Current studies on tumor progression focus on the roles of cytokines in the tumor microenvironment (TME), and recent research shows that transforming growth factor-β1 (TGF-β1) released from TME plays a pivotal role in tumor development and malignant transformation. The alteration in cellular metabolism is a hallmark of cancer, which not only provides cancer cells with ATP for fuel cellular reactions, but also generates metabolic intermediates for the synthesis of essential cellular ingredients, to support cell proliferation, migration, and invasion. Interestingly, we found a distinct metabolic change during TGF-β1-induced epithelial-mesenchymal transition (EMT) in glioblastoma cells. Indeed, TGF-β1 participates in metabolic reprogramming, and the molecular basis is still not well understood. NADPH oxidases 4 (NOX4), a member of the Nox family, also plays a key role in the biological effects of glioblastoma. However, the relationship between NOX4, TGF-β1, and cellular metabolic changes during EMT in glioblastoma remains obscure. Here, our findings demonstrated that TGF-β1 upregulated NOX4 expression accompanied by reactive oxygen species (ROS) through Smad-dependent signaling and then induced hypoxia-inducible factor 1α (HIF-1α) overexpression and nuclear accumulation resulting in metabolic reprogramming and promoting EMT. Besides, inhibition of glycolysis reversed EMT suggesting a causal relationship between TGF-β1-induced metabolic changes and tumorigenesis. Moreover, TGF-β1-induced metabolic reprogramming and EMT which modulated by NOX4/ROS were blocked when the phosphoinositide3-kinase (PI3K)/AKT/HIF-1α signaling pathways were inhibited. In conclusion, these suggest that NOX4/ROS induction by TGF-β1 can be one of the main mechanisms mediating the metabolic reprogramming during EMT of glioblastoma cells and provide promising strategies for cancer therapy.

Background Increased expression of the transcription factor Forkhead box M1 (FOXM1) has been reported to play an important role in the progression and development of multiple tumors, but the molecular mechanisms that regulate FOXM1 expression remain unknown, and the role of FOXM1 in aerobic glycolysis is still not clear. Methods The expression of FOXM1 and NADPH oxidase 4 (NOX4) in normal brain tissues and glioma was detected in data from the TCGA database and in our specimens. The effect of NOX4 on the expression of FOXM1 was determined by Western blot, qPCR, reactive oxygen species (ROS) production assays, and luciferase assays. The functions of NOX4 and FOXM1 in aerobic glycolysis in glioblastoma cells were determined by a series of experiments, such as Western blot, extracellular acidification rate (ECAR), lactate production, and intracellular ATP level assays. A xenograft mouse model was established to test our findings in vivo. Results The expression of FOXM1 and NOX4 was increased in glioma specimens compared with normal brain tissues and correlated with poor clinical outcomes. Aberrant mitochondrial reactive oxygen species (ROS) generation of NOX4 induced FOXM1 expression. Mechanistic studies demonstrated that NOX4-derived MitoROS exert their regulatory role on FOXM1 by mediating hypoxia-inducible factor 1α (HIF-1α) stabilization. Further research showed that NOX4-derived MitoROS-induced HIF-1α directly activates the transcription of FOXM1 and results in increased FOXM1 expression. Overexpression of NOX4 or FOXM1 promoted aerobic glycolysis, whereas knockdown of NOX4 or FOXM1 significantly suppressed aerobic glycolysis, in glioblastoma cells. NOX4-induced aerobic glycolysis was dependent on elevated FOXM1 expression, as FOXM1 knockdown abolished NOX4-induced aerobic glycolysis in glioblastoma cells both in vitro and in vivo. Conclusion Increased expression of FOXM1 induced by NOX4-derived MitoROS plays a pivotal role in aerobic glycolysis, and our findings suggest that inhibition of NOX4-FOXM1 signaling may present a potential therapeutic target for glioblastoma treatment.

Background: Low-grade glioma (LGG) is a common malignant tumor of the central nervous system.The clinical prognosis of different patients varies greatly, so exploring appropriate markers that affect the prognosis and treatment of LGG is important. The purpose of this study was to identify the potential effect of autophagy-related DNA methylation on the prognosis and immune microenvironment in LGG. Methods:The methylation profile, transcription data and corresponding clinical information of 451 patients with LGG were obtained from The Cancer Genome Atlas (TCGA). Another methylation data and clinical information of 110 patients with LGG from Chinese Glioma Genome Atlas (CGGA) were used as the validation set. Through univariate and multivariate COX regression analysis, we identified the autophagy-related genes (ARGs) associated with methylation levels and prognosis, and established a risk assessment signature. The receiver operating characteristic (ROC) and Kaplan-Meier (KM) survival curve were used to verify the model's effectiveness in predicting prognosis. Patients were divided into low-and high-risk groups based on risk scores. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis were used to explore the differences in biological functions between the two groups. ESTIMATE and CIBERSORT algorithms were used to explore differences in immune infiltration and immunotherapy sites. Pearson correlation analysis was used to analyze the relative relationship between methylated cg sites and corresponding genes.Results: A total of 6 ARGs (ARSB, CFLAR, WIPI2, RB1, ERN1, RAB24) were selected that were associated with methylation levels and prognosis. The area under the curve (AUC) =0.96, and the KM survival curve P<0.0001, which proves that the risk assessment model has a good effect in predicting the prognosis ofLGG. GO and KEGG enrichment analysis showed that the model mainly involved major histocompatibility complex (MHC) II receptors, antigen processing and presentation, and immune cell differentiation. In addition, we also found differences in immune infiltration and immune checkpoints between high-and lowrisk groups. Conclusions:The methylation levels of these 6 ARGs have a strong predictive potential for LGG, and the methylation regulation of ARGs has an important impact on the immune microenvironment of LGGs.