Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Expression of leukocyte-endothelial adhesion molecules during acute inflammation in the lung Burns, Alan R. 1993

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

Item Metadata

Download

Media
831-ubc_1993_spring_phd_burns_alan.pdf [ 15.68MB ]
Metadata
JSON: 831-1.0086351.json
JSON-LD: 831-1.0086351-ld.json
RDF/XML (Pretty): 831-1.0086351-rdf.xml
RDF/JSON: 831-1.0086351-rdf.json
Turtle: 831-1.0086351-turtle.txt
N-Triples: 831-1.0086351-rdf-ntriples.txt
Original Record: 831-1.0086351-source.json
Full Text
831-1.0086351-fulltext.txt
Citation
831-1.0086351.ris

Full Text

THE UNIVERSITY OF BRITISH COLUMBIEXPRESSION OF LEUKOCYTE-ENDOTHELIAL ADHESION MOLECULESDURING ACUTE INFLAMMATION IN THE LUNGbyALAN ROBERT BURNSB.Sc. The University of British Columbia, 1978M.Sc. The University of British Columbia, 1981A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Pathology)We accept this thesis as conformingJanuary 1993© Alan Robert Burns, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of 1/4 7/16 lo The University of British ColumbiaVancouver, CanadaDateDE-6 (2/88)ABSTRACTNeutrophil (PMN)-endothelial cell adhesion is a critical step in the response of PMNsto inflammation. In the systemic circulation, the leukocyte adhesion molecule, L-selectin,facilitates PMN adhesion to inflamed endothelium while CD11/CD18 is required for PMNemigration into extravascular tissues. An inducible endothelial ligand for CD18 is intercellularadhesion molecule-1 (ICAM-1). In the pulmonary circulation, PMNs emigrate by either aCD 18-independent or CD 18-dependent mechanism. The objective of this thesis was toquantitate and compare the surface expression of L-selectin, CD 18, and ICAM-1 during CD18-independent and CD 18-dependent emigration. Rabbits and mice received airway instillates ofStreptococcus pneumoniae or Escherichia coil endotoxin to induce CD 18-independent or CD18-dependent emigration, respectively. Ultrathin cryosections of frozen lung were immunogoldlabeled for L-selectin, CD 18, and ICAM-1. Gold particles on the plasma membranes werequantitated by transmission electron microscopy. In rabbits, CD 18-independent emigration wasassociated with L-selectin downmodulation and CD18 upmodulation on intravascular PMNs.A similar alteration of L-selectin and CD18 expression was observed during CD 18-dependentemigration but only after PMNs emigrated into the interstitium. Alterations in L-selectin andCD18 expression were only observed on PMNs within the inflammatory focus. In mice,capillary endothelial ICAM-1 expression was unchanged during CD 18-independent emigration.During CD 18-dependent emigration, ICAM-1 expression increased 4.2-fold and this increasebordered on statistical significance, suggesting that the mechanism of adhesion may beregulated by the expression of endothelial rather than PMN adhesion molecules. ICAM-1 wasalso constitutively expressed on alveolar Type I but not Type II pneumocytes, the precursorsof Type I cells. During pneumonia, Type II but not Type I pneumocytes showed increasedICAM-1 expression, suggesting that ICAM-1 expression represents an early event indifferentiation preceding proliferation. In vitro studies of unstimulated human PMNs showedthat L-selectin was preferentially expressed on the PMN surface microvilli that mediate initialcontact with endothelium. During transendothelial migration, L-selectin downmodulation istemporally correlated with PMN-endothelial contact. These studies describe the ultrastructurallocalization of adhesion molecules in normal and inflamed lungs and increase our understandingof the correlation between expression and function of adhesion molecules.ivTABLE OF CONTENTSPageAbstract^ iiTable of Contents^ ivList of Abbreviations viList of Tables viiiList of Figures^ ixAcknowledgements xivDedication xvCHAPTER ONE. INTRODUCTION1.1^Introduction: Overview ^  21.2^The history of leukocyte adhesion and emigration ^ 31.3^Leukocyte-endothelial adhesion molecules ^ 91.4^Summary and objectives  27CHAPTER TWO MATERIALS AND METHODS2.1^Sources of Materials ^  322.2 Rabbit pneumonia experiments ^  332.3 Mouse pneumonia experiments  352.4 Human leukocyte-rich plasma experiments ^ 372.5 Human PMN transendothelial migration experiments^382.6^Cryoultramicrotomy ^  412.7^Immunogold labeling of cryosections ^ 432.8 Immunoelectron microscopy  462.9^Statistical analysis  50CHAPTER THREE RESULTS3.1^Electron microscopic observations of PMNsin rabbit lung ^  583.2^Immunoelectron microscopy of PMNs in rabbit lung .^603.3^Light microscopy of mouse lung tissue ^ 633.4 Immunoelectron microscopy of mouse lung tissue ^ 653.5 Immunoelectron microscopy of human PMNs  67VCHAPTER FOUR DISCUSSION4.1^Expression of PMN, endothelial, and epithelial adhesionmolecules during pneumonia ^  714.2^Observations on L-selectin  864.3^Sensitivity and limitations of the immunogold technique . ^ 91CHAPTER FIVE CONCLUSIONS5.1^Summary and conclusions ^  955.2^Future prospects ^  995.3^Significance of this thesis  101TABLES  103FIGURES ^  109REFERENCES  179viLIST OF ABBREVIATIONSitg^Microgram(s)60.3 Anti-CD18 MAbANOVA^Analysis of varianceC3bi Split product of the third component of complementC5a^Split product of the fifth component of complementC5f Fragment of the fifth component of complementCD^Cluster of differentiation designationCD 11a/CD18^A leukocyte,82-integrinCD11b/CD18^A leukocyte 02-integrinCD11c/CD18^A leukocyte 02-integrinCHO^Chinese hamster ovary cell lineCOS Monkey kidney cell lineCSLEX-1^Anti-sLex MAbDREG-200^Anti-L-selectin MAbE. coil Escherichia coliE-selectin^Endothelial-selecting^Gram(s)h HoursHL60^Myeloid/monocytic cell lineHUVEC^Human umbilical vein endothelial cellsICAM-1^Intercellular adhesion molecule-1ICAM-2^Intercellular adhesion molecule-2IFN-y Interferon-gammaIL-1^Interleuldn-1IL-6 Interleukin-6viiIL-8^Interleukin-8kD Kilodaltonkg^Kilogram(s)LAD Leukocyte adhesion deficiencyLRP^Leukocyte-rich plasmaL-selectin^Leukocyte-selectinLTB4 Leukotriene B4MAb^Monoclonal antibodymg Milligram(s)min^MinutesNZW rabbit^New Zealand White rabbitN2^ NitrogenP-selectin^Platelet-selectin (also found on endothelial cells)PAF Platelet activating factorPBS^Phosphate buffered salinePBS-FCS^Phosphate buffered saline containing 5% fetal calf serumPMN Polymorphonuclear leukocyte (or neutrophil)SecondsS. pneumoniae^Streptococcus pneumoniaeSD^Standard DeviationSE Standard ErrorsLex^Sialylated Lewis X antigenTNF-a^Tumour necrosis factor-alphaYN1/1/7/4^Anti-ICAM-1 MAbZAP^Zymosan activated plasmaZAS Zymosan activated serumviiiLIST OF TABLESTable 1.^L-selectin study: Summary of morphometric data on rabbitneutrophils (PMNs) in control lungs and lungs instilled with aninflammatory stimulus (S. pneumoniae or E. coil endotoxin) andcolloidal carbon   103Table 2.^CD18 study: Summary of morphometric data on rabbit neutrophils(PMNs) in control lungs and lungs instilled with an inflammatorystimulus (S. pneumoniae or E. coil endotoxin) and colloidalcarbon   104Table 3.^L-selectin immunoreactivity on rabbit neutrophils (PMNs) incontrol lungs and lungs instilled with an inflammatory stimulus (S.pneumoniae or E. coil endotoxin) and colloidal carbon ^ 105Table 4.^CD18 immunoreactivity on rabbit neutrophils (PMNs) in controllungs and lungs instilled with an inflammatory stimulus (S.pneumoniae or E. coil endotoxin) and colloidal carbon ^ 106Table 5.^ICAM-1 study: Summary of morphometric data on control mouselungs and lungs instilled with colloidal carbon + an inflammatorystimulus (S. pneumoniae or E. coil endotoxin) ^  107Table 6.^ICAM-1 immunoreactivity in control mouse lungs and lungsinstilled with colloidal carbon + an inflammatory stimulus (S.pneumoniae or E. coil endotoxin) ^  108ixLIST OF FIGURESFigure 1:^Figure showing the leukocyte-endothelial adhesion molecules thathave been characterized at the molecular level and shown to beimportant for PMN adhesion and migration during acuteinflammation   109Figure 2:^Figure showing the "Specimen Pin Jig" specially constructed forcryosectioning ^  111Figure 3:^Electron micrograph of an intravascular PMN in a rabbit thatreceived a bronchial instillate of E. coil endotoxin mixed withcolloidal carbon ^  113Figure 4:^Electron micrograph of an elongated intravascular PMNin a control rabbit  113Figure 5:^Electron micrograph of an interstitial PMN in a rabbit thatreceived a bronchial instillate of E. coil endotoxin mixed withcolloidal carbon ^  115Figure 6:^Electron micrograph of an airspace PMN in a rabbit that receiveda bronchial instillate of S. pneumoniae mixed with colloidalcarbon ^  115Figure 7:^Electron micrograph of an airspace PMN in a rabbit that receiveda bronchial instillate of E. coil endotoxin mixed with colloidalcarbon ^  117Figure 8:^Electron micrograph of an airspace PMN in a rabbit that receiveda bronchial instillate of S. pneumoniae mixed with colloidalcarbon ^  117Figure 9:^Electron micrograph of a pneumonic lung in a rabbit that receiveda bronchial instillate of S. pneumoniae mixed with colloidalcarbon ^  119Figure 10:^Immunoelectron microscopic detection of L-selectin on aspherically-shaped intravascular PMN in control rabbit lung ^ 121Figure 11:^Immunoelectron microscopic detection of L-selectin on anelongated intravascular PMN in control rabbit lung ^ 121Figure 12:^Immunoelectron microscopic detection of CD18 on anintravascular PMN in control rabbit lung ^  123Figure 13:^Immunoelectron microscopic detection of L-selectin on anintravascular PMN in rabbit lung following a bronchial instillateof S. pneumoniae mixed with colloidal carbon ^  125Figure 14:Figure 15:Figure 16:Figure 17:Figure 18:Figure 19:Figure 20:Figure 21:Figure 22:Immunoelectron microscopic detection of L-selectin on anintravascular PMN in rabbit lung following a bronchial instillateof S. pneumoniae mixed with colloidal carbon   125Immunoelectron microscopic detection of L-selectin on anairspace PMN in rabbit lung following a bronchial instillate of S.pneumoniae mixed with colloidal carbon   127Immunoelectron microscopic detection of L-selectin on aninterstitial PMN in rabbit lung following a bronchial instillate ofS. pneumoniae mixed with colloidal carbon   129Immunoelectron microscopic detection of L-selectin on anintravascular PMN from the contralateral, non-pneumonic lung ofa rabbit that received a bronchial instillate of S. pneumoniaemixed with colloidal carbon   131Immunoelectron microscopic detection of L-selectin on anintravascular PMN from the contralateral, non-pneumonic lung ofa rabbit that received a bronchial instillate of E. coil endotoxinmixed with colloidal carbon   131Immunoelectron microscopic detection of L-selectin on aninterstitial PMN in rabbit lung following a bronchial instillate ofE. coil endotoxin mixed with colloidal carbon   133Immunoelectron microscopic detection of L-selectin on anairspace PMN in rabbit lung following a bronchial instillate of E.coil endotoxin mixed with colloidal carbon   133Immunoelectron microscopic detection of CD 18 on anintravascular PMN in rabbit lung following a bronchial instillateof S. pneumoniae mixed with colloidal carbon   135Immunoelectron microscopic detection of CD 18 on an interstitialPMN in rabbit lung following a bronchial instillate of S.pneumoniae mixed with colloidal carbon   137Figure 23:^Immunoelectron microscopic detection of CD18 on an airspacePMN in rabbit lung following a bronchial instillate of S.pneumoniae mixed with colloidal carbon ^  139xiFigure 24:Figure 25:Figure 26:Figure 27:Figure 28:Figure 29:Figure 30:Figure 31:Figure 32:Figure 33:Figure 34:Figure 35:Immunoelectron microscopic detection of CD18 on anintravascular PMN in the contralateral, non-pneumonic lung of arabbit that received a bronchial instillate of S. pneumoniae mixedwith colloidal carbon   141Immunoelectron microscopic detection of CD18 on an interstitialPMN in rabbit lung following a bronchial instillate of E. coilendotoxin mixed with colloidal carbon   143Immunoelectron microscopic detection of CD18 on an airspacePMN in rabbit lung following a bronchial instillate of E. conendotoxin mixed with colloidal carbon   145Light micrograph of mouse lung tissue 4h after the trachealinstillation of a saline solution of colloidal carbon ^ 147Light micrograph of alveolar macrophages in mouse lung tissue4h after the tracheal instillation of a saline solution of colloidalcarbon   147Light micrograph of mouse lung tissue 24h after the trachealinstillation of a saline solution of colloidal carbon ^ 149Light micrograph of alveolar macrophages in mouse lung tissue24h after the tracheal instillation of a saline solution of colloidalcarbon   149Light micrograph of mouse lung tissue 4h after the trachealinstillation of a S. pneumoniae mixed with colloidal carbon ^ 151Light micrograph of an alveolar PMN infiltrate in mouse lung 4hafter the tracheal instillation of a S. pneumoniae mixed withcolloidal carbon ^  151Light micrograph of mouse lung tissue 24h after the trachealinstillation of a E. coil endotoxin mixed with colloidal carbon ^ 153Light micrograph of an alveolar PMN infiltrate in mouse lungtissue 4h after the tracheal instillation of a E. coli endotoxinmixed with colloidal carbon   153Electron micrograph showing a cross-section of a pulmonarycapillary in normal mouse lung tissue ^  155Figure 36:^Immunoelectron microscopic detection of ICAM-1 on thepulmonary capillary endothelium in normal mouse lung ^ 155XIIFigure 37:^Immunoelectron microscopic detection of ICAM-1 on the alveolarepithelial cell surface in normal mouse lung ^  157Figure 38:^Electron micrograph contrasting the difference in ICAM-1immunoreactivity between a Type I and II pneumocyte in normalmouse lung ^  159Figure 39: Immunoelectron microscopic detection of ICAM-1 in mouse lung24h after a tracheal instillation of E. coil endotoxin mixed withcolloidal carbon ^  161Figure 40: Immunoelectron microscopic detection of ICAM-1 on a Type IIpneumocyte in mouse lung 4h after a tracheal instillation of S.pneumoniae mixed with colloidal carbon   163Figure 41:^Immunoelectron microscopic detection of L-selectin on normalhuman PMNs in a suspension of leukocyte-rich plasma ^ 165Figures 42: Immunoelectron microscopic detection of total sLeximmunoreactivity on normal human PMNs in a suspension ofleukocyte-rich plasma   165Figure 43:^Immunoelectron microscopic detection of CD18 immunoreactivityon normal human PMNs in a suspension of leukocyte-richplasma ^  165Figure 44:^Electron micrograph of an IL-1 stimulated human endothelialmonolayer 5 minutes after the addition of unstimulated humanPMNs ^ 167Figure 45:^Immunoelectron microscopic detection of L-selectin on a humanPMN that has not yet adhered to an IL-1-stimulated endothelialmonolayer ^  169Figure 46:^Immunoelectron microscopic detection of L-selectin on a humanPMN that has adhered to an IL-1-stimulated endothelialmonolayer ^  171Figure 47: Electron micrograph of a human PMN that was engaged intransmigrating the IL-1 stimulated endothelial monolayer ^ 173Figure 48:^Immunoelectron microscopic detection of L-selectin on a portionof a transmigrating human PMN that has not yet penetrated theIL-1-stimulated endothelial monolayer ^  173Figure 49:^Immunoelectron microscopic detection of L-selectin on a portionof a transmigrating human PMN that has penetrated an IL-1-stimulated endothelial monolayer ^  173Figure 50:Figure 51:Figure 52:Electron micrograph of several human PMNs that have penetratedthe IL-1 stimulated endothelial monolayer within 5 minutes ^ 175Immunoelectron microscopic detection of L-selectin on PMNs thathave completely penetrated the IL-1-stimulated endothelialmonolayer   175Immunoelectron microscopic detection of non-immune mouse IgGlabeling of a non-adherent human PMN as a control for non-specific gold labeling   177xivACKNOWLEDGMENTSI wish to acknowledge my supervisor, Dr. James Hogg, for encouraging me to do aPh.D. in the Pulmonary Research Laboratory (PRL) and for supporting my interest incryoultramicrotomy. I am very fortunate to have trained with Dr. Hogg and I consider thepeople in the PRL to be among the finest that I have ever known. I would also like to thankDr. Claire Doerschuk for allowing me to work with her on this project and I am especiallythankful for the guidance and friendship that she showed me over the years. Special thanks areextended to my committee members Dr. F. Takei, Dr. W. Ovalle, and Dr. G. Krystal for theirconstructive criticism and support. I also would like to thank Dr. David Walker for sharinghis thoughts with me on cell ultrastructure, philosophy, and life. David was very supportiveof me during my Ph.D. programme and I will always be grateful for his kindness andfriendship. To Dr. John Smith I owe a special debt of gratitude for guiding me through thestatistical analysis of the data. To Dr. Stephan van Eeden I am thankful for the manywonderful hours that we spent discussing adhesion molecules. The animal experimentsdescribed in this thesis were greatly facilitated by the surgical expertise of Dean English. I amalso indebted to Fanny Chu for her skilful preparation of tissue samples for light microscopy.The photographs and graphic illustrations used in this thesis were expertly arranged by StuartGreene, a gifted photographer and a good friend. My father, Robert Burns, deserves a specialacknowledgement for designing the "Specimen Pin Jig" that was critical to the success of theproject. I would like to acknowledge the assistance of the office staff, Wendy Cunningham,Kent Webb, Maureen Brooks, and Joan Dixon, for their help in keeping me financed andorganized over the years. I also acknowledge the help of the technical support staff, JoeComeau, Randy Thomson, Barry Wiggs, Jennifer Hards, Harvey Coxson, and LorraineVerburght. To André MacKenzie, a truly resourceful laboratory manager and friend, a specialacknowledgement is given for his imaginative and practical solutions to my endless queries.Finally, I am very grateful for the scholarship support that I received from the University ofBritish Columbia and the Medical Research Council of Canada during my Ph.D. programme.XVDEDICATIONTo my best friend and wife, Elizabeth Donnachie.1CHAPTER ONEINTRODUCTION1.1 INTRODUCTION: OVERVIEW ^ 21.2 THE HISTORY OF LEUKOCYTE ADHESION AND EMIGRATION1.2.1 Leukocytes adhere preferentially to veins, not arteries ^  31.2.2 The mechanism of leukocyte adherence and emigration  61.3 LEUKOCYTE-ENDOTHELIAL ADHESION MOLECULES1.3.1 An introduction to the leukocyte-endothelial adhesion molecules ^ 91.3.2 PMN adhesive glycoproteins that bind to endothelial cells  91.3.3 Endothelial adhesive glycoproteins that bind to PMNs ^  141.3.4 L-selectin as a counter-receptor for P- and E-selectin  201.3.5 PMN margination and emigration in the pulmonary circulation ^ 221.4 SUMMARY AND OBJECTIVES ^  2721.1 INTRODUCTION: OVERVIEWInflammation can be broadly classified as being either acute or chronic. Chronicinflammation differs from acute inflammation not only in duration, but also in the nature ofthe response. By definition, chronic inflammation is persistent. It is generally, but notnecessarily, preceded by acute inflammation and it is characterized by the histological presenceof lymphocytes, monocytes, and plasma cells. Moreover, it is associated with a proliferationof fibroblasts and vascular endothelial cells (Cotran et al., 1989). In contrast, acuteinflammation is of short duration lasting from minutes to several days (Ibid). In response tofocal injury or infection, it is characterized by an increased passage of fluid and leukocytes,mainly polymorphonuclear leukocytes (PMNs), out of the blood and into the affected tissues(Lackie and Smith, 1980; Harlan 1985). The PMNs must first adhere to the vascularendothelium in preparation for emigration. This adherence is mediated by the complexinteraction of surface adhesion molecules expressed on both the PMN and the endothelial cell(Smith, 1990). The expression of surface adhesion molecules can be modulated by the releaseof chemotactic mediators and inflammatory cytokines from injured tissues (Carlos and Harlan,1990). The purpose of this thesis is to investigate the expression of adhesion molecules inthe lung during acute inflammation.The following introduction is intended to highlight those studies and observations onleukocyte biology and acute inflammation that prompted this investigation. The introductionwill focus on some of the historical observations that led to the concept of endothelial- andleukocyte-dependent mechanisms of adhesion and emigration. This will be followed by adiscussion of the proposed role of proinflammatory adhesion molecules on both the PMN andthe endothelium. Evidence gathered from the literature will be used to show that PMN3emigration in the pulmonary circulation is fundamentally distinct from PMN emigration in thesystemic circulation. Based on this distinction, a working hypothesis will be presented alongwith a statement of the specific aims of this thesis.1.2 THE HISTORY OF LEUKOCYTE ADHESION AND EMIGRATION1.2.1 Leukocytes adhere preferentially to veins, not arteriesThe history associated with the study of leukocyte adhesion and emigration is adistinguished one with origins that can be traced back to the late 17th century and the discoveryof the blood leukocyte by Antoni van Leeuwenhoek in 1669 (Schierbeek, 1962). The earlyinvestigators who studied the phenomenon of adhesion and emigration, did so without anyknowledge of the existence of adhesion molecules. In fact, it is only within the last 10 yearsthat adhesion molecules have been demonstrated to play a role in PMN adhesion andemigration (Harlan et al., 1992).The Marquis Henri Dutrochet, is frequently cited as being the first investigator todescribe the adherence and emigration of blood leukocytes in the year 1824 (Atherton andBorn, 1972; Grant, 1973; Williams et al. 1984; Harlan, 1985; Harlan et al., 1992). However,a re-examination of the literature shows that Dutrochet was not the first observer of leukocyteemigration. In fact, by his own admission, he never actually observed the phenomenon ofleukocyte emigration with which he is credited. To begin with, the evidence favouringDutrochet as the discoverer of leukocyte emigration is based upon the following passage fromhis manuscript of 1824:" ... the vesicular globules contained in the blood are added to the tissues of the organsand become fixed there to augment and repair them so that nutrition consists of a4veritable intercalation of fully formed and extremely tiny cells. This opinion, thoughit may seem strange, is however well founded, since observation favours this view.Many times I have seen blood globules leaving the blood stream, being arrested andbecoming fixed to the organic tissue. I have seen this phenomenon, which I was farfrom suspecting, when I observed the movement of the blood in the transparent tail ofyoung tadpoles under the microscope... Observing the movement of the blood, I haveseen many times a single globule escape laterally from the blood vessel and move in thetransparent tissue.., with a slowness which contrasted strongly with the rapidity of thecirculation from which the globule had escaped. Soon afterwards, the globule stoppedmoving and remained fixed in the transparent tissue. A comparison with thegranulations which this tissue contained showed that they were in no way different.There is no doubt that these semi-transparent granulations were also blood globuleswhich had previously become fixed." [Dutrochet, 1824; italics added].The "vesicular globules" have been interpreted as white blood cells (Grant 1973); however,this interpretation is probably incorrect and it is more likely that the blood globules, to whichDutrochet refers, were in fact red blood cells and not white blood cells. In his later (1837)memoirs on the subject, Dutrochet wrote:"...The blood globules are composed of two parts, to be more specific: a centralnucleus which is opaque and a semi-transparent and very delicate envelope whichdissolves rapidly in water, and which contains the coloured material to which the bloodowes its red colour... " [Dutrochet, 1837].This is an excellent description of what we would recognize today as the nucleated red bloodcell of a submammalian vertebrate, such as a tadpole (Andrew, 1965). More importantly,these later memoirs also contain Dutrochet's own admission that he had misinterpreted whathe saw in 1824. He wrote:"Some observations lead me to think that the blood globules are fixed to the organs byan intercalary aggregation, and it is these fixed globules that form the elementaryglobules that give form to all the organs; but this theory seems inadmissible to metoday. I saw, it is true, often enough the blood globules suspending their movementand remaining fixed in the transparent organic tissue of the tails of tadpoles, but thatobservation probably resulted from those globules that were engaged in very smallvessels; the point being that this wasn't a phenomenon of nutrition and growth."[Dutrochet, 1837; italics added].It is clear that Dutrochet's final thoughts on the matter were that the blood globules (red5blood cells) were still within the microvasculature (capillaries) of the tadpole tail and had notemigrated into the tissues. Therefore, it seems unreasonable to continue to credit Dutrochetwith the distinction of being the first investigator to observe leukocyte adhesion and emigration.If Dutrochet was not the first investigator to observe leukocyte adhesion and emigration,then who was? A re-examination of the literature suggests that Albrecht von Haller rightfullydeserves this distinction and that he made his observations almost 70 years before Dutrochet'spublication of 1824. Von Haller documented the adhesion of blood cells to the inner wall ofa mesenteric vein in a frog when he wrote:"Experiment CXXIV. On four frogs. 22 July.I followed with my eyes for a long time the movement of blood, in an arteryand vein of considerable size: it was almost as swift in the vein as in the artery. It wassimilarly swift in the small vessels of the venous network, where the globules followedin single file and far enough apart that they were separated from each other. Theglobules stuck out along the edge of the veins, they gave rise to a half semicircularcircumference, and took the shape of a string of beads, because the membrane of theveins is so thin as to become invisible: this same phenomenon does not exist in arteries,where the membranes are thicker." [von Haller, 1756; italics added].It is well established that, during inflammation, leukocytes preferentially adhere to theveins and not arteries (for a review, see Harlan et al., 1992). While von Haller failed todescribe the colour of these blood globules, his morphological description of blood globulessticking to veins and not arteries is reason enough to think that he was observing leukocyteadhesion. In 1757, in a dissertation on the movement of blood and the effects of hemorrhagein the mesenteries of various species of animals, von Haller described leukocyte emigrationwhen he reported that:"he was struck with the appearance of globules coating the veins like a chaplet of beadsand the extravascular appearance of spherical and yellow cells." [From Grant, 1973;italics added].Grant (1973) correctly points out that red cells appear pale yellow when viewed individually6and he therefore concludes that von Hailer's "yellow cells" were probably red blood cells andnot leukocytes at all (Grant, 1973). However, von Haller may not have been describing theextravascular appearance of spherical yellow cells, but rather the extravascular appearance ofspherical cells (white cells) and yellow cells (red cells). The amphibian white blood cell isspherical, but the red blood cell is flat and ellipsoid (Andrew, 1965). The suggestion that vonHaller did not assign a colour to the spherical cells is in keeping with the idea that they haveno colour. Indeed, years later, Cohnheim would refer to white blood cells as "colourlesscorpuscles" (Cohnheim, 1882).In conclusion then, there is strong evidence to support the idea that von Haller describedthe adhesion and emigration of leukocytes from the mesenteric veins of frogs. This conclusionis important for two reasons. First, it gives von Haller priority on the point of being the firstinvestigator to describe leukocyte adherence and emigration. Secondly, and more importantly,it establishes that the process of leukocyte adhesion and emigration preferentially occurs in theveins and not the arteries of the systemic circulation; an observation that is repeatable even tothis day (Harlan et al., 1992).1.2.2 The mechanism of leukocyte adherence and emigration By the beginning of the 1840's, the early English investigators had begun to use anexperimental approach to show that leukocytes adhere to and emigrate from blood vessels inresponse to inflammatory stimuli (reviewed by Grant, 1973). However, Grant (1973) notes,that no explanation was forthcoming as to the mechanism of leukocyte adhesion and emigrationuntil Cohnheim (1882) and Metchnikoff (1893) considered the problem.Julius Cohnheim's writings elegantly describe leukocyte adhesion and emigration in the7frog (mesentery and tongue) and rabbit (ear). Like von Haller, he draws the reader's attentionto the fact that leukocytes primarily and preferentially adhere to and emigrate from veins ratherthan arteries, although he acknowledges that they sometimes emigrate from small capillariesas well (Cohnheim, 1882). But perhaps Cohnheim's most influential contribution to the studyof inflammation was not what he saw but how he interpreted it. In 1882, he suggested that:"Inflammation is the expression and consequence of a molecular alteration in the vesselwalls...it is only and solely the vessel wall which is responsible for the entire series ofevents..." [Cohnheim, 1882].This suggestion is remarkable in light of the fact that evidence of a molecular alterationwould not be documented for almost another 100 years. However, Cohnheim's theory is onlypartially correct; both the endothelium and the leukocyte are now known to undergo molecularalterations that contribute to the adhesion and emigration process (Smith, 1990). Interestingly,Cohnheim argued so strongly for the role of the endothelium in this process, that he consideredleukocyte emigration to be the result of mechanical filtration. He did not believe that theleukocyte actively participated in adhesion or emigration.In contrast, Elias Metchnikoff believed that the whole process of adhesion andemigration could be attributed to the activity of the leukocyte and he constructed elaborate andlogical arguments to disprove Cohnheim's theory about the role of the endothelium in thisprocess. Metchnikoff believed that the accumulation of leukocytes at sites of inflammation waseffected by their attraction (sensibility) towards a chemotactic substance, a theory firstadvanced by Leber in 1888, to whom he gives full credit. Metchnikoff wrote:"...leukocytes are not sticky and do not become attached on account of their consistencybut solely by means of their amoeboid properties." [Metchnikoff, 1893]."