The genome of th17 cell-inducing segmented filamentous bacteria reveals extensive auxotrophy and adaptations to the intestinal environment - PubMed (original) (raw)

The genome of th17 cell-inducing segmented filamentous bacteria reveals extensive auxotrophy and adaptations to the intestinal environment

Andrew Sczesnak et al. Cell Host Microbe. 2011.

Abstract

Perturbations of the composition of the symbiotic intestinal microbiota can have profound consequences for host metabolism and immunity. In mice, segmented filamentous bacteria (SFB) direct the accumulation of potentially proinflammatory Th17 cells in the intestinal lamina propria. We present the genome sequence of SFB isolated from monocolonized mice, which classifies SFB phylogenetically as a unique member of Clostridiales with a highly reduced genome. Annotation analysis demonstrates that SFB depend on their environment for amino acids and essential nutrients and may utilize host and dietary glycans for carbon, nitrogen, and energy. Comparative analyses reveal that SFB are functionally related to members of the genus Clostridium and several pathogenic or commensal "minimal" genera, including Finegoldia, Mycoplasma, Borrelia, and Phytoplasma. However, SFB are functionally distinct from all 1200 examined genomes, indicating a gene complement representing biology relatively unique to their role as a gut commensal closely tied to host metabolism and immunity.

Copyright © 2011 Elsevier Inc. All rights reserved.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Circular representation of the SFB genome

Wheel

: The 5 contigs were arranged in order in a circular pseudochromosome (see Methods). Circles from outside in are: (i) ORF homology. Each ORF was color coded according to the genera most prevalent in its top ten PSI-BLAST hits (see left bottom corner for color legend). The SFB genome is dominated by ORFs homologous to Clostridium spp.; (ii) KEGG BRITE functional categories of annotated coding sequences (CDS) colored by type of category (listed on the left side of the figure); (iii) CDS on the forward strand; (iv) CDS on the reverse strand; (v) non-coding RNAs; (vi) markers delineating end of contigs; (vii) G-C content, and (viii) G-C skew, defined as (G − C) / (G + C).

Top Left:

KEGG BRITE categories in SFB and the 30 available Clostridium spp. genomes. Percentage of genes in each BRITE category. Top column in each category - percentage in the SFB genome, bottom column - average percentage in Clostridium spp. Arrows indicate the two categories that differ in SFB.

Figure 2

Figure 2. Genome-wide metabolic comparison between SFB and all sequenced genomes

Analysis of microbial functional similarities based on shared orthologous gene families (A,B) and modules (C). A/B. The 1,209 genomes in KEGG and the 13,118 KEGG Orthology gene families (KOs) are reported as circles and small cyan triangles, respectively. Organisms are connected by edges to all gene families contained within their genome. A) Global network of all genomes for visual overview. SFB (large red circle) lies outside any cluster but is close to groups of several Firmicutes genera and in particular Clostridium, Thermoanaerobacter, Staphylococcus, and Streptococcus. Other genera, including Mycoplasma, Borrelia, Treponema, Finegoldia and Gardnerella are quantitatively similar to SFB (see text) but located in the network periphery due to overall reduced gene content. Despite their similarity to SFB in terms of genome size and host environment, Helicobacter and Campylobacter are located in different regions of the network suggesting different functional specialization. B) Sub-network of SFB and the 20 most similar organisms (Tversky index 0.75, Table 2); KOs included in at least two organisms are depicted. SFB is functionally distinct from both the cluster of Clostridia (further differentiated as C. botulinum, C. perfringens, and “other”) and the Thermoanaerobacteria. C) Metabolic comparison based on functional genomic potential (Table 2) highlights SFB’s similarity to several Myco/Phyto/Acholeplasma and to Finegoldia magna (for which only one genome is available). The width and color (white = low, green = medium, red = high) of edges reflect relationship strength. Again, despite considering only the most functionally similar organisms, SFB is not directly included in the clusters they form.

