UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The cardiac actions of platelet activating factor : possible involvement in endotoxix shock Pugsley, Michael Kenneth 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_fall_pugsley_michael_kenneth.pdf [ 2.83MB ]
JSON: 831-1.0103798.json
JSON-LD: 831-1.0103798-ld.json
RDF/XML (Pretty): 831-1.0103798-rdf.xml
RDF/JSON: 831-1.0103798-rdf.json
Turtle: 831-1.0103798-turtle.txt
N-Triples: 831-1.0103798-rdf-ntriples.txt
Original Record: 831-1.0103798-source.json
Full Text

Full Text

THE CARDIAC ACTIONS OF PLATELET ACTIVATING FACTOR: POSSIBLE INVOLVEMENT IN ENDOTOXIC SHOCK by MICHAEL KENNETH PUGSLEY B.Sc. University of British Columbia, British Columbia A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF PHARMACOLOGY & THERAPEUTICS FACULTY OF MEDICINE We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1992 (c) Michael Kenneth Pugsley In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. (Signature) Department of Pharmacology The University of British Columbia Vancouver, Canada Date October 14, 1992 DE-6 (2/88) ii ABSTRACT Endotoxic shock involves impairment of the cardiovascular system. Shock is a complex pathophysiological process involving many mechanisms, including interactions between endotoxin, the production of eicosanoids and PAF, and involvement of the immune system, specifically the monocytic cell types e.g. macrophages. Many studies, especially those of Salari and Walker (1989), have shown that endotoxin-stimulated macrophages produce substances which mediate cardiac dysfunction in isolated rat hearts and that PAF antagonists afford the greatest protection against these substances. The studies described in this thesis began with characterization of the pharmacological actions of PAF on rat hearts since Salari and Walker showed that PAF was probably being produced as the result of endotoxin action on macrophages. Rat hearts were used in the Salari and Walker study and the literature is sparse concerning the actions of PAF on the rat heart. Both in vitro and in vivo actions of PAF were examined on the rat heart and circulatory system. To assess whether PAF was acting directly or indirectly on the heart, a series of experiments were performed. It was determined that PAF acted directly on coronary vasculature and on myocytes. The effects of a series of antagonists of PAF, leukotrienes, thromboxanes as well as cyclooxygenase inhibitors on responses to PAF were studied. Each iii antagonist was examined at two concentrations (0.01 and 1.0 uM). Only ibuprofen had a protective action. In view of the lack of action of PAF antagonists on the cardiac actions of PAF, the LD50 for PAF was determined in a population of Swiss CD1 mice in order to determine the ED50 value for novel PAF antagonists. The PAF antagonist, RP 59227 was also examined for antiarrhythmic activity in anesthetised acutely prepared rats. The drug did not significantly reduce the incidence of ventricular tachycardia; however, ventricular fibrillation was reduced to 30% of control. We have also examined the actions of endotoxin on the heart in vitro and in vivo. Endotoxin had very little action in vitro and in vivo on the ECG. However, endotoxin produced the expected and sustained decrease in blood pressure and this was attenuated by ibuprofen. On the other hand, ibuprofen did not change PAF actions in vivo. These studies suggest that endotoxin does not mimic the actions of PAF. However, the possibility remains that a slow release of PAF, via the stimulation of monocytes, is involved in endotoxin action. iv TABLE OF CONTENTS CHAPTER Page Abstract ii Table of Contents iv List of Figures vi List of Tables viii List of Abbreviations ix Dedication xi Acknowledgements xii 1 Introduction 1 1.1 Endotoxic shock - an overview 1 1.1.1 Leukocytes and cardiac dysfunction 2 1.2 Platelet-Activating Factor (PAF) and other 6 autacoids producing Cardiac Dysfunction 1.2.1 General History 6 1.2.2 Eicosanoids - an overview 8 1.2.3 Platelet-Activating Factor 11 Chemistry of PAF 11 Biochemistry of PAF 13 Chemical Analogues of PAF 17 Methylcarbamyl-PAF 20 1.3 Pharmacology of PAF - General Overview 21 1.3.1 Systemic Activity 22 1.3.2 Pathophysiological actions of PAF 25 1.3.3 Cardiac actions of PAF 25 Actions in whole hearts 26 Actions on isolated cardiac tissues 28 1.3.4 Arrhythmogenesis 3 0 1.4 PAF binding sites and PAF receptors 32 1.5 Endotoxic Shock 35 1.5.1 Pharmacology of shock 35 1.5.2 Cardiovascular dysfunction 39 1.6 Objectives 40 2 Methods 43 2.1 Cardiac preparations 43 2.1.1 Isolated rat hearts 43 Perfusion apparatus 44 Concentration-response studies for PAF and 45 nifedipine PAF and Eicosanoid blockers 47 PAF's action on contraction independent 48 of flow PAF action on coronary flow 48 E. coli endotoxin studies 49 Statistical analysis 49 Intact rat studies 51 Surgical preparation 52 Design and analysis 53 Evaluation of a new ECG measure (RSh) 55 Toxicological studies with PAF 56 Lethal-dose curves in mice 56 Design and analysis 57 Ischaemia-induced arrhythmias 58 Surgical preparation in acute studies 58 Design and analysis 59 Pre- and post-occlusion ECG changes 61 Analysis of arrhythmias 62 Results 64 Actions of PAF 64 In vitro effects of PAF 64 In vivo effects of PAF 65 Actions of PAF in the presence of 66 eicosanoid blockers Actions of PAF on coronary arteries 67 Actions of PAF on ventricular contraction 68 PAF activity in rat and rabbit hearts 69 Effectiveness of methylcarbamyl-PAF 69 vs. PAF Lethality of PAF in mice 70 Cardiovascular, ECG and Antiarrhythmic 70 actions of RP59227, a new PAF antagonist Actions of E. coli endotoxin 71 In vitro effects of endotoxin 71 In vivo effects of endotoxin 72 Discussion 107 PAF actions on the heart 107 PAF actions on the cardiovascular system 110 Antiarrhythmic actions of RP59227 against 114 ischaemia-induced arrhythmias 4.4 Actions of E. coli endotoxin on the heart 117 and cardiovascular system 5 Summary 121 2 2 2 2 2 2 2 2 2 2 2, 2 , 2 . 2 . 3 3 , 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 3 . 4 4 . 4 . 4 . . 1 . . 1 . 2 . 2 , . 2 , . 2 , . 3 . 3 . . 3 . . 4 . 4 . . 4 . . 4 . . 4 . . 1 . 1 . . 1 . .2 ,3 ,4 .5 6 ,7 8 9 9 . 9 . 1 2 3 . 1 . 6 . 2 . 1 . 2 .3 . 1 .2 . 1 .2 ,3 ,4 1 2 1 2 6 References 124 vi LIST OF FIGURES Figures Page 1 Effects of PAF on peak systolic and end- 80 diastolic left ventricular pressure in vitro. 2 Effects of PAF on (A) the maximum rate of 82 intraventricular pressure development (+dP/dtmax) and (B) coronary flow in vitro. 3 Actions of PAF and Lyso-PAF on the P-R interval 84 and Heart Rate in vitro. 4 The actions of PAF on mean blood pressure in 86 intact rats. 5 The actions of PAF on heart rate in intact rats. 88 6 The in vivo actions of PAF on (A) the P-R 90 interval and (B) the QRS interval of the ECG. 7 The in vivo actions of PAF on (A) blood pressure 92 and (B) the P-R interval in the absence and presence of ibuprofen. 8 Action of PAF on coronary flow in contracting vs. 94 non-contracting isolated hearts. 9 Comparison of reductions in coronary flow induced 96 by PAF in non-contracting hearts with those induced by vasopressin and ergonovine. 10 Nifedipine dose-response curve for coronary flow 98 and systolic pressure. 11 Effects of PAF in the absence and presence of 100 nifedipine. 12 PAF and MC-PAF actions on (A) blood pressure and 102 (B) heart rate in intact rats. 13 The lethality-dose curve for PAF. 104 14 The time effects of PAF and endotoxin on (A) 106 blood pressure and (B) the P-R interval in the absence and presence of ibuprofen. viii LIST OF TABLES Table Page I The order and purpose of experiments 42 performed. II Percent reversal of responses to PAF by 75 various antagonists. Ill The haemodynamic and ECG responses to 76 RP 59227. IV Antiarrhythmic actions of RP 59227 during 77 coronary artery occlusion in anaesthetised rats. V The actions of endotoxin on isolated hearts. 78 LIST OF ABBREVIATIONS AA AEGPC a-level ANOVA APRL AS B.P. BSA cm °C ECG ED50 EDRF g hr(s) HZ i.p. i.v. kg LD50 < LPS arachidonic acid alkyl ether glycero-phosphorylcholine level of significance analysis of variance antihypertensive polar renomedullary lipid arrhythmia score blood pressure bovine serum albumin centimetre degree celcius electrocardiogram dose of drug producing half-maximal response endothelium-derived relaxation factor gram hour(s) hertz intraperitoneally intravenous kilogram dose of drug producing 50% lethality less than 1ipopolysaccharide leukotriene X mAb monoclonal antibody /xM micromolar ml millilitre mm millimetre mmHg millimetres of mercury mM millimolar min minute MW molecular weight NED normal equivalent deviation NO nitric oxide PAF platelet activating factor PEG polyethylene glycol 400 PG prostaglandin pH hydrogen ion concentration PKC protein kinase C PLP-A2 phospholipase A 2 % percentage SEM standard error of the mean sec second TNF tumor necrosis factor TTX tetrodotoxin TX thromboxane VPB ventricular premature beat VF ventricular fibrillation VT ventricular tachycardia xi Dedication This thesis is dedicated to my beloved parents xii Acknowledgements I wish to thank all the members of my thesis committee (Drs Godin, Salari and Sutter) for all their help in the completion of this thesis. As well I want to thank all the members of the department (George, Christian, "Jiffy" Jeff, Elaine, Janelle, Margaret and last but not least Lorraine)-for all the little things which helped me complete my degree. I want to thank the B.C. & Yukon Heart and Stroke Foundation for providing me with financial support. lam indebted forever to both Dr. Greg Beatch and Paisley Howard for teaching me everything they know. Again I want to express my sincere gratitiude to my parents, Ken and Mary, my brother, David, sister Christina and the rest of my family for having faith in me and my career in heart research. To all my "monkey-men" friends both in the lab (Bill, Eric, Paul, Jeffrey etc.) and not (Ian, Gord, Wayne, Craig, Spike, etc.) I say "thanks guys" for all the encouragement throughout the last few years. Last but not least I want to thank my supervisor, Dr. M.J.A. Walker (a.k.a. "Homey"), for offering me the opportunity to work with him on this project and for transforming/ transmogrifying me (I hopel) into a real pharmacologist-to you lam indebted forever. 1 1 INTRODUCTION 1.1 Endotoxic shock- an overview Endotoxic or septic shock is a clinical term which is most aptly used to describe a progressive collapse of the cardiovascular system leading to failure of tissue perfusion resulting in profound cellular hypoxia, ischemia, and general dysfunction (Ball et al., 1986). Circulatory shock is of major clinical relevance as mortality rates are high (30-70%) amongst patients with septic shock. This results in over 100,000 deaths yearly in the U.S. (Ziegler et al., 1991) despite the use of treatments which include antibiotics and pharmacological agents such as steroids (Fletcher and Ramwell, 1980). The clinical features of endotoxic shock include fever, leukopenia, and hypotension -regardless of the type of bacteria causing the infection (Sanderson, 1990). It is now agreed that it is not the endotoxin produced by the bacterium which initiates systemic responses resulting in shock states, but rather the lipopolysaccharide (LPS) cell wall component of disintegrating gram negative bacteria (Ulevitch, 1991). LPS contains a series of components which convey both structural and antigenic diversity to a bacterium. A lipid component, Lipid-A, which is highly conserved amongst gram negative bacteria confers 2 the toxicity of endotoxin (reviewed by Glauser et al., 1991). The marked similarities between the pathophysiology of live gram negative bacteria and LPS endotoxin has led to the use of purified LPS as a means by which to produce and study shock in animals (Guenter et al., 1969). The time-course of experimental endotoxemia is well defined and reproducible in a variety of animal models (Yellin et al., 1986). Since endotoxins activate many humoral and cellular targets it is now possible to experimentally delineate the pathophysiological sequelae of LPS action and examine its involvement in activating the complement system, endothelial cells and macrophages, as well as other monocytic and granulocytic cell types of the inflammatory process. 1.1.1 Leukocytes and cardiac dysfunction The pathophysiology of endotoxic shock is complicated and remains to be elucidated. Many chemical mediators such as histamine, kinin, serotonin (Vyhnanek, 1979) and more recently discovered vasoactive lipids derived from arachidonic acid (AA), including leukotrienes (LT), thromboxanes (TX) and prostaglandins (PG), have been extensively studied (Bult and Herman, 1982; Fletcher and Ramwell, 1980 and Feuerstein and Hallenback, 1987), and have been implicated as pathophysiological mediators in endotoxic shock. By activating phospholipase-A2 (PLP-A2), a number of 3 endogenous and pathologic stimuli are able to liberate free AA from cell membranes. The release of PGs, TXs and LTs from a cell is determined by the activity of their metabolizing enzymes, although AA is the substrate for all eicosanoids. Pre-treatment of animals or isolated hearts with indomethacin and other selective blockers of the cyclooxygenase and 5-lipoxygenase enzyme pathways provides partial protection against LPS toxicity, implicating these enzymes, and most importantly their metabolites, in endotoxic shock states (Parratt, 1983; Lefer, 1983; Etienne et al., 1986). However, the relationship between the release of these vasoactive substances and the changes induced by endotoxin remains controversial (Terashita et al., 1985). This, in turn, has prompted intensive investigation into the involvement of inflammatory processes as a means by which endotoxin may mediate endotoxic shock. Macrophages and other monocytic or granulocytic cell types play important roles in inflammation and defence against bacterial infection. Many studies have shown that LPS has multiple effects, including triggering the secretion of small quantities of AA metabolites, such as PGs, TXs and LTs in many cell types of various species including human blood monocytes and murine peritoneal macrophages (Kurland and Bockman, 1978). LPS also enhances the synthesis and release of tumor necrosis factor (TNF), and growth factors from monocytes (reviewed by Morrison and Ulevitch, 1978). Aderem et al. (1986) reported that LPS, at concentrations 4 similar to those found in the blood of gram negative bacteremic patients, was minimally effective in triggering the AA cascade pathway; instead, it "primed" macrophages for enhanced AA metabolite release when challenged with a second stimulus. Platelet Activating Factor (PAF), an alkyl-ether phospholipid, was found to be released from rabbit basophils after immunopharmacological challenge (Benveniste et al., 1972). PAF has many biological actions beside enhancement of platelet aggregation. These include hypotension in endotoxemic rats (Caillard et al., 1982; Doebber et al., 1985), bronchoconstriction (Page and Robertson, 1987) and cardiac dysfunction (Felix et al., 1990; Pugsley et al., 1991). Numerous studies indicate that PAF can be released from a variety of cell types in many species by both immunological and non-immunological processes. Stewart and Phillips (1983) showed that guinea pig macrophages contain cell-associated PAF whose levels increase following endotoxin stimulation. Circulating blood levels of PAF are markedly elevated during endotoxemia (Doebber et al., 1985). A PAF-like lipid compound has also been identified in exudates of peritoneal macrophages and splenic cells of rats innoculated with E. coli (Innarea et al., 1985) suggesting a potential cellular source of PAF in endotoxaemia. Furthermore, PAF and endotoxin significantly increase TXA2, TXB2 and PGI2 levels in equine macrophages, responses which, rather than being blocked by the PAF antagonist SRI 63-441, 5 are enhanced (Morris and Moore, 1989). Endotoxin has also been shown to increase levels of the cytokines and TNF in human platelet-free monocytes and this activity is significantly enhanced by PAF over the concentration range of 0.01-1.0/zM. Many other studies have confirmed this endotoxin-PAF interaction (see Barthelson et al., 1988; Pignol et al., 1990). PAF is also involved in cardiac manifestations of immediate hypersensitivity reactions (Levi et al., 1984). PAF induces a cardiac dysfunction which resembles that observed in anaphylaxis. This dysfunction has been ascribed to the release of TXA2 and LTC4 (Benveniste et al., 1983; Piper and Stewart, 1986). The serial-perfusion experiments of Salari and Walker (1989) provided evidence that endotoxin-stimulated macrophages generate many mediators capable of influencing the ventricular function of the heart. Use of blockers for AA metabolites, free-radicals and PAF showed that only PAF antagonists significantly prevented the action of macrophage-derived products in the heart. Thus, activation of macrophages, and perhaps other monocytes, acting synergistically with endotoxin may be the mechanism by which severe cardiac dysfunction occurs with endotoxemia. This dysfunction results in a high incidence of mortality in patients with toxaemia. 6 1.2 Platelet-Activating Factor (PAF) and other autacoids producing cardiac dysfunction 1.2.1 General History According to J. Benveniste (1982), the history of Platelet-Activating Factor (PAF) is an example of "the imprecision of scientific boundaries". PAF is a phospholipid. Phospholipids are usually inert and their biological role is predominantly confined to regulating the composition of cell membranes. PAF is one of the first phospholipids shown to possess biological activity and to be involved in many physiological processes, such as influencing ion channel regulation, specifically K+ channels (Ordway et al., 1991). PAF is also involved in many pathophysiological processes such as asthma, cardiac anaphylaxis and endotoxic shock (Hanahan, 1986). PAF was discovered in 1966 when Barbaro and Zvaifler demonstrated that a leukocyte-dependent substance possessed the ability to release vasoactive amines from rabbit platelets. However, it was Sirganian and Ostler (1971) who were the first to detect the presence of a soluble intermediate produced as a result of interaction between rabbit leukocytes and platelets. They considered this substance to be "lytic" in action but did not characterize it. In 1972, Benveniste, Henson and Cochrane showed that this "soluble intermediate" was released from rabbit 7 basophils through an IgE-dependent process and coined the term "Platelet-Activating Factor" or acronym PAF. They continued their attempts to elucidate the chemical nature of this compound - speculating that it was a lipid. PAF was then detected in human leukocytes (Benveniste, 1974) and shown to both aggregate human platelets and to release vasoactive mediators from them (Benveniste et al., 1975). Benveniste et al. (1977) revolutionized the PAF story by elucidating the chemical nature of this compound. It was shown to be a glycerophospholipid which possessed a choline polar head group at C3, an ether-linked acyl moiety at the C2 position and which was without an ester-link at the C^ position. The first report on the partial synthesis of PAF was published in 1979 by Demopolous et al.. They reported the synthesis of l-0-alkyl-2-acetyl-sn-glyceryl-3-phosphoryl-choline (AGEPC) which mimicked exactly the biological activity of the natural material. Shortly thereafter, Benveniste et al. (1979) reported on an alternative semi-synthetic pathway yielding the same lipid derivative. Concommitantly, Blank et al. (1979) published a semisynthetic approach to AGEPC and considered it to be a potent hypotensive agent. Their entirely different but parallel line of research culminated in the description of a renal medullary-derived compound called antihypertensive polar renomedullary lipid (APRL). This and PAF are the same molecule. At the same time, the total synthesis of PAF was 8 achieved thus making this lipid readily available for experimental study (Godfroid et al., 1980). Hanahan et al., (1980) successfully elucidated the total structure of the naturally-occurring lipid from sensitized rabbit basophils. Then, in 1982, Benveniste et al. showed that PAF was sensitive to phospholipase-A2 (PLP-A2) • Since this finding the complete, complex, synthetic and metabolic pathways for PAF have been determined (Snyder, 1990) . Research in this area is multi-faceted and now involves elucidating the pharmacological properties of PAF. 1.2.2 Eicosanoids - an overview Arachidonic acid (AA) is a basic cell constituent (an essential fatty acid) and is the main precursor in the synthesis of eicosanoids. All cell membranes have the potential to produce eicosanoids when AA is released from phospholipid sources. This involves hydrolysis of the ester link joining AA to the phospholipid glycerol backbone by phospholipase enzymes, especially PLP-A2. The release of AA is the rate-limiting step in the synthesis of PG's, LT's, and TX's. PLP-A2 is activated by non-specific stimuli including cell damage or injury (Piper and Vane, 1971) and, since it is Ca2+-dependent, the influx of Ca2+ into cells will cause enzyme activation resulting in the production of eicosanoids. 9 The best studied pathway for AA metabolism involves the conversion of AA to prostaglandins (PG's). The PG endoperoxide synthetase (or cyclo-oxygenase) enzyme converts AA to an extremely unstable endoperoxide intermediate. Several other enzymes convert endoperoxides to relatively stable end-products. The endoperoxides PGG2 and PGH2 are precursors for the production of thromboxane-A2 (TXA2) via the thromboxane synthetase enzyme. This enzyme is found in blood platelets. TXA2 is a potent platelet-aggregator and vasoconstrictor in many species (Hamberg et al., 1975). However, TXA2 is unstable and is rapidly converted to TXB2-PGH2 can be converted via prostacyclin synthetase into prostacyclin (PGI2)• PGI2, a powerful vasodilator and inhibitor of platelet aggregation (Moncada et al., 1976), is produced by endothelial cells. AA can also be metabolized by the Ca2+-dependent lipoxygenase enzymes, especially 5-lipoxygenase. The resulting hydroperoxy derivatives are short-lived intermediates which are converted to several hydroxy-fatty acids, the monohydroxyeicosaenoic (HETE) acids. 