Effect of indomethacin on arachidonic acid metabolism in human leukocytes stimulated ex vivo We had previously shown that inhibition of cyclooxygenase in vitro by indomethacin can cause increased formation of products of the 5-lipoxygenase pathway of arachidonic acid metabolism in leukocytes. To determine if this effect also occurred in vivo, we studied leukocyte arachidonic acid metabolism in 12 volunteers before and after ingestion of 150 mg indomethacin daily for 3 days. Blood was collected before treatment and 2 hours, 2 days, and 5 days after the final dose of indomethacin. Serum thromboxane B2, a measure of platelet cyclooxygenase activity, was profoundly suppressed 2 hours after the final dose of indomethacin but had recovered to control values at 2 days. Mixed leukocyte suspensions and purified neutrophil suspensions were prepared and stimulated with calcium ionophore A23187 and the resultant 5-lipoxygenase metabolites were quantified by HPLC. Two hours after the final dose of indomethacin, the stimulated levels of 5-hydroxyeicosatetraenoic acid, leukotriene 134, and leukotriene C, were significandy increased to 247% ± 68%, 135% ± 14%, and 149% ± 23% of pretreatment values, respectively. Two days after the final dose of indomethacin, 5-hydroxyeicosatetraenoic acid levels were still significandy elevated. By 5 days all parameters had returned to baseline Similar effects were not observed in purified neutrophil suspensions, probably because of the loss of indomethacin from the cells during the multiple washing procedures used in their preparation. This is in accord with the reversible nature of the inhibitory effect of indomethacin on cyclooxygenase. We conclude that indomethacin at a commonly used dose increases the ability of circulating leukocytes to produce 5-lipoxygenase products. (CLIN PHARMACOL
THER
1991;49:294-9.)
John C. Docherty, PhD, and Thomas W. Wilson, MD Metabolism of arachidonic acid yields a variety of products, many of which appear to be important mediators of the inflammatory reaction." Thus the cyclooxygenase pathway gives rise to the prostaglandins, prostacyclin, and thromboxane, whereas lipoxygenases produce the hydroxyeicosatetraenoic acids (HETEs), 5-, 12-, and 15-HETE. In addition, activation of the 5-lipoxygenase pathway leads to formation of the dihydroxy-derivative leukotriene B4 (LTB4) and the sulfidopeptide leukotrienes LTC4, LTD4, and LTE4.1'3
From the Departments of Pharmacology and Medicine, University of Saskatchewan. Supported by a grant from the Saskatchewan Heart Foundation, Saskatoon, and a Fellowship from the Saskatchewan Health Research Board, Saskatoon (J.C.D.). Received for publication June I, 1990; accepted Sept. 19, 1990. Reprint requests: John Docherty, PhD, Department of Pharmacology, University of Saskatchewan, Saskatoon, Saskatchewan, S7N OWO Canada. 13/1/25575
294
Saskatoon, Saskatchewan, Canada
Much of the therapeutic benefit of the nonsteroidal antiinflammatory drugs (NSAIDs) is thought to be derived from their ability to inhibit the cyclooxygenase pathway of arachidonic acid metabolism.4'5 The NSAIDs are the drugs of choice for relief of the symptoms of the acute inflammatory component of the rheumatic diseases6 but are less effective in the management of the progressive and chronic aspects of these disorders.7 Indomethacin, the forerunner of the modern NSAIDs, remains one of the most widely prescribed of this group of drugs8 and is often employed as a yardstick against which the efficacy of newly developed NSAIDs is measured. However, this potent cyclooxygenase inhibitor has been shown to potentiate the formation of leukotrienes by neutrophils (PMN)9-11 or bronchial tissue12 stimulated in vitro. A recent study13 has demonstrated indomethacin-mediated enhancement of plasma levels of leukotrienes during anaphylaxis in guinea pigs. Given the powerful inflammatory activity of leukotrienes, it is possible that this aug-
VOLUME 49 NUMBER 3
mentation of their production offsets the benefits of cyclooxygenase inhibition. Studies on the effects of leukotrienes on human cells in vitro have employed a wide range of concentrations of the drug. There is no direct evidence that similar effects would be observed at therapeutic plasma concentrations of the drug in humans. To address this problem we have studied the effect of indomethacin on arachidonic acid metabolism in human peripheral blood leukocytes stimulated ex vivo. The cells were stimulated with the calcium ionophore A23187 in the presence and absence of exogenous arachidonic acid to differentiate effects occurring before and after the release of arachidonic acid from membrane phospholipids. The studies were performed on mixed leukocyte preparations in the presence of circulating plasma levels of indomethacin. Additional studies were performed on purified PMN preparations to determine if the effects of indomethacin could survive the numerous steps involved in the purification of PMN. Glucuronidase release and superoxide generation were also assayed as measures of PMN activation.
