Stimulation of Phospholipase A2 by Toxic Main Group Heavy Metals: Partly Dependent on G-proteins?

Organometals induce platelet aggregation and inorganic metal ions such as Cd2+ or Pb2+ sensitise human blood platelets to aggregating agents and this action is associated with the liberation of arachidonic acid and eicosanoid formation. The same mechanism is observed using human leukaemia cells (HL-60) when treated with MeHgCl or Et3PbCl. The fatty acid liberation within human platelets and HL-60 cells could only be inhibited with phospholipase A2 inhibitors of different specificity. Preincubation of the cells with pertussis toxin reduces the activation induced by Et3PbCl to a great extent. The non-catalytic B subunit, that only mediates the binding of the toxin to the cell membranes, has no effect at all. When summarised, these results suggest that one possible mechanism for the stimulation of phospholipase A2 by Et3PbCl functions via a G-protein dependent pathway.


INTRODUCTION
The cascade of arachidonic acid liberation and its metabolisation becomes more and more hnportant within physiological and pathological processes. With regard to immunological and inflammatory reactions there are some hints for the involvement of xenobiotics in these mechanisms.L2 Cd 2+ ions sensitise human blood platelets to aggregating agents 3 and the organometallic compounds methylmercury (MeHgC1), triethyllead (Et3PbCI) and triethyltin induce platelet aggregation. 4"6 This activation of platelets is associated with the liberation of arachidonic acid and eicosanoid formation. As these metals accumulate in the environment 7 and the biosphere s the effects of these compounds and their ability to increase lipid mediators of inflammatory reactions are of great interest. The possible induction of the cellular lipid metabolism by xenobiotics and the following increase of available precursors of lipid mediators could possibly lead to immunotoxic effects.
These experiments gave a greater insight into the mechanism by which the lipid metabolism is affected. The results shown here corroborate the assumption that the activation of phospholipase A 2 is at least partly triggered via a G-protein.
Platelet-rich plasma and determination of aggregation Fresh human blood from healthy donors (3.8% citrate/blood, 1:9, v/v) was centrifuged at 340 g for 10 min at 22C to get platelet-rich plasma (PRP). Donors must have abstained from all drugs for more than 2 weeks. Platelet-poor plasma was prepared by further centrifugation of the remaining blood at 5 000 g for 15 rain at 4C. After counting the platelets in PRP their number was adjusted with autologous platelet-poor plasma to 300 000/pl. PRP was added to each cuvette, stirred at 1 000 rpm at 37C and the light transmission during aggregation was monitored by use of an Elvi aggregometer. Heavy metal compounds were added as aqueous or ethanolic solution to give the concentrations indicated. Incubation of ilL-60 cells HL-60 cells were grown in suspension culture in RPMI 1640 medium supplemented with 15% foetal calf sennn. They were induced to differentiate to mature granulocytes by the addition of 1.3% dimethyl sulphoxide for 5 days. The cells were harvested by centtifugation, washed once with RPMI without any additives and f'mally resuspended in medium containing 1% dimethyl sulphoxide and 3.3% foetal calf serum at a concentration of 1 107 cells/ml. Experiments were started after 30min standing. The cell suspensions (3 ml) were then incubated at 37C with 10ttM calcium ionophore A23187 or Et3PbCI as indicated. In the case of radioactive prelabelling, [t4C]-arachidonic acid was dissolved in dimethyl sulphoxide, added at day 4 (92.5 kBq/50 ml) to the culture medium and the cells were incubated overnight. The labelled cells were washed twice with RPMI and resuspended as described above.