5-lipoxygenase converts AA via 5-hydroperoxyeicosa-tetraenoic (5-HPETE) acid into the 5,6 epoxide known as leukotriene-A4 (LTA4) . LTA4 is a substrate for both a hydrolase enzyme which produces 5,12-dihydroxyeicosa-tetraenoic acid or LTB4 (a potent leukocyte chemotaxin and inflammatory mediator) and a glutathione-S-transferase enzyme which produces LTC4 which is then quickly converted by gamma-glutamyl 10 transpeptidase into LTD4. Cysteinyl glycinase converts LTD4 into LTE4, (see Piper, 1983; Feuerstein and Hallenbeck, 1987; Williams and Higgs, 1988). These eicosanoids are rapidly and efficiently cleared from the blood. PG's are inactivated after a single pass through the pulmonary circulation while the LT's and TX's are metabolized by the liver and excreted into the bile as inactive metabolites. PAF has a controversial relationship to eicosanoids since observations both favor and negate its involvement in many in vivo/in vitro immunological and non-immunological systems. For example, coronary vasoconstriction induced by PAF and subsequent cardiac failure seems to be partially mediated by TXA2/ since indomethacin reduces the cardiac effects of PAF (Feuerstein et al., 1983). Levi et al. (1984) showed that indomethacin and FPL 55712, a LTC4/LTD4 receptor antagonist, did not have any significant effect on PAF-mediated cardiac dysfunction. However, Piper and Stewart (1986) argue that LTC4 may be responsible for coronary vasoconstriction by PAF since the 5-lipoxygenase inhibitor dimethylcarbazine attenuates the coronary vasoconstriction mediated by PAF while other eicosanoids, such as TXA2, may be responsible for decreased cardiac contractility. The literature abounds with such contradictory findings. However, the reconciling link has been speculated to involve PAF interaction with cellular blood components. Activation and aggregation of white blood cells causes 11 release of secondary mediators such as serotonin (5-HT), TXA2/ LT's, PG's, vasoactive peptides and oxygen radicals. These secondary mediators, as well as PAF itself, have deleterious actions on all cell membranes and vasculature (Feuerstein and Hallenbeck, 1987). Thus, much work has still to be performed to determine the significance of eicosanoids and their interrelationship to PAF. Studies into the chemical nature of PAF have led to structural analogs of PAF which can be used to examine such problems. 1.2.3 Platelet Activating Factor Chemistry of PAF PAF is an unsymmetrically substituted D-glycerol possessing a chiral carbon. It is isolated as l-0-alkyl-2-acetyl-sii-glycero-3-phosphocholine. PAF is currently the most potent biologically active lipid mediator known, and acts at concentrations as low as 10~14M. However, its amphiphilic nature results in high levels of non-specific binding (Snyder, 1990). Since PAF is structurally similar to naturally occurring plasmalogens, these compounds are used as the chiral precursors for PAF synthesis by catalytic reduction of the C^ ether side chain and acetylation of the C2-OH moiety (Braquet et al., 1987). The synthesis of PAF is a lengthy procedure, usually involving 11-15 steps, due 12 to the many complex blocking and unblocking steps required. Yields are usually between 5-25% (Demopolous et al., 1980). Attempts have been made to resolve PAF into its optical enantiomers; however, overall yields are only approximately 5% (Braquet et al., 1987). PAF is a very hygroscopic molecule and hydrolysis can occur rapidly in solution. In blood of various animals, i.v. bolus injections of [3H]-PAF show a short t^/2 of 30 sec. It is rapidly cleared from blood (Yamashita et al., 1983) with a parallel increase in radioactivity in various organs and tissues (Braquet et al., 1987). The ether-linked C± moiety of the glycerol backbone conveys significant activity for PAF. Analogs lacking this ether side chain possess <1% of the potency of PAF. The short acetate residue at the C2 position is also required for PAF activity; hydrolysis of this group produces lyso-PAF, an inactive metabolite. The polar head group at C3 is unimportant, since altering it does not significantly modify PAF activity. Many synonyms for PAF exist (a generic term for this class of ether phospholipids which may be misleading in view of its diverse biological actions) and are based on structural characteristics. Diverse terms such as alkyl-ether glycero-phosphocholine (AGEPC), alkylacetyl-GPC and PAF-acether (acet for acetate and ether for the alkyl bond) are used interchangeably for PAF, and all are referred to by researchers in the field. 13 Biochemistry of PAF Since the stimulation of a diverse number of cells and tissues including, human and rabbit platelets (Chignard et al., 1980), alveolar macrophages (Arnoux et al., 1980), endothelial cells (Whatley et al., 1990) and the isolated guinea-pig heart (Montruccio et al., 1989) can lead to the formation of PAF, it is important to examine its metabolic origin and fate. PAF is not a preformed mediator since disruption of cells mechanically does not cause PAF release (Camussi et al., 1983) while in endothelial cells it is formed but not secreted (Prescott et al., 1990). So, despite the fact that PAF was originally described as a soluble factor in blood it is only released when cells are activated. Two enzymatic pathways have been documented in PAF synthesis. The first pathway, remodelling, involves structural modification of a pre-existing ether-linked lipid, usually a structural component of the membrane. The de novo pathway, which is analogous to phosphatidylcholine synthesis, uses highly specific, sequential biochemical modification beginning with an ether-linked metabolic intermediate formed soon after the ether bond is created on the PAF molecule. The enzymes involved in each pathway for the transformation will be described briefly. The first pathway to be described for the synthesis of PAF was the remodelling pathway (Wykle et al., 1980). This 14 pathway increases PAF synthesis and secretion when stimulated. It is implicated in pathophysiological reactions and involves two highly-specific and Ca2+-dependent enzymes. The first event in the two-step sequence involves catalytic hydrolysis of the C2 fatty acyl moiety from a l-alkyl-2-acyl-sn-glycero-3-phosphocholine precursor by a cytosolic arachidonoyl (or polyenol) PLP-A2 enzyme, and yields a free fatty acid and the intermediate 1-0-alkyl-sn-glycero-3-phosphocholine, or lyso-PAF. This enzyme is highly specific for alkylacyIphosphocholine and prefers AA for optimal activity at the C2 position in the precursor. It is believed that this enzyme is activated by protein kinase C (PKC) and Ca2+ influx (Prescott et al., 1990), since it has been shown to be inhibited by mepacrine and ethyleneglycol-tetraacetic acid, EGTA (Benveniste et al., 1990). A second enzyme catalyzes the conversion of lyso-PAF to PAF via acetylCoA:lyso-PAF acetyltransferase (Wykle et al., 1980). This enzyme is associated with the endoplasmic reticulum and has been partially purified and characterized (Snyder, 1990). It is also Ca2+-dependent; however, no potent inhibitors are available. It has been shown to be activated by rapid phosphorylation of an inactive precursor via PKC or cAMP-dependent protein kinase (Braquet et al., 1990). These enzymatic reactions result in the substitution of an acetate group for the long chain acyl moiety at the C2 position of the alkylether-phosphocholine molecule. The 15 long chain fatty acids at the C2 position are usually polyunsaturated alkylacylglycero-phosphocholines consisting mainly of AA. Therefore, PAF formation via the remodelling pathway is accompanied by the release of AA which may be rapidly converted to eicosanoids. This emphasizes the complex interrelationship between PAF and AA and therefore interpretation of biological response must be made with reservation in systems that form PAF. The de novo pathway to PAF synthesis is responsible for basal levels of PAF (Hanahan, 1986) . It consists of the initial enzymatic transfer of an acetyl moiety to the C2 position of an alkyllyso-glycerophosphate precursor via an acetylCoA:alkyllysophosphate acetyl-transferase. The resulting intermediate then has its C3 phosphate moiety hydrolyzed by an alkylacetyl-glycerophosphate phosphohydrolase and subsequent transfer of a phosphatidylcholine residue occurs at this position by a dithiothreitol-insensitive choline phosphotransferase which completes the synthesis of a PAF molecule (Renooij and Snyder, 1981). These de novo enzymes are all inhibited by Ca2+ influx and are highly membrane bound (Snyder, 1990). Inactivation of PAF occurs via an acid-labile factor identified as the enzyme, acetylhydrolase, and also by an arachidonic or polyenolate transacylase, which together ultimately hydrolyze the C2 acetyl moiety. Acetylhydrolase is a highly specific enzyme for PAF which acts independently of Ca2+. It cleaves only short-16 chain acyl phosphocholines at the C2 position, and is present in intra- and extracellular forms, both of which possess identical catalytic activity, despite differences in some properties, such as molecular weight (MW) and susceptibility to inhibition and proteases (Blank et al., 1981). Little is known about the intracellular form; however, the extracellular (or plasma) form has been isolated as a protein which is not susceptible to protease and which has a low MW (Yamashita et al., 1983; Blank et al., 1983). The active plasma enzyme is associated with low density lipoprotein (LDL) and that associated with high density lipoprotein (HDL) is inactive (Stafforini et al., 1990), yet it has the same enzyme kinetic constants when isolated from either lipoprotein. The resultant product is lyso-PAF which is either metabolized or released to extracellular fluid. Lyso-PAF does not possess the biological activity of PAF but can be cytotoxic since, being a lysophosphocholine, it possesses detergent-like properties (Hanahan, 1986). Arachidonate or polyenolate transacylase then introduces a long chain fatty acid into the C2 position of lyso-PAF, the resulting alkylacyl-GPC becomes a component of the cell membrane. This lyso-PAF is an obligatory intermediate in the synthesis and inactivation of PAF in a bicyclic metabolic pathway (Braquet et al., 1987). Exogenous lyso-PAF is preferentially converted to alkylacyl-GPC rather than to PAF; therefore, acyltransferase has a 17 greater affinity for lyso-PAF than acetyltransferase. PAF has a very complex, metabolism which involves divalent cations, synthesizing and degradative enzymes, and other compounds including endogenous phosphorylated lipocortins and inhibitory lipids such as unsaturated fatty acids, all of which inhibit PLP-A2. However, even as we try to determine these controlled regulatory processes, evidence shows that such processes may not always be easily understood. Recent evidence using oxidative fragmentation of synthetic 2-arachidonoyl phosphatidylcholines in an attempt to shorten the C2 residue, and studies involving fragmentation of this synthetic compound by ozonolysis and uncontrolled free-radical reactions showed that only short or intermediate length C2 residues are specific substrates for acetylhydrolase (Stremler et al., 1991). Therefore, rather than by the route of enzyme reactions, oxidative fragmentation of various phosphocholines containing polyunsaturated acyl residues may form vasoactive phosphocholines with the potential for unregulated activation of target cells, i.e. macrophages, leading to inflammation of tissue and injury (Smiley et al., 1991). Chemical analogs of PAF The synthesis of analogs and the elucidation of the structure-activity relationships (SAR) of PAF were performed 18 with various aims, including examining the relationship between pharmacological activity and molecular structure; improving therapeutic applications, especially antihypertensive properties; searching for antagonists; determining molecular mechanisms of action and obtaining information regarding characterization, identification and isolation of PAF receptors. The naturally occurring PAF molecule has an R configuration at C2, and changing this to the S-configuration greatly reduces PAF activity (Hanahan, 1986). The ether bond at the C^ position is also critical for PAF activity, so that replacement of the ether bond with ester or methylene groups (Braquet et al., 1984), or substitution of sulfur for oxygen (Wissener et al., 1984) inactivates PAF. Alterations have also been made on the alkyl chain length. Maximal activity, as assessed by platelet aggregation, was observed for Ci6~ci8 analogs of PAF (Godfroid et al., 1987). This observation was confirmed with C16-C18 alkyl chain homologs of PAF, compounds which were the most potent of all agents tested, for vasoconstriction and negative inotropism in isolated guinea pig hearts (Levi et al., 1989). Elucidation of the action of acetylhydrolase on deacetylation of PAF at the C2 position, rendering PAF inactive, prompted interest in substitutions at this position. Of the many compounds synthesized, the methylcarbamyl (see discussion below) and ethoxy analogs 19 (Braquet et al., 1987) have been the most intensely studied. The methylcarbamyl group renders PAF resistant to serum acetylhydrolase, while the molecule retains the potency of PAF (0*Flaherty et al., 1987). The ethoxy analogs retain only about 4% of the PAF activity (Braquet et al., 1987) which may indicate that no transfer of the C2 acetyl moiety is required for activity. Effectiveness of PAF relates mainly to the length and size of the functional C2 moiety. With regard to the C3 phosphoryl group, replacement by phosphonate does not modify the efficacy of PAF. However, if an ethoxide or carboxyl group is used, PAF potency is reduced (Braquet et al., 1987). In the choline series of analogs, substitutions at the methyl moiety show a decrease in efficacy with decreasing numbers of methylamino groups on the choline chain, i.e. the methylamino analog is less effective than the trimethylammonium analog (Coeffier et al., 1986). Coeffier et al. (1986) also showed that replacing the quaternary ammonium group by cyclic derivatives produces analogs more efficacious than PAF. Analogs in which the methylene bridge separating the phosphate and trimethylammonium groups is lengthened shows a progressive decrease in platelet aggregation and hypotensive efficacy (Wissener et al., 1986). The last structural component to be considered is that of the glycerol backbone of PAF. Extension of the glycerol backbone by the incorporation of a methylene group between 20 either C1-C2 or C2-C3 results in a decreased activity (Wissener et al., 1985) compared to PAF. However, as with all structural studies, problems arise. These include uncertainties concerning the 3-dimensional structure (or spatial conformation) of agents since the conformation of the crystal may not be the pharmacologically active one at the receptor level. Many synthetic PAF antagonists have been developed. Ironically, most of the potent antagonists are naturally occurring plant alkaloids (Handley, 1990). Methylcarbamyl-PAF This analog of PAF has been shown to compete with PAF for binding to receptors on human polymorphonuclear leukocytes (O'Flaherty et al., 1987) and also to stimulate synthesize of PAF in human neutrophils (Tessner et al., 1989). Unlike PAF, methylcarbamyl-PAF is resistant to hydrolysis by neutrophils and human serum. This analog has been used to study the binding, subcellular distribution and the in vivo actions of PAF, which would have otherwise been limited by the rapid degradation of the parent molecule (O'Flaherty et al., 1987). 21 1.3 Pharmacology of PAF- General Overview As outlined earlier, PAF is synthesized and sometimes secreted from cells in both the systemic circulation (which includes monocytes, eosinophils, basophils, neutrophils) and in the reticuloendothelial system (which includes endothelial cells, macrophages and mast cells). Various organs also produce PAF , including the heart, lung, kidney, brain, and intestine (see review by Goldstein et al., 1991). PAF effects are both route- and species-dependent (Braquet et al., 1987). The actions of aerosolized PAF on lung tissue is platelet-independent, while parenterally, activity on the lung is platelet-dependent (Rabinovici et al., 1991). Generally, it is believed that the overall actions of PAF are dependent upon the cell type first making contact with PAF. PAF has actions on many tissues and cells concerned with normal cardiovascular function. A single i.v. injection of PAF produces marked systemic effects in all species studied (from rodents to humans), including pulmonary hypertension, increased vascular permeability, bronchoconstriction, thrombocytopenia, and hypotension (see reviews by Braquet et al., 1987; Page and Abbott, 1989; Koltai et al., 1991). The mechanisms by which these responses occur are not clearly defined, since reliable methods for the localization and measurement of endogenous PAF are currently inadequate. Of all the actions of PAF, 22 the most extensively studied and least understood is the prominent hypotension. It has been shown that the decrease in blood pressure is dose-dependent (Pinckard et al., 1985), is rapid in onset, and is maximal 30-60 sec after injection (Doebber at al., 1985). Tachyphylaxis does not occur (Doebber at al., 1985). Many studies have ruled out renin, vasopressin, angiotensin II inhibition, central mechanisms and a-adrenoceptor blockade as factors contributing to PAF-mediated hypotension (Kamitani et al., 1984; Rabinovici et al., 1991). However, recent data suggest that the combined actions of PAF in the heart, microcirculation and vasculature play a role in the hypotensive effect. At the level of individual organs PAF alters cardiac and renal function, produces pulmonary edema, pulmonary arterial hypertension, pulmonary and circulatory anaphylaxis, and is associated with many inflammatory diseases (Williams and Higgs, 1988; Goldstein et al., 1991). Therefore, attempts have been made to implicate PAF in a great many disease states. 1.3.1 Systemic Activity Studies have characterized the actions of PAF on blood vessels in disease states. Initial hypotensive studies were performed in normotensive and spontaneously hypertensive rats (Caillard et al., 1982). PAF dose-dependently decreases blood pressure (BP) which is not associated with 23 tachycardia or modified by anticholinergics, antihistamines, 6-blockers or cyclooxygenase inhibitors. However, perfusion of low-doses of PAF into organs produces vasodilation. This has been demonstrated in the coronary circulation of pigs (Feuerstein et al., 1983) and dogs (Fiedler et al., 1989). Mesenteric blood vessels were shown to produce biphasic responses to PAF. Vasodilation occurred at a low dose while at high doses the vessels were constricted. These effects were blocked with the naturally occurring PAF antagonist BN 52021 (Siren and Feuerstein, 1989). Blood vessels constricted by PGF^a vasodilate in response to PAF when the endothelium is intact but remain constricted when the endothelium is removed (Kamitani et al., 1984). More recently, Chiba et al. (1990) have shown that PAF-induced nitric oxide release from endothelial cells produces relaxation of norepinephrine-contracted mesenteric artery strips. Despite the above actions, other studies report PAF-induced vasoconstriction in both in vitro and in vivo preparations. Bessin et al. (1983) reported that PAF produces a dose-dependent increase in portal and pulmonary vein hypertension in anesthetised dogs together with an increase in TXB2 and 6-keto-PGFla (the PGI2 metabolite)-features of hypovolemic acute circulatory collapse. Goldstein et al. (1986) showed similar vasoconstriction was induced by PAF in pig pulmonary vessel preparations while Piper and Stewart (1986) showed coronary vasoconstriction in 24 isolated guinea pig hearts. Some models show that TXA2 and peptido-leukotrienes play a pivotal role in mediating PAF effects on vessels, either vasoconstriction or vasodilation (Feuerstein et al., 1982; Piper and Stewart, 1987). Biochemical elucidation of the metabolic pathway of PAF degradation supports the theory that eicosanoids are intimately involved in PAF effects. PAF increases microvascular permeability in all species studied (Sanchez-Crespo et al., 1982). Intraplantar injection of PAF into the rat paw (0.025-16 /xg/paw) produces dose-dependent oedema (Martins et al., 1987). Changes in capillary permeability result in decreased intravascular volume and hence haemoconcentration. These two events were considered to be coupled by Bessin et al. (1983) who suggested that the decreased intravascular volume contributed to hypotension and reduced cardiac output. However, Handley et al. (1984) found that hypotension is produced immediately and that the extravasation response takes 4-10 min to peak and several hours to reverse. The mechanism by which this alteration in capillary permeability occurs is not known, yet it is generally agreed that PAF does interact with endothelial cells. This may involve leukocyte adhesion and activation which then sets into motion a positive feedback signalling system in which PAF induces the release of eicosanoids and vice versa (Rabinovici et al, 1991) . The result is damage to blood vessels and the microcirculation. 25 1.3.2 Pathophysiological actions of PAF PAF has been implicated as having a role in many diseases, including anaphylaxis, asthma, arthritis, cardiac and cerebrovascular ischemia, encephalomyelitis, heart failure, ventricular arrhythmias, glomerulonephritis, liver cirrhosis, ocular keratitis, pancreatitis, ulcers, ischemic bowel necrosis, transplant rejection, cold urticaria, endotoxic shock, pregnancy, artherosclerosis and many inflammatory skin disorders (Braquet et al., 1987; Koltai et al., 1991). The associated diseases are grouped according to endogenous PAF production produced by, e.g. endotoxin, or via immune complexes. As an effective approach to elucidating the relationships between PAF and disease, pharmacological antagonists of PAF, including structural analogs and naturally occurring alkaloids have been investigated. 