.4375 V ii! L. .5 2.... y . .73 .2: _...?>: )1. V 3. Lia , . l 2...... v) .1! h 2. 1 i.» sham a. . . , 3 36%.. .- >A : my _. , e. ,.m..mmfiwn.w%“ 7% 1L Pa. mm 9. i! ,3vfi .i v) I .f-v it 1. .1: _ :3 Luau; .5 in}... , 19-1 ;:..‘¥Lv: ‘ .~4 “WE: . v3.5.3? \1 la)? a p f "a vywmfinmhfifi. .1... hymns ‘ § L. . ‘2: £11 a til 1...- iv. ‘13; fl .“ B I .u 2 3.13.}. \xii‘. $..\w|.: gig P 7'! . 2:83;: 23‘58 :‘ :m g’ 33%;“ :2; a“ ‘ l i I $‘ ' ,sta 1‘: t 3 trig“. u ; . i .11 A nil 1.1:»: .- 14 -: u ‘ 3h . ! « i=3. rflflshhknnfluahuha ‘ L‘s..." . 01 . :1 a... 3. ilbixail 32:2: :51... v 5:... i . {if Inca)»- 9. 9.. n 04.3.]... 3. x .3. infill. “it: 5. n r!( Inuit.“ 911.4 997 X. 4%“. f ‘ “fez-”1:3” r'fififl' “'9‘“? 1:. 1: Lu ‘ “W“ 1}”, 5 on 31),... 31293 01415 3393 This is to certify that the thesis entitled CYTOKINE INVOLVEMENT IN PULMONARY OXYGEN TOXICITY presented by Hugh Jeffrey Lindsey has been accepted towards fulfillment of the requirements for Master degree in Science Major pr% Date Juiy 14, 1994 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State University PLACE Ii RETURN BOXtonmovombcinckoutm mm TOAVOIDFINESMunonorbdonddodm. DATE DUE DATE DUE DATE DUE ._____——~___ MSUI AnM‘inndivo O i a ActioNEquui maturity mason ‘ —— CYTOKINE INVOLVEMENT IN PULMONARY OXYGEN TOXICITY BY Hugh Jeffrey Lindsey A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Surgery 1994 ABSTRACT CYTOKINE INVOLVEMENT IN PULMONARY OXYGEN TOXICITY BY Hugh Jeffrey Lindsey Inspiration of high oxygen concentrations can result in severe pulmonary injury. The cytokines - tumor necrosis factor (TNF), interleukin-1 (IL-1), and.interleukin-6 (IL-6) - have been implicated in the pathophysiology of various pulmonary inflammatory processes. However, direct evidence of their involvement in oxygen toxicity has been lacking. Consequently, murine alveolar macrophages were exposed to high concentrations of oxygen both in vitrg and in vivg. Increased productive capacities for IL-1 and TNF were demonstrated after exposure to hyperoxia for 1 hour in vitro and 2 hours in yivg, respectively. Further, rats exposed to hyperoxia for 52 hours demonstrated increased concentrations of IL-6 in the bronchoalveolar lavage supernatant. Pretreatment with pentoxifylline (PTX) followed by immediate prolonged exposure of the rats to hyperoxia resulted in marked attenuation of oxygen-induced lung injury as determined by indices of pulmonry injury and decreased pulmonary supernatant and systemic serum levels of IL-6. Consequently, TNF, IL-1 and IL-6 appear to be involved in the development of pulmonary oxygen toxicity. Further, the oxygen-induced lung injury which develops as a result of prolonged exposure may be attenuated by pretreatment with PTX. Copyright by Hugh J. Lindsey, MD 1994 This thesis is dedicated to my wife and best friend, Barb, whose love, support, and encouragement were always present during the course of this project and to my parents, Hugh and Beverly, who instilled in me the importance of education and the gift of perseverance. ACKNOWLEDGEMENTS I am greatly indebted to Dr. Irshad Chaudry for allowing me to become part of his laboratory team and learning, not only from him, but also from the fine group he has assembled. I would also like to extend my thanks to Drs. John Kisala, Al Ayala, James Harkema, Keith Apelgren, and Robert Roth for serving on my graduate committee and offering valuable guidance during the course of this work. In particular, I would like to express my sincere gratitude Dr. Kisala for providing the original stimulus for this project and serving as the mentor of this work. Further, I would also like to extend a special thank-you to Dr. Ayala who not only freely gave numerous helpful suggestions but was always willing to listen to other thought processes. Finally, I would like to express my gratitude to Drs. Ping Wang and P.J. O’Neill for their helpful suggestions and camaraderie during the course of this project. ii II. III. IV. TABLE OF CONTENTS List of Tables . . . . . List of Figures . . . . Key to Abbreviations . . Introduction . . . . . . A. Clinical Significance B. The Pathology of Pulmonary Oxygen 1. Initiation Stage . 2. Inflammatory Stage 3. Destructive Stage 4. Proliferative Stage 5. Fibrotic Stage . . C. Mechanism of Pulmonary Oxygen D. Cytokines . . . . . . 1. Tumor Necrosis Factor 2. Interleukin-1 . . 3. Interleukin-6 . . Toxicity Toxicity . . E. Protection Against Pulmonary Oxygen Toxicity F 0 Purpose 0 O O O O O 0 Materials and Methods . A. Animal Model . . . . B. General . . . . . . . iii Page Number vi vii 8-9 9-11 11-17 12-15 15-16 16-17 17-19 19-20 21-40 21 22 VI. H. In_yitzg Exposure of Alveolar Macrophages to Hyperoxia and Normoxia . . . . . . . . In 2120 Exposure of Alveolar Macrophages to Hyperoxia and Normoxia . . . . . . . . Determination of Cell-associated TNF . . CYtOkine Assays O O O O O O O O I O O O O 1. Purpose and Cell Line Maintenance of Cytokine Assays . . . . . . . . . . . 2 o TNF Assay o o o o o o o o o o o o o o 3 o III-1 Assay o o o o o o o o o o o o o o 4. 11-6 Assay . . . . . . . . . . . . . . Determination of the Effects of Pretreatment with Pentoxifylline on Prolonged Hyperoxia 1. The Effects of Pentoxifylline on Mean Arterial Pressure . . . . . . . . . . 2. The Effects of Hyperoxia on Rats Pretreated with Pentoxifylline . . . . 3. Assays for Indicators of Pulmonary Injury Statistical Analyses . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . A. Alveolar Macrophage Cytokine Productive Capacities After Exposure to Hyperoxia or Normoxia In yitrg . . . . . . . . . . Exposure of Animals to Hyperoxia and Normoxia . . . . . . . . . . . . . . 1. Quantity of Alveolar Macrophages Obtained by Bronchoalveolar Lavage . . . . . . 2. Arterial Blood Gas Values 1 Hour After Exposure to Normoxia or Hyperoxia . . 3. Alveolar Macrophage Cytokine Productive Capacities After Exposure to Normoxia or Hyperoxia In Viyo . . . . . . . . . 4. Serum Cytokine Levels After Exposure to Normoxia or Hyperoxia In V119 . . . . iv 22-24 24-28 28-29 30-34 30 31-32 32-33 33-34 34-38 34 34-37 39-40 41-67 41-45 46-52 46 47 48-51 52 VII. VIII. IX. C. Results of Cell-associated TNF Bioassay After Exposure to Normoxia or Hyperoxia In VIyQ D. The Effects of Hyperoxia on Rats After . Pretreatment with PTX . . . . . . . . . . 1. The Effects of Pentoxifylline on Mean Arterial Pressure . . . . . . . . . . 2. The Results of Prolonged Hyperoxia After Pretreatment with Pentoxifylline . . . DiscuSSion O I O O O O O O O I O O O O O O O A. B. C. D. General Discussion of Oxygen-induced Lung Injury 0 C O O C O O O O O O C O O O C . Alveolar Macrophage Cytokine Productive Capacities After Exposure to Hyperoxia . Determination of Cell-associated TNF After Exposure to Either Hyperoxia or Normoxia Effects of Exposure to Lethal and Without Pretreatment with Hyperoxia mx 0 O O 0 Summary and Conclusions . . . . . . . . . . A. B. S umma ry O I O O O O O O O O O O O O O O 0 Conclusions . . . . . . . . . . . . . . . Appendices O O O O O I O O O O O O O O O O 0 Appendix A - Addendum . . . . . . . . . . Appendix B - Dulbecco’s Modification of Eagle’s Medium (DMEM) . . . . . . . . . . Appendix C - RPMI o o o o o o o o o o o 0 Appendix D - Publications, Presentations, and Awa rd 8 O O O O O O O O O O O O O O 0 Bibliography . . . . . . . . . . . . . . . . With 53 54-67 54-56 57-67 68-85 68-70 71-73 74 75-85 86-88 86-88 88 89-96 89 90-91 92-93 94-96 9Tfl22 Table Table Table Table Table LIST OF TABLES Viable Alveolar Macrophages . . . . . . Arterial Blood Gas Values . . . . . . . Serum Cytokine Levels After 1, 2, and 4 Hour Exposures . . . . . . . . . . . . Cell-associated TNF Values . . . . . . Parameters Determined Following 52 Hour Exposures O O O O O O O O O O O O O O 0 vi Page 46 47 52 53 62 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 10 11 12 13 14 15 LIST OF BIGURES IL-l Productive Capacities 60 Minutes In Vit; Q C I O O O O O O O O O O O O O 0 IL-1 Productive Capacities 30 Minutes m o o o o o o o o o o o o o o o o TNF Productive Capacities - In VItrg . . IL—6 Productive Capacities - In_yIt;Q . TNF Productive Capacities - In yivg . . IL-l Productive Capacities - In VIVQ . . Il-6 Productive Capacities - In yIvg . . Time Course of Changes in MAP Following Injection with PTX . . . . . . . . . . . Supernatant LDH Concentrations . . . . . Supernatant Protein Concentrations . . . Supernatant IL—6 Concentrations . . . . Serum IL-6 Concentrations . . . . . . . Serum IL-1 Concentrations . . . . . . . Serum LDH Concentrations . . . . . . . . Serum Protein Concentrations . . . . . . vii Page 42 43 44 45 49 50 51 56 59 60 61 64 65 66 67 ABG ACAS ARDS ATA BW Con A DAD DMEM D-PBS ELAM-l FBS F102 1502 ICAM-1,2 ICU IL-1 IL-6 IM IP KEY TO ABBREVIATIONS arterial blood gas anchored cell analysis station adult respiratory distress syndrome atmosphere bronchoalveolar lavage body weight concanavalin A diffuse alveolar damage Dulbecco’s modified eagle’s medium Dulbecco's phosphate buffered saline endothelial-leukocyte adhesion molecule-1 fetal bovine serum forced inspiratory concentration of oxygen hyperoxia hydrogen peroxide Na-hypothanxine-thymidine intercellular adherence molecule-1,2 intensive care unit interleukin-1 interleukin-6 intramuscular intraperitoneal viii IV kd RBC SC SDS SEM SOD VCAM-l intravenous kilodalton lactate dehydrogenase liters per minute lipopolysaccharide leukotriene B, mean arterial pressure 3-(4,S-dimethylthiazol-z-yl)-2,5- diphenyltetrazolium bromide; thiazol blue normoxia normal saline oxygen superoxide anion hydroxyl radical prostaglandin E2 pentoxifylline red blood cell subcutaneous sodium dodecylsulfate standard error of the mean superoxide dismutase tumor necrosis factor vascular endothelial adhesion molecule-1 ix INTRODUCTION "Though pure [oxygen] might be very useful as a medicine, . . . as a candle burns out much faster in [oxygen] . . . so we might ... 00 s ." - J. Priestly, 1775'. CLINICAL SIGNIFICANCE Inspiration of high oxygen concentrations can result in severe pulmonary damage. Today, this usually occurs when the individual is mechanically ventilated as in the intensive care unit (ICU). Barber g;_nII demonstrated declining pulmonary function in irreversibly brain-damaged, mechanically ventilated patients exposed to a forced inspiratory concentration of oxygen of 1.0 (FiO2 1.0) for 40 hoursz. The deterioration of pulmonary function.was evident by decreasing arterial oxygen tension, increased ratio of dead space to tidal volume, and increased intrapulmonary shunt. Chest roentgenograms revealing diffuse multilobar infiltrates and autopsy specimens showing grossly heavier lungs further supported the above evidence for oxygen toxicity. 1 2 Further work has demonstrated an early breakdown of the lung’s defense system. Mucociliary function becomes depressed which may lead to increased adherence of gram-negative organisms to the lower respiratory tract epithelium}. Neutrophil chemotaxins are released by alveolar macrophages, which ultimately results in pulmonary damage by the neutrophils‘. Also, migration of the alveolar macrophages is inhibited, which may impair movement into and within the bronchoalveolar spacess. Over the years, the time course of pulmonary oxygen toxicity in otherwise healthy humans after exposure to 100% oxygen has been delineated“: 6 hrs decreased tracheal mucus velocity 14 hrs - symptoms of tracheobronchitis 24-48 hrs - physiologic changes such as decreased forced vital capacity 30 hrs - gas exchange abnormalities 72-96 hrs - morphologic abnormalities >96 hrs - pulmonary fibrosis. Moreover, other factors can lower the toxic threshold of oxygen. Consequently, lesser concentrations can lead to pulmonary injury. For example, the duration of respiratory support rather than the duration of illness correlated better with the degree of pulmonary fibrosis upon a review of the pathologic specimens obtained from patients receiving 3 conventional therapy or extracorporeal membrane oxygenation’. Further, histology demonstrated that the pattern of injury was quite similar to that seen with oxygen toxicity despite attempting to minimize inspired oxygen concentrations. Administration of certain pharmacologic agents, particularly bleomycin, in combination with oxygen may result in pulmonary toxicity. Bleomycin is a chemotherapeutic agent used ‘to ‘treat several neoplasms including' squamous cell cancer, lymphoma, and testicular cell cancer”. Its mode of action is through generation of free radicals that have a deleterious effect on DNA“. Generation of these free radicals is thought to induce pulmonary toxicity6'3"0. Bleomycin, alone and coupled with various inspired concentrations of oxygen.ranging from 33%-70%, has been reported to cause severe pulmonary injury in patientsmfih. Animal studies have confirmed that bleomycin exacerbates pulmonary oxygen toxicity secondary to hyperoxiawnz. Animal research has also demonstrated that concurrent acute pulmonary inflammation lowers the toxic threshold of oxygen and exacerbates pulmonary inflammation. Haschek and Witschi have shown this in'miceux Immediately after inducing acute alveolitis by butylated hydroxytoluene, the mice were exposed to 70% oxygen for 24 hours. Extensive interstitial fibrosis developed in two weeks. Oxygen or butylated hydroxytoluene administered alone resulted in no or minimal fibrosis, respectively. Moreover, after inducing alveolitis as stated above, administration of 50% oxygen for six days 4 resulted in increased fibrosis. Apparently, oxygen and butylated hydroxytoluene act synergistically. Finally, the morphologic changes of pulmonary oxygen toxicity, termed. diffuse alveolar' damage (DAD), and. the mechanism of oxygen toxicity, believed to be the overwhelming production of reactive oxygen species, are present in other conditions. DAD has been described following sepsis, hemorrhagic shock, severe trauma, complicated intraabdominal surgery, ARDS (adult respiratory distress syndrome) and other insults1 ' u . The production of oxygen-derived free radicals has been implicated in the development of the acute lung injury secondary to ischemia-reperfusion, thermal injury, pancreatitis, and sepsis or TNF infusionfiqm. Further, ARDS may be due to free radical production”. Therefore, despite the fact that pure pulmonary oxygen toxicity is a relatively rare event in the clinical setting, delivering any increased concentration of oxygen may be deleterious to the lungs when other factors and insults are involved. THE PATHOLOGY OF PULMONARY OXYGEN TOXICITY The development of pulmonary oxygen toxicity can be divided into stages. ‘When lethal doses of oxygen are administered, such as 100% 02 at 1 atmosphere (ATA) , the phases are, in chronologic order, the initiation, inflammatory, and destructive stages“: The same chronologic order is maintained when sublethal doses of oxygen, generally 5 regarded as _<_85% Oz in rodents, are administered except the destructive stage is followed by a proliferative and possibly fibrotic stage. The order of morphologic changes is quite similar in different species: however, the severity and duration in each stage varies considerably from species to species. Since lethal concentrations of oxygen were administered to the rat and their alveolar macrophages for these studies, the stages of oxygen toxicity will be discussed predominantly from this framework. Initiation Stage The first 40-48 hours mark the initiation phase of pulmonary oxygen toxicity in the rat. During this time, the rat demonstrates few, if any, clinical signs of distresszs’”. This phase was initially thought to be a period characterized by the production of reduced oxygen species associated with no significant morphologic changes“. However, recent histologic work has demonstrated otherwise”. Wistar rats exposed to 100% O2 1 ATA for 24 hours demonstrated changes in the lung morphology. With light microscopy, alveolar collapse and vascular congestion with accompanying bronchiolar and vascular constriction in scattered areas were demonstrated. Further, the alveolar septae were folded and compressed which increased the local tissue density and caused adjacent alveolae to be overdistended by air. Arteriolar changes included adventitial distention by edema and fibrin deposition within the walls. Electron microscopy confirmed these findings. In addition, 6 arteriolar changes such as vacuolization of endothelial cells and fibrin deposition within the elastic layers were noted. Also, some capillary endothelial cells were separated from their basement membrane, and some capillaries demonstrated a pericyte reaction. No marked changes occurred in the alveolar epithelium. During the initiation phase, then, the rat appears to be tolerating the hyperoxia rather well upon gross physical examination. However, histologic evidence of oxygen toxicity is already present. Apparently, the initiating events are well underway within 24 hours of continuous exposure to hyperoxia. Inflammatory Stage After approximately 40-48 hours of continuous exposure to 100% 02, the rat begins to exhibit signs of distress26'27. Progressive lethargy, fur bristling, decreasing oral consumption , and respiratory distress develop . Histologically, at 40 hours post-exposure, the capillary endothelium begins to show evidence of damage25'27. This damage to the endothelial cell takes the form of vacuolization, swollen mitochondria, and plasma membrane detachment. Interestingly, platelets and RBCs aggregate at the areas of endothelial damage. Associated interstitial edema is noted. Over the next 20 hours, the alveolar capillary endothelium is reduced in number by 30%, alveolar and interstitial edema worsen, and increased interstitial cellularity, specifically increased numbers of neutrophils and indeterminate cells, is 7 seen”. No significant morphologic changes in the alveolar Type I and Type II epithelial cells are noted. Destructive Stage At 60 hours post-exposure, the physical signs which began in the inflammatory stage worsen. Usually, the rat succumbs within 72 hours of continuous exposure to 100% 0225'“'2°'3‘. Histologically, the capillary endothelium is further damaged. Capillary endothelial cells are reduced in number by 44%, and the surviving cells demonstrate evidence of further damage25'27. Although the Type I and II epithelial cells demonstrate some ultrastructural changes, no significant change in cell number is noted. The interstitium continues to increase in cellularity. Inflammatory cells, consisting of neutrophils, macrophages, monocytes, fibroblasts, and lymphocytes, become more predominant. The accumulation of neutrophils in the microvasculature and interstitium has been thought to herald the onset of this phase. The neutrophil influx has been associated with a rapid increase in the severity of lung injuryfin Further, evidence exists that this influx is due at least in jpart to ‘the endothelial damage with resultant increase in neutrophil adherence and subsequent intrapulmonary retention32'33 . However, neutrophils do not appear absolutely necessary for the deleterious effects of hyperoxia. The increase in alveolar permeability and fibrinopurulent intra-alveolar 8 inflammation which occurs initially appears to be independent of the accumulation of neutrophilsu'”. Moreover, induced neutropenia does not attenuate pulmonary damage nor prolong survivar“. Finally, a lung, devoid of neutrophils, appears to be "primed" for further damage by exposure to hyperoxia when the neutrophils are subsequently replacedyl Proliferative Stage The proliferative stage is present when sublethal concentrations of oxygen, such as ,5, 85% O2 1. ATA" are administered continuously“. When 85% O2 is administered to rats, the initiation phase is prolonged to 72 hours, the inflammatory phase to 5 days and the destructive phase to 7 daysz"25'3°. By day 7, the number of Type II alveolar cells has nearly doubled, and 41% of the capillary endothelial cells are destroyed. However, no further loss of endothelial cells is noted at the end of the following week”. Also by day 7, the proliferative response has begun. This response, which is characterized primarily by an interstitial influx of fibroblasts and monocytes and.is devoid.of:neutrophils, may be responsible for the survival of the anima12"25. Fibrotic Stage After lethal and sublethal exposure to hyperoxia, fibrosis has occurred within the interstitium. Fibrin deposition is noted by 48 hours and quite obvious by 72 hours of continuous exposure to 100% 0227. Also, after exposure to 9 70-85% 02 for 7 days, a fibrotic response is observed‘3-25. Although this stage is thought to occur later in oxygen toxicity, a fibrotic reaction apparently precedes the influx of neutrophils ‘when lethal concentrations of oxygen .are administered and follows the acute inflammatory response when sublethal concentrations are inspired2"25'27. MECHANISM OF PULMONARY OXYGEN TOXICITY Pulmonary oxygen toxicity is thought to occur due to the profuse production of reactive oxygen species - superoxide anion (02') , hydrogen peroxide (H202) , and hydroxyl radical (OH?) - which overwhelms the antioxidant defense systemhéfi9. During cellular respiration in a normoxic environment, oxygen is metabolized in the mitochondria to water. Oxygen free radicals are produced in small quantities and are metabolized enzymatically - by superoxide dismutase, catalase, and glutathione peroxidase and reductase - and nonenzymatically - by antioxidants such as vitamins E and C and beta- 1-6'39. However, as the concentration of inspired carotene oxygen increases, the production of superoxide and hydrogen peroxide, which occur in the mitochondria and microsomes and possibly other intracellular organelles, also increasesw43. This ultimately overwhelms the antioxidant defense system. Superoxide and hydrogen peroxide undergo further reaction to hydrogen peroxide and hydroxyl radical, respectively, via 10 a superoxide-driven Fenton reaction6'9'39'“: Fe” + 11202 = Fe3+ + on' + on- These reactive species may damage the cellular integrity and possibly cause cell death. Consequently, pulmonary damage occurs. Cellular damage and death by oxygen free radicals occurs by several different means‘“. Oxidation of DNA by superoxide and hydrogen peroxide results in chromosome aberrations, chromatid exchanges, breaks in the DNA and interference with repair and replication. Lipid peroxidation damages the cell membrane by decreasing the fluidity, increasing permeability, injuring the structural integrity, and altering transport. Sulfhydryl-containing enzymes, which include succinate dehydrogenase, xanthine oxidase, papain, glyceraldehyde-3- phosphate dehydrogenase among others, are oxidized during exposure to hyperoxia resulting in inactivation. DNA and protein synthesis are both inhibited as is the synthesis or transport of surfactant phospholipids. Finally, the microsomal enzymes involved in the cytochrome P,50 system may be induced leading to increased production of reactive oxygen species. The exact source(s) of the reactive oxygen species is not fully known. Presently, during acute hyperoxia, early significant lung injury is probably due to the increased intracellular generation of free radicals by the endothelial 11 cells resulting in their own demise9'2"25-‘5. Neutrophils, macrophages, and platelets have been identified as cells which release superoxide extracellularly“. Neutrophils probably augment the hyperoxic pulmonary injury secondary to free radical production, but this occurs in the latter stages of oxygen toxicity and may not be a requirement for lethalityzs'32'35'3a"7"9. Alveolar macrophages do not appear to increase in number, and evidence exists that they inhibit free radical productionzs'so. However, platelets have been shown to accumulate at the endothelial surface of the pulmonary capillary bed within 40 hours of continuous exposure to 100% (5 and.may, at least in part, be responsible for the pulmonary injury at this point-“"5. CYTOKINES Cytokines, the generic term that incorporates both lymphokines and monokines, are soluble polypeptide mediators51 . These antigenically nonspecific, intercellular mediators are produced primarily, but not exclusively, by a variety of immune cells, most notably T lymphocytes and macrophages. They are able to recruit other cells and modulate cellular/physiological responses. Macrophages are an important component of the innate immune system. They have been found to be significant producers of tumor necrosis factor (TNF), interleukin-1 (IL- 1), and interleukin-6 (IL-6)”. All three have been implicated 12 in the ”acute phase response" of the inflammatory process”. Each will be discussed in general terms and with regard to their involvement in pulmonary inflammation and oxygen toxicity. Tumor Necrosis Factor TNF is a 17 kd, 157 amino acid peptide produced by activated macrophages as well as many other cell types following