Agents Actions 38 (1993)

0065-4299/93/040202-10 $1.50+0.20/0 9 1993 Birkh/iuser Verlag, Basel

Differential inhibition of human secretory and cytosolic phospholipase A2 * F. Mfirki 1'*, W. Breitenstein 1, E. Beriger 2, R. Bernasconi 1, G. Caravatti 1, J. E. Francis 3, R. Paioni a, H. U. Wehrli ~ and R. Wiederkehr ~ 1Research Department, Pharmaceuticals Division and 2Agro Division, Ciba-Geigy Limited, Basel, Switzerland, and 3Research Department, Pharmaceuticals Division, Ciba-Geigy Corporation, Summit, New Jersey, USA

Abstract

The roles and relative contributions of secretory and cytosolic phospholipases A2 in physiology and pathology are not precisely known. In a search for differential inhibitors of these enzymes, which could serve as tools to clarify this issue, we evaluated the potencies of reference compounds and three series of new compounds, viz. substrate analogues, 1,2-amino alcohols and enolized/~-tricarbonyl derivatives, as inhibitors of secretory phospholipase A 2 from human polymorphonuclear leukocytes (sPLA2) and of cytosolic phospholipase Az from human U937 cells (cPLA2). With few exceptions, the compounds selected are potent inhibitors of sPLA2 with ICso values (concentration inhibiting 50%) in the low micromolar range. Inhibition of cPLA2 was only observed with some phosphate-free substrate analogues, with 1,2-amino alcohols and two of seven reference compounds, These results suggest that inhibition of secretory and of cytosolic phospholipases A2 are independent effects. Several inhibitors could be identified with a marked selectivity for sPLA/.

Introduction

Phospholipases A2 (PLA2s) are lipolytic enzymes of widespread occurrence that have recently been subdivided into two main classes, the secretory and the cytosolic PLA2s [1]. Mammalian secretory

Abbreviations: PLA2, phospholipase A2; cPLA2, cytosolic phos-

pholipase A2 from human U937 cells; sPLA2, secretory phospholipase A2 from human polymorphonuclear leukocytes; PMN, polymorphonuclear leukocyte; 1C50, concentration inhibiting 50%. * This work has been presented in part at the Symposium "Phospholipase A2, Basic and Clinical Aspects in Inflammatory Diseases" Reisensburg Castle (FRG), November 18-20, 1992. * Address for correspondence: F. Mfirki, Schweissbergweg 5, CH-4102 Binningen, Switzerland.

PLA2s of group II [2] have a molecular weight of 14 kDa and are activated and secreted as soluble proteins from a variety of physiologically stimulated cells such as platelets [3, 4], neutrophils [5], renal mesangial cells [6] or smooth muscle ,cells [7]. These enzymes hydrolyze the ester bond at the sn-2 position of phospholipids without preference for a particular fatty acid [8] and require millimolar concentrations of Ca 2 + for optimal activity [9]. Secretory PLA2s are unusually resistant to acidic pH [10-12], but are readily inactivated by thiol compounds reducing their essential disulfide bonds [1, 13, 14]. The abundance of secretory PLA2 in inflammatory exudates, e.g. in synovial fluid of patients with rheumatoid arthritis [15] and the prominent pro-inflammatory activity following

203

Agents Actions 38 (1993)

injection in an animal model [16], suggest an involvement in the pathogenesis of inflammatory diseases [15, 17]. Intracellular PLA2s occurring in the cytosol have molecular weights in the range 60-110kDa [18-21] and differ from secretory PLA2s in several

respects. They preferentially hydrolyze phospholipids with arachidonic acid in the sn-2 position [19, 20], they are activated by (sub-)micromolar concentrations of Ca 2§ [I, 19-21], and they are resistant to reducing conditions due to the presence of free thiol groups [1, 20, 21], but they are rather

ioi [-O(CH2)lsCH3 CH3(CH2)15-P-O--I 0 6H LOPO(CH2)2NH3 O 9

/

II

Oii I-(CH2)2CH3 CH3(CH2)lo-CNHA 0II

+

LOPO(CH2)20H

OH

1

~ H

2

C00c2H5 COOH 2)IOCH3

Cl'~~

(CH2)?.~COOC2H5

3

4

OH

ONH(CH2)15CH3

i

so S

H2NH( H 2 ) 3 C H - - ~ -F

H 5

\1 HO" _'N--g. (CH2)~ "~0 0

7

I

N(CH3) 2

17

Figure 1 Chemical structures of published reference compounds, 1 7, and compound 17. For references on synthesis and/or biological properties see Materials and Methods and Table la.

