Details

Autor(en) / Beteiligte

• Sharon, Itai
• Morowitz, Michael J
• Thomas, Brian C
• Costello, Elizabeth K
• Relman, David A
• Banfield, Jillian F

Titel

Time series community genomics analysis reveals rapid shifts in bacterial species, strains, and phage during infant gut colonization

Ist Teil von

• Genome research, 2013-01, Vol.23 (1), p.111-120

Ort / Verlag

United States

Erscheinungsjahr

2013

Quelle

MEDLINE

Beschreibungen/Notizen

• The gastrointestinal microbiome undergoes shifts in species and strain abundances, yet dynamics involving closely related microorganisms remain largely unknown because most methods cannot resolve them. We developed new metagenomic methods and utilized them to track species and strain level variations in microbial communities in 11 fecal samples collected from a premature infant during the first month of life. Ninety six percent of the sequencing reads were assembled into scaffolds of >500 bp in length that could be assigned to organisms at the strain level. Six essentially complete (∼99%) and two near-complete genomes were assembled for bacteria that comprised as little as 1% of the community, as well as nine partial genomes of bacteria representing as little as 0.05%. In addition, three viral genomes were assembled and assigned to their hosts. The relative abundance of three Staphylococcus epidermidis strains, as well as three phages that infect them, changed dramatically over time. Genes possibly related to these shifts include those for resistance to antibiotics, heavy metals, and phage. At the species level, we observed the decline of an early-colonizing Propionibacterium acnes strain similar to SK137 and the proliferation of novel Propionibacterium and Peptoniphilus species late in colonization. The Propionibacterium species differed in their ability to metabolize carbon compounds such as inositol and sialic acid, indicating that shifts in species composition likely impact the metabolic potential of the community. These results highlight the benefit of reconstructing complete genomes from metagenomic data and demonstrate methods for achieving this goal.