The 5-lipoxygenase of human ... ation of arachidonic acid via the cyclooxygenase pathway ... trations which completely inhibit the PMN 5-lipoxygenase as.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1984 by The American Society of Biological Chemists, Inc.
Vol. 259, No. 11, Issue of June 10, pp. 6752-6756,19&( Printed in U.S. A.
Activation of a 15-Lipoxygenase/LeukotrienePathway in Human Polymorphonuclear Leukocytesby the Anti-inflammatory Agent Ibuprofen* (Received for publication, July 8, 1983)
Jack Y. Vanderhoek andJ. Martyn Bailey From the Department of Biochemistry, The George Washington University, Schoolof Medicine and Health Sciences, Washington, D.C. 20037
Human peripheral blood polymorphonuclear leuko- and allergic reactions (3). PMN’ leukocytes contain both 5cytes (PMNs) metabolized [“Clarachidonic acid pre- and 15-lipoxygenases (4-6). The 5-lipoxygenase pathway prodominantly by lipoxygenasepathways. The major duces 5-HETE, diHETEs, LTB4, LTC,, LTD,, and LTE, (2). productswere 5-hydroxy-6,8,11,14-eicosatetraenoic LTB, is a very potent chemokinetic and chemotactic agent acid (5-HETE) and 15-HETE. These and otherlipox- for neutrophils(5-HETEisabout a100-foldless active), ygenase products, including theirderived leuko- whereas LTC,, LTD.,, and LTE4 are the active components trienes, have been implicated as mediators of inflam- of slow reactingsubstance of anaphylaxis (2, 7). The 15matory and allergicreactions. In human platelets, the lipoxygenase pathway initially forms15-HPETE which is nonsteroidal anti-inflammatory drug ibuprofen inhib- either reduced to 15-HETE (by cellular peroxidases) or meited production of the cyclooxygenase product throm- tabolized to other products including those of the 15-series boxane B, (160 = 65 FM), whereas the lipoxygenase leukotrienes (8, 9). 15-HPETE has been reported to inhibit vascular prostacyclin synthesis (10) and platelet aggregation product 12-HETE was not appreciably affected even at 5 mM ibuprofen. The5-lipoxygenase of human (11). Other studies indicate that both 15-HPETE and 15PMNs (measured by 5-HETE formation) was inhibitedHETEmodulateseveral cellularlipoxygenases as well as by ibuprofen but was about six times less sensitive (Iso inhibit lymphocyte mitogenesis (6, 12-14). = 420 PM) than the platelet cyclooxygenase. The unIn 1971, Vane (15) and Ferreira et al. (16) reported that biosynthesis. expected observation was made that the human PMN aspirin and indomethacin inhibit prostaglandin 15-lipoxygenase/leukotrienepathway was selectively This observation led to the concept that nonsteroidal anticyclooxygenase enzyme, activated by 1-5 mM ibuprofen. Metaboliteswere iden- inflammatorydrugsinhibitthe of cyclooxygenase pathway tified by ultraviolet spectroscopy, by radioimmunoas- thereby preventing the formation say, orby retention timeson highpressure liquid chro- products (17). Variousinvestigators have reported that certain NSAIDs also inhibited theconversion of arachidonic acid to matography in comparison with authentic standards. The major product was 15-HETE; and in all of 19 lipoxygenase metabolites. It was found that some of these donors tested, 15-HETE formation was stimulated up drugs were more effective in inhibiting one pathway than the were dual cyclooxygenasef lipoxyto 20-fold by 5 mM ibuprofen. Other identified products other, whereas other drugs included 12-HETE and 15- and 12-hydroperoxyeicos- genase inhibitors (18-21). We report here that the NSAID atetraenoic acid. Activation of the 15-lipoxygenase by ibuprofen (2-(4-isobutylphenyl)propionicacid) selectively acibuprofen occurred within 1 min and was readily re- tivates a 15-lipoxygenaseactivity in human PMNs at concentrations which completely inhibit the PMN 5-lipoxygenase as versible. The effects of aspirin, indomethacin, and ibuprofen well as the human plateletcyclooxygenase pathways. on the PMN 15-lipoxygenase were compared in six MATERIALS AND METHODS donors. Ibuprofen produced an average 9-fold stimulation of the enzyme, whereas aspirin andindomethaBlood was obtained from human donors who had not taken any cin resulted in an average 1.5- and 2-fold enhance- aspirin-like compounds during the preceding 2 weeks. PMN leukocytes from human heparinized blood were isolated using a modified ment, respectively.
