factor - Reproduction

2 downloads 22 Views 650KB Size Report
M.J., Wykle, R.L., Dechatelet, L.R. & Waite, B.M.. (1982) Platelet activating factor. ... Morrison, J. (1977) An early pregnancy factor: tis¬ sues involved in its ...

Partial characterization of the PAF-induced soluble factors which mimic the activity of 'early pregnancy factor' F. M. Division

Clarke, C. Orozco, A.

V. Perkins and I. Cock

of Science and Technology, Griffith University, Nathan, Queensland 4111, Australia

Summary. Platelet-activating factor (PAF) stimulated mouse spleen cells to release soluble factors (termed S2 factors) which were capable of inducing increased rosette inhibition titres when applied to fresh mouse spleen cells in the rosette inhibition assay. In this ability the S2 factors mimic that of pregnancy serum, an action previously ascribed to 'early pregnancy factor'. The PAF-stimulated production of these S2 factors was not influenced by inhibitors of cyclooxygenase metabolism, but was completely inhibited by the lipoxygenase inhibitors, diethyl carbamazine and nordihydroguaiaretic acid. The S2 factors had a lipid-like character in that they were extractable in organic solvents. The calcium ionophore A23187 also stimulated the production of these factors which may well be products of the lipoxygenase pathway of arachidonic acid metabolism.

Keywords: early

pregnancy

eicosanoids; leukotrienes

factor; EPF; platelet activating factor; PAF;

rosette inhibition assay;

Introduction Orozco

et al. (1990) extended their studies (Orozco et ai, 1986) on the link between plateletactivating factor (PAF) and expression of activity of'early pregnancy factor' (EPF) by demonstrat¬ ing that a combination of PAF and sera from mice in oestrus could mimic the action of pregnancy sera by inducing an elevated rosette inhibition titre (RIT) when applied to mouse spleen cells in vitro. Neither component could induce this effect if applied to cells alone, the combination was required. PAF stimulated the spleen cells to release soluble factors (termed S2 factors) which were themselves capable of inducing elevated RITs if applied to fresh spleen cells even in the absence of serum from oestrous mice. However, the PAF-stimulated cells were rendered refractory to the action of these S2 factors and failed to display elevated RIT values unless oestrous mouse serum was also added, the serum components acting to reverse the refractory state. The discovery of these PAF-induced S2 factors which are directly active in the rosette inhibition assay is an important and germinal observation as their identification could contribute greatly to our understanding not only of the mechanism of action of PAF but also that of the so-called 'early pregnancy factor' in the

inhibition assay. The observed potency of the S2 factors coupled with the known primary action of PAF to stimulate arachidonic acid metabolism in a wide variety of cell types (Snyder, 1985; Kawaguchi & Yasuda, 1984; Chilton et ai, 1984; Macconi et ai, 1985) led us to investigate whether arachidonic acid metabolites generated via the cyclooxygenase and/or lipoxygenase pathways are involved in the generation of the S2 factors. rosette

Materials and Methods Synthetic PAF (l-0-octadecyl-2-acetyl-j«-glycero-3-phosphocholine) was obtained (Mannheim, FRG) and prepared and used as previously described (Orozco et al, 1990). PAF.

from

Boehringer

Aspirin. Aspirin (acetylsalicylic acid: Sigma Chemical Co., St Louis, MO, USA) was prepared freshly before use as

a

10 mM stock solution in PBS.

