om 3 0 20013 102700 OVERDUE FINES: 25¢ per day per item 'an'xus LIBMRY MATERIALS. Place in book return to move charge from circulation fiscal-d: MICROVASCULAR PERMEABILITY: EFFECTS OF PROSTAGLANDIN E1 BY James John Maciejko A DISSERTATION . Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology 1980 ABSTRACT MICROVASCULAR PERMEABILITY: EFFECTS OF PROSTAGLANDIN El BY James John Maciejko A wide variety of experimental evidence indicates that prosta- glandins are indirectly involved in the inflammatory process. This evidence includes the fact that prostaglandins of the "E" series can be recovered from a sundry of inflammatory exudates, and that non— steroidal, anti-inflammatory agents (cyclo—oxygenase inhibitors) can attenuate all forms of inflammation. However, discrepancy prevails concerning the direct microvascular permeability inducing effect of exogenous PGE1 or PGE when administered into a tissue or organ. 2 To study the direct effects of exogenous PGE1 on vascular resistance and microvascular permeability, the canine forelimb tech- nique was employed. The sixty minute, local intra-arterial infusion of PGE1 (2.0 to 16.0 ug/min) into forelimbs perfused either naturally or at constant brachial artery inflow produced profound vasodilation and increases in lymph total protein concentration, yet only minute dose independent increases in lymph flow rate. The increase in lymph flow rate was similar in magnitude regardless of the method of forelimb perfusion, i.e., natural or constant brachial artery inflow. PGE 1 (16.0 or 32.0 Ug/min) also generated marked elevations in forelimb James John Maciejko weight due to edema formation. The weight increases were substantially greater under natural inflow conditions. Skin small vein pressures were markedly elevated in the fore- limbs perfused at natural inflow, although they failed to change relative to control in the limbs perfused at constant inflow. This suggests an increase in capillary hydrostatic pressure with natural inflow and no increase with constant inflow. Interestingly, local, intra—arterial infusion rates between 2.0 to 16.0 ug/min yielded sig- nificant increases in mean systemic arterial blood pressure. In contrast, increasing the local intra—arterial infusion rate to 32.0 ug/min caused sustained reductions in mean systemic arterial blood pressure. All infusion rates studied produced a significant increase in hematocrit. Following pre-treatment with indomethacin (5 mg/kg) for one hour, the local, intra-arterial infusion of PGE1 (16.0 ug/min) pro- duced vasodilation and increases in lymph total protein concentration similar to the values obtained in the absence of indomethacin. There were no significant changes in lymph flow rate observed in the absence of indomethacin, but dramatic increases in lymph flow rate were obtained under both natural and constant brachial artery inflow condi- tions following pre-treatment with indomethacin. Therefore, these data suggest that PGE may well function as an 1 inflammatory mediator in the dog, yet its role may be different and its mechanism of action considerably more complex than the putative mediators of inflammation (i.e., histamine and bradykinin). DEDICATION To Linda and my parents their love and understanding have lit my way to accomplishment and joy ii ACKNOWLEDGEMENTS I would like to express my deep appreciation to the following people for their many contributions in this endeavor: Mr. Donald Anderson, Miss Amylou Davis, Mr. Charles L. Brooks, Dr. Douglas L. Marciniak, and Dr. Daniel Sak. A very special thanks is extended to Dr. Thomas G. Bell, Dr. George J. Grega, and Dr. William L. Frantz. iii INTRODUCTION. . . . . . . SURVEY OF THE LITERATURE. STATEMENT OF THE PROBLEM. METHODS I O O O O O I O O ESULTS O O O O O O O I O Tale III 0 O O O O T¢le IV I O O O 0 Table V. I O O O 0 Table VI 0 O O O 0 Table VII 0 O O O O Captopril Studies. DISCUSSION. . . . . . . . APPENDICES. . . . . . . . BIBLIOGRAPHY. . . . . . . TABLE OF CONTENTS 0 O O O O O O O O O O O O O O O O O O 1 O O O O O O O O O O O O O O O O O O O 3 O O O O O O O O O O O O O O O O O O O 41 O O O O O O O O O O O O O O O O O O O 43 . . . . . . . . . . . . . . . . . . . 50 O O O O O O O O O O O O O O O O O O O 50 O O O O O O O O O O O O O O O O O O O 51 O I O O O I O O O O I O O O O O O O O 51 O O O 0 O O O O O O O O O O O O O O O 52 O O O O O O O O O O O O O O O C O O O 53 O O O O O O O O O O O O O O I O O O O 54 O O O O O O O O O O O O O O O O O O O 63 . . . . . . . . . . . . . . . . . . . 71 O O O O O O O O O O O O O O O O I O O 110 iv Table II III IV VI VII A-III LIST OF TABLES Ultrastructure of the capillary wall in various tissues and organs (130) . . . . . . . . . . . . . . . . . Stimulants of acylhydrolase (121). . . . . . . . . . . . Effects of locally infused PGEl (2.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph pro- tein transport, protein concentration of lymph and plasma, vascular pressures and hematocrit (n=6) (NF = natural flow, CF = constant flow). . . . . . . . . . Effects of locally infused PGEl (4.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph pro- tein transport, protein concentration of lymph and plasma, vascular pressures and hematocrit (n=6) (CF = constant flow) . . . . . . . . . . . . . . . . . . . Effects of locally infused PGEl (16.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph pro- tein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (NF = natural flow, CF = constant flow). . . . . . . . . . Effects of locally infused PGEl (32.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph pro- tein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (NF = natural flow, CF = constant flow). . . . . . . . . . Effects of indomethacin (5 mg/kg) injected intravenously followed by PGEl (16.0 ug/min, I.A.) infused locally into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hemato- crit (n=6) (NF = natural flow, CF = constant flow) . . . Effects of locally infused PGEl (2.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures and hematocrit (n=6) (natural flow) . . Page 26 55 56 S7 59 61 72 Table B-III B-IV A-VI B-VI A-VII B-VII Page Effects of locally infused PGEl (2.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures and hematocrit (n=6) (constant flow). . 76 Effects of locally infused PGEl (4.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures and hematocrit (n=6) (constant flow). . 80 Effects of locally infused PGEl (16.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (natural flow) . . . . . . . . . . . . . . . . . . . . . . 84 Effects of locally infused PGEl (16.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (constant flow). . . . . . . . . . . . . . . . . . . . . . 88 Effects of locally infused PGEl (32.0 UQ/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (natural flow) . . . . . . . . . . . . . . . . . . . . . . 92 Effects of locally infused PGEl (32.0 ug/min, I.A.) into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (constant flow). . . . . . . . . . . . . . . . . . . . . . 96 Effects of indomethacin (5 mg/kg) injected intravenously followed by PGEl (16.0 ug/min, I.A.) infused locally into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (natural flow) . . . . . . . . . . . . . . . . . . . . . . 101 Effects of indomethacin (5 mg/kg) injected intravenously followed by PGEl (16.0 ug/min, I.A.) infused locally into the forelimb for 60 minutes on lymph flow, lymph protein transport, protein concentration of lymph and plasma, vascular pressures, forelimb weight and hematocrit (n=6) (constant flow). . . . . . . . . . . . . . . . . . . . . . 105 vi LIST OF FIGURES Figure Page I Linear relation between capillary hydrostatic pressure and fluid movement across the microvascular membrane (67) 0 o o o o o o o o o o o o o o o o o o o o o o o o o o 12 II Structure of prostanoic acid . . . . . . . . . . . . . . . 22 III Biosynthesis of prostaglandins from dihomo-y-linolenic acid (C20:3), arachidonic acid (C20.4) and eicosopentae- nOiC aCid (C20:5) o (53) o o o o o a o o o o o o o o o o o 24 IV Metabolic pathways of arachidonic acid . . . . . . . . . . 29 V Schematic diagram of the canine forelimb technique . . . . 46 vii LIST OF SYMBOLS AND ABBREVIATIONS NF = naturally perfused forelimb CF = forelimb perfused at constant inflow cm = centimeter mm = millimeter um = micrometer gms = grams mgm = milligram pg = microgram m1 = milliliter min = minute A = Angstrom mm Hg = millimeters of mercury pressure k = Capillary Filtration Coefficient (CFC) viii I NTRODUCT ION The prostaglandins are a group of metabolic products from membrane unsaturated fatty acids which possess diversified physiological and pathological roles. Since their discovery in 1935 by Euler (28,29), there has accumulated a variety of evidence implicating these autocoids in the inflammatory process (118). The supportive documentation includes that: l) exogenously administered prostaglandins, primarily of the "E" series, increase microvascular permeability in several species; 2) prostaglandins are released and can be recovered from several different types of inflammatory exudates; 3) non-steroidal, anti-inflammatory drugs (i.e., indomethacin, aspirin, naproxen) inhibit the biosynthesis of prostaglandins, thus attenuating inflammation. Although no controversy prevails among investigators concerning prosta- glandin release in inflammation and the ability of non-steroidal anti—inflammatory drugs to inhibit prostaglandin biosynthesis, disagree- ment does exist with regard to the extent that exogenously administered PGE or PGE can augment microvascular permeability. l 2 In the rat, mouse, hamster and guinea pig, PGE1 causes an increase in permeability due to a direct action on the microvascular membrane resulting in the formation of gaps between endothelial cells in the post-capillary venules (24,59,113,ll4). This appears to be similar to the mechanism by which histamine and bradykinin increase microvascular peremeability. In contrast, however, others argued that PGEl has no permeability inducing effects (52,120). 1 2 Even though most reports show that PGE1 can cause edema in the rodent, the concentrations required are much higher than those recovered from inflammatory exudates. This observation has led to the hypothesis that prostaglandin generation contributes to the edema of inflammation by potentiating or increasing the vascular leakage induced by other putative mediators, including histamine and bradykinin (113,126). In the few permeability studies reported for dogs, PGE1 failed to increase microvascular permeability to macromolecules (25,41,58). However, in these experiments doses of PGE1 were small and the duration of infusions was short. Also, a possible interference by other prosta- glandins or eicosanoids produced in response to the exogenous PGE 1 could have antagonized the effect of PGE on the microvascular membrane. 1 Consequently, these factors may have contributed to the failure to obtain positive results. With the preponderance of supportive evidence in the rat, mouse, hamster and guinea pig, and the valid criticisms for the canine studies being considered, a systematic examination of the effects of PGE on- 1 fluid filtration and macromolecular efflux is warranted. Therefore, the present study is an attempt to systematically characterize the permeability properties of one hour, local infusions of PGE1 into the canine forelimb over a dose range of 2.0 ug/min. to 32.0 UQ/min. Several experiments will test systemic indomethacin pretreatment prior to the onset of the local PGEl infusion. This was designed to suppress endogenous prostaglandin synthesis and release which might interfere with the authentic microvascular effects of exogenous PGEl. SURVEY OF THE LITERATURE The microcirculation is the end organ of the cardiovascular system and performs two vital functions. The primary one is cell nutrition and the second is body fluid balance. In general, this system operates by delivering blood to tissues and organs commensurate with their metabolic demands, and by regulating the exchange of fluid between the intravascular and extravascular compartments. To understand these processes better, investigators have attempted to define the anatomical features of capillary and lymphatic vessels, and to experimentally manipulate the distribution of plasma across the vessel walls. As better microscopes became available, additional potential pathways for solute and/or solvent movement were postulated. The term permeability was introduced to explain the end result of solute and solvent distribution across capillaries within the limits of the experimental conditions. However, such studies do not directly explain the relationship between the known microstructure and the steady-state values of body fluid distribution. This review of the literature will deal first with the current status of the anatomical features of capillary and lymphatic vessels, and then with some of the analytical processes of plasma and interstitial fluid dynamics that are pertinent to this dissertation. Microcirculation implies blood flow through small vessels at the level of the capillary. The term capillary was first introduced in 4 1661 by Marcello Malpighi in the frog lung (73). Malpighi demonstrated that these "porosities in the flesh" were long, thin-walled tubes, whereby blood flows from artery to vein. Even though centuries have elapsed since Malpighi's initial observations, a persistent diversity of opinions still exists regarding the descriptions of the patterns of distribution and structure of the microcirculatory system. The capillary vessel of most mammalian tissues and organs con- sists of a single layer of flattened endothelial cells bound together within an acellular basal lamina. The materials comprising the basal lamina (mainly mucopolysaccharides) are presumably secreted by these cells. Substances may penetrate the endothelium by passing directly through the cells, that is, through two thicknesses of cell membrane and a layer of cytoplasm, or by penetration without entering the cell. This latter route occurs by extracellular channels or through junctions between cells which are referred to as pores (94). The work of Pappenheimer (90), Renkin (93) and Solomon (110) provides the foundation for the pore theory. According to this theory, hydrOphilic materials traverse the endothelial barrier through water- filled channels or pores, under the driving force of hydrostatic pressure and concentration gradients. This concept unifies many experimental observations and provides the physical basis for the phenomenon of molecular sieving, or the restriction of molecules whose diameter is larger than the pore. The mechanism of molecular sieving has been characterized by Grotte (43) and Arturson (4), who studied the exchange of dextran frac- tions of different molecular sizes. Their work leads to the conclusion that there must be a "large pore system" and a "small pore system." The diameter of the large pores ranges from 250 A to 700 A. These pores 5 provide the main pathway for the permeation of plasma proteins and other macromolecules (43). Large pores are presumed to be present mostly in the venous portion of the microcirculation and to account for the higher permeability of the venular segment to proteins. Using a similar experimental approach, Landis and Pappenheimer (67) con- cluded that the small pores occur as cylinders or slits with a radial range of 45 A to 90 A. The so-called “large pore system" appears to be only in a l:34,000 ratio to the "small pore system" (53). The pore theory implies only that the capillary wall acts as if it had pores. Electron microscopists have sought evidence for such pathways in view of the postulated pore diameters. Electron micro- graphs have revealed that the ultrastructure of capillaries is not the same in all parts of the circulation (77). Three different types of capillary walls have been identified based on the number of pores in the endothelia: they are termed continuous (non-fenestrated), dis- continuous, and fenestrated. Table I summarizes the various capillary types in tissues and organs. In analogy with larger blood vessels, the capillary wall of the continuous type consists of three layers or tunics (86,88). The come plete inner tunic is formed by the endothelial cells, which are markedly attenuated--O.20 to 0.40 uM thick over all portions except in the nuclear regions. A second, also complete, middle tunic is formed by the basement membrane, which envelopes scattered pericytes. Pericytes are precursors of smooth muscle cells, and very little is known about their role within the general functional scheme of the vascular wall. The outer layer consists of scattered bundles of collagen and elastic fibers, and of single connective tissue cells. Detailed accounts of Table I. Ultrastructure of the capillary wall in various tissues and organs (130) Tissue Endothelium All endocrine glands fenestrated Kidney: fenestrated Glomerulus fenestrated Tubules fenestrated Medulla fenestrated CNS blood-brain barrier continuous Muscle: continuous Skeletal continuous Cardiac continuous Smooth continuous Lung continuous Sinusoids of: Liver discontinuous Spleen discontinuous Bone marrow discontinuous Adipose tissue continuous Connective tissue continuous Intestinal mucosa fenestrated Dermis continuous & fenestrated 7 of the ultrastructure of continuous endothelium are given in the literature (17,97,106). The most characteristic feature of continuous endothelial cells is the presence of varying numbers of micro-pinocytotic vesicles, 700 A in diameter (87). Most of these vesicles appear as invaginations of the plasma membrane on either surface of the cell and therefore are designated as plasmalemmal vesicles. A large number of closed vesicles not in contrast with the plasma membrane also can be found in the endothelial cell cytoplasm. The margins of adjacent capillary endothelial cells overlap or interdigitate (88). Generally, the intercellular space is obliterated within its third tunic (luminal side) by an area of close approximation or an apparent fusion of the outer leaflets of the adjacent plasma membranes (87,88). Differentiations of the structure of the junctions have been reported in consecutive segments of the microvasculature (87). Simionescu et al. (104,105) studied the distribution after intravascular injection of three tracer materials of various effective radii (10 A—ZO A) within the endothelium of rat skeletal muscle capil— laries. All three tracers were excluded from the narrow junctional regions, but did appear within cytoplasmic vesicles. Within 30 seconds of administration, the vesicles on the luminal (internal) surfaces of the endothelial cells were labeled with the tracer. At 60 seconds the labels were distributed uniformly in external and internal surface vesicles. Passage across the cell was concluded to occur entirely via the vesicles, either singly or through openings formed by chains of vesicles. In other studies utilizing colloidal carbon (radius: <150 A), penetration of the endothelium did not occur (23). After injury, 8 however, it appeared to penetrate through widened intercellular clefts between adjacent endothelial cells in post-capillary venules. The structural identity of a small pore system (40—45 A) is a matter of dispute. Based on the experiments of Karnovsky (62), small pores were represented in skeletal muscle capillary endothelium. These findings have not been confirmed, although Simionescu