The miR-20-Rest-Wnt signaling axis regulates neural progenitor cell differentiation

Abstract

Increasing evidence suggests that three dimensional (3-D) cell cultures are an improvement over traditional two dimensional (2-D) cell cultures. Current researches have extensively focused on the study of utilizing biomaterial-based 3-D culture systems to study and direct stem-cell fate both in vitro and in vivo. Here in our study, we screened the differential expression patterns of miRNAs between 2-D cultured and 3-D cultured NPCs using microarray analysis. Among these differentially expressed miRNAs, miR-20 was found to increase during differentiation of NPCs. Specifically, the facilitative effect on neural differentiation of miR-20 is mediated, at least in part by directly target the Rest gene, which is essential for preventing neural differentiation and maintaining NPCs self-renewal. Furthermore, the expression of miR-20 was decreased when the WNT pathway was inhibited by knock down of β-catenin or by exogenous Dkk protein, whereas it increased when the WNT pathway was activated by exogenous Wnt3a protein. Overall, miR-20, Rest and Wnt signaling are suggested to be involved in a regulatory circuit that can modulate the neural differention of NPCs. This novel regulatory circuit provides additional insight into how microRNAs interact with signaling molecules during neural differentiation of NPCs, allowing for fine-tuning of intricate cellular processes.

Figure 1. The morphology and characteristic of collagen sponge scaffold and 3-D cultured NPCs.

(A) Scanning electron microscopy (SEM) images of the collagen sponge scaffold. (B) Evaluation of pore size distribution and porosimetry by mercury porosimetry. (C) The morphology of the 3-D cultured NPCs was observed by SEM. (D) MiRNA array profiles of differentially regulated miRNAs and their target genes, which were identified in NPCs seeded on 2-D and 3-D substrates using bioinformatic analysis. The node sizes and line colors are correlated with expression changes of the miRNAs. MiRNAs covered by red nodes were up-regulated in 3D cultured NPCs and miRNAs covered by green nodes were down-regulated in 3D cultured NPCs.

Figure 2. MiR-20 modulates Rest expression.

(A) A schematic representation of putative miRNA-binding sites (shown in red) in the 3′UTR sequence of Rest. Sequence alignment indicated that miR-20 and its predicted binding site in the Rest 3′UTR are 100% conserved in vertebrates. The seed region is underlined. Results of the dual luciferase reporter assay using HeLa cells (B) and NPCs (C). Luciferase activity of Rest wild-type 3′UTR vectors (wt) or its mutant derivative lacking the miRNA binding sites (mut) in HeLa cells. The results were normalized with the pRL-CMV-Renilla Luciferase control. Relative luciferase level = (S luc/S renilla)/(C luc/C renilla). Luc, raw firefly luciferase activity; Renilla, internal transfection control renilla activity; S, sample; C, WT + NC group. The data are shown as the means ± SD. From 3 independent repetitions. *P < 0.05 versus WT + NC and **P < 0.01 versus the corresponding WT + NC. The western blot analysis showed that MiR-20 negatively regulated Rest protein expression in HeLa cells (D) and NPCs (E). (ctr: Control vector transfection; Inhibitor: miR-20 inhibitor; Mimics: miR-20 mimics; SiRNA: Rest siRNA;NC: negative control; INNC: inhibitor negative control) (F) Western blot assay indicated that the miR-20 inhibitor may rescue the inhibitory effect on the expression of Rest resulted by Rest siRNA. The datas are shown as the means ± SD. From 3 independent repetitions. *P < 0.05 versus ctr and **P < 0.01 versus ctr.

Figure 3. The expression pattern of miR-20 and Rest in 2-D and 3-D cultured NPCs.

Quantitative real-time PCR analysis showed the time-dependent elevation of miR-20 mRNA levels during the differentiation process (A). In contrast, the mRNA levels of Rest were reduced in a time-dependent manner (B). The datas are shown as the means ± SD. From 3 independent repetitions. *P < 0.05 versus 0 and **P < 0.01 versus 0.

Figure 4. The regulatory circuit of miR-20, Rest and Wnt signaling.

(A) Activation of Wnt signaling induced miR-20 activation. NPCs were treated with Wnt-3a or DKK1 and were harvested at the indicated times. Total RNA was extracted and miR-20 expression was measured by qPCR. The results were normalized to U6 RNA as an internal control. (B) A proposed model for the regulatory loop between miR-20, Rest and Wnt signaling in NPCs. The arrows represent Wnt activation and the bars represent repression. (C) The expression level of miR-20 was significantly attenuated when β-catenin was knocked down by siRNA in NPCs in a dose-dependent manner. (D) A working model for the relationship between miR-20, Rest and Wnt signaling involved in the neuronal differentiation of 3-D cultured NPCs. The data represent the means ± S.D. (n = 3). *P < 0.05 versus ctr and **P < 0.01 versus ctr.

Figure 5. MiR-20 regulated NPCs differentiation.

(A) qPCR data showing mRNA levels of Nestin, Sox2, Vimentin, Tuj1 and Map2 genes during NPCs differentiation. (B–F) Immunostaining images and quantified data of Nestin (B), Sox2 (C), Vimentin (D), Tuj1 (E) and Map2 (F) positive cells in NPCs transfected with miRNA mimics, miRNA inhibitor or Rest siRNA alone in differentiation medium or differentiation medium containing Wnt3a or DKK1 for 96 h. Scale bar, 50 μm (Top panel: immunostaining images; Bottom panel: quantified data from positive immunostaining cells). Quantitation and representative photomicrographs showed that miR-20 promotes cell differentiation in NPCs. Bars show mean ± SD. All experiments were repeated three times. *P < 0.05 vs. ctr, **P < 0.01 vs. ctr, ***P < 0.001 vs. ctr.

Figure 6. The percentage of Nestin, Sox2, Vimentin, Tuj1, Map2 and GFAP positive cells determined by Fluorescence-activated sorting (FACS) analysis.

Representative images showed the expression level of these genes in NPCs transfected with miRNA mimics, miRNA inhibitor or Rest siRNA alone in differentiation medium or differentiation medium containing Wnt3a or DKK1 for 96 h. An isotype control is needed to determine whether fluorescence emitted is due to non-specific binding of the fluorescent antibody. The datas are shown as the means ± SD. From 3 independent repetitions. *P < 0.05 versus ctr, **P < 0.01 versus ctr, ***P < 0.001 vs. ctr.

Figure 7. MiR-20 promoted neuronal differentiation in 3-D cultured NPCs.

(A,B) Immunofluorescence detection of Tuj1 (A) and Map2 (B) positive cells in 3-D cultured NPCs after transfection with miR-20 mimics, inhibitor alone, or cultured in medium containing Wnt3a or DKK1. Scale bar, 250 μm (Left panel: immunostaining images; Right panel: quantified data from positive immunostaining cells). Bars show mean ± SD. All experiments were repeated three times. *P < 0.05 vs. ctr, **P < 0.01 vs. ctr, ***P < 0.001 vs. ctr.