The ribosome, which constitutes one of the most complex and.docx

The ribosome, which constitutes one of the most com-
plex and sophisticated macromolecules in the bacterial
cell, lies at the centre of translation. In bacteria, the small
30S ribosomal subunit associates with the large 50S sub-
unit to form a functional 70S ribosome. The 30S subunit
consists of the 16S ribosomal RNA (rRNA) and 21 pro-
teins (denoted S1–S21; prefix S for ‘small’), whereas the
50S subunit contains two rRNAs (the 23S and 5S rRNAs)
and 33 different proteins (known as L proteins; prefix L
for ‘large’)1. All components are present in one copy, with
the exception of L7/L12, which is present in four or six
copies per ribosome in bacteria2,3 and archaea4,5 (L7 is the
N-acetylated form of L12). These proteins are the only
ribosomal proteins that do not directly interact with
rRNA; their binding is mediated by L10, and together
they form a stable pentameric or heptameric complex 6
known as the L7/L12 stalk (referred to hereafter as the L12
stalk). This stalk is an essential component of the docking
site for the translational guanosine-nucleotide-binding
proteins (G proteins), which assist the ribosome at vari-
ous stages of translation. Despite the large number of
ribosomal proteins, rRNA is the dominant component
in terms of both structure and function (FIG. 1). Decoding
of the mRNA is carried out by elements of the 16S
rRNA7,8, and peptide-bond formation is carried out by
nucleotides of the 23S rRNA9–11 (reviewed in REF. 12).
Ribosomal proteins have important roles in ribosome
biogenesis13,14, in maintaining the overall architecture of
the rRNA, and they have also been implicated in a num-
ber of important functional activities, including mRNA
helicase activity (for S3, S4 and S5)15, decoding (for S12)7

and peptidyltransferase activity (for L27 (REF. 16) and
L2 (REF. 17)).
The ribosome passes through four functional phases
for the synthesis of a single protein: initiation, elonga-
tion, termination and recycling (FIG. 2). All phases are
mediated by specific factors, some of which are bacteria-
specific, whereas others (such as the elongation factors
EF-Tu and EF-G) are universally conserved. The amino
acid substrates that are attached to tRNAs (known as
aminoacyl-tRNAs (aa-tRNAs)) are delivered to the ribo-
some in a ternary complex with EF-Tu and GTP, and
the tRNAs move through three distinct binding sites
(the aminoacyl- (A-), peptidyl- (P-) and exit- (E-) sites)
located at the interface of the 30S and 50S subunits.
After initiation — which involves placement of the
mRNA start codon and the specific initiator tRNA
(formyl methionine tRNA; fMet-tRNA) at the P-site
of the 30S subunit, followed by association of the 50S
sub unit — the elongation cycle ensues. The ribo-
some moves along an mRNA in the 5ʹ to 3ʹ direction
and decodes each consecutive codon with the help of
the incoming aa-tRNAs. After successful decoding, the
aa-tRNA swings fully into the A-site (in a process that is
known as accommodation). Decoding and accommo-
dation are often collectively referred to as ‘Asite occu
pation’. The swing docks the aminoacyl residue into the
peptidyltransferase centre, resulting in rapid peptide
bond formation. The nascent chain is transferred from
the peptidyl-tRNA at the P-site to the charged tRNA at
Decoding
Selection of the cognate
ternary complex of aminoacyl-
tRNA–EF-Tu–GTP on the basis

of correct codon-anticodon
interactions between the
mRNA and tRNA, respectively.
EF‑G and EF4: translocation and
back‑translocation on the bacterial
ribosome
Hiroshi Yamamoto1*, Yan Qin2*, John Achenbach3*,
Chengmin Li2, Jaroslaw Kijek4,
Christian M. T. Spahn1 and Knud H. Nierhaus1,4
Abstract | Ribosomes translate the codon sequence of an mRNA
into the amino acid
sequence of the corresponding protein. One of the most crucial
events is the translocation
reaction, which involves movement of both the mRNA and the
attached tRNAs by one codon
length and is catalysed by the GTPase elongation factor G
(EF‑G). Interestingly, recent
studies have identified a structurally related GTPase, EF4, that
catalyses movement of the
tRNA
2
–mRNA complex in the opposite direction when the ribosome
stalls, which is known as
back‑translocation. In this Review, we describe recent insights
into the mechanistic basis of
both translocation and back‑translocation.
1Institut für Medizinische
Physik und Biophysik, Charité
– Universitätsmedizin Berlin,
Charitéplatz 1,10117 Berlin,
Germany.

2Laboratory of noncoding
RNA, Institute of Biophysics,
Chinese Academy of Science;
15 Datun Road, Beijing
100101, China.
3NOXXON Pharma AG,
Max-Dohrn-Strasse 8–10,
10589 Berlin, Germany.
4Max Planck Institut für
molekulare Genetik,
Ihnestrasse 73, D-14195
Berlin, Germany.
*These authors contributed
equally to this work.
Correspondence to K.H.N.
e-mail: [email protected]
mpg.de
doi:10.1038/nrmicro3176
Published online
23 December 2013
REVIEWS
NATURE REVIEWS | M I C R O B I O LO GY VO LU M E 1
2 | F E B RUA RY 2 0 1 4 | 89
© 2014 Macmillan Publishers Limited. All rights reserved
mailto:nierhaus%40molgen.mpg.de?subject=
mailto:nierhaus%40molgen.mpg.de?subject=
the A-site, thus deacylating the P-site tRNA and extend-
ing the nascent chain by one amino acid. The tRNAs
must then be moved in a step known as translocation. In
this Review, we classify all tRNA conformational states

after peptide bond formation and before translocation
as pre-translocational states (PRE-states). To accom-
modate the next incoming aa-tRNA, the peptidyl-tRNA
at the A-site and the deacylated tRNA at the P-site are
translocated to the P- and E-sites, respectively, and this
is catalysed by EF-G–GTP. The resulting state, in which
the P- and E-sites are occupied and the A-site is vacant,
is called the post-translocational state (POST-state)
(reviewed in REF. 18). The release of the deacylated tRNA
from the E-site is thought to occur after trans location19,20
or, alternatively, on occupation of the A-site with the
next aa-tRNA21–23.
It is possible that ribosomes mistranslocate, which
leads to an arrest in protein synthesis as the ribosome
stalls and thereby blocks the progression of other ribo-
somes on the same mRNA. Recent studies suggest that
such stalled ribosomes can be rescued by a GTPase
known as EF4, which is structurally related to EF-G. This
factor recognizes stalled ribosomes that have a deacylated
tRNA in the E-site and a peptidyl-tRNA in the P-site
(the POST-state) and catalyses a back-translocation
reaction (FIG. 2). The tRNAs are dragged back into the
P- and A-sites, thereby giving the ribosome a second
chance to properly translocate24–26. Other studies suggest
that EF4 can also bind to and mobilize ribosomes that
are stalled in the PRE-state27 (see below). Translation is
terminated when a ribosome encounters a stop codon on
the mRNA, which is recognized by a release factor that
triggers release of the nascent polypeptide. During the
final phase of translation, which is known as recycling,
the 70S ribosome is thought to dissociate into its 30S and
50S subunits, which are re-used for subsequent rounds
of initiation (reviewed in REF. 18).

In this Review, we discuss a number of recent struc-
tural and biochemical studies in bacteria, primarily
Escherichia coli and Thermus thermophilus, that have
enhanced our understanding of the mechanisms of bac-
terial translocation and back-translocation. The binding
modes and functional roles of EF-G and EF4 are dis-
cussed, as well as the proposed physiological relevance
of back-translocation.
EF‑G and EF4
Structural similarities. EF-G and EF-Tu are universal
translation factors, whereas EF4 is found in almost all
bacteria, in mitochondria and chloroplasts, but is absent
in archaea and the cytoplasm of eukaryotes. EF4 is the
third most highly conserved bacterial protein after
EF-Tu and EF-G, with a 55–68% amino acid identity
between different bacterial species24.
The three-dimensional structures of EF-G and the
ternary complex (aa-tRNA–EF-Tu–GTP) are highly
similar (FIG. 3a,b). The five structural domains of EF-G
(FIG. 3a) fold into a structure that resembles the ternary
complex, and domain IV of EF-G corresponds to the
anticodon stem–loop of the tRNA within the ternary
Nature Reviews | Microbiology
50S subunit
L12
L12 stalk
L1
L1 stalk

CP
PTC
Head
Body
Platform
S13
S12E P
A
30S subunit
E
P
A
L11
L10
SRL
23S rRNA 16S rRNA
5S rRNA
Figure 1 | Overall architecture of the large and small subunits of
the bacterial ribosome. Both subunits are shown
from the interface side. The large 50S subunit contains the 23S
ribosomal RNA (rRNA) and 5S rRNA (light grey and dark

grey, respectively), and the small 30S subunit is composed of
the 16S rRNA (light grey). Ribosomal proteins are
represented as coloured ribbons, and those that have specific
roles in translocation, as well as the sarcin–ricin loop (SRL) of
the 23S rRNA and the acceptor ends of A‑ and P‑site tRNAs
within the peptidyl‑transferase centre (PTC), are highlighted
by surface representation. The A‑site, P‑site and E‑site tRNAs
are also shown. For clarity, only the anticodon stem‑loops of
the tRNAs are shown on the 30S subunit. The structures were
produced using coordinates from Protein Data Bank
accessions 2WRL31, 2QA4 (REF. 112), 3A1Y5, 1RQU113 and
3J0T (50S subunit), and 2WRK31 and 3J0U46 (30S subunit).
CP,
central protuberance.
R E V I E W S
90 | F E B RUA RY 2 0 1 4 | VO LU M E 1 2 w w
w.nature.com/reviews/micro
http://www.rcsb.org/pdb/home/home.do
http://www.rcsb.org/pdb/explore.do?structureId=2WRL
http://www.rcsb.org/pdb/explore.do?structureId=2QA4
http://www.rcsb.org/pdb/explore.do?structureId=3A1Y
http://www.rcsb.org/pdb/explore.do?structureId=1RQU
http://www.rcsb.org/pdb/explore/explore.do?structureId=3J0T
http://www.rcsb.org/pdb/explore/explore.do?structureId=2wrk
http://www.rcsb.org/pdb/explore/explore.do?structureId=3J0u
Single‑turnover experiments
Experiments in which the
conditions are set such that the
catalyst (for example, the

ribosome) only undergoes a
single round of catalysis.
complex (FIG. 3b). This is probably the most famous
example of molecular mimicry, which highlights the
need for both EF-G and the ternary complex to occupy
a similar site at the interface of the ribosomal subunits.
Similarly, the domain structure of EF4 is highly related
to that of EF-G (FIG. 3a). Both factors share domains I
(known as the G domain), II, III and V, which are
responsible for ribosome binding and GTPase activ-
ity. In addition, both factors have specific domains:
EFG contains Gʹ (which is a subdomain of domain I)
and domain IV, whereas EF4 has a unique carboxy-
terminal domain (CTD)24. Domain IV of EF-G and the
CTD of EF4 are responsible for mediating the opposing
roles of these two factors in translation (FIG. 3c).
First contacts with the ribosome. The first contacts of
EF-G and EF4 with the ribosome involve the L12 stalk
and seem to follow the same pathway. The substrate for
EF-G is the 70S ribosome in the PRE-state, whereas the
substrate for EF4 is still unclear. One study suggests that
EF4 preferentially binds to the POST-state ribosome,
owing to observations that EF4 binds to the POST-state
with higher affinity than to the PRE-state, and that
EF4-dependent GTP hydrolysis has a higher turnover
rate with POST-state ribosomes than with PRE-state
ribosomes28. However, single-turnover experiments and
single-molecule FRET (Förster resonance energy trans-
fer) measurements suggest that the PRE-state is the
preferential but not the exclusive target of EF4. In this
study, EF4 could compete with EF-G for binding to the
PRE-state27. Thus, EF-G recognizes a specific functional
state, whereas EF4 seems to be more promiscuous in its