et al. (105) have presented evidence indicating that the small pores are patent transendothelial channels built up by the vesicles in muscle capillaries. In post-capillary venules, on the other hand, junctions between adjacent endothelial cells did occur (102). The existence of the small membrane-lined vesicles containing water-soluble tracers inside the endothelial cells supports the concept of vesicular transport proposed by Bruns and Palade (17). According to this theory, the vesicles first form invaginations on the luminal side of the cell membrane, develop a neck, and then are pinched off, to presumably drift across the cytosol and empty their contents at the opposite cell membrane. A basic drawback of the vesicular theory of transport is its inability to account for the observed rate of fluid exchange in tissues and for molecular sieving. Although it may parti- cipate in normal fluid and macromolecular transport in altered fluid- flux states, such as inflammation where loss of fluid and protein from the vascular compartment is greatly augmented, this mechanism as a normal regulator is improbable. The endothelial cells of fenestrated capillaries display numerous intracellular openings (200-250 A) which may be either opened or closed by a very delicate plasma membrane diaphragm (34). These occur pri- marily in the renal glomeruli, glands and intestinal mucosa. Tracer electron micrograph studies of capillaries in rat and mouse intestinal 9 mucosa showed concentrations of dextrans and glycogens outside a large fraction of the fenestrae (103). These tracers also were present in cytoplasmic vesicles: however, transit was considerably slower than through the fenestrae. The fenestrae, although usually closed by a diaphragm, are not only permeable to large proteins such as ferritin but also to large amounts of bulk fluid including solutes (127). This high protein permeability may be important in post-capillary venules since this would directly influence the reabsorptive capacity of the blood vessels. Discontinuous capillaries possess endothelial cells with obvious intercellular gaps. This type of capillary is adapted for the trans- mural exchange of not only macromolecules but even blood cells. There- fore, a blood transfusion can be administered via a cannula inserted into the bone marrow. Liver, spleen and bone marrow are characterized by this type of capillary. The exchange of materials between blood and interstitial fluid involves diffusive and convective processes that occur at the level of the microvascular membrane. Reciprocity between these two fluid compartments separated by the membrane takes place primarily by the diffusive process, which is determined by concentration gradients, and the physical properties of the microvascular membrane. The convective process, which results in the continuous flow of fluid through the microvascular membrane, is a consequence of the interaction between hydrostatic and osmotic forces. Fluid movement across the enormous surface area presented by the microcirculation is essential for the maintenance of osmotic equilibrium in the tissue environment, the facilitation of the transport of large molecules through the intersti- tium, and the regulation of blood volume. 10 In the tissue milieu, the movement of fluid is maintained by the convective transfer of blood through the capillary network and the radial transfer of blood ultra-filtrate through the tissue by a diffusive process. These two flows, namely the convective (intra- vascular) and the diffusive (extravascular), have a precise relation— ship to each other which is summarized by the concepts involved in the Starling hypothesis of fluid exchange (111,67). The Starling Equation for the radial movement of fluid between the blood and extravascular compartment is expressed as JV = k |(PC + “1’ - (Pi + np)|. where: Jv = rate of fluid movement, k = capillary filtration coefficient, PC = capillary hydrostatic pressure, "1 = tissue colloid osmotic pressure, Pi = tissue hydrostatic pressure, and up = plasma colloid osmotic pressure. If the rate of fluid movement (Jv) is positive, net fluid filtration occurs; if this value is negative, net fluid reabsorption ensues. According to this theory, the balance of fluid between the two compartments is a result of the exchange of fluid arising from the interaction of hydrostatic and oncotic pressures, mediated by the permeability characteristics of the blood-tissue interface. Classi- cally, net filtration occurs at the arterial end of the capillary and net absorption at the venular end. This occurs because the transmural hydrostatic pressure gradient exceeds the transmural oncotic pressure ll gradient at the arteriolar end, whereas the reverse is the premise at the venular end of the capillary. The primary measurements necessary to examine the validity of the Starling hypothesis were first carried out by Landis in 1927 using micromanipulation techniques (66,67). These studies were per— formed on single capillaries in the frog mesentery. The results from these experiments showed a direct linear relation between capillary hydrostatic pressure and fluid movement across the microvascular membrane (Figure I). When the capillary hydrostatic pressure exceeded 12 cm H 0, fluid passed from the vasculature to the interstitium. At 2 pressures below 10 cm H 0, fluid was withdrawn from the interstitium 2 and into the capillary. At pressures between 10 and 12 cm H 0, little 2 or no movement of fluid occurred. This was the result of the capillary hydrostatic pressure being balanced by the osmotic pressure of the plasma proteins. In addition to supporting the basic tenets of Starling's hypothesis, these data also provided a measure of the permeability of the ndcrovascular ummbrane to isotonic fluid. The permeability of the microvascular wall to isotonic fluid or the hydrodynamic conduc- tivity of the capillary wall is called the capillary filtration coef- ficient (k or CFC). This proportionality constant was computed by Landis from the slope of a straight line drawn through the data points (Figure I). Experimentally, the capillary filtration coefficient is measured in ml/min/mm Hg/100 gms tissue, by recording the increase in weight of an organ or limb that is caused by a known increase in capillary hydro- static pressure (34). The coefficient varies from tissue to tissue with an average whole body value for man of .0061 ml/min/mm Hg/100 gms Cubic micra of fluid per sec per square micron .08 .06 .04 .02 .02 .04 12 L filtration . slope = capillary filtra— ‘E:;__,S—’——-_— tion coefficient (CFC) . absorption 0 5 10 15 20 25 Capillary Hydrostatic Pressure (cm H20) Figure I. Linear relation between capillary hydrostatic pressure and fluid movement across the microvascular membrane (67). 13 tissue (67). Under experimental circumstances the capillary filtra- tion coefficient is used as an index of capillary permeability, which includes surface area available for diffusion. Capillary hydrostatic pressure (Pc) is directly dependent upon capillary blood volume and compliance. Clough et al. (22) have indi- cated that capillaries are quite rigid. They report a change of only 0.1 pm in radius in capillaries of the cat mesentery during systole. This rigidity results from the environmentcircumjacenttm>the capil— laries. The basement membrane and gel matrix surrounding these microvessels give the capillaries little, if any, compliance (36). Since compliance is relatively constant in capillaries, changes in capillary blood volume are the primary factor in determining Pc. Capillary blood volume is influenced by systemic arterial pressure, venous pressure and the pre- and post-capillary resistances. The interrelationship of these factors is expressed in the following equa- tion of Pappenheimer and Soto-Rivera (67): —- Rv = - -——————-+ Pv Pc (Pa Pv) Ra + RV , where: Pa = systemic arterial pressure, Pv = venous pressure, Ra = arterial resistance (pre-capillary), and Rv = venous resistance (post-capillary). An increase in Pa or Pv will increase Pc. Increasing Rv will raise Pc, whereas increasing Ra will lower Pc. Vessel resistances are inversely related to vessel caliber. This caliber is determined mainly by active changes in vascular smooth muscle activity and passively by changes in effective transmural pressure. Effective transmural pressure is the l4 pressure in the interstitial fluid environment of the capillary sub- tracted from the intraluminal pressure in the capillary. Changes in blood viscosity also affect resistance to blood flow. An increase in the viscosity of the blood will increase the resistance to flow. Blood viscosity is proportional to the hematocrit and the dissolved materials in the plasma. The hydrostatic pressure of the interstitial spaces is deter- mined by tissue compliance and interstitial fluid volume. Classically, it is accepted that this pressure is positive and will therefore oppose fluid filtration out of the capillaries. While this is the case for many tissues (kidney, brain, bone marrow, abdominal viscera), recent studies have suggested that it is sub-atmospheric in subcutaneous tissue (44). Guyton, using implanted perforated spheres in a variety of tissues (44), has concluded that this pressure is sub-atmospheric (-7 mm Hg). This conclusion, however, is under much criticism and further investigation is needed to resolve the question. Plasma colloid osmotic pressure (oncotic pressure) is the pressure due to the concentration of dissolved proteins in the blood. Under normal conditions, the total osmotic pressure of plasma is about 6,000 mm Hg, with the oncotic pressure contributing about 25 mm Hg (12). This oncotic pressure is responsible for vascular fluid retention. It is achieved because the plasma proteins are largely confined to the intravascular space and therefore create an active tonicity within the vasculature. The electrolytes which constitute the bulk of the remaining total osmotic pressure pass freely through the capillary membrane, creating no tonicity between the two body fluid compartments. Albumin, globulin and fibrinogen comprise the major plasma proteins. Albumin, which is found in the greatest abundance, has an 15 average molecular weight of 69,000 and a concentration of 4.6 gms %. Globulin has an average molecular weight of about 140,000 with a concentration of 2.5 gms %. Fibrinogen is the largest of the plasma proteins, having a molecular weight of about 400,000, but is found in a concentration of only 0.3 gms %. 0f the total oncotic pressure, 19 mm Hg are attributable purely to the proteins, and the remaining 6 mm Hg are due to the cations which bind to the proteins by electronega- tive forces. This phenomenon is known as the "Donnan Effect." Since albumin is in the greatest concentration of the three proteins, it constitutes the largest fraction (70%) of the total oncotic pressure (12). Interstitial fluid colloid osmotic pressure is contingent on the protein concentration of the interstitial fluid. The concentration of the interstitial protein is not uniform: it varies from 0.4 gms % to 3.3 gms %, depending on the tissue (67). In skin and skeletal muscle, the average interstitial protein concentration is about 2.0 gms %, which yields an oncotic pressure of about 5 mm Hg. More recent findings suggest that the total protein concentration of interstitial fluid is about 3.0 gms % and the colloid osmotic pressure about 10 mm Hg (125). In the liver, where capillary protein permeability is high, large amounts of plasma proteins cross the microvascular membrane, producing an interstitial fluid oncotic pressure of about 16 mm Hg or more. This corresponds to a minimal protein concentration of about 3.3 gms %. Discussion of absolute values for interstitial fluid colloid osmotic pressure is under considerable debate. Measurements using implantable devices such as perforated capsules that theoretically equilibrate with interstitial fluid may be inaccurate because of the l6 possibility of contamination by plasma, or that the fluid sampled may not contain all the osmotically active substances. The most common method for estimating interstitial oncotic pressure is lymph fluid analysis. This method makes the assumption that lymph is a true reflection of the interstitial fluid contents. Critics of this view argue that changes could occur in lymph composition as it flows from terminal lymphatics upward to larger vessels due to gradients of protein concentration within the interstitial spaces. Renkin and Garlick (96) have shown that dextran molecules of known molecular weight and size are in equal concentration between lymph and intersti- tial fluid when injected into a tissue. This observation allowed them to conclude that there is no significant protein concentration gradient beyond the capillaries. Garlick and Renkin (37) have performed studies showing that exchange occurs only at lymph nodes and not in the lymphatic trunks. If lymph is sampled before it reaches a node, it should be a true reflection of what is at the terminal lymphatic vessel. The terminal vessels of the lymphatics consist of a widely distributed closed-end network of highly permeable lymph capillaries that are similar to blood capillaries. In general, there are no tight junctions between endothelial cells, but they possess fine filaments that secure them to the surrounding connective tissue. With muscular contraction, these filaments may distort the lymphatic vessel such that spaces will open between adjacent endothelial cells, allowing the entrance of proteins, macromolecules and cells present in the inter- stitial fluid. The terminal lymphatics drain into larger vessels that ultimately enter the right and left subclavian veins. l7 Valves exist throughout all lymphatic channels. When a lymph vessel is compressed by pressure, lymph in the channel is squeezed in both directions. Since the valves are only unidirectional to flow, the fluid will be transported from the terminal lymphatics in the central direction past the valve. Factors that can compress the lymphatics and evoke the movement of lymph are: muscle contraction, arterial pulsations, passive movements of body parts and compression of body tissues from the outside. During exercise, lymph flow can increase substantially, about 15 times normal. When either the volume of interstitial fluid exceeds the drainage capacity of the lymphatics or the lymphatic vessels become blocked, interstitial fluid accumulates and gives rise to edema. The accumulation of fluid occurs primarily in the more compliant tissues (i.e., subcutaneous tissue). The most fundamental mechanism by which substances are relocated between the vascular and interstitial compartments is by the dif- fusive process. Diffusion is a distributive process which is dependent on the thermal motion of molecules: therefore, the greater the con- centration difference along an axis, the greater the rate of diffusion. This mode of exchange can be described by the Fick Law of Diffusion, which states that the quantity of substance moved per unit time is equal to the free diffusion coefficient Of the molecule, the concen- tration gradient, and the area of the capillary membrane. These factors are related in the following equation: ds/dt = D - A - dc/dt 18 where: ds/dt = amount of substance moved per unit time, D = free diffusion coefficient for a molecule (this value is inversely proportional to the square root of the molecular weight), A = area of the capillary membrane, dc/dt = concentration gradient. The site of diffusion of a molecule depends on whether the sub- stance is water-soluble or lipid-soluble. Water-soluble substances pass through small pores in the endothelial cell. For small molecules such as water, ions, and urea, diffusion is free and rapid. However, for lipid-insoluble molcules of increasing size, diffusion becomes progressively more restricted, such that molecules above a molecular weight of 60,000 are almost completely impermeable to the capillary endothelium. Lipid-soluble molecules such as CO , 0 and anesthetic 2 2 gases pass freely through the intact cell membrane. Small, lipid- soluble molecules pass with great ease and rapidity between capillary and interstitium. Micropinocytosis (cytopemphysis) is a slow transport mechanism that is probably responsible for the movement of large lipid-insoluble molecules between plasma and interstitial fluid. The process has been described previously in this survey of the literature. It appears that there should be little doubt that such a mechanism is involved in the normal or physiologic transport of certain large macromolecules across the microvascular membrane, based on numerous data and observations of this phenomenon by several well respected medical scientists (19,62, 95,104,105,127). However, the importance of this mechanism in patho- logic conditions, such as inflammation, is questionable. 19 This dissertation focuses on the second of two vascular changes associated with acute inflammation: 1) hemodynamic adjustment and 2) alterations in vascular permeability. The increased vascular permeability with the escape of plasma proteins and fluid is known as exudation. It accounts for an increase in the volume of interstitial fluid (edema) and tissue swelling at the site of injury. The increased permeability first affects venules but rapidly extends to capillaries (98). The mechanisms leading to this increase in permeability in inflammation are still incompletely understood. Several classes of mediators affecting permeability of blood vessels have been described, and most appear to have their effect on post-capillary venules. The mechanism by which permeability agents accomplish their biological effects is not known. The suggestion that permeability changes are due to increases in hydrostatic pressure in the capillaries, caused by contraction of venules, has not been sub- stantiated. It has recently been suggested that permeability-inducing agents cause contraction of periendothelial cells (pericytes), which surround capillaries. Since these cells seem to be structurally attached to the endothelial cells, their contraction could result in pulling apart endothelial cells at their junctional zones, allowing the escape of fluids and solutes. Electron microscopic evidence of contractile-like agents in pericytes has been reported (107). However, this theory of changes in permeability awaits further study. Classical permeability inducing agents are histamine, bradykinin, SRS-A (slow reacting substance of anaphylaxis) and several basic peptides released from neutrophils (98,107). The prostaglandins are biologically active compounds whose role in the inflammatory process and the induction of permeability is currently under a great deal of speculative investigation. 