Structural analysis, supported by biochemical, mutagenesis and computational evidence, indicates that the peptidyltransferase centre of the contemporary ribosome is a universal symmetrical pocket composed solely of rRNA. This pocket seems to be a relic of the proto-ribosome, an ancient ribozyme, which was a dimeric RNA assembly formed from self-folded RNA chains of identical, similar or different sequences. This could have occurred spontaneously by gene duplication or gene fusion. This pocket-like entity was capable of autonomously catalysing various reactions, including peptide bond formation and non-coded or semi-coded amino acid polymerization. Efforts toward the structural definition of the early entity capable of genetic decoding involve the crystallization of the small ribosomal subunit of a bacterial organism harbouring a single functional rRNA operon.
- Mycobacterium smegmatis
- symmetrical region
The ribosome is the universal naturally occurring ribozyme that performs translation of the genetic code into nascent proteins. These cellular catalytic machines are multi-component riboproteins of molecular mass 2.5 and 4 MDa (for prokaryotic and eukaryotic sources respectively). Remarkably, despite the size difference, the functional regions of ribosomes, the decoding centre and the PTC (peptidyltransferase centre) composed solely of rRNA, are highly conserved across all domains of life.
Involvement of RNA-rich particles in genetic expression was suggested over five decades ago [1,2], when RNA-rich ‘Palade particles’ were observed in cells, in close proximity to the endoplasmic reticulum membrane [1,3]. The discovery of the ribosomes and the realization of their complexity stimulated attempts to define their precursors, by comparative sequence analogy , in vitro selection , functional and mutagenesis investigations [6,7], and analyses of the nature of internal ribosome interactions [8–10]. These studies illuminated key ribosomal properties and stimulated hypotheses on the way the ribosome developed, but did not reveal the entity that could be assigned as the proto-ribosome.
In the present paper, we present arguments supporting the existence of ancestral dimeric entity that may have acted as machinery capable of catalysing various reactions, including peptide bond formation.
Peptide bonds are formed in a symmetrical region within the asymmetric ribosome
The heart of the contemporary ribosome hosts the PTC, i.e. the site of peptide bond formation. This centre is located in the midst of a symmetrical ‘pocket-like’ structure [11–17] (Figure 1) that encompasses 180 nucleotides of which the backbone folds, irrespective of the nucleotide sequences, are related by an internal pseudo-2-fold symmetry. This is an extremely unusual feature, as the ribosome, composed of over 4500 nucleotides and over 50 proteins, is highly asymmetric. This internal symmetry exists in all known high-resolution ribosome structures [18–22] and its exceptionally high sequence conservation indicates its ancient origin.
The architecture of this structural element provides the framework for all of the ribosome's catalytic contributions towards peptide bond formation, since it positions the ribosome's substrates in favourable stereochemistry for peptide bond formation and substrate-mediated catalysis. Furthermore, by encircling the PTC, the symmetrical region confines the void required for the motions involved in the translocation of the 3′-end of the aminoacylated tRNA from the A-site into the site of the peptidylated tRNA molecules (P-site). Hence the symmetrical region also provides the mechanism facilitating the polymerase activity of the ribosome [11,13,14].
On the basis of the detection of the internal pseudo-symmetry within the ribosomal catalytic region, we have proposed that, as in modern life, the ancient machine for forming peptide bonds was made exclusively from RNA. It is widely accepted that such RNA oligomers could have formed in the primordial soup [23–25] by various mechanisms, including chemical (non-enzymatic) RNA synthesis and polymerization under conditions mimicking those assumed to govern the prebiotic era [26–28].
It appears that the ancient short RNA chains acquired common motifs that were sufficiently stable to survive environmental alterations and evolutionary stress. Such motifs have been identified in various natural ribozymes, some of which are believed to be relics from the prebiotic world. These include gene regulators, riboswitches, RNA polymerases and ribozymes catalysing phosphodiester cleavage, RNA processing and RNA modification .
Emergence of the proto-ribosome
Among the existing RNA chains, those forming entities that could later evolve to become the precursor of the proto-ribosome underwent self dimerization and acquired a ‘pocket-like’ configuration, capable of hosting variety of substrates. Subsequently, RNA chains of lengths and compositions sufficient for the construction of basic elements of the modern ribosome active site may have existed in the prebiotic period, becoming like pocket-like molecular machines with either non-specific or specialized catalytic capabilities [30,31].
