Integrating genomes and qSIP to probe the activity of Omnitrophota

Weaving between different data types allowed us to integrate otherwise discrete observations and build an incisive and testable picture of this enigmatic phylum.
Published in Microbiology
Integrating genomes and qSIP to probe the activity of Omnitrophota

Share this post

Choose a social network to share with, or copy the shortened URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Omnitrophota: an exemplar among uncultivated microbes

In 1998, Hugenholtz et al.1 described a then-large-scale analysis of bacterial 16S rRNA gene amplicons from Obsidian Pool in Yellowstone National Park, revealing 12 new deep-branching lineages. Among these was OP3, later dubbed “Omnitrophica” or Omnitrophota, which has subsequently been detected in many cultivation-independent studies. Yet, to date, no representative of the Omnitrophota has been cultivated in isolation. The first report of Omnitrophota in co-culture was published just recently in 2022. That organism, Velamenicoccus archaeovorus LiM2, is a small coccus that attaches to, and possibly kills, co-enriched cells of the archaeal genus Methanosaeta. Only one other species of Omnitrophota has even been microscopically observed, named SKK-013,4. SKK-01 is a large, ovoid, magnetic coccus that was physically enriched by magnetic cell sorting. SKK-01 contained intracellular sulfur, probably resulting from chemolithotrophic sulfide oxidation, which was consistent with interpretations of its genome.

In parallel with these important studies providing both phenotypic and genotypic data, were others describing genomes from the phylum. In the first, Rinke et al., 20135 described four single-amplified genomes (SAGs) of Omnitrophota from the terrestrial subsurface and saline environments. And in the second, Williams et al., 20216 described 14 additional Omnitrophota genomes from an Antarctic lake, this time binning metagenomic data into metagenome-assembled genomes (MAGs). These genomes, and those from V. archaeovorus LiM and SKK-01, provided important context to begin to understand the nature of the phylum, yet they were remarkably different and were not interpreted in a comparative framework.

The phylogeny of Omnitrophota based on the 'Bac120' marker set. SKK-01 and Velamenicoccus archaeovorus are indicated with stars.

BPH and I, as part of the Joint Genome Institute’s Microbial Dark Matter II (MDM II) Project, sought to change this by gathering a much larger collection of Omnitrophota genomes from a variety of biomes and studying their genomic diversity and potential physiology in toto. This was exciting enough for us, but as the project unfolded, we increasingly noticed a wealth of information associated with the genomes that could be leveraged to better contextualize them. Importantly, we were able to estimate the sizes of 36 Omnitrophota cells based on forward scatter measurements made during fluorescence-activated cell sorting that was done by EDB, RS, and others at the Single Cell Genomics Center at Bigelow Ocean Sciences as part of the MDM II Project. A range of cell sizes could also be interpreted based on serial filtration followed by metagenomics and or 16S rRNA gene amplicon sequencing. The latter could be integrated within our genomic framework because we generated a custom 16S rRNA gene taxonomy based on harmonizing our phylogenomic analyses with 16S rRNA gene phylogenetic analyses. Together, these different lines of evidence showed that hundreds of species across the entire phylum were very small – in most cases, <450 nm. This was hinted at previously, but these other studies did not assess relationships between cell size and Omnitrophota phylogeny.

Omnitrophota cell sizes. The dotted line is shown at 450 nanometers. Except for the class Gorgyraia, cells are predominantly nanobacteria.

To us, the finding that most Omnitrophota are nanobacteria was both surprising and exciting. Given this exciting news, along with exciting work we eagerly read out of the Max Planck Institute in Bremen about the probable predatory lifestyle of V. archaeovorus LiM, we hypothesized that a symbiotic lifestyle might be the rule across the Omnitrophota, which we could nominally support by our discovery of many genes associated with symbiotic lifestyles – namely those encoding tight-adherence apparatuses, type-IV pili collocated with type-2 and type-3 F-type ATPases, ATP/ADP translocases, and bizarre open reading frames that in some cases exceeded 100 kb in length.

