Eur Respir J 1192, 5, 915- 918 EDITORIAL

Does airway smooth muscle care about platelet-activating factor?

C. Kroegel*

In 1966, BARBARO and ZvAIFLER [l] demonstrated the molecule biologically inactive. The enzyme is associ­ release of histamine from rabbit platelets during an ated with both low and high density lipoproteins and acute allergic response. Five years later, it was shown is present in blood, plasma or serum, as well as in that a soluble factor from leucocytes, capable of in­ various cells and tissues [11]. Removal of lhe acetyl ducing platelet activation, was responsible for this group is a critical step, since other PAF metabolizing effect [2, 3). In subsequent years, the factor was rec­ enzymes do not act on PAF directly. In view of its ognized for its potent ability to stimulate platelet high biological activity, rapid inactivation of PAF may aggregation and secretion and was consequently named represent a vital physiological mechanism. "platelet-activating factor" (PAF) [4]. By the end of PAF comprises a family of multiple molecular spe­ the decade, three independent groups had elucidated its cies, including both saturated and unsaturated 1-0-alkyl chemical structure as 1-alkyi-2(R)-acetyl-glycerol-3- homologues, 1-0-acyl analogues, and acetylated phosphocholine. On the basis of its structural and phosphoglycerides having polar head groups other than physiological properties, the lipid was subsequently choline. Different molecular species of PAF can be termed as P AF-acether ("ace" for acetate and "ether" produced by an inflammatory cell. The biological sig­ for the alkyl bond) [5], acetylglyceryletherphos­ nificance of the molecular heterogeneity is not yet phorylcholine (AGEPC) [6], or antihypertensive polar clear. However, given the variable biological activity renomedullary lipid (APRL) [7]. of each member of the PAP family [11), the inflam­ P AF is an unique phosphoglyceride, which possesses matory cell may be able to alter the predominate many potent biological acti vi ties relevant for various "type" of PAF, thereby determining the degree of in­ physiological ~111d pathophysiological conditions [8- 10 J. flammatory activity. Although originally described as an activator of plate­ To complicate matters further, the molecular lets 14), PAF is now known to have a wide range of heterogeneity of PAF is met by the expression of biological activities, acting on a long list of inflam­ apparently distinct PAF receptors. Several independ­ matory cells that includes mast cells, mononuclear ent studies have proposed the existence of at least two cells, neutrophils and eosinophils. In addition, PAF distinct PAF receptor subtypes in platelets and has potent biological effects on noninflammatory cells, leucocytes [14, 15]. HwANG [16] found that the rela­ such as smooth muscle cells, fibroblasts, endothelial tive potencies of PAF agonists and P AF antagonists on cells, type II alveolar cells, and some neurologically human platelets differ from those on neutrophils, and derived cell lines. that the cellular responses to PAF in these cells could Two principal pathways for the biosythesis of PAF be differentiated by pertussis toxin, cholera toxin and have been discovered [11]. In the deacylation- reac­ the presence of monovalent cations Na• and Li•. ylation pathway, PAF is derived from l-0-alkyl-2- Furthermore, the expression of multiple receptor acyl.