Exosomal EPHA2 derived from highly metastatic breast cancer cells promotes angiogenesis by activating the AMPK signaling pathway through Ephrin A1-EPHA2 forward signaling

Abstract

Rationale: Angiogenesis is a fundamental process of tumorigenesis, growth, invasion and metastatic spread. Extracellular vesicles, especially exosomes, released by primary tumors promote angiogenesis and cancer progression. However, the mechanism underlying the pro-angiogenic potency of cancer cell-derived exosomes remains poorly understood. Methods: Exosomes were isolated from breast cancer cells with high metastatic potential (HM) and low metastatic potential (LM). The pro-angiogenic effects of these exosomes were evaluated by in vitro tube formation assays, wound healing assays, rat arterial ring budding assays and in vivo Matrigel plug assays. Subsequently, RNA sequencing, shRNA-mediated gene knockdown, overexpression of different EPHA2 mutants, and small-molecule inhibitors were used to analyze the angiogenesis-promoting effect of exosomal EPHA2 and its potential downstream mechanism. Finally, xenograft tumor models were established using tumor cells expressing different levels of EPHA2 to mimic the secretion of exosomes by tumor cells in vivo, and the metastasis of cancer cells were monitored using the IVIS Spectrum imaging system and Computed Tomography. Results: Herein, we demonstrated that exosomes produced by HM breast cancer cells can promote angiogenesis and metastasis. EPHA2 was rich in HM-derived exosomes and conferred the pro-angiogenic effect. Exosomal EPHA2 can be transferred from HM breast cancer cells to endothelial cells. Moreover, it can stimulate the migration and tube-forming abilities of endothelial cells in vitro and promote angiogenesis and tumor metastasis in vivo. Mechanistically, exosomal EPHA2 activates the AMPK signaling via the ligand Ephrin A1-dependent canonical forward signaling pathway. Moreover, inhibition of the AMPK signaling impairs exosomal EPHA2-mediated pro-angiogenic effects. Conclusion: Our findings identify a novel mechanism of exosomal EPHA2-mediated intercellular communication from breast cancer cells to endothelial cells in the tumor microenvironment to provoke angiogenesis and metastasis. Targeting the exosomal EPHA2-AMPK signaling may serve as a potential strategy for breast cancer therapy.

Keywords: EPHA2; angiogenesis; breast cancer; exosomes; high metastatic potential cells.

Figure 1

Exosomes derived from breast cancer cells with high metastatic potential promote angiogenesis. A. Transmission electron microscopy showed the morphology of exosomes derived from high or low-metastatic breast cancer cells. Scale bar: 200 nm. B Nanoparticle tracking analysis (NTA) of exosomes derived from high or low-metastatic breast cancer cells. The results showed that exosomes derived from 4 different breast cancer cells were comparable in size. C Western blot analysis of the isolated exosomes derived from indicated cell lines. Equal amounts of protein (100 μg) was loaded and analyzed. Alix, TSG101 and CD81 are specific markers for exosomes, while Calnexin was used to monitor cytoplasm contamination of isolated exosomes. D PKH-26-labeled exosomes can be endocytosed into the recipient cells. Scale bar: 200 μm. E HM-Exos significantly enhanced the tube-forming ability of endothelial cells compared to LM-Exos and control. Scale bar: 200 μm. F Compared with LM-Exos and control, HM-Exos significantly enhanced the capillary sprouting ability of rat arterial rings. Scale bar: 200 μm. G, H HM-Exos significantly enhanced the migration ability of endothelial cells compared to LM-Exos and control. Scale bar: 200 μm. Data are expressed as mean ± SD, and all experiments were repeated at least three times. *P < 0.05, **P < 0.01, ***P < 0.001 and ns P > 0.05 indicate no statistical significance. High Metastatic, HM; Ligh Metastatic, LM; Exosomes, Exos.

