Biochimlca et BJophysica Attn. 1085 (1901) 191-200

191

~3 1991 Elscvier Science Publishers B.V. All righls reserved 0Da5-2760/91/SII3 ~0 A D O N I S 0005276[)91~)244T BBALIP 53703

Arachidonic acid metabolism during antigen and ionophore activation of the mouse bone marrow derived mast cell T s u t o m u N a k a m u r a ~,2 A l f r e d N . F o n t e h 3 W a l t e r C . H u b b a r d ~, M a s s i m o T r i g g i a n i 3 N a o k i I n a g a k l L2, T e r u k o l s h i z a k a 1.2 a n d F l o y d H . C h i l t o n 3 ! Departmen! ojt~h, dicille, Tll~]dhtE~ I]ol~kltl~ [dllil','~kt}" &'llool o]Medicim', Baltimore. MD (US, A,L 2 l.aJolla h~liltll~']orAllt'v~,~ • and b):nulnologv, LaJolht, CA fUS.A ) at)d ¢ Tile J¢lhns tIopl~itls Asthma & .+lllt'rt~ ('enter. T71e]¢)hns Hopkins IJ'tzitcrsay &'hool o f Mtdit hle, Baovnvre. M D I~,S.A.)

(Received f, February 1991)

Key words: Arachidonleacid metabolism:IonophoreA23187 Immunologicalchallenge:Antigen; Gf'-MS: (Mast cell)

This study has examiaed the metabolism of arachidonic acid in the mouse bone real rnw.derived mast cell (BMMCI during immunologic and nonimmunologic activation. The predominant pools of endogenous arachidonate in the mast ceils were found in ethnnolamine (,16%), choline (39%) and inositol (14%) containing glycerolipids. Initial stadles established conditions where equilibrium labelling of these major pllmspnolipids in the BMMC could bo reached. Upon challenge, arachidonate was lost from all major phosphnlipid classes (phosphatidylethanolamine > phosphatidyleholine > phosphatidylinositol). There was a small hut significant inct~rase in the amount of label associated with phosphatldic acid during cell activation. Arachidonate was distribotcd among l-aeyl, l.alkyl and l-alk-l-enyl-Iinked subclasses of PC and PE. The rank order of loss of labelled arachidonate from the major PE and PC subclasses during antigen and iunophore activation was I.alk.enyl-2.acuchidonoyI-GPE > 1-acyl-2nrachldonoyI-OPC > l-acyl-.?,-arochidonoyI-GPE > l-alkyl-2-arachidonoyI-GPC. Labelled products released into the xupernatant fluids and free arachidouic acid within the cell accounted for the bulk of arachidonate lost from phosphonpids. Labelled products in the supernatant fluids were composed of LTB4, LTC4, PGD z and free arachidouic acid. BMMC phospholipids were also labelled for 24 hr with ]3H]choline, ]3H]myoinositol or [ 14H]ethanolamine and labelled 2-1yso phospholipids were measured after cell activation. Radioactivity in lysopbosphonpids from PC, PE and PI increased significantly between 30 s and 2 rain after antigen aetivalion and then declined. Taken together, these studies suggest that arachidouate is mobilized predominantly from PE and in particular 1-alk-l-enyl-2-arachidouoyI-GPE by Ihe direct removal of arachidonate from the s o - 2 position of the molecule. Most of Ihis arachidoaate is then released from cells as eicusanoids or free fatty acid.

Introduction

Abbreviations: BMMC, mouse bone marrow-derived InlLMcell; PC. choline-linked phnsphogl¥cerides; PE, uthanolamine-linked phosphoglyceridcs; PI, inositol-linked pho~phoglyceride~: GPC, s,giyccro-3-phosphocholine; GPE, sn-glycero-3-phasphoelhamdaminc: GPL sn-glycero-3-phosphoino~itol; GC-MS, gas chromatographyma~ spectrometod: I-IPLC, high-pressure !iquld chn)matography: HBSS He,nk's balanced sah solutlori; HSA. fatty acld-free human serum albumin; LTB4, leukotrien¢ Ba; HETE, bydroxyeicosatetracnoic ~cid: LTC 4 leukotriene C4; PGDz. prostaglandin D 2 and TxB2 Ihromboxanc B~. Correslalndencc: Floyd H. Chihon,Section on PuMaonuryMedicine. Medical Cenlcr Boulevard. Winstun-Salcm, NC 27t57, US.A

It has b c c n r e c o g n i z e d for sometime that one of the initial events in the activation of mast cells is the crosslinking of [gE o n Fc r e c e p t o r s ( F e e R I ) with specific a n t i g e n [I,2], This e v e n t leads to a c o m p l e x series of biochemical changes which includes plasma m e m b r a n e depolarization, phospholipid turnover, calcium influx a n d c h a n g e s in cyclic aucleotides [3-6]. In addition to th.:sc events, m a s t cells p r o d u c e a n d secrete a unique sea of arachidonic a~.id metabolites (eicosanoids)

during cell activation [7-10], the quantities and nature

192 of the mctabolites depend on which enzymes and precursor pools are available in the various mast cells. Eicosanoids are produced in cells by a series of enzymatic steps all working in c~.~ncert. The initial step involves tile mobilization of arachldonic acid from the sn 2 position of membrane phospholipids [11]. This is thought Io be accomplished directly by phospholipase A 2 which cleaves araehidonie acid from thc sn-2 position of the moleculc [12-14]/Alternatively, enzymes such as phospholipasc C remove the phosphobase moiety of phospholipid molecules to generate diacylglyceridcs [15.16]. Diacylglyccrides have also r,~cently been shown to be formed from chollne-containlag phospholipids via phospholipasc D followed by phosphohydrolase [17-20]. Arachid(mic acid may then be mohilized front thcsc diaeylglyccrides by the action of diaeylglycerol lipasc Is) a n d / o r monoglyccrol lipasc [21-23]. Although many potential mechanisms for arachidonate release have been proposed, little is known about the arachidonate-containing phospholipids which are hydrolyzed during cell activation. In the mast cell, phosphoglyceridc sources have typically been identified by studies in which cells were prelabellcd with radioactive arachidonic acid before stimulation. Utilizing this approach, phosphatidyleholine arid phosphatidylinositol have been suggested to contribute the bulk of the substrate arachldonatc for elcosanold biosynthesis durtug cell activation [7,24 26]. In contrast to data in the mast cell. it has recently been demonstrated that in the ncutruphils stimulated with the ionophore A23187, phosphatidylethanolamine provides 3 to 4-times as much araehidonatc as choline or inositol-linked phospholipids [27 29]. The apparent differences in which phospholipids release arachidonate during cell activation between mast cell and neutrophil may arise for several reasons. Perhaps the more physiologic stimulus, antigen mobilizes different sources of arachidonate than ionophorc A23187. Alternatively, there may be fundamental differences in the mechanism of re!ease tlf araehidonulc for the mast cell and the human neutrophil. Finally, there may have been fundamental differences between the methods utilized in the neutrophil and mast cell studies that led to discrepancies in the sources, hi particular, the mole quantities of arachidorute mobilized from phospholipids were determined by GC-MS in the ncutrophil study while radiolabellcd arachidonate was analyzed in the mast ccll studies. In the present study, we have examined the release of [3H]araclaidonate fl'om phospholipids during antigen and ionophorc A23187 stimulation in the mouse b{~ne marrow-derived mast cell to better understand which arachidonate-containing phospholipid classes are utilized during immunologic activation. In these experiments, wc have utilized GC-MS to quantify unlahclled arachidonate in complex lipids to assure that the amount of labelled arachidonate in a