1.3.3 Cardiac actions of PAF Vasoconstriction and cardiac dysfunction due to immunological and non-immunological causes have been ascribed to the release of PAF as well as to a synergistic interaction of PAF with TXA2 and LTC4 during inflammatory and hypersensitivity reactions (Piper and Stewart, 1987). Biochemical evidence shows that PAF can be released from 26 many cell types, including injured cardiomyocytes (Janero and Burghardt, 1990). PAF induces a cardiac dysfunction which is similar to that seen in cardiac anaphylaxis (Benveniste et al., 1983; Levi et al., 1984; Piper and Stewart, 1986). The effects of PAF on cardiac contractility and coronary circulation have been assessed many times. However, considerable debate exists regarding the mechanisms involved because of the inconclusive nature of the experimental evidence. There have bben as many studies published implicating PAF in cardiac dysfunction due to various causes as there are those showing it has no actions. The latter studies tend to implicate the eicosanoids as mediators of cardiac dysfunction (Braquet et al., 1987). Actions in whole hearts The cardiac actions of PAF have been well defined using many in vitro and in vivo preparations from many species including guinea pigs (Benveniste et al., 1983; Piper and Stewart, 1986, 1987; Felix et al., 1990), rabbits (Aloatti et al., 1990), dogs (Kenzora et al., 1984) and rats (Piper and Stewart, 1986; Hu et al., 1991; Pugsley et al., 1991). These studies show that PAF effects are common to many different mammalian species. PAF is consistently reported to lead to dose-dependent coronary vasoconstriction and negative inotropism. 27 The above two cardiac actions of PAF have been associated with the release of peptido-leukotrienes and TXA2 (Piper and Stewart, 1986; Hu et al., 1987), The use of eicosanoid antagonists to block the actions of PAF or blockers to inhibit eicosanoid synthesizing enzymes has provided evidence for and against involvement of AA metabolites in the actions of PAF. The demonstration of the importance of eicosanoids in mediating PAF-induced cardiac dysfunction rests upon the specificities of antagonists which have yet to be fully confirmed (Braquet et al., 1987; Braquet and Godfroid, 1986). Piper and Stewart (1986) demonstrated that indomethacin inhibits LTC4 vasoconstriction in isolated hearts and that the LTC4/LTD4 antagonist FPL 55712 blocked PAF, leading to the idea that LT's play a role in mediating cardiac actions of PAF. However, Stahl et al. (1987) showed that PAF did not vasoconstrict the vascular smooth muscle of arteries, leading to the speculation that PAF may constrict only small resistence vessels of the coronary circulation. They found no detectable levels of LT's or TX's in coronary effluent of hearts treated with PAF. In a second series of experiments, Stahl et al. (1987) confirmed their initial observations by showing that OKY-1581, a thromboxane synthetase inhibitor, and LY-171,883, a LTD4 antagonist, did not prevent the increase in coronary perfusion pressure or decrease in coronary flow produced by PAF and that CV-6209, a PAF antagonist at platelets, did abolish the effects of PAF. 28 They concluded that PAF directly mediates its cardiac actions (Stahl et al., 1987). Interestingly, another study (Sybertz et al., 1985) showed that intracoronary injection of PAF reduced coronary flow by direct negative inotropism and not by changes in coronary vascular resistence. Hu et al. (1991) showed that PAF-mediated vasoconstriction is dose-dependent, so that high doses produce coronary vasoconstriction while low doses produce coronary vasodilation. Thus, PAF has a biphasic effect on coronary resistance. The biphasic action of PAF has also been observed following direct bolus injection into the coronary circulation of the pig heart in situ (Feuerstein et al., 1984). This biphasic nature is characterized by an initial increase in coronary blood flow due to coronary vessel dilation, followed by a prolonged decrease in coronary blood flow due to coronary constriction (Feuerstein et al., 1984; Ezra et al., 1987). Actions on isolated cardiac tissues Isolated cardiac tissue studies provide evidence for a direct negative inotropic effect of PAF. PAF caused a dose-dependent negative inotropism in electrically-paced bathed left atrium as well as in right ventricular papillary muscles from guinea pigs and humans. This negative inotropism was not blocked by indomethacin or FPL 55712 29 (Levi et al., 1984; Alloatti et al., 1986). The direct nature of PAF action was further substantiated by Robertson et al. (1987) who showed that only CV-3988, a PAF antagonist, blocked negative inotropism. Camussi et al. (1984), using paced left-ventricular papillary muscles, showed that PAF produced a transient positive inotropic effect which was preceded by an increased action potential duration (APD). This was followed by marked negative inotropism and a decreased APD which led to the conclusion that PAF interfered in a dose-dependent manner with slow calcium channels or changed trans-sarcolemmal calcium entry mediated by PAF receptors. An effect on calcium channels was verified by Tamargo et al. (1985) , Nakaya and Tohse (1986), and Diez et al. (1990) all of whom examined the electrophysiological actions of PAF on guinea pig papillary muscle, canine Purkinje fibers and spontaneously beating atria, respectively. PAF produced a dose-dependent increase in the amplitude and rate of maximum upstroke (Vmax) of Phase 0 of the action potential, decreased APD and hyperpolarized the resting membrane potential due to either activation of calcium influx through slow inward calcium channels or by decreasing calcium efflux. A biphasic effect on ventricular contractile force accompanied these effects on the action potential. Studies with calcium channel blockers such as verapamil showed that these PAF-induced changes in the action potential could be blocked (Tamargo et al., 1985). Robertson et al. (1988) failed to show that PAF 30 affects slow calcium currents since PAF did not prevent histamine-induced contractile effects in paced K+-depolarized ventricular papillary muscles. In an extension of these studies, Robertson et al. (1988) determined that the negative inotropic effect of PAF was associated with a shortening in the APD and a decrease in intracellular sodium currents (a^ua); hence, as ama falls, intracellular calcium may be lost via the Na+/Ca2+ exchanger, thereby reducing contractility. However, the decrease in APD may be due to PAF-induced increase in gK+ which would decrease a^ jja* Vornovitskii et al. (1989) abolished the depressor actions of PAF by blocking the outward flow of K+ ions with 4-aminopyridine, thereby causing an increase in calcium ion flow due to the increased duration of the action potential plateau. However, recent patch-clamp studies on guinea pig ventricular myocytes provided evidence that PAF inhibits *kl/ the inward rectifying background potassium channel current, primarily due to shortening of channel mean open time (Wahler et al., 1990). 1.3.4 Arrhythmogenesis The controversy over the direct vs. indirect actions of PAF continues in coronary occlusion and ischaemia-reperfusion studies where PAF has been implicated in arrhythmogenesis. Mickelson et al. (1988) showed that intracoronary administration of PAF during ischaemia-31 reperfusion worsened hemodynamic status and reduced contractility in ischaemic hearts. These authors concluded that the reduced cardiac contractility was due to release of TXA2 and peptido-leukotrienes. In ouabain-induced arrhythmias in guinea pigs, PAF markedly increased the severity of fatal ventricular arrhythmias, an effect which was not blocked by acetyl-salicylic acid, but which was completely abolished by the TXA2 antagonist BM 13-177 and the 5-lipoxygenase inhibitor, esculatin (Riedel and Mest, 1987). PAF itself was found to be released intravascularly during atrial pacing in patients with coronary artery disease (Montrucchio et al., 1986) and accumulation of PAF was detected in myocardia of baboons subjected to ischaemia (Annable et al., 1985). Pharmacological studies with several PAF antagonists support the possibility of direct effects of PAF on cardiac function. Wainwright et al. (1989) showed that the PAF antagonists SRI-63441 and BN 52021 markedly protected against ischaemia/reperfusion arrhythmias in dogs and observed that PAF was arrhythmogenic only in vivo. They concluded that PAF may interact with platelets and blood leukocytes and produce arrhythmias through microvascular thrombosis. The PAF antagonists BN 520739 and BN 52021 have also been shown to reduce myocardial infarct size in dogs and rabbits subjected to myocardial ischaemia and reperfusion (Schaer et al., 1991; Chakrabarti et al., 1991). CV-6209 protected rats subject to coronary occlusion against 32 the progression of ischaemic damage monitored as a loss of both free-amino-nitrogen and cathepsin-D (Stahl et al., 1988). Auchampach et al. (1991) observed the protective effect of RP 59227, a novel PAF antagonist, against ventricular arrhythmias in dogs subjected to coronary artery occlusion. These observations underlie the potential of PAF antagonists for the prevention and treatment of patients with myocardial infarction. They may provide a cardioprotective effect by preventing fatal arrhythmias attributed to PAF acting either indirectly or via the AA cascade production of eicosanoids or directly. 1.4 PAF binding sites and PAF receptors The nature of PAF effects including the fact that the natural stereoisomer (R) is the only active stereoisomer, the high potency (threshold often 1.0 pM) in eliciting responses in immunological cells, the desensitization to PAF and the specific inhibition of PAF by many antagonists all suggest the existence of specific binding sites (receptors) for PAF on or in target cells and tissues (Braquet et al., 1987; Peplow and Mikhailidis, 1990; Snyder, 1990; Venuti, 1990). The existence of saturable, high affinity PAF binding sites has been confirmed in binding studies using [3H]PAF. These high affinity sites have been found in human, rabbit, 33 and porcine platelets (Hwang et al., 1984; Innarea et al., 1984; Duronio et al., 1990), human polymorphonuclear leukocytes (Valone and Goetzel, 1983), human and guinea-pig lung membrane (Hwang et al., 1983, 1985; Dent et al., 1989) but not in rat platelets (Innarea et al., 1984). The presence of a low affinity binding site also was detected in studies with human platelets and polymorphonuclear leukocytes (Innarea et al., 1984). The PAF binding site (isolated from plasma membrane) is heat labile, and protease-sensitive. Platelets exposed to low concentrations of PAF for as little as 5 min exhibit marked desensitization (Venuti, 1990). This is characterized by a decrease in affinity of the high affinity binding sites for PAF and not by a reduction in the number of binding sites (Snyder, 1990). A large number of "specific" PAF antagonists have been developed for use in the treatment of diseases linked to PAF, such as asthma and inflammation. Two such antagonists (BN 52022 and WEB 2086) are in Phase II clinical trials (Hosford and Braquet, 1990). Many antagonists resemble PAF in structure, whereas others, such as CV-3988, do not. A computer-generated model of the PAF receptor has been developed (Braquet and Godfroid, 1986; Godfroid and Braquet, 1986) using data regarding the putative geometry and 3-D electrostatic potentials for many PAF antagonists which share spatial features and possess similar lipophilic and hydrophilic regions. This model has been aptly named the 34 "Cache-oreilles" or ear-muff model because two strong electronegative wells are 180° apart and reside at a constant distance from each other (Godfroid and Braquet, 1986). More recent studies with newly developed PAF antagonists and agonists (structurally similar to PAF) suggest that the receptor may be a multipolarized cylinder or, alternatively, have electronegative and electropositive sites with at least two hydrophobic zones (Godfroid et al., 1991; Lamotte-Brasseur et al., 1991). Molecular biologists have recently cloned the putative "PAF-receptor" (Honda et al., 1991). Isolated cDNA for the purported PAF receptor was expressed in Xenopus laevis oocytes. The PAF antagonists WEB 2086 and CV-6209 were shown to inhibit PAF-induced cellular responses of the cloned receptor. It was also shown that this receptor belongs to the superfamily of G-protein-coupled receptors and that a series of serine and threonine amino acids were present on the DNA sequence which may be sites for phosphate acceptance and hence desensitization (Honda et al., 1991). Additionally, use of the cDNA from human leukocytes shows that activation of the putative PAF receptor yields inositol triphosphate production and the injection of guanosine bisphosphate into oocytes inhibits a PAF-induced chloride current, thus providing some evidence of intracellular second messengers involved in translation of PAF responses (Nakamura et al., 1991). 35 Elucidation of the protein sequence of the PAF receptor may aid in the development of novel and selective antagonists against PAF. 1.5 Endotoxic Shock 1.5.1 Pharmacology of shock Endotoxic shock develops as a result of a severe gram negative bacterial infection. Although bacteria can produce exotoxins, it is now recognized that it is the bacterial cell wall components (endotoxins) which are responsible for the development of shock (Ulevitch, 1991). The lipid-A component of the LPS molecule conveys toxicity. Analogues of lipid-A (in which additions and/or deletions have been made of specific sugars and acyl residues) have been developed which reduce the activity of the parent endotoxin and behave as specific endotoxin antagonists (Glauser et al., 1991). The mechanism by which lipid-A is recognized by host cells or plasma proteins is largely unknown; however, recent biochemical studies show that there are several plasma membrane proteins which specifically bind to lipid-A. These pathways facilitate rapid stimulation of PAF synthesis by LPS and entry of this molecule into monocytes. When LPS is exposed to a variety of plasma proteins, it binds to many types, including one known as lipopolysaccharide-binding 36 protein (LBP) (Tobian et al., 1986; Schumann et al., 1990). LBP is a 60 kD serum glycoprotein which binds lipid-A of LPS forming a tight, high affinity complex which then binds to the surface of a monocyte via a receptor, recently identified as CD-14 (Wright et al., 1990). The result of the activation of the LBP/CD-14 dependent pathway is to enhance the sensitivity of the monocyte to LPS. The classical pathway by which LPS activates the humoral system involves complexes of antibodies and cell-wall components. This complement pathway has gained in interest since the anaphylatoxins, C3a and C5a, have been shown to produce the characteristics of septic shock, such as hypotension and increased vascular permeability (Glauser et al., 1991). Neutrophil activation also occurs (see review by Malech and Gallin, 1987). Activated neutrophils may adhere to vascular endothelium or can aggregate, releasing vasoactive AA metabolites and granular enzymes (Hack et al., 1989). These activated neutrophils contribute to tissue and vascular injury via their binding to specific integrin receptors on endothelial cells and adhesion molecules, or selectins, on inflammatory cells (reviewed by Springer, 1990). Hypotension in septic shock may be caused by release of endothelium-derived relaxant factor (EDRF), nitric oxide (NO), AA metabolites or the complement system (Julou-Schaeffer et al., 1991; Rees et al. 1990). Urinary excretion of nitrate, the metabolite of NO, is increased 37 both in animals given E. coli LPS and in patients during infection (Julou-Schaeffer et al., 1991). LPS is also known to stimulate macrophages and neutrophils via endotoxin-activated complement (C5a), in addition to endothelial cells and isolated myocytes (Grisham et al., 1988; Schulz et al., 1991). However, the importance of NO in the pathophysiology of shock is not completely understood. Advances in the pharmacological treatment of shock have occurred over the last 10 years. These include the use of novel, and more powerful, antibiotics, steroids and monoclonal antibody (mAb) therapy. Combination antibiotic therapy with benzylpenicillin and flucloxacillin provides the most effective regimen in the treatment of endotoxic shock (Sanderson, 1990). However, it is now argued that treatment with antibiotics may aggravate the situation by causing lysis of bacterial cell walls and thereby enhance the release of LPS and subsequent release of various toxic intermediates (Shenep and Morgan, 1984). Corticosteroids are immunosuppressive, antipyretic and anti-inflammatory. Clinically, these actions provide patients with subjective improvement; however, this "feeling" of improvement does not translate into a clinically significant improvement. Many clinical trials have determined the efficacy of dexamethasone and methylprednisolone in endotoxic shock, and conclude that steroids only delay death but do not reduce overall 38 mortality (Sprung et al., 1984; Bone et al., 1987). The latter concluded that patients on steroids have significantly more secondary bacterial infections. Hoffman et al. (1984) reported a 10% mortality rate in patients treated with dexamethasone compared to 39% in controls. This is the only favorable report to date for use of steroids in endotoxic shock. As a result of such data, there have been no recommendations for use of high-dose steroids in the treatment of septic shock (Kass, 1984). The development of primary immunization against gram negative bacteria began with studies showing that when a polyclonal antiserum to an E. coli mutant (J5) was administered prophylactically, it reduced the mortality in infected patients (Baumgartner et al., 1985). Antibodies against both the lipid-A and core glycolipid of endotoxin have been developed. In a study involving E5, a murine IgM anti-lipid-A mAb, there was a significant decrease in mortality amongst patients in endotoxic shock states (Cohen and Glauser, 1991). HA-1A is a human mAb developed against the core glycolipid of endotoxin. In a study designed similarly to that for E5, HA-1A also reduced mortality in gram negative endotoxic shock (Ziegler et al., 1991). Despite these promising therapeutic agents, there is some doubt as to the specificity of these antibodies since immunoglobulins tend to bind non-specifically and, most importantly, these agents have not been shown to actually neutralize endotoxin (Wolff, 1991). 39 1.5.2 Cardiovascular dysfunction LPS administration produces profound changes in the cardiovascular system, including pulmonary hypertension, increased vascular permeability and chronic systemic hypotension, associated with a decreased vascular resistance and cardiac failure. In chronically catheterized conscious rats, LPS administration transiently decreased mean arterial blood pressure (Doebber et al., 1985; Hosford and Braquet, 1989). LPS also impairs vascular reactivity to pressor agents such as norepinephrine. Pithed rats respond similarly to conscious rats except that they are more prone to developing hypotension (Gray et al., 1990). The mechanism of both impaired vascular reactivity and heightened sensitivity to hypotension by LPS has not yet been defined. Endotoxin may exert its effects directly via modification of membrane phospholipids in response to PLP-A2 activation (Terashita et al., 1985), alterations in membrane enzymes or via receptors, as has been observed in human monocytes (Ulevitch et al., 1986). Effects of endotoxin administration on the Na+/Ca2+ exchanger in canine cardiac sarcolemma have been characterized (Liu and Xuan, 1986). Endotoxin alters the stoichiometry of the membrane pump from 3(Na+):l(Ca2+) to 2(Na+):l(Ca2+), which may contribute to the cardiac dysfunction during endotoxic shock. The exchanger is a protein situated in the phospholipid membrane; therefore, 40 disruption of the membrane could affect the integrity of this protein. LPS elevates PLP-A2 activity during shock. In dogs, activation of this enzyme may impair the myocardial Na /Ca^ exchanger via modification of the membrane lipid environment. Endotoxin stimulation of human monocytes may occur via a receptor dependent mechanism since specific binding of picomolar concentrations of many endotoxins has been detected (Ulevitch et al., 1986; Williams et al., 1990). Studies by Innarea et al. (1985) showed that a dose-dependent increase in vascular permeability was associated with the appearance of PAF in endotoxic rats. Doebber et al. (1985) also reported significantly elevated blood levels of PAF shortly after i.v. administration of endotoxin. These findings suggest that PAF antagonists may be useful in endotoxic shock, and such is the case. In guinea pigs injected with S. typhimurium endotoxin, both CV-3988 and WEB 2086, dose-dependently inhibited lethality produced by the endotoxin (Terashita et al., 1985; Casals-Stenzel, 1987). Such findings have been confirmed in the dog and rat (Braquet and Hosford, 1990). 1.6 Objectives Previous studies in this laboratory have shown that endotoxin-activated macrophages produce substance(s) which cause cardiac dysfunction (Salari and Walker, 1989) in 41 isolated rat hearts. Subsequent analysis showed Platelet-Activating Factor (PAF) to be one of the substances released. We, therefore, began to investigate the actions of PAF on isolated rat hearts. The literature concerning the actions of PAF on rat hearts is limited (Piper and Stewart, 1986). Most studies have involved sensitized guinea pig hearts and the immunological production of PAF by exposing the heart to an antigenic stimulus. We chose to examine the effects of PAF perfused directly into the heart. The questions asked included the following: 1. Are the cardiac actions of PAF the same as those of the substances released by the interaction between endotoxin and macrophages? 2. What are the cardiac actions of PAF in vivo and in vitro? 