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.

One of the key differences between Bacteria and Archaea is their canonical membrane phospholipids, which are synthesized by distinct biosynthetic pathways with nonhomologous enzymes. This “lipid divide” has important implications for the early evolution of cells and the type of membrane phospholipids present in the last universal common ancestor. One of the main challenges in studies of membrane evolution is that the key biosynthetic genes are ancient and their evolutionary histories are poorly resolved. This poses major challenges for traditional rooting methods because the only available outgroups are distantly related. Here, we address this issue by using the best available substitution models for single-gene trees, by expanding our analyses to the diversity of uncultivated prokaryotes recently revealed by environmental genomics, and by using two complementary approaches to rooting that do not depend on outgroups. Consistent with some previous analyses, our rooted gene trees support extensive interdomain horizontal transfer of membrane phospholipid biosynthetic genes, primarily from Archaea to Bacteria. They also suggest that the capacity to make archaeal-type membrane phospholipids was already present in last universal common ancestor.

Bacteria are the most abundant and metabolically diverse cellular lifeforms on Earth. A rooted bacterial phylogeny provides a framework to interpret this diversity and to understand the nature of early life. Inferring the position of the bacterial root is complicated by incomplete taxon sampling and the long branch to the archaeal outgroup. To circumvent these limitations, we model bacterial genome evolution at the level of gene duplication, transfer and loss events, allowing outgroup-free inference of the root1. We infer a rooted bacterial tree on which 68% of gene transmission events are vertical. Our analyses reveal a basal split between Terrabacteria and Gracilicutes, which together encompass almost all known bacterial diversity. However, the position of one phylum, Fusobacteriota, could not be resolved in relation to these two major clades. In contrast to recent proposals, our analyses strongly reject a root between the Candidate Phyla Radiation (CPR) and all other Bacteria. Instead, we find that the CPR is a sister lineage to the Chloroflexota within the Terrabacteria. We predict that the last bacterial common ancestor was a free-living flagellated, rod-shaped cell featuring a double membrane with a lipopolysaccharide outer layer, a Type III CRISPR-Cas system, Type IV pili, and the ability to sense and respond via chemotaxis.

One of the key differences between Bacteria and Archaea are their canonical membrane phospholipids, which are synthesized by distinct biosynthetic pathways with non-homologous enzymes. This “lipid divide” has important implications for the early evolution of cells and the type of membrane phospholipids present in the last universal common ancestor (LUCA). One of the main challenges in studies of membrane evolution is that the key biosynthetic genes are ancient and their evolutionary histories are poorly resolved. This poses major challenges for traditional rooting methods because the only available outgroups are distantly related. Here, we address this issue by using the best available substitution models for single gene trees, by expanding our analyses to the diversity of uncultivated prokaryotes recently revealed by environmental genomics, and by using two complementary approaches to rooting that do not depend on outgroups. Consistent with some previous analyses, our rooted gene trees support extensive inter-domain horizontal transfer of membrane phospholipid biosynthetic genes, primarily from Archaea to Bacteria. They also suggest that the capacity to make archaeal-type membrane phospholipids was already present in LUCA.