CircESRP1 enhances metastasis and epithelial-mesenchymal transition in endometrial cancer via the miR-874-3p/CPEB4 axis

Abstract

Background: Metastasis is critical for endometrial cancer (EC) progression and prognosis. Accumulating evidence suggests that circular RNAs (circRNAs) can operate as independent functional entities. However, the functional regulatory mechanisms of circRNAs in EC remain unclear.

Methods: The levels of circESRP1, miR-874-3p, and CPEB4 mRNA in EC tissues and cells were determined by qRT-PCR. Sanger sequencing, PCR with divergent primers, an actinomycin D assay, and RNase R treatment were applied to verify the circular properties. Fluorescence in situ hybridization (FISH) and nuclear-cytoplasmic fractionation were used to determine the localization of circESRP1. CCK-8, EdU incorporation, colony formation, Transwell, and wound healing assays were applied to assess the effects of circESRP1 on cell proliferation, migration, and invasion. The mutual regulatory mechanism of ceRNAs was investigated using dual-luciferase reporter, RNA pulldown, RNA immunoprecipitation (RIP), and Western blot assays. The biological effects were further validated in vivo in nude mouse xenograft models.

Results: circESRP1 was highly expressed in EC tissues and cells and was mainly localized in the cytoplasm. Silencing circESRP1 inhibited the proliferation, migration, and invasion of EC cells in vitro and in vivo; however, overexpression of circESRP1 had the opposite effects. Mechanistically, circESRP1 sponged miR-874-3p to upregulate CPEB4 expression and ultimately contribute to EC cell proliferation and metastasis. Furthermore, circESRP1 regulated tumour growth in xenograft models.

Conclusions: CircESRP1 can interact with miR-874-3p to regulate EMT in endometrial cancer via the miR-874-3p/CPEB4 axis. CircESRP1 may serve as a promising therapeutic target for endometrial cancer.

Keywords: CPEB4; CircESRP1; EMT; Endometrial cancer; Proliferation; miR-874-3p.

Fig. 1

Characterization of circESRP1 in EC. a CircRNAs with significant upregulation were selected based on the criteria of fold change > 2 and P value < 0.01 by using a heatmap and volcano plot. In the volcano plot, red dots indicate circRNAs with increased expression in EC, and green dots indicate circRNAs with decreased expression in EC. b The expression of the differentially highly expressed circRNAs was validated in EC and normal tissues by RT-qPCR. c The figure illustrates that circESRP1 is backspliced into a loop by exons 7–9 of ESRP1, and Sanger sequencing confirms the back-spliced junction sites. d The presence of circESRP1 was verified by qRT-PCR in Ishikawa cells and RL952 cells. Different primers amplified circESRP1 in cDNA rather than genomic DNA (gDNA). e The relative RNA levels of Ishikawa cells upon exposure to actinomycin D were analysed by qRT-PCR. f The relative RNA levels of Ishikawa cells and RL952 cells treated with or without RNase R were analysed by qRT-PCR. g Cytoplasmic and nuclear distribution of circESRP1 in Ishikawa cells was analysed by qRT-PCR. 18S and U6 served as positive controls for the cytoplasm and nucleus, respectively. h Identification of circESRP1 location by FISH in Ishikawa cells and RL952 cells. 18S and U6 were used as positive controls in the cytoplasm and nucleus, respectively. i qRT-PCR analysis of circESRP1 in tumour tissues and normal endometrial tissues. Scale bar: 10 μm. n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 2

Knockdown of circESRP1 suppressed EMT and EC cell proliferation, migration, and invasion. a qRT-PCR analysis of circESRP1 expression in Ishikawa cells and RL952 cells after transfection of the expression shRNA. b–d CCK-8, EdU incorporation, and colony formation assay were carried out to assess the proliferation of Ishikawa cells and RL952 cells. e, f Transwell and wound healing assays (magnification, ×40) were performed to assess the migration and invasion of Ishikawa cells and RL952 cells. g EMT-related proteins (E-cadherin and Vimentin) were detected by Western blot analysis in Ishikawa cells and RL952 cells transfected with sh-circESRP1. Scale bar: 100 μm. n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 3

