H‘.,.. M IIIIIIIIIIIIIIIIIIIIIIIIIII WNW! I!lNWIll!H!fill/lili’fli’lll!11WW 3 1293 00908 1781 This is to certify that the dissertation entitled Qoia a? In (it/.mmdomf magician \IV\ EMcQo'i-bx I’V\ Haffifo+oxr\ci\l© presented by {iowvxe/S Aims/x New Q‘H—d has been accepted towards fulfillment of the requirements for DmiiforoA-L degree in Qi: CVC‘WLOQO 825 (12M A 72 :5! \ 7 Major professor ‘/ / Date 11/18/91 —fi m-— _ .. 4 r’ T i LIBRARY j Michigan State University x o- J *7 A PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or betore date due. I DATE DUE DATE DUE DATE DUE L_Jl_i_= MSU Is An Affirmative Action/Equal Opportunity Institution amoral” ROLE OF INFLAMMATORY MEDIATORS IN ENDOTOXIN HEPATOTOXICITY BY James Alan Hewett A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology 1991 632' 5837 Copyright by JAMES ALAN HEWETT 1991 ABSTRACT ROLE OF INFLAMMATORY MEDIATORS IN ENDOTOXIN HEPATOTOXICITY BY James Alan Hewett Endotoxin (lipopolysaccharide, LPS) is an outer cell wall component of gram-negative bacteria. It is thought to be an important contributing factor to the pathophysiologic alterations associated with infections by these bacteria. A myriad of adverse effects are attributed to LPS, including injury to the liver. The overall objective of this dissertation was to examine the role of several endogenous inflammatory mediators in the pathogenesis of LPS hepatotoxicity. Liver injury occurred between 3 and 6 hr after bolus iv injection of LPS in rats. This was preceded by an increase in hepatic neutrophil (PMN) numbers, an elevation of circulating tumor necrosis factor (TNF)-alpha concentration, and activation of the coagulation system. Protection against liver injury was afforded by depletion of circulating PMNs, which attenuated hepatic PMN accumulation, by either TNF- alpha antiserum or pentoxifylline, which attenuated the increase in circulating TNF-alpha concentration, and by the anticoagulant that PMHS, T2: to the pathog Pmis alo liver inju pentoxifyllir. accumulation Similarly, 1: concentratio suggesting t manifestatio anticoagulants, heparin and warfarin. These results indicate that PMNs, TNF-alpha, and the coagulation system contribute to the pathogenesis of LPS hepatotoxicity. PMNs alone are not sufficient for full manifestation of liver injury since neither' TNF-alpha antiserum, pentoxifylline, nor heparin pretreatment prevented the accumulation of PMNs in the liver after LPS exposure. Similarly, PMN depletion enhanced circulating TNF—alpha concentration by more than 3-fold after LPS exposure suggesting that TNF-alpha alone is not sufficient for full manifestation of liver injury. These results are consistent with an interaction between PMNs and TNF-alpha in the pathogenesis of ITS-induced liver injury. This interaction may contribute to liver injury by a mechanism which is dependent on the coagulation system since PMN depletion, TNF-alpha antiserum and pentoxifylline inhibited activation of the coagulation system after LPS exposure. The results described in this dissertation provide new insight into the mechanisms of liver injury from bacterial LPS by showing that blood PMNs, circulating TNF-alpha and the coagulation system each play important roles in the pathogenesis. While much remains unknown about the specific mechanisms by which each of these inflammatory mediators is involved, the results suggest that complex interactions among them may be necessary for full manifestation of liver injury. To my family Especially to my wife, Sandy, my grandmother, Mildred, and my parents, John and Eileen iv I am es .obert Roth, an independ instrumental Dr. Leon Bru me to resear I also Irshad Chau- their 9‘uida including Stachlewitz I Bahia, Dr. E CIQated a u: A Mr Contributic Stachlewitz responsiblel meticulouS analYSeg, Chapter 3 I Uni. Y ACKNOWLEDGMENTS I am especially grateful to my mentor and friend, Dr. Robert Roth, who never seemed to doubt my ability to become an independent, researcher; His abiding' support. has been instrumental in making it attainable. I am also grateful to Dr. Leon Bruner, who played an important role in introducing me to research and to the field of Toxicology. I also thank. members of :my ‘thesis committee, Drs. Irshad Chaudry, Lawrence Fischer, and Kenneth Moore, for their guidance, and everyone in 8346/7 Life Sciences, including Jim Wagner, Eric Shobe, Simi VanCise, Rob Stachlewitz, Dr. Marc Bailie, Dr. Cindy Hoorn, Dr. Lonnie Dahm, Dr. Eric Schultze and Dr. Patty Ganey who together created a unique and enjoyable working environment. A number of individuals made important technical contributions for which I am thankful, including Rob Stachlewitz and Simi VanCise, who provided proficient and responsible technical assistance, Dr. Eric Schultze, who's meticulous histopathologic, morphometric, and cytologic analyses contributed greatly to the results presented in Chapter 3, and Dr. Steven L. Kunkel and Pam Lincoln at the University of Michigan, who provided tumor necrosis factor- alpha ar tumor ne: Fina this dis: stipend i 07404. Fina companions alpha antiserum and who performed measurements of plasma tumor necrosis factor-alpha. Financial support for much of the work presented in this dissertation was provided by USPHS Grant ESO4139. My stipend was provided in part by USPHS Training Grant HL 07404. Finally, I am grateful to my wife, Sandy, for her companionship and love, which. made the difficult times bearable and the good times even better. vi LIST OF TAB LIST OF FIG LIST 01" ASE CHI-PIER 1.1 S 1.2 E 1 1 1.3 C 1.4 p h 1 l 1.5 I C l *h‘ TABLE OF CONTENTS Page LIST OF TABLES...0.0.0.0000...OOOOOOOOOOOOIOOOOOOOOOOOOOXiii LIST OF FIGURESOOOOOOOOOOO0.00.00.00.00.000000000000000QOXiv LIST OFABBREVIATIONS...OOOOOOOOOOOOOOOOO0.0...OOOOOOOOOOXVi CHAPTER 1 GENERAL INTRODUCTION............................l 1.1 1.2 Structure of LPS.................................2 Exposure to LPS..................................4 1.2.a Gram-negative bacterial infection.........4 1.2.b Absorption from the gastrointestinal tract.....................................7 Clearance and detoxification of LPS..............8 Pathogenic role of LPS in gram-negative bacterial sepsis.................................9 1.4.a Lethality.................................9 1.4.b Tissue injury............................10 Interactions between LPS and mammalian cells in vitro..................................11 1.5.a Cytotoxicity.............................11 1.5.b Stimulation of mediator release..........13 1.5.c LPS receptor.............................18 Binding of LPS to cell surface receptors.............................18 Binding mediated by LPS-binding protein in serum......................20 Nonspecific interactions with membranes.............................21 vii CHAPTER 2 2.1 2.2 TABLE OF CONTENTS (continued) EASE Pathophysiologic effects of LPS..... .......... ..22 1.6.a Circulatory shock.......... ........... ...22 Decreased cardiac output... ........ ......23 Hymtens-ion.‘...OOOOOOOOOO0.0... ...... .0023 other circulatory alterations. .......... .24 1.6.b Disseminated intravascular coagulationOOOOOOOOOOOOOOO. ........ 00.25 1.6.c Multiple organ damage....................25 Gastrointestinal tract... .............. ..26 LungSOOOOOOOOOIOOOOOOO OOOOOOOOOOOOOOOOO .027 LPS-INDUCED LIVER INJURY ....................... 30 Normal liver function and structure ............. 31 Morphologic and functional alterations in the liver after LPS exposure.................34 Morphologic a1terations.. .............. ..35 Sinusoidal changes............. ..... .....35 Changes in parenchymal cells... ....... ...36 Functional alterations..... ............ ..37 Mechanisms of LPS-induced liver injury..........38 2.3.3 Direct effects of LPS on the liver.......38 LPS-induced cholestasis........... ..... ..38 LPS-induced hepatic lipid metabolism............................40 Indirect (host-mediated) effects of LPS on the liver............. ...... ...40 Role of PMNs.............................4O Role of Kupffer cells....................48 Role of cytokines........................54 Role of the coagulation system...........58 Role of arachidonic acid metabolites.....59 Overall objective...............................62 viii M CHAPTER 13."; 3.4 (A, L.) (..l (J ’71 CHAPTER 3 TABLE OF CONTENTS (continued) Eégfi SpeCific aimSIOOOOOOIO0.0.0.0000...... ....... 00.63 2.5.a Specific aim 1.......... ........... ......63 2.5.b SpeCific aim ZOOOOOOOOOOOOOOOOO00.0.0000064 2.5.a Specific aim 3.. ...... ... ..... ...........64 ROLE OF NEUTROPHILS IN LPS-INDUCED LIVER INJIIRYOOOO....I...0.0.0........OOOIOOOOOOOOO00.0.000066 Abstract........................................67 Introduction....................................68 Materials and methods. ........................ ..70 3.3.a Animals..................................70 3.3.b Treatment protocols......................7O 3.3.c Anti-PMN and anti-LC 19 u u u u u u Hotbfla u u in uuw O www 00 tip-:1 preparation.................... ........ ..72 Evaluation of liver injury...............74 Histopathologic evaluation...............74 Morphometric analysis of hepatic lesions......................... ......... 75 Quantification of total and individual WBC numbers...................75 Isolation of hepatic NPCs................76 Cytologic examination of NPCs............77 Data analysis................... ..... ....78 ResultSOOOOO......OOOOOOOOOOOOIOOO....0.0.0.0...78 3.4.a Characterization of control and anti-PMN Ig..........................78 Effects of LPS on circulating WBC........8O Effect of PMN depletion on liver PMN number...... ..... ...........83 Effect of PMN depletion on LPS-induced liver injury..............90 Ineffectiveness of LC depletion on LPS hepatotoxicity...................101 ix 3.5 GHETER 4 LIVER 4.1 4.2 4.3 l AAAA 4-4 Re \ 4,5 TABLE OF CONTENTS (continued) 229E 3.5 Discussion.... ..... ..... ....................... 104 CHAPTER 4 ROLE OF TNF-ALPHA IN LPS-INDUCED LIVER INJURY.......................... .............. 110 4.1 Abstract ....................................... 111 4.2 Introduction.. ................................. 112 4.3 Materials and methods.. ......... ...... ......... 114 4.3.a Animals.................................114 4.3.b Treatment protocols.....................115 4.3.c Quantification of plasma TNF-alpha concentration............... ............ 116 4.3.d Enumeration of hepatic PMN numbers.................... ............. 116 4.3.e Evaluation of liver injury..............117 4.3.f Quantification of circulating WBCs......117 4.3.g Anti-PMN Ig preparation.................117 4.3.h Data analysis...........................118 4.4 Results................................... ..... .118 4.4.a Effect of antiserum to TNF-alpha on LPS hepatotoxicity...................118 4.4.b Effect of pentoxifylline on circulating TNF-alpha concentration after LPS exposure..... .......... . ...... 121 4.4.c Effect of pentoxifylline on LPS hepatotoxicity............ .............. 121 4.4.d Effect of PMN depletion on circulating TNF-alpha concentration after LPS exposure......................125 4.4.e Effect of pentoxifylline on hepatic PMN number......................... ..... 125 4.5 Discussion.................... ........... ......129 9‘ qmv—fi CELAI J. er CRAP”: r L. TABLE OF CONTENTS (continued) Page CHAPTER 5 ROLE OF THE COAGULATION SYSTEM IN LPS-INDUCED LIVER INJURY............. ............... 134 5.1 Abstract .......... ..... ........................ 135 5.2 IntrOductionOOOOOOOOOOO ....... O. OOOOOOOOOOOOOOO 136 5.3 Materials and methods... ...... . ............... .139 5.3.a AnimalsOOOOOOOOOOOOOO......OOOOOO ....... 139 5.3.b Treatment protocols............... ..... .139 5.3.c Quantification of plasma fibrinogen concentration and liver injury..........142 5.3.d Anti-PMN Ig preparation........ ....... ..142 5.3.e Quantification of hepatic NPC and PMN number................ ......... .142 5.3.f Data analysis.. ....................... ..144 5.4 Results.‘0.0.0.0000.........OOOOOOIOOOOO ....... 145 5.4.a Role of coagulation system in LPS hepatotOXiCitYOOOO0......0.0.00.0...0.0.145 5.4.b Mechanism of liver injury by the coagulation system after LPS exposureOOOOOOOOOOOO.....OOOOOOO0.00.00.153 5.4.c Mechanism of activation of the coagulation system after LPS............158 5.5 DiscuSSionOOOO......OOOOOOOOOOOOOO00.0.00000000161 CHAPTER 6 6.1 SUMMARY AND CONCLUSIONS.......................168 Host-derived mediators in LPS-induced liver injury............................ ...... .169 6.1.a Role of blood PMNs in LPS-induced liver injury............................170 6.1.b Role of TNF-alpha in LPS-induced liver injury............................170 6.1.c Role of the coagulation system in LPS-induced liver injury........ ........ 170 xi LIST 0 TABLE OF CONTENTS (continued) BASS 6.2 Relationship between host-derived mediators in LPS-induced liver injury....................171 6.2.a Relationship between PMNs and TNF-alpha.......OOOOOOOO00.0.0.0... ..... 171 6.2.b Relationship between PMNs, TNF-alpha and the coagulation system........ ...... 172 6.3 Proposed mechanism of LPS-induced liver injury: an hypothesis..........................173 LIST OF REFERENCES .............. . .......... . . . . . . . . ...... 178 xii 3.2 3.3 4.1 4.2 5.1 LIST OF TABLES 1921; Page 1.1 Inflammatory mediators released by LPS ............... 14 3.1 Effect of control I9 and anti-PMN Ig on circulating blood cell numbers and hematocrit in the absence of LPS................................79 Cytologic evaluation of NPC fraction from liver digests after exposure to LPS..... ..... ........84 Morphometric analysis of livers from control Ig- or anti-PMN Ig-preteated rats after exposure to LPS....................... ............ ...96 Effect of pentoxifylline on the LPS-induced increase in circulating TNF-alpha concentration.....122 Effect of PMN depletion on the LPS-induced increase in circulating TNF-alpha concentration ..... 126 Effect of anticoagulants on circulating fibrinogen concentration in the presence and absence of LPS exposure................. ....... .152 xiii 3.7 N e lit LIST OF FIGURES figure Base 1.1 Routes of exposure to and clearance of LPS... ....... ..5 1.2 Interactions of LPS with mammalian cells ............ .16 2.1 The liver lobule................... ................. .32 2.2 Neutrophil (PMN) response to infection by gram-negative bacteria.................. ........... ..42 Changes in circulating total white blood cells (WBCs), neutrophils (PMNs) and lymphocytes (LC) after LPS administration to either control Ig- or anti-PMN Ig-pretreated rats.......... ........ .81 Photomicrographs of NPC fractions from rat liver digests after exposure to LPS vehicle ........ ..85 Photomicrographs of NPC fractions from rat liver digests after exposure to LPS..................88 Effect of pretreatment with either control Ig or anti-PMN Ig on LPS-induced liver injury. ........ ..91 Photomicrographs of liver after administration of LPS to rats pretreated with control Ig.... ..... ...93 Photomicrographs of liver after administration of LPS to rats pretreated with anti-PMN Ig... ...... ..97 Time—course of liver injury after administration of LPS to rats pretreated with either control Ig or anti-PMN IgOOOOOOOOOOOOOOOO0.0.0.000...O000000.99 xiv figure 3.8 4.1 Ef in LIST OF FIGURES (continued) Figure Page 3.8 Comparison of the effect of anti-PMN Ig and anti-LC Ig on plasma AST activity after LPS exposure ........................................ 102 4.1 Effect of TNF antiserum on LPS hepatotoxicity in rats ............................ . ................. 119 Effect of pentoxifylline on LPS hepatotoxicity......123 Effect of pentoxifylline on hepatic nonparenchymal cell (NPC) and neutrophil (PMN) numbers after LPS exposure... ................. 127 Development of liver injury and activation of the coagulation system after LPS exposure ........... 146 Effect of heparin on LPS-induced liver injury. ...... 148 Effect of warfarin on LPS-induced liver injury ...... 150 Effect of ancrod on LPS-induced liver injury ........ 154 Effect of heparin on hepatic neutrophil (PMN) numbers after LPS administration .............. 156 Effect of depletion of circulating neutrophils (PMNs) on activation of the coagulation system after LPS exposure....... ..... .............. 159 Effect of pentoxifylline on LPS-induced activation of the coagulation system...... ........ ..162 Proposed mechanism of LPS-induced liver injury ...... 175 ALT ANOVA Anti-LC Ig Anti-PMN Ig AST Control Ig DDW DIC d1 E. coli EGTA HBSS HEPES hr 19 IL ip LIST OF ABBREVIATIONS Alanine aminotransferase Analysis of variance Immunoglobulin fraction from serum of rabbits immunized with rat lymphocytes Immunoglobulin fraction from serum of rabbits immunized with rat neutrophils Aspartate aminotransferase Immunoglobulin fraction from serum of non- immunized rabbits Double distilled water Disseminated intravascular coagulation Deciliter Escherichia coli Ethylene glycol-bis(beta-aminoethyl ether) N,N,N’N'-tetraacetic acid Hank’s balanced salt solution N-z-hydroxyethylpiperazine-N'-2- ethanesulfonic acid Hour Immunoglobulin Interleukin Intraperitoneal xvi iv kila 15? 13? n1 ng NPC Plfli po SF U TNF-51; ul EEC LIST OF ABBREVIATIONS (continued) iv Intravenous kDa Kilodalton kg Kilogram LPS Endotoxin (lipopolysaccharide) mg Milligram ml Milliliter ng Nanogram NPC Nonparenchymal cell 02' Superoxide anion PMN Neutrophil (polymorphonuclear leukocyte) po Oral (per 05) SF U Sigma-Frankel Units of transaminase activity TNF-alpha Tumor necrosis factor-alpha ul Microliter WBC White blood cell xvii CHAPTER 1 GENERAL INTRODUCTION sectic the r: circul altera bacter reSPOn. this EXtrah, dESCri] disSert in the e2Phasj endOGEr 1.1 St LP' tactErie bactEria A myriad of pathophysiologic alterations are associated with infection by gram-negative bacteria and many of these alterations can be attributed to gram-negative bacterial endotoxin, or lipolysaccharide (LPS). The purpose of the initial sections of Chapter 1 is to introduce LPS. These sections provide brief descriptions of the structure of LPS, the routes of exposure to and clearance of LPS from the circulation, and the role of LPS in the pathogenesis of alterations associated with infections by gram-negative bacteria. This is followed by a detailed discussion of the response of mammalian cells to LPS. The final sections of this chapter will describe the effects of LPS on extrahepatic tissues. The effects of LPS on the liver are described in detail in Chapter 2. Because the focus of this dissertation is on the role of certain host-derived factors in the pathogenesis of LPS-induced liver injury, the emphasis of these two chapters will be on the role of endogenous mediators in the response of host tissues to LPS. 1.1 Structure of LPS LPS is an endotoxin derived from gram-negative bacteria. It is a major component of the cell wall of these bacteria and is one of the distinguishing features between varie negat polys. regio: terne< the e pclysa polyse noietj consti the C( the ur struct differ gram-negative and gram-positive bacteria which do not contain LPS in ‘their cell wall. Although the structure varies slightly among different species and strains of gram- negative bacteria, all LPS molecules are composed of polysaccharide and lipid domains (1). The polysaccharide region consists of a repeating oligosaccharide structure, termed the O-antigen polysaccharide, which extends toward the extracellular environment of the bacterium, and a core polysaccharide which covalently links the O-antigen polysaccharide with the lipid region. Unusual dideoxysugar moieties, such as colitose and paratose, are often constituents of the O-antigen oligosaccharide unit whereas the core polysaccharide is characterized by the presence of the unique sugar, 2-keto-3-deoxyoctonate. In contrast to the structure of the core polysaccharide, which is similar among different strains and species of bacteria, the composition of the O-antigen polysaccharide is usually more variable. The lipid region of LPS is often refered to as lipid A. This amphipathic structure is a major component of the outer leaflet of the cell wall lipid bilayer. It consists of several long-chain fatty acids linked by amide and ester bonds to two glucosamine residues. These glucosamine residues are joined by a .beta-1-6 linkage to form the backbone of the lipid A molecule. Charged phosphate or pyrophosphate groups bound to this glucosamine backbone contribute to the amphipathic nature of lipid A along with the hydrophobic fatty acids. DODOC 1.2 ] (Figur during integr‘ bacter. QHVirOr CirCUhs followi StudieS infecti abd0min. cell~fr£ levels infectio LPS in t. Many’ of the biological effects associated with LPS appear to be mediated by lipid A. This is supported by evidence which indicates that exposure to purified lipid A produces many of the same effects as LPS. Furthermore, most of the biological activities of LPS can be neutralized by substances which bind specifically to this region of the LPS molecule such as the antibiotic, polymyxin B, and certain monoclonal antibodies to lipid A (2). 1.2 Exposure to LPS 1.2.a Gram-negative bacterial infection Exposure to LPS may occur by several different routes (Figure 1.1). Perhaps the most obvious route of exposure is during infection by gram-negative bacteria. LPS is an integral component of the cell wall of gram-negative bacteria and is not normally released into the extracellular environment. However, it can be liberated under certain circumstances, particularly' during' cell division. and following death of the bacteria. Evidence from numerous studies indicates that a rise in cell-free LPS accompanies infection by gram-negative bacteria. For example, following abdominal infection of rabbits by Pasteurella multocida, cell-free LPS in the plasma increased from non-detectable levels prior to infection to 100 ug/ml by 6 hr after infection (3). Similarly, increases in the concentration of LPS in the cerebrospinal fluid have been associated with P; 939.1 .um>HH on» >n >Humawum omummao mun mmq .cofluoasonflo on» cue mono 43 «name may we whoHu Howumuonn m>flummocnsnuc msocmmwocfl scum cot/«not mmq mo uncuu Hocwummfifiouummm on» scum coaumuomnm ca mmmmuocw .HNH «Mahmuonn m>wummmc nacho xn newuomucfl mcwuso cwuouonn comp can onwofiép Eouu Ummmmamu mag .72 .mmq mo moccummao can 0» musmomxm no mousom HA musowm com: .0 a _ 2.; v 56> . Eton. L . cozflzozo 02.8616 333 p - 3:223 ll..._ mai. .......... o>=mmocéwto 3 9:9“. gram-mega: the disea: disease (E Gran- usinq bac antimicrol the bacte: ray cause (617,8,9'1 exposure t 7 gram-negative bacterial meningitis both in animal models of the disease (4) as well as in humans suffering from the disease (5). Gram—negative bacterial infections are often treated using bactericidal antibiotics. While several of these antimicrobial agents have proven effective in eliminating the bacteria, results from recent studies suggest that they may cause the release of large amounts of cell-free LPS (6,7,8,9,10). Thus, antibiotic therapy may enhance the exposure to LPS during gram-negative bacterial infections. 1.2.b Absorption from the gastrointestinal tract In contrast to exposure during gram-negative bacterial infection, it has been proposed that LPS exposure could occur in the absence of bacterial infection through an increase in the absorption of LPS from the gastrointestinal tract. The gastrointestinal tract normally contains a large concentration of LPS which is presumably derived from the indigenous gram-negative bacterial flora of the gut (11) . Under normal conditions, the intestinal wall acts as a formidable barrier to the passage of LPS from the gastrointestinal tract into the blood stream. However, disruption of this barrier during certain pathophysiologic conditions can lead to endotoxemia. For example, occlusion of the portal vein results in an increase in the LPS concentration in both portal venous and systemic blood (12). Whereas LPS was not detectable in blood from sham operated animals occlusi observe endoto> increas tract. absorpi implica injury (16), ; Provide condit; gaStro: 1.3 c CIEarel large Circul. this gradua. aziEiTli: SPIQED‘ dCQUnuJ with 8 animals, it ranged from 100-300 ng/ml after portal vein occlusion. Because this increase in. plasma LPS was not observed in germ-free animals, it was concluded that endotoxemia induced by portal vein occlusion was due to an increase in absorption. of LPS from. the gastrointestinal tract. In addition to portal vein occlusion, increased absorption of LPS from the gastrointestinal tract has been implicated in certain instances of chemically induced liver injury (13,14), dietary cirrhosis (15), partial hepatectomy (16), and following intestinal ischemia (17). These studies provide strong evidence that exposure to LPS can occur under conditions in which the intestinal barrier to gastrointestinal LPS is compromised. 1.3 Clearance and detoxification of LPS Following bolus intravenous (iv) administration, LPS is cleared from the circulation in a biphasic manner (18,19). A large fraction of the initial dose disappears from the circulation within minutes after injection. Subsequent to this rapid clearance phase, LPS elimination progresses gradually over a period of hours. Although intravenously administered LPS is found in numerous tissues, including the spleen, lungs, kidneys, and adrenal glands, the majority accumulates in the liver where it is associated primarily with Kupffer cells (20,21,22,23). These fixed hepatic macrophaqE clearance Clea: plasma li high dens the half-i complexed biological function (25). LPS contained (25)- This the lipid as baCter. lyse and the Surfa 1-4 Path Sng 1.4. Dea‘ is often HEgatiVQ hospital bacteria Contribu macrophages are thought to play an important role in the clearance and detoxification of LPS from the circulation. Clearance of LPS from the blood stream is influenced by plasma lipoprotein particles. LPS binds predominately to high density lipoprotein particles in plasma. This prolongs the half-life of circulating LPS (24). However, because LPS complexed with high density lipoprotein particles is less biologically active, it has been proposed that this may function as a protective mechanism against LPS toxicity (25). LPS can also be neutralized by an LPS-binding protein contained within specific granules of neutrophils (PMNs) (26). This 50-60 kDa protein, which specifically recognizes the lipid A region of the LPS molecule (27), is referred to as bacterial/permeability increasing protein because it can lyse and kill gram-negative bacteria by binding to LPS on the surface of the bacteria (28,29,30). 1.4 Pathogenic role of LPS in gram-negative bacterial sepsis 1.4.a Lethality Death resulting from shock and multiple organ failure is often a consequence of overwhelming infection by gram- negative bacteria. Indeed, a large percentage of deaths of hospitalized patients can be attributed to gram-negative bacterial sepsis. It has been proposed that LPS is a major contributing factor to the high mortality associated with gram—i evidei with prote bacte hav in l bacte contr gran- 10 gram-negative bacterial infection. This is supported by evidence from studies in animal models in which treatment with specific neutralizing antibodies to LPS afforded protection against the lethal effects of gram-negative bacterial sepsis (31,32,33,34). Specific antibodies to LPS have also proven to be effective in reducing the mortality in hospitalized patients suffering from gram-negative bacterial infections (35,36,37). Thus, LPS appears to contribute to the mortality associated with infection by gram-negative bacteria. 1.4.b Tissue injury In addition to its role in lethality, LPS appears to play a role in the development of tissue injury during local gram-negative Ibacterial infections. For’ example, LPS ‘has been implicated in the pathogenesis of tissue injury in gram-negative bacterial meningitis. A large percentage of clinical cases of bacterial meningitis can be attributed to certain gram-negative bacteria, including Haemophilus influenzae type. b, .Escherichia (E.) coli, and..Neisseria meningitidis. Infection of the cerebrospinal fluid by these bacteria is accompanied by inflammation of the meninges, which is characterized by the appearance of large numbers of leukocytes and by alterations in blood-brain barrier permeability (38). The adverse effects associated with gram- negative bacterial meningitis appear to be mediated by LPS since intracisternal administration of LPS produces many of these polyr} induc: injec infil lesic {'1 injui Poly: injU] the baCti 1.5 11 these same effects and since neutralization of LPS with polymyxin B afforded protection against experimentally induced gram-negative bacterial meningitis (39,40). LPS has also been implicated in the pathogenesis of tissue injury following gram-negative bacterial infection of the skin. Lesions in the skin following intradermal injection of E. coli are characterized by hyperemia, increased vascular permeability, and hemorrhage and are infiltrated by large numbers of neutrophils (41). Similar lesions occurred following intradermal injection of killed E. coli or' purified..Eu coli LPS (42). Furthermore, the injury was attenuated by pretreatment of the bacteria with polymyxin B (43). Thus, like gram-negative bacteria-induced injury to the blood-brain barrier, microvascular injury in the skin following intradermal injection of gram-negative bacteria is mediated by LPS. 1.5 Interactions between LPS and mammalian cells in vitro 1.5.a Cytotoxicity Results from several studies indicate that LPS is cytotoxic to cells in vitro under certain conditions. For example, marked degenerative morphological changes are observed in cultured vascular endothelial cells after exposure to LPS (44). These morphological changes are accompanied. by increased. cell detachment. and leakage of cytoplasmic contents, and by decreased DNA, RNA, and protein synthesi on both concentr presence conjugat cytotoxi injury peroxida that th. scavenge: xanthine allopurir afforded recent 5: to LPS‘ir In this POUChe$ . Was Preve allopurii In cells, L and rib: Cyt°toxi metabol: 12 synthesis. Injury to cultured endothelial cells is dependent on both the duration of LPS exposure (>4 hr) and the concentration of LPS (>0.1 ug/ml) and is enhanced by the presence of serum (45,46). Oxygen radical production and conjugated. diene formation are associated. with the cytotoxicity, suggesting that LPS-induced endothelial cell injury is mediated by oxygen radical-dependent lipid peroxidation (47). This is supported by evidence indicating that the cytotoxicity is attenuated by oxygen radical scavengers. The source of the oxygen radicals appears to be xanthine oxidase since the xanthine oxidase inhibitor, allopurinol, attenuated the oxygen radical production and afforded protection against the cytotoxicity. Results from a recent study suggest that a similar mechanism may contribute to LPS-induced injury to vascular endothelial cells in vivo. In this study, LPS administered locally to hamster cheek pouches caused an increase in microvascular leakage which was prevented by the antioxidant, dimethyl sulfoxide, and by allopurinol (48). In addition to its effects on vascular endothelial cells, LPS is also cytotoxic to cultured macrophages (49) and fibroblasts (50). However, whether the mechanisms of cytotoxicity in these cells is dependent on reactive oxygen metabolites is not known. 