EXPERIMENTAL AND MOLECULAR Platelet-Activating PATHOLOCiY Factor JON C. LEWIS,',~ ROBERT Effects on Pulmonary in Rabbits JOSEPH L. 38, loo- 108(1983) T. WYKLE,~ O’FLAHERTY, AND CHARLES Ultrastructure E. MCCALL, M. GENE BONDS Departments of ‘Pathology, Medicine, 3Biochemisrty, and 4Comparative Medicine, Wake Forest Universily, Bowman Gray School of Medicine, Winston-Salem, North Carolina 27103 Received June 8, 1982 Morphometric analysis of endothehum in rabbit lungs demonstrated a dramatic effect of platelet-activating factor (PAF) on the ultrastructure of pulmonary vasculature. Flasmalemmat vesicles in capillaries were increased both in size (638 A FAF vs 538 A control) and in number (386/pm3 cytoplasm PAF vs 125/pm3 cytoplasm control) when PAF was administered as a single acute low dose (x < 3 /&kg). High concentrations of PAF (X > 3 &kg) as a single acute dose also increased vesicle number (203/pm3 cytoplasm), but frequently precipitated the respiratory distress syndrome. Chronic administration of PAF with daily doses over periods of either 3.5 or 7 weeks resulted in changes paralleling the acute observations, but did not lead to more extensive lung disease. INTRODUCTION Platelet-activating factor (1-O-alkyl-2acetyl-sn-glycero-3-phosphocholine; neutropenia, and ultimately with severe respiratory distress resulting in death (McManus ef al., 1980; Pinckard et al., 1977). Furthermore, it has been demonstrated that PAF is the mediator of basophil action in IgE-induced systemic anaphylactic shock in rabbits (Benveniste et al., 1977), and may be associated with a variety of allergic and nonallergic disease processes (Pinckard et al., 1977). Several investigators have reported the response of platelets and neutrophils to PAF either in vitro (Benveniste et al., 1977; Cazenave et al., 1979; Fesus et al., 1977; O’Flaherty et al., 1981) or in viva (McManus et al., 1980; Pinckard et al., 1977), and it has recently been shown that aggregation of circulating platelets following intravascular administration of PAF was accompanied within 30 set by an increase in plasma levels of platelet factor 4 and transient sequestration of platelet aggregates within the lung (McManus et al., 1980). Relatively little, however, is known about the cellular pathology of the pulmonary response. In large part, this lack of information has been due in the past to the limited availability of naturally occurring (cellular derived) PAF (McManus et al., 1980). Alleviation of this limitation has recently occurred, for PAF from rabbit basophils has been characterized and subsequently synthesized from beef heart plasmalogen (O’Flaherty et al., 1981; Hanahan et al., 1980). The present study was undertaken to determine the effects of PAF on vascular endothelium. Our studies, based upon the use of synthetic PAF, clearly document an acute effect of PAF when administered as a single dose, but suggest that this acute response does not lead to more extensive lung disease. (PAF) in rabbits has been associated with acute thrombocytopenia, MATERIALS AND METHODS Thirty mature New Zealand white rabbits ranging in weight from 3.1 to 4.2 kg were used in these studies. The animals, obtained from Franklin’s Rabbitry in 2 To whom all correspondence should be addressed. 100 0014-4800/83iO10100-09$03.00/O Copyright 0 1983 by Academic Press, Inc. All rights of reproduction in any form reserved. PLATELET-ACTIVATING FACTOR EFFECTS ON RABBIT LUNGS 101 Raleigh, North Carolina were medically checked by the comparative pathology unit at Bowman Gray School of Medicine prior to use in one of the two groups described below. The first group, composed of 18 animals was used in acute studies involving a single administration of PAF. The remaining 12 animals were entered into a study involving chronic PAF administration over a period of 7 weeks. PAF for all of the studies was prepared from beef heart choline phospholipids by a modification of the method of Blank et al. (1979), was then purified by preparative thin-layer chromatography, and before use was analyzed by gas-liquid chromatography and mass spectroscopy (Vargaftig er al., 1980). The purified PAF was evaporated under a stream of nitrogen, taken up in saline containing rabbit serum albumin (RSA, 2.5 mg/ml), and administered via an ear vein. The 18 animals used for the acute studies were handled as follows: (1) 4 animals served as controls and were given an infusion of RSA (2.5 mg in 1 ml of saline); (2) 8 animals received a low dose of PAF (0.15 -2.4 pg/kg): and (3) 6 animals received a high dose of PAF (3.0- 10 pgkg). Fourteen