Mechanical ventilation (MV) with tidal volumes of 10–12 ml/kg is considered safe in the absence of acute lung injury (ALI). However, recent studies show that, when lung injury is already present, tidal volumes of this magnitude increase inflammation and injury in the lungs. We hypothesized that MV with tidal volumes of 10-ml/kg can also function as a cofactor in the initiation of ALI by modulating the transcriptional response to bacterial products. To test this hypothesis, we developed a mouse model in which MV did not independently cause inflammation or injury but augmented the inflammatory response to low-dose aspirated LPS and promoted development of ALI. We analyzed gene expression in lungs from 24 mice assigned to four different groups: control, MV only, intratracheal LPS only, and MV + LPS. There were twice as many differentially regulated genes in the MV + LPS group compared with the LPS-only group and 10 times as many differentially regulated genes compared with the MV-only group. For genes up-regulated by LPS treatment alone, the addition of MV further augmented expression. Cytokine concentrations in bronchoalveolar lavage fluid and tissue distribution of an intracellular protein, GADD45-γ, correlated with mRNA levels. We conclude that MV with conventional tidal volumes enhanced the transcriptional response to LPS and promoted development of ALI.

Although mechanical ventilation (MV)3 is a life-saving intervention for critically ill patients, it has also been recognized as a potential contributor to the pathogenesis of acute lung injury (ALI). MV is typically initiated with moderate tidal volumes of 8–12 ml/kg. Although this magnitude of tidal volume exceeds that typically observed at rest during spontaneous respiration, it is consistent with tidal volumes during exercise, is used routinely for prolonged periods in spinal cord injury patients (1), and has been considered safe in the absence of lung injury. Recently, however, tidal volumes of this magnitude have been associated with increased mortality as compared with smaller tidal volumes in patients with severe ALI (the acute respiratory distress syndrome (ARDS)) (2).

Whether MV using moderate tidal volumes can also function as a cofactor in the development of ALI in critically ill patients is an area of active research. In ex vivo nonperfused rat lungs, MV has been shown to augment expression of TNF-α in response to LPS (3, 4). We and others have shown that, in an in vivo rabbit model, MV with moderate tidal volumes of 10–15 ml/kg augments cytokine expression in bronchoalveolar lavage (BAL) fluid in response to different doses of systemic LPS (5, 6). Additionally, MV with moderate tidal volumes augmented inflammation following intratracheal acid instillation, suggesting that this response is not specific to stimulation with microbial products (7).

Because ALI is always associated with an excessive inflammatory response in the lungs (8, 9), it is plausible to speculate that the observed interaction between MV and LPS promotes the development of lung injury over time. Therefore, understanding the mechanism by which MV functions as an immunomodulator may facilitate development of novel therapeutic strategies for respiratory failure. We hypothesized that MV augments the transcriptional response to LPS resulting in enhanced inflammation and lung injury. To test this hypothesis, we developed a murine model of lung inflammation and injury that was dependent on concurrent low-dose LPS exposure and MV with moderate tidal volumes. Using this model we then evaluated transcriptional responses to MV, LPS, and the combination of MV and LPS using high-density oligonucleotide expression microarrays. We found that MV with a tidal volume of 10 ml/kg resulted in no significant pulmonary inflammation or injury and only modest differential gene expression compared with nonventilated controls. However, when MV was combined with LPS there was broad augmentation of gene transcription, which was associated with enhanced inflammation and the development of ALI.

A stock solution of LPS derived from Escherichia coli serotype 0111:B4 (List Biological Laboratories) in PBS was stored in small aliquots at −20°C. Immediately before each experiment, an aliquot of LPS was thawed, sonicated, and diluted in PBS to a working concentration of 2.5 μg/ml with PBS.

The University of Washington Office of Animal Welfare approved these experiments. C57BL/6 8- to 10-wk-old male mice (Harlan) were housed for a minimum of 5 days in a modified specific pathogen-free facility before all experiments. Mice were randomly assigned to one of four groups: 1) PBS aspiration without MV (control); 2) PBS with MV; 3) LPS aspiration without MV (LPS), or 4) LPS with MV (MV + LPS).

Each mouse was anesthetized with 4% isoflurane and suspended by its front teeth at a 60° angle. After extending the jaw and tongue, 0.005-μg/gm body weight LPS (2 μl/gm body weight of diluted LPS) or an equal volume of pyrogen-free PBS was deposited in the oropharynx with a pipette (10). Aspiration of the liquid was visually confirmed, and the mouse returned to its cage.

Thirty minutes after aspiration, each mouse was reanesthetized with ketamine (1.3 mg) and xylazine (0.9 mg) i.p. followed by 4% isoflurane for 3 min. Mice assigned to the control or LPS groups were returned to their cages. Mice assigned to MV were positioned at a 60° angle for intubation. After the vocal cords were visualized by extending the jaw and transilluminating the larynx with an external fiber optic light source, the mice were intubated with a 20-gauge angiocatheter (BD Biosciences). The angiocatheter was connected to a 1-ml syringe with a cotton plug in the tip and a small drop of sterile saline in the syringe barrel. Tracheal insertion was confirmed by respiratory motion of the saline droplet at which point the catheter was positioned to a depth of 20 mm at the teeth and secured in place with a drop of superglue (Loctite). The hub of the angiocatheter was removed 25 mm above the catheter tip, and the mice were connected to a ventilator (MiniVent; Harvard Biosciences). Mice were ventilated with a tidal volume of 10 μl/gm body weight, a respiratory rate of 150/min, 0 cm H2O end-expiratory pressure, and an FIO2 of 0.21. In preliminary studies this ventilation strategy produced normal arterial blood pH values (7.36 ± 0.08, n = 4). MV was continued for either 4 or 6 h with anesthesia maintained by 1% isoflurane. Mixed expired CO2 (Novametrix) and airway pressure was continuously monitored. Body temperature was monitored by a rectally inserted thermistor and regulated by a heating lamp. At the end of the MV period, the mouse was euthanized by i.p. sodium pentothal and exsanguination by cardiac puncture.

For one series of experiments, mice were euthanized after either 4 or 6 h of ventilation, and BAL of both lungs was performed using four 1-ml aliquots of PBS containing 0.6 mM EDTA. Total cell count was determined using a hemacytometer. Differential cell count was determined by cytocentrifugation and Wright-Giemsa staining. The remaining fluid was spun at 1200 × g for 15 min, and the supernatant was removed and stored in small aliquots at −80°C for batch protein analysis.

BAL fluid concentrations of MIP-2, mouse growth-related oncogene homologue (KC), IL-1β, IL-6, IL-10, IL-12, TNF-α, GM-CSF, and IFN-γ were determined by multiplex immunoassay (Luminex 100) using cytokine-specific bead kits (R&D Systems) according to the vendors’ protocols. BAL fluid concentrations of IgM and albumin were determined by ELISA (Bethyl Laboratories).

