Tải bản đầy đủ - 0 (trang)
6 The Ribosome, a Ribozyme Machine

6 The Ribosome, a Ribozyme Machine

Tải bản đầy đủ - 0trang


6. RNA Structures and Their Diversity

Figure 6.31 Structure of the thiamine pyrophosphate riboswitch, with the thiamine pyrophosphate

shown in ball and stick form (Edwards and Ferré-D’Amaré, 2006).

Figure 6.32 An enlarged view of the bound thiamine pyrophosphate in the riboswitch complex, with

hydrogen bonds shown as dashed lines.


6.6 The Ribosome, a Ribozyme Machine

Electron microscopy has enabled the outlines of ribosome structure to be defined, which

has been of considerable help to the more recent single-crystal analyses. Crystals of diffraction quality were first obtained well over 20 years ago, but only relatively recently

have the various necessary technologies become available for tackling such a massive and

complex crystallographic task. It was necessary to collect huge data sets from the very

large unit cells involved, with crystals that diffracted only weakly and decayed rapidly in

the X-ray beam. This has been achieved using synchrotron radiation together with lowtemperature methods to stabilize the crystals from excessive decay.

Determining the phases of these large structures has also been a formidable challenge. Rather than using single heavy atoms, the use of clusters of heavy metals has

been successful, especially for the initial structures, at less than atomic resolution.

The progress of ribosome structure determination has been rapid (Table 6.7). The

first structure, at 9 A˚ resolution, was reported in 1998 (Ban et al., 1998), culminating

in electron-density maps of the small (Tocija et al., 1999) and large (Yusupov et al.,

2001) subunits at 3.0 A˚ and 2.4 A˚ resolution respectively, at which point individual

side-chains and bases become apparent and can be refined. Recently, several complete

ribosomal structures have been determined, of the E. Coli ribosome (Schuwirth et al.,

2005), and from T. thermophilus with tRNA (Korostelev et al., 2006) and with tRNA

and mRNA (Selmer et al., 2006). These structures are some of the largest determined

by X-ray crystallography, and dissection of the electron density maps has only been

possible with the aid of the large body of biochemical and other data accumulated on

the ribosome and its subunits. It is remarkable that the secondary structures for the

RNAs in these complexes are close to those predicted by analyses of RNA across species, using the methods of phylogenetic analysis, which are based on the reasonable

assumption that RNA structure is largely conserved throughout evolution.

A number of ribosome coordinate data sets are now available from the Data Banks.

Although only backbone Cα carbon atoms were deposited for the protein components

in the earlier lower-resolution structures, protein side-chain atoms have been identified

in the more recent analyses, as well as complete coordinate sets for the RNA components. The sheer size of the PDB files involved has resulted in a number being split

over more than one entry – this is reflected in Table 6.8. The size and complexity of

ribosome structures is such that there is some advantage when viewing these very

large assemblies, in restricting initial viewing to backbone atoms. The necessarily brief

survey below of the structures cannot do justice to their complexity, and the reader is

strongly encouraged to read the primary literature on them.

Table 6.7

Progress in Ribosome Crystallographic Structure Determination















H. marismortui

H. marismortui

T. thermophilus

T. thermophilus

T. thermophilus

T. thermophilus

T. thermophilus

H. marismortui

T. thermophilus

T. thermophilus

T. thermophilus

E. Coli

Resolution (A˚)














Ban et al., 1998

Ban et al., 1999

Clemons et al., 1999

Cate et al., 1999

Tocija et al., 1999

Wimberley et al., 2000

Ban et al., 2000

Yusupov et al., 2001

Carter et al., 2000

Korostelev et al., 2006

Selmer et al., 2006

Schuwirth et al., 2005


6. RNA Structures and Their Diversity

Table 6.8 Selected Ribosome Subunit Crystal Structures

Subunit and resolution

30S, 3.0 A˚

50S, 2.4 A˚

30S with mRNA + tRNA, 3.3 A˚

70S+ tRNA + mRNA, 5.5 A˚

70S + tRNA, 3.71 A˚

70S + tRNA + mRNA, 2.8 A˚

70S ribosome from E. Coli

PDB code no





2OW8, 1VSA

2J01, 2J02, 2J03, 2J04

2AVY, 2AW4, 2AW7, 2AW8

NDB code no



RR0029, RR0030

RR0031, RR0032

RR0163, RR0164



The initial crystal structure of the complete 70S ribosome (Yusupov et al., 2001),

although at relatively low resolution, revealed much about the interactions between the

subunits that are an essential element of the protein-synthesis cycle. It is notable that even

though the resolution precluded detailed study of the interactions involved, the bound

tRNA molecules were all seen to be extensively contacted by ribosomal RNA in addition

to the necessary established functional interactions such as codon-anticodon recognition.