20 The prostaglandins are a group of endogenous, unsaturated fatty acids to which are attributed a diversified range of physiological activity. The first descriptive report of an effect ascribed to the prostaglandins was in 1930, when two American gynecologists, Drs. Raphael Kurzrok and Charles Lieb, observed that uterine muScle strips would relax or contract when exposed to human semen (65). In an attempt to procure a treatment for human sterility, these investigators recognized that uteri obtained from women who had completed successful pregnancies responded with relaxation when 1 cc of semen was applied. However, the uteri from women who had histories of complete or extended periods of sterility contracted with the addition of human seminal fluid. The significance of these findings not only indicated that the active constituent in the human semen was dependent upon the endocrine status of the uterus, but that this new substance possessed differen- tial biological activity. Although Kurzrok and Lieb made these preliminary observations, the credit for the discovery and isolation of prostaglandins belongs to the world-renowned Swedish medical scientist, Ulf S. von Euler. In 1935, it was Euler who clearly established that the active principle in human semen and extracts from sheep vesicular glands was a lipid- soluble acid which he called "prostaglandin" (28,29,30). The following excerpt is a brief account from the records of Euler regarding the identification of prostaglandins: Starting with the findings that the vesicular gland of the sheep was a rich source of the active compound, we found it only natural to seek more information about its chemical nature. The observation that it could be extracted with lipid solvents from an acid solution but was freely water soluble in an alkaline medium not only suggested that it was of acidic character, but also provide a simple means of purification. It thus appeared to have the properties 21 of a fatty acid, which was somewhat surprising since very little was known about biologically active substances of this kind, and certainly nothing of their activity in the body. Through the good offices of Hugo Theorell in the chemistry department of the Karolinska Institute, who had just developed an apparatus for preparative electrophoresis allowing separation of various fractions, an active extract was separated. The results showed as expected, that the active principle migrated toward the anode. Since it appeared that the active compound was a new and previously unknown chemical, it was thought that it should receive a name of its own. Owing to its occur- rence in extracts of the prostate and vesicular gland, it was called prostaglandin. (31) For the next two and one-half decades, the investigation into this area was limited because of inadequate experimental techniques. The impetus for further biological research occurred in 1960, when Bergster and Sj6vall isolated and chemically characterized the first pure samples of prostaglandin (10,11). They initially isolated two prostaglandin compounds which behaved differently, based on their partition between ether and an aqueous phosphate buffer. The compound soluble in ether was called prostaglandin E or PGE (11), and the compound soluble in the phosphate buffer was called prostaglandin F or POP (10). Preceding the identification of PGE and PCP, Bergstrdm and his associates were able to differentiate and isolate 13 structurally different compounds which were all derivatives of the parent-skeleton compound, prostanoic acid (9) (see Figure II). All prostaglandins contain 20 carbon atoms, a cyclopentane ring, two aliphatic side chains and a terminal hydroxy group. With the recent discoveries of many otherprostaglandins and prostaglandin-like compounds, E. J. Corey has suggested that the term "eicosanoids" be used to describe all of the 20-carbon derivatives, whereas "prostanoids" refers only to those with a prostanoic acid skeleton (81). 22 x/\/‘ <17 A All prostanoids contain 20 carbon atoms, a cyclopentane ring, two ali- phatic side chains and a C—terminal hydroxy group. Figure II. Structure of prostanoic acid. 23 Physiologically, prostaglandins are synthesized from 20-carbon polyunsaturated fatty acids containing three, four or five double bonds. These fatty acids are present mainly in the phospholipid portion of the mammalian cell membrane, although serum triglycerides and cholesterol are also sources (79,53). Furthermore, prostaglandins have been identified in birds, reptiles, amphibians, fish and various members of the plant kingdom (53). The principal fatty acid precursor of prostaglandins in man is eicosatetraenoic acid, which is commonly referred to as arachidonic acid (see Figure III). This compound contains four double bonds and generates those eicosanoids with two double bonds in their aliphatic side chains (i.e., PGE , PGF a' PGD 2 2 acid, which possesses three double bonds, yields eicosanoids that contain one double bond (i.e. , PGE , PGF a' PGD 1 1 , TXAl). Eicosan01ds 1 with one double bond are synthesized only in trace amounts in most mammalian tissues. Although it was originally assumed that these substances were of paramount importance in physiological control systems, recent advances have demonstrated that their significance is less than the eicosanoids derived from arachidonic acid (79). Eicosa- pentaenoic acid” the harbinger substance for the-eicosanoids with three double bonds, incorporates five double bonds in its structure. The physiologic relevance of these eicosanoids is ambiguous and con- siderable investigation is necessary to eluéidate their significance (79). The situation regarding the biosynthesis of the eicosanoids as of 1979 is summarized in Figure III. Since the arachidonic acid cascade is of greater importance in mammalian tissues than the other two biosynthetic pathways, this 24 6.5 53.258 8.2 .fiuomov Boo 3652:3388 uco Atom“: Boo £52596 .Amuomov Eco 85.057.510.356 Eat 385.023... to smog—£35 5.96235 mad man. $cox0nES£ » 4 ma. 7c. § «5303233 on“. mm mm gum. «w No 4 B... .m 5 _whwa+|_ _mwwa_ 4 336.385 mnou » » [luxoaun .Illxoa » » Nooa .ooa 4 s omocmozxo All. -228 v “om m ”ON a muse...— 25 treatise will concentrate on the significance of these compounds. Reference to other pathways will be made when necessary. Arachidonic acid is obtained directly from the diet or by ana- bolic desaturation and chain elongation from dietary linolenic aCid- It is transported in blood, largely bound to albumin, and is incor- porated as a structural component of phospholipids in cell membranes and other subcellular structures of all tissues in the body (81). It is commonly accepted that the majority of free arachidonic acid, as well as the other precursor substances (i.e., di-homo-y- linolenic acid and eicosapentaenoic acid) are liberated from the membrane phospholipids by enzymatic cleavage. The enzyme or group of enzymes responsible are called acylhydrolases, which include mainly the phospholipases. Evidence remains nebulous as to the specificity of these enzymes for a particular precursor (68). The specific acylhydrolases associated with prostaglandin synthesis are activated by a number of different physical and chemical stimuli (Table II). The accepted reaction sequence for prostaglandin synthesis is: stimulus -——+ acylhydrolase activation --+-release of fatty acid precursor from various lipid sources -—-+-eicosanoid biosynthesis and release. The acylhydrolase(s) is the rate-limiting enzyme in the prostaglandin synthetic pathway (121). This assumption is based on the fact that the activity of the other eicosanoid synthesizing enzymes in intact tissues are limited by the amount of available arachidonic acid. In addition, arachidonic acid can be provided from endogenous fatty acids by the administration of exogenous acylhydrolase (121). 26 Table II. Stimulants of acylhydrolase (121) Stimulant Tissue Amines: catecholamines spleen, stomach, brain, fat cells serotonin stomach, brain, fat cells acetylcholine skin, intestine, adrenals Hormones: TSH thyroid gland ACTH adrenal glands LH uterus glucagon liver Peptides: bradykinin lung, skin angiotensin spleen Miscellaneous: collagen platelets thrombin platelets dust particles nerve stimulation anaphylaxis inflammation distention distilled H20 lung, spleen brain, kidney, lung, fat cells lung, spleen skin lung, uterus, stomach intestine 27 Based on the substantial number of diverse acylhydrolase stimulants (Table II), a perplexing question arises as to the mechanism or mechanisms by which acylhydrolases can be activated to augment arachidonic acid release. Currently, there are three proposed mechanisms which may account for the activation process (121). The first mechanism was forwarded from the observation that in the pancreas, phospholipase A is stored as a zymogen granule and is activated by 2 tryptic cleavage. Consequently, a stimulus would convert trypsinogen to trypsin and this would cleave phospholipase A from its zymogen 2 storage granule. This type of action is irreversible and, in most cases, not suitable for physiological control of prostanoid biosyn- thesis, which must depend on reversible activation mechanisms. However, in the platelet only, this could provide a mechanism of action since platelets may undergo irreversible damage subsequent to their activa- tion (49). A second proposed mechanism involves a reversible activation achieved by changes in the concentration of free intracellular Ca++. Some acylhydrolases of plasma membranes are Ca++ dependent. The optimal Ca++ concentration for maximal acylhydrolase activity of liver plasma membranes is 5 uM. This is considerably higher than the normal intracellular concentration of free Ca++. Consequently, an induced increase of Ca++ would activate an acylhydrolase and enhance the cleav- ing activity. Supporting this theory is the fact that a variety of prostaglandin stimulants facilitate Ca++ release into the cell. These stimuli include thrombin, acetylcholine and the catecholamines. The third mechanism has been proposed by Verger and his co-workers (119). They suggest that the activity of acylhydrolases, particularly phos- pholipases, depends to a large extent on the structural arrangement of the phospholipid molecules in the cell membrane. When the normal lipid 28 bi-layer structure is altered as in perturbation of the cell, cleavage of arachidonic acid from phospholipids will ensue. Acylhydrolases are found in various organelles in the cell, including mitochondria, lysosomes, microsomes and the plasma membrane (38). Since acylhydrolases have several sub-cellular locations, it is assumed that its stimulation and activity are modulated by more than one mechanism depending on the state of the particular tissue or cells. It has also been noted that mepacrine is a specific inhibitor of acylhydrolase, and thereby prevents the release of free arachidonic acid (49). Once released from the fatty acid pools, arachidonic acid is metabolized by two types of enzyme systems. One is a series of lipoxy- genases that peroxidize arachidonic acid to form unstable hydroperoxides. The hydroperoxides are then rapidly reduced enzymatically by gluta- thione peroxidase to inactive hydroxyacids (Figure IV) (81). Different lipoxygenases have been described in different tissues depending on what carbon atom the peroxidation occurs (68). For example, the lipoxygenases in the lung peroxidize arachidonic acid at carbon atoms 11 and 12, whereas white cell lipoxygenase peroxidizes at carbon 5. The other enzyme system has a cyclo-oxygenase that forms the prostaglandin endoperoxide PGG (Figure IV). This is converted to 2 another endoperoxide, PGH , by the action of a peroxidase enzyme (85). 2 The half-lives of both endoperoxides are about five minutes at 37°C (78). The classical prostaglandins, PGE PGF and PGD , are formed 2' 2a 2 directly from PGE either by the action of an isomerase or by a non- 2 enzymatic reduction which will yield PGF PGD is formed from PGH 2a' 2 2 in the presence of ll-keto—isomerase, which is a cytoplasmic (soluble) enzyme (68). PGD2 is a major prostanoid identified in brain (1) and 29 FIGURE 1! ARACHIDONIC ACID I'GYDROPETE PEROXIDE HYDROXYACID HETE CYCLIC ENDOPEROXIDES COO 0490' r\,\/coou on O W HO OH moon :&Wcoou OO’MTAcooa moo“ Cl? HYDROXYACID mosmscimom F2. mosmamom oz mosnsuuom :2 go UALONDIALDEHYDE M D A MERBOLIC PII'HWAYS OF ARACHIOONIC ACID (81) 30 platelets (2). Although PGD has been regarded as biologically 2 inactive, it does inhibit platelet aggregation (68). Endoperoxide- e-isomerase is a membrane bound enzyme which converts PGH2 to PGE2 (68). Although PGF can be formed directly from PGH by a non- 2a 2 enzymatic reduction, speculation exists for another route of formation. PGE2-9-keto-reductase, which reversibly converts PGE2 to PGan, is a cytoplasmic enzyme that has been isolated from vascular tissue, kidney, lung, brain, heart, spleen, liver and blood cells (112,69). This enzyme requires NADP as a cofactor and may function as the major synthetic route for the formation of PGF (69). Consequently, PGF 20 2a generation may not be directly from PGH , but rather through PGE 2 2 acting as an intermediate. It is interesting to speculate on the importance of this enzyme in regulating the diverse physiological processes involving these two classes of prostaglandins, yet no defini- tive evidence is available to elucidate this issue. The prostaglandin endoperoxides are also transformed enzymati- cally into two other unstable products, prostacyclin (PGIZ) and thromboxane A2 (TXAZ) (47,78,82). The half-lives of these compounds are three minutes and 30 seconds, respectively. The cascade of arachidonic acid metabolism that leads to the formation of the cyclic endoperoxide PGG is catalyzed by the enzyme 2 complex known as cyclo-oxygenase. The enzyme catalyzes the reaction which incorporates molecular oxygen into the polyunsaturated fatty acid, arachidonate. The cyclo-oxygenase complex is composed of several sub-units and has a molecular weight of about 70,000 daltons (68). The activity of the enzyme is completely dependent upon adequate heme availability. Since competition for heme by various other proteins 31 exists within the cell, this could provide a means of regulating pros- taglandin biosynthesis (68). In 1971, Vane and others discovered that non-steroidal, anti- inflammatory agents, notably aspirin and indomethacin, inhibited prostaglandin biosynthesis (33,117). Since this initial discovery, much evidence has emerged to indicate that the antipyretic, analgesic and anti-inflammatory actions of these drugs are mainly mediated via inhibition of prostaglandin biosynthesis at the cyclo-oxygenase step (80). Because aspirin and indomethacin prevent the biosynthesis of all known eicosanoids, they have been used extensively as experimental tools to clarify whether prostaglandins are involved in various physiological or pathological events. In 1975, Hamberg and his associates discovered a new group of biologically active compounds, thromboxanes (47). Since these sub- stances occurred in the thrombocytes (platelets) and contained an oxane ring in their structure, the word thromboxane was proposed. TXA 2 is derived from PGH2 (Figure IV) and stimulates platelet aggregation. It has also been shown that TXA2 is identical with rabbit aorta con- tracting substance (RCS) originally described by Piper and Vane (91). TXA2 generation is the major product of prostaglandin metabolism in platelets; however, other tissues possess thromboxane synthetase and can therefore produce TXA This list includes iris, umbilical artery, 2. placenta, leukocytes and lung fibroblasts (81). TXA2 produces vaso- constriction and contraction of the trachea (85). TXA2 is potentially capable of causing myocardial infarction. Shimamoto and his co-workers (100) demonstrated that in rabbits intra— coronary artery injection of TXA could cause massive platelet aggre- 2 gation capable of producing myocardial necrosis and typical 32 electrocardiogram changes associated with infarction. Pre-treatment with phthalazinol, which inhibits TXA , prevented the myocardial 2 (101). Other TXA2 inhibitors (specifi- cally thromboxane synthetase inhibitors) include N0164, L-8027, infarction induced by TXA2 dipyridamole and imidazole (68). The platelet pro-aggregatory phenomenon is by far the most important effect of TXA , which reduces platelet adenylate cyclase 2 activity, and leads to an increase in platelet cyclic AMP (79). Therefore, Vane (79) has proposed that cyclic AMP exerts a regula- tory function on platelets. It is known that increasing levels of cyclic AMP will render platelets less aggregable. Consequently, when TXA2 decreases cyclic AMP, platelets aggregate. The counterpart eicosanoid of TXA is the platelet anti- 2 aggregatory compound, PGI PGI was discovered by the Wellcome group 2' 2 from England in 1976 (78,82). In that laboratory, Dr. Moncada and his associates were studying the vasoconstriction induced by TXA2 in porcine aortas. When arachidonic acid, PGEZ, PGFZQ or P602 was added to porcine microsome preparations, no prostaglandin-like activity was detectable by smooth muscle bioassay. However, when PGG2 or PGH2 was incubated in the same system, a substance called PGX was found which relaxed blood vessels and inhibited platelet aggregation (78). The structure of PGX was quickly solved by the joint efforts of the Upjohn group in the United States and the Wellcome group in England, and the name prostacyclin (PGIZ) was adopted (55). Prostacyclin inhibits platelet aggregation by stimulation of adenylate cyclase (79). This effect leads to an increase in platelet cyclic AMP, thereby rendering the platelets less aggregable. Since TXA2 reduces adenylate cyclase activity, Moncada and Vane (79) have 33 proposed a theory that cyclic AMP exerts a regulatory function on platelets, which depends on a balance between TXA2 and PGIZ. Prostacyclin is the major known product of arachidonic acid metabolism in all vascular tissues studied (79). A recent discovery of interest is that prostacyclin can be released into the pulmonary vein by the lungs and can therefore be present in arterial blood. However, unlike the traditional prostaglandins (E2, F and D2), 2a prostacyclin is not degraded to an inactive metabolite by the lung. This results from the absence of an active uptake process for prosta- cyclin by pulmonary capillary endothelial cells (122). Consequently, it has been proposed that prostacyclin is a circulating hormone which continually activates platelet adenylate cyclase causing platelets to be less aggregable. This hypothesis suggests a homeostatic system in which platelet aggregation is controlled in vivo by a prostacyclin mechanism that involves both circulating and locally produced prostacyclin (79). Prostacyclin, thromboxane A and the classical prostaglandins 2 are considered to be major metabolites of the cyclic endoperoxide, PGH2. However, under some conditions this endoperoxide may break down into a l7-carbon hydroxy acid (HHT) and malondialdehyde (MDA) (Figure IV) (81). The reaction occurs non-enzymatically by spontaneous degra— dation of PGH . It is thought that the biosynthetic pathways of TXA 2 2 and HHT formation are somehow linked since thromboxane synthetase also catalyzes HHT formation (81). The rapid disappearance of physiologically active prostaglandins is catalyzed by intracellular enzymes (68). Tritiumrlabeled PGE2 and PGan, administered intravenously to human subjects, are rapidly con— verted to the corresponding 15-keto-13,l4 dihydro metabolites by the 34 enzymes prostaglandin dehydrogenase and A13 reductase (99). Initially, prostaglandin dehydrogenase catalyzes the oxidation of the lS-hydroxyl group to form a conjugated ketone at this position. This is usually followed by a reduction of the l3-l4-trans double bond, catalyzed by A13 reductase. The resulting dihydro-keto derivative is the major prostaglandin metabolite in the peripheral circulation. Subsequent oxidations will follow yielding the mono- and dicarboxylic acids, which are predominant metabolites in the urine (68). Spontaneous hydrolytic cleavages of TXA and PGI 2 2, the major degradative reaction sequences for these eicosanoids, yield the major metabolic by-products TXB2 and 6-keto—PGFla, respectively. Although many regard prostaglandins as freely diffusible agents, current available evidence shows that some cell membranes represent a barrier to the paséive diffusion of prostaglandins. Experiments have shown that cellular uptake in many tissues is an active or carrier- mediated process (16). Interference with the mediated transport by drugs, such as non-steroidal anti-inflammatory agents (indomethacin, naproxen, phenylbutazone), will increase the circulating levels of prostaglandins (15). Such active or carrier-mediated processes have been shown to exist in the brain (15), eye (15), lung (14), vagina (15), skin (70), kidney (13), and adipose tissue (68). Thus, active transport or carrier-mediated transport of eicosanoids across the cell membrane can be prerequisite to their degradation. The first experimental evidence for a prostaglandin receptor was reported in 1972 by Kuehl and Humes (64). These investigators demon- strated a receptor for PGEl on the cell membranes of rat adipocytes. Using a binding assay technique, specific binding of tritiated PGE 1 was revealed in a particulate fraction of isolated rat adipocytes. 