a variety of stimulants”. A complex cytokine network has become increasingly understood over the years. Specifically, IL-l, as well as other cytokines, stimulates macrophages to produce TNF”: Additionally, TNF induces IL-l release from macrophages and endothelial cells“. TNF is an important inflammatory mediator. Systemically, TNF is involved in the physiologic response to endotoxin shock, induction of fever, and induction of the hepatocytes to produce the acute phase reactants‘s'". Locally, TNF affects endothelial cells, neutrophils, and.macrophages. In response to TNF or IL-1, endothelial cells express more intercellular adherence molecule-1 (ICAM-1, ICAM-2), vascular endothelial cell adhesion molecule-1 (VCAM—l), and endothelial-leukocyte adhesion molecule-1 (ELAM-l)“: These molecules increase adherence of the endothelial cell for leukocytes and stimulate transmigration of leukocytes. Procoagulant activity is increased and anticoagulant activity is suppressed by the vascular endothelial cells in response to TNF”. TNF induces 13 respiratory burst and degranulation in neutrophils and induces proliferation of fibroblasts. TNF appears to be an important mediator of pulmonary inflammation which may have both local and systemic effects. Serum TNF is increased in children with acute lower respiratory tract infections”. Those with elevated TNF concentrations were found to have longer duration of fever and higher levels of C-reactive protein but no increase in the clinical severity of respiratory tract infection. Animal studies have demonstrated that the lung, after inoculation with bacteria, may be a major source of systemic TNF“. Warren g; n1, has demonstrated that TNF participates in the development of immune-complex alveolitisflz TNF was shown to originate primarily from alveolar macrophages and was a necessary component for the full development of this pathology. Others have demonstrated a decreased TNF productive capacity of alveolar macrophages in septic patients and mice”. However, the TNF assays were performed only after lipopolysaccharide (LPS) stimulation and consequently, the macrophages may have already been, maximally stimulated. Levels of TNF in bronchoalveolar lavage (BAL) fluid were found to be elevated in those individuals with ARDS compared to normal subj ects”. Evidence for the involvement of TNF and IL-1 in pulmonary oxygen toxicity has been growing. Reactive oxygen species, particularly hydrogen peroxide, have demonstrated the ability to increase the LPS-induced TNF release from macrophages in 14 vitro. These reactive oxygen species also facilitated the LPS-induced TNF release in the sermd”. Further, TNF, IL-1, and LPS increased the mRNA levels of manganese-superoxide dismutase (Mn—SOD), the ‘mitochondrial SOD, of pulmonary epithelial cells with only minor increases noted in the fibroblasts“. Hyperoxia appeared to have no effect within 24 hours on the Mn-SOD mRNA levels of the pulmonary epithelial cells or fibroblasts. Also of note is that LPS has been shown to induce SOD levels in endothelial cells and monocytes‘z'“. Immune—complex activated alveolar macrophages demonstrated increased superoxide production following addition of IL-1 and, to a lesser extent, TNF“u Pretreatment with TNF and/or IL-l prior to exposure to hyperoxia appears to attenuate pulmonary injury. White §t_nII has shown in rats that administration of both TNF and IL-1 intravenously and intraperitoneally prior to administration of >99% 02 decreased lung injury and increased survival which was associated with a small increase in antioxidant enzymes early on (4-16 hours after exposure) and a much larger increase at 72 hours‘s'“. White and Ghezzi have also shown in a similar model that the beneficial effects of pretreatment could be partially blocked by lysine acetylsalicylate and, to a lesser extent, ibuprofen. This suggests that cyclooxygenase products may be responsible for the induction of antioxidant enzymes". Others have administered TNF alone via tracheal insufflation prior to exposure to hyperoxia and demonstrated a protective effect as evidenced by decreased lung injury and also increased 15 antioxidant enzymes levelsul This is in contrast to White's work in which evidence was presented that pretreatment with IL-1 played a more important role in protecting against hyperoxia than TNF. Finally, sera from endotoxin-treated rats conferred protection in recipients exposed to hyperoxia“. TNF and IL-1 levels were elevated in the sera of the recipients 60 hours post-exposure whereas the antioxidant enzymes were not. Consequently, TNF and IL-1 by themselves appear to have protected against pulmonary oxygen toxicity. Interleukin-1 IL-1 is a 17 kd, 159 amino acid peptide produced primarily by macrophages but also by endothelial cells and other cell typessm". As mentioned previously, IL-1 and TNF are capable of inducing macrophages to produce each other5"°9. Further, IL-l may induce the cell that produced it to make more IL—1 in a positive feedback loop 5"”. Although TNF and IL-1 have many similar properties and often act synergistically, IL—1 is distinctly different from TNF in that it participates directly in the activation of lymphocytes. Like TNF, IL-1 is an important inflammatory mediator. Systemically, IL-1 has been shown to induce shock, fever and production of acute phase proteins by the liver51'69'70. IL—l administered intravenously has been able to induce a shock- like state in rabbitsn'n. Further, when a low dose of both IL-1 and TNF were administered together, but not separately, a shock-like state was induced. This synergistic property 16 also resulted in marked accumulation of protein, red blood cells (RBCs) , and neutrophils in the alveolar space secondary to the disruption of the pulmonary vascular endothelium”. However, TNF has been found to induce greater lung injury than IL-l". This may be why TNF and not IL-1 appeared to be the primary mediator of acute pulmonary dysfunction in a canine model challenged with TNF and IL-173. On the other hand, IL-1 appears to play a greater role than TNF in the neutrophil accumulation despite similar abilities to induce increased adherence to endothelial cells”. Other studies have demonstrated that IL-1 initiates vascular congestion, cellular infiltration, and endothelial leakage". Finally, bleomycin- stimulated alveolar macrophages release IL-1 in the early phases of the fibrogenic response”. Possibly, IL-l plays a critical role in the initial development of pulmonary fibrosis. Interleukin-6 IL-6 is a 22 kd, 190 amino acid. peptide produced primarily by macrophages and activated T lymphocytes as well as fibroblasts and endothelial cells51'76. IL-6, along with TNF and IL—1, are all involved in inducing fever and.production of acute phase proteins, part of the acute phase response of the inf lammatory processs"77. Moreover, activated macrophages can release IL-6, along with IL-1 and TNF, in a coordinated fashion. Specifically, TNF and IL-1 induce production of IL- 6 in fibroblasts, and IL-1 induces production of IL-6 in 17 endothelial cells, fibroblasts, and monocytesn’”. Elevated serum levels of IL-6 have been identified in lung transplant patients undergoing rejection or experiencing pulmonary infection“. The source of the IL-6 was presumably considered to be the activated alveolar macrophages. PROTECTION AGAINST PULMONARY OXYGEN TOXICITY To date, other than to decrease the F102, no effective method to protect against oxygen toxicity exists in the clinical setting. However, this is often not a practical consideration. A number of experimental methods have protected against oxygen toxicity in animals. These methods have been aimed at the mechanism and inflammatory processes of oxygen toxicity. The antioxidants, vitamin E and butylated hydroxyanisole, and antioxidant enzymes, polyethylene glycol- conjugated superoxide dismutase and catalase, have been given prior to exposure to hyperoxia and found to decrease lung injury in rabbitsv'sz. Particularly, a decrease in the alveolar-capillary permeability was noted. Others have administered cimetidine to lambs resulting in attenuated oxygen-induced lung injury, decreased oxidant stress, and increased ratio of reduced to oxidized glutathione concentrations“. Cimetidine's mechanism of action is through competitive and noncompetitive inhibition of the cytochrome P450 system and subsequent decrease production of reactive oxygen species. In infant mice exposed to 80% oxygen, l8 dexamethasone was administered and found not to decrease lung collagen, thickness of the air-blood barrier, change the epithelial or basement membrane components, or neutrophils in the lungza'u'as. However, dexamethasone was found to decrease lung wet weight and lung water. Low-dose endotoxin administered prior to exposure to hyperoxia has also been found to extend survival in adult mice“. Similarly, endotoxin conferred protection from oxygen toxicity in adult ratsw'". This is believed to be due to the production of TNF and IL-1 after induction by endotoxin. While some have suggested that the protection against oxygen toxicity from TNF and IL-1 is due to the induction of antioxidant enzymes, particularly Mn- SOD, others have demonstrated that induction of these enzymes is not necessary for protection“"8'°7. Studies have shown that leukotriene B, (LTB‘) protects rats from hyperoxic lung injury. One possible explanation is that TNF and IL—1 stimulate prostaglandin E2 (PGEZ) production which subsequently leads to decreased production of LTB, and formation of free radicals from neutrophils”. Unfortunately, many of these treatments are not practical. Administration of antioxidants needs to take place approximately 2 days prior to exposure to hyperoxia in order to confer protection. One does not usually desire to give steroids or endotoxin to an individual who may already be suffering from an infection or sepsis. Finally, administration of antioxidant enzymes seems effective but its clinical applicability is unknown. Consequently, no effective 19 treatment is available at this time. PURPOSE The purpose of this project was to determine if cytokines ‘were involved in oxygen-induced lung injury; Toidate, TNF and IL-1 have been implicated in the development of pulmonary inflammation due to a variety of etiologies but not oxygen. Further, though TNF and IL-1 appear to play a role indigeggly in the pathogenesis of pulmonary oxygen toxicity, it is difficult to determine from previous work whether it is a protective or detrimental one. Finally, little is known with regard to the role of IL-6 in pulmonary inflammation, let alone oxygen toxicity, despite its involvement in the acute inflammatory process and close interactions with TNF and IL-1. The aims of this project, therefore, were to determine if TNF, IL—l, and/or IL-6 were involved in oxygen-induced lung injury utilizing the following methods in the rat model: 1. Expose alveolar macrophages In_yi§xg to lethal hyperoxia for 30 minutes and 1 hour followed by assessment of their ability to produce TNF, IL-1, and IL-6 in response to LPS challenge. 2. Expose the animals In vIvo to hyperoxia for 1, 2, and 4 hours. The alveolar macrophages will then be harvested and their ability to produce TNF, IL-1, and IL-6 in response to LPS challenge determined. 3. Obtain serum samples from animals exposed to 20 hyperoxia for 1, 2, and 4 hours to determine the TNF, IL-1, and IL-6 concentrations. Assess membrane-associated levels of the particular cytokine(s) that demonstrated increased alveolar macrophage productive capacity. If the cytokine(s) is elevated, block or downregulate the production and determine if this has protective or detrimental effects with regard to pulmonary oxygen toxicity. MATERIALS AND METHODS ANIMAL MODEL The unmanipulated, adult male Sprague-Dawley rat was chosen as the animal model for studying the involvement of cytokines in pulmonary oxygen toxicity. The first and foremost reason is that the rat demonstrates a pattern of pulmonary injury similar to primates but over a shorter period of time9"3. Also, the animal is small, easy to house, and relatively inexpensive. Adult animals were chosen over neonates because neonates appear to have induction of their antioxidant enzymes which results in longer survival times compared to the adult animals”. Additionally, adult rats, as opposed to adult mice, allow harvesting of an adequate number of alveolar macrophages via bronchdalveolar lavage”. 