...the leukocytes, led by their sensibility and by means of their amoeboid movements,themselves proceed towards the injured spot instead of passively filtering through avessel-wall." [Metchnikoff, 1893].8Thus, by the end of the 19th century, the study of leukocyte adhesion and emigrationwas dominated by two seemingly incompatible theories. However, as Harlan and colleagues(1992) point out, it would turn out that both theories are correct and it is now appreciated thatthere are endothelial- and leukocyte-dependent mechanisms of adhesion and emigration.Some progress was made in this direction in 1935 by Eliot and Eleanor Clark. In astudy of inflammation in the tadpole tail and the rabbit ear they reported that:"...individual leukocytes were seen to stick at localized points of a vessel wall, toapproach such a point without sticking, and when dislodged by the blood stream tomove on once more without sticking, demonstrated that a change in the endotheliumitself is an essential preliminary to the sticking of leukocytes. Our studies showed alsothat although the migration of leukocytes through the vessel wall is carried out throughthe activity of individual leukocytes, a further change in the wall, beyond that ofstickiness alone appears to be necessary before such cells can successfully penetrate theendothelium, as was shown that a reversal from this latter phase may occur so abruptlyas to trap leukocytes in the act of emigrating, in addition to preventing furtherdiapedesis of the leukocytes adherent to the inner wall." [Clark and Clark, 1935; italicsadded].Their conclusions brought together Cohnheim's views on the importance of an alterationin the constitution of the endothelium with Metchnikoff's concept of the amoeboid nature ofthe leukocyte. Twenty years later, Allison and colleagues made the additional observation thatthe leukocytes also became more adhesive during the inflammatory response. They reportedthat:"...during the course of the inflammatory reaction leukocytes were frequently seen tostick to one another, indicating that the increased adhesiveness characteristic of theinflammatory response is not limited to the endothelium." [Allison, et al., 1955; italicsadded].The critical observation that the leukocyte also undergoes important adhesive changesthat enable it to adhere to and migrate across an inflamed endothelium would have to await thediscovery and characterization of leukocyte-endothelial adhesion molecules.91.3 LEUKOCYTE-ENDOTHELIAL ADHESION MOLECULES1.3.1 An introduction to the leukocyte-endothelial adhesion moleculesTo facilitate the reading of this thesis, an adhesion molecule diagram is shown in Figure1 and it illustrates the leukocyte-endothelial molecules that are considered to be important forPMN adhesion and migration during acute inflammation. It is hoped that this referencediagram will prove useful to the reader as one reads through the body of this thesis.1.3.2 PMN adhesive glycoproteins that bind to endothelial cellsAn intravital microscopy study in the rabbit mesentery showed that PMNs adhere toinflamed (activated) postcapillary and collecting venules in a two-step process (von Andrianet al., 1991). First, PMNs within the flowing blood make transient contacts (adhesions) withthe vascular endothelium which causes them to slowly roll and slide along it (Atherton andBorn, 1972, 1973; Arfors et al., 1987; von Andrian et al., 1991; Ley et al., 1991). In thesecond step of this process, the PMNs cease rolling and become more firmly attached to theendothelium. The firm attachment of PMNs to the endothelium is essential to PMN emigrationand it is mediated by the leukocyte integrins (Harlan, 1985; see below). The rolling of PMNsalong the venular endothelium is largely (80%) dependent upon the leukocyte adhesionmolecule L-selectin (von Andrian et al., 1991; Ley et al., 1991).L-selectin (also known as LECAM-1, Leu 8 Ag, MEL-14, Dreg-56, TQl , gp9OMEL,or LAM-1) is expressed on the surface of PMNs, lymphocytes, monocytes, and eosinophils(Lewinsohn et al., 1987) and it is a member of the selectin family of adhesion molecules(reviewed by Paulson, 1992). Selectins are a unique family of adhesion molecules10characterized by the juxtaposition of an N-terminal C-type lectin domain, an epidermal growthfactor (EGF) domain, and variable numbers of complement regulatory protein (CRP)-likerepeating units (Siegelman et al., 1989; Lasky et al., 1989; Bevilacqua et al., 1989; Johnstonet al., 1989). L-selectin is a 37 kD polypeptide and when expressed as a glycoprotein, it hasa relative molecular mass (under reducing conditions) of 90-100 kD and 74 kD on PMNs andlymphocytes, respectively (reviewed by Kansas et al., 1991). L-selectin was originallyidentified in the mouse by the rat monoclonal antibody, MEL-14 (Gallatin et al., 1983); MEL-14 binds to the amino terminus of the lectin binding domain of L-selectin (reviewed by: Kansaset al., 1991 and Lasky, 1992). In vitro, MEL-14 specifically blocked the adherence oflymphocytes to frozen sections of the peripheral lymph node high venule endothelium, but notto Peyer's patches; in vivo, it inhibited lymphocyte migration into peripheral lymph nodes, andthis suggested that the lymphocyte cell surface antigen (L-selectin) recognized by the MEL-14antibody functions as a "homing" receptor for peripheral lymph nodes (Gallatin et al., 1983;Kansas et al., 1991). Significantly, the MEL-14 antibody also binds to PMNs suggesting thatPMN L-selectin may allow PMNs to home into sites of acute inflammation (Lewinsohn et al.,1987). In vivo, the MEL-14 antibody reduced PMN migration into sites of acute inflammationin the peritoneum and skin of the mouse by 50-70% (Lewinsohn et al., 1987; Jutila et al.,1989, 1991). This reduced migration is likely related to an inhibition of PMN marginationalong the inflamed endothelium (Smith, 1992) as suggested by the in vitro findings that anti-L-selectin antibodies inhibit PMN adhesion to interleukin- 1-stimulated (Smith et al., 1991) andendotoxin-stimulated (Abbassi et al., 1991) endothelial monolayers by 60% under conditionsof flow. Moreover, anti-L-selectin antibodies did not inhibit the transendothelial migration ofadherent PMNs in vitro (Smith et al., 1991; Abbassi et al., 1991).11Within seconds, L-selectin is rapidly downmodulated on the PMN cell surface followingactivation with a variety of inflammatory agents (Kishimoto et al., 1989a; Jutila et al., 1990,1991; Smith et al., 1991; von Andrian et al., 1991). Downmodulation is caused by the rapidshedding of the L-selectin molecule from the cell surface (Kishimoto et al., 1989a). In mouse,the apparent molecular mass of the shed L-selectin molecule is only about 4 kD smaller thanthe plasma membrane-associated form of L-selectin (Ibid) suggesting that L-selectin isproteolytically cleaved at a site very close to the transmembrane domain of the molecule(reviewed by Lasky, 1992). Collectively, these observations predict that if after adhering tothe vascular endothelium, PMNs become activated by tissue-based inflammatory stimuli, thenL-selectin will be shed from the PMN cell surface. Significantly, PMN adhesion to andmigration across an interleukin-1 (IL-1)-stimulated endothelial monolayer in vitro is associatedwith L-selectin downmodulation (shedding) (Kuijpers et al., 1992a, 1992b). IL-1 is aninflammatory cytokine that stimulates the endothelial monolayer to produce platelet activatingfactor (PAF) and interleukin-8 (IL-8) (Kuijpers et al., 1992a). Significantly, both PAF andIL-8 induce L-selectin downmodulation (Smith et al., 1991; Kuijpers et al., 1992a) and PMNsthat have transmigrated an IL-1-stimulated endothelial monolayer are essentially L-selectinnegative (Kuijpers et al., 1992a, 1992b). However, the transmigrated PMN is an endpoint andthe dynamics of L-selectin shedding have not been studied in this model. It is unclear as towhether PMNs lose their L-selectin prior to, during, or shortly after transendothelial migration.Normally, PMNs begin transmigration within a few minutes after adhering to the IL-1-stimulated endothelium. However, when transendothelial migration is intentionally blocked,sustained (30-60 min) PMN adhesion to the IL-1-stimulated endothelium results in a significant,but incomplete, 4-fold loss of L-selectin from the PMN cell surface (Smith et al., 1991). To12what extent transendothelial migration is associated with L-selectin shedding remains to bedetermined.Significantly, studies conducted both in vivo and in vitro show that the same activatingagents that cause L-selectin shedding also induce an increased expression and functionalupregulation in another class of PMN adhesion molecules. These adhesion molecules areknown as leukocyte integrins and they appear to strengthen the adhesive interaction betweenactivated PMNs and the activated endothelium (Arfors et al., 1987; von Andrian et al., 1991).Leukocyte integrins are essential for PMN emigration in the systemic circulation. This pointis best illustrated by a group of patients that suffer from a rare disease known as leukocyteadhesion deficiency (LAD) (reviewed by: Kishimoto et al., 1989b and Arnaout, 1990a, 1990b).In these patients, PMNs fail to emigrate from the systemic circulation in response to infectionor injury and despite having a chronic leukocytosis, these patients suffer from life threatening,recurrent bacterial and fungal infections (Ibid). The molecular basis for this disease is aninherited deficiency in the expression of a high molecular weight leukocyte surface protein(Crowley et al., 1980). In 1984, several investigators simultaneously reported that leukocytesfrom LAD patients are deficient in a class of heterodimeric proteins that are now referred toas the leukocyte integrins (Kishimoto et al., 1989b; Arnaout, 1990a, 1990b).Leukocyte integrins are evolutionarily related to the integrin receptors that mediate cell-cell and cell-matrix adhesion (Kishimoto et al., 1989b; Arnaout, 1990a, 1990b; Hynes, 1992).The term integrin signifies that " ...these are membrane receptors that integrate informationfrom the extracellular environment (extracellular matrix or other cells) with the intracellularcytoskeletal network." (Kishimoto et al., 1989b). As many as twenty different integrins havebeen reported and all of them are afl heterodimers (Hynes, 1992). Integrins are found on a13wide variety of cell types and to date, 8 different 0-subunits and 14 different a-subunits havebeen identified (Ibid). Most a-subunits only associate with one particular type of 0-subunit(the exceptions are «4,0/6, and av) and based on the identity of the 0-subunit, 8 subfamilies (01to 138) of integrins are currently recognized (Ibid). Most integrin subfamilies recognizeextracellular matrix proteins (one exception is «4131, which also recognizes an endothelial ligandknown as vascular cell adhesion molecule-1 (VCAM-1)) and they are considered to beimportant in cell-substrate adhesion (Ibid). However, the leukocyte integrins (32-integrins) arelargely involved in the recognition of cell surface determinants and have a well established rolein cell-cell adhesion (see below). The [32-integrin subfamily consists of 3 different heterodimerswith a variable a-subunit non-covalently linked to a common 0-subunit. The CD (cluster ofdifferentiation designation) nomenclature assigned to each a/13 subunit combination isCD11a/CD18, CD11b/CD18, and CD11c/CD18 (also known as LFA-1, Mac-1, and p150,95,respectively) (Kishimoto et al., 1989b; Arnaout, 1990b; Hynes, 1992). CD11a/CD18 andCD11b/CD18 are important in PMN adhesion and migration (reviewed by Smith, 1992). Thebasal expression of CD11a/CD18 and CD11b/CD18 on the PMN is sufficient for adhesion(Vedder and Harlan, 1988; Schleiffenbaum et al., 1989), but the total CD11/CD18 expressioncan increase 2-10 fold upon stimulation with PMN activating agents such as formyl peptides,phorbol esters, and the complement split product C5a (Arnaout et al., 1984; Bainton et al.,1987; Detmers et al., 1987). This rapid (minutes) increase in CD11/CD18 expression occursthrough the exocytosis of an intracellular granular pool of CD1 lb/CD18 and CD11c/CD18 thatis stored within the PMN (Arnaout et al., 1984; Springer et al., 1986; Bainton et al., 1987;Jones et al., 1988); the majority of this increase is thought to be accounted for byCD 1 lb/CD18 (Bainton et al., 1987). The contribution of CD11c/CD18 to PMN adhesion and14emigration is largely unknown. Although endothelial cells stimulated with interleuldn-1 (IL-1)or endotoxin for 18h show increased binding to purified CD11c/CD18 absorbed to plastic,minimal binding was observed when the CD11c/CD18 density fell to < 1000 binding sites/gm2 (Stacker and Springer, 1991). The current evidence suggests that CD11c/CD18 plays aminor role in the process of PMN adhesion to endothelial cells (reviewed by Smith, 1992).1.3.3 Endothelial adhesive glycoproteins that bind to PMNsStimulation of the vascular endothelium by inflammatory mediators and cytokines resultsin both transient (minutes) and sustained (hours) increases in endothelial adhesiveness forPMNs. Transient increases in endothelial adhesion are largely attributable to P-selectin,whereas sustained increases are effected by E-selectin and ICAM-1. P- and E-selectin areclassed as vascular selectins and together with the leukocyte selectin, L-selectin, they comprisethe three known members of the selectin family (reviewed by Paulson, 1992). ICAM-1, orintercellular adhesion molecule-1, belongs to the immunoglobulin supergene family (Stauntonet al., 1988). The distinct roles played by P- and E-selectin and ICAM-1 in leukocyte-endothelial interactions will now be considered.P-selectin (also known as PADGEM, GMP-140, and CD62) was first identified by aMAb to a surface membrane protein expressed on activated platelets, but subsequent studieshave confirmed that it is also expressed by endothelial cells (reviewed by McEver, 1991).Immunocytochemistry of normal human tissues demonstrated that endothelial P-selectin isfound in specialized endothelial cytoplasmic granules known as Weibel-Palade bodies (Bonfantiet al., 1989; Hattori et al., 1989a, 1989b; McEver et al., 1989). In vitro stimulation of humanumbilical vein endothelial cells with thrombin, histamine, calcium ionophore A23187,15complement proteins C5b-9, or phorbol ester results in a rapid translocation of P-selectin ontothe cell surface (Hattori et al., 1989a, 1989b; Geng et al., 1990). The time course forendothelial P-selectin surface expression is transient; following histamine stimulation, P-selectinexpression is maximal by 3 minutes and declines to near-basal levels by 20-30 minutes as aresult of endocytosis (Hattori et at, 1989b). Unstimulated PMNs rapidly bind to P-selectinin vitro and this binding can be blocked by anti-P-selectin antibodies directed against the lectin-binding domain of P-selectin (McEver, 1991) or by fluid phase P-selectin (Geng et al., 1990).These data suggest that P-selectin expression by activated endothelial cells might promote rapidtargeting of unstimulated PMNs to sites of acute inflammation (Geng et al., 1990; McEver,1991). As mentioned above, during an inflammatory response in the systemic circulation,PMNs preferentially adhere to and emigrate from veins rather than arteries (von Andrian et al.,1991). Significantly, immunoperoxidase localization of P-selectin shows that it is synthesizedby a wide variety of tissues and that the majority of P-selectin is found in small veins andvenules rather than arteries, arterioles, or capillaries (McEver et al., 1989). Although the roleof P-selectin in PMN adhesion and emigration in vivo has not yet been determined, it istempting to speculate that the characteristic expression of P-selectin by veins and not arteriesexplains the PMN's preferred adherence to these vessels. The PMN counter-receptor for P-selectin is unknown (reviewed by Zimmerman et al, 1992), but a possible candidate for thisfunction is PMN L-selectin (Picker et al., 1991; see below).E-selectin, also known as endothelial leukocyte adhesion molecule-1 (ELAM-1), wasfirst identified by MAbs generated against cytokine stimulated endothelial cells (Pober et al.,1986a, 1986b; Bevilacqua et al., 1987); unstimulated endothelial cells do not express E-selectin(Cotran et al., 1986; Munro et al., 1991). E-selectin is expressed on the surface of16endothelium only after the induction of E-selectin mRNA, an event that is preceded byactivation of the transcription factor NF-kB (Montgomery et al., 1991). A wide range ofinflammatory mediators have been shown to induce E-selectin expression and these include:interleukin-1 (11-1), tumor necrosis factor-a (TNF-a), lymphotoxin, bacterial endotoxin, andthe neuropeptide, substance P (reviewed by: Bevilacqua and Gimbrone, 1990, Bevilacqua etal., 1991, and Paulson, 1992). Following the exposure of endothelial monolayers to IL-1,TNF-a, or endotoxin, E-selectin is on the endothelial surface by 1 h, is maximally expressedby 4-8h, and is no longer detectable by 24-48h (Pober et al., 1986a, 1986b; Bevilacqua et al.,1987). E-selectin expression on cultured endothelial cells supports PMN adhesion in vitro(Bevilacqua et al., 1987); however, studies on the role of E-selectin in PMN transmigrationhave produced conflicting results (Smith, 1992). One study provides evidence that PMNmigration across an IL-1-stimulated endothelial monolayer is maximal at 4h and can beinhibited by >90% with anti-E-selectin, anti-ICAM-1, or anti-CD 18 antibodies. When anti-E-selectin and anti-ICAM-1 antibodies were used together, they were found to have an additiveinhibitory effect on PMN adhesion (Luscinskas et al., 1991). Moreover, when the endotheliumwas stimulated with IL-1 for 24h, a time-point when E-selectin expression is low and ICAM-1expression is maximal, the level of PMN transendothelial migration had returned to baseline(Ibid). These findings indicate that PMN adhesion is dependent upon both E-selectin andICAM-1 and that successful PMN transendothelial migration requires the participation of E-selectin, ICAM-1, and CD18 (Ibid). However, in contrast to this study, several other studiesshow that anti-E-selectin antibodies have little or no effect on PMN transmigration and thattransmigration is only CD 18- and ICAM-1-dependent (reviewed by Smith, 1992). The reasonfor these different results is not known, but it may relate to technical differences between the17studies (Ibid). The most convincing arguement that E-selectin does not play a role in PMNmigration and that CD18/ICAM-1 interactions are more important is the finding that CD18-deficient PMNs migrate poorly, if at all, across a 4h endotoxin-stimulated endothelialmonolayer, while 80% of normal PMNs successfully transmigrate the endothelium (Smith etal., 1988).Immunohistochemical studies of human tissues from a variety of disease states haveshown that E-selectin, like P-selectin, is preferentially expressed by venules (Cotran et al.,1986). In vivo, the time course of E-selectin expression appears to correlate with PMNadhesion and emigration. E-selectin was induced on subcutaneous venules in baboons 2 h afteran injection of bacterial endotoxin into the skin; 2h after the endotoxin injection, adherent andextravasated PMNs were associated with these venules (Munro et al., 1991). In response toan intraperitoneal injection of glycogen, PMN emigration was reduced (70%) in rats treatedsystemically with CL-3, an anti-E-selectin antibody (Mulligan et al., 1991). Moreover, CL-3also reduced PMN emigration by approximately 50% into the skin and alveolar airspaces inresponse to IgG immune complex deposition (Ibid). Significantly, immunohistochemistryrevealed that 3-4h after IgG immune complex deposition in the lung, there was a strikingupregulation of E-selectin not only in the pulmonary venules, but also in the alveolar capillaries(Ibid). Alveolar capillaries are an important site of PMN emigration in the lung (see below,Section 1.3.5). A functional role for E-selectin in acute inflammation in the lung is furtherstrengthened by the finding that in a cynomolgus monkey model of extrinsic asthma, PMNemigration in the lung is blocked by pretreatment with the anti-E-selectin antibody CL-2, butnot with the anti-ICAM-1 antibody RR1/1 (Gundel et al., 1991). The in vivo studiesdemonstrate that E-selectin plays a prominent role in acute inflammation; however, E-selectin18expression has also been detected in certain chronic inflammatory diseases includingrheumatoid arthritis, psoriasis, and allergic cutaneous inflammation (reviewed by Paulson,1992). The role of E-selectin in chronic inflammation is unknown, but it may be related toits ability to bind to a variety of different types of leukocytes including T-cells, monocytes,eosinophils, and PMNs (Ibid). The PMN counter-receptor for E-selectin is unknown (reviewedby Zimmerman et al., 1992), but one possible candidate is PMN L-selectin (Picker et al.,1991; see below).ICAM-1, also known as CD54, was identified by Rothlein and colleagues during a studyof phorbol ester induced lymphocyte aggregation. This type of aggregation is mediated byCD1 la/CD18 and it does not involve an increase in CD11a/CD18 surface expression, showingthat CD1 la/CD 18-dependent lymphocyte aggregation is regulated by some other mechanism(Rothlein and Springer, 1986). In addition, the mechanism does not involve like-likeinteractions between CD11a/CD18 adhesion molecules on adjacent cells because CD11a/CD18deficient cells from LAD patients do not aggregate in the presence of phorbol ester, but theywill aggregate with CD11a/CD18 positive cells (Ibid). This finding demonstrated that theligand for CD11a/CD18 was on CD11a/CD18 deficient cells and therefore, monoclonalantibodies (MAbs) were raised against these cells and screened for their ability to inhibitCD11a/CD18-mediated aggregation (Rothlein et al., 1986). One MAb, designated as RR1/1,inhibited cell aggregation and identified a novel cell surface molecule with a molecular weightof 90kD (Ibid). This molecule was termed intercellular adhesion molecule 1 (ICAM-1) (Ibid).Dustin and colleagues (1986) used the RR1/1 antibody to isolate and purify ICAM-1from a variety of cell types. They found that the apparent molecular mass of ICAM-1 variedbetween 90 and 114 kD and that it was heavily glycosylated. Immunohistochemistry19established that both hematopoietic as well as non-hematopoietic cells expressed ICAM-1.ICAM-1 expression on dermal fibroblasts increased 3 to 5 fold after IL-1 or IFN-7 stimulation.Moreover, they found that this increase was dependent on both mRNA and protein synthesis.Of importance to the study of PMN adhesion and emigration was the observation that ICAM-1is constitutively expressed on large and small blood vessels. A substantial body of evidenceshows that ICAM-1 is upregulated on endothelial cells, both in vitro and in vivo, in responseto inflammatory cytokines (IL-1, TNF-a, and IFN-7) or endotoxin (reviewed by Lobb, 1992).In vitro, the time course for ICAM-1 expression initially parallels that of E-selectin and PMNadherence to and migration across an IL-1 stimulated endothelium is maximal at 4h (Smith etal., 1988, 1989; Luscinskas et al., 1991). ICAM-1 appears to mediate CD18-dependentadhesion and is essential to transendothelial migration (Smith et al., 1988, 1989), whereas E-selectin mediates CD 18-independent PMN adhesion (Luscinskas et al., 1991) and, as mentionedabove, there is some controversy over whether it plays a role (Ibid) or not (reviewed by Smith,1992) in transendothelial migration. Unstimulated PMN adherence to ICAM-1 is mediated byCD1 la/CD18, whereas PMNs stimulated with a chemotactic agent (e.g. formyl peptide) showenhanced adherence to ICAM-1 and this enhancement is mediated by the additional engagementof CD11b/CD18 with ICAM-1 (Smith et al., 1989). Significantly, in vivo studies in the rabbitmesentery have established a role for ICAM-1 in the rapid attachment of the PMN to theendothelium following superfusion of the mesentery with zymosan-activated serum (ZAS), arich source of the complement split product C5a (Argenbright et al., 1991). This study showedthat pretreatment with an anti-ICAM-1 antibody, or antibodies directed against the CD18glycoprotein complex, inhibits ZAS-induced PMN adhesion (Ibid). Because of the shortduration of their experiment (10 minutes) and their unpublished observation that exposure of20human umbilical endothelial cells to ZAS for 24h did not induce ICAM-1 expression, theysuggest that basal expression of ICAM-1 is sufficient for stimulated PMN adhesion (Ibid).CD11a/CD18 and CD11b/CD18 bind to different regions of the ICAM-1 molecule. ICAM-1has 5 tandem immunoglobulin domains (Staunton et al., 1988) and CD11a/CD18 binds to thefirst N-terminal domain while CD 1 lb/CD18 binds to the third N-terminal domain (Stauntonet al., 1990; Diamond et al., 1991). Similar observations on the related adhesion moleculeICAM-2 confirm the specificity and importance of these immunoglobulin domains. ICAM-2contains only 2 immunoglobulin domains and these domains are 34% homologous to the first2 immunoglobulin domains of ICAM-1 (Staunton et al., 1989); significantly, it binds only toCD11a/CD18 and not to CD 1 lb/CD18 (Staunton et al., 1989; Diamond et al., 1990). Currentevidence shows that although ICAM-2 is constitutively expressed on endothelial cells, it doesnot show enhanced expression in response to inflammatory stimuli (reviewed by Lobb, 1992;Staunton et al., 1989).1.3.4 L-selectin as a counter-receptor for P- and E-selectinA recent review article suggests that the PMN carbohydrate-bearing ligands for P- andE-selectin are unknown and that they still await characterization at the molecular level(Zimmerman et al., 1992). However, a potential ligand candidate for E-selectin is PMN L-selectin. The evidence is that blocking MAbs to L-selectin or E-selectin inhibit unstimulatedCD 18-deficient PMN binding to 3h IL-1 stimulated endothelium in vitro (Kishimoto et al.,1991). In combination, the effects of these blocking MAbs were not additive, implying thatthey inhibit a common adherence mechanism (Ibid). Formyl peptide stimulation of theseCD 18-deficient PMNs reduced the percentage of adherent cells from 74% to 29%, suggesting21that the PMN ligand for E-selectin was downmodulated (Ibid). Finally, normal unstimulatedPMNs bound avidly to E-selectin-transfected mouse fibroblasts and this binding was equallyinhibitable by either anti-E-selectin or anti-L-selectin blocking MAbs (Ibid). The mechanismby which L-selectin binds to E-selectin was not determined. However, E-selectin (Bevilacquaet al. 1987, 1989; Munro et al., 1989) and P-selectin (Moore et al., 1991, 1992) are thoughtto facilitate PMN adhesion via recognition of sialylated derivatives of the Lewis X (sLex)oligosaccharide (reviewed by Paulson, 1992). In fact, recognition of the sLex oligosaccharideis a property shared by all three members of the selectin family, including L-sele,ctin (Foxallet al., 1992). The importance of the sLex oligosaccharides to selectin-mediated adhesion isclearly illustrated in a recent report on two patients with a rare congenital defect in fucosemetabolism (Etzioni et al., 1992). These patients suffer from recurrent soft tissue infectionswithout the formation of pus (Ibid). The failure of PMNs to emigrate is likely explained bythe finding that these cells do not express the sLex (a fucosylated oligosaccharide) determinant;in vitro, sLex-deficient PMNs do not adhere to E-selectin (Ibid). This type of leukocyteadhesion deficiency (LAD) has been termed LAD II to distinguish it from LAD caused by adeficiency in 62-integrin expression (Ibid; see page 12).Interestingly, P- or E-selectin transfected COS cells bind to purified PMN L-selectinand not to purified lymphocyte L-selectin; this binding can be inhibited up to 70% by an anti-L-selectin MAb (Picker et al., 1991). Similarly, when chymotrypsin was used to selectivelycleave PMN L-selectin from the cell surface, PMN adhesion to E-selectin transfected COS cellswas inhibited by 68% (Ibid). Significantly, PMN L-selectin, unlike lymphocyte L-selectin, ismodified with the sLex oligosaccharide and this observation suggested that PMN L-selectinmay "present" this carbohydrate to these vascular selectins (Ibid). However, L-selectin-22associated sLex accounts for less than 5% of the total cell surface sLex (Ibid); sLex is alsoexpressed by a variety of glycolipids and glycoproteins found on the surface of the PMN(reviewed by Paulson, 1992). Collectively, these data imply that if L-selectin were to functionas an important presenter of sLex to the vascular selectins, then sLex determinants on othercell surface molecules might be functionally less efficient or relatively inaccessible to thevascular selectins (Kishimoto et al., 1991; Picker et al., 1991). The investigation of this latterpossibility, that L-selectin may have a unique topographical distribution on the surface of thePMN favouring its sLex presentation to the vascular selectins, is one of the specific aims ofthis thesis (see below).1.3.5 PMN margination and emigration in the pulmonary circulationAs mentioned above, intravital microscopy studies in the systemic circulation haveshown that PMNs can roll (i.e. marginate) along the vascular endothelium of post-capillary andcollecting venules and this phenomenon is greatly increased during acute inflammation (Smith,1992). L-selectin is largely responsible for this phenomenon (Arfors et al., 1987; von Andrianet al., 1991; Ley et al., 1991) that allows the PMNs to leave the blood stream and marginatealong the vascular endothelium (Atherton and Born, 1972, 1973). Under conditions of flowand shear, both in vitro (Lawrence et al., 1990) and in vivo (Arfors et al., 1987; von Andrianet al., 1991; Ley et al., 1991) L-selectin makes a significantly larger contribution to PMNmargination than CD 18.A different type of PMN margination occurs in the pulmonary circulation and the roleof PMN adhesion molecules in this phenomenon is less clear. In the lungs of normal humans(Peters et al., 1985) and animals (Doerschuk et al., 1987; Martin et a., 1987) there is a large23pool of marginated PMNs and this pool is estimated to contain 0.6-3 times the total numberof circulating PMNs (Doerschuk et al., 1987). In vivo studies in animals have determined thatthe marginated pool of PMNs is located within the capillary bed (Doerschuk et al., 1987; Lienet al., 1987, 1990). Using direct in vivo microscopy, PMNs are observed to temporarily stopmoving (marginate) at discrete locations within the pulmonary capillary network, rather thanmoving through the capillaries at constant rate (Lien et al., 1990). In humans, the diameterof the PMN ranges from 4.9 to 8.1 /Am (Schmid-Schonbein et al., 1980a). In comparison, thediameter of the pulmonary capillaries ranges from 1 to 15 Am (Weibel et al., 1963) andapproximately half of the capillaries have diameters smaller than that of the PMN (Hogg,1987). Although, PMNs and red blood cells are similar in diameter (Schmid-Schonbein et al.,1980a), the PMN is approximately "...1000 times less deformable than the red blood cell andrequires about 1 s to deform and enter a narrow capillary compared with 1/500 s for a redblood cell." (reviewed by Doerschuk et al., 1989). The pulmonary capillary bed consists ofa dense network of short tubular segments (reviewed by Hogg, 1987). In a single pass throughthe lung, it has been estimated that a PMN negotiates approximately 100 tubular segments totravel from the pulmonary artery to the pulmonary vein (Ibid). Because of the large numberof capillary segments and the size discrepancy between the PMN and lumen of the capillary,it has been suggested that PMN transit through the normal lung may be delayed, relative to thered blood cell, not because of the engagement of adhesion molecules, but because of the extratime required for the cell to deform and pass through the narrow capillary segments (Hogg,1987; Doerschuk et al., 1987; Downey and Worthen, 1988). This hypothesis is supported bythe recent observation that in the dog, the pulmonary capillary transit times of CD 18-deficientPMNs are similar to those of normal PMNs (Yoder et al., 1990).24Previous work in our laboratory established that the normal marginated pool of PMNsin the rabbit lung is located primarily within the alveolar capillaries (Doerschuk et al., 1987).The mean diameter of the rabbit PMN is 6-7 Am (Schmid-Schonbein et al., 1980b; Beyers etal., 1989) and this is larger than the mean pulmonary capillary diameter of 5.8 + 0.4 (SD) Am(Beyers et al., 1989). In fact, only 45% of the measured capillary diameters were greater thanthe mean PMN diameter (Ibid). As early as 1894, it was recognized that PMN sequestrationin the rabbit lung could be enhanced by an intravenous injection of peptone (Bruce, 1894).Since this report, numerous investigators have shown, in other animal species and humans, thatPMN sequestration in the lung is also enhanced by infusion of various chemotactic stimuli(reviewed by Harlan et al., 1992). In all cases, there is a rapid fall in the circulating PMNcount with a return to normal after several hours (Ibid). Several investigators have suggestedthat the enhanced sequestration of activated PMNs in the lung is due to a decrease in PMNdeformability (Hogg, 1987; Doerschuk et al., 1989; Worthen et al., 1989). Experimental datasupporting this hypothesis include: the observation that rabbit PMNs are less deformable(stiffer) following activation with formyl peptide and this correlates with their increasedretention in the lung (Worthen et al., 1989); in the rabbit lung, the transient PMN sequestrationproduced by an intravenous injection of formyl peptide was not blocked by pretreatment withan anti-CD18 MAb and is therefore, CD 18-independent (Lundberg and Wright, 1990); thegreatest number of PMNs sequester in the alveolar capillaries following an infusion of zymosanactivated plasma (ZAP) or C5f (a fragment of the fifth component of complement) in rabbits(Doerschuk et al., 1989) and dogs (Lien et al., 1991), respectively.However, decreased PMN deformability may not be the only mechanism of enhancedPMN sequestration. Systemic activation of complement in humans is associated with increased25(5-fold) expression of CD11b/CD18 on PMNs in vivo and it has been suggested that thisincreased CD 18 expression may be related to PMN sequestration in the lung (Arnaout et al.,1985). Direct evidence for CD18 involvement in this process was provided by an in vivovideofiuorescence microscopy study of the effect of ZAP infusion on PMN sequestration in doglung (Yoder et al., 1990). This study showed that the pulmonary capillary transit time ofnormal dog PMNs is increased by ZAP infusion, but more importantly, that ZAP infusion hadno effect on the transit time of CD 18-deficient dog PMNs (Ibid). Additional studies in therabbit suggest that ZAP induced PMN sequestration in the lung is a two-step process thatprobably depends on both decreased PMN deformability and increased PMN adhesion(Doerschuk, 1992). The data supporting this conclusion are that administration of the anti-CD18 antibody, MAb 60.3, did not prevent the initial fall in the circulating PMN count thatoccurs within 1 minute following ZAP infusion. Significantly, MAb 60.3 did prevent thesustained accumulation of PMNs in the lung that occured between 7 and 15 minutes (Ibid).The data show that the initial fall in the circulating PMN count is CD 18-independent and likelymediated by a decrease in PMN deformability. In contrast, sustained PMN sequestration inthe lung may depend upon a transient increase in PMN adhesiveness that is CD 18-dependent(Doerschuk