Many cellsmetabolize arachidonic acidvia two distinct pathways, the cyclooxygenase and the lipoxygenase. Oxygenation of arachidonic acid via the cyclooxygenase pathway leads to prostaglandins,thromboxanes,andprostacyclin, whereaslipoxygenase-catalyzed metabolism producespolyunsaturated hydroxy fatty acids and leukotrienes( 1 , Z ) . These metabolites probably have important roles in inflammatory
Hypaque-Ficoll technique (22). Cell purity was >95% as determined by Wright’s stain. The PMNs were resuspended in Dulbecco’s phosphate-buffered saline, pH 7.0, containing 1 mM glucose at a concentration of 20-25 X lo6 cells/ml. Human platelets were obtained using citrate/dextrose as the anticoagulant, centrifuging for 8 min at 120 x g, and removing the platelet-rich plasma. After the addition of 1 volume of ice-cold Tris/EDTA/NaCl buffer, pH 7.4, the platelet pellet was isolated by centrifugation at 1500 X g for 10 min at 4 “C. Platelets were resuspended in Krebs-Henseleit buffer at a concentration of 3.3 X 10’ cells/ml. Unlabeled arachidonic acid was obtained from NuCheck Prep, Elysian, MN, and [1-”C]arachidonic acid was from Amersham-
* This work was supported by The Upjohn Co. and by grants from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The abbreviations used are: PMN, polymorphonuclear leukocytes; HETE, hydroxyeicosatetraenoic acid; HPETE, hydroperoxyeicosatetraenoic acid LT, leukotriene; NSAID, nonsteroidal anti-inflammatory drug; SP-HPLC, straight phase high pressure liquid chromatography; RIA, radioimmunoassay.
6752
’
Activation of a 15-Lipoxygenme Pathway Human inPMNs Searle.Ibuprofenwasagiftfrom The Upjohn Co. 5-HETE,12HETE,12-HPETE,15-HETE,and15-HPETE were preparedas described previously (6, 13).8-HETE and 9-HETEwere obtained by singlet oxygen oxidation of arachidonic acid (23). 5,15-diHETE and 8.15-diHETE were prepared asdescribed (12,24). LTB, was gift from Dr. J. Rokach, Merck-Frost,Canada, Inc. Indomethacin was purchased from Sigma, and aspirin from the Monsanto Chemical Co. Antiserum to 15-HETEwas a gift from Dr. D. H. Hwang, Louisiana State University, Baton Rouge, LA. [3H]15-HETE wasgenerously provided by New England Nuclear. Ethanolic solutions of the NSAIDs were prepared prior to each experiment. Using this procedure, no precipitation of the drugs was observed whentheethanolicsolutions were added tothePMN suspensions. For most experiments, PMN leukocytes (0.5 ml) were preincubated with different concentrations of ibuprofen, aspirin, indomethacin, or ethanol vehicle for 2 min at 37 “C. [“CIArachidonic acid (35-55 pCi/mol, 16.4 p~ final concentration) was then added and the incubation continued for 5 min.The reaction was terminated by the addition of 1.5 ml of methanol. The cells were removed by centrifugation,andthesupernatant was extracted with 3 ml of chloroform. The chloroform extract was evaporated under Nt. and the residue was spotted on Silica Gel G TLC plates. The plates were developed either in petroleum ether/ether/acetic acid (50501, Solvent A) or inchloroform/methanol/acetic acid/water (90:81:0.8, Solvent B). Labeled products were located by autoradiography, scraped, and counted ina liquid scintillation counter. Experimentswith platelets were carried out in a similar manner using TLC Solvent B for chromatographic analysis. SP-HPLC analyses were performed on a Hewlett-Packard 1084A instrument equipped with a constant wavelength detector (254 nm) andfitted withaLichrosorb-100 column (1.0 X 25 cm,Alltech Associates, Inc.). The column was eluted with solvent system (90% hexane/acetic acid (999:1, SolventC)and 10% hexane/isopropyl alcohol/acetic acid (899100:1, Solvent D))a t a flow rate of 4 ml/min. After 15 min, a linear solvent gradient was established to reach the composition of 10% Solvent C and 90% Solvent D by 30 min after injection. Fractions (0.3 min) were collectedand aliquotswere assayed for radioactivity by liquid scintillation counting. To some samples to be analyzed, authentic HETE, diHETE, and LTB4 standards were added to correlate HPLC retention times of the standardswith those of the radioactive products formed in the incubation mixtures. In larger scale incubations, the product mixture was subjected to SPHPLC, and themajor product fraction elutingbetween 8 and 13 min was collected. This fraction wasreduced with NaBH4to convert any HPETE to HETE (25). An aliquot was rechromatographed on SPHPLC. The peak with retention time of 11.1 min had disappeared, and there was a concomitant increase in the peaks with retention times of 9.2 and 10.1 min. In those experiments where the metabolites were assayed for 15HETE content by RIA, the chloroform extracts of the arachidonic acid metabolites or HPLC fractions were first evaporated under NP. The residues wereredissolved in ethanol, and aliquots of the ethanolic solutions were then assayed by RIA as described (26). To determine the distributionof radioactivity in cellular lipids, the PMN incubation mixtureswere extracted with chloroform/methanol (2:l). After evaporation of the chloroform extract, the residue was chromatographed onsilicic acid as described (27). Two fractionswere obtained, a neutral fraction containing glycerides, unreacted arachidonic acid, and arachidonic acid metabolites and a polar fraction containing phospholipids. The radioactive component of the glyceride and phospholipid fractions was identified as [14C]arachidonic acidby alkaline deacylation (28). Both the glyceride and phospholipid subclasses were separated by TLC in Solvent A and chloroform/methanol/concentrated ammonium hydroxide/water (6535:2.5:2.5), respectively. Appropriate standards were run on the same plate. Labeled products were located by autoradiography, scraped, and counted. RESULTS
Human peripheralblood P M N leukocytes metabolized [“C] arachidonic acid to [I4C]5-HETE (1.8 f 0.19%, 19 donors), 5,12-diHETE isomers, including LTB, (0.42 f 0.05%), and [I4C]15-HETE (0.98 & 0.18%) as shown in Fig. 1, left. Less than 0.2% of the exogenous arachidonic acid is metabolized via the cyclooxygenase pathway (results notshown). Furthermore, 6.9% of the radioactivity on the TLC plateswas in the
6753
CONTROL
IBUPROFEN 1 mM
5 mM
TG-
AA-
15-HETE-
5-HETE-
DIHETEORIGIN-
FIG. 1. Activation of human PMN 15-lipoxygenase by ibuprofen. Autoradiographic profile of [“Clarachidonic acid metabolites formed by human peripheral blood PMN leukocytes (25 X 106/ml) in the absence (kit)or presence (middle, 1mM; right, 5 mM) ofibuprofen. The products were extracted and separated by TLC (Solvent A) as described under “Materials andMethods.” AA, arachidonic acid; TG; triglycerides.
/
15.HETE
0
10
100
1000
10 000
IWPROFEN IPMI
FIG. 2. Comparative effects of ibuprofen on humanplatelet cyclooxygenase and on human PMN 5- and 15-lipoxygenase activities. Platelet cyclooxygenase activity was assayed by measuring formation of thromboxane B, (TXB,)(0).PMN 5-lipoxygenase activity wasmeasured by 5-HETE formation (A) and15-HETE production (0)was used as an indicatorof 15-lipoxygenase activity. [“CIArachidonic acid was used as substrate in all experiments. Detailed procedures are described under “Materials andMethods.” The data pointfor 15-HETE stimulation by 5 mM ibuprofen is anaverage of seven determinations. Values are means f S.E.