-glycerophosphocholine cellular pools through lhe subtypes in the neutrophil [ 17) and the eosinophil action of a phospholipase A,. The resulling 1-0-alkyl- [18] have been postulated, based on the presence of 2-lyso-glycerophosphocbolirie is then acylated by an pertussis toxin-sensitive and pertussis toxin-insensitive acetyi-CoA-requiring enzyme to form PAF [12J. PAF-dependent activity in each cell type. Binding Alternatively, PAF can be synthesized via a de novo studies identified a high affinity (Kd ""0.2 nM pathway [13], which involves the transfer of phos­ for neutrophils, 0.3 nM for eosinophils) and a low phocholine from CDP-choline to 1-0-alkyl-2- affinity (Kd ""200 nM and J 1.5 nM, respectively) acetylglycerol. The former pathway predominates in receptor on these leucocytes [17, 19]. More recently, activated inflammatory cells, whereas the latter has microinjection o f size-fractionated messenger been shown to be particularly important in a number ribonucleic acid (mRNA) from the promyelocytic of noninflammatory cells and isolated tissues. leukemia cell line HL60 into the Xenopus oocyte Metabolic degradation of PAF occurs via an acetyl showed that PAF receptor activity was broadly distrib­ hydrolase, which removes the acetyl group leaving the uted in several mRNA fractions ranging from 3.5 to 6 kb [20]. This finding suggests that leucocytes " Dept Pneumology, Medical Clinics, University of Freiburg, Hug­ may encode for distinct multiple PAF receptor stetter Str. 55, D-7800, Germany. (Director: Prof. Dr. H. Matthys). subtypes. 916 C. KROEGEL

These results are in keeping with the recent mo­ followed by methacholine (1 mM) and was then ex­ lecular identification and cloning of the PAF receptor amined histologically. The authors conclude, in con­ from guinea-pig lung by HoNDA et al. [21]. Using a trast to other studies cited above [25-28]. that ligand specific cloning strategy, which involved the contraction of human isolated bronchus induced by expression of complementary deoxyribonucleic acid PAF is not related to the release of mediators [Tom (cDNA) on Xenopus oocytes, they demonstrated that secondary cells. Hence, the variability of the responses the molecular architecture of the receptor shows ho­ to PAP, observed in this study and previous studies mology to the G-protein-coupled receptors, with seven [25, 30] may not be due to bronchial tissue dwelling transmembrane spanning segments. The P AF recep­ cells but rather to a direct effect of the mediator. tor(s) is (are) linked to different intracellular pathways, What could be the reasons for these somewhat including activation of glutamyl transpeptidase conflicting results? There are several important factors (GTPase), phospholipid turnover via phospholipases which the experimental design of the study did not A2, C, and D pathways, as well as activation of protein control, thus weakening the findings of the study. kinase C and tyrosine kinase. The fact that various Firstly, whilst no correlation was observed between ions can modulate the binding properties and affinities, leucocyte numbers and the contractile response, the the presence of heterogeneity in the P AF molecule it­ activation of tissue dwelling cells and neuronal struc­ self and the recently noted oxidatively fragmented PAF tures releasing spasmogenic mediators, such as prosta­ molecule, all of which have PAF like activity, strongly glandins [27]. major basic protein (MBP) [31], support the view of an extremely complex mediator­ tachykinins [32], or acetylcholine [33], cannot be ruled receptor-effector interaction. PAF may not only use out. Hence, in order to strengthen the conclusions different molecular species, with different biological drawn from the experiments, it would have been in­ activities, but also different receptor subtypes, with teresting to evaluate the activation status of the cells. various affinity states and different signalling mecha­ Using immunofluoresence techniques, this is a com­ nisms, to differentially regulate (patho )physiological paratively easy procedure. Activation of the eosinophil, processes, for instance in bronchial tissue in asthma. for instance, can be determined using EGl or EG2 Asthma is characterized by episodic, reversible antibodies [34] or by staining for CDllb [35]. In airway narrowing, accompanied by an eosinophil­ addition, transmission electron microscopy could have dominated inflammatory