Lipid extraction and separation of lipid classes After incubation of the cell suspensions the lipids were extracted as reported earlier. 9 The extract was dried under nitrogen, taken up in chloroform, applied to bonded phase aminopropyl columns (Waters) and separated into a phospholipid-, free fatty acidand triacylglyceml-fraction. After hydrolysis of the triacylglycerols and the phospholipids, the fatty acids were estered with pBPB. The phenacyl esters of the fatty acids were then analysed using HPLC. Radioactive lipids were extracted as described above. The dried lipids were taken up in CHCI 3 and spotted onto SIL G polyester plates (20 cm 20 cm) and separated by thin-layer chromatography as described elsewhere. 9 The Rf-values for the lipid classes were determined by comparison of their migration with that of commercial standards. This system gives good separation of all cellular lipids. HPLC analysis The experimental equipment consisted of two HPLC pumps (Waters, model 510), an automated sample processor (Waters, WISP), a programmable multiwavelength detector (Waters, model 490), and the chromatograms were evaluated with a Waters Maxima 820 chromatography data station. The analysis was carried out as reported earlier. 12 The fatty acid esters were detected at 254 nm Heavy metal compounds were added to PRP to give the final concentrations as indicated. CdO was dispersed by soniflcation in buffer. After a preincubation pedod of 5 rain, collagen or arachidonic acid were added in sub-threshold concentrations and aggregation was measured using an Elvi aggregometer and constant stirdng for a further 15 rain. Values of aggregation are given in % of light transmission as compared to stimulation with optimal concentrations of collagen or arachidonic acid, respectively. (R): sub-threshold concentrations of collagen (0.32 pg/ml) or amchidonic acid (0.26 mM) that were not able to induce platelet aggregation by themselves.

Platelet aggregation studies
Human blood platelets can be stimulated in vitro with exogenous arachidonic acid or collagen to aggregate. However, low concentrations of these agents, 0.26 mM arachidonic acid or 0.32 pg/ml collagen, were not sufficient to stimulate platelets. On the other hand, inorganic metal ions as Pb 2+ or Cd 2+ showed synergistic action in activation of blood platelets together with these low concentrations of physiological inducers (Tab. 1). Although other inorganic compounds like Hg 2+ or CdO failed to induce platelet aggregation, in solution or suspension of small particles, respectively, the organic heavy metal compounds Et3PbCI and MeHgCI were able to stimulate this reaction without the addition of any physiological agent (Tab. 1).

Inhibitors of platelet functions andphospholipase A 2
Previous reports have shown that human platelets, as well as HL-60 cells, liberate arachidonic acid from phospholipids and metabolise it to eicosanoids following incubation with the organometallic compounds. 5,9 Compared to the controls with or without thrombin, Et3PbCI and MeHgCI induced the liberation and metabolisation of arachidonic acid to a stupendous extent (Tab. 2). To elucidate which step(s) in the reaction cascade is(are) affected, [3H]-arachidonic acid labelled platelets were incubated with different compounds capable of affecting enzyme or metabolic reactions. Forskolin, a potent stimulator of adenylate cyclase, inhibits platelet aggregation by raising the intracellular level of cyclic AMP. Pretreatment of platelets with forskolin and Partly Mediated by G-Proteins? subsequent incubation with either Et3PbCI or MeHgCI, however, could not prevent liberation of arachidonic acid and its metabolisation (Tab. 2) in spite of total inhibition of platelet aggregation.
The action of acetylsalicylic acid (ASA) was tested; ASA is a known inhibitor of cyclooxygenase. Even in this case only the aggregation was prevented because the formation of the cyclooxygenase products thromboxane B 2 and 12-hydroxy-5,8,10-heptadecatrienoic acid was drastically reduced, whereas the lipoxygenase product 12-hydroxy-5,8,10,14-eicosatetraenoic acid was increased (Tab. 2). Only the third compound, quinacrine, was able to inhibit both aggregation and arachidonic acid liberation induced by the heavy metal compounds (Tab. 2). 'Prelabelled platelets wer preincubated'with 100 IJM forsk01in (15 min), mM acetylsalicylic acid (ASA,15 min) or mM quinacrine (5 rain) and prelabelled HL-60 cells were preincubated with mM quinacrine (5 min) or 50 pM pBPB (30 rain) before thrombin, fMLP, Et3PbCI or MeHgCI were added and the incubation was continued for 15 min (platelets) or 30 rain (HL-60). Lipids were extracted and separated by t.l.c. Values are the mean of three to five experiments _+s.e.m.
