Novel activation of neutrophil functions

Chronic neutrophil activation results in severe tissue damage leading to loss of limb and life as illustrated from events ranging from adult respiratory distress syndrome (ARDS) to the inflammatory skin lesions associated with Staphylococcal and Streptococcal infections. Conversely, PMNs are critical for host protection, as individuals lacking functional PMNs, in the case of Leukocyte Adhesion Deficiency (LAD), neutropenia, and Chronic Granulomatous Disease (CGD), succumb to persistent bacterial infections. The efficacy of PMNs in host protection derives from their specialized functions. Activation of PMNs upregulates the quantity and affinity of their ?2-integrins with a concomitant loss of L-selectin. The quantitative increase in ?2-integrin upon activation is dependent upon fusion of intracellular granules containing ?2-integrin with the plasma membrane at the site of noxious insult. This also delivers proteases and essential components of the NADPH oxidase complex to the membrane that is necessary for efficient bacterial killing. This dissertation examines novel mechanisms of PMN activation. In the first study neutrophil activation is dissected into two distinct pathways. I demonstrate that sphingomyelinase C (SMC) treatment with the subsequent increase in the intracellular signaling molecule, ceramide, illuminates a bifurcation in PMN activation: ceramide induces degranulation and superoxide generation in the absence of increased ?2-integrin function. SMC induces an increase in ?2-integrin surface expression but, unusually, the integrin is maintained in a low affinity state. This correlates with in vivo data showing SMC mediates dysfunctional PMN sequestration within the vasculature. In the second study I show ICAM-3 engagement by specific epitope-mapped antibodies induces cytoskeletal dependent functional upregulation of the ?2-integrins. Signaling through this adhesion molecule in leukocytes was previously poorly defined. We demonstrate phosphorylation events by PKC and tyrosine kinases of several proteins after engagement, whereas inhibitors of these kinases result in inhibition of the signal emanating from ICAM-3. In the final study, data are presented illustrating that oncostatin M is a proinflammatory cytokine whose effects are mediated through activation of endothelium. In conclusion, these studies illustrate the complex regulation of PMN activation by defining diverse signaling pathways resulting in activation of specific effector functions.

NOVEL ACTIVATION OF NEUTROPHIL FUNCTIONS by Michael John Feldhaus A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Experimental Pathology Department of Pathology The University of Utah August 1996 THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by Michael J. Feldhaus This dissertation has been read by each member of the following supervisory committee and by majority vote has been found to be satisfactory. I ~ ~n F. Bohnsack THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of the University of Utah: I have read the dissertation of Michael John Feldhaus in its [mal fonn and have found that (1) its fonnat, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the supervisory committee and is ready for submission to The Graduate School. Date Thomas M. McIntyre .' Chair, Supervisory Committee Approved for the Major Department Approved for the Graduate Council Dean of The Graduate School Copyright © Michael John Feldhaus 1996 All Rights Reserved ABSTRACT Chronic neutrophil activation results in severe tissue damage leading to loss of limb and life as illustrated from events ranging from adult respiratory distress syndrome (ARDS) to the inflammatory skin lesions associated with Staphylococcal and Streptococcal infections. Conversely, PMNs are critical for host protection, as individuals lacking functional PMNs, in the case of Leukocyte Adhesion Deficiency (LAD), neutropenia, and Chronic Granulomatous Disease (CGD), succumb to persistent bacterial infections. The efficacy of PMNs in host protection derives from their specialized functions. Activation of PMNs upregulates the quantity and affinity of their f32- integrins with a concomitant loss of L-selectin. The quantitative increase in f32- integrin upon activation is dependent upon fusion of intracellular granules containing f32-integrin with the plasma membrane at the site of noxious insult. This also delivers proteases and essential components of the NADPH oxidase complex to the membrane that is necessary for efficient bacterial killing. This dissertation examines novel mechanisms of PMN activation. In the first study neutrophil activation is dissected into two distinct pathways. I demonstrate that sphingomyelinase C (SMC) treatment with the subsequent increase in the intracellular signaling molecule, ceramide, illuminates a bifurcation in PMN activation: ceramide induces degranulation and superoxide generation in the absence of increased f32-integrin function. SMC induces an increase in f32-integrin surface expression but, unusually, the integrin is maintained in a low affinity state. This correlates with in vivo data showing SMC mediates dysfunctional PMN sequestration within the vasculature. In the second study I show ICAM-3 engagement by specific epitope-mapped antibodies induces cytoskeletal dependent functional upregulation of the ~2-integrins. Signaling through this adhesion molecule in leukocytes was previously poorly defined. We demonstrate phosphorylation events by PKC and tyrosine kinases of several proteins after engagement, whereas inhibitors of these kinases result in inhibition of the signal emanating from ICAM-3. In the final study, data are presented illustrating that oncostatin M is a proinflammatory cytokine whose effects are mediated through activation of endothelium. In conclusion, these studies illustrate the complex regulation of PMN activation by defining diverse signaling pathways resulting in activation of specific effector functions. v To Mom and Dad TABLE OF CONTENTS ABSTRACT .................................................................................................................... i v ABBREVIATIONS... ............ ........... .... ................... ... ....... ... ....................... .......... ... ..... x ACKN"OWLEDGMENTS ............................................................................................. xi Chapter I. IN''IRODUCTION ....... ...................... ......................... ............. ......... ..... ............. 1 The physiology of neutrophils and inflammation.............................. 1 P:MN adhesion molecule expression.................. .... .............. ........ ... ....... 4 Cytoskeleton........ .............. ........................... ......... .......................... ... ........... 8 Degranulation .............................................................................................. 10 Signal transduction in PMNs ................................................................... 13 Conclusions .................................................................................................. 17 References...... ........... .............. ........... ........ ..... ........ ....... ..... ..... ........... .... ...... 18 II. MATERIALS AND METHODS ...................................................................... 29 Materials ........................................................................................................ 29 Methods ......................................................................................................... 30 References ..................................................................................................... 34 ill. CERAMIDE SIGNALIN'G IN NEUTROPHILS ........................................... 35 Introduction ................................................................................................. 35 Results ........................................................................................................... 37 Discussion ..................................................................................................... 63 References ..................................................................................................... 68 IV. ICAM-3 OUTSIDE IN SIGNALING IN HUMAN NEUTROPHILS ....... 74 Introduction ................................................................................................. 74 Results ........................................................................................................... 78 Discussion .................................................................................................... 1 00 References .................................................................................................... 1 06 V. ONCOSTATIN M IS A PROINFLAMMATORY CYTOKINE ................ 112 Introduction ................................................................................................ 112 Results .......................................................................................................... 114 Discussion .................................................................................................... 123 References .................................................................................................... 125 viii BRSV CD DAG EC ECL ELISA ERK fMLP FITC HBSS HBSS/A HPLC HUVEC IL-XX ICAM LFA-l LPS mAbs MAP mg ABBREVIATIONS Brown Recluse Spider Venom Cluster of Differentiation Diacylglycerol Endothelial Cells Enhanced chemiluminescence Enzyme Linked Immunosorbent Assay Extracellular stimulus Regulated Kinase formyl-methionoyl-Ieucyl-phenyalanine Fluorescein Isothiocyana te Hank's Balanced Salt Solution Hank's Balanced Salt Solution with 0.5% albumin High Performance Liguid Chromatography Human Umbilical Vein Endothelial Cell InterLeukin number designation Intercellular Adhesion Molecule Leukocyte Function Associated antigen 1 Lipopolysaccharide Monoclonal antibodies Mitogen Activated Protein Milligram ml M NADPH OSM PAF PAGE PBS PKC PLA2 PMN PMA SDS SMC SMD TBST TNF ug ul milliliter Molar Nicotinamide Adenine Dinucleatide Phopshate Oncostatin M Platelet -activating Factor Polyacrylamide Gel Electrophoresis Phosphate Buffered Saline Protein Kinase C Phospholipase A2 Polymorphonuclear Neutrophil Phorbol Myristate Acetate Sodium Dodecyl Sulfate Sphingomyelinase C Sphingomyelinase D Tris Buffered Saline with 0.05% Tween 20 Tumor necrosis Factor microgram microliter x ACKNO~EDGEMrnNTS I wish to thank Drs. Thomas McIntyre, my research advisor, Steve Prescott, and Guy Zimmerman for guidance, support, and years of helpful discussions during the development of this thesis. I am also grateful to my committee members and all merrLbers of the Experimental Pathology Department that have provided a rich resource of knowledge, technical expertise, and help throughout the years. Research in this dissertation was greatly expedited by the expert technical assistance from Donelle Benson, Debbie Dykstra, Wen Hua Li. I am thankful to them and many other colleagues at CVRTlfor their help, time, insightful discussions, and expertise. I wish to thank my parents for all the love and for holding on. I know I was a more than difficult son at times, but their unconditional love, care, and support, when understanding was at a loss, are what has brought me here today. My brothers and sisters: Marty, Barbara, Nancy, Eric, Tony, Usa, and Ann, thank you all for the love and understanding. Finally, I greatly appreciate Jane for the strength to speak her mind, delivery of helpful critiques, and the relaxing interludes in the great outdoors. The work in this dissertation was funded by awards from Nora Eccles Treadwell Foundation and United States Public Health grant R01 HL 44513, and P50 HL50153. CHAPTER 1 INTRODUCTION The physiology of neutrophils and inflammation Immunologists today rarely consider the neutrophil more than a mundane effector cell. As Patrick Murphy so eloquently put it: "Like the legendary jackass of Missouri, he (the PMN) is of uncertain ancestry and without hope of progeny, and there is no mechanism by which the experience of an individual neutrophil could translate into altered function on the part of its decedents"(l). However, the importance of the PMN for human existence is well appreciated by those lacking functional PMNs as in the case of patients with leukocyte adhesion deficiency (LAD) - characterized by dysfunctional or absent ~2-integrins, or those patients suffering from chronic granulomatous disease (CGD), a disease characterized by dysfunctional generation of oxygen radicals (2-6). The neutrophil is a terminally differentiated cell with a half life of several hours in the blood stream and several days in tissues (7). The specificity of destruction of pathogens or damaged tissue by the PMN appears to be a stereotyped response independent of stimulus or target. However, recent advances in the elucidation of PMN activational states have painted a more complete picture of a complex cell. In fact, PMNs are now appreciated as cells able to synthesize proteins with the transcriptional regulation demonstrated by longer lived cells. Over 2000 years ago the classical macroscopic features of inflammation were laid down: redness, heat, swelling, pain, and loss of function (7). The 2 redness is due to dilatation of the blood vessels at the inflammatory foci as the small vessels become dilated and more permeable. The dilatation leads to increased blood flow although with lower shear forces that facilitate cell-cell interaction. It also results in temperatures in the area reaching that of aortic blood, thereby responsible for the "heat." The increased vascular permeability leads to plasma loss and red cell aggregates in the vessels that slow the flow of blood. The plama loss and red cell aggregation lead to the red flare and the swelling. The pain can in part be due to the production of the inflammatory mediators, serotonin, bradykinin, and prostaglandin E1. Finally, destruction of tissue at the inflammatory site can lead to loss of function. These five macroscopic features also coincide with activation of endothelial cells, with subsequent PMN activation and recruitment of several types of inflammatory cells to the locus. Postcapillary venules are the major site where leukocytes exit from the systemic vasculature to reach the underlying tissue bed. Upon appropriate activation, endothelial cells lining these vessel walls become pro-adhesive for quiescent circulating leukocytes. Leukocytes adhere to activated endothelial cells initially via shear resistant selectin molecules, and subsequently become activated by endothelial cell-produced agonists such as P AF and 1L-8 (Fig. 1.1). These agonists upregulate the affinity of the ~2-integrins present on the surface of the PMNs that are required for transmigration out of the vasculature, a process termed "inside-out signaling." A key function of neutrophils in the body is to ingest and destroy invading microorganisms. Elie Metchnikoff first described the phenomenon of phagocytosis by macrophages in 1891 in his Lectures on the Comparative Pathology of Inflammation delivered at the Pastuer 1nstituteand more formally presented in 1901 in Immunity in Infective Disease (8,9). When PMNs Post-Capillary Venule Blood Flow P2-Integrin Mediated Adhesion 3 CD31 and P2-Integrin Dependent Transnrigration Figure 1.1. Leukocyte extravasation. Quiescent PMNs in the blood roll upon activated endothelial cells expressing P- or E-selectin. PSGL-1 and other unidentfied fucosylated ligands interact with the selectins newly expressed on the endothelial cell. In addition, activated endothelial cells express the PMN agonists PAF or IL-8 on their surface, which stimulate PMN and upregulates the affinity of the ~2-integrins on the PMN. The activated integrins halt the rolling and facilitate blood flow-resistant adhesion. The activated PMNs then migrate to an endothelial cell junction where CD31 and ~2-integrins are required in exiting the vasculature. 