Extraordinary phylogenetic diversity and metabolic versatility in aquifer sediment

Abstract

Microorganisms in the subsurface represent a substantial but poorly understood component of the Earth's biosphere. Subsurface environments are complex and difficult to characterize; thus, their microbiota have remained as a 'dark matter' of the carbon and other biogeochemical cycles. Here we deeply sequence two sediment-hosted microbial communities from an aquifer adjacent to the Colorado River, CO, USA. No single organism represents more than ~1% of either community. Remarkably, many bacteria and archaea in these communities are novel at the phylum level or belong to phyla lacking a sequenced representative. The dominant organism in deeper sediment, RBG-1, is a member of a new phylum. On the basis of its reconstructed complete genome, RBG-1 is metabolically versatile. Its wide respiration-based repertoire may enable it to respond to the fluctuating redox environment close to the water table. We document extraordinary microbial novelty and the importance of previously unknown lineages in sediment biogeochemical transformations.

Figure 1. Rank abundance and taxonomic affiliations.

Top: rank abundance curve for the microbial community in the 5-m depth sample featuring the 161 organisms for which at least 8 of the 16 selected ribosomal proteins (Rpl2, 3, 4, 5, 6, 14, 15, 16, 18, 22, 24 and RpS3, 8, 10, 17, 19) could be recovered. The most abundant organism, RBG-1, represents <1% of the community. RBG-1 lineage members are denoted by red crosses. Bottom: summary of taxonomic affiliations for the 161 community members, based on the concatenated ribosomal protein tree (Supplementary Fig. S1 and Methods). The first row denotes the taxonomic assignment of each organism to a phylum or, for the Proteobacteria, class, based on placement and bootstrap support on the tree. Roman numbers i to xii indicate distinct novel clades within each taxonomic division. Novel clades with the same number identifier in different major groups are unrelated. There are 15 new potentially phylum-level groups (i to xii for Bacteria (in black), i to iii for Archaea (in dark blue)). Note that abundance (%) reflects DNA fraction rather than genome copy number (OP11 and OD1 have very small genome sizes). For details, see Methods, Supplementary Fig. S1 and Supplementary Data 1.

Figure 2. Maximum likelihood concatenated 16 ribosomal protein phylogeny.

RBG-1 lineage bacteria (now referred to as the Zixibacteria phylum) in the 5 m sample are shown in red. Bootstrap support values greater than 55 are displayed, black circles indicate nodes with greater than 95% bootstrap support. Each individual gene data set was aligned using Muscle version 3.8.31 (ref. 54) and then manually curated to remove end gaps and single-taxon insertions. The concatenated alignment contained 3,010 unambiguously aligned positions and 1,021 taxa. Maximum likelihood phylogeny was conducted using Phyml under the LG+α+γ model of evolution and with 100 bootstrap replicates (Methods).

Figure 3. Reconstructed energy metabolism of RBG-1.

For full gene information and for box numbers see Supplementary Table S1. Where protein modelling was possible, three-dimensional structural representations are shown (Methods). ArrA, arsenate reductase; ArxA, arsenite oxidase; Blue Cu, blue copper protein; But-CoA, Bbtyryl-CoA; Cit, citrate; CuNir, copper-nitrite reductase; D-ribo-5P, D-ribose-5-phosphate; Fd, ferredoxin; Fum, fumarate; G3P, glyceraldehyde-3-phosphate; 3-HB CoA, 3-hydroxybutyryl-CoA; 3-HP-CoA, 3-hydroxypimelyl-CoA; Iso, isocitrate; 2-oxo, 2-oxoglutarate; 3-Keto CoA, 3-ketoacyl-CoA; Mal, malate; Methylmal-CoA, methylmalonyl-CoA; NXR, nitrite/nitrate oxidoreductase; Oaa, oxaloacetate; Pro-CoA, propionyl-CoA; Suc-CoA, Succinyl-CoA; Succ, succinate.

Figure 4. Phylogenetic analysis of the catalytic subunits of the dimethyl sulphoxide (DMSO) reductase superfamily.

Genes that were newly assigned to the DMSO reductase superfamily (this study) are indicated by boxes. Red circles indicate 100% bootstrap support, green circles indicate nodes with greater than 90% bootstrap support. ACTB1, alternative complex III, domain 1 of subunit B; ArrA, arsenate reductase; ArxA, arsenite oxidase; DmsA, DMSO reductase, alpha subunit; NarG, nitrate reductase, alpha subunit; NrxA, nitrite oxidoreductase, alpha subunit; PsrA/PhsA, polysulphide reductase, subunit A, thiosulphate reductase, subunit A; QrcB, quinone reductase complex, subunit B; TtrA, tetrathionate reductase, subunit A. Accession numbers and amino acid sequences can be found as Supplementary Data 5.