miR-342-5p Regulates Neural Stem Cell Proliferation and Differentiation Downstream to Notch Signaling in Mice

Abstract

Notch signaling is critically involved in neural development, but the downstream effectors remain incompletely understood. In this study, we cultured neurospheres from Nestin-Cre-mediated conditional Rbp-j knockout (Rbp-j cKO) and control embryos and compared their miRNA expression profiles using microarray. Among differentially expressed miRNAs, miR-342-5p showed upregulated expression as Notch signaling was genetically or pharmaceutically interrupted. Consistently, the promoter of the miR-342-5p host gene, the Ena-vasodilator stimulated phosphoprotein-like (Evl), was negatively regulated by Notch signaling, probably through HES5. Transfection of miR-342-5p promoted the differentiation of neural stem cells (NSCs) into intermediate neural progenitors (INPs) in vitro and reduced the stemness of NSCs in vivo. Furthermore, miR-342-5p inhibited the differentiation of neural stem/intermediate progenitor cells into astrocytes, likely mediated by targeting GFAP directly. Our results indicated that miR-342-5p could function as a downstream effector of Notch signaling to regulate the differentiation of NSCs into INPs and astrocytes commitment.

Keywords: Notch; RBP-J; glia; intermediate neural progenitors; miR-342-5p; neural stem cells; neuron.

Graphical abstract

Figure 1

miR-342-5p Was a Downstream miRNA of the Notch Pathway in NSCs (A) Screening of Notch downstream miRNAs using microarray hybridization. Primary neurospheres were cultured with cells derived from the ganglion eminences (GE) of NesCre-Rbp-j^f/f (cKO) and NesCre-Rbp-j^+/f embryos (Ctrl) (E11.5). Differentially expressed miRNAs were compared by using microarray hybridization and clustered. RNA samples were derived from three different pairs of littermates. (B) The expression of Evl mRNA, miR-342-3p, miR-342-5p, Hes1 mRNA, and Hes5 mRNA in Rbp-j cKO and control neurospheres was determined by qRT-PCR. Neurospheres were derived from four different pairs of littermates. (C) Primary neurospheres were cultured by using cells derived from GE of wild-type embryos (E15.5). Cells were treated with 75 μM GSI for 12 hr, and the expression of miR-342-3p and miR-342-5p was examined by qRT-PCR. DMSO was used as a control. RNA samples were extracted from four pairs of GSI- and DMSO-treated neurospheres. (D) The expression of miR-342-5p, miR-342-3p, Hes1 mRNA, and Hes5 mRNA in the GE and striatum during neural development in mice was determined by qRT-PCR. The brain tissues at the specific time point came from four mice). (E) The expression of miR-342-5p in the telencephalon of E14.5 embryo was determined by in situ hybridization using a locked nucleic acid probe. The brain sections were derived from three different wild-type embryos. (a–d) Coronal sections of E14.5 telencephalon. (c′ and c″) The cortex and GE regions of the telencephalon section in (c) are shown in magnification, respectively. VZ, ventricular zone; SVZ, subventricular zone; IZ, intermediate zone; CP, cortical plate. (F) The expression of miR-342-5p and miR-342-3p in neurospheres of the first-generation (1°) and third-generation (3°) was determined by qRT-PCR. RNA samples were extracted from four pairs of passaged neurospheres. Bars, means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ns, not significant.

