Mechanism of N6-threonylcarbamoyladenonsine (t6A) biosynthesis: Isolation and characterization of the intermediate threonylcarbamoyl-AMP

Article abstract of DOI:10.1021/bi301233d

Genetic and biochemical studies have recently implicated four proteins required in bacteria for the biosynthesis of the universal tRNA modified base N6-threonylcarbamoyl adenosine (t⁶A). In this work, t⁶A biosynthesis in Bacillus subtilis has been reconstituted in vitro and found to indeed require the four proteins YwlC (TsaC), YdiB (TsaE), YdiC (TsaB) and YdiE (TsaD). YwlC was found to catalyze the conversion of l-threonine, bicarbonate/CO₂ and ATP to give the intermediate l-threonylcarbamoyl- AMP (TC-AMP) and pyrophosphate as products. TC-AMP was isolated by HPLC and characterized by mass spectrometry and ¹H NMR. NMR analysis showed that TC-AMP decomposes to give AMP and a nearly equimolar mixture of l-threonine and 5-methyl-2-oxazolidinone-4-carboxylate as final products. Under physiological conditions (pH 7.5, 37 °C, 2 mM MgCl₂), the half-life of TC-AMP was measured to be 3.5 min. Both YwlC (in the presence of pyrophosphatase) and its Escherichia coli homologue YrdC catalyze the formation of TC-AMP while producing only a small molar fraction of AMP. This suggests that CO₂ and not an activated form of bicarbonate is the true substrate for these enzymes. In the presence of pyrophosphate, both enzymes catalyze clean conversion of TC-AMP back to ATP. Purified TC-AMP is efficiently processed to t⁶A by the YdiBCE proteins in the presence of tRNA substrates. This reaction is ATP independent in vitro, despite the known ATPase activity of YdiB. The estimated rate of conversion of TC-AMP by YdiBCE to t⁶A is somewhat lower than the initial rate from l-threonine, bicarbonate and ATP, which together with the stability data, is consistent with previous studies that suggest channeling of this intermediate.

Full text of DOI:10.1021/bi301233d

Article
pubs.acs.org/biochemistry
Mechanism of N6-Threonylcarbamoyladenonsine (t6A) Biosynthesis:
Isolation and Characterization of the Intermediate
Threonylcarbamoyl-AMP
Charles T. Lauhon*
Pharmaceutical Sciences Division, School of Pharmacy, University of Wisconsin, Madison, Wisconsin 53705, United States
S
* Supporting Information
ABSTRACT: Genetic and biochemical studies have recently implicated four proteins required in bacteria for the biosynthesis of
the universal tRNA modified base N6-threonylcarbamoyl adenosine (t6A). In this work, t6A biosynthesis in Bacillus subtilis has
been reconstituted in vitro and found to indeed require the four proteins YwlC (TsaC), YdiB (TsaE), YdiC (TsaB) and YdiE
(TsaD). YwlC was found to catalyze the conversion of L-threonine, bicarbonate/CO2 and ATP to give the intermediate L-
threonylcarbamoyl-AMP (TC-AMP) and pyrophosphate as products. TC-AMP was isolated by HPLC and characterized by mass
spectrometry and 1H NMR. NMR analysis showed that TC-AMP decomposes to give AMP and a nearly equimolar mixture of L-
threonine and 5-methyl-2-oxazolidinone-4-carboxylate as final products. Under physiological conditions (pH 7.5, 37 °C, 2 mM
MgCl2), the half-life of TC-AMP was measured to be 3.5 min. Both YwlC (in the presence of pyrophosphatase) and its
Escherichia coli homologue YrdC catalyze the formation of TC-AMP while producing only a small molar fraction of AMP. This
suggests that CO2 and not an activated form of bicarbonate is the true substrate for these enzymes. In the presence of
pyrophosphate, both enzymes catalyze clean conversion of TC-AMP back to ATP. Purified TC-AMP is efficiently processed to
t6A by the YdiBCE proteins in the presence of tRNA substrates. This reaction is ATP independent in vitro, despite the known
ATPase activity of YdiB. The estimated rate of conversion of TC-AMP by YdiBCE to t6A is somewhat lower than the initial rate
from L-threonine, bicarbonate and ATP, which together with the stability data, is consistent with previous studies that suggest
channeling of this intermediate.
initial work to isolate the biosynthetic enzymes involved led to
ells devote significant energy and information content to
Cthe modification of nucleosides in their tRNA. These the enrichment of t6A activity and partial characterization of
proteins required.15,16 However, complete characterization
proved challenging because of the number and likely low
expression levels of the proteins involved.