More recent analyses, at 2.8 A˚ (Selmer et al., 2006) and at 3.7 A˚ (Korostelev et al., 2006),

together with that of the complete E. Coli ribosome at 3.5 A˚ (Schuwirth et al., 2005), have

enabled atomic-level details to be unambiguously defined.

6.6.1 The Structure of the 30S Subunit

The 3.0 A˚ crystal structure (Wimberley et al., 2000), together with the more recent

higher-resolution analyses, have located the complete 16S ribosomal RNA of 1511

nucleotides together with the ordered regions of 20 ribosomal proteins, altogether organized into four well-defined domains. The implication is that there is considerable flexibility between them, which is needed in order to ensure the movement of messenger

and transfer RNAs. The overall shape of the 30S particle is dominated by the structure

of the folded RNA (Fig. 6.33), and the proteins serve merely to fill up the gaps and

hold the whole assembly together. The secondary structure of the RNA shows a very

large number (>50) helical regions. The numerous loops are mostly small and do not

disrupt the runs of helix in which they are embedded. There are extensive interactions

between helices, mostly involving coaxial-stacking via the minor grooves. In one type

of helix–helix interaction two minor grooves abut each other, with consequent distortions from A-type geometry. These distortions, which tend to involve runs of adenines,

are facilitated by both extrahelical bulges and noncanonical base pairs, as have been

observed in simple RNA structures. Less commonly, perpendicular packing of one helix

against another (also via the minor groove) is mediated by an unpaired purine base. This

mode is of special importance since it involves the functionally significant helices in

the 30S subunit. The motifs of RNA tertiary structure such as non-Watson–Crick base

pairs, base triplets and tetraloops, all contribute to the overall structure.

The majority of the 20 ribosomal proteins in the structure each consist of a globular region and a long flexible arm. The latter have been too flexible to be observed in

structural studies on the individual proteins, but have been located in the 30S subunit,

where they play important roles in helping to stabilize the RNA folding, by essentially

filling in the numerous spaces in the RNA folds.

The structure identifies the three sites where tRNA molecules bind and function,

and where the essential proofreading checks for fidelity of code-reading and translation occur. These sites are:

6.6 The Ribosome, a Ribozyme Machine


Figure 6.33 Crystal structure of the RNA component in the 30S ribosome subunit (Wimberley

et al., 2000).

The P (peptidylation) site, when a tRNA anticodon base pairs with the appropriate

codon in mRNA, and where the peptide chain is covalently linked to a tRNA

The A (acceptor) site, when peptide bonds are eventually formed (the actual

peptidyl transferase steps occur in the 50S subunit)

The E (exit) site, for tRNAs to be released from the subunit as part of the proteinsynthesis cycle.

tRNA itself was not present in the first 30S crystal structure, but the RNA from a symmetry-related 30S subunit effectively served to mimic it as the anticodon stem-loop. Its

interactions with the 30S RNA are also mediated via (i) minor-groove surfaces, helped

by some contacts with ribosomal proteins, and (ii) via backbone contacts. Interestingly,

the exit site of the tRNA is almost exclusively protein-associated, whereas the other

functional sites are composed of RNA and not protein. It is thus the ribosomal RNA

that mediates the functions of the 30S subunit, and not the ribosomal proteins.

The determination (Ogle et al., 2001) of the crystal structure at 3.3 A˚ of the 30S

subunit complete with the anticodon stem-loop of a tRNA and a short RNA acting as

a mRNA, has enabled the details of codon-anticodon recognition and checking to be

visualized. A key role is played by two adjacent adenines from the 16S RNA, which

are in contact with the minor groove of the codon-anticodon helical stem, so as to

verify the correctness of the first two Watson–Crick base pairs of the triplet, and selection of the correct codon. The third codon position (the “wobble” position), is not as

involved in interactions with the 16S RNA and a tRNA anticodon is thus able to cope

with hydrogen-bonding to a range of 3rd-position codon bases. The codon-anticodon


6. RNA Structures and Their Diversity

stringency is disrupted by the antibiotic paromomycin, which has also been cocrystallized with the 30S subunit.

