Elevated expression of miR-10b in glioma cells and tumors. A, miR-10b expression in normal brain and glioma tumors. miR-10b expression was measured by qRT-PCR in gliomas of different grades (A2 and A3, astrocytoma; O3, oligodendroglioma; GBM, glioblastoma) and normal brain cortical tissues (Norm). B, miR-10b expression in normal human neural and glioma cells. Glioma cell lines (A172, U87, U251, U373, LN229, LN215, LN308, and LN464), low-passage tumorigenic GBM cells (BT74, BT70, and GBM8), and normal brain cells (mixed primary neuro-glial culture, NNC, and normal neural progenitors, hNP) were analyzed for miR-10b expression by qRT-PCR.

miR-10b silencing leads to cells cycle arrest and reduces glioma cell growth. A, Cell viability assay was performed on various glioma (A172, U87, U251, U373, LN229, LN215, LN308, LN464, and BT74) and non-glioma (MCF7, SH-SY5Y, and human NNC) cells transfected with either control (Control) or miR-10b 2′-O-MOE with phosphate backbone inhibitor (Anti-miR-10b). Error bars represent SD for 3-6 independent transfections; p<0.05 for all glioma lines. B, Inhibition of miR-10b in A172 glioma cells reduces BrdU incorporation. C, miR-10b inhibition leads to cell cycle arrest in glioma cells. A172 and U87 cells were analyzed by PI staining and FACS three days following transfections. D, miR-10b modulates expression of cyclins B1 and D1 in glioma cells. A172 cells were transfected with miR-10b inhibitor (Anti-miR-10b), mimic (miR-10b), and corresponding controls, and analyzed by Western blotting.

miR-10b inhibition leads to glioma cell death. A, Caspase 3/7 activity is induced in glioma cells treated with miR-10b inhibitor; *p<0.05, **p<0.005. B, miR-10b silencing results in caspase cleavage. Indicated glioma cells were analyzed by Western blots for cleaved (cl.) caspase-3 and -7. Anti-actin was used for loading control and representative immunoblot is shown. C, Inhibition of miR-10b leads to accumulation of glioma cells in sub-G1. U251 cells were analyzed by PI staining. Fractions of apoptotic cells (sub-G1) are indicated by arrows. D, miR-10b inhibition leads to autophagy, as determined by immunoblots with anti-LC3 antibody.

Validation of miR-10b targets Bim, TFAP2C, p16, and p21. A, Inhibition of miR-10b causes up-regulation of target mRNA expression. Glioma cells (A172 for detecting p21, TFAP2C, and Bim; LN464 for p16) were analyzed by qRT-PCR. p<0.05 for all mRNAs. B, Inhibition of miR-10b causes up-regulation of target proteins Bim, p16, p21, and TFAP2C. Indicated (in italics) glioma cells were analyzed by Western blot. Elevated levels of p16 were detected upon miR-10b inhibition in all tested cell lines that express p16 (LN215, LN464, and LN308). Representative immunoblots for actin expression are shown for the loading control. C, Response of luciferase reporters containing 3′UTRs of miR-10b targets. The following luciferase reporters were used: empty plasmid (Empty), with miR-10b perfect binding site (miR-10b), with Bim, TFAP2C, p21, and p16 wild-type and corresponding mutant (Bim mut, TFAP2C mut, p21 d-mut [double mutant], and p16 mut) 3′UTR constructs. Glioma A172 cells (for Bim, p21, and TFAP2C) and HEK cells (for p16) were co-transfected with luciferase reporters and either miR-10b inhibitor or control oligos. Error bars represent SEM from 3–5 independent transfections; *p<0.05. D, Bim siRNA partially rescues apoptosis caused by miR-10b inhibition. Glioma LN308 cells were co-transfected with anti-miR-10b or its control oligo, and either siRNA to Bim, control siRNA or no siRNA (Mock), as indicated. The cells were stained with PI to examine DNA content by FACS.

Inhibition of miR-10b reduces glioma growth in vivo in U87 mouse model. A, Mice with established subcutaneous U87 tumors were injected intratumorally with either inhibitor of miR-10b or corresponding control oligo at days indicated by arrows. Tumor growth was monitored by in vivo Fluc bioluminescence imaging. Each data point represents the mean ±SD. B, Established U87 tumors were injected with either control or miR-10b inhibitor for two days in a row. Two days later, mice were sacrificed and tumors were removed, sectioned, and stained for nuclei using DAPI (blue) and Ki-67 (green).

Analysis of miR-10b function in GBM as suggested by TCGA data. A, Expression of cell cycle genes positively correlates with miR-10b expression in GBM patients. Significantly correlated genes (p<0.0001) were assessed for enrichment level of GO biological processes terms and the enrichment scores are presented in the form of heatmaps, showing the terms in rows. The gradient of red color shows the level of enrichment and black corresponds to no enrichment. Enrichment of all correlated (All), negatively correlated (Negative), and positively correlated (Positive) genes is demonstrated for two miR-10b probes (Probe A and Probe B) used on the TCGA arrays. Correlation for miR-21 and miR-10a is demonstrated for comparison. B, Expression of “positive regulators of cell death” anti-correlates with miR-10b expression in GBM patients. C, Shorten survival of GBM patients is strongly associated with high miR-10 expression. Only three (top 10 miR-10 expressors out of 252 patients, top 63, and top 135) from numerous miR-10-based stratification strategies that exhibit significant association are shown. Log rank test p-values for the difference between two survival curves for the miR-10-high and -low expressor GBM patients are indicated.