triglyceride band (the radioactive component was [“Clarachidonic acid; results not shown), and 4.3% of the radioactivity was at theorigin (phospholipids). Pretreating thePMNs with increasing concentrations of ibuprofen (up to 1 mM) prior to the addition of [’4C]arachidonicacid resulted in progressively decreased formation of [I4C]5-HETEwith little effect on [“C] 15-HETE formation (Fig. 2). The concentration of ibuprofen resulting in half-maximalinhibition of the 5-lipoxygenase was found to be 0.42 f 0.06 mM (13 donors). Whenthe PMNs
in Human PMNs
Activation15-Lipoxygenme of a Pathway
6754
were pretreated with ibuprofen, 1 mM (Fig. 1, middle) and 5 mM (Fig. 1, right), metabolism of [14C]arachidonicacid to [l4C]15-HETEprogressively increased with the concomitant appearance of several minor new bands withRF values of 0.33 (unidentified) and 0.10 (diHETEs). Thedistribution of radioactive products formed by the 5 mM ibuprofen-treated cells was0.46% triglycerides, 19.4% 15-HETE, 0.20% 5-HETE, 2.5% diHETE, and 1.4% at theorigin. In addition to inhibiting incorporation of ['4C]arachidonic acid into triglycerides, ibuprofen suppressed incorporation of [I4C]arachidonic acid into phospholipids. Ibuprofen (5 mM) decreased incorporation of ['4C]arachidonic acid into phosphatidylcholine by 94%, into phosphatidylinositol by 90%, into phosphatidylethanolamine and phosphatidylserine by 76% (each). Trypan blue staining of PMNs treated with 5 mM ibuprofen indicated no loss in viability compared to untreated cells. SP-HPLC analysis of the product mixture obtainedafter PMNswere treated with 5 mM ibuprofen and ['4C]arachidonic acid revealed one major peak and several minor peaks (Fig. 3). The major peak co-migrated with authentic15-HETE(retention time 10.1 min). The retention times of two other peaks corresponded to authentic 12-HETE(9.2 min) anda mixture of 12- and 15HPETE (11.1min; these HPETEs are not resolved under the HPLC conditions used). Further support for the assignment of these HPETEs to the 11.1-min peak is that this peak disappeared when the sample was reduced with NaBH4 (25), and a concomitant increase in the 9.2 and 10.1 peaks was observed. In a representative experiment,the relative proportions of 12-HETE,15-HETE,and12/15-HPETE formed were 25.1:63.9:11.0. After NaBH, treatment, the proportions of both 12-HETE and 15-HETE was increased to 27.2 and 72.8, respectively. The increase in each HETE after reduction was nearly in the same proportion as the original percentage of 12-HETE and 15-HETEin the untreated sample, further confirming the identity of the components of the HPETE peak. Untreated PMNs did not produce 12-HETE, and plate-
let contamination of PMNs was less than 5%,which quantitatively is too low to account for the amount of 12-HETE formed (no [14C]12-HETEformation was observed by autoradiographic analysis from lo6 platelets/ml under these experimental conditions).Ultraviolet spectroscopy and RIA (see below) confirmed the assignment of 15-HETE. The effect of ibuprofen on the formation of 15-HETE from endogenous arachidonic acid was investigated. The product mixture obtained from [14C]arachidonicacid and ibuprofentreated PMNs was analyzed by SP-HPLC. HPLC fractions eluting in the 12/15-HETE region were collected and combined, and themass amount of 15-HETE was assayed by RIA using antiserum specific for 15-HETE (26). The specific activity of the 15-HETE formed was greater than 90% of the specific activity of the starting [I4C]arachidonic acid. Furthermore, 21 f 3 ng of 15-HETE were