In two mice per group, the lungs were removed after 6 h and fixed at 20 cm H2O with 4% paraformaldehyde. After fixation, the lungs were embedded in paraffin, cut into 4-μm sections, and stained with H&E. Immunohistochemistry was performed with Vectastain ABC kit (Vector Laboratories). Briefly, slides were deparaffinized by washing twice in xylene for 5 min and rehydrated by washing twice in 100% ethanol for 3 min, twice in 95% ethanol for 3 min, twice in 70% ethanol for 3 min, and twice in distilled H2O for 5 min. Slides were rinsed twice with PBS for 5 min. Ag unmasking was performed by bringing Citrate Buffer (Vector Laboratories) to a boil and then allowing the slides to simmer for an additional 10 min. After the 10-min boil, the solution was allowed to cool for 20 min before removing the slides. After digestion, slides were rinsed three times in distilled water for 5 min each. The slides were then incubated with 1% H2O2 in methanol for 30 min to block endogenous peroxidases. Slides were again washed with distilled water three times for 5-min each and once with PBS for 5 min before blocking with normal goat serum (Vector Laboratories) for 60 min. Then the samples were labeled with chicken anti-human growth arrest and DNA damage 45-γ (GADD45-γ) (Chemicon International) overnight in a moist chamber at 4°C. Next, the slides were rinsed three times with PBS and labeled with rabbit anti-chicken biotinylated Ab (Vector Laboratories) at a 1/200 dilution for 30 min at room temperature. The slides were rinsed twice with PBS, labeled with the Vectastain ABC kit, incubated in a moist chamber for 30 min at room temperature, and rinsed three times with PBS. Slides were developed in a moist chamber with diaminobenzidine substrate (Sigma-Aldrich) for 8–12 min. The slides were rinsed twice with deionized H2O and counterstained with 1% methyl green for 5 min.

In a second set of experiments, mice were euthanized after 4 h of ventilation. The lungs were flushed with 2 ml of iced saline via the right ventricle, and then homogenized in guanidinium isothiocyanate lysis solution with a rotor-stator homogenizer (Omni International). Total RNA was extracted using the RNeasy midi kit (Qiagen), and any remaining genomic DNA was removed with DNase I treatment (DNA-free; Ambion). RNA quality was assessed by determining the 26S/18S ratio (Agilent 2100 Bioanalyzer).

Complete details of experiment design, hybridization procedures, and raw microarray data in compliance with Minimum Information About a Microarray Experiment (MIAME) have been deposited at the Gene Expression Omnibus web site (〈www.ncbi.nlm.nih.gov/projects/geo/〉, query GSE 2411). Briefly, for each mouse, labeled cRNA was prepared from total RNA and hybridized to an Affymetrix GeneChip 430A oligonucleotide microarray with minor modifications from Affymetrix recommended protocols as detailed at the University of Washington’s Center for Expression Arrays web site (〈http://ra.microslu.washington.edu〉). Microarrays were scanned using the GeneChip Scanner and image analysis was performed with Affymetrix MAS 5.0 software. Background adjustment and quantile normalization of microarray data was performed using the Robust Multichip Average algorithm (11). Statistically significant differential gene expression for each perturbation (MV, LPS, MV + LPS) relative to control mice was determined using significance analysis of microarrays (12) and a predetermined false discovery rate of 1%. To calculate gene expression changes relative to the control group, all data were log2-transformed. For each gene, the average of the log2-transformed results from the six control mice was subtracted from each data point within all four of the groups, and the mean value and upper and lower limits of the 95% confidence interval were calculated. The data were transformed back to the linear domain to give the average fold change in expression with 95% confidence interval for each gene within each group relative to the control group.

For confirmatory quantitative RT-PCR, first-strand cDNA was transcribed from 2 μg of total RNA using random hexamer primers, and Superscript II (Invitrogen Life Technologies). Quantitative cDNA amplification was performed using an ABI 5700 real-time PCR machine (Applied Biosystems), gene-specific TaqMan primer-probe mixes (Applied Biosystems), and PCR conditions as specified by the manufacturer. All samples were run in duplicate. Target gene expression was quantified relative to 18s RNA from the same reverse transcription reaction.

Nonmicroarray data were compared among groups by ANOVA with posthoc comparisons using Tukey’s HSD test. Statistical significance was assigned at p < 0.05. All analyses were done using JMP software (SAS). Data are presented as mean ± SEM.

We measured concentrations of cytokines in BAL fluid at both the 4- and 6-h times. IL-10, IL-12, and IFN-γ were below the threshold of detection in all samples at both times. After 4 h, the combination of MV and LPS (MV + LPS group) caused increased concentrations of MIP-2 (745 ± 107 pg/ml), KC (958 ± 100 pg/ml), and IL-6 (1041 ± 243 pg/ml) compared with the other three groups (p < 0.05, Fig. 1). A similar trend was observed for both IL-1β and GM-CSF, although the differences between the MV + LPS group and the LPS group were not statistically significant. In contrast, mice in the LPS-only group had significant increases in the concentrations of three cytokines, KC, GM-CSF, and TNF-α, as compared with the control group. There were no differences between the MV group and the control group for any of the cytokines measured.

FIGURE 1.

BAL fluid cytokine concentrations in the control group, MV-only group, LPS-only group, and MV + LPS group: A, MIP-2; B, KC; C, IL-6; D, GM-CSF; E, TNF-α; and F, IL-1β. n = 6/group; p ≤ 0.05 compared with the control (∗), the MV-only (†), and the LPS-only (#) groups.

FIGURE 1.

BAL fluid cytokine concentrations in the control group, MV-only group, LPS-only group, and MV + LPS group: A, MIP-2; B, KC; C, IL-6; D, GM-CSF; E, TNF-α; and F, IL-1β. n = 6/group; p ≤ 0.05 compared with the control (∗), the MV-only (†), and the LPS-only (#) groups.

Close modal

At 4 h, there was early lung injury in the MV + LPS group compared with the control group as indicated by total polymorphonuclear leukocyte (PMN) count in the BAL fluid (279 ± 67 × 104 vs 1.2 ± 0.5 × 104, p < 0.05) and by permeability to IgM (39.6 ± 4.7 vs 17.1 vs 0.4 ng/ml, p < 0.05). PMN counts and IgM concentrations of the MV-only and the LPS-only groups were all significantly less than the MV + LPS group. The PMN count from the LPS-only group trended higher than the control group, but the difference was not statistically significant (111 ± 15 × 104 vs 1.2 ± 0.5 × 104, p > 0.05).

