Colorectal cancer-secreted exosomal circ_001422 plays a role in regulating KDR expression and activating mTOR signaling in endothelial cells by targeting miR-195-5p

Abstract

Background: As non-coding RNAs, exosomal circular RNAs (circRNAs) regulate colorectal cancer (CRC) progression, although the functional mechanisms by which such molecules affect the tumor microenvironment are still elusive. Herein, we aimed to explore the potential clinical significance of a signature of five serum-derived circRNAs in CRC and investigated the mechanisms underlying endothelial cell angiogenesis mediated by CRC-secreted exosomal circ_001422.

Methods: The expression of a signature of five serum-derived circRNAs (circ_0004771, circ_0101802, circ_0082333, circ_0072309, and circ_001422) were measured by RT-qPCR, and their associations with tumor staging and lymph node metastasis were further evaluated in CRC patients. In silico analysis was used to show the relationship between circ_001422, miR-195-5p, and KDR, validated by dual-luciferase reporter and Western blotting assays. CRC cell-derived exosomes were isolated and characterized by scanning electron microscopy and Western blotting. Endothelial cell uptake of PKH26-labeled exosomes was demonstrated using a spectral confocal microscope. In vitro genetic strategies were used to exogenously alter the expression level of circ_001422 and miR-195-5p expression. Cell proliferation assay, transwell migration assay, and capillary tube formation assay were conducted to explore the role of CRC-secreted exosomal circ_001422 in endothelial cell function in vitro.

Results: The expression levels of serum-derived circ_0004771, circ_0101802, circ_0082333, and circ_001422 were significantly higher in CRC and were positively correlated with the lymph node metastasis status. However, circ_0072309 showed a significant down-regulation in CRC than in healthy individuals. Furthermore, a higher expression level of circ_001422 in both cellular and exosomal fractions was found in HCT-116 CRC cells. We found that HCT-116 exosomes considerably enhanced proliferation and migration of endothelial cells through shuttling of circ_001422. We also observed that exosomes derived from HCT-116 cell, but not non-aggressive Caco-2 CRC cells, increased in vitro tubulogenesis of endothelial cells. Importantly, knockdown of circ_001422 impaired the capability of endothelial cells to form the capillary-like tube structures. CRC-secreted circ_001422 acted as an endogenous miR-195-5p sponge to inhibit miR-195-5p activity, which led to increased KDR expression and mTOR signaling activation in endothelial cells. Importantly, ectopic expression of miR-195-5p mimicked the effect of circ_001422 silencing on KDR/mTOR signaling in endothelial cells.

Conclusion: This study attributed a biomarker role for circ_001422 in CRC diagnosis and proposed a novel mechanism whereby circ_001422 up-regulates KDR through sponging miR-195-5p. These interactions may give rise to the activation of mTOR signaling and may be a possible clarification for the pro-angiogenesis effects of CRC-secreted exosomal circ_001422 on endothelial cells.

Keywords: Circ_001422; Colorectal cancer; Endothelial cell angiogenesis; Exosomes; miR-195-5p.

Fig. 1

The expression levels of serum-derived circular RNAs in CRC patients compared to healthy controls. A The serum levels of circ_0004771 (***p value < 0.001), circ_0101802 (***p value < 0.001), circ_0082333 (***p value < 0.001), and circ_001422 (*** p value < 0.001) were significantly up-regulated in CRC patients than in healthy individuals, while the level of circ_0072309 was down-regulated (**p value < 0.01). B To evaluate the diagnostic accuracy of each candidate circRNA, we analyzed the AUC receiver operating characteristic (AUC-ROC) curve with a confidence interval of 95%. ROC curves of circ_0004771 (p value < 0.0001, AUC = 0.84), circ_0101802 (p value < 0.0001, AUC = 0.87), circ_0082333 (p value < 0.0001, AUC = 0.89), circ_001422 (p value < 0.0001, AUC = 0.92), and circ_0072309 (p value < 0.0001, AUC = 0.79) suggest that all of these circRNAs may discriminate CRC patients from healthy individuals. C The expression levels of circ_0004771 (p value < 0.01), circ_0101802 (p value < 0.01), circ_0082333 (p value < 0.05), and circ_001422 (p value < 0.05) were higher in advanced-stage CRC patients (i.e., stage III according to TNM scores). Circ_0072309, on the other hand, did not show a significant association with TNM status (p value: not significant, ns). D The expression levels of circ_0004771 (p value < 0.05), circ_0101802 (p value < 0.05), circ_0082333 (p value < 0.05), and circ_001422 (p value < 0.05) were higher in LNM-positive CRC patients. Circ_0072309, in contrast, did not follow a significant trend in terms of association with TNM status (p value: not significant, ns). The level of statistical significance was set at a p value of 0.05

Fig. 2

Exosomes derived from CRC cells are enriched by circ_001422 and internalized into endothelial cells. A, B Circ_001422 expression levels were measured in Caco2, SW480, HT-29, and HCT-116 cells and their corresponding exosomes. The higher levels of circ_001422 were detected in both cellular (A) and exosomal (B) fractions of HCT-116 cells. Columns, mean of three different experiments; bars, SD. ***p < 0.001. C Scanning electron micrographs of purified exosomes obtained from the culture media of HCT-116 and Caco2 CRC cells showing spherical and membrane-encapsulated particles with diameters of 50 to 150 nm. D Western blot analysis of purified exosomes obtained from HCT-116 and Caco2 CRC cells demonstrating the presence of the exosome-specific markers CD9 and CD81. E Exosomes derived from HCT-116 CRC cells are internalized by endothelial cells. Endothelial cells exhibit a red fluorescence within their cytoplasm, indicating that they are absorbing significant amounts of PKH-labeled exosomes. The nuclear staining was done by using DAPI