3. Are any of the cardiac actions of the material released by the interaction of endotoxin with macrophages mimicked by endotoxin alone? A short summary of the sequence of experiments conducted in this thesis can be found in Table I. Table I Table showing the order and purpose of experiments performed 1. Does PAF mediate endotoxin effects? Endotoxin dose-response curves To determine whether endotoxin produces a similar dose-dependent effect as PAF on cardiac contractility and coronary flow in vitro 3. Endotoxin and ibuprofen To assess endotoxin action in vivo and determine whether these can be blocked by ibuprofen What are PAF actions on the heart and cardiovascular system? PAF dose-response curves in vitro PAF dose-response curves in vivo PAF and contractility PAF and coronary flow To assess the actions of PAF on contractility and coronary flow in isolated hearts To compare the cardiac effects of PAF in vivo with those in vitro To determine whether or not PAF reduces cardiac contractility independently of coronary flow To determine whether PAF produces coronary vasoconstriction independently of contractility What is the mechanism of PAF action? Actions of antagonists PAF and ibuprofen PAF and Methylcarbamyl-PAF PAF toxicity in mice PAF and arrhythmias To examine the effects of PAF and eicosanoid antagonists on PAF action in the isolated heart To assess whether ibuprofen is as effective in vivo as in vitro To compare PAF actions to a stable, acetylhydrolase resistant PAF analog in vivo To determine the LD50 value of PAF for use in assessing novel PAF antagonists To assess whether PAF produces arrhythmias to 43 2 Methods 2.1 Cardiac preparations The accurate assessment of the pharmacological actions of drugs on the heart and cardiovascular system in vivo and in vitro requires an understanding of the dose-response relationship between the drug under investigation and tissue response. These studies were performed to ascertain the ED50 of PAF and its toxicity on the heart and cardiovascular system of the rat. This was done as is described below. 2.1.1 Isolated rat hearts The isolated perfused warm-blooded heart has many advantages which lends itself to study of the actions of drugs on both mechanical and electrical properties of the heart. The isolated heart is a simple preparation with which to screen for the cardiac actions of drugs as first demonstrated by Langendorff in 1895. This preparation resolved many problems encountered when using preparations which involved blood. It allowed, for the first time, the use of a physiological solution to maintain normal heart function. Obvious problems with preparations using blood and blood-borne constituents include clotting and frothing. In the Langendorff heart, a filtered Krebs-Henseleit solution is used to replace the blood. The isolated heart 44 is also free of both neural and humoral factors which may alter drug activity (Doring and Dehnert, 1988). Wiggers first critically appraised the Langendorff isolated heart method in 1909. More recently, Broadly (1979) carefully discussed the advantages and limitations of the method. For example, he showed that the perfusate solution had a low oxygen-carrying capacity and that a lack of patency of the aortic valves (and the ease with which they are damaged) allowed perfusion fluid to enter and distend the left ventricle. We have, therefore, developed a perfusion apparatus which reduces or eliminates some of these problems. Perfusion apparatus A modified perfusion apparatus for the study of the actions of drugs on the mechanical and electrophysiological behaviour of hearts from small animals such as rats and guinea pigs was developed in this laboratory (Curtis et al., 1986) . Nine chambers (each of a 250 mL capacity) were machined into a plexiglass block and placed into a bath containing circulating warm water (37°C- maintained by an external heater). Krebs-Henseleit perfusate from within individually controlled chambers flows via separate silastic tubes to a common manifold and then to the aortic cannula. Since total dead-space for each chamber is less than 0.1 mL this allowed for a rapid switching of perfusate while an 45 external carbogen (5% CO2 in O2) gas mixture maintained an aortic root pressure between 100-125 mmHg, as reguired. Concentration-response studies for PAF and nifedipine Male Sprague-Dawley rats (300-400 g) were killed by a blow to the head, exsanguinated and the heart rapidly removed from the chest cavity. Hearts were perfused with 5 ml of ice-cold Krebs-Henseleit solution to remove remaining blood prior to being mounted, via an aortic cannula, on the modified Langendorff perfusion apparatus (Curtis, 1986). Within two minutes of sacrifice, hearts were perfused with an oxygenated Krebs-Henseleit solution at 37°C and pH 7.4. The composition (mM) of the Krebs-Henseleit solution was: NaCl, 118; KC1, 4.74; CaCl2«2H20, 2.5; KH2P04, 0.93; NaHC03, 25; D-Glucose, 10; MgSO4«7H20, 1.2. Bovine serum albumin (BSA) (Sigma Chem. Co.) (0.25%) was added to the Krebs-Henseleit solution as a carrier in all studies involving PAF. The left atrium was removed and a small compliant balloon made of plastic wrapping film (Saran Wrap) inserted into the left ventricle and adjusted to give an initial left ventricular end-diastolic pressure of 10 mmHg (balloon volume was approximately 0.5 mL) . The aortic root of the heart was perfused at a constant pressure of 100 mmHg. Ventricular pressure was measured by a pressure transducer 46 and a Grass Polygraph while the maximum rate of intraventricular pressure development (+dP/dtmax) was obtained by differentiating left ventricular pressure using a Grass Polygraph differentiator (model 7P20C). The ECG was recorded from the epicardial surface of the heart with atraumatic, silver-ball electrodes (Curtis, 1986) placed on the right atrium and left-ventricle, i.e., approximating a Lead II configuration. The variability in the rat ECG makes determination of drug effects difficult, since the P-, QRS and T-waves do not share a common baseline (Detweiler, 1981). We have, therefore, adopted the commonly used reference marker as the point at which the P-R interval terminates and the QRS interval begins (since this is subject to the least variation) as the determinant of the ECG isoelectric line (Driscoll, 1981). Recordings were made on a Grass Polygraph (model 7D) at a bandwidth of 0.1-40 Hz. Coronary flow perfusate was measured by collecting effluent at one minute intervals in a graduated cylinder. Changes in flow were expressed as percent of control. Therefore, normalization of coronary flow for heart weight was not necessary. All hearts were of similar weight (l.l-1.4g) and normalization for weight did not improve the data expressed as percent of control. Platelet Activating Factor (l-0-hexadecyl-2-acetyl-rac-glycero-3-phosphocholine) was solubilized in 70% ethanol and serial dilutions prepared in Krebs-Henseleit solution. The hearts were perfused with Krebs solution for 15 min prior to 47 administration of PAF at concentrations of 0.0001 /zM to 1 /LtM (using 5 concentration steps) for a period of 2 min at each concentration. The exposure time was chosen as that during which a steady-state response to PAF occurred. Nifedipine concentration-response curves were constructed in a manner similar to that for PAF. A stock solution of nifedipine was prepared by dissolving the drug in a Krebs-Henseleit/ethanol (3:1) solution. This was added to perfusate to produce suitable concentrations for concentration studies. The nifedipine concentration used in further studies was that which produced maximal vasodilation without negative inotropism. Suitable vehicle controls had no effects during the experimental time period. PAF and Eicosanoid blockers Experiments with blockers were performed by adding each antagonist to the perfusate for a period of 5 min before adding 1 /iM PAF for 2 min. All antagonists were dissolved in saline and gently warmed. Antagonists were chosen on the basis of a literature survey and the concentrations tested were ones which could be expected to produce the specific actions for which they were developed. In view of the fact that the actions of 1 JUM PAF were essentially irreversible under the experimental conditions, we were forced to adopt a strategy of comparing responses in groups of hearts treated with PAF plus antagonist with those 48 in a control group treated with PAF alone. Thus, each heart did not act as its own control. Altogether, 14 groups (each consisting of 5 hearts) were examined. PAF's action on contraction independent of flow Concentration-response curves for nifedipine were constructed in a similar manner to that for PAF. Nifedipine is a vaso-selective calcium channel blocker and concentrations exist which maximally dilate coronary arteries without producing negative inotropism. It was chosen in an attempt to examine PAF action on contractility independent of coronary flow. A stock solution of nifedipine was prepared by dissolving the drug in a Krebs/ethanol (3:1) solution. This was added to perfusate to produce suitable concentrations for dose-response studies. The drug was protected from light at all times by storing stock solutions in amber bottles and placing aluminum foil around the chambers of the heart perfusion apparatus. PAF action on coronary flow To examine the actions of PAF on coronary flow in the absence of effects on contractility, isolated rat hearts were prevented from contracting by using either a low concentration (1/xM) of the sodium channel blocker, 49 tetrodotoxin (TTX) in the presence of a high concentration (24 mM) of K+, or a high concentration of TTX (100/iM) in the presence of a low concentration (lOmM) of K+. No differences were seen between the two regimens; both abolished contractions and reduced coronary flow. To compare the efficacy of PAF in reducing coronary flow, IJUM PAF was compared to two potent coronary vasoconstrictors, namely ergonovine (3/xM) and vasopressin (6.3/xM) in the presence of a high K+/TTX solution. Hearts were exposed to either of the TTX regimens for 10 min prior to the addition of PAF. E. coli endotoxin studies Although PAF has been implicated as a major mediator in shock, the mechanism by which endotoxin produces a state of shock is unknown. Therefore, we examined the effects of Escherichia coli lipopolysaccharide endotoxin on isolated hearts to see whether endotoxin produces the same actions as PAF. As above, isolated rat hearts received 0.1, 0.5, 1.0, 5.0, 10.0 and 50/xg/mL concentrations of endotoxin. The ECG and contractility of the heart were monitored as above. 2.1.2 Statistical analysis All studies were performed according to a randomized block experimental design. This block design allows for 50 homogeneity in the experiment as well as control of variability which may arise from experimental error. This type of error in experimentation is quite common, as it reflects a combination of both random error and variability. The random block design essentially reduces the variability in the system and hence decreases experimental error (Montgomery, 1984). Since the only randomization that occurs is confined to the treatments within the blocks, the blocks represent a restriction on randomization and thus yields a simple linearly additive statistical model (Li, 1964). Experiments with such a design lend themselves to the Analysis of Variance (ANOVA) statistical test, particularly as this test demands that treatments are from as uniform an environment as possible (Li, 1964). ANOVA allows one to compare many treatments, thus making it a most useful statistical test (Zar, 1984; Gad and Weil, 1988). Therefore, all statistical significance was determined at an a level of 0.05 using the General Linear Model ANOVA (GLM ANOVA) from the NCSS Statistical Package (Hintze, 1981). A post hoc or multiple comparison test is one which is performed when the experimenter would like to determine which treatment means differ after ANOVA. A large number of tests exist for this purpose; we chose Duncan's multiple comparison test (Duncan, 1955). This test compares groups of continuous and randomly distributed data of equal sample 51 size and is a powerful test for detecting differences between means (Montgomery, 1984). For data which did not conform to a normal distribution, a "variance-stabilizing transformation" was applied (Montgomery, 1984; Zar 1984). Such nonconstant variance occurs with VPB data. Data were, therefore, logarithmically (base 10) transformed (Winkle, 1979) to bring both the variance and error distributions closer to normal (Zar, 1984). 2.2 Intact rat studies The cardiovascular actions of PAF, especially the potent hypotensive properties which occur in all species, have been most intensely studied and characterized in acutely prepared anaesthetised rats (Goldstein, 1991). The rat provides reliable, accurate cardiovascular measures (Schroeder et al., 1981), and is routinely used in drug investigation. As a small animal, it is relatively inexpensive, readily available, and provides reproducible results. An extensive biochemical, physiological and anatomical data base exists for this species. Furthermore, the rat is amenable to surgical intervention and the study of drug actions on various organs and/or systems (Curtis et al., 1987). However, like many models it has drawbacks; as with many animal models, it is very unclear as to exactly how crucial models relate to disease in man. However, this 52 does not detract from the usefulness of the rat in the pharmacological characterization of drugs. It is no less valid an animal model than any others which have been used to assess drug actions in the heart and cardiovascular system (Curtis et al., 1987). 2.2.1 Surgical preparation Male Sprague-Dawley rats (150-350 g) were used in accordance with the guidelines established by the University of British Columbia Animal Care Committee. Rats were anaesthetised with sodium pentobarbital (60 mg/kg, i.p.). All animals had their right jugular vein and right carotid artery cannulated for administration of drugs and blood pressure monitoring, respectively. The ECG was recorded using a special lead configuration. The superior needle electrode was placed 0.5 cm from the midline of the trachea at the level of the right clavicle while the lower needle electrode was placed 0.5 cm from the midline at the level of the 9th and 10th ribs (Penz et al., 1992). The trachea was cannulated for artificial ventilation at a stroke volume of 10 mL/kg and rate of 60 strokes/min to ensure adequate blood-gas levels (MacLean and Hiley, 1988). Animals were placed in a supine position and body temperature was monitored by rectal thermometer and maintained between 37-38°C with a heating lamp. Blood pressure and ECG were 53 recorded on a Grass polygraph (model 7D) at a bandwidth of 0.1-40 Hz. and a chart speed of 100 mm/sec. 2.2.2 Design and analysis Cumulative in vivo dose-response curves for PAF were obtained in artificially-ventilated, pentobarbital anaesthetised rats over the dose-range of 0.5 to 20 /xg/kg, i.v.. PAF was first solubilized in 70% ethyl alcohol as a stock solution and serial dilutions were made in saline vehicle containing 0.25% BSA. Since PAF is a lipid, it tends to adhere to glassware and syringes; therefore, BSA was used as a carrier for PAF. Animals (n=5) were randomly assigned to receive either PAF or vehicle control at the end of a 15 min control period. All doses were injected i.v. over 30 sec and blood pressure, heart rate and ECG were recorded 2 min later, immediately prior to addition of the next dose. In a separate experiment, and according to a random and blind design, either vehicle or ibuprofen (5 mg/kg) was given 5 min prior to either PAF (0.5 /xg/^ 9) or E. coli endotoxin (10 mg/kg). Again, blood pressure, heart rate and ECG were recorded 2, 5, 10, 15, 20, and 25 min after and immediately prior to addition of the next cumulative dose. In a final series of studies in vivo dose-response curves examined the effects of methylcarbamyl-PAF (a PAF derivative in which the sn-2 acetyl moiety of the lipid has 54 been modified) on blood pressure, heart rate and the ECG. Animals (n=6) were treated, via a random block design, with either vehicle, PAF or methylcarbamyl-PAF over the dose-range of 0.5-20 /ig/kg, i.v. bolus infusion over 30 sec. Blood pressure, heart rate and the ECG were recorded 2 and 5 min after administration. No significant changes occurred in any measurements between 2 and 5 min of drug administration, hence, all doses were given cumulatively. Accurate analysis of drug-induced changes in the rat ECG presents certain difficulties. Driscoll (1981) exhaustively outlines many anomalies which occur with the rat ECG, as outlined above. Briefly, the P, QRS and T waves do not share a common baseline and therefore a reference point for determination of the isoelectric line is required. Driscoll (1981) suggests that this point is that at which the P-R interval terminates and the QRS complex begins since it is subject to the least variation. In association with this, we have also adopted the ECG measures of Budden et al., (1981) in an attempt to maintain interlaboratory consistency. Another important factor which influences the rat ECG is electrode position. In all our studies of acutely prepared anaesthetised rats, the positioning of electrodes was as explained previously. Admittedly, there will always be some anatomical differences in the position of the rat heart but we were able to minimize these variations with our method of electrode placement. 55 2.2.3 Evaluation of a new ECG,measure (RSh) Before starting our studies with PAF we had developed a novel ECG measure (RSh) for the detection of possible sodium channel blockade in artificially ventilated, anaesthetised rats. This measure was of value in the PAF studies since much controversy exists as to whether PAF receptors are linked to either sodium (Robertson et al., 1988) or calcium channels (Camussi et al., 1984). Conventional measures of sodium channel blockade (QRS complex widening and/or P-R interval prolongation) are limited in their sensitivity and incapable of detecting small effects of PAF. The new measure, termed "RSh" or RS-height, quantifies the height from the peak of the R wave to the bottom of the S wave. It is more sensitive to sodium channel blockade than conventional measures (Penz et al., 1992). In order to illustrate this, we compared the ECG effects of various Class I sodium channel blockers with other antiarrhythmics. Representative drugs from the three subclasses of Class I, i.e. quinidine, lidocaine and flecainide, were tested. In each case, changes in RSh occurred before changes in QRS or P-R. The other antiarrhythmics (Class II, Class III and Class IV) only influenced RSh if they had known sodium channel blocking properties and then only at high doses, e.g. propranolol (Class II) and tedisamil (Class III). Other physiological manoeuvres, such as changing vagal activity, administration of catecholamines, or direct pacing 56 of the right atrium, did not change RSh. Thus, RSh is a useful measure with which to detect possible sodium channel blocking actions, in rats, of cardiovascular drugs. All cumulative dose-response curves for PAF in vivo were analyzed for changes in the RSh measurement in addition to P-R and QRS intervals. 2.3 Toxicological studies with PAF 2.3.1 Lethal-dose curves in mice In order to assess the safety of a substance adequately it is necessary to have both a quantifiable method of measuring and a precise means of expressing the observed toxicity (Klassen and Doull, 1980). For this study we chose death as the toxic end-point for measurement. Lethality was chosen because it provides a measure of comparison amongst many different substances which act at markedly different sites through presumably different mechanisms. Lethality is a precise, quantal and unequivocal measure. With such a conclusive end-point, we are able to apply this method for use in quantal dose-response curves to obtain the lethal dose (LD50)• Quantal dose-responses for effects such as lethality exhibit a normal gaussian distribution and hence if the percentage of animals that die at each dose is examined the data should approximate a normal frequency distribution. This distribution arises because of normal 57 biological variation in the susceptibility of the animals to the drug being tested. Data obtained from dose-response curves for lethality were subject to probit analysis. Briefly, since quantal dose-response phenomena are normally distributed, one can convert the percent response to units of deviation from the mean or normal equivalent deviations (NED) (Gad and Weil, 1988). Bliss (1935) suggested that units of NED be converted by the arbitrary addition of 5 to the value to avoid negative numbers and that these be called probit units. In essence, what is accomplished is an adjustment of mortality to an assumed normal population distribution. In an attempt to determine the lethal dose (LD50) of PAF for use in future studies with PAF antagonists, we constructed a dose-mortality curve for PAF in Swiss CDi mice. 2.3.2 Design and analysis Eight week old male Swiss CD^ mice (30-40 g) were randomly assigned to 5 groups, each of which would receive a single dose of PAF or its vehicle. The doses of PAF given were 100, 400, 800 and 1200 jug/kg. Each animal (n=6 mice/group) was injected (total volume of not more than 0.2 mL) via the lateral tail vein with either PAF or vehicle and monitored for 90 mins following injection. Animals unsuccessfully injected were excluded and immediately 58 replaced. For each dose, the percent mortality (number of animals which died per group) was determined (at 90 min post-injection). 2.4 Ischaemia-induced arrhythmias A novel PAF antagonist, RP 59227 (a pyrrolo-[l,2-c] thiazole derivative) (Lave et al., 1988), was examined as a potential antiarrhythmic. Previous studies showed RP 59227 to be a competitive antagonist of PAF in rabbit, dog and human platelets (Cavero et al., 1989). In studies involving coronary occlusion and reperfusion in dogs, the drug was shown to be effective in reducing arrhythmias when given at a dose of 2.5 mg/kg i.v. (Floch et al., 1991). In view of this evidence, we examined the antifibrillatory actions of RP 59227 in acutely prepared anaesthetised rats. 2.4.1 Surgical preparation in acute studies The surgical procedures used were similar to those previously employed by Au et al. (1979) and Paletta et al. (1989). In brief, rats were initially anaesthetised with pentobarbitone (60 mg/kg, i.p.) and supplemental doses (6mg/kg) were given i.v. when necessary to ensure an adequate level of anaesthesia. The trachea was cannulated and all animals were artificially ventilated. The left carotid artery was cannulated for measurement of mean 59 arterial blood pressure and withdrawal of blood samples for determination of serum K+ concentrations (Ionetics Potassium Analyzer). The right jugular vein was also cannulated for administration of drugs. The thoracic cavity was opened and a polyethylene occluder placed loosely around the left anterior decending coronary artery. The chest cavity was closed and body temperature was maintained between 35-37°C using a surgical lamp. In order to obtain the best ECG signal for detection of S-wave changes, needle electrodes were placed subcutaneously along the suspected anatomical axis (right atrium to apex) of the heart determined by palpation. The superior electrode was placed at the level of the right clavicle about 0.5 cm from the midline, while the inferior electrode was placed on the left side of the thorax at the point of maximum impulse approximately 0.5cm from the midline and at the level of the ninth rib. The animal was allowed to recover for 30 min prior to drug administration. 