Inflammation and ALI were also evaluated in 18 mice divided among the MV, LPS, and MV + LPS groups (n = 6/group) after 6 h of MV or spontaneous respiration. These groups were compared with the previous control group. Total PMN counts in BAL fluid followed the same pattern as at 4 h except that the LPS-only group was significantly higher than the control and MV-only group (Fig. 2,A). There were no changes within each group as compared with the 4-h time. The IgM concentrations in BAL fluid were higher for the MV-only, LPS-only, and MV + LPS groups compared with the 4-h time; however, the MV + LPS group remained the only group statistically different from Control (93.2 ± 9.9 vs 17.1 ± 0.4 ng/ml, p < 0.05, Fig. 2 B). Albumin concentrations in BAL fluid had the same pattern as IgM concentration across all four groups. There was a statistically significant difference between the MV + LPS group at 6 h compared with the control group (122.4 ± 20.5 vs 23.5 ± 1.1 μg/ml, p < 0.05). Mean albumin concentration tended to be higher in the MV-only (72.7 ± 9.2 μg/ml) and LPS-only (66.3 ± 21.9 μg/ml) groups compared with the control group. Peak airway pressures in the mechanically ventilated animals were similar at baseline between the MV + LPS and MV-only groups (10.8 ± 0.4 vs 11.7 ± 1.4 cm H2O). After 6 h, airway pressures had increased in the MV + LPS group but not in the MV-only group (14.0 ± 1.3 vs 11.8 ± 1.0 cm H2O, p = 0.008), suggesting decreased compliance in the MV + LPS group but not in the MV-only group.

FIGURE 2.

BAL fluid total PMN cell count (A) and IgM concentration (B) from the MV-only, LPS-only, and MV + LPS groups at 6 h vs the 4 h control group. n = 6/group; p ≤ 0.05 compared with control (∗), MV-only (†), and LPS-only (#) groups.

FIGURE 2.

BAL fluid total PMN cell count (A) and IgM concentration (B) from the MV-only, LPS-only, and MV + LPS groups at 6 h vs the 4 h control group. n = 6/group; p ≤ 0.05 compared with control (∗), MV-only (†), and LPS-only (#) groups.

Close modal

Lungs were prepared for histological evaluation for two animals in each of the four groups at the 6-h time. Lungs from the MV-only group were not qualitatively different from those from the control group (Fig. 3). In the LPS-only group, there was increased cellularity without edema or septal thickening. In the MV + LPS group, there was increased cellularity, septal thickening, and perivascular and peribronchial edema consistent with early lung injury.

FIGURE 3.

Representative histological images from the control group at 4 h and the MV-only, LPS-only, and MV + LPS groups at 6 h. The higher magnification insert for the MV + LPS histology reveals the PMN component of the cellular infiltrate as well as fibrin stranding in the alveolar space.

FIGURE 3.

Representative histological images from the control group at 4 h and the MV-only, LPS-only, and MV + LPS groups at 6 h. The higher magnification insert for the MV + LPS histology reveals the PMN component of the cellular infiltrate as well as fibrin stranding in the alveolar space.

Close modal

Gene expression was assessed in six mice from each group after 4 h of MV or spontaneous breathing. Comparisons were made between the Control group and the MV, LPS, and MV + LPS groups. At a false discovery rate of 1%, a total of 6136 genes were differentially expressed in at least one of the MV, LPS, or MV + LPS groups compared with the control group. There were 324 genes up-regulated and 169 genes down-regulated in the MV group, and 1099 genes up-regulated and 1832 genes down-regulated in the LPS group. The greatest number of differentially expressed genes was in the MV + LPS group with 1976 genes up-regulated and 3753 genes down-regulated. Thus, the combination of MV with LPS stimulation resulted in twice as many differentially regulated genes as compared with the LPS-only group and over 10 times as many genes as in the MV-only group.

If only highly up-regulated genes were considered, there were 217 genes with >3-fold expression in at least one of the intervention groups compared with the control group. Within this subgroup of highly expressed genes, 98.6% had >3-fold up-regulation in the MV + LPS group as compared with 48.8% in the LPS-only group and 5.5% in the MV group. When the expression of this subgroup of genes was examined across the three interventions, there was a pattern of increasing up-regulation comparing the LPS-only group with the MV + LPS group (Fig. 4). A number of these genes did not have well-annotated biological functions; however, when a subset of 135 genes was grouped by biological function, the largest number (53) were found to be involved in immunity/inflammation response (online supplemental data).4 Included within this highly up-regulated group of genes were chemokines (e.g., CCL3, CXCL2), other cytokines (e.g., IL-6, IL-1β), and genes that participate in the acute transcriptional response to stress, such as components of the MAPK cascade (e.g., GADD45-γ) and transcription factors (e.g., IFN regulatory factor-7 (IRF-7)). Quantitative RT-PCR confirmed augmented expression of these genes when MV was combined with LPS exposure (Fig. 4).

FIGURE 4.

Heat map showing row normalized expression patterns across the control, MV, LPS, and MV + LPS groups for the 217 genes expressed at >3-fold above the control group in at least one of the other groups (left side). Independent confirmation of gene expression patterns relative to the control group and normalized to 18s RNA for selected genes by using quantitative RT-PCR (right side). n = 6/group; p ≤ 0.05 compared with the MV-only (#) or the LPS-only (∗) groups.

FIGURE 4.

Heat map showing row normalized expression patterns across the control, MV, LPS, and MV + LPS groups for the 217 genes expressed at >3-fold above the control group in at least one of the other groups (left side). Independent confirmation of gene expression patterns relative to the control group and normalized to 18s RNA for selected genes by using quantitative RT-PCR (right side). n = 6/group; p ≤ 0.05 compared with the MV-only (#) or the LPS-only (∗) groups.

Close modal

We chose to further investigate GADD45 because it is a cytoplasmic protein that can be localized to specific cell types by immunohistochemistry. The GADD45 family includes three isoforms (-α, -β, and -γ), which are transcriptionally up-regulated following exposure to mitogenic stimuli such as ionizing or UV radiation (13). Both GADD45-γ and -β bind to and activate MEK kinase 4 (MEKK4) ultimately leading to phosphorylation and activation of the MAPKs p38 and JNK (14, 15, 16). Quantitative RT-PCR analysis confirmed that all three isoforms of GADD45 were up-regulated (Figs. 4 and 5) in our model. GADD45-γ, by immunohistochemistry, showed protein expression corresponding with the up-regulation in mRNA (Fig. 6). In the MV + LPS group, GADD45-γ was identified in alveolar wall cells and in large intra-alveolar cells suggesting up-regulation in both alveolar epithelium and macrophages, but not in neutrophils recruited to the air spaces (Fig. 7).

FIGURE 5.

Expression of GADD45-α (A), and GADD45-β (B) in whole lung homogenate. Expression is relative to the control group normalized to 18s RNA. n = 6/group; p ≤ 0.05 compared with the MV-only (#) or the LPS-only (∗) groups.

FIGURE 5.

