Effect of the purified phospholipases A2 from snake and bee venoms on rabbit platelet function

Toxicort, Vol. 22, No. !, pp. Punted in Great Britain .

70s-719, 1994.

0041-0101/84 ß .00+ .00 ® 1984 Pergamon Prat Ltd.

EFFECT OF THE PURIFIED PHOSPHOLIPASES A2 FROM SNAKE AND BEE VENOMS ON RABBIT PLATELET FUNCTION CHAOHO OUYANG and TÙR-FU HUANG Pharmacological Institute, College of Medicine, National Taiwan University, Taipei, Taiwan 100, R.O .C. (Amepted for publication 9 February 1984) OUYANG and Tux-Ftl Hun1vG. Effect of the purified phospholipases A, from snake and bce venons on rabbit platelet function. Toxicon 1.2, 70s- 718, 1984. - Effects of seven purified phospholipases A, from the venons of snakes (Ngja ngja atra, Trimerrsurus mucrasquamatus and T, gmmineus) and honey bee (Apis melljfera) on rabbit washed platelet suspension in the absence of bovine serum albumin have been studied. Only phoapholipases A, from N. n. atra, T. mucrosquamatus and A. melljfera venons induced platelet aggregation with small amounts of '~-serotonin release . They showed tachyphylaxis and also cross-tachyphylaxis in inducing platelet aggregation . The former two phospholipasea A, exhibited biphasic responses in which irreversible aggregations appeared at concentrations of 1-10 pg/ml. At higher concentrations, they elicited the reversible aggregation . Exogenous Ca" was essential to their activity . Indomethacin and EDTA completely abolished both phospholipase A, induced platelet shape change and aggregation, while mepacrine, prostaglandin E verapamil and nitroprusaide inhibited only the aggregation response . p-Bromophenacyl bromide-modified phospholipasea A,, which almost completely lost enzymatic activity, failed to induce platelet aggregation . Phosphatidylcholine, phoaphatidylethanolamine and phoaphatidylinositol inhibited the phospholipase A,-induced platelet aggregation . These phoapholipases A, induced thromboxane B, formation which was inhibited by EDTA and indomethacin, but not by prostaglandin E, . Pre-treatment of platelet suspension with phospholipase A, from N. n. atra or A. melljjera venom (s0 pg/ml) inhibited platelet aggregation induced by sodium arachidonate or collagen, but not that induced by thrombin or ionophore A-23187 . Exogenous sodium arachidonate or lysophosphatidylcholine also showed unaltered inhibitory spectrum on platelet aggregation . It is concluded that phospholipases A, induce platelet aggregation by virtue of their enzymatic activity, cleaving the membrane phoapholipids resulting in arachidonic acid release and formation of thromboxane A, . On the other hand, the cleaved products, lysophoaphatidylcholine, arachidonic acid or arachidonate metabolites (via lipoxygenase pathway) may be responsible for anti-platelet activity . CIiAOFIO

INTRODUCTION PHOSPHOLIPIDS, major constituent of platelet plasma membrane, play important roles in blood coagulation and platelet aggregation (ZWAAL, 1978). Phospholipid metabolism gives rise to arachidonic acid, the precursor of thromboxane A, (HAMBERO et al., 1975), phosphatitüc acid, lysophosphatidic acid (GERRARD et al., 1979) and platelet activating factor (PAF-asether) (DEMOPOULOS et al., 1979; CHIGNARD et al., 1979) all of which are involved in the activation of platelets. Exogenous phospholipases A,, purified from various sources, including snake and bee venoms, have been used to investigate the los