1.5. In e the rele certain substance potent, l in the exposure LPS culture 1 the relea appears t induced a release (51). In CYtOkine, macrOphag (51,52). “diene: product i ”115. A moiECulE and Prone Ritz-0pm} The; “Sin ;: d 13 1.5.b Stimulation of mediator release In addition to its cytotoxic effects, LPS stimulates the release of a variety of inflammatory mediators from certain cells in culture. Many of these endogenous substances and their sources are listed in Table 1.1. These potent, biologically active substances have been implicated in the pathophysiologic alterations associated with LPS exposure (see below and Table 1.1). LPS stimulates the release of mediators from cells in culture by several mechanisms. For example, stimulation of the release of arachidonic acid metabolites from macrophages appears to be mediated at least in part by the direct, LPS- induced activation of phospholipase A2, which catalyzes the release of arachidonic acid from membrane phospholipids (51). In contrast, LPS stimulates the production of the cytokine, tumor necrosis factor (TNF)-alpha, from macrophages by inducing TNF-alpha gene transcription (51,52). This is indicated by the increase in TNF-alpha mRNA which occurs prior to the onset of TNF-alpha release. Induction of gene expression by LPS contributes to the production of numerous other inflammatory mediators by cells. Among these are the increased expression of adhesion molecules for PMNs on vascular endothelial cells (53,54), and production of certain interleukins and nitric oxide by macrophages (55,56) and smooth muscle cells (57,58,59). The signal transduction pathways involved in the LPS-induced alteration in cell function are not completely Mediator Couplenc Clottim Tissue Oxygen LYSOsor. Nitric Arachic metal PIQIEIQ facti Tumor ‘ fact IDIerl 14 Table 1.1 Inflammatory mediators released by LPS Mediator Possible Sourcesa Likely Target Tissues l bl I Complement factors Clotting factors Tissue factor Oxygen radicals Lysosomal enzymes Nitric oxide Arachidonic acid metabolites Platelet Activating factor Tumor necrosis factor Interleukins Plasma Plasma Endothelial cells Macrophages Endothelial cells Macrophages Neutrophils Macrophages Endothelial cells Smooth muscle cells Macrophages Hepatocytes Endothelial Macrophages Neutrophils Platelets cells Endothelial Macrophages Neutrophils cells Endothelial Macrophages cells Endothelial cells Smooth muscle cells Macrophages Cardiovascular Kidneys Liver Lungs Plasma Cardiovascular Lungs Liver Liver Cardiovascular Liver Cardiovascular Gut Lungs Liver Cardiovascular Gut Lungs Cardiovascular Liver Lungs Cardiovascular See text of Chapter 1 for description and references See text of Chapter 2 for description of mediators involved in LPS-induced liver injury and Chapter 1 for mediators involved in alterations in extrahepatic tissues define to be eviden expres line c lipoxy altera S-lipo I intrac. (PK-C) eXposu: additit of TNF. of mic actiVa. the r 15 defined (59a). However, arachidonic acid metabolites appear to be involved in certain instances. This is supported by evidence which indicates that the induction of TNF gene expression by LPS in the HL-60 macrophage-like tumor cell line can be blocked by inhibitors of phospholipase A2 or 5- lipoxygenase (51). This suggests that LPS-induced alterations in TNF-alpha gene expression are dependent on a 5-lipoxygenase product of arachidonic acid. Increases in phosphatidylinositol metabolism (60), intracellular Ca2+ concentration (60), and protein kinase-C (PK-C) activity (61,62) have also been reported after exposure of cells in vitro to either LPS or lipid A. In addition, inhibitors of protein kinase C blocked expression of TNF-alpha and interleukin (IL)-1 mRNA in primary cultures of mice macrophages treated with LPS in vitro (62a). Thus, activation of CaZT-dependent PK-C appears to play a role in the response of cells to LPS in certain instances. Calmodulin kinase also appears to contribute to expression of IL-1 mRNA in LPS-treated macrophages (62a). Signal transduction after LPS exposure by these second messenger pathways is 'probably’ mediated. by' alterations in 'protein phosphorylation (62b). Possible signal transduction pathways contributing to the response of cells to LPS are illustrated in figure 1.2. 16 /. @203th Iii-'l UQHUOOimQJ Nip 0L30.l < .Hmu mmufiaonnuoa neon owcoofinomum o>fluon waanofimoH0an mo :ofiunsuom ecu CH ocwuasmmu Hag mowmfiaonmmonm manunama Bonn capo oficooflconum henchmen: Nd mmomfiaosmmocm mo cowun>fluo< .2; :oflunfiauocmmosm swououm commuac ou wasnfiuucoo >9: Hog +moo an SUV momncfix cHHonano uo :oflum>fiuoo uo\ocn ”mg +~no can meg Ode wn Auxmv O omncfix samuoum no cofluo>fluon “HmH +~mo unasaawoouucfi aw ommmuosfl so museums can mmH “Hum Loans Houmosauasoaac new HHL AmmHv ouanomonomauu Houfimocfl mmumumcmm Emflaoomuoa Houwmocflonmmonm mo cofluoasfifium umwnznumm cowuoscmcnuu Hmcmflm “manaauoouucfl Houo>om ha uncommon m. on wood aamumfiwuas >9: nouns: Amocfia oonmog .3:on mcmunamfimcmuu moumauwcw muoumooou macho—Ems named“ cues mma mo cowuomuoucH "m .mpflmwa msmunfima mammam spas. mmq mouwlaaoo mo _co«uooumuca Damaoommcoc .ng «maaoo cfimuuoo Mo moouusm ecu co .>Ho>fluoommou .mcwmuoum max cm can mm 0» mag mouuuaaoo mo mcwccflo wombat .meu can ”mg «mmmcnmouonfi so somfiucn eaoo ou mmq mouuiaaoo no moaofluucm coucOOimQQ Ho mcwocwn ooumwpmaicflmuoum masocflnimma asumm .HNH umHku ofiuhoommnm :0 advance maau ou Amwuouomn m>flunooCIBouo powwow .mmv moHOAuuom poumooimmq mo onwocflm .HHH ”c .mHHou smeawssoa sues mmq Ho mcowuonuoucH N.H ensues 17 52.0338 Eon 02022092 cozflbocamoca £305 no.2? ovah u xo 7 \N NmOé a + Boo 0:. Econ}. v more 9.56 a 2.6 .— «/ ax a 83.02 sec. N< omoazocqwocd _8_moc_ocawozd m a“, A, A, a. $223022: «Ea: m m m m 0555...: 2:8 «9.3 See _ _ Eco; > $22.25:me n v a a u w 5808 mEIIV seesaw“... EDCQw / «23th 82893.. m; 2:5 < 1.5. LPS receptor- interact. identifi' binding LPS (ie, receptor associat bacteria associat (Couplex of LPS interact Bir apparent have be identifj (63), A. that th and 0n bl°0d line 95 kDa J774A.1 18 1.5.c LPS receptor LPS can interact with cells by either specific, receptor-mediated binding or by non-specific membrane interactions. Several cell surface receptors have been identified which bind LPS, and the type of receptor-mediated binding is dependent on the nature of LPS. Thus, cell-free LPS (ie, LPS purified from gram-negative bacteria) binds to receptors which are distinct from those that bind cell- associated LPS (ie, LPS associated with gram-negative ' bacteria). Also, complexes formed between cell-free or cell- associated LPS and a specific serum LPS-binding protein (complexed LPS) bind to different receptors than either form of LPS in the absence of complex formation. The different interactions of LPS with cells are summarized in Figure 1.2. Binding of LPS to cell surface receptors. At least two apparently different cellular receptors for cell-free LPS have been identified. One receptor, which was first identified in splenocyte isolates, is an 80 kDa protein (63) . An analysis of subpopulations of splenocytes showed that this receptor was present on both B and T lymphocytes and on macrophages. However, it was not found on human red blood cells or the undifferentiated murine myeloma cell line, Sp2/0. The second receptor has a molecular weight of 95 kDa and was initially identified in the RAW 264.7 and J774A.1 macrophage-like cell lines (64). Subsequent studies with. human blood leukocytes indicated. that an identical receptor neutrophi Clio-K1 a‘. This ind receptor recogniz associat proteina saturabi Sew the hind is the MacrOpha and thi: CYtokine against macrophE macrophe induced 19 receptor was present on human blood monocytes (65) . Human neutrophils (PMNs) as well as the fibroblast cell lines, CHO-Kl and L929, did not have detectable 95 kDa receptor. This indicates that, like the 80 kDa receptor, the 95 kDa receptor exhibited cell specificity. Both receptors recognize the lipid. A region of the LPS molecule, are associated with the cell membrane fraction, are sensitive to proteinase K digestion, and exhibit specificity and saturability (64,66). Several LPS-induced responses appear to be mediated by the binding of cell-free LPS to cell receptors. Among these is the stimulation of tumoricidal activity by macrophages. Macrophages lyse tumor cells after exposure to LPS in vitro, and this tumoricidal activity is markedly enhanced by the cytokine, gamma-interferon. A. monoclonal antibody raised against the 80 kDa receptor blocks binding of LPS to macrophages and stimulates tumoricidal activity by macrophages (67,68). As with LPS, the tumoricidal activity induced by this monoclonal antibody is enhanced by gamma- interferon. These results are consistent with the role of the 80 kDa receptor in LPS-induced macrophage tumoricidal activity. Binding to the 80 kDa protein also appears to be required for the induction of B lymphocyte mitogenesis by LPS in vitro (66). The interaction of cell-associated LPS with certain mammalian cells depends on the presence of the C018 group of adhesion molecules (69 , 70) . These cell surface glyco macro inclu parti and a C018 derivi for t} cells Altho' direc kDa inter forme (COn‘ is p is s duri affi Stra. LPs n which the me which 20 glycoproteins, which are present on phagocytic cells such as macrophages and PMNs, mediate a variety of functions including adherence to surfaces and phagocytosis of particles coated (opsonized) with complement-derived factors and antibodies (71). Interaction of cell-associated LPS with CD18 does not require prior opsonization. by complement- derived factors or antibodies and appears to be important for the phagocytosis of gram—negative bacteria by phagocytic cells. Binding mediated by LPS-binding protein in serum. Although both cell-free and cell-associated LPS can interact directly with specific mammalian cell receptors (80 and 95 kDa and CD18 receptors, respectively), the cellular interaction of both forms of LPS may be modified by complex formation with an LPS-binding protein found in serum (complexed LPS). This 60 kDa acute phase glycoprotein, which is present in low concentrations in serum of normal animals, is synthesized by hepatocytes and increases in concentration during the acute phase response (72,73). It has a high affinity for LPS from a variety of gram-negative bacterial strains and appears to recognize the lipid A region of the LPS molecule (74). Interactions between LPS and macrophages which are mediated by this serum protein are dependent on the macrophage CD14 antigen. This is illustrated by evidence which indicates that binding can be blocked by monoclonal antibodies to CD14 (75). bindi: macro; activ. induc‘ couple of co: surfac is al: as an the b can inter; by Cc This Other resp the bind EEmbr inter 21 Formation of a complex with the 60 kDa acute phase LPS— binding protein markedly enhances the response of isolated macrophages to LPS as indicated. by enhanced. tumoricidal activity. Monoclonal antibodies to CD14 antigen block the induction of macrophage tumoricidal activity by these complexes, providing further evidence that the interaction of complexed LPS with macrophages is mediated by this cell surface protein (76). Phagocytosis of gram-negative bacteria is also enhanced by 60 kDa LPS-binding protein, which acts as an opsonin apparently by binding to LPS on the surface of the bacteria (77). These studies indicate that, while LPS can affect macrophage function through the direct interaction with cell receptors, these effects are modified by complex formation with the serum LPS-binding protein. This LPS-binding protein also may mediate effects of LPS on other cells, including neutrophils (78). Nonspecific interactions with membranes. Although the response of cells to LPS in many instances is mediated by the presence of specific cell surface receptors, LPS can bind to cells by nonspecific interactions with the plasma membrane (79). However, it is not clear whether or not this interaction results in a biological response in these cells. 1.6 Path An attributl dissenin several gastroir of path high n: death 0; Th! these Neverth of host médiato induced along 3 they ma meChani Several derived 22 1.6 PathOphysiologic effects of LPS An array of pathophysiologic alterations have been attributed to LPS. Among these are circulatory shock, disseminated intravascular coagulation (DIC), and damage to several organs, including the heart, kidneys, gastrointestinal tract, lungs and liver. With the multitude of pathophysiologic effects, it is not surprising that, a high.'morta1ity’ rate laccompanies severe. endotoxemia, with death often ensuing within 48 hours. The mechanisms contributing to the manifestation of these alterations are diverse and vary among tissues. Nevertheless, a common motif appears to be the involvement of host-derived, soluble mediators. A variety of endogenous mediators have been implicated in the pathogenesis of LPS- induced alterations. Many of these are listed in Table 1.1 along with their possible sources and example tissues which they may affect. The following section describes briefly the mechanisms contributing to ITS-induced alterations in several tissues with an emphasis on the role of these host- derived mediators. 1.6.a Circulatory shock Changes in the circulatory system during severe endotoxemia are characterized by marked alterations in cardiac output, mean arterial blood pressure, and organ blood flow. Decr injectior (80,81). approxina followed which is phase. I to be me can be i (82). contract later 5 factors decreasi HY. Pressur LPS-ind Vascula contraC hYPOter Vascul; I depend metabc, (85) a nitric 23 Decreased cardiac output. Within 1 hour after bolus iv injection of LPS in pigs, cardiac output falls dramatically (80,81). A brief rebound period is often observed approximately 1 hour after LPS administration. This is followed by a second, sustained decrease in cardiac output which is more gradual in onset in comparison to the initial phase. The early, transient change in cardiac output appears to be mediated by metabolites of arachidonic acid, since it can be blocked by inhibitors of arachidonic acid metabolism (82). In contrast, LPS-induced decrease in myocardial contractility seems to play an important role in the the later stage of altered cardiac output (83). Several other factors have been proposed to be involved as well, including decreased venous return. Hypotension. A marked decrease in mean arterial blood pressure is also observed during severe endotoxemia (84) . LPS-induced hypotension is associated with a decrease in vascular contractility. This alteration in vascular contractility’ occurs prior' to ‘the ‘manifestation of hypotension and is characterized by hyporesponsiveness of vascular preparations to vasoconstrictors such as KCl, norepinephrine, and 5-hydroxytryptamine (84). It is not dependent on vascular endothelium or arachidonic acid metabolism. However, it does depend on protein synthesis 2+ (85) and is accompanied by an increase in Ca -independent nitric oxide (NO) synthase activity in vascular smooth alt 24 muscle cells (59,86). Because NO is a potent vasodilator, it has been proposed that the aberrant production of nitric oxide (NO) by smooth muscle cells contributes to the alterations in vascular contractility after LPS exposure. Indeed, antagonists of NO vasodilation block the inhibitory effects of LPS on vascular contractility in vitro (59) and attenuate the hypotension observed during LPS exposure in vivo (87). These results support a role for endogenously derived NO in. ITS-induced hypotension. Several other endogenous, vasoactive mediators may be involved as well. These include lipid mediators such as platelet activating factor (88) and arachidonic acid metabolites (89,90,91), endothelium-derived reactive oxygen metabolites (92) , and cytokines such as tumor necrosis factor (93) and interleukin 1 (94,95). other circulatory alterations. In addition to alterations in cardiac output and vascular contractility, LPS exposure is often accompanied by a redistribution of blood flow (96,97). An increase in vascular permeability in several tissues , particularly in the lungs and gastrointestinal tract, also occurs. Several vasoactive agents have been implicated in the pathogenesis of these changes, including arachidonic acid metabolites, such as thromboxane A2 and sulfidopeptide leukotrienes, and platelet activating factor (see below). both relea XII, coagu coagi durir actii reduc clott fibr; Alth: and 1 cont, Undez (See 2.3.k fibri tubul 25 1.6.b Disseminated intravascular coagulation LPS can activate the coagulation system in vitro by both the extrinsic and intrinsic pathways through the release of tissue factor (98) and the activation of factor XII, respectively (99). Disseminated intravascular coagulation (DIC) resulting from. activation, of the coagulation system in viva occurs under certain conditions during' LPS exposure, indicating ‘that coagulation can. be activated by LPS in vivo (100). DIC is characterized by a reduction in circulating platelet numbers, depletion of clotting factors and plasma fibrinogen, and deposition of fibrin within the microcirculation of various tissues. Although its role in the pathogenesis of circulatory shock and lethality is not clear (101,102,103), DIC does appear to contribute to alterations in tissues during LPS exposure under certain circumstances, including injury to the lungs (see Section 1.6.c), to the liver (see Chapter 2, Section 2.3.b) and to the kidneys, where it can cause deposition of fibrin in the glomerular capillary bed with subsequent acute tubular necrosis (100). 1.6.c Multiple organ damage In addition to circulatory effects and DIC, alterations in numerous tissues are often associated with severe, systemic endotoxemia. in va pa th- ex; in. SL1 Va: pe: 181 in: SU: 311. 26 Gastrointestinal tract. A spectrum of changes in the gastrointestinal tract are associated with LPS exposure. These changes resemble ischemic bowel necrosis and are characterized by the detachment of the epithelial cells from the mucosal basement membrane , inf lammatory cel l infiltration, vascular congestion, and an increase in vascular permeability (104) . The mechanisms involved in the pathogenesis of tissue injury have not been clearly elucidated. However, results from several studies suggest that injury to the gastrointestinal tract during LPS exposure may be mediated by certain endogenous factors, including leukotrienes. Leukotrienes are synthesized from arachidonic acid via the 5-lipoxygenase pathway. These sulfidopeptide metabolites of arachidonic acid are potent vasoactive substances and promote increased vascular permeability in certain tissues (105). In the gut, leukotriene D4 and E4 receptor antagonists attenuated the increase in vascular permeability during LPS exposure, suggesting that this alteration is mediated by sulfidopeptide leukotrienes (106). PAF also appears to contribute to the pathogenesis of LPS-induced gastrointestinal injury. This is supported by studies showing that PAF concentrations in the gastrointestinal tract are increased during LPS exposure (107), that PAF receptor antagonists afford protection against the tissue injury (104), and that tissue injury resembling LPS—induced injury to the gut is produced by em to st me ha‘ al' ha) ma] vas phe bre thi the in} inc Cor C05 DUI fag lur fol 038 16a; 27 exogenously administrated PAF (108). PAF may mediate injury to the gastrointestinal tract in part through the stimulation of production of the arachidonic acid metabolite, thromboxane A2 (109). Lungs. Alterations in the lungs during LPS exposure have been well characterized. Pulmonary hemodynamic alterations during LPS exposure are biphasic in nature (80,81,82). The initial changes in the lung occur prior to 1 hour after LPS administration and are characterized by a marked increase in pulmonary artery pressure and pulmonary vascular resistance. This early, pulmonary hypertensive phase is associated with an increase in the lung lymph of breakdown products of thromboxane A2 (110) suggesting that this vasoactive arachidonic acid metabolite contributes to the early, pulmonary hypertension. The demonstration that inhibitors of arachidonic acid metabolism attenuate the increase in pulmonary artery pressure during this phase is consistent with this view (82). Inhibitors of the coagulation system markedly attenuate the increase in pulmonary vascular resistance, suggesting that clotting factors may contribute to the hemodyamic changes in the lungs following LPS exposure as well (111). The early' changes in. pulmonary artery' pressure are followed by an increase in vascular permeability which is associated with PMN infiltration, increased plasma protein leakage, and increased lung lymph flow. Although pulmonary t} ‘1 th- al pep pufi an) per all EEC) 28 artery pressure returns toward normal, it remains elevated throughout this second, permeability phase of lung injury. Unlike the gastrointestinal tract, leukotrienes do not appear to mediate the LPS-induced increase in vascular permeability in the lungs in the rat (106). Because catalase and depletion of circulating PMNs attenuate the increase in protein leakage and lung lymph flow, it has been proposed that PMN-derived oxygen radicals play an important role in the pathogenesis of LPS-induced vascular injury in the lungs (81,112). PAF also appears to contribute to the LPS-induced alterations in lung vascular permeability, since the permeability' changes are associated. with. an increase in pulmonary' PAF“ concentration and since PAF receptor antagonists attenuate the increase in pulmonary vascular permeability (113,114,115,116). In addition to alterations in pulmonary hemodynamics, alterations in pulmonary mechanics are observed during LPS exposure. Among these are a decrease in lung dynamic compliance, an increase in the alveolar/arterial oxygen gradient and an increase in alveolar dead space ventilation. Many of these alterations in pulmonary mechanics are attenuated by the 5-hydroxytryptamine (5-HT) receptor antagonist, ketanserin, suggesting that 5-HT contributes to their development (80). Activation of the clotting system by LPS may also contribute to alterations in pulmonary mechanics (117). An: resp nse (us 119, bolt; in differer. eff~cts Genzrat seczion) 29 .A number of factors have been shown to influence the response to LPS in vivo including species differences (118,119), route of exposure (120), dosing protocol (ie, bolus injection or infusion), and age (121). Also, large differences in potency are observed between LPS obtained from different species and strains of gram-negative bacteria. While the outcome of LPS exposure may vary under different conditions, it is clear that many of the adverse effects of LPS are dependent on endogenous mediators generated in response to LPS exposure (see previous section). Indeed, in many instances it seems that a network of host-derived. mediators 'may be required for the full manifestation of the response to LPS. The relationship between these mediators in the pathogenesis of LPS-induced effects on mammalian tissues, however, remains to be determined. The overall objective of this dissertation was to examine the mechanisms contributing to the pathogenesis of LPS-induced liver injury. Therefore, the following chapter will describe in somewhat greater detail the effects of LPS on the liver. Because many of the adverse effects of LPS appear to be mediated by endogenous factors, attention will be given to the possible contribution of certain host- derived mediators. CHAPTER 2 LPS-INDUCED LIVER INJURY 30 CC fr cel (Pi C611 Port hepa) Ce“tr. 31 2.1 Normal liver function and structure The liver performs a number of important functions. Among these are maintenance of blood glucose levels, storage of nutrients, fat metabolism, and synthesis and secretion of bile constituents and plasma proteins. It also plays an important. role in ‘the elimination and detoxification of xenobiotic agents. These functions are performed primarily by hepatic parenchymal cells, which in humans comprise approximately 80 % of the cells in the liver (122). The remaining liver cells consist of a heterogeneous population of cells ‘which includes fenestrated. vascular endothelial cells, resident vascular macrophages (Kupffer cells), fat storage cells (Ito cells), and bile duct epithelial cells. Collectively, these cells comprise the non-parenchymal cell fraction of the liver (123). The basic structural unit of the liver, where hepatic cells come in close contact with blood, is the liver lobule (Figure 2.1). The lobule consists of a region of parenchymal cells surrounding a central vein. It is bounded by several portal triads which consist of branches of the portal vein, hepatic artery' and. bile duct. Parenchymal cells in the central vein and portal triad regions of the liver lobule are referred to as centrilobular or periportal parenchymal 32 F.N QLJDWC ifii .Aomv uoso eHHc ecu mo mecoceuc oucfl eascoH ue>wa ecu scum meuumse use Assocm uocv 3:039:80 eaflc oucu ceueuoem we... mum ac penumecucwm eaflm .e>e0 ece> Heuueucw ecu ou aaeueEwuas oewuueo mu oce Cw“: cHe> Heuuceo e oucfi mcuomscwm ecu aouw mewumfie «00on .moflomscwm ecu cucuuz ocsou eue £9: mHHeo penumsx no .memecmouoee amazemetreuucw euuemec cemecmieuezeum msoueasc u 83 maaeo deflaecuopce oeueuumeceu >c oecwa eue cowcs mowomscwm owuemec cmsoucu £05 meu>oouemec no .mHHeo anacoceuem ofluemec mo mouoo ceezuec mae>euu cce Racy mueuue ofluemec oce A>mv sue> Heuuom ecu mo mecoceuc scum eascoH near: ecu mueuce oooam .eascoa ue>fin eca Um eugwm Figure 2.1 Sinusoid 33 at or 34 cells, respectively. Midzonal parenchymal cells are those cells located between the periportal and centrilobular regions. Blood flows through sinusoids from the periportal to the centrilobular region of the liver lobule between cords of parenchymal cells. In humans, the portal vein and hepatic artery provide approximately 75 and 25 % of the total hepatic blood supply, respectively (124). Fenestrations in endothelial cells lining the sinusoids facilitate contact between parenchymal cells and plasma constituents. Kupffer' cells are scattered. throughout 'the sinusoids. These fixed vascular macrophages remove senescent blood cells as well as debri and foreign substances, such as LPS, from the circulation. After flowing through the liver lobule, blood empties into the central vein and eventually into the inferior vena cava. Bile produced by parenchymal cells flows toward the periportal region of the liver lobule in bile canaliculi formed by tight junctions between adjacent parenchymal cells. Canaliculi empty into twanches of the bile duct located in the periportal region. 2.2 Morphologic and functional alterations in the liver after LPS Severe gram-negative bacterial infection results in a variety of changes in the liver (125). Because these changes resemble those produced by purified LPS (121,126,127), and because LPS has been shown to be an important contributing <- s1. 35 factor to the manifestation of many of the pathophysiologic effects associated with gram-negative infections (see above), it seems likely that LPS contributes to the pathogenesis of liver injury associated with gram-negative bacterial infections. 2.2.a Morphologic alterations Marked morphologic changes occur in the liver following exposure to LPS. These include changes in the sinusoids as well as in parenchymal cells (126,127,128). Sinusoidal changes. Morphologic alterations in Kupffer cells are among the earliest changes in the hepatic sinusoids. Initially, these cells appear moderately swollen and contain an increased number of cytoplasmic lysosomal granules and phagocytic vacuoles. Platelets and PMNs are occasionally observed within phagocytic vacuoles of Kupffer cells. Dilation of Kupffer cell endoplasmic reticulum and damage to the plasma and nuclear membranes are also apparent. The early changes in Kupffer cell morphology are accompanied by injury to endothelial cells and by the appearance of fibrin clumps and platelet thrombi in the sinusoids. The 'platelets often_ appear’ deformed and show signs of degranulation. Large numbers of PMNs also accumulate in the sinusoids during LPS exposure. These morphologic alterations in sinusoids, which are evident as 36 early as 30 min following LPS administration, become augmented by 4 hr and often persist for at least 24 hr. Changes in parenchymal cells. Morphologic changes in liver parenchymal cells are associated with the sinusoidal changes. Initially, subtle alterations occur in the mitochondria and endoplasmic reticulum. These changes, which are characterized. primarily’ by ‘moderate: dilation of the organelles, occur prior to 1 hr and become progessively worse after 1 hr of LPS exposure. Other early morphologic changes in parenchymal cells include swelling of the microvilli on the sinusoidal border and dilation of bile canaliculi. Subsequent changes in parenchymal cells become apparent betweem 4 and 24 hr after LPS administration and are characterized by signs of hepatocellular degeneration and necrosis. Lesions are multifocal, frequently involve parenchymal cells in the midzonal region of the liver lobule, and are often infiltrated by neutrophils. Increases in plasma activities of liver-specific enzymes, such as AST and ALT, occur between 4 and 8 hrs after LPS exposure, suggesting membrane damage to parenchymal cells (102,129) . The morphologic changes in the liver during LPS exposure are more marked in older animals, indicating that the sensitivity of the liver to LPS increases with age (121). 37 2.2.b Functional alterations Glycogen content in parenchymal cells was markedly decreased 30 min after LPS exposure and remained reduced for 24 hr (126). This is probably due to both a decrease in gluconeogenesis and to an increase in glycogenolysis (130,131). In contrast, parenchymal cell fat content gradually increased over the 24 hr exposure period. Increased hepatic lipogenesis as well as inhibition of parenchymal cell protein synthesis may contribute to this effect (132,133). These changes indicate that LPS induces functional and metabolic alterations in parenchymal cells that are rapid in onset and persistent. Alterations in parenchymal cell protein synthesis are observed during LPS exposure. Protein synthesis can be either increased or decreased following LPS administration depending on the specific protein measured. However, the net effect of LPS on hepatic parenchymal cells in vitro is to decrease protein synthesis (133). Liver cytochrome P-450 content is reduced 24 hr after LPS administration in mice and rats (134,135,136). This is accompanied by a decrease in hepatic mixed function oxidase activity. Other functional changes int the liver include cholestasis (23) and circulatory alterations such as decreased hepatic blood flow (137,138) and portal hypertension (139,140). 2.3 Mech Wh. U611 Ch: of LPS 2 the lid indireci l Daren aPPEE fact< rais. alter 38 2.3 Mechanisms of LPS-induced liver injury While the hepatic alterations produced by LPS have been well characterized, much remains unknown about the mechanism of LPS hepatotoxicity. Evidence suggests that alterations in the liver during LPS exposure may be due to both direct and indirect (host-mediated) effects of LPS on the liver. 2.3.a Direct effects of LPS on the liver After intravenous administration of large doses of LPS, the majority of liver-associated LPS is found in Kupffer cells (20,21,22,23). However, small quantities are detected in other liver cells, including endothelial cells and parenchymal cells. Although. many' of the effects. of LPS appear to be mediated indirectly by endogenously derived factors, LPS binds to liver parenchymal cells in vitro (79), raising the possibility that, direct, LPS-induced alterations may also contribute to some hepatic alterations. LPS-induced cholestasis. Among the hepatic alterations which appear to be directly mediated by LPS is cholestasis (23). Perfusion of isolated livers with concentrations of LPS greater than 5 ug/ml results in cholestasis (141). This is characterized by a decrease in bile flow as well as by a decrease in bile clearance of the dyes, sulfobromophthalein (BSP) and indocyanine green (141,142). LPS also caused a decrease in the perfusate flow, suggesting that the cholesta alterati observe< mechani' not pla were us observe aminotl were h' foma (142)‘ inhib~' Canaz appea LPS 0 39 cholestatic effect and decreased dye clearance were due to alterations in tissue perfusion. However, no cholestasis was observed when flow was reduced to a similar degree mechanically in the absence of LPS (141). Blood elements did not play a role in this effect since buffer-perfused livers were used. Furthermore, because decreases in bile flow were observed in the absence of significant changes in perfusate aminotransferase activity, the cholestatic effects of LPS were not due to hepatocellular injury. Bile is produced by liver parenchymal cells. Bile production can be divided into two fractions, bile salt- dependent. and. bile salt-independent fractions (143). The bile salt-dependent fraction is formed by an osmotic gradient established in the bile canaliculus by the active secretion of bile acids. In contrast, active transport of Na+ into the bile canaliculus by parenchymal cell Na+/K+— ATPase presumably mediates the formation of the bile salt- independent fraction. Because LPS reduces bile flow without affecting secretion of bile salts, the cholestatic property of LPS is thought to be due to an alteration in the formation. of the ‘bile salt-independent fraction of bile (142) . This is supported by evidence indicating that LPS inhibits Na+/K+-ATPase activity in partially purified bile canalicular membranes (144). Thus, LPS-induced cholestasis appears to be due, at least in part, to a direct effect of LPS on parenchymal cell bile formation. 