Guy$ Hospital, London Bridge, London, SEI 9RT, United Kingdom, the (Jlnstitute of ... London. United Kingdom and the **Department of Zoology, University of ...
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc
Vol. 266, No. 14, Issue of May 15, pp. 8720-8726 1991 Printed in S.A.
Synthesis of Lipoxins and Other Lipoxygenase Products by Macrophages fromthe Rainbow Trout, Oncorhynchus myhiss* (Received for publication, November 20,1990)
Trevor R. PettittS, Andrew F. RowleySQ,Susan E. Barrowli, Anthony I. Malletl), and Christopher J. Secombes** From the $Biomedical and Physiological Research Group, School of Biological Sciences, University College of Swansea, Singleton Park, Swansea, SA2 8PP, United Kingdom, the TDepartment of Clinical Pharmacology, United Medicaland Dental Schools, Guy$ Hospital, London Bridge, London, SEI 9RT, United Kingdom, the (Jlnstituteof Dermatology, United Medical and Dental Schools, St. Thomas’ Hospital. London. United Kingdom and the **Department of Zoology, University of Aberdeen, Tillydrone Avenue, Aberdeen, AB9 2TN;United Kingdom
Rainbowtrout macrophages maintained in short term culture when incubated with either calcium ionophore, A23187, or opsonized zymosan synthesize a range of lipoxygenase products including lipoxins and leukotrienes. These cells areunusual in that they generate more lipoxin than leukotriene following such challenge. The main lipoxin synthesized was lipoxin (LX) A,. This compound was identified by cochromatography with authentic standard during reversephase high performance liquid chromatography, by ultra violet spectral analysis, radiolabeling following incorporation of [ 14C]arachidonic acid substrate into macrophage phospholipids, and gas chromatography electron impact massspectrometry of the methyl ester, trimethylsilyl ether derivative. Other 4-series lipoxins synthesized by trout macrophages were identified as 11-trans-LXA,, 7-cis-1 1-trans-LXA4,and 6(S)-LXA4. These cells also produced 5-series lipoxins tentatively identified as LXAS, ll-trans-LXASand possibly 6(S)LXAs. No LXB4 or LXBS was, however, detected. The dynamics of leukotriene and lipoxin release were also determined. Lipoxin generation was slower then leukotriene generation the latter reaching a maximum after 30 min of exposure to ionophore (5 pM, 18 “C) compared with 46 min for the former.
Many studies have investigated the synthesis and function of various eicosanoids leading to a wealth of information, particularly on the leukotrienes, prostaglandins, and thromboxanes (e.g. 1-6). However, one group of eicosanoids which remains relatively poorly characterized in terms of their function are the lipoxins, first described by Serhan andcolleagues in 1984 (7,8). These trihydroxytetraenesmay bederived from arachidonic acid or eicosapentaenoic acid via one or more biosynthetic route(s) which involve specific introduction of oxygen to form hydroperoxide and epoxide intermediates (Fig. 1).The epoxide pathway involves formation of a 5,6- or 14,15epoxide from an intermediate such as 15-hydroperoxyeicosa-
* This work was supported by the Oliver Bird Fund for Research into Rheumatism and theScience and Engineering Research Council Grant GR/G/05179. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. To whom correspondence should be addressed.
tetraenoic acid (15-HPETE)’ (9). There is also evidence for a triple lipoxygenase pathway to lipoxins in some cell systems which introduces the hydroxyl groups via three separate lipoxygenase reactions (10). An alternative mechanism which has been identified in platelets involves the biosynthesis of leukotriene (LT) A4 which is then converted to an epoxide intermediate (11, 12). A number of cell types from different species are capable of performing the later steps in the biosynthesis of the lipoxins in uitro, although most require the addition of a suitable substrate, usually 15-HPETE or 15-hydroperoxyeicosapentaenoic acid (15-HPEPE), before any production is detected (e.g. 7,8,13,14). Several more recent resultshave shown that bovine and porcine leukocytes and rabbit reticulocytes can synthesize lipoxins from exogenous fatty acid (15-17), while mixed human platelet-granulocyte suspensions (11, 12), human neutrophils (la), and human eosinophilic granulocytes (19) also generate them from endogenous substrates, as can porcine leukocytes when incubated with a snakevenom phospholipase Az (20). In these species, the lipoxins are minor lipoxygenase products of relatively