Content uploaded by Jaanus Remme
Author content
All content in this area was uploaded by Jaanus Remme on Jun 20, 2014
Content may be subject to copyright.
Subribosomal particle analysis reveals the stages
of bacterial ribosome assembly at which
rRNA nucleotides are modified
TRIINU SIIBAK
1,2
and JAANUS REMME
1
1
Institute of Molecular and Cell Biology, University of Tartu, Tartu 51010, Estonia
2
Insitute of Technology, University of Tartu, Tartu 51010, Estonia
ABSTRACT
Modified nucleosides of ribosomal RNA are synthesized during ribosome assembly. In bacteria, each modification is made by
a specialized enzyme. In vitro studies have shown that some enzymes need the presence of ribosomal proteins while other
enzymes can modify only protein-free rRNA. We have analyzed the addition of modified nucleosides to rRNA during ribosome
assembly. Accumulation of incompletely assembled ribosomal particles (25S, 35S, and 45S) was induced by chloramphenicol or
erythromycin in an exponentially growing Escherichia coli culture. Incompletely assembled ribosomal particles were isolated
from drug-treated and free 30S and 50S subunits and mature 70S ribosomes from untreated cells. Nucleosides of 16S and 23S
rRNA were prepared and analyzed by reverse-phase, high-performance liquid chromatography (HPLC). Pseudouridines were
identified by the chemical modification/primer extension method. Based on the results, the rRNA modifications were divided
into three major groups: early, intermediate, and late assembly specific modifications. Seven out of 11 modified nucleosides of
16S rRNA were late assembly specific. In contrast, 16 out of 25 modified nucleosides of 23S rRNA were made during early steps
of ribosome assembly. Free subunits of exponentially growing bacteria contain undermodified rRNA, indicating that a specific
set of modifications is synthesized during very late steps of ribosome subunit assembly.
Keywords: modified nucleosides; ribosome assembly; chloramphenicol; erythromycin; rRNA
INTRODUCTION
Ribosome assembly involves several coordinated reactions.
Folding, nucloelytic processing, and post-transcriptional mod-
ification of rRNA occur concomitantly with the r-protein
association with the rRNA. Modification of rRNA is thereby
an integral part of ribosome assembly process. Escherichia coli
16S rRNA contains 11 and 23S rRNA contains 25 modified
nucleosides (Ofengand and Del Campo 2004). Eleven pseu-
douridines and 21 base methylations make up the bulk of
modified nucleosides in E. coli rRNA. Each rRNA modification
is made by a specific enzyme. In E. coli there are 32 rRNA
modification enzymes in total, 25 of them are methyltrans-
ferases and seven pseudouridine synthases. All the pseudour-
idine synthases and most of the rRNA methyltransferases have
been identified (Ofengand and Del Campo 2004; Purta et al.
2009). The substrate specificity of the rRNA modification
enzymes has been studied mostly by cell-free experiments using
purified enzymes. The specificity of specific rRNA modifica-
tion has been shown to depend on the presence of r-proteins
(Ofengand and Del Campo 2004). Substrate specificities of
rRNA modification enzymes with respect to ribosome assem-
bly are summarized in Table 1.
Ribosome assembly in bacteria is fast and efficient. In wild-
type bacteria ribosome subunits are formed within 2–3 min
at 37°C (Lindahl 1975). Chloramphenicol is known to in-
hibit assembly of both ribosome subunits (Dagley and Sykes
1959; Kurland et al. 1962). More recently many other anti-
biotics were shown to cause ribosome assembly defects (for
review, see Champney 2006). Chloramphenicol and eryth-
romycin were shown to inhibit assembly of both ribosome
subunits due to the unbalanced synthesis of ribosomal com-
ponents (Dodd et al. 1991; Siibak et al. 2009). In the presence
of these antibiotics incompletely assembled ribosome sub-
units accumulate in the cells (Siibak et al. 2009). Here
we report the results of quantitative analysis of modified
nucleosides of 16S and 23S rRNA in the antibiotic-induced
Reprint requests to: Jaanus Remme, Institute of Molecular and Cell
Biology, University of Tartu, Riia 23, Tartu 51010, Estonia; e-mail:
jremme@ebc.ee; fax: +372-7420286.
Article published online ahead of print. Article and publication date are at
http://www.rnajournal.org/cgi/doi/10.1261/rna.2160010.
RNA (2010), 16:2023–2032. Published by Cold Spring Harbor Laboratory Press. Copyright Ó2010 RNA Society. 2023
ribosomal particles stalled at different assembly stages us-
ing reverse-phase, high-performance liquid chromatography
(HPLC). Pseudouridines at specific positions of 23S rRNA
were identified by the chemical modification/primer exten-
sion method. The results allowed dividing the rRNA mod-
ification enzymes with respect to ribosome assembly in vivo
into three classes: early, intermediate, and late assembly spe-
cific enzymes.
RESULTS
Isolation of ‘‘chloramphenicol and erythromycin
particles’’
In order to analyze the temporal relationship between rRNA
modification and ribosome subunit assembly, rRNA from
incompletely assembled ribosomal subunits was analyzed
with respect to the modified nucleoside composition. Accu-
mulation of assembly defective ribosomal subunits was in-
duced by chloramphenicol or erythromycin in an exponen-
tially growing E. coli culture.
Ribosomal particles were separated by sucrose gradient
centrifugation (Fig. 1). Addition of chloramphenicol or
erythromycin to the growth medium leads to the appearance
of three unusual ribosomal particles termed the 25S, 35S, and
45S particles. 35S and 45S particles are related to the 50S
subunit and 25S particles are related to the 30S subunit (Fig.
1), in agreement with the earlier observations (Usary and
Champney 2001; Siibak et al. 2009). In the absence of drugs
the ribosome small subunit has one precursor (21S) and the
large subunit has two precursors (34S and 43S) (Nierhaus
1991). It is possible that the 25S, 35S, and 45S assembly
defective particles formed upon addition of antibiotics are
related to the precursor particles found in vivo. The precursor
rRNA from the ‘‘chloramphenicol particles’’ is converted into
normal 30S and 50S ribosomes without prior degradation
upon removal of the drug (Nomura and Hosokawa 1965;
Adesnik and Levinthal 1969). Therefore, the subribosomal
particles formed in the presence of chloramphenicol are
ribosome subunit assembly intermediate particles.
