Molecular Cell, Vol. 11, 103–112, January, 2003, Copyright 2003 by Cell Press
The Critical Role of the Universally Conserved
A2602 of 23S Ribosomal RNA in the Release of the
Nascent Peptide during Translation Termination
Norbert Polacek,1 Maria J. Gomez,1 Koichi Ito,2
Liqun Xiong,1 Yoshikazu Nakamura,2
and Alexander Mankin1,*
1
Center for Pharmaceutical Biotechnology-M/C 870
University of Illinois
900 S. Ashland Avenue
Chicago, Illinois 60607
2
Department of Basic Medical Sciences
Institute of Medical Science
University of Tokyo
4-6-1 Shirokanedai
Minato-ku
Tokyo 108-8639
Japan
Summary
The ribosomal peptidyl transferase center is responsible for two fundamental reactions, peptide bond formation and nascent peptide release, during the elongation and termination phases of protein synthesis,
respectively. We used in vitro genetics to investigate
the functional importance of conserved 23S rRNA nucleotides located in the peptidyl transferase active site
for transpeptidation and peptidyl-tRNA hydrolysis.
While mutations at A2451, U2585, and C2063 (E. coli
numbering) did not significantly affect either of the
reactions, substitution of A2602 with C or its deletion
abolished the ribosome ability to promote peptide release but had little effect on transpeptidation. This
indicates that the mechanism of peptide release is
distinct from that of peptide bond formation, with
A2602 playing a critical role in peptide release during
translation termination.
Introduction
The large ribosomal subunit is responsible for two critical chemical reactions of protein synthesis—polymerization of amino acids at the elongation stage of translation and peptidyl-tRNA hydrolysis during translation
termination. These reactions occur in the peptidyl transferase center whose active site is composed entirely of
RNA and thus can be viewed as a ribozyme (Nissen et
al., 2000; Ban et al., 2000).
During translation elongation, the peptidyl transferase
center promotes the formation of new peptide bonds,
thereby linking single amino acids into a polypeptide
chain. Specific interactions between rRNA nucleotides
constituting the peptidyl transferase center and tRNA
molecules places the donor (peptidyl-) and acceptor
(aminoacyl-tRNA) substrates in an orientation suitable
for peptidyl transfer. This allows the ␣-amino group of
the A site bound aminoacyl-tRNA to attack the carbonyl
carbon atom of the ester bond linking the peptidyl residue to the 3⬘ end of the P site bound tRNA, resulting in
*Correspondence: [email protected]
the formation of a new peptide bond and the transfer
of the peptidyl residue to the amino acid of A site bound
tRNA. The entropic contribution of the peptidyl transferase ribozyme apparently provides the main catalytic
factor of peptide bond formation (Nierhaus et al., 1980;
Polacek et al., 2001). Additional acceleration may come
from chemical catalysis, though its molecular mechanism and contribution to the overall rate of the peptidyl
transfer are unclear (Nissen et al., 2000; Barta et al.,
2001; Polacek et al., 2001; Muth et al., 2000, 2001;
Thompson et al., 2001; Bayfield et al., 2001; Katunin et
al., 2002; Parnell et al., 2002).
Much less is known about the molecular events involved in translation termination. Similar antibiotic sensitivity and pH dependence profiles of the peptidyl-tRNA
hydrolysis and transpeptidation reactions suggest that
both reactions are a function of the ribosomal peptidyl
transferase center (Vogel et al., 1969; Caskey et al., 1971;
Tate and Brown, 1992). However, the functioning of the
ribosomal peptidyl transferase during translation elongation and termination are markedly different. In contrast to peptide bond formation, the release of the completed polypeptide from the peptidyl-tRNA is directly
assisted in vivo by proteins called class 1 release factors
(RFs). When a RF binds to the ribosome in response to
the presence of a stop codon in the decoding site of
the small ribosomal subunit, it is thought to reach into
the peptidyl transferase center and, through a poorly
understood mechanism, trigger the hydrolysis of the
peptidyl-tRNA ester bond (Frolova et al., 1999). The exact contribution of the ribosome and the RF to the catalysis of peptide release remains unclear. The class I RFs
contain a universally conserved GGQ sequence, which
is thought to interact with the peptidyl transferase center
(Frolova et al., 1999; Ito et al., 2000), an assumption
generally compatible with the recently solved crystallographic structures of eukaryotic and eubacterial class I
release factors (Song et al., 2000; Vestergaard et al.,
2001; Ehrenberg and Tenson, 2002). Some mutations in
or near the GGQ sequence interfere with peptidyl-tRNA
hydrolysis without impeding factor binding (Frolova et
al., 1999; Song et al., 2000; Zavialov et al., 2002), leading
to the proposal that the glutamine residue of the GGQ
sequence might play a direct role in the peptide release
by positioning the water molecule for attack on the ester
bond of the P site bound peptidyl-tRNA (Frolova et al.,
1999; Song et al., 2000). However, more recent mutational studies challenged this theory (Seit Nebi et al.,
2000, 2001; Zavialov et al., 2002). Other models of termination, which focus more on the ribosome itself, view
the reaction of peptide release as a modified version of
the peptidyl transfer reaction where the peptidyl residue
is transferred to a water molecule instead of the ␣-amino
group of an aminoacyl-tRNA (Tate and Brown, 1992;
Vogel et al., 1969; Caskey et al., 1971; Maden and Monro,
1968). In this scenario, the RF serves an auxiliary function as a modulator of the peptidyl transferase center’s
activity.
In order to understand the relationship between the
two main reactions of the ribosomal peptidyl trans-
Molecular Cell
104
Figure 1. The Mutations of the Universally Conserved Nucleotides in the Ribosomal Peptidyl Transferase Center Investigated in This Study
(A) The secondary structure of the central region of domain V of T. aquaticus 23S rRNA (Gutell et al., 2002) with the mutated positions indicated
by red characters and the mutations shown by arrows.
(B) Positions of the four nucleotides, C2063, A2451, U2585, and A2602, located in the crystallographic structure of the large ribosomal subunit
within 5.5 Å of the putative location of the tetrahedral carbon atom (green sphere) of the peptidyl transfer transition state intermediate (Nissen
et al., 2000).
ferase, transpeptidation and peptide release, and to
identify the nucleotides that might be functionally relevant for the peptidyl-tRNA hydrolysis, we investigated
the effects of mutations at four residues of the peptidyl
transferase center of 23S rRNA, C2063, A2451, U2585,
and A2602 (Figure 1A). The nucleotide residues that
were included in our studies were chosen on the basis
of their proximity to the active site of the peptidyl transferase center (Figure 1B) as well as on phylogenetic and
biochemical evidence for their functional importance.
In the atomic structure of the large ribosomal subunit
complexed with a transition state analog, these are the
only nucleotides whose reactive groups are positioned
within 5 Å of the putative tetrahedral carbon center (Figure 1B, green sphere) at which the reaction of peptide
bond formation takes place (Nissen et al., 2000). All
four nucleotides are universally conserved. A plethora
of biochemical and structural data implicate at least
three of these nucleotides as functionally important for
the activity of the peptidyl transferase center. The accessibility of A2451, U2585, and A2602 to chemical modification is affected by binding of the peptidyl transferase
ligands (Moazed and Noller, 1989; Noller et al., 1990),
and chemical modification of A2451 or U2585 was
shown to weaken the binding of peptidyl-tRNA to the P
site (Bocchetta et al., 1998). Crystallographic studies
show that the orientation of these residues is affected
by the binding of the peptidyl transferase substrates
(Nissen et al., 2000; Schmeing et al., 2002), with A2602
undergoing the most dramatic conformational transition
upon substrate binding (Nissen et al., 2000; A. Yonath
and C. Duarte, personal communications). In the X-ray
structure of the 50S ribosomal subunit functional complexes, the N3 of A2451 is located within 4 Å from the
tetrahedral center of the transition state analog and was
proposed to directly participate in the chemical catalysis
of the peptidyl transfer, acting as a general base-acid
to promote the formation and subsequent resolution of
the tetrahedral intermediate (Nissen et al., 2000; Muth
et al., 2000). Additionally, A2451, U2585, and A2602 are
involved in binding antibiotics that affect peptidyl transferase function (Schlünzen et al., 2001; Porse et al., 1999;
Vester and Garrett, 1988; Lázaro et al., 1996; A.M., unpublished data). Though less is known about the functional significance of C2063, the formation of the C2063A2450 wobble base pair was recently proposed to be
important for the pH-dependent component of catalysis
of peptide bond formation (Katunin et al., 2002).
