The ubiquity and conservation of tRNA in the replication strategies of a variety of contemporary genomes suggest a functional phylogeny for tRNA. This phylogeny is unique in placing the origin of tRNA in replication, prior to the advent of templated protein synthesis. In this scenario, aminoacyl-tRNA synthetase activities would have arisen next, to facilitate or regulate replication, and both tRNA and the aminoacyl-tRNA synthetase activities would have predated the anticodon and mRNA. A corollary is that the top half of modern tRNA may have had a more ancient origin than the bottom half bearing the anticodon.
The genomic tag hypothesis has "explanatory power" (Popper, 1963). It makes sense of—and establishes possible relationships between—otherwise puzzling structures and functions including RNase P, the CCA-adding enzyme, telomerase, contemporary synthetases, and the terminal tRNA-like structures themselves. It is also robust. A number of key experiments alluded to above in support of the genomic tag model were carried out after our original proposal (Weiner and Maizels, 1987), including studies of the Mauriceville retroplasmid (Wang and Lambowitz, 1993), the internal RNA template of telomerase (Blackburn, 1991), cleavage of plant virus tRNA-like structures by RNase P (Green et al., 1988; Guerrier-Takada et al., 1988; Mans et al., 1990), stereospecific binding of an amino acid by RNA (Connell et al., 1993), the ability of a ribozyme to work on a carbon center (Piccirilli et al., 1992), and the division of contemporary synthetases into two classes (Eriani et al., 1990).
What do we expect to learn from this model in the future? One prediction is that there are likely to be other transitional genomes that employ tRNA or tRNA-like structures in their replication. Another is that detailed functional and structural studies of contemporary tRNAs (White, 1982; Pan et al., 1991) will further support the independence of the top and bottom halves of the molecule, explain why contemporary tRNA is a cloverleaf rather than the pseudoknotted structure found in plant viruses (Mans et al., 1990), and unlock the evolutionary history that must lie in the location and function of the (almost) universally modified bases in tRNA. But especially exciting is the possibility that plausible phylogenies will emerge for other key biochemical pathways, grounded in the structure and function of contemporary molecules.
We propose a phylogeny for the evolution of tRNA that is based on the ubiquity and conservation of tRNA-like structures in the replication of