LPS (Escherichia coli LPS serotype 0111:B4), sodium pentobarbital, Evans blue dye and dihydroethidium (DHE), goat serum (G9023) and bovine serum albumin (A2153) were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Recombinant human vaspin (rh-vaspin) protein was purchased from GeneTex, Inc. (Irvine, CA, USA). The following primary antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA): Anti-ICAM-1 (cat. no. 4915), anti-NF-κB Rel (cat. no. 8242), anti-phosphorylated (p)-NF-κB Rel (Ser536; E1Z1T; cat. no. 13346), anti-Akt (cat. no. 4685) and anti-p-Akt (Ser473; cat. no. 4060), anti-rabbit IgG, HRP-linked antibody (cat. no. 7074) and Alexa Fluor 488-labeled anti-rabbit secondary antibodies (cat. no. 4412). Anti-vascular endothelial (VE)-cadherin (ab205336) and anti-NADPH oxidase antibodies (ab133303) were purchased from Abcam (Cambridge, UK). Anti-GAPDH (cat. no. AP0063), anti-β-catenin (cat. no. BS6879), anti-glycogen synthase kinase (GSK)-3β (cat. no. BS1402) and anti-p-GSK3β (Ser9; cat. no. BS4084P) antibodies were purchased from Bioworld Technology (Nanjing, China). Adenoviral vectors (Ad) expressing β-galactosidase (Ad-β-gal) and full-length vaspin (Ad-vaspin) were constructed by Shanghai GeneChem Co., Ltd. (Shanghai, China) (primer sequences available upon request). Ad-β-gal was used as a control.

C57BL/6 mice (weighing 20–25 g; 10 weeks old; male; of specific-pathogen-free grade; Department of Laboratory Animal Center, Chongqing Medical University, Chongqing, China) were housed under light-controlled conditions (12-h light/dark cycle) at room temperature (25°C) with 60% humidity and were granted ad libitum access to food and water. All animal experimental protocols were implemented in accordance with the instructions of the National Institutes of Health Guide for the Care and Use of Laboratory Animals (29). The present study was approved by the Ethics Committee of the Second Affiliated Hospital of Chongqing Medical University (Chongqing, China). Mice were randomly assigned to 4 groups as follows: the control group not pre-treated with Ad-vaspin (n=24), the control group pre-treated with Ad-vaspin (n=24), the ARDS group not pre-treated with Ad-vaspin and the ARDS group treated with Ad-vaspin (n=24). Mice were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) prior to exposure of the trachea and right internal jugular vein (IJV). Subsequently, 3×10⁷ PFU Ad-vaspin or Ad-β-gal per mouse was injected into the IJV for 3 days prior to LPS or vehicle (PBS) intratracheal instillation. Mice were anesthetized with sodium pentobarbital (50 mg/kg, i.p.). As described previously (18), to establish a mouse model of ARDS, mice were intratracheally instilled with 5 mg/kg LPS (E. coli LPS serotype 0111:B4) in 50 µl sterile PBS, or PBS alone (control group) using an 18-G catheter. The mice were sacrificed 4 h after the LPS injection and lung tissues were harvested and stored at −80°C until further analysis.

Left lung lobes were harvested, fixed in 3.7% paraformaldehyde (24 h at room temperature), embedded in paraffin wax and cut into 5-µm sections. Subsequently, the sections were stained with hematoxylin and eosin. Histological lung injury in each mouse was evaluated in 5 random fields (×200 and ×400 magnification) using an inverted microscope (TE2000-U; Nikon Corporation, Tokyo, Japan). A standardized scoring system, published by the American Thoracic Society (30), was used to assess histological lung injury in mice.

Lung homogenates were used to determine the levels of TNF-α (MTA00B), interleukin (IL)-6 (M6000B) and IL-10 (M1000B) using commercially available ELISA kits (R&D Systems, Inc., Minneapolis, MN, USA) according to the manufacturer's protocols. In cultured cells, basal levels of secretion were determined by measuring TNF-α, IL-6 and IL-10 concentration in cell lysates. The effectiveness of the adenoviral vector expression system was confirmed by measuring mean plasma vaspin concentrations in the treated mice using a commercial ELISA kit (DSA120; R&D Systems, Inc.).