This hypothesis got some unexpected wind behind its sails when my dad sent me a link to Hungate et al., 20218 describing their work that leveraged quantitative stable-isotope probing (qSIP) to show unusually high metabolic rates – hyperactivity – to be common among well-known bacterial predators and parasites such as Bdellovibrionales, Vampirovibrionales, Lysobacter, and Myxococcales. Knowing my suspicions that Omnitrophota are symbiotic, he suggested that Omnitrophota might also be hyperactive. He was right! We reached out to BAH and colleagues and asked to use their raw data. We reasoned that if Omnitrophota were symbiotic, then we could expect to see metabolic rates comparable to predators and parasites highlighted by their original paper. Although Omnitrophota were rare in the qSIP datasets, the volume of data proved sufficient. The three families of Omnitrophota present in the datasets – mapping to two different classes – were all hyperactive. We then also noticed that Omnitrophota from two additional classes were among the most abundant taxa in qSIP experiments done in anoxic Black Sea sediments9. Together these cell size and qSIP data paint a clearer picture: across the phylum most Omnitrophota are hyperactive nanobacteria. We now look forward to renewed interest in the Omnitrophota as part of a schema of prokaryote-on-prokaryote symbiosis not only among Omnitrophota but also diverse and pervasive Patescibacteria and DPANN archaea.

Behind the Paper

It is a rite-of-passage among microbial ecologists to find uncultivated phyla in their data and lack the knowledgebase necessary to contextualize their presence. I participated in this ritual during the Summer of 2017, combing through the taxonomy of an environmental 16S RNA gene Illumina amplicon survey to find unclassified microbes that needed to be updated with Candidatus names. “Omnitrophica” stood out to me. As the thirteenth-most abundant phylum in the entire dataset, “Omnitrophica” were outnumbered only by taxa that had “real” names. At the 2019 Joint Genome Institute User Meeting, I had the opportunity to speak with other microbial ecologists about their work. Despite how little we collectively knew about Omnitrophota, nearly every poster at the conference mentioned the lineage. The central pillar of my motivation was to fix this problem by dragging the lineage out of the dark, so to speak.

In early 2018, BPH offered to mentor me for graduate school. He gave me a choice of two projects for my thesis, but I made up my mind as soon as he mentioned “Omnitrophica”. We hatched a plan: (i) collect as many Omnitrophota MAGs and SAGs as we could, (ii) organize them into a systematic framework, and then (iii) run a robust suite of comparative genomics analyses (or, as described by BPH: “throw everything that we can at them”). There were a lot of data at our fingertips that nobody else seemed interested in using. Even if we were unable to find a feature to exploit to promote cultivation of the phylum, we knew that we would still provide a thorough analysis of their evolution and natural history. But, as described above, by being greedy for data we incorporated other types of data, which greatly improved our understanding of the biology of the phylum.


  1. Hugenholtz, P., Pitulle, C., Hershberger, K. L. & Pace, N. R. Novel Division Level Bacterial Diversity in a Yellowstone Hot Spring. Journal of Bacteriology 180, 366–376 (1998).
  2. Kizina, J. et al. Methanosaeta and “Candidatus Velamenicoccus archaeovorus”. Applied and Environmental Microbiology (2022) doi:10.1128/aem.02407-21.
  3. Kolinko, S. et al. Single-cell analysis reveals a novel uncultivated magnetotactic bacterium within the candidate division OP3. Environmental Microbiology 14, 1709–1721 (2012).
  4. Kolinko, S., Richter, M., Glöckner, F.-O., Brachmann, A. & Schüler, D. Single-cell genomics of uncultivated deep-branching magnetotactic bacteria reveals a conserved set of magnetosome genes. Environmental Microbiology 18, 21–37 (2016).
  5. Rinke, C. et al. Insights into the phylogeny and coding potential of microbial dark matter. Nature 499, 431–437 (2013).
  6. Williams, T. J., Allen, M. A., Berengut, J. F. & Cavicchioli, R. Shedding Light on Microbial “Dark Matter”: Insights Into Novel Cloacimonadota and Omnitrophota From an Antarctic Lake. Frontiers in Microbiology 12, 2947 (2021).
  7. Proctor, C. R. et al. Phylogenetic clustering of small low nucleic acid-content bacteria across diverse freshwater ecosystems. ISME J 12, 1344–1359 (2018).
  8. Hungate, B. A. et al. The Functional Significance of Bacterial Predators. mBio 12, (2021).
  9. Suominen, S., Dombrowski, N., Sinninghe Damsté, J. S. & Villanueva, L. A diverse uncultivated microbial community is responsible for organic matter degradation in the Black Sea sulphidic zone. Environ Microbiol 23, 2709–2728 (2021).


Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in