Activation by Saturated and Monounsaturated Fatty Acids of the 0

THEJOURNALOF BIOLOGICAL CHEMISTRY 0 1988 by The American Society for Biochemistry and Molecular Biology, Inc

Vol. 263,No. 27. Issue of September 25, pp. 136?0-13676,1988 Printed in U.S.A.

Activation by Saturated andMonounsaturated Fatty Acids of the 0;generating System in a Cell-free Preparation from Neutrophils* (Received for publication, April 8, 1988)

Torahiko Tanaka, Ryu Makino, Tetsutaro IizukaS, Yuzuru Ishimuraj, and ShiroKanegasakin From the Department of Biochemistry, Schoolof Medicine, Keio University, Shinjuku-ku, Tokyo I60 and the (Institute of Medical Science, the University of Tokyo, Minato-ku, Tokyo 108, Japan

Saturated and monounsaturatedfatty acids with appropriate chain lengthsuch as laurate andoleate activated an 0;-generating enzyme system in a cell-free preparation from porcine neutrophils. The activated preparation catalyzed a stoichiometric conversion of 0 2 to 0, by utilizing NADPH as the electron donor. The preparation contained both membrane and soluble fractions and, upon separation into subfractions, the 0;-generating activity resided exclusively in the membrane fraction. Polyunsaturated fatty acids including arachidonate also activated the system, but they concurrently stimulatedNADPH-independent O2 consuming reactions which yield neither 0; nor H202. The amount of such a non-0,”producing O2 consumption often reached twice as much as that of 0; production. For the activation of the 0;-generating system in the membrane, the presence of the soluble fraction was essential. However, the soluble fraction wasno longer effective when once usedfor the activation, suggesting that the effective component(s) in the fraction was consumed or translocated to themembrane during the activation. When the activated membrane was incubated with delipidated albumin, the activity was lost with concomitant decreases in the amount of membrane-associated fatty acids. The lost activity was restored by the replenishment of the fatty acid in the presence of a freshsoluble fraction. We also found that Ca2+augmented a non-0;-producing 0 2 consumption in thecell-free preparation by unsaturated fatty acids and interfered with the activation of the 02-generating system, especially that by saturated fattyacids.

&-Unsaturated fatty acids such asarachidonate’ and oleate

* This investigation was supported in part by grants from Keio University, by a grant from Takeda Science Foundation, and by grants from the Ministry of Education, Science and Culture, Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This articlemust therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $Present address: The Institute of Physical and Chemical Research, Wako-shi, Saitama, 351-01, Japan. $ To whom all correspondence should be addressed. ‘The fatty acids are expressed in this paper by their common names: caprylate, n-octanoate; caprate, n-decanoate; laurate, n-dodecanoate; myristate, n-tetradecanoate; palmitate, n-hexadecanoate; stearate, n-octadecanoate; arachidate, n-eicosanoate; palmitoleate, cis-9-hexadecenoate; oleate, cis-9-octadecenoate; linoleate, cis-9-cisar12-octadecadienoate; linolenate, 9,12,15-all-cis-octadecatrienoate; elaidate, tram-9-ocachidonate, 5,8,11,14-all-cis-eicosatetraenoate; tadecenoate.