Furthermore, HL-60 cells, differentiated with dimethyl sulphoxide to mature granulocytes, were incubated for 24 h in the presence of [t4C]-arachidonic acid. As shown for human blood platelets, the liberation of arachidonic acid is well inducible by Et3PbCI and could be totally inhibited in these cells by the inhibitors of phospholipase A 2, quinacrine and pBPB (Tab. 2).

Fatty acid liberation and lipid remodelling in HL-60 cells
In another set of experiments without prelabelling of the cells, the fatty acid composition of the cellular lipids were charactedsed by HPLC. When HL-60 cells were treated with lower concentrations of the xenobiotic for longer periods of time, a loss of fatty acids within the phospholipids was obvious, although no free arachidonic acid could be detected. All liberated fatty acids were re-esteritied continuously back into the phospholipids or triacylglycerols. The re-incorporation into triacylglycerols reveals information on nearly all of the fatty acids present in the cell. Above all, the non-cytotoxic concentration of Et3PbCI (0.5 txM) induced a substantial transfer from cellular phospholipids to triacylglycerols not only of arachidonic acid but also of linoleic, oleic and palmitic acid as shown in Fig. 1 Lipids were extracted and separated using bonded phase columns into a phospholipid-, free fatty acid-and tdacylglyceml-fraction. After hydrolysis of the phospholipid-and triacylglycerol-fractions, the fatty acid phenacyl esters were prepared and submitted to HPLC analysis. Ordinate: absorbance units full scale (AUFS) at 254 nm; abscissa: time (min). C16:0 palmilic acid; C18:0 stearic acid; C18:1 oleic add; C18:2 linoleic acid; C20:3 eicosatrienoic acid; C20:4 arachidonic acid; C17:0 internal standard margadc add.
Effects of pertussis toxin and its B-oligomer on Et3PbCI stimulated fiberation of arachidonic acid [14C]-arachidonic acid prelabelled HL-60 cells were incubated for 3 h at Partly Mediated by G-Proteins?
37C with 1 000 ng/ml pertussis toxin or an equivalent amount of its B-oligomer. During this period of time no alteration of incorporation and distribution of [4C]-arachidonic acid within the lipid classes could be detectexl (clam not shown). Fig. 2 shows that Et3PbCl-stimulation is higly sensitive to pertussis toxin.
Whereas the radioactivity within the neutral lipids is absolutely identical within the three samples, the arachidonic acid spot of the pertussis toxin treated cells contains only 35% of arachidonic acid as compar with B-oligomer or not pretreated cells.

DISCUSSION
The quantity of free unsaturated fatty acids within human blood platelets and HL-60 cells is very low 3 but these cell types respond to exogenous stimuli, e.g. thrombin, collagen, A 23187, or fMLP, with a rapid increase above all of free arachidonic acid. This is an important metabolic pathway and thus, these cells are provided with an efficient regulatory mechanism in controlling free fatty acid concentration. Involved in these processes are the fatty acid liberating enzymes, phospholipaseC and diacylglycerol lipase or phospholipase A, and the reacylating enzymes, arachidonoyl=CoA synthetase, lysophospholipid acyltransferase and diacylglycerol acyltransferase. 4 +B-Oligomer [14C]_arachidonic acid prelabelled and differentiated HL-60 cells were preincubated for 3 h with 000 ng/ml pertussis toxin (PT) or an equivalent amount of its B-oligomer before the cells were stimulated with 50 HM Et3PbCI (20 min). Lipids were extracted and separated by t.l.c. Shown is an autoradiography of that part of the OIM Et3Pb + t.l.c, plate containing the spots of the free arachidonic acid (fAA) and the neutral lipids (NLs).
