RNA evolution conjectured from tRNA and riboswitches


The structures of tRNAs and riboswitches comprise distinct domains. A tRNA molecule is L-shaped, and each arm may reflect its evolution. The amino acid-nonspecific minihelix domain may have emerged before the amino acid-specific anticodon-containing domain was acquired. The glycine riboswitch of Bacillus subtilis functions in a “glycine-independent” manner in the presence of polyethylene glycol or ethylene glycol. The effect is dependent only on the existence of a terminator stem within the expression platform of the riboswitch and is independent of the aptamer domain. Similar to the hypothesis that explains the evolution of tRNA function, the specificity of riboswitch functional domains may have increased during evolution. Collectively, these views provide new perspectives on the evolution of RNA.


RNA plays crucial roles in current biological systems. Crick’s prediction of an adaptor molecule that mediates translation of messenger RNAs (mRNAs) (1) was proven by Zamecnik and collaborators’ discovery of transfer RNA (tRNA) (2). Protein biosynthesis occurs on the ribosome, which is composed of ribosomal RNAs (rRNAs) and proteins, according to the information on mRNA. mRNA, tRNA and rRNA are related to the expression of genetic information and they are produced through RNA processing. RNA splicing is a typical reaction of RNA processing and the discovery of self-splicing RNA (3) gave rise to the concept of the ribozyme. The discovery of other RNAs such as micro-RNAs dramatically expanded our knowledge of the functions of noncoding RNAs (4).

The property of self-replication is a crucial characteristic for defining life itself and must be defined in detail. RNA, in contrast to protein, may self-replicate due to intrinsic interactions mediated through Watson-Crick base pairing. Moreover, the discovery of ribozymes (3,5) strongly suggests the existence of an “RNA world” as the initial form of life on Earth (6). Here, the size of RNA and its possible evolutionary pathway are the main issues in considering the formation of functional RNA in the RNA world as a fundamental process underlying life. This article focuses on the modular structures of tRNAs and riboswitches as well as their evolutionary relationships.

Origin of self-replicating RNA
What was the maximum size of self-replicating RNAs upon their emergence? As the size of these RNAs increases, their complexity and stored information also increase. Errors in replication cause so-called “error catastrophe,” which is often cited in the extinction of an organism because of excessive RNA mutations (7). During enzyme-free nucleotide polymerization, the chain length required for accurate replication under Darwinian selection is no more than approximately 100 nucleotides when the intrinsic molecular properties of nucleic acids are considered. Interestingly, this size corresponds to that of tRNA. To increase the length of the nucleic acid to ensure accurate replication, a set of cooperating self-replicators, such as a “hypercycle” (8), is required. Thus, when considering the evolution of RNA, at least before the advent of the hypercycle self-replicating system, RNAs approximately 75 nucleotides long (the size of a tRNA) may have existed during Earth’s early history.

Origin of tRNA and its function
The L-shaped three-dimensional structure of tRNA is enabled by tertiary interactions between D- and T-arms. Each half of the “L” is composed of acceptor stem plus T-stem, and D-stem plus anticodon stem, respectively (9,10). The termini of arms of the L are separated by approximately 75 Å, and an aminoacylation site is present on one and an anticodon on the other.

One half of the L-shaped tRNA structure is often called a “minihelix” (11), and the domains, even when isolated, are substrates for aminoacylation by many aminoacyl tRNA synthetases (aaRSs) (12,13,14). These enzymes are classified into two groups according to their amino acid sequences and structures of their catalytic domains (15). Moreover, the tertiary structures of aaRSs reveal structural and functional correlations with tRNAs. Aminoacylation of tRNA occurs through the activated aminoacyl adenylate form of amino acids (16). The domains of aaRSs required for amino acid activation are structurally similar and conserved within the same classes, whereas the anticodon-binding domains comprise diverse nonconserved structures. The top half of the L-shaped tRNA structure (minihelix) interacts with the conserved domains of aaRSs, while the bottom half of the L (anticodon-containing domain) interacts with the nonconserved domains of aaRSs (Figure 1).