have been known to stimulate the release of superoxide anion (0;)’ from intact neutrophils and macrophages (1-4). The stimulation was reversed by the removal of fatty acids from the cells with delipidated albumin and was restored by the replenishment of the fatty acids (2). It was also known that the stimulation was accompanied by morphological changes of the cells (2). Saturated fatty acids such as myristate and palmitate and trans-unsaturated fatty acids including linolelaidate were described as inert both in stimulating the 0; release andin provoking the morphological changes. The effects of fatty acids were thus interpreted to be due to the perturbation of plasma membrane caused by the intercalated fatty acids with the double bond(s) in the &-configuration (1,2). Recently, an NADPH-dependent 0;-generating system in cell-free preparations from neutrophils or macrophages was shown to be activated by unsaturated fatty acids (5-10) and sodium dodecyl sulfate (ll), butagain not by saturated fatty acids such as stearate and palmitate(6,9, 10). In contrast to above observations, Kakinuma and her associates (12-15) have repeatedly reported that saturated fatty acids with appropriate chainlength were also good stimuli for the 0;-generating system in intact neutrophils from various sources. Recently, we were also able to show that, besides unsaturated fatty acids, saturated fattyacids were effective in stimulating the 0 2 release in human andporcine neutrophils (16). In these papers, evidence was provided that thediscrepant results on the effects of saturated fatty acids are due at leastin part to the different Ca2’ concentrationsin the medium employed in each study. It has been known that Ca’+ interacts with fatty acids affecting their solubilities (17) or altering the physicochemical states of fatty acids which were inserted in the biological membrane (18). In the present study, we examined the activation of the 0;-generating system in a cell-free preparation from porcine neutrophils by various fatty acids with special emphasis on the stoichiometry between 0; generation and 02 consumption in the activated reaction. The stoichiometry was determined by using a heme-substituted horseradish peroxidase as a trapping reagent for both 02 and Hz02 (19). The results show that all kinds of straight-chain fatty acids with appropriate carbon numbers (Clz-C20)are capable of activating the 0 2 generating system in the cell-free preparation irrespective of theirstructures. The reported inertness of saturatedand trans-monounsaturated fatty acids in the activation of 0 2 generating system was presumably due to the way of adding them to thereaction mixture. A difference was noted among the effects of various kinds of fatty acids. The activation by The abbreviations used are: 0 2 , superoxide anion; HBSS, HEPES-buffered Hanks’ balanced salt solution; HEPES, N-8-hydroxyethylpiperazine-N‘-2-ethanesulfonate;EGTA, [ethylenebis(oxyethylenenitri1o)ltetraacetic acid.