Indomethacin (Sigma) was prepared as a 25 mM stock solution in ethanol and of usually 0-5 mM in 50% ethanol in PBS before use.

freshly diluted to a working solution

Diethylcarbamazine (Sigma) was prepared freshly in PBS before use. Calcium ionophore, A23187 (Sigma) was prepared as a 5 mg/ml stock solution in ethanol and freshly diluted to the working solution (usually 1 mg/ml) in ethanol before use. Nordihydroguaiaretic acid (Sigma) was prepared as a 4 mM stock solution in ethanol and freshly diluted to the working solution of 0-4 mM in PBS before use. Preparation of S2 fractionfrom PAF-stimulated mouse spleen cells. Mouse spleen cells (15 IO6) were incubated in the presence of 5nM-PAF in 200 µ PBS. After 30 min at 37°C the cells were sedimented, washed twice with 2 ml Hanks balanced salt solution (HBSS: Gibco, Grand Island, NY, USA), resuspended in 200 µ PBS and incubated for a further 30 min at 37°C. At the end of this period the cells were sedimented and the supernatant fraction, designated S2, was collected. Then 10 µ of the S2 fraction and aliquants of 10-fold serial dilutions in PBS/BSA (0-2 mg/ml) were immediately taken for testing in the rosette inhibition assay as described below. Effect ofinhibitors on generation of S2 factors. Mouse spleen cells (15 106) were incubated at 37°C in 200 µ PBS in the presence or absence of the desired concentration (see text) of inhibitor (aspirin, indomethacin. diethylcarbama¬ zine or nordihydroguaiaretic acid). After 15 min PAF was added to a final concentration of 5 nM and the incubation

continued for a further 30 min, after which the cells were washed and reincubated in PBS to generate the S2 fraction as described above.

Cell incubations and the rosette inhibition assay. For the standard spleen cell incubation for testing in the rosette inhibition assay, 1 ml aliquants of spleen cells (15 106) were sedimented and the supernatant aspirated. The cells were resuspended in a final volume of 200 µ PBS containing the test samples which were added as 1 (ionophore), 5 (PAF, oestrous serum) or 10 µ (S2 fractions) aliquants. After incubation at 37°C for 30 min, the cells were sedi¬ mented, washed twice with 2 ml HBSS, then resuspended in 1-5 ml HBSS and dispensed into the ALS dilutions for the rosette inhibition assay as described by Orozco et al (1990). Extraction of the S2 fraction. S2 fractions prepared as described above were subjected to the two step extraction procedure of Clancy & Hughli (1983). The first extraction step separates total lipids from substances such as proteins and buffer salts, while the second step performed on the initial total lipid fraction separates polar from less polar lipids. In the first step the S2 fractions was combined with isopropanol and 5 M-formic acid in the ratio 10:0-5:003 (by vol.). After 5 min, 1-5 volumes of ether were added, resulting in the development of two phases. The upper organic phase consisting of an isopropanol/ether mixture was removed and combined with 10n-NH4OH in a ratio of 20: 0015 (v/v). This was evaporated to dryness under a stream of N2. For the second extraction step, the residue was dissolved in 0-5ml chloroform:methanol (1:1, v/v) and 0-22 ml 10mM-NH4OH was added, resulting in a two-phase separation. The upper aqueous methanol and the lower chloroform phases were collected and dried under a stream of nitrogen. The residues were resuspended in 20 µ (per 200 µ original S2 fraction) 65% methanol, 0-1% acetic acid, pH 5-6 (with NH3). Then 10 µ aliquants of 10-fold serial dilutions prepared in PBS/BSA (0-2 mg/ml) were taken for testing in the rosette inhibition assay. Other methods. Collection of sera and preparation of spleen cells were performed as described previously (Orozco et

al. 1986).

Results

lipoxygenase pathway inhibitors on generation of S 2 factors When PAF-stimulated mouse spleen cells were reincubated in PBS after PAF treatment, a soluble fraction (S2) was generated which was extremely potent in inducing an elevated RIT when applied to fresh mouse spleen cells see (Orozco et ai, 1990; and Fig. la). If the mouse spleen cells were treated with the cyclooxygenase inhibitors aspirin and indomethacin before PAF stimulation, then there was little or no effect on the ability of these cells to generate an active S2 fraction (Fig. lb). On the other hand if the cells were treated with the lipoxygenase pathway inhibitors, diethyl¬ carbamazine (Mathews & Murphy, 1982) or nordihydroguaiaretic acid (Salari et ai, 1984) before PAF stimulation, the ability to produce an active S2 fraction was completely abolished (Fig. la), indicating that the PAF-stimulated generation of the S2 factors was dependent upon the activity of the lipoxygenase pathway of arachidonic acid metabolism. Effect of cyclooxygenase and

