N6-methyladenosine reader YTHDF3 contributes to the aerobic glycolysis of osteosarcoma through stabilizing PGK1 stability

N6-methyladenosine (m6A) modification is a pivotal transcript chemical modification of eukaryotics, which has been identified to play critical roles on tumor metabolic reprogramming. However, the functions of m6A-reading protein YTH N6-methyladenosine RNA-binding protein 3 (YTHDF3) in osteosarcoma is still unclear. This research planned to investigate the bio-functions and mechanism in osteosarcoma tumorigenesis. The aerobic glycolysis of osteosarcoma cells were calculated by glucose uptake, lactate production analysis, ATP analysis and metabolic flux analysis for extracellular acidification rate (ECAR). Molecular binding was identified by RIP-qPCR, RNA decay analysis. Results indicated that YTHDF3 is upregulated in the osteosarcoma tissue samples and cells, and closely correlated to the poor prognosis of osteosarcoma patients. Functionally, gain and loss-of-functional assays illustrated that YTHDF3 promoted the proliferation and aerobic glycolysis of osteosarcoma cells in vitro, and accelerated the tumor growth in vivo. Mechanistically, a m6A-modified PGK1 mRNA functioned as the target of YTHDF3, and YTHDF3 enhanced the PGK1 mRNA stability via m6A-dependent manner. In conclusion, these findings indicated that YTHDF3 functioned as an oncogene in osteosarcoma tumorigenesis through m6A/PGK1 manner, providing a therapeutic strategy for human osteosarcoma.


Introduction
Osteosarcoma is considered as the most common primary bone malignance. In recent years, although the survival rates of osteosarcoma have been improved, the curative ratio of osteosarcoma is still low-rise due to the metastasis and rapid progression (Müngen et al. 2021;Sheng et al. 2021). Currently, the traditional treatments still have limited success in clinical treatment. Herein, the question that which therapeutic strategy is more suit for osteosarcoma is worthy of further discussion, which may reveal new directions for osteosarcoma clinical treatment (Zhang et al. 2021a).
N 6 -methyladenosine (m 6 A) acts as the most abundant and common mRNA's modification, which plays crucial roles and functions in series of biological processes (Dang et al. 2021;Fang et al. 2021). The effects of m 6 A on mRNA methylation and its biological functions in osteosarcoma are of great significance (Wu et al. 2022;Zhang et al. 2021b). For example, m 6 A writer WTAP (Wilms' tumor 1 associated protein) enhanced the stability of FOXD2-AS1 transcripts to promote the methylation modification and accelerate the osteosarcoma progression through FOXD2-AS1/m 6 A/ FOXM1 complex manner (Ren et al. 2022). METTL3 overexpression promotes the malignant proliferation of osteosarcoma cells, and upregulates the histone deacetylase-5 (HDAC-5) expression in by increasing the m 6 A level via the HDAC5/miR-142-5p/ARMC8 axis (Jiang et al. 2022). Overall, the data indicates that m 6 A regulator regulates the osteosarcoma progression in m 6 A-dependent manner.
The critical regulation of m 6 A modification in human cancer is still complicated and confusing. In the osteosarcoma tumorigenesis, m 6 A regulators exert diverse functions. For instance, upregulated METTL14 plays an oncogenic role in facilitating osteosarcoma progression via methylating MN1 mRNA in its coding sequence (CDS) regions by recognizing m 6 A reader insulin-like growth factor 2 mRNA binding protein 2 (IGF2BP2) . Moreover, in osteosarcoma, METTL3 upregulated in tissue and cells and increased the m 6 A level, which upregulated the histone deacetylase 5 (HDAC5) expression by reducing the enrichment of H3K9/K14ac on miR-142 promoter (Jiang et al., 2022). Thus, the functions of m 6 A on osteosarcoma are fascinating.
Given that the underlying mechanisms of YTHDF3 in human osteosarcoma are controversial, our present research tried to investigate its biological function and regulation pathway in osteosarcoma. Bio-information analyses and functional experiments were applied to reveal the potential glycolysis-associated target mRNA regulated by YTHDF3. In present research, our study found that YTHDF3 positively regulated the osteosarcoma aerobic glycolysis via binding PGK1.

Quantitative RT-PCR (qRT-PCR)
Firstly, the total RNA in cancer cells was extracted by Trizol reagent (Invitrogen). Then, reverse transcriptions were performed as previously described. In brief, the extracted RNA was utilized to synthesize cDNA with the instructions of PrimerScript RT Reagent Kit (TaKaRa, Kyoto, Japan). The real-time PCR was performed according to SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA) on the 7900 Real-time PCR System using SYBR Premix Ex Taq (Takara). RT-qPCR primers were purchased from RiboBio Biotechnology Co., Ltd (Guangzhou, China), and listed in Supplementary Table S1. GAPDH was used as endogenous controls.

Proliferation CCK-8 assays
Transfected osteosarcoma cells at logarithmic growth phase were seeded into a 96-well plate (2 × 103 cells in each well) and cultured in 5% CO 2 incubation at 37 ℃. After incubation for indicated time, CCK-8 reagent (10 μl) was added to each well and the absorbance was measured at 450 nm. Each well had three repetitions.

Colony formation assay
Transfected cells were seeded at 6-well plate in density of 1000 cells per well. After 14 days, cells were fixed by 4% paraformaldehyde 30 min and stained by 0.1% crystal violet (Beyotime) for 30 min. Then, plates were mildly washed by PBS and then stained. The photographs were taken using high-resolution camera. The experiments were performed in triplicate.

Glucose consumption, lactate production and ATP generation
The procedures in this research were done according to the instruction manual. Glucose consumption and lactate production were analyzed by ELISA kits (BioVision, Milpitas, CA, USA). Respectively, glucose uptake was detected by Glucose Uptake Colorimetric Assay Kit (BioVision, Cat. # K676). The lactate production was measured by Lactate Colorimetric Assay Kit (BioVision, Cat. #K627). ATP level was detected by ATP Determination Kit (Thermo Fisher Scientific, Cat. #A22066) according to the manufacturer protocol.

Metabolic flux analysis for extracellular acidification rate (ECAR)
ECAR in cells were examined on Seahorse XF-96 metabolic flux analyzer (Seahorse Bioscience, North Billerica, MA, USA) using Seahorse XF Glycolysis Stress Test Kit in accordance with the manufacturer's instructions. In brief, cells (3 × 10 4 cells per well) were seeded in the XF-96 cell culture microplates. Wells were added with glucose 10 mM, oligomycin 1 mM and 2-DG 80 mM. Finally, data was evaluated by Seahorse XF-96 Wave software.

Immunoprecipitation (RIP) PCR assay
The RIP assay for YTHDF3 and PGK1 mRNA was performed using Magna RIP RNA-Binding Protein Immunoprecipitation Kit (cat. #17-701, Millipore, Billerica MA, USA) in accordance with the manufacturer's instructions. Anti-YTHDF3 (Proteintech, Cat No: 25537-1-AP, 1:500) were applied for RIP detection. After washing by P buffer for elution, subject RNA in the precipitation was detected by qRT-PCR.

RNA decay analysis
Osteosarcoma cells (MG63, Saos-2) with stable transfection were incubated by actinomycin D (Cat. #HY17559, Med-Chem) for 0 h, 3 h or 6 h and then the RNA was extracted from cells. Then, as previously described, the half-life of PGK1 mRNA was analyzed by quantitative RT-PCR.

Methylated RNA immunoprecipitation (MeRIP)-qPCR
MeRIP-qPCR analysis was performed to examine the m 6 A modification on PGK1 mRNA using Magna MeRIP Kit (Millipore, Massachusetts, USA, cat. CR203146) according to the manufacturer's instructions. Osteosarcoma cells were harvested and washed using ice-cold PBS twice. Subsequently, after removing the supernatant, RNA was collected by centrifugation at 1500 rpm at 4 °C and then mixed with RIP lysis buffer (100 μL) and incubated with the lysate on ice for 5 min. Magnetic beads was incubated with m 6 A antibody (5 μg). After rotation at 4 °C overnight, the beads were washed by salt buffer, and extracted with RIP wash buffer. The enrichment was analyzed by qRT-PCR.

