A scenario on the stepwise evolution of the genetic code

Abstract

It is believed that in the RNA world the operational (ribozymes) and the informational (riboscripts) RNA molecules were created with only three (adenosine, uridine, and guanosine) and two (adenosine and uridine) nucleosides, respectively, so that the genetic code started uncomplicated. Ribozymes subsequently evolved to be able to cut and paste themselves and riboscripts were acceptive to rigorous editing (adenosine to inosine); the intensive diversification of RNA molecules shaped novel cellular machineries that are capable of polymerizing amino acids-a new type of cellular building materials for life. Initially, the genetic code, encoding seven amino acids, was created only to distinguish purine and pyrimidine; it was later expanded in a stepwise way to encode 12, 15, and 20 amino acids through the relief of guanine from its roles as operational signals and through the recruitment of cytosine. Therefore, the maturation of the genetic code also coincided with (1) the departure of aminoacyl-tRNA synthetases (AARSs) from the primordial translation machinery, (2) the replacement of informational RNA by DNA, and (3) the co-evolution of AARSs and their cognate tRNAs. This model predicts gradual replacements of RNA-made molecular mechanisms, cellular processes by proteins, and informational exploitation by DNA.

Fig. 1

The R–Y (A–U) code (A) and its encoded amino acids (B). R and Y stand for purine and pyrimidine, respectively. Start and stop codons are indicated with Sr and St, respectively. RYR or AUA is assumed as start codon rather than encodes isoleucine since RYY or AUU already does so.

Fig. 2

The extended early code in two scenarios: (A) incorporation of G but avoiding AG and GU (both dinucleotides were used as splicing signals) and (B) extended through base editing from A to I. The codons overlapping with splice signals and the original codons are underlined.

Fig. 3

The organization of the genetic code and AARSs. The code is divided into two halves, pro-diversity (unshaded area) and pro-robustness (shaded area), according to sensitivity of the codons to purine (AG) content changes. AARSs are also divided into two types and the Type II enzymes are underlined. There are both types of AARSs for lysine although it is underlined in this figure. After C was recruited as an essential building block, the code was extended to include more amino acids with its C-containing codons. The rule of extension for AARSs followed a G–C conversion trend except the six-fold codons (L, R, and S). For instance, the pairs, such as CAR and GAR, CAY and GAY, GCN and GGN, share the same class of AARSs.

Fig. 4

Hypothetical schemes illustrating how RNAs were spliced into informational and operational molecules in the RNA world with the involvement of cellular machineries, such as spliceosome and reverse-transcriptosome (A). In the DNA–RNA–protein world, replisome and repairosome were created for managing DNA processes (B).

Fig. 5

Multiple sequence alignments of the RNA-binding domain in ribonuclease 3. The sequences were retrieved from public databases from the top to the bottom as abbreviated names of the following: Pseudomonas solanacearum, Dechloromonas aromatic, Shigella boydii serotype 4, Psychrobacter arcticum, Ehrlichia canis, Listeria monocytogenes serotype 4b, Staphylococcus aureus, Dehalococcoides ethenogenes, Dehalococcoides sp., Desulfotalea psychrophila, Aquifex aeolicus, Leptospira interrogans, Fusobacterium nucleatum, Treponema denticola, Campylobacter jejuni, Chlorobium chlorochromatii, Mimivirus, Treponema pallidum, Mycobacterium leprae, Rhodopirellula baltica, Caenorhabditis elegans, human, Tropheryma whipplei, Anabaena sp., Chlamydia trachomatis, and yeast. Strictly conserved and less strictly conserved amino acids are indicated with stars and solid doubled dots, respectively.