low potency, and there has been some discussion as to whether these compounds have any physiological or pathological function. However, we have recently demonstrated that the lipoxins are a major class of eicosanoid synthesized by rainbow trout macrophages and that they have potent chemotactic/chemokinetic activities in this species (21, 22). This paper reports on the full identification of lipoxins and other lipoxygenase products synthesized by rainbow trout, Oncorhynchus mykiss, macrophages in response to both call The abbreviations used are: 5-HPETE, 5-hydroperoxy-5,8,11,13eicosatetraenoic acid 5-HPEPE, 5-hydroperoxy-5,8,11,13,17-eicosapentaenoic acid; HETE, hydroxyeicosatetraenoic acid; HEPE, hydroxyeicosapentaenoic acid; diHETE, dihydroxyeicosatetraenoic acid; diHEPE, dihydroxyeicosapentaenoic acid; PGB2, prostaglandin 13EBz; LXA,, lipoxin A,, 5(S),6(R),l5(S)-trihydroxy-7E,9E,llZ, eicosatetraenoic acid; LXAS, lipoxin A,, 5(S),6(R),15(S)-trihydroxy7E,9E,llZ,13E,17Z-eicosapentaenoicacid; 6(S)-LXA4, 5(S),6(S),15 (S)-trihydroxy-7E,9E,11Z,13E-eicosatetraenoic acid; 11-trans-LXA, 5(S),6(R),15(S)-trihydroxy-7E,9E,llE,l3E-eicosatetraenoic acid; 7cis-11-trans-LXA,, 5(S),6(R),15(S)-trihydroxy-7Z,9E,llE,l3E-eicosatetraenoic acid; LXB,, lipoxin B,, 5(S),l4(R),l5(S)-trihydroxy6E,8Z,lOE,12E-eicosatetraenoic acid; LTA, leukotriene A LTB,, leukotriene B4, 5(S),12(R)-dihydroxy-6Z,8E,lOE,l4Z-eicosatetraenoic acid g-trans-LTB,, 5(S),12(R)-dihydroxy-6E,8E,lOE,l4Z-eicosatetraenoic acid; 6-trans-12(S)-LTB4, 5(S),12(S)-dihydroxy6E,8E,lOE,14Z-eicosatetraenoicacid; LTBs, leukotriene Bg; LPS, lipopolysaccharide; NDGA nordihydroguaiaretic acid; HBSS, Hank’s balanced salt solution; RP-HPLC, reverse-phase high performance liquid chromatography; GC/MS, gas chromatography/mass spectrometry; 1, liter.
Lipoxins Lipoxygenase and Products Synthesized
by 0. mykiss
an internal standard and theeicosanoids extracted with C1, Sep-Pak minicolumns (Millipore/Waters, Watford, U. K.) and finally separated by RP-HPLC using an UltrasphereCl, column (5-pm packing; 4.6 mm X 25 cm; Beckman RIIC) as described previously (24, 25). Briefly, the lipoxygenase products were eluted witha solvent gradient of either 100% water/methanol/acetonitrile/acetic acid (45:30:25:0.05 MATERIALS AND METHODS (v/v) pH 5.7), or methanol/water/acetic acid (70:300.05 (v/v), pH 5.7) both changing to 100% methanol in40 min using a flow rate of Chemicals-Calcium ionophore, A23187, prostaglandin (PG) B,, 0.6 ml. rnin", or 0.4 ml.min", respectively. Detection was a t 235, Escherichia coli Olll.B4 lipopolysaccharide (LPS), Saccharomyces 280, and 301 nm. Full UV spectra of the major peaks were obtained cereuisiae zymosan, nordihydroguaiaretic acid (NDGA),andindoduring HPLC separation. Further identificationwas achieved where methacin were obtained from SigmaChemical Co. Ltd., Poole, United possible by comparing the retention times with those of authentic Kingdom (U. K.), LXA, and 12-hydroxyeicosatetraenoic acid (12standards. Material from the major peaks was collected, dried under HETE) from Peninsula Laboratories, Merseyside, U. K., 6(S)-LXA, NP, resuspended in 100 p1 of the second solvent system, and rerun from Cascade Biochem Ltd., Reading, U. K., leukotriene (LT) B4, 6(with collection) in this solvent to purify the components for subsetrans-LTB,, 6-trans-12(S)-LTB4, 5(S),G(R)-diHETE, 5-HETE, and quent structural analysis. 15-HETE fromNovabiochem, Nottingham, U. K. [l-'4C]arachidonic In some experiments 0.25,0.5,1.0 2.5, 5.0, or 10 p~ NDGA (a acid(58 mCi.mmol-') and[l-'4C]eicosapentaenoicacid(59mCi. lipoxygenase inhibitor) or 1,10, or 50 p M indomethacin (a cyclommol") were purchased from Amersham International, Amersham, oxygenase inhibitor) was added 5 min before the calcium ionophore. U. K. while [l-l*C]docosahexaenoic acid (57 mCi.mmo1") was obQuantification of lipoxygenase products separated by RP-HPLC tained from Du Pont-New EnglandNuclear. Leukotriene B5 (LTB,) (PGB,) using c$gG2 = 28,000 and lipoxin B, (LXB,) were kindly provided by Dr. John Williams, was by reference to the internal standard Cardiff Royal Infirmary and 11-trans-LXA, and 7-cis-ll-trans-LXA4 M.Cm", eggBz=13,000 M .cm", ~ 5 : ~="25,000 M.Cm-l, e301 = 50,000 by Dr. Charles N. Serhan,Harvard MedicalSchool, Boston. All M .cm" for lipoxins and ezm = 44,000 M .cm" for conjugated trienes (including the leukotrienes). solvents were of HPLC grade and/or redistilled in glass. GC/MS Analysis-Components isolated by HPLC thatshowed