rinsed, blocked, and reprobed with anti--actin antibody. All immunoblots were quantified using densitometric analysis (Multi-Analyst, Bio-Rad).

Measurement of PLA₂ activity. PLA₂ activity in isolated myocytes was measured as described previously (26). Briefly, after incubation of isolated myocytes with TNF- and/or IL-1, myocytes were suspended in ice-cold homogenization buffer (0.25 M sucrose, 10 mM KCl, 10 mM imidazole, 5 mM EDTA, and 2 mM dithiothreitol) and sonicated on ice for three bursts of 10 s. The resultant suspension was centrifuged at 14,000 g for 10 min, and the supernatant was centrifuged at 100,000 g to separate cytosol and membrane fractions. PLA₂ activity in the two fractions was measured with 100 µM (16:0,[³H]18:1) plasmenylcholine, (16:0,[³H]18:1) phosphatidylcholine, or (16:0,[³H]18:1) alkylacyl glycerophosphorylcholine in assay buffer containing 100 mM Tris, 4 mM EGTA, and 5% glycerol (pH = 7.0) at 37°C for 5 min in a final volume of 200 µl. Under these reaction conditions, exogenous radiolabeled phospholipid substrate was present in at least 10-fold molar excess compared with endogenous membrane and cytosolic phospholipid. Reactions were terminated by the addition of 100 µl butanol. Released radiolabeled fatty acid was quantified by application of 25 µl of the butanol extract to silica gel G plates, development in petroleum ether-diethyl ether-acetic acid (70:30:1), and liquid scintillation spectrometry of the radioactivity recovered in the fatty acid region. Radiolabeled synthetic phospholipids were prepared as described previously (25). The use of these three substrates in the presence of EGTA determined the Ca²⁺-independent substrate selectivity of PLA₂ present in rat ventricular myocytes.

Measurement of total arachidonic acid release. Arachidonic acid release was determined by measuring [³H]arachidonic acid released into the surrounding medium from myocyte suspensions labeled with [³H]arachidonic acid as described previously (26). Briefly, myocyte suspensions (10⁶ myocytes in 10 ml culture medium) were incubated at 37°C with 3 µCi [³H]arachidonic acid for 18 h. Eighty-five percent of incorporated radioactivity was recovered from phosphatidylcholine or phosphatidylethanolamine phospholipids. After incubation, myocyte suspensions were washed three times with Tyrode solution containing 3.6% BSA to remove unincorporated [³H]arachidonic acid. Myocytes were incubated at 37°C for 15 min before being subjected to experimental conditions. At the end of the stimulation period, myocyte suspensions were centrifuged, and the supernatant was removed. Myocyte pellets were dissolved in 10% SDS, and radioactivity in both supernatant and pellet was quantified by liquid scintillation spectrometry.

