Meta gene set enrichment analyses link miR-137-regulated pathways with schizophrenia risk.

Background Glioblastoma multiforme (GBM) is an invariably fatal central nervous system tumor despite treatment with surgery, radiation, and chemotherapy. Further insights into the molecular and cellular mechanisms that drive GBM formation are required to improve patient outcome. MicroRNAs are emerging as important regulators of cellular differentiation and proliferation, and have been implicated in the etiology of a variety of cancers, yet the role of microRNAs in GBM remains poorly understood. In this study, we investigated the role of microRNAs in regulating the differentiation and proliferation of neural stem cells and glioblastoma-multiforme tumor cells. Methods We used quantitative RT-PCR to assess microRNA expression in high-grade astrocytomas and adult mouse neural stem cells. To assess the function of candidate microRNAs in high-grade astrocytomas, we transfected miR mimics to cultured-mouse neural stem cells, -mouse oligodendroglioma-derived stem cells, -human glioblastoma multiforme-derived stem cells and -glioblastoma multiforme cell lines. Cellular differentiation was assessed by immunostaining, and cellular proliferation was determined using fluorescence-activated cell sorting. Results Our studies revealed that expression levels of microRNA-124 and microRNA-137 were significantly decreased in anaplastic astrocytomas (World Health Organization grade III) and glioblastoma multiforme (World Health Organization grade IV) relative to non-neoplastic brain tissue (P < 0.01), and were increased 8- to 20-fold during differentiation of cultured mouse neural stem cells following growth factor withdrawal. Expression of microRNA-137 was increased 3- to 12-fold in glioblastoma multiforme cell lines U87 and U251 following inhibition of DNA methylation with 5-aza-2'-deoxycytidine (5-aza-dC). Transfection of microRNA-124 or microRNA-137 induced morphological changes and marker expressions consistent with neuronal differentiation in mouse neural stem cells, mouse oligodendroglioma-derived stem cells derived from S100β-v-erbB tumors and cluster of differentiation 133+ human glioblastoma multiforme-derived stem cells (SF6969). Transfection of microRNA-124 or microRNA-137 also induced G1 cell cycle arrest in U251 and SF6969 glioblastoma multiforme cells, which was associated with decreased expression of cyclin-dependent kinase 6 and phosphorylated retinoblastoma (pSer 807/811) proteins. Conclusion microRNA-124 and microRNA-137 induce differentiation of adult mouse neural stem cells, mouse oligodendroglioma-derived stem cells and human glioblastoma multiforme-derived stem cells and induce glioblastoma multiforme cell cycle arrest. These results suggest that targeted delivery of microRNA-124 and/or microRNA-137 to glioblastoma multiforme tumor cells may be therapeutically efficacious for the treatment of this disease.

Introduction Neurogenesis in adult mammalian brains occurs throughout life. This process has been observed at two locations under normal conditions: the subventricular zone of the lateral ventricles and the subgranular zone of the dentate gyrus (DG) in the hippocampus (Zhao et al., 2008). The cellular basis for adult neurogenesis is adult neural stem cells (aNSCs), which exhibit the two essential properties of stem cells: self-renewal and multipotency. Adult neurogenesis is defined as the process of generating new neurons from NSCs, which consists of the proliferation and fate determination of aNSCs, the migration and survival of young neurons, and the maturation and integration of newly produced neurons (Ming and Song, 2005; Zhao et al., 2008). Adult neurogenesis is regulated at many levels by both extrinsic factors, such as physiological and pathological conditions, and intrinsic factors, such as genetic and epigenetic programs. The maintenance and differentiation of stem cells is tightly controlled by intricate molecular networks (Li et al., 2008). Uncovering these regulatory mechanisms is crucial to understanding the functions and plasticity of adult brains. Epigenetic regulation, including DNA methylation and histone modification, is known to play significant roles in the modulation of stem cell proliferation and differentiation, including NSCs (Abel and Zukin, 2008; Zhao et al., 2008). Recent genome-wide analyses have demonstrated a clear role for DNA methylation