UV Fish-Rainbow trout, 0. mykiss, were obtained from Whitesprings Fisheries, Pontardulais, SouthWales, U. K. Fish were maintained in spectra characteristic of a conjugated tetraene structure were taken large outdoor tanks and fed on Mainstream Expanded Trout Diet (B.to dryness under a stream of nitrogen at ambient temperature. The P. Nutrition (UK) Ltd., Cheshire, U. K.). Adult triploid fish were residues containing lipoxins were converted to methyl esters, trimethylsilyl ethers using ethereal diazomethane, and bis-(trimethylused in all studies. MacrophageIsolation and Culture-Macrophages were isolated sily1)-trifluoroacetamide (26). Samples were reconstituted in n-hepfrom the hemopoietic head kidney of rainbow trout and placed in tane containing 1%bis-(trimethylsily1)-trifluoroacetamideand anashort term culture using a procedure modified from that described lyzed by GC/electron impact MS usinga VG-70 SEQ instrument previously (23). Briefly, thehead kidneywasremoved, pressed equipped with a Gerstel injector (Mulheim, Germany) anda 6-m SEthrough a fine mesh into ice-cold Eagle's basal medium containing 30 capillary GC column (0.33 mm innerdiameter).Samples were 0.01% (final concentration) penicillin/streptomycinmix (1:1), centri- introduced into the injector at an initial temperature of 80 "C. This solvent was purged during heating to 90"C (0.2 "C.s"). The sample fuged (1,000 x g, 10 min, 4 "C) to pellet the cells, resuspended in Eagle's medium, loadedonto 54% Percoll continuous gradientsCa/ in was then vaporized in the injector by heating to 280 "C (12 "C .s-'). The GC column was maintained at an initial temperature of 150 "C Mg-free Hank's balanced salt solution (HBSS; final concentration: for 2 min and then programmed to320 "C a t a rate of 30 "C.min". KC1, 0.10 g.1" KH,PO,, 1.27 g.1" 8.20 g.1" NaC1, 0.40 g.1" Radiolabe~ing-[ l-'4C]Arachidonic, eicosapentaenoic, or docosaNaHCOs, 2.00 g.1" glucose, pH 7.2; gradients formed by centrifugation at 22,000 X g for 15 min at4 "C) and centrifuged(2,800 X g, 30 hexaenoicacids (0.5 pCi) were added to each washedmacrophage min, 4 "C). The whitecell bands (containing macrophages and other culture, incubated for 1 h at 18 "C in Ca/Mg-containing HBSS, the leukocytes) were removed,washedtwice with Ca/Mg-free HBSS, supernatant removed, and the cells washed with Ca/Mg-containing HBSS to removefreeradiolabel. Subsequently, they were covered resuspended in Eagle's basal medium (containing 0.01% penicillin/ streptomycin),platedoutinto 25 cm3 flasks(5 ml/flask; Nunc, with fresh Ca/Mg-containing HBSS and challenged with 5 p~ calcium ionophore for 30 min a t 18 "C, and extracted and separated as Denmark), and left to adhere for 2 h at 18 "C. Subsequently, nonadherent cells were removedbywashing and the remaining cells, before. Radiolabeled HPLC fractions (0.3 ml) were collected, mixed mostly macrophages, maintained for 48-72 h at 18 "C in5 ml of fresh with 4.5 ml of Pico-fluor 40, and analyzed on a Beckman LS 3801 were made usingan external Eagle's medium containing 5% (final concentration) heat-inactivated scintillation counter. Quench corrections fetal bovine serum and antibioticsas before. Immediately before use, standard method. Biological Challenge-Zymosan was opsonized for 60 min a t 18 "C any nonadherent cells remaining were removed by vigorous washing with Ca/Mg-containing HBSS (8.00 g.1" NaC1, 0.40 g.1" KCl, 0.10 using heterologous rainbow trout serum and washed several times in HBSS before further use. Macrophage cultures in Ca/Mg-containing g.1" KH,PO,, 1.27 g.1" NaHC03, 2.00g.1" glucose, 0.41 g.1" HBSS were incubated with opsonized zymosan (0.25, 1.0, 2.5, 5.0, or MgS0,.7 H20, 0.26 g.1" CaC12.2 H20, pH 7.2). Theremaining adherent cells (195% macrophages) were covered with 5 ml of fresh 10 mg.ml") for 2 h a t 18 "C followed by eicosanoid extraction and separation as described above. Macrophage cultures were also chalCa/Mg-containing HBSS. lenged with E. coli Olll.B4 LPS (0.1, 1.0, and 10 pg.rn1-I) for 2 h at Ionophore Challenge-Macrophage cultures (-1 X lo7 cells/flask) 18 "C. were incubated with 5 p~ calcium ionophore, A23187, for 5-60 min at 18 "C, the medium removed, and centrifuged to pellet any cells (10,000 X g, 5 min, room temperature). PGB, (100 ng) was added as RESULTS
cium ionophore and biological stimuli, and their subsequent identification using cochromatography with authentic standards, fatty acid radiolabeling and gas chromatography/mass spectrometry (GC/MS).
Identification of Lipoxins-Rainbow trout macrophages synthesized a range of oxygenated fatty acid metabolites in response to both ionophore and opsonized zymosan but not LPS. A typical RP-HPLC chromatogram is shown in Fig. 2. IEW I5-HPETE ---UV analysis of peaks 1-7 showed maxima at 289, 301, and J-w 316 nmcharacteristic of a conjugated tetraenestructure. 5,15diH(l')ETE ----Peaks 8-17 showed UV maxima at 260,270, and 280 nm I characteristic of a conjugated triene structure. The major peaks were identified as LXA,, 11-trans-LXA,, 5,6-epoxy-15-HETE LTB4, LTBB, and 12-HETE on the basis of cochromatography (14.15-epoxy-5-HETE) with authentic standards using the two different solvent systems. Material in the smaller peaks coeluted with 7-cis-11Lipdxins I trans-LXA,, 6(S)-LXA,, 6-trans-LTB,, 6-trans-12(S)-LTB4, FIG. 1. Possible biosynthetic routes to lipoxins derived from 5(S),G(R)-di-HETE,and5-HETE (Fig. 2). There was no arachidonic acid. Lipoxygenase ( L O ) . detectable LXB,, or 15-HETE. LTA4
by 0. mykiss
Lipoxins and Lipoxygenase Products Synthesized 100 90
'5 h m t l o n tlnw (rnlnl ' 0
FIG.2. Lipoxygenase metabolites produced by calcium ionophore-challenged rainbow trout macrophages. Rainbow trout macrophage cultures were stimulated withcalcium ionophore A23187 by RP(5 p ~ 30, min, 18 "C), Sep-Pak extracted, and separated HPLC on a Cls ODS column (5 pm, 4.6 mm X 25 cm, Ultrasphere) using a solvent gradient changing from 100% water/methanol/acetonitrile/acetic acid (45:3025:0.05 (v/v), pH 5.7) to 100% methanol in 40 min (flow rate, 0.6 ml.min-'). 100 ng of PGB, was added as an internal standard. Peaks 1-7 all show spectra characteristicsof lipoxins, while peaks 8-17 all show spectra characteristic of conjugated trienes. Labeled peaks were identified by cochromatography with authentic standards. The elution positions for authentic LXB, and 15-HETE are marked although these compounds were not detected in the samples.
FIG.4. Authentic LXA, (methyl ester, trimethylsilyl ether derivative). Electron impact mass spectra of GC peaks elutingwith retention times ( a ) 5 min 46 s, ( b ) 6 min 16 s. TIC 5:24
. ."_" . ~ ~ _-------. _ ~~
_ ~ ~ ~ _
Mass 5 8 0 Mass 2 0 3
5 :"4 6 5 : 00
FIG.3. GC chromatogram of authentic LXA, (methyl ester, trimethylsilyl ether derivative). TIC = total ion current. GC/electron impact MS was used to elucidate the structure of the lipoxins in HPLC components 1, 2, 4, and 6. A GC chromatogram of authentic LXA, is shown in Fig. 3. Mass spectra of the peaks eluting with retention times of 5 min 46 s and 6min 16 s (carbon values 25.6 and 26.9, respectively) are shown in Fig.4. The later eluting peak (C-26.9) has a retention time identical to authentic 11-trans-LXA, and is presumably a product of on column LXA, isomerization. The peak with a retention time of 5 min 46 s is most likely the parent LXA,. Both mass spectra show a molecular ion at m/ z (mass to charge ratio) 582 and a base peak at m/z 203 (CH(OSiMe3)-(CH2),-COOCH3), the latterpeak representing cleavage of the C5-C6 bond between vicinal hydroxyl groups. loss of CH,), 551 ("31, loss Further ions at m/z 567 ("15, of CH30), 492("90; loss of Me3SiOH), 482("100, rearrangement then loss of (OCH-(CHJ4-CH3),461 (M-31-90), and 171 (203379 (M-203), 173 (CH(OSiMe3)-(CH2),-CH3), CH30H) are consistent with the structure of LXA, or its 11-
Mass 2 0 3
6 : 15
FIG.5. GC chromatogram of LXAa (methyl ester, trimethylsily! etherderivative) obtained from trout macrophages (HPLC component 2). TIC = total ion current. trans isomer. The GC chromatogram and mass spectra of component 6 isolated by HPLC were identical to those of authentic LXA, andits11-trans isomerization product. HPLC component 4 showed a single GC peak with a retention time of 6 min 16 s (carbon value 26.9) and a mass spectrum consistent with a 4-series lipoxin and was identified as 11trans-LXA, by comparison of its retention time with authentic standard. The GC chromatogram of HPLC component 2 is shown in Fig. 5. This component shows similar chromatographic characteristics to LXA, with two peaks with retention times of 5 min 45 s and6 min 15 s (carbon values 25.6 and 26.8).
