Endothelial cell-specific loss of eNOS differentially affects endothelial function

Abstract

The endothelium maintains and regulates vascular homeostasis mainly by balancing interplay between vasorelaxation and vasoconstriction via regulating Nitric Oxide (NO) availability. Endothelial nitric oxide synthase (eNOS) is one of three NOS isoforms that catalyses the synthesis of NO to regulate endothelial function. However, eNOS's role in the regulation of endothelial function, such as cell proliferation and migration remain unclear. To gain a better understanding, we genetically knocked down eNOS in cultured endothelial cells using sieNOS and evaluated cell proliferation, migration and also tube forming potential in vitro. To our surprise, loss of eNOS significantly induced endothelial cell proliferation, which was associated with significant downregulation of both cell cycle inhibitor p21 and cell proliferation antigen Ki-67. Knockdown of eNOS induced cell migration but inhibited formation of tube-like structures in vitro. Mechanistically, loss of eNOS was associated with activation of MAPK/ERK and inhibition of PI3-K/AKT signaling pathway. On the contrary, pharmacologic inhibition of eNOS by inhibitors L-NAME or L-NMMA, inhibited cell proliferation. Genetic and pharmacologic inhibition of eNOS, both promoted endothelial cell migration but inhibited tube-forming potential. Our findings confirm that eNOS regulate endothelial function by inversely controlling endothelial cell proliferation and migration, and by directly regulating its tube-forming potential. Differential results obtained following pharmacologic versus genetic inhibition of eNOS indicates a more complex mechanism behind eNOS regulation and activity in endothelial cells, warranting further investigation.

Fig 1. Genetic knockdown of eNOS promotes proliferation and migration but inhibits formation of tube-like structures in vitro in HUVECs.

(A) HUVECs were transfected with sieNOS or scrambled control for 24, 48 and 72 hrs and the knockdown effect was confirmed by qPCR (B) western blot. (C) Endothelial cell proliferation was evaluated 24, 48 and 72 hrs post-transfection using proliferation kit and (D) by counting the cells using Cyto Smart cell counter. Each triplicate was counted twice, and average of all triplicates was calculated for each biological replicate. (E) Later, RNA was extracted 24 hrs post-transfection and qPCR for p21 and (F) protein was extracted 24, 48 and 72 hrs post-transfection and immunoblot was performed for ki67, p21 and GAPDH. (G) ENOS-knockdown and control HUVECs were seeded on Matrigel, and pictures were taken 6 hrs post-seeding (tubes are marked by arrows) and (H) then quantification for mesh area and (I) tube-length were performed. (J) Scratch assay was performed in sieNOS and scrambled control-transfected HUVECs and pictures were taken at 0, 8 and 20 hrs and (K) the cell migration was quantified. *p<0.05, **p<0.01, ***p<0.001 vs. Scr Control. N = 3–5 in triplicates.

Fig 2. Genetic knockdown of eNOS promotes proliferation and migration but inhibits formation of tube-like structures in vitro in HPAECs.

Human pulmonary artery endothelial cells (HPAECs) were transfected with sieNOS for 24, 48 and 72 hrs in HPAECs and (A) the knockdown effect was confirmed by qPCR. (B) Endothelial cell proliferation was evaluated 24, 48 and 72 hrs post-transfection using using Cyto Smart cell counter. Each triplicate was counted twice, and average of all triplicates was calculated for each biological replicate. (C) Protein was extracted 48 hrs post-transfection and immunoblot was performed for p21 and β-tubulin. (D) RNA was extracted 24 hrs post-transfection and qPCR for p21 was performed. (E) Scratch assay was performed in sieNOS and scrambled control-transfected HPAECs and pictures were taken at 0, 8 and 20 hrs (F) and the cell migration was quantified. *p<0.05, **p<0.01, ***p<0.001 vs. Scr Control. N = 3–5 in triplicates.

Fig 3. Genetic knockdown of endothelial cell-specific eNOS inhibits AKT activation.

(A) HUVECs were transfected with sieNOS or scrambled control for 24 and 48 hrs. Total proteins were extracted 24 and 48 hrs post-transfection immunoblotting for p-AKT, AKT, p-ERK, ERK was performed. Immunoblots for (B, C) p-AKT, AKT, and (D, E) pERK1/2 and ERK1/2 were performed and quantified. *p<0.05, **p<0.01 vs. Scr Control. N = 3–5 in triplicates.

Fig 4. Pharmacologic inhibition of eNOS-activity by L-NAME inhibits endothelial cell proliferation and formation of tube-like structures in vitro, but promotes endothelial cell migration.

HUVECs were seeded until they reached 60–70% confluency and were treated with different dose of L-NAME (10, 50 and 100uM). (A) Cells were collected and counted at each dose using Cyto Smart cell counter. Each triplicate was counted twice, and average of all triplicates was calculated for each biological replicate. (B) HUVECs were seeded on Matrigel, and pictures were taken 6 hrs post-seeding in the presence of L-NAME or vehicle (tubes are marked by arrows) and (C) then quantification for mesh area and (D) tube-length were performed. (E) HUVECs were seeded until they reached 70–80% confluency and were treated with L-NAME or vehicle. A scratch was made and pictures were taken at 0, 8 and 20 hrs and (F) the cell migration was quantified. *p<0.05, ***p<0.001 vs. vehicle. #p<0.05, ###p<0.001 vs. 10μM L-NAME. N = 3–5 in triplicate.

Fig 5. Pharmacologic inhibition of eNOS-activity by L-NMMA inhibits endothelial cell proliferation but promotes endothelial cell migration.

(A) HUVECs were seeded until they reached 60% confluency and were treated with L-NMMA (50uM) or vehicle and 24 hrs later cells were collected and counted using Cyto Smart cell counter. Each triplicate was counted twice, and average of all triplicates was calculated for each biological replicate. (B) Cell migration assay was performed at 0, 8 and 20 hrs. HUVECs were seeded until they reached 70–80% confluency and were treated with L-NMMA or vehicle. A scratch was made and pictures were taken at 0, 8 and 20 hrs and (C) the cell migration was quantified. *p<0.05, ***p<0.001 vs. vehicle. N = 3–5 in triplicate.