Saccharomyces cerevisiae cells are killed by zymocin, a tRNase ribotoxin complex from Kluyveromyces lactis, which cleaves anticodons and inhibits protein synthesis. Zymocin’s action requires specific chemical modification of uridine bases in the anticodon wobble position (U34) by the Elongator complex (Elp1-Elp6). Hence, loss of anticodon modification in mutants lacking Elongator or related KTI (K. lactis Toxin Insensitive) genes protects against tRNA cleavage and confers resistance to the toxin. Here, we show that zymocin can be used as a tool to genetically analyse KTI12, a gene previously shown to code for an Elongator partner protein. From a kti12 mutant pool of zymocin survivors, we identify motifs in Kti12 that are functionally directly coupled to Elongator activity. In addition, shared requirement of U34 modifications for nonsense and missense tRNA suppression (SUP4; SOE1) strongly suggests that Kti12 and Elongator cooperate to assure proper tRNA functioning. We show that the Kti12 motifs are conserved in plant ortholog DRL1/ELO4 from Arabidopsis thaliana and seem to be involved in binding of cofactors (e.g. nucleotides, calmodulin). Elongator interaction defects triggered by mutations in these motifs correlate with phenotypes typical for loss of U34 modification. Thus, tRNA modification by Elongator appears to require physical contact with Kti12, and our preliminary data suggest that metabolic signals may affect proper communication between them.

The genetic code conceals a ‘code within the codons’, which hints at biophysical interactions between amino acids and their cognate nucleotides. But research over decades has failed to corroborate systematic biophysical interactions across the code. Using molecular dynamics simulations and NMR, we have analysed interactions between the 20 standard proteinogenic amino acids and four RNA mononucleotides in three charge states. Our simulations show that 50% of amino acids bind best with their anticodonic middle base in the -1 charge state common to the backbone of RNA, while 95% of amino acids interact most strongly with at least one of their codonic or anticodonic bases. Preference for the cognate anticodonic middle base was greater than 99% of randomized assignments. We verify a selection of our results using NMR, and highlight challenges with both techniques for interrogating large numbers of weak interactions. Finally, we extend our simulations to a range of amino acids and dinucleotides, and corroborate similar preferences for cognate nucleotides. Despite some discrepancies between the predicted patterns and those observed in biology, the existence of weak stereochemical interactions means that random RNA sequences could template non-random peptides. This offers a compelling explanation for the emergence of genetic information in biology.

In addition to the canonical loss-of-function mutations, mutations in proteins may additionally result in gain-of-function through the binary activation of cryptic ‘structural capacitance elements’. Our previous bioinformatic analysis allowed us to propose a new mechanism of protein evolution - structural capacitance – that arises via the generation of new elements of microstructure upon mutations that cause a disorder-to-order (DO) transition in previously disordered regions of proteins. Here we propose that the DO transition is a necessary follow-on from expected early codon-anticodon and tRNA acceptor stem-amino acid usage, via the accumulation of structural capacitance elements - reservoirs of disorder in proteins. We develop this argument further to posit that structural capacitance is an inherent consequence of the evolution of the genetic code.

As a matter of course, the first gene must be formed in the absence of gene. On the other hand, many biopolymers including gene are produced under the genetic system in extant organisms. Thus far, no idea explaining how the first gene was formed in the absence of gene, has been proposed except the idea based on the GADV hypothesis. GADV means four amino acids; Gly [G], Ala [A], Asp [D] and Val [V]. In this article, a reliable answer to the question, how the first gene was generated, is provided. The idea is as follows. The first gene (genetic information) was formed by random joining of anticodons, GNCs, which were carried by primeval anticodon stem-loop (AntiC-SL) tRNAs produced during repeated cycles of random joining of nucleotides and degradation of oligonucleotides. However, it might be difficult for many persons to accept the idea, because tRNAs and metabolic pathways using various proteins, which are produced under genetic functions, must be used to form the first gene. Then, it is reconsidered based on definition of genetic information and minimum necessary things to generate the first gene in this article, how the first gene was generated. Consequently, it could be reconfirmed that five prototypes of members (genetic code, tRNA, metabolism, cell structure and protein) composing the fundamental life system, except gene, are certainly necessary to generate the first gene. Namely, it can be concluded that the first gene was formed through nucleotide metabolism with immature [GADV]-proteins, primeval AntiC-SL tRNA formation with immature [GADV]-proteins, formations of single-stranded (GNC)n codon sequence by random joining of anticodons, GNCs, carried by AntiC-SL tRNAs, double-stranded (ds)-(GNC)n RNA and maturation of an immature [GADV]-protein, which was produced from one strand of the ds-RNA. Thus, the first gene was formed using the five primitive or prototypes of members, which were produced in the absence of gene. There would be no other way for generating the first gene.