Statistics. Values are presented as means ± SE; n represented the number of experiments. The myocytes used in each experiment were collected from two to five hearts to obtain sufficient amounts of cell mass for assays. Statistical significance was evaluated by the two-tailed paired Student's t-test or, when more than two conditions were compared, by one-way analysis of variance with Duncan's multiple range test. Differences with P < 0.05 were considered significant.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Different isoforms of PLA₂ in adult rat ventricular myocytes. Data from our previous study (26) suggest the existence of more than one isoform of PLA₂ in adult rat ventricular myocytes. Thus we used Western blot analysis to detect the possible coexistence of different PLA₂ isoforms in these preparations. Figure 1 demonstrates differential subcellular localization of separate isoforms of PLA₂ as detected by two specific anti-PLA₂ antibodies. In Fig. 1A, a membrane that contained 2-40 µg of total protein and cytosolic protein samples was first probed with antibodies against cPLA₂ (110 kDa) and displayed a concentration-dependent immunoreactive density of cPLA₂ in both samples. In Fig. 1B, a membrane that contained 2-40 µg of total protein and particulate samples was probed with anti-iPLA₂ (85 kDa) antibodies and showed a concentration-dependent density of iPLA₂ in both samples. Two distinctive bands recognized by anti-iPLA₂ antibody were observed, one (at ~82 kDa) in Fig. 1, B and D, and the other in the cytosolic fraction (at ~40 kDa) in Fig. 1D. Figure 1C shows a good correlation between amounts of protein loaded and transferred onto membranes and densities of specific PLA₂ immunoblots from total protein samples and cytosolic and particulate fractions on membranes in Fig. 1, A and B. In this particular experiment, the membranes in Fig. 1, A and B, were rinsed without being stripped, blocked, and reprobed with anti-iPLA₂ and anti-cPLA₂ antibody, respectively. Results show that only total protein, not the cytosolic fractions, displayed the PLA₂ isoform (at ~82 kDa) recognized by anti-iPLA₂ antibody (see Fig. 1A), whereas the particulate fractions did not contain cPLA₂ (Fig. 1B). The location specificity and loading quantity of protein were confirmed by reprobing membranes with anti--actin antibody, which showed a concentration-dependent density of -actin immunoblots (~40 kDa) in the total protein sample and the cytosolic fraction but not in the particulate fraction. Figure 1D shows iPLA₂ immunoblots on a membrane that contained cytosolic and particulate fractions of ventricular myocytes isolated from five different hearts. The membrane was probed only with anti-iPLA₂ antibody and displayed two distinctive bands, one in the cytosolic fraction at ~40 kDa and another in the particulate fraction that is the same as those shown in Fig. 1B. Note that low-molecular-mass bands detected by anti-iPLA₂ antibody in total protein and cytosolic fractions in Fig. 1A and in total protein in Fig. 1B were masked by -actin immunoblots. These results demonstrate that adult rat ventricular myocytes possess different isoforms of PLA₂ located at different compartments and suggest minimal contamination of each compartment by another after the subcellular fractionation procedure.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Western blot analysis of phospholipase A2 (PLA2) isoforms in adult rat ventricular myocytes. Two to forty micrograms of total protein sample (C + P) and cytosolic (C) fractions (A) and the same amounts of total protein sample and particulate (P) fractions (B) of ventricular myocytes obtained from the same heart were loaded on 10% polyacrylamide gels. Protein was separated by standard SDS-PAGE and transferred to nitrocellulose membranes, followed by washing and blocking (see METHODS). Membranes in A and B were incubated with antibodies against Ca2+-dependent (cPLA2) and -independent (iPLA2) PLA2, respectively (both with 1:1,000 dilution). After membranes were washed, they were incubated for 1 h at room temperature with a horseradish peroxidase-linked antibody (with 1:50,000 dilution). Immunoblots on membranes were detected with enhanced chemiluminescence method and exposed to a film. Specific cPLA2 protein migrated between 113 and 82 kDa compared with molecular mass markers and appeared in total protein samples and cytosolic fractions (A). Two specific protein bands were detected by anti-iPLA2 antibodies, one near 82-kDa marker in total protein sample and particulate fractions (B) and another near 40-kDa marker in total protein sample only (not shown, but see Fig. 1D). Without being stripped, membranes in A and B were washed, blocked, and then reprobed with anti-iPLA2 antiserum and anti-cPLA2 antibodies, respectively. After detection of immunoblots, followed by stripping, both membranes were then reprobed with anti--actin antibody, for which immunoblots appeared at ~40 kDa in total protein sample and cytosolic fractions. C: correlation between amounts of protein loaded onto gels and densities of PLA2 immunoblots from total protein sample and cytosolic and particulate fractions on membranes in A and B. D: 20 µg of cytosolic and particulate fractions of same ventricular myocytes obtained from 5 different hearts were loaded on a 10% SDS polyacrylamide gel and transferred onto a nitrocellulose membrane. Membrane was probed only with anti-iPLA2 antibodies and displayed two distinct bands, one with high molecular mass in particulate fraction and another with low molecular mass in cytosolic fraction.

Time course and substrate specificity of TNF--induced stimulation of Ca²⁺-independent PLA₂ activity. We have previously demonstrated that the majority of PLA₂ activity in isolated rat ventricular myocytes is Ca²⁺-independent (26). In this study, we report changes in cytosolic and membrane-associated PLA₂ activity in cytokine-stimulated myocytes measured in the absence of Ca²⁺ (+4 mM EGTA) measured using (16:0,[³H]18:1) plasmenylcholine, alkylacyl glycerophosphorylcholine, and phosphatidylcholine substrates.