Figure 2

Notch Signaling Regulated the Evl Promoter (A) Schematic illustration of the mouse Evl promoter, the reporter constructs, and the position of primers used in ChIP. Thick line, thin lines, and filled boxes represent 5′ upstream sequence, introns, and exons, respectively. The miR-342-5p gene and miR-342-3p gene are represented by two pink arrows. The putative RBP-J-binding site, N boxes, E boxes, and a TATA box are indicated by a red dot, orange quadrangles, yellow quadrangles, and a blue box, respectively. Different promoter fragments were cloned into the pGL3-basic vector to construct the reporters including Evl-RepL (−5,000 → +100 bp), Evl-RepM (−3449 → +100 bp), and Evl-RepS (−1,227 → +100 bp). (B and C) Reporter assays. NIH3T3 cells were transfected with different reporters in combination with increasing amounts of pEFBOS-NICD (B) or pCMV-RBP-J^R218H (C). A Renilla luciferase (luc) expression vector was co-transfected as an internal control. Cell lysates were prepared 48 hr after the transfection, and luciferase activity was determined. Reporter assays were done by four independent transfections, and three wells of 96-well plates were prepared as replicate samples in each group. (D and E) ChIP assays. Crosslinked chromatins were prepared from cultured primary neurospheres (neurospheres derived from three different wild-type embryos), and immunoprecipitated with immunoglobulin G (IgG) and anti-HES1 (D) or anti-HES5 (E) antibody. After washing, the co-precipitated DNA was extracted and amplified by PCR primers targeting three different N boxes and an E box; numbers underlined in (A). (F) Neurospheres cultured from E15.5 cortex and GE of NesCre-ROSA-Stop^f/+-NIC (NICD) and ROSA-Stop^f/+-NIC embryos (Ctrl) were harvested and total RNAs were extracted. The expressions of Hes1, Hes5, Evl, and miR-342-5p were determined by qRT-PCR. Neurospheres were derived from three different pairs of littermates. (G) Neurospheres were transfected with siHes5 and control oligonucleotide (three independent transfections performed), and the expressions of Hes5, Evl, and miR-342-5p were analyzed by qRT-PCR. Bars, means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001.

Figure 3

Overexpression of miR-342-5p Repressed the Formation of Both Suspended Neurospheres and Adherent Colonies, and Repressed the Proliferation of NS/PCs (A) Single-cell suspensions were prepared from the GE of E15.5 embryos and cultured on poly-L-lysine (PLL)-coated dishes in DF12 medium supplemented with N², 20 ng/mL bFGF, and 20 ng/mL EGF. After 12 hr, NS/PCs were transfected with 50 or 100 nM miR-342-5p mimics or control oligonucleotides (Ctrl). Cells were detached 4 hr later and cultured under neurosphere conditions for 7 days. The cultures were photographed and the number of neurospheres was determined under a microscope. The transfections were done five times independently. (B and C) NS/PCs derived from normal embryos were transfected with 100 nM miR-342-5p mimics or control as in (A) (five independent transfections performed). Cells were then replated at clonal density on PLL-coated dishes for 7 days. The number of colonies (B) and cell number per colony (C) were determined. (D–I) NS/PCs were transfected with miR-342-5p mimics and the control, and were cultured adherently at clonal density for 7 days (D–G) (five independent transfections performed) or at normal density for 48 hr (H and I) (five independent transfections performed). BrdU was added into the medium 18 hr before the end of the experiment. Cells were fixed and stained with anti-BrdU and anti-NESTIN antibodies by immunofluorescence staining (D) or with anti-BrdU, anti-NESTIN, and anti-Ki67 antibodies (H). Nuclei were counterstained with a Hoechst stain. The percentages of BrdU+ (E), NESTIN+ (F), and NESTIN+ BrdU+ (G) cells within one colony were determined (50 colonies counted in each group). The percentages of NESTIN+ BrdU+ cells and NESTIN+ Ki67+ cells within one field were determined (I) (ten fields counted in each group). Bars, means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001.

Figure 4

miR-342-5p Promoted the Differentiation of NS/PCs into INPs In Vitro (A) NS/PCs were transfected with miR-342-5p mimics or control (five independent transfections performed), and cultured under the adherent condition to proliferate at clonal density for 7 days. The newly formed colonies were further induced to differentiate for 48 hr with low mitogens (DF12 containing 1/2 × N², 5 ng/mL bFGF and 5 ng/mL EGF) supplemented with 5% FBS. Cells were then fixed and stained with anti-GFAP and anti-TUJ1. The frequencies of GFAP+, TUJ1+, and GFAP+ TUJ1+ colonies were compared (50 colonies counted in each group). (B and C) Adherently cultured NS/PCs were stained with anti-PAX6 and anti-NESTIN (B), and percentages of PAX6+ cells in total cells or NESTIN+ cells were determined (C) (eight fields counted in each group). (D and E) NS/PCs were stained with anti-TBR2 and anti-NESTIN (D), and percentage of TBR2+ cells were determined (E) (eight fields counted in each group). Bars, means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001.