30

S2 26

22

18

14

E IO DC

30

S2

+

DEC

S2

+

NDGA

(b)

26

22-

18

14

10· 10

10"

10

10

10

10

Dose

10°

10'

(µ )

Dose-responses in the rosette inhibition assay of PAF-induced S2 fractions prepared spleen cells in the absence (S2) or presence of diethylcarbamazine (DEC 10µg/ ml), nordihydroguaiaretic acid (NDGA: 20 µ ), aspirin (0-25 mM) and indomethacin (Indo, 10 µ ). Values are means ± s.d. of 3 independent experiments with different spleen cell preparations. Fig.

1.

from

mouse

Lipid extraction of S2 factors When the S2 fraction generated from PAF-stimulated cells was extracted by the two-step pro¬ cedure of Clancy & Hughli (1983), essentially all of the S2 factor activity was extracted with the total lipid fraction into the isopropanol/ether phase (Fig. 2). Subsequent fractionation of this total lipid extract into polar and less polar fractions by the second step of the procedure resulted in a partitioning of S2 factor activity between both phases. While the bulk of the activity partitioned with the more polar lipids into the aqueous methanol phase, a significant activity was still associated with the less polar lipid fraction in the chloroform phase (Fig. 2). Calcium

ionophore stimulation of production of S2 factors

As shown in Fig. 3, treatment of mouse spleen cells with the calcium ionophore A23187 resulted in the production of an S2 fraction capable of inducing elevated RIT values when applied to fresh mouse spleen cells. Comparison of the S2 fractions generated by the action of A23187 (5 µg/ml) and by PAF (5 µ) indicated that the ionophore-induced S2 fractions were a little less potent than those induced by PAF action. In this series of experiments the ionophore induced S2 fraction was active down to a dose equivalent to 10~3, compared to 10~5 µ for the PAF-induced S2 fraction.

14

3

10

Dose

io-1

io-

(µ )

Fig. 2. Dose-responses in the rosette inhibition assay of fractions obtained by the two-step extraction of PAF-induced S2 fractions. Values are mean of 2 experiments; ip. ether isopropanol ether phase; aq. = aqueous phase; aq. met. aqueous methanol phase; chlor. chloroform phase. =

=

=

30

26-

22

14

10

10

10"" Dose

io

10

10

(µ )

Fig. 3. Dose-responses in the rosette inhibition assay of S2 fractions generated from mouse spleen cells by the action of PAF (5 µ) and the calcium ionophore A23187 (5 µg/ml). Values are mean ± s.d. from 4 different experiments in which actions of PAF and A23187 were compared on the same spleen cell preparations.

Ionophore-stimulated cells fail to display an increased RIT While the ionophore-treated cells were capable of producing S2 factors able to induce elevated RIT values when applied to fresh cells, the ionophore-treated cells themselves were refractory to the action of the S2 factors and this refractory state could not be reversed by addition of serum from oestrous mice as was previously found to be the case with PAF-treated cells (Orozco et ai, 1990). The mean RIT value determined on spleen cells incubated with ionophore (5-40 pg/ml) was 12 (22

determinations), the same as that determined on cells incubated with ionophore in the (0-5 µ ) of oestrous mouse serum (10 determinations).

