Tumor-Derived EV-Encapsulated miR-181b-5p Induces Angiogenesis to Foster Tumorigenesis and Metastasis of ESCC

Abstract

Pathological angiogenesis is necessary for tumor development and metastasis. Tumor-derived extracellular vesicles (EVs) play an important role in mediating the crosstalk between cancer cells and vascular endothelial cells. To date, whether and how microRNAs (miRNAs) encapsulated in tumor-derived EVs affect angiogenesis in esophageal squamous cell carcinoma (ESCC) remains unclear. Here, we showed that miR-181b-5p, an angiogenesis-promoting miRNA of ESCC, can be transferred from ESCC cells to vascular endothelial cells via EVs. In addition, ESCC-derived EVs-miR-181b-5p dramatically induced angiogenesis by targeting PTEN and PHLPP2, and thereby facilitated tumor growth and metastasis. Moreover, miR-181b-5p was highly expressed in ESCC tissues and serum EVs. High miR-181b-5p expression level in ESCC patients was well predicted for poor overall survival. Our work suggests that intercellular crosstalk between tumor cells and vascular endothelial cells is mediated by tumor-derived EVs. miR-181b-5p-enriched EVs secreted from ESCC cells are involved in angiogenesis that control metastasis of ESCC, providing a potential diagnostic biomarker or drug target for ESCC patients.

Keywords: angiogenesis; esophageal squamous cell carcinoma; extracellular vesicle; metastasis; miR-181b-5p; prognosis.

Figure 1

EVs Secreted from ESCC Cells Contribute to Angiogenesis In Vitro and In Vivo (A) EVs released by Eca-109 and TE-13 cells were observed by electron microscopy. Scale bars, 100 nm. (B) The concentration of EVs was detected by NanoSight particle tracking analysis. (C) Indicated proteins in EVs were detected by western blot. (D) The delivery of PKH26-labeled (red) EVs to Hoechst-labeled HUVECs (blue) was shown by confocal imaging. Yellow arrows represented delivered EVs, and representative images were presented. Scale bar, 100 μm. (E) Migration assays of HUVECs treated with equal quantities of EVs derived from HEEC and ESCC cell lines or blank control. Migrated cells were counted, and representative images were shown. Scale bar, 200 μm. (F) Tube formation assays of HUVECs treated with equal quantities of EVs derived from HEEC and ESCC cell lines or blank control. Branches of tubes were counted, and representative images were shown. Scale bar, 200 μm. (G) Matrigel mixed with ESCC-derived EVs or PBS was subcutaneously injected into nude mice (n = 5). After 14 days, the Matrigel plugs were harvested and performed by hematoxylin and eosin staining and immunohistochemistry for CD31. Representative micrographs were shown. Scale bars, 2 mm (top panels); 100 μm (middle and bottom panels). (H) The number of microvessel density (MVD) significantly increased in the Matrigel plugs containing ESCC-derived EVs. (I) The expression of mouse CD31 was significantly increased in the Matrigel plugs containing ESCC-derived EVs. Experiments were performed at least in triplicate, and results are shown as mean ± SD. Student’s t test was used to analyze the data. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

Figure 2

EVs-miR-181b-5p Is Highly Expressed in ESCC and Mediates Angiogenesis In Vitro (A) Microarray analysis of significantly expressed miRNAs in ESCC tissues and adjacent normal tissue miRNA from seven ESCC patients was presented in a heatmap. (B) Hierarchical clustering analysis of 10 of the most upregulated and 21 of the most downregulated miRNAs (fold change > 2.5 or < −2.5; p < 0.001) was shown. (C) The expression of miR-181b-5p in 58 pairs of ESCC tissues was detected by qRT-PCR. ΔΔCt=–((Ct_miR-181b-5p-Ct_U6)_ESCC-(Ct_miR-181b-5p-Ct_U6)corresponding non-malignant tissue). (D) The expression of miR-181b-5p in 10 ESCC cell lines compared with normal esophageal epithelia cells (HEEC) was detected by quantitative relative real-time PCR. (E) The expression of miR-181b-5p in EVs from the cell lines referred above was detected by quantitative relative real-time PCR. (F) Significant correlation of miR-181b-5p expression in ESCC cells and ESCC cell-derived EVs was calculated with Spearman’s test. (G–O) Migration (G and K), wound healing (H and L), tube formation (I and M), cell cycle (J and N), and cell viability (O) assays of HUVECs treated with miR-181b-5p mimic, inhibitor, or negative control in vitro. Representative images and quantitative analysis were shown. Scale bars, 200 μm. Each experiment was performed three times independently, and data are shown as mean ± SD. Student’s t test was used to analyze the data (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001).

Figure 3

PTEN and PHLPP2 Are Direct Downstream Targets of miR-181b-5p in Mediating Angiogenesis (A) Target gene prediction of miR-181b-5p with two bioinformatics tools, miRDB and TargetScan. (B) Quantitative relative real-time PCR and western blot assays of PTEN and PHLPP2 expression in HUVECs treated with miR-181b-5p mimic, inhibitor, or negative control. (C) The wild-type and a mutated type of binding site between miR-181b-5p and PTEN/PHLPP2. (D) Relative luciferase activity of HUVECs in the presence of indicated treatments. (E) Migration of HUVECs was affected by miR-181b-5p in the presence of PTEN and PHLPP2 or not. Migrated cells were counted, and representative images were shown. Scale bars, 200 μm. (F) Tube formation ability of HUVECs was affected by miR-181b-5p in the presence of PTEN and PHLPP2 or not. Branches of tubes were counted, and representative images were shown. Scale bars, 200 μm. Experiments were performed at least in triplicate, and results are shown as mean ± SD. Student’s t test was used to analyze the data (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001).