Expression of GADD45-α (A), and GADD45-β (B) in whole lung homogenate. Expression is relative to the control group normalized to 18s RNA. n = 6/group; p ≤ 0.05 compared with the MV-only (#) or the LPS-only (∗) groups.

Close modal
FIGURE 6.

Immunohistochemistry identifying expression of GADD45-γ in animals from the MV-only, the LPS-only, and the MV + LPS groups. Arrows identify positive staining cells. All three animals have expression in airway epithelium. GADD45-γ staining is increased in the MV + LPS animal.

FIGURE 6.

Immunohistochemistry identifying expression of GADD45-γ in animals from the MV-only, the LPS-only, and the MV + LPS groups. Arrows identify positive staining cells. All three animals have expression in airway epithelium. GADD45-γ staining is increased in the MV + LPS animal.

Close modal
FIGURE 7.

GADD45-γ staining in a lung section from an animal treated with both MV and LPS. GADD45-γ is present in airway epithelium (open arrowheads), large intra-alveolar cells (open arrows), and alveolar wall cells (solid arrows). Inset, A negative control in which the anti-GADD45-γ Ab was excluded.

FIGURE 7.

GADD45-γ staining in a lung section from an animal treated with both MV and LPS. GADD45-γ is present in airway epithelium (open arrowheads), large intra-alveolar cells (open arrows), and alveolar wall cells (solid arrows). Inset, A negative control in which the anti-GADD45-γ Ab was excluded.

Close modal

We hypothesized that MV with a noninjurious and clinically relevant tidal volume would augment the transcriptional response to LPS, resulting in enhanced inflammation and early ALI. The main findings of this study were that: 1) MV with a tidal volume of 10 ml/kg for 6 h did not cause significant cytokine expression, PMN infiltration into the lung, or increased permeability of the alveolar-capillary barrier; 2) low dose LPS exposure by pharyngeal aspiration caused modest cytokine expression at both mRNA and protein levels and PMN migration to the alveolar space but did not cause significant lung injury; 3) MV with a clinically noninjurious tidal volume augmented lung inflammation and ALI in response to low-dose LPS exposure, and 4) this augmented response was associated with broad modulation of the acute transcriptional response in the lungs.

The current study significantly extends previous reports of interaction between MV and LPS (5, 6). By using intratracheal instillation of LPS instead of systemic injection, we eliminated the possibility that augmented inflammation resulted from MV-induced systemic-to-alveolar translocation of LPS or LPS-induced early response cytokines (17, 18). Additionally, we used microarray analysis to evaluate the role of transcription. We found that differential gene expression was greatest in the group that received both MV and LPS. Although alterations in the posttranscriptional mRNA half-life cannot be excluded in this in vivo model, increases in both the total number of differentially expressed genes and in the magnitude of fold change in expression for multiple genes in the MV + LPS group suggested that gene transcription was modulated by combining MV with LPS. Transcriptional modulation by MV is supported by several published studies, which show that stretching lung-derived cells in vitro activates transcription factors, which promote proinflammatory cytokine transcription (19, 20, 21). Activation of the transcription factors NF-κB and AP-1 occurs when lungs are ventilated ex vivo and in vivo with moderate to large tidal volumes (5, 22, 23).

We compared the results of our gene expression data with those from two previously published models of lung injury to identify genes that were consistently differentially expressed and, therefore, were more likely to have an important role in the pathogenesis of ALI. In one study, microarray results from ventilator-induced models of lung injury in mice, rats, and dogs and from human pulmonary endothelial cell cultures subjected to in vitro cyclical stretch were compared to identify 69 candidate genes with differential expression (64 up-regulated and 5 down-regulated) across multiple species (24). Of these 69 genes, we identified 32 genes that were differentially expressed in the same direction (29 increased, 3 decreased) in the MV + LPS group of our study (Table I). Included within this list are multiple genes involved with inflammation (CCL2, IL-1β, IL-6), regulators of transcription (activating transcription factor-3, CCAAT/enhancer binding protein-β and -δ), MAPK cascade elements (GADD45-α), and other genes associated with lung injury (F3 or tissue factor and Serpine1 or plasminogen activator inhibitor-1) (25, 26). For the majority of these genes, combining MV and LPS resulted in a larger change in expression compared with either LPS or MV alone. Using a different model of intratracheal bleomycin-induced pulmonary fibrosis, Kaminski et al. (27) evaluated the temporal pattern of gene expression in the lung. Several genes identified by these investigators were also up-regulated in our model (Table II), including GADD45 (isoform not identified), regulators of transcription (STAT-1, IRF-7), and chemokines (CCL2, CCL7). Notably, GADD45 expression was increased in all three studies. We further examined GADD45 expression and confirmed up-regulation of all three GADD45 isoforms. Additionally, we showed that GADD45-γ protein was expressed in lung tissue. GADD45-γ was present in lung resident cells as opposed to recruited PMN cells.

Table I.

Expression pattern of genes that were also identified in a separate cross-species study of ventilator-induced (large tidal volume) lung injury (24 )

