American Journal of Respiratory and Critical Care Medicine

Rationale: The resolution of pulmonary inflammation seen in various inflammatory lung conditions depends on the clearance of apoptotic cells to prevent permanent tissue damage or progressive disease. Uptake of apoptotic cells by alveolar macrophages is suppressed by oxidants through the activation of Rho signaling.

Objectives: We hypothesized that antioxidant exposure would increase the ability of alveolar macrophages to clear pulmonary apoptotic cells through the inhibition of RhoA.

Methods: The effects of the antioxidant N-acetylcysteine (NAC) on the pulmonary immune response were seen in mice treated intratracheally with LPS, LPS + NAC, or saline. Apoptotic cell clearance, RhoA activity, and changes in the lung inflammatory responses were analyzed in vivo or ex vivo.

Measurements and Main Results: Neutrophil accumulation, apoptosis, necrosis, and oxidant production peaked at 3 days post LPS treatment. NAC enhanced the clearance of apoptotic cells and inhibited RhoA activity in alveolar macrophages at 3 days post LPS treatment. NAC suppressed LPS-induced proinflammatory mediators, enhanced the production of transforming growth factor-β1, reduced the accumulation of inflammatory cells, and reduced levels of protein and lactate dehydrogenase in bronchoalveolar lavage fluid. In the presence of ex vivo apoptotic cells, alveolar macrophages exposed to LPS or LPS + NAC had reduced tumor necrosis factor-α levels and increased transforming growth factor-β1 levels. A Rho kinase inhibitor mimicked the effects of NAC on the clearance of apoptotic cells and the inflammatory responses.

Conclusions: These results indicate that NAC can expedite the resolution of LPS-induced pulmonary inflammation through the inhibition of RhoA activity and the enhancement of apoptotic cell clearance.

Scientific Knowledge on the Subject

Uptake of apoptotic cells by alveolar macrophages is suppressed by oxidants through the activation of Rho signaling. However, this finding has not been examined in an in vivo model of lung inflammation associated with oxidant stress.

What This Study Adds to the Field

There is a relative suppression of apoptotic cell clearance under oxidant stress in acute lung injury. N-acetylcysteine promotes apoptotic cell clearance through down-regulation of the RhoA/Rho kinase pathway, resulting in resolution of lung inflammation.

The rapid accumulation of neutrophils in the lung induced by a proinflammatory stimulus is a central process in the pathogenesis of various inflammatory lung diseases (1). The mechanisms underlying the resolution of the inflammatory cell infiltrate and fibrosis are unclear. Apoptosis is believed to be a major mechanism for the clearance of neutrophils from sites of inflammation. Recently, Medan and colleagues (2) reported that in LPS-treated mice, the number of apoptotic neutrophils increased concomitantly with the influx of neutrophils into the lung. In an animal model of LPS-induced intense lung inflammation, Rydell-Törmänen and colleagues (3, 4) showed direct evidence that extensive neutrophil infiltration was followed by apoptosis and secondary necrosis of neutrophils. If the apoptotic neutrophils are not cleared away efficiently, they undergo secondary necrosis resulting in induction of the proinflammatory cascade (5). The failure of this highly conserved process of apoptotic cell clearance may contribute to pulmonary disease pathogenesis by impeding both the resolution of inflammation and the repair of alveolar integrity (6).

Reactive oxygen species (ROS) produced by activated neutrophils, alveolar macrophages, and stimulated pulmonary epithelial and endothelial cells play a major role in inflammatory lung injury. Previously, McPhillips and colleagues (7) demonstrated that tumor necrosis factor (TNF)-α specifically inhibits macrophage clearance of apoptotic cells in vitro through oxidant-dependent activation of the guanosine triphosphatase (GTPase), Rho, which is a key negative regulator of phagocytosis. This finding has not been examined in an in vivo model of intense lung inflammation associated with oxidant stress.

N-acetylcysteine (NAC) is a thiol compound that can replenish intracellular glutathione and act as a direct scavenger of ROS. The effectiveness of NAC administration in animal models of sepsis- or bleomycin-induced lung injury has been previously reported (8, 9). In the current study, the time course of neutrophil accumulation, oxidant production, apoptosis, and necrosis of bronchoalveolar lavage (BAL) cells and alveolar macrophages was analyzed in an attempt to characterize the intensive lung inflammatory response to intratracheal LPS treatment. The in vivo effects of an antioxidant, NAC, on apoptotic cell clearance and RhoA activity in alveolar macrophages, the changes in pro- or antiinflammatory mediator production, and inflammatory cell accumulation during intense lung inflammation associated with oxidant stress also were studied. Furthermore, the role of the RhoA/Rho kinase pathway in apoptotic cell clearance and the inflammatory responses was confirmed in this mouse model using a Rho kinase inhibitor, Y-27632.

See the online supplement for detailed methods.

Animal Protocols

Pathogen-free male BALB/C mice (Daehan Biolink, Eumsung, South Korea) weighing 20 to 25 g were used in all experiments. The Animal Care Committee of the Ewha Medical Research Institute approved the experimental protocol.

Mouse pharyngeal aspiration was used for administration of LPS (4.5 mg/kg in 30 μl) (10). Animals were given NAC (200 mg/kg), intraperitoneally 2 hours before LPS treatment and then every 12 hours (11, 12) and killed at 1, 3, or 6 days post LPS treatment. For RhoA/Rho kinase pathway inhibition experiments, a Rho kinase inhibitor, Y-27632 (1 or 10 mg/kg) was administered intraperitoneally 30 minutes before LPS treatment and once a day thereafter (13). Animals were killed at 3 days post LPS treatment.

BAL Cells, Lung Tissue, and Cell Counts

BAL was done, and cell counts were determined using an electronic Coulter Counter fitted with a cell sizing analyzer (Coulter Model ZBI with a channelizer 256; Coulter Electronics, Bedfordshire, UK) (14). BAL cells were isolated and cytospins were made to assess the phagocytic index (PI) (15).

Primary Cell Isolation and Culture

Suspended alveolar macrophages were more than 95% viable, as determined by trypan blue dye exclusion. Alveolar macrophages were isolated by adhesion (60 min) and cultured in serum-free X-vivo medium (Biowhittaker, Wakersville, MA). Neutrophils were isolated and prepared from normal blood using Percoll gradient centrifugation, as previously described (16).

