Circular RNA CircSHKBP1 accelerates the proliferation, invasion, angiogenesis, and stem cell-like properties via modulation of microR-766-5p/high mobility group AT-hook 2 axis in laryngeal squamous cell carcinoma

Abstract

Laryngeal squamous cell carcinoma (LSCC) is a common malignancy in head and neck. Circular SHKBP1 (circSHKBP) exerts momentous functions in the occurrence of many cancers including LSCC. Thus, we investigated the oncogenic capacities of circSHKBP1 in LSCC, and revealed the underlying mechanism as a competing endogenous RNA. The expression levels of circSHKBP1, miR-766-5p, and high mobility group AT-hook 2 (HMGA2) were examined by quantitative real-time PCR and their influences on the overall survival were measured by Kaplan-Meier method. The correlations between circSHKBP1 and miR-766-5p or HMGA2 were detected by Spearman's rank correlation analysis. In vitro, the influences of circSHKBP1/miR-766-5p/HMGA2 axis on the tumorigenesis of LSCC were examined by CCK-8, transwell, sphere formation, and angiogenesis assays, respectively. circSHKBP1 expression was up-regulated in the LSCC specimens and cell lines. And elevated circSHKBP1 expression was closely linked to poor prognosis. Silencing circSHKBP1 expression restrained cell proliferation, invasion, angiogenesis, stem cell-like properties and tumor growth. We observed that miR-766-5p was down-regulated and negatively correlated to circSHKBP1 in LSCC samples. However, HMGA2 was highly expressed and positively associated with circSHKBP1 in these specimens. Importantly, the levels of circSHKBP1, miR-766-5p, and HMGA2 were closely associated with patients' clinical parameters including lymph nodes metastasis and TNM stages. Mechanistic analysis clarified that circSHKBP1 sponged miR-766-5p to regulate HMGA2, the target of miR-766-5p. Moreover, miR-766-5p inhibition and overexpression of HMGA2 rescued the tumor-suppressing roles of circSHKBP1 downregulation in LSCC. In conclusion, circSHKBP1 accelerated the tumorigenesis of LSCC via modulating HMGA2 by targeting miR-766-5p.

Keywords: CircSHKBP1; HMGA2; MiR-766-5p; laryngeal squamous cell carcinoma; tumorigenesis.

Graphical abstract

Figure 1.

CircSHKBP1 was up-regulated in LSCC and involved in patients’ poor prognosis. (a) The circSHKBP1 expression in the 60 LSCC tissues and 60 adjacent normal samples (NC) was examined by qRT-PCR, ***P <0.001; (b) Kaplan–Meier method analysis for the influence of circSHKBP1 on the overall survival of LSCC patients; (c) qRT-PCR method for circSHKBP1 expression in TU686, AMC-HN-8 and 16-HBE cells, ***P <0.001, compared with 16-HBE cells; (d) The structure of circSHKBP1 was shown; (e and f) The stability of circSHKBP1 after exposure to actinomycin D and RNase was tested by qRT-PCR. **P <0.01, compared with GAPDH or Mock group. Experiments were conducted in triplicate. LSCC, laryngeal squamous cell carcinoma.

Figure 2.

Silencing of circSHKBP1 restrained proliferation, invasion, angiogenesis and stem cell-like properties of LSCC cells. (a) Transfection efficacy of circSHKBP1 siRNA in cells was examined by qRT-PCR; (b) Cell viability was measured by CCK-8 assay; (c) The invasive ability of TU686 and AMC-HN-8 cells was measured by transwell assay after transfecting with si-circSHKBP1; (d) The percentage of tube formation of HUVEC cells treated with tumor-conditioned medium from si-circSHKBP1-transfected LSCC cells was shown; (e) Sphere formation assay for stem cell-like properties; (f) Levels of indicated proteins (PCNA, MMP2, VEGFA, and OCT4) were examined by western blotting. *P <0.05, **P <0.01, ***P <0.001, compared with si-NC group. Experiments were conducted in triplicate.

Figure 3.

CircSHKBP1 sponged miR-766-5p. (a) The complementary binding sites of circSHKBP1 and miR-766-5p were predicated by Starbase 3.0; (b) MiR-766-5p expression in 60 LSCC tissues was lower than the adjacent normal (NC) samples, ***P <0.001; (c) The correlation between circSHKBP1 and miR-766-5p in LSCC samples; (d) The association between miR-766-5p and overall survival of LSCC patients was analyzed; (e) miR-766-5p expression in cells was tested by qRT-PCR, ***P <0.001, compared with 16-HBE cells; (f) Relative luciferase activity in AMC-HN-8 cells under different treatments was measured, **P <0.01, compared with miR-NC group; (g) RNA pull-down assay for the direct binding relationship between circSHKBP1 and miR-766-5p in AMC-HN-8 cells, ***P <0.001, compared with Bio-NC group; (h) miR-766-5p expression in si-circSHKBP1-transfected AMC-HN-8 cells; **P <0.01, compared with si-NC group. Experiments were conducted in triplicate. LSCC, laryngeal squamous cell carcinoma.

Figure 4.

HMGA2 was the target of miR-766-5p. (a) The targeting relationship between miR-766-5p and HMGA2 was predicated by Starbase 3.0 and examined by dual-luciferase reporter assay; (b) miR-766-5p decreased HMGA2 protein level in AMC-HN-8 cells, **P <0.01, compared with miR-NC group; (c) Transfection efficacy of miR-766-5p inhibitor was determined in AMC-HN-8 cells, **P <0.01, compared with miR-NC group; (d) HMGA2 protein level in AMC-HN-8 cells transfecting with si-circSHKBP1, or si-circSHKBP1 and miR-766-5p inhibitor, **P <0.01, ***P <0.001; (e) qRT-PCR assay for HMGA2 expression in LSCC specimens, ***P <0.001; (f) High expression of HMGA2 was associated with LSCC patients’ shorter overall survival; (g, h) The association between HMGA2 and miR-766-5p or circSHKBP1 in LSCC patients; (i) HMGA2 protein levels in both LSCC cells, *P <0.05, **P <0.01, compared with 16-HBE cells. Experiments were conducted in triplicate. LSCC, laryngeal squamous cell carcinoma.

Figure 5.

Silencing of circSHKBP1 restrained the tumorigenesis of LSCC by modulating miR-766-5p/HMGA2 axis. (a) Western blotting assay for transfection efficacy of HMGA2 in LSCC cells, *P <0.05, compared with the vector group; (b-f) The influences of si-circSHKBP1/miR-766-5p/HMGA2 on cell viability, invasion, angiogenesis, stem cell-like properties were measured by CCK-8, transwell, angiogenesis, and sphere formation assays, respectively; (g) The protein levels of PCNA, MMP2, VEGFA and OCT4 were determined. *P <0.05, **P <0.01, ***P <0.001. Experiments were conducted in triplicate.

Figure 6.

Silencing of circSHKBP1 suppressed the tumorigenesis of LSCC in vivo. (a-c) The tumor size, body weight and tumor weight were shown in mice after injection of AMC-HN-8 cells transfected with sh-circSHKBP1 or sh-NC; (d) qRT-PCR assay for levels of circSHKBP1, miR-766-5p and HMGA2 in tumor tissues; (e) Immunohistochemistry method was used to examine levels of HMGA2, PCNA, MMP2, VEGFA and OCT4 in mouse tumor tissues. *P <0.05, **P <0.01, compared with the sh-NC group. Experiments were conducted in triplicate.