A recent flurry of genetic and biochemical work in Escherichia
coli and yeast has identified multiple genes required for t6A
formation in these strains. Initially, the E. coli yrdC gene and its
yeast homologue SUA5 were demonstrated to be required for
t6A formation.17 YrdC is indispensable for growth in E. coli,
while yeast sua5Δ null strains can be constructed, but the
mutation results in a severe growth defect. Consistent with the
modified bases are essential for proper codon recognition,
aminoacylation, and reading frame maintenance during trans-
lation.1,2 The modified nucleoside N6-threonylcarbamoyl
adenosine (t6A, Scheme 1)3 is universally conserved in nature
and found at position 37 in nearly all cytosolic tRNAs that
recognize 5′-ANN-3′ codons. The importance of the biological
role for t6A has been related in vitro to increased stability of
adjacent base pairing in the anticodon−codon interaction, as
well as maintenance of anticodon loop shape and flexibility.4−9
In vivo, t6A has been shown to be important for start codon
recognition and reading frame maintenance.10,11 Early work on
t6A biosynthesis eliminated the involvement of carbamoyl
phosphate and pointed toward ATP and bicarbonate as sources
of the activated carbamoyl carbon (Figure 1).12−14 Impressive
Received: September 10, 2012
Revised: October 15, 2012
Published: October 16, 2012
© 2012 American Chemical Society
8950
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
Scheme 1. Requirements for the Biosynthesis of t6A in B.
subtilis
carbamoyl group is transferred to the substrate in a reaction
that utilizes a novel mode of iron chelation in the active site.
Despite the recent studies on these systems, several questions
remain, including the mechanism of formation of the putative
carbamoyl adenylate intermediate, the characterization of its
reactivity, and the precise role of each bacterial protein in t6A
biosynthesis. For example, because of it is demonstrated
binding to RNA,34 YrdC was proposed to catalyze the addition
of either activated bicarbonate or an activated threonylcarba-
moyl group to the tRNA. Because of the noted difficulty in
expression and stability of E. coli YgjD, the Bacillus subtilis
homologues were targeted in this work as a possible tractable
system to reconstitute and study the mechanism of t6A
biosynthesis in vitro (Figure 1). In B. subtilis, the homologues
for E. coli yjeE/tsaE, yeaZ/tsaB and ygjD/tsaD (ydiB, ydiC and
ydiE, respectively) are found as sequential genes in a single
operon. The yrdC/tsaC homologue ywlC is located at a separate
site in the genome. The ydiB, ydiC and ywlC genes were initially
categorized as essential in B. subtilis;35 however a reinvesti-
gation showed that deletion mutants for ydiB and ywlC can be
prepared, but the mutants share a serious growth defect.36 The
phenotypes for the two mutants are very similar, which was an
early clue that they may both be involved in the same pathway.
Similar to its E. coli homologue YjeE, YdiB has measurable
ATPase activity that is required for normal growth.37
Preliminary work in this laboratory found that a B. subtilis
ywlC conditional mutant dependent on IPTG for expression
produced diminished levels of t6A when grown in the absence
of the inducer. YwlC, YdiC and YdiE have not been purified
and characterized to date. In this paper, the expression,
purification and reconstitution of t6A biosynthesis in B. subtilis
is reported with a particular focus on the characterization of L-
threonylcarbamoyl AMP (TC-AMP) as the product of the
YwlC/TsaC protein.
universal presence of t6A in tRNA, homologues of this gene
family have been identified in every complete genome
sequenced to date.17 More recently, the E. coli ygjD gene19
and the yeast homologue KAE118−20 were also found to be
required. Initial structural work of Kae1 homologues in archaea
showed it to be a novel iron-containing ATPase with
metalloendonuclease activity.21 Subsequently, additional re-
quired genes have been identified that reveal a functional
divergence in t6A biosynthesis between bacteria and eukaryotes.
