Cell culture

bEnd.3 cells (ATCC® CRL-2299™) were cultured in Dulbecco’s modified Eagle medium (DMEM, Gibco, catalog number: C11995500BT) supplemented with 10% fetal bovine serum (FBS, Gibco, catalog number: 10099-141) and 1% penicillin/streptomycin (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) in a humidified 5% CO2 incubator at 37 °C and used for at most ten passages.

Irradiation

Microvascular ECs were seeded at 1 × 104 cells/cm2 density and treated with gamma (γ) radiation after reaching 80% confluence using a GSR C1 137 cesium gamma irradiator (Gamma-Service Medical, Bautzner, Germany) at a dose of 10 Gy with a dose rate of 1.88 Gy/min. The irradiated EC cells were cultured in humidified 5% CO2 incubators at 37 °C until further analysis. Non-irradiated control samples were treated similarly (i.e., culture medium, transport to the accelerator, and incubation conditions).

Preparation of plasmids

Mouse Robo4 was amplified by PCR using the specific primers (Invitrogen, Carlsbad, CA, USA) containing Age 1 (CCGG) and EcoR 1 (AATTCAAAAA) restriction sites. The synthesized mRobo4 oligonucleotides: 5′-GCTGACTGTGTCTTCACTGAT CTCGAG ATCAGTGAAGACACAGTCAGC-3′ (shRobo4#1), 5′-GCCAACAACCTATGGCTATAT CTCGAG ATATAGCCATAGGTTGTTGGC-3′ (shRobo4#2) and 5′-GCCACCAACAATGCTGGGCAA CTCGAG TTGCCCAGCATTGTTGGTGGC-3′ (shRobo4#3) with or without the 3′-UTR (TTTTTG) were inserted and cloned into GV248 vector (hU6-MCS-Ubiquitin-EGFP-IRES-puromycin), (Genechem, Shanghai, China) following Genechem standard procedures and confirmed by restriction digestion and DNA sequencing. A plasmid carrying a non-targeting control sequence (TTCTCCGAACGTGT CACGT) was used as a transfection control (shCont.). For overexpression, lentiviral vectors harboring Robo4 overexpression vectors (Robo4-OX) and its control vector (CON) were obtained from Genechem. The recombinant GV 248 vector plasmid (20 μg) containing shRobo4 and Robo4-OX vectors, 15 μg pHelper 1.0 vector plasmid (pGAG-POL), and 10 μg pHelper 2.0 vector plasmid (pCMV-VSVG) were co-transfected into the packaging cell line 293 T cells under the mediation of Lipofectamine. Viral supernatant collected from the HEK293T cells two days after transfection were passed through a 0.45 μm syringe filter (Thermo Fisher) and ultracentrifuged (Beckman).

Stable establishment of Robo4 knockdown and overexpression in microvascular ECs

Microvascular ECs (3 × 105) were plated to attain 60–70% confluence in a 60 mm culture dish and allowed to adhere overnight. The shRobo4, Robo4-OX, and corresponding control vectors were transduced with 1× HitransG P (Genechem, Shanghai, China) in 10% FBS DMEM into the adherent ECs. After 72 h, infected lentiviral cells were selected with 1 ug/ml puromycin (Genechem, Shanghai, China), and GFP positivity and transduction levels were checked using fluorescent microscopic and FACS analysis. Target expression alteration was assessed by quantitative real-time PCR (qRT-PCR) and western blot.

Total RNA isolation and quantitative reverse transcription-polymerase chain reaction (qRT-PCR)

Total RNA was isolated from the adherent microvascular EC monolayer 24 h post-irradiation using TRIzol reagent (Invitrogen/Life Technologies, Carlsbad, CA, USA). After quantification using a NanoDrop 2000c spectrophotometer (NanoDrop Technologies, Rockland, DE, USA) to obtain an OD ratio of 260/280 between 1.8 and 2.0, reverse transcription was performed with 1.0 µg RNA using PrimeScript reverse transcriptase master mix (Takara Biotechnology, Dalian, China). During qRT-PCR, samples were prepared in triplicates in 96-well plates. Each well contained 2.0 μL of cDNA, 200 nM forward primer, 200 nM reverse primer, 1 × SYBR Green I Master mix, and up to 20 μL of DNase/RNase-free distilled water. The 96-well plates were loaded into a Roche LightCycler 480 II real-time PCR System (Roche Life Sciences). The following PCR cycling program was used: 10 min at 95 °C, 40-cycles of 30 s at 95 °C, 30 s at 60 °C and 15 s at 72 °C; 15 s at 95 °C using LightCycler® 480 software 1.5.0 SP4 for the analysis. The primers, designed with NCBI Primer-BLAST and Primer3 prepared by Thermo Fisher Scientific and Invitrogen, were used in this study as tabulated in Table 1. Triplicate real-time PCR analyses were performed for each sample, and the resulting threshold cycle (CT) values were averaged. Target mRNA expression was normalized to the averaged expression of the housekeeping genes (GAPDH and beta-actin), yielding the ∆CT value. Each target gene’s relative mRNA expression levels were calculated by 2−∆∆Ct.

Table 1 Real-time quantitative PCR primer sequencesAntibodies

Primary antibodies against ZO-1 (Cat. No. 21773-I-AP), CD31 (Cat. No. 28083-I-AP), ROBO4 (Cat. No. 20221-I-AP), connexin 43 (Cat. No. 26980-I-AP), and beta-actin (Cat. No. 20536-I-AP) were purchased from Proteintech Group, Inc (Rosemont, IL, USA). Rabbit polyclonal claudin 5 (Cat. No. AF5216) and P-VE-cadherin (Cat. No. AF3265) antibodies were purchased from Affinity Biosciences (Cincinnati, OH, USA). In addition, rabbit polyclonal anti-occludin (Cat. GB11149-2) and rabbit polyclonal anti-β-tubulin (Cat. No. 2128S) were purchased from Servicebio (Nanjing, China) and Cell Signaling Technology (Danvers, MA, USA), respectively. Alexa Fluor® 594 anti-rabbit secondary antibody (Cat. no. ab15008), goat pAb to IgG (FITC) (Cat. no. ab97050), and rabbit monoclonal anti-VE-cadherin (Cat. No. ab205336) were also purchased from Abcam (Cambridge, MA, USA). The anti-rabbit IgG HRP-linked antibody (Cat. 7074) was purchased from Cell Signaling Technology (Danvers, MA, USA).

Protein isolation and immunoblot analysis

To extract proteins, 200 μL of ice-cold RIPA lysis buffer (Beyotime, China), comprising 150 mM NaCl, 50 mM Tris–HCl pH 7.4, 1% NP-40/IGEPAL CA-630, 0.5% sodium deoxycholate, 0.1% SDS, phosphatase and protease (PMSF) inhibitors (Solarbio®, China), was added to the adherent monolayer ECs (106 cells) 24 h post-radiation and incubated for 5 min on ice. Next, adherent cells were gently scraped and transferred into sterile 1.5 mL EP tubes. The cell lysates were collected at 15,000 rcf for 20 min at 4 °C, and the protein concentrations were evaluated using a bicinchoninic acid (BCA) Protein Assay Kit (Beyotime, China). Subsequently, the protein lysates were supplemented with Laemmli buffer (one fourth of the total volume) (Beyotime, China) and heated at 95 °C for 5 min. 20 μg of the protein lysates were separated using 6–12% SDS-PAGE and blotted onto polyvinylidene difluoride (PVDF) membranes (Life Science, Germany) in a tank blot unit (Mini-PROTEAN II, Bio-Rad, Hercules, CA, USA). The blotted membranes were blocked for 1 h at room temperature using either 5% non-fat dry milk (BD, Le Pont de Claix, France) or 5% BSA (Solarbio®, China) dissolved in a tris-buffered saline solution containing 0.1% (v/v) Tween 20 (Sigma-Aldrich Co. LLC, St. Louis, MO, USA). Afterward, membranes were incubated overnight at 4 °C with the appropriate primary antibody. After washing with 1× TBST, the membranes were incubated for 1 h at room temperature with the appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies (CST, Danvers, MA, USA). The HRP-immunoreactive bands were detected with the enhanced chemiluminescence (ECL) detection kit (GE Healthcare Life Sciences, Little Chalfont, UK). Band signal intensities were analyzed using the Fiji software. Protein loading was normalized by β-actin or β-tubulin protein expression.

Immunofluorescence staining

Microvascular ECs (20,000 cells) with or without Robo4 gene manipulation were seeded on 12-mm round coverslips in 24-well plates and incubated for 24 h to attain 60–70% confluence in a humidified 5% CO2 incubator at 37 °C. Twenty-four hours after irradiation treatment, the cells were washed with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde or 100% ice-cold methanol for 10 min at room temperature, permeabilized for 15 min with 0.1% (v/v) Triton X-100. After three washes with PBS, the cells were blocked with 1% BSA, 22.52 mg/mL glycine in PBST (PBS + 0.1% Tween 20) at room temperature for 30 min. Coverslips were incubated with primary antibodies for claudin-5, connexin-43, VE-cadherin, and ZO-1 overnight at 4 °C. After three additional washes with PBST (5 min/wash), cells were incubated with secondary antibodies conjugated with Alexa Fluor® 488 or Alexa Fluor® 594 (1:500 dilution) at room temperature for 2 h in a moist chamber while protected from light. Lastly, the cells were stained with 1 mg/ml of 4′,6-diamidino-2-phenylindole (DAPI) for nuclei labeling after adequate rinsing in PBS. The coverslips were mounted with an anti-fading mounting medium, and the fluorescent signals were detected using a Zeiss confocal fluorescence microscope (ZEISS LSM 880, Munich, Germany). We used Fiji software (Bethesda, MD, USA) to analyze the acquired images. Figures were assembled with Adobe Illustrator. Adjustments of brightness and contrast were performed in the figure preparations. For an accurate comparison, sample images of the same antigen were acquired under constant acquisition settings.

Flow cytometry

Following plasmid transfection and γ-radiation exposure, microvascular ECs (5 × 105 cells) inoculated in a 60-mm sterile dish were incubated at 37 °C, 5% CO2 for 24 h. The resulting cells were washed with PBS and harvested using a 0.25% trypsin/EDTA reagent. The cells were then incubated with R-phycoerythrin (PE)-conjugated rabbit anti-mouse CD31 (16B1; Thermo Fisher Scientific, Waltham, MA, USA) or PE-conjugated rabbit IgG1 (Thermo Fisher Scientific) for 30 min in the dark at 4 °C and then washed with ice-cold PBS containing 2% FBS. The fluorescence intensity of the cells was determined by a flow cytometer (BD LSRFortessa; BD Biosciences, San Jose, CA, USA), and the data obtained were analyzed by FlowJo version 10 software.

Cycloheximide chase assay

The turnover of VE-Cadherin protein was evaluated using cycloheximide inhibition of protein synthesis. Cells were treated with 25 µg ml−1 cycloheximide (Aladdin Biochemical Tech., Shanghai, China) and lysed at 0, 6, and 12 h time points. The expression of VE-cadherin was analyzed by immunoblotting.

Cell counting kit-8 (CCK-8)

Following Robo4 gene manipulation, microvascular ECs (1 × 104/well) were plated in 96-well plates and subjected to 10 Gy gamma radiation. Twenty-four hours post-irradiation, 10 μL cell counting kit-8 (Beyotime, Shanghai, China) was added, and cells in the culture plate were incubated for 4 h. The absorbance was measured at 450 nm with a microplate reader (BioTek, Synergy H1 microplate reader; Winooski, VT, USA).

Endothelial permeability assessment

The endothelial permeability assay was performed in 24-well plates containing 12 Transwell inserts (6.5-mm diameter, 0.4-mm pore size polycarbonate filters, Corning Costar Corporation), according to the manufacturer’s instructions. One hundred thousand microvascular ECs with or without Robo4 gene manipulation were seeded and cultured in a complete growth medium for 72 h before being exposed to gamma radiation. Twenty-four hours post-irradiation, the medium was removed entirely from apical and basal compartments, 500 µL of fresh medium was added to the bottom chamber, and 150 µL of medium containing 10.0 µg/mL 40 kDa FITC-dextran (MW4000, Sigma, Shanghai, China) was loaded into the apical compartment of inserts. Aliquots of 100 μL were removed from the basolateral chamber every 30 min for 2 h after the application of FITC-dextran and transferred into a black 96-well plate to quantify the fluorescence intensity (excitation and emission at 485 and 535 nm, respectively) using a fluorescence spectrophotometer (BioTek, Synergy H1 microplate reader, Winooski, VT, USA). The permeability assay of the endothelial monolayer was repeated three times for each time point for three samples. The following equation calculated the absolute permeability P [cm/s]: P = [C(t) – C(t0)] × V/A × t × C0, where C(t) is the concentration [μg/ml] of FITC-dextran in the samples that were taken from the basal chambers after 30, 60, 90 or 120 min. C(t0) is the FITC-dextran concentration [μg/ml] of the samples at 0 min, and t is the duration of the flux(es). V is the volume [cm3] in the lower compartment; A is the surface of the Transwell® membrane [cm2]. C0 is the initial concentration [μg/ml] of FITC-dextran in the apical chamber.

Endothelial monolayer staining

EC suspensions of 200 μL were added to Transwell inserts at a density of 5 × 105 cells/mL and incubated for 72 h to form monolayer confluence before irradiation. Twenty-four hours later, media from the inserts were carefully removed without disturbing the cell monolayer. Afterward, 100 μL of 0.5% crystal violet staining solution was added to the insert, covered, and kept at room temperature for approximately 20 min. After removing the cell stain, the inserts were rinsed twice with PBS, filling the insert and receiver plate well with 200 μL and 1 ml of washing buffer. Stained cells were brightfield-imaged at 5× objective magnification on a Nikon Eclipse Ti inverted microscope, and data were analyzed with NIS-Elements AR Microscope Imaging Software (Nikon, Japan).

Gap junction-mediated intercellular communication (GJIC) assay

Semi-confluent microvascular ECs were treated with 10 Gy of γ-radiation. After 24 h, the irradiated cells were co-cultured with double stained (2 µM PKH26/3 µM Calcein Blue AM; MK Biotechnology, Shanghai, China) donor ECs in the ratio of 1:50 donor/recipient. After a co-culture duration of 4 h, cells were detached by Accutase, and the single-cell suspension was evaluated by BD LSRFortessa (BD Biosciences, San Jose, CA, USA) and analyzed using FlowJo v10 software. Calcein (blue) positive and PKH26 (orange) negative cells represent the cell population that has established GJIC.

Statistical analysis

Results analysis was performed using GraphPad Prism® 8.0 to conduct statistical analysis and draw graphs. The analyzed results were presented as the mean ± standard deviation (x ± std.). The variations between the two groups were analyzed using the non-parametric Student t-test for variable data and analysis of variance (ANOVA) for comparison between multiple data sets. Confidence level at α = 0.05 (degree of freedom), P < 0.05 indicates a statistically significant difference.