4O LPS-induced hepatic lipid metabolism. Alterations in hepatic lipid. metabolism. may also result from. a direct effect of LPS on the parenchymal cell. Exposure of isolated parenchymal cells to low concentrations of LPS results in an increase in cellular neutral lipid content (132). Increased secretion of neutral lipids is also observed following exposure of parenchymal cells to LPS in vitro. These changes are accompanied. by’ an increase in 'the incorporation of radiolabeled acetate into neutral lipids, suggesting that they are due to an increase in de novo lipid synthesis. Thus, direct, LPS-induced alterations in parenchymal cell lipid metabolism may contribute to the accumulation of lipids in the liver as well as to the hyperlipidemia associated with LPS exposure in vivo. 2.3.b Indirect (host-mediated) effects of LPS on the liver Like LPS-induced injury to other tissues (see Chapter 1), several host cells and endogenous mediators may play a role in the pathogenesis of LPS hepatotoxicity. These include circulating PMNs, hepatic fixed macrophages, cytokines, clotting factors, and arachidonic acid metabolites. Role of PMNs. PMNs originate in the bone marrow from promyelocytic stem cells. After maturation, which occurs primarily in the bone marrow of normal individuals, these cells lymph the nunbe numbe agail nega‘ defe illr bacl of at of in SE ir ti CO at ECG and dire! SPEQi S the 41 cells are released into the blood where they join lymphocytes, monocytes, eosinophils and basophils to form the circulating blood leukocyte population. Circulating numbers of PMNs normally range from 20-30% of the total number of white blood cell in rats to 40-75% in humans. PMNs play an important role in the defense of the host against infection by certain pathogens including gram- negative bacteria. The importance of PMNs in the host’s defense against gram-negative bacterial infections is illustrated by studies which show that inhibition of bacterial growth is temporally related to the accumulation of PMNs at the site of infection, and that bacterial growth at these sites is uninhibited in animals previously depleted of circulating PMNs (145). Also, patients with deficiencies in PMN numbers or function exhibit an increased susceptibility to infection by bacteria (146,147,148). The response of PMNs to gram-negative bacterial infections has been studied extensively, particularly in tissue such as the skin (145). The PMN response involves a complex series of events which include accumulation of PMNs at the site of infection followed by microbial killing. PMN accumulation can be subdivided into chemotaxis, adherence, and diapedesis (Figure 2.2). Chemotaxis is the process of directed cell movement where cells such as PMNs migrate to specific locations along concentration gradients established by specific, soluble chemotactic factors. For example, in the case of gram-negative bacterial infections, chemotactic 42 N.N e39... .nmg EmuceoHOOMOuE mcwce>cfl ecu HHflx cce Aemouaoomece masoce ou TL asuuuumueucfl ecu aw oeum>wuoe eeooec mzznw .7; Amwmeoemeflcv EnuuuumueuCM ecu ou cefiza Meadow?» ecu Bonn counueuocé one H3 Um: mHHeo Heuaecuooce Meadow“; ou eoceuecoe .3; Amuxeuoeecov coflueumfia :eo oeuoeuuc ecsHocw cowcs muce>e mo meuuem e xc couuoeucH mo euum ecu um ceuusuoeu eue mzzm mcwueadoufio .euueuoec e>uumoecnamuo ac :oHuoeucfl ou emcommeu £sz Hucmouusez m.~ enamwm uc____x 6:223 e>=emec -ESG V E::_uw._euc_ ceE:_ ..m_:omm> we 2:9“. 44 factor concentration gradients direct PMNs toward the site of infection. Diapedesis is the process by which PMNs migrate from the vascular lumen to the interstitium. This is thought to require adherence of PMNs to the vascular endothelium. The microbicidal activity of PMNs can be subdivided into phagocytosis and cytotoxicity. Upon arrival at sites of infection, PMNs become activated and begin to engulf bacteria by the processes of phagocytosis. Bacteria contained within phagocytic vacuoles are subsequently killed and degraded by cytotoxic factors released into the phagocytic vacuole. LPS appears to play an important role in eliciting the PMN response to infection by gram-negative bacteria. For example, LPS likely contributes to the accumulation of PMNs at sites of infection. This is supported by observations that polymyxin B and anti-LPS antibodies attenuate PMN accumulation and that PMN accumulation can be induced by injection of purified LPS (41,42,145). LPS is not directly chemotactic for PMNs in vitro. However, it can stimulate the production by host tissues of chemotactic factors such as the cytokine, interleukin 8 (149,150) and the complement factor, C5a (1). This suggests that LPS-induced production of endogenous chemotactic factors may“ contribute to PMN accumulation at the site of infection by gram-negative bacteria. LPS may also contribute to PMN diapedesis during gram- negative bacterial infections by facilitating PMN adherence 45 to endothelial cells. This is supported by evidence indicating that LPS exposure induces the expression of PMN adhesion molecules on the surface of endothelial cells in vitro and increases the adherence of PMNs to cultured endothelial cell monolayers (53,54). LPS-induced production of the cytokines, IL-1 and TNF-alpha, may contribute to the adherence of PMNs to vascular endothelium. as 'well (see below). Thus, LPS appears to mediate the accumulation of PMNs at sites of gram-negative bacterial infections by promoting adherence to the endothelium as well as by inducing production of chemotactic factors. LPS has been shown to enhance the release of cytotoxic agents from activated PMNs in vitro in a serum-dependent manner (151). This raises the possibility that, in addition to its role in PMN accumulation, LPS may promote killing of gram-negative bacteria at sites of infection by inducing the activation of PMNs to produce microbicidal agents. Following activation, PMNs are capable of releasing a variety of bactericidal agents intracellularly into phagocytic vacuoles or into the extracellular mileu. Among these agents are highly reactive oxygen metabolites, such as superoxide anion (02'). Activation of PMNs is characterized in part by an increase in hexose monophosphate shunt activity and a burst of nonmitochondrial oxygen consumption (152,153). These activities reflect the activation of NADPH oxidase. This plasma membrane-bound enzyme utilizes NADPH derived from metabolism of glucose via the hexose ix 46 monophosphate shunt pathway to catalyze a one electron transfer to molecular oxygen to form 02'. 02" is released into the phagosome where it contributes to bacterial killing or extracellularly, where it may injure host tissue (see below). While 02' may possess direct antimicrobial activities, its cytotoxic effects are thought to be due to the generation of more potent oxidants, such as hydrogen peroxide (H202) and hydroxyl radical. H202 can be formed from 02' by either a spontaneous or enzyme-catalyzed dismutation process. “OH is thought to be generated from 02' and H202 by an iron-catalyzed pathway. These oxygen metabolites can initiate peroxidation of lipids resulting in membrane damage. They may also cause disruption of cell function by inducing alterations in protein and DNA. PMN activation is also accompanied by the fusion of lysosomal granules with phagocytic vacuoles. This process is refered to as degranulation and results in the release of lysosomal contents into the phagosome. Lysosomes contain a variety of proteases which. may contribute to the microbicidal potential of PMNs. In addition, lysosomal enzymes, such as lysozyme, digest the cell wall of certain bacteria, whereas myeloperoxidase utilizes NADPH oxidase- generated 02' and chloride anions to catalyze the production of the powerful oxidant, hypochlorous acid, and chloramines. Other antibacterial agents released from the lysosomes include lactoferin, which chelates iron, and 60 kDa LPS- 47 binding protein, which can permeabilize gram-negative bacteria. As indicated above, release of cytotoxic mediators from activated PMNs is not restricted to the phagosome. Extracellular release of these agents has been shown to occur in vitro under certain circumstances, and evidence from numerous studies suggests that this can result in damage to host cells in vitro and in vivo (153,154). Instances in which PMNs have been implicated in the pathogenesis of injury to host tissue include myocardial ischemia/reperfusion injury (155), complement-mediated lung injury (156), injury to the gastrointestinal tract induced by nonsteroidal anti-inflammatory drugs (156), certain instances of chemically induced liver injury (158), and liver injury following hypovolumic shock (159). In addition to these examples, PMNs mediate ‘tissue injury following infection by gram-negative bacteria. For example, PMN depletion affords protection against injury’ to the microvasculature of the skin following intradermal injection of live E. coli (160). Similarly, PMN depletion attenuates the injury to the blood-brain barrier in gram-negative bacterial meningitis (40). This presents a paradox with respect to the role of PMNs at sites of gram-negative bacterial infections in that while they limit the growth of bacteria, they contribute: to the jpathogenesis of tissue injury associated with gram-negative bacterial infections. 48 In addition to their role in the pathogenesis of tissue injury at sites of gram-negative bacterial infections, PMNs mediate tissue injury in the lungs after intravenous administration of purified LPS (112, Chapter 1). Thus, exposure to LPS in the absence of gram-negative infection results in tissue injury which is mediated by PMNs. Because PMNs accumulate in the liver and are associated with foci of hepatocellular necrosis during exposure to LPS, it seems possible that these cells contribute to the pathogenesis of LPS hepatotoxicity as well. Results from one study which showed that heat killed Pseudomonas aeruginosa injure primary cultures of hepatic parenchymal cells in a PMN- dependent manner support this hypothesis (161). However, the role of PMNs in LPS hepatotoxicity in vivo remains to be determined. Role of Kupffer cells. Kupffer cells are fixed, vascular macrophages which are scattered throughout the hepatic sinusoids. Like other resident tissue macrophages, Kupffer cells originate from promyloblast stems cells in the bone marrow (162). These precursor stem cells differentiate into monocytes in the bone marrow and are released into the circulation. Monocytes may circulate for several days before becoming sequestered in various tissues including the brain, skin, kidneys, lungs, spleen, and liver, where they undergo terminal differentiation to become tissue macrophages. Following differentiation, macrophages may reside in tissues 49 for extended periods of time. This contrasts with PMNs which are short-lived after emigrating from the vasculature. Macrophages are phagocytic cells and, like PMNs, they contribute to the nonspecific immune response by phagocytosing and killing invading pathogenic microorganisms (162,163). The macrophage activation process and killing mechanisms are similar to those described for PMNs and include increased nonmitochondrial respiration and hexose monophosphate shunt activity, NADPH oxidase-catalyzed 02' production, and lysosomal enzyme release. Nitric oxide and various cytokines, including TNF-alpha, are also produced by activated macrophages under certain conditions. These agents possess tumoricidal activities in vitro and may contribute to the capacity of macrophages to kill neoplastic cells in vivo. In addition to their role in nonspecific immunity, macrophages 'perform. important. accessory functions in specific immune responses by processing and presenting antigens to B and T lymphocytes as well as by releasing lymphocyte mitogenic and activating factors, including the cytokine, IL-l. It has been proposed that an important function of Kupffer cells is to filter the portal circulation of foreign substances and neoplastic cells derived from the gastrointestinal tract. This function is thought to be particularly important in preventing LPS originating from the gut from reaching the systemic circulation. en t1”. p5 m 50 Although Kupffer cells may protect against systemic endotoxemia, evidence from several studies suggests that these cells also contribute to the manifestation of certain pathophysiologic alterations associated with LPS exposure including lethality. Several strains of mice are exceptionally hyporesponsive to the lethal effects of LPS. Among these is the C3H/HeJ mouse strain (164). Although the mechanism for this has not been clearly elucidated, genetic studies suggest that a defect in a single, autosomal dominant gene locus on chromosome 4, which is referred to as the lps locus, may be involved (165). In any case, results from several studies in vitro suggest that alterations in macrophage function may contribute to the altered LPS responsiveness exhibited by C3H/HeJ mice. For example, macrophages isolated from C3H/HeJ mice are defective in the release of several mediators following LPS exposure in vitro. Because these mediators, which include certain arachidonic acid metabolites (166), TNF-alpha (167) and nitric oxide (168), have been implicated in the pathophysiologic alterations associated with LPS exposure in vivo, this suggests that defects in macrophage function may contribute to the reduced responsiveness of LPS-resistant mice and that normal macrophage functions may be required for the manifestation of many LPS-induced pathophysiologic alterations. A defect in phagocytic activity has also been observed by macrophages elicited from C3H/HeJ mice (169). In vivo, 51 fewer numbers of Kupffer cells exhibited phagocytic activity in C3H/HeJ mice compared to LPS-sensitive mice strains (170). This decrease was not due to a reduction in absolute numbers of Kupffer cells since histological stains specific for Kupffer cells revealed that their numbers in liver sections from sensitive and resistant mice were similar. Furthermore, phagocytic activity by Kupffer cells in the resistant mouse strain occured at a lower rate than in LPS sensitive mice strains, indicating that the efficiency of Kupffer cell function in the resistant mice was reduced as well. Similar differences in Kupffer cell phagocytic activity were observed in vivo among animal species (118). A large variation in LPS sensitivity also exists among species. For example, guinea pigs are nearly 10 times more sensitive to the lethal effects of LPS than are rats. This species difference in LPS-induced lethality correlated with the number of Kupffer cells in the liver exhibiting phagocytic activity. Thus, guinea pigs had greater numbers of phagocytic Kupffer cells than rats. Furthermore, phagocytosis by Kupffer cells in less sensitive species occured at lower rates. It was proposed from these studies and from the studies in C3H/HeJ mice that sensitivity to LPS was dependent on Kupffer cell function. Consistent with this is evidence indicating that substances which increase Kupffer cell activity also increase LPS lethality (171,172). Conversely, substances which decrease Kupffer cell activity 52 protect against LPS-induced lethality. Interestingly, pretreatment with Bacillus Calmette Guerin, which markedly increases macrophage activity in normal mice, increased the sensitivity of LPS-resistant mice to LPS (173). These results are consistent with the role of hepatic macrophages in the lethal effects of LPS. Kupffer cells may play a role in the pathogenesis of LPS hepatotoxicity as well. This is supported by studies showing that the increase in numbers of activated macrophages in the liver following pretreatment with Corynebacterium parvum is associated with a dramatic increase in the hepatotoxicity of LPS (174,175). Also, inhibition of Kupffer cell function with methyl palmitate affords protection against LPS hepatotoxicity (175). Because activated Kupffer cells can release several cytotoxic mediators (162), it has been suggested that damage to hepatic parenchymal cells following LPS exposure may be mediated by Kupffer cell-derived substances. Further support for the role of Kupffer cells in LPS hepatotoxicity has been provided by studies in vitro using cocultures of primary liver parenchymal cells and Kupffer cells. Inhibition of protein synthesis by and cytotoxicity to cultured liver parenchymal cells induced by exposure to LPS occurs only in the presence of Kupffer cells (177). These effects are dependent on L-arginine, are associated with an increase in nitric oxide (NO) production in parenchymal cells, and are prevented by inhibitors of NO 53 synthesis (177). Because similar effects are observed by liver parenchymal cells incubated with supernatant from LPS- treated Kupffer cells, it was concluded that LPS induced the release of soluble factors from Kupffer cells which increased the production of NO by liver parenchymal cells (178,179). The increased production of NO subsequently resulted in decreased protein synthesis in and cytotoxicity to liver parenchymal cells. Although the Kupffer cell- derived factor has not been identified, evidence suggests that these effects may be mediated at least in part by cytokines released from LPS-treated Kupffer cells (179). Although some studies support the role of Kupffer cells in the pathogenesis of LPS hepatotoxicity, results from other studies argue against this hypothesis. For example, NO synthesis inhibitors do not afford protection against liver injury following exposure to LPS in vivo suggesting that, whereas NO mediates injury to hepatic parenchymal cells in vitro, NO production does not contribute to parenchymal cell damage in vivo (180). Indeed, inhibition of NO synthesis actually enhanced LPS hepatotoxicity. Furthermore, intravenous administration to mice of liposome-encapsulated dichloromethylene diphosphonate, which effectively eliminates the Kupffer cell population of the liver, did not protect against LPS hepatotoxicity (181). Thus, despite evidence supporting the role of Kupffer cells in LPS-induced liver injury, the contribution of Kupffer cells remains unclear. 54 Role of cytokines. Interleukins, interferons (IFNs) , and TNF are examples of soluble mediators comprising cytokines. These proteins are produced and secreted by a wide variety of cell types including monocytes and macrophages, lymphocytes, endothelial and vascular smooth muscle cells, fibroblasts, and hepatocytes (182). .A number of important physiologic functions have been attributed to cytokines (182,183). For example, they form a complex communication network among cells of the hematopoietic system in which they control the differentiation and maturation of blood cells from precusor stem cells in the bone marrow (184) . Among the cytokines that have been implicated in the process of hematopoiesis are interleukins such as IL-3, IL-4, IL-5 and IL—7, and colony' stimulating' factors. Cytokines. also» contribute 'to both the specific and nonspecific immune responses by regulating lymphocyte proliferation, differentiation, and activation and by increasing the resistance to viral infection and neoplasia (185, 186, 187,188). Cytokines with immune system functions include various interleukins (most notably IL-1, IL-2, and IL-6), TNF, and the interferons. Following tissue injury or infection, a shift in protein synthesis by the liver occurs which is characterized by a decrease in synthesis of certain plasma proteins such as albumin, and an increase in synthesis of the class of plasma proteins termed acute phase proteins (182) . These proteins contribute to the regulation of the inflammatory 55 response and their synthesis is mediated by certain cytokines, including IL-1, IL-6 and TNF-alpha. This cytokine-induced alteration in plasma protein synthesis by the liver is thought to play a role in the acute phase response to gram-negative bacterial infection and LPS exposure. Several cytokines, particularly IL-1, IL-6, IL-8, and TNF-alpha, appear to play an important regulatory role at sites of inflammation (188). For example, because IL-1 and TNF-alpha both induce procoagulant activity in cultured vascular endothelial cells, these cytokines may contribute to the initiation of coagulation at sites of inflammation in vivo (189). Furthermore, IL-1 and TNF-alpha induce the expression. of PMN’ adhesion 1molecules on the surface of vascular endothelial cells, thus raising the possiblity that these cytokines contribute to the accumulation of PMNs at sites of inflammation (53,190,191). TNF-alpha induces the expression of CD11b/CD18 complex on the surface of PMNs as well (192). In as much as this complex mediates PMN adhesion to surfaces, TNF-alpha may mediate PMN accumulation at sites of inflammation by affecting PMNs in addition to affecting vascular endothelium. Finally, these cytokines stimulate the production of the potent cytokine chemotactic factor, IL-8, from host cells suggesting that they may promote PMN accumulation at sites of inflammation by inducing the production of endogenous chemotactic factors (150,193). 56 While cytokines perform beneficial functions in the hematopoietic and immune systems as well as during inflammation, evidence from numerous studies suggests that they may also mediate tissue injury under certain circumstances. Among the cytokines implicated in the pathogenesis of tissue injury is TNF-alpha. A wide variety of pathophysiologic alterations have been attributed to TNF- alpha. Exposure to recombinant TNF-alpha can produce severe hypotension, shock, and injury to various organs including the lungs and. gastrointestinal tract (194,195). Pathophysiologic conditions in which TNF-alpha has been implicated include neurologic alterations associated with cerebral malaria (196), injury to the liver and lungs following hepatic ischemia/reperfusion (197) and lung injury following intestinal ischemia/reperfusion (17). Exposure to LPS is associated with a transient increase in circulating TNF-alpha concentration which peaks approximately 1-2 hr after LPS administration (198). Results from several studies suggest that TNF-alpha may contribute to the pathophysiologic alterations accompanying LPS exposure. For example, neutralization of TNF-alpha with specific TNF antiserum affords protection againsts many LPS- induced alterations, including hypotension (93) , lethality (199), and lung injury (17,200). Results from one study suggest that TNF-alpha may contribute to the pathogenesis of LPS hepatotoxicity as well (201). 57 The mechanism by which TNF-alpha mediates tissue injury is not known. However, lung injury that follows hepatic ischemia/reperfusion is accompanied by accumulation of PMNs in. the lungs, and neutralization of TNF-alpha with TNF antiserum attenuates both the pulmonary PMN infiltration as well as lung injury, suggesting that TNF-alpha-induced PMN accumulation may contribute to the pathogenesis (197). This view is supported by the observation that PMN depletion attenuates lung injury induced by exposure to recombinant TNF-alpha (202). Thus, the ability of TNF-alpha to promote PMN accumulation in tissues may predispose the tissue to PMN-dependent injury. Although PMNs accumulate in the liver during LPS exposure, the relationship between hepatic PMN accumulation and TNF-alpha in the pathogenesis of LPS hepatotoxicity remains to be elucidated. In addition to TNF-alpha, IL-1 and IL-6 may contribute to certain pathophysiologic alterations that accompany LPS exposure. Circulating levels of IL-1 and IL-6 are markedly elevated 2-6 hr after LPS exposure (198), and IL-1 receptor antagonists (94) and antiserum to IL-6 (202a) attenuated the lethality associated with exposure to purified LPS or infection by gram-negative bacteria, respectively. Thus, these cytokines appear to contribute to the pathogenesis of severe shock induced by LPS or gram-negative bacteria. They have also been implicated in the pathogenesis of tissue injury after LPS exposure in certain instances (202b, 58 202c,248). However, their role in LPS-induced liver injury, if any, remains to be determined. Role of the coagulation system. It has been proposed that liver injury during LPS exposure is mediated by circulatory disturbances resulting from activation of the clotting system. Circulating fibrinogen concentration falls more than. 90 % 3 hr after LPS administration to rats (203,204). This is accompanied by the appearance of fibrin clumps in the microcirculation of the liver. The role of the coagulation system in the pathogenesis of LPS hepatotoxicity has been implicated by evidence from several studies. For example, infusion of thrombin into the portal vein results in morphologic changes in the liver which resemble those produced by portal venous infusion of LPS. These changes include fibrin deposition and PMN accumulation in the sinusoids and hepatocellular necrosis (129) . An evaluation of the time-course of changes in circulating transaminase activity indicated that the onset of liver injury was more rapid following thrombin infusion compared to LPS infusion. This may be due to the time required for generation of endogenous thrombin following activation of the coagulation system by LPS. The role of thrombin and the coagulation system in the pathogenesis of LPS hepatotoxicity was strengthened by evidence from studies with heparin. Heparin is a naturally occuring anticoagulant which is widely distributed 59 throughout animal tissues. This complex proteoglycan is synthesized primarily by tissue mast cells and consists of mucopolysaccharide chains of varying length and number which are linked to a core protein (205). The anticoagulant properties of heparin are related to the mucopolysaccharide component. of ‘the.:molecule, which inhibits the ‘thrombin- catalyzed conversion of fibrinogen to fibrin (206). This inhibitory effect is due to a heparin-induced increase in affinity of antithrombin III for thrombin (207). Results from several studies indicate that pretreatment with heparin prevents the fall in circulating fibrinogen concentrations after LPS exposure (203,204) and affords protection against the hepatotoxic effects of LPS (102). This protective effect is not due to an enhanced rate of clearance of LPS from the circulation in heparin-treated animals (208). These results are consistent with the hypothesis that the coagulation system plays a role in the pathogenesis of liver injury following LPS exposure. Role of arachidonic acid metabolites. A number of products of arachidonic acid are released from cells treated with LPS in vitro (Table 1.1). Also, exposure to LPS in vivo results in marked increases in the circulating concentrations of several arachidonic acid metabolites, including thromboxane A2 and several prostaglandins and leukotrienes (209). Arachidonic acid, or cis-5,8,11,14- eicosatetraenoic acid, is a 20-carbon, polyunsaturated fatty 6O acid containing 4 double bonds. It is an essential fatty acid since it can only be obtained from dietary sources or by synthesis from other essential fatty acids such as linoleic acid. Arachidonic acid is incorporated into membranes where it is esterified to phospholipids. Under certain conditions, such as LPS exposure, it is cleaved from membrane phospholipids by phospholipase A2 and subsequently metabolized by two major pathways. The cyclooxygenase pathway generates prostaglandins and thromboxane whereas the 5-lipoxygenase pathway generates leukotrienes. These arachidonic acid products have potent, biological activities and, although they are important mediators of a variety of physiologic processes, aberrant production can result in pathophysiologic alterations in tissues (210). Numerous alterations associated with LPS exposure have been attributed to arachidonic acid metabolite (Table 1.1). Evidence from several studies suggests that arachidonic acid metabolites may contribute to the pathogenesis of LPS- induced liver injury. For example, dexamethasone, which inhibits phospholipase A2, afforded protection against LPS hepatotoxicity in galactosamine-sensitized mice (211). Rats maintained on a diet deficient in essential fatty acids were also resistant to the hepatotoxic effects of LPS (212). These rats were depleted of arachidonic acid, suggesting that the protective effect was due to the absence of arachidonic acid metabolites (213). This view is supported by evidence that essential fatty acid-deficiency prevents 61 the LPS-induced rise in circulating concentrations of certain metabolites of arachidonic acid (214). Among the arachidonic acid metabolites that have been implicated in the pathogenesis of LPS hepatotoxicity is the potent vasoconstrictor and platelet aggregator, thromboxane A2. Pretreatment with the thromboxane synthase inhibitors, imidazole and 7-(1-imidazolyl)-heptanoic acid, markedly attenuated the rise in circulating concentrations of thromboxane 32 (a stable breakdown product of thromboxane A2) as well as the increase in transaminase activity that followed LPS administration, suggesting that thromboxane A2 may contribute to LPS hepatotoxicity (215) . In contrast, liver injury following' administration. of LPS to animals pretreated with galactosamine, which markedly increases the sensitivity of animals to LPS, was not attenuated by either aspirin or ibuprofen (211). Since these cyclooxygenase inhibitors block thromboxane sythesis, this suggests that thromboxane A2 does not play a role in LPS hepatotoxicity in mice sensitized to LPS by galactosamine. It also raises the concern that different mechanisms may be involved in the pathogenesis of LPS-induced liver injury in nonsensitized and galactosamine-sensitized animals. Although thromboxane does not appear to play a role in LPS-induced liver injury in galactosamine-sensitized mice, leukotrienes, particularly leukotriene D4, do (211). This is indicated by studies with leukotriene synthesis inhibitors, which attenuated liver injury in LPS-treated animals, and by 62 results with exogenously administered leukotriene D, which produced liver injury resembling LPS. The role of leukotrienes in LPS-treated animals in the absence of galactosamine pretreatment remains to be determined. 2.4 Overall aim It is clear from the evidence presented in this chapter and in the previous chapter that a broad spectrum of pathophysiologic alterations accompany LPS exposure. Although some of these may be mediated by the direct effect of LPS on tissues, many appear to be mediated by host- derived factors. Indeed, in many instances, multiple endogenous mediators appear to be involved (Table 1.1). This seems to be the case with certain LPS-induced alterations in the liver where factors derived from Kupffer cells, arachidonic acid and the coagulation system have been implicated in the pathogenesis. The association of PMNs with foci of hepatocellular necrosis raise the possibility that factors released from these phagocytic cells may contribute to injury to the liver as well. It is possible that these mediators act by separate, parallel mechanisms. That is, alterations in tissues after LPS exposure result from the sum of the effects of several different mediators. Alternatively, it seems possible that the pathophysiologic alterations associated with LPS exposure are the result of complex interactions among 63 endogenous mediators. In other words, each mediator by itself is not sufficient to. cause the alteration. Such interactions might involve the potentiatian of the effects of one mediator by a second mediator. Alternatively, one mediator may stimulate the production or activation of a second mediator. Thus, a series or cascade of interactions among several mediators may be required for the manifestation of LPS-induced alterations in vivo. The overall objective of the following studies was to examine the involvement of several endogenous mediators in LPS hepatotoxicity, including circulating PMNs, TNF-alpha, and coagulation factors. Emphasis will be on the possible interactions among these various mediators in the pathogenesis of liver injury. 2.5 Specific aims 2.5.a Specific aim 1 Morphologic changes in the liver are characterized by the infiltration of large numbers of PMNs which are associated with foci of hepatocellular necrosis. Because PMNs have been implicated in the pathogenesis of tissue injury during LPS exposure in certain instances, this raises the possibility that these phagocytic cells contribute to the pathogenesis of LPS hepatotoxicity. Studies were performed to test this hypothesis. The results from these studies are presented in Chapter 3. 64 2.5.b Specific aim 2 TNF-alpha has been implicated in the pathogenesis of LPS hepatotoxicity. However, the mechanism by which this cytokine mediates liver injury is not known. TNF-alpha has several effects on PMN function which could contribute to the pathogenesis of LPS-induced liver injury. For example, TNF-alpha increases the adherence of PMNs to vascular endothelial cells and enhances the release of cytotoxic substances from activated PMNs in vitro. Thus, TNF-alpha may contribute to LPS hepatotoxicity by a PMN-dependent mechanism. Studies were performed to test this hypothesis. The results from these studies are presented in Chapter 4. 