MATERIAL AND METHODS Material. Calcium ionophore A23187, n formyl methionyl leucyl phenylalanine (fMLP), ficol, sodium diatrizoate, and substrates for enzyme assays were purchased from Sigma Chemical Co. (St. Louis, Mo.). The lipoxygenase standards LTB4, LTC4, 5-HETE, 12-HETE, and 15-HETE were gifts of Merck Frosst (Pointe Claire, Dorval, Quebec). Arachidonic acid was obtained from Nu-Check Prep (Elysian, Minn.), and dextran T500 was from Pharmacia Fine Chemicals (Piscataway, N.J.). Indomethacin (Indocin) capsules were supplied by Merck Frosst. All other reagents and solvents were of analytic grade and were obtained from either BDH Chemicals (Toronto, Ontario, Canada) or Fisher Scientific Co. (Pittsburgh, Pa.). Subjects. Twelve volunteers (eight men and four women; age range, 22 to 45 years) participated in the study. All subjects denied use of any drugs in the preceding 2 weeks. Indomethacin capsules (50 mg t.i.d.) were administered for 3 days with one further capsule being administered on the morning of the fourth day. Venous blood for leukocyte preparation was drawn into 7 ml vacutainers (Beckton Dickinson and Co., Cockeysville, Md.) containing EDTA anticoagulant before drug treatment and 2 hours, 2 days, and 5 days after the final dose. For determination of serum thromboxane levels, blood was collected into vacutainers
Arachidonic acid metabolism in WBCs
295
with no anticoagulant and allowed to clot at 37° C. All procedures were approved by the University of Saskatchewan President's Advisory Committee on Ethics in Human Experimentation. Stimulation of arachidonic acid metabolism ex vivo. Two cell systems were used for the study of arachidonic acid metabolism ex vivo: mixed leukocyte suspensions and purified PMN. Mixed leukocyte suspensions were prepared by sedimentation of erythrocytes from whole blood by treatment with 0.2 vol of a 5% solution of dextran T500 in phosphate-buffered saline solution (PBS). Aliquots (0.2 ml) of the leukocyteenriched plasma were diluted with 0.3 ml PBS before incubation. PMN were purified by centrifuging leukocyteenriched plasma over a cushion of ficol-sodium diatrizoate. Residual erythrocytes were removed by lysis with NH4CL.14 PMN were washed twice in PBS and resuspended in the same buffer at a concentration of 2 to 4 x 106 cells/ml. Cells were counted in a hemocytometer. Staining with Wright's stain revealed the cell population to be greater than 95% PMN with less than 1 platelet/PMN. These procedures had no significant effect on cell viability as assessed by trypan blue exclusion. Calcium ionophore A23187 and arachidonic acid were stored in aliquots in alcohol at 20° C at concentrations of 1 and 10 mg/nil, respectively. Immediately before use, they were diluted in PBS. Cells were preincubated at 37° C for 10 minutes, after which CaCl2, MgCl2, and dextrose were added to achieve final concentrations of 1.8, 0.5, and 5.6 mmol/L, respectively. After addition of A23187 and arachidonic acid, where appropriate, cells were incubated for 5 minutes at 37° C. Reactions were quenched by addition of 2 vol methyl alcohol, the mixture was held on ice for 1 hour, and supernatants were obtained by centrifugation. After drying under nitrogen at 37° C, residues were resuspended in methanol for HPLC analysis. Serum thromboxane levels were determined as described previously. /5 Briefly, blood was allowed to clot at 37° C for 1 hour and serum was collected after centrifugation. Thromboxane B2 (TXB2) was extracted into chloroform, purified by chromatography over Sephadex LH 20 (Pharmacia Fine Chemicals), and quantified with a specific radioimmunoassay. The limit of detection was 30 pg/ml serum. HPLC of lipoxygenase metabolites. Leukotrienes were separated on a Waters Novapak C18 reversephase column (Waters Associates, Milford, Massachusetts) with acetonitrile : methanol : water: acetic acid (33.6:5.4:60.0:1.0; pH adjusted to 5.6 with am-
CLIN PHARMACOL THER MARCH 1991
296 Docherty and Wilson Table I. Control values of arachidonic acid metabolites from leukocytes stimulated in vitro Lipoxygenase product (ng1106 cells) Cell preparation
Stimulus
Mixed leukocytes Mixed leukocytes PMN PMN
A23187 A23187 + AA A23187 A23187 + AA
LTB4
5.0 5.2 41.0 37.4
± ± ± ±
0.8 0.7 3.7 3.4
LTC4
5-HETE
12-HETE
14.3 ± 3.1 14.2 ± 3.2 7.8 ± 3.4 5.1 ± 1.4
5.4 ± 1.0 7.7 ± 1.6
81.9 ± 20.6 244.0 ± 50.0* ND ND
ND ND
Data are expressed as mean values ± SEM (n = 12). LTB4, Leukotriene B4; LTC4, leukotriene C4; 5-HETE, 5-hydroxyeicosatetraenoic acid; 12-HETE, 12-hydroxyeicosatetraenoic acid; AA, arachidonic acid; PMN, neutrophils; ND, not determined. Suspensions of mixed leukocytes or purified PMN were prepared as described in the Material and Methods section. Calcium ionophore (5 innol/L) and arachidonic acid (16 ilmol/L) were added as indicated. Incubation was for 5 minutes at 37° C. Lipoxygenase products were quantified by HPLC as described in Material and Methods. *p < 0.05 compared with mixed leukocytes plus A23187.
monium hydroxide), with ultraviolet detection at 280 nm.16 HETEs were separated on the same column with methanol water acetic acid (75:25:0.01) with detection at 234 nm.14 Lipoxygenase products were quantified by integration of peak area (Waters Data Module M730) and comparison with authentic standards. Limits of detection for all compounds were approximately 0.5 ng and reproducibility of all procedures was within 10%. Assays for superoxide formation and glucuronidase release. Superoxide generation was assayed as the superoxide dismutase inhibitable reduction of cytochrome C.17 The reaction was monitored continuously in a Bausch and Lomb spectronic 2000 spectrophotometer after addition of 5 ilmol/L A23187. Initial activity was calculated from the linear portion of the curve. Glucuronidase release was assayed after stimulation of cells at 37° C for 5 minutes with 5 iimol/L A23187. Cells were removed by centrifugation and supernatants were assayed for glucuronidase activity with phenolphthalein glucuronide as substrate." Total cellular glucuronidase activity was determined by treating cells with 0.2% Triton X100. Agonistinduced enzyme release has been expressed as percent of total cellular activity. Statistical analysis. Formation of lipoxygenase products was expressed as nanograms/106 cells. Indomethacin-induced effects on these levels were expressed as percent changes from pretreatment levels. Results were analyzed by two-way analysis of variance and level of statistical significance was determined by the Student t test for paired observations.
RESULTS Measurement of serum thromboxane levels was used as an index of cyclooxygenase activity. Circulating plasma levels of TxB2 were less than 15 pg/m1,19
a value at or below the limit of detection of methods for routine measurement of TxB2. The clotting of whole blood is a simple and reproducible procedure for stimulating thromboxane synthesis and assessing the gross effects of drugs on cyclooxygenase activity. Pretreatment levels of serum TxB2 were 78 ± 28 ng/ml. The serum TxB2 levels (percent of preindomethacin levels, mean ± SEM) were 3.6% ± 0.7% (p < 0.001), 99% ± 14.3%, and 163.1% ± 34.3% 2 hours, 2 days, and 5 days, respectively, after the final dose of indomethacin. The profound inhibition of TxB, formation 2 hours after indomethacin administration and its full recovery within 2 days are fully compatible with the known effects of the drug. Stimulation of mixed leukocyte suspensions with the calcium ionophore A23187 (5 p,mol/L) resulted in the formation of significant quantities of the 5lipoxygenase products LTB4, LTC4, and 5-HETE, with much larger quantities of the 12-lipoxygenase product 12-HETE also being formed (Table I). Addition of exogenous arachidonic acid (16 iimol/L) had little effect on the formation of 5-lipoxygenase products but caused a threefold increase in the formation of 12-HETE. Similar treatment of purified PMN resulted in the formation of large quantities of LTB4, with smaller amounts of LTC4 also being formed (Ta-
ble I).