et al., 1992). However, this study does not entirely exclude the possibility thatthe initial CD 18-independent sequestration produced by ZAP may be due to the rapid andreversible induction of a CD 18-independent adhesive process (Ibid). In vitro and in vivo datashow that the CD 18-independent adhesive processes are largely mediated by the selectin familyof adhesion molecules (i.e. L-selectin, E-selectin, and P-selectin; see above). A role for L-selectin in normal or enhanced PMN margination in the lung has not yet been determined.Transmission electron microscopic studies have clearly demonstrated that PMNs can26emigrate directly from alveolar capillaries during acute inflammation in rabbits (Walker et al.,1992), mice (Loosli and Baker, 1962), and dogs (Lien et al., 1991). Significantly, in responseto an airway administration of C5f in the dog, 90% of PMN emigration in the lung was fromcapillaries rather than venules (Lien et al., 1991). This implies that, during acuteinflammation, PMN emigration in the pulmonary circulation is fundamentally different fromthat in the systemic circulation (Ibid) where the major site of emigration is in the veins(Allison, Jr. et al. 1955; Harlan et al., 1992). Interestingly, PMN emigration in the pulmonaryand systemic circulations may not just differ in the preferred site of emigration, but also in thepreferred mechanism of emigration.In the systemic circulation, PMN emigration towards all inflammatory stimuli examinedto date is mediated by CD18 (Harlan et al., 1992). However, previous work in our laboratoryestablished that PMN emigration in the lung is different and can occur through either a CD18-independent or CD 18-dependent mechanism, depending on the nature of the inflammatorystimulus (Doerschuk et al., 1990a, 1990b). Specifically, the data showed that systemicadministration of the anti-CD18 monoclonal antibody MAb 60.3 did not inhibit PMNemigration in response to an intrabronchial instillation of Streptococcus pneumoniae orhydrochloric acid, but it completely inhibited PMN emigration when these stimuli wereinstilled into the peritoneal cavity. Conversely, MAb60.3 largely inhibited PMN emigrationinto the rabbit lung in response to an intrabronchial instillation of phorbol ester, Escherichiacoli organisms and E. coli endotoxin. The mechanism of CD 18-independent PMN emigrationin the rabbit is unknown, but it has been suggested that it may be due to the large number ofalveolar macrophages within the lung (Mileski et al., 1990); S. pneumoniae may induce therelease of a macrophage product that is chemotactic or proadhesive for PMNs (Ibid). The27evidence is that in the rabbit peritoneum, PMN emigration towards S. pneumoniae is normallyCD18-dependent, but when the number of peritoneal macrophages was experimentallyincreased, a significant number of PMNs emigrated by a CD18-independent mechanism (Ibid).Importantly, PMN emigration towards E. coli organisms was CD 18-dependent, even when thenumber of macrophages was increased (Ibid). This observation established that the presenceof the macrophage is essential to the induction of CD18-independent emigration (Ibid).Moreover, it was proposed that macrophages might ingest streptococcal bacteria and releasechemotactic or proadhesive bacterial degradation products (Ibid).1.4 SUMMARY AND OBJECTIVES In summary, there is a large body of evidence to suggest that PMN adherence to andemigration across endothelial cells is largely effected by three distinct classes of adhesionmolecules, the selectins (P- and E-selectin on the endothelium and L-selectin on the PMN), theleukocyte integrins (CD11a/CD18 and CD11b/CD18) on the PMNs, and a member of theimmunoglobulin super family (ICAM-1) expressed on the endothelium. In the systemiccirculation, the PMNs preferentially adhere to and emigrate from post-capillary and collectingvenules rather than arteries or arterioles. The initial adhesion of the PMN to the inflammedvenular endothelium is thought to be mediated primarily by L-selectin on the PMN and P- andE-selectin on the endothelium; L-selectin functions to allow PMNs to leave the axial flow ofthe blood stream and marginate along the blood vessel wall. Although the endothelial ligandfor PMN L-selectin at sites of acute inflammation has not been identified, in vitro data suggestthat the PMN L-selectin contains an sLex oligosaccharide that can be "presented" to either P-or E-selectin. Once the PMN makes contact with the endothelium, local cytokine activation28promotes the shedding of L-selectin and the firm adhesion of the PMN; firm adhesion isnecessary for PMN emigration, a process mediated by the CD18 glycoprotein complex.ICAM-1 functions as an endothelial ligand for both CD1 1a/CD18 and CD11b/CD18. In vivo,in the systemic circulation, PMN emigration towards all inflammatory stimuli examined to dateis mediated by CD18.In the pulmonary circulation, PMNs sequester (marginate) in the normal lung, but thismargination appears to be determined by physical factors rather than leukocyte-endothelialadhesion molecules. During an inflammatory response, enhanced and sustained sequestrationof PMNs in the lung is mediated by the CD18 glycoprotein complex. Whether otherleukocyte-endothelial adhesion molecules, like L-selectin or ICAM-1, participate in this processis unknown. Depending upon the nature of the inflammatory stimulus, PMNs emigrate byeither a CD18-dependent or CD18-independent mechanism. The mechanism of CD18-independent PMN emigration in the lung has not yet been determined, nor has a role forICAM-1 in CD 18-dependent or -independent PMN emigration been established.The main objective of this thesis is to quantitate and compare the surface expressionof L-selectin, CD 18, and ICAM-1 during CD 18-independent and CD 18-dependent PMNemigration in the lung.My working hypothesis is that, during an acute inflammatory response, the adherenceof PMNs to the pulmonary microvasculature and their subsequent emigration into the alveolarairspace are dependent upon the complex interaction of cell surface adhesion moleculesexpressed on both the PMN and the endothelium. The principal site of PMN emigrationduring acute inflammation in the lung is the alveolar capillaries. PMNs that sequester in thecapillary bed near the injured site are activated by the release of local inflammatory mediators29and cytokines. Some of these stimuli will also activate the capillary endothelial cells.Therefore, PMN emigration in response to inflammatory stimuli that evoke either a CD18-independent or CD 18-dependent PMN emigration is associated with a downmodulation of L-selectin, an upmodulation of CD18, and an upmodulation of ICAM-1.The method of approach used in this study entailed the successful application ofmonoclonal antibodies and colloidal gold markers to the ultrastructural localization andquantitation of L-selectin, CD18 and ICAM-1 in normal and acutely inflamed rabbit and mouselungs. In addition, for comparative purposes, this method was used to study the in vitroexpression of L-selectin on unstimulated human PMNs and on human PMNs engaged intransmigration of an IL-1-stimulated endothelial monolayer.The Specific Aims of this thesis are:1. To determine, in the rabbit lung, whether L-selectin expression is downmodulatedand whether CD18 expression is upmodulated in response to stimuli that induceCD18-independent and CD18-dependent PMN emigration.2. To determine, in the mouse lung, whether ICAM-1 expression is upmodulated inresponse to stimuli that are known to induce CD18-independent and CD18-dependent PMN emigration in the rabbit lung.3. To determine, in vitro, whether L-selectin is topographically positioned on thehuman PMN surface such that it renders L-selectin-associated sLex more"bioavailable" for the vascular selectins P- and E-selectin.4. To determine, in vitro, whether the process of human PMN transendothelialmigration is associated with L-selectin downmodulation.30CHAPTER TWOMATERIALS AND METHODS2.1 SOURCES OF MATERIALS2.1.1 Animals ^  322.1.2 Monoclonal antibodies ^  322.2 RABBIT PNEUMONIA EXPERIMENTS2.2.1 Induction of pneumonia in the rabbit ^  332.2.2 Preparation of rabbit lung tissue for cryoultramicrotomy ^ 332.3 MOUSE PNEUMONIA EXPERIMENTS2.3.1 Induction of pneumonia in the mouse ^  352.3.2 Preparation of mouse lung tissue for cryoultramicrotomy ^ 362.3.3 Preparation of mouse lung tissue for light microscopy  372.4 HUMAN LEUKOCYTE-RICH PLASMA EXPERIMENTS2.4.1 Collection of human LRP ^  372.4.2 Preparation of human LRP for cryoultramicrotomy ^  382.5 HUMAN PMN TRANSENDOTHELIAL MIGRATION EXPERIMENTS2.5.1 Preparation of acellular amniotic tissue for cell culture ^ 382.5.2 Culturing human umbilical vein endothelial cell (HUVEC)monolayers on amniotic tissue ^  392.5.3 Isolation of PMNs ^  402.5.4 Induction of PMN transmigration across HUVEC monolayers ^ 402.5.5 Preparation of transmigrated HUVEC monolayers for cryoultramicrotomy^. ^ 402.6 CRYOULTRAMICROTOMY2.6.1 Fabrication of brass specimen pins for cryosectioning ^  412.6.2 Mounting of cell and tissue samples to the brass specimen pins ^ 422.6.3 Cryosectioning and section retrieval ^  432.7 IMMUNOGOLD LABELING OF CRYOSECTIONS2.7.1 Immunogold labeling of PMNs in rabbit lung ^  432.7.2 Immunogold labeling of mouse lung  442.7.3 Immunogold labeling of human LRP  442.7.4 Immunogold labeling of transmigrating human PMNs  45312.7.5 Post-fixation and contrast enhancement of labeled cryosections ^ 462.8 IMMUNOELECTRON MICROSCOPY2.8.1 Immunogold quantitation of L-selectin and CD18 on rabbit PMNs ^ 462.8.2 Immunogold quantitation of ICAM-1 in mouse lung ^  482.8.3 Immunogold localization of L-selectin in human LRP  492.8.4 Immunogold localization of L-selectin on transmigrating human PMNs ^ 502.9 STATISTICAL ANALYSIS2.9.1 Determination of the minimum number of PMN profiles required to accuratelyestimate the mean gold label density within an animal ^ 502.9.2 Determination of intra- and interobserver error associated withgold particle counts on PMNs ^  522.9.3 Statistical analysis of L-selectin and CD18 immunoreactivity onrabbit PMNs ^  532.9.4 Statistical analysis of ICAM-1 immunoreactivity in the mouse lung ^ 55322.1 SOURCES OF MATERIALS 2.1.1 Animals New Zealand White (NZW) female rabbits (2-3 kg) were purchased from R & RRabbitry (Stanwood, WA, USA) and BALB/c female white mice (18-20 g) were purchasedfrom Charles River Canada Ltd. (St. Constant, Que., Canada).2.1.2 Monoclonal antibodiesThe four monoclonal antibodies used in the studies described in this thesis werereceived as generous gifts. The DREG-200 mouse monoclonal IgG antibody, against humanL-selectin (Jutila et al., 1990; Kishimoto et al., 1990) and against rabbit L-selectin (vonAndrian et al., 1991), was provided by Dr. E.C. Butcher (Stanford University School ofMedicine, Stanford, CA, USA). The 60.3 mouse monoclonal IgG antibody, against humanCD18 (Beatty et al., 1983; Wallis et al., 1986) and against rabbit CD18 (Lindbom et al.,1990), was provided by Dr. J.M. Harlan (University of Washington, Seattle, WA, USA). TheYN1/1.7.4 rat monoclonal IgG antibody, against mouse ICAM-1 (Takei, 1985; Prieto et al.,1989), was provided by Dr. F. Takei (University of British Columbia, Terry Fox Laboratory,Vancouver, B.C., Canada). The CSLEX-1 mouse monoclonal IgM antibody, against humansLex (Picker et al., 1991), was provided by Dr. L. Picker (University of Texas SouthwesternMedical Center, Dallas, TX, USA).332.2 RABBIT PNEUMONIA EXPERIMENTS 2.2.1 Induction of pneumonia in the rabbitStreptococcal or endotoxin pneumonias were induced in 2-3 kg NZW female rabbits(Doerschuk et al., 1990a; 1990b). Briefly, the rabbits were anesthetized with an intra-muscularinjection of ketamine hydrochloride (Ketalean, MTC Pharmaceuticals, Cambridge, Ontario,Canada, 25-40 mg/kg) and acepromazine maleate (Atravet, Ayerst Laboratories, Montreal,Quebec, Canada, 2-3 mg/kg). The trachea was surgically exposed and a small incision wasmade just below the larynx. A narrow, flexible infant feeding tube (Argyle Feeding Tube, 31/2 FR X 12", X-ray opaque, Argyle, Division of Sherwood Medical, St. Louis, MO, USA)was inserted through the tracheostomy and manoeuvred into a distal bronchus of the right lungusing fluoroscopy. A unilateral pneumonia was induced by instilling intrabronchially 0.15ml/kg of either Streptococcus pneumoniae (109 organisms/ml saline, a clinical isolate providedby Microbiology, St. Paul's Hospital, Vancouver, B.C., Canada, (n=3)) or endotoxin(Escherichia coil serotype 055:B5, Sigma, St. Louis, MO, USA, 10 gg/m1 saline, (n =3)).Both inflammatory instillates contained 5% colloidal carbon (Pelikan, Fount India, Hannover,Germany) to mark the distribution of the instillate in the lung. Following instillation, allincisions were sutured and the pneumonias were allowed to develop for 4h. Control rabbits(n =3) received no anesthesia and no instillates.2.2.2 Preparation of rabbit lung tissue for cryoultramicrotomyAt the end of the 4h treatment period, each rabbit received an intravenous overdose ofpentobarbital. The chest and pericardial sac were opened and the base of the heart was ligated34to keep the pulmonary blood volume. The thoracic organs were removed and the right hilumwas clamped with a hemostat. The left lung was then inflated at 20 cm H20 pressure with0.025% glutaraldehyde in ice cold phosphate buffered saline (PBS), which consisted (in mM)of 138 NaC1, 13.2 KH2PO4, and 53.2 Na2PO4 (pH 7.2), and the left hilum was clamped.Because previous studies in this laboratory have determined that approximately half of thealveolar space within the pneumonic region is filled with edematous fluid (Doerschuk,unpublished observation), a fixative instilled into this edematous region would be diluted byapproximately 50%. To compensate for this dilutional effect, the right hilum was undampedand inflated with ice cold PBS containing double strength (0.05%) glutaraldehyde and thenclamped once more. Control rabbits were similarly treated except that the lungs were notclamped prior to instillation of fixative and both lungs were inflated with ice cold PBScontaining 0.025% glutaraldehyde.The lungs were fixed for lh at 4°C and then cut into 2 mire tissue blocks. Thepneumonic regions were identified by colloidal carbon staining and tissue blocks were takenfrom these "blackened" regions and fixed for an additional lh at 4°C in 0.05% bufferedglutaraldehyde. Comparable tissue blocks were also taken from control rabbit lungs and fromthe uninvolved, contralateral "non-blackened" lungs of each rabbit that had received aninflammatory instillate. These tissue pieces were fixed for an additional lh at 4°C in 0.025%buffered glutaraldehyde. Following the fixation period, all lung tissue blocks were rinsed inice cold PBS and then cryoprotected by immersion in PBS containing 2.3M sucrose overnightat 4°C.The next day, cryoprotected lung tissue blocks (n =16 blocks from each control,pneumonic, and contralateral lung) were mounted on brass specimen pins and frozen in liquid35nitrogen (Section 2.6.2).2.3 MOUSE PNEUMONIA EXPERIMENTS 2.3.1 Induction of pneumonia in the mouseStreptococcal or endotoxin pneumonias were induced in BALB/c female mice (18-20g, n=13) using a modification of the rabbit pneumonia protocol (Section 2.2.1). Thismodification was necessary for two reasons. The first is that the small size of the mouse,relative to the rabbit, precluded the possibility of catheterizing a distal bronchus and inducinga unilateral pneumonia. The second reason is that mice are approximately 1000 times moreresistant to the effects of endotoxin than rabbits (Beutler et al., 1985). Therefore, threechanges were introduced to the protocol as follows: 1) bilateral, rather than unilateral,pneumonias were induced by tracheal instillation, 2) the dose of E. coil endotoxin deliveredto each mouse was increased approximately 1000X relative to the rabbit and, 3) the endotoxinpneumonias were allowed to develop over 24h instead of 4h because, even with the elevateddose of endotoxin, an appreciable pneumonia did not develop until 24h.Mouse pneumonias were induced using the following protocol: BALB/c female mice(18-20 g, n=13) were anesthetised with halothane (Fluothane, Ayerst Laboratories, Montreal,Que., Canada), the trachea surgically exposed and a 25 gauge needle attached to a 1 c.c.syringe containing the instillate was inserted just below the larynx. A bilateral pneumonia wasinduced by instilling S. pneumoniae (0.1 ml, 109 organisms/ml, n= 3) or E. coil endotoxin(0.1 ml, 2 mg/ml, n=3) mixed with 5% colloidal carbon in saline intratracheally. The neckincision was closed, but not sutured, and the mice were allowed to recover for 4h and 24h,36respectively. Control mice were similarly anesthetised and received sterile tracheal instillatescontaining 5% colloidal carbon in saline and were sacrificed 4h (n=2) and 24h (n =2) later.Additional controls included mice (n=3) that received no anesthesia and no instillates.2.3.2 Preparation of mouse lung tissue for cryoultramicrotomyAt the end of the treatment period (4h or 24h), each mouse was killed with an overdoseof halothane, the chest and pericardial sac were opened, and the thoracic organs were removed.The lungs were fixed by instilling ice cold PBS containing 0.4% glutarladehyde down thetrachea.Fixation was allowed to proceed for lh at 4°C, after which time the lungs were cut into2 mm3 tissue blocks. The pneumonic lung regions were identified by colloidal carbon stainingand tissue blocks were taken from these "blackened" regions and fixed for an additional lh at4°C in 0.4% buffered glutarladehyde. In those control mice that received colloidal carboninstillates or no instillates at all, tissue blocks from "blackened" or "pink" regions of lungrespectively, were taken and also fixed for an additional lh at 4°C in 0.4% glutaraldehyde.Following the fixation period, all lung tissue blocks were rinsed in ice cold PBS and thencryoprotected overnight by immersion at 4°C in PBS containing 2.3M sucrose.The next day, lung tissue blocks (n=20 blocks/mouse) were placed on filter paper strips(20 mm X 7 mm) cut from Whatman #1 (Whatman International Limited, Maidstone, Eng.UK) that had been pre-moistened in cryoprotectant (PBS containing 2.3M sucrose). Excesscryoprotectant was removed and the filter paper strips carrying the lung tissue blocks wereimmersed in liquid N2 and rapidly frozen. The frozen filter strips were then transferred toNalgene cryovials and stored under liquid nitrogen for up to two years prior to cryosectioning.37In order to prepare the lung tissue blocks for cryosectioning, they were retrieved from liquidnitrogen and rapidly thawed by immersion (with constant swirling) in room temperaturecryoprotectant (PBS containing 2.3M sucrose). The thawed lung tissue blocks were thenmounted on brass specimen pins as described in Section 2.6.2.2.3.3 Preparation of mouse lung tissue for light microscopyTo verify that S. pneumoniae and E. coli endotoxin instillation caused PMN emigrationinto the alveolar airspace, five to ten lung tissue blocks fixed in 0.4% glutaraldehyde from eachcontrol and treated mouse were re-fixed in fresh PBS containing 2% glutaraldehyde for lh.These tissue blocks were dehydrated through a graded alcohol series (30%, 50%, 70%, 90%,and 100% ethanol), infiltrated and embedded in glycol methacrylate (JB4 Embedding Kit,Polysciences, Warrington, PA, USA), sectioned at 2 Am on a Sorvall J134 microtome (Porter-Blum) using a glass knife and stained with Toluidine blue 0. Photomicrographs were madeon a Zeiss Universal light microscope using Fujichrome Tungsten 64 ASA film.2.4 HUMAN LEUKOCYTE-RICH PLASMA EXPERIMENTS 2.4.1 Collection of human LRPLeukocyte-rich plasma (LRP) was obtained from adult human volunteers (n=3) usinga modification (Doerschuk, 1987) of a method originally described by Boyum (1974). Briefly,to prevent coagulation, venous blood (24 ml) was drawn into acid citrate dextrose (ACD, 6ml). Erythrocyte (RBC) sedimentation was induced by the addition of 25 ml of 4% dextran(100-200 IcD) in PMN buffer, which consisted (in mM) of 138 NaCl, 27 KC1, 8.1 Na2HPO4,38and 5.5 glucose (pH 7.4). A sharp interface appeared between the sedimented RBC and theLRP within 30-40 min. The RBC fraction was discarded, and the LRP fraction wascentrifuged at 1000 rpm (200 g) (Beckman Model TJ-6 centrifuge) for 8 min, the supernatantdiscarded and the leukocytes were resuspended in 1 ml of PBS. Because the LRP suspensionscontained large numbers of PMNs that were easily identified by transmission electronmicroscopy, no further purification of the leukocytes was required.2.4.2 Preparation of human LRP for cryoultramicrotomyThe resuspended LRP fraction was fixed in 0.05% buffered glutaraldehyde for 5 minat room temperature, washed in PBS, and resuspended in warm (37°C) PBS containing 2%agarose (Ultrapure, low melting point ( < 30°C), Bethesda Research Laboratories, Gaithersburg,MD, USA). After cooling to room temperature, the LRP-agarose gel was cut into 2 mm3blocks and these blocks were then immersed in cryoprotectant (PBS containing 2.3M sucrose)overnight at 4°C.The next day, cryoprotected LRP-agarose blocks (n=16 blocks/human subject) weremounted on brass specimen pins and frozen in liquid N2 as described in Section 2.6.2.2.5 HUMAN PlVIN TRANSENDOTHELIAL MIGRATION EXPERIMENTS2.5.1 Preparation of acellular amniotic tissue for cell cultureHuman amniotic tissue was prepared by a well established protocol in Dr. C.W. Smith'slaboratory at Baylor College of Medicine, Houston Texas (Furie et al., 1984; Furie andMcHugh, 1989). In brief, placentas were obtained from vaginal and Cesarean deliveries. The39amnion was separated from the chorion by blunt dissection and fastened to Teflon rings (16mm I.D., 22 mm 0.D., 9.5 mm high) with Viton (vinylidene fluoride-hexafluoropropylene)0-rings (C.E. Conover, Fairfield, NJ, USA). The amniotic epithelium was removed by lysiswith 0.25N NH4OH for lh at room temperature and then stored for up to one month at 4°Cin HEPES-buffered saline (137 mM NaC1, 4mM KCL, 11 mM glucose, 10 mM HEPES, 500U/ml penicillin, and 200 Ag/m1 streptomycin, pH 7.4). Just prior to seeding with humanumbilical vein endothelial cells (HUVECs), the stored amniotic preparation was incubated inM199 (GIBCO Laboratories, New York, NY, USA) containing 20% fetal calf serum (FCS).2.5.2 Culturing human umbilical vein endothelial cell (HUVEC) monolayers on preparedamniotic tissueThe endothelial cells were harvested from 5 to 10 human umbilical veins by collagenaseperfusion according to Huang et al. (1988), and characterized as endothelial cells by the abilityto bind acetylated low-density lipoprotein (Voyata et al., 1984), and to express factor VIII(Jaffe et al., 1973). Pooled cells were resuspended in M199 supplemented with 20% heat-inactivated fetal bovine serum (HyClone Laboratories Inc., Logan UT), 100 U/ml penicillin,100 ug/ml streptomycin, and 2 ug/ml amphotericin B. Four or five days after plating, cellsfrom confluent cultures were detached with 0.125% trypsin (ICN Biomedicals, Cleveland, OH,USA) and 1 mM EDTA in Ca2+- and Mg'-free PBS, pooled and seeded on the stromalsurface of the human amniotic tissue at a density of 1.5 X 105 cells; each ring contained 2 cm2of exposed tissue. The HUVEC cultures generally reached confluence in 2 days, and wereused for PMN migration experiments from 7 to 10 days post-seeding. Confluence wasdetermined by silver nitrate staining according to Furie et al. (1984).402.5.3 Isolation of PMNsOn two separate occasions, PMNs were isolated from adult human volunteers (n=2)using established protocols (English and Andersen, 1974). Briefly, PMNs were isolated fromheparinized (10U/m1) venous blood that was sedimental and centrifuged at 1000g for 30 minover Ficoll-Hypaque gradients. Contaminating erythrocytes were removed from thepreparation by hypotonic lysis in 0.2% NaCl for 45 s. The PMNs were resuspended in coldDulbecco's PBS (Gibco Laboratories, Grand Island, NY) containing 0.2% dextrose at 4°C, pH7.4 and added to the IL-1 activated HUVEC monolayer (Section 2.5.4) within 2h. The PMNsuspensions were 95% pure and trypan blue exclusion showed greater than 97% viability.2.5.4 Induction of PMN transmigration across HUVEC monolayersHUVEC monolayers were treated with human monocyte derived interleukin-1 (IL-1,Genzyme Inc., Boston, MA, USA) for 3-4h, a time previously shown to be optimal for theenhancement of PMN adherence (Smith et al., 1988). The IL-1 was added directly to themedia in which the monolayers were cultured at a final concentration of 3 U/ml. Themonolayers (n=9 per experiment) were then rinsed with serum-free M199 at 37°C for 20 to30 minutes (Lawrence et al., 1990) and the PMNs were added to the upper chamber (a totalof 1 X 106 cells/2-cm' ring). This yields an expected ratio of 5 PMNs/endothelial cell atconfluence (Huang et al. 1988). The PMNs were allowed to settle onto and migrate acrosseach HUVEC monolayer for 5 min.2.5.5 Preparation of transmigrated HUVEC monolayers for cryoultramicrotomyAt the end of the 5 min PMN transmigration period, the monolayer preparations were41fixed at 4°C with PBS buffer (pH 7.2) containing 0.1% glutaraldehyde for 15 min and thenrinsed in ice cold PBS for 10 min. The HUVEC monolayer preparations were immersed incryoprotectant (PBS buffer containing 2.3M sucrose) at 4°C overnight and then, on the nextday, each HUVEC monolayer was detached from its Teflon ring using a pair of fine forceps,re-immersed in cryoprotectant and shipped on ice from Houston, Texas to Vancouver, BritishColumbia.Upon arrival in Vancouver, the HUVEC monolayers were rapidly frozen by plungingthem in liquid nitrogen. The frozen monolayers were stored in Nalgene cryovials under liquidnitrogen for up to one year. In order to prepare the monolayers for cryosectioning, they wereretrieved from liquid nitrogen and rapidly thawed by immersion (with constant swirling) inroom temperature cryoprotectant (PBS containing 2.3M sucrose). The thawed monolayerswere then mounted on brass specimen pins as described in Section 2.6.2.2.6 CRYOULTRAMICROTOMY2.6.1 Fabrication of brass specimen pins for cryosectioningIn the rabbit study, all of the cryoprotected lung tissue samples were mounted onindividual specimen pins (n=240), frozen in liquid N2, and stored until they were retrieved forcryosectioning (Section 2.2.2). Additional specimen pins were also required for the mountingof mouse and human cryoprotected samples (Sections 2.3.2, 2.4.2, and 2.5.5). Since only 10specimen pins were supplied with the cryoultramicrotome (RMC, Tuscon, AZ, USA), andbecause the cost of purchasing additional specimen pins was prohibitive ($12.00 each, USfunds), another source of specimen pins was required. After much trial and error, the final42solution to the problem was to develop a "Specimen Pin Jig" (Figure 2) that could be used inthe laboratory to produce inexpensive, brass specimen pins.One-half inch (12.5 mm) long brass specimen pins were cut from 1/8 inch (3.2 mm)brazing rod using a metal hacksaw with 32 teeth/inch (13 teeth/cm) and a custom builtaluminium "Specimen Pin Jig" (Figure 2). The jig was mounted in a table-top vice and thebrazing rod was inserted into the jig up to the "swing-stop". The hacksaw blade was theninserted into the guide slot and the rod was cut. The newly cut pin was retrieved from theblock by moving the swing-stop out of the way and pushing the pin forward with the remainingportion of the brazing rod. The swing-stop was then lowered and the operation repeated.Thirty specimen pins could be made in 15 minutes at a material cost of just under $1.00.2.6.2 Mounting of cell and tissue samples to the brass specimen pinsBrass specimen pins were placed vertically into holes (3.3 mm diameter X 5 mm deep)that had been drilled into a modified nylon kitchen cutting board (70mm X 70mm X lOmm).Each cryoprotected specimen (lung tissue, agarose blocks, HUVEC monolayers) was trimmedunder cryoprotectant into the shape of a pyramid on the nylon cutting board using a sharp razorblade. The trimmed specimen was then carefully positioned onto the specimen pin with itsapex pointing away from the newly cut surface of the pin. Excess cryoprotectant waswithdrawn with the ragged edge of a torn piece of filter paper (Whatman #1) leaving a thinfilm of cryoprote,ctant behind. The prepared specimen was then rapidly frozen at -196°C byplunging it into liquid nitrogen. The hacksaw blade, that was used to cut the specimen pins,imparted a serrated pattern to the mounting surface of the pin and these serrations strengthenedthe adhesive bond that formed between the specimen and the pin during freezing. Once frozen,43the mounted specimens were either cryosectioned immediately, or catalogued and maintainedunder liquid nitrogen in Nalgene cryovials (Nalge Company, Rochester, NY, USA) for up tothree years before being retrieved for cryosectioning.2.6.3 Cryosectioning and section retrieval All cryoprotected, frozen specimens mounted on brass specimen pins werecryosectioned according to established procedures (Tokuyasu, 1973, 1978, 1983; Griffiths etal., 1984). Ultrathin (80-100 nm) cryosections were cut at -85°C (using a diamond knife onan RMC MT6000-XL ultramicrotome equipped with a CR2000 cryochamber (RMC, Tuscon,AZ, USA)), collected on cryoprotectant droplets (PBS containing 2.3M sucrose) andtransferred to Formvar coated grids. The grids containing the cryosections were stored (30min to 5 h, section side down) on PBS containing 5% fetal calf serum (PBS-FCS, 4°C) priorto immunolabeling. This procedure removes the sucrose and blocks potential non-specificantibody binding sites.2.7 EVIMUNOGOLD LABELING OF CRYOSECTIONS 2.7.1 Immunogold labeling of PMNs in rabbit lungTo determine the effect of S. pneumoniae and E. coli endotoxin instillation on L-selectinand CD18 immunoreactivity on rabbit PMNs, lung cryosections were immunogold labeled atroom temperature with monoclonal antibodies (MAbs) against L-selectin and CD18.Cryosections that had been incubated in PBS-FCS (Section 2.6.3) were immunolabeled witheither the primary mouse MAb DREG-200 (against L-selectin) or MAb 60.3 (against CD18)44at a concentration of 0.01 mg/ml in PBS-FCS for 30 min. Additional cryosections were cutand immunolabeled with non-immune mouse IgG (0.01 mg/ml in PBS-FCS) obtained from theSigma Chemical Company (St. Louis, MO, USA) in place of the primary MAb for estimationof non-specific gold label. Cryosections were then washed in PBS-FCS (15 min) andimmunolabeled with a secondary polyclonal goat-anti-mouse IgG (whole molecule) conjugatedto 10 nm colloidal gold (Sigma Chemical Company, St. Louis, MO, USA) diluted in PBS-FCS(A52o = 0.1) for 30 min.2.7.2 Immunogold labeling of mouse lungTo determine the effect of S. pneumoniae and E. coli endotoxin instillation on ICAM-1immunoreactivity on endothelial, Type I, and Type II cells in the mouse lung, cryosectionswere immunogold labeled at room temperature with a MAb against ICAM-1. Cryosections thathad been incubated in PBS-FCS (Section 2.6.3) were immunolabeled with the primary ratmonoclonal antibody YN1/1.7.4 (against mouse ICAM-1) at a 1:100 dilution of the hybridomasupernatant in PBS-FCS for 30 min. To assess non-specific gold labeling, additionalcryosections were incubated with non-immune rat IgG (0.01 mg/ml in PBS-FCS) obtained fromthe Sigma Chemical Company (St. Louis, MO, USA) in place of the MAb. Cryosections werethen washed in PBS-FCS (15 min) and immunolabeled with a secondary polyclonal goat-anti-ratIgG (H + L) conjugated to 10 nm colloidal gold (Amersham Canada Limited, Oakville, Ont.,Canada) diluted in PBS-FCS (A520 = 0.1) for 30 min.2.7.3 Immunogold labeling of human LRPTo determine whether the topographical distribution of L-selectin is different than that45of CD18 and total sLex, ultrathin cryosections of glutaraldehyde fixed human LRP wereimmunogold labeled at room temperature with MAbs against L-sele,ctin, CD18 and sLex.Cryosections of human LRP that had been incubated in PBS-FCS (Section 2.6.3) wereimmunolabeled with mouse monoclonal antibodies MAb DREG-200 (against L-selectin, mouseIgG), MAb 60.3 (against CD18, mouse IgG), or MAb CSLEX-1 (against sLex, mouse IgM)at a concentration of 0.01 mg/ml in PBS-FCS for 30 minutes. To assess non-specific goldlabeling, additional cryosections were cut and immunolabeled with non-immune mouse IgG(0.01 mg/ml in PBS-FCS) or the non-immune murine IgM myeloma protein TEPC-183 (SigmaChemical Company, St. Louis, MO, USA) in place of the primary MAb. Cryosections werethen washed in PBS-FCS (15 min) and immunolabeled with a secondary polyclonal goat-anti-mouse IgG conjugated (whole molecule) to 10 nm colloidal gold (A520 = 0.1, Sigma ChemicalCompany (St. Louis, MO)) in PBS-FCS for 30 min.2.7.4 Immunogold labeling of transmigrating human PMNsTo investigate the possibility that L-selectin downmodulation is coincidental with PMNadherence to an activated endothelium, cryosections of PMNs transmigrating an IL-1 stimulatedHUVEC monolayer were immunolabeled at room temperature with mouse MAb DREG-200(against L-selectin) at a concentration of 0.01 mg/ml in PBS-FCS for 30 min. To assess non-specific gold labeling, additional cryosections were cut and immunolabeled with non-immunemouse serum (0.01 mg/ml in PBS-FCS) in place of the primary MAb. Cryosections were thenwashed in PBS-FCS (15 min) and immunolabeled with a