2.5.c Specific aim 3 Activation of the coagulation system with deposition of fibrin in the microvasculature of the liver occurs after LPS administration, and anticoagulants afford protection against LPS hepatotoxicity. This suggests that clotting factors may contribute to the pathogenesis of LPS hepatotoxicity. The objective of studies presented in Chapter 5 was to test the hypothesis that the coagulation system contributes to LPS- induced liver injury by a mechanism which is dependent on circulating fibrinogen. Interactions between PMNs, TNF- alpha, and the coagulation system were examined. Results from studies in this dissertation contribute to the elucidation of the mechanisms of LPS hepatotoxicity. 65 Since LPS has been implicated in the pathophysiologic effects associated with gram-negative bacterial infections, and since alterations in the liver may play an important role in the high mortality associated with severe, systemic, gram-negative bacterial infections, an understanding of the mechanism of LPS-induced liver injury may facilitate the development of therapeutic interventions that could improve survival from infections by these bacteria. CHAPTER 3 ROLE OF NEUTROPHILS IN LPS-INDUCED LIVER INJURY 66 67 3.1 Abstract PMN infiltration is an early occurrence in the liver following exposure to hepatotoxic doses of LPS. The purpose of this study was to test the hypothesis that PMNs contribute to the pathogenesis of LPS hepatotoxicity. The immunoglobulin (Ig) fraction from serum of rabbits immunized with rat PMNs (anti-PMN Ig) was administered iv to rats 18 and 6 hr prior to exposure to an hepatotoxic dose of LPS. This protocol caused a >95 % reduction in circulating PMNs, which was maintained for the duration of the study. The Ig fraction from nonimmunized rabbits was used as a control (control Ig) . Rats pretreated with control Ig exhibited a marked increase in the number of PMNs in the liver 1.5 hr after LPS exposure. This increase in hepatic PMNs was significantly reduced by pretreatment with anti-PMN Ig. Marked elevations in both ALT and AST activities were observed in plasma from control Ig-treated rats 6 hr after iv administration of LPS. The response to LPS was greatly attenuated in animals receiving anti-PMN Ig. Pretreatment of rats with Igs to rat lymphocytes (LCs) reduced numbers of circulating LCs but did not afford protection against the hepatotoxic effects of LPS. These results suggest that PMNs contribute to the pathogenesis of LPS hepatotoxicity. 68 3.2 Introduction An array of morphologic and functional alterations in the liver have been attributed to LPS (216). Early morphologic changes occur in the sinusoids and include Kupffer cell swelling, formation of platelet thrombi, fibrin clumping and PMN infiltration (126,127). These changes become evident within 1 hr after LPS exposure and precede degenerative changes in parenchymal cells. Increases in serum activities of liver-derived enzymes also occur after LPS exposure, suggesting damage to liver parenchyma (102). Among the hepatic functional alterations associated with LPS exposure are decreased hepatic blood flow (137,138,141), cholestasis (23,141), and alterations in hepatic protein synthesis (217). Although the effects of LPS have been well characterized, much remains unknown about the mechanisms of LPS hepatotoxicity. The early accumulation of PMNs in the liver after LPS exposure raises the possibility that these phagocytic cells contribute to LPS-induced liver injury. PMNs are an. important cellular component of the host’s defense system against invading pathogens. However, they have been shown. to :mediate tissue injury ‘under certain conditions. For example, intradermal injection of E. coli LPS produces an acute inflammatory response which is associated with hyperemia, increased vascular permeability, and hemorrhage (41) . PMN infiltration occurs concurrently 69 with the onset of these changes, suggesting that PMNs contribute to their development (40). This is supported by studies which indicate that damage to the skin does not occur in animals depleted of circulating PMNs (160). Evidence from similar studies suggests that PMNs also contribute to lung injury following iv administration of LPS. The initial response of the lungs to LPS is increased pulmonary artery pressure (82). This hypertensive phase is transient in nature and is followed by injury to the vasculature which is characterized by an increase in vascular permeability (218). PMNs have been implicated in the pathogenesis of this second phase of the response, since PMN infiltration is temporally associated with the increased vascular permeability (219) and since injury to the vasculature is prevented by PMN depletion (112). Although evidence suggests that PMNs contribute to the pathogenesis of LPS-induced injury in some tissues, their role in LPS hepatotoxicity in vivo has not been demonstrated. Results from one study showed that isolated hepatocytes were injured by heat-killed Pseudomonas aeruginosa only in the presence of PMNs (161). It was proposed that PMNs may contribute to the liver injury associated with exposure to this gram negative bacterium in vivo. Since many of the biological effects associated with gram negative bacterial infection are thought to be caused by LPS, it seemed possible that these effects in the liver were also mediated by LPS. The purpose of this study was to 70 test the hypothesis that PMNs contribute to the development of liver injury that follows exposure to LPS in vivo. 3.3 Materials and methods 3.3.a Animals Female Sprague-Dawley rats (Charles River, Crl:CD BR (SD) VAF/plus, Portage, MI) weighing 200-250g’ were maintained on a 12 hr light-dark cycle for at least one week prior to use. Food (Wayne Lab-Blox, Allied Mills, Chicago, IL) and water was allowed ad libitum throughout the studies. Female New Zealand white rabbits weighing 2-3 kg were obtained from Bailey’s Rabbitry (Alto, MI) and maintained on high fiber Purina Lab Rabbit Chow (Purina Mills, St. Louis, MO) . 3.3.b Treatment protocols LPS (E. coli 0128:B12, Sigma Chemical Co, St. Louis, MO) was dissolved in sterile saline immediately prior to administration. It was administered in the tail vein as a single bolus injection in a volume of 5 ml/kg. The response to LPS varied considerably over the course of the studies. As a result, a dose of LPS was chosen for each study which produced consistent liver injury in the absence of significant lethality. The dose of LPS are indicated in the figure legends and ranged from 2 mg/kg in the time-course studies to 8 mg/kg in the lymphocyte (LC)-depletion studies. 71 A dose of 5 mg/kg LPS produced reproducible liver injury with minimal lethality over a 6 hr exposure period and was used exclusively in subsequent chapters. It should be noted that manipulations prior to LPS exposure which stress the animals (ie. increasing' body’ temperature) 'may alter ‘the sensitivity of animals to the hepatotoxic and lethal effects of LPS. In an initial study to assess the effect of PMN depletion on LPS hepatotoxicity, rats were divided into two treatment groups. One group of rats received immunoglobulins (Ig) isolated from serum obtained from untreated rabbits (control Ig). The second group received 19 isolated from rabbits immunized ‘with. rat PMNs (anti-PMN Ig). 19’ were administered in the tail vein in two separate injections of 0.4 and 0.3 ml, 18 and 6 hr prior to LPS administration, respectively. Both groups received LPS. Rats were anesthetized with diethyl ether 6 hr after administration of LPS (3 mg/kg), and blood samples anticoagulated with sodium citrate (3.8 % w/v sodium citrate in DDW diluted 1/10 with whole blood in the syringe) were obtained from the inferior vena cava for quantification of liver injury and WBC numbers. Liver injury was quantified by changes in plasma ALT and AST activities and total plasma bilirubin concentration. Liver samples were also obtained for morphologic evaluation of liver injury. A similar experimental design was used to examine the effect of PMN depletion on the development of LPS-induced liver injury 72 over a 24 hr period. In this study, rats were treated with 2 mg/kg LPS to ensure that the animals survived for the duration of the study. In a separate study, PMN infiltration in the liver was assessed 1.5 hr after LPS exposure. Changes in the cellular composition of the NPC fraction of liver digests, which includes PMNs, were determined after administration of 5 mg/kg LPS or saline vehicle to rats pretreated with either control Ig or anti-PMN Ig. Liver digests were prepared, and NPCs were quantified as described in "Methods". 3.3.c Anti-PMN and LC Ig preparation Rat PMNs were elicited from the peritoneum of male, Sprague-Dawley, retired breeder rats using a 1 % solution of glycogen in sterile saline as described previously (220). Rabbits were immunized against rat PMNs by injecting PMNs (1.0 x 106 cells/m1 suspended in complete Freund’s adjuvent) so into the footpads. This was followed by booster injections of PMNs (1.0 x 106 PMNs suspended in incomplete Freund's adjuvent) in the dorsum area of the rump two and four weeks after the initial immunization. Blood (40-50 ml) was obtained from the central ear artery one week after the second PMN booster injection and allowed to clot. Total Igs were precipitated from serum using ammonium sulfate. Anti-rat LC serum raised in rabbits was obtained from Accurate Chemical and Scientific Co (Westbury, NY). Total Ig were isolated from ‘the serum. by ammonium. sulfate 73 precipitation as described above. 19 fractions from both control Ig and anti-LC Ig serum were incubated on ice for 1 hr with rat peritoneal PMNs (3.5 x 106 cells/ml) to remove Ig to PMNs. Following incubation, PMNs were pelleted by centrifugation for 10 min at 1000 x g, and the supernatant was stored at -20°C prior to use. The total Ig fraction of serum from rabbits was isolated by ammonium sulfate precipitation (221). A saturated ammonium sulfate solution was added drop-wise to serum using a separatory funnel until a 2:1 ratio of serum:ammonium sulfate solution was attained. Serum. was mixed thoroughly with a magnetic stir bar during addition of ammonium sulfate. The suspension was stirred for an additional hour at 4°C, spun in a centrifugation at 1000xg for 10 min, and the supernatant was discarded. The pellet was dissolved in sterile saline to the original volume of serum. This solution was precipitated as described above except that a 1:1 ratio of saturated ammonium sulfate solution was used. The precipitate was resuspended in saline to approximately half the orginal volume and dialyzed against saline using Spectrapor 2 dialysis tubing (12-14000 m. w. cutoff, Spectrum Medical Ind., Los Angeles, CA) for 2- 3 days at 4°C to remove contaminating ammonium sulfate. After dialysis, the volume was adjusted to its original volume with saline. Flocculent material was removed by spinning' in. a centrifuge at 1000 x: g for 10 min. The supernatant, which contained lg to rat PMNs (anti-PMN Ig), 74 was stored at -20°C prior to use. Total Ig from untreated rabbits were isolated in a similar manner and used as a control (control Ig). 3.3.d Evaluation of liver injury Liver injury was quantified by changes in plasma ALT and AST activities and by changes in total plasma bilirubin concentration. Plasma ALT and AST activities were measured using the spectrophotometric procedure of Reitman and Frankel (222). Total plasma bilirubin concentration was measured spectrophotometrically using an extinction coefficient of 73000 following conjugation with p- diazobenzenesulfonic acid (223). 3.3.e Histopathologic evaluation Samples of liver were fixed in 10 % buffered formalin and processed for histopathologic examination. Paraffin embedded sections were cut at 6 um and were stained with hematoxylin and eosin (H and E). Slides were randomized, coded and evaluated with light microscopy. A detailed description of the histologic lesions was provided by a pathologist. The severity' of each lesionl was graded as follows: no lesion (0), mild necrosis (1), moderate necrosis (2), marked necrosis with mild hemorrhage (3), severe necrosis with moderate hemorrhage (4), severe necrosis with marked hemorrhage (5). 75 3.3.f Morphometric analysis of hepatic lesions Morphometric assessment of hepatic lesions was performed. on coded liver sections. The area of hepatic tissue examined, the area of individual lesions and the number of segmented PMNs per area of tissue were determined with the aid of a Jandel Video Analysis System (Jandel Scientific, Corte Madera, CA) attached to a Nikon Microphot- FX microscope (Nikon Instrument Group, Oak Park, IL). The area of liver with lesions was calculated and expressed as % liver affected, and the number of lesions per area of liver was also determined. The injury index was defined as the product of the mean lesion area (m2) and the mean lesion severity score (see "Histopathologic evaluation"). The numbers of PMNs per unaffected and affected area of liver were determined by direct, microscopic examination of tissue and were expressed as cells per area of liver. 3.3.g Quantification of total and individual WBC numbers Circulating total WBC numbers (cells/uL blood) were determined using a Coulter Counter Model ZM. Differential counts were performed on blood smears stained with buffered, differential wright-Giemsa stain to obtain the fraction of PMNs and LCs in each blood sample. The total number of WBCs was multiplied by this fraction to obtain numbers of PMNs and LCs (cells/uL blood). 76 3.3.h Isolation of hepatic NPCs Rats were anesthetized with sodium pentobarbital (50 mg/kg, ip) prior to surgery. The abdominal cavity was opened and a loose ligature was placed around the inferior vena cava proximal to the kidneys. It was severed distal to the ligature, and the liver was cleared of blood by infusing 40 ml of Ca2+—free Hank's balanced salt solution (HBSS, pH 7.4) containing 0.5 M HEPES and 0.5 mM EGTA through a cannula (Intramedic PE-160 polyethylene tubing, Clay Adams, Parsippany, NJ) in the hepatic portal vein. The thoracic cavity was opened, and the superior vena cava was ligated. The liver was flushed with 10 ml HBSS containing Pronase E (0.2 % w/v Type XIV protease, Sigma Chemical Co, St Louis, MO) , and the ligature around the inferior vena cava was tightened to isolate the liver from the systemic circulation. An additional 5 ml of HBSS containing Pronase E was infused into the portal vein. The liver was removed and minced with scissors into pieces that could be drawn up into a 5 ml pipette (Pipetman, Rainin Instrument Co, Woburn, MA) without plugging it. Thorough mincing was critical to the complete digestion of the liver. The minced liver was transfered to a 250 ml plastic flask containing 100 ml Ca2+- free HBSS with 0.2 % Pronase E and incubated between 1.0 and 1.5 hr at 37°C under 95 % 02/5 % C02 in a shaking water bath. The solution was triturated twice during the incubation period using a 5 ml pipette to facilitate the disruption of the liver. After the incubation period, the 77 solution was filtered through gauze into 50 ml plastic tubes. The amount of tissue retained by the gauze was routinely small and consisted mostly of connective tissue and portions of the diaphram which were not separated from the liver prior to digestion. The cell suspension was spun in a centrifuge for 20 min at 500 x g, and the pellet was resuspended in 50 ml Ca2+-free HBSS containing 0.5 M HEPES, 0.5 mM EGTA, and 1000 U/L sodium heparin (grade II, Sigma Chemical Co, St Louis, MO). This suspension, which contained hepatic non-parenchymal cells (NPC) and a small number of blood cells, was spun again for 20 min at 500 x g, and the 2+ pellet was resuspended in 25 ml Ca -free HBSS containing heparin and used for cell enumeration. 3.3.i Cytologic examination of NPCs Total nucleated NPCs/liver were quantified using a Unopette (test 5859, Becton-Dickinson, Rutherford, NJ) and a hemacytometer. Samples of the NPC isolates were diluted in Ca2+-free HBSS containing heparin and 1 % bovine serum albumin (Sigma chemical Co, St Louis, MO), concentrated on a microscope slide using a cytocentrifuge (Cytospin 2, Shandon Southern Instruments, Sewickley, PA) and stained 'with. a modified Wright's stain. Stained slides were coded, randomized, and a detailed cytologic analysis, including a 300 unit differential count, was performed. Absolute cell numbers were determined by multiplying the relative fraction 78 of each cell type by the total number of nucleated cells/liver. 3.3.j Data analysis Results are expressed. as means i SEM. Comparisons between control 19 and anti-PMN Ig treatment groups were made using the Wilcoxon rank-sums test. In the PMN depletion time-course study, multiple comparisons within groups were made with the Wilcoxon rank-sums test utilizing Bonferroni’s correction factor. Data from cytologic evaluations of NPCs isolated from liver digests were analyzed following transformation using a completely random 2 x 2 factorial analysis of variance. Between group comparisons were performed using Tukey’s omega test. The rank sums test was used when variances were not homogeneous (eg, eosinophils). For histologic and morphometric analyses, data were analyzed by Student's t-test after appropriate transformation to render variances homogeneous or by the rank sums test in cases where variances were not homogenous after transformation. The criterion for significance was p < 0.05. 3.4 Results 3.4.a Characterization of control and anti-PMN Ig The effects of Ig administration on circulating numbers of WBCs, platelets, and red blood cells (hematocrit) in otherwise untreated rats are shown in Table 3.1. In blood 79 Table 3.1 Effect of control Ig and anti-PMN 19 on circulating blood cell numbers and hematocrit in the absence of LPS Treatment Control Ig Anti-PMN Ig Total WBC 12800 i 1310 7380 i 1050* (cells/ul) PMNs 2320 i 306 92 i 28* (cells/ul) Lymphocytes 10500 i 1230 7140 i 1070 (cells/ul) Monocytes 46 i 23 95 i 22 (cells/ul) Eosinophils 36 i 16 36 i 19 (cells/ul) Platelets 790000 i 38300 681000 1 15400 (cells/ul) Hematocrit 40 i 0.8 42 i 0.7 (’3) Rats were pretreated with either control or anti-PMN Ig as described in Section 3.3.b. Changes in circulating blood cell numbers and hematocrit were assessed 6 hr after administration of the second dose of Ig. *, significantly different from respective control Ig treatment value. 80 from control Ig-pretreated rats, neither numbers of total WBCs nor numbers of individual WBCs were different from values reported for normal, 8 week old, female, Sprague- Dawley rats (224). In contrast, rats pretreated with anti- PMN Ig exhibited a marked decrease in total WBCs. This appeared to be due to a decrease in both circulating PMNs and LCs, although the decrease in LCs was not statistically significant in this experiment. The reduction in PMNs was pronounced. PMN numbers amounted to <5 % of those from rats pretreated with control Ig. Red blood cells (hematocrit) , other WBCs, and platelets were not affected by the anti-PMN Ig. These results are consistent with results obtained by others which showed that antiserum to rat PMNs raised in rabbits using a similar procedure exhibited high antibody titer to PMNs and low antibody titer to LCs, platelets and red blood cells (225,226). 3.4.b Effects of LPS on circulating WBCs Changes in circulating total and individual WBC numbers in control Ig- and anti-PMN Ig-pretreated rats at various times after exposure to LPS are shown in Figure 3.1. LPS administration to control Ig-pretreated rats caused leukopenia within 3 hr (Figure 3.1A). The LPS-induced decrease in total WBCs was more modest in rats pretreated with anti-PMN Ig and was due primarily to a decrease in circulating LCs (Figure 3.1C). After 3 hr, total WBCs returned to pre-LPS values in both groups. Circulating PMNs 81 Figure 3.1 Changes in circulating total white blood cells (WBCs), neutrophils (PMNs) and lymphocytes (LCs) after LPS administration to either control Ig- or anti-PMN Ig- pretreated rats. Ig was administered in the tail vein of rats as described in Section 3.3.b. Changes in total WBCs (A), PMN (B), and LCs (C) were measured as described in Section 3.3.g at various times after exposure to LPS (2 mg/kg, iv). Results are expressed as means i SEM; N=6. a, significantly different from respective value at 0 hr. b, significantly different from respective control Ig pretreatment group. i LR. Tlfl 1* a.” 1 bag U m g. h. 01H mum War b \\\\\\\\_. mm L as ..a w h. o 1 1 Ana. x 1}]; .8; .33 838.8 A D Control I -entl-Pflfl lg Pretreatment O 1 5 “no. a 1318 .22.. 83:26 0 C Prctreatment i _ \\\\\\\\\\x 2 a b 3 a - m mm b .\\\\\\\\\\\\ a mg l a m 107 5 O Pb. x 1).-8 8.. 8.8.326 Tlme (hr) after LPS 83 gradually increased to more than 2-fold the pretreatment value in control Ig-pretreated rats 24 hr after exposure to LPS (Figure 3.1B). In contrast, blood PMN numbers in rats pretreated with anti-PMN Ig remained reduced for 24 hr after LPS exposure. Blood monocyte and eosinophil numbers did not appear to be altered by LPS exposure in these studies (data not shown). However, since circulating' numbers of these cells are normally <1 % of the total circulating WBCs number in female, Sprague-Dawley rats (224), alterations in their numbers in the blood were difficult to assess precisely. 3.4.c Effect of PMN depletion on liver PMN number Table 3.2 summarizes the changes in cellular composition of the NPC fraction of liver digests from control Ig- or anti-PMN Ig-pretreated rats 1.5 hr after administration of either LPS or saline vehicle. The cellular composition of liver digests was similar in saline-treated rats, irrespective of Ig pretreatment (Table 3.2 and Figure 3.2). The primary cell type ‘was a round to polygonal, mononuclear cell with eosinophilic, slightly granular cytoplasm. These cells, which are designated as "other cells", were variable in size and had large, round or oval, basophilic nuclei with homogeneous chromatin. The nucleus was often eccentrically located and had indistinct nucleoli. These morphologic characteristics are consistent with those of endothelial cells (see Section 3.5). The preparations 84 Table 3.2 Cytologic evaluation of hepatic NPC fraction from liver digests after LPS exposure Cells/liver Control Ig anti-PMN Ig (x 10°) Saline LPS Saline LPS Total NPCS 170 262 150 135 i 10 i 21‘ i 17 at 8-0° Neutrophils 10.5 192 11.1 58.4 i 3.0 i 21ll i 7.0 + 11”": Lymphocytes 11.4 11.6 11.5 8.1 i 3.3 i 1.8 i 1.5 i 1.2 Macrophages 19.6 16.6 23.6 27.2 i 2.7 i 4.0 .t 1.1 i 4.0 Eosinophils 0.60 0.57 0.00 0.45 i 0.60 i 0.57 i 0.00 i 0.29 Plasma cells 0.60 1.30 0.40 0.63 i 0.60 i 0.59 i 0.40 i 0.42 Other cells 127 40.2 103 40.3 b i 11 : 11.2II i 10 i 9.3 Rats were pretreated with either control 19 or anti-PMN Ig as described in Section 3.3.b. Cells were quantified in the hepatic NPC fraction 1.5 hr after LPS (5 mg/kg, iv) as decribed in Section 3 . 3 . i . SEM. Data are expressed as means 1; ° Significantly different from respective control Ig/saline group. b Significantly different from respective anti-PMN Ig/saline group. c group. Significantly different from respective control Ig/LPS 85 Figure 3.2 Photomicrographs of NPC fractions from rat liver digests after exposure to LPS vehicle. A-B, NPCs from rats pretreated with control Ig; C-D, NPCs from rats pretreated with anti-PMN Ig. Rats were pretreated with Ig as described in Section 3.3.b. Hepatic NPC fractions were isolated as described in Section 3.3.h 1.5 hr after treatment with saline. Magnifications: A and C = 216x; B and D = 540x. Arrows indicate PMNs. 863 Figure 3.2 O O (7" 006” A °° 2 0' . o _',' ‘3 a: c ,3? . ”a" O 6 8% 5‘ f. C ’0 :35-90 ”we; “ ~e - '0. o ‘. w?" o 4. 9 ‘3 ‘ ¢ "1 87 contained lesser numbers of macrophages (ie., Kupffer cells), PMNs and small LCs. Plasma cells and eosinophils were seen occasionally and averaged 5 1 % of the cell population. Occasional mast cells and hepatic parenchymal cells were observed in some digests (< 1 %). One and a half hr after exposure to LPS, the liver digests from rats pretreated with control Ig had pronounced alterations in cell quantity and cytomorphologic characteristics (Table 3.2, Figure 3.3). These preparations were more cellular than those from rats that received either control Ig and saline or anti-PMN Ig and LPS. This increased cellularity was due to an increase in PMNs (Table 3.2), many of which had degenerate cytomorphologic characteristics including pyknosis, karyolysis and karyorrhexis. The liver digests from ‘these. animals also Ihad significantly fewer "other cells" than those from saline controls (Table 3.2). In livers from LPS-treated rats, the absolute number of PMNs was markedly reduced by pretreatment with anti-PMN Ig (Table 3.2). The Idegenerate cytomorphology that characterized PMNs from liver digests of rats treated with control Ig and LPS was less common in digests from animals treated with anti-PMN Ig and LPS (Figure 3.3). Although the number and morphology of the macrophages, LCs, eosinophils and plasma cells were unaffected by Ig or LPS treatment (Table 3.2), the number of "other cells" was decreased to a degree similar to that observed in liver digests from rats treated with control Ig and LPS. 88 Figure 3.3 Photomicrographs of NPC fractions from rat liver digests after exposure to LPS. A—B, NPCs from rats pretreated with control Ig; C-D, NPCs from rats pretreated with anti-PMN Ig. Igs were administered as described in Section 3.3.b. Hepatic NPC fractions were isolated as described in Section 3.3.h 1.5 hr after exposure to LPS (5mg/kg, iv). Magnifications: A and C = 216x; B and D = 540x. Arrows indicate PMNs. 89a Figure 3.3 ' a.-. 21% e ‘hu;_ 'i'éc . '6‘fit e. _, 3 g: ,. *3 a. - 00:. .. 89b Figure 3.3 9O 3.4.d Effect of PMN depletion on LPS-induced liver injury To assess the contribution of PMNs to LPS hepatotoxicity, LPS was administered to rats depleted of blood PMNs. In this study, the number of circulating PMNs 6 hr after LPS administration to rats pretreated with control Ig was 2792 j; 576 cells/uL blood compared to 168 i 50 cells/uL blood in anti-PMN Ig-pretreated rats. Figure 3.4 shows the plasma AST and ALT activities and total plasma biliubin concentration 6 hr after intravenous administration of LPS. In rats pretreated with anti-PMN Ig, LPS exposure resulted in plasma AST and ALT activities which were markedly less than those exhibited by rats treated with control Ig and LPS. Plasma bilirubin concentration in the anti-PMN Ig pretreatment group tended also to be less than in the control Ig pretreatment group after LPS exposure; however, this trend was not statistically significant. Liver samples obtained from animals in this study were assessed for histopathologic changes. Liver sections from rats pretreated with control Ig had multifocal to coalescing, moderate, acute, neutrophilic hepatitis 6 hr after LPS exposure (Figure 3.5). The sinusoids contained many PMNs, few plump Kupffer cells and small amounts of an eosinophilic, proteinaceous precipitate. Multifocal, coalescing, irregularly shaped areas of hepatocellular degeneration were observed. These lesions were primarily midzonal and were characterized by slightly swollen, weakly 91 .msonm usesumenuenc 6H Honucoo e>nuoemmen Eonu uceneuuuc >Huce0uuucmnm .e .oauz “cam H mcmea me cemmenmxe ene muadmem .A>n .mchE 3 mad ou masono uceaueenuenm 0H cuoc no enamomxe neuue nc e c.n.m coauoem cw cecwnomec we censmmes ene3 conumnuceocoo uncannauc mameum Heuou can meuufi>wuoe Baa use emd mammHm .c.m.m c0nuoem an cecunomec me on ZSmInuce no Honucoo necune cun3 ceumenuenm ene3 muem .ansmcn ne>na ceosUCulmmA co 0H ZZquuce no 0H Honucoo necuue cunz uceaueenuenm no uoeuum v.m ensmum 92 Total plosmo bilirubin (mg/dl) r2.0 1 Figure 3.4 Pretreatment [:| Control lg n anti—PMN lg Bilirubin I I I I r I I I O. 0. (\I 1— (90L x IUJ/n as) KIIAIIOD l'IV/lSV Dwsold 0 ALT AST 93 Figure 3.5 Photomicrographs of liver after administration of LPS to rats pretreated with control Ig. Liver samples from rats in Figure 3.4 were prepared for light microscopy as described in Section 3.3.e. Magnifications: A = 54x; B = 216x. 94 Figure 3.5 I'" .u' ‘ "‘\ V » £ a: 13.“: ~t with» *h ".“ .a =:~,_ :. _- ‘ ‘ . '1' ':-...-a . ' r g ‘ I. , I . :3' ‘ ‘ " ' , g . e ,V n" ' q ,. r r in! V) . M’ . I?“ -94 ' \¥.{ "“f‘vbir 33359? 1 __.z V3.1.» o . ‘ a? .37.. . A” \ 95 basophilic hepatocytes with swollen, round or oval nuclei. In many areas, the lesions progressed to necrosis. Hepatocytes in necrotic areas were hypereosinophilic with small, pyknotic nuclei or were large and pale with faint nuclei and indistinct cytoplasmic borders. In several affected areas, normal architecture was completely lost. Necrotic foci contained. a :moderate number of degenerate PMNs. The number of PMNs in these affected areas was greater than that in unaffected areas (Table 3.3, Control Ig/LPS group: PMNs/affected and PMNs/unaffected areas). Hemorrhage was associated with foci in some samples. Few small foci of single cell necrosis were also observed. Livers from LPS-exposed animals that were pretreated with anti-PMN Ig had similar but significantly fewer lesions per area of liver compared to livers from rats pretreated with control Ig and exposed to LPS (Figure 3.6, Table 3.3). The ‘total area. of liver' affected. with lesions ‘was also significantly less in animals pretreated with anti-PMN Ig. There was a trend for the lesions in rats pretreated with anti-PMN 19 to be smaller and less severe as indicated by a reduction in mean lesion area and individual lesion severity. Rats pretreated with anti-PMN Ig also had a significant reduction in PMNs within both affected and unaffected areas of the liver as well as fewer PMNs per lesion compared to control Ig-pretreated rats. The development of LPS-induced liver injury is shown in Figure 3.7. Plasma AST activity was less than 50 SF units/ml 96 Table 3.31. Morphometric analysis of livers from control Ig- or anti-PMN Ig- pretreated rats after exposure to LPS Treatment Control Ig/LPS Anti-PMN Ig/LPS % liver affectedb 6.1 i 2.6 1.0 i 0.3* Lesions/area liver 51.6 i 14.2 14.5 i 3.4* (f/cmz) Mean lesion area 0.094 i 0.015 0.058 i 0.012 (mmz) Lesion severity 2.37 i 0.30 1.89 i 0.24 Injury indexc 0.213 1 0.043 0.110 1 0.025* PMNs/affected area 289 i 31 94 i 11* (cells/mmz) PMNs/unaffected area 127 i 14 63 i 11* (cells/mmz) PMNs/lesion 27 i 4 7 i 1* Rats were pretreated with either control Ig or anti-PMN Ig as described in "Materials and Methods". Morphometric analysis was performed as described in Section 3.3.f on liver sections from studies shown in figures 3.4-3.6, 6 hr after LPS exposure (3 mg/kg, iv). Data are expressed as means i SEM. ', significantly different from control Ig/LPS treatment group. , data obtained by Dr. Eric Schultze b, determined as lesion area/total area examined for each rat multiplied by 100. , determined as the product of severity score (0-5) and mean lesion area for each rat. C 97 Figure 3.6 Photomicrographs of liver after administration of LPS to rats pretreated with anti-PMN Ig. Liver samples from rats in Figure 3.4 were prepared from light microscopy 54X; B = as described in Section 3.3.e. Magnifications: A 216x. 98 Figure 3.6 - . I l. .c In) 61!} . . 1.! -.».O. '4 ~JKL I wgwmvmmfie . .... ...??? +3 1. 1f . . . . .4: I. . I . .1 I .I v A .. .. . - sewage...» _ gee»... .. .. ewe . , ....M .. weave sheet“. .1. 5.3.0.1.. 36...».1-.. .5 t o. . ,. ensue. e... B O ... , c 3‘4 ”or/....I . . . . .v . satin .v . .7 v... n . _ 1) 5m 1W... .0 . «1" en. Pun». .. .0 Eyeing ... ifavoa .....x o e. 2 9 ..2 r... .. ..Q..«. . .b M... .. “km w ..79HVWOQWMWWW13 . . . snafu , . . ... .... 3-..... .....- a 4 .... ... as»? wepewwefimwbeenmmw 99 Figure 3.7 Time-course of liver injury after administration of LPS to rats pretreated with either control Ig or anti—PMN Ig. Ig pretreatment was as described in Section 3.3.b. Plasma aspartate aminotransferase (AST) activity (A) and total plasma bilirubin concentration (B) were measured as described in Section 3.3.d at various times after exposure of both 19 pretreatment groups to LPS (2 mg/kg, iv). Results are expressed as means i SEM; N=6. a, significantly different from respective value at 0 hr. > d a J an. 9 Plasma AST activity (SF U/ml x 103) .