Figure 3

Figure 3. KEGG modules over- and under-represented in the SFB genome compared to its 20 most similar organisms

Overview of the KEGG functional hierarchy; highlighted leaf nodes represent metabolic modules over- (green) or under-enriched (red) in SFB relative to the 20 most functionally similar organisms. Black circles represent modules present at a similar level in SFB and these organisms (absolute z-score <1) and white circles are absent in both. Similar organisms were computed using A) Tversky index α = 0.75 (representing larger related genomes) and B) Tversky index α = 0.25 (smaller, more minimal related organisms) and are listed in Table 2. A complete list of KEGG modules differentiating SFB from functionally similar organisms can be found in Table S4A and S4B.

Figure 4

Figure 4. Predicted SFB metabolic pathways

Overview of SFB metabolic pathways. SFB are highly auxotrophic and have a few complete essential pathways mostly for utilization of glycans and monosaccharides. They have complete glycolysis and pentose phosphate pathways, but lack the TCA cycle. Although fatty acid biosynthesis pathways are present, fatty acid metabolism pathways are absent. Absent as well are most pathways for co-factor and amino acid biosynthesis with the exception of the interrelated pathways for lysine, aspartate, glutamate, asparagine, and glutamine as noted in the figure. In contrast, multiple oligosaccharide and metal ion (in particular iron) transport and utilization mechanisms are present in SFB, including PTS, ABC, and other transporters, as well as extracellular peptidases and glycosyl hydrolases, which are shown interacting with extracellular glycans. SFB appear to be able to digest the glycoprotein components of the mucus layer and import sugars released in the process via ABC transporters and TCS. Once imported, several enzymes prepare these substrates for glycolysis. All these pathways provide SFB with the ability to acquire multiple metabolites from the surrounding environment and the host. Within the cell, essential pathways leading from import of polysaccharides through production of peptidoglycan, fatty acids, reduced ferredoxin, and acetate are shown.

Figure 5

Figure 5. Presence of genomes of intestinal microorganisms in the MetaHIT human metagenome database

The WGS Illumina reads of 124 individual fecal samples in the MetaHIT database (Qin et al., 2010) were aligned to the SFB genome and six other reference genomes (Clostridium perfringens ATCC13124, Enterococcus faecalis V583, Enterococcus faecium TX1330, Escherichia coli MG1655, Lactobacillus johnsonii NCC533, and Methanobrevibacter smithii ATCC35061). Reads with alignment identity of 95% or higher were used to calculate the relative abundance (the percentage of mapped reads in total reads) and genome coverage (percentage of genome bases aligned to reads) for each genome. Heatmaps representing the relative abundance (A) and coverage (B) in each human sample for each of the seven organisms are shown. All organisms were detected in multiple (albeit not all samples) with the exception of SFB and E. faecium. SFB was the only organism not detected in any sample at thresholds of 0.02% abundance and 0.5% coverage (see text).

Similar articles

Cited by

References

    1. Atarashi K, Umesaki Y, Honda K. Microbiotal influence on T cell subset development. Seminars in immunology. 2011;23:146–153. - PubMed
    1. Ausiello CM, Cerquetti M, Fedele G, Spensieri F, Palazzo R, Nasso M, Frezza S, Mastrantonio P. Surface layer proteins from Clostridium difficile induce inflammatory and regulatory cytokines in human monocytes and dendritic cells. Microbes Infect. 2006;8:2640–2646. - PubMed
    1. Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, Formsma K, Gerdes S, Glass EM, Kubal M, et al. The RAST Server: rapid annotations using subsystems technology. BMC Genomics. 2008;9:75. - PMC - PubMed
    1. Barketi-Klai A, Hoys S, Lambert-Bordes S, Collignon A, Kansau I. Role of fibronectin binding protein A in Clostridium difficile intestinal colonization. J Med Microbiol. 2011;60:1155–1161. - PubMed
    1. Besemer J, Borodovsky M. Heuristic approach to deriving models for gene finding. Nucleic Acids Res. 1999;27:3911–3920. - PMC - PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources