MicroRNA-92b-3p is a prognostic oncomiR that targets TSC1 in clear cell renal cell carcinoma

Abstract

Although several studies have reported that microRNA (miR)-92b-3p is involved in various cellular processes related to carcinogenesis, its physiological role in clear cell renal cell carcinoma (ccRCC) remains unclear. To clarify the role of miR-92b-3p in ccRCC, we compared miR-92b-3p expression levels in ccRCC tissues and adjacent normal renal tissues. Significant upregulation of miR-92b-3p was observed in ccRCC tissues. Overexpression of miR-92b-3p using a miRNA mimic promoted proliferation, migration, and invasion activities of ACHN cells. Functional inhibition of miR-92b-3p by a hairpin miRNA inhibitor suppressed Caki-2 cell growth and invasion activities in vitro. Mechanistically, it was found that miR-92b-3p directly targeted the TSC1 gene, a known upstream regulator of mTOR. Overexpression of miR-92b-3p decreased the protein expression of TSC1 and enhanced the downstream phosphorylation of p70S6 kinase, suggesting that the mTOR signaling pathway was activated by miR-92b-3p in RCC cells. Importantly, a multivariate Cox proportion hazard model, based on TNM staging and high levels of miR-92b-3p, revealed that miR-92b-3p expression (high vs. low hazard ratio, 2.86; 95% confidence interval, 1.20-6.83; P = .018) was a significant prognostic factor for overall survival of ccRCC patients with surgical management. Taken together, miR-92b-3p was found to act as an oncomiR, promoting cell proliferation by downregulating TSC1 in ccRCC.

Keywords: TSC1; ccRCC; miR-92b-3p; oncomiR; proliferation.

Figure 1

MicroRNA (miR)‐92b‐3p is upregulated in clear cell renal cell carcinoma (ccRCC) and inversely correlates with patient overall survival (OS). A, miR‐92b‐3p expression in ccRCC tissues. Expression of miR‐92b‐3p was examined in 20 matched‐pair samples of ccRCC by quantitative real‐time PCR. Data are the relative expression normalized to U6 snRNA. Comparison was undertaken by Mann‐Whitney U test. ***P < .001. B, Probability estimates of OS of patients. ccRCC samples were divided into 2 groups, high or low levels, based on real‐time PCR results. Kaplan‐Meier survival estimates stratified by miRNA level. Statistical analysis was carried out using the Gehan‐Breslow‐Wilcoxon test

Figure 2

MicroRNA (miR)‐92b‐3p mimic significantly upregulated cell growth, migration, and invasion activities in ACHN cells. A, Expression of miR‐92b‐3p in 4 renal cell carcinoma (RCC) cell lines was examined by quantitative real‐time PCR. B, ACHN cells transfected with miR‐92b‐3p mimic or negative control (NC) miRNA mimic for 24 h were reseeded in a 96‐well plate, incubated for the indicated times, and examined by MTS assay. Values are means ± SD of 3 independent experiments. *P < .05, **P < .01 vs. control mimic. C, ACHN cells were transfected with the miR‐92b‐3p mimic or a negative control miRNA mimic for 72 h. Cell migration was measured 18 h after a wound was created by scraping. Representative results of cell motility in the scratch wound‐healing assay are shown. Scale bar = 200 μm. Results are expressed as mean ± SD of 3 independent experiments. *P < .05 vs. control mimic. D, ACHN cells were transfected with the miR‐92b‐3p mimic or a NC miRNA mimic for 72 h. The transfected cell suspension was added to the upper chamber of Matrigel‐coated Transwell membrane inserts, and the lower chamber was filled with the complete medium and then cultured for 48 h. Invasive cells that had penetrated the Matrigel membrane were fixed and stained. Scale bar = 1.2 mm. Values are means ± SD of 3 independent experiments. *P < .05 vs. control mimic

Figure 3

MicroRNA (miR)‐92b‐3p inhibitor downregulated cell growth and invasion activities in Caki‐2 cells. A, Caki‐2 cells were transfected with the miR‐92b‐3p inhibitor or a negative control (NC) miRNA inhibitor for 72 h. Cell migration was measured 18 h after a wound was created by scraping. Scale bar = 200 μm. Results are expressed as mean ± SD of 3 independent experiments. B, Caki‐2 cells transfected with miR‐92b‐3p inhibitor or NC miRNA inhibitor for 24 h were reseeded in a 96‐well plate, incubated for the indicated times, and examined by MTS assay. Values are means ± SD of 4 independent experiments. **P < .01 vs. control inhibitor. C, Caki‐2 cells were transfected with miR‐92b‐3p inhibitor or NC miRNA inhibitor for 72 h. The transfected cell suspension was added to the upper chamber of Matrigel‐coated Transwell membrane inserts, and the lower chamber was filled with the complete medium and then cultured for 48 h. Invasive cells that had penetrated the Matrigel membrane were fixed and stained. Scale bar = 1.2 mm. Values are means ± SD of 3 independent experiments. *P < .05 vs. control mimic

Figure 4

MicroRNA (miR)‐92b‐3p targets TSC1 in renal cell carcinoma (RCC) cells. A, Predicted miR‐92b‐3p‐binding site within the 3′‐UTR of the human TSC1 gene. B, ACHN cells were cotransfected with the luciferase reporter construct containing the predicted miR‐92b‐3p‐binding site within the TSC1 3′‐UTR and miR‐92b‐3p mimic, or a negative control (NC) miRNA mimic. C, Caki‐2 cells were cotransfected with the luciferase reporter construct containing the predicted miR‐92b‐binding site in the TSC1 3′‐UTR and miR‐92b inhibitor, or NC miRNA inhibitor. Relative luciferase activities were calculated as ratios of firefly luminescence / Renilla luminescence. Values presented as mean ± SD of 7 independent experiments (B,C). One‐way ANOVA with Tukey’s post hoc tests. *P < .05, **P < .01, ***P < .001

Figure 5

MicroRNA (miR)‐92b‐3p regulates the mTOR signaling pathway by targeting TSC1, and TSC1 is downregulated in clear cell renal cell carcinoma (ccRCC) specimens compared to adjacent normal tissues. A, Representative images of western blot analysis from 3 independent experiments showing the protein levels of TSC1 and p70S6K(Thr389) and S6K in ACHN cells transfected with miR‐92b‐3p mimic, or a negative control (NC) miRNA mimic, and also in Caki‐2 cells transfected with miR‐92b‐3p inhibitor, or NC miRNA inhibitor. B, Proliferation assay of ACHN cells transfected with a miR‐92b‐3p mimic or NC mimic after 96 h incubation with either everolimus or vehicle control (0.1% DMSO). Values are means ± SD of 3 independent experiments. Comparison was carried out by unpaired Student’s t test. **P < .01. C, Representative images of immunohistochemical staining for TSC1 in ccRCC tissues and normal tissues. Paired tissues (30 samples of each) were examined and TSC1 staining intensity was classified into 2 levels: weak = 1 and moderate = 2. Mean scores of normal tissues and ccRCC tissues are shown in the bar graph. Scale bar = 100 μm