21 22 GENERAL Pathogen and viral antibody free male Sprague-Dawley rats (from Charles River Labs., Portage, MI) weighing 275-350 g were used throughout the studies. They were housed and acclimatized for at least 72 hours at the University Laboratory Animal Research (ULAR) building prior to experimentation. The rats were kept on.a 12 hour light on and off cycle. Prior ‘to experimentation, they ‘were fasted overnight (approximately' 16 .hours) but allowed.‘water"ng Iinitnn. All protocols were carried out in accordance with the guidelines set forth in the Animal Welfare Act and as outlined in the Guide for Care and Use of Laboratory Animals by the National Institutes of Health Publications. m EXPOSURE OF ALVEOLAR MACROPHAGES TO HYPEROXIA AND NORMOXIA Alveolar macrophages were harvested by bronchoalveolar lavage (BAL) immediately after each rat was euthanized by overdose of ether. This was accomplished by cannulating the exposed trachea with a dulled 18 gauge needle under direct vision and lavaging the lungs with ice-cold Dulbecco's Modified Eagle’s Medium (DMEM), supplemented with glutamine (GIBCO Labs., Grand Island, NY) and gentamicin (GIBCO Labs.). Three separate, consecutive 5 ml lavages were carried out, and 23 the effluents combined in a 15 ml sterile conical centrifuge tube (Falcon 2095, Becton.Dickenson.and.Co., Lincoln Park, NY) kept on ice. Though precise amounts of each total collected effluent were not recorded, this was usually 1211 ml. Once all the effluents were collected for each animal, they were then centrifuged at 320 x g for 15 minutes at 4° C (Sorvall RT 6000B and D models, DuPont Biotechnology Systems, Wilmington, DE). The cells were resuspended in 1 ml DMEM. The number of viable macrophages were determined by morphologic criteria and tryphan blue exclusion. The cell suspension was then diluted to 2 ml. One ml of the each cell suspension was placed in one of two 24 well plastic plates, and the other ml was placed in the remaining plate. The macrophages were allowed to adhere on plastic for 2 hours by placing them in an incubator (Steri- Cult 200 Incubator, Forma Scientific Inc., Marietta, OH) at 37° C, 5% C02, and >90% humidity. The cells were washed with DMEM, and 1 ml DMEM was then added to each well. One plate with adherent cells was placed in an air-tight container (Modular Incubator Chamber, Voss Ind. Inc., Cleveland, OH) with equal sized inlet and outlet ports and exposed to 95% O2 and 5% CO2 for the selected time period (30 or 60 minutes). The other plate was placed back in the incubator and exposed to 21% O2 and 5% CO2 for the same length of time. The 02 concentration was verified by an oxygen gas analyzer (5525 Oxygen Analyzer, Hudson Ventronics Division, Temecula, CA) and the C02 concentration, by a Pyrite carbon dioxide gas analyzer (Bacharach Inc., Pittsburgh, PA). After the time of exposure 24 was complete, the plates were removed from their respective environments and the DMEM in each well was then replaced with 1 ml of DMEM containing 10% by volume heat inactivated fetal bovine serum (FBS, Biologos Inc., Naperville, IL) and lipopolysaccharide (LPS, 10 pg LPS/ml DMEM) (LPS from Escherichia coli 055:85, Difco Labs. , Detroit, MI). The macrophages were then incubated for 24 hours. Following the incubation period, the supernatants were collected, aliquoted, and stored at -70° C until cytokine determination by bioassay was to be performed. IN VIVQ EXPOSURE OF ALVEOLAR MACROPHAGES TO HYPEROXIA AND NORMOXIA Alveolar macrophages were exposed to hyperoxia and normoxia In vIvg by placing unmanipulated rats into one of two equal compartments of a specially designed 3 x 1 x 1' plexiglass chamber. One rat was placed in each compartment which contained bedding with wood chips. Both compartments had inlet and outlet ports of 5/16" diameter. After flushing the compartments for 3 minutes at 15 liters per minute (LPM) with either room air or medical grade 02 (AGA Medical Gas, 99.5% purity, AGA Gas Inc., Cleveland, OH), the animals were then exposed to the appropriate gas, as determined by prior computer randomization, at 8.5 LPM for the selected time period (1, 2, or' 4 hours). This allowed simultaneous 25 exposures to be carried out. The humidity and temperature of each compartment were constantly monitored and ranged from 48- 70% and 19.4-24.4? C, respectively. The atmospheric pressure was checked immediately prior to the beginning of an experiment. Due to the fact that the inlet and outlet ports were of the same size, the pressure within the chamber would not be expected to be vastly different from the atmospheric pressure. The 02 concentration on the hyperoxic side was continuously monitored and was 298% at all times. The 02 concentration on the normoxic side was measured at selected intervals and found to be 21% at all times. The CO2 concentration was determined at 2 and 4 hours post-exposure and was always 50.2%. During the period the animals were exposed to hyperoxia or normoxia, they were allowed food (standard rat chow) and water ng_linitnn. Immediately upon completion of exposure, euthanasia of each rat was accomplished by overdose of ether within 4 minutes. A midline neck incision was extended through the thorax. The thoracic cavity and pulmonary parenchyma were grossly inspected. A 1.0 ml blood sample was obtained by cardiac aspiration, transferred to sterile plastic Microtainer tubes containing a serum separator (Becton Dickenson and Co., Rutherford, NJ), and placed on ice. BAL was then carried out in the fashion described earlier. Briefly, the trachea was cannulated with a dulled 18 gauge needle which was secured in place with 3-0 silk. Three consecutive 5.0 ml lavages were performed with ice-cold DMEM. The combined effluents were 26 transferred to a 15 ml sterile conical centrifuge tube. Though precise amounts of the total effluent were not recorded, it was usually 1211 ml. Blood samples and BAL effluents were prepared in the following manner. The blood was separated by centrifugation at 12,000 x g for 15 minutes at 4° C (Model 235C Micro- Centrifuge, Fisher Scientific, Chicago, IL). The serum samples were aliquoted and stored at -70° C until cytokine bioassay could be performed. The BAL effluents were centrifuged at 320 x g for 15 minutes at 4° C. The cells were then resuspended in 1 ml supplemented DMEM. The number of viable macrophages were determined by morphologic criteria and tryphan blue exclusion. Counts >7 x 105 cells/ml or <1 x 105 cells/ml were excluded. Large counts indicated that the animal likely had an underlying respiratory infection which was not apparent when the animal was examined by ULAR personnel. Small counts were likely secondary to inadequate lavage. Viable macrophages were diluted to 1 x 105 cells/ml in supplemented DMEM, and 1 ml of the macrophage suspension was placed in a 24 well plastic plate. The cells were allowed to adhere on plastic for 2 hours by placing them in an incubator (37° C, 5% C02, >90% humidity). The cells were washed with DMEM and then checked for adherence microscopically. One ml 10% FBS/DMEM with LPS (10 pg LPS/ml DMEM) was added to each well. The macrophages were then placed in an incubator for 24 hours. At the conclusion of this time period, the supernatants were 27 collected, aliquoted, and stored at -70° C until assayed. To verify further that the animals were breathing the predetermined concentration of oxygen, a group of rats underwent arterial blood gas (ABG) analysis. This was accomplished in the following manner: Under light ether anesthesia, the right carotid artery was catheterized via a midline neck incision with a hybrid catheter consisting of a 3.0 cm sialastic tip (size 0.02" I.D.x 0.037"O.D, Baxter Healthcare Corporation, Scientific Products Division, McGraw Park, IL) secured to PE 50 tubing (Becton Dickenson and Co., Parsipanny, NJ). The tip was secured to the artery with 6-0 nylon suture. The tubing was tunnelled subcutaneously posteriorly and brought through a separate small skin incision at the nape of the neck. The catheter was then threaded through a miniature single channel fluid swivel (Harvard Apparatus, South Natick, MA). This swivel was sutured to the musculature with 4-0 nylon, and the skin was approximated with the same material. After an adequate recovery period from the anesthesia, the animal’s temperature and hematocrit were determined, and baseline arterial blood gas analysis obtained. Utilizing a Harvard pump (Model 2400003, Harvard Apparatus, South Natick, MA), heparinized normal saline (2U heparin/ml NS) (bovine heparin from Upjohn Company, Kalamazoo, MI) was infused continuously at 0.358 ml/hr to replace the blood loss due to blood draws at a 3:1 ratio of solution infusionzblood loss and to keep the catheter patent . The animal was then exposed to either a normoxic or hyperoxic condition as 28 described above. At l-hour post-exposure, another ABG analysis was performed. DETERMINATION OF CELL-ASSOCIATED TNF Since the TNF productive capacity of the alveolar macrophage from the In yIvg exposure to hyperoxia was increased at 2 hours (please see RESULTS section), cell- associated.TNF‘was assessed near this time period.according to the methods of Hogan et nl.9‘. Following exposure of the animals to normoxia or hyperoxia for 2, 3, or 4 hours and BAL performed with subsequent centrifugation of the effluents as described previously, the number of viable alveolar macrophages were determined by morphologic characteristics and trypan blue exclusion. Cells from each sample were then diluted to 5 x 105983 92 LL92 >9§§ 02 m2 m 1 ND ND 46:16 96:37 14:4 19:4 2 ND ND 658:8? 577:55 8:2 8:2 4 ND ND 7:2 20:9 18:3 23:7 53 RESULTS OF THE CELL-ASSOCIATED TNF BIOASSAY AFTER EXPOSURE TO NORMOXIA OR HYPEROXIA It! 2129 Preliminary data revealed no marked increase in macrophage/eell-associated TNF following 2, 3, and 4 hours of in vivo exposure to hyperoxia compared to their respective normoxic controls. Moreover, the amount of fluorescence/cell was barely above background for each group. Consequently, no further studies utilizing this technique were performed. Presented below are the results of two typical experiments on alveolar macrophages harvested following 3 hour in vivo exposures (Table 4). Table 4: Cell-associated TNF values after 3 hour exposures to hyperoxia or normoxia. Results are presented as amount of fluorescence/cell from each random reading and as meantsmt with 8:2. Mae—W11. ...—N 2L- 02 22.9.”; 1 29.11 29.30 23.49 14.85 42.82 27.55 2 2.84 15.63 3.84 8.32 4.96 8.03 MEAN:SEM 17.84:7.40 17.28:4.12 54 THE EFFECTS OF HYPEROXIA 0N RATS AFTER PRETREATMENT WITH PENTOXIFYLLINE Since PTX has been shown to downregulate macrophage production of TNF and superoxide anion, inhibit macrophage chemotaxis, decrease the inflammatory effects of TNF and IL-1, and attenuate sepsis/TNF-induced lung injury, it was chosen as the treatment agent in this project involving cytokines and pulmonary oxygen toxicity‘““". However, PTX has been primarily administered intravenously, and Wang ML have demonstrated that PTX can cause marked hypotensionm. Consequently, an acceptable administration route and dosage had to be determined - one which would not appreciably change the healthy rat model nor result in marked hypotension while, at the same time, hopefully retaining the salutary effects of PTX. The Effects of PTX on MAP Prior to and after subcutaneous administration of PTX 50 mg/kg BW - a route which would not appreciably alter the healthy rat model, and a dosage which has been shown to have salutatory effects, serial determinations of the MAP were mademv 112-114 . Readings were taken every minute for 15 minutes initially - the time interval around the lowest pressure, and then periodically until the MAP returned to its original baseline level. No significant differences were found among 55 the MAP determinations at any of the time points with this dosage and route of administration (Figure 8). The lowest average MAP was 95.8 mm Hg which occurred 5 minutes after administration of PTX. This represented a decrease of 11.5 mm Hg from the pre-injection reading. The MAP progressively returned to its baseline level within 1 hour of PTX administration. .Also at 1 hour following PTX administration, the subcutaneous *weal which. had. been. raised. wee» barely appreciable. 