Dimerization in a symmetrical manner could have occurred spontaneously, utilizing chemical complementarity obtained by tertiary interactions, such as the common GNRA (guanine–any nucleotide–a purine–adenine) system  that includes the abundant and ubiquitous ‘A-minor’ structural motif [9,33,34]. This seemingly simple process could be assisted by other molecules acting as small chaperones  offering stabilization. Additional structural support could be obtained from short peptides that have high affinity for RNA , or from longer poly-amino acids, similar to protein–RNA interactions within the contemporary ribosome, or from interactions of low-molecular-mass compounds with exposed regions of RNA. These short peptides could have been produced by uncoded elongation by early versions of the proto-ribosome or by other means. Following this mechanism, a proto-ribosome that produced peptides with amino acid compositions that provided with better fitness, thus had higher probability of being retained.
On the basis of the presumed ability of the prebiotic construct to enable peptide bond formation, it is suggested that the symmetrical region of the contemporary ribosome originated from a symmetrical proto-ribosome [14,15,17,37]. Its substrates could have been activated amino acids or spontaneously produced amino acids conjugated with single or short oligonucleotides [38–41].
What could have been the driving force for the construction of a machine that forms peptide bonds in the RNA world? It is conceivable that a pocket-like entity existed in the RNA world for performing required chemical reactions, involving mainly RNA molecules. An examples could be a primitive RNA replisome, an entity that was proposed to be the ancestor of the ribosome  and could perform RNA polymerization and/or simple RNA replication. Similarly, the original ancient machines could have been an integral part of the RNA metabolic pathways  within some sort of a proto-cell. Regardless of the tasks of these hypothetical ancient molecular machines, it appears that polymerization of amino acids became the task of highest preference. Consequently, stable pocket-like molecular dimers capable of peptide bond formation were evolutionarily favoured, surviving under various environmental conditions, and later becoming the ancestor of the active site of the contemporary ribosome. As it is assumed that, in the prebiotic era, RNA chains could self-replicate [44–48], it is conceivable that phenotypes with favourable properties could have been synthesized in many copies. It is therefore likely that some of these phenotypes originated by the fusion of two different or duplication of two identical sequences.
We assessed the feasibility of a dimeric proto-ribosome capable of peptide bond formation biochemically, by testing the tendency of various RNA chains to dimerize and to form peptide bonds. Among the various RNA sequences that have been constructed, several, but not all, RNA chains with sequences resembling those observed in the current ribosome, are capable of forming dimers that may adopt a pocket-like structure . Specifically, so far, a marked preference was detected for the dimerization of sequences resembling the PTC P-site, including those that underwent mutational alterations by in vitro site-directed mutagenesis. This noticeable preference may indicate higher stability of this portion of the symmetrical region, or a higher significance in the process of peptide bond formation, in accordance with the contemporary accommodation of the initial tRNA at the P-site of the PTC.
Emergence of coded translation
What would have been the selective benefit for a ribozyme functioning in the RNA world to begin the process of amino acid ligation? It is clear a posteriori that proteins are much better catalysts than their ribozyme cousins. However, evolution has no ability to discern future functionality from its present state. Accordingly, there must have been an immediate reason for the production of short ‘uncoded’ peptides by an emerging proto-ribosome.
It has been postulated that, among the first amino acids to be incorporated into a primeval genetic code, were the amino acids with the highest ‘catalytic potential’ . This hypothesis yields insight into what type of amino acids would have been preferentially elongated by the proto-ribosome, as they presumably gave enough selection advantage to warrant the formation of their coded elongation. Encouragingly, these same amino acids (histidine, lysine, arginine) are the ones that interact strongly with RNA in extant biology.
It is conceivable, then, that the proto-ribosome would have initially favoured peptidyl transfer and elongation of histidine, lysine and arginine because of the favourable electrostatic and hydrogen-bonding interactions that these amino acids exhibit with RNA. Upon elongation by the proto-ribosome, a pool of short peptides would have been generated, some of which would be rich in the amino acids that can be attracted to the negatively charged RNA. Plausibly, a small portion of this peptide pool would have displayed the ability to bind to the ribozyme that produced them. Of these RNA-binding peptides, some would have stabilized further the RNA structure, or caused it to adopt an even more active conformation for peptidyl transfer (Figure 2).
Simultaneously, the small peptide pool generated by a proto-ribosome must have had the ability to assist a portion of the RNA machinery that was in the ribocyte (a postulated ancient RNA-based cell), as observed in modern ribozymes that are supported by protein assemblies that enhance their reactivity (i.e. group I intron ribozyme) . Hence a proto-ribosome that could manufacture a set of small peptides that stabilized its own function and/or assist the other surrounding ribozymes would have imparted a significant selection advantage.
