Integrated Analysis of the Fecal Metagenome and Metabolome in Bladder Cancer in a Chinese Population

Abstract

Bladder cancer (BLCA) is a common malignancy of the urinary system. The gut microbiome produces various metabolites that play functional roles in tumorigenesis and tumor progression. However, the integrative analysis of the gut microbiome and metabolome in BLCA has still been lacking. Thus, the aim of this study was to identify microbial and functional characteristics and metabolites in BLCA in a Chinese population. Metagenomics, targeted metabolomics, bioinformatics, and integrative analysis were used in fecal samples of BLCA patients and healthy individuals. We found gut microbiomes were significantly dysregulated in BLCA patients, including Bifidobacterium, Lactobacillus, Streptococcus, Blautia, and Eubacterium. We also found 11Z-eicosenoic acid, 3-methoxytyrosine, abrine, aniline-2-sulfonate, arachidic acid, conjugated linoleic acids, elaidic acid, glycylleucine, glycylproline, leucyl-glycine, linoelaidic acid, linoleic acid, nicotinamide hypoxanthine dinucleotide, oleic acid, petroselinic acid, and ricinoleic acid to be significantly decreased, while cholesterol sulfate was significantly increased in BLCA patients. Integration of metagenomics and metabolomics revealed interactions between gut microbiota and metabolites and the host. We identified the alterations of gut microbiomes and metabolites in BLCA in a Chinese population. Moreover, we preliminarily revealed the associations between specific gut microbiomes and metabolites. These findings determined potential causative links among gut dysbiosis, dysregulated metabolites, and BLCA.

Keywords: bladder cancer; gut microbiome; metabolomics; metagenomics; omics integration.

Figure 1

Gene number differences and correlation analysis (BLCA patients, EXP group; healthy individuals, CON group). (a) Box plot of gene number between groups. The horizontal coordinate stands for each group, and the vertical coordinate stands for the gene numbers. (b) Venn graph. Each circle represents a group. The number in overlapping circles represents the gene numbers shared between groups; the number in non-overlapping circles represents the number of genes unique to the group. (c) Sample correlation coefficient heat map. Different colors represent the level of Spearman correlation coefficient. The leftward bias of the ellipse indicates that the correlation coefficient is positive, and the rightward bias is negative.

Figure 2

Relative abundance analysis of gut microbiota (BLCA patients, EXP group; healthy individuals, CON group). (a) At the genus level, the top 10 microbiomes with maximum relative abundance in two groups are shown. (b) At the species level, the top 10 microbiomes with maximum relative abundance in the two groups are shown. The number in non-overlapping circles represents the number of genes unique to the group. (c) Heat map of the top 35 microbiomes of abundance ranking in the two groups at genus level. (d) Heat map of the top 35 microbiomes of abundance ranking in the two groups at the species level.

Figure 3

Dimension-reducing analysis based on species abundance (BLCA patients, EXP group; healthy individuals, CON group). Each point in the graph indicates a sample, and samples from the same group were represented using the same color. The distance between points indicated the degree of difference. (a) At the genus level, PCoA is shown. (b) At the species level, PCoA is shown. (c) At the genus level, NMDS is shown. Stress point = 0.136. (d) At the species level, NMDS is shown. Stress point = 0.127.

Figure 4

LEfSe and Metastat analysis of species differing between groups (BLCA patients, EXP group; healthy individuals, CON group). (a) Species with an LDA score greater than a set value (set to 4 by default) are shown. The length of the bar graph represents the effect size of the differential species (i.e., the LDA score). (b) Cluster heat map based on differential species. The clustering tree on the left is the species clustering tree, and the values corresponding to the middle heat map are the Z values obtained by normalizing the relative abundance of species in each row. (c) Metastat analysis of butyrate-producing bacterium SS3_4. * indicates q-value < 0.05. (d) A Metastat analysis of Clostridiales bacterium 36_14. * indicates q-value < 0.05.

Figure 5

Functional characterization of gut microbiota (BLCA patients, EXP group; healthy individuals, CON group). (a) Distributions of relative abundances of KEGG pathway categories in EXP and CON groups. (b) Distributions of relative abundances of eggNOG pathway categories in EXP and CON groups. (c) Distributions of relative abundances of CAZy pathway categories in EXP and CON groups.

Figure 6

Changes of gut-microbiome-associated metabolites in BLCA patients (BLCA patients, EXP group; healthy individuals, CON group). (a) Quantitative statistics of the identified metabolites in each chemical classification. Each bar in the figure stands for a different chemical classification attribution entry. (b) Volcano plot was used to visualize the differential analysis for all metabolites detected. The horizontal coordinate was the log2 value of the fold change (FC), and the vertical coordinate was the log10 value of the significant p-value. Significantly dysregulated metabolites: metabolites meeting FC > 1, p-value < 0.05, are shown in red; metabolites meeting FC < 1, p-value < 0.05 are shown in blue. Non-significantly dysregulated metabolites are shown in black. (c) The fold change of identified metabolites with statistical significances are shown. Upregulation of differential metabolites are shown in red. Downregulation of differential metabolites are shown in green. (d) Heatmap of correlation analysis for these significantly dysregulated metabolites. Red indicates positive correlation, blue indicates negative correlation, and white indicates non-significant correlation. The size of the dot is related to the significance of the correlation.

Figure 7

Integrative analysis of metagenomics and metabolomics (BLCA patients, EXP group; healthy individuals, CON group). (a) Redundancy analysis (RDA) at the species level. Dots of different colors or shapes indicate different samples. Arrows indicate metabolite information. A projection was made from the point to the arrow of the metabolite, and the distance of the projection point from the origin represents the relative influence of the metabolite on the distribution of the sample community. (b) Heat map of correlation coefficients between significantly different microbiomes and metabolites in two groups. Each row indicates a microbiome at the species level. Each column indicates a significantly different metabolite. The correlation coefficient R is expressed in color. R > 0 indicates positive correlation (red); R < 0 indicates negative correlation (blue). * indicates p-value < 0.05; ** indicates p-value < 0.01; *** indicates p-value < 0.001.