ERG-Associated lncRNA (ERGAL) Promotes the Stability and Integrity of Vascular Endothelial Barrier During Dengue Viral Infection via Interaction With miR-183-5p

Abstract

Dengue virus (DENV) continues to be a major public health problem. DENV infection will cause mild dengue and severe dengue. Severe dengue is clinically manifested as serious complications, including dengue hemorrhagic fever and/or dengue shock syndrome (DHF/DSS), which is mainly characterized by vascular leakage. Currently, the pathogenesis of severe dengue is not elucidated thoroughly, and there are no known therapeutic targets for controlling the disease effectively. This study aimed to further reveal the potential molecular mechanism of severe dengue. In this study, the long non-coding RNA, ERG-associated lncRNA (lncRNA-ERGAL), was activated and significantly up-regulated in DENV-infected vascular endothelial cells. After knockdown of lncRNA-ERGAL, the expression of ERG, VE-cadherin, and claudin-5 was repressed; besides, cell apoptosis was enhanced, and cytoskeletal remodeling was disordered, leading to instability and increased permeability of vascular endothelial barrier during DENV infection. Fluorescence in situ hybridization (FISH) assay showed lncRNA-ERGAL to be mainly expressed in the cytoplasm. Moreover, the expression of miR-183-5p was found to increase during DENV infection and revealed to regulate ERG, junction-associated proteins, and the cytoskeletal structure after overexpression and knockdown. Then, ERGAL was confirmed to interact with miR-183-5p by luciferase reporter assay. Collectively, ERGAL acted as a miRNA sponge that can promote stability and integrity of vascular endothelial barrier during DENV infection via binding to miR-183-5p, thus revealing the potential molecular mechanism of severe dengue and providing a foundation for a promising clinical target in the future.

Keywords: ERGAL; dengue virus; lncRNA; miRNA-microRNA; permeability; severe dengue infection.

Figure 1

The expression of ERGAL was up-regulated after DENV infection. (A) The viral load in HUVECs showed to be in dose-dependent and time-dependent manner. The expression of E gene of dengue virus in HUVECs was detected by RT-qPCR at different infectious MOIs (a MOI from 0.1 to 10) and time intervals (24 hpi, 48 hpi, and 72 hpi) (n = 3) compared with those mock-infected groups. (B) The expression of ERGAL was increased after DENV infection. The expression levels of ERGAL were detected by RT-qPCR in HUVECs infected with DENV(MOI = 10, 24 hpi). Values were means ± SD (n = 6). **P < 0.01 vs. mock-infected control group. (C) ERGAL genome was located on chromosome 21(start site from 38350109 to end site 38364839) predicted at Ensembl website (http://asia.ensembl.org/index.html). (D) The expression pattern of ERGAL was in a time-dependent manner but virus dose-independent. The expression levels of ERGAL were assessed in HUVECs by RT-qPCR at different infectious time intervals (2hpi, 6 hpi, 12 hpi, and 24 hpi with MOI = 10) or different infectious MOIs (MOI = 0.1, 1, 5, and 10 with 24 hpi). Values were means ± SD (n = 4). *P < 0.05, **P < 0.01 vs. uninfected control groups. (E,F) ERGAL had no protein-coding potential. The bioinformatics prediction of coding protein function of lncRNA-ERGAL was analyzed with Coding-Potential Assessment Tool (http://lilab.research.bcm.edu/cpat/index.php) and PhyloCSF output value assessment at UCSC Genome Browser Gateway (http://genome-asia.ucsc.edu/cgi-bin/hgGateway). The experiments were performed independently at least three times with similar results.

Figure 2

LncRNA-ERGAL was correlated with ERG expression. (A) The expression of ERG was up-regulated after DENV infection. The expression levels of ERG were detected in HUVECs by RT-qPCR (MOI = 10, 24 hpi). Values were means ± SD (n = 6). **P < 0.01 vs. mock-infected groups. (B) The expression pattern of ERG was in a time-dependent manner but virus dose-independent. The expression levels of ERG were assessed in HUVECs by RT-qPCR at different infectious time intervals (2 hpi, 6 hpi, 12 hpi, and 24 hpi with MOI = 10) or at different infectious MOIs (MOI = 0.1, 1, 5, and 10 with 24 hpi). Values were means ± SD (n = 4). *P < 0.05, **P < 0.01 vs. uninfected group. (C) ERGAL was strongly correlated with ERG expression. The relationship between ERGAL and ERG was assessed by Pearson correlation analysis. (D) The expression of ER GAL was remarkably inhibited after DENV infection by si-RNA transfection. The efficiency of knocking down ERGAL was tested by RT-qPCR after transfecting siRNA 24 h then infecting with MOI = 10 at 24 hpi. Values were means ± SD (n = 4). (E) The expression of ERG was correspondingly reduced in line with knockdown of ERGAL after infection. The expression of ERG in HUVECs was detected by RT-qPCR after transfecting 24 h then infecting with MOI = 10 at 24 hpi. Values were means ± SD (n = 4). *P < 0.05; **P < 0.01 vs. si-NC groups (negative siRNA control groups) and NC groups (infected control groups). (F) The protein level of ERG was decreased with knockdown of ERGAL. The protein level of ERG in DENV-infected HUVECs was evaluated by western blotting. ***P < 0.001; ****P < 0.0001. The experiments were performed independently at least three times with similar results.