given phospholipid reflects the mass of unlabelled arachidonic acid in that same phospholipid. Using conditions where equilibrium labelling is obtained, the present study suggest that during the actiwition of the mast cell with antigen and ionophore A23197, ethanolamine-containing phospholipids prtwide the bulk of the arachidonate released from the ceil. A large percentage of this released arachidonatc is then converted to a variety of products.

Materials and Methods

Mast cell preparations Bone marrow-derived mouse mast cells (BMMC) were obtained from suspension cultures of bone marrow cells of C B A / J mice (Jackson Laboratories, Bar Harbor, ME) in RPMI 1640 culture medium (GIBCO, Grand Island, NY) supplemented with 10% (v/v) fetal calf scrum, 50 p.M 2-mercaptoethanol, 2 mM L-glutamine and antibiotics. The medium was enriched with 5% (v/v) of medium conditioned with St5 72F-DII which consistently produce mouse IL-3 [30]. The IL-3 producing cell line was kindly supplied by Dr. N. Arai, D N A X Institute of Molecular Biology (Paid Alto, CA). After 4 weeks of culture, more than 95% of non-adherent cells in the culture were mast cells, with a viability of higher than 98% as determined by Trypan blue exclusion test. The cells were recovered at 4 weeks and employed in these experiments.

Mouse lgE antibody, antigen and passice sensitization Purified monoelonal mouse IgE anti-DNP antibody and dinitrophenyl derivatives of human serum albumin (DNP-HSA) were the same preparations as those described in previous studies [31,32]. BMMC were passively sensitized by culture of the cells with 10 ,o,g/ml mouse IgE anti-DNP antibody overnight. After washing, sensitized cells were suspended in Tyrode solution (pH 7.4) containing 124 mM NaCI, 4 mM KCI,0.64 mM NaH2PO4, 1.6 mM CaCI e, 1 mM MgCI.,, 5.5 m M glucose, 10 mM Hepes and 0.05% (v/v) gelatin or Hank*s Balanced Salt Solution (HBSS).

Incorpocation of /~H/arachidonic acid into glycerolipids BMMC were labelled tapproJ:. I " 10"/ml) by adding 5 /zCi of [3H]AA (209 C i / m m o l ) eomplexed to albumin to a culture medium at 3 7 ° C in a humidified atmosphere of 5% C0,./95% air. For antigen stimulation, BMMC were incubated with [JH]AA together with 10 p.g/ml mouse anti-DNP lgE antibody for 24 h. Labelled cells were washed 3-times with Tyrodc's or HBSS containing (I.25 m g / m l humt:n serum albumin, and resuspended in Tyrode's or HBSS solution.

lq3

Cell actication Cell suspension containing BMMC sensitized with anti-DNP IgE antibody were incuoated for the indicated times at 3 7 ° C with DNP-HSA or ionophore A23187 (2 #M). An optimal concentration of the DNP-I-ISA for maximal histamine release was 10 n g / m l (DNP21-HSA) or 2 ,ug/ml (DNPv-HSA). We confirmed that antigen stimulation did not significantly effect the viability of the cells (viability > 98% before stimulation and > 96% after stimulation). The reactions were stopped by centrifugation (225 × g, 4 ° C). Supernatant fluids were added to ethanol at a final proportion of supernatant fluid/ethanol (1 : 3, v/v). Lipids in the cell pellet were extracted by the method of Bligh and Dyer [331.

Chromatography of phosphoh)Tids Phospholipids were separated using several different chromatography methods. In initial experiments, it was important to separate all major phosphoglyceride classes, including phosphatidyllnosltol from phosphatidylserine. This was accomplished using a two-dimensional TLC system on layers of Silica Gel G developed in c h l o r o f o r m / m e t h a n o l / 2 8 % N H 4 O H / H . ~ O ( 6 5 : 3 5 : 3 : 2 , v/v) in the first dimension and chlorof o r m / a c e t o n e / m e t h a n o l / g l a c i a l acetic a c i d / H z O ( 1 0 : 4 : 2 : 2 : 1 , v / v ) in the second dimension [34]. Under these conditions, the respective R~. values for the first and second developments were 0.10 and 0.65 for PA, 0.15 and 0.45 for PS, 0.19 and 0.34 for PL 0.31 and 0.51 for PC, 0.46 and 0.60 for PE. Once it had been established that PS eonlailmd < 2% of the total arachidonate, a one-dimensional system was utilized to separate phospholipid classes. Here phospholipids were separated on Silica Gel G developed in chloroform/ methanol/acetic a c i d / w a t e r (50 : 25 : 8 : 2.5, v/v). Neutral lipids were separated by TLC on layers of Silica Gel G developed in hexane/ethylether/formic acid 0 0 : 6 0 : 6 , v/v). In the specific activity experiments, it was crucial to obtain h~gh recoveries of the phospholipid after chromatography. Here phospholipid classes wcrc separated by normal phase HPLC using an Ultrasphere-Si column (4.6 X 250 ram; Rainin Instrument Co., Woburn, MA) eluted with 2-propanol/Z~; raM-phosphate buffer (pH 7 . 0 ) / h e x a n e / e t h a n o l / a c e t i c acid (367:30:490: 100:0.6, v/v) at a flow rate of l.ll m l / m i n for 5 min [35,36]. After 5 min the solvent was changed to 2-propanol/25 mM phosphate b u f f e r / h c x a n e / e t h a n o l / acetic acid (367 : 50 : 491) : i 0(.~: l).6, v/v). The amount of label in each of the glycerolipids was determined by liquid scintillation counting. Isolated phospholipids were then prepared for gas chromatography-mass spectrometry (GC-MS) analysis. Purified choline and ethanulamine phospholipids were further separated into l-acyl, l-alkyl and I-alk-l-enyl subclasses as de-

scribed by Nakagawa et al. [37]. Briefly, these phospholipids were hydrolyzed to l-sadyl-2-acyl-glyccrols with phospholipase C (from Bacillus cereus) in 100 mM Tris-HCl buffer (pH 7.4). The l-radyl-2-acyl-glycerols were then extracted and converted into 1,2-diradyl-3acetylglycerols with acetic anhydride and pyridine for 18 h at 37°C. The 1,2-diradyl-3-acerylglycerols were then separated into 1-acyl, l-alkyl and 1-alk-l-enyl subclasses by TLC on layers of Silica O¢1 0 developed in be nz e ne / hc xa ne / di e t hyl ether (50: 45 : 4, v/v).

