MEG3 sponges miRNA-376a and YBX1 to regulate angiogenesis in ovarian cancer endothelial cells

Abstract

Objectives: Recent studies have demonstrated maternally expressed gene 3 (MEG3) as a tumor suppressor across multiple malignancies. Meanwhile, the role of MEG3 in ovarian cancer needs further investigation. We aim to study the effects of MEG3 on angiogenesis in ovarian cancer and the underlying mechanisms.

Methods: The transcript levels of MEG3 in ovarian cancer samples from the GEPIA database were analyzed and compared to those in normal samples. The effect of MEG3 on the tube formation ability was quantified in ovarian carcinoma-derived microvascular endothelial cells (ODMECs). Through sequence analysis, we identified miR-376a as a major candidate to bind to MEG3. A MEG3-miR-376a binding site was identified via genetic modulation methods. RAS p21 protein activator 1 (RASA1) was screened as a middle player to bridge the role of miR-376a and angiogenesis. The regulation between miR-376a and RASA1 was confirmed via a dual-luciferase reporter assay. Finally, the competition was explored between Y-box binding protein 1 (YBX1) and miR-376a in binding to MEG3.

Results: MEG3 was significantly downregulated in ODMECs compared with normal ovarian endothelial cells. Overexpression of MEG3 led to reduced tube formation of ODMECs. The MS2 hairpin assay showed that MEG3 acted as a platform to sponge miR-376a. RASA1, a key suppressor of tube formation, was directly targeted by miR-376a. Further, MEG3 suppressed angiogenesis through the miR-376a/RASA1 axis in ODMECs. Finally, YBX1 and miR-376a were competitively bound to MEG3.

Conclusion: This study uncovered a novel mechanism that MEG3 sponged miRNA-376a and YBX1 to regulate the expression of RASA1 and exert an effect on the angiogenesis of ovarian cancer.

Keywords: Endothelial cells; LncRNAs; MEG3; ODMECs; Ovarian cancer; RASA1; Tube formation; YBX1; miR-376a.

Fig. 1

MEG3 expression is downregulated in ODMECs, and overexpression of MEG3 inhibits tube formation by ODMECs. (A) MEG3 RNA levels in ovarian cancer and normal tissues based on the GEPIA database. (B) Kaplan-Meier analysis on overall survival rates based on MEG3 expression levels in 1435 patients with ovarian cancer. The mean value of MEG3 mRNA levels was applied to stratified patients into two groups. Log-rank test was applied to analyze the differences between the two groups. (C) Expression of MEG3 in ODMECs and normal ovarian microvascular endothelial cells. (D) qRT-PCR detection of lentivirus-mediated expression of MEG3 in ODMECs. (E) Analysis of tube formation by ODMECs infected with Lv-control or Lv-MEG3. Scale bar = 100 μm. (F) Quantified data from E.

Fig. 2

MEG3 sponges miR-376a. (A) Schematic plotting of binding sites for predicted miRNAs on MEG3. (B) Schematic of the MEG3 conjugate recombinant fusion protein that recognizes the MS2 hairpins. (C) Agarose gel electrophoresis indicated that MBP-MCP–conjugated amylose resin could be used to pull down MEG3. (D) Quantified data of MEG3 enrichment by qRT-PCR. (E) Six different miRNAs were detected by qRT-PCR among the precipitates pulled down with MEG3. (F) miR-376a and miR-485-5p expression were dectected by qRT-PCR in Lv-control and Lv-MEG3 letivirus infected ODMECs. (G) Schematic representation of the MEG3-binding sequence in the 3′-UTR of miR-376a. A mutated sequence was generated in the wild-type miR-376a–binding sequence as indicated. (H) miR-376a enrichment levels in ODMECs transfected with MEG3, MEG3-MS2, and MEG3 Mut (miR-376a)-MS2. (I) Steady-state levels of MEG3 in ODMECs transfected with miR-376a mimics. The full, not adjusted blot image for Fig. 2C was included as supplementary materials.

Fig. 3

MEG3/miR-376a loop regulates tube formation. (A) TargetScan prediction genes were collected from the online prediction website. Predicted candidate genes were then analyzed according to the PANTHER classification system. The listed genes are associated with the process of angiogenesis. (B) qRT-PCR analysis depicting the changes in angiogenesis-related genes in ODMECs treated with control mimics and miR-376a mimics. (C) Prediction of miR-376a target sequences in the 3′-UTR of the RASA1 gene and its mutant. (D) Luciferase assay after co-transfection of ODMECs with the wild-type or mutant RASA1 reporter and miR-376a mimics or the control mimics. (E) qRT-PCR analysis of RASA1 expression in ODMECs treated with MEG3 or miR-376a mimics. (F) Tube formation by ODMECs treated with MEG3, si-control, or si-RASA1. Scale bar = 100 μm. (G) Quantified data from F.

Fig. 4

Competitive binding of YBX1 and miR-376a to MEG3. (A) YBX1 co-precipitated with MEG3. The MEG3-MS2-pulldown assay was performed. Western blot analysis of YBX1 present in the MEG3-MS2 precipitates. (B) RIP analysis of YBX1 interaction with MEG3 in ODMECs. Following YBX1 IP or control using IgG IP, MEG3 levels were quantified by qRT-PCR. (C) MEG3 RNA expression in ODMECs transfected with si-control or si-YBX1. (D) RIP analysis and qRT-PCR detection of MEG3 enrichment in ODMECs treated with mimics control and miR-376a mimics. The full, not adjusted blot images for Fig. 4A and Western blot quantification analysis were included as supplementary materials.

Fig. 5

YBX1 and miR-376a competitively bind to MEG3. (A, B) mRNA level (A) and protein level (B) of RASA1 in ODMECs transfected with si-control or si-YBX1. (C, D) RASA1 mRNA expression (C) and protein level (D) in si-YBX1 transfected ODMECs treated with control or miR-376a mimics. The full, not adjusted blot images (for Fig. 5B and D) and Western blot quantification analysis were included as supplementary materials.

Fig. 6

A model depicting the roles of MEG3 was found to inhibit tube formation of ODMECs by sponging miR-376a. RASA1, a key angiogenesis suppressor, was identified as a direct target gene of miR-376a. YBX1 and miR-376a competitively bind to MEG3 and regulate the expression of RASA1.