formed (measured by RIA) by ibuprofen-treated PMNs (1.9 X lo6 cells) and exogenous arachidonic acid, whereas only 0.31 f 0.12 ng of 15HETE was produced by untreated PMNs and added arachidonic acid. In theabsence of exogenous arachidonic acid, less than 0.1 ng of 15-HETE was formed. The time course of ibuprofen stimulation of the PMN 15lipoxygenase is shown in Fig. 4. Ibuprofen was either added before (left) or after (right) the addition of ['4C]arachidonic acid to the PMNs. All incubations were terminated 5 min after the addition of arachidonic acid since preliminary studies (not shown) indicatedthat a plateau in productformation was reached at this time. The results indicate that the stimulation of the 15-lipoxygenase by ibuprofen occurs rapidly and reaches a maximum when the time interval between the addition of ibuprofen and arachidonic acid substrate is less than 1 min. Fig. 5 shows that thestimulation of the PMN 15lipoxygenase by ibuprofen is reversible. Cells treated with ibuprofen exhibited a markedly enhanced formation of ["C] 15-HETE anddecreased production of [14C]5-HETErelative to thecontrol. Cells treated with ibuprofen, then washed with 1%bovine serum albumin to remove ibuprofen (29) showed 5- and 15-lipoxygenase activities comparable to control values. Readdition of ibuprofen tothe washed PMNs again resulted in increased formation of [I4C]15-HETE and negli-
*cI-AA
IBUPROFEN ADDITION
DDlTlON
1
1
t
L 0
5
10
15
20
25
30
35
40
ELUTION TIME (MIN)
FIG. 3. SP-HPLC of [14C]arachidonicacid metabolites obtained after pretreatment of human PMN leukocytes with ibuprofen. The figure shows the SP-HPLC product profile formed from human PMNs (25 X 106/ml) pretreated with ibuprofen (5 mM) followedby incubation with [14C]arachidonicacid (16 p ~ 40, pCi/ mol) for 5 min. Sequential (0.3 min) aliquots of the HPLC fractions were assayed for radioactivity. The chromatographic mobilities of authenticstandardsare indicated by arrows. See "Materials and Methods" for the chromatographic conditions used.
0
1
3
TIME IMIN) OF ["Cl-AA
5 ADDITION
TIME IMIN) OF IBUPROFEN ADDITION
FIG. 4. Time course of ibuprofen activation of the 15-lipoxygenase in human PMNleukocytes. Human PMNs (0.5 ml, 11 X lo6)were incubated with ibuprofen (5 mM) either before (left)or after (right)the addition of [14C]arachidonicacid ([14C]AA,16 pM, 44 pCi/ pmol). All incubations were stopped 5 min after the addition of arachidonic acid, and [14C]HETE formation was determined as described under "Materials and Methods." Formation of [14C]15-HETE and [I4C]5-HETEby PMNs in the absence of ibuprofen is indicated by 0 and A (left).
Activation15-Lipoxygenme of a Pathway Human inPMNs @ 5-HETE
0 15-HETE
UI
’07
FIRST IBUPROFEN (5rnM) WASH TREATMENT SECOND IBUPROFEN (5mM)
-
-
+
-
+
+
-
+
FIG. 5. Reversibility ofibuprofen activation of human PMN 15-lipoxygenase. 15-HETE and 5-HETE formation from [“CC]
arachidonicacidacid (16 p ~ 44, pCi/pmol)in untreated human PMNs (0.5 ml, 11 X lo6) or PMNs (4 ml, 22 X 106/ml; a 0.5-ml aliquot was removed for product analysis) pretreatedwith ibuprofen (5 mM) for 2 min. Products were quantitated as described under “Materials andMethods.”TheremainingPMNs(3.5 ml), which had been pretreated with ibuprofen, were then washed with 1%bovine serum albumin to remove the original ibuprofen. After resuspending these washed cells in buffer (22 X 106/ml),one aliquot (0.5 ml) was incubated with [“C]arachidonic acid (16 p M ) and a second 0.5-ml aliquot was treated with ibuprofen(5 mM) prior to the addition of arachidonic acid.
