RNA-Protein Machines
Nucleic Acids and Molecular Biology Group Colloquium Organized by M. Hartley (Department of Biological
Sciences, University of Warwick), J. McCarthy (Department of Biomolecular Sciences, UMIST) and M. Gait
(MRC Laboratory of Molecular Biology, Cambridge), and Edited by M. Gait. 675th Meeting held at the
University of York, 17-1 9 December 200 I, Joint with The Physiological Society.
Functions and interplay of the tRNA-binding sites of the ribosome
V. Marquez, D. N . Wilson and K. H. Nierhaus'
Max-Planck-lnstitut fur Molekulare Genetik AG Ribosomen, lhnestrasse 73, D- I 4 I95 Berlin, Germany
Abstract
signal of the mRNA with the help of specific
protein-initiation factors and an initiator tRNA,
thus forming an initiation complex. Following
the recruitment of the large ribosomal subunit the
ribosome is now prepared for the second phase,
the elongation cycle. T h e ribosome decodes genetic information as it 'travels' along the mRNA,
simultaneously elongating the nascent peptide
chain by addition of amino acids. T h e appearance
of a stop codon in the A site signals the third
phase of protein synthesis, the termination phase,
where the nascent peptide chain is cleaved and
released from the ribosome.
tRNAs play a crucial role during the initiation
and elongation phases. These adapter molecules
form the connection between the RNA and protein
worlds by linking the codon of the mRNA with
the corresponding amino acid. There are three
tRNA-binding sites on the ribosome. T h e A
site, containing the decoding centre, binds an
aminoacyl-tRNA corresponding to the codon
displayed at this site. T h e P site binds the peptidyltRNA, the tRNA carrying the nascent chain before
peptide-bond formation, and the E site binds
specifically deacylated tRNAs, tRNAs that have
already participated in chain elongation. In the
course of two elongation cycles a tRNA moves in
succession through the ribosomal A, P and E sites.
Figure 1 depicts the elongation cycle in the frame
of the CI-E model, a current model that best
satisfies the existing structural and biochemical
data (see Figure 1 legend and [1,2] for reviews).
There are two elongation factors that aid tRNA
movement during an elongation cycle. Elongation
factor (EF)-Tu delivers the aminoacyl-tRNA to
the A site in the form of a ternary complex
aminoacyl-tRNA. E F - T u . GTP. EF-G is involved in the translocation reaction, which basically moves the tRNAs into the adjacent binding
site on the ribosome, thereby moving the ribo-
T h e ribosome translates the genetic information
of an mRNA molecule into a sequence of amino
acids. T h e ribosome utilizes tRNAs to connect
elements of the RNA and protein worlds during
protein synthesis, i.e. an anticodon as a unit of
genetic information with the corresponding amino
acid as a building unit of proteins. Three tRNAbinding sites are located on the ribosome, termed
the A, P and E sites. In recent years the tRNAbinding sites have been localized on the ribosome
by three different techniques, small-angle neutron
scattering, cryo-electron microscopy and X-ray
analyses of 70 S crystals. These high-resolution
glimpses into various ribosomal states together
with a large body of biochemical data reveal an
intricate interplay between the tRNAs and the
three ribosomal binding sites, providing an explanation for the remarkable features of the ribosome, such as the ability to select the correct
ternary complex aminoacyl-tRNA. E F - T u . G T P
out of more than 40 extremely similar tRNA
complexes, the precise movement of the
tRNA, . mRNA complex during translocation
and the maintenance of the reading frame.
Introduction
T h e ribosome translates the genetic information
contained within an mRNA in three phases. First,
the small ribosomal subunit recognizes the start
Key words: A site, E site, P site, protein synthesis.
Abbreviations used: EF, elongation factor: PTF, peptidyltransferase; SANS, small-angle neutron scattering; PRE, pretranslational: POST, post translational: cryo-EM, cryo-electron
microscopy: SD, Shine-Dalgamo: RF2, release factor 2.
'To whom correspondence should be addressed (e-mail
[email protected]).
I33
0 2002 Biochemical Society
Biochemical Society Transactions (2002) Volume 30, part 2
codons with a length of l O A each, although a
tRNA as an RNA helix measures 20 A in diameter,
and at the other end both tRNAs must contact the
peptidyl-transferase (PTF) centre. We will see
that the simultaneous interaction of both tRNAs
with the mRNA is of utmost importance for the
accuracy of aminoacyl-tRNA selection, the translocation reaction and the maintenance of the
reading frame, i.e. for all the essential features of
protein synthesis.
some by one codon length towards the 3’-end of
the mRNA.