4 phagocytose they employ three antimicrobial systems to remove and destroy invading organisms: adhesion molecules, granular constituents, and generation of hydrogen peroxide by the NADPH oxidase system. The interrelated nature of PMN adhesion, integrin activation, degranulation, and superoxide generation is a focus of this thesis. Events and causes of incomplete and differential PMN activation are also examined. PMN adhesion molecule expression Selectins. The three member selectin family of adhesion molecules all contains an extracellular lectin (sugar-binding) domain (10). L-selectin (Leukocyte-selectin) is present on lymphocytes, monocytes, and neutrophils. On PMNs, L-selectin is regulated by a proteolytic mechanism that leads to a rapid shedding from the surface of activated PMNs (11). P-selectin (Platelet-selectin) is stored in the alpha-granules of platelets and also the Wibel-Palade bodies of endothelial cells (12,13). P-selectin is expressed within minutes on the surface of platelets and endothelial cells upon activation by thrombin, histamine and other cell specific mediators (14,15). Newly synthesized E-selectin (Endothelial­selectin) is expressed on activated endothelial cells within several hours of activation by a separate group of endothelial cell agonists (16). All members are thought to participate in the selectin-dependent, shear-resistant rolling of PMNs along inflamed endothelium (17-19). In fact, there is evidence that L-selectin and E-selectin interact directly as counter receptors (20-22). The agonist induced proadhesive state of endothelial cells is, in part, mediated by the surface expression of P- and E-selectin adhesion molecules at early and late times, respectively (23). The P-selectin (and perhaps E-selectin) counterreceptors PSGL- 1 and other poorly defined counterreceptors are always present on unactivated 5 PMNs and therefore it is the selectin upregulation on endothelial cells that is responsible for localizing the inflammatory response (10,24). Integrins. Integrin family members are cation-dependent heterodimeric transmembrane glycoprotein receptors, consisting of an a and ~ subunit. The 16 known a subunits and 8 ~ subunits produce more than 20 different integrin receptors that mediate a plethora of functions for a myriad of cells (25,26). The major integrin expressed on PMNs is the ~2-integrin Mac1, also known as Mol, or CD1Ib/18. However, PMNs also express CDI1a/CD18, CD11c/CD18, and the recently identified CD11d/CD18 (27). ~l-integrins a4/~l, a5/~l, and a6/~1 are also expressed at low levels and are implicated in neutrophil adhesion to HUVEC, fibronectin and laminin, respectively (28-30). Further discussion will specifically address the ~2-integrins and their function as related to the human PMN. The a subunit of integrins is thought to impart substrate specificity, whereas the ~ subunit interacts with cytoskeletal components through a short (40-60 amino acid) cytoplasmic tail (Fig. 1.2)(25,31- 33). The major ~integrin present on PMNs, CD11b/CD18, is on the surface of quiescent PMNs in a low affinity state that does not support adhesion. However, upon activation of the PMN, CD11b/18 undergoes an affinity modulation into a higher affinity state that does support adhesion. How this is accomplished is not completely understood, but the importance of conformational changes, cations, and the cytoskeleton is demonstrated by their effects on the affinity of the receptor (29,34-36). Additionally, many cytoskeletal proteins associate with the integrin when in the high affinity state, suggesting other modulating influences may also occur. The majority of the CD11b/CD18 is stored in intracellular granules in the quiescent PMN, although the precise identity of these are the subject of some debate (37). Upon activation a 75% - 90% increase in surface expression is Low Affinity ~ a 6 High Affinity pp12SFAK pp60src Figure 1.2. Integrin structure and association with the cytoskeleton. Integrins are heterodimeric proteins containing regulatory cations binding domains. Substitution of Mn++ for Ca++ in the cation binding site induces one xhigh affinity state. The ~ subunits interact with the cytoskeleton when in the integrin is in the high affinity state. Many proteins that localize with the high affinity integrins are phosphorylated when associated. Other nonstructural proteins such as the protein kinases, pp125FAK and pp60src, also become associated in this complex. 7 observed that coincides with an increase in the adhesiveness of the PMN (38). The increased adhesiveness, at least under some conditions, is not facilitated by the newly arrived CD11b/18 but instead is due to functional upregulation of the preexisting CD11b/CD18 on the surface of quiescent PMNs (39,40). In the case of an initial suboptimal stimulation newly recruited CD11 b /18 does become involved in adhesion when the PMNs are exposed to subsequent higher concentration of chemotactic stimuli (41). However, in these experiments PMNs have been manipulated by nonphysiologic temperature and pharmacologic agents (41). These manipulations are important to consider when compared to studies using larger concentrations of agonist that stimulate recruitment and activation simultaneously. ICAMs. There are three members in the ICAM subfamily of immunoglobulin superfamily of proteins. PMNs express a small number of ICAM-1 molecules, no ICAM-2 and a large amount of ICAM-3 (42,43). ICAM-1, -2 -3 are all counterreceptors for CD11a/CD18 (44-46). ICAM-1 has been shown to be an undisputed receptor for CD11b/CD18, the most prevalent integrin on the neutrophil (47). However, there is some controversy as to whether or not ICAM-2 is also a counterreceptor for CD11b/CD18 (48,49). Endothelial cells express ICAM-1 constitutively and upregulate its surface expression when activated (50,51). ICAM-1 is also expressed by neutrophils and many other hematopoetic and nonhematopoetic cells. ICAM-1 upregulation on endothelial cells facilitates CD11/18-dependent shear-resistant binding prior to transmigration (50). ICAM-1 as a receptor for LFA-1 has been shown to be dependent on homodimerization of ICAM-1 and this dimerization may also be important for CD11b/18 binding to ICAM-1 (52,53). The glycosylation state of ICAM-1 also greatly affects binding by CD11b/CD18 and, as discussed below, 8 may affect certain conclusions regarding ICAM function based on experiments using recombinant ICAMs (52). The high level of ICAM-3 expression suggests it may have important immune functions for PMN s. However, the functional importance of this recently described member of the ICAM family has not been reported. This thesis addresses several novel aspects of the functionality of this receptor on PMNs. CD31 (PECAM-l). CD31 is a single polypeptide containing six Ig-like domains, a transmembrane domain, and a cytoplasmic taiL It is localized at cell­cell junctions of endothelial cells in the vascular compartment, and is also expressed by platelets, monocytes, macrophages, neutrophils and naive lymphocytes (54). Cell-cell adhesion mediated by CD31 can be homophilic (CD31 binding to another CD31) or heterophilic (CD31 binding to a different receptor) in nature (55-57). It has been shown that CD31 is important in leukocyte transendothelial migration at a step distal to CDllb/18 dependent event as a-CD31 monoclonal antibodies block transmigration, and so ligation and clustering participate in "outside-in signaling" (58). Cross-linking of CD31 has been shown to upregulate the affinity of CDllb/18 on PMNs, an example of "outside-in" signaling (59). CD31 may direct the PMN to the endothelial cell junction where the recently activatedCDllb/18 allows migration out of the vasculature. Cytoskeleton F-actin and bridging proteins. The first definition of a membrane skeleton originated with the observation that the nonionic detergent, Triton X-I00, disrupts hydrophobic, but not polar, protein-protein and protein-lipid interactions in the membrane of human red blood cells. The proteins that were 9 not solubilized by this treatment remained, creating an "empty shell" or ghost of the red blood cell, and are considered the cytoskeleton. More stringent criteria are now used to identify proteins that interact with the actin based cytoskeleton. Filamentous actin or f-actin is a multimer of globular-actin or g-actin. The rapid polymerization and depolymerization of f-actin to g-actin are an important feature that allows the dynamics of the cytoskeleton to change quickly and reversibly (60-63). The cytoskeleton can be described as a mesh work of highly organized f-actin and associated proteins just under, and closely associated with, the plasma membrane of cells. The f-actin based cytoskeleton interacts with the plasma membrane through a series of bridging proteins and is regulated through phosphorylation cascades. Talin, paxillin, <x-actinin, zyxin, and tensin are all involved in the interaction of f-actin and plasma membrane bound receptors (Figl.2)(64). The regulation of some of these proteins are mediated by kinases such as pp125FAK, PKC-<x and pp60src. The cytoplasmic domains of two adhesion proteins, integrins and ICAM-l in particular, have been shown to directly associate with the cytoskeleton through talin and <x-actinin (65-68). Disruption of the cytoskeleton with pharmacologic agents, such as cytochalasins, changes the function of these receptors and their associated kinase cascades (65,66,68,69). Cytochalasins are a group of naturally occurring fungal metabolites that bind actin and alter its polymerization. Functionally, cytochalasins resemble capping proteins, which bind the barbed end of actin filaments, thus inhibiting both association and dissociation of actin monomers (70). Some cytochalasins, A and B, are not as specific for actin as they also inhibit monosaccharide transport. Cytochalasins are powerful tools in elucidating cytoskeletal involvement of membrane receptors and responses. For example, cytochalasin D treatment of PMNs decreases the apparent affinity of PSGL-l for P-selectin as well as CD18 10 for fibrinogen, from which it is possible to infer PSGL-l and CDIS integrins must interact with the cytoskeleton to establish firm adhesion (71,72). In human PMNs, cytochalsin D potentiates degranulation, suggesting actin redistribution is important for granule fusion with the plasma membrane (73,74). The widely held belief that cytochalasins depolymerize actin is certai?ly not true in light of the report that cytochalsin D induces an increase in f-actin (75). In conclusion, cytochalasins that are used to determine the involvement of the cytoskeleton in specific cellular responses are supported, but by no means proven, by experimental observation. Degranulation Granule compartments and components. Exocytosis, degranulation and secretion are terms for the discharge of proteinaceous materials, found in lysosomes and other granule structures from granulocytes, into the extracellular milieu. The granule compartments in PMNs have been classified into three to four specific organelles based on their constituents (Fig. 1.3). Azurophilic, or primary granules, contain many of the microbicidal enzymes such as myeloperoxidase, lysozyme, elastase, and an assortment of other proteases. Specific, or secondary granules, contain a large intracellular reserve of CDllb/CDlS, gelatinase, collagenase, lactoferrin, vitamin B12, and the specific cytochrome b that is required for the formation of the NADPH oxidase electron transport chain and the production of superoxide (76). Secretory vesicles, or tertiary granules, contain alkaline phosphatase and several other overlapping constituents with the other storage organelles previously listed. A fourth granule compartment, containing only gelatinase, has been described, but the existence of such a unique compartment is disputed (76). It is important to note that rarely are the components of the different classes of granules actually excluded from the Quiescent eo 6) = 1°,2°,3° granules ~ = P2 -integrins r = L-selectin 1° Lysozyme Cathepsin Elastase 11 Activated 2° 3° Lysozyme Gelatinase Lactoferrin Cathepsin Collagenase Alkaline- ~2 -integrins phosphatase Cytochrome b558 Figure 1.3. PMN degranulation and granule constituents. Quiescent PMNs have an abundance of L-selectin on their surface that is shed upon PMN activation and subsequent delivery of an unidentified granule protease to the surface. In contrast, the ~2-integrin surface expression is greatly increased upon PMN activation. Release of granule constituents from the three classes of granules is dependent upon PMN activation. 12 others. A more accurate picture is that there is a continuum of granules with populations that have characteristics of other granules. This interpretation is supported by the plethora of papers that demonstrate, with a variety of methods, that specific granular constituents are present in differing amounts within several granule subsets (76-79). Degranulation-dependent quantitative CD11b/18 upregulation and L­selectin shedding. PMNs constitute the primary mobile cellular defense against intruding microorganisms. Essential components of their defense mechanisms are stored within the intracellular granules. The majority ofPMN CD11b/CD18 is stored in the secondary granules, and is only expressed on the surface when an agonist induces this granule subset to fuse with the plasma membrane and thereby increase its surface expression (37,38). The enhanced surface expression is important for the phagocytosis of invading microorganisms and transmigration through the endothelium. The enzyme responsible for the shedding of L-selectin from the surface of activated PMN s has not been identified, but L-selectin shedding correlates with degranulation from the gelatinase containing secondary granules (8,80). L-selectin shedding correlates with an inability to roll on E-selectin and P-selectin surfaces (20-23). The loss of L-selectin has been postulated to be necessary for efficient PMN transmigration into inflammatory sites by allowing the initial cell-cell interaction to mature by decreasing selectin mediated adhesion with the endothelial cell apical surface, and allowing integrin-dependent transmigration to and between endothelial cell junctions (81). Delivery of NADPH oxidase components to the plasma membrane. The respiratory burst is a distinguishing property of phagocytes. It results from the assembly of the NADPH oxidase, a transmembrane electron transport chain that reduces extracellular oxygen to superoxide with concomitant oxidation of 13 cytosolic NADPH generated by the hexose mono phosphate shunt (82,83). Oxidase activity-dependent formation of oxygen radicals, hydrogen peroxide, hypochlorous acid, and other oxidizing species is essential for the bactericidal function of neutrophils. Oxygen and other radicals modify unsaturated fatty acids and hydrogen peroxide and with the help of myeloperoxidase, also oxidatively deaminate and decarboxylate amino acids that aid in bacterial killing (84,85). The toxic nature of this pathway mandates that the respiratory burst be tightly controlled. Cytochrome b558 was the first component of the NADPH oxidase system to be identified (86). It is a heme-binding integral membrane protein comprised of a 22kd a-subunit and a 91kd ~-subunit (87,88). Cytochrome b558 is localized to the specific granules of resting PMNs and upon activation of the PMN is rapidly translocated to the plasma membrane through fusion of specific granules (Fig. 1.3)(89-91). It becomes part of the active NADPH oxidase which additionally contains the translocated cytosolic factors p47phox, p67phox, gp91 phox, and p22phox (92,93). Deletion or loss of function mutations in any of the four