Characterizations of the multi-kingdom gut microbiota in Chinese patients with gouty arthritis

Abstract

Objective: The gut microbial composition has been linked to metabolic and autoimmune diseases, including arthritis. However, there is a dearth of knowledge on the gut bacteriome, mycobiome, and virome in patients with gouty arthritis (GA).

Methods: We conducted a comprehensive analysis of the multi-kingdom gut microbiome of 26 GA patients and 28 healthy controls, using whole-metagenome shotgun sequencing of their stool samples.

Results: Profound alterations were observed in the gut bacteriome, mycobiome, and virome of GA patients. We identified 1,117 differentially abundant bacterial species, 23 fungal species, and 4,115 viral operational taxonomic units (vOTUs). GA-enriched bacteria included Escherichia coli_D GENOME144544, Bifidobacterium infantis GENOME095938, Blautia_A wexlerae GENOME096067, and Klebsiella pneumoniae GENOME147598, while control-enriched bacteria comprised Faecalibacterium prausnitzii_G GENOME147678, Agathobacter rectalis GENOME143712, and Bacteroides_A plebeius_A GENOME239725. GA-enriched fungi included opportunistic pathogens like Cryptococcus neoformans GCA_011057565, Candida parapsilosis GCA_000182765, and Malassezia spp., while control-enriched fungi featured several Hortaea werneckii subclades and Aspergillus fumigatus GCA_000002655. GA-enriched vOTUs mainly attributed to Siphoviridae, Myoviridae, Podoviridae, and Microviridae, whereas control-enriched vOTUs spanned 13 families, including Siphoviridae, Myoviridae, Podoviridae, Quimbyviridae, Phycodnaviridae, and crAss-like. A co-abundance network revealed intricate interactions among these multi-kingdom signatures, signifying their collective influence on the disease. Furthermore, these microbial signatures demonstrated the potential to effectively discriminate between patients and controls, highlighting their diagnostic utility.

Conclusions: This study yields crucial insights into the characteristics of the GA microbiota that may inform future mechanistic and therapeutic investigations.

Keywords: Gouty arthritis; Gut bacteriome; Gut mycobiome; Gut virome; Microbiota dysbiosis; Multi-kingdom signatures; Whole-metagenome shotgun sequencing.

Fig. 1

Difference in the gut bacteriome between GA patients and healthy controls. A, Rarefaction curve analysis of the number of observed species in two groups. The number of species in different groups is calculated based on a randomly selected specific number of samples with 30 replacements, and the median and quartile values are plotted. B, C, Boxplot shows the distributions of Shannon diversity index (B) and the Simpson index (C) of gut bacteriome for two groups. D, PCoA analysis of Bray–Curtis distance based on the composition of gut bacteriome, revealing the separations between two groups. The location of samples (represented by nodes) in the first two principal coordinates is shown. Lines connect samples in the same group, and circles cover samples near the center of gravity for each group. E, Composition of gut bacteriome at the species level. F, Boxplot shows the representative differential gut bacterial species when compared between patient and control groups. G, H, Boxplot shows the Simpson index (G) and Shannon diversity index (H) of gut functional composition that significantly differs between patients and controls. I, PCoA analysis of Bray–Curtis distance based on the gut functional composition, revealing the separations between two groups. For boxplots, boxes represent the interquartile range between the first and third quartiles and median (internal line); whiskers denote the lowest and highest values within 1.5 times the range of the first and third quartiles, respectively; and nodes represent outliers beyond the whiskers. The significance level is calculated based on the Student’s t-test

Fig. 2

Difference in gut mycobiome between GA patients and healthy controls. A, Rarefaction curve analysis of the number of observed species in each group. The number of species in different groups is calculated based on a randomly selected specific number of samples with 30 replacements, and the median and quartile values are plotted. B, C, Boxplot shows the Shannon diversity index (B) and the Simpson index (C) of gut mycobiome that differ between two groups. D, PCoA analysis of Bray–Curtis distance based on the composition of gut mycobiome, revealing the separations between two groups. The location of samples (represented by nodes) in the first two principal coordinates is shown. Lines connect samples in the same group, and circles cover samples near the center of gravity for each group. E, Composition of gut mycobiome at the family level. F, Boxplot shows the GA-associated gut fungal species when compared between GA patients and healthy controls. For boxplots, boxes represent the interquartile range between the first and third quartiles and median (internal line); whiskers denote the lowest and highest values within 1.5 times the range of the first and third quartiles, respectively; and nodes represent outliers beyond the whiskers. The significance level is calculated based on the Student’s t-test

Fig. 3

Characteristics of the gut virus catalog and gut virome. A, Pie plot shows the proportions of complete, high-quality, medium-quality, and low-quality vOTUs in the non-redundance virus catalog. B, Venn plot shows the overlap of the current virus catalog and the other three public gut virus catalogs. C, Pie plot shows the family-level taxonomic annotation of the virus catalog. D, Rarefaction curve analysis of the number of observed vOTUs on each group of samples. The number of species in different groups is calculated based on a randomly selected specific number of samples with 30 replacements, and the median and quartile values are plotted. E, F, Boxplot shows the Shannon diversity index (E) and the Simpson index (F) of gut virome that differ among two groups. Boxes represent the interquartile range between the first and third quartiles and median (internal line); whiskers denote the lowest and highest values within 1.5 times the range of the first and third quartiles, respectively; and nodes represent outliers beyond the whiskers. The significance level is calculated based on the Student’s t-test. G, PCoA analysis of Bray–Curtis distance based on the composition of gut virome, revealing the separations between two groups. The location of samples (represented by nodes) in the first two principal coordinates is shown. Lines connect samples in the same group, and circles cover samples near the center of gravity for each group. H, Volcano plot shows the fold change vs. q-values for all vOTUs. The X-axis shows the ratio of vOTU abundance in GA patients compared with that in healthy controls. The Y-axis shows the q-value (-log10 transformed) of a vOTU. The vOTUs that enriched in GA patients and healthy controls are shown in red and blue points, respectively. I, J, Pie plots show the taxonomic distribution of GA-enriched (J) and control-enriched (K) vOTUs. K, The occurrence rate of the KOs differed in frequency between the GA-enriched and ocntrol-enriched vOTUs

Fig. 4

Interactions among gut bacteriome, mycobiome, and virome. A, The inter-omics effect sizes for the gut bacteriome, mycobiome, and virome. Numbers show the combined effect sizes between two datasets. B, Network showing the co-abundance correlations between gut bacteriome, mycobiome, and virome. All species and vOTUs are grouped at the family level. C, D, Barplots showing the number of top 20 gut bacterial species (C) and vOTUs (D) with the largest number of connections in the network. E, F, Pie plots showing the taxonomic distribution of bacterium-dependent (E) and bacterium-independent vOTUs (F)

Fig. 5

Classification of GA status by the compositions of gut multi-kingdom signatures. A, Receiver operating characteristic (ROC) analysis for classification of GA status using gut bacterial, fungal, and viral signatures. B-D, The 10 most important bacterial signatures (B), as well as the 20 most discriminant fungal (C) and viral signatures (D), in models aimed at classifying GA patients and healthy controls. The bar lengths indicate the importance of the variables and the label colors indicate the enriched trend of the microbial signatures (red: GA-enriched; green: control-enriched)