We used the in vitro genetics approach to investigate
the relationship between the mechanisms of peptidyl
transfer and peptide release and to gain insights into
the possible function role of C2063, A2451, U2585, and
A2602 in these reactions. We found that while peptidyl
transfer tolerates base changes at any of the studied
nucleotides, some mutations at A2602 specifically eliminated the peptide release activity. This suggests a critical role of A2602 in the hydrolysis of peptidyl-tRNA and
hints at a distinct mechanism for the peptide release
reaction as compared to transpeptidation.
Results
Effects of Mutations in the 23S rRNA
on Peptidyl Transfer and RF-Dependent
Peptidyl-tRNA Hydrolysis
In agreement with their invariant nature, central location
in the active site of the peptidyl transferase center, and
putative functional importance, base alterations at at
least three out of the four investigated positions, A2451,
U2585, and A2602 (Figure 1A), are lethal in vivo (Thompson et al., 2001; Green et al., 1997; Muth et al., 2000;
K.R. Lieberman and A.E. Dahlberg, personal communication). Therefore, in order to study the effects of the
mutations, we used an in vitro reconstitution system.
Functionally active T. aquaticus large ribosomal subunits were assembled from in vitro transcribed 23S rRNA
Critical Role of A2602 in Translation Termination
105
(wild-type or mutant), 5S rRNA, and 50S subunit proteins
(Khaitovich et al., 1999), and the reconstituted 50S subunits were reassociated with native 30S subunits to form
70S ribosomes. The reconstituted wild-type 50S subunits were shown previously to be active in several transpeptidation assays as well as in a poly(U)-dependent
cell-free translation system (Khaitovich et al., 1999;
Khaitovich and Mankin, 2000; Polacek et al., 2001). Although the peptidyl transferase activity of the in vitro
reconstituted subunits was usually only 10%–20% of
that of the native large ribosomal subunits when measured in a classic “fragment reaction” assay, it was sufficient to assess the general ability of the in vitro reconstituted wild-type or mutant ribosomes to catalyze the
reaction (Khaitovich et al., 1999). Furthermore, control
experiments showed that characteristic features of the
peptide release activity of reconstituted 50S subunits
also remained unchanged: the ribosome-catalyzed peptidyl-tRNA hydrolysis was stop codon- and RF-dependent and readily inhibited by sparsomycin or hygromycin
A (data not shown). The RF-dependent peptidyl-tRNA
hydrolyzing activity of ribosomes containing in vitro assembled subunits was about one half (52% ⫾ 10%) of
the activity of the native ribosomes.
Previously we have shown that T. aquaticus ribosomes harboring mutations at A2451, a nucleotide that
has been proposed to directly participate in the chemical
catalysis of transpeptidation (Nissen et al., 2000), retain
high levels of peptidyl transferase activity in several in
vitro assays employing puromycin derivatives as an acceptor substrate (Polacek et al., 2001). In the current
study we used a more physiological two-tRNA SPARK
assay (Polacek et al., 2002) in which both donor
(N-biotinyl-Phe-tRNA) and acceptor ([3H]Phe-tRNA) substrates contained full-size tRNAs and were bound to the
ribosome in the presence of poly(U) mRNA. In agreement
with our previous results, all A2451 mutants efficiently
promoted the peptidyl transferase reaction, leading to
formation of the reaction product, N-biotinyl-Phe[3H]Phe-tRNA (Table 1).
For measuring the RF-dependent peptide release activity, radiolabeled formyl-Met-tRNA was bound to the
ribosomes in the presence of the mRNA analog UUC
AUG UAA, the peptidyl-tRNA hydrolysis was initiated
by the addition of Thermus thermophilus RF1 (Ito and
Nakamura, 1997; Ito et al., 2000), and the release of
formyl-methionine was quantified by scintillation counting or by paper electrophoresis. Similar to their activity
in the peptidyl transferase reaction, ribosomes with any
base substitution at A2451 were readily able to hydrolyze peptidyl-tRNA (Figures 2A and 2B). Though the
activity of the mutant ribosomes was somewhat decreased compared to the wild-type, the initial rate of
peptidyl-tRNA hydrolysis differed no more than 6-fold at
25⬚C. The 2451U mutation is known to render ribosomes
resistant to chloramphenicol (Kearsey and Craig, 1981;
Thompson et al., 2001). Indeed, while 250 ␮M chloramphenicol inhibited the peptide release activity of reconstituted wild-type ribosomes by ⵑ50%, the ability of the
2451U, 2451C and 2451G mutants to hydrolyze peptidyltRNA remained unaffected (data not shown). This confirms that the peptide release was indeed the function
of the mutant ribosomes rather than a result of contamination with wild-type ribosomes.
Table 1. Relative Activity of Reconstituted Wild-Type and Mutant
T. aquaticus Ribosomes in the Reactions of Peptide Bond Formation
and RF-Dependent Peptidyl-tRNA Hydrolysis
2602
A
⌬
C
G
U
2585
U
A
C
G
2451
A
C
G
U
2063
C
A
G
U
Peptide Bond
Formationa,b
RF1-Mediated
Hydrolysisa,c
100
91
73
106
103
100
0
0
8⫾1
16 ⫾ 1
⫾
⫾
⫾
⫾
1
2
12
9
100
52 ⫾ 0
35 ⫾ 5
30 ⫾ 2
100
75 ⫾ 11
14 ⫾ 2
21 ⫾ 4
100
50 ⫾ 11
44 ⫾ 10
33 ⫾ 6
100
27 ⫾ 6
52 ⫾ 1
47 ⫾ 1
100
75 ⫾ 14
65 ⫾ 7
82 ⫾ 1
100
71 ⫾ 9
78 ⫾ 10
23 ⫾ 7
a
Activity of reconstituted wild-type ribosomes in each experiment
was taken as 100%. Values shown represent an average of 2–3
independent experiments.
b
Peptidyl transferase activity was measured in a two-tRNA SPARK
assay (Polacek et al., 2002). The incubation time, 25 min, corresponded to the slope of the kinetic curve of the reaction catalyzed
by reconstituted wild-type ribosomes.
c
RF1-dependent hydrolysis assay was performed with radiolabeled
formyl-Met-tRNA as a peptidyl-tRNA analog under the conditions
described in the Experimental Procedures. The incubation time, 2
min, corresponded to the slope of the kinetic curve of the reaction
catalyzed by reconstituted wild-type ribosomes.
The effects of mutations at U2585 and C2063 were
qualitatively similar to those of the A2451 mutations: the
mutant ribosomes exhibited reduced, but still significant, levels of peptidyl transferase and peptide release
activities (Table 1). High transpeptidation activities of
the U2585 mutants were in agreement with the results
obtained previously with the Bacillus stearothermophilus in vitro reconstitution system (Green and Noller,
1999). These results suggest that neither U2585 nor
C2063 are vitally critical for any of the two reactions
catalyzed by the peptidyl transferase center.