13670

Activation of Neutrophil 08-generating System

by Fatty Acids

13671

constant stirring (19). The standard reaction mixture (0.4 ml) consaturated and monounsaturated fatty acids was specific to the 0;-generating enzyme system in the sense that they did n o t tained the cell-free preparation obtained from 4 X lo7 neutrophils, 5 mM EGTA, 0.16-0.25 mM diacetyldeuteroheme-substituted horsestimulate other 02-consuming reactions, while polyunsaturadish peroxidase and other ingredients in Ca2+-freeHBSS at pH7.3. rated fatty acids stimulated both 02-generating and non-02- For the activation of the 07-generating system, a desired amount of generating O2 consumptions. We also found that the OF- a fatty acid was added slowly to the reaction mixture with vigorous stirring a t a rate of 5-10 nmol/s, which usually took about 2 min. If generating system in the membrane fraction could recurrently be activated and deactivated in vitro by the addition and the fatty acid, especially a saturated fatty acid, was added more removal of a fatty acid, respectively. The presence of fresh rapidly, activation of the 0;-generating system could hardly be observed. After further incubation for 5 min at 25 "C, NADPH (0.1 soluble fraction was alwaysrequired for the activation,but i t mM) was added to initiate the reaction. was not necessary to maintain the activity. The significance Binding of "C-Fatty Acids to the Membrane Fraction-The cellof these results is discussed. A preliminary report pertinent free preparation (4 X lo7 cells/0.4 ml) was incubated with 2 mM 114C-labeledlaurate (0.04 mCi/mmol, Du Pont-New England Nuclear) t o a portion of this work has appeared (16). or 1.5 mM [l-'4C]oleate (0.04 mCi/mmol, Amersham Corp.) for 5 min at 25 "C. The mixture was centrifuged a t 200,000 X g for 30 min at MATERIALS AND METHODS 4 "C, and thesurface of the resulting pellet was rinsed with Ca2+-free Isolation of Neutrophils-Porcine neutrophils were isolated as de- HBSS. Radioactivities in both pellet and supernatantfractions were scribed previously (20) with minor modifications. The whole blood (1 determinedina liquid scintillationcounter (Beckman Model LS liter) anticoagulated with heparin was mixed with an equal volume 3800) using a Triton X-100-toluene-based scintillation fluid (21). Gas Chromatographic Analysis of Metabolites from Fatty Acidsof phosphate-buffered saline at pH7.4, followed bysedimentation of erythrocytes in 0.9% dextran T-500 (Pharmacia LKBBiotechnology Metabolism of laurate or oleate, if any, during the activation process Inc.) a t 1 X g, 25 "C. The granulocyte-rich supernatant was collected of the cell-free system was evaluated by gas chromatography with a and centrifuged a t 200 X g for 10 min. The pellets were suspended in Shimadzu GC-14A gas-chromatograph equipped with a fused silica 280 ml of a Ca2+-freeHEPES-buffered Hanks' balanced salt solution capillary column (Supelco Inc., SP-2330,300 cm).The cell-free prepcontaining 137 mM NaC1,5.4 mMKC1, 0.81 mM MgSO4,0.44 mM aration (4 X lo7 cells/0.4 ml) was incubated with 2.0 mM laurate or KH2P04,0.34 mM NazHP04, 5.6 mM glucose, and 25 mM HEPES- 1.5 mM oleate for 5 min at 25 'C. Then 1/19 volume of delipidated NaOH at pH 7.3, which is hereafter called as Ca2+-freeHBSS. The albumin solution (6.0 mM) in Ca2+-free HBSS was added to the suspension (10 ml) was centrifuged at 400 X g for 30 min over 12 ml mixture followed by further incubation for 10 min. The mixture was of Ficoll/Conray solution (IBL, Japan). Each precipitate was sus- cooled and centrifuged a t 400,000 X g for 5 min a t 4 "C. The lipids in pended in 30 ml of distilled water for 30 s to lyse the remaining the resultant supernatant fraction were extracted by chloroform/ erythrocytes, and the isotonicity of the medium was restored by the methanol (107, v/v) as described previously (22). Fatty acids thus addition of 3 ml of 10% NaCI. The isotonicity was further assured by extracted were methylated by diazomethane and quantitated with the addition of an appropriate amount of physiological saline. The myristic acid or capric acid as thestandard. granulocytes were obtained by centrifugation at 200 X g and washed Chemicals-Extra pure preparations of lauric acid (99.8% pure), a t least twice with Ca2+-freeHBSS. The cell preparation thus ob- oleic acid (99.9% pure), myristic acid (99.7% pure), palmitic acid tained consisted of 88-97% neutrophils, and theyield was 2-10 X loB (99.5% pure), stearic acid (99.9% pure), linoleic acid (99.3% pure) cells from 1 liter of the blood. The cell suspension (5 X lo7 cells in were the gifts of Nippon Oil & Fats Co. Ltd. (Hyogo, Japan). Other 0.5 ml of Ca2+-free HBSS) in a glass culture tube (12 X 75 mm, fatty acids were of the highest grade available from Sigma (98-100% Corning Glass Works) was frozen in liquid N2 and stored at -70 "C. pure).These fatty acids were dissolved in absolute ethanol and After the storage for 2 months, loss of activity was less than 10% as neutralized with a minimal amount of 1N NaOH except for arachidic judged by the fattyacid-stimulated 0;generation. acid. Arachidic acid was used without neutralization. The concentraPreparation of Cell-free System, Its Fractionation and Reconstitu- tions of ethanol in the stock solution were adjusted to 50 or 70% (v/ tion-To obtain acell-free preparation, frozen suspensions of porcine v). Usually 10-12.5 p1 of these fatty acid solutions were added to 0.4 neutrophils (5 X IO7cells/0.5 ml) were quickly thawed in awater bath ml of the reaction mixture slowly as described already. Ethanol up to at 39 "C with vigorous shaking. The cells were found to be disrupted the final concentration of 4% (v/v) in the reaction mixture had no by this procedure and showed no 0;-generating activity upon stimu- detectable effect on the 0; generation activity. Horseradish peroxilation by phorbol 12-myristate 13-acetate. In most experiments, the dase was purchased from Toyobo Co. Ltd. (Osaka). Diacetyldeuterofreeze-thawed preparation was further disrupted by sonic oscillation heme-substituted horseradish peroxidase was prepared as described (for conditions see below) and was used as the cell-free preparation. previously (23). Delipidated albumin (bovine, prepared from fraction To obtain unstimulated membrane and soluble fractions, the freeze- V) was from Sigma. BW-755C was the gift of Wellcome Research thawed preparation before sonic oscillation was fractionated into Laboratories (England). Other chemicals were of the highest grade precipitate and supernatant by centrifugation at 6,000 X g for 5 s commercially available. with a Beckman Microfuge B. The precipitate, which contained 8090% of the total alkaline phosphatase activity and less than 3% of RESULTS the total lactate dehydrogenase activity, was used as the membrane fraction. The soluble fraction was obtained by centrifuging further Stimulation of 0;-generating System in a Cell-free Prepathe supernatant fraction at 400,000 x g for 5 min at 4 "C by a micro- ration by Fatty Acids-Fig. 1 shows effectsof various concenultracentrifuge (Hitachi CP100H). In reconstitution studies, the 6,000 trations of fatty acids with t h e chain length from Clo t o C18 X g precipitate was used in combination with the 400,000 X g supernatant. When membrane and soluble fractions were prepared on NADPH-dependent 0; generation in a cell-free preparafrom apreviously sonicated preparation (see text), they were obtained tion from porcine neutrophils. As seen, saturated fatty acids by 400,000 X g centrifugation of the preparation. After mixing, the exerted their maximal effects at concentrations over 2 mM, preparations were dispersed evenly by sonic oscillation at 0 "C for 30 whileunsaturatedfattyacids had t h e maximal effect at times with 0.2-s pulses using a microprobe of the Branson sonifier at around 1.5 mM and thereafter became inhibitory. In these 20% of the full power. experiments, fatty acids were added slowly under vigorous Determination of 0; and H202 Generation and O2Consumptwnstirring to the reaction mixture which contained the cell-free The generation of both 0;and H202were determined spectrophotometrically by using diacetyldeuteroheme-substitutedhorseradish per- preparation derived from 4 X lo7 cells (approximately 2 m g oxidase as the trapping reagent (19). The heme-substituted horse- as protein) in 0.4 ml of Ca2+-free HBSS. The use of such a radish peroxidase reacts stoichiometrically with 0;and H202to form high concentration of the cell-free preparation was to obtain compounds I11 and 11, respectively, which are stable enough for our an accurate stoichiometry between O2 consumption and 0; measurements and are spectrally distinguishable from each other generation. Under comparable conditions,the concentration (19). 0 2 consumption was measured with a Clark-type oxygen electrode (YSI Inc., 5331 probe) with a high sensitivity membrane (YSI of arachidonate which induced the maximal 0; generation Inc., 5776 membrane). 0;and H202 generation and 02 consumption was also around 1.5 mM. Saturated fatty acids with more were measured simultaneously in a single cuvette which was set in a longer or shorter chain length such as arachidate (Czo) and Hitachi 557 dual wavelength-double beam spectrophotometer under caprylate (C,) were not effective at similar concentrations.

