This is to certify that the thesis entitled THE EFFECT OF SYNTHETIC PROTEASE INHIBITORS 0N LYSOSOMAL ENZYME RELEASE FROM HUMAN NEUTROPHILS presented by David L. McCloy has been accepted towards fulfillment of the requirements for Masters degree in PathologL W , 1w Major professor Date July 11, 1978 0-7639 LIBRARY Michigan Sm: University THE EFFECT OF SYNTHETIC PROTEASE INHIBITORS ON LYSOSOMAL ENZYME RELEASE FROM HUMAN NEUTROPHILS By David L. McCloy A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Pathology 1978 ABSTRACT THE EFFECT OF SYNTHETIC PROTEASE INHIBITORS ON LYSOSOMAL ENZYME RELEASE FROM HUMAN NEUTROPHILS By David L. McCloy Serine esterase activity has been implicated in the neutrOphil function of lysosomal enzyme release. To better characterize this esterase role and its possible plurality, the effect of synthetic protease inhibitors on complement- and noncomplement-mediated enzyme secretion was investigated. Noncytotoxic, selective degranulation was achieved by exposing human neutrophils to either 10% Zymosan-Activated Plasma (ZAP) or 10‘4 M N-formyl-Methionyl-Phenyla1anine (fMP) in the presence of 5 ug/ml of Cytochalasin B. The majority of the releasing capability generated With ZAP was shown to be CS-related using CS-deficient plasma. Lysosomal markers assayed were Beta-Glucuronidase for primary granule response and Lysozyme for secondary granule involvement. The inhibitors were used to pretreat the cells or were added concomitantly with the releasing stimulus. The tryptic inactivator TLCK inhibited both primary and secondary granule release regardless of the stimulus and inactivation protocol. The chymotryptic inhibitor, TPCK, only blocked fMP-induced David L. McCloy release in the absence of inducer; however, when present with either stimulus, secondary granule secretion alone was inhibited. The sulfonyl halide, PMSF, did not impede release under any of the experimental conditions. The chymotrypsin model substrate, BTEE, blocked secondary release by both ZAP and fMP only when present with the secretory stimuli. Its counterpart for tryptic activity, TAME, had no effect on release. These studies suggest the involvement of tryptic and chymotryptic activated esterases in lysosomal enzyme release from human neutrophils. Their role may be discrimi- natory with regard to both the releasing stimulus and granule reSponse. To Christine and Jennifer whom I dearly love ii ACKNOWLEDGEMENTS Verbal or written expression of my gratitude to Dr. Richard A. Patrick seems inadequate. As my major profes- sor, this compassionate individual provided continual support throughout my research but, more importantly, he Opened my eyes to a true understanding of the educational experience beyond that offered in the classroom. For this invaluable gift and his constructive insight, I am eternally thankful. My special thanks to the other members of my committee, Dr. Wayne Smith and Dr. Robert Leader, for their continual support. Dr. Smith's assistance throughout my experimenta- tion was deeply appreciated. Mrs. Martha Thomas, my academic advisor, gave me the aid and advice needed to establish my graduate curriculum. For her endeavors and help in my achieving this goal, I am especially grateful. Ms. Barbara Giese and Mr. James Hollers provided technical advice when needed. A well-deserved expression of thanks is extended to both of them. A special note of thanks to Mr. Hollers for his excellent preparation of the figures shown in this thesis. Ms. Giese is responsible for the preparation of CS-deficient plasma used. iii My thanks to Ms. Janice Fuller for the exceptional typing quality and arrangement shown in this manuscript. Although this thesis has been dedicated to my wife and daughter, I must mention my heartfelt appreciation for the patience and understanding that my spouse, Christine, demonstrated during difficult times. She deserves as much credit as I for the accomplishments made during my graduate studies. iv TABLE OF CONTENTS INTRODUCTION LITERATURE REVIEW. NeutrOphil Lysosomes. Selective Degranulation . Cytochalasin B (CB) . . . Inhibition of NeutrOphil Functions. MATERIALS AND METHODS. Reagents and Solutions. Neutrophil Isolation. Enzyme Assays Enzyme Secretion Protocol Inhibitor Treatment of Cells. . Statistical Procedure for Data Presentation Preparation of CS- Deficient Plasma. RESULTS. Inhibition of Enzyme Release with the Trypsin-Like Inhibitor TLCK . Inhibition of Enzyme Release with the Chymotrypsin- Like Inhibitor TPCK. . . Time Course of TLCK and TPCK Inhibition of Lysozyme Release with fMP . . . . . Inhibition of Enzyme Release with BTEE. Lack of Inhibition of Enzyme Release with PMSF and Cycoheximide . Association of CS with the Active Principle in ZAP. DISCUSSION SUMMARY AND CONCLUSIONS. BIBLIOGRAPHY VITA . 24 28 33 33 36 39 44 45 52 LIST OF TABLES Table Page 1 TLCK inhibition of lysosomal enzyme release . . 27 2 Lysozyme release after TPCK and PMSF pretreatment. . . . . . . . . . . . . . . . . . 31 3 TPCK inhibition of lysosomal enzyme release . . 32 4 Effect of PMSF on enzyme release. . . . . . . . 