Exosomal miR-9 inhibits angiogenesis by targeting MDK and regulating PDK/AKT pathway in nasopharyngeal carcinoma

Abstract

Background: Exosomes are small vesicles containing a wide range of functional proteins, mRNA and miRNA. Exosomal miRNAs from cancer cells play crucial roles in mediating cell-cell communication and tumor-microenvironment cross talk, specifically in enabling metastasis and promoting angiogenesis. We focused on miR-9 that was identified as a tumor suppressor previously in nasopharyngeal carcinoma (NPC) tumorigenesis.

Methods: Differential centrifugation, transmission electron microscopy and nanoparticle tracking analysis were used to isolate and identify exosomes. Quantitative PCR and western blotting analysis were used to detect miR-9, pri-miR-9, CD63, TSG101, MDK, P70S6K P-Ser424 and PDK1 P-Ser241 expression. Laser confocal microscopy was used to trace exosomal miR-9 secreted by NPC cells into HUVECs. The effect of exosomal miR-9 on cell migration and tube formation of HUVECs in vivo and vitro was assessed by using migration assay, tube formation assay and matrigel plug assay, respectively. Bioinformatics analysis and luciferase reporter assay were utilized to confirm the binding of exosomal miR-9 to the 3'untranslated region (3'-UTR) of MDK, while Phosphorylation Array was performed to identify AKT Pathway in HUVECs treated with exosomal miR-9. Furthermore, Immunohistochemistry (IHC) and in situ hybridization (ISH) was used to detected miR-9, CD31 and MDK expression in human NPC tumor samples.

Results: NPC cells transfected with miR-9-overexpressing lentivirus, released miR-9 in exosomes. Exosomal miR-9 directly suppressed its target gene - MDK in endothelial cells. Mechanistic analyses revealed that exosomal miR-9 from NPC cells inhibited endothelial tube formation and migration by targeting MDK and regulating PDK/AKT signaling pathway. Additionally, the level of MDK was upregulated in NPC tumor samples and was positively correlated with microvessel density. Notably, the level of exosomal miR-9 was positively correlated with overall survival, and MDK overexpression was positively associated with poor prognosis in NPC patients, suggesting the clinical relevance and prognostic value of exosomal miR-9 and MDK.

Conclusions: Taken together, our data identify an extracellular anti-angiogenic role for tumor-derived, exosome-associated miR-9 in NPC tumorigenesis and prompt further investigation into exosome-based therapies for cancer treatment.

Keywords: Angiogenesis; Exosome; MDK; Nasopharyngeal carcinoma; miR-9.

Fig. 1

miR-9 was reduced in exosomes derived from cultured NPC cells and plasma samples. a Representative images of transmission electron microscopy for exosome derived from NP69, 5-8F and CNE1 cells. Scale bars, 50 nm. b CD63 and TSG101 (common exosomal markers) immunoblots of exosomes derived from above three cell lines. c The expression of miR-9 in the exosomes derived from NP69, 5-8F and CNE1 cells. d Examination of exosomal miR-9 level in plasma samples of NPC patients (n = 110) and healthy controls (n = 60). e A high level of plasma exosomal miR-9 predicted good survival of NPC patients. **, P < 0.01

Fig. 2

Transfer of miR-9 derived from NPC cells to HUVECs via exosomes. a The nanoparticle concentration and size distribution of the exosomes derived from 5-8F/con and 5-8F/miR-9 cells. b Transfer of miR-9 derived from NPC cells to HUVECs via exosomes. HUVECs were cultured with PKH67-labeled exosomes derived from the parental cells (5-8F) transfected with Cy3-pre-miR-9. The Cy3-miR-9 signals were detected in the cytoplasm and nucleus of HUVECs (red), and green signals indicated PKH67-labeled exosomes. Cy3-miR-9 signals are colocalized with PKH67 in HUVECs (yellow). Nuclear counterstaining was performed using 4′,6-diamidino-2-phenylindole (DAPI) (blue). The scale bar indicated 10 μm. c The expression of miR-9 in the exosomes derived from 5-8F/con, 5-8F/miR-9, CNE1/con and CNE1/miR-9 cells. d The expression of miR-9 in HUVEC cells which were co-cultured with exosomes derived from 5-8F/con, 5-8F/miR-9, CNE1/con and CNE1/miR-9 cells respectively. e The expression of pri-miR-9 in HUVEC cells which were co-cultured with exosomes derived from 5-8F/con, 5-8F/miR-9, CNE1/con and CNE1/miR-9 cells respectively. **, P < 0.01

