Double-stranded RNA attenuates the barrier function of human pulmonary artery endothelial cells

Abstract

Circulating RNA may result from excessive cell damage or acute viral infection and can interact with vascular endothelial cells. Despite the obvious clinical implications associated with the presence of circulating RNA, its pathological effects on endothelial cells and the governing molecular mechanisms are still not fully elucidated. We analyzed the effects of double stranded RNA on primary human pulmonary artery endothelial cells (hPAECs). The effect of natural and synthetic double-stranded RNA (dsRNA) on hPAECs was investigated using trans-endothelial electric resistance, molecule trafficking, calcium (Ca(2+)) homeostasis, gene expression and proliferation studies. Furthermore, the morphology and mechanical changes of the cells caused by synthetic dsRNA was followed by in-situ atomic force microscopy, by vascular-endothelial cadherin and F-actin staining. Our results indicated that exposure of hPAECs to synthetic dsRNA led to functional deficits. This was reflected by morphological and mechanical changes and an increase in the permeability of the endothelial monolayer. hPAECs treated with synthetic dsRNA accumulated in the G1 phase of the cell cycle. Additionally, the proliferation rate of the cells in the presence of synthetic dsRNA was significantly decreased. Furthermore, we found that natural and synthetic dsRNA modulated Ca(2+) signaling in hPAECs by inhibiting the sarco-endoplasmic Ca(2+)-ATPase (SERCA) which is involved in the regulation of the intracellular Ca(2+) homeostasis and thus cell growth. Even upon synthetic dsRNA stimulation silencing of SERCA3 preserved the endothelial monolayer integrity. Our data identify novel mechanisms by which dsRNA can disrupt endothelial barrier function and these may be relevant in inflammatory processes.

Figure 1. Double-stranded RNA decreased transendothelial electric resistance and increased FITC-dextran permeability through hPAECs monolayers.

(A) Both natural double-stranded RNA (squares) and synthetic analogue (Poly I:C, circles) caused a significant, time dependent, linear decrease of the transendothelial electric resistance (TEER) of the human pulmonary artery endothelial cell (hPAEC) monolayer. Λ-DNA served as a control and had no significant effect. The graph summarizes the results of 6 independent measurements for each treatment and the linear fit of the data. The values were normalized to the TEER of hPAECs at the start of the experiment (*p<0.05, ***p<0.001 compared to Vehicle control). (B) Histogram summarizing the effect of the 24 h treatments on the FITC-dextran permeability of hPAEC. Double-stranded RNA significantly increased FITC-dextran traffic. Similar results were obtained after Poly I:C and total RNA treatment, but not with Λ-DNA. Values represent 3 independent experiments, each performed in triplicates and they were normalized to the vehicle control (*p<0.05 compared to Vehicle control).

Figure 2. Actin rearrangement and stiffening of hPAECs under stimulation with Poly I:C.

Actin rearrangement (green) showing changes in the cell shape (A - control, B - 24 h after Poly I:C). Nuclei were counterstained with DAPI (blue). (C) Representative 80×80 µm² amplitude atomic force microscope image of hPAEC culture without (Control) and with 24 h Poly I:C treatment (D). (E) The force measurements (performed on 5 different points in the central region of numbered cells) showed a significant increase in the Young’s moduli of the cells (***p<0.001 compared to Control).

Figure 3. Involvement of PI3 kinase in the Poly I:C induced cell-cell contact disruption and permeability increase.

(A) 24 h LY-294002 (PI3 kinase blocker) treatment along with Poly I:C (third column) reduced the VE-cadherin and ZO-1 signal similarly to 24 h Poly I:C treatment (second column) compared to control (first column). (B) LY-294002 significantly increased the FITC-dextran permeability of hPAEC, comparably to Poly I:C effect. The bar graph summarizes 3 independent experiments each performed in triplicates. (C). 24 h LY-294002 treatment significantly reduced hPAEC proliferation compared to Vehicle control. A similar and additive effect of 24 h Poly I:C treatment has been observed (**p<0.01, ***p<0.001 compared to Vehicle control).

Figure 4. Proliferation inhibition of hPAEC by natural and synthetic dsRNA.

(A) Concentration dependent inhibition of the hPAEC proliferation upon Poly I:C administration. The line is the best fit to the Hill equation with an IC₅₀ of 2.0±0.3 µg/mL. The bar graph summarizes the effect on hPAEC proliferation of Poly I:C, double-stranded RNA and Λ-DNA treatment. The graphs represent 3 independent experiments, each performed in triplicates (**p<0.01, ***p<0.001 compared to Vehicle control). (B) Histogram summarizing the effect of Poly I:C treatment on the cell number distribution in G1, S and G2/M phase of cell cycle. The graph represents 3 independent experiments of flow-cytometric analysis of propidium iodide stained cells.

Figure 5. Prolonged histamine-induced Ca²⁺ plateau in hPAECs after Poly I:C incubation accompanied by SERCA downregulation and phospholamban dephosphorylation.

Representative traces of 100 µM histamine induced intracellular Ca²⁺ rise under Ca²⁺-free (A) and Ca²⁺ (C) conditions showed a prolonged decay of intracellular Ca²⁺ level in hPAECs after 24 h Poly I:C (red line), dsRNA (green line) or total RNA (orange line) treatment. As a control, Λ-DNA had no effect (blue line). Arrow indicates the application of histamine (His). (B) The histamine induced transient Ca²⁺ plateau duration was significantly longer in the case of treatment (*p<0.05, **p<0.01, ***p<0.001 compared to Control, untreated sample) in the absence (B) and presence (D) of 1.8 mM extracellular Ca²⁺. Bar graphs show expression of SERCA3 (E) and SERCA2b (F) isoform of the sarco-endoplasmic reticulum Ca²⁺ ATPase pump. Results are from 3 independent experiments each performed in triplicates (**p<0.01, ***p<0.001 compared to Vehicle control). (G) Phospholamban phosphorylation upon 24 h Poly I:C and dsRNA treatment (p-PLB - phosphorylated phospholamban). (H) Bar graph represents p-PLB/α-tubulin ratio from 3 independent western blot experiments (***p<0.001 compared to Vehicle control).

Figure 6. Inhibition of hPAEC proliferation, increase of FITC-dextran permeability and disruption of intercellular junctions by the SERCA blocker.

(A) The SERCA blocker, BHQ inhibited hPAECs proliferation in a concentration-dependent manner. The bar graph summarizes 3 independent experiments each performed in triplicates. (B) BHQ significantly increased the FITC-dextran permeability of hPAECs. The bar graphs summarize 3 independent experiments each performed in triplicates. (**p<0.01, ***p<0.001 compared to Vehicle control). (C) Confocal microscopic images revealed that BHQ reduced the VE-cadherin signal compared to control, similar to Poly I:C and LY-294002. Nuclei were counterstained with DAPI (blue).

Figure 7. siRNA treatment against SERCA3 protects the hPAECs from Poly I:C induced permeability and junctional changes.

(A) siRNA silencing of SERCA3 abolished the FITC-dextran permeability increase of hPAECs caused by Poly I:C. The bar graphs summarize 4 independent experiments each performed in triplicates. (*p<0.05 compared to siCTL - Vehicle control, ^#p<0.001 compared to siCTL - Poly I:C). (B) mRNA and protein level of SERCA3 upon treatment of siCTL and siSERCA3. (C) Representative confocal microscopic images reveal that siSERCA3 treated hPAECs respond with less VE-cadherin signal loss compared to siCTL treated hPAECs upon 24 h Poly I:C stimulation. Nuclei were counterstained with DAPI (blue).