° 8" O 100 Figure 3.7 Pretreatment :lControl lg f? -anti—PMN lg ° J a a n... ma 8 a s N o J ..5 0| 1 ..e O l .0 a: L Total plasma bilirubin (mg/dl) Pretreatment 0 El Control lg anti—PMN lg ...... Fl§ e m 'l‘Ime (hr) after LPS 101 in rats treated with either control or anti-PMN Ig prior to the administration. LPS (Figure 3.7A, 0 hr). In animals pretreated with control Ig, plasma AST activity increased between 3 and 6 hr after administration of LPS and remained elevated thereafter. In rats pretreated with anti—PMN Ig, a much more modest response occurred in the same time frame. Prior to LPS administration, total plasma bilirubin concentration was less than 0.1 mg/dL in both 19 pretreatment groups (Figure 3.7B, 0 hr). It became maximally elevated 6 hr after LPS administration and returned to preLPS-exposure values by 24 hrs in both pretreatment groups. No lethality occurred at any time in this study. 3.4.e Ineffectiveness of LC depletion on LPS hepatotoxicity The decrease in circulating lymphocytes prior to LPS exposure in rats preteated with anti-PMN Ig (Table 3.1 and Figure 3.1C, 0 hr) indicated that the anti-PMN Ig preparation was highly selective but not completely specific for PMNs. Since it was possible that reduction of lymphocyte numbers contributed to the protection by the anti-PMN Ig preparation, a study was undertaken to test the effect of an anti-lymphocyte Ig' preparation (anti-LC Ig) on LPS hepatotoxicity. The results from this study are shown in Figure 3.8 along with the results from a parallel study using anti-PMN Ig. Consistent with earlier findings (above), pretreatment with the anti-PMN Ig preparation largely 102 .maonm uceEuaenuenQ 0H Honucoo e>nuoemmen Eonu uceneuuflc aauceouMflcmnm .e .maloauz “2mm H mceea me cemmenmxe ene muaamem .A>w .ax\va we was ce>ueoen masonm ads .c.n.m coauoem nu cecunomec me enamomxe mag neune nc w censeaee mes wun>nuoa Bmd eEmeHm .couuenumwcHEce mad ou nannm nc n canuoemcfl as a .eaach a me cwe> Hueu ecu cu ceneumncnece ma3 mH Onluucw .c.m.m cofluoem Cw cecanomec we me3 0H zzmInuce cuu3 ucesueenuenm .ensmomxe mmq neuue wuu>uuoe Bm¢ efimeam co OH Ogluucm ace cH z=mIfluce no uoeuue ecu no comnnemfiou w.n enemam 103 .532an 360:an cozeaec 2.21 % ... wan 939.1 a. wmsueca a _ 35.30 D uceEuaebeta (90L x Iw/n as) KIIAIIGD 15v owsola 104 prevented the increase in plasma AST activity 6 hr after LPS administration. In contrast, although pretreatment with anti-LC Ig resulted in >90 % depletion of circulating lymphocytes at the time of LPS administration (7204 i 738 cells/uL and 561 i 112 cells/uL in control Ig- and anti-LC Ig-treated rats, respectively; N=4), no protection was observed. The anti-LC Ig did not affect circulating PMN numbers (data not shown). 3.5 Discussion Evidence from several studies indicates that PMN depletion affords protection against LPS-induced injury to the skin (160) and lungs (112), suggesting that these phagocytic cells contribute to the pathogenesis of injury from LPS in certain tissues. Results from the present study indicate that PMN depletion attenuates liver injury associated with LPS exposure. This suggests that PMNs contribute to the pathogenesis of LPS hepatotoxicity. PMN depletion markedly attenuated LPS hepatotoxicity. However, it did not afford complete protection. This is indicated by the significant elevation in plasma AST activity and total bilirubin concentration 6 hr following LPS administration in PMN-depleted rats (Figure 3.7). Also, foci of parenchymal cells exhibiting mild degenerative changes were observed in livers from PMN-depleted rats after LPS administration (Figure 3.6). Although pretreatment with 105 the anti-PMN Ig caused a marked reduction in circulating PMNs, it is possible that enough PMNs remained in the liver after PMN depletion to cause some injury. Alternatively, PMN-independent mechanisms may contribute to LPS hepatotoxicity. Cholestasis and hyperbilirubinemia are often associated with LPS exposure. LPS has been shown to decrease bile flow from the isolated, buffer-perfused liver (141,142) and evidence suggests that LPS may directly affect bile secretion by inhibiting Na+,K+-ATPase on the bile canalicular plasma membrane of hepatocytes (144). Thus, hyperbilirubinemia following LPS exposure may be due to a PMN-independent alteration in bile secretion. This may explain the ability of anti-PMN Ig to attenuate the rise in plasma AST and ALT activities without causing a significant reduction in total plasma bilirubin concentration. Cytocentrifuge preparations of liver digests contained a large number of cells which were described as "other cells". Bautista et al. reported values for hepatic endothelial cell numbers which closely approximate the number of "other cells" observed in the present study, suggesting that these cells are hepatic endothelial cells (227). Additionally, these cells had morphologic characteristics that are similar to cytocentrifuge preparations of cultured rat and bovine vascular endothelial cells (not shown). Exposure to LPS caused a significant decrease in "other cells" which was not prevented by PMN 106 depletion. Thus, the decrease in "other cells" does not appear to be mediated by PMNs. This finding is consistent with. the observation. that LPS is directly cytotoxic to endothelial cells grown in culture (45,46). Further studies are necessary to elucidate the mechanism of this decrease in "other cells" following LPS exposure. LPS is cleared from the circulation primarily by Kupffer cells in the liver (18,20). Because these hepatic macrophages exhibit enhanced activity following LPS exposure in vivo (228) and because macrophages exposed to LPS in vitro release cytotoxic substances, including reactive oxygen metabolites (228,229,230), it has been proposed that activated macrophages may contribute to the development of LPS-induced liver injury (231). This is supported by evidence indicating that treatments which alter Kupffer cell function alter the response of the liver to LPS (176). Also, studies of cocultures of Kupffer cells and hepatocytes indicate that LPS is cytotoxic to hepatocytes in a Kupffer cell-dependent manner (177). Thus, Kupffer cells might contribute to LPS hepatotoxicity through the extracellular release of cytotoxic mediators. LPS-activated macrophages release chemotactic factors for PMNs in vitro (149), and macrophage-derived chemotactic factors have been implicated in the accumulation of PMNs in the pleural cavity (149) and peritoneum (232) after LPS exposure in vivo. Thus, it is possible that Kupffer cells contribute to LPS hepatotoxicity by mediating the recruitment of PMNs. 107 The anti-PMN Ig caused a significant reduction in circulating‘ LCs. Since LCs have been implicated in the pathogenesis of liver injury under certain circumstances (233,234), it could be argued that the protective effect of this Ig preparation was due at least in part to LC depletion. However, administration of an anti-LC Ig preparation resulted in depletion of circulating LC but did not afford protection against LPS hepatotoxicity. Therefore, it is unlikely that the decrease in circulating LCs contributed to the protective effect of the anti-PMN Ig preparation. This result also indicates that the ability of the anti-PMN Ig preparation to protect against LPS hepatotoxicity ‘was not due to the method used for PMN depletion (ie, generation of immune complexes), since depletion of a different blood cell by a similar method did not afford protection. This strengthens the conclusion that the anti-PMN’ Ig jprotected against. LPS hepatotoxicity' by depletion of PMNs and supports the hypothesis that PMNs contribute to the pathogenesis of LPS hepatotoxicity. Liver injury following LPS exposure is associated with an acute inflammatory response, a prominant feature of which is an early accumulation of PMNs at the site of inflammation. PMN infiltration is usually followed by the appearance of macrophages that are thought to be derived from circulating blood monocytes (235,236). Although macrophages can mediate tissue injury under certain circumstances” it. seems ‘unlikely' that. these infiltrating 108 macrophages contribute to the early development of liver injury that follows LPS exposure since they usually appear later in the acute inflammatory response (235). This is supported by our results (Table 1.1) and results from a recent study which indicated that the total number of mononuclear phagocytes in the liver was not changed prior to the onset of LPS-induced liver injury (227). By contrast, large numbers of PMNs were observed in the liver as early as 45 min after LPS (data not shown), well before the onset of severe liver injury. Accordingly, the time course of hepatic PMN infiltration after LPS exposure is consistent with their role in the pathogenesis. Recently, PMNs have been implicated in the pathogenesis of liver injury in other models. For example, depletion of PMNs afforded protection against the hepatotoxicity caused by alpha-naphthylisothiocyanate (158), and liver injury following resuscitation from hypovolemic shock was attenuated by monoclonal antibodies to PMN adhesion molecules (159). The mechanisms by which PMNs mediate injury to the liver in these instances are not known. PMNs might cause tissue injury by several mechanisms. For example, activation. of PMNs is associated. with the extracellular release of reactive oxygen metabolites, including 02' and H202 (153). These cytotoxic substances can initiate peroxidation of lipids and oxidation of protein thiols which might contribute to tissue injury. Products of lipid peroxidation have been detected in the liver following 109 LPS exposure (237,238), raising the possibility that oxygen radicals released by activated PMNs may contribute to LPS- induced liver injury. This is supported by the observation that 02" production in the liver was increased after LPS exposure (239). Activation of PMNs may also result in the extracellular release of lysosomal enzymes (240,241,242). These enzymes have been implicated in certain instances of PMN-induced tissue injury (243,244,245). They can also injure isolated hepatocytes under certain circumstances (246). Thus, it is possible that LPS-induced liver injury is mediated by lysosomal enzymes, reactive oxygen metabolites or other agents released from activated PMNs. Further studies are necessary to determine which PMN-derived cytotoxic agents are involved in LPS-induced liver injury. In summary, the PMN accumulation in liver that follows administration of an hepatotoxic dose of LPS occurs before the onset of liver injury, and prior depletion of circulating PMNs attenuates the accumulation of PMNs in the liver and protects against the hepatotoxic effects of LPS. Although additional mechanisms may be involved in LPS hepatotoxicity, the substantial protective effect of PMN depletion indicates that PMNs play an important role in the pathogenesis of LPS hepatotoxicity in the rat. CHAPTER 4 ROLE OF TUMOR NECROSIS FACTOR-ALPHA IN LPS-INDUCED LIVER INJURY 110 111 4.1 Abstract TNF-alpha and blood PMNs have been implicated in the pathogenesis of LPS hepatotoxicity. However, the mechanism by which these endogenous factors mediate liver injury during LPS exposure is uncertain. The objective of this study was to test the hypothesis that TNF-alpha contributes to LPS hepatotoxicity indirectly by a PMN-dependent mechanism. Pretreatment with pentoxifylline (100 mg/kg, iv) attenuated circulating TNF-alpha concentration 1.5 hr after administration of 5 mg/kg LPS and afforded protection against liver injury. Pretreatment of rats with an antiserum to TNF-alpha also afforded protection against LPS hepatotoxicity. Hepatic PMN accumulation 1.5 hr after exposure to LPS ‘was not significantly reduced by pretreatment with either pentoxifylline or TNF-alpha antiserum. Depletion of circulating PMNs, which protects against LPS hepatotoxicity, enhanced circulating TNF-alpha concentration more than 3-fold compared to control rats 1.5 hr after LPS exposure. These results support the hypothesis that TNF-alpha mediates LPS-induced liver injury in the rat by a PMN-dependent mechanism. 112 4.2 Introduction TNF-alpha is one of several cytokines which are thought to play a role in the acute inflammatory response to tissue injury or infection. Among its proinflammatory activities are induction of 'procoagulant activity in 'vascular endothelial cells (189) and the stimulation of PMN adherence to surfaces in vitro, including endothelial cell monolayers (247,248,249). This action may be important in the accumulation of these phagocytic cells at sites of inflammation in vivo. TNF-alpha also induces the synthesis of acute phase proteins by the liver (182,250), suggesting that. TNF-alpha may' contribute to ‘the regulation. of the inflammatory response under certain conditions. Although TNF-alpha mediates certain important beneficial functions during inflammation, results from numerous studies suggests that this cytokine contributes to tissue injury associated with certain pathophysiologic conditions, including cerebral malaria (196) and ischemia/reperfusion in the liver (197) and in the gastrointestinal tract (17). Exposure to LPS is accompanied by a marked rise in circulating concentration of TNF-alpha (251), and evidence from numerous studies suggests that many of the pathophysiologic effects associated with exposure to LPS may be mediated by TNF-alpha, including LPS-induced liver injury. The notion that TNF-alpha mediates the liver injury that occurs after exposure of animals to LPS is supported by 113 studies showing that neutralization of circulating TNF-alpha with a specific antiserum attenuated liver injury in galactosamine-sensitized mice given LPS (201). Futhermore, administration of recombinant murine TNF-alpha to galactosamine-sensitized mice in the absence of LPS produced alterations in the liver which resembled LPS-induced liver injury (252). Although TNF-alpha appears to play a role in the pathogenesis of LPS hepatotoxicity, the mechanisms by which it mediates tissue injury remains unclear. Depletion of PMNs afforded protection against lung injury induced. by recombinant human TNF-alpha suggesting that TNF-alpha mediates tissue injury in vivo by a PMN-dependent mechanism (202). Because TNF-alpha increases the adherence of PMNs to vascular endothelial cells, it has been proposed that TNF- alpha may mediate tissue injury by promoting accumulation of PMNs in tissues (192). Consistent with this is evidence which indicates that pretreatment with antiserum to TNF- alpha or with pentoxifylline, which inhibits the LPS-induced increase in circulating TNF-alpha (251), attenuated the accumulation of PMNs in lungs as well as the pulmonary injury' after exposure: to LPS (200,253). Recent evidence indicates that PMNs accumulate in livers of LPS-exposed rats and that PMN depletion attenuates hepatic injury from LPS, indicating that these cells contribute to liver injury induced by LPS (254 and Chapter 3). This raises the 114 possibility that TNF-alpha may contribute to LPS-induced liver injury by a PMN-dependent mechanism. In contrast, results from other studies suggest that TNF-alpha may mediate liver injury by mechanisms independent of blood PMNs. For example, LPS is cytotoxic to cocultures of Kupffer cells and hepatic parenchymal cells (255). This cytotoxicity was associated with a rise in TNF-alpha in the culture supernatant and could be attenuated by antiserum to TNF-alpha. Since PMNs were not included in this culture system, these results indicate that TNF-alpha can contributed to cytotoxicity to liver cells in vitro by a PMN-independent mechanism. Whether TNF-alpha contributes to the pathogensis of LPS-induced liver injury by a similar mechanism in vivo remains unknown. The purpose of this study was to gain insight into the mechanism by which TNF-alpha contributes to liver injury after LPS exposure in vivo. Specifically, the hypothesis that TNF-alpha mediates liver injury after LPS exposure by a PMN-dependent mechanism was tested. 4.3 Materials and methods 4.3.a Animals Female, Sprague-Dawley rats (Charles River, Crl:CD BR (SD) VAF/plus, Portage, MI) weighing 200-250 kg were used in all studies. Animals were maintained on a 12 hr light-dark cycle for at least one week prior to use. Food (Wayne Lab- 115 Blox, Allied Mills, Chicago, IL) and water were allowed ad libitum. Female New Zealand White rabbits (Bailey's Rabbitry, Alto, MI) were maintained on high fiber Purina Lab Rabbit Chow (Purina Mills, St. Louis, MO). 4.3.b Treatment protocols LPS from E. coli, 0128:B12 (Sigma Chemical Co, St Louis, MO) was administered in the tail vein. To test the effect of antiserum to TNF-alpha on liver injury and hepatic PMN accumulation after LPS administration, rats were divided into two treatment groups. One group was pretreated 1 hr prior to LPS exposure with serum from rabbits immunized against recombinant murine TNF-alpha (TNF-alpha antiserum). The second group was pretreated with serum from untreated rabbits (control serum). Rabbit serum was diluted 1:1 with sterile saline and administered in the tail vein in a volume of 2 ml. Both treatment groups received 5 mg/kg LPS in the tail vein. TNF-alpha antiserum was a gift from Dr. Steven L. Kunkel. A 2 x 2 factorial design was used to test the effect of pentoxifylline (Sigma Chemical Co, St Louis, M0) on LPS hepatotoxicity. Pentoxifylline (100mg/kg) or saline vehicle (0.5 ml/0.1 kg body weight) was administered as a bolus in the tail vein 1 hr prior to administration of either 5 mg/kg LPS or saline vehicle. A similar design was used to examine the effect of pentoxifylline on circulating TNF-alpha 116 concentration and hepatic PMN infiltration after LPS administration. To determine the relationship between circulating PMNs and TNF-alpha in the pathogenesis of LPS hepatotoxicity, the effect of PMN depletion on circulating TNF-alpha concentration after LPS exposure was determined. In this study, rats were pretreated 18 and 6 hr prior to LPS (5 mg/kg, iv) exposure with total Ig fractions isolated from rabbit serum. One group of rats received Igs from rabbits immunized with rat PMNs (anti-PMN Ig). A second group of rats received Igs from untreated rabbits (control Ig). This treatment protocol with anti-PMN Ig markedly reduces circulating PMN numbers prior to LPS exposure and affords protection against LPS hepatotoxicity (254 and Chapter 3). 4.3.c Quantification of plasma TNF-alpha concentration The concentration of TNF-alpha in plasma samples obtained from the inferior vena cava of anesthetized rats (sodium pentobarbital, 50 mg/kg, ip) was determined by the lysis of 'the fibrosarcoma. cell line, ‘WEHI 164 clone 13 (256). Plasma from blood anticoagulated with sodium citrate was used in these studies since heparin can interfere with the assay (257). 4.3.d Enumeration of hepatic PMN numbers PMN numbers in liver tissue were estimated in the NPC fraction of the liver as described in 3.3.h (254). 117 4.3.e Evaluation of liver injury Liver injury was quantified 6 hr after LPS administration except in the TNF-alpha antiserum study in which liver injury was also quantified 1.5 hr after exposure to LPS. Rats were anesthetized with sodium pentobarbital (50 mg/kg, ip), and anticoagulated blood samples (1/10 dilution in whole blood of 3.8 % w/v sodium citrate in DDW) for quantification of liver injury were obtained from the inferior vena cava. Liver injury was assessed by changes in plasma AST activity and by changes in total plasma bilirubin concentration as described in section 3.3.d. 4.3.f Quantification of circulating WBCs Total circulating WBC numbers were quantified in blood anticoagulated with sodium citrate as described in 3.3.9. Circulating numbers of PMNs and LCs were obtained by multiplying the total WBC number by the fraction of each cell type, which was determined by differential counts made on blood smears stained with buffered, differential Wright- Giemsa stain. 4.3.g Anti-PMN Ig preparation Igs to rat PMNs ‘were raised in female rabbits as described in section 3.3.c (254). This polyclonal Ig preparation selectively depleted circulating PMNs and afforded protection against LPS hepatotoxicity (254 and Chapter 3). 118 4.3.h Data analysis Results are expressed as means : SEM. In most studies, comparisons were made using the t-test (p<0.05). In studies in which variances were non-homogeneous, analyses were performed on log transformed data. 4.4 Results 4.4.a Effect of antiserum to TNF-alpha on LPS hepatotoxicity Antiserum to TNF-alpha affords protection against LPS hepatotoxicity in mice sensitized with galactosamine (201). Initial studies were performed to determine if antiserum to this cytokine also affords protection against LPS-induced liver injury in the rat without galactosamine sensitization. The results of these studies are shown in Figure 4.1. Liver injury, as indicated by elevations in plasma AST activity (Figure 4.1A) and total plasma bilirubin concentration (Figure 4.1B), occured by 6 hr after LPS administration in .rats pretreated with control serum. This LPS-induced liver injury was markedly attenuated by pretreatment with antiserum to TNF-alpha. Regardless of pretreatment, plasma AST activity and total plasma bilirubin concentration were unchanged 1.5 hr after administration of LPS, indicating that the onset of liver injury occured between 1.5 and 6 hr after LPS administration. 119 Figure 4.1 Effect of TNF antiserum on LPS hepatotoxicity in rats. Rats were pretreated with either TNF antiserum or control serum in the tail vein 1 hr prior to LPS exposure. LPS (5mg/kg, iv) was administered to all animals. Liver injury was assessed 1.5 and 6 hr later by changes in plasma aspartate aminotransferase (AST) activity (A) and by changes in total plasma bilirubin. concentration (B) as described in Section 4.3.e. Results are expressed as means 1 SEM; N=3-5. a, significantly different from respective saline treatment group at 1.5 hr. b, significantly different from respective control serum pretreatment groups. Plasma AST activity (SF U/ml x 103) Total plasma bilirubin (mg/dl) 120 Figure 4.1 2'5" Pretreatment . CJControl serum 0 -TNF antiserum I 2.0- 1.5- 1.0- a.b 0,5. -l 0 r-‘-1 m 2'5‘ Pretreatment . EZJControl serum -TNF antiserum 2.0- a 1.5- 1.04 0.5.. b . .... .... s 1.5 6.0 Time (hr) after LPS 121 4.4.b Effect of pentoxifylline on circulating TNF- alpha concentration after LPS exposure A previous study showed that the concentration of circulating TNF-alpha was maximally increased 1.5 hr after administration of LPS to rats and that pretreatment with pentoxifylline attenuated this increase (251). In a confirmatory study (Table 4.1), a pronounced increase in plasma TNF-alpha concentration was observed 1.5 hr after administration of LPS to rats. This increase was markedly attenuated by pretreatment of rats with 100 mg/kg pentoxifylline. The concentration of TNF-alpha in plasma from rats pretreated with either pentoxifylline or saline vehicle was less than 1 ng/ml in the absence of LPS exposure (Table 4.1). 4.4.c Effect of pentoxifylline on LPS hepatotoxicity To test the possibility that pentoxifylline affords protection against LPS hepatotoxicity, rats were pretreated 1 hr prior to LPS administration with pentoxifylline or saline vehicle. The results from this study are shown in Figure 4.2. Rats pretreated with saline then treated with LPS exhibited marked elevations in plasma AST activity and total plasma bilirubin concentration by 6 hr after exposure to LPS. Both indices of liver injury were markedly attenuated in animals pretreated with pentoxifylline. Pentoxifylline did not affect either plasma AST activity or 122 Table 4.1 Effect of pentoxifylline on the LPS-induced increase in circulating TNF-alpha concentration. Pretreatment Treatment TNF-alpha (mg/m1) Saline Saline < 1 Pentoxifylline Saline < 1 Saline LPS 34 i 13 x 103‘! Pentoxifylline LPS 5.4 i 2.2 x 103°'b Rats were pretreated with either pentoxifylline (100 mg/kg, iv) or saline vehicle 1 hr prior to treatment with either LPS (5 mg/kg, iv) or saline vehicle. TNF-alpha was measured in plasma 1.5 hr after LPS exposure as described in Section 4.3.c. a, significantly different from respective saline treatment group. b, significantly different from respective vehicle- pretreated control. 123 Figure 4.2 Effect of pentoxifylline on LPS hepatotoxicity. Rats were pretreated with either pentoxifylline (100mg/kg, iv) or saline vehicle and then treated 1 hr later with either LPS (5mg/kg, iv) or saline vehicle. Liver injury was assessed by changes in plasma aspartate aminotransferase (AST) activity (A) and by changes in total plasma bilirubin concentration (B) 6 hr after exposure to LPS as described in Section 4.3.e. Results are expressed as means 1 SEM; N= 4 and 10 in the saline and LPS treatment groups, respectively. a, significantly different from respective saline treatment group. b, significantly different from respective vehicle- pretreated control. Plasma AST activity (SF U/ml x 103) Total plasma bilirubin (mg/dl) 124 Figure 4.2 2'0' Pretreatment E: Saline ‘ Pentoxifylline a 1 .5- l 1 .0- 0.5" a.b o r—fi m § 3'0“ Pretreatment a a 1:1 Saline Pentoxifylline 2.5~ * * 2.0. 1 .5- 1.0- « b 0.5- 0 l—.—1 : Saline LPS Treatment 125 total plasma bilirubin concentration in the absence of LPS exposure. 4.4.d Effect of PMN depletion on circulating TNF-alpha concentration after LPS exposure Results from Chapter 3 indicate that depletion of c ircul at ing PMNs a f f ords protect ion aga inst LPS hepatotoxicity (254). Accordingly, studies were performed to assess whether PMN depletion modulated the increase in circulating TNF-alpha concentration after LPS exposure. Pretreatment with anti-PMN Ig caused a marked reduction in circulating PMNs compared to control Ig-pretreated animals 1.5 hr after exposure to saline (77 i 40 and 1150 i 62 PMNs/ul, respectively). As indicated in 'Table 4.2, administration of LPS to rats pretreated with control Ig was associated with a pronounced increase in plasma TNF-alpha concentration 1.5 hr after LPS exposure. In comparison, this increase was more than 3-fold greater in PMN-depleted rats. 4.4.e Effect of pentoxifylline on hepatic PMN number Figure 4.3 shows the effect of pentoxifylline pretreatment on LPS-induced alterations in hepatic NPC and PMN numbers. NPC numbers were significantly increased 1.5 hr after administration of LPS to saline-pretreated rats (Figure 4.3A). A trend toward increased NPCs also occurred after exposure of pentoxifylline-pretreated rats to LPS: however, this did not attain statistical significance. The 126 Table 4.2 Effect of PMN depletion on the LPS-induced increase in circulating TNF-alpha concentration. Pretreatment Treatment TNF-alpha (rig/ml) Control Ig Saline < 1 Anti-PMN Ig Saline < 1 Control 19 LPS 46 i 8.5 x 10Ba Anti-PMN Ig LPS 177 i 40 x 103‘!” Rats were pretreated with either control Ig or anti-PMN 19 18 and 6 hr prior to treatment with either LPS (5 mg/kg, iv) or saline vehicle. TNF-alpha was measured in plasma 1.5 hr after LPS exposure as described in Section 4.3.c. Values represent the means : SEM; N=3-5. a, significantly different from respective saline treatment group. b, significantly different from respective control Ig pretreatment group. 127 Figure 4.3 Effect of pentoxifylline on hepatic nonparenchymal cell (NPC) and neutrophil (PMN) numbers after LPS exposure. Pretreatments with either pentoxifylline (100 mg/kg, iv) or saline vehicle were administered in the tail vein 1 hr prior to treatment with either LPS (5 mg/kg, iv) or saline vehicle. Hepatic NPCs (A) and PMNs (B) were quantified 1.5 hr after treatment with LPS as described in Section 3.3.h. Results are expressed as means 1 SEM; N=3. a, significantly different from respective saline treatment group. Total hepatic NPCs (cells/liver x 10°) Hepatic PMNs (cells/liver x 108) 4.0- 2.0- 128 Figure 4.3 Pretreatment [:1 Saline Pentoxyfilline 0 2'0“ Pretreatment I: Saline Pentoxyfilline a % i a 1 .0a 0 ' ' NS“ Saline LPS Treatment 129 LPS-induced increase in NPCs in both pretreatment groups was due largely to an increase in hepatic PMN numbers (Figure 4.3B). Hepatic PMN numbers in pentoxifylline-pretreated rats were not significantly different from those of rats pretreated with saline either in the presence or absence of LPS. Similarly, pretreatment of rats with antiserum to TNF- alpha did not significantly alter hepatic PMN accumulation 1.5 hr after LPS administration (11 i 1.3 x 107 and 8.7 1; 1.7 x 107 PMNs/liver in rats pretreated with control serum and TNF-alpha antiserum, respectively). 4.5 Discussion Exposure to LPS is associated with an elevation in circulating TNF-alpha concentration (251, Table 4.1), and evidence from numerous studies suggests that this cytokine may mediate many of the pathophysiologic alterations accompanying LPS exposure (199,93,258). The observation that antiserum to TNF-alpha afforded protection against LPS hepatotoxicity in galactosamine-sensitized. mice (201) suggested that TNF-alpha may play a role in the pathogenesis of LPS-induced liver injury. However, several differences exist between normal and galactosamine-sensitized animals with respect to LPS hepatotoxicity. For example, thromboxane synthase inhibitors afforded protection against LPS hepatotoxicity in rats in the absence of galactosamine sensitization, suggesting that the potent vasoactive 130 arachidonic acid metabolite, thromboxane A2, contributes to the pathogenesis of liver injury in these animals (215). In contrast, this arachidonic acid metabolite does not appear to contribute to LPS hepatotoxicity in galactosamine- sensitized animals, since cyclooxygenase inhibitors did not afford protection (211). Although differences in the mechanism of liver injury in nonsensitized and galactosamine-sensitized. animals may' exist, TNF-alpha appears to be an important mediator of LPS-induced liver injury in both instances. This is supported by results from the present study, which showed that LPS hepatotoxicity in rats was attenuated by pretreatment with TNF-alpha antiserum. Pretreatment with pentoxifylline also afforded protection against liver injury after exposure of rats to LPS. Pentoxifylline is a methylxanthine derivative which has been used clinically to treat circulatory disorders (259). In addition, it has been shown to protect against certain pathophysiologic alterations associated. with exposure to LPS. For example, pretreatment with pentoxifylline improves survival of rats exposed to lethal doses of LPS (251). It was proposed that this protective effect was due to the inhibition of TNF-alpha production since pentoxifylline markedly attenuated the rise in circulating concentration of TNF-alpha after exposure to LPS. Pentoxifylline also inhibited production of TNF-alpha from macrophages exposed to LPS in culture. This inhibitory effect was presumably 131 mediated by a cAMP-dependent mechanism (284) . Results from studies presented in this chapter confirm that pentoxifylline attenuates the increase in circulating TNF- alpha concentration caused by LPS (Table 4.1) and suggest further that TNF-alpha contributes to LPS-induced liver injury in rats. An argument can be made for both PMN-dependent and PMN- independent mechanisms in the TNF-alpha-mediated liver injury that occurs after LPS exposure. For example, results indicating that TNF-alpha contributes to cytotoxicity after cocultures of Kupffer cells and hepatic parenchymal cells are exposed to LPS raises the possibility that TNF-alpha mediates LPS-hepatotoxicity' by’ a direct, PMN-independent mechanism (255) . Conversely, a PMN-dependent mechanism is suggested by the observation that PMN depletion affords protection against LPS hepatotoxicity in vivo (254 and Chapter 3). In the present study, administration of LPS to PMN-depleted rats was associated with a marked increase in circulating TNF-alpha concentration compared to LPS-treated controls (Table 4.2). Inasmuch as the increase occured under conditions that afford protection against LPS hepatotoxicity (ie. PMN depletion), this result indicates that TNF-alpha does not directly mediate LPS-induced liver injury in vivo. Furthermore, it is consistent with the hypothesis that TNF- alpha contributes to liver injury by a PMN-dependent mechanism. Such an interaction between TNF-alpha and blood PMNs has been implicated in the pathogenesis of lung injury 132 after exposure of guinea pigs to recombinant human TNF-alpha (202). The mechanism for the marked increase in circulating TNF-alpha. concentration caused. by PMN’ depletion. in LPS- treated rats is not known. Since circulating TNF-alpha concentration was not increased by PMN depletion in the absence of LPS exposure, PMN depletion does not by itself stimulate TNF-alpha production. Kupffer cells are thought to be a major source of circulating TNF-alpha in rodents exposed to LPS, since LPS stimulates the production of TNF- alpha by rat Kupffer cells in vitro (255) and since TNF- alpha gene expression in Kupffer cells is temporally related to the increase in circulating TNF-alpha after exposure of mice to LPS in vivo (198). Several substances, including interferon-gamma and the yeast cell wall component, glucan, prime macrophages for production of TNF-alpha after exposure to LPS (259a,259b). Kupffer cells are fixed hepatic vascular phagocytic cells and may contribute to the removal of circulating immune complexes formed between PMNs and anti- PMN Ig during the process of PMN depletion. Thus, it seems possible that phagocytosis of immune complexes by Kupffer cells during PMN depletion may prime Kupffer cells for LPS- induced TNF-alpha production. Alternatively, PMN depletion may alter the clearance of TNF-alpha from the circulation. Further studies are required to evaluate these possibilities. 