specificity.
It is thought that EF-G makes its first ribosomal con-
tact with the CTD of L12 using the Gʹ domain3. The next
step might be shared by other factors (such as EF-Tu and
EF4) and involves contact with the base of the L12 stalk,
resulting in interactions between the L12 CTD and the
amino-terminal domain (NTD) of L11, as demonstrated
by cryo-electron microscopy(cryo-EM)29,30 and X-ray
crystallography31,32. This interaction is controlled by the
universally conserved Pro22 residue of L11, which is in
a trans-configuration when the ribosome is free of GTP-
binding proteins or when a non-GTPase factor is bound
(Supplementary information S1 (figure)). However, when
a G-protein factor such as EF-G, EF-Tu or EF4 binds to
the ribosome, Pro22 adopts the cis-configuration, which
facilitates the L11–L12 interaction. Interestingly, the
trans–cis transition is catalysed by a peptidyl-prolyl cis–trans
isomerase (PPIase) centre, comprising amino acyl residues
that reside mainly in the G domain of translational fac-
tors. Before the factor dissociates from the ribosome after
GTP hydrolysis and inorganic phosphate (Pi) release, the
PPIase activity of the factor stimulates reversion of Pro22
to the trans-configuration33,34.
The early contacts of EF-G with the ribosome pre-
sent a conundrum: EF-G triggers the movement of the
tRNA2–mRNA complex from a PRE-state to the POST-
state, but the initial EF-G contacts with the ribosome
that are essential for activating the ribosome and setting
the tRNA2–mRNA complex in motion are currently
unknown. When EF-G is added to a PRE-state ribo-
some and its dissociation from the ribosome is inhibited
(using the antibiotic fusidic acid or the non-cleavable
GTP analogues GDPNP (guanosine 5ʹtetrahydro
gen triphosphate) or GDPCP (5ʹguanosylmethylene

triphosphate), X-ray and cryo-EM structures have dem-
onstrated that the peptidyl-tRNA has left the A-site
and approaches the P-site, and domain IV of EF-G is
flipped into the A-site, where it functions as a doorstop
to prevent back-translocation of the tRNA2–mRNA
aa-tRNA–EF-Tu–GTP
EF-Tu–GDP + P
i
A-site occupationTranslocation
Peptidyl transfer
EF-G–GTP
E-tRNA
EF-G–GDP + P
i
Elongation
cycle
Initiation
Termination
Recycling
70S
initiation complex

mRNA
fMet-tRNA
50S
30S
APE
APE
EF4–GDP + P
i
EF4–GTP
APE
APE
APE APE
Figure 2 | The functional phases of the ribosome during
translation. The 70S
initiation complex contains the initiator tRNA
(formylmethionine tRNA (fMet‑tRNA)) at
the ribosomal P‑site, which interacts with the start codon
(typically AUG) of the mRNA
via the formation of a codon–anticodon duplex. The 70S
initiation complex enters the
elongation cycle on binding the ternary complex
aminoacyl‑tRNA–elongation factor
Tu–GTP (aa‑tRNA–EF‑Tu–GTP). After successful decoding,
GTP is hydrolysed, EF‑Tu–GDP
and inorganic phosphate (P

i
) leaves the ribosome, and the aa‑tRNA swings into the A‑site
(A‑site occupation). The nascent peptide chain is transferred
from the peptidyl‑tRNA in
the P‑site to the aa‑tRNA in the A‑site, extending the peptide
chain by one amino acid, in
a reaction known as peptidyl transfer. Facilitated by EF‑G–GTP,
the tRNA
2
–mRNA
complex is translocated by a distance of one codon from the A‑
and P‑sites to the P‑ and
E‑sites. EF4–GTP can catalyse a reversal of this step, termed
back‑translocation, in order
to mobilize stalled ribosomes (dashed arrows). When a stop
codon enters the A‑site,
termination of protein synthesis occurs, which is assisted by
release factors. The
ribosome can now enter the recycling phase, after which a 70S
initiation complex is
formed again.
R E V I E W S
2 | F E B RUA RY 2 0 1 4 | 91
http://www.nature.com/nrmicro/journal/v12/n2/full/nrmicro3176
.html#supplementary-information

Single‑molecule FRET
(Single-molecule Förster
resonance energy transfer). A
phenomenon in which energy
induced by light excitation is
transferred from one
fluorophore to another in a
distance-dependent manner,
observed on a single complex
or molecule.
complex 31,35–39 (FIG. 3c; Supplementary information S2
(figure)). In other words, in all previous ribosome struc-
tures with EF-G, the factor has already triggered a first
step of translocation. However, a recent report describes
the structure of a pre-translocational EF-G—ribosome
complex with two tRNAs in hybrid positions. The com-
plex was prepared in the presence of GTP; EF-G disso-
ciation was blocked with the antibiotic fusidic acid and
translocation of the tRNA2–mRNA complex was inhib-
ited with the antibiotic viomycin115. In this PRE-state, the
tip of EF-G domain IV makes strong contacts with the
anticodon loop of the A-site tRNA. A comparison of
the EF-G structure in the POST state31 revealed that
EF-G undergoes a ~20° rotation around the sarcin–ricin
loop (SRL) of the 23S rRNA. This rotation results in
a movement of the tip of domain IV by 20 Å into the
decoding centre during the transition from the PRE- to
the POST-state. Although this study reveals important
insights, it is still unclear what triggers the dramatic
conformational change of EF-G and which contacts
between EF-G and the ribosome (or its ligands) set the
tRNA2-mRNA in motion.

When EF4 is added to POST-state ribosomes, the
structures that are available show the peptidyl-tRNA in
a back-translocated position, having established either
an intermediate state (possibly identical with a trans-
location intermediate25) or a PRE-state28. Thus, a struc-
ture in which EF4 is bound to the POST-state before the
onset of back-translocation is currently lacking.
The specific domains of EF‑G and EF4. Both factors
reduce the activation-energy barrier between PRE- and
POST-states, but the binding of each factor induces
one distinct state of the tRNA2–mRNA complex; EF-G
favours the POST-state and EF4 favours the PRE-state.
EF-G flips domain IV into the A-site, resulting in a door-
stop effect that stabilizes the POST-state. This suggests
that domain IV is essential for translocation. Indeed,
Thermus thermophilus EF-G fragments that lack this
domain are unable to translocate, but they retain GTPase
activity and are able to bind to the ribosome40. As men-
tioned above, EF4 lacks domain IV of EF-G and, as such,
lacks the doorstop function, which is considered to be a
prerequisite to allow for the back-movement of tRNAs
from the POST-state to the PRE-state. This is clearly seen
in the cryo-EM structure28 (Supplementary information S2
(figure), left panel), in which the back-translocated
peptidyl-tRNA in the A-site is attached to the unique
CTD of EF4, whereas domain IV of EF-G would prevent
movement into this position. After movement back into
the A-site, the CTD of EF4 halts the peptidyl-tRNA in
this position, thereby re-establishing the PRE-state. This
halting effect is caused by surface patches of strong posi-
tive charges on EF4 that attract the negative charges of
the A-site tRNA28,41. The CTD of EF4 contacts the inner
side of the elbow and the acceptor-stem down to the
CCA end of the A-site tRNA (Supplementary information
S2 (figure), right panel).

To preserve the reading frame during back-translocation,
maintenance of codon–anticodon interactions is essen-
tial. The presence of a cognate E-site tRNA is crucial
for EF4-mediated back-translocation24 because a back-
translocated tRNA in the P-site must sustain codon–
anticodon interactions; without such interactions, a P-site
tRNA cannot be fixed on the 30S subunit42.
Mechanism of translocation
A wealth of recent structural data describing the dynam-
ics and structural transitions of the ribosome during
translocation now allows for a comprehensive overview
of the mechanisms involved. In this section, we describe
EF4
a
b
P/P
P/P
E/E
A/L
EF4
EF-G
EF-Tu

G Gʹ G II III IV V CTD
c
1–158 159–253 254–289 290–404 405–482 483–603
1– –212 213–313 314–405 tRNA
604–691
1– –188 189–281 291–371 398–486 487–599
EF-G
EF4
G
IIIII
VV
CTD
EF-Tu
G
II
III
A/T-tRNA
EF-G
G

II
III
IV
Gʹ
Common domainsSpecific domains
Backwards
Forwards
Figure 3 | Structure, binding sites and functions of the
elongation factors. a | Domain
organization of elongation factor G (EF‑G), EF4 and EF‑Tu. b |
EF‑G, EF4 and EF‑Tu have a
highly similar domain organization and fold into similar
three‑dimensional structures
(EF‑G, Protein Data Bank (PDB) accession 2WRI31; EF4, PDB
accession 3DEG28; and the
ternary complex aminoacyl‑tRNA−EF‑Tu−GTP, PDB accession
2WRN70). c | EF‑G and EF4
bind to a similar site on the ribosome, but their specific
domains promote opposing effects.
EF‑G catalyses forward movement of the tRNAs from the A/A
and P/P sites to the P/P and
E/E sites, whereas EF4 can reverse this reaction to promote
back translocation, moving
the tRNAs from E/E to P/P and from P/P even beyond the A/A
site toward the L12 stalk.
The latter position is only seen in the presence of EF4 and is
referred to as the A/L position.

R E V I E W S
http://www.rcsb.org/pdb/explore.do?structureId=2WRI
http://www.rcsb.org/pdb/explore/explore.do?structureId=3DEG
http://www.rcsb.org/pdb/explore/explore.do?structureId=2WRN
Peptidyl‑prolyl cis–trans
isomerase
An enzyme that belongs to the
peptidyl-prolyl isomerase
(PPIase) family that catalyses
the transition of a proline
residue between cis and trans
conformations by reducing the
activation-energy barrier that
separates these two
conformations.
Sarcin–ricin loop
(SRL). The loop of helix H95
(G2654–A2665; E. coli
nomenclature), which contains
the longest universally
conserved ribosomal RNA
(rRNA) sequence. Its name
derives from the observations
that removing base A2660 by

the N-glycosidase ricin or
cleaving the 23S rRNA after
G2661 by the RNase α-sarcin
impairs the binding and GTPase
activity of both elongation
factor Tu (EF-Tu) and EF-G,
thereby blocking translation.
Activation‑energy barrier
The energy barrier that
separates reactants and
products in a chemical
reaction.
the role of intersubunit rotation (formerly called ‘ratch
eting’43) and swivelling of the head of the 30S subunit in
translocation, as well as recent insights into the role of
GTP hydrolysis.
The PRE‑states. After peptidebond formation, the ribo-
some can adopt at least three PRE-states; in each state,
both the A- and P-sites on the 30S subunit are occupied
by a tRNA-anticodon stem, whereas the CCA ends of
the tRNAs on the 50S subunit can vary in their location.
In the classical PRE-state, the anticodon stem and the
CCA end of the two tRNAs are positioned in the same
site on each ribosomal subunit (known as A/A for the
A-site tRNA and P/P for the P-site tRNA). The ribosome
spontaneously fluctuates between this classical state and
a rotated state44. Rotation involves a 4–7 ° anticlockwise
rotation of the 30S subunit relative to the 50S subunit,
around a pivot axis close to the middle of helix 44 (h44)43
(FIG. 4a). The intersubunit rotation is coupled to a move-
ment of the CCA end of the P-site tRNA on the 50S sub-
unit to the E-site; simultaneous movement of the CCA
end of the A-site tRNA into the 50S P-site may occur

but is not strictly coupled. The tRNA positions within
the 30S subunit remain unchanged, giving rise to hybrid
sites45. The functional state of a ribosome with a tRNA
in an A/P hybrid site (anticodon stem in the A-site on
the 30S subunit and the CCA end in the P-site on the
50S subunit), and a deacylated tRNA in a P/E hybrid site
(anticodon stem in the P-site of the 30S and the CCA
end in the E-site of the 50S) is known as hybrid state 1
(H1). The third PRE-state (A/A and P/E), which corre-
sponds to movement of the P-site tRNA only, is known
as hybrid state 2 (H2)44,46 (FIG. 4b). Back-rotation of the
30S subunit re-establishes the tRNAs in the classical A/A
and P/P binding positions.
These fluctuations between the various PRE-states
only occur in the absence of EF-G47. All three PRE-
states are substrates for EF-G; EF-G can enter the
sequence of PRE-states (classical, H2 and H1) at any
stage in order to move the tRNA2–mRNA complex to the
POST-state, although EF-G–GTP seems to favour
the 30S rotated state with tRNAs in hybrid positions48,49.
In other words, this sequence of PRE-states is the
only route to the transition state and is thus essential
4–7°
Non-rotated
30S head
30S head
Rotated

Classical
H1 H2
18°
P/P A/A
P/E A/P P/E A/A
30S body
50S subunit
L1 stalk
L12 stalk
L12 stalk
L1 stalk
a
b
Swivelled
Non-swivelled
c
30S body
50S subunit

