Three data sets (GSE11434, GSE58169 and GSE7742) associated with VILI in mice were collected from the GEO database (www.ncbi.nlm.nih.gov/geo) (22–24).

GEO2R (www.ncbi.nlm.nih.gov/geo/info/geo2r.html), an interactive web tool for comparing two or more groups of samples in a GEO series, was used to identify DEGs across experimental conditions. P<0.05 and log fold change >1 were chosen as the cut-off criteria. A Venn diagram, heatmap, and cluster profiler analysis were performed with R 3.5.1 software (cran.r-project.org/bin/windows/base/) in the 'limma' package.

KEGG Orthology Based Annotation System (KOBAS) 3.0 (kobas.cbi.pku.edu.cn) is a web server for gene/protein functional annotation and for identifying the enrichment of gene functional sets. KEGG pathway analysis was performed using the KOBAS 3.0 online tool (25,26). P<0.05 was set as the cut-off criterion.

The animal study protocol was approved by University of Pittsburgh institutional animal care and use committee. Experiments were performed in accordance with the recommendations of the National Institutes of Health Guidelines for the Use of Laboratory Animals (27). Sensitive A/J mice (n=36; 8-9 weeks; 1:1 male:female; 20-25 g) were anesthetized by intraperitoneal pentobarbital injection (70 mg/kg). The trachea was exposed, cut and intubated. To analyze the effects of mechanical ventilation on lung injury, mice (n=6 each) were randomized to low tidal volume (LTV; 6 ml/kg, 140 breaths/min, 0 positive end-expiratory pressure), MTV (12 ml/kg, 100 breaths/min, 0 positive end-expiratory pressure), and spontaneous breathing groups for 6 h interventions, as described previously (28). To examine the Wnt pathway's effects on MTV-induced lung injury, MTV-exposed mice were randomized and intratracheally administered Y27632 [inhibitor of Rho-associated protein kinase 1 (ROCK1); 5 mg/kg], SP600125 [inhibitor of C-Jun N-terminal kinase (JNK); 6 mg/kg] prior to MTV, or were ventilated with MTV alone. Mice were subsequently sacrificed and one section of the right lung was stored in liquid nitrogen for western blotting and ELISAs; other lung sections were fixed in 10% formaldehyde for 24 h at room temperature for immunohistochemistry (IHC) analysis.

The alveolar-capillary permeability was quantified by Evan's blue albumin (EBA) staining as previously described (13). In brief, EBA was prepared by adding bovine serum albumin (BSA; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) to 0.5% EB to a final concentration of 4% (0.6 mM). The solution was thoroughly dissolved by gently stirring with a magnet bar, and the EBA was then sterilely filtered through a 0.22 µm syringe filter and stored in aliquots at -80°C until use. Each aliquot was used only once for each animal to prevent cross-contamination. To evaluate the alveolar capillary barrier function, EBA (20 mg/kg) was administered via the internal jugular vein 1 h before sacrifice and tissue harvesting. At the termination of each experiment, all animals were euthanized, and blood samples were obtained via the right ventricle for plasma EBA measurement. The pulmonary vasculature was then flushed with PBS to remove blood components. The right lung was ligated at the level of the right mainstem bronchus, excised, and weighed and stored in liquid nitrogen for subsequent EB analysis. Once thawed, the lung tissue was homogenized in 2 ml PBS and then incubated with 2 ml formamide (Sigma-Aldrich; Merck KGaA) at 60°C for 18 h. Formamide extracts were centrifuged (Beckman TLX; Beckman Coulter, Inc., Brea, CA, USA) at 15,000×g for 30 min at 4°C, and the centrifuged supernatants were collected to quantify lung EBA content by dual-wavelength spectrophotometry (Beckman DU-640; Beckman Coulter) at 620 and 740 nm (29). EBA permeability index was calculated by dividing the corrected pulmonary tissue EBA absorbance at 620 nm (per g of lung tissue) by the corrected plasma EBA absorbance at 620 nm.

For histological examination, the lung of each animal was fixed in 4% paraformaldehyde overnight at 4°C. Then, 4 mm thick sections were paraffin embedded and stained with hematoxylin and eosin [(H&E) cat. no. P032IH; Auragene Bioscience] at room temperature for 10 min. As described previously, the histological score of the degree of lung injury was graded on a scale from 0 to 4 (0, absent and appeared normal; 1, light; 2, moderate; 3, strong; and 4, intense) for the following pathological features: Alveolar congestion, neutrophil infiltration, hemorrhage and thickness of alveolar wall membrane formation. Five fields of view were assessed (30).

