High-throughput library screening identifies IGF2BP3 as a core m6A regulator in LIHC.
A total of 28 m6A regulators including 10 writers, 3 erasers and 15 readers were finally identified based on TCGA data in this study. The process of m6A methylation was dynamically and reversibly regulated by these m6A regulators (Fig. 1a and Table-S1). The investigation of CNV alteration frequency revealed a prevalent CNV alteration in 28 regulators. VIRMA, HNRNPC, and METTL3 showed copy number deletions, whereas ZC3H13 and YTHDF2 had CNV amplification frequencies (Fig. 1b). Compared to normal tissues, m6A regulators with amplificated CNV demonstrated remarkedly higher expression in LIHC tissues, and vice versa (Fig. 1b and c). The comprehensive landscape of m6A regulator interactions, regulator connection and their prognostic significance for LIHC patients was depicted with the m6A regulator network. We found that not only exhibited significant correlations in expression within the same functional category but also among writers, erasers, and readers. (Fig. 1d and Table S). A survival analysis was performed for the prognostic values of 21 m6A regulators in patients with LIHC, the results showed that patients with high IGF2BP3 expression exhibited worse overall survival (Fig. 1e). Taken together, these data further shed light on the oncogenic role of IGF2BP3 in tumour progression.
IGF2BP3 is an independent prognostic factor in LIHC.
Based on the TCGA dataset, receiver operating characteristics (ROC) curve analysis of the promising predictive value for IGF2BP3 expression showed that the areas under the curves (AUCs) for 0.5-, 1-, and 1.5-year OS were 0.682, 0.669, and 0.619, respectively (Fig. 2a). The multi-index ROC analysis revealed that the AUC of IGF2BP3 was significantly better than those of other clinicopathological indicators (Fig. 2a) (such as age, gender, and stage). In addition, high IGF2BP3 expression was related with higher risk score and poorer OS status in LIHC patients (Fig. 2b). In univariate Cox analysis, stage (p < 0.01) and risk score (p < 0.001) were significantly correlated with OS (Fig. 2c). In multivariate Cox analysis, only IGF2BP3 (p < 0.001) was an independent prognostic factor (Fig. 2d). Together, these data illustrate IGF2BP3 is an independent prognostic factor in LIHC.
IGF2BP3 promotes tumor progression in LIHC.
To explore the biological function of IGF2BP3 in LIHC, we transfected IGF2BP3 knockdown or overexpression in HCCL-M3 and Hub-7 cells. Transfection efficiency was evaluated by western blot and qPCR (Fig. 3a-b and Figure S1). Colony formation assays and CCK-8 showed that IGF2BP3 played a vital role in cell proliferation and colony formation ability (Fig. 3c-d). In addition, IGF2BP3 also impacted cell migratory ability in HCCL-M3 and Hub-7 cells (Fig. 3e-g). Therefore, these datas illuminated that IGF2BP3 promotes tumor progression in LIHC.
Functional annotations of IGF2BP3 in LIHC.
The correlations between IGF2BP3 expression and clinical properties were examined in the TCGA dataset. High IGF2BP3 expression level was related with older age, gender, and tumor stage (Fig. 4a). Single-sample GSEA (ssGSEA) algorithm was used to evaluate the associations between IGF2BP3 expression and immune cell infiltration, the results showed that high-IGF2BP3 patients were slightly increased immune activity than the low-IGF2BP3 patients (Fig. 4b). Additionally, Gene set variation analysis (GSVA) was performed to explore the underlying molecular mechanisms differing in the high-IGF2BP3 and low-IGF2BP3 subgroups of LIHC patients. The results showed that high-IGF2BP3 patients were mainly related with mitotic spindle, G2M checkpoint, E2F targets, and cell cycle signaling pathway in the TCGA dataset (Fig. 4c-d).
Protein–protein interaction network and univariate cox regression analyses
To further investigate the potential biological behavior of IGF2BP3 modification pattern, we determined the top 200 DEGs between high-IGF2BP3 and low-IGF2BP3 expression groups using limma package. The PPI network of the interactions among 200 DEGs was performed using STRING database (confidence value > 0.8) (Fig. 5a), which were visualized in Cytoscape v3.8.2 (Fig. 5b). The top 30 genes were represented based on the number of nodes using bar plots, which may serve as hub nodes in the network (Fig. 5c and Table S2). In addition, univariate Cox regression analyses showed that 23 genes were of prognostic significance among the 200 DEGs (Fig. 5d and Table S3).
IGF2BP3 regulates MCM10 expression in an m6A-dependent manner.
Based on previous studies, 51 genes were identified as transcripts recognized and regulated by IGF2BP3 through m6A modification[17]. MCM10 was identified based on the intersection of the prognostic significance, the top 30 hub genes in the PPI network, and 51 genes of IGF2BP3 through m6A modification (Fig. 6a). Subsequently, the expression and potential role of MCM10 were further explored using the TCGA dataset. The expression of MCM10 was significantly increased in LIHC comparing normal tissue (Fig. 6b-c). K-M analysis showed that LIHC patients with MCM10 overexpression had poor survival (Fig. 6d). RIP and RT–qPCR was used to evaluate RNA enrichment, the results showed that the mRNA of MCM10 was enriched by the anti-IGF2BP3 antibody compared with IgG in the HCCL-M3 and Hub-7 cell lines, which confirmed the direct interaction between IGF2BP3 and MCM10 (Fig. 6f). In addition, the distinct m6A sites in MCM10 at single-base resolution were predicted using m6A site prediction tool SRAMP (26896799) (Fig. 6e). MeRIP-RT–qPCR was performed to investigate whether gene expression affected m6A modification, the results indicated MCM10 mRNA enrichment in the m6A-specific antibody precipitate (Fig. 6g). Moreover, IGF2BP3 overexpression markedly prolonged (Fig. 6h), while IGF2BP3 knockdown obviously shortened the half-life of MCM10 mRNA in LIHC cells (Fig. 6i); silencing mRNA stabilizer HuR dramatically diminished IGF2BP3-induced MCM10 upregulation (Fig. 6j), which demonstrated IGF2BP3 regulated MCM10 expression by modulating its mRNA stability. Taken together, these dates reveal that IGF2BP3 mediates the degradation of MCM10 mRNA by m6A modification.
IGF2BP3 promotes LIHC progression through MCM10 expression.
Since IGF2BP3 could regulate MCM10 expression, rescue experiments were carried out to verify the interaction between IGF2BP3 and MCM10 in LIHC progression. As expected, MCM10 could partially counteract the antitumor effects on cell viability (Fig. 7a), colony formation (Fig. 7b) and migration (Fig. 7c-d) mediated by shIGF2BP3. In addition, overexpression of MCM10 also promoted cell proliferation, colony formation and migration in HCCL-M3 and Hub-7 cells, confirming the oncogenic effects of MCM10 (Fig. 7a-d). Collectively, IGF2BP3 promotes LIHC progression through MCM10 expression.