Human granulocyte generation of hydroxyl radical

Activity of phagocytic granulocytes in patients with prostatic cancer

Chemilumenescence (CL) occurs due to the phagocytosis of bacteria and of tumor cells by polymorphonuclear neutrophils (PMN). Levels of CL were measured in patients with prostatic cancer and from normal subjects. Patients with advanced disease (stage C, D) showed no elevated CL levels as compared to healthy individuals or patients with minimal disease (stage A, B). Following external radiation therapy in patients with stage A-C prostatic carcinoma high levels of CL were recorded. Estrogen medication also resulted in increased CL levels, while estramustine did not affect phagocytic activity. Intradermal BCG vaccination caused increased PMN activity. Progressive prostatic cancer in hormone treated patients was associated with increased CL as compared to patients with stable or regressive disease.

Interference of target cell catalase with an early step of the neutrophil cytolytic pathway

The hypochlorous acid (HOCL)-dependent lysis of human red blood cells (HRBC) targets by neutrophils, activated with opsonized zymosan particles (OPZ), was increased by inhibiting HRBC catalatic activity with aminotriazole (AT; HRBCAT). The inhibition of HRBC glutathione cycle activity with carmustine (BCNU; HRBCBCNU) had no effect. In addition, the recovery of hydrogen peroxide (H2O2) and HOCL from neutrophils, activated under conditions similar to those used for cytotoxicity assay, was reduced by the presence of HRBC and restored by replacing HRBC with HRBCAT, but not with HRBCBCNU. Linear relationships were found between the increments in the neutrophil-mediated lysis, observed by using HRBCAT instead of HRBC, and the increments in the H2O2 or HOCL recovery, detected by replacing HRBC with HRBCAT. Together these data, coupled with the results obtained by probing neutrophil cytolysis with chemical agents, suggest that the increased cytolytic efficiency displayed by neutrophils against HRBCAT, inhibited in their catalatic activity, is due to an enhanced availability of neutrophil-derived H2O2, with a consequent enhancement in the HOCL production (according to the following reaction: (formula; see text). Thus it appears that HRBC catalase restrains the neutrophil cytolytic activity, by interfering with an early step of the pathway through which neutrophils generate cytotoxins.

Etiology of the pulmonary pathophysiology associated with inhalation injury

This study describes an experimental model of smoke inhalation injury in sheep in which the same pathophysiologic alterations occur as with clinical inhalation injury in man. Diffuse pulmonary mucosal sloughing with atelectasis and emphysema with concomitant development of pulmonary edema results in a decrease in arterial oxygen and progressive pulmonary deterioration which results in a substantial mortality. Increased pulmonary edema fluid is shown to be caused by an increased microvascular permeability to protein with pulmonary lymph and tracheobronchial fluid, a filtrate of plasma. Concomitant with this increase in microvascular permeability is an influx of neutrophils, release of proteolytic enzymes and an identified presence of the metabolite of the prostanoid thromboxane A2 which are postulated as contributors to the progressive pulmonary dysfunction post inhalation injury.

Myeloperoxidase as an effective inhibitor of hydroxyl radical production. Implications for the oxidative reactions of neutrophils

Hydroxyl radicals have been generated from hydrogen peroxide and superoxide (produced with xanthine oxidase), and an iron (EDTA) catalyst, and detected with deoxyribose, or in some cases with benzoate or alpha-keto-gamma-methiolbutyric acid. Purified myeloperoxidase, and neutrophils stimulated with fMet-Leu-Phe and cytochalasin B, strongly inhibited this hydroxyl radical production in a concentration-dependent manner. Supernatants from stimulated cells also inhibited, and inhibition by cells or supernatant was prevented by azide. There was much less inhibition by myeloperoxidase-deficient neutrophils. Inhibition thus was due to myeloperoxidase released by the cells. With neutrophils stimulated with phorbol myristate acetate, which release very little myeloperoxidase, hydroxyl radical production was enhanced due to the additional superoxide produced by the cells. It is concluded that under conditions where neutrophils release myeloperoxidase as well as superoxide and hydrogen peroxide, breakdown of hydrogen peroxide by myeloperoxidase would make conditions unfavorable for hydroxyl radical production.

Modulation of phagocytic cell function

Phagocytosis is an important factor in the defense of the host against all kinds of microorganisms. The process of phagocytosis of microorganisms by phagocytes can be separated into distinct but interrelated phases: adherence, chemotaxis, opsonization, attachment, ingestion, degranulation and killing. Phagocytosis is accompanied by an increase in oxygen metabolism in which H2O2 and activated oxygen species are generated. Modulation of phagocytic cell function can be brought about by a variety of substances. Microorganisms produce and contain components which influence the process of phagocytosis. Surrounding tissue cells and the phagocytes themselves produce biologically active molecules that modulate phagocytosis.