1

1

1

1

1

1

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Time course for effects of tumor necrosis factor (TNF)- on cytosolic (A) and membrane-associated (B) PLA2 activity in adult rat ventricular myocytes. Myocytes were exposed to 10 ng/ml TNF- from 1 to 15 min, and PLA2 activity was measured using (16:0,[3H]18:1) plasmenylcholine (n = 5), (16:0,[3H]18:1) alkylacyl glycerophosphorylcholine (n = 4), or (16:0,[3H]18:1) phosphatidylcholine (n = 5) in absence of Ca2+. Substrates were incubated with 200 µg cytosolic protein or 8 µg membrane protein for 5 min at 37°C in presence of 4 mM EGTA. Data are presented as means ± SE for independent results; each experiment uses myocytes collected from 2 to 5 separate hearts. * P < 0.05, ** P < 0.02 compared with control values at 15 min.

1

1

1

1

Concentration-dependent effects of TNF- on Ca²⁺-independent PLA₂ activity. TNF- induced peak changes in cytosolic PLA₂ activity were observed with all substrates in 2 min in four of five experiments. Thus we stimulated isolated ventricular myocytes with increasing concentrations of TNF- for 2 min and measured PLA₂ activity in the cytosolic and membrane fractions. Figure 3A shows that cytosolic PLA₂ activity measured with plasmenylcholine was significantly increased with TNF- concentrations >2.5 ng/ml (from control of 0.36 ± 0.03 to 0.81 ± 0.13 nmol · mg protein¹ · min¹, n = 3, P < 0.05) and was maximal at concentrations 10 ng/ml (1.07 ± 0.16 nmol · mg protein¹ · min¹ at 50 ng/ml, n = 3, P < 0.02). As measured with alkylacyl glycerophosphorylcholine, PLA₂ activity in the cytosolic fraction was also significantly increased over control values at TNF- concentrations of 10 ng/ml (from control of 0.31 ± 0.08 to 0.85 ± 0.03 nmol · mg protein¹ · min¹, n = 3, P < 0.02) (Fig. 3A). In contrast, cytosolic PLA₂ activity was significantly decreased at TNF- concentrations >10 ng/ml as measured with phosphatidylcholine (from control of 0.53 ± 0.03 to 0.29 ± 0.03 nmol · mg protein¹ · min¹, n = 3, P < 0.02) (Fig. 3A). Figure 3B shows that membrane-associated PLA₂ activity was unaffected by the 2-min exposure to TNF- at all concentrations tested using all three phospholipid substrates. Thus, in adult rat ventricular myocytes, TNF- alters cytosolic Ca²⁺-independent PLA₂ activity in a concentration- and time-dependent manner as early as 2 min and has no effect on membrane-associated PLA₂ activity during this period. Because we have previously demonstrated an IL-1-enhanced membrane-associated PLA₂ that is selective for plasmalogen substrates (26), these results show that TNF- modulates different PLA₂ isoforms in ventricular myocytes.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Concentration dependence of cytosolic PLA2 activity in ventricular myocytes exposed to TNF-. Cytosolic (A) and membrane-associated (B) PLA2 activities were measured with (16:0,[3H]18:1) plasmenylcholine, (16:0,[3H]18:1) alkylacyl glycerophosphorylcholine, or (16:0,[3H]18:1) phosphatidylcholine in absence of Ca2+. Myocytes were exposed to different concentrations of TNF- for 2 min. Substrates were incubated with 200 µg cytosolic protein or 8 µg membrane protein for 5 min at 37°C in presence of 4 mM EGTA. Data are presented as means ± SE for independent results from 3 separate experiments; open symbols represent untreated values. * P < 0.05, ** P < 0.02 compared with control values.