components leads to chronic granulomatous disease, indicating the generation of hydrogen peroxide, and its metabolites, is important for the ultimate PMN function (93,94). The GTP binding proteins Rap1, Rap2, and Rac1, have also been implicated in the activation of the NADPH oxidase (95-99). These are located in the specific granule with cytochrome b558 and may play an active role in regulating degranulation and assembly of the oxidase complex. Signal transduction in PMNs Intracellular free Ca++. Adherence, chemotaxis, phagocytosis, and bactericidal activities of PMNs are differentially regulated at the receptor and post receptor levels (100). Increases in intracellular Ca++ levels have been 14 implicated in degranulation and the oxidative burst; Ca++ increase is one of the first events detected upon chemoattractant receptor engagement (63,101,102). The intracellular concentration of Ca++ in resting PMNs is about 100nM, approximately 20,000-fold lower than extracellular levels (103,104). However, upon activation the intracellular concentration of uncomplexed Ca++ can rise to micromolar concentrations (105). The mechanism responsible for this agonist­induced increase in Ca ++ utilizes IP3-induced mobilization of intracellular stores, and enhanced Ca++ entry through nonselective ion transporter in the plasma membrane (106,107). Additional mechanisms of intracellular Ca++ release are mediated via cyclic ADP-ribose and sphingosine-l-phosphate (108- 111). The dependence of degranulation on Ca ++ levels has been well documented. Experiments have shown that the different granule subsets require different levels of Ca++ to induce release of their contents (107,112). How Ca++ mediates these effects is largely unknown. However PMNs contain the Ca ++­dependent Protein Kinase C ~ (PKC) which translocates to the cellular membrane upon an increase in intracellular Ca++. Once associated with the membrane it interacts with Diacylglycerol (DAG) formed from Ca ++ -dependent PLC hydrolysis of membrane phospholipids, and then is able to phosphorylate a number of proteins (113). This is one wayan increase in intracellular Ca++ can serve as a link between agonist receptors and PKC mediated phosphorylation cascades. PKC and tyrosine kinases. Protein kinase C (PKC) activity results from a family of closely related enzymes: the tissue specific distribution of the various isoforms suggests distinct functions for the various enzymes. When activated, PKC phosphorylates its substrates on serine or threonine. Some of its substrates are tyrosine kinases that further amplify the phosphorylation cascade. PMNs 15 express examples of all three classes of PKC. These include: 1) ~I and ~II, from the conventional PKC category, which require Ca++ and are activated by DAG and PMA, 2) n, from the novel isoenzymes category, which is Ca++ independent but is activated by DAG and phorbol myristate acetate (PMA), and 3) ~, an atypical PKC, which is Ca++ independent and is not activated by DAG or PMA (114,115). The roles of the specific isoforms of PKC have not been well defined in PMN activation, although nPKC is involved in actin assemble, and an unidentified PKC is linked to activation of the MAP kinase cascade (116,117). Inhibitors of PKC (H7 and chelerythrine), and of tyrosine kinases (herbimycin A and genistein), have been used to examine their involvement in a specific agonist induced signaling pathway. In conjunction with these inhibitors, immunoblots with phospho-tyrosine and phospho-serine specific mAbs can identify proteins phosphorylated on tyrosine or serine when PMNs are exposed to the agonist of interest. Used together, these techniques can help elucidate members of the signal transduction pathways, but to date have unveiled a complex, interacting matrix. Sphingomyelin metabolism. Sphingomyelin, a sphingolipid present in the outer leaflet of the plasma membrane, has been implicated in signal transduction pathways over the past decade (reviewed in 2608}. Sphingomyelinases are a group of phospholipases, some which are specific for sphingomyelin, that catalyze the hydrolysis of sphingomyelin (Fig. 1.4). Ceramide activates an unidentified kinase, a phosphatase, and is also reported to activate PKC~ (118- 120). It induces apoptosis in leukemic cell lines, and is a second messenger in some aspects of TNF-a and dihydroxyvitamin D3 mediated signaling (118,121). Exogenous short chain ceramide can regulate oxidant release from PMNs, and a metabolite of ceramide, sphingosine, has been shown to inhibit PKC function (122-124). Another metabolite of ceramide, sphingosine 1-phosphate, has also Sphingomyelin Containing Plasma Membrane HO Phospho choline I o Membrane Sphingomyelin Receptor Choline + Choline phosphate + N HO OH HO o Ceramide Phosphate 16 second --.....~.m. essenger?? Figure 1.4. Signal transduction through sphingomyelin metabolism. Binding of TNF to its receptor activates a membrane associated sphingomyelinase C which hydrolizes membrane sphingomyelin to ceramide, a putative second messenger. Sphingomyelinase D generates cerami de-phosphate from sphingomyelin which can be further metabolized to ceramide and then to sphingosine. 17 been implicated in increasing cytoplasmic free Ca++ (110). Recently a SMD, which generates ceramide phosphate and choline, has been shown to activate endothelium with concomitant dysregulated PMN adhesion and activation, in vivo this toxin causes a severe necrotic inflammatory lesion (125,126). Furthermore production of intracellular ceramide in PMNs has been shown to occur when they are stimulated with the agonist fMLP for long periods of time, and this coincides with a decrease in hydrogen peroxide production (123). These data suggest that ceramide generation in PMNs may lead to a less activated state that negatively affects, at a minimum, the respiratory burst. Conclusions It is evident that agonist-induced PMN responses, adhesion modulation, degranulation, and the respiratory burst, are coordinately regulated. Degranulation from specific granules is involved in three aspects of PMN activation: 1) it upregulates CD18 surface expression necessary for transmigration and an increase in phagocytic function 2) delivers of the enzyme responsible for L-selectin shedding which down regulates PMN rollin.g on endothelium and may enable more efficient transmigration, and finally 3) delivers cytochrome b558 and regulatory GTP proteins, RAP and Rac, to the plasma membrane where an active electron transport chain is assembled. This dissertation examines novel signaling mechanisms through endogenous ceramide production, and ICAM-3 receptor engagement, that differentially affect individual PMN responses. 18 References 1. Murphy, P. 1976. The neutrophil. Plenum Medical Book Company, New York. 1 pp. 2. Malech, H. L. and J. I. Gallin. 1987. Neutrophils in human diseases. Nw Eng J Med 317:687-694. 3. Dinauer, M. C., E. A. Pierce, G. A. Bruns, J. T. Curnutte, and S. H. Orkin. 1990. Human neutrophil cytochrome b light chain (p22-phox). J Clin Invest 86:1729- 1737. 4. Curnutte, J. T., P. J. Scott, and L. A. Mayo. 1989. Cytosolic components of the respiratory burst oxidase. Proc NatI Acad Sci USA 86:825-829. 5. Okamura, N., J. T. Curnutte, R. L. Roberts, and B. M. Babior. 1988. Relationship of protein phosphorylation to the activation of the respiratory burst in human neutrophils. J BioI Chern 263:6777-6782. 6. Curnutte, J. T., R. L. Berkow, R. L. Roberts, S. B. Shurin, and P. J. Scott. 1988. Chronic granulomatous disease due to a defect in the cytosolic factor required for nicotinamide adenine dinucleotide phosphate oxidase activation. J Clin Invest 81:606-610. 7. Malech, H. L. 1988. Phagocytic cells: egress from marrow and diapedesis. In Inflammation: Basic Principles and Clinical Correlates. J.I. Gallin, I.M. Goldstein, and R. Snyderman, editors. Raven Press, Ltd., New York. 297-308. 8. Molad, Y., K. A. Haines, D. C. Anderson, J. P. Buyon, and B. N. Cronstein. 1994. Immunocomplexes stimulate different signalling events to chemoattractants in the neutrophil and regulate L-selectin and ~2-integrin expression differently. Biochem J 299:881-887. 9. Metchnikoff, E. 1893. Lectures on the comparative pathology of inflammation. Trubner and Co. 10. Bevilacqua, M. P. and R. M. Nelson. 1993. Selectins. J Clin Invest 91:379-387. 11. Kishimoto, T. K., M. A. Jutila, E. L. Berg, and E. C. Butcher. 1989. Neutrophil Mac-l and MEL-14 adhesion proteins inversely regulated by chemotactic factors. Science 245:1238-1241. 12. Bonfanti, R., B. C. Furie, B. Furie, and D. D. Wagner. 1989. PADGEM (GMP 140) is a component of Weibel-Palade bodies of human endothelial cells. Blood 73:1109-1112. 19 13. McEver, R. P., J. H. Beckstead, K. L. Moore, L. Marshall-Carlson, and D. F. Bainton. 1989. GMP-140, a platelet a-granule membrane protein, is also synthesized by vascular endothelial cells and is localized in Weibel-Palade bodies. J Clin Invest 84:92-99. 14. Stenberg, P. E., R. P. McEver, M. A. Shuman, Y. V. Jacques, and D. F. Bainton. 1985. A platelet alpha-granule membrane protein (GMP-140) is expressed on the plasma membrane after activation. J Cell Bioi 101:880-886. 15. Lorant, D. E., M. K. Topham, R. E. Whatley, R. P. McEver, T. M. McIntyre, S. M. Prescott, and G. A. Zimmerman. 1993. Inflammatory roles of P-selectin. J Clin Invest 92:559-570. 16. Bevilacqua, M. P., J. S. Pober, D. L. Mendrick, R. S. Cotran, and M. A. Gimbrone,Jr .. 1987. Identification of an inducible endothelial-leukocyte adhesion molecule. Proc Natl Acad Sci USA 84:9238-9242. 17. Toothill, V. J., J. A. van Mourik, H. K. Niewenhuis, M. J. Metzelaar, and J. D. Pearson. 1990. Characterization of the enhanced adhesion of neutrophil leukocytes to thrombin-stimulated endothelial cells. J ImmunoI145:283-291. 18. Geng,J-G, M. P. Bevilacqua, K. L. Moore, T. M. McIntyre, S. M. Prescott, J. M. Kim, G. A. Bliss, G. A. Zimmerman, and R. P. McEver. 1990. Rapid neutrophil adhesion to activated endothelium mediated by GMP-140. Nature 343:757-760. 19. Bevilacqua, M. P., S. Stengelin, M. A. Gimbrone,Jr., and B. Seed. 1989. Endothelial leukocyte adhesion molecule 1: An inducible receptor for neutrophils related to complement regulatory proteins and lectins. Science 243:1160-1165. 20. Patel, K. D., K. L. Moore, M. U. Nollert, and R. P. McEver. 1995. Neutrophils use both shared and distinct mechanisms to adhere to selectins under static and flow conditions. J Clin Invest 96:1887-1896. 21. Picker, L. J., R. A. Warnock, A. R. Burns, C. M. Doerschuk, E. L. Berg, and E. C. Butcher. 1991. The neutrophil selectin LECAM-1 presents carbohydrate ligands to the vascular selectins ELAM-1 and GMP-140. Cell 66:921-933. 22. Lawrence, M. B., D. F. Bainton, and T. A. Springer. 1994. Neutrophil tethering to and rolling on E-selectin are separable by requirement for L-selectin. Immunity 1:137-145. 23. Buttrum, S. M., R. Hatton, and G. B. Nash. 1993. Selectin-mediated rolling of neutrophils on immobilized platelets. Blood 82:1165-1174. 24. Moore, K. L., K. D. Patel, R. E. Bruehl, L. Fugang, D. A. Johnson, H. S. Lichenstein, R. D. Cummings, D. F. Bainton, and R. P. McEver. 1995. P-selectin 20 glycoprotein ligand-l mediates rolling of human neutrophils on P-selectin. J Cell Bioi 128:661-671. 25. Hynes, R. 0.1992. Integrins: Versatility, modulation, and signaling in cell adhesion. Cell 69:11-25. 26. Clark, E. A. and J. S. Brugge. 1995. Integrins and signal transduction pathways: The road taken. Science 268:233-239. 27. Van der Vieren, M., H. L. Trong, C. L. Wood, P. F. Moore, T. St.John, D. E. Staunton, and W. M. Gallatin. 1995. A novelleukointegrin, adb2, binds preferentially to ICAM-3. Immunity 3:683-690. 28. Bohnsack, J. F., S. K. Akiyama, C. H. Damsky, W. A. Knape, and G. A. Zimmerman. 1990. Human neutrophil adherence to laminin in vitro. Evidence for a distinct neutrophil integrin receptor for laminin. J Exp Med 171:1221-1237. 29. Bohnsack, J. F. and X. -N. Zhou. 1992. Divalent cation substitution reveals CDI8- and very late antigen-dependent pathways that mediate human neutrophil adherence to fibronectin. J ImmunoI149:1340-1347. 30. Kubes, P., X. F. Niu, C. W. Smith, M. E. Kehrli,Jr., P. H. Reinhardt, and R. C. Woodman. 1995. A novel bl-dependent adhesion pathway on neutrophils. FASEB J 9:1103-1111. 31. Springer, T. A. 1990. Adhesion receptors of the immune system. Nature 346:425-434. 32. Sastry, S. K. and A. F. Horwitz. 1993. Integrin cytoplasmic domains: Mediators of cytoskeletallinkages and extra- and intracellular initiated transmembrane signaling. Curr Opin Cell Bioi 5:819-831. 33. Reszka, A. A., Y. Hayashi, and A. F. Horwitz. 1992. Identification of amino acid sequences in the integrin ~ 1 cytoplasmic domain implicated in cytoskeletal association. J Cell Bioi 117:1321-1330. 34. Altieri, D. C. and T. S. Edgington. 1988. A monoclonal antibody reacting with distinct adhesion molecules defines a transition in the functional state of the receptor CDllb/CDI8 (Mac-I). J ImmunoI141:2656-2660. 35. Pavalko, F. M. and S. M. LaRoche. 1993. Activation of human neutrophils induces an interaction between the integrin (32-subunit (CDI8) and the actin binding protein a-actinin. J ImmunoI151:3795-3807. 21 36. Jesaitis, A. J., R. W. Erickson, K. -N. Klotz, R. K. Bommakanti, and D. W. Siemsen. 1993. Functional molecular complexes of human N-formyl chemoattractant receptors and actin. J ImmunoI151:5653-5665. 37. Petrequin, P. R., R. F. Todd III, L. J. Devall, L. A. Boxer, and J. T. Curnette III. 1987. Association between gelatinase release and increased plasma membrane expression of the Mol glycoprotein. Blood 69:605-610. 38. Jones, D. H., D. C. Anderson, B. L. Burr, H. E. Rudloff, C. W. Smith, S. S. Krater, and F. C. Schmalsteig. 1988. Quantitation of intercellular Mac-1 (CD11b/CD18) pools in human neutrophils. J Leukoc Bioi 44:535-544. 39. Vedder, N. B. and J. M. Harlan. 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. 40. Philips, M. R., J. P. Buyon, R. Winchester, G. Weissmann, and S. B. Abramson. 1988. Up-regulation of the iC3b receptor (CR3) is neither necessary nor sufficient to promote neutrophil aggregation. J Clin Invest 82:495-501. 41. Hughes, B. J., J. C. Hollers, E. Crockett-Torabi, and C. W. Smith. 1992. Recruitment of CD11b/CD18 to the neutrophil surface and adherence-dependent cell locomotion. J Clin Invest 90:1687-1696. 42. de Fougerolles, A. R., S. A. Stacker, R. Schwarting, and T. A. Springer. 1991. Characterization of ICAM-2 and evidence for a third counter-receptor for LFA-1. J Exp Med 174:253-267. 43. Vazeux, R., P. A. Hoffman, J. K. Tomita, E. S. Dickinson, R. L. Jasman, T. St.John, and W. M. Gallatin. 1992. Cloning and characterization of a new intercellular adhesion molecule ICAM-R. Nature 360:485-488. 44. Staunton, D. E., S. D. Marlin, C. Stratowa, M. L. Dustin, and T. A. Springer. 1988. Primary structure of ICAM-1 demonstrates interaction between members of the immunoglobulin and integrin supergene families. Cell 52:925-933. 45. de Fougerolles, A. R. and T. A. Springer. 1992. Intercellular adhesion molecule 3, a third adhesion counter-receptor for lymphocyte function-associated molecule 1 on resting lymphocytes. J Exp Med 175:185-190. 46. Staunton, D. E., M. L. Dustin, and T. A. Springer. 1989. Functional cloning of ICAM-2, a cell adhesion ligand for LFA-1 homologous to ICAM-1. Nature Lond. 339:61. 22 47. Diamond, M. S., D. E. Staunton, A. R. de Fougerolles, S. A. Stacker, J. Garcia­Aguilar, M. L. Hibbs, and T. A. Springer. 1990. ICAM-1 (CD54): A counter­receptor for Mac-1 (CD11b/CD18). J Cell BioI 111:3129-3139. 48. Xie, J., L. Rui, P. KotoVllori, C. Vermot-Desroches, J. Wijdenes, M. A. Arnaout, P. Nortamo, and C. G. Gahmberg. 1995. Intercellular adhesion molecule-2 (CD102) binds to the leukocyte integrin CD11b/CD18 through the A domain. J ImmunoI155:3619-3628. 49. de Fougerolles, A. R., M. S. Diamond, and T. A. Springer. 