Effects of Exogenous Nitric Oxide and Hyperoxia on Lung Fibroblast Viability and DNA Fragmentation

Biochemical and Biophysical Research Communications 262, 685– 691 (1999) Article ID bbrc.1999.1216, available online at http://www.idealibrary.com on

Effects of Exogenous Nitric Oxide and Hyperoxia on Lung Fibroblast Viability and DNA Fragmentation Nandkishore Raghuram,* James D. Fortenberry,* ,1 Marilyn L. Owens,* and Lou Ann S. Brown† *Division of Critical Care and †Division of Neonatology, Department of Pediatrics, Emory University School of Medicine, Atlanta, Georgia; and Egleston Pediatrics Group, Egleston Children’s Hospital, Atlanta, Georgia 30322

Received July 26, 1999

Effective lung repair after acute injury requires elimination of proliferating mesenchymal and inflammatory cells without inducing an acute inflammatory response or disturbing concomitant repair of lung microvasculature. Previous studies have shown that endogenous NO regulates programmed cell death in fibroblasts and can modulate wound fibroblast synthetic function. We hypothesized that exposure of human lung fibroblasts to NO gas would decrease viability and induce apoptotic cell death. Primary cultures of normal human lung fibroblasts were exposed for 4 h to room air (RA), 80% oxygen, NO (at either 20 or 50 ppm) blended with RA, or NO blended with 80% O 2, then incubated for 24 to 72 h. Cell viability was determined by fluorescence viability/cytotoxicity assay and DNA fragmentation by TUNEL assay. Peroxynitrite formation was assessed using immunoblotting for S-nitrosotyrosine. NO plus O 2 induced significant cell death at 20 and 50 ppm NO when compared to either RA or O 2 alone at both 24 and 72 h (p < 0.05). Incubation with superoxide dismutase (SOD), catalase (CAT) or SOD 1 CAT significantly decreased cell death in fibroblasts treated with NO 20/O 2 and NO 50/O 2 compared with controls (p < 0.05). NO 20/O 2 and NO 50/O 2 exposure significantly increased TUNEL mean fluorescence intensity (MFI), consistent with increased DNA fragmentation, compared to RA at 24 and 72 h (p < 0.05). Antioxidants decreased MFI in cells exposed to NO 20/O 2 (CAT and SOD 1 CAT) compared to controls at 24 h (p < 0.05). Western blot analysis for S-nitrosotyrosine showed increased signal intensity in fibroblasts exposed to NO at 20 and 50 ppm plus O 2 compared to RA or O 2 alone. Incubation with SOD 1 CAT reduced sig1 To whom correspondence and reprint requests should be addressed at Egleston Children’s Hospital, 1405 Clifton Road NE, Atlanta, GA 30322. Fax: (404) 325-6233. Abbreviations used: ARDS, acute respiratory distress syndrome; BPD, bronchopulmonary dysplasia; CAT, catalase; EDTA, ethylenediaminetetraacetate; Hepes, N-2-hydroxyethylpiperazine-N9-2ethanesufonic acid; MFI, mean fluorescence intensity; NO, nitric oxide; O 2, oxygen; SOD, superoxide dismutase; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling.

nal intensity for peroxynitrite in cells exposed to NO 20/ O 2. We conclude that NO in hyperoxic conditions induces fibroblast cell death and DNA fragmentation, which could be partially mediated by peroxynitrite synthesis. © 1999 Academic Press Key Words: fibroblast; nitric oxide; inhaled nitric oxide; DNA fragmentation; apoptosis; lung injury.