Cell Lines and Culture

Murine J774 macrophages (American Type Culture Collection, Rockville, MD) were cultured in Dulbecco's modified Eagle's medium (DMED) supplemented with 10% heat-inactivated FBS plus penicillin-streptomycin-glutamine at 37°C in 10% CO2. A human leukemia Jurkat T-cell line (American Type Culture Collection) was cultured in RPMI 1640 containing 10% heat-inactivated fetal bovine serum (FBS) supplemented with penicillin-streptomycin-glutamine at 37°C in 5% CO2.

Histone-Bound DNA Fragmentation Assay of BAL Cells

BAL cells (105/well) were analyzed for histone-bound DNA fragments using an ELISA kit (Roche Molecular Biochemicals, Indianapolis, IN) as detailed in the online supplement.

Flow Cytometry

To measure cell membrane changes associated with apoptosis, BAL cells were incubated with annexin V-FITC (Annexin V Apoptosis Detection Kit; PharMingen, Becton Dickinson, San Diego, CA) and propidium iodide for 15 minutes in the dark. The cells were analyzed on a FACSCalibur system (BD Biosciences, San Jose, CA) with CellQuest software (BD Biosciences) and ModFitLT software (Verity Software House, http://www.vsh.com). See online supplement for details.

Lactate Dehydrogenase Assays

BAL fluid lactate dehydrogenase (LDH) activities were determined by monitoring the LDH-catalyzed oxidation of pyruvate coupled with the reduction of nicotinamide adenine dinucleotide (NAD) at 340 nm using a commercial kit (Roche Diagnostic Systems, Montclair, NJ). See online supplement for details.

Viability Assays of Alveolar Macrophages

Suspended alveolar macrophages were stained with trypan blue (0.08%) and examined with light microscopy. DNA fragmentation of apoptotic alveolar macrophages was detected using terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) kit (Promega, Madison, WI). See online supplement for details.

ROS Production

BAL cells were stained with dichlorodihydrofluorescein diacetate (H2DCF-DA) (Molecular Probes, Eugene OR). DCF fluorescence intensity was measured as previously described (17). Alveolar macrophages on glass coverslips were stained with dihydroethidium (DHE). Images were collected, and the fluorescent mean intensity of DHE staining was measured as previously described (18). Oxidant production in zymosan A-stimulated alveolar macrophages was measured using DCF-DA assays. Unopsonized zymosan A has been shown to stimulate oxidant production from alveolar macrophages but not neutrophils (19, 20). See online supplement for details.

Induction of Apoptosis

Human neutrophils were stained with 2 μM calcein-AM (Molecular Probes, Inc., Eugene, OR) for 20 minutes in phosphate-buffered saline before inducing apoptosis by 24-hour culture in RPMI with 10% FBS (21). This method allowed us to distinguish them from the endogenous apoptotic neutrophils when the PI was determined. See online data supplement for additional details. Apoptosis of Jurkat cells was induced by ultraviolet irradiation as previously described (22).

In Vitro Phagocytosis Assays

Alveolar macrophages (105) from mice treated with saline, LPS, or LPS + NAC were plated in each well of a four-chamber slide glass plate at 37°C for 2 hours. Primary alveolar macrophages from naive mice and J774 macrophages (105 AM/24-well plates) were treated with NAC (0.5 mM) or Y-27632 (10 μM) for 30 minutes before LPS (10 μg/ml) or H2O2 (10 μM) stimulation for 20 minutes (7). Apoptotic human neutrophils labeled fluorescently or Jurkat T cells were added to macrophage cultures in a ratio 10:1 in 1 ml of medium and incubated for 90 minutes. Phagocytosis was quantified using the PI previously described (23). See online supplement for details.

Rho Activity Assays

RhoA activity was measured in the lysates collected from alveolar macrophages (106/well) using an ELISA-based RhoA Activation Assay Biochem Kit (G-LISA; Cytoskeleton, Denver, CO) according to the manufacturer's instructions. Rho activation pull-down assay was performed in lung tissue homogenate according to the manufacturer's indications (Millipore, Temecula, CA). The methods are described in the online supplement.

ELISAs

TNF-α, macrophage inflammatory protein (MIP)-2, and transforming growth factor (TGF)-β1 protein levels were measured from the first BAL fluid or the 18-hour supernatants of cultured alveolar macrophages. The activity of myeloperoxidase (MPO) in lung homogenates was estimated by using a specific ELISA kit. See online supplement for details.

Zymographic Analysis of Matrix Metalloproteinase-9

The gelatinolytic activities of BAL fluid were determined by zymography as described previously (24). See online supplement for details.

Measurement of Total Protein

BAL sample protein concentration was used as an indicator of blood-pulmonary epithelial cell barrier integrity. Total protein was measured according to the method of Hartree (25).

Lung Histology

Lungs were fixed with 4% formalin and embedded in optimal cutting temperature mold. Sections (5 μm) were stained with hematoxylin and eosin. See online supplement for details.

Western Blot

Lung tissue homogenate samples and total cell lysates were separated on 10% SDS-polyacrylamide gels. Separated proteins were electrophoretically transferred onto nitrocellulose paper. The membranes were probed with specific antibodies to phospho-myosin phosphatase target subunit (MYPT-1) at threonine 853, phospho-IκB-α (Ser32) and IκB-α and visualized by chemiluminescence (ECL). See online supplement for details.

Nuclear Extracts

Nuclear extracts from lung tissue were prepared using a modification of the method described by Lee and colleagues (26).

Electrophoretic Mobility Shift Assay

Electrophoretic mobility shift assay was performed as reported previously (26). As a probe for electrophoretic mobility shift assay, double-stranded DNA fragment, containing the nuclear factor (NF)-κB consensus sequence (5′-CCTGTGCTCCGGGAATTTCCCTGGCC-3′), was used and labeled with [α-32P]-dATP (Amersham, Buckinghamshire, UK) using DNA polymerase Klenow fragment (Life Technologies, Gaithersburg, MD). See online supplement for details.

Statistical Analysis

Values are expressed as mean ± SEM. Analysis of variance was applied for multiple comparisons, and Tukey post hoc test was applied where appropriate. Student t test was used for comparisons of two sample means. A P value of less than 0.05 was considered statistically significant.

