Para pensar cum granum salis: os tRNAs são mais antigos do que suas sintetases?

terça-feira, outubro 14, 2008

Genetic code origins: tRNAs older than their synthetases?

1. Lluís Ribas de Pouplana,
2. Robert J. Turner,
3. Brian A. Steer, and
4. Paul Schimmel*
Author Affiliations
1. The Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037
1. Contributed by Paul Schimmel

We present a phylogenetic analysis to determine whether a given tRNA molecule was established in evolution before its cognate aminoacyl-tRNA synthetase. The earlier appearance of tRNA versus their metabolically related enzymes is a prediction of the RNA world theory, but the available synthetase and tRNA sequences previously had not allowed a formal comparison of their relative time of appearance. Using data recently obtained from the emerging genome projects, our analysis points to the extant forms of lysyl-tRNA synthetase being preceded in evolution by the establishment of the identity of lysine tRNA.

The hypothesis of an RNA world postulates that self-replicating RNA molecules preceded the use of DNA and proteins, and that this world existed before the appearance of the universal ancestor of the extant tree of life (1). The existence of an RNA world has been supported by the biochemical characterization of catalytic RNA molecules, either from contemporary metabolic pathways or after in vitro selection of RNA ribozymes (2–6). Viral RNA genomes and the role or tRNA-like structures in viral replication are also indicative of the ancestral existence of an RNA world (7). A more direct proof of an RNA world could come from the direct comparison of the evolutionary time of appearance of protein and RNA molecules involved in a universal metabolic pathway. If this analysis was possible, then the RNA world theory would predict that the moment of appearance of the RNA component would precede the appearance of the protein elements involved in the same reaction. Here we present a phylogenetic analysis that suggests that, in an RNA-protein interaction essential for the elucidation of the genetic code, the RNA molecule is ancestral to its associated enzyme.

Aminoacyl-tRNA synthetases (aaRSs) evolved as two distinct classes (I and II), each containing 10 enzymes (8–14). Each aaRS is responsible for establishing the genetic code by specifically aminoacylating only its cognate tRNA isoacceptors, thereby linking an amino acid with its corresponding anticodon triplets. Because the aminoacylation of tRNA establishes the genetic code, a strong coevolution exists between the enzymes and their cognate tRNAs (15). The aminoacylation reaction precedes the first split of the tree of life, resulting in almost invariable conservation of aaRSs and their cognate tRNAs in all living organisms (16).

The strict conservation of aaRS and tRNA sequences across the whole phylogenetic tree prevented the analysis of initial events in the evolution of the system, because no sequences exist from precursors of the extant aaRSs. Without this kind of sequence information, the relative age of the duplications that gave rise to the current set of aaRSs could not be calculated. Moreover, the relative time of appearance of aaRSs and tRNAs could not be analyzed, because no extant organism are known presently where earlier, simpler sets of aaRS or tRNAs are used. As a result, it has not been possible to calculate whether the final evolutionary events that gave rise to modern aaRSs had taken place after the time when tRNAs had already evolved.

This situation changed with the sequencing of the genome of the archaebacterium Methanococcus jannaschii (17) and with the exponential growth of sequence data from other genome sequencing projects. In an initial analysis, M. jannaschii’s genome was found to lack an ORF coding for a canonical class II LysRS. Two reports by Ibba et al. (18, 19) established that the aminoacylation of tRNALys in a subset of archaebacteria (i.e., M. jannaschii, a member of the euryarchae) and bacteria of the spirochete group (i.e., Treponema pallidum and Borrelia burgdorferii) appears to be catalyzed by a class I-type LysRS. This is the first example of a class switch by an aaRS.

The origin of this new enzyme must lie, presumably, within the set of duplication events that gave rise to the rest of class I aaRSs. However, its distribution within the phylogenetic tree (it is present in a limited number of archaeal and bacterial species) can be initially explained by three different evolutionary models (Fig. 1). A first possibility would be a late duplication event from a class I aaRS in one of these branches, followed by horizontal gene transfer. Another potential model would require a late duplication event that, independently, gave rise to two different class I LysRSs in a subgroup of archaea and of bacteria. Finally, the observed distribution also can be explained by an early duplication event, at the base of the phylogenetic tree, which produced a class I LysRS that later was conserved only in limited groups of organisms, while the majority of species adopted a class II LysRS. The later scheme of events would imply the coexistence of class I and II LysRS enzymes in an organism ancestral to all existing species (Fig. 1). The phylogenetic relationships between class I LysRS sequences and the rest of class I aaRSs would be different in each model. As a result, cladistic analysis can be used to test each of the three possible evolutionary schemes (Fig. 1).



Aqui nós temos uma situação do tipo: "Quem veio primeiro, o ovo ou a galinha?" E eles têm a desonestidade acadêmica de retratar o fato, Fato, FATO da evolução nos livros didáticos de Biologia do ensino médio e superior como sendo corroborado científicamente. E alguns na mídia reverberam: "Tão certa quanto a lei da gravidade."

Leia o artigo baixando o PDF gratuito aqui no site do PNAS.