133 The nature of the interaction between TNF-alpha and blood PMNs in the pathogenesis of LPS hepatotoxicity is not known. Because TNF-alpha increases the adherence of PMNs to vascular endothelial cell monolayers in vitro (248,249,250), it seemed possible that TNF-alpha contributed to LPS-induced liver injury by mediating the accumulation of PMNs in the liver. However, although both TNF-alpha antiserum and pentoxifylline markedly attenuated LPS hepatotoxicity, neither significantly altered the accumulation of PMNs in the liver. This suggests that TNF-alpha does not contribute to LPS hepatotoxicity by recruiting PMNs into the liver. In addition, it is clear from this result that hepatic PMN accumulation alone was not sufficient for the full manifestation of liver injury. TNF-alpha can stimulate the release of reactive oxygen metabolites and lysosomal enzymes from PMNs in vitro (260). It also enhances the release of these cyototoxic agents from PMNs stimulated with chemotactic peptides and phorbol myristate acetate (261). Accordingly, TNF-alpha might contribute to LPS hepatotoxicity by stimulating or enhancing the release of cytotoxic substances from PMNs. In conclusion, these results provide evidence for a role for TNF-alpha in the pathogenesis of LPS hepatotoxicity. Furthermore, TNF-alpha appears to mediate liver injury after LPS exposure by an indirect mechanism which may involve interactions between TNF-alpha and PMNs. CHAPTER 5 ROLE OF THE COAGULATION SYSTEM IN LPS-INDUCED LIVER INJURY 134 135 5.1 Abstract The coagulation system has been implicated in the pathogenesis of liver injury after exposure to LPS. The purpose of this study was to test the hypothesis that the coagulation system contributes to LPS hepatotoxicity by a mechanism. which is dependent on circulating fibrinogen. Relationships between PMNs, TNF-alpha and the coagulation system were also examined. A marked reduction in plasma fibrinogen concentration occurred in rats after LPS exposure. This preceded the onset of liver injury. Pretreatment with heparin or warfarin attenuated the decrease in plasma fibrinogen and afforded protection against liver injury in LPS-treated rats. The decrease in circulating fibrinogen. after' LPS exposure *was also attenuated by depletion of circulating PMNs, pretreatment with pentoxifylline, or pretreatment with TNF-alpha antiserum, all of which protect against LPS hepatotoxicity. In contrast, pretreatment with ancrod 4 and 2 hr prior to LPS exposure, which reduced circulating fibrinogen to < 20 mg/dl in. rats ‘treated. with. LPS ‘vehicle, did. not. afford protection against LPS hepatotoxicity. Heparin did not prevent accumulation of PMNs in the liver after LPS exposure. These results suggest that the coagulation system 136 contributes to the pathogenesis of LPS-induced liver injury by a mechanism which is independent of circulating fibrinogen and recruitment of PMNs. Also, activation of the coagulation system after LPS is mediated by both PMNs and TNF-alpha. 5.2 Introduction Activation. of the coagulation system. is often associated with exposure to LPS. This is characterized in part by a decrease in circulating concentrations of clotting factors (204) and by an increase in the prothrombin and partial thromboplastin times (262,263). These LPS-induced changes are accompanied by a marked decrease in circulating fibrinogen concentration (111,204,262,264) and by the deposition of fibrin in the microvasculature of various tissues, including the kidneys (262,264,265), lungs (264,265) and liver (129,264,265). The. decrease in circulating fibrinogen after’ LPS exposure: is thought to reflect the proteolytic action of thrombin, which catalyzes the formation of fibrin monomers from fibrinogen (266). Evidence suggests that certain pathophysiologic alterations associated with exposure to LPS may be mediated by the coagulation system. Among these is injury to the liver. For example, pretreatment with the anticoagulant, heparin, prevented the reduction in circulating coagulation factors (204) and attenuated liver injury after LPS exposure 137 (102,203). Also, infusion of thrombin into the portal vein of rats produced morphologic changes in the liver, including fibrin deposition in the hepatic sinusoids and hepatocellular necrosis, which resemble those produced by portal venous infusion of LPS (129). Although these results point to the coagulation system in the pathogenesis of LPS hepatotoxicity, the mechanisms of liver injury by the coagulation system during LPS exposure have not been elucidated. Results from several studies suggest that the coagulation system may mediate tissue injury in certain instances by a mechanism which is dependent on fibrinogen. For example, pretreatment of sheep with the thrombin-like enzyme, ancrod, which depletes circulating fibrinogen without the concomitant formation of insoluble fibrin clots (267,268), afforded protection against lung injury after intravenous infusion of thrombin (269). Similarly, hirudin, an anticoagulant that binds to and inhibits thrombin (270), attenuated. the. decrease in plasma fibrinogen concentration, reduced the deposition of fibrin in the pulmonary microvasculature, and protected against lung injury after LPS exposure (111,264). As in the lungs, it seems possible that the coagulation system could contribute to LPS-induced liver injury by a mechanism which is dependent on fibrinogen. Alternatively, the coagulation system may contribute to LPS hepatotoxicity by a mechanism which is dependent on blood PMNs. Activated PMNs cause tissue injury under certain 138 circumstances, presumably through the release of cytotoxic substances such as oxygen metabolites and lysosomal enzymes (153). Among the early morphologic changes in the liver after LPS exposure is the accumulation of large numbers of PMNs (126,127), and these phagocytic cells contribute to the pathogenesis of LPS-induced liver (Chapter 3). Because PMN chemotactic factors are generated by activation of the coagulation system (271), it seemed possible that the coagulation system could mediate liver injury after LPS exposure by promoting the recruitment of PMNs. In addition to PMNs, TNF-alpha, plays an important role in LPS-induced hepatotoxicity (Chapter 4). Indeed, the full manifestation of liver injury after exposure to LPS appears to be dependent on an interaction between this cytokine and PMNs, inasmuch as preventing the rise in plasma TNF-alpha protected the liver but did not decrease hepatic PMN infiltration. Similarly, depletion of circulating PMNs afforded protection against LPS hepatotoxicity (Chapter 3) but. did not attenuate the increase in jplasma TNF-alpha (Chapter 4). Thus, both PMNs and TNF-alpha have important roles in LPS hepatotoxicity, but neither PMNs nor TNF-alpha alone are sufficient to cause liver injury during LPS exposure. TNF-alpha stimulates PMNs in 'vitro to release oxygen metabolites (260,261), which appear to activate the coagulation system during exposure of animals to LPS (272). Accordingly, it seems possible that an interaction between 139 PMNs and TNF-alpha could mediate liver injury after LPS exposure through activation of the coagulation system. The objective of the present study was to test the hypothesis that the coagulation system contributes to the pathogenesis of LPS-induced liver injury by a mechanism which is dependent on circulating fibrinogen and on recruitment of PMNs into the liver. The possible roles of PMNS: and. TNF-alpha. in activating’ the coagulation system during exposure to LPS were also examined. 5.3 Materials and methods 5.3.a Animals Female, Sprague-Dawley rats (Charles River, Crl:CD BR (SD) VAF/plus, Portage, MI) weighing ZOO-2509 were used in all studies. Rats were maintained on a 12 hr light-dark schedule for 1 week prior to use. Food (Wayne Lab-Blox, Allied Mills, Chicago, IL) and water were allowed ad libidum throughout the studies. 5.3.b Treatment protocols Heparin (Type II), warfarin [3-(alpha-acetonylbenzyl)- 4-hydroxycoumarin], ancrod (from Agkistrodon rhodostoma venom), and pentoxifylline were obtained from Sigma Chemical Company (St. Louis, MO). An initial study was performed to evaluate the temporal relationship between activation of the coagulation system 140 and the onset of liver injury after exposure to LPS. Activation of the coagulation system and liver injury were quantified 0, 1.5, 3, and 6 hr after exposure to LPS by changes in plasma fibrinogen concentration and by changes in plasma AST activity and total plasma bilirubin concentration, respectively (see below). Studies were also performed to assess the effects of the anticoagulants, heparin and warfarin, on LPS-induced activation of the coagulation system and liver injury. Heparin (2000 U/kg) or its saline vehicle was administered in the tail vein 0.5 hr prior to LPS exposure. The maximum effects of warfarin on the coagulation system in rats were observed between 16 and 24 hr after po administration of warfarin (273). Thus, warfarin (20 mg/kg, po) or its saline vehicle was administered 16 hr prior to exposure to LPS. To assess the effects of defibrinogenation on LPS hepatotoxicity, rats were pretreated with ancrod. Ancrod (50 U/kg per administration) or its saline vehicle was administered in the tail vein 4 and 2 hr prior to LPS exposure. This protocol causes an 89 % reduction in plasma fibrinogen concentration in rats 4‘ hr after the initial exposure to ancrod (274). In a study to test the effect of depletion of blood PMNs on the LPS-induced activation of the coagulation system, rats were pretreated with the total Igs fraction of serum from rabbits immunized with rat PMNs (anti-PMN Ig) (see below). Total Ig fraction of serum from non-immunized 141 rabbits was used as a control (control Ig). Igs were administered in the tail vein of rats in a volume of 0.5 ml 18 and 6 hr prior to exposure to LPS. This protocol markedly reduces circulating PMNs and affords protection against LPS hepatotoxicity (Chapter 3). The effect of pentoxifylline or antiserum to TNF-alpha on the coagulation system was also assessed after LPS exposure. Pentoxifylline (100 mg/kg) or saline vehicle was administered in the tail vein 1 hr prior to LPS administration. Neutralizing antiserum to recombinant mouse TNF-alpha, which was raised in rabbits and which cross- reacts with rat TNF-alpha (275), was a gift from Dr. Steven L. Kunkel. It was diluted 1:1 in sterile saline and administered in a 2 ml volume in the tail vein of rats 1 hr prior to LPS exposure. Serum from non-immunized rabbits was used as a control. These treatments obtund the increase in plasma TNF-alpha that occurs after LPS administration (Chapter 4). In the above studies, LPS (E. coli 0128:B12, Sigma Chemical Co, St. Louis, MO) or saline vehicle were administered in the tail vein of rats in a volume of 5 ml/kg. Rats were anesthetized with sodium pentobarbital (50 mg/kg, ip) 6 hr after exposure to LPS unless stated otherwise. Blood samples anticoagulated with sodium citrate (1/10 dilution in whole blood of 3.8 % w/v sodium citrate in DDW) were obtained from the inferior vena cava for the 142 quantification of plasma fibrinogen concentration and liver injury. 5.3.c Quantification of plasma fibrinogen concentration and liver injury Changes in plasma fibrinogen concentration were measured to quantify activation of the coagulation system after LPS exposure and to verify defibrinogenation in studies with ancrod. Plasma fibrinogen concentration was determined from the thrombin clotting time of diluted samples using a fibrometer and a commercially available kit (Data-Fi, B4233-15, American Dade, Aguada, Puerto Rico). The procedure is based on that described by Clauss (276). Liver injury was quantified by changes in plasma AST activity and total plasma bilirubin concentration as described in Section 3.3.d. 5.3.d Anti-PMN Ig preparation Igs to rat peritoneal PMNs were raised in New Zealand white rabbits as detailed in Chapter 3 (254). The total Ig fraction of serum from immunized (anti-PMN Ig) and non- immunized (control Ig) rabbits was isolated by ammonium sulfate precipitation (221 and Chapter 3). 5.3.e Quantification of hepatic NPC and PMN number To assess the role of the coagulation system in hepatic PMN infiltration after LPS exposure, rats were pretreated 143 with heparin or saline vehicle and treated with LPS as described above. Hepatic PMNs were quantified 1.5 hr after LPS exposure as detailed in Chapter 3 (254). Rats were anesthetized with sodium pentobarbital (50 mg/kg, ip), and 40 ml of ca2+-free Hank's balanced salt solution (HBSS, pH 7.4) containing EGTA (0.5 mM) and HEPES (0.5 M) was infused in situ through a cannula placed in the portal vein to flush the liver of blood. The liver was removed, minced thoroughly with scissors and incubated for 1-1.5 hr at 37°C in Ca2+- free HBSS containing heparin (1000 U/ml, grade II, Sigma Chemical Co, St. Louis, MO) and Pronase E (0.2 % w/v, Type XIV protease, Sigma Chemical Co, St. Louis, M0) to lyse parenchymal cells. The remaining cells, which include hepatic NPCs and PMNs, were pelleted in a centrifuge (500xg, 20 min), washed twice with HBSS, and resuspended in 25 ml Ca2+-free HBSS containing heparin. Total NPCs were quantified using a Unopette (test 5859, Becton-Dickinson, Rutherford, NJ) and a hemocytometer. The 2+-free HBSS containing 1 % NPC isolates were diluted in Ca BSA (Sigma Chemical Co, St. Louis, MO) and concentrated on a microscope slide using a cytocentrifuge (Cytospin 2, Shandon Southern Inc, Sewickley, PA). The fraction of‘ PMNs was quantified by differential count after the slides were stained with a modified Wright's stain. The total number of NPCs was multiplied by the fraction of PMNs to obtain the number of PMNs/liver. 144 5.3.f Data analysis Data from studies to determine changes in circulating fibrinogen and total plasma bilirubin concentration with time after LPS exposure were anaylzed using the rank sums test and Bonferroni’s correction factor. A one-way, completely random ANOVA was used to analyze time-dependent changes in plasma AST activity after LPS exposure. Comparisons among group means were made using the least significant difference test (277). Data from studies to test the effects of heparin and warfarin on LPS-induced changes in plasma AST activity and total plasma bilirubin concentration were analyzed using the rank sums test (277). Data from studies of the effect of ancrod on changes in plasma AST activity and total plasma bilirubin as well as from studies of the effects of heparin, warfarin, ancrod and pentoxifylline on circulating fibrinongen concentration were analyzed using a 2 x 2 factorial, completely random ANOVA. Changes in hepatic PMN number after administration of LPS to rats pretreated with heparin were also assessed using a 2 x 2 factorial, completely random ANOVA. ANOVAs were performed on log transformed data in instances in which variances were non- homogeneous. Comparisons among group means were made using the least significant difference test. Results are expressed as means + SEM. The criterion for significance was p < 0.05 for all studies. 145 5.4 Results 5.4.a Role of the coagulation system in LPS hepatotoxicity The temporal relationship between liver injury and activation of the coagulation system is shown in Figure 5.1. Liver injury, as measured by elevations in plasma AST activity and total plasma bilirubin concentration, occurred between 3 and 6 hr after LPS exposure (Figure 5.1A and B). Activation of the coagulation system preceded the onset of liver injury as indicated by the marked decrease in plasma fibrinogen concentration that occurred between 1.5 and 3 hr after LPS administration (Figure 5.1C). Circulating fibrinogen concentration was reduced > 85 % 6 hr after exposure to LPS. The effects of anticoagulants on LPS-induced liver injury and activation of the coagulation system are shown in Figures 5.2 and 5.3 and in Table 5.1, respectively. Pretreatment with heparin attenuated the elevation in plasma AST activity and total bilirubin concentration 6 hr after administration of LPS (Figure 5.2). This was associated with a complete inhibition of the LPS-induced reduction of circulating fibrinogen concentration (Table 5.1). Like heparin, warfarin afforded protection against liver injury (Figure 5.3) and attenuated the LPS-induced reduction in circulating fibrinogen :concentration (Table 5.1). In contrast to heparin, a significant decrease in circulating 146 Figure EL]. Development of liver injury and activation of the coagulation system after LPS exposure. LPS (5 mg/kg, iv) or saline vehicle was administered at 0 hr. Liver injury and activation of the coagulation system.‘were quantified. by changes in plasma aspartate aminotransferase (AST) activity (A) and total plasma bilirubingconcentration (B) and by changes in plasma fibrinogen concentration (C), respectively, as described in Section 5.3.c. Results are expressed as the means i SEM; N=3-6. a, significantly different from the respective value at 0 hr. b, significantly different from the respective vehicle- treated control. 147 Figure 5.1 Time (hr) after LPS .M- II. 0 .. M. u 0 A! r A. ”a . 4 a . a + . .fi. u. m m . m all: b n a." an . am em 0 e r 0 e r 0 e : __ __ 0 e 0 e 0 e u d o: e a u I e d d d On 4 d d I 3 o: o. q a .. u . a .... .. a . z .. #2 a .5} he 55%.. 54 2.8: Assay crass 25% .38 f: x secs 5623: 2:8... A B C ....H... - u t .: .m. l .... m... w a m. .... .a .1... 148 Figure 5.2 Effect of heparin on LPS-induced liver injury. Rats were pretreated with either heparin (2000 U/kg, iv) or saline vehicle 30 min prior to treatment with either LPS (5 mg/kg, iv) or saline vehicle. Liver injury was quantified 6 hr after LPS exposure by changes in plasma aspartate aminotransferase (AST) activity (A) and total plasma bilirubin concentration (B) as described in Section 5.3.c. Results are expressed as means i SEM; N=5-10. a, significantly different from the respective saline treatment group. b, significantly different from the respective vehicle- pretreated control. 149 A Figure 5.2 A 1'5" Pretreatment no l:lSaline v- ‘ -Heparin 1‘ a E 3 1.0- L 21 g. - .2 *6 ° 05: infl'li 5 Hi: 2 2: g a.b , IT. . o m m m ‘3er pile: B 3.3.3 2'0' Pretreatment l:lSaline o -Heparin '0 a a \E’ 1.5“ g- C .5 .3 g 1.0- O E m 2 3 0.5- .3 :9 b o Ham Saline LPS Treatment 150 Figure 5.3 Effect of warfarin on LPS-induced liver injury. Rats were pretreated with either warfarin (20 mg/kg, po) or distilled water (DDW) vehicle 16 hr prior to treatment with either LPS (5 mg/kg, iv) or saline vehicle. Liver injury was assessed 6 hr after LPS exposure by changes in plasma aspartate aminotransferase (AST) activity (A) and total plasma bilirubin concentration (B) as described in Section 5.3.c. Results are expressed as means : SEM; N=6-11. a, significantly different from the respective saline treatment group. b, significantly different from the respective vehicle- pretreated control. 'er iiffli k9: PO}: :tment i3 injul‘l'i‘ in phi and iii in SEC-:3 .1. .ine liCle' Plasma AST activity (sr U/ml x 103) Total plasma bilirubin (mg/0|) 151 Figure 5.3 1 '5' Pretreatment 1:] DDW . Worfari n a 1 .0- "l— 0.5- ‘ a.b O r—L m g 2'5“ Pretreatment , 1:] DDW Worfarin a 2.0- l 1.5- 1 .0- i a.b 0.5- (3 l l .rtsiitsrl Saline LPS Treatment 152 Table 5.1 Effect of anticoagulants on circulating fibrinogen concentration in the presence and absence of LPS exposure. Pretreatment LPS Plasma fibrinogen (mg/d1) Heparin - - 196 :_14(4) + - 195 i 23(4) - + 46 : 17(7)a + + 207 1 13(8) Warfarin - - 182 i 26(6) + - 194 i 10(6) - + 40 i 10(12)a + + 97 i 10(10)"'b Ancrod - - 147 1 12(3) + - 18 i 3(3)b - + 38 i 13(7)a + + 12 1 1(8) Rats were pretreated with either heparin (2000 U/kg, iv), warfarin (20 mg/kg, po), ancrod (50 U/kg x 2, iv) or their respective vehicles (-) and then treated with either LPS (+) or saline vehicle (-). Plasma fibrinogen was quantified 6 hr after exposure to LPS (5 mg/kg, iv) as described in Section 5.3.0. Results are expressed as means : SEM. Number of measurements are in parentheses. a, Significantly different from the respective saline treatment group. b, Significantly different from respective vehicle- pretreated control. 153 fibrinogen concentration occurred in warfarin-pretreated rats after LPS exposure. However, this decrease was less than that which occurred in LPS-exposed rats that received no warfarin. 5.4.b Mechanism of liver injury by the coagulation system after LPS exposure Ancrod causes a marked reduction in circulating fibrinogen concentration without significantly altering circulating concentrations of other clotting factors (278). Rats were pretreated with ancrod to assess the role of circulating fibrinogen in the pathogenesis of liver injury after LPS exposure. Ancrod pretreatment reduced circulating fibrinogen concentration by more than 85 % in the absence of LPS exposure (Table 5.1) but did not cause hepatotoxicity (Figure 5.4). Elevations in plasma AST activity and total plasma bilirubin concentration produced by administration of LPS were not significantly attenuated by pretreatment of rats with ancrod. In an earlier report, we presented evidence indicating that PMNs mediate liver injury from LPS (254 and Chapter 3). It seemed possible that activation of the coagulation system by LPS exposure might contribute to recruitment of PMNs into the liver. This possibility was tested by examining hepatic PMN accumulation in LPS-exposed rats after pretreatment with heparin anticoagulant. The results from this study are shown in Figure 5.5. A significant increase in hepatic PMNs was 154 Figure 5.4 Emfect of ancrod on lPS-induced liver injury. Rats were pretreated with ancrod (50 U/kg, iv) or saline vehicle 2 and 4 hr prior to treatment with either LPS (5 mg/kg, iv) or saline vehicle. Plasma aspartate aminotransferase (AST) activity (A) and total plasma bilirubin concentration (B) were quantified 6 hr after LPS exposure as described in Section 5.3.c. Results are expressed as the means : SEM; N=3-8. a, significantly different from the respective saline treatment group. Plasma AST activity (SF U/ml x 103) Total plasma bilirubin (mg/dl) 155 Figure 5.4- 1 '5‘ Pretreatment a l: Saline ‘ -Ancrod 1.0a 0.5- o I l m 2'5' Pretreatment a 4 El Saline -Ancrod 2.0- a 1.5+ .‘L w 1.0- 0.54 o f—L'I m Saline LPS Treatment 156 .osono ucesuaenu ennaem e>uuoeomen Bonn uceneumuo xauceouuucmum .e .mleuz «2mm H mceea we oemmenmxe ene muaamem .e.m.m one nlc.m.m chnuoem an oecnnomeo me enamooxe meg neune nc m.a oeuuuuceao ene3 mnecesc 22m omuemem .N.m enomflm Cu oecnnomeo mm eaoece> essaem no mod necune cun3 oeueenu cecu one eaowce> ennamm no auneoec necune cuns oeueenueno ene3 muem .couuenumncHEoe men neume mnecfisc €va uncoonusec owuemec co suneoec no uoenum m.m ensmum cceEuoefi csam 157 EFL crane: a ec__om _IIU uceEcoebeta no 83$ 1 L“. 0 (9m x JeAll/SIIGO) SNWd allodeH r O. 158 observed after administration of LPS alone. The accumulation of PMNs was not significantly attenuated by pretreatment with. heparin. Thus, heparin affords protection. from LPS hepatotoxicity but does not prevent influx of PMNs into the liver. 5.4.c Mechanism of activation of the coagulation system after LPS To test the possibility that activation of the coagulation system during LPS exposure is dependent on PMNs, the effects of PMN depletion on the LPS-induced reduction in plasma fibrinogen concentration was examined. The results from this study are shown in Figure 5.6. A time-dependent decrease in plasma fibrinogen concentration was observed in LPS-treated rats that were pretreated with control Ig. Plasma fibrinogen concentration was maximally decreased 6 hr after LPS exposure and returned to pre-exposure values thereafter. Pretreatment of rats with anti-PMN Ig resulted in a modest but statistically significant elevation in plasma fibrinogen concentration prior to LPS exposure and attenuated the decrease in plasma fibrinogen concentration 6 hr after administration of LPS. Pretreatment of rats with either pentoxifylline or antiserum to TNF-alpha obtunds the LPS-induced increase in circulating TNF-alpha concentration and affords protection against LPS hepatotoxicity (Chapter 4). Rats were pretreated with either pentoxifylline or antiserum to TNF-alpha to 159 .msono uceaueenuenm mH Honucoo e>wuoemmen ecu Bonn ucenemuuo hauCMOHuwcoum .8 .nc o um esHe> e>wuoeomen ecu Eonu ucenewuflo wauceOfluwcmHe .e .mnz «Sum H names we oemmenexe ene muaamem .co.m.m couuoem eemv cofluenuceocoo ceoocuncuu mameae :H meoceco xc oeuuuuceso me3 A>u .ox\ma my men cuu3 uceaueenu neune meeuu ceueomocH um Eeumwm cauueasomoo ecu mo canue>wuo¢ .c.m.m cauuoem Cu oecunomec we OH gmlwuce no oH Honucoo necuue cufi; oeueenuenm ene3 muem .enamooxe won neuue Eeumwm couueaammoo ecu mo scuueefiuom co szzmv mancoonuaec ocwueasonua mo coaueameo mo uoemum o.m enamflm 160 Flgure 5 6 a Time (hr) after LPS -95 °7 5'65 5%.}: .o 30 4 till 3'7 ' f2- 7 SE. ’ 53 ' o (30" x lp/fiul) uefioullqll owsold 24 12 161 examine the possibility that TNF-alpha contributes to activation of the coagulation system after LPS exposure. Pentoxifylline pretreatment attenuated the decrease in circulating fibrinogen concentration 6 hr after LPS administration (Figure 5.7) . Similarly, pretreatment with antiserum to TNF-alpha significantly attenuated the LPS- induced decrease in circulating fibrinogen (19 i 5 and 60 i 3 mg/dl plasma in LPS-treated rats pretreated with control serum or TNF-alpha antiserum, respectively). 5.5 Discussion Results from the this study indicate that activation of the coagulation system after LPS exposure occurs prior to the onset of liver injury. Furthermore, inhibition of the coagulation system by pretreatment with either heparin or warfarin afforded protection against LPS-induced liver injury. These results support the hypothesis that the coagulation system is important in the pathogenesis of LPS hepatotoxicity. It has been suggested that circulatory disturbances from occlusive fibrin thrombi in the hepatic microvasculature contribute to the pathogenesis of LPS- induced liver injury (129). However, pretreatment with ancrod, which markedly reduced circulating fibrinogen concentration, did not afford protection against liver injury after LPS exposure. Since ancrod acts by converting 162 .Honucoo oeueenuenoleHOMce> e>uuaeomen ecu Eonn ucenenuwo hauGMOHmucon .n .osono uceaueenu essaem e>fluoeomen ecu Eonu ucenemuflo >Huceouuucomm .e .oalvuz “2mm + mcmee we oemmenmxe ene muasmem .o.m.m cauuoem cw .oecunomec_ me mad mo cowuenumwcuaoe neuue nc o ceoocwncuu nausea Cw meoceco ac oemmemwe me3 Eeumxm cowueasmeoo ecu mo cauum>nu0< .eaoncecw ennaem no 2% .919: me when necune cunz ucesueenu ou nanno nc H eaonce> ecuaem no A>n .mx\ofi ooav ecuafiauuxoucem necune cud; oeueenueno ene3 muem .Eeumenm coaueasmeoo ecu uo cauue>uuoe oeosocfllmmq co ecflaaxuuxouceo mo uoeuum >.m ensmflm Figure 5.7 s/ [le Saline Q (ZOL x |p/5u.l) uefiouqull DLUSD|d LPS Treatment 164 fibrinogen to soluble fibrin monomers which do not form clots (267,268), this result argues against the role of fibrin thrombi-induced circulatory disturbances in LPS- induced liver injury and strongly suggests that the coagulation system contributes to LPS hepatotoxicity by a mechanism which is independent of fibrinogen. Taken together, the results with ancrod, heparin and warfarin suggest that components of the coagulation pathway proximal to fibrinogen are important in the pathogenesis of liver injury. Because intraportal infusion of thrombin produces liver injury which resembles that caused by LPS (129), it seems possible that thrombin, independent of its action on fibrinogen, may contribute to LPS hepatotoxicity. Thrombin has a number of biologic activities that are independent of its proteolytic action on circulating fibrinogen. Among these is the stimulation of aggregation of and thromboxane A2 release from platelets (279,280,281,282). Circulating concentrations of thromboxane are markedly increased within 2 hr after exposure to LPS (214), and this vasoactive arachidonic acid metabolite may contribute to the pathogenesis of LPS-induced liver injury (212,215). Thus, thrombin might play an important role in LPS hepatotoxicity by stimulating release of thromboxane A2 from platelets. Thrombin also binds to specific receptors on the surface of hepatic parenchymal cells (283). However, the pathophysiologic consequenses of this binding remain to be determined. 165 Exposure to LPS is associated with the accumulation of large numbers of PMNs in the liver (Chapter 3 and Figure 5). Because PMNs have been implicated in the pathogenesis of liver injury after LPS exposure (Chapter 3), it seemed possible that the coagulation system could contribute to LPS hepatotoxicity by promoting the accumulation of PMNs in the liver. However, pretreatment with heparin, which inhibited the reduction in circulating fibrinogen after LPS exposure and afforded protection against liver injury, did not significantly reduce accumulation of PMNs in the liver. Thus, the coagulation system does not appear to contribute to the pathogenesis of LPS hepatotoxicity solely by mediating the recruitment of PMNs. It has been suggested that leukocytes may be involved in the activation of the coagulation system during LPS exposure (263). In the present study, depletion of circulating PMNs markedly attenuated the decrease in circulating fibrinogen concentration associated with LPS exposure. This suggests that activation of the coagulation system after exposure to LPS is mediated by PMNs. The mechanism of activation of the coagulation system by PMNs after LPS administration is not known. The prolongation of prothrombin and partial thromboplastin times, reduction in circulating fibrinogen concentration, and accumulation of fibrin in tissues after LPS exposure are attenuated by administration of superoxide dismutase and catalase (272). This suggests that oxygen metabolites contribute to the 166 activation of the coagulation system after LPS exposure. Because extracellular release of reactive oxygen metabolites can accompany activation of PMNs (153), it is possible that PMNs mediate activation of the coagulation system during LPS exposure by an oxygen radical-dependent mechanism. Additional studies are necessary to determine the role of PMN-derived oxygen metabolites in activation of the coagulation system after LPS exposure. Results from several studies suggest that TNF-alpha contributes to the pathogenesis of liver injury during LPS exposure. For example, neutralization of circulating TNF- alpha with antiserum to this cytokine afforded protection against LPS-induced liver injury (252,254). Pentoxifylline, which inhibits TNF-alpha production from LPS-stimulated macrophages in vitro and attenuates the rise in circulating TNF-alpha in vivo (251,284), also afforded protection against LPS hepatotoxicity (Chapter 4) . Both pentoxifylline and TNF-alpha antiserum attenuated the decrease in circulating fibrinogen concentration after LPS exposure. This result suggests that TNF-alpha may contribute to the pathogenesis of liver injury by mediating the activation of the coagulation system. TNF-alpha directly induces procoagulant activity from vascular endothelial cells in culture (189). Perhaps TNF- alpha contributes to activation of the coagulation system after LPS exposure in vivo by a similar mechanism. However, PMN depletion enhanced circulating TNF-alpha concentration 1 L.