CYTOKINE- AND CALCIUM IONOPHORE A23187-MEDIATED ARACHIDONIC ACID METABOLISM IN NEUTROPHILS Thomas R. Ulich,13*

Katie Busser,’

Kenneth

J. Longmuir*‘*

Arachidonic acid (AA) metabolism is implicated as an intracellular and/or intercellular second messenger system for the transmission of cytokine-initiated signals that affect neutrophils and mediate systemic toxicity. The purpose of the present study is to ascertain if cytokines that are known to affect neutrophil function in vivo and in vitro directly stimulate neutrophil AA metabolism in vitro. The recombinant human cytokines multi-colony stimulating factor, granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin 1, tumor necrosis factor (TNF), and interleukin 6 and the calcium ionophore A23187 were incubated with purified 14C-AA radiolabeled human peripheral blood neutrophils and the effects were assayed by oneand two-dimensional thin layer lipid chromatography. None of the cytokines appeared to induce the release of cell-incorporated AA or to increase the level of radiolabeled phosphatidic acid. TNF induces severe systemic toxicity that is inhibited by cyclooxygenase inhibitors, which suggests a role for AA metabolites in the pathophysiologic effects of TNF; we have confirmed that TNF and endotoxin act synergistically to induce indomethacin-inhibitable fatal shock in rats. However, when in 3H-AA radiolabeled human neutrophils were incubated with TNF in kinetic, cold-chase, and TNF preincubation experiments, TNF was not found to increase AA metabolism, although changes in the intracellular neutral lipid content were noted. GM-CSF, which has been reported by previous investigators to directly induce the release of AA, did not release neutrophil-associated 3H-AA. In conclusion, the direct release of AA from membraneassociated phospholipids does not appear to be a major second messenger pathway for cytokine-initiated activation of neutrophils. The therapeutic effects of indomethacin in TNF and endotoxin-associated fatal shock would not appear to be related to any inhibitory effects on cytokine-mediated AA metabolism in neutrophils. o 1990 by W.B. Saunders Company.

The number of circulating neutrophils in peripheral blood is tightly regulated under physiologic conditions. When neutropenia or neutrophilia do occur, they are clinically important signs of infection or of inflammatory disease. The number of circulating neutrophils under such pathophysiologic conditions is controlled by the rate of myelopoiesis and neutrophil release from the marrow, by the relative proportion of marginated and freely circulating neutrophils, and by the intravascular half-life of the neutrophils. The processes of cellular proliferation, differentiation, and activation that deter-

Departments of Pathology’ and Physiology: University at Irvine School of Medicine, Irvine, CA 92717. *To whom reprint requests should be addressed. o 1990 by W.B. Saunders Company. 1043-4666/90/0204-0004$05.00/O KEY WORDS: Neutrophils

280

Cytokines/Arachidonic

acid/Calcium

of California

ionophore/

mine the number of circulating neutrophils are mediated in large part by a family of endogenous mediators known as cytokines. Among the cytokines that affect circulating neutrophil numbers are the classical colonystimulating factors multi-colony stimulating factor (multi-CSF),’ granulocyte-macrophagecolony-stimulating factor (GM-CSF),* and granulocyte colony-stimulating factor (G-CSF),3 the proinflammatory cytokines interleukin 1 (IL 1)4 and tumor necrosis factor (TNF),4 and interleukin 6 (IL 6),5 a molecule that appears to be identical to a previously described myeloid blood cell differentiation-inducing protein and which supports granulocytic differentiation of hematopoietic progenitor cells.‘j Cytokines have been reported to prime or activate neutrophils in various ways and to promote neutrophil spreading.‘-” Arachidonic acid (AA) metabolites also appear to be involved in the regulation of the number of circulating neutrophils and in the activation of neutrophils during acute inflammation. Prostaglandins and leukoCYTOKINE,

Vol. 2, No. 4 (July),

1990: pp 280-286

Arachidonic

TABLE 1. Extracellular human neutrophils Cytokine

release of radiolabeled

Carrier

A23187

(4)

(4)

aracbidonic

acid metabolites

Multi-CSF

GM-CSF

(2)

(2)

acid metabolism

in neutrophils

/ 281

by cytokine-treated

“C-arachidonic

TNF

IL 1

IL 6

(2)

(2)

G-CSF

acid-labeled

(number of experiments)

cpm released* *cpm f standard

4,386 deviation,

+ 2,139

+ 1,338

17,163

per mL of culture

3,214

i 9

4,302

RESULTS Recombinant human cytokines incubated for 30 min with r4C-AA-prelabeled human neutrophils did not induce any significant release of radiolabeled AA metab-

2.