BAL Inflammatory Cell Yield, Apoptosis, and Necrosis of BAL Cells after LPS Treatment

A high dose of LPS (4.5 mg/kg body weight) was administered intratracheally to BALB/c mice to model intense lung inflammation. Neutrophil accumulation occurred in the lungs within 24 hours, peaked at 3 days post LPS treatment, and declined by Day 6 (Figure 1A). The quantity of alveolar macrophages in BAL fluid increased at 3 days and slightly declined by 6 days post LPS treatment (Figure 1A).

Apoptosis in BAL cells from LPS-treated mice were quantified with a histone-bound DNA fragmentation assay. LPS-induced apoptosis peaked at 3 days post LPS treatment (Figure 1B). To measure cell membrane changes associated with apoptosis, BAL cells were labeled with annexin V and propidium iodide and analyzed by flow cytometry. Similarly, the percentage of apoptotic BAL cells (annexin V positive/propidium iodide negative) (41%) maximally increased at 3 days post LPS treatment (Figures 1C and 1D). Necrosis of BAL cells was determined by percentage of late apoptotic/necrotic cells (positive in both annexin V and propidium iodide signals analyzed by flow cytometry) (Figures 1C and 1E) and LDH activity in BAL fluid (Figure 1F). Both assays yielded similar results that necrosis was slightly developed at 1 day and peaked at 3 days post LPS treatment.

The viability of alveolar macrophages was assessed by trypan blue dye exclusion (Figures 1G and 1H), and TUNEL assay was used to detect DNA fragmentation, which is an early marker of apoptosis (Figures 1I and 1J). Alveolar macrophages from saline or LPS-treated mice showed low levels of cells positively stained with trypan blue, consistent with high viability, and low levels of DNA fragmentation by TUNEL assay 1 to 6 days post treatment. The earlier finding showed a great increase in the number of neutrophils after the LPS treatment (Figure 1A). These data suggest that apoptotic and necrotic cells in BAL fluid are mainly recruited neutrophils, but not alveolar macrophages.

Reduction of ROS Production after NAC Treatment

Intracellular ROS production was directly measured using a fluorescence probe (H2DCF-DA) in BAL cells. ROS production in BAL cells from LPS-treated mice rapidly increased at 1 day, peaked at 3 days, and declined at 6 days post LPS treatment (Figure 2A). Intracellular ROS production in alveolar macrophages from LPS-treated mice was qualitatively measured in the presence of the DHE probe, using confocal microscopy. The levels of nuclear fluorescence in alveolar macrophages markedly increased at 3 days post LPS treatment (Figure 2B). Using the laser-scanning microscope (LSM) image examiner software, intracellular oxidant staining levels were quantified. Oxidant staining reached peak fluorescence at 3 days (3.9-fold versus saline control) and declined at 6 days post LPS treatment (2.7-fold versus saline control) (Figure 2C). We also measured ROS production in alveolar macrophages taken from mice at 1 to 6 days after in vivo stimulation with LPS (with or without NAC treatment), which were then secondarily stimulated in vitro with zymosan. We observed that alveolar macrophages stimulated with zymosan significantly increased their production of ROS over levels produced by the saline-treated group at 1 to 6 days post LPS (P < 0.05) (Figure 2D). Similarly, peak ROS production in zymosan-stimulated alveolar macrophages was seen at 3 days post LPS treatment. Taken together, this series of experiments demonstrates that oxidant production in BAL cells and alveolar macrophages or zymosan-stimulated alveolar macrophages after the LPS treatment increased and peaked at 3 days post LPS treatment. However, inoculation with NAC before and after LPS treatment significantly reduced the production of ROS from BAL cells (Figure 2A), alveolar macrophages (Figures 2B and 2C), and alveolar macrophages on in vitro stimulation with zymosan (Figure 2D) at each time point (P < 0.05), and maximal inhibition in all cell types was seen at 3 days post LPS.

NAC Treatment Enhances Apoptotic Cell Clearance by Alveolar Macrophages

We then looked at whether apoptotic cell uptake by alveolar macrophages was altered by changes in oxidant levels during intense lung inflammation. Similar to the findings of Janssen and colleagues (27), the PI in BAL cytospin alveolar macrophages obtained from LPS-stimulated lungs was increased over the levels seen with saline controls at 1 day and 3 days post LPS treatment (Figures 3A and 3B). Furthermore, NAC treatment significantly increased the PI in BAL cytospin alveolar macrophages at 3 days post treatment (compared with LPS treatment alone, P < 0.05). These data suggest that phagocytic efficiency of alveolar macrophages significantly increased at Day 3 post treatment because the antioxidant activity of NAC also peaked at this time point. This caused a critical change in the inflammatory environment that made it less oxidant rich. Thus, phagocytic activities of alveolar macrophages obtained on Days 1 through 6 post LPS treatment were assessed ex vivo. This approach has the advantage that it controls for the number and ratio of alveolar macrophages to apoptotic cells (28). Freshly obtained alveolar macrophages from saline or LPS-treated mice were cocultured with apoptotic human neutrophils labeled fluorescently, to distinguish them from the endogenous apoptotic neutrophils. The PI in alveolar macrophages taken from LPS-stimulated lungs was significantly enhanced ex vivo compared with saline controls at 3 days post LPS treatment (P < 0.05). Exposure to NAC significantly increased the ability of alveolar macrophages to phagocytose apoptotic neutrophils ex vivo by 3 days post LPS treatment (P < 0.05) (Figures 3C and 3D).

NAC Treatment Enhances Apoptotic Cell Clearance through Inhibition of RhoA Activity in Alveolar Macrophages

Previous studies indicate that oxidant-dependent activation of RhoA negatively regulates apoptotic cell clearance in macrophage cell lines (7, 29). Thus, RhoA activation during oxidant-rich, intense pulmonary inflammation was examined. RhoA activity in alveolar macrophages was significantly increased at 3 days post LPS treatment (2.8-fold) (P < 0.05), and it declined at 6 days (Figure 4A). NAC treatment significantly inhibited the LPS-induced increase in RhoA activity in alveolar macrophages by 51% at 3 days post LPS treatment (P < 0.05) (Figure 4A). Similarly, a significant increase in the amount of active guanosine triphosphate–bound Rho in lung tissue from LPS-treated mice occurred only at 3 days post LPS treatment (P < 0.05) (Figure 4B). This increase was significantly suppressed by NAC treatment (P < 0.05).