Fig. 3

Exosomes derived from aggressive CRC cells enhanced proliferation, migration, and tubularization of endothelial cells through shuttling of circ_001422. A A significant increase in circ_001422 level in HUVECs was observed 24 h following incubation with 100 µg/ml HCT-116 derived exosomes, compared to that of the HUVECs control group. B The knockdown of circ_001422 by using si-circ_001422 resulted in a significant reduction in its expression in HUVECs (24 h after transfection). C The cell proliferation assay revealed that HUVEC proliferation was significantly increased at different time intervals after being treated with 100 µg/ml HCT-116 exosomes, but not Caco-2 exosomes. Additionally, reducing the level of circ_001422 by RNAi strategy led to decreased endothelial cell proliferation. Of note, the incremental effects of HCT-116 exosomes on the proliferation rate of HUVECs were significantly reduced when circ_001422 was knocked down. D The transwell migration assay demonstrated that HCT-116 exosomes (100 µg/ml) had positive impact on the endothelial cell migration, while Caco-2 exosomes did not. CRC-secreted circ_001422 was identified as a critical factor in promoting endothelial cell migration and reducing its levels decreased the migration rate of HUVECs incubated with HCT-116 exosomes. These observations suggest that the exosomal transfer of circ_001422 is responsible for the enhancing effects of HCT-116 exosomes on endothelial cell migration. E Quantitative assessment of migrated cells indicated that exosomes derived from aggressive CRC cells may boost endothelial cell migration by transferring circ_001422. F Representative micrographs showing the effects of exosomes derived from aggressive CRC cells on the capability of endothelial cells to form the capillary-like tube structures. Observations indicated that treatment of exosomes derived from HCT-116 CRC cells induced the formation of tube-like structures by HUVECs in vitro, but when si-circ_001422 was transfected, the ability of endothelial cells to form tubular structures was hindered. Quantitative analyses of the tube length (G) and branch (H) of HUVECs were calculated in three randomly selected fields. Columns, mean of three different experiments; bars, SD. ***p < 0.001

Fig. 4

Circ_001422 regulates KDR expression by sponging miR-195-5p. A The putative binding sites between miR-195-5p and circ_001422. B Graphs showing luciferase reporter activity of circ_001422 luciferase reporter construct in HUVECs co-transfected with miR-195-5p or negative control. C Having silenced circ_001422, relative expression of miR-195-5p was increased in endothelial cells. D Based on the miRTargetLink Human algorithm (https://ccb-web.cs.uni-saarland.de/mirtargetlink), the interaction networks between KDR and putative target miRNAs have been constructed. E The putative binding sites between miR-195-5p and the 3′-UTR of KDR, as predicted by TargetScan 7.2 (http://www.targetscan.org). F Graphs showing luciferase reporter activity of KDR 3′-UTR luciferase reporter construct in HUVECs co-transfected with miR-195-5p or negative control. miR-195-5p decreased the luciferase activity in the cells that were transfected by KDR 3′-UTR luciferase reporter construct, suggesting that KDR is directly targeted by miR-195-5p. Columns, mean of three different experiments; bars, SD. **p < 0.01. G Western blot analysis showed that transfection of miR-195-5p mimic caused a significant reduction of KDR protein level in HUVECs, suggesting that miR-195-5p may regulate KDR in endothelial cells. As a loading control, GAPDH was used. A minimum of three independent experiments were conducted in order to obtain Western blot images

Fig. 5

CRC-secreted circ_001422 regulates KDR expression and mTOR signaling activation by sponging miR-195-5p in endothelial cells. A As time passed, the level of circ_001422 by adding 100 µg/ml HCT-116 exosomes to HUVECs was increased. B Alternatively, miR-195-5p expression was decreased at different time intervals after HTC-116 exosomes (100 µg/ml) were added to HUVECs. C The mean normalized ratio for KDR mRNA levels were measured by RT-qPCR in different conditions, 48 h after incubation. Results demonstrated that KDR mRNA levels were considerably increased in HUVECs incubated with 100 µg/mL HCT-116 exosomes compared to negative control. In contrast, suppression of circ_001422 in HUVECs partially prevented the promoting effect of HCT-116 exosomes on KDR expression. Besides, to confirm that CRC-secreted circ_001422 has a functional competitive role, we transfected HUVECs with miR-195-5p mimics and observed that the inhibitory effect of miR-195-5p on KDR expression was significantly hindered when HUVECs were incubated with HCT-116 exosomes. However, when si-circ_001422 was transfected into exosome-treated HUVECs, the inhibitory effect of miR-195-5p on KDR expression was partially restored. This highlights the importance of exosomal circ_001422 in regulating KDR expression through miRNA competition. Columns, mean of three different experiments; bars, SD. *p < 0.05, **p < 0.01, ***p < 0.001. D Western blot analysis showed up-regulation of the phosphorylated levels of PI3K, AKT, and mTOR in HUVECs, 48 h after treatment with 100 µg/mL HCT-116 exosomes compared with the corresponding control cells. Results indicated that the promoting effects of HCT-116 exosomes on mTOR signaling were partially abolished when the cells were transfected with the si-circ_001422. Interestingly, the introduction of miR-195-5p into HUVECs that were treated with HCT-116 exosomes resulted in significantly elevated levels of p-PI3K, p-AKT, and p-mTOR protein expression compared to cells that were only transfected with miR-195-5p. As a loading control, GAPDH was used. A minimum of three independent experiments were conducted in order to obtain Western blot images