Chromatography of eicosanoids Leukotricnes, prostaglandins, thromlx, xanes and free arachidonic acid fcmnd in supernatant fluids were separated by reversed-phase HPLC using modification of a previously described method [38], Labelled components in the supernatant fluids were loaded on an Ultraspherc O D S column 14.6 × 2511mm) and eluted at l m l / m i n with acetonilrile/water/trifluoroacetic acid (33:67:(I.I. v/v) for 35 rain. The percentage of acetonitrile was then increased from 33% to 45% over 5 min. Then, the pcreemage of acetonitrile was increased from 45% to 100% over 6t) rain. The eluate was continuously monitored for absorbance at 210 nM. I ml fractions were collected and the radioactivity determined by liquid scintillation counting.

GC-MS of arachidonate The mole quantity of unlabelled arachidonate was determined by hydrolyzing the glycerolipids with 2 MKOH in methanol/water (3: I, v/v) for 30 rain at h0 ~C to liberate the esterificd arachidonate. ]z H s]ara" chidonic acid (250 ng) as an internal standard was added to the reaction mixture. After 30 min. additional water was adned and the pH was adjusted to 3.0 with 6 M HCI. The ?tee arachidonic acid was extracted with hexane. The arachidonic acid was then converted into the t-butyldiraethylsilyl ester &s de~ribed previously [27]. The solvents were removed from the sample with a stream of N, and the sample suspended in hexane. The /-butyldimethylsilyl m uchidonate was analyzed by GC-MS using selected ion-monitoring techniques to monitor the M - 5 7 ion at m / z 361 and the tbutyldimethylsilyl ['iHs]arachidonate M - 5 7 ion at m / z 369. GC-MS was carried out on an HP selective mass detection system (HP 5790). G C was pelformed using a cros~s-Iinked methylsilicone column. This col. umn was threaded into the mass spectrometric ion source. The initial column temperature was 80 ° C and programmed to 2 2 0 ° C at 3 0 ° C / r a i n . At that point, the temperature was increased I t l ° C / m i n to 270~C. The injector temperature was 250 ° C and the GC-MS interface line was at 281)°C. Helium was used a~ a carrier gas, and I - 2 ,~1 of the samples was injected in the splifiess mode. Specific radi.activities of tacit arachidonatc-containing class were determined from

Ig4 the label attd mass measurements and expressed as CPM/nmol.

GC-MS of prostaglandins Allquots of the medium for prostaglandin measurement by GC-MS were immediately transferred to silanized glass vials containing 0.65 to 3 5 4 ng each of 3,3',4,4'-tctradeuterated (2H 4) analogues of PGF~., PGE., and 6-keto-POFl., PGD 2 and TxB 2 as internal standards. The vial contents were dried under a nitrogen stream and tile sample residue was treated in sequence with reagents for synthesi~ of methyloxlmepentafluorobenzyl ester-trimethylsilyl ether derivatives as described [39]. Combined capillary gas chromatography-negative ion mass spectrometry was performed with a Finoigan M A T T S Q 700 G C / M S / M S / D S (Finnigan M A T Corp., San Jose, CA) operated as a single quadrupolc system. Negative ions were acquired at m / z 524 (PGE2 and PGDz), m / z 569 (9a, 11/3-PGE 2 and PGFz.) and m / z 614 (TxB2 and 6kPGFz.). Signal ions 'H., PG analogs employed were recorded at m / z 528, 573 and 618.

Determ#lalion of l}'sophosphatidylcholine (lysoPC') lysophosphatidylinosilol (ly~oPl) and lysophosphatidylethanolamine (lysoPE) BMMC (2" 10 (' ceils/roll were incubated with I0 uCi [Me3H]choline chloride (81.80 Ci / mmol , Amersham) for lysoPC, Ig ~Ci myo-[2-3H]inositol (10-20 C i / m m o l , Amersham) for lysoPl, or 2 ~Ci [2~4C]ethan-l-ol-2-aminc hydrochloride (50-60 m C i / mmol, Amersham) for lysoPE. After 18 to 24 h, the cells were washed and resuspended in fresh Tyrode solution (10 I' ceils/400 txl) challenged with DNP-HSA at 3 7 ° C . The reactions were terminated by the addition of 5(10 a l of ice-cold methanol followed by I ml of chloroform. The extracts were vortexed, centrifuged at 500 × g for 10 rain and the upper phase was discarded, The lower phase was cvaporatcd under Nz, dissolved in 30 p.l of chlorolorm containing lysoPC, lysoPI and lysoPE (300 /xg), and analyzed by TLC using chlorof o r m / m e t h a n o l / a c e t i c a c l d / H 2 0 (51):30:8:4, v/v). The R~ values of lysoPC, lysoPI and lysoPE were 0.10, I).25 and 0.37, respectively.

Reagents 2-Mercaptoethanol,

1,2-diolcoyl-rac-glycetol,

l-

monoolcoyl-rac-glycerol, formic acid, phosphatidic acid, phosphatidylserine, phosphatidylinositol, phosphatidylcholine, phosphat[dylethanolamine, lysophosphatidylcholine, lysophosphatidylinositol and lysophosphatldylethanolamine werc purchased from Sigma (St. Louis, MO). myo[-2-3H]lnositol, [Me-~H]choline chloride, [5,6,7,8,11,12,14,15-3H]arachidunic acid, [2-t4C]ethan1-ol-2-amine hydlochloride were purchased from Amcrsham (Arlington Heights, ILl. The 2H 4 analogues

of PGE 2, PGF2, ,, 6KPGF],,. PGD 2 and TxB 2 used as internal standards, were purchased from Biomol Research Laboratories (Plymouth Meeting, PAl. Unlabelled 9a. I Ifl-PGF 2, was also procured from Biomol Research Laboratories (Philadelphia, PAL The calcium ionophorc A23187 and unlabelled standards of PGE_,, PGD 2, PGF2, ,, TxB 2 and 6KPGF~. were supplied by Sigma Chemical Company. Methoxamine HCL and pyridine (acetylation grade) used in the synthesis of methyloxime derivatives were products of AIItech/ Applied Science Associates. Reagents used for synthesis of pentafluorobenzyl esters, (diisopropylethylamine. pentafluorobenzyl bromide, and silylation grade acetonitrile) and trimethylsilyl esters (BSTFA) were purchased from Pierce Chemical Co. All solvents ,vere purchased from Fisher Scientific. Results