Previous in vivo and ex vivo phagocytosis assays demonstrate that because apoptotic cell clearance actually increases post LPS treatment in mice, there is a relative suppression of apoptotic cell clearance under oxidant stress in acute lung injury (Figures 3A–3D). Thus, the relationship between RhoA activity and relatively restrained phagocytic activity of alveolar macrophages was assessed using Y-27632 to inhibit the downstream effector, Rho kinase. Mice were treated with Y-27632 that promoted apoptotic cell uptake by alveolar macrophages in a dose-dependent manner at 3 days post LPS treatment (Figure 4C). These data indicate that oxidant-dependent activation of RhoA/Rho kinase pathway is responsible for a relative suppression of clearance of apoptotic cells in acute lung injury.

To determine inhibitory effect of Y-27632 on the activation of Rho kinase, the phosphorylation of myosin phosphatase target subunit-1 (MYPT-1) was measured because Rho kinase phosphorylates its substrate, MYPT-1 at threonine 853 (30). LPS treatment induced a significant increase in phosphorylation of MYPT-1 at Thr853 in alveolar macrophages (Figure 4D) and lung tissue (Figure 4E). These increases were substantially blocked by Y27632 treatment, indicating inhibition of Rho kinase activity.

The increase in apoptotic cell clearance by macrophages induced by NAC or a Rho kinase inhibitor was confirmed in vitro when alveolar macrophages taken from naive mice (Figure 4F) or a murine macrophage cell line, J774 (Figures 4G and 4H), were exposed to LPS or H2O2.

NAC Treatment Reduces Proinflammatory Mediators but Increases TGF-β1 in BAL Fluid

Protein levels of the proinflammatory cytokine, TNF-α, and a chemokine, MIP-2, were measured by ELISA. Levels of these cytokines in BAL fluid were greatest at 1 day after LPS treatment, remained significantly increased at 3 days after LPS treatment, and declined to basal levels 6 days after LPS treatment. NAC significantly suppressed LPS induction of TNF-α (approximately 45% inhibition at 1 day and 3 days post LPS treatment) (Figure 5A) and MIP-2 (approximately 51 and 66% inhibition at 1 day and 3 days post LPS treatment, respectively) in BAL fluid (P < 0.05) (Figure 5B). Matrix metalloproteinase (MMP)-9 activity in BAL fluid was analyzed by gelatin zymography. Peak MMP-9 activity induced by LPS stimulation occurred at 3 days post LPS treatment. NAC treatment significantly inhibited LPS-induced MMP-9 activity in BAL fluid (32 and 42% inhibition at 1 and 3 days post LPS, respectively, P < 0.05) (Figure 5C). The levels of MMP-9 activity in all treatment groups rapidly declined after 6 days post LPS treatment.

Production of TGF-β1, an antiinflammatory mediator, in BAL fluid from LPS-treated mice was 2.2-fold higher than controls at 3 days and slightly declined by 6 days after LPS treatment (Figure 5D). NAC treatment further increased LPS-induced TGF-β1 production in BAL fluid by 68, 96, and 112% at 1, 3, and 6 days post LPS treatment, respectively.

NAC Treatment Reduces TNF-α Production but Enhances TGF-β1 by Alveolar Macrophages Cocultured with Apoptotic Cells

In vitro uptake of apoptotic cells by macrophages actively suppresses LPS-induced production of proinflammatory cytokines, but enhances TGF-β1 production (31). Because the PI was also enhanced when alveolar macrophages from the LPS + NAC group were cocultured with apoptotic cells at 3 days post LPS, changes in TNF-α and TGF-β1 production were assayed from alveolar macrophages in the presence of apoptotic cells. Alveolar macrophages taken from mice at 3 days post LPS treatment produced less TNF-α when exposed to apoptotic cells than those exposed to viable cells or media only (Figure 5E). Similarly, TNF-α levels produced by alveolar macrophages from mice treated with LPS + NAC were further reduced on apoptotic cell exposure. However, TGF-β1 levels in alveolar macrophages from the LPS or the LPS + NAC group at 3 days post LPS treatment were further increased when exposed to apoptotic Jurkat T cells (Figure 5F).

Reduction in Inflammatory Cell Recruitment, and Levels of Protein and LDH in BAL Fluid after NAC Treatment

NAC treatment significantly inhibited LPS-induced increases in intraalveolar neutrophils and macrophages at 1 to 6 days post LPS treatment (P < 0.05) (Figures 6A and 6B). Inhibition of inflammatory cell accumulation was also confirmed by pulmonary levels of MPO, indicating neutrophil infiltration in the lung parenchyma. NAC treatment reduced MPO activities in lung tissue after LPS treatment at 1 to 6 days (Figure 6C). NAC treatment also inhibited LPS-induced increases in levels of BAL protein and LDH at days 1 to 6 (Figures 6D and 6E). Histological sections of lung tissue from these mice 3 days after LPS treatment confirmed BAL findings. Hematoxylin and eosin staining of lung tissue fixed with formalin revealed significant reductions in parenchymal and intraalveolar cells in the lungs after LPS + NAC treatment compared with those treated with LPS alone (Figure 6F).

Reduction in Inflammatory Responses after Treatment with Rho Kinase Inhibitor

Present data suggest that the down-regulation of RhoA activity induced by NAC treatment is required for the enhancement of apoptotic cell clearance leading to antiinflammatory signaling. Experiments were performed with the Rho kinase inhibitor Y-27632 to confirm whether directly blocking RhoA/Rho kinase pathway leads to antiinflammatory cell signaling. Levels of inflammatory mediators and antiinflammatory cytokine TGF-β1 in BAL fluid were analyzed at 3 days post LPS treatment. Treatment with Y-27632 suppressed LPS-induced production of TNF-α and MIP-2 and MMP-9 activity (Figures 7A–7C). However, the Rho kinase inhibitor enhanced TGF-β1 production in a dose-dependent manner (Figure 7D).

Y-27632 treatment also inhibited the other lung inflammatory responses, such as increases in intraalveolar neutrophils and macrophages, MPO activity in lung tissue, BAL protein, and LDH activity at 3 days post LPS treatment in a dose-dependent manner (Figures 7E–7I). Histological findings confirmed reduction in alveolar and parenchymal infiltration of inflammatory cells by Y-27632 treatment at 3 days post LPS (Figure 7J). These results indicate that a Rho kinase inhibitor has similar effects on the lung inflammatory response as those seen after NAC treatment.