Effect of TNF- on total arachidonic acid release. Figure 4 shows that exposure of ventricular myocytes to 10 ng/ml TNF- for increasing time intervals resulted in a significant increase in [³H]arachidonic acid release within 1 min (from control of 1.14 ± 0.03% to 2.31 ± 0.10% for TNF-, n = 3, P < 0.01). The total arachidonic acid release remained significantly greater than control values over the 20-min exposure to TNF- (3.32 ± 0.67%, n = 3, P < 0.05, compared with 1.12 ± 0.04% for time controls). These findings suggest that TNF--induced activation of cytosolic PLA₂ results in total arachidonic acid release from the myocytes. To distinguish which isoform of PLA₂ was involved in the TNF--induced increase in arachidonic acid release, we used specific PLA₂ inhibitors to determine whether TNF- activates a distinct PLA₂ isoform from the IL-1-induced iPLA₂. BEL is a potent, irreversible specific inhibitor of myocardial iPLA₂ (1, 6) and has been shown to completely block IL-1-induced arachidonic acid release (26). Figure 4 shows that preincubation of myocytes with 10 µM BEL for 30 min had no significant effect on TNF--stimulated arachidonic acid release. MAFP is a selective, active site-directed, irreversible inhibitor of both cytosolic cPLA₂ and iPLA₂, but not sPLA₂ (23, 29). Preincubation of myocytes with 5 µM MAFP for 15 min caused a complete inhibition of the TNF--induced increase in arachidonic acid release over the 20-min interval (compared with time control, Fig. 4). Taken together, these results suggest that TNF--stimulated arachidonic acid release involves a different PLA₂ isoform from that involved in the IL-1-stimulated arachidonic acid release.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Time course for TNF--induced total arachidonic acid release. [3H]arachidonic acid release from ventricular myocytes was measured in response to exposure for 2 min to 10 ng/ml TNF- (+TNF-, n = 3). Myocytes were preincubated with 5 µM methyl arachidonyl fluorophosphonate (MAFP) for 15 min (+TNF- + MAFP, n = 3) or 10 µM bromoenol lactone (BEL) for 30 min (+ TNF- + BEL, n = 3) before exposure to TNF-. TNF--enhanced arachidonic acid release was completely abolished by MAFP but unaffected by BEL. Dashed line represents time control of arachidonic acid release in absence of TNF- or PLA2 inhibitors (n = 3); open symbols represent untreated control values. * P < 0.05, ** P < 0.02 compared with control or time control values.  P < 0.05,  P < 0.02 compared with TNF- exposure in absence of PLA2 inhibitors.

Synergistic effects of IL-1 and TNF- on total arachidonic acid release. We have shown that both IL-1 and TNF- increased arachidonic acid release, probably via activation of different PLA₂ isoforms. Because IL-1 and TNF- synergistically activate many biological events (7-9, 22), we examined the synergistic effect of these cytokines on arachidonic acid release and PLA₂ activity.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Effect of PLA2 inhibitors on synergistic stimulation of arachidonic acid release induced by interleukin (IL)-1 and TNF-. [3H]arachidonic acid release was measured in myocytes exposed to 10 ng/ml TNF- for 1 min, followed by addition of 5 ng/ml IL-1 to TNF- for another minute (shaded bar). Preincubation with 5 µM MAFP for 15 min (+MAFP) or 10 µM BEL for 30 min (+BEL) partially inhibited arachidonic acid release. Preincubation with both MAFP and BEL (+MAFP & BEL) completely blocked arachidonic acid release induced by IL-1 and TNF-. Solid horizontal line represents presence of IL-1 + TNF-. Data are means ± SE; nos. in parentheses represent no. of experiments. * P < 0.01 compared with control values.  P < 0.05,  P < 0.01 compared with corresponding value of IL-1 + TNF- in absence of inhibitors.