Although the ribosome ensures correct
codon-anticodon (mRNA-tRNA) interactions, it
is actually the synthetases that ensure that each
codon in the mRNA is correlated with the correct
amino acid. U p to 20 different synthetases exist in
a cell, one for each amino acid. T h e synthetases
form this link by ensuring that each tRNA species,
a tRNA with a unique anticodon, is charged with
an amino acid that correlates with the codon that
the anticodon recognizes.
T h e universal L-shape of the tRNAs is a
consequence of the absolute requirement of
codon-anticodon interactions of both tRNAs
present on the ribosome, which are found in the A
and P sites before translocation and in the P and E
sites afterwards. T h e reason is that the tRNAs
have to make contact, with one end, two adjacent
Localization of the tRNA-binding sites
on the ribosome
T h e first reliable localization of tRNA-binding
sites on the ribosome was achieved by means of
small-angle neutron scattering (SANS) [3]. T h e
two main functional states of the ribosome during
elongation were constructed under near-in vivo
Figure I
The
Q-E
model of the elongation cycle, according to [ I71
The essential feature is a movable ribosomal a-E domain that connects both subunits through the intenubunit space, binds both tRNAs of
an elongating ribosome and carries them from the A and P sites to the P and E sites, respectively, during translocation. The model explains
the reciprocal linkage between the A and E sites by the fact that the a-E domain moves out of the A site during translocation, leaving the
decoding centre alone at the A site. Yellow and green, the two binding regions of the a-Edomain: blue, the decoding centre at the A site.
y3
hk=
T
EF-TU GTP
EF-G GDP + Pi
EF-G GTP
ar
POST
l a
I
E PE Aa
u
PRE
POST
PRE
30s
50s
POST
0 2002 Biochemical Society
I34
RNA-Protein Machines
tonated (‘stained’) ligands, in this case tRNAs,
within a large deuterated complex like the ribosome. T h e two tRNAs at A and P sites of the
PRE state and at P and E sites of the P O S T
state could be localized unequivocally between
the subunits. An important outcome was that the
anticodons of both tRNAs neighbour each other,
thus allowing adjacent codon-anticodon interaction in the PRE and P O S T states, and that the
ionic conditions, that is the state prior to translocation (pre-translocational or PRE state) and the
state after (post-translocational or P O S T state).
Ribosomeswerefullydeuteratedand bothmRNA
and tRNAs were protonated (their normal state in
Nature). Since neutrons are scattered differently
by protons and deuterons (protons and deuterons
are the atom nuclei of normal and heavy hydrogen,
respectively), this allows localization of the pro-
Figure 2
tRNA locations on the ribosome
(A) tRNAs in the PRE and POST states ofthe E. coli ribosome. Shown are cryo-EM reconstructions oftRNAs bound to the 70 S ribosome
[6]. PRE state, tRNAs bound to the A site (pink) and P site (green). POST state, tRNAs bound to the P site (green) and E site (yellow). The
small 30 S (yellow) and large (blue) subunit are shown. The 70 S ribosome is presented as a semi-transparent surface to better illustrate
the tRNA positions. Abbreviations: CP, central protuberance; ST, L7/LI 2 stalk base; Ib, long bridge (helix 38 of 23 S rRNA); h, head; ch,
mRNA channel: sh. shoulder region: sp, spur. (6)Relative positions of the three tRNAs bound at the A, P and E sites. The P and C I ’ atoms
The two
of the P-site tRNAs are used t o align the molecules of the cryo-EM (brown) [6] with that from the X-ray work (dark green)
studies agree on the position of the tRNAs on the ribosome. From this comparison it can be seen that the A-site tRNAs (cryo-EM in olive
green, X-ray in cyan) are shifted relative to each other along the anticodon stem axis (vertical arrow), and the E-site-bound tRNAs along
the acceptor stem axis (arrow; cryo-EM in red and X-ray in blue). Note that the anticodon regions of all three tRNAs are in close proximity
t o each other. Panel (B)is reproduced from [23] with permission. 0 (2002) Bentham Science Publishers.
m.
B
I
Ch
.