No free 30S or 50S subunits were found in the presence
of chloramphenicol or erythromycin (Fig. 1), suggesting that
the limiting step of ribosome subunit assembly in the
presence of protein synthesis inhibitors is ribosomal protein
production. Sucrose gradient fractions containing particles
of interest were combined, concentrated by ultrafiltration,
and repurified by a second sucrose gradient. It is worth
mentioning that sedimenting the subribosomal particles
leads to hardly soluble pellet. Therefore, ultrafiltration is
the preferable method for concentrating the subribosomal
particles. The second sucrose gradient centrifugation yielded
homogenous particles. Isolation of chloramphenicol in-
duced particles is shown in Figure 1D–F. rRNA was obtained
by phenol extraction. The 25S and 45S particles contained
only 16S and 23S rRNA, respectively (data not shown). The
35S particles contained a mixture of 16S and 23S rRNAs. The
TABLE 1. Modified nucleosides in Escherichia coli rRNAs and the specificities of the corresponding enzymes
Modification Enzyme
Stage
of assembly Comment Reference
16S rRNA
C516 RsuA Intermediate Some proteins Wrzesinski et al. (1995)
m
7
G527 RsmG Late 30S Okamoto et al. (2007)
m
2
G966 RsmD Late Requires S7 and S19 Weitzmann et al. (1991)
m
5
C967 RsmB Late Blocked by S7 and S19 Weitzmann et al. (1991)
m
2
G1207 RsmC Late 30S Tscherne et al. (1999)
m
4
C1402 RsmH Late 30S Kimura and Suzuki (2010)
Cm1402 RsmI Late 30S Kimura and Suzuki (2010)
m
5
C1407 RsmF Late 30S Andersen and Douthwaite (2006)
m
3
U1498 RsmE Late 30S Basturea and Deutscher (2007)
m
62
A1518 RsmA Late 30S Poldermans et al. (1979)
m
62
A1519 RsmA Late 30S Poldermans et al. (1979)
23S rRNA
m
1
G745 RlmA Early 23S rRNA Hansen et al. (2001)
m
6
A1618 RlmF Intermediate 3.5M LiCl particle Sergiev et al. (2008)
m
2
G1835 RlmG Early 23S rRNA Sergiev et al. (2006)
C1911 RluD Late 50S Leppik et al. (2007)
c1915 RluD Late 50S Leppik et al. (2007)
C1917 RluD Late 50S Leppik et al. (2007)
m
3
c1915 RlmH Late 70S Ero et al. (2008)
m
5
C1962 RlmI Early 23S rRNA Purta et al. (2008)
m
2
G2445 RlmL Early 23S rRNA Lesnyak et al. (2006)
Cm2498 RlmM Early 23S rRNA Purta et al. (2009)
Um2552 RlmE Late 50S, 70S Caldas et al. (2000); Bu
¨gl et al. (2000)
Siibak and Remme
2024 RNA, Vol. 16, No. 10
two rRNA species were separated from each other by sucrose
gradient centrifugation (Fig. 2B).
Free 30S and 50S subunits were isolated from exponen-
tially growing E. coli cells (at 25°C) by two consecutive
sucrose gradient centrifugation methods as described above
for subribosomal particles. It must be noted that the free
ribosome subunits in these conditions are mostly assembly
intermediate particles (z80%) with low functional activity
(Peil et al. 2008). The limiting step of large ribosome subunit
assembly is the final maturation at the level of the 50S
particles (Lindahl 1975; Peil et al. 2008). Therefore, it was
interesting to analyze the modified nucleoside composition
of rRNA of the free 30S and 50S subunits. Ribosomal RNA
was deproteinized by phenol extraction. Mature 16S and
23S rRNA species were isolated from 70S ribosomes by
phenol extraction followed by sucrose gradient centrifuga-
tion (Fig. 2A).
Modification of 16S rRNA during 30S subunit
assembly
Analytical reverse-phase HPLC (RP-HPLC) allows identify-
ing all nucleotides of E. coli 16S rRNA and nearly all
nucleosides of E. coli 23S rRNA (Gehrke and Kuo 1989).
Moreover, this method enables quantitative estimation of
nucleosides. Therefore we have used RP-HPLC for determi-
nation of modified nucleosides in rRNA species. Nucleosides
were prepared by treatment of the rRNA with nuclease P1
and bacterial alkaline phosphatase. Chromatographic peak
assignments to the specific modified
nucleosides were derived from relative
retention times according to Gehrke and
Kuo (1989). The A
260
/A
280
ratio was used
to confirm the peak identities. The RP-
HPLC peak surface area corresponding
to each nucleoside was calculated and
compared with the respective nucleoside
peak area of mature 16S rRNA of 70S
ribosomes (Fig. 3; Table 2). Mature 16S
rRNA isolated from 70S ribosomes has
been shown to contain stochiometric
amounts of modified nucleosides within
10% error limit (Gehrke and Kuo 1989).
At least three independent particle prep-
arations were analyzed with respect to
modified nucleoside content.
The 25S particles formed in the pres-
ence of Cam or Ery contain only 16S
rRNA that is incompletely processed
(Siibak et al. 2009). 16S rRNA of both
Cam and Ery 25S particles clearly contain
several modified nucleosides, albeit at a
lower level compared with the mature
30S particles (Fig. 3; Table 2). Both Cam
and Ery 25S particles have similar nucle-
oside composition (Fig. 3; Table 2). Pseudouridine, m
5
C, and
m
4
Cm are present in 25S particles between 40% and 60%
compared with the mature 16S rRNA. m
7
G and m
2
G are
found in 25S particles in <25%. m
5
U and m
62
A are found in
25S particles only in trace amounts (Table 2). The results
were well reproducible, except in m
4
Cm, which exhibited
significant variation in different preparations. Thus, 16S
rRNA is modified only at low levels during early events of 30S
subunit assembly.
In the bacteria grown at 25°C, 70%–80% of the free 30S
particles contain the 16S rRNA precursor with 115 extra
nucleotides at the 59end (Siibak et al. 2009), indicating that
the majority of these particles are incompletely assembled
precursors of the ribosome small subunit. The modification
level of 16S rRNA is still incomplete compared with the
mature rRNA. Free 30S subunits contain nearly 90% of
pseudouridine and 80% of m
7
G (Table 1). Thus the majority
of these modifications are introduced into 16S rRNA during
intermediate stages of small subunit assembly. Formation of
m
4
Cm seems also to occur during intermediate assembly
stages, although only 70% was found in the free 30S fraction
and a large variability of results was observed (Table 2). m
5
U
and m
62
A are present in the free 30S particles at z20% level
(Table 2), which is about the same fraction as the mature 59end
of 16S was found. Therefore, m
5
Uandm
62
A can be classified as
late assembly specific modifications. m
2
G (positions G966,
G1207, and G1516) and m
5
C (C967 and C1407) are present at
multiple positions of E. coli 16S and therefore cannot be clearly
assigned to a specific ribosome assembly stage.