TABLEI Influence o f 5 mM ibuprofen on the incorporation of [‘4C]arachidonic acid into human PMN cellular lipids and the lipoxygenme-catalyzed metabolism of (“C]arnchidonic acid Cellular lipids and [“Clarachidonic acid metabolites were analyzed as described under“Materialsand Methods.” Subject B
Subject A Control -
TABLEI1 Relative potencies of ibuprofen, aspirin, and indomethacin in stimulating the PMN 15-lipoxygenase in different humansubjects Subject
Stimulation of (“C115-HETE’ formation by Ibuprofen, Aspirin, Indomethacin, 5 mMb 5 mMr 0.7 mMd -fold
W. M. G. K.
T. H. T.S. J. S. J. V.
12.4 1.810.4 2.2 13.6 6.4 6.6
1.1
1.1
2.0’ 3.4
1.o 1.3 0.58‘
0.79 4.7
5.4 l.ld “Stimulation of [14C]15-HETEformationfrom[14C]arachidonic acid is described under “Materials and Methods.” Values obtained are relative to controls withoutthe drug. *Concentration range tested was 0.2-5 mM. At 5 mM, ibuprofen exhibited highest stimulationof the 15-lipoxygenase. ‘Concentration range tested was 0.2-8 mM. A t 5 mM, aspirin exhibited highest stimulation except for subjects G. K. and J. V., where 0.4 and 1 mM, respectively, were found to be the optimum concentrations. dConcentrationrange tested was 0.05-0.7 mM and was not further extended due to insolubility of indomethacin at higher concentrations. At 0.7 mM, indomethacin exhibited highest stimulationof the 15-lipoxygenase except for subjectJ. V. where 0.05 mM was found to be the optimum concentration. tained with ibuprofen, whereas the average enhancement observed with aspirin and indomethacinwas only 1.5- and 2fold, respectively. When the effect of ibuprofen on the 12-lipoxygenase and cyclooxygenase pathways in human platelets was examined, it was found that production of the lipoxygenase metabolite 12-HETE was not appreciably affected even a t 5 mM (results not shown).However, ibuprofen inhibited thecyclooxygenase pathway, and the drug concentrationrequired for half-maximal inhibitionof thromboxane B2formation was about 65 WM (Fig. 2 ) .
Ibuprofen Control Ibuprofen
DISCUSSION
% dpm
[“CIArachidonic acid incorporation into: Diglycerides Triglycerides Phospholipids Total [“C]Arachidonic acid metabolism into: [I4C]5-HETE [I4C]15-HETE
6755
0.42 0.73 1.92
0.10
0.30
0.33 0.45
1.25
0.64 2.04 3.38
0.11
3.07 1.29 0.10
0.44 8.22
1.04
0.52
0.40 7.08
0.32 0.22
1.08
gible production of [14C]5-HETE. Since ibuprofen inhibited uptake of [‘4C]ara~hid~nic acid by cellular lipids, the possibility that this decreased incorporation could account for the increase in [14C]15-HETE formation was examined. Table I shows that the decreasein radioactivity in cellular lipids in the ibuprofen-treated cells relative to the controlswas 3-4-fold less than the increase in [14C]15-HETE production. Even taking into account the decreased formation of [I4C]5-HETE in the presence of ibuprofen, the total decrease in radioactivity was still 2-3-fold less than the observed increase in 15-HETE formation. In order to determine whether the stimulation of the PMN 15-lipoxygenase was unique to ibuprofen, theeffects of other NSAIDs were investigated. As shown in Table 11, the effects of aspirin, indomethacin, and ibuprofen on PMN [I4C]15HETE production in six different donors were compared. An average9-fold increase in 15-lipoxygenase activity wasob-