Figure 4 | The three PRE-states of tRNAs on the ribosome
during translocation. a | Intersubunit rotation of the 30S
subunit, viewed from the 30S solvent side with the 50S subunit
in a fixed position. Rotation of the 30S subunit occurs in
an anticlockwise direction by 4–7 ° and does not depend on
elongation factor G (EF‑G). b | After peptidyl transfer, the
tRNAs can shift between classical and hybrid states. In the
classical pre‑translocational state (PRE‑state) the tRNAs are
located in A/A and P/P positions, in the post‑translocational
state (POST‑state), the tRNA adopts the P/P and E/E
positions. However, in hybrid state 1 (H1), the tRNAs occupy
the A/P and P/E positions and in hybrid state 2 (H2), they are
located in the A/A and P/E positions. c | The second major
conformational change that the 30S undergoes during
translocation is termed swivelling. This movement is
EF‑G‑dependent and involves an anticlockwise rotation of the
30S
head towards the E‑site, which opens the A790 gate and moves
the tRNA
2
–mRNA complex to the POST‑state.
R E V I E W S
2 | F E B RUA RY 2 0 1 4 | 93
Polysomes
mRNAs to which more than
one ribosome is bound.

for translocation47. Inhibition of intersubunit rotation
by crosslinking the 30S and 50S subunits blocks trans-
location50, which shows that this is an essential step in
translocation. Single-molecule FRET measurements have
revealed that there are two populations of pre-translocation
complexes: one in which the ribosome rapidly fluctu-
ates between classical and hybrid states, and another
in which the tRNA positions are long-lived in either
the classical or hybrid state configuration. Following the
addition of EF-G, both populations of pre-translocation
complexes are translocated47, but it is currently unclear
whether only one or both populations exist in vivo.
The transition from PRE‑states to the POST‑state. After
binding to the A-site, a tRNA must translocate twice
(from the A-site to the P-site and from the P-site to the
E-site) during the course of translation, which involves
five distinct combinations of tRNA binding sites:
A/A, A/P, P/P, P/E and E/E. Analyses of ribosomes in
polysomes51,52 or during poly(Phe) synthesis53 have
revealed that at least two tRNAs are always present
on the ribosome during the elongation cycle; in the
PRE-state this corresponds to either the classical state
(A/A and P/P) or the hybrid states (H1 or H2). By con-
trast, only one POST-state exists, which is characterized
by a peptidyl-tRNA in the P/P site and a deacylated tRNA
in the E/E site (Supplementary information S3 (figure)).
A transition intermediate between the PRE- and POST-
states is observed when EF-G is trapped on the ribosome
either by using GDPNP or fusidic acid. This intermedi-
ate is characterized by another large-scale movement of
the ribosome, this time exclusively within the small sub-
unit. It involves an anticlockwise rotation of the 30S head
relative to the 30S body, termed swivelling, which turns
the head by about 18 ° towards the E-site35–39,54–56

(FIG. 4c).
In agreement with measurements of head rotation and
mRNA movement 57, structural data show an almost
complete translocation of the tRNA2–mRNA complex in
the POST-state transition intermediate (TIPOST)35,58. EF-G
dependent GTP hydrolysis is not required for translo-
cation, however, it must occur to ensure that EF-G is
released from the ribosome. A reversal of the head swivel
and 30S back-rotation ensues, thereby establishing the
stable POST-state, in which the tRNAs fully occupy
the P/P and E/E sites.
It is important to note that during translocation of
the tRNA2–mRNA complex, it is the tRNAs that are
physically moved by the ribosome, whereas the mRNA
co-migrates with the tRNAs, mainly owing to codon–
anticodon interactions. This conclusion is supported
by the observation that the main physical contacts
between the mRNA and the ribosome during elongation
are mediated by the codon–anticodon interactions59.
This highlights the importance of codon–anticodon
inter actions not only during decoding at the A-site but
also at the P-site31,60,61 and the E-site22,32,62.
Activation‑energy barrier between PRE‑ and POST‑states.
The PRE-states are separated from the POST-state by
a high activation-energy barrier of 90 kJ mol–1 (REF. 63).
EF-G reduces this barrier by establishing the TIPOST state
and accelerates the translocation rate by 104- to 106-fold
compared with spontaneous translocation (reviewed
in REF. 64). Structures that possibly have a role in estab-
lishing the energy barrier are the bridges that connect
the 30S and 50S subunits at the intersubunit face and the
ribosomal proteins S12 and S13 (REF. 65), which are
located close to the A-site and P-site tRNAs. However,

studies have shown that disruption of some of the
bridges66 or removal of S12 and S13 (REF. 65) only con-
fer a modest increase in the rates of both spontaneous
translocation and back-translocation, which indicates
that they have only a marginal role in establishing the
energy barrier.
By contrast, it has been proposed that a structural
element of the 16S rRNA might have a decisive role in
creating the activation-energy barrier. A ridge of four
bases, G1338-A-N-U1341 (where N represents any
base), in the 30S head and the nucleotide A790 of the 30S
platform form a gate that blocks movement of the tRNA
anticodon stem between the P- and E-sites67 (FIG. 5a,b).
Four of the five nucleotides of this gate, which is referred
to as the A790 gate, are universally conserved in all three
domains of life. The A790 gate is 13.8 Å in width in the
absence of EF-G (closed gate), which is too narrow to
allow the passage of an RNA duplex, such as the anti-
codon stem of the P-site tRNA (which has a diameter
of 20 Å). Therefore, this gate needs to open in order to
enable movement of a P-site tRNA to the E-site. A series
of published functional complexes in the absence and
presence of EF-G have been analysed, which suggest that
the A790 gate is closed in the absence of EF-G and in the
POST-state31,46, but that it opens to a width of approxi-
mately 24 Å exclusively in the intermediate TIPOST state35.
These findings are in clear agreement with a recent crys-
tal structure of translocation intermediates of bacterial
ribosomes68 as well as with a first cryo-EM structure of
a TIPOST ribosome containing two tRNAs116. Opening of
the gate is accompanied and probably caused by the 18 °
swivel of the 30S head68, as the gate is closed in the non-
swivelled PRE-states (FIG. 5b). Swivelling of the 30S head
not only opens the A790 gate, but also induces move-
ment of the tRNA2–mRNA complex on the 30S subunit

from the A- and P-sites to the P- and E-sites, respectively,
as recently shown by ensemble stopped-flow FRET57.
X-ray structures of EF-G–70S complexes have shown
that EF-G remains on the ribosome until the POST-state
is reached31,32. In the POST-state, the A790 gate is closed
(the width of the opening decreases to approximately
15 Å), which indicates that the energy barrier is re-estab-
lished before EF-G leaves the ribosome, thus preventing
back-translocation of the tRNA2–mRNA complex to a
PRE-state. Opening of the A790 gate in the TIPOST transi-
tion state is currently the most attractive explanation for
how EF-G accelerates the translocation reaction, and the
observations that are described here add a key structural
correlate to this hypothesis.
A recent study suggests that transport of the
tRNA2–mRNA complex through the A790 gate is facili-
tated by two universally conserved residues of the 16S
rRNA, C1397 and A1503, which intercalate with mRNA
bases only in the TIPOST transition state. A1503 inserts
R E V I E W S
Swivelling 18°

A790
A1338-
U1341
7°
Rotated
EF-G
EF-G
POST-states
Non-rotated
L1 stalk
Head
4°
P/P A/A P/E A/P pe/E ap/P E/E P/P
Platform
Head
ASL 13.8 Å P/P P/E P/Ppe/E15.5 Å
23.6 Å
E/E 14.7 Å
L1 stalk
Open

PRE-states
POSTClassical
a
b
c
H1 TI POST
Closed Intermediate Intermediate
Figure 5 | Ribosomal conformational changes during
translocation.
a | After peptidyl ‑transfer, the tRNAs are in the classical state
(A/A and
P/P), which establishes an equilibrium with the hybrid states H1
and H2
(H2 not shown) owing to intersubunit rotation. When elongation
factor G
(EF‑G) binds to one of these three PRE‑states, swivelling of the
30S head
is induced, leading to the formation of the translocation
intermediate
TIPOST, which later resolves into the post‑translocational state
(POST‑state)
after a reversal of the head swivel and 30S back‑rotation. Top
row, view of
the 70S ribosome from the 30S solvent side showing the
intersubunit
movements. Bottom row, view from above the 70S ribosome
showing the
tRNA positions. b | Positions of the 16S rRNA base A790,
which forms an

important component of the A790 gate, corresponding to the
ribosomal
states that are shown in part a. The A790 gate is wide enough
(23.6 Å) only
in the TIPOST intermediate state to allow passage of the
anticodon stem of
the tRNA from the P‑ to the E‑site on the 30S subunit during
translocation.
c | Positions of the L1 stalk in the open conformation
(corresponding to
the classical state of the tRNAs), closed conformation
(corresponding
to the hybrid states H1 and H2) and intermediate conformation
(TIPOST and POST); the pivot point for rotation of the L1 stalk
is indicated
by the red dot. The following Protein Data Bank accessions
were used
for parts b and c: PRE classical (column 1), 3J0T and 3J0U46;
PRE H1
(column 2), 3J10, 3J14 (REF. 46) and 3J0L114; TIPOST
(column 3), 2XUX and
2XUY35; POST (column 4), 2WRI and 2WRJ31. ASL,
anticodon stem‑loop;
pe/E, pe indicates that the codon‑anticodon duplex takes a
position
between the P and E sites35; ap/P, indicates a position between
the
A‑ and P‑sites
R E V I E W S
2 | F E B RUA RY 2 0 1 4 | 95

http://www.rcsb.org/pdb/explore.do?structureId=3J0T
http://www.rcsb.org/pdb/explore.do?structureId=3J0U
http://www.rcsb.org/pdb/explore.do?structureId=3J10
http://www.rcsb.org/pdb/explore.do?structureId=2J14
http://www.rcsb.org/pdb/explore.do?structureId=3J0L
http://www.rcsb.org/pdb/explore.do?structureId=2XUX
http://www.rcsb.org/pdb/explore.do?structureId=2XUY
http://www.rcsb.org/pdb/explore.do?structureId=2WRJ
Exocyclic group
A chemical group attached to a
cyclic structure. For example,
adenine contains an exocyclic
amino group at position 6, and
guanine contains a hydroxyl
group at the same position.
between the second and third nucleotide of the E-site
codon and C1397 between the +9 and +10 nucleotides68
(assuming the first nucleotide of the P-site codon is +1).
Both of these 16S rRNA residues might be important for
translocation by preventing back-sliding, thus function-
ing as ‘pawls’ as long as the gate is open (Supplementary
information S4 (figure)), thereby cooperating with the
‘doorstop’ effect of EFG.
Role of the L1 stalk. The L1 stalk undergoes dynamic
structural transitions during the various stages of trans-
location. It can swing by approximately 30 ° around a
pivot point of the stalk (located at the base of helix 76;
(H76)), whereas the tip of the stalk can move by about
50 Å towards the intersubunit space. Three different

