Figure 1. The codon sun. The first codon position is in the centre, the second and third one follow concentrically. The initiation codon (start) and the stop codons are indicated. *, Arg, Leu and Ser are decoded by six codons each.

Figure 2. Types of molecular recognition. Type 1: an antibody binds its substrate (pink) at a region (thick red line) that is specific for the substrate and is indifferent to competing substrates. Therefore, the free energy of binding is almost identical to the discriminating energy. Type 2: most of the surface of a ribosomal protein inside the ribosome can be regarded as the binding region. Therefore, the binding energy is still more or less identical with the discriminating energy. The large discriminatory energy drives complicated processes such as ribosomal assembly. Type 3: this is the unfavourable case, where the discriminating energy is only a small fraction of the total binding energy. The large nondiscriminating binding area (green dotted line) is identical for different substrates.

Figure 3. The three classes of ternary complexes (only tRNAs are shown) that compete for the codon in the decoding centre of the A site located on the small ribosomal subunit. A mRNA runs through the ribosome where the nucleotides are shown with a binary code (red and blue) that are considered as being complementary to each other. A hypothesis is presented explaining the importance of the reciprocal linkage between the A and the E sites for both fast and accurate protein synthesis. If the A site is a high-affinity state and the E site free (a) all ternary complexes interfere with the selection process. An occupied E site induces a low-affinity A site (b), abolishing the nondiscriminatory interactions, thus preventing interference with the noncognate complexes (90% of all complexes).

Figure 4. Principles of decoding in the A site of the ribosome. (a) The first base pair of codon–anticodon interaction (position 1) exemplifies as Type I A-minor motif: A1493 recognizes the minor groove of the A36–U1 base pair via H-bonds. (b) Position 2 illustrates a Type II A-minor motif: A1492 and G530 acting in tandem to recognize the stereochemical correctness of the A35–U2 base pair using H-bonds. (c) The third (or wobble) base pair (G34–U3) is less rigorously monitored. C1054 stacks against G34 while U3 interacts directly with G530 and indirectly with C518 and proline 48 of S12 through a magnesium ion (magenta). All nucleotides involved in monitoring positions 1 and 2 are universally conserved. Adapted from Ogle et al. (1993).

Figure 5. A simple model for kinetic proofreading. The transition from the low- to the high-affinity state of the A site (rate k₂) is thought to be slow compared to the decoding reaction (recognition of the partial Watson–Crick structure of the codon–anticodon interaction) governed by k₁ and k_–1. Therefore, the decoding reaction runs under equilibrium conditions and fully exploits the discrimination potential of the decoding reaction. With k_a > k₂ the concentration of the ribosomal complex with the near-cognate ternary complex will decay with the rate up to k_a, thus increasing the accuracy. Rounded ribosomes are in the post-translocational (POST) state and rectangular ribosomes in the pre-translocational (PRE) state. The aminoacyl-tRNA (with the filled circle) in complex 2 and 2¢ is loosely bound to the low-affinity A site and in complex 3 to the high-affinity A site.