Self-perpetuating mechanisms of immunoglobulin G aggregation in rheumatoid inflammation

When human IgG is exposed to free radical generating systems such as ultraviolet irradiation, peroxidizing lipids, or activated human neutrophils, characteristic auto-fluorescent monomeric and polymeric IgG is formed (excitation [Ex], 360 nm, emission [Em], 454 nm). 1 h ultraviolet irradiation of IgG results in the following reductions in constituent amino acids; cysteine (37.0%), tryptophan (17.0%), tyrosine (10.5%), and lysine (3.6%). The fluorescent IgG complexes, when produced in vitro, can stimulate the release of superoxide from normal human neutrophils. In the presence of excess unaltered IgG, further fluorescent damage to IgG occurs. Measurement and isolation of fluorescent monomeric and polymeric IgG by high performance liquid chromatography, from in vitro systems and from fresh rheumatoid sera and synovial fluid, indicates that identical complexes are present in vivo; all these fluorescent complexes share the property of enhancing free radical production from neutrophils. The results described in this study support the hypothesis that fluorescent monomeric and aggregated IgG may be formed in vivo by oxygen-centered free radicals derived from neutrophils, and that in rheumatoid inflammation this reaction may be self-perpetuating within the inflamed joint.

Myeloperoxidase oxidation of sulfur-centered and benzoic acid hydroxyl radical scavengers

Myeloperoxidase (MPO) oxidizes sulfur-centered and benzoate hydroxyl radical scavengers through formation of HOCl. Sulfur-centered hydroxyl radical scavengers compete with benzoate as antioxidants of HOCl. We conclude from these observations that competition experiments between benzoate and sulfur-centered hydroxyl radical scavengers are not sufficiently specific to infer participation of hydroxyl radicals in oxidative reactions mediated by neutrophils because of the unique action of MPO in affecting oxidation of the test radical scavengers.

Effect of interferon on chemiluminescence and hydroxyl radical production in murine macrophages stimulated by PMA

Considerably augmented chemiluminescence (CL) occurred when murine peritoneal resident macrophages (MPs), pretreated with murine interferon (MuIFN)-alpha within 24 hr, were stimulated by 4-beta-phorbol, 12-beta-myristate, 13-beta-acetate (PMA). Augmentation of CL generation ceased when incubation in the presence of MuIFN was continued for 48 hr. As 12 hr preincubation with MuIFN procured optimal CL generation, the various reactive oxygen species (OH, O2-., H2O2) were measured at that point. The hydroxyl radical (OH.) level in MuIFN-treated MPs was 19.44 times higher than in MuIFN-untreated MPs. However, the levels of O2-. and H2O2 generation were the same in both MuIFN-treated and untreated MPs. Moreover, by using the inhibitors lipoxygenase and cyclo-oxygenase, we established clearly that CL and OH. generation in MuIFN-treated MPs is due to the lipoxygenase pathway of arachidonic acid metabolism.

In vivo production of ethylene from 2-keto-4-methylthiobutyrate in mice

The use of 2-keto-4-methylthiobutyric acid (KMB), the alpha-keto analog of methionine, was studied as a potential means of detecting free radical generation in vivo. KMB-dependent ethylene production (presumably from free radical interception), and ethane production from in vivo lipid peroxidation, were monitored simultaneously by measuring the rate of exhalation of these hydrocarbons by mice. Injection of KMB (1 g/kg) into mice resulted in an 8-fold increase in ethylene production above endogenous levels seen in saline-injected controls (1.47 +/- 0.35 vs 0.17 +/- 0.02 nmoles/100 g/hr respectively). Administration of CCl4 (3.0 g/kg) to initiate hepatic lipid peroxidation, 20 min prior to KMB injection, augmented the production of ethylene (2.37 +/- 0.10 nmoles/100 g/hr). Lipid peroxidation following injection of CCl4 was monitored via the increased exhalation of ethane. Pretreating the mice with vitamin E (100 mg/kg daily for 3 days), an inhibitor of lipid peroxidation, did not result in a significant change in ethylene production from KMB by itself or after prior injection of CCl4. However, vitamin E did suppress ethane production initiated by CCl4. Similar results were obtained with mouse liver slices studied in vitro. Metyrapone (150 mg/kg), an inhibitor of hepatic mixed function oxidase activity, also suppressed significantly the CCl4-stimulated production of ethane, but not the CCl4-stimulated production of ethylene from KMB. It appears that ethylene production from KMB does not derive from free radicals generated during in vivo lipid peroxidation since suppression of lipid peroxidation by vitamin E or metyrapone did not suppress ethylene production.

Sulfonate salts of amino acids: novel inhibitors of the serine proteinases

A series of amino acid-derived sulfonate salts have been synthesized. They were found to inactivate efficiently and selectively human leukocyte elastase. The sulfonate salts of the methyl esters of L-norleucine, L-norvaline and L-valine were the most potent. The enzyme is inactivated irreversibly with concomitant release of bisulfite ion. The results demonstrate for the first time that ionic compounds can indeed function as novel inhibitors for the serine proteinases.

