[Back to Number 1 ToC] [Back to Journal Contents] [Back to Biokhimiya Home page]

REVIEW: The Bioregulatory Role of Platelet-Activating Factor in Intracellular Processes and Cell--Cell Interactions

V. I. Kulikov* and G. I. Muzya

Research and Development Center for Medical Biotechnology, Ministry of Health of the Russian Federation, ul. Shchukinskaya 6, Moscow, 123182 Russia; fax: (095) 190-0100

* To whom correspondence should be addressed.

Received May 28, 1997
The role of platelet-activating factor (PAF, a phospholipid compound) in regulation of cell functions and cell--cell interactions is reviewed. The biological effects of PAF on platelets, neutrophils, basophils, eosinophiles, lymphocytes, and endothelial cells are described. Mechanisms of cell activation by PAF are discussed. Interactions of PAF with other biological regulators (prostaglandins, leukotrienes, NO, tumor necrosis factor, and interleukins) are considered.
KEY WORDS: platelet-activating factor, cells

Abbreviations: 5-HETE) 5-hydroxyeicosatetraenoic acid; IL-1, IL-2, IL-5, IL-6) interleukins 1, 2, 5, and 6, respectively; LTB4) leukotriene B4; LDL) low-density lipoproteins; PGI2) prostacyclin; TxA2) thromboxane A2; TNF) tumor necrosis factor; PAF) platelet-activating factor (1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine).

Determination of the chemical structure of the phospholipid compound PAF (platelet-activating factor) in 1979 resulted in the discovery of new biological mechanisms of cell regulation mediated by this unique phospholipid.

The versatility of PAF-mediated mechanisms of cell regulation is primarily related to the fact that PAF occurs in a broad variety of living organisms, such as bacteria [1], protozoa [2], fungi [3], plants [4], invertebrates [5], and lower vertebrates [6] that occupy various positions on the evolutionary scale. The versatility of PAF as a bioregulator is also related to its property to regulate diverse vital systems of the body of mammals (cardiovascular, immune, reproductive, and central nervous systems). Finally, at the cellular level PAF is a key bioregulator of cell--cell interactions and is part of a complex of finely tuned cell bioregulators, including prostaglandins, leukotrienes, tumor necrosis factor (TNF), interleukins, histamine, and serotonin [7]. This review is concern with the role of PAF in intracellular processes and cell--cell interactions as well as the relationships of PAF with other cellular bioregulators.


PAF is one of the most potent inducers of platelet aggregation. At a concentration of 10-10-10-7 M, PAF can activate rabbit, cat, guinea pig, dog, sheep, horse, and human platelets [8]. PAF-induced platelet activation is accompanied by: changes in the cell shape; release of vasoactive amines, aggregation factor 4, and beta-thromboglobulin; stimulation of thromboxane A2 (TxA2) synthesis; and platelet aggregation [8].

Processes of platelet activation have species-specific features. Differences mainly concern the sensitivity of platelets to PAF. Rabbit platelets display high sensitivity to PAF: 50% aggregation occurs at 1.4·10-10 M PAF [9]. Human platelets display a lower sensitivity to PAF; their reversible aggregation occurs within the range 10-9-10-8 M PAF, and irreversible aggregation occurs at 10-8-10-6 M [10]. In rabbit platelets, PAF is a strong inducer of aggregation [9], secretion of vasoactive amines [8], and TxA2 synthesis [11]. In relation to human platelets PAF is a strong inducer of aggregation and release of vasoactive amines but a relatively weak inducer of TxA2 synthesis [10]. The effects of PAF on platelet activation are not mediated by TxA2, the level of which remains low during aggregation and increases only by the end of the process [12]. PAF bound to platelets is then deacetylated by platelet acetylhydrolase [13], and the lysoPAF generated in this process is further acylated to 1-O-alkyl-2-acyl-sn-glycero-3-phosphocholine [8].

When applied to isolated neutrophils at a concentration of 10-9-10-6 M, PAF stimulates their chemotaxis, degranulation, generation of superoxide radicals, and aggregation [14]. PAF-induced neutrophil degranulation releases lysozyme and beta-glucuronidase associated with intracellular granules; however, PAF does not cause the release of cytoplasmic lactate dehydrogenase [14]. PAF also stimulates neutrophils to release other inflammation mediators such as eosinophil chemotaxis factor, leukotriene B4 (LTB4), and 5-hydroxyeicosatetraenoic acid (5-HETE) [7].

Generation of superoxide radicals by neutrophils is an important factor of the body defense system. PAF and its cellular analogs (1-acyl-PAF and 1-plasmalogen-PAF) were shown to stimulate the release of superoxide radicals into the extracellular medium [7, 15]. PAF and its cellular analogs act as priming (facilitating) agents in these processes [16].

PAF is involved in immediate hypersensitivity responses in humans and serves as a mediator of allergic reactions. Even before the determination of the chemical structure of PAF, experiments showed that interaction of antigens with basophils releases a mediator that stimulates platelet aggregation and histamine secretion from basophils; this factor was identified later as PAF [7]. When added to isolated basophils at 10-9-10-6 M, PAF releases histamine from histamine-containing granules [17]; cellular analogs of PAF also display a histamine-releasing activity [18]. Eosinophiles, which are also targets of PAF, play an important role in the pathogenesis of certain inflammatory processes and allergic diseases [7]. PAF stimulates eosinophil chemotaxis, as well as the formation of superoxide radicals, TxA2, and prostaglandins of the E group; it also releases certain enzymes and specific proteins (peroxidase, eosinophil cationic protein, and eosinophil major basic protein) from their specific granules [19, 20]. The release of these proteins and mediators probably damages endothelial cells in asthma and other lung diseases [20]. PAF affects the functional activity of lymphocytes; thus, it is involved in the complex mechanisms of immune reactions [21].