56 120 — 110 - 100 P m 0 \ WV Mean Ar’reriol Pressure (mm Hg) 4— .1 I l l l O i I l I 0 10 20 30 40 $0 60 Time (min.) Figure 8: Time course of changes in the MAP following subcutaneous injection with PTX (50 mg/kg BI) . Meanisml, fl=4/grp, one-way ANOVA, significance set at p<0.05. 57 The Results of Prolonged Hyperoxia After Pretreatment With PTX In this study, the animals were divided into 4 groups: 1) those that were exposed to normoxia and were administered NS (N/NS) , 2) those that were exposed to normoxia and were administered PTX subcutaneously (50 mg/kg BW) (N/PTX) , 3) those that were exposed to hyperoxia (>95% 02) and were administered NS (H/NS) and 4) those that were exposed to hyperoxia and received PTX (H/PTX). Pretreatment with PTX appeared to attenuate the pulmonary injury that occurred with sustained lethal hyperoxia. The BAL supernatant LDH concentration of the H/PTX group was significantly decreased compared to the level of the hyperoxic/NS treated group but significantly increased compared to normoxic controls [3.67:0.40 (H/PTX) vs 4.97:0.40 (H/NS), 2.32:0.20 (N/NS), and 2.33:0.19 (N/PTX) IU/lOO ml, p<0.05, respectively] (Figure 9). As expected, the untreated H/NS group demonstrated markedly elevated levels compared to all groups (4.97:0.40 (H/NS) vs 3.67:0.40 (H/PTX), 2.32:0.20 (N/NS), and 2.33:0.19 (N/PTX) IU/lOO ml, p<0.05, respectively]. The supernatant protein concentration of the H/PTX group was markedly decreased compared to the concentration of the H/NS group, and was not significantly different from the levels of the normoxic controls [2.456:O.531 (H/NS) vs 0.477:0.136 (H/PTX), 0.105:0.016 (N/NS), and 0.097:0.013 (N/PTX) mg/ml, p<0.05, respectively] (Figure 10). Further, the supernatant IL—6 concentration of 58 the H/PTX group was significantly decreased compared to the level of’ the H/NS group and, moreover, not appreciably different when compared to the normoxic controls (8.90:3.58 (H/NS) vs 2.07:0.71 (H/PTX), 0.0192io.0142 (N/NS), and 0.0250:0.0189 (N/PTX) U/ml, p<0.05, respectively] (Figure 11) . Additionally, the amount of pleural fluid present in the H/PTX group was not significantly different from the normoxic controls and significantly decreased from the level of the H/NS group [1.90:0.47 (H/NS) vs 0 (N/NS and N/PTX) and 0.14:0.01 ml (Ii/PTX) ml, p<0.05, respectively] (Table 5). The percentage of the lavage fluid recovered was >80% for all groups (Table 5). The percentage recovered from the H/PTX group was not appreciably different from the normoxic controls, but the percentage of effluent recovered from the H/NS group was significantly greater when compared to all other groups [86.77:0.74 (H/NS) vs 82.32:0.98 (H/PTX), 82.62:l.22 (N/NS), and 82.17:0.94 %, p<0.05, respectively]. Finally, of note, no TNF or IL-1 was detected in the supernatant of any of the groups at this time point. 59 311-535322aaéaix ? c :::: MW / I} g: :2: Z :: 23:: T \\ % 1:: § % z: k 6 exposures to normoxia or hyperoxia with or without PTX pretreatment. Meanisnx, N=5-6/grp, one-way ANOVA then Tukey's test, p<0.05 vs all groups. 60 3.0 - C] Normoxio/ NS Normoxia/PTX 2 5 _ Hyperoxia/NS m Hyperoxia/PTX A EAZeo- * \c 0.0 ..— Ec VE15_ c0 '80 ...: 0") LV (L 0.5 - \\\\\\\\\\\\\\\\\\\—~ 0.0 I I m Figure 10: Supernatant protein concentrations after 52 hr exposures to normoxia or hyperoxia with and without PTX pretreptment. neanisml, N=s-6/grp, one-way ANOVA then Tukey's test, p<0.05 vs all groups. 61 12 .- [:l Normoxia/NS Normoxia/PTX 10 P Hyperoxia/NS 3 _ m Hyperoxia/PTX lL-6 (U/ml) (Supernaioni) \\V \\\\‘ 5 ‘% e e ee 0 0 0 0 0 0 00000 0 0 0.0 0 .0 0 0 0 .0 0 0 0 0 .0 0 0 0 0 ‘% 0 0 0 0 .0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 .0 0 e e e e e e e '% e 0 0 0 0 0 0 0 0 0 0 0 0 .0 0 0 0 o “a; eeee eee eeee eee eeee eee fvv5 eee Figure 11: Supernatant IL—G concentrations after 52 hr exposures to normoxia or hyperoxia with and without PTX pretregtment. Meanisml, N=5-6/grp, one-way ANOVA then Tukey's test, p<0.05 vs all groups. 62 Table 5: Parameters determined after 52 hr exposures with and without PTX. :Mean:szx, N=5-6/grp, one-way'ANOVA then.Tukey's test, p<0.05 vs all groups and p<0.05 vs normoxic controls. 21% Q2 >96i_Q2 Treatment HS 213 NS PIX Hematocrit 50.2io.7 49.3:0.6 59.0iz.2* 53.3:l.5 Pleural fluid (ml) 0 o 1.910.5* 0.6:0.6 Ascites (ml) 0 0 0 0 % change in wt 9.4:O.8 9.5:o.7 l.6:l.1’ 2.430.6’ % lavage fluid 82.6:1.2 82.2:O.9 86.8:0.7* 82.3:1.0 recovered Pretreatment with PTX also appeared to attenuate some of the systemic effects of sustained hyperoxia. The serum IL-6 concentration of the H/PTX group was significantly decreased compared to the level of H/NS group and was not appreciably different from the values of the normoxic controls [12.33:4.61 (H/NS) vs 3.34:1.40 (H/PTX), 0.0680:0.0446 (N/NS), and 0.860:0.397 (N/PTX) U/ml, p<0.05, respectively] (Figure 12). The serum IL—1 level of the H/PTX group [701:125 U/ml] was not significantly elevated compared to the values of the normoxic controls [143:73 (N/NS) and 237:162 (N/PTX) U/ml] nor was it significantly decreased compared to the concentration of the H/NS group [910:256 U/ml]. However, the serum IL-1 concentration of the H/NS group was significantly greater compared to the normoxic controls [910:256 (H/NS) vs 143:73 (N/NS) and 237:162 (N/PTX) U/ml, p<0.05, respectively]. 63 Further, the hematocrit of the H/PTX group was markedly decreased and closer to within normal limits compared to the value of the H/NS (Table 5). Moreover, the hematocrit of the H/PTX was not significantly different from the normoxic controls while the hematocrit of the H/NS group was dramatically increased compared to all groups [S9.0:2.2 (H/NS) vs 53.311.5 (H/PTX), 50.2io.7 (N/NS), and 49.3:O.6 (N/PTX) %, p<0.05, respectively]. However, pretreatment with PTX failed to demonstrate an increase in weight similar to controls [1.58:l.12 (H/NS) and 2.38:0.56 (H/PTX) vs 9.42:0.77 (N/NS) and 9.47:0.66 (N/PTX) %, p<0.05, respectively] (Table 5). No differences in the serum LDH and protein concentrations were found among the groups (Figures 14 and 15) . No TNF was detected in the serum of any of the animals, and no ascites was present in any of the animals. 64 0 18 - 16 _ [:1 Normoxia/NS 14 r Normoxia/PTX 12 __ Hyperoxia/NS 10 _ m Hyperoxia/PTX 8 r- 6 ,_ C.‘ E A Z ' E r- 3 3 / 3 0.5 / “I, m V z." x\\\\§ \‘ WT Figure 12: Serum IL-G concentrations after 52 hr exposures to normoxia or hyperoxia with and without *PTx. Mean:8ml, N=5- 6/grp, one-way ANOVA then Tukey's test, p_95% 02 suffered significant 79 pulmonary damage as evidenced by increased BAL supernatant concentrations of LDH and protein and increased accumulation of pleural fluid. These indices have been utilized by others as evidence of pulmonary injury secondary to hyperoxia as well as other various in8u1t828,29,31,32,37,67,68,73,101-103,137,138. Additionally, markedly increased BAL supernatant concentrations of IL-6 were present in those animals that were exposed to hyperoxia indicating that this cytokine probably is associated with of oxygen-induced lung injury. Pretreatment with PTX led to marked attenuation of pulmonary injury. Those rats pretreated with PTX and exposed to hyperoxia showed a marked decrease in the supernatant concentrations of LDH and protein as well as a decline in the amount of pleural fluid present compared to those who were exposed to hyperoxia and did not receive treatment. Further, except for the LDH supernatant level, these decreased values were not significantly different from the normoxic controls. Moreover, the supernatant concentration of IL-6 of those exposed to hyperoxia and pretreated with PTX was markedly reduced compared to those which suffered exposure to hyperoxia without treatment. This reduced level, too, was not significantly different from the normoxic controls. This further supports the involvement of IL-6 in the development of pulmonary oxygen toxicity. To our knowledge, this is the first direct evidence that IL-6 may be may be associated with this type of injury. Exposure to hyperoxia also led to other systemic changes. 80 Hemoconcentration.as evidenced.by increased.hematocrit values in those exposed to hyperoxia was present in this study and has been recognized by others25'67. This most likely was secondary to extravasation of intravascular fluid into the thoracic cavity and probably the pulmonary parenchyma as evidenced by the accumulation of pleural fluid and lack of ascites. Additionally, hyperoxia led to failure to thrive as evidenced by lack of weight gain; others have observed weight 1033'”. As noted earlier by Roth, hyperoxia did not result in increased serum LDH and protein concentrations‘o‘. Perhaps most important from a systemic perspective, hyperoxia led to increased serum IL-1 and IL—6 concentrations. This is the first study showing that elevated levels of these cytokines have been observed in oxygen-induced lung injury. PTX attenuated some of these systemic effects. Those pretreated with PTX and exposed to hyperoxia developed a hematocrit value and serum IL-6 level that were not significantly different from controls. The serum IL-1 level, though not significantly different from the normoxic controls, was also not significantly decreased from the untreated hyperoxia-exposed animals. Further, PTX did not improve the animals’ weight gain amidst exposure to high oxygen concentrations. These findings raise several interesting questions. First, which cell(s) are responsible for the increased BAL concentrations of IL-6? Secondly, what is the relationship between the earlier findings of increased IL-1 and. TNF 81 productive capacities of alveolar macrophages after a short exposure to hyperoxia and the increased BAL IL-6 concentrations after exposure to hyperoxia for 52 hours? Thirdly, what is the mechanism by which PTX attenuates oxygen- induced lung injury and decreases BAL supernatant IL-6 concentrations? Fourthly, what do the increased serum IL—1 and IL-6 concentrations after exposure to sustained hyperoxia mean? Finally, what does the attenuation of serum IL—6 concentration by pretreatment with PTX tell about the role of cytokines in pulmonary oxygen toxicity? As the increase in BAL IL-6 concentration occurred within 52 hours, i_,_e_,_, during the initiation or inflammatory phases, alveolar macrophages and endothelial cells seem to be the most likely sources of IL-6. Alveolar macrophages are components of the innate immune system, and endothelial cells are among the first cells morphologically affected by hyperoxiazsvu. Fibroblasts are unlikely as they are thought to be a critical component later on in those situations when lesser concentrations of oxygen are inspired , jug, during the prol iferat ive phasez’"ZS . Moreover, it seems unlikely that the BAL supernatant IL-6 levels were elevated due to the increased serum IL-6 concentrations diffusing through capillary-alveolar barrier as the comparative elevation of serum IL-1 levels did not result in increased supernatant IL-l despite the fact that IL-1 is a smaller polypeptide than IL-G”. The earlier findings of increased alveolar macrophage productive capacities of IL-1 and TNF shortly after exposure 82 to hyperoxia also suggest in part that alveolar macrophages or endothelial cells may be responsible for the increased supernatant concentration of IL—6. Others have demonstrated that activated macrophages can release IL-6, along with TNF and IL-1, in a coordinated fashion. More precisely, TNF can induce IL-l release from macrophages or endothelial cells, and IL-1 can induce production of IL-6 from macrophages and endothelial cells‘s-n’”. Though TNF or IL—1 were not detected in the supernatant at 52 hours, they possibly could have been produced earlier and are no longer present in significant amounts at that time59'119. Additionally, IL—6 may be the mediator through which neutrophils inflict further damage on the pulmonary parenchyma since IL-6 has been shown to stimulate neutrophilic lysozyme secretion and prime the neutrophils to generate an enhanced respiratory burst”). Possibly, then, a sequence of cytokine production is involved. This cytokine involvement may be responsible for the priming of the lung for further damage - independent of neutrophils - to which Krieger m were referring to in their studies of pulmonary oxygen toxicity”. This priming may involve interaction between cytokines and LTB‘ with a resultant effect on neutrophils. LTB‘ is known to induce the generation of TNF in BAL supernatant and is thought to be an intermediary responsible for the neutrophil