Mouse left lung tissue sections from each group were deparaffinized with xylene, rehydrated in gradient ethanol and incubated in 3% H₂O₂ at 37°C for 15 min. Subsequently, the sections were rinsed three times in PBS (10 min/wash). Antigen retrieval was performed by immersing the slices in citrate buffer in a microwave at 96°C for 30 min. Tissues were blocked with goat serum albumin in an incubator at room temperature for 1 h and the sections were then incubated with anti-ICAM-1 primary antibodies (1:50 dilution) at 4°C overnight. Sections were washed a further three times with PBS (10 min/wash) and were incubated with a biotin-labeled secondary antibody (1:1,000; ZB-2010; ZSBIO Biotech Co., Ltd., Beijing, China) at 37°C for 30 min, and were then stained with DAB. Sections were counterstained with hematoxylin, dehydrated in gradient ethanol, vitrified with xylene and sealed with neutral resins. Images were captured using an inverted microscope (TE2000-U; Nikon Corporation).

Pulmonary capillary permeability was assessed by determining EBDA concentrations. The right IJV of mice was injected with EBDA (30 mg/kg). Then mice were sacrificed and lungs free of blood were excised, weighed and homogenized in 1 ml PBS, after which they were extracted in 2 ml formamide (24 h, 60°C) and centrifuged at 5,000 x g for 30 min at 20°C. The absorbance of the supernatants was measured by spectrophotometry at 620 and 740 nm, plotted against a standard curve, normalized, and converted to µg EBDA/g lung tissue.

A trimmed 18-G catheter was inserted into the trachea. A syringe was connected to the catheter, and 1 ml of sterile normal saline was infused into the airway. Four hours after LPS administration, BALF was collected by the intra-tracheal instillation of 1 ml of sterile normal saline followed by repeated aspiration 3 times and centrifugation at 500 x g for 10 min at 4°C. According to manufacturer's instructions, the protein concentrations in the BALF supernatants were determined using a bicinchoninic acid protein assay (BCA) kit (KeyGen Biotech Co., Ltd., Nanjing, China).

The right upper lung lobes were harvested and weighed to determine wet lung weight. Subsequently, tissues were dried in an oven at 80°C for 24 h and were weighed again to calculate the W/D ratios.

Human pulmonary microvascular ECs (HPMECs) were cultured in EC medium (both ScienCell Research Laboratories, Inc., San Diego, CA, USA) supplemented with 10% fetal bovine serum (FBS; cat. no. 0025), 1% endothelial cell growth supplement (ECGS; cat. no. 1052) and 1% penicillin/streptomycin (P/S; cat. no. 0503) (all from ScienCell Research Laboratories, Inc.) in a 5% CO₂ incubator at 37°C. Cells between passages 4 and 10 were grown as a monolayer and starved (1% serum) for 6 h prior to each treatment. Human serum concentrations of vaspin are reported to range between 0.1 and 100 ng/ml; therefore, 10 ng/ml vaspin was used in the present study. Cells were pretreated with rh-vaspin (10 ng/ml) or PBS as a control for 24 h, after which they were washed with PBS and exposed to LPS or vehicle (PBS) at 100 ng/ml for 2 h. Cell lysates were collected for subsequent analysis at the indicated time intervals.

Permeability was determined based on the paracellular permeability of 70 kDa fluorescein isothiocyanate (FITC)-dextran into the lower chamber as described previously (18). Briefly, HPMECs were grown on 0.4 µm Transwell inserts at 1×10⁵ cells per well. Following the indicated time interval for each treatment, 0.5 ml FITC-dextran (1 mg/ml) was added to the upper wells, and 1.5 ml medium was added to the bottom chamber. Following 1 h incubation in the dark, 50 µl medium was aspirated and absorbance was measured using a luminometer (BioTek Instruments, Inc., Winooski, VT, USA) at an excitation wavelength of 488 nm and an emission wavelength of 520 nm. The basal FITC-dextran permeability for unstimulated monolayers was set at 100%.

Cell viability was measured using Cell Counting kit-8 (CCK-8). Cell suspensions of each group were seeded in 96-well plates at 2×10⁴ cells/well and were preincu-bated at 37°C in a humidified atmosphere containing 5% CO₂. Subsequently, 10 µl CCK-8 solution was added to each well, and the plates were incubated for 2 h in an incubator. The absorbance of each well was measured using a microplate reader at 450 nm (BioTek Instruments, Inc.) at room temperature (25°C). Cell viability was calculated using the following equation: Viability = (OD_{test group} − OD_{blank group}) / (OD_{control group} − OD_{blank group}) × 100%; where OD refers to optical density.