Clearly, in modern life, only a small portion of proteins functionally interact with RNA, whereas the majority of them are self-sufficient in their function. Modern enzymes adopt highly specific tertiary structures to achieve their high rates of catalysis, but these complicated folded structures would not have been available to the emerging proto-peptide pool. Probably, the preferentially elongated amino acids were the positively charged histidine, arginine and lysine, which are also rich in chemical reactivity. Although, presumably, these peptides were unfolded (or largely unstructured), those rich in these moieties could have enriched their neighbourhood and served as the first non-RNA catalysts . For example, imidazole rings (of histidine) can act as a proton-transfer agent and nucleophilic catalyst , guanidine (found in arginine) can stabilize transitions states by favourable hydrogen-bonding interactions [52,53], and the primary amine in lysine can act as general acid–base catalyst. Although the reaction rates of these small peptide catalysts would have been incomparable with modern enzymatic catalysts. The RNA world was a ‘slow-living entity’ , so the addition of any new chemical functionality would have been vital to add selection advantage to this primordial life.
The transition from a molecule forming peptide bonds to an elaborate apparatus capable of decoding genetic information was coupled with the evolving genetic code and the development of the substrates. A higher level of efficiency could have been achieved by the increase of the proto-ribosome size and the creation of a supporting environment coupled with the elongation of the substrates forming the minimal mono- or oligo-nucleotide–amino acid conjugates towards the modern tRNA. The conversion into longer compounds with a contour that can complement the inner surface of a larger reaction pocket occurred concurrently with proto-ribosome mutational optimization, aimed at accurate substrate positioning that, in the modern ribosome, is governed by remote interactions beyond the PTC [11,54].
Later stages of this development were coupled with the advance from non- or pseudo-genetic peptide bond formation towards performing genetically driven translation and with the optimization of selected portions of the evolving machine for their specific tasks. Additionally recruited structural elements, which could be oligopeptides or oligonucleotides, interacted with the proto-ribosome and its surroundings in a manner resembling the protein–RNA interactions in the modern ribosome. Mutational optimization that facilitated distinction between the two sides of the active site allowed differentiation between the two substrates. Besides functional optimization, the nucleotide identities and conformations evolved for enhancing the stability of the symmetrical region. Consequently, orientations of the RNA bases within the contemporary symmetrical region are not related by the internal symmetry, consistent with the finding that, in the contemporary ribosome, the symmetry is related to the backbone fold and not to the nucleotide sequence, thus emphasizing the superiority of functional requirement over sequence conservation. With the selection pressure for increased stability and efficiency, the proto-ribosome served as the entity providing all the activities required for non-coded nascent protein elongation. The modern ribosome could have evolved gradually around the symmetrical region until it acquired its final shape, either hierarchically  or by another mechanism (e.g. ).
The growing complexity of the ribosome in size and shape allowed it to perform programmed translation. Encouraged by the identification of the ancient peptide bond formation machinery within the large ribosomal subunit, we searched for remnants of a primitive decoding region within the small subunit. Because decoding does not require symmetry or any other distinct structural motif, unambiguous identification of the ancestor region was so far not possible. We therefore selected the small ribosomal subunit from Mycobacterium smegmatis, a strain of a bacterial organism that was genetically manipulated to harbour a single functional rRNA operon , in order to approach this issue using mutagenesis combined by X-ray crystallography. The microcrystals (Figure 3) obtained so far indicated feasibility of crystallographic studies, as they are of a quality comparable with the first ribosomal crystals, obtained almost 30 years ago , which could be improved to yield high-resolution structures.
Support was provided by the National Institutes of Health [grant number GM34360], the German–Israeli Foundation [grant number G-853-284.9/2004] and the Kimmelman Centre for Macromolecular Assemblies. C.D. is supported by the Adams Fellowship Program of the Israel Academy of Sciences and Humanities. A.Y. holds the Martin and Helen Kimmel Professorial Chair.
Thanks are due to all members of the ribosome group at the Weizmann Institute for continuous interest, Shosh Tel-Or and Donna Matzov for technical assistance, the electron microscope unit, especially Sharon Wolf and Helena Sabanay, for providing experimental support and to Dr Erik Bottger for encouraging the M. smegmatis crystallization. Crystallographic data were collected at ID23-2, European Synchrotron Radiation Facility, Grenoble, France.
Machines on Genes: Enzymes that Make, Break and Move DNA and RNA: A Biochemical Society Focused Meeting held at Robinson College, Cambridge, U.K., 12–14 August 2009. Organized and Edited by Richard Bowater (University of East Anglia, U.K.), Ben Luisi (University of Cambridge, U.K.) and Marshall Stark (University of Glasgow, U.K.).
Abbreviations: PTC, peptidyltransferase centre
- © The Authors Journal compilation © 2010 Biochemical Society