204 labile towards acidic pH (Reference [12] and F. Mfirki, unpublished observation). Finally, the amino acid sequences of secretory and cytosolic PLA2 display no homology [20, 22]. The roles and relative contributions of secretory and cytosolic PLA2 in the release of arachidonic acid, i.e. the rate-limiting precursor for eicosanoid mediator generation [23], in normal physiology and under pathological conditions are not precisely known. The intracellular localization of cytosolic PLA2, its sensitivity to (sub-)micromolar concentrations of Ca/+, the preference for release of arachidonic acid versus other long-chain fatty acids, and the presence of a potential phosphorylation site [20] make cytosolic PLA2 a reasonable candidate for a role in cellular signal transmission (for current views on cytosolic PLAz as a receptor-activated and G-protein-transduced effector system that mediates the release of arachidonic acid, and on the role of arachidonic acid and its metabolites as intracellular second messengers, refer to the brief reviews by Axelrod [24] and Burgoyne et al. [25]). On the other hand, many pieces of evidence support a function of secretory PLA2 in the initiation and propagation of pathological processes such as local and systemic inflammation [26]. To delineate further the respective roles of secretory and cytosolic PLA2, selective inhibitors of these enzymes might serve as valuable tools. However, no such compounds have been described to date. Therefore, we analyzed seven known reference compounds and three series of newly synthesized PLAz inhibitors for possible selectivity by determining their potencies to inhibit a human secretory and a human cytosolic PLA2.

Materials and methods

Materials

Materials and reagents used were purchased from the following sources: U937 cells (human monocyte-like histiocytic lymphoma, code CRL 1593) from American Type Culture Collection, Rockville, MD; Dulbecco's modified Eagle's medium (DMEM/F12) from Northumbria Biologicals Ltd., Cramsington, Scotland; fetal calf serum (FCS) from Gibco Life Technologies, Basel, Switzerland; bovine serum albumin (BSA), essentially fatty-acidfree, from Sigma, St. Louis, MO, silicic acid for column chromatography, 100-200 mesh (Bio-Sil A)

Agents Actions38 (1993) from Bio-Rad, Anaheim, CA; dioleoylglycerol and 1-stearoyl-2-arachidonyl-sn-glycero-3-phosphocholine from Avanti Polar Lipids, Birmingham, AL and 1-stearoyl-2-[ 1-14C]arachidonyl-sn-glycero-3phosphocholine, 58 mCi/mmol, from Amersham International, Amersham UK Test compounds

Reference compounds 1-6 (Fig. l) were synthesized in our laboratories according to published procedures as follows: l: [27]; 2: [28]; 3: European Patent Appl. 0142145; 4: US Patent 4,788,304; 5: Intern. Patent WO 88/06885; 6: [31]. Manoalide ( S K F # 190093, compound 7) was obtained from Allergan Inc., Irvine, CA. Syntheses and analytical data of compounds 8-23, 32 37 and 40-42 are described in an Appendix which may be ordered free of charge from the authors. For compound 24, refer to Ger. Often. 2,404,328 and US Patent 3.906.110, for 38 and 39 to Ger. Often. 2,700,876. Data of 25-31 will be published elsewhere (G. Caravatti et al., manuscript in preparation). Methods U937 cells. Cells were cultured as described by Kramer et al. [21] with the following modifications: The medium contained 50 gg/ml gentamycin (Gibco) and 0.02% Synperonic PE/F68 (ICI) instead of tobramycin (Lilly) and Pluronic F68 (BASF), respectively. Suspension cultures in tissue culture flasks were transferred to Spinner flasks when a cell density of 5x 105/ml had been attained. Cells (2 x 109) were harvested by centrifugation (10 min, 300 rpm, IEC Centra-7R centrifuge), washed by suspension in balanced salt solution, and suspended in 15 ml ice-cold lysis buffer (25 mmol/l Tris/HC1 pH 7.4, 140 mmol/1 NaC1, 5 mmol/1 KC1, 2 mmol/1 EDTA, 1 mmol/1 phenylmethanesulfonyl fluoride and 100 p.mol/1 leupeptin). Enzyme preparations, cPLA2 was prepared from

U937 cells (2 x 109/15 ml lysis buffer) by nitrogen cavitation, high-speed centrifugation of the lysate, and filtration of the supernatant through a sterile 0.22 tam filter as described previously [21]. This enzyme preparation had a specific activity of 0.16 nmol min -1 mg -1.

Agents Actions 38 (1993)

205

sPLA2 was prepared from human P M N s by acid extraction and dialysis as described by M/irki and Franson [11]. The specific activity of the enzyme was 0.29 nmol m i n i m g - 1.

