The Domain Structure of the Ribosomal Subunits

The shear complexity of protein synthesis forces any participating component to maintain their structure and function through evolution. This principle justifies the assumption that not only all tRNAs, but also all 16S (and 16S-like) and 23S (and 23S-like) rRNAs have the same general secondary and tertiary structures [55]. Therefore, the secondary structures of 16S, 23S, and 5S rRNA could be derived by analyzing the pattern of variation within aligned rRNA sequences from different species (see Figs. 2-4A and C; [56]). The resulting secondary-structure diagrams consist of a complex arrangement of A-form helices and non-helical regions (loops or bulges) [55]

In the 16S rRNA the different domains branch from a central pseudo-knot and, beginning from the 5'-end, are termed the 5', central, 3'-major and 3'-minor domains (see Fig. 2-4A). In striking contrast to the 50S subunit (see following), the domains are not interwoven in the tertiary fold and can be assigned easily to the structural landmarks of the 30S subunit (Fig. 2-4B). The 5'-domain forms the 30S body, starting from the neck of 30S subunit it goes down to the toe and finally turns back to form the shoulder. The central domain constitutes the platform, the 3'-major domain the head. The 3'-minor domain consists of h44 and h45; h44 runs down the 30S along the inter-subunit surface and returns back to the neck, followed by h45 and a single-stranded 3'-end containing the anti-Shine-Dalgarno sequence.

In the 23S rRNA secondary structure the 5'- and 3'-terminal ends are brought together to form a helix (H1 in Fig. 2-4D). Radiating from the loop of this helix are 11 stem-loop structures of differing degrees of complexity. These stem-loop structures

Figure 2-4 Secondary structures of 16S, 23S, and 5S rRNAs. (A) Secondary structure of T. thermophilus 16S rRNA, with its 5', central, 3'-major, and 3'-minor domains shaded in blue, magenta, red, and yellow, respectively. (B) Three-dimensional fold of 16S rRNA in 70S ribosomes, with its domains colored as in (A). (C) Secondary structures of T. thermophilus 23S and 5S rRNAs, indicating domains I (blue), II (cyan), III (green), IV (yellow), V (red), and VI (magenta) of 23S rRNA. The rRNAs are numbered according to E. coli. (D) Three-dimensional folds of 23S and 5S rRNAs, with their domains colored as in (C).

Figure 2-4 Secondary structures of 16S, 23S, and 5S rRNAs. (A) Secondary structure of T. thermophilus 16S rRNA, with its 5', central, 3'-major, and 3'-minor domains shaded in blue, magenta, red, and yellow, respectively. (B) Three-dimensional fold of 16S rRNA in 70S ribosomes, with its domains colored as in (A). (C) Secondary structures of T. thermophilus 23S and 5S rRNAs, indicating domains I (blue), II (cyan), III (green), IV (yellow), V (red), and VI (magenta) of 23S rRNA. The rRNAs are numbered according to E. coli. (D) Three-dimensional folds of 23S and 5S rRNAs, with their domains colored as in (C).

are bundled into six different domains and, in analogous fashion to the helices, are numbered from the 5'- to 3'-end. The last change in the assignment of the secondary structure to the various domains was contributed by crystallography. Helix 25, which was originally considered as being part of domain I, was reassigned to domain II, because it exhibits stronger interactions with this domain than with the elements of domain I [11]. The six domains of 23S and 5S rRNA all have compact shapes, which are intertwined (Fig. 2-4D). The domains form structural units, as the vast majority of

Figure 2-4 From Ref. [1] with permission. (E, F) Comparison of the current comparative structure models for the 16S and 23 S rRNAs with the corresponding ribosomal subunit crystal structures. (E) 16S rRNA versus the T. thermophilus structure (GenBank accession no. M26923; PDB code 1 FJF). (F) 23S rRNA, 5'-half and 3'-half versus the H. marismortui structure (GenBank accession no. AF034620; PDB code 1JJ2). Nucleotides are replaced with colored dots that show the sources of the interactions: red, present in both the covariation-based structure model and the crystal structure; green, present in the comparative structure and not present in the crystal structure; blue, not present in the comparative structure and present in the1 crystal structure; purple, positions that are unresolved in the crystal structure.

Figure 2-4 From Ref. [1] with permission. (E, F) Comparison of the current comparative structure models for the 16S and 23 S rRNAs with the corresponding ribosomal subunit crystal structures. (E) 16S rRNA versus the T. thermophilus structure (GenBank accession no. M26923; PDB code 1 FJF). (F) 23S rRNA, 5'-half and 3'-half versus the H. marismortui structure (GenBank accession no. AF034620; PDB code 1JJ2). Nucleotides are replaced with colored dots that show the sources of the interactions: red, present in both the covariation-based structure model and the crystal structure; green, present in the comparative structure and not present in the crystal structure; blue, not present in the comparative structure and present in the1 crystal structure; purple, positions that are unresolved in the crystal structure.

interactions involving two or more hydrogen bonds occur within domains, rather than between them. This is also the reason for that the interchange of the domain V from E. coli against Staphylococcus aureus in the 23S rRNA of E. coli was possible, even though it introduces 132 changes into the E. coli rRNA sequence, only one additional mutation of U1782C for was necessary for viability [57].

Nearly all of the secondary-structure base pairings and a few of the tertiary base pairs observed in the crystal structure had already been predicted by comparative structure models. Specifically, more than 1250 base pairs predicted were indeed present in the 16S and 23S rRNA crystal structures. The ~35 predicted base pairs, which were not found in the crystal structures, could simply not occur at all or possibly only at certain stages of protein synthesis, for example, in the 30S subunit, the A-A base pair between positions 1408 and 1493 is broken upon binding of tRNA and mRNA (cf. 1FJF and 1IBM) [58, 59]. The crystal structures of small and large sub-units enriched the secondary-structure diagrams by ~170 base pairs in 16S and ~415 in 23S rRNA, i.e., these were not predicted by comparative methods. Essentially, all the "mis-assigned" base pairs have no significant amount of variation (Ref. [55], see Figs. 2-4E and F).

The fact that the domain secondary structures form well-defined structural domains of quaternary structure in the small subunit, but not in the large subunit, may result from two reasons that need not be mutually exclusive: (i) the small subunit might require larger flexibility for ribosomal functions [2], and the difference in the organization of the secondary structures might indicate that (ii) the 50S subunit is older in evolutionary terms, because more time would be required to evolve such a complicated interwoven structure similar to that of the 50S subunit. As Schimmel and Henderson [60] noted, three elements of protein synthesis, viz. tRNA, synthetases and the ribosome, separate into two domains of different evolutionary ages that have probably co-evolved. The "old" domain of the tRNA is the aminoacyl stem (the short arm of the L-shaped tRNAs or mini-helix), which corresponds to the catalytic domain of synthetases in charging the tRNAs and to the large ribosomal subunit involved in peptide-bond formation. The "young" domains are represented by the long arm of tRNAs bearing the anticodon loop, the recognition domain of the synthetases, and the small ribosomal subunit that interacts with the anticodon loop and some of the stem base-pairs [61, 1].

0 0

Post a comment