Spinal cord injuries severely affect motor functions. Regeneration of the injured nerve is an increasingly promising approach in which stem cells are induced into neuron-like cells on a scaffold and transplanted. Bone mesenchymal stem cells (bMSCs) can be induced to differentiate into neuron-like cells in vitro1–3. Throughout the neuronal differentiation process, a few transcription factors play an important role such as mammalian achaete-scute homolog 1 (Mash1) 4, mammalian atonal homolog 1 (Math1) 5, and neurogenic differentiation 1 (NeuroD1) 6. T-box brain 1 (Tbr1) is a transcription factor gene of the T-box family that is expressed soon after cortical progenitors begin to differentiate, and in the Tbr1 mutant cortex, the laminar organization of neurons showed developmental abnormalities 7,8. Tbr1 also regulates the regional and laminar identity of postmitotic neurons in developing neocortex and promotes corticothalamic neuron identity by preventing the expression of subcerebral projection neuron characteristics 9,10.
MicroRNA (miRNA) is an endogenous small noncoding RNA, which can negatively regulate gene expression by fully or partially complementary with the 3′-untranslated region (3′UTR) of target gene mRNA 11–13. Several studies have shown a few miRNAs highly expressed in neuron-like cells 14,15. However, it is unclear whether miRNAs may regulate Tbr1 expression during the neural differentiation of bMSCs. In the pre-experiment, miR-122-5p was found by bioinformatics and luciferase assay. In this investigation, expressions of miR-122-5p and the Tbr1 gene were determined and the molecular biological effects after overexpression of miR-122-5p were also detected.
Isolation of bone mesenchymal stem cells and neuronal induction
The procedures were performed in accordance with the guidelines for animal experimentation of Jinzhou Medical University. Mouse bMSCs were isolated as published previously 16. Simply, 6-week-old C57BL/6 male mice were sterilized by immersion in 70% ethanol. After removal of the remaining skin and muscles, bMSCs were obtained from the femurs and tibias and then cultured. Adherent cells gaining 80% confluence were passaged and the cells of third to fifth passages were used for neuronal induction, which was performed using a modification of previous neuronal induction protocols 17. Briefly, the cells of third passages were seeded in dishes and then cultured in the basic medium supplemented with 100 μM butylated hydroxyanisole (BHA; Sigma St. Louis, Missouri, USA), 0.5 μM retinoic acid (RA; Sigma St. Louis, Missouri, USA), 10 ng/ml basic fibroblast growth factor (bFGF; Invitrogen Carlsbad, California, USA), and 10 ng/ml nerve growth factor (NGF; R&D Systems Minneapolis, Minnesota, USA). The culture medium was changed every 3 days.
Bone mesenchymal stem cells immunophenotype analysis
Standard flow cytometry was used to determine the surface antigen expression of bMSCs. Cells were detached with 0.125% trypsin and resuspended into a single-cell suspension at a concentration of 106/ml. Cells were labeled with the following monoclonal antibodies: CD29-PE, CD34-PE, CD44-PE, CD45-PE, and CD105-PE (Biolegend San Diego, California, USA). The labeled cells were acquired on a BD LSR II flow cytometer and the data were analyzed using the WinMDI software (Purdue University Cytometry Laboratories, West Lafayette, USA).
Induced cells were fixed with ice-cold methanol for 8 min, permeabilized with 0.05% Triton X-100 (Sigma) for 5 min, and blocked with 5% normal horse serum in PBS for 1 h at room temperature. The cells were then incubated with the antineuron-specific enolase (NSE) antibody (1 : 100, cat. no. sc-7455; Santa Cruz Biotechnology, Shanghai, China) or the anti-β-III-tubulin antibody (1 : 200, cat. no. sc-31782; Santa Cruz Biotechnology) overnight at 4°C. After three washes with PBS, the cells were further incubated with the secondary antibody for 1 h at room temperature. After washing, the cells were exposed to avidin–biotin complex. 3,3′-diaminobenzidine served as the chromogen for immunocytochemistry. The nuclei were stained with 4′,6-diamidino-2-phenylindole. Images were captured under a fluorescent microscope for immunofluorescence.
Prediction of target gene and Dual-Luciferase assay
MiR-122-5p that may regulate Tbr1 gene expression was predicted by TargetScan (http://www.targetscan.org/) and miRanda (http://www.microrna.org/).
