Mettl3 Regulates Osteogenic Differentiation and Alternative Splicing of Vegfa in Bone Marrow Mesenchymal Stem Cells

1N6-methyl-adenosine (m6A) is a methylation modification at the N6 position of adenosine in both coding and noncoding RNAs and has been identified as the most prevalent internal chemical modification of mRNAs in eukaryotes. Although its presence was first described in the 1970s, the function of m6A remained a mystery until recently. The biological function and significance of m6A modification have recently begun to enter the limelight as a result of an m6A RNA immunoprecipitation approach followed by high-throughput sequencing. m6A modifications are frequently found near stop codons, at the beginning of 3′ untranslated regions (3′UTRs) and within internal long exons. All m6A sites are found within sequences conforming to the consensus sequence RR m6A CH([G/A/U][G>A] m6A C[U>A>C]). The m6A modification occurs by a methyltransferase complex, including methyltransferase-like 3 (METTL3), methyltransferase-like 14 (METTL14) and Wilms’ tumor 1-associated protein (WTAP) in mammalian cells. The modification is removed by fat-mass and obesity-associated protein (FTO) or α-ketoglutarate-dependent dioxygenase alkB homolog 5 (ALKBH5). The addition of m6A affects different aspects of mRNA metabolism, such as RNA stability, translation, alternative polyadenylation and pre-mRNA splicing. Previous studies have demonstrated that m6A methylation has a profound effect on cell differentiation, embryonic development and stress responses.

2Mesenchymal stem cells (MSCs) are a heterogeneous population of stem cells that can be harvested from many different sources and differentiate into mesoderm-type lineages, including osteoblasts, chondrocytes, and adipocytes. MSCs have been widely used in stem cell transplantation, gene therapy, tissue engineering, and immunotherapy. Bone mesenchymal stem cells (BMSCs) are commonly isolated and characterized from bone marrow aspirates and have been extensively investigated as the most promising cellular source for bone regeneration for both research and clinical purposes. Numerous studies have demonstrated the beneficial effects of BMSCs on bone regeneration and repair, including stimulation of their differentiation into osteoblasts, stimulation of angiogenesis, immunomodulatory effects, antiapoptotic effects on osteogenic lineage cells, and recruitment of host MSCs/progenitor cells. BMSCs have high affinity to differentiate into osteogenic lineages, and their differentiation into osteoblast cells is a pivotal step during bone regeneration and repair.

3Bone reconstruction is a complex process of overlapping phases, including inflammation, repair, and remodeling, and involves many intracellular signaling pathways and growth factors. Multiple signaling pathways have been reported to be involved in the osteogenic differentiation of BMSCs, such as bone morphogenetic protein (BMP)-Smad pathways, Wnt/β-catenin pathways, and PI3K-Akt pathways. Osteogenesis and angiogenesis are highly coupled in the process of bone formation, and angiogenesis is a prerequisite step for bone tissue regeneration. In tissue-engineered bone, BMSCs not only directly differentiate into endothelial cells to participate in new blood vessel formation in the bone defect area but also secrete angiogenic factors, such as vascular endothelial growth factor (VEGF) and angiopoietin, to promote local angiogenesis. As one of the most potent inducers of angiogenesis, VEGF is particularly intriguing due to its dual role in bone and endothelial development and plays an important role in bone formation. Previous studies have shown that VEGF has been successfully used for enhanced bone regeneration due to its capacity to promote both osteogenic and angiogenic differentiation. The angiogenic activity of BMSCs has been suggested to contribute to their regenerative capability.

4While much evidence has demonstrated that the osteogenic differentiation of BMSCs is intimately associated with multiple genetic factors, such as signaling molecules and growth factors, the epigenetic regulatory mechanisms underlying cell differentiation have attracted increasing attention. Recent evidence has illustrated that epigenetic modifications, such as histone acetylation and DNA methylation, are involved in the cell differentiation process of BMSCs during bone regeneration. As another layer of epigenetic regulation (RNA epigenetics), m6A modification has recently been reported to play important roles in cell function and the differentiation of embryonic stem cells (ESCs) and various cancer cell types. However, little is known about the role of m6A modification in the cell differentiation of BMSCs. Whether m6A affects the angiogenic factor Vegf during the osteogenic differentiation process of BMSCs also remains to be determined. Therefore, the aims of this study were to investigate the effect of m6A methylation on the osteogenic differentiation of BMSCs and to determine its role in regulating Vegf.

