Polysaccharides utilization in human gut bacterium Bacteroides thetaiotaomicron: comparative genomics reconstruction of metabolic and regulatory networks

Abstract

Background: Bacteroides thetaiotaomicron, a predominant member of the human gut microbiota, is characterized by its ability to utilize a wide variety of polysaccharides using the extensive saccharolytic machinery that is controlled by an expanded repertoire of transcription factors (TFs). The availability of genomic sequences for multiple Bacteroides species opens an opportunity for their comparative analysis to enable characterization of their metabolic and regulatory networks.

Results: A comparative genomics approach was applied for the reconstruction and functional annotation of the carbohydrate utilization regulatory networks in 11 Bacteroides genomes. Bioinformatics analysis of promoter regions revealed putative DNA-binding motifs and regulons for 31 orthologous TFs in the Bacteroides. Among the analyzed TFs there are 4 SusR-like regulators, 16 AraC-like hybrid two-component systems (HTCSs), and 11 regulators from other families. Novel DNA motifs of HTCSs and SusR-like regulators in the Bacteroides have the common structure of direct repeats with a long spacer between two conserved sites.

Conclusions: The inferred regulatory network in B. thetaiotaomicron contains 308 genes encoding polysaccharide and sugar catabolic enzymes, carbohydrate-binding and transport systems, and TFs. The analyzed TFs control pathways for utilization of host and dietary glycans to monosaccharides and their further interconversions to intermediates of the central metabolism. The reconstructed regulatory network allowed us to suggest and refine specific functional assignments for sugar catabolic enzymes and transporters, providing a substantial improvement to the existing metabolic models for B. thetaiotaomicron. The obtained collection of reconstructed TF regulons is available in the RegPrecise database (http://regprecise.lbl.gov).

Figure 1

Comparison of the motifs for the reconstructed regulons for SusR-like proteins. Maximum-likelihood phylogenetic tree for the predicted HTH domains of the SusR-like proteins in studied genomes and sequence logos for the predicted binding DNA-motifs. SusR-like proteins with reconstructed regulons are shown in bold.

Figure 2

Comparison of the motifs for the reconstructed regulons for HTCS regulatory systems. (A) Maximum-likelihood phylogenetic tree for the HTH domains of B. thetaiotaomicron HTCSs; HTCSs with reconstructed regulons are shown in bold (the tree for all HTCSs in the studied genomes is shown in Additional file 4: Figure S1). (B) Logos for the predicted DNA binding motifs. (C) Heat map of e-values for the predicted HTCS binding motifs.

Figure 3

Reconstructed metabolic and regulatory pathways in B. thetaiotaomicron. (A) pathways for utilization of dietary and host glycans; (B) pathways for utilization of monosaccharides in the cytoplasm. Regulators and proteins from the corresponding regulons are shown by matching background colors. Abbreviations for monosaccharides: Ara, L-arabinose; ddGlcA, 5-dehydro-4-deoxy-D-glucuronate; Fuc, L-fucose; Fru, D-fructose; Gal, D-galactose; GalA, galacturonate; Glc, D-glucose; GlcA, glucuronate; GlcNAc, N-acetyl-D-glucosamine; KDG, 2-keto-3-deoxy-D-gluconate; Man, D-mannose; Rha, L-rhamnose.

Figure 4

Sequence logos for the TFs regulating cytoplasmic monosaccharide utilization and for the global Crp-like regulator.

Figure 5

Comparison of the reconstructed metabolic network with other metabolic models of B. thetaiotaomicron. Top bar shows the number of novel genes in the reconstructed metabolic network in comparison with the previous genome-scale reconstruction in B. thetaiotaomicron[10] and with automatically generated metabolic reconstruction using the Model SEED tool [53], whereas the other two bars show numbers of genes that are shared between these reconstructions.

Figure 6

Bioinformatics workflows used for reconstruction of TF regulons. (A) Workflow 1 used for identification of DNA binding sites and regulon reconstruction for SusR-like regulators and HTCSs controlling polysaccharide utilization pathways. (B) Workflow 2 used for inference of binding site motifs and regulons for TFs controlling monosaccharide utilization pathways.