Lipoxins Lipoxygenme and Products Synthesized
by 0. mykiss
trienes produced by trout macrophages in response to challenge by calcium ionophore are presented in Table I. Both the 4- and 5-series lipoxins were quantitatively more important than their LTB counterparts. There was some marked variation in the relative amounts of LXA4 and LXA, derived from different rainbow trout, some producing more 5-series than 4-series lipoxins. Furthermore, 7-cis-11-trans LXA, (peak 5 ) and 6(S)-LXA, (peak 7) were often completely absent. Lipoxin generation was slower than leukotriene generation with the latter reaching a maximum after 30 min of exposure to ionophore compared with 45 min for the former (Fig. 8). Inhibition of LipoxinlLTB Biosynthesis-Biosynthesis of lipoxins and LTBby macrophages stimulated with ionophore was reducedina concentration-dependent manner by the lipoxygenase inhibitor, NDGA. Inhibition was 100% at conM/ 2 centrations >5 p ~ No . inhibition or enhancement of lipoxin and LTB synthesiswas seen in the presence of the cyclooxygenase inhibitor indomethacin even at 50 pM. ii. 00 'U:50.001 Stimulation of LipoxinlLeukotriene Biosynthesis by Biological Agents-The use of opsonized zymosan and E. coli LPS as physiological stimulants to evoke lipoxin and leukotriene synthesis, demonstrated that while zymosan was a potent stimulator of lipoxygenase product formation (Fig. 91, LPS had little, or no such activity even at 100 pg.ml" (data not 308 shown). Opsonized zymosan (5.0 mg.ml-') was as effective at I r stimulating overall lipoxin production a t 5 p M calcium ionoi phore, although therelative amounts of the differentlipoxins were altered with more LXA, and LXA, being generated but less ll-truns-LXA5and11-trans-LXA,inthe presence of zymosan. Surprisingly, opsonizedzymosancausedslightly greater enhancementof the LXAs and LTB, synthesis relative M/ 7. to LXA, and LTB, production ascompared to that seen with FIG. 6. LXAs (methyl ester, trimethylsilyl ether derivative) calcium ionophore. Leukotriene generation was stimulated to obtained from trout macrophages (HPLC component 2). Elec- a greaterextentthan lipoxin generation at low zymosan tron impact mass spectraof GC peaks eluting with retention times5 concentrations (0.25 mg.ml" and 1.0 mg.ml-').
min 45 s ( a ) ,6 min 15 s ( b ) .
However, the mass spectra (Fig. 6) show a molecular ion 2 In the many studiesof eicosanoid generation in mammals, atomic mass unitslower at mlz580 and base peakat mlz 203 very few havereported cell types capable of synthesizing (CH(OSiMe3)-(CH,),-COOCH3) indicating that these prodlipoxins entirely from endogenous fatty acids (12, 15, 18, 20) ucts arelikely to be &serieslipoxins. The peak with retention time 5 min 45 s is tentatively identified as LXA,. Further ions and then sometimes only as part of a mixed cell population (12, 19).Even when mammalian cells can produce these at mlz 565 ("15, loss of CH,), 549 ("31, loss of CH30),459 compounds from eitherarachidonic, or eicosapentaenoic ("31-90, loss of MeBiOH), 377 (M-203), and 171 acids, they do so in far smaller amounts thanfor the equiva(CH(OSiMe3)-(CH2),-CH3or 203-CH30H)areconsistent lent leukotrienes (15, 18, 19) which raises the question as to with the structureof a 5-series lipoxin. The later eluting peak whether the lipoxins have a true physiological or pathological (C"26.8) is likely to represent oncolumn isomerization to 11role. In contrast, we have found that rainbow trout macrotrans-LXA5 by analogy with the chromatography ofLXA,. phages synthesize lipoxins as majoreicosanoid products. No standards of authentic 5-serieslipoxins were available for These macrophages were of at least 95% purity as judged by comparison. HPLC component 1 showed a single GC peak a variety of histochemical and morphological criteria (data with a retention time of 6 min16 s (C-26.8) and a mass not shown) and in many experiments of apparently greater spectrum consistent with a 5-series lipoxin and was tenta- than 