Mouse tumor necrosis factor (TNF)-α (cat. no. ab100747) and interleukin (IL)-6 (cat. no. ab100713) ELISA kits were purchased from Abcam (Cambridge, UK) and used to measure the concentrations of TNF-α and IL-6 in BALF. All reagents were equilibrated to room temperature and prepared prior to use according to the manufacturer's protocol. The standards and samples were added into wells and incubated for 2.5 h at room temperature. The wells were washed with wash buffer four times and biotinylated TNF-α and IL-6 were added into the wells and incubated 1 h at room temperature with gentle shaking. Following four washes, horseradish peroxidase (HRP)-streptavidin solution was added into the wells and incubated for 45 min at room temperature. The wash steps were repeated, and TMB substrate solution was added into the wells and incubated for 30 min in the dark at room temperature. Then, stop solution was added into the wells and a microplate reader was used to determine the optical density of each well at 450 nm.

Total protein concentration in BALF was detected by using bicinchoninic acid (BCA) protein assay kit (cat. no. P001B, Auragene Bioscience) according to the manufacturer's instructions.

Frozen lung tissues were thawed and lysed in radioimmunoprecipitation assay buffer (cat. no. P002A; Auragene Bioscience, Changsha, China) with protease inhibitors (cat. no. P019A; Auragene Bioscience) and phenylmethylsulfonyl fluoride (cat. no. P018A; Auragene Bioscience). Protein concentration was measured with a standard BCA protein assay. Proteins were separated by 10% SDS-PAGE and transferred to nitrocellulose membranes. Membranes were blocked in Tris-buffered saline with 0.5% Tween-20 (TBST) and 3% BSA for 1 h at room temperature and incubated overnight with WISP1 (cat. no. 18166-1-AP; 1:1,000; ProteinTech Group, Inc., Chicago, IL, USA), Ras homolog gene family, member A (RhoA; cat. no. ab54835; 1:1,000; Abcam), phosphorylated (p-)RhoA (cat. no. ab41435; 1:1,000; Abcam), JNK (cat. no. sc-7345; 1:500; Santa Cruz Biotechnology, Inc., Dallas, TX, USA), p-JNK (cat. no. sc-6254; 1:500; Santa Cruz Biotechnology, Inc.) or β-actin (cat. no. ab8226, 1:5,000; Abcam) antibodies. The membranes were washed with TBST five times and incubated with HRP-conjugated goat anti-mouse (cat. no. SA001, 1:15,000) and rabbit (cat. no. SA009; 1:15,000) IgG secondary antibodies (Auragene Bioscience) for 40 min at room temperature. Proteins were detected by chemiluminescence (cat. no. P020WB; Auragene Bioscience) and quantified using ImageJ version 6.0 software (National Institutes of Health, Bethesda, MA, USA).

For histological examination, a lung from each animal was fixed in 4% paraformaldehyde overnight at 4°C. Then, the samples were embedded in paraffin and cut into 3-µm-thick slices. Tissue slides were heated for 2 h at 65°C and subsequently incubated in xylene for 30 min at room temperature and in graded alcohol (100, 95, 85 and 75% alcohol for 5 min each). The slides were blocked in 3% H₂O₂ for 15 min at room temperature to inhibit endogenous peroxidase activity and placed in 1 M sodium citrate buffer (cat. no. P019IH; Auragene Bioscience) for antigen retrieval. The slides were incubated with the primary antibodies against WISP1 (cat. no. 18166-1-AP; 1:100; ProteinTech Group, Inc.), or p-Jnk1/2/3 (cat. no. YP0157; 1:200; ImmunoWay Biotechnology Company, Plano, TX, USA) overnight at 4°C. Then, the slides were rinsed with PBS five times and incubated with HRP-conjugated secondary antibodies (cat. no. SA009; 1:1,000; Auragene Bioscience) for 30 min. The slides were stained with DAB (cat. no. P013IH; Auragene Bioscience) solution for 1 min, counter-stained with hematoxylin (cat. no. C0107; Beyotime Institute of Biotechnology, Haimen, China) for 15 min at room temperature and dehydrated in a graded alcohol (85 and 95% for 5 min each) at room temperature prior to coverslipping.

Data are presented as the mean ± standard error of the mean, and were analyzed by one-way analysis of variance followed by Tukey's post-hoc test. Statistical analyses were performed using GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

A total of 157, 757 and 1,086 DEGs were identified from GSE11434, GSE58169, and GSE7742, respectively. There were 10 common DEGs from the three datasets that were associated with lung injury (Fig. 1), including IL-1β, IL-1 receptor 2, stratifin, C-type lectin domain family 4 member D, amphiregulin, suppressor of cytokine signaling 3, S100 calcium binding protein A1, matrix metalloproteinase 8, solute carrier family 7 member 11 and early growth response 1.