FIGURE 1. Isolation of ribosomal particles from E. coli grown in the presence of chloram-
phenicol or erythromycin or in the absence of a drug. Exponentially growing bacterial cells
without the drug (A), or treated with chloramphenicol (B), or erythromycin (C)werelysed
and centrifuged in 10%–25% sucrose. 45S, 35S, and 25S fractions of chloramphenicol and
erythromycin treated cells were combined (only chloramphenicol particles are indicated by gray
zones in B). Ribosomal particles were purified by a second sucrose gradient centrifugation (D–F).
Indicated fractions were collected.
rRNA modification during ribosome assembly
www.rnajournal.org 2025
Modification of 23S rRNA
23S rRNA was isolated from the Ery- and Cam-induced 35S
and 45S assembly intermediate particles, from the free 50S
subunits, and from the mature 70S ribosomes of untreated
cells (Figs. 1, 2). It must be noted that the
free 50S subunits are in the majority
(80%) assembly intermediate particles
(Peil et al. 2008; Al Refaii and Alix
2009), The RP-HPLC method used al-
lows identifying all 23S rRNA modifica-
tions except three. m
3
C(present at
position 1915 of 23S rRNA) and m
5
C
(1962) have identical retention times
(Kowalak et al. 1996). However, taking
into account that the formation of m
3
C
occurs on the level of 70S ribosomes (Ero
et al. 2008), the RP-HPLC peak at 11.4
min of 35S and 45S particles probably
contain only m
5
C. The third nucleoside,
which was not analyzed, is an incomplete
modification at C2501 whose retention
time is not known.
The 35S particles formed in the pres-
ence of Ery or Cam exhibited similar
nucleoside composition. The particles
contain several modified nucleosides in nearly stochiometric
amounts. m
1
G, m
6
A, m
2
G, m
7
G, and m
2
A are already present
in the first assembly intermediate particles by 80% or more
(Table 3). This suggests that these modifications are formed
during an early step of large subunit assembly. Pseudour-
idines, m
5
U, m
5
C, Gm, and Cm are present in the 35S
particles by 40%–70% (Table 3). Um was not found in the
23S rRNA isolated from 35S particles (Fig. 4). Thus, m
1
G,
m
6
A, m
2
G, m
7
G, m
5
C, and m
2
A are formed during the early
steps of 50S subunit assembly.
A significant increase of the m
5
U, Gm, Cm, and Um levels
was observed in the 45S particles compared with the 35S
particles (Table 3). However, the Um level is still only 35% of
mature 23S rRNA (Table 3). We conclude that synthesis
of m
5
U, Gm, and Cm occurs during the intermediate steps
of large ribosome subunit assembly.
Free 50S particles isolated from exponentially growing
cultures represent z80% large subunit assembly intermedi-
ate particles according to the processing status of the 23S
rRNA 59end and the translational activity (Peil et al. 2008).
Most of the modified nucleosides are present in the free 50S
particles by 90%–100% in comparison with the mature 23S
rRNA (Table 3). However, Um is present in the free 50S
particles by 63% of the mature 23S rRNA level (Table 3). This
result is in agreement with the free 50S particles being
incompletely assembled and shows that Um together with
m
3
C(see below) are added during the late steps of ribosome
large subunit assembly.
Pseudouridines in 23S rRNA
Pseudouridylation is the most abundant modification in
stable RNAs. There are 10 pseudouridines in E. coli 23S
rRNA, one of them is methylated (m
3
C1915). Cresidues
FIGURE 3. HPLC analysis of 16S rRNA. 16S rRNA was prepared from mature 70S ribosomes,
free 30S subunits, and Cam 25S particles. Nucleoside composition was determined by
RP-HPLC on a Supelcosil LC-18-S. Peaks corresponding to three standard nucleosides
(U, G, and A) and m
5
C, m
7
G, m
3
U, m
4
Cm, m
2
G, and m
22
A are indicated. X corresponds to
an unknown compound.
FIGURE 2. Purification of ribosomal RNA by sucrose gradient
centrifugation. Ribosomal RNA from 70S ribosomes (A) and 35S
particles (B) was deproteinized by phenol extraction and centrifuged in
5%–20% (w/w) sucrose gradient (buffer 20 mM Na-acetate, 100 mM
NaCl, 1 mM EDTA) at 25,700 rpm in an SW28 rotor (Beckman) for
16 h. 16S and 23S rRNA were collectedasindicatedbythegrayzones.
Siibak and Remme
2026 RNA, Vol. 16, No. 10
were quantitated by RP HPLC. According to the HPLC
analysis, 35S particles contain 65%–70%, 45S particles
contain 80%, and free 50S subunits contain 90% of pseu-
douridines compared with the 23S rRNA of 70S ribosomes
(Table 2). It must be taken into account that the methylated
pseudouridine (m
3
C) co-elutes with m
5
C and therefore does
not contribute to the pseudouridine peak. Thus, the mature
23S rRNA contains nine pseudouridines. The 23S rRNA of
precursor particles lacking methylation at C1915 can contain
10 pseudouridine residues.
The chromatographic analysis does not answer the question
of which Cresidues are underrepresented in the subribosomal
particles. To answer this question, chemical modification of
23S rRNA followed by reverse transcriptase-directed primer
extension was used. The presence of pseudouridine at a par-
ticular position is indicated by the primer extension stop on
the CMCT-treated RNA (+ lane) and its absence on the
control RNA (lane). We used five primers to analyze all nine
pseudouridines of 23S rRNA of 35S and 45S particles, free 50S
subunits, and 70S ribosomes for comparison.
All three uridines in helix 69 of E. coli 23S rRNA are
isomerized to pseudouridines (C1911,
m
3
C1915, and C1917) by RluD (Huang
et al. 1998; Raychaudhuri et al. 1998).