fAA +PT + + + In various cell types, the thiol-blocking activity of heavy metals leads to an inhibition of the reacylation of frcc fatty acids into phospholipids. ]5,16 Et3PbCl inhibits the incorporation of exogenously added [14C]-arachidonic acid into cellular lipids 9 as MeHgC115, but at low concentrations (< 1 gl no inhibition of [14C]-arachidonic acid incorporation could be observed. 9 However, the liberation and subsequent redistribution of fatty acids still occurs at these low concentrations (Fig. 1). As shown by the use of various inhibitors it could be demonstrated that the heavy metal induced effects are dependent on fatty acid liberation from phospholipids. Inhibitors of phospholipasc A 2, such as quinacrine or pBPB7, could prevent this reaction in both cell types, human blood platelets and HL-60 cells, indicating a central role of this enzyme. In order to detect whether a direct stimulation of phospholipase A 2 by organic lead compounds occurs or any preceding components in the signal transduction mechanism is affected by Et3PbC1, the pertussis toxin sensitivity of this mechanism was tested. These experiments clearly show that only the holotoxin and not its membrane binding subunit, the B-oligomcr, is responsible for nearly 70% reduction of the metal-effect. It becomes more and more evident that phospholipase A 2 is coupled to membrane receptors via G-proteins 18 and these results point to a G-protein dependent mechanism for the stimulation of phospholipase A 2 by the heavy metals.
Phospholipases are important enzymes within regulatory processes inducible by external signals. Their products are second messengers with a multitude of functions, intra-as well as intercellular. Especially neutmphilic granulocytes are able to interact with various cell types, such as macrophages, mast cells, platelets, polymorphonuclear leukocytes and many others, e.g. via their products of the phospholipase A 2 cascade. 9. All three phospholipases shown in Fig. 3 can be affected by various stimulators from outside the cell 2-22 and may affect each other. The substances produced as the result of phospholipase A2 activity, i.e. the eicosanoids and the platelet activating factor, have been studied as potent mediators of immunological reactions. 19,23,24 The enhanced production of PAF as well as the rise in intracellular calcium concentration could be demonstrated in our laboratory. 2,2 The concentrations of organic lead used in the experiments come close to those in normal human brains as reported by Nielsen et al. (1978). They have found organic lead in quantifies up to 50 ng Pb/g wet weight and the lowest amount used to stimulate HL-60 cells was 70 ng Pb/ml. Such low concentrations of lead compounds are able to induce enzyme activities as shown here or reported elsewhere. 27 / PAFI E,C, Stimulation of Phospholipase A 2 (PLA2) by heavy metal compounds (G)) leads to potent cell stimulators such as free arachidonic acid (fAA) and subsequently to the eicosanoicls (pmstaglandins, leukotrienes, thmmboxanes) and/or via the lysophospholipids (IPL) to the platelet-actJvating factor (PAF). These cellular signalling mediators induce other phospholipases via membrane receptors such as phospholipase C (PLC) or phospholipase D (PLD). PLD mainly hydmlyses phosphatidylcholine to give phosphatidic add (PA). PLC hydrolysis of phosphatidylinositolbisphosphate yields diacylglyceml (DG) and inositol triphosphate (IP3). Whereas DG stimulates protein kinase C (PKC), the IP 3 regulates intmcellular calcium concentration. 1T = increased parameters after heavy metal treatment.
A large number of hnportant cellular processes, such as lipid mediator release or membrane functions, are dependent on signal transduction mechanisms that could be induced by physiological agents. On the other hand, more .and more evidence arises that xenobiotics, e.g. heavy metal compounds, may be involved in immunological reactions or in hypersensitisation of organisms to natural products 2,2s by lowering the threshold of cellular sensibility possibly via inducing the phospholipase A 2 to a higher level of basal activity. ACKNOWLEDGMENTS I am grateful to Helga Steegbom for superb technical assistance and Lindsay Yule for reviewing the manuscript before its submission.