35 They also observed that PGE2 would bind to the receptor with approxi- mately the same affinity as PGE , while the binding capacity of A- and l F-type prostaglandins to this receptor was insubstantial. Reports of a high-affinity PGE receptor have subsequently appeared for stomach, thyroid gland, uterus, liver, corpus luteum, kidney (99), skin (71), and the vasculature (56). These receptors which mediate the various physiologic responses initiated by E-type prostaglandins are located on the cellular plasma membranes (99). High affinity receptors for PGFZa have also been demonstrated in cell membranes of corpus luteum, uterus and oviduct (99). Although it appears likely that receptors for the various eicosanoids exist in virtually all tissues, these receptor relationships for the different cells must be learned. Also, it is not yet understood the extent to which prostaglandins affect cells other than the ones catalyzing their biosynthesis (68). The importance of prostaglandins in the inflammatory response has been implicated by three lines of research: 1) prostaglandins of the E series have been shown to have potent pro-inflammatory effects: 2) they are released in several different types of inflammation; 3) non-steroidal anti-inflammatory drugs (aspirin and indomethacin) inhibit the biosynthesis of prostaglandins. Increased concentrations of prostaglandins, especially of the E series, have been recovered from inflammatory exudates in man (40,92), dog (3,57), and rat (128,129). Greaves et al. (40), using a skin perfusion technique in viva, studied the prostaglandin activity of inflamed skin in patients with allergic contact eczema. Perfusates from 35 patients contained a mixture of E and F prostaglandins. Concentrations of PGE recovered from inflamed tissues averaged 1.0 1 ng PGEl/ml of perfusate. Control levels averaged 0.2 ng PGEl/ml of 36 perfusate. The approximate distribution of total activity of PGE1 was: E 12%, E l 12%, F 24%, and F 49%. This is suggestive that 2 la Za prostaglandins participate in the pathogenesis of inflammation in man. In a similar study using human subjects, Plummer et a1. (92) showed that an exudate from inflamed skin displayed augmented prostaglandin activity compared with exudates from contralateral non-inflamed skin. PGE2 and PGE1 activity increased 70% and PGFZa increased 73% as com- pared to control. Inflammation was produced by application of Trafuril cream to normal abdominal skin. Normal human plasma levels of prostaglandins as measured by radioimmunoassay are: PGE1 2.6 pg/ml, PGE 4.5 pg/ml and PGF 2 12 pg/ml (26). 2d Although prostaglandins have been extracted from various inflam- matory exudates and inflammatory diseases involving acute or chronic inflammation of the joints, skin and eyes can be controlled though not cured by cyclo-oxygenase inhibitors, there is no direct evidence for prostaglandins as putative mediators of inflammation. The applica- tion of a prostaglandin directly to a tissue causing edema remains a controversial conclusion. Considerable debate exists among investigators concerning the relative capacity to which PGE and/or PGE are involved 2 in the transcapillary movement of fluid and plasma proteins. Possible 1 reasons for this uncertainty include experimental techniques, species variability, PGE dose and routes of administration, and magnification of conclusions deduced from the data. Greenberg and Sparks (41) studied the effects of PGE1 on the vascular resistance, capacitance and capillary filtration coefficient in the isolated canine hind limb by measuring total outflow and pressure of the popliteal vein and changes in plethysmographic recordings in an enclosed hind limb. They found that the intra-arterial administration 37 of 0.001 to 1.0 pg/min of PGE caused a decrease in resistance, an 1 increase in the capillary filtration coefficient and an increase in the vascular capacity of the isolated hind limb. Constant inflow studies and experiments on isolated vascular smooth muscle strips indicated that PGE1 caused relaxation, whereas PGFZa caused contraction. They suggested that the increased capillary filtration coefficient during PGE1 infusion is due to a decrease in pre-capillary sphincter tone, since 1) there was no net filtration associated with the increased capillary filtration coefficient and 2) the relative changes in resistance and the capillary filtration coefficient were similar in magnitude to those observed during exercise when there is no increase in capillary permeability. More recent work in the hind limb of the dog by Joyner (58) presents evidence that the subcutaneous injection of PGE1 (0.01 to 100 ug/15 min) causes net fluid filtration but does not increase permeability, as ascertained from the protein concentra- tion of collected lymph. However, Joyner proposed that PGEl enhanced transcapillary fluid movement, as demonstrated by an enhanced lymph flow, primarily by its vasodilatory properties. The vasodilation resulted in a decrease in the pre- to post-capillary resiStance ratio resulting in net fluid filtration. In contrast, Daugherty (25) studied the effects of PGE1 on skin and skeletal muscle vascular beds in the canine forelimb and observed little effect on filtration. In these studies, PGE1 (2.0 to 10 ug/min) infused into the brachial artery produced large increases in both skin and muscle blood flows and decreases in skin and muscle vascular resistances. The small effect on filtration, as measured by forelimb weight, was postulated to be due to a proportional dilation of the pre- and post-capillary vascular segments. The prostaglandin infusion period was only three to five 38 minutes, which could have accounted for no observable changes in capillary filtration and/or permeability. In general, it has been demonstrated that the systemic (i.v.) administration of PGE1 decreases total peripheral resistance and systemic arterial pressure in rats (50,123), guinea pigs (83), dogs (75,84), and man (18). The cardiovascular effects of PGE1 are markedly affected by the route of its administration. Bergstrdm et a1. (8) is showed that the depressor effect of intravenously injected PGEl markedly smaller than that of intra—arterially injected PGEl in anesthetized dogs. These studies indicate that PGEl is effectively inactivated by a single circulation through the lungs. Other investigators employing the vital dye technique in the rat and guinea pig have presented conflicting data concerning the effect of PGE1 on capillary fluid and protein movement. The vital dye technique involves the injection of Evans Blue into the systemic circulation of an animal, and then an estimation of the amount of dye leaking at the administration site of the prostaglandin. The injection site for the prostaglandin is usually skin. The dye is assumed to bind to circulating plasma proteins, and will therefore be transported with the escaping protein. Microvascular permeability changes are quanti- tated by either measuring the diameter of the dye-leakage in the area of application, or more precisely by recovering the extravasated dye by extraction from the injected tissue, and measuring the concen- tration of the dye spectrophotometrically. Another approach using the vital dye technique involves the intravenous injection of radioactively labeled albuminanuithe subsequent measurement of the radioactivity of extravasated protein at the site of prostaglandin injection. 39 Utilizing the vital dye technique in the rat and guinea pig, several investigators have shown that PGE responds variably and l unpredictably with respect to vascular permeability changes (24,35,126). PGE1 had been observed to increase vascular permeability to various degrees (24,35), while other studies showed no apparent change in microvascular permeability (126,51). It is difficult to attempt to explain the reasons for these differences: however, the vital dye method has three apparent disadvantages which may have contributed: l) The concentrations of PGE1 at the testing site cannot be well controlled. Depending upon the rate of penetration and diffusion of PGE1 in the tissue, concentration differences could have occurred. 2) The extent to which the dye binds to the plasma protein is uncertain and variable. 3) Finally, the effect of PGE would be instantaneous and not a dynamic 1 change in vascular permeability. This last effect is caused by the rapid metabolization of the autocoid by degradative enzymes located in the tissue. Although criticisms exist for the vital dye procedure, the preponderance of the data utilizing this technique show increases in vascular permeability in response to PGE in the rat and guinea pig 1 (20,24,35,89,118). Kaley (61) noted increases in permeability, especially in venules, while Pangides (89) observed a direct relation- ship between the dose of PGE used and the average diameter of the l lesion. Another group of investigators has proposed that prosta- glandins (El and E2) contribute to the edema of inflammation by potentiating or increasing the vascular leakage induced by other mediators such as histamine and bradykinin (33,70,126). This poten- tiation can be inhibited by indomethacin pre-treatment (33). 40 The hamster cheek pouch preparation has been employed to study permeability changes in the microcirculation. In this procedure, FITC fluorescein-labeled dextran (M.w. = 145,000) is administered intravenously. Then a portion of the hamster cheek pouch is dissected and viewed microscopically. Leakage sites are counted after topical application of various agents by areas of fluorescence. This tech~ nique is of more value than the vital dye technique, in that it allows for the dynamic study of macrmolecular leakage in vivo. Using the more reliable hamster cheek pouch preparation, Joyner et al. (59) and Svensjo (113) have shown that PGE , in doses ranging 1 from 1 to 100 ng, increases the leakage of FITC-dextran from post- capillary venules. These investigators do not believe that the aug- mented macromolecular leakage occurs as a result of vesicular transport. They suggest that interendothelial gaps form as adjacent endothelial cells "round up" in response to the prostaglandin or other permeability inducing agent in a manner similar to that of histamine or bradykinin. STATEMENT OF THE PROBLEM A wide variety of experimental evidence demonstrates that the prostaglandins, especially PGE , are intimately involved in the inflame 1 matory process (53,118). PGE1 causes an increase in microvascular permeability"to macromolecules owing to a direct action on the micro- vascular membrane which results in the formation of large venular gaps similar to that produced by bradykinin and histamine (59,113,114). Some evidence suggests that the direct macromolecular permeability increasing action of PGE is considerably less potent than that produced 1 by histamine or bradykinin (113,118). This lesser direct effect is apparently offset by the unique ability of PGE to potentiate the l actions of histamine and bradykinin on the microvascular membrane (33,70,113,127). These actions of PGE1 on the microvascular membrane, coupled with the elevated concentration of prostaglandins in inflamma- tory exudates (3,40,92,128,129), suggests an important role for these autocoids in inflammation. This hypothesis is further supported by the observation that pre-treatment with indomethacin or aspirin reduces the intensity of inflammatory responses (5,20,92). The evidence discussed above implicating a role for the prosta- glandins in inflammation is primarily from studies in rats, hamsters and guinea pigs. The few studies performed in the dog (25,41,58) provide no conclusive evidence that PGE1 causes an increase in micro- vascular permeability to macromolecules, suggesting the possibility of 41 42 species variability. If true species variability exists, the relevance of conclusions based on studies in rats, hamsters, and guinea pigs to inflammatory processes in humans is questionable. In the dog studies, the doses of PGE1 employed were small and the duration of the infusions were usually short. These factors could have contributed to the lack of definitive results. Because of the overwhelming evidence supporting a role for prostaglandins in the rodent studies, and the grounds for valid criticism of the canine studies, a systematic study of the effects of PGE1 on fluid filtration and macromolecular efflux is clearly warranted. These observations prompted the present study to characterize the effects of one hour infusions with a dose range of PGE1 from 2.0 to 32.0 ug/min, i.a., on fluid filtration and macro- molecular efflux in the canine forelimb. Also, several experiments will test the effect of systemic indomethacin prior to the onset of a local PGE1 infusion on fluid filtration and macromolecular efflux. It is the interest of this study to correlate these pharmaco- logical data with the morphological data already available from the Swedish investigators. METHODS These experiments were conducted on 51 mongrel dogs of either sex weighing approximately 30 kilograms (range: 20-40 kg). Anesthesia was induced with 30 mg/kg of sodium pentobarbital and was supplemented periodically throughout the duration of the experi- ment as required. A Harvard respiratory pump was used to ventilate each animal with room air at a frequency of 10 cycles per minute and a tidal volume that was adjusted according to the animal's weight. The virtually intact canine forelimb perfused at either constant brachial artery or natural inflow was employed as the test organ to collect subcutaneous lymph and measure lymph total protein concentra— tion. The surgical process necessary to prepare the forelimb con- sisted of electrocautery to make small incisions superficial to the axillary portion cf the brachial artery, the cephalic vein (5 cm distal to the elbow), the second superficial dorsal metacarpal vein of the right forelimb and, in the constant brachial artery inflow preparations, an incision over the right femoral triangle. Additionally, an incision was made in the neck lateral to the trachea, for isolation and cannulation of the left common carotid artery and jugular vein. Ensuing the neck vessel cannulations, a side branch of the brachial artery, a skin lymph vessel in the region of the cephalic vein and the dorsal paw vein were isolated. Following the intravenous administration of 10,000 U.S.P. units of heparin via the jugular vein 43 44 catheter, the small blood vessels were cannulated in an upstream direction with polyethylene tubing, thereby allowing for the measure- ment of brachial artery perfusion pressure and skin small vein pressure. The brachial artery side branch was cannulated with PE-50 tubing inserted to the point of bifurcation of this vessel from the major artery. PE-60 tubing was used to cannulate the skin small vein. The cannulated small vessel acts as an extension of the catheter and, therefore, the pressure measured is that which is present in the vessel or vessels to which the cannulated vessel unites. This pressure is a true lateral pressure only if the cannulated vessel is patent, as verified by the ability to freely withdraw blood from and to flush physiologic saline into the vessel. The presence of the catheter does not measurably alter the pressure in the arterial or venous system because, in the canine forelimb, the cannulated vessel is a negligible fraction of the total cross-sectional area of the arterial and venous vascular bed, and there are abundant artery to artery and vein to vein anastomoses (25,42,45). In the natural inflow experiments the cannulated side branch of the brachial artery was used for the local infusion of PGE1. The lymphatic vessel was cannulated in an upstream direction with polyethylene tubing for lymph collection. These lymph vessels in the cephalic vein area below the elbow drain forelimb skin and paw (76). Two or three were usually tied centrally and one of them was cannulated distally with a beveled 10 cm length of PE-lO tubing. The walls of these vessels were quite substantial, necessitating a needle puncture (22 gauge) prior to cannulation. In the experiments employing a constantly perfused brachial artery forelimb, the brachial artery was isolated, tied and transected 45 about 5 cm proximal to the previously cannulated side branch. Blood that was obtained from a cannulated femoral artery was pumped at a controlled flow into the transected brachial artery. A Sigmamotor pump (Model T68H, Sigmamotor Inc., Middleport, NY) was used to keep inflow constant at a value which produced a perfusion pressure similar to systemic pressure (see Figure V for summary). Lymph was collected in 0.5 ml graduated cylinders constructed from plastic pipettes. Forelimb skin small vein pressure, brachial artery perfusion pressure (constant flow only), systemic blood pressure, and skin lymph flow rate were continuously monitored and recorded at 10 minute intervals. All pressures were measured with Statham pressure transducers (Model P23Gb, Statham Instruments, Inc., Oxnard, CA), connected to a direct writing oscillograph (Model 7754A, Hewlett- Packard Co., Palo Alto, CA). After one 10 minute control period, followed by two consecutive 10 minute vehicle-control periods, the local intra—arterial adminis- tration of PGE was begun. In the constant inflow preparations, PGE 1 1 introduction was accomplished by infusing directly into the circuit behind the Sigmamotor pump to insure adequate mixing of the drug. In the natural inflow studies, PGE was infused directly into a side 1 branch of the brachial artery. All concentrations of PGE1 were administered at a delivery rate of 0.2 ml/min with a Harvard Apparatus infusion/withdrawal pump. Arterial blood samples (5 ml aliquots) for ascertaining hemato- crit and plasma protein concentrations were drawn from the cannula measuring systemic blood pressure. The blood samples were taken 10 minutes prior to the infusion of PGE , and at 30 minute intervals 1 throughout the duration of the experiment. The protein concentrations 46 FIGURE I BRACI-IIAL ARTERY \ \\ \ \ \ \ CONSTANT FLOW \ \ CANNULA := ‘ \ (BRACHIAL ART. . ~ \ y."- \\.\—-- PERFUSION \\ PRESSURE AAAAALVAESSEL' ‘ ‘ (SIDE BRANCH BRACHIAL ART.) SMALL SKIN VEIN PRESSUR Schematic diagram of the canine forelimb technique. 