Gene SymbolGene ID (Entrez gene)ControlMVLPSMV + LPS
Admr 11536 1.0 (0.9–1.1) 0.7 (0.6–0.8) 0.7 (0.6–0.8) 0.5 (0.5–0.6) 
Aqp1 11826 1.0 (0.9–1.1) 0.8 (0.8–0.9) 0.9 (0.8–1.0) 0.5 (0.5–0.6) 
Arg2 11847 1.0 (0.8–1.3) 2.0 (1.7–2.2) 3.5 (3.3–3.8) 4.6 (4.1–5.2) 
Atf3 11910 1.0 (0.9–1.1) 1.5 (1.4–1.7) 2.3 (1.9–2.8) 4.9 (3.6–6.6) 
Bhlhb2 20893 1.0 (0.8–1.2) 1.3 (1.2–1.5) 1.4 (1.2–1.6) 2.2 (1.8–2.8) 
Btg1 12226 1.0 (0.9–1.2) 1.1 (1.1–1.2) 1.6 (1.5–1.8) 2.0 (1.9–2.2) 
Ccl2 (MCP-1) 20296 1.0 (0.8–1.3) 0.8 (0.8–0.9) 2.8 (2.2–3.6) 5.4 (3.2–9.1) 
Cdkn1a 12575 1.0 (0.8–1.3) 1.3 (1.2–1.5) 2.0 (1.6–2.6) 2.1 (1.9–2.2) 
Cebpb 12608 1.0 (0.9–1.2) 1.5 (1.3–1.8) 1.7 (1.5–2.1) 2.7 (2.0–3.5) 
Cebpd 12609 1.0 (0.9–1.2) 2.2 (1.9–2.5) 2.8 (2.6–3.1) 3.5 (3.1–3.9) 
Cycs 13063 1.0 (1.0–1.1) 1.2 (1.1–1.2) 1.3 (1.2–1.3) 1.6 (1.5–1.7) 
Dnaja1 15502 1.0 (0.9–1.1) 1.4 (1.3–1.6) 1.1 (1.0–1.2) 1.6 (1.4–1.8) 
Egfr 13649 1.0 (0.9–1.1) 0.9 (0.9–1.0) 1.6 (1.5–1.6) 1.5 (1.4–1.7) 
Eif2s1 13665 1.0 (0.9–1.2) 1.2 (1.2–1.3) 1.3 (1.2–1.4) 1.8 (1.7–2.0) 
F3 14066 1.0 (0.8–1.3) 1.8 (1.5–2.1) 1.5 (1.3–1.8) 3.0 (2.2–4.1) 
Gadd45α 13197 1.0 (0.9–1.2) 2.1 (1.8–2.4) 2.0 (1.7–2.3) 2.6 (2.2–3.0) 
Gbp2 14469 1.0 (0.6–1.7) 1.0 (0.9–1.1) 4.6 (3.9–5.5) 7.3 (5.7–9.3) 
Gch1 14528 1.0 (0.8–1.3) 1.2 (1.1–1.4) 2.5 (2.4–2.7) 3.7 (3.1–4.4) 
Gclc 14629 1.0 (0.8–1.4) 1.2 (1.1–1.4) 2.4 (2.3–2.6) 2.1 (2.0–2.1) 
Gja1 14609 1.0 (0.8–1.3) 1.2 (1.2–1.3) 2.3 (2.0–2.5) 2.9 (2.6–3.2) 
Hspa8 15481 1.0 (0.9–1.2) 1.5 (1.3–1.6) 1.1 (1.0–1.2) 1.9 (1.8–2.1) 
Ifrd1 15982 1.0 (0.9–1.1) 1.1 (1.0–1.3) 1.4 (1.3–1.6) 1.5 (1.3–1.8) 
Il1b 16176 1.0 (0.7–1.6) 1.4 (1.2–1.7) 7.2 (6.2–8.3) 10.3 (7.7–13.8) 
Il1r2 16178 1.0 (0.7–1.4) 3.9 (3.1–5.0) 8.9 (6.9–11.5) 13.7 (11.3–16.6) 
Il6 16193 1.0 (0.9–1.2) 1.3 (1.1–1.5) 3.3 (2.6–4.3) 9.1 (5.2–15.8) 
Lgals3 16854 1.0 (0.9–1.2) 1.4 (1.2–1.5) 1.2 (1.1–1.2) 1.8 (1.6–2.1) 
Plaur 18793 1.0 (0.9–1.2) 1.7 (1.6–1.8) 2.0 (1.8–2.2) 3.5 (2.8–4.3) 
Plcg2 234779 1.0 (0.9–1.1) 1.1 (1.0–1.1) 1.1 (1.0–1.2) 1.5 (1.3–1.6) 
Ptgs2 (Cox-2) 19225 1.0 (0.9–1.1) 1.0 (0.9–1.2) 1.3 (1.2–1.4) 2.8 (1.9–4.3) 
S100a9 20202 1.0 (0.6–1.8) 2.6 (2.2–3.1) 4.3 (3.6–5.2) 4.7 (4.0–5.5) 
Serpine1 (PAI-1) 18787 1.0 (0.9–1.2) 2.1 (1.9–2.3) 4.0 (3.5–4.6) 7.7 (5.0–11.9) 
Tcf21 21412 1.0 (0.9–1.2) 1.1 (1.0–1.2) 0.7 (0.6–0.8) 0.6 (0.4–0.8) 
Gene SymbolGene ID (Entrez gene)ControlMVLPSMV + LPS
Admr 11536 1.0 (0.9–1.1) 0.7 (0.6–0.8) 0.7 (0.6–0.8) 0.5 (0.5–0.6) 
Aqp1 11826 1.0 (0.9–1.1) 0.8 (0.8–0.9) 0.9 (0.8–1.0) 0.5 (0.5–0.6) 
Arg2 11847 1.0 (0.8–1.3) 2.0 (1.7–2.2) 3.5 (3.3–3.8) 4.6 (4.1–5.2) 
Atf3 11910 1.0 (0.9–1.1) 1.5 (1.4–1.7) 2.3 (1.9–2.8) 4.9 (3.6–6.6) 
Bhlhb2 20893 1.0 (0.8–1.2) 1.3 (1.2–1.5) 1.4 (1.2–1.6) 2.2 (1.8–2.8) 
Btg1 12226 1.0 (0.9–1.2) 1.1 (1.1–1.2) 1.6 (1.5–1.8) 2.0 (1.9–2.2) 
Ccl2 (MCP-1) 20296 1.0 (0.8–1.3) 0.8 (0.8–0.9) 2.8 (2.2–3.6) 5.4 (3.2–9.1) 
Cdkn1a 12575 1.0 (0.8–1.3) 1.3 (1.2–1.5) 2.0 (1.6–2.6) 2.1 (1.9–2.2) 
Cebpb 12608 1.0 (0.9–1.2) 1.5 (1.3–1.8) 1.7 (1.5–2.1) 2.7 (2.0–3.5) 
Cebpd 12609 1.0 (0.9–1.2) 2.2 (1.9–2.5) 2.8 (2.6–3.1) 3.5 (3.1–3.9) 
Cycs 13063 1.0 (1.0–1.1) 1.2 (1.1–1.2) 1.3 (1.2–1.3) 1.6 (1.5–1.7) 
Dnaja1 15502 1.0 (0.9–1.1) 1.4 (1.3–1.6) 1.1 (1.0–1.2) 1.6 (1.4–1.8) 
Egfr 13649 1.0 (0.9–1.1) 0.9 (0.9–1.0) 1.6 (1.5–1.6) 1.5 (1.4–1.7) 
Eif2s1 13665 1.0 (0.9–1.2) 1.2 (1.2–1.3) 1.3 (1.2–1.4) 1.8 (1.7–2.0) 
F3 14066 1.0 (0.8–1.3) 1.8 (1.5–2.1) 1.5 (1.3–1.8) 3.0 (2.2–4.1) 
Gadd45α 13197 1.0 (0.9–1.2) 2.1 (1.8–2.4) 2.0 (1.7–2.3) 2.6 (2.2–3.0) 
Gbp2 14469 1.0 (0.6–1.7) 1.0 (0.9–1.1) 4.6 (3.9–5.5) 7.3 (5.7–9.3) 
Gch1 14528 1.0 (0.8–1.3) 1.2 (1.1–1.4) 2.5 (2.4–2.7) 3.7 (3.1–4.4) 
Gclc 14629 1.0 (0.8–1.4) 1.2 (1.1–1.4) 2.4 (2.3–2.6) 2.1 (2.0–2.1) 
Gja1 14609 1.0 (0.8–1.3) 1.2 (1.2–1.3) 2.3 (2.0–2.5) 2.9 (2.6–3.2) 
Hspa8 15481 1.0 (0.9–1.2) 1.5 (1.3–1.6) 1.1 (1.0–1.2) 1.9 (1.8–2.1) 
Ifrd1 15982 1.0 (0.9–1.1) 1.1 (1.0–1.3) 1.4 (1.3–1.6) 1.5 (1.3–1.8) 
Il1b 16176 1.0 (0.7–1.6) 1.4 (1.2–1.7) 7.2 (6.2–8.3) 10.3 (7.7–13.8) 
Il1r2 16178 1.0 (0.7–1.4) 3.9 (3.1–5.0) 8.9 (6.9–11.5) 13.7 (11.3–16.6) 
Il6 16193 1.0 (0.9–1.2) 1.3 (1.1–1.5) 3.3 (2.6–4.3) 9.1 (5.2–15.8) 
Lgals3 16854 1.0 (0.9–1.2) 1.4 (1.2–1.5) 1.2 (1.1–1.2) 1.8 (1.6–2.1) 
Plaur 18793 1.0 (0.9–1.2) 1.7 (1.6–1.8) 2.0 (1.8–2.2) 3.5 (2.8–4.3) 
Plcg2 234779 1.0 (0.9–1.1) 1.1 (1.0–1.1) 1.1 (1.0–1.2) 1.5 (1.3–1.6) 
Ptgs2 (Cox-2) 19225 1.0 (0.9–1.1) 1.0 (0.9–1.2) 1.3 (1.2–1.4) 2.8 (1.9–4.3) 
S100a9 20202 1.0 (0.6–1.8) 2.6 (2.2–3.1) 4.3 (3.6–5.2) 4.7 (4.0–5.5) 
Serpine1 (PAI-1) 18787 1.0 (0.9–1.2) 2.1 (1.9–2.3) 4.0 (3.5–4.6) 7.7 (5.0–11.9) 
Tcf21 21412 1.0 (0.9–1.2) 1.1 (1.0–1.2) 0.7 (0.6–0.8) 0.6 (0.4–0.8) 
Table II.