A different picture emerged with ribosomes containing mutations at position A2602. The 2602C mutation
as well as the deletion of A2602 (A2602⌬) completely
eliminated RF1-dependent peptidyl-tRNA hydrolysis
(Figure 2C). Even with prolonged incubation, we were
unable to detect any ribosome-catalyzed release of formyl-methionine from formyl-Met-tRNA. The other two
mutants (2602U and 2602G) also showed diminished
peptide release activities, with the initial rates of the
reaction reduced by a factor of 25 and 250, respectively
(Figure 2C).
The severe effect of some of the 2602 mutations on
the peptide release drastically contrasted the virtual lack
of an effect of any of the 2602 mutations on the peptidyl
transferase activity of ribosomes coming from the same
Molecular Cell
106
Figure 2. Peptidyl-tRNA Hydrolyzing Activity of the Ribosomes Carrying Mutations at Positions A2451 and A2602
(A) Phosphorimager exposure of an electrophoretic analysis of formyl-[35S]methionine released from formyl-[35S]Met-tRNA by reconstituted
wild-type or mutant (A2451) 70S ribosomes, or isolated 30S ribosomal subunits at indicated time points (seconds) of the reaction.
(B) The time course of formyl-methionine release catalyzed by ribosomes containing mutations at A2451. The plot points were obtained by
quantifying formyl-[35S]methionine spots on phosphorimager autoradiogram (average of two independent experiments). The background values
(amount of formyl-methionine released in the reaction containing only 30S subunits) were subtracted from all experimental data points.
(C) The time course of formyl-methionine release catalyzed by reconstituted wild-type or mutant ribosomes containing mutations at A2602
or the A2602 deletion (⌬).
(D) Dependence of peptide release catalyzed by wild-type or 2602 mutant ribosomes on RF1 concentration. Incubation time was 2 min.
For the peptide release assay, T. aquaticus 50S subunits reconstituted with wild-type or mutant 23S rRNA were reassociated with native 30S
subunits and programmed with UUCAUGUAA mRNA; hydrolysis of the P site bound formyl-[35S]Met-tRNA (A, B, and C) or formyl-[3H]MettRNA (D) was initiated by addition of RF1 (0.4 ␮M). The reactions were stopped by adjusting pH to 1 (HCl); free formyl-methionine was extracted
into ethyl acetate phase and resolved by paper electrophoresis (A and B) or directly quantified by scintillation counting (C and D).
reconstituted pool. The mutant ribosomes containing
any of the three nucleotide substitutions at A2602 or the
deletion of this nucleotide (2602⌬) exhibited essentially
wild-type levels of transpeptidation activity in a twotRNA SPARK assay (Table 1) at least within the limitations of the in vitro reconstitution system (see Discussion). Previous data implicated A2602 in binding of sparsomycin (Porse et al., 1999), a drug known to strongly
inhibit peptidyl transfer (Monro et al., 1969). Indeed,
while the peptidyl transferase activity of wild-type native
ribosomes (see Figure 4) or in vitro assembled ribosomes was severely inhibited by sparsomycin, mutant
ribosomes carrying the 2602C, 2602U, or 2602⌬ mutations were resistant to the drug in the SPARK-Pmn reaction (data not shown). Therefore, we are confident that
the high levels of the peptidyl transferase activity in our
experiments were indeed attributable to the 2602 mutant
ribosomes rather than wild-type contamination.
A2602 Mutations Do Not Significantly
Perturb Ligand Binding
The dramatically different effect of base changes at
A2602 on peptide bond formation and peptide release
suggests a critical and distinct role of A2602 in hydrolysis of peptidyl-tRNA during translation termination. One
can envision at least two different scenarios of A2602
involvement: A2602 might be critical for RF binding and/
or proper positioning of the RF in the peptidyl transferase active site; alternatively, A2602 could be directly
involved in the catalysis of peptidyl-tRNA hydrolysis.
In order to discriminate between the two models of
A2602 involvement in peptide release, we investigated
the rate of peptidyl-tRNA hydrolysis as a function of RF
concentration. In the standard assay with reconstituted
ribosomes, the RF concentration (0.4 ␮M) exceeds 20fold the concentration required to saturate the reaction
catalyzed by the native T. aquaticus ribosomes. Further
(up to additional 10-fold) increase of the RF concentration did not compensate for the decreased peptide release activity of the mutant ribosomes carrying the
A2602 mutations (Figure 2D). Thus, the compromised
activity of the 2602 mutants does not appear to be due
to diminished RF binding.
To further verify whether the A2602 mutations directly
affect the ability of the ribosome to hydrolyze peptidyltRNA as opposed to ligand binding, the activity of the
Critical Role of A2602 in Translation Termination
107
Table 2. Relative Activity of Reconstituted Wild-Type and 2602
Mutant T. aquaticus Ribosomes in the Reactions of Peptide Bond
Formation and RF-Independent Peptidyl-tRNA Hydrolysis
in the Presence of 30% Acetone
2602
A
⌬
C
G
U
Peptide Bond
Formationa,b
RF-Independent
Hydrolysisa,c
100
78
52
79
62
100
0
0
0
4
a
Activity of reconstituted wild-type ribosomes in each experiment
was taken as 100%.
b
Peptidyl transferase activity was measured in a two-tRNA SPARK
assay (Polacek et al., 2002) in the presence of 30% acetone, 25 min
incubation.
c
RF1-independent hydrolysis was performed with radiolabeled formyl-Met-tRNA as a peptidyl-tRNA analog under the conditions described in the Experimental Procedures. The incubation time, 1 hr,
corresponded to the reaction plateau.
2602 mutant ribosomes was tested in an RF-independent peptide release reaction. The assumption was that
if mutations affected RF binding, then the mutant ribosomes should be active in the factor-independent assay,
while mutations affecting the catalytic mechanism
should severely compromise the activity irrespective of
whether RF-dependent or RF-independent assay is
used. The RF-independent peptidyl-tRNA hydrolyzing
activity of the ribosome can be activated in the presence
of deacylated tRNA and 30% acetone (Caskey et al.,
1971). While reconstituted wild-type ribosomes showed
considerable activity in the RF-independent peptide release reaction (77% of native ribosomes), the 2602C as
well as the 2602 deletion mutant remained completely
inactive similar to what was observed in the RF1-dependent reaction (Table 2). In addition, the 2602G mutant,
which in the RF1-mediated assay showed some activity,
was inactive in the RF-independent reaction. Only the
2602U mutant exhibited low residual activity (4%). This
observation agreed well with the previous conclusion
that the inability of the 2602 mutants to hydrolyze peptidyl-tRNA was not due to reduced affinity of the RF to
the mutant ribosome in the standard assay. In contrast
to the 2602 mutants, ribosomes with mutations at A2451,
U2585, or C2063 were active in the RF-independent
assay; the relative activities were generally comparable
to those measured in the RF-mediated peptide release
(data not shown).