Several other antibiotics that are known to inhibit protein synthesis have been

examined bound to the 30S subunit, following cocrystallization experiments (Carter

et al., 2000; Brodersen et al., 2000). For example, streptomycin has been found to

bind tightly to four distinct sites on the 16S RNA, mostly via phosphate groups. This

stabilizes the A site in an open conformation, thereby enabling noncognate tRNAs to

bind and reducing proofreading capability. These and other studies are relevant to the

structure-based design of new antibiotics to be rationally designed, and are further discussed in Sect. 6.5.2. A second crystal structure (Schlunzen et al., 2000), also at 3.3 A˚

resolution, has shown that the decoding region of the structure, is highly conserved.

Furthermore the A and P site tRNAs and the codon part of the bound mRNA do not

interact with any of the proteins, again emphasizing the dominant role of RNA in

ribosome structure and function.

6.6.2 The Structure of the 50S Subunit

In the 2.4 A˚ crystal structure (Ban et al., 2000) 2711 out of the total of 2923 ribosomal

RNA nucleotides have been observed, together with 27 ribosomal proteins, and 122

nucleotides of 5S RNA, with the subunit being about 250 A˚ in each dimension. Again,

the function of the proteins appears to be mainly to act as a cement, helping to maintain the integrity of the folded RNA. Although the RNA itself can be arranged into six

domains on the basis of its secondary structure, overall the 50S subunit is remarkably

globular, reflecting its greater conformational rigidity compared to the 30S subunit,

consist with its functional need to be more flexible.

The principal function of the 50S subunit is peptide bond formation. The active

site where this occurs is within domain 5. Its precise location was identified by the

analysis of crystals containing the antibiotic puromycin, an inhibitor of peptide synthesis. Remarkably, there is no protein structure within 18 A˚ of this molecule, bound

in its active site, and accordingly it has been proposed that peptidyl transferase catalytic activity is entirely carried out by RNA, analogous to the function of RNAs as

ribozymes. The reaction is an acid-base hydrogen transfer, with an adenine being suggested as the key residue, accepting a hydrogen atom from an aminoacylated tRNA

so that it can bond to the carbonyl group of an esterified peptidyl-tRNA carrying the

peptide chain to which the new amino acid is to be attached. All nucleotides involved

in the active site are highly conserved in nature, strongly suggesting that RNA catalysis

is a fundamental mechanism acting throughout evolution.

6.6.3 Complete Ribosome Structures

Structural studies on complete ribosomes are important for full understanding of

mechanism since they do not function as individual subunits, but as complete 70S

particles. The crystals of the E. Coli ribosome (Schuwirth et al., 2005) have provided

two structures for this intact ribosome, since there are two independent ribosome

assemblies per crystallographic asymmetric unit. There are some significant differences

between the two structures, notably that the 30S subunit is rotated about 6° around

the so-called neck region (which connects the 30S and 50S subunits), in one structure

6.7 RNA-Drug Complexes


relative to the other, and is also different from the conformation seen in the 2.8 A˚ crystal structure of the intact 70S T. thermophilus ribosome (Selmer et al., 2006), in which

tRNA molecules are clearly seen (Fig. 6.34) together with part of the mRNA (some of

it is disordered and is not visible in the electron density). Figures 6.35 and 6.36 shows

the high level of detail observed in this structure.

6.7 RNA-Drug Complexes

A variety of RNA molecules and assemblies are targets for therapeutic intervention by

small molecules. An important target that has received much attention, is the HIV

Trans-activator of transcription (Tat) protein and Trans Activation Responsive region

(TAR) RNA. This has an arginine-rich group Tat-recognizing the base sequence and

the conformation of TAR RNA. NMR studies have been extensively used for this

system, and has, for example, identified the binding site of the drug acetylpromazine

on TAR RNA (Du, Lind, and James, 2002).

Figure 6.34 (a–c) Three views of the 30S ribosomal subunit, as found in the crystal structure of the

complete 70S particle (Selmer et al., 2006). The protein secondary structure elements are shown colored

dark grey. The RNA backbones are shown in tube form and bases are represented as sticks.


Figure 6.34 Cont’d.

6.7 RNA-Drug Complexes


Figure 6.35 The RNA components of the 30S ribosomal subunit, as found in the crystal structure

of the complete 70S particle (Selmer et al., 2006). Only the backbone is shown for the ribosomal and

mRNAs; the incoming and outgoing tRNA molecules are see at bottom left.

