Fig 1.
EV treatment of mouse endothelial cells causes increased cell permeability and macrophage migration.
Monolayers of mouse endothelial cells (SVEC4-10) in transwell plates were treated with 40μg/mL EVs derived from non-infected or Mtb-infected RAW 264.7 macrophages. 4kD FITC-labeled dextran (A) or 70kD Rhodamine-labeled dextran (B) was added to top chamber and culture medium was taken from bottom chamber at various time points and the dextran concentration quantified. All the data points were generated from three independent replicates. *p<0.05 when compared to untreated endothelial cells (RC). UnEV: EVs from non-infected Raw264.7 cells, RvEV: EVs from H37Rv-infected cells. (C) Endothelial cell monolayers were stimulated with EVs derived from non-infected or Mtb-infected macrophages for 3 hours or left untreated. CFSE-labeled mouse BMMs were added to the SVEC4-10 cells. The fluorescently-labeled macrophages which migrated through the SVEC4-10 cell monolayer into the bottom of the Transwell filter were imaged 4 hrs after their addition to the SVEC-10 cells (representative images of two independent experiments). (D) The number of BMMs in seven randomly selected fields were counted for each condition and the total number of cells calculated. Shown is the average from two experimental replicates + SD with asterisk (*) indicating a p value < 0.05 when compared to untreated endothelial cells (RC).
Table 1.
Read and alignment information for each sample sequenced.
Roughly 76% of all high quality reads was used for differential expression analysis.
Fig 2.
Analysis of the endothelial cell gene expression profile following treatment with EVs isolated from Mtb-infected and uninfected macrophages.
Total RNA was sequenced for two independent biology replicates. (A) Venn diagram of the genes that showed a minimum two fold up- or down-regulation in endothelial cells following treatment with EVs from Mtb-infected (RvEV) or uninfected (UnEV) macrophages compared to each other and to untreated cells. (B) Hierarchical cluster analysis of differentially expressed genes in endothelial cells treated with EVs isolated from Mtb-infected or uninfected macrophages. The analysis was conducted with a minimal 2-fold change compared with resting endothelial cells.
Table 2.
Shared pathways significantly upregulated in endothelial cells treated with EVs.
Fig 3.
EV-induced gene expression in endothelial cells.
(A) Endothelial cells were treated 4hrs with 40μg/mL EVs derived from either non-infected or Mtb-infected macrophages or left untreated. Total RNA was sequenced for two independent biology replicates. Venn diagram indicating the number of genes whose expression was >2-fold upregulated with a false discover rate (q<0.05) for each indicated comparison. (B) Quantitative RT-PCR was performed for a subset of genes defined as upregulated in the sequence analysis. Endothelial cells were left untreated or treated with EVs derived from uninfected or Mtb-infected macrophages. Total RNA was extracted followed by cDNA synthesis. Fold change of gene expression was calculated by comparative Ct method. Dnaja2 whose expression by sequence analysis did not change before and after EV treatment was selected as a negative control. Expression data was normalized to gapdh. RC: Untreated cells, RvEV: Treatment with EVs from H37Rv-infected macrophages, UnEV: Treated with EVs from non-infected macrophages. Graph indicates fold change of gene expression from two independent experiments +/- SD.
Fig 4.
Upregulation of TLR2, VCAM-1 and CCL2 on endothelial cells following exposure to EVs from Mtb-infected macrophages.
Endothelial cells were left untreated or treated for 16 hrs with LPS (1μg/mL) or EVs derived from non-infected or Mtb-infected macrophages. (A) Cells were stained with FITC conjugated anti-mouse TLR2 antibody or FITC conjugated anti-mouse IgG1 antibody as an isotype control. (B) Cells were surface-stained with FITC-labeled rat anti-mouse VCAM1 or FITC labeled anti-rat IgG2a antibody as an isotype control. (C) Cells were permeabilized and stained for intracellular CCL2 using PE-conjugated anti-mouse CCL2 antibody or PE-labeled IgG as an isotype control. Gates were set to approximately 1% for isotype control and were maintained for all subsequent analysis. RC: untreated cells, RvEV: Treatment with EVs from H37Rv-infected macrophages, UnEV: Treated with EVs from non-infected macrophages. Data is representative of the protein expression from three independent experiments.
Fig 5.
EVs derived from the serum of Mtb-infected mice can activate endothelial cells ex vivo.
(A) SVEC4-10 cell monolayers were left untreated or stimulated for 3 hrs with EVs derived from non-infected or Mtb-infected mice. CFSE-labeled mouse BMMs were added to SVEC4-10 cells. The fluorescently-labeled macrophages which migrated through the SVEC4-10 cell monolayer into the bottom of the Transwell filter were imaged. The number of BMMs in seven randomly selected fields were counted and the total number of cells for each condition defined. The data is the average of three independent mouse Mtb infections +SD with (*) indicating a p value < 0.05 compared to RC. (B) Quantitative RT-PCR was performed on endothelial cells that were left untreated or treated for 4 hours with EVs derived from uninfected or Mtb-infected macrophages. Total RNA was extracted followed by cDNA synthesis. Fold change of gene expression was calculated by comparative Ct method. Data is from two independent mouse Mtb infections. (C) Scatter plots of flow cytometry analysis of CCL2 expression. Endothelial cells were left untreated or treated for 16 hours with EVs derived from non-infected or Mtb-infected macrophages. Permeabilized cells were stained with PE-conjugated anti-mouse CCL2 antibody or PE-labeled IgG as an isotype control. Gate was set for isotype control and was maintained for all subsequent analysis. RC: untreated cells. Un-EV: serum-derived EVs from uninfected mice, D7-EV, D14-EV, D21-EV: serum derived EVs from mice infected for 7, 14 and 21 days respectively. Data is representative of the CCL2 expression from two independent experiments.
Fig 6.
Treatment of SVEC4-10 cells with EVs from Mtb-infected macrophages induces NF-κB nuclear localization.
SVEC4-10 cells were seeded in collagen-coated cover slips and incubated for three days to produce a cell monolayer. The cells were left untreated or treated for 4hrs with EVs released from non-infected or Mtb-infected macrophages (40μg/mL). Cells were fixed and stained with rabbit anti-mouse NF-κB antibody. FITC-conjugated goat anti rabbit was used as the secondary antibody. DAPI was used for nuclear staining. Cover slips were mounted on slides in mounting media and observed at 40x using a Nikon c2 confocal fluorescent microscope. (A) Representative images of the different treatment groups from two independent experiments. (B) Quantification of the number of cells with NF-κB nuclear localization. Approximately 100 cells in 4–5 randomly selected fields per coverslip were counted. (*) indicates a p value < 0.05 compared to RC or UnEV treatment. RC: untreated cells, RvEV: EVs from H37Rv-infected macrophages, UnEV: EVs from non-infected macrophages.