5To confirm the type of the separated BMSCs, third-passage cells were identified by flow cytometry to recognize the surface markers. The results showed expression of CD29+, CD44+, CD90+, CD34−, and CD45− (Figure 1A,B), indicating that the BMSCs were of high purity. Moreover, the BMSCs were able to be differentiated into osteoblast-like cells and adipose-likes cell when cultured in differentiating induction medium, as determined by Alizarin Red S staining and Oil Red O staining, respectively (Figure 1C–E). These data confirmed the status of the isolated and cultured BMSCs.

6During the osteogenic differentiation of BMSCs, cells were cultured in osteogenic induction medium at the indicated time points, and alkaline phosphatase (ALP) activity and mineralized nodule formation were estimated after induction. As shown in Figure 2C, cells cultured with osteogenic induction medium displayed significantly higher ALP activity on Day 7 than on Day 0. Mineralized matrix deposition was higher after 21 days of osteogenic induction than after 0, 7, and 14 days (Figure 2A,B). The expression levels of several osteogenic markers were also upregulated, as shown by quantitative real time polymerase chain reaction (qRT-PCR) and Western blotting, further confirming the osteogenic differentiation of BMSCs (Figure 2D,E).

7To investigate the role of m6A modification in the osteogenic differentiation potential of BMSCs, the expression patterns of an m6A methyltransferase (Mettl3) and demethylases (Fto, Alkbh5) were measured during the osteogenic induction of BMSCs. Both the mRNA and protein levels of Mettl3 increased after 7 and 14 days of osteogenic induction (Figure 2F,G). Although Fto increased at the mRNA level after 7 and 14 days, its protein level showed no significant difference. Alkbh5 showed no significant difference at either the mRNA or protein level. Accordingly, the expression pattern of Mettl3 during osteogenic differentiation was consistent with those of the mineralization-related markers.

8To explore the function of Mettl3 in the osteogenic differentiation process of BMSCs, specific shRNAs were applied to knock down its expression in BMSCs. Quantification of lentiviral gene transfer efficiency in BMSCs was measured via the proportion of fluorocytes. A transfer efficiency up to 90% was achieved at 48 h after transfection (Figure 3A). The Mettl3 protein level exhibited an approximate 60% decrease in the shRNA group (#sh1) compared with the negative control group, suggesting that Mettl3 was effectively silenced in BMSCs (Figure 3B). Mettl3-sh1 was thus chosen for further experiments.

9To investigate the differentiation potential of BMSCs after Mettl3 knockdown, the expression levels of several osteogenic markers were measured. The results showed that Mettl3 knockdown reduced the mRNA levels of Alp and Ocn in BMSCs after osteogenic differentiation induction for 7 and 14 days (Figure 3C). The mRNA and protein expression of Runx2 and Osterix also decreased in Mettl3-shRNA cells (Figure 3D).

10To verify the effect of Mettl3 knockdown on the mineralization potential of BMSCs, ALP activity and the formation of mineralized nodules were then examined in Mettl3-shRNA cells. ALP activity decreased on Day 7, and the formation of mineralized nodules was reduced on Days 7, 14, and 21 after osteogenesis induction (Figure 3E,F).

11To further determine the effect of Mettl3 on the differentiation of BMSCs, Mettl3-shRNA and Mettl-shCtrl cells undergoing osteogenic induction for 14 days were subjected to RNA sequencing. The genes that were differentially expressed following Mettl3 knockdown with a Log2 ratio >2 or <−2 were revealed in the heat map (Figure 4A). Among these genes, 556 genes were upregulated and 641 genes were downregulated (Figure 4B). Furthermore, gene oncology (GO) enrichment and KEGG pathway analyses were utilized to cluster the biological processes and pathways that were differentially inhibited between shMettl3 and shCtrl cells. In the biological process analysis of genes downregulated by Mettl3 knockdown with a Log2 ratio <−2, the significant GO terms were related to ossification, cell adhesion, bone mineralization, and positive regulation of cytosolic calcium ion concentration (Figure 4C). From KEGG pathway analysis of the genes downregulated by Mettl3 knockdown, the top ten enriched pathways were shown in Figure 4D. Among the signaling pathways that are closely related to osteogenic differentiation, the phosphatidylinositol 3-kinase (PI3K)-Akt signaling pathway appeared to be one of the most enriched pathways.

12To confirm the inhibition of the PI3K-Akt pathway following Mettl3 knockdown in BMSCs, Western blotting was performed to evaluate the phosphorylation level of Akt. Compared to the control group, the group with Mettl3 knockdown exhibited a significant decrease in Akt phosphorylation (Figure 4E), indicating that PI3K-Akt signaling was suppressed by Mettl3 knockdown in BMSCs during the osteogenic differentiation process.