47 of the lymph and plasma samples were analyzed by the modified Biuret reaction (7). The samples were read spectrophotometrically in grams percent with an ACCU-STAT Blood Chemistry Analyzer (Clay Adams, Model 2000) which had been calibrated with samples of known protein concentration. Modifications to this standard protocol were instituted in 15 of the 57 total experiments. In 12 studies 5 mg/kg of indomethacin was given as pretreatment. After 60 minutes this was followed by 16 ug/min of PGE infused into the brachial artery of the forelimb l for 60 minutes. The values for lymph flow rate and lymph protein transport were compared to those values obtained with 16.0 ug/min of PGE1 alone using a group t-test. The indomethacin was given over a five minute interval into the jugular vein. In another series, the angiotensin I to angiotensin II converting enzyme blocker, SQ 14225 (captopril), was administered intravenously 20 minutes prior to, and throughout, the 60 minute duration of the local infusion of PGE1. These experiments were performed on constant inflow preparations using three animals. At the conclusion of all experiments the animals were sacri- ficed. The experimental (right) and control (left) forelimbs were then severed approximately two centimeters proximal to the humeral condyle for measurement of forelimb weights. The brachialis, biceps and triceps muscles were carefully dissected down to their tendons of insertions on the radial and ulnar tuberosities. Diligent care was assumed to insure that the limbs were always severed at the same points on the humeral condyle. The forelimbs were then exsanguinated and weighed. Limb weights were compared (experimental vs. control) using a paired t-test. Previous work by our group, in the same preparation, 48 produced no significant changes in limb weights. Therefore, neither the ethanol-saline vehicle nor the trauma produced by the surgery were responsible for the significant increases in limb weights seen in dogs with marked increases in lymph total protein concentration. Moreover, our group has shown that when lymph total protein concentration did not significantly increase, the experimental and contralateral control limb weights were similar. Consequently, this method appears to be valid for ascertaining changes in limb weights. All other data were statistically analyzed by the Analysis of Variance (Randomized Complete Block Design) and the means were compared to control by the Least Significant Difference Test (10). PGEl was procured from The Upjohn Company, Kalamazoo, Michigan, compliments of Dr. John E. Pike (U-10136, Lot No. 12874-JHK-102C and 11894-VOV-77). Stock solutions were prepared by dissolving 10 mg of PGE into 10 ml of absolute ethanol. Appropriate dilutions were made 1 using normal saline as the solvent. Infusion rates of PGE1 used in these experiments were: 2 ug/min, 4 ug/min, 16 ug/min, and 32 ug/min. Indomethacin solutions (Sigma Chemical Co., I 7378; (l- p- chlorobenzoyl -5-methoxy indole-3-acetic acid) were prepared by dissolving an appropriate amount of indomethacin, based on the animal's weight, into 100 ml of 1N saline containing 90 mg of NaHCO Dissolu— 3. tion of the indomethacin took approximately two hours of continual stirring. The NaHCO3 raised the pH of the 1N saline to 8.2-8.4, which aided in the dissolution of the indomethacin and augmented its half- 1ife to about three hours at room temperature. Captopril was obtained from the Squibb Laboratories, New Brunswick, New Jersey. It was prepared for infusion by dissolving it 49 it in normal saline to an appropriate concentration for an infusion at a rate of 14 ug/kg/min. RESULTS The data in this section are presented in Tables III through VII. Measurements of vascular pressures and collection of lymph samples were made over 10 minute periods. Arterial blood samples were taken every 30 minutes. In Tables III through VI, three control readings followed by six experimental readings were acquired. The control readings were of two types. The -20 minute control reading was measured following a 10 minute period where no drug or vehicle was infused. The -10 and 0 minute readings were taken during the brachial artery infusion of a vehicle solution (ethanol-saline). At 0 minutes the vehicle infusion was terminated and replaced with the intra- brachial artery infusion of PGE1 for 60 minutes. In Table VII, the format is essentially the same as in Tables III through VI, except that at -60 minutes a "solution and/or drug free" control reading was taken followed by the intravenous injection of indomethacin. The -10 and 0 minute readings were acquired during the vehicle infusion period. 262212.11}. In the natural and constant brachial artery inflow forelimb preparations, PGE (2.0 ug/min, i.a.) moderately increased systemic l . arterial blood pressure. Perfusion pressure markedly decreased in the constant inflow experiments, while skin small vein pressure minimally 50 51 increased. Skin small vein pressure rose substantially in the naturally perfused forelimbs. Lymph total protein concentration and lymph protein transport increased in both forelimb preparations, yet lymph flow rate changed minimally. Plasma protein concentrations remained unaltered relative to control values, while the hematocrit ratios increased. Table IV Four micrograms per minute of PGE1 infused into the brachial artery of constantly perfused forelimbs for 60 minutes augmented systemic arterial blood pressure to the same extent as 2.0 Ug/min of PGE1 (Table III). Perfusion pressure markedly decreased, while skin small vein pressure remained unchanged. Lymph total protein concen- tration and lymph protein transport increased with only a very slight enhancement of lymph flow rate. These lymph parameters are almost identical to those obtained at an infusion rate of 2.0 ug/min of PGE1 (Table III). Plasma protein concentration remained unaltered, while the hematocrit ratio increased relative to control values. Table V At 16.0 ug/min, PGE1 did not significantly alter systemic arterial blood pressure in either the natural or constant brachial artery inflow forelimb preparations. In the constant flow studies, perfusion pressure decreased markedly as skin small vein pressure remained unchanged. Skin small vein pressure increased in the naturally perfused forelimbs. In both preparations, lymph total protein concentration and lymph protein transport increased relative to control levels, yet lymph flow rate increased only slightly. The total lymph protein 52 concentration in the constant inflow experiments increased by almost 2.0 gms %, which is greater than the values obtained with 2.0 or 4.0 ug/min of PGE (=l.2 gms %). Hematocrit ratios increased in both 1 preparations, while the plasma protein concentration remained unchanged in the constantly perfused forelimbs. Interestingly, the plasma protein concentration slightly increased relative to control in the naturally perfused forelimbs. Forelimb weight measurements were obtained only from naturally perfused forelimbs. These forelimbs significantly increased in weight by 55 grams. Table VI The local, brachial artery infusion of PGE (32.0 ug/min) l moderately decreased systemic arterial blood pressure in naturally per- fused forelimbs and slightly lowered this pressure in the constant inflow forelimbs. Skin small vein pressure did not change in the constant inflow experiments, while it increased in the natural flow studies. Perfusion pressure was greatly reduced in the constant brachial artery inflow preparations. The total protein concentrations of the lymph rose with both experimental procedures: however, the increment was somewhat lower than the levels achieved at an infusion rate of 16.0 ug/min of PGE1 (Table V). In the naturally perfused forelimbs, lymph flow rate increased relative to control, which is different from the lymph flow rates obtained with the other three infusion rates (Tables III, IV and V). At constant inflow, however, lymph flow rate increased minimally, which is similar to the results obtained with the other infusion rates. Lymph protein transport increased in both forelimb procedures. 53 Plasma protein concentration increased in both preparations, whereas at other infusion rates plasma protein concentration levels remained relatively unchanged (Tables III, IV and V). Hematocrit ratios also increased in both forelimb procedures. In the naturally perfused forelimbs, weights were signifi- cantly augmented by an increment of 50 grams and in the constant inflow forelimbs, weights increased significantly by 26 grams. Table VII The systemic administration of 5 mg/kg of indomethacin failed to alter significantly the control values of the measured parameters in both naturally and constantly perfused forelimbs. Following this pre-treatment with indomethacin, the local, intra-arterial infusion of 16.0 ug/min of PGE for 60 minutes did alter the vascular lymph and 1 plasma parameters quantified. In the constant brachial artery inflow studies, systemic arterial blood pressure increased significantly (=10 mm Hg), while little change was observed at natural inflow. These results are somewhat similar to those obtained with the infusion of 16.0 HQ/min of PGE1 alone (Table V). Perfusion pressure markedly decreased, being almost iden- tical to the results shown in Table V. Skin small vein pressure increased at natural inflow but did not change at constant inflow. Lymph total protein concentration increased in both forelimb pre- parations. Interestingly, lymph flow rates were statistically different from those obtained when PGE1 was infused into the brachial artery without prior systemic pre-treatment with indomethacin (Table V). The flow rates of the lymph in both preparations increased markedly, with a larger increase observed at natural inflow. As a result, lymph protein 54 transports increased significantly from those values obtained at 16 ug/min of PGE1 alone. Hematocrit ratios increased significantly and plasma protein concentrations failed to change in both natural and constant inflow forelimbs. Forelimb weights also significantly increased by 64 grams in the natural flow forelimbs and by 14 grams in the constant flow forelimbs. Captopril Studies In three constantly perfused brachial artery forelimbs, 15 ug/kg of captopril (SQ 14225) was administered intravenously 20 minutes prior to and throughout the 60 minute duration of the intra-brachial artery infusion of 16.0 ug/min of PGE No significant change from 1. control was Observed in any of the vascular lymph or plasma parameters measured. 55 .oEfiu oumu ou m>HumHmu mo.o w m .muflu oumu Op w>flumeu Ho.o w m .1 awv «mv mm m0 . «Nv «av mm m2 ufiHOOumEmm 0.0 0.0 m.m no Aw msmume 0.m 0.m n.m m2 cumuoum mammam we. +mm. +00. «mm. «an. 00. am. 0m. 0m. mo Aces 0H\0ev +Nm.m +m0.~ 0m.H 00.H mH.H m0. 0m. mm. «0. uz unommcmue cumuoum edema .m.m +a.~ «0.m .0.m e.m +0.m m.~ H.m 0.~ no in meaumv «m.m «0.m «~.m «H.m m.m e.~ m.m m.~ v.m m2 cflwuoua Hence sass; +N0. +m0. «m0. «m0. «m0. +m0. H0. H0. H0. no Aces 0H\Hav +n0. +50. 00. m0. 00. m0. H0. H0. «0. m2 mumm scam amasq A0: see 40H «0H 40H «ma Ama .NH HH Ha Ha no musmmmum «AH 40H «AH Aha ,mH «be He AH NH m2 cflm> Haaam :Axm Ame see «m0 «m0 «mm «00 ,AN0 «00 mNH «NH HNH mo musmmwum seemsuumm rum sac «mma aema «mma «mma .0mH +0mH AMH 0mm and mo musmmmnm wooam mma +0mH +HMH +~mH mma ANH 0NH mfla mHH mz Haaumuua ermumsm 00 0m 00 0m H 0m 0H 0 0H- .Immwwuu Ammuscflev mafia UOHHOQ 30¢.me CH mwm HOHHCOU OHUfl£w> HOHHCOU Asoam ucoumcoo u m0 .30Hm Housumc u mzv Amucv uwuooumemn can monommmum Hoasumn> .mEmmHm pom £QE>H mo coflumuucwucoo samuoum .puommnmuu cflmuoum nmESH .30Hm nmESH co mepacaE 00 now bEwkuow on» oucw A.<.H .cfis\mn o.mv Hmum penance zaauooH mo muowmmm .HHH canoe 56 .OEHU OHON OH 0>flUmHOH mooo .mfiwu Oman 0» w>wuuaou Ho.o w m V + .1. «N0 «Hm hm mo uHHooumfiwm Aw msmumv 0.0 H.0 0.0 mo :Hmuoum mammHN HcHs 0H\0ev 0N. 00.H +0H.H «0m.H +Hm.H N0. 0m. Nm. mm. 00 uuommcmue :Hmuoum saaNH Aw mauumv «m.m «v.m «m.m «0.m «0.m N.N 0.N 0.N 0.N mo :Hmuoum Hmuoe sassH . HcHs 0H\Hee N0. m0. mo. «00. «00. m0. H0. N0. N0. mo mumm onm nmssq Am: 220 whammwum NH NH NH NH NH HH HH HH HH no ch> HHmam :me A0: 220 .00 400 «N0 «00 «mm «00 mHH mHH 0NH no wusmmmnm consuumm H00 sea A whammmum coon +NOH +HOH +HOH +0mH +HOH +mmH NNH 0NH 0NH mo HmHumuAa oHsmumNm 00 00 00 0m H 0N 0H 0 0H: 0N- AmmuscHa. maHe pewuwm COHmsmcH mum Houucou oHoH£o> Houucou Ason acmumcoo u mov Aoflfiv UflHOOHMEOS UGO mmnzmmoum nodaomu> .uEmem pom £QE>H mo :oflumuucmucoo camuoum .uuommcmuu :Hououm smaha .3OHM :QESH co mwuacfls om How bEmeuom on» oucH A.¢.H .cHE\ma o.vv Hmwm penance aaauooH mo muoommm .>H manna 57 av? «mv hm av .00 «00 00 02 HHAooumemm 0.0 0.0 0.0 00 A0 mamuma ,N.0 0.0 0.0 02 :kuoum mammHm ,0N.H A00.H «0H.N «00.N .00.H N0. 0N. NN. 0N. no AcHa 0meee A00.H +0N.H +NN.H +00.H «HN.H 00. 00. N0. 02 uuommcmue :Hmuoum rmst ,0.0 ,0.0 45.0 «N.m AH.m 0.N 0.H 0.H 0.N 00 i0 msmuma «N.N AN.N «0.N «0.N 0.N H.N 0.N 0.H 02 :Hmuoum Houoe amaNH 400. +00. «00. «00. «00. .00. H0. H0. H0. 00 AcHa 0H\Hsa 00. 00. 00. 00. 00. 00. N0. N0. mz 0000 onm :aaNH I00 250 0H 0H 0H 0H 0H 0H 0H 0H 0H 00 musmmmum «NH «0H ,0H .0H .0H ,0H 0 0H 02 ch> HHoam :me H00 see 400 «00 ,N0 ,00 .00 «N0 N0H 00H 00H 00 musmmmum nonsuuma i0: see NNH HNH HNH 0HH 0HH 0HH 0HH 0HH NHH no musmmwum coon 00H 0NH 0NH 0NH NNH 00H NNH HNH Nz HmHAmuAN oHamumNm 00 00 00 00 4 0N 0H 0 0H- 0N- AmmuscHee msHe noHAoa :onsucH .000 Honucoo oHoHnm> Houucoo Hzoam unaumcoo u mu .30Hm Hmuauu: u mzv Hmucv uwuooucsoc can unmwoz bEHHouom .meSmmonm Hudsomm> .mEmmHm pom £QE>H mo cofiumuucoocou :flwuoum .uuomwcunu :Hououm smaxa .3OHM smama so mouncfifi 00 How nawamuom can one“ H.¢.H .cefi\m: o.wHV Hmum pomsmcfi adamooH mo muowumm .> wanna 58 .OEHU OHQN OD 0>HUMHQH m0.0 V m + .080» Owen ou w>HuuHmu Ho.o w m i +mm mucmummmwo mmb HmucmEHHumxm mom. Honucou .mz Sun “9:33 unmfloz nfiwaouom Apmzcwucoov > magma 59 «vv *vv mm m0 «00 +N0 00 02 HHuooumsmm «N.® «N.© b.m m0 Aw mEMHmv «v.0 «m.® m.m mz chuoum mammHm N0. +HN. +NN. «00. +00. 00. 00. 00. 00. mo HcHe 0H\0sv «0H.0 00H.0 «00.0 «NN.0 «00.0 00.N 00. 00. 02 un0dmcmue chuoum smaNH «N.N «N.N +0.N 00.N +0.N +0.N 0.N 0.N 0.N mo H0 msmumv «H.0 «0.0 «H.m .H.0 +0.N 0.N N.N N.N 02 :kuonm Hmuoe rme>H N0. 00. 00. +00. 00. N0. N0. N0. N0. mo AcHe 0H\Hev «0H. 00H. 00H. «0H. 00H. +00. N0. N0. 02 wumm onm ndssq H00 550 0H 0H 0H NH NH NH NH 0H NH mo musmmmum «0H .0N «0H 00H «0N «0N NH NH 02 ch> HHmem :me H00 see «N0 «H0 «H0 «00 000 «N0 00H 00H 00H mo musmmmnm :onsmumm 100 sec NHH HHH 00H 00H 00H 00H 0HH 0HH 0HH mo musmmmum ooon 000H .00H 000H «00H .00H «00H 0NH HNH 02 HmHumuua oHamumam 00 00 00 00 4 0N 0H 0 0H- 0N: AmmuscHa. meHe ooHumm consmcH .000 Houucoo mHoHnm> Houucoo H30Hw ucmumcoo u mu .30Hm Hausum: u mzv Honey uwuooumsmn pom uzmfim3 QEHHwHom .mwusmmoum HuHsomm> .mEmmHm poo smewa mo coflumuucmucoo cfimuoum .uuommcmuu Cwmuoum SQESH .30Hu smeaa co mwuscfle om How QEHHmHom 020 case A.¢.H .cws\m: 0.va Hmwm penance >HHMUOH mo muommwm .H> canoe 60 .wEHu OHoN on w>Huunu 00.0 V m + .meu Open on w>wumeu Ho.o w m k. +©N «om oucmumwwflo Hob vmb Hmucoswuomxm own vow Houucoo mu Hon: «msmumv .02 ”£303 neHHonm Aposcwucoov H> manna 61 0.0 0.0 0.0 0.0 mo H0 050000 0.0 0.0 0.0 0.0 02 ckuoum mammHm +H0.N 3+00.N 3000.0 0000.0 300N.0 0N0.N 00. N0. 00. 00 HcHe 0H\0av 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 00. 00. 00. 02 uuommcmna chuoum :050H 00.0 00.0 00.0 00.0 00.0 N.0 0.N 0.N 0.N 00 H0 msmu00 00.0 00.0 00.N 0H.0 00.0 +0.N N.N H.N 0.N 02 chuoum Hmuoe :020H 300. 300. 000. 00H. 30NH. 000. N0. N0. H0. 00 HcHs 0H\Ha0 300N. 300N. 000N. 00N0. 0000. 000N. 00. 00. 00. 02 0000 30H0 00E0H H00 250 HH HH HH HH HH HH 0H 0H HH mo ousmmmum 00N 00N 00N 00N 00N 00N NH NH 0H 02 0H0> HHmem 0H00 H00 sea 0H0 000 000 000 000 000 NOH HOH 0NH 00 ousmmwum consmumm H00 250 NOH +00H +00H 000H 000H +NOH 00H 00H NOH 00 wunmmmnm 000H0 0HH 0HH 0HH 0HH 0HH HNH 0HH 0NH 0HH mz HmHumuuc oHEOum0m 00 00 00 00 0N 0H 0 0H- 00- immuscHav meHe H pofluwm cOMmsmcH mum HOHDCOU 0H00£o> cofluooflcH .Och HBOHM ucmumcoo u no .3OHM Hmuaumc n hzv Houcv ufluooumEm: pom 020003 nsflaouom .wmusmmoum Hwasomm> .mEmon tam Shama mo cowumuucoucoo :kuoum .uuommcmuu cwmuoum SQESH .30Hm smawa co moussfla 00 How nsflamuom ecu 00:0 >Hamooa pomsmcw H.¢.H 0206\01 o.mHv Hmom >3 pmonHom >Hmsocm>muusw popconC0 me\ms mv cwomnuoEOpCM mo muommmm .HH> OHQMB 62 .H> 000090 00000 :0E\0mwm on 0.00 £003 00000000 m0:00> 0:0 00 0>000000 00.0 w m .H> 0Hnmav meoHu cHE\H000 0: 0.0H 0003 00000000 00=H0> 000 00 0>H00H00 H0.0 w 00 .0500 000m 00 0>000000 mo.o w 0+ .0500 000N 00 0>000000 00.0 w m .1 +0H 000 000 00 A0": 0050000 000 000 000 02 000003 nsHHmuom 0000000000 000c0a0u0mxm 0000000 «m0 0M0 0m 0m mo «mm «mm 0m mm 02 0000000803 op 00 00 00 ON 00 o 00: cm: H000sc0av 0E0B 000000 :00mDMCH HMOQ 0000000 0000£0> :000o0flcH .OUCH Ap0sc00coov 00> 00009 DISCUSSION The role of "E" series prostaglandins in the inflammatory process has evoked substantial controversy for several years (35,58, 70,118,120,117,126). Although these prostanoids have been recovered from various inflammatory exudates, and inflammation can be attenuated by cyclo-oxygenase inhibitors (i.e., PG synthetase inhibitors), variable results have been obtained concerning the ability of PGE1 or PGE2 to mimic the edema of inflammation when given exogenously into a tissue. The data in the present study demonstrate increases in forelimb weight in the dog following local, intra-brachial artery infusions of PGE1 for 60 minutes. This provides evidence for edema formation subsequent to an increase in net fluid filtration. The rise in lymph total protein concentration results from an increase in micro- vascular permeability to macromolecules. The precise mechanism by which the augmented microvascular permeability occurs cannot be ascertained from these experiments. However, work carried out by Swedish laboratories (59,113,114) suggests that it probably results from a direct action of PGE1 on the microvascular membrane causing the formation of large venular gaps or pores. The heightened skin small vein pressure observed in the naturally perfused forelimbs suggests an increase in capillary hydrostatic pressure, since skin small vein pressure represents a minimum for capillary hydrostatic pressure (45,74). These observations imply that the mechanism of edema formation is by a 63 64 decrease in the transmural oncotic pressure gradient and an increase in the transmural capillary hydrostatic pressure gradient. In the constant brachial artery inflow preparations, the edema was attribu- table primarily to a decrease in the transmural oncotic pressure gradient, since skin small vein pressure failed to change from control in these forelimbs. PGEl produces only increases in microvascular permeability in doses exceeding those needed to produce vasodilation. Theoretically, this is physiologically important, since it permits a means for "E" series prostaglandins to function in the local regulation of blood flow without increasing microvascular permeability to macromolecules (125). It would be unprofitable if these agents simultaneously increased net fluid filtration. is The increase in microvascular permeability yielded by PGE1 considerably less than that produced by bradykinin and/or histamine (72,74). The average maximal increase in lymph total protein concen- tration produced by PGE1 is about 1.5 grams percent (Table V). Large doses of histamine or bradykinin increase the total protein concentra- tion of