secondary polyclonal goat-anti-mouseIgG (whole molecule) conjugated to 10 nm colloidal gold (A520 = 0.1, Sigma ChemicalCompany (St. Louis, MO)) in PBS-FCS for 30 min.462.7.5 Post-fixation and contrast enhancement of labeled cryosections Thirty minutes after the incubation with the secondary polyclonal antibody-goldconjugate (Sections 2.7.2 to 2.7.4), all cryosections were washed in PBS (10 min), fixed in 1%glutaraldehyde in PBS (10 min) to immobilize the immunogold complexes (G. Griffiths,personal communication), washed in distilled water (20 min) and finally embedded andcontrasted with 1.8% methylcellulose (25 cps, Sigma Chemical Company, St. Louis, MO,USA) containing 0.3% uranyl acetate (Polysciences, Warrington, PA, USA) in distilled water(10 min, 4°C).2.8 IM1VIUNOELECTRON MICROSCOPY2.8.1 Immunogold quantitation of L-selectin and CD 18 on rabbit PMNsAn analysis of the surface immunoreactivity of L-selectin and CD18 on rabbit PMNswas carried out using a systematic, random sampling procedure. A minimum of two and amaximum of eight lung tissue blocks were randomly selected from the sixteen blocks that wereavailable for each rabbit. These blocks were cryosectioned (Section 2.6), immunolabeled(Section 2.7), and viewed at a primary magnification of 1,650X on a Philips EM400transmission electron microscope. At this magnification, gold particles were not visible butit was possible to identify colloidal carbon blackened areas and to recognize PMNs within theblood microvasculature, interstitium, and airspace of pneumonic lung tissue. Thus, withoutany knowledge of the gold labeling density on the cells, the first ten PMNs that wereencountered in each lung region (microvasculature, interstitium, and airspace) werephotographed at a magnification of 10,000-12,500X.47Gold particles were counted directly from the photographic negatives (magnification10,000-12,500X). Each negative was placed on a light table and gold particles were viewedthrough a photographer's loupe (8X, Agfa) and manually counted. Gold particles were onlycounted on lengths of free plasma membrane; PMN plasma membrane surfaces in close contactwith other cells or tissue components were not counted because of the potential for physicalexclusion of the immunogold reagents (i.e. MAbs and colloidal gold). The plasma membranelengths were measured on photographically enlarged (27,500X) prints using a digitizing tablet(SummaSketch II, Summagraphics, Seymour, CT, USA) connected to an IBM-compatiblepersonal computer (Packard Bell, Model PB686, Korea) running Bioquant System IV software(R & M Biometrics, Nashville, TN, USA). The digitizing tablet was exactly calibrated on adaily basis using a similarly enlarged photographic image of an electron microscope diffractiongrating. As an additional check for measurement accuracy and reproducibility, a randomlyselected cell profile that had been previously measured was measured again at the beginningof each session. The difference between the two measurements was always less than 1%.The gold label density on each PMN profile examined was calculated by dividing thenumber of gold particles counted by the length of plasma membrane examined. The amountof gold that was specific for L-selectin and CD18 was calculated by subtracting the amount ofnon-specific gold label on rabbit PMNs labeled with non-immune mouse IgG (0.3 goldparticles/Am in the L-selectin study, for all PMNs; 0.1 gold particles/Am in the CD18 study,for all PMNs) from the amount of gold label on rabbit PMNs labeled with MAbs DREG-200and 60.3. The change in L-selectin and CD18 immunoreactivity on the PMN plasmamembrane, relative to the amount of specific gold label on control intravascular PMNs, wasdetermined using a least squares estimate for the mean specific gold label in a given group.48The fold difference was taken to be the ratio of a treatment group's least squares estimate tothe control group's least squares estimate (Mood et al., 1974).2.8.2 Immunogold quantitation of ICAM-1 in mouse lungAn analysis of the surface immunoreactivity of ICAM-1 on mouse pulmonary endotheliaand epithelia was carried out using a systematic, random sampling procedure. Two to threelung tissue blocks were randomly selected from the twenty blocks that were available for eachmouse. The blocks were cryosectioned (Section 2.6), immunolabeled (Section 2.7), andviewed at a primary magnification of 1,650X on a Philips EM400 transmission electronmicroscope. At this magnification, gold particles were not visible but it was possible toidentify regions of lung tissue in which the alveolar wall was not twisted or obscured byoverlapping sections. In normal mice (no tracheal instillations), this process allowed randomselection of alveolar wall segments without apriori knowledge of the gold labeling density. Inmice that had received tracheal instillates, the lung regions were considered for goldquantitation only if colloidal carbon were detected in the alveolar airspace or within alveolarmacrophages.Gold particles were manually counted at 125,000X magnification using the electronmicroscope. At this magnification, the microvascular endothelial and alveolar type I and IIepithelial cell surfaces were identified and gold particles were only counted on the plasmamembrane lengths that were unobstructed and not in close contact with other cells or tissuecomponents. This ensured that the cell surface being counted had been fully exposed to theimmunogold reagents (i.e. MAbs and colloidal gold).Enlarged (27,500X) photographic prints were made from the negatives and taped49together to construct photographic montages of each alveolar wall segment. Four to twelvedifferent alveolar wall segments were photographed from each mouse. The plasma membranelengths that had been gold counted in the electron microscope were measured on a calibrateddigitizer connected to a personal computer (Section 2.8.1). Within each alveolar wall segment,the total length of plasma membrane was recorded for both the pulmonary capillaryendothelium and the alveolar epithelium (Type I and II pneumocytes).Data from the photographic montages were combined and the gold label density on eachcell type was calculated by dividing the total number of gold particles counted by the totallength of plasma membrane examined. The amount of gold label that was specific for ICAM-1was calculated by subtracting the non-specific gold label obtained with non-immune rat IgG(0.2 gold particles/Am plasma membrane, for all endothelial and epithelial cells) from the goldlabel obtained with MAb YN1/1.7.4. The change in ICAM-1 immunoreactivity on endothelialand epithelial cell surfaces, relative to the amount of specific gold label on control mouseendothelial and epithelial cell surfaces, was determined using a least squares estimate for themean specific gold label in a given group. The fold difference was taken to be the ratio of atreatment group's least squares estimate to the control group's least squares estimate (Moodet al., 1974).2.8.3 Immunogold localization of L-selectin in human LRPTwo to three LRP-agarose blocks were randomly selected, out of a possible sixteen,from each of the three human subjects. These blocks were cryosectioned (Section 2.6),immunolabeled (Section 2.7), and viewed on a Philips EM400 transmission electronmicroscope. Approximately twenty PMNs were examined from each individual subject for50each MAb studied. Representative photomicrographs were taken at a primary magnificationof 10,000-12,500X.2.8.4 Immunogold localization of L-selectin on transmigrating human PMNsSix tissue blocks were randomly selected from each of the two PMN transmigrationexperiments. These blocks were cryosectioned (Section 2.6), immunolabeled (Section 2.7),and viewed on a Philips EM400 transmission electron microscope. Every PMN that wasencountered was photographed at a primary magnification of 10,000- to 12,500X.2.9 STATISTICAL ANALYSIS 2.9.1 Determination of the minimum number of PMN profiles required to accurately estimatethe mean gold label density within an animalTo determine whether a sample of ten PMNs from each lung region accuratelyrepresents an animal's response to the treatment, it was necessary to perform a theoreticalcalculation. It was assumed that the ten PMNs examined within a particular lung region (e.g.intravascular, interstitial, or airspace) represent a random sample of the total number(population) of PMNs within that lung region. An important concept in statistics (Glantz,1987; Zar, 1984) is that the standard error (SE) associated with the sample mean decreases,relative to the standard deviation (SD) of the population, as the sample size (n) increasesaccording to the equationSE= SDVi2EFFECT OF MULTIPLE SAMPLES ONTHE ACCURACY OF THE MEAN6^10^15^25Number of samples51This equation can be re-arranged to yieldSE 1SDThe magnitude of SE/SD decreases as a function of the square root of n. Therefore, anincrease in the sample size causes a decrease in the magnitude of SE/SD which is indicativeof an increase in the accuracy with which the sample mean reflects the population mean. Thisis graphically illustrated below:The ratio of SE/SD decreases (increasing accuracy of the mean) rapidly from 1.0 to 0.3as the sample size increases from 1 to 10. Because the standard error is influenced by thesquare root of the sample size, doubling the sample size to 20 would only marginally improvethe accuracy of the mean (SE/SD ratio falls from 0.3 to 0.2). Actual gold labeling data valueswere used to confirm this prediction (data obtained from the L-selectin study of intravascular52PMNs within the pneumonic lung of one rabbit 4h after a bronchial instillate of E. coliendotoxin) and the results are shown below:# PMN mean (gold SE SD Predicted Actual(n) particles/Am) SE/SD SE/SD10 1.2 0.4 1.3 0.3 0.320 1.1 0.3 1.3 0.2 0.2Notice that the mean, SE, and SD change very little (or not at all) as a result ofincreasing the number of PMNs (n) from 10 to 20. Importantly, the actual and predictedSE/SD ratios are in agreement. Based on these findings, an examination of 10 PMNs fromeach lung region was considered sufficient and appropriate for estimating each animal'sresponse to the treatment.2.9.2 Determination of intra- and interobserver error associated with goldparticle counts To determine the intraobserver error associated with the counting of gold particles,20 photographic negatives of L-selectin and CD18 immunogold labeled rabbit PMNs wererandomly selected. The gold particles on the cell profiles were re-counted and these newcounts were compared with the original counts using a least squares linear regression.Correlation analysis of these data resulted in an r value of 0.98 (P < 0.0001). This resultshowed that gold particle counts obtained by this observer (AB) were highly reproducible.To determine the interobserver error associated with the counting of gold particles,53a separate set of 20 photographic negatives of L-selectin and CD18 immunogold rabbitPMNs were randomly selected. The second observer (MK) was asked to count the goldparticles on the cell profiles and then these counts were compared with those of the firstobserver (AB) using a least squares linear regression. Correlation analysis of these dataresulted in an r value of 0.91 (P < 0.0001) that showed that the gold particle countsbetween observers were highly correlated.2.9.3 Statistical analysis of L-selectin and CD18 immunoreactivity on rabbit PMNsParametric statistical methods were used to analyse the data; the data aresummarized as mean + SD. There are two underlying assumptions for the use ofparametric statistics. The first assumption is that the variance between the groups must beapproximately equal (Glantz, 1987; Zar, 1984) and therefore, in this study, a mathematicaltransformation was performed on each data set as recommended by Zar (1984). For the L-selectin study, each data point (x) was transformed by the mathematical expressionwhereas, a log transformation was performed on the CD18 data set. The secondassumption is that the data are normally distributed. In this study, The transformed datasets were checked for normality using a quantile plot. A quantile plot displays the relativefraction of data versus the response values. By way of illustration, the quantile plot of thetransformed CD18 data is shown on the next page:QUANTILE PLOT: TRANSFORMED CD 18 DATA1^wNENfw0.84I-4' 0.6u_0z0E 0.44w1,_0.2^ )1(A)1(Ao0.3^-0.2^-0.1^0.0^0.1^0.2^0.3(ANIMAL RESPONSE) - (GROUP MEAN)54In this graphical analysis, the exposure group mean was subtracted from eachanimal's response. This forces the data to consist of a set of responses all centred about azero mean value. If the entire data set, from all of the groups, is normally distributed thena quantile plot will display an "S-shaped" or "ogive" curve, and this quantile plot is clearlyS-shaped (Wilkinson, L. 1990).In both the L-selectin study and the CD18 study, a one-way ANOVA was used todetect differences in specific gold label on PMNs in the various regions (intravascular,interstitial, airspace, contralateral) of the lungs of rabbits that received no instillate, S.pneumoniae, or E. coli endotoxin. Statistical significance for differences between groupswas accepted at P < 0.05., with corrections made for multiple comparisons by the DuncanMultiple Range test. Calculations were done using SAS version 6.03, (SAS Institute, Inc.,Cary, NC) on a 386 computer with a numeric coprocessor.552.9.4 Statistical analysis of ICAM-1 immunoreactivity in the mouse lungThe data are summarized as mean + SD. To make use of parametric statisticaltests, the variance between the groups was made approximately equal by a mathematicaltransformation of each data set as recommended by Zar (1984). To compare ICAM-1immunoreactivity on different cell types (endothelial, Type I, and Type II) within thecontrol mice that received no instillate, each data point (x) was transformed using themathematical expression1 (x+1 . 0 )whereas, the equation used to compare ICAM-1 immunoreactivity on Type I pneumocytesafter different exposures wasFinally, the mathematical transformation used to compare ICAM-1 immunoreactivity onendothelial cells, as well as ICAM-1 immunoreactivity on Type II pneumocytes, afterdifferent instillate exposures was^1 (x+0 .5)A quantile plot (see Section 2.9.3) of each transformed data set confirmed that each set wasnormally distributed (Wilkinson, L. 1990).Four different one-way ANOVAs were used to detect overall differences in specificgold particles/gm plasma membrane. The first ANOVA tested for differences betweenendothelial, Type I and Type II pneumocytes in control mice. The second ANOVA testedfor differences between endothelial cells exposed to different instillates. The third ANOVAtested for differences between Type I pneumocytes exposed to different instillates. Finally,the fourth ANOVA tested for differences between Type II pneumocytes exposed to different56instillates. Because there is an increased chance of a Type I error when using "multiple"ANOVAs, a Bonferroni-like correction was employed to minimize this possibility. As aresult, the overall statistical significance for differences between groups was accepted at P< 0.05/4 (i.e. P < 0.0125) with corrections made for multiple comparisons by the DuncanMultiple Range test. Calculations were done using SAS version 6.03, (SAS Institute, Inc.,Cary, NC) on a 386 computer with a numeric coprocessor.57CHAPTER THREERESULTS3.1 ELECTRON MICROSCOPIC OBSERVATIONS OF PMNs IN RABBIT LUNG3.1.1 Morphology of intravascular PMNs ^  583.1.2 Morphology of interstitial PMNs  583.1.3 Morphology of airspace PMNs  593.2 IMMUNOELECTRON MICROSCOPY OF PMNs IN RABBIT LUNG3.2.1 Summary of morphometric data on rabbit PMNs^  603.2.2 Localization and quantitation of L-selectin & CD 18 immunoreactivity in controlrabbit PMNs   603.2.3 Effect of pneumonia on L-selectin cell surface immunoreactivity on rabbit PM . 613.2.4 Effect of pneumonia on CD18 cell surface immunoreactivity on rabbit PMNs . . 623.3 LIGHT MICROSCOPY OF MOUSE LUNG TISSUE ^  633.4 IMMUNOELECTRON MICROSCOPY OF MOUSE LUNG TISSUE3.4.1 Summary of morphometric data on mouse lung tissue ^  653.4.2 Localization and quantitation of ICAM-1 immunoreactivityin control mouse lung ^  653.4.3 Effect of colloidal carbon instillation on ICAM-1 immunoreactivityin mouse lung  663.4.4 Effect of pneumonia on pulmonary endothelial ICAM-1 immunoreactivity . . . 663.4.5 Effect of pneumonia on alveolar epithelial ICAM-1 immunoreactivity ^ 673.5 IMMUNOELECTRON MICROSCOPY OF HUMAN PMNs3.5.1 Localization of L-selectin, CD18, and sLex on human PMNs in LRP ^ 673.5.2 Localization of L-selectin immunoreactivity on transmigrating human PMNs . . ^ 68583.1 ELECTRON MICROSCOPIC OBSERVATIONS OF PMNs IN RABBIT LUNG3,1.1 Morphology of intravascular PMNsThe general ultrastructural appearance of intravascular PMNs in the lungs of controlrabbits was not different from that of intravascular PMNs in rabbit lungs exposed to colloidalcarbon mixed with S. pneumoniae or E. coil endotoxin. The PMNs were easily distinguishedfrom the other circulating blood leukocytes (i.e. mononuclear cells, eosinophils and basophils)by their large segmented nucleus that contained characteristic accumulations of heterochromatinassociated with the inner nuclear membrane (Figure 3). The cytoplasm of the PMN, incomparison to the other blood leukocytes, was very electron dense and contained largenumbers of electron translucent, membrane-bound granules. Few mitochondrial profiles werepresent and the Golgi apparatus was small and rarely encountered.The shape of the intravascular PMNs was influenced by the size of the blood vessel.When the diameter of the blood vessel lumen was large, relative to the size of the PMN, thePMN was generally spherical in shape (Figures 3). The PMN was usually positioned at somedistance from the endothelium and its plasma membrane was frequently ruffled due to thepresence of numerous small microvillar projections. However, when the diameter of bloodvessel lumen was small, relative to the size of the PMN, the PMN had an elongated shape andthe contour of its plasma membrane was smooth and closely apposed to the endothelium liningthe vessel (Figure 4).3.1.2 Morphology of interstitial PMNsWithin the alveolar wall, an interstitial region separated the endothelium of the59microvasculature from the alveolar epithelium (Type I and II cells) that lines the alveolarspace. In the control rabbit lung, this interstitium consisted of a loose array of collagen fibres,elastin and connective tissue cells. Four hours after the instillation of S. pneumoniae or E. coilendotoxin mixed with colloidal carbon, PMNs were found within the interstitium of thepneumonic lung, but never within the interstitium of the contralateral lung. Interstitial PMNswere irregular in shape (Figure 5) and often had cytoplasmic projections (pseudopods) thatextended out into the connective tissue matrix. The cells were usually surrounded by anelectron translucent region in which no interstitial matrix components (collagen or elastin)could be detected.3.1.3 Morphology of airspace PMNsPolymorphonuclear leukocytes (PMNs) were never observed within the alveolar airspaceof control rabbits. However, airspace PMNs were frequently encountered in the pneumoniclungs, but not the contralateral lungs, of rabbits exposed to S. pneumoniae (Figure 6) or E. coilendotoxin (Figure 7) mixed with colloidal carbon for 4h. In both instances, the airspace PMNswere generally spherical in shape and showed extensive ruffling of the plasma membrane. Insome cases, colloidal carbon particles were associated with the plasma membrane surface, butonly rarely were these particles ingested. Unlike the endotoxin instillate, the streptococcalinstillate contained live bacteria. These bacteria were frequently ingested by the airspacePMNs and, in some cases, the PMNs contained so many phagocytosed bacteria that the PMNcytoplasm was almost entirely obscured (Figure 8). This contrasts sharply with the behaviourof the resident alveolar macrophages. Alveolar macrophages ingested more colloidal carbonthan the airspace PMNs and, in the case of the streptococcal instillate, they showed no60evidence of bacterial phagocytosis (Figure 9).3.2 IMMUNOELECTRON MICROSCOPY OF PMNs IN RABBIT LUNG3.2.1 Summary of morphometric data on rabbit PMNsThe total number of gold particles counted was 12,617 in the L-selectin study and69,578 in the CD18 study (n=9 rabbits). The length of PMN plasma membrane examined was5.0 mm in the L-selectin study and 3.4 mm in the CD18 study (n=9 rabbits). For bothstudies, the mean length of plasma membrane examined on intravascular PMNs tended to beless than that examined on interstitial and airspace PMNs (Tables 1 and 2). This finding mayrelate to the observation that intravascular PMNs were occasionally in close contact with theendothelium (Figure 4). As mentioned earlier (Section 2.8.1), those portions of the PMNmembrane in close contact with the endothelium were not measured and thus, shorter lengthsof plasma membrane were recorded for these PMNs. Alternatively, if degranulationaccompanies PMN emigration then this may increase the surface area and the surface:volumeratio of interstitial and airspace PMNs.3.2.2 Localization and quantitation of L-selectin & CD18 immunoreactivity in control rabbitPMNsL-selectin immunoreactivity was detected only on the plasma membrane of PMNs; therewas no detectable intracellular pool of L-selectin. Two distinct topographical distributions ofgold particles were observed on the cell surface of L-selectin positive PMNs. When the PMNsurface displayed microvillar processes, L-selectin immunoreactivity was consistently localized61to these microvilli and not to the "flat" regions of the plasma membrane between the microvilli(Figure 10). When the PMN surface lacked microvillar processes and was relatively smooth,the gold particles appeared to be grouped into small clusters and these clusters were detectedover the entire surface of the PMN (Figure 11).By comparison, only one topographical distribution of gold particles was observed onCD18 positive rabbit PMNs. The gold particles were found along the entire length of theplasma membrane and showed no preference for microvillar processes. In addition, unlike theimmunolocalization of L-selectin, CD18 immunoreactivity was frequently detected in many ofthe cytoplasmic granules within the PMN (Figure 12).On control rabbit PMNs, L-selectin immunoreactivity was much more variable thanCD18 immunoreactivity (5.6 + 3.2 (SD) vs 8.6 + 1.8 gold particles/pm, in three rabbits); thecoefficient of variation for L-selectin and CD18 immunoreactivity was 57% and 21%respectively.3.2.3 Effect of pneumonia on L-selectin cell surface immunoreactivity on rabbit PMNsFour hours after the instillation of S. pneumoniae, L-selectin immunoreactivity onintravascular (Figures 13 and 14) and airspace PMNs (Figure 15) was reduced by 78% and99%, respectively, when compared to control intravascular PMNs (Table 3, each P < 0.05).The L-selectin immunoreactivity on the pneumonic intravascular PMNs was extremely variable;the coefficient of variation was much greater compared to control PMNs (167% vs. 57%). L-selectin positive intravascular PMNs were detected in only one of the three pneumonic rabbitlungs examined. From this lung, two tissue blocks were examined and all of the L-selectinpositive cells came from one tissue block (Figure 13); the other tissue block was L-selectin62negative. Of the 30 PMNs examined from the three pneumonic lungs, 24 of them showednegligible ( < 0.3 gold particles/Am) or no L-selectin gold label (Figure 14). Because L-selectin immunoreactivity was negligible on intravascular PMNs from two of the three rabbits,L-selectin immunoreactivity was only determined on interstitial PMNs from the pneumoniclung of the one rabbit which had L-selectin positive intravascular PMNs. Nine of the 10interstitial PMNs examined came from the one tissue block that had L-selectin positiveintravascular PMNs. Despite a high coefficient of variation (367%) L-selectinimmunoreactivity on these interstitial PMNs (Figure 16) was significantly less than that onintravascular PMNs from the same tissue blocks (0.3 + 1.1 vs. 2.5 + 2.7, in 10 PMNs,unpaired t test, P < 0.05). Streptococcal pneumonia in the right lung had no effect on L-selectin immunoreactivity on intravascular PMNs in the contralateral left lung (Figure 17); L-selectin immunoreactivity on these contralateral PMNs was not different than that on controlintravascular PMNs (Table 3).Four hours after the instillation of E. coil endotoxin, L-selectin immunoreactivity onintravascular PMNs within the pneumonic right lung or contralateral left lung (Figure 18) wasnot different from that on control intravasacular PMNs (Table 3). However, L-selectinimmunoreactivity on PMNs that had emigrated into the interstitium (Figure 19) and alveolarairspace (Figure 20) was reduced by 86% and 98%, respectively, compared to intravascularcontrol PMNs (Table 3, each P < 0.05).3.2.4 Effect of pneumonia on CD18 cell surface immunoreactivity on rabbit PMNsFour hours after the instillation of S. pneumoniae, CD18 immunoreactivity on PMNsin each region of the pneumonic lung was significantly greater than that on control63intravascular PMNs (Table 4). Within the pneumonic lung, CD18 immunoreactivity onintravascular (Figure 21) and interstitial PMNs (Figure 22) was virtually identical and 2.6 timesgreater than that on control intravascular (Figure 12) PMNs (each P < 0.05). The maximumincrease in CD18 immunoreactivity was observed on those PMNs that had emigrated into thealveolar airspace (Figure 23). The level of CD18 immunoreactivity on these PMNs was 5.5times greater than that on the control intravascular PMNs (Table 4, P < 0.05). In addition,the gold label density on these airspace PMNs was significantly greater than that on theintravascular and interstitial PMNs (each P < 0.05). CD18 immunoreactivity on intravascularPMNs in the contralateral lung (Figure 24) was not different from that on control intravascularPMNs (Table 4; compare Figures 12 and 24).Instillation of E. coil endotoxin, in contrast to S. pneumoniae, did not result inupmodulation of CD18 immunoreactivity on the intravascular PMNs compared to controlintravascular PMNs (Table 4, P = 1.0). However, CD18 immunoreactivity was increased 2-fold on intersitial PMNs (Table 4, P < 0.05; Figure 25) and 3.2-fold on airspace PMNs(Table 4, P < 0.05; Figure 26) when compared to that on control intravascular PMNs.Because CD18 immunoreactivity was similar on intravascular PMNs in the pneumonic lungregion and in the control lung region, CD18 immunoreactivity on PMNs in the contralaterallung was not quantitated.3.3 LIGHT MICROSCOPY OF MOUSE LUNG TISSUELight microscopy showed that PMNs did not emigrate into the alveolar airspace duringthe 4 or 24h period following intratracheal instillation of a sterile saline solution of colloidalcarbon. Lung tissue sections from mice that received colloidal carbon in saline for 4h (Figures6427 and 28) or 24h (Figures 29 and 30) did not differ from those of control mice that receivedno instillates except for the presence of ingested carbon particles within the cytoplasm of thealveolar macrophages.In contrast, large numbers of PMNs were found in the alveolar airspace 4h afterintratracheal instillation of S. pneumoniae mixed with colloidal carbon (Figures 31 and 32).In some alveoli, an equally large number of extravasated red blood cells could also be seen(Figure 32). The presence of red blood cells in the airspace was unique to streptococcalpneumonia as it was never observed in the lungs of mice instilled with E. coil endotoxin.Streptococcal bacteria were frequently ingested by airspace PMNs, but it was only the alveolarmacrophages that ingested the colloidal carbon. Interstitial PMNs were frequently observedin the connective tissue surrounding the bronchial arteries (Figure 31), but it could not bedetermined by light microscopy whether PMNs had also emigrated into the interstitium of thealveolar walls.By comparison, PMNs were found in the alveolar airspace 24h after intratrachealinstillation of E. coli endotoxin and colloidal carbon (Figures 33 and 34). In general, unlikethe airspace PMNs, the alveolar macrophages ingested large amounts of colloidal carbon.Extravascular PMNs were observed in the bronchovascular interstitium (Figure 33), but onceagain, the resolution of the microscope was insufficient to determine whether interstitial PMNswere also present within the alveolar walls.653.4 IMMUNOELECTRON MICROSCOPY OF MOUSE LUNG TISSUE3.4.1 Summary of moThometric data on mouse lung tissueIn the ICAM-1 study of mouse lung, 75% of the total number of gold particles (41,166)counted were localized to the Type I pneumocytes (Table 5). In total, 13 mice were examinedand the cumulative length of plasma membrane examined in this study for the endothelial, TypeI, and Type II cells was 2.5 mm, 3.6 mm, and 2.0 mm, respectively. The mean length ofplasma membrane examined on each cell type in this study was: endothelial, 192 + 38 (SD)Am; Type I, 263 + 116 Am; Type II, 158 + 36 Am.3.4.2 Localization and quantitation of ICAM-1 immunoreactivity in control mouse lungICAM-1 was constitutively expressed on the lumenal surfaces of both pulmonarycapillary endothelial (Figures 35 and 36) and alveolar epithelial cells (Figure 37); there wasnegligible ICAM-1 immunoreactivity on the ablumenal and lateral surfaces of these cells andthere were no detectable intracellular pools of ICAM-1. Epithelial ICAM-1 immunoreactivitywas primarily limited to the Type I pneumocytes (Figure 37). By comparison, very few goldparticles were detected on the surface of Type II pneumocytes (Figures 37 and 38). There wasno obvious pattern to the distribution of these gold particles on either endothelial or Type Icells. Quantitation of the gold label density showed that ICAM-1 immunoreactivity on TypeI pneumocytes was 22 times greater than that found on the endothelial cells (10.6 + 3.0 (SD)vs. 0.5 + 0.1 gold particles/Am, Bonferroni-adjusted P < 0.0125) and 212 times greater thanthat found on the Type II pneumocytes (Table 6, Bonferroni-adjusted P < 0.0125). The largedifference in Type I and Type II pneumocyte ICAM-1 immunoreactivity was best appreciated66at the cellular junction between these cells, where the rarity of gold particles on the Type IIpneumocyte contrasted sharply with the large numbers of gold particles on the Type Ipneumocyte (Figure 38).3.4.3 Effect of colloidal carbon instillation on ICAM-1 immunoreactivity in mouse lungAn examination of lungs 4h and 24h after the intratracheal instillation of colloidalcarbon showed, that in the absence of an inflammatory stimulus (S. pneumoniae or E. coilendotoxin), there was no significant change in ICAM-1 expression on pulmonary endothelialcells or Type I pneumocytes (Table 6). By comparison, 24h after, but not 4h after, colloidalcarbon instillation, ICAM-1 immunoreactivity on Type II pneumocytes was 52 times greaterthan that found on control Type II pneumocytes (Table 6, Bonferroni-adjusted P < 0.0125).3.4.4 Effect of pneumonia on pulmonary endothelial ICAM-1 immunoreactivityFour hours after intratracheal instillation of S. pneumoniae and colloidal carbon, ICAM-1 immunoreactivity on pulmonary capillary endothelial cells was not significantly different fromthat on control pulmonary endothelium (Table 6). Twenty four hours after intratrachealinstillation of E. coli endotoxin and colloidal carbon into the lungs of mice, ICAM-1immunoreactivity on pulmonary capillary endothelial cells (Table 6, Figure 39) was estimatedto be 4.2 times greater than that on pulmonary endothelial cells of control mice and thisincrease was statistically significant using the Duncan Multiple Range test. However, theoverall Bonferroni-adjusted P value for the ANOVA was 0.017 and this exceeded theacceptable level of significance (P < 0.0125). Zar (1984) points out that it is not an actualrequirement that multiple comparison testing be performed only if the ANOVA rejects a67multisample hypothesis of equal means. On the advise of a professional statistician (Dr. G.E.I.Smith, Consulting Statistician for the Canadian Wildlife Service, Delta, B.C.) it is best toreport that the 4.2-fold increase in ICAM-1 expression borders on statistical significance, ratherthan simply dismissing the observation as statistically insignificant.3.4.5 Effect of pneumonia on alveolar epithelial ICAM-1 immunoreactivityNeither streptococcal nor endotoxin pneumonias had any significant effect on ICAM-1immunoreactivity on Type I pneumocytes. However, there was a downward trend towards adecrease in gold particle density on the surface of the Type I pneumocytes following thetracheal instillation of S. pneumoniae or E. coil endotoxin and colloidal carbon (Table 6).Four hours after intratracheal instillation of S. pneumoniae mixed with colloidal carbon,ICAM-1 immunoreactivity on Type II pneumocytes (Figure 40) was 42 times greater than thaton Type II pneumocytes in the control mouse lung (Table 6, Bonferroni-adjusted P < 0.0125).Similarly, but more striking, 24h after tracheal instillation of colloidal carbon mixed with E.coil endotoxin, ICAM-1 immunoreactivity on Type II pneumocytes (Figure 39) was 178 timesgreater than that on Type II pneumocytes in the control mouse lung (8.9 + 4.8 (SD) vs. 0.1+ 0.1 gold particles/pm, in 3 mice, Bonferroni-adjusted P < 0.0125).3.5 1MMUNOELECTRON MICROSCOPY OF HUMAN PMNs 3.5.1 Localization of L-selectin. CD18. and sLex on human PMNs in LRPAn electron microscopic examination of fixed and cryosectioned human LRP showedthat most of the leukocytes were PMNs. These PMNs were generally spherical in shape, but68their surface was frequently ruffled due to the presence of numerous small microvillarprojections. Immunogold labeled cryosections showed that L-selectin immunoreactivity on thePMN cell surface was consistently localized to microvillous processes (Figure 41). Only asmall fraction of L-sele,ctin labeling was identified in "flat" regions of the PMN surface, andno appreciable labeling was observed in cytoplasmic granules. This preferential distributionof L-selectin was apparent in all human PMNs examined (60 from three different individuals).In contrast, the immunoreactivity of total sLex and CD18 appeared randomly distributed overthe plasma and granule membranes of the PMNs from all three donors (Figures 42 and 43,respectively).3.5.2 Localization of L-selectin immunoreactivity on transmigrating human PMNsAn electron microscopic examination of fixed and cryosectioned IL-1 stimulatedendothelial monolayers showed PMNs in various stages of transmigration. L-selectinimmunoreactivity on human PMNs that had not yet contacted the IL-1-stimulated endothelialsurface (Figures 44 and 45) was similar to that observed in the human LRP preparations(Figure 41). As before, the L-selectin immunoreactivity was preferentially localized to themicrovillar processes and no appreciable gold label was observed in the cytoplasmic granules.Adherent PMNs, those that had not yet penetrated the endothelial monolayer but were in closecontact with its surface, were positive for L-selectin immunoreactivity. However, L-selectinimmunoreactivity was restricted to the portion of the PMN surface that was not in contact withthe endothelium (Figure 46); L-selectin immunoreactivity could not be demonstrated on thePMN surface that was in close contact with the endothelium.Several PMNs were also observed in the process of transmigrating the endothelium69(Figures 47 to 49) where the advancing portion (pseudopod) of the PMN had just penetratedthe endothelial monolayer. Qualitatively, fewer gold particles were present on the surface ofthe advancing pseudopod (Figure 49) relative to that on the rest of the cell (Figure 48). L-selectin immunoreactivity was still detected on PMNs that had completely transmigrated theendothelium (Figures 50 and 51). These PMNs accumulated between the endothelium and itsbasal lamina (Figure 51). L-selectin immunoreactivity on these cells was qualitatively less thanthat on non-migrated PMNs, but greater than that on PMNs labeled with non-specific mouseIgG (Figure 52).70CHAPTER FOURDISCUSSION4.1 EXPRESSION OF PMN, ENDOTHELIAL, AND EPITHELIAL ADHESIONMOLECULES DURING PNEUMONIA4.1.1 Alterations in the expression of PMN L-selectin and CD18 during pneumonia . . 