706

CHOAHO OUYANG and TUR-FU HUANG

transverse distribution of plasma membrane phospholipids (CHAP et al ., 1977 ; PERRET et al ., 1979 ; Scx1cK et al., 1981 and ZWAAL et al ., 1975). However, the effects of the purified phospholipases A~ on platelet function and phospholipid metabolism are still obscure. In this paper, we found a biphasic effect of certain phospholipases A, on rabbit platelets. On the other hand, an inhibitory activity of phospholipases A~ on platelet aggregation induced by sodium arachidonate and collagen was also demonstrated . MATERIALS AND METHODS Phospholipase A, (PLA,) from Ngja ngja atra (Formosan cobra) was purified according to the procedure described by Lo et al. (1972) . PLA, from Apis mell(fera (honey bee) venom was purified by the method as described by LAN et al. (1983) . PLA, from Trimeresurus mucrosquamatus snake venom was further purified by the method described by OUYANO et al. (1978) . The PLA, (Fraction 16) was purified by Sephadex G-7S, and G50 gel filtration successively and its amino acid composition was different from that of PLA, (Fraction 20) with anticoagulant activity (OuvnNC et al., 1978). Four phospholipases A, from T. gramineus snake venom (Fractions I, VI, VIII and XII) were fractionated by DEAF-Sephadex A-SO column chromatography (OuvnNc and HUANG, 1979), and they were further purified by Sephadex G-SO successively . The purifies of all these phospholipases A, were verified by both SDS-polyacrylamide gel electrophoresis and isoelectrofocusing (LKB multiphore, using ampholyte, pH 3.3 -10.0). PLA, from the venoms of N. ngja atra, Apis mell(jera and T. mucrosquamatus were modified with pbromophenacyl bromide. PLA, (1 mg/ml) in aqueous solution was mixed with 1/100 (v/v) ofp-bromophenacyl bromide (0 .1 M, in dimethylsulfoxide) and stood at room temperature for 6 hr. Then the reaction mixture was dialysed against 1000 ml of redistilled water at 4°C for 20 hr. The dialysed reaction mixture was frozen and lyophilized. Amino acid analysis was performed on a Beckman 121 M Analyzer using a two column system . Sodium dodecyl sulfate (SDS}polyacrylamide gel electrophoresis was performed according to the method of WessR and OSnORN (1969) . Phospholipase A, enzymatic activity was determined usinga pH-stet titration method (STRONG et al., 1976), in which phosphatidylcholine (2 mg/ml) was used as substrate. Phospholipase A, activity was expressed as the liberation of fatty acid (ianoles/min) per mg of PLA, in the first 6 min. Lactate dehydrogenase assay was carried out according to the method of WROS~ewsxi and LnDue (1955) . Thromboxane B, was assayed as follows. At the sixth min of platelet aggregation response, 50 pl of a mixture consisting of equal volumes of EDTA (20 mM) and indomethacin (200 pM) wasadded to terminate the reaction . After centrifugation (Eppendorf Model 5414, 8,000 g, 3 min) the supernatant was assayed for thromboxane B, ('H-thromboxane B, RIA kit, New England Nuclear Co ., U.S.A .) . Platelet suspension from rabbits was prepared as described by MusTnRn et al., (1972) and modified by OuY.we and HunNC (1983) . The platelet pellets were washed twice with Tyrode's solution containing bovine serum albumin (1 mg/ml) and apyrase (0 .1 mg/ml) . The washed pellets were finally suspended in albumin-free Tyrode's solution . The final concentration of Ca" in platelet suspension was 1 mM . p-Bromophenacyl bromidetreated platelet suspension was prepared as follows . The once-washed platelet suspension was treated with pbromophenacyl bromide (PBPB, 0.1 M) at 37°C for 30 min and centrifuged at 1,500 g for 10 min. Platelet pellets were washed once with Tyrode's solution and finally suspended in albumin-free Tyrode's solution . Platelet aggregation was measured turbidimetrically by the method of BORN and CROSS (1963), using a Payton dual channel aggregometer (Ontario, Canada). Aggregation (%) was calculated as described in a previous paper (OuvnNC et al., 1983). Platelet suspension was pre-warmed at 37°C for 2 min in the siliconized glass cuvette and stirred at 900 rev/min constantly before addition of PLA, or aggregation inducer. When the inhibitory effect of PLA, was being tested, it was usually pre-incubated with platelet suspension at 37°C before the addition of aggregation inducer. Platelet release reaction was measured as follows . EDTA platelet-rich plasma was labeled with S-hydroxy (side chain) 2-'~-tryptamine creatinine sulfate (0.6 NCi/10° platelets), following the method of ARntie et al. (1974) . The labeled platelets were washed twice and finally suspended in albumin-free Tyrode's solution . Release was calculated as % of control (Triton X-100, 0.3% v/v, treated platelet suspension). The venoms of T. mucrosquamatus, T, gramineus and Ngja ngja atra were lyophilized and stored at -20°C. Bovine thrombin was purchased from Parke, Davis and Co ., U.S .A . DEAF-Sephadex A-S0, CM-Sephadex CS0, Sephadex G-7S, G-SO and G-23 were purchased from Pharmacia, Sweden . Bee (Apis melljfera) venom, adenosine diphosphate, collagen (bovine, tendon), sodium arachidonate, p-bromophenacyl bromide, bovine serum albumin, apyrase, lactate dehydrogenase kit, prostaglandin E mepacrine, indomethacin, EDTA, L-aphosphatidylcholine (dipalmitoyl), L-c-phoephatidylethanolamine (dipahnitoyl), L-a-phosphatidylinositol (sodium salt), lysophosphatidylethanolamine and lysophosphatidylcholine were purchased from Sigma Chemical Co ., U.S .A . Ionophore A-23187 was obtained from Calbiochem-Behring Co ., U.S .A.

Venom Phospholipases A, on Platelet Function

F10. l .

EFFECT OF PHOSPHOLIPASE

A, (PLA,)

FROM N. A . afro VENOM PLATELET SUSPENSION .

(NNAV-PLA)

707

ON RABBIT

NNAV-PLA (1- 50 pg/ml) was added ( ~ ) to trigger platelet aggregation measured turbidimetri cauy (~T, change in light transmission) in washed rabbit platelet suspension in the absence of bovine serum albumin . RESULTS