Equilibrium labelling of arachidonate pools in BMMC Preliminary experiments were carried out to determine the distribution of endogenous stores of arachidonate within glycerolipid classes and the kinetics of uptake of exogenously-provided [3H]arachidonie acid i,lto these classes. The predominant pools of endogenous arachldonate within BMMC as delivered by G C / M S were found in ethanolamine (46%), choline (39%) and inosifol (14%) containing glyeerolipids (Table I). Phosphatidylscrinc contained less than 2% of the arachidonate found in BMMC. The araehidonic acid provided to the cells was initially incorporated into PC and PI (data not shown). In contrast, there was little incorporation of exogenous arachidonic acid into the large endogenous pools of arachidonate found within PE at the early time points. However. the levels of [3H]araehidonate in PE increased throughout the 24 h cell culture period. In most experiments, the distribution of labelled araehidonate mimicked the distribution

TABLE I

Specific aeality analysisof phospholipid classes Mast cells were labelled with laH]AA for 2,1 h as described in the Ex!7~rimental Pro~dures. The amount of radioaenvay and arachldonote in each pho~phogl~eerideclass was Jetermined by liquid scinlillotion counting and GC-MS, respectiveP, as described. These data {Ire fTOlUone experiment and arc r~pl:~cnlativc of two separate expcriment~ Glycemlipid class Neulral lipid PE PI PS PC

[~H]AA (cpm- 10s/ 5" Illr' cells) I).,17

AA {ampb 5' 10t~~ells)

Spe¢. act. (cpm" lOS/ nmoD

23.10

3.40 1.00 n.I I 2.79

6.79 5.73 s.18 ~,.sl

5.73 P.t)lJ Iq.(~l

195 of e n d o g e n o u s a r a c h i d o n a f e w i t h i n various p h o s p h o glyceride classes a f t e r i n c u b a t i n g t h e cells w i t h [ ] H ] A A for 24 h. This fact is reflected by t h e u n i f o r m r a d i o s p e cific activities o f t h e p h o s p h o g l y c e r i d e classes (Table I). T h e s e e x p e r i m e n t s provided c o n f i d e n c e t h a t in B M M C labelled for 24 h, any loss of labellcd a r a c h i d o n a t e f r o m phospht}lipids d u r i n g cell activation r e p r e s e n t s a loss of e n d o g e n o u s a r a c h i d o n a t e . T h e r e was s o m e variability a m o n g d i f f e r e n t cell b a t c h e s in t h e rate at w h i c h e q u i l i b r i u m w a s r e a c h e d ; s o m e of the B M M C cultures h a d n o t r e a c h e d e q u i l i b r i u m within 24 h. Only d a t a f r o m B M M C cultures t h a t h a d r e a c h e d e q u i l i b r i u m labelling c o n d i t i o n s w e r e used in this study. T h i s was f o u n d to be crucial because B M M C labelled for s h o r t e r p e r i o d s o f t i m e yielded inconsistent results c o n c e r n i n g t h e p h o s p h o l i p i d sources o f a r a c h i d o n a t c . P h o s p h o lipid classes w e r e also ,separated into individual molecu l a r species by reverse p h a s e H P L C to assure t h a t t h e labelled a r a c h i d o n a t e provided to t h e cells e l u t e d with a r a e h i d o n a t e - c o n t a i n l n g p h o s p h o l l p i d s . D a t a f r o m this e x p e r i m e n t indicated t h a t t h e bulk of t h e label in t h e mast cell w a s i n c o r p o r a t e d i n t o p h o s p h o l i p i d s as a r a c h i d o n a r e a n d h a d n o t b e e n c o n v e r t e d to a significant e x t e n t to o t h e r fatty acids ( d a t a n o t s h o w n ) . Mobilization o f arachidonate f r o m pho~pholipids d , ring antigen aclit'ation U t i l i z i n g t h e labelling strategy described above to o b t a i n e q u i l i b r i u m labelling, B M M C w e r e s t i m u l a t e d w i t h a n t i g e n for 10 rain a n d t h e a m o u n t of radioactivity in e a c h p h o s p h o l i p i d class w a s d e t e r m i n e d . As illust r a t e d in T a b l e II, radiolabelled a r a c h i d o n a t e w i t h i n P E . P C a n d PI was r e d u c e d by a n t i g e n challenge. In c o n t r a s t , t h e r e w a s a small b u t significant increase in t h e a m o u n t of labelled a r a c h i d o n a t e a s ~ c i a t e d w i t h PA. In fact, t h e a m o u n t o f a r a c h i d o n a t e in P A inc r e a s e d g r e a t e r t h a n 25-fold a f t e r cell activation w i t h a n t i g e n . D a t a in t h e s e e x p e r i m e n t s s u g g e s t e d t h a t P E

0

1

2

3

5

10

Fig I. Kinutic~ ot ;mtigen-induccd generation of l-rad¥1-2-1ysoGPE {IvsaPE). I radyl.2dys;oGPC (I~soPC)and l.radyl-2dygoGPI (l~,goPl), Sensitized BMMC v.ere blosynt hetically Iabelled with 13| Ikholine for the determination oF ly~P(" to, o). ~ith m~o-[3Lgin~ilol for lysoPl ( A,~ ) and with Itlg?lelhanolamine for lysoPE t [].D ) then ineubal':d with (closed symbols) or withoul (open symbols) antigen. The reaction was stopped at indicated intel~'als by the addition of ice-cold mctha,ol, lysoPC, lysoPl and lysoPE were anahrzed by thin-layer chrt*matograpby. Data shown in lhe figure are the mean valu~ of three independent cxperimenls. provided t h e bulk of t h e a r a c b i d o n a t e lost f r o m all p h o s p h o l i p i d s d u r i n g a n t i g e n activation. Logic h o l d s t h a t if a r a c h i d o n i e acid is lost f r o m t h e sn-2 position of phospholipids, 2-1ysoPE. 2-1ysoPC a n d 2-1ysoPI s h o u l d be f o r m e d in this reaction. H o w e v e r , it h a s b e e n quite difficult in a n u m b e r o f cell systems to m e a s u r e these m e t a b o l i c intermediates. This is probably d u e to t h e rapid reaeylation o f 2-1ysopbospholipid w i t h a fatty acyl chain. I n a n a t t e m p t to m e a s u r e t h e ~ i n t e r m e d i a t e s , B M M C w e r e labelled for 24 h w i t h choline, myo-inositol o r - e t h a n o l a m i n e . Ceils labelled in this fashion w e r e t h e n s t i m u l a t e d w i t h a n t i g e n a n d t h e a m o u n t of radioactivity m i g r a t i n g by T L C with 2-1yso p b o s p h o l i p i d w a s d e t e r m i n e d . Fig. I shows the kinetic analysis of t h e g e n e r a t i o n o f l-radyl-2-tyst~GPE. I-radyl-2-1ysoGPC a n d I-radyl-2-1ysoGPI following

TABLE it Distribu.on o / I ~1I]urach,dtmic at &l in ph.~phohpld~ hefi,.¢ and after ,mtrgtgl stt,nl,hlm.l []HIAA labelled BMMC ~nsitizL,d with ami-DNP IgE antibody were incubated in the presence and absenc~ of DNP liSA (If) ng/ml) al 37 °C for Ill rain, and the distribution of [ 3IItAA in 0hugoholipids was analyzud DNP-HSA

(- ) (+ )

[3HlArachidonic acid ill " (cpm' IO~/lU" cells) PA PS PI ().bh

28,65

I 13 3fi

(()1~)8)

( I"i7)

( l l'hO)

14.87* * {3,17J

27 85

g 1.38* (b.42)

I 12 " )

PC"

PE

23~J92

482.79

(8"75!

(53"KS}

2(11.I0 * (13.20)

367,63 * (32.g0)

t listamin¢ release (C~I 1.7 ± 0.1 74.5 + 2.9

" Each number in this column is a mean (ff three experiments with S £ in paremhcsis. The changes in [3HIAA cpm in each ph~pholipid aher antigen cha!lenge v.~re ¢valualed by unpaired t-test, P %,;d11¢,;ire ().(h)15for PA, 0.45 for PS (insignificant), a.ol4 f()r Pl. IOH7 for P(' and 0.g34 for PE.

196

antigen challenge. The radioactivity in all three of these lysophospholipids increased significantly within 30 s after antigen challenge. The amount of 2-1ysoGPC and 2-1ysoGPI reached a maximum between 60 and 90 s and rapidly dcclined. Ly.soGPE showed a slower kinetic reaching a maximum at 2 rain with a more gradual decline. Fate o f arachidonic acid released f r o m phospholipid classes The aklrcmcnlioned experiments measured the loss of arachidonatc from thc major phospholipid classcs of B M M C upon antigen stimuhltion. This loss can arise from the direct deacylation of arachidonate or from the conversion of araehidonate-containing phospbolipids to other products such as dlacylglycerol and phosphalidic acid. In order to gain a better understanding of the fate of arachidonate mobilized from phospholipids during cell activation, a ~eparate set of cxpcrintcnts wcrc undcrtakcn to accoulit for the arachidonatc in various glycerolipids and compartments {cell vs. supcrnatant fluid) after cell activation. H m c . B M M C labelled with [ ~ H ] ~ to equilibrium wcrc stimulated with antigen or ionophore A23187. The previous set of experiments, measuring the 2lysophospholipids, and other e x p e r i m e n t s ' m e a s u r i n g the kinetics of loss of arachidonate from phospholipids (data not shown) revealed that tile major losses of arachldonic acid probably take place within the first 5 min. Therefore, in these experiments, arachidonatecontaining products were measured 5 min after cell activation. Fig. 2 shows that after both antigen and ionophorc A23187 stimulation, there is a 10 and 2 0 % increase above control, respectively, in the percentage of the total radioactivity found in the supcrnatant fluids. If labelled arachidonalc is in equilibrium with endogenous pools, as Table 1 implies, these data suggest that relatively large quantities of the arachidonate fmm endogenous phospholipids are eventually found outside the ceil during cell activation. In addition to the labelled products in the supernatant fluids, cell

{ go g0

-tO 20

NL P~"

Cell Sup

Anligen A231S7

PItPS

Fig, 2. Accnnnting tar Ihe loss of labelled alachldonate from phospholipids. BMMC labelled as described were incubated for 5 rain wilh no stimulus, antigen or ionophore A23187. The BMMC were removed from Ihe supernalam fluids by eentrifugation. The lipids were exlraetetl from the ~ll pellet and Ihe glycerolipid classes ~paraled by TLC. The amount of radioactivity in supematant fluids and each glyeerolipid class "~as determined and expressed as a percentage of Ihe total radioactivity. The data are expressed as the differences in the percentage of total radioactivity within each compartment between unstimulated Icon{roll and stimulated lantigen or A23187)cells. activation also induces an increase in the quantiq,' of arachidonate associated with neutral lipids inside the cell (Fig. 2). Labelled products in the supernatant fluids and neutral llpids in t h e cell increased concomitantly with a decrease in labelled arachidonate primarily observed in P E and PC. In fact, the loss of arachidohate from P E accounts for 6 0 - 7 0 % of the total arachidonate found as free arachidonic acid or converted to eieosanoids during B M M C activation with antigen or ionophore. In order to get additional information on the utilization of arachidonate released from phospholipids upon cell activation, the radioactive products in the cell and in the supernatant fluids were further investigated. Analysis of the neutral lipids in the cell pellet was performed to determine the percentage of araehidonic acid released from phospholipids that had remained as free arachidonie acid or had been converted to other

lain 1" 111 IJt~fril~utiotr of tadio, J¢'Hri;j l..~(',qhehtr m..trul bpid~'

eelN labelled ;is de,~'rlhed, were incubated fi~r ~ mi,~ x~hb anligen ,,r ilm~)phore A23187. Cells were removed from the supernalanl fluids by c~ntril'tlgalion I'he lipids ~cre uxlracted flora Ihe pellet and the neutral hpl~JC[ds~ey~parated by TIC as described. The amourU of radioagliviW in each neulr;d lipid class is expressed a~ at pe role111of total. The perccnl change was determined by tile differe~lce between control and antigen or A23187 challenge value~ These dulls are train tt ~inglc experiment and are representative of three separate experiments M:~t