According to the manufacturer's protocol, TUNEL staining was conducted using an in situ cell death detection kit (Roche Diagnostics GmbH, Mannheim, Germany) to detect apoptosis of HPMECs, as described previously (18). DAPI (Nanjing KeyGen Biotech Co., Ltd.) was used to stain nuclei. In situ apoptosis detection (Abcam, Cambridge, MA, USA) was used to assess the apoptotic rate of paraffin-embedded lung sections according to the manufacturer's protocols. DAB reacted with the horseradish peroxidase (HRP)-labeled sample to generate a dark brown signal at the site of DNA fragmentation. TUNEL-positive cells were counted in 5 randomly selected fields (×400 magnification) under an inverted microscope (TE2000-U; Nikon Corporation).

Following the indicated treatments, cells were collected and resuspended in 500 µl 1X Annexin V binding buffer, after which they were incubated with 5 µl Annexin V-FITC and 5 µl PI for 15 min at room temperature in the dark. Flow cytometry (FCM; BD Biosciences, Franklin Lakes, NJ, USA) was performed to detect apoptotic cells.

Radioimmunoprecipitation assay buffer was used to extract total protein, and a Membrane and Cytoplasmic Protein Extraction kit (both Nanjing KeyGen Biotech Co., Ltd.) was used to extract protein from cells and the left lung tissues of mice from each treatment group according to the manufacturer's protocols. Bicinchoninic acid kit was used to measure protein concentration. Equivalent amounts of protein (30 µg) were separated by SDS-PAGE and were electrotransferred to polyvinylidene fluoride membranes. Subsequently, the membranes were blocked with 5% dry milk/bovine serum albumin (BSA) at room temperature (25°C) for 1 h, and were immunoblotted with anti-NF-κB Rel (1:1,000), anti-phospho-NF-κB Rel (Ser536; 1:1,000), anti-β-catenin (1:500), anti-VE-cadherin (1:1,000), anti-NADPH oxidase antibody (1:500), anti-GSK-3β (1:500), anti-phospho-GSK-3β (Ser9; 1:500) and anti-GAPDH (1:8,000) primary antibodies overnight at 4°C, followed by incubation with the corresponding HRP-conjugated secondary antibodies (1:5,000). Protein bands were detected according to an enhanced chemiluminescence (ECL) method (EMD Millipore, Billerica, MA, USA) using a Bio-Rad Gel Imaging system and were analyzed with Quantity One software version 4.4.0 (both Bio-Rad Laboratories, Inc., Hercules, CA, USA). The expression levels were determined by measuring the corresponding band intensities.

Total RNA was isolated from cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's protocol. The relative RNA levels were quantified using a NanoDrop 2000 spectrophotometer (NanoDrop; Thermo Fisher Scientific, Wilmington, DE, USA). Subsequently, 1 µg RNA was used as a template for the generation of cDNA using a HiScript First Strand cDNA synthesis kit (Vazyme, Piscataway, NJ, USA). The reverse transcription reaction conditions were 25°C for 5 min, 42°C for 15 min, and 85°C for 5 min. Gene expression levels of TNF-α, IL-6, VE-cadherin and β-catenin were detected using a HiScript^® II One Step qRT-PCR SYBR^®-Green kit (Vazyme) and StepOne Real-Time PCR apparatus (Applied Biosystem; Thermo Fisher Scientific, Inc., Waltham, MA, USA). Polymerase chain reaction conditions were as follows: pre-denaturation at 95°C for 30 sec, 40 cycles of denaturation at 95°C for 5 sec, annealing at 55°C for 30 sec, and polymerization at 72°C for 30 sec. Relative gene expression levels were normalized to GAPDH using CFX manager softwareversion 3.0 (Bio-Rad Laboratories, Inc.) according to the comparative Cq (ΔΔCq) method (31). The primer sequences are presented in Table I.