Assay of PLA2 activity. The activity of cPLA2 was assayed according to the method of Kramer et al. [21], except that the separation of the released [1-14C]arachidonic acid was performed by the procedure used for [1-14C]oleic acid [11]. sPLA2 was assayed as described by Mfirki and F r a n s o n [11], using autoclaved [1-t4C]oleate-labelled Escherichia coli as substrate.

Inhibition of PLA2 activity. Inhibitors were dissolved in D M S O , diluted with water to a concentration of 10% D M S O (v/v) immediately before use and added to the assay mixtures at a final D M S O concentration of 1%. Percent inhibition relative to solvent-containing controls was determined at various concentrations of the test compounds. ICso values were determined graphically. Protein was determined by the Pierce Coomassie protein assay kit using bovine serum albumin as a standard.

10

E a

-

y

"6 E

>- 6 I-m I--

o ,r

4,

04

.< .-I D_

'0

=

i

l

I

20

40

60

80

100

E N Z Y M E CONC. [jug PROTEIN]

Figure 2 Dependence of PLA2 activity on enzyme concentration: ( i ) secretory PLA2 from human PMNs; (A) cytosolic PLA2 from human U937 cells. Values given are means of triplicate determinations.

Results

Enzyme kinetics. The dependence of the activity of the two PLAzs used in this study on enzyme concentration is shown in Fig. 2. A linear increase of enzyme activity was observed with both enzymes up to about 3 pmol/min. Subsequent experiments with inhibitors were performed at enzyme concentrations within this linear range. Figure 3 shows the dependence of enzyme activities on (bulk) concentration of substrates. The substrate concentrations used for the inhibition assays, i.e. 5 I-tM for sPLA2 and 2 I.tM for cPLA2, were selected from the subsaturation range. Inhibition by reference compounds. The inhibition of sPLA2 and cPLA2 by the reference compounds 1 7, which were selected from previously described inhibitors of PLA2, is shown in Table la. While all seven compounds were found to inhibit sPLA2 with ICso values in the low micromolar range, only two compounds, i.e. the alkylamine 5 (WY-48,489) and the dehydroabietylamine derivative 6, also inhibited the cytosolic enzyme. The other five compounds were either inactive or marginally active

0.35

0.3

E d E '5 E r

0.25

0.2

>j.,_

>

0.15

I--

o,<

0.1

(.,q .< ,_1 Q.

0.05

0

2

4

6

8

10

S U B S T R A T E [pM]

Figure 3 Substrate dependence of PLA2 activity. For symbols see Fig. 2.

Values given are means of triplicate determinations.

206

Agents Actions 38 (1993)

Table 1 Inhibition of secretory PLA2 from human PMNs (sPLA2) and of cytosolic PLA2 from human U937 cells (cPLA2). ICso values (laM) are means ,+ SEM of n experiments (a) Reference compounds"

Compound

sPLA2

Phosphonate-containing phospholipid analogue [27] Acylamino phospholipid analogue [28] Arachidonic acid analogue [29] Glycidic ester; WY-49,422 [30] Alkylamine; WY-48,489 [30] Dehydroabietylamine derivative [31] Manoalide [32 34]

(b) Phosphate-free substrate analogues Compound

8 9 10 ll 12 13 14 15 16 17

R

n-C3H7 n-CI4H29 n-C14H29 n-C14H29 n-C~4H29 C6H5 C6H 5 n-CI4H29 n-C14H29 (structure see Fig.

X

F F F H F F F F F 1)

cPLA2

n

IC5o

n

ICso

3 3 3 5 3 3 3

0.3_+ 0.1 1.5 _+0. ! 4.3-+ 0.2 1.3 _+0.3 5.5-t-0.1 15.0+_0.6 3.2+0.5

2 3 3 2 4 4 4

i.a.b 10 > 30 i.a. 30 i.a. 30 15.0_+0.5 10.5_+2.3 i.a. 30

R-CX 2 Y CH2CH2-Z CONH(CHz)3N(CH3) 2 Y

CO CO CHOH CHOH CHOOCCH3 CHOH CHOOCCH 3 CHOH CHOOCCH3

Z

sPLA2

CH20 CH20 CH20 CH20 CHzO CH20 CH20 --

cPLA2

n

ICso

n

IC5o

2 3 3 3 4 2 2 3 3 3

> 100 3.2+0.8 3.2_+0.7 4.7_+0.1 2.2+0.3 i.a.30 i.a.30 2.0+0.1 4.5_+0.7 2A -+0.3

2 4 3 3 3 2 2 3 3 3

i.a. 100 7.7-+0.6 15.2,+0.6 14.7+0.9 7.1 • i.a.30 i.a.30 9.0-+ 1.0 9.7-+0.6 4.7_+0.6