P\sp
PRE
I35
E
P
A
E
P
A
0 2002 Biochemical Society
Biochemical Society Transactions (2002) Volume
30,part 2
sites at a resolution of 5.5-7.5 A was the next
important achievement [7]. T h e crystal structures
had a very similar arrangement of the tRNAs
when compared with the cryo-EM reconstructions
(Figure 2B), even though the complex (i) was
obtained from the thermophilic bacterium Thermus thermophilus instead of Escherichia coli, and
(ii) does not represent a functional state of the
ribosome, since three deacylated tRNAs are never
found on the ribosome during protein synthesis.
In addition, the tRNA-ribosome contacts could
be determined in molecular detail by utilizing the
higher-resolution subunit structures (30 S structure [8,9] and 50 S structure [lo]). In particular,
the fixation of the anticodon region of the P-site
tRNA by a number of ribosomal fingers was
revealed. Another interesting observation was that
the deacylated tRNA was buried deeply in the
ribosomal matrix. This suggests that an active
mechanism is necessary to release the E sitebound tRNA. Precisely this point was a topic of
intense discussion during the past 15 years.
Results obtained with near-in vivo conditions
clearly indicated binding of a tRNA at the E-site,
invoking an active release mechanism, whereas
results gathered under conventional conditions
suggested a weak binding and an easy release from
the E site (for discussion, see [l]).
T h e three tRNA-binding sites differ in their
capability to bind tRNAs with different charging
states, namely deacylated tRNA, aminoacyltRNA and peptidyl-tRNA. These features are
summarized in Table 1 (according to [ l 13).
mutual arrangement of both tRNAs was similar in
both states. We stress the usage of near-in vivo
ionic conditions (Mg’+ at 3-6 mM, 150 m M
NH:/Kt, 2 m M spermidine and 0.05 m M spermine), since it has been demonstrated that the usage
of conventional buffer systems (10-15 m M Mg2+,
z 150 m M N H I / K + and no polyamines) can
induce artifacts. During the last two decades these
artifacts have provided controversy with regard to
the importance of the E site and have given rise
to an alternative model for the elongation cycle
([4]; for review, see [l]).
However, the resolution power of the SANS
method is limited. An estimate of the resolution
obtainable with the SANS method would fall
within a range of 2 5 4 0 A. In comparison, cryoelectron microscopy (cryo-EM) and X-ray crystallographic analyses of 70 S ribosomes and
subunits have produced resolutions of 12-1 8 A
and 2.3-5.5 A, respectively. T h e emergence of
cryo-EM within the field of ribosome research [5]
provided an exciting opportunity to re-analyse
functional complexes at near-in vivo conditions in
collaboration with the Frank group (Wadsworth
Center, Albany, NY, U.S.A.). T h e result was
the first clear description of both the PRE and
P O S T elongation states in molecular terms [6]
(Figure 2A) and the mutual arrangement of the
three tRNAs on the ribosome (Figure 2B). T h e
tRNA-binding sites can now be seen at higher
resolution ( z15 A), generally confirming the
positions and the angle between the tRNAs observed in the aforementioned SANS study, but
with important improvements. (i) T h e exact
tRNA positions could be assessed. A SANS
study principally cannot distinguish between a
distinct mutual arrangement of the two tRNAs
and its mirror arrangement. (ii) T h e anticodon
distances of the two adjacent tRNAs could be
determined precisely, clearly indicating that a
simultaneous codon-anticodon interaction of both
tRNAs is possible. (iii) T h e CCA 3’-ends of the
two tRNAs in the PRE state are in close proximity
to each other, as was expected since both ends
must contact the P T F centre on the large subunit.
T h e PTF centre is responsible for peptide-bond
formation between the peptidyl residue of the
peptidyl-tRNA at the P site and the aminoacyl
residue of the aminoacyl-tRNA at the A site. In
contrast, the CCA ends of the tRNAs in the P O S T
in the
state are far from each other ( x 17
PRE state and 60 A in the P O S T state).
T h e crystal structure of 70 S ribosomes
carrying three deacylated tRNAs at the A, P and E
Functional features of the ribosome:
selection of the correct aminoacyltRNA at the decoding centre of the
A site
Two features are of vital importance for the
intricate functional spectrum of the ribosome. (i)
T h e first and the third tRNA-binding sites, the A
and the E sites, are coupled in a reciprocal fashion.