Specificity ofphospholipases A~ activity on platelet function All the phospholipases Az (PLA,, 50 ~g/ml) failed to elicit platelet aggregation if platelets were suspended in Tyrode's solution containing bovine serum albumin (3.5 mg/ ml). They did not affect the thrombin-induced platelet aggregation even after 30 min incubation with platelet suspension in the same medium. Only the acidic PLA, from T. gramineus venom (Fraction XIn noncompetitively inhibited platelet aggregations induced by thrombin, collagen, sodium arachidonate and ionophore A-23187 (OUYANG and HUANG, 1983). However, in the absence of bovine serum albumin, PLAs from venoms of N. n. atra, A. mellifera and T. mucrosquamatus elicited aggregation response, while the PLA, from venom of T. gramineus did not. Phospholipases A3-induced platelet aggregation PLA, from the venoms of N. n. atra and A . mell(fera showed biphasic response on rabbit washed platelets suspended in albumin-free Tyrode's solution (Fig . 1). The optimal concentrations producing maximal effect varied with different lots of platelet suspension . However, they were calculated to be approximately 1-10 ~g/ml (nil 1). This irreversible aggregation lasted longer than an hour . At concentrations higher than 20 Fig/ml, they elicited reversible aggregation . In the process of platelet aggregation, no lactate dehydrogenase activity was detected in the supernatant of suspensions, indicating that no

708

CHOAHO OUYANG and TUR-FU HUANG

FIG .

i.

EFFECT OF

PLA,

FROM

T. mucrosquamatus

SUSPENSION .

vENOIK

(TMV-PLA)

ON RABBIT PLATELET

TMV-PLA (1-SO pg/ml) induced platelet aggregation measured turbidimetrically in rabbit platelet suspension in the absence of bovine serum albumin. cell lysis occurred . On the other hand, PLAN from T. mucrosquamatus venom usually elicited reversible aggregation (Fig . 2, n = ~. The higher the concentration, the faster the disaggregation . Only limited "C-serotonin release reaction (< 11 %) was observed in "Cserotonin preloaded platelets with all these three phospholipases A, (S - 50 ~tg/ml, n~3).

F~ffect of antiplatelet compounds on phospholipases A1-induced platelet aggregation The platelet aggregation induced by PLA= from N. n. atra venom was very susceptible

to inhibition by EDTA (2 mM), indomethacin (5 PM), mepacrine (25 pIvl), prostaglandin E, (1 ~M), nitroprusside (S mM) and verapamil (SS ~. However, the platelet shape change (the initial decrease of light transmission) produced by this PLA, was inhibited only by EDTA and indomethacin (Fig . 3) . Similarly, the platelet aggregation induced by PLAz from the venom of A. mell fera and T. mucrosquamatus was susceptible to inhibition by these anti-platelet compounds (not shown) .

Chemical mod cation of phospholipases Az When PLAz from the venoms of N. n. atra, A. mell~fera and T. mucrosquamatus were

modified with p-bromophenacyl bromide, the modified enzymes (20 Ng/ml) lost their aggregation response (Fig . 4) . Amino acid analysis showed that p-bromophenacyl

Venom Phospholipases A, on Platelet Function

70 9

"T 10'k 1 Min IND

a

FIG . 3 . THE SUSCEPTIBILITY OF PLATELET AGGREGATION INDUCED HY PLA FROM N. A. (NNAV-PLA) IN THE ABSENCE OF ALBUMIN .

Qtla

VENOM

Indomethacin (IND, 5 ~, mepacrine (MEP, 25 ~, EDTA (2 mM), PGE, (1 ~, nitroprus side (NP, S mM) or verapamil (VER, SS ~ was added 1 min prior to the addition of NNAV-PLA (10 iIg/ml).

bromide alkylated the only histidine residue in T. mucrosquamatus PLAN. Although the modified bee venom (Apis mell(fera) PLA1 (20 pg/ml) lost its aggregation activity, it still induced a platelet shape change (Fig . 4, c panel), and at 50 pg/ml it induced platelet aggregation . This may result from the residual enzymatic activity (6%). When the platelet suspensions were pretreated with their respective phenacyl bromide modified PLAN enzymes (SO pg/ml), the platelet aggregations induced by the native PLA, enzymes from N. naja atra or T. mucrosquamatus venoms were not affected . The specific enzymatic activity of phospholipases A,

The enzymatic activities of PLA= from venoms of NaEja n. atra, A. melllfera and T. phosphatidylcholine were estimated to be 490 t 10, 720 t 23, and 862 t 26 Emloles/min per mg (n = 3), respectively, and those of PLA= from T. gramineus venom (Fraction I, VI, VIII and XII) were 46 t 10, 342 t 9, 472 t 32 and 260 t 17 Eanoles/min per mg (n = 3), respectively . After phenacyl bromide modification, muscrosquamatus towards

71 0

CHOAHO OUYANG and TUR-FU HUANG

FIO . 4 . EFFECT OF jNBROMOPHENACYL BROMIDE-MODIFIED PLAT ON RABBIT PLATELET SUSPENSION . Native (upper tracing, 10 pg/ml) or p-bromophenacyl bromide-modified PLA, (lower tracing, 20 ~rg/ml) was added to trigger platelet aggregation, NNAV-PLA (a panel), TMV-PLA (b panel) or BV-PLA {c panel) denotes PLA, from N. n. atra, T. mucrosquamatus or A. mellifera (honey bee) venom, respectively .