Label r0igraling with

Conlrnl (i tff mlal

Amig~:n '; ol unal

9~ Change from control

MG/LTB 4 5,S,'L I I I { Er'E~ DG/12.1 "~ IIETE~

a.3 ~1.3 112

I,h 1.7 ll.5

+ 1,3 + 1,4 +0.3

A23187 e~ of tolul 3,2 7.a I.II

r4 Change from conlml +~.9 + b.7 --a,8

FFA

(1.5

25

+ 2,11

0.4

11.5

+{hi

12,a 0.8

+ I 1.5

TO

+().4

197 oroduets. Table III shows the distribution of radioactivity in various neutral lipid fractions before and after cell activation. The majority of this radioactivity migrated by TLC with free fatty acid, HETEs and monoglyceride/leukotriene Ba. There was also a small but reproducible increase of radioactivity in the diradylglycerol region of the TLC plate. It was also critical to account [br the radioactivity found in supernatant fluids since this fraction contained the majority of the araehidonic acid lost from phospholipids. GC-MS analysis indicated that supernatant fluids contained PGD,. The synthesis of POD_+ increased from 1.26 to 3 . 4 f following antigen challenge(antigen) and to 4.98 pmol/lI) ~ cells following ionophore activation. In addition to PGD ,, small quantities of other cyelooxygenase products, TxB, in particular, were detected by GC-MS in supernatants of slimulated BMMC. An additional analysis of arachidonic acid metabolites including the lipoxygenase products was performed by HPLC. In this ease, we determined the radioactive products obtained from BMMC Inbetted to equilibrium and stimulated by antigen or ionc,phore. The radioactivity in supernatam fluids was primarily in association with the lipoxygenase products, LTB+ and LTC~, along with free arachidonic acid [Fig. 3). Radioactive peaks were also found migrating with prostaglan.ain D 2 and, to a lesser extent, with throm+ boxane ,~., in supernatant fluids from ionophore

TL~l~Em Fig. 3. Distribution of labelled products from B M b l C supernatural fluids upon antigen and ionophllre ~itimulaunn. BMM("labelled as described were challenged with antigen, ionophnre c~rnil slimulatoD' agcnl. Cell~ were remtwed from the supernal~mllluidsby centrifugaIk)n al 4 ° C. Producl~were separated from Ihe extractilmmixlureby reverse+phase I-|PLC i~ described. The retenlion timt:of cicusannid standards are illdiculed un tile chromatograph. The tlal~l tram Ibis experiment ;ire from one experiment ~hlch are representalive in three separate experiments.

PE

t.alkn+

Fig. 4 Distributio~t,f labelled ~lraehidon;ncin P(" and PE ~ubcla~,~cs in the BMMC nMMU were hlbelled tar 24 h with [~It]AA as de~ribctl PC" and PIE cla~e~ were ist,laled and sepJrated into I-;Icyl, I-alkvl, and (-alk-l-cnyl subclas,scs as described. The,c dzaa arc expressed a~ the pcrcent;igeof the total labelled;irachidonut¢ in each PC t~r PE ~ubcla~.These data arc ~prexentalive tff ~ix differeat experiments. A23187-stimulated BMMC, confirming the data obtained with G C / M S . The rehaively large amount of free araeludonie acid found in supernatant fluids along with that found in cell pellets of BMMC activated with A23187 suggest that a smaller proportion of lhe total arachidonic acid released from phospholipids is converted to clcosanolds under this condition as compared to antigen stimulation.

Mobdization o~ nraehidonate ~?om pho~ptzohpid snbchtsse.I Three major phospholipid subclasses I I-acyl, I-alkyl, l-aik-I-enyl) contained labelled urachidonute within ethanolamine and choline+linked pbospholipids+ Greater than (gl% of [3H]araehidonat¢ found within PC was a~soeialed with I-aeyl nnd I-alkyl-linked molecular species (Fig. 4h In contrast to PC, l-alk-I-enyl-linkcd molecular species clearly contained the bulk of labelled aracbidonate in PE. The remaining arachidonate was tkmnd in I-acyl-2-arachidonoyI-GPE with the l-alkyl species containing les,~ than 6% of the labelled arachidonate. Since PE clearly contains most of the arachidonatc in the mast cell, these results suggest thai l-alk-l-cnyl-2-arachidonoyI-GPE is the largest arachidnnate-eontaining subclass in the BMMC, Subsequent experiments were perlL~rmcd in order to dctcrntine which of the major arachidonatc-containing subclasses provided labelled arachidonate during cell activation. Fig. 5 illustrates the loss of arachidonate from the varkms PC and PE subclasses of the BMMC 5 min after stimulation with antigen and A23187. Arachidonate was lost from all the PC ,utd PE subclasses. The rank order of loss of arachidonatc from phospholipid subclasses was 1-alk-l-enyl-2-arachidonoyI-GPE> Iacyl-2-arachidonoyI-GPC _> l-acyl-2-arachidonoyl-GPE > I-alkyl-2-arachidonoyI-GPC. As indicated in Fig. 4, l-alkyl-2-arachidonoyI-GPE and 1-idk-l-enyl-2-arachi,ionoyl-GPC contained little of the labelled arachidopatc and contributed relatively small quantities of arachidomqe. There were no major differences in the rank order of release seen in antigen and ionophorc

198 PC

PE

r--ii

~1 ~1 ~

anligen A231S7

suhclass~ Fig. 5. Loss ot labelled arachidonale from Pg and PC subclasses up~m challenge with ut*tigen or ionophore A23187. BMMC were labelled for 24 h with [3H]AA as described. PC and PE class were isolaled and separalcd inlo I-acyl. I-alkyl iir I-alk-l-enyl subclasses us described. These dala tire expressed as Ihc Io~., of [3H]arachidonale [rum each PC or PE suhclass upon challenge with antigen or i(mophure A23187.These data are represenlative of five :.eparatc experiments.

stimulated cells. In general, with both stimuli, those phospholipid classes and subclasses, which contained the largest amounts of labelled arachidonate, released most of ttle arachidonate during cell activation. Discussion

In light of the intense interest in oxygenated derivatives of arachidonic acid, the acylation-deacylation of arachidonic acid into and from phospholipids has been the subject of numerous studies. Phospholipid sources of arachidonic acid have typically been identified in experiments which utilized cells prelabelled with arachidonic acid followed by celI activation. A major critcrion which must be fulfilled in these experiments is that the labclled arachidonate in each phospbolipid class represents the mole quanlitigs of endogenous araehidonatc in those same phosphnlipids. There can be major discrepancies between studies using mass or label to determine sources of arachidonate for eicosanoids when equilibrium labelling conditions are not achieved. In the present study, BMMC that were labelled for 24 h in culture with arachidonic acid distribute that arachidonic acid in phospholipids in a proportion similar to the endogenous mass of arachidonate. These initial studies provided confidence that the turnover of labelled arachidonatc in BMMC labelled for 24 h would accurately reflect the turnover of endogenous arachidonatc, Our results demonstrated that large quantities of arachidonate were lost front PE, PC and Pl during cell activation. Ethanolamine-linked glycerolipids were the major arachidonate-containing phospholipid which wcrc degraded during BMMC activation. It is important to point out that there was no preferential loss of