The clinical outcome of patients after acute lung injury is dependent on the biological response of lung mesenchymal cells, including alveolar epithelial cells and fibroblasts involved in remodeling and repair (1, 2). The acute phase of diffuse alveolar damage is followed by a marked fibroproliferative response (3). Excessive fibroproliferation contributes to bronchopulmonary dysplasia (BPD) following hyaline membrane disease in the neonate and is associated with chronic lung injury and worsened outcome in acute respiratory distress syndrome (ARDS) in children and adults (3, 4). Hyperoxia, barotrauma, inflammation, and infection are major pathogenic factors in acute lung injury (3). Progressive pulmonary fibroproliferation often leads to alveolar fibrosis culminating in respiratory failure and death (5). Directly, pulmonary fibrosis can contribute to refractory hypoxemia and hypercarbia, a cause of death in a small proportion of ARDS patients (5). Indirectly, pulmonary fibrosis enhances ventilator dependency and compromises pulmonary defense mechanisms, which can lead to nosocomial pneumonia, extrapulmonary infections and other septic complications. Sepsis-related factors are responsible for the majority of late deaths in ARDS patients (5). Current strategies to combat these diseases are directed toward nonspecific suppression of inflammation, defending the parenchyma against injury and suppressing the response of mesenchymal cells to exogenous growth signals (2). For example, treatment with corticosteroids, a known inhibitor of fibroblast growth, has been suggested to decrease pulmonary fibroproliferation in patients with late ARDS (5). Short term

685

0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

Vol. 262, No. 3, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 1. Effects of NO on fibroblast viability following exposure. NO significantly decreased viability at 20 and 50 ppm in presence of hyperoxia at both 24 and 72 h compared to RA control (*p , 0.05, mean 6 standard deviation). NO 50/O 2 exposure significantly decreased viability at both time points compared to hyperoxia alone (**p , 0.05).

treatment of BPD with dexamethasone has been shown to improve pulmonary function, suppress biochemical indices of inflammation and decrease the level of respiratory support required by ventilatordependent premature infants with BPD (3). Effective lung repair and concomitant reepithelialization and recanalization of the lung microvasculature requires the elimination of proliferating mesenchymal and inflammatory cells from the alveolar air space and interstitium (6). Apoptosis, a form of programmed cell death, has been recognized as a potential mechanism of clearance of unwanted cells without inciting additional inflammatory response. Apoptotic cells are present within airspace granulation tissue of patients with ARDS, indicating that apoptosis is found in remodeling after lung injury (4). Modulation of fibroblast apoptosis thus could potentially affect the resolution of lung injury. Inhaled nitric oxide (NO) is an intriguing new therapy for the treatment of pulmonary hypertension in a variety of disease states, including persistent pulmonary hypertension of the newborn, management fol-

lowing repair of congenital heart defects, and ARDS (7). NO may also have immunomodulatory effects, including control of fibroblast proliferation (8, 9, 10). Endogenous NO regulates apoptosis in fibroblasts (11). IL-1-stimulated human articular chondrocytes underwent apoptosis in presence of endogenous NO and oxygen radicals, while necrosis was observed when the simultaneous production of NO was reduced (12). We have found that NO gas at clinically relevant concentrations can inhibit neutrophil oxidative function and induce neutrophil death and DNA fragmentation consistent with apoptosis (13, 14). The mechanism of NOinduced death is unknown, but peroxynitrite, a toxic metabolite from the interaction of NO and superoxide anions (15), can induce apoptotic cell death (16). We hypothesized that exposure of human lung fibroblasts to NO gas would also induce cell death and DNA fragmentation consistent with apoptosis. We also hypothesized that fibroblast cell death from NO and hyperoxia is associated with peroxynitrite exposure and that inhibition of peroxynitrite formation would attenuate DNA fragmentation. MATERIALS AND METHODS Fibroblast culture. Primary cultures of normal human lung fibroblasts were obtained from commercially available cryopreserved cell samples (Clonetics, San Diego, CA). Cells were grown in a fibroblast basal medium supplemented with human fibroblast growth factor (1 mg/ml), fetal bovine serum (final concentration 2%), insulin (5 mg/ ml), gentamicin (50 mg/ml) and amphotericin B (50 mg/ml) (Clonetics). The fibroblast medium used is a modification of medium MDCDB 202, as described by McKeehan and Ham (17; Clonetics, personal communication). Iron is reportedly present in this formulation as ferrous sulfate at a concentration of 5 3 10 26 M. Cells were incubated in 5% CO 2 and room air at 37°C and were allowed to reach near-confluence in tissue culture flasks by light microscopy prior to subcultivation. Trypsin (0.025%), EDTA (0.01%), trypsin neutralizing solution and Hepes buffered saline solution (Clonetics) were used to detach and recover cells for transfer. Experiments were performed on cells between the fourth and eighth passage. Cell exposures. Transferred cells were plated on six well culture plates (Becton Dickinson) and incubated in 5% CO 2 and room air at 37°C and serum free media for 24 h to allow for attachment. Each six-well dish was then exposed for 4 h to one the following conditions: 21% oxygen (room air, RA); 80% oxygen (O 2); room air blended with NO in concentrations of 20 (NO 20) or 50 ppm (NO 50); 80% oxygen blended with NO in concentrations of 20 (NO 20/O 2) or 50 ppm (NO 50/ O 2). NO gas concentrations were determined using a NO chemiluminescence analyzer (Model 42 H, Thermo Environmental Instruments Inc, USA) and oxygen concentration was determined using a portable analyzer (Miniox I Oxygen Analyzer).