Still, the possibility remained that base alterations at
A2602 interfered with binding of the reaction substrate,
peptidyl-tRNA, and/or binding of deacylated tRNA in the
factor-independent hydrolysis reaction. To address this
question, it was useful to compare the substrates of
the RF-independent peptide release, where several of
A2602 mutants were completely inactive, with the twotRNA SPARK transpeptidation reaction, where the same
mutants showed high activities (Tables 1 and 2). In the
two-tRNA SPARK, N-protected aminoacyl-tRNA binds
to the ribosomal P site and acceptor aminoacyl-tRNA
binds in the A site. Under the conditions of the RFindependent peptide release assay, peptidyl-tRNA is
bound to the P site while deacylated tRNA can poten-
Figure 3. Binding of Deacylated tRNA to the A Site Stimulates RFIndependent Hydrolysis of Formyl-Met-tRNA
RF-independent release of formyl-methionine from formyl-[35S]MettRNA bound to the ribosomal P site in response to the presence of
deacylated tRNAPhe (squares) or tRNATyr (triangles) cognate (filled
symbols) or noncognate (open symbols) to the A site codon. 0.2
pmol of formyl-[35S]Met-tRNA were bound to 7 pmol of native T.
aquaticus ribosomes programmed with mRNAs placing an AUG
codon in the P site and either Gly (GGU), Tyr (UAC), or Phe (UUC)
codons in the A site. RF-independent hydrolysis was initiated by
the addition of 30% acetone and 10 pmol tRNAPhe or tRNATyr; the
release of formyl-methionine was monitored as described in Figures
2C and 2D. Legend is as follows: filled square, Phe (UUC) codon in
the A site/tRNAPhe; filled triangle, Tyr (UAC) codon in the A site/
tRNATyr; open square, Gly (GGU) codon in the A site/tRNAPhe; open
triangle, Gly (GGU) codon in the A site/tRNATyr.
tially interact with either the E or A site of the ribosome
(Rheinberger et al., 1981; Kirillov et al., 1983; Moazed
and Noller, 1989). In order to clarify occupation of which
site (A or E) by deacylated tRNA stimulates the peptide
release, native T. aquaticus 70S ribosomes containing
P site bound formyl-[35S]Met-tRNA were programmed
with synthetic mRNAs that placed different codons in
the ribosomal A site, and limiting amounts of cognate
or noncognate tRNA were used to stimulate formyl-MettRNA hydrolysis (Figure 3). The results clearly demonstrated that the presence of tRNA in the A site is required
to stimulate the peptide release—only deacylated tRNA
cognate to the A site codon was able to stimulate the
reaction rate. Thus, the ligands of the RF-independent
peptide release assay are similar to those of the twotRNA SPARK transpeptidation reaction, with the only
difference that aminoacyl-tRNA is required for transpeptidation, while deacylated tRNA is needed for the
peptide release. Since the 2602 mutant ribosomes from
the same reconstitution pool showed high activities in
the peptidyl transferase assays (in the absence or in the
presence of acetone; Tables 1 and 2), but were inactive
in RF-independent peptide release, the effect of the
mutations on the peptidyl-tRNA hydrolysis could hardly
be explained by altered tRNA binding to either the P or
the A site.
Thus, the complete loss (with 2602C or 2602⌬ mutants) or the dramatic reduction (with 2602G and 2602U
mutants) of the peptide release activity observed in both
the RF-mediated and the RF-independent assays does
not appear to be related to the reduced ligand binding
and is more consistent with the direct involvement of
A2602 in peptide release. Furthermore, the divergent
effects of A2602 mutations on peptidyl transfer and peptide release indicate that these two reactions are pro-
Molecular Cell
108
Figure 4. Antibiotic Sensitivity of the Reactions of Transpeptidation, RF-Dependent
Formyl-Met-tRNA Hydrolysis, and RF-Independent Formyl-Met-tRNA Hydrolysis
SPARK-Pmn assay (Polacek et al., 2002) was
used to monitor peptidyl transferase activity
(light gray bars). RF-dependent (black bars)
and RF-independent (dark gray bars) assays
are described in the Experimental Procedures. The activity of the corresponding reactions without antibiotics was taken as 100%.
Antibiotics were grouped according to their
inhibitory action on peptidyl transfer and peptide release. Group I, strong inhibition of all
the reactions; group II, strong inhibition of
peptidyl transfer and moderate-to-weak inhibition of RF-dependent or RF-independent
peptide release; group III, no strong inhibition of any of the reactions. All drugs were tested at 100 ␮M final concentration except for
chloramphenicol, which was tested at 250 ␮M. Antibiotics used were HygA, hygromycin A; Spa, sparsomycin; Cam, chloramphenicol; Crb,
carbomycin; Cln, clindamycin; Lnc, lincomycin; Pri II, pristinamycin II; Pri I, Pristinamycin I; Qnp, quinupristin; Ery, erythromycin; Tel, telithromycin; RU, RU69874; Evn, evernimicin; Lin, linezolid.
moted by distinct elements of the ribosomal peptidyl
transferase center.
Differential Effects of Antibiotics on Peptide Bond
Formation and Peptide Release
The latter conclusion was corroborated by idiosyncratic
inhibition pattern of these two reactions by some antibiotics. Following early experiments with E. coli ribosomes
(Vogel et al., 1969), we examined the sensitivity of peptide bond formation and nascent peptide release catalyzed by native T. aquaticus ribosomes toward an extended collection of 50S subunit-targeted antibiotics. In
order to facilitate comparison with the previous data, a
puromycin-based SPARK-Pmn assay was used to monitor peptide bond formation (Polacek et al., 2002) while
a standard formyl-Met release assay was used to monitor the peptide release activity (see above). Finally, in
order to eliminate possible effects of the drugs on RF
binding, we tested the antibiotic effects in the RF-independent peptide release assay (Caskey et al., 1971). In
all three assays the P site was occupied by formyl-MettRNA; the reactions were performed in the presence of
excess amounts (100 ␮M) of antibiotics (250 ␮M for
chloramphenicol) and stopped at a time point corresponding to the slope on the time curve of the corresponding reactions.
In a general agreement with the data obtained previously with E. coli ribosomes (Caskey et al., 1971), some
antibiotics, in particular hygromycin A and sparsomycin,
strongly inhibited the activity of T. aquaticus ribosomes
in all three assays (⬍3% of the control) at the tested
concentrations (Figure 4, group I). Several other drugs
(group III), including macrolides (erythromycin, telithromycin, RU 69874), evernimicin, and linezolid, did not
inhibit any of the studied reactions. Notably, however,
a number of antibiotics (group II) strongly interfered with
the peptidyl transfer (less then 4% of the remaining
activity) but were notably less active in inhibiting peptide
release, especially in the RF-independent assay (more
than 50% of the remaining activity). The difference was
especially dramatic with lincomycin and its derivative
clindamycin, which completely failed to inhibit the factor-independent peptidyl-tRNA hydrolysis. In addition,
streptogramin B-type drugs (pristinamycin I and quinupristin) that stimulated transpeptidation had no effect
on the peptidyl-tRNA hydrolysis. Though the disparity
in action of some antibiotics could be potentially due
to the different nature of the A site substrates in the
peptidyl transfer and peptide release assays, the results
were generally consistent with the idea that distinctive
features of the peptidyl transferase center may be responsible for these two activities.
Discussion
We studied the functional effects of mutations at four
universally conserved nucleotides located in close proximity to the peptidyl transferase active site and found
that only mutations at A2602 of 23S rRNA abolish the
hydrolysis of peptidyl-tRNA. Neither of the studied mutations abolish peptide bond formation. These results
suggest a critical and distinct role for A2602 in peptide
release during termination of protein synthesis.
The in vitro reconstitution system used in this study
provides a valuable advantage of obtaining ribosomes
carrying mutations at functionally important rRNA positions that are known to be lethal in vivo. It is important
to point out, however, the intrinsic limitations of this
system. Only small quantities of mutant 23S rRNA can
be prepared by in vitro transcription, which puts limits
on the amount of assembled ribosomes available for
biochemical and kinetic studies. Additionally, due to
relatively inefficient in vitro assembly, the activities of
in vitro reconstituted ribosomal subunits never reach
those of native ribosomes (⬍20% and ⵑ50% for peptide
bond formation and peptide release, respectively). Furthermore, the rates of the reactions catalyzed by in vitro
assembled ribosomes varied sometimes between different reconstituted pools (up to 5-fold). Therefore, a special effort was made to minimize the effects of such
variations. All mutant 23S rRNAs used in any experimental series were always transcribed in parallel reactions,
and wild-type and mutant ribosomes were assembled
simultaneously. Most importantly, in every experiment,
ribosomes from the same assembly pool were always
tested in parallel in peptidyl transferase and peptide
Critical Role of A2602 in Translation Termination
109
Figure 5. The
Putative
Conformational
Switch at A2602 as a Trigger for Changing the
Mode of Activity of the Ribosomal Peptidyl
Transferase Center
Orientation of A2602 during translation elongation allows for proper positioning of peptidyl- and aminoacyl-tRNAs in the peptidyl
transferase center that makes peptidyl transfer and a new peptide bond formation possible. Binding of the class 1 release factor (RF1
in the figure) in response to the presence of
a stop codon in the decoding site reorients
A2602. This places it in a position where its
reactive groups can potentially activate a water molecule, facilitating its nucleophilic attack on the carbonyl carbon atom of the peptidyl-tRNA ester bond, thereby accelerating
the rate of peptidyl-tRNA hydrolysis.
release assays, so that the lack of activity in only one
of the two assays could not be attributed to assembly
failures.