Two hours after the final dose of indomethacin, there were statistically significant increases in the formation of the 5-lipoxygenase products by mixed leukocyte suspensions stimulated with A23187 (Fig. 1). There was no significant increase in 12-HETE formation. Two days after the final dose of indomethacin, the levels of 5-HETE formed during stimulation of mixed leukocyte suspensions with A23187 were still significantly increased. By 5 days after the final dose of indomethacin, all values had returned to control levels.
VOLUME 49 NUMBER 3
Arachidonic acid metabolism in WBCs
Table II. Effect of indomethacin on lipoxygenase product formation in neutrophils stimulated ex vivo
350
,:i:";',
300
Lipoxygenase product (% change from preindomethacin level)
Time after
indomethacin
LTB4
2 Hr 2 Days 5 Days
86 ± 11 99 ± 8 94 -± 9
2 HOURS 2 DAYS 5 DAYS
250
LTC4
200
87 -± 20 106 ± 22
150 100
76 ± 16
-± SEM In = 12). LTB4, Leukotriene 134: LTC4, leukotriene C4. Purified neutrophils, obtained from normal volunteers at the times indicated after administration of indomethacin (50 mg t.i.d. for 3 days), were stimulated with calcium ionophore A23187. and the resultant lipoxygenase products were quantified by HPLC as described in Material and Methods.
1:1
297
50
Data are mean values
0
A
LIB
4
LTC 4
5-H ETE
12-H ETE
5-HETE
12-HETE
350
El
Table III. Effect of indomethacin on neutrophil activity
300 250
Time after
Superoxide generation (nmol1106 cellslmin)
Glucuronidase release (% of total release)
200
indomethacin
A23187
fMLP
(A23187)
150
Before 2 Hr 2 Days 5 Days
1.6 ± 0.7 2.1 ± 1.2 1.8 ± 0.7
4.4 ± 1.2 6.5 ± 0.6 5.8 ± 1.2 ND
42.2 ± 3.8 37.3 ± 4.9 46.1 -± 4.7 49.7 ± 2.1
100
ND
6). Data are mean values ± SEM In fMLP, ii Formyl methionyl leucyl phenylalanine: ND. not determined. Suspensions of neutrophils were prepared from venous blood obtained at the times indicated before or after administration of indomethacin (50 mg t.i.d. for 3 days). Superoxide generation and glucuronidase release accompanying stimulation by calcium ionophore A23187 (5 ilmol/L) or fMLP (I ilmol/L) were assayed as described in the Material and Methods section.
There were no significant differences in the levels of LT134 and LTC4 formed during stimulation of purified PMN with A23187 2 hours, 2 days, and 5 days after the final dose of indomethacin (Table II). Similarly, the indomethacin treatment had no significant effect on the generation of superoxide or release of glucuronidase after stimulation of purified PMN with A23187 or fMLP (Table III).
DISCUSSION Two hours after the final dose of indomethacin, there was a significant increase in the levels of all of the 5-lipoxygenase products on stimulation of mixed leukocyte suspensions with calcium ionophore A23187. This observation suggests that indomethacin has its major effect at or before the 5-lipoxygenase step rather than by altering the enzymatic conversion to, or further metabolism of, the leukotrienes themselves. In addition to leukocytes, the mixed leukocyte
0
2 HOURS
2 DAYS 5 DAYS
50
B
LTB 4
LTC 4
Fig. 1. Effect of indomethacin on lipoxygenase product formation in leukocytes stimulated ex vivo. Mixed leukocyte suspensions were stimulated with calcium ionophore A23187 (5 iimol/L) in the absence (A) or presence (B) of exogenous arachidonic acid (16 iimol/L). Lipoxygenase metabolites were quantified by HPLC. Times refer to the time elapsed since the final dose of indomethacin. *p < 0.05 compared with preindomethacin
suspensions contained platelets (about 30 platelets/leukocyte). It has been observed that addition of platelets to leukocyte suspensions can augment lipoxygenase product formation in vitro.20 This effect of platelets is maximal at a plateletleukocyte ratio of 15:1. It is therefore unlikely that the observed effects of indomethacin could be caused by minor variations in platelet number. In contrast to the striking effects on mixed leukocyte suspensions, the indomethacin treatment had no significant effect on purified PMN with respect to both arachidonic acid metabolism and respiratory burst and granule release. This lack of effect on purified PMN likely reflects the reversible nature of the inhibitory action of indomethacin. Whereas the mixed leukocyte
298 Docherty and
CLIN PHARMACOL THER MARCH 1991
Wilson
suspensions are prepared with minimal manipulation, the cells in effect being stimulated in diluted autologous plasma, the preparation of purified PMN involves multiple washing and centrifugation steps. Thus most of the indomethacin is likely to be washed out of the cells before stimulation of the purified PMN suspensions. This is supported by the observation that a peak corresponding to authentic indomethacin was resolved during HPLC analysis of leukotrienes formed from mixed leukocyte suspensions. No such peak (limit of detection, 500 pg, 1.5 pmol) was observed during analysis of samples obtained from purified PMN suspensions. PMN are capable of being directly affected by indomethacin, as shown in our previous studies in vitro," wherein indomethacin at concentrations of 0.1 to 1.0 )J,mol/L caused a threefold increase in the formation of lipoxygenase products after stimulation with A23187. The mechanism by which indomethacin increases leukotriene formation is presently unknown. It is unlikely to be a simple redirection of substrate arachidonic acid from the cyclooxygenase to the lipoxygenase pathway, because this potentiating effect on leukotriene formation is not a common feature of cyclooxygenase inhibitors.9-11 Most of the arachidonic acid released from membrane phospholipids after cell stimulation appears to be reacylated into phospholipids rather than serving as substrate for further metabolism.21.22 Indomethacin may inhibit these reacylation reactions, thereby leading to a local increased availability of arachidonic acid to serve as substrate for the 5-lipoxygenase. It has recently been reported that activation of 5-lipoxygenase on cell stimulation is accompanied by translocation of the 5-lipoxygenase to the membrane fraction.2324 The activated enzyme would then be in a suitable location to use the available substrate arachidonic acid. Alternatively, indomethacin potentiation of 5lipoxygenase activity may be mediated through increased protein kinase C activity. Indomethacin appears to augment superoxide generation in human PMN in a manner similar to that of the diacyclglycerol kinase inhibitor R59022.25 The increased levels of diacyclglycerol may then serve to stimulate protein kinase C activity. Although activation of protein kinase C alone is not capable of inducing LTB4 formation in human PMN, it does appear to be required to couple increased intracellular Ca to LTB4 formation.26 Either or both of these mechanisms may contribute to the observed increase in leukotriene formation ex vivo after administration of therapeutic doses of indomethacin.
This study has made use of the calcium ionophore A23187 to stimulate lipoxygenase product formation in leukocytes. This compound differs from more physiologically relevant agonists in that A23187 can activate PMN without binding to a specific receptor. Thus cell activation can occur without following the usual sequence of postreceptor occupancy signaling events. In spite of these shortcomings, A23187 has proved useful in many studies on the control of the enzymes responsible for arachidonic acid metabolism14,23 and for the screening of compounds as potential inhibitors of these reactions.27 The difficulties involved in performing this study with more physiologic stimuli (e.g., opsonized zymosan or fMLP) include the low and very variable levels of lipoxygenase products formed in response to these agonists. With these reservations about the use of A23187 in mind, it remains reasonable to think that the potentiating effect of indomethacin on leukotriene formation also occurs in vivo and may contribute to the inability of indomethacin to alter the long-term progression of inflammatory disease states in addition to alleviating the acute symptoms of these disorders. We thank Fran McCauley for TX132 assays.
References Samuelsson B. Leukotrienes: mediators of immediate hypersensitivity reactions and inflammation. Science 1983;220:568-75. Williams TJ. Interactions between prostaglandins, leukotrienes and other mediators of inflammation. Br Med Bull 1983;39:239-42. Ford-Hutchinson AW. Leukotrienes: their formation and role as inflammatory mediators. Fed Proc 1985;44:25-9. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature 1971;231:232-5. Smith JB, Willis AL. Aspirin selectively inhibits prostacyclin production in human platelets. Nature 1971;231:235-7. Ehrlich GE. Basis and current scope of rational antiinflammatory therapy. Am J Med 1985; 79(suppl):2-6 Katz WA. Modern management of rheumatoid arthritis. Am J Med 1985;79(suppl):24-31. Baum B, Kennedy DL, Forbes MB. Utilization of nonsteroidal anti-inflammatory drugs. Arthritis Rheum 1985;28:686-92. Myers RF, Siegel MI. Differential effects of antiinflammatory drugs on lipoxygenase and cyclooxygenase activities of neutrophils from a reverse passive Arthus reaction. Biochem Biophys Res Commun 1983;112:586-94.
VOLUME 49 NUMBER 3
Tavares IA, Capasso F, Vine ND, Bennet A. Effect of isoxicam and other non-steroidal anti-inflammatory drugs on arachidonic acid metabolism by rat peritoneal leukocytes. J Pharm Pharmacol 1985;37:587-8. Docherty JC, Wilson TW. Indomethacin increases the formation of lipoxygenase products in calcium ionophore-stimulated human neutrophils. Biochem Biophys Res Commun 1987;148:543-8. Undem BJ, Pickett WC, Lichtenstein LM, Adams GK. The effect of indomethacin on immunologic release of histamine and sulfidopeptide leukotrienes from human bronchus and lung parenchyma. Am Rev Respir Dis 1987;136:1183-7. Lee TK, Israel E, Drazen JM, et al. Enhancement of plasma levels of biologically active leukotriene B compounds during anaphylaxis in guinea pigs pretreated by indomethacin or by a fish oil-enriched diet. J Immunol 1986;136:2575-82. Borgeat P, Samuelsson B. Arachidonic acid metabolism in polymorphonuclear leukocytes: effects of ionophore A23187. Proc Nat! Acad Sci USA 1979;76:2148-52. Wilson TW, McCauley FA, Wells HD. Effects of lowdose aspirin on responses to furosemide. J Clin Pharmacol 1986;26:100-5. Abe M, Kawazoe Y, Shigematsu N. Influence of salts on high-performance liquid chromatography of leukotrienes. Anal Biochem 1985;144:417-22. Korchak HM, Weismann G. Changes in membrane potential of human granulocytes antecede the metabolic responses to surface stimulation. Proc Natl Acad Sci USA 1978;75:3818-22. Levy GA, Conchie J. Mammalian glycosidases and their inhibition by aldonolactones. Methods Enzymol 1966;8:571-84. FitzGerald GA, Pedersen AK, Patrono C. Analysis of
Arachidonic acid metabolism in WBCs
299
prostacyclin and thromboxane biosynthesis in cardiovascular disease. Circulation 1983;67:1174-7. Kanaj K, Okuma M, Uchino H. Deficient induction of leukotriene synthesis in human neutrophils by lipoxygenase-deficient platelets. Blood 1986;67:903-8. Kroner EE, Peskar BA, Fischer H, Ferber E. Control of arachidonic acid accumulation in bone marrow-derived macrophages by acyltransferases. J Biol Chem 1981;256:3690-7 Irvine RF. How is the level of free arachidonic acid controlled in mammalian cells? Biochem J l982;204:3.
16.
