| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biochemistry and Molecular Pharmacology, Program in Cell Dynamics, University of Massachusetts Medical School, Worcester, Massachusetts, USA
2Department of Biochemistry and Molecular Pharmacology, Program in Cell Dynamics, University of Massachusetts Medical School, 377 Plantation St., Worcester, MA 01605, USA. E-mail: thoru.pederson{at}umassmed.edu
ABSTRACT
SUMMARY In 1962, one of the most creative and cogent experiments on the protein coding problem was published. Now it has been discovered that archaebacteria had been doing a related kind of "experiment" all along. Both involve a trick: changing an amino acid that is already attached to a "correct" transfer RNA.
The discoveries of the double helix and messenger RNA were monumental milestones, but they immediately posed the problem of how the gene directs an amino acid sequence. For this, a third discovery was needed. Francis Crick predicted it—an "adaptor" (1
, 2)
and Mahlon Hoagland and Paul Zamecnik found it—transfer RNA (3)
. A few years later (1960–62) Heinrich Matthei and Marshall Nirenberg were refining their use of polynucleotides of defined sequence as templates for incorporation of specific amino acids. Sydney Brenner and Francis Crick were coming up with various coding schemes, as was the outlaw George Gamow, with the former two coming up with the correct solution: a non-overlapping triplet code. Meanwhile, the always prescient Seymour Benzer was carrying out acridine mutagenesis studies that defined the concept of the translational reading frame. This was all seminal and key (4
, 5)
but there was another experiment done at this time that was exceptionally clever (amidst this landscape of very creative people and the breakthrough experiments underway).
Francois Chapeville and colleagues had the idea that if they could link an amino acid to its appropriate tRNA and then chemically change the attached amino acid, they could ask whether the specificity of coding resided in the tRNA or in the amino acid. They fed cysteine into a cell extract that contained all the enzymes (later termed aminoacyl-tRNA synthetases) that attach amino acids to tRNAs. They then carried out a reductive chemical reaction that converted the attached cysteine to alanine (Fig. 1
A). Their triumphant finding was that the alanine was incorporated as if it were still cysteine, proving that the coding mechanism resided in the particular transfer RNA, not in the attached amino acid (6)
.
|
Fast-forward three and a half decades. In 1996, the genome of the archaebacterium Methanococcus jannaschii was sequenced, and Dieter Söll and colleagues went on to demonstrate that in this organism tRNAAsn and tRNAGln are misaminoacylated with aspartyl and glutamyl moieties, respectively, by synthestases displaying relaxed specificity. The free carboxyl ends of the attached aspartyl and glutamyl groups are then subjected to transamidation (Fig. 1B
; refs. 7
8
9
). The existence of these "unorthodox" pathways to asparaginyl-tRNAAsn and glutaminyl-tRNAGln in this archaebacterium implies that these relaxed specificity aminoacyl-tRNA synthetases and the requisite amidotransferases catalyzing the second step were subject to co-selection, as either one alone would produce amino acid substitutions, of which some would be neutral but many would be deleterious. The fixation of these co-selected enzymes has allowed the archeabacteria to endure without any "cognate" aminoacyl-tRNA synthetases for either asparagine or glutamine.
One is struck by the conceptual resemblance of this pathway (Fig. 1B
) to the experiment of Chapeville et al. (Fig. 1A
). The obvious difference of course is that the M. jannaschii pathway attaches the wrong amino acid to the right tRNA and then the amino acid is made right, whereas Chapeville et al. attached the right amino acid to the right tRNA and then made the amino acid wrong.
Desulfurization of cysteine to alanine à la Chapeville et al. is not thought to occur in vivo (10)
and we can only guess as to whether this reaction was ever tried in evolution (on or off tRNA). Ironically, cysteinyl-tRNA formation itself has been the subject of further surprises in the aminoacyl-tRNA synthetatse field, coming again from archaebacteria. A bioinformatics study had revealed the presence of a class II cysteinyl-tRNA synthetase in M. jannaschii, and orthologs in other euryarchaebacteria indicated that these are a gene family evolutionarily unrelated to class I cysteinyl-tRNA syntheases (11)
. Söll and colleagues demonstrated that this synthetase first attaches O-phosphoserine (the usual biosynthetic precursor of serine) to tRNACys followed by its conversion to cysteine (Fig. 1C
; ref. 12
).
The discoveries of these pathways for converting "wrongly" aminoacylated tRNAs to the cognate amino acid-tRNAs obviously speak to the evolution of the synthetases themselves, and indicate that their history is far more complex and thus more interesting than once imagined. Most recently, comparative phylogenetic studies have led Carl Woese and colleagues to the conclusion that this newly discovered pathway for cysteinyl-tRNA formation (Fig. 1C
) may have existed as far back as the last universal common ancestor (13)
. It is possible that other cases of these "unorthodox" pathways of misaminoacylation followed by enzymatic correction remain to be discovered.
It is now is clear that the original "standard model" of one, high specificity cognate synthetase for each amino acid was far from the full story. Thus, as with these attached amino acids, our understanding has been corrected. We have Söll and Woese and their respective colleagues to thank.
REFERENCES
Related Articles
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