NAC Reduces Effects on IκB-α and NF-κB Induced at Early Time Points by LPS

LPS induces NF-κB activation in lungs that peaks after 2 hours and is then down-regulated (24). Activation of NF-κB is regulated by redox-sensitive signaling pathways (32). Thus, it is likely that LPS-induced proinflammatory mediators would be up-regulated by oxidant-dependent NF-κB activation at 1 day post LPS treatment. The kinetics of NF-κB activation in lung tissue and the pathways leading to NF-κB activation were analyzed. LPS-induced activation of NF-κB significantly increased at Day 1, but was completely down-regulated at 3 days post LPS treatment (Figures 8A and 8B). Likewise, phosphorylation and degradation of IκB-α were significantly increased at 1 day, but not at 3 days, after LPS treatment (Figures 8C–8F). The inhibitory effects of NAC on NF-κB activation and the phosphorylation and degradation of IκB-α were only seen 1 day after LPS treatment, and this suggests that increases in the production of proinflammatory mediators at 1 day post LPS treatment are mediated through oxidant-dependent activation of the NF-κB pathway. Additionally, NAC treatment did not impact apoptotic cell clearance at this time point. Thus, the suppressive effects of NAC on LPS-induced inflammatory responses may depend more on the down-regulation of NF-κB pathways rather than the up-regulation of apoptotic cell clearance at earlier time points.

Based on previous in vitro observation of TNF-α–induced oxidant production leading to increased active Rho and decreased clearance of apoptotic cells by macrophages (7), we hypothesized that antioxidant exposure would increase the ability of alveolar macrophages to clear pulmonary apoptotic cells through the inhibition of RhoA and consequently induce antiinflammatory process. To investigate this hypothesis, we used a mouse model of LPS-induced acute lung injury and examined the in vivo effects of an antioxidant, NAC, on apoptotic cell clearance, RhoA activity in alveolar macrophages, and the associated changes in lung inflammatory responses at Days 1 to 6 after intratracheal treatment with LPS. Furthermore, we also focused on the role of RhoA/Rho kinase pathway in apoptotic cell clearance and the changes in inflammatory responses in this model using an inhibitor of Rho kinase (Y-27632).

Our time-course study indicates that neutrophil accumulation, apoptosis and necrosis of BAL cells (mainly neutrophils), and ROS activity in BAL cells and alveolar macrophages peaked at 3 days after intratracheal LPS treatment. Clearance of apoptotic cells by alveolar macrophages from LPS-treated mice was enhanced 1 and 3 days post LPS treatment. This may be because apoptosis of recruited neutrophils occurs quite rapidly and alveolar macrophages may be globally activated in response to LPS at these time points (33). However, consistent with electron microscopic findings from LPS-exposed mice (3, 4), unengulfed apoptotic neutrophils were found in several microscopic fields of BAL cytospins particularly at Day 3, indicating that clearance of apoptotic cells by alveolar macrophages may not be sufficient after acute intense lung inflammation.

Data from our present study and that of others (2) showed that the number of apoptotic neutrophils in BAL fluid increased concomitantly with the increase in neutrophil influx and the levels of necrosis in BAL cells (mainly neutrophils) changed correspondingly to those of apoptosis and peaked at the same exposure day in a mouse model of LPS-induced lung inflammation. Similarly, by using the same LPS model, Rydell-Törmänen and colleagues (3) showed direct evidence in BAL studies and lung histology that neutrophil infiltration was followed by apoptosis and secondary necrosis of neutrophils. These important findings imply that neutrophil apoptosis may exacerbate rather than resolve lung inflammation if apoptotic cells were not efficiently cleared. Unlike these murine LPS models, the proportion of apoptotic neutrophils in BAL fluid from patients with acute respiratory distress syndrome (ARDS) was low throughout the course, perhaps because of the presence of antiapoptotic factors such as granulocyte colony-stimulating factor, granulocyte-macrophage colony-stimulating factor, and possibly IL-8 and IL-2 (34, 35). The antiapoptotic effect of ARDS BAL fluid is highest during early ARDS and decreases during late ARDS (36). Thus, it is possible that the neutrophils undergo secondary necrosis or are rapidly cleared by alveolar macrophages at later stages of human ARDS. Further studies are needed to demonstrate a definitive causal relationship between changes in the phagocytic system in alveolar macrophages associated with the rate of neutrophil apoptosis and necrosis and outcome in human ARDS.

Previous in vitro data indicate that endogenous and exogenous oxidants can inhibit uptake of apoptotic cells by macrophages (7, 37) and a variety of antioxidants normalized apoptotic cell uptake (7). Data from our in vivo study support this finding because in vivo treatment with NAC inhibited oxidant levels in BAL cells and alveolar macrophages at 3 days post LPS, and at this same time point, in vivo and ex vivo apoptotic cell clearance by alveolar macrophages was markedly enhanced. Our in vitro models wherein alveolar macrophages taken from naive mice and a murine macrophage cell line, J774, were exposed to LPS or H2O2 confirmed the enhancement of apoptotic cell clearance by exposure to NAC.

LPS stimulation induces oxidant up-regulation through the interaction of NADPH oxidase with cytosolic Toll/IL-1R regions of Toll-like receptor 4 (TLR4) (38), through effects on cytosolic phospholipase A2, and through the release of arachidonate, which seem to be affected by the induction of TNF-α production (7). Oxidants reportedly lead to the activation of Rho family GTPases, including RhoA, Rac1, and cdc-42 (39, 40). Heo and Camphell (39) identified a unique redox-active motif (GXXXXGK[ST]C) present in 50% of all Rho GTPases that is important for redox-mediated regulation of guanine nucleotide exchange activity. Several reports indicate that redox signaling leads to activation or deactivation of RhoA, depending how phosphorylation or dephosphorylation of p190RhoGAP occurs (4143).