the residual enzymatic activities of the PLA, from the venoms of N. n. atra, A. mellifera and T. mucrosquamatus were 14 t 3, 42 t 2 and 3 t 2 ~tnoles/min per mg, respectively . Tachyphylaxis ofphospholipases A, When platelet suspensions were pre-treated (37°C, 3 min) with inactive phospholipases A, from T. gramineus venom (Fractions I, VI and VIII), the aggregation activity of N. n. atra and A. mellifera PLA, were not affected . However, when platelet suspensions were pretreated with N. n. atra PLA, (SO !`g/ml), the successive addition of A . mellifera PLA, (101tg/ml) failed to elicit an aggregation response (Fig . Sa, n = 3) . Conversely, when A . mellifera PLA, (2 pg/ml) was added prior to N. n. atra PLA, (10 ~g/ml), the latter failed to induce platelet aggregation (Fig. Sb, n =12). Similarly, when T. mucrosquamatus PLA, (20 ttg/ml) was added prior to A. mellifera (2 Ng/ml) or N. n. atra (10 tag/ml), the latter two failed to induce aggregation (Fig. Sc, d; n = 6 and 3, respectively). Tachyphylaxis also rapidly developed with the PLA, enzymes . Pretreatment of platelet suspension with a low

Venom Phospholipases A, on Platelet Function

NNAV-PLA

TMV-PLA

e.

1 NNAV-PLA

"T~10'~f, 1Min

~ NNAV~PLA

FIG . S . CROSS-TACHYPHYLAXIS AMONG PLA, ENZYMES FROM N. n . atfa (NNAV-PLA), A. mel/jjefa (BV-PLA) AND T. mucrasquamatus vENOM (TMV-PLA) 1N INDUCING AGGREGATION OF RABHri PLATELET SUSPENSION . (a) NNAV-PLA (50 pg/ml) was added 6.S min prior to BV-PLA (10 pg/ml) . (b) BV-PLA (2 fig/

ml) was added S min prior to NNAV-PLA (10 pg/ml) . TMV-PLA (20 ~g/ml) was added 3 min prior to BV-PLA (10 pg/ml) (c) or NNAV-PLA (10 pg/ml) (d) . (e) NNAV-PLA (2 Ng/ml) was added prior to the successive addition of NNAV-PLA (10 pg/ml) (left panel) ; NNAV-PLA (10 pg/ml, control, right panel), was added to trigger platelet aggregation .

TABLE

1 . THROMHOXANE Ba FORMATION INDUCED BY PHOSPHOLIPASES A, (PLA,) FROM N. n . atra, A. melJjfera AND T. mucrosquamatus vENOMs

Dose

Thromboxane B, formation (ng/ml) "NNAV-PLA

"BV-PLA

"TMV-PLA

lOt4 lOf4 0 lOt4 (S) (S) (S) 1 72 t 8 (3) 100 t 12(3) 65 t S (3) 10 254 t 24(3) S17 t 50(3) 158 t 20(3)

Values are presented as mean t S .E . (n) . "NNAV-PLA : PLA, from N. n . atra venom. BV-PLA: PLA, from A . mell~era venom . TMV-PLA: PLA, from T. mucrasquomatus venom.

dose of N. n. atra PLAs (2 pg/ml) inhibited the aggregation activity of a subsequent administration of a high dose of the same enzyme (10 pg/ml) (Fig. Se, n=6). When platelet suspensions were pretreated with a low dose of A. mell~fera PLAz (0.5 - 2.5 pg/ml), the subsequent addition of A. mell(jera PLA= (10 Fig/ml) also failed to induce aggregation (data not shown) . Thromboxane B~ formation PLA, from venoms of N. n. atra, A . mell(fera and T. mucrosquamatus dosedependently induced thromboxane B, formation in platelet suspension in the absence of albumin (Table 1). PLA, from A. mell(fera was the most potent in this respect, followed

CHAOHO OUYANG and TUR-FU HUANG

71 2

1Min

FIG .

6.

EFFECT OF PLA3 FROM

N.

)f .

Rtra VENOM (NNAV-PLA, AGGREGATION .

SO /rg/ml) ON PLATELET

The washed rabbit platelet suspension was pretreated with NNAV-PLA at 37°C for 5 or 10 min before the addition of thrombin (THR, 0 .05 U/ml), collagen (COL, 10 pg/ml), sodium arachidonate (SA, 10 ~ or ionophore A-23187 (1 ~ . The upper tracing in each panel denotes the control (induucr alone in the absence of PLA,).

by N. n. atra and T. mucrosquamatus PLA,. At concentration of 10 pg/ml, PLA, from A . mellifera venom increased thromboxane B, formation about SO-fold, while PLA, from venoms of N. n. atra and T. mucrosquamatus increased formation about 2S and 15-fold, respectively, when compared to controls (10 t 4 ng/ml) . p-Bromophenacyl bromidemodified PLA, (20 pg/ml) from venoms of N. n. atra, A . mellifera and T, mucrosquamatus induced only 9 t 1, 19 t 6 and 18 ± 4 ng of thromboxane B, formation per ml. Indomethacin (5 ~ almost completely inhibited the thromboxane B, formation induced by PLA, from venoms of N. n. atra and T. mucrosquamatus (all at 10 pg/ml, residual thromboxane B,, 17 and 14 ng/ml respectively). EDTA (2 mM) prevented the PLA,induced thromboxane B, formation. Mepacrine (25 NM) partially inhibited thromboxane B, formation induced by PLA, from venoms of N. n. atra, T. mucrosquamatus and A . mell(fera (all at 10 pg/ml, residual thromboxane B,, 85, 82 and 108 ng/ml respectively). PGE, (1 ~ had no significant effect on the PLA,-induced thromboxane B, formation. The non-aggregating PLA, from T. gramineus venom did not induce any significant increase of thromboxane B, formation. The yields of thromboxane B, were 10 ng/ml (Fraction I), 8 ng/ml (Fraction VI), 26 ng/ml (Fraction VIII) and 10 ng/ml (Fraction XIn. Sodium arachidonate (10 and 25 ~ induced thromboxane B, formation (360 and 576 ng/ml, respectively). Inhibitory activity ofphospholipases A, on platelet aggregation When the washed platelet suspension was pretreated with N, n. atra PLA, in the absence of bovine serum albumin, it inhibited platelet aggregation induced by sodium arachidonate or collagen (Fig. 6) . Sodium arachidonate (4-10 pIvl}induced platelet aggregation was completely inhibited by N. n. atra PLA, (10 Ng/ml), A . mell(fera PLA,