0-

(c) Phosphate-containing substrate analogues

I LI

R--O--P--O--R'

o Compound

R

R'

sPLA2 n

18

n-C16H33NHCOCH (NHSO2CH3)CH2

19

n-C j,,HzgCF2CO(CH 2)3

20

n-C~8HsTNHCOO-~

cPLA 2 ICso

n

ICso

6.0-+0.5

>30

-CH2CH2N + C S

5.0-+0.6

26.3•

-CH zCHzN + ~

0.6,+0.1

>30

CH2CH2N + ~ / / S

207

Agents Actions 38 (1993) Table 1 Continued (d) 1,2-Amino alcohols (open chain)

R - f H O H - C H R ' NHR"

Compound

R'

R

sPLA2

R"

cPLA 2

n

IC5o

n

ICso

21

1-Adamantyl---@CH2-

H

H

3

14.0•

3

>30

22

l-Adamantyl--@CH2

H

CHzC6H5

3

I8.7•

3

31.0•

23

l-Adamantyl@--OCH2

H

i-C3H

3

14.8•

3

16.3•

CH3

CH2CH2CH(C6Hs)2

5

12.5•

4

13.9•

24

@ C H

2 0 ~

~"

0

ci /

(e) 1,2-Amino alcohols (hydroxyalkyl piper•

I

Compound

x

25 26 27 28 29 30 31

/CH2NH 2 92HCI X

CH3(CHz)~-CHOHCH2N~ 2

~

\C~,H s

sPLA2

1 5 7 9 11 I R ( - ) 11 IS(+) 15

cPLA2

n

IC5o

n

1Cso

4 2 5 5 3 3 4

i.a. 100 i.a. 100 60• 10 5.0• 4.7+0.2 4.3 • 4.9•

3 3 3 3 5 4 5

i.a. 100 i.a. 100 105 • 8.5• 8.3• 1.1 8.4• 5.6_+0.6

NH

/ \ (CHz)x

\

(f) 1,2-Amino alcohols (cyclic)

CH

/, HO

Compound

x

y

R

(CH2)),

/

CH

2\ NHR

sPLA 2 n

32 33 34 35 36 37

2 2 2 2 1 l

1 1 1 1 2 1

n-CI2H25 n-C,sH3~ n-C,sH~7 (CH2CH20)3CsH 17 n-C,sH3~ n-C~sH~t

cPLA 2 1C50 5.1• 1.5• 4.7• i.a.50 4,2• 4.3•

n

IC5o 19.3• 8.1• 10.5• i.a.30 9.9• 14.8•

Agents Actions38 (1993)

208 Table 1 Continued OH (g) Enolized /3-tricarbonyl compounds

Compound

38 39 40 41 42

n

4 5 5 5 6

X

CI C1 CI Br Br

Y

H H Cl H H

sPLA2

Z

CI C1 H Br Br

cPLAz

n

ICso

n

IC50

3 3 3 3 3

5.6• 2.8• 21.7• 0.9• 0.6•

3 3 3 3 3

>30 >30 >30 >30 i.a. 30

Chemical structures are shown in Fig. 1. b i.a., inactive at the concentration indicated (highest concentration tested). on cPLA2 at the highest concentrations tested. Manoalide (7), a nonsteroidal sesterterpenoid from a marine sponge, described as an irreversible inhibitor of several secretory PLAzs [32, 33], inhibited sPLA2, but not cPLA2.

Inhibition by substrate analoyues. A series of new phosphate-free substrate analogues containing a medium-to-long-chain alkyl residue (9-12, 15, 16) markedly inhibited sPLA2 and, with only slightly reduced potency, cPLAz (Table lb). However, close analogues with a short-chain alkyl or an aryl substituent (8, 13, 14) were inactive on both enzymes. A closely related substrate analogue, 17, with the cyclic structure shown in Fig. 1, inhibited sPLA2 and cPLAz with potencies comparable to the open-chain analogues. The phoshate-containing substrate analogues 18-20 with a mediumto-long-chain alkyl substituent are highly potent inhibitors of sPLA2 with weak or marginal effects on cPLA2 (Table lc).

in the hydroxyalkyl piper• series showed a gradual decrease of inhibitory potency for both sPLA2 and cPLAz as the chain length was reduced to less than 12 carbon atoms (x<9; Table le). The stereochemistry at the chiral carbon atom (position 1) in this series has no influence on the inhibitory activity; the two stereoisomers 29 and 30 of the compound with x = 11 had comparable potencies.

