Fig 2: Transcriptome profiling of ELMO1 knockdown endothelial cells before and after stimulation with a TLR3 agonist poly(I:C).a Experimental design. EA.hy926 human endothelial cells were transfected with ELMO1 siRNA (siELMO1) and incubated for 48 h, followed by stimulation with 1 μg/ml of poly(I:C) for 24 h. Cells before (0 h) and after (24 h) the poly(I:C) stimulation were analyzed by RNAseq. Cells treated with the transfection reagent without siRNA (Mock) and those transfected with non-targeting siRNA (siNT) were used as controls. The image was created with BioRender.com. b Gene ontology (GO) analyses of genes highly expressed in siELMO1 compared to Mock and siNT at 0 hour. c GO analyses of genes less expressed in siELMO1 compared to Mock and siNT at 0 hour. d GO analyses of genes highly expressed in siELMO1 compared to Mock and siNT at 24 h. Genes were subjected to GO analyses using DAVID (false discovery rate (FDR) < 0.05). FDR for each gene set is indicated in the graphs. N = 3 per group. No significant GO terms were detected for the genes expressed at lower levels in siELMO1 cells at 24 h.

Fig 3: Knockdown of ELMO1 increases ECM and inhibits endothelial cell proliferation.a Transcript levels of collagen genes before (0 h) and after (24 h) the poly(I:C) treatment. Data are based on the RNAseq analysis shown in Fig. 1b. Transcript levels of each gene are shown as TPM (transcript per million). Data are mean ± SE (N = 3). **p < 0.01, ***p < 0.001. b Western blot of COL5A1 before (0 h) and 8 h after the stimulation with poly(I:C). Beta-Actin was detected as a loading control. c Transcript levels of selected genes associated with cell cycle before (0 h) and after (24 h) the poly(I:C) treatment. Data are based on the RNAseq analysis shown in Fig. 1c. Transcript levels of each gene are shown as TPM (transcript per million). Data are mean ± SE (N = 3). **p < 0.01, ***p < 0.001, n.s. not significant. d Cell numbers of untreated, Mock, siNT and siELMO1 cells after the transfection of siRNA. 3 × 104 cells were plated on a 6-well plate and siRNA transfection was performed the next day. Cell numbers were determined by hemocytometer at the indicated time point after the transfection. Data are mean ± SE (N = 4). ***p < 0.001 vs. Untransfected and Mock, ††p < 0.01 vs. siNT. e Representative images of the siNT and siELMO1 cells at 48 h in (d) are shown. Bar = 100 μm. f Proliferation of siRNA-transfected cells detected by incorporation of EdU into newly synthesized DNA. Cells were labeled with 5 μM of EdU 44–48 h after the transfection. Data are mean ± SD (N = 4). **p < 0.01, ***p < 0.001. g Cell cycle analysis at 48 h after the transfection of siRNA. The histograms are representative of three independent experiments.

Fig 4: Knockdown of ELMO1 inhibits apoptosis by suppressing caspase-8 activation downstream of TLR3.a Transcript levels of selected genes associated with apoptosis before (0 h) and after (24 h) the poly(I:C) treatment. Transcript levels of each gene are shown as TPM (transcript per million). Data are mean ± SE (N = 3). *p < 0.05, **p < 0.01, ***p < 0.001, n.s. not significant. b Representative images of Mock, siNT and siELMO1 cells 24 h after the poly(I:C) stimulation. Cells were treated with or without 20 ng/ml of caspase-8 inhibitor Z-IETD-FMK throughout the experiment. Bar = 20 μm. c Cell death in (b) was quantified as in Fig. 4d. Data are mean ± SD (N = 3). ***p < 0.001. d Western blot of full-length and cleaved caspase-8, -3 and -7 before (0 h) and after (4 h, 8 h) the stimulation with poly(I:C). Beta-Actin was detected as a loading control. Images are representative of three independent experiments.

Fig 5: Effects of ELMO1 knockdown on inflammatory response.a Transcript levels of selected genes associated with inflammatory responses in Mock, siNT and siELMO1 transfected EA.hy926 cells before (0 h) and after (24 h) the poly(I:C) treatment. Data are based on RNAseq analysis shown in Fig S4. Transcript levels of each gene are shown as TPM. Data are mean ± SE (N = 3). *p < 0.05, **p < 0.01, ***p < 0.001, n.s. not significant. b Effects of three different non-targeting siRNAs (Control siRNA-A, B and C) on the expression of CXCL8 in EA.hy926 cells before (0 h) and after (24 h) the poly(I:C) treatment. Data are mean ± SE (N = 4). ***p < 0.001 vs. Untransfected and Mock, ††p < 0.01 vs. siELMO1. c Effects of ELMO1 knockdown on the protein levels of CXCL8 and IFNβ in the medium detected by ELISA. Cells were transfected with siRNA (siNT or siELMO1) and 48 h later further stimulated with 1 μg/ml of poly(I:C) for 24 h. Data are mean ± SE (N = 4). ***p < 0.001 vs. siNT. Effects of ELMO1 knockdown on the mRNA levels of CXCL8 after the poly(I:C) treatment in the early (d) and late (e) time points. Transcript levels of CXCL8 were determined by qRT-PCR. Data are mean ± SE (N = 3). *p < 0.05, **p < 0.01, ***p < 0.001 vs. siNT at 0 hour; †††p < 0.001 vs. siELMO1 at 0 hour ; ##p < 0.01, ###p < 0.001 vs. siNT at the same time point.