Rouzer CA, Kargman S. Translocation of 5-lipoxygenase to the membrane in human leukocytes challenged with ionophore A23187. J Biol Chem 1988;263: 10980-8. Wong A, Hwang SM, Cook MN, Hogaboom GK, Crooke ST. Interactions of 5-lipoxygenase with membranes: studies on the association of soluble enzyme with membranes and alterations in enzyme activity. Biochemistry 1988;27:6763-9. Dale MM, Penfield A. Comparison of the effects of indomethacin, RHC80267 and R59022 on superoxide production by 1, oleoy1-2 acetyl glycerol and A23187 in human neutrophils. Br J Pharmacol 1987; 92:63-8. McIntyre TM, Reinhold SL, Prescott SM, Zimmerman GA. Protein Icinase C activity appears to be required for the synthesis of platelet-activating factor and leukotriene 134 by human neutrophils. J Biol Chem 1987;262:15370-6. Sweeney FJ, Eskra JD, Carty Ti. Development of a system for evaluating 5-lipoxygenase inhibitors using human whole blood. Prostaglandins Leukot Med 1987;28:73-93.
THER
1991;49:294-9.)
John C. Docherty, PhD, and Thomas W. Wilson, MD Metabolism of arachidonic acid yields a variety of products, many of which appear to be important mediators of the inflammatory reaction." Thus the cyclooxygenase pathway gives rise to the prostaglandins, prostacyclin, and thromboxane, whereas lipoxygenases produce the hydroxyeicosatetraenoic acids (HETEs), 5-, 12-, and 15-HETE. In addition, activation of the 5-lipoxygenase pathway leads to formation of the dihydroxy-derivative leukotriene B4 (LTB4) and the sulfidopeptide leukotrienes LTC4, LTD4, and LTE4.1'3
From the Departments of Pharmacology and Medicine, University of Saskatchewan. Supported by a grant from the Saskatchewan Heart Foundation, Saskatoon, and a Fellowship from the Saskatchewan Health Research Board, Saskatoon (J.C.D.). Received for publication June I, 1990; accepted Sept. 19, 1990. Reprint requests: John Docherty, PhD, Department of Pharmacology, University of Saskatchewan, Saskatoon, Saskatchewan, S7N OWO Canada. 13/1/25575
294
Saskatoon, Saskatchewan, Canada
Much of the therapeutic benefit of the nonsteroidal antiinflammatory drugs (NSAIDs) is thought to be derived from their ability to inhibit the cyclooxygenase pathway of arachidonic acid metabolism.4'5 The NSAIDs are the drugs of choice for relief of the symptoms of the acute inflammatory component of the rheumatic diseases6 but are less effective in the management of the progressive and chronic aspects of these disorders.7 Indomethacin, the forerunner of the modern NSAIDs, remains one of the most widely prescribed of this group of drugs8 and is often employed as a yardstick against which the efficacy of newly developed NSAIDs is measured. However, this potent cyclooxygenase inhibitor has been shown to potentiate the formation of leukotrienes by neutrophils (PMN)9-11 or bronchial tissue12 stimulated in vitro. A recent study13 has demonstrated indomethacin-mediated enhancement of plasma levels of leukotrienes during anaphylaxis in guinea pigs. Given the powerful inflammatory activity of leukotrienes, it is possible that this aug-
VOLUME 49 NUMBER 3
mentation of their production offsets the benefits of cyclooxygenase inhibition. Studies on the effects of leukotrienes on human cells in vitro have employed a wide range of concentrations of the drug. There is no direct evidence that similar effects would be observed at therapeutic plasma concentrations of the drug in humans. To address this problem we have studied the effect of indomethacin on arachidonic acid metabolism in human peripheral blood leukocytes stimulated ex vivo. The cells were stimulated with the calcium ionophore A23187 in the presence and absence of exogenous arachidonic acid to differentiate effects occurring before and after the release of arachidonic acid from membrane phospholipids. The studies were performed on mixed leukocyte preparations in the presence of circulating plasma levels of indomethacin. Additional studies were performed on purified PMN preparations to determine if the effects of indomethacin could survive the numerous steps involved in the purification of PMN. Glucuronidase release and superoxide generation were also assayed as measures of PMN activation.
MATERIAL AND METHODS Material. Calcium ionophore A23187, n formyl methionyl leucyl phenylalanine (fMLP), ficol, sodium diatrizoate, and substrates for enzyme assays were purchased from Sigma Chemical Co. (St. Louis, Mo.). The lipoxygenase standards LTB4, LTC4, 5-HETE, 12-HETE, and 15-HETE were gifts of Merck Frosst (Pointe Claire, Dorval, Quebec). Arachidonic acid was obtained from Nu-Check Prep (Elysian, Minn.), and dextran T500 was from Pharmacia Fine Chemicals (Piscataway, N.J.). Indomethacin (Indocin) capsules were supplied by Merck Frosst. All other reagents and solvents were of analytic grade and were obtained from either BDH Chemicals (Toronto, Ontario, Canada) or Fisher Scientific Co. (Pittsburgh, Pa.). Subjects. Twelve volunteers (eight men and four women; age range, 22 to 45 years) participated in the study. All subjects denied use of any drugs in the preceding 2 weeks. Indomethacin capsules (50 mg t.i.d.) were administered for 3 days with one further capsule being administered on the morning of the fourth day. Venous blood for leukocyte preparation was drawn into 7 ml vacutainers (Beckton Dickinson and Co., Cockeysville, Md.) containing EDTA anticoagulant before drug treatment and 2 hours, 2 days, and 5 days after the final dose. For determination of serum thromboxane levels, blood was collected into vacutainers
Arachidonic acid metabolism in WBCs
295