the lymph to values approaching that in plasma (=5.5 gms %) (72,74). The inability of others to demonstrate an effect of PGE1 on macromolecular permeability in the dog is probably related to the doses studied (25,41,58). These investigators performed the majority of their experiments at infusion rates below 2.0 ug/min. In addition, the short duration of the infusion and the failure to directly measure the efflux of protein or macromolecular tracer substances may account for their conclusions. However, the data presented in this 65 dissertation using levels of PGE above 2.0 ug/min are consistent with l the data from many other laboratories using rats, mice and hamsters (24,35,113). PGE1 affects lymph flow rate differently from bradykinin and histamine (46,72,74). At infusion rates between 2.0 and 16.0 ug/min, PGE1 produced very small dose independent increases in lymph flow rate. These lymph flow rates were similar in magnitude under both natural and constant brachial artery inflow conditions. In contrast, 2 ug histamine base/min produces an increase in the protein concentration of the lymph similar to that elicited by PGEl, yet produces extremely large increases in lymph flow rate (46). The lymph flow rates induced by histamine were substantially greater in the naturally perfused forelimbs than those perfused at constant brachial artery inflow. Following pre-treatment with indomethacin, PGE (16.0 ug/min, i.a.) 1 yielded an increase in the total protein concentration of the lymph similar to that produced without indomethacin, but produced very dramatic increases in lymph flow rate (Table VII). These elevated lymph flow rates were statistically greater from the flow rates obtained at 16.0 ug/min of PGE1 alone. Moreover, these flow rates were pro- portionally greater in forelimbs naturally perfused than in those per- fused at constant brachial artery inflow. It is interesting to speculate the reasons why indomethacin pre- treatment enhanced the rate of lymph flow produced by PGE Since 1. indomethacin inhibits cyclo-oxygenase, it is tenable to conclude that in the absence of indomethacin one or more eicosanoids produced subsequent to the cyclo-oxygenase step (Figures III and IV) could be liberated in response to exogenous PGE and actively constrict pre- 1 nodal lymphatic vessels, thereby impeding drainage. It has long been 66 known that the topical application of various vasoactive agents can affect lymphatic contractility (108,116). The mechanism is believed to occur by contraction of smooth muscle cells, since the anatomical structure of lymphatic vessels includes, in addition to an external adventitia and an internal endothelium, a tunica media made of smooth muscle cells arranged in several layers (116). This PGEl induced lymphatic contraction might occur via the action of certain enzymes in the prostaglandin biosynthetic cascade beyond the cyclo-oxygenase step, which would increase the relative amounts of various eicosanoids in the body. For example, stimulation of 9-keto-reductase by PGE1 would result in enhanced levels of PGF Also, this enzyme could convert 20' the exogenous PGE to PGF . PGF and PGF are vasoconstrictors 1 la 20 la (27,31) and are antagonists of PGE induced permeability in the rat 1 (24). It is possible that either of these prostaglandins could be actively constricting pre-nodal lymphatics. Thus, not only would indo-' methacin prevent endogenous PGF formation by inhibiting cyclo- 20 oxygenase, it would also prevent PGF , since it also inhibits 9-keto- la reductase (80). PGE1 could likewise stimulate thromboxane synthetase in the platelet causing excessive release of TXA TXA , like PGF or 2' 2 2a PGFla' is a vasoconstrictor (79,81) and could possibly cause active constriction of lymphatic vessels. The theory that exogenous prostaglandin introduced into a tissue to cause an imbalance of the prostaglandin synthetic cascade, thereby increasing the relative amounts of other eicosanoids, is not a novel idea. Chahl and Chahl (20) have shown that part of the action of PGE1 in inflammation is produced indirectly by releasor stimulation of the synthesis of other prostaglandins and their precursors. Their 67 experiments show that in rats pre-treated with indomethacin the area of swelling on the skin induced by PGE is significantly reduced from 1 the amount of swelling produced by PGEl alone. Conceivably, in the indomethacin pre-treated rats, drainage was not impeded, resulting in an attenuated area of edema. When cyclo-oxygenase is blocked, this terminates the competition for substrate (i.e., arachidonic acid, di-homo-y-linolenic acid and eicosopentaenoic acid) by cyclo—oxygenase and the lipoxygenases (Figure IV), allowing lipoxygenases to act freely and yield larger amounts of hydroperoxides and hydroxy-acids. Indomethacin could therefore be indirectly stimulating the production of these latter substances, which might participate in the opening of lymphatic vessels. Another possible explanation for these data would be that normal circulating levels of TXA , PGF and PGF antagonize the permeability 2 la 20 effects of exogenous PGE (24) and may also exert a normal tonic 1 effect on the lymphatics. Administration of indomethacin would pre- vent further generation of PGF 0 and PGF 0 by the vascular endothelium 2 l (51) and TXA2 by the platelet (79,81), thereby reducing their circu- lating levels. Consequently, the tonic effects on the lymphatics would be removed along with any microvascular antagonizing effects. Thus, after one hour, the local infusion of PGE would effectively 1 increase lymph flow rate concomitant with the increased total protein concentration of the lymph. It should be noted that at 32.0 ug/min, PGE produced very pro- 1 nounced increases in lymph flow rate (Table VI), yet the increase in lymph total protein concentration was no greater than that produced at 16.0 ug/min (Table V). Perhaps at this dose the direct action of PGE1 largely counteracts the effects of the presumed endogenously released 68 lymphatic constrictor substance, the existing prostaglandin micro- vascular antagonist, or a lymphatic, tonically active substance. The local intra-arterial infusions of PGE1 at 2.0 and 4.0 ug/min produced significant increases in mean systemic arterial blood pressure (Tables III and IV). The elevated mean arterial pressure was somewhat less at 16.0 ug/min of PGE1 (Table V). Since the augmented blood pressure is not attributable to the ethanol-saline vehicle (39)} it must be ascribed to an action of PGE1. It is well known that PGE1 exerts an effect on the central cardiovascular centers which causes an elevation in arterial blood pressure (51). This could explain the observed hypertensive effects of the lower PGEl doses. When the PGEl infusion rate was increased to 32.0 ug/min, mean systemic arterial blood pressure decreased. It is possible that at this high dose, the central pressor effect may be overwhelmed.by the peripheral depressor effect. This is supported by the fact that Kadowitz et al. (60) have shown that PGEl decreases the release of norepinephrine from adrenergic neurons via stimulation of presynaptic receptors. Captopril (SQ 14225) inhibits the converting enzyme which trans- forms Angiotensin I to Angiotensin II. This converting enzyme, which is richly abundant in the lung, is also believed to catalyze the con- version of bradykinin to inactive fragments. In a single passage of bradykinin through the pulmonary circulation, the lungs destroy up to 90% of the active peptide. It has been suggested that the permeability inducing effects by PGE or PGE in the microcirculatory bed is mediated 1 2 in part via bradykinin (52). This hypothesis is currently unsettled and requires additional experimentation. In this study, three dogs were treated with 15 ug/kg/min of captopril 20 minutes prior to and throughout the 60 minute local, brachial artery infusion of PGE (16 l 69 ug/min). If the microcirculatory effects of PGE1 are mediated in part by bradykinin, then a more dramatic effect on permeability should be observed when bradykinin inactivation is prevented. However, no significant change from control was observed in any of the vascular lymph or plasma parameters quantified, indicating that the effects of PGE1 are probably not mediated by bradykinin. Examining the data generated from the 51 dogs used in these studies, there is more variability in the lymph parameters than that which is observed for either bradykinin or histamine (72,74). This unusual variability of response has also been noted by Freeman and West (35) in their studies of inflammation on rats. The reason for this variation is not immediately clear, although it could be related to the release of an endogenous substance by PGE (PG and/or catechola- 1 mine). Additional experimentation is needed to resolve this point. In summary, the local intra-arterial infusions of PGE (2.0 to 1 16.0 ug/min, i.a.) for 60 minutes into the canine forelimb perfused either naturally or at constant inflow produced marked vasodilation and increases in lymph total protein concentration, yet only minute dose independent increases in lymph flow rate. The increase in lymph flow rate was similar in magnitude with either the natural or constant inflow preparation. FGE1 (16.0 or 32.0 ug/min, i.a.) also produced increases in forelimb weight due to edema formation. The weight increases were substantially greater under natural inflow conditions. Following pre-treatment with indomethacin, PGE produced vasodilation l and increases in lymph total protein concentration similar to that in the absence of indomethacin and also produced significant increases in the flow rate of lymph. Therefore, these data suggest that PGE may 1 well function as an inflammatory mediator in the dog. However, its 7O role is probably different and its mechanism of action more complex than the putative mediators of inflammation (i.e., histamine and bradykinin). APPENDICES APPENDICES This appendix lists, in the form of tables, all the individual observations for the experiments performed in this study. Also listed are the means, standard error of the mean and the statistical significance. 71 72 00 00 00 00 00 00 00 00 00 00000 0000:000 «00 «m0 «50 «00 «m0 000 00 00 m0 0:005 00 00 00 00 00 b0 00 m0 m0 m0 m0 m0 m0 m0 v0 00 00 O0 00 00 50 00 00 m0 m0 m0 N0 00 00 00 ON on 00 00 00 00 :00 see 00 00 00 00 00 00 00 00 00 00500000 00 om 0m 0m mm on m 0 N0 :00> 00050 :0xm m0 v0 00 m0 m0 m0 00 00 00 00000 0000:000 m~0 +om0 +0m0 +Nm0 mm0 bm0 0m0 000 000 0:005 hm0 hm0 mm0 mm0 hm0 0m0 mm0 mm0 om0 om0 om0 mm0 om0 om0 mm0 N00 000 . 000 hm0 mm0 mm0 mm0 0m0 mm0 0m0 mm0 mm0 om0 om0 0m0 mm0 mm0 N00 mm mm om 0mm 55V 000 0N0 0N0 o~0 0N0 0m0 Om0 om0 Om0 00500000 0000m 000 000 mv0 mm0 om0 om0 om0 om0 mm0 00000000 005000>m 00 on 00 Om on 00 o 00: om: 0000::050 0509 000000 :00000:H umom 0000:00 0000£0> 0000:00 _ A3000 000500:V Amu:v 000000050: 0:0 000500000 0009000> .05000m 0:0 :05x0 00 :000000:00:00 :00000m .000m0:000 :000000 £05>0 .3000 00500 no 0000:02 00 000 02000000 0:0 o0c0 0.0.0 .c0sxm: 0.00 0000 000:0:0 0000000 00 0000000 .000-< 00000 73 0.0 m.0 N.0 m.0 N.0 N.0 N.0 0.0 0.0 06000 0000:000 00.0 0v.m «m.m «0.0 0.N 0.N m.m 0.N 0.N 0:005 0.m 0.m 0.m >.m ~.m 0.N 0.0 0.0 0.0 0.N 0.N 0.N 0.N 0.N 0.N 6.0 6.m 0.0 0.0 0.0 0.N 0.N 0.N 0.N 0.0 6.0 0.N m.v N.0 m.m 0.m 0.m ~.m 0.N 0.N 0.N 0.m. m.m 0.0 0.0 m.m .m.m 0.N 0.0 0.0 :0 050000 0.0 m.m m.m 0.m- m.m m.m 6.m 0.N 0.0 0000600 00060 20500 06.0 06.0 06.0 06.0 06.0 06.0 66.0 66.0 66.0 06000 00005000 +06.6 +06.6 06.6 06.6 06.6 No.6 06.6 06.6 No.6 0:005 06.6 m6.6 06.6 06.6 06.6 «6.6 06.6 06.6 «6.6 06.6 06.6 06.6 06.6 06.6 06.6 06.6 06.6 no.6 No.6 «6.6 no.6 06.6 no.6 06.6 06.6 06.6 «6.6 06.6 06.6 06.6 06.6 06.6 06.6 06.6 06.6 06.6 00.6 00.6 06.6 06.6 06.6 06.6 06.6 06.6 06.6 0:05 60x050 06.6 no.6 no.6 06.6 no.6 06.6 06.6 06.6 «6.6 0000 3600 £0500 60 60 60 on 0 60 60 6 60: on: 00005505. 0500 “000me COHmSMGH mwm 00.000000 wHOHSm> HOHHCOU 0009:00coov 000.0 00000 74 N.0 N.0 0.0 00000 00005000 0.0 o.m h.m 05008 0.0 0.0 m.m 0.0 0.0 0.0 0.0 0.0 v.m 0.0 m.m 0.0 6.0 0.0 0.0 00 050006 0.m o.m m.m 5000000 080000 om.00 m~.00 om.0 mm.0 hm.“ m0.0 00.0 00.0 00.0 00000 00005000 +mm.m +mv.m vN.H ov.H mH.H m®.o vm.o mm.o Nv.o mcmwfi 00.6 00.0 60.0 00.0 00.0 00.6 00.6 00.6 00.6 00.6 00.0 00.0 00.0 00.0 00.6 60.6 60.6 00.6 00.6 60.6 00.6 00.0 00.6 00.6 00.6 60.6 00.6 00.0 60.0 00.0 00.0 00.0 00.6 00.6 00.6 00.6 66.0 00.0 00.6 00.6 00.6 00.6 00.6 00.6 00.6 0005 60\056 60.0 06.0 06.0 00.0 00.6 60.0 60.6 00.6 00.6 000000000 0000000 00500 60 60 60 60 0 60 60 6 60: 60: 000000056 0500 000000 00000000 mom 00000oo 0000n0> 0000000 00000000006 000-0 00000 .msfiu cums on 0>wuuamu mo.o w m .mEflu cums on w>wumamu H0.0 w m 75 + * HH HM NH uouuo oumocmum «Nv «av mm mamms mv me ov hm wm cm mm mm mm mv mv mm vv vv H¢ vv mv ov afiuooumsmm om om ow om H om 0H 0 OH: own Amwuscflsv mafia oofluwm :meamcH mum Houucoo maofl£m> Houucou Acmncflucoov HHHu« wanna 76 «mm «mm *mo «vm *mo *om mmfl vmfl HNH mcwms no mo om pm mm mm ONH oma bma mm mm mm no no mo NmH OMH NMH No oo «0 oo oo pm mod moa mod mm on on 05 mo om omH mHH mad om om om om om om oma mwa oma no: say mm om om om mv mv mma ONH mHH ousmmmum cofimsmumm *mma *nma *mma «mma gomfl «cma hmH «ma Hma mcwws oma omH omH omH omH bud mud oea mma onH ova 05H ona 05H ova ova pmH mma NmH cod «ma cod cod oma mma .FNH mmH ooa cod mod mmH mmfl ova omH mmH mud Am: 25v oma omH omH mmH mmH mmfl cod mmfl ova musmmmum cooflm mmH cma mma omfl mqa mMH mmfl ONH oma Hmfiumuua afiampmam om om ov on on OH 0 0H- om- “mopsafls. mafia ©0wumm coflmsmcH mum Houucoo mauwnm> Houucou Azoam ucmumcoov “oucv uwuooumamn can mmusmmmum Hmaaomm> .msmmHm wan smiwa mo :oHumnucmocou Camuonm .uuommcmuu :Hmuoum zmfi>a .30Hm nmfi>a co mmuscws om HOH abflamuom any oucfi A.¢.H .cwfi\mn 0.NV Hmwm Ummsmcfl madmoofl mo muommmm .HHHIm manna 77 «N0.0 «O0.0 «O0.0 «OO.O *OO.O OOO.O HO.O HO.O HO.O OOOOE H0.0 m0.0 OO.O OO.O OO.O HO.O HO.O H0.0 H0.0 H0.0 O0.0 O0.0 O0.0 OO.O ~0.0 H0.0 H0.0 H0.0 HO.O HO.O OO.O OO.O HO.O HO.O HO.O HO.O HO.O O0.0 O0.0 OO.O OO.O OO.O O0.0 H0.0 H0.0 H0.0 OO.O NO.O O0.0 ~0.0 OO.O OO.O ~O.O OO.O H0.0 HOHE OHxHaO OO.O OO.O OO.O OO.O OO.O OO.O ~O.O HO.O HO.O OOOO soHO OOEOO OOH «OH «OH OOH «OH OH HH HH HH menus OH OH OH OH OH OH O O O HH HH HH HH OH OH O O .O OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH HH HH HH HOO EEO OH OH OH OH OH OH OH OH OH Ousmmmum OH OH OH. OH HH OH OH OH OH OHO> HHmsO :me OO OO OO OO Om OH O OH- Om- HOOOOOHEO OEHO woflumm coflmsmcwldmwm Houucoo mHoH£w> Houucou HOOOOHucooO HHHnm OHnme 78 mv.o +mm.o +mo.o «Nw.o «¢h.o mv.o hm.o «N.0 om.o mammE om.o mm.o mN.H Om.m vH.H mm.o vm.o ¢m.o vm.o mN.o mh.o Hm.o om.o mm.o 00.0 «N.0 vm.o HN.O 0N.o mm.o wv.o ov.o mN.o vm.o mH.o mH.o hH.o mm.o vv.o 00.0 90.0 no.0 00.0 Hm.o Hm.o mH.o mm.o ov.o «v.0 ov.o Nb.o Nb.o mm.o mm.o mH.o Acflfi OH\OEV mo.H Hm.o vm.o om.o mo.H om.o om.o VH.O mH.o uuommcmufi afimuOHm smeaq «m.N +b.m «m.m «m.N +b.N +©.N N.N H.N 0.N mammfi o.m v.¢ w.v w.v m.m m.m v.m v.m v.m m.N m.m b.N m.N 0.N o.m ¢.N ¢.N H.N 0.N m.N m.N m.N m.N ¢.N m.H m.H h.H m.H N.N 0.N N.N H.N N.N H.N H.N m.H m.H 0.N N.N m.N v.N v.~ m.H w.H w.H Aw mfimhmv N.N h.N N.N m.m H.N m.H m.H v.H m.H cflmuonm Hmuoa nmfiwq ow om 06 on ON OH o CHI ON! Ammuncwfiv OBOE vofiuwm cofimamcH mwm HOHHCOU 0H0fl£m> Houucoo wascHuaoov HHHum OHnOa 79 .msau ouwN ou w>wumamu mo.o w m .meu oumn ou m>HumHmH H0.0 I m + v « O¢v «mv mm mammfi mv Nv 0v mv vv mm hm mm mm mm om mw mv wv vv mm Om mm panooumfimm 0.0 0.0 m.m mammfi m.o 0.9 v.m N.0 v.0 0.0 5.0 0.0 v.0 m.m m.m m.m m.m v.m m.m A» mamumv 00v 00¢ N. o¢ GHGHOHAH 33H“ 00 om ow on H ON 0H 0 can own Ammuacwev QEHB coHHmm :oflmsmcH mom Houucou mHUO£m> Houucou HOOOOHucoo. HHHnm OHnOa 80 OH OH OH OH OH OH OH OH OH HOHHO OHOOOOHO OOO OOO OOO OOO «OO .OO OHH OHH OOH OOOOE OO OO OO OO OO OO OHH OHH OOH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OO OO OO O OO OO OO OO OO OO OOH OOH OOH AOO EEO OO OO OO OO OO OO OOH OOH OOH OHOOOOHO OOHOOOHOO OHH HHH HHH OHH OHH HHH OH OH OH HOHOO OOOOOOHO HOOH +HOH +HOH +OOH +HOH HOOH OOH OOH OOH OOOOE OOH OOH OOH OHH OHH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OHH OHH OOH OO HO: EEO OOH OOH OOH OOH OOH OOH OOH OOH OOH OHOOOOHO cooHO OOH OOH OOH OHH OOH OOH OHH OHH OOH HOHHOHHO OHEOHOOO OO OO OO OO H OO OH O OH- OO- HOOHOOHEO maHa muOHH mm GOHmDMQH "mom HOHHCOU mHUfl£0> HOHuCOU Hzoam unnumcoov Honey ufluooumEmn can mmusmmwum umHsomO> .nEmmHm can SQE>H mo coflumuucmocoo awwuoum .uuommcmuu cflwuoum amaaa .30Hm smexa co mmuscfie om How nEOkuom wsu Gaza A.¢.H .cwfi\m: o.vv Hmwm comSMGO madmooa mo muommmm .>Hlm 0HQMB 80 OH OH OH OH OH OH OH OH OH HOHHO OHOOOOHO OOO OOO «OO OOO OOO OOO OHH OHH OOH OOOOE OO OO OO OO OO OO OHH OHH OOH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OO OO OO O OO OO OO OO OO OO OOH OOH OOH HOO EEO OO OO OO OO OO OO OOH OOH OOH OHOOOOHO :onOHHOO OHH HHH HHH OHH OHH HHH OH OH OH HOHHO OHOOOOHO +OOH +HOH +HOH +OOH +HOH OOOH OOH OOH OOH OOOOE OOH OOH OOH OHH OHH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OHH OHH OOH OO HOO EEO OOH OOH OOH OOH OOH OOH OOH OOH OOH OHOOOOHO nooHO OOH OOH OOH OHH OOH OOH OHH OHH OOH HOHHOHHO OHEOHOOO OO OO OO OO H OO OH O OH- OO- HOOHOOHEO OOHO 00H-H mm COflmHHM CH NOAH HOHUfiOU 0 Howfiwkw HOHHGOU, ABOHM unnumcoov Hmncv ufiuooumfimc can mwuammmum unasomm> .mEmmHm new zmexa mo coflumuucwocoo :Owuoum .uuommcmuu :Owuoum smaha .30Hm £QE>H co OOHDGOE 00 new nEHHmHom ms» oucfi H.¢.H .cflfi\m: o.vv Hmwm Ummsmcfl adamooH mo muommmm .>Hlm wanna 81 HO.OH HO.OH HO.OH OO.OH OO.OH HO.O OH OH OH HoHHO OHOOOOHO OO.O O0.0 O0.0 «O0.0 «O0.0 OO.O HO.O OO.O OO.O memos OO.O OO.O OO.O OO.O OO.O HO.O OO.O OO.O O0.0 OO.O O0.0 OO.O OO.O OO.O OO.O OO.O HO.O HO.O HO.O H0.0 HO.O HO.O HO.O HO.O H0.0 H0.0 H0.0 OO.O OO.O OO.O OO.O OO.O OO.O HO.O HO.O H0.0 OO.O O0.0 OO.O O0.0 O0.0 OO.O HO.O OO.O OO.O HOHs OHstO OO.O O0.0 O0.0 OH.O OH.O OO.O HO.O HO.O HO.O OHOO soHO :OsOH HH HH HH OH OH OH HH HH HH HOHHO OHOOOOHO OH OH OH OH OH HH HH HH HH mcnma O O O O O .O O O O OH OH OH OH O O O O O OH OH OH OH OH OH OH OH OH OH OH OH OH OH O O O O .OO EEO HH HH OH OH HH HH OH OH OH OHOOOOHO OH OH OH OH OH OH OH OH OH ch> HHOaO OHHO OO OO OO OO H OO OH O OH- OO- HOOHOOHEO OsHe 00.“me GOHmSH CH mum HOH UCOU 0H 0 H£0> HOHuGOU HOOOOHHOOOO >H-m OHnOs 82 HN.OH mN.OH Hm.OH mb.OH on.OH mv.0H No.0“ O0.0H v0.0“ Houum vwwocmum 05.0 v0...” ma .H «mm 2.” Hm .H Fwd Om .0 mm .0 mm .o 9.39: om.o mm.o ©m.0 vm.o 00.0 om.o Nm.o mv.o Nv.o m®.o mm.o mN.H Nm.H ON.H mv.o vm.o mH.o mH.o mm.o mm.o vm.o Hm.o mm.o Hm.o Hm.o am.o mm.o om.o mh.o om.o mm.o Nh.o 00.0 om.o 0N.o 0N.o ON.H mv.H mm.H om.o mm.o Nm.o om.o ov.o vv.o ACOE OH\mEV ON.H ON.N ON.N ov.m mN.m NH.m mm.o mm.o mm.o uuommcmua :wmuoum EQE>A N.“ m.“ N.“ M.“ v.« m.“ m.“ M.“ m.H Houum Ohmncwum «m.m «v.m «m.m ¢¢.m «o.m h.N v.m v.N v.m mammfi 0.N 0.N m.N N.N 0.N 0.N ©.H m.H v.H m.m m.m N.m m.m v.N win OHH min m...“ m.m m.m ¢.m H.m ~.m in H.m H.m m.m w.m m.m o.v H.v ®.m o.m 0.N 0.N 0.N H.m m.N m.m v.N m.m 0.N 0.N 0.N N.N Aw mEmumv m.v ¢.