714.1.2 Alterations in the expression of ICAM-1 during pneumonia ^ 794.2 OBSERVATIONS ON L-SELECTIN4.2.1 The topographical distribution of L-selectin on the PMN surface ^ 864.2.2 Downmodulation of L-selectin during PMN transmigration  894.3 SENSITIVITY AND LIMITATIONS OF THE IMMUNOGOLD TECHNIQUE . . 91714.1 EXPRESSION OF PMN. ENDOTHELIAL AND EPITHELIAL ADHESIONMOLECULES DURING PNEUMONIA4.1.1 Alterations in the expression of PMN L-selectin and CD18 during pneumoniaThe first specific aim of this thesis was to determine whether L-selectin expressionis downmodulated and whether CD18 expression is upmodulated in response to stimulithat induce CD18-independent and CD18-dependent PMN emigration in the rabbit lung.The results clearly show that CD 18-independent PMN emigration towards a bronchial instillateof S. pneumoniae is associated with L-selectin downmodulation and CD18 upmodulation onintravascular, interstitial, and airspace PMNs. A similar alteration in the expression of L-selectin and CD18 is observed during CD 18-dependent PMN emigration towards a bronchialinstillate of E. coli endotoxin, but only after PMNs emigrated into the interstitium. Regardlessof the mechanism of PMN emigration, the change in L-selectin and CD18 expression occursonly on PMNs within the pneumonic site. The evidence for this finding is that L-selectin andCD18 expression on PMNs within the pulmonary microvasculature of the non-pneumoniccontralateral lung is not different from control. This finding implies that PMN activation isa focal phenomenon and restricted to the site of acute inflammation.The DREG-200 and 60.3 MAbs were made against human L-selectin (Kishimoto et al.,1990) and human CD18 (Beatty et al., 1983; Wallis et al., 1986), respectively. On rabbitPMNs, the DREG-200 MAb (von Andrian et al., 1991) and the 60.3 MAb (Arfors et al.,1987, Price et al., 1987; Doerschuk et al., 1990a, 1990b) recognize a functional L-selectinequivalent and a functional CD18 equivalent, respectively. The rabbit glycoprotein recognizedby MAb 60.3 is GP150/GP85 and this glycoprotein is the functional equivalent of human72CD11/CD18 (Lindbom et al., 1990). Ultrastructural (Bainton et al., 1987; Jones et al., 1990a)and biochemical (Arnaout et al., 1984; O'Shea et al., 1984; Todd et al., 1985; Springer et al.,1986; Jones et al., 1988, 1990a, 1990b) studies conducted on unstimulated human PMNpreparations have shown that the majority of CD18 is stored as a latent intracellular granularpool (largely CD11b/CD18 and some CD11c/CD18) with lesser amounts being expressed onthe cell surface (CD11a/CD18 , CD11b/CD18 , and CD11c/CD18). During PMN activation,CD18 surface expression can increase 2-10 fold and fusion of the CD18-rich cytoplasmicgranules with the plasma membrane provides the mechanism for this increase (Arnaout et al.,1984; Bainton et al., 1987; Jones et al., 1988, 1990a, 1990b). In the present study, themajority of the CD18 in normal rabbit PMNs is localized within the intracellular granules,supporting the presence of a latent pool that can be mobilized to the surface during PMNactivation. Direct evidence of increased CD18 surface expression during PMN activation isshown by the increased numbers of CD 18-specific gold particles on the cell surface ofemigrating PMNs in both streptococcal and endotoxin pneumonias. Human PMNs showdecreased L-selectin expression during activation (Smith et al., 1991) and this is most likelydue to shedding of L-selectin from the cell surface (Kishimoto et al., 1989a). Importantly,during streptococcal and endotoxin pneumonias, the rabbit antigen recognized by the DREG-200 MAb is also downmodulated as PMNs are activated and CD18 expression increases.Collectively, these observations are consistent with the concept that the rabbit antigensrecognized by the DREG-200 and 60.3 MAbs are the functional equivalents of human L-selectin and CD 18, respectively.The present ultrastructural finding that the shape of intravascular capillary PMNs issometimes elongated agrees with the idea that PMNs must deform in order to pass through73narrow capillary segments (Hogg, 1987; Martin et al., 1987; Doerschuk et al., 1987; Downeyand Worthen, 1988). However, in capillaries with diameters greater than that of the PMN,the PMNs are generally spherical in shape and display microvillar projections. A strikingobservation regarding the microvilli is that L-selectin immunoreactivity is consistentlyassociated with these surface processes, rather than the flatter regions of the cell. Importantly,this finding may be common to other animal species as L-selectin expression on human PMNsis also associated with microvillar processes (Section 4.2). In the systemic circulation,spherically-shaped PMNs "roll" (marginate) along the vessel wall of inflamed post-capillaryvenules (Atherton and Born, 1972, 1973; Arfors et al., 1987; von Andrian et al., 1991; Leyet al., 1991; Tangelder and Arfors, 1991). The association of L-selectin with surfacemicrovilli may facilitate this rolling phenomenon (discussed in Section 4.2). However, thephysical constraints of narrow capillary segments probably inhibit PMN rolling and the presentfinding that elongated PMNs show very few microvillar processes supports this hypothesis.Of particular interest is the additional finding that L-selectin-specific gold particles are detectedin small clusters at discrete sites along the flattened PMN surface. This observation impliesthat L-selectin is not redistributed on the surface during PMN elongation, because if it were,then a more even distribution of gold particles would be expected. Whether microvilli reformat these L-selectin clusters once the PMN passes into a larger vessel (i.e. post-capillary venule)is unknown.During streptococcal pneumonia, L-selectin expression is downmodulatal and CD18expression is upmodulated on intravascular PMNs and this confirms that these PMNs areactivated (Kishimoto et al., 1989a; Jutila et al., 1989, 1990, 1991; Smith et al., 1991; vonAndrian et al., 1991). Importantly, these activated PMNs do not appear to circulate in74appreciable numbers as evidenced by the observation that PMNs within the contralateral lungexpress normal amounts of L-selectin and CD 18. These observations imply that L-selectin isnot involved in the retention of activated PMNs within the pulmonary microvasculature of thepneumonic region. Although CD18 expression is enhanced on these intravascular PMNs, itis unlikely to play a role in PMN retention based on the finding that PMN emigration is notinhibited by the anti-CD18 MAb 60.3 (Doerschuk et al., 1990a, 1990b). Retention of activatedPMNs in response to S. pneumoniae within the lung must involve CD18-independentmechanisms of adhesion and/or a decrease in PMN deformability that prevents the PMN frompassing through narrow capillary segments (Hogg, 1987; Doerschuk et al., 1989; Worthen etal., 1989).The activation of intravascular PMNs within the pulmonary microvasculature impliesthat PMNs within the capillary lumen of the pneumonic lung are being stimulated by the localrelease of soluble inflammatory mediators and cytokines (e.g. IL-8, LTB4, PAF, TNF-a, C3a,and C5a). This postulate assumes that the soluble stimuli are not washed away and diluted bycapillary blood flow. In support of this is the recent finding that capillary blood flow withinthe pneumonic region is reduced by 80% at 4h (Coxson et al., 1990; Doerschuk, Markos,Coxson, English, and Hogg. (Submitted)). Reduced capillary blood flow would potentiallyallow the soluble mediators to accumulate and activate the intravascular PMNs. Moreover,because blood flow is not entirely obliterated, it may continue to deliver unstimulated PMNsto the pneumonic region. This could explain the present finding that although most (80%) ofthe intravascular PMNs were L-selectin negative, 20% (6/30 cells) of these PMNs expressednormal levels of L-selectin. The mechanism of blood flow reduction is unknown, observationsmade in the present study suggest that it may be caused, in part, by a loss of microvascular75integrity. The evidence is that red blood cells and fibrin were frequently detected within thealveolar airspace and this suggests that there has been a gross disruption of the alveolarcapillary wall. Importantly, activated PMNs are less deformable (more "stiff") thanunstimulated PMNs (Worthen et al., 1989). The combined effects of no or low blood flow anddecreased PMN deformability may make it very difficult for activated PMNs to pass throughnarrow capillary segments and rejoin the circulation.In response to S. pneumoniae, emigrated airspace PMNs show increased CD18 surfaceexpression (5.5-fold greater than control intravascular PMNs) and this may contribute to thepresent finding that airspace PMNs ingest large numbers of bacteria. This increase in CD18expression is likely due to upregulation of CD11b/CD18 adhesion proteins onto the cell surfaceby specific granule fusion with the plasma membrane (Arnaout et al., 1984; Bainton et al.,1987; Jones et al., 1988, 1990a, 1990b). During the inflammatory response, streptococcalbacteria are likely to activate both the alternative (Winkelstein and Tomasz, 1978; Hummel etal., 1981, 1985) and classical (Loos et al., 1986) complement pathways. BecauseCD11b/CD18, in addition to binding ICAM-1, binds to the complement split product C3bi(reviewed by Wright et al., 1990), then binding of CD1 lb/CD18 to C3bi-opsonized bacteriashould greatly facilitate PMN phagocytosis (Wright and Meyer, 1986). Consistent with thishypothesis is the recently published in vitro finding that PMN phagocytosis of serum-opsonizedStreptococcus agalactiae is largely inhibited by the anti-CD 18 MAb R15.7 (Sherman et al.,1992). Significantly, PMN emigration in the rabbit lung towards S. agalactiae is also CD18-independent (Ibid). Collectively, these data imply that although CD18 expression is notnecessary for PMN emigration during streptococcal pneumnonia, it may be important tobacterial phagocytosis. It is curious that although the phagosomal membrane surrounding the76ingested bacteria is derived (presumably) from an invagination of the plasma membrane, it doesnot show CD18 immunoreactivity. This observation implies that CD18 is either excluded fromthe phagosomal membrane during its formation or, that fusion of the phagosome with aprimary lysosome results in enzymatic degradation of phagosomal-associated CD 18.Endotoxin pneumonia, in contrast to streptococcal pneumonia, is not associated with asignificant change in L-selectin expression on intravascular PMNs. However, PMNs thatemigrate into the interstitium show a significant decrease (P < 0.05) in L-selectin expression.These findings imply that L-selectin downmodulation is associated with transendothelialmigration and that it occurs only after the PMN adheres to the capillary endothelium. A recentstudy demonstrated that IL-1-stimulated endothelial cells express a surface membrane-associatedform of platelet activating factor (PAF) (Kuijpers et al., 1992a). Importantly, in vitro, thismembrane-associated PAF induces L-selectin downmodulation on transmigrating human PMNs(Ibid). It would be of great interest to determine whether the pulmonary capillary endotheliumexpresses membrane-associated PAF during endotoxin pneumonia in rabbits. This questioncould be addressed by immunogold labeling lung cryosections with a MAb directed againstPAF. However, such a study would require raising an anti-PAF MAb, because at present,none are commercially available.In the pulmonary circulation, PMN adherence to the endothelium may occur by eithera CD18-dependent or CD18-independent mechanism that is specific to the inflammatorystimulus (Doerschuk et al., 1990a, 1990b). S. pneumoniae elicits CD 18-independent PMNemigration into the alveolar space (Ibid). Yet, our results indicate that CD18 expression onthe PMN surface increases prior to emigration and increases further once the PMNs havereached the airspace in response to S. pneumoniae. In contrast, CD18 expression is not77increased on intravascular PMNs prior to CD 18-dependent emigration in response to E. coilendotoxin. Increased CD18 expression is only observed once the PMNs have left thepulmonary microvasculature and are present within the interstitium and alveolar airspace.Thus, the levels of CD18 expression on PMNs do not correlate with the function of thismolecule during PMN emigration in response to CD 18-independent and CD 18-dependentstimuli. This finding is not really surprising, because there have been several independent invitro studies showing that increased CD18 expression on the PMN surface is not a prerequisitefor adherence to endothelial cells (Vedder and Harlan, 1988; Lo et al., 1989a, 1989b;Schleiffenbaum 1989). Furthermore, there is increasing evidence that the existing CD18 onthe surface of the PMN is sufficient for adhesion and migration. Although both CD11a/CD18and CD11b/CD18 contribute to this process (reviewed by Smith, 1992), only CD11b/CD18shows increased surface expression following PMN activation (Arnaout et al., 1984). Manyin vitro studies have shown that newly expressed CD1 lb/CD18 on the surface of the neutrophilappears to be non-active (Wright and Meyer, 1986; Buyon et al., 1988; Philips et al., 1988,Nourshargh et al., 1989; Lo et al., 1989b). Activation of CD11b/CD18 into a competentadhesion receptor is thought to require a conformational change (Altieri and Edgington, 1988;Robinson et al., 1992). The mechanism(s) responsible for this conformational change islargely unknown, but it may involve a phosphorylation event (Wright and Meyer, 1986; Buyonet al., 1990). Alternatively, it may be induced by a lipid known as integrin modulating factor-1 (IMF-1) (Hermanowski-Vosatka et al., 1992). IMF-1 is transiently expressed on activatedPMNs and it induces an equally transient increase in binding avidity on CD11b/CD18 for thecomplement fragment C3biThe present finding that interstitial PMNs in both streptococcal- and endotoxin-induced78pneumonias express very little L-selectin implies that L-selectin is not important for PMNmigration through the interstitial matrix. This agrees with the generally held concept that L-selectin functions in the initial contact between the PMN and the inflamed vascular endothelium(Lewinsohn et al., 1987; Jutila et al., 1989, 1991; von Andrian et al., 1991; Ley et al., 1991).Interestingly, CD18 can serve as a receptor for Type I collagen (Monboisse et al., 1991) andin the present study, interstitial PMNs were often in close contact with fibrillar (Type I)collagen. Because PMN emigration towards S. pneumoniae is CD 18-independent (Doerschuket al., 1990a, 1990b), CD 18-mediated adherence to collagen can not be essential for PMNmigration through the interstitium. Conversely, because PMN emigration towards E.coliendotoxin is CD18-dependent, the possibility is raised that part of this CD18-dependency mayreside in a requirement for PMN adhesion to collagen during interstitial migration.In the systemic circulation, PMN emigration towards all acute inflammatory stimuliexamined to date is dependent upon the expression of CD18 adhesion molecules (Harlan et al.,1992). The reason why PMN emigration towards S. pneumoniae in the lung is CD18-independent is unknown, but it has been argued that the presence of macrophages is essentialto the development of the CD18-independent response (Mileski et al., 1990). In response toa peritoneal instillate of S. pneumoniae, PMN emigration is normally CD 18-dependent;however, if peritoneal macrophages are activated and their numbers increased prior to thedelivery of the instillate, then PMN emigration can proceed by a CD 18-independent mechanism(Ibid). This observation suggests that macrophage ingestion of streptococcal bacteria resultsin the release of chemotactic or proadhesive bacterial degradation products (Ibid). However,in the present study of the lung, this postulate is not supported because alveolar macrophagesrarely ingest streptococcal bacteria. Instead, it is the emigrated PMNs that ingest large79numbers of bacteria. It must be pointed out that these observations are only valid up to thefirst 4h following the instillation of S. pneumoniae. Whether macrophages ingest more bacteriaas the pneumonia develops was not determined in this study, however, the present findings areconsistent with Metchnikoff s (1905) early observations on acute inflammation. He showedthat PMNs in general are far more efficient at bacterial phagocytosis than macrophages. Infact, he remarked:"Macrophages can also ingest the bacteria of acute diseases, but, save inexceptional cases, their intervention is of little moment."[Metchnikoff, 1905].These observations suggest that S. pneumoniae activates alveolar macrophages by someother mechanism that does not involve bacterial phagocytosis. Although the present studyoffers no insight into the mechanism of streptococcal activation of alveolar macrophages, itseems likely macrophage activation is accompanied by the release of a monokine(s) that iseither chemotactic or proadhesive for PMNs (Mileski et al., 1990). This hypothesis could betested using the present rabbit model of streptococcal pneumonia. Rabbit lungs infected withstreptococcal organisms could be lavaged, and the recovered lavage fluid examined forchemotactic or proadhesive macrophage products that promote CD 18-independenttransendothelial migration in vitro.4.1.2 Alterations in the expression of ICAM-1 during pneumoniaThe second specific aim of this thesis was to determine, in the mouse lung, whetherICAM-1 expression is upmodulated in response to stimuli that are known to induce CD18-independent and CD18-dependent PMN emigration in the rabbit lung. The results clearlyshow that, in response to the CD18-independent stimulus, S. pneumoniae, capillary endothelialICAM-1 expression is not upmodulated during PMN emigration. Conversely, endothelial80ICAM-1 expression is upmodulated (a 4.2-fold increase over control levels) in response to theCD 18-dependent stimulus, E. coil endotoxin; however, this finding must be interpreted withcaution because the increase only borders on statistical significance. An additional new findingin the present study is that in normal lung, constitutive ICAM-1 expression is 22-fold greateron the alveolar epithelium compared to the capillary endothelium and the epithelial expressionis mainly restricted to the Type I pneumocytes. Interestingly, epithelial ICAM-1 expressionis not increased on Type I pneumocytes during streptococcal and endotoxin inducedpneumonias, but a significant increase in ICAM-1 expression is observed on Type IIpneumocytes.In response to S. pneumoniae, PMN emigration in the rabbit lung is known to be CD18-independent (Doerschuk et al., 1990a, 1990b); however, it is not known if emigration is alsoa CD 18-independent phenomenon in the mouse lung. Constitutive pulmonary capillaryendothelial ICAM-1 expression, as determined by the immunogold method, is very low andbarely above background. Moreover, capillary endothelial ICAM-1 expression is not increasedat 4h when significant numbers of PMNs are in the alveolar airspace. These observationsimply that constitutive endothelial ICAM-1 expression is very low and that increased ICAM-1expression is not required for PMN emigration. This is consistent with the concept of CD18-independent emigration, because ICAM-1 is an adhesive ligand for CD11a/CD18 andCD11b/CD18 (Rothlein et al., 1986; Smith et al., 1989; Diamond et al., 1990, 1991).Following cytoldne stimulation, both in vitro and in vivo, endothelial cells become moreadhesive for PMNs and a significant portion of this increased adhesiveness is due to an increasein ICAM-1 expression (Smith et al., 1988, 1989). However, regardless of whether PMNsemigrate by a CD 18-independent mechanism, the results clearly demonstrate that increased81ICAM-1 expression is not required for PMN diapedesis from the pulmonary microvasculatureduring streptococcal pneumonia in the mouse.In response to lipopolysaccharide (endotoxin), PMN emigration in the mouse lung isknown to be dependent upon the expression of CD18 (Rosen and Gordon, 1990). The presentstudy suggests, but does not statistically confirm, that ICAM-1 expression is increased 4.2-foldon pulmonary capillary endothelial cells (see Section 3.4.4). An increase in ICAM-1expression would be consistent with the concept of CD 18-dependent PMN emigration and,importantly, this increase is coincidental with the presence of PMNs in the alveolar space 24hafter the instillation of E. coli endotoxin. In vitro, ICAM-1 expression on endothelialmonolayers typically increases 6-fold over baseline following exposure to IL-1 or endotoxinand is nearly maximal by 4h (Smith et al., 1988, 1990); it remains elevated for at least 48hfollowing exposure to IL-1 (Luscinskas et al., 1991). Four hours after endothelial stimulation,PMN adhesion and transendothelial migration in vitro is maximal (Smith et al., 1988;Luscinskas et al., 1991). By 8h, although PMNs continue to adhere by an ICAM-1/CD18-dependent mechanism, they seldom migrate (Smith et al., 1988; Luscinskas et al., 1991). Inthe present study, PMN emigration was negligible at 4h following endotoxin instillation.Although the present study did not examine ICAM-1 expression at 4h, the lack of PMNemigration implies that the level of ICAM-1 expression was insufficient to support CD18-dependent PMN emigration. Interestingly, rabbits are approximately 1000 times more sensitiveto the effects of endotoxin than mice (Beutler et al., 1985). The present study confirms thisresistance by showing that BALB/c mice required a higher dose of endotoxin than that usedin the rabbit (10 mg/kg versus 1.5 gg/kg, respectively) to elicit PMN emigration into thealveolar airspace. Moreover, CD18-dependent PMN emigration towards endotoxin in the82rabbit occurs within 4h of instillation (this study; Doerschuk et al., 1990a, 1990b). Theseobservations predict that the time course for increased ICAM-1 expression on rabbit alveolarcapillaries will be much more rapid than that observed in the mouse.The alveolar macrophage is probably an important and major target for endotoxin(Christman et al., 1989; Sylvester et al., 1990; Rankin et al., 1990; Rietschel et al., 1991;Issekutz et al., 1991). Significantly, BALB/c mouse peritoneal macrophages respond poorlyto several different stimuli including IFN-7 and endotoxin (Oswald et al., 1992). Presumably,BALB/c mouse alveolar macrophages also respond poorly to endotoxin. The binding ofendotoxin to monocytes is known to result in the synthesis and release of lymphokines such asIL-1, TNF-a, and IL-6 (reviewed by Rietschel et al., 1991). IL-1 and TNF-a are both capableof inducing endothelial ICAM-1 expression (reviewed by Lobb, 1992). Moreover, endotoxin-stimulated alveolar macrophages can also release products that act directly on PMNs. Forexample, human alveolar macrophages stimulated with endotoxin release potent neutrophilchemoattractants such as LTB4 and IL-8 (Sylvester et al., 1990; Rankin et al., 1990). Morerecently, it was shown that 3h after endotoxin-stimulation, rabbit alveolar macrophages alsorelease a novel protein factor with an apparent molecular mass of 22 to 32 1(13, which undergel filtration conditions probably exists as a dimer with an apparent molecular mass of 45 to60 kD (Issekutz et al., 1991). When this protein is injected intradermally into the rabbit, itinduces PMN infiltration (Ibid). Interestingly, this protein factor does not induce PMNmigration in vitro, suggesting that it may induce endothelial cells to elaborate chemotacticfactors (Ibid). If CD 18-dependent PMN emigration were dependent upon the release ofalveolar macrophage-derived chemotactic factors for PMNs and/or cytokines that induceICAM-1 upregulation, then sub-optimal production of these macrophage products in BALB/c83mice could account for the observed hyporesponsiveness towards endotoxin.To our knowledge, this is the first study to quantitate and compare the expression ofICAM-1 on specific cell types in peripheral lung tissue. ICAM-1 has been detected on trachealepithelial cells during allergic airway inflammation in cynomolgus monkeys, but its constitutiveexpression appears to be negligible, or very low at best (Wegner et al., 1990). In contrast,the high level of constitutive epithelial ICAM-1 expression on the Type I pneumocyte impliesthat it is required for normal lung function. One intriguing possibility is that epithelial ICAM-1 participates in alveolar macrophage locomotion over the epithelial cell surface and thereby,assists the macrophage in scavenging inhaled particulates. The data from this study show thatICAM-1 is constitutively expressed on alveolar Type I pneumocytes and not Type IIpneumocytes. Although the alveolar surface is comprised of nearly equal numbers of Type Iand II pneumocytes, the squamous Type I pneumocytes make up approximately 95% of thealveolar surface (Thurlbeck and Miller, 1988). Collectively, these observations predict thatthe majority of the alveolar surface expresses ICAM-1. It is well established that ICAM-1 isan adhesive ligand for the leukocyte integrins CD11a/CD18 and CD11b/CD18 (Rothlein et al.,1986; Smith et al., 1989; Diamond et al., 1990, 1991). Alveolar macrophages obtained bylung lavage from humans (Albert et al., 1992), monkeys (Ibid), and mice (Rosen and Gordon,1990) express high surface levels of CD11a/CD18 and low or insignificant levels ofCD1 lb/CD18 and CD11c/CD18. The basal adherence of human alveolar macrophages to anunstimulatal alveolar epithelial cell monolayer is CD 18-independent (Albert et al., 1992).However, stimulation of the monolayer with TNF-a increased epithelial ICAM-1 expression9-fold and increased numbers (5-10% more) of macrophages were found to adhere to thisstimulated monolayer. Significantly, this increased adherence was CD 18-dependent and84completely inhibitable by MAb 60.3 (Ibid). The contribution of ICAM-1 to this CD18-dependent adherence was not determined, but a role for ICAM-1 is implied.In the present study, normal Type II pneumocytes express little or no ICAM-1 on theircell surfaces. However, acute inflammation in the lung is accompanied by a significantincrease (48-fold in response to S. pneumoniae and 178-fold in response to E. coli endotoxin)in Type II pneumocyte ICAM-1 expression. Twenty-four hours after endotoxin instillation,the levels of ICAM-1 expression are similar to those on normal Type I pneumocytes.Increased surface expression of ICAM-1 is considered to be a time dependent process requiringmRNA and protein synthesis (Dustin et al., 1986). Induction of mRNA and protein synthesisreflects a change in cellular activity. In the Type II pneumocyte, this may also reflect a changein cellular function.In normal lung, Type II pneumocytes have two very important functions. First, theysynthesize and secrete pulmonary surfactant which forms a continuous lining layer that coversthe alveolar epithelial surface (Kuhn, III and Finke, 1972; Junqueira et al., 1989; Uhal et al.,1991); surfactant functions to decrease the surface tension in the lung and prevent alveolarcollapse during expiration (Deavers, 1984; Junqueira et al., 1989; Uhal et al., 1991). Second,they are considered to be the progenitor cells that give rise to both Type I and II pneumocytesduring normal epithelial cell turnover (Adamson and Bowden, 1975; Uhal et al., 1991).Importantly, when the alveolar epithelium is injured during an inflammatory response, TypeII pneumocyte proliferation increases to provide new sister cells for epithelial cell replacement.This conclusion is based on the observation that the lung injury produced by inhalation of 90%oxygen (Adamson and Bowden, 1974) or 15-17 ppm nitrogen dioxide (Evans et al., 1973) inmice and rats, respectively, resulted in increased mitotic activity and increased incorporation85of tritiated thymidine in Type II pneumocytes. Of particular relevance to the present study isthe published finding that in the rat, Type I pneumocytes are injured during streptococcalpneumonia (Rhodes et al., 1989). Moreover, the injured Type I pneumocytes are replaced byproliferating Type II pneumocytes (Ibid). In the present study, Type II pneumocyte ICAM-1expression increases during both streptococcal and endotoxin pneumonias and this implies thatincreased ICAM-1 expression may be an early differentiation event that precedes Type IIpneumocyte proliferation.It has been reported that tracheal instillation of colloidal carbon (4-fold more than wasused in the present study) into the lungs of mice typically results in acute inflammation asevidenced by PMN emigration into the alveolar space (Adamson and Bowden, 1978, 1980,1982; Bowden and Adamson, 1978). In the present study, smaller amounts of colloidal carbonwere instilled and did not induce PMN emigration, but did result in a 52-fold increase inICAM-1 surface expression on Type II pneumocytes. Interestingly, significant epithelial injurydoes not seem to be associated with colloidal carbon instillation (this study; Adamson andBowden, 1978, 1980, 1982) and yet ICAM-1 expression is increased. If increased ICAM-1expression is associated with Type II pneumocyte proliferation, then instillation of colloidalcarbon should result in increased Type II pneumocyte proliferation as measured by the tritiatedthymidine method (Evans et al., 1973; Adamson and Bowden, 1974). To date there are nostudies to confirm or reject this hypothesis.Assuming, for the moment, that increased ICAM-1 expression is an early differentiationevent that is linked to and precedes Type II pneumocyte mitosis, then the stimulus for increasedType II pneumocyte proliferation may also be the stimulus for increased ICAM-1 expression.In vitro, alveolar macrophages stimulated with opsonizet1 zymosan particles produce a Type86II pneumocyte growth-factor that increases cell proliferation (Brandes and Finkelstein, 1989).The apparent molecular mass of this growth factor is > 25 kD and its ability to promote TypeII pneumocyte proliferation in vitro is unique when compared with other known growthpromoting macrophage products (Ibid). In the present study, alveolar macrophages ingest largeamounts of carbon particles. Ingestion of carbon can induce macrophages to produce solubleactivation factors for PMNs (Adamson and Bowden, 1982) and perhaps for Type IIpneumocytes. In