13The angiogenic factor VEGF is closely related to the osteogenic differentiation potential of BMSCs. The Vegfa expression in Mettl3-knockdown BMSCs was detected, and the results showed that Mettl3 knockdown decreased the expression level of Vegfa during the osteogenic differentiation of BMSCs (Figure 5B).

14Alternative splicing is the most common event in transcript isoforms. Mettl3 has been reported to play an important role in regulating alternative splicing of mRNA. Previous studies reported that the alternative splicing of Vegfa affects the osteogenic differentiation of BMSCs. Vegfa contains three homologous splice variants: Vegfa-120, Vegfa-164, and Vegfa-188. To analyze whether Vegfa and its alternative splicing are modulated by Mettl3 in BMSCs undergoing osteogenic differentiation, we performed RT-PCR to detect three Vegfa isoforms. The results indicated that Mettl3 knockdown decreased the mRNA expression of Vegfa-164 and Vegfa-188 but did not significantly alter Vegfa-120 mRNA level (Figure 5A,C–E).

15m6A is a prevalent epitranscriptomic modification in mRNAs and is widely conversed across single-cell organisms, plants and vertebrates. m6A methylation occurs during nascent pre-mRNA processing by a methyltransferase complex consisting of METTL3, METTL14, and WTAP and is removed by FTO and ALKBH5. This modification has been shown to regulate the pluripotency and differentiation of ESCs and somatic cell reprogramming. m6A methylation on some transcripts in ESCs, particularly those encoding developmental regulators, blocks HuR binding and maintains ESC self-renewal capability. Depletion of m6A in the mRNA of Mettl3−/− ESCs limits naïve marker priming and differentiation competence, which leads to a “hyper”-naïve pluripotency phenotype. Overexpression of Mettl3 in mouse embryonic fibroblasts (MEFs) increases m6A levels and significantly increases the reprogramming efficiency by enhancing the expression of key pluripotent factors.

16BMSCs are the most suitable and widely used progenitor cell population for bone regeneration due to their capacity for multilineage differentiation and the potential to increase osteoinduction and osteogenesis. BMSCs are known to stimulate bone tissue regeneration, but the mechanism by which the effect occurs has not yet been elucidated. To explore the role of m6A-dependent RNA methylation in the osteogenic differentiation of BMSCs, BMSCs were cultured using osteogenic induction medium to establish an osteogenesis model. The expression levels of main m6A methyltransferase and demethylases, including Mettl3, Fto and Alkbh5, were then evaluated. The results showed that Mettl3 mRNA and protein levels increased after osteogenic induction, but the protein expression of Fto and Alkbh5 levels did not significantly differ with or without osteogenic differentiation. These findings indicated that Mettl3-dependent RNA methylation might be involved in the osteogenic differentiation of BMSCs.

17METTL3 (also known as MTA70) is identified as the main methyltransferase critical for m6A methylation. Deletion or overexpression of METTL3 changes the total m6A methylation level, which has a direct effect on cell survival, stem cell maintenance, and lineage determination. Depletion of the Mettl3 homolog in Arabidopsis thaliana gives rise to plants with altered growth patterns and reduced apical dominance. The inhibition of the Drosophila Mettl3 homolog (Dm ime4) has been proven to inhibit oogenesis. METTL3 inhibition leads to a concomitant decrease in the cellular m6A level and apoptosis of human HeLa cells. To elucidate the role of Mettl3 in the osteogenic differentiation of BMSCs, Mettl3 was silenced in BMSCs, and the effect of Mettl3 knockdown on cell differentiation was investigated in the present study. The results showed that Mettl3 knockdown reduced the accumulation of osteogenic differentiation markers such as Runx2 and Osterix. The ALP activity and the formation of mineralized nodules were also decreased after Mettl3 knockdown. Differentially expressed genes in Mettl3 knockdown cells were subjected to RNA sequencing, and the results suggested that many genes were regulated by Mettl3 knockdown. GO term analysis indicated that downregulated genes were associated with ossification, cell adhesion, and bone mineralization. KEGG pathway analysis revealed that the signaling pathways associated with the downregulated genes, such as PI3K-Akt signaling pathway, AMPK signaling pathway, and PPAR signaling pathway, were mostly involved in protein digestion and absorption, regulation of lipolysis in adipocytes, and osteogenic differentiation. Among the signaling pathways that are closely relevant to the osteogenic differentiation, the PI3K-Akt signaling pathway appeared to be one of the most enriched pathways. The PI3K-Akt signaling pathway is known to play a regulatory role in the survival, proliferation, migration, and differentiation of rabbit and human MSCs. PI3K-Akt signaling also acts as a crucial regulator in bone tissue metabolism and the osteogenic differentiation of BMSCs. Our Western blotting results confirmed a significant reduction in Akt phosphorylation after Mettl3 knockdown, which was consistent with the RNA sequencing results. The above data suggested that Mettl3 increases the osteogenic differentiation of BMSCs.