Activation of Neutrophil 0;-generating System

13672

3.0

1

.

1.0

by Fatty Acids

g

, , I I I

I a

I

J05 generation

Y

0 C

;.

3.0

I 4

C

i

al

rn 2.0

I

0

Il.4

0 10

0 0

1.0 2.0 Fatty Acid ( m M )

3.0

FIG. 1. 0;-generating activity induced by different concentrations of fatty acids in the cell-free system. The reaction mixture (0.4 ml) contained the cell-free preparation (approximately 2 mg as protein obtained from 4 X 10' neutrophils), 5 mM EGTA, 0.16-0.25 mM diacetyldeuteroheme-substituted horseradish peroxidase in Ca2'-free HBSS, pH 7.3, at 25 "C. After incubation of the mixture with an indicated concentration of fatty acid for 5 min, the reaction was initiated by the addition of 0.1 mM NADPH except for linoleate and linolenate. In thelatter cases, NADPH was added immediately after the addition of fatty acid. Initialrates of 0; generation were determined by measuring the formation of compound I11 of diacetyldeuteroheme-substituted horseradish peroxidase and plotted against the fatty acid concentrations. Panel A , with saturated fatty acids; caprate (4),laurate (O), myristate (O),palmitate (W), stearate (V),and arachidate (a).Panel B, with unsaturated fatty linoleate (A),linolenacids; palmitoleate (O), oleate (A), elaidate (O), ate (V).

