Research Article

A rooted phylogeny resolves early bacterial evolution

See allHide authors and affiliations

Science  07 May 2021:
Vol. 372, Issue 6542, eabe0511
DOI: 10.1126/science.abe0511

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Reconstructing ancestral bacteria

The origin of the eubacteria and phylogenetic relationships between subgroups have been difficult to resolve. Applying a phylogenetic analysis and recent computational methods to the expanded diversity of bacterial sequences from metagenomic analyses, Coleman et al. infer the root of the eubacterial tree (see the Perspective by Katz). The root was determined without using the Archaea as an outgroup, to avoid the possibility of a false result due to long branch attraction. This method places the eubacterial root in the neighborhood of Fusobacteriota. Using this information, the authors reconstructed the eubacterial ancestor, identifying that this organism likely had a double-membrane cell envelope, flagellum-mediated motility, antiphage defense mechanisms, and diverse metabolic pathways.

Science, this issue p. eabe0511; see also p. 574

Structured Abstract


Bacteria are the most diverse and abundant cellular organisms on Earth, and in recent years environmental genomics has revealed the existence of an enormous diversity of previously unknown lineages. Despite the abundance of genomic sequence data, the root of the bacterial tree and the nature of the most recent common ancestor of Bacteria have remained elusive. The problem is that even with the help of new data, tracing billions of years of bacterial evolution back to the root has remained challenging because standard phylogenetic models do not account for the full range of evolutionary processes that shape bacterial genomes. In particular, standard models treat horizontal gene transfer as an impediment to the reconstruction of the tree of life that must be removed from analyses. But if horizontal gene transfer is modeled appropriately, it can provide information about the deep past that is unavailable to standard methods.


We reconstructed and rooted the bacterial tree by applying a hierarchical phylogenomic approach that explicitly uses information from gene duplications and losses within a genome as well as gene transfers between genomes. This approach allowed us to root the tree without including an archaeal outgroup. Outgroup-free rooting is a promising approach for Bacteria, both because the position of the universal root is uncertain and because the long branch separating Bacteria from Archaea has the potential to distort the reconstruction of within-Bacteria relationships. Outgroup-free gene tree-species tree reconciliation allowed us to quantitatively model both the vertical and horizontal components of bacterial evolution and integrate information from 11,272 gene families to resolve the root of the bacterial tree. Notably, these analyses also provided estimates of the gene content of the last bacterial common ancestor.


Our analyses place the root between two major bacterial clades, the Gracilicutes and Terrabacteria. We found no support for a root between the Candidate Phyla Radiation (CPR), a lineage comprising putative symbionts and parasites with small genomes, and all other Bacteria. Instead, the CPR was inferred to be a member of the Terrabacteria and formed a sister lineage to the Chloroflexota and Dormibacterota. This suggests that the CPR evolved by reductive genome evolution from free-living ancestors. Gene families inferred to have been present at the root indicate that the last bacterial common ancestor was already a complex double-membraned cell capable of motility and chemotaxis that possessed a CRISPR-Cas system. Although ~92% of gene families have experienced horizontal transfers during their history, tracing their evolution along the most likely rooted tree revealed that about two-thirds of gene transmissions have been vertical. Thus, bacterial evolution has a major vertical component, despite a profound impact of horizontal gene transfer through time. Horizontal gene flows can also provide insight into the temporal sequence of events during bacterial diversification, because donor lineages must be at least as old as recipients. Analysis of gene transfers in our dataset suggests that the diversification of the Firmicutes, CPR, Acidobacteriota, and Proteobacteria is the oldest among extant bacterial phyla.


The vertical and horizontal components of genome evolution provide complementary sources of information about bacterial phylogeny. The vertical component provides a robust framework for interpreting species diversity and allows us to reconstruct ancestral states, while the horizontal component helps to root the vertical tree and orient it in time. The inferred Gracilicutes-Terrabacteria root will be useful for investigating the tempo and mode of bacterial diversification, metabolic innovation, and changes in cell architecture such as the evolutionary transitions between double (diderm) and single (monoderm) membranes. Future development of methods that harness the complementarity of vertical and horizontal gene transmission will continue to further our understanding of life on Earth.

A rooted phylogeny of Bacteria.

The reconciliation of bacterial gene phylogenies places the root between the major clades of Gracilicutes (including Proteobacteria and Bacteroidota) and Terrabacteria (including Firmicutes and Cyanobacteria). On the basis of this hypothesis, ancestral genome reconstruction predicts that the last bacterial common ancestor (LBCA) was a complex, double-membraned cell and that, on average, two-thirds of gene transmissions have been vertically inherited along this rooted tree.


A rooted bacterial tree is necessary to understand early evolution, but the position of the root is contested. Here, we model the evolution of 11,272 gene families to identify the root, extent of horizontal gene transfer (HGT), and the nature of the last bacterial common ancestor (LBCA). Our analyses root the tree between the major clades Terrabacteria and Gracilicutes and suggest that LBCA was a free-living flagellated, rod-shaped double-membraned organism. Contrary to recent proposals, our analyses reject a basal placement of the Candidate Phyla Radiation, which instead branches sister to Chloroflexota within Terrabacteria. While most gene families (92%) have evidence of HGT, overall, two-thirds of gene transmissions have been vertical, suggesting that a rooted tree provides a meaningful frame of reference for interpreting bacterial evolution.

View Full Text

Stay Connected to Science