37 5 Effect of cycloheximide on lysozyme release . . 38 6 Summary of the inhibitory effects on lysosomal enzyme release. . . . . . . . . . . . . . . . . 41 vi LIST OF FIGURES Figure Page 1 TLCK pretreatment . . . . . . . . . . . . . . . 26 2 TPCK pretreatment then fMP release. . . . . . . 30 3 BTEE inhibition of enzyme release . . . . . . . 35 vii BCF BG BTEE CIINA CB DFP EACA fMP HBSS LDH LYS PMN PMSF TAME TLCK TPCK ZAP LIST OF ABBREVIATIONS Bacterial Chemotactic Factor Beta-glucuronidase Benzoyl Tyrosine Ethyl Ester Cl inactivator Cytochalasin B Diisopropylfluorophosphate Epsilon Amino Caproic Acid N-formyl~Methionyl-Phenylalanine Hank's Balanced Salt Solution Lactic Dehydrogenase Lysozyme Polymorphonuclear leukocyte Phenylmethylsulfonylfluoride Tosyl Arginine Methyl Ester N-tosyl-L-lysyl-chloromethyl-ketone N-tosylamide-phenylethyl-chloromethy1-ketone Zymosan-activated plasma viii INTRODUCTION An inflammatory reaction culminates the body's cellu- lar response to an infectious challenge; however, during the process the protective function of the neutrophil is overshadowed by the deleterious effect of its lysosomal constituents on the surrounding tissues (18,22,34). Enzymes are released either during phagocytosis by regurgi- tation during feeding or when the cells encounter a non- phagocytosable surface by reverse endocytosis (69). Factors which are Chemotactic for neutrophils will also promote lysosomal degranulation. These include complement- derived factors such as the active fragments of the third and fifth components of complement, C3a and C53 (10,11) as well as serum or plasma activated through the alternate complement pathway by zymosan (27): a bacterial Chemotactic factor (BCF) and synthetic di- and tripeptides formylated at their N-termini (10,55). The release induced by these Chemotactic substances is enhanced significantly when the neutrophils are exposed to the fungal metabolite Cytochalasin B (27,55). This drug disrupts contractile microfilament function and inhibits chemotaxis as well as phagocytosis at concentrations that cause increased lysosomal secretion (9,75,76). Thus, the use of CB-treated neutrophils provides 2 a model system for the study of lysosomal degranulation in the absence of phagocytosis. By their inhibition with DFP, the neutrophil functions of chemotaxis, phagocytosis and lysosomal enzyme release demonstrate the role of one or more serine esterases in the activation of these inter- related processes (11,51,67). The synthetic inhibitors TPCK and TLCK, specific for chymotrypsin-like and trypsin- like esterases, respectively, irreversibly inactivate the Chemotactic response to C3a and C5a (2,25). These inhibi- tors also block the ability of neutrophils to hydrolyze Chemotactic peptides, while inhibiting neutrophil directed migration to CSa, BCF, and fMP. The Chemotactic response to these factors is competitively inhibited by the model substrate for chymotrypsin activity, BTEE (3 ). Since correlation coefficients are high between chemotaxis and lysosomal secretion when cells are stimulated by N-formylated synthetic peptides (13), perhaps these cell functions demon- strate different or similar inhibitory responses to more specific proteinase inhibitors. The purpose of this study is to discern the effect of certain proteinase inhibitors on the lysosomal enzyme release of CB-treated human neutrophils stimulated with ZAP and fMP. This may further delineate the characteristics of the serine esterase involvement with this neutrophil process. LITERATURE REVIEW Neutrophil Lysosomes The normal polymorphonuclear leukocyte (PMN) con- tains two granule types based upon staining with Wright- Giemsa stain which is used routinely for making peripheral blood smears. During granulopoietic maturation, azurophilic (primary) granules are formed first at the promyelocyte stage while the specific (secondary) granules are not found until the more mature myelocyte stage. Two-thirds of the granules of the mature PMN are specific with the remainder being azurophilic. Cytochemical studies using electron microsc0py show that the primary lysosomes contain peroxidase and acid hydrolases such as acid phosphatase and beta- glucuronidase. The secondary lysosomes lack acid hydrolases and peroxidase; however, they do contain alkaline phosphatase, some neutral proteases, and bactericidal substances such as lysozyme (muramidase). One-third of this enzyme's total lysosomal content is also found in the primary granule (5,6). The degranulation of lysosomal enzymes is a secre- tory process that occurs in a sequential manner with secondary granule secretibn preceding primary (7,32). This may be related to the fact that secondary granules are particularly accessible for extracellular release (45). Morphological 4 studies using a discontinuous sucrose gradient demonstrate a heterogeneity of lysosome populations in rabbit and human PMNs beyond the specific and azurophilic staining of granules. Small, low-density granules which are speci- fic contain alkaline phosphatase and most of the lysozyme activity. The large high-density granules which are azuro- philic contain