— {ca-nu 167 by more than 3-fold during LPS exposure (Chapter 4) but attenuated the decrease in circulating fibrinogen concentration. This indicates that TNF-alpha by itself is not sufficient to activate the coagulation system during LPS exposure. Furthermore, this result suggests that TNF-alpha promotes activation of the coagulation system by a PMN- dependent mechanism. In summary, these results indicate that the coagulation system contributes to the pathogenesis of liver injury after LPS exposure by' a mechanism *which is not dependent on circulating fibrinogen or the recruitment of PMNs. In addition, results from PMN depletion studies and studies with pentoxifylline and TNF-alpha antiserum suggest that activation of the coagulation system after exposure to LPS is dependent on both PMNs and TNF-alpha. CHAPTER 6 SUMMARY AND CONCLUSIONS 168 169 The overall aim of this dissertation was to examine the mechanisms of LPS-induced liver injury. Specifically, the role of host-derived inflammatory mediators in the pathogenesis was assessed. Liver injury was quantified in most studies by changes in liver specific enzymes activities in the plasma, which are indicative of hepatic parenchymal cell injury, and by changes in total plasma bilirubin concentation, which is consistent with alterations in parenchymal cell function and cholestasis. Evidence from the studies presented in this dissertation suggests that 1) liver injury after LPS exposure is mediated. largely by certain host-derived factors including circulating PMNs, TNF-alpha and the coagulation system, and 2) that full manifestation of LPS-induced liver injury is dependent on complex interactions between these endogenous factors. 6.1 Host-derived mediators in the pathogenesis of liver injury after exposure to LPS Liver injury, as measured by changes in total plasma bilirubin concentration and liver-specific enzyme activities in the plasma, occurred between 3 and 6 hr after administration of LPS in the tail vein of rats (Figures 3.7 and 5.1). 170 6.1.a Role of blood PMNs in LPS-induced liver injury The results from studies presented in Chapter 3 indicated that large numbers of PMNs accumulate in the liver prior to the onset of liver injury and that depletion of circulating PMNs afforded protection against liver injury after LPS exposure. These results support the hypothesis proposed in Specific Aim 1 (Section 2.4.a) which stated that LPS-induced liver injury was dependent on circulating PMNs. 6.1.b Role of TNF-alpha in LPS-induced liver injury Circulating TNF-alpha concentration was markedly increased after exposure to LPS (Chapter 4). This increase preceded the onset of liver injury suggesting that TNF-alpha contributes to the pathogenesis. Consistent with this are results from studies with pentoxifylline which attenuated the rise in circulating TNF-alpha concentration after LPS exposure and protected against LPS hepatotoxicity (Chapter 4). Neutralization of circulating TNF-alpha using a specific antiserum also afforded protection against LPS hepatotoxicity. 6.1.c Role of the coagulation system in LPS-induced liver injury In addition to PMNs and TNF-alpha, results from studies presented in chapter 5 indicated that the coagulation system contributes to the pathogenesis of LPS-induced liver injury. Activation of the coagulation system by LPS administration 171 occurred prior to the onset of liver injury. Anticoagulants inhibited activation of the coagulation system and afforded protection against liver injury after LPS exposure. Although much remains unknown about the mechanism of liver injury by the coagulation system after LPS exposure, the actions of specific anticoagulants pointed to the importance of a coagulation factor(s) proximal to fibrinogen in the coagulation cascade. It is tempting to speculate that this factor may by thrombin. 6.2 Relationship between host-derived mediators in LPS hepatotoxicity. 6.2.a Relationship between PMNs and TNF-alpha An interaction between PMNs and TNF-alpha appears to be necessary for the full manifestation of LPS hepatotoxicity. For example, PMN depletion afforded protection against LPS- induced liver injury but did not prevent the LPS-induced increase in circulating' TNF-alpha concentration. Indeed, exposure to LPS after PMN depletion was associated with a > 3-fold enhancement of TNF-alpha concentration in plasma. Conversely, attenuation of the increase in plasma TNF-alpha concentration after LPS exposure protected against liver injury but did not prevent the accumulation of PMNs in the liver. Taken together, these results suggest that neither TNF-alpha nor’ PMNs. alone are sufficient. to: cause liver injury after exposure to LPS. 172 6.2.b Relationship between PMNs, TNF-alpha and the coagulation system In addition to an interaction between PMNs and TNF- alpha, evidence from chapter 5 suggests that both PMNs and TNF-alpha interact with the coagulation system after LPS exposure. This is suggested by studies which showed that activation of the coagulation system was inhibited by PMN depletion or by manipulations that attenuated the increase in plasma TNF-alpha concentration after LPS administration. The nature of the interactions between PMNs, TNF-alpha and the coagulation system in the pathogenesis of LPS hepatotoxicity remains unknown. TNF-alpha promotes the release of reactive oxygen metabolites from PMNs in vitro. Because reactive oxygen metabolites have been implicated in the activation of the coagulation system after LPS exposure, it seems possible that TNF-alpha may contribute to the activation of the coagulation system by inducing PMNs to release reactive oxygen metabolites. Further studies are necessary to test this possibility. The mechanism of liver injury by the coagulation system after LPS exposure remains to be elucidated. However, this system does not appear to contribute to liver injury by promoting hepatic PMN accumulation since heparin afforded protection against LPS hepatotoxicity without attenuating PMN accumulation in the liver. Also, because defibrinogenation with ancrod did not afford protection against LPS hepatotoxicity, the coagulation system probably 173 does not mediate liver injury by a mechanism which is dependent on the thrombin-catalyzed conversion of fibrinogen to fibrin monomers. This result argues against the hypothesis proposed in Specific Aim 3 which stated that the coagulation system mediates liver injury by a mechanism which is dependent on circulating fibrinogen. Thrombin has a number of biologic activities which appear to be independent of its proteolytic action on circulating fibrinogen. Among these is the stimulation of thromboxane A2 release from platelets. Evidence from several studies suggests that this potent, vasoactive arachidonic acid metabolite may contribute to the pathogenesis of LPS- induced liver injury. Although the source of thomboxane A2 is not known, it seems possible that it could be derived from activated platelets. Indeed, an early morphologic change in the hepatic sinusoids after LPS administration is the appearance of aggregates of activated platelets. Thus, the elaboration of thrombin during activation of the coagulation system after LPS exposure could result in liver injury by a mechanism which was dependent on platelet- derived thromboxane A2. Further studies are necessary to test this possibility. 6.3 Proposed mechanism of LPS hepatotoxicity: an hypothesis The results from this dissertation indicate that PMNs, TNF-alpha and the coagulation system contribute to the 174 pathogenesis of LPS hepatotoxicity. Furthermore, full manifestation of liver injury after LPS exposure appears to be dependent on complex interactions between these host- derived factors. Figure 6.1 summarizes the results from this dissertation and shows proposed relationships between PMN activation, TNF-alpha release and activation of the coagulation system in LPS-induced liver injury. Exposure to LPS is associated with the accumulation of PMNs in the liver and increased circulating TNF-alpha, which is probably derived from hepatic Kupffer cells. The mechanism of recruitment of PMNs in the liver remains unknown. However, it does not appear to be mediated by the coagulation system since hepatic PMN accumulation was not prevented by heparin. Both blood PMNs and TNF-alpha appear to be required for activation of the coagulation system after LPS exposure. Because TNF-alpha stimulates PMNs to release oxygen metabolites in vitro and because reactive oxygen metabolites have been implicated in the activation of the coagulation system after LPS exposure, TNF-alpha may contribute to the activation of the coagulation system by inducing PMNs to release reactive oxygen metabolites. The coagulation system mediates liver injury by an unknown mechanism which might involve thrombin but which is independent of the thrombin-catalyzed conversion of fibrinogen to fibrin monomers. 175 .memeneumcenuocwse ecuceae\eueuneome .BA<\Bm< «ecQHeInouoeu mumonoec nossu .mze “mawcoonucec .zzm umaaeo Hefiaecuooce .Um umHHeo newness .ox ..meuhooueoec .Om “eounecooemhaomooua .mmq "ecofiuefirenccd $3 coHuenuceocoo cflcsnuauc .oce hufifiuoe emeneumcenuocuae cemeenocw me eaceuoeueo H3 hnoflcfl Haeo deacocenem ou .mEchccer oeCuueo >Hnooo >c .omea aauceooemcam Eeumxm cowueasmeoo ecu aonn oe>nneo mnouoem .emncecoefi uceoceoeolea>nce Hmsomomaa no lHeofloen ceo>xo cm >c wacwmmom .3; seumxm couueadoeoo ecu no c0uue>wuoe ou mowed canue>nuoe 22m .ne>: ecu nu mzznm no H3 cowue>nuoe one Hm“ ucefiuwsnoen meuoaono caucus enamelmze no emeeaen ecu Cu ocwuasmen .3 Ge: 330 .8392 maucfiuoc mag acnucnsonno .mnmecuoamc cm ">nsfl2n ne>ua oeoaocnlmmq no Emncecoee oemooonm .H.o ensmum 176 _ . _ _\||._ 0 1 2 0 an sagas a 5355 582m 283:5 e O S was“. 177 The results described in this dissertation provide new insight into the mechanisms of liver injury from bacterial LPS by showing that blood PMNs, circulating TNF-alpha and the coagulation system each play important roles. While much remains unknown about the specific mechanisms by which each of these factors is involved, it is clear that they interact to result in liver injury. In addition, evidence suggests that PMNs, TNF-alpha, and the coagulation system may contribute to other pathophysiologic alterations associated with systemic endotoxemia. Among these is injury to the lungs. Thus, the results from studies in this dissertation may provide insight into the mechanisms contributing to LPS- induced alterations in extrahepatic tissues. LI ST OF REFERENCES 178 LIST OF REFERENCES 1. Morrison, D.C., and R.J. Ulevitch. 1979. The effects of bacterial endotoxins on host mediation systems. Am. J. Pathol. 93:527-616. 2. Rietschel, E.T., and C. Galanos. 1977. Lipid. A antiserum-mediated protection against lipopolysaccharide- and lipid A-induced fever and skin necrosis. Infect. Immun. 15:34-49. 3. Corrigan, J.J. and J.F. Kiernat. 1979. Effect of polymyxin B sulfate on endotoxin acitivity in a gram- negative septicemia model. Pediatr. Res. 13:48-51. 4. Tauber, M.B., A.M. Shibl, C.J. Hachkbarth, J.W. Larrick, and M.A. Sande. 1987. Antibiotic therapy, endotoxin concentration in cerebrospinal fluid, and brain edema in experimental Escherichia coli meningitis in rabbits. J. Infect. Dis. 156:456-462. 5. Berman, N.S., S.E. Siegel, R. Nachum, A. Lipsey, and J. Leedom. 1976. Cerebrospinal fluid endotoxin concentration in gram-negative bacterial meningitis. J. Pediat. 88:553- 556. 6. Anderson, B.M., and O. Solberg. 1980. The endotoxin- liberating effect of antibiotics on meningoccocci in vitro. Acta Pathol. Microbiol. Scand. [B] 88:231-236. 7. Goto, H., and. S. Nakamura. 1980. Liberation of endotoxin from Escherichia coli by addition of antibiotics. Japan J. Exp. Med. 50:35-43. 8. Shenep, J.L., and K.A. Morgan. 1984. Kinetics of endotoxin release during antibiotic therapy for experimental gram-negative bacterial sepsis. J. Infect. Dis. 150:380-388. 179 9. Shenep, J.L., R.P. Barton, and K.A. Morgan. 1985. Role of antibiotic class in the rate of liberation of endotoxin during therapy for experimental gram-negative bacterial sepsis. J. Infect. Dis. 151:1012-1018. 10. Walterspiel, J.W., S.L. Kaplan, and E.O. Mason, Jr. 1986. Protective effect of subinhibitory polymyxin B alone and in combination with ampicillin for overwhelming Haemophilus influenzae type B infection in the infant rat: evidence for in vivo and in vitro release of free endotoxin after ampicillin treatment. Ped. Res. 20:237-241. 11. Ditter, B., R. Urbaschek, and B. Urbaschek. 1983. Ability of various adsorbents to bind endotoxin in vitro and to prevent orally induced endotoxemia in mice. Gastroenterology 84: 1547-1552. 12. Olcay, I., A. Kitahama, R.H. Miller, T. Drapanas, R.A. Trejo, and R. DiLuzio. 1974. Reticuloendothelial dysfunction and endotoxemia following portal vein occlusion. Surgery 75:64-70. 13. Grun, M., H. Liehr, and U. Rasenack. 1976. Significance of endotoxaemia in experimental "galactosamine- hepatitis" in the rat. Acta Hepato-Gastroenterol. 23:64-81. 14. Nolan, J.P., and A.I. Leibowitz. 1978. Endotoxin and the liver. III. Modification of acute carbon tetrachloride injury by polymyxin B- an antiendotoxin. Gastroenterology 75:445-449. 15. Broitman, S.A., L.S. Gottlieb, and N. Zamcheck. 1964. Influence of neomycin and ingested endotoxin in the pathogenesis of choline deficiency cirrhosis in the adult rat. J. EXp. Med. 119:633-641. 16. Cornell, R.P. 1985. Gut-derived endotoxin elicits hepatotrophic factor secretion for liver regenertion. Am. J. Physiol. 249:R551-R562. 180 17. Caty, M.G., K.S. Guice, K.T. Oldham, D.G. Remick, and S.L. Kunkel. 1990. Evidence for tumor necrosis factor- induced pulmonary microvascular injury after intestinal ischemia-reperfusion injury. Ann. Surg. 212:694-700. 18. Mathison, J.C., and R.J. Ulevitch. 1979. The clearance, tissue distribution, and cellular localization of intravenously injected lipopolysaccharide in rabbits. J. Immunol. 123:2133-2143. 19. DiLuzio, N.R., and C.G. Crafton. 1969. Influence of reticuloendothelial function on vascular clearance and tissue distribution of S. enteritidis endotoxin. Proc. Soc. Exp. Biol. Med. 132:686-690. 20. Praaning-van. Dalen, D.P., A, Brouwer, and. D.L. Knook. 1981. Clearance capacity of rat liver Kupffer, endothelial and parenchymal cells. Gastroenterology 81:1036- 1044. 21. Ruiter, D.J., J. Van Der Meulen, A. Brouwer, W.J.R. Hummel, B.J. Mauw, J.C.M. van Der Ploeg, and E. Wisse. 1981. Uptake by liver cells of endotoxin following its intravenous injection. Lab. Invest. 45:38-45. 22. Uragoh, H., K. Sueishi, T. Nakamura, and S. Iwanaga. 1988. A novel immunohistochemical method for in vivo detection of endotoxin using horseshoe crab Factor C. J. Histochem. Cytochem. 36:1275-1283. 23. Van Bossuyt, H., R.B. DeZanger, and E. Wisse. 1988. Cellular and subcellular distribution of injected lipopolysaccharide in rat liver and its inactivation by bile salts. J. Hepatol. 7:325-337. 24. Ulevitch, R.J., A.R. Johnston, and D.B. Weinstein. 1981. New function for high density lipoprotein (HDL): II. Isolation and characterization of a bacterial lipopolysaccharide-BBL complex formed in rabbit plasma. J. Clin. Invest. 67:827. 181 25. Ulevitch, RJJ., J.C. Mathison, R.R. Schumann, and P.S. Tobias. 1990. A new model of macrophage stimulation by bacterial lipopolysaccharide. J. Trauma 30:8189-3192. 26. Marra, M.N., C.G. Wilde, J.E. Griffith, J.L. Snable, and. R.W. Scott. 1990. Bactericidal/permeability- increasing protein has endotoxin neutralizing activity. J. Immunol. 144:662-666. 27 . Tobias, P. S . , J. C. Mathison, and R.J. Ulevitch. 1988. A family of lipopolysaccharide binding proteins involved. in. responses to :gram-negative sepsis. J} Biol. Chem. 263:13479-13481. 28. Elsbach, P., and J. Weiss. 1983. A reevaluation of the roles of the 02-dependent and 02-independent microbicidal systems of phagocytes. Rev. Infect. Dis. 5:843- 29. 001, C.E., J. Weiss, P. Elsbach, B. Frangione, and B. Mannion. 1987. A 25 kDa NHz-terminal fragment carries all the antibacterial activities of the human neutrophil 60-kDa bactericidal/permeability-increasing protein. J. Biol. Chem. 262:14891-14894. 30. Weiss, J., K. Muello, M. Victor, and P. Elsbach. 1984. The role of lipopolysaccharides in the action of the bactericidal/permeability increasing neutrophil protein on the bacterial envelope. J. Immunol. 132:3109-3115. 31. Dunn, D.L., P.A. Mach, R.M. Condie, and F.B. Cerra. 1984. Anticore endotoxin F(ab')2 equine immunoglobulin fragments protect against lethal effects of gram-negative bacterial sepsis. Surgery 96:440-446. 32. Dunn, D.L., and R.M. Ferguson. 1982. Immunotherapy of gram-negative bacterial sepsis: enhanced survival in a guinea pig model by use of rabbit antiserum to Becherichia coli J5. Surgery 92:212-219. 33. Priest, B.P., D.N. Brinson, D.A. Schroeder, and D.L. Dunn. 1989. Treatment of experimental gram-negative bacterial sepsis with murine monoclonal antibodies directed against lipopolysaccharide. Surgery 106:147-154. 182 34. Teng, N.N.H., H.S. Kaplan, J.M. Hebert, C. Moore, H. Douglas, A. Winderlich, and A.I. Braude. 1985. Protection against gram-negative bacteremia and endotoxemia with human monoclonal IgM antibodies. Proc. Natl. Acad. Sci. 82:1790- 1794. 35. Baumgartner, J-D., M.P. Glauser, J .A. McCutchan, E.J. Ziegler, G. van Melle, M.R. Klauber, M. Vogt, E. Muehlen, R. Luethy, R. Chiolero, and S. Geroulanos. 1985. Prevention of gram-negative shock and death in surgical patients by antibody to endotoxin core glycoprotein. Lancet 2:59-63. 36. Ziegler, E.J., C.J. Fisher, C.L. Sprung, R.C. Straube, J.C. Sadoff, G.E. Faulde, C.H. Wortel, M.P. Fink, P. Dellinger, N.N.H. Teng, I.E. Allen, H.J. Berger, G.L. Knatterud, A.F. LoBuglio, and C.R. Smith. 1991. Treatment of gram-negative bacteremia and septic shock with HA-lA human monoclonal antibody against endotoxin. N. Engl. J. Med. 324:429-436. 37. Ziegler, E.J., J.A. McCutchan, J. Fierer, M.P., Glauser, J.C. Sadoff, H. Douglas, and A.I. Baude. 1982. Treatment of gram-negative bacteremia and shock with human antiserum to mutant Escherichia coli. N. Engl. J. Med. 307:1226-1230. 38. Quagliarello, V.J., W.J. Long, and W.M. Scheld. 1986. Morphologic alterations of the blood-brain barrier with experimental meningitis in the rat. Temporal sequence and role of encapsulation. J. Clin. Invest. 77:1084-1095. 39. Mustafa, M.M., O. Ramilo, G.A. Syragiannopoulos, K.D. Olsen, G.H. McCracken Jr., and E.J. Hansen. 1989. Induction of meningeal inflammation by outer membrane vesicles of Haemophilus influenzae Type b. J. Infect. Dis. 159:917-922. 40. Wispelwey, B., A.J. Lesse, E.J. Hansen, and W.M. Scheld. 1988. Haemophilus influenzae lipopolysaccharide- induced blood-brain barrier permeability during experimental meningitis in the rat. J. Clin. Invest. 82:1339-1346. 183 41. Kopaniak, M.Mg, A.C. Issekutz, and. H.Ac Movat. 1980. Kinetics of acute inflammation induced by E. coli in rabbits. Am. J. Pathol. 98:485-498. 42. Issekutz, A.C., and S. Bhimji. 1982. Role for endotoxin in the leukocyte infiltration accompanying Escherichia coli inflammation. Infect. Immun. 36:558-566. 43. Issekutz, A.C. S. Bhimji, and R. Borolussi. 1982. Effect of immune serum or polymyxin B on Eocherichia coli- induced inflammation and vascular injury. Infect. Immun. 36:548-557. 44. Gartner, S.L., D.G. Sieckmann, Y.H. Kang, L.P. Watson, and L.D. Homer. 1988. Effects of lipopolysaccharide, lipid A, lipid X, and phorbol ester on cultured bovine endothelial cells. Lab. Invest. 59:181-191. 45. Harlan, J.M., L.A. Harker, M.A. Reidy, C.M. Gajdusek, S.M. Schwartz, and G.E. Striker. 1983. Lipopolysaccharide-mediated endothelial cell injury in vitro. Lab. Invest. 48:269-274. 46. Meyrick, B.O. 1986. Endotoxin-mediated pulmonary endothelial cell injury. Fed. Proc. 45:19-24. 47. Brigham, K.L., B. Meyrick, L.C. Berry, Jr., and J.E. Repine. 1987. Antioxidants protect cultured bovine lung endothelial cells from injury by endotoxin. J. Appl. Physiol. 63:840-850. 48. Matsuda, T., G.A. Eccleston, I. Rubinstein, S.I. Rennard, and W.L. Joyner. 1991. Antioxidants attenuate endotoxin-induced microvascular leakage of macromolecules in vivo. J. Appl. Physiol. 70:1483-1489. 49. Glode, L.M., A. Jacques, S.E. Mergenhagen, and D.L. Rosenstreich. 1977. Resistance of macrophages from C3H/HeJ mice to the in vitro cytotoxic effects of endotoxin. J. Immunol. 119:162-166. 184 50. Ryan, J.L., and K.P.W.J. McAdam. 1977. Genetic non- responsiveness of murine fibroblasts to bacterial endotoxin. Nature 269:153-155. 51. Mohri, M., D.R. Spriggs, and D. Kufe. 1990. Effects of lipopolysaccharide on phospholipase A2 activity and tumor necrosis factor expression in HL-60 cells. J} Immunol. 144:2678-2682. 52. Zuckerman, S.H., G.E. Evans, Y.M. Snyder, and W.D. Roeder. 1989. Endotoxin-macrophage interaction: post- translational regulation of tumor necrosis factor expression. J. Immunol. 143:1223-1227. 53. Bevilacqua, M.P., J.S. Pober, D.L. Mendrick, R.S. Cotran, and M.A. Gimbrone, Jr. 1987. Identification of an inducible endothelial-leukocyte adhesion molecule. Proc. Natl. Acad. Sci. 84:9238-9242. 54. Schleimer, R.P., and B.K. Rutledge. 1986. Cultured human vascular endothelial cells acquire adhesiveness for neutrophils after stimulation with interleukin 1, endotoxin, and tumor-promoting phorbol diesters. J. Immunol. 136:649- 654. 55. Doide, S. and R.M. Steinman. 1987. Induction of murine interleukin 1: Stimuli and responsive primary cells. Proc. Natl. Acad. Sci. 84:3802-3806. 56. Marletta, M.A., P.S. Yoon, R. Iyengar, C.D. Leaf, and. J.S. Wishnokn Macrophage oxidation. of L-arginine ‘to nitrite and nitrate: nitric oxide is an intermediate. Biochemistry 27:8706-8711. 57. McCall, T.B., M. Feelisch, R.M.J. Palmer, and S. Moncoda. 1991. Identification of N-iminoethyl-L-ornithine as an irreversible inhibitor of nitric oxide synthase in phagocytic cells. British J. Pharmacol. 102:234-238. 58. Libby, P., J.M. Ordovas, L.K. Birinyi, K.R. Auger, and C.A. Dinarello. 1986. Inducible interleukin-1 gene expression in human vascular smooth muscle cells. J. Clin. Invest. 78:1432-1438. 185 59. Rees, D.D., S. Cellek, R.M.J. Palmer, and S. Moncada. 1990. Dexamethasone prevents the induction by endotoxin of a nitric oxide synthase and the associated effects on vascular tone: an insight into endotoxin shock. Biochem. Biophys. Res. Commun. 173:541-547. 59a. Kovacs, E.J. 1991 Control of IL-1 and TNF-alpha production at the level of second messenger pathways. In Cytokines and Inflammation. E.S. Kimball, Ed. CRC Press, Inc. pp 89-107. 60. Prpic, V., J.E. Weiel, S.D. Somers, J.D. Guiseppi, S.L. Gonias, S.V. Pizzo, J.A. Hamilton, B. Herman, and D.O. Adams. 1987. Effects of bacterial lipopolysaccharide on the hydrolysis of phosphatidylinositol-4,5-bisphosphate in murine peritoneal macrophages. J. Immunol. 139:526-533. 61. Wightman, P.D., and C.R.H. Raetz. 1984. The activation of protein kinase C by biologically active lipid moieties of lipopolysaccharide. J. Biol. Chem. 259:10048- 10052. 62. Bosca, L., and M.J.M. Diaz-Guerra. 1988. Activation of protein kinase C from B lymphocytes by lipid A. Biochem. Biophys. Res. Commun. 152:149-154. 62a. Kovacs, E.J., D. Radzioch, H.A. Young, and L. Varesio. 1988. Differential inhibition of IL-1 and TNF-alpha mRNA expression by agents which block second messenger pathways in murine macrophages. J. Immunol. 141:3101-3105. 62b. Weiel, J.E., T.A. Hamilton, and D.O. Adams. 1986. LPS induces altered phosphate labeling of proteins in murine peritoneal macrophages. J. Immunol. 136:3012-3018. 63. Lei, M-G., and D.C. Morrison. 1988. Specific endotoxic lipopolysaccharide-binding proteins on murine splenocytes. I. Detection of lipopolysaccharide-binding sites on splenocytes and splenocyte subpopulations. J. Immunol. 141:996-1005. 186 64. Hampton, R.Y., D.T. Golenbock, and C.R.H. Raetz. 1988. Lipid A binding sites in membranes of macrophage tumor cells. J. Biol. Chem. 263:14802-14807. 65. Golenbock, D.T., R.Y. Hampton, C.R.H. Raetz, and S.D. Wright. 1990. Human phagocytes have multiple lipid A- binding sites. Infect Immun. 58:4069-4075. 66. Lei, M-G., and D.C. Morrison. 1988. Specific endotoxic lipopolysaccharide-binding proteins on murine splenocytes. II. Membrane localization and binding characteristics. J. Immunol. 141:1006-1011. 67. Bright, S.W., T-Y. Chen, L.M. Flebbe, M-G. Lei, and D.C. Morrison. 1990. Generation and characterization of hamster-mouse hybridomas secreting monoclonal antibodies with. specificity' for lipopolysaccharide: receptor. .1. Immunol. 145:1-7. 68. Chen, T.Y., S.W. Bright, J.L. Pace, S.W. Russell, and D.C. Morrison. 1990. Induction of macrophage-mediated tumor cytotoxicity by a hamster monoclonal antibody with specificity for lipopolysaccharide receptor. J. Immunol. 145:8-12. 69. Wright, S.D., and. M.T.C. Jong. 1986. Adhesion- promoting receptors on human macrophages recognize EScherichia coli by binding to lipopolysaccharide. J. Ekp. Med . 164 : 1876-1888 . 70. Wright, S.D., S.M. Levin, M.T. Jong, Z. Chad, and L.G. Kabbash. 1989. CR3 (CD11b/CD18) expresses one binding site for Arg-Gly-Asp containing peptides and a second site for bacterial lipopolysaccharide. J. Exp. Med. 169:175-183. 71. Wright, S.D., and P.A. Detmers. 1988. Adhesion- promoting receptors on phagocytes. J. Cell Sci Suppl. 9:99- 120. 72. Ramador, G., K-H. Meyer zum Buschenfelde, P.S. Tobias, J.C. Mathison, and R.J. Ulevitch. 1990. Biosynthesis of lipopolysaccharide-binding protein in rabbit hepatocytes. Pathobiol. 58:89-94. 187 73. Tobias, P.S., K. Soldau, and R.J. Ulevitch. 1986. Isolation of a lipolysaccharide-binding acute phase reactant from rabbit serum. J. Exp. Med. 164:777-793. 74. Tobias, P.S., K. Soldau, and R.J. Ulevitch. 1989. Identification of a lipid A binding site in the acute phase reactant lipopolysaccharide binding protein. J. Biol. Chem. 264:10867-10871. 75. Schumann, R.R., S.R. Leong, G.W. Flaggs, P.W. Gray, S.D. Wright, J.C. Mathison, P.S. Tobias, R.J. Ulevitch. 1990. Structure and function of lipopolysaccharide binding protein. Science 249:1429-1431. 76. Wright, S.D., R.A. Ramos, P.S. Tobias, R.J. Ulevitch, and J .C. Mathison. 1990. CD 14, a receptor for complexes of lipopolysaccharide (LPS) and LPS binding protein. Science 249:1431-1433. 77. Wright, S.D., P.S. Tobias, R.J. Ulevitch, and R.A. Ramos. 1989. lipopolysaccharide (LPS) binding protein opsonizes LPS-bearing particles for recognition by a novel receptor on macrophages. J. EXp. Med. 170:1231-1241. 78. Vosbeck, K., P. Tobias, H. Mueller, R.A. Allen, K- E. Arfors, R.J. Ulevitch, and L.A. Sklar. 1990. Priming of polymorphonuclear’ granulocytes by lipopolysaccharides and its complexes with lipopolysaccharide binding protein and high density lipoprotein. J. Leuk. Biol. 47:97-104. 79. Pagani, R., M.T. Portoles, and A.M. Municio. 1981. The binding of Escherichia coli endotoxin to isolated rat hepatocytes. FEBS Lett. 131:103-107. 80. Olson, N.C. 1987. Role of 5-hydroxytrytamine in endotoxin-induced respiratory failure in pigs. Am. Rev. Respir. Dis. 135:93-99. 81. Milligan, S.A., J.M. Hoeffel, I.M. Goldstein, and M.R. Flick. 1988. Effect of catalase on endotoxin-induced acute lung injury in unanesthetized sheep. Am. Rev. Respir. Dis. 137:420-428. 188 82. Olson, N.C., T.T. Brown Jr., and D.L. Anderson. 1985. Dexamethasone and indomethacin modify endotoxin- induced respiratory failure in pigs. J. .Appl. Physiol. 58:274-284. 83. Parker, J.L., and H.R. Adams. 1985. Develompment of myocardial dysfunction in endotoxin shock. Am. J. Physiol. 248:H818-H826. 84. Wakabayashi, I., K. Hatake, E. Kakishita, and K. Nagai. 1987. Diminution of contractile response of the aorta from endotoxin-injected rats. Eur. J. Pharmacol. 141:117- 122. 85. Mckenna, T.M. 1990. Prolonged exposure of rat aorta to low levels of endotoxin in vitro results in impaired contractility. Association with vascular cytokine release. J. Clin. Invest. 86:160-168. 86. Knowles, R.G., M. Salter, S.L. Brooks, and S. Moncada. 1990. Anti-inflammatory glucocorticoids inhibit the induction by endotoxin of nitric oxide synthase in the lung, liver and aorta of the rat. Biochem. Biophys. Res. Commun. 172:1042-1048. 87. Thiemermann, C., and J. vane. 1990. Inhibition of nitric oxide synthesis reduces the hypotension induced by bacterial lipopolysaccharide in the rat in vivo. Eur. J. Pharmacol. 182:591-595. 88. Doebber, T.W., M.S. Wu, J.C. Robbins, B.M. Choy, M.N. Chang, and T.Y. sheu. 1985. Platelet activating factor (PAF) involvement in endotoxin-induced hypotension in rats. Studies with PAF-receptor antagonist Kadsurenone. Biochem. Biophys. Res. Commun. 127:799-808. 89. Etemadi, A.R., G.E. Tempel, B.A. Farah, W.C. Wise, P.V. Halushka, and J.A. Cook. 1987. Beneficial effects of a leukotriene antagonist on endotoxin-induced acute hemodynamic alterations. Circ. Shock 22:55-63. 189 90. Matera, G., J.A. Cook, R.A. Hennigar, G.E. Tempel, W.C. Wise, T.D. Oglesby, and P.V.Halushka. 1988. Beneficial effects of a 5-lipoxygenase inhibitor in endotoxic shock in the rat. J. Pharmacol. EXp. Ther. 247:363-371. 91. Mikashima, H., and Y. Muramoto. 1990. Protective effect of Y-20811, a long lasting thromboxane synthetase inhibitor, on endotoxic shock in rabbits. Thromb. Res. 57:499-505. 92. Goligorsky, M.S. 1989. Role of endothelium in endotox in blockade of vo ltage-sens itive Ca2+ channels in smooth muscle cells. Am. J. Physiol. 257:C875-C881. 93. Mathison, J.C., E. Wolfson, and. R.J; Ulevitch. 1988. Participation of tumor necrosis factor in the mediation of gram negative bacterial lipopolysaccharide- induced injury in rabbits. J. Clin. Invest. 81:1925-1937. 94. Ohlsson, K., P. Bjork, M. Bergenfeldt, R. Hageman, and R.C. Thompson. 1990. Interleukin-1 receptor antagonist reduces mortality from endotoxin shock. Nature 348:550-552. 95. Beasley, D., R.A. Cohen, and N.G. Levinsky. 1989. Interleukin 1 inhibits contraction of vascular smooth muscle. J. Clin. Invest. 83:331-335. 96. Ferguson, J.L., J.J. Spitzer, and. H.I. Miller. 1978. Effects of endotoxin on regional blood flow in the unanesthetized guinea pig. J. Surg. Res. 25:236-243. 97. Breslow, M.J., C.F. Miller, S.D. Parker, A.T. Walman, and R.J. Traystman. 1987. Effect of vasopressors on organ blood flow during endotoxin shock in pigs. Am. J. Physiol. 252:H291-H300. 98. Colucci, M., G. Balconi, R. Lorenzet, A. Pietra, D. Locati, M.B. Donati, and N. Semeraro. 1983. Cultured human endothelial cells generate ‘tissue factor' in response 'to endotoxin. J. Clin. Invest. 71:1893-1896. 190 99. Morrison, D.C., and C.G. Cochrane. 1974. Direct evidence for Hageman factor (Factor XII) activation by bacterial lipopolysaccharides (endotoxins). J} EXpu Med. 140:797-811. 100. Warr, T.A., L.V.M. Rao, and S.I. Rapaport. 1990. Disseminated intravascular coagulation in rabbits induced by administration of endotoxin or tissue factor: effect of anti-tissue factor antibodies and measurement of plasma extrinsic pathway inhibitor activity. Blood 75:1481-1489. 101. Filkins, J.P. , and N.R. DiLuzio. 1968 . Heparin protection in endotoxin shock. Am . J. Physiol . 214 : 1074- 1077 . 102. Sleeman, H.K., and R.M. Hardaway. 1967. Evaluation of biochemical changes associated with experimental endotoxemia. I. Transaminase activity. Surgery 61:945-950. 103. Priano, L.L., R.D. Wilson, and D.L. Traber. 1974. Lack of significant protection afforded by heparin during endotoxic shock. Am. J. Physiol. 220:901-905. 104. Wallace, J.L., G. Steel, B.J.R. Whittle, V. Lagente, and. B. ‘Vargaftig. 1987. Evidence for jplatelet- activating factor as a mediator of endotoxin-induced gastrointestinal damage in the rat. Effects of three platelet-activating factor’ antagonist. Gastroenterology 93:765-773. 105. Dahlen, S-E., J. Bjork, P. Hedqvist, K-E. Arfors, S. Hammarstrom, J-Ac Lindgren, and. B. Samuelsson. 1981. Leukotrienes promote plasma leakage and leukocyte adhesion in postcapillary venules: in vivo effects with relevance to the. acute inflammatory’ response. Proc. .Natl. Acad. .Sci. 78:3887-3891. 106. Cook, J.A., E.J. Li, R.M. Spicer, W.C. Wise, and P.V. Halushka. 1990. Effect of leukotriene receptor antagonists on vascular permeability during endotoxic shock. Circ. Shock 32:209-218. 191 107. Whittle, B.J.Rc, N.K. iBoughton-Smith, I.R. Hutcheson, J.V. Esplugues, and J.L. Wallace. 1987. Increased intestinal formation of PAF in endotoxin-induced damage in the rat. Br. J. Pharmacol. 92:3-4. 108. Rosam, A-C, J.L. Wallace, and J.R. Whittle. 1986. Potent ulcerogenic actions of platelet-activating factor on the stomach. Nature 319:54-56. 109. Boughton-Smith, N.K., I. Hutcheson, and B.J.R. Whittle. 1989. Relationship between PAF-acether and thromboxane A2 biosythesis in endotoxin-induced intestinal damage in the rat. Prostaglandins 38:319-333. 110. Ogletree, M.L., C.J. Begley, G.A. King, and K.L. Brigham. 1986. Influence of steroidal and nonsteroidal anti- inflammatory agents on the accumulation of arachidonic acid metabolites in plasma and lung lymph after endotoxemia in awake sheep. Measurement of prostacycline and thromboxane metabolites and 12 HETE. Am. Rev. Respir. Dis. 133:55-61. 111. Hoffmann, H., M. Siebeck, M. Spannagl, M. Weis, R. Geiger, M. Jochum, and H. Fritz. 1990. Effect of recombinant hirudin, a specific inhibitor of thrombin, on endotoxin- induced intravascular coagulation and acute lung injury in pigs. Am. Rev. Respir. Dis. 142:782-788. 112. Heflin, A.C., and K.L. Brigham. 