presence

Discussion soluble factors (termed S2 factors) which are fresh spleen cells. In this action they mimic that of pregnancy serum. PAF is a potent stimulus activating a wide variety of cells to produce a diverse range of mediators (Snyder, 1985). The primary mode of action of PAF, however, on almost all cell types appears to be via stimulation of phosphatidylinositol turnover leading to activation of phospholipase C and mobilization of Ca2+ ions (Shukla & Hanahan, 1983; Mauco et ai, 1983; Kawaguchi & Yasuda, 1984). A common consequence is the activation of phospholipase A2 and stimulation of arachidonic acid metabolism which may lead to eicosanoid production via the cyclooxygenase and/or lipoxygenase pathways (Shaw et ai, 1981; Chilton et ai, 1982; Roubin et ai, 1983; Kawaguchi & Yasuda, 1984). The present studies explored the possible involvement of these pathways in S2 factor production. The results of the inhibitor studies strongly implicated the lipoxygenase pathway. Specific inhibitors of the cyclooxygenase pathway, aspirin and indometha¬ cin, had little or no influence on the PAF-stimulated production of S2 factors. The lipoxygenase pathway inhibitors, diethylcarbamazine and nordihydroguaiaretic acid, however, completely blocked the production of these factors, indicating that activation of this pathway of arachidonic acid metabolism was necessary for the generation of these factors. The extraction studies per¬ formed on the S2 fraction indicated that the S2 factors were not only dependent on the lipoxy¬ genase pathway but may well be products ofthat pathway. Their extraction into the organic phase indicating a lipid-like character and their extreme potency described in this and the previous paper would be consistent with this interpretation. Because diethylcarbamazine and nordihydroguaiare¬ tic acid inhibit the initial steps in the lipoxygenase pathway, the conversion of arachidonic acid to 5hydroperoxyeicosatetraenoic acid and subsequently to leukotriene A4 (Mathews & Murphy, 1982; Salari et ai, 1984), a range of lipoxygenase products, including LTB4, 5-hydroxyeicosatetraenoic acid (5-HETE) and the peptidoleukotrienes, LTC4, LTD4 and LTE4 for example, must be con¬ sidered as possible candidates. As the bulk of the S2 factor activity extracts into the polar aqueous methanol phase during the second step of the Clancy & Hughli (1983) procedure, it may be that activity expression is associated with the more polar leukotrienes such as the peptidoleukotrienes which are known (Clancy & Hughli, 1983) to extract preferentially into this phase. LTB4, and particularly 5-HETE, preferentially extract into the chloroform phase, but this phase still displays some activity. Given that the partitioning of these metabolites into one phase or the other is not absolute (Clancy & Hughli, 1983) and that they are all molecules with potentially potent biological activities, it is clear further characterization studies are necessary. The basic conclusions we draw at the present time are that the generation of the S2 factors is dependent on the lipoxygenase pathway, they are lipid in character, and so may possibly be lipoxygenase products. These conclusions are supported by the fact that the calcium ionophore A23187 can also stimu¬ late the production of S2 factors. This ionophore, by raising intracellular Ca2+ levels, can activate phospholipase A2, stimulating arachidonic acid metabolism and causing many cell types to release eicosanoids, particularly lipoxygenase products. For example, it stimulates human polymorphonuclear leucocytes to release principally LTB4, LTC4 and 5-HETE (Borgeat & Samuelsson, 1979; Hansson et ai, 1981; Shak & Goldstein, 1984). The ionophore stimulation of S2 factor production is also inhibited by diethylcarbamazine and nordihydroguaiaretic acid, indicating lipoxygenase involvement, and the activity is extracted into the organic phase. Ionophore stimulation of human polymorphonuclear leucocytes also produces lipid-extractable factors capable of inducing elevated RIT values (F. M. Clarke, unpublished). Taken together these observations support the conclusions above as to the character and mode of production of the S2 factors. PAF can stimulate mouse spleen cells to produce capable of inducing elevated RIT values if applied