Alternatively, several lines of evidence indicate ROS induces an increase in RhoA activation and an inhibition of apoptotic cell uptake (7, 31). TNF-α (7) and pyocyanin, a toxic metabolite of Pseudomonas aeruginosa (44), have been reported to inhibit uptake of apoptotic cells by murine alveolar macrophages, a murine macrophage cell line, and human monocyte-derived macrophages through oxidant-dependent activation of RhoA. Recently, work by Richens and colleagues (28) supported this concept in in vivo cigarette smoke exposure models. In this study, we propose that a critical level of oxidants is required for RhoA activation in alveolar macrophages and lung cells during LPS-induced intense lung inflammation. This is supported by the fact that significant increases in RhoA activity were observed only at 3 days post LPS treatment, when the oxidant levels in BAL cells and alveolar macrophages peaked. Down-regulation of oxidant-dependent RhoA activity by NAC contributed to the enhancement of apoptotic cell clearance at 3 days post LPS treatment. Additionally, in vivo studies using a Rho kinase inhibitor Y-27632 showed a direct impact of the RhoA/Rho kinase pathway on the insufficient apoptotic cell uptake at this time point. Enhancement of apoptotic cell clearance by the Rho kinase inhibitor was also induced in in vitro systems of primary murine alveolar macrophages and J774 macrophages treated with LPS or H2O2. Thus, our in vivo and in vitro data suggest that relative suppression of apoptotic cell clearance in acute lung injury is induced by oxidant-dependent activation of RhoA/Rho kinase.

In vitro and in vivo experiments on macrophages from LPS-stimulated lungs that ingest apoptotic cells demonstrate that this activity suppresses proinflammatory processes while up-regulating the production of antiinflammatory mediators TGF-β1 and PGE2 (31, 45, 46). In the present study, we found that in vivo NAC treatment also inhibits LPS induction of proinflammatory mediator production and activity (TNFα, MIP-2, and MMP-9), but enhances an antiinflammatory mediator, TGF-β1, in BAL fluid at 1 to 6 days post LPS. Furthermore, up-regulation of TGF-β1 induction and down-regulation of TNF-α secretion were also shown when alveolar macrophages from LPS + NAC mice were exposed to ex vivo apoptotic cells. These results are consistent with previous findings, which demonstrated an increase in TGF-β1 induction from peritoneal macrophages stimulated with thioglycollate after in vivo injection of apoptotic cells (45). Collectively, these data suggest that inhibition of oxidant stress by NAC treatment results in active antiinflammatory signals in intense lung inflammation through the up-regulation of apoptotic cell clearance, resulting in TGF-β1 induction.

Likewise, in vivo treatment with NAC resulted in a significant reduction of inflammatory cells, such as neutrophils and alveolar macrophages, and levels of protein and LDH in BAL fluid and MPO activity in lung tissue at 1 to 6 days after LPS treatment. The increased resolution of inflammation was also confirmed by lung tissue histology showing improved alveolar and interstitial inflammatory changes after NAC treatment at 3 days post LPS treatment. It is possible that the enhanced apoptotic cell clearance induced by NAC may exhibit a general suppression of inflammation by Day 3, as its impact was greatest at this time point.

Several previous studies demonstrated the antiinflammatory effects of NAC on LPS- or carrageenan-induced acute lung injury at 6 or 24 hours after stimulation (8, 47). Redox-sensitive transcription factors expressed during acute lung injury, such as NF-κB, which regulates gene expression of proinflammatory mediators, have been the focus of studies describing the protective effects of NAC. In the present study, NF-κB activation and the phosphorylation and degradation of IκB-α were only induced by LPS treatment at 1 day, but not 3 days, post LPS treatment. This indicates that NAC uses different mechanisms to down-regulate proinflammatory mediator production with regard to its timing. Alternatively, these data may indicate that antiinflammatory signaling by NAC at 3 days after LPS exposure is related to enhanced apoptotic cell clearance due to down-regulation of the RhoA/Rho kinase pathway. This conclusion was supported by evidence that directly blocking RhoA/Rho kinase pathway exerts similar effects to that of NAC on LPS-induced inflammatory responses, proinflammatory mediators and TGF-β1 in BAL fluid, neutrophil infiltration, BAL protein levels, and LDH activity at 3 days post LPS. Interestingly, statins (3-hydroxyl-3-methylglutaryl coenzyme A-reductase inhibitors) also enhance apoptotic cell clearance in vitro and in vivo, in part, through inhibition of RhoA (27, 48). Furthermore, recent studies show that statins decrease LPS-induced pulmonary inflammation in healthy volunteers (49). Thus, these data suggest that there are common therapeutic roles for NAC, for Rho kinase inhibitors, or for statins via blocking the RhoA/Rho kinase pathway and enhancing apoptotic cell clearance in acute lung injury/ARDS.

Although TGF-β produced in response to apoptotic cells can lead to resolution of inflammation, these antiinflammatory effects appear to be overwhelmed in severe inflammatory circumstances (i.e., acute lung injury) (33). Oxidants selectively reverse TGF-β suppression of proinflammatory mediator production. Xiao and coworkers (50) demonstrated that oxidative stress reversed TGF-β suppression of proinflammatory mediators in macrophages by inducing p38 mitogen-activated protein kinase, which then prevented the up-regulation of extracellular signal-regulated kinase and mitogen-activated protein kinase phosphatase-1 normally induced by TGF-β. Therefore, it seems likely that NAC can dampen the effect of oxidative stress on TGF-β signaling in the context of LPS stimulation of acute lung inflammation.

In conclusion, we found that there is a relative suppression of apoptotic cell clearance through oxidant-dependent activation of RhoA/Rho kinase in acute lung injury. Our findings also demonstrate how an antioxidant, NAC, enhances apoptotic cell clearance through the down-regulation of RhoA in alveolar macrophages, resulting in concomitant reduction of proinflammatory mediators and increases in TGF-β1 production as well as reduction of inflammatory cell accumulation. Thus, these data suggest that by hastening the resolution of lung inflammation through enhancement of apoptotic cell clearance, NAC or RhoA/Rho kinase blockers may be potential therapeutic agents for inflammatory lung injury.

The authors thank Dr. Ju-Young Seoh and Ki-Hwan Han for their assistance with the analysis of flow cytometry and the histological study, respectively.