Inhibition by enolized fl-tricarbonyl

compounds.

Highly potent inhibitors of sPLA2 without significant activity on cPLA2 were obtained in the series of enolized fl-tricarbonyl compounds (Table lg). Seemingly minor changes of the chemical structure, e,g. 39 and 40, had a dramatic effect on the biological activity, suggesting that the phenyl ring with its halogen substituents may occupy a critical area of the active site of sPLA2. A structure-activity relationship of about 30 additional compounds of this series will be presented elsewhere (manuscript in preparation).

Inhibition by 1,2-amino alcohols. Three subtypes of 1,2-amino alcohol derivatives were found to inhibit sPLA2 as well as cPLA2: the open-chain compounds 22-24 (Table ld), the hydroxyalkyl piper• 27-31 (Table le), and the cyclic compounds 32-34, 36, 37 (Table lf). The most potent inhibitors were found among those containing a medium-to-long-chain alkyl substituent (28-34, 36, 37). It was interesting to note that compound 35, whose alkyl chain of 14 carbon atoms is interrupted with three oxygens, was inactive on both enzymes. A systematic variation of the chain length

Discussion Selective inhibitors of sPLA2 and cPLA2 would provide a possible means of clarifying the roles of these enzymes in physiology and pathology at different levels of complexity, from studies with intact cells in vitro to animal models in vivo. It seems that little attention has been devoted to this approach so far, since specific information on the selectivity of PLA2 inhibitors for the secretory and cytosolic enzyme is lacking. Although many inhibitors of

Agents Actions38 (1993) (secretory) PLA2 (of, as yet, unspecified selectivity) are described in the literature (for recent reviews see [35, 36] ), we are aware of only four compounds reported to inhibit a cytosolic PLA2: nordihydroguaiaretic acid [37], quercetin [37], manoalide [33, 38] and a synthetic substrate analogue [38]. These compounds are, however, not selective for cytosolic PLA2; besides other enzymes they also inhibit secretory PLA2, some with rather high potency [29, 33, 34, 38, 39]. In order to contribute to the scarce knowledge on differential inhibition of PLA2s, and with the aim to identify PLA2 inhibitors with selectivity for either the secretory or the cytosolic enzyme, we systematically determined the inhibitory potencies for sPLA2 and cPLA2 of several reference compounds and a large series of new PLA2 inhibitors that we had recently synthesized. Analysis of the results indicates that about twothirds of the compounds reported, including six of seven previously published inhibitors chosen as reference compounds, are highly active (IC5o _<10 taM) on sPLA2. Several of our new compounds, viz. 12, 15, 17, 20, 33, 39, 41 and 42, have ICs0 values around or below 2 jaM and rank among the most potent inhibitors of a human secretory. PLA 2 known. By contrast, comparable potent inhibition of cPLA2 was less frequently observed. This is perhaps not surprising since the compounds were designed and optimized as inhibitors of secretory PLA 2. None of the seven reference compounds and less than one-third of the new compounds had a potency below l0 jaM for the cytosolic enzyme. In addition, the distribution of inhibitors of cPLA2 among the different chemical classes did not parallel that of sPLA2. For instance, while all phosphate-free substrate analogues active on sPLA2 also inhibited cPLA2, none of the inhibitors of sPLA 2 among enolized fl-tricarbonyl compounds was active on cPLA 2 (Tables lb and lg, respectively). Consequently, the potency ratio sPLA2/cPLA 2 of the inhibitors varied over a wide range, from >50 for compounds 20 and 42 to about 1 for compounds 6, 23, 24 and 31. Thus, the former compounds display a remarkably high degree of selectivity for sPLA2, whereas the latter lack selectivity. The wide variation of the potency ratio of different compounds shows that there is no general correlation between inhibition of sPLA2 and cPLA2, suggesting that the two inhibitory activities are independent of each other. Secretory and cytosolic PLA2s are different proteins without amino acid sequence homology and

