Egr-1 activation by cancer-derived extracellular vesicles promotes endothelial cell migration via ERK1/2 and JNK signaling pathways

Abstract

Various mammalian cells, including cancer cells, shed extracellular vesicles (EVs), also known as exosomes and microvesicles, into surrounding tissues. These EVs play roles in tumor growth and metastasis by promoting angiogenesis. However, the detailed mechanism of how cancer-derived EVs elicit endothelial cell activation remains unknown. Here, we provide evidence that early growth response-1 (Egr-1) activation in endothelial cells is involved in the angiogenic activity of colorectal cancer cell-derived EVs. Both RNA interference-mediated downregulation of Egr-1 and ERK1/2 or JNK inhibitor significantly blocked EV-mediated Egr-1 activation and endothelial cell migration. Furthermore, lipid raft-mediated endocytosis inhibitor effectively blocked endothelial Egr-1 activation and migration induced by cancer-derived EVs. Our results suggest that Egr-1 activation in endothelial cells may be a key mechanism involved in the angiogenic activity of cancer-derived EVs. These findings will improve our understanding regarding the proangiogenic activities of EVs in diverse pathological conditions including cancer, cardiovascular diseases, and neurodegenerative diseases.

Figure 1. In vivo and in vitro angiogenesis induced by SW480-derived EVs.

(A, B) EVs released by SW480 cells were purified from culture supernatants by a combination of differential centrifugation, ultracentrifugation onto sucrose cushions, and iodixanol density gradients. Each fraction of iodixanol density gradients was analyzed by Western blotting to detect CD81 and CD63, marker proteins of EVs (A). The purified EVs in fraction 3 were analyzed by Western blotting to detect non-EV marker proteins (GM130 and cytochrome c). SW480-derived whole cell lysate (WCL; 10 µg) and SW480-derived EVs (EVs; 10 µg) were loaded for Western blotting analysis (B). (C, D) Matrigel in the presence or absence of SW480-derived EVs (20 µg) was injected subcutaneously into C57BL/6 mice. After 7 days, whole-mount staining of Matrigel with anti-CD31 antibody was conducted (n = 5). Representative confocal Z-stack photographs of whole mounts stained for CD31 (green) are shown in panel C. Scale bars represent 100 µm. Fluorescence intensities of CD31 staining of the Z-stack plane of the Matrigel were measured as described in the Methods (D). (E) Migratory activity of SW480-derived EVs was evaluated by a scratch wound-healing assay. Confluent HMEC-1s were scratched and treated with SW480-derived EVs (1 µg/mL); then the number of migrated cells in the denuded zone was evaluated after 12 h (n = 3). (F) Proliferative activity of SW480-derived EVs (1 µg/mL) was evaluated by assessing the mitosis marker PH3. After 24 h, the PH3 and nuclei were stained with anti-PH3 antibody and Hoechst, respectively and evaluated by confocal microscopy. The percentage of PH3-positive cells in HMEC-1s was quantified by counting the cells with co-localized fluorescence signals (n = 3). As a positive control, HMEC-1s were treated with EGM-2 medium supplemented with diverse angiogenic factors such as EGF, FGF2, VEGF, and IGF1. Data are represented as mean ± SD. *, P<0.05; **, P<0.01; ***, P<0.001; n.s., not significant.

Figure 2. Egr-1 activation in endothelial cells by SW480-derived EVs.

(A) HMEC-1s and HUVECs were incubated with SW480-derived EVs (1 µg/mL) or untreated control. mRNA was isolated from untreated control cells or cells treated with EVs for 0, 0.5, 1, 2, and 4 h and analyzed using real time RT-PCR (n = 3). Values represent Egr-1 mRNA/GAPDH mRNA normalized to untreated control cells. (B) HMEC-1s were treated with EVs (1 µg/mL) derived from HCT116 colorectal carcinoma, A549 lung adenocarcinoma, HT1080 fibrosarcoma, PC3 prostate carcinoma, SH-SY5Y neuroblastoma, and BEAS-2B normal bronchial epithelial cells for 0.5 h (n = 3). (C, D) In HMEC-1s, nuclear translocation of Egr-1 protein after stimulation with SW480-derived EVs (1 µg/mL) for 0.5, 1, 2, and 4 h was analyzed under confocal microscopy (n = 3). Nuclei and Egr-1 proteins were stained with Hoechst (blue) and anti-Egr-1 antibody (red), respectively. Co-localized fluorescence signals (purple) indicate the translocation of Egr-1 into the nucleus. Representative photographs are shown in panel C. Scale bars represent 30 µm. The percentage of Egr-1-positive nuclei was determined by measuring the number of cells with nucleus colocalized signals over that of total cells (D). Data are represented as mean ± SD. *P<0.05; **P<0.01; **P<0.001; n.s., not significant.

Figure 3. Inhibition of Egr-1 activation and endothelial cell migration by Egr-1 siRNA.

(A) HMEC-1s were transfected with 50 nM of scrambled siRNA or Egr-1 siRNA-1, Egr-1 siRNA-2, or Egr-1 siRNA-3. mRNAs were isolated from the cells after 48 h and the level of Egr-1 mRNA was analyzed using real time RT-PCR (n = 3). (B) Nuclear translocation of Egr-1 protein after stimulation with SW480-derived EVs (1 µg/mL) for 1 h was analyzed in scrambled siRNA (50 nM) or Egr-1 siRNA-1 (50 nM) transfected HMEC-1s (n = 3). (C, D) Confluent HMEC-1s transfected with scrambled siRNA (50 nM) or Egr-1 siRNA-1 (50 nM) were scratched and treated with SW480-derived EVs (1 µg/mL); then the number of migrated cells in the denuded zone was evaluated after 12 h (n = 3). The number of migrated cells in the denuded zone of each group is shown in panel D. Scale bars represent 100 µm. Data are represented as mean ± SD. *P<0.05; **P<0.01; *** P<0.001; n.s., not significant.

Figure 4. Role of ERK1/2 and JNK signaling pathways in SW480-derived EV-mediated endothelial cell migration.

(A, B) HMEC-1s were pretreated with signaling inhibitors for 1 h and then stimulated for 1 h with SW480-derived EVs (1 µg/mL). Nuclear translocation of Egr-1 protein was analyzed using confocal microscopy (n = 3). Nuclei and Egr-1 proteins were stained with Hoechst (blue) and anti-Egr-1 antibody (red), respectively. Co-localized fluorescence signals (purple) indicate the translocation of Egr-1 into the nucleus. Representative photographs are shown in panel A. The percentage of Egr-1-positive nuclei was determined by measuring the number of cells with nucleus colocalized signals over total cells (B). (C, D) Confluent HMEC-1s were scratched and treated with SW480-derived EVs (1 µg/mL) in the presence or absence of signaling inhibitors; then the number of migrated cells in the denuded zone was evaluated after 12 h (n = 3). Representative photographs of confocal microscopic imaging are shown in panel C and the number of migrated cells in the denuded zone of each group is shown in panel D. ERK1/2 inhibitor, PD98059 (20 µM); p38 MAPK inhibitor, SB203580 (10 µM); JNK inhibitor, SP600125 (20 µM); Akt inhibitor, BML-257 (20 µM). (E, F) C57BL/6 mice were subcutaneously injected with Matrigel containing SW480-derived EVs (20 µg) with PD98059 (20 µM) or SP600125 (20 µM). After 7 days, whole-mount staining of Matrigel with anti-CD31 antibody was conducted (n = 5). Representative confocal Z-stack photographs of whole mounts stained for CD31 (green) are shown in panel E. Fluorescence intensities of CD31 in the Z-stack plane of the Matrigel were measured as described in the Methods (F). Scale bars in panels A, C, and E represent 30, 100, and 100 µm, respectively. Data are represented as mean ± SD. ***P<0.001.

Figure 5. Inhibition of Egr-1 activation and endothelial cell migration by MβCD.

(A) HMEC-1s were treated with DiI-labeled SW480-derived EVs (1 µg/mL) for 1 h in the presence or absence of MβCD (10 mM) (n = 3). SW480-derived EVs and nuclei were stained with DiI (red) and Hoechst (blue) respectively. Representative photographs are shown in panel A. Scale bars represent 10 µm. (B) Confluent HMEC-1s were scratched and treated with SW480-derived EVs (1 µg/mL) in the presence or absence of MβCD (10 mM); then the number of migrated cells in the denuded zone was evaluated after 12 h (n = 3). (C, D) HMEC-1s were pretreated with MβCD (10 mM) for 1 h and then stimulated with SW480-derived EVs (1 µg/mL) for 1 h. Scale bars represent 30 µm. Nuclear translocation of Egr-1 protein was analyzed using confocal microscopy and the percentage of Egr-1-positive nuclei was determined by measuring the number of cells with nucleus co-localized signals over that of total cells (n = 3). Data are represented as mean ± SD. ***P<0.001.