This means that a tRNA bound to the E site
induces a low-affinity state for an aminoacyltRNA or a ternary complex at the A site and vice
versa; occupying the A site promotes a low-affinity
state of the E site thus triggering the release of the
deacylated tRNA from this site ([12,13]; see also
Figure 1). A consequence of this feature is that
statistically only two tRNAs are found on the
ribosome during protein synthesis, either at the A
and P sites (PRE state) or at the P and E sites
A
0 2002 Biochemical Society
I36
E
n
3.
n,
85
m
h,
0
0
h,
0
+Deacylated tRNA
Peptidyl-tRNA at the P site
and a deacylated tRNA at
the E site
+Aminoacyl-tRNA .EF-Tu .GTP
+Aminoacyl- or peptidyl-tRNA
+Deacylated tRNA
+Peptidyl-tRNA
+Aminoacyl-tRNA. EF-TU* GTP
A peptidyl-tRNA at the P site
-6
37 "C: A site induces complete release of E-site-bound tRNA.
bound tRNA.
0 "C: binds t o the A site, incomplete release of the E-site-
E-site-bound tRNA.
0 "C: no binding t o the A site.
37 " C : binding t o the A site induces quantitative release of the
release of the E-site-bound tRNA.
0 "C: no binding.
37 "C: binding t o the A site does not trigger quantitative
site without EF-G.
0/37 " C : binds directly t o the E site.
0/37 "C: does not bind t o the A site (exclusion principle [24]).
0 "C: binds directly t o the A site, dipeptide formation.
37 "C: dipeptidyl-tRNA slides within several minutes t o the P
site; antibiotic viomycin prevents sliding.
0137 "C: binds directly t o the P site.
0137 "C: binds directly t o the P site.
0 "C: binds directly t o the A site; EF-G promotes translocation
t o the P site at 0 "C.
37 " C : binds t o the A site and slides (without EF-G) t o the P
+Deacylated tRNA
Empty ribosome with mRNA
+Peptidyl-tRNA
+Aminoacyl-tRNA
fEF-TU.GTP
Binding event
Added tRNA
State of the ribosome
Binding features of the tRNA-binding sites
Table I
Biochemical Society Transactions (2002) Volume 30, part 2
more, every one of the 41 ternary complexes
would interfere with the selection process and thus
could by chance be incorporated erroneously. This
is never seen; instead, only three or four ternary
complexes can be accepted by the A site, those
tRNAs with either the correct (cognate) or similar
(near-cognate) anticodon. Thus the majority
( z 90 yo) of the non-cognate ternary complexes
that carry an aminoacyl-tRNA with a dissimilar
anticodon are never mis-incorporated during protein synthesis.
The solution to this selection riddle is the
reciprocal coupling between A and E sites. An
occupied E site (POST state, see Figure 1) induces
a low-affinity state in the A site, and this state
mainly allows codon-anticodon interaction but
abolishes all interaction outside the anticodon.
This ribosomal trick has two consequences. First,
equilibrium can be reached in a short time in
agreement with that observed in protein synthesis
in the cell. Second, after the decoding has been
successfully completed, the high-affinity state of
the A site has to be adjusted and the aminoacyltRNA adapted at the A site, probably a gross
conformational change that lasts significantly
longer than the preceding decoding step. A corollary of such a reaction, where the first partial
reaction is fast and the second slow, implies that
the first reaction reaches equilibrium even under
steady-state conditions, thus allowing full exploitation of the discrimination potential of codonanticodon interaction at the A site. This mechanism also explains why non-cognate ternary
complexes are never mis-incorporated : at the first
step, non-cognate anticodons dissimilar to the
cognate one that cannot interact with the A site
codon are thus not selected, and the potential
contact regions outside of the anticodon are not yet
available.
(POST state). (ii) As mentioned already, both
tRNAs simultaneously undergo codon-anticodon
interaction in the PRE and POST states. Interaction in the POST state was a surprise, since
decoding does not take place here. Instead,
codon-anticodon interaction at the E site is the
signal for the ribosome to switch into the POST
mode with a low A-site affinity [12]. As will
become evident, codon-anticodon interaction of
the adjacent tRNAs on the ribosome is a cornerstone for the translocation mechanism as well
as for the maintenance of the reading frame (see
the next two sections).
Here we will discuss the importance of
the reciprocal coupling of A and E sites for the
accuracy of tRNA selection at the A site. This
point has been reviewed extensively ([14,15]), and
therefore we only present a brief summary.