+ 1,019

3,298

zi 626

3,991

f 1,395

5 174

4,299

k 1,082

3,573

medium.

trienes themselves induce experimental neutropenia efand/or neutrophilia. 13,14The neutrophilia-inducing fects of some cytokines are inhibited by inhibitors of AA metabolism,” suggesting that AA metabolites may act as second messengers. Cytokines have been reported to induce directly the release of AA metabolites from neutrophils,‘5-16 macrophages,“-I8 and from non-hematologic cells.1gS21 Cytokines induce systemic toxicity that is clinically responsive to inhibitors of AA metabolism. The morbidity associated with the administration of high doses of TNF has especially been reported to respond favorably to pharmacologic inhibitors of AA metabolism.22 Although many lines of evidence thus support the hypothesis that AA metabolites play a role as second messengers for the in vivo effects of cytokines, in vitro experiments employing individual cytokines and purified cell populations are necessary to determine whether cytokines directly initiate AA metabolism within a given cell type. The purpose of the present study is to investigate the interaction of recombinant human cytokines with AA-radiolabeled human peripheral blood neutrophils.

TABLE

(2)

(3)

Lipid profile of cytokine-treated

“C-arachidonic

olites into the culture supernatant as compared to carrier controls (Table 1). The calcium ionophore A23187 was used as a positive control (Table 1) and induced a significant (p < 0.05) release of radiolabeled AA metabolites. One-dimensional thin layer chromatography (TLC) was performed on all culture supernatants of all experiments and the results rule out the possibility that the radiolabel in the supernatant of cytokinetreated neutrophils might be differently distributed than in the supernatant of carrier controls. The intracellular phospholipid, free fatty acid, and neutral lipid content of the cytokine-treated neutrophils was determined by two-dimensional TLC (Table 2). None of the cytokines induced significant release of free fatty acid as compared to carrier (negative) controls. Calcium ionophore A23 187 (positive control) induced a 19-fold increase in intracellular free fatty acid and a large increase in phosphatidic acid as well as decreases in neutrophil phospholipid content. In A23 187-treated neutrophils the phospholipids whose radiolabeled content decreased were phosphatidylcholine and phosphatidylinositol. TNF and endotoxin are each known to induce fatal shock at sufficient doses, endotoxin is known to induce the release of TNF in vivo, and TNF and endotoxin are known to act synergistically to induce fatal shock in mice.23 TNF- and endotoxin-induced shock are inhibited by indomethacin and other cyclooxygenase inhibi-

acid-labeled

human neutrophils

PMNs prelabeled with 14C-AA were treated with various cytokines for 30 min. 14C-labeled lipids were extracted from the cell pellet and resolved by two-dimensional thin-layer chromatography (tic). Values are the amount of radioactivity in each lipid class, expressed as a percentage of the total radioactivity recovered from the tic plate, mean f standard error of independent experiments conducted with different preparations of PMNs. Lipid Treatment (number of exoeriments)

Carrier

(5)

A23187(5)

IL 1 (2) IL 3 (2) IL 6 (2) TNF (2) G-CSF (3) GM-CSF (2)

Phosphatidic acid 0.07 0.47 0.06 0.11 0.14 0.02 0.07 0.08

t 0.01 i 0.10 f 0.04 + 0.03 Y? 0.02

f 0.01 zk 2

0.03 0.04

Phosphatidylserine 0.26 0.32 0.23 0.22 0.21 0.53 0.22 0.21

+ 0.03 f 0.03 f 0.10 + 0.04

+ * f i

0.00 0.17 0.02 0.04

Phosphatidylinositol 3.26 1.87 2.84 3.00 2.79 1.63 2.70 3.62

k 0.14 * 0.30

L 1.10 f f f f +

1.02 0.41 0.46 0.49 0.53

Phosphatidylethanolamine and plasmalogens 2.61 2.16 3.02 2.07 1.95

f 0.28 + 0.21 + 0.60 + 0.35 * 0.90

1.66 & 0.29 2.03 2.31

i i

0.55 0.22

Phosphatidylcholine 5.65 + 0.69 1.42 + 0.18 6.06 t 1.45 4.75 + 1.41 4.62 t 0.19 4.88 i 0.93 5.62 i 0.60 4.98 f 1.17

Free fatty acid 0.35 5.84 0.23 0.18 0.89 0.26 0.19 0.17

+ 0.10 k 1.32 + 0.10 + 0.06 + 0.53 i

0.00

t 0.04 f 0.05

Neutral lipid 87.79 87.91 87.56 89.68 89.40 91.02 89.17 88.64

+ + + + k + + +

1.03 1.67 3.19 2.91 0.85 0.95 1.03 2.05

282

/

Ulich.

