Extensive microbial diversity within the chicken gut microbiome revealed by metagenomics and culture

Abstract

Background: The chicken is the most abundant food animal in the world. However, despite its importance, the chicken gut microbiome remains largely undefined. Here, we exploit culture-independent and culture-dependent approaches to reveal extensive taxonomic diversity within this complex microbial community.

Results: We performed metagenomic sequencing of fifty chicken faecal samples from two breeds and analysed these, alongside all (n = 582) relevant publicly available chicken metagenomes, to cluster over 20 million non-redundant genes and to construct over 5,500 metagenome-assembled bacterial genomes. In addition, we recovered nearly 600 bacteriophage genomes. This represents the most comprehensive view of taxonomic diversity within the chicken gut microbiome to date, encompassing hundreds of novel candidate bacterial genera and species. To provide a stable, clear and memorable nomenclature for novel species, we devised a scalable combinatorial system for the creation of hundreds of well-formed Latin binomials. We cultured and genome-sequenced bacterial isolates from chicken faeces, documenting over forty novel species, together with three species from the genus Escherichia, including the newly named species Escherichia whittamii.

Conclusions: Our metagenomic and culture-based analyses provide new insights into the bacterial, archaeal and bacteriophage components of the chicken gut microbiome. The resulting datasets expand the known diversity of the chicken gut microbiome and provide a key resource for future high-resolution taxonomic and functional studies on the chicken gut microbiome.

Keywords: Bacterial nomenclature; Biodiversity; Candidatus; Chickens; Gut microbiome; Metagenome-assembled genome; Metagenomics; Uncultured bacteria.

Figure 1. Analytical Workflow.

An asterisk (*) indicates read numbers are detailed post-filtering of diet and host associated reads.

Figure 2. Genome synteny of recovered phage genomes.

Synteny plots comparing four novel coliphage genomes recovered from chicken faecal metagenomes (in red) to closest reference genomes. The figure was generated using EasyFig.

Figure 3. Coliphage abundance within chicken faecal samples.

Coverage of four coliphages and of putative host bacterial species. Only samples in which at least one genome had ≥1× coverage are shown (n = 29). All coverage values have been Log₁₀ transformed with blue depicting low abundance and red high abundance.

Figure 4. Phylogenetic tree of draft MGS genomes derived from 820 metagenomic samples of the chicken gut and draft genomes of 93 species cultured from chicken faecal samples.

Phylum, generally as assigned by GTDB, is indicated by colour range. Data symbols in the outer layers have been used to describe further characteristics for each draft genomes. Triangles indicate sequence novelty and status of binomial designation within publicly available databases or published research with filled symbols indicating novel species assigned a binomial as part of this research, hollow symbol indicated a known species assigned a binomial as part of this research and no symbol indicated a known species with a well-formed binomial already assigned. Stars are used to indicate isolation source, with filled symbols indicating isolation of species in both culture and metagenomic assembly and hollow symbols indicating isolation in culture alone. Tree branches have been collapsed where duplicate species have been identified by different methodologies. The tree was reconstructed using PhyloPhlAn 3.0.58 against 400 marker genes before reconstruction using FastTree and RAxML of a MAFFT sequence alignment and visualised using the online iTOLv5.7 tool including provision of a heat map according to individual genome length.

Figure 5. Phylogenetic tree showing the relationships between Escherichia marmotae, Escherichia whittamii and the other Escherichia species and cryptic clades.

The tree was constructed by RAxML maximum likelihood analysis of a core genome alignment generated using Mugsy. The scale bar indicates the number of substitutions per site represented by the branch length shown. Numbers on branches indicate the percentage bootstrap support out of 100 replicates. Strains sequenced as part of this study are highlighted in red.

Figure 6. Sequence novelty.

(A) Venn diagram showing shared and unique taxonomic species among three data sources; cultured isolates derived from six chicken faecal samples (Cultured species), metagenomic species identified from a combined dataset of >630 chicken gastrointestinal metagenome samples (Metagenomic species); MAGs also found by Glendinning et al. (2020). (B) Percentage of classified metagenomic reads derived from 50 chicken faecal samples according to a standard Kraken 2 database (Previously) and to a standard Kraken 2 database with the addition of the 2,344 genomic and metagenomic sequences derived from this study (Now).

Figure 7. UpSet plots depicting presence of 820 metagenomic species across all BioProjects included within this study.

(A) 1× coverage b. (B) 10× coverage. Bars are stacked according to taxonomic species novelty, with black-stacked bars depicting novel species and grey depicting species previously described in public databases or published studies. Only intersections with five or more species are shown.