diapedesis‘“. Moreover, TNF and IL-1 have demonstrated the ability to induce synthesis of an endothelial cell surface factor(s) which promotes increased neutrophil adherencew". So 83 possibly the finding by others that LTB, is responsible for vthe accumulation of neutrophils in pulmonary oxygen toxicity actually involves the interaction of LTB‘ and cytokines”. PTX was shown to attenuate oxygen-induced lung injury as evidenced by decreased levels of supernatant LDH and protein and accumulation of pleural fluid, Moreover, PTX resulted in decreased supernatant concentrations of IL-6. As mentioned earlier, this drug is known to have a short half-life and to decrease the TNF production by inhibiting TNF mRNA production‘“"33. Wang M: demonstrated decreased serum TNF concentrations as well as IL-6 levels with PTX treatment in a trauma-hemorrhage :model"2. Since PTX: was administered immediately prior to exposure to hyperoxia, and earlier it was demonstrated that the TNF productive capacity of the alveolar macrophage was increased shortly after exposure to hyperoxia, possibly PTX lowered the IL-6 supernatant concentration by inhibiting TNF production. Unfortunately, the precise mechanisms by which IL-6 is involved in oxygen toxicity and PTX attenuates oxygen-induced lung injury cannot be established from this work. Prolonged hyperoxia also resulted in increased serum concentrations of IL—1 and IL-6. Further, pretreatment with PTX demonstrated a serum IL-6 concentration which was decreased and an IL-l concentration which appeared to be trending upward though not significantly different from the normoxic controls and untreated hyperoxia-exposed animals. Taken together, this suggests an association and chronology 84 between IL—1 and Il-6, 143:, IL-6 elevation appears to follow the IL-1 increase. Serum TNF levels possibly could have been elevated prior to 52 hours though this was not demonstrated. Indeed this would not be unexpected in light of studies involving a hemorrhage model in which TNF is increased shortly after hemorrhage is induced‘”. The TNF level then declines to insignificant levels but is responsible for initiating a cytokine cascade. Further, if PTX inhibits TNF production, as hypothesized earlier, then the decreased IL-6 level would consistent with the work of Wang gt al,"2. The source(s) of the increased serum cytokines were not determined in this project. The lung appears to have the capability to produce TNF - and possibly other cytokines -— systemically“. However, others have shown that although ARDS may result in increased BAL supernatant TNF and alveolar macrophage production of IL-1, the increased serum concentrations of TNF and monocyte production of IL-1 may be the result of ongoing shock and sepsis as opposed to the ARDS f59'1‘3'1“. Therefore, possibly another insult was process itsel involved which led to the increased serum levels of IL-1 and IL-6. Certainly in this work, a degree of hypovolemia is suggested in 'those animals exposed to Jhyperoxia by' the accumulation of intravascular fluid into the thoracic cavity, the increased hematocrit values, and the lack of weight gain. Whether this amount of hypovolemia is sufficient to result in increased serum concentrations of cytokines - as occurs in 85 hemorrhagic and septic/endotoxin-induced shock - cannot be ‘determined. from this work, although it should. be investigatedm'r’v"'9°'"2"‘9"23'“5"5°. Virtually all the rat models utilized, including the one in this project, have been given food and water gg_libitgm. Certainly as the rat becomes more distressed his oral intake is inadequate as evidenced by the lack of weight gain in those animals exposed to hyperoxia. Further studies should be done with the MAP being continuously monitored and adequate intravascular fluid administered to determine if this hypovolemia is a significant concern and the model heretofore employed adequate. SUMMARY AND CONCLUSIONS SUMMARY This project determined that the cytokines, TNF, IL—1, and IL-6, are involved in pulmonary oxygen toxicity. The IL—1 and TNF productive capacities of the alveolar macrophages were upregulated shortly after exposure to hyperoxia in_gitrg and in yiyo, respectfully. Also, those animals exposed to sustained hyperoxia demonstrated an increased BAL supernatant concentration of IL-6. Pretreatment with PTX led to attenuation of the BAL supernatant IL-6 concentration as well as the indices of pulmonary injury. This suggests that supernatant IL-6 concentration is associated with the severity of pulmonary injury. Further, those animals exposed to prolonged hyperoxia demonstrated increased serum IL-1 and IL-6 concentrations with associated hemoconcentration and accumulation of pleural fluid. Pretreatment with PTX led to attenuation of serum IL-6 levels and decreased the degree of hemoconcentration and pleural fluid accumulation. All animals exposed to hyperoxia for prolonged periods failed to gain weight despite being 86 87 allowed food and water ag_;1bitgm. These findings suggest that hypovolemia may be involved. If so, this model of hyperoxia - which is widely utilized - may need to be modified. In summary, the findings are the following: 1. Alveolar macrophages demonstrated an enhanced IL-l productive capacity after exposure to hyperoxia.1n_gi§rg for 1 hour. 2. Alveolar macrophages demonstrated an enhanced TNF productive capacity after exposure to hyperoxia in yivo for 2 hours. 3. Serum TNF, IL-1, and IL—6 concentrations were not elevated after exposure to hyperoxia for 1, 2, or 4 hours. 4. Prolonged hyperoxia resulted in increased BAL supernatant concentrations of IL-6 which may be associated with the severity of pulmonary oxygen toxicity. 5. Prolonged hyperoxia led to increased systemic serum concentrations of IL-6 and IL-1. 6. PTX attenuated oxygen-induced lung injury and decreased the BAL supernatant concentrations of IL-6 which further supports its association with the severity of oxygen- induced lung injury. 7. PTX attenuated the systemic serum concentration of IL-6. 8. The hemoconcentration, leakage of intravascular fluid, and lack of weight gain indicate that inadequate oral intake has occurred and that hypovolemia may be involved. 88 Consequently, allowing food and water mm only may be inadequate. CONCLUS I ONS These studies demonstrated.that cytokines - specifically TNF, IL-1, and IL-6 - are involved in oxygen toxicity. The alveolar macrophages are primed shortly after exposure to hyperoxia to produce IL-1 in vitro and TNF m. Additionally, both the BAL supernatant and serum IL-6 levels were elevated in those animals exposed to prolonged hyperoxia. Further, PTX pretreatment was shown to markedly attenuate pulmonary oxygen toxicity as well as decrease the BAL supernatant and serum IL-6 concentrations. Though the precise mechanism requires further investigation, this may occur through inhibiting production of TNF with resultant decrease in IL-6 levels. In contrast to other injuries in which the time of insult cannot be predicted and consequently adequate pretreatment is virtually impossible, administration of high oxygen concentrations is a predetermined, controlled event. Therefore, PTX may have a role in protecting patients from oxygen-induced lung injury. APPENDIX APPENDIX A ADDENDUM This work has been presented in part at the Surgical Forum section of the 78th annual Clinical Congress of the American College of Surgeons (10/13/92) and published in Surgical Forum XLIII 15‘. At that time, we presented our initial findings, .i:§:, ”Alveolax' macrophages, exposed. to hyperoxia demonstrate early enhanced cytokine productive capacity." During the same year, Jensen §;_al: demonstrated that TNF, in particular, and possibly IL-1 and IL-6 may play a role in oxygen toxicity ”2. Although their work supports the basic hypothesis that these cytokines are involved in oxygen toxicity, the model and methodology differ significantly from ours. 89 APPENDIX B Amino acids Arginine HCl Cystine . . Glutamine . Glycine . . Histidine HCljgo Isoleucine . Leucine . . Lysine HCl . Methionine . Phenylalanine Serine . . . Threonine . Tryptophan . Tyrosine . . Valine . . . Vitamins Choline chloride Folic acid . Nicotinamide Calcium d-pantothenate . . . . Pyridoxal HCl 90 30.0 42.0 104.8 104.8 146.2 30.0 66.0 42.0 95.2 16.0 72.0 93.6 91 Riboflavin . . . . . . . . . . . . . . . . . 0.4 Thiamine HCl . . . . . . . . . . . . . . . . 4.0 Inositol . . . . . . . . . . . . . . . . . . 7.0 Inorganic Salts mg/ml NaCl . . . . . . . . . . . . . . . . . . . . 6400.0 KCL . . . . . . . . . . . . . . . . . . . . 400.0 NaHzPO"2HZO................125.0 MgSO{7HgJ . . . . . . . . . . . . . . . . . 200.0 Cac12 . . . . . . . . . . . . . . . . . . . 200.0 NaHCO3 . . . . . . . . . . . . . . . . . . . 3700.0 Fe(NO3)3'9HZO................0.l Other Components n-Butyl-p-hydroxybenzoate . . . . . . . . . 0.2 Glucose . . . . . . . . . . . . . . . . . . 4500.0 Penicillin G potassium . . . . . . . . . . . 5000 units Phenol red . . . . . . . . . . . . . . . . . 15.0 Sodium pyruvate . . . . . . . . . . . . . . 110.0 Streptomycin sulfate, equivalent base . . . 100.0 APPENDIX C L-Amino Acids mg/ml Arginine . . . . . . . . . . . . . . . . . . 200.0 Asparagine . . . . . . . . . . . . . . . . . 50.0 Aspartic acid . . . . . . . . . . . . . . . 20.0 Cystine . . . . . . . . . . . . . . . . . . 50.0 Glutamic acid . . . . . . . . . . . . . . . 20.0 Glutamine . . . . . . . . . . . . . . . . . 300.0 Glycine . . . . . . . . . . . . . . . . . . 10.0 Histidine . . . . . . . . . . . . . . . . . 15.0 Hydroxyproline . . . . . . . . . . . . . . . 20.0 Isoleucine (allo-free) . . . . . . . . . . . 50.0 Leucine (methionine—free) . . . . . . . . . 50.0 Lysine HCl . . . . . . . . . . . . . . . . . 40.0 Methionine . . . . . . . . . . . . . . . . . 15.0 Phenylalanine . . . . . . . . . . . . . . . 15.0 Proline (hydroxy-L-proline-free) . . . . . . 20.0 Serine . . . . . . . . . . . . . . . . . . . 30.0 Threonine (allo-free) . . . . . . . . . . . 20.0 Tryptophan . . . . . . . . . . . . . . . . . 5.0 Tyrosine . . . . . . . . . . . . . . . . . . 20.0 Valine . . . . . . . . . . . . . . . . . . . 20.0 Vitamins 92 Biotin . . . Vitamin B12 Calcium pantothenate Vitamins (continued) Choline Chloride Folic acid . i-Inositol . Nicotinamide . p-Aminobenzoic Pyridoxine HCl Riboflavin . Thiamine HCl Inorganic Salts NaCl . . . . KCl . . . . NaZHP0,-7H20 Mgso,-7H20 . Ncho3 . . . Ca (N03) 2-4H20 other Components Glucose . . Glutathione (reduced) Phenol red . 93 0.2 0.005 0.25 Ins/m1 3.0 1.0 35.0 1.0 1.0 6000.0 400.0 1512.0 100.0 2000.0 100.0 2000.0 1.0 5.0 APPENDIX D Publications 1. Lindsey, HJ, Kisala, JM, Ayala, A, Lehman, D, Herdon, CD, and Chaudry, IH. Pentoxifylline attenuates oxygen-induced lung injury. J Surg Res 1994;56:543-548. 2. Lindsey, HJ, Kisala, JM, Ayala, A, and Chaudry, IH. Alveolar macrophages exposed to hyperoxia demonstrate early enhanced cytokine productive capacity. W 1992;43:66- 68. Presentations and Awards We]. 1. "Pentoxifylline attenuates oxygen-induced lung injury" presented November 11, 1993 at the 27th.annual meeting of the Association for Academic Surgery, Hershey, PA. 2. "Alveolar macrophages exposed to hyperoxia demonstrate early enhanced cytokine productive capacity" presented October 13, 1992 at the Surgical Forum section of the 78th Clinical Congress of the American College of Surgeons, New Orleans, LA. 94 95 m3; 1. "Pentoxifylline attenuates oxygen-induced lung injury” presented May 6, 1993 at the 42nd.annual Coller Day of the Michigan Chapter, American College of Surgeons, Grand Rapids, MI. 2. "Hyperoxia enhances interleukin-1 production by alveolar“macrophages” presented May 2, 1993 at the 40th.annual Coller Day of the Michigan Chapter, American College of Surgeons, East Lansing, MI. Legal 1. "Pentoxifylline attenuates oxygen-induced lung injury" presented.March 11, 1993 at the Ninth.Annual Research Day Forum of the Michigan State University Department of Surgery, East Lansing, MI, and won first place in the basic science competition. 2. "Alveolar macrophages exposed to hyperoxia in_yiyg demonstrate enhanced TNF productive capacity" presented May 10, 1992 at Flint Academy of Surgery, Flint, MI, and won second place in resident research competition. 3. "Alveolar macrophages exposed to hyperoxia in yivo demonstrate enhanced TNF productive capacity" presented.March 12, 1992 at the Eighth Annual Research Day Forum of the Michigan State University Department of Surgery, East Lansing, MI, and won first place in the basic science competition. 