response and by persistent been used to evaluate the state of the granularity of hyperresponsiveness of the airways, to a variety of the ceJis, such as the eosinophil [36]. stimuli. Various different inflammatory mediators, such Secondly, whilst neutrophils, eosinophils, lympho­ as leukotrienes, prostaglandins or histamine, have been cytes, plasma cells and epithelium were included in the implicated in asthma. Although these mediators may study, macrophages, basophils and mast cells were not account for some of the pathophysiological features of accounted for. Since basophils and macrophages can asthma, PAF has several properties which may be be activated by PAF, the smooth muscle contraction relevant to asthma, including effects on inflammatory observed may be due to the release of histamine or cells, blood vessels, bronchial hyperresponsiveness, other biologically active mediators. Although there is airway secretion, mucociliary transport, and broncho­ no evidence as yet that mast cells can be activated by constriction [8, 9, 22]. PAF directly [37], it is conceivable that PAF may P AF is recognized as being one of the most potent induce the secretion of tachykinins, which in turn spasmogens both in experimental animals and man. In could stimulate the mast cells to secrete histamine and man, PAF is considerably more potent than established other mediators. spasmogens, such as peptidoleukotrienes or histamine. Thirdly, a possible way of evaluating whether PAF In artificially ventilated humans suffering from cerebral has a direct effect on smooth muscle cells, would be death, intratracheal instillation of PAF caused broncho­ to expose human bronchial tissue to stimuli such as constriction [23]. In addition, normal subjects showed immunoglobulin E (IgE) or formyl-methionyl-leucyl­ an immediate dose-dependent bronchoconstriction with phenylalanine (fMLP) that activate inflammatory tissue 0.1 ~ol PAF causing a 40% fall in the partial ex­ dwelling cells but not smooth muscle cells in the piratory flow rate at 30% vital capacity [24]. However, presence of a P AF receptor antagonist. A response of no late-phase bronchoconstriction and no correlation smooth muscle to such a treatment would not only rule with the responsiveness to methacholine was observed. out a direct effect of P AF but clearly implicate other In vitro, human airway only showed a variable con­ mediators in bronchoconstriction. tractile response to P AF in the presence of platelets Finally, the concentration close to I ~. used to and it failed to respond to PAF alone [25]. A depen­ induce a contractile response of the bronchial tissue, dency on platelets for bronchoconstriction was also is comparatively high and a nonspecific or (sub)toxic reported in guinea-pig [26, 27] and rabbit [28], sugg­ effect was not excluded by JoHNSON et al. [29J. It esting that PAF may exert its effects indirectly via re­ might have been helpful to employ PAF receptor an­ leasing spasmogens from other cells, such as platelets. tagonists in order to exclude this effect. Furthermore, The study by JoHNSON et al. [29] in this issue of the since PAF receptors may exist in different affinity Journal examined eight specimens of bronchial tissue states, it would be interesting to know whether the obtained from five patients resected at thoracotomy. bronchial smooth muscle contracts in a concentration­ Each specimen was exposed to PAF (700 nM) dependent manner. EDffORIAL 917

Therefore, does airway smooth muscle cell seriously 11. Chilton FH. - Platelet-activating factor: synthesis, care about PAF? Clearly, as outlined here, there is a metabolism and relationship to arachidonate products. In: considerable body of evidence that PAF is "only" act­ Henson PM, Murphy RC, eds, Handbook of Inflammation, ing upon airway smooth muscle cells indirectly, via the Vol. 6, Elsevier Science Publishers BY, 1989; pp. 77- 112. release of bronchoconstricting mediators from second­ 12. Wykle RL, Kraemer WF, Schremmer JM. ary cells. However, given the considerable heteroge­ Specificity of lysophospholipase D. Biochim Biophys Acta, 1980; 619: 58- 67. ne ity of PAF molecules, PAF receptors and 13. Renooji W. Snyder F. - Biosynthesis of 1-alkyl-2- transmembrane pathways, the possibility that other, as acetyl-sn-glycero-3-phosphocholine (platelet-acti vating factor yet less established, mechanisms may be involved, has and a hypotensive lipid) by cholinc-phosphotransferase in to be considered. For instance, in keeping with the vorious rat ti ssues. Biocltim Biapltys A era. 19XI ; 663: effects of P AF on other cells or tissue [8-10], higher 545- 556. in vitro concentrations used in this study may elicit a 14. Hwang SB, Lam MH. - Species differences in the direct effect, whilst lower concentrations may not. The specific receptors of platelet-activating factor. Biochem heterogeneity or PAF may also help to explain another, Pharmacal, 1986; 35: 4511-4518. as yet unanswered, question relating to biological 15. Lambrecht G, Pamham MJ. - Kadsurenone distin­ guishes between different platelet-activating factor receptor variability among asthmatic subjects. subtypes on macrophages and polymorphonuclear leuko­ Clearly, more studies are needed before the conclu­ cytes. Br J Pharmacal, 1986; 87: 287-291. sions drawn by JoHNSON et al. [29, 30] can be fully 16. Hwang SB. - Identification of a second putative accepted. Whether or not airway smooth muscle is rcceptor of platelet-activating factor from human polymor­ affected by PAF directly, the potential significance of phonuclear leukocytes. J Bioi Clzem, 1988; 263: 3225- 3233. PAF remains. However, given Lhe number and variety 17. Naccache PH, Molski MM, Volpi M, Becher EL, of inflammatory mediators, it seems unlikely that any Sha'afi RI. - Unique inhibitory profile of platelet-activat­ single mediator accounts for all pathophysiological ing factor induced cal cium mobilization. phosphoinositide processes associated with bronchoconstriction. turnover and granular enzyme secretion in rabbit neutrophils towards pertussis toxin ilnd phorbol ester. Biochem Biophys Res Commun, 1985; 130: 677-684. 18. Kroegel C, Warner JA, Lichtenstein LM, Hubbard WC. References - Platelet activating factor (PAF) stimulates eosinophils through both pertussis toxin-sensitive and pertussis toxin­ I. Barbaro JF, Zvaiflcr NJ. Antigen-induced insensitive transduction mechanisms. FASEB J, 1990; 4: histamine release from platelets and rabbits producing A1942. homologous antibody. Proc Soc Exp Bioi Med, 1966; 122: 19. Kroegel C. Warner JA, Gicmbycz MA, Matthys H, 1245-1247. Lichtenstein LM, Barnes PJ. Dual transmembrane 2. Henson PM. - Release of vasoactive amines from signaling mechanisms in eosinophils: evidence for two ra bbit pla te le ts induced by sens itized mono nuclear functionally distinct receptors for platelet activating ·factor. leukocytes and antigen. J E:rp Med. 1970: 13 1: 2tn-306. In! Arch Allergy Appl lmmuno/, 1992; (in press). 3. Siraganian RP. Osier A. - Destruction of rabbit 20. Murphy PM, Gallin EK, Tiffany HL. - Characteri­ platelets in the allergic response of sensitized l eukocy t ~:S: zation of human phagocytic cell receptors and platelet­ demonstration of a fluid phase interm ediate. J l111111un rJl. activating factor expressed in Xenopus oocyte. J lmmunol, 1971; 106: 1244-1251. 1990; 145: 2227-2234. 4. Benveniste J, Henson PM, Cochrane CC. 21. Honda Z, Nakamura M, Miki l, Minami M, Watanabe Leukocyte-dependent histamine release from rabbit platelets: T, Seyama Y, et al. - Cloning and functional expression the role of IgE, basophils and a platelet-activating factor. J of platelet-activating factor receptor from guinea-pig lung. Exp Med, 1972; 136: 1356-1377. Nature, 1991; 349: 342-346. 5. Beneviste J, Tence M, Varenne P, Bidault J, Boullet 22. Nieminen MM, Moilanen EK, Nyholm JEJ, Koskinen C, Polonsky J. - Scmi-syntbese et structure du facteur MO, Karvonen JI, Metsa-Ketela TJ, et al. - Platelet­ activant les plaquettes (PAF): PAF-acether, un alkyl ether activating factor impairs mucociliary transport and increases analogue de la lysophosphatidylcholine. C Rend Acad Sci, plasma leukotriene B4 in man. Eur Respir J, 1991; 4: 551- 1979; 289: 1037-1040. 560. 6. Oemopoulos CA, Pinckard RN , Hunuhrw DJ. - .Plate­ 23. Gateaou 0, Amoux B, Deriaz H, Viars P, Benveniste let-acti va ting factor: evidence for 1-0 -ulkyl-2-accty l­ J. - Acute effects of intratracheal administration of PAF sn-glycerylcholine as the acti ve component (a new t ins of acether (platelet-activating factor) in humans. Am Rev Respir lipid mediators). J Bioi Chem. 1979: 254: 9335-9358. Dis, 1984; 129: A3. 7. Blank ML. Snyder F, Bycrs WL, Bronks B. Muirhead 24. Cuss FM, Dixon CMS, Bames PJ. - Effects of in­ EE. - Antihypertensive activity of an alkyl ether analog haled platelet-activating factor on pulmonary function and of phosphatidylcholine. Biochem Biophys Res Commun, bronchial responsiveness in man. Lancet, 1986; ii: 189- 192. 1979; 90: 523-534. 25. Schellenberg RR, Walker B, Snyder F. - Platelet­ 8. Bames PJ, Chung KF, Page CP. - Platelet-activating dependent contn1c1ion of human bro nchi by platelet­ factor as a mediator of allergic disease. J Allergy Clin activating factor. J Allergy C/in lmmunol, 1983: 71: Al45 Immunol, 1988; 81: 919- 934. 26. Vargafti g BB , Lefon J, Chingard M, Benveniste J. - 9. Braquet PL, Touqui L, Shen TY, Vargaftig BB. - Plate let-activating fac tor indu ce ~ a platelet-dependent Perspectives in platelet-activating factor research. Phwmaco/ bronchoconstriction unrelated to the formation of prostag­ Rev, 1987; 39: 97-145. landin derivates. Eur J Pharmacal, 1980; 65: 185- 192. 10. Prescot SM, Zimmerman GA, Mclntyre TM. - Plate­ 27. Lefort J, Rotilio 0, Vargaftig BB. - The platelet­ let-activating factor. J Bioi Chem, 1990; 265: 17381-17384. independent release of thromboxane A2 by PAF-acether from 918 C. KROEGEL guinea-pig lungs involves mechanisms distinct from those 33. Stimler-Gerard NP. - Parasympathetic stimulation as for leukotrienes. Br J Pharmacol, 1984; 82: 565-575. a mechanism for platelet-activating factor-induced contrac­ 28. Halonen M, Palmer JD, Lohman C, McManus LM, tile responses in lung. J Pharmacal Exp Ther, 1986; 237: Pinckard RN. - Differential effects of platelet depletion 209-213. on the physiologic alterations of IgE anaphylaxis and acyl 34. Bousquet J, Chanez P, Lacoste JY, Barneon G, glyceryl ether phosphorylcholine infusion in the rabbit. J Ghavanian N, Enander I, et al. - Eosinophilic inflamma­ lmmunol, 1981; 124: 41~21. tion in asthma. N Eng J Med, 1990; 323: 1033-1039. 29. Johnson PRA, Black JL, Armour CL. - Platelet­ 35. Kroegel C, Liu MC, Lichtenstein LM, Bochner BS. - activating factor-induced contraction of human isolated Antigen-induced eosinophil activation and recruitment in the bronchus. Eur Respir J, 1992; 5: 970-974. lower airways. J Allergy Clin lmmunol, !991; 87: 303. 30. Johnson PRA, Armour CL, Black JL. - The action of 36. Kroegel C, Dewar A, Yukawa T, Venge P, Barnes PJ, platelet-activating factor and its antagonism by WEB 2086 Chung KF. - Ultrastructural characterization of platelet­ on human isolated airways. Eur Respir J, 1990; 3: 55-60. activating factor-stimulated human eosinophils from subjects 31. Gundel RH, Letts LG, Gleich GJ. - Human eosi­ with asthma. Am J Pathol, 1992; (submitted). nophil major basic protein induces airway constriction and 37. Columbo M, Casolaro V, Warner JA, MacCiashan DW, airway hyperresponsiveness in primates. J Clin Invest, 1991; Kagey-Sobotka, Lichtenstein LM. - The mechanism 87: 1470-1473. of mediator release from human basophils induced by 32. Barnes PJ. - Asthma as an axon reflex. Lancet, 1986; platelet-activating factor. J lmmunol, 1990; 145: 3855- i: 242-245. 3861.