vitro, a particulate stimulus (opsonized zymosan) induces alveolarmacrophage production of a Type II pneumocyte growth factor (Brandes and Finkelstein,1989). The amount of growth factor produced is dependent upon the concentration of thestimulus and the length of time that the macrophages are exposed to the particles (Ibid). Ifcolloidal carbon, in the absence of S. pneumoniae or E. coli endotoxin, also stimulatesmacrophages to produce this growth factor, then the observation that the Type II pneumocyteICAM-1 expression is increased at 24h, but not at 4h, following colloidal carbon instillationmay simply relate to the length of time that the macrophages are exposed to the carbonparticles. It remains to be determined whether colloidal carbon particles induce macrophagesto produce Type II pneumocyte growth factor and whether Type II pneumocyte exposure to thisgrowth factor results in increased ICAM-1 expression.4.2 OBSERVATIONS ON L-SELECTIN4.2.1 The topographical distribution of L -selectin on the PMN surfaceThe third specific aim of this thesis was to determine, in vitro, whether L-selectinis topographically positioned on the PMN surface such that it renders L-selectin-associated87sLex more "bioavailable" for the vascular selectins P- and E-selectin. The results on humanPMNs clearly show that L-selectin is topographically positioned on the PMN surfacemicrovilli. This observation is consistent with the hypothesis that L-selectin-associated sLexcan be preferentially "presented" to the vascular selectins, even though it accounts for only 5%of protein associated surface sLex (Picker et al., 1991), because its concentration on themicrovilli renders it more "bioavailable." Consistent with this proposal is the in vitro findingthat microvilli mediate the initial contact between the PMN and the endothelium (Beesley etal., 1979). During acute inflammation in the systemic circulation, PMNs make transientcontacts with the endothelium and they "roll" along the vessel wall. Intravital microscopicstudies in the systemic circulation have demonstrated that the "rolling" of PMNs along theinflamed venular endothelium is largely mediated by L-selectin (von Andrian et al., 1991; Leyet al., 1991). Rolling PMNs are essentially spherical (Tangelder and Arfors, 1991) and thepresent study suggests that this rolling behaviour and its dependence on L-selectin is explainedby the unique association of L-selectin with the microvillar processes. The uniqueness of thisassociation is suggested by the present finding that CD18 and total sLex are not preferentiallyexpressed by surface microvilli.Although PMN L-selectin-associated sLex is an important presenter of sLex to thevascular selectins P- and E-selectin (Kishimoto et al., 1991; Picker et al., 1991), it may notbe the only PMN carbohydrate ligand for these adhesion molecules. Both glycolipids andglycoproteins on the leukocyte surface express the sLex epitope (reviewed by Paulson, 1992)and several studies have examined the possibility that these other sLex-bearing molecules alsobind to P- and E-selectin. Protease treatment of PMNs cleaves L-selectin from the surface andabolishes soluble 121-P-selectin binding (Moore et al., 1991). This observation suggests that88the PMN ligand for P-selectin is a surface protein or glycoprotein and this is consistent withL-selectin presentation of sLex to P-selectin. However, the binding of soluble P-selectin toPMNs is not affected by PMN activation (Moore et al., 1991) and this implies that PMN L-selectin can not be the major ligand for P-selectin, because PMN activation is associated withL-selectin downmodulation (this study; Kishimoto et al., 1989a; Jutila et al., 1989, 1990,1991). This contradicts an independent collaborative finding that the anti-L-selectin MAbDREG-56 inhibited PMN binding to P-selectin-transfected COS (monkey kidney cell line) cellsby 70% (Picker et al., 1991). The reason for this discrepancy is unclear.Several other investigators have used the myeloid/monocytic cell line HL60 to study theleukocyte ligand for E-selectin. Interestingly, HL60 cells bind to E-selectin-transfected CHO(chinese hamster ovary) cells and this binding is unaffected by protease digestion of HL60 cells(Larsen et al., 1992). Furthermore, when HL60 cells are grown in the presence of inhibitorsthat block the synthesis of N-linked glycosylatal glycoproteins, binding to E-selectin is notdiminished (Leeuwenberg et al., 1991). Collectively, these findings imply that the HL60ligand for E-selectin is a glycolipid. This conclusion appears to contradict two other studiesdemonstrating that the PMN L-selectin glycoprotein functions as an important ligand for E-selectin (Kishimoto et al., 1991; Picker et al., 1991). However, HL60 cells are not directlycomparable to PMNs since they do not express L-selectin (Tedder et al., 1989). Furthermore,PMNs bind E-selectin-transfected cells more avidly than HL60 cells (Kishimoto et al., 1991)and this strengthens the concept that L-selectin is the major PMN "presenter" of sLex to E-selectin (Picker et al., 1991).894.2.2 Downmodulation of L-selectin during PMN transmigrationThe fourth, and last, specific aim of this thesis addressed whether the process ofPMN transendothelial migration is associated with L-selectin downmodulation. The resultsshow that, in vitro, L-selectin downmodulation is associated with PMN transendothelialmigration and the evidence supporting this conclusion is that there is a qualitative reduction inL-selectin immunoreactivity on PMNs as they transmigrate an IL-1-stimulated endothelialmonolayer. This observation implies that L-selectin downmodulation is initiated by PMNcontact with and migration through an activated endothelium. Interestingly, completedownmodulation of L-selectin was never observed over the 5 minute period of adhesion andmigration; PMNs continue to express low levels of L-selectin even after they transmigrate theendothelium.The application of large numbers of PMNs GM to an IL-1-stimulated endothelialmonolayer is an ideal way to induce synchronous transendothelial migration. In theory,random sections through this monolayer should show many PMN profiles in the process ofadhesion and migration. In practice, the endothelial monolayers were very difficult tocryosection and the ultrastructural preservation was less than optimal. However, L-selectinantigenicity was maintained and the ultrastructural preservation was adequate for identifyingPMNs in various stages of transendothelial migration. Surprisingly few adherent PMNs wereobserved as most had already transmigrated the endothelium by 5 minutes. However, in thefew cases where an adherent PMN was observed, the portion of the PMN surface in contactwith the endothelium was L-selectin negative. One interpretation of this observation is that L-selectin is downmodulated at the point of contact with the endothelium. Indeed, prolonged (30minutes) adherence of PMNs to an IL-1-stimulated endothelium results in L-selectin90downmodulation (Smith et al., 1991). Alternatively, it is just as likely that L-selectinmolecules are expressed on the PMN surface in contact with the endothelium, but because theyare engaged with their endothelial ligands, they are not accessible to immunodetection with theDREG-200 MAb. Moreover, the surface of the PMN appears to be tightly apposed to theendothelium at the point of contact and this may preclude the penetration of immunogoldreagents at the contact site. By comparison, the free-surface of the adherent PMN was L-selectin positive and the intensity and distribution of gold label was not appreciably differentfrom that on the surface of non-adherent PMNs. If L-selectin were downmodulated on thePMN surface only at the point of contact with the endothelium, then L-selectindownmodulation on adherent PMNs is focal and contact dependent. One possible mechanismof L-selectin downmodulation is suggested by an independent study which showed that IL-1stimulates endothelial monolayers to express cell surface-associated platelet activating factor(PAF). Importantly, this membrane-associated form of PAF contributes to L-selectindownmodulation on transmigrating PMNs (Kuijpers et al., 1992a). Although the present studydoes not confirm that L-selectin is downmodulated on the PMN surface that is in contact withthe IL-1-stimulated endothelium, the lack of L-selectin immunoreactivity on the contact surfaceis consistent with a model of endothelial-associated PAF-induced L-selectin downmodulation.Within 5 minutes, adherent PMNs quickly become activated as evidenced by theirability to transmigrate the endothelium and by the large numbers of PMNs that accumulatebeneath the endothelium. Transmigration begins when the PMN extends a pseudopod throughthe endothelial monolayer. Importantly, the pseudopod is less immunoreactive for L-selectinthan the portion of the PMN that has not yet passed through, or contacted, the endothelium.This observation is again consistent with the idea that L-selectin downmodulation is initiated91on the PMN surface in contact with the endothelium. The mechanism of L-selectindownmodulation on the PMN pseudopod is likely explained by L-selectin shedding duringPMN activation (Kishimoto et al., 1989a). In vitro studies have demonstrated that followingcell activation, PMNs shed > 95% of their L-selectin within 5 minutes (Kishimoto et al.,1989a; Jutila et al., 1990). This time course for L-selectin shedding is compatible with thetime course of the present study in which PMN transmigration was terminated byglutaraldehyde fixation 5 minutes after the addition of PMNs to the IL-1-stimulated endothelialmonolayer. Furthermore, in the present study, there was no evidence of L-selectininternalization during PMN transmigration, supporting the concept that L-selectin is shed fromthe surface during PMN transendothelial migration.4.3 SENSITIVITY AND LIMITATIONS OF THE IMMUNOGOLD TECHNIQUEVery little information is available for comparing the sensitivity of the immunogoldtechnique to conventional methods of surface antigen detection. One FACS (fluorescence-activated cell sorter) study has compared the labeling of cell surface antigens with primarymonoclonal antibodies followed by secondary polyclonal antibodies that are tagged with eithera gold particle or a fluorescent marker. Importantly, when the secondary antibody is taggedwith a colloidal gold particle (40 nm in diameter), it can be detected in the side scatter (90°)channel in logarithmic amplification mode (Totterman and Festin, 1989). The results clearlyshowed that the colloidal gold method "...recognizes a population identical to that detected bythe fluorochromes..." (Ibid). An independent study has shown that the immunogold labelingof cell surface antigens can be used to successfully study lymphocyte subsets by transmissionelectron microscopy (Hoogeveen et al., 1988).92In the present study, the immunogold technique was coupled to cryoultramicrotomy.The sensitivity of the cryotechnique is far superior to that obtained with epoxy or methacrylateembedded sections (personal experience; reviewed by van Bergen en Henegouwen, 1989).Moreover, surface detection of antigens on both PMNs and endothelial cells was enhanced inthe present study because plasma proteins within the pulmonary microvasculature were notpreserved during fixation. Gold labeling is enhanced, in the absence of plasma proteins,because the immunogold reagents had complete access to the endothelial or PMN cell surfacethroughout the depth (80-100 nm) of the cryosection. This was verified by tipping thegoniometer stage in the electron microscope and observing the plasma membrane profiles enface. Importantly, gold particle counts were made on plasma membrane surfaces that had bothlength and depth; therefore, this is really a surface area count rather than a two-dimensionallinear count. However, because the section thickness was held constant throughout the study,it was not a confounding variable in this study.Two limitations of the immunogold technique were encountered in the present study.One was that it was not possible to detect colloidal gold particles at the site of close contactor adhesion between the PMN and the endothelium. Immunogold reagents are likely excludedfrom these regions because the adhesion molecule epitope recognized by the MAb is alreadyengaged in cell-cell adhesion. Alternatively, the immunogold reagents may not be able tophysically penetrate the close contact region between the cells. Obviously, an understandingof the expression of adhesion molecules at the contact region between the PMN and theendothelium is of great interest and importance. Receptors and ligands may be clustered atthese sites as has been described for CD 1 lb/CD18 during PMN adhesion to C3bi-coatederythrocytes (Detmers et al., 1987). Future prospects for studying the contact region with the93immunogold technique will require raising specific MAbs against cytoplasmic epitopes on theadhesion molecules.A second limitation in this study relates to the small numbers of animals that wereexamined (9 rabbits; 13 mice); the labour intensive nature of these immunogold studiesprecluded the possibility of studying more animals. This limitation is best illustrated by thefinding that it was not possible to determine with statistical certainty whether ICAM-1expression on the pulmonary capillary endothelium is significantly increased during CD18-dependent PMN emigration in the mouse lung. This is an important question and one that canonly be answered by studying larger numbers of mice.94CHAPTER FIVECONCLUSIONS5.1 SUMMARY AND CONCLUSIONS^  955.2 FUTURE PROSPECTS ^  995.3 SIGNIFICANCE OF THIS THESIS ^  101955.1 SUMMARY AND CONCLUSIONSThe main objective of this thesis has been to quantitate and compare the in vivo surfaceexpression of three leukocyte-endothelial adhesion molecules (L-selectin and CD18 on PMNs;ICAM-1 on endothelial and epithelial cells) during CD 18-independent and CD 18-dependentPMN emigration in the lungs of rabbits and mice.The working hypothesis of this thesis was that during an acute inflammatory response,the adherence of PMNs to the pulmonary microvasculature and their subsequent emigration intothe alveolar airspace are dependent upon the complex interaction of cell surface adhesionmolecules expressed on both the PMN and the endothelium. The principal site of PMNemigration during acute inflammation in the lung is the alveolar capillaries. PMNs thatsequester in the capillary bed near the injured site are activated by the release of localinflammatory mediators and cytokines. Some of these stimuli will also activate the capillaryendothelial cells. Therefore, PMN emigration in response to inflammatory stimuli that evokeeither CD 18-independent or CD 18-dependent emigration, is associated with a downmodulationof L-selectin, an upmodulation of CD18, and an upmodulation of ICAM-1.In this study, rabbits and mice received airway instillates of Streptococcus pneumoniaeor Escherichia coil endotoxin to induce CD 18-independent or CD 18-dependent PMNemigration, respectively. Following the development of pneumonia, the lungs were fixed inglutaraldehyde, cryoprotected in 2.3M sucrose, frozen in liquid nitrogen, and sectioned on acryoultramicrotome. Ultrathin cryosections were immunogold labeled with MAbs DREG-200(against L-selectin) and 60.3 (against CD18) in the rabbit, or YN1/1.7.4 (against ICAM-1) inthe mouse. Gold particles on the plasma membrane surfaces of PMNs, endothelial cells, andepithelial cells were quantitated by transmission electron microscopy and expressed as gold96particles/Am of plasma membrane.The results clearly show that CD 18-independent PMN emigration towards a bronchialinstillate of S. pneumoniae is associated with L-selectin downmodulation and CD18upmodulation on intravascular PMNs, prior to emigration. A similar alteration in theexpression of L-selectin and CD18 is observed during CD 18-dependent PMN emigrationtowards a bronchial instillate of E. coli endotoxin, but only on PMNs that have emigrated intothe interstitium and airspace. It is concluded that increased CD18 expression on PMNs is notrequired for CD 18-dependent emigration in the lung and this finding supports the concept thatqualitative (conformational) changes rather than quantitative changes may be important toCD 18-dependent adhesion and migration.Interestingly, CD18 expression is increased 5.5-fold on PMNs within the alveolarairspace during streptococcal pneumonia and yet, CD18 is not required for CD 18-independentPMN emigration. Because CD11b/CD18 is also capable of binding the complement splitproduct C3bi, increased expression of CD11b/CD18 may facilitate PMN phagocytosis of C3bi-opsonized bacteria. In evidence is the observation that large numbers of streptococcal bacteriawere ingested by airspace PMNs.An important new finding is that regardless of the mechanism of PMN emigration, thechange in PMN L-selectin and CD18 expression occurs only on PMNs within the pneumonicregion. The evidence supporting this conclusion is that within the pulmonary microvasculatureof the non-pneumonic contralateral rabbit lung, L-selectin and CD18 expression onintravascular PMNs is not different from control. Because downmodulation of L-selectin andupmodulation of CD18 occurs during PMN activation, these observations imply, in the caseof streptococcal pneumonia, that PMNs become activated as they transit through the97inflammatory lesion and that once activated, they do not rejoin the circulation in appreciablenumbers. In the case of endotoxin pneumonia, PMN activation occurs during transendothelialmigration, as evidenced by an 86% reduction in L-selectin expression and a 100% increase inCD18 expression on interstitial PMNs. This observation is consistent with a role for L-selectinin PMN adhesion to the vascular endothelium and argues that L-selectin is not involved inPMN migration through the interstitial matrix.PMN emigration towards lipopolysaccharide (endotoxin) in the mouse lung is knownto be CD 18-dependent. ICAM-1 expression increases 4.2 fold in response to E. coli endotoxinand this finding borders on statistical significance, suggesting that ICAM-1 may be animportant adhesive ligand for CD18 adhesion molecules during CD 18-dependent emigrationin the lung. Although it remains to be determined whether PMN emigration towards S.pneumoniae in the mouse lung is CD 18-independent, the observation that PMN emigrationoccurred in the absence of an increase in capillary endothelial ICAM-1 expression is consistentwith the concept of a CD 18-independent emigration mechanism.A potentially important new observation is that in normal mouse lung, constitutiveICAM-1 expression is 22-fold greater on the alveolar epithelium compared to the capillaryendothelium and epithelial expression is mainly restricted to the Type I pneumocytes. Thefunction of this epithelial ICAM-1 expression is unknown. However, alveolar macrophagesexpress high surface levels of CD1 la/CD18 and it is conceivable that macrophages adhere toepithelial ICAM-1 as they travel over the epithelial surface scavenging inhaled particulates.Interestingly, Type II pneumocytes constitutively express very little ICAM-1, yet duringpneumonia, ICAM-1 expression increased as much as 178-fold. The reason for this increasedexpression is unknown, but it may be an early differentiation event that precedes Type II98pneumocyte proliferation. When the epithelium is injured, Type II pneumocytes proliferate andreplace the damaged epithelium (Adamson and Bowden, 1974; Evans et al., 1973). It isconceivable that during pneumonia, Type II pneumocytes are stimulated by inflammatorymediators, possibly released by the alveolar macrophage. However, epithelial injury, asmeasured by morphologic criteria, is not a prerequisite for Type II pneumocyte ICAM-1induction since the epithelium appeared to be undamaged at 24h following the instillation ofcolloidal carbon alone, yet the Type II pneumocytes showed a 52-fold increase in ICAM-1expression.For comparative purposes, the immunogold method was used to study the in vitroexpression of L-selectin on unstimulated human PMNs and on human PMNs engaged intransmigration of an IL-1-stimulated endothelial monolayer. There were two specific aims tothis part of the study and the question addressed by the first was whether L-selectin istopographically positioned on the PMN surface such that it renders L-selectin-associated sLexmore "bioavailable" for the vascular selectins P- and E-selectin. In vitro results on humanPMNs demonstrate that L-selectin is topographically positioned on the PMN surface microvilli.Remarkably, the same observation was made for PMNs in the rabbit and this suggests that itmay be a common finding in other animal species as well. Although L-selectin-associated sLexaccounts for only 5% of protein associated surface sLex (Picker et al., 1991), its concentrationon the microvilli may render it more "bioavailable" for the vascular selectins, P- and E-selectin. The other specific aim addressed whether L-selectin downmodulation is coincidentalwith PMN transendothelial migration. The results show that L-selectin downmodulation iscoincidental with PMN transendothelial migration and it appears to be induced by contact withthe endothelium. Interestingly, L-selectin was also downmodulated on rabbit PMNs during99transendothelial migration towards E. coli endotoxin and, as mentioned above, L-selectindownmodulation is an indication of PMN activation. The mechansim of PMN activation wasnot determined in this thesis, but it may involve an endothelial-associated form of PAF that hasbeen shown to contribute to L-selectin downmodulation (Kuijpers et al., 1992a).5.2 FUTURE PROSPECTS The findings reported in this thesis have suggested several potential avenues of futureresearch. As already mentioned, it remains to be determined whether ICAM-1 expression issignificantly increased on pulmonary capillary endothelial cells during CD 18-dependent PMNemigration in the mouse lung. This question could be best addressed by repeating the presentstudy, but with larger numbers of mice. Five control mice and 5 endotoxin-treated mice wouldprovide sufficient statistical power to determine whether the 4.2-fold increase in endothelialICAM-1 expression during endotoxin pneumonia is significant. If ICAM-1 expression issignificantly increased at 24h, it would be worth examining earlier time points to determinewhether PMN emigration can occur prior to an increase in endothelial ICAM-1 expression.Finally, a group of mice could be pretreated with the blocking anti-ICAM-1 MAb YN1/1.7.4(Doerschuk et al., 1992) to determine whether ICAM-1 expression is essential for CD18-dependent PMN emigration in the lung.A second avenue of research would be to determine whether the presence of ICAM-1on the Type I pneumocytes facilitates alveolar macrophage motility and thereby enhances themacrophages ability to scavenge and ingest inhaled particulates. The working hypothesis is thatalveolar macrophages crawl over the epithelial surface in order to phagocytose inhaledparticulates and that the adhesive interaction between the macrophage and the epithelium is100both CD 18- and ICAM-1-dependent. This hypothesis predicts that macrophages will ingestfewer particulates if their motility is inhibited by an anti-ICAM-1 blocking MAb, likeYN1/1/7.4 (Doerschuk et al., 1992). An aerosol mixture of colloidal carbon particles in salinecould be delivered into the lungs of mice and the numbers of macrophages with ingested carbonparticles could be determined by light microscopic histology at various time intervals (e.g. 0,10, 30, and 60 minutes). The experiment could be repeated, but this time the colloidal carbonparticles would be mixed with the YN1/1.7.4 MAb. In the presence of this MAb, fewermacrophages should show signs of colloidal carbon ingestion and those that ingest carbon, willcontain only minor amounts because they will only have access to carbon particles that havebeen deposited on or adjacent to the macrophage. If this experiment proved successful, thestudy could be expanded to determine which member of the CD 18-family was interacting withthe ICAM-1. This question could be addressed by substituting anti-CD11a, anti-CD11b, oranti-CD1 lc MAbs for the anti-ICAM-1 MAb.A third avenue of research would be to determine whether the induction of ICAM-1expression on Type II pneumocytes is an early differentiation event that precedes proliferation(mitosis). The working hypothesis is that Type II pneumocytes show enhanced proliferationfollowing injury to the alveolar epithelial cell surface and this enhanced proliferation will bepreceded by increased ICAM-1 expression. To test this hypothesis, Type II pneumocytes couldbe induced to proliferate by exposing the lungs of mice to 90% oxygen (Adamson and Bowden,1974). The rate of Type II pneumocyte proliferation could be monitored by tritiated thymidineincorporation (Ibid) and lung tissue could be simultaneously processed for immunogold labelingof ICAM-1 as described in this thesis.A fourth, but by no means final, avenue of research would be to determine whether L-101selectin and ICAM-1 are required for CD18-independent emigration in the lung. Themorphological data presented in this thesis suggest that ICAM-1 and L-selectin are unlikely tobe involved in CD 18-independent PMN emigration because endothelial ICAM-1 expression isnot increased and L-selectin is downmodulated on intravascular PMNs that have not emigrated.The working hypothesis is that neither L-selectin nor ICAM-1 is required for CD18-independent PMN emigration. This hypothesis could be tested by pretreating rabbits with amixture of the blocking MAbs 60.3 (anti-CD 18)and DREG-200 (anti-L-selectin) or 60.3 andR6.5 (anti-ICAM-1) (Barton et al., 1989; Argenbright et al., 1991) and then instilling S.pneumoniae to induce CD 18-independent PMN emigration, as described in this thesis. The60.3 MAb would inhibit any possible contribution of CD18 to the PMN emigration process.PMN emigration could be quantitated by either bronchial lavage or by light microscopichistology. The prediction is that neither the DREG-200 or the R6.5 MAb will have any effecton CD 18-independent PMN emigration.5.3 SIGNIFICANCE OF THIS THESIS The adherence and emigration of leukocytes at sites of injury or infection is an earlyand essential step in the defense and repair of the host tissues. However, there are severalclinical disorders that arise as complications of an unwanted inflammatory or immune response,including: rheumatoid arthritis, adult respiratory distress syndrome, ischemia-reperfusion injuryand allograft rejection (reviewed by Harlan et al., 1992). Current research in this area isfocused on the development of anti-adhesion therapies to prevent leukocyte-mediated vascularand tissue injuries (Ibid). The study of leukocyte-endothelial interactions in vivo using animalmodels is a necessary prerequisite to understanding "...the cell biology, physiology, and102clinical relevance of leukocyte-endothelial interactions in humans." (Ibid). The findingspresented in this thesis provide new information that confirm and extend our presentunderstanding of leukocyte-endothelial (and epithelial) adhesive interactions. Moreover, thesefindings offer the possibility of new avenues of research that in themselves will lead to a morecomplete understanding of the role of adhesion molecules in health and disease.103TABLE 1. L-selectin study: Summary of moiphometric data on rabbit neutrophils (PMNs) incontrol lungs and lungs instilled with an inflammatory stimulus (S. pneumoniae orE. coli endotoxin) and colloidal carbonTreament Lung regionexaminedTotal # goldparticlesTotal memb.length (p,m)Mean memb.length (tim)No instillate(n=3 rabbits)intravascular 3,189a 581b 19.4 + 3.6c4h S. pneumoniae(n=3 rabbits)intravascular 561 515 17.2 + 1.1interstitial' 124 207 20.7airspace 93 691 23.0 + 2.0contralateralintravascular3,700 446 14.9 + 4.44h E. coliendotoxin(n=3 rabbits)intravascularinterstitial1,56956360673320.2 + 1.224.4 + 4.0airspace 161 722 24.1 ±0.3contralateralintravascular2,657 486 16.2 + 2.9a total number of gold particles counted on 30 PMN profiles.I) total plasma membrane length examined from 30 PMN profiles.c mean + (SD) PMN plasma membrane length examined/rabbit.d only 10 PMN profiles were examined in one rabbit; all other regions had 30 PMNprofiles from three rabbits.104TABLE 2. CD18 study: Summary of morphometric data on rabbit neutrophils (PMNs) incontrol lungs and lungs instilled with an inflammatory stimulus (S. pneumoniae orE. coli endotoxin) and colloidal carbonTreatment Lung regionexaminedTotal # goldparticlesTotal memb.length (Am)Mean memb.length (pcm)No instillate(n=3 rabbits)intravasculara 3,619b 397C 13.7 + 3.1d4h S. pneumoniae(n=3 rabbits)intravascular 6,312 281 9.4 + 0.9interstitial 10,183 460 15.3 + 0.6airspace 18,053 396 13.0 + 1.6contralateralintravascular4,274 434 14.5 + 4.24h E. coliendotoxin(n=3 rabbits)intravascularainterstitial3,8139,70434057811.7 + 2.219.3 + 3.3airspace 13,620 538 17.9 + 1.3a only 29 PMN profiles examined from three rabbits; all other regions had 30 PMNprofiles from three rabbits.b total number of gold particles counted on 30 PMN profiles.c total plasma membrane length examined from 30 PMN profiles.d mean + (SD) PMN plasma membrane length examined/rabbit.105TABLE 3. L-selectin immunoreactivity on rabbit neutrophils (PMNs) in control lungs andlungs instilled with an inflammatory stimulus (S. pneumoniae or E. coli endotoxin) andcolloidal carbonTreatmentLung region examinedintravascular interstitial airspace contralateralintravascularNo instillate 5.6 + 3.2 NA NA NA(Control)4h 0.9 + 1.5' ND 0.04 + 0.02' 8.1 + 5.3S. pneumoniae (0.22) (0.01)4h 2.7 + 3.7 0.6 + 1.0a'b 0.1 + 0.2° 5.6 + 2.7E. coliendotoxin(0.14) (0.02)Three rabbits were studied for each treatment. Values outside of closed brackets are the mean+ (SD) gold partilces/Am plasma membrane. Values within closed brackets are the estimatedfold difference in L-selectin immunoreactivity relative to that on the control PMNs. NA, notapplicable; ND, not determined.a P < 0.05 vs. intravascular (Control).b P < 0.05 vs. intravascular (E. coli endotoxin)106TABLE 4. CD18 immunoreactivity on rabbit neutrophils (PMNs) in control lungs and lungsinstilled with an inflammatory stimulus (S. pneumoniae or E. coli endotoxin) and colloidalcarbonTreatmentLung region examinedintravascular interstitial airspace contralateralintravascularNo instillate 8.6 + 1.8 NA NA NA(Control)4h 21.7 + 6.1a 21.5 + 4.5a 45.4 + 7.6a,b,c 9.1 + 2.7S. pneumoniae (2.6) (2.6) (5.5)4h 11.0 + 4.3 16.7 + 2.5a 26.4 + 13.2a'd NDE. coliendotoxin(2.0) (3.2)Three rabbits were studied for each treatment. Values outside of closed brackets are the mean+ (SD) gold partilces/Am plasma membrane. Values within closed brackets are the estimatedfold difference in CD18 immunoreactivity relative to that on the control PMNs. NA, notapplicable; ND, not determined.a P < 0.05 vs. intravascular (Control).b P < 0.05 vs. interstitial (S. pneumoniae).C P < 0.05 vs. intravascular (S. pneumoniae).d P < 0.05 vs. intravascular (E. coli endotoxin).107TABLE 5. ICAM-1 study: Summary of morphometric data on control mouse lungs and lungsinstilled with colloidal carbon + an inflammatory stimulus (S. pneumoniae or E. coliendotoxin)Treatment Lung celltypeTotal # goldparticlesTotal memb.length (pm)Mean memb.length (/.4m)No instillate(n=3 mice)endothelial 502k 666b 222 + 74'Type I 15,355 1326 442 + 147Type II 110 337 112 ±554hcolloidal carbon(n=2 mice)endothelialType I2825,058343522172 + 65261 + 69Type II 101 392 196 + 5124hcolloidal carbon(n=2 mice)endothelialType I2201,990268281140 + 63134 + 28Type II 718 270 135 + 234h endothelial 799 699 233 + 57S. pneumoniae(n=3 mice) Type I 5,298 847 282 + 91Type II 1,140 468 156 + 3024h endothelial 1,201 586 195 ±28E. coli endotoxin(n=3 mice) Type I 3,258 586 195 + 38Type II 5,134 572 191 + 37a total number of gold particles counted on the electron microscope.I) total plasma membrane length examined on the electron micrographs.' mean + (SD) plasma membrane length examined in each mouse.108TABLE 6. ICAM-1 immunoreactivity in control mouse lungs and lungs instilled with colloidalcarbon + an inflammatory stimulus (S. pneumoniae or E. coli endotoxin)TreatmentLung cell typeendothelial Type I Type IINo instillate 0.5 + 0.1 10.6 + 3.0arb 0.1 + 0.1a(Control)4hcolloidal carbon0.6 + 0.0 9.4 + 0.6 0.1 + 0.124h 0.6 + 0.3 7.7 + 2.3 2.6 + 0.8bcolloidal carbon (52)4h 1.0 ±0.3 6.1 ±0.6 2.1 + 1.1bS. pneumoniae (2.1) (0.6) (42)24h 2.0 + 1.2' 5.1 + 2.0 8.9 + 4.8bE. coli endotoxin (4.2) (0.5) (178)Three mice were studied for each treatment, except for 4h and 24h colloidal carbon where twomice were studied for each treatment. Values outside of closed brackets are the mean + (SD)gold partilces/Am plasma membrane. Values within closed brackets are the estimated folddifference in ICAM-1 immunoreactivty relative to that on the corresponding control cell type.a Bonferroni-adjusted P < 0.0125 vs. endothelial (Control).b Bonferroni-adjusted P < 0.0125 vs. Type II (Control).' Bonferroni-adjusted P = 0.017 vs. endothelial (Control).109Figure 1: Figure showing the leukocyte-endothelial adhesion molecules that have beencharacterized at the molecular level and shown to be important for PMNadhesion and migration during acute inflammation. Note that the