Figure 3

Knockdown of lncRNA-ERGAL impaired the adherens junction and tight junction of HUVECs infected with DENV. (A) The gene expression of VE-cadherin and claudin-5 was reduced after DENV infection (MOI = 10, 24 hpi) with knockdown of lncRNA-ERGAL. Relative mRNA expression of VE-cadherin and Claudin-5 was detected in HUVECs by RT-qPCR with knockdown of lncRNA-ERGAL. Values were means ± SD (n = 3). (B) The protein level of VE-cadherin and claudin-5 was inhibited after infection (MOI = 10, 24 hpi) with knockdown of lncRNA-ERGAL. Relative protein of VE-cadherin and claudin-5 were measured by western blotting. (C,D) The distribution and expression of VE-cadherin and claudin-5 decreased after infection with knockdown of lncRNA-ERGAL. Immunofluorescence analysis of VE-cadherin or claudin-5 in HUVECs. Nuclei were stained blue (DAPI), and VE-cadherin or claudin-5 were stained green or red, respectively. Positive signal represented the area of protein distribution, and the mean fluorescence intensity (Integrated density/Area) represented Semi-quantitative protein expression. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 vs. siNC groups or NC groups. Original magnification: 40x. The experiments were performed independently at least three times with similar results.

Figure 4

Knockdown of lncRNA-ERGAL promoted vascular endothelial cells apoptosis, increased monolayer cell permeability, and obstructed cytoskeleton remodeling. (A) Knockdown of ERGAL enhanced early apoptosis induced by DENV. Apoptotic statues of HUVECs were assessed by flow cytometry. Values were means ± SD (n = 3). (B) The arrangement of DENV-infected HUVECs with the siRNA-mediated knockdown was sparsely and loosely rearranged, and the gaps among cells were observed to become notably expanded. The gaps among cells were shown under optical microscope (40x), and the gap size was drawn in red. (C) The knockdown of ERGAL coincided with increased permeability of HUVECs monolayer to 40 kDa FITC-dextran after DENV infection(MOI = 10, 24 h post-infection). The permeability assays were performed by detected the FITC-dextran permeability coefficient. Values were means ± SD (n = 3). (D) The distribution and expression of F-actin decreased, and cytoskeleton remodeling disordered after infection with knockdown of ERGAL. Immunofluorescence analysis of F-actin cytoskeleton was performed in HUVECs. Nuclei were stained blue (DAPI), and F-actin were stained green. Positive signal represented the area of protein distribution, and the mean fluorescence intensity (Integrated density/Area) represented Semi-quantitative protein expression. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 vs. siNC groups or NC groups. Original magnification: 40x. The experiments were performed independently at least three times with similar results.

Figure 5

miR-183-5p involved in functioning the endothelial barrier via regulating junction associated proteins and cytoskeleton. (A) ERGAL was mainly expressed in the cytoplasm. Localization of ERGAL was detected by FISH in HUVECs. Nuclei were stain blue (DAPI) and ERGAL were stained green (white arrow). Scale bars: 5 μm. (B) ERGAL sequence contained multiple miRNAs binding site. The putative microRNAs binding ERGAL were predicted on RegRNA₂₀ (http://regrna2.mbc.nctu.edu.tw/). (C) The expression of miR-183-5p was significantly increased after DENV infection. The expression levels of miR-183-5p were detected in HUVECs by RT-qPCR (MOI = 10, 24 hpi). Values were means ± SD (n = 6). (D) The protein level of VE-cadherin and claudin-5 was inhibited in HUVECs with miR-183-5p overexpression and DENV infection. The protein level of VE-cadherin and claudin-5 was measured by western blotting. (E) The protein level of ERG was decreased with overexpression of miR-l 83-5p. The protein level of ERG in DENV-infected HUVECs was evaluated by western blotting. (F, G) MiR-183-5p regulated the expression of VE-cadherin and claudin-5, F-actin, and cytoskeletal remodeling after DENV infection. Immunofluorescence analysis of VE-cadherin, claudin-5 in DENV-infected HUVECs transfected with microRNA mimic or inhibitor. Nuclei were stained blue (DAPI), VE-cadherin, and F-actin were stained green, claudin-5 were stained red and F-actin were stained green. Positive signal represented the area of protein distribution, and the mean fluorescence intensity (Integrated density/Area) represented Semi-quantitative protein expression. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Original magnification: 40x.

Figure 6

LncRNA-ERGAL interacted with miR-183-5p to regulate the vascular barrier. (A) ERGAL was predicted to bind to miR-183-5p from 8842 to 8862 sequence site. The panel showed schematic representation of the predicted binding site for miR-l 83-5p in ERGAL. (B) LncRNA could act as miRNA sponge by binding with miR-183-5p. Upper panel presents the alignment or mutation of potential miR-183-5p binding sites in ERGAL transcript. Lower panel showed the interaction between lncRNA-ERGAL and miR-183-5p was proven by luciferase assay. Values were means ± SD (n = 3). ****P < 0.0001.

Figure 7

LncRNA-ERGAL was induced by DENV and involved in regulating vascular endothelial barrier during early infection. Dengue virus can activated ERGAL, which could bind with miR-183-5p to attenuate its inhibition on expression of ERG, VE-cadherin, and claudin-5, besides ERGAL could reduce the early apoptosis and promote cytoskeleton remodeling for promoting the stability and integrity of endothelial barrier against DENV challenge.