PAF at concentrations 10-12-10-10 M inhibits the proliferation of isolated human peripheral lymphocytes stimulated by phytohemagglutinin HA16 and concanavalin A [21]. This effect is not related to PAF-induced cytotoxicity because the cell survival rates do not change [22]. PAF also suppresses the mitogen-stimulated production of IL-2 by human lymphocytes [21]. Inhibition of lymphocyte proliferation and IL-2 production results from activation of suppressor cells by PAF. This induction of suppressor cells is accompanied by an increase in the number of CD8+ T-cells and a decrease in the population of CD4+ T-cells [9]. After treatment with PAF, the number of IL-2 receptors on the cell surface decreases [23].

PAF at a concentration of 10-13-10-7 M stimulates the secretion of immunoglobulins E and A by human lymphoblastoma cells; this effect is not accompanied by an increase in their proliferation rates or cell counts [24]. The mechanism whereby PAF stimulates immunoglobulin secretion remains unclear; however, it was shown that in lymphoblastoma B cells PAF activates the ribosomal S6 peptide kinase and microtubule-associated protein-2-kinase [25].

The activity of natural killer cells toward erythromyeloleukemia K-562 cells is considerably increased in the presence of PAF [21]. The activity of natural killer cells, as well as cytotoxic properties of monocytes and macrophages are know to be determined, at least partly, by their liberation of TNF [9]. PAF was shown to stimulate the release of TNF from monocytes and macrophages [7]. The antitumor activity of several synthetic alkoxylipids may be determined to a certain extent by their structural similarity to PAF [26], which explains their ability to stimulate the cytotoxic properties of certain immune cells [27].

Endothelial cells are one of the most important targets of PAF. Within the concentration range of 10-10-10-8 M, PAF induces changes in the shape of endothelial cells, including cell retraction, the disappearance of reciprocal contacts, and the appearance of elongated cells [7]. In PAF-stimulated endothelial cells, the concentration of Ca2+ and the rate of phosphoinositide turnover increase, the activity of adenylate cyclase decreases, and protein kinase C is activated [7].

Mobilization of Ca2+ induces changes in ion transport, resulting in hyperpolarization of endothelial cells through activation of Ca2+-dependent potassium channels [7]. In endothelial cells, PAF alters beta-integrin receptors, thereby decreasing the cell adhesion to vitronectin and fibrinogen [7]. These biochemical changes increase the permeability of vascular walls [28], increase platelet and neutrophil adhesion to vascular walls, and facilitate the development of vascular pathology [7].


Activation of various cells by PAF is primarily mediated by their high-affinity PAF receptors. These receptors were found on human [29] and rabbit [30] platelets, human [31] and rabbit [30] polymorphonuclear leukocytes, guinea pig eosinophiles [32] and peritoneal macrophages [30], human lung cells [33], guinea pig smooth muscle cells [34], plasma membranes of rat hepatocytes [35], and human keratinocytes [36].

Several lines of evidence support the existence of PAF-specific receptors on platelets: a) high-affinity PAF binding sites were found on intact platelets and platelet plasma membranes; b) the affinity of various PAF analogs for PAF binding sites depends directly on their ability to induce platelet aggregation; and c) PAF antagonists can inhibit the specific binding of PAF to platelets and PAF-induced platelet activation with parallel dose-response curves in both cases.

Rabbit platelets contain 150-400 PAF receptors per cell with a Kdof 0.5 nM [8]. Human platelets contain a single class of PAF receptors with a Kd of 0.3-3 nM whose number varies from 85 to 1400 per cell [8].

PAF is one of the strongest known agonists of platelet aggregation. For example, rabbit platelets are activated at a PAF concentration of 1 nM [8], whereas human platelets are activated within the range of 1 to 10 nM [8]. The affinity of PAF receptors (Kd of 1-3 nM) for PAF lies within the same PAF concentration range (1-10 nM) that causes changes in platelet shape and primary monophase aggregation. PAF binding by intact platelets does not depend on the presence of extracellular Ca2+ [29], whereas PAF binding to platelet membrane receptors increases 8-10-fold in the presence of Mg2+, Ca2+, or Mn2+ [8]. In contrast, Zn2+ inhibits the specific binding of PAF [34] and PAF-induced platelet activation [37]. Na+ and GTP specifically decrease the affinity of PAF receptors for PAF. It was suggested that PAF receptors can exist in a high-affinity state in the presence of Mg2+ and a low-affinity state in the presence of Na+ [8].

A known phenomenon is a decrease in the sensitivity of PAF receptors, whereby platelets subjected to short-term (2-3 min) treatment with low PAF concentrations become unresponsive to its higher concentrations [8]. This phenomenon is not prevented by treatment of platelets with 0.5% BSA, which removes the surface-bound PAF [8]. The decrease in platelet sensitivity to PAF can be caused by a decrease in the affinity or the number of binding sites [38].

Many lipid compounds are agonists of PAF receptors. The platelet-activating ability of PAF receptor agonists depends on their structure and the structural similarity to PAF. Studies of the biological activity of PAF analogs containing no oxygen atoms at sn-2 (2-n-propyl-PAF, 2-isobutyl-PAF, and 2-isopropyl-PAF) showed that PAF-induced platelet activation does not require the transfer of an acetyl group to any receptor [39]. However, an increase in the length of the aliphatic residue (more than three carbon atoms) sharply decreases the activity of PAF analogs [9]. The absence of an oxygen atom at sn-1 or sn-2 causes a more than 100-fold decrease in the platelet-activating ability of the compound [40].