FIG. 2. Representative fluorescence cytotoxicity/viability assay photomicrographs (203) of NO-exposed fibroblasts. Live, esterasebearing cells stain blue-green and permeable necrotic cells take up red ethidium homodimer fluorescent dye. Photographs of fibroblasts 24 h following 4 h exposure to: (A) RA control, (B) 80% oxygen, (C) NO 50/RA, (D) NO 50/O 2. FIG. 4. (A) Percent of TUNEL positive fibroblasts (mean 6 SD) 72 h after NO 50/O 2 exposure expressed as percent of total fibroblasts in the same field. TUNEL positive cells significantly increased compared with RA control and O 2 alone (*p , 0.03). (B) Phase contrast and TUNEL fluorescence microscopy in NO-exposed fibroblasts (original magnification 203). Phase contrast photomicrographs of RA control (A) and (C) NO 50/O 2-exposed cells. Fluorescence photomicrographs of RA control (B) and (D) NO 50/O 2-exposed cells. Note increased TUNEL positive fibroblasts with NO exposure; cells appear shrunken with evidence of membrane blebbing indicative of apoptosis. 686

Vol. 262, No. 3, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

687

Vol. 262, No. 3, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS dium citrate at 4°C. Samples were then incubated for 60 minutes at 37°C in the absence or presence of exogenous TdT and incubated with fluorescein-conjugated dUTP for repair of nicked 39OH DNA ends. Cells were photographed under both phase and fluorescent microscopy, then counted to determine a percentage of TUNEL positive cells to total cells. Mean cell fluorescence intensity (MFI) of 5000 fibroblasts was also assessed by flow cytometry (FACScan, Becton Dickinson) for each condition. Increasing MFI has been shown to correlate with increasing DNA fragmentation (14).

FIG. 3. Effects of NO and hyperoxia on fibroblast DNA fragmentation: TUNEL assay. Increase in DNA fragmentation, as represented by percent increase of mean fluorescence intensity (MFI; mean 6 SD) compared to daily RA control samples. MFI was significantly increased in NO 20/O 2- and NO 50/O 2-exposed samples compared with RA at 24 h (*p , 0.05) and compared with O 2 alone at 24 or 72 h (**p , 0.05). NO 50/O 2 exposure increased MFI compared with NO 50/RA (#) and NO 20/O 2 (##) at 72 h (p , 0.05).

To determine the interaction of oxygen radical production and potential peroxynitrite formation on NO-induced effects, an additional set of experiments was performed. Prior to the gas exposures described above, cells were also incubated in the absence or presence of superoxide dismutase (300 units/ml; SOD, Sigma, St. Louis, MO), catalase (1000 units/ml; CAT, Sigma, St. Louis, MO), or SOD and CAT in the above concentrations. These treated cells were then incubated for 24 or 72 h in 5% CO 2 and room air oxygen at 37°C prior to analysis for cell death and DNA fragmentation. Fluorescence microscopy for cell viability. Cell necrosis was evaluated by fluorescence viability/cytotoxicity assay (Eukolight, Molecular Probes) (13, 14). Briefly, after exposures, cells were stained with a mixture of the fluorescent probes calcein AM and ethidium homodimer. After uptake only viable cells containing functioning esterases can cleave the ester group on calcein AM to generate a characteristic blue-green fluorescence under fluorescent microscopy. Ethidium homodimer penetrates the permeable membranes of nonviable cells and binds with nucleic acids identifiable by red-orange fluorescence. Cells with green and red fluorescence were counted under a fluorescent microscope (Zeiss) from three high power fields, averaged and expressed as the percentage of calcein (blue-green) positive cells/total calcein (blue-green) and ethidium (red) positive cells counted. TUNEL assay. Specific 39-hydroxyl ends of DNA fragments generated by endonuclease-mediated apoptosis are preferentially repaired by terminal deoxynucleotidyl transferase (TdT) (18). The TdT-mediated dUTP nick end labeling (TUNEL) assay has been developed to label these strand breaks with fluorescent nucleotides and provide a sensitive and specific measure of DNA fragmentation, consistent with apoptosis, within individual cells (19). Cell samples were analyzed after 24 –72 h of incubation following gas exposure. The supernatant and detached cells were rinsed with Hepes buffered saline solution and removed from individual wells with trypsin-EDTA and TNS. Cells were then fixed in 4% paraformaldehyde and permeabilized with 1% Triton X-100 and 0.1% so-