arachidonate from a phospholipid class with antigen or ionophore. For example, BMMC stimulated with antigen released about 15% of PE, PC and PI pools of arachidonate. Similarly, approx. 50% of the arachidohate in all three phospholipids was lost during ionophore A23187 stimulation. The major arachidonate-containing subcL~scs of ethanolamine and choline-linked glyeerolipids were also those who lost the majority of arachidonatc during cell activation. These subclasses were 1-alk-l-enyl-2-arachidonoylGPE, l-aeyl-2-arachidonoyI-GPC, l-aeyl-2arachidonoyI-GPE and 1-allcyl-2-arachidonoyl-GPC. These results differ from a number of other studies utilizing mast cells from other sources. While other studies suggest that PC and PI are the major sources of arachidonate [7,24-26], our data indicate that most of the arachidonate is released from PE and particularly from l-alk-l-enyl-2-arachidonoyl-GPP- [7,24-26]. Potential explanations for these discrepancies include the fact that different mast cell preparations are obtained from a variety of sources and, therefore, may have fundamenta; differences in their arachidonic acid metabolism. Alternatively, mast cells in the present study were labelled to a point where the labelled arachidonate mimicked the mass of araehidonate in all phospholipid classes. To confirm this hypothesis, we labelled BMMC for shorter periods of time ( 1 - 1 2 h). In these conditions, PC and PI appeared to be the major sources of arachidonate (data not shown). Therefore, it appears critical that cellular lipids are labelled to equilibrium in order to obtain label results which represent mass. The present data in the B M M C concur with findings in the human neutrophil which indicate that PE is the major phospholipid source of arachidonate [27-19]. In previous studies within this laborator~ and others, there has been some question as to whether the nature by which ionophorc stimulates cells is merely to amplify tile physiologic response of the cell or to alter the pattern of activation. For example, t,asophilic leukemia cells have been reported to release labelled araehidonate from predominantly PI during lgE stimulation and from PE after ionophore [40]. In contrast to the aforementioned study, these data suggest that in BMMC, ionophore~A23187 amplifies and does not alter the pattern of loss of arachidohate from phospholipid classes when compared to antigen. In addition, they revealed that the phospholipid which contributes the most arachidonate during call activation (i.e., PE) is the phospholipid which contains the largest pools of araehidonate. The release of araehidonate from phospholipids could arise directly via a phospholipase A , mechanism or indirectly by conversion of phospholipids to acylglycerols and phosphatidic acid via phospholipase C a n d / o r phospholipase D. In order to better understand which of the enzyme systems play an important

i9") role in this process in BMMC, the various intermediates from these reactions were isolated and quantified. During antigen and ionophore stimulation of BMMC. most of the arachidonate lost from phospholipids was found as free arachidonic acid or eicosanoid products. Small amounts of the arachidonic acid was found in diradylglyceride species. There was also a small but significant increase in the arachidonic acid found in phosphatidic acid after cell activation. More indirect evidence that the direct removal of arachidonatc from the sn-2 position is a major means by which arachidouate is mobilized was obtained from studies which indicated that a portion of PE, PC and P[ pools labelled at the sn-3 position of the molecule was converted to labelled 2-1yso compounds during cell activalion. These 2-1yso-phospholipids were then rapidly reacy~ated with a long t~ttty acyl chain at the sn-2 position of the molecule. Historically, it has been ~x*lemcly difficult to measure the formation of lysophospholipids during cell activation. A recent study by Tessner and colleagues [29] indicates that relatively large amounts of lysoPE are formed during neutrophil activation. The present data, aiong with the aforementioned study. provide clear evidence for the direct removal of araehidonate from the sn-2 position of arachidonate-comaining phospholipids during cell activation. In addition, otlr study suggests that all major arachldonate-comaining phospholipids can be acted upon, in large part. by enzymes (i.e., phospholipaso A : , transacylase) that remove the laity acid from the sn-2 position. Because of the possibility that different 2-1ysophospholipids are formed and acylated at different rates and by difl;erent mechanism, the formation of these compounds ohserved in this experiment may not reflect the actual amounts of each 2-1ysophosphulipids formed. In fact, these data show that although PE clearly provide.~ the bulk of araehidonate lost from phospholipids, similar quantities of lysoPE, lysoPC or lysoPl ate produced. This may be due to differences in the specific activity of different phosphotipids utilizing the three diflcrent radiolabels (i.e., [3H]cholioe chloride, myo-[3 Hlinusitol or [l~C]ethanolamine). Alternatively. it may suggest that lysoPE is preferentially acyhucd feint!re to lysoPC and lysoPl. In addition, it is still pos,~ible that the differences in the rates of formation of lysoPE relative to lysoP'C and lysoPI are the result of different mechanisms involved in the generation of these lymph-, nholipids. These important questions will be the basis of future wurk in this and other laboratories. In any event, the formation of 2-1yzopbosphulipids together with the release of arachidonic acid suggest that the direct removal of arachidonate from the sn-2 position of the phospholipid molecules plays a major role in the mobilization of araehidonate during mast cell activation. In addition, the increases of arachidonate in both D G and PA suggest that a smaller portion of the

arachidunlc acid lost from phospholipids is the result of phospholipasc D a n d / o r pho~pholipasc C type reactions. Several products were formed from the arachidohate released from phosphulipids during cell activation. Products (free araehidonic and eicosanoids) released into the supernatant fluids accounted for a large percentage of the loss of arachidonate-containing phosphotipids during cell activation. Several products mcludiug Icukottienc B 4. leukotricno C a, prostaglandin D, and free arachidonie acid were among the labelled components found in the supernatant fluids during cell actNatiun. In addition, arachidonic acid and compounds migrating in the H E T E region were detected within the cell after activation. A large percentage of tbe additional arachidonic acid released from BMMC phospholipids upon ionophore stimulation was not converted to products but remained as free fatty acid. This sugge,~t that, at some point after cell activation, the BMMC reach their capacity to convert arachidonic acid to elcosanoids. The reason for this phenomenon is not clear, but it may be due to the inactivation of enzymes such a~ the 5-1ipoxygenaxe during cell activation. Taken together, this study suggest that PE, and in particular, I-alk-l-enyl-2-arachidonoyI-GPE, is the major source of araehidonic acid released from BMMC during activation. Enzymes which remove arachidonate from the .sn-2 position of the phospholipid plug a major role in mobilizing that arachidonate from phospholipids. A large proportion of the araehidonate released during autigcn activation is converted to eicosanoids including [eukotricn¢ B4, Icukotricnc C 4 and prostaglandin D.,, Sludics are presently being performed to determine if araehidonatc which is liberated from individual phuspholipid molecular species is utilized unitbrmly in the 6.~rmatlon of cyclooxygenase and liooxygenase prt)duets.

