Fig 1.
LDL increases endothelial permeability in an LDLR and cholesterol-dependent way.
(A) HUVECs were plated on top of 0.4 μm pores size transwell inserts and cultured in order to form a confluent and mature monolayer. Cells were either incubated with 100 μg/ml LDL or with the same volume of the control buffer. 24 hours later, cells were washed with serum-free media and the permeability of the monolayer to 70 kDa FITC-dextrans was assessed two hours later, by measurement of the fluorescence at the bottom chamber of the culture system. (B) The same experiment as in (A) with the addition of 2 μg/ml of anti-LDLR or the IgG control at day 7, one hour before the addition of LDL. Fluorescence at the bottom chamber was measured 15 minutes upon the addition of the dextrans. (C) The same experiment as in (B) with the addition of 50 μg/ml of nystatin or the same volume of vehicle at day 8, one hour before the addition of LDL. All the data, except from panel (A), which is a representation of an experiment performed twice with similar results, represent the averages ± standard deviation of at least three independent experiments. Significance values have been calculated using a two-tailed unpaired student t test at the 95% confidence interval (* P<0.05).
Fig 2.
The effect of LDL on endothelial permeability is not caused by disruption of junctions.
(A) Immunofluorescence for the junctional proteins VE-cadherin and (B) ZO-1, two hours after washing from control and LDL conditions. The number of intercellular gaps (absence of ZO-1 staining at cell-to-cell contacts) per field was quantified in 10 different fields of view (A—right panel). (C) TEER assay two hours after washing from control and LDL conditions. (D) Permeability assay to sodium fluorescein (NaF) two hours after washing from control and LDL conditions. All the data represent the average ± standard deviation of at least three independent experiments. Significance values have been calculated using a two-tailed unpaired student’s t test at the 95% confidence interval (* P<0.05).
Fig 3.
LDL favors the transcytosis of high molecular weight dextrans.
The same experiment as in Fig 1A was performed. (A) At the end of the experiment, cells were washed twice with PBS and fixed with PFA. A confocal image showing dextrans inside cells is presented. (B) The number of dextran-containing vesicles per cell, two hours after washing from control and LDL conditions was assessed by widefield fluorescence microscopy. Representative images of each condition are shown and the chart represents the quantification of three independent blind experiments in which 50 cells were analyzed. (C) The same experiment as in Fig 1A with addition of 70 kDA dextrans for 15 minutes, followed by three washes and re-incubation with fresh media both at the top and bottom chambers, 15 minutes later a sample from the bottom chamber was collected and analyzed. (D) The same experiment as in Fig 1A with the addition of either vehicle or 5 μg/ml of Brefeldin A (BFA). Significance values have been calculated using a two-tailed unpaired student t test at the 95% confidence interval (* P<0.05).
Fig 4.
Knock down of Caveolin 1 reverts the effect of LDL on endothelial permeability and 70 kDa dextrans colocalize with cholera toxin subunit B.
(A) HUVECs were transfected with either scramble siRNA (Sc siRNA) or siRNAs for Caveolin1 (CAV1) and Clathrin Heavy Chain (CHC) and the day after were either plated on top of 0.4 μm pores size transwell inserts or on 96 well plates. Cells were then cultured as in Fig 1A. At the end of the experiment cells on transwells were used for the permeability assay to 70 kDa FITC-dextrans (left panel). Cells on 96 well plates were lysed and used to check for protein levels by immunobloting (right panel). (B) HUVECs grown on transwells and cultured as in Fig 1A were incubated at the same time with 70 kDa FITC-dextrans (1 mg/ml) and Alexa594-cholera toxin subunit B (CTxB, 20 μg/ml) for 15 minutes. Transwells were then fixed and analyzed by confocal microscopy (63X magnification) for the presence of colocalization events. The graphs represent the average +/- the standard deviation of either the number of colocalization events (top-left panel), or the number of CTxB-containing vesicles (top-right panel) present in control and LDL-treated cells. Representative images of three independent experiments in which 30 randomly selected cells have been analyzed, per condition, are shown (lower panel). Significance values have been calculated using a two-tailed unpaired student t test at the 95% confidence interval (* P<0.05).