Measurement of nitrosylated tyrosine (S-nitrosotyrosine). Peroxynitrite has a half-life of seconds, but its existence can be “footprinted” by demonstration of nitrosylated proteins such as nitrosylated tyrosine (S-nitrosotyrosine) (20). S-Nitrosotyrosine was demonstrated by Western blotting of fibroblast cytosolic protein extracts using a specific monoclonal antibody for S-nitrosotyrosine. Briefly, whole cell lysates were obtained from exposed cell samples that were washed and resuspended in buffer containing 5 mM Hepes (pH 7.9), 26% glycerol, 1.5 mM MgCl 2, 0.2 mM EDTA, 0.5 M DTT and 0.2 M PMSF. Samples were then homogenized, incubated on ice and centrifuged. Aliquots of supernatant were snap-frozen in liquid nitrogen and stored at 270°C for later use. Protein concentration was determined in thawed specimens by Bradford assay and samples diluted to 100 mg/ml of protein. Whole cell extracts were then incubated overnight with rabbit polyclonal IgG primary anti-human nitrotyrosine (Upstate Biotechnology) at 4°C in RIPA buffer. Extracts were incubated with protein Agarose A, centrifuged and placed in sample buffer containing 0.5 M Tris-HCl (pH 6.8), 20% glycerol, 10% SDS, 10% 2-b-mercaptoethanol, and 0.05% bromophenol blue. Extracts were heated at 95°C and centrifuged, then loaded and separated by SDS-polyacrylamide gel electrophoresis. Proteins were transferred onto polyvinylidene difluoride membranes (Millipore) and incubated at 4°C overnight in solution containing primary anti-human nitrotyrosine, 3% skim milk and 0.1% Tween-20 in Tris-buffered saline. The exposed membranes were washed and then treated with secondary anti-rabbit IgG antibody (1:5000, Sigma) coupled to horseradish peroxidase. Immunodetection was performed using the Enhanced Chemiluminescence kit as described previously (NEN, DuPont) (20). Statistical analysis. Statistical data were obtained using one way analysis of variance (ANOVA), ANOVA on ranks and the Student-Newman-Keuls method for multiple comparisons. Results are expressed as mean 6 SD. Values of P , 0.05 were considered significant.

RESULTS Effects of NO and Hyperoxia on Fibroblast Viability Nitric oxide in the presence of hyperoxia (NO/O 2) induced significant cell death with increasing NO concentrations (Figs. 1 and 2). At 50 ppm NO exposure, cell death approached 100% when compared to RA controls at 24 and 72 h of incubation post exposure (n 5 3–10 for each group; p , 0.05). NO/O 2 at 20 and 50 ppm induced significantly greater death than O 2 alone (n 5 3–10; p , 0.05). NO/RA alone did not significantly increase fibroblast death after 4 h exposure at either NO concentration. NO at 5 ppm with hyperoxia induced less cell death than 20 or 50 ppm (data not shown). Effects of NO and Hyperoxia on DNA Fragmentation and Apoptosis Induction of fibroblast DNA fragmentation was demonstrated by two TUNEL assay indicators. TUNEL