of the eighteen rabbits included in the acute studies (4 RSA controls and 10 PAF treated) were anesthetized by intramuscular ketamine (100 mg/kg body wt) plus rompum (8.9 mg/kg body wt) approximately 4 min after PAF or saline injection. Ten minutes thereafter, these animals were sacrificed by cardiac intraventricular perfusion at 110 mm Hg with phosphate (0.1 M)-buffered (pH 7.2) paraformaldehyde (4%) containing glutaraldehyde (0.25%). The remaining four animals, receiving 3, 3.8, 7.5, and 10 pg PAF/kg, respectively, experienced severe respiratory distress; therefore, they were anesthetized immediately and sacrificed without the lo-min period described above. The intraventricular perfusion was complemented by low-pressure infusion of fixative into the lungs to preserve alveolar distention. The lungs were excised following primary fixation, and a uniformly selected peripheral lobe was dissected. The segments were then secondarily fixed in 0.1 M phosphate-buffered (pH 7.2) 1% osmium tetroxide, dehydrated, and embedded in epoxy resin for transmission electron microscopy (TEM). Morphometric analysis of the TEM micrographs to determine the distribution of cellular organelles was done by the point-intercept method. Tissue segments from spleen, liver, lung, and aorta from each of the animals were excised and immersed in buffered formalin for histologic analysis. Sections from each of the tissues were stained with routine H & E in addition to Sudan IV, Carstair’s trichrome, and Mallory’s PTAH. Preliminary results from the acute studies described above indicated an adverse effect of PAF on the integrity of pulmonary vasculature (see Results for full details) with the most dramatic changes found in endothelial integrity. Since endothelial changes have been associated with exacerbation of atherosclerosis in major arteries and may be related to both edema and interstitial fibrosis in the lungs, we decided to conduct a study to explore the long-term effects of PAF administration on both the lungs and major arteries. Twelve mature rabbits were used in the experiment. The animals were placed on a rabbit chow diet supplemented with 0.5% cholesterol in lard for 2 weeks prior to administration of PAF. Following the diet equilibration period, eight of the animals received injections of PAF (1.5 pg PAFikg body wt) on a daily basis for 5 days (Monday-Friday) followed by 2 days with no PAF administration (Saturday and Sunday). This 7-day cycle was repeated until half of the animals had received 17 PAF injections. The other four PAF-treated animals remained on the cycle until they had received 35 injections of PAF. The final four animals received no PAF and served as hyper- LEWIS ET AL. FIG. 1. Representative transmission electron micrograph of lung from a control animal which had been infused with rabbit serum albumin, the carrier used for PAF studies. The septa are characterized by uniform capillaries containing red blood cells (R). x4000. lipidemic controls for the PAF rabbits. Total plasma cholesterol in all animals was monitored on a weekly basis. On the day of the final PAF treatment the animals were necropsied as described for the acute experiments. RESULTS Acute PAF Administration Consistent with the reports of others (McManus et al., 1980; Pinckard et al., 1977) PAF administration resulted in a clinical pattern which resembled IgE anaphylaxis. This was transient in the four animals receiving low levels of PAF (0.15, 0.5, 1.5, and 2.4 pg/kg), but resulted in acute respiratory distress in four of the six animals given higher concentrations of PAF (3.0, 3.8, 7.5, and 10 pg/kg). The remaining two animals, which received high PAF (3.0 and 5.0 pg/kg), did not display the acute reaction; however, they did remain in mild respiratory distress throughout the experimental period. This clinical pattern, observed in the animals prior to sacrifice, paralleled the appearance of the lungs upon necropsy, at which time pulmonary contraction resulting in atelectasis was noted in lungs from five of the six animals receiving the high PAF concentrations. Multiple diffuse petechia were also observed on the lungs of PAF-treated animals. Histologic evaluation revealed no PAF effects in the spleen, but kidney, liver, and lung were characterized by the presence of platelet emboli and focal hyperemia. Occasionally, focal atelectasis was observed in the lungs of the treated animals. When observed by electron microscopy, a dramatic contrast was noted among the treatment groups. Whereas, the lungs from the RSA control animals (Fig. 1) had normally