2.4.2 Design and analysis During the 30 min allowed for the animal to recover from open chest surgery, RP 59227 was prepared as a 50 mg/mL stock solution by solubilizing it in acidified polyethylene glycol (PEG400) containing the antioxidant a-tocopherol. 60 The solution was bubbled with nitrogen and stored under nitrogen during all experiments. A random and double-blind experiment was performed (n=6) in which animals received either vehicle, or RP 59227 (at either 10 or 25 mg/Jcg) as an i.v. infusion. A control trace (taken at a chart speed of 100 mm/sec) was recorded 15 min before occlusion and 1 min prior to drug administration. Drugs were infused at a rate of 0.09 ml/min and traces were taken at 1 min intervals over a period of 5 min, post-infusion. A blood sample (approximately 0.25 mL) was then taken. Thereafter, the occluder was pulled so as to produce coronary artery occlusion. ECG, blood pressure, heart rate, arrhythmias and mortality were monitored for 30 min after occlusion. Arrhythmias were identified as ventricular premature beats (VPB), ventricular tachycardia (VT) or ventricular fibrillation (VF) and the number of each was recorded and expressed as an arrhythmia score (AS) as described by Curtis and Walker (1988), see discussion below. At the end of the 30 min period, if the animal survived, a second blood sample was taken. After death, hearts were removed and perfused by the Langendorff technique (Langendorff, 1895) with Krebs-Henseleit solution to wash out all remaining blood. This was followed by perfusion with saline containing 1 mg/ml indocyanine (Fast green dye, BDH) for 60 sec which revealed the underperfused and occluded zone (zone-at-risk). The 61 occluded zone was then cut away from the non-occluded ventricular tissue, blotted, and weighed on an analytical balance. The occluded zone was then expressed as a percentage of the total ventricular weight. 2.4.3 Pre- and post-occlusion ECG changes ECG traces were taken over the time interval of 1, 2, 5, 10, 15 and 30 min post-occlusion. Prior to both drug administration and occlusion, the ECG showed a positive ST-segment with respect to the isoelectric baseline, as defined above. This allowed the monitoring of signs of drug-mediated changes in the RSh measure, indicative of sodium channel blockade in the heart. Immediately after ligation of the coronary artery, many changes occur in the ECG. There is a rapid increase in the size of the ECG signal, particularly a large increase in the R-wave amplitude. This gradually reduces to baseline (pre-occlusion) with time (Johnston et al., 1981; Kane et al., 1981). ST-segment elevation follows an initial depression, and this is maintained for the duration of the experiment (Curtis, 1986). It is expressed as a percentage of the R-wave amplitude. Associated with these changes in the ECG is the later development of Q-waves. The appearance of Q-waves occurs 1.8±0.4 hrs (Walker et al., 1991) post-occlusion and is usually associated with fatal ventricular arrhythmias. 62 2.4.4 Analysis of arrhythmias The analysis of ischaemic arrhythmias produced by occlusion of the coronary artery is complex. Statistical analysis depends upon how the arrhythmic data is both categorized and treated. We have developed a classification scheme by which to summarize the arrhythmic history in one single value- the Arrhythmia Score (Curtis and Walker, 1988). Arrhythmia appearance is biphasically time-dependent (Johnston et al., 1981). In the following studies, we quantified the severity and incidence of arrhythmias during the initial phase (5-15 min post-occlusion). Arrhythmias were categorized according to guidelines established by the Lambeth conventions (Walker et al., 1988). Ventricular premature beats (VPB) were defined as single QRS complexes which occurred before any identifiable P wave. Doublets (bigeminal) or triplets (trigeminal), variations in the single complex, were not classed as distinct arrhythmias but rather were pooled (Howard, 1990). Ventricular tachycardia (VT) was defined as 4 or more consecutive VPB's and not subclassified according to rate. We were also able to classify VT by its characteristic change in BP, usually a sharp fall in mean BP. Ventricular fibrillation (VF) was defined as a chaotic ECG pattern in which no distinguishable QRS complexes could be discerned accompanied by a precipitous fall in blood 63 pressure to less than 10 mmHg. Animals were not defibrillated by thump-version (Curtis, 1986). If VF was irreversible the animal died. All arrhythmic incidences were recorded as above and summarized according to the Arrhythmia Score (Curtis and Walker, 1988). Briefly, this is an arbitrary numerical grading of ventricular arrhythmias which considers their severity with time post-occlusion. 64 3 RESULTS 3.1 Actions Of PAF Cumulative, steady-state, concentration- or dose-response curves for PAF were constructed in isolated rat hearts over the concentration range of 0.0001-1.0/xM (n=15) and in vivo over the dose range of 0.5-20 /xg/kg (n=6) . 3.1.1 In vitro effects of PAF PAF effects on cardiac function (electrical and mechanical) were dose-dependent, as can be seen in Figures 1-3. Systolic pressure was decreased by 23±5% by 1 /xM PAF (maximal concentration) while diastolic pressure remained unchanged (Figure 1) . Lyso-PAF, an inactive metabolite of PAF used as a control, had no effects. The maximum rate of intraventricular pressure development (+dp/dtmax) was decreased 39 + 61 by 1 /xM PAF (Figure 2 A), while the maximal rate of intraventricular relaxation (-dP/dtmax) was not significantly affected (data not shown). Again, Lyso-PAF had no significant effect. Coronary flow (Figure 2 B) was also reduced by PAF in a dose-dependent manner such that the reduction reached 42±7% with 1 /xM PAF. The ventricular QRS complex was not altered by PAF treatment, whereas PAF prolonged the PR-interval of the ECG and reduced heart rate in a dose-dependent manner (Figure 3) . The PR-interval 65 increased 29±7% with 1 izM PAF while heart rate decreased 29±4% with the same concentration. The above results established the dose-dependent actions of PAF which were essentially irreversible in nature. Irreversibitity occurs in our preparation and is defined in terms of duration of PAF response in the heart. For isolated rat hearts the PAF response (at the highest concentrations) is produced and remains for the duration of the experiment, approximately one to two hours. 3.1.2 In vivo effects of PAF In vivo dose-response curves for PAF were obtained in artificially-ventilated, pentobarbital-anaesthetised rats. PAF dose-dependently reduced blood pressure and produced marked changes in the ECG. A dose of 0.5 /xg/kg PAF (ED50 value) reduced blood pressure to 48±5 mmHg (Figure 4) . Heart rate was not affected by PAF except at the highest doses (Figure 5) . However, the PR-interval of the ECG was maximally prolonged at a dose of 20.0 /xg/kg (Figure 6). Further analysis of the ECG revealed a slight elevation in the ST-segment, especially at the higher doses, but no change in the QRS interval was apparent. Ibuprofen, a cyclooxygenase enzyme inhibitor which attenuates the actions of PAF more effectively than any other eicosanoid or PAF blocker examined in vitro (see later) , was also tested in intact animals. At a dose of 5 66 mg/kg, i.v., ibuprofen pre-treatment, which itself had no cardiovascular actions, did not attenuate the hypotensive actions of PAF nor its ability to prolong the P-R interval of the ECG (Figure 7 A and B). 3.2 Actions of PAF in the presence of eicosanoid blockers In an attempt to verify whether PAF was acting directly, or via the release of mediator(s), a series of experiments were performed with various antagonists of leukotrienes, thromboxanes, PAF and the cyclooxygenase enzyme. The effectiveness, or lack thereof, of blockade was expressed as percentage reversal of control responses to 1 /xM PAF obtained in a separate group of hearts. It proved impossible, in the same heart, to record the actions of PAF both in the absence, and presence, of blockers since the actions of the test concentration of PAF (lfM) were essentially irreversible. Apart from the exceptions noted below, none of the antagonists themselves had actions on isolated hearts. As can be seen in Table II, both PAF antagonists and eicosanoid inhibitors only moderately attenuated the actions of IJUM PAF on the heart despite their being given at concentrations of 0.01 and 1.0/zM. Of the two PAF antagonists tested, Web 2086 was the most efficacious. The cyclooxygenase inhibitor ibuprofen was the most effective of all the drugs in attenuating responses to PAF. 67 Ono 3708 at a dose of 0.01/xM antagonized PAF effects by about 60%, but in the presence of 1/uM ONO 3708, ventricular fibrillation was induced within two minutes of adding PAF. To determine whether 5-Lipoxygenase products are involved as mediators of PAF action, we used the leukotriene receptor antagonists FPL-55712 and L-655,240. These were similar in action at both concentrations and gave only limited protection. For all inhibitor drugs shown in Table II, the pattern of protection varied with the response considered, e.g., pressure, heart rate, etc.. Sometimes the protection was dose-related, but often it was not. As indicated above, at the concentrations used, blockers alone had no action on hearts. 3.3 Actions of PAF on coronary arteries In order to identify actions of PAF on myocardial cells distinct from those on coronary smooth muscle cells, we examined coronary flow responses in non-contracting hearts. Two methods were used to prevent cardiac contractions: either high K+ and low TTX, or low K+ and high TTX. In the presence of either solution, PAF reduced coronary flow, but to a lesser extent than in contracting hearts (Figure 8) . In order to assess the relative efficacy of PAF in reducing coronary flow in non-contracting isolated hearts, the 68 actions of PAF were compared under the same conditions with two known coronary vasoconstrictors. Doses of ergonovine (3/zM) and vasopressin (6.3/zM) were chosen which would produce maximum coronary vasoconstriction. The vasoconstrictor efficacy of these agents reduced coronary flow to the same extent as 1/xM PAF (Figure 9) . Under the same conditions, lyso-PAF did not produce coronary vasoconstriction. 3.4 Actions of PAF on ventricular contraction In order to isolate the direct effects of PAF on myocardial contractile cells from those resulting from reduced coronary flow, we performed experiments in hearts whose coronary vessels were maximally dilated. Nifedipine dose-response curves were constructed and a dose of nifedipine chosen which would maximally vasodilate coronary vessels without producing negative inotropism (Figure 10) . Figure 11 shows that PAF administered in the presence of 0.03JUM nifedipine still reduced contractility although nifedipine did not abolish the effects of PAF on heart rate. This study also showed that nifedipine prevented PAF-induced coronary vasoconstriction. 69 3.5 PAF activity in rat and rabbit hearts The study in rat hearts described above showed that PAF, at a concentration of 1/xM, produced dramatic reductions in both contractility and coronary flow and that these actions were independent of each other. We examined the actions of PAF in two different species, rat and rabbit, in an attempt to determine whether PAF possessed similar activity on blood pressure, heart rate and the ECG. After 10 min of equilibration, each heart was perfused with 1/xM PAF for 2 min. PAF prolonged the PR-interval of the rat ECG by 29±7% (n=5) and in the rabbit by 16±4% (n=4). PAF also reduced the left-ventricular peak systolic pressure and coronary flow in the rat by 2 3 ±5% and 4 2 ±7%, respectively, while in the rabbit hearts these were reduced by 20±6% and 17±4% from control. 3.6 Effectiveness of methyl carbamyl-PAF vs. PAF Analysis of the cumulative dose-response curves for the comparison of PAF and MC-PAF, a long acting PAF agonist (half-life=60 min), (Figure 12) given over the range of 0.5 to 20.0/xg/kg showed that this analog produces very similar responses to PAF. Both produced a dose-dependent decrease in blood pressure (Figure 12 A) ; however, neither compound altered heart rate (Figure 12 B) . MC-PAF and PAF elevated the ST-segment of the ECG to the same extent. 70 3.7 Lethality of PAF in mice Probit analysis of the dose-lethality curves in mice yielded an LD5n of 400/xg/kg (Figure 13). Intravenous injection of PAF at its lethal dose produces a rapid rate of respiration, convulsions, spasms and finally death which occurred within 30 sec of injection. Sublethal doses of PAF produce slowed rates of respiration, docility and slight muscle twitch. 3.8 Cardiovascular, ECG and Antiarrhythmic actions of RP 59227, a new PAF antagonist In view of the evidence concerning the effectiveness of RP 59227 in dogs subjected to coronary artery ligation, it was decided to examine the antifibrillatory actions of this drug in our coronary artery occlusion rat model. Rats were randomly assigned to receive either RP 59227 (10 or 25mg/kg) or vehicle. When compared with vehicle, RP 59227 (lOmg/kg) transiently prolonged the P-R interval and exhibited no other ECG changes, while at the 25mg/kg dose both the Q-T and P-R intervals were moderately prolonged (Table III) . Blood pressure and heart rate were not altered by vehicle or the low dose RP 59227. However, at the high dose RP 59227 elevated blood pressure and reduced heart rate (Table III). After occlusion, all animals had VPB's, the frequency of which was not reduced by drug treatment at either dose 71 (Table IV) . Vehicle and both RP 59227 groups had a high (100% and 90%) incidence of VT; however, only the high dose RP 59227-treated group showed significantly reduced incidence of VF (from 100% in the control group to 30%). 3.9 Actions of E. coli endotoxin 3.9.1 In vitro effects of endotoxin in isolated hearts The exposure of rat hearts to E. coli lipopoly-saccharide endotoxin shows that there is a dose-dependent relationship between endotoxin dose and electrical and mechanical effects in isolated rat hearts. Hearts received concentrations of the E coli lipopolysaccharide endotoxin over the concentration range of 0.1-50/xg/ml (n=5 in each group). Only the highest concentration reduced heart rate significantly from the control of 266±7 to 236±6 beats/min (Table IV) . Endotoxin also prolonged the P-R and QRS intervals from 45±1 to 52±3 msec and 36±1 to 39±1 msec when given at the maximum dose of 50jug/mL, respectively. Left-ventricular peak systolic pressure was reduced from 111±5 to 86±8 mmHg; however, left-ventricular peak diastolic pressure, was slightly elevated at the highest concentration (Table V). The rate of ventricular contraction (+dP/dtinax) and relaxation (-dP/dtmax) was reduced dose-dependently and coronary flow decreased by 31%. Thus, endotoxin alone, at 72 high concentrations, produced some of the effects of PAF on rat heart, although not to the same degree. 3.9.2 In vivo effects of endotoxin In order to compare in vivo effects of endotoxin with those of PAF, and to determine whether endotoxin induces PAF release in vivo, we examined the actions of these two substances on blood pressure and ECG. As previously seen, PAF (0.5/zg/kg, i.v.) produced profound hypotension; blood pressure fell from 100±12 to 45±5 mmHg. Concomitant changes in the ECG included P-R interval prolongation and elevation of the S-T segment. On the other hand, endotoxin (10 mg/kg, i.v.) produced a brief hypotensive episode approximately 20 sec after administration followed by a delayed hypotensive effect (Figure 14 A) . Thereafter, the blood pressure returned to normal before falling inexorably with time. Neither heart rate nor the QRS interval of the ECG were changed relative to control. However, the P-R interval was prolonged and this effect was maximal 15 min after administration of endotoxin (Figure 14 B). Ibuprofen (5mg/kg, i.v.), which alone had no cardiovascular actions, did not prevent the action of PAF with respect to blood pressure and ECG changes. However, both the initial and delayed hypotensive actions of endotoxin were attenuated by treatment with ibuprofen (Figure 14A). The accompanying ECG changes, most notably 73 the prolongation of the P-R interval, were also attenuated by ibuprofen treatment (Figure 14B). 74 Table II: For each concentration of inhibitor the ability of the drug to inhibit responses to 1/zM PAF was expressed as percent protection of PAF effects in control hearts (line 3). The reductions in PAF response due to the presence of various antagonists and inhibitors were all statistically significant at p<0.05, using t-tests corrected for multiple sampling. The first three rows give absolute values for the different variables. Control 1 and control 2 are for the same group of hearts. 1 refers to the time before the experimental period and 2 to the time at end of the experimental time period and is a measure of the stability of hearts within the experimental time frame (i.e. a time control). All PAF responses were statistically different (p<0.05) from controls 1 and 2. None of the blockers alone reduced any of the variables. Results are the mean ± s.e. (n=5 for each of the fourteen groups). 75 Table II; Percent protection of responses to 1/xM PAF by various blockers (PAF antagonists and eicosanoid inhibitors). Drug Heart PR Systolic +dP/dtmax Coronary Rate interval pressure flow rate (beat/min) (msec) (mmHg) (mmHg/sec) (ml/min) control 1 control 2 PAF ( 1/xM ) 291 ± 12 279 ± 10 214 ± 16 32 ± 2 34 ± 3 48 ± 3 140 ± 6 132 ± 5 105 ± 5 3650±160 3425±150 2550±145 12 ± 2 11 ± 2 6.5 ± 2 Heart PR Systolic +dP/dtmax Coronary Rate interval pressure flow rate % % % % % CV 3988 (1/XM) CV 3988 (.01/iM) WEB 2086 (1/XM) WEB 2086 (.OljLlM) Ibuprofen (1/xM) Ibuprof en (. OljiiM) ONO 3708 (lMM) * ONO 3708 (.01/iM) FPL 55712 (1/iM) FPL 55712 (.OljuM) L-655,240 (1/XM) L-655,240 (.OljUM) 74 ± 26 ± 81 ± 58 ± 96 ± 67 ± 11 6 13 18 2 9 N/A 55 ± 91 ± 59 ± 84 ± 63 ± 20 3 16 13 11 42 ± 25 ± 55 ± 50 ± 89 ± 58 ± 11 10 6 9 4 8 N/A 68 ± 81 ± 62 ± 85 ± 60 ± i . - . i 14 6 10 6 8 25 + 26 ± 82 ± 54 ± 75 ± 77 ± 1 17 4 16 7 8 N/A 67 ± 58 ± 25 ± 85 ± 18 ± 11 10 7 9 6 i . * 61 ± 40 ± 74 ± 77 ± 82 ± 58 ± 11 9 11 7 5 9 N/A 60 ± 81 ± 57 ± 77 ± 47 ± 5 7 10 6 8 44 ± 9 37 ± 9 86 ± 10 80 ± 9 91 ± 3 85 ± 8 N/A 72 ± 7 87 ± 6 75 ± 8 92 ± 3 63 ± 9 76 Table III: Haemodynamic and ECG responses of RP 59227 as compared to vehicle control in pentobarbital-anaesthetised rats. Group BP HR PR QRS QT Vehicle 101±5 331±10 61±2 31±1 29±1 RP 59227 95+5 326±13 65±2 31±1 30±1 (10 mg/kg) RP 59227 121±6* 289±10* 66±2* 32±1 31±2 (25 mg/kg) The effects of RP 59227 on blood pressure (BP), heart rate (HR) and ECG measures in pentobarbital-anaesthetised rats. Values were recorded 5 min after drug administration. Doses were administered in a random and blind manner according to a randomized block design. All values are meants.e., n=6. * indicates p<0.05 from vehicle values. 77 Table IV; The antiarrhythmic properties of RP 59227 in pentobarbital-anaesthetised rats. GROUP INCIDENCE OF VT & VF (%) A.S. VT VF VT and/or VF Vehicle 100 100 100 6.3±0.5 RP 59227 100 50 100 5.1±0.2 (10 mg/kg) RP 59227 90 30* 90 4.2±0.6 (25 mg/kg) The antiarrhythmic properties of RP 59227 compared to vehicle on arrhythmia incidence during coronary artery occlusion in pentobarbitone-anaesthetised rats. Values are expressed as the percent of animals experiencing particular arrhythmias or mean±s.e.mean for arrhythmia score (AS). Doses were administered in a random and blind manner according to a randomized block design. All values are meants.e., n=6. * indicates p<0.05 from vehicle values. 78 TABLE V; Heart rate, ventricular pressure and ECG actions of E. coli lipopolysaccharide endotoxin GROUP WASH o.i 0.5 1.0 5.0 10.0 50.0 H.R. P-R QRS SP DP +dP/dt -dP/dt FLOW 266±7 257±11 246±17 254±9 241±6* 239±6* 236±6* 45±1 36±1 111±5 10±2 42±2 35±2 108±3 9±1 50±1 35±1 98±2 10±2 50±1 36±1 92±3 8±3 51±2 37±3 90±7 10±1 50±2 38±1 89±6* 12±1 52±3 39±1* 86±8* 14±2 2600±130 2450±120 2150±100 2100±150 2080±130 2050±120 1950±130 2250±120 2300±130 1950±100 1890±170 1800±100 1810±180 1780±140 11±0.6 9.5±0.6 8.9±0.1 8.2±0.2 8.0±0.1 7.5±0.5* 6.9±0.1* The dose-related effects of E. coli lipopolysaccharide endotoxin (|ig/ml) in isolated perfused rat hearts. Values are meanis.e.mean, n=5. *P<0.05 for means after treatment compared to wash or pretreatment means. SP= left-ventricular systolic pressure; DP= left-ventricular end-diastolic pressure. 79 Figure 1 Effects of PAF on peak systolic and end-diastolic left ventricular pressure. Measurements were made before (pre-drug) and 2 min after exposure to each concentration of PAF. PAF effects are expressed as percentage changes from pre-drug values. Each point is a meanis.e. (n=5) for peak systolic pressure with PAF (A) and Lyso-PAF (Jk). The symbol (O) indicates end-diastolic pressure with PAF; (•) is with Lyso-PAF. Pre-drug (control) values were 145±6 (mean±s.e.mean) mmHg for systolic pressure and 10±3 mmHg for diastolic pressure. % of Pre-drug ai o o £> L D —s c (Q O O D O CD r o (Q •Nl G) ai 08 81 Figure 2 Effects of PAF and Lyso-PAF on the maximum rate of intraventricular pressure development (+dP/dtmax) (A) and coronary flow (B). Measurements were made before drug administration (pre-drug) and after 2 min at each concentration. PAF effects are expressed as percentage changes from pre-drug values. Each point is a meants.e. (n=5) for PAF (A) and Lyso-PAF (A) . The pre-drug (control) value for +dP/dtmax for both groups was 3,7501160 mmHg/sec. B shows actions on coronary flow where values were determined as above. Each point represents the mean±s.e. (n=5) for PAF (A) and Lyso-PAF (A). The pre-drug value was 10±2 ml/min. Figure 2A 82 125 §> 100 CD a S 75 V 50 10 9 8 7 Drug cone. (-Log M) Figure 2 B 125 r 9 8 7 Drug conc.