L1 positions are observed31,35,37,46,69,70 (FIG. 5c): it adopts
an open position during decoding and in the classical
PRE-state; a closed position in the hybrid PRE-states
(H1 and H2); and an intermediate position in the
TIPOST and POST-state. Thus, the L1 stalk is proposed
to function as a gate for the deacylated E-site tRNA,
blocking release of the tRNA when it is in the closed
position, but enabling free dissociation when it is in the
open position71. This hypothesis is consistent with the
allosteric three-site model for the elongation cycle72,
which posits that the E-site tRNA is only released
when the A-site becomes occupied with the next
aa-tRNA21–23,73, coinciding with opening of the L1 stalk
during decoding. The coupling of different transloca-
tional states to distinct positions of the L1 stalk is clearly
visible in X-ray and cryo-EM structures46, whereas FRET
measurements have indicated that, at least under the
in vitro conditions that were used, anticlockwise subunit
rotation and L1 closure are only loosely coupled74,75.
As the L1 stalk is in contact with the deacylated
tRNA in the H1, TIPOST and the POST-states (FIG. 5c), it
has been suggested that it might carry the tRNA from
the P-site to the E-site during translocation37,69. However,
L1 is not an essential protein and its removal only leads
to a 50% reduction in the growth rate of E. coli, which
corresponds to a 50% reduction in poly(Phe) synthesis
in vitro76. Furthermore, deletion of the L1 gene has no
effect on EF-G-dependent translocation77, which sug-
gests that the L1 protein is unlikely to have an active role
in tRNA transport from the P-site to the E-site. However,
the importance of the L1 rRNA-binding site, which also
makes contact with the tRNA, is unknown.
GTP hydrolysis. GTP hydrolysis on EF-G and EF4 is

mediated by domains that are shared by both factors
(FIG. 3c) and therefore probably follows identical path-
ways. GTP cleavage is not essential for tRNA movement,
although EF-G-mediated translocation occurs at least
fourfold faster with GTP compared with GDPNP78–80.
How this acceleration is achieved is unclear, but it is
modest, considering that EF-G-dependent transloca-
tion (with or without GTP hydrolysis) is at least four
orders of magnitude faster than spontaneous transloca-
tion64 (BOX 1). GTP hydrolysis is primarily thought to be
important for fast and efficient release of EF-G, which
is required to enable the incoming ternary complex to
bind to the ribosome. Although EF-G dependent GTP
cleavage can precede translocation78, GTP hydrolysis and
Pi release are not strictly coupled to the movement of the
tRNA2–mRNA complex81.
Residues in the SRL of the 50S sub unit are impor-
tant for factor binding and are involved in trig-
gering GTP cleavage36,38,39,82,83. The SRL comprises
the 2660 loop of H95 of the 23S rRNA, which contains the
longest universally conserved stretch of 12 RNA nucleo-
tides82,84. Ribosomes in which the SRL is cleaved by the
RNase toxin αsarcin, as well as studies of SRL mutants,
have revealed that the SRL is important for EF-Tu binding
and essential for anchoring EF-G to the ribosome during
the various conformational changes of the translocation
process82,85,86. It has been shown that the exocyclic group
of A2660, rather than the actual chemistry of this base,
is crucial for GTP hydrolysis87, although the effects are
indirect, as A2660 points away from the GTPase centre.
Our current understanding for the mechanism that
triggers GTPase activity involves the hydrophobic resi-
dues Ile19 and Ile61 (E. coli nomenclature) of EF-G. These
two amino acids are proposed to form a hydrophobic gate,

which needs to open to enable His92 to approach GTP.
Box 1 | Spontaneous translocation and back‑translocation
in vitro
Spontaneous translocation has been observed by several
groups101,102, but it occurs at a
rate that is at least four orders of magnitude slower than
translocation catalysed by
elongation factor G (EF‑G)–GTP (reviewed in REF. 64).
Thiol‑modifying reagents, such as
p‑chloromercuribenzoate103, or the absence of the ribosomal
proteins S12 and S13 from
the small ribosomal subunit65 accelerate the rate of
spontaneous translocation, but the
rate is still orders of magnitude slower than translocation
catalysed by EF‑G–GTP.
Addition of deacylated tRNAs cognate to the codon at the E‑site
can induce
back‑translocation of ribosomes from the post‑translocational
state (POST‑state) to a
pre‑translocational state (PRE‑state)104,105. However, direct
binding of a deacylated
tRNA to the E‑site does not occur in vivo because deacylated
tRNAs are always
complexed with components of the translational machinery,
such as the ribosomes or
tRNA synthetases106. This is true despite the large fraction
(30%) of deacylated tRNAs
that are observed in minimal media107; in rich media, the
percentage might be
substantially lower. Thus, there is almost no pool of free
deacylated tRNA under
non‑starvation conditions because most of the tRNAs that are
not bound to ribosomes

or synthetases are fully charged with amino acids106,108.
Interestingly, when EF‑G is removed from a population of
ribosomes in the
post‑translocational state (POST‑state), the ribosomes partially
fall back into the pre‑
translocational state (PRE‑state)95,104. This suggests that the
energetic levels of PRE‑
and POST‑states are very similar, and that, in some cases, the
PRE‑state might be
slightly thermodynamically favoured over the POST‑state. The
rates of spontaneous
forward and reverse translocation are similar (about 0.5 to 2 ×
10–3 s–1), which suggests
that even small energetic increments could shift the equilibrium
to either side. Such
shifts are observed with antibiotics, which was first noted with
sparsomycin‑triggered
translocation109. Other examples are streptomycin, neomycin,
paromomycin and
viomycin, which shift the ribosome from the POST‑state to a
PRE‑state, whereas
hygromycin favours the POST‑state95,104.
The induction of back translocation by the addition of
deacylated tRNAs to the
POST‑state has been analysed in a time‑resolved cryo‑electron
microscopy study, and
the observed structures have been used to describe the
conformational changes that
occur during canonical forward translocation110. However, the
validity of these
interpretations is questionable for two main reasons. First, the
induced back translocation
is more than four orders of magnitude slower than an enzymatic
translocation. Second,

the energetic barriers between the various identified states are
low (the energy
landscape is flat, in striking contrast to EF‑G‑dependent
translocation, which has
high‑energy barriers between PRE‑ and POST‑states70,111).
Therefore, there might only
be a partial overlap between the structural intermediates of
enzymatic translocation
and non‑enzymatic back translocation.
R E V I E W S
His92 positions a water molecule to attack the γphosphate
of GTP. Three recent studies show His92 in an identical
orientation pointing to the γphosphate of the GTP ana-
logue, GDPCP38,39,88 (FIG. 6a,b), which provides compelling
evidence that these structures represent an active state of
the GTPase centre. The studies also suggest how inter-
actions between residues of P-loop and switch I and II of
EF-G cooperate with the SRL to open the hydrophobic
gate. This enables His92 to move towards the γphosphate
of GDPCP (reaching a distance of ~3 Å), which is stabi-
lized by hydrogen bonding to A2662. As a similar His92
arrangement was observed in crystal structures of isolated
EF-Tu–GTP89, it is thought that GTPase activation

follows the same mechanism in EF-G and EF-Tu.
Because the ‘active’ orientation of His92 is only
observed in three translocation intermediates38,39,88 and
the essential residues of the GTPase centre are positioned
so that they are ready to cleave GTP, the time of GTP
cleavage can now be identified: it occurs just before, or
during, the formation of TIPOST (REF. 35), before the A790
gate fully opens39. Interestingly, His92 occupies a dif-
ferent orientation in one of the recent structures of the
transition intermediates68: it is located 9 Å away from
the γphosphate and points away from the bound nucleo
tide, which indicates an inactive GTPase centre (FIG. 6b),
similar to two unrotated states with an inactive GTPase
centre, the POST-state31 and the EF-Tu–70S complex70
after GTP cleavage. The observation of an open A790
gate in the translocation intermediate38,39,88 and an inac-
tive GTPase centre (which occurs in the POST-state31)
suggests that this structure represents a late transition
intermediate just before arriving at the POST-state.
EF4 and back‑translocation
The data available on 70S–EF4 complexes and the mecha-
nism of EF4 dependent back-translocation are still insuf-
ficient to provide a detailed description of the structural
transitions that occur during this reaction. For example,
the molecular basis by which EF4 might open the A790
gate to facilitate a reversal of the E-site tRNA to the
P-site is unknown. However, a model for EF4-mediated
back-translocation has been proposed28. By examining
EF4-mediated back-translocation of POST-state ribo-
somes, the tRNAs were observed in a PRE-state that
was unique to back-translocation. In this state, a
deacylated tRNA was found in the P/P site, whereas the
peptidyl-tRNA had moved beyond the A/A site to a posi-

tion known as the A/L site (L for LepA, the original name
of EF4 (REF. 28)). In this position, the elbow of the A-site
tRNA is displaced by ~14 Å towards the L12 stalk (FIG. 3c).
When EF4 is released, the peptidyl-tRNA is predicted to fall
back into the A/A position, which might be an important
step for the re-mobilization of a stalled ribosome. These
data indicate that EF4-dependent back-translocation
is not a simple reversal of translocation; this view is also
supported by FRET analysis of back -translocation25.
SRL
His92
Ile61
P-loop
a b
GTPase
centre
of EF-G
GDPCP
His18
Ile19
Asp20
Inactive His92
Active GTPase
conformation

Inactive GTPase
conformation
Active His92
Active His92
SW II
γ-Ph
A2662G2661
γ-Ph
SW I
Figure 6 | Mechanism of GTP hydrolysis on EF-G. a | The active
GTPase centre of EF‑G in complex with a translocation
intermediate in the presence of the non‑cleavable GTP analogue
GDPCP (5ʹ‑guanosyl‑methylene‑triphosphate). The
functional motifs of EF‑G are shown, namely the P‑loop, switch
I (SW I) and switch II (SW II), together with a portion of the
ribosomal sarcin–ricin loop (SRL). Interactions of His18 and
the ‘catalytic’ His92 (Escherichia coli nomenclature) with
nucleotides of the SRL are shown as dashed lines. In the active
GTPase state, the catalytic His92 is oriented towards the
γ‑phosphate (γ‑Ph) of GDPCP (distance 3 Å). Note that His18
and His92 interact with the backbone of the SRL
(phosphate‑OH groups of G2661 and A2662, respectively;
Protein Data Bank (PDB) accessions 4BTC and 4BTD32).
b | Left panel, His92 from three crystal structures of the
translocation intermediate38,39,88 have been aligned according
to
the bound nucleotide; His92 occupies an almost identical
position in all three structures, which corresponds to an active

GTPase centre (PDB accessions 4BTC38, 4JUW39 and
4KIX88). Right panel, in one translocation intermediate (PDB
accession
3SFS68), His92 points away from the γ‑phosphate, similarly to
the His92 (orange) in the inactive GTPase centre of the
POST‑state (PDB accession 2WRI31).
R E V I E W S
2 | F E B RUA RY 2 0 1 4 | 97
http://www.rcsb.org/pdb/explore.do?structureId=4BTC
http://www.rcsb.org/pdb/explore.do?structureId=4btd
http://www.rcsb.org/pdb/explore.do?structureId=4BTC
http://www.rcsb.org/pdb/explore.do?structureId=4JUW
http://www.rcsb.org/pdb/explore.do?structureId=4KIX
http://www.rcsb.org/pdb/explore.do?structureId=3SFS
Physiological relevance of back‑translocation. What is the
physiological relevance of a factor that can reverse
the canonical translocation reaction? The wide distribu-
tion and high conservation of EF4 in bacteria argue for
an important function. However, deletion of the encod-
ing gene (lepA) in E. coli has no phenotype when cells
are grown in either rich or poor medium90. A first hint
of the importance of EF4 came from a report showing
that lepA is one of ten genes that are essential for survival
of Helicobacter pylori in the hostile acidic environment of
the stomach muscosa, which has a pH of 4 (REF. 91). Low
pH is equivalent to high H+ concentrations, suggesting

that EF4 could have an important physiological role
at high ionic strength, which could be caused by high
intracellular levels of K+ and Mg2+. For example, under
hyperosmotic conditions, the intracellular concentra-
tions of Mg2+ and K+ (together with glutamate) increase
three- to sevenfold92,93. A change in K+ concentration
over a wide range has only a marginal effect on protein
synthesis in vitro. By contrast, an increase in Mg2+ leads
to the ribosome becoming more compact and less flex-
ible, resulting in an increase in error rate and a decrease
in translation rate owing to both decelerated ribosome
movement and an increase in the number of stalled
ribosomes on mRNAs94,95.
A recent analysis showed that EF4 has no effect on the
rate of elongation under physiological Mg2+ concentra-
tions (4.5 mM), whereas it accelerates protein synthesis by
about fivefold when the Mg2+ concentration is increased
threefold in vitro26. These data suggest that EF4 might
function in recognizing ribosomes that are stalled either
in the PRE- or the POST-state, and that it re-mobilizes
them, thus recycling both the mRNA and the associ-
ated ribosomes of the polysome. It was shown in vivo
and in vitro that EF4 does not reduce misincorporation
errors26,96, whereas a previous study 24 showed that EF4
increases the fraction of functional proteins produced in
the cell (which could be due to a reduction in misincor-
poration rate). However, this effect was only observed at
increased Mg2+ concentrations, but not in the presence
of aminoglycosides, which are known to increase the
misincorporation rate97. A possible explanation is that
EF4 indirectly leads to increased synthesis of functional
proteins by preventing the misfolding of proteins (rather
than counteracting misincorporations). Consistent with
this hypothesis, protein misfolding is known to occur
when the ribosome is subject to unscheduled stalls98.