dilated alveoli and capillaries containing randomly dispersed red blood cells, the capillaries from the animals given PAF at low concentrations PLATELET-ACTIVATING FACTOR EFFECTS ON RABBIT LUNGS 103 FIG. 2. A portion of lung from an animal following administration of PAF at the low dose level (1.5 &kg). The septal capillaries are filled with aggregated platelets (P) and distorted erythrocytes (R). Compare to Fig. 1. x4000. contained aggregated platelets and marginated leukocytes (Fig. 2). In marked contrast to lungs from both the control and low PAF-treated animals, the lungs from animals receiving high concentrations of PAF were contracted, an effect which obscured identification by light microscopy of alveoli and capillaries alike (Fig. 3). The most consistent effect of acute PAF treatment was evidenced by TEM. As shown in Fig. 4, perturbations of plasma membranes of capillary endothelial cells were noted in all PAF-treated rabbits. This resulted in lobular protrusion of the cells into the luminal spaces. Although the endothelial alterations in animals given high concentrations of PAF could not be dissociated from pulmonary contraction, these two phenomena seemed to occur independently with low concentration PAF (x < 3 pg/kg) treatment. Under these conditions endothelial alterations were typically observed in vessels which had no apparent reduction in luminal size (Figs. 2, 4). The most dramatic ultrastructural effect of PAF was a vacuolation and increased vesiculation of endothelial cells. As shown in Fig. 4, this hypervesiculation of endothelial cells, particularly in areas of luminal protrusions, resulted in a foamy cytoplasmic appearance. Although increased vesiculation was most pronounced in the areas of cytoplasmic lobes, increases in both the number and size of plasmalemmal vesicles were consistently found in pulmonary capillaries from all PAF-treated animals. When analyzed morphometrically (see Table I), the number of vesicles in the low PAF-treated animals was three times that found in RSA-treated control rabbits. This increase in vesicle number, which was significant at the P < 0.01 level, when combined with a 20% increase in average vesicle 104 LEWIS ET AL. FIG. 3. A transmission electron micrograph showing a portion of lung from an animal following administration of PAF at the high dose level (5 /.&kg). The septal configuration is distorted due to the tortuous shape of endothelial cells (E) shown in oblique/longitudinal section. The undulated endothelial cell plasma membranes protrude into vessel lumina, which contain leukocytes (L) and aggregated platelets (P). x8000. PLATELET-ACTIVATING FACTOR EFFECTS ON RABBIT LUNGS 105 FIG. 4. Transmission electron micrographs illustrating the endothelial cell hypervesiculation characteristic to PAF treatment. (A) Low magnification micrograph in which the foamy appearance characteristic to the post PAF endothelium (E) is evident. A leukocyte (L) is within the lumen of the vessel. x 10,500. (B) High magnification micrograph of normal lung capillary endothelium illustrating relatively normal vesiculation. ~43,000. (C) Capillary endothelium from a low dose PAF treated animal. Compare the degree of vesiculation to that illustrated in B. x 19,000. (D) Isolated region of endothelium from a high dose PAF treated animal illustrating the membrane (M) damage observed in these animals. x33,500. 106 LEWIS Morphometric Treatment n’L Control PAF < 3 /&kg PAF 2 3 &kg 3 4 4 Analysis ET AL. TABLE I of Vesiculation Vesicle (4 Vesicles/km* 7.5 k 1.2 23.2 f 2.5 12.2 f 2.2 of Endothelial size Cells Percentage surface* 538 638 534 fl Number of animals from which lung ultrastructure was morphometrically * Represents the percentage ofendothelial cytoplasm which was occupied ’ Figure shown was extrapolated from data and assumes an average TEM Vesicles/pm3 1.7 7.4 2.7 analyzed. by plasmalemmal section thickness 125 386 203 vesicles. of 650 A. diameter (Table I) resulted in a fourfold increase in the percentage of cytoplasm occupied by vesicles. The size of vesicles in animals receiving high concentrations of PAF did not differ from the control level, but the number of vesicles was significantly increased (P < 0.01). This resulted in a 58% increase in cytoplasmic vesicle volume in the high PAF group. Chronic PAF administration The administration of PAF over periods of either 22 days (17 injections) or 48 days (35 injections) had no consistent effect upon either the extent of cholesterol-induced atherosclerosis or on the degree of lung fibrosis. Tissues from the chronic treatment groups