Venom Phospholipaaes A, on Platelet Fundion

FIG .

7.

71 3

EFFECT OF SODIUM ARACHIDONATE ($A, 1 - 100 ~M) ON WASHED RABBIT PLATELET SUSPENSION IN THE ABSENCE OF BOVINE SERUM ALEUMW.

Platelet aggregation was measured turbidimetrically (DT, change in light transmission) .

(5 ~g/ml) or T. mucrosquamatus PLA, (20 Fig/ml) (n = 3 - 9). After 10 min incubation of platelet suspension with these phospholipases A,, the percent inhibition produced by N. n. atra PLA=, A. mell(fera PLA, or T. mucrosquamatus PLA3 (all at 50 I,~g/ml) on collagen (10 ~g/ml)-induced platelet aggregation was 90 t 1(n = 4), 76 t 9 (n = 4) or 17 t 8 (n = 3), respectively. The longer the incubation time of N. n. atra PLA, with platelet suspension, the greater its inhibitory effect on collagen-induced platelet aggregation (S or 10 min preincubation; 39 t 9 or 90 t 1 % inhibition, n = 3 - 4). However, these phospholipases A= (50 ~tg/ml) did not significantly inhibit the platelet aggregation induced by thrombin (0 .05 U/ml) or ionophore A-23187 (1 ~ (Fig. 6). Inhibitory effect of sodium arachidonate and lysophosphatidylcholine on platelet aggregation Sodium arachidonate (1-100 ~ also showed a biphasic response on platelet aggregation (Fig. 7). The maximal aggregation response was induced at about 10 pM. At higher concentrations (20 ~M), it elicited reversible aggregation . Sodium arachidonate (SO pM) inhibited platelet aggregation induced by collagen (20 Fig/ml) p 80% inhibition), but not that induced by thrombin (0.05 U/ml) (Fig. 8). When platelet suspension was pretreated with sodium arachidonate (2 ~, the sucxessive addition of N. n. atra, A. mell~fera or T. mucrosquamatus PLA, (all at 10 ~Ig/ml) failed to induce platelet

71 4

CHAOHO OUYANG and TUR-FU HUANG

FIG. H. EFFECT OF SODIUM ARACHIDONATE (SA, SO PM) OR LYSOPHOSPHATIDYLCHOLINE (LySO-PC, 42 ~ON PLATELET AGGREGATION INDUCED BY THROMBIN (THR, O.OS U/IBI) OR COLLAGEN (COL, 20 pg/ml) .

aggregation. Lysophosphatidylcholine (42 pIN) inhibited the platelet aggregations induced by collagen (20 pg/ml), N. n. atra, A. mellifera or T. mucrosquamatus PLA2 (all at 10 Fig/ml), but not that induced by thrombin (0.05 U/ml) (Fig. 8). Phosphatidylinositol (20 pg/ml), phosphatidylethanolamine (SO pg/ml) as well as phosphatidylcholine (SO fig/ ml) inhibited the phospholipases A,-induced platelet aggregations . Lactate dehydrogenase assay No significant lactate dehydrogenase activity was detected in the supernatant of platelet suspensions treated with N. n. atra, A . mellifera or T. mucrosquamatus PLAN enzymes (all at 10-SO pg/ml) . DISCUSSION

Three PLA, enzymes, purified from the venoms of N. naja atra, Apis mellifera and T. mucrosquamatus, induced platelet aggregation of rabbit platelet suspension in the absence of bovine serum albumin. The former two PLA, enzymes showed a biphasic response; at concentrations less than 10 pg/ml, they induced dose-dependently an