Aminoacylated tRNAs participate in peptide synthesis on the ribosome by decoding mRNA triplets through codon-anticodon interactions. Precise codon-anticodon interactions ensure the fidelity of the genetic code. Analogous to the structure and function of aaRSs, the large and small ribonucleoprotein subunits are functionally separated (17). The minihelix domain of tRNA interacts with the large subunit where peptide bond formation occurs between the peptidyl-tRNA and the aminoacyl-tRNA situated at the peptidyl transferase center (PTC). The amino group of an amino acid of the latter attacks the carbonyl carbon of the former, which occurs in a manner that is not specific for an amino acid. The PTC is composed exclusively of RNA molecules (18,19), proving that the ribosome is a ribozyme (20). Conversely, the anticodon-containing domain of tRNA interacts with the small subunit and serves to decode the mRNA triplets through codon-anticodon binding. This interaction contributes to amino acid-specific translation (Figure 1).


Figure 1 | Evolution of tRNAs, aaRSs, and ribosomes. The minihelix (half domain of tRNA with the amino acid attachment site) interacts with the conserved domain of aaRSs for amino acid activation, and with the large ribosomal subunit for peptide bond formation. The other half domain of tRNA interacts with the nonconserved domain of aaRSs for the specific recognition of the anticodon, and with the small ribosomal subunit for decoding the mRNA triplets through codon-anticodon interactions. Thus, the minihelix may have evolved to the current form of tRNA by acquiring an anticodon-containing domain.


Thus, the domain structures and functions of tRNA synthetases and ribosome subunits are closely related to each half of the L-shaped tRNA structure. These two helical arms of the “L” possibly appeared at different times during evolution, and the minihelix may have preceded the anticodon-containing domain (21-29). Consistent with this hypothesis, the formation of tRNAs by combinations of all possible nucleotide sequences (475) is inconceivable considering the huge mass required (one percent of that of the entire Earth).

Evolution of riboswitches
The discovery of riboswitches is significant because they are “natural” RNA aptamers (molecules that bind to specific target molecules). The riboswitch is located within the untranslated regions of the mRNA and functions as a cis-acting RNA-based genetic control apparatus (30,31,32). Structural changes that occur following the binding of various ligands to riboswitches cause transcriptional or translational ON/OFF switching (33,34).

Attention should be paid to the variety of ligands and the contrasting functional uniformity of riboswitches through evolution. Riboswitches generally comprise an aptamer region and an expression platform. The structures of the aptamer region vary according to the type of ligand. In contrast, structures of the expression platform are less variable.

Here, using a glycine riboswitch as an example, I consider the recent discovery of riboswitches and how they evolved. The two similar tandemly arranged aptamer regions in the glycine riboswitch of Bacillus subtilis recognize glycine (33). The binding of glycine through the conformational change of a putative terminator stem on the expression platform, increases the transcription of the gcvT operon located downstream, which eventually affects the expression of enzymes involved in glycine metabolism (33).

Although the crystal structure of the aptamer regions of the glycine riboswitch of Fusobacterium nucleatum reveals the glycine-binding site (35,36), the regulatory mechanism underlying the effect induced by the expression platform is unknown. The glycine riboswitch of Bacillus subtilis functions in a “glycine-independent” manner in the presence of polyethylene glycol (PEG) or ethylene glycol (EG) (37). Synthesis of a full-length transcript is inhibited by PEG and facilitated by EG and occurs as well using an aptamer-deleted template. These findings indicate that the formation of a terminator stem within the expression platform, irrespective of the presence of the aptamer domain, determines the fate of transcription. Moreover, PEG stabilizes the structure of the terminator stem while EG destabilizes it (37). EG is one of the most abundant polyols found in meteorites (38), suggesting that under specific conditions present during the Earth’s early history, polyols may have played a crucial role in inducing structural changes of RNAs before RNA specificity was established (Figure 2).

HJ380_IMG2Figure 2 | Evolution of riboswitches. A simple hairpin loop structure may have been utilized for simple conformational changes (e.g. stem-structure formation/deformation) induced by solutes such as ethylene glycol (EG). Diverse ligand-binding aptamer domains may have been added to these primitive expression platforms. For example, the glycine riboswitch of B. subtilis exhibits “glycine-independency” in the presence of PEG or EG37, which suggests the transition from primitive ligand-independent riboswitches to the contemporary ligand-dependent form.


Relation of riboswitches to tRNA evolution
The evolution of tRNA likely proceeded from a minihelix hairpin by the addition of an anticodon-containing hairpin (Figure 1). Similarly, a commonly functioning expression platform, which e.g. consists only of the simple stem-loop structure of riboswitches, emerged first followed by the ligand-binding aptamer domains of riboswitches. Regulating transcription is a function shared by riboswitches, and their specificity is imparted by the addition of various aptamer ligand-binding regions (Figure 2). Further, the size of riboswitches (aptamer plus expression platform) exceeds the maximum size predicted by Eigen for self-replicating molecules (8).

There are similar examples of the artificial evolution of bimodular ribozymes using in vitro selection. For example, joining an ATP-binding RNA to a self-cleaving ribozyme creates a ribozyme that is allosterically regulated by ATP (39). Autocatalytic aptazymes are constructed by combining catalytic and ligand-binding aptamer domains. Two independent aptamer domains (one for theophylline and another for flavin mononucleotide) mediate ligand-dependent ligation of RNAs (40).

Taken together, these findings support the hypothesis proposed here that riboswitches and tRNA evolved through similar pathways.

Without a cooperative self-reproducing system, the self-replication of RNA proceeds up to approximately 100 nucleotides, similar to the size of tRNA. The existence form of domain structures and functional evolution of tRNAs and riboswitches are similar. Molecule-nonspecific functional domains (i.e. minihelix domain of tRNA and expression platform of riboswitches) may have preceded the generation of high specificity (i.e. the anticodon-containing domain of tRNA and aptamer domain of riboswitches, respectively). These conclusions are consistent with experimental evidence and provide new perspectives on the evolution of RNA.H

This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture, Japan (Grant No. 23657160 and No. 25291082).

Author declares no conflicts of interest.

About the author
Dr. Koji Tamura is a Professor in the Department of Biological Science and Technology, Tokyo University of Science. Research in his laboratory centers on investigations of the origin of life and the genetic code.


  1. Crick FHC. On degenerate templates and the adapter hypothesis. A note for the RNA Tie Club. 1955.
  2. Hoagland MB, Stephenson ML, Scott JF, Hecht LI, Zamecnik PC. A soluble ribonucleic acid intermediate in protein synthesis. J Biol Chem. 1958;231:241-57.
  3. Kruger K, Grabowski PJ, Zaug AJ, Sands J, Gottschling DE, Cech TR. Self-splicing RNA: auto-excision and autocyclization of the ribosomal RNA intervening sequence of Tetrahymena. Cell. 1982;31:147-57.
  4. Couzon J. Small RNAs make big splash. Science. 2002;298:2296-7.
  5. Guerrier-Takada C, Gardiner K, Marsh T, Pace N, Altman S. The RNA moiety of ribonuclease P is the catalytic subunit of the enzyme. Cell. 1983;35:849-57.
  6. Gilbert W. The RNA world. Nature. 1986;319:618.
  7. Eigen M. Selforganization of matter and the evolution of biological macromolecules. Naturwissenschaften. 1971;58:465-523.
  8. Eigen M, Schuster P. Hypercycle. A principle of natural self-organization. Part A: emergence of the hypercycle. Naturwissenschaften. 1977;64:541-65.
  9. Kim SH, Suddath FL, Quigley GJ, McPherson A, Sussman JL, Wang AH, Seeman NC, Rich A. Three-dimensional tertiary structure of yeast phenylalanine transfer RNA. Science. 1974;185:435-40.
  10. Robertus JD, Ladner JE, Finch JT, Rhodes D, Brown RS, Clark BF, Klug A. Structure of yeast phenylalanine tRNA at 3 Å resolution. Nature. 1974;250:546-51.
  11. Schimmel P, Giegé R, Moras D, Yokoyama S. An operational RNA code for amino acids and possible relationship to genetic code. Proc Natl Acad Sci USA. 1993;90:8763-8.
  12. Francklyn C, Shi J-P, Schimmel P. Overlapping nucleotide determinants for specific aminoacylation of RNA microhelices. Science. 1992;255:1121-5.
  13. Frugier M, Florentz C, Giegé R. Efficient aminoacylation of resected RNA helices by class II aspartyl-tRNA synthetase dependent on a single nucleotide. EMBO J. 1994;13:2218-26.
  14. Saks ME, Sampson JR. Variant minihelix RNAs reveal sequence-specific recognition of the helical tRNASer acceptor stem by E.coli seryl-tRNA synthetase. EMBO J. 1996;15:2843-9.
  15. Eriani G, Delarue M, Poch O, Gangloff J, Moras D. Partition of tRNA synthetases into two classes based on mutually exclusive sets of sequence motifs. Nature. 1990;347:203-6.
  16. Schimmel P. Aminoacyl tRNA synthetases: general scheme of structure-function relationships in the polypeptides and recognition of transfer RNAs. Annu Rev Biochem. 1987;56:125-58.
  17. Noller HF. On the origin of the ribosome: coevolution of subdomains of tRNA and rRNA. The RNA World, Cold Spring Harbor Laboratory Press, Plainview, NY. 1993;137-56.
  18. Ban N, Nissen P, Hansen J, Moore PB, Steitz TA. The complete atomic structure of the large ribosomal subunit at 2.4 Å resolution. Science. 2000;289:905-20.
  19. Nissen P, Hansen J, Ban N, Moore PB, Steitz TA. The structural basis of ribosome activity in peptide bond synthesis. Science. 2000;289:920-30.
  20. Noller HF, Hoffarth V, Zimniak L. Unusual resistance of peptidyl transferase to protein extraction procedures. Science. 1992;256:1416-9.
  21. Di Giulio M. On the origin of the transfer RNA molecule. J Theor Biol. 1992;159:199-214.
  22. Rodin S, Rodin A, Ohno S. The presence of codon-anticodon pairs in the acceptor stem of tRNAs. Proc Natl Acad Sci USA. 1996;93:4537-42.
  23. Schimmel P, Ribas de Pouplana L. Transfer RNA: from minihelix to genetic code. Cell. 1995;81:983-6.
  24. Tamura K. Origin of amino acid homochirality: relationship with the RNA world and origin of tRNA aminoacylation. BioSystems. 2008;92:91-8.
  25. Tamura K. Molecular handedness of life: significance of RNA aminoacylation. J Biosci. 2009;34:991-4.
  26. Tamura K. Molecular basis for chiral selection in RNA aminoacylation. Int J Mol Sci. 2011a;12:4745-57.
  27. Tamura K. Ribosome evolution: emergence of peptide synthesis machinery. J Biosci. 2011b;36:921-8.
  28. Tamura K, Schimmel P. Chiral-selective aminoacylation of an RNA minihelix. Science. 2004;305:1253.
  29. Tamura K, Schimmel PR. Chiral-selective aminoacylation of an RNA minihelix: mechanistic features and chiral suppression. Proc Natl Acad Sci USA. 2006;103:13750-2.
  30. Nudler E, Mironov AS. The riboswitch control of bacterial metabolism. Trends Biochem Sci. 2004;29:11-7.
  31. Roth A, Breaker RR. The structural and functional diversity of metabolite-binding riboswitches. Annu Rev Biochem. 2009;78:305-34.
  32. Winkler WC, Breaker RR. Regulation of bacterial gene expression by riboswitches. Annu Rev Microbiol. 2005;59:487-517.
  33. Mandal M, Lee M, Barrick JE, Weinberg Z, Emilsson GM, Ruzzo WL, Breaker RR. A glycine-dependent riboswitch that uses cooperative binding to control gene expression. Science. 2004;306:275-9.
  34. Yarnell WS, Roberts JW. Mechanism of intrinsic transcription termination and antitermination. Science. 1999;284:611-5.
  35. Butler EB, Xiong Y, Wang J, Strobel SA. Structural basis of cooperative ligand binding by the glycine riboswitch. Chem Biol. 2011;18:293-8.
  36. Huang L, Serganov A, Patel DJ. Structural insights into ligand recognition by a sensing domain of the cooperative glycine riboswitch. Mol Cell. 2010;40:774-86.
  37. Hamachi K, Hayashi H, Shimamura M, Yamaji Y, Kaneko A, Fujisawa A, Umehara T, Tamura K. Glycols modulate terminator stem stability and ligand- dependency of a glycine riboswitch. BioSystems. 2013;113:59-65.
  38. Cooper G, Kimmich N, Belisle W, Sarinana J, Brabham K, Garrel L. Carbonaceous meteorites as a source of sugar-related organic compounds for the early Earth. Nature. 2001;414:879-83.
  39. Tang J, Breaker RR. Rational design of allosteric ribozymes. Chem Biol. 1997;4:453-9.
  40. Lam BJ, Joyce GF. Autocatalytic aptazymes enable ligand-dependent exponential amplification of RNA. Nat Biotechnol. 2009;27:288-92.

Share your thoughts

Leave a Reply

You must be logged in to post a comment.