The presentinvestigation shows that ibuprofen is about six times more effectivein inhibiting the human platelet cyclooxygenase pathway (150 = 65 W M ) than the human peripheral blood PMN leukocyte 5-lipoxygenase (I5,, = 420 p ~ Fig. ; 2). This correlates with other reports that ibuprofen is an effective cyclooxygenase inhibitor in sheepvesicular glands (17). When human PMNs were treated with5 mM ibuprofen (and in certain donors, even with 1 mM), the stimulation of the PMN 15-lipoxygenase was selective since the PMN 5lipoxygenase activity was inhibited by at least 80% in 13 donors tested. Furthermore, the platelet cyclooxygenase was alsocompletely inhibited at thesedrug levels. Themajor metabolite formed from exogenously added [ l4C]arachidonic acid was 15-HETE, although other metabolites formed included 12-HETE, WHPETE, and 12-HPETE. Very little 15HETE (and probably other metabolites) was formedfrom endogenous arachidonic acid under these experimental conditions. The presence of the HPETEsin the product mixture could be due toa partial inhibition of the cellular peroxidases by ibuprofensinceother NSAIDs havebeen reportedto inhibit these enzymes (30). Since purified reticulocyte lipoxygenase oxygenates arachidonic acid at both the n-6 and n-9 positions in a 4:l ratio (31, 32), the formation of 12-HETE and 12-HPETEis probably due to dual Iipoxygenase activities of the humanPMN 15-lipoxygenase. However, the possibility that ibuprofen stimulates a separate PMN 12-lipoxygenase enzyme cannot be excluded. The amount of 5-HETE formed
6756
in Human PMNs
Activation15-Lipoxygenase of a Pathway
in the presence of ibuprofen (Fig. 3) represents residual 5lipoxygenase activity and correlates well with that obtained by TLC autoradiographic analysis. Since untreated PMNs produce much less LTB4 than 5-HETE, it is not surprising that no LTB, (or other 5,12-diHETEs) was detected when ibuprofen-treated PMNs were analyzed by HPLC. Although ibuprofen also inhibitedincorporation of ['4C]arachidoni~ acid in cellular lipids, this decreased incorporation cannot account for the observed increase in 15-lipoxygenase metabolites. The activation of the 15-lipoxygenase by ibuprofen occurred within 1 min of exposure tothis drug and was reversible. These observations suggest that the enhancement was not due to increased protein synthesis or covalent modification of the enzyme. It should be noted, however, that the concentration of ibuprofen needed to activate the 15-lipoxygenase is at least an order of magnitude greater than that found in human serum following administration of a normal pharmacological dose of ibuprofen (29). In comparing the relative amounts of ibuprofen-induced activation of the 15-lipoxygenase in PMNs from 19 different subjects, eight donors showed an 11-20-fold increase and six exhibited a 5-10-fold increase, whereas in five subjects 15HETE formation was increased 2-4-fold. Since ibuprofen was recently reported to inhibit the soybean 15-lipoxygenase(33), theseresults suggest that the mammalian 15-lipoxygenase may have different properties from the plant enzyme. When the effects of aspirin and indomethacin were compared to ibuprofen, it was found that with five of six donors, ibuprofen is 4-11-fold more effective than either aspirin or indomethacin. Studiesare in progress to determine which structuraI functionalities are essentialfor this activation process. Recently, several naturally occurring compounds have been implicated as activatorsof lipoxygenase enzymes. For example, 15-HETE was reported to enhance the 5-lipoxygenase in PT-18 mastfiasophils, whereas both 12-HPETE and ATP stimulated the 5-lipoxygenase in PMN leukocytes (12,34,35). The presentreportindicates thatthe 15-lipoxygenase in human PMNs is normally in a relatively inactive state. No mechanism has yet been described for the selective regulation of the 15-lipoxygenasepathway. It is possible that ibuprofen interacts with a physiological activation process that is normally presentinthese cells. Alternatively, ibuprofen may mimic the action of endogenous activator or possibly displace a naturally occurring inhibitor of the 15-lipoxygenase.
