Pleiotropic effects of miR-183~96~182 converge to regulate cell survival, proliferation and migration in medulloblastoma

Abstract

Medulloblastomas are the most common malignant brain tumors in children. Several large-scale genomic studies have detailed their heterogeneity, defining multiple subtypes with unique molecular profiles and clinical behavior. Increased expression of the miR-183~96~182 cluster of microRNAs has been noted in several subgroups, including the most clinically aggressive subgroup associated with genetic amplification of MYC. To understand the contribution of miR-183~96~182 to the pathogenesis of this aggressive subtype of medulloblastoma, we analyzed global gene expression and proteomic changes that occur upon modulation of miRNAs in this cluster individually and as a group in MYC-amplified medulloblastoma cells. Knockdown of the full miR-183~96~182 cluster results in enrichment of genes associated with apoptosis and dysregulation of the PI3K/AKT/mTOR signaling axis. Conversely, there is a relative enrichment of pathways associated with migration, metastasis and epithelial to mesenchymal transition, as well as pathways associated with dysfunction of DNA repair in cells with preserved miR-183 cluster expression. Immunocytochemistry and FACS analysis confirm induction of apoptosis upon knockdown of the miR-183 cluster. Importantly, cell-based migration and invasion assays verify the positive regulation of cell motility/migration by the miR-183 cluster, which is largely mediated by miR-182. We show that the effects on cell migration induced by the miR-183 cluster are coupled to the PI3K/AKT/mTOR pathway through differential regulation of AKT1 and AKT2 isoforms. Furthermore, we show that rapamycin inhibits cell motility/migration in medulloblastoma cells and phenocopies miR-183 cluster knockdown. Thus, the miR-183 cluster regulates multiple biological programs that converge to support the maintenance and metastatic potential of medulloblastoma.

Figure 1

miR-183∼96∼182 is differentially expressed in medulloblastoma cell lines and regulates cell proliferation. a Quantitative Real Time PCR (q-RT-PCR) analysis of miR-183, miR-96, and miR-182 expression in medulloblastoma cell lines. Raw data were normalized to the endogenous control RNU6B. The errors bars were derived from 3 independent experiments done in triplicate. b LNA-mediated knockdown of individual components of the miR-183∼96∼182 cluster results in a sponge effect. c LNA-mediated knockdown of individual or pooled components of the miR-183 cluster results in decreased cluster formation and overall cell viability. Bar graphs reflect combined data from three independent experiments with each condition plated at least in duplicate.

Figure 2

Knockdown of miR-183∼96∼182 cluster induces apoptosis and G1 and G2 arrest. a Cell cycle analysis of medulloblastoma cells transfected with LNAs directed towards individual components of the miR-183 cluster shows an increase in the G0/G1 and G2 fractions upon miR-183∼96∼182 pooled knockdown and miR-96 knockdown in particular. b Immunofluorescence staining shows an increase of apoptotic markers BAD and BAK1 (red) and BAX (green) after miR-183∼96∼182knockdown; all cells counterstained with DAPI (blue). P-values (Fisher's exact test) indicate significance relative to scramble control, *p-value<0.05, **p-value<0.01, ns=not significant.

Figure 3

miR-183∼96∼182 inhibits double strand break repair. a Immunofluorescence staining and western blot (b) of BRCA1 (red) and DAPI counterstain (blue) shows increased, predominantly cytoplasmic BRCA1 staining in D458 cells upon knockdown of miR183∼96∼182. c Representative COMET assay showing decreased ‘tail moment’ in the miR-183 cluster knockdown compared to the scrambled control. d Dot plot indicative of a decreased level of DNA double-strand breaks in miR-183 cluster knockdown relative to scrambled control (***P-value<0.0001).

Figure 4

miR-182 induces cell migration and invasion in medulloblastoma cells. a Scratch/wound healing assay in UW426 medulloblastoma cells shows knockdown of miR-182 inhibits wound closure. Notably, knockdown of miR-183 results in a sponge effect-mediated reduction of miR-182 leading to impeded wound healing. b Representative images from a transwell matrigel invasion assay shows reduction of cell invasion upon knockdown of miR183∼96∼182. Migrated cells are quantified in the bar graph reflecting data from three independent experiments. c Scratch assay after stable overexpression of individual or pooled miR-183∼96∼82 cluster components in DAOY cells shows that miR-182 alone is sufficient to enhance scratch closure. P-values (Fisher's exact test) indicate significance relative to scramble control, ***p-value<0.0001.

Figure 5

miR-183∼96∼182 promotes epithelial to mesenchymal transition (EMT). a Heatmap showing genes associated with EMT are downregulated upon miR-183∼96∼182 knockdown in D458 cells. b Immunofluorescence staining and Western Blot show decreased expression of vimentin in D458 cells after miR-183∼96∼182 knockdown. c Increased neurite/axonal outgrowth and elevated TUJ1 staining upon knockdown of miR-183∼96∼182. d Knockdown of miR-183∼96∼182 cluster induces the expression of pro-neural genes NeuroD2 and NeuroG1. P-values (Fisher's exact test) indicate significance relative to scramble control, **p-value<0.01, ***p-value<0.0001.

Figure 6

miR-183∼96∼182 cluster regulates the PI3K/AKT/mTOR signaling axis. a Immunofluorescence staining shows elevated AKT1 (red) and decreased AKT2 (green) upon knockdown of the microRNA cluster in D458 medulloblastoma cells. b Western Blot showing increased AKT1 and decreased AKT2 in pooled miR-183∼96∼182 knockdown compared to the scrambled control. c Increased migration upon forced expression of miR-183∼96∼182 cluster in DAOY cells is negated by rapamycin. d Impeded migration upon rapamycin treatment in UW426 cells comparable to the effects of miR-183∼96∼182 knockdown. Bar graphs represent migration into wounds measured in an unbiased manner based on pixel intensity. P-values (Fisher's exact test) indicate significance relative to scramble control, **p-value<0.01, ***p-value<0.0001.