1981. Prevention by granulocyte depletion of increased vascular permeability of sheep lung following endotoxemia. J. Clin. Invest. 68:1253-1260. 113. Chang, S-W., C.O. Feddersen, P.M. Henson, and N.F. Voelkel. 1986. Platelet-activating' factor' :mediates hemodynamic changes and lung injury in endotoxin-treated rats. J. Clin. Invest. 79:1498-1509. 114. Chang, S-W., S. Fernyak, and N.F. Voelkel. 1990. Beneficial effect of a platelet-activating factor antagonist, WEB 2086, on endotoxin-induced lung injury. Am. J. Physiol. 258:H153-H158. 192 115. Christman, B.W., P.L. Lefferts, I.A. Blair, and J .R. Snapper. 1990. Effect of platelet-activating factor receptor antagonism on endotoxin-induced lung dysfunction in awake sheep. Am. Rev. Respir. Dis. 142:1272-1278. 116. Olson, N.C., P.B. Joyce, and L.N. Fleisher. 1990. Role of platelet-activating factor and eicosanoids during endotoxin-induced lung injury in pigs. Am. J. Physiol. 258:H1674-H1686. 117. Emerson, T.E.Jr., M.A. Fournel, T.B. Redens, and F.B. Taylor Jr. 1989. Efficacy of antithrombin III supplementation in animal models of fulminant Escherichia coli endotoxemia or bacteremia. Am. J. Med. 87:278-338. 118. McCuskey, R.S., R.A. McCuskey, R. Urbaschek, and B. Urbaschek. 1984. Species differences in Kupffer cells and endotoxin sensitivity. Infect. Immun. 45:278-280. 119. Kuida, H., R.P. Gilbert, L.B. Hinshaw, J.G. Brunson, and M.B. Visscher. 1961. Species differences in effect of gram-negative endotoxin on circulation. Am. J. Physiol. 200:1197-1202. 120. Jacobson, E.D., and W.E. Farrar Jr. 1963. Influence of route of administration on hemodynamic effects of endotoxin. Am. J. Physiol. 205:799-802. 121. Durham, S.K., A. Brouwer, R.J. Barelds, M.A. Horan, and D.L. Knook. 1990. Comparative endotoxin-induced hepatic injury in young and aged rats. J. Pathol. 162:341- 349. 122. Gates, G.A., K.S. Henley, H.M. Pollard, E. Schmidt, and F.W. Schmidt. 1961. The cell population of human liver. J. Lab. Clin. Med. 57:182. 123. Blouin, A., R.P. Bolender, and E.R. Weibel. 1977. Distribution of organelles and membranes between hepatocytes and non hepatocytes in the rat liver parenchyma. J. Cell. Biol. 72:441-455. 193 124. Greenway, C.V., and 2R.D. Stark. 1971. Hepatic vascular bed. Physiol. Rev. 51:23-65. 125. Balis, J.U., J.F. Paterson, S.A. Shelley, C.H. Larson, J. Fareed, and L.I. Gerber. 1979. Glucocorticoid and antibiotic effects on hepatic microcirculation and associated host responses in lethal gram-negative bacteremia. Lab. Invest. 40:55-65. 126. Levy, E., F.C. Path, and B.H. Ruebner. 1967. Hepatic changes produced by a single dose of endotoxin in the mouse. Light microscopy and histochemistry. Am. J} Pathol. 51:269-285. 127. Hirata, K., A. Kaneko, K. Ogawa, H. Hayasaka, and T. Onoe. 1980. Effect of endotoxin on rat liver. Analysis of acid phosphatase isozymes in the liver of normal and endotoxin-treated rats. Lab. Invest. 43:165-171. 128. Levy, E., F.C. Path, R.J. Slusser, and B.H. Ruebner. 1968. Hepatic changes produced by a single dose of endotoxin in the mouse. Electron microscopy. Am. J. Pathol. 52:477-502. 129. Shibayama, Y. 1987. Sinusoidal. circulatory disturbance by microthrombosis as a cause of endotoxin- induced hepatic injury. J. Pathol. 151:315-321. 130. Filkins, J.P., and B.J. Buchanan. 1977. In viva vs in vitro effects of endotoxin on glycogenolysis, gluconeogenesis, and glucose utilization. Proc. Soc. Exp. Biol. Med. 155:216-218. 131. Filkins, J.P., and R.P. Cornell. 1974. Depression of hepatic gluconeogenesis and the hypoglycemia of endotoxin shock. Am. J. Physiol. 227:778-781. 132. Victorov, A.V., B.M. Gladkaya, D.K. Novikov, V.A. Kosykh, and V.A. Yurkiv. 1989. Lipopolysaccharide toxin can directly stimulate the intracellular accumulation of lipids and their secretion into medium in the primary culture of rabbit hepatocytes. FEBS Lett 256:155-158. 194 133. Keller, G.A., M.A. West, L.A. Wilkes, F.B. Cerra, and R.L. Simmons. 1985. Modulation of hepatocyte protein synthesis by endotoxin-activated Kupffer cells. II. Mediation by soluble transferrable factors. Ann. Surg. 201:429-235. 134. Goradischer, R., J. Krasner, J.J. MCDevitt, J.P. Nolan, S.J. Yaffe. 1976. Hepatic microsomal drug metabolism after administration of endotoxin in rats. Biochem. Pharmacol. 25:351-353. 135. Williams, J.F., S. Lawitt, and A. Szentivanyi. 1980. Endotoxin depression of hepatic mixed-function oxidase system in C3H/HeJ and C3H/HeN mice. Immunapharmacal. 2:285- 291. 136. Yaffe, S.J., and B.R. Sonawane. 1978. Inhibition of rat hepatic ‘microsomal drug' metabolizing enzymes by endotoxin. Fed. Proc. 37:304. 137. Nolan, J.P., and C.J. O'Connell. 1965. Vascular response in the isolated rat liver. I. Endotoxin, direct effects. J. Exp. Med. 122:1063-1073. 138. Filkins, J.P. 1969. Hepatic vascular response to endotoxin. Proc. Soc. EXp. Biol. Med. 131:1235-1238. 139. Rangel, D.M. A. Dinbar, G.H. Stevens, R. Cooper, and E.W. Fankalsrud. 1970. The hepatic response to endotoxin shock; hemodynamic and enzymatic observations. J. Surg. Res. 10:181-188. 140. Hinshaw, L.B., D.A. Reins, and R.J. Hill. 1966. Response of isolated liver to endotoxin. Can. J. Physiol. Pharmacol. 44:529-541. 141. Utili, R., C.O. Abernathy, and H.J. Zimmerman. 1976. Chalestatic effects of Escherichia coli endotoxin on the isolated perfused rat liver. Gastraenteralagy 70:248- 253. 195 142. Utili, R., C.O. Abernathy, and H.J. Zimmerman. 1977. Studies on the effects of E. coli endotoxin an canalicular’ bile formation. in ‘the isolated. perfused. rat liver. J. Lab. Clin Med. 89:471-482. 143. Erlinger, S., and D. Dhumeaux. 1974. Mechanisms and control of secretion of bile water and electrolytes. Gastraenteralagy 66:281-304. 144. Utili, R., C.O. Abernathy, and H.J. Zimmerman. 1977. Inhibition of Na+/K+-adenasine triphosphatase by endotoxin: A possible mechanism for endotoxin-induced cholestasis. J. Infect. Dis. 136:583-587. 145. Movat, H.Z., M.I. Cybulsky, I.G. Colditz, M.K.W. Chan, and C.A. Dinarello. 1987. Acute inflammation in gram- negative infection: endotoxin, interleukin 1, tumor necrosis factor, and neutrophils. Fed Proc. 46:97-104. 146. Dale, D.C., D. Guerry IV, J.R. Wewerka, J.M. Bull, and M.J. Chusid. 1979. Chronic neutropenia. Medicine 58:128- 144. 147. Tauber, A.I. 1981. Current views of neutrophil dysfunction. an integrated clinical perspective. Am. J. Med. 70:1237-1246. 148. Malech, H.L., and J.I. Gallin. 1987. Current concepts: immunology. Neutrophils in human disease. N. Engl. J. Med. 317:687-694. 149. Issekutz, A.C. P. IMegyeri, and. T.B. Issekutz. 1987. Role of macrophage products in endotoxin-induced polymorphonuclear leukocyte accumulation during inflammation. Lab. Invest. 56:49-59. 150. Strieter, R.M., S.L. Kunkel, H.J. Showell, D.G. Remick, S.H. Phan, P.A. Ward, and R.M. Marks. 1989. Endothelial cell gene expression of a neutrophil chemotactic factor by TNF-alpha, LPS, and IL-1 beta. Science 243:1467- 1469. 196 151. Aida, Y., and M.J. Pabst. 1990. Priming of neutrophils by lipopolysaccharide for enhanced release of superoxide. Requirement for plasma but not for tumor necrosis factor-alpha. J. Immunol. 145:3017-3025. 152. Sandborg, R.R. and J.E. Smolen. 1988. Biology of disease. Early biochemical events in leukocyte activation. Lab. Invest. 59:300-320. 153. Fantone, J.C., and P.A. Ward. 1982. Role of oxygen-derived free radicals and metabolites in leukocyte- dependent inflammatory reactions. Am. J. Pathol. 107:397- 418. 154. Weiss, S.J. Tissue destruction by neutrophils. N. Engl. J. Med. 320:365-376. 155. Romson, J.L., B.G. Hook, S.L. Kunkel, G.D. Abrams, M.A. Schark, and. B.R. Lucchesi. 1983. Reduction. of the extent of ischemic myocardial injury by neutrophil depletion in the dog. Circulation 67:1016-1023. 156. Till, G.O., K.J. Johnson, R. Kunkel, and P.A. Ward. 1982. Intravascular activation of complement and acute lung injury. Dependency' on. neutrophils. and. toxic :oxygen metabolites. J. Clin. Invest. 69:1126-1135. 157. Wallace, J.L., C.M. IKeenan, and. D.N. Granger. 1990. Gastric ulceration induced by nonsteroidal anti- inflammatory drugs is a neutrophil-dependent process. Am. J. Physiol. 259:6462-6467. 158. Dahm, L.D., A.E. Schultze, and R.A. Roth. 1991. An antibody to neutrophils attenuates alpha- naphthylisothiocyanate-induced liver injury. J. Pharmacol. Exp. Ther. 256:412-420. 159. Vedder, N.B., E.W. Fouty, R.R. Winn, J.M. Harlan, and C.C. Rice. 1989. Role of neutrophils in generalized reperfusion injury associated with resuscitation from shock. Surgery 106:509-516. 197 160. Kopaniak, M.M. and H.Z. Movat. 1983. Kinetics of acute inflammation induced by Escherichia coli in rabbits. II. The effect of hyperimmunization, complement depletion, and depletion of leukocytes. Am. J. Pathol. 110:13-29. 161. Holman Jr., J.M., and T.M. Saba. 1988. Hepatocyte injury during post-operative sepsis: activated neutrophils as potential mediators. J. Leuk. Biol. 43:193-203. 162. Decker, K. 1990. Biologically active products of stimulated liver macrophages (Kupffer cells). Eur. J. Biochem. 192:245-261. 163. Adams, D.O., and T.A. Hamilton. 1984. The cell biology of macrophage activation. Ann. Rev. Immunol. 2:283- 318. 164. Sultzer, B.M. 1968. Genetic control of leukocyte response to endotoxin. Nature 219:1253-1254. 165. Rosenstreich, D.L. 1985. Genetic control of endotoxin response: C3H/HeJ mice. In Handbook of endotoxin vol 3: cellular biology of endotoxin. L.J. Berry, Ed. Elsevier Science Pub. pp82-122. 166. Wahl, L.M., D.L. Rosenstreich, L.M. Glade, A.L. Sandberg, and S.E. Mergenhagen. 1979. Defective prostaglandin synthesis by C3H/HeJ mouse macrophages stimulated with endotoxin preparations. Infect. Immun. 23:8- 13. 167. Beutler, B., N. Krochin, I.N. Milsark, C. Luedke, and A. Cerami. 1986. Control of cachectin (tumor necrosis factor) synthesis: mechanisms of endotoxin resistance. Science 232:977-980. 168. Stuehr, D.J., and M. A. Marletta. 1985. Mammalian nitrate biosynthesis: Mouse macrophages produce nitrite and nitrate in response to Escherichia coli lipopolysaccharide. Proc. Natl. Acad. Sci. 82:7738-7742. 198 169. Vagel, S.N., and D.L. Rosenstreich. 1979. Defective Fc receptor-mediated phagocytosis in C3H/HeJ macrophages . I . Correction by lymphokine induced stumulatian. J. Immunol. 123:2842-2850. 170. McCuskey, R.S., R. Urbaschek, P.A. McCuskey, N. Sacco, W.T. Stauber, C.A. Pinkstaff, and B. Urbaschek. 1984. Deficient Kupffer cell phagocytosis and lysosomal enzymes in the endotoxin-law-respansive C3H/HeJ mouse. J. Leuk. Biol. 36:591-600. 171. DiLuzio, N.R., and C.G. Crafton. 1970. A consideration of the role of the reticuloendothelial system (RES) in endotoxin shock. In Shack: biochemical, pharmacological and clinical aspects. A. Bertelli and N. Back, Eds, Plenum Press, New York, pp 7-57. 172. Crafton, C.G., and N.R. DiLuzio. 1969. Relationship of reticuloendothelial functional activity to endotoxin lethality. Am. J. Physiol. 217:736-742. 173. Vagel, S.N., R.N. Moore, J.D. Sipe, and. D.L. Rosenstreich. 1980. BCG-induced enhancement of endotoxin sensitivity in C3H/HeJ mice. I. In viva studies. J. Immunol. 124:2004-2009. 174. Arthur, M.J.P., P. Kowalski-Saunders and R. Wright. 1986. Carynebacterium ,parvum-elicited hepatic macrophages demonstrate enhanced respiratory burst activity compared. with resident. Kupffer cells in the rat. Gastraenteralagy 91:174-181. 175. Arthur, M.J.P., I.S. Bentley, .A.R, Tanner, P. Kowalski-Saunders, G.H. Millward-Sadler, and R. Wright. 1985. Oxygen-derived free radicals promote hepatic injury in the rat. Gastraenteralagy 89:1114-1122. 176. DiLuzio, N.R., R.A. Trejo, and C.C. Crafton. 1972. Influence of endotoxin administration on hepatic fucntion of rats with altered endotoxin sensitivity. J. Reticuloendothel. Sac. 11:637-653. 199 177. Billiar, T.R., R.D. Curran, M.A. West, K. Hofmann, and R.L. Simmons. 1989. Kupffer cell cytotoxicity to hepatocytes in culture requires L-arginine. Arch. Surg. 124:1416-1421. 178. Curran, R.D., T.R. Billiar, D.J. Stuehr, K. Hofmann, and RJL. Simmons. 1989. Hepatocytes produce nitrogen oxides from L-arginine in response to inflammatory products of Kupffer cells. J. Exp. Med. 170:1769-1774. 179. Curran, R.D., T.R. Billiar, D.J. Stuehr, J. B. Ochoa, B.G. Harbrecht, S.G. Flint, and RIL. Simmons. 1990. Multiple cytokines are required to induce hepatocyte nitric oxide production and inhibit total protein synthesis. Ann. Surg. 212:462-471. 180. Billiar, T.R., R.D. Curran, B.G. Harbrecht, D.J. Stuehr, A.J. Demetris, and R.L. Simmons. 1990. Modulation of nitrogen oxide synthesis in viva: NG-monomethyl-L-arginine inhibits endotoxin-induced. nitrite/nitrate biosynthesis while promoting hepatic damage. J. Leuk. Biol. 48:565-569. 181. Groeneveld, P.H.P., E. Claassen, C.F. Kuper, and N. Van Rooijen. 1988. The role of macrophages in LPS-induced lethality and tissue injury. Immunol. 63:521-527. 182. Andus, T., J. Bauer, and W. Gerok. 1991. Effects of cytokines on the liver. Hepatolagy 13:369-375. 183. Balkwill, F.R., and F. Burk. 1989. The cytokine network. Immunol. Today 10:299-304. 184. Thomas, D.J., M.J. Reasor, D. Wierda. 1989. Macrophage regulation of myelopoiesis is altered by exposure to the benzene metabolite hydroquinone. Toxicol. Appl. Pharmacol. 97:440-453. 185. O'Garra, A. 1989. Interleukins and. the immune system 1. Lancet 1:943-947. 186. O’Garra, A. 1989. Interleukins and. the immune system 2. Lancet 1:1003-1005. 200 187. Aiello, F.B., D.L. Longo, R. Overton, L. Takacs, and S.K. Durum. 1990. A. role for cytokines in antigen presentation: IL-1 and IL-4 induce accessory function of antigen presenting cells. J. Immunol. 144:2572-2581. 188. Akira, S., T. Hirano, T. Taga, and T. Kishimoto. 1990. Biology of multifunctional cytokines: IL 6 and related molecules (IL 1 and TNF). FASEB J. 4: 2860-2867. 189. Bevilacqua, M.P., J.S. Pober, G.R, Majeau, ‘W. Fiers, R.S. Cotran and MIA. Gimbrone Jr. 1986. Recombinant tumor necrosis factor induces procoagulant activity in cultured human vascular endothelium: Characterization and comparison with the actions of interleukin 1. Proc. Natl. Acad. Sci. 83:4533-4537. 190. Bevilacqua, M.P., J.S. Pober, M.E. Wheeler, R.S. Cotran, and M.A. Gimbrone Jr. 1985. Interleukin 1 acts on cultured human vascular endothelium to increase the adhesion of polymorphonuclear leukocytes, monocytes, and related leukocyte cell lines. J. Clin. Invest. 76:2003-2011. 191. Vadas, M.A., and J.R. Gamble. 1990. Regulation of the adhesion of neutrophils to endothelium. Biochem. Pharmacol. 40:1683-1687. 192. Salyer, J.L., J.F. Bohnsack, 'W.Ac Knape, A.O. Shigeoka, E.R. Ashwood, and H.R. Hill. 1990. Mechanisms of tumor necrosis factor-alpha alteration of PMN adhesion and migration. Am J. Pathol. 136:831-841. 193. Thornton, .A.J., R.M. Strieter, I. Lindleyy M. Baggiolini, and S.L. Kunkel. 1990. Cytokine-induced gene expression of a neutrophil chemotactic factor/IL-8 in human hepatocytes. J. Immunol. 144:2609-2613. 194. Tracey, K.J., B. Beutler, S.F. Lowry, J. Merryweather, S. Wolpe, I.W. Milsark, R.J. Hariri, T.J. Fahey, A. Zentella, J.D. Albert, G.T. Shires, A. Cerami. 1986. Shock and tissue injury induced by recombinant human cachectin. Science 234:470-474. 201 195. Remick, D.G., R.G. Kunkel, J.W; Larrick, S.L. Kunkel. 1987. Acute in vivo effects of human recombinant tumor necrosis factor. Lab. Invest. 56:583-590. 196. Grau, G.E., L.F. Fajardo, P-F. Piguet, B. Allet, P-H. Lambert, P. Vassalli. 1987. Tumor necrosis factor (cachectin) as an essential mediator in murine cerebral malaria. Science 237:1210-1212. 197. Colletti, L.M., D.G. Remick, G.D. Burtch, S.L. Kunkel, R.M. Strieter, and D.A. Campbell Jr. 1990. Role of tumor necrosis factor-alpha in the pathophysiologic alterations after hepatic ischemia/reperfusion injury in the rat. J. Clin. Invest. 85:1936-1943. 198. Chensue, S.W., P.D. Terebuh, D.G. Remick, W.E. Scales, and S.L. Kunkel. 1991. In vivo biologic and immunohistochemical analysis of interleukin-1 alpha, beta and tumor necrosis factor during experimental endotoxemia. Kinetics, Kupffer cell expression, and giucocorticoid effects. Am. J. Pathol. 138:395-402. 199. Beutler, B., I.Wt IMilsark, A.C. Cerami. 1985. Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin. Science 229:869-871. 200. Remick, D.G., R.M. Strieter, M.K. Eskandari, D.T. Nguyen, M.A. Genord, C.L. Raiford, and S.L. Kunkel. 1990. Role of tumor necrosis factor-alpha in lipopolysaccharide- induced pathologic alterations. Am. J. Pathol. 136:49-60. 201. Tiegs, G., M. Niehorster, and A. Wendel. 1990. Leukocyte alterations do not account for hepatitis induced by endotoxin or TNF-alpha in galactosamine-sensitized mice. Biochem. Pharmacol. 40:1317-1322. 202. Stephens, K.E., A. Ishizaka, Z. Wu, J.W. Larrick, and T.A. Raffin. 1988. Granulocyte depletion prevents tumor necrosis factor-mediated acute lung injury in guinea pigs. Am. Rev. Respir. Dis. 138:1300-1307. 202 202a. Starnes, H.F., M.K. Pearce, A. Tewari, J.H. Yim, J-C. Zou, and J.S. Abrams. 1990. Anti-IL-6 monoclonal antibodies protect against lethal Escherichia coli infection and lethal tumor necrosis factor-alpha challenge in mice. J. Immunol. 145:4185-4191. 202b. Movat, H.Z., C.E. Burrowes, M.I. Cybulsky, and C.A. Dinarello. 1987. Acute inflammation and a Shwartzman- like reaction induced by interleukin-1 and tumor necrosis factor. Synergistic action of the cytokines in the induction of inflammation and microvascular injury. Am. J. Pathol. 129:463-476. 202C. Waage, A., A. Halstensen, R. Shalaby, P. Brandtzaeg, P. Kierulf, and T. Espevik. 1989. Local production of tumor necrosis factor-alpha, interleukin 1, and interleukin 6 in meningococcal meningitis. Relation to the inflammatory response. J. Exp. Med. 170:1859-1867. 203. Margaretten, W., D.G. Mckay, and L.L. Phillips. 1967. The effect of heparin on endotoxin shock in the rat. Am. J. Pathol. 51:61-68. 204. Prager, R.L., E.L. Dunn, M.M. Kirsh, and J.A. Penner. 1979. Endotoxin-induced intravascular coagulation (DIC) and its therapy. Adv. Shack Res. 2:277-287. 205. Yurt, R.W: R.W. Leid, and. K.F. Austen, 1977. Native heparin from rat peritoneal mast cell. J. Biol. Chem. 252:518-521. 206. Rosenberg, R.D., and L.H. Lam. 1979. Correlation between structure and function of heparin. Proc. Natl. Acad. Sci. 76:1218-1222. 207. Rosenberg, R.D., and P.S. Damus. 1973. The purification and mechanism of action of human antithrombin- heparin cofactor. J. Biol. Chem. 248:6490-6505. 208. Gans, H. 1975. Mechanism of heparin protection in endotoxin shock. Surgery 77:602-606. 203 209. Ball, H.A., J.A. Cook, W.C. Wise and P.V. Halushka. 1986. Role of thromboxane, prostaglandins and leukotrienes in endotoxic and septic shock. Int. Care Med. 12:116-126. 210. O’Flaherty, J.T. 1982. Biology of disease. Lipid mediators of inflammation and allergy. Lab. Invest. 47:314- 329. 211. Tiegs, G., and A. Wendel. 1988. Leukotriene- mediated liver injury. Biochem. Pharmacol. 37:2569-2573. 212. Cook, J.A., W.C. Wise, and C.S. Callihan. 1979. Resistance of essential fatty acid-deficient rats to endotoxic shock. Ciro. Shock 6:333-342. 213. Cook, J.A., W.C. Wise, D.R. Knapp, and P.V. Halushka. 1981. Sensitization of essential fatty acid- deficient rats to endotoxin by arachidonate pretreatment role of thromboxane A2. Ciro. Shack 8:69-76. 214. Cook, J.A., W.C. Wise, and P.V. Halushka. 1980. Elevated thromboxane levels in the rat during endotoxic shock. Protective effects of imidazole, 13-azaprostanoic acid, or essential fatty acid defiency. J. Clin. Invest. 65:227-230. 215. Wise, W.C., J.A. Cook, P.V. Halushka, and DR. Knapp. 1980. Protective effects of thromboxane synthetase inhibitors in rats in endotoxic shock. Circ. Res. 46:854- 859. 216. Utili, R., C.O. Abernathy, H.J. Zimmerman. 1977. Minireview. Endotoxin effects on the liver. Life Sci. 20:553-568 . 217. Keller, G.A., M.A. West, F.B. Cerra, and R.L. Simmons. 1985. Modulation of hepatocyte protein synthesis by endotoxin activated Kupffer cells. Ann. Surg. 201:87-95. 218. Brigham, K.L., R.E. Bowers, and J. Haynes. 1979. Increased sheep lung vascular permeability caused by Escherichia coli endotoxin. Circ. Res. 45:292-297. 204 219. Snapper, J.R., G.R. Bernard, J.M. Hinson, A.A. Hutchison, J.E. Loyd, K.L. Ogletree, and K.L. Brigham. 1983. Endotoxin-induced leukopenia in sheep. Correlation with lung vascular permeability and hypoxemia but not with pulmonary hypertension. Am. Rev. Respir. Dis. 127:306. 220. Hewett, J.A., and R.A. Roth. 1988. Dieldrin activates. rat. neutrophils. in vitro. Toxicol. Appl. Pharmacol. 96:269-278. 221. Heide, K., and R.G. Schwick. 1978. Salt fractionation of immunoglobul ins . In Handbook of experimental immunology. Vol 1, Immunochemistry. D.M. Weir, ed., Blackwell Scientific Publications, Oxford, pp 7.1-7.11. 222. Reitman, S., and S. Frankel. 1957. A colorimetric method for the determination of serum glutamic oxaloacetic and glutamic pyruvic transaminases. Am. J. Clin. Pathol. 28:56. 223. Michaelson, M. 1969. Billirubin determination in serum. and urine. Scand, J} Clin. Lab. Invest. 13(supp1 56):1. 224. Ringler, D.B., and L. Dabich. 1979. Hematology and Clinical Biochemistry. In The laboratory rat. Vol 1: Biology and Disease, H.J. Baker, J.R. Lindsey, and S.H. Weisbroth, eds, Orlando, Academic Press, Inc. p 105. 225. Sandler, H., H. Hogstorp, C. Lundberg, and B. Gerdin. 1987. Antiserum-induced neutropenia in the rat: characterization of a rabbit anti-rat neutrophil serum. Br. J. Exp. Pathol. 68:71-80. 226. Smith, S.M., L. IHolm—Rutili, M.A. Perry; M.B. Grisham, K.E. Arfors, D.N. Granger, P.R. Kvietys, and J.M. Russell . 1987 . Role of neutrophils in hemorrhagic shock- induced gastric mucosal injury in the rat. Gastroenterology 93:466. 205 227. Bautista, A.F., K. Meszazos, J. Bojta, and J.J. Spitzer. 1990. Superoxide ‘anion. generation. in 'the liver during the early stage of endotoxemia in rats. J. Leuk. Biol. 48:123. 228. Arredondo, M.I. and R.P. Kampschmidt. 1963. Effect of endotoxins on phagocytic activity of the reticuloendothelial system of the rat. Poc. Soc. Exp. Biol. Med. 112:78. 229. Maier, R.V., and R.J. Ulevitch. 1981. The response of isolated rabbit hepatic macrophages (H-mo) to lipopolysaccharide (LPS). Circ. shock 8:165. 230. Pabst, M.J., and R.B. Johnston. 1980. Increased production of superoxide anion by macrophages exposed in vitro to muramyl dipeptide on lipopolysaccharide. J. Exp. Med. 151:101. 231. Pilaro, A.M., and D.L. Laskin. 1986. Accumulation of activated mononuclear phagocytes in the liver following lipopolysaccharide treatment in rats. J. Leuk. Biol. 40:29. 232. Russo, M. 1980. The role of macrophages in the chemotactic response of polymorphonuclear leukocytes to bacterial lipopolysaccharides. Proc. Soc. Exp. Biol. Med. 164:326. 233. Kakumu, 8., and C.M. Leevy. 1977. Lymphocyte cytotoxicity in alcoholic hepatitis. Gastroenterology 72:594. 234. Mizoguchi, Y. F. Ohnishi, T. Monna, S. Yamamoto, S. Otani, and S. Morisawa. 1981. Sudies on intrahepatic cholestasis in drug-induced allergic hepatitis: intrahepatic cholestasis induced in the rat by the culture supernatant of activated lymphocytes. Hepato-gastroenterology 28:147. 235. Ryan, G.E. 1967. The origin and sequence of the cells found in the acute inflammatory response. Aust. J. Exp. Biol. Med. Sci. 45:149. 206 236. Ryan, 6.3., and G. Majno. 1977. Acute inflammation. A Review. Am. J. Pathol. 86:185. 237. Ogawa, R. T. Morita, F. Kunimoto, and T. Fugita. 1982. Changes in hepatic lipoperoxide concentration in endotoxemic rats. Circ. Shock 9:369. 238. Sugino, K., K. Dohi, K. Yamada, T. Kawasaki. 1987. The role of lipid peroxidation in endotoxin-induced hepatic damage and the protective effect of antioxidants. Surgery 101:746. 239. Bautista, A.P. and J.J. Spitzer. 1990. Superoxide anion generation by in situ perfused rat liver: effect of in vivo endotoxin. Am. J. Physiol. 259:6907. 240. Becker, E.L., and H.J. Showell. 1974. The ability of chemotactic factors to induce lysosmal enzyme release. II. the mechanism of release. J. Immunol. 112:2055. 241. Hawkins, D. and S. Peeters. 1971. The response of polymorphonuclear leukocytes to immune complexes in vitro. Lab. Invest. 24:483. 242. Hensen, P. 1971. The immunological release: of constituents from neutrophil leukocytes. I. The role of antibody and complement on nonphagocytosable surfaces or phagocytosable particles. J. Immunol. 107:1535. 243. Cochrane, C.G., and B.S. Aikin. 1966. Polymorphonuclear leukocytes in immunological reactions. The destruction of vascular basement membrane in vivo and in vitro. J. Exp.Med. 124:733. 244. Hawkins, D., and C.G. Cochrane. 1968. Glomerular basement membrane damage in immunologic glomerulonephritis. Immunol. 14:665. 245. Wasi, 8., R.K. Murray, D.R.L. Macmorine, and H.Z. Movat. 1966. The role of PMN-leukocyte lysosomes in tissue injury, inflammation and hypersensitivity. II. Studies on the proteolytic activity of PMN-leukocyte lysosomes of the rabbit. Br. J. Exp. Pathol. 47:411. 207 246. Mavier, P., A. Preaux, B. Guigui, M. Lescs, E. Zafrani, D. Dhumeaux. 1988. In vitro toxicity of polymorphonuclear neutrophils to rat hepatocytes: evidence for a proteinase-mediated mechanism. Hepatol. 8:254. 247. Gamble, J.R., J.M. Harlan, S.J. Klebanoff, and M.A. Vadas. 1985. Stimulation of the adherence of neutrophils to umbilical vein endothelium by human recombinant tumor necrosis factor. Proc. Natl. Acad. Sci. USA 82:8667-8671. 248. Movat, H.Z. 1987. Tumor necrosis factor and interleukin-1: role in acute inflammation and microvascular injury. J. Lab. Clin. Med. 110:668-681. 249. Pohlman, T.H., K.A. Stannes, P.G. Beatty, H.D. Ochs, and J.M. Harlan. 1986. An endothelial cell surface factor(s) induced in vitro by lipopolysaccharide, interleukin 1 and tumor necrosis factor-alpha increase neutrophil adherence by a CDw18-dependent mechanism. J. Immunol. 136:4548-4553. 250. Baumann, H., V. Onorato, J. Gauldie, and G.P. Jahreis. 1987. Distinct sets of acute phase plasma proteins are stimulated by separate human hepatocyte-stimulating factors and monokines in rat hepatoma cells. J. Biol. Chem. 262:9756-9768. 251. Noel, P., 8. Nelson, R. Bokulic, G. Bagby, H. Lippton, G. Lipscomb, and W. Summer. 1990. Pentoxifylline inhibits lipopolysaccharide-induced serum tumor necrosis factor and mortality. Life Sci. 47:1023-1029. 252. Tiegs, G., M. Wolter, and A. wendel. 1989. Tumor necrosis factor is a terminal mediator in galactosamine]endotoxin-induced hepatitis in mice. Biochem. Pharmacol. 38:627-631. 253. Welsh, C.H., D. Lien, G.S. Worthen, and J.V. Weil. 1988. Pentoxifylline decreases endotoxin-induced pulmonary neutrophil sequestration and extravascular protein accumulation in the dog. Am Rev. Respir. Dis. 138:1106-1114. 208 254. Hewett, J.A., A.E. Schultze, S. Commen, and R.A. Roth. 1992. Neutrophil depletion protects against liver injury from bacterial endotoxin. Lab. Invest. in press. 255. Hartung, T., and A. Wendel. 1991. Endotoxin- inducible cytotoxicity in liver cell cultures-1. Biochem. Pharmacol. 42:1129-1135. 256. Espevik, T., and J. Nissen-Meyer. 1986. A highly sensitive cell line, WEHI 164 clone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes. J. Immunol. Methods 95:99-105. 257. Morris, P., A. Wheeler, and K. Brigham. 1991. Heparin reduces human recombinant tumor necrosis factor (TNF-alpha)-induced L929 cell cytotoxicity. FASEB J. 5:A1621 (Abstr.). 258. Waage, A., A. Halstensen, R. Shalaby, P. Brandtzaeg, P. Kierulf, and T. Espevik. 1989. Local production of tumor necrosis factor-alpha, interleukin 1, and interleukin 6 in meningococcal meningitis. Relation to the inflammatory response. J. EXp. Med. 170:1859-1867. 259. Ward, A., and S.P. Clissold. 1987. Pentoxifylline. A review of its pharmacodynamic and pharmacokinetic properties and its therapeutic efficacy. Drugs 34:50-97. 259a. Lorence, R.M, C.K. Edwards, R.J. Walter, K.W. Kelley, and J.A. Greager. 1990. In vivo effects of recombinant interferon-gamma: augmentation of endotoxin- induced necrosis of tumors and priming of macrophages for tumor necosis factor-alpha production. cancer Lett 53::223- 229. 259b. Moore, J.N. J.A. Cook, D.D. Morris, P.V. Halushka, and W.C. Wise. 1990. Endotoxin-induced procoagulant activity, eicosanoid synthesis, and tumor necrosis factor production by rat peritoneal macrophages: effect of endotoxin tolerance and glucan. Circ. Shock 31:281-295. 209 260. Klebanoff, S.J., M.A. Vadas, J.M. Harlan, L.H. Sparks, J.R. Gamble, J.M. Agosti, and A.M. Waltersdorph. 1986. Stimulation of neutrophils by tumor necrosis factor. J. Immunol. 136:4220-4225. 261. Berkow, R.L., D. Wang, J.W. Larrick, R.W. Dodson, and T.H. Howard. 1987. Enhancement of neutrophil superoxide production by preincubation with recombinant human tumor necrosis factor. J. Immunol. 139:3783-3791. 262. Yoshikawa, T., Y. Furukawa, M. Murakami, S. Takemura, and M. Kondo. 1981. Experimental model of disseminated intravascular coagulation induced by sustained infusion of endotoxin. Res. Exp. Med. 179:223-228. 263. Lipinski, B., V. Gurewich, and E. Hyde. 1976. Thrombin elaboration in endotoxin-induced intravascular fibrin deposition. A leukocyte dependent process distinct frmm systemic hypercoagulability. Thromb. Haemost. 36:615- 622. 264. Nowak, G., F. Markwardt. 1980. Influence of hirudin on endotoxin-induced disseminated intravascular coagulation (DIC) in weaned pigs. Exp. Pathol. 18:438-443. 265. DePalma, R.G., J. Coil, J.H. Davis, W.D. Holden. 1967. Cellular and ultrastructural changes in endotoxemia: A light and electron microscopic study. Surgery 62:505-515. 266. Fenton, J.W. 1981. Thrombin specificity. Ann. N.Y. Acad. Sci. 370:468-495. 267. Barlow, G.H., W.H. Holleman, and L. Lorand. 1970. The action of arvin on fibrin stabilizing factor (Factor XIII). Res. Commun. Chem. Pathol. Pharmacol. 1:39-42. 268. Pizzo, S.V., M.L. Schwartz, R.L. Hill, and P.A. McKee. 1972. Mechanism of ancrod anticoagulation. A direct proteolytic effect on fibrin. J. Clin. Invest. 51:2841-2850. 269. Johnson, A., M.V. Tahamont, and A.B. Malik. 1983. Thrombin-induced lung vascular injury. Roles of fibrinogen and fibrinolysis. Am. Rev. Respir. Dis. 128:38-44. 210 270. Stone, S.R., and J. Hofsteenge. 1986. Kinetics of the inhibition of thrombin by hirudin. Biochemistry 25:4622- 4628. 271. Bizios, R., L. iLai, J.W. Fenton II, and. A.B. Malik. 1986. Thrombin-induced Chemotaxis and aggregation of neutrophils. J. Cell Physiol. 128:485-490. 272. Yoshikawa, T., M. Murakami, N. Yoshida, 0. Seto, and M. Kondo. 1983. Effects of superoxide dismutase and catalase on disseminated intravascular coagulation in rats. Thromb. Haemost. 50:869-872. 273. Fasco, M.J. and L.M. Principe. 1982. R- and S- Warfarin inhibition of vitamin K and Vitamin K 2,3-epoxide reductase activities in the rat. J. Biol. Chem. 257:4894- 4901. 274. Van Pelt-Verkuil, E., P. van de Ree, and J.J. Emeis. 1989. Defibrinogenation by arvin reduces air-drying- induced arteriosclerosis in rat carotid artery. Thromb. Haemostas. 61:246-249. 275. Chensue, S.W., D.G. Remick, C. Shmyr-Forsch, T.F. Beals, and S.L. Kunkel. 1988. Immunohistochemical demonstration of cytoplasmic and membrane-associated tumor necrosis factor in murine macrophages. Am. J. Pathol. 133:564-572. 276. Clauss, A. 1957. Gerinnungsphysiologische schnell- methode zur bestimmung des fibrinogens. Acta. Haemat. 17:237-246. 277. Steel, R.G.D., and J.H. Torrie. 1980. In Principles and procedures of statistics. A Biomedical approach. 2nd edition. McGraw-Hill, Inc. New York. 633 pp. 278. Bell, W.R., G. Bolton, and W.R. Pitney. 1968. The effect of arvin. on ‘blood. coagulation factors. Brit. J. Haemat. 15:589-602. 211 279. Feinstein, M.B. E.L. Becker, C. Fraser. 1977. Thrombin, collagen and A23187 stimulated endogenous platelet arachidonate metabolism: differential inhibition by PGEl, local anesthetics and a serine-protease inhibitor. Prostaglandins 14:1075-1093. 280. Grodzinska, L., and E. Marcinkiewicz. 1979. The generation of TXA2 in human platelet rich plasma and its inhibition by nictindole and prostacyclin. Pharmacol. Res. Commun. 11:133-146. 281. Sinha, A.K., J. Willis, and R.W. Colman. 1983. Inhibition of thromboxane A2 synthesis in human platelets by coagulation factor Xa. Proc. Natl. Acad. Sci. 80:6086-6090. 282. Kito, M., H. Narita, M. Ishinaga, H.J. Park, and H. Takamura. 1985. Phosphoinositide breakdown as an indirect link between stimulation and aggregation of rat platelets by thrombin and collogen. J. Biochem. (Tokyo) 97:765-772. 283. Weyer, B., T.E. Petersen, and 0. Sonne. 1988. Characterization of the binding of bovine thrombin to isolated rat hepatocytes. Thromb. Haemost. 60:419-427. 284. Strieter, R.M., D.G. Remick, P.A. Ward, R.N. Spengler, J.P. Lynch III, J. Larrick, and S.L. Kunkel. 1988. Cellular and molecular regulation of tumor necrosis factor- alpha production by pentoxifylline. Biochem. Biophys. Res. Commmun. 155:1230-1236. "IT'ElliljlilfllflflflllfiliITS fl----‘ _-_-_-_-_