Fig. 3

Exogenous miR-9 inhibited recipient HUVEC cell migration and tube formation. Exosomes were extracted from the medium of 5-8F/con or 5-8F/miR-9 cells and co-cultured with HUVECs. Then cell migration was measured by Transwell migration assay as shown in (a). Representative pictures of transwell migration (Left) were taken at 16 h postplating and quantified for migratory cells (Right). Tube formation of HUVECs was examined by in vitro tube formation assay as shown in (b). Representative pictures of tube formation (Left) were taken at 18 h postplating and quantified for tubule length (Right). Tube formation of HUVECs was also examined by in vivo matrigel plug assay. Representative pictures of matrigel plug were taken at 2 weeks postplating as shown in (c). Representative pictures of tube formation were shown in (d) and quantified for tubule length as shown in (e). **, P < 0.01

Fig. 4

Exosomal miR-9 derived from NPC cells targeted MDK in endothelial cells. a Seventeen angiogenesis-related factors were overlapped predicted by Targetscan, miRBase and Pictar. b The mRNA level of MDK in HUVEC cells after miR-9-overexpressing exosomes treatment. c Diagram of MDK 3′UTR-containing reporter constructs. d Luciferase reporter assays in HUVECs, with cotransfection of wt or mt MDK 3′UTR and exosomes derived from 5-8F/con or 5-8F/miR-9 cells as indicated. These experiments were performed in triplicate, and the results were shown as Mean ± SD. e Luciferase reporter assays in HUVECs, with cotransfection of wt or mt MDK 3′UTR and exosomes derived from CNE1/con or CNE1/miR-9 cells as indicated. f The protein levels of MDK in HUVEC cells after treatment with different exosomes as indicated. The intensity of each band was normalized by GAPDH. g The protein levels of MDK in NPC cells after transfection with miR-9 mimic or negative control. h After direct knockdown of MDK in HUVEC cells by siRNA, cell migration was measured and quantified by Transwell migration assay. i After direct knockdown of MDK in HUVEC cells by siRNA, tube formation of HUVECs was examined by in vitro tube formation assay and quantified for tubule length. **, P < 0.01

Fig. 5

Exosomal miR-9 derived from NPC cells inhibited angiogenesis by regulating the MDK-PDK/AKT signaling pathway. a Antibody array analysis of HUVECs co-cultured with exosomes derived from 5-8F/miR-9 or 5-8F/con cells. Blue and red squares indicated more than two-folds of changes in protein phosphorylation. b HUVECs were co-cultured with exosomes derived from 5-8F/miR-9 and 5-8F/con cells, or were transfected with siMDK or control, respectively. Then, the expression of P70S6K and PDK1 were analysed by western blot. c After inhibiting PDK1 in HUVECs with OSU-03012 (AR-12), cell migration was measured and quantified by Transwell migration assay. d After inhibiting PDK1 in HUVECs with OSU-03012 (AR-12), tubule formation of HUVECs was examined by in vitro tube formation assay and quantified for tubule length. **, P < 0.01

Fig. 6

Expression of miR-9 was negatively correlated with microvessel density in NPC. a Representative images of immunohistochemical staining for CD31 with low or high levels of miR-9. Scale bars, 100 μm. b A negative correlation between miR-9 expression and MVD in NPC tissues. c A schematic diagram illustrated how exosomal miR-9 from NPC cells inhibited angiogenesis by targeting MDK-PDK/AKT pathway in NPC. **, P < 0.01

Fig. 7

MDK was overexpressed and positively correlated with MVD in NPC. a Representative images of immunohistochemical staining for MDK in NPC specimens and healthy control. Magnification, × 20. b Examination of MDK expression in the NPC tissues with different clinical stage. c A high expression of MDK was significantly associated with a poor OS. d Representative images of immunohistochemical staining for MVD with low or high levels of MDK in stage 1 and stage 4 NPC tissues. e A positive correlation between MDK expression and MVD in NPC tissues. **, P < 0.01