KOBAS 3.0 was used to analyze the pathway enrichment of identified DEGs. KEGG analysis revealed that the upregulated DEGs were predominantly enriched in 'HTLV-1 infection', 'Wnt signaling pathway' and 'MAPK signaling pathway', as well as pathways associated with VILI (31,32) (Fig. 2A and B). It was demonstrated that 'HTLV-1 infection' and 'MAPK signaling pathway' were associated with 'Wnt signaling pathway' (Fig. 3). It has been reported that Wnt/β-catenin signaling pathway is activated in early VILI, and inhibition of the Wnt/β-catenin signaling pathway suppresses VILI (31,33) Thus, the Wnt pathway was selected for further study. Heatmap analysis revealed that Wnt1 and GSK3β were significantly increased in VILI, compared with the control groups (Fig. 4A). In addition, the non-canonical signaling pathway factors Wnt5a, Wnt11 and ROCK2 were activated in VILI (Fig. 4B). The KEGG pathway enrichment demonstrated a relationship between the identified DEGs and VILI. These analyses support the reliability of the results in the present study.

To investigate the role of non-canonical Wnt signaling in VILI, animal experiments in sensitive A/J mice. To assess VILI, alveolar-capillary permeability was measured, H&E staining was performed, and the expression of TNF-α and IL-6, as well as total protein, was measured. Permeability was increased in the lungs of A/J mice ventilated with LTV and MTV compared with the spontaneous breathing group (Fig. 5A). Histological examination illustrated that ventilation with LTV and MTV significantly increased inflammatory cell infiltration, pulmonary edema and alveolar septal thickening, findings which were confirmed by histological scoring (Fig. 5B and C). Similarly, increased TNF-α (Fig. 5D), IL-6 (Fig. 5E), and total protein (Fig. 5F) expression was detected in the LTV and MTV groups. Collectively, these results demonstrated that mechanical ventilation with LTV or MTV induced lung injury in A/J mice. To investigate the role of non-canonical Wnt signaling in VILI, the protein levels of important non-canonical Wnt signaling pathway components were examined via western blotting and IHC. RhoA and JNK are important proteins in non-canonical Wnt signaling (34). Mechanical ventilation did not alter the expression of RhoA or JNK. However, the expressions of p-RhoA, ROCK1 and p-JNK were elevated in the LTV and MTV groups, compared with the spontaneous breathing group (Fig. 6A and B). Furthermore, IHC revealed that p-JNK was dramatically increased in the LTV and MTV groups, compared with the spontaneous breathing group (Fig. 6C). Taken together, these data indicated that non-canonical Wnt signaling pathway participated in the pathogenesis of VILI in A/J mice.

To further investigate the functional role of non-canonical Wnt signaling in the pathogenesis of VILI, inhibitors of target proteins in the pathway were used. Sensitive A/J mice were given Y27632 (ROCK1 inhibitor) or SP600125 (JNK inhibitor) intratracheally before undergoing ventilation with MTV for 6 h. Y27632 and SP600125 significantly decreased MTV-induced EBA permeability (Fig. 7A). Histological analysis and scoring illustrated that Y27632 and SP600125 attenuated MTV-induced inflammatory cell infiltration, pulmonary edema and alveolar septal thickening (Fig. 7B and C). Total protein concentration, as well as TNF-α and IL-6 expression in BALF were notably decreased following treatment with Y27632 or SP600125 (Fig. 7D and E). In addition, the protein expression of JNK, p-JNK, RhoA, p-RhoA, and ROCK1 was measured by western blotting and IHC. Western blot analysis showed that inhibition of ROCK1 (Y27632) and JNK (SP600125) dramatically decreased p-JNK, p-RhoA, and ROCK1 levels compared to the MTV group (Fig. 7F and G). IHC confirmed that p-JNK (Fig. 7H) expression was attenuated in VILI following administration of Y27632 or SP600125. Collectively, these data indicated that inhibition of non-canonical and canonical Wnt signaling pathways may effectively protect lung tissues against VILI in A/J mice.

Western blotting was performed to analyze the regulation of WISP1 by the non-canonical Wnt pathway. The results revealed that WISP1 expression was increased in mice ventilated with LTV and MTV, compared with the control group (Fig. 8A). The inhibition of ROCK1 and JNK markedly decreased MTV-induced WISP1 expression (Fig. 8B and C). These results revealed that non-canonical Wnt signaling promoted VILI by upregulating WISP1 expression in A/J mice.