Mice xenografts
Twelve 6-week-old BALB/c male nude mice (Vitalriver, Beijing, China) were randomly divided to two groups. One group nude mice were subcutaneously injected with MG63 cells of YTHDF3 knockdown (sh-YTHDF3) or controls (sh-NC) at the right flank. The mice xenografts had been approved by the Laboratory Animal Welfare & Ethics Committee of Xi'an Jiao Tong University. All these animal procedures and handling care were performed in accordance with the Health guide for the care National Institutes and Laboratory animals.

Statistical analysis
The SPSS 18.0 software package and GraphPad Prism 7.0 (GraphPad Software, La Jolla, CA, USA) were utilized to perform the statistical analysis. Data are expressed as the mean ± SD from at least three independent experiments, and statistic was calculated using Students t tests or one-way analysis of variance (ANOVA). Value-p less than 0.05/0.01 was statistically considered as significance. The experiments were performed in triplicate.

YTHDF3 upregulated in the osteosarcoma tissue and cells
In the large cohort of human osteosarcoma, the expression level of YTHDF3 increased in the tumor samples as comparing to the normal samples (Fig. 1A). Besides, in osteosarcoma cells (Saos-2, MG63), YTHDF3 mRNA upregulated in the cells as compared with normal cells (Fig. 1B). As regarding to the YTHDF3 protein, western blot analysis showed that the protein levels increased in osteosarcoma cells comparing to normal cell (hFOB) (Fig. 1C). For the survival of osteosarcoma, the survival distributions were described by the Kaplan-Meier survival curve and the logrank test (http:// kmplot. com/ analy sis/). Survival data indicated that the patients with high YTHDF3 level illustrated a poor prognosis, and low YTHDF3 level was correlated to better prognosis (Fig. 1D). Taken together, YTHDF3 might 1 3 acted as an oncogene in osteosarcoma, and up-regulate in the osteosarcoma tissue and cells.

YTHDF3 promoted the progression of osteosarcoma cells
To test the function of YTHDF3 on osteosarcoma cells, the gain-of-function assays and loss-of-function assays were performed using osteosarcoma cell lines (Saos-2, MG63). The upregulated transfection efficiency and silencing efficiency were detected, suggesting the potential vectors for following cellular experiments ( Fig. 2A, B). Proliferation analysis by CCK-8 revealed that YTHDF3 overexpression promoted the proliferation of MG63 cells, while the silencing of YTHDF3 repressed the proliferation of Saos-2 cells (Fig. 2C). Proliferation analysis by colony formation assay suggested that YTHDF3 overexpression promoted the clones of MG63 cells, while the silencing of YTHDF3 repressed the clones of Saos-2 cells (Fig. 2D,E). Generation of mice xenografts assays illustrated that YTHDF3 overexpression accelerated the tumor weight (Fig. 2F)

Fig. 1 YTHDF3 upregulated in the osteosarcoma tissue and cells. A
In the large cohort of human osteosarcoma (http:// gepia. cancer-pku. cn/ index. html), data showed a high-expression level of YTHDF3 in the tumor samples as comparing to the normal samples. B RT-PCR was performed in osteosarcoma cell lines (Saos-2, MG63) and nor-mal osteoblastic cell line (hFOB) to detect the YTHDF3 mRNA levels. C Western blot analysis showed the protein levels of YTHDF3 in osteosarcoma cells comparing to normal cell (hFOB). D The survival distributions were described by the Kaplan-Meier survival curve and the log-rank test (http:// kmplot. com/ analy sis/). *p < 0.05. **p < 0.01

Fig. 2 YTHDF3 promoted the progression of osteosarcoma cells.
A RT-PCR and B western blot assays were performed to detect the transfection efficiency and silencing efficiency in osteosarcoma cell lines (Saos-2, MG63). C Proliferation analysis by CCK-8 was performed to test the proliferation of MG63 cells and Saos-2 cells. (D, E) Colony formation assay was performed to calculate the clones of MG63 cell and Saos-2 cells. F Tumor weight (F) and volume (G) in mice xenografts assays injected with osteosarcoma cell lines with YTHDF3 overexpression. *p < 0.05. **p < 0.01 and volume (Fig. 2G) of osteosarcoma cell lines. Taken together, YTHDF3 promoted the progression of osteosarcoma cells.