The peptidyl transferase assay showed that in spite
of the central position of the residues C2063, A2451,
U2585, and A2602 in the peptidyl transferase active site
(Figure 1B), mutations at these residues did not dramatically affect the overall in vitro rates of peptide bond
formation (Table 1). Although the assays used in this
work do not directly allow addressing the catalytic rate
of peptidyl transfer, the mere fact that the mutant ribosomes are readily able to catalyze transpeptidation
shows that the nucleotide alterations do not impede the
main acceleration factor provided by the ribosome for
this reaction. Similarly, mutations at C2063, A2451, and
U2585 did not dramatically affect the overall in vitro
rates of peptidyl-tRNA hydrolysis. In striking contrast,
some base changes at A2602 had a profound negative
effect on the peptide release: the 2602C mutation and
the deletion of A2602 abolished the ribosome-mediated
peptidyl-tRNA hydrolysis in an RF-dependent or RFindependent reactions while ribosomes harboring the
2602G mutation exhibited notably reduced activities
(Tables 1 and 2).
It seems unlikely that mutations at 2602 interfere with
binding of the reaction substrates to the ribosome. Increasing the RF concentration could not rescue the activity of A2602 mutants (Figure 2D), indicating that at
least with the mutants whose residual activity could
be measured (2602U and 2602G), RF binding was not
limiting. Additionally, the 2602G, 2602C, and the 2602
deletion mutants were inactive in an RF-independent
peptide release assay where hydrolysis of peptidyltRNA in the P site is stimulated by the binding of deacylated-tRNA in the A site (Table 2 and Figure 3). At the
same time, these mutants possessed high levels of
transpeptidation activity in an assay where the reaction
substrates, peptidyl-tRNA and aminoacyl-tRNA, closely
resemble those employed in the RF-independent peptide release reaction (Tables 1 and 2). This suggests
that neither P nor A site tRNA binding was significantly
perturbed in the 2602 mutant ribosomes. Such conclusion is in agreement with the previous observation that
A2602 methylation by dimethyl sulfate does not interfere
with tRNA binding in the ribosomal P site (Bocchetta et
al., 1998). Since the A2602 mutations do not interfere
with ligand binding, it seems reasonable to conclude
that these mutations directly affect the ability of the
ribosome to catalyze the peptide release reaction.
The remarkable difference in the effects of A2602 mutations on peptidyl transfer and peptide release may
reflect the idiosyncratic nature of the chemical transformation that takes place during these two reactions. The
attack of the aminoacyl-tRNA ␣-aminogroup, a strong
nucleophile, on the ester bond during peptide bond formation can occur with reasonably high rates upon
proper substrate orientation (Nierhaus et al., 1980).
Hence, the entropic factor—correct positioning of the
peptidyl- and aminoacyl-tRNA substrates—may be the
main acceleration factor of peptide bond formation.
Since substrate binding requires multiple interactions
between the ribosome and tRNAs (Yusupov et al., 2001),
alteration of single nucleotides in the peptidyl transferase active site may not produce a dramatic effect on
the overall rate of the reaction. In contrast, hydrolysis
of the ester bond during peptide release, driven by a
less nucleophilic water oxygen, should depend more
strongly on chemical (possibly, general base) catalysis
involving activation of the water molecule (Jencks,
1969). In the latter case, alteration of a catalytic nucleotide will affect the reaction rate much more severely,
resulting in a dramatic drop in product formation—
similar to what we have observed in our experiments
with 2602 mutant ribosomes. Thus, it is possible that
the two critical reactions mediated by the peptidyl transferase center during protein synthesis, transpeptidation
and peptide release, may be promoted through different
catalytic mechanisms potentially involving distinct elements of the peptidyl transferase center. In line with this
interpretation are the differential effects of antibiotics
on these two reactions (Figure 4). The most significant
differences were evident with lincosamides and streptogramin A-type antibiotics. Apparently, these antibiotics
do not affect proper coordination of the water molecule
required for the peptide release but interfere with a precise positioning of the A site ligand (the amino group
of aminoacyl-tRNA)—a conclusion compatible with the
location of lincosamides binding site on the ribosome
(Schlünzen et al., 2001).
In the cell, the reaction of the nascent peptide release
strictly depends on the action of class 1 RFs (Ito et al.,
1998; Ryden and Isaksson, 1984). However, the ability
Molecular Cell
110
of the ribosome to hydrolyze peptidyl-tRNA in vitro in
the absence of RF (Caskey et al., 1971; Zavialov et al.,
2002) and our mutational data point to the ribosome,
and more specifically to 23S rRNA, as the main catalyst
of this reaction. What is then the role of the class 1 RFs
in peptide release? We favor the view that the release
factor plays a regulatory or trigger function in peptidyltRNA hydrolysis while the chemistry of the reaction is
primarily driven by the ribosome itself. The structural
flexibility (Nissen et al., 2000; A. Yonath and C. Duarte,
personal communications) and the central location of
A2602 in the peptidyl transferase active site is compatible with its possible role in coordinating a water molecule for the attack onto the carbonyl carbon atom of
the ester bond linking the nascent peptide to tRNA.
A2602 should assume such an orientation required for
the nascent peptide release only at the termination step
of translation, not in the elongation phase when this
would lead to a premature nascent peptide release. The
switch in the mode of peptidyl transferase function—
from peptide bond formation to peptidyl-tRNA hydrolysis—occurs only in response to RF binding and may
involve repositioning of A2602 into an orientation suitable for peptide release (Figure 5). The universally conserved GGQ motif of the class 1 RFs may directly contribute to placing A2602 in the catalytically active
conformation. The possibility of interaction of the GGQ
motif with A2602 is supported by crystallographic structures of eukaryotic and bacterial RFs as well as by biochemical data showing that the GGQ segment of the RF
can reach into the ribosomal peptidyl transferase center
where it will be positioned close to A2602 (Song et al.,
2000; Vestergaard et al., 2001; Zavialov et al., 2002). The
proximity of the GGQ loop of RF to A2602 is further
revealed by hydroxyl radical mapping: A2602 was the
only nucleotide in the ribosomal peptidyl transferase
center that was directly hit by a probe attached near
the GGQ motif of E. coli RF1 (Wilson et al., 2000). Since
A2602 in the large ribosomal subunit RNA and the GGQ
motif in the class 1 release factors are universally conserved, our model is equally applicable to translation
termination in all evolutionary domains. In vitro, in the
RF-independent peptide release reaction, the presence
of deacylated tRNA in the ribosomal A site and acetone
may induce a similar orientation of A2602, thus making
the peptidyl-tRNA hydrolysis possible even in the absence of the protein factor. Indeed, RNA probing
showed that the accessibility of A2602 to chemical modification is affected by both tRNA and acetone (our unpublished results).