688

Vol. 262, No. 3, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

creased apoptosis (Fig. 6; n 5 4 – 6 for each group; p , 0.05). Similar observations were noted at 72 h following exposure; however, the differences were not statistically significant due to small sample numbers. Evidence of Peroxynitrite in NO-Exposed Fibroblasts Immunoblot analysis of fibroblasts exposed to NO 20/O 2 and NO 50/O 2 showed increased signal for S-nitrosotyrosine compared to RA controls, NO/RA or O 2 alone (Fig. 7). Coincubation of cells with SOD and SOD/catalase also produced a decrease in signal intensity compared to cells treated with NO and hyperoxia alone, while an effect of CAT alone was not obvious (Fig. 8). DISCUSSION FIG. 5. Effects of antioxidants on NO/O 2-induced fibroblast death. Viability in the presence of NO 20/O 2 and NO 50/O 2 was significantly increased by coincubation with superoxide dismutase (SOD), catalase (CAT), and SOD 1 CAT compared to controls (mean 6 SD, *p , 0.05). The effects of SOD 1 CAT with 20 ppm NO exposure were significantly greater than SOD or CAT alone (**p , 0.05).

MFI increased significantly in cells exposed to NO 20/O 2 and NO 50/O 2 when compared to RA at 24 h (n 5 3–5; for each group; p , .05) (Fig. 3). Similar trends were seen in cells exposed to NO at 50 ppm. A significant difference between the NO 20/O 2 and NO 50/O 2 groups was observed at 72 h (n 5 3–5 for each group; p , 0.05). Exposure to NO 50/O 2 also significantly increased the percentage of TUNEL-positive fibroblasts at 72 h compared to RA or O 2 alone (19 6 10%, 6 6 4% and 7 6 3%, respectively; p , 0.04, n 5 4 – 6 in each group) (Fig. 4A). Morphologically, fibroblasts exposed to NO/O 2 underwent changes characteristic of apoptosis (Figs. 2 and 4B). Unlike the normal spindle shape typically seen in control cells, NO-exposed fibroblasts were shrunken with bleblike appendages.

Our findings demonstrate that NO gas at clinically relevant concentrations in the presence of hyperoxia induces fibroblast death. Neither NO/RA nor O 2 alone significantly induced death, suggesting that either synergistic effects or a reaction product of NO and O 2 was responsible for producing this death. Oxidative stress is considered a major mediator of apoptosis in several cellular systems (12). Cell death in the milieu of NO and hyperoxia could be mediated by peroxynitrite formation. Peroxynitrite, a highly toxic oxidant formed by the reaction of nitric

Effects of Antioxidants on NO-Induced Fibroblast Death and DNA Fragmentation Incubation with SOD significantly decreased cell death at 24 h in fibroblasts incubated in 20 ppm NO with hyperoxia (Fig. 5; n 5 4 – 6 for each group; p , 0.05). Similar protection was seen with CAT treatment. Coincubation with SOD and CAT synergistically decreased NO/hyperoxia-associated fibroblast death compared to samples treated with CAT or SOD alone (n 5 4 – 6 for each group; p , 0.05). Similar findings were seen with fibroblasts exposed to NO 50/O 2 (n 5 4 – 6 for each group; p , 0.05). Incubation with CAT alone and SOD/CAT also significantly reduced fibroblast MFI, consistent with de-

FIG. 6. Effects of antioxidants on NO/O 2-induced fibroblast DNA fragmentation. Catalase (CAT) and superoxide (SOD) 1 CAT significantly decreased DNA fragmentation, expressed as percent of control mean fluorescence intensity (MFI; mean 6 SD), induced by NO 20/O 2 at 24 h following exposure (*p , 0.05). Decreases observed at 72 h were not statistically significant.

689

Vol. 262, No. 3, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 7. Evidence for peroxynitrite in NO-exposed fibroblasts. Western blot analysis shows increased signal intensity for S-nitrosotyrosine in fibroblasts exposed to (A) NO 20/O 2 and (B) NO 50/O 2 when compared to RA and O 2 alone.