Acknowledgements This work was supported by Health and Human Services grants AI Ill060, Al 24985 and AI 26771,

Referent~s I I~hiT;tk;t,T and lshlzaka, K. (1975) Pnlg Allergy It~. fil 121 2 Ishizaka,T ~md Ishiz~tka,K. (1984}pnlg. Allergy34. I~1~-235 3 Sollix~tn,T,J. Parker. K.L, Kulezyckl. A.. Jr ~md P:lrker, t.'W. t1~176)t tmmunol. 117.713-7|6. 4 Mclzgcr. tl. AIc~waz.C., Hohman. R,. KineL J-P.. Pribluda. V add QU~IlULR. IlUSb)Annu. Rev. trnmunal. 4, 4pl-417U Oli~cr, J M. Sc;lgrave. J.C., Stump. RF.. I'll:liter. J.n ~md Deanln. G.G. (l'~Sgl Prog. Allergy42, 185 245. 6 Beavcn. M.A. and Cunha-McM. J.R. (19881 Pnlg. Allergy J.2 t23 te,4, 7 Peters. S.P., MacGlubhan.D.W, Scholm;m.ES. Schleimer.RP. II:lw'~, E£'.. Rokach. I.. Adkinson. N.F. a0d Lichlenstcirl 1 M ~l*t84) .L ImmunoL 132. IU72~i'~79.

2{10 ~; Murphy. R C . Ilammarslrlim, S and Samuelsson, B (1979) Prt~c. Nal A c a d Sei USA 76. 4275-4279. 9 Mcncia-Hucrhi, J-M. Razin. E . Ringcl. E.W.. Corcy. E.L, Hoover, D,. Austen. K F and Lewis. R . A (19831 J ImmunoL 1311. 18g5 Igl)(I I11 Ri~!m. E,. M c a c l a - l l u e r l a , J-M.. Lewis. R.A Corey E.J and At:~4cn, K.F. Prec. Natl. Acad. ~Jci. U S A 79. 4665 4667. 11 Chilt(~n, F.tL lind Lichtenstcin. L M ( 199111Chem. l m m u n ~ L 49. 173 2fl5 12 BilFs. T K.. Smith. J B and Silver. M J . (1977) J. Clin. InvesL 6[h l-b. 13 Wal~h. ( . E . . DcChat¢ict. L.bL. ( h i l t o n . F,II.. Wykle. R . L and W~dl¢. M (1983} Bk~chim, Bk~phys. Acta. 7311, 32-4O. 14 GiIrci;l-flil, M, alld Siragillliitn. R,P. 11986) J, | n l r a u n o h 13@ 259-2(~3 15 Rittcnh(~llsv Simmt~n~ S (19791 J C l i n Invest. 63, 58()-5[47 16 Kenned'.. D,A. (1987) J. BioL (.'hem. 262, 153(15 16313. 17 Chilmn, F, tl.. Elba, J . M . Olson, SC. and Wykle. R . L (1984) J, Biol. Chem. 2 5 9 12(114-12111(L I~ Dimiel. L W . W l i t e , M. , n d Wykle. R L (198(d J. Bkfl. Chem. 2bL 912g 9132 19 PaL J - K , Sig¢l. M.L.. Egan. R . W and Billah. M,M. (1988)J Biol, Chcnl. 2f,3 12472 12477 211 Kennedy, D.A. tl99[])J. ]rllaluno], 144, 3912-3919, 21 Bell, R,L., Kcnnerly. D . A , Stanlilrd. N, and Majerus, P, [1979) Prt~c. Natl. Ac[id. Sci, U S A 76, 3238-3241. 22 Presc41tt. S.M. and MaJerus. P.W. (19831 ]. Biol. Chem. 258, 784 7 6 q 23 Kcnnerly, D.A.. Sullivan, ~.J.. Sylve~ter. P. and P~rkcr, C.W. (][)7qlJ Exp M e d 15(I, 1t139 Iil4J. 24 Dkano. Y., Ishizuka. Y.. Naka~him~, S., Tohmalsu, T.. Takagi, H, lind Nozawa, Y {P~85) Biochem. Biophys. Res, Commun, 127. 726 732,

25 Yamada K.. '~k:mo. Y.. Miura. K and Nl~za~a, "Y. (It~.R7} Biochim. Binphys. Acta. 917, 29~]-29f. 26 Yamada. K., Okano, Y,. M}ura, I~.. and Nozawa, Y. (I';87) B i o c h e m J 247.95 t)9. 27 Chillon, F.H. and Ctnu~elL T.R. (1988) J, Bkll, Chem. 263. ,"; 52hI~-5265. 2, Chittlm. F M (1!)891 Bk~CllCnl. J. 25tL 327 333. 29 Tessner, T.G., G r e e n D O and Wvkle R.L (199D J. Bird C h e m . 265, 21[)32-21038, 31~ Yokota. T , Lee, F.. Arai. N., Rennick. D , Z l o m i c L A.. Mosmann. F.. Miyajima. A., Takehc. ¥., Kaslelein. R.. Zurawski. C, and Arai, K. (19861 in G e n e Cloning in Lymphokine Research. (Wehb, G.. ed.k pp. 63, Academic Pres,,. N e w York, 31 Saito, H.. Okajima. F.. Molski, T.F.P, Sha'afi, R.I., Ui. M. and Ishizaka, T. (1987)J. Immunol. 138. 3927-3934. 32 SaiTo. | i . [shizaka. K, a n d Ishizaka, T. (19891 J. Immtmol. 145. 251}-258 33 Bligh. E.G. and Dyer. W.J, (lq5t-*) Can. J. Biochem. Physiol. 37, 911-917. 34 Kennerly. D,A., Sullivan. T - 1. and Parker. C.W. (1979) J. Immunol. 122. 152-159. 35 Pallon. O.M.. Fasulo. J.M. and Robins. S,J. 11982)21. Lipid Res. 23, l @ l - 19b. 36 Chilton, F.H. [1991)) Methods Eazymol. 187. 157-167. 37 Nakagawa. Y., lshima, Y. and Waku, K, 11982) Biochim. Biophys, A¢la, 7 1 2 . 6 6 7 6 7 6 38 P c l e r ~ G n l d e n , M. and Thebert, P (19871 Am. Rev. Respir, Dis. 135, 1(120 1026. 39 Hubbard. W.C.. Alley, M.C., Gray, G.N., Green, K . McLemore. T . L and Boyd, M.R. (1989) Cancer Res. 49. 826-832.