peroxidase, the remainder of the lysozyme activity, and most of the acid hydrolases. A morphologi- cally heterogeneous fraction contains the remainder of the acid hydrolases, but no myeloperoxidase (5,71). A discontinuous sucrose gradient of human neutrophil lysates separates the granules into two high-density azurophilic bands while the specific lysosomes collate into a band of lesser density (74). All of these studies indicate there are two populations of azurOphilic lysosomes and one popu- lation of specific granules. Besides the digestive and bactericidal prOperties of lysosomal enzymes, they are also capable of interacting with components of the complement system. Incubation of lysosomal enzymes from human PMNs with C1, C3, or C5 at a neutral pH has shown alterations usually associated with the sequential immunologic activa- tion of the complement system (61). Cl esterase can be activated (61) and then subsequently inactivated (62) with incubation. C3 is cleaved into large and small fragments and more recently it has been shown that human leukocyte elastase cleaves C3 into fragments which can be partially characterized (63). Within the neutrOphil lysosomes are 5 neutral proteases which both activate C5 (68) and inacti- vate CSa, the Chemotactic fragment of C5 (73). Both of these enzymes are found in separate granules and subject to sequential release. The C5a generation activity is found in the specific granules while CSa inactivation capability is contained in the azurophilic granules (74). These studies suggest an interplay between lysosomal enzymes and complement components in the generation and exacerbation of the inflammatory response. Selective Degranulation The secretion of lysosomal enzymes from neutrophilic granulocytes involves migration of the granules to the localized area of contact with the foreign substance or surface. Fusion of the lysosomal membrane with the plasma membrane occurs and the enzymes are released into a pre- formed vacuole in actual phagocytosis or to the cell exterior in "frustrated" phagocytosis, i.e., no ingestion (69). This degranulation with leakage or release of enzymes to the cell exterior causes tissue damage (19,22,34). The most injurious constituents are proteases, collagenase, elastase, and permeability factors--all capable of digesting and injuring tissues. Peripheral blood leukocytes contain IgG receptors as well as trypsin-sensitive complement receptors for C3 and C5. Neutrophils when exposed to phagocytosable and nonphagocytosable immune complexes selectively release their enzymes in a noncytotoxic, non- cytolytic manner (30,32,69). In human PMNs, the selective 6 release caused by complex uptake resembles that induced by zymosan particle ingestion; both particle and complex uptake are accompanied by an increase in hexose mono- phosphate shunt activity (69). Aggregated human myeloma proteins of all IgG and IgA subclasses and normal IgG react with human neutrOphils in a serum-free medium causing enzyme release. The other immunoglobulin classes (IgD, IgE, IgM) do not stimulate degranulation. The insoluble aggregates adhere to the neutrOphils, are ingested, and concomitantly enzymes leak to the cell exterior prior to vacuole closure. Soluble aggregates do not induce secre- tion when in suspension with neutrOphils but do effect secretion when bound to a nonphagocytosable surface (35,36). Under the appropriate conditions substances which are Chemotactic for neutrOphils may also induce enzyme release. The complement-derived factors include the active fragments of the third and fifth components of complement, C3a and C5a (10,11), and serum or plasma activated through the complement alternate pathway by zymosan (27). C3a, CSa, and a bacterial Chemotactic factor (BCF) are capable of activating Proesterase I in rabbit neutroPhils. Activation of this proenzyme is believed to be an obligatory step in the Chemotactic activity induced by these substances (8). The surface charge of human granulocytes is diminished when incubated with CSa. This suggests that a decrease in cell surface charge is a prerequisite for normal cell movement (24). Synthetic peptides formylated at the N-termini are 7 strong attractants for neutrOphils and macrophages (54). These di- and tripeptides can also induce enzyme release and their potency is greatest when the hydrophobic amino acid phenylalanine is positioned at the C-terminus (55). A statistical and linear correlation exists between enzyme release and the chemotactic response evoked by these pep- tides. A higher peptide concentration is needed to induce enzyme secretion than the corresponding chemotactic response (13,55). Specific receptor sites for chemotactic peptides have been demonstrated on rabbit as well as human neutrOphils and the binding is rapid yet reversible (3,72). Other effects can be observed with neutrOphils upon exposure to chemotactic factors. The rate of ingestion of sheep erythrocytes sensitized with IgG and complement is reduced significantly with CSa, BCF, and the peptide formyl- Methionyl-Leucine (47). An excellent correlation exists between calcium influx and lysosomal enzyme release. In addition, a large and rapid potassium efflux is observed under conditions which give rise to enzyme secretion using the peptide formyl-Methionyl-Leucyl-Phenylalanine (fMLP). This suggests a possible ionic basis for degranulation (48). In rabbit PMNs, fMLP stimulates protein carboxymethylation but not protein synthesis, revealing a high correlation between chemotactic responsiveness and specific carboxymethylation (50). 8 Cytochalasin B (CB) This biologically active mold metabolite is obtained from the fungus Helminthosporium dematoideum. The literal translation of cytochalasin is "cell relaxation" and this was the original description of its characteristic effect and not its mode of action (1,17). Cytochalasin B has a unique and novel macrolide structure in which a lactone ring is joined to a bicyclic lactam system. Cultured mammalian cells that have been exposed to CB show an increased rate of attachment to glass, a reversible inhi- bition of cell locomotion, and total inhibition of cyto- plasmic cleavage, without interfering with nuclear division during mitosis. Although multinucleated, these cells remain viable for days in the presence of cytochalasin B (17). Human neutrophils and rabbit macrOphages Show a disruption and depletion of microfilaments when treated with cytochalasin B (46). The influence of CB on neutrOphil functions is reversible and diverse. Phagocytosis (l6,21,46,75) and chemotaxis (9,15,75) are either stimulated or inhibited by cytochalasin B depending on the drug's concentration. Rabbit and human PMNs show suppression and eventual inhibition of both particle and bacteria uptake at CB concentrations above 5 ug/ml (16,75). When drOpped below 1 ug/ml, CB stimulates the uptake of both live staphylococci and starch particles in human neutrophils (16). A similar pattern exists with the PMN chemotactic response. At a concentration range of 2 to 4 ug/ml, CB reversibly inhibits the directed migration 9 and locomotion of rabbit and human PMNs (9,75); however, below 1 ug/ml chemotactic stimulation is generated (9). Lysosomal enzyme release is enhanced with CB regardless of the inducing stimulus (ll,20,37,76), including phago- cytosable and nonphagocytosable substances. The enhanced release occurs above a 5 ug/ml concentration of CB with a maximal effect at 10 ug/ml (37,76). Another study has shown that this drug delays and decreases primary granule release stimulated by zymosan particles at 10 ug/ml (57). In any case, CB at its highest concentration does not affect the total activities of assayed lysosomal enzymes even for incubation periods of 72 hours (20). Electron micrograph examination of CB-treated human neutrophils shows nuclear and cytOplasmic spreading with a linear arrangement of the granules. One hypothesis for the enhancing effect that CB has on enzyme release is the removal of normal constraints for the merger of granules with each other or the plasma membrane (76). Besides its effects on the neutrOphil functions of phagocytosis, chemotaxis, and lysosomal enzyme release, CB also strongly inhibits sugar uptake (75), oxygen consumption and hexose monophosphate shunt activity (37,52). The fungal metabolite enhances phospholipid metabolism in guinea pig peritoneal leukocytes. In fact, kinetic studies show that the drug's influence on the increased incorporation of inorganic phosphorus into phosphatidic acid and phosphoinositides and the enhanced release of beta-glucuronidase are comparable. 10 The increased synthesis of phosphoinositides in CB-treated cells may interact with cationic protein and facilitate the fusion of lysosomal and plasma membranes (64). Studies have shown that chemotactic factors increase PMN volume; however, the cell volume decreases when the cells are exposed to CB alone or CB with a chemotactic factor. There is no quantitative correlation between this phenomenon and the secretion of lysosomal enzymes (41,42). Microtubules have been implicated in granule movement and lysosomal enzyme release (28). Colchicine disrupts or eliminates microtubules which are demonstrable in mature PMNs (45); it also inhibits enzyme secretion and microtubule assembly in a dose-related fashion (40). Concanavalin-A, a plant lectin and lymphocyte mitogen, induces microtubule assembly and specific granule discharge in human PMNs (39). Thus, the importance of microfilaments and/or microtubules is apparent with regard to lysosomal enzyme release. The use of CB-treated neutrOphils provides a model system for the study of selective degranulation in the absence of phagocytosis. Inhibition of Neutrophil Functions As noted earlier, the inflammatory functions of the neutrophil involve an energy source, calcium and/or magnesium in the external medium, and possible involvement of the con- tractile elements of the cell. Any substance which may inhibit or block any of these parameters may obstruct any or all of the neutrOphil functions concerned with inflammation. ll Nonphagocytic lysosomal enzyme release is inhibited by cAMP and theophylline but the effect in combination is no