MiR-218 Induces Neuronal Differentiation of ASCs in a Temporally Sequential Manner with Fibroblast Growth Factor by Regulation of the Wnt Signaling Pathway

Differentiation of neural lineages from mesenchymal stem cells has raised the hope of generating functional cells as seed cells for nerve tissue engineering. As important gene regulators, microRNAs (miRNAs) have been speculated to play a vital role in accelerating stem cell differentiation and repairing neuron damage. However, miRNA roles in directing differentiation of stem cells in current protocols are underexplored and the mechanisms of miRNAs as regulators of neuronal differentiation remain ambiguous. In this study, we have determined that miR-218 serves as crucial constituent regulator in neuronal differentiation of adipose stem cells (ASCs) through Wnt signaling pathway based on comprehensive annotation of miRNA sequencing data. Moreover, we have also discovered that miR-218 and Fibroblast Growth Factor-2 (FGF2) modulate neuronal differentiation in a sequential manner. These findings provide additional understanding of the mechanisms regulating stem cell neuronal differentiation as well as a new method for neural lineage differentiation of ASCs.

Mesenchymal stem cells are the ideal candidates for regenerative medicine and tissue engineering 1 . Generating neuronal cells from stem cells is an attractive approach given the limited intrinsic capacity of neurons in repairing neural tissue. In vitro studies have shown that the mesenchymal stem cells could differentiate into mature neurons expressing neuronal specific markers after exposure to various chemical agents [2][3][4] . However, these chemical induction methods are usually of low efficiency and considerable cytotoxicity. Recently, gene therapy has developed to meet this challenge. The approach involves the use of multipotential cells such as bone marrow-derived mesenchymal stem cells (BMSCs), muscle-derived stem cells (MDSCs) and adipose-derived stem cells (ASCs), which are engineered to overexpress factors that are of crucial roles of neurogenesis for promoting neuronal differentiation [5][6][7] .
Gene expression and related function of stem cell are controlled by a newly discovered class of short 22 nucleotides Micro-RNAs (miRNAs). MiRNAs interact with complex signal transduction pathways, including those involved in neuronal formation and development, by regulating the protein translation of specific cellular mRNAs and mRNAs degradation [8][9][10][11] . In the last decade, there has been an increase in our understanding of the role of miRNAs in neuronal development and stem cell neuronal differentiation, where miRNAs have shown to be involved in important genes that control cell pluripotency. Meanwhile, Researchers endeavor to manipulate the expression of particular miRNAs in order to promote stem cells differentiation into neural progenitor cells or authentic neural cells [12][13][14][15] . For example, miR-146a has been shown to be a key regulator of stem cell survival when

Results
MiRNA expression profile analysis revealed that the Wnt signaling pathway and miR-218 were crucial for neuronal differentiation of ASCs. After 15 days of incubation with RA, the neurite outgrowth has been observed when the ASCs is differentiated into the neuronal lineage. The differentiation of ASCs into the neuronal lineage is confirmed by the expression of the neural terminal differentiation marker, βIII-Tubulin, using immunofluorescence (Fig. 1a). The protein expression of differentiation markers such as OCT4, SOX2, β III-TUBULIN and MAP2 are also monitored at different time points (Day 0, Day 2, Day 5, Day 10 and Day 15) of RA treatment. (Figure 1b,c). The reduction of stemness markers (OCT4 and SOX2) is accompanied by enhancing the neural cell markers (β III-TUBULIN and MAP2). The percentage of cells quantifies for this transformation (see Supplementary Figs S1 and S2). To elucidate the expression pattern of miRNAs during neuronal differentiation, high-throughput deep sequencing is performed using an Applied Biosystems SOLiD System. From the miRNA profiling results, about dysregulated 654 miRNAs are summarized. With the fold-change and Z-test analysis in the sequencing results, we have found that the expression levels of miRNAs are widely affected while the ASCs are differentiated