Busser,

24

t:. .

CYTOKINE,

Longmuir

:: ..

TNF- and endotoxin-induced shock motivated us to investigate further the possibility that TNF directly stimulates AA metabolism in neutrophils. In a kinetic study the effect of TNF on 3H-AA prelabeled neutrophils was examined after 1,2 and 4 hr of incubation with TNF. TNF did not cause increased release of radiolabeled AA metabolites into the supernatant compared to controls after 1, 2, or 4 hr of incubation (data not shown). The intracellular phospholipid and neutral lipid profile (Table 3) showed that TNF caused a slight increase in AA content of the neutral lipid pool and a slight decrease in AA content of the phospholipid pool, an effect that suggests inhibition of triacylglycerol lipase and is reminiscent of the known inhibitory effect of TNF on lipoprotein lipase activity. The diacylglycerol: triacylglycerol ratio within the neutral lipid pool of TNF-treated neutrophils was approximately the same (a significant elevation of the ratio might indicate phospholipase C activation). The possibility that the radiolabeled AA from the phospholipid pool of neutrophils was being released and then rapidly reesterified was investigated by a coldchase study in which an excess of unlabeled AA was added to 3H-AA-prelabeled neutrophils at the same time as TNF or carrier was added. The cold-chase study (Table 4) did not reveal any difference in the phospholipid profile of TNF- versus carrier-treated neutrophils at 1,2 or 4 hr, suggesting that reesterification of AA did not account for the apparent lack of any direct effect of TNF on AA release. The effect of TNF on the initial uptake and esterification of AA within neutrophils was also examined by preincubating neutrophils for 30 min with TNF, then adding 3H-AA to the incubation medium. Phospholipid profiles were determined by one-dimensional thin layer chromatography in triplicate at various timepoints. TNF did not significantly affect the uptake and incorporation of 3H-AA into phospholipids but did cause a significant increase in the incorporation of 3H-AA within

..

1

: i i

LPS

TNF

LPS + TNF

LPSfTNF fSaline

Vol. 2, No. 4 (July 1990: 280-286)

LPS+TNF + lndomethacin

Figure 1. Lipopolysaccharide (LPS) (100 pg) and TNF (10 pg) act synergistically to cause fatal shock that is inhibited by pretreatment of rats with indomethacin. Each dot represents one rat.

tots. The cell type(s) on which indomethacin acts to inhibit shock is unknown, but the neutrophil might a priori be considered a potentially very important cellular site of interaction between endotoxin, cytokines, AA metabolism, and cyclooxygenase inhibitors. In order to confirm in our laboratory the previously reported inhibitory effect of indomethacin on shock, a model of synergistic TNF- and endotoxin-induced fatal shock was developed in the Lewis rat. TNF (10 bg) in combination with endotoxin (100 pg) was fatal to 100% of rats within less than 8 hr whereas either TNF or endotoxin alone was never fatal (Fig. 1). Indomethacin, an inhibitor of the cyclooxygenase pathway and of phospholipase A, in rabbit neutrophils, completely abrogated the lethal effect of the combination of TNF and endotoxin (Fig. 1). The striking inhibitory effect of indomethacin on

TABLE 3. Kinetic study of intracellular phospholipid and neutral lipid profile of TNF-treated 3H-arachidonic acid radiolabeled human neutrophils PMNs prelabeled with ‘H-AA were incubated with TNF for 1, 2, and 4 hr at 37°C following removal of radiochemical label. ‘H-labeled lipids were extracted from the cell pellet and resolved by one-dimensional TLC. Values are mean + standard error of 3 TLC determinations of a single experiment. Diacylglycerol and triacylglycerol were resolved using a separate tic solvent system. Values are the ratio of the radioactivity in the diacylglycerol and triacylglycerol fractions. Carrier Lipid* PI (%)

PC (%) PE (%) NL (%) DG/TG *PI, phosphatidylinositol;

1 hr

25.7 k 2.4 18.3 + 2.8 8.6 f 1.4 47.4 k 1.8 5.7 PC, phosphatidylcholine;

TNF

2hr

27.3 z~4.8 18.4 f 1.0 9.6 i 0.5 44.8 i 4.2 5.7

4 hr

26.1 2 18.1 + 9.1 * 46.2 + 4.9

PE, phosphatidylethanolamine;

2.9 1.7 1.0 1.0

1 hr

2 hr

4 hr

23.2 + 3.2 14.9 * 1.1 7.0 f 0.5 54.9 f 2.4 6.0

22.0 i 2.8 14.4 * 1.0 8.2 f 0.3 55.3 & 2.0 7.0

24.5 t 3.5 14.3 f 1.3 8.3 f 1.2 53.0 k 2.4 N.D.