YTHDF3 promoted the aerobic glycolysis of osteosarcoma cells
As regarding to the aerobic glycolysis of osteosarcoma cells, series of glycolytic experiments were performed. Firstly, the glucose uptake analysis found that YTHDF3 overexpression promoted the glucose uptake level of MG63 cells, while the silencing of YTHDF3 inhibited the glucose uptake of Saos-2 cells (Fig. 3A). Then, lactate production analysis found that YTHDF3 overexpression promoted the lactate level of MG63 cells, while the silencing of YTHDF3 inhibited the lactate of Saos-2 cells (Fig. 3B). Moreover, ATP analysis revealed that YTHDF3 overexpression upregulated the ATP level of MG63 cells, while the silencing of YTHDF3 repressed the ATP of Saos-2 cells (Fig. 3C). Metabolic flux analysis for extracellular acidification rate (ECAR) analysis found that YTHDF3 overexpression potentiated the glycolysis rate of MG63 cells (Fig. 3D), while the silencing of YTHDF3 reduced the glycolysis rate of glycolysis rate of Saos-2 cells (Fig. 3E). Taken together, YTHDF3 promoted the aerobic glycolysis of osteosarcoma cells.

YTHDF3 interacted with PGK1 in a direct m 6 A binding manner
Given that YTHDF3 regulated the aerobic glycolysis of osteosarcoma cells, we aimed to investigate the potential mechanism by which YTHDF3 regulated the osteosarcoma cells' glycolysis. Online database (http:// gepia. cancer-pku. cn/ index. html) suggested that PGK1 was a high-expressed gene in osteosarcoma (Fig. 4A). It is a proven fact that PGK1 acts as an essential element in the tumor aerobic glycolysis Fig. 3 YTHDF3 promoted the aerobic glycolysis of osteosarcoma cells. A The glucose uptake analysis was performed to test the glucose uptake level of MG63/Saos-2 cells with YTHDF3 overexpression/silencing. B Lactate production analysis was performed to test the lactate level of MG63/Saos-2 cells with YTHDF3 overexpression/ silencing. C ATP analysis was performed to determine the ATP generation of MG63/Saos-2 cells with YTHDF3 overexpression/silencing. D, E Metabolic flux analysis for extracellular acidification rate (ECAR) analysis showed the glycolysis rate of MG63/Saos-2 cells with YTHDF3 overexpression/silencing. *p < 0.05. **p < 0.01 and functions as an oncogene (Fu and Yu, 2020;Liu et al. 2020;Zhang et al. 2020). Predictive analysis (http:// www. cuilab. cn/ sramp) inspired that there were potential possible m 6 A-modified sites in the PGK1 genome (Fig. 4B). Correlation analysis revealed that PGK1 positively correlated to the level of YTHDF3 in osteosarcoma cohort (Fig. 4C). In the genomic sequence of PGK1, the possible m 6 A-modified sites is AUG GAC U (Fig. 4D). Taken together, YTHDF3 interacted with PGK1 in a direct m 6 A binding manner.