Effect of specific PLA₂ inhibitors on TNF--stimulated cytosolic PLA₂ activity. To further examine the TNF--induced PLA₂ isoform, we determined the effect of these two specific PLA₂ inhibitors, MAFP and BEL, on the PLA₂ activity in cytosolic and membrane fractions during a 2-min exposure to 10 ng/ml TNF- using different substrates. Figure 6A shows that pretreatment of myocytes with 5 µM MAFP for 15 min significantly decreased basal cytosolic PLA₂ activity (in the absence of cytokine) with plasmenylcholine substrate (without MAFP: 0.34 ± 0.05 vs. with MAFP: 0.13 ± 0.03 nmol · mg protein¹ · min¹, n = 4, P < 0.02) and completely blocked the TNF--induced twofold increase in PLA₂ activity (without MAFP: 0.70 ± 0.04 vs. with MAFP: 0.20 ± 0.03 nmol · mg protein¹ · min¹, n = 4, P < 0.02; Fig. 6A). In contrast, neither basal nor TNF--induced increase in cytosolic PLA₂ activity was significantly altered by preincubation with 10 µM BEL (control: 0.25 ± 0.02 vs. TNF-: 0.50 ± 0.07 nmol · mg protein¹ · min¹, n = 4 each). Preincubation with 5 µM MAFP had no effect on the membrane-associated PLA₂ activity in untreated or TNF--treated myocytes (Fig. 6B). However, preincubation with BEL decreased the PLA₂ activity in the membrane fraction in the absence and presence of TNF-, a result consistent with a significant basal plasmalogen-selective iPLA₂ activity in mammalian cardiac myocytes (13).

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Effect of PLA2 inhibitors on cytokine-induced changes in PLA2 activity using plasmenylcholine as substrate. Cytosolic (A) and membrane-associated (B) PLA2 activities were measured using (16:0,[3H]18:1) plasmenylcholine in absence of Ca2+ (open bars). Myocytes were exposed to 10 ng/ml TNF- for 2 min in absence (shaded bars) or presence of 5 ng/ml IL-1 for 1 min (solid bars). A: TNF--increased cytosolic PLA2 activity was completely blocked by preincubation with 5 µM MAFP for 15 min but unaffected by preincubation with 10 µM BEL for 30 min. (Note that PLA2 activities in presence of TNF- are not significantly different in absence and presence of BEL). Increased cytosolic PLA2 activity in TNF- + IL-1 was also completely blocked by MAFP or MAFP + BEL. B: increased membrane PLA2 activity in TNF- + IL-1 was attenuated by MAFP and completely abolished by BEL or MAFP + BEL. Data are presented as means ± SE for independent results from 4 separate experiments. * P < 0.05, ** P < 0.02, *** P < 0.01 compared with control values.  P < 0.05,  P < 0.01 compared with corresponding TNF- or TNF- + IL-1- stimulated levels. ¶ P < 0.02 compared with control in presence of corresponding PLA2 inhibitors.

1

1

1

1

1

1

View larger version (27K): [in this window] [in a new window]   Fig. 7.   Effect of PLA2 inhibitors on cytokine-induced change in PLA2 activity using alkylacyl glycerophosphorylcholine as substrate. Cytosolic (A) and membrane-associated (B) PLA2 activities were measured using (16:0,[3H]18:1) alkylacyl glycerophosphorylcholine in absence of Ca2+ (open bars). A: myocytes were exposed to 10 ng/ml TNF- for 2 min in absence (shaded bars) or presence (solid bars) of 5 ng/ml IL-1 for 1 min. TNF--increased cytosolic PLA2 activity was completely blocked by preincubation with 5 µM MAFP for 15 min but unaffected by 10 µM BEL for 30 min. Enhanced cytosolic PLA2 activity in TNF- + IL-1 was completely blocked by MAFP, BEL, or both. B: increased membrane PLA2 activity induced by TNF- + IL-1 was attenuated by MAFP but completely abolished by BEL or MAFP + BEL. Data are presented as means ± SE for independent results from 4 separate experiments. * P < 0.05, ** P < 0.02, *** P < 0.01 compared with control values.  P < 0.05,  P < 0.01 compared with corresponding TNF-- or TNF- + IL-1-stimulated levels. ¶ P < 0.01 compared with control in presence of corresponding PLA2 inhibitors.