Coverslips (3 from each group) of the HPMECs (1×10⁵ cells per well) were fixed with 3.7% paraformaldehyde, permeabilized with 0.5% Triton X-100, blocked with PBS containing 5% goat serum and incubated with anti-NF-κB p65 antibodies (1:400) at 4°C overnight. Subsequently, coverslips were incubated with Alexa Fluor 488-labeled secondary antibodies (1:500) for 1 h in the dark. Coverslips were rinsed three times with PBS, and the nuclei were stained with DAPI (Nanjing KeyGen Biotech Co., Ltd.) for 5 min. Images were captured by inverted microscopy (TE2000-U; Nikon Corporation) after washing.

HPMECs (1×10⁵ cells per well) were seeded on coverslips (3 from each group) and were fixed in 3.7% paraformaldehyde in PBS. The coverslips were then rinsed with PBS, permeabilized with 0.1% Triton X-100 in PBS, preincubated with PBS containing 1% BSA and stained with 200 µl Fluorescent-iFluor™ 594 phalloidin solution (Nanjing KeyGen Biotech Co., Ltd.) for 30 min at room temperature. Finally, coverslips were washed, sealed and images were captured by inverted microscopy (TE2000-U; Nikon Corporation).

Intracellular ROS production was detected using the superoxide indicator dihydroethidium (DHE; D1168; Invitrogen; Thermo Fisher Scientific, Inc.). HPMECs were pretreated with rh-vaspin (10 ng/ml) or PBS as a control for 24 h, and were then exposed to PBS or LPS (100 ng/ml) for 4 h. HPMECs were washed twice with PBS, incubated in fresh culture medium without FBS, and were incubated with DHE (10 µM dissolved in dimethyl sulfoxide) for 15 min at 37°C. Once the DHE probe is oxidized by ROS to 2-hydroxyethidium, it intercalates within DNA, resulting in the fluorescent red staining of cell nuclei. Images of the cells were captured using an inverted fluorescence microscope (TE2000-U; Nikon Corporation) from three different fields of view. Fluorescence intensity was measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA), averages were calculated and were normalized to the control.

Following exposure of HPMECs to PBS or LPS (100 ng/ml) for 4 h in the absence or presence of rh-vaspin (100 ng/ml) for 24 h, total cell lysates were harvested. Nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (NOX) activity was determined using a lucigenin assay. As a chemiluminescent probe, lucigenin is used to indicate the presence of superoxide anion radicals in cells. The reaction was performed in a total volume of 200 µl assay buffer containing 10 µM lucigenin (L-6868; NSC-151912; MedChem Express, Monmouth Junction, NJ, USA), 1 mM NADPH, and 15 µg cell lysates. Following incubation for 30 min at 37°C in the dark, absorbance was measured using a luminometer (BioTek Instruments, Inc.) at an excitation wavelength of 540 nm and an emission wavelength of 605 nm. Samples were well mixed and chemiluminescence was continuously measured for 30 min. Chemiluminescence of relative light units per second (RLU/sec) was obtained every 10 sec and the results were calculated as area under the curve and normalized to the control.

Data are presented as the mean ± standard deviation. Results are representative of at least three independent experiments performed in triplicate. Unpaired Student's t-test was performed for comparisons between two independent groups. One-way analysis of variance (ANOVA) followed by the Student-Newman-Keuls post hoc test was performed to compare continuous variables with normal distribution from three or more independent groups, and to detect significant differences between particular groups. Kruskal-Wallis ANOVA followed by Mann-Whitney U test was performed to compare continuous variables with abnormal distribution from three or more independent groups, and to detect significant differences between particular groups. P<0.05 (95% confidence interval) was considered to indicate a statistically significant difference at. All statistical analyses were conducted using GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, USA).