Figure 4

EVs-miR-181b-5p Regulates Angiogenesis via the Akt Signaling Pathway (A) The expression of indicated proteins in HUVECs treated with CM from HEEC and ESCC cell lines was detected by western blot. (B) The expression of indicated proteins in HUVECs treated with EVs from HEEC and ESCC cell lines was detected by western blot. (C) The expression of indicated proteins in HUVECs treated with miR-181b-5p mimic, siRNA of PTEN/PHLPP2, or control was detected by western blot. (D) The expression of indicated proteins in HUVECs incubated with Eca-109/miR-NC EVs, Eca-109/miR-181b-5p EVs, Eca-109/miR-181b-5p EVs + Annexin V, Eca-109/miR-181b-5p EVs + inh-181b-5p, Eca-109/miR-181b-5p EVs + PTEN, or Eca-109/miR-181b-5p EVs + PHLPP2 was detected by western blot. (E) The expression of indicated proteins was detected by immunofluorescence. Representative images were shown. Scale bar, 100 μm. Experiments were performed at least in triplicate.

Figure 5

EVs-miR-181b-5p Induces Angiogenesis to Foster ESCC Tumorigenicity and Proliferation In Vivo (A and B) Xenograft assays of KYSE30 cells mixed with EVs derived from Eca-109/miR-181b-5p cells, Eca-109/inh-181b-5p cells, or Eca-109/miR-NC cells were performed on nude mice (n = 5). Representative tumors (A) and tumor growth curves (B) were shown. Scale bar, 5 mm. (C) The expression of miR-181b-5p in tumor tissues was determined by quantitative relative real-time PCR. (D) The expression of PTEN and PHLPP2 in tumor tissues was determined by quantitative relative real-time PCR and western blot. (E) The expression of miR-181b-5p in serum EVs was determined by quantitative relative real-time PCR. (F) The tumor tissues were performed by immunohistochemistry staining for CD31, and microvessel density was analyzed. Representative micrographs were shown. Scale bar, 50 μm. (G) The number of liver and lung metastatic sites (indicated by arrowheads) was determined by H&E staining and counted under the microscope. Scale bars, 5 mm (G, right panels); 50 μm (F and G, left panels). Each experiment was performed in triplicate, and data are presented as mean ± SD. Student’s t test was used to analyze the data (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001).

Figure 6

EVs-miR-181b-5p Induces Angiogenesis to Foster ESCC Metastasis In Vivo (A) The flow diagram for creating the metastatic tumor model. (B) Representative images of liver and lung metastasis of indicated mice treated with EVs derived from Eca-109/miR-181b-5p cells, Eca-109/miR-NC cells, or Eca-109/inh-181b-5p cells were determined by GFP-based fluorescence imaging (n = 5). (C) The number of liver metastatic sites (indicated by arrows) was determined by H&E staining and counted under microscope. Scale bars, 5 mm (left panels); 100 μm (right panels). (D) The number of lung metastatic sites (indicated by arrows) was determined by H&E staining and counted under microscope. Scale bars, 5 mm (left panels); 100 μm (right panels). (E) The livers were performed by immunohistochemistry staining for CD31, and microvessel density was analyzed. Scale bars, 100 μm. (F) The lungs were performed by immunohistochemistry staining for CD31, and microvessel density was analyzed. Scale bars, 100 μm. Each experiment was performed three times independently, and data are presented as mean ± SD. Student’s t test was used to analyze the data (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001).

Figure 7

miR-181b-5p Correlates with Angiogenesis in ESCC Patients (A) EVs in serum from ESCC patients and healthy controls were observed by electron microscopy. Scale bar, 100 nm. (B) The concentration of EVs in serum was detected by NanoSight particle tracking analysis. (C) miR-181b-5p expression in serum EVs from healthy donors and ESCC patients was detected by quantitative relative real-time PCR. Data are presented as mean ± SD. Student’s t test was used to analyze the data (∗∗∗p < 0.001). (D) In situ hybridization (ISH) of miR-181b-5p with ESCC tissue microarrays (TMAs) was performed to detect miR-181b-5p expression in normal and tumor tissues. Representative images were shown. Scale bar, 50 μm. (E) Kaplan-Meier plots of overall survival of 87 patients with ESCC, stratified by expression of miR-181b-5p. Survival data were analyzed by the Kaplan-Meier method and log rank test. (F) ISH of miR-181b-5p in combination with IHC staining of endothelial cell markers (CD31) on serial sections of human ESCC tissues and adjacent normal tissues. Scale bar, 100 μm. (G) The correlation between CD31 and miR-181b-5p in ESCC tissues using Spearman’s test. (H) Proposed schematic diagram of tumor-derived EVs-miR-181b-5p mediating angiogenesis to promote ESCC progression and metastasis.