1

1

1

1

View larger version (30K): [in this window] [in a new window]   Fig. 8.   Effect of PLA2 inhibitors on cytokine-induced change in PLA2 activity using phosphatidylcholine as substrate. Cytosolic (A) and membrane-associated (B) PLA2 activities were measured with (16:0,[3H]18:1) phosphatidylcholine in absence of Ca2+ (open bars). A: exposure to 10 ng/ml TNF- for 2 min significantly decreased cytosolic PLA2 activity (shaded bars) that was reversed by addition of 5 ng/ml IL-1 in 1 min (solid bars). Preincubation with 5 µM MAFP for 15 min suppressed basal level without effect on cytosolic PLA2 activity in either TNF- or TNF- + IL-1. Preincubation with 10 µM BEL for 30 min reversed TNF--reduced cytosolic PLA2 activity without an effect on PLA2 activity in TNF- + IL-1. Similarly, MAFP + BEL had no effect on cytosolic PLA2 activity in TNF- + IL-1. B: BEL decreased membrane PLA2 activities in control and in TNF- + IL-1. MAFP + BEL also significantly decreased membrane PLA2 activity in TNF- + IL-1. Data are presented as means ± SE for independent results from 4 separate experiments. * P < 0.05, ** P < 0.02, *** P < 0.01 compared with control values.  P < 0.05,  P < 0.01 compared with corresponding TNF-- or TNF- + IL-1-stimulated levels. ¶ P < 0.01 compared with control in presence of corresponding PLA2 inhibitors.

Effects of PLA₂ inhibitors on synergistic actions of IL-1 and TNF- on PLA₂ activity. We previously showed a IL-1-activated membrane-associated iPLA₂ activity that was completely blocked by BEL (26). Because Fig. 5 showed a synergistic stimulation of arachidonic acid release by IL-1 and TNF- together, we then examined the synergistic effect of these cytokines on cytosolic and membrane-associated PLA₂ activity.

1

1

1

1

1

1

1

1

1

1

1

1

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The majority of PLA₂ activity in the mammalian myocardium is that of iPLA₂ (2, 10, 26, 29). This isoform has been shown to play an important role in signal transduction in response to several agonists (10, 25, 26) and in cell injury during myocardial ischemia (2, 10, 25). Both IL-1 and TNF- have also been shown to be closely linked to injury- and/or immune-mediated cardiovascular disorders (22, 31). We have previously shown that IL-1 enhances a membrane-associated iPLA₂ that preferably uses plasmenylcholine as substrate and is sensitive to BEL (26). The present study shows that at least two different PLA₂ isoforms exist in rat ventricular myocytes and that TNF- activates a MAFP-sensitive, BEL-insensitive, cytosolic iPLA₂ that is selective for sn-1 ether-linked phospholipids.

Different PLA₂ isoforms exist in rat ventricular myocytes. Different PLA₂ isoforms were recognized by two distinctively specific antibodies against cPLA₂ and iPLA₂, respectively. The monoclonal antibody specific for cPLA₂, which does not cross-react with other phospholipases, detects an isoform only in cytosolic fractions of rat ventricular myocytes with a molecular mass between 82 and 113 kDa, similar to that of cPLA₂ as reported previously by others (27, 36). The iPLA₂ polyclonal antiserum cross-reacts with human iPLA₂ from smooth muscle cell lines and murine iPLA₂ from P388D1 macrophages but not cPLA₂ or sPLA₂. This anti-iPLA₂ antibody constantly recognizes two proteins of ventricular myocytes, one in cytosolic fractions and another in particulate fractions. The membrane-associated iPLA₂ isoform has a molecular mass in the range of 80-85 kDa as reported for P388D1 macrophages and Chinese hamster ovary cells (1, 4, 5). However, the location of iPLA₂ in P388D1 macrophages was not specified because the whole cell homogenate (except nuclei) was used (3, 5). The molecular mass of cytosolic iPLA₂ protein is ~40 kDa, which is similar to that of the purified cytosolic iPLA₂ as reported for canine myocardium (14). Apparently, these two iPLA₂ proteins contain the same sequence that is recognized by an anti-iPLA₂ antibody.