Venom Phospholipases A, on Platelet Function

71 5

increasing aggregation response, however, they elicited a decreasing, reversible aggregation at concentrations higher than 20 pg/ml (Fig. 1). They did not induce platelet aggregation when the platelets were suspended in bovine serum albumin. These PLA, enzymes induced a negligible amount of thromboxane B, formation in the albumincontaining medium. The isoelectric points of A . mellifera and N. n. atra PLA, were reported to be 10.5 (SHIPOLINI et al., 1974) and 5.4 (CHANG et al ., 1975) respectively, while those of T. mucrosquamatus and T, gramineus PLA, (Fractions I, VI, VIII and XII) were 5.6, 10, 5 .4, 4.2 and 3 .6, respectively (unpublished data). The specific activity of PLA, toward phosphatidylcholine was in the following decreasing order : T. mucrosquamatus > A. mell{fera > N. n. atra > T. gramineus (Fraction VIII > Fraction VI > Fraction XII > Fraction I). The platelet-activating actions of PLA, enzymes were not parallel with either their enzymatic activities or their net charges . The lateral surface tension of platelet membrane was reported to be greater than that of red blood cell membrane (ZWAALL et al., 1975 ; CHAP et al., 1977). Therefore, PLA, exerts a more limited hydrolysis on platelet membrane phospholipids than on red blood cell membrane phospholipids. PLA, from the venons of N. n. atra, A. mellifera and T. mucrosquamatus dose-dependently induced thromboxane B, formation (Table 1). Both the aggregation activity and the increase of thromboxane B, formation were inhibited in parallel by indomethacin and EDTA. Chemical modification of PLA, with p-bromophenacyl bromide abolished both enzymatic and aggregation activities and thromboxane B, formation. Therefore, PLA, probably induced platelet aggregation by an enzymatic action on platelet membrane phospholipids, leading to release of arachidonic acid and subsequent thromboxane formation, which is in agreement with the suggestion made by CHAP et al. (1977, 1982). The non-aggregating PLA, enzymes from T. gramineus (Fractions I, VI, VIII and XII) may belong to the another group, which has no action on intact platelet membranes (CHAP et al. 1977, 1982). This speculation is supported by our data that these non-aggregating PLA, enzymes from T. gramineus venom caused no significant increase in thromboxane B, formation. Therefore, the thromboxane B, levels produced by these PLA, enzymes seems to be in parallel with their potency in inducing platelet aggregation. The specificity of PLA, on platelet aggregation may depend on the substrate specificity, the penetrating activity of these PLA, enzymes and the compactness of platelet membrane phospholipids, in accordance with suggestions made by other authors (ZWAAL et al., 1975; DEI~r. et al., 1975; CHAP et al., 1979, 1982) . PLA, from the venons of N, n. atra, A, mellifera and T. mucrosquamatus all showed tachyphylactic effects on rabbit platelets and cross-tachyphylaxis to each other. On the other hand, PLA, from T. gramineus venom did not show these properties . These phenomena may be due to the limited accessible domains of phospholipids (BILLAH et al., 1980; KANA(3I et al., 1981) or to platelet cyclooxygenase inactivation caused by PLA,, as described with trypsin and thrombin (EGAN et al., 1976; LAPETINA and CuATxECASAS, 1979). Sodium arachidonate showed a biphasic response, however, it did not show tachyphylaxis when a low dose was initially given (unpublished observation) . Therefore, the limited accessible phospholipid domains may result in cross-tachyphylaxis among these active PLA, enzymes . PLA,-induced platelet aggregation was very susceptible to inhibition by EDTA, which may be due to the known Ca'* dependence of PLA,. Verapamil, a blocker of Ca'* influx (MSYER et al., 1972), inhibited PLA,-induced platelet aggregation, thus, PLA, may cause Ca=* influx indirectly, resulting in platelet aggregation. Prostaglandin El, a cAMP elevating compound, and nitroprusside, an inhibitor of intracellular Ca=* flux

71 6

CHAOHO OUYANG and TUR-F[1 HUANG

and SHEPHERD, 1976), all inhibited PLA,-induced platelet aggregation. These results indicate that intracellular Ca'+ mobilization is also involved in this process. Mepacrine, an inhibitor of the activation of endogenous PLAN (BLACKWELL et al., 1978), also partially and non-specifically inhibited thromboxane B~ formation. PLAz retained full aggregation activity in p-bromophenacyl bromide-treated platelet suspension, in which the endogenous PLAN was partially inactivated. Thus, PLA=-induced platelet aggregation was not through activation of endogenous PLAN. The PLA,-induced platelet shape change was inhibited by EDTA and indomethacin but not by mepacrine, prostaglandin E,, verapamil or nitroprusside . EDTA and indomethacin almost completely inhibited thromboxane B~-formation produced by these active PLA2 enzymes while prostaglandin E, had no effect . This implies that the platelet shape change induced by PLAz is also related to thromboxane formation . At higher concentrations, PLA, from the venoms of N. n. atra, A. mellifera and T. mucrosquamatus (> 20 ~g/ml) and sodium arachidonate caused a decreasing, reversible aggregation response (Fig. 1, 2 and 7), the arachidonate effect being in agreement with the report Of FRATANTONI and POINDEXTER (1981) . PLA, at a lower concentration (10-20 pg/ml) preferentially inhibited platelet aggregation induced by sodium arachidonate . At higher concentration (50 pg/ml), PLAz from the venoms of N. n. atra and A. mellifera also inhibited platelet aggregation induced by collagen, but not that induced by thrombin or ionophore A-23187. The inhibitory activity may be caused by the cleaved products which include arachidonic acid, 12-hydroperoxy-5,8,10,14eicosatetraenoic acid (HPETE), prostaglandins and lysophospholipids ; the latter (especially lysophosphatidylcholine) exerts a preferential inhibition on sodium arachidonate or collagen-induced platelet aggregation (Fig. 8). Lysophosphatidylcholine inhibits platelet aggregation (Jo1sT et al., 1977). Indomethacin and mepacrine did not overcome the inhibitory effect of PLAN (unpublished data), indicating that prostaglandins formed via the cyclooxygenase pathway were not involved . However, 12-hydroperoxy-5,8,10,14, eicosatetraenoic acid (12-HPETE), an arachidonate metabolite (via the lipoxygenase pathway), may be responsible for the inhibitory action, since it is inhibitory to platelet cyclooxygenase (SIEGEL et al., 1979). Venom PLAN enzymes may be classified into three groups according to their effects on platelet function . (1) Active PLA, enzymes which induce platelet aggregation of rabbit platelet suspensions in the absence of albumin and exhibit biphasic responses on platelet aggregation in the presence of Ca'+ . This platelet activating action is intimately related to thromboxane formation. This group includes PLA, from the venoms of A. mellifera, N. n. atra and T. mucrosquamatus. (2) PLAN enzymes which do not induce platelet aggregation even in the absence of albumin and non-competitively inhibit platelet aggregation even in the presence of albumin. Their activity is independent of their enzymatic activity . This group includes PLA, from the venoms of T. gramineus (Fraction XIn and Agkistrodon halys (OUYANG and HUANG, 1983 ; OUYANG et al., 1983). (3) Nonaggregating PLA, enzymes which exert no apparent effect on platelet function . Included in this group are three PLA, enzymes from T. gramineus venom (Fractions I, VI and VIII). They induced negligible amounts of thromboxane B, formation in platelet suspensions . (VERHAEGHE