to

The identity of the cells producing these factors in the spleen cell preparation is not known. There is no reason to suppose, for example, that the cell population stimulated to produce the S2 factors is that which responds to them. Indeed, it is quite probable that they are different popu¬ lations. For example, the rosette-forming cells are T-lymphocytes (Morton et ai, 1975; Rolfe et ai, 1984), in fact a select subpopulation of T-lymphocytes, because under our assay conditions only 3% of the total spleen cell population are involved in rosette formation. The assay therefore moni¬ tors the ultimate response of only a small subpopulation of T-lymphocytes. While there is evidence that T-lymphocytes may respond to and metabolize lipoxygenase products (Rola-Pleszezynski et ai, 1982; Payan et ai, 1984; Goetzl et ai, 1988; Ödländer et ai, 1988), there is little or no evidence that they can be stimulated to produce them (Poubelle et ai, 1987; Goodwin & Behrens, 1988). Other leucocytes, monocytes and macrophages are known to produce eicosanoids on stimulation (Snyder, 1985) and so may be the source of S2 factors in the spleen cell preparation. It is quite possible that while PAF may stimulate one cell population to produce S2 factors, it may, by direct or indirect means, render another cell population refractory to their action. Of course, if S2 factors are added directly to fresh spleen cells, no PAF-induced block would exist in the responding cell population and so they could respond to the S2 factors with the ultimate expression of an elevated RIT. The present studies have only begun to unravel the complex networks involved; however, by defining the mode of action of PAF and the oestrous mouse serum components, and by providing insights into the identity of some of the effector molecules, future studies aimed at elucidating mechanistic detail should be facilitated. The present studies also have other implications concerning the appreciation of both the 'early pregnancy factor' phenomenon and the physiological significance of production of embryo-derived PAF. We have shown in these studies that (i) a combination of PAF and serum from oestrous mice, and (ii) PAF or ionophore-induced S2 factors can both mimic the action of pregnancy sera by inducing the expression of an elevated rosette inhibition titre. This ability or activity of pregnancy sera has long been ascribed to the presence in these sera of a so-called 'early pregnancy factor' (Morton et ai, 1976, 1977, 1987). The choice of this term to describe an activity in a complex biological assay, while understandable in an historical context, has been a little unfortunate, as it implies the existence of a single, unique factor in sera responsible for this activity expression. This need not be so. As we have pointed out previously, characterization studies (Clarke et ai, 1980; Clarke & Wilson, 1985) indicate that a complex system of components is involved and so we have emphasized (Clarke & Wilson, 1985) that the basis for this activity expression may well be multifactorial. The present results serve again to emphasize this point, by demonstrating that comparable activity expression can be achieved by applying either active factors (the S2 factors), or an appro¬ priate stimulatory/regulatory system (PAF plus serum from oestrous mice). It is quite possible that pregnancy sera could contain either active factors or an appropriate stimulatory/regulatory system or both. The presence of one or both of them would allow the sera to express 'EPF' activity. An appreciation of these possibilities and therefore the likely nature of the serum 'EPF' system should help to explain and resolve the apparent complexities which currently surround the characterization of this phenomenon. Using PAF antagonists, Spinks & O'Neill (1988) provided evidence that embryo-derived PAF activity was necessary for the establishment of pregnancy in the mouse but suggested that the mechanism by which PAF influences implantation may not be solely mediated via platelet acti¬ vation. The present results suggest that one role of PAF may be that of stimulating eicosanoid production, with these potent vasoactive agents co-operating with PAF to promote the proinflammatory reactions which precede and accompany implantation in this and other species. Malathy et ai (1986) have noted a surge in leukotriene production in the rat uterus during the periimplantation period. In addition to their proinflammatory actions, leukotrienes are also becoming recognized as potent regulators of T-lymphocyte function (Goetz et ai, 1988; Goodwin & Behrens, 1988). LTB4 for example, is known to induce proliferation of suppressor T-cells at the same time inhibiting the generation of T-helper cells (Rola-Pleszezynski et ai, 1982; Payan et ai, 1984). The

of embryo-derived PAF or other leukotriene-stimulating agents at the feto-maternal interface may therefore contribute to regulating the maternal immune response to the fetal allograft and thus the successful estalishment of pregnancy.

production

This work was supported by grants from the National Health and Medical Research Council of Australia.

References P. & Samuelsson, B. (1979) Arachidonic acid metabolism in polymorphonuclear leukocytes: effect of ionphore A23187. Proc. natn. Acad. Sci. USA 76, 2148-2151.