NL, neutral lipids; DG, diacylglycerol;

TG, triacylglycerol.

Arachidonic

acid metabolism

in neutrophils

/ 283

TABLE 4. Cold-chase study of intracellular phospholipid and neutral lipid profile of TNF-treated 3H-arachidonic acid radiolabeled human neutrophils PMNs prelabeled with ‘H-AA were incubated with TNF arachidonate/BSA complex [4:1 mol/mol]) for 1, 2, or 4 Individual phospholipids and the neutral lipid fraction from Values are mean i standard error of 3 TLC determinations

plus unlabeled arachidonic acid (0.1 pmol/ml hr at 37OC following removal of radiochemical the cell pellet were resolved by one-dimensional of a single experiment.

Carrier Lipid* PI PC PE NL

TNF

0 hr

1 hr

2 hr

18.2 & 1.3 15.6 * 1.6

14.4 + 2.3 12.9 t 0.7

16.9 * 2.4

14.9 + 0.5

16.3 f 1.6

15.6 + 1.7

14.5 t

11.9 + 0.5

10.1 i 0.8

11.7 * 0.2

10.0 * 0.7

5.4 + 0.8 60.9 f 1.1

5.7 i 0.3 67.1 f 1.9

5.9 2 0.4 65.4 f 2.1

4.7 i 0.4 70.3 i 1.0

5.4 f 0.2 66.7 i 1.8

11.1 + 0.4 6.1 r 0.1

*PI, phosphatidylinositol;

PC, phosphatidylcholine;

4 hr

PE, phosphatidylethanolamine;

the neutral lipid pool (Fig. 2), the latter observation again suggesting that TNF might inhibit triacylglycerol lipase. GM-CSF has previously been reported by some investigators to induce the release of small amounts of radiolabeled AA from neutrophils. Under the experimental conditions described, GM-CSF did not cause the release of radiolabeled AA (six experiments) as compared to carrier-treated neutrophils (six experiments), whereas A23 187 (two experiments) released large amounts of AA (Fig. 3).

DISCUSSION The purpose of this study was to determine whether cytokines directly stimulate the metabolism of AA

30 20

t

as an label. TLC.

I

Figure 2. Uptake and incorporation of rH-AA in the presence and absence of TNF was determined by one-dimensional TLC. Cells were harvested at 15 min, 30 min, 1 hr, 2 hr, and 3 hr following the addition of ‘H-AA. Values are mean f SE of three TLC determinations of a single experiment. Open symbols, carrier. Closed symbols, TNF.

NL, neutral

2 hr

1 hr

lipids; DG, diacylglycerol;

67.2

i: 1.3

4 hr 1.2

6.2 t 0.3 69.3 + 0.9

TG, triacylglycerol.

esterified to phospholipids. All of the cytokines studied are known to induce peripheral neutrophilia and in most cases have been shown to prime or activate mature neutrophils in various in vitro assays. Despite the close association between neutrophil activation and AA metabolism and despite some previous reports to the contrary, none of the cytokines were found to release arachidonic acid within the detection limits of our experimental system. The relationship to AA metabolism is perhaps more strongly established for TNF than for any other cytokine. TNF has been reported to stimulate directly AA release from synovial cells, macrophages, and a number of transformed mesenchymal and epithelial cell lines.20 TNF-related morbidity in patients and mortality in experimental animal models is inhibited by inhibitors of AA metabolism.22 In the present study, we confirmed in rats the previous report of Rothstein and Schreiber23 that TNF and endotoxin in mice synergistically induce fatal shock. We also confirmed the previous report of Kettelhut et a1.22 that indomethacin prevents TNFrelated fatal shock. Indomethacin inhibits the cyclooxygenase pathway of AA metabolism and has also been reported to inhibit phospholipase A2 in rabbit neutrophils,24 implicating AA metabolites in the pathogenesis of fatal shock. On the other hand, indomethacin also is known to act as a calcium antagonist25 and the possibility cannot be overlooked that the therapeutic effect of indomethacin may be related to this later property. Dexamethasone, which is known to inhibit phospholipase A, via the induction of an endogenous inhibitor called macrocortin,26 was not as therapeutically effective as indomethacin (data not shown). The temporal relationship between TNF and AA metabolite expression during endotoxemia is unclear since enhanced phospholipase A, activity in rat plasma appears within three min after injection2’ whereas TNF serum levels do not increase until 30 to 60 min2* The most signifieant therapeutic effect of indomethacin on endotoxemia might, therefore, occur independently of and before any