and chromatin remodeling, particularly by the Polycomb group (PcG) proteins, in defining the properties and regulating the functions of stem cells (Bernstein et al., 2007). The importance of epigenetic regulation in brain development and neurological disorders has been well documented (Shahbazian and Zoghbi, 2002; Abel and Zukin, 2008). For example, de novo mutations in MECP2 give rise to neurodevelopmental disorders, including Rett syndrome (Amir et al., 1999; Chahrour and Zoghbi, 2007). MeCP2 belongs to a family of DNA methyl-CpG–binding proteins (MBDs) that translate DNA methylation into gene expression regulation (Bird, 2002). Two members of the MBD family of proteins, MBD1 and MeCP2, influence either the proliferation and differentiation of aNSCs or the maturation of young neurons (Zhao et al., 2003; Kishi and Macklis, 2004; Smrt et al., 2007). Nonetheless, how these epigenetic factors regulate adult neurogenesis is unclear because of the difficulty in identifying downstream targets via classical gene expression analyses (Bienvenu and Chelly, 2006). MicroRNAs (miRNAs) are small, noncoding RNAs that regulate gene expression and development by post-transcriptionally targeting RNA-induced silencing complex (RISC) to cognate messenger RNA (Bartel, 2004). The loss of components of the miRNA pathway, including Dicer and DGCR8, can alter the proliferation and differentiation of stem cells (Bernstein et al., 2003; Wang et al., 2007). Furthermore, specific miRNAs are known to play important roles in modulating the proliferation and differentiation of many types of stem cells (Ivey et al., 2008; Yi et al., 2008). Here, we show that MeCP2 could epigenetically regulate specific miRNAs in mouse aNSCs. The absence of MeCP2 binding to the genomic region proximal to one such miRNA, miR-137, correlates with an altered chromatin state that is reflective of miR-137 expression. In addition, we demonstrate that the MeCP2-mediated effect on miR-137 expression could be performed through a mechanism involving Sox2, a core transcription factor regulating stem cell self-renewal (Zappone et al., 2000; Avilion et al., 2003; Ferri et al., 2004). Furthermore, we found that miR-137 influences aNSC proliferation and differentiation both in vitro and in vivo. Lastly, we identified Ezh2, a histone H3 lysine 27 methyltransferase and a member of the PcG protein family, as one of the post-transcriptionally regulated targets of miR-137; and we found that the miR-137–mediated repression of Ezh2 subsequently caused a global decrease in trimethyl H3-lysine 27 (H3-K27-Tri-Me). Functionally, coexpression of Ezh2 rescued the phenotypes associated with miR-137 overexpression. These results demonstrate that cross talk between epigenetic regulation and the miRNA pathway could play important roles in the modulation of adult neurogenesis. Furthermore, our data suggest that the loss of functional MeCP2 could alter the expression of specific miRNAs, potentially contributing to the molecular pathogenesis of Rett syndrome. Results Identification of miRNAs with altered expression in Mecp2-deficient aNSCs To identify miRNAs potentially regulated at the epigenetic level and determine whether Mecp2 could influence the expression of miRNAs in the context of adult neurogenesis, we profiled the expression of 218 miRNAs in primary NSCs derived from wild-type (WT) and Mecp2-deficient mice (Chen et al., 2001) using multiplex reverse transcription and miRNA-specific TaqMan assays (Fig. 1 and Table S1). Nearly all cultured NSCs were positive for the NSC markers Nestin and Sox2, which suggests a relative homogeneity in these primary aNSCs. These aNSCs incorporate the thymidine analogue BrdU under proliferating conditions and produce both β-III tubulin (TuJ1)-positive neuronal cells and glial fibrillary acidic protein (GFAP)-positive glial cells under differentiating conditions, demonstrating that they possess the same essential properties as NSCs (Fig. 1 A). We identified a subset of miRNAs that consistently displayed altered expression in the absence of Mecp2, relative to WT aNSCs (Fig. 1, C and D; and Table S1). When considering 95% confidence intervals (CIs) on mean relative quantities (RQs), we identified four miRNAs with expression decreased by ≥2.5-fold and three miRNAs with a ≥2.5-fold increase in MeCP2-/y aNSCs (Fig. 1 D). These results suggest that the loss of functional MeCP2 leads to the dysregulation of a