Inter-compatibility of eukaryotic and Asgard archaea ribosome-translocon machineries.

In all domains of life, the ribosome-translocon complex inserts nascent transmembrane proteins into, and processes and transports signal peptide-containing proteins across, membranes. Eukaryotic translocons are anchored in the endoplasmic reticulum, while the prokaryotic complexes reside in cell membranes. Phylogenetic analyses indicate inheritance of eukaryotic Sec61/OST/TRAP translocon subunits from an Asgard archaea ancestor. However, the mechanism for translocon migration from a peripheral membrane to an internal cellular compartment (the proto-endoplasmic reticulum) during eukaryogenesis is unknown. Here we show compatibility between the eukaryotic ribosome-translocon complex and Asgard signal peptides and transmembrane proteins. We find that Asgard translocon proteins from Candidatus Prometheoarchaeum syntrophicum strain MK-D1, a Lokiarchaeon confirmed to contain no internal cellular membranes, are targeted to the eukaryotic endoplasmic reticulum on ectopic expression. Furthermore, we show that the cytoplasmic domain of MK-D1 OST1 (ribophorin I) can interact with eukaryotic ribosomes. Our data indicate that the location of existing ribosome-translocon complexes, at the protein level, determines the future placement of yet to be translated translocon subunits. This principle predicts that during eukaryogenesis, under positive selection pressure, the relocation of a few translocon complexes to the proto-endoplasmic reticulum will have contributed to propagating the new translocon location, leading to their loss from the cell membrane.

The eukaryotic-like characteristics of small GTPase, roadblock and TRAPPC3 proteins from Asgard archaea.

Membrane-enclosed organelles are defining features of eukaryotes in distinguishing these organisms from prokaryotes. Specification of distinct membranes is critical to assemble and maintain discrete compartments. Small GTPases and their regulators are the signaling molecules that drive membrane-modifying machineries to the desired location. These signaling molecules include Rab and Rag GTPases, roadblock and longin domain proteins, and TRAPPC3-like proteins. Here, we take a structural approach to assess the relatedness of these eukaryotic-like proteins in Asgard archaea, the closest known prokaryotic relatives to eukaryotes. We find that the Asgard archaea GTPase core domains closely resemble eukaryotic Rabs and Rags. Asgard archaea roadblock, longin and TRAPPC3 domain-containing proteins form dimers similar to those found in the eukaryotic TRAPP and Ragulator complexes. We conclude that the emergence of these protein architectures predated eukaryogenesis, however further adaptations occurred in proto-eukaryotes to allow these proteins to regulate distinct internal membranes.

Short Peptide Amyloids Are a Potential Sequence Pool for the Emergence of Proteins.

Under prebiotic conditions, peptides are capable of self-replication through a structure-based template-assisted mechanism when they form amyloids. Furthermore, peptide amyloids can spontaneously form inside fatty acid vesicles creating membrane enclosed complex structures of variable morphologies. This is possible because fatty acid vesicle membranes act as filters allowing passage of activated amino acids while some amino acids derived from the activated species become non-permeable and trapped in the vesicles. Similarly, nascent peptides derived from the condensation of the activated amino acids are also trapped in the vesicles. It is hypothesized that such preselected peptide amyloids become a sequence pool for the emergence of proteins in life and that after billions of years of cellular evolution, the sequences in the current proteome have diverged significantly from these original seed peptides. If this hypothesis is correct, it could be possible to detect the traces of these seed sequences in current proteomes. Here, we show for all possible 3, 6, 7, 8 or 9 residue sequence motifs that those motifs that are most amyloidogenic/aggregation prone are over-represented in extant proteomes compared to a sequence-randomized proteome. Furthermore, we find that there is a greater proportion of amyloidogenic sequence motifs in archaea proteomes than in the larger primate proteomes. This suggests that the evolution towards larger proteomes leads to smaller proportion of amyloidogenic sequences.