The relationship between increased Mg2+ concentra-
tion and EF4 activity is consistent with the pheno type
that is associated with LepAdepleted (ΔlepA) E. coli
mutants grown in competition with wild-type cells in
media containing 100 mM Mg 2+ at pH 6. Wild-type
cells show a strong growth advantage under these con-
ditions, whereas there was no substantial difference
between wildtype and ΔlepA mutants in medium that
contains 1 mM Mg2+ at pH 7 (REF. 26). Surprisingly, the
intracellular concentration of EF4 in vivo is the same
during growth under physiological and hyperosmotic
conditions. However, during physiological growth con-
ditions, almost all EF4 proteins are associated with the
membrane, whereas the majority of EF4 is found in
the cytoplasm under hyperosmotic conditions26. This sug-
gests that the membrane is a storage vessel for EF4 under
optimal growth conditions and that EF4 is liberated when
the Mg2+ concentration rises to unfavourable levels.
The lack of EF4 orthologues in archaea and the cyto-
plasm of eukaryotes might be related to the fact that
hyperosmotic conditions generally leave the intra cellular
concentrations of K+ and Mg 2+ largely unchanged99.
However, the EF4 orthologue in mitochondria and
chloro plasts might have the same function as EF4 in
bacteria. Depending on the rates of respiration and
photo synthesis, the inner membrane potential of these
organelles can change sharply, which affects the pH of the
cytosol close to the membrane where protein synthesis
occurs. Similarly to E. coli EF4, the mitochondrial homo-
logue Guf1 is found at the inner membrane. A Δguf1
yeast strain has a reduced growth rate under suboptimal
temperatures and starvation conditions. Protein synthe-
sis is only marginally perturbed in the knockout strain,

but the production of functional proteins is reduced100.
Similarly to bacterial EF4 (REF. 98), this would suggest that
Guf1 might also reactivate stalled ribosomes and thereby
enhance the production of functional proteins. The pro-
posed ability of EF4 to resolve stalled ribosomes when the
pH and Mg2+ concentrations are unfavourable has two
important consequences: it could accelerate protein syn-
thesis by mobilizing stalled ribosomes and it could also
prevent co-translational misfolding. However, it should
be noted that the evidence of a role for EF4 in rescuing
stalled ribosomes is suggestive rather than direct, thus
further studies are required to confirm this potential role.
Summary and outlook
The opposing functions of EF-G and EF4, which trig-
ger translocation and back-translocation, respectively,
are mediated by their specific domains (domain IV
of EF-G and the CTD of EF4 (FIG. 3)). During trans-
location, EF-G reduces the activation-energy barrier
between the PRE- and POST-states, probably by open-
ing of the A790 gate during swivelling (FIG. 5B), which
enables the tRNAs to translocate to the POST-state.
Domain IV of EF-G enters the A-site as soon as the
tRNAs have moved from the PRE- to the POST-state
and thereby blocks back translocation. The exact details
of the mechanism of EF4-mediated back-transloca-
tion of the tRNA2–mRNA complex have not yet been
resolved. Deacylated tRNA and peptidyl-tRNA in the
E- and P-sites are moved to the P- and A-sites, respec-
tively, and it seems as though the CTD of EF4 halts the
peptidyl-tRNA at the A-site and drags the elbow of
the peptidyl-tRNA beyond the A-site to the A/L posi-
tion (FIG. 3c; Supplementary information S2 (figure)).
The data suggest that EF4-triggered back translocation
is not a simple reversal of translocation. However, we
have much to learn about the structural transitions

that occur during this reaction before the principles
of back-translocation can be elucidated. Furthermore,
evidence so far suggests that EF4 can bind to both
PRE- and POST-state ribosomes, but whether one or
the other is the preferential target of EF4 remains an
unanswered question27,28.
R E V I E W S
1. Kaltschmidt, E. & Wittmann, H. G. Ribosomal proteins
XII. Number of proteins in small and large subunits of
Escherichia coli as determined by two‑dimensional gel
electrophoresis. Proc. Natl Acad. Sci. USA 67,
1276–1282 (1970).
2. Ilag, L. L. et al. Heptameric (L12)6/L10 rather than
canonical pentameric complexes are found by tandem
MS of intact ribosomes from thermophilic bacteria.
Proc. Natl Acad. Sci. USA 102, 8192–8197 (2005).
3. Diaconu, M. et al. Structural basis for the function of
the ribosomal L7/12 stalk in factor binding and
GTPase activation. Cell 121, 991–1004 (2005).
4. Maki, Y. et al. Three binding sites for stalk protein
dimers are generally present in ribosomes from
archaeal organism. J. Biol. Chem. 282, 32827–32833
(2007).

5. Naganuma, T. et al. Structural basis for translation
factor recruitment to the eukaryotic/archaeal
ribosomes. J. Biol. Chem. 285, 4747–4756 (2010).
6. Pettersson, I., Hardy, S. J. S. & Liljas, A. The ribosomal
protein L8 is a complex of L7/L12 and L10. FEBS Lett.
6, 135–138 (1976).
7. Ogle, J. M. et al. Recognition of cognate transfer RNA
by the 30S ribosomal subunit. Science 292, 897–902
(2001).
8. Demeshkina, N., Jenner, L., Westhof, E., Yusupov, M.
& Yusupova, G. A new understanding of the decoding
principle on the ribosome. Nature 484, 256–259
(2012).
9. Nissen, P., Hansen, J., Ban, N., Moore, P. B. &
Steitz, T. A. The structural basis of ribosome activity in
peptide bond synthesis. Science 289, 920–930
(2000).
10. Voorhees, R. M., Weixlbaumer, A., Loakes, D.,
Kelley, A. C. & Ramakrishnan, V. Insights into
substrate stabilization from snapshots of the peptidyl
transferase center of the intact 70S ribosome. Nature
Struct. Mol. Biol. 16, 528–533 (2009).
11. Hiller, D. A., Singh, V., Zhong, M. & Strobel, S. A. A
two‑step chemical mechanism for ribosome‑catalysed
peptide bond formation. Nature 476, 236–239
(2011).
12. Pech, M. & Nierhaus, K. H. The thorny way to the
mechanism of ribosomal peptide‑bond formation.
Chembiochem 13, 189–192 (2012).

13. Nomura, M. Assembly of bacterial ribosomes. Science
179, 864–873 (1973).
14. Nierhaus, K. H. The assembly of prokaryotic
ribosomes. Biochimie 73, 739–755 (1991).
15. Takyar, S., Hickerson, R. P. & Noller, H. F. mRNA
helicase activity of the ribosome. Cell 120, 49–58
(2005).
16. Wower, I., Wower, J. & Zimmermann, R. Ribosomal
protein L27 participates in both 50S subunit
assembly and the peptidyl transferase reaction.
J. Biol. Chem. 273, 19847–19852 (1998).
17. Diedrich, G. et al. Ribosomal protein L2 is involved in
the association of the ribosomal subunits, tRNA
binding to A and P sites and peptidyl transfer. EMBO J.
19, 5241–5250 (2000).
18. Schmeing, T. & Ramakrishnan, V. What recent
ribosome structures have revealed about the
mechanism of translation. Nature 461, 1234–1242
(2009).
19. Robertson, J. M. & Wintermeyer, W. Mechanism of
ribosomal translocation. tRNA binds transiently to an
exit site before leaving the ribosome during
translocation. J. Mol. Biol. 196, 525–540 (1987).
20. Uemura, S. et al. Real‑time tRNA transit on single
translating ribosomes at codon resolution. Nature
464, 1012–1017 (2010).
21. Chen, C. et al. Allosteric versus spontaneous exit‑site

(E‑site) tRNA dissociation early in protein synthesis.
Proc. Natl Acad. Sci. USA 108, 16980–16985
(2011).
22. Triana‑Alonso, F. J., Chakraburtty, K. & Nierhaus, K. H.
The elongation factor 3 unique in higher fungi and
essential for protein biosynthesis is an E site factor.
J. Biol. Chem. 270, 20473–20478 (1995).
23. Dinos, G., Kalpaxis, D. L., Wilson, D. N. &
Nierhaus, K. H. Deacylated tRNA is released from the
E site upon A site occupation but before GTP is
hydrolyzed by EF‑Tu. Nucleic Acids Res. 33,
5291–5296 (2005).
24. Qin, Y. et al. The highly conserved LepA is a ribosomal
elongation factor that back‑translocates the ribosome.
Cell 127, 721–733 (2006).
This paper reports the discovery that EF4 is a
ribosomal back-translocase.
25. Liu, H., Pan, D., Pech, M. & Cooperman, B. S.
Interrupted catalysis: the EF4 (LepA) effect on back‑
26. Pech, M. et al. Elongation factor 4 (EF4/LepA)
accelerates protein synthesis at increased Mg2+
concentrations. Proc. Natl Acad. Sci. USA 108,
3199–3203 (2011).
An in vivo and in vitro study suggesting that EF4
rescues stalled ribosomes.
27. Liu, H. et al. The conserved protein EF4 (LepA)
modulates the elongation cycle of protein synthesis.
This study suggests that the PRE-state is the

preferential target for EF4.
28. Connell, S. R. et al. A new tRNA intermediate revealed
on the ribosome during EF4‑mediated back‑
translocation. Nature Struct. Mol. Biol. 15, 910–915
(2008).
This study shows the cryo-EM structure of EF4 in
action.
29. Datta, P. P., Sharma, M. R., Qi, L., Frank, J. &
Agrawal, R. K. Interaction of the Gʹ domain of
elongation factor G and the C‑terminal domain of
ribosomal protein L7/L12 during translocation as
revealed by cryo‑EM. Mol. Cell 20, 723–731 (2005).
30. Yokoyama, T. et al. Structural insights into initial and
intermediate steps of the ribosome‑recycling process.
EMBO J. 31, 1836–1846 (2012).
31. Gao, Y. G. et al. The structure of the ribosome with
elongation factor G trapped in the posttranslocational
state. Science 326, 694–699 (2009).
This paper shows the crystal structure of an
EF-G–70S complex in the POST-state.
32. Feng, S., Chen, Y. & Gao, Y. G. Crystal structure of 70S
ribosome with both cognate tRNAs in the E and P
sites representing an authentic elongation complex.
PLoS ONE 8, e58829 (2013).
33. Wang, L. et al. A conserved proline switch on the
ribosome facilitates the recruitment and binding of
trGTPases. Nature Struct. Mol. Biol. 19, 403–410
(2012).
34. Zhang, D. et al. Common chaperone activity in the

G‑domain of trGTPase protects L11‑L12 interaction on
the ribosome. Nucleic Acids Res. 40, 10851–10865
(2012).
35. Ratje, A. H. et al. Head swivel on the ribosome
facilitates translocation by means of intra‑subunit
tRNA hybrid sites. Nature 468, 713–716 (2010).
This paper reports the first visualization of distinct
structural intermediates during translocation.
36. Connell, S. R. et al. Structural basis for interaction of
the ribosome with the switch regions of GTP‑bound
elongation factors. Mol. Cell 25, 751–764 (2007).
37. Khade, P. K., Shi, X. & Joseph, S. Steric
complementarity in the decoding center Is Important
for tRNA selection by the ribosome. J. Mol. Biol. (2013).
38. Chen, Y., Feng, S., Kumar, V. & Gao, Y.‑G. Structure of
EF‑G–ribosome complex in a pre‑translocation state.
Nature Struct. Mol. Biol. 20, 1077–1084 (2013).
This paper (along with references 39, 68 and 88) is
one of four simultaneously published crystal
structures of translocation intermediates.
39. Tourigny, D. S., Fernandez, I. S., Kelley, A. C. &
Ramakrishnan, V. Elongation factor G bound to the
ribosome in an intermediate state of translocation.
Science 340, 1235490 (2013).
40. Martemyanov, K. A. & Gudkov, A. T. Domain IV of
elongation factor G from Thermus thermophilus is
strictly required for translocation. FEBS Lett. 452,
155–159 (1999).
41. Evans, R. N., Blaha, G., Bailey, S. & Steitz, T. A.