subset of specific miRNAs in the context of neurogenesis. Figure 1. Identification of miRNAs with altered expression in Mecp2-deficient adult NSCs. (A) Adult NSCs cultured under proliferating conditions express Sox2 (nuclear, green) and Nestin (cytoplasmic, red), and incorporate BrdU (nuclear, red). Adult NSCs used in this study were multipotent and, when subjected to differentiation, expressed neuron-specific TuJ1 (red) and astrocyte-specific GFAP (green, DAPI is shown in blue), Bar, 50 µm. (B) Western blot showing expression of MeCP2 in WT aNSCs and the absence of MeCP2 in MeCP2-/y aNSCs (ab2828 antibody; Abcam). (C) Heat map of miRNA with a ≥2.5-fold change in expression in proliferating MeCP2-/y aNSCs. Quantities relative to WT aNSCs from each of four independent miRNA profiling experiments are shown. Relative quantity scale is shown below for reference. (D) Relative quantity of miRNA with ≥2.5-fold change in expression shown for MeCP2-/y proliferating aNSCs, calibrated to WT proliferating aNSCs. WT relative quantity = 1, mean relative quantity from three WT/MeCP2-/y pairs plus one pooled sample per genotype is plotted, with error bars representing a 95% CI. Dotted lines indicate a threshold of 2.5-fold change in expression. To assess the potential for epigenetic regulation of these altered miRNAs, we first evaluated the genomic context of each miRNA, including CpG content and phylogenetic conservation, two hallmarks of functional nonprotein-coding regulatory elements. The region immediately upstream of one conserved single copy miRNA, miR-137, is highly conserved, contains multiple CpG-rich regions, and has a consensus binding site for Sox2, an Sry-related HMG-box transcription factor that plays important roles in stem cell function and adult neurogenesis (see Fig. 3, A–C; Jaenisch and Young, 2008; Zhao et al., 2008). Elevated expression of miR-137 in the absence of Mecp2 was verified using independent assays (Fig. 2 A). Overall, miR-137 was 6.5-fold higher in proliferating MeCP2-/y aNSCs (Fig. 2 A). Furthermore, pri–/pre–miR-137 expression was significantly increased in the absence of MeCP2, which indicates that expression of miR-137 was in fact influenced at the level of transcription (Fig. S1 C). Figure 2. Expression of miR-137 is epigenetically regulated by MeCP2. (A) Verification of increased expression of miR-137 in MeCP2-/y proliferating aNSCs using independent real-time PCR (n = 6, mean relative quantity ± SEM, P = 0.022). (B) Schematic of the 7 kb proximal to miR-137 on chromosome 3qG1 that were assayed in ChIP experiments. The region 2.5 kb upstream is indicated, along with a previously identified transcriptional start site that lies 2.2 kb upstream of miR-137 (Shiraki et al., 2003; Carninci et al., 2005). (C) MeCP2-specific ChIP indicates the enrichment of DNA 2.5 kb upstream of the miR-137 genomic locus in WT aNSCs, but not MeCP2-/y aNSCs. Relative enrichment is calculated relative to IgG-only nonspecific control and normalized to the directly adjacent 1.5-kb upstream region (n = 3, two-way analysis of variance [ANOVA], Bonferroni post-test; ***, P 0.05). (C) Coexpression of Ezh2 rescued the decreased neuronal differentiation caused by the overexpression of miR-137 in aNSCs, as determined by the percentage of TuJ1-positive cells among infected, GFP-positive cells (unpaired t test, n = 3; *, P = 0.0016). (D) The miR-137 7mer-1A target site in the Ezh2 3′ UTR as predicted by TargetScan. (E) Ezh2-3′ UTR–dependent expression of a luciferase reporter gene was suppressed by miR-137 in HEK293FT cells. MiR-137–mediated suppression of luciferase was specific, as deletion of the miR-137 target site in the Ezh2 3′ UTR abolished repression by miR-137. Renilla luciferase-Ezh2–3′ UTR expression was normalized to firefly luciferase (n = 6 for HEK293FT cells, unpaired t test; **, P 0.5–1 × 109 infectious viral particles/ml. In vivo retroviral grating was performed as described previously (Zhao et al., 2006; Smrt et al., 2007). In brief, 7- to 8-wk-old C57B/L6 male mice were anesthetized with isoflurane, and virus (1.5 µl with titer >5 × 105/µl) was injected stereotaxically into the DG using the following coordinates relative to bregma: anteroposterior, −(1/2) × d mm; lateral, ±1.8 mm (if d > 1.6), or otherwise ±1.7 mm; and ventral, −1.9 mm (from dura). For each mouse, the sh-control virus was injected into the left DG, and the miR-137 virus was injected into the right DG. Mice received two BrdU injections per day (50 mg/kg, i.p.) for a total of seven injections, immediately after viral grafting. 