284

/ Ulich,

Busser,

Longmuir

Time after addition of GM-CSF to neutrophils

3. GM-CSF (solid lower line, six experiments) does not cause neutrophils to release ‘H-AA metabolites into the culture medium as compared to carrier-treated neutrophils (dashed line, six experiments). A23 187 (solid upper line, two experiments) causes a substantial release of jH-AA metabolites.

Figure

TNF-related activation of AA metabolism, although the same could not be said in the case of experimental administration of TNF. TNF did not cause the release of AA from neutrophils, which is consistent with the findings of Berkow and Dodson” and of Roubin et al.* Dipersio et a1.,16on the other hand, reported that TNF induced a slight but significant release of 3H-AA from human neutrophils in data expressed as a percent increase after 1 hr. It is noteworthy that Roubin et al.* observed a greater release of AA from neutrophils primed with TNF and subsequently stimulated with A23 187 than from neutrophils stimulated with A23187 alone. Berkow and Dodson” did not observe any such TNF-induced priming effect when they used fMLP as the final stimulating agent. In the present study, attempts to detect a direct effect of TNF on AA metabolism in kinetic, cold-chase, and TNF preincubation experiments were negative, although an increase in the intracellular neutral lipid content was noted in repeated experiments, suggesting an inhibitory effect of TNF on triacylglycerol lipase remminiscent of the known inhibitory effect of TNF on lipoprotein lipase. Laudanna et a1.2g recently reported that TNF activates the respiratory burst of human neutrophils in the absence of the formation of 3Hinositol phosphates, 32P-phosphatidic acid, or 3Harachidonic acid. GM-CSF has been reported to induce directly the

CYTOKINE,

Vol. 2, No. 4 (July 1990: 280-286)

release of AA from neutrophils and these reports prompted us to reexamine the effects of GM-CSF. Under the conditions employed in our laboratory, GMCSF did not cause the release of radiolabeled AA from neutrophils when compared to controls in a careful kinetic study. Sullivan et a1.15 studied the effects of G-CSF and GM-CSF on signal transduction pathways in human granulocytes and reported that both cytokines caused prompt release of AA from plasma membrane phospholipids. However, the statistical significance compared to controls of the relatively slight average release of 3H-AA was not given. Dispersio, et al.“j reported that GM-CSF, like TNF, primes neutrophils for a subsequently greater release of arachidonate. They also reported that GM-CSF directly induced a two- to fivefold increase in neutrophil leukotriene B, (LTB,) synthesis as measured by radioimmunoassay.16 In conclusion, the present study suggests that several recombinant human cytokines that are known to affect neutrophil function in vitro and in vivo do not mediate their effects on neutrophils via the release of neutrophil-derived esterified AA as a major transmembrane signaling pathway. Indomethacin is, however, an effective inhibitor of TNF-associated shock, suggesting that AA from a cellular source other than neutrophils may play an important role in the pathogenesis of septic shock. Of interest is that IL 8, a cytokine not investigated in our study, has recently been reported by Schroeder3’ to stimulate human neutrophil arachidonate5-lipoxygenase, but not to stimulate the release of cellular arachidonate. Cytokines may therefore contribute to the synthesis of inflammatory arachidonate metabolites by the upregulation of enzymes other than phospholipase A, with the arachidonate coming from an exogenous rather than a cell membrane-derived pool.

MATERIALS AND METHODS Isolation of Human Peripheral Neutrophils and Radiolabeling Acid

Blood with Arachidonic

Neutrophils were isolated from the venous blood of healthy adults. Whole heparinized blood obtained from the

antecubital vein was mixed 1:l with hetastarch (Hespan, DuPont Pharmaceuticals,

Wilmington,

DE) in 50-mL conical-

tipped tubes and allowed to sediment for 30 min. The erythrocyte-poor supernatants were layered over a Ficoll-Hypaque (Sigma, St. Louis, MO) gradient and centrifuged at 1,500 rpm at room temperature for 15 min. The polymorphonuclear leukocyte (PMN)-enriched cell pellet was depleted of any remaining erythrocytes by hypotonic lysis with sterile water for 1 min after which the cells were resuspended in Dulbecco’s phosphate-buffered saline (PBS) (Irvine Scientific, Santa

Ana, CA). The purity and viability of the PMN preparations were always greater than 95% as judged by modified Wright’s staining (Diff Quick, American Scientific Products, McGaw Park, IL) and by Trypan blue dye exclusion.