1995. Heterogenous glycosylation of ICAM-3 and lack of interaction with Mac-1 and p150,95. Eur J ImmunoI25:1008-1012. 50. Smith, C. W., S. D. Marlin, R. R. Rothlein, C. J. Toman, and D. C. Anderson. 1988. Mac-1, LFA-1, and ICAM-1 in neutrophil adherence to endothelial cells in vitro. Effects of chemotactic stimulation. Blood 72:I-1531.(Abstr.) 51. Collins, T., M. A. Read, A. S. Neish, M. Z. Whitley, D. Thanos, and T. Maniatis. 1995. Transcriptional regulation of endothelial cell adhesion molecules: NF-kB and cytokine-inducible enhancers. FASEB J 9:899-909. 52. Diamond, M. S., D. E. Staunton, S. D. Marlin, and T. A. Springer. 1991. Binding of the integrin Mac-1 (CD11b/CD18) to the third immunoglobulin-like domain of ICAM-1 (CD54) and its regulation by glycosylation. Cell 65:961-971. 53. Reilly, P. L., J. R. Woska,Jr., D. D. Jeanfavre, E. McNally, R. Rothlein, and B. -J. Barmann. 1995. The native structure of intercellular adhesion molecule-1 (ICAM-1) is a dimer. J ImmunoI155:529-532. 54. Delisser, H. M., W. A. Muller, P. J. Newman, and S. M. Albelda. 1991. PECAM-1 (CD31) mediates heterophilic cell-cell adhesion. J Cell BioI 115:70a.(Abstr.) 55. Delisser, H. M., J. Chilkotowsky,H. C. Yan, M. L. Daise, C. A. Buck, and S. M. Albelda. 1994. Deletions in the cytoplasmic domain of platelet-endothelial cell adhesion molecule-1 (PECAM-1, CD31) result in changes in ligand binding properties. J Cell BioI 124:195-203. 56. Delisser, H. M., H. C. Yan, P. J. Newman, W. A. Muller, C. A. Buck, and S. M. Albelda. 1993. Platelet/endothelial cell adhesion molecule-1 (CD31)-mediated cellular aggregation involves cell surface glycosammoglycans. J BioI Chern 268:16037-16046. 57. Fawcett, J., C. Buckley, C. L. Holness, J. H. Spragg, J. Saunders, A. Harris, and D. L. Simmons. 1995. CD31 adhesion in the formation of interendothelial cell contacts. J Cell BioI 128:1229-1241. 23 58. Muller, W. A., S. A. Weigl, X. Deng, and D. M. Phillips. 1993. PECAM-1 is required for transendothelial migration of leukocytes. J Exp Med 178:449-460. 59. Berman, M. E. and W. A. Muller. 1995. Ligation of platelet/ endothelial cell adhesion molecule 1 (PECAM-1/CD31) on monocytes and neutrophils increases binding capcity of leukocyte CR3 (CD11b/CD18). J ImmunoI154:299-307. 60. Janmey, P. A. 1994. Phosphoinositides and calcium as regulators of cellular actin assembly and disassembly. Annu Rev PhysioI56:169-191. 61. Omann, G. M., R. Rengan, J. F. Hoffman, and J. J. Linderman. 1995. Rapid oscillations of actin polymerization/ depolymerization in polymorphonuclear leukocytes stimulated by leukotriene B4 and platelet-activating factor. J Immunol 155:5375-5381. 62. Packman, C. H. and M. A. Lichtman. 1990. Activation of neutrophils: measurement of actin conformational changes by flow cytometry. Blood Cells 16:193-207. 63. Korchak, H. M., L. E. Rutherford, and G. Weissmann. 1994. Stimulus response coupling in the human neutrophil. J BioI Chern 259:4070-4075. 64. Luna, E. J. and A. L. Hitt. 1992. Cytoskeleton-plasma membrane interactions. Science 258:955-964. 65. Schaller, M. D., C. A. Otey, J. D. Hildebrand, and J. T. Parsons. 1995. Focal adhesion kinase and paxillin bind to peptides mimicking beta integrin cytoplasmic domains. J Cell BioI 130:1181-1187. 66. Bussolino, F., F. Gremo, C. Tetta, P. Pescarmosa, and G. Camussi. 1986. Production of platelet-activating factor by chick retina. J BioI Chem 261:16502- 16508. 67. Pavalko, F. M., C. A. Otey, K. O. Simon, K. Burridge, and S. Burridge. 1991. Alpha-actinin: a direct link between actin and integrins. Biochern Soc Trans 19:1065-1069. 68. Carpen, 0., P. Pallai, D. E. Staunton, and T. A. Springer. 1992. Association of intercellular adhesion molecule-1 (ICAM-1) with actin-containing cytoskeleton and alpha-actinin. J Cell BioI 118:1223-1234. 69. Miyamoto, S., H. Teramoto, O. A. Coso, J. S. Gutkind, P. D. Burbelo, S. K. Akiyama, and K. M. Yamada. 1995. Integrin function: molecular hierarchies of cytoskeletal and signaling molecules. J Cell BioI 131:791-805. 24 70. Cooper, J. A. 1987. Effects of cytochalasin and phalloidin on actin. J Cell BioI 105:1473-1478. 71. Lorant, D. E., R. P. McEver, T. M. McIntyre, K. L. Moore, S. M. Prescott, and G. A. Zimmerman. 1995. Activation of polymorphonuclear leukocytes reduces their adhesion to P-selectin and causes redistribution of ligands for P-selectin on their surfaces. J Clin Invest 95:1452-1461. 72. Peter, K. and T. E. O'Toole. 1995. Modulation of cell adhesion by changes in alpha L beta 2 (LFA-1, CD11a/CD18) cytoplasmic domain/cytoskeleton interaction. J Exp Med 181 :315-326. 73. Hawkins, D. 1973. Neutrophilic leukocytes in immunologic reactions in vitro effect of cytochalasin B. J ImmunoI110:294-296. 74. Henson, P. M. and Z. G. Oades. 1973. Enhancement of immunologically induced granule exocytosis from neutrophils by cytochalasin B. J Immunol 110:290-293. 75. Keller, H. and V. Niggli. 1995. Effects of cytochalasin D on shape and fluid pinocytosis in human neutrophils as related to cytoskeletal changes (actin, alpha­actinin and mierotubules). Eur J Cell BioI 66:157-164. 76. Sengelov, H., L. Kjeldsen, M. S. Diamond, T. A. Springer, and N. Borregaard. 1993. Subcellular localization and dynamics of Mac-1 (am~2) in human neutrophils. J Clin Invest 92:1467-1476. 77. Kjeldsen, L., O. W. Bjerrum, J. Askaa, and N. Borregaard. 1992. Subcellular localization and release of human neutrophil gelatinase confirming the existence of separate gelatinase containing granules. Biochem J 287:603. 78. Hibbs, M. S. and D. F. Bainton. 1989. Human neutrophil gelatinase is a component of specific granules. J Clin Invest 84:1395-1402. 79. Henson, P. M., J. E. Henson, C. Fittschen, D. L. Bratton, and D. W. H. Riches. 1992. Degranulation and secretion by phagocytic cells. In Inflammation: Basic Principles and Clinical Correlates. J.I. Gallin and I.M. Goldstein, editors. Raven Press,New York. 511-539. 80. Diaz-Gonzalez, F., I. Gonzalez-Alvaro, M. R. Campanero, F. Mollinedo, M. A. del Pozo, C. Munoz, J. P. Pivel, and F. Sanchez-Madrid. 1995. Prevention of in vitro neutrophil-endothelial attachment through shedding of L-selectin by nonsteroidal antiinflammatory drugs. J Clin Invest 95:1756-1765. 25 81. Kishimoto, T. K., M. A. Jutila, and E. C. Butcher. 1990. Identification of a human peripheral lymph node homing receptor. Proc Natl Acad Sci USA 87:2244- 2248. 82. Babior, B. M. 1987. The respiratory burst oxidase. Trends Biochem Sci 12:241- 243. 83. Cross, A. R. and O. T. G. Jones. 1991. Enzymic mechanisms of superoxide production. Biochim Biophys Acta 1057:281-298. 84. Klebanoff, S. J. 1974. Role of the superoxide anion in the myeloperoxidase­mediated antimicrobial system. J BioI Chem 249:3724. 85. Cline, M. J. and R. I. Lehrer. 1969. D-amino acid oxidase in leukocytes. Proc Natl Acad Sci USA 62:756. 86. Cross, A. R., O. T. G. Jones, A. M. Harper, and A. W. Segal. 1981. Oxidation­reduction properties of the cytochrome-b found in the plasma-membrane fraction of human neutrophils. Biochem J 194:599-607. 87. Segal, A. W., A. R. Cross, R. C. Garcia, N. Borregaard, N. H. Valerius, J. F. Soothill, and O. T. G. Jones. 1983. Absence of cytochrome b-245 in chronic granulomatous disease. New Eng J Med 308:245-251. 88. Segal, A. W. and O. T. G. Jones. 1978. Novel cytochrome b system in phagocytic vacuoles of human granulocytes. Nature Lond 276:515-517. 89. Nakamura, M., S. Sendo, R. Van Zweiten, T. Koga, D. Roos, and S. Kanegasaki. 1988. Immunochemical discovery of the 22- to 23-kD subunit of cytochrome b558 at the surface of human peripheral phagocytes. Blood 72:1550- 1552. 90. Borregaard, N., J. M. Heiple, E. R. Simons, and R. A. Clark. 1983. Subcellular localization of the b-cytochrome component of the human neutropihl microbicidal oxidase. J Cell Bio197:52-61. 91. Woodman, R. C., J. M. Ruedi, A. J. Jesaitis, N. Okamura, M. T. Quinn, R. M. Smith, J. T. Curnutte, and B. M. Babior. 1991. Respiratory burst oxidase and three of four oxidase-relatedpolypeptides are associated with the cytoskeleton of human neutrophils. J Clin Invest 87:1345-1351. 92. Clark, R. A., B. D. Volpp, K. G. Leidal, and W. M. Nauseef. 1990. Two cytosolic components of the human neutrophil respiratory burst oxidase translocate to the plasma membrane during cell activation. J Clin Invest 85:714- 721. 26 93. Thelen, M., B. Dewald, and M. Baggiolini. 1993. Neutrophil signal transduction and activation of the respiratory burst. Physiol Rev 73:797-821. 94. Whitin, J. C., C. E. Chapman, E. R. Simons, M. E. Chovaniec, and H. J. Cohen. 1980. Correlation between membrane potential changes and superoxide production in human granulocytes stimulated by phorbol myristate acetate. ] BioI Chem 255:1874-1878. 95. Kreck, M. L.:! D. J. Uhlinger, S. R. Tyagi, K. L. Inge, and J. D. Lambeth. 1994. Participation of the small molecule weight GTP-binding protein Rac1 in cell-free activation and assembly of the respiratory burst oxidase. ] BioI Chern 269:4161- 4168. 96. Diekmann, D., A. Abo/C. Johnston, A. W. Segal, and A. Hall. 1994. Interaction of Rac with p67phox and regulation of phagocytic NADPH oxidase activity. Science 265:531-533. 97. Prigmore, E., S. Ahmed, A. Best, R. Kozma, E. Manser, A. W. Segal, and L. Lim. 1995. A 68-kDa kinase and NADPH oxidase component p67phox are targets for Cdc42Hs and Racl in neutrophils. ] BioI Chern 270: 1 0717-10722. 98. Mollinedo, F., D. Perez-Sal a, C. Gajate, B. Jimenez, P. Rodriguez, and J. C. Lacal. 1993. Localization of rap1 and rap2 proteins in the gelatinase-containing granules of human neutrophils. FEBS Lett 326:209-214. 99. Abo, A., M. R. Webb, A. Grogan, and A. W. Segal. 1994. Activation of NADPH oxidase involves the dissociation of p21 rac from its inhibitory GDP /GTP exchange protein (rhoGDI) followed by its translocation to the plasma membrane. Biochem ] 298:585-591. 100. Lew, P. D., A. Monod, F. A. Waldvogel, B. Dewald, M. Baggiolini, and T. Pozzan. 1986. Quantitative analysis of the cytosolic free calcium dependency of exocytosis from three subcellular compartments in intact human neutrophils. ] Cell BioI 102:2197-2204. 101. Lew, P. D., C. B. Wollheim, F. A. Waldvogel, and T. Pozzan. 1984. Modulation of cytosolic-free calcium transients by changes in intracellular calcium-buffering capacity. ] Cell BioI 99:1212-1220. 102. Lew, P. D. 1990. Receptors and intracellular signaling in human neutrophils. Am Rev Respir Dis 141:127-131. 103. Lagast, H., T. Pozzan, F. A. Waldvogel, and P. D. Lew. 1984. Phorbol myristate acetate stimulates the ATP-dependent calcium transport by the plasma membrane of neutrophils. ] Clin Invest 73:878-883. 27 104. Krause, K. H. and P. D. Lew. 1987. Subcellular distribution of Ca2+ pumping sites in human neutrophils. J Clin Invest 80:107-116. 105. Jaconi, M. E. E., J. M. Theler, W. Schlegel, and P. D. Lew. 1993. Cytosolic free Ca2+ signals in single adherent human neutrophils. Eur , Pediatr 152:26-32. 106. Andersson, T., C. Dahlgren, T. Pozzan, O. Stendahl, and P. D. Lew. 1986. Characterization of f-met-Ieu-phe receptor mediated Ca2+ influx across the plasma membrane of human neutrophils. Mol Pharmacol 30:437-43. 107. Von Tscharner, V., B. Prod'hom, M. Baggiolini, and H. Reuter. 1986. Ion channels in human neutrophils activated by a rise in free cytosolic calcium concentration. Nature 324:369-372. 108. Clapper, D. L., T. F. Walseth, P. J. Dargie, and H. C. Lee. 1987. Pyridine nucleotide metabolites stimulate calcium release from sea urchin egg micro somes desensitized to inositol trisphosphate., BioI Chern 262:9561-9568. 109. Choi, O. H., J. H. Kim, and J. P. Kinet. 1996. Calcium mobilization via sphingosine kinase in signaling by the Fc&RI antigen receptor. Nature 380:634- 636. 110. Mattie, M., G. Brooker, and S. Spiegel. 1994. Sphingosine-1-phosphate, a putative second messenger, mobilizes calcium from internal stores via an inositol triphosphate-independent pathway. J BioI Chern 269:3181-3188. 111. Ghosh, T. K., J. Bian, and D. L. Gill. 1990. Intracellular calcium release mediated by sphingosine derivatives generated in cells. Science 248:1653-1656. 112. Jaconi, M. E. E., J. M. Theler, W. Schlegel, R. D. Appel, S. D. Wright, and P. D. Lew. 1991. Multiple elevations of cytosolic-free Ca2+ in human neutrophils: Initiation by adherence receptors of the integrin family., Cell BioI 112:1249-1257. 113. Majumdar,S., M. W. Rossi, T. Fujiki, W. A. Phillips, S. Disa, C. F. Queen, R. B. Johnston,Jr., O. M. Rosen, B. E. Corkey, and H. M. Korchak. 1991. Protein kinase C isotypes and signaling in neutrophils., BioI Chern 266:9285-9294. 114. Nishizuka, Y. 1986. Studies and perspectives of protein kinase C. Science 233:305-312. 115. Merrill, A. H.,Jr. and V. L. Stevens. 1989. Modulation of protein kinase C and diverse cell functions by sphingosine - a pharmalogically interesting compound linking sphingolipids and signal transduction. Biochim Biophys Acta 1010:131-139. 28 116. Downey, G. P., C. K. Chan, P. Lea, A. Takai, and S. Grinstein. 1992. Phorbol ester-induced actin assembly in neutrophils. J Cell Bioi 116:695-706. 117. Grinstein, S., J. R. Butler, W. Furuya, G. L'Allemain, and G. P. Downey. 1994. Chemotactic peptides induce phosphorylation and activation of MEK-1 in human neutrophils. J Bioi Chern 269:19313-19320. 118. Hannun, Y. A. 1994. The sphingomyelin cycle and the second messenger function of ceramide. J Bioi Chern 269:3125-3128. 119. Dobrowsky, R. T., C. Kamibayashi, M. C. Mumby, and Y. A. Hannun. 1993. Ceramide activates heterotrimeric protein phosphatase 2A. J Bioi Chem 268:15523- 15530. 120. Raines, M. A., R. N. Kolesnick, and D. W. Golde. 1993. Sphingomyelinase and cerami de activate mitogen-activated protein kinase in myeloid HL-60 cells. J Bioi Chern 268:14572-14575. 121. Kolesnick, R. and D. W. Golde. 1994. The sphingomyelin pathway in tumor necrosis factor and interleukin-1 signaling. Cell 77:325-328. 122. Jones, M. J. and A. W. Murry. 1995. Evidence that ceramide selectively inhibits protein kinase C-a translocation and modulates bradykinin activation of phospholipase D. J Bioi Chern 270:5007-5013. 123. Nakamura, T., A. Abe, K. J. Balazovich, D. Wu, S. J. Suchard, L. A. Boxer, and J. A. Shayman. 1994. Ceramide regulates oxidant release in adherent human neutrophils. J Bioi Chern 269:18384-18389. 124. Hannun, Y. A., C. R. Loomis, A. H. Merrill,Jr., and R. M. Bell. 1986. Sphingosine inhibition of protein kinase C activity and of phorbol dibutyrate binding in vitro and in human platelets. J Bioi Chern 261:12604-12609. 125. Kurpiewski, G., L. J. Forrester, J. T. Barrett, and B. J. Campbell. 1981. Platelet aggregation and sphingomyelinase D activity of a purified toxin from the venom of Loxosceles reclusa. Biochim Biophys Acta 678:467-476. 126. Patel, K. D., V. Modur, G. A. Zimmerman, S. M. Prescott, and T. M. McIntyre. 