had platelet emboli, focal hyperemia, and focal atelectasis as observed in the acute animals. Lipid accumulation as evidenced by Sudan staining was evident in the livers and aortas from both control and PAFtreated rabbits. This observation was consistent with the rise in the TPC from the baseline mean value of 78 mg/dl to the necropsy values of 347 mg/dl (17 injections/animals) and 1038 mg/dl (animals receiving 35 injections of PAF). DISCUSSION The effects of PAF on general pulmonary morphology and endothelial ultrastructure, as documented in this and other reports, are consistent with the pathophysiology of anaphylaxis. Systemic infusion of antigen into sensitized rabbits causes thrombocytopenia, neutropenia, constriction of pulmonary arteries and bronchi, increased vascular permeability, and cardiorespiratory collapse (Vargaftig et al., 1980; Henson et al., 1971). PAF has induced these same changes (McManus et al., 1980; Pinckard et al., 1977; Vargaftig et al., 1980; Henson et al., 1971; O’Flaherty et al., 1981). In addition to these previously reported effects, we found in the present study that at the cellular level PAF causes a reduction in pulmonary capillary luminal area and vacuolation of the cytoplasm. This suggests a pathologic response of these cells to PAF administration. Based upon the current understanding of endothelial function including its complex role in plasma homeostasis, plasma protein transport, and the hemostatic process, the alterations reported in this study have far-reaching implications. Particularly noteworthy is the increase in plasmaIemma1 vesicles. Since similar vesicles have been associated with transendothelial transport of several plasma constituents (Palade et al., 1979), the two- to fourfold increase associated with PAF treatment could explain the enhanced permeability to immune complexes reported by Hensen (1971). Although we do not provide direct evidence to substantiate increased capillary PLATELET-ACTIVATING FACTOR EFFECTS ON RABBIT LUNGS 107 permeability with PAF treatment, the increased number and size of vesicles documented in the present report parallel the observations of DeFouw and Berendsen (1979), who reported a twofold increase in plasmalemmal vesicle volume with pulmonary edema in isolated perfused dog lungs. The morphometric technique used in our studies differed from that employed by DeFouw and Berendsen, therefore direct comparison of their data to OUR 09 @an not be made. Our morphometric data can, however, be compared to tha$r$cently reported by Gil and Silage (1980), who found in normal rabbit lungs 13,‘\ vesicles/pm3 of endothelial cytoplasm. This value compares favorably to the 12%yesicleslpm3 in our own study and further emphasizes the magnitude of vesicle changes observed with PAF administration. The temporal relationships among pulmonary constriction, platelet aggregation, leukocyte margination, and endothelial alteration are not defined in our study; however, the contrasting results obtained with low PAF (<3 pg/kg) as compared to high PAF (23 pg/kg) treatment suggests that the endothelial response, platelet aggregation, and perhaps leukocyte margination can occur independent of pulmonary constriction. Whereas with high concentrations of PAF the most pronounced effect was the immediate alveolar and capillary constriction, it was with low concentration PAF that the greatest degree of endothelial vesiculation was found. Thus, the endothelial cell may, along with platelets and neutrophils be an important target for PAF, and alteration of endothelial function may underlie the breakdown in pulmonary capillary integrity and lung physiology seen in anaphylaxis. Although hypervesiculation has been consistently found with acute PAF administration, the long-term significance of this endothelial change remains unknown. Since plasmalemmal vesicles have been implicated in transendothelial transport (Palade et al., 1979) and since pulmonary fibrosis and atherosclerosis are associated with enhanced movement of plasma constituents both into and across the endothelium, it seemed reasonable to suggest that either fibrosis or atherosclerosis or both would be exacerbated by PAF treatment. The hypervesiculation resulting from PAF administration, however, appears not to contribute to these disease processes. Conceivably, the duration of our chronic PAF administration studies (48 days maximum) was too short for PAF effects to be manifested. This probably is not the case, for recent studies in our laboratories using other agents suggest that the time (48 days) was sufficient