Surfactant protein A prevents silica-mediated toxicity to rat alveolar macrophages.

Silicosis is a serious occupational lung disease associated with irreversible pulmonary fibrosis. The interaction between inhaled crystalline silica and the alveolar macrophage (AM) is thought to be a key event in the development of silicosis and fibrosis. Silica can cause direct injury to AMs and can induce AMs to release various inflammatory mediators. Acute silicosis is also characterized by a marked elevation in surfactant apoprotein A (SP-A); however, the role of SP-A in silicosis is unknown. We investigated whether SP-A directly affects the response of AMs to silica. In this study, the degree of silica toxicity to cultured rat AMs as assessed by a (51)Cr cytotoxicity assay was shown to be dependent on the time of exposure and the concentration and size of the silica particles. Silica directly injured rat AMs as evidenced by a cytotoxic index of 32.9 +/- 2.5, whereas the addition of rat SP-A (5 microg/ml) significantly reduced the cytotoxic index to 16.6 +/- 1.2 (P < 0. 001). This effect was reversed when SP-A was incubated with either polyclonal rabbit anti-rat SP-A antibody or D-mannose. These data indicate that SP-A mitigates the effect of silica on AM viability, and this effect may involve the carbohydrate recognition domain of SP-A. The elevation of SP-A in acute silicosis may serve as a normal host response to prevent lung cell injury after exposure to silica.

Surfactant protein A suppresses reactive nitrogen intermediates by alveolar macrophages in response to Mycobacterium tuberculosis.

Mycobacterium tuberculosis attaches to, enters, and replicates within alveolar macrophages (AMs). Our previous studies suggest that surfactant protein A (SP-A) can act as a ligand in the attachment of M. tuberculosis to AMs. Reactive nitrogen intermediates (RNIs) play a significant role in the killing of mycobacteria. We have demonstrated that RNI levels generated by AMs were significantly increased when interferon-gamma-primed AMs were incubated with M. tuberculosis. However, the RNI levels were significantly suppressed in the presence of SP-A (10 microg/ml). The specificity of SP-A's effect was demonstrated by the use of F(ab')2 fragments of anti-SP-A monoclonal antibodies and by the use of mannosyl-BSA, which blocked the suppression of RNI levels by SP-A. Furthermore, incubation of deglycosylated SP-A with M. tuberculosis failed to suppress RNI by AMs, suggesting that the oligosaccharide component of SP-A, which binds to M. tuberculosis, is necessary for this effect. These results show that SP-A-mediated binding of M. tuberculosis to AMs significantly decreased RNI levels, suggesting that this may be one mechanism by which M. tuberculosis diminishes the cytotoxic response of activated AMs.

Gamma interferon stimulates rat alveolar macrophages to kill Pneumocystis carinii by L-arginine- and tumor necrosis factor-dependent mechanisms.

Pneumocystis carinii pneumonia remains a serious complication for immunocompromised patients. In the present study, P. carinii organisms interacted with gamma interferon (IFN-gamma)-stimulated alveolar macrophages (AMs) to activate the L-arginine-dependent cytocidal pathway involving reactive nitrogen intermediates (RNI) that were assayed as nitrite (NO2-). Unstimulated cultures of AMs produced negligible quantities of RNI. Addition of P. carinii organisms to IFN-gamma-primed AMs resulted in greatly enhanced production of RNI. NO2- levels increased from 0.8 +/- 0.4 to 11.1 +/- 3.8 microM as early as 6 h after P. carinii organisms were incubated with IFN-gamma-stimulated AMs and to 35.1 +/- 8.9 microM after a 24-h incubation, a near-maximum level. High levels of NO2- were produced by AMs primed with as little as 10 U of IFN-gamma per ml in the presence of P. carinii, and a 20-fold increase in IFN-gamma concentration resulted in only a further 65% increase in NO2- production. RNI-dependent killing of P. carinii was demonstrated by both a 51Cr release assay and a [35S]methionine pulse immunoprecipitation assay. Addition of either monoclonal tumor necrosis factor alpha (TNF-alpha) neutralizing antibody or 200 microM NG-monomethyl-L-arginine (L-NGMMA), a competitive inhibitor of the L-arginine-dependent pathway, significantly decreased NO2- production and reduced P. carinii killing. TNF-alpha alone had no effect on P. carinii viability. These results suggest that (i) the specific interaction of P. carinii organisms with IFN-gamma-primed AMs triggers the production of RNI, (ii) RNI are toxic to P. carinii, and (iii) TNF-alpha likely plays a central role in mediating P. carinii killing by IFN-gamma-stimulated AMs.