1994. The necrotic venom of the brown recluse spider induces dysregulated endothelial cell-dependent neutrophil activation. Differential induction of GM-CSF, IL-8 and E-selectin expression. J Clin Invest 94:631-642. CHAPTER 2 MATERIALS AND METHODS Materials Sphingomyelinase D from Loxosceles deserta venom was obtained from Spider Pharm (Feasterville, P A). Venom was stored in small aliquots at -800 , thawed, diluted 100 fold into HBSS containing 0.5% human serum albumin from Miles Laboratories (Elkhart, IN). HBSS and M199 were obtained from Whittaker Bioproducts (Walkersville, MA). Staphylococcus au reus sphingomyelinase C (SMC), staurosporin, herbimycin A, gene stein, N,N-Dimethylspingosine, lysphosphatidylcholine, FITC-phalloidin, were obtained from Sigma Chemical Company (St. Louis, MO). N-octanylsphingosine (CS-ceramide) was from Avanti Polar Lipids (Alabaster, AL). PKC inhibitors H7 and 1004 were obtained from Seikagaku Corporation (Japan). mAb 60.3 and mAb 60.1 were gifts from Dr. Patrick Beatty, (University of Utah). Chelerythrine was obtained from Calbiochem (La Jolla CA). Enhanced chemiluminescence (ECL) western blotting reagents were obtained from Amersham (England). The anti-ICAM-3 mAbs CAL3.10, CAL3.34 and oncostatin M were obtained from R&D Systems (Minneapolis, MN). All secondary antibodies were obtained from Biosource International (Camarillo, CA). FITC conjugated anti-CD18, rabbit polyclonal (l­phospho tyrosine, and anti-L-selectin were obtained from Immunotech (Marseille, France). Phycoerythrin conjugated anti-CDllb was obtained from Becton Dickinson (Bedford, MA). Indo-l AM was obtained from Molecular Probes 30 (Eugene, OR). Matrigel was obtained from Collaborative Biomedical Products (Bedford, MA). Recombinant TNF-alpha was provided by Genetech (San Francisco, CA). Female Swiss 44 mice were obtained from Charles River (Hollister, CA). Costar Transwells were from Corning Costar Corp. (Cambridge, MA). Methods In vivo studies. Female Swiss 44, 6-8 week old mice were utilized to demonstrate in vivo inflammatory responses to SMC, SMD, and OSM. The hair on the backs of the mice was removed and a 32-gauge needle was used to deliver 10ul quantities of agonist intradermally. The mice were sacrificed by cervical dislocation at desired time points, the site of agonist delivery excised and fixed in 10% formalin overnight before embedding in paraffin, sectioning and staining with hematoxylin and eosin. Cells and adhesion assays. Human neutrophils were isolated from blood as described (1). 4 well plates from Nunclone (Roskilde, Denmark) were coated with 0.2% gelatin, 100ug/ml fibronectin, or 100ug/mllaminin, in nanopure water at 370 for 3 hours and then blocked with HBSSA for 1 hour at 370 . Plates were washed twice with HBSSA before addition of PMNs, which were diluted to S x 105/ml in HBSSA. Agonist at concentrations as reported in figure legends were then added in 2Sul aliquots. The plates were incubated at 370 for S minutes and nonadherent PMNs were removed by aspiration, followed by two SOOul HBSSA washes. The cells were immediately quantified using a custom video imaging system connected to a microscope. Three random fields/well were read. PMNs at S x 105/ml were pretreated with O.S V/ml SMC for 30 minutes at 370 before addition to adhesion assays. Ion substitution assays were done as 31 described with two minor alterations (2). Unlabeled PMNs were used; and video imaging analysis was used to quantify adherent cells. Western blotting of PMN proteins. PMNs, after specified treatments, were pelleted by centrifugation before boiling in Laemmli sample buffer with ~­mercaptoethanol. Proteins were electrophoresed in a 10% SDS polyacrylamide gel and transferred to PVDF membrane (Immobilon P, Millipore Corporation, Bedford, MA) (3). The membrane was blocked overnight with Tris buffered saline with 0.05% Tween-20 (TBST) containing 10% human serum albumin. Primary and secondary Abs were in appropriate sera at 10%. Staining was detected by ECL (Amersham) with subsequent exposure to Kodak X-OMAT film. PMN transmigration assays. One hundred microliters of sterile 0.2% gelatin from porcine skin were placed on the transwell membrane of Costar 24 well Transwell Plates for 1 hour at 370 . The gelatin solution was then removed and SOOul of complete endothelial cell medium was placed on the well bottom. The upper chamber containing the gelatin pretreated membrane was inserted and 100ul of freshly isolated HUVEC cell suspension were then placed on the membrane in complete endothelial cell medium (M199 + 20% pooled human sera). Twelve to twenty-four hours later, complete endothelial cell medium was aspirated off the bottom of the well and SOOul of fresh endothelial cell medium was added. Media were changed every 3-4 days and HUVEC reach confluence over approximately 8 days. Upon confluency the media were removed and replaced with HBSSA containing an agonist or control buffer, and incubated at 370 for 4 hours. Two hundred fifty microliters of Inlll labeled PMNs at 2 x 106/ml in HBSSA were then overlaid on the HUVEC. For some wells a stimulus was added before plates were incubated at 370 for 30 minutes. To terminate the assay plates were chilled in an ice bath, then the transwells removed. The fluid in the upper chamber was removed and combined with the excised membrane 32 after the membrane bottom had been swabbed to remove any transmigrated PMNs adherent to the underside of the membrane. The PMNs in the bottom well were collected and combined with the swab of the membrane bottom. The samples were counted on a gamma counter for 111 In .. PMN aggregation and Ca++ flux measurements. PMNs at 5 xl06/ml were loaded in HBSS with luM Indo-l AM (Molecular Probes) for 30 minutes at 370 shielded from light. Cells were washed with HBSSA before assaying in a stirred cuvette at 370 in a Shimadzu 5000RF spectrofluoremeter with the specified agonist. PMNs were treated with SMC or buffer at 370 for 30 minutes prior to addition to the aggregation assay. Pretreatment of PMNs with PKC and tyrosine kinase inhibitors was as follows. PMNs were pretreated for 10 minute at 370 with H7 [10uM], chelerythrine [10uM], 1004 [10uM], herbimycin A at [10uM] and genestein at [80uM]. Pretreatment of PMNs with mAbs was as stated in figure legends. A Payton and Associates Aggrecorder was used to measure real time aggregation by loss of light dispersion. Flow cytometry, imaging, ELISA, and superoxide quantification. Surface expression of adhesion molecules and f-actin content was quantified by FACscan Analysis. 1 x 106 PMNs were treated as described in figure legends, and pelleted and resuspended in 40 PBS 0.1 % Na Azide containing 10% goat serum for 10 minutes. Cells were centrifuged and resuspended for 1 hour at 40 in 10ug/ml of mAb 60.3, 60.1, CAL3.1, CAL3.34. Cells were washed thrice before resuspension in FITC-conjugated goat anti-mouse polyclonal antibody in same buffer. Cells were fixed in 0.5% formaldehyde overnight at 40 before analysis by flow cytometry. An aliquot of cells was removed and centrifuged onto microscope slides for fluorescence microscopic imaging using a cooled Charged Couple Device (CCD) camera and BDS Image analysis software. An aliquot of PMN that were being analyzed in the aggregometer after anti-ICAM-3 treatment were 33 stained for the presence of L-selectin, CDllb, or CD18 after fixation in 1 % ice cold paraformaldehyde in PBS for 30 minutes. PMNs were washed in PBS, and then resuspended in 100ul PBS 10% goat serum with the directly conjugated antibody of choice. The f-actin content was determined using FITC-phalloidin. 200 ul of 8% formaldehyde, 100ug/ml LPC, 2ug/ml FITC-phalloidin in PBS, was added to 200ul of lxl06 PMNs. These were incubated for 4 hours on ice, washed with PBS and resuspended in lml PBS before analysis by FACScan. Lactoferrin and elastase were detected using sandwich ELISA with rabbit anti-human lactoferrin and sheep anti-human elastase as described (4). Superoxide was detected by the reduction of ferricytochrome C at a wavelength of 490nm. Fifty microliters of PMN [5 x 106ml] were added to a well in a 96 well plate containing 20ul of agonist, 20 ul of cytochrome C [2.5mg/ml], and 120 ul of HBSSA. Plates were incubated, mixed, and read at 370 in a Molecular Devices Thermo max microplate reader for 90 minutes and the data analyzed using Softmax software. Sphingomyelin and ceramide Q.uantification. Lipids isolated from 1 x 107 PMNs were extracted after the stated treatment (5). Fifty percent of this lipid was subjected to diacylglycerol kinase assay as described by the manufacturer (Amersham) with several modifications: Only glass tubes were used. Purified C16-ceramide was included as a standard in addition to diacylglycerol and the reaction was extended to 3 hours. The completed reaction mixtures were separated by high performance thin layer chromatography and liquid scintillation counting of 32P-Iabeled ceramide and 32P-phosphatidic acid (diacylglycerol) allowed quantification by comparison to 32P-ceramide and 32p_ diacylglycerol standard curves. Phosphorimage analysis was also used to confirm the intensity of ceramide phosphate and phosphatidic acid bands. An aliqout (10% of sample) was analyzed for inorganic phophorous content of 34 individual phospholipid fractions by separation on an HPLC using 4.6mm x 300mm silica gel column and the following gradient: Buffer A = acetonitrile:water (97.5:2.5) Buffer B = acetonitrile:water (85:15). 100% A for three minutes followed by a gradient to 100% B over 12 minutes and held for 10 minutes (6). Peaks were detected at 203nm collected and dried under nitrogen before the phosphate content was quantified (7). References 1. Zimmerman, G. A., T. M. McIntyre, and S. M. Prescott. 1985. Thrombin stimulates the adherence of neutrophils to human endothelial cells in vitro. J Clin Invest 76:2235-2246. 2. Bohnsack, J. F. and X. -N. Zhou. 1992. Divalent cation substitution reveals CDI8- and very late antigen-dependent pathways that mediate human neutrophil adherence to fibronectin. J ImmunoI149:1340-1347. 3. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. 4. Topham, M., S. M. Prescott, G. A. Zimmerman, and T. M. MCIntyre. 1989. Neutrophil secretion while adherent to activated endothelium is differentially regulated. Clin Res 37:119A. 5. Bligh, E. G. and W. J. Dyer. 1959. A rapid method of total lipid extraction and purification. Can J Biochem PhysioI37:911-917. 6. Nissen, H. P. and H. W. Kreysel. 1983. Analysis of phospholipids in human semen by high-performance liquid chromatography. J Chromatogr 276:29-35. 7. Ames, B. N. and D. T. Dubin. 1960. The role of polyamines in the neutralization of bacteriophage deoxyribonucleic acid. J BioI Chern 235:769-775. CHAPTER 3 CERAMIDE SIGNALING IN NEUTROPHILS Introduction Leukocytes constitute the body's first line of defense against invading organisms by responding to physiologic and pathologic inflammatory stimuli. The initial step in this process is adhesion of quiescent leukocytes to the surface of activated endothelium; this occurs through endothelial cell­expressed adhesion molecules that account for leukocyte rolling along the vasculature (1,2). Activated endothelial cells simultaneously express leukocyte agonists that bind to receptors on the leukocyte plasma membrane and induce a rapid rise in intracellular calcium, inside-out signaling, resulting in functional upregulation of leukocyte adhesion proteins, and cellular polarization. These functional changes are required for the leukocytes to migrate to endothelial cell junctions (3). Transmigration of leukocytes through the endothelium further affects function in that it primes them to respond in a more sensitive fashion to a second stimulus. This carefully orchestrated series of events depends on a complex interaction of the leukocyte with its environment, as it is now apparent that the leukocyte's environment sends modulatory and enabling stimuli, generally classified as outside/inside signaling, to alter cellular responses. The best defined example of this is the effect of TNF on adherent compared to those in suspension. Generation of oxygen radicals and release of granule contents by this agonist 36 is reported to depend on adherence of the leukocytes via CD11 b / 18 integrins (4). CD18 receptors are normally uniformly dispersed in a non-functional state over the surface of quiescent cells (5,6). However, upon activation these adhesion molecules, as well as those that are newly recruited to the surface through degranulation, aggregate in clusters and undergo a conformational change to a state capable of interacting with many ligands (5,6). Clustering and activation not only serve to strengthen adhesive interactions but may directly affect both calcium signaling and other events leading to the respiratory burst (5,7-12). Appropriate ligation, clustering, and outside-in signaling coordinate the leukocyte response in a fashion appropriate to its extracellular environment. The nature of the signal transmitted by appropriately engaged integrins is not well defined, nor is the nature of the response leading to integrin clustering well defined. TNF induction of differentiation or programmed cell death in cells of the myeloid lineage derives from activation of sphingomyelinase(s) activity, leading to ceramide generation. Ceramide activates kinase cascades and a key cellular phosphatase required for these responses (13-15). A role for ceramide signaling in leukocytes, as opposed to the leukocytic cell lines, is currently undefined, although TNF results in a rapid rise in ceramide content in neutrophils (16). In contrast agonists such as fMLP, which do not require adhesion to an appropriate surface for activation, also result in ceramide accumulation, but only at prolonged times as the respiratory burst activity is progressively depressed (17). We injected sphingomyelinase C into the skin of mice and found massive inappropriate PMN accumulation. We therefore examined the role of ceramide in leukocyte function, and took advantage of our preliminary observation that sphingomyelinase D depletes the PMN plasma membrane of 37 sphingomyelin but does not directly activate PMNs. This allowed us to inhibit the neutral plasma membrane sphingomyelinase activity to determine if cerami de induces or alters PMN function, and to determine if endogenous sphingomyelinase C has an essential role in signaling. We found that ceramide is a powerful activator of PMN degranulation, resulting in near complete granular enzyme release in the absence of a priming event. Additionally, ceramide-induced degranulation caused ~2-integrin accumulation on the surface, but this was expressed in a nonactive form that could not support adhesion. This state appears to correlate with decreased f­actin content and a depressed ability of the ~2-integrin to form clusters. Despite these responses, depletion of endogenous substrate shows that ceramide is not an essential signaling molecule in the response of PMN to either PAF or fMLP. Results Sphingomyelinase C is an inflammatory mediator. Bacterial exotoxins released from Clostridium perfringens, the causative agent of gas gangrene, includes a phospholipase C that hydrolyzes membrane lipids and causes pores to form in the plasma membrane. This subsequently activates endothelial cells and causes massive leukocyte accumulation in vivo (18-21). The phospholipase C from C. perfringens (formerly C. welchii) shows little selectivity for complex lipids, and hydrolyzes both phosphatidylcholine and sphingomyelin to diglyceride and ceramide, respectively (22). As both these neutral lipids are signaling molecules in various receptor-coupled signal cascades, we sought to determine if a similar massive, inappropriate inflammatory response occurs when just ceramide is generated by an activity that specifically hydrolyzes sphingomyelin to ceramide. We injected 38 sphingomyelinase C from Staphylococcus aureus, which is specific for this reaction, into the skin of a mouse and then excised skin patches from these mice for analysis after 6 and 24 hours. Figure 3.1A shows a typical small vessel, present in a saline injected animal, demonstrating their usual morphology. There is continuous endothelium, numerous erythrocytes within the vessel lumen, with only the occasional staining of the less abundant leukocytes. In contrast, 6 hours after injection of sphingomyelinase C the microcirculation demonstrates (Fig. 3.1B) swollen endothelial cells, leukocytes adhering to them, and perhaps to each other in the vessel lumen. The abnormal nature of this inflammatory response is clear from the remarkable exclusion of erythrocytes, the plugging of the vessel by leukocytes, and leukocyte accumulation with essentially no extravasation. This unusual response results in further abnormal changes as 24 hours postinjection there was inflammation with a heavily damaged vessel, leukocyte extravasation into the tissue from the disrupted vessel, and marked tissue destruction (Fig. 3.1C). Exogenous sphingomyelinase C in vivo mimics many of the disjointed responses of C. perfringens exotoxins, and had a more profound effect on vessel integrity than would be expected from its effects on isolated endothelial cells (20). Instead, this picture is one of an overly exuberant, perhaps abnormal, leukocyte response. Sphingomyelinase C, but not Dr increases ceramide levels in PMN. In our search to determine whether ceramide affects PMN function, we first determined whether we could specifically alter ceramide levels in isolated human neutrophils by treatment with exogenous sphingomyelinases. We found (Fig. 3.2a) that exogenous sphingomyelinase C depleted two-thirds of the cellular sphingomyelin without affecting phosphatidylcholine levels. The latter point is important as it clearly demonstrates the specificity of this 39 Figure 3.1 Intradermal injection of sphingomyelinase C induces an intense leukocytic infiltration. Inflammatory response in mouse skin after intradermal injection of PBS after 6 hours (A), and SMC after 6 hours (B), and 24 hours (C). Shown are 10um sections of paraffin embedded mouse skin after hematoxylin and eosin staining at 156x magnification. 