1 wk after viral grafting, mice were deeply anesthetized with pentobarbital and perfused with saline followed by 4% PFA. Brains were dissected out, postfixed overnight in 4% PFA, and then equilibrated in 30% sucrose. 40-µm brain sections were generated using a sliding microtome and were stored in a −20°C freezer as floating sections in 96-well plates filled with cryoprotectant solution (glycerol, ethylene glycol, and 0.2 M phosphate buffer, pH 7.4, 1:1:2 by volume). Immunohistochemistry and confocal imaging analysis were performed as described previously (Smrt et al., 2007). Floating brain sections containing EGFP+ cells were selected for staining and matched by DG region. Sections were pretreated with 1 M HCl, as described previously (Tang et al., 2007). The primary antibodies used were chicken anti-GFP (A10262; Invitrogen), rat anti-BrdU (ab-6326; Abcam), and rabbit anti-DCX (4604; Cell Signaling Technology). The secondary antibodies used were anti–chicken Alexa Fluor 488 (A11039; Invitrogen), goat anti–rat Alexa Fluor 647 (A21242; Invitrogen), and goat anti–rabbit Alexa Fluor 568 (A11036; Invitrogen). The z-stack images of GFP-BrdU-DCX staining were taken at 1-µm resolution using a TE2000 (Nikon) equipped with a spin disc confocal microscope with an oil-immersion objective lens (40×, NA = 1.3; Carl Zeiss, Inc.) and MetaMorph quantification software (MDS Analytical Technologies); we then counted the proportion of GFP+DCX+ or GFP+BrdU+ out of total GFP+ cells. For colocalization analysis, roughly 70 GFP+ cells per animal were imaged. The data were analyzed using the Student’s t test. 3′ UTR dual luciferase assays of candidate miR-137 target mRNA 3′ UTR sequences of candidate mRNAs were PCR amplified directly from proliferating aNSC first-strand cDNA generated from 5 µg of TRIZOL-isolated total RNA using oligo-dT SuperScript III reverse transcription, according to the manufacturer’s instructions (Invitrogen). All primers were designed incorporating XhoI and NotI restriction sites and 4 bp of extra random sequence to aid in restricting digestion. XhoI- and NotI-digested PCR products were cloned into XhoI- and NotI-digested psiCHECK-2 dual luciferase vector (Promega). As a primary screen of candidate miR-137 targets, 293FT cells (3 × 103 cells per well, 96-well plate, grown overnight before transfection) were transfected with sh-miR-137 cloned into a pCR2.1 TOPO vector (sh-miR-137 TOPO) and psiCHECK-2–3′ UTR using TransFast Transfection reagent (Promega) according to the manufacturer’s instructions. As a control, psiCHECK-2 plasmid with no 3′ UTR and U6-neg-shRNA were cotransfected with U6-mir-137-shRNA TOPO or psiCHECK-2–3′ UTR, respectively. All transfections used a total of 1 µg of plasmid DNA. The ratio of luciferase–3′ UTR/shRNA plasmid was 1:2 for all experiments. Luciferase expression was detected using the Dual-Luciferase Reporter 1000 System (Promega) according to the manufacturer’s instructions. 48 h after transfection, hRLuc activity was normalized to hLuc+ activity to account for variation in transfection efficiencies, and miR-137–mediated knockdown of hRLuc activity was calculated as the ratio of normalized hRLuc activity in the U6-miR-137-shRNA treatments to normalized hRLuc activity in the U6-neg-shRNA treatments. All luciferase readings were taken from either three or four individual wells for each psiCHECK-2–3′ UTR construct and control construct tested. Each transfection experiment was repeated at least three times. The miR-137 target site in the Ezh2 3′ UTR was deleted using the QuikChange Site-Directed Mutagenesis kit (Agilent Technologies) to delete 5 bases (AAUAA) from the 7mer-1a miR-137 seed site in the Ezh2 3′ UTR luciferase reporter. Target site deletion was verified by Sanger sequencing. To confirm the specificity of miR-137 targeting the Ezh2 3′ UTR, the Ezh2 3′ UTR and Ezh2 3′ UTRΔmiR-137 were transferred by XhoI–NotI double digestion and T4 DNA ligation from psiCHECK2 into a pIS2 renilla luciferase vector modified with the addition of an XhoI restriction site and deletion of the SpeI restriction site. MiR-137–dependent Ezh2 3′ UTR luciferase assays were performed as described previously using 10 pmol of miR-137 duplex RNA or control miRNA duplex, and Promega Dual Luciferase Reporter System and pIS0 