with no anticoagulant and allowed to clot at 37° C. All procedures were approved by the University of Saskatchewan President's Advisory Committee on Ethics in Human Experimentation. Stimulation of arachidonic acid metabolism ex vivo. Two cell systems were used for the study of arachidonic acid metabolism ex vivo: mixed leukocyte suspensions and purified PMN. Mixed leukocyte suspensions were prepared by sedimentation of erythrocytes from whole blood by treatment with 0.2 vol of a 5% solution of dextran T500 in phosphate-buffered saline solution (PBS). Aliquots (0.2 ml) of the leukocyteenriched plasma were diluted with 0.3 ml PBS before incubation. PMN were purified by centrifuging leukocyteenriched plasma over a cushion of ficol-sodium diatrizoate. Residual erythrocytes were removed by lysis with NH4CL.14 PMN were washed twice in PBS and resuspended in the same buffer at a concentration of 2 to 4 x 106 cells/ml. Cells were counted in a hemocytometer. Staining with Wright's stain revealed the cell population to be greater than 95% PMN with less than 1 platelet/PMN. These procedures had no significant effect on cell viability as assessed by trypan blue exclusion. Calcium ionophore A23187 and arachidonic acid were stored in aliquots in alcohol at 20° C at concentrations of 1 and 10 mg/nil, respectively. Immediately before use, they were diluted in PBS. Cells were preincubated at 37° C for 10 minutes, after which CaCl2, MgCl2, and dextrose were added to achieve final concentrations of 1.8, 0.5, and 5.6 mmol/L, respectively. After addition of A23187 and arachidonic acid, where appropriate, cells were incubated for 5 minutes at 37° C. Reactions were quenched by addition of 2 vol methyl alcohol, the mixture was held on ice for 1 hour, and supernatants were obtained by centrifugation. After drying under nitrogen at 37° C, residues were resuspended in methanol for HPLC analysis. Serum thromboxane levels were determined as described previously. /5 Briefly, blood was allowed to clot at 37° C for 1 hour and serum was collected after centrifugation. Thromboxane B2 (TXB2) was extracted into chloroform, purified by chromatography over Sephadex LH 20 (Pharmacia Fine Chemicals), and quantified with a specific radioimmunoassay. The limit of detection was 30 pg/ml serum. HPLC of lipoxygenase metabolites. Leukotrienes were separated on a Waters Novapak C18 reversephase column (Waters Associates, Milford, Massachusetts) with acetonitrile : methanol : water: acetic acid (33.6:5.4:60.0:1.0; pH adjusted to 5.6 with am-
CLIN PHARMACOL THER MARCH 1991
296 Docherty and Wilson Table I. Control values of arachidonic acid metabolites from leukocytes stimulated in vitro Lipoxygenase product (ng1106 cells) Cell preparation
Stimulus
Mixed leukocytes Mixed leukocytes PMN PMN
A23187 A23187 + AA A23187 A23187 + AA
LTB4
5.0 5.2 41.0 37.4
± ± ± ±
0.8 0.7 3.7 3.4
LTC4
5-HETE
12-HETE
14.3 ± 3.1 14.2 ± 3.2 7.8 ± 3.4 5.1 ± 1.4
5.4 ± 1.0 7.7 ± 1.6
81.9 ± 20.6 244.0 ± 50.0* ND ND
ND ND
Data are expressed as mean values ± SEM (n = 12). LTB4, Leukotriene B4; LTC4, leukotriene C4; 5-HETE, 5-hydroxyeicosatetraenoic acid; 12-HETE, 12-hydroxyeicosatetraenoic acid; AA, arachidonic acid; PMN, neutrophils; ND, not determined. Suspensions of mixed leukocytes or purified PMN were prepared as described in the Material and Methods section. Calcium ionophore (5 innol/L) and arachidonic acid (16 ilmol/L) were added as indicated. Incubation was for 5 minutes at 37° C. Lipoxygenase products were quantified by HPLC as described in Material and Methods. *p < 0.05 compared with mixed leukocytes plus A23187.
monium hydroxide), with ultraviolet detection at 280 nm.16 HETEs were separated on the same column with methanol water acetic acid (75:25:0.01) with detection at 234 nm.14 Lipoxygenase products were quantified by integration of peak area (Waters Data Module M730) and comparison with authentic standards. Limits of detection for all compounds were approximately 0.5 ng and reproducibility of all procedures was within 10%. Assays for superoxide formation and glucuronidase release. Superoxide generation was assayed as the superoxide dismutase inhibitable reduction of cytochrome C.17 The reaction was monitored continuously in a Bausch and Lomb spectronic 2000 spectrophotometer after addition of 5 ilmol/L A23187. Initial activity was calculated from the linear portion of the curve. Glucuronidase release was assayed after stimulation of cells at 37° C for 5 minutes with 5 iimol/L A23187. Cells were removed by centrifugation and supernatants were assayed for glucuronidase activity with phenolphthalein glucuronide as substrate." Total cellular glucuronidase activity was determined by treating cells with 0.2% Triton X100. Agonistinduced enzyme release has been expressed as percent of total cellular activity. Statistical analysis. Formation of lipoxygenase products was expressed as nanograms/106 cells. Indomethacin-induced effects on these levels were expressed as percent changes from pretreatment levels. Results were analyzed by two-way analysis of variance and level of statistical significance was determined by the Student t test for paired observations.
RESULTS Measurement of serum thromboxane levels was used as an index of cyclooxygenase activity. Circulating plasma levels of TxB2 were less than 15 pg/m1,19
a value at or below the limit of detection of methods for routine measurement of TxB2. The clotting of whole blood is a simple and reproducible procedure for stimulating thromboxane synthesis and assessing the gross effects of drugs on cyclooxygenase activity. Pretreatment levels of serum TxB2 were 78 ± 28 ng/ml. The serum TxB2 levels (percent of preindomethacin levels, mean ± SEM) were 3.6% ± 0.7% (p < 0.001), 99% ± 14.3%, and 163.1% ± 34.3% 2 hours, 2 days, and 5 days, respectively, after the final dose of indomethacin. The profound inhibition of TxB, formation 2 hours after indomethacin administration and its full recovery within 2 days are fully compatible with the known effects of the drug. Stimulation of mixed leukocyte suspensions with the calcium ionophore A23187 (5 p,mol/L) resulted in the formation of significant quantities of the 5lipoxygenase products LTB4, LTC4, and 5-HETE, with much larger quantities of the 12-lipoxygenase product 12-HETE also being formed (Table I). Addition of exogenous arachidonic acid (16 iimol/L) had little effect on the formation of 5-lipoxygenase products but caused a threefold increase in the formation of 12-HETE. Similar treatment of purified PMN resulted in the formation of large quantities of LTB4, with smaller amounts of LTC4 also being formed (Ta-
ble I).