Acknowledgements - This work was supported by a National Science Council reuarch grant of the Republic of

China (NSC71-0412-ß002-0g) . The authors also thank Dr . Lowst .i E. McCoy, Department of Physiology, wayne State University, Detroit, for his cooperation in the amino acid analysis on these native and ~ bromophenacyl bromide-modified phospholipases A, .

Venom Phospholipases A, on Platelet Function

71 7

REFERENCES Hwx, (1974) AxnLte, N. G. and P. Platelet release reaction : enzymatic basis for platelet aggregation and release. The significance of the platelet atmosphere and the relationship between platelet function and blood coagulation. Br. J. Haematol. 26, 331. BtLt.wij, M. M., Lwpsrtxw, E. G., CuwTxacwsws, P. (1980) Phospholipase A, and phospholipases C activities of platelets; Differential substrate specificity, Cai' requirement, pH dependence, and cellular localization . J. biol. Chem . 255, 10227. BLwcxwet,L, G. J., FwweR, R. J ., NIJKAMP, F. P. and VAxs, J. R. (1978) Phospholipase A, activity of guineapig isolated perfused lungs: stimulation and inhibition by anti-inflammatory steroids . Br. J. Pharmacol. 62, 79. Boxx, G. V. R. and Cxoss, M . T. (1963) The aggregation of blood platelet . J. PhystoL, Lond. 168, 178. Ctuwe, W. C., CHwrte, C. S., Hsu, H. P. and Lo, T. B. (1975) Isoenzymes of phosphoGpase A from Formosan Cobra (Ngja ngja atra) venom. J. Chinese Biochem. Soc. 1, 77 . CttwP, H., ZwwwL, R. F. A. and Vwx Deetvex, L. L. M. (1977) Action of highly purified phospholipases on blood platelets. Evidence for an asymmetric distribution of phospholipids in the surface membrane . Biochim. biophys. Acta 467, 146 . CFIAP, H., PsxxEr, B., Mwuco, G., PLANrAVID, M., LAFFONT, F., SiMOx, M. F. and Dousre-Bt.wzv, L. (1982) Organization and role of platelet membrane phospholipids as studied with phospholipases A, from various venoms and phospholipases C from bacterial origin. Toxicon 20, 291 . CHtexwxn, M., Le Countc, J . P., TExc$, M., VAROAFr~G, B. B. and Bewveav~sre, J. (1979) The role of plateletactivating factor in platelet aggregation. Nature, Lortd. 279, 799. Dam., R. A., GEUx~rs Vwx KasssL, W. S. M., ZwwwL, R. F. A., RoewFSSx, B. and Vwx DeExux, L. L. M. (1975) Relation between various phospholipase actions on human red cell membranes and the interfacial phospholipid pressure in monolayers . Biochim. biophys. Acta 406, 97 . DEMOPOtn.os, C., P~xcxwxn, R. N. and HANAHAN, D. J. (1979) Platelet-activating factor. Evidence for 1~ alkyl-2-acetyl-sn-glyceryl-3-phosphorylcholine as the active component (a new class of lipid chemical mediators) . J. biol. Chem . 254, 9335 . Eowx, R. W., PwxTOx, J. and KunttL, F. A. Jx . (1976) Mechanism for irreversible self~eactivation of prostaglandin synthetase . J. biol. Chem . 251, 7329 . FRATANTONI, J. C. and PonvnexrEx, B. J. (1981) Characterization of the platelet response to exogenous arachidonic acid. 77rromb. Res. 22, 157. GF.xxwxn, J. M., K[Nnont, S. E., PL~ttsox, D. A., Pe~t.Ex, J., Kxwx~z, K. E. and Ww~rE, J. G. (1979) Lysophosphatidic acids. Influence on platelet aggregation and intracellular calcium flux . Am . J. Path. 96, 423. HwMaBxe, M., Svewssox, J. and Swnstmtssox, B. (1975) Thromboxanes : a new group of biologically active compounds derived from prostaglandin endoperoxides. Proc. Natn. Acad. Sci. U.S.A. 72, 2994 . Jo~sr, J. H., Dot..ezzeL, G., Cucuiwxu, M. P., Ntst~azwww, E. E. and Musrwxo, J. F. (1977) Inhibition and potentiation of platelet function by lysolecithin . Blood 49, 101. Kwxxwat, R., KOIZUMI, K. and M~sunw, T. (1981) Limited hydrolysis of platelet membrane phospholipids on the proposed phospholipase-susceptible domain in platelet membranes. J. biol. Chem. 256, 1177 . Lwrt?~rtxw, E. G. and Cuwrxecws~s, P. (1979) Rapid inactivation of cyclooxygenase activity after stimulation of intact platelets. Proc. Natn. Acad. Sci. U.S.A . 76, 121. Lnv, S.-C., Huwxc, T. F. and Ouvwxe, C. (1983) Characterization of the anticoagulant principles from Apis melljfera (honey bee) venom. J. Formosan Med. Ass. 82, 629. Lo, T. B., Ct~twxe, W. C. and Crtwrtc, C. S. (1972) Chemical studies on phospholipaseA from Formosan cobra (Ngja ngja atra) venom. J. Formosan Med. Ass. 71, 138. MavEtt, C. J., Vwx BoRp~N, C. and CASrEEI$, R. (1972) The action of lanthanum and D-600 on the calcium exchange in the smooth muscle cells of guinea pig taenia coli . Pflügers Arch . ges. Physiol. 337, 333. Musrwxn, J. F., PexxY, D. W., Axni.te, N. G. and PwcxttwM, M. A. (1972) Preparation of suspensions of washed platelets from humans . Br. J. Haematol. 22, 193. Ouxwxa, C. and Huwxe, T. F. (1979) a- and ß-fibrinogcnases from Trimerrsurus gramineus snake venom. Biochim. biophys. Acts 571, 270. Ouvwxe, C. and Huwxe, T. F. (1983) Potent platelet aggregation inhibitor from 7Yimercsurus gramirreus snake venom. Btochim. biophys. Acts 757, 332. OuYwxe, C., TExG, C. M., Ctw, Y.-C. and Lnv, S.-C. (1978) Purification and characterization of the anticoagulant principle of 7i7merrsurus mucrasquamatus venom. Biochim. biophys. Acts 541, 394. Ouxwxe, C., Ysrt, H.-I. and Huwxe, T .-F. (1983) A potent platelet aggregation inhiAgkistrodon halys (Mamushi) snake venom. Taxitort 21, 797. PeRxe'r, B., Ctiwr, H. and Donsra-Bt.wzv, L. (1979) Asymmctric distribution of arachidonic acid in theplasma membrane of human plateleù . A determination using purified phoapholipases A and a rapid method for membrane isolation. Biochim. biophys. Acts 556, 434. SCHICIC, P. K., Sctttctt, H. P., Bxwxnats, G. and Mn.ts, D. C. B. (1981) Distribution of phosphatidylethanolamine arachidonic acid in platelet membrane. Biochim. biophys. Acts 643, 659.