1. Nick JA, Young SK, Brown KK, Avdi NJ, Arndt PG, Suratt BT, Janes MS, Henson PM, Worthen GS. Role of p38 mitogen-activated pulmonary inflammation. J Immunol 2000;164:2151–2159.
2. Medan D, Wang L, Yang X, Dokka S, Castranova V, Rojanasakul Y. Induction of neutrophil apoptosis and secondary necrosis during endotoxin-induced pulmonary inflammation in mice. J Cell Physiol 2002;191:320–326.
3. Rydell-Törmänen K, Uller L, Erjefält JS. Direct evidence of secondary necrosis of neutrophils during intense lung inflammation. Eur Respir J 2006;28:268–274.
4. Rydell-Törmänen K, Uller L, Erjefält JS. Neutrophil cannibalism – a back up when the macrophage clearance system is insufficient. Respir Res 2006;7:143–149.
5. Haslett C. Granulocyte apoptosis and its role in the resolution and control of lung inflammation. Am J Respir Crit Care Med 1999;160:S5–S11.
6. Henson PM, Vandivier RW, Douglas IS. Cell death, remodeling, and repair in chronic obstructive pulmonary disease. Proc Am Thorac Soc 2006;3:713–717.
7. McPhillips K, Ghosh M, Byrne A, Gardai S, Remigio L, Bratton DL, Kang JL, Henson P. TNFα inhibits macrophage clearance of apoptotic cells via cPL2 and oxidant dependent mechanisms. J Immunol 2007;178:8117–8126.
8. Kao SJ, Wang D, Lin HI, Chen HI. N-acetylcysteine abrogates acute lung injury induced by endotoxin. Clin Exp Pharmacol Physiol 2006;33:33–40.
9. Hagiwara SI, Ishll Y, Kitamura S. Aerosolized administration of N-acetylcysteine attenuates lung fibrosis induced by bleomycin in mice. Am J Respir Crit Care Med 2000;162:225–231.
10. Rao GVS, Tinkle S, Weissman DN, Antonini JM, Kashon ML, Salmen R, Battelli LA, Willard PA, Hoover MD, Hubbs AF. Efficacy of a technique for exposing the mouse lung to particles aspirated from the pharynx. J Toxicol Environ Health A 2003;66:1441–1452.
11. Mizuguchi S, Takemura S, Minamiyama Y, Kodai S, Tsukioka T, Inoue K, Okada S, Suehiro S. S-allyl cysteine attenuated CCL4-induced oxidative stress and pulmonary fibrosis in rats. Biofactors 2006;26:81–92.
12. Demiralay R, Gürsan N, Özbilim G, Erdoğan G, Demirci E. Comparison of the effects of erdosteine and N-acetylcysteine on apoptosis regulation in endotoxin-induced acute lung injury. J Appl Toxicol 2006;26:301–308.
13. Tasaka S, Koh H, Yamada W, Shimizu M, Ogawa Y, Hasegawa N, Yamaguchi K, Ishii Y, Richer SE, Doerschuk CM, et al. Attenuation of endotoxin-induced acute lung injury by the Rho-associated kinase inhibitor, Y-27632. Am J Respir Cell Mol Biol 2005;32:504–510.
14. Castranova V, Jones T, Barger MW, Afshari A, Frazer DJ. Pulmonary responses of guinea pigs to consecutive exposures to cotton dust. In: Jacobs RR, Wakelyn PJ, Domelsmith LN, editors. Proceedings of the 14th Cotton Dust Research Conference. Memphis, TN: National Cotton Council; 1990. Pp. 131–135.
15. Wenzel SE, Szefler SJ, Leung DY, Sloan SI, Rex MD, Martin RJ. Bronchoscopic evaluation of severe asthma: persistent inflammation associated with high dose glucocorticoids. Am J Respir Crit Care Med 1997;156:737–743.
16. Savill JS, Wyllie AH, Henson JE, Walport MJ, Henson PM, Haslett C. Macrophage phagocytosis of aging neutrophils in inflammation. Programmed cell death in the neutrophil leads to its recognition by macrophages. J Clin Invest 1989;83:865–875.
17. Qin L, Block ML, Liu Y, Bienstock RJ, Pei Z, Zhang W, Wu X, Wilson B, Burka T, Hong JS. Microglial NADPH oxidase is a novel target for femtomolar neuroprotective against oxidative stress. FASEB J 2005;19:550–557.
18. Sanlioglu S, Williams CM, Samavati L, Butler NS, Wang G, McCray PB Jr, Ritchie TC, Hunninghake GW, Zandi E, Engelhard JF. Lipopolysaccharide induces Rac1-dependent reactive oxygen species formation and coordinates tumor necrosis factor-α secretion through IKK regulation of NF-κB. J Biol Chem 2001;276:30188–30198.
19. Allen RC. Evaluation of serum opsonic capacity by quantitating the initial chemiluminescent response from phagocytizing polymorphonuclear leukocytes. Infect Immun 1977;15:828–833.
20. Hill HR, Hogan NA, Bale JF, Hemming VG. Evaluation of nonspecific (alternative pathway) opsonic activity by neutrophil chemiluminescence. Int Arch Allergy Appl Immunol 1977;53:490–497.
21. Golpon HA, Fadok VA, Taraseviciene-Stewart L, Scerbavicius R, Sauer C, Welte T, Henson PM, Voelkel NF. Life after corpse engulfment: phagocytosis of apoptotic cells leads to VEGF secretion and cell growth. FASEB J 2004;18:1716–1733.
22. Hoffman PR, deCathelineau AM, Ogden CA, Leverrier Y, Bratton DL, Daleke DL, Ridley AJ, Fadok VA, Henson PM. Phosphatidylserine (PS) induces PS receptor-mediated macropinocytosis and promotes clearance of apoptotic cells. J Cell Biol 2001;155:649–659.
23. Fadok VA, Savill JS, Haslett C, Bratton DL, Doherty DE, Campbell PA, Henson PM. Different populations of macrophages use either the vitronectin receptor or the phosphatidylserine receptor to recognize and remove apoptotic cells. J Immunol 1992;149:4029–4035.
24. Kang JL, Lee HW, Lee HS, Pack IS, Chong Y, Castranova Y, Koh Y. Genistein prevents nuclear factor-kappa B activation and acute lung injury induced by lipopolysaccharide. Am J Respir Crit Care Med 2001;164:2206–2212.
25. Hartree EF. Determination of protein: a modification of the Lowry method that gives a linear photometric response. Anal Biochem 1972;48:422–427.
26. Lee HS, Moon C, Lee HW, Park EM, Cho MS, Kang JL. Src tyrosine kinases mediate activations of NF-κB and integrin signal during lipopolysaccharide-induced acute lung injury. J Immunol 2007;179:7001–7011.