The nucleotide sequence of the 3′UTR of mouse Tbr1 mRNA was retrieved from GenBank (NM_009322). The predicted miR-122-5p target region was cloned by PCR and inserted into the Sac I and Xba I sites of the pmirGLO luciferase vector (cat. no. E1330; Promega Beijing, China). The oligonucleotides for PCR were 5′-CCGAGCTCTAGATTC-CAGTCTATGGGG-3′ (sense) and 5′-GCTCTAGACACAGACGATACAAACGAG-3′ (antisense). A mutated version (3′UTRmu) was constructed using the QuickChange site-directed mutagenesis kit (cat. no. 200519; Stratagene Hangzhou, China). HEK293 cells were transfected with 3′UTR, 3′UTRmu, or empty reporter by Lipofectamine 2000 (Invitrogen) and then cotransfected with control miRNA, miR-122-5p mimic (cat. no. C-310464-07-0020; Dharmacon, Lafayette, Colorado, USA) or inhibitor (cat. no. IH-310464-08-0020; Dharmacon). Luciferase activity was measured 24 h after transfection using the Dual-Glo luciferase assay system (cat. no. E1910; Promega).
Quantitative real-time PCR
Total RNAs, including miRNAs, were extracted and reverse transcribed using the miScript system (Qiagen, Redwood City, California, USA) in accordance with the manufacturer’s instructions. The real-time PCR was performed in the miScript system, which included an SYBR Green PCR kit (Qiagen). GAPDH and U6 were used as internal controls, respectively. The PCR primers are listed in Table 1 and the reverse primer of miRNA was universal.
Cell infection of miR-122-5p
The lentiviral vector containing miR-122-5p was obtained from Invitrogen. For the transduction with lentivirus, 1×106 bMSCs were plated and infected with miR-122-5p or the negative control vector at a multiplicity of infection of 50. The cells were placed for 2 h at 4°C and then transferred to a plate. After culture in 5% CO2 at 37°C for 48 h, green fluorescent protein (GFP) expressions were observed under a fluorescence microscope.
Western blot analysis
The cells were lysed in a lysis buffer and the lysates containing 15 μg of protein were electrophoresed on SDS-polyacrylamide gels and transferred to supported nitrocellulose membrane (Bio-Rad Beijing, China). Membranes were blocked for 1 h in 5% nonfat dry milk in 1×TBS with 0.1% Tween-20 (TBST), rinsed, and incubated with anti-Tbr1 (Santa Cruz Biotechnology), anti-NSE (cat. no. sc-7455; Santa Cruz Biotechnology), anti-TAU (cat. no. sc-5587; Santa Cruz Biotechnology), or anti-β-actin antibody (cat. no. sc-47778; Santa Cruz Biotechnology) overnight at 4°C. Blots were washed in TBST and incubated with HRP-conjugated secondary antibodies for 1 h. The blots were detected with enhanced chemiluminescence and protein bands were visualized.
Data are presented as mean±SEM. Statistical analysis was carried out using one-way analysis of variance with the post-hoc Dunnett test for comparisons. A significant difference was considered if P values were less than 0.05 or less than 0.01.
Characterization of bone mesenchymal stem cells
Flow cytometry was performed to analyze the surface antigen expression of bMSCs. We found that the cells lacked expressions of CD34 (0.08%) and CD45 (0.83%), and they positively expressed CD29 (97.35%), CD44 (100%), and CD105 (98.59%). CD29, CD44, and CD105 represented three of the major markers characteristic of MSCs, whereas CD34 and CD45 were two of the hematopoietic markers (Fig. 1).
Differentiation of bone mesenchymal stem cells into neuron-like cells
Here, we induced bMSCs to differentiate into neuron-like cells. BMSCs of third to fifth passages showed a short spindle-shaped morphology with a uniform distribution (Fig. 2a). After 14 days of neural induction the cells transformed into a neuron-like morphology and increasingly showed neuronal traits of a pyramidal appearance (Fig. 2b). The expressions of NSE (Fig. 2c) and β-III-tubulin (Fig. 2d–f) could be detected in neuron-like cells after 14 days of neural induction by immunocytochemistry and immunofluorescence staining. The mRNA expressions of the neuronal marker NSE, TAU, and microtubule-associated protein 2 (MAP2) were determined by real-time PCR and normalized to GAPDH. If the gene expression in bMSCs was defined as 1, after 7 and 14 days of neuronal induction, we detected whether there was a significant increase in neuronal marker expression levels. NSE or TAU expression increased rapidly after 7 days (1.52±0.36, P>0.05; 2.71±0.62, P<0.05) and increased to the maximum after 14 days (2.49±0.57, P<0.05; 4.38±0.83, P<0.01), but after 7 days, MAP2 expression increased to the maximum (4.67±0.74, P<0.01) and then decreased (3.25±0.56, P<0.05, Fig. 3).