Nitrated SP-A does not enhance adherence of Pneumocystis carinii to alveolar macrophages.

We investigated whether nitration of surfactant apoprotein (SP) A alters its ability to bind to mannose-containing saccharides on Pneumocystis carinii and its potential role in the mediation of P. carinii adherence to alveolar macrophages. Human SP-A was nitrated by incubation with tetranitromethane at pH 8.0 or synthetic peroxynitrite (ONOO-) at pH 7.4, which resulted in significant nitration of tyrosines in its carbohydrate recognition domain [0.63 +/- 0.001 (SE) and 1.25 +/- 0.02 mol nitrotyrosine/mol monomeric SP-A, respectively; n = 3 samples]. Binding of SP-A to P. carinii was calcium dependent and competitively inhibited by alpha-methyl-D-mannopyranoside. Nitration of SP-A by ONOO- or tetranitromethane decreases its binding to P. carinii by increasing its dissociation constant from 7.8 x 10(-9) to 1.6 x 10(-8) or 2.4 x 10(-8) M, respectively, without significantly affecting the number of binding sites (7.1 x 10(6)/P. carinii organisms, assuming that the native molecular mass of oligomeric SP-A is 650 kDa). Furthermore, ONOO--nitrated SP-A failed to mediate the adherence and phagocytosis of P. carinii to rat alveolar macrophages as observed with normal SP-A. Binding of SP-A to rat alveolar macrophages was not altered by nitration. These results indicate that nitration of SP-A interferes with its ability to serve as a ligand for P. carinii adherence to alveolar macrophages at the site of the SP-A moleculeP. carinii interaction.

Surfactant protein A (SP-A) mediates attachment of Mycobacterium tuberculosis to murine alveolar macrophages.

Attachment of Mycobacterium tuberculosis organisms to alveolar macrophages (AMs) is an essential early event in primary pulmonary tuberculosis. Surfactant protein A (SP-A) is a nonimmune opsonin present in the alveolar spaces that binds carbohydrate residues such as mannose. It was hypothesized that SP-A attaches to M. tuberculosis and serves as a ligand between M. tuberculosis and AMs. [125I]SP-A was found to bind to M. tuberculosis in a time- and [Ca2+]-dependent manner with a Kd of 1.9 x 10(-9) M and an apparent number of 6.3 x 10(2) SP-A binding sites/organism. Further, deglycosylated SP-A had minimal binding to M. tuberculosis, indicating that sugar moieties are important in this interaction. SP-A specifically binds to a 60-kD cell-wall protein from M. tuberculosis. SP-A-mediated attachment of 51Cr-labeled M. tuberculosis organisms to AMs is dependent on time, SP-A concentration, and Ca2+. M. tuberculosis attachment to murine AMs in the absence of SP-A was 12.8 +/- 0.9%; however, in the presence of 5 microg/ml SP-A the attachment increased to 38.6 +/- 2.9% (P < 0.001). SP-A-mediated attachment was significantly decreased from 38.6 +/- 2.9% to 18.7 +/- 3.3% (P < 0.05) in the presence of antihuman SP-A antibodies. When the attachment assay was repeated in the presence of alpha-methylene-D-mannosepyranosidase (mannosyl-BSA) and type V collagen, SP-A-mediated attachment decreased from 38.6 +/- 2.9% to 16.6 +/- 1.5% (P < 0.001) and 19.1 +/- 1.4% (P < 0.05), respectively. Further, deglycosylated SP-A had only a minimal effect on M. tuberculosis attachment to AMs. These data indicate that SP-A can mediate M. tuberculosis attachment to AMs, and suggest possible underlying mechanisms for this.

Assessment of in vivo attachment/phagocytosis by alveolar macrophages.

Alveolar macrophages (AMs) are recognized as an important first line of cellular host defense within the lung. Although mechanisms underlying AM response to microorganisms or particulates are well characterized in vitro, experimental approaches to the study of AMs in vivo are limited. To circumvent these limitations, a new assay was developed using fluorescently labelled liposomes or Pneumocystis carinii (PC) organisms which were administered intratracheally into mechanically ventilated rats. After 30 min, the lungs were lavaged and the percentage of administered liposomes or PC bound to AMs was determined by quantifying fluorescence. Factors known to enhance attachment/phagocytosis by AMs in vitro were assayed to determine their effect in vivo. For example, vitronectin (VN)-coated liposomes increased attachment from 25.2 +/- 2.4% to 47.2 +/- 3.0% (p < 0.001), while addition of VN increased the binding of PC to AMs from 16.5 +/- 1.7% to 24.5 +/- 2.2% (p < 0.05). Confocal laser microscopy of cells obtained by lavage provided morphologic evidence of attachment/phagocytosis by AMs. This model will permit the quantitative assessment of the interaction of fluorescently labelled liposomes or microorganisms with AMs in the lower respiratory tract of living animals.