99% purity. Thissuggests that while a second cell type tively identified at ll-trans-LXA5. may possibly be required for lipoxin synthesis, it is unlikely Radiolabeling Studies-Macrophages, incubated with [ 1- that this is the case. Support for this comes from the obser14C]arachidonic acidand challenged with ionophore,produced vation that stimulated crudemacrophage preparations, used radiolabeled HPLC peaks as shownin Fig. 7a. This confirms directly after separation on Percoll density gradients, and 24 that peaks identified as 11-transLXA, (peak 4 ) , LXA, (peak h cultures, synthesized less lipoxin than 48-72 h macrophage 6),6(S)-LXA4 (peak7), LTB, (peak I 3 ) , 6-trans-LTB4 (peak cultures (data not shown). Trout peripheral blood leukocytes II), and 6-trans-12(S)-LTB, (peak 12) were derived from can also synthesize lipoxin but at much reduced levels comarachidonic acid. Similarly, [l-'4C]eicosapentaenoicacid pro- pared to themacrophages (22). duced radiolabeled peaks that cochromatographed with peak Of the lipoxins detected, peak 6 was identified as LXA, and 1 (tentatively identified as ll-trans-LXAa), LXA, (peak 2 ) peak 4 as 11-trans-LXA,by HPLC with referenceto authentic and LTB, (peak 8) (Fig. 76). No lipoxin-like or leukotriene- standards.Furtherconfirmation of the structure of these like peaks derived from docosahexaenoic acid were detected lipoxins was sought using GC/MS. It is well recognized that (Fig. 7c). there are difficulties inanalyzing lipoxins by this method Quantification and the Dynamics of Lipoxygenuse Product because breakdownand isomerizationoccur to avariable Formation-Therelative amounts of lipoxins and leuko- extent on GC columns. We sought to ameliorate this problem
Products Synthesized by
Lipoxins Lipoxygenase and
FIG. 7. a-c, radiolabeling of lipoxygenase products using [l-14C]fatty acids. Trout macrophage cultures (-1 X lo' cells/flask) were incubated with 1 ,uCi of arachidonic acid ( a ) ,eicosapentaenoic acid ( b ) ,or docosahexaenoic acid ( c ) for 1 h at 18 "C. After washing to remove excess radiolabel, the cells were calcium ionophore-challenged (5 PM, 30 min, 18 "C), the supernatants Sep-Pak extracted, separated on RPHPLC (100% water/methanol/acetonitrile/acetic acid; 45:30:25:0.05 (v/v), pH 5.7, changing to 100% methanol in 40 min; flow rate 0.6 ml'rnin-'; detection a t 280 and 301 nm), and collected in 0.3-ml fractions for scintillation counting. The large radiolabeled peaks on the radioactivity traces with retention times -35-45 min probably represent monohydroxy-fatty acid derivatives andunmodified fatty acids.
TABLEI Formation of lipoxins and othermajor lipoxygenase products by calcium ionophore-challenged rainbow trout macrophages Concentrations are mean values f S.D. (n = 12). Trout macrophages (-10' cells/flask) were challenged with calcium ionophore A23187 (5 FM,30 min, 18 "C), Sep-Pak extracted with 100 ng of PGB, as internal standard and separated by RP-HPLC. Identification is based on cochromatography with authentic standards, spectral analysis, radiolabeling, and for components 1,2, 4, and 6 by GC/MS. Lipoxygenase product
Conc. ngl10' cells
11-trans-LXAb LXA, 11-trans-LXA, 7-cis-ll-trans-LXA4 LXA, 6(S)-LXA, LTBs 6-trans-LTB, 6-trans-12(S)-LTB4 LTB, 5-HETE 12-HETE
1 2 4 5 6 7 8 11 12 13
2.5 f 0.1 12.9 f 4.0 5.0 f 1.9 0.6 f 0.2 23.3 f 7.4 2.4 f 1.3 6.5 f 1.8 1.5 f 0.5 2.3 f 0.8 17.7 f 4.6 2.2 +- 0.7 32.7 2.2
by using short (6 m) capillary columns which allow relatively low elution temperatures but still maintain adequateresolution. This minimizes but does not eliminate the problem of breakdown entirely. Nevertheless, it was possible to demonstrate an identical chromatographicprofile and mass spectra for authentic LXA, and for HPLC component 6 derived from trout macrophages. This taken together with the HPLC data
LT 8 4
FIG. 8. Time course of leukotriene and lipoxin generation. Trout macrophage cultures were exposed to 5 F M ionophore for 5, 15, 30,45, and 60 min. Mean values k S.D., n = 4.