into neuronal lineage and some miRNAs expression levels are more tempestuously regulated, including miR-146a, miR-196b, miR-31, miR-218, miR-214, miR-203, miR-124, miR-26a, miR-222, miR-375, miR-9, and let-7 family (Fig. 1d). The expression levels of some miRNAs are implicated in the development of neurons, such as miR-9, miR-214 and the let-7 family (the expression levels of miR-9, miR-146a and miR-214 are detected at 3 time points, see supplementary Fig. S3). Meanwhile, we evaluate the target genes of this miRNA pool by bioinformatics and subject to DAVID database. The functions of the target genes predicted by obviously altering miRNAs are annotated with KEGG signaling pathway analysis. From the P-Value analysis in the terms of the biological process, the Wnt signaling pathway (P -Value = 6.3) is likely to be critical for ASCs neuronal differentiation (Fig. 1e). The key gene expressions in Wnt signaling pathway (Wnt3a, Tcl4, Lef1, β-Catenin and Axin2 in Wnt/β -Catenin pathway) predicted with bioinformatics are validated by qRT-PCR (Fig. 1f). After addition of ICG-001 protein and subsequently adding RA for 15 days (anti-Wnt/RA group), Wnt signaling pathway is effectively inhibited by decreasing the expression levels of phosphorylation FZD (p-FZD) and β -CATENIN (Fig. 1g,h). As expected, OCT4 and SOX2 protein levels are unchanged and β III-TUBULIN is undetectable in anti-Wnt/RA group (Fig. 1i,j).
We further investigate the Wnt signaling pathway genes. From KEGG analysis results, the red pentacles reveal the key genes closely related to the ASCs neuronal differentiation process (Fig. 2a). The genes involving in Wnt signaling (red pentacles) or regulation of Wnt signaling (blue pentacles) are targeted by miRNAs (blue circles). From our sequencing and predicted data, the top level among the differentially expressed miRNAs is highlighted and reveals that miR-218 has significantly up-regulation after RA treatment (Z-test = 42.3, Fig. 2b), which is accord with the previous reports about the key regulator in Wnt Signaling [21][22][23] . Therefore, miR-218 is considered to be crucial for ASCs neuronal differentiation. Indeed, we find RA supplementation in culture medium increases the endogenous miR-218 expression by almost 8.5-fold (Fig. 2c) and downregulates the expression of the OCT4 and SOX2 simultaneously (Fig. 2d,e). However, after anti-miR-218 transfection, subsequent RA treatment does not enhance β III-TUBULIN expression (in anti-miR-218/RA group, Fig. 2d,e).
Taken together, these data demonstrated that the Wnt signaling pathway and miR-218 both participate and positively promote ASCs neuronal differentiation.

MiR-218 regulates Wnt signaling pathways but is insufficient to induce ASCs differentiation into neural cells.
MiR-218 is specifically active in developing motor neurons. The robust upregulation of miR-218 in ASCs, differentiate ASCs into the neural lineage inspires us to investigate whether over expressing miR-218 may induce the neuronal differentiation of ASCs through Wnt signaling pathway. The miR-218 transfection markedly increases the endogenous miR-218 levels by almost 100-fold (Fig. 3a). While the transfection of anti-miR-218 significantly decrease the miR-218 expression. The expression of miR-218 target genes like Robo1, Robo2 and Lamb3 and the Wnt signaling pathway antagonist genes such as Sfrp2 and Dkk2 are validated by qRT-PCR with the cells transfected with miR-218 and anti-miR-218 (Fig. 3b). In comparison to the transfection Scientific RepoRts | 7:39427 | DOI: 10.1038/srep39427 of anti-miR-218 and the controls (miR-NC), transfection of miR-218 dramatically enhances p-FZD levels and elevates nuclear accumulation of β -CATENIN (Fig. 3c,d). These results indicate that miR-218 transfection can activate Wnt signaling pathway. Meanwhile, it seems that the anti-Wnt does not have any effect on the expression of miR-218 (see Supplementary Fig. S4).
However, both the mRNAs levels of Oct4, Sox2, βIII-Tubulin, Map2 and Nestin (Fig. 3e) and the protein expression studies (Fig. 3f,g) prove that, overexpression of miR-218 alone cannot induce βIII-Tubulin, Map2 and Nestin expression in the absence of RA. These results confirm that, although miR-218 positively regulates Wnt signaling pathway, which alone is insufficient to induce ASCs differentiation into neural cells.