Although coordination of a water molecule by the base
or sugar of A2602 seems plausible, other scenarios,
such as coordination of a metal (Mg2⫹) ion, which can
potentially participate in the catalysis of peptidyl-tRNA
hydrolysis, cannot be excluded. Additionally, we cannot
discard the possibility that mutations at A2602 may have
an indirect allosteric effect on the other nucleotides in
the peptidyl transferase center that may be critically
involved in peptide release.
Our results highlighting the potential involvement of
a 23S rRNA nucleotide in peptide release are consistent
with the hypothesis that peptidyl-tRNA hydrolysis is an
intrinsic function of the ribosome itself rather than that
of a protein factor. Therefore, it appears that termination
is yet another reaction of protein synthesis which possibly is inherent to the ribosomal RNA and may be a relic
of the RNA world. The assisting protein factors probably
evolved to fine-tune and enhance the ribosome performance in the tightly regulated process of protein synthesis in the modern cells.
Experimental Procedures
tRNAs
Formyl-[3H]Met-tRNA (14,300 dpm/pmol) was prepared by NEN Life
Science products (Boston, MA). Formyl-[35S]Met-tRNA, biotin-PhetRNA, and [3H]Phe-tRNA were prepared and HPLC-purified as described (Marcker, 1965; Polacek et al., 2002).
Reconstitution of 50S Subunits
Mutations at positions C2063, A2451, A2602, and U2585 were engineered in the cloned Thermus aquaticus 23S rRNA gene (Khaitovich
et al., 1999) by PCR mutagenesis as described (Polacek et al., 2001,
2002). Wild-type and mutant 23S rRNA were in vitro transcribed and
together with 5S rRNA and total 50S subunit proteins assembled into
50S subunits (in the presence of 0.5 mM of the macrolide antibiotic
RU69874) (Khaitovich and Mankin, 1999). The in vitro reconstituted
50S subunits were reassociated with native T. aquaticus 30S subunits as previously described (Polacek et al., 2001).
Peptide Bond Formation
The transpeptidation activity of 70S ribosomes containing in vitro
reconstituted 50S subunits was measured using the SPARK technology employing two complete tRNA substrates (Polacek et al., 2002).
14 pmol of ribosomes were incubated with 50 ␮g poly(U) and 14
pmol N-biotin-Phe-tRNA for 15 min at 37⬚C in a buffer containing
11.7 mM Tris/HCl (pH 7.5), 8.3 mM HEPES/KOH (pH 7.5), 11.7 mM
MgCl2, 233 mM NH4Cl, 3.5 mM spermidine, 0.05 mM spermine, 4
mM ␤-mercaptoethanol, and 0.12 mM EDTA. The reaction was initiated by the addition of 15 pmol [3H]Phe-tRNA (6000 cpm/pmol) and
performed at 37⬚C for 25 min in a final volume of 30 ␮l. The reaction
was stopped by the addition of SPA beads in solution containing 125
mM EDTA and the product analyzed using the scintillation proximity
technique as described (Polacek et al., 2002). Under these conditions, 11% ⫾ 4% of the tRNA substrates were converted into dipeptidyl-tRNA product.
The transpeptidation activity of native T. aquaticus ribosomes (19
pmol) was assessed by using the SPARK-Pmn reaction employing
54 pmol f[3H]Met-tRNA and 37 pmol biotin-puromycin (BiotPmn) as
reaction substrates (Polacek et al., 2002). In antibiotic sensitivity
experiments, the drugs were added after P-site tRNA binding to a
final concentration of 100 ␮M (250 ␮M for chloramphenicol) and
incubated 5 min at 25⬚C prior to initiating the reaction with the
addition of BiotPmn. The assay was performed at 37⬚C for 25 min in
a final volume of 30 ␮l.
RF1-Mediated Peptidyl-tRNA Hydrolysis
13.5 pmol ribosomes containing in vitro assembled 50S subunits
were programmed with 0.8 nmol synthetic mRNA (UUCAUGUAA)
and incubated with 0.25–0.5 pmol formyl-[35S]Met-tRNA (46,000–
110,000 cpm/pmol) or formyl-[3H]Met-tRNA (30,000 cpm/pmol) for
15 min at 30⬚C in 25 ␮l buffer containing 15 mM Tris/HCl (pH 7.5),
10 mM HEPES/KOH (pH 7.5), 15 mM MgCl2, 280 mM NH4Cl, 4.2 mM
spermidine, 0.05 mM spermine, 3.5 mM ␤-mercaptoethanol, and
0.14 mM EDTA. The peptidyl-tRNA hydrolysis reaction was initiated
by the addition of 20 pmol Thermus thermophilus RF1 (Ito and
Nakamura, 1997; Ito et al., 2000) and performed at 25⬚C in a final
volume of 50 ␮l. The final buffer was 7.5 mM Tris/Cl (pH 7.5), 12.5
mM HEPES/KOH (pH 7.5), 7.5 mM MgCl2, 140 mM NH4Cl, 2.1 mM
spermidine, 0.05 mM spermine, 1.8 mM ␤-mercaptoethanol, and
0.07 mM EDTA. The reaction was stopped by the addition of 5
volumes 0.1 M HC; the product, radiolabeled formyl-methionine,
was extracted into 1 ml of ethyl acetate (extraction efficiency was
36% ⫾ 4%) and either analyzed directly by liquid scintillation counting or resolved by paper electrophoresis in 5% acetic acid, 0.5%
Critical Role of A2602 in Translation Termination
111
pyridine (pH 3.5) (Monro and Marcker, 1967) and then quantified
using a Molecular Dynamics phosphorimager.
In reactions with native ribosomes, T. aquaticus ribosomes (7
pmol) were programmed with 250 pmol UUCAUGUAA mRNA and
incubated with 0.2 pmol formyl-[35S]Met-tRNA for 10 min at 30⬚C in
5 ␮l buffer containing 20 mM HEPES/KOH (pH 7.6), 6 mM MgAc2,
150 mM NH4Cl, 2 mM spermidine, 0.05 mM spermine, and 4 mM
␤-mercaptoethanol. Subsequently the volume was increased to 50
␮l with the same buffer and the release reaction was initiated by
addition of 20 pmol RF1. The hydrolysis reaction was terminated
after 2 min at 25⬚C as described above and analyzed by scintillation
counting. In antibiotic studies, 100 ␮M (250 ␮M for chloramphenicol)
of the drug was added and incubated for 4 min at 25⬚C before
RF1 addition. In order to avoid ethanolysis of the peptidyl-tRNA,
antibiotics that were originally dissolved in ethanol were first dried
down in a SpeedVac concentrator and then resuspended directly
in the reaction mixture.
RF-Independent Peptidyl-tRNA Hydrolysis
14 pmol reassociated ribosomes containing in vitro reconstituted
50S subunits were incubated with 70 pmol synthetic mRNA AAGGA
GAUAUAACAAUGGGU (Dharmacon Research) and 0.63 pmol formyl-[35S]Met-tRNA or formyl-[3H]Met-tRNA for 15 min at 25⬚C in 50
␮l buffer containing 7 mM Tris/HCl, 13 mM HEPES/KOH (final pH
ⵑ8.0), 30 mM MgCl2, 150 mM NH4Cl, 2.1 mM spermidine, 0.05 mM
spermine, 4 mM ␤-mercaptoethanol, and 0.07 mM EDTA. Subsequently, 130 pmol total E. coli tRNA was added and the reaction
was initiated by the addition of 30% acetone. The hydrolysis reaction
proceeded at 25⬚C and was stopped as described for the RF1mediated release. Under these conditions 61% ⫾ 11% of the starting
amount of peptidyl-tRNA was hydrolyzed by reconstituted ribosomes.