CircESRP1 promoted the proliferation, migration, and invasion abilities of EC cells in vitro. a qRT-PCR analysis of circESRP1 expression in Ishikawa cells and RL952 cells after transfection of the expression vector. b–d CCK-8, EdU incorporation, and colony formation assay were carried out to assess the proliferation of Ishikawa cells and RL952 cells. e, f Transwell and wound healing assays (magnification, ×40) were performed to assess the migration and invasion of Ishikawa cells and RL952 cells. g EMT-related proteins (E-cadherin and Vimentin) were detected by Western blot analysis in Ishikawa cells and RL952 cells transfected with the circESRP1 vector. Scale bar: 100 μm. n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 4

CircESRP1 acts as a sponge for miR-874-3p. a Venn diagram suggested that 3 miRNAs might be targets of circESRP1 sponge action. b, c RNA pulldown was performed in Ishikawa cells and RL952 cells, followed by qRT-PCR to detect the enrichment of circESRP1 and related miRNAs. d Anti-AGO2 RIP was performed in Ishikawa cells and RL952 cells, followed by qRT-PCR to detect the capacity for AGO2 enrichment on circESRP1 and miR-874-3p compared to IgG. e The luciferase reporter assay functionally validated the interaction between circESRP1 and miR-874-3p in in Ishikawa cells and RL952 cells. f Transwell assays demonstrated the effects of the miR-874-3p mimic and inhibitor on Ishikawa cells and RL952 cell migration and invasion. g The relative RNA levels of miR-874-3p in EC tissues and normal tissues were determined by qRT-PCR. h The negative interaction between miR-874-3p and circESRP1. Scale bar: 100 μm, n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 5

MiR-874-3p targets CPEB4 in EC cells. a Venn diagram suggested that 2 genes might be targets of miR-874-3p. b, c The relative RNA and protein levels of CRCP and CPEB4 were determined by qRT-PCR and Western blot analyses, respectively, after transfection with miR-874-3p mimic and inhibitor in Ishikawa cells and RL952 cells. d The relative RNA levels of CPEB4 in EC tissues and normal tissues were determined by qRT-PCR. e The luciferase reporter assay functionally validated the interaction between miR-874-3p and CPEB4 in Ishikawa cells and RL952 cells. f The CPEB4 expression in Ishikawa cells and RL952 cells after transfection of shRNA was determined by qRT-PCR analysis. g The impacts of CPEB4 knockdown on the migration and invasion of Ishikawa cells and RL952 cells were examined by using Transwell assays. h The expression of EMT-related proteins was detected by Western blot analysis. Scale bar: 100 μm, n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 6

The circESRP1/miR-874-3p/CPEB4 axis promoted EC progression by activating EMT. a The Transwell assay indicated that the migration and invasion abilities of Ishikawa cells and RL952 cells transfected with the circESRP1 vector were counteracted when the cells were cotransfected with the miR-874-3p mimic. b Western blot analysis showed that the levels of EMT-related proteins and CPEB4 transfected with the circESRP1 vector were rescued when Ishikawa cells and RL952 cells were cotransfected with the miR-874-3p mimic. c The results of Transwell assays indicated that sh-CPEB4 inhibited the promotive effect of circESRP1 on Ishikawa cells and RL952 cell metastasis. d Western blot analysis showed that the level of EMT-related proteins transfected with circESRP1 vector was rescued when Ishikawa cells and RL952 cells were cotransfected with sh-CPEB4. Scale bar: 100 μm, n = 3, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 7

CircESRP1 enhances tumour growth and metastasis in vivo. a, e Images of subcutaneous injection of BALB/c nude mice. b, f Images of xenograft tumours of each group. c, g Tumour volume and weight measurement in BALB/c nude mice. d, h The relative protein levels of E-cadherin, Vimentin, and CPEB4 were determined in subcutaneous xenograft tumours by IHC. Scale bar, 100 μm. n = 3, *P < 0.05, **P < 0.01, ***P < 0.001