¢ v.v m.v v.v m.m m.m m.m m.m Gwououm HmuOB nmfihq om om O¢ Om H ON OH o OH! 0N! AmmuSCOEV mEHE cofluwm :onSMQH mom Houucou maofism> Houucou mescHucooO >H-m OHnOs 83 .wEHu oumn ou m>HumHmu mo.o I m v + .mEHu Chou on m>HunHmu HO.O w m .1 NH HH NH Houum oumcamum «NO «HO Om mamas OO OO MO 0O mO mm HO NO HO mO MO mm hm hm om Om Om mm uHuooumEmm O.O m.H m.H Mouuo vumwcwum 0.0 H.w 0.0 mauve 0.0 0.0 0.0 0.0 O.m o.m m.w N.0 0.0 m.m m.m ~.m O.O O.O 0.0 HO OEOHO. ®.m H.o N.0 cwmuoum mammHm om om CO on H om 0H 0 oH omI AmmusaHsv «EH9 voHuwm :onsm:H mom Houucoo 0H0H2w> Houucoo HOOsOHHcooO >H-m mHnOe 84 OH OH OH OH OH OH HH HH HoHHm OHOOOOHO «OH «mH OmH «wH «wH «0H m CH mammfi OH OH OH OH OH OH O O OO OO OO OO OO OO OH OH HH OH HH HH HH OH O O OH OH OH OH OH OH O O HOO EEO OH OH OH OH OH OH OH OH OHOOOOHO OH OH OH OH OH OH O O OHO> HHOEO cHxO OH OH OH OH OH OH OH OH HOHHO OHOOOOHO OOH OOH OOH OOH OOH OOH OOH HOH OOOOE OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OHH OHH OHH OOH OOH OOH OOH OOH OOH OOH OOH HO: EEO OO OO OO OOH OOH OHH OHH OHH «HOOOOHO cooHO OOH OOH OOH OOH OOH OOH OOH OHH HOHHOHHO OHEOHOOO OO OO OO OO H OO OH O OH- OO- HOOHOOHEO OsHa GOMHGQ COHmDH CH mg HOHHCOU ”HUflSwNV HOHHGOU Honm H3335 3&3 uHHooumfimn can ”£9.63 QOHHm-Hom .mmHHHmmw-Hm HOHHHomm> .wEmOHm can :95: mo coHumHucmocoo :Hmuoum .uuommcmuu chuoum :95: .onm OOEOH co OOHOOHE OO HOH nsHHOHOO OOH ouOH H.O.H .OHE\O: 0.0HO HOOO OOOOOOH OHHOooH mo OHOOOOO .>-O OHOOO 85 O.H O.H O.H O.H O.H O.H O.H O.H HOHHO OHOOOOHO «0.0 OO.O OO.O OO.O 0.0 H.O 0.0 O.H 233 0.0 0.0 0.0 0.0 0.0 0.0 H.O H.O O.H H.H O.H O.H O.H O.H O.H O.H H.O 0.0 0.0 0.0 O.H O.H O.H O.H 0.0 H.O 0.0 0.0 O.H O.H O.H O.H 0.0 0.0 0.0 0.0 0.0 0.0 H.O H.O HO OEOHOO 0.0 0.0 H.O 0.0 0.0 0.0 0.0 0.0 :HOHoHO Hmuoa cmemq OO.H OO.H OO.H OO.H OO.H HO.H HO.H HO.H HOHHO_OHOOOOHO O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 Ocmws O0.0 O0.0 O0.0 O0.0 O0.0 H0.0 H0.0 H0.0 O0.0 O0.0 OH.O O0.0 O0.0 OH.O O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 H0.0 H0.0 H0.0 O0.0 O0.0 O0.0 O0.0 OH.O O0.0 H0.0 H0.0 HOHE OH\Hs. O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 H0.0 H0.0 OHOO onO nae»; OO OO OO OO H OO OH O OH- OO- HOOHOOHSO OEHO OOHHOO :onsmcH mom HoHHcoo OHOHOO> Houucoo HOOOOHHOooO >-< OHQOO 86 H Mouuw cumocmum .m wcwms Aw mfimumv :Hmuoum mammHm OOH OO.H OO.H OO.H OO.H HH.H OO.H HOHHO OHOOOOHO +OO.H +OO.H +OO.H «HO.H OO.O OO.O O0.0 OOOOE OO.H OO.H OO.H OO.O OO.H HO.O HO.O HcHs OH\OaO OO.H O0.0 OO.H OO.H O0.0 O0.0 O0.0 Huommcmua :HOHOHO :OEOO om 0O om om 0H 0 OHI omI Amwuacwfiv mEHB UoHuwm consmcH Hmom Houucoo 0HOHnm> Houucoo HOOOOHHOOOO >-O OHOOO 87 .OEHH oumu ou m>HumHmH OO.O w m .OEHH ouwn cu m>HumHmu H0.0 I a # + V mNH OOH ONH Houum cumccmum .Em OOO MOO mzwma OO mOO HNO OOH MNm OOO mH mOO oMO OH OOO OOO HO»: NOOO-HOV OO HHO OOO EOE: QOHHOHOO wocmummmHo HmucwEHummmw Houucou NH NH NH Houum cumwcwum OOO OOO 0O mauve OM OM OM mO 0m OO om mO mO mO OO NO 0O MO Om wO mO CO HHHooumEmm ow om 0O 0M ON 0H 0 CHI ONI AmmuncHEv QEHB ooHumm conswcH Hmwm Honucou wHoHnm> Houucou Acmscwucoov >I¢ mHnt 88 HHH OHH OH OH OH OH OH OH OH HOHHO OHOOOOHO OOO OOO OOO OOO OOO OOO OOH OOH OOH OOOOE OO OO OO OO OO OO OHH OHH OHH OO OO OO OO OO OO OO OO OO OOH OOH OO OO OO OO OOH OOH OHH OO OO OO OO OO OO OO OO OO OO OO OO OO OO OO OOH OO OO HOO EEO OO OO OO OO OO OO OHH OOH OOH OHOOOOHO OOHOOOHOO OHH OHH OHH OHH HHH HHH OH OH OH HOHHO OHOOOOHO HOH HOH HOH OHH OHH OHH OHH OHH OHH Ocmms OOH OOH OOH OOH OOH OOH OOH OOH OOH OO OO OO OO OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OO OO OO OO OO OO OO OO OO HOO EEO OOH OOH OOH OOH OOH OO OOH OOH OOH OHOOOOHO OooHO OOH OOH OOH OOH OOH OOH OOH OOH OOH HOHHOHHO oHaOHOOO OO OO OO OO OO OH O OH- OO- HOOHOOHEO OEHO , UOHuwm GOHWDMCH HWQOH HOHHGCU wHOflSw> HOHuGOU H303 Hcmuwcoov Gus HHHUOHOEOS HEM H2085 naHHmuow .mmusmmwum HOHsomm> .mEmmHm van. 5953 m0 coHumHucmucoo :Hmuoum .Huommcmuu chHoum :95; .303 £953.. :0 $5358 om How HHEHHmHom may OHCH A.<.H .cHE\mO: o .OHV Hmum owmdmcH OHHHmooH mo Bommmm .>Im OHS-O.H. 89 OO.H HO.H OO.H HO.H OO.H OO.H OO.H OO.H OO.H HOHHO OHOOOOHO «OO.O +O0.0 «O0.0 «OO.O «O0.0 .O0.0 HO.O HO.O H0.0 mauve OO.O O0.0 HH.O OH.O OH.O OH.O HO.O H0.0 OO.O OO.O OO.O O0.0 OO.O OH.O OO.O OO.O OO.O HO.O HO.O HO.O HO.O OO.O HO.O HO.O H0.0 HO.O HO.O OO.O OO.O OO.O OO.O OO.O OO.O H0.0 H0.0 H0.0 HO.O H0.0 HO.O OO.O HO.O H0.0 HO.O HO.O H0.0 .OHs OH\HEO O0.0 OO.O OO.O OO.O OO.O OO.O HO.O HO.O OO.O OHOO soHO :OEOA HH HH HH HH HH HH HH HH HH HOHHO OHOOOOHO OH OH OH OH OH OH OH OH OH wanes OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH HH OH OH OH OH HH HH HH HOO EEO O O O O O O O O O OHOOOOHO OH OH OH OH OH OH OH HH OH ch> HHOsO :me OO OO OO OO H OO OH O OH- OO- HOOHOOHEO OEHO UOHme CO.“ mDMCH mum HO.HHCOU ”HUEmNV HO.H-thou “wwSCHucoov >Im anwB 90 OO.H OO.H OO.H OO.H OO.H OO.H OO.H OO.H OO.H HOHHO OHOOOOHO «OO.H «OO.H «OH.O .OO.O «OO.H OO.O OO.O O0.0 OO.O OOOOE OO.O OO.H OO.O O0.0 OO.O OO.H OH.O OH.O O0.0 OO.O OO.O O0.0 OO.O O0.0 OO.H O0.0 OO.O O0.0 H0.0 OO.O H0.0 OO.O OO.O OO.O O0.0 O0.0 H0.0 O0.0 OO.O HO.O O0.0 O0.0 OO.O O0.0 O0.0 O0.0 OO.O O0.0 OO.O OO.H OO.O OO.O OH.O OH.O HO.O HOHE OHxOsO O0.0 O0.0 OO.H OH.H OO.O O0.0 OH.O OH.O OO.O HHOOOOOHO OHOHOHO OOEOH O.H O.H O.H O.H O.H O.H H.H H.H O.H HOHHO OHOOOOHO «0.0 «0.0 OO.O OO.O OH.O 0.0 O.H O.H O.O OOOOE H.O 0.0 0.0 0.0 0.0 O.H O.H O.H O.H 0.0 0.0 H.O 0.0 O.O H.O O.O O.O 0.0 H.O 0.0 H.O 0.0 O.O O.O O.O 0.0 H.O 0.0 0.0 0.0 0.0 O.O O.O 0.0 O.O O.O O.O 0.0 0.0 0.0 O.O O.O O.H O.H H.O HO OEOHOO H.O O.O O.O O.O O.O O.H O.H O.H O.H OHOHOHO HOHoa OOEOH OO OO OO OO H OO OH O OH- OO- HOOHOOHEO «EH9 OOHHOO :oHOuHOH OOO HOHHOOO OHOHOO> HOHHOOO awn: Hucoov >Im mH HO.H. 91 .OEHH cums on m>HuOH0H O0.0 w m .mEHH cums on m>HumHmH HO.O w m + O NH NH HH Houum Oumccmum OOO «MO Om mcmwa MO NO OM OO OO OO mO OO Om mm mm Om HO OO Om Nm OO OO HHHooumfimm N.H H.H N.H Houum Oumccmum 0.0 0.0 0.0 mcmwe H.O 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 H.O O.O 0.0 HO OEOHOO 0.0 H.O 0.0 :kuoum MEOOHO OO Om OO Om H ON OH O OHI ONI AmmuacHfiv wEHB ooHumm OOHmsmcH mwm Houucou mHonw> Houucoo AOOOOHHOOOO >-O «Hams 92 OH OH OH OH OH OH HH HH HOHHO OHOOOOHO {ma «ON «OH «mH «ON «ON NH NH mwaE OO OO OO OO OO OO OH OH OH OH OH OH OH OO OH OH OH OH OH OH OH HO OH OH OH OH OH OH OH OH OH HH HO: EEO OH OH OH OH OH OH O O OHOOOOHO OH OH OH OH OH OH OH OH ch> HHOsO cHxO OH OH OH OH OH OH OH OH HOHHO OHOOOOHO «OOH «OOH «OOH «OOH «OOH «OOH OOH HOH OOOOE OOH OOH OOH OOH OOH OOH OOH OOH OHH OOH OOH OOH OOH OOH OOH OOH OO OO OO OO OO OO OOH OOH OOH OOH OOH OOH OOH OOH OOH OHH HO: EEO OO OO OO OO OO OO OOH OOH OHOOOOHO cooHO OOH OHH OHH OHH OHH OHH OOH OOH HOHHOHHO OHEOHOOO OO OO OO OO 4 OO OH O OH- OO- HOOHOOHEO OEHO 00.?me COHmDM EH Omwm HOHHCOU mHUfifiwxw HO.H-«COO Aonm HOHOHOGV HOHCO HHuooumEms OOO HSOHOB QEHHOHOO .wmusmmmum HOHsomw> .mEmmHm cam SOEOH mo coHumuchocoo chuoum .Huommcmuu chuoum SQEOH .3OHm OOEOH Oo OOHOOHE OO H0O OEHHOHOO OOH OHOH H.O.H .OH5\OO O.OOO HOOO OOOOOOH OHHOooH Ho OHOOOOO .H>I< mHan 93 OH.O OH.O OH.O HN.O no.0 MN.O OH.O ON.O OH.O OH.O m0.0 NN.O OH.O Hounm Oumncmum memes Aw mEOHOO OH 38E H38. :93 Houuw Oumncwum OO Om OH OOHumm :onamcH MOO Houucou wHoH£m> madmfi ACHE OH\HEO mumm onh £QE>H ONI Ammuacflfiv mEHB Houucoo Avmanucoov H>I< mHnma 94 N.H N.H N.H Momma Oumwcwum «0.0 Om.O m.m mammfi N.0 0.0 N.m H.O N.0 O.m 0.0 m.O 0.0 0.0 0.0 0.0 H.O H.O O.O HO OEOHOO H.O N.O 0.0 :Hmuoum mEmmHm OH.HH NH.HH OH.HH HN.HH mm.HH HO.H mO.H HH.H HOHHQ UHOOCOHO OOH.m «OH.m «OO.m «NN.m amm.m OO.N Om.O mm.O mammfi OO.N OO.m ON.m mm.N OO.N ON.N Om.O ON.O O0.0 OH.m OO.m O0.0 Nm.N OO.H Om.O OH.O O0.0 O0.0 Om.m O0.0 OH.N mm.O Om.O O0.0 OM.O mw.m Om.O Hm.m mv.m O0.0 ON.O ON.O O0.0 OO.H O0.0 m0.0 mO.H OM.H HN.O HN.O ASHE OH\OEV Om.O Nm.O Hm.m N0.0 Om.OH mm.m O0.0 O0.0 HHOQOGMHB GHmuOHm £QE>H OO Om OO OM H ON OH O OHI ONI HOOUSCHEV wEHB 00.."me COH WDMCH WU“ HO.HUCOU wHUMSQNV HOHUCOU HOOOOHHOooO H>-< OHOOO 95 .wEHu oumu 0H w>HHmHmH OO.O w m + .mEHH ouwu ou m>HuunH HO.O w m .4. OH OMH OmH Houuw Oumvcmum OOO OOO OOO mcmwfi OO HOO mmO Om OOO HMO ON OHO MOO NO mOm Mmm HO HOO OOO HO": «mamumv OO OOO OOO HOOHOS nsHHOHOO wocmummmHO HmucmEHHmmwm HOHHCOU NH HH NH Houum Oumwcmum «mO +NO OM mcwms OO OO Nm mO OO OO OO OO Om OO OO Om OO OO NO mm OM OM HHHUOHMEOO OO Om OO Om H ON OH O OHI .ONI AmmuncHfiv wEHB OOHme consmcH mum Houucoo wHUHnm> Houucou HOOOOHHOooO H>IO OHOOO 96 O+ OH OH OH OH OH OH OH OH HOHHO OHOOOOHO «OO «HO «HO OOO OOO OOO OOH OOH OOH OOOOE OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OO OOH OHH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OO OO OO HOO EEO OO OO OO OO OO OO OOH OO OO OHOOOOHO coHOOOHOO OH OH OH OH OH OH OHH OH OH OH HOHHO OHOOOOHO OHH HHH OOH OOH OOH OOH OOO OHH OHH OHH OOOOE OHH OHH OHH OHH OOH OOH OO OOH OOH OOH OO OO OO OO OO OO OO OOH OOH OOH OOH OOH OOH OHH OHH OHH OO OHH OHH OHH OOH OOH OOH OOH OOH OOH OOH OHH OHH OHH HOO EEO OOH OHH OOH OOH OOH OOH OO OO OO OO OHOOOOHO OOOHO OO OO OO OO OO OO OO OOH OOH OOH HOHHOHHO OHEOHOOO OO OO OO OO OO OH O O OH- OO- HOOHOOHEO msHa COHHwnH COH mauve—H Ammom HOHHCOU wHUfi£®> HOHHCOU 2;on Hcmumcoov HOnnv HHHooumEmn OOO ucmHm3 nEHHmHom .mwusmmmum Hanomm> .MEmmHm can :mEOH mo :oHumHucmocoo chuoum .Huommcmuu :Hmuoum LQEOH .3OHH LQEOH :0 mmuscHfi OO H0O QEHHwHom mp» oucH H.O.H .CHE\O: O.va Hmom OmmnmcH OHHmooH mo muommmm .H>Im OHQMB 97 HO.H HO.H HO.H OO.H OO.H HO.H HO.H HO.H HO.H HOHHO OHOOOOHO OO.O OO.O OO.O +OO.O OO.O OO.O OO.O OO.O OO.O OOOOE OO.O OO.O OO.O HO.O HO.O HO.O HO.O OO.O HO.O HO.O HO.O OO.O OO.O OO.O HO.O OO.O HO.O H0.0 OO.O OO.O OO.O OO.O OO.O HO.O H0.0 HO.O H0.0 HO.O HO.O HO.O HO.O HO.O HO.O H0.0 HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HOHE OH\HeO OO.O OH.O OH.O OH.O OH.O OO.O OO.O OO.O OO.O OHOO soHO OOOOO HH HH HH HH HH HH HH HH HH HOHHO OHOOOOHO OH OH OH OH OH OH OH OH OH mamas O O O O O O OH O O OH OH OH HH HH HH OH HH HH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH HOO EEO OH OH OH OH OH OH OH OH OH OHOOOOHO OH OH OH OH OH HH OH OH OH OHO> HHOsO :me OO OO OO OO OO OH O OH- OO- HOOHOOHEO OEHO UOHHOQ COHODMCH HNOOH HOHUCOU mHUfl£w> HOHHCOU HOOOOHHOOOO H>-m mHnOe 97 HO.H HO.H HO.H OO.H OO.H HO.H HO.H HO.H HO.H HoHHm OHOOOOHO OO.O OO.O OO.O +OO.O OO.O OO.O OO.O OO.O OO.O menus OO.O OO.O OO.O HO.O HO.O HO.O HO.O OO.O HO.O HO.O HO.O OO.O OO.O OO.O HO.O OO.O HO.O HO.O OO.O OO.O OO.O OO.O OO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HO.O HcHe OHxHeO OO.O OH.O OH.O OH.O OH.O OO.O OO.O OO.O OO.O OHOO soHO OOOOH HH HH HH HH HH HH HH HH HH HOHHO OHOOOOHO OH OH OH OH OH OH OH OH OH Ocmma O O O O O O OH O O OH OH OH HH HH HH OH HH HH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH HO: EEO OH OH OH OH OH OH OH OH OH OHOOOOHO OH OH OH OH OH HH OH OH OH ch> HHOsO :HxO OO OO OO OO OO OH O OH- OO- HOOHOOHEO OEHO UOHHmnH COHwSMGHJfimUnH HOHHCOU 0HOM5w> HOHHCOU HUOSGHHCOUV H>Im anmB 98 OH.H OO.H OO.H OO.H OO.H OH.H OO.H OH.H OH.H HOHHO OHOOOOHO O0.0 +H0.0 +O0.0 «O0.0 +O0.0 O0.0 O0.0 O0.0 O0.0 memes O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 H0.0 H0.0 O0.0 H0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 H0.0 O0.0 O0.0 H0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 O0.0 HOHE OH\Os. OO.H O0.0 OH.O O0.0 O0.0 OO.H O0.0 O0.0 O0.0 HHOOOOOHO OHOHOHO :OEOO O.H O.H O.H O.H O.H O.H O.H O.H O.H HOHHO OHOOOOHO «0.0 OO.O +0.0 «0.0 H0.0 O0.0 0.0 0.0 0.0 OOOOE 0.0 0.0 H.O 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 H.O H.O 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 H.O 0.0 0.0 H.O 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 HO OEOHOO H.O 0.0 H.O O.H O.H 0.0 O.H O.H O.H chHoHO HOHoO :OEOO OO OO OO OO OO OH O OH- OO- HOOHOOHEO OEHO ooHHOO :onsucH HOOO HoHHcoo OHOHOO> Houucoo HOOOOHHOOOO H>-m OHOOO 99 NM NM HH HO.H.HG @Hmmucmum «OO «OO OO OOOOE OO OO OO OO OO OO OO OO OO OO OO OO OO OO OO OO HO OO HHHOOHOEOO O.H O.H O.H HOHHO OHOOOOHO OO.O «O.O O.O Ocmma O.O O.O O.O O.O O.O O.O O.O O.O O.O O.O O.O O.O O.O 0.0 O.O HO OEOHOO O.O O.O O.O OHOHOHO OsOOHO OO OO OO OO OO OH O OH- OO- HOOHOOHEO OeHe muOflHmm COHmDMr—HJHIWUOH HOHUCOU wHUH£w> HOHUCOU HOOOOHHGOUO H>Im OHQMB 100 .wEHH ouwu OH w>HumeH O0.0 V m + .wEHu cums ou 0>HHmHmH HO.O w m .1 HHH +ON N OO mH MO N OH wocmHmHmHO OOH HOO NmO OOO OHO OONH OOO OOO kucmeHummxm OOH HOHHO Oumvcmum OOO mcmwe mmO NOO OOO ONHH NOO HO": Omfimumv OOO uanmz nEHHmHom Houucoo HOOSGHHCOUV H>Im mHnma 101 HH HH HH OH OH OH HH HH HH HOHHO OHOOOOOO «OO OOO «OO «OO «OO «OO OH OH OH mamma OO OO OO OH OH OH HH OH OH OO OO OO OO OO OO OH O OH OO OO OO OO OO OO OH OH OH OO OO OO OO OO OO OH OH OH HOO EEO OO OO OO OO OO OO OH OH OH OHOOOOHO OO OO OO OO OO OO OH OH OH OHO> HHOeO OHOO OH OH OH OH HHH OHH OH OH. OH HOHHO OHOOOOHO OHH OHH OHH OHH OHH HOH OHH OOH OHH Ocmms OO OO OO OO OO OO OO OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OHH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH .OO EEO OHH OOH OOH OOH OOH OOH OOH OHH OOH OHOOOOHO OOOHO OOH OOH OOH OOH OOH OOH OHH OHH OOH HOHHOHHO OHEOHOOO OO OO OO OO OO OH O OH- OO- HOOHOOHEO «EH9 J1 OOHHmm conswcH rmwm Houucoo OHOHnw> coHHomth .OOOH HonO HOHOHOOO .OnOO HHHuoumen can uanm3 QEHHOHOO .mmusmmwum HOHaomm> .mEmmHm can SQEOH mo coHumHucmocoo :Hmuoum .Huommcmuu chHoum SOEOH .3OHO smEOH co mwuscHE OO How QEHHmHom map OHGH OHHOOOH OOmOHCH H.O.H .OH2\OO 0.0HO HOOO On OmgoHHoH OHOOOOO>OHHOH OOHOOOOH HOH\OE OO cHomnHmaoOOH Ho OHOOHHO .HH>-O OHOOO 102 H OOHHmm conamcH mum O.H O.H O.H O.H O.H O.H O.H O.H O.H HOHHO OHOOOOHO «O.O OO.O «O.O «H.O «0.0 HO.O O.O H.O 0.0 menus O.O O.O O.O O.O O.O O.H O.H O.H O.H O.O O.O O.O O.O O.O H.O 0.0 O.O O.O O.O O.O O.O O.O O.O O.H O.H O.H O.H O.O O.O O.O O.O O.O 0.0 H.O O.H O.H O.O 0.0 0.0 0.0 0.0 H.O 0.0 0.0 H.O HO OsOHOO O.O O.O O.O O.O O.O 0.0 H.O H.O O.H OHOHOHO HOHOO OOEOH OO.H OO.H OO.H OO.H OO.H OO.H HO.H HO.H HO.H HOHHO OHOOOOHO OOOO.O OOOO.O OOOO.O OOOO.O O«OO.O O.OO.O OO.O OO.O OO.O mamma OO.O OO.O OO.O OO.O HO.O OO.O OO.O OO.O OO.O O0.0 O0.0 OO.O O0.0 H0.0 HH.O OO.O HO.O HO.O OO.O O0.0 OO.O OO.O OO.O OO.O OO.O O0.0 OO.O OO.O OO.O OO.O O0.0 OO.O OO.O OO.O OO.O OO.O OH.O OH.O OH.O OH.O OH.O HH.O OO.O OO.O HO.O HOHE OH\HaO OH.O OH.O OH.O OO.O OO.O HO.O OO.O HO.O OO.O OHOO onO OOEOH OO OO OO OO OO OH O OH- OO- HOOHOOHEO «EH9 Houucoo wH0H£m> :oHuumflcH .OOCH HowscHucooO HH>-O OHOOO 103 O.H O.H O.H O.H HOHHO OHOOOOHO O.O O.O O.O 0.0 mamas O.O O.O O.O O.O O.O O.O O.O O.O 0.0 O.O O.O O.O O.O O.O O.O O.O O.O O.O O.O 0.0 HO OEOHOO O.O O.O O.O O.O OHOHOHO OsmOHO OO.HH OO.HH OO.HH OO.HH OO.HH OO.H OH.H OH.H OH.H HOHHO OHOOOOHO O.OO.O O«OO.O OOOO.O O«OO.O OOOO.O OOOO.O OO.O OO.O OO.O Ocmma OH.O OO.O O0.0 O0.0 O0.0 OO.O O0.0 OO.H OO.O O0.0H O0.0H OO.HH O0.0H O0.0H H0.0 OO.O O0.0 O0.0 OO.O O0.0 O0.0 OO.O O0.0 OO.O OO.O H0.0 O0.0 HO.O O0.0H OO.OH O0.0H OO.OH OO.O OO.H OO.H OH.H OO.O O0.0 OH.O OO.O O0.0 H0.0 O0.0 O0.0 HO.O HcHe OHan. OO.O O0.0 O0.0 OO.O OO.O OO.O O0.0 HO.O OO.O HHOOOOOHO :HmHoHO OOEOO OO OO OO OO H OO OH O OH- OO- HOOHOOHEO OEHO OoHumm :onsmcH mom Houucou OHUH£w> :oHHowficH .occH HOOOOHucooO HH>IO mHnOa .H> anva mcon CHE\Hmom O O.OH £HH3 OwcHOHnO mmsHO> may on w>HumHmH HO.O w m O .mEHH cums on m>HumHmH OO.O w m+ .mEHu OHON 0H m>HHmHmH H0.0 w m k. OOO OOO mmO mammE OO MOO OOO Om ONO mOO Hm OOO OHO MO NOO OOO OO OOO OOO AOuO uOOOHOO Om NOO OOO HLOHOB nEHHmHom wocwHwOOHO HmucmEHHmmmm HOHHGOU 4. nu 1 OH OH OH OH HoHHm OHOOOOHO «mm «OM OM mm 9308 OO Om OO Om OO OO mm mm mO OO Om mm mO OO mm mm ON ON NN NN Nm mm ON ON HHHooumamm OO Om OO Om ON OH O OHI OOI HOOHSCHEV wEHB H OOHme conDMcH mom HOHHGOO wHUH£m> COHHomth .OOCH HcmscHucooO HH>-< mHnOa 105 O O O O O m m O O HOHHO Oumwcmum «HO «OO «OO «OO «OO «OO OOH HOH OOH mcmws OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OOH OOH OHH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OOH OOH OOH OO OO OO OO OO OO OOH OOH OOH AOO EEO OO OO OO OO OO OO OOH OOH OOH OHOOOOHO :onsHHOO OH OH OH OH OH OH OH OH OH HOHHO OHOOOOHO OOOH OOOH +OOH «OOH «OOH +OOH OOH OOH OOH OOOOE OOH OOH OOH OOH OOH OOH OHH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OHH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH OOH HO: EEO OOH OOH OOH OOH OOH OOH OOH OOH OOH OHOOOOHO cooHO OOH OOH OOH OOH OOH OHH OOH OOH OOH HOHHOHHO OHEOHOOO OO OO OO OO OO OH O OH- OO- HOOHOOHEO OsHe H OOHme GOHmecH mwm Houucou 0H0H5m> :oHuochH .OOGH HHHooumEms OOO uanwz QEHHOHOO .Huomwcmuu :kuoum LQEOH A3OHH Hcmumcoov HOucv .mmnsmmmum.HmHsomm> .OEmOHm OOO SQEOH mo :oHHmHucwocoo OHwHOHm .BOHM :OEOH :0 mmuscHE OO How nEHkuow may. oucH OHHOOOH OOOSOOH H.O.H «CHE\O: 0.0HV Hmom On OmonHOO aHmaocw>muucH OwuomncH HOx\OE my :HownquOOcH mo muommmm .HH>Im mHnma 106 OOHme consmcH mum OO.H HO.H OO.H OO.H OO.H OO.H OH OH OH HOHHO OHOOOOHO 3+OO.O 3+OO.O «OO.O «OH.O 3«OH.O «OO.O OO.O OO.O HO.O mcmwa OO.O OO.O OO.O OO.O OO.O OO.O OO.O HO.O OO.O OO.O OO.O OO.O OO.O OO.O OO.O HO.O HO.O HO.O OO.O OO.O OO.O OO.O OH.O OO.O OO.O OO.O HO.O OO.O OO.O OO.O OO.O OO.O OO.O OO.O OO.O OO.O OO.O HH.O OH.O OO.O OO.O OH.O HO.O HO.O HO.O HcHs OH\HsO OH.O OH.O OO.O OH.O HH.O OH.O HO.O HO.O HO.O OHOO soHO OOEOH OH OH OH OH OH OH HH HH HH HOHHO OHOOOOHO HH HH HH HH HH HH OH OH HH memos OH OH OH OH OH OH OH OH OH O O O O O O O O O O O O O O O O O O OH OH OH OH HH HH OH OH OH 1O: EEO OH OH OH OH HH HH OH OH HH OHOOOOHO OH OH OH OH OH OH OH OH OH ch> HHOsO :me OO OO OO OO H OO OH O OH- OO- HOOHOOHEO OEHO Houucou mH0H5m> coHHUWHOH .occH 828382 HH>-m OHHOO. 107 OO.H OO.H OO.HH HO.HH OO.OH HO.HH OO.H OO.H OO.H HOHHO OHOOOOHO +H0.0 3+OO.O 3«OO.O 3«OO.O O«O0.0 aO0.0 OO.O O0.0 OO.O Ocmms OH.O OO.O O0.0 HO.O OO.H O0.0 OO.O OO.O OO.O OO.O O0.0 O0.0 OH.O O0.0 O0.0 O0.0 O0.0 O0.0 OO.H OO.H OO.H OO.O O0.0 OO.H O0.0 O0.0 O0.0 OO.H OO.H OO.H OO.O OO.O OO.H OO.O OO.O O0.0 O0.0 O0.0 O0.0H OO.HH O0.0H O0.0 O0.0 O0.0 H0.0 HOHs OH\OsO OO.O OO.O OO.O OO.O OO.O OO.O OO.O OO.O OH.O HHomchHa OHOHoHO :OEOH O.H O.H O.H O.H O.H O.H O.H O.H O.H HOHHO OHOOOOHO «0.0 «0.0 «0.0 OO.O «O.O 0.0 O.O O.O O.O mamms O.O 0.0 O.O O.O O.O O.O O.O O.O O.O 0.0 O.O 0.0 O.O 0.0 0.0 0.0 0.0 0.0 O.O O.O O.O O.O 0.0 O.O O.O O.O O.O O.O O.O O.O 0.0 O.O O.O O.O O.O H.O 0.0 0.0 0.0 O.O 0.0 H.O O.O O.O H.O HO OEOHOO O.O O.O H.O O.O O.O O.O O.O 0.0 O.H OHOHOHO Hmuoe OOEOH OO OO OO OO H OO OH O OH- OO- HOOHOOHEO OEHO OOHHOO :oHOsOOH OOO HOHHOoo OHOHOO> :oHHomncH .oocH HOOOOHHOOOO HH>-O OHOOO 108 mH mH NH NH HOHHO Oumwcmum «mO «MO Om Om mcwma OO OO OO HO OO OO OO OO OO OO HO HO OO OO OO. OO OO OO OO OO OO OO OO OO HHHooHOsOO N. H N.H N .H N.H HO.H-H0 myHMmquum 0.0 0.0 0.0 0.0 Ocmws O.O 0.0 0.0 0.0 O.O 0.0 O.O 0.0 O.O H.O O.O 0.0 0.0 O.O 0.0 O.O O.O 0.0 0.0 O.O HO OEOHO. N.m N.m 0.0 0.0 chuoum OEmem OO OO OO OO H OO OH O OH- OO- HOOHOOHEO maHa coHHOO OOHOOHOH OOO Houucou OHOHOO> :oHHownnH .occH HOOOOHHOoOO HH>Im OHOOO .H> OHQOBV wcon CHE\HmOm m: O.OH £HH3 OwcHano mmsHu> may 0» 0>HumHmH OO.OMHQO .H> mHnOBO mcoHO cHE\Hmwm O: 0.0H £HH3 OOOHOHQO mwsHo> map on m>HumeH HO.OMHOO 109 .mEHH oumu 0» m>HHOHmH O0.0 w m+ .mEHu OHON 0H 0>HHOHOH HO.O w m .1 OH mmH OOH Houum Oumccmum +OH OOO OOO mcwms N OOO MOO O mOm Omm m NOO OOO MH HOm OOO HO OOO OOO HOH: OmEMHmO MH ONO OOO uanmz AEHHOHOO mocwuwmmHO HmucmEHHmmwm Houucoo HumscHucoov HH>-m OHOOO BIBLIOGRAPHY 10. BIBLIOGRAPHY Abdel-Halim, M. S., M. Hamberg, B. Sjoquist, and E. Knggard: Identification of prostaglandin D2 as a major prostaglandin in homogenates of rat brain. Prostaglandins 14:633-643, 1977. Ali, M., A. L. Cerskus, J. Zamecnik, and J. W. D. McDonald: Synthesis of prostaglandin D2 and thromboxane 32 by human platelets. Thromb. Res. 11:485-496, 1977. Knggard, E. and C. E. Jonsson: Efflux of prostaglandins in lymph from scalded tissue. Acta Physiol. Scand. 81:440- 447, 1971. Arturson, G., F. Groth, and G. Grotte: The functional ultra- structure of the blood-lymph barrier. Computer analysis of data from dog heart—lymph experiments using theoretical models. Acta Physiol. Scand. Suppl. 374:1-30, 1972. Arturson, G., and C; Johnson: Effects of indomethacin on the transcapillary leakage of macromolecules and the efflux of prostaglandin in paw lymph following experimental scalding injury. Uppsala JOurnal of Medical Sciences 78:181-188, 1973. Baez, 8.: Microvascular terminology. In: Microcirculation, Vol.