96 4. "Hyperoxia enhances interleukin-1 production by alveolar macrophages" presented May 14, 1991 at the Flint .Academy of Surgery, Flint, MI, and.won first.place in resident research competition. 5. ”Hyperoxia enhances interleukin-1 production by alveolar macrophages" presented May 2, 1991 at the Seventh Annual Research Day Forum of the Michigan State University Department of Surgery, East Lansing, MI. BIBLIOGRAPHY 97 1. Lodato RF. Oxygen toxicity. Crit Care Clinics 1990:6:749-765. 2. Barber RE, Lee J, Hamilton WK. Oxygen toxicity in man. A prospective study in patients with irreversible brain damage. NEJM 1970;283:1478-1484. 3. Sackner MA, Landa J, Hirsch J, Zapata A. Pulmonary effects of oxygen breathing: A 6-hour study in normal men. Ann Int Med 1975;82:40-43. 4. Fox RB, Shasby DM, Harada RN, Repine JE. A novel mechanism for pulmonary oxygen toxicity: Phagocyte mediated lung injury. Chest 1981;80:Supp1:3S-4S. 5. Bowles AL, Dauber JH, Daniele RP. The effect of hyperoxia on migration of alveolar macrophages 1n_yitrg. Am Rev Respir Dis 1979;120:541-545. 6. Jackson RM. Molecular, pharmacologic, and clinical aspects of oxygen-induced lung injury. Clinics in Chest Medicine 1990;11:73-86. 7. Pratt PC, Vollmer RT, Shelburne JD, Crapo JD. Pulmonary morphology in a multihospital collaborative extracorporeal membrane oxygenation project. Am J Pathol 1979;95:191-208. 98 8. Ingrassia TS,III, Ryu JH, Trastek VF, Rosenow EC,III. Oxygen-exacerbated bleomycin pulmonary toxicity. Mayo Clin Proc 1991;66:173-178. 9. Fanburg BL, Deneke SM. Hyperoxia: Toxicity and Adaption. In: Massaro D, ed. Lung Cell Biology. New York, NY: Marcel Dekker, Inc., 1989:1199-1226. lO. Berend N. The effect of bleomycin and oxygen on rat lung. Pathology 1984;16:136-139. 11. Hartman LC, Frytak S, Richardson RL, Coles DT, Cupps RE. Life-threatening bleomycin pulmonary toxicity with ultimate reversibility. Chest 1990;98:497-499. 12. Rinaldo J, Goldstein RH, Snider GL. Modification of oxygen toxicity after lung injury by bleomycin in hamsters. Am Rev Respir Dis 1982;126:1030-1033. 13. Haschek WM, Witschi H. Pulmonary fibrosis-a possible mechanism. Toxicol Appl Pharmacol 1979;51:475-487. 14. Katzenstein AA, Bloor CM, Leibow AA. Diffuse alveolar damage-the role of oxygen, shock, and related factors. Am J Pathol 1976;85:210-222. 99 15. Beutler B, Cerami A. Cachectin: More than a tumor necrosis factor. NEJM 1987;316:379-385. 16. Tracey KJ, Beutler B, Lowry SF, et al. Shock and tissue injury induced by recombinant human cachectin. Science 1986;234:470-474. 17. Starnes HF,Jr., Warren RS, Jeevanandam M, et al. Tumor necrosis factor and the acute metabolic response to tissue injury in man. J Clin Invest 1988;82:1321-1325. 18. Punch J, Rees R, Cashmer B, Oldham K, Wilkins E, Smith DJ,Jr.. Acute lung injury following reperfusion after ischemia in the hind limbs of rats. J Trauma 1991;31:760-767. 19. Guice KS, Oldham KT, Caty MG, Johnson KJ, Ward PA. Neutrophil-dependent, oxygen-radical mediated lung injury associated with acute pancreatitis. Ann Surg 1989;210:740-747. 20. Till GO, Hatherill JR, Tourtellotte WW, Lutz MJ, Ward PA. Lipid peroxidation and acute lung injury after thermal trauma to skin. Am J Pathol 1985;119:376-384. 100 21. Guice KS, Oldham KT, Johnson KJ, Kunkel RG, Morganroth ML, Ward PA. Pancreatitis-induced acute lung injury. Ann Surg 1988;208:71-77. 22. Till GO, Beauchamp C, Menapace D, et a1. Oxygen radical dependent lung damage following thermal injury of rat skin. J Trauma 1983;23:269-277. 23. Snider MT. Adult respiratory distress syndrome in the trauma patient. Crit Care Clinics 1990;6:103-110. 24. Crapo JD. Morphologic changes in pulmonary oxygen toxicity. Ann Rev Physiol 1986;48:721-731. 25. Crapo JD, Barry BE, Foscue HA, Shelburne J. Structural and biochemical changes in rat lungs occurring during exposures to lethal and adaptive doses of oxygen. Am Rev Respir Dis 1980;122:123-143. 26. Royston BD, Webster NR, Nunn JF. Time course of changes in lung permeability and edema in the rat exposed to 100% oxygen. J Appl Physiol 1990;69:1532-1537. 27. Porte A, Mantz J, Hindelang C, Asmarats L, Viragh S, Stoeckel ME. Early bronchopulmonary lesions in rat lung after 100% oxygen exposure and their evolution. Virchows Archiv A Pathol Anat 1989;414:135-145. 101 28. Koizumi M, Frank L, Massaro D. Oxygen toxicity in rats. Am Rev Respir Dis 1985;131:907-911. 29. Turrens JF, Crapo JD, Freeman BA. Protection against oxygen toxicity by intravenous injection of liposome-entrapped catalase and superoxide dismutase. J Clin Invest 1984;73:87-95. 30. Frank L. Endotoxin-tolerant rats are still protected from oxygen toxicity by low-dose endotoxin treatment. J Appl Physiol 1985;58:819-822. 31. White CW, Ghezzi P. Protection against pulmonary oxygen toxicity by interleukin-1 and tumor necrosis factor: Role of antioxidant enzymes and the effect of cyclooxygenase inhibitors. Biotherapy 1989;1:361-367. 32. Bowman CM, Butler EN, Repine JE. Hyperoxia damages cultured endothelial cells causing increased neutrophil adherence. Am Rev Respir Dis 1983;128:469-472. 33. Rinaldo JE, English D, Levine J, Stiller R, Henson J. Increased intrapulmonary retention of radiolabeled neutrophils in early oxygen toxicity. Am Rev Respir Dis 1988;137:345-352. 102 34. Verghese M, Snyderman R. Signal transduction and intracellular messengers. In: Zembala M, Asherson GL, eds. Human Monocytes. San Diego: Academic Press, 1989:101-112. 35. Laughlin MJ, Wild L, Nickerson PA, Matalon S. Effects of hyperoxia on alveolar permeability of neutropenic rabbits. J Appl Physiol 1986;61:1126-1131. 36. Raj JU, Hazinski TA, Bland RD. Oxygen-induced lung microvascular injury in neutropenic rabbits and lambs. J Appl Physiol 1985;58:921-927. 37. Krieger BP, Loomis WH, Czer GT, Spragg RG. Mechanisms of interaction between oxygen and granulocytes in hyperoxic lung injury. J Appl Physiol 1985;58:1326-1330. 38. Barry BE, Crapo JD. Patterns of accumulation of platelets and neutrophils in rat lungs during exposure to 100% and 85% oxygen. Am Rev Respir Dis 1985;132:548-555. 39. Jamieson D. Oxygen toxicity and reactive oxygen metabolites in mammals. Free Radical Biology & Medicine 1989;7:87-108. 103 40. Turrens JF, Freeman BA, Levitt JG, Crapo JD. The effect of hyperoxia on superoxide production by lung submitochondrial particles. Arch Biochem Biophys 1982;217:401-410. 41. Turrens JF, Freeman BA, Crapo JD. Hyperoxia increases H55 release by lung mitochondria and microsomes. Arch Biochem Biophys 1982;217:411-421. 42. Freeman BA, Crapo JD. Hyperoxia increases oxygen radical production in rat lungs and lung mitochondria. J Biol Chem 1981;256:10986-10992. 43. Jamieson D. The relation of free radical production to hyperoxia. Ann Rev Physiol 1986;48:703-719. 44. Deneke SM, Fanburg BL. Normobaric oxygen toxicity of the lung. NEJM 1980;303:76-86. 45. Frank L, Massaro D. Oxygen toxicity. Am J Med 1980;69:117-126. 46. McCord JM. Reviews in Biochemical Toxicology. Amsterdam: Elsevier, 1979:109-124. 104 47. Shasby DM, Fox RB, Harada RN, Repine JE. Reduction of the edema of acute hyperoxic lung injury by granulocyte depletion. J Appl Physiol 1982;52:1237-1244. 48. Steinberg H, Das DK, Cerreta JM, Cantor JD. Neutrophil kinetics in OZ-exposed rabbits. J Appl Physiol 1986;61:775-779. 49. Boyce NW, Campbell D, Holdsworth SR. Granulocyte independence of pulmonary oxygen toxicity. Exp Lung Res 1989;15:491-498. 50. Johnson KJ, Fantone JC,III, Kaplan J, Ward PA. In vivo damage of rat lungs by oxygen metabolites. J Clin Invest 1981;67:983-993. 51. Durum SK, Oppenheim JJ. Macrophage-Derived Mediators: Interleukin 1, Tumor Necrosis Factor, Interleukin 6, Interferon, and Related Cytokines. In: Paul WE, ed. Fundamental Immunology. 2nd ed. New York, NY: Raven Press Ltd., 1989:639-661. 52. Golub ES, Green DR. Immunology: A Synthesis. 2nd ed. Sunderland, MA: Sinauer Associates, Inc., 1991: 105 53. Philip R, Epstein LB. Tumour necrosis factor as immunomodulator and mediator of monocyte cytotoxicity induced by itself, gamma-interferon and interleukin-1. Nature 1986;323:86-89. 54. Schleimer RP, Benenati SV, Friedman B, Bochner BS. Do cytokines play a role in leukocyte recruitment and activation in the lungs? Am Rev Respir Dis 1991;143:1169-1174. 55. Nohynek H, Teppo A, Laine E, Leinonen M, Eskola J. Serum tumor necrosis factor-a concentrations in children hospitalized for acute lower respiratory tract infection. J Infect Dis 1991;163:1029-1032. 56. Fukushima R, Alexander JW, Gianotti L, Ogle C. Cytokine production in the lung: lung as a major source of systemic TNF. Presented at the 1992 annual meeting of the Society of University Surgeons. 1992 (unpublished). 57. Warren JS, Yabroff KR, Remick DG, et al. Tumor necrosis factor participates in the pathogenesis of acute immune complex alveolitis in the rat. J Clin Invest 1989;84:1873-1882. 106 58. Simpson SQ, Modi HN, Balk RA, Bone RC, Casey LC. Reduced alveolar macrophage production of tumor necrosis factor during sepsis in mice and men. Crit Care Med 1991;19:1060-1066. 59. Hyers TM, Tricomi SM, Dettenmeier PA, Fowler AA. Tumor necrosis factor levels in serum and bronchoalveolar lavage fluid of patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1991;144:268-271. 60. Chaudhri G, Clark IA. Reactive oxygen species facilitate the in vitro and in vivo lipopolysaccharide-induced release of tumor necrosis factor. J Immunol 1989;143:1290-1294. 61. Visner GA, Dougall WC, Wilson JM, Burr IA, Nicks HS. Regulation of superoxide dismutase by lipopolysaccharide, interleukin-1, and tumor necrosis factor. J Biol Chem 1990;265:2856-2864. 62. Asayama K, Janco RL, Burr IM. Selective induction of manganous superoxide dismutase in human monocytes. Am J Physiol 1985;249:C393-C397. 63. Shiki Y, Meyrick BO, Brigham KL, Burr IM. Endotoxin increases superoxide dismutase in cultured bovine pulmonary endothelial cells. Am J Physiol 1987;252:C436-C440. 107 64. Warren JS, Kunkel SL, Cunningham TW, Johnson KJ, Ward PA. Macrophage-derived cytokines amplify immune complex-triggered 02 responses by rat alveolar macrophages. Am J Pathol 1988;130:489-495. 65. White CW, Ghezzi P, Dinarello CA, Caldwell SA, McMurtry IF, Repine JE. Recombinant tumor necrosis factor/cachectin and interleukin 1 pretreatment decreases lung oxidized glutathione accumulation, lung injury, and mortality in rats exposed to hyperoxia. J Clin Invest 1987;79:1868-1873. 66. White CA, Ghezzi P, McMahon S, Dinarello CA, Repine JE. Cytokines increase rat lung antioxidant enzymes during exposure to hyperoxia. J Appl Physiol 1989;66:1003-1007. 67. Tsan M, White JE, Santana TA, Lee CY. Tracheal insufflation of tumor necrosis factor protects rats against oxygen toxicity. J Appl Physiol 1990;88:1211-1219. 68. Berg JT, Allison RC, Prasad VR, Taylor AE. Endotoxin protection of rats from pulmonary oxygen toxicity: possible cytokine involvement. J Appl Physiol 1990;68:549-553. 69. Dinarello CA. Biology of interleukin-1. FASEB J 1988 ; 2 : 108-115 . 108 70. Dinarello CA, Mier JW. Lymphokines. NEJM 1987;317:940-945. 71. Okusawa S, Gelfand JA, Ikejima T, Connolly RJ, Dinarello CA. Interleukin 1 induces a shock-like state in rabbits. J Clin Invest 1988;81:1162-1172. 72. Wakabayashi G, Gelfand JA, Burke JF, Thompson RC, Dinarello CA. A specific receptor antagonist for interleukin 1 prevents Egghe;ighig_ggli—induced shock in rabbits. FASEB J 1991;5:338-343. 73. Eichacker PQ, Hoffman WD, Farese A, et al. TNF but not IL-1 in dogs causes lethal lung injury and multiple organ dysfunction similar to human sepsis. J Appl Physiol 1991;71:1979-1989. 74. Wankowicz z, Megyeri P, Issekutz A. Synergy between tumour necrosis factora and interleukin-1 in the induction of polymorphonuclear leukocyte migration during inflammation. J Leukocyte Biol 1988;43:349-356. 75. Suwabe A, Takahashi K, Yashui S, Arai S, Sendo F. Bleomycin-stimulated hamster alveolar macrophages release interleukin-1. Am J Pathol 1988;132:512-520. 109 76. Kishimoto T. The biology of interleukin-6. Blood 1989;74:1-10. 77. Helle M, Brakenhoff JPJ, DeGroot ER, Aarden LA. Interleukin 6 is involved in interleukin 1-induced activities. Eur J Immunol 1988;18:957-959. 78. Sironi M, Brevario F, Proserpio P, et al. IL-l stimulates IL-6 production in endothelial cells. J Immunol 1989;142:549. 79. Kohase M, May LT, Tamm I, Vilcek J, Sehgal PB. A cytokine network in human diploid fibroblasts: Interaction of fi-interferons, tumor necrosis factor, platelet derived growth factor, and interleukin-1. Mol Cell Biol 1987;7:273. 