In order to explore the effects of vaspin on lung histological injury in an LPS-induced mouse model of ARDS, lung histopathological examination was conducted. Mice were systemically instilled with Ad-vaspin or Ad-β-gal as control (3×10⁷ PFU/mouse for 3 days) and were subjected to intratracheal injection with LPS (5 mg/kg) to establish a mouse model of ARDS or with PBS as a control. Plasma vaspin levels were enhanced to 99.0±7.9 ng/ml on day 3 after Ad-vaspin injection, which could not be detected in the control mice treated with Ad-β-gal (data not shown). Mice that were injected with LPS developed ARDS as early as 4 h after LPS insult, as indicated by histopathological alterations in the lungs, including increased inflammatory cell infiltration, thickened alveolar septum, intra-alveolar and interstitial edema fluid, and patchy areas of hemorrhage (Fig. 1A). These histopathological alterations were significantly attenuated following administration of Ad-vaspin (Fig. 1A and B). Since lung injury in ARDS is associated with an exaggerated inflammatory response, the present study further examined the effects of vaspin on pulmonary inflammation. Consistent with the results of lung injury assessment, expression levels of the proinflammatory cytokines TNF-α (Fig. 1C) and IL-6 (Fig. 1D) were reduced in Ad-vaspin-pretreated mice compared with in Ad-β-gal-pretreated mice 4 h post-LPS insult, whereas expression of the anti-inflammatory cytokine IL-10 (Fig. 1E) was increased. In addition, the expression levels of the adhesion molecule, ICAM-1 (Fig. 1F), were reduced in Ad-vaspin-pretreated mice compared with in Ad-β-gal-pretreated mice 4 h post-LPS insult. These findings suggested that vaspin may exert an anti-inflammatory effect on lung injury in a mouse model of LPS-induced ARDS.

Vaspin has previously been reported to exert anti-inflammatory effects on various types of vascular ECs (24–27). Therefore, it may be hypothesized that vaspin mediates mitigation of inflammation in pulmonary ECs. To assess the anti-inflammatory effects of vaspin in vitro, the mRNA expression levels of inflammatory cytokines (TNF-α and IL-6) and endothelial-specific adhesion markers (VCAM-1 and E-selectin) were analyzed in HPMECs 2 h after LPS insult. Consistent with the LPS-mediated inflammatory response in lung tissue, the expression levels of TNF-α (Fig. 2A), IL-6 (Fig. 2B), VCAM-1 (Fig. 2C) and E-selectin (Fig. 2D) were significantly increased in HPMECs following LPS administration, which were reversed by pretreatment with rh-vaspin. The observed anti-inflammatory effects of vaspin were further confirmed by examining the activation of NF-κB, which is a pivotal inflammatory mediator; the results indicated that the phosphorylation and nuclear translocation of the NF-κB Rel subunit were reduced following treatment of HPMECs with rh-vaspin 2 h after LPS insult (Fig. 2E and F). Taken together, these findings indicated that vaspin may exert a protective role during the early stage of LPS-induced ARDS via the suppression of EC inflammation.

Pulmonary microvascular hyperpermeability is a common feature of ARDS; therefore, it was assessed in vitro and in vivo in the present study (Fig. 3). To address whether vaspin modulates pulmonary endothelial barrier function in vivo, total bronchoalveolar lavage fluid (BALF) protein concentrations, EBDA extravasation and lung W/D ratios were analyzed in mice with LPS-induced ARDS. Lung histological damage was observed to be associated with pulmonary microvascular hyperpermeability following LPS instillation, as manifested by increases in BALF protein concentrations (Fig. 3A), EBDA extravasation (Fig. 3B) and W/D ratios (Fig. 3C) at 4 h after LPS treatment. The time point selected to evaluate the effects of vaspin on ARDS was based on its coincidence with the histological onset of lung injury in mice. Pretreatment with Ad-vaspin significantly attenuated pulmonary microvascular hyperpermeability in a murine model of ARDS (Fig. 3A–C). In addition, to further confirm the ability of vaspin to mitigate LPS-induced EC hyperpermeability in vitro, HPMECs were cultured in the presence or absence of rh-vaspin (10 ng/ml) for 24 h, and the influx of FITC-dextran was measured. Treatment with a physiological concentration of rh-vaspin prevented LPS-induced increases in the influx of FITC-dextran (Fig. 3E). These data indicated that vaspin may significantly attenuate LPS-challenged pulmonary microvascular hyperpermeability in vivo and EC barrier dysfunction in vitro.

