Search

The secondary structure of the 5S ribosomal RNA (5S rRNA) molecule shows a high degree of symmetry. In order to explain the origin of this symmetry, it has been conjectured that one half of the 5S rRNA molecule was its precursor and that an indirect duplication of this precursor created the other half and thus the current symmetry of the molecule. Here, we have subjected to an empirical test both the indirect duplication model, analysing a total of 684 5S rRNA sequences for complementarity between the two halves of the 5S rRNA, and the direct duplication model analysing in this case the similarity between the two halves of this molecule. In intra- and inter-molecule and intra- and inter-domain comparisons, we find a high statistical support to the hypothesis of a complementarity relationship between the two halves of the 5S rRNA molecule, denying vice versa the hypothesis of similarity between these halves. Therefore, these observations corroborate the indirect duplication model at the expense of the direct duplication model, as reason of the origin of the 5S rRNA molecule. More generally, we discuss and favour the hypothesis that all RNAs and proteins, which present symmetry, did so through gene duplication and not by gradualistic accumulation of few monomers or segments of molecule into a gradualistic growth process. This would be the consequence of the very high propensity that nucleic acids have to be subjected to duplications.

Evolutionary models of molecular structures must incorporate molecular information at different levels of structural complexity and must be phrased within a phylogenetic perspective. In this regard, phylogenetic trees of substructures that are reconstructed from molecular features that contribute to order and thermodynamic stability show that a gradual model of evolution of 5S rRNA structure is more parsimonious than models that invoke large segmental duplications of the molecule. The search for trees of substructures that are most parsimonious, by their very nature, defines an objective strategy to select models of molecular change that best fit structural data. When combined with additional data, such as the age of protein domains that interact with RNA substructures, these trees can be used to falsify unlikely hypotheses.

5S rRNA is the smallest nucleic acid component of the large ribosomal subunit, contributing to ribosomal assembly, stability, and function. Despite being a model for the study of RNA structure and RNA-protein interactions, the evolution of this universally conserved molecule remains unclear. Here, we explore the history of the three-domain structure of 5S rRNA using phylogenetic trees that are reconstructed directly from molecular structure. A total of 46 structural characters describing the geometry of 666 5S rRNAs were used to derive intrinsically rooted trees of molecules and molecular substructures. Trees of molecules revealed the tripartite nature of life. In these trees, superkingdom Archaea formed a paraphyletic basal group, while Bacteria and Eukarya were monophyletic and derived. Trees of molecular substructures supported an origin of the molecule in a segment that is homologous to helix I (alpha domain), its initial enhancement with helix III (beta domain), and the early formation of the three-domain structure typical of modern 5S rRNA in Archaea. The delayed formation of the branched structure in Bacteria and Eukarya lends further support to the archaeal rooting of the tree of life. Remarkably, the evolution of molecular interactions between 5S rRNA and associated ribosomal proteins inferred from a census of domain structure in hundreds of genomes established a tight relationship between the age of 5S rRNA helices and the age of ribosomal proteins. Results suggest 5S rRNA originated relatively quickly but quite late in evolution, at a time when primordial metabolic enzymes and translation machinery were already in place. The molecule therefore represents a late evolutionary addition to the ribosomal ensemble that occurred prior to the early diversification of Archaea.

This study was designed to characterize further the nontranscribed intergenic spacers (NTSs) of the 5S rRNA genes of fish and evaluate this marker as a tool for comparative studies. Two members of the closely related North American Great Lakes cisco species complex (Coregonus artedi and C. zenithicus) were chosen for comparison. Fluorescence in situ hybridization found the ciscoes to have a single multicopy 5S locus located in a C band-positive region of the largest submetacentric chromosome. The entire NTS was amplified from the two species by polymerase chain reaction with oligonucleotide primers anchored in the conserved 5S coding region. Complete sequences were determined for 25 clones from four individuals representing two discrete NTS length variants. Sequence analysis found the length variants to result from presence of a 130-bp direct repeat. No two sequences from a single fish were identical. Examination of sequence from the coding region revealed two types of 5S genes in addition to pseudogenes. This suggests the presence of both somatic and germline (oocyte) forms of the 5S gene in the genome of Coregonus. The amount of variation present among NTS sequences indicates that accumulation of variation (mutation) is greater in this multicopy gene than is gene conversion (homogenization). The high level of sequence variation makes the 5S NTS an inappropriate DNA sequence for comparisons of closely related taxa.

The chloroplast 5S rRNA gene of the brown alga Pylaiella littoralis (L.) Kjellm has been cloned and sequenced. The gene is located 23 bp downstream from the 3' end of the 23S rRNA gene. The sequence of the gene is as follows: GGTCTTG GTGTTTAAAGGATAGTGGAACCACATTGAT CCATATCGAACTCAATGGTGAAACATTATT ACAGTAACAATACTTAAGGAGGAGTCCTTTGGGAAGATAGCTTATGCCTAAGAC. A secondary structure model is proposed, and compared to those for the chloroplast 5S rRNAs of spinach and the red alga Porphyra umbilicalis. Cladograms based on chloroplast and bacterial 5S rRNA and rRNA gene sequences were constructed using the MacClade program with a user-defined character transformation in which transitions and transversions were assigned unequal step values. The topology of the resulting cladogram indicates a polyphyletic origin for photosynthetic organelles.