Arachidonic acid metabolism in neutrophils / 285

Radiolabeling of Neutrophils with Archidonic Acid

Extraction and Thin Layer Chromatography of Lipids from Neutrophils

Bovine serum albumin/arachidonic acid complexes were prepared by first mixing ethanolic solutions of [ l-‘4C]arachidonic acid (52 mCi/mmol) or [5,6,8,9,11,12,14,15‘Hlarachidonic acid (230 mCi/mmol, New England Nuclear Corporation), with excess sodium bicarbonate in sterile water. After evaporation of solvent with N, gas, the arachidonate was mixed with a sterile solution of fatty acid-free bovine serum albumin in PBS or Hepes-buffered minimal essential medium (HMEM) buffer. Neutrophils were labeled with 14C- or 3H-arachidonic acid by resuspending a pellet of 50 to 100 x lo7 PMNs in two mL PBS or HMEM with 0.5 mM choline, ethanolamine, inosital, and serine and containing the BSA/arachidonate complex. Final fatty acid/BSA concentrations were 1 &i 14C-arachidonate or 2.5 &i ‘H-arachidonate per lo7 cells, and 1 mg serum albumin per lo7 cells. After 1 hr at 37OC in a shaking water bath, free radiolabel was removed by washing the cells three times in buffer.

Cytokine-treated or control PMN pellets were resuspended in 1 mL D-PBS and then added to 4.5 mL of 1:2 chloroform/methanol with 100 ~1 6N HCl for subsequent extraction by the Bligh-Dyer procedure.32 Briefly, 1.5 mL of 2M KC1 and 1.5 ml of chloroform were added, centrifuged for 5 min at 1,500 rpm, and the upper phase discarded. After two washes with Folch upper phase (chloroform/methanol/H,O, 3:48:47), the lower phase was evaporated with nitrogen and the residue redissolved in 95:5 chloroform:methanol. One-dimensional TLC was performed on 20 x 20 cm silica gel-60 plates (E. Merck) using 10 to 20 kg of phosphatidylcholine, phosphatidylinositol, phosphatidylserine, phosphatidylethanolamine, and free fatty acid as standards. The plates were spotted with lipid residue in 95:5 chloroform: methanol and run in an acidic solvent system of 50:20:10:10:5 chloroform/acetone/methanol/acetic acid/water. Diacylglycerol and triacylglycerol were resolved by a separate TLC analysis using a solvent system of hexane/diethyl ether/acetic acid, 4O:lO:l. Two-dimensional TLC was performed on 10 x 10 cm silica gel-G plates (Analtech) according to the procedure of Yavin-Zutra.33 Briefly, the first dimension was run in a solvent system of 130:60: 15 chloroform/methanol/methylamine and the second dimension in sequential solvent systems of 95:5 diethyl ether/acetic acid and 100:40:20:30:10 chloroform/ acetone/methanol/acetic acid/water. The plates were exposed to X-ray film and developed after 3 d.

Cytokines Recombinant human cytokines were generous gifts from the following sources: IL 3, G-CSF, GM-CSF and IL 6, Dr. Lawrence Souza of Amgen in Thousand Oaks, CA; IL lot, Drs. Peter LoMedico and Tim Andersen of Hoffman La Roche in Nutley, N.J.; and TNF-o(, Dr. Michael Shepard of Genentech in South San Francisco, CA. Cytokines in the following amounts were added to 10’ neutrophils in 2 mL of culture medium at 37%: IL 3, 10 pg; GM-CSF, 25 pg; G-CSF, 100 yg; IL 6,25 pg; IL la, 2 x lo5 U; TNFa, lo5 U.