FGF2 and miR-218 co-operate sequentially in ASCs neural differentiation. Previous studies
demonstrate that the FGF signaling pathway participates in neurogenesis and central nervous system formation [24][25][26] . In our earlier studies, we identified that addition of FGF2 (10 ng/mL) may work as a pre-induction factor and affect ASCs neuronal differentiation 27 . Herein, we speculate that FGF2 may interact with miR-218 to induce   neuronal differentiation. To confirm this, we pre-treat ASCs with FGF2 (10 ng/mL) for 10 days followed by transfecting with miR-218. Subsequently, elevation of miR-218 levels in ASCs (+ FGF2/miR-218 group) increases the expression of Wnt signaling pathway markers (p-FZD and β -CATENIN) in the + FGF2/miR-218 group compared to the -FGF2/-miR-218 and + FGF2/anti-miR-218 groups (Fig. 4a,b). This indicates, FGF2 and miR-218 work synergistically for enhancing Wnt signaling pathway. The morphological transformation to neural-like cells and the expression of βIII-Tubulin are confirmed by immunofluorescent imaging and photomicrograph (Fig. 4c). At the same time, the two-color Flow Cytometry (flow cytometric dot plots) shows, there is an increase (0.1% to 41.4%) of double-positive (βIII-Tubulin + cells) cells in + FGF2/miR-218 group compared to + FGF2/anti-miR-218 group (Fig. 4d). These results indicate, the temporal relationship between FGF2 and miR-218 on the neuronal differentiation. The FGF2 pretreatment cooperatively interacts with miR-218 to induce ASCs into neural lineage.
To evaluate in sequential manner, we treat cells with FGF2 followed by transfecting with miR-218 (miR-218/+ FGF2 group) or anti-miR-218 (anti-miR-218/+ FGF2 group) (Fig. 4e). Interestingly, neither morphological changes nor β III-TUBULIN expression can be detected (Fig. 4f,g) in the above treated cells. These results indicate that pre-induction with FGF2 is necessary to facilitate the neuronal induction effect of miR-218 in ASCs. In addition, we find that miR-218 and FGF2 does not form the negative feedback loop (see Supplementary Fig. S5).
Taken together, our data indicate, elevation of FGF2 and miR-218 cooperatively induces stem cells to differentiate in a temporally sequential manner via the Wnt signaling pathway.
Similarly, we predicted from bioinformatics analysis [37][38][39] , and then found, that Wnt signaling pathway was closely involved and played a pivotal role in the neural differentiation process of ASCs. In the Wnt signaling pathway (involving Fzd/β -Catenin pathway), the combination of Wnt proteins and the receptors led to an increase in activity of glycogen synthase kinase 3β (Gsk3β) and Axin2. Then, the β-Catenin undergoes a nuclear translocation where it accumulated and formed complexes with transcription factors, activating a number of intracellular signaling pathways [40][41][42] . Previous evidence revealed that Wnt signaling pathway promoted stem cell self-renewal and participated in neurogenesis 43  Furthermore, certain miRNAs have shown to be involved with important genes that controlled the cell pluripotency and mediated the induction of pluripotent stem cells by targeting the Wnt signaling pathway. For example, considerable evidence suggested that the Wnt signaling pathway has been regulated by miR-499, miR-355, miR-375, miR-27, miR-29, miR-17, miR-142 and miR-218 [46][47][48][49][50][51][52][53][54] . The Dkk2, Sfrp2 and Sost are reported as the Wnt signaling pathway inhibitors and miR-218 targets these inhibitors as a positive feedback loop with Wnt signaling pathway 23,55 . Based on the sequencing data, we selected miR-218 for further investigation, not only because of its extreme differential expression in comparison to other miRNAs ( Fig. 1d) but mainly of its involvement in the Wnt signaling pathway (Fig. 2b-d).