greater than additive. The anti-inflammatory agents hydrocortisone and colchicine are also inhibitory (31,77). Of particular interest is the effect that phosphonate esters have on PMN functions, especially the inhibitor diisopropylflurophosphate (DFP). This irreversible inactivator functions through phosphorylation of the serine hydroxyl group within the active site of the affected enzyme. The charge relay between the serine, aspartic acid and histidine residues within the active site is blocked, thereby rendering the serine esterase nonfunc- tional. In studies done with rabbit neutrophils, DFP was shown to inhibit chemotaxis whether the cells were pre- treated or the inhibitor was present with the chemotactic factor. The same inhibitory effect seen under different conditions suggests thatth>esterases--one activated and the other activatable--are in or on the cell and are involved in the chemotactic response. This is reinforced by the difference in inhibition profiles exhibited by either cell-dependent or chemotactic factor-dependent inactivation (12,67). One approach toward observing phago- cytosis is treatment of guinea pig peritoneal cells with EACl423 using sheep erythrocytes and guinea pig serum as a complement source. This complement-dependent erythrOphago- cytosis is inhibited by DFP. The effect is irreversible, temperature-dependent, and prOportional to the inhibitor 12 concentration and the duration of exposure to the inhibitor. Cell pretreatment with DFP gives the greatest effect, suggesting that a critical proteinase or esterase is in or on the cell in the "activated" state and a second activatable enzyme is also required for the phagocytic process (51). Lysosomal enzyme release from rabbit neutrophils by BCF is inhibited with DFP, but only when the inhibitor and chemotactic factor are together and not when the cells are treated before release (11). Human neutrophil degranu- lation using zymosan particles (phagocytic stimulus) or aggregated IgG bound to a surface (nonphagocytic stimulus) is inhibited by DFP. Some effect is seen when the cells are preincubated with the inhibitor prior to release; however, the strongest result occurs with the stimulus and inhibitor together (38). The observations with DFP show that serine esterase activity is a critical part of chemotaxis, phagocytosis, and enzyme secretion. This has led to studies with more specific, irreversible serine esterase inhibitors such as N-tosyl-L-lysyl-chloromethyl- ketone (TLCK), N-tosylamide-phenylethyl-chloromethy1-ketone (TPCK), phenyl-methyl-sulfonyl-f1uoride (PMSF), Cl inhibitor (CIINH), and alpha-l-antitrypsin. CllNH is a biological broad-spectrum inactivator exhibiting control over the coagulation, fibrinolysis, complement, and kallikrein blood systems and it weakly inhibits the pancreatic enzymes trypsin and chymotrypsin. Alpha-l-antitrypsin is also a biological proteinase inhibitor that blocks the fibrinolytic and 13 kallikrein systems, but it strongly inactivates both trypsin and chymotrypsin (33). The synthetic serine esterase inhibitors TPCK and TLCK function by alkylation of the histidine residue within the active site of the enzyme. Functioning as pseudosubstrates, TLCK inactivates chymotrypsin- like proteinases while TPCK affects trypsin-like proteinases. The synthetic inhibitor PMSF functions like DFP; however, it sulfonylates the serine hydroxyl in the enzyme's active site. In contrast to the organophosphates, sulfonyl halides show great differences in reactivity depending on the enzyme and the structure of the sulfonyl halide (26). Fibroblasts that have been treated with Simian Virus 40 (SV-40) lose their ability to exhibit contact inhibition and consequently grow proliferatively in tissue cultures. Treatment of these transformed cells with TPCK inhibits their growth; an identical "growth plateau" effect is seen also with cyclo- heximide, an inhibitor of de novo protein synthesis in eukaryotes. This suggests that the growth inhibitory effect of TPCK on transformed cell growth involves the inhi- bition of protein synthesis instead of proteolytic inacti- vation (18). TLCK, TPCK, and PMSF inhibit the attachment of Baby Hamster Kidney (BHK) cells to a substratum only in the presence of serum; that is, significant inhibition occurs when the substratum is coated with serum. The effect is achieved by preincubating the cells with the inhibitor at room temperature; after washing the cells are resuspended in fresh attachment medium. Under these conditions, the l4 findings demonstrate that the cellular function of attach- ment also requires cell-related esterase activity (29). Human ClINH reversibly enhances human neutrophil chemo- tactic response to activated plasma or serum without affect- ing spontaneous motility (59). This same proteinase inactivator also shows chemotactic inhibition to trypsin- activated CS while significantly enhancing the response to N-formyl-methionyl-phenylalanine (44). These effects with ClINH occur only when the inhibitor is present with the cells in the upper portion of the Boyden chamber. Preincubation of human neutrophils with