-I, edited by G. Kaley and B. M. Altura, Univ. Park Press, Baltimore, pp. 23-34, 1977. Bauer, J. D., P. G. Ackerman, and G. Toro: Clinical Laboratory Mbthods (8th ed.). C. V. Mosby Co., St. Louis, pp. 402- 403, 1974. Bergstrom, S., L. A. Carlson, and L. Oro: Effects of prostaglandins on catecholamine induced changes in the free fatty acids of plasma and in blood pressure in the dog. Acta Physiol. Scand. 60:170-178, 1964. Bergstrbm, S., R. Ryhage, B. Samuelsson, and J. Sjavall: The structures of prostaglandin E1, Fla! and F13. J. Biol. Chem. 238:3555-3564, 1963. Bergstrom, S. and J. Sjovall: The isolation of prostaglandin F from sheep prostate glands. Acta Chem. Scand. 14:1693-1700, 1960. 110 11. 12. 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 111 Bergstrbm, S. and J. Sjovall: The isolation of prostaglandin E from sheep prostate glands. Acta Chem. Scand. 14:1701- 1705, 1960. Berne, R. M. and M. N. Levy: Cardiovascular Physiology. The C. V. Mosby Co., St. Louis, p. 121, 1977. Bito, L. 2.: Accumulation and apparent active transport of prostaglandins by some rabbit tissues in vitro. J. Physiol. (London) 221:371—387, 1972. Bito, L. Z. and R. Baroody: Inhibition of pulmonary prostaglandin metabolism by inhibitors of prostaglandin biotransport (probenecid and bromcresol green). Prostaglandins 10:633- 638, 1975. Bito, L. Z. and E. V. Salvador: Effects of anti-inflammatory agents and some other drugs on prostaglandin biotransport. J. Pharm. Exp. Ther. 198:481-488, 1976.. Bito, L. Z., M. Wallenstein, and R. Baroody: The role of trans- port processes in the distribution and disposition of prostaglandins. Adv. Prost. Throm. Res. 1:297-303, 1976. Bruns, R. R. and G. E. Palade: Studies on blood capillaries. I. General organization of blood capillaries in muscle. J. Cell Biol. 37:244-276, 1968. Carlson, L. A., L. G. Ekelund, and L. Oro: Circulatory and respiratory effects of different doses of PGE1 in man. Acta Physiol. Scand. 75:161—167, 1969. Carter, R. D., W. L. Joyner, and E. M. Renkin: Effect of histamine and some other substances on molecular selectivity of the capillary wall to plasma protein and dextran. Microvas. Res. 7:31-48, 1974. Chahl, J. S. and L. A. Chahl: The role of prostaglandins in chemically induced inflammation. Br. J. Exptl. Path. 57: 689-695, 1976. Chambers, R. and B. W. Zweifach: The topography and function of the mesenteric capillary microcirculation. Am. J. Anat. 75:173-205, 1944. Clough, G., P. A. Fraser, and L. H. Smaje: Compliance measure- ment in single capillaries of the cat mesentery. J. Physiol. 240:9P, 1974. Cotran, R. S., E. R. Suter, and G. Majno: The use of colloidal carbon as a tracer for vascular injury. Vasc. Dis. 4:107- 127, 1967. Crunkhorn, P. and A. L. Willis: Cutaneous reactions to intra- dermal prostaglandins. Br. J. Pharmacol. 41:49-56, 1971. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 112 Daugherty, R. N.: Effects of i.v. and i.a. PGEl on dog forelimb skin and muscle blood flow. Amer. J. Physiol. 220:392-396, 1971. Dray, F., B. Charbonnel, and J. Maclouf: Primary prostaglandins in human peripheral plasma by radioimmunoassay. Adv. Prosta. DuCharme, D. W., J. R. Weeks, and R. G. Montgomery: Studies on the mechanism of the hypertensive effect of prostaglandin FZQ. J. Pharm. Exp. Ther. 160:1-10, 1968. Euler, U. S. von: A depressor substance in the vesicular gland. J. Physiol. (London) 84:21P-22P, 1935. Euler, U. 5. von: On the specific vasodilating and plain muscle stimulating substances from accessory genital glands in man and certain animals (prostaglandin and vesiglandin). J. Physiol. (London) 88:213-234, 1936. Euler, U. S. von: fiber die spezifische blutdrucksenkende substanz des menschlichen prostata- und samenblasensckretes. Klinische Wbchenschrift 14:1182-1183, 1935. Euler, U. 8. von and R. Eliasson: Prostaglandins. Madicinal Chemistry Mbnographs, Vol. 8, Academic Press, New York- Ferreira, S. H., S. Moncada, and J. R. Vane: Indomethacin and aspirin abolish prostaglandin release from the spleen. Nature New Biol. 231:237-239, 1971. Ferreira, S. H., S. Moncada, and J. R. Vane: Prostaglandins, aspirin-like drugs and the oedema of inflammation. Nature 246:217—219, 1973. Folkow, B. and E. Neil: Circulation. Oxford University Press, New York, pp. 97-131, 1971. Freeman, P. C. and G. B. West: Skin reactions to prostaglandins. J. Physiol. Pharmac. 24:407—408, 1972. Fung, Y. C., B. W. Zweifach, and M. Intaglietta: Elastic environ- ment of the capillary bed. Circ. Res. 19:441-461, 1966. Garlick, D. G. and E. M. Renkin: Transport of large molecules from plasma to interstitial fluid and lymph in dogs. Amer. J. Physiol. 219:1595-1605, 1970. Goracci, G., G. Porcellati, and H. Woelk: Subcellular localiza- tion and distribution of phospholipases A in liver and brain tissue. Adv. Prost. Thrombox. Res. 3:55-67, 1978. Gorman, A. N.: The inter-relationship between prostaglandin E1, various vasoactive substances, and macromolecular permeability in the canine forelimb. M.S. Thesis, Michigan State University, 1979. ' 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 113 Greaves, M. W., J. Sondergaard, and W. McDonald-Gibson: Recovery of prostaglandins in human cutaneous inflammation. Brit. Med. J. 2:258-260, 1971. Greenberg, R. A. and H. V. Sparks: Prostaglandins and consecutive vascular segments of the canine hindlimb. Amer. J. Physiol. 216:567-571, 1969. Grega, G. J., D. E. Dobbins, P. B. Parker, and F. J. Haddy: Effects of intravenous histamine on forelimb weight and vascular resistances. Am. J. Physiol. 223:353—360, 1972. Grotte, G.: Passage of dextran molecules across the blood-lymph barrier. Acta Chir. Scand. Suppl. 211:1—84, 1956. Guyton, A. C., H. J. Granger, and A. E. Taylor: Interstitial fluid pressure. Physiol. Rev. 51(3):527—563, 1971. Haddy, F. J.: Effect of histamine on small and large vessel pressures in dog foreleg. Am. J. Physiol. 220:1448-1462, 1961. Haddy, F. J., J. B. Scott, and G. J. Grega: Effects of histamine upon lymph protein concentration and flow in dog forelimb. Am..J. Physiol. 223:1172-1177, 1972. Hamberg, M., J. Svensson, and B. Samuelsson: Thromboxanes: a new group of biologically active compounds derived from prostaglandin endoperoxides. Proc. Natl. Acad. Sci. USA 72:2994—2998, 1975. Hammerson, F.: The pattern of the terminal vascular bed and the ultra-structure of capillaries in skeletal muscle. In: Oxygen Transport in Blood and Tissues, edited by D. W. Lfibbers, Grune and Stratton, New York, pp. 184-197, 1968. Harris, R. H. and P. W. Ranwell: Cellular mechanisms of prosta- glandin action. Ann. Rev. Physiol. 41:653-668, 1979. Holmes, S. W., E. W. Horton, and I. H. M. Main: The effects of PGE1 on responses of smooth muscle to catecholamines, angio- tensin and vasopressin. Br. J. Pharmacol. Chemotherap. 21: 538-546, 1963. Horrobin, D. F.: Prostaglandins: Physiology, Pharmacology and Clinical Significance. Eden Press Inc., Montreal, Quebec, pp. 90-101, 1978. Horton, E. W.: Action of PGE1 on tissues which respond to bradykinin. Nature (London) 200:893, 1963. Horton, E. W.: Prostanoids in health and disease. In: Chemistry, Biochenfistry, and Pharmacological Activity of Prostanoids, edited by S. M. Roberts and F. Scheinmann, Pergamon Press Ltd., New York, pp. 1-16, 1979. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 114 Intaglietta, M.: Transcapillary exchange of fluid in single microvessels. In: Microcirculation, Vol. I, edited by G. Kaley and B. M. Altura, University Park Press, Baltimore, pp. 197-212, 1977. Johnson, R. A., D. R. Morton, J. H. Kinner, R. R. Gorman, J. C. McGuire, F. F. Sun, N. Whittaker, S. Bunting, J. Salmon, S. Moncada, and J. R. Vane: The chemical structure of prostaglandin x (prostacyclin). Prostaglandins 12:915-928, 1976. Jones, R. L.: Actions of prostaglandins D and E: evidence for two distinct receptors. Adv..Prost. Thrombox. Res. 4:905, 1978. Jonsson, C. E.: Smooth muscle stimulating lipids in peripheral lymph after experimental burn injury. J. Reconstr. Surg. 5:1-5, 1971. Joyner, W. L.: Effects of prostaglandins on macromolecular transport from blood to lymph in the dog. Amer. J. Physiol. 232(6):H690-H696, 1977. Joyner, W. L., E. Svensjo, and K. A. Arfors: Simultaneous measure- ments of macromolecular leakage and arteriolar blood flow as altered by PGEl and Bz-receptor stimulant in the hamster cheek pouch. Microvas. Res. In Press. Kadowitz, P. J., C. S. Sweet, and M. J. Brody: Differential effects of prostaglandins E1, E2, Fla and F26 on adrenergic vasoqonstriction in the dog hindpaw. J. Pharm. Exp. Ther. 177:641-649, 1971. Kaley, G. and R. Weiner: PGEl: a potential mediator of the inflammatory response. Ann. N.Y. Acad. Sci. 180:338-350, 1971. Karnovsky, M. J.: The ultrastructural basis of capillary permea- bility studied with peroxidase as a tracer. J. Cell Biol. 35:213-236, 1967. Kline, R. L., D. P. Sak, F. J. Haddy, and G. J. Grega: Pressure- dependent factors in edema formation in canine forelimbs. Jo phaIm. Exp. Ther- 193:452'459, 19750 Kuehl, F. A. and J. L. Humes: Direct evidence for a prostaglandin receptor and its application for prostaglandin measurements. Proc. Nat..Acad. Sci. USA 69:480-484, 1972. Kurzok, R. and C. C. Lieb: Biochemical studies of human semen. The action of semen on the human uterus. Proc. Soc. Exp. Biol. Med. 28:268-272, 1930. ’ 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 115 Landis, E. M.: Micro-injection studies of capillary permeability. II. The relation between capillary pressure and the rate at which fluid passes through the walls of single capillaries. Amer. J. Physiol. 82:217-238, 1927. Landis, E. M. and J. R. Pappenheimer: Exchange of substances through the capillary walls. Handbook of Physiology, Section 2: Circulation, Vol. 2, American Physiological Society, Washington, DC, pp. 961-1034, 1963. Lands, W. E. M.: The biosynthesis and metabolism of prosta- glandins. Ann. Rev. Physiol. 41:633-652, 1979. Lee, S. C. and L. Levine: Cytoplasmic reduced nicotinamide adenine dinucleotide phosphate-dependent and microsomal reduced nicotinamide adenine dinucleotide-dependent prostaglandin E-9-keto-reductase activities in monkey and pigeon tissues. J. Biol. Chem. 249:1369-1375, 1974. Lewis, A. J., D. J. Nelson, and M. F. Sugrue: Potentiation by PGEl and arachidonic acid of oedema in the rat paw induced by various phlogogenic agents. Br. J. Pharmacol. 50: 468P-469P, 1974. Lord, J. T., V. A. Ziboh, and 8. Warren: Prostaglandin binding to membrane fractions of rat skin. Adv. Prost. Throm. Res. 1:291-296, 1976. Maciejko, J. J., D. L. Marciniak, E. F. Gersabeck, and G. J. Grega: Effects of locally and systemically infused brady- kinin on transvascular fluid and protein transfer in the canine forelimb. J. Pharmacol. Exp. Therap. 205:221—235, 1978. Malpighi, M.: De Pulmonibus Observationes Anatomicae. Bologna. 1661. Marciniak, D. L., D. E. Dobbins, J. J. Maciejko, J. B. Scott, F. J. Haddy, and G. J. Grega: Antagonism of histamine edema formation by catecholamines. Amer. J. Physiol. 233: H180—H187, 1978. Mazwell, G. M.: The effect of PGE1 upon the general and coronary hemodynamics and metabolism of the intact dog. Br. J. Pharm. Chemother. 31:162-169, 1967. Miller, M. E., G. C. Christensen, and H. E. Evans: Anatomy of the Dog. W. B. Saunders Company, Philadelphia, p. 432, 1964. Milnor, W. R.: Capillaries and lymphatic vessels. In: Medical Physiology, edited by V. B. Mountcastle, The C. V. Mosby Co., St. Louis, pp. 984-992, 1974. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 116 Moncada, S., R. Gryglewski, S. Bunting, and J. R. Vane: An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits plate- let aggregation. Nature 263:663-665, 1976. Moncada, S. and J. R. Vane: Arachidonic acid metabolites and the interactions between platelets and blood vessel walls. New Eng. J. Med. 300:1142-1147, 1979. Moncada, S. and J. R. Vane: Mode of action of aspirin-like drugs. In: Advances in Internal Medicine, Vol. 24, edited by G. H. Stollerman, Year Book Medical Publishers, Inc., Chicago, pp. 1-22, 1979. Moncada, S. and J. R. Vane: Pharmacology and endogenous roles of prostaglandin endoperoxides, thromboxane A2, and prosta- cyclin. Pharm. Rev. 30:293-331, 1979.. Moncada, S., J. R. Vane, and E. A. Higgs: Human arterial and venous tissues generate prostacyclin (prostaglandin X), a potent inhibitor of platelet aggregation. Lancet 1:18-20, 1977. Nakano, J.: Cardiovascular actions. In: The Prostaglandins, Vol. I, edited by P. W. Ramwell, Plenum Press, New York, pp. 239-316, 1973. Nakano, J. and J. R. McCurdy: Cardiovascular effects of PGE1. J. Pharmacol. Exp. Ther. 156:538-564, 1967. Nalbandian, R. M. and R. L. Henry: Metabolic maps of structures and actions of prostaglandins, prostacyclin, thromboxane and cyclic AMP. Sem. Throm. Hemo. 5(2):87-111, 1978. Palade, G. E.: Blood capillaries of the heart and other organs. Circulation 24:368-384, 1961. Palade, G. E.: Fine structure of blood capillaries. J. Appl. Physiol. 24:1424, 1953. Palade, G. E. and R. R. Bruns: Structural modulations of plasma- lemmal vesicles. J. Cell Biol. 37:633—649, 1968. Panagides, J. and E. Tolman: Effect of various pharmacological agents on prostaglandin induced increases in microvascular permeability. Res.-Comm. Chemical Pathol. Pharmacol. 12: 609-612, 1975. Pappenheimer, J. R., E. M. Renkin, and L. M. Borrero: Filtration, diffusion and molecular sieving through peripheral capillary membranes. Am. J. Physiol. 167:13-46, 1951. Piper, P. J. and J. R. Vane: Release of additional factors in anaphylaxis and its antagonism by antiinflammatory drugs. Nature 223:29-35, 1969. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 117 Plummer, N. A., C. N. Hensby, A. Kobza-Black, and M. W. Greaves: Prostaglandin activity in sustained inflammation of human skin before and after aspirin. Clin. Sci. Mbl. Med. 52: 615-620, 1977. Renkin, E. M.: Filtration, diffusion and molecular sieving through porous cellulose membranes. J. Gen. Physiol. 38: 225-243, 1954. Renkin, E. M.: Multiple pathways of capillary permeability. Circ. Res. 41:735-743, 1977. Renkin, E. M., R. D. Carter, and W. L. Joyner: Mechanism of the sustained action of histamine and bradykinin on transport of large molecules across the capillary walls in dog paw. Microvas. Res. 7:49-60, 1974. Renkin, E. M. and D. G. Garlick: Blood-lymph transport of macromolecules. Microvas. Res. 2:392-398, 1970. Rhodin, J. A. G.: The ultrastructure of mammalian arterioles and precapillary sphincters. J. Ultrastruc. Res. 18:181- 223, 1967. Robbins, S. L. and M. Angell: Basic Pathology. W. B. Saunders Co., Philadelphia, pp. 31-67, 1976. Samuelsson, E., E. Granster, K. Green, M. Hamberg, and S. Hammerstrdm: Prostaglandins. Ann. Rev. Biochem. 44:669- 695, 1975. Shimamoto, T., M. Kobayashi, T. Takahashi, T. Motomiya, and F. Numano: Heart attack induced experimentally with thromboxane A2. Proc. Jpn. Acad. 53:38-42, 1977. Shimamoto, T., T. Takahashi, Y. Takashima, T. Motondya, F. Numano, and M. Kobayashi: Prevention by pthalazinol of thromboxane A2--induced myocardial infarction in rabbits. Proc. Jpn. Acad. 53:43-46, 1977. Simionescu, N., M. Simionescu, and G. E. Palade: Open junctions of venular endothelium: size limit of permeant molecules. J. Cell Biol. 75:373a, 1977. Simionescu, N., M. Simionescu, and G. E. Palade: Permeability of intestinal capillaries: pathway followed by dextrans and glycogens. J. Cell Biol. 53:365-392, 1972. Simionescu, N., M. Simionescu, and G. E. Palade: Permeability of muscle capillaries to exogenous myoglobin. J. Cell Biol. 57:424-452. 1973. Simionescu, N., M. Simionescu, and G. E. Palade: Permeability of muscle capillaries to small heme-peptides: evidence for the existence of patent transendothelial channels. J. Cell Biol. 64:586-607. 1975. 106. 107. 108. 109. 110.- 111. 112. 113. 114. 115. 116. 117. 118. 119. 118 Simionescu, M., N. Simionescu, and G. E. Palade: Segmental differentiations of cell junctions in the vascular endo- thelium: the microvasculature. J. Cell Biol. 67:863-885, 1975. Slauson, D. 0.: The inflammatory process. 27th Annual Meeting of the American College of Pathologists, Miami, Florida, December 1976. Smith, R. 0.: Lymphatic contractility - A possible intrinsic mechanism of lymphatic vessels for the transport of lymph. J. Exp. Med. 90:497-509, 1949. Sokal, R. R. and F. J. Rohlf: Biometry. Freeman, San Francisco, 1969. Solomon, A. K.: Characterization of biological membranes by equivalent pores. J. Gen. Physiol. 51:3355—3645, 1968. Starling, E. H.: On the absorption of fluids from the connective tissue spaces. J. Physiol. (London) 19:312-326, 1896. Subbiah, M. T. R.: 0n the characterization of prostaglandin E2-9-ketoreductase from aorta and regional differences in its activity. Proc. Exp. Biol. Med. 161:158-161, 1979. Svensjo, E.: Bradykinin and prostaglandin E1, E2, and an-induced macromolecular leakage in the hamster cheek pouch. Prosta- glandins and Med. 1:397-410, 1978. Svensjo, K.: Characterization of leakage of macromolecules in postcapillary venules. Dissertation, Acta Universitatis Upsaliensis, Uppsala, Sweden, 1978. Svensjo, K., K. E. Arfors, R. M. Raymond, and G. J. Grega: Morphological and physiological correlation of bradykinin- induced macromolecular efflux. Am. J. Physiol. 236(4):H600- H606, 1979. Tirone, P., P. Schiantarelli and G. Rosati: Pharmacological activity of some neuro-transmitters in the isolated thoracic duct of dogs. Lymphology 6:65-68, 1973. Vane, J. R.: Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature (London) 231:232- 235, 1971. - Vane, J. R.: Prostaglandins as mediators of inflammation. Adv. Prost. Thrombox. Res. 2:791-801, 1976. Verger, R., J. Rietsch, M. C. E. van Dam-Mieras, and G. H. de Haas: Comparative studies of lipase and phospholipase A2 acting on substrate monolayers. J. Biol. Chem. 251:3123- 3128,1976. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 119 Viguera, M. G. and F. Sunahara: Microcirculatory effects of prostaglandins. Can. J. Physiol. Pharm. 47:627, 1969. Vogt, W.: Role of phospholipase A2 in prostaglandin formation. Adv. Prost. Thrombox. Res. 3:89-95, 1978. Weeks, J. R.: Upjohn Co., Kalamazoo, MI 49001 (personal communi- cation, April 1979). Weeks, J. R. and F. Wingerson: Cardiovascular actions of PGE1 evaluated using unanesthetized relatively unrestrained rats. Proc. Fed. Am. Soc. Exp. Biol. 23:327, 1964. Weiner, R. and G. Kaley: Influence of prostaglandin E1 on the terminal vascular bed. Am.J. Physiol. 217:563-566, 1969. Wiederheilm, C. A.: The interstitial space and lymphatic pressures in the bat wing. In: Pulmonary Circulation and Interstitial Space, edited by A. P. Fishman and H. H. Hecht, University of Chicago Press, Chicago, pp. 29-41, 1969. Williams, T. J. and J. Morely: Prostaglandins as potentiators of increased vascular permeability in inflammation. Nature (London) 248:215—217, 1973. Williams, M. C. and S. L. Wissig: The permeability of muscle capillaries to horseradish peroxidase. J. Cell Biol. 66: 531-555, 1975. Willis, A. L.: Identification of PGE2 in rat inflammatory exudate. Pharm. Res. Comm. 2:297—304, 1970. Willis, A. L.: Parallel assay of prostaglandin-like activity in rat inflammatory exudate by means of cascade superfusion. J. Physiol. Pharm. 21:126-128, 1969. Wolff, J. R.: Ultrastructure of the terminal vascular bed as related to function. In: Microcirculation, edited by G. Kaley and B. M. Altura, University Park Press, Baltimore, pp. 95-130, 1977. llllllllllll“