Figure 2

Exosomal EPHA2 promotes endothelial cell angiogenesis. A Expression of EPHA2 in exosomes and cell lysates was analyzed by using Western blotting; β-actin was used as the loading control. B Western blot analysis showed that EPHA2 expression was silenced in high metastatic potential cells and exosomes after infection with lentivirus expressing EPHA2-specific shRNAs; β-actin was used as the loading control. C, D EPHA2-silenced HM-exos failed to promote the tube formation of endothelial cells and the microvascular outgrowth of rat arterial rings. E Matrigel plug assay showed that exosomal EPHA2 promotes angiogenesis in vivo. For the in vivo matrix gel plug assay, Matrigel plugs with or without VEGF were used as positive or negative controls. HE and CD31 staining were carried out on the dissected Matrigel plugs. F, G EPHA2-silenced HM-exos failed to promote the migration of endothelial cells as measured by Transwell and Wound healing assays. H Western blot analysis showed the expression of EPHA2 in HEK-293T cells and exosomes in cells that were transfected with control or EPHA2 expression vectors. β-actin was used as the loading control. I, J Exosomes from HEK-293T cells expressing EPHA2 significantly promote the tube-forming capacity of endothelial cells compared with control exosomes. Data are expressed as mean ± SD, and all experiments were repeated at least three times. *P < 0.05, **P < 0.01, ***P < 0.001 and ns P > 0.05 indicate no statistical significance. Scale bar: 200 μm.

Figure 3

Exosomal EPHA2 promotes endothelial cell angiogenesis via kinase-dependent forward signaling pathway. A Uptake of HM-Exos by endothelial cells can enhance EPHA2 phosphorylation at Tyr588, whereas HM-Exos from EPHA2-silenced cells failed to induce an upregulation of EPHA2 Tyr588 phosphorylation (Left). EPHA2-rich exosomes from HEK-293T cells can induce a dramatic increase in EPHA2 Tyr588 phosphorylation in endothelial cells (Middle). HM breast cancer cells pre-treated with ALW-II-41-27, an inhibitor of EPHA2, can inhibit EPHA2 phosphorylation at Tyr588 when the cells were treated with HM-Exos (Right). B ALW-II-41-27 treated HM-Exos could not promote the tube formation of endothelial cells. C Schematic diagram of the structure of EPHA2 and its mutants. Full-length EPHA2 and its three mutants EPHA2-ΔL, EPHA2-D739N and EPHA2-S897A were cloned into pCDH-mCherry vector (Left). The expression of EPHA2 and its mutants in cell lysates and exosomes was detected by Western blotting. β-actin was used as a loading control (Right). D, E Exosomes carrying EPHA2 and EPHA2-S897A, but not EPHA2-ΔL and EPHA2-D739N, promote the tube-forming ability of endothelial cells and the sprouting capacity of arterial rings. Data are expressed as mean ± SD, and all experiments were repeated at least three times. *P < 0.05, **P < 0.01, ***P < 0.001 and ns P > 0.05 indicate no statistical significance. Scale bar: 200 μm.

Figure 4

Exogenous EPHA2 promotes endothelial cell angiogenesis through a ligand-dependent forward signaling. A Western blot analysis of Ephrin A1 expression in lentivirus-infected HUVEC expressing control or Ephrin A1-specific shRNA (Left). HM-Exos failed to induce an upregulation of Tyr588 in Ephrin A1-KD HUVECs compared to control cells (Middle). Exosomes from EPHA2-expressing HEK-293T cells failed to induce an upregulation of Tyr588 in Ephrin A1-KD HUVECs compared with control cells (Right). B HM-Exos or EPHA2-rich exosomes failed to promote the tube-forming ability of Ephrin A1-KD cells. C Western blot analysis of EPHA2 expression in lentivirus-infected HUVEC expressing control or EPHA2-specific shRNA (Left). HM-Exos induces upregulation of Tyr588 in EPHA2-KD HUVECs compared to control cells (Middle). Exosomes from EPHA2-expressing HEK-293T cells induce an upregulation of Tyr588 in EPHA2-KD HUVECs compared with control cells (Right). D HM-Exos and EPHA2-rich exosomes still promote the tube-forming ability of EPHA2-KD HUVECs. All experiments were repeated at least three times. *P < 0.05, **P < 0.01,***P < 0.001 and ns P > 0.05 indicate no statistical significance. Scale bar: 200 μm.