oxide with superoxide, has been shown in high concentrations to induce apoptosis in primary murine fibroblasts (16) and human chondrocytes (12). SOD and CAT decreased NO/O 2-induced fibroblast cell death and DNA fragmentation. In these experiments, SOD and CAT could act a sink for superoxide and hydrogen peroxide radicals, thus decreasing the formation of peroxynitrite. This hypothesis is supported by our qualitative findings of decreased nitrosylated tyrosine on immunoblot analysis in SOD/CAT-treated fibroblasts. While not proving that peroxynitrite was the cause of cell death, the association between protective effects of antioxidants and decreased peroxynitrite implies such a possibility. Of note, CAT alone also showed a protective effect on fibroblast cell death. This effect would be expected to result directly from decreased hydrogen peroxide formation and would not be primarily a result of decreased interaction with NO, as would be the case with SOD-induced decreases in superoxide generation and peroxynitrate formation. It is therefore likely that CAT-mediated protection occurs through an effect on non-apoptotic hydrogen peroxide-mediated cell death. SOD contamination of CAT could also be responsible, but CAT used in these studies was not tested for SOD activity. However, the lack of effect of CAT alone on nitrosylated tyrosine formation (Western blot studies) provides indirect evidence that our CAT was not significantly contaminated with SOD. The difference in effect of SOD compared to CAT in these tryrosine assay results is predictable, given the known involvement of superoxide in the formation of peroxynitrite from NO.

DNA fragmentation and histologic findings suggest that NO and hyperoxia also increased cell death through apoptotic pathways. TUNEL assays demonstrated increased DNA fragmentation with NO and hyperoxia exposure at both 24 and 72 h. The lack of an appreciable difference in the amount of DNA fragmentation between these two time periods may have been due to progression of initially apoptotic cells to secondary necrosis in the absence of the typically rapid phagocytic clearance of apoptotic cells in vivo. While TUNEL results suggest the induction of apoptosis, the specificity of TUNEL for apoptosis has been questioned (21). However, morphologic evaluation of dying cells in our studies also suggests changes consistent with apoptosis. Fibroblast death in vivo may be more complicated. Human adult lung fibroblasts cultured with bronchoalveolar lavage fluid obtained from patients with ARDS more than 10 days after lung injury appeared to form mulberry-like structures with ruffled borders and contracted nuclei not completely consistent with either apoptosis or necrosis in the course of dying (22). Our findings are are also consistent with previous studies associating endogenous NO production and fibroblast apoptosis. Cellular senescence among embryonic rat lung fibroblasts has been observed with a dramatic increase in NO synthesis in response to the inflammatory cytokines interleukin-1b and tumor necrosis factor-a, and to lipopolysaccharide, and apoptosis can be blunted with NO synthase inhibition (23). Our study has several inherent limitations. We performed our experiments on an isolated human fibroblast cell line. It is uncertain whether NO exposure would have effects on the other resident cell types of the lung in a more complex milieu. Recent studies have shown differential effects of hyperoxia on the type and degreee of cell death in small airway epithelial cells compared to an alveolar epithelial cell line (24). Our laboratory has previously performed studies demonstrating that NO gas enhances DNA fragmentation in alveolar epithelial cells (25) and in human neutrophils (13). Further studies are necessary to determine the in vivo significance of these in vitro findings. Our experiments were conducted on adult human lung fibroblasts; neonatal fibroblasts in the setting of BPD could

FIG. 8. Effect of antioxidants on S-nitrosotyrosine formation in NO-exposed fibroblasts. Incubation of NO 20/O 2-exposed fibroblasts with both superoxide dismutase (SOD) and catalase (CAT), but not SOD or CAT alone, reduced Western blot anti-S-nitrotyrosine signal intensity compared to NO 20/O 2 alone.

690

Vol. 262, No. 3, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

potentially respond differently. We also used a relatively high concentration of oxygen, although for a short period of time (4 h). Further studies would be helpful to determine if apoptosis or cell necrosis predominates at lower oxygen concentrations for longer periods of duration in the presence of NO. The effects of lower concentrations of oxygen on other resident cell types may be of greater relevance in the clinical setting. Similarly, prolonged exposure to lower NO concentrations may cause greater DNA fragmentation than that seen in our studies. The potential significance for these findings is twofold. With increasing interest in the use of inhaled NO, it is important to determine functional effects and potential toxicity on resident cells. If found to have a differential effect on fibroblast viability and mode of death, inhaled NO could potentially serve as an agent to arrest excessive fibroproliferative response leading to organ dysfunction after injury. However, any positive effect of inhaled NO would have to be balanced with the negative effects on other resident cells. Further in vivo studies in models of fibrotic injury will be helpful to assess whether NO is harmful or beneficial in subacute lung injury. ACKNOWLEDGMENTS The authors thank Frank Harris for his technical assistance. This study was funded by an American Lung Association grant (J.D.F.) and by the Goddard Research Scholarship of Emory Egleston Children’s Research Center (J.D.F.).