209 have distinct enzymatic characteristics, e.g. acyl specificity and C a 2 + sensitivity (see Introduction). Therefore, it is no surprise that these enzymes were found to respond in different ways - quantitatively or qualitatively - to some of the inhibitors tested. On the other hand, these different enzymes catalyze the same chemical reaction, i.e. the hydrolysis of the fatty acyl ester bond at the sn-2 position of phosphoglycerides. One may, therefore, presume that the identical biochemical function of the enzymes would depend on a common or similar structure of their active sites. So far, X-ray structure analysis of several secretory PLAzs has revealed that the functionally important amino acid residues of snake venom, bovine pancreatic and recombinant human synovial fluid PLA2 are virtually superimposable, indicating that secretory PLAzs from different sources have a common active site [40-42]. The potent inhibition of secretory PLA2 from human PMNs by all seven reference compounds, and, especially, by 1 and 2, provides additional support for this concept. Compounds 1 and 2 are substrate analogues that were rationally designed and optimized as competitive inhibitors of secretory PLA2 from cobra venom and porcine pancreas, respectively [27, 28]. When assayed with these enzymes and synthetic phospholipid substrates dispersed in detergent micelles, they are able to compete with the substrate at a mole fraction as low as 0.001 (1) or even 0.0002 (2), indicating extremely high binding affinity and suggestive of highly specific interaction with these enzymes [27, 28]. As we observed, both compounds also inhibit sPLA2, a secretory PLA2 from still another source (human PMNs) and assayed with the more "natural" substrate E. coli (Table la). Furthermore, the effect of 2 on sPLA2 is stereospecific, as previously reported for pancreatic PLA2 [28]; inhibition occurs exclusively with the (R)-stereoisomer, whereas the (S)enantiomer is totally inactive ([28]; our data not shown). These results are interpreted as further evidence for the similarity of the active sites of diverse secretory PLA2s [42]. As no structural analysis comparable to that of secretory PLAzs has yet been published for the active site of cytosolic PLAzs, it remains a challenge to ascertain to what extent secretory and cytosolic PLAzs utilize a common active site. Our observation that many potent inhibitors of sPLA2, including the reference compounds 1 and 2 which have been successfully tailored for (stereo-)specific

210

interaction with the well-defined active site of secretory PLA2, did not inhibit cPLA2 would suggest that the structure of the active sites of secretory and cytosolic PLA2s, and, presumably, the reaction mechanism of the enzymes; will not be identical. In conclusion, in this study we analyzed a variety of new inhibitors of PLA2 and some reference compounds for differential inhibition of secretory and cytosolic PLA2. We observed that the potency ratio for the two enzymes ranged from > 50 in favor of sPLA2 to about 1. The obvious lack of a correlation between the two inhibitory activities indicates that they are independent effects and suggests that the active sites of sPLA2 and cPLA2 are presumably not identical. While compounds selective for cytosolic PLA2 could not be identified so far, several potent inhibitors with high selectivity for secretory PLA2 are now available for further studies.

Acknowledgements We thank Silvio Roggo for the synthesis of the reference compound 2 and its enantiomer, Walter Baumgartner, Beatrice Bohrer, Kurt B6rgi, Hansj6rg Haas, Peter Sch~iublin and Roland Wermuth for their synthetic work, Val6rie Hanulak for technical assistance with enzyme preparation and assays, Kathrin Wagner for advice and help with the culture of U937 cells, and Jonathan Green for critical reading of the manuscript.

Agents Actions 38 (1993)

[7] J. Pfeilschifter, W. Pignat, F. Marki and I, Wiesenberg,

[8] [9] [I0]

[11]

[12]

[13]

[14]

[15]

Received 17 November 1992; accepted by I. Ahnfelt-R~nne 20 January 1993

[16]

References

[17]

[1] J. D. Clark, N. Milona and J. L. Knopf, Purification of a 1lO-kilodalton cytosolic phospholipase A 2from the human monocytic cell line U937. Proc. Natl. Acad. Sci. USA 87, 7708-7712 (1990). [2] R. L. Heinriksou, E. T. Krueger and P. S. Keim, Amino acid sequence of phospholipase A2-~ from the venom of Crotalus adamanteus. A new classification of phospholipases A 2 based upon structural determinants. J. Biol. Chem. 252, 4913~4921 (1977). [3] K. Horigome, M. Hayakawa, K. lnoue and S. Nojima, Purification and characterization of phospholipase A2 released from rat platelets. J. Biochem. 101,625 631 (1987). [4] M. Murakami, 1. Kudo and K. Inoue, In vivo release and clearance of rat platelet phospholipase A 2. Biochim. Biophys. Acta 1005, 270-276 (1989). [5] C. Lanni and E. L. Becker, Release of phospholipase A2 activity from rabbit peritoneal neutrophils by f-Met-LeuPhe. Am. J. Pathol. 113, 90-94 (1983). [6] J. Pfeilschifter, W. Pignat, K. Vosbeck and F. M~irki, Interleukin 1 and tumor necrosis factor synergistically stimulate prostaglandin synthesis and phospholipase A2 release from rat renal mesanyial cells. Biochem. Biophys. Res. Commun. 159, 385-394 (1989).