Native T. aquaticus ribosomes (7 pmol) were incubated with 0.2
pmol formyl-[35S]Met-tRNA for 10 min at 25⬚C in 25 ␮l hydrolysisbuffer containing 20 mM HEPES/KOH (pH 8.0), 30 mM MgAc2, 150
mM NH4Cl, 2 mM spermidine, 0.05 mM spermine, and 4 mM
␤-mercaptoethanol. In inhibition experiments, 100 ␮M of the drug
(250 ␮M for chloramphenicol) was added and incubated for 5 min
at 25⬚C prior to initiating the reaction by the addition of 65 pmol
total E. coli tRNA and 30% acetone. The reaction was performed
at 25⬚C and stopped as described above.
In experiments aimed at verifying the site of interaction of the
deacylated tRNA with the formyl-Met-tRNA/70S complex, the MgAc2
concentration was reduced to 3.5 mM and 10 pmol specific tRNAPhe,
or tRNATyr (Sigma) was used. Under these conditions, the reaction
became strictly mRNA dependent. However, the reaction yields
dropped and only ⵑ25% of the peptidyl-tRNA was hydrolyzed. 38
pmol of the synthetic mRNA analogs AAGGAGAUAUAACAAUG
GGU, AGGAGGUCACACAUGUACUAA (Dharmacon Research), or
of the in vitro transcribed MFV-mRNA GGGAAAAGAAAAGAAAA
GAAAAUGUUCGUUAAAAGAAAAGAAAAGAAAU (Polacek et al.,
2000) were used to program native T. aquaticus 70S (the A site
codons of the three used mRNAs, coding for Gly, Tyr, or Phe, respectively, are shown in bold and the AUG codon is underlined).
Antibiotic Sensitivity Testing
SPARK-Pmn assay (Polacek et al., 2002) was used to monitor antibiotic sensitivity of peptidyl transferase activity, and RF-dependent
and RF-independent formyl-Met-tRNA hydrolysis assays described
above were used to monitor peptide release activity. All experiments
were performed with native T. aquaticus ribosomes. The antibiotics
were obtained from the following sources: hygromycin A from Pfizer;
sparsomycin from National Institutes of Health; chloramphenicol,
lincomycin, clindamycin, and erythromycin from SIGMA; carbomycin was a gift of B. Weisblum; pristinamycin I and II, quinupristin,
telithromycin, and RU 69874 from Aventis Pharma; evernimicin from
Schering-Plough; and linezolid from Pharmacia.
Acknowledgments
We thank S. Swaney and D. Shinabarger for providing formyl[3H]Met-tRNA and SPA beads, K. Nierhaus for MFV-mRNA clone,
B. Weisblum for carbomycin, S.R. Connell for critical reading of the
manuscript and useful suggestions, and M. Ehrenberg, C. Duarte,
and A. Yonath for sharing their results prior to publication. This work
was supported by a Grant from the National Institutes of Health (GM
59028) to A.M. and grants from the Ministry of Education, Sports,
Culture, Science, and Technology of Japan (MEXT) and the Human
Frontier Science Program to Y.N. N.P. was supported by an Erwin
Schrödinger fellowship (J2127) from the Austrian Science Foundation (FWF).
Received: July 26, 2002
Revised: October 17, 2002
References
Ban, N., Nissen, P., Hansen, J., Moore, P.B., and Steitz, T.A. (2000).
The complete atomic structure of the large ribosomal subunit at
2.4 Å resolution. Science 289, 905–920.
Barta, A., Dorner, S., Polacek, N., Berg, J.M., Lorsch, J.R., Nissen,
P., Hansen, J., Muth, G.W., Ban, N., Moore, P.B., et al. (2001). Mechanism of ribosomal peptide bond formation. Science 291, 203.
Bayfield, M.A., Dahlberg, A.E., Schulmeister, U., Dorner, S., and
Barta, A. (2001). A conformational change in the ribosomal peptidyl
transferase center upon active/inactive transition. Proc. Natl. Acad.
Sci. USA 98, 10096–10101.
Bocchetta, M., Xiong, L., and Mankin, A.S. (1998). 23S rRNA positions essential for tRNA binding in ribosomal functional sites. Proc.
Natl. Acad. Sci. USA 95, 3525–3530.
Caskey, C.T., Beaudet, A.L., Scolnick, E.M., and Rosman, M. (1971).
Hydrolysis of fMet-tRNA by peptidyl transferase. Proc. Natl. Acad.
Sci. USA 68, 3163–3167.
Ehrenberg, M., and Tenson, T. (2002). A new beginning of the end
of translation. Nat. Struct. Biol. 9, 85–87.
Frolova, L.Y., Tsivkovskii, R.Y., Sivolobova, G.F., Oparina, N.Y., Serpinsky, O.I., Blinov, V.M., Tatkov, S.I., and Kisselev, L.L. (1999).
Mutations in the highly conserved GGQ motif of class 1 polypeptide
release factors abolish ability of human eRF1 to trigger peptidyltRNA hydrolysis. RNA 5, 1014–1020.
Green, R., and Noller, H.F. (1999). Reconstitution of functional 50S
ribosomes from in vitro transcripts of Bacillus stearothermophilus
23S rRNA. Biochemistry 38, 1772–1779.
Green, R., Samaha, R.R., and Noller, H.F. (1997). Mutations at nucleotides G2251 and U2585 of 23 S rRNA perturb the peptidyl transferase center of the ribosome. J. Mol. Biol. 266, 40–50.
Gutell, R.R., Lee, J.C., and Cannone, J.J. (2002). The accuracy of
ribosomal RNA comparative structure models. Curr. Opin. Struct.
Biol. 12, 301–310.
Ito, K., and Nakamura, Y. (1997). Cloning and overexpression of
polypeptide release factor 1 of Thermus thermophilus. Biochimie
79, 287–292.
Ito, K., Uno, M., and Nakamura, Y. (1998). Single amino acid substitution in prokaryote polypeptide release factor 2 permits it to terminate
translation at all three stop codons. Proc. Natl. Acad. Sci. USA 95,
8165–8169.
Ito, K., Uno, M., and Nakamura, Y. (2000). A tripeptide ’anticodon’
deciphers stop codons in messenger RNA. Nature 403, 680–684.
Jencks, W.P. (1969). Catalysis in Chemistry and Enzymology (New
York: McGraw-Hill, Inc.).
Katunin, V.I., Muth, G.W., Strobel, S.A., Wintermeyer, W., and Rodnina, M.V. (2002). Important contribution to catalysis of peptide bond
formation by a single ionizing group within the ribosome. Mol. Cell
10, 339–346.
Kearsey, S.E., and Craig, I.W. (1981). Altered ribosomal RNA genes
in mitochondria from mammalian cells with chloramphenicol resistance. Nature 290, 607–608.
Khaitovich, P., Tenson, T., Kloss, P., and Mankin, A.S. (1999). Reconstitution of functionally active Thermus aquaticus large ribosomal
subunits with in vitro-transcribed rRNA. Biochemistry 38, 1780–
1788.
Khaitovich, P., and Mankin, A.S. (2000). Reconstitution of the 50S
Molecular Cell
112
subunit with in vitro transcribed 23S rRNA: a new tool for studying
peptidyl transferase. In The Ribosome. Structure, Function, Antibiotics and Cellular Interactions, R.A. Garrett, S.R. Douthwaite, A. Liljas,
A.T. Matheson, P.B. Moore, and H.F. Noller, eds. (Washington, DC:
ASM Press), pp. 229–243.
Khaitovich, P., and Mankin, A.S. (1999). Effect of antibiotics on large
ribosomal subunit assembly reveals possible function of 5S rRNA.
J. Mol. Biol. 291, 1025–1034.
Kirillov, S.V., Makarov, E.M., and Semenkov, Yu.P. (1983). Quantitative studies of interaction of deacylated tRNA with Escherichia coli
ribosomes: role of 50S subunits in formation of the E-site. FEBS
Lett. 157, 91–94.