Enzymes achieve an astonishing precision of
substrate recognition and binding (a precision
of one error in 103-106 binding events), a precision that is necessary in order to avoid the degeneration of metabolism into complete chaos.
Enzymes attain this accuracy mostly or exclusively by binding their substrate using the ‘molecular corner’ that is unique amongst its structural
relatives. Therefore, the binding energy is identical to the discrimination energy in the best cases.
Since the binding constant describes an equilibrium state, it is clear that the discrimination potential can only be exploited under equilibrium
conditions.
With this in mind it is evident that the
ribosome faces an almost unsolvable selection
problem. In E. coli, for example, 41 ternary
complexes containing tRNA with different anticodons all compete for the codon displayed at the
A site. The discrimination is achieved solely via
codon-anticodon interaction, although a tRNA
makes many other contacts with the ribosome
[7,16,17]. In addition, it is not the aminoacyltRNA but rather the ternary complex of
aminoacyl-tRNA . EF-Tu . G T P that is a substrate for the A site. The elongation factor EF-Tu
is twice as large as a tRNA molecule and increases
the affinity to the A site by at least two orders of
magnitude [1 13, indicating additional contacts
with the A site. Taken together, this means that
the discrimination (or binding) energy due to
codon-anticodon interaction is only a tiny fraction
of the total binding energy of a ternary complex,
and therefore the time to reach equilibrium would
take hours rather than milliseconds (in Nature,
one elongation cycle lasts % 50-100 ms). Further-
0 2002 Biochemical Society
Functional features of the ribosome:
the translocation reaction
As mentioned already, tRNAs are L-shaped molecules the functional groups of which are located
at the tips of the arms and separated by about
75A; the anticodon is at one end and the acyl
residue (aminoacyl or peptidyl group) is at the
other. The anticodon interacts with the codons on
the mRNA at the decoding centre on the small
ribosomal subunit and the acyl residues with the
P T F centre on the large ribosomal subunit. A
tRNA is the size of an average enzyme, and the
I38
RNA-Protein Machines
problem with the translocation of tRNAs is that
both tips of this large molecule have to be moved
with high precision. Imprecise movement at the
anticodon end would lead to a loss of the reading
frame, and at the acyl end would preclude proper
placement at the PTF centre, thus preventing
peptide-bond formation.
T h e analysis of contact patterns of tRNAs in
various ribosomal sites has provided an answer to
this problem. I n brief, it was found that the contact
patterns of both tRNAs at the A and P sites in the
PRE state were remarkably different, but that
both patterns did not change upon translocation to
the P and E sites, respectively, i.e. to the P O S T
state. We interpreted this finding to suggest that
the microtopography of the tRNA-ribosome interactions does not change during translocation.
This implies that a movable ribosomal domain
should exist that tightly binds both tRNAs and
moves or guides them from one tRNA-binding
site to the adjacent one (a-E model for the
elongation cycle; Figure 1). Ribosomal components as candidates for the movable domain
have been identified with both cryo-EM [18]
and X-ray analysis of ribosome crystals [7],
It follows that the tRNAs are the handle that
is moved actively by the ribosome and that the two
associated codon-anticodon interactions are required to concomitantly draw the mRNA through
the ribosome. T h e mRNA follows passively the
actively moved tRNAs during translocation. This
scenario explains why the number of base pairs
during codon-anticodon interactions at the A site
can induce frameshifts : tRNAs with modified
anticodon loops that allow either 2 or 4 bp instead
of the canonical 3 b p provoke a -1 or +1
frameshift during translocation [19,20].
whereas during normal translation the error
frequency for frameshifts is not higher than 1 case
in 30000 amino acid incorporations [22], i.e. the
frameshift on the RF2 mRNA occurs with a
frequency that is more than four orders of magnitude larger than normal. What mechanism
operates at this recoding site to override the strict
maintenance of the reading frame?
T h e recoding site has a number of special
features; in particular, the presence of a ShineDalgarno (SD)-like sequence upstream of the
UGA stop codon. Normally, the SD sequence is
associated with initiation in bacteria and is found
upstream of the initiation A U G codon. T h e S D
sequence is complementary to a sequence of the
3’-end of 16 S rRNA (anti-SD). Duplex formation
is a prerequisite for bacterial initiation. What has
been overlooked in a number of publications is
that here the SD-like sequence would be predicted
to extend its interaction into the E site, i.e. basepairing with the first base of the E-site codon. Our
hypothesis is that this duplex formation between
the anti-SD-like sequence and the E site codon
would prevent codon-anticodon interaction and
thus trigger the release of the deacylated tRNA
from the E site, leaving the ribosome with only a
single peptidyl-tRNA at the P site. This situation
does not occur during elongation and might facilitate a slip of the ribosome on the mRNA, thus inducing, with high probability, a + 1 frameshift.