Changes in the hydrophobicity of the alkyl chain of PAF caused by introduction of a phenylene residue results in no considerable change in the activity of PAF analogs provided that the spatial configuration of the modified alkyl chain in relation to the glycerol moiety of the molecule does not change considerably [40]. The structure of the polar part of the molecule strongly determines the properties of PAF analogs and their interactions with PAF receptors. Introduction of three to five methylene groups between atoms P and N of choline or substitution of ethyl groups for methyl groups in the trimethylammonium part of choline considerably changes the platelet-activating capacity of PAF analogs [41, 42]. However, substitution of N-methylpiperidine or N-methylpyrrolidine for the trimethylammonium group did not decrease the activity of the analogs but rather increased it more than tenfold in comparison to PAF [42].

Studies of properties of PAF resulted in the discovery of a range of compounds that act as PAF antagonists and block its effects at the receptor level. The large family of PAF antagonists can be divided into classes of nonlipid compounds and those whose structure contains certain lipid fragments. The group of nonlipid PAF antagonists includes many natural and synthetic compounds (kadsurenone, BN 52021, L-652,731, and L-659,989) containing a tetrahydrofuran ring [43], triazolobenzodiazepines (alprazolam, triazolam, and WEB 2086) [44], derivatives of pyrrolo-[1,2-c]thiazole (RP 48740 and RP 52770) [45], 1,4-dihydropyridines [46], and other compounds (reviewed in [44]). CV 3988 was among the first synthetic antagonists of lipid nature [47]. This compound inhibited [3H]PAF binding to platelets and their secretion and aggregation at a concentration of 0.1-0.15 µM [47]. CV 3988 displays a 1000-fold lower affinity for PAF receptors in comparison to PAF [47].

Studies of the biological activity of synthetic and cellular analogs of PAF showed that various modifications of the bond type at sn-1, as well as modification of residues at sn-2 and sn-3 caused the appearance of PAF-antagonistic properties in lipid analogs [48, 49]. The available data on the properties of PAF agonists and antagonists resulted in a concept of the molecular structure of PAF receptors [50]. Current data suggest that the long-chain PAF residue is deeply inserted in the hydrophobic region, whereas the acetyl residue is involved in "fixation" of PAF to receptors, which results in optimal binding of the choline residue to the receptor. After binding, PAF or its agonist affect the receptor conformation by transferring electrons from the oxygen atom at C1 to the acceptor in the PAF receptor [50]. A model of PAF receptor was proposed [50] to explain the mechanism of action of PAF antagonists containing a tetrahydrofuran ring and structurally not related to PAF. This model suggests that PAF receptor is a bipolar cylinder with two strongly electronegative surfaces positioned at an angle of 180° to each other and 10-12 Å apart [50].

A great diversity of chemical compounds displaying PAF-antagonistic properties can be related with the existence of a hydrophobic region (pocket) in the structure of PAF receptors. Most PAF antagonists sharing no structural homology with PAF can probably have a hydrophobic residue that fits the hydrophobic pocket of the PAF receptor, and certain groups of antagonists can interact with molecular components of PAF receptors. Rabbit platelet PAF receptors were found to contain a protein of 350 or 220 kD, which can be an association of several polypeptides of 52 kD [44]. The presence of protein(s) in PAF receptors are consistent with data indicating that thermal treatment or trypsin treatment of platelet membranes inactivate PAF receptors [44].

Binding of PAF to PAF receptors causes a cascade of biochemical events resulting in signal transduction in the cell and its activation. One of the earliest biochemical responses to PAF is the stimulation of polyphosphoinositide turnover. As soon as 5 sec after PAF application to platelets, membrane-associated phospholipase C degrades phosphatidylinositol-4,5-bisphosphate to inositol-1,4,5-trisphosphate [51]. This reaction generates two second messengers, inositol-1,4,5-trisphosphate and diglyceride, that increase the intracellular concentration of Ca2+ and activate platelet protein kinase C, respectively [52]. Activation of phospholipase C by PAF is mediated by G-proteins [52]. After PAF-receptor binding, GTP binds to the alpha-subunit of the G-protein oligomeric form and causes its dissociation from beta- and gamma-subunits; the dissociated alpha-subunit then interacts with phospholipase C and activates the enzyme [52]. Further, PAF activates GTPase of the alpha-subunit [53], which promotes the reassociation of alpha-, beta-, and gamma-subunits into the initial state of the G-protein. PAF antagonists (CV 6209, BN 52021, and SRI 63441) inhibit both the binding of [3H]PAF to platelets and the formation of inositol trisphosphate in platelets [52]. Activators of adenylate cyclase (forskolin and prostacyclin (PGI2)) that increase the level of cAMP, inhibit the PAF-stimulated breakdown of phosphatidylinositol-4,5-bisphosphate and an increase in intracellular Ca2+ level, whereas PAF inhibits adenylate cyclase in platelets [52].

Activation of protein kinase C by phorbol myristate acetate inhibits the polyphosphoinositide turnover, probably by modulating phospholipase C [52]. These mechanisms of transmembrane signaling from PAF were found in many cell types (neutrophils, endothelial cells, macrophages, hepatocytes, etc.) and can be, to a certain extent, universal mechanisms of transmembrane transduction of signals from PAF [44, 52].


Biological functions of PAF require complex interactions with virtually all cellular bioregulators. The various types of these interactions include: a) direct stimulation of synthesis or liberation of cellular bioregulators by PAF; b) potentiation by PAF of liberation of biological regulators induced by various stimuli; c) autocrine stimulation of synthesis of cellular bioregulators; d) synergistic effects of PAF and other bioregulators; e) priming effects of PAF and other bioregulators; and f) stimulation or inhibition of receptor expression in cells.

Various components of blood plasma strongly affect the intercellular biological effects of PAF. Plasma lipoproteins, albumin [54], phospholipid-transporting proteins and certain platelet-derived microparticles [55] can transport PAF from cell to cell. The activity of PAF in the intracellular [13] and intercellular media are regulated by acetyl hydrolase, a special enzyme whose major fraction in blood plasma is associated with low-density lipoproteins (LDL) [56]. The biological effects of PAF are closely related to prostaglandins and leukotrienes. PAF was shown to stimulate the synthesis of prostaglandins in rabbit [11] and human [10] platelets, macrophages [57], eosinophiles [20], smooth muscle cells [58], dental pulp cells [59], and human amnion and endometrium [60]. However, these data indicating that PAF causes direct stimulatory effects on prostaglandin synthesis usually do not reflect actual biochemical events; in most cases, PAF is probably an intermediate agent of regulation of prostaglandin biosynthesis in cells. In isolated endothelial cells, guinea pig peritoneal macrophages, and rabbit neutrophils, the synthesis of PAF was shown to be stimulated by various agents (bradykinin, ATP, fMet-Leu-Phe, and the ionophore A23187 before an increase in the rate of synthesis of PGI2 by the same agents [57]. PAF receptor blockers (WEB 2086 and CV 6209) decreased the rate of bradykinin-stimulated synthesis of PGI2 but did not affect the synthesis stimulated by A23187 [57].

In rat smooth muscle cells, endothelin-1 (10-11-10-6 M) stimulated the synthesis of TxA2 and PGI2, whereas exogenous PAF caused a 2-3.5-fold increase in the synthesis of TxA2 and PGI2 [58]. CV 6209, a PAF antagonist, inhibited the endothelin-1-stimulated biosynthesis of TxA2 and PGI2 in these cells [58]. This mechanism can probably operate in vivo because the PAF antagonist can block biological effects (bronchoconstriction and increase in vascular permeability) caused by administration of endothelin-1 to rats [61]. In rat dental pulp cells in vitro, PAF stimulated the production of TxA2 and PGI2 simultaneously [59]. The stimulation of PGI2 synthesis was observed at PAF concentrations higher than 10 nM and reached a plateau at 1 µM, whereas the stimulatory effect on TxA2 synthesis was observed at PAF concentrations higher than 100 nM and had no plateau on the dose-response curve [59]. PAF antagonists (CV 3988 and BN 52021), a Ca2+ antagonist (TMB-8), and a protein kinase C inhibitor (H-7) were shown to inhibit the PAF-stimulated synthesis of TxA2 and PGI2 in these cells [59]. The simultaneous activation of TxA2 and PGI2 synthesis induced by PAF in dental pulp cells was suggested to be mediated by interactions of PAF with various subtypes of PAF receptors connected to independent systems of transmembrane signaling [59]. Thus, data available on the stimulation of prostaglandin synthesis by PAF suggest that PAF works as a mediator causing mobilization of cellular arachidonic acid pools for synthesis of prostaglandins [57-59].

PAF also stimulates the synthesis of leukotrienes in rabbit and human neutrophils [14] and human monocytes [62]. A comparative study of the abilities of various lipid mediators of inflammation, such as PAF, LTB4, leukotrienes C4 and D4, and 5-HETE to stimulate leukocyte degranulation showed that only LTB4 has a stimulatory effect comparable with PAF [63].

PAF and LTB4 can act as autocrine activators of various cell functions. For example, polymerization of actin in neutrophils induced by A23187 can be prevented by LY-223982, an antagonist of LTB4 receptors, or PAF antagonists [64]. PAF antagonists inhibit the fMet-Leu-Phe-stimulated synthesis of leukotrienes and generation of superoxide radicals in rabbit neutrophils [64]. Antagonists of LTB4 and PAF can inhibit the A23187-stimulated synthesis of leukotrienes in human neutrophils [64]. LTB4 and PAF affect leukotriene synthesis by activating phosphorylation of phospholipase A2, thereby increasing the amounts of arachidonic acid available for leukotriene biosynthesis [64].

Endothelium-dependent vascular relaxation factor (EDRF), which is identical to nitric oxide (NO), is, in addition to PGI2, an important regulator of blood vessels [65]. NO released from endothelium stimulated by bradykinin, thrombin, PAF, and ATP causes relaxation of blood vessels by affecting soluble guanylate cyclase of endothelial cells [65]. PAF is known to cause either vasoconstriction or vasodilatation, depending on the type of blood vessels [44]. The vasodilator effects of PAF and acetylcholine on rat mesenteric arteries are mediated by the release of NO from endothelial cells [66]. Similarly, PAF causes relaxation of rat thoracic aorta [67]. The same concentrations of PAF that relax blood vessels stimulated a considerable increase in cGMP levels in endothelial cells. Antagonists of PAF (CV 3988 and CV 6209) and N-monoethyl-L-arginine (an inhibitor of NO synthase) inhibited the vasodilator effect of PAF [67]. These data suggest that PAF stimulates the production of NO by activating PAF receptors in endothelial cells [67].

Oxidized LDL were shown to inhibit NO synthase in endothelial cells [68] and platelets [69] and inhibit the uptake of L-arginine, a substrate of NO synthesis, by these cells [69]. This can deteriorate the normal regulation of blood vessels [28]. Lipoproteins are oxidized by reactive oxygen species generated by neutrophils and monocytes, for which PAF is the most potent activator [70]. In addition, lipid peroxidation in LDL generates 1-acyl PAF-similar lipids [71]. 1-Acyl-PAF can stimulate the formation of superoxide radicals in human neutrophils [15] and prime, together with PAF, the induction of superoxide radical generation by neutrophils stimulated by fMet-Leu-Phe or C5a [16]. These data provide a link between PAF and NO synthesis in endothelium: PAF activates neutrophils and monocytes that generate superoxide radicals and other reactive oxygen species. Reactive oxygen species oxidize LDL lipids; oxidized LDL lipids inhibit the synthesis of NO in endothelium, whereas PAF-similar lipids produced from oxidized LDL [71] can trigger a new "wave" of neutrophil and monocyte activation. These mechanisms are especially important in disease; therefore, many PAF antagonists are now regarded as potential means of treatment of cardiovascular diseases [72].

PAF interacts with various factors of cell growth and differentiation, as well as with cytokines. These interactions are especially important and evident in various states of disease (septic shock, ischemia, and airway hyperreactivity). Braquet (see [7] for review) proposed a theory of PAF--cytokine interactions in pathophysiological states. TNF, one of the most important cytokines, is synthesized by monocytes and macrophages stimulated by bacterial endotoxins or other agents liberated during inflammation [7]. TNF stimulates the synthesis of PAF by human endothelial cells and monocytes [7] and expression of the gene encoding PAF receptors in human monocytes [73]. In turn, PAF can stimulate the release of TNF from human monocytes [74] and rat alveolar macrophages [7]; it also considerably increases the production of TNF by monocytes stimulated by gamma-interferon, IL-1, muramyl dipeptide, or lipopolysaccharides [7].

Interactions of PAF and cytokines were observed in activation of neutrophil oxidative burst. Preincubation of human neutrophils with TNF or granulocyte-macrophage colony-stimulating factor increases the rate of PAF-stimulated generation of superoxide radicals by neutrophils [7]. PAF at a concentration of 10-16-10-8 M does not stimulate the neutrophil oxidative burst; however, the release of superoxide radicals by neutrophils is considerably increased when PAF is added at the end of the cell incubation with TNF; the maximal increase was observed at a PAF concentration of 10-12 M [75]. PAF antagonists partially inhibited the TNF-stimulated production of superoxide radicals by neutrophils [75]. These data indicate that PAF can mediate the TNF-induced activation of neutrophils [75].

In endothelial cells, certain agents (leukotrienes, histamine, ATP, H2O2, and bradykinin) rapidly stimulate the synthesis of PAF [7]. In contrast to these substances, TNF and IL-1 activate the de novo PAF synthesis, which requires an increase in the rate of protein synthesis. More than 30% of de novo synthesized PAF is liberated in the extracellular milieu [7]. It was suggested that PAF released from endothelial cells can probably increase the TNF and IL-1-stimulated production of granulocyte-macrophage colony-stimulating factor and IL-6 from endothelium, thereby potentiating the adhesion of neutrophils to endothelium mediated by E-selectin and ICAM-1 [76]. In these processes, PAF and TNF cause synergistic effects and stimulate the adhesion of neutrophils to endothelium [77]. Priming-effects of TNF and granulocyte-macrophage colony-stimulating factor in PAF-induced neutrophil activation may be related to modifications of functional states of certain subtypes of G-proteins, modulation of intracellular cAMP level, and decreases in the threshold concentration of Ca2+ required for cell activation [7].

Interactions of PAF with interleukins are important factors of body defenses. IL-1 is a mediator of acute and chronic inflammation [7]. The formation of PAF and IL-1 in cells are closely related processes. IL-1 was shown to stimulate the formation of PAF in endothelial cells [78], whereas PAF regulates the lipopolysaccharide-stimulated production of IL-1 by rat spleen macrophages [79]. Adding PAF at a concentration of 10 fM to lipopolysaccharide-stimulated spleen macrophages caused an increase in the activity of IL-1 in the intercellular medium [79]. L-651,392, an inhibitor of lipoxygenase, considerably attenuated this effect of PAF; therefore, the PAF-induced priming of IL-1 liberation can be mediated by leukotriene formation because LTB4 is known to stimulate the production of IL-1 by spleen macrophages [79]. In human monocytes, PAF alone cannot stimulate the synthesis of IL-1; however, adding PAF to monocytes stimulated by muramyl dipeptide or lipopolysaccharide caused a twofold to threefold increase in IL-1 secretion by monocytes [80]. The ability of PAF to regulate the production of IL-1 may be a new mechanism responsible for regulation of inflammatory and immune responses [80].

PAF (4-10 µM) is known to inhibit the phytohemagglutinin HA-16-stimulated proliferation of CD4+ T-cells; however, this process is not related to inhibition of production of IL-2 required for T-cell proliferation [23]. Studies of mechanisms responsible for this proliferation-inhibiting effect showed that PAF induces a considerable (more than 50%) decrease in the number of high-affinity IL-2 receptors in CD4+ cells without changing the affinity of these receptors for IL-2 [23]. The decrease in the number of IL-2 receptors can be a factor facilitating the PAF-induced inhibition of proliferation of CD4+ cells, and this inhibition can be an important part of the homeostatic mechanism of limiting T-cell activation by IL-2 in inflammatory responses [23].

During eosinophil activation, PAF interacts with IL-5, which stimulates differentiation and proliferation of eosinophiles and antibody-dependent cytotoxicity of mature eosinophiles [81]. IL-5 causes a priming effect on activation of eosinophiles by PAF resulting in eosinophil migration and activation of their oxidative metabolism; however, PAF does not produce such an effect in relation to eosinophiles treated with LTB4 or TNF [81]. Effects of PAF and IL-5 on eosinophiles are inhibited by PAF antagonists, whereas PAF-insensitive eosinophiles cannot respond to subsequent treatment with IL-5 [81]. A similar increase in the effect of PAF after treatment with IL-5 was observed in isolated perfused guinea pig lung [82]. Such an interaction between PAF and IL-5 may be a mechanism of development of bronchopulmonary hyperreactivity that accompanies airway inflammation [82].

Biological effects of PAF and IL-6 are also closely related in neutrophils [83, 84], alveolar macrophages [85], and fibroblasts [86]. Within the range of 10-10 to 10-8 M, PAF considerably increases the production of IL-6 by rat alveolar macrophages stimulated by muramyl dipeptide [85]. Secretion of IL-6 by macrophages is inhibited by antagonists of PAF and lipoxygenase [85]. Exogenous LTB4 blocks the inhibitory effect of lipoxygenase inhibitors on IL-6 secretion by macrophages. These data suggest that PAF regulates the secretion of IL-6 by macrophages and its activity is mediated by 5-lipoxygenase metabolites [85]. Because PAF alone cannot stimulate IL-6 secretion by macrophages, a double signal, such as muramyl dipeptide + PAF, is necessary to trigger the processes of synthesis and secretion of IL-6 in macrophages [85]. In mouse fibroblasts, PAF (1 µM) stimulated the secretion of IL-6; however, the PAF antagonist BN 52021 did not inhibit this secretion [87]. IL-6 is known to regulate the production of TNF and IL-1; PAF can act as a mediator that inhibits inflammatory processes. For example, IL-6 was shown to be able to inhibit lung inflammation caused by lipopolysaccharide administration [85].

Thus, data reviewed here suggest that PAF interacts with many cellular bioregulators during interactions between cells involved in the formation of body defenses.

PAF is known to be not the only a phospholipid metabolite formed during cell activation; cells also produce its analogs 1-acyl-PAF (1-acyl-2-acetyl-sn-glycero-3-phosphocholine) and 1-plasmalogen-PAF (1-O-alk-1´-enyl-2-acetyl-sn-glycero-3-phosphocholine). Data on biological activities of PAF cellular analogs and their interactions with other bioregulators have been reviewed recently [88]. In addition to the above-described effects of interactions between PAF and cellular bioregulators, PAF interacts with or is involved in the effects of histamine [89], fibronectin [90], tissue plasminogen activator [91], gamma-interferon [92], interleukin-8 [86, 93], and steroid hormones [94].

The data on interactions between PAF and other cell bioregulators reviewed here indicate that isolating a certain bioregulator as a key factor in this ensemble is an impossible and probably unnecessary task. A multitude of effectors interact with each other in regulation of complex and diverse biochemical processes in cells. Certain pathways for the same biochemical processes may be interchangeable. It is conceivable that disorders in interactions of cell bioregulators at certain stages can deprive cells of some types of functional activity.


1.Denizot, Y., Dassa, E., Benveniste, J., and Thomas, Y. (1989) Biochem. Biophys. Res. Commun., 161, 939-943.
2.Bussolino, F., Sordano, C., Benfenati, E., and Bozzaro, S. (1989) Eur. J. Biochem., 196, 609-615.
3.Nakayama, R., Kumagai, H., and Saito, K. (1994) Biochim Biophys. Acta, 1199, 137-143.
4.Scherer, G. F. E., and Stoffel, B. (1987) Planta 172, 127-130.
5.Sugiura, T., Yamashira, A., Kudo, N., Fukuda, T., Miyamoto, T., Chang, N. N., Kishimoto, S., Waku, K., Tanaka, T., Tsukatani, H., and Tokumura, A. (1995) Biochim. Biophys. Acta 1258, 19-26.
6.Terner, M. R., and Lumb, R. H. (1989) Biochim. Biophys. Acta 1004, 49-52.
7.Braquet, P., Paubert-Braquet, M., Bourgain, R. H., Bussolino, F., and Hosford, D. (1989) J. Lipid Mediators 1, 75-112.
8.Siess, W. (1989) Pharmacol. Rev., 69, 58-178.
9.Tence, M., Coeffier, E., Heumans, F., Polonsky, J., Codfroid, J. J., and Benveniste, J. (1981) Biochimie 63, 723-727.
10.Marcus, A. J., Safier, L. B., Ullman, H. L., and Wong, K. T. H. (1981) Blood, 58, 1027-1031.
11.Shaw, J. O., Klusick, S. J., and Hanahan, D. J. (1981) Biochim. Biophys. Acta, 663, 222-229.
12.McCulloch, R. K., and Vandongen, R. (1990) Prostaglandins 39, 13-21.
13.Karpouza, A. P., and Vakirtzi-Lemonias, C. (1997) Biochim. Biophys. Acta, 1323, 12-22.
14.O'Flaherty, J. T., Wykle, R. L., Miller, C. H., Levis, J. C., Waite, M., Bass, D. A., McCall, C. E., and DeChatlet, C. R. (1981) Am. J. Pathol., 103, 70-78.
15.Kuznetsova, T. I., and Kulikov, V. I. (1992) Biokhimiya, 57, 16-20.
16.Pinckard, R. N., and Priboda, T. J. (1996) J. Leukoc. Biol. 59, 219-228.
17.Okuda, Y., Tsuyuguchi, I., and Yamatodani, A. (1988) Int. Arch. Appl. Immunol., 85, 341-345.
18.Orlov, S. M., Kulikov, V. I., Polner, A. A., and Bergelson, L. D. (1985) Biokhimiya, 50, 680-685.
19.Kroegel, C., Yukawa, T., Dent, G., Venge, P., Chung, K. F., and Barnes, P. J. (1989) J. Immunol., 142, 3518.
20.Giembycz, M. A., Kroegel, C., and Barnes, P. J. (1990) J. Immunol., 144, 3489-3497.
21.Braquet, P., and Rola-Pleszczynski, M. (1987) Immunol. Today 8, 345-352.
22.Orlov, S. M., and Kulikov, V. I. (1987) Immunologiya, No. 3, 53-55.
23.Dulioust, A., Duprez, V., Pitton, C., Salem, P., Hemar, A., Benveniste, J., and Thomas, Y. (1990) J. Immunol., 144, 3123-3129.
24.Mazer, B., Clay, K. L., Renz, H., and Gelfand, E. W. (1990) J. Immunol., 145, 2602-2607.
25.Franklin, R. A., Mazer, B., Sawami, H., Mills, G. B., Terada, N., Lucas, J. J., and Gelfand, E. W. (1993) J. Immunol., 151, 1802-1810.
26.Muzya, G. I., Orlov, S. M., Berdyshev, E. V., and Kulikov, V. I. (1994) Biochemistry (Moscow), 59, 1054-1061 (Russ.).
27.Im, S. Y. (1996) Cancer Res., 56, 2662-2665.
28.Noel, A. A., Hobson, R. W., and Duran, W. N. (1996) Microvasc. Res., 52, 210-220.
29.Kloprogge, E., and Akkerman, J. W. N. (1984) Biochem. J. 233, 901-909.
30.Stewart, A. G., and Dusting, G. J. (1988) Br. J. Pharmacol., 94, 1225-1233.
31.Marquis, O., Robaut, C., and Cavero, I. (1988) J. Pharmacol. Exp. Therap. 250, 293-300.
32.Ukena, D., Krogel, C., Denot, G., Yukawa, T., Sybrecht, G., and Barnes, P. J. (1989) Biochem. Pharmacol., 38, 1702-1705.
33.Hwang, S. B., Lam, M. H., and Shen, T. Y. (1985) Biochem. Biophys. Res. Commun., 128, 972-979.
34.Hwang, S. B., Lee, C. S. C., Cheah, M. J., and Shen, T. Y. (1983) Biochemistry, 22, 4756-4763.
35.Hwang, S. B. (1987) Arch. Biochem. Biophys., 257, 339-344.
36.Travers, J. B., Huff, J. C., Rola-Pleszczynski, M., Gelfand, E. W., Morelli, J. G., and Murphy, R. C. (1995) J. Invest. Dermatol. 105, 816-823.
37.Nunez, D., Kumar, R., and Hanahan, D. (1989) Arch. Biochem. Biophys. 272, 466-475.
38.Valone, F. H., and Johnson, B. (1985) Thromb. Res., 40, 385-392.
39.Wykle, R. L., Surles, J. R., Piantadosi, C., Salzer, W. L., and O'Flaherty, J. T. (1982) FEBS Lett., 141, 29-32.
40.Wissner, A., Sum, P. E., Schaub, R. E., Kohler, C. A., and Goldstein, B. M. (1984) J. Med. Chem., 27, 1174-1181.
41.Hoffman, B. Ostermann, G., Kertscher, H. P., and Till, U. (1988) Tromb. Res., 49, 415-423.
42.Ohno, M., Fujita, K., Nakai, H., Kobayashi, S., Inoue, K., and Nojima, S. (1985) Chem. Pharm. Bull., 33, 572-582.
43.Rollsen, W. N. A. (1990) Drugs Future, 15, 597-603.
44.Koltai, M., Guinot, P., Hosford, D., and Braquet, P. (1994) Adv. Pharmacol., 28, 81-167.
45.Lave, D., James, C., Rajoharison, H., Bost, P. E., and Cavero, I. (1989) Drugs Future, 14, 891-898.
46.Ortega, M. P., Garcia, M. D. C., Gijon, M. A., Fau de Casa-Juana, M., Priego, J. G., Sanchez-Crespo, M., and Sunkel, C. (1990) J. Pharmacol. Exp. Therap. 255, 28-33.
47.Terashita, Z. I., Tsushima, S., Yoshioka, Y., Nomura, H., Inada, Y., and Nishikawa, K. (1983) Life Sci., 32, 1975-1982.
48.Melikyants, A. G., Koloskova, E. E., Kuznetsova, T. I., Lipatkina, L. Yu., and Kulikov, V. I. (1991) Farmakol. Toksikol., 54, 43-45.
49.Muzya, G. I., Shestakov, K. A., Bol'shakova, N. S., Chernyad'eva, I. F., Kukharchuk, V. V., and Kulikov, V. I. (1995) Khim.-Farm. Zh., 29, 34-35.
50.Godfroid, J. J., Dive, G., Lamotte-Brasseur, J., Batt, J. P., and Haymans, F. (1991) Lipids, 26, 1162-1166.
51.Shukla, S. D., and Hanahan, D. J. (1983) Arch. Biochem. Biophys., 227, 626-629.
52.Shukla, S. D. (1991) Lipids, 26, 1028-1033.
53.Avdonin, P. V., Svitina-Ulitina, I. V., and Kulikov, V. I. (1985) Biochem. Biophys. Res. Commun., 131, 307-313.
54.Kulikov, V. I., and Bergelson, L. D. (1984) Biokhimiya, 49, 1310-1315.
55.Iwamoto, S., Kawasaki, T., Kambayashi, J., Artyoshi, H., and Monden, M. (1996) Biochem. Biophys. Res. Commun., 218, 940-944.
56.Muzya, G. I., and Kulikov, V. I. (1991) Biokhimiya, 56, 1140-1144.
57.Stewart, A. G., and Phillips, W. A. (1989) Br. J. Pharmacol., 98, 141-148.
58.Takayasu-Okishio, M., Terashita, Z.-I., and Kondo, K. (1990) Biochem. Pharmacol., 40, 2713-2717.
59.Hirafuji, M., and Ogura, Y. (1990) Eur. J. Pharmacol., 185, 81-90.
60.Muzya, G. I., Kulikov, V. I., Van'ko, L. V., and Sukhikh, G. T. (1996) Akusherstvo Ginekol., No. 3, 12-16.
61.Filep, J. G., Sirois, M. G., Rousseau, A., Fournier, A., and Sirois, P. (1991) Br. J. Pharmacol., 104, 797-804.
62.Fauler, J., Sielhorst, G., and Frolich, J. C. (1989) Biochim. Biophys. Acta, 1013, 80-85.
63.Dahl, M. L. (1985) Int. Arch. Allergy Appl. Immunol., 76, 145-150.
64.McDonald, P. P., McCall, S. R., Braquet, P., and Borgeat, P. (1994) Br. J. Pharmacol., 111, 952-860.
65.Brenner, B. M., Troy, J. L., and Ballerman, B. J. (1989) J. Clin. Invest., 84, 1373-1378.
66.Chiba, Y., Mikoda, N., Kawasaki, H., and Ito, K. (1990) Naunyn-Schmeideberg's Arch. Pharmacol., 341, 68-73.
67.Moritoki, H., Hisayama, T., Takeuchi, S., Miyano, H., and Kondoh, W. (1992) Br. J. Pharmacol., 107, 196-201.
68.Cox, D. A. (1996) Pharmacol. Rev., 48, 3-19.
69.Chen, L. Y. (1996) Circulation, 93, 1740-1746.
70.Muzya, G. I., Bol'shakova, N. S., Chernyad'eva, I. F., Kukharchuk, V. V., and Kulikov, V. I. (1995) Vopr. Med. Khim., 41, 45-48.
71.Tokumura, A., Toujima, M., Yoshioka, Y., and Fukuzawa, K. J. (1996) Lipids, 31, 1251-1258.
72.Feuerstein, G., Rabinovici, R., Leor, J., Winkler, J. D., and Vonhaf, S. (1997) J. Lipid Mediators Cell Signalling 15, 255-284.
73.Dagenais, P., Thivierge, M., Parent, J. L., Stankova, J., and Rola-Pleszczynski, M. (1997) J. Leukoc. Biol., 61, 106-112.
74.Bonavida, B., Mencia-Huerta, J. M., and Braquet, P. (1990) J. Lipid Mediators 2, 65-76.
75.Paubert-Braquet, M., Hosford, D., Klotz, P., Guilband, J., and Braquet, P. (1990) J. Lipid Mediators 2, 1-14.
76.Schmid, E., Mulert, H., Budrinski, R. M., and Binder, K. (1995)  J. Interferon Cytokine Res., 15, 819-825.
77.Sterner-Kock, A. (1996) Immunology, 87, 454-460.
78.Bussolino, F., Breviario, F., Tetta, C., Aglietta, M., Mautovani, A., and Dejana, E. (1986) J. Clin. Invest., 77, 2027-2033.
79.Pignol, B., Henane, S., Chaumeron, S., Mencia-Huerta, J. M., and Braquet, P. (1990) J. Lipid Mediators, 2, 93-99.
80.Salem, P., Deryckx, S., Dumais, C., Dinarello, C. A., and Thomas, Y. (1990) J. Immunol., 144, 1338-1344.
81.Coeffier, E., Joseph, D., and Vargaftig, B. B. (1991) J. Immunol., 147, 2595-2602.
82.Pretolani, M., Lefort, J., Leduc, D., and Vargaftig, B. B. (1992) Br. J. Pharmacol., 106, 677-684.
83.Biffl, W. L., Moore, E. E., Moore, F. A., and Barnett, C. C. (1996) J. Trauma, 40, 575-578.
84.Biffl, W. L., Moore, E. E., Moore, F. A., Barnett, C. C., Silliman, C. C., and Petersen, V. M. (1996) J. Leukoc. Biol., 59, 569-574.
85.Thivierge, M., and Rola-Pleszczynski, M. (1992) J. Allergy Clin. Immunol., 90, 796-802.
86.Roth, M. (1996) J. Exp. Med., 184, 191-201.
87.Braquet, P., Pignol, B., Maisonnet, T., and Mencia-Huerta, J. M. (1991) Int. Arch. Allergy Appl. Immunol., 94, 165-166.
88.Kulikov, V. I., and Muzya, G. I. (1996) Biochemistry (Moscow), 61, 387-403 (Russ.).
89.Tomeo, A. C., Egan, R. W., and Duran, W. N. (1991) FASEB J., 5, 2850-2855.
90.Stanislawski, L., Huu, T.-P., and Perianin, A. (1990) Inflammation, 14, 523-530.
91.Emeis, J., and Kluft, C. (1985) Blood, 66, 86-91.
92.Howard, A. D., and Enckson, K. L. (1996) Cell Immunol., 174, 155-164.
93.Rainger, G. E., Fisher, A. C., and Nash, G. B. (1997) Amer. J. Physiol., 41, 114-122.
94.Giussani, D. A., Parks, J. E., Hough, S. R., Jenkins, J. L., Winter, J. A., and Natanielsz, P. W. (1997) Biol. Reprod., 56, 247-252.