strongly suggests that this product is LXA,. We found that the less labile 11-trans-LXA, elutes as a single GC peak, and we were further able to confirm the identity of HPLC peak 4 as the 11-trans product by comparison with authentic standard. Our HPLC studies including those using [l-'4C]eicosapentaenoic acid incorporation strongly suggested production of 5-series lipoxins. The GC/MS studies also showed mass spec-
Lipoxins and Lipoxygenase Products
Synthesized by 0. mykiss
lized to lipoxins in this system. It is not possible to determine the biosynthetic route to lipoxin A from the results of the c present study, but it is interesting to note theabsence of any E 40 detectable amounts of LXB, in the cell supernatants. Many 9 n studies of mammalian cells report concomitant biosynthesis of both LXA and LXB. Exceptions to this have been reported, LXA4 and in these cases it has been proposed that either LXA or LXB is formed via an epoxide pathway involving specific enzymatic hydrolysis of 5,6-epoxy-15-HETE (9). An epoxide hydrolase has recently been reported in human granulocytes T that specifically catalyzes LXB, biosynthesis from epoxide intermediate(s) provided by platelets (28) or nasal polyp cells (29). It is possible that such a system may operate in trout macrophages in this case toproduce LXA, and LXAS. The incorporation of exogenousradiolabeled fatty acids into macrophage phospholipids and their subsequent calcium ionophore-stimulated release and metabolismshowed that while arachidonic and eicosapentaenoic acids are good lipoxLT E4 ygenase substrates, docosahexaenoic acid is poor in this respect. Only a few minor components were radiolabeled with this fatty acid which is in agreement with our earlier work using trout peripheralblood leukocytes (25) and other studies with human platelets and neutrophils(30). Whereas calcium ionophore can be employed as a useful tool to investigate the capacityof cells to synthesize various mediators,itrepresents anonphysiological stimulus.We therefore looked at two more appropriate biological stimuli for macrophage activation and found that opsonized zymosan caused biosynthesis of eicosanoids to a similarextentas ionophore. However, the pattern of lipoxygenase products Opsonized Zymosan ( m g / m l ) differed. Zymosan apparently enhanced LTB, and LTB, proFIG. 9. Leukotriene and lipoxin generation by trout macrophages challenged withopsonized zymosan relative to that duction to a greater extent than LXA, and LXA5. These obtained with calcium ionophore. Trout macrophages were stim- results with foreign material suggest that lipoxins are probtroutduringinflammatoryreulated with opsonized zymosan (0.25, 1.0, 2.5, 5.0, or 10 mg.rn1-l for ably generated inrainbow 2 h a t 18 "C), the supernatants extracted together with 100 ng of sponses. Ourinitialchemotaxis/chemokinesisexperiments PGB2 as internal standard and separatedby RP-HPLC. The results with leukocytes from this species (22) also suggestthat lipox(mean values ? S.E., n = 4) are expressed as percentages of that ins are more potent than LTBl at mediating such events. Ill-m-LXA4
obtained using calcium ionophore (5WM,30 min, 18 "C).
Acknowledgments-Special thanks aregiven to Dr. Charles Serhan,
tra consistent with lipoxinsderived from eicosapentaenoic HematologyDivision,Brigham and Women's Hospital,Harvard provision of acid. By analogy with the chromatography of LXA, and 11- Medical School,Boston for hisencouragementand trans-LXA,, we were able to tentatively identify LXA, and lipoxin standards. The diligent technical assistance of A. Hopkins is also gratefully acknowledged. 11-trans-LXA, in HPLC fractions2 and 1, respectively. HPLC peak 5 coeluted with the recently identified 7 4 11-trans-LXA, (27) in two different solvent systems but inREFERENCES sufficient material was available forconfirmation by GC/MS. 1. Belch, J. J. F. (1989) Prostaglandins Leukotrienes Essent. Fatty Similarly,peak 7coeluted with authentic 6(S)-LXA, and Acids 36,219-234 2. Cowell, A. M., andBuckingham, J. C. (1989) Prostaglandins peak 3 from its radiolabeling, and its HPLC characteristics Leukotrienes Essent. Fatty Acids36, 235-250 may be the equivalent 5-series lipoxinisomer, 6(S)-LXAs. 3. Lagarde, M., Gualde, N., and Rigaud, M. (1989) Biochem. J. 257, These compounds were present inonly a minority of samples 313-320 and thereforemay be of limited biological significance at least 4. Lote, C. J., and Haylor, J. (1989) ProstaglandinsLeukotrienes in this species. Essent. Fatty Acids36, 203-217 5. Needleman, P., Turk, J., Jakschik, B. A,, Morrison, A. R., and Variation in lipoxin production between macrophages obLefkowith, J. B. (1986) Annu. Reu. Biochem. 