[18]

[19] [20]

[21]

[22]

Release of phospholipase A 2 activity from rat vascular smooth muscle cells mediated by cAMP. Eur. J. Biochem. 181,237-242 (1989). C. G. Schalkwijk, F. M~irki and H. Van den Bosch, Studies on the aeyl-chain selectivity of cellular phospholipases A 2. Biochim. Biophys, Acta 1044, 139-146 (1990). R. L. Biltonen, B. K. Lathrop and J. D. Bell, Phospholipases. In Methods in Enzymology, Vol. 197. (Ed. E. A. Dennis) pp. 237 241, Academic Press, New York 1991. P. Elsbach, J. Weiss, R. C. Franson, S. Beckerdite-Quagliata, A. Schneider and L. Harris, Separation and purification of a potent bactericidal~permeability-increasing protein and a closely associated phospholipase A 2 from rabbit polymorphonuclear leukocytes. J. Biol. Chem. 254, 11000 11009 (1979). F. Mfirki and R. Franson, Endogenous suppression of neutral-active and calcium-dependent phospholipase A 2 in human polymorphonuclear leukocytes. Biocbim. Biophys. Acta 879, 149-156 (1986). R.M. Kramer, B. Johansen, C. Hession and R. B. Pepinsky, Structure and properties of a secretable phospholipase A2 from human platelets. In Phospholipase A2. (Ed. P. Y.-K. Wong and E. A. Dennis) pp. 35 53, Plenum Press, New York 1990. G. J. M. Van Scharrenburg, G. H. De Haas and A. K. Slotboom, Regeneration of full enzymatic activity by reoxidation of reduced pancreatic phospholipase Az. HoppeSeyler's Z. Physiol. Chem. 361,571-576 (1980). S. Hara, I. Kudo, H. W. Chang, K. Matsuta, T. Miyamoto and K. Inoue, Purification and characterization of extracellular phospholipase A2from human synovial fluid in rheumatoid arthritis. J. Biochem. 105, 395 399 (1989). W. Pruzanski and P. Vadas, Secretory synovialfluid phospholipase A 2 and its role in the pathogenesis of inflammation in arthritis. J. Rheumatol. 15, 1601 1603 (1988). J. S. Bomalaski, P. Lawton and J. L. Browning, Human extracellular recombinant phospholipase A z induces an inflammatory response in rabbit joints. J. lmmunol. 146, 3904-3910 (1991). W. Pruzanski and P. Vadas, Phospholipase A2 a mediator between proximal and distal effectors of inflammation. Immunol. Today 12, 143 146 (1991). J. H. Gronich, J. V. Bonventre and R. A. Nemenoff, Identification and characterization of a hormonally regulated form of phospholipase A2 in rat renal mesangial cells. J. Biol. Chem. 263, 16645-16651 (1988). D. K. Kim, I. Kudo and K. lnoue, Purification and characterization of rabbit platelet cytosolic phospholipase A 2. Biochim. Biophys. Acta 1083, 80 88 (1991). J. D. Clark, L.-L. Lin, R. W. Kriz, C. S. Ramesha, L. A. Sultzman, A. Y. Lin, N. Milona and J. L. Knopf, A novel arachidonic acid-selective cytosolic PLA 2 contains a Ca 2 +dependent translocation domain with homology to PKC and GAP. Cell 65, 1043 1051 (1991). R. M. Kramer, E. F. Roberts, J. Manetta and J. E. Putnam, The Ca2+-sensitive cytosolic phospholipase A2 is a 100kDa protein in human monoblast U937 cells. J. Biol. Chem. 266, 5268-5272 (1991). J. D. Sharp, D. L. White, X. G. Chiou, T. Goodson, G. C. Gamboa, D. McClure, S. Burgett, J. Hoskins, P. L. Skatrud, J. R. Sportsman, G. W. Becket, L. H. Kang, E. F. Roberts and R. M. Kramer, Molecular cloning and expression of human Ca 2 +-sensitive cytosolic phospholipase A 2. J. Biol. Chem. 266, 14850-14853 (1991).