The structure of LepA, the ribosomal back translocase.
This paper shows the crystal structure of EF4.
42. Schäfer, M. A. et al. Codon–anticodon interaction at
the P site is a prerequisite for tRNA interaction with
the small ribosomal subunit. J. Biol. Chem. 277,
19095–19105 (2002).
43. Frank, J. & Agrawal, R. K. A ratchet‑like inter‑subunit
reorganization of the ribosome during translocation.
Nature 406, 318–322 (2000).
44. Munro, J. B., Altman, R. B., O’Connor, N. &
Blanchard, S. C. Identification of two distinct hybrid
state intermediates on the ribosome. Mol. Cell 25,
505–517 (2007).
This study is the first demonstration of the two
ribosomal hybrid states, H1 and H2.
45. Moazed, D. & Noller, H. F. Intermediate states in the
movement of transfer RNA in the ribosome. Nature
342, 142–148 (1989).
46. Agirrezabala, X. et al. Structural characterization of
mRNA–tRNA translocation intermediates. Proc. Natl
Acad. Sci. USA 109, 6094–6099 (2012).
47. Chen, C. et al. Single‑molecule fluorescence
measurements of ribosomal translocation dynamics.
Mol. Cell 42, 367–377 (2011).
48. Cornish, P. V., Ermolenko, D. N., Noller, H. F. & Ha, T.
Spontaneous intersubunit rotation in single
ribosomes. Mol. Cell 30, 578–588 (2008).

49. Munro, J. B., Altman, R. B., Tung, C. S.,
Sanbonmatsu, K. Y. & Blanchard, S. C. A fast dynamic
mode of the EF‑G‑bound ribosome. EMBO J. 29,
770–781 (2010).
50. Horan, L. H. & Noller, H. F. Intersubunit movement is
required for ribosomal translocation. Proc. Natl Acad.
Sci. USA 104, 4881–4885 (2007).
51. Warner, J. R. & Rich, A. The number of soluble RNA
molecules on reticulocyte polyribosomes. Proc. Natl
Acad. Sci. USA 51, 1134–1141 (1964).
52. Remme, J., Margus, T., Villems, R. & Nierhaus, K. H.
The third ribosomal tRNA‑binding site, the E site, is
occupied in native polysomes. Eur. J. Biochem. 183,
281–284 (1989).
53. Rheinberger, H.‑J. & Nierhaus, K. H. Testing an
alternative model for the ribosomal peptide elongation
cycle. Proc. Natl Acad. Sci. USA 80,
4213–4217 (1983).
54. Spahn, C. M. et al. Localization of the ribosomal
protection protein Tet(O) on the ribosome and the
mechanism of tetracycline resistance. Mol. Cell. 7,
1037–1045 (2001).
55. Spahn, C. M. et al. Domain movements of elongation
factor eEF2 and the eukaryotic 80S ribosome facilitate
tRNA translocation. EMBO J. 23, 1008–1019 (2004).
Although we now know a great deal about the
mechanistic details of translocation, there is one major
step about which we still know very little, although it is
crucial for a complete understanding of both translo-

cation and back-translocation: the early contacts of an
elongation factor with the ribosome. A recent paper115
provides some answers for EF-G: a PRE-state containing
EF-G and two tRNAs in the hybrid sites shows an ~12°
rotation of the 30S subunit but a negligible swivelling
of head by only 3°. These are features of a ribosome–
EF-G complex before the formation of a translocational
intermediate. The tip of EF-G domain IV makes strong
contacts with the anticodon loop of the A-site tRNA but
has not yet entered the A-site. Despite these important
insights, it is still unclear which ribosomal contacts are
required to initiate the translocation process, includ-
ing 30S head swivelling, and to coordinate movement
of both the tip of domain IV and the tRNA2–mRNA
complex.
Solving this problem will represent major progress in
our understanding of translocation, which will also have
implications for back-translocation.
R E V I E W S
2 | F E B RUA RY 2 0 1 4 | 99
56. Zhang, W., Dunkle, J. A. & Cate, J. H. Structures of the
ribosome in intermediate states of ratcheting. Science
325, 1014–1017 (2009).
57. Guo, Z. & Noller, H. F. Rotation of the head of the 30S
ribosomal subunit during mRNA translocation. Proc.

Natl Acad. Sci. USA 109, 20391–22034 (2012).
58. Ramrath, D. J. et al. The complex of tmRNA‑SmpB
and EF‑G on translocating ribosomes. Nature 485,
526–529 (2012).
59. Alexeeva, E. V., Shpanchenko, O. V., Dontsova, O. A.,
Bogdanov, A. A. & Nierhaus, K. H. Interaction of
mRNA with the Escherichia coli ribosome: accessibility
of phosphorothioate‑containing mRNA bound to
ribosomes for iodine cleavage. Nucleic Acids Res. 24,
2228–2235 (1996).
60. Wurmbach, P. & Nierhaus, K. H. Codon–anticodon
interaction at the ribosomal P (peptidyl‑tRNA) site.
61. Lührmann, R., Eckhardt, H. & Stöffler, G. Codon–
anticodon interaction at the ribosomal peptidyl‑site.
Nature 280, 423–425 (1979).
62. Jenner, L., Rees, B., Yusupov, M. & Yusupova, G.
Messenger RNA conformations in the ribosomal E site
revealed by X‑ray crystallography. EMBO Rep. 8,
846–850 (2007).
63. Schilling‑Bartetzko, S., Bartetzko, A. & Nierhaus, K. H.
Kinetic and thermodynamic parameters for transfer
RNA binding to the ribosome and for the translocation
reaction. J. Biol. Chem. 267, 4703–4712 (1992).
64. Shoji, S., Walker, S. E. & Fredrick, K. Ribosomal
translocation: one step closer to the molecular
mechanism. ACS Chem. Biol. 20, 93–107 (2009).
65. Cukras, A. R., Southworth, D. R., Brunelle, J. L.,

Culver, G. M. & Green, R. Ribosomal proteins S12 and
S13 function as control elements for translocation of
the mRNA:tRNA complex. Mol. Cell 12, 321–328
(2003).
66. Liu, Q. & Fredrick, K. Contribution of intersubunit
bridges to the energy barrier of ribosomal translocation.
Nucleic Acids Res. 41, 565–574 (2013).
67. Schuwirth, B. S. et al. Structures of the bacterial
ribosome at 3.5 Å resolution. Science 310, 827–834
(2005).
This paper reports the first crystal structure of an
E. coli ribosome and also demonstrates the closure
of the A790 gate.
68. Zhou, J., Lancaster, L., Donohue, J. P. & Noller, H. F.
Crystal structures of EF‑G‑ribosome complexes
trapped in intermediate states of translocation.
Science 340, 1236086 (2013).
This paper reports the crystal structure of a
translocation intermediate showing 16S rRNA
bases C1397 and A1503 intercalating with the
mRNA, thereby preventing back-translocation.
69. Valle, M. et al. Locking and unlocking of ribosomal
motions. Cell 114, 123–134 (2003).
70. Schmeing, T. M. et al. The crystal structure of the
ribosome bound to EF‑Tu and aminoacyl‑tRNA.
Science 326, 688–694 (2009).
71. Trabuco, L. G. et al. The role of L1 stalk‑tRNA
interaction in the ribosome elongation cycle. J. Mol.
Biol. 402, 741–760 (2010).

72. Nierhaus, K. H. The allosteric three‑site model for the
ribosomal elongation cycle: features and future.
Biochemistry 29, 4997–5008 (1990).
73. Rheinberger, H.‑J. & Nierhaus, K. H. Allosteric
interactions between the ribosomal transfer RNA‑
binding sites A and E. J. Biol. Chem. 261, 9133–9139
(1986).
74. Cornish, P. V. et al. Following movement of the L1
stalk between three functional states in single
ribosomes. Proc. Natl Acad. Sci. USA 106,
2571–2576 (2009).
75. Munro, J. B. et al. Spontaneous formation of the
unlocked state of the ribosome is a multistep process.
76. Subramanian, A. R. & Dabbs, E. R. Functional studies
on ribosomes lacking protein L1 from mutant
Escherichia coli. Eur. J. Biochem. 112, 425–430
(1980).
77. Sander, G. Ribosomal protein L1 from Escherichia coli.
Its role in the binding of tRNA to the ribosome and in
elongation factor G‑dependent GTP hydrolysis. J. Biol.
Chem. 258, 10098–10103 (1983).
78. Rodnina, M. V., Savelsbergh, A., Katunin, V. I. &
Wintermeyer, W. Hydrolysis of GTP by elongation
factor G drives tRNA movement on the ribosome.
Nature 385, 37–41 (1997).
79. Zavialov, A. V. & Ehrenberg, M. Peptidyl‑tRNA
regulates the GTPase activity of translational factors.
Cell 114, 113–122 (2003).

80. Pan, D., Kirillov, S. V. & Cooperman, B. S. Kinetically
competent intermediates in the translocation step of
protein synthesis. Mol. Cell 25, 519–529 (2007).
81. Peske, F., Savelsbergh, A., Katunin, V. I.,
Rodnina, M. V. & Wintermeyer, W. Conformational
changes of the small ribosomal subunit during
elongation factor G‑dependent tRNA–mRNA
82. Hausner, T. P., Atmadja, J. & Nierhaus, K. H. Evidence
that the G2661 region of 23S rRNA is located at the
ribosomal binding sites of both elongation factors.
Biochimie 69, 911–923 (1987).
83. Clementi, N. & Polacek, N. Ribosome‑associated
GTPases: the role of RNA for GTPase activation.
RNA Biol. 7, 521–527 (2010).
84. Endo, Y. & Wool, I. G. The site of action of α‑sarcin on
eukaryotic ribosomes. The sequence at the α‑sarcin
cleavage site in 28S rRNA. J. Biol. Chem. 257,
9054–9060 (1982).
85. Garcia‑Ortega, L., Alvarez‑Garcia, E., Gavilanes, J. G.,
Martinez‑del‑Pozo, A. & Joseph, S. Cleavage of the
sarcin‑ricin loop of 23S rRNA differentially affects
EF‑G and EF‑Tu binding. Nucleic Acids Res. 38,
4108–4119 (2010).
86. Shi, X., Khade, P. K., Sanbonmatsu, K. Y. & Joseph, S.
Functional role of the sarcin–ricin loop of the 23S
rRNA in the elongation cycle of protein synthesis.
J. Mol. Biol. 419, 125–138 (2012).

87. Clementi, N., Chirkova, A., Puffer, B., Micura, R. &
Polacek, N. Atomic mutagenesis reveals A2660 of 23S
ribosomal RNA as key to EF‑G GTPase activation.
Nature Chem. Biol. 6, 344–351 (2010).
88. Pulk, A. & Cate, J. H. Control of ribosomal subunit
rotation by elongation factor G. Science 340,
1235970 (2013).
89. Berchtold, H. et al. Crystal structure of active
elongation factor Tu reveals major domain
rearrangements. Nature 365, 126–132 (1993).
90. Dibb, N. J. & Wolfe, P. B. lep operon proximal gene is
not required for growth or secretion by Escherichia
coli. J. Bacteriol. 166, 83–87 (1986).
91. Bijlsma, J. J., Lie, A. L. M., Nootenboom, I. C.,
Vandenbroucke‑Grauls, C. M. & Kusters, J. G.
Identification of loci essential for the growth of
Helicobacter pylori under acidic conditions. J. Infect.
Dis. 182, 1566–1569 (2000).
92. Froschauer, E. M., Kolisek, M., Dieterich, F.,
Schweigel, M. & Schweyen, R. J. Fluorescence
measurements of free [Mg2+] by use of mag‑fura 2 in
Salmonella enterica. FEMS Microbiol. Lett. 237,
49–55 (2004).
93. Richey, B. et al. Variability of intracellular ionic
environment of Escherichia coli. J. Biol. Chem. 262,
7157–7164 (1987).
94. Bartetzko, A. & Nierhaus, K. H. A simple Mg2+/NH
4

+/
polyamine system for poly(U) dependent poly(Phe)
synthesis with near in vivo characteristics. Methods
Enzymol. 164, 650–658 (1988).
95. Szaflarski, W. et al. New features of the ribosome and
ribosomal inhibitors: non‑enzymatic recycling,
misreading and back‑translocation. J. Mol. Biol. 380,
193–205 (2008).
96. Shoji, S., Janssen, B. D., Hayes, C. S. & Fredrick, K.
Translation factor LepA contributes to tellurite
resistance in Escherichia coli but plays no apparent
role in the fidelity of protein synthesis. Biochimie 92,
157–163 (2010).
97. Ogle, J. M. & Ramakrishnan, V. Structural insights
into translational fidelity. Annu. Rev. Biochem. 74,
129–177 (2005).
98. Zhang, G., Hubalewska, M. & Ignatova, Z. Transient
ribosomal attenuation coordinates protein synthesis
and co‑translational folding. Nature Struct. Mol. Biol.
16, 274–280 (2009).
99. Kunte, H. Osmoregulation in bacteria: compatible
solute accumulation and osmosensing. Environ. Chem.
3, 94–99 (2006).
100. Bauerschmitt, H., Funes, S. & Herrmann, J. M. The
membrane‑bound GTPase Guf1 promotes mitochondrial
protein synthesis under suboptimal conditions. J. Biol.
Chem. 283, 17139–17146 (2008).
101. Gavrilova, L. P. & Spirin, A. S. “Nonenzymatic”
translation. Methods Enzymol. 30, 452–462