endothelialadhesion molecules E- and P-selectin recognize PMN molecules bearing thesialyl Lewis X antigen (sLex*); however, the identity of these PMN moleculesawaits characterization. Interestingly, L-selectin on the PMN also bears an sLexepitope and there is some evidence that it can bind to endothelial E-selectin andP-selectin. L-selectin also has its own lectin-binding domain and can potentiallyinteract with additional, and as yet uncharacterized, endothelial carbohydrate-bearing ligands. The leukocyte integrins CD11a/CD18 and CD 1 lb/CD18 canboth bind to the endothelial ligand ICAM-1 (intercellular adhesion molecule-1)and CD1 la/CD18 can also bind to ICAM-2. The endothelial ligand recognizedby CD11c/CD18 has not been characterized, nor has its role in the inflammatoryprocess been clearly defined. See Introduction for more information on eachadhesion molecule.LEUKOCYTE-ENDOTHELIAL ADHESION MOLECULESIMPORTANT TOPMN ADHESION AND EMIGRATIONCD11c/CD18CD11b/CD18ICAM-1ICAM-2E-selectinP-selectinsLex© A. Burns 1991INFLAMED TISSUE111Figure 2:^Figure showing the "Specimen Pin Jig" that was specially constructed tomanufacture large numbers of inexpensive brass specimen pins forcryosectioning. The jig was made from a rectangular-shaped block ofaluminium. A central guide hole (1/8 inch diameter) was drilled through thelong axis of the jig. In addition, two hacksaw blade guide slots were cut intothe top surface of the jig at right angles to the guide hole axis. A moveable"swing stop" was mounted at either end of the jig using a nut and bolt. In thisway, the jig could be used by either a left-handed or right-handed operator. Tofabricate the specimen pins, the jig was secured in a table-top vice and a 2-3foot long piece of brass brazing rod (1/8 inch diameter) was inserted into theguide hole until it butted up against the aluminum "swing stop." Holding therod firmly in one hand, a metal hacksaw (32 teeth/inch) was then used to cutthrough the brazing rod. To retrieve the newly cut pin, the swing stop wasmoved aside and the length of brazing rod pushed forward until the pin fell outonto the table. With the brazing rod in position, the swing stop was loweredonce more and another pin was cut. Numerous specimen pins were producedin only a few minutes using this method.I I aFigure 2^Specimen Pin Jig113Figure 3: Figure 4: Electron micrograph of an intravascular PMN in a rabbit that received abronchial instillate of E. coli endotoxin mixed with colloidal carbon. Threenuclear lobes (N) are visible within this PMN and electron dense accumulationsof heterochromatin are evident. Note that the diameter of the blood vessel isgreater than that of the PMN and that the PMN is approximately spherical inshape. The surface of the PMN is convoluted and it possesses numerous shortmicrovilli (arrowheads). The cytoplasmic granules of this PMN appeared empty(electron translucent) suggesting that the granule contents have been lost duringtissue processing. Scale = 1.0 Am.Electron micrograph of an elongated intravascular PMN in a control rabbit.Because the diameter of the capillary is very small (3.0 Am), the PMN iselongated in shape and it has a flattened surface. This surface is tightly apposed(black arrows) to the endothelium lining the pulmonary capillary. Thisparticular piece of lung tissue was not subjected to immunogold labeling becauseit was fixed with a relatively high concentration of glutaraldehyde (0.1% , ratherthan 0.025%), at 0.1% glutaraldehyde, the affinity of MAb 60.3 for CD18 issignificantly reduced. However, this micrograph illustrates the important pointthat, in cryosectioning, a small change in the concentration of the fixative candramatically affect the ultrastructural preservation of the tissue. In evidence isthe improved preservation of the PMN cytoplasmic granule contents which areno longer electron translucent (see Figure 3). A single nuclear lobe (N) and asingle centriole (white arrow) can also be appreciated in the cytoplasm of thisPMN. Scale = 1.0 Am.115Figure 5: Figure 6: Electron micrograph of an interstitial PMN in a rabbit that received a bronchialinstillate of E. coil endotoxin mixed with colloidal carbon. Note the elongatedshape of this PMN and the filipodal extension (arrow). Collagen fibres(arrowheads) confirm the interstitial location of the PMN. N = nuclear lobe.Scale = 1.0 Am.Electron micrograph of an airspace PMN in a rabbit that received a bronchialinstillate of S. pneumoniae mixed with colloidal carbon. The shape of this PMNis essentially spherical and the cell surface is highly ruffled. Note that, althoughthere were colloidal carbon particles (arrowhead) on the cell surface of thePMN, there was no evidence of carbon ingestion. However, there was evidenceof bacterial ingestion (arrows). N = nuclear lobe. Scale = 1.0 Am.lt6s117Figure 7:^Electron micrograph of an airspace PMN in a rabbit that received a bronchialinstillate of E. coli endotoxin mixed with colloidal carbon. The shape of thisPMN is essentially spherical and the cell surface is quite ruffled. Note that, alarge amount of colloidal carbon was associated with the cell surface (arrows),but there was no evidence of carbon ingestion. N = nuclear lobe. Scale = 1.0Am.Figure 8: ^Electron micrograph of an airspace PMN in a rabbit that received a bronchialinstillate of S. pneumoniae mixed with colloidal carbon. Note the large numbersof bacteria that have been ingested by this PMN (arrows) and the lack of carboningestion. N = nuclear lobe. Scale = 1.0 Am.118119Figure 9: Electron micrograph of a pneumonic lung in a rabbit that received a bronchialinstillate of S. pneumoniae mixed with colloidal carbon. Four PMNs (P), analveolar macrophage (M), a hemolysed red blood cell (R), and colloidal carbon(*) can be seen within the alveolar airspace. In contrast to the PMNs, thealveolar macrophage ingested more colloidal carbon (arrowheads) than it didbacteria (arrow). Scale = 2.0 Am.0121Figure 10:^Immunoelectron microscopic detection of L-selectin on a spherically-shapedintravascular PMN in control rabbit lung. Note that the majority ofimmunogold particles are localized to the surface microvilli (arrowheads). Nointracellular pool of L-selectin was associated with the cytoplasmic granules (*).N = nuclear lobe. Scale = 0.5 Am.Figure 11:^Immunoelectron microscopic detection of L-selectin on an elongatedintravascular PMN in control rabbit lung. Note the generally smooth andflattened appearance of the cell surface. Despite the lack of surface microvilli,immunogold particles appeared to be grouped into small clusters (arrowheads),rather than being distributed as single particles over the cell surface. Thecytoplasmic vesicles (*) did not show evidence of gold labeling. N = nuclearlobe. Scale = 0.5 Am.I123Figure 12: Immunoelectron microscopic detection of CD18 on an intravascular PMN incontrol rabbit lung. Immunogold particles were detected on the microvilli(arrowheads) and flatter regions (arrows) of the cell surface. A largeintracellular pool of CD18 was present as evidenced by the intense gold labelingover the cytoplasmic granules. The majority of these gold particles wereassociated with the inner surface of the granule membrane. N = nuclear lobe.Scale = 0.25 pm.lat+125Figure 13: Figure 14: Immunoelectron microscopic detection of L-selectin on an intravascular PMNin rabbit lung following a bronchial instillate of S. pneumoniae mixed withcolloidal carbon. Clusters of immunogold particles (arrowheads) are seen overthe entire free surface of the cell. Six out of ten PMNs examined in this rabbitwere L-selectin positive, the other four PMNs showed no evidence of goldlabeling. Arrows = colloidal carbon particles associated with the alveolarepithelium. N = nuclear lobe. Scale = 0.5 Am.Immunoelectron microscopic detection of L-selectin on an intravascular PMNin rabbit lung following a bronchial instillate of S. pneumoniae mixed withcolloidal carbon. Note the absence of gold labeling on the surface microvilli(arrowheads), the flatter regions of the cell surface (arrows), and thecytoplasmic granules (*). In contrast to the findings reported in Figure 14, ofthe ten PMNs examined in this rabbit, all of them were L-selectin negative. N= nuclear lobe. Scale = 0.25 Am.lab127Figure 15:^Immunoelectron microscopic detection of L-selectin on an airspace PMN inrabbit lung following a bronchial instillate of S. pneumoniae mixed withcolloidal carbon. Note the lack of gold particles on the microvilli (arrowheads)and the flatter regions of the cell surface (arrows). There is evidence ofbacterial ingestion (star) in preference to colloidal carbon (*). N = nuclearlobe. Scale = 0.5 Am.las129Figure 16: Immunoelectron microscopic detection of L-selectin on an interstitial PMN inrabbit lung following a bronchial instillate of S. pneumoniae mixed withcolloidal carbon. This is the same rabbit in which L-selectin positiveintravascular PMNs (Figure 14) were found. The interstitial location of thisPMN is confirmed by the presence of collagen fibres (arrowheads) and elastin(star). Note the scarcity of gold particles on the cell surface (arrows) andcomplete lack of gold label within the cytoplasmic vesicles (*). N = nuclearlobe. Scale = 0.5 Am.130131Figure 17: Figure 18: Immunoelectron microscopic detection of L-selectin on an intravascular PMNfrom the contralateral, non-pneumonic lung of a rabbit that received a bronchialinstillate of S. pneumoniae mixed with colloidal carbon. Note the abundant goldlabeling on the cell surface and its preference for the microvilli (arrows). Goldlabel was not detected within the cytoplasmic granules (*). N = nuclear lobe.Scale = 0.5 Am.Immunoelectron microscopic detection of L-selectin on an intravascular PMNfrom the contralateraL non-pneumonic lung of a rabbit that received a bronchialinstillate of E. coil endotoxin mixed with colloidal carbon. Large numbers ofgold particles are present on the cell surface (arrows) but none are located overthe cytoplasmic granules (*). N = nuclear lobe. Scale = 0.25 Am.isa133Figure 19: Figure 20: Immunoelectron microscopic detection of L-selectin on an interstitial PMN inrabbit lung following a bronchial instillate of E. coil endotoxin mixed withcolloidal carbon. No gold particles are located on the surface of this PMN orwithin the cytoplasmic granules (*). Collagen fibres (arrows) adjacent to thesurface of the PMN clearly establish the interstitial location of this PMN. N =nuclear lobe. Scale = 0.5 gm.Immunoelectron microscopic detection of L-selectin on an airspace PMN inrabbit lung following a bronchial instillate of E. coli endotoxin mixed withcolloidal carbon. No gold particles are present on the surface of this PMN orwithin the cytoplasmic granules (*). Interestingly, this PMN appears to haveformed an adhesive contact (arrows) with the alveolar epithelial surface. N =nuclear lobe. Scale = 0.5 gm.134661,135Figure 21: Immunoelectron microscopic detection of CD18 on an intravascular PMN inrabbit lung following a bronchial instillate of S. pneumoniae mixed withcolloidal carbon. Large numbers of gold particles are evenly distributed all overthe entire cell surface (arrows), as well as in the cytoplasmic granules (*). Thelabeling density on the cell surface is greater than that on control intravascularPMNs (see Figure 12). N = nuclear lobe. Scale = 0.25 gm.136137Figure 22: Immunoelectron microscopic detection of CD18 on an interstitial PMN in rabbitlung following a bronchial instillate of S. pneumoniae mixed with colloidalcarbon. The gold label density on the cell surface is similar to that onintravascular PMNs within the pneumonic region (see Figure 21). Many of thecytoplasmic granules (*) show gold particle labeling along the inner membranesurface. Some portions of the PMN cell surface appear to be in contact withthe interstitial collagen fibres (Type 1 collagen, arrows). This observation isinteresting because CD18 can serve as a receptor for Type I collagen. AlthoughCD 18-mediatedadherence to collagen can not be required for CD 18-independentPMN emigration, it may augment PMN migration through the interstitium. N= nuclear lobe. Scale = 0.25 gm.I38139Figure 23: Immunoelectron microscopic detection of CD18 on an airspace PMN in rabbitlung following a bronchial instillate of S. pneumoniae mixed with colloidalcarbon. The gold label density on the cell surface is much greater than thatobserved on intravascular (Figure 21) or interstitial (Figure 22) PMNs withinthe pneumonic region. The gold particles are widely distributed over the entirecell surface of this airspace PMN, except where the surface is in contact withcolloidal carbon (arrow). A few cytoplasmic granules (*) show gold labelingalong their inner membrane surface. Note the absence of gold particles on thephagosomal membrane surrounding an ingested bacterium (star). N = nuclearlobe. Scale = 0.25 Am.tL4-0141Figure 24: Immunoelectron microscopic detection of CD18 on an intravascular PMN in thecontralateral, non-pneumonic lung of a rabbit that received a bronchial instillateof S. pneumoniae mixed with colloidal carbon. Few gold particles are presenton the cell surface (arrowheads) and the density of gold label is similar to thatof control intravascular PMNs (Figure 12). Remarkably, the vast majority ofgold particles are located over the cytoplasmic granules (*) and most of theseparticles are associated with the inner membrane surface of the granule. N =nuclear lobe. Scale = 0.25 Am.14P..143Figure 25:^Immunoelectron microscopic detection of CD18 on an interstitial PMN in rabbitlung following a bronchial instillate of E. coil endotoxin mixed with colloidalcarbon. Along the entire free surface of this PMN, numerous CD 18-specificgold particles can be seen (arrowheads). The labeling density appears to besomewhat higher than that on control intravascular PMNs (Figure 12). Notethat the cell surface of the PMN is in contact with another interstitial cell(arrowheads). The identity of this second cell is unknown, but it is probablyanother leukocyte as evidenced by the surface gold labeling of its plasmamembrane. Interestingly, in spite of the observation that both of these cells areCD18 positive, no gold particles are found along the contact region (whitearrows). N = nuclear lobe. Black arrows = collagen fibres. Scale = 0.5 Am.144145Figure 26: Immunoelectron microscopic detection of CD18 on an airspace PMN in rabbitlung following a bronchial instillate of E. coil endotoxin mixed with colloidalcarbon. Large numbers of gold particles are present along the entire cellsurface (arrowheads). The gold label density is much greater than that oncontrol intravascular PMNs (Figure 12). In spite of this increase in surfaceCD 18 immunoreactivity, large numbers of gold particles are still detected withinthe cytoplasmic granules (*) and these particles are clearly associated with theinner membrane surface of the granules. N = nuclear lobe. Scale = 0.5 Am.146fr147Figure 27:^Light micrograph of mouse lung tissue 4h after the tracheal instillation of asaline solution of colloidal carbon. A few alveolar macrophages (arrows) arepresent, but the the alveolar airspace shows no evidence of PMN infiltration andthese lungs are not different from that of normal mice (not shown). Ar =bronchial artery. Br = bronchiole. Stain = Toluidine blue 0. Magnification= 200X.Figure 28: Light micrograph of alveolar macrophages in mouse lung tissue 4h after thetracheal instillation of a saline solution of colloidal carbon. Each macrophage(arrows) ingested colloidal carbon as evidenced by the golden-browncytoplasmic accumulations. Br = bronchiole. Stain = Toluidine blue 0.Magnification = 500X.149Figure 29:Figure 30: Light micrograph of mouse lung tissue 24h after the tracheal instillation of asaline solution of colloidal carbon. The appearance of the lung is similar to thatof the lung at 4h after colloidal carbon instillation (Figures 27 and 28). Again,a few alveolar macrophages are seen within the airspace (arrows) and there isno evidence of PMN infiltration. Ar = bronchial artery. Br = bronchiole.Stain = Toluidine blue 0. Magnification = 200X.Light micrograph of alveolar macrophages in mouse lung tissue 24h after thetracheal instillation of a saline solution of colloidal carbon. Note the cytoplasmicaccumulations of golden-brown colloidal carbon particles within each of thealveolar macrophages (arrows). Stain = Toluidine blue 0. Magnification =500X.150151Figure 31: Light micrograph of mouse lung tissue 4h after the tracheal instillation of a S.pneumoniae mixed with colloidal carbon. Alveolar macrophages ingestedcolloidal carbon (large arrowhead) and this confirmed the delivery of S.pneumoniae to the lung region shown. Large numbers of interstitial (smallarrowheads) and airspace (arrows) PMNs can be seen. Airspace PMNs showlittle evidence of colloidal carbon ingestion. The dotted-line denotes theboundary between the bronchovascular interstitium and the alveolar airspace.Ar = bronchial artery. Br = bronchiole. Stain = Toluidine blue 0.Magnification = 500X.Figure 32: Light micrograph of an alveolar PMN infiltrate in mouse lung 4h after thetracheal instillation of a S. pneumoniae mixed with colloidal carbon. NumerousPMN aggregates can be seen within the alveolar airspace (arrows) and thesePMNs again show no evidence of colloidal carbon ingestion. The alveolar wallsare poorly defined amongst the extravasated red blood cells (*). Stain =Toluidine blue 0. Magnification = 800X.153Figure 33: Figure 34: Light micrograph of mouse lung tissue 24h after the tracheal instillation of a E.coil endotoxin mixed with colloidal carbon. The dotted-line denotes theboundary between the bronchovascular interstitium and the alveolar airspace.Note the presence of PMNs within the bronchovascular interstitium(arrowheads). Alveolar macrophages containing ingested colloidal carbon arepresent and this confirms the delivery of E. coil endotoxin to this region of thelung. Ar = bronchial artery. Br = bronchiole. Stain = Toluidine blue 0.Magnification = 500X.Light micrograph of an alveolar PMN infiltrate in mouse lung tissue 4h after thetracheal instillation of a E. coil endotoxin mixed with colloidal carbon. Theairspace contains both PMNs (arrows) and alveolar macrophages (arrowheads).Unlike the streptococcal pneumonias (Figures 31 and 32), no extravasated redblood cells are present. The presence of colloidal carbon within the alveolarmacrophages confirms the delivery of E. coil endotoxin to the lung region.Note the minimal amount of carbon ingestion by the airspace PMNs.Magnification = 500X.155Figure 35: Figure 36: Electron micrograph showing a cross-section of a pulmonary capillary in normalmouse lung tissue. This section has been immunogold labeled for ICAM-1, butat this magnification, gold particles can not be distinguished. A small regionof this pulmonary capillary has been enlarged in Figure 36 for the purpose ofillustrating ICAM-1 expression on the endothelium. R = red blood cell. Scale= 1.0 gm.Immunoelectron microscopic detection of ICAM-1 on the pulmonary capillaryendothelium in normal mouse lung. Note that this micrograph is an enlargementof the pulmonary capillary shown in Figure 35. Very few gold particles(arrows) are found on the lumenal surface of the capillary endothelium. R =red blood cell. Scale = 0.25 gm.157Figure 37: ^Immunoelectron microscopic detection of ICAM-1 on the alveolar epithelial cellsurface in normal mouse lung. The expression of ICAM-1 is largely restrictedto the Type I pneumocytes (arrows). Very few gold particles are present on thesurface of the Type II pneumocyte (II). Scale = 0.5 Am.1St159Figure 38:^Electron micrograph contrasting the difference in ICAM-1 immunoreactivitybetween a Type I and II pneumocyte in normal mouse lung. Large numbers ofimmunogold particles are located on the surface of the Type I pneumocyte(black arrows) but not on the Type II pneumocyte (II). Note how theimmunogold labeling abruptly ceases at the cellular junction (white arrows)between the two cells. A lamellar body (*) containing surfactant can beappreciated within the cytoplasm of the Type II pneumocyte. Scale =0.25 Am.I0161Figure 39: Immunoelectron microscopic detection of ICAM-1 in mouse lung 24h after atracheal instillation of E. coli endotoxin mixed with colloidal carbon.Immunogold labeling is strictly extracelluar, and the gold particles areexclusively detected on lumenal cell surfaces. Significantly, in contrast tonormal mouse lung (Figures 37 and 38), large numbers of immunogold particlesare present on the surface (arrows) of the Type II pneumocyte (II). ICAM-1immunoreactivity is also increased on the pulmonary capillary endothelium(small arrowheads) when compared to normal mouse lung (Figure 36). Thegold label density on the surface of the Type I pneumocyte appears to be lessthan that observed in normal mouse lung (Figures 37 and 38). Scale = 0.5 Am.1 4, a163Figure 40: Immunoelectron microscopic detection of ICAM-1 on a Type II pneumocyte inmouse lung 4h after a tracheal instillation of S. pneumoniae mixed with colloidalcarbon. In contrast to normal mouse lung (Figures 37 and 38), numerous goldparticles are found (arrows) on the surface of the Type II pneumocyte (II). Thisincrease in ICAM-1 immunoreactivity is confined to the lumenal cell surface.No immunogold labeling is seen over the cytoplasm or intracellular organelles.165Figure 41: Figures 42: Figure 43: Immunoelectron microscopic detection of L-selectin on normal human PMNs ina suspension of leukocyte-rich plasma. L-selectin immunoreactivity isconcentrated on microvilli (arrows) of the PMN surface and is absent fromgranule membranes (*). Scale = 0.25 Am.Immunoelectron microscopic detection of total sLex immunoreactivity on normalhuman PMNs in a suspension of leukocyte-rich plasma. In contrast to L-selectin(Figure 41), total sLex immunoreactivity is found on both the plasma membraneand granule membranes (*). Gold particles do not show a preference forsurface microvilli (arrows) and are often associated with the flatter regions ofthe cell surface (arrowheads). Scale = 0.25 Am.Immunoelectron microscopic detection of CD18 immunoreactivity on normalhuman PMNs in a suspension of leukocyte-rich plasma. In contrast to L-selectin(Figure 41), CD18 immunoreactivity is found on both the plasma membrane andgranule membranes (*). Gold particles do not show a preference for surfacemicrovilli (arrow) and are often associated with the flatter regions of the cellsurface (arrowheads). Scale = 0.25 Am.4I 6 b167Figure 44:^Electron micrograph of an IL-1 stimulated human endothelial monolayer 5minutes after the addition of unstimulated human PMNs. The endothelialmonolayer (arrows) is resting on a layer of collagen fibres (*). Two PMNs arein the field of view. One is non-adherent (A), while the other (B) has adheredto the endothelial surface. This cryosection was immunogold labeled for L-selectin, but gold particles can not be appreciated at this magnification. For thepurpose of viewing the L-selectin immunoreactivity, magnified views of cell"A" and cell "B" are shown in Figures 45 and 46, respectively. Scale = 2.0Am.101169Figure 45:^Immunoelectron microscopic detection of L-selectin on a human PMN that hasnot yet adhered to an IL-1-stimulated endothelial monolayer. Note that thismicrograph is an enlargement of cell "A" in Figure 44. L-selectinimmunoreactivity is restricted to the cell surface. Gold particles are largelyconcentrated on the surface microvilli (arrows), with fewer gold particles beingdetected on the flatter regions of the cell surface (arrowheads). Scale = 0.5m.140171Figure 46: Immunoelectron microscopic detection of L-selectin on a human PMN that hasadhered to an IL-1-stimulated endothelial monolayer. Note that this micrographis an enlargement of cell "B" in Figure 44. L-selectin immunoreactivity(arrowheads) is restricted to that portion of the cell surface that is not in contactwith the endothelium. The adherent portion of the PMN (arrows) is flattenedand tightly apposed to the endothelium. No gold particles can be seen along thiscontact region. N = nuclear lobe. Scale = 0.5 pm.1 14. a173Figure 47: Electron micrograph of a human PMN that was engaged in transmigrating theIL-1 stimulated endothelial monolayer (E). The PMN has extended a pseudopodthrough the endothelial monolayer (arrows). A gap exists between themonolayer and the layer of collagen fibres (*). Note that this cryosection wasimmunogold labeled for L-selectin and magnified views of these immunogoldparticles are provided in Figures 48 and 49. Scale 2.0 Am.Figure 48: Immunoelectron microscopic detection of L-selectin on a portion of atransmigrating human PMN that has not yet penetrated the IL-1-stimulatedendothelial monolayer. Note that this micrograph is an enlargement of a portionof the transmigrating PMN in Figure 47. L-selectin immunoreactivity isconspicuously present on the surface microvilli (arrows), as well as flatterregions of the cell surface (arrowheads). Gold particles are not found within thecytoplasmic granules (*). N = nuclear lobe. Scale = 0.25 Am.Figure 49: Immunoelectron microscopic detection of L-selectin on a portion of atransmigrating human PMN that has penetrated an IL-1-stimulated endothelialmonolayer. Note that this micrograph is an enlargement of a portion of thetransmigrating PMN in Figure 47. L-selectin immunoreactivity is not detectedwithin the cytoplasmic granules (*). The gold label density on the cell surfaceappears to be reduced compared to that portion of the cell that had not yetpenetrated the endothelium (Figure 48). Importantly, L-selectinimmunoreactivity is still detected on the most distal tip of the pseudopod thathas penetrated the endothelium (arrows). Scale =0.25 it .t34.•4kg• •175Figure 50: Electron micrograph of several human PMNs that have penetrated the IL-1stimulated endothelial monolayer (E) within 5 minutes. Three PMN profiles (A,B, and C) can be seen lying within a large space between the endothelialmonolayer and the layer of collagen fibres (*). This cryosection wasimmunogold labeled for L-selectin and a magnified view of the immunogoldparticles is provided in Figure 51. Scale = 3.0 Am.Figure 51: Immunoelectron microscopic detection of L-selectin on PMNs that havecompletely penetrated the IL-1-stimulated endothelial monolayer. Note that thismicrograph is an enlargement of cells "A" and "B" in Figure 50. At thismagnification, it is apparent that these PMNs occupy a space that is between theendothelium and the basal lamina (arrowheads). Small amounts of L-selectinimmunoreactivity are still detected on the cell surface (arrows) even thoughthese PMNs have completely transmigrated the endothelium. No gold particlesare located over the cytoplasmic granules (*). Scale 1.0 Am.•177Figure 52: Immunoelectron microscopic detection of non-immune mouse IgG labeling ofa non-adherent human PMN as a control for non-specific gold labeling. Notethe lack of gold particle labeling on the surface microvilli (arrows) and flatterregions of the cell surface (arrowheads). In addition, no gold particles can beseen over the cytoplasmic granules (*). These observations confirm thespecificity and sensitivity of the immunogold labeling technique for the detectionof L-selectin immunoreactivity on human PMNs. Scale = 0.5 Am.179REFERENCES Abbassi, 0., Lane, C.L., Krater, S., Kishimoto, T.K., Anderson, D.C., McIntire, L.V.,and Smith, C.W. (1991). Canine neutrophil margination mediated by lectin adhesionmolecule-1 in vitro. J. Immunol. 147:2107-2115.Adamson, I.Y.R. and Bowden, D.H. (1974). The type 2 cell as progenitor of alveolarepithelial regeneration. Lab. Invest. 30:35-42.Adamson, I.Y.R. and Bowden, D.H. (1975). Derivation of Type 1 epithelium from Type2 cells in the developing rat lung. Lab. Invest. 32:736-745.Adamson, I.Y.R. and Bowden, D.H. (1978). Adaptive responses of the pulmonarymacrophage system to carbon. II. Morphologic studies. Lab. Invest. 38:430-438.Adamson, I.Y.R. and Bowden, D.H. (1980). Role of monocytes and interstitial cells in thegeneration of alveolar macrophages. II. Kinetic studies after carbon loading. Lab.Invest. 42:518 -524.Adamson, I. Y.R. and Bowden, D.H. (1982). Chemotactic and mitogenic components ofthe alveolar macrophage response to particles and neutrophil chemoattractant. Am. J.Pathol. 109:71 -77.Albert, R.K., Embree, L.J., McFreely, J.E., and Hickstein, D.D. (1992). Expression andfunction of 82 integrins on alveolar macrophages from human and nonhuman primates.Am. J. Respir. Cell Mol. Biol. 7:182 -189.Allison, F.Jr., Smith, M.R., and Wood, W.B.Jr. (1955). Studies on the pathogenesis ofacute inflammation. J. Exp. Med. 102:655-668.Altieri, D.C. and Edgington, T.S. (1988). A monoclonal antibody reacting with distinctadhesion molecules defines a transition in the functional state of the receptorCD11b/CD18 (Mac-1). J. Immunol. 141:2656-2660.Andrew, W. (1965). Comparative Hematology. Grune and Stratton, New York.Arfors, K.-E., Lundberg, C., Lindbom, L., Lundberg, K., Beatty, P.G. and Harlan, J.M.(1987). A monoclonal antibody to the membrane glycoprotein complex CD18 inhibitspolymoiphonuclear leukocyte accumulation and plasma leakage in vivo. Blood 69:338-340.Argenbright, L.W., Letts, L.G. and Rothlein, R. (1991). Monoclonal antibodies to theleukocyte membrane CD18 glycoprotein complex and to intercellular adhesionmolecule-1 inhibit leukocyte-endothelial adhesion in rabbits. J. Leuk. Biol. 49:253-257.180Arnaout, M.A. (1990a). Leukocyte adhesion molecules deficiency: Its structural basis,pathophysiology and implications for modulating the inflammatory response. Immunol.Rev. 114:145-180.Arnaout, M.A. (1990b). Structure and function of the leukocyte adhesion moleculesCD11/CD18. Blood 75:1037-1050.Arnaout, M.A., Spits, H., Terhorst, C., Pitt, J., and Todd III, R.F. (1984). Deficiency ofa leukocyte surface glycoprotein (LFA-1) in two patients with Mol deficiency. J. Clin.Invest. 74:1291-1300.Arnaout, M.A., Hakim, R.M., Todd III, R.F., Dana, N., and Colten, H.R. (1985).Increased expression of an adhesion-promoting surface glycoprotein in thegranulocytopenia of hemodialysis. New. Engl. J. Med. 312:457-462.Atherton, A. and Born, G.V.R. (1972). Quantitative investigations of the adhesivenessof circulating polymorphonuclear leucocytes to blood vessel walls. J. Physiol. 222:447-474.Atherton, A. and Born, G.V.R. (1973). Relationship between the velocity of rollinggranulocytes and that of the blood flow in venules. J. Physiol. 233:157-165.Bainton, D.F., Miller, L.J., Kishimoto, T.K., and Springer, T.A. (1987). Leukocyteadhesion receptors are stored in peroxidase-negative granules of human neutrophils.J.Exp.Med. 166:1641-1653.Barton, R.W., Rothlein, R., Ksiazek, J., and Kennedy, C. (1989). The effect of anti-intercellular adhesion molecule-1 on phorbol-ester-induced rabbit lung inflammation.J. Immunol. 143:1278-1282.Beatty, P.G., Ledbetter, J.A., Martin, P.J., Price, T.H., and Hansen, J.A. (1983).Definition of a common leukocyte cell-surface antigen (Lp95-150) associated withdiverse cell-mediated immune functions. J. Immunol. 131:2913-2918.Beesley, J.E., Pearson, J.D., Hutchings, A., Carleton, J.S. , and Gordon, J.L. (1979).Granulocyte migration through endothelium in culture. J. Cell Sci. 38:237-248.Beutler, B., Milsark, I.W., and Cerami, A.C. (1985). Passive immunization againstcachectin/tumor necrosis factor protects mice from lethal effects of endotoxin. Science229:869-871.Bevilacqua, M.P., Pober, J.S. , Mendrick, D.L., Cotran, R.S., and Gimbrone Jr., M.A.(1987). Identification of an inducible endothelial-leukocyte adhesion molecule. Proc.Natl. Acad. Sci. USA 84:9238-9242.181Bevilacqua, M.P., Stengelin, S., Gimbrone Jr. M.A., and Seed, B. (1989). Endothelialleukocyte adhesion molecule 1: An inducible receptor for neutrophils related tocomplement regulatory proteins and lectins. Science 243:1160-1165.Bevilacqua, M.P. and Gimbrone Jr., M.A. (1990). Identification and characterization ofendothelial-leukocyte adhesion molecule 1. In Leukocyte Adhesion Molecules. Eds.T.A. Springer, D.C. Anderson, A.S. Rosenthal and R. Rothelein. Springer-Verlag,New York. pp. 215-223.Bevilacqua, M.P., Corless, C., and Lo, S.K. (1991). Endothelial-leukocyte adhesionmolecule-1 (ELAM-1): A vascular SELECTIN that regulates inflammation. In:Cellular and Molecular Mechanisms of Inflammation. Vascular Adhesion Molecules.Vol.2. Eds. C.G. Cochrane and M.A. Gimbrone, Jr.. Academic Press, Inc., SanDiego. pp. 1-13.Beyers, N., Doerschuk, C.M., Coxson, H.O., and Hogg, J.C. (1989). Neutrophil volumeand capillary segment diameter in the rabbit, the dog and the human. Am. Rev. Respir.Dis. 139:A297.Bonfanti, R., Furie, B.C., Furie, B., and Wagner, D.D. (1989). PADGEM (GMP140) isa component of Weibel-Palade bodies of human endothelial cells. Blood 73:1109-1112.Bowden, D.H. and Adamson, I.Y.R. (1978). Adaptive responses of the pulmonarymacrophagic system to carbon. I. Kinetic studies. Lab. Invest. 38:422-429.Boyum, A. (1974). Separation of blood leukocytes, granulocytes, and lymphocytes.Tissue Antigens 4:269-274.Brandes, M.E. and Finkelstein, J.N. (1989). Stimulated rabbit alveolar macrophagessecrete a growth factor for Type II pneumocytes. Am. J. Respir. Cell Mol. Biol.1:101 -109.Bruce, D. (1894). On the disappearance of the leukocytes from the blood, afterinjection of peptone. Communicated by Professor Horsley. Proc. Royal Soc. 55:295-299.Buyon, J.P., Abramson, S.B., Philips, M.R., Slade, S.G., Ross, G.D., Weissman, G., andWinchester, R.J. (1988). Dissociation between increased surface expression ofGp165/95 and homotypic neutrophil aggregation. J. Immunol. 