Cancer-associated fibroblasts-derived exosomal miR-3656 promotes the development and progression of esophageal squamous cell carcinoma via the ACAP2/PI3K-AKT signaling pathway

Abstract

Esophageal squamous cell carcinoma (ESCC) is one of the most common gastrointestinal tumors, accounting for almost half a million deaths per year. Cancer-associated fibroblasts (CAFs) are the major constituent of the tumor microenvironment (TME) and dramatically impact ESCC progression. Recent evidence suggests that exosomes derived from CAFs are able to transmit regulating signals and promote ESCC development. In this study, we compared different the component ratios of miRNAs in exosomes secreted by CAFs in tumors and with those from normal fibroblasts (NFs) in precancerous tissues. The mRNA level of hsa-miR-3656 was significantly upregulated in the former exosomes. Subsequently, by comparing tumor cell development in vitro and in vivo, we found that the proliferation, migration and invasion capabilities of ESCC cells were significantly improved when miR-3656 was present. Further target gene analysis confirmed ACAP2 was a target gene regulated by miR-3656 and exhibited a negative regulatory effect on tumor proliferation. Additionally, the downregulation of ACAP2 triggered by exosomal-derived miR-3656 further promotes the activation of the PI3K/AKT and β-catenin signaling pathways and ultimately improves the growth of ESCC cells both in vitro and in xenograft models. These results may represent a potential therapeutic target for ESCC and provide a new basis for clinical treatment plans.

Keywords: ACAP2; Cancer-associated Fibroblast (CAF); Esophageal Squamous Cell Carcinoma (ESCC); Exosome; MicroRNA miR-3656.

Figure 1

miR-3656 upregulates CDEs to enhance ESCC development. (A) Electron microscopy images (bar=100 nm) and (B) nanoparticle particle size distribution of CDEs and NDEs. (C) Western blot analyses of the exosomal protein markers in CDEs and NDEs from three different patients. Whole cell lysis was used as an input control; GAPDH was used as a loading control. (D) Plots showing the relative growth rates of EC18 and KYSE30 cells treated with exosomes for 48 hr. (E) Representative images (top) and quantification (down) of migratory or (F) invasive EC18 or KYSE30 cells per field after treatment with exosomes for 48 hr. Bars in plots represent mean ± SD. Scale bars in images equal to 50 µm. p > 0.05; *, p ≤ 0.05; **, p ≤ 0.01; and ***, p ≤ 0.001. (G) Plots showing the relative levels of a list of miRNAs in CDEs and NDEs prepared from 5 patients. (H) Distribution map drawn by expression level and false discovery rate. (I) The relative concentration comparison of miR-3656 in NDEs and CDEs as measured by RT-qPCR. (J) Kaplan-Meier analysis showing that individuals with high miR-3656 expression presented shorter DFS in 184 EC patients.

Figure 2

miR-3656 impacts the proliferation and metastasis of ESCC cells in vitro. (A) Plots showing the relative levels measured by RT-qPCR of miR-3656 among a panel of untreated ECSS cells and (B) between EC18 or KYSE30 cells treated with chemically synthesized miR-3656 and negative control miRNA without homology. (C) Plots showing the impacts of miR-3656 on the growth rates of EC18 and KYSE30 cells determined by CCK-8 assay. (D) Representative images and quantification showing the impacts of miR-3656 mimics on the rates of migration and invasion of EC18 and KYSE30 cells determined by Transwell assays (bar=50 µm), (E) colony formation assays and (F) wound healing assays (bar=250 µm). (G-I) Similar promotive roles were authentically repeated in lentivirus-transfected ESCC cells stably overexpressing miR-3656 (bar=250 µm).

Figure 3

miR-3656 but exosomal vehicle or CAFs enhance the rates of ESCC tumor growth in vivo. (A) Plots showing tumor volume within 30 days post cell injection. (B) The appearance of tumor mass in transplanted tumor mice and the tumor tissue after dissection. (C) Plots showing tumor weights after harvesting. (D) Comparison of the expression levels of miR-3656 in fast-growing tumor tissue and normal tumor tissue. (E) The expression of miR-3656 in exosomes derived from the HEK293 cell line overexpressing miR-3656 constructed by gene editing was higher than that in the control group. (F) Representative images showing that exosomes labeled with the exosome-specific marker PKH67 (green) entered and were contained in ESCC cells. (G-H) The growth rate, invasion intensity and migration activity of ESCC cells were further compared, and ESCC cell lines treated with HEK293-derived exosomes overexpressing miR-3656 had an increased proliferation ability.

Figure 4

miR-3656 directly targets ACAP2 and represses its expression. (A) Venn diagram showing the miR-3656 target genes associated with tumor progression. (B) Heat map showing the expression changes of 7 candidate genes. (C) A diagram showing the 3′-UTR fragment of the human ACAP2 gene containing wild-type or mutated miR-3656 in the binding sequence of luciferase reporter vectors. (D) The relative luciferase activities in EC18 and KYSE30 cells expressing wild-type (WT) or mutant miR-3656 binding site (MU) untreated or treated with negative control (Mimic) or miR-3656 mimics (miR3656). (E) Comparison of the relative expression level of ACAP2 mRNA between transient (exogenous) transfection and lentivirus stable (endogenous production) cell lines. (F) The protein levels of ACAP2 in EC18 and KYSE30 cells treated with mimic or miR-3656.

Figure 5

ACAP2 mediates the impacts of miR-3656 on the proliferation and metastasis of ESCC cells. Respectively, reducing ACAP2 expression with siRNA, overexpressing ACAP2 expression by lentiviral vectors pLenti-ACAP2, and repression of ACAP2 expression in overexpressing cell lines by transfecting miR-3656. (A-C) The mRNA level of each group was determined with RT-qPCR and (D-F) the ACAP2 protein level in each group were observed via Western Blot. After the level of ACAP2 was intervened, (G-I) the proliferation ability, (J-L) migration and invasion intensity of ESCC cells showed a negatively correlation with the level of ACAP2.

Figure 6

The miR-3656/ACAP2 axis impacts both PI3K/AKT and classical Wnt signaling pathway cascades in ESCC cells. (A-D) Representative immunoblots and quantitation of relative levels of phosphorylated AKT (p-AKT) and total AKT and β-catenin in EC18 and KYSE30 cells treated with miR-3656, ACAP2-targeted siRNA, ACAP2 overexpression and expression reversion by miR-3656 transfection after overexpression. (E) Schematic diagram of the inference based on existing data.