Methylation of C1915 causes a primer
extension stop independent of CMCT
treatment (Fig. 5). It must be noted that
m
3
Ccan form a Watson–Crick-like base
pair in the syn conformation of glycosidic
bond allowing a low level of readthrough
by reverse transcriptase. Therefore, it was
not possible to identify pseudouridyla-
tion at position 1915 of 50S and 70S but
the upstream C1911 was still detectable.
In the 35S particles of both drugs, the
C-specific signals at 1911 and 1917 are
not detectable. In the 45S particles all
three pseudourindines appear to be pres-
ent at low levels. The fraction of C1911 and C1917 is further
increased in the free 50S particles but is clearly lower
compared with the pseudouridylation level of 70S ribosomes
(Fig. 5). The results indicate that RluD is specific to the late
assembly of the large ribosome subunit. This conclusion
is supported by the earlier findings that RluD-specific C
residues are made on the level of 50S particles in the RNA
helicase DeaD deletion strain (Leppik et al. 2007) and that
the RluD modifies 50S subunits and 70S ribosomes more
specifically and efficiently than protein-free 23S rRNA
(Vaidyanathan et al. 2007).
A CMCT-independent reverse transcriptase stop site at
position 1915 of 23S rRNA indicates the presence of m
3
C.
23S rRNA of 35S and 45 particles exhibit a very low CMCT-
independent stop signal. In the 50S the stop is strong and in
the mature 23S rRNA of 70S ribosomes the stop is very strong
(Fig. 5). This shows that the enzyme RlmH responsible for
methylation of C1915 at C3 is specific to the very late steps
of large ribosome subunit assembly. This conclusion is sup-
ported by the fact that RlmH requires the presence of the 30S
subunit (Ero et al. 2008).
All other pseudouridines were present in the 35S and 45S
subribosomal particles of both antibiotics. In the free 50S
subunits the level of pseudouridines was similar to that of 70S
ribosomes (data not shown). Taking into account the HPLC
results described above, this suggests that the Cresidues
outside of helix–loop 69 are made during the early steps of
large ribosome subunit assembly. Thus, the corresponding
pseudouridine synthases RluA (C746), RluB (C2605), RluC
(C955, C2504, C2580), RluE (C2457), and RluF (C2604)
are specific to the early assembly particles.
DISCUSSION
The nucleoside composition of subribosomal particles dem-
onstrates that the modification pattern of rRNA is similar in
both erythromycin and chloramphenicol-induced particles
of the same size. Modifications are added gradually during
TABLE 3. Nucleoside composition of 23S rRNA of different stages of large ribosome
subunit assembly
Nucleoside Free 50S Ery 45S Cam 45S Ery 35S Cam 35S
c89 637864806269666563
m
1
G (745) 95 648964926292619269
m
5
U (747, 1939) 84 636865726254615265
m
6
A (1618, 2030) 100 63 101 629 92 6178618063
m
2
G (1835, 2445) 90 6382638062836187610
m
5
C (1962) 66 635667486341624267
m
7
G (2069) 103 667966806455616 83 610
Gm (2251) 98 645965386149624661
Cm (2498) 89 647761896442695261
m
2
A (2503) 100 629269926185628060
Um (2552) 63 611 32 611 35 660 0
Details as in Table 1.
TABLE 2. Nucleoside composition of 16S rRNA of the different
stages of small ribosome subunit assembly
Nucleoside Free 30S Ery 25S Cam 25S
c(516) 87 611 56 6561617
m
7
G (527) 83 616 15 652268
m
2
G (966, 1207, 1516) 45 615 21 622463
m
5
C (967, 1407) 58 610 48 614861
m
4
Cm (1402) 70 615 55 627 54 629
m
3
U (1498) 20 613 4 62363
m
62
A (1518, 1519) 23 618 3 62361
Ribosomal particles were isolated from E. coli grown in the
presence of chloramphenicol or erythromycin or in the absence
of drugs. Nucleosides of the 16S rRNA were separated and
quantified by reverse-phase HPLC. Amount of nucleosides is
expressed as percentage of mature 70S ribosomes as average of
three independent experiments.
rRNA modification during ribosome assembly
www.rnajournal.org 2027
association of r-proteins with the rRNA and thereby with the
growing S-value indicating that the synthesis of rRNA
modified nucleosides depends on the progression of ribo-
some assembly.
Two modified nucleosides are added to the rRNA at equal
levels during all steps of ribosome assembly. The modifica-
tions m
4
Cm1402 of 16S rRNA are synthesized in an appar-
ently stochastic way throughout the ribosome subunit as-
sembly. Kimura and Suzuki (2010) reported recently that in
the cell-free assay 30S subunit, and not free 16S rRNA, is the
substrate for RsmH and RsmI in m
4
Cm1402 formation.
Other rRNA modifications can be divided into three groups:
early, intermediate, and late assembly specific modifications.
In the 16S rRNA of 25S particles the fraction of m
5
Cis
z50% and in the free 30S subunits z60% (Fig. 3; Table 1).
Thus, about half of m
5
C is added to the 16S rRNA during the
early step and the second half during the late steps of
ribosome small subunit assembly. Notably, only a small
fraction of m
5
C is added during 25S to 30S transition of the
ribosome small subunit. The fact that the first half of m
5
Cis
added during the early step and the second half during the
late stages of ribosome small subunit assembly suggests that
the RsmB (methylates C967) and RsmF (C1407) have dif-
ferent specificity with respect to ribosome assembly. Cell-free
studies have revealed that RsmB can methylate C967 only on
the protein-free 16S rRNA and is blocked when the ribo-
somal proteins S7 and S19 are added (Weitzmann et al.
1991). Methylation of C1407 is thought to be a late event
during ribosome assembly as the in vitro substrate of RsmF is
a 30S subunit rather than naked 16S rRNA (Andersen and
Douthwaite 2006). Based on the results of the nucleoside
composition of assembly intermediate particles and enzyme
specificity in vitro, we propose that m
5
C967 is made during
the early step and m
5
C1407 during the late events of small
subunit assembly.
A significant fraction of C516 is added
during early and intermediate steps
(transition 25S to 30S) of small subunit
assembly. This observation agrees with the
known specificity of RsuA (Wrzesinski
et al. 1995). Formation of m
7
G527 of
16S rRNA is clearly an intermediate
assembly event. Other modified nucleo-
sides of 16S rRNA (m
2
G at 966, 1207,
1516; m
5
C1407; m
3
U1498; m
62
A at 1518
and 1519) are synthesized at the level of
the 30S subunit during the late steps of
ribosome assembly. Thus, seven out of 11
modified nucleosides of 16S rRNA are
late assembly specific. This tendency
was noted earlier by Kaczanowska and
Ryde
´n-Aulin (2007), based on a compre-
hensive review by Ofengand and Del
Campo (2004). The results on the 16S
rRNA-specific modification enzymes de-
termined in this study are summarized in Figure 6A.