Borgeat,

Chilton, F.H., O'Flaherty, J.T., Walsh, CE., Thomas, M.J., Wykle, R.L., Dechatelet, L.R. & Waite, B.M.

(1982) Platelet activating factor. Stimulation of the lipoxygenase pathway in polymorphonuclear leuko¬ cytes by I-0-alkyl-2-0-acetyl-sn-glycero-3-phosphocholine. J. biol. Chem. 257, 5402-5409. Chilton, F.H., Ellis, J.M., Olson, S.C. & Wykle, R.L. 1 -0-alkyl-2-arachidonyl-sn-glycero-3-phos( 1984) phocholine. A common source of platelet activating factor and arachidonate in human polymorphonu¬ clear leukocytes. J. biol. Chem. 259, 12014-12019. Clancy, R.M. & Hughli, T.E. (1983) The extraction of leukotrienes (LTC4, LTD4 and LTE4) from tissue fluids. The metabolism of these mediators during IgE-dependent hypersensitivity reactions in lung.

Analyt. Biochem. 133, 30-39. Clarke, F.M. & Wilson, S. (1985) In search of EPF. In Early Pregnancy Factors, pp. 165-177. Eds F. Ellendorff & E. Koch. Perinatology Press, New York.

Clarke, F.M., Morton, H., Rolfe, B.E. & Clunie, G.J.A. (1980) Partial characterisation of early pregnancy factor in sheep. J. Reprod. Immunol. 2, 151-162. Goetzl, E.J., Sherman, J.W., Ratnoff. W.D., Harvey, J.P., Eriksson, E., Seaman, W.E., Baud, L. & Koo, CH. (1988) Receptor-specific mechanisms for the

responses of human leukocytes to leukotrienes. Ann. N. Y. Acad. Sci. 524, 345-355. Goodwin, J.S. & Behrens, T. (1988) Role of lipoxygenase metabolites of arachidonic acid in cell activation. Ann. N. Y. Acad. Sci. 524, 201-207. Hansson, (... Lindgren, J.A., Dahlen, S.E., Hedquist, P. & Samuelsson, B. (1981) Identification and biological activity of novel -oxidized metabolites of leukotriene B4 from human leukocytes. FEBS Letters 130, 107-112. Kawaguchi, H. & Yasuda, H. (1984) Platelet activating factor stimulates phospholipase in quiescent Swiss mouse 3T3 fibroblasts. FEBS Letters 176, 93-96.

Macconi, D., Morzenti, G., Livio, M., Morelli, C, Cassina, G. & Remuzzi, G. (1985) Acetyl Glycerylphosphorylcholine aggregates human platelets two distinct pathways, both dependent on arachidonic acid metabolism. Lab. Invest. 52, 159-168. Malathy, P.V., Cheng, H.C. & Dey, S.K. (1986) Produc¬ tion of leukotrienes and prostaglandins in rat uterus

through

during peri-implantation period. Prostaglandins 32, 605-612.