(-Log M) 83 Figure 3 Actions of PAF and Lyso-PAF on P-R interval and heart rate. Measurements were made before drug administration (pre-drug) and after 2 min at each concentration. PAF effects are expressed as percentage change from the pre-drug value. Each point represents the meanis.e. (n=5) for heart rate with PAF (A) and with Lyso-PAF (A)• The symbol (O) is for P-R interval in the presence of PAF and (•) is for Lyso-PAF. Pre-drug (control) values were 248±16 beats/min for heart rate and 37±3 msec for P-R interval. % of Pre-drug Ol o o ID D c (Q O O 15 O 00 r o (Q -si 0> fr8 85 Figure 4 The dose-response data for PAF actions (hatched columns) on mean arterial blood pressure compared with vehicle in artificially-ventilated, pentobarbital-anaesthetised rats (filled columns). Measurements were made 0.5, 1.0, 2.0 and 5.0 mins after drug administration. Intravenous bolus doses were given at 5 min intervals, cumulatively. PAF vehicle did not alter blood pressure, indicates P<0.05 vs. control (con). 86 E a In vivo actions of PAF Blood Pressure 125 r 100 -75 ~ 50 -25 -0 M. con 0.5 1 2 4 8 16 PAF dose (pg/kg) 87 Figure 5 The in vivo dose-response data for PAF actions on heart rate (hatched columns) of artificially-ventilated, pentobarbital-anaesthetised rats. Measurements were made 0.5, 1.0, 2.0 and 5.0 mins after drug administration. Intravenous bolus doses were given at 5 min intervals, cumulatively. PAF vehicle (filled columns) did not alter heart rate. 88 CD * - » CC CO CD X In vivo actions of PAF Heart rate 500 r 400 -E CO £ 300 -CO CD .O 200 -100 -con 0.5 1 2 4 8 PAF dose (Mg/kg) 16 89 Figure 6 The in vivo dose-response data for PAF actions (hatched columns) on the P-R (A) and the QRS interval (B) of the ECG in artificially-ventilated, pentobarbital-anaesthetised rats. Measurements were made 0.5, 1.0, 2.0 and 5.0 mins after drug administration. Intravenous bolus doses of either PAF or vehicle control (filled columns) were given at 5 min intervals, cumulatively. P<0.05 vs. control (con). In vivo actions of PAF P-R interval o £ > © c DC 0_ 9 0 con 0.5 1 2 4 8 PAF dose (pg/kg) 16 B QRS interval con 0.5 1 2 4 8 PAF dose (pg/kg) 16 91 Figure 7 The actions of PAF on blood pressure (A) and the P-R interval (B) of the ECG in the absence (filled columns) and presence (hatched columns) of ibuprofen (5mg/kg, i.v.). Intravenous bolus doses of PAF were given 5 min after ibuprofen and monitored at 0.5, 1.0, 2.0 and 5.0 min after administration. indicates P<0.05 vs. control (con). Blood pressure 92 150 r X E E a.' CO 100 r con 0.5 1 2 4 8 16 PAF dose (pg/kg) B P-R intrval 150 o W E > • Q. 100 con 0.5 1 2 4 8 16 PAF dose (pg/kg) 93 Figure 8 Action of PAF on coronary flow in contracting and non-contracting hearts. PAF (1/xM) was perfused into control hearts in Krebs-Henseleit solution (I) or in hearts perfused with Kreb's-Henseleit solutions containing either of the potassium/tetrodotoxin solutions (II). PAF effects were determined after 2 min of PAF perfusion following an initial 10 min perfusion with the appropriate Krebs-Henseleit solution. Each point is the mean±s.e. (n=5). Changes in coronary flow are expressed as percentage change from pre-drug values which were 12±2 ml/min for (I) and 10±3 mL/min for (II). P<0.05 for difference from I and + for difference from 100 percent. % of Pre-drug o ro o 4^ o ® o 00 o o o * + fr6 95 Figure 9 Comparison of reductions in coronary flow induced by PAF in non-contracting hearts with those induced by vasopressin and ergonovine. Hearts were perfused with Krebs-Henseleit containing potassium/tetrodotoxin plus vehicle (I); 1/xM Lyso-PAF (II); 1/xM PAF (III); 6.3/xM vasopressin (IV); or 3.0/zM ergonovine (V). Values are those obtained 2 min after beginning perfusion with drug and are expressed as a percentage of the appropriate pre-drug values. Coronary flow did not change with time as is shown by the values for I and II. Each point is the meanls.e. (n=5). *P<0.05 vs. I. % of Pre-drug o en en O en o o < < 96 97 Figure 10 Shows the dose-response data for isolated rat hearts exposed to nifedipine over the concentration range of 0.001-0.1/iM. (A) shows increases in coronary flow while (B) shows the positive and negative inotropic effects. These data allowed for the determination of concentrations of nifedipine which would provide vasodilation without negative inotropism (0.03/xM). indicates P<0.05 vs. control (con). Nifedipine D-R curve coronary flow 98 200 150 E 3 E x Ctf E 100 con .001 .01 .1 [Nifedipine] (JJM) B Systolic pressure 200 r 150 100 con .001 .01 .1 [Nifedipine] (yM) 99 Figure 11 Effects of 1.0/xM PAF on systolic pressure, +dP/dtmax/ heart rate and coronary flow in the absence (filled columns) and presence (hatched columns) of nifedipine (0.03JJM) . Results are the meanis.e. (n=5) of percent changes from pre-drug. P<0.05 for comparison between response in the absence of nifedipine. 2 0 0 ® 100 a Peak Sys. Pres. +dP/dt Heart Rate Coronary Flow o o 101 Figure 12 Dose-response data comparing methylcarbamyl-PAF (hatched columns) to PAF (filled columns) on mean arterial blood pressure (A) and heart rate (B) in pentobarbital-anaesthetised rats. indicates P<0.05 vs. control. Heart Rate (beats per min) o r o o o o CO o o o o en o o 00 Blood pressure (mmHg) en o o o en o o to 103 Figure 13 A Probit analysis of the lethal-dose curve for PAF in Swiss CD-I mice. Animals were given a single i.v. injection of PAF (n=10 mice per dose tested) and monitored for 60 min post-injection. Data are expressed as a probit percentage of animals which died after 60 min. The plot produces an LD50 value of 400/xg/kg for PAF with a correlation coefficient of 0.98. 104 PAF Probit Analysis r=0.98 0 Log of PAF dose 105 Figure 14 Shows the time-response actions of PAF and E. coli lipopolysaccharide endotoxin in pentobarbital-anaesthetised rats (n=6 per group). 0.5/ig/kg PAF (solid columns) and lOmg/kg endotoxin (narrow hatched columns) were administered intravenously and measurements of the mean arterial blood pressure (A) and the P-R interval (B) of the ECG recorded for 30 min. Time-response data for PAF (wide hatched columns) and endotoxin (spotted columns) were also observed after the administration of 5mg/kg ibuprofen. indicates P<0.05 vs. PAF and + vs. endotoxin at time 0. VO o *t—G O > .£ 5 .£ « X © 2a HI 5 < a. t u m m m m m m m m m s M w ' • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • i w ~Z~D + iHM^mssmsssmssmsmmssmssssm'sa CM l U W M M J W W M M M ^ f f l j M m d J Iml i l l b l I I I ! ' ' • ^\^^\x\\\v\\\vv\\\^^wxvv\^^^\\»^^ in • • i ' 11 • 11 i • 1 1 1 1 M ' i ^ i l ^ W ' i X ^ I o ® to >»w«ss;«ss^«w^»wciccciisjjx^^ o to o o o to I o (BHIUUJ) ejnssejd pooie > w <D •** C tc Q. " • • • • • • • • • • • • • • • • • • • • • • • • • • • • ' • • I o * • • • • • • • = » i o i f t m m m » m m m » » « w w ^ " -- --- ---•- -•- -•-•-•-•-•--.•-•-•.•i Q „, . , , , . , , , , . . , , , , , . ., , .£ T- ,»J, z $ss>x>iiiiiiS8idiii8$^^ 'v\'vVv?v.yv'v'vt^V^^V^\'^vh^'>.\'vWl'J :<b»»Kcaiicsaia»^aKa^^x^^ ; ^ ^ ^ » i » K ^ t S i ^ ^ ^ ^ i « ^ ^ ^ ^ ! ^ ^ ^ ^ » CO to CM O O O IO U) CM (oesai) |BAJ6)U| y-d 107 4 DISCUSSION 4.1 PAF actions in the heart This study showed that PAF had two distinct actions when it was perfused into isolated rat hearts. It acted on cardiac cells to reduce contractility and on coronary vasculature to reduce coronary flow. These two actions interact in a complex manner, since a reduction in contractility will reduce flow because of reduced oxygen consumption, while direct coronary vasoconstriction will reduce oxygen availablity and thereby reduce cardiac contractility. Our findings are in accord with those reported for cardiac tissue from various species under different conditions. Thus, a direct effect to reduce cardiac contractility has been reported by a number of workers (Camussi et al., 1984; Robertson et al., 1988) and direct coronary vasoconstriction by others (Benveniste et al., 1983; Feuerstein et al., 1984; Sybertz et al., 1985). This study is the first in which both actions have been unequivocally produced in the heart. The use of tetrodotoxin combined with elevated potassium concentrations to prevent contractility allowed the direct effects of PAF on coronary vessels to be demonstrated. PAF had the same vasoconstrictor action in the presence of either tetrodotoxin/K+ regimens. It has been shown that elevated 108 potassium concentrations can potentiate blocking actions of toxins selective for the sodium channel, e.g., tetrodotoxin (Baer et al., 1976; Lazdunski, 1982). On the other hand, in isolated hearts perfused with an appropriate concentration of nifedipine alone, contractility was not impaired whereas the direct effects of PAF on coronary flow were abolished. Under these conditions, the fact that cardiac contractility was still reduced by PAF indicated that negative inotropism was independent of vasoconstriction. It is well recognized that nifedipine is vasoselective and that concentrations exist which maximally dilate coronary arteries without producing negative inotropism. Negative inotropism normally occurs at nifedipine concentrations above 0.1 /iMolar (Curtis and Walker, 1988 and coronary flow studies here). The present studies showed that when PAF was continuously perfused for 2 min a steady state effect could be achieved and that this did not reach a maximum even at 1/xM. Comparisons of our data with those obtained when PAF was given as a single bolus (Piper and Stewart, 1988), show that the bolus dose was associated with greater decreases in cardiac function and severe tachyphylaxis with repeated bolus doses. Observations made with perfusions may better reflect the pathological release of PAF. The ECG was also affected by PAF. The decrease in heart rate may have been due to vasoconstrictor actions of PAF or due to direct release of eicosanoids induced by PAF. The P-R prolongation can be explained by vasoconstriction 109 resulting in partial ischemia in the atrio-ventricular node. The A-V node is particularly sensitive to such ischemia and responds with slowed conduction and P-R prolongation. On the other hand, if the direct negative inotropism of PAF were due to inhibition of calcium currents (Scherf and Cohen, 1964; Diez et al., 1990; Valone, 1991), the inhibition of such currents in the atrio-ventricular node would also slow conduction. Evidence has accumulated which suggests that the vasoconstriction mediated by PAF results from leukotriene release (Feuerstein et al., 1984; Sybertz et al., 1985; Stewart and Piper, 1986), or thromboxane production (Ezra et al., 1987). Recently described agents with purported PAF receptor antagonist properties (Terashita et al., 1983; Braquet and Godfroid, 1986; Braquet et al., 1987; Casals-Stenzel, 1987) were tested against PAF in this study. Studies with PAF antagonists showed only partial protection over a 100-fold concentration range, a pattern not in agreement with classical pharmacological competitive antagonism. The degree to which these antagonists are specific for PAF receptors in the heart is not known. To determine whether the production of eicosanoids by PAF contributed to cardiac effects several agents were examined. Ibuprofen, the specific cyclo-oxygenase inhibitor, attenuated, but did not abolish, the detrimental effects of PAF indicating that while cyclo-oxygenase products may contribute to the actions of PAF they are not 110 essential. The use of the thromboxane antagonist 0N0 3708 gave equivocal results in that high concentrations in the presence of IJUM PAF induced irreversible ventricular fibrillation, while a low concentration afforded only modest protection. The agent FPL 55712 has been shown to be quite effective in reducing leukotriene effects (Piper and Stewart, 1986) but it did not give complete protection in our studies. Levi et al. (1984) suggested that PAF acts, in the heart, independently of the release of arachidonic acid metabolites on the basis of observations of a lack of effect of either indomethacin or FPL 55712 on PAF responses. In conclusion, our study has shown that PAF has a number of actions in the isolated rat heart. It directly reduces contractility as well coronary flow and atrio-ventricular conduction. These actions can be attentuated by eicosanoid inhibitors but cannot be completely abolished suggesting that some of the actions of PAF may involve actions on a PAF receptor which still lacks definition. 4.2 PAF actions on the cardiovascular system All studies agree that PAF is a potent vasoactive compound in the cardiovascular system. The prominent hypotension is characterized by dose-dependency and rapid-onset while it mimics systemic anaphylaxis (Siren and Feuerstein, 1989). The infusion of PAF into rats produces widespread actions on the cardiovascular system which are Ill essentially platelet-independent (Caillard et al., 1982; Sanchez-Crespo et al., 1982). These actions were closely mimicked by the methyl-carbamyl PAF analog. The mechanism by which PAF produces these actions probably reflects a variety of complex mechanisms. The most prominent in vitro action of PAF is coronary vasoconstriction; however, PAF effects on the coronary circulation in vivo are more complex. In the heart, PAF and its methyl-carbamyl analog produce a biphasic coronary vasoconstrictor response (Feuerstein et al., 1984). An initial, brief increase in coronary flow is followed by a sustained decrease. In addition to these vascular changes, deterioration of cardiac output, stroke volume, and blood distribution in both the ventricles and periphery occur, enhancing the risk for coronary underperfusion and heart failure (Bessin et al., 1983). The mechanism of action of PAF on coronary blood flow is difficult to interpret and it is speculated that PAF may produce vasodilation via an endothelially-derived substance. The observed release of prostacyclin by PAF in canine experiments in vivo (Bessin et al., 1983) and from injured cardiac myocytes in vitro (Janero and Burghardt, 1990) may represent an attempt by the coronary vasculature to defend itself against underperfusion. The fact that the PAF effect is immediate, and that indomethacin does not block the early vasodilator effect of PAF, indicates that cyclooxygenase products do not participate in vasodilation. Several 112 studies show that the lipoxygenase pathway may be involved (Piper and Stewart, 1987), while others suggest a role for relaxation factors, such as EDRF (Chiba et al., 1990). However, most recent studies suggest a direct interaction of PAF with ion channels (Wahler et al., 1990). The sustained vasoconstriction which follows the initial vasodilation may be a direct effect of PAF itself (Feuerstein et al., 1984; Ezra et al., 1987; Pugsley et al., 1991) or a superimposed action of released vasoconstrictor elements such as TXA2 (Bessin et al., 1983) or peptido-leukotrienes (Piper and Stewart, 1987). During the decrease in coronary blood flow there is a concommitant change in the ECG, indicative of ischaemia. In vivo these ischaemic changes in the ECG may be enhanced by the peripheral haemodynamic actions of PAF. PAF produces profound increases in blood vessel permeability which result in a significant depletion of blood volume (Plante et al., 1985). In addition, PAF has a potent direct vasodilatory action on peripheral blood vessels which may, as in the heart, be due to the release of a relaxant factor. These systemic changes in blood flow may lead to decreased coronary perfusion pressure and hence ischaemia. The significant dose-dependent prolongation of the P-R interval and elevation of the S-T segment of the ECG may be consistent with ischaemia (Tamargo et al., 1985; Nakaya and Tohse, 1986). In this study, ibuprofen did not reduce any of the effects of PAF at the dose used. However, it has 113 been shown that the early phase of PAF-induced vasodilation in the heart is not blocked by the leukotriene antagonist FPL 55712 (Stahl et al., 1987) but is blocked by the PAF antagonists CV-6209 and BN 52021 (Stahl et al., 1987; Piper and Stewart, 1986). FPL 55712 and eicosanoid inhibitors do attenuate the sustained vasoconstriction (Piper and Stewart, 1987). This does not suggest an association between eicosanoids and either phase of coronary vasoaction of PAF. It has also been suggested that PAF-mediated cardiac dysfunction in intact rats is caused by asphyxia or a decrease in arterial pC>2 (Felix et al., 1990). Since all animals were ventilated, the authors differentiate between hypoxic and cardiac damage, since despite normoxia, cardiac dysfunction develops (Felix et al., 1990). Goldstein et al. (1991) observed that the cardiovascular dysfunction produced by PAF in pigs was due to an increase in pulmonary artery resistance which consequently produced right ventricular failure. A sudden release of PAF (as associated with systemic anaphylaxis or bacterial infection) may lead to a marked coronary vasoconstriction, ischaemia and low cardiac output, all of which are the hallmarks of cardiac anaphylaxis. 114 4.3 Antiarrhythmic actions of RP 59227 against ischaemia-induced arrhythmias. Studies have shown that PAF may act as an endogenous arrhythmogen. PAF has been shown to be released during atrial pacing in patients with coronary artery disease (Montrucchio et al., 1986), while in isolated hearts PAF also produces arrhythmias (Goldstein et al., 1991). In our in vitro studies, there was no significant increase of arrhythmia incidence with PAF except when ONO-3708 was combined with PAF, both at 1/iM concentrations, and ventricular fibrillation was induced. ONO-3708 is a PAF-related antagonist which possesses a non-constrained glyceryl backbone (Toyofuka et al., 1986). This is the first report of a potentially lethal synergism between PAF and a structurally-related antagonist (Pugsley et al., 1991). PAF may contribute to myocardial ischaemia and arrhythmogenesis in a number of ways. Firstly, as a potent platelet-aggregator it may promote the platelet-aggregation and trapping which occurs in ischaemic tissue of dogs and baboons (Annable et al., 1985; Wainwright et al., 1989). This may be true for mammals in which PAF affects platelets, such as the pig and dog. Hence, PAF antagonists may show promise in platelet-dependent disease states. However, since rat platelets are refractory to PAF, this postulated 115 mechanism of arrhythmogenesis seems not to play a significant role in this species. A second possible mechanism by which PAF may be involved in arrhythmogenesis is by a direct effect on myocardial muscle cells. PAF is a powerful negative inotropic agent (Kenzora et al., 1984). It is suggested that PAF interacts with a K+:K+ exchange pump within the lipid membrane of platelets and thereby produces changes in the platelet membrane potential. K+ imbalance increases Ca2+ conductance through Ca2+-mediated slow channels and elevates free cytosolic Ca2+ levels which is postulated to lead to a more permanent hyperpolarization through the opening of Ca2+-dependent K+ channels (Garay and Braquet, 1986). Although Ca2+ blockers such as verapamil inhibit PAF-induced changes in APD (Tamargo et al., 1985), other studies show that one effect of PAF, an increased sensitivity to ouabain-induced arrhythmias, is not blocked by diltiazem. Thus PAF interactions with Ca2+ channels may not be the major mechanism of PAF involvement in arrhythmias. Evidence is accumulating which implicates PAF as having direct actions on Na+ and K+ currents in cardiac tissue (Robertson et al., 1988; Vornovitskii et al., 1989). The precise mechanism and the site of this interaction are not known. However, PAF, as well as many other lipids, may indeed fulfil the requirement as endogenous arrhythmogens during ischaemic or reperfusion arrhythmias. 116 As previously described, PAF alters coronary flow in a biphasic manner. The initial increase in coronary blood flow may be due to PAF-mediated release of a relaxation factor such as EDRF and the coronary vasoconstriction which follows may be a direct effect of PAF or involve eicosanoids, probably TXA2 and peptido-leukotrienes. TXA2 plays a significant role in arrhythmogenesis (Mehta, 1990) and PAF may mediate its arrhythmogenic actions via TXA2, since studies with PAF and TXA2 antagonists reduce the incidence of PAF-induced ventricular arrhythmias (Riedel and Mest, 1987; Wainwright et al., 1989). The novel PAF antagonist RP 59227 dose-dependently decreases the severity and incidence of ventricular arrhythmias. Doses of RP 59227 which are antiarrhythmic in dogs (Auchampach et al., 1991) are not antiarrhythmic in rats subjected to occlusion of the main left-anterior descending (LAD) coronary artery. To be a moderately effective antiarrhythmic in rats, RP 59227 must be given at doses of 25 mg/kg. This may be due to the fact that cellular PAF levels contribute insignificantly as an endogenous arrhythmogen in ischaemic tissue of the rat. However, it might also indicate that platelet-activation is necessary for PAF-induced arrhythmogenesis. Evidence is accumulating which suggests that PAF can influence cardiac activity by stimulating the accumulation of PMN leukocytes in the myocardium. PMNL sequestration by PAF may exacerbate the ischaemic condition and mediate fatal 117 ventricular arrhythmias. The accumulation of PMNL's has been demonstrated to occur within 15 minutes of a decrease in coronary blood flow (Engler, 1987; Engler and Covell, 1987), which coincides with the incidence of life-threatening arrhythmias in the rat (Curtis et al., 1987). PAF dose-dependently killed mice when injected intravenously. PAF toxicity in mice is used for in vivo screening of PAF antagonists and our lethal-dose curve for PAF in Swiss CDj mice indicated an LD50 of 400 g/kg. Death was apparently cardiovascular in origin, involving circulatory shock and hemoconcentration (Myers and Ramwell, 1990). 