Ribosome Paleontology


The initial timelines indicate that the peptidyl transferase center (PTC) and parts of the 50S subunit emerged before any aspects of the 30S subunit. As the Fox group recently reported at AbSciCon and the NAI Workshop without walls this has the important implication that the machinery was initially non-coding and hence the ability to make coded peptides was likely not the immediate driving force behind the emergence of the ribosome.


Previous efforts of Bokov and Steinberg (1) have emphasized that the tertiary contacts between different regions of the RNA represent potential timing events. Their analysis utilizes the A minor motif but there are many other types of interactions that may also be employed for this purpose. For example, the terminal A of GNRA tetraloops frequently forms distant interactions that one may be able to use to assign timing. Likewise in many instances tertiary base-base interactions are comprised of a Watson-Crick pair in a stem region involved with a distance base. As in the case of A-minor interactions, the directionality is likely such that the distant interaction is newer than the pairing. Our ongoing efforts are utilizing this additional data to increase the resolution of the hierarchical model that Bokov and Steinberg have created for evolution of the large subunit RNA.


As separate timelines for various subsystems emerge, it is important to examine their compatibility of one approach with others in the hope reaching consistent conclusions. Detailed examination of the exit tunnel, Figure 1, in the past year illustrated that this can be done. The tunnel likely began at the PTC and grew longer over time. In order to accommodate the growing ribosome, the tunnel would have to get longer over time in order for the peptide to emerge. Hence the tunnel length can be correlated with the timing of the enlargement of the RNA and the various additions correlated with and in fact mapped onto a hierarchical model of 23S rRNA evolution. Thus, in the past year the tunnel was found to pass through the Domains in the order VàIVàIIàIàIII which therefore is their likely relative times of addition. This is in agreement with the Bokov and Steinberg model as well as the Williams group’s onion model (2) and the Fox group’s earlier results based on connectivity (3). The mRNA channel in the 30S subunit may provide similar insights. In this case, the timing is likely bidirectional emanating from the decoding center.


It is our view that the level of complexity that characterizes LUCA is not accidental, but rather reflects that state of things when a globally important event in biological evolution occurred. The most likely candidate for this event is the greatly improved control of the characteristic movements of the ribosome that would be contributed by the addition of the GTPase Center to the ribosome. This would have accelerated peptidyl transfer by at least an order of magnitude and thus allowed major expansion in the capabilities of the primitive living systems then extant. The GTPase center would, however, have needed to be integrated into the ribosome inner workings and to do this it is likely that a signaling system would be needed to facilitate timing of GTP cleavages.  Thus, all the ribosomal processes and components that are evolving at the LUCA boundary are likely to be of special interest. One such component is 5S rRNA, which is centrally located on a path, Figure 2, which ultimately connects the PTC in the 50S subunit to the decoding center in the 30S subunit. Previously a large number of 5S rRNA mutants were constructed for studies of the structure of RNA sequence space. Ongoing efforts are focused on characterizing these variants from the perspective of ribosome function. Mutations at U78 where the A-site finger crosses 5S rRNA are especially deleterious.

Primary and secondary structures of rRNAs (ribosomal RNAs) and rProteins (ribosomal proteins) are available for a large number of species. Atomic-resolution three-dimensional (3D) structures of ribosomes are available for a few species (Figure 3). These 3D structures are information dense, revealing tertiary structural information as well as primary and secondary structural data. Information contained in the small subset of available 3D structures, can be extrapolated to other species.

To visualize the structural and evolutionary data we have been developing a software that readily and quickly transfers information between 1D (primary structure), 2D (secondary structure), and 3D (tertiary structure) spaces. Primary structure is used for sequence-alignments and therefore contains information on genetic variability. Secondary structure contains local organizational information and describes folding, motifs, functional and evolutionary units (helices and domains). Tertiary structure contains atomic level detail, including conformation and molecular interactions. Additionally, diverse information such as results of footprinting experiments, evolutionary data, and physicochemical properties can also be mapped onto any of the three spaces. 


The simultaneous representation of multiple types of data (Figure 4) for the whole ribosome in several dimensional spaces provides a powerful approach for detecting correlations between the variables. Our procedure is automated, allowing for high throughput. Through creative use of colors, symbols, and hyperlinks many types of information can be analyzed simultaneously (Figure 5).


Our method allows us to present novel information about the roles of rRNA and rProteins, which were elucidated from analyzing rRNA-rRNA and protein-rRNA contact maps projected onto the secondary structure of the ribosome. Our automated data processing and visualization software allows us to filter the contact maps based on sequence conservation, the conservation of a physicochemical property, or any other residue level property. The data can be projected onto ribosomal secondary structure per rRNA residue basis as well as grouped using coarser-grained representations of rRNA (per helix or per domain bases).




1.         Bokov K, Steinberg SV (2009) A Hierarchical Model for Evolution of 23S Ribosomal RNA. Nature 457:977-980.

2.         Hsiao C, Mohan S, Kalahar BK, Williams LD (2009) Peeling the Onion: Ribosomes Are Ancient Molecular Fossils. Mol Biol Evol 26:2415-2425.

3.         Hury J, Nagaswamy U, Larios-Sanz M, Fox GE (2006) Ribosome Origins: The Relative Age of 23S rRNA Domains. Orig Life Evol Biosph 36:421-429.