Identification of miR-122-5p regulating Tbr1 expression
To determine the role of miR-122-5p targeting Tbr1 3′UTR, we predicted using miRanda and TargetScan. MiR-122-5p had one target site broadly conserved among vertebrates on Tbr1 mRNA 3′UTR (Table 2). MiRanda showed that the mirSVR score of miR-122-5p was −1.0637 and TargetScan showed that the seed match of miR-122-5p was 8-mer and the context plus score percentile was 95. These all indicated that miR-122-5p was well complementary with the target site of Tbr1 mRNA 3'UTR.
Dual-Luciferase assay was used to verify whether miR-122-5p may regulate Tbr1 expression. The pmirGLO vector itself included two luciferases: Firefly luciferase and Renilla reniformis luciferase. Renilla reniformis luciferase activity served as the internal control. Relative luciferase activity from the 3′UTR was suppressed by miR-122-5p mimic (0.42±0.04, P<0.01) and increased by the miR-122-5p inhibitor (1.21±0.07, P<0.05). Instead, the miR-122-5p mimic (0.95±0.06, P>0.05) or the miR-122-5p inhibitor (1.03±0.08, P>0.05) did not affect relative luciferase activity from the 3′UTRmu (Fig. 4a). MiR-122-5p mimic (0.69±0.06, P<0.05) decreased the Tbr1 mRNA level and the miR-122-5p inhibitor (1.35±0.07, P<0.05) increased it as shown by real-time PCR (Fig. 4b). Similar changes in Tbr1 protein level were showed by western blot: miR-122-5p mimic (0.41±0.03, P<0.05) and miR-122-5p inhibitor (1.28±0.04, P<0.05) (Fig. 4c). All of these indicated that miR-122-5p could inhibit Tbr1 expression by binding to its 3′UTR.
Quantitative analysis of Tbr1 and miR-122-5p expression
Real-time PCR was used to measure Tbr1 mRNA and miR-122-5p expression levels during the differentiation of bMSCs into neuron-like cells and normalized to GAPDH and U6, respectively. If the gene expression in bMSCs was defined as 1, Tbr1 mRNA expression level after 7 days was (5.43±0.79) fold higher than that in bMSCs and increased to (8.68±0.96) fold after 14 days, and then declined and still remained (3.67±0.53) fold after 21 days compared with that in bMSCs. In contrast, the expression of miR-122-5p after 7 days was (43.52±4.85%) lower than that in bMSCs and decreased to (24.69±3.01%) after 14 days, and sequentially decreased to (12.47±1.54%) after 21 days (Fig. 5).
The role of miR-122-5p overexpression in the differentiation of bMSCs
The expression of Tbr1 mRNA in bMSCs was much lower than that after neuronal induction and miR-122-5p was reversed as shown by real-time PCR; thus, the cells after 7 days of neural induction were selected for infection with lentivirus. The fluorescence was observed at 48 h after infection in the GFP-transfected group and the miR-122-5p-transfected group (Fig. 6a). The expression of miR-122-5p was almost five-fold higher in the miR-122-5p-transfected group than that in the nontransfected group, but there was no significant difference between the nontransfected group and the GFP-transfected group (Fig. 6b).
To explore the role of miR-122-5p overexpression in the differentiation of bMSCs, we compared the expression of neuronal markers using real-time PCR and western blot. If mRNA expression was arbitrarily defined as 100% in the nontransfected group, we found that Tbr1 mRNA expression in the miR-122-5p-transfected group was (51.92±3.13%) lower than that in the nontransfected group. Similarly, NSE mRNA expression decreased to (67.51±4.11%) and TAU mRNA expression decreased to (59.82±4.64%) (Fig. 6c). The protein levels of Tbr1, NSE, and TAU in the miR-122-5p-transfected group were significantly decreased compared with the nontransfected group as shown by western blot (Fig. 6d).