Expression pattern of genes that were also identified as up-regulated by day +2 following intratracheal bleomycin instillation (27 )

Gene SymbolGene ID (Entrez gene)ControlMVLPSMV + LPS
Ccl2 (MCP-1) 20296 1.0 (0.8–1.3) 0.8 (0.8–0.9) 2.8 (2.2–3.6) 5.4 (3.2–9.1) 
Ccl7 (MCP-3) 20306 1.0 (0.8–1.3) 0.9 (0.9–1.0) 1.8 (1.5–2.1) 2.9 (1.9–4.4) 
Cdkn1a (CDK11) 12575 1.0 (0.8–1.3) 1.3 (1.2–1.5) 2.0 (1.6–2.6) 2.1 (1.9–2.2) 
G1p2 (ISG15) 53606 1.0 (0.7–1.4) 0.9 (0.8–1.0) 6.2 (4.9–7.9) 12.1 (9.3–15.8) 
Gadd45α 13197 1.0 (0.9–1.2) 2.1 (1.8–2.4) 2.0 (1.7–2.3) 2.6 (2.2–3.0) 
Ifi1 15944 1.0 (0.8–1.4) 1.0 (0.9–1.2) 3.4 (3.0–3.9) 3.8 (3.1–4.8) 
Ifit1 (GARG16) 15957 1.0 (0.6–1.7) 1.1 (0.9–1.4) 9.4 (8.0–11.0) 12.1 (9.3–15.8) 
Irf7 54123 1.0 (0.9–1.2) 1.1 (1.0–1.3) 3.6 (3.1–4.3) 6.9 (5.2–9.1) 
Lcn2 (Lipocalin 2) 16819 1.0 (0.6–1.7) 0.9 (0.7–1.2) 2.8 (2.7–2.9) 2.7 (2.6–2.9) 
Saa3 20210 1.0 (0.6–1.8) 2.3 (1.7–3.3) 15.3 (14.0–16.7) 22.8 (20.3–25.6) 
Stat1 20846 1.0 (0.7–1.5) 1.0 (1.0–1.1) 1.8 (1.4–2.3) 3.3 (2.6–4.1) 
Timp1 21857 1.0 (0.8–1.3) 1.0 (0.9–1.1) 2.6 (2.3–3.0) 5.9 (4.3–8.0) 
Gene SymbolGene ID (Entrez gene)ControlMVLPSMV + LPS
Ccl2 (MCP-1) 20296 1.0 (0.8–1.3) 0.8 (0.8–0.9) 2.8 (2.2–3.6) 5.4 (3.2–9.1) 
Ccl7 (MCP-3) 20306 1.0 (0.8–1.3) 0.9 (0.9–1.0) 1.8 (1.5–2.1) 2.9 (1.9–4.4) 
Cdkn1a (CDK11) 12575 1.0 (0.8–1.3) 1.3 (1.2–1.5) 2.0 (1.6–2.6) 2.1 (1.9–2.2) 
G1p2 (ISG15) 53606 1.0 (0.7–1.4) 0.9 (0.8–1.0) 6.2 (4.9–7.9) 12.1 (9.3–15.8) 
Gadd45α 13197 1.0 (0.9–1.2) 2.1 (1.8–2.4) 2.0 (1.7–2.3) 2.6 (2.2–3.0) 
Ifi1 15944 1.0 (0.8–1.4) 1.0 (0.9–1.2) 3.4 (3.0–3.9) 3.8 (3.1–4.8) 
Ifit1 (GARG16) 15957 1.0 (0.6–1.7) 1.1 (0.9–1.4) 9.4 (8.0–11.0) 12.1 (9.3–15.8) 
Irf7 54123 1.0 (0.9–1.2) 1.1 (1.0–1.3) 3.6 (3.1–4.3) 6.9 (5.2–9.1) 
Lcn2 (Lipocalin 2) 16819 1.0 (0.6–1.7) 0.9 (0.7–1.2) 2.8 (2.7–2.9) 2.7 (2.6–2.9) 
Saa3 20210 1.0 (0.6–1.8) 2.3 (1.7–3.3) 15.3 (14.0–16.7) 22.8 (20.3–25.6) 
Stat1 20846 1.0 (0.7–1.5) 1.0 (1.0–1.1) 1.8 (1.4–2.3) 3.3 (2.6–4.1) 
Timp1 21857 1.0 (0.8–1.3) 1.0 (0.9–1.1) 2.6 (2.3–3.0) 5.9 (4.3–8.0) 

The role of GADD45 up-regulation in the pathogenesis of ALI is unknown; however, one function of GADD45 is to activate MEKK4, which in turn is an upstream regulator of JNK MAPK. JNK phosphorylates and activates members of the AP-1 family of transcription factors and has been implicated in ventilator-induced lung injury (23, 28). Thus, it is a biologically plausible hypothesis that GADD45 may contribute to the pathogenesis of ALI, and further research into its role is warranted.