Extraction and Thin Layer Chromatography of Lipids from Culture Medium Incubation mixtures (2 mL containing lo7 cytokinetreated or control PMNs) were centrifuged at 200 x g for 5 min. The supernatant was removed and the archidonic acid metabolites extracted using modifications of the procedure of Salmon and Flower.3’ The supernatant was mixed with 4 mL of acetone and centrifuged at 800 x g for 10 min at 2°C. This supernatant was transferred to a clean tube and extracted twice with 2 mL of hexane. The acetone/water mixture was acidified with 60 ~1 of 1 M citric acid and the lipids extracted twice into 5 mL of ethyl acetate/acetone (4:l). The ethyl acetate/acetone mixture was washed twice with water and the solvents evaporated with N, gas. The residue was redissolved in chloroform/methanol (95:5) for subsequent scintillation counting or thin-layer chromatography analysis. One-dimensional TLC was performed on 20 x 20 cm silica gel 60 plates using 5 to 10 pg of 5-hydroxyeicosatetraenoic acid (HETE), 12-HETE, LTB,, prostaglandin E,, and prostaglandin F,, as standards. The plates were spotted with the culture medium lipid residue in 95:5 chloroform/methanol and run in a solvent system with the upper phase of 110:50:20: 100 ethyl acetateliso-octane/acetic acid/water. After drying for 30 min, the plates were placed in X-ray film cassettes and exposed for 1 week.

Acknowledgments Supported by NIH award ROl-AI2655 1 and by the Long Beach Memorial-UC Irvine Cancer Foundation (TRU) and by ACS #IN- 166 and University of California Cancer Research Coordinating Committee (KJL).

REFERENCES 1. Metcalf D, Begley CG, Johnson GR, Nicola NA, Lopez AF, Williamson DJ (1986) Effects of purified bacterially synthesized murine multi-CSF (IL-3) on hematopoiesis in normal adult mice. Blood 68:46-57. 2. Donahue RE, Wang EA, Stone DK, Kamen R, Wong GG, Sehgal PK, Nathan DG, Clark SC (1986) Stimulation of hematopoiesis in primates by continuous infusion of recombinant human GMCSF. Nature 321:872-875. 3. Ulich TR, de1 Castillo J, Souza L (1988) Kinetics and mechanisms of recombinant human granulocyte-colony stimulating factor induced neutrophilia. Am J Path01 133:630-638. 4. Ulich TR, de1 Castillo J, Keys M, Granger GA, Ni RX (1987) Kinetics and mechanisms of recombinant human interleukin-1 and tumor necrosis factor a-induced changes in circulating numbers of neutrophils and lymphocytes. J Immunol 139:3406-3415. 5. Ulich TR, de1 Castillo J, Guo K (1989) In vivo hematologic effects of interleukin-6 on hematopoiesis and circulating numbers of red and white blood cells. Blood 73:108-l 10. 6. Caracciolo D, Clark SC, Rovera G (1989) Human interleukin-6 supports granulocytic differentiation of hematopoietic progenitor cells and acts synergistically with GM-CSF. Blood 73:666-670. 7. Sullivan GW, Carper HT, Sullivan JA, Murata T, Mandel GL

(1989)

Both

recombinant

IL-10

and purified

human

monocyte

286

/ Ulich,

Busser,

IL-1 prime human promote neutrophil

CYTOKINE,

Longmuir

neutrophils spreading.

for increased oxidative J Leuk Biol45:389-395.

activity

and

8. Roubin R, Elsas PP, Fiers W, Dessein AJ (1987) Recombinant human tumor necrosis factor enhances leukotriene biosynthesis in neutrophils and eosinophils stimulated with Ca’+ ionophore A23187. Clin Exp Immunol70:484-490. 9. Shalaby MR, Palladino MA, Hirabayashi SE, Eesalu TE, Lewis GD, Shepard HM, Aggarwal BB (1987) Receptor binding and activation of polymorphonuclear neutrophils by tumor necrosis factor alpha. J Leuk Biol41:196-204. 10. Berkow RL, involved in the priming Biol44:345-352.

Dodson MR of neutrophils

(1988) Biochemical by tumor necrosis

mechanisms factor. J Leuk

21. Last-Barney K, Homon CA, Faanes RB, (1988) Synergistic and overlapping activities of tumor tor-a and IL-l. J Immunol 141:527-530.

Merluzzi necrosis

VJ fac-

22. Kettelhut IC, Fiers W, Goldberg AL (1987) The toxic effects of tumor necrosis factor in vivo and their prevention by cyclooxygenase inhibitors. Proc Nat1 Acad Sci USA 84:4273-4277.