The modified nucleoside composition of 23S rRNA in the
early assembly particles (35S) reveals the early assembly
specific modifications in the large ribosome subunit. All
pseudouridines, except the RluD-specific ones in helix–loop
69 of 23S rRNA, m
1
G745, m
6
A1618, m
6
A2030, m
7
G2069,
m
2
G1835, m
2
G2495, and m
2
A2503 are clearly early assembly
specific (Fig. 6B).
Single m
5
C residue at position 1962 of E. coli 23S rRNA is
the product of RlmI (YccW) (Purta et al. 2008). The peak at
11.4 min of the HPLC derived from 35S, 45S particles, and
from free 50S subunits constitutes 40%, 50%, and 70%,
respectively (Fig. 3; Table 2). This peak contains two
nucleosides m
5
C and m
3
C(Gehrke and Kuo 1989). The
absorbance at 260 nm of m
5
C and m
3
Cis roughly equal. We
anticipate that in 70S both nucleosides contribute to the peak
FIGURE 5. Primer extension analysis of the pseudouridines in helix–
loop 69 of 23S rRNA. 23S rRNA was isolated from mature 70S
ribosomes, free 50S subunits, and Cam 45S, Ery 45S, Cam 35S, and
Ery 35S particles and analyzed for pseudouridines by CMCT/alkali
and reverse transcriptase directed primer extension. +, CMCT/alkali
treatment; , untreated RNA. Sequence of the 23S rRNA around
helix–loop 69 is shown in sequencing lanes (A, C, G, T). Note that the
CMCT induced stop site is one nucleotide below the actual site.
FIGURE 4. HPLC analysis of 23S rRNA. 23S rRNA was isolated from mature 70S ribosomes,
free 50S subunits, Cam 45S, and Cam 35S particles. Peaks corresponding to four standard
nucleosides (C, U, G, and A) and m
5
C, Cm, m
7
G, m
5
U, Um, Gm, m
1
G, m
2
G, m
2
A, and m
6
A are
indicated. Peak corresponding to the m
5
C contains also m
3
C. X corresponds to an unknown
compound.
Siibak and Remme
2028 RNA, Vol. 16, No. 10
at 11.4 min by 50%. Thus, the distribution of the two
nucleosides is not immediately clear from the data. The
intensity of the reverse transcriptase stop site at positio n 1915
shows that 23S rRNA of 35S particles does not contain m
3
C,
and 45S particles contain only a small fraction of m
3
C(Fig.
5). The formation of m
3
Cwas shown to occur on the level of
the 70S ribosomes (Ero et al. 2008). Therefore, the 11.4-min
peaks of the 35S and 45S particles probably contain mostly
m
5
C. Moreover, z80% of the free 50S subunits are in-
completely assembled precursors of the large ribosome
subunit (Peil et al. 2008; Al Refaii and Alix 2009). Thus, the
11.4-min peak of the free 50S can contain only 20% of mature
23S rRNA and the corresponding amount of m
3
C. There-
fore, m
5
C is present in the free 50S particles by z100%. These
results show that C1962 is methylated at C5 during early
assembly. This is consistent with the in vitro results showing
that methyltransferase RlmI is active on the naked 23S rRNA
molecules. No methylation was detected on the 50S subunits
or tight-couple 70S ribosomes (Purta et al. 2008).
E. coli 23S rRNA contains two m
5
U residues (m
5
U747,
m
5
U19389). m
5
U is found in the 35S particles in z50%
(Table 3). The amount of m
5
U methylation increases from
35S to 45S particles by 20% and reaches about 85% in the free
50S subunits (Fig. 3; Table 1). This suggests that one m
5
Uis
made during the early step and the second m
5
U is made
during the intermediate steps of 50S maturation. The
nucleotide m
5
U747 is located close to the peptide exit tunnel
deep inside the mature 50S subunit (Schuwirth et al. 2005).
Other modified nucleosides in the 750 loop of 23S rRNA
(m
1
G745, C746) are made during early assembly (Fig. 6B).
Therefore it is reasonable to assume that m
5
U747 is an early
assembly specific modification, too.
Taken together, early assembly specific modifications
are m
1
G745, C746, m
5
U747, C955, m
6
A1618, m
6
A2030,
m
7
G2069, m
2
G1835, m
5
C1962, C2457, m
2
G2495, m
2
A2503,
C2504, C2580, C2604, and C2605 (Fig. 6B). Thus, 17 out of
25 modified nucleosides of 23S rRNA are made during the
early steps of ribosome assembly. The limited time window
for synthesis of nucleoside modifications during the large
ribosome subunit assembly requires efficient coordination of
ribosome assembly events.
As discussed above, the nucleotide m
5
U747 is synthesized
during the early stages of 50S subunit assembly. Therefore,
the second m
5
U at position 1939 is made during the
intermediate assembly events. We conclude that nucleotides
m
5
U1939, Gm2251, and Cm2498 are synthesized during the
intermediate steps of the large ribosome subunit assembly.
It is evident that the corresponding regions of 23S rRNA are
accessible for the modification enzymes RlmD, RlmB, and
RlmM not only on the protein-free precursor-23S rRNA but
also during the intermediate steps of large ribosome subunit
assembly. Substrate specificity is known only for the RlmM,
which is able to catalyze methylation of C2498 in the protein-
free 23S rRNA but not in the tight-coupled 70S ribosomes
or 50S subunits (Purta et al. 2009).
23S rRNA of free 50S subunits contains z60% of Um
(Table 3) and low levels of U1911 and U1917 (Fig. 5). Thus,
the late assembly events of the large ribosome subunit involve
formation of Um2552, and isomerization of the uridines
in h69.
It is important to stress that the assembly dependence
of rRNA modification in vivo is in very good agreement
with the published specificities of modification enzymes
determined in vitro, as is evident by comparing Table 1 and
Figure 6.
Coordination of the rRNA modification and the associa-
tion of ribosomal protein with the subunit is important for
several reasons. First, rRNA folding is a step-wise process
where transient structures play an important role (Besancon
and Wagner 1999; Liiv and Remme 2004; Adilakshmi et al.