Figure 5

Exosomal EPHA2 promotes endothelial cell angiogenesis through AMPK signaling pathway. A KEGG enrichment analysis of the RNA-Seq data showed that the AMPK signaling pathway was significantly activated in EPHA2-rich exosomes treated HUVECs compared with control exosome treated cells. B The expression level of p-AMPK was significantly higher in the HM-Exos-treated endothelial cells than in the control and LM-Exos-treated cells, while the phosphorylation levels of Akt, mTOR and ERK were not altered. C Exosomes from EPHA2 stably silenced HM breast cancer cells failed to induce an upregulation of AMPK phosphorylation. D HM-Exos still induced an increase in AMPK phosphorylation in EPHA2-KD HUVECs compared with control cells. E HM-Exos failed to induce an upregulation of p-AMPK in Ephrin A1-KD HUVECs compared with control cells. F Compared with control cells, EPHA2-rich exosomes from HEK-293T cells could induce an increase in AMPK phosphorylation in EPHA2-KD HUVECs, but failed to induce an upregulation of phosphorylated AMPK in Ephrin A-KD HUVECs. G Compound C, an AMPK inhibitor, significantly inhibited AMPK phosphorylation at a concentration of 10 μM; STO609, a CaMKKβ inhibitor, significantly inhibited AMPK phosphorylation at a concentration of 5μM. H Compound C and STO609 eliminated phosphorylation of AMPK in HUVECs after incubation with HM-Exos. I Inhibition of AMPK signaling by Compound C and STO609 reduced tube-forming ability of HUVECs treated with HM-Exos. J Inhibition of AMPK signaling by Compound C and STO609 decreased the ability of microvascular outgrowth in HM-Exos-treated rat arterial rings. All experiments were repeated at least three times. *P < 0.05, **P < 0.01, ***P < 0.001 and ns P > 0.05 indicate no statistical significance. Scale bar: 200 μm.

Figure 6

Exosomal EPHA2 promotes the metastasis of breast cancer cells in vivo. A Schematic diagram of the in vivo experiment. B Representative images showed the lung metastasis signals detected by IVIS Spectrum imaging systems. C The metastatic foci in the lungs of mice were investigated using Computed Tomography (CT) (Left). The number of metastatic foci on the lung surface was significantly reduced in the EPHA2 knockdown group compared with the control group (Middle). H & E staining showed that the number of micro-metastases in lung tissue of mice inoculated with HM breast cancer cells was significantly higher than that in the group inoculated with LM breast cancer cells (Right). D The IHC analysis showed that the microvessel density (MVD) in MDA-231 and MDA-231-shControl tumors was significantly increased compared with the T47D and MDA-231-shEPHA2 tumors. Data are shown as mean ± SD. Statistical analysis was performed by one-way ANOVA. ***P < 0.001 and nsP > 0.05 indicate no statistical significance.

Figure 7

EPHA2 is an indicator of poor prognosis and progression in breast cancer. A Patients with high EPHA2 expression showed a lower survival rate in BRCA in the TCGA database. B TCGA data analysis suggested that EPHA2 expression was positively correlated with angiogenesis-related genes in BRCA. C The expression level of EPHA2 was higher in drug-resistant tissues in BRCA in the TCGA database. D Immunohistochemical staining to examine the expression of EPHA2 in a breast cancer tissue microarray. E Immunohistochemical staining to examine the expression of CD31 in a breast cancer tissue microarray. F The percentage score of EPHA2 in the paracancer group (Control), nonmetastatic group and metastatic group. G The number of CD31-stained microvesicles in the paracancer group (Control), nonmetastatic group and metastatic group. H The number of CD31-stained microvesicles in EPHA2 low expression and high expression tissues. I ELISA assays showed the concentration of exosomal EPHA2 in plasma collected from healthy donors (n=25), breast cancer patients without metastasis (n=25), or breast cancer patients with metastasis (n=25). Data are shown as mean ± SD. Statistical analysis was performed by one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, and ns P > 0.05 indicate no statistical significance.

Figure 8

Schematic model of the proposed mechanisms of this study. Compared with LM breast cancer cells, the exosomes released by HM breast cancer cells are rich in EPHA2. Exosomal EPHA2 is delivered from HM breast cancer cells to endothelial cells and mediates activation of the Ephrin A1-EPHA2 forward signaling, which further activates the downstream AMPK-HIF-1α signaling pathway and promotes angiogenesis.