18Reconstructing local microcirculation and increasing blood vessel density are essential changes for effective bone regeneration. VEGF is the most attractive and well-characterized factor for the therapeutic induction of new blood vessel growth. VEGF can be secreted by different MSC populations in a paracrine manner and influences endothelial cell behavior. This growth factor is not only involved in angiogenesis but is also implicated in the maturation of osteoblasts, ossification, and bone turnover. Therefore, VEGF is widely used to induce angiogenesis and osteogenesis in bone tissue engineering constructs. In rodents, Vegfa contains three homologous splice variants: 120, 164, and 188 amino acids. Vegfa-120, the shortest variant, is less effective in promoting MSC growth. Vegfa-164 plays a regulatory role in improving MSC proliferation and chondrogenic differentiation, and Vegfa-188 is more effective in promoting the osteogenic differentiation of MSCs. The present study indicated that Mettl3 inhibition decreased the expression level of Vegfa during the osteogenic differentiation of BMSCs. The low expression of Vegfa might inhibit the osteogenic differentiation of BMSCs in turn. Mettl3 localizes predominantly in nuclear speckles, which are sites of mRNA splicing and storage. Several studies have implied that Mettl3 plays an important role in regulating alternative splicing of mRNA. To clarify whether Mettl3 affects the alternative splicing pattern of Vegfa, the expression levels of Vegfa variants were determined using RT-PCR in Mettl3-depleted BMSCs. The results demonstrated that Mettl3 knockdown did not significantly alter Vegfa-120 mRNA level but significantly decreased the mRNA levels of Vegfa-164 and Vegfa-188, which are related to MSC proliferation and differentiation. These findings indicated that Mettl3 knockdown limited the expression of Vegfa and its bone formation-related splice variants in osteoblast-induced BMSCs.

19In conclusion, the present study estimated the expression pattern of m6A methyltransferase and demethylases during the osteogenic differentiation of BMSCs and demonstrated that Mettl3 is highly expressed in osteogenically differentiated BMSCs. Investigating the role of Mettl3 in regulating cell differentiation showed that the loss of Mettl3 suppressed the osteogenic differentiation potential of BMSCs. Mettl3 knockdown also decreased the expression of Vegfa and its splice variants in BMSCs. These findings might contribute to novel progress in the role of the epitranscriptome in osteogenic differentiation and provide a promising perspective for the development of innovative therapeutic strategies for bone regeneration.

20BMSCs were obtained from the femurs and tibias of 2–3-week-old Sprague-Dawley male rats (Animal Center of Sun Yat-sen University) as described previously by Liu et al.. BMSCs were seeded in 25-cm2 culture flasks and were maintained by Dulbecco’s Modified Eagle Medium: Nutrient Mixture F-12 (DMEM/F12, Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA) and 1% penicillin/streptomycin (Invitrogen, Carlsbad, NM, USA). The medium was renewed every 3 days. When reaching 90% confluence, BMSCs were harvested using trypsin/EDTA (Gibco, Grand Island, NY, USA) and reseeded into dishes (10 cm in diameter). The expression levels of different cell surface markers, including CD29+, CD44+, CD90+, CD34− and CD45−, were determined by flow cytometry. Cells from the third passage were used in further experiments.

21For osteogenic differentiation, BMSCs were cultured using osteogenic induction medium, which contained DMEM/F12 supplemented with 100 nmol/L dexamethasone, 50 g/L ascorbic acid and 10 nmol/L β-glycerophosphate (Sigma-Aldrich, St. Louis, MO, USA) for 7 and 14 days. Cells without osteogenic induction medium were used as the control.

22Short hairpin RNA (shRNA) against Mettl3 shRNA: AGTCACAAACCAGATGAAATA and a nonspecific shRNA construct were designed and cloned into a hU6-MCS-Ubiquitin-EGFP-IRES-puromycin vector. The recombinant lentiviral vector was transfected into 293FT cells. The supernatant was harvested after 48 h and transduced into BMSCs. The cells were then incubated in solution with 1 μg/mL puromycin (Sigma-Aldrich, St. Louis, MO, USA) for 2 weeks. The stable clones were then maintained in 0.5 μg/mL puromycin and were observed and counted with a fluorescence microscope to ensure that the transfection rate was over 90%, and the knockdown effect was confirmed using a Western blotting assay. The cells were independently divided into 3 groups: control (untransduced group), shCtrl (negative control group), and shMettl3.