40 41 enzyme for sphingomyelin even though both complex lipids possess the same polar headgroup. Similarly, sphingomyelinase D activity in Loxosceles reclusa venom hydrolyzed two-thirds of the sphingomyelin without affecting phosphatidylcholine levels (23). We also found sequential treatment with sphingomyelinase D followed by sphingomyelinase C failed to further deplete sphingomyelin content of the treated cells. We infer that this residual content is not accessible to extracellular sphingomyelinase activity, and might therefore represent intracellular pools. In support of this, we find neither of the sphingomyelinases affected cellular viability, or permeability as measured by trypan blue exclusion (data not shown). We next determined what effect sphingomyelin hydrolysis by exogenous sphingomyelinases had on cellular ceramide content, and found (Fig. 3.2b) that exogenous sphingomyelinase C resulted in a seven and half-fold increase over basal ceramide levels within 10 minutes of exposure to the enzyme. This represents a nearly quantitative conversion of cellular sphingomyelin to ceramide and, notably, maintenance of this increased ceramide pool over the subsequent 20 minutes. It appears the cells were incapable of rapidly metabolizing this amount of ceramide, at least in the continued presence of extracellular enzyme. In contrast, sphingomyelinase D treatment failed to increase cellular cerami de levels, confirming that the product of the sphingomyelinase D reaction, ceramide-phosphate, is not readily metabolized to ceramide. The data from the above experiment demonstrated that sphingomyelinases C and D have access to the same pool of sphingomyelin, and it confirms that the product of the sphingomyelinase D reaction, ceramide-phosphate, is not a substrate for sphingomyelinase C (24). We took advantage of this inability to rapidly metabolize ceramide-phosphate to ceramide by using a pretreatment with sphingomyelinase D to convert 42 Figure 3.2. Sphingomylinase C and D hydrolize PMN membrane sphingomyelin. Panel A demonstrates a loss of sphingomylelin from whole PMN membrane preparations after treatment with SMC, SMD, or SMD and then SMC. Lipids were quantified by inorganic phosphate analysis of the phospholipids separated by HPLC. Panel B) Analysis of ceramide in samples before separation by HPLC using Diacylglycerol Kinase Assay system by Amersham. 30 ~ ~ 24 t'-­Q T""I )( ~ 18 C.fJ -Q.I Q e ..s 12 'tS '.s..... . :8 ~ 6 Q f o A Control SMC SMD SMD->SMC 8 Z f 6 t'--Q T""I Cis Q.I - 4 Q ~ ~ '.t.S... ~ 2 .... Q.I U 0 B 0 5 10 15 20 25 30 Time in Minutes 43 Phosphotidylcholine ~ Sphingomyelin SMC SMD SMD->SMC C8-ceramide - fMLP 44 sphingomyelin to ceramide-phosphate and thereby deplete the substrate for sphingomyelinase C. This key observation will be employed in subsequent experiments to specifically block the effects of sphingomyelinase C on leukocytes. We also examined the effect of a short-chain water soluble synthetic ceramide analog on cellular ceramide levels, and found that Cs­ceramide did not cause an increase. Thus, the short-chain ceramide analogs did not stimulate accumulation of long-chain ceramide derived from cellular sources. We also confirmed that fMLP does not stimulate an increase in cellular ceramide levels over the first 30 minutes of exposure to this agonist (17). This inability to detect an increase in ceramide content argues that ceramide is not a component in the fMLP signaling pathway. Ceramide is a leukocyte agonist. Ceramide accumulation in response to the agonist fMLP occurs late in the activation response, and correlates with a decrease in the respiratory burst (17). TNF, in contrast, induces a rapid accumulation of ceramide at a time when a burst of oxidizing species begins (4,16). To determine whether cerami de is a positive, negative, or neutral molecule in PMN activation, we first determined if ceramide alters intracellular calcium levels. Using the fluorescent indicator Indo-1 we found that sphingomyelinase C induced an increase in intracellular calcium, although the rate of increase was slower than that induced by fMLP (Fig 3.3). The increase in intracellular Ca++ did not result from nonspecific permeabilization of the cell as these cells still responded to a subsequent stimulation with fMLP in a fashion indistinguishable from that of control PMN. We found the water soluble cerami de analog Cs-ceramide also induced a transient increase in intracellular calcium. In contrast sphingomyelinase D treatment of leukocytes had no effect on intracellular calcium levels, so neither the loss of plasma membrane sphingomyelin, nor its conversion to 45 INDO-l Fluorescence 1 SMD fMLP +. ~ ~ l ~I.~~~~ .. ~ H 1 minute Figure 3.3. Sphingomyelinase C and ceramide induce a Ca++ flux as measured by an increase in INDO-l fluorescence. PMN were loaded with Indo-lAM as detailed under "Materials and Methods" and exposed to sphingomyelinase C (SMC) [O.5U /ml], sphingomyelinase D (SMD) [O.lul/ml], a ceramide analog (C8) [lOuM], or fMLP [lO-7M] while in a spectrofluoremeter cuvette at 370 . PMN were exposed to SMD and, after 5 minutes without a rise in Ca++, SMC was added. After an additional 5 minutes without a rise in Ca++, fMLP was then added to the SMD- and SMC-treated PMNs. 46 cerami de phosphate, directly alters calcium metabolism. We used this property to determine if ceramide accumulation accounted for the effect of sphingomyelinase C on intracellular Ca++ levels, or if this was a response to either contaminating materials or a nonenzymatic effect of sphingomyelinase C. We depleted the substrate for sphingomyelinase C digestion by first treating with sphingomyelinase D (as cerami de-phosphate is not a substrate for the C activity) and then examining changes in Ca++ levels evoked by sphingomyelinase C (24). We found that in the absence of its substrate, sphingomyelinase C did not affect Ca++ metabolism. We next used this property to deplete the pool of plasma membrane sphingomyelin to determine if this pool has an essential role in fMLP-induced Ca++ accumulation. However, we found that sphingomyelinase D pretreatment did not alter the Ca++ transient induced by a subsequent exposure to fMLP (Fig.3.3). We conclude from these studies that the ceramide produced by sphingomyelinase C treatment activates at least one signaling event in PMN, intracellular Ca++ accumulation, and that ceramide is a true leukocyte agonist. Additionally, these studies show fMLP induction of intracellular free Ca ++ accumulation does not require, nor is inhibited by, the pool of ceramide derived from plasma membrane sphingomyelinase. Ceramide accumulation causes complete degranulation of resting PMN. We determined whether ceramide accumulation has a functional consequence in leukocytes. We examined a Ca++-dependent process, release of hydrolytic enzymes from primary and secondary granules (25), and found that ceramide accumulation is a remarkable agonist for this response. We found (Fig. 3.4) that almost 90% of cellular elastase was released from the primary granules of sphingomyelinase C-treated cells, a level of degranulation that greatly exceeded the release of this marker in response to 47 Primary Granules 100 Control 80 ~ Q,I fMLP \'I.) ftS Q,I ~ Q,I 60 = Q,I \'I.) ..f.t.S. \'I.) 40 ftS ~ JJ.l Q~ 20 Control SMC SMD C8 SMD/SMC 100 Secondary Granules Control ~ fMLP o Control SMC SMD C8 SMD/SMC Figure 3.4. Sphingomyelinase C induces PMNs to degranulate. ELISA was used to detect the amount of elastase and lactoferrin released into the supernatant of PMNs treated with buffer, SMC [O.5U /mll, SMD [O.lul/mll, or Cs-ceramide [lOuM]; + / - fMLP [lO-7M]. Totallactoferrin and elastase content was determined by sonication of control PMNs with the insoluble material removed by centrifugation. 48 fMLP. An identical response to sphingomyelinase C was observed when lactoferrin release from secondary granules was determined (Fig. 3.3). As these cells had not been primed by pre-exposure to an initial agonist or by disruption of the cytoskeleton by pretreatment with a cytochalasin, the low level of degranulation in response to fMLP was anticipated. What was not expected was that ceramide accumulation resulted in almost complete degranulation, a phenomenon not previously reported. That this was indeed due to cerami de accumulation was demonstrated using sphingomyelinase D pretreatment. We found sphingomyelinase D treatment on its own did not induce enzyme release from either primary or secondary granules, but that pretreatment with this enzyme completely abolished enzyme release in response to sphingomyelinase C. PMN exposed to sphingomyelinase C continued to exclude the dyes trypan blue and propidium iodide (and retain the Ca++ indicator dye), so appearance of primary and secondary granule contents into the medium does not reflect cellular permeabilization. Water soluble analogs of cerami de are frequently employed to identify a potential role for cellular ceramide, whose use is made difficult by its insolubility. In light of the strong degranulating activity of cellular ceramide, it was surprising to find that the synthetic Cs-ceramide analog failed either to induce degranulation by itself, or even to enhance fMLP-stimulated degranulation. It appears this water soluble analog either does not sufficiently mimic the structure of long chain ceramide, or cannot access appropriate compartments to mimic the powerful effect of cellular ceramide on degranulation. Ceramide stimulates production of 02-. The respiratory burst of leukocytes resulting in 02- and H202 release depends on appropriate assembly and activation of the oxidase complex. Since the cytochrome b component of the complex is present in specific granules, we determined whether ceramide- 49 induced membrane fusion and granule secretion also allowed oxidase complex assembly and activation. We found (Fig. 3.5a) that sphingomyelinase C treatment of PMN stimulated 02- accumulation, and that wheras a distinct delay in the onset of 02- production was evident following sphingomyelinase treatment, the amount of superoxide generated was comparable to that induced by fMLP. We also found (Fig. 3.5b) pretreatment with sphingomyelinase C neither enhanced nor inhibited the fMLP induced accumulation of superoxide, suggesting ceramide is as complete an agonist as this bacterial peptide. Exposure to neither sphingomye1inase D nor Cs-ceramide stimulated 02-, nor did these agents affect fMLP-induced 02- accumulation. Again we found sphingomyelinase D pretreatment of leukocytes abolished sphingomyelinase C, but not fMLP, induction of 02- accumulation, so fMLP stimulation of this process is independent of ceramide signaling from the plasma membrane pool of sphingomyelin. Cerami de inhibits. rather than stimulates, CD18-dependent adhesion to a planar surface. Leukocytes adhere and migrate using CDll / CD18 ~2 integrin complexes. Normally this heterodimer is constitutively expressed on leukocytes, although upon appropriate stimulation additional ~2-integrin is recruited from intracellular compartments. This complex is maintained in an inactive state on quiescent cells, and is then activated to an adhesive state in ways that are currently ill defined. We determined whether sphingomyelinase C would functionally upregulate CDll / CD18 and enhance leukocyte adhesion in addition to its other agonistic effects. Surprisingly, we found that it did not do so (Fig. 3.6a). In each of several dozen experiments the adherence of sphingomyelinase C-treated PMN to immobilized gelatin, which is a measure of the ~2-integrin function, was no greater than 50 Figure 3.5. SMC induces superoxide generation. Panel A is the change in mOD over time demonstrating that SMC induction of the respiratory burst is delayed compared to fMLP or PMA induction. Panle B quantitates the amount of superoxide generation over 1 hour. Each pair of bars indicate superoxide generated by the specified treaments with (hatched bars) and without [10-7M] fMLP (filled bars). Under these conditions the generation of superoxide was equal for fMLP and SMC. The change in mOD Ihour is measured by the reduction of ferricytochrome C by the superoxide produced by the PMNs. SMD or C-s ceramide treatment of PMNs did not induce a respiratory response. 0.7 0.6 § S 0.5 .S ~ ~ 0.4 ..r:: U 0.3 SMC ---"""""'~"""==========::::::::fMLP I===:!""~_""""",,_"""'-----""'-- Control 0.2 -+---..----__ --_--...--_-__ A -J..4 ;:::1 ] o 0.2 0' 0.15 -~ "~"'"' 0.1 j:Q e­.9 Cd 0.05 .~ 0... ~ o B 40 80 120 Time (min.) Buffer fMLP 51 52 that of quiescent unstimulated cells. In contrast, stimulation with either fMLP of P AF resulted in an approximately 20-fold increase in adherent PMN. Moreover, we found that not only was sphingomyelinase C not a positive effector for adhesion, it actually completely blocked the ability of fMLP to functionally upregulate the ~2 integrin complex. This inhibition of ~2 integrin function was not unique to fMLP as adhesion in response to P AF was also inhibited by sphingomyelinase C pretreatment. Thus the block created by sphingomyelinase C pretreatment was not likely to be at the receptor level. That the event affected by sphingomyelinase C actually was ~2-dependent adhesion was demonstrated by the complete inhibition of fMLP-induced adhesion by a blocking monoclonal antibody directed against CD18 (not shown). We also demonstrated that this inhibition of CD18/CDll function was due to ceramide accumulation as treating the cells with sphingomyelinase D, which had no effect on its own, prior to sphingomyelinase C blocked its effect and restored agonist-induced adhesion. We wished to determine whether the inhibition of adhesion by ceramide was in the series of events that led to activation of ~2-integrin function, or if ceramide acted distal to the signal cascade and was a direct negative effector of ~2-integrin functions. In the latter case, we expected that ceramide would reverse CDll/CD18-dependent adhesion once the signaling events had a chance to operate and upregulate ~2-integrin function. We therefore stimulated PMN with fMLP, allowed them to adhere to a gelatinized surface, and then increased ceramide levels with exogenous sphingomyelinase C. We found (Fig. 3.6b) that treatment of adherent PMNs with SMC caused these PMNs to release from the surface. This reversal of adhesion and release of PMN from gelatin was mimicked by treatment with 53 Figure 3.6. Ceramide inhibits fMLP- and PAF-induced ~2-integrin mediated PMN adhesion to gelatin coated wells. Panel A demonstrates that SMC or C8-ceramide does not induce PMN adhesion, and inhibits fMLP-and P AF-induced adhesion. The assays were done as detailed under "Materials and Methods. II Panel B demonstrates that SMC or C8-ceramide induces release of adherent P:MNs. P:MNs were stimulated with fMLP [IO-7M] for 5 minutes at 370 . The nonadherent PMN were removed, and then either HBSSA, or HBSSA containing SMC [O.5U/ml] or C8-ceramide [IOuM], was added for an additional 25 minutes at 370 before removal of the nonadherent PMN sand quantification by video microscopy. 