Genetically Predicted 3-Methoxytyrosine Mediates the Causal Association between Fibroblast Growth Factor 21 and Glioblastoma Multiforme

Abstract

Background: Glioblastoma multiforme (GBM) is one of the most common brain malignancies characterized by an inflammatory microenvironment and metabolic reprogramming. This study aims to investigate the causal relationship between inflammatory factors (IFs) and GBM, as well as the potential mediating effects of specific plasma metabolites. Methods: We used a bidirectional two-sample Mendelian randomization (MR) approach to investigate the causal associations between 91 IFs and GBM. We employed a two-step MR technique to identify significant mediators in this relationship, followed by a mediation analysis to explore and quantify the mediating effects of specific metabolites on the causal relationship between IFs and GBM. In vitro experiments were conducted to verify the effects of specific IF and metabolite on GBM cells. The response of cells to treatment was examined using a series of assays, including colony formation, cell proliferation, and migration assays. Results: Three IFs showed significant associations with GBM. Among them, fibroblast growth factor 21 (FGF21) had a protective effect against GBM [odds ratio (OR): 0.42; 95% confidence interval (CI): 0.25, 0.71; p=1.00×10^-3]. There was no strong evidence that genetically predicted GBM had an effect on FGF21 (OR: 1.04; 95% CI: 0.83, 1.31; p = 0.692). Mediation analysis identified 3-methoxytyrosine (3-MTyr) level (mediation effect of 11.50%) as a significant intermediary. The in vitro study demonstrated that FGF21 inhibited proliferation and migration in GBM cells, whereas 3-MTyr exerted the opposite effects. Conclusion: FGF21 was causally associated with a reduced risk of GBM, and this relationship is partially mediated by 3-MTyr. This identified regulatory network offers a novel avenue for further research into the pathogenic mechanisms of GBM and provides a theoretical foundation for the development of relevant therapeutic regimens.

Keywords: 3-methoxytyrosine; fibroblast growth factor 21; glioblastoma multiforme; mediation analysis; mendelian randomization.

Figure 1

(A) Diagrams illustrating the associations examined in this study are presented. The total effect (β) between IF and GBM was decomposed into two parts: (i) the indirect effect, which was calculated using a two-step approach (where β1 represents the effect of IF on metabolite, and β2 represents the effect of metabolite on GBM) and the coefficient difference method (β1×β2); and (ii) the direct effect (β-β1×β2). (B) A flow chart outlining the methodology applied in our study is provided. SNP selection criteria were applied for MR analysis to determine causal relationships and identify significant IF. Mediation analysis quantified the potential influence of 3-MTyr on the IF-GBM association. Abbreviations: GWAS, genome-wide association study; MR, mendelian randomization; IF, inflammatory factor; GBM, glioblastoma multiforme; SNP, single nucleotide polymorphism; IVW, inverse variance weighted; MR-PRESSO, MR-Pleiotropy Residual Sum and Outlier, 3-MTyr, 3-methoxytyrosine.

Figure 2

(A) Scatter plot for the causal association between FGF21 and GBM. (B) Forest plot of the causal effects of SNPs associated with FGF21 on GBM. Red lines represent estimations from the MR Egger and IVW test. (C) Leave-one-out plots for the causal association between FGF21 and GBM. Red lines represent estimations from the IVW test. Abbreviations: IVW, inverse variance weighted; MR, mendelian randomization; SNP, single nucleotide polymorphism; GBM, glioblastoma multiforme; FGF21, fibroblast growth factor 21.

Figure 3

Forest plot to visualize the causal effects of 3-MTyr with FGF21 and GBM. Three methods: IVW, MR Egger and weighted median. Abbreviations: FGF21, fibroblast growth factor 21; GBM, glioblastoma multiforme; 3-MTyr, 3-methoxytyrosine; IVW, inverse variance weighted; OR, odds ratio; CI, confidence interval.

Figure 4

(A) EdU proliferation assay of U118 and U251 cells. Cells were labeled with Apollo 567 and the nuclei were counterstained with DAPI. Scale bar = 200 μm. (B) Colony formation of U118 and U251 cells after treatment with FGF21 or 3-MTyr. (C) Transwell assay of U118 and U251 cells. Data are presented as the mean ± SD (n=3). *p < 0.05. Abbreviations: EdU, 5-Ethynyl-2´-deoxyuridine; FGF21, fibroblast growth factor 21; 3-MTyr, 3-methoxytyrosine.