27. Janssen WJ, McPhillips KA, Dickinson MS, Linderman DJ, Morimoto K, Xiao UQ, Oldham KM, Vandivier RW, Henson PM, Gardai SJ. Surfactant proteins A and D suppress alveolar macrophage phagocytosis via interaction with SIRPα. Am J Respir Crit Care Med 2008;178:155–167.
28. Richens TR, Linderman DJ, Horstmann SA, Lambert C, Xiao YQ, Keith RL, Boe DM, Morimoto K, Bowler RP, Day BJ, et al. Cigarette smoke impairs clearance of apoptotic cells through oxidant-dependent activation of RhoA. Am J Respir Crit Care Med 2009;179:1011–1021.
29. Jin L, Ying Z, Webb RC. Activation of Rho/Rho kinase signaling pathway by reactive oxygen species in rat aorta. Am J Physiol 2004;287:H1495–H1500.
30. Mong PY, Petrulio C, Kaufman HL, Wang Q. Activation of Rho kinase by TNF-α is required for JNK activation in human pulmonary microvascular endothelial cells. J Immunol 2008;180:550–558.
31. Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY, Henson PM. Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-β, PGE2, and PAF. J Clin Invest 1998;101:890–898.
32. Haddad JJ. Science review: redox and oxygen-sensitive transcription factors in the regulation of oxidant-mediated lung injury: role for nuclear factor-κB. Crit Care 2002;6:481–490.
33. Huynh ML, Malcolm KC, Kotaru C, Tilstra JA, Westcott JY, Fadok VA, Wenzel SE. Defective apoptotic cell phagocytosis attenuates prostaglandin E2 and 15-hydroxyeicosatetraenoic acid in severe asthma alveolar macrophages. Am J Respir Crit Care Med 2005;172:972–979.
34. Matute-Bello G, Liles WC, Radella F II, Steinberg KP, Ruzinski JT, Jonas M, Chi EY, Hudson LD, Martin TR. Neutrophil apoptosis in the acute respiratory distress syndrome. Am J Respir Crit Care Med 1997;156:1969–1977.
35. Matute-Bello G, Martin TR. Science review: apoptosis in acute lung injury. Crit Care 2003;7:355–358.
36. Matute-Bello G, Liles WC, Radella F II, Steinberg KP, Ruzinski JT, Hudson LD, Martin TR. Modulation of neutrophil apoptosis by granulate colony-stimulating factor and granulocyte/macrophage colony-stimulating factor during the course of acute respiratory distress syndrome. Crit Care Med 2000;28:1–7.
37. Anderson HA, Englert R, Gursel I, Shacter E. Oxidative stress inhibits the phagocytosis of apoptotic cells that have externalized phosphatidylserine. Cell Death Differ 2002;9:616–625.
38. Park HS, Chun JN, Jung HY, Choi C, Bae YS. Role of NADPH oxidase 4 in lipopolysaccharide-induced proinflammatory responses by human aortic endothelial cells. Cardiovasc Res 2006;72:447–455.
39. Heo J, Campbell SL. Mechanism of redox-mediated guanine nucleotide exchange on redox-active Rho GTPases. J Biol Chem 2005;280:31003–31010.
40. Alexandrova AY, Kopnin PB, Vasiliev JM, Kopnin BP. ROS up-regulation mediates Ras-induced changes of cell morphology and motility. Exp Cell Res 2006;312:2066–2073.
41. Nimnual AS, Taylor LJ, Bar-Sagi D. Redox-dependent downregulation of Rho by Rac. Nat Cell Biol 2003;5:236–241.
42. Buricchi F, Giannoni E, Grimaldi G, Parri M, Raugei G, Ramponi G, Chiarugi P. Redox regulation of ephrin/integrin cross-talk. Cell Adhesion Migration 2007;1:33–43.
43. Shinohara M, Shang WH, Kubodera M, Harada S, Mitsushita J, Kato M, Miyazaki H, Sumimoto H, Kamata T. Nox1 redox signaling mediates oncogenic Ras-induced disruption of stress fibers and focal adhesions by down-regulating Rho. J Biol Chem 2007;282:17640–17648.
44. Bianchi SM, Prince LR, McPhillips K, Allen L, Marriott HM, Taylor GW, Hellewell PG, Sabroe I, Dockrell DH, Henson PM, et al. Impairment of apoptotic cell engulfment by pyocyanin, a toxic metabolite of Pseudomonas aeruginosa. Am J Respir Crit Care Med 2008;177:35–43.
45. Huynh ML, Fadok VA, Henson PM. Phosphatidylserine-dependent ingestion of apoptotic cells promotes TGF-beta1 secretion and the resolution of inflammation. J Clin Invest 2002;109:41–50.
46. Hoffmann PR, Kench JA, Vondracek A, Kruk E, Daleke DL, Jordan M, Marrack P, Henson PM, Fadok VA. Interaction between phosphatidylserine and the phosphatidylserine receptor inhibits immune responses in vivo. J Immunol 2005;74:1393–1404.
47. Cuzzocrea S, Mazzon E, Dugo L, Serraino I, Ciccolo A, Centorrino T, De Sarro A, Caputi AP. Protective effects of N-acetylcysteine on lung injury and red blood cell modification induced by carrageenan in the rat. FASEB J 2001;15:1187–1200.
48. Morimoto KW, Janssen WJ, Fessler MB, McPhillips KA, Borges VM, Bowler RP, Xiao YQ, Kench JA, Henson PM, Vandivier RW. Lovastatin enhances clearance of apoptotic cells (efferocytosis) with implications for chronic obstructive pulmonary disease. J Immunol 2006;176:7657–7665.
49. Shyamsundar M, McKeown STW, O'Kane CM, Craig TR, Brown V, Thickett DR, Matthay MA, Taggart CC, Backman JT, Elborn JS, et al. Simvastatin decreases lipopolysaccharide-induced pulmonary inflammation in healthy volunteers. Am J Respir Crit Care Med 2009;179:1107–1114.
50. Xiao YQ, Freire-de-Lima CG, Janssen WJ, Morimoto K, Lyu D, Bratton DL, Henson PM. Oxidants selectively reverse TGF-β suppression of proinflammatory mediator production. J Immunol 2006;176:1209–1217.
Correspondence and requests for reprints should be addressed to Jihee Lee Kang, M.D., Ph.D., Department of Physiology, School of Medicine, Ewha Womans University 911-1 Mok-6-dong, Yangcheon-gu, Seoul 158-056, Korea. E-mail:

Related

No related items
American Journal of Respiratory and Critical Care Medicine
181
4

Click to see any corrections or updates and to confirm this is the authentic version of record