Although we did not directly test whether lung injury in the MV + LPS group was a consequence of altered gene expression, many of the up-regulated genes are associated with inflammation, which is thought to play a central role in the pathogenesis of ALI. Several animal models of lung injury have shown a partial protective benefit of blocking proinflammatory cytokine activity (29, 30, 31, 32). However, in human sepsis trials, targeting individual cytokines has not reduced the observed incidence of ALI (33). Strategies directed at inhibiting the immunomodulatory role of MV may have an improved chance of reducing the development of lung injury because of a more general attenuation of inflammatory responses to infections or other inflammatory stimuli.

Several additional factors must be considered in interpreting the results of this study. MV in the MV-only and MV + LPS groups required intubation and prolonged anesthesia as compared with the spontaneously breathing mice in the control and LPS-only groups. These differences could have contributed to the augmented cytokine responses seen with MV. Our previous study in rabbits (34), which allowed for better control of these confounding factors, supports the hypothesis that MV is an independent modifier of lung response to LPS. Arterial CO2 partial pressure was not directly measured in these animals, and, the possibility that a respiratory acidosis or alkalosis contributed to the outcome measures cannot be excluded (35, 36). However, our preliminary data showed that the chosen ventilation strategy resulted in a normal arterial pH in normal mice. Thus, the most likely explanation for our data remains that MV enhances the lung inflammatory response to LPS.

We extracted mRNA from whole lung homogenate. Because substantially more PMN cells were recruited to the lungs of animals in the MV + LPS group, some of the gene expression data may have resulted from these changes in the lung cell population. However, analysis of the transcriptionally regulated, intracellular protein GADD45-γ using immunohistochemistry indicated that resident alveolar cells participated in at least some of the observed differential gene expression (Fig. 7). A potential role for alveolar epithelial cells is supported by modulation of TNF-α-induced gene expression in A549 cells after concurrent in vitro cell stretching (37).

In conclusion, we report that a noninjurious, clinically relevant MV strategy can augment the inflammatory response to LPS and promote early ALI. This effect was associated with both an increase in the total number of differentially expressed genes and in modulation of the fold change of most genes compared with the control group. These findings suggest that MV may contribute to the development of ALI in patients with risk factors for ARDS but without pre-existing overt lung injury. Understanding the signaling pathways by which MV augments the transcriptional response to bacterial pathways is an essential first step for developing strategies to disrupt this interaction.

We thank Dowon An and Shen-Sheng Wang for their assistance with animal experiments and Wilson Slaid Jones for his assistance with multiplex cytokine assays.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This research was supported by National Institutes of Health Grants HL71020, HL73996, HL62995, and HL72370 and by the American Heart Association.

3

Abbreviations used in this paper: MV, mechanical ventilation; ALI, acute lung injury; BAL, bronchoalveolar lavage; PMN, polymorphonuclear leukocyte; IRF-7, IFN regulatory factor-7; GADD, growth arrest and DNA damage.

4

The online version of this article contains supplemental material.

