An asymmetric underlying rule in the assignment of codons: possible clue to a quick early evolution of the genetic code via successive binary choices

Abstract

Aminoacyl-tRNA synthetases (aaRSs) are responsible for creating the pool of correctly charged aminoacyl-tRNAs that are necessary for the translation of genetic information (mRNA) by the ribosome. Each aaRS belongs to either one of only two classes with two different mechanisms of aminoacylation, making use of either the 2'OH (Class I) or the 3'OH (Class II) of the terminal A76 of the tRNA and approaching the tRNA either from the minor groove (2'OH) or the major groove (3'OH). Here, an asymmetric pattern typical of differentiation is uncovered in the partition of the codon repertoire, as defined by the mechanism of aminoacylation of each corresponding tRNA. This pattern can be reproduced in a unique cascade of successive binary decisions that progressively reduces codon ambiguity. The deduced order of differentiation is manifestly driven by the reduction of translation errors. A simple rule can be defined, decoding each codon sequence in its binary class, thereby providing both the code and the key to decode it. Assuming that the partition into two mechanisms of tRNA aminoacylation is a relic that dates back to the invention of the genetic code in the RNA World, a model for the assignment of amino acids in the codon table can be derived. The model implies that the stop codon was always there, as the codon whose tRNA cannot be charged with any amino acid, and makes the prediction of an ultimate differentiation step, which is found to correspond to the codon assignment of the 22nd amino acid pyrrolysine in archaebacteria.

FIGURE 1.

Main differences in the two aminoacylation mechanisms as seen in the present-day two classes of aaRSs (Eriani et al. 1990). The primary sites of aminoacylation for Class I aaRS (2′OH, green, left) and Class II aaRS (3′OH, red, right) are highlighted, corresponding to two different types of tRNA recognition through the minor groove (Class I) or the major groove (Class II) of the tRNA. The members of each class of aaRSs that stand out as exceptions in their own class are printed with a different color (PheRS and TyrRS) as discussed in the text.

FIGURE 2.

The binary partition of the genetic code deduced from the aminoacylation mechanism of each corresponding tRNA. (Left half). The 2′OH mechanism (Class I) is in green, and the 3′OH (Class II) mechanism is in red. Ambiguous cases are both red and green, in the hash mode. The stop codon box is in white. Two variants of the mitochondrial codes have been adopted (printed in blue), so as to reduce the number of codons to 32 (only the pyrimidine/purine character of the third base matters), leaving only one stop codon. (Right half). Same as left half but with a different order for the bases of the three positions of the codon: U and C are permuted for the second base, and A, G, C, U order is adopted for the first base, instead of the usual U, C, A, G order. In addition, the rare AGR codons have been assigned to Gly/Ser as in most mitochondrial variants of the code (printed in white).

FIGURE 3.

Histogram of distances of random binary partitions of the codon table with the closest asymmetric distribution. Some 10⁷ different binary partitions of the table of codons were generated randomly, with a probability of 1/2 for each color. For each one of them, the distance to the closest perfectly asymmetric table was recorded, scanning all 24 × 24 × 2 × 6 × 2 = 13,824 possible ones, and a histogram was built.

FIGURE 4.

A series of binary choices leading to the colored table of codons as in Figure 2 (left half). At each differentiation step, one daughter is a copy of the original phenotype (dark blue), while the other daughter (light blue) will differentiate at the next generation into two codons of different colors (green and red). From then on, red or green codons can never switch back. Only a few instances of the fifth differentiation are shown because of space limitations. There is a unique order of differentiation that reproduces the pattern of Figure 2. The two exceptions in comparing Figure 2 (left half) and this figure are Asp and possibly Phe codons, although the latter one can be considered both green (2′OH) and red (Class II sequence motifs and tRNA recognition mode). The underlying mechanism implies that the dark blue codon is the stop codon (i.e., the one that can be charged neither on the 2′OH nor on the 3′OH sites) while the light blue codon is an ambiguous one that can be charged on both 2′OH and 3′OH of A76 of tRNA.

FIGURE 5.

Decoding each codon in a single (color) bit. Upon each differentiation step, the daughter on the left receives a “1,” the daughter on the right receives a “0.” The order of the bits is the one encountered when descending down the tree (two bits for second base, then two bits for first base). The color bit is the bit immediately after the first nonzero bit, reading from left to right: 1 is for red and 0 is for green. Only four divisions are shown here, but the same rule is valid if there are more divisions. The special roles of the stop codon and the ambiguous codon can also be deduced from the same rule.