Here, we induced mouse bMSCs to differentiate into neuron-like cells with BHA, RA, bFGF, and NGF in vitro. Woodbury et al. have previously shown that bMSCs could be induced into neuron-like cells with β-mercaptoethanol, dimethyl sulfoxide, and BHA 18, but these antioxidants can lead to cytotoxic changes 19. Recently, Ye et al.20 induced bMSCs to differentiate into neuron-like cells with EGF and bFGF. In our study, after 14 days of neural induction, bMSCs transformed into a neuron-like morphology and increasingly showed neuronal traits with a pyramidal appearance. The expressions of the neuronal marker NSE and β-III-tubulin could be detected in neuron-like cells by immunocytochemistry and immunofluorescence staining. The mRNA expressions of NSE, TAU, and MAP2 were determined by real-time PCR after 7 and 14 days of neuronal induction. We found that there was a significant increase in neuronal marker expression levels after neuronal induction. However, the differentiation efficiency was not high and the neuron-like cells were immature.
To determine the role of miR-122-5p targeting Tbr1 3′UTR, we predicted using miRanda and TargetScan and found that miR-122-5p had one target site broadly conserved among vertebrates on Tbr1 mRNA 3′UTR. Both databases showed that miR-122-5p was well complementary with the target site of Tbr1 mRNA 3'UTR. Dual-Luciferase assay was used to verify whether miR-122-5p can regulate Tbr1 expression. The pmirGLO vector itself included two luciferases: Firefly luciferase to monitor mRNA regulation and Renilla reniformis luciferase for normalization of gene expression. The results showed that miR-122-5p could inhibit Tbr1 expression by binding to its 3′UTR and further identified by real-time PCR and western blot.
Before transfection of miR-122-5p, real-time PCR was used to measure Tbr1 mRNA and miR-122-5p expression levels during differentiation of bMSCs into neuron-like cells. If the gene expression in bMSCs was defined arbitrarily as 1, we found that after 7 days, Tbr1 mRNA expression increased rapidly and increased slowly to the maximum after 14 days, and then decreased and still the higher level after 21 days. Instead, the expression of miR-122-5p after 7 days was rapidly decreased and then decreased sequentially. These results showed that the expression of Tbr1 mRNA was almost reciprocal to that of miR-122-5p. To explore the role of miR-122-5p overexpression in the differentiation of bMSCs, we compared expressions of neuronal marker Tbr1, NSE, and TAU using real-time PCR and western blot. The cells after 7 days of neural induction were infected and the expression of miR-122-5p was significantly increased. Then, the mRNA and protein levels of Tbr1, NSE, and TAU decreased significantly. The results indicated that miR-122-5p may directly or indirectly regulate the expression of neuronal markers to control the differentiation of bMSCs.
In summary, we induced bMSCs to differentiate into neuron-like cells in vitro. TargetScan and miRanda showed that miR-122-5p was well complementary with the target site of Tbr1 mRNA3′UTR. Identified by the Dual-Luciferase assay, we found that miR-122-5p could inhibit Tbr1 expression by binding to its 3′UTR. Furthermore, the expressions of Tbr1 mRNA and protein were decreased by real-time PCR and western blot. Overexpression of miR-122-5p downregulated the expressions of Tbr1, NSE, and TAU. Therefore, miR-122-5p may directly or indirectly regulate the expression of neuronal markers to control the differentiation of bMSCs.
Conflicts of interest
Yue Guo is currently receiving a grant (no. 81601727) from the National Natural Science Foundation of China. For the remaining authors there are no conflicts of interest.
1. Wislet-Gendebien S, Hans G, Leprince P, Rigo JM, Moonen G, Rogister B. Plasticity of cultured mesenchymal stem cells: switch from nestin-positive to excitable neuron-like phenotype. Stem Cells 2005; 23:392–402.
2. Guan M, Xu Y, Wang W, Lin S. Differentiation into neurons of rat bone marrow derived mesenchymal stem cells. Eur Cytokine Netw 2014; 25:58–63.
3. Kim EY, Lee KB, Yu J, Lee JH, Kim KJ, Han KW, et al. Neuronal cell differentiation of mesenchymal stem cells originating from canine amniotic fluid. Hum Cell 2014; 27:51–58.