Lázaro, E., Rodriguez-Fonseca, C., Porse, B., Ureña, D., Garrett,
R.A., and Ballesta, J.P.G. (1996). A sparsomycin-resistant mutant
of Halobacterium salinarium lacks a modification at nucleotide
U2603 in the peptidyl transferase centre of 23 S rRNA. J. Mol. Biol.
261, 231–238.
Maden, B.E., and Monro, R.E. (1968). Ribosome-catalyzed peptidyl
transfer. Effects of cations and pH value. Eur. J. Biochem. 6,
309–316.
tant of Escherichia coli that shows enhanced misreading of UAG/A
and increased efficiency for some tRNA nonsense suppressors. Mol.
Gen. Genet. 193, 38–45.
Schlünzen, F., Zarivach, R., Harms, J., Bashan, A., Tocilj, A., Albrecht, R., Yonath, A., and Franceschi, F. (2001). Structural basis
for the interaction of antibiotics with the peptidyl transferase centre
in eubacteria. Nature 413, 814–821.
Schmeing, T.M., Seila, A.C., Hansen, J.L., Freeborn, B., Soukup, J.K.,
Scaringe, S.A., Strobel, S.A., Moore, P.B., and Steitz, T.A. (2002). A
pre-translocational intermediate in protein synthesis observed in
crystals of enzymatically active 50S subunits. Nat. Struct. Biol. 9,
225–230.
Seit Nebi, A., Frolova, L., Ivanova, N., Poltaraus, A., and Kisselev,
L. (2000). Mutation of a glutamine residue in the universal tripeptide
GGQ in human eRF1 termination factor does not cause complete
loss of its activity. Mol. Biol. 34, 899–900.
Seit-Nebi, A., Frolova, L., Justesen, J., and Kisselev, L. (2001).
Class-1 translation termination factors: invariant GGQ minidomain
is essential for release activity and ribosome binding but not for
stop codon recognition. Nucleic Acids Res. 29, 3982–3987.
Moazed, D., and Noller, H.F. (1989). Interaction of tRNA with 23S
rRNA in the ribosomal A, P, and E sites. Cell 57, 585–597.
Song, H.W., Mugnier, P., Das, A.K., Webb, H.M., Evans, D.R., Tuite,
M.F., Hemmings, B.A., and Barford, D. (2000). The crystal structure
of human eukaryotic release factor eRF1—mechanism of stop codon recognition and peptidyl-tRNA hydrolysis. Cell 100, 311–321.
Monro, R.E., and Marcker, K.A. (1967). Ribosome-catalysed reaction
of puromycin with a formylmethionine-containing oligonucleotide.
J. Mol. Biol. 25, 347–350.
Tate, W.P., and Brown, C.M. (1992). Translational termination: “stop”
for protein synthesis or “pause” for regulation of gene expression.
Biochemistry 31, 2443–2450.
Monro, R.E., Celma, M.L., and Vazquez, D. (1969). Action of sparsomycin on ribosome-catalysed peptidyl transfer. Nature 222,
356–358.
Thompson, J., Kim, D.F., O’Connor, M., Lieberman, K.R., Bayfield,
M.A., Gregory, S.T., Green, R., Noller, H.F., and Dahlberg, A.E. (2001).
Analysis of mutations at residues A2451 and G2447 of 23S rRNA in
the peptidyl transferase active site of the 50S ribosomal subunit.
Proc. Natl. Acad. Sci. USA 98, 9002–9007.
Marcker, K.A. (1965). The formation of N-formyl-methionyl-sRNA. J.
Mol. Biol. 14, 63–70.
Muth, G.W., Ortoleva-Donnelly, L., and Strobel, S.A. (2000). A single
adenosine with a neutral pKa in the ribosomal peptidyl transferase
center. Science 289, 947–950.
Muth, G.W., Chen, L., Kosek, A.B., and Strobel, S.A. (2001). pHdependent conformational flexibility within the ribosomal peptidyl
transferase center. RNA 7, 1403–1415.
Nierhaus, K.H., Schulze, H., and Cooperman, B.S. (1980). Molecular
mechanisms of the ribosomal peptidyl transferase center. Biochem.
Int. 1, 185–192.
Nissen, P., Hansen, J., Ban, N., Moore, P.B., and Steitz, T.A. (2000).
The structural basis of ribosome activity in peptide bond synthesis.
Science 289, 920–930.
Noller, H.F., Moazed, D., Stern, S., Powers, T., Allen, P.N., Robertson,
J.M., Weiser, B., and Triman, K. (1990). Structure of rRNA and its
functional interactions in translation. In The Ribosome: Structure,
Function, and Evolution. W.E. Hill, A. Dahlberg, R.A. Garrett, P.B.
Moore, D. Schlessinger, and J.R. Warner, eds. (Washington, DC.:
ASM Press), pp. 73–92.
Parnell, K.M., Seila, A.C., and Strobel, S.A. (2002). Evidence against
stabilization of the transition state oxyanion by a pKa-perturbed
RNA base in the peptidyl transferase center. Proc. Natl. Acad. Sci.
USA 99, 11658–11663.
Polacek, N., Patzke, S., Nierhaus, K.H., and Barta, A. (2000). Periodic
conformational changes in rRNA: monitoring the dynamics of
translating ribosomes. Mol. Cell 6, 159–171.
Polacek, N., Gaynor, M., Yassin, A., and Mankin, A.S. (2001). Ribosomal peptidyl transferase can withstand mutations at the putative
catalytic nucleotide. Nature 411, 498–501.
Polacek, N., Swaney, S., Shinabarger, D., and Mankin, A.S. (2002).
SPARK—a novel method to monitor ribosomal peptidyl transferase
activity. Biochemistry 41, 11602–11610.
Porse, B.T., Kirillov, S.V., Awayez, M.J., Ottenheijm, H.C.J., and Garrett, R.A. (1999). Direct crosslinking of the antitumor antibiotic sparsomycin, and its derivatives, to A2602 in the 23S rRNA of Escherichia
coli ribosome. Proc. Natl. Acad. Sci. USA 96, 9003–9008.
Rheinberger, H.-J., Sternbach, H., and Nierhaus, K.H. (1981). Three
tRNA binding sites on Escherichia coli ribosomes. Proc. Natl. Acad.
Sci. USA 78, 5310–5314.
Ryden, S.M., and Isaksson, L.A. (1984). A temperature-sensitive mu-
Vester, B., and Garrett, R.A. (1988). The importance of highly conserved nucleotides in the binding region of chloramphenicol at the
peptidyl transfer center of Escherichia coli 23S ribosomal RNA.
EMBO J. 7, 3577–3587.
Vestergaard, B., Van, L.B., Andersen, G.R., Nyborg, J., Buckingham,
R.H., and Kjeldgaard, M. (2001). Bacterial polypeptide release factor
RF2 is structurally distinct from eukaryotic eRF1. Mol. Cell 8, 1375–
1382.
Vogel, Z., Zamir, A., and Elson, D. (1969). The possible involvement of
peptidyl transferase in the termination step of protein biosynthesis.
Biochemistry. 8, 5161–5168.
Wilson, K.S., Ito, K., Noller, H.F., and Nakamura, Y. (2000). Functional
sites of interaction between release factor RF1 and the ribosome.
Nat. Struct. Biol. 7, 866–870.
Yusupov, M.M., Yusupova, G.Z., Baucom, A., Lieberman, K., Earnest, T.N., Cate, J.H., and Noller, H.F. (2001). Crystal structure of
the ribosome at 5.5 Å resolution. Science 292, 883–896.
Zavialov, A.V., Mora, L., Buckingham, R.H., and Ehrenberg, M.
(2002). Release of peptide promoted by the GGQ-motif of class 1
release factors regulates the GTPase activity of RF3. Mol. Cell 10,
789–798.