The stabilization of interendothelial AJs and the actin cytoskeleton are essential for a restrictive pulmonary EC barrier; lung endothelium permeability can increase due to alterations in AJs and the endothelial cyto-skeleton. Therefore, to further address the effects of vaspin on vascular homeostasis, the expression levels of VE-cadherin and β-catenin, which are two important AJ proteins of ECs, were evaluated in the lung tissue of mice. Notably, analysis of AJs expression did not detect a difference in the Ad-vaspin-pretreated mice compared with those treated with LPS alone (Fig. 3D). In addition, LPS caused a reduction in the mRNA expression levels of AJs (Fig. 3F and G), as well as cell retraction, F-actin reorganization and stress fiber formation in HPMECs, as determined by phalloidin staining (Fig. 3H). However, consistent with the results of the in vivo experiments, AJs gene expression and distribution of the actin cytoskeleton in vaspin-treated HPMECs were not altered compared with the controls treated with LPS alone (Fig. 3F–H). Collectively, these findings indicated that vaspin may exert beneficial effects on pulmonary microvascular hyperpermeability via restoring EC barrier function; however, these effects may not depend on the ability of vaspin to regulate interendothelial AJs or the endothelial cytoskeleton but may be dependent on other mechanisms associated with endothelial barrier integrity.

Vaspin has been reported to inhibit the apoptosis of ECs, including human aortic ECs, HUVECs and smooth muscle cells (SMCs). Furthermore, the function of the pulmonary endothelial barrier depends on its integrity (24,25,27). Focusing on the beneficial effects of vaspin on the function and integrity of pulmonary ECs at the cellular level, HPMECs were treated with LPS in the presence or absence of rh-vaspin (10 ng/ml, pretreatment for 24 h). CCK-8, TUNEL staining and FCM analyses were performed to ascertain whether vaspin modulates the survival and apoptosis of pulmonary ECs following exposure to LPS (Fig. 4). In HPMECs, pretreatment with rh-vaspin significantly promoted pulmonary EC survival and suppressed pulmonary EC apoptosis under LPS stimulus compared with those pretreated with PBS, as determined by increased cell viability (Fig. 4B), a reduction in the number of TUNEL-positive cells (Fig. 4A and C) and reduced apoptotic rate, as determined by FCM (Fig. 4A and D). In situ apoptotic detection was further performed to verify the effects of vaspin on apoptosis in lung tissues. Compared with in the control group, Ad-vaspin markedly reduced the number of TUNEL-positive cells in lung tissues post-LPS insult (Fig. 4E). These results suggested that vaspin exerts a protective role on HPMECs, at least partially via its prosurvival and anti-apoptotic properties.

A previous study demonstrated that ROS and NOX mediated EC apoptosis under various insults. Furthermore, vaspin has been demonstrated to inhibit ROS generation by suppressing NOX activation in ECs, including HUVECs. Therefore, the present study further evaluated the effects of vaspin on intracellular ROS production by measuring the fluorescence intensity of the intracellular fluorescent probe, DHE. Compared with HPMECs exposed to LPS alone, pretreatment with rh-vaspin significantly counteracted LPS-induced ROS production in HPMECs (Fig. 5A and B). Constitutive NOX functions as an oxygen sensor that regulates intracellular superoxide organization. Therefore, to elucidate the upstream mechanisms, the effects of vaspin on NOX were further assessed by measuring NOX activity and expression. The results indicated that LPS induced an increase in NOX activity (Fig. 5C) and expression (Fig. 5D and E); however, this was markedly abrogated by pretreatment of HPMECs with vaspin. These findings indicated that the protective effects of vaspin against LPS insults in ECs may be associated with antioxidative properties.

The PI3K/Akt signaling pathway acts as a compensatory regulator of ARDS through its anti-inflammatory, anti-apoptotic and antioxidant effects in response to numerous growth factors. Therefore, to assess the effects of vaspin on the activation of Akt-related signaling in vivo and in vitro, the phosphorylation of Akt and its downstream target GSK3β were assessed by western blotting. The results indicated that p-Akt and p-GSK3β levels were low under non-stressed conditions but were increased in a mouse model of LPS-induced ARDS; however, the differences between these groups were not statistically significant, thus suggesting an endogenous negative feedback mechanism underlying the effects of LPS on the PI3K/Akt pathway. Notably, administration of vaspin enhanced the phosphorylation of Akt and GSK3β in mouse lungs subjected to LPS (Fig. 6A). At the cellular level, rh-vaspin stimulated the phosphorylation of Akt and GSK3β (Fig. 6B). Collectively, these data suggested that vaspin acts as a stimulatory factor for the Akt-GSK3β signaling pathway, and vaspin-induced protection of pulmonary ECs during ARDS may be mediated, at least partially, by activating the Akt-GSK3β signaling pathway.