Two hours after the final dose of indomethacin, there were statistically significant increases in the formation of the 5-lipoxygenase products by mixed leukocyte suspensions stimulated with A23187 (Fig. 1). There was no significant increase in 12-HETE formation. Two days after the final dose of indomethacin, the levels of 5-HETE formed during stimulation of mixed leukocyte suspensions with A23187 were still significantly increased. By 5 days after the final dose of indomethacin, all values had returned to control levels.
VOLUME 49 NUMBER 3
Arachidonic acid metabolism in WBCs
Table II. Effect of indomethacin on lipoxygenase product formation in neutrophils stimulated ex vivo
350
,:i:";',
300
Lipoxygenase product (% change from preindomethacin level)
Time after
indomethacin
LTB4
2 Hr 2 Days 5 Days
86 ± 11 99 ± 8 94 -± 9
2 HOURS 2 DAYS 5 DAYS
250
LTC4
200
87 -± 20 106 ± 22
150 100
76 ± 16
-± SEM In = 12). LTB4, Leukotriene 134: LTC4, leukotriene C4. Purified neutrophils, obtained from normal volunteers at the times indicated after administration of indomethacin (50 mg t.i.d. for 3 days), were stimulated with calcium ionophore A23187. and the resultant lipoxygenase products were quantified by HPLC as described in Material and Methods.
1:1
297
50
Data are mean values
0
A
LIB
4
LTC 4
5-H ETE
12-H ETE
5-HETE
12-HETE
350
El
Table III. Effect of indomethacin on neutrophil activity
300 250
Time after
Superoxide generation (nmol1106 cellslmin)
Glucuronidase release (% of total release)
200
indomethacin
A23187
fMLP
(A23187)
150
Before 2 Hr 2 Days 5 Days
1.6 ± 0.7 2.1 ± 1.2 1.8 ± 0.7
4.4 ± 1.2 6.5 ± 0.6 5.8 ± 1.2 ND
42.2 ± 3.8 37.3 ± 4.9 46.1 -± 4.7 49.7 ± 2.1
100
ND
6). Data are mean values ± SEM In fMLP, ii Formyl methionyl leucyl phenylalanine: ND. not determined. Suspensions of neutrophils were prepared from venous blood obtained at the times indicated before or after administration of indomethacin (50 mg t.i.d. for 3 days). Superoxide generation and glucuronidase release accompanying stimulation by calcium ionophore A23187 (5 ilmol/L) or fMLP (I ilmol/L) were assayed as described in the Material and Methods section.
There were no significant differences in the levels of LT134 and LTC4 formed during stimulation of purified PMN with A23187 2 hours, 2 days, and 5 days after the final dose of indomethacin (Table II). Similarly, the indomethacin treatment had no significant effect on the generation of superoxide or release of glucuronidase after stimulation of purified PMN with A23187 or fMLP (Table III).
DISCUSSION Two hours after the final dose of indomethacin, there was a significant increase in the levels of all of the 5-lipoxygenase products on stimulation of mixed leukocyte suspensions with calcium ionophore A23187. This observation suggests that indomethacin has its major effect at or before the 5-lipoxygenase step rather than by altering the enzymatic conversion to, or further metabolism of, the leukotrienes themselves. In addition to leukocytes, the mixed leukocyte
0
2 HOURS
2 DAYS 5 DAYS
50
B
LTB 4
LTC 4
Fig. 1. Effect of indomethacin on lipoxygenase product formation in leukocytes stimulated ex vivo. Mixed leukocyte suspensions were stimulated with calcium ionophore A23187 (5 iimol/L) in the absence (A) or presence (B) of exogenous arachidonic acid (16 iimol/L). Lipoxygenase metabolites were quantified by HPLC. Times refer to the time elapsed since the final dose of indomethacin. *p < 0.05 compared with preindomethacin
suspensions contained platelets (about 30 platelets/leukocyte). It has been observed that addition of platelets to leukocyte suspensions can augment lipoxygenase product formation in vitro.20 This effect of platelets is maximal at a plateletleukocyte ratio of 15:1. It is therefore unlikely that the observed effects of indomethacin could be caused by minor variations in platelet number. In contrast to the striking effects on mixed leukocyte suspensions, the indomethacin treatment had no significant effect on purified PMN with respect to both arachidonic acid metabolism and respiratory burst and granule release. This lack of effect on purified PMN likely reflects the reversible nature of the inhibitory action of indomethacin. Whereas the mixed leukocyte
298 Docherty and
CLIN PHARMACOL THER MARCH 1991
Wilson
suspensions are prepared with minimal manipulation, the cells in effect being stimulated in diluted autologous plasma, the preparation of purified PMN involves multiple washing and centrifugation steps. Thus most of the indomethacin is likely to be washed out of the cells before stimulation of the purified PMN suspensions. This is supported by the observation that a peak corresponding to authentic indomethacin was resolved during HPLC analysis of leukotrienes formed from mixed leukocyte suspensions. No such peak (limit of detection, 500 pg, 1.5 pmol) was observed during analysis of samples obtained from purified PMN suspensions. PMN are capable of being directly affected by indomethacin, as shown in our previous studies in vitro," wherein indomethacin at concentrations of 0.1 to 1.0 )J,mol/L caused a threefold increase in the formation of lipoxygenase products after stimulation with A23187. The mechanism by which indomethacin increases leukotriene formation is presently unknown. It is unlikely to be a simple redirection of substrate arachidonic acid from the cyclooxygenase to the lipoxygenase pathway, because this potentiating effect on leukotriene formation is not a common feature of cyclooxygenase inhibitors.9-11 Most of the arachidonic acid released from membrane phospholipids after cell stimulation appears to be reacylated into phospholipids rather than serving as substrate for further metabolism.21.22 Indomethacin may inhibit these reacylation reactions, thereby leading to a local increased availability of arachidonic acid to serve as substrate for the 5-lipoxygenase. It has recently been reported that activation of 5-lipoxygenase on cell stimulation is accompanied by translocation of the 5-lipoxygenase to the membrane fraction.2324 The activated enzyme would then be in a suitable location to use the available substrate arachidonic acid. Alternatively, indomethacin potentiation of 5lipoxygenase activity may be mediated through increased protein kinase C activity. Indomethacin appears to augment superoxide generation in human PMN in a manner similar to that of the diacyclglycerol kinase inhibitor R59022.25 The increased levels of diacyclglycerol may then serve to stimulate protein kinase C activity. Although activation of protein kinase C alone is not capable of inducing LTB4 formation in human PMN, it does appear to be required to couple increased intracellular Ca to LTB4 formation.26 Either or both of these mechanisms may contribute to the observed increase in leukotriene formation ex vivo after administration of therapeutic doses of indomethacin.