FIGURE 6. Summary of the specificity of the rRNA modification
enzymes with respect to ribosome subunit assembly. Ribosome assembly
is divided into three stages (early, intermediate, and late), which are
shown by white and gray zones. Activity of rRNA modification
enzymes is shown by black bars. (A) Modification of 16S rRNA. (B)
Modification of 23S rRNA.
rRNA modification during ribosome assembly
www.rnajournal.org 2029
2008). Therefore, the structures that are recognized by
modification enzymes are formed during the progression
of ehe ribosome subunit assembly. The second aspect is the
role of ribosomal proteins in directing rRNA folding upon
binding to the rRNA (Holmes and Culver 2004; Talkington
et al. 2005). In this way, r-proteins could help to create the
recognition sites for rRNA modification enzymes. On the
other hand, r-proteins can inhibit rRNA modification by
shielding the modification site. The dual role of r-proteins in
rRNA modification is illustrated by proteins S7 and S19,
which are necessary for the m
2
G966 formation by RsmD but
block the synthesis of m
5
C967 by RsmB (Weitzmann et al.
1991). The modification sites that are buried deep inside the
subunit must be modified during early assembly. Finally, it
is noteworthy that a specific set of r-proteins are exchange-
able in vivo (Pulk et al. 2010). When the ribosome bound
r-proteins can be exchanged with the free r-proteins, they
could be displaced by the rRNA modifications as well. In this
way, the rRNA modification sites that are already masked
by r-proteins, can become accessible to the modification
enzymes during the late stages of ribosome assembly.
MATERIALS AND METHODS
Preparation of rRNA
Escherichia coli strain MG1655 (Blattner et al. 1997) was used in all
experiments. Cells were grown at 25°C in 200 mL 23YT medium
(Sambrook and Russell 2001) until the A
600
reached 0.2. At this
point, either erythromycin (final concentration of 100 mg/mL) or
chloramphenicol (final concentration of 7 mg/mL) was added,
followed by incubation for 2 h at 25°C. The control culture was
grown without antibiotics. Bacterial cells were collected by
centrifugation in a Sorvall GS-3 rotor at 4000 rpm at 4°C for 10
min and were resuspended in 1 mL lysis buffer (60 mM KCl, 60
mM NH
4
Cl, 50 mM Tris-HCl [pH 8], 6 mM MgCl
2
,6mM
b-mercaptoethanol,16%sucrose);lysozymeandDNaseI
(Amresco, GE Healthcare) were added to final concentrations of
1 mg/mL and 20 U/mL, respectively. Cells were frozen for 15 min
at –70°C and subsequently thawed in ice-cold water for 30 min.
The freeze–thaw cycle was repeated twice, followed by centrifu-
gation at 13,000gand 4°C for 20 min. Clear lysate was diluted
twofold with buffer A (60 mM KCl, 60 mM NH
4
Cl, 10 mM Tris-
HCl [pH 8], 12 mM MgCl
2
,6mMb-mercaptoethanol). Diluted
lysates (2 mL) were loaded onto 30 mL, 10%–25% (w/w) sucrose
gradient in buffer A. Ribosomal particles were separated by
centrifugation at 23,000 rpm in an SW28 rotor (Beckman) at
4°C for 13.5 h. Ribosome profiles were detected by continuous
monitoring of absorbance at 254 nm. 70S ribosomal particles were
precipitated with 2.5 volumes of ice-cold ethanol and collected by
centrifugation at 5000 rpm for 30 min in an HS4 rotor (Sorvall).
50S and 30S subunit fractions from untreated cells, and 45S, 35S,
and 25S particle fractions from antibiotic-treated cells were
diluted twofold with buffer A and concentrated by ultrafiltration
using Amicon Ultracel-100k filters. Ribosomal particles were
loaded onto sucrose gradients and separated as described above.
Ribosomal particles were precipitated by addition of ethanol
(2 volumes). rRNA was deproteinized by extraction with phenol
and chloroform followed by ethanol precipitation. RNA was
dissolved in water and stored at 80°C. 16S and 23S rRNA from
70S ribosomes and 35S particles were separated by centrifugation
in 5%–20% (w/w) sucrose gradient (buffer 20 mM Na-acetate,
100 mM NaCl, 1 mM EDTA) at 25 700 rpm in an SW28 rotor
(Beckman) at 4°C for 16 h. Purified 16S and 23S rRNA were
precipitated with ethanol.
High-performance liquid chromatography
For HPLC analysis, 2–4 A
260
Units rRNA (70–100 pmol) was
digested with nuclease P1 (MP Biochemicals) and bacterial
alkaline phosphatase (Fermentas Life Sciences) according to the
method of Gehrke and Kuo (1989). Nucleoside composition was
determined by RP-HPLC on a Supelcosil LC-18-S HPLC column
(25 cm 34.6 mm, 5 mm) equipped with a precolumn (4.6 mm 3
20 mm) at 30°C on a SHIMADZU Prominence HPLC system. The
following buffers were used: buffer A (10 mM NH
4
H
2
PO
4
, 2.5%
methanol at pH 5.3); buffer B (10 mM NH
4
H
2
PO
4
, 20% methanol
at pH 5.1); and buffer C (10 mM NH
4
H
2
PO
4
, 35% acetonitrile at
pH 4.9). RP-HPLC analysis was performed using the gradient
conditions of Gehrke and Kuo (1989): flow rate 1.0 mL/min held
at 0% buffer B 12 min, to 10% buffer B over 8 min, to 25% buffer
B over 5 min, to 60% buffer B over 8 min, to 64% buffer B over
4 min, to 100% buffer B over 9 min, 0%–100% buffer C over
35 min, held at 100% buffer C for 10 min, and equilibration with
0% buffer B for 30 min. Nucleoside absorbance profiles were
recorded at 260 and 280 nm, and peak areas were integrated.
Quantitative calculations were according to the following formula:
X% = (X
P
/N
P
)/[Average(X
70S
/N
70S
)]100, where X% is percentage
of modified nucleoside in the subribosomal particle; X
P
is-
modified nucleoside area of the subribosomal particle; X
70S
is
modified nucleoside area of the 70S ribosomes; N
P
is area of
corresponding unmodified nucleoside of the subribosomal parti-
cles; and N
70S
is area of corresponding unmodified nucleoside of
the 70S ribosomes.