firefly luciferase as a control (Yekta et al., 2004). To test whether Ezh2 knockdown modulates NSC differentiation, Ezh2 shRNA plasmid was electroporated into WT NSCs along with NeuroD1-luciferase DNA and internal control E1α-Rluc DNA plasmids. Both Ezh2 mRNA and NeuroD1-luciferase activity were determined after 24 h of differentiation. Western blot analyses Protein samples were separated on SDS-PAGE gels and then transferred to polyvinylidene fluoride membranes (Millipore). Membranes were processed according to the ECL Western Blotting Protocol (GE Healthcare). anti-MeCP2 (ab2828; Abcam), anti-Ezh2 (4905; Cell Signaling Technology), anti–tri-methyl-histone H3 (Lys27, C36B11; 9733; Cell Signaling Technology), and anti-histone H4 (ab10158; Abcam) were used as primary antibodies at a 1:1,000 dilution. HRP-labeled secondary antibodies were obtained from Sigma-Aldrich (A0545) and were used at a dilution of 1:5,000. For loading controls, membranes were stripped and re-probed with the antibody against glyceraldehyde 3-phosphate dehydrogenase (GAPDH; Applied Biosystems AM4300). All Western blot quantifications were performed using ImageJ software from the National Institutes of Health. Real-time PCR relative quantification of Ezh2 mRNA 500 ng of total RNA isolated by TRIZOL from lentiviral sh-miR-137 and lentiviral sh-control–infected aNSCs was reverse transcribed using random hexamers to generate first-strand cDNA with SuperScript III according to the manufacturer’s instructions. 1 µl of cDNA was used directly in 20 µl SYBR Green real-time PCR reactions that consisted of 1× Power SYBR Green Master Mix, 0.5 µM forward and reverse primers, and nuclease-free water. 18S rRNA was used as an endogenous control for all samples, with 1 µl of cDNA diluted 1:10 in the nuclease-free water used. Reactions were run on an Applied Biosystems SDS 7500 Fast Instrument using the Standard 7500 default cycling protocol and SDS 7500 Fast System Software version 1.3.1 without the 50°C incubation. Primers for Ezh2 mRNA and 18S rRNA were designed using Primer Express 3.0 software (Applied Biosystems) and were as follows. Ezh2: forward, 5′-GGTGAAGAGTTGTTTTTTGATTACAGA-3′; and reverse, 5′-TCTCGTTCGATGCCCACATA-3′. 18S: forward, 5′-CGGCTACCACATCCAAGGAA-3′; and reverse, 5′-CCTGTATTGTTATTTTTCGTCACTACCT-3′. All real-time PCR reactions were performed in triplicate, and RQs were calculated using the ΔΔCt method (95% CI) with calibration to sh-control–treated samples. All primer sets were subjected to a dissociation curve analysis and produced single peaks on a derivative plot of raw fluorescence. Online supplemental materials Fig. S1 shows miR-137 and primary/precursor miR-137 expression in aNSCs. Fig. S2 shows determination of the epigenetic state of the miR-137 genomic locus using additional histone ChIP assays. Fig. S3 shows efficient delivery of control and miR-137 shRNA constructs by lentivirus. Fig. S4 shows verification of a functional interaction between miR-137 and Ezh2. Fig. S5 shows that miR-137 has distinct effects on the expression of CDK6 in different cell types. Table S1 shows miRNA profiling of aNSCs. Table S2 lists candidate miR-137 targets. Online supplemental material is available at http://www.jcb.org/cgi/content/full/jcb.200908151/DC1.

The maturation of young neurons is regulated by complex mechanisms and dysregulation of this process is frequently found in neurodevepmental disorders. MicroRNAs have been implicated in several steps of neuronal maturation including dendritic and axonal growth, spine development, and synaptogenesis. We demonstrate that one brain-enriched microRNA, miR-137, has a significant role in regulating neuronal maturation. Overexpression of miR-137 inhibits dendritic morphogenesis, phenotypic maturation, and spine development both in brain and cultured primary neurons. On the other hand, a reduction in miR-137 had opposite effects. We further show that miR-137 targets the Mind bomb one (Mib1) protein through the conserved target site located in the 3' untranslated region of Mib1 messenger RNA. Mib1 is an ubiquitin ligase known to be important for neurodevelopment. We show that exogenously expressed Mib1 could partially rescue the phenotypes associated with miR-137 overexpression. These results demonstrate a novel miRNA-mediated mechanism involving miR-137 and Mib1 that function to regulate neuronal maturation and dendritic morphogenesis during development.