Agents Actions 38 (1993) [23] R. F. Irvine, How is the level offree arachidonic acid controlled in mammalian cells? Biochem. J. 204, 3 16 (1982). [24] J. Axelrod, Receptor-mediated activation of phospholipase A2 and arachidonic acid release in signal transduction. Biochem. Soc. Trans. 18, 503 507 (1990). [25] R. D. Burgoyne, T. R. Cheek and A. J. O'Sullivan, Receptor-activation of phospholipase A2 in cellular signalling. TIBS 12, 332 333 (1987). [26] P. Vadas and W. Pruzanski, Phospholipase A2 activation is the pivotal step in the effector pathway of inflammation. In Phospholipase A 2. (Eds. P. Y.-K. Wong and E. A. Dennis) pp. 83-101, Plenum Press, New York 1990. [27] W. Yuan and M. H. Gelb, Phosphonate-containing phospholipid analogues as tight-binding inhibitors of phospholipase A2. J. Am. Chem. Soc. 110, 2665 2666 (1988). [28] G. H. de Haas, R. Dijkman, S. Ransac and R. Verger, Competitive inhibition of lipolytic enzymes. IV. Structural details of aeylamino phospholipid analogues important for the potent inhibitory effects on pancreatic phospholipase A 2. Biochim. Biophys. Acta 1046, 249-.257 (1990). [29] K. A. Foster, D. R. Buckle, K. L. Crescenzi, A. E. Fenwick and J. E. Taylor, Arachidonic acid analogues as inhibitors of phospholipase A2 activity. Biochem. Soc. Trans. 15, 418~-19 (1987). [30] L. A. Marshall and J. Y. Chang, Pharmacological control of phospholipase A z activity in vitro and in vivo. In Phospholipase Az. Adv. Exptl. Med. & Biol., Vol. 275. (Eds. P. Y. K. Wong and E. A. Dennis) pp. 169 182, Plenum Press, New York 1990. [31] W. Wilkerson, 1. De Lucca, W. Galbraith, K. Gans, R. Harris, B. Jaffee and J. Kerr, Anti-ir~flammatory phospholipase-A2 inhibitors. 1. Eur. J. Med. Chem. 26,667 676 (1991). [32] D. Lombardo and E. A. Dennis, Cobra venom phospholipase A 2 inhibition by manoalide. A novel type of phospholipase inhibitor. J. Biol. Chem. 260, 7234 7240 (1985). [33] C. F. Bennett, S. Mong, M. A. Clarke, L. 1. Kruse and S. T.

211

[34]

[35]

[36] [37] [38]

[39]

[40] [4l]

[42]

Crooke, Differential effects of manoalide on secreted and intracellular phospholipases. Biochem. Pharmacol. 36, 733-740 (1987). L. A. Marshall, J. Bauer, M. L. Sung and J. Y. Chang, Evaluation of antirheumatic drugs Jor their effect in vitro on purified human synovial fluid phospholipase A 2. J. Rheumatol. 18, 59 65 (1991). D. Mobilio and L. A. Marshall, Recent advances in the design and evaluation of inhibitors of phospholipase A 2. Ann. Rep. Med. Chem. 24, 157 166 (1989). W. W. Wilkerson, Anti-inflammatory phospholipase A 2 inhibitors. Drugs Future 15, 139 148 (1990). J. Wijkander and R. Sundler, A phospholipase A2 hydrolyzing arachidonoyl-phospholipids in mouse peritoneal macrophaqes. FEBS Lett. 244, 51-56 (1989). L. A. Marshall, B. Bolognese, W. Yuan and M. Gelb, Phosphonate-phospholipid analogues inhibit human phospholipase A2. Agents and Actions 34, 106 109 (1991). C. Lanni and E. L. Becker, Inhibition of neutrophil phospholipase A 2 by p-bromophenylacyl bromide, nordihydroguaiaretic acid, 5,8,11,14-eicosatetrayenoic acid and quercetin. Int. Archs. Allergy Appl. lmmunol. 76, 214 217 (1985). B. W. Dijkstra, K. H. Kalk, W. B. J. Hol and J. Drenth, Structure of bovine pancreatic phospholipase A 2 at 1.7 resolution. J. Mol. Biol. 147, 97 123 (1981). D. L. Scott, S. P. White, Z. Otwinowski, W. Yuan, M. H. Gelb and P. B. Sigler, InterJdcial catalysis: The mechanism o/'phospholipase A2. Science 250, 1541 1546 (1990). J.-P. Wery, R. W. Schevitz, D. K. Clawson, J. L. Bobbitt, E. R. Dow, G. Gamboa, T. Goodson, Jr., R. B. Hermann, R. M. Kramer, D. B. McClure, E. D. Mihelich, J. E. Putnam, J. D. Sharp, D. H. Stark, C. Teater, M. W. Warrick and N. D. Jones, Structure of recombinant human rheumatoid arthritic synovial.fluid phospholipase A z at 2.2 A resolution. Nature 352, 79 82 (1991).