TLCK, TPCK, and alpha-l-antitrypsin inhibits their chemotactic responsive- ness to C3a and C53. TPCK inhibits random migration while TLCK and alpha-l-antitrypsin enhance it (25). The poten- cies of fMet peptides as chemotactic agents is related to the rate at which they are hydrolyzed. TPCK and TLCK inhibit chemotaxis as do the hydrolysis products of fMet peptides. NeutrOphils pretreated with TPCK and TLCK inhibit peptide hydrolysis and the chemotactic response to CSa, BCF, and fMP. In addition, benzoyl tyrosine ethyl ester (BTEE), a model substrate specific for chymotrypsin activity, inhibits the chemotactic response to all three attractants. Tosyl arginine methyl ester (TAME), which is used to identify trypsin-like activity, has little or no effect on chemotaxis (2). Rabbit neutrophils have chymotrypsin- like esterase activity in their cytosol and lysosomes. The esterase is not inhibited by CAMP, nor is it stimulated by 15 BCF or C5a. High chemotactic activity is found in the partially purified fraction of the enzyme. Both chemo— tactic and enzymatic activities are inhibited by a phosphonate ester (65). Unlike rabbit PMNs, essentially all the chymotrypsin-like esterase activity of human neutrophils is in the lysosomal fraction, but more specifically the majority is in the primary granule. Human esterase activity has a sharper Optimum pH range and the specific activity may be as high as 40—fold that observed in rabbit (66). Since the neutrophil functions of chemotaxis, phagocytosis, and lysosomal enzyme release require one or more surface esterases, a suggestive activation scheme can be hypothe- sized. Interaction of a chemotactic factor with a specific receptor could activate the surface esterase(s) leading to interdependent events causing chemotaxis or enzyme release (12,13). Included in these interrelated events could be a transient increase in membrane permeability to sodium and calcium, membrane depolarization and the involvement of microfilaments and/or microtubules (48). MATERIALS AND METHODS Reagents and Solutions Sodium Hypaque (50%) was obtained from Winthrop Laboratories, New York, NY 10016. Triton X-100 was pur- chased from Research Products International Corporation, Elk Grove Village, IL 60007; Goat anti-human C5, Meloy Laboratories, Alexandria, VA; Protein A-Sepharose, Pharmacia Fine Chemicals, Piscataway, NJ. The remaining materials were procured from Sigma Chemical Company, St. Louis, MO 63178. These included bovine serum albumin (BSA), human serum albumin (HSA), benzoyl-tyrosine ethyl ester (BTEE), Cytochalasin B (CB), cycloheximide, Dextran, dimethyl- sulfoxide (DMSO), Epsilon Amino Caproic Acid (EACA), Ficoll (MW 400,000), Lysozyme Egg White Standard, Micrococcus Zysodeikticus, N-formylmethionylphenylalanine (fMP), B- nicotinamide adenine dinucleotide (B-NADH), phenolphthalein beta-glucuronic acid, phenylmethylsulfonylfluoride (PMSF), tosyl-arginine methyl ester (TAME), L-tosylamide-phenyl- ethylchloromethyl ketone (TPCK), N-tosyl-L-lysine chloromethyl ketone (TLCK), and zymosan A. A stock solution of Cytochalasin B (5 mg/ml in DMSO) was kept at 4°C. 'The stock solution was diluted with Hank's Balanced Salt Solution, pH 7.4 (HBSS) to a concentration of 16 17 50 ug/ml. Aliquots of 10'3M fMP in HBSS and 10 mg/ml BSA in HBSS were stored at -20°C until used. TLCK and cyclo- heximide readily dissolved in HBSS to the desired concen- trations. TPCK and PMSF did not go into aqueous solution 3 below 10' M. They were therefore dissolved in anhydrous methanol at 0.1M and then diluted with HBSS to concentra- tions of 10'3 M or less. Ficoll—Hypaque Solution was made by combining 24 parts of 9% Ficoll with 10 parts of 33.9% Hypaque. Human CS was isolated according to Nilsson (49) without modification. NeutroPhil Isolation Heparinized whole blood was obtained from healthy human volunteers. Nine parts of blood was mixed with 1 part of 6% Dextran (in HBSS) and allowed to settle for 1 hour. Four milliliters of the cell-rich plasma was layered over 4 ml of Ficoll-Hypaque Solution in siliconized 16 x 100 mm glass tubes. After centrifugation at 800 x g for 20 minutes, the polymorphonuclear cells pelleted to the bottom along with some erythrocytes. The supernatant which contained mononuclear cells and platelets was discarded. The cell pellet was suspended in HBSS, transferred to a clean siliconized tube, and washed once. Cell counts were accomplished on a hematocytometer with the white cell concen- tration being adjusted to 20 x 106 cells/m1. These prepara- tions contained approximately 95% neutrophils with an estimated red to white cell ratio of 5:1 (58). Aliquots (0.1 ml) of the cell suspension were placed in 10 x 75 mm 18 glass tubes. The final reaction volume and cell concentra- tion throughout experimentation was 2 x 106 cells/ml/tube. Enzyme Assays Lactic Dehydrogenase (LDH), the cytoplasmic marker for cell viability, was determined by the method of Bergemeyer et a1. (14). This bisubstrate reaction measured the change in absorbance at 366 nm due to the oxidation of NADH. The absorbance decrease of the 3-ml reaction mixture