This study has made use of the calcium ionophore A23187 to stimulate lipoxygenase product formation in leukocytes. This compound differs from more physiologically relevant agonists in that A23187 can activate PMN without binding to a specific receptor. Thus cell activation can occur without following the usual sequence of postreceptor occupancy signaling events. In spite of these shortcomings, A23187 has proved useful in many studies on the control of the enzymes responsible for arachidonic acid metabolism14,23 and for the screening of compounds as potential inhibitors of these reactions.27 The difficulties involved in performing this study with more physiologic stimuli (e.g., opsonized zymosan or fMLP) include the low and very variable levels of lipoxygenase products formed in response to these agonists. With these reservations about the use of A23187 in mind, it remains reasonable to think that the potentiating effect of indomethacin on leukotriene formation also occurs in vivo and may contribute to the inability of indomethacin to alter the long-term progression of inflammatory disease states in addition to alleviating the acute symptoms of these disorders. We thank Fran McCauley for TX132 assays.
References Samuelsson B. Leukotrienes: mediators of immediate hypersensitivity reactions and inflammation. Science 1983;220:568-75. Williams TJ. Interactions between prostaglandins, leukotrienes and other mediators of inflammation. Br Med Bull 1983;39:239-42. Ford-Hutchinson AW. Leukotrienes: their formation and role as inflammatory mediators. Fed Proc 1985;44:25-9. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature 1971;231:232-5. Smith JB, Willis AL. Aspirin selectively inhibits prostacyclin production in human platelets. Nature 1971;231:235-7. Ehrlich GE. Basis and current scope of rational antiinflammatory therapy. Am J Med 1985; 79(suppl):2-6 Katz WA. Modern management of rheumatoid arthritis. Am J Med 1985;79(suppl):24-31. Baum B, Kennedy DL, Forbes MB. Utilization of nonsteroidal anti-inflammatory drugs. Arthritis Rheum 1985;28:686-92. Myers RF, Siegel MI. Differential effects of antiinflammatory drugs on lipoxygenase and cyclooxygenase activities of neutrophils from a reverse passive Arthus reaction. Biochem Biophys Res Commun 1983;112:586-94.
VOLUME 49 NUMBER 3
Tavares IA, Capasso F, Vine ND, Bennet A. Effect of isoxicam and other non-steroidal anti-inflammatory drugs on arachidonic acid metabolism by rat peritoneal leukocytes. J Pharm Pharmacol 1985;37:587-8. Docherty JC, Wilson TW. Indomethacin increases the formation of lipoxygenase products in calcium ionophore-stimulated human neutrophils. Biochem Biophys Res Commun 1987;148:543-8. Undem BJ, Pickett WC, Lichtenstein LM, Adams GK. The effect of indomethacin on immunologic release of histamine and sulfidopeptide leukotrienes from human bronchus and lung parenchyma. Am Rev Respir Dis 1987;136:1183-7. Lee TK, Israel E, Drazen JM, et al. Enhancement of plasma levels of biologically active leukotriene B compounds during anaphylaxis in guinea pigs pretreated by indomethacin or by a fish oil-enriched diet. J Immunol 1986;136:2575-82. Borgeat P, Samuelsson B. Arachidonic acid metabolism in polymorphonuclear leukocytes: effects of ionophore A23187. Proc Nat! Acad Sci USA 1979;76:2148-52. Wilson TW, McCauley FA, Wells HD. Effects of lowdose aspirin on responses to furosemide. J Clin Pharmacol 1986;26:100-5. Abe M, Kawazoe Y, Shigematsu N. Influence of salts on high-performance liquid chromatography of leukotrienes. Anal Biochem 1985;144:417-22. Korchak HM, Weismann G. Changes in membrane potential of human granulocytes antecede the metabolic responses to surface stimulation. Proc Natl Acad Sci USA 1978;75:3818-22. Levy GA, Conchie J. Mammalian glycosidases and their inhibition by aldonolactones. Methods Enzymol 1966;8:571-84. FitzGerald GA, Pedersen AK, Patrono C. Analysis of
Arachidonic acid metabolism in WBCs
299
prostacyclin and thromboxane biosynthesis in cardiovascular disease. Circulation 1983;67:1174-7. Kanaj K, Okuma M, Uchino H. Deficient induction of leukotriene synthesis in human neutrophils by lipoxygenase-deficient platelets. Blood 1986;67:903-8. Kroner EE, Peskar BA, Fischer H, Ferber E. Control of arachidonic acid accumulation in bone marrow-derived macrophages by acyltransferases. J Biol Chem 1981;256:3690-7 Irvine RF. How is the level of free arachidonic acid controlled in mammalian cells? Biochem J l982;204:3.
16.
Rouzer CA, Kargman S. Translocation of 5-lipoxygenase to the membrane in human leukocytes challenged with ionophore A23187. J Biol Chem 1988;263: 10980-8. Wong A, Hwang SM, Cook MN, Hogaboom GK, Crooke ST. Interactions of 5-lipoxygenase with membranes: studies on the association of soluble enzyme with membranes and alterations in enzyme activity. Biochemistry 1988;27:6763-9. Dale MM, Penfield A. Comparison of the effects of indomethacin, RHC80267 and R59022 on superoxide production by 1, oleoy1-2 acetyl glycerol and A23187 in human neutrophils. Br J Pharmacol 1987; 92:63-8. McIntyre TM, Reinhold SL, Prescott SM, Zimmerman GA. Protein Icinase C activity appears to be required for the synthesis of platelet-activating factor and leukotriene 134 by human neutrophils. J Biol Chem 1987;262:15370-6. Sweeney FJ, Eskra JD, Carty Ti. Development of a system for evaluating 5-lipoxygenase inhibitors using human whole blood. Prostaglandins Leukot Med 1987;28:73-93.