4. Wang K, Long Q, Jia C, Liu Y, Yi X, Yang H, et al. Over-expression of Mash1 improves the GABAergic differentiation of bone marrow mesenchymal stem cells in vitro. Brain Res Bull 2013; 99:84–94.
5. Srivastava R, Kumar M, Peineau S, Csaba Z, Mani S, Gressens P, et al. Conditional induction of Math1 specifies embryonic stem cells to cerebellar granule neuron lineage and promotes differentiation into mature granule neurons. Stem Cells 2013; 31:652–665.
6. Jeon SJ, Kim JW, Kim KC, Han SM, Go HS, Seo JE, et al. Translational regulation of NeuroD1 expression by FMRP: involvement in glutamatergic neuronal differentiation of cultured rat primary neural progenitor cells. Cell Mol Neurobiol 2014; 34:297–305.
7. Bulfone A, Smiga SM, Shimamura K, Peterson A, Puelles L, Rubenstein JL. T-brain-1: a homolog of Brachyury whose expression defines molecularly distinct domains within the cerebral cortex. Neuron 1995; 15:63–78.
8. Bulfone A, Wang F, Hevner R, Anderson S, Cutforth T, Chen S, et al. An olfactory sensory map develops in the absence of normal projection neurons or GABAergic interneurons. Neuron 1998; 2:1–20.
9. Bedogni F, Hodge RD, Elsen GE, Nelson BR, Daza RA, Beyer RP, et al. Tbr1 regulates regional and laminar identity of postmitotic neurons in developing neocortex. Proc Natl Acad Sci USA 2010; 107:13129–13134.
10. McKenna WL, Betancourt J, Larkin KA, Abrams B, Guo C, Rubenstein JL, et al. Tbr1 and Fezf2 regulate alternate orticofugal neuronal identities during ocortical development. J Neurosci 2011; 31:549–564.
11. Mu CZ, Wang T, Wang XM, Tian H, Liu Y. Identification of microRNAs regulating Hlxb9 gene expression during the induction of insulin-producing cells. Cell Biol Int 2016; 40:515–523.
12. Liu Q, He H, Zeng T, Huang Z, Fan T, Wu Q. Neural-specific expression of miR-344-3p during mouse
embryonic development. J Mol Histol 2014; 45:363–372.
13. Park JH, Ahn S, Kim S, Lee J, Nam JW, Shin C. Degradome sequencing reveals an endogenous microRNA
target in C. elegans. FEBS Lett 2013; 587:964–969.
14. Ning H, Huang YC, Banie L, Hung S, Lin G, Li LC, et al. MicroRNA
regulation of neuron-like differentiation of adipose tissue-derived stem cells. Differentiation 2009; 78:253–259.
15. Zhuang H, Zhang R, Zhang S, Shu Q, Zhang D, Xu G. Altered expression of microRNAs in the neuronal differentiation of human Wharton’s Jelly mesenchymal stem cells. Neurosci Lett 2015; 600:69–74.
16. Hisanaga E, Park KY, Yamada S, Hashimoto H, Takeuchi T, Mori M, et al. A simple method to induce differentiation of murine bone marrow mesenchymal cells to insulin-producing cells using conophylline and betacellulin-delta4. Endocr J 2008; 55:535–543.
17. Cardozo AJ, Gómez DE, Argibay PF. Neurogenic differentiation of human adipose-derived stem cells: relevance of different signaling molecules, transcription factors, and key marker genes. Gene 2012; 511:427–436.
18. Woodbury D, Schwarz EJ, Prockop DJ, Black IB. Adult rat and human bone marrow stromal cells differentiate into neurons. J Neurosci Res 2000; 61:364–370.
19. Neuhuber B, Gallo G, Howard L, Kostura L, Mackay A, Fischer I. Reevaluation of in vitro differentiation protocols for bone marrow stromal cells: disruption of actin cytoskeleton induces rapid morphological changes and mimics neuronal phenotype. J Neurosci Res 2004; 77:192–204.
20. Ye Y, Zeng YM, Wan MR, Lu XF. Induction of human bone marrow mesenchymal stem cells differentiation into neural-like cells using cerebrospinal fluid. Cell Biochem Biophys 2011; 59:179–184.