The comparative genomics of Bifidobacterium callitrichos reflects dietary carbohydrate utilization within the common marmoset gut

Abstract

Bifidobacterium is a diverse genus of anaerobic, saccharolytic bacteria that colonize many animals, notably humans and other mammals. The presence of these bacteria in the gastrointestinal tract represents a potential coevolution between the gut microbiome and its mammalian host mediated by diet. To study the relationship between bifidobacterial gut symbionts and host nutrition, we analyzed the genome of two bifidobacteria strains isolated from the feces of a common marmoset (Callithrix jacchus), a primate species studied for its ability to subsist on host-indigestible carbohydrates. Whole genome sequencing identified these isolates as unique strains of Bifidobacterium callitrichos. All three strains, including these isolates and the previously described type strain, contain genes that may enable utilization of marmoset dietary substrates. These include genes predicted to contribute to galactose, arabinose, and trehalose metabolic pathways. In addition, significant genomic differences between strains suggest that bifidobacteria possess distinct roles in carbohydrate metabolism within the same host. Thus, bifidobacteria utilize dietary components specific to their host, both humans and non-human primates alike. Comparative genomics suggests conservation of possible coevolutionary relationships within the primate clade.

Keywords: bifidobacteria; commensalism; comparative genomics; gut microbiota; non-human primate.

Fig. 1.

Relative taxon abundance in the marmoset gut. Shown are pie and bar chart representations of the relative abundance values at the (a) phylum and (b) genus level of microbial diversity in the marmoset gut using 16S rRNA gene sequencing. Each color represents a phylum (a) and the top 16 genera (b) identified (at >0.2 % abundance) in the marmoset gut.

Fig. 2.

Genomic diversity of B. callitrichos. Venn diagrams showing the number of genes shared and unique between (a) B. callitrichos JCM 17296^T, UMA51804, and UMA51805, and (b) genes shared between B. callitrichos strains (UMA51804, UMA51805) and closely related species B. aesculapii JCM 18761^T and B. stellenboschense JCM 17298^T. The B. callitrichos strains UMA51804 and UMA51805 shared a higher number of genes with B. callitrichos JCM 17296^T than with each other and other type strains included in the analysis.

Fig. 3.

Maximum-likelihood phylogenetic tree of members of the genus Bifidobacterium originating from non-human primates. The tree was created using the bcgTree pipeline with 107 essential single-copy core genes, found in a majority of bacteria, using hidden Markov models based on a partitioned maximum-likelihood analysis. Bootstrap confidence values were obtained with 1000 resamplings and are provided at branch points. The scale bar represents the number of amino acid substitutions per site. Microbial isolates identified in this study are shown as blue circles closely clustered with B. callitrichos JCM 17296^T (blue triangle). Alloscardovia macacae (UMA81212) was used as the outgroup species within the family Bifidobacterium.

Fig. 4.

Proposed mechanism for the utilization of N-acetylmuramic acid (NAM) and N-acetylglucosamine (NAG). NAM and NAG serve as the precursors for fructose 6-phosphate, which feeds into the bifid shunt pathway. The proposed mechanism is depicted to show the structure and EC number for the enzymes involved in the pathway. Text in red represents the locus tags for UMA51804, UMA51805, and JCM 17296^T, in that order.

Fig. 5.

Growth of B. callitrichos JCM 17296^T, UMA51804, and UMA51805 on various sole carbohydrate sources. Shown are growth profiles on acacia gum, arabinose, cellobiose, fructose, galactose, glucose, lactose, maltose, mannose, N-acetylglucosamine (NAG), rhamnose, sorbitol, tamarind gum, trehalose, cranberry xyloglucan, xylan, and xylose as a sole carbohydrate source. Bars represent the average final OD₆₀₀ of biological triplicates, and error bars show the standard deviation. Sole carbohydrate sources and OD₆₀₀ values are shown on the x- and y-axis, respectively. Significant differences among the growth profiles of strains on each carbohydrate source are computed using two-way ANOVA with significance at *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Fig. 6.

Growth of B. callitrichos JCM 17296^T, UMA51804, and UMA51805 on 2′-fucosyllactose and pooled HMOs. Shown are growth profiles on 2-fucosyllactose and pooled HMOs as sole carbohydrate sources. Curves represent the average OD₆₀₀ of biological triplicates with error bars showing the standard deviation. Growth times and OD₆₀₀ values are shown on the x- and y-axis, respectively.