Figure 5

miR-342-5p Promoted Precocious Differentiation of NSCs In Vivo (A) The forebrain areas of E15.5 mouse embryos were electroporated in utero with pcDNA6.2-GW/EmGFP-miR-342-5p (miR-342-5p) that could express miR-342-5p and EGFP, or with pcDNA6.2-GW/EGFP-negative control (Ctrl) that expressed only EGFP as a control. At E18.5, the mice were perfused and brains were fixed, cryosectioned, and stained with anti-PAX6 using immunofluorescence. Nuclei were counterstained with a Hoechst stain. Arrows and arrowheads represent EGFP- and PAX6-double-positive cells and EGFP-single-positive cells, respectively. Five pairs of littermates successfully electroporated with miR-342-5p and Ctrl separately were obtained. (B) The percentage of PAX6-positive cells in EGFP-positive cells was compared (12 fields counted in each group). (C) The EGFP-positive cells in each bin of five bins of the transfected cortical region were counted, and the percentage of EGFP-positive cells of each bin in total EGFP-positive cells was determined. Asterisks represent statistical analysis of the difference between the miR-342-5p-transfected and the control groups in the bin 1 and bin 2 areas (ten fields counted in each group). (D) Samples in (A) were stained with anti-TBR2. The EGFP-positive cells were compared in the SVZ (TBR2+ zone), VZ (inside SVZ), and cortex (outside the SVZ) regions (ten fields counted in each group). Bars, means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ns, not significant.

Figure 6

miR-342-5p Targeted GFAP and Regulated Astrocyte Differentiation (A) The sequence of the 3′ UTR of Gfap was aligned with the seed sequence of miR-342-5p. The recognized sequences (nucleotides 202–207 and 1,246–1,251) are marked in red. (B) Adherently cultured NS/PCs were transfected with miR-342-5p mimics or the control (five independent transfections performed), and 4 hr later 5% FBS medium was applied to induce the differentiation for another 48 hr. Cells were collected and cell lysates were analyzed by western blot using anti-GFAP, anti-TUJ1, and anti-β-actin. (C) Reporter assays. Cos7 cells were transfected with different reporter constructs, pcDNA6.2-GW/EmGFP-miR-342-5p and pcDNA6.2-GW/EmGFP-miR-negative control, and phRL-TK-expressing Renilla luciferase as an internal control. Cells were lysed 48 hr after the transfection, and luciferase activities in the cell lysates were determined. Reporter assays were done by four independent transfections, and three wells of a 96-well-plate were prepared as replicate samples in each group. (D) Primary neurospheres (NS/PCs), oligodendrocytes, astrocytes, and neurons were isolated from rats (each type of cells derived from three different rat samples). Total RNA was extracted from cells and the expression of miR-342-5p was determined by qRT-PCR, with U6 RNA as an endogenous reference. (E and F) NS/PCs were transfected with miR-342-5p or the control (five independent transfections performed), and cells were induced to differentiate by 5% FBS for 48 hr, and then fixed and stained with anti-GFAP and anti-TUJ1. Nuclei were counterstained with a Hoechst stain. The percentages of GFAP+ cells and TUJ1+ cells were compared (eight fields counted in each group). (G) After differentiation, cells were collected, and the expressions of genes related with progeny cells were analyzed by qRT-PCR. Transfections of miR-342-5p mimic and the negative control were performed with 50 and 100 nM dosages (three independent transfections performed). Bars, means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ns, not significant.

Figure 7

Inhibition of miR-342-5p Could Rescue Part of the Phenotypes Caused by Notch Blockade (A and B) Adherent NS/PCs were treated with GSI or DMSO, and miR-342-5p inhibitor or control were transfected into these cells (four independent transfections performed). After 48 hr, NS/PCs were fixed and stained with anti-NESTIN antibody (A). The percentage of NESTIN+ cells within cells of one field was calculated (B) (eight fields counted in each group). (C) NS/PCs treated as those in (A) were resuspended to form neurospheres. After 7 days, neurospheres were counted and compared. Four independent transfections were performed, and three wells of a 12-well-plate were counted as replicate samples in each group. (D and E) NS/PCs treated as those in (A) were induced to differentiate for another 48 hr. Cells were then collected and stained with anti-GFAP antibody (D). The percentage of GFAP+ cells within cells of one field were analyzed (E) (eight fields counted in each group). Bars, means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ns, not significant.