23The BMSCs were harvested using RIPA lysis buffer (Cell Signaling Technology, Boston, MA, USA) containing protease inhibitor cocktail (Cwbiotech, Beijing, China). Forty micrograms of protein were electrophoresed in 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and then transferred onto polyvinylidene fluoride membranes (PVDF) (Millipore, Billerica, MA, USA) in transfer buffer with 10% methanol. The membrane was blocked in TBST containing 5% nonfat milk for 1 h at room temperature and then incubated with the following primary antibodies overnight at 4 °C: Mettl3 (1:1000; Proteintech, Chicago, IL, USA), Fto (1:1000; Abcam, Cambridge, UK), Alkbh5 (1:1000; Proteintech, Chicago, IL, USA), Runx2 (1:1000; Abcam, Cambridge, UK), Osterix (1:1000; Abcam, Cambridge, UK), Akt and P-Akt (1:1000; Cell Signaling Technology, Boston, MA, USA), Gapdh (1:1000; Abcam, Cambridge, UK), and Vinculin (1:1000; Cell Signaling Technology, Boston, MA, USA). The membrane was then incubated with secondary antibodies (Abcam, Cambridge, UK) for 1 h at room temperature after washing. The blot showing antibody binding was developed with an enhanced chemiluminescence system (Millipore, Billerica, MA, USA) and captured via an ImageQuant LAS 4000 mini system (GE Healthcare Life Sciences, Illinois, NJ, USA). Band densities were quantified using ImageJ v1.47 software (National Institutes of Health, Bethesda, MD, USA).

24To detect mineralized nodule formation, BMSCs seeded in 6-well plates were induced in osteogenic medium for 0, 7, and 14 days. After culturing, the cells were rinsed with PBS and fixed in 4% paraformaldehyde solution for 20 min, stained with 1% Alizarin Red S (GL Biochem, Shanghai, China) solution at room temperature for 10 min, and washed 5 times. Mineralized nodules were then photographed under an inverted phase contrast microscope (Axiovert 40; Zeiss, Jena, Germany).

25Cells seeded in 24-well plates were cultured in osteogenic differentiation medium for 7 days. Alkaline phosphatase activity was determined utilizing an ALP assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Cells were lysed with 1% Triton X-100 for 30 min at 4 °C. The optical density (OD) value of the solution was quantified by measuring the absorbance at 520 nm by a microplate reader (Tecan, Hombrechtikon, Switzerland). The protein content was examined by a bicinchoninic acid protein assay (Beyotime Biotechnology, Haimen, China), and ALPase levels were standardized to the total protein concentration.

26Total RNA from BMSCs was extracted using TRIzol reagent (Invitrogen, Carlsbad, NM, USA). RNA samples were reverse-transcribed for cDNA synthesis using Superscript III Reverse Transcriptase (Invitrogen, Carlsbad, NM, USA). Subsequently, cDNA was used in PCR. PCR was performed and monitored using the Light Cycler 480. Samples were analyzed using primers listed in Table 1.

27To determine the expression of Vegfa splice variants, semiquantitative RT-PCR followed by agarose gel electrophoresis was performed. The pair of primers generated a different sized product for each of the splicing forms of VEGF mRNA. The predicted PCR products for the major forms, Vegf-120, Vegf-164, and Vegf-188 were 431, 563, and 635 bp, respectively. One microgram of RNA was prepared and subjected to reverse transcription as described above. Reverse-transcribed cDNA was used for PCR using Taq DNA polymerase (Invitrogen, Carlsbad, NM, USA). PCR products were electrophoresed using 1.5% agarose gels, and images were captured using the FluorChem™ Q system (Alpha Innotech, San Jose, CA, USA).

28Approximately 20 ng of Poly (A) RNA were purified from total RNA and converted to cDNA. The cDNA samples were sequenced with a HiSeq 2000 system (Illumina Inc., San Diego, CA, USA). The expression of transcripts was quantified as reads per kilobase per million reads (RPKM). Genes were considered differentially expressed genes when the expression showed twofold changes using RPKM. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed with the Database for Annotation, Visualization and Integrated Discovery (DAVID: http://david.abcc.ncifcrf.gov/) to identify the significantly enriched pathways.

29All experiments were performed at least in triplicate. The data are shown as the mean ± standard deviation and were analyzed using the SPSS 20.0 software program (SPSS, Chicago, IL, USA). One-way analysis of variance was applied to compare the experimental groups. Differences were considered statistically significant differences for p < 0.05.