Many antibacterial antibiotics act directly at the ribosomal level, binding to RNA

structural features and thereby blocking correct translation (Böttger, 2006; Poehlsgaard and Douthwaite, 2005). As outlined in the previous section, there have been

a number of crystal structure analyses of ribosome structures with bound drugs. For

example, the crystal structures of complexes of the large ribosomal subunits from the

organism H. marismortui (Tu et al., 2005) with bound erythromycin, azithromycin,

clindamycin, virginiamycin S, and telithromycin, have provided some insights into

the resistance mechanisms involving these antibiotics. These all contain the mutation

G2099A, which confers increased affinity on erythromycin, possibly as a result of the

loss of solvent at the N2 position on going from G to A. Crystallographic analyses

(Hansen et al., 2002) of the four macrolide antibiotics (carbomycin A, spiramycin,

tylosin, and azithromycin) with the same subunit have shown that these drugs bind in

the polypeptide tunnel adjacent to the peptidyl transferase center, and therefore that

their action involves blocking a growing peptide chain from exiting.

Since the crystal structure of the ribosomal subunit for a particular microorganism is

not always available, or is often at only moderate resolution (ca 3 A˚), a more rapid and


6. RNA Structures and Their Diversity

Figure 6.36 Detail of the interactions (Selmer et al., 2006) between the anticodon stems of the two

tRNA molecules and codons in a fragment of mRNA (on the right-hand side, adopting an extended


precise approach to antibiotic drug design has been to analyze cocrystals between the antibiotic and the RNA that constitutes its minimal binding site. The power of this approach

has been shown by the analyses of complexes involving the aminoglycoside antibiotics

gentamicin C1A, kanamycin A, ribostamycin, lividomycin A, neomycin B (Franỗois

et al., 2004), and several paromomycin derivatives (Hanessian et al., 2007). This latter

study itself follows on from earlier analyses (Vicens and Westhof, 2001) of a complex of

the native paromomycin antibiotic and two synthetic derivatives (Franỗois et al., 2004)

with the decoding A-site RNA of E. Coli. Paromomycin itself is in an L-shape when bound

in this site, with a large number of hydrogen-bond and electrostatic contacts (Fig. 6.37),

that are shared with most of the A-site bound aminoglycoside antibiotics. It was reasoned

(Franỗois et al., 2004) that appropriate chemical modification of paromomycin would lead

to a distinct bound conformation and set of contacts, with enhanced potency and activity

against resistant strains, as was subsequently found experimentally and structurally. Structure-based methods have also been successfully employed (Fig. 6.38) in the design of three

synthetic derivatives of the antibiotic enamine, which similarly binds to the A-site (Murray

et al., 2006). Confirmation of the bound conformation was obtained by a crystal structure

analysis of one derivative bound to a complete 30S ribosome subunit.

Table 6.9 Selected RNA-drug Crystal Structures

RNA type and drug

Neomycin + A-site

Paromomycin + A-site

Streptomycin + 40-mer aptamer

Designer antibiotic1 + A-site

Designer antibiotic2 + A-site

Designer antibiotic3 + A-site

PDB code no

NDB code no













6.7 RNA-Drug Complexes


Figure 6.37 Paromomycin bound in the decoding A-site RNA of E. Coli this site, showing the hydrogen

bond and electrostatic contacts (Franỗois et al., 2004).

Figure 6.38 Crystal structure of a synthetic derivative of the antibiotic enamine, bound to the A-site

(Murray et al., 2006).

Streptomycin (Fig. 6.39) was the first antibiotic with activity against tuberculosis,

although it is no longer the first-line treatment as a consequence of its side effects. It acts

by binding to 16S ribosomal RNA, and a crystal structure of a complex with the 30S

subunit has been determined (Carter et al., 2000), which involves the active recognition of all three rings of the drug. Its cocrystal structure with a 40-mer RNA aptamer

(Tereshko, Skripkin, and Patel, 2003), shows a very different arrangement, with an Lshaped RNA fold (Fig. 6.40). Only one part of the drug molecule (the six-membered

carbohydrate streptose ring) is fully buried in the binding pocket that is formed from the

two asymmetric internal loops of the RNA, and held by several hydrogen bonds.


6. RNA Structures and Their Diversity

Figure 6.39

The structure of streptomycin.

Figure 6.40 (a,b) Two views of the streptomycin complex with a 40-mer RNA aptamer (Tereshko,

Skripkin, and Patel, 2003). The enlarged view shows the hydrogen-bonding between the partially bound

streptomycin molecule and RNA sites.

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

6 The Ribosome, a Ribozyme Machine

Tải bản đầy đủ ngay(0 tr)