5 5 , 69-102 tained from different trout was quite marked witha few fish 6. Smith, W. L. (1989) Biochem. J. 259, 315-324 producingmoreLXAB than LXA,, or more 11-trans-LXA, 7. Serhan, C. N., Hamherg, M., and Samuelsson,B. (1984) Biochem. than LXA,. We were unable to accountfor this by differences Biophys. Res. Commun. 118,943-949 in extraction efficiencies: neither was lipoxin isomerization 8. Serhan, C. N., Hamberg, M., and Samuelsson, B. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, 5335-5339 during these procedures a cause since it did not occur a t any 9. Puustinen, T., Webber, S. E., Nicolaou, K. C., Haeggstrom, J., detectable rate using our extraction method. Less variation Serhan, C. N., and Samuelsson, B. (1986) FEBS Lett. 207, was seen in leukotriene and monohydroxy fatty acid genera127-132 tion suggesting that lipoxin biosynthesis may be more sensi- 10. Rokach, J., and Fitzsimmons,B. (1988) Int. J . Biochem. 20,753tive to genetic, dietary, and/or environmentaldifferences. 758 Trout macrophages contain both 5- and 12-lipoxygenases 11. Serhan, C. N., and Sheppard, K-A. (1990) J . Clin.Inuest. 85, 772-780 as judged by the presenceof 5- and 12-HETEs in supernatants 12. Edenius, C., Haeggstrom, J., and Lindgren, J. A. (1988) F E B S of ionophore and zymosan challenged cells. No 15-HETE was Lett. 228, 167-171 detected but this does not rule out 15-lipoxygenase activity 13. Kim, S. J. (1988) Biochem. Biophys. Res. Commun. 150, 870because 15-HPETE may well be a substrate readily metabo876
Lipoxins Lipoxygenase and Products Synthesized
by 0. mykiss
14. Lam, B. K., Hirai, A., Yoshida, S., Tamura, Y., and Wong, P. Y.24. Pettitt, T. R., Rowley, A. F., and Barrow, S. E. (1989) Biochim. -K. (1987) Biochim. Biophys. Acta 917, 398-405 Biophys. Acta 1003, 1-8 15.Walstra,P., Verhagen, J., Vermeer,M. A., Klerks, J. P. M., 25. Pettitt, T. R., and Rowley, A. F. (1990) Biochim. Biophys. Acta 1042,62-69 Veldink, G. A., and Fliegenthart, J. F. G. (1988) FEBS Lett. 26. Barrow, S. E., and Taylor, G. W. (1987) in Prostaglandins and 228, 167-171 RelatedSubstances:aPracticalApproach (Benedetto, C., 16.Lam,B. K., and Wong, P. Y.-K. (1988) Adu. Exp. Med. Biol. McDonald-Gibson, R. G., Nigan, S., and Slater, T. F. eds) pp. 229,51-59 99-141, IRL Press, Oxford. 17. KGhn, H., Ludwig, p.,Salzmann-Reinhardt, U., Hohne, M., and 27, ~ i ~K, c,, ~ Ml ~ ~ B,~ E.,~ ~veale,,~ c, A,,~webber, ~ s, ,E., Rapoport, S. M. (1987) Biomed. Biochim. Acta 46, S156-Sl59 Dahlen, B., Samuelsson, B., and Serhan, C. N.(1989) Biochim. 18. Serhan, C. N. (1989) Biochim. Biophys. Acta 1004, 158-168 Biophys. Acta 1003, 44-53 19. Serhan! c. N.* Hirsch! J.* andSamuelsson~ B. 28. Edenius, C., Forsberg, I., Stenke, L., andLindgren, J. A. (1991) (1987) FEBS Lett. 217, 242-246 in Advances in Prostaglandin, Thromboxane, and Leukotriene 20. Lam, B. K., Serhan, C. N., Samuelsson, B., and Wong, P. Y.-K. Research (Samuelsson, B., Ramwell, P. W., Poaletti, R., Folco, (1987) Biochem. Biophys. Res. Commun. 144, 123-131 G., and Granstrom,E., eds) Vol. 21A, pp. 97-100, Raven Press, 21. Pettitt, T. R., Rowley, A. F., and Secombes, C. J. (1989) F E B S NY Lett. 259, 168-170 29. Lindgren J. A., Edenius, C., Kumlin, M., Dahlen, B., and Anggard, 22. Rowley, A. F., Pettitt, T. R., Secombes, C. J., Sharp, G. J. E., A. (1991) in AdvancesinProstaglandin,Thromboxane,and Barrow, S. E.,andMallet, A. I. (1991) in AdvancesinProsta-LeukotrieneResearch (Samuelsson, B., Ramwell, P. W., Poalglandin,ThromboxaneandLeukotrieneResearch (Samuelsson,etti,R., Folco, G., andGranstrom. E., eds) Vol. 21A, pp. 89-92, B., Ramwell, P. W., Poaletti, R., Folco, G., and Granstrom, E., Raven Press, NY eds) Vol. 21B, pp. 557-560, Raven Press, NY 30. Fischer, S., Schacky, C. V., Siess, W., Strasser, Th., and Weber, 23. Secombes, C. J. (1985) J . Fish Dis. 8,461-464 P. C. (1984) Biochem. Biophys. Res. Commun. 120, 907-918 u.l