Determination of pseudouridines
Pseudouridines were determined according to the method of
Ofengand et al. (2001). Fifteen micrograms rRNA was dissolved in
20 mL water, 80 mL of BEU buffer (7 M urea, 4 mM EDTA, 50
mM Bicine/NaOH [pH 8.5]), and 20 mL of CMCT/BEU buffer
(1 m CMCT in BEU buffer) (CMCT; Sigma-Aldrich Chemie
GmbH) were added. One hundred microliters of BEU buffer was
added to 15 mgofrRNAin20mLofwaterservingasthenegative
control. Both samples were incubated at 37°C for 20 min for CMCT
modification of U, G, and Cresidues. Reaction was stopped by
addition of 38 mL 4 M NaOAc and 600 mL of cold 96% ethanol.
Samples were kept at 20°Cfor10min,andtheRNAprecipitate
was collected by centrifugation at 6000gand 4°C. The supernatant
was carefully removed and RNA was washed twice with 600 mL
of 70% ethanol. The precipitate was dried at 37°C for 10 min. rRNA
was dissolved in 50 mL of NPK buffer (20 mM NaHCO
3
,30mM
Na
2
CO
3
, 2 mM EDTA), and the samples were incubated at 37°Cfor
4 h to allow removal of CMCT from U and G residues. After
incubation, rRNA was precipitated and washed as described above.
The precipitate was dissolved in 20 mL of water and stored
at 20°C. Pseudouridine sequencing of rRNA was carried out by
Siibak and Remme
2030 RNA, Vol. 16, No. 10
primer extension using primers U1 (CAGCCTGGCCATCATTA
CGCC), C7 (ACACCAGTGATGCGTCCAC), C17 (CCACTTTAA
ATGGCGAAC), C33 (GTTTGATTGGCCTTTCACCC), and T3
(GCTTTCTTTAAATGATGGCTGCTT), and AMV reverse transcrip-
tase (Seikagaku Corp.) in the presence of [a-
32
P]dCTP (Amersham
Biosciences). The resulting DNA fragments were resolved in 7%
polyacrylamide-urea gel. RadioactivitywasvisualizedbyaTyphoon
PhosphorImager (GE Healthcare).
ACKNOWLEDGMENTS
We thank Aivar Liiv, Lauri Peil, Margus Leppik, Tanel Tenson,
Kai Viruma
¨e, and Rya Ero (all from the University of Tartu) for
help and advice. This work was supported by Estonian Science
Foundation Grant No. 7509.
Received March 4, 2010; accepted July 15, 2010.
REFERENCES
Adesnik M, Levinthal C. 1969. Synthesis and maturation of ribosomal
RNA in Escherichia coli.J Mol Biol 46: 281–303.
Adilakshmi T, Bellur DL, Woodson SA. 2008. Concurrent nucleation
of 16S folding and induced fit in 30S ribosome assembly. Nature
455: 1268–1272.
Al Refaii A, Alix J. 2009. Ribosome biogenesis is temperature-
dependent and delayed in Escherichia coli lacking the chaperones
DnaK or DnaJ. Mol Microbiol 71: 748–762.
Andersen N, Douthwaite S. 2006. YebU is a m5C methyltransferase
specific for 16 S rRNA nucleotide 1407. J Mol Biol 359: 777–786.
Basturea G, Deutscher M. 2007. Substrate specificity and properties of
the Escherichia coli 16S rRNA methyltransferase, RsmE. RNA 13:
1969–1976.
Besancon A, Wagner R. 1999. Characterization of transient RNA–
RNA interactions important for the facilitated structure formation
of bacterial ribosomal 16S RNA. Nucleic Acids Res 27: 4353–4362.
Blattner FR, Plunkett G III, Bloch CA, Perna NT, Burland V, Riley M,
Collado-Vides J, Glasner JD, Rode CK, Mayhew GF, et al. 1997.
The complete genome sequence of Escherichia coli K-12. Science
277: 1453–1462.
Bu
¨gl H, Fauman E, Staker B, Zheng F, Kushner S, Saper M, Bardwell J,
Jakob U. 2000. RNA methylation under heat shock control. Mol
Cell 6: 349–360.
Caldas T, Binet E, Bouloc P, Costa A, Desgres J, Richarme G. 2000.
The FtsJ/RrmJ heat shock protein of Escherichia coli is a 23 S
ribosomal RNA methyltransferase. J Biol Chem 275: 16414–16419.
Champney W. 2006. The other target for ribosomal antibiotics:
Inhibition of bacterial ribosomal subunit formation. Infect Disord
Drug Targets 6: 377–390.
Dagley S, Sykes J. 1959. Effect of drugs upon components of bacterial
cytoplasm. Nature 183: 1608–1609.
Dodd J, Kolb J, Nomura M. 1991. Lack of complete cooperativity of
ribosome assembly in vitro and its possible relevance to in vivo
ribosome assembly and the regulation of ribosomal gene expres-
sion. Biochimie 73: 757–767.
Ero R, Peil L, Liiv A, Remme J. 2008. Identification of pseudouridine
methyltransferase in Escherichia coli.RNA 14: 2223–2233.
Gehrke C, Kuo K. 1989. Ribonucleoside analysis by reversed-phase high-
performance liquid chromatography. J Chromatogr 471: 3–36.
Hansen L, Kirpekar F, Douthwaite S. 2001. Recognition of nucleotide
G745 in 23 S ribosomal RNA by the rrmA methyltransferase. J Mol
Biol 310: 1001–1010.
Holmes KL, Culver GM. 2004. Mapping structural differences
between 30S ribosomal subunit assembly intermediates. Nat Struct
Mol Biol 11: 179–186.
Huang L, Ku J, Pookanjanatavip M, Gu X, Wang D, Greene P, Santi
D. 1998. Identification of two Escherichia coli pseudouridine
synthases that show multisite specificity for 23S RNA. Biochemistry
37: 15951–15957.
Kaczanowska M, Ryde
´n-Aulin M. 2007. Ribosome biogenesis and the
translation process in Escherichia coli.Microbiol Mol Biol Rev 71:
477–494.
Kimura S, Suzuki T. 2010. Fine-tuning of the ribosomal decoding
center by conserved methyl-modifications in the Escherichia coli
16S rRNA. Nucleic Acids Res 38: 1341–1352.
Kowalak JA, Bruenger E, Hashizume T, Peltier JM, Ofengand J, McCloskey
JA. 1996. Structural characterization of U*-1915 in domain IV from
Escherichia coli 23S ribosomal RNA as 3-methylpseudouridine. Nucleic
Acids Res 24: 688–693.