(1974).
102. Bergemann, K. & Nierhaus, K. H. Spontaneous,
elongation factor G independent translocation of
Escherichia coli ribosomes. J. Biol. Chem. 258,
15105–15113 (1983).
103. Gavrilova, L. P. & Spirin, A. S. Stimulation of ‘non‑
enzymic’ translocation in ribosome by
p‑chloromercuribenzoate. FEBS Lett. 17, 324–326
(1971).
104. Shoji, S., Walker, S. E. & Fredrick, K. Reverse
translocation of tRNA in the ribosome. Mol. Cell 24,
931–942 (2006).
105. Konevega, A. L. et al. Spontaneous reverse movement
of mRNA‑bound tRNA through the ribosome. Nature
Struct. Mol. Biol. 14, 318–324 (2007).
106. Stapulionis, R. & Deutscher, M. P. A channeled tRNA
cycle during mammalian protein synthesis. Proc. Natl
Acad. Sci. USA 92, 7158–7161 (1995).
107. Yegian, C. D., Stent, G. S. & Martin, E. M. Intracellular
condition of Escherichia coli transfer RNA. Proc. Natl
Acad. Sci. USA 55, 839–846 (1966).
108. Dittmar, K. A., Sorensen, M. A., Elf, J., Ehrenberg, M.
& Pan, T. Selective charging of tRNA isoacceptors
induced by amino‑acid starvation. EMBO Rep. 6,
151–157 (2005).
109. Fredrick, K. & Noller, H. F. Catalysis of ribosomal
translocation by sparsomycin. Science 300,
1159–1162 (2003).

110. Fischer, N., Konevega, A. L., Wintermeyer, W.,
Rodnina, M. V. & Stark, H. Ribosome dynamics and
tRNA movement by time‑resolved electron
cryomicroscopy. Nature 466, 329–333 (2010).
111. Munro, J. B., Sanbonmatsu, K. Y., Spahn, C. M. &
Blanchard, S. C. Navigating the ribosome’s metastable
energy landscape. Trends Biochem. Sci. 34, 390–400
(2009).
112. Kavran, J. M. & Steitz, T. A. Structure of the base of
the L7/L12 stalk of the Haloarcula marismortui large
ribosomal subunit: analysis of L11 movements. J. Mol.
Biol. 371, 1047–1059 (2007).
113. Bocharov, E. V. et al. From structure and dynamics of
protein L7/L12 to molecular switching in ribosome.
J. Biol. Chem. 279, 17697–17706 (2004).
114. Budkevich, T. et al. Structure and dynamics of the
mammalian ribosomal pretranslocation complex.
Mol. Cell 44, 214–224 (2011).
This paper reports cryo-EM analyses of various
functional states of rabbit liver ribosomes.
115. Brilot, A.F., Korostelev, A.A., Ermolenko, D.N. &
Grigorieff, N. Structure of the ribosome with
elongation factor G trapped in the pretranslocation
state. Proc. Natl Acad. Sci USA http://dx.doi.
org/10.1073/pnas.1311423110 (2013).
116. Ramrath, D.J. et al. Visualization of two transfer RNAs
trapped in transit during elongation factor G‑mediated
translocation. Proc. Natl Acad. Sci. USA http://dx.doi.
org/10.1073/pnas.1320387110 (2013).

Acknowledgements
The authors thank J. Harms (Hamburg) N. Polacek (University
Bern, Switzerland) and T. Sprink (Charité, Berlin) for help and
discussions. H.Y. and C.M.T.S acknowledge the support of the
Deutsche Forschergruppe (DFG), Forschergruppe 1805, and
Y.Q. is grateful for research grants from the Major State Basic
Research of China 973 project (grant 2012CB911000) and
the National Natural Science Foundation of China (grants
31270847 and 31322015).
Competing interests statement
The authors declare no competing interests.
DATABASES
Protein Data Bank: http://www.rcsb.org/pdb/home/home.do
SUPPLEMENTARY INFORMATION
See online article: S1 (figure) | S2 (figure) | S3 (figure) | S4
(figure)
ALL LINKS ARE ACTIVE IN THE ONLINE PDF
R E V I E W S

http://www.nature.com/nrmicro/journal/vaop/ncurrent/full/nrmic
ro3176.html#supplementary-informationAbstract | Ribosomes
translate the codon sequence of an mRNA into the amino acid
sequence of the corresponding protein. One of the most crucial
events is the translocation reaction, which involves movement
of both the mRNA and the attached tRNAs by one coFigure 1 |
Overall architecture of the large and small subunits of the
bacterial ribosome. Both subunits are shown from the interface
side. The large 50S subunit contains the 23S ribosomal RNA
(rRNA) and 5S rRNA (light grey and dark grey, respectively),
aEF‑G and EF4Figure 2 | The functional phases of the ribosome
during translation. The 70S initiation complex contains the
initiator tRNA (formylmethionine tRNA (fMet-tRNA)) at the
ribosomal P‑site, which interacts with the start codon (typically
AUG) of the mRNA via tFigure 3 | Structure, binding sites and
functions of the elongation factors. a | Domain organization of
elongation factor G (EF‑G), EF4 and EF‑Tu. b | EF-G, EF4 and
EF-Tu have a highly similar domain organization and fold into
similar three-dimensional stMechanism of translocationFigure 4
| The three PRE-states of tRNAs on the ribosome during
translocation. a | Intersubunit rotation of the 30S subunit,
viewed from the 30S solvent side with the 50S subunit in a
fixed position. Rotation of the 30S subunit occurs in an
anticlockwise Figure 5 | Ribosomal conformational changes
during translocation.
a | After peptidyl -transfer, the tRNAs are in the classical state
(A/A and P/P), which establishes an equilibrium with the hybrid
states H1 and H2 (H2 not shown) owing to intersubunit rotBox
1 | Spontaneous translocation and back-translocation
in vitroEF4 and back-translocationFigure 6 | Mechanism of GTP
hydrolysis on EF‑G. a | The active GTPase centre of EF‑G in
complex with a translocation intermediate in the presence of the
non-cleavable GTP analogue GDPCP (5ʹ-guanosyl-methylene-
triphosphate). The functional motifs of EF‑G aSummary and
outlook

APPLIED AND ENVIRONMENTAL MICROBIOLOGY,
0099-2240/98/$04.0010
July 1998, p. 2554–2559 Vol. 64, No. 7
Copyright © 1998, American Society for Microbiology. All
Rights Reserved.
Use of Green Fluorescent Protein To Tag and Investigate
Gene Expression in Marine Bacteria
SERINA STRETTON, SOMKIET TECHKARNJANARUK,
ALAN M. MCLENNAN,
AND AMANDA E. GOODMAN*
School of Biological Sciences, The Flinders University of South
Australia, Adelaide 5001, Australia
Received 22 December 1997/Accepted 20 April 1998
Two broad-host-range vectors previously constructed for use in
soil bacteria (A. G. Matthysse, S. Stretton,
C. Dandie, N. C. McClure, and A. E. Goodman, FEMS
Microbiol. Lett. 145:87–94, 1996) were assessed by
epifluorescence microscopy for use in tagging three marine
bacterial species. Expression of gfp could be
visualized in Vibrio sp. strain S141 cells at uniform levels of
intensity from either the lac or the npt-2 promoter,
whereas expression of gfp could be visualized in Psychrobacter
sp. strain SW5H cells at various levels of
intensity only from the npt-2 promoter. Green fluorescent
protein (GFP) fluorescence was not detected in the

third species, Pseudoalteromonas sp. strain S91, when the gfp
gene was expressed from either promoter. A new
mini-Tn10-kan-gfp transposon was constructed to investigate
further the possibilities of fluorescence tagging of
marine bacteria. Insertion of mini-Tn10-kan-gfp generated
random stable mutants at high frequencies with all
three marine species. With this transposon, strongly and weakly
expressed S91 promoters were isolated.
Visualization of GFP by epifluorescence microscopy was
markedly reduced when S91 (mini-Tn10-kan-gfp) cells
were grown in rich medium compared to that when cells were
grown in minimal medium. Mini-Tn10-kan-gfp
was used to create an S91 chitinase-negative, GFP-positive
mutant. Expression of the chi-gfp fusion was
induced in cells exposed to N*-acetylglucosamine or attached to
chitin particles. By laser scanning confocal
microscopy, biofilms consisting of microcolonies of chi-
negative, GFP1 S91 cells were found to be localized
several microns from a natural chitin substratum. Tagging
bacterial strains with GFP enables visualization of,
as well as monitoring of gene expression in, living single cells
in situ and in real time.
The gene encoding green fluorescent protein (GFP) has
recently become an important visual marker of gene expres-
sion in eukaryotic organisms, as it is more sensitive than other
reporter genes, requires no special cofactors for detection (7),
and can be quantitated with a spectrofluorimeter (24). GFP
has not been as widely applied to prokaryotic organisms be-
cause of a lack of constructs useful for diverse groups of bac-
teria, although GFP vectors are available for specialized bac-
terial systems (13, 24, 33, 41, 42). The wild-type gfp gene has
been mutated to improve detection and expression of the flu-
orescent protein in prokaryotes (10, 18, 30), and both the
wild-type and mutated forms have been used to construct less
specialized bacterial GFP vectors.

A broad-host-range plasmid expressing the improved gfp-
(mut2) (10) gene from either a lac or an npt-2 promoter has
been used successfully to tag gram-negative soil bacteria with
GFP (27). Escherichia coli-Pseudomonas spp. shuttle vectors
containing gfp(mut2) expressed from lac and tac promoters
have been constructed (5), and a GFP cloning cassette con-
taining a similarly improved gfp gene is available for creating
transcriptional fusions in prokaryotes (30). A suicide plasmid
containing a promoterless gfp that recombines wholly with the
bacterial chromosome has been constructed to create genomic
gfp fusions in a diverse range of gram-negative bacteria (22).
Transposons provide an alternative method to insert reporter
genes directly into the genomic DNAs of target strains. Several
Tn5-based transposons containing either a promoterless gfp
gene or a gfp gene expressed from a broad-host-range pro-
moter have been generated for use in tagging diverse bacterial
species (6, 9, 27, 40). Tn5 reporter gene systems, however, are
not effective in all gram-negative bacteria (2, 36).
GFP-tagged bacteria have been used in ecological studies to
monitor single cells or cell populations in activated sludge
communities (14) during symbiosis with plant cells (15), during
infection of macrophages (24), during plasmid conjugation on
semisolid surfaces (9), and in survival studies of E. coli in
aquatic environments (26). There have been recent advances
in detecting the presence of specific genes in single cells,
thereby enabling identification of specific cells in mixtures by
in
situ PCR (20, 21), as well as in detecting the expression of
genes in single cells by in situ PCR after an initial reverse
transcription step targeting the mRNA in the cell (8, 39).
Although these methods are useful approaches for many ex-
periments, they all involve killing the cells because of the
fixation step in the procedure and the heating steps in the PCR

regimen.
We are interested in studying regulation of gene expression
(37, 38), surface colonization behavior (11), and plasmid trans-
fer (3) in several diverse marine bacterial species in situ. To
investigate these processes in living microbial communities, it
is necessary to find methods of tagging the different marine
species so that cells may be visualized in situ and in real time.
For this purpose we compared levels of expression of gfp-
(mut2) from two different bacterial promoters on a broad-host-
range plasmid (27) in three gram-negative marine bacterial
species. As Tn5-based transposons do not work with the ma-
rine bacteria we have tested and Tn10-based transposons do
(2), we also constructed a Tn10-based reporter transposon
containing promoterless gfp(mut2) (hereafter referred to as
gfp) to create transcriptional fusions in order to investigate
gene expression in marine bacteria further.
Here we show the differences in levels of expression of gfp in
three vector-transposon constructs in three marine species:
* Corresponding author. Mailing address: School of Biological
Sci-
ences, The Flinders University of South Australia, GPO Box
2100,
Adelaide, SA 5001, Australia. Phone: (61-8) 8201-5134. Fax:
(61-8)
8201-3015. E-mail: [email protected]
2554
Pseudoalteromonas sp. strain S91, Vibrio sp. strain S141, and
Psychrobacter sp. strain SW5H. GFP-tagged S91 cells were
used to investigate initial biofilm formation on a natural bio-
degradable substratum, squid pen, and laser scanning confocal

