Fig. 2: Mfn2 at the plasma membrane promotes endothelial barrier integrity.

HLMVECs were infected with doxycycline inducible lenti-Mfn2 shRNA virus for 48 h and treated with doxycycline for 72 h to deplete endogenous Mfn2 (Mfn2-KD). DMSO was used as a vehicle for control ECs. a Barrier integrity of control and Mfn2-KD ECs was examined by VE-cadherin and β-catenin immunostaining using confocal microscopy (Zeiss LSM880, Plan 1.46NA, ×63 magnification). b The levels of VE-cadherin and β-catenin at the cell surface in (a) were determined by analyzing area (upper panel) or fluorescence intensity (lower panel) using ImageJ and represented as % of control. Data are mean values ± SEM for n = 5 biologically independent samples. *p = 0.0184, **p = 0.0024 vs control by paired or unpaired, two-tailed t-test. c EC barrier resistance (Ω) of control and Mfn2-KD ECs was measured with TER. The data are presented without normalizing to show baseline difference. Data are presented with box and whiskers plot and whiskers are Min to Max. n = 5 biologically independent experiments (TER was measured every 8 s for 40 min). ****p = 0.0001 vs control by paired, two-tailed t-test. d Control and Mfn2-KD ECs were grown on transwell plates (0.4 µm pore) and treated with a FITC-albumin tracer (0.5 mg/mL) for 4 h. For rescue experiments, Mfn2-KD ECs were overexpressed with GFP-Mfn2 and the cells (Mfn2-KD + GFP-Mfn2 OE) were examined by permeability assay. The concentrations of FITC-albumin tracer were determined in the lower chamber media at excitation 488 nm and emission 520 nm. The data are presented by normalizing to the baseline. Data are mean values ± SEM for three independent experiments. *p = 0.0298 for control vs Mfn2-KD, *p = 0.0399 for Mfn2-KD vs Mfn2-KD + GFP-Mfn2 OE by unpaired, one-tailed t-test. e Barrier integrity of confluence control, Mfn2-KD ECs, and Mfn2-KD + GFP-Mfn2 OE ECs was examined by VE-cadherin and β-catenin immunostaining using confocal microscopy (Zeiss LSM7700, Plan 1.46NA, ×63 magnification). f The levels of VE-cadherin and β-catenin at the cell surface in (e) were determined by analyzing area (upper panel) or fluorescence intensity (lower panel) using ImageJ and represented as % of control. Data are mean values ± SEM for n = 3–6 biologically independent samples. **p = 0.0053 for area of VE-cadherin in control vs Mfn2-KD, *p = 0.0211 for area of VE-cadherin in Mfn2-KD vs Mfn2-KD + GFP-Mfn2 OE, *p = 0.0256 for area of β-catenin in control vs Mfn2-KD, *p = 0.0354 for area of β-catenin in Mfn2-KD vs Mfn2-KD + GFP-Mfn2 OE by unpaired, two-tailed t-test.