Functional impairment of bronchoalveolar lavage phospholipids in early Pneumocystis carinii pneumonia in rats.

Surfactant abnormalities may contribute to the impairment of gas exchange observed in Pneumocystis carinii pneumonia. Analysis of rat bronchoalveolar lavage (BAL) lipid extracts from normal controls, steroid controls, trimethaprim-sulfamethoxazole (TMP-SMX) controls, TMP-SMX/P. carinii pneumonia controls, and P. carinii pneumonia animals reveal similar total phospholipid and total protein levels. However, there was a marked reduction in phosphatidylglycerol (PG) from the BAL of P. carinii pneumonia rats as compared with control animals, with a decrease from 4.91 +/- 1.29 nmol/mg protein to 0.46 +/- 0.57 nmol/mg protein (p<0.05) and a decrease, as a percent of total phospholipids, from 7.7% +/- 0.88% to 0.91% +/- 0.59% (p<0.001). Furthermore, in vitro surface activities of BAL lipid extracts from control and P. carinii pneumonia rats revealed minimum surface tension increases from 9.38 +/- 1.71 mN/m in controls to 16.36 +/- 0.83 mN/m in P. carinii pneumonia rats (p<0.05) and likewise maximum surface tension increases from 22.14 +/- 4.34 mN/m to 38.57 +/- 2.07 mN/m (p<0.01). Of interest, the surface activity of PG-deficient P. carinii pneumonia BAL lipid extracts is completely restored to that of normal controls by the addition of exogenous PG. These findings suggest that a functionally abnormal surfactant occurs in P. carinii pneumonia and that this may account, in part, for the impairment of gas exchange observed in this disorder.

Cyclophosphamide-induced lung toxicity: mechanism of endothelial cell injury.

Cyclophosphamide (CP) is associated with significant pulmonary toxicity; however, the mechanism of toxicity is unknown. An in vitro endothelial model of injury was developed to assess the direct toxic effects of CP, CP derivatives and CP metabolites on cultured endothelial cells. Injury to 51Cr-labeled bovine artery pulmonary endothelial (BPAE) cells was quantified by the release of 51Cr from BPAE cells incubated for 18 h with injury expressed as a cytotoxic index. Because CP activation and metabolism occurs primarily in liver, assays assessing CP effects were conducted in the presence of an hepatic microsomal enzyme system. Upon activation, CP produces 4-hydroxycyclophosphamide, acrolein (ACR) and the alkylating metabolite, phosphoramide mustard. Nonactivated CP demonstrated no toxicity to BPAE cells within 18 h; whereas, activated CP induced significant BPAE cell injury in a concentration-dependent manner. Specific metabolites of CP 4-hydroxycyclophosphamide and ACR were markedly more toxic to BPAE cells than phosphoramide mustard. Sulfhydryl-rich compounds, S-2-(3-aminopropylamino)ethylphosphoric acid (WR-2721) and N-acetylcysteine, significantly reduced 4-hydroxycyclophosphamide- and ACR-induced injury but had no significant protective effect against phosphoramide mustard-induced toxicity. These studies suggest 1) CP is not metabolized within pulmonary artery endothelial cells, 2) ACR may be the principal CP metabolite involved in mediating direct injury to pulmonary artery endothelial cells and 3) sulfhydryl-rich agents may be effective in reducing CP-induced damage to critical endothelial cell barriers.

Mechanism of putrescine transport in human pulmonary artery endothelial cells.