Kurland C, Nomura M, Watson J. 1962. The physical properties of the
chloromycetin particles. J Mol Biol 4: 388–394.
Leppik M, Peil L, Kipper K, Liiv A, Remme J. 2007. Substrate
specificity of the pseudouridine synthase RluD in Escherichia coli.
FEBS J 274: 5759–5766.
Lesnyak D, Sergiev P, Bogdanov A, Dontsova O. 2006. Identification
of Escherichia coli m
2
G methyltransferases: I. the ycbY gene
encodes a methyltransferase specific for G2445 of the 23 S rRNA.
J Mol Biol 364: 20–25.
Liiv A, Remme J. 2004. Importance of transient structures during
post-transcriptional refolding of the pre-23S rRNA and ribosomal
large subunit assembly. J Mol Biol 342: 725–741.
Lindahl L. 1975. Intermediates and time kinetics of the in vivo
assembly of Escherichia coli ribosomes. J Mol Biol 92: 15–37.
Nierhaus KH. 1991. The assembly of prokaryotic ribosomes. Biochimie
73: 739–755.
Nomura M, Hosokawa K. 1965. Biosynthesis of ribosomes: Fate of
chloramphenicol particles and of pulse-labeled RNA in Escherichia
coli.J Mol Biol 12: 242–265.
Ofengand J, Del Campo M, Kaya Y. 2001. Mapping pseudouridines
in RNA molecules. Methods 25: 365–373.
Ofengand J, Del Campo M. 2004. Modified nucleosides of Escherichia
coli ribosomal RNA. In EcoSal—Escherichia coli and Salmonella:
Cellular and molecular biology (ed. G Bjo
¨rk), ASM Press, Wash-
ington, DC. http://www.ecosal.org/.
Okamoto S, Tamaru A, Nakajima C, Nishimura K, Tanaka Y, Tokuyama
S, Suzuki Y, Ochi K. 2007. Loss of a conserved 7-methylguanosine
modification in 16S rRNA confers low-level streptomycin resistance
in bacteria. Mol Microbiol 63: 1096–1106.
Peil L, Viruma
¨e K, Remme J. 2008. Ribosome assembly in Escherichia
coli strains lacking the RNA helicase DeaD/CsdA or DbpA. FEBS
J275: 3772–3782.
Poldermans B, Roza L, Van Knippenberg PH. 1979. Studies on the
function of two adjacent N6,N6-dimehyadenosines near the 39end
of 16 S ribosomal RNA of Escherichia coli. III Purification and
properties of the methylating enzyme and methylase-30 S in-
teraction. J Biol Chem 254: 9094–9100.
Pulk A, Liiv A, Peil A, Maiva
¨li U
¨, Nierhaus KH, Remme J. 2010.
Ribosome reactivation by replacement of damaged proteins. Mol
Microbiol 75: 801–814.
Purta E, O’Connor M, Bujnicki J, Douthwaite S. 2008. YccW is the
m5C methyltransferase specific for 23S rRNA nucleotide 1962.
J Mol Biol 383: 641–651.
Purta E, O’Connor M, Bujnicki J, Douthwaite S. 2009. YgdE is the
29-O-ribose methyltransferase RlmM specific for nucleotide C2498
in bacterial 23S rRNA. Mol Microbiol 72: 1147–1158.
Raychaudhuri S, Conrad J, Hall B, Ofengand J. 1998. A pseudouridine
synthase required for the formation of two universally conserved
pseudouridines in ribosomal RNA is essential for normal growth
of Escherichia coli.RNA 4: 1407–1417.
Sambrook J, Russell DW. 2001. Molecular cloning: A laborator y manual,3rd
ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Schuwirth B, Borovinskaya M, Hau C, Zhang W, Vila-Sanjurjo A,
Holton J, Cate J. 2005. Structures of the bacterial ribosome at 3.5 A
˚
resolution. Science 310: 827–834.
rRNA modification during ribosome assembly
www.rnajournal.org 2031
Sergiev P, Lesnyak D, Bogdanov A, Dontsova O. 2006. Identification
of Escherichia coli m2G methyltransferases: II. The ygjO gene
encodes a methyltransferase specific for G1835 of the 23 S rRNA.
J Mol Biol 364: 26–31.
SergievP,SerebryakovaM,BogdanovA,DontsovaO.2008.TheybiNgene
of Escherichia coli encodes adenine-N6 methyltransferase specific for
modificationofA1618of23Sribosomal RNA, a methylated residue
located close to the ribosomal exit tunnel. JMolBiol375: 291–300.
Siibak T, Peil L, Xiong L, Mankin A, Remme J, Tenson T. 2009.
Erythromycin- and chloramphenicol-induced ribosomal assembly
defects are secondary effects of protein synthesis inhibition.
Antimicrob Agents Chemother 53: 563–571.
Talkington M, Siuzdak G, Williamson J. 2005. An assembly landscape
for the 30S ribosomal subunit. Nature 438: 628–632.
Tscherne J, Nurse K, Popienick P, Ofengand J. 1999. Purification,
cloning, and characterization of the 16 S RNA m2G1207 methyl-
transferase from Escherichia coli.J Biol Chem 274: 924–929.
Usary J, Champney W. 2001. Erythromycin inhibition of 50S
ribosomal subunit formation in Escherichia coli cells. Mol Micro-
biol 40: 951–962.
Vaidyanathan P, Deutscher M, Malhotra A. 2007. RluD, a highly
conserved pseudouridine synthase, modifies 50S subunits more
specifically and efficiently than free 23S rRNA. RNA 13: 1868–
1876.
Weitzmann C, Tumminia S, Boublik M, Ofengand J. 1991. A
paradigm for local conformational control of function in the
ribosome: Binding of ribosomal protein S19 to Escherichia coli 16S
rRNA in the presence of S7 is required for methylation of m2G966
and blocks methylation of m5C967 by their respective methyl-
transferases. Nucleic Acids Res 19: 7089–7095.
Wrzesinski J, Nurse K, Bakin A, Lane B, Ofengand J. 1995. A dual-
specificity pseudouridine synthase: An Escherichia coli synthase
purified and cloned on the basis of its specificity for c746 in 23S
RNA is also specific for c32 in tRNA
phe
.RNA 1: 437–448.
Siibak and Remme
2032 RNA, Vol. 16, No. 10