140:3156-3160.Buyon, J.P. , Abramson, S.B., Philips, M.R., Slade, S.G., Ross, G.D., Weissman, G., andWinchester, R.J. (1990). Mechanism regulating recruitment of CD11b/CD18 to thecell surface is distinct from that which induces adhesion in homotypic neutrophilaggregation. In: Leukocyte Adhesion Molecules. Eds. T.A. Springer, D.C. Anderson,A.S. Rosenthal and R. Rothelein. Springer-Verlag, N.Y. pp. 72-83.182Carlos, T.M. and Harlan, J.M. (1990). Membrane proteins involved in phagocyteadherence to endothelium. Immunol. Rev. 114:5-28.Christman, J.W., Petras, S.F., Vacek, P.M., and Davis, G.S. (1989). Rat alveolarmacrophage production of chemoattractants for neutrophils: Response to Escherichiacoli endotoxin. Infect. Immun. 57:810-816.Clark, E.R. and Clark, E.L. (1935). Observations on changes in blood vascularendothelium in the living animal. Amer.J.Anat. 57:385-438.Cohnheim, J. (1882). Lectures on General Pathology. A Handbook for Practitionersand students, Vol. 1, New Sydenham Soc., London (translated into English, 1889).Coxson, H.O., Markos, J., English, D., and Doerschuk, C.M. (1989). Neutrophilmargination and migration in streptococcal pneumonia in rabbits. Amer. Rev. Respir.Dis. 141:A652.Cotran, R.S., Gimbrone Jr., M.A., Bevilacqua, M.P., Mendrick, D.L., Pober, J.S.(1986). Induction and detection of a human endothelial activation antigen in vivo.J.Exp. Med. 164:661-666.Cotran, R.S., Kumar, V., Robbins, S.L. (1989). Robbins Pathologic Basis of Disease. 4thedition. W.B. Saunders Co., Philadelphia. pp. 39-86.Crowley, C.A., Curnutte, J.T., Rosin, R.E., Andre-Schwartz, J., Gallin, J.I., Klempner,M., Snyderman, R., Southwick, F.S., Stossel, T.P., and Babior, B.M. (1980). Aninherited abnormality of neutrophil adhesion: Its genetic transmission and its associationwith a missing protein. New Eng.J.Med. 302:1163-1180.Deavers, S.I. (1984). The mechanical aspects of respiration. Chpt. 28. In: Contemporary Medical Physiology. Ed. Vick, R.L.. Addison-Wesley Publishing Company. pp.424-447.Detmers, P.A., Wright, D.D., Olasen, E., Kimball, B. and Cohn, Z.A. (1987).Aggregation of complement receptors on human neutrophils in the absence of ligand.J.Cell Biol. 105:1137-1144.Diamond, M.S., Staunton, D.E., de Fougerolles, A.R., Stacker, S.A., Garcia-Aguilar, J.,Hibbs, M.L., and Springer, T.A. (1990). ICAM-1 (CD54): A counter-receptor forMac-1 (CD11b/CD18). J. Cell Biol. 111:3129-3139.Diamond, M.S., Staunton, D.E., Marlin, S.D., and Springer, T.A. (1991). Binding of theintegrin Mac-1 (CD11b/CD18) to the third immunoglobulin-like domain of ICAM-1(CD54) and its regulation by glycosylation. Cell 65:961-971.183Doerschuk, C.M. (1992). The role of CD 18-mediated adhesion in neutrophil sequestrationinduced by infusion of activated plasma in rabbits. Am. J. Respir. Cell Mol. Biol.7:140-148.Doerschuk, C.M., Allard, M.F., Martin, B.A., MacKenzie, A., Autor, A.P. and Hogg,J.C. (1987). Marginated pool of neutrophils in rabbit lungs. J. App!. Physiol.63:1806-1815.Doerschuk, C.M., Allard, M.F., and Hogg, J.C. (1989). Neutrophil kinetics in rabbitsduring infusion of zymosan-activated plasma. J. App!. Physiol. 67:88-95.Doerschuk, C.M., Winn, R.K. and Harlan, J.M. (1990a). Mechanisms of neutrophilemigration. In: Leukocyte Adhesion Molecules.  Eds. T.A. Springer, D.C. Anderson,A.S. Rosenthal and R. Rothelein. Springer-Verlag, New York. pp. 87-94.Doerschuk, C.M., Winn, R.K., Coxson, H.O. and Harlan, J.M. (1990b). CD18-dependent and independent mechanisms of neutrophil emigration in the pulmonary andsystemic microcirculation of rabbits. J.Immunol. 144:2327-2333.Doerschuk, C.M., Quinlan, W.M., and Takei, F. (1992). The role of CD11a, CD11b, andICAM-1 in leukocyte emigration and edema induced by intraperitoneal instillation ofS. pneumoniae in mice. FASEB J. 6:A1430.Downey, G.P. and Worthen, G.S. (1988). Neutrophil retention in model capillaries:deformability, geometry, and hydrodynamic forces. J. Appl. Physiol. 65:1861-1871.Dustin, M.L., Rothlein, R., Bhan, A.K., Dinarello, C.A., and Springer, T.A. (1986).Induction by IL-1 and interferon-gamma: Tissue distribution, biochemistry, and functionof a natural adherence molecule (ICAM-1). J. Immunol. 137:245-254.Dutrochet, M.H. (1824). Recherches anatomiques et physiologiques sur la structure intimedes animaux et des vegetaux, et sur leur motilite.  J.-.B Bailliere, Paris. pp. 214-215.Dutrochet, M.H. (1837). Memoires pour servir a l'histoire anatomique et physiologiquedes vegetaux et des animaux. J.-.B Bailliere, Paris. pp.502-503.English, D. and Andersen, B.R. (1974). Single-step separation of red blood cells,granulocytes and mononuclear leukocytes on discontinuous density gradients of Ficoll-Hypaque. J. Immunol. Meth. 5:249-252.Evans, M.J., Cabral, L.J., Stephens, R.J., and Freeman, G. (1973). Renewal of alveolarepithelium in the rat following exposure to NO2. Amer. J. Path. 70:175-198.Etzioni, A., Frydman, M., Pollack, S., Avidor, I., Phillips, M.L., Paulson, J.C., andGershoni-Baruch, R. (1992). Brief Report: Recurrent severe infections caused by anovel leukocyte adhesion deficiency. New Engl. J. Med. 327:1789-1792.184Foxall, C., Watson, S.R., Dowbenko, D., Fennie, C., Lasky, L.A., Kiso, M., Hasegawa,A., Asa, D., and Brandley, B.K. (1992). The three members of the selectin receptorfamily recognize a common carbohydrate epitope, the sialyl Lewis' oligosaccharide.J. Cell Biol. 117:895-902.Furie, M.B., Cramer, E.B., Naprstek, B.L., and Silverstein, S.C. (1984). Culturedendothelial cell monolayers that restrict the transendothelial passage of macromoleculesand electrical current. J. Cell Biol. 98:1033 -1041.Furie, M.B. and McHugh, D.D. (1989). Migration of neutrophils across endothelialmonolayers is stimulated by treatment of the monolayers with interleukin-1 or tumornecrosis factor-a. J. Immunol. 143:3309 -3317.Furie, M.B., Burns, M.J., Tancinco, M.C.A., Benjamin, C.D., and Lobb, R.R. (1992).E-selectin (endothelial-leukocyte adhesion molecule-1) is not required for the migrationof neutrophils across IL-1 stimulated endothelium in vitro. J. Immunol. 148:2395-2404.Gallatin, W.M., Weissman, II., and Butcher, E.C. (1983). A cell-surface moleculeinvolved in organ-specific homing of lymphocytes. Nature 304:30-34.Geng, J.-G., Bevilacqua, M.P., Moore, K. L., McIntyre, T. M., Prescott, S.M., Kim,J.M., Bliss, G.A., Zimmerman, G.A., and McEver, R.P. (1990). Rapid neutrophiladhesion to activated endothelium mediated by GMP-140. Nature 343:757-760.Gimbrone Jr., M.A. (1975). Culture of vascular endothelium. Prog. HemostasisThromb. 3:1-28.Glantz, S.A. (1987). Primer of Biostatistics. 2nd edition. McGraw-Hill Book Co., NewYork. 379 pages.Grant, L. (1973). The sticking and emigration of white blood cells in inflammation.In: The Inflammatory Process, 2nd edition. Eds. B.W. Zweifach, L. Grant, and R.T.McCluskey, Academic Press, Orlando. pp. 205-249.Griffiths, G., McDowall, A., Back, R. and Dubochet, J. (1984). On the preparation ofcryosections for immunocytochemistry. J. Ultrastruct. Res. 89:65 -78.Gundel, R.H., Wegner, C.D., Torcellini, C.A., Clarke, C.C., Haynes, N., Rothlein, R.,Smith, C.W., and Letts, L.G. (1991). Endothelial leukocyte adhesion molelcule-1mediates antigen-induced acute airway inflammation and late-phase airway obstructionin monkeys. J. Clin. Invest. 88:1407-1411.Harlan, J. M. (1985). Leukocyte-Endothelial Interactions. Blood 65:513 -525.185Harlan, J.M., Winn, R.K., Vedder, N.B., Doerschuk, C.M., and Rice, C.L. (1992). Invivo models of leukocyte adherence to endothelium. In: Adhesion. Its Role inInflammatory Disease. Eds. J.M. Harlan and D.Y. Liu. W.H. Freeman & Company,New York. pp. 117-150.Hattori, R., Hamilton, K.K., Fugate, R.D., McEver, R.P., and Sims, P.J. (1989a).Complement proteins C5b-9 induce secretion of high molecular weight multimers ofendothelial von Willebrand factor and translocation of granule membrane protein GMP-140 to the cell surface. J. Biol. Chem. 264:9053-9060.Hattori, R., Hamilton, K.K., Fugate, R.D., McEver, R.P., and Sims, P.J. (1989b).Stimulated secretion of endothelial von Willebrand factor is accomplished by rapidredistribution to the cell surface of the intracellular granule membrane protein GMP-140. J. Biol. Chem. 264:7768-7771.Hermanowsld-Vosatka, A., Van Strijp, J.A.G., Swiggard, W.J., and Wright, S.D. (1992).Integrin modulating factor-1: A lipid that alters the function of leukocyte integrins.Cell 68:341-352.Hogg, J. (1987). Neutrophil kinetics and lung injury. Physiol. Rev. 67:1249-1295.Hoogeveen, Y.L., Smit, J.W., Blom, N.R., and Halle, R.M. (1988). Gold probe choice insimultaneous detection of human lymphocyte surface antigens at the ultrastructurallevel. J. Histochem. Cytochem. 36:821-825.Huang, A.J., Furie, M.B., Nicholson, S.C., Fischbarg, J., Liebovitch, L.S., andSilverstein, S.C. (1988). Effects of human neutrophil chemotaxis across humanendothelial cell monolayers on the permeability of these monolayers to ions andmacromolecules. J. Cell. Physiol. 135:355-366.Hummel, D.S., Berninger, R.W., Tomasz, A., and Winkelstein, J.A. (1981). The fixationof C3b to pneumococcal cell wall polymers as a result of activation of the alternativecomplement pathway. J. Immunol. 127:1287-1289.Hummel, D.S., Swift, A.J., Tomasz, A., and Winkelstein, J.A. (1985). Activation of thealternative complement pathway by pneumococcal lipoteichoic acid. Infect. Immun.47:384-387.Hynes, R.O. (1992). Integrins: Versatility, modulation, and signaling in cell adhesion.Cell 69:11-25.Issekutz, A.C., Morzycki, W., and Sadowska, J. (1991). Rabbit alveolar macrophagesstimulated with endotoxin and lung fragments from endotoxemic rabbits produce aleukocyte infiltration-inducing factor that lacks IL-1, TNFa, or chemotactic activity.Exp. Lung. Res. 17:803-819.186Jaffe, E.A., Hoyer, L.W., and Nachman, R.L. (1973). Synthesis of antihemophilic factorby cultured human endothelial cells. J. Clin. Invest. 52:2757-2764.Johnston, G.I., Cook, R.G., and McEver, R.P. (1989). Cloning of GMP-140, a granulemembrane protein of platelets and endothelium: Sequence similarity to proteins involvedin cell adhesion and inflammation. Cell 56:1033-1044.Jones, D.H., Anderson, D.C., Burr, B.L., Rudolff, H.E., Smith, C.W., Krater, S.S., andSchmalstieg, F.C. (1988). Quantitation of intracellular Mac-1 (CD11b/CD18) poolsin human neutrophils. J. Leuk. Biol. 44:535-544.Jones, D.H., Schmalstieg, F. C., Hawkins, H.K., Burr, B.L., Rudloff, H.E., Krater, S.,Smith, C.W. and Anderson, D.C. (1990a). Characterization of a new mobilizableMac-1 (CD11b/CD18) pool that co-localizes with gelatinase in human neutrophils. InLeukocyte Adhesion Molecules,  Eds. T.A. Springer, D.C. Anderson, A.S. Rosenthaland R. Rothelein. Springer-Verlag, New York. pp. 106-124.Jones, D.H., Schmalstieg, F. C., Dempsey, K., Krater, S. 5, Nannen, D.D., Smith, C.W.and Anderson, D.C. (1990b). Subcellular distribution and mobilization of MAC-1(CD11b/CD18) in neonatal neutrophils. Blood 75:488-498.Junqueira, L.C., Carneiro, J., and Kelley, R.O. (1989). Respiratory system. Chpt. 17.Basic Histology. 6th edition. Appelton and Lange. pp.334-353.Jutila, M.A., Rott, L., Berg, E.L. and Butcher, E.C. (1989). Function and regulation ofthe neutrophil MEL-14 antigen in vivo: Comparison with LFA-1 and MAC-1.J.Immunol. 143:3318-3324.Jutila, M.A., Kishimoto, T.K. and Butcher, E.C. (1990). Regulation and lectin activityof the human neutrophil peripheral lymph node homing receptor. Blood 76:178-183.Jutila, M.A., Kishimoto, T.K. and Finken, M. (1991). Low-dose chymotrypsintreatment inhibits neutrophil migration into sites of inflammation in vivo: Effects onMac-1 and MEL-14 adhesion protein expression and function. Cell.Immunol. 132:201-214.Kansas, G.S., Spertini, 0., and Tedder, T.F. (1991). Leukocyte adhesion molecule-1(LAM-1): Structure, Function, Genetics, and Evolution. In: Cellular and MolecularMechanisms of Inflammation. Vascular Adhesion Molecules.  Vol.2. Eds. C.G.Cochrane and M.A. Gimbrone, Jr.. Academic Press, Inc. San Diego. pp. 31-59.Kishimoto, T.K., Jutila, M.A., Berg, E.L., and Butcher, E.C. (1989a). Neutrophil Mac-1and MEL-14 adhesion proteins inversely regulated by chemotactic factors. Science245:1238-1241.187Kishimoto, T.K., Larson, R.S., Corbi, A.L., Dustin, M.L., Staunton, D.E. and Springer,T.A. (1989b). The leukocyte integrins. Adv. Immunol. 46:149-182.Kishimoto, T.K., Jutila, M.A., and Butcher, E.C. (1990). Identification of a humanperipheral lymph node homing receptor: A rapidly down-regulated adhesion molecule.Proc. Natl. Acad. Sci. USA. 87:2244-2248.Kishimoto, T.K., Warnock, R.A., Jutila, M.A., Butcher, E.C., Lane, C., Anderson,D.C., and Smith, C.W. (1991). Antibodies against human neutrophil LECAM-1(LAM-1/Leu-8/DREG-56 antigen) and endothelial cell ELAM-1 inhibit a commonCD18-independent adhesion pathway in vitro. Blood 78:805-811.Kuhn III, C. and Finke, E.H. (1972). The topography of the pulmonary alveolus: Scanningelectron microscopy using different fixations. J. Ultrastruct. Res. 38:161-173.Kuijpers, T.W., Hakkert, B.C., Hart, M.H.L., and Roos, D. (1992a). Neutrophilmigration across monolayers of cytokine-prestimulated endothelial cells: a role forplatelet-activating factor and IL-8. J. Cell Biol. 117:565-572.Kuijpers, T.W., Hoogerwerf, M., and Roos, D. (1992b). Neutrophil migration acrossmonolayers of resting or cytokine-activated endothelial cells. Role of intracellularcalcium changes and fusion of specific granules with the plasma membrane. J.Immunol. 148:72-77.Lackie, J.M. and Smith, R.P.C. (1980). Interactions of leukocytes and endothelium. In:Cell Adhesion and Motility, (Symposium of the British Society for Cell Biology; 3).Eds. A.S.G. Curtis and J.D. Pitts. Cambridge University Press, Cambridge. pp. 235-272.Larsen, G., Sako, D., Ahern, T.J., Shaffer, M., Erban, J., Sajer, S.A., Gibson, R.M.,Wagner, D.D., Furie, B.C., and Furie, B. (1992). P-selectin and E-selectin. Distinctbut overlapping leukocyte ligand specificities. J. Biol. Chem. 267:11104-11110.Lasky, L.A. (1992). The homing receptor (LECAM 1/L Selectin): A carbohydrate-bindingmediator of adhesion in the immune system. In: Adhesion. Its Role in InflammatoryDisease. Eds. J.M. Harlan and D.Y. Liu. W.H. Freeman & Company, New York.pp. 43-63.Lasky, L.A., Singer, M.S., Yednock, T.A., Dowbenko, D., Fennie, C., Rodriguez, H.,Nguyen, T., Stachel, S., and Rosen, S.D. (1989). Cloning of a lymphocyte homingreceptor reveals a lectin domain. Cell 56:1045-1055.188Lawrence, M.B., Smith, C.W., Eskin, S.G. and McIntire, L.V. (1990). Effect of venousshear stress on CD18-mediated neutrophil adhesion to cultured endothelium. Blood75:227-237.Leeuwenberg, J.F.M., Tan, A., Jeunhomme, T.M.A.A., Ploegh, H.L., and Buurman,W.A. (1991). The ligand recognized by ELAM-1 on HL60 cells is not carried by N-linked oligosaccharides. Eur. J. Immunol. 21:3057-3059.Lewinsohn, D.M., Bargatze, R.F., and Butcher, E.C. (1987). Leukocyte-endothelial cellrecognition: Evidence of a common molecular mechanism shared by neutrophils,lymphocytes, and other leukocytes. J. Immunol. 138:4313-4321.Ley, K., Gaehtgens, P., Fennie, C., Singer, M.S., Lasky, L.A. and Rosen, S.D. (1991).Lectin-like cell adhesion molecule 1 mediates leukocyte rolling in mesenteric venulesin vivo. Blood 77:2553-2555.Lien, D.C., Wagner, W.W., Capen, R.L., Haslet, C., Hanson, W.L., Hofmeister, S.E.,Henson, P.M. and Worthen, S.G. (1987). Physiological neutrophil sequestration in thelung: visual evidence for localization in capillaries. J.Appl.Physiol. 62:1236-1243.Lien, D.C., Worthen, S.G., Capen, R.L., Hanson, W.L., Checkley, L.L., Janke, S.J. ,Henson, P.M. and Wagner, W.W. (1990). Neutrophil kinetics in the pulmonarymicrocirculation. Effects of pressure and flow in the dependent lung.Amer.Rev.Resp.Dis. 141:953 -959.Lien, D.C., Henson, P.M., Capen, R.L., Henson, LE., Hanson, W.L., Wagner,Jr.,W.W., and Worthen, G.S. (1991). Neutrophil kinetics in the pulmonarymicrocirculation during acute inflammation. Lab. Invest. 65:145-159.Lindbom, L., Lundberg, C., Prieto, J., Raud, J., Nortamo, P., Gahmberg, C.G., andPatarroyo, M. (1990). Rabbit leukocyte adhesion molecules CD11/CD18 and theirparticipation in acute and delayed inflammatory responses and leukocyte distribution invivo. Clin. Immunol. Immunopath. 57:105-119.Lo, S.K., Van Seventer, G.A., Levin, S.M., and Wright, S.D. (1989). Two leukocytereceptors (CD 11a/CD18 and CD1 lb/CD18) mediate transient adhesion to endotheliumby binding to different ligands. J. Immunol. 143:3325-3329.Lo, S.K., Detmers, P.A., Levin, S. M., and Wright, S.D. (1989b). Transient adhesion ofneutrophils to endothelium. J. Exp. Med. 169:1779- 1793.Lobb, R.R. (1992). Integrin-immunoglobulin superfamily interactions in endothelial-leukocyte adhesion. In: Adhesion. Its Role in Inflammatory Disease.  Eds. J.M. Harlanand D. Y. Liu. W.H. Freeman & Company, New York. pp. 1-18.189Loos, M., Clas, F., and Fischer, W. (1986). Interaction of purified lipoteichoic acid withthe classical complement pathway. Infect. Immun. 53:595-599.Loosli, C.G. and Baker, R.F. (1962). Acute experimental pneumococcal (Type I)pneumonia in the mouse: The migration of leucocytes from the pulmonary capillariesinto the alveolar spaces as revealed by the electron microscope. Trans. Am. Clin.Climatol. Assoc. 74:15 -28.Lundberg, C. and Wright, S.D. (1990). Relation of the CD11/CD18 family of leukocyteantigens to the transient neutropenia caused by chemoattractants. Blood 76:1240-1245.Luscinskas, F.W., Cybulsky, M.I., Kiely, J.-M., Peckins, C.S., Davis, V.M., andGimbrone Jr., M.A. (1991). Cytokine-activated human endothelial monolayers supportenhanced neutrophil transmigration via a mechanism involving both endothelial-leukocyte adhesion molecule-1 and intercellular adhesion molecule-1. J. Immunol.146:1617 - 1625.McEver, R.P., Beckstead, J.H. , Moore, K.L., Marshall-Carlson, L., and Bainton, D.F.(1989). GMP-140, a platelet a-granule membrane protein, is also synthesized byvascular endothelial cells and is localized in Weibel-Palade bodies. J. Clin. Invest.84:92-99.McEver, R.P. (1991). Leukocyte interactions mediated by GMP-140. In: Cellular andMolecular Mechanisms of Inflammation. Vascular Adhesion Molecules.  Vol.2. Eds.C.G. Cochrane and M.A. Gimbrone, Jr.. Academic Press, Inc. Harcourt BraceJovanovich, Publishers. San Diego, CA., pp. 15-29.Martin, B.A., Wiggs, B., Lee, S. and Hogg, J.C. (1987). Regional differences inneutrophil margination in dog lungs. J.Appl.Physiol. 63:1253 -1261.Metchnikoff, E. (1893). Lectures on the Comparative Pathology of Inflammation.Delivered at the Pasteur Institute in 1891. Translated by F.A. Starling and E.H.Starling. Kegan, Paul, Trench Trubner and Co., Ltd., London. pp. 1-218.Metchnikoff, E. (1905). Immunity in infective diseases. Translated by F.G. Binnie.Cambridge University Press, London. pp. 544-569.Mileski, W., Harlan, J., Rice, C., and Winn, R. (1990). Streptococcus pneumoniae-stimulated macrophages induce neutrophils to emigrate by a CD 18-independentmechanism of adherence. Circ. Shock 31:259-267.Monboisse, J.-C., Garnotel, R., Randoux, A., Dufer, J., and Borel, J.-P. (1991).Adhesion of human neutrophils to and activation by Type-I collagen involving 132integrin. J. Leuk. Biol. 50:373 -380.190Montgomery, K.F., Osborn, L., Hession, C., Tizard, R., Goff, D., Vassallo, C., Tarr,P.I., Bomsztyk, K., Lobb, R., Harlan, J.M., and Pohlman, T.H. (1991). Activationof endothelial-leukocyte adhesion molecule 1 (ELAM-1) gene transcription. Proc. Natl.Acad. Sci. USA 88:6523-6527.Mood, A.M., Graybill, F.A. and Boes, D.C. (1974). Introduction to the Theory ofStatistics. 3rd edition, McGraw Hill, New York, New York. pp. 180-181.Moore, K.L., Varki, A., and McEver, R.P. (1991). GMP-140 binds to a glycoproteinreceptor on human neutrophils: evidence for a lectin-like interaction. J. Cell Biol.112:491-499.Moore, K.L., StuIts, N.L., Diaz, S., Smith, D.F., Cummings, R.D., Varki, A., andMcEver, R.P. (1992). Identification of a specific glycoprotein ligand for P-selectin(CD62) on myeloid cells. J. Cell Biol. 118:445-456.Mulligan, M.S., Varani, J., Dame, M.K., Lane, C.L., Smith, C.W., Anderson, D.C. andWard, P.A. (1991). Role of endothelial-leukocyte adhesion molecule 1 (ELAM-1) inneutrophil-mediated lung injury in rats. J. Clin. Invest. 88:1396-1406.Munro, J.M., Pober, J.S., and Cotran, R.S. (1991). Recruitment of neutrophils in thelocal endotoxin response: Association with de novo endothelial expression of endothelialleukocyte adhesion molecule-1. Lab Invest. 64:295-299.Nourshargh, S., Rampart, M., Hellewell, P.G., Jose, P.J., Harlan, J.M., Edwards, A.J.,and Williams, T.J. (1989). Accumulation of 111In-neutrophils in rabbit skin in allergicand non-allergic inflammatory reactions in vivo. Inhibition by neutrophil pretreatmentin vitro with a monoclonal antibody recognizing the CD18 antigen. J. Immunol.142:3193-3198.O'Shea, J.J., Brown, E.J., Seligmann, B.E., Metcalf, J.A., Frank, M.M., and Gallin, J.I.(1985). Evidence for distinct intracellular pools of receptors for C3b and C3bi inhuman neutrophils. J. Immunol. 134:2580-2587.Oswald, I.P., Afroun, S., Bray, D., Petit, J.-F., and Lemaire, G. (1992). Low responseof BALB/c macrophages to priming and activating signals. J. Leuk. Biol. 52:315-322.Paulson, J. C. (1992).^Selectin/Carbohydrate-mediated adhesion of leukocytes.In:Adhesion. Its Role in Inflammatory Disease. Eds. J.M. Harlan and D.Y. Liu. W.H.Freeman & Company, New York. pp. 19-42.Peters, A.M., Saverymuttu, S.H., Bell, R.N. and Lavender, J.P. (1985). Quantification ofthe distribution of the marginating granulocyte pool in man. Scand. J. Haematol.4:111-120.191Philips, M.R., Buyon, J.P., Winchester, R., Weissman, G., and Abramson, S.B. (1988).Up-regulation of the iC3b receptor (CR3) is neither necessary nor sufficient to promoteneutrophil aggregation. J. Clin. Invest. 82:495-501.Picker, L.J. , Warnock, R.A., Burns, A.R., Doerschuk, C.M., Berg, E.L. and Butcher,E.C. (1991). The neutrophil selectin LECAM-1 presents carbohydrate ligands to thevascular selectins ELAM-1 and GMP-140. Cell 66:921-933, 1991.Pober, J.S. , Bevilacqua, M.P., Mendrick, D.L., Lapierre, L.A., Fiers, W., and Gimbrone,Jr., M.A. (1986a). Two distinct monoldnes, interleukin-1 and tumor necrosis factor,each independently induce biosynthesis and transient expression of the same antigen onthe surface of cultured human vascular endothelial cells. J. Immunol. 136:1680 -1687.Pober, J.S., Gimbrone, Jr., Lapierre, L.A., Mendrick, D.L., Piers, W., Rothlein, R. andSpringer, T.A. (1986b). Overlapping patterns of activation of human endothelial cellsby interleukin 1, tumor necrosis factor, and immune interferon. J. Immunol. 137:1893 -1896.Price, T.J., Beatty, P.G. and Corpuz, S.R. (1987). In vivo inhibition of neutrophilfunction in the rabbit using monoclonal antibody to CD 18. J.Immunol. 139:4174-4177.Prieto, J., Takei, F., Gendelman, R., Christenson, B., Biberfield, P., and Patarroyo, M.(1989). MALA-2, mouse homologue of human adhesion molecule ICAM-1 (CD54).Eur. J. Imunol. 19:1551-1557.Rankin, J.A., Sylvester, I., Smith, S., Yoshimura, T., and Leonard, E.J. (1990).Macrophages cultured in vitro release leukotriene 134 and neutrophil attractant/activationprotein (Interleukin-8) sequentially in response to stimulation with lipopolysaccharideand zymosan. J. Clin. Invest. 86:1556-1564.Rhodes, G.C., Tapsall, J.W., and Lykke, A.W.J. (1989). Alveolar epithelial responses inexperimental streptococcal pneumonia. J. Pathol. 157:347-357.Rietschel, E.Th., Kirikae, T., Feist, W., Loppnow, H., Zabel, P., Brade, L., Ulmer, A.J.,Brade, H., Seydel, U., Zahringer, U., Schlaak, M., Flad, H.D., and Schade, U.(1991). Molecular aspects of the chemistry and biology of endotoxin. In: 42.Colloquium - Mosbach. Molecular Aspects of Inflammation.  Eds. Sies, H., Flohe, L.,and Zimmer, G.. Springer-Verlag. pp. 207-231.Robinson, M.K., Andrew, D., Rosen, H., Brown, D., Ortlepp, S., Stephens, P., andButcher, E.C. (1992). Antibody against the Leu-CAM fl-chain (CD18) promotes bothLFA-1 and CR3-dependent adhesion events. J. Immunol. 148:1080 -1085.192Rosen, H. and Gordon, S. (1990). The role of the type 3 complement receptor in theinduced recruitment of myelomonocytic cells to inflammatory sites in the mouse. Am.J. Respir. Cell Mol. Biol. 3:3-10.Rothlein, R. and Springer, T. A. (1986). The requirement for lymphocyte function-associated antigen 1 in homotypic leukocyte adhesion stimulated by phorbol ester. J.Exp. Med. 163:1132-1149.Rothlein, R., Dustin, M.L., Marlin, S.D., and Springer, T.A. (1986). A humanintercellular adhesion molecule (ICAM-1) distinct from LFA-1. J. Immunol. 137:1270-1274.Schierbeek, A. (1962).^Antoni Van Leeuwenhoek on the Circulation of Blood.Nieuwkoop, B. DeGraaf, page 33.Schleiffenbaum, B., Moser, R., Patarroyo, M. and Fehr, J. (1989). The cell surfaceglycoprotein Mac-1 (CD11b/CD18) mediates neutrophil adhesion and modulatesdegranulation independently of its quantitative cell surface expression. J.Immunol.142:3527-3545.Schmid-Schonbein, G. W. , Shih, Y. Y. , and Chien, S. (1980a). Morphometry of humanleukocytes. Blood 56:866-875.Schmid-Schonbein, G.W., Usami, S., Skalak, R., and Chien, S. (1980b). The interactionof leukocytes and erythrocytes in capillary and postcapillary vessels. Microvasc. Res.19:45-70.Sherman, M.P., Johnson, J.T. , Rothelein, R., Hughes, B.J., Smith, C.W., and Anderson,D.C. (1992). Role of pulmonary phagocytes in host defense against group Bstreptococci in preterm versus term rabbit lung. J. Infect. Dis. 166:818-826.Siegelman, M.H., Van De Rijn, M., and Weissman, I.L. (1989). Mouse lymph nodehoming receptor cDNA clone encodes a glycoprotein revealing tandem interactiondomains. Science 243:1165-1172, 1989.Smith, C.W. (1990). Molecular determinants of neutrophil adhesion. Am. J. Respir.Cell Mol. Biol. 2:487-489.Smith, C.W. (1992). Transendothelial migration. In: Adhesion. Its Role in InflammatoryDisease. Eds. J.M. Harlan and D. Y. Liu. W.H. Freeman & Company, New York. pp.83-115.Smith, C.W., Rothlein, R., Hughes, B.J. , Mariscalo, M.M., Rudloff, H.E., Schmalstieg,F.C., and Anderson, D.C. (1988). Recognition of an endothelial determinant forCD 18-dependent human neutrophil adherence and transendothelial migration. J. Clin.Invest. 82:1746-1756.193Smith, C.W., Marlin, S.D., Rothlein, R., Toman, C., and Anderson, D.C. (1989).Cooperative interactions of LFA-1 and Mac-1 with intercellular adhesion molecule-1in facilitating adherence and transendothelial migration of human neutrophils in vitro.J. Clin. Invest. 83:2008-2017.Smith, C.W., Kishimoto, T.K., Abbassi, 0., Hughes, B., Rothlein, R., McIntire, L.V.,Butcher, E., and Anderson, D.C. (1991). Chemotactic factors regulate lectin adhesionmolecule 1 (LECAM-1)-dependent neutrophil adhesion to cytokine-stimulatedendothelial cells in vitro. J. Clin. Invest. 87:609-618.Springer, T.A., Miller, L.J., and Anderson, D.C. (1986). p150,95, the third member ofthe Mac-1, LFA-1 human leukocyte adhesion glycoprotein family. J. Immunol.136:240-245.Stacker, S.A. and Springer, T.A. (1991). Leukocyte integrin P150,95 (CD11c/CD18)functions as an adhesion molecule binding to a counter-receptor on stimulatedendothelium. J. Immunol. 146:648-655.Staunton, D.E, Marlin, S.D., Dustin, M.L. and Springer, T.A. (1988). Primary structureof ICAM-1 demonstrates interaction between members of the immunoglobulin andintegrin supergene families. Cell 52:925-933.Staunton, D.E, Dustin, M.L. and Springer, T.A. (1989). Functional cloning of ICAM-2,a cell adhesion ligand for LFA-1 homologous to ICAM-1. Nature 339:61-64.Staunton, D.E., Dustin, M.L., Erickson, H.P., and Springer, T.A. (1990).^Thearrangement of the immunoglobulin-like domains of ICAM-1 and the binding sites forLFA-1 and rhinovirus. Cell 61:243-254.Sylvester, I., Rankin, J.A., Yoshimura, T., Tanaka, S., and Leonard, E.J. (1990).Secretion of neutrophil attractant/activation protein by lipolysaccharide-stimulated lungmacrophages determined by both enzyme-linked immunosorbent assay and N-terminalsequence analysis. Am. Rev. Respir. Dis. 141:683-688.Takei, F. (1985). Inhibition of mixed lymphocyte response by a rat monoclonal antibody toa novel murine lymphocyte activation antigen (MALA-2). J. Immunol. 134:1403-1407.Tangelder, G.J. and Arfors, K.-E. (1991). Inhibition of leukocyte rolling in venules byprotamine and sulfated polysaccharides. Blood 77:1565-1571.Tedder, T.F., Penta, A.C., Levine, H.B., Freedman, A.S. (1989). Expression of thehuman leukocyte adhesion/homing molecule, LAM-1: Identity with the TQl and Leu-8differentiation antigens. J. Immunol. 144:532-540.194Thurlbeck, W.M. and Miller, R.R. (1988). The respiratory system. In: Pathology. Eds.Rubin, E. and Farber, J.L. J.B. Lippincott Company. page 546.Todd III, R.F., Arnaout, M.A., Rosin, R.E., Crowley, C.A., Peters, W.A., and Babior,B.M. (1984). Subcellular localization of the large subunit of Mol (Mola; formerly gp110), a surface glycoprotein associated with neutrophil adhesion. J. Clin. Invest.74:1280-1290.Tokuyasu, K.T. (1973). A technique for ultramicrotomy of cell suspensions and tissues.J. Cell Biol. 57:551-565.Tokuyasu, K.T. (1978). A study of positive staining of ultrathin frozen sections.J. Ultrastruct. Res. 63:287-307.Tokuyasu, K.T. (1983). Present state of immunocryoultramicrotomy. J. Histochem.Cytochem. 31:164-167.Totterman, T.H. and Festin, R. (1989). Immunogold labeling for the single-laser FACSanalysis of triple antibody-binding cells. In: Colloidal Gold. Principles. Methods. andApplications. Vol. 2. Ed. M.A. Hayat. Academic Press, Inc., San Diego. pp.431-442.Uhal, B.D., Flowers, K.M., and Rannels, D.E. (1991). Type II pneumocyte proliferationin vitro: problems and future directions. Am. J. Physiol. Suppl. 261:110-117.Van Bergen en Henegouwen, P.M. (1989). Immunogold labeling of ultrathin cryosections.In: Colloidal Gold. Principles. Methods, and Applications. Vol. 1. Ed. M.A. Hayat.Academic Press, Inc., San Diego. pp.191-216.Vedder, N.B. and Harlan, J.M. (1988). Increased surface expression of CD11b/CD18(Mac-1) is not required for stimulated neutrophil adherence to cultured endothelium.J.Clin.Invest. 81:676-682.von Andrian, U.H., Chambers, J.D., McEvoy, L.M., Bargatze, R.F., Arfors, K.-E. andButcher, E.C. (1991). Two-step model of leukocyte-endothelial cell interaction ininflammation: Distinct roles for LECAM-1 and the leukocyte 02-integrins in vivo.Proc.Natl.Acad.Sci. USA 88:7538-7542 .von Haller, A. (1756). Deux Memoires sur le Mouvement du Sang, et sur les Effetsde la Saignee: Fondes sur des Experiences faites sur des Animaux. M.-M. Bousquet,Paris, France, pp.248-249.Voyata, J.C., Via, D.P., Butlerfield, C.E., and Zetter, B.R. (1984). Identification andisolation of endothelial cells based on their increased uptake of acetylated low densitylipoprotein. J. Cell Biol. 99:2034-2040.195Walker, D.C., Nisbet, C.L., and Chu, F. (1992). Polymorphonuclear leukocyte (PMN)migration through alveolar interstitium during streptococcal pneumonia in rabbits.FASEB J. 6:A742.Wallis, W.J., Hickstein, D.D., Schwartz, B.R., June, C.H., Ochs, H.D., Beatty, P.G.,Klebanoff, S.J., and Harlan, J.M. (1986). Monoclonal antibody-defined functionalepitopes on the adhesion-promoting glycoprotein complex (CDw18) of humanneutrophils. Blood 67:1007-1013.Wegner, C.D., Gundel, R.H., Reilly, P., Haynes, N., Letts, L.G., and Rothlein, R.(1990). Intercellular adhesion molecule-1 (ICAM-1) in the pathogenesis of asthma.Science 247:456-459.Weibel, E.R. (1963). Morphometry of the Human Lung.  Academic Press, New York. pp.78-82.Wilkinson, L. (1990). SYSTAT: The System for Statistics. Evanston, IL: SYSTAT, Inc.,1990.Williams, T.J., Jose, P.J., Forrest, M.J., Wedmore, C.V., and Clough, G.F. (1984).Interactions between neutrophils and microvascular endothelial cells leading to cellemigration and plasma protein leakage. In: White Cell Mechanics. Basic Science andClinical Aspects. H.J. Meiselman, M.A. Lichtman, and P.L. LaCelle, eds. Alan R.Liss, Inc., New York, pp. 195-208.Winkelstein, J.A. and Tomasz, A. (1978). Activation of the alternative complementpathway by pneumococcal cell wall teichoic acid. J. Immunol. 120:174- 178.Worthen, S.G., Schwab, B., Elson, E.L. and Downey, G.P. (1989). Mechanicalproperties of stimulated neutrophils: cell stiffening induces retention in capillaries.Science 245:183 -186.Wright, S.D. and Meyer, B.C. (1986). Phorbol esters cause sequential activation anddeactivation of complement receptors on polymorphonuclear leukocytes. J. Immunol.136:1759 -1764.Wright, S.D., Lo, S.K., and Detmers, P.A. (1990). Specificity and regulation of CD18-dependent adhesions. In Leukocyte Adhesion Molecules. Eds. T.A. Springer, D.C.Anderson, A.S. Rosenthal and R. Rothelein. Springer-Verlag, New York. pp. 190-207.Yoder, M.C., Checldey, L.L., Giger, U., Hanson, W.L., Kirk, K.R., Capen, R.L., andWagner, W.W. Jr. (1990). Pulmonary microcirculatory kinetics of neutrophilsdeficient in leukocyte adhesion-promoting glycoproteins. J. Appl. Physiol. 69:207-213.196Zar, J.H. (1984). Biostatistical Analysis. 2nd edition. Prentice-Hall, Inc., EnglewoodCliffs. 718 pages.Zimmerman, G.A., Prescott, S.M., and McIntyre, T.M. (1992). Endothelial cellinteractions with granulocytes: tethering and signaling molecules. Immunol. Today13:93-100.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086351/manifest

Comment

Related Items