REFERENCES 1. Chen, B., Polunovsky, V., White, J., Blazar, B., Nakhleh, R., Jessurun, J., Peterson, M., and Bitterman, P. (1992) J. Clin. Invest. 90, 1778 –1785. 2. Wolff, G., and Crystal, R. G. (1997) in The Lung: Scientific Foundations, pp. 2509 –2524. Lippincott-Raven, Philadelphia. 3. Abman, S. H., and Groothius, J. R. (1994) Pediatr. Clin. North Am. 41, 277–315.

4. Henke, C., Bitterman, P., Roongta, U., Ingbar, D., and Polunovsky, V. (1996) Am. J. Pathol. 149, 1639 –1650 5. Meduri, G. U., Chinn, A. J., Leeper, K. V., Wunderink, R. G., Tolley, E., Winer-Muram, H. T., Khare, V., and Eltorky, M. (1994) Chest 105, 1516 –27. 6. Bitterman, P. B. (1992) Am. J. Med. 92(Suppl. 6A), 39 – 43. 7. Zapol, W. M., Rimar, S., Gillis, N., Marletta, M., and Bosken, C. H. (1994) Am. J. Respir. Crit. Care Med. 149, 1375–1380. 8. Moncada, S., and Higgs, E. A. (1995) FASEB J. 9, 1319 –1330. 9. Rodeberg, D. A., Chaet, M. S., Bass, R. C., Arkovitz, M. S., and Garcia, V. F. (1995) Am. J. Surg. 170, 292–303. 10. Mizutani, T., and Layon, A. J. (1996) Chest 110, 506 –524. 11. Zhang, H., Gharaee-Kermani, M., and Phan, S. H. (1997) J. Immunol. 158, 1392–1399. 12. Blanco, F. J., Ochs, R. L., Schwarz, H., and Lotz, M. (1995) Am. J. Pathol. 146, 75– 85. 13. Daher, A. H., Fortenberry, J. D., Owens, M. L., and Brown, L. A. (1997) Am. J. Respir. Cell Mol. Biol. 16, 407– 412. 14. Fortenberry, J. D., Owens, M. L., Brown, M. R., Atkinson, D. , and Brown, L. A. (1998) Am. J. Respir. Cell Mol. Biol. 18, 421– 428. 15. Kooy, N. W., Royall, J. A., Ye, Y. Z., Kelly, D. R., and Beckman, J. S. (1995) Am. J. Respir. Crit. Care Med. 151, 1250 –1254. 16. Shin, J. T., Barbeito, L., MacMillan-Crow, L. A., Beckman, J. S., and Thompson, J. A. (1996) Arch. Biochem. Biophys. 335, 32– 41. 17. McKeehan, W. L., and Ham, R. G. (1977) Dev. Biol. Standard 37, 97–108. 18. Gavrieli, Y., Sherman, Y., and Ben-Sasson, S. A. (1992) J. Cell Biol. 119, 493–501. 19. Gorcyza, W., Gong, J., and Darzynkiewicz, Z. (1993) Cancer Res. 53, 1945–1951. 20. Peng, H., Libby, P., and Liao, J. K. (1995) J. Biol. Chem. 270(23), 14214 –14219. 21. Hetts, S. W. (1998) JAMA 279, 300 –307. 22. Polunovsky, V. A., Chen, B., Henke, C., Snover, D., Wendt, C., Ingbar, D., and Bitterman, P. B. (1993) J. Clin. Invest. 92, 388 –397. 23. Lavnikova, N., and Laskin, D. L. (1995) J. Leukocyte Biol. 58, 451– 458. 24. Jyonouchi, H., Sun, S., Abiru, T., Chareancholvanich, S., and Ingbar, D. (1998) Am. J. Respir. Cell Mol. Biol. B 426 – 436. 25. Piazza, A., Harris, F., Brown, L., and Jain, L. (1998) Pediatr. Res. B 294A. [Abstract]

691