Sprague-Dawley (SD) rats (male, 5–7 weeks old, 200±20 g; n=32) were obtained from the Animal Center of Fourth Military Medical University (Xi'an, China). The feeding environment of rats includes temperature-controlled house with 12 h light-dark cycles, free access to standard laboratory diet and water ad libitum. All the animal experiments were approved by the Animal Care and Use Committee of the Fourth Military Medical University and in accordance with the Declaration of the National Institutes of Health Guide for Care and Use of Laboratory Animals (Bethesda, MD, USA; publication No. 85-23, revised 1985).

Seawater (osmolality 1,300 mmol/l, pH 8.2, relative density 1.05, NaCl 6.518 g/l, MgSO₄ 3.305 g/l, MgCl₂ 2.447 g/l, CaCl₂ 1.141 g/l, KCl 0.725 g/l, NaHCO₃ 0.202 g/l, NaBr 0.083 g/l, diluted by deionized water) was obtained according to the major composition of the East China Sea provided by Chinese Ocean Bureau (Beijing, China). ELISA kits for tumor necrosis factor (TNF)-α (cat. no. PRTA00) and interleukin (IL)-1β (cat. no. PRLB00) were obtained from R&D Systems, Inc. (Minneapolis, MN, USA). Recombinant rat SEMA7A (rSEMA7A) was obtained from R&D Systems, Inc. Antibody for plexin C1 (cat. no. AF5375) and β1 integrin (cat. no. AF2405) were obtained from R&D Systems, Inc. Anti-NF-κB p65 (cat. no. 3033), anti-pNF-κB p65 (cat. no. 3036), anti-p-cofilin (cat. no. 5175), anti-cofilin (cat. no. 3313) antibodies were obtained from Cell Signaling Technology, Inc. (Cell Signaling Technology, Inc., Danvers, MA, USA). The anti-β-actin (cat. no. sc-47778) antibody were obtained from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Endothelial cell growth supplement was obtained from Upstate Biotechnology, Inc. (Lake Placid, NY, USA).

Primary RPMVEC isolation and culture were performed according to previous methods with some modification (19). Firstly, the pleura and outer edges of washed (with PBS) fresh rat lung lobe were cut off. Then, the 1.5 mm³ specimens of tissue cut from lung surface were carefully plated into cell culture dishes (containing DMEM supplemented with 20% FBS, 25 µg/ml endothelial cell growth supplement and 100 U/ml penicillin-streptomycin). These tissues were cultured at 37°C in a humidified atmosphere with 5% CO₂ and 95% air. The residue specimens were removed following 60 h. The cells were passed (with 0.25% trypsin) when monolayer cells were achieved and all experimental cells were between passage 2 and 3. Primary RPMVECs were identified according to their characteristic morphology and staining with anti-CD31 antibody (18). In some of these experiments, RPMVECs were pre-treated with or without anti-plexin C1 antibody (5 µg/ml) or isotype antibody (5 µg/ml) for 1 h at 37°C before stimulation. The plexin C1 antibody based blocking was performed as previously described (14). SEMA7A small interfering (si)RNA (the sequence is presented in Table I) and scramble control siRNA were obtained from GeneChem Co., Ltd. (Shanghai, China). The siRNA was transiently transfected into RPMVECs with Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to manufacturer's instructions and suppression efficiency was detected by reverse transcription-quantitative polymerase chain reaction. Total RNA was extracted with TRIzol reagent (cat no. 9108; Takara Biotechnology Co., Ltd., Dalian, China) according to the manufacturer's protocol. RNA concentration was tested by spectrometric analysis. The PrimeScript RT reagent kit (cat no. RR037A; Takara Biotechnology Co., Ltd.) and the One Step SYBR PrimeScript RT-PCR kit (cat no. RR066A; Takara Biotechnology Co., Ltd.) were used. SEMA7A and β-actin were examined using a Real Time PCR instrument following the manufacturer's protocol (Takara Biotechnology Co., Ltd.). Amplification and detection were carried out using a Bio-Rad My iQ detection system (Edinburgh Biological Science and Technology Development co. Berkeley, CA, USA). The PCR cycling conditions were as follows: Pre-denaturation for 1 min at 94°C, denaturation for 30 sec at 94°C, annealing for 30 sec at 55°C, extension for 1 min at 72°C, for a total of 30 cycles. The sequences of the rat SEMA7A primers were: 5′-CGAGTGGCCCAGTTATGCAG-3′ (forward) and 5′-AAACCAGCATGGCTTTCAGGA-3′ (reverse). The sequences of the rat β-actin primers were 5′-ACGGTCAGGTCATCACTATCGG-3′ (forward) and. 5′-GCACTGTGTTGGCATAGAGGTC-3′ (reverse). The 2^−∆∆Cq method was used for quantification (20). The knockdown efficiency of the siRNA was verified in our previous study (18). After 48 h transfection, RPMVECs were stimulated with 25% seawater or co-culture, and 25% deionized water stimulation was used as a control.