Effective lung repair requires optimal replication of critical cell populations in the lung. Endogenous polyamines such as putrescine, spermidine and spermine play important roles in cell proliferation and differentiation, and may arise due to intracellular synthesis or transport into the cell. To determine the mechanism of polyamine transport in lung endothelial cells, the uptake of putrescine in human pulmonary artery endothelial cells was examined. Putrescine (7 nM) uptake into the cells approached equilibrium at 1 hr and was inhibited by methylglyoxal bis(guanylhydrazone). Kinetic studies revealed that uptake occurred via both a high- and low-affinity system. The effect of several amines (700 microM) on the 15-min uptake of putrescine was examined and a rank order of inhibition was determined: methylglyoxal bis(guanylhydrazone) > putrescine > spermine > spermidine > gentamicin > mepiperphenidol. alpha-Aminoisobutyric acid, a prototype system A amino acid, and tetraethylammonium, an organic cation, had no effect. N-ethylmaleimide inhibited transport 71%, whereas dinitrophenol did not. A reduction in temperature from 37 degrees C to 5 degrees C resulted in a 42% decrease in putrescine transport. Additionally, removing fetal bovine serum from the uptake medium reduced transport 38%. These data indicate that human pulmonary artery endothelial cells possess a specific transport system for polyamines. An improved understanding of this pathway in pulmonary endothelial cells may permit development of strategies to facilitate growth and repair of this critical cell population.

Amiodarone-induced injury of human pulmonary artery endothelial cells: protection by alpha-tocopherol.

Amiodarone is a potent antidysrhythmic drug that is associated with severe pulmonary toxicity. The mechanism of amiodarone pulmonary toxicity is poorly understood. To investigate the possible involvement of oxygen-derived metabolites in amiodarone-induced injury, 51Cr-labeled human pulmonary artery endothelial (HPAE) cells were incubated with amiodarone for 18 hr in the presence of various antioxidants and in hypoxic and hyperoxic conditions with cell injury quantified by 51Cr release, expressed as cytotoxic index. Amiodarone (10-50 microM) directly injured HPAE cells in a concentration-dependent manner, but the injury was not modulated by altering ambient oxygen concentrations. Furthermore, amiodarone-induced injury (30 microM) was not reduced by the following antioxidants: catalase, superoxide dismutase, ascorbic acid, dimethyl sulfoxide and ethanol. In contrast, toxicity from 30 microM amiodarone was significantly reduced by alpha-tocopherol (alpha-TOC) at 10, 20 and 40 microM from a cytotoxic index of 41.6 +/- 3.5 to 25.5 +/- 7.9, 10.61 +/- 5.4 and 3.1 +/- 2.8, respectively. As revealed by phase microscopy, alpha-TOC (40 microM) prevented any evidence of toxicity to the amiodarone-treated cells. Amiodarone concentrations in the HPAE cells incubated in the presence and absence of alpha-TOC were not significantly different, indicating that alpha-TOC did not interfere with the uptake of the drug by the cells. Similarly, amiodarone did not interfere with the uptake of alpha-TOC by the HPAE cells. Although the specific mechanism of action remains unclear, alpha-TOC affords nearly complete protection in vitro from the cellular injury induced by amiodarone.

Mechanism of phospholipidosis in amiodarone pulmonary toxicity.

Amiodarone therapy can be complicated by life-threatening pulmonary toxicity and is invariably associated with characteristic morphologic changes in the lung consistent with a severe phospholipidosis. To determine possible mechanisms, this study utilizes a unique in vitro cell culture model of amiodarone pulmonary toxicity and demonstrates that amiodarone can directly induce an accumulation of phospholipids within bovine pulmonary artery endothelial (BPAE) cells during the first 24 hr using amiodarone concentrations equivalent to concentrations found in the blood and lungs of human subjects. For example, amiodarone at 7.5 microM during a 24-hr incubation increases [32P]orthophosphate incorporation into phospholipids from 193 +/- 10 X 10(3) dpm/10(6) cells to 266 +/- 19 X 10(3) dpm/10(6) cells (P less than .01). A generalized increase in all phospholipids occurs after amiodarone treatment of the cultured cells; however, several specific phospholipids including phosphatidylinositol, phosphatidic acid and bis(monoacylglycerol) phosphate are all significantly increased to a greater extent than other phospholipids. Furthermore, the data indicate that amiodarone is a potent inhibitor of lysosomal phospholipase A1 and A2 activities derived from the BPAE cells; whereas, amiodarone has no effect on phospholipase A1 and A2 activities from the BPAE microsomal fraction. Thus, this study demonstrates phospholipids accumulate in cultured BPAE cells exposed to amiodarone and provides direct evidence that the drug initiates a specific and nearly complete inhibition of phospholipid degradation by lysosomal phospholipase A1 and A2 suggesting a similar process may occur in vivo in the lungs of subjects with amiodarone pulmonary toxicity.

Oxygen-mediated impairment of human pulmonary endothelial cell growth: evidence for a specific threshold of toxicity.