71 8

CHAOHO OUYANG and TUR-FU HUANG

R. A., CALLEWAERT, G. L., COTTRELL, R . C. and VERNON, C. A. (1974) The amino-acid sequence and carbohydrate content of phospholipase A, from bee venom. Eur. J. Biochem. 48, 465. SIEGEL, M. L, MCCONNELL, R. T., AHRAHAMS, S. L., PORTER, N. A . and Cti.uTRECwsws, P . (1979) Regulation of arachidonate metabolism via lipoxygenase and cyclo-oxygenase by 12-HPETE, the product of human platelet lipoxygenase . Biochem. biophys. Res. Commun. 89, 1273 . STRONG, P. N., GOERKE, J., OBERO, G. and KELLY, R. B. (1976) ß-Bungarotoxin, pre-synaptic toxin with enzymatic activity. Pros. Natn . Acad. Sci. U.S .A . 73, 178 . VERHAEGHE, R. H. 8IId SHEPHERD, J. T. (1976) Effect of nitroprusside on smooth muscle and adrenergic nerve terminals in isolated blood vessels. J. Pharmacol. Exp. Ther. 199, 269. WEBER, K. and OseoRN, M. (1%9) The reliability of molecular weight determination by dodecyl-sulfatepolyacrylamide gel electrophoresis. J. bioi. Chem . 244, 4406 . WROHLEWSKI, F. and Lw DUE, J. S. (1955) Lactic dehydrogenase activity in blood. Proc. Soc. exp. Bioi. Med. 90, 210. ZwwwL, R. F. A. (1978) Membrane and lipid involvement in blood coagulation . Biochim. biophys. Acta 515, 163. ZwwwL, R. F. A., ROELOFSEN, B., COMFURIUS, P. and VwN DEENEN, L. L. M. (1975) Organization of phospholipids in human red cell membranes as detected by the action of various purified phospholipases . Biochim. biophys. Acta 406, 83 . SHIPOLINI,