microscopy (LSCM) was used to visualize the hydrated struc-
ture of the biofilm at the squid pen surface.
MATERIALS AND METHODS
Bacteria, plasmids, and growth conditions. The bacterial strains
and plasmids
used in this study are described in Table 1. The gfp reporter
transposon con-
structed in this study is shown in Fig. 1. The plasmid construct
pLOFKmgfp in E.
coli SM10 has been lodged with the American Type Culture
Collection. E. coli
strains were grown in Luria-Bertani broth (LB) (28) at 37°C.
All other strains
were grown at 30°C in either tryptone soy broth (Oxoid)
containing NaCl (0.26
M), MgCl2 (1 mM), and CaCl2 (0.33 mM) (TS); LB containing
NaCl (0.26 M),
MgCl2 (1 mM), and CaCl2 (0.33 mM); or artificial seawater
minimal medium
(32) supplemented with 20 mM glutamate (MMMglt) for strains
S91 and SW5H
or 20 mM glucose for strain S141. Agar plates contained 15 g of
Bitek agar
(Sigma) liter21 unless otherwise indicated. The following
antibiotics (Sigma) and
concentrations were used: ampicillin (50 mg ml21), kanamycin
(600 mg ml21 for
S91 and 100 mg ml21 for all other strains), and streptomycin
(100 mg ml21).
DNA manipulations. Plasmid extractions, restriction enzyme
digestions, liga-
tions, transformations, and agarose gel electrophoresis were
carried out by stan-

dard methods (35) and according to the manufacturers’
instructions where ap-
propriate. Restriction and other enzymes were obtained from
New England
Biolabs Inc.
PCR. PCR amplification of the 740-bp gfp fragment from
pBCgfp was done as
previously described (27) with the following primers: gfpSfiI-F
(59-CTCCTCGG
CCGCCTAGGCCGATTTCTAGATTTAAGAAGG) and gfpSfiI-
R (59-CTCC
TCGGCCTAGGCGGCCTCATTATTTGTATAGTTCATC). PCR
amplifica-
tion of the 1.3-kb fragment from pLOFKmgfp transformants was
carried out as
described above with gfpSfiI-F and Kmseq-F (59-
TACAATCGATAGATTGTC
GC), a primer designed to amplify sequence upstream from the
kanamycin
resistance (Kmr) gene of pLOFKm.
Conjugations. Mobilization of p519gfp, p519ngfp, and
pLOFKmgfp from E.
coli hosts with E. coli(pNJ5000) as a helper was done by plate
matings as
previously described (2). The numbers of transconjugants,
donors, and recipients
from matings between E. coli SM10(pLOFKmgfp) and S91 were
determined by
plating a dilution series of each cell mix to MMMglt
(kanamycin and strepto-
mycin) and TS (kanamycin and streptomycin) to select
transconjugants, LB
(ampicillin) to select donors, and TS (streptomycin) to select

recipients.
Screening for extracellular chitinase activity. Chitinase-
negative mutants were
screened on MMMglt supplemented with 0.1% yeast extract and
0.1% colloidal
chitin as previously described (38).
Identification of the transposon-interrupted chitinase gene from
S91CGFP.
Part of the transposon-interrupted chitinase gene from S91CGFP
was amplified
by PCR amplification with a primer, PLOFOUT (59-
CACTGATGAATGTTCC
GTTGC-39), designed to extend outward from the 39 end of the
Kmr gene (38)
and a primer, CHIAR1 (59-ACCAATGTTGATGCGACC-39),
designed to ex-
tend inward from the 39 end of a chitinase gene. The 500-bp
PCR product
obtained was used as a template for DNA sequencing with the
PLOFOUT
primer by the Taq–Dye-Terminator method on an automated
DNA sequencer
(Applied Biosystems model 373; DNA Sequence and Synthesis
Facility, West-
mead Hospital, Sydney, Australia).
Detection of GFP fluorescence and microscopy. Bacterial
colonies on solid
media were exposed to blue light in a light box constructed to
contain a 100-W
quartz-halogen lamp with an infrared filter and a 480-nm-band-
pass filter (An-
dover Corp. part no., FS10-50).

An Olympus BX50 microscope, fitted with epifluorescence and
differential
interference contrast (DIC) optics, was used to visualize cells
grown in liquid
with and without colloidal chitin particles. Images were
generated by either DIC
or epifluorescence (excitation, 488 nm; emission, 520 nm)
optics with a 403 oil
objective lens, numerical aperture of 1.0. Images were recorded
with a Panasonic
digital closed-circuit television camera (model WV-BP510/A)
and captured and
prepared with NIH Image (version 1.59) and Adobe Photoshop
(version 3.0.4)
software, respectively, running on a model 7600/120 Power
Macintosh. For
photomicrography, slides were coated with gelatin (3%) to
prevent cell move-
ment.
For experiments involving LSCM, squid pen, which consists of
40% chitin and
60% protein (wt/wt) (16) was collected from a fish market in
Sydney and stored
at 280°C as described previously (38). Biofilms of S91CGFP
cells were grown on
1-cm2 pieces of squid pen suspended in MMMglt at 30°C. After
24 h and then 7
days, small slices were cut aseptically from a piece of squid pen
and placed
without further treatment on a glass slide and covered by a
coverslip, with
MMMglt as the mounting medium.
A Bio-Rad MRC-1000 LSCM system in combination with a
Nikon Diaphot

300 inverted microscope was used to obtain LSCM images of
bacterial micro-
colonies attached to squid pen. The microscope was equipped
with a 403,
1.15-numerical-aperture water immersion lens and a krypton-
argon laser. Exci-
tation at 488/10 nm was used for GFP and chitin. Due to
autofluorescence of
squid pen at an emission of .515 nm, both GFP and the squid
pen surface could
be imaged. At an emission of 522/35 nm, only GFP was
visualized. Images of
microcolonies attached to squid pen were collected as xy and xz
sections and
TABLE 1. Strains and plasmids used in this study
Bacterial strain or plasmid Relevant characteristic(s)a
Reference or source
Bacterial strains
Pseudoaltermonas sp.
Strain S91 Smr 2, 38
Strain S91CGFP S91::mini-Tn10-gfp-kan, Smr Kmr GFP1,
chitinase-negative This study
Vibrio sp. strain S141 Smr 32
Psychrobacter sp. strain SW5H Smr 34
Escherichia coli
DH5a supEDlac (f80lacZDM15) hsdR recA endA gyrA thi relA
35
C600 supE hsdR thi thr leu lacY tonA 35
SM10 thi thr leu tonA lacY supE (lpir) recA::RP4-2-Tc::Mu Km
29

Plasmids
pBCgfp Cmr gfp1 ATCC 87451
27
p519gfp RSF1010 derivative, Kmr mob1, gfp cloned
downstream of the lac promoter ATCC 87452
27
p519ngfp p519gfp with npt-2 promoter in front of gfp ATCC
87453
27
pNJ5000 Tcr, tra1 17
pLOFKm oriR6K mob1 RP4 Apr lacIq mini-Tn10 19
pLOFKmgfp pLOFKm with promoterless gfp cloned upstream
of kan This study
a Smr, streptomycin resistant; Cmr, chloramphenicol resistant;
Tcr, tetracycline resistant.
FIG. 1. Diagrammatic representation of mini-Tn10-gfp-kan.
Tn10 inverted-
repeat ends are shown as filled boxes at either end. Genes and
relevant restric-
tion enzyme sites are indicated; large arrows show the direction
of gene tran-
scription. Primers used in construction are shown above the
boxes, with small
arrows indicating the 59-to-39 direction. The diagram is not to
scale.
VOL. 64, 1998 USE OF GFP TO TAG MARINE BACTERIA
2555

captured as digital computer files, and quantitative examination
was performed
with CoMOS (Bio-Rad) computer image analysis software.
RESULTS AND DISCUSSION
Expression of GFP from a lac or an npt-2 promoter. p519gfp
and p519ngfp are broad-host-range mob1 plasmids, derived
from the broad-host-range RSF1010 derivative pDSK519 (23),
that were constructed to contain gfp expressed from a lac or an
npt-2 promoter, respectively (27). The npt-2 promoter is known
to be more effective than lac in gram-negative bacteria other
than E. coli (4, 25, 27, 31). E. coli DH5a carrying either
p519gfp
or p519ngfp was conjugated separately to each of the three
marine strains with the E. coli(pNJ5000) helper. Transconju-
gants were selected on TS (kanamycin and streptomycin)
plates. Each marine strain carrying either p519gfp or p519ngfp
was grown to exponential phase and assessed by epifluores-
cence microscopy for expression of gfp. More than 99% of S141
cells expressed GFP uniformly at high intensity from either
promoter. Fluorescence was so strong that colonies of S141
carrying either promoter could be easily identified on TS plates
by eye. Although more than 99% of SW5H(p519ngfp) cells
expressed GFP, fluorescence was not uniform. Of cells express-
ing GFP, only 14% (42 of 296) did so at high intensity. Vari-
ation in fluorescence of SW5H cells may have resulted from
plasmid instability in this strain. pDSK519 was maintained
poorly in SW5H cells, whereas the plasmid was well main-
tained in S141 and S91 cells (data not shown). Cells expressing
GFP were not detected in SW5H(p519gfp) cultures, suggesting
that the lac promoter was not functional in this strain. GFP
fluorescence of S91(p519gfp) or S91(p519ngfp) cells during
exponential growth could not be detected. Weak GFP fluores-
cence was detected, however, in S91(p519ngfp) cells after 2

days of growth as colonies on TS plates. It is possible that the
npt-2 promoter functioned poorly in S91 such that a relatively
longer time was required for GFP to accumulate to levels
sufficient for visibility by epifluorescence microscopy but that
the lac promoter was not functional at all. Bloemberg et al.
reported that gfp was expressed poorly from a tac promoter in
Pseudomonas aeruginosa (at a level 10 times lower than in E.
coli) and Pseudomonas fluorescens (at a level 20 times lower
than in E. coli) and was not expressed from a lac promoter in
P. fluorescens (5).
Construction of pLOFKmgfp(mini-Tn10-gfp-kan). As gfp ex-
pressed from either promoter on pDSK519 derivatives was not
useful for all three marine species tested, each species was
tagged with gfp by transposon delivery direct to the chromo-
some. It is known that mini-Tn10 (19) yields stable transcon-
jugants in our marine strains (2) but that various mini-Tn5,
including mini-Tn5-gfp (27), or Tn5 transposons yield no
transconjugants (reference 2 and data not shown). It was nec-
essary therefore, to construct a mini-Tn10 with gfp as the re-
porter gene for use with the marine bacteria. The plasmid
pLOFKm contains the mini-Tn10 transposon which carries the
Kmr gene (19). A promoterless gfp gene was inserted in a
position similar to that of the promoterless lacZ gene in mini-
Tn10-lac-kan carried by plasmid pLBT, which was described
previously (2). A promoterless gfp fragment, including the T7
(gene 10) ribosome binding site, was amplified from pBCgfp
with primers containing an SfiI restriction enzyme site at their
59 ends. The 740-bp product was digested with SfiI and ligated
to pLOFKm, also digested with the same enzyme. The ligation
mixture was transformed into E. coli SM10 competent cells,
and transformants were selected for ampicillin resistance (Apr)
and Kmr. Plasmid DNA was extracted from each transformant
and linearized with SfiI, and fragments were separated by gel
electrophoresis. One transformant that contained a correctly

The ribosome, which constitutes one of the most complex and.docx

The ribosome, which constitutes one of the most complex and.docx

Recommended

Recommended

More Related Content

Similar to The ribosome, which constitutes one of the most complex and.docx

Similar to The ribosome, which constitutes one of the most complex and.docx (20)

More from oreo10

More from oreo10 (20)

Recently uploaded

Recently uploaded (20)