By employing laurate and oleate as the representatives of saturatedandmonounsaturatedfatty acids, we compared characteristics of the 0; generating reaction. The K , values for NADPH in the laurate (2.0 mM)- and oleate (1.5 mM)induced 0; generation were 73.4 t 4.9 and 66.9 t 5.6 p ~ respectively, and the Vmaxvalues for the 0; generation were 0.32 0.02 and 0.26 +- 0.01 mM/min, respectively (mean +S.E. in three experiments). An approximate K,,, for NADPH in the arachidonate (1.5 mM)-induced 0; generation was 50 PM, although a large amount of O2 consumption unrelated to 0; production (see below) obscured the determination of the accurate value. These V,,, values were comparable to that induced by an optimal concentration of phorbol 12-myristate 13-acetate in intact porcine neutrophils, whichwas 20-30 nmol/min/107 cells at 25 "C. There was a low level of NADPH-consuming activity (6.8 +. 0.4 pM/min in five experiments) in the cell-free preparation in the absence of fatty acid. No significant 0; generation was observed by the addition of 0.1 mM NADH in both laurate- andoleate-stimulated preparation. The pH optimum was between 6.8 and 7.3 for both laurate- and oleate-stimulated activities. Stoichiometry between O2 Consumption and 0; Production Induced by Various Fatty Acids-Fig. 2 shows a simultaneous record of oxygen consumption and 0; generation by the cellfree preparation which was activated with 2.5 mM laurate. During the preincubation with laurate for about 5 min, the preparation showed neither O2 consumption nor 0; forma-

*

1.o

Time ( min

2.0

I

FIG. 2. Rates of 0;generation and 0 2 consumption during the stimulation of the cell-freesystem by laurate. The reaction mixture and other conditions were the same as described under Fig. 1.The concentration of diacetyldeuteroheme-substituted horseradish peroxidase was 0.20 mM. Laurate (2.5 mM) was added a t arrow a, and, after 5min of incubation, NADPH (0.1 mM) was added at arrow b. 02 generation was assayed by using diacetyldeuteroheme-substituted horseradish peroxidase, and O2 consumption was determined polarographically (see "Materials and Methods").

tion. Upon addition of 0.1 mM NADPH, it immediately began to consume 0 2 and produced 0 2 , where the initial rates of 0 2 consumption and 0; generation were both 0.21 mM/min. Formation of Hz02was not detected throughout the reaction. When the reaction was carried out in the absence of laurate, both 0 2 consumption and 0;production were negligible in the presence of the same concentration of NADPH. The reaction practically ceased at 2 min afterthe addition of NADPH,and the total amounts of 0 2 consumed and 0; generated during the reaction were 57.6 and 60.0 nmol, respectively. From these results, it is evident that the cell-free preparation activated by laurate converts molecular O2 stoichiometrically to 0;. Table I summarizes the initial rates of O2consumption and 0; generation upon stimulation by various fatty acids. When the cell-free preparation was stimulated by saturated, and cis,and trans-monounsaturated fatty acids, the ratios of the two reaction rates were all approximately 1:l. The ratios of total O2 consumption to total 0; production were also about 1:l. On the other hand, the ratios of 0; production to O2 consumption were as low as 0.27,0.53, and 0.49, when the preparation was stimulated with cis-polyunsaturated fatty acids such as arachidonate, linolenate, and linoleate, respectively. Furthermore, a massive O2 consumption was observed with these cis-polyunsaturated fatty acids, when they were added to thecell-free preparation in the absence of NADPH (data notshown). During such NADPH-independent 0 2 consumption, no formation of 0; was detected, and the amount of O2 consumed was roughly equal to that of the non-0;producing O2 consumption observed in the presence of NADPH. Thus both saturated and unsaturated fatty acids are capable of activating the 0;-generating system, but the latter fatty acids also induced a large amount of non-02producing O2 consumption. The NADPH-independent 0 2 consumption induced by 0.25 mM linoleate was inhibited by nordihydroxyguaiaretic acid and BW755C lipoxygenase inhibitors, with apparent Ki values of 8 and 9 pM, respectively. Porcine neutrophils have been known to contain a significant amount of 12-lipoxygenase which metabolizes cis-polyunsat-

13673

Activation of Neutrophil 0;-generating Systemby Fatty Acids TABLE I Initial rates of 0; generation and 0 2 consumption in the cell-freepreparation activated by various fatty acids ExDerimental conditions were the same as those in Fie. 1. ~

~~

~~

Rate of reactions

Fatty Acid (chain length) and concentration employed

Ratio 01/02

0; generation

O2consumption pM/min X I O - 2

mllP

Saturated fattyacid