100-1 -~ 80- o .U.... o 60- ~ d ..o... ..~c 40- ~ ~ 20- ~ • "I" o~~~ I Control A 100 ~ ~ 80 0 U ..... 0 ~ 60 d ..0.. . 1'1.) 40 Q.I ..c 't'S -< 20 ~ 0 B 54 Buffer fMLp PAF . _I - - 'Z - % - ~~WA ... IIIIIIIII'- 'II. .... F'If.. I I - I - I fMLp~--------------~• . SMC C8-c:eramide 55 Cs-ceramide, and so clearly is the result of ceramide on a process (s) required to maintain the ~2-integrin in an activated state. Ceramide increases surface 1l2-integrin in an inactive state. We sought to determine the basis for ceramide inhibition of ~2-dependent adhesion, and so we first determined whether PMN exposed to sphingomyelinase C actually demonstrated increased amounts of this integrin on their surface. We analyzed ~2 surface expression by FACScan, and determined that quiescent neutrophils demonstrated modest CD18 staining consistent with the small pool of constitutively expressed ~2-integrin (Fig. 3.7a). As expected, fMLP stimulation distinctly enhanced this expression. We found that sphingomyeUnase C treatment also increased ~2-surface expression, and that the level of surface expression of this integrin clearly exceeded that induced by fMLP. This finding is consistent with the powerful effect of ceramide on degranulation, arid it demonstrates that the failure of sphingomyelinase C­treated PMN to adhere is not due to a lack of surface adhesion protein. We infer that the ~2-integrin present on the surface of a sphingomyelinase-treated PMN exists in an unactivated state, like the constitutively expressed integrin, that does not support adhesion. \lYe sought to determine if the newly expressed molecules, like the constitutively expressed molecules, could be activated by cation switching. It has been determined that manganese, when substituted for the inhibitory calcium ion, converts inactive ~2-integrin complexes to ones capable of supporting adhesion (26,27). We therefore treated PMN with sphingomyelinase C, or not, removed integrin-bound cations with EDTA and then reconstituted surface ~2 with either Mn++, or Mg++ and Ca++. We found substitution of Mn++ for endogenous divalent cations allowed unactivated PMN to adhere to 56 Figure 3.7. SMC induces quantitative up regulation of CD18 in the abscence of receptor clustering and F-actin accumulation. Panel A is flow cytometric analysis of CD18 expression on PMNs that were treated as indicated and stained with anti-CD18 mAb and FITC-conjugated goat anti­mouse as detailed under "Materials and Methods". These same cells were also centrifuged on to microscope slides and the CD 18 staining visualized by fluorescent microscopy (Panel B). Panel 1 are unstimulated control cells. Panel 2 are fMLP stimulated demonstrating clustering of CD18. SMC treament, panel 3, with fMLP, panel 4, do not have this punctate pattern of CD18. The lack of CD 18 clusters maybe due to the loss of F-actin induced by SMC treatment of PMNs (Panel C). PMNs were treated as indicated and immediately put into a fixative containing FITC-conjugated phalloidin and lysophosphatidylcholine to permeablize the cells as described in "Materials and Methods." To detect and quantify the f-actin content in PMNs FITC­phalloidin and flowcytometry was 500 o A 750 o c B 57 102 CD18 Fluorescence SMC 25' + fMLP 5' 101 F-actin Fluorescence 58 a laminin-coated surface just as if they had been activated by an agonist (Fig. 3.8). Indeed, this adhesion was blocked with the inhibitory 60.3 monoclonal antibody against CD18. The cation replacement procedure per se did not stimulate latent ~2 function as the replacement of the endogenous cations with calcium and magnesium did not increase leukocyte adhesion in the absence of an agonist. Conversely adding fMLP to PMN that had been through this stripping and Ca++, Mg++ replacement procedure stimulated adhesion, so this process did not adversely affect stimulated ~2-integrin function. When we subjected sphingomyelinase C-treated neutrophils to the cation switching procedure, we found that Mn++ substitution induced ~2- dependent adhesion. Interestingly the level of adhesion of sphingomyelinase C-treated PMN was not different from untreated control leukocytes following this ion switching, so the level of adhesion was not proportioned to its surface expression following SMC treatment. It is important to note that reconstitution with Ca++ and Mg++ retained the phenotype of the unprocessed SMC treated cells; sphingomyelinase C-treated PMN failed to adhere to the laminin surface after fMLP stimulation. This experiment also shows the defect in adhesion is not limited to a gelatinized surface as this experiment employed a surface coated with laminin and SMC treated PMNs could not adhere after fMLP stimulation. These results demonstrate that the ~2-integrin expressed on sphingomyelinase C-treated PMN is expressed in a state that is unable to support adhesion but is not irretrievably locked in this state. Ceramide does not block ~2-integrin dependent aggregation. To further determine if the CDll/CD18 complex on sphingomyelinase C-treated PMNs was expressed in a nonfunctional state we sought to determine if the binding of other ligands was similarly depressed. We employed an agonist-induced 59 2000 Control 1600 Z ~ ~ 1200 t: ~ QJ ..s= "<'0 800 ""o'"' =It: 400 o ~ SMC Figure 3.8. Mn++ induces ~2-integrins on SMC treated cells to adhesion to laminin-coated wells. Mab 60.3 blocks this adhesive event demonstrating it is ~2-integrin mediate adhesion. Readdition of Mg++ or Ca++ ions does not induce adhesion demonstrating this is a Mn++ the surface CD18 can be transformed into the high affinity state that facilitates adhesion. Subsequent exposure of SMC treated PMNs, after readdition Mg++ and Ca++, to fMLP do not become adherent demonstrating this is an ion++ dependent adhesive event. 60 PMN CD18-dependent aggregation assay (Fig. 3.9Al). However we were surprised to find that PMN aggregation was not blocked by ceramide accumulation. The tracings of alterations in light dispersion as a consequence of PMN aggregation clearly show that fMLP induces aggregation (Fig. 3.9Al), and that a similar process occurs in sphingomyelinase C-pretreated cells (Fig. 9A2). Both these events were suppressed by a blocking antibody against CD18 integrin, 60.3, so the aggregation was dependent on ~2-integrin function. While sphingomyelinase C did not directly stimulate aggregation (Fig. 3.9B), it also does not suppress it. Because of the complete disparity between adhesion to a planar surface and aggregation as measured by this assay, we considered the possibility that changes in light transmission actually reflected changes in leukocyte optical properties, due to the massive degranulation induced by sphingomyelinase C, rather than true aggregation. However PMA, which is a better degranulating agent than fMLP, produced similar recordings (Fig. 3.9A). Moreover Cytospin® analysis (Fig. 3.9B) clearly demonstrates the presence of small aggregates following fMLP stimulation and massive aggregates induced by PMA in both control and sphingomyelinase C-treated populations. Leukocytes employed in these aggregation assays were examined in parallel for their adhesiveness to a gelatinized surface (not shown), with results completely concordant with our previous results (Fig. 3.6). Therefore ceramide blocks ~2-integrin dependent planar adhesion but not ~2-integrin dependent aggregation, and so this integrin can exist in a state that can distinguish between these events. Ceramide alters F-actin content and 62-integrin clustering. A change in cytoskeletal actin content in response to ceramide was intriguing as leukocyte adhesion to a surface is followed by cellular spreading, and by sequestration of ~2-integrins to the contact surface (28). Clustering of these adhesion Al Control fMLP Jc=-:=-=+mA 60.3 B Quiescent • Untreated '•"• .. ~ • f •• '." " . ... ., •• SMC • .... ' • Pretreated •• " .- +mAb 60.3 fMLP Stimulated I • .~iJ' • • • .it >,~., .. 2 ,?.',L ..·... .' • " .' .... , ": it .~ .. ,,'.':" . • ... .-. ~'.', -. ~ 4 . ' .. • ., . .. ~ ~ . f • 61 +mA L..-L,...-----''''''""-;:;---;60.3 6 PMA Stimulated +mAb 60.3 Figure 3.9. Aggregation of PMNs is not affected by SMC pretreatment. Panel A) Aggregation recorder tracings of PMN exposed to fMLP [10-7M] or PMA [10-7M] after prior exposure to buffer or SMC. mAb60.3 inhibits the majority of aggregation demonstrating that ~2-integrin accounts for stimulated aggregation. Panel B) Microscopic analysis, using Normarsky optics, of PMNs from the aggregation assay demonstrates that PMN aggregates are present and that SMC affects PMN morphology. PMNs were removed from the aggregometer at the end of the aggregation assay and centrifuged on to glass microscope slides and stained with DifQuick. 62 molecules strengthens adhesion, so disruption of the cytoskeleton may affect adhesion to a greater extent than aggregation which may require smaller contact areas. We probed the location of ~2-integrins on leukocytes with an anti-CD18 antibody followed by a fluorescent secondary antibody, and determined that resting cells demonstrated weak, but detectable, fluorescence that appeared as punctate areas embedded in a field of hazy staining (Fig. 3.7B). When these cells were activated by fMLP there was a marked increase in fluorescence intensity, a marked increase in numbers of clusters, and a marked increase in the fluorescent intensity of the clusters. When we examined quiescent PMN after sphingomyelinase C treatment, the intensity of CD18 staining was enhanced, consistent with the increase demonstrated upon FACScan analysis. Some punctate staining was apparent, but much of the newly recruited CD18 was uniformly spread on the surface generating a luminescent background. Leukocytes pretreated with sphingomyelinase C and then activated with fMLP demonstrate a marked reduction in the number and intensity of stained integrin clusters. Sphingomyelinase C treatment did not abolish agonist-stimulated clustering of the adhesion complex, but clearly had a significant depressing effect on this event. Inspection of the low power photomicrographs obtained in the above Cytospin® analyses left us with the impression that PMN exposed to sphingomyelinase C were slightly larger than control cells, and displayed a more distinct cytoplasm. This suggested that the cytoskeleton was affected by the sphingomyelinase C treatment, and that perhaps this in tum differentially affect planar adhesion compared to aggregation. Accordingly, we assayed f-actin content by F ACScan analysis of FITC -phalloidin stained cells, and found (Fig. 3.7c) that the f-actin content of sphingomyelinase C treated, quiescent PMN was less than control cells. Furthermore, it was 63 apparent that while fMLP stimulation increased f-actin content in sphingomyelinase-treated cells, sphingomyelinase treatment attenuated this response so that there was only a small difference over that of quiescent control cells. We conclude that sphingomyelinase C treatment affects cytoskeletal reorganization, and, likely, through this event affects localization of ~2-integrins on the surface of the cell that may be essential for adhesion to a planar surface. Cytochalasin D mimics cerami de effects on adhesion and aggregation. The foregoing observations suggest that the primary effect of ceramide in modulating surface ~2-integrin function derives from its effect on the cytoskeleton. To support this conclusion, we examined an agent that directly affects the cytoskeleton, cytochalasin D. We found (Fig. 3.10A) that, like ceramide, cytochalasin D treatment abolished the ability of PMN to adhere to a gelatin surface in response to fMLP simulation. Cytochalasin D pretreatment also greatly decreased PMN adhesion in response to the powerful stimulus, PMA. We next examined the effect of cytochalasin D on aggregation, and found (Fig. 3.10B) that while cytochalasin D did not by itself induce aggregation neither did it suppress aggregation. Again, this is similar to the effects of ceramide. We infer from this that the disparity in ~2-integrin function revealed by comparing planar adhesion and aggregation reflects underlying cytoskeletal alterations. Discussion Complex cellular lipids provide cells with the precursors for intra and intercellular communication. The sequestration of these signaling lipid precursors consisting of a variety of phospholipids and sphingomyelin, in the membrane, in a spatially distinct, nonaqueous environment offers several 64 A 100 .-- Buffer 0 Jj 80 § U ~ fMLP ...... 0 ;;:g 60 e....- EI PMA .§ (e!n) 40 ..c:: "0 < Z 20 ~ p., a Control SMC CytoD B H Control SMC CytoD 1 minute Figure 3.10. SMC and Cytochalasin D pretreatment of PMNs inhibit agonist induced adhesion but not aggregation. Panel A) PMNs were pretreated with SMC or cytochalasin D (cytoD) and then aded to the gelatin coated wells with or without fMLP [10-7M], PMA [10-8M], or buffer for 5 minutes at 370 as detailed under "Materials and Methods". Panel B) Half the PMNs pretreated for the adhesion assay in Panel A were used to determine their ability to aggregate in response to the same agonist. 65 advantages to the cell in this regard, but also poses several challenges. One of these is the necessity to distinguish between the lipid moiety of phosphatidylcholine and sphingomyelin, both of which have the same polar headgroup exposed to solvent. This is particularly intriguing as sphingomyelin is localized to the outer leaflet of the plasma membrane (24). The neutral lipid ceramide, however, readily translocates across the membrane bilayer and the neutral, stimulatable sphingomyelinase exists on the outer leaflet of the plasma membrane (29,30). A role for the stimulated hydrolysis of phosphatidylcholine has been well established. A similar role for the stimulated cleavage of sphingomyelin is expanded in this study. Hydrolysis of sphingomyelin by cellular sphingomyelinases generates ceramide and phosphorylcholine. The neutral lipid ceramide clearly activates signal transduction cascades in numerous cell types, although the relevant ceramide-binding activity and its downstream effectors are not well defined (16,31-35). For instance, it is established that TNF through its TNF-R1 receptor activates a neutral plasma membrane sphingomyelinase activity, even in isolated preparations of cellular membranes, and ceramide in cell free preparations activates IKB degradation (30,36,37). This is important as IKB associates with NF-kB in the cytoplasm, and disassociation from NF-kB, allows translocation of NF-kB to the nucleus, where it is instrumental in many TNF responses. A second sphingomyelinase activity, with an acidic pH optimum, also can be activated in TNF-treated cells and plays an essential role in some responses (36). Ceramide is reported to activate protein kinase C ~, a phosphatase of the 2A family, and ceramide accumulation results in phosphorylation and activation of a 97 kDa protein kinase (15,38-40). The precise role for ceramide in cellular stimulation by TNF, and other receptor derived signaling, has been difficult to establish as there are no 66 genetic or pharmacologic tools to abolish sphingomyelinase activity. However, we have recently been able to divorce much of TNF signaling from ceramide accumulation in primary cultures of human umbilical vein endothelial cells (41). Here, we have been able to take advantage of the inability of sphingomyelinase C to metabolize ceramide-1-phosphate, the product of sphingomyelinase D, to show that plasma membrane ceramide is not essential for agonist-induced PMN responses. Thus even though ceramide is uniquely effective at inducing degranulation, it is not essential for