80. Tosato G, Jones KD. Interleukin-1 induces interleukin-6 production in peripheral blood monocytes. Blood 1990;75:1305-1310. 81. Yoshida Y, Iwaki Y, Pham S, et al. Benefits of posttransplantation monitoring of interleukin 6 in lung transplantation. Ann Thorac Surg 1993;55:89-93. 110 82. Jacobson JM, Michael JR, Jafri MH,Jr., Gurtner GH. Antioxidants and antioxidant enzymes protect against pulmonary oxygen toxicity in the rabbit. J Appl Physiol 1990;68:1252-1259. 83. Hazinski TA, France M, Kennedy KA, Hansen TN. cimetidine reduces hyperoxic lung injury in lambs. J Appl Physiol 1989;67:2586-2592. 84. Ohtsu N, Ariagno RL, Sweeney TE, et al. The effect of dexamethasone on chronic pulmonary oxygen toxicity in infant mice. Pediatr Res 1989;25:353-359. 85. Sahebjami H, Gacad G, Massaro D. Influence of corticosteroids on recovery from oxygen toxicity. Am Rev Respir Dis 1974;110:566-571. 86. Berg JT, Allison RC, Taylor AE. Endotoxin extends survival of adult mice in hyperoxia. Proc Soc Exp Biol Med 1990;193:167-170. 87. Berg JT, Smith RM. Protection against hyperoxia by serum from endotoxin treated rats: absence of superoxide dismutase induction. Proc Soc Exp Biol Med 1988;187:117-122. 111 88. Taniguchi H, Taki F, Takagi K, Satake T, Sugiyama S, Ozawa T. The role of leukotriene B4 in the genesis of oxygen toxicity in the lung. Am Rev Respir Dis 1986;133:805-808. 89. Yam J, Frank L, Roberts RJ. Oxygen toxicity: Comparison of lung biochemical responses in neonatal and adult rats. Pediat Res 1978;12:115-119. 90. Ayala A, Perrin MM, Kisala JK, Ertel W, Chaudry IH. Polymicrobial sepsis selectively activates peritoneal but not alveolar macrophage to release inflammatory mediators (IL-1, IL-6, and TNF). Circ Shock 1992;36:191-199. 91. Hogan M, Perera PY, Vogel SN. Examination of macrophage cell surface antigen regulation by rIFN-gamma and IFN-alpha/beta utilizing digital imaging by a novel laser detection system. Anchored cell analysis station (ACAS) 470. J Immunol Methods 1989;123:9. 92. Espevik T, Nissen-Meyer J. A highly sensitive cell line, WEHI 164 clone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes. J Immunol Methods 1986;95:99-105. 112 93. Kaye J, Gillis S, Mizel SB, et al. Growth of a cloned helper T cell line induce by a monoclonal antibody specific for the antigen receptor: interleukin 1 is required for the expression of receptors for interleukin 2. J Immunol 1984;133:1339. 94. Van Snick J, Cayphas J, Vink A, et al. Purification and NH2-terminal amino acid sequence of a T-cell-derived lymphokine with growth factor activity for B-cell hybridomas. Proc Natl Acad Sci USA 1989;83:9679-9683. 95. Ayala A, Perrin MM, Meldrum DR, Ertel W, Chaudry IH. Hemorrhage induces an increase in serum TNF which is not associated with elevated levels of endotoxin. Cytokine 1990;2:170. 96. Eskandari MK, Nguyen DT, Kunkel SL, Remick DG. WEHI 164 sublcone 13 assay for TNF: sensitivity, specificity, and reliability. Immunol Invest 1990;19:69-79. 97. Ayala A, Perrin MM, Wagner MA, Chaudry IH. Enhanced susceptibility to sepsis following simple hemorrhage: depression of Fe and C3b receptor mediated phagocytosis. Arch Surg 1990;125:70. 113 98. Mizel SB. Production and quantitation of lymphocyte-activating factor (interleukin 1). In: Herscowitz HB, Holden HT, Bellanti JA, Ghaffar A, eds. Manual of Macrophage Methodology. New York: Marcel Dekker,Inc., 1981:407-441. 99. Hultner L, Szots M, Welle M, Moeller J, Dormeu P. Mouse bone marrow-derived interleukin-3-dependent mast cells and autonomous sublines produce interleukin-6. Immunology 1989;67:408-413. 100. Mauderly JL. Bronchopulmonary lavage of small laboratory animals. Lab Animal Sci 1977;27:255-261. 101. Roth RA. Effect of pneumotoxicants on lactate dehydrogenase activity in airways of rats. Toxicol Appl Pharmacol 1981;57:69-78. 102. Henderson RF, Damon EG, Henderson TR. Early damage indicators in the lung. I. Lactate dehydrogenase activity in the airways. Toxicol Appl Pharmacol 1978;44:291-297. 103. Henderson RF, Muggenburg BA, Mauderly JL, Tuttle WA. Early damage indicators in the lung. II. Time sequence of protein accumulation and lipid loss in the airway of Beagle dogs with beta-irradiation of the lung. Radiat Res 1978;76:145-158. 114 104. Bergmeyer HU, Bernt E. Methods of Enzymatic Analysis. New York: Academic Press, 1974:574-579. 105. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-275. 106. Doherty GM, Jensen JC, Alexander HR, Buresh CM, Norton JA. Pentoxifylline suppression of tumor necrosis factor gene transcription. Surgery 1991;110:192-198. 107. Williams JH,Jr., Heshmati S, Tamadon S, Guerra J. Inhibition of alveolar macrophages by pentoxifylline. Crit Care Med 1991;19:1073-1078. 108. Schade UF. Pentoxifylline increases survival in murine endotoxin shock and decreases formation of tumor necrosis factor. Circ Shock 1990;31:171-181. 109. Schade UP. The role of prostacyclin in the protective effects of pentoxifylline and other xanthine derivatives in endotoxin action in mice. Eicosanoids l989;2:183-188. 110. Lilly CM, Sandhu JS, Ishizaka A, et al. Pentoxifylline prevents tumor necrosis factor-induced lung injury. Am Rev Respir Dis 1989;139:1361-1368. 115 111. Mandell GL. ARDS, neutrophils, and pentoxifylline. Am Rev Respir Dis 1988;138:1103-1105. 112. Wang P, Ba ZF, Morrison MH, Ayala A, Chaudry IH. Mechanism of the beneficial effects of pentoxifylline on hepatocellular function following trauma-hemorrhage and resuscitation. Surgery 1992;112:451-458. 113. Welsh CH, Lien D, Worthen GS, Weil JV. Pentoxifylline decreases endotoxin-induced pulmonary neutrophil sequestration and extravascular protein accumulation in the dog. Am Rev Respir Dis 1988;138:1106-1114. 114. Noel P, Nelson S, Bokulic R, et al. Pentoxifylline inhibits lipopolysaccharide-induced serum tumor necrosis factor and mortality. Life Sci 1990;47:1023-1029. 115. Kennedy TP. Pulmonary Oz toxicity: Validation of the "60%" rule in cultured human lung endothelium. J Lab Clin Med 1989;113:399-400. 116. Rinaldo JE, Rogers RM. Adult respiratory-distress syndrome. NEJM 1982;306:900-909. 116 117. Hazinski TA, Kennedy KA, France ML, Hansen TN. Pulmonary O2 toxicity in lambs: physiological and biochemical effects of endotoxin infusion. J Appl Physiol 1988;65:1579-1585. 118. Ulich TR, Yin S, Guo K, Yi ES, Remick D, Del Castillo J. Intratracheal injection of endotoxin and cytokines: II. Interleukin-6 and transforming growth factor beta inhibit acute inflammation. Am J Pathol 1991;138:1097-1101. 119. Chaudry IH, Ayala A. Immunological Aspects of Hemorrhage. Austin, TX: Medical Intelligence Unit; R.G. Landes Company, 1992:1-132. 120. Kisala JM, Ayala A, Stephan RN, Chaudry IH. A model of pulmonary atelectasis in rats: activation of alveolar macrophage and cytokine release. Am J Physiol 1993; 121. Tipping PG, Campbell DA, Boyce NW, Holdsworth SR. Alveolar macrophage procoagulant activity is increased in acute hyperoxic lung injury. Am J Pathol 1988;131:206-212. 122. Garner WL, Downs JB, Reilly TE, Frolicher D, Kargi A, Fabri PJ. The effects of hyperoxia during fulminant sepsis. Surgery 1989;105:747-751. 117 123. Ayala A, Perrin MM, Wang P, Ertel W, Chaudry IH. Hemorrhage induces enhanced Kupffer cell cytotoxicity while decreasing peritoneal or splenic macrophage capacity. J Immunol 1991;147:4147-4154. 124. Warren JS. Intrapulmonary interleukin 1 mediates acute immune complex alveolitis in the rat. Biochem Biophys Res Comm 1991;175:604-610. 125. Lindegarde F, Jelnes R, Bjorkman H. Conservative drug treatment in patients with moderately severe chronic occlusive peripheral arterial disease. Circulation 1989;80:1549. 126. Physicians’ Desk Reference. 47th ed. Montvale, NJ: Medical Economics Data, 1993:1125-1126. 127. Ishizaka A, Wu 2, Stephens KE, et al. Attenuation of acute lung injury in septic guinea pigs by pentoxifylline. Am Rev Respir Dis 1988;138:376-382. 128. Yonemaru M, Hatherill JR, Hoffmann H, Zheng H, Ishii K, Raffin TA. Pentoxifylline does not attenuate acute lung injury in the absence of granulocytes. J Appl Physiol 1991;71:342-351. 118 129. McDonald RJ. Pentoxifylline reduces injury to isolated lungs perfused with human neutrophils. Am Rev Respir Dis 1991;144:1347-1350. 130. Bertocchi F, Prosperio P, Lampugnami MC, Dejana E. Pentoxifylline and leukocyte function. Weston, CT: Haber and Flora, 1988:68-74. 131. Bessler H, Gilgal R, Djaldetti M, Zahavi I. Effect of pentoxifylline on the phagocytic activity, c-AMP levels and superoxide anion production by monocytes and polymorphonuclear cells. J Leukocyte Biol 1986;40:747-754. 132. Zabel P, Wolter DT, Schonharting MM, Schade UF. Opentoxifylline in endotoxaemia. Lancet l989;2:1474-1477. 133. Han J, Thompson P, Beutler B. Dexamethasone and pentoxifylline inhibit endotoxin-induced cachectin/tumor necrosis factor synthesis at separate points in the signaling pathway. J Exp Med 1990;172:391-394. 134. Seear MD, Hannam VL, Kaapa P, Raj JU, O’Brodovich HM. Effect of pentoxifylline on hemodynamics, alveolar fluid reabsorption, and pulmonary edema in a model of acute lung injury. Am Rev Respir Dis 1990;142:1083-1087. 119 135. Hoffmann H, Hatherill JR, Crowley J, et al. Early post-treatment with pentoxifylline or dibutyryl cAMP attenuates Eggnerignia_§gli-induced acute lung injury in guinea pigs. Am Rev Respir Dis 1991;143:289-293. 136. Chalkiadakis GE, Kostakis A, Karayannacos PE, et al. Pentoxifylline in the treatment of experimental peritonitis in rats. Arch Surg 1985;120:1141-1144. 137. White CW, Jackson JH, Abuchowski A, et al. Polyethylene glycol-attached antioxidant enzymes decrease pulmonary oxygen toxicity in rats. J Appl Physiol 1989;66:584-590. 138. Eckenhoff RG, Dodia C, Tan 2, Fisher AB. Oxygen-dependent reperfusion injury in the isolated rat lung. J Appl Physiol 1992;72:1454-1460. 139. Roberts JD,Jr., Oh W. Pulmonary oxygen toxicity in the guinea pig. Dev Pharmacol Ther 1989;12:106-112. 140. Borish L, Rosenbaum R, Albury L, Clark S. Activation of neutrophils by recombinant interleukin 6. Cell Immunol 1989;121:280-289. 120 141. Goldman G, Welbourn R, Kobzik L, Valeri CR, Shepro D, Hechtman HB. Lavage with leukotriene B, induces lung generation of tumor necrosis factor-a that in turn mediates neutrophil diapedesis. Surgery 1993;113:297-303. 142. Pohlman TH, Stanness KA, Beatty PG, Ochs HD, Harlan JM. An endothelial cell surface factor(s) induced in vitro by lipopolysaccharide, interleukin 1, and tumor necrosis factor-a increases neutrophil adherence by a CleB-dependent mechanism. J Immunol 1986;136:4548-4553. 143. Jacobs RF, Tabor DR, Burks AW, Campbell GD. Elevated interleukin-1 release by human alveolar macrophages during the adult respiratory distress syndrome. Am Rev Respir Dis 1989;140:1686-1692. 144. Roten R, Markert M, Feihl F, Schaller M, Tagan M, Perret C. Plasma levels of tumor necrosis factor in the adult respiratory distress syndrome. Am Rev Respir Dis 1991;143:590-592. 145. Fong Y, Moldawer LL, Marano M, et al. Endotoxemia elicits increased circulating fii-IFN/IL-6 in man. J Immunol 1989;142:2321-2324. 121 146. Schirmer WJ, Schirmer JM, Fry DE. Recombinant human tumor necrosis factor produces hemodynamic changes characteristic of sepsis and endotoxemia. Arch Surg 1989;124:445-448. 147. Waage A, Brandtzaeg P, Halstensen A, Kierulf P, Espevik T. The complex pattern of cytokines in serum from patients with meningococcal septic shock. Association between interleukin 6, interleukin 1, and fatal outcome. J Exp Med 1989;169:333-338. 148. Tracey KJ. Tumor necrosis factor (cachectin) in the biology of septic shock syndrome. Circ Shock 1991;35:123-128. 149. Calandra T, Gerain J, Heumann D, Baumgartner JD, Glauser MP. High circulating levels of interleukin-6 in patients with septic shock: evolution during'sepsis, prognostic value, and interplay with other cytokines. Am J Med 1991;91:23-29. 150. Ertel W, Morrison MH, Ayala A, Perrin MM, Chaudry IH. Anti-TNF monoclonal antibodies prevent haemorrhage induced suppression of Kupffer cell antigen presentation and MHC class II antigen expression. Immunology 1991;74:290-297. 122 151. Lindsey HJ, Kisala JM, Ayala A, Chaudry IH. Alveolar macrophages exposed to hyperoxia demonstrate early enhanced cytokine productive capacity. Surg Forum 1992;43:66-68. 152. Jensen JC, Pogrebniak HW, Pass HI, et al. Role of tumor necrosis factor in oxygen toxicity. J Appl Physiol 1992;72:1902-1907.