This hypothesis can be tested and, if true, it not
only provides a satisfying explanation for the extremely high frameshift frequency during the
translation of the RF2 mRNA, but also suggests
that the codon-anticodon interaction at the E site
has the important role of maintaining the reading
frame during translation. We have already shown
in vitro (V. Marquez and K. H. Nierhaus,
unpublished work) that the presence of a S D
sequence positioned as in the RF2 mRNA in
front of an UGA is instrumental for a 1 frameshift with an almost 100% efficiency. Removal
of the S D sequence without changing the amino
acid sequence of the corresponding peptide lowers
the frameshift frequency to background levels. It
remains to be seen whether or not removal of the
deacylated tRNA from the E site is the essential
function of the unusual location of a S D sequence
in front of a stop codon (V. Marquez and K. H.
Nierhaus, unpublished work).
Functional features of the ribosome:
maintenance of the reading frame
+
T h e simple feedback regulation of the synthesis
of the termination factor release factor 2 (RF2)
might shed some light on this aspect. T h e codon
position 26 of the RF2 mRNA is taken by a
U G A stop codon that is specifically recognized
by RF2. If the concentration of RF2 in the cell
is sufficient, the stop codon will be recognized
and translation of the RF2 mRNA will abort
after the synthesis of a 25-mer peptide that is
quickly degraded. However, if there is a shortage
of RF2 in the cell, a + 1 frameshift will occur,
enabling the translation of the complete RF2
protein. This frameshift can occur with an
astonishingly high efficiency of up to 100 yo [21],
We thank the Deutsche Forschungsgemeinschaft for support
(DFG Ni I74/8-2). D.N.W. would like t o recognizethe support of
the Alexander von Humboldt Foundation in awarding a Fellowship
that made this work possible.
I39
0 2002 Biochemical Society
Biochemical Society Transactions (2002) Volume 30, part 2
References
I
2
3
4
5
6
7
8
9
10
II
Nierhaus, K. H., Spahn, C. M. T., Burkhardt, N.. Dabrowski,
M., Diedrich, G., Einfeldt E., Kamp. D., Marquez, V., Patzke,
S., Schafer, M. A. et al. (2000) in The Ribosome. Structure,
Function, Antibiotics, and Cellular Interactions (Garrett,
R A., Douthwaite, S. R., Liljas, A,, Matheson, A. T., Moore,
P. B. and Noller, H. F., eds), pp. 3 19-335, American
Society for Microbiology Press, Washington D C
Blaha. G. and Nierhaus, K. H. (2001) in Ribosome, Cold
Spring Harbor Press, Cold Spring Harbor, NY, in the press
Nierhaus, K. H., Wadzack J., Burkhardt, N.,Junemann,R,
Meerwinck W., Willumek, R and Stuhrmann, H. B. ( I 998)
Proc. Natl. Acad. Sci. U.S.A. 95, 945-950
Moazed, D. and Noller, H. F. ( I 989) Nature (London) 342,
142- I 48
Frank I.,
Verschoor, A.. Wagenknecht, T., Radetmacher, M.
and Carazo, J. M. (I988) Trends Biochem. Sci. 13, 123- I27
Agrawal, R K., Spahn, C. M. T., Penczek P., Grassucci, R A,,
Nierhaus, K. H. and Frank J. (2000) J. Cell Biol. 150,
447-459
Yusupov, M. M., Yusupova, G. Z., Baucom, A,, Lieberman,
K., Earnest, T. N., Cate, J. H. and Noller, H. F. (2001)
Science 292, 883-896
Schluenzen, F., Tocilj, A, Zarivach. R. Harms. 1..
Gluehmann, M., Janell. D., Bashan, A,, Bartels, H., Agmon, I.,
Franceschi, F. and Yonath, A. (2000) Cell 102, 6 15-623
Wimberly, B. T., Brodersen, D. E.,Clemons, W. M.,
Morgan-Warren, R I.,
Carter, A. P., Vonrhein, C., Hartsch,
T. and Ramakrishnan, V. (2000) Nature (London) 407,
327-339
Ban, N., Nissen, P., Hansen, J., Moore, P. B. and Steitz, T. A.