Effects of TNF- on PLA₂ activity differ from IL-1. As measured with plasmenylcholine or alkylacyl glycerophosphorylcholine, the TNF--increased cytosolic iPLA₂ activity and the concomitant increase in arachidonic acid release are completely blocked by MAFP but unaffected by BEL. In contrast, BEL blocks IL-1-induced increases in both membrane-associated iPLA₂ and arachidonic acid release (26). Accordingly, although both cytokines can elicit arachidonic acid release from ventricular myocytes, they appear to activate different isoforms of iPLA₂. This is further supported by the results showing that the additive effect of TNF- and IL-1 on the arachidonic acid release is attenuated by either MAFP or BEL, presumably inhibiting each iPLA₂ isoform. This additive effect of cytokines can be completely blocked by inhibition of both iPLA₂ isoforms in the presence of MAFP plus BEL. Moreover, only when both TNF- and IL-1 are added together is the concomitant increase in cytosolic and membrane-associated PLA₂ activity attenuated by either MAFP or BEL. Under these conditions, no increase in PLA₂ activity in either subcellular fraction is detected when myocytes are pretreated with MAFP plus BEL. These results are consistent with the additive stimulation of total arachidonic acid release induced by TNF- and IL-1 together via activation of separate iPLA₂ isoforms.

Does TNF- act on different cytosolic PLA₂ isoforms? In contrast to the increase in cytosolic iPLA₂ activity measured with the use of sn-1 ether-linked phospholipid substrates, TNF- decreases the cytosolic PLA₂ activity when phosphatidylcholine is used as substrate, without any effect on membrane-associated PLA₂ activity. Interestingly, when the same substrates are used, IL-1 also decreases cytosolic PLA₂ activity, which returns to the basal level after a 10-min exposure (26). The TNF--induced decrease in the cytosolic PLA₂ activity is blocked by pretreatment with BEL, but not MAFP, suggesting a cytosolic iPLA₂. Therefore, it is conceivable that both cytokines decrease an iPLA₂ that preferably acts on phosphatidylcholine. Results showing that the TNF--induced suppression of iPLA₂ activity is reversed by IL-1 (Fig. 8A) could be attributed to the possibility that IL-1 competes with TNF- for the regulatory site of cytosolic iPLA₂. These results further support the data from Western blots showing that cardiac myocytes possess multiple isoforms of iPLA₂ that have different substrate specificity. Modulation of this iPLA₂ activity appears to be cytokine specific.

Comparison with other isozymes of PLA₂. BEL is a potent irreversible inhibitor of myocardial and macrophage iPLA₂ with an IC₅₀ of 100 nM, which is >1,000-fold specific for inhibition of iPLA₂(s) compared with multiple cPLA₂(s) or sPLA₂(s) (1, 2, 15, 29). We have shown that BEL completely inhibits the IL-1-activated membrane-associated iPLA₂ and arachidonic acid release in rat ventricular myocytes (26). Similarly, the present study demonstrated that BEL decreases the basal and IL-1-induced membrane PLA₂ activities in the absence or presence of TNF- (Figs. 6B, 7B, and 8B), consistent with the involvement of the previously described myocardial iPLA₂ (1). In contrast, BEL has no effect on the TNF--activated cytosolic iPLA₂ and arachidonic acid release, suggesting that TNF- acts on a iPLA₂ distinct from the BEL-sensitive, membrane-associated iPLA₂ with a preference for plasmenylcholine.

Pathophysiological implications. As occurs in many injury- and immune-associated cardiovascular disorders, the presence of TNF- and IL-1 activates different classes of intracellular PLA₂ with plasmenylcholine, the natural and major substrate, as well as other phospholipids as substrate in cardiac ventricular myocytes. They synergistically increase the arachidonic acid release and, thereby, a resultant eicosanoid production. Both arachidonic acid and concomitantly produced lysophospholipids reduce cardiac L-type Ca²⁺ channel currents (I_Ca,L; unpublished observation). The increased arachidonic acid has also been shown to activate sphingomyelinase, thereby increasing ceramide production (16, 17). Ceramide and its metabolite, sphingosine, which both inhibit cardiac I_Ca,L (32, 35), probably mediate IL-1- and TNF--induced suppression of I_Ca,L and contractility (21, 32). Thus activation of PLA₂ signaling pathways during cytokine stimulation results in depressed myocardial contractile function observed in many immune- and cell injury-mediated pathophysiological conditions. The role of the cytokine-decreased phosphatidylcholine-selective cytosolic iPLA₂ in this event and the identification of TNF--activated cytosolic PLA₂ requires further investigation. The present and future studies should provide an insight into the development of specific therapeutical strategy for these cytokine-mediated diseases.