1
Padman, R., M. Alexander, C. Thorogood, S. Porth.
2003
. Respiratory management of pediatric patients with spinal cord injuries: retrospective review of the duPont experience.
Neurorehabil. Neural. Repair.
17
:
32
.-36.
2
The ARDS Network.
2000
. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network.
N. Engl. J. Med.
342
:
1301
.-1308.
3
Tremblay, L., F. Valenza, S. P. Ribeiro, J. Li, A. S. Slutsky.
1997
. Injurious ventilatory strategies increase cytokines and c-fos m-RNA expression in an isolated rat lung model.
J. Clin. Invest.
99
:
944
.-952.
4
Whitehead, T. C., H. Zhang, B. Mullen, A. S. Slutsky.
2004
. Effect of mechanical ventilation on cytokine response to intratracheal lipopolysaccharide.
Anesthesiology
101
:
52
.-58.
5
Altemeier, W. A., G. Matute-Bello, C. W. Frevert, Y. Kawata, O. Kajikawa, T. R. Martin, R. W. Glenny.
2004
. Mechanical ventilation with moderate tidal volumes synergistically increases lung cytokine response to systemic endotoxin.
Am. J. Physiol.
287
:
L533
.-L542.
6
Bregeon, F., S. Delpierre, B. Chetaille, O. Kajikawa, T. R. Martin, A. Autillo-Touati, Y. Jammes, J. Pugin.
2005
. Mechanical ventilation affects lung function and cytokine production in an experimental model of endotoxemia.
Anesthesiology
102
:
331
.-339.
7
Gurkan, O. U., C. O’Donnell, R. Brower, E. Ruckdeschel, P. M. Becker.
2003
. Differential effects of mechanical ventilatory strategy on lung injury and systemic organ inflammation in mice.
Am. J. Physiol.
285
:
L710
.-L718.
8
Ranieri, V. M., P. M. Suter, C. Tortorella, R. De Tullio, J. M. Dayer, A. Brienza, F. Bruno, A. S. Slutsky.
1999
. Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: a randomized controlled trial.
J. Am. Med. Assoc.
282
:
54
.-61.
9
Ware, L. B., M. A. Matthay.
2000
. The acute respiratory distress syndrome.
N. Engl. J. Med.
342
:
1334
.-1349.
10
Foster, W. M., D. M. Walters, M. Longphre, K. Macri, L. M. Miller.
2001
. Methodology for the measurement of mucociliary function in the mouse by scintigraphy.
J. Appl. Physiol.
90
:
1111
.-1117.
11
Bolstad, B. M., R. A. Irizarry, M. Astrand, T. P. Speed.
2003
. A comparison of normalization methods for high density oligonucleotide array data based on variance and bias.
Bioinformatics
19
:
185
.-193.
12
Tusher, V. G., R. Tibshirani, G. Chu.
2001
. Significance analysis of microarrays applied to the ionizing radiation response.
Proc. Natl. Acad. Sci. USA
98
:
5116
.-5121.
13
Takekawa, M., H. Saito.
1998
. A family of stress-inducible GADD45-like proteins mediate activation of the stress-responsive MTK1/MEKK4 MAPKKK.
Cell
95
:
521
.-530.
14
Lu, B., H. Yu, C. Chow, B. Li, W. Zheng, R. J. Davis, R. A. Flavell.
2001
. GADD45γ mediates the activation of the p38 and JNK MAP kinase pathways and cytokine production in effector TH1 cells.
Immunity
14
:
583
.-590.
15
Mita, H., J. Tsutsui, M. Takekawa, E. A. Witten, H. Saito.
2002
. Regulation of MTK1/MEKK4 kinase activity by its N-terminal autoinhibitory domain and GADD45 binding.
Mol. Cell Biol.
22
:
4544
.-4555.
16
Chi, H., B. Lu, M. Takekawa, R. J. Davis, R. A. Flavell.
2004
. GADD45β/GADD45γ and MEKK4 comprise a genetic pathway mediating STAT4-independent IFNγ production in T cells.
EMBO J.
23
:
1576
.-1586.
17
Haitsma, J. J., S. Uhlig, R. Goggel, S. J. Verbrugge, U. Lachmann, B. Lachmann.
2000
. Ventilator-induced lung injury leads to loss of alveolar and systemic compartmentalization of tumor necrosis factor-α.
Intensive Care Med.
26
:
1515
.-1522.
18
Murphy, D. B., N. Cregg, L. Tremblay, D. Engelberts, J. G. Laffey, A. S. Slutsky, A. Romaschin, B. P. Kavanagh.
2000
. Adverse ventilatory strategy causes pulmonary-to-systemic translocation of endotoxin.
Am. J. Respir. Crit. Care Med.
162
:
27
.-33.
19
Pugin, J., I. Dunn, P. Jolliet, D. Tassaux, J. L. Magnenat, L. P. Nicod, J. C. Chevrolet.
1998
. Activation of human macrophages by mechanical ventilation in vitro.
Am. J. Physiol.
275
:
L1040
.-L1050.
20
Vlahakis, N. E., M. A. Schroeder, A. H. Limper, R. D. Hubmayr.
1999
. Stretch induces cytokine release by alveolar epithelial cells in vitro.
Am. J. Physiol.
277
:
L167
.-L173.
21
Oudin, S., J. Pugin.
2002
. Role of MAP kinase activation in interleukin-8 production by human BEAS-2B bronchial epithelial cells submitted to cyclic stretch.
Am. J. Respir. Cell Mol. Biol.
27
:
107
.-114.
22
Held, H.-D., S. Boettcher, L. Hamann, S. Uhlilg.
2001
. Ventilation-induced chemokine and cytokine release is associated with activation of nuclear factor-κB and is blocked by steroids.
Am. J. Respir. Crit. Care Med.
163
:
711
.-716.
23
Uhlig, U., J. J. Haitsma, T. Goldmann, D. L. Poelma, B. Lachmann, S. Uhlig.
2002
. Ventilation-induced activation of the mitogen-activated protein kinase pathway.
Eur. Respir. J.
20
:
946
.-956.
24
Grigoryev, D. N., S. F. Ma, R. A. Irizarry, S. Q. Ye, J. Quackenbush, J. G. Garcia.
2004
. Orthologous gene-expression profiling in multi-species models: search for candidate genes.
Genome Biol.
5
:
R34
.
25
Eitzman, D. T., R. D. McCoy, X. Zheng, W. P. Fay, T. Shen, D. Ginsburg, R. H. Simon.
1996
. Bleomycin-induced pulmonary fibrosis in transgenic mice that either lack or overexpress the murine plasminogen activator inhibitor-1 gene.
J. Clin. Invest.
97
:
232
.-237.
26
Carraway, M. S., K. E. Welty-Wolf, D. L. Miller, T. L. Ortel, S. Idell, A. J. Ghio, L. C. Petersen, C. A. Piantadosi.
2003
. Blockade of tissue factor: treatment for organ injury in established sepsis.
Am. J. Respir. Crit. Care Med.
167
:
1200
.-1209.
27
Kaminski, N., J. D. Allard, J. F. Pittet, F. Zuo, M. J. Griffiths, D. Morris, X. Huang, D. Sheppard, R. A. Heller.
2000
. Global analysis of gene expression in pulmonary fibrosis reveals distinct programs regulating lung inflammation and fibrosis.
Proc. Natl. Acad. Sci. USA
97
:
1778
.-1783.
28
Li, L. F., L. Yu, D. A. Quinn.
2004
. Ventilation-induced neutrophil infiltration depends on c-Jun N-terminal kinase.
Am. J. Respir. Crit. Care Med.
169
:
518
.-524.
29
Standiford, T. J., S. L. Kunkel, N. W. Lukacs, M. J. Greenberger, J. M. Danforth, R. G. Kunkel, R. M. Strieter.
1995
. Macrophage inflammatory protein-1α mediates lung leukocyte recruitment, lung capillary leak, and early mortality in murine endotoxemia.
J. Immunol.
155
:
1515
.-1524.
30
Yamamoto, T., O. Kajikawa, T. R. Martin, S. R. Sharar, J. M. Harlan, R. K. Winn.
1998
. The role of leukocyte emigration and IL-8 on the development of lipopolysaccharide-induced lung injury in rabbits.
J. Immunol.
161
:
5704
.-5709.
31
Imai, Y., T. Kawano, S. Iwamoto, S. Nakagawa, M. Takata, K. Miyasaka.
1999
. Intratracheal anti-tumor necrosis factor-α antibody attenuates ventilator-induced lung injury in rabbits.
J. Appl. Physiol.
87
:
510
.-515.
32
Belperio, J. A., M. P. Keane, M. D. Burdick, V. Londhe, Y. Y. Xue, K. Li, R. J. Phillips, R. M. Strieter.
2002
. Critical role for CXCR2 and CXCR2 ligands during the pathogenesis of ventilator-induced lung injury.
J. Clin. Invest.
110
:
1703
.-1716.
33
Cheng, I. W., M. A. Matthay.
2003
. Acute lung injury and the acute respiratory distress syndrome.
Crit. Care Clin.
19
:
693
.-712.
34
Altemeier, W. A., S. McKinney, M. Krueger, R. W. Glenny.
2004
. Effect of posture on regional gas exchange in pigs.
J. Appl. Physiol.
97
:
2104
.-2111.
35
Broccard, A. F., J. R. Hotchkiss, C. Vannay, M. Markert, A. Sauty, F. Feihl, M. D. Schaller.
2001
. Protective effects of hypercapnic acidosis on ventilator-induced lung injury.
Am. J. Respir. Crit. Care Med.
164
:
802
.-806.
36
Sinclair, S. E., D. A. Kregenow, W. J. Lamm, I. R. Starr, E. Y. Chi, M. P. Hlastala.
2002
. Hypercapnic acidosis is protective in an in vivo model of ventilator-induced lung injury.
Am. J. Respir. Crit. Care Med.
166
:
403
.-408.
37
dos Santos, C. C., B. Han, C. F. Andrade, X. Bai, S. Uhlig, R. Hubmayr, M. Tsang, M. Lodyga, S. Keshavjee, A. S. Slutsky, M. Liu.
2004
. DNA microarray analysis of gene expression in alveolar epithelial cells in response to TNFα, LPS, and cyclic stretch.
Physiol. Genomics
19
:
331
.-342.

Supplementary data