(2000) Science 289,905-920
12
13
14
15
16
17
18
19
20
21
22
23
24
Schilling-Bartetzko, S., Franceschi, F., Sternbach, H. and
Nierhaus, K. H. ( I 992) J. Biol. Chem. 267, 4693-4702
Geigenmuller, U. and Nierhaus, K. H. ( I 990) EMBO J. 9,
452745 33
Triana-Alonso, F. J., Chakraburtty, K. and Nierhaus, K H.
( I 995) J. Biol. Chem. 270, 20473-20478
Nierhaus, K. H. ( I 990) Biochemistry 29,4997-5008
Nierhaus, K. H. ( 1993) Mol. Microbiol. 9, 66 1-669
Dabrowski, M., Spahn, C. M. T. and Nierhaus. K. H. ( I 995)
EMBO J, 14,48724882
Dabrowski, M., Spahn, C. M. T., Schafer, M. A,, Patzke, S.
and Nierhaus, K. H. ( 1998) J. Biol. Chem. 273,
32793-32800
VanLoock M. S., Agrawal, R. K., Gabashvili. I. S., Qi, L.,
Frank J. and Harvey, S. C. (2000) J. Mol. Biol. 304, 507-5 I5
Culbertson, M., Leeds, P., Sandbaken, M. and Wilson, P.
( 1990) in The Ribosome: Structure, Function, & Evolution
(Hill, W. E., Dahlberg, A. E., Garrett, R A,, Moore, P. B.,
Schlessinger, D. and Warner, J. R., eds), pp. 559-570,
American Society for Microbiology Press, Washington D C
Atkins. J. F., Weiss, R B., Thompson, S. and Gesteland, R F.
( I 99 I ) Annu. Rev. Genet. 25,20 1-228
Donly, B. C., Edgar, C. D., Adamski, F. M. and Tate, W. P.
( 1990) Nucleic Acids Res. 18, 65 I 7-6522
Jorgensen,F. and Kurland, C. G. ( 1990) J. Mol. Biol. 2 15,
5 I 1-521
Wilson, D. N.. Blaha, G., Connell, S. R, Ivanov, P. V., Jenke,
H., Stelzl, U.,
Teraoka, Y. and Nierhaus, K. H. (2002) Cur.
Protein Peptide Sci., 3, 1-53
Rheinbetger, H.-J.,Sternbach. H. and Nierhaus, K. H.
( I 98 I) Proc. Natl. Acad. Sci. U.S.A. 78, 53 I C-53 I 4
Received 22 November 200 I
The hepatitis C virus internal ribosome-entry site: a new target for
antiviral research
J. Gallego' and G. Varan?
MRC Laboratoty of Molecular Biology, Hills Road, Cambridge CB2 2QH, U.K.
Abstract
translation-initiationmechanism. The viral RNA
contains an internal ribosome-entry site (IRES)
that is recognized specifically by the small ribosomal subunit and by eukaryotic initiation factor 3,
and these interactions allow cap (7-methylguanine nuc1eotide)-independent initiation of
viral protein synthesis. In this article, we review
the structure and mechanism of translation initiation of the HCV IRES, and its potential as a
target for novel antivirals.
The hepatitis C virus (HCV) is the main causative
agent of non-A, non-B hepatitis in humans and a
major cause of mortality and morbidity in the
world. Currently there is no effective treatment
available for the infection caused by this virus,
whose replication depends on an unusual
Key words: elF3, HCV, IRES, translation.
Abbreviations used: cap, 7-methyl-guanine nucleotide; elF,
eukaryotic initiation factor: EM, electron microscopy: HCV,
hepatitis C virus; IRES, internal ribosome-entry Site; UTR
untranslated region.
'To whom correspondence should be addressed (e-mail
[email protected]).
2Present address: Departments of Biochemistry and Chemistry,
University of Washington, Seattle, W A 98 195- 1700, U S A
0 2002 Biochemical Society
Introduction
The hepatitis C virus (HCV) is the main causative
agent of post-transfusion hepatitis. This human
pathogen is a positive (messenger) sense singlestranded RNA virus (Figure 1) that relies on an
I40