was observed over a 2-minute interval. Enzyme activity was tabulated and recorded as units/m1. Lysozyme, the secondary granule marker, was measured according to the Shugar method as described in the Sigma Bulletin for Lysozyme Egg White Standard (56). This assay also involved a change in absorbance over a fixed time period. One-tenth milliliter of the unknown was added to a 2.5 m1 suspension of Micrococcus Zysodeikticus and the decrease in absorbance at 450 nm after 2 minutes was recorded. The enzyme's activity was expressed as units/m1. The primary granule indicator, B-Glucuronidase, was assayed by the Fishman method (23). A sample of the unknown (0.1 ml) was added to 0.1 ml of phenolphthalein glucuronic acid and 0.8 ml of acetate buffer making the final reaction volume and pH 1.0 ml and 4.5, respectively. After 17 hours of incubation at 37°C, the solution was made alkaline with a glycine buffer (pH 11.2),trichloroacetic acid and water. The phenolphthalein released due to sub- strate hydrolysis caused the development of a pink color 19 whose intensity was dependent on the enzyme concentration. The solution's absorbance was measured at 540 nm and compared to 25 ug/ml and 50 ug/ml standards made by diluting in HBSS aliquots from a stock phenolphthalein solution of 1 mg/ml in ethanol. BG values were reported as ug phenolphthalein/m1. Enzyme Secretion Protocol Stimulated release of primary and secondary neutrophil granules was effected with ZAP and fMP in the presence of 5 ug/ml cytochalasin B (13,27). A suspension of zymosan A (10 mg/ml) was made in phosphate-buffered saline, pH 7.4, ionic strength 0.15, containing 0.15M magnesium chloride. .One-tenth milliliter of the zymosan suspension was added to 0.9 ml autologous plasma containing .ZSM EACA. In our hands EACA allowed substantially greater enzyme release than plasma alone when utilizing ZAP, as previously shown by Goldstein et a1. (27). After incubation at 37°C for 30 minutes the zymosan particles were removed by centrifugation at 1000 x g for 5 minutes. One-tenth milliliter of ZAP and 0.1 m1 CB (50 ug/ml) were added to 0.8 ml HBSS containing 2 x 106 cells. After mixing, the suspension was incubated at 37°C for 30 minutes. Supernatants were removed for enzyme assays after centrifugation at 1000 x g for 1 minute. Noncytotoxic enzyme release occurred with ZAP alone and CB alone; however, together their effect was signifi- cantly more than additive. A ZAP dilution control (1:10 in 20 HBSS) in the absence of neutrOphils was assessed for enzyme levels in order to negate any effects due to plasma alone. 6 cells When utilizing the formylated peptide, 2 x 10 in 0.7 ml HBSS were treated with 0.1 ml BSA (10 mg/ml) and held at 4°C until ready for use. One-tenth milliliter CB (50 ug/ml) and 0.1 m1 fMP (10'3M) were then added and mixed. After incubating at 37°C for 5 minutes the super- natants were removed as before and enzyme determinations performed. Relative secreted enzyme levels were calculated as the percent of total enzyme content of the cells. Total enzyme content was ascertained by treatment of cells with 0.2% Triton X-100. Baseline or unstimulated enzyme levels were assessed on cells incubated in HBSS. Inhibitor Treatment of Cells Two approaches were taken to evaluate the effect of inhibitors on enzyme release. The first involved incuba- tion of inhibitor with cells in the presence of the releasing stimulus. In this instance the inhibitor was added to cells, followed immediately by addition of the releasing agent. After an appropriate incubation period (30 minutes for ZAP and 5 minutes for fMP), the supernatants were analyzed for enzyme content. The second protocol utilized preincubation of the cells with inhibitor and a subsequent single washing in HBSS before exposure to ZAP or fMP. The inactivators or inhibitors were added to the PMNs in HBSS, mixed, incubated at 37°C for 5 minutes and centrifuged 21 (30 seconds at 1000 x g). The supernatants were carefully removed and discarded. The cell pellets were washed once in HBSS and subjected to ZAP or fMP as already described. Control experiments indicated none of the inhibitors elicited enzyme release alone, nor did they inactivate the marker enzymes. Experiments using TPCK or PMSF required that initial solubilization be effected in anhydrous methanol. Stock solutions of each agent at 0.1M were used for subsequent dilutions in HBSS. Appropriate methanol dilutions did not affect the enzyme assays nor alter stimulated or resting enzyme release. Statistical Procedure for Data Presentation Enzyme release values are presented as % 1 Standard Error of the Mean (SEM); 100% designates the total enzyme content determined by treatment with 0.2% Triton X-100. Statistical significance was determined by a paired Student's t-test with n signifying the number of degrees of freedom and p the level of significance. Calculated p values of less than .05 were considered to be statistically significant. Preparation of CS-Deficient Plasma Goat antiserum to human CS was prepared as follows. Twelve precipitin bands were excised from replicate agar immunoelectrophoresis gels which were developed with puri- fied C5 and commercial anti-C5 in the presence of 0.01M EDTA. The precipitin bands were allowed to stand for 22 several days at 4°C in several changes of physiologic saline containing 0.02% sodium azide. The washed opaque immune precipitates were emulsified in Freund's Complete Adjuvant, and injected subcutaneously into multiple sites in an individual goat. Two weeks later the injection pro- cedure was repeated and 6 weeks after the initial injec- tion the antiserum was harvested, heat inactivated, made 0.02% with sodium azide, and frozen at -20°C or -80°C. In immunoelectrOphoresis (53) and Ochterlony analyses the antiserum was found to possess strong anti-C5, anti- HSA, and anti-IgG activities. If immunodiffusion patterns were allowed to deve10p for more than 2 days with NHS, an additional faint band was seen that possessed slow beta motility and a slow rate of diffusion. Anti-HSA activity was removed with stoichiometric amounts of HSA and anti- IgG activity was removed by absorption with IgG-Sepharose 6B. The antiserum was rendered monospecific for C5 when observing precipitin bands within 24 hours at room tempera- ture. Moreover, this anti-C5 did not demonstrate reactivity with serum genetically deficient in C5, kindly given to us by Dr. John Leddy. The IgG fraction of the anti-C5 was isolated by caprylic acid precipitation according to the method of Steinbach and Audran (60). After adjusting the protein content to ~2%, insolubilization of anti-C5 activity was effected by treatment with Protein-A Sepharose. Packed Protein A-anti C5 contained approximately 20 mg protein/ml resin. 23 Fifteen milliliters of normal human plasma was treated by stirring for 45 minutes at room temperature with 5 ml of packed resin in the presence of 0.01M EDTA. A second treatment for 90 minutes at 4°C and 30 minutes at room temperature rendered the plasma deficient in C5 as ascertained by immunoelectrOphoresis with our anti—C5 as well as with commercial anti-C5. Rocket electrophoresis (4) was also done on the commercial anti-C5. The unidenti- fied precipitating activity described above was still present in the absorbed CS-deficient plasma. The CS-deficient plasma was dialyzed against HBSS and frozen at -80°C until ready for use. An untreated source of plasma was treated in an identical fashion with regard to incubations, dialysis and freezing. This plasma subsequently served as a positive stimulus control when activated with zymosan. RESULTS Inhibition of Enzyme Release with the Trypsin-LikeTInHibitor TECK When PMNs were treated with TLCK prior to stimulated enzyme release with ZAP and fMP, significant prevention was observed. As shown in Figure 1, both lysozyme and BG release were significantly inhibited. When ZAP was used as the secretory stimulus, BG release was significantly 3 4 depressed at 10' and 10- M TLCK and lysozyme release was 5 inhibited at 10- M as well. Similar results were obtained when the formylated peptide was used to stimulate granule secretion. In this instance significant inhibition was 3 5 observed for both granule markers from 10' to 10- M TLCK. Table 1 shows the results of experiments designed to detect inhibition when inhibitor was present simultaneously 3 with either ZAP or fMP. Significant inhibition at 10- M of both lysozyme and BC was observed when using ZAP. Similar results were obtained with fMP-stimulated cells. Signifi- 3 cant inhibition of both enzymes was affected at 10- M TLCK 4M TLCK as well. and BG release was inhibited at 10' The effect of TLCK on enzyme release is apparently non- cytotoxic since LDH levels remained low throughout these experiments. 24 25 Figure l. TLCK pretreatment. PMNs (2 x 106/ml) were preincubated with the inhibitor for 5 minutes at 37°C, washed once, then treated with (A) 10% ZAP + 5 ug/ml CB with incubation at 37°C for 30 minutes or (B) 10-4M fMP + 5 pg/ml CB + 1 mg/ml BSA with incubation at 37°C for 5 minutes. Results presented as % + SEM of total enzyme content determined by treatment of 2 x 106 cells/ml with 0.2% Triton X-100 (B-Glucuronidase = 128 i 10 ug phenolphthalein/m1, Lysozyme = 219 i 13 units/m1, LDH = 388 + 13 units/m1). Graphic presentation of enzyme leveIs shown as solid bars (BG), hashed bars (LYS), and Open bars (LDH) with tolerances less than 0.5 not dis- played. Negative controls for all 3 enzymes with both stimuli were not greater than 5.4 + .40%. Student's t-test p values for BC and LYS compared to tubes receiv- ing no pretreatment, but subjected to stimuli. LDH p values (n=4), although not significant, were compared to the negative control. In part A, significant inhibi- tion was seen at 10'3M and 10'4M TLCK for both BG and LYS. At both concentrations, p<.05 for BC (n=4); for LYS (n=6) p<.OOS at 10'3M and p<.OZ at 10'4M TLCK. In part B significant inhibition was exhibited from 10-3M to 10'§M TLCK. At the hi her 2 concentrations, p<.005 for both enzymes. At 10‘ M TLCK, p<.05 for BC (n=4) and p<.02 for LYS (n=5). 26 6M a 7// . m .— Mm 7////////////.////////////.M.v i IM /////////////,w ”M a. m IE 2 f w m w m m w m w m m mm