In yeast, Kae1 is part of the EKC/KEOPS complex, which
contains five proteins, three of which (Kae1, Bud32, and Pcc1)
are required for the formation of t6A.19 The biological activities
linked to this complex are wide ranging  from telomere
stability22,23 to transcription defects24 and genomic insta-
bility.25,20 In E. coli, Handford, et al. found evidence that YgjD
functions as a complex with the YjeE and YeaZ proteins.26
YeaZ was suspected to be involved in proteolytic degradation of
YgjD.26 YjeE and its H. influenza ortholog have low but
measurable ATPase activity,27 and both yeaZ and yjeE are listed
as essential in E. coli.28 The bacterial system differs from that in
yeast and other eukaryotes since no homology exists between
YjeE and YeaZ and the proteins of the EKC/KEOPS complex,
although it was shown that two other proteins in the yeast
complex, Pcc118,19 and Bud3219 are required for t6A
modification in vivo. Armed with this information, Deutsch et
al recently reported the first in vitro reconstitution of t6A
biosynthesis using purified components in E. coli.29 The in vitro
modification reaction requires YrdC, the putative protein
complex of YgjD, YeaZ and YjeE, in addition to L-threonine,
ATP and bicarbonate. The authors used ATPase assays to show
that YrdC catalyzes L-threonine dependent AMP formation and
that both AMP and ADP are produced during the reaction. The
genes have therefore been renamed using the abbreviation tsa
(yeaZ = tsaB, yrdC = tsaC, ygjD = tsaD, and yjeE = tsaE).29
Finally, recent structural studies have shed light on
homologous systems that catalyze the transfer of carbamoyl
groups to both small molecule and protein acceptors. First, the
structure of a YrdC/SUA5 homologue from S. tokodaii was
solved in a complex with L-threonine and an analogue of
ATP.30 Subsequently, the structures of HypF,31,32 an enzyme
MATERIALS AND METHODS
■
Materials and General Methods. DNA oligonucleotides
were from IDT Technologies (Coralville, IA) and purified on
polyacrylamide gels before use. Biochemicals and enzymes were
from Sigma-Aldrich, except for L-threonine, which was from
ChemImpex. Radiolabeled α-32P-ATP (3000 Ci/mmol) was
from Perkin-Elmer Life Sciences. T7 RNA polymerase was
purified from strain BL21(pAR219), which was a generous gift
of F. Studier. B. subtilis ywlC mutant strain BFS215 was a
generous gift of P. Stragier. Mass spectrometry and NMR
spectroscopy was performed at the Analytical Instrumentation
Center at the University of Wisconsin School of Pharmacy.
DNA sequencing was performed at the University of Wisconsin
Biotechnology Center.
Cloning, Expression and Purification of Proteins. The
ywlC and ydiB genes were amplified by PCR from B. subtilis 168
genomic DNA, digested and ligated into pET15b. The ligation
mixtures were used to transform NovaBlue cells (Novagen) to
give N-terminal His6 fusion expression plasmids pCL219
(YwlC) and pCL223 (YdiB). Oligonucleotides used for cloning
ywlC were 5′- GTT TGG TCACAT ATG AAAACG AAA
AGATGG TTT-3′ (N-terminal) and 5′- ATC TGA GGATCC
TCA GCGAAT CAC TCTTCC TCC-3′ (C-terminal) and for
ydiB 5′- CGG GTG ACTCAT ATG AAGCAA TTA AAATGG
AGA-3′ (N-terminal) and 5′- TAA TTT GGATCC TCA
ATTGCT AAT ATTGTC ATG-3′ (C-terminal). Plasmids
were isolated using a spin miniprep kit (Qiagen), verified for
sequence and used to transform E. coli chemically competent
required for hydrogenase maturation, and TobZ,33
a
carbamoyltransferase involved in secondary metabolism show
a mechanism for activation and transfer of a carbamoyl moiety
starting with carbamoyl phosphate. In