The archaeal ancestor scenario (AAS) for the origin of eukaryotes implies the emergence of a new kind of organism from the fusion of ancestral archaeal and bacterial cells. Equipped with this "chimeric" molecular arsenal, the resulting cell would gradually accumulate unique genes and develop the complex molecular machineries and cellular compartments that are hallmarks of modern eukaryotes. In this regard, proteins related to phagocytosis and cell movement should be present in the archaeal ancestor, thus identifying the recently described candidate archaeal phylum "Lokiarchaeota" as resembling a possible candidate ancestor of eukaryotes. Despite its appeal, AAS seems incompatible with the genomic, molecular, and biochemical differences that exist between Archaea and Eukarya. In particular, the distribution of conserved protein domain structures in the proteomes of cellular organisms and viruses appears hard to reconcile with the AAS. In addition, concerns related to taxon and character sampling, presupposing bacterial outgroups in phylogenies, and nonuniform effects of protein domain structure rearrangement and gain/loss in concatenated alignments of protein sequences cast further doubt on AAS-supporting phylogenies. Here, we evaluate AAS against the traditional "three-domain" world of cellular organisms and propose that the discovery of Lokiarchaeota could be better reconciled under the latter view, especially in light of several additional biological and technical considerations.
Bacteria and Eukarya have cell membranes with sn-glycerol-3-phosphate (G3P), whereas archaeal membranes contain sn-glycerol-1-phosphate (G1P). Determining the time at which cells with either G3P-lipid membranes or G1P-lipid membranes appeared is important for understanding the early evolution of terrestrial life. To clarify this issue, we reconstructed molecular phylogenetic trees of G1PDH (G1P dehydrogenase; EgsA/AraM) which is responsible for G1P synthesis and G3PDHs (G3P dehydrogenase; GpsA and GlpA/GlpD) and glycerol kinase (GlpK) which is responsible for G3P synthesis. Together with the distribution of these protein-encoding genes among archaeal and bacterial groups, our phylogenetic analyses suggested that GlpA/GlpD in the Commonote (the last universal common ancestor of all extant life with a cellular form, Commonote commonote) acquired EgsA (G1PDH) from the archaeal common ancestor (Commonote archaea) and acquired GpsA and GlpK from a bacterial common ancestor (Commonote bacteria). In our scenario based on this study, the Commonote probably possessed a G3P-lipid membrane synthesized enzymatically, after which the archaeal lineage acquired G1PDH followed by the replacement of a G3P-lipid membrane with a G1P-lipid membrane.