4.4 Actions of E. coli endotoxin on the heart and cardiovascular system In order to examine whether the cardiac actions of PAF are the same as those of the material released by the interaction of endotoxin with macrophages or whether these are due to direct endotoxin effects alone, we performed studies with E. coli LPS given directly to isolated rat hearts. These studies were followed by studies in intact rats with or without pre-treatment with ibuprofen. Although PAF has been implicated as a major mediator in endotoxin shock, the mechanism by which endotoxin produces a state of shock is unknown. LPS produces a dose-dependent reduction in coronary flow and ventricular pressure in 118 vitro; however, the actions occur to a lesser degree than with PAF. As outlined above, the mechanisms involved are unknown but it may be a direct effect on membrane phospholipids in response to PLP-A2 activation (Terashita et al., 1985), or alterations in membrane enzymes in the heart or via receptors. Therefore, we conclude that the initial studies in which peritoneal macrophages were exposed to endotoxin (Salari and Walker, 1989) released substances such as PAF, LT's and TX's and that these produced a severe coronary vasoconstriction and general failure of cardiac contract i1ity. The endotoxic shock model in which the LPS endotoxin is given intrvenously as a bolus dose is the subject of some controversy. It has been suggested that studies would be more relevant to clinical endotoxic shock if shock were the result of a slow and sustained release of endotoxin (Ball et al., 1986). Despite such arguments, we have shown that a bolus injection of LPS produces a hypotensive response composed of at least 2 possible interactive components which are in agreement with studies by Doebber et al. (1985). The initial phase of endotoxic shock in animal models is a rapid and short-lived decrease in blood pressure which returns to baseline for a period of time before falling gradually with time. This first effect involves endotoxin-induced release of AA metabolites such as TXA2 and peptido-leukotrienes and in vivo this effect was attenuated by pre-treatment with ibuprofen. Many studies have shown that non-steroidal anti-119 inflammatory drugs (NSAIDs), especially ibuprofen, alter the pathophysiological sequelae of experimental endotoxic shock (Wise et al., 1980). Generally, ibuprofen reduces the production of TXA2, LTC4 and LTD4. We cannot fully explain the mechanism of action and degree of involvement of each autacoid until specific blocking drugs for each eicosanoid are made available. Recent studies have shown that endotoxin has a direct effect on blood vessels (Chiba et al., 1990; Vallance and Moncada, 1991). Endotoxin may stimulate nitric oxide production from vascular endothelium so as to produce the short-lived hypotensive effect. On the other hand, endotoxin may directly activate leukocytes and macrophages (and other inflammatory cells) which in turn produce PAF and eicosanoids (Innarea et al., 1985; Williams et al., 1990) . This is believed to be the mechanism by which the delayed hypotensive actions occur (Doebber et al., 1985), and ibuprofen only delays the onset of this second phase which may be due to PAF release. It is also conceivable that the time delay reflects the time required for PAF synthesis and secretion. Most studies have concentrated on the role of PAF in endotoxic shock and few have actually measured the levels of circulating PAF, although Levin et al. (1970) did develop a limulus test for endotoxin and bacteremia due to gram negative organisms. However, PAF antagonists such as ONO 6240, CV-3988 and WEB 2086 have been shown to reverse the hypotension and hemoconcentration 120 produced by endotoxin (Toyofuko et al., 1986; Terashita et al., 1983; Casals-Stenzel, 1987). Thus, much active research is now being conducted towards the development of more selective antagonists for PAF. In the above studies we were able to show that PAF alone produced cardiovascular dysfunction both in vivo and in vitro. However, both the literature and previous studies in the laboratory with activated macrophages showed the unequivocal release of substances which induce a direct and profound cardiovascular dysfunction and that one of these substances may be PAF. 121 5 Summary Our studies have shown that PAF has a number of actions in the isolated rat heart. It has a direct vasoconstrictor action on coronary vessels and also a direct negative inotropic action on myocardial cells, both independent of each another. The cardiac actions of PAF may also, in part, involve secondary release of mediators. The apparent non-specific nature of the reported eicosanoid antagonists made identification of the nature of the arachidonic acid metabolites difficult. However, the eicosanoid antagonists, as well as the PAF antagonists, did attenuate PAF actions on isolated rat hearts, suggesting two components: an indirect one and a direct one involving a PAF receptor which as yet lacks definition in molecular and pharmacological terms. Studies with endotoxin in vivo confirmed that it has profound actions on blood pressure. In vivo, its actions are attenuated by ibuprofen which implicates arachidonic acid metabolites as being involved in the pathophysiological process. Endotoxin may produce its actions by affecting endothelial cells directly (through nitric oxide production), by priming macrophages for the enhanced release of the arachidonic acid metabolites or by interacting with a putative endotoxin-binding site on neutrophils and monocytes which then produce PAF. 122 The ability of PAF to produce cardiac dysfunction in both isolated rat and rabbit hearts suggests that there are only slight differences in the degree of response of cardiac vasculature to the coronary vasoconstricting properties of PAF in these two species. Finally, the in vivo actions of PAF and its analogue, methylcarbamyl-PAF, provide important information regarding activity on both the cardiac and cardiovascular systems. PAF and methylcarbamyl-PAF are very potent hypotensive agents. A dose of 1/zg/kg of either compound is sufficient to reduce the blood pressure of an animal by at least 50%. Interestingly, both compounds decrease the height of the R-wave but whether this is a direct action on the ventricles or the indirect result of systemic hypotension is not known. PAF and methylcarbamyl-PAF also significantly elevated the S-T segment and prolonged the P-R interval which may indicate ischaemia of the A-V node due to the severely decreased systemic blood pressure. Sodium channel blockade in the rat is not responsible for changes in the ECG, since the RSh measure did not show changes characteristic of such channel blockade. However, PAF could have been causing calcium channel blockade. PAF has a number of actions on the heart and cardiovascular systems and further pharmacological investigation is necessary to fully characterize all of its actions on these systems. Current research involves using 123 isolated tissues in an attempt to analyze how PAF influences electrophysiological currents in tissues and cells. 124 References Aderem, A.A., Cohen, D.S., Wright, S.D., and Cohn, Z.A. Bacterial lipopolysaccharides prime macrophages for enhanced release of arachidonic acid metabolites. J. Exp. Med. 164; 165-179, 1986. Alloatti, G., Montrucchio, G., Marrano, F., Tetta, C , DePaulis, R., Morea, M., Emmanuelli, G., and Camussi, G. Effect of platelet-activating factor (PAF) on human cardiac muscle. Int. Archs. Allergy Appl Immunol. 79: 108-112, 1986. Alloatti, G., Montrucchio, G., and Camussi, G. Prostacyclin inhibits the platelet-dependent effects of platelet-activating factor in the rabbit isolated heart. J. Cardiovascular Pharmacol. 15: 745-751, 1990. Annable, C.R., McManus, L.M., Carey, K.D., and Pinckard, R.N. Isolation of platelet-activating factor (PAF) from ischemic baboon myocardium. Fed. Proc. 44: 1271, 1985. Arnoux, B., Duval, D, and Benveniste, J. Release of platelet-activating factor (PAF-acether) from alveolar macrophages by the Ca2+ ionophore A23187 and phagocytosis. Eur. J. Clin. Invest. 10: 437-441, 1980. Au, T.L.S., Collins, G.A., Harvie, C.J., and Walker, M.J.A. The actions of prostaglandins 12 and E2 on arrhythmias produced by occlusion in rat and dog. Prostaglandins 18: 207-220, 1979. Auchampach, J.A., Maruyama, M., Cavero, I., and Gross, G.J. Blocking platelet-activating factor receptors with RP 59227 decreases myocardial infarct size and the incidence of ventrucular fibrillation. J. Molec. Cell. Cardiol. 23(suppl III): S-52, 1991. Ball, H.A., Cook, J.A., Wise, W.C., and Halushka, P.V. Role of thromboxanes, prostaglandins, and leukotrienes in endotoxic and septic shock. Int. Care Med. 12: 116-126, 1986. Barbaro, J.F. and Zvaifler, N.J. Antigen-induced histamine release from platelets and rabbits producing PCA antibody. Proc. Soc. exp. Biol. Med. 122: 1245-1247, 1966. Barthelson, R., Valone, F., Debs, R., and Philips, R. Synergy in interleukin 1 (IL-1) release by human monocytes stimulated with platelet-activating factor (PAF) plus gamma interferon (INFgamma) or tumor necrosis factor (TNF). FASEB J. 2: 1228, 1988. 125 Batt, J.-P., Redevich, C., Heymans, F., Dive, G., Lamouri, A., Hosford, D., Braquet, P., and Godfroid, J.J. PAF receptor and "Cache-oreilles" effect. Simple PAF antagonists. Lipids. 26: 1167-1171, 1991. Baumgartner, J.-D., Glauser, M.P., McCutchan, J.A., Ziegler, E.J., vanMelle, G., Klauber, M.R., Vogt, M., Muehlin, E., Luethy, R., Chiolero, R., and Geroulanos, S. Prevention of gram-negative shock and death in surgical patients by antibody to endotoxin core glycolipid. Lancet. 2: 59-63, 1985. Benveniste, J., Henson, P.M., and Cochrane, C.G. Leukocyte-dependent histamine release from rabbit platelets: the role of IgE basophils, and platelet activating factor. J. Exp. Med. 13_6: 1356-1377, 1972. Benveniste, J. Platelet activating factor, a new mediator of anaphylaxis and immune complex deposition from rabbit and human basophils. Nature (Lond.) 249: 581-582, 1974. Benveniste, J., LeCouedic, J.P., Kamoun, P. Aggregation of human platelets by platelet-activating factor. Lancet 344: 211-212, 1975. Benveniste, J., LeCouedic, J.P., Polonsky, J., and Tence, M. Structural analysis of purified platelet-activating factor by lipases. Nature (Lond.) 269: 1170-171, 1977. Benveniste, J., Tence, M., Varenne, P., Bidault, J., Boullet, C , and Polonsky, C. Proc. Acad. Paris 289: 1037-1040, 1979. Benveniste, J. Platelet activating factor (PAF-acether): Hitorical background and definitions. Agents and Actions. 12: 700-701, 1982. Benveniste, J., Chignard, M., LeCouedic, J.P., and Vargaftig, B.B. Biosynthesis of platelet-activating factor (PAF-acether). II Involvement of phospholipase A2 in the formation of PAF-acether and Lyso-PAF-acether from rabbit platelets. Thromb Res. 25: 375-385, 1982. Benveniste, J., Boullet, C , Brink, C., and Labat, C. The actions of PAF-acether (platelet-activating factor) on guinea-pig isolated heart preparations. Brit. J. Pharmacol. 80: 81-83, 1983. Bessin, P., Bonnet, J., Apffel, P., Soulard, C., Desgrou, L., Pelassi, I., and Benveniste, J. Acute circulatory collapse caused by Platelet Activating Factor (PAF) in dogs. Eur. J. Pharmacol. 86: 403-413, 1983. 126 Blank, M.L., Snyder, F., Byers, L.W., Brooks, B., and Muirhead, E.E. Antihypertensive activity of an alkyl ether analog of phosphatidylcholine. Biochem. Biophys. Res. Commun. 90: 1194-1200, 1979. Blank, M.L., Lee, T.C., Fitzgerald, V., and Snyder, F. A specific acetylhydrolase for ll-alkyl-2-acetyl-sn-glycero-3-phosphocholine (a hypotensive and platelet-activating lipid). J. Biol. Chem. 256: 175-178, 1981. Bliss, C.I. The calculaton of the dosage-mortality curve. Ann. Appl. Biol. 22,: 134-167, 1935. Bone, R.C., Fisher, C.J., Clemmer, T.P., Slotman, G. J., Metz, C.A., and Balk, R.A. A controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. New Eng. J. Med. 317: 653-658, 1987. Braquet, P. and Godfroid, J.J. PAF-acether specific binding sites: 2. Design of specific antagonists. Trends Pharmacol Sci. 7: 397-403, 1986. Braquet, P., Touqui, L., Shen, T.Y., and Vargaftig, B.B. Perspectives in Platelet-activating Factor Research. Pharmacol. Rev. 3_9: 97-145, 1987. Broquet, C., Teulade, M.P., Borghero, C , Heymans, F., Godfroid, J.J., and Lefort, J. Structural analogs of PAF-acether. Eur. J. Med. Chem. 19: 229-233, 1984. Broadly, K.J. The Langendorff heart preparation Reappraisal of it's role as a research and teaching model for coronary vasoactive drugs. J. Pharmacol. Methods 2: 143-156, 1979. Budden, R., Buschmann, G., and Kuhl, U.G. The rat ECG in acute Pharmacology and Toxicology. In: Budden, R., Detweiler, D.K., Zbinden, G. (Eds.) The Rat Electrocardiogram in Pharmacology and Toxicology. Pergamon Press Ltd., U.K. 1981. Bult, H., and Herman, A.G. Prostaglandins and circulatory shock. In: Herman, A.G., Vanhoutte, P.M., Denolin, H., and Goossens, A. (Eds.) Cardiovascular Pharmacology of Prostaglandins. Raven Press, New York, 1982. Caillard, C.G., Mondot, S., Zundel, J.L., and Julou, L. Hypotensive activity of PAF-acether in rats. Agents and Actions 12: 725-730, 1982. Camussi, G., Aloatti, G., Montrucchio, G., Meda, M., and Emanuelli, G. Effect of platelet-activating factor on guinea-pig papillary muscle. Experentia 4_0: 697-699, 1984. 127 Camussi, G., Bussolino, F., Tetta, C , Piacibello, W., and Aglietta, M. Biosynthesis and release of platelet-activating factor from human monocytes. Int. Arch. Allergy Appl. Immunol. 70: 245-251, 1983. Casals-Stenzel, G. Protective effect of WEB 2086, a novel antagonist of platelet activating facttor, in endotoxin shock. Eur. J. Pharmacol. 135: 117-122, 1987. Chakrabarty, S., Thomas, P., and Sheridan, D.J. Contribution of platelets and platelet activating factor (PAF) to the arrhythmogenic, haemodynamic and necrotic effects of acute myocardial ischaemia. Eur. Heart J. 12: 583-589, 1991. Chiba, Y., Mikoda, N., Kawasaki, H. and Ito, K. Endothelium dependent relaxant action of platelet-activating factor in the rat mesenteric artery. Naun. Schmied. Arch Pharmacol. 341: 68-73, 1990. Chignard, M., LeCouedic, J.P., Vargafttig, B.B., and Benveniste, J. Platelet-activating Factor (PAF-acether) secretion from platelets: effects of aggregating agents. Br. J. Haematol. 46: 455-464, 1980. Cohen, J., and Glauser, M.P. Septic shock: treatment. Lancet. 338: 736-739, 1991. Coiffier, E., Burrel, M.C., Lefort, J., Chignard, M., Braquet, C., Heymans, F., Godfroid, J.J., and Vargaftig, B.B. Effects of PAF-acether and structural analogs on platelet activation and bronchoconstriction in guinea-pigs. Eur. J. Pharmacol. 131: 179-188, 1986. Curtis, M.J., MacLeod, B.A., Tabrizchi, R., and Walker, M.J.A. An improved perfusion apparatus for small animal hearts. J. Pharmacol. Methods 15: 87-94, 1986a. Curtis, M.J. PhD Thesis. The Actions of Calcium Antagonists on Arrhythmias and Other Responses to Myocardial Ischaemia in the Rat. The University of British Columbia, July 1986. Curtis, M.J., MacLeod, B.A., and Walker, M.J.A. Models for the study of arrhythmias in myocardial ischaemia and infarction: the use and abuse of the rat. J. Mol. Cell. Cardiol. 19: 399-419, 1987. Curtis, M.J. and Walker, M.J.A. Quantification of arrhythmias using scoring systems: an examination of seven scores in an in vivo model of regional myocardial ischaemia. Cardiovasc. Res. .22: 656-665, 1988. 128 Demopolous, C.A., Pinckard, R.N., and Hanahan, D.J. Platelet-activating factor. Evidence for l-0-alkyl-2-acetyl-sn-glyceryl-3-phosphorylcholine as the active component. (A new class of lipid chemical mediators). J. Biol. Chem. 354! 9355-9358, 1979. Dent, F., Ukena, D., Sybrecht, G.W., and Barnes, P.J. [3H]WEB 2086 labels platelet activating factor receptors in guinea-pig and human lung. Eur. J. Pharmacol. 169: 313-316, 1989. Detweiler, D.K. The use of electrocardiography in Toxicological studies with rats. In: Budden, R., Detweiler, D.K., Zbinden, G. (Eds.) The Rat Electrocardiogram in Pharmacology and Toxicology. Pergamon Press Ltd., U.K. 1981. Diez, J., Delpon, E., and Tamargo, J. Effects of platelet-activating factor on contractile force and 45Ca fluxes in guinea-pig isolated atria. Br. J. Pharmacol. 100: 305-311, 1990. Doebber, T.W., Wu, M.S., Robbins, J.C., Choy, B.M., Chang, M.N., and Shen, T.T. Platelet Activating Factor (PAF) involved in endotoxin-induced hypotension in rats. Studies with PAF-receptor antagonists kadsurenone. Biochem. Biophys. Res. Commun. 2jJ: 799-808, 1985. Doring, H.J. and Dehnert, H. The Isolated Perfused Warm-Blooded Heart according to Langendorff. Methods in Experimental Physiology and Pharmacology. Biological Measurement Techniques V. (1st English ed.) Biomesstechnik-Verlag, West Germany 1988. Driscoll, P. The normal rat electrocardiogram (ECG). In: Budden, R., Detweiler, D.K., Zbinden, G. (Eds.) The Rat Electrocardiogram in Pharmacology and Toxicology. Pergamon Press Ltd., U.K. 1981. Duncan, D.B. Multiple Range and multiple F tests. Biometrics 11: 1-42, 1955. Duronio, V., Reany, A., Wong, S., Bigras, C., and Salari, H. Characterization of platelet-activating factor receptors in porcine platelets. Can. J. Physiol. Pharmacol. 68: 1514-1519, 1990. Engler, R. Consequences of activation and adenosine-mediated inhibition of granulocytes during myocardial ischaemia. Fed. Proc. 46: 2407-2412, 1987. Engler, R., and Covell, W. Granulocytes cause reperfusion ventricular dysfunction after 15-minutes ischemia in the dog. Circ. Res. 61: 20-28, 1987. 129 Etienne, A., Hecquet, F., Soulard, C., Touvay, C., Clostre, F., and Braquet, P. The relative role of PAF-acether and eicosanoids in septic shock. Pharmacol. Res. Commun. 18: 71-79, 1986. Ezra, D., Laurindo, F.R.M., Czaja, J.F., Snyder, F., Goldstein, R.E., and Feuerstein, G. Cardiovascular and coronary coonsequences of intracoronary platelet-activating factor infusion in the domestic pig. Prostaglandins. 34; 41-57, 1987. Farr, R.S., Wardlow, M.L., Cox, C.P., Meng, K.E., and Greene, D.E. Human serum acid-labile factor is an acyl-hydrolase that inactivates Platelet-Activating Factor. 42; 3120-3122, 1983. Felix, S.B., Steger, A., Baumann, G., Busch, R., Ochsenfeld, G., and Berdel, W.E. Platelet-activating factor-induced coronary constriction in the isolated perfused guinea-pig heart and antagonist effects of the PAF antagonist WEB 2086. J. Lipid Med. 2: 9-20, 1990. Felix, S.B., Baumann, G., Ahmed, Z., Hashemi, T., Niemczyk, M., and Berdel, W.E. Effects of platelet-activating factor on myocardial contraction and myocardial relaxation of isolated perfused guinea-pig hearts. J. Cardiovasc. Pharmacol. 16: 750-756, 1990. Felix, S.B., Baumann, G., Raschke, P., Maus, C , and Berdel, W.E. Cardiovascular reactions and respiratory events during platelet-activating factor-induced shock. Bas. Res. Cardiol. 85: 217-226, 1990. Feuerstein, G., Boyd, L.M., Ezra, D., and Goldstein, R.E. Effect of platelet-activating factor on coronary circulation of the domestic pig. Am. J. Physiol. 246: H466-H471, 1983. Feuerstein, G and Hallenbeck, J.M. Prostaglandins, Leukotrienes and PAF in shock. Ann. Rev. Pharmacol. Toxicol. 2_7: 301-313, 1987. Fiedler, V.B., Mardin, M. and Abram, T.S. Comparison of cardiac and hemodynamic effects of platelet-activating factor-acether and leukotriene D4 in anesthetised dogs. Bas. Res. Cardiol. 82: 197-208, 1987. Fletcher, J.R., and Ramwell, P.W. Indomethacin treatment following baboon endotoxic shock improves survival. Adv. Shock Res. 4: 103-111, 1980. Gad, S., Weil, C.S. Statistical and Experimental Design for Toxicologists. Telford Press, Caldwell, N.J., 70-93, 1988. 130 Garay, R., and Braquet, P. Involvement of K movements in the membrane signal induced by PAF-acether. Biochem. Pharmacol. 35: 2811-2815, 1986. Glauser, M.P., Zanetti, G.,, Baumgartner, J.-D., and Cohen, J. Septic shock: pathogenesis. Lancet 338: 732-736, 1991. Godfroid, J.J., Heymans, F., Michel, E., Redeuilh, H.C., Steiner, E., and Benveniste, J. Platelet Activating Factor (PAF-acether): total synthesis of l-0-octadecyl-2-0-acetyl-sn-glycero-3-phosphorylcholine. FEBS Letters. 116; 161-164, 1980. Godfroid, J.J., and Braquet, P. PAF-acether binding sites. 1. Quantitative SAR study of PAF-acether isosteres. Trends Pharmacol. Sci. 7: 368-373, 1986. Godfroid, J.J., Braquet, C , Jouquey, S., Lebbarn, M., Heymans, F., Steiner, E., Michel, E., Coeffier, E., Fichelle, J., and Worcel, M. Structure-Activity Relationship in PAF-acether 3. Hydrophobic contribution to the agonistic activity. J. Med. Chem. 3_0: 792-797, 1987. Goldstein, R.E., Feuerstein, G.Z., Bradley, L.M., Stambouly, J.J., Laurindo, F.R.M., and Davenport, N.J. Cardiovascular effects of Platelet-Activating Factor. Lipids 26: 1250-1256, 1991. Grey, G.A., Furman, B.L., and Parratt, J.R. Endotoxin-induced impairment of vascular reactivity in the pithed rat: Role of arachidonic acid metabolites. Circ. Shock. 31: 395-406, 1990. Grisham, M.B., Eversen, J., and Janssen, H.F. Endotoxemia and neutrophil activation in vivo. Am. J. Physiol. 254: H1017-H1022, 1988. Guenter, C.A., Fiorcica, v., and Hinshaw, L.B. Cardiorespiratory and metabolic responses to live E. coli and endotoxin in the monkey. J. Appl. Physiol. 2_6: 780-786, 1969. Hack, C.E., Nuijens, J.H, and Felt-Bergma, R.J.F., Schreuder, W.O., Eerenberg-Belmer, A.J.M., Paaredkooper, J., and Bronsveld, W. Elevated plasma levels of the anaphylatoxins C3a and C4a are asssociated with a fatal outcome in sepsis. Am. J. Med. 86: 20-26, 1989. Hamberg, M., Svensson, J., and Samuelsson, B. Thromboxanes: a new group of biologically active compounds derived from prostaglandin endoperoxides. Proc. Natl. Acad. Sci. USA. 72: 2994-2998, 1975. 131 Hanahan, D.J., Demopolous, C.A., Liehr, J., and Pinckard, R.N. Identification of platelet-activating factor isolated from rabbit neutrophils. J. Biol. Chem. 255: 5514-5516, 1980. Hanahan, D.J. Platelet-Activating Factor: A biologically active phosphoglyceride. Ann. Rev. Biochem. 55: 483-509, 1986. Handley, D.A. Preclinical and clinical pharmacology of platelet-activating factor receptor antagonists. Med. Res. Rev. 10: 351-370, 1990. Hintze, J.L. Number Cruncher Statistical System: Version 5.01. J.L. Hintze, Kaysville, Utah, 1987. Hoffman, S.L., Punjabi, N.H., Kumala, S., Moechtar, M.A., Pulungsih, S.P., Rivai, A.R., Rockhill, R.C., Woodward, T.E., and Loedin, A.A. Reduction of mortality in chloramphenicol-treated severe typhoid fever by high-dose dexamethasone. New Eng. J. Med. 310: 82-88, 1984. Honda, Z., Nakamura, M., Miki, I., Minami, M., Watanabe, T., Seyama, Y., Okado, H., Toh, H., Ito, K., Miyamoto, T., and Shimizu, T. Cloning by functional expression of platelet activating factor receptor from guinea-pig lung. Nature. 349: 342-346, 1991. Hosford, D., Braquet, M.-P., and Braquet, P. Platelet-activating factor and cytokine interaction in shock. Resuscitation. 18: 207-218, 1989. Hosford, D., and Braquet, P. Antagonists of Platelet-Activating Factor: Chemistry, Pharmacology and Clinical Applications. In Ellis, G.P., and West, G.B. (Eds.) Progress in Medicinal Chemistry. Elsevier Science Publishers, B.V. (Biomedical Division), Amsterdam, The Netherlands, 1990. Howard, P.G. MSc Thesis. Pharmacological and Antiarrhythmic Properties of Quinacainol - A new sodium channel blocker? The University of British Columbia. November, 1990. Hu, W., Kinnaird, A.A.A., and Man, R.Y.K. Mechanism of the coronary vascular effects of platelet-activating factor in the rat perfused heart. Br. J. Pharmacol. 103: 1097-1102, 1991. Hwang, S.B., Lee, C.S.C., Cheah, M.J., and Shen, T.Y. Specific receptor sites for l-0-alkyl-2-0-acetyl-sn-glycero-3-phosphocholine (PAF) on rabbit platelet and guinea-pig smooth muscle membranes. Biochemistry. 22.: 4756-4763, 1983. 