Oxygen toxicity to the lung is characterized by injury of the pulmonary capillary endothelium with progressive loss of functioning alveolar-capillary units. Current concepts suggest that the risk of O2 toxicity in human subjects is greatly increased with O2 concentrations exceeding 50% to 60%, although there are no data to support a cellular basis for this apparent threshold of toxicity. Our study suggests that a cellular threshold may exist in human pulmonary endothelial cells for O2 toxicity. Hyperoxia was directly toxic to cultured human pulmonary artery endothelial (HPAE) cells, with impairment of replicative function, expressed as growth impairment (GI) index, monitored by two independent parameters: cell number determination and tritiated thymidine incorporation. Impaired cell growth occurred as early as 8 hours after beginning exposure to 95% O2 and with concentrations as low as 60% during a 48-hour incubation. For example, 60% O2 resulted in an impairment of HPAE cell growth at 48 hours with a GI index (cell number) of 37.5 +/- 2.1 (p less than 0.001, comparison with control cells in normoxia). Furthermore, 95% O2 impaired cell growth, as monitored by tritiated thymidine incorporation, as early as 8 hours after exposure (GI index of 43.6 +/- 4.9) however, the injury was completely reversible when cells were reincubated in normoxia for 6 hours (GI index of 4.2 +/- 4.7), p less than 0.001. O2 toxicity was associated with an increase in cellular glutathione levels but was not associated with a detectable loss of antioxidant enzyme activity.(ABSTRACT TRUNCATED AT 250 WORDS)

Importance of hydrogen peroxide in nitrofurantoin-induced cytotoxicity: evidence from an inbred catalase-deficient strain of mice.

Significant pulmonary toxicity is associated with the use of nitrofurantoin; however, the mechanism of cellular toxicity remains poorly characterized. By using a novel in vitro red blood cell (RBC) chromium 51 cytotoxicity assay, cell injury induced by nitrofurantoin was quantified with normocatalasemic BALB/c RBCs and hypocatalasemic (but otherwise genetically identical) CCN RBCs as target cell populations. Nitrofurantoin at concentrations of 2 x 10(-4) and 4 x 10(-4) mol/L resulted in significant injury to normocatalasemic RBCs with a cytotoxic index (CI) of 21.7% +/- 3.7% and 65.3% +/- 3.7% (p less than 0.05, both comparisons). This injury was substantially increased when nitrofurantoin (2 x 10(-4) and 4 x 10(-4) mol/L was incubated with hypocatalasemic RBCs, resulting in CIs of 59.0% +/- 7.4% and 91.0% +/- 2.0% respectively (p less than 0.05, both comparisons with normocatalasemic RBCs). Direct oxidant-mediated cytotoxicity induced by either H2O2 or the superoxide anion radical (as generated by xanthine-xanthine oxidase) also resulted in more significant injury to hypocatalasemic RBCs than to normocatalasemic RBCs (p less than 0.05, both comparisons). Catalase levels of CCN RBCs were approximately 7% of control BALB/c RBC values; however, the activities of superoxide dismutase and glutathione peroxidase were identical in both populations of RBCs. This model, using genetically defined target cell populations, clearly demonstrates the importance of endogenous catalase in protecting against nitrofurantoin-induced cytotoxicity, suggesting that H2O2 is a critical intermediary in the direct cell injury mediated by the drug.

Iprindole reverses the lamellar body deficiency of cultured L-2 cells. Possible implications in the reversal of surfactant deficiency.

Type II alveolar epithelial cells in long-term culture typically lose the ability to synthesize surfactant together with a loss of the characteristic lamellar bodies in the cytoplasm of the cells. Iprindole, a cationic amphiphilic drug, induces lamellar body formation in cultured L-2 cells, a cell line derived from rat Type II cells, but devoid of lamellar bodies. With concentrations of 10(-7)-10(-5) M iprindole, which approximate therapeutic plasma concentrations in human subjects, the drug induced the formation of lamellar bodies within 8 hours of incubation. This effect on cell morphology was rapidly lost after withdrawal of the drug. At concentrations of iprindole which induced lamellar body formation, there was a significant increase in phospholipid content of the L-2 cells as well as an increase in disaturated phosphatidylcholine, an important constituent of surfactant. These studies suggest that the use of drugs such as iprindole may represent a novel approach in the augmentation of phospholipid (and possibly surfactant) content of Type II cells in the lung.

Bleomycin-induced pulmonary endothelial cell injury: evidence for the role of iron-catalyzed toxic oxygen-derived species.

Bleomycin, an effective cancer chemotherapeutic agent, is associated with serious pulmonary toxicity. As an in vitro model of bleomycin pulmonary toxicity, this study examined the ability of bleomycin to injure chromium 51-labeled bovine pulmonary artery endothelial (BPAE) cells in an 18-hour cytotoxicity assay. The data indicate that bleomycin-mediated injury to cultured BPAE cells can be quantified by 51Cr release, expressed as cytotoxic index (CI). Bleomycin-mediated injury to 51Cr-labeled BPAE cells (CI 19.4 +/- 1.6) could be significantly reduced by the iron chelator deferoxamine, 10(-3) mol/L (CI 7.5 +/- 1.1, P less than 0.001), but not by ethylenediaminetetraacetic acid, 10(-5) mol/L (CI 19.8 +/- 2.2). Similarly, bleomycin-mediated injury to BPAE cells (monitored by lactate dehydrogenase release) with a CI 27.1 +/- 4.8 could be reduced by 10(-3) mol/L deferoxamine to CI 10.5 +/- 2.6 (P less than 0.01). In contrast, hyperoxia (95% O2) accelerated bleomycin (1 to 100 mU/ml) toxicity to BPAE cells (P less than 0.01, all comparisons). This study suggests that bleomycin-induced injury of pulmonary endothelial cells may be dependent in part on two critical factors in the cellular environment: the availability of iron to the cell and the ambient O2 concentration.

Reduction of neutrophil-mediated injury to pulmonary endothelial cells by dapsone.

The presence of activated neutrophils in the alveolar structures is thought to contribute to parenchymal cell injury in various acute and chronic lung disorders. This study indicates that dapsone (30 micrograms/ml), an agent with anti-inflammatory properties, can significantly reduce neutrophil-mediated injury to 51Cr-labeled bovine pulmonary artery endothelial (BPAE) cells with a reduction in the injury (expressed as a cytotoxic index) from 65 +/- 3 to 33 +/- 3 (p less than 0.001). Dapsone was unable to protect 51Cr-labeled BPAE cells injured by the chemical generation of superoxide, hydrogen peroxide, or neutrophil-derived, myeloperoxidase-dependent hypohalite ion. In contrast, dapsone significantly inhibited the respiratory burst of the neutrophil, with a reduction in the generation of superoxide, hydrogen peroxide, and conversion of nitroblue tetrazolium to formazan (p less than 0.01, all comparisons). Thus, dapsone appears to protect lung parenchymal cells such as endothelial cells from neutrophil-mediated injury by directly inhibiting the respiratory burst of the neutrophil, with a consequent diminution in the generation of toxic, oxygen-derived radicals.

Nitrofurantoin-stimulated oxidant production in pulmonary endothelial cells.

Nitrofurantoin, a urinary antiseptic, is associated with significant pulmonary toxicity. Our study indicates that nitrofurantoin may produce lung injury by directly stimulating lung parenchymal cells to generate toxic oxygen species such as superoxide and hydrogen peroxide, which can overwhelm cellular antioxidant defenses and result in permanent injury to the cells. Nitrofurantoin (10(-3) mol/L) stimulates bovine pulmonary artery endothelial cells to release 3.7 +/- 0.4 mumol/L superoxide per 10(5) cells and 4.4 +/- 0.5 mumol/L hydrogen peroxide per 10(5) cells (p less than 0.001 compared with endothelial cells without nitrofurantoin). Endothelial cells treated with nitroblue tetrazolium, a yellow dye reduced by superoxide to insoluble blue formazan, can be monitored both spectrophotometrically and morphologically. Nitrofurantoin (10(-5), 10(-4), and 10(-3) mol/L) stimulated pulmonary endothelial cells to reduce nitroblue tetrazolium monitored spectrophotometrically as 0.022 +/- 0.001, 0.032 +/- 0.002, and 0.071 +/- 0.004 delta A515 per 10(5) cells per hour, respectively (p less than 0.001, comparison of cells with 10(-4) and 10(-3) mol/L nitrofurantoin vs. control cells. From the same dose-response curve, endothelial cells incubated with nitrofurantoin morphologically demonstrated formazan granules in the cytoplasm in 17% +/- 6%, 71% +/- 9%, and 92% +/- 5% of cells. Nitrofurantoin (10(-5), 10(-4), and 10(-3) mol/L) directly injures 51Cr-labeled pulmonary endothelial cells, with injury expressed as a cytotoxic index of 1 +/- 1, 20 +/- 4, and 51 +/- 3, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)