To further support the hypothesis that Wnt signaling pathway and miR-218 were closely involved in the process of ASCs neural differentiation, we used anti-miR-218 and the inhibitor ICG-001 to block ASCs neuronal differentiation and supplemented the ASCs culture medium with RA. Our experimental results verified the predictive identification of miR-218 and Wnt signaling in neural differentiation and using our modified bioinformatics analysis method further demonstrated their function as key cellular triggers of neuronal differentiation. Interestingly, miR-218 did not appear to play an instructive role neither in mesenchymal stem cell fate determination nor in motor neuron fate determination on its own. Given that the overexpression of miR-218 alone was not sufficient to induce the formation of motor neurons in chick neural tube or mouse embryonic stem cells (ESCs) and to induce the neuronal differentiation of ASCs as well 56,57 . Therefore, we speculated that there might exist a method of combinative regulation in the ASCs neuronal differentiation.
FGF2 belongs to the family of heparin-binding growth factors and has been described as a mitotic activator in the stem cells differentiations 27,58 . The FGF2 treatment of cultured stem cells provided mitogenic support and predominated in the induction. In the present study, we included FGF2 in the culture medium as a pre-induction factor for neural differentiation. With this pre-induction and the overexpression of miR-218, Wnt signaling pathway was stimulated to a greater extent compare to other conditions such as + FGF2/anti-miR-218 or miR-218. Additionally, we demonstrated that, synergistically, the supplementation of FGF2 and overexpression of miR-218 prompted the differentiation of ASCs into neural cells (Fig. 4c,d).
A  regulators (i.e., miRNAs). Therefore, we classified mesenchymal stem cells neural differentiation could be divided into two sequential stages: "induction" and "differentiation". The delivery of FGF2 was used for the induction of differentiation in the initial step, during which ASCs displayed a stage of increased activity triggered by FGF2 and were "conditioned" for the subsequent elevation of miR-218. Finally, we validated the sequential link between bFGF (FGF2) and miR-218 during ASC neural differentiation. Further investigations of the molecular mechanisms underpinned this link and the synergistic effects of FGF2 and miR-218 on the Wnt signaling pathway during neuronal differentiation were warranted.
In this study, we identified that Wnt signaling pathway and miR-218 were closely related to ASCs neuronal differentiation. We also demonstrated that miR-218 overexpression alone failed to induce ASCs neuronal differentiation and FGF2 pretreatment cooperatively interacted with miR-218 to generate neural cells. Furthermore, FGF2 and miR-218 were shown to operate in a temporally sequential manner to promote the differentiation of ASCs into the neural lineage. Our results augmented, the current understanding of the developmental processes of neural differentiation and provided important insights into how miRNAs contributed to this process, which could assist the development of novel inductive approaches for neural tissue regeneration.

Materials and Methods
ASCs isolation and differentiation. ASCs were obtained from four-week old female Sprague-Dawley rats (weight 100-130 g), as previously described 61 . The use of all animal samples were approved by and carried out in accordance to the medical ethics committee of Southeast University, China. ASCs were digested and seeded. The cells were cultured in basal medium composed of Dulbecco's modified Eagle's medium (DMEM; Thermo Fisher Scientific, USA), 5% fetal bovine serum (FBS; Gibco Lab., USA), 1% penicillin/streptomycin (Thermo Fisher Scientific) with or without FGF2 (Peprotech, USA). The medium was replaced every 3 days for a period of 10 days. ASCs were then seeded into 12-well plates and neuronal differentiation (RA-treated, + RA group) was performed over a period of 15 days using basal medium supplemented with 100 ng/mL Retinoic Acid (RA; Sigma Aldrich, USA). The non-treated group (control, −RA group) was cultured contemporaneously.
MiRNA analysis. Total RNA were extracted from RA-treated and non-treated cells and altered miRNAs expression were detected by sequencing of the Applied Biosystems SOLiD System, as previously described 37,62,63 . Altered expression of miRNAs and prediction of effects on their target genes were analyzed following the method (based on the Z-test calculation method, a mathematical model to evaluate the comprehensive repression rate of specific mRNAs using total miRNA expression profiling) reported previously 37,64,65 . Briefly, the target genes levels between RA-treated (+ RA group) and non-treated (−RA group) cells were predicted by the tools of TargetScan version 6.2 database. The identified lists of genes were subjected to functional annotation, clustering and analysis using the KEGG analysis based on the Database for Annotation, Visualization and Integrated Discovery (DAVID) Bioinformatics Database (https://david.ncifcrf.gov/).