for fibrosis to have occurred. We have found in our recent studies that administration of Phorbol myristate acetate to rabbits over a period of 2 weeks results in significant pulmonary fibrosis as evidenced by histologic evaluation (McCall et al., unpublished observations). Similarly, numerous investigators have documented atherosclerotic changes in aortas of rabbits following short-term diet of cholesterol. Although the experimental time period in the chronic administration studies appears to have been sufficient, it is possible that the endothelial hypervesiculation is a transient phenomenon and is evidenced only immediately following PAF administration. All of the animals in the present experiment were necropsied within 15 min of the final PAF injection. Both thrombocytopenia and neutropenia are also found at this time. However, circulating platelet and neutrophil levels in PAF-treated rabbits return to normal within 20 min McManus et al., 1980). Conceivably, endothelial integrity is restored within the same period. If this is the case, it is possible that the transient endothelial hypervesiculation (20-30 min) has no long-term consequence . 108 LEWIS ET AL. ACKNOWLEDGMENTS The authors are grateful to Ms. Melanie S. White, Ms. Sue Cousert, and Mr. Richard G. Taylor for their technical assistance, and to Mrs. Bobbie Lindsay for assistance with manuscript preparation. This work was supported by National Institutes of Health Grants AI-09169. HL-16769, HL-26257, HL-26818. HL-14164 (SCOR). and a grant from the National Dairy Council. REFERENCES J., CAMUSSI, J., and POLONSKY, J. (1977). Platelet activating factor. In “Monographs of Allergy 12” (P. Kall6s, T. M. Inderbitzin, 2. Trnka, A. L. de Week, and B. H. Waksman, eds.) pp. 138 142. Karger, Basel. BLANK, M. L., SNYDER, F., BYERS, L. W., BROOKS, B., and MUIRHEAD, E. E. (1979). Antihypertensive activity of an alkyl ether analog of phosphatidylcholine. Biochem. Biophys. Res. Commun. 90, 1194- 1200. CAZENAVE, J., BENVENISTE, J., and MUSTARD, J. F. (1979). Aggregation of rabbit platelets by platelet-activating factor is independent of the release reaction and the arachidonate pathway and inhibited by membrane-active drugs. Lab. Invest. 41, 275-285. DEFOIJW, D. 0. and BERENDSEN, P. G. (1979). A morphometric analysis of isolated perfused dog lungs after acute oncotic edema. Microvas. Res. 17, 90- 103. F&Us, L., SCABA, B., and MUSZBEK, L. (1977). Platelet activating factor, the trigger of haemostatic alterations in rat anaphylaxis. C&n. Exp. Immunot. 27, 512-515. GIL, J., and SILAGE, D. A. (1980). Morphometry of pinocytotic vesicles in the capillary endothelium of rabbit lungs using automated equipment. Circ. Res. 47, 384-391. HANAHAN, D. J., DEMOPOULOS, C. A., LIEHR, J., and PINCKARD, R. N. (1980). Identification of platelet activating factor isolated from rabbit basophils as acetyl glyceryl ether phosphorylcholine. J. Biol. Chem. 255, 5514-5516. HENSON, P. M., and COCHRANE, J. G. (1971). Acute immune complex disease in rabbits: The role of complement and of a leukocyte-dependent release of vasoactive amines from platelets. j. Exp. Med. BENVENISTE, 133,.554-571. L. M., HANAHAN, D. J., DEMOPOULOS, C. A., and PINCKARD, R. N. (1980). Pathobiology of the intravenous infusion of acetyl glyceryl ether phosphorylcholine (AGEPC), a synthetic platelet-activating factor (PAF), in the rabbit. J. Immunol. 124, 2919-2924. O’FLAHERTY, J. T., MILLER, C. H., LEWIS, J. C., WYKLE, R. L., BASS, D. A., MCCALL, C. E., WAITE, M., and DECHATELET, L. R. (1981). Neutrophil responses to platelet-activating factor. Inflammation 5, 193-201. O’FLAHERTY, J. T., WYKLE, R. L., MILLER, C. H., LEWIS, J. C., WAITE, M., BASS, D. A., MCCALL, C. E., and DECHATELET, L. R. (1981). 1-0-Alkyl-sn-glyceryl-3-phosphorylcholines: A novel class of neutrophil stimulants. Amer. J. Pathol. 103, 70-78. PALADE, G. E., SIMIONESCU, M., and SIMIONESCU, N. (1979). Structural aspects of the permeability of the microvascular endothelium. Acta Physiol. Stand. Suppl. 463, 11. PINCKARD, R. N., HALONEN, M., PALMER, J. D., BUTLER, C., SHAW, J. O., and HENSON, P. M. (1977). Intravascular aggregation and pulmonary sequestration of platelets during IgE-induced systemic anaphylaxis in the rabbit: Abrogation of lethal anaphylactic shock by platelet depletion. J. MCMANUS, Immunol. VARGAFTIG, 119, 2185-2193. B. B., LEFORT, induces a platelet-dependent rivatives. Eur. J. Pharmacol. M., and BENVENISTE, J. (1980). Platelet-activating factor bronchoconstriction unrelated to the formation of prostaglandin de65. 185- 192. J., CHIGNARD,