Metagenomics and transcriptomics analysis of aspartame's impact on gut microbiota and glioblastoma progression in a mouse model

Abstract

Aspartame, a widely used artificial sweetener, has been extensively studied for its potential health effects. Emerging evidence suggests that aspartame intake may directly impact the composition and function of the intestinal microbiota, which could subsequently influence the risk, progression, and treatment of glioblastoma multiforme (GBM) within the tumor microenvironment. However, it remains unclear whether aspartame intake affects intestinal flora, gene expression, and epigenetic regulation during tumor progression. To address these gaps in knowledge, we conducted a comprehensive metagenomics and transcriptomics analysis of aspartame's impact on gut microbiota and glioblastoma progression in a mouse model. Using a well-established mouse model and a rigorous metagenomics and transcriptomics approach, our results demonstrated that although the aspartame diet did not significantly affect tumor growth, it induced changes in the composition of the gut microbiota, particularly a decrease in the relative abundance of the Rikenellaceae family. Additionally, key N6-methyladenosine (m⁶A)-regulated genes, such as cyclin-dependent kinase inhibitor 1A (CDKN1A), MYC (myelocytomatosis) oncogene, and transforming growth factor-β (TGFB1), were significantly upregulated in GBM tumors exposed to aspartame. Notably, the expression of TGFB1 (transforming growth factor-β) suggested a critical role in the progression of GBM mediated by aspartame-induced m⁶A modifications. Our integrative analysis offered novel perspectives on the intricate interplay between dietary aspartame intake, gut microbiota, and tumor biology.

Keywords: Aspartame; Glioblastoma; Metagenomics; N6-methyladenosine; RNA.

Fig. 1

(A) Experimental group (aspartame diet) versus the control group (normal diet). (B) Biophotonic imaging of glioma regions. (C) Families in the top ten rankings. (D) Heatmap illustrating the disparity in the relative abundance of each family within the microbial community. (E) The relative abundance of Rikenellaceae families was significantly lower in the aspartame diet group (P < 0.05).

Fig. 2

Differential analysis of m⁶A methylation and mRNA expression. (A) Volcano plot of differentially expressed genes between the aspartame diet group and the normal diet group. Genes significantly upregulated are shown in red, genes significantly downregulated in blue, and genes with no significant difference in expression are shown with gray dots. (B) Volcano plot showing differential m⁶A methylation of transcripts between the two groups. Transcripts with high methylation are shown in red, and those with low methylation are shown in blue. (C) Heatmap illustrating differences in gene expression between the two groups. (D) Heatmap demonstrating different mRNA methylation patterns between the two groups.

Fig. 3

Correlation analyses between differentially expressed mRNAs and differentially methylated m⁶A-mRNAs. (A) Downregulation-Hypermethylation; (B) Upregulation-Hypermethylation; (C) Downregulation-Hypomethylation; (D) Upregulation-Hypomethylation; (E) Nine-quadrant plot depicting the correlation between log2-fold changes (log2FC) in differentially expressed transcripts and log2FC in differential m⁶A methylation between the two groups.

Fig. 4

GO enrichment and pathway analysis. (A) GO enrichment of 6,146 differential genes between the two groups; (B) KEGG pathway analysis of 6,146 differential genes between the two groups; (C) KEGG pathway analysis of 1,314 differentially expressed genes in the first quadrant (transcripts upregulated in MeRIP-seq but downregulated in RNA-seq); (D) KEGG pathway analysis of 1,169 differentially expressed genes in the third quadrant (transcripts upregulated in both MeRIP-seq and RNA-seq); (E) KEGG pathway analysis of 1,504 differentially expressed genes in the seventh quadrant (transcripts downregulated in both MeRIP-seq and RNA-seq); (F) KEGG pathway analysis of 1,207 differentially expressed genes in the ninth quadrant (transcripts downregulated in MeRIP-seq but upregulated in RNA-seq) (G). GO enrichment of differentially expressed genes in the ninth quadrant.

Fig. 5

Analysis of m⁶A modification sites. (A) Overlapping m⁶A peaks between the two groups; (B) distribution of m⁶A peaks; (C) m⁶A motif; (D-F) profiles of m⁶A modification; (G-I) comparisons of Myc gene expression between the two groups (p < 0.001).

Fig. 6

A: Boxplot showing the expression of CDKN1A, MYC and TGFB1 . CDKN1A (The 95% confidence intervals for gene expression show little overlap (Tumor: 6.16–6.64, Normal: 4.28–4.72), indicating a significant difference. The effect size (Cohen’s d = 1.22) suggests higher expression in the tumor group), MYC, The confidence interval for the tumor group is (4.8, 5.6), while for the normal group, it is (2.3, 3.0). The intervals barely overlap, and Cohen’s d is 1.5, indicating significantly higher MYC expression in GBM. TGFB1 The confidence interval for the tumor group is (5.2, 5.9), while for the normal group, it is (2.6, 3.3). There is no overlap between the intervals, and Cohen’s d is around 1.6, suggesting significantly higher TGFB1 expression in GBM. B: Kaplan‒Meier curve showing survival rates based on the expression levels of CDKN1A, MYC, and TGFB1.