RPMVECs (5×10⁵ cells/well in a six-well plate) were stimulated with or without 25% seawater for 24 h and fixed with 2% paraformaldehyde. Following washing, fixed RPMVECs were co-cultured with NR8383 cells (American Type Culture Collection, Manassas, VA, USA; 5×10⁵ cells/ml) for 24 h. In some experiments, NR8383 cells were pre-treated with or without anti-β1 integrin antibody or isotype antibody (5 µg/ml) for 1 h before co-culture. The β1 integrin antibody based blocking was performed as previously described (14). In addition, RPMVECs or NR8383 cells were preconditioned by culturing with rSEMA7A (5 µg/ml) and incubated for 24 h before stimulation in other experiments.

TNF-α and IL-1β obtained from culture medium were detected with ELISA kits according to the manufacturer's protocol. To detect the proinflammatory cytokine concentration in cells, the culture medium (100 µl) from NR8383 cells was collected. The concentration of proinflammatory cytokines was detected in each sample by microplate reader at 450 nm.

Cells proteins (1×10⁶ cells/ml) were extracted according to instructions of protein extraction kit (cat no. P0013; Beyotime Institute of Biotechnology, Haimen, China) and protein concentration was determined with a bicinchoninic acid protein assay kit (cat no. PA115; TianGen Biotech Co., Ltd., Beijing, China). Proteins (35 µg per sample) were separated in 10% SDS-PAGE and electrotransferred to nitrocellulose membranes. The membrane was incubated at 4°C overnight with rabbit monoclonal antibodies for β-actin (1:5,000 dilution), SEMA7A (1:1,000 dilution), VEGF (1:1,000 dilution), cofilin (1:1,000 dilution), p-cofilin (1:1,000 dilution), NF-κB (1:1,000 dilution) or pNF-κB (1:1,000 dilution). The secondary antibody (horse radish peroxidase-goat anti-rabbit IgG; 1:5,000 in TBST; cat no. A0216; Beyotime Institute of Biotechnology) was incubated at 37°C for 1 h. Detection of target protein uses enhanced chemiluminescence detection system (GE Healthcare Life Sciences, Chalfont, UK).

RPMVECs (2×10⁵ cells/well in a six-well plate) were seeded in growth medium onto sterile glass slides and incubated for 24 h. Following administration, RPMVECs were fixed with 4% paraformaldehyde (pH 7.4) for 10 min, permeabilized with 0.1% Triton X-100 for 10 min, and incubated with 2% BSA for 30 min. Then cells were stained with Alexa488-Phalloidin (Molecular Probes; Thermo Fisher Scientific, Inc.) for 60 min. Nuclei were stained with DAPI. Slides were mounted with ProLong Gold anti-fade reagent and read with confocal microscope.