132 Hwang, S.B., Cheah, M.J., Lee, C.S.C., and Shen, T.Y. Effects of non-steroidal anti-inflammatory drugs on the specific binding of platelet-activating factor to membrane preparations of rabbit platelets. Thrombosis Res. 34: 519-5311, 1984. Hwang, S.B., Lamb, M.H., and Shen, T.Y. Specific binding sites for platelet-activating factor in human lung tissues. Biochem. Biophys. Res. Commun. 128: 972-979, 1985. Innarea, P., Gomez-Cambronero, J., Nieto, M., and Sanchez-Crespo, M. Characteristics of the binding of platelet-activating factor to platelets of different animal species. Eur. J. Pharmacol. 105: 309-315, 1984. Innarea, P., Gomez-Cambronero, J., Pascual, J., delCarmen Ponte, M., Hernando, L., and Sanchez-Crespo, M. Synthesis of PAF-acether and blood volume changes in gram (-) sepsis. Immunopharmacology. _9: 45-52, 1985. Jackson, C.V., Schumacher, W.A., Kunkel, S.L., Driscoll, E.M., Pope, T.K., and Lucchesi, B.R. Platelet-activating factor release of a platelet derived coronary artery vasodilator substance in the canine. Circ. Res. 58: 218-229, 1986. Janero, D.R. and Burghardt, C. Production and release of platelet-activating factor by the injured heart-muscle cell (cardiomyocyte). Res. Commun. Chem. Path. Pharmacol. 67: 201-218, 1990. Johnston, K.M., Macleod, B.A., and Walker, M.J.A. ECG and other responses to ligation of a coronary artery in the conscious rat. In: Budden, R., Detweiler, D.K., Zbinden, G. (Eds.) The Rat Electrocardiogram in Pharmacology and Toxicology. Pergamon Press Ltd., U.K. 1981. Julou-Schaeffer, G., Gray, G.A., Fleming, I., Schott, C , Parratt, J.R., and Stochet, J.-C. Loss of vascular responsiveness induced by endotoxin involves L-arginine pathway. Am. J. Physiol. 259: H1038-H10„3, 1990. Kane, K.A., McDonald, F.M., and Parratt, J.R The anaesthetised rat as a model for investigating early post-ligation dysrhythmias. In: Budden, R., Detweiler, D.K., Zbinden, G. (Eds.) The Rat Electrocardiogram in Pharmacology and Toxicology. Pergamon Press Ltd., U.K. 1981. Kamitani, T., Katamoto, M., Tatsumi, M., Katsuta, K., Ono, T., Kikuchi, H and Kumpa, S. Mechanisms of the hypotensive effects of synthetic 1-0-octadecyl-2-0-acetyl-glycero-3-phosphorylcholine. Eur. J. Pharmacol. 98: 357-366, 1984. 133 Kass, E.H. High dose corticosteroids for septic shock. New Eng. J. Med. 311: 1178-1179, 1984. Kenzora, J.L., Perez, J.E., Bergmann, S.R., and Large, L.G. Effects of Acetyl Glycero Ether of Phosphorylcholine (PAF) on ventricular preload, afterload, and contractility in dogs. J. Clin. Invest. 74: 1193-1203, 1984. Klaassen, CD., and Doull, J. Evaluation of Safety: Toxicologic Evaluation. In: Doull, J., Klaassen, CD. and Amdur, M.O. (Eds.) Casarett and Doull's Toxicology: The Basic Science of Poisons. Macmillan Publishing Co., Inc. New York, N.Y. 1980. Koltai, M., Hosford, D., Guinot, P., Esanu, A. and Braquet, P. Platelet-activating factor (PAF). A review of its Effects, Antagonists and possible future Clinical Implications (Part I) Drugs 42(1): 9-29, 1991. Koltai, M., Hosford, D., Guinot, P., Esanu, A. and Braquet, P. Platelet-activating factor (PAF). A review of its Effects, Antagonists and possible future Clinical Implications (Part II) Drugs 42 (2): 174-204, 1991. Kurland, J.I., and Bockman, R. Prostaglandin E production by human blood monocytes and mouse peritoneal macrophages. J. Exp. Med. 147: 952-957, 1978. Lamotte-Brasseur, J., Dive, G., Lamouri, A., Heymans, F., and Godfroid, J.J. PAF-receptor. III. Conformational and electronic properties of PAF-like agonists and antagonists. Biochem. Biophys. Acta. 1085: 91-105, 1991. Langendorff, 0. Untersuchung am uberlebenden saugetierherzen. Pflug. Arch. Ges. Physiol. 61: 291-295, 1895. Lave, D., James, C , Fabre, J.L., Floch, A., and Robaut, C Synthesis of new pyrrolo [1,2-c] thiazole derivatives as potent and specific PAF-antagonists. Prostaglandins 35: 802, 1988. Leaver, H.A., Qu, J.M., Smith, G., Howie, A., Ross, W.B., and Yap, P.L. Endotoxin releases platelet-activating factor from human monocytes in vitro. Immunopharmacology. 20: 105-113, 1990. Lefer, A.M. Role of prostaglandins and thromboxanes in shock states. In: Altura, B.M., Lefer, A.M., and Schumer, W. (Eds.) Handbook of shock and trauma. Raven Press, New York, N.Y., 1986. Levi, R., Burke, J.A., Guo, Z.G., Hattori, Y., Hoppens, CM., McManus, L.M., Hanahan, D.J., and Pinckard, R.N. 134 Acetyl-glyceryl ether phosphocholine (AGEPC). A putative mediator of cardiac anaphylaxis in the guinea-pig . Circ. Res. 54: 117-124, 1984. Levi, R., Genovese, A., and Pinckard, R.N. Alkyl chain homologs of platelet-activating factor and their effects on the mammalian heart. Biochem. Biophys. Res. Commun. 161: 1341-1347, 1989. Levin, J., Poore, T.E., Zauber, N.P., and Osier, R.S. Detection of endotoxin in the blood of patients with sepsis due to gram-negative bacteria. New Eng. J. Med. 283; 1313-1316, 1970. Li, J.C.R. Statistical Inference I. Edwards Brothers, Inc., Ann Arbor, MI., 1964. Liu, M.S., and Takeda, H. Endotoxin-induced stimulation on phospholipase-A2 activity in dog hearts. Biochem. Med. 25; 62-69, 1982. Liu, M.S., and Xuan, Y.T. Mechanism of endotoxin-induced impairment in Na/Ca exchange in canine myocardium. Am. J. Physiol. 251: R1078-R1086, 1986. Ludwig, J.C., McManus, L.M., Clar, P.O., Hanahan, D.J. and Pinckard, R.N. Modification of platelet-activating factor (PAF) synthesis and release from human polymorphonuclear leukocytes: role of extracellular Ca2+. Arch. Biochem. Biophys. 23_2: 102-110, 1984. MacLean, M.R., and Hiley, C.R. Effects of artificial respiratory volume on the cardiovascular responses to an Al and A2 adrenoceptor agonist in the air-ventilated pithed rat. Br. J. Pharmacol. 78: 165-16 , 1988. Malech, H.L., and Gallin, J.I. Neutrophils in human diseases. New Eng. J. Med. 317; 687-694, 1987. Mehta, J.L. Platelet-Activation in Unstable Angina: Role of thromboxane-A2 and other mediators of vasoconstriction. J. Am. Coll. Cardiol. 15: 727-729, 1990. Meyers, A.K., and Ramwell, P.W. Modelling of PAF pathophysiology: Lethal shock induced by PAF in rodents. In: O'Flaherty, J.T., and Ramwell, P.W. (Eds.) PAF Antagonists: New developments for clinical application. Portfolio Publ. Co., Houston, TX, 1990. Mickelson, J.K., Simpson, P.J., and Lucchesi, B.R. Myocardial dysfunction and coronary vasoconstriction induced by platelet-activating factor in the post-infarcted rabbit isolated heart. J. Mol. Cell. Cardiol. 2J): 547-561, 1988. 135 Moncada, S., Gryglewski, R.J., and Bunting, S. A lipid peroxide inhibits the enzyme in blood vessel microsomes that generates from prostaglandin endoperoxides the substance (PG X) which prevents platelet aggregation. Prostaglandins. 12: 715-733, 1976. Montgomery, D.C. Design and Analysis of Experiments. John Wiley & Sons, New York, N.Y., 43-85, 1984. Montruccio, G., Camussi, G., Tetta, G., Emmanuelli, G., Orzan, F., Libero, L., and Brusca, A. Intravascular release of platelet-activating factor during atrial pacing. Lancet. 2: 293, 1986. Montruccio, G., Aloatti, G., and Tetta, G. Release of platelet-activating factor from ischemia-reperfused rabbit heart. Am. J. Physiol. 256; H1236-H1246, 1989. Morris, P.D and Moore, J.N. Equine peritoneal macrophage production of thromboxane A2 and prostacyclin in response to platelet-activating factor and its receptor antagonist SRI 63-441. Circ. Shock. 28: 149-158, 1989. Morrison, D.C, and Ulevitch, R.J. The effects of bacterial endotoxin on host mediated systems. Am. J. Path. 93: 527-536, 1978. Nakamura, M., Honda, Z., Izumi, T., Sakanaka, C , Mutoh, H., Minami, M., Bito, H., Seyama, Y., Matsumoto,T., Noma, M. and Shimizu, T. Molecular cloning and expression of Platelet-Activating Factor receptor from human leukocytes. J. Biol. Chem. 266: 20400-20405, 1991. Nakaya, H. and Tohse, N. Electrophysiological effects of Acetyl Glycero Ether Phosphocholine on cardiac tissues: comparison with lysophosphatidylcholine and long-chain acyl carnitine. Br. J. Pharmacol. 89: 745-757, 1986. O'Flaherty, J.T. and Wykle, R.L. Metabolic origin and fate of platelet-activating factor. Agents and Actions Suppl. 11: 59-66, 1987. O'Flaherty, J.T., Redman, J.F., Schmitt, J.D., Marshall Ellis, J., Surles, J.R., Marx, M.H., Piantadosi, C., and Wykle, R.L. l-O-alkyl-2-N-methylcarbamylglycero-phosphocholine: a biologically potent, non-metabolizable analog of platelet-activating factor. Biochem. Biophys. Res. Commun. 147: 18-24, 1987. Ordway, R.W., Singer, J.J., and Walsh, J.V. Direct regulation of ion channels by fatty acids. TINS. 14: 96-100, 1991. 136 Paletta, M.J., Abraham, S., Beatch, G.N., and Walker, M.J.A. Mechanisms underlying the antiarrhythmic properties of B-adrenoceptor blockade against ischaemia-induced arrhythmias in acutely prepared rats. Br. J. Pharmacol. 98: 87-94, 1989. Page, C.P. and Robertson, D.N. Properties of PAF relevent to asthma. Agents and Actions [suppl.] 21: 67, 1987. Page, C.P. and Abbott, A. PAF: new antagonists, new roles in disease and a major role in reproductive biology. T.I.P.S. 10: 256-257, 1989. Parratt, J.R. Neurohumoral agents and their release in shock. In: Altura, B.M., Lefer, A.M., and Schumer, W. (Eds.) Handbook of shock and trauma. Raven Press, New York, 1986. Penz, W.P., Pugsley, M.K., Hsieh, M.Z. and Walker, M.J.A. A new ECG measure (RSh) for detecting possible sodium channel blockade in vivo in rats. J. Pharmacol. Methods. 27: 51-58, 1992. Peplow, P.V. and Mikhailidis, D.P. Platelet-Activating Factor (PAF) and its relationship to prostaglandins, leukotrienes and other aspects of arachidonic acid metabolism. Prostaglandins, Leukotrienes and Essential Fatty Acids. 41: 71-82, 1990. Pignol, B., Henane, S., Chaumeron, S., Mencia-Huerta, J.M. and Braquet, P. Modulation of the priming effects of platelet-activating factor on the release of interleukin-1 from lipopolysaccharide-stimulated rat spleen macrophages. J. Lipid Med. 2: S93-S100, 1990. Piper, P.J. and Vane, J.R. The release of prostaglandins from the lungs and other tissues. Ann. N.Y. Acad. Sci. 180: 363-385, 1971. Piper, P.J. Pharmacology of leukotrienes. Br. Med. Bull. 39: 255-259, 1983. Piper, P.J and Stewart A.G. Evidence for a role for platelet-activating factor in antigen-induced coronary vasoconstriction in guinea-pig perfused hearts. Br. J. Pharmacol. 88: 81-83, 1986. Piper, P.J and Stewart A.G. Coronary vasoconstriction in the rat isolated perfused heart by platelet-activating factor is mediated by leukotriene C4. Br. J. Pharmacol. 88: 595-605, 1986. Piper, P.J and Stewart A.G. Antagonism of vasoconstriction induced by platelet-activating factor in guinea-pig perfused 137 isolated hearts by selective PAF receptor antagonists. Br. J. Pharmacol. 90: 771-783, 1987. PIante, G.E., Hebert, R.L., Lamoureaux, C., Braguet, P., and Sirois, P. Hemodynamic effects of PAF-acether. Pharmacol. Res. Commun. 18: 173-179, 1986. Prescott, S.M., Zimmerman, G.A., and Mclntyre, T.M. Platelet-activating factor. J. Biol. Chem. 265: 17381-17384, 1990. Pugsley, M.K., Salari, H., and Walker, M.J.A. The cardiac actions of platelet-activating factor in isolated rat hearts. Circ. Shock. 35: 207-214, 1991. Rabinovici, R., Yue, T-L. and Feuerstein, G. Platelet-activating factor in cardiovascular stress situations. Lipids. _26: 1257-1263, 1991. Rees, D.D., Cellek, S., Palmer, R.M.J., and Moncada, S. Dexamethasone prevents the induction by endotoxin of a nitric oxide synthase and the associated effects on vascular tone. An insight into endotoxic shock. Bioc. Biophys. Res. Commun. 173: 541-547, 1990. Rennoij, W., and Snyder, F. Biosynthesis of l-alkyl-2-acetyl-sn-glycero-3-phosphocholine (platelet-activating factor and a hypotensive lipid) by cholinephosphotransferase in various rat tissues. Biochem. Biophys. Acta 663: 545-556, 1981. Riedel, A., and Mest, H.J. The effect of PAF (platelet-activating factor) on experimental cardiac arrhythmias and its inhibition by substances influencing arachidonic acid metabolites. Prost. Leuk. Med. 28: 103-109, 1987. Robaut, C , Durand, G., James, C., Lave, D., Sedivy, P., Floch, A., Mondot, S., Pacot, D., Cavero, I., and Le Fur, G. PAF binding sites: characterization by [3H]52770 RP, a pyrrolo[l,2-c]-thiazole derivative, in rabbit platelets. Biochem. Pharmacol. 36: 3221-3229, 1987. Robertson, D.A., Genovese, A., and Levi, R. Negative inotropic effect of platelet-activating factor on human myocardium: a pharmacological study. J. Pharmacol. Exp. Ther. 243: 834-839, 1987. Robertson, D.A., Wang, D.Y., Lee, CO., and Levi, R. Negative inotropic effect of platelet-activating factor: Association with a decrease in intracellular sodium activity. J. Pharmacol. Exp. Ther. 245: 124-128, 1988. 138 Salari, H and Walker, M.J.A. Cardiac dysfunction caused by factors released from endotoxin-activated macrophages. Circ. Shock. 27: 263-272, 1989. Sanchez-Crespo, M., Alonso, F., Inarrea, P., Alvarez, V., and Egido, J. Vascular actions of synthetic-PAF-acether (a synthetic platelet-activating factor) in the rat: evidence for a platelet independent mechanism. Immunopharmacology. 4: 173-185, 1982. Sanderson, P. Do streptococci cause toxic shock? Br. Med. J. 301: 1006-1007, 1990. Schaer, G.L., Hursey, T.L., Rowan, D.A., Campbell, D.R. Heesch, CM., Mannabat, N., and Parrillo, J.E. Platelet Activating Factor is an important mediator in myocardial reperfusion injury. J. Am. Coll. Cardiol. 17: 228A, 1991. Scherf, D. and Cohen, J. The atrio-ventricular node and selected cardiac arrhythmias. Grune and Stratton, New York, 1964. Schroeder, G., Maass, B., Bartels, D., and Mannesmann, G. Measurement of the cardiodynamics, haemodynamics and the ECG in the anaesthetised rat. In: Budden, R., Detweiler, D.K., Zbinden, G. (Eds.) The Rat Electrocardiogram in Pharmacology and Toxicology. Pergamon Press Ltd., U.K. 1981. Schulz, R., Nava, E., and Moncada, S. Endotoxin and cytokines induce a calcium-dependent nitric oxide synthase in myocardium and isolated myocytes. Br. J. Pharmacol. 104(suppl.): 325, 1991. Shenep, J.L., and Mogan, K.A. Kinetics of endotoxin release during antibiotic therapy for experimental gram negative bacterial sepsis. J. Inf. Dis. 150: 380-388, 1984. Siren, A.L. and Feuerstein, G. Effects of PAF and BN 52021 on cardiac function and regional blood flow in conscious rats. Am. J. Physiol. 257: H25-H32, 1989. Smiley, P.L., Stremler, K.E., Prescott, S.M., Zimmerman, G.A., and Mclntyre, T.M. Oxidatively fragmented phosphatidylcholines activate human neutrophils through the receptor for PAF. J. Biol. Chem. 266: 11104-11110, 1991. Snyder, F. Platelet-Activating Factor and related acetylated lipids as potent biologically active cellular mediators. Am. J. Physiol. 259: C697-C708, 1990. Springer, T.A. Adhesion receptors of the immune system. Nature. 346: 425-434, 1990. 139 Sprung, C.L., Caralis P.V., Marcial, E.H., Pierce, M., Gelbard, M.A., Long, W.M., Duncan, R.C., Tendler,, M.D., and Karpf, M. The effects of high-dose corticosteroids in patients with septic shock. New Eng. J. Med. 311: 1137-1143, 1984. Stafforini, D.M., Mclntyre, T.M, and Prescott, S.M. Platelet-activating factor acetyl-hydrolase from human plasma. In: Abelson, J.N. and Simon, M.I. (Eds.) Methods in Enzymology: Arachidonate related lipid mediators. Academic Press, N.Y. 1990. Stahl, G.L., Lefer, D.J., and Lefer, A.M. PAF-acether induced cardiac dysfunction in the isolated perfused guinea-pig heart. Nauyn. Schmiedeberg Arch. Pharmacol. 336: 459-463, 1987. Stahl, G.L., Terashita, Z.I., and Lefer, A.M. Role of platelet-activating factor in propagation of cardiac damage during myocardial ischaemia. J. Pharmacol. Exp. Therap. 244: 898-904, 1988. Stewart, A.G and Phillips, W.A. Intracellular PAF regulates eicosanoid generation in guinea-pig resident peritoneal macrophages. Br. J. Pharmacol. 98: 141-148, 1989. Stewart A.G. and Piper, P.J. Platelet-activating factor induced vasoconstriction in the rat isolated perfused hearts: contribution of cyclo-oxygenase and lipoxygenase arachidonic acid metabolites. Pharmacol. Res. Commun. 18: 163-172, 1986. Stremler, K.E., Stafforini, D.M., Prescott, S.M., and Mclntyre. Human plasma platelet-activating factor hydroxylase. Oxidatively fragmented phospholipids as substrates. J. Biol. Chem. 266: 11095-11103. Sybertz, E.J., Watkins, R.W., Baum, T., Pula, K., and Rivelli, M. Cardiac coronary and peripheral vascular effects of acetyl wether phosphorylcholine in the anaesthetised dog. J. Pharmacol. Exp. Therap. 232: 156-162, 1984. Tahraoui, L., Floch, A., Mondot, S., and Cavero, I. High affinity specific binding sites for [3H]PAF in canine platelet membranes: counterparts of PAF receptors mediating platelet aggregation. Mol. Pharmacol. 34.: 145-151, 1988. Tahraoui, L., Floch, A., and Cavero, I. Functional validation of PAF receptor sites characterized biochemically by a specific and reproducible [3H]PAF binding in human platelets. J. Pharmacol. Exp. Ther. 252: 1221-1227, 1990. 140 Tamargo, J., Tejerina, T., Delgado, C., and Barrigon, S. Electrophysiological effects of platelet-activating factor (PAF-acether) in guinea-pig papillary muscles. Eur. J. Pharmacol. 109: 219-227, 1985. Terashita, Z.I., Tsushima, S., Yoskioska, Y., Nomura, H., Inada, Y., and Nishikawa, K. CV-3988 a specific antagonist of platelet-activating factor (PAF). Life Sci. 32: 1975-1982, 1983. Terashita, Z.I., Imura, Y., Nishikawa, K., and Sumida, S. Is platelet-activating factor (PAF) a mediator of endotoxic shock? Eur. J. Pharmacol. 109: 257-261, 1985. Tessner, T.G. and Wykle, R.L. Stimulated neutrophils produce an ethanolamine plasmalogen analog of platelet-activating factor. J. Biol. Chem. 262: 12660-12664, 1987. Toyofuku, T., Kubo, K., Kobayashi, T., and Kusama, S. Effects of ONO-6240, a platelet-activating factor antagonist on endotoxic shock in unanesthetised sheep. Prostaglandins. 31: 271-281, 1986. Ulevitch, R.J. Recognition of bacterial endotoxin in biological systems. Lab. Invest. 65: 121-122, 1991. Vallance, P., and Moncada, S. Hyperdynamic circulation in cirrhosis: a role for nitric oxide? Lancet. 337: 776-778, 1991. Valone, F.H., and Goetzl, E.J. Specific binding by human polymorphonuclear leukocytes of the immunological mediator, platelet-activating factor. Immunology. 48: 141-149, 1983. Venuti, M.C. Platelet Activating Factor Receptors. In: Hansen, C. (Ed.) Comprehensive Medicinal Chemistry. Pergamon Press, New York, N.Y., 1990. Vornovitskii, E.G., Ignateva, V.B., Gollash, M., Kulikov, V.I., and Lipatkina, L.I. Cardiodepressive effect of platelet-activating factor. Bull. Exp. Biol. Med. 107: 27-30, 1989. Vyhnanek, F. Acute endotoxic shock in the dog and an attempt at its therapy. Acta Univ. Med. 94: 1-11, 1979. Wahler, G.M., Coyle, D.E., and Sperelakis, N. Effects of platelet-activating factor on single potassium channel currents in guinea-pig ventricular myocytes. Mol. Cell. Biochem. 93: 69-76, 1990. Wainwright, C.C., Parratt, J.R., and Bigaud, M. The effect of PAF antagonists on arrhythmias and platelets during acute 141 myocardial ischaemia and reperfusion. Eur. Heart J. 10: 235-243, 1989. Walker, M.J.A., Curtis, M.J., Hearse, D.J., Campbell, R.W.F., Janse, M.J., Yellon, D.M., Cobbe, S.M., Coker, S.J., Harness, J.B., Harron, D.W.G., Higgins, A.J., Julian, D.G., Lab, M.J., Manning, A.S., Northover, B.J., Parratt, J.R., Riemersma, R.A., Riva, E., Russell, D.C., Sheridan, D.J., Winslow, E., and Woodward, B. The Lambeth Conventions: guidelines for the study of arrhythmias in ischaemia, infarction, and reperfusion. Cardiovasc. Res. 22: 447-455, 1988. Walker, M.J.A., Macleod, B.A., and Curtis, M.J. Animal models of human disease: Myocardial ischaemia and infarction. Comp. Path. Bull. 23.: 3-4, 1991. Whatley, R.E., Fennell, D.F., Kurrus, J.A., Zimmerman, G.A., Mclntyre, T.M., and Prescott, S.M. Synthesis of platelet-activating factor by endothelial cells: the role of G-proteins. J. Biol. Chem. 265: 15550-15559, 1990. Wiggers, C.J. The innervation of the coronary vessels. Am. J. Physiol. 24: 391-405, 1909. Williams, K.I. and Higgs, G.A. Eicosanoids and inflammation. J. Path. 156: 101-110, 1988. Williams, P.E., Leaver, H.A., Smith, G., Pacitti, N., Parratt, J.R., Ramsay, G., Ledingham, I.P., and Yap, P.C. The endotoxin receptor on monoclonal phagocytes. Microb. Immunol. 64: 11-12, 1990. Winkle, E.R. Measuring antiarrhythmic drug efficacy by suppression of asymptomatic ventricular arrhythmias. Ann. Int. Med. 91: 480-482, 1979. Wise, W.C., Cook, J.A., Eller, T., and Halushka, P.V. Ibuprofen improves survival from endotoxic shock in the rat. J. Pharmacol. Exp. Therap. 215: 160-164, 1980. Wissener, A., Sum, P.E., Schaub, R.E., Kohler, C.A., and Goldstein, B.M. Analogs of platelet-activating factor (PAF). 1. Some modifications of the alkoxy chain. J. Med. Chem. 27: 1174-1181, 1984. Wissener, A., Sum, P.E., Schaub, R.E., Kohler, C.A., and Goldstein, B.M. Analogs of platelet-activating factor (PAF). 2. Some modifications of the glycerine backbone. J. Med. Chem. 28: 1181-1187, 1985. Wissener, A., Sum, P.E., Schaub, R.E., Kohler, C.A., and Goldstein, B.M. Analogs of platelet-activating factor 142 (PAF). 3. Some modifications of the phosphocholine moiety. J. Med. Chem. 29: 328-337, 1986. Wolff, S.M. Monoclonal antibodies and the treatment of gram negative bacteremia and shock. Lancet. 324: 486-487, 1991. Wykle, R.L., Malone, B., and Snyder, F. Enzymatic synthesis of l-alkyl-2-acetyl-sn-glycero-3-phosphocholine, a hypotensive and platelet aggregating lipid. 255: 10256-10260, 1980. Yamashita, M., Homma, H., Inoue, K., and Nojimi, S. The metabolism of platelet-activating factor in platelets and plasma of various animals. J. Toxicol. Sci. 8: 177-188, 1983. Yellin, S.A., Nguyen, D., Quinn, J.V., Burchard, K.W., Crowley, J.P., and Slotman, G.J. Prostacyclin and thromboxane A2 in septic shock: species differences. Circ. Shock 20: 291-297, 1986. Zar, J.H. Biostatistical Analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J., 162-202, 1984. Zbinden, G. Spontaneous and induced arrhythmias in rat toxicity studies. In: Budden, R., Detweiler, D.K., Zbinden, G. (Eds.) The Rat Electrocardiogram in Pharmacology and Toxicology. Pergamon Press Ltd., U.K. 1981. Ziegler, E.J., Fisher Jr., C.J., Sprung, C.J., Straube, R.C., Sadoff, J.C., Foulke, G.E., Wortel, C.H., Fink, M.P., Dellinger, R.P., Teng, N.N.H., Allen, I.E., Berger, H.Y., Knatterud, G.L., LoBuglio, A.F., Smith, C.R., and HA-1 Sepsis Study Group: Treatment of gram negative bacteremia and septic shock with HA-IA human monoclonal antibodies against endotoxin. A randomized, double-blind, placebo-controlled trial. New Eng. J. Med. 324: 429-436, 1991. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items