4. Borgeat, P., Hamberg, M., and Samuelsson, B. (1976) J. Biol. Chem. 251,7816-7820 5. Borgeat, P., and Samuelsson, B. (1979) Proc. Natl. Acad. Sci. U. S. A. 76,2148-2152 6. Vanderhoek, J. Y., Bryant, R. W., and Bailey, J. M. (1980) J. Biol. Chem. 2 5 5 , 10064-10066 7. Ford-Hutchinson, A. W., Bray, M. A., Doig, M. V., Shipley, M. E., and Smith, M. J. H. (1980) Nature (Lond.)286,264-265 8. Maas, R. L., Brash, A. R., and Oates, J. A. (1981) Proc. Natl. Acad. Sei. U. S. A. 78,5523-5527 9. Jubiz, W., Ridmark, O., Lindgren, J. A., Malmsten, C., and Samuelsson, B. (1981) Biochim. Biophys. Res. Commun. 9 9 , 976-986 10. Moncada, S., Gryglewski, R. J., Bunting, S., and Vane, J. R. (1976) Prostaglandins 12,715-737 11. Vericel, E., and Lagarde, M. (1980) Lipids 15,472-474 12. Vanderhoek, J. Y., Tare, N. S., Bailey, J. M., Goldstein, A. L., and Pluznik, D. H. (1982) J. Bid. Chen. 257,12191-12195 13. Vanderhoek, J. Y., Bryant, R. W., and Bailey, J. M. (1980) J. Biol. Chern. 255,5996-5998 14. Bailey, J. M., Bryant, R. W., Low,C. E., Pupillo, M. B., and Vanderhoek, J. Y. (1982) Cell. Immunol. 67, 112-120 15. Vane, J. R. (1971) Nat. New Biol. 231,232-235 16. Ferreira, S. H.,Moncada, S., and Vane, J. R. (1971) Nut. New Bid. 231,237-239 17. Rome, L. H., and Lands, W. E. M. (1975) Proc. Natl. Acad. Sci. U. S. A. 72,4863-4865 18. Walker, J. R., and Dawson, W. (1979) J. Pharm. Pharmacol. 3 1 , 778-780 19. Randall, R. W., Eakins, K. E., Higgs, G. A., Salmon, J. A., and Tateson, J. E. (1980) Agents Action 10, 553-555 20. Radmark, O., Malmsten, C., and Samuelsson, B. (1980) FEBS Lett. 110, 213-215 21. Myers, R. F., and Siegel, M. I. (1983) Biochem. Biophys. Res. Commun. 112,586-594 22. Ferrante, A., and Thong, Y. H. (1978) J. Immunol. Methods 2 4 , 389-393 23. Porter. N. A.. Logan, J.. and Kontoviannidou, V. (1979) J. O x . cheA 44,3177-3181 24. Van Os. C. P. A.. Riike-Schilder. G. P. M.. Van Halbeek. H.. Verhagen, J., and\iliegenthart,'J. F. G. (1981) Biochim. '%io: phys. Acta 663,177-193 25. Narumiya, S., Salmon, J. A., Cottee, F. H., Weatherley, B. C., and Flower, R. J. (1981) J. Biol. Chem. 256,9583-9592 26. Bryant, R. W., and Hwang, D. H. (1983) Prostaglandins 26,375385 27. Vanderhoek, J. Y., and Feinstein, M. B. (1979) Mol. Pharmucol. 16,171-180 28. Kates. M. (1975) Technioues Elsevier/North-Hol. of. Lipidology, .
Acknowledgments-we wish to thankS. L. Ekborg, M. Winne, and Dr. A. R. Brash for technical assistance.
31.
REFERENCES 1. Samuelsson, B., Goldyne, M., Granstrom, E., Hamberg, M., Hammarstrom, s.,and Malmsten, C . (1978). Annu. Reu. Biochem. 47.997-1029 2. Samielsson, B. (1983) Science (Wash. D.C.) 220,568-575 3. Kuehl, F. A,, and Egan, R. W. (1980). Science (Wash. D.C.) 2 1 0 , 978-984
~
I .
29. 30.
32. 33. 34. 35.
land, New York Davies. E. F.. and Averv. G . S. (1971) Drum 2,416-446 Siege1,'M. I.,'McConneil; R. T.,'and CuatGcasas, P. (1979) Pruc. Natl. Acad. Sci. U. S. A. 7 6 , 3774-3778 Bryant, R. W., Bailey, J. M., Schewe, T., and Rapoport, S. M. (1982) J. Bid. Chem. 257, 6050-6055 Kuhn, H., Wiesner, R., Schewe, T., and Rapoport, S.M. (1983) FEBS Lett. 153,353-356 Sircar, J. C., Schwender, C. F., and Johnson, E. A. (1983) Prostaglandins 25,393-396 Maclouf, J., DeLaclos, B. F., and Borgeat, P. (1982) Proc. Nutl. Acad. Sci. U. S. A. 79.6042-6046 Ochi, K.,Yoshimoto, T., Yamamoto, S., Taniguchi, K., and Miyamoto, T. (1980) J. Biol. Chem. 258,5754-5758