The monolayer permeability was measured according to previous method (21). RPMVECs (2×10⁵ cells/well) were cultured on collagen coated Transwell insert which were placed into 24-well plates containing 500 µl medium. 100 µl fluorescein isothiocyanate (FITC)-dextran (2,000-kDa, Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) was added into the insert following monolayer formation. At 1 h, the insert was removed and 100 µl medium was obtained from the bottom chamber. The fluorescent density was measured at a wavelength of 520 nm using a fluorospectrophotometer (LS-50B; PerkinElmer, Waltham, MA, USA) and GraphPad Prism version 5.0 software (GraphPad software, Inc. La Jolla, CA, USA).

All data are expressed as mean ± standard deviation and statistical analysis were performed with one-way analysis of variance followed by Dunnett's test. P<0.05 was considered to indicate a statistically significant difference.

VEGF is vascular permeability- and cell transmigration-related protein. To assess the role of SEMA7A on endothelial permeability, the authors explored the effect of SEMA7A in seawater-stimulated VEGF expression in RPMVECs using SEMA7A siRNA. The expression of VEGF increased significantly in seawater group. However, in SEMA7A group, the expression of VEGF was significantly inhibited in seawater stimulated RPMVECs (Fig. 1).

To investigate the role of receptor plexin C1 on cytoskeletal remodeling in RPMVECs, the authors stained F-actin on cells (Fig. 2A). Seawater or rSEMA7A exposure both induced evident motile phenotype characterized by contraction of stress fibers and enlargement of gap between cells. However, cells were pretreated with the plexin C1 antibody led to a significant reduction in cytoskeletal remodeling.

To assess the role of plexin C1 on monolayer RPMVECs permeability, the FITC-dextran flux across the monolayer was detected (Fig. 2B). Seawater stimulation and rSEMA7A both resulted in FITC-dextran flux increase (P<0.001). Blockage of plexin C1 inhibited the increase of FITC-dextran flux (P<0.05).

In addition, the phosphorylation of cofilin was assessed (Fig. 2C). Seawater stimulation or rSEMA7A both led to the phosphorylation of cofilin (P<0.001). Blockage of plexin C1 inhibited the increase of p-cofilin (P<0.001).

On the basis of increasing expression of endothelial SEMA7A stimulated by seawater, and given that seawater stimulation resulted in robust inflammatory responses, the authors investigated the role of endothelial SEMA7A on the expression of proinflammatory cytokines. In inflammation, the vast majority of proinflammatory cytokines are released by inflammatory cells. Therefore, RPMVECs were co-cultured with rat alveolar macrophage cell line-NR8383 cells. In this co-culture system, the RPMVECs were fixed with paraformaldehyde to inhibit cytokine production and preserve cell surface molecules. As presented in the Fig. 3, expression of TNF-α and IL-1β were significantly increased in seawater group (P<0.001). Administration of SEMA7A siRNA significantly reduced TNF-α and IL-1β produced by NR8383 cells (P<0.001).

Endothelial SEMA7A promotes proinflammatory cytokine production in seawater stimulated NR8383 cells via interaction with β1 integrin and activation of NF-κB pathway. To investigate the role of NF-κB pathway and β1 integrin on the proinflammatory cytokines production by NR8383 cells, the β1 integrin antibody was used in this co-culture cell system. NF-κB activity in NR8383 cells was measured by western blotting. In the co-culture system, seawater induced phosphorylation of NF-κB (Fig. 4A) and the expression of TNF-α and IL-1β (Fig. 4B) in NR8383 cells. Blockage of β1 integrin attenuated the increase of pNF-κB and the expression of TNF-α and IL-1β following seawater stimulation. In addition, as a control, the rSEMA7A was directly administrated to the cultured NR8383 cells. As presented in Fig. 4, the rSEMA7A also significantly led to the phosphorylation of NF-κB and the expression of TNF-α and IL-1β in NR8383 cells. Blockage of β1 integrin attenuated the increase of pNF-κB and the expression of TNF-α and IL-1β following rSEMA7A administration.