m 6 A reader YTHDF3 enhanced PGK1 mRNA stability
To investigate the regulation of YTHDF3 on PGK1 mRNA, the RNA decay analysis was performed in osteosarcoma cells. Results indicated that YTHDF3 overexpression upregulated the PGK1 mRNA level upon Act D treatment (Fig. 5A), while the silencing of YTHDF3 repressed the PGK1 mRNA level upon Act D treatment (Fig. 5B). To detect the m 6 A level of PGK1 mRNA, MeRIP-PCR was performed and results showed that the m 6 A level of PGK1 mRNA (Fig. 5C). Besides, the interaction within YTHDF3 and PGK1 was detected by RIP assay, and results indicated that YTHDF3 significantly interacted with PGK1 mRNA in osteosarcoma cells (Fig. 5D). Then, the western blot assay indicated that YTHDF3 overexpression upregulated the PGK1 protein and silencing of YTHDF3 repressed the PGK1 protein (Fig. 5E). Taken together, these data suggested that m 6 A reader YTHDF3 enhanced PGK1 mRNA stability. Discussion N 6 -methyladenosine (m 6 A) is the most common and abundant mRNA modification in eukaryocytes, playing crucial roles in many biological processes (Zheng et al. 2021). It is a proven fact that m 6 A mRNA methylation could affect the energy metabolism of cancer cells (An and Duan, 2022;Liu and Jia, 2014;Zhao et al. 2014). The function of m 6 A with its underlying mechanisms for human osteosarcoma remains obscure. As a key m 6 A reader, the roles of YTHDF3 on osteosarcoma aerobic glycolysis are still elusive.
It is a proven fact that m 6 A modification significantly modulates the tumorigenesis via epigenetic regulation, including osteosarcoma (Hu and Zhao, 2021;Zhang et al. 2021c). For instance, ALKBH5 is upregulated in RNA decay analysis was performed in osteosarcoma cells (MG63, Saos-2) treated with Act D. C MeRIP-PCR was performed using anti-m6A antibody to detect the m6A level of PGK1 mRNA. D RIP following PCR showed the PGK1 RNA expression precipitated by anti-YTHDF3 and anti-IgG in osteosarcoma cells (MG63, Saos-2). E The western blot assay indicated the PGK1 protein in osteosarcoma cells (MG63, Saos-2) transfected with YTHDF3 overexpression (YTHDF3) and YTHDF3 silencing (sh-YTHDF3). *p < 0.05. **p < 0.01 osteosarcoma cells and ALKBH5-mediated m 6 A modification decreasing inactivates the STAT3 pathway by increasing SOCS3 expression via m 6 A-YTHDF2-dependent manner (Yang et al. 2022). METTL3 is highly expressed in osteosarcoma and METTL3 overexpression promotes the malignant proliferation of osteosarcoma cells, besides, METTL3 upregulates the expression of histone deacetylase 5 (HDAC5) to reduce H3K9/K14ac/miR-142/ARMC8 (Jiang et al. 2022). Therefore, the data indicates the critical function of m 6 A in osteosarcoma.
Here, present research revealed that m 6 A reader YTHDF3 significantly upregulated in the osteosarcoma tissue and cells. Functionally, YTHDF3 promoted the proliferation and aerobic glycolysis of osteosarcoma cells. Mechanistically, a m 6 A-modified PGK1 mRNA functioned as the target of YTHDF3, and YTHDF3 enhanced the PGK1 mRNA stability via m 6 A-dependent manner. As a m 6 A reader, YTHDF3 might interact with other regulators to exert their roles on transcripts metabolism. For example, in hematopoietic stem cells, Ythdf3 and Mettl3 regulates HSCs by transmitting the m 6 A RNA methylation of Ccnd1 mRNA on the 5ʹ-UTR (Zhang et al. 2022). Thus, this finding suggested the oncogenic role of YTHDF3 in osteosarcoma.
Numerous signal pathways mediated by m 6 A are associated with human tumor prognosis. Besides, m 6 A can modulates the expression of transcripts to mediate the target in signaling pathway, leading to methylation change by different m 6 A-correlated enzymes (Wu et al. 2022). YTHDF3 is a canonical m 6 A reader in the tumorigenesis. In ocular melanoma stem-like cells, YTHDF3 is highly expressed and related to poor clinical prognosis. Moreover, YTHDF3 knockdown exhibits the inhibitory tumor proliferation and migration abilities through targeting CTNNB1 (Xu et al. 2022). In triple-negative breast cancer, the YTHDF3 is correlated with poorer disease-free survival and overall survival, and YTHDF3 positively promotes the cell migration, invasion of TNBC cells by enhancing ZEB1 mRNA stability (Lin et al. 2022). Combined with the above literatures, data revealed that YTHDF3 could promote the tumor progression.
As a member of m 6 A catalyzing enzymes, YTHDF3 could recognize the m 6 A site on target mRNA installed by m 6 A methyltransferase. Consistent with our analysis, we found that YTHDF3 functioned as an oncogene in osteosarcoma tumorigenesis through m 6 A/PGK1 manner (Fig. 6), providing a therapeutic strategy for human osteosarcoma.
Funding The authors have no any funding.

Data availability
No research data shared.

Declarations
Conflict of interest All authors declare no conflicts of interest.