25. Northover BJ (1977) Gen Pharmacol8:293-296.

13. Ulich TR, Dakay EB, Williams J, Ni RX (1986) In vivo induction of neutrophilia, lymphopenia and diminution of neutrophil adhesion by stable analogs of prostaglandins E,, E, and F,,. Am J Pathol 124:53-58. 14. Ulich TR, Keys M, Ni RX, de1 Castillo J, Dakay E (1988) The contributions of adrenal hormones, hemodynamic factors and the endotoxin-related stress reaction to stable prostaglandin analoginduced peripheral lymphopenia and neutrophilia. J Leuk Biol 43:510. 15. Sullivan R, Griffin JD, Simons ER, Schafer AI, Meshulam T, Fredette JP, Maas AK, Gadenne AS, Leavitt JL, Melnick DA (1987) Effects of recombinant human granulocyte and macrophage colony-stimulating factors on signal transduction pathways in human granulocytes. J Immunol 139:3422-3430. 16. Dipersio JF, Billing P, Williams R, Gasson JC (1988) Human granulocyte-macrophage colony stimulating factor and other cytokines prime human neutrophils for enhanced arachidonic acid release and leukotrine B4 synthesis. J Immunol 140:4315-4322. Tumor necrosis is regulated by

19. Godfrey RW, Johnson WJ, Hoffstein ST (1987) Recombinant tumor necrosis factor and interleukin-I both stimulate human synovial cell arachidonate release and phospholipid metabolism. Biochem Biophys Res Comm 142:235-241.

20. Elias JA, Gustilo K, Baeder W, Freundlich B (1987) Synergistic stimulation of fibroblast prostaglandin production by recombinant interleukin-1 and tumor necrosis factor. J Immunol 138:3812-3816.

24. Kaplan L, Weiss J, Elsbach P (1978) Low concentrations indomethacin inhibit phospholipase A, of rabbit polymorphonuclear leukocytes. Proc Nat1 Acad Sci USA 75:2955-2958.

12. Tsujimoto M, Yokota S, Vilcek J, Weissmann G (1986) Tumor necrosis factor provokes superoxide anion generation from neutrophils. Biochem Biophys Res Comm 137:1094-l 100.

18. Kunkel SL, Spengler M, May MK, Spengler R, Larnck Zemick D (1988) Prostaglandin E, regulates macrophage-derived tumor necrosis factor gene expression. J Biol Chem 263:5380-5384.

1990: 280-286)

23. Rothstein JL, Schreiber H (1988) Synergy between tumor necrosis factor and bacterial products causes hemorrhagic necrosis and lethal shock in normal mice. Proc Nat1 Acad Sci USA 85:607-611.

11. Atkinson YH, Marasco WA, Lopez AF, Vadas MA (1988) Recombinant human tumor necrosis factor-a. Regulation of N-FMLP receptor affinity and function on human neutrophils. J Clin Invest 81:759-765.

17. Lehman V, Benninghoff B, Droge W (1988) factor induced activation of peritoneal macrophages prostaglandins E, and CAMP. J Immunol 141:587-591.

Vol. 2, No. 4 (July

J,

Indomethacin-A

calcium

of

antagonist.

26. Vadas P, Pruzanski W (1986) Role of secretory phospholipase A, in the pathobiology of disease. Lab Invest 55:391-404. 27. Shakir KMM, O’Brian JT, Gartner SL (1985) Enhanced phospholipase A, activity in rat plasma, liver and intestinal mucosa following endotoxin treatment: A possible explanation for the protective effect of indomethacin in endotoxic shock. Metabolism 34:176187. 28. Waage A (1987) Production and clearance of tumor necrosis factor in rats exposed to endotoxin and dexamethasone. Clin Immunol Immunopathol45:348-355. 29. Laudanna C, Miron S, Berton G, Rossi F (1990) Tumor necrosis factor-a/cachectin activates O,-generating system of human neutrophils independently of the hydrolysis of phosphoinositides and the release of arachidonic acid. Biochem Biophys Res Comm 166:308315. 30. Schroeder JM (1989) The monocyte derived neutrophil activating peptide (NAP/IL-S) stimulates human neutrophil arachidonate-5-lipoxygenase, but not the release of cellular arachidonate. J Exp Med 170:847-863. 31. Salmon chromatography 86477-493.

JA, Flower of arachidonic

32. Bligh EG, Dyer extraction and purification.

J (1982) Extraction and acid metabolites. Methods

thin-layer Enzymol

WJ (1959) A rapid method of total Can J Biochem Physiol37:911-917.

lipid

33. Yavin E, Zutra A (1977) Separation and analysis of 32P-labeled phospholipids by a simple and rapid thin-layer chromatographic procedure and its application to cultured neutroblastoma cells. Anal Biochem 80:430-437.

Cytokine- and calcium ionophore A23187-mediated arachidonic acid metabolism in neutrophils.

Arachidonic acid (AA) metabolism is implicated as an intracellular and/or intercellular second messenger system for the transmission of cytokine-initi...
803KB Sizes 0 Downloads 0 Views