this response in cells stimulated by G protein-linked agonist receptors. The unusual response of PMNs to ceramide likely are critical to the dysregulated inflammation engendered by S. aureus or C. perfringens exotoxins (19,42). The unusual and massive leukocyte sequestration within the vascular lumen in response to the specific sphingomyelinase C of S. aureus followed by destruction of the vessel is consistent with defective PMN CD18 function coupled with the release of priming and secondary granules into the bloodstream. At least for C. perfringens, the <x-toxin that cleaves both sphingomyelin and phosphatidylcholine is an essential exotoxin in the development of gas gangrene (24). Previously, ceramide analogs with short N-acyl residues such as N~ acetylsphingosine, which in contrast to cerami des are water soluble, were found to inhibit the respiratory burst of PMN (32,43). These, however, may also model sphingosine - a protein kinase C inhibitor, and this may contribute to the ability of N-acetylsphingosine to potentiate or stimulate 02- generation (44). In contrast, we find the long chain ceramides generated from the hydrolysis of cellular sphingomyelin were agonistic for certain events. Cerami de accumulation induced a rapid rise in intracellular free Ca+2, a respiratory burst, and release of primary and secondary granules. That this 67 was due to ceramide was shown by the complete blockade of these responses by depleting the cells of sphingomyelin by pretreatment with sphingomyelinase D. Both types of sphingomyelinase act on the same pool of sphingomyelin as shown by an experiment in which a combination of the two did not hydrolyze more than the individual enzymes (Fig. 3.2). The reason that approximately one-third of the cellular sphingomyelin was resistant to hydrolysis is not defined, but a similarly resistant pool of sphingomyelin is found in the promyelocytic cell line HL60 (45). In one report this was documented as sequestration of plasma membrane sphingomyelin within the cell due to sphingomyelinase -induced vesiculation (46). It is also likely that SMC is not able to access the internal sphingomyelin containing membranes and therefore represents the SMC resistant pool. Our results show that ceramide, in the absence of a change in diglyceride levels, activates the respiratory burst. However, our novel method to specifically block this event shows cerami de formation from this route is not required for activation of the burst, or degranulation, in response to the receptor-mediated agonists P AF or fMLP. The effector functions of leukocytes depend on the regulated release of oxidizing agents, hydrolytic enzymes, and nutrient binding proteins. The induction of this effector function in PMNs is subject to stringent regulation, as shown by the tissue destruction that ensues from the excessive recruitment and activation of PMNs after sphingomyelinase injection. As initial exposure to an agonist in physiologic inflammation would be in the vascular compartment, it is therefore understandable that a requirement for other stimuli to effectively induce degranulation. Degranulation is enhanced when primed PMNs are exposed to a second agonist (47), adhere to an appropriate surface (48) transmigrate (49)/ or are exposed to cytoskeletal disrupting agents 68 like cytochalasin D (50,51). In fact, in vitro studies of agonist-induced degranulation routinely employ cytochalasin as a recurring component of the assay (52) even though cytochalasin D does not have a physiologic correlate. As a practical matter, it raises the level of enzyme release to soluble agonists from a few percent to a few tenths of the cellular enzyme content. Our observation that ceramides generated from endogenous sphingomyelin disrupt the cytoskeleton, and decrease f-actin content, coupled with the uniquely complete degranulation, suggest that cerami de could be the physiologic correlate of cytochalasin D. For adherent, TNF-a exposed PMN, this may be a relevant route ensuring the release of hydrolytic and cytostatic mediators occurs at an appropriate level in an appropriate environment. References 1. Lawrence, M. B. and T. A. Springer. 1991. Leukocytes roll on a selectin at physiologic flow rates: Distinction from and prerequisite for adhesion through integrins. Cell 65:859-873. 2. Ley, K., P. Gaehtgens, C. Fennie, M. S. Singer, L. A. Lasky, and S. D. Rosen. 1991. Lectin-like cell adhesion molecule 1 mediates leukocyte rolling in mesenteric venules in vivo. Blood 77:2553-2555. 3. Anderson, D. C. and T. A. Springer. 1987. Leukocyte adhesion deficiency: An inherited defect in the Mac-1, LFA-1, and p150,95 glycoproteins. Ann Rev Med 38:175-194. 4. Frank, M. 0., G. W. Sullivan, H. T. Carper, and G. L. Mandell. 1994. Inhibition of phospholipase D blocks activation of fibrinogen-adherent neutrophils by tumor necrosis factor. Infect Irnrnun 62:2622-2624. 5. Detmers, P. A., S. D. Wright, E. Olsen, B. Kimball, and Z. A. Cohn. 1987. Aggregation of complement receptors on human neutrophils in the absence of ligand. J Cell BioI 105:1137-1145. 6. Hermanowski-Vosatka, A., P. A. Detmers, O. Gotze, S. C. Silverstein, and S. D. Wright. 1988. Clustering of ligand on the surface of a particle enhances adhesion to receptor-bearing cells. J BioI Chern 263:17822-17827. 69 7. Altieri, D. C. 1991. Occupancy of CD11b/CD18 (Mac-I) divalent ion binding sites(s) induces leukocyte adhesion. J Immunol 147:1891-1898. 8. Wright, S. D. and P. A. Detmers. 1988. Adhesion-promoting receptors on phagocytes. J Cell Sci 9:99-120. 9. Altieri, D. C., S. J. Stamnes, and C. G. Gahmberg. 1992. Regulated Ca2+ signaling through leukocyte CD11b/CD18 integrin. Biochem J 288:465-473. 10. Suchard, S. J. and L. A. Boxer. 1994. Exocytosis of a subpopulation of specific granules coincides with H202 production in adherent human neutrophils. J Immunol 152:290-300. 11. Dapino, P., F. Dallegri, L. Ottonello, and C. Sacchetti. 1993. Induction of neutrophil respiratory burst by tumour necrosis factor-alpha; priming effect of solid-phase fibronectin and intervention of CD11b-CD18 integrins. Clin Exp Immunol 94:533-538. 12. Shappell, S. B., C. Toman, D. C. Anderson, A. A. Taylor, M. L. Entman, and C. W. Smith. 1990. Mac-1 (CD11b/CD18) mediates adherence-dependent hydrogen peroxide production by human and canine neutrophils. J Immunol 144:2702-2711. 13. Kolesnick, R. and D. W. Golde. 1994. The sphingomyelin pathway in tumor necrosis factor and interleukin-1 signaling. Cell 77:325-328. 14. Raines, M. A., R. N. Kolesnick, and D. W. Golde. 1993. Sphingomyelinase and ceramide activate mitogen-activated protein kinase in myeloid HL-60 cells. J Biol Chem 268:14572-14575. 15. Dobrowsky, R. T., C. Kamibayashi, M. C. Mumby, and Y. A. Hannun. 1993. Ceramide activates heterotrimeric protein phosphatase 2A. J BioI Chem 268: 15523-15530. 16. Ohta, H., Y. Yatomi, E. A. Sweeney, S. -I. Hakomori, and Y. Igarashi. 1994. A possible role of sphingosine in induction of apoptosis by tumor necrosis factor-a in human neutrophils. FEBS Lett 355:267-270. 17. Nakamura, T., A. Abe, K. J. Balazovich, D. Wu, S. J. Suchard, L. A. Boxer, and J. A. Shayman. 1994. Cerami de regulates oxidant release in adherent human neutrophils. J BioI Chem 269: 18384-18389. 18. Bryant, A. E. and D. L. Stevens. 1996. Phospholipase C and perfringolysin 0 from Clostridium perfringens upregulate endothelial cell-leukocyte adherence molecule 1 and intercellular leukocyte adherence molecule 1 70 expression and induce interleukin-8 synthesis in cultured human umbilical vein endothelial cells. Infect Immun 64:358-362. 19. Bryant, A. E., R. Bergstrom, G. A. Zimmerman, J. L. Salyer, H. R. Hill, R. K. Tweten, H. Sato, and D. L. Stevens. 1993. Clostridium perfringens invasiveness is enhanced by effects of theta toxin upon PMNL structure and function. FEMS Immunol Med Microbiol 7:321-336. 20. Patel, K. D., V. Modur, G. A. Zimmerman, S. M. Prescott, and T. M. McIntyre. 1994. The necrotic venom of the brown recluse spider induces dysregulated endothelial cell-dependent neutrophil activation. Differential induction of GM-CSF, IL-8 and E-selectin expression. J Clin Invest 94:631-642. 21. Awad, M. M., A. E. Bryant, D. L. Stevens, and J. I. Rood. 1995. Virulence studies on chromosomal a-toxin and a-toxin mutants constructed by allelic exchange provide genetic evidence for the essential role of a-toxin in Clostridium perfringens-mediated gas gangrene. Molecular Microbiology 15:191-202. 22. Titball, R. W. 1993. Bacterial phospholipases C. Microbiological Reviews 57:347-366. 23. Kurpiewski, G., L. J. Forrester, J. T. Barrett, and B. J. Campbell. 1981. Platelet aggregation and sphingomyelinase D activity of a purified toxin from the venom of Loxosceles reclusa. Biochim Biophys Acta 678:467-476. 24. Spence, M. W. 1993. Sphingomyelinases. Adv Lipid Res 26:3-23. 25. Lew, P. D., A. Monod, F. A. Waldvogel, B. Dewald, M. Baggiolini, and T. Pozzan. 1986. Quantitative analysis of the cytosolic free calcium dependency of exocytosis from three subcellular compartments in intact human neutrophils. J Cell BioI 102:2197-2204. 26. Bohnsack, J. F. and X. -N. Zhou. 1992. Divalent cation substitution reveals CD18- and very late antigen-dependent pathways that mediate human neutrophil adherence to fibronectin. J Immunol 149:1340-1347. 27. Dransfield, I., C. Cabanas, A. Craig, and N. Hogg. 1992. Divalent cation regulation of the function of the leukocyte integrin LFA-1. J Cell BioI 116:219- 226. 28. Clark, E. A. and J. S. Brugge. 1995. Integrins and signal transduction pathways: The road taken. Science 268:233-239. 71 29. Das, D. V., H. W. Cook, and M. W. Spence. 1984. Evidence that neutral sphingomyelinase of cultured murine neuroblastoma cells is oriented externally on the plasma membrane. Biochirn Biophys Acta 777:339-342. 30. Dressler, K. A., S. Mathias, and R. N. Kolesnick. 1992. Tumor necrosis factor-a activates the sphingomyelin signal transduction pathway in a cell­free system. Science 255:1715-1718. 31. Chen, C. -S., A. G. Rosenwald, and R. E. Pagano. 1995. Ceramide as a modulator of endocytosis. J BioI Chern 270:13291-13297. 32. Wilson, E., M. C. Olcott, R. M. Bell, A. H. Merrill,Jr., and J. D. Lambeth. 1986. Inhibition of the oxidative burst in human neutrophils by sphingoid long-chain bases: Role of protein kinase C in activation of the burst. J BioI Chern 261:12616-12623. 33. Jones, M. J. and A. W. Murry. 1995. Evidence that ceramide selectively inhibits protein kinase C-a translocation and modulates bradykinin activation of phospholipase D. J BioI Chern 270:5007-5013. 34. Obeid, L. M., C. M. Linardic, L. A. Karolak, and Y. A. Hannun. 1993. Programmed cell death induced by ceramide. Science 259:1769-1771. 35. Olivera, A., N. E. Buckley, and S. Spiegel. 1992. Sphingomyelinase and cell­permeable ceramide analogs stimulate cellular proliferation in quiescent swiss 3T3 fibroblasts. J BioI Chern 267:26121-26127. 36. Wiegmann, K., S. Schutze, T. Machleidt, D. Witte, and M. Kronke. 1994. Functional dichotomy of neutral and acidic sphingomyelinases in tumor necrosis factor signaling. Cell 78:1005-1015. 37. Machleidt, T., K. Wiegmann, T. Henkel, S. Schutze, P. Baeuerle, and M. Kronke. 1994. Sphingomyelinase activates proteolytic IkB-a degradation in a cell-free system. J BioI Chern 269:13760-13765. 38. Lozano, J., E. Berra, M. M. Municio, M. T. Diaz-Meco, I. Dominguez, L. Sanz, and J. Moscat. 1994. Protein kinase C ~ isoform is critical for kB­dependent promoter activation by sphingomyelinase. J BioI Chern 269:19200- 19202. 39. Dobrowsky, R. T. and Y. A. Hannun. 1992. Ceramide stimulates a cytosolic protein phosphatase. J BioI Chern 267:5048-5051. 72 40. Liu, J., S. Mathias, Z. Yang, and R. N. Kolesnick. 1994. Renaturation and tumor necrosis factor-a stimulation of a 97-kDa ceramide-activated protein kinase. J BioI Chem 269:3047-3052. 41. Modur, V., G. A. Zimmerman, S. M. Prescott, and T. M. McIntyre. 1996. Endothelial cell inflammatory responses to TNFa: ceramide-dependent and independent MAP kinase cascades. J BioI Chem (In Press) 42. Ward, P. D., C. Adlam, A. C. McCartney, J. P. Arbuthnott, and C. M. Thorley. 1979. A histopathological study of the effects of highly purified staphylococcal a and ~ toxins on the lactating mammary gland and skin of the rabbit. J Comp Path 89:169-177. 43. Wong, K., X. -B. Li, and N. Hunchuk. 1995. N-acetylsphingosine (C2- ceramide) inhibited neutrophil superoxide formation and calcium influx. J BioI Chem 270:3056-3062. 44. Merrill, A. H.,Jr. and V. L. Stevens. 1989. Modulation of protein kinase C and diverse cell functions by sphingosine - a pharmalogically interesting compound linking sphingolipids and signal transduction. Biochim Biophys Acta 1010:131-139. 45. Linardic, C. M. and Y. A. Hannun. 1994. Identification of a distinct pool of sphingomyelin involved in the sphingomyelin cycle. J BioI Chem 269:23530- 23537. 46. Allan, D. and C. M. Walklin. 1988. Endovesiculation of human erythrocytes exposed to sphingomyelinase C. Biochim Biophys Acta 938:403- 410. 47. Elbim, C., S. Chollet-Martin, S. Bailly, J. Hakim, and M. A. Gougerot­Pocidalo. 1993. Priming of polymorphonuclear neutrophils by tumor necrosis factor a in whole blood. Blood 82:633-640. 48. Henson, P. M. 1971. The immunologic release of constituents from neutrophil leukocytes. J Immunol 107:1547-1557. 49. Kubes, P., X. F. Niu, C. W. Smith, M. E. Kehrli,Jr., P. H. Reinhardt, and R. C. Woodman. 1995. A novel ~l-dependent adhesion pathway on neutrophils. FASEB J 9:1103-1111. 50. Henson, P. M. and Z. G. Oades. 1973. Enhancement of immunologically induced granule exocytosis from neutrophils by cytochalasin B. J Immunol 110:290-293. 73 51. Hawkins, D. 1973. Neutrophilic leukocytes in immunologic reactions in vitro effect of cytochalasin B. J Immunol 110:294-296. 52. Henson, P. M., J. E. Henson, C. Fittschen, D. L. Bratton, and D. W. H. Riches. 1992. Degranulation and secretion by phagocytic cells. In Inflammation: Basic principles and clinical correlates. J.1. Gallin and LM. Goldstein, editors. Raven Press, New York. 511-539. CHAPTER 4 ICAM-3 OUTSIDE IN SIGNALING IN HUMAN NEUTROPHILS Introduction The interactions between adhesion molecules and their counterreceptors playa key role in effector functions of the immune response by regulating the specificity and strength of cell-cell interactions. Not only are adhesion molecules involved in tethering cells to each other and to extracellular matrix components, but recently the cell adhesion protein ICAM-3 has been implicated in signal transduction events that lead to gene expression and altered adhesive functions (1-4). An established example of an adhesion molecule transmitting intracellular signals is antibody crosslinking of CDll/18 integrins. This crosslinking primes neutrophils for superoxide production, and induces an increase in intracellular free Ca++, exocytosis of azurophilic granules, up regulation of CD18, L-selectin shedding, and actin polymerization (5,6), Furthermore, cross linking of this molecule on monocytes induces nuclear translocation of NF-kB that is important in inflammatory gene transcription (7). Three members of the immunoglobulin superfamily, ICAM-l (CD54), ICAM-2 (CDI02), and ICAM-3 (CD50), are highly glycosylated type-I transmembrane proteins that are counterreceptors for CDlla/CD18 (LFA-l)(8- 10). ICAM-l and -3 contain five extracellular Ig-like domains, whereas ICAM-2 contains only two (Fig. 4.1)(8-12). The two Ig-like domains in ICAM-2 are most ICAMs Extracellular Domains ICAM-l CDlla/ CD18 CDllb/ CD18 75 actin cytoskeleton Figure 4.1. Proposed structures and interactions between ICAMs 1, 2, 3 and the ~2-integrins CDlla/CD18 and CDllb/CD18. ICAM-l, 2, and 3 are members of the immunoglogulin superfamily. The lightly shaded domain 3 of ICAM-l and ICAM-3 illustrate highly homologous domains. The hair like structures depict hetergeneous glycosylation of the ICAMs. The arrows pointing from the ~