Page 27 - Read Online

Ladeira et al. Microbiome Res Rep 2023;2:9 https://dx.doi.org/10.20517/mrr.2023.01 Page 9 of 15

Figure 4. Heatmap of Mash distances between the B. bifidum MAGs. The heatmap is annotated with the prevalence of the 15 OGs
related to phages. The gradient from blue to red indicates increasing Mash genetic distances between the MAGs. Hierarchical clustering
was performed by the Ward2 method. MAGs derived from subjects without Bifidobacterium partition information are shown in white.

We then studied the functions of Bifidobacterium associated with previously identified Bifidobacterium
partitions (health-associated ado_lon, ado_lon_bif, and lon_ado_cat or others (lon_pse, lon, ado_pse, and
no_Bif) by pairing MAG content to a Bifidobacterium partition for each subject. This analysis compared the
gene content of MAGs from specific Bifidobacterium species regardless of their differential abundance
between health groups. We found 38 OGs significantly associated with these two types of Bifidobacterium
partitions (chi-squared, FDR < 0.1). Fifteen of these OGs were less prevalent in health-associated
Bifidobacterium partitions (< or > 25% prevalence in health-associated and others, respectively)
[Supplementary Table 3]. These OGs were assigned to phages and included integrases, transposases, and
helicases. We then investigated the phylogenetic relationships between the MAGs harboring these 15 OGs
in B. bifidum. We computed Mash distances between 198 B. bifidum MAGs and visualized the prevalence of
15 significant OGs in the 198 B. bifidum MAGs. These 15 OGs had a higher prevalence in a cluster of MAGs
associated with the Bifidobacterium partitions most frequently detected in individuals with diseases
[Figure 4]. This finding suggests that a B. bifidum subspecies or strain may be enriched in phage-related
genes in subjects with more altered gut microbiomes.

DISCUSSION
In this study, we performed a large-scale analysis of the Bifidobacterium community in the human gut
microbiome, with data from a large adult population, including individuals with various health conditions.
By combining ecological and functional analyses of the Bifidobacterium community, we identified variable
associations of partitions of Bifidobacterium species and functions associated with gut microbiome features
and human health. Overall, our results confirm and extend previous findings on the ecological and
functional relevance of the Bifidobacterium community for the gut microbiome and human health.

The human gut microbiome varies significantly between subjects, and this variation may obscure the effect
of diet or treatment. Stratification of the gut microbiome on the basis of its composition has been used to