Quantitative real-time PCR. The miRNAs and mRNAs expression were measured by quantitative
Real-Time PCR (qRT-PCR) using ABI 7500 System (Life Technologies, USA) and standard protocols (All the Primers are listed in Supplementary Table S1 online). Relative amounts were determined using the 2 −ΔΔCt method.
ICG-001 treatment. ICG-001 protein inhibits the Wnt signaling pathway by binding to the element-binding protein (CBP) 66 . ASCs were seeded into 12-well plates and maintained at CO2 incubator. After 24 h of seeding, the final concentrations of 5 μ M ICG-001 (Selleckchem, USA) was added into the ASCs medium for 2 days, followed by 100 ng/mL of RA was added and further cultured for 15 days (anti-Wnt/RA group).
Western blotting. Total cellular proteins and nuclear proteins were separately extracted using the Total Protein Isolation Kit (Sangon Company, China) and Nuclear Protein Isolation Kit (Sangon Company). Proteins were quantified by Bicinchoninic acid assay kit (Sangon Company). The protein samples were loaded on SDS-PAGE gels and electrophoresed under standard conditions. Western blotting was performed using nitrocellulose membranes. After blocking, membranes were incubated with primary antibodies (1:200-1:500) at 4 °C overnight. After rinsing, incubation was conducted with secondary horseradish peroxidase-conjugated goat anti-rabbit or mouse IgG antibody (Bioss Biotechnology, China) and exposed to film. Primary antibodies included anti-Frizzled MiRNA transfection. ASCs were seeded into 12-well plates, after 24h of seeding, 5 μ L Superfectin Transfection Reagent (Qiagen, Canada) was added per well, according to the manufacturer's instructions. Transfections were performed using 20 nM of miR-NC, anti-miR-218 or miR-218, respectively. The miRNA plasmids were designed and enhanced green fluorescence protein (eGFP) was used as a reporter gene. After 48 h, transfected cells and transfection efficiency was measured and assayed by detection of eGFP. In the anti-miR-218/ RA group, the anti-miR-218 plasmids were transfected for 2 days and the ASCs medium was supplemented with 100 ng/mL RA for 13 days. The β III-tubulin positive expressions were analyzed by Flow Cytometry using C-Flow software.
Flow cytometry. Cells were trypsinized with 0.25% trypsin solution (Sigma Aldrich) and fixed with 4% formaldehyde (Sangon Company) in PBS at 37 °C for 0.5 h. After rinsing, the cells were incubated with rabbit anti-rat β III-Tubulin primary antibody (1:200 in 1% BSA; Santa Cruz Biotechnology) at 37 °C for 4 h then incubated with the appropriate amount of the secondary antibody, goat anti-rabbit Alexa-Fluor 647 (1:500 in 1% BSA; Invitrogen, USA) for 1 h at 37 °C. During the whole experiment the cells were protected from light. Finally, the cell samples were subjected to Flow Cytometry and corresponding data were analyzed by C-Flow software. Immunofluorescent analysis. After induction of RA and/or miR-218 for a period of time, ASCs were subjected to immunofluorescent staining by rabbit anti-rat β III-Tubulin primary antibody to detect differentiated neural cells. After washing with PBS, the samples were incubated with the appropriate secondary antibody, mouse anti-rabbit Alexa-Fluor 647 (1:200 in 1% BSA; Cell Signaling Technology) in the dark. Then samples were washed twice with PBS, the nuclei stained with 10 μ g/mL Hoechst 33342 (HOE, Sigma Aldrich,) for 0.5 h, and images were obtained using a Revolution XD confocal laser scanning microscope (Andor, Belfast, Northern Ireland).
Statistical analysis. All data were expressed as mean ± SD. Differences were compared using the Student's t-test; p-values < 0.05 were considered statistically significant (*p < 0.05, **p < 0.01).