W.R. & Murphy, R.C. (1982) Inhibition of leukotriene biosynthesis in mastocytoma cells by diethyl carbamazine. Biochem. Pharmacol. 31, 2129-2123. Mauco, G., Chap, H. & Douste-Blazy, L. (1983) Platelet activating factor (PAF-acether) promotes an early degradation of phosphatidylinositol-4,5-biphosphate in rabbit platelets. FEBS Letters 153, 361-365. Morton, H., Hegh, V. & Clunie, G.J.A. (1975) Antilym¬ phocyte serum: Studies of the rosette inhibiting anti¬ body in rabbit anti-mouse ALS. Aust. J. exp. Biol. Med. Sci. 53, 373-380. Morton, H., Hegh, V. & Clunie, G.J.A. (1976) Studies of the rosette inhibition test in mice: evidence of immunosuppresion? Proc. R. Soc. 193, 413-419. Morton, H., Rolfe, B.E., Clunie, G.J.A., Anderson, M.J. & Morrison, J. (1977) An early pregnancy factor: tis¬ sues involved in its production in the mouse. J. Reprod. Immunol. 2, 73-82. Morton, H., Rolfe, . & Cavanagh, A.C. (1987) Ovum factor and early pregnancy factor. Curr. Topics Devi Biol. 23, 73-92. Ödländer, ., Jakobsson, P.J., Rosen, A. & Claesson, H.E. (1988) Human and lymphocytes convert leukotriene A4 into leukotriene B4. Biochem. Biophys. Res. Commun. 153, 203-208. Orozco, C, Perkins, T. & Clarke, F.M. (1986) Platelet activating factor induces the expression of early preg¬ nancy factor activity in female mice. J. Reprod. Fert. 78, 549-555. Orozco, C, Cock, I., Perkins, A.V. & Clarke, F.M. (1990) Platelet-activating factor and serum compo¬ nents from oestrous mice co-operate to mimic the activity of "early pregnancy factor" in the rosette inhibition assay. J. Reprod. Fert. 88, 447-457. Payan, D.G., Missirian-Bastian, A. & Goetzl, E.J. (1984) Human T-lymphocyte subset specificity of the regu¬ latory effects of leukotriene B4. Proc. natn. Acad. Sci. USA 81, 3501-3505. Poubelle, P.E., Borgeat, P. & Rola-Pleszczynski, M. (1987) Assessment of leukotriene B4 synthesis in human lymphocytes by using high performance liquid chromatography and radioimmunoassay methods. J. Immunol 139, 1273-1277. Rola-Pleszeczynski, M., Borgeat, P. & Sirois, P. (1982) Leukotriene B4 induces human suppressor lympho¬ cytes. Biochem. Biophys. Res. Commun. 106, 1531-1537.

Mathews,

Rolfe, B.E., Cavanagh, A.C., Förde, C, Bastin, F., Chen, C & Morton, H. (1984) Modified rosette inhibition test with mouse lymphocytes for detection of early pregnancy factor in human pregnancy Immunol. Methods 70, 1-11.

serum.

J.

Roubin, R., Tence, M., Mencia-Hverta, J.M., Arnoux, ., Ninio, E. & Benveniste, J. (1983) A chemically

defined monokine: macrophage-derived platelet acti¬ vating factor (PAF-acether). In Lymphokines, pp. 249-276. Ed. E. Pike. Academic Press, New York. Salari, H., Braquet, P. & Borgeat, P. (1984) Comparative effects of indomethacin, acetylenic acids, 15-HETE, nordihydroguaiaretic acid and BW755C on the metabolism of arachidonic acid in human leucocytes and platelets. Prostaglandins, Leukotrienes Med. 13, 53-60. Shak, S. & Goldstein, I.M. (1984) -oxidation is the major pathway for the catabolism of leukotriene B4 in human polymorphonuclear leukocytes. J. biol. Chem.259, 10181-10187. Shaw, J.O., Klusick, S.J. & Hanahan, D.J. (1981) Activation of rabbit platelet phospholipase and thromboxane synthesis by 1-0-hexadecyl/octadecyl-

2-acetyl-sn-glyceryl-3-phosphorylcholine activating factor). Biochem. Biophys.

(platelet

Ada

663,

222-228. Shukla, S.D. & Hanahan, D.J. (1983) An early transient decrease in phosphatidylinositol 4,5-bisphosphate upon stimulation of rabbit platelets with acetylgly-

cerylether phosphorylcholine (platelet activating factor). Archs Biochem. Biophys. 227, 626-629. Spinks, N.R. & O'Neill, C (1988) Antagonists of embryo-derived platelet-activating factor prevent implantation of mouse embryos. /. Reprod. Fert. 84, 89-98.

Snyder, F. (1985) Chemical and biochemical aspects of platelet activating factor: a novel class of acetylated ether-linked choline phospholipids. Med. Research Rev. 5, 107-140.

Received 27

April

1989