Modulation of pulmonary clearance of bacteria by antioxidants

To further delineate the mechanisms underlying murine pulmonary defenses against bacterial infection, we studied the effects of antioxidant enzymes and hydroxyl radical scavengers on pulmonary clearance processes. Intratracheal injection of catalase and superoxide dismutase resulted in prolonged intraalveolar residence of the enzymes, but caused no decrease in rates of clearance of either Staphylococcus aureus 502A or Pseudomonas aeruginosa PAO1. In contrast, dimethylsulfoxide and dimethylthiourea caused significant depression of clearance of P. aeruginosa without altering clearance of S. aureus. These results provide further differentiation between clearance processes affecting gram-negative and gram-positive bacteria and suggest that murine clearance of gram-negative organisms may be in part mediated by reactions which generate hydroxyl anion. In vivo administration of agents which inhibit hydrogen peroxide-, superoxide-, or hydroxyl anion-mediated reactions do not alter normal clearance of S. aureus.

Human neutrophil elastase modulates platelet function by limited proteolysis of membrane glycoproteins

During blood coagulation human polymorphonuclear leukocytes release elastase in amounts that can exceed 100 nmol/liter. We therefore studied the effect of elastase on platelet structure and function. Physiologic concentrations of elastase specifically inhibited thrombin-induced platelet aggregation and ristocetin-induced agglutination of washed platelets in a time- and dose-dependent manner. This was associated with a decrease in the number of high affinity thrombin binding sites on the platelet surface (analysis by "Ligand" program) from 31 per platelet to 12 per platelet (P less than 0.05). As analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, treatment of 3H-labeled platelets with elastase resulted in a decrease in the percent glycoprotein at 130,000-150,000 Mr = and an increase in the percent protein at Mr = 102,000. The supernatant from elastase-treated platelets contained a Mr = 88,000 glycoprotein not found in the supernatant from untreated platelets. Immunoprecipitation studies with monoclonal antiglycoprotein Ib demonstrated that treatment of whole platelets with physiologic concentrations of elastase resulted in proteolytic cleavage of glycoprotein Ib. Elastase treatment of glycoprotein immunoisolated with monoclonal antiglycoprotein Ib antibody resulted in formation of a glycopeptide with the same electrophoretic mobility as the Mr = 102,000 membrane-related glycopeptide. In contrast, analysis by Western blot technique using antiglycoprotein IIb and IIIa antibodies demonstrated that elastase did not degrade glycoproteins IIb or IIIa. We conclude that elastase inhibition of thrombin-induced platelet stimulation is accompanied by (a) a reduction in the number of thrombin binding sites per platelet and (b) proteolysis of glycoprotein Ib.

Canine myocardial reperfusion injury. Its reduction by the combined administration of superoxide dismutase and catalase

Therapy directed against the toxic effects of reactive oxygen species may reduce the final extent of ischemic injury in otherwise viable tissue irreversibly injured by the abrupt reoxygenation of reperfusion. In four groups of dogs, superoxide dismutase plus catalase (groups I-III) or saline (controls) (group IV) was infused into the left atrium. Group I received the infusion for 2 hours, beginning 15 minutes before occlusion of the left circumflex coronary artery (90 minutes) and ending 15 minutes after reperfusion. Group II received the infusion for 1 hour starting 15 minutes before reperfusion. Group III received the infusion for 1 hour beginning 40 minutes after reperfusion. Dogs were killed the next day, and infarct size was determined by dissection and weighing, and confirmed histologically. Infarct size expressed as percent of the anatomic area at risk was: group I, 19.4 +/- 5.0; group II, 21.8 +/- 3.3; group III, 47.6 +/- 10.3; group IV, 43.6 +/- 3.5 (mean +/- SEM). Analysis of variance followed by Duncan's multiple range test showed that ultimate infarct size as assessed in groups I and II differed significantly (P less than 0.05) from that observed in the control animals in group IV, whereas infarct size between groups III and IV did not differ significantly (P greater than 0.05). The percent of left ventricle at risk did not differ between the four groups. The beneficial effects of superoxide dismutase plus catalase could not be explained by hemodynamic differences. Similar protection of jeopardized myocardium in groups I and II suggest that potentially viable tissue is salvaged by scavenging free radicals during early reperfusion.(ABSTRACT TRUNCATED AT 250 WORDS)

The toxicology of molecular oxygen

Molecular oxygen, itself not very reactive, can be converted by photosensitization to electronically excited singlet states, and by partial reduction to the superoxide and hydroxyl free radicals and to hydrogen peroxide. The very considerable toxicity of oxygen, which is due primarily to the properties of these derivatives, is ordinarily overlooked because aerobes have evolved an elaborate system of defenses which is reasonably adequate under ambient conditions. This toxicity becomes all too apparent when these defenses are overwhelmed at elevated pO2 or through the action of compounds which increase the conversion of oxygen to its more reactive derivatives. We will here consider the threat posed by oxygen and the defenses which make aerobic life possible.

The chemiluminescent response of bovine polymorphonuclear leucocytes isolated from milk and blood

Polymorphonuclear leucocytes (PMN) were isolated from milk and blood of healthy cows, and the generation of reactive oxygen by the two cell populations was compared by measuring chemiluminescence (CL) after stimulation with zymosan. The ratio of milk to blood PMN CL was relatively constant in a given animal, but varied widely between different cows, ranging from 0.3 to 1.3. The relative contributions of various oxygen species to CL was studied by measuring quenching using different oxygen scavengers. While the relative contributions of H202, -02 and '02 seemed to be similar in both milk and blood PMN, the OH. radical was clearly more prominent in PMN isolated from milk than from blood. In addition, blood PMN CL was more dependent on the presence of glucose in the reaction medium than milk PMN CL. Furthermore, the CL response to phorbol myristate acetate, to the Ca ionophore A23187 and to Sendai virus was different in the two cell types. The results suggest that CL generation in milk PMN differs from that in blood PMN in quantitative as well as qualitative aspects.

Effect of cimetidine, hydrocortisone superoxide dismutase and catalase on the development of oedema after thermal injury

A new hypothesis for the pathophysiological mechanism underlying the development of oedema after a thermal injury has been tested in an experimental burn model. Support was given to the suggestion that oxygen-derived free radicals produced by invading leucocytes which upon activation release the superoxide radical (O2-), may be partly responsible for the increase in microvascular permeability seen after thermal injury. By removal of oxygen-derived free radicals with radical scavengers (superoxide dismutase and catalases) it was possible to reduce significantly the post-burn oedema formation. For comparison, one series of rats was pretreated with hydrocortisone and another with the histamine H2-blocker cimetidine. Hydrocortisone reduced the very early post-burn oedema formation which might partly be due to its membrane-stabilizing influence and partly to a direct effect on the microvasculature, causing a reduction of the vasodilatation observed post-burn. The inhibition of post-burn oedema formation by cimetidine, earlier demonstrated in animal burn models, was confirmed in the present study. The mean arterial blood pressure (MAP) in the cimetidine-treated rats decreased, however, after treatment. It is therefore difficult to determine to what extent the concentration of oedema is attributable to histamine H2-receptor blockade and to what extent to the reduced blood pressure. The influence of MAP on post-burn odema formation was further illustrated in two series of rats anaesthetized with Inactin an Hypnorm/Valium respectively. The results underline the importance of using the same anaesthetic throughout the experimental series.

Lactoferrin-catalysed hydroxyl radical production. Additional requirement for a chelating agent

The ability of lactoferrin to catalyse hydroxyl radical production was determined by measuring ethylene production from methional (2-amino-4-methylthiobutyraldehyde) or 4-methylthio-2-oxobutyrate. Lactoferrin, isolated from human milk and saturated by adding the exact equivalents of Fe3+-nitrilotriacetic acid and dialysing, give little if any catalysis of the reaction between H2O2 and either O2-. or ascorbic acid at either pH 7.4 or pH 5.0. However, in the presence of chelating agents such as EDTA or nitrilotriacetic acid that can complex with lactoferrin, hydroxyl radical production by both mechanisms was observed.

Damage to Aspergillus fumigatus and Rhizopus oryzae hyphae by oxidative and nonoxidative microbicidal products of human neutrophils in vitro

Our previous studies established that human neutrophils could damage and probably kill hyphae of Aspergillus fumigatus and Rhizopus oryzae in vitro, primarily by oxygen-dependent mechanisms active at the cell surface. These studies were extended, again quantitating hyphal damage by reduction in uptake of (14)C-labeled uracil or glutamine. Neither A. fumigatus nor R. oryzae hyphae were damaged by neutrophils from patients with chronic granulomatous disease, confirming the importance of oxidative mechanisms in damage to hyphae. In contrast, neutrophils from one patient with hereditary myeloperoxidase deficiency damaged R. oryzae but not A. fumigatus hyphae. Cell-free, in vitro systems were then used to help determine the relative importance of several potentially fungicidal products of neutrophils. Both A. fumigatus and R. oryzae hyphae were damaged by the myeloperoxidase-hydrogen peroxide-halide system either with reagent hydrogen peroxide or enzymatic systems for generating hydrogen peroxide (glucose oxidase with glucose, or xanthine oxidase with either hypoxanthine or acetaldehyde). Iodide with or without chloride supported the reaction, but damage was less with chloride alone as the halide cofactor. Hydrogen peroxide alone damaged hyphae only in concentrations >/=1 mM, but 0.01 mM hypochlorous acid, a potential product of the myeloperoxidase system, significantly damaged R. oryzae hyphae (a 1 mM concentration was required for significant damage to A. fumigatus hyphae). Damage to hyphae by the myeloperoxidase system was inhibited by azide, cyanide, catalase, histidine, and tryptophan, but not by superoxide dismutase, dimethyl sulfoxide, or mannitol. Photoactivation of the dye rose bengal resulted in hyphal damage which was inhibited by histidine, tryptophan, and 1,4-diazobicyclo(2,2,2)octane. Lysates of neutrophils or separated neutrophil granules did not affect A. fumigatus hyphae, but did damage R. oryzae hyphae. Similarly, three preparations of cationic proteins purified from human neutrophil granules were more active in damaging R. oryzae than A. fumigatus hyphae. This damage, as with the separated granules and whole cell lysates, was inhibited by the polyanion heparin. Damage to R. oryzae hyphae by neutrophil cationic proteins was enhanced by activity of the complete myeloperoxidase system or by hydrogen peroxide alone in subinhibitory concentrations. These data support the importance of oxidative products in general and the myeloperoxidase system in particular in damage to hyphae by neutrophils. Cationic proteins may also contribute significantly to neutrophil-mediated damage to R. oryzae hyphae.

Proteolysis by neutrophils. Relative importance of cell-substrate contact and oxidative inactivation of proteinase inhibitors in vitro

Polymorphonuclear leukocytes have been implicated in connective tissue injury in a variety of disease processes. To gain insight into mechanisms by which neutrophils might degrade connective tissue macromolecules in the presence of proteinase inhibitors, we have used a model system that allows neutrophils to be held in vitro under physiologic conditions in close proximity to a very proteinase-sensitive substrate, (125)I-labeled fibronectin. We have found: (a) neutrophils spread rapidly on the fibronectin substrate; (b) fibronectin proteolysis by neutrophils is largely attributable to released elastase, and is linearly related to cell number over the range of 2,000 to 30,000 cells per assay; (c) oxidants released from neutrophils stimulated by opsonized zymosan or phorbol myristate acetate do not protect released elastase from inhibition by alpha(1)-proteinase inhibitor or alpha(2)-macroglobulin; (d) neutrophil myeloperoxidase and enzymatically generated superoxide anion render alpha(1)-proteinase inhibitor ineffective against fibronectin proteolysis when neutrophils are added 30 min later; and (e) alpha(1)-proteinase inhibitor and alpha(2)-macroglobulin incompletely inhibit fibronectin proteolysis by neutrophils (79.8+/-6.3 and 73.5+/-12.0%, respectively.) The data suggested that proteolysis due to neutrophils that are in contact with susceptible macromolecules may occur due to partial exclusion of inhibitors from the cell-substrate interface. Although confirming that alpha(1)-proteinase inhibitor is ineffective against neutrophil-derived proteolysis after exposure to oxidants, these studies did not support the hypothesis that oxidants released from stimulated neutrophils enhance activity of proteinases they release in the presence of alpha(1)-proteinase inhibitor. We anticipate that further studies with this test system will be helpful in defining conditions that modulate inflammatory connective tissue injury in diseases such as pulmonary emphysema and rheumatoid arthritis.

Chlorination of taurine by human neutrophils. Evidence for hypochlorous acid generation

The model hydrogen peroxide-myeloperoxidase-chloride system is capable of generating the powerful oxidant hypochlorous acid, which can be quantitated by trapping the generated species with the beta-amino acid, taurine. The resultant stable product, taurine chloramine, can be quantitated by its ability to oxidize the sulfhydryl compound, 5-thio-2-nitro-benzoic acid to the disulfide, 5,5'-dithiobis(2-nitroben-zoic acid) or to oxidize iodide to iodine. Using this system, purified myeloperoxidase in the presence of chloride and taurine converted stoichiometric quantities of hydrogen peroxide to taurine chloramine. Chloramine generation was absolutely dependent on hydrogen peroxide, myeloperoxidase, and chloride and could be inhibited by catalase, myeloperoxidase inhibitors, or chloride-free conditions. In the presence of taurine, intact human neutrophils stimulated with either phorbol myristate acetate or opsonized zymosan particles generated a stable species capable of oxidizing 5-thio-2-nitrobenzoic acid or iodide. Resting cells did not form this species. The oxidant formed by the stimulated neutrophils was identified as taurine chloramine by both ultraviolet spectrophotometry and electrophoresis. Taurine chloramine formation by the neutrophil was dependent on the taurine concentration, time, and cell number. Neutrophil-dependent chloramine generation was inhibited by catalase, the myeloperoxidase inhibitors, azide, cyanide, or aminotriazole and by chloride-free conditions, but not by superoxide dismutase or hydroxyl radical scavengers. Thus, it appears that stimulated human neutrophils can utilize the hydrogen peroxide-myeloperoxidase-chloride system to generate taurine chloramine. Based on the demonstrated ability of the myeloperoxidase system to generate free hypochlorous acid we conclude that neutrophils chlorinate taurine by producing this powerful oxidant. The biologic reactivity and cytotoxic potential of hypochlorous acid and its chloramine derivatives suggest that these oxidants play an important role in the inflammatory response and host defense.

Lung cell oxidant injury. Enhancement of polymorphonuclear leukocyte-mediated cytotoxicity in lung cells exposed to sustained in vitro hyperoxia

The oxidant damage of lung tissue during in vivo hyperoxic exposure appears to be amplified by neutrophils that release toxic amounts of oxygen metabolites. In our studies cloned lung epithelial cells (L2 cells), lung fibroblasts, and pulmonary artery endothelial cells were cultured under either ambient (Po(2) approximately 140 torr) or hyperoxic (Po(2) approximately 630 torr) conditions for 48 h (24 h for endothelial cells). After cultivation, phorbol myristate acetate- or opsonized zymosan-stimulated neutrophils were added to the cultivated monolayers for 4 h, and lung cell damage was quantitated using (51)Cr release as an index. The data show that stimulated neutrophils are able to injure the three lung cell lines tested, with endothelial cells being highly susceptible to this injury and L2 cells being slightly more susceptible than lung fibroblasts. The studies also demonstrate that all three lung cell lines exposed to sustained hyperoxia are more susceptible to neutrophil-mediated cytotoxicity than their time-matched air controls. Hydrogen peroxide was the main toxic oxygen metabolite because catalase (2,500 U/ml) completely protected the target cells. Equivalent quantities of hydrogen peroxide generated by glucose oxidase instead of by neutrophils gave a similar degree of target cell injury. Superoxide dismutase at high concentrations (250 mug/ml) provided some protection. Other systems that detoxify oxygen metabolites were without protective effect. These findings indicate that the increase in susceptibility of lung cells to neutrophil-mediated oxidant damage is a toxic effect of hyperoxia on lung cells. This specific manifestation of oxygen damage provides insight into the integration between primary mechanisms (oxygen exposure) and secondary mechanisms (release of oxygen metabolites by neutrophils) with respect to the cellular basis for pulmonary oxygen toxicity.

Role of oxygen-derived free radicals and metabolites in leukocyte-dependent inflammatory reactions

Images

Enhanced production of hydroxyl radicals by the xanthine-xanthine oxidase reaction in the presence of lactoferrin

The generation of hydroxyl radicals by the xanthine-xanthine oxidase reaction (C. Beauchamp and I. Fridovich (1970) J. Biol. Chem. 245, 4641-1616) has been shown to be increased by iron-saturated lactoferrin isolated from pig neutrophils. Hydroxyl radical production, measured by EPR spin trapping and by ethylene production from alpha-keto-gamma-methiol butyric acid, has been demonstrated to be produced by a Fenton-type Haber-Weiss reaction catalysed by lactoferrin. The possibility that lactoferrin catalyses such a reaction in vivo is considered.

How important is the myeloperoxidase microbicidal system of phagocytic cells?

The myeloperoxidase system is presented by most immunology textbooks as a major microbicidal system of phagocytic cells. This theory, however, has not bee subjected to vigorous testing in the clinical arena. Of 14 patients with primary myeloperoxidase deficiency, only 3 had infectious complication. All 3 patients have more plausible explanation than myeloperoxidase deficiency for their infectious complications. Two of these patients were healthy until middle age when they developed systemic candidiasis after the onset of diabetes mellitus. The third patient was an infant with a maturational defect in neutrophil chemotaxis whose infectious complications ceased after the normalization of the chemotactic defect. The results of these "experiments of nature" indicate that the meyloperoxidase system is not a major microbicidal mechanism of phagocytic cells.

Monocyte and granulocyte-mediated tumor cell destruction. A role for the hydrogen peroxide-myeloperoxidase-chloride system

Human monocytes stimulated with phorbol myristate acetate were able to destroy a T lymphoblast cell target (CEM). Stimulated human granulocytes were also capable of mediating CEM cytotoxicity to a comparable degree as the monocyte. CEM destruction was dependent on the pH and the effector cell number. Both monocyte or granulocyte mediated cytotoxicity were inhibited by the addition of catalase, whereas superoxide dismutase had no inhibitory effect. In addition, CEM were protected from cytolysis by the effector cells by the myeloperoxidase inhibitors, azide and cyanide, or by performing the experiment under halide-free conditions. Glucose oxidase, an enzyme system capable of generating hydrogen peroxide, did not mediate CEM cytotoxicity, while the addition of purified myeloperoxidase dramatically enhanced cytolysis. Hypochlorous acid scavengers prevented CEM destruction by the glucose oxidase-myeloperoxidase-chloride system but neither hydroxyl radical nor singlet oxygen scavengers had any protective effect. These hypochlorous acid scavengers were also successful in inhibiting monocyte or granulocyte-mediated CEM cytotoxicity. Based on these observations we propose that human monocytes or granulocytes can utilize the hydrogen peroxide-myeloperoxidase-chloride system to generate hypochlorous acid or species of similar reactivity as a potential mediator of CEM destruction.

Polymorphonuclear leucocyte membrane-stimulated oxidative metabolic activity---the effect of divalent cations and cytochalasins

The effect of divalent cations and cytochalasins on the oxidative metabolic response of polymorphonuclear leucocytes (PMNL) following membrane stimulation by opsonized zymosan, phorbol myristate acetate (PMA), digitonin and n-formyl-methionyl-leucyl-phenylalanine (FMLP) was investigated using several techniques. For optimal ferricytochrome C reduction, oxygen consumption, luminol- and lucigenin-dependent chemiluminescence by PMNL on exposure to opsonized zymosan, extracellular magnesium was necessary. In contrast FMLP- and digitonin-induced PMNL metabolic activities were less dependent on extracellular magnesium than on calcium. Phorbol myristate acetate-induced PMNL oxidative metabolic activity occurred independently of extracellular magnesium and calcium concentration. Incubation of PMNL with cytochalasins B and E had little effect on PMA-induced metabolic activity, enhanced FMLP-induced metabolic activity but inhibited digitonin-induced metabolic activity. Investigation of zymosan-induced PMNL metabolic activity after incubation with cytochalasins demonstrated a dissociation between the different parameters of metabolic activity measured. The results described in this study support the suggestion that PMNL oxidative metabolism can be activated by at least two different mechanisms depending on the stimulus used, and that these mechanisms can in part be differentiated by their dependence on extracellular divalent cations.

Evidence for the generation of hydroxyl radical during arachidonic acid metabolism by human platelets

Reactive oxygen species, probably hydroxyl radicals (OH.), have been suggested to be generated during arachidonic acid (AA) metabolism and, once released, these species can modify the rate and extent of various reactions involved in AA metabolism. We have studied this phenomenon in washed human platelets. OH. generation was quantitated using 14C-benzoic acid as a specific trap in a continuous ionization chamber system. Resting platelets did not produce any detectable signal, whereas addition of AA resulted in gradual OH. production with peak values detected at approximately 20 min. Similar studies conducted under nitrogen or after boiling the platelets almost abolished OH. generation. Aspirin had no significant effect, whereas 5,8,11,14-eicosatetraynoic acid decreased the signal by greater than 90%, thus suggesting that OH. is produced primarily through the lipoxygenase pathway. Superoxide dismutase (SOD) and catalase had no effect and, as expected, phenol and mannitol decreased OH. production considerably, by greater than 50% and 90%, respectively. Azide and cyanide also reduced the OH. generation by about two thirds. We conclude that OH. is generated during AA metabolism by human platelets. It is primarily produced via the lipoxygenase pathway and may require a heme-dependent peroxidase. This highly reactive oxidant may play an important role in normal and abnormal hemostasis.

Role of hydrogen peroxide in neutrophil-mediated destruction of cultured endothelial cells

Human neutrophils stimulated with phorbol myristate acetate were able to destroy suspensions or monolayers of cultured human endothelial cells. Neutrophil-mediated cytotoxicity was related to phorbol myristate acetate concentration, time of incubation and neutrophil number. Cytolysis was prevented by the addition of catalase, while superoxide dismutase had no effect on cytotoxicity. The addition of the heme-enzyme inhibitors, azide or cyanide, markedly stimulated neutrophil-mediated damage while exogenous myeloperoxidase failed to stimulate cytolysis. Neutrophils isolated from patients with chronic granulomatous disease did not destroy the endothelial cell targets while myeloperoxidase-deficient neutrophils successfully mediated cytotoxicity. Endothelial cell damage mediated by the myeloperoxidase deficient cells was also inhibited by catalase but not superoxide dismutase. The addition of purified myeloperoxidase to the deficient cells did not stimulate cytotoxicity. Glucose-glucose oxidase, an enzyme system capable of generating hydrogen peroxide, could replace the neutrophil as the cytotoxic mediator. The addition of myeloperoxidase at low concentrations of glucose oxidase did not increase cytolysis, but at the higher concentrations of glucose oxidase it stimulated cytotoxicity. The destruction of endothelial cells by the glucose oxidase-myeloperoxidase system was inhibited by the addition of hypochlorous acid scavengers. In contrast, neutrophil-mediated cytolysis was not effectively inhibited by the hypochlorous acid scavengers. Based on these observations, we propose that human neutrophils can destroy cultured human endothelial cells by generating cytotoxic quantities of hydrogen peroxide.

Hydroxyl radical production in body fluids. Roles of metal ions, ascorbate and superoxide

Hydroxyl radical production, detected by ethylene formation from methional, has been investigated in plasma, lymph and synovial fluid. In the presence of added iron--EDTA, addition of either H2O2 or xanthine and xanthine oxidase gave rise to hydroxyl radical formation that in most cases was not superoxide-dependent. The ascorbate already present in the fluid appeared to participate in the reaction. In the absence of added catalyst, the reaction was hardly detectable, the rate being less than 5% of that observed with 1 microM-iron--EDTA added. This implies that the fluids had little if any capacity to catalyse hydroxyl radical production via this mechanism.

Chemiluminescence by polymorphonuclear leukocytes adhering to surfaces

Stimulation of the plasma membranes of granulocytes results in an oxidative metabolic response. This response can be measured by measuring the reduction of oxidizable substrates, such as Nitro Blue Tetrazolium, as well as by measuring the energy released as light (chemiluminescence). While investigating the oxidative response of human granulocytes, we observed a marked variation in the chemiluminescence response when leukocytes were suspended in a balanced salt solution without gelatin or any other protein. We performed systematic study to investigate the role of protein in suspensions of human polymorphonuclear leukocytes. Final results were identical with human serum, albumin, fetal calf serum, and gelatin; gelatin was used as the protein source in most experiments. Polymorphonuclear leukocytes suspended in Hanks balanced salt solution without gelatin decreased in numbers during incubation at room temperature (approximately 50 percent after 60 min). Cell structures were observed on the walls of the tubes containing leukocyte suspensions without gelatin. Numbers of polymorphonuclear leukocytes were stable in suspensions containing gelatin. A chemiluminescence response which peaked at approximately 10 min and was sustained for at least 30 min was observed in suspensions of polymorphonuclear leukocytes without gelatin. This surface attachment-stimulated chemiluminescence occurred in the absence of either soluble or particulate stimuli. Chemiluminescence was inhibited by either superoxide dismutase or sodium azide and did not occur with suspensions of granulocytes from patients with chronic granulomatous disease. We postulate that both superoxide- and myeloperoxidase-dependent oxidative metabolic reactions are induced during the adherence of polymorphonuclear leukocytes to surfaces. Gelatin or other proteins in leukocyte suspending media are necessary when assays are performed to evaluate the metabolic responses of these cells to particulate or soluble stimuli.

In vivo damage of rat lungs by oxygen metabolites

The intrapulmonary instillation into rat lung of enzymes that generate oxygen metabolites results in acute lung injury. The injection of xanthine oxidase and xanthine produces acute lung injury that, in the presence of superoxide dismutase, but not in the presence of catalase, can be significantly diminished, suggesting that O2- has the capacity to injure the lung. Instillation of a generator of H2O2, namely glucose oxidase, will, in sufficient quantities, produce acute injury that is not neutrophil-dependent. When either a low dose of glucose oxidase alone or lactoperoxidase alone is employed, little lung injury occurs. However, instilling the combination of the two enzymes produces severe, acute injury that can be blocked in a dose-dependent manner by catalase, but not by superoxide dismutase. Purified human leukocytic myeloperoxidase, but not horseradish peroxidase, will substitute for lactoperoxidase in the model of lung injury. The lung damaging effects of these enzymes cannot be attributed to the presence of contaminating proteases. Acute lung injury produced by the instillation of glucose oxidase and lactoperioxidase progresses to interstitial fibrosis. These studies represent a direct application of generators of oxygen metabolites to the in vivo induction of lung injury. The data suggest that rat lung is susceptible to injury by a variety of oxygen metabolites, including O2-, H2O2 and its lactoperoxidase or myeloperoxidase-produced derivatives. The studies also indicate that lung injury produced by oxygen metabolites can result in interstitial pulmonary fibrosis.

Chemiluminescence and superoxide anion production by leukocytes from chronic hemodialysis patients

During phagocytosis or in response to a soluble stimulus, polymorphonuclear leukocytes (PMN) undergo a burst of oxidative metabolism involved intimately in antimicrobial activity. Superoxide anion produced during the burst is bactericidal either directly or as an intermediate metabolite. In addition, stimulated PMN's emit light or chemiluminescence (CL). CL is a sensitive measure of PMN oxidative potential and correlates with antimicrobial activity. Because of the increased susceptibility to infection observed in patients on chronic hemodialysis (CHD), we studied superoxide anion production and luminol-amplified CL in PMN's from CHD patients and in age-matched controls in the resting state and in response to phorbol myristate acetate (PMA). Studies in autologous serum showed higher CL resting values in PMN's from CHD patients, and crossincubation studies indicated that this is a result of factor(s) in the patients' serum. In response to PMA, PMN's from patients on CHD in autologous serum had significantly (P less than 0.05) less of an increase (phorbol-stimulated minus resting values) in CL as compared with controls (peak time values, 60 +/- 5 x 10(3) cpm for patients vs. 76 +/- 5 x 10(3) cpm for controls). With crossincubation studies, these differences persisted, suggesting that there is a defect intrinsic to the patient PMN. Studies done after dialysis showed a significant reduction in resting values. The decreased response to PMA stimulation remained, however. No difference in superoxide anion production (superoxide-dismutase-inhibitable ferricytochrome c reduction) was seen in PMN's from patients on CHD compared with controls. This along with other studies indicating normal nitroblue tetrazolium (NBT) reduction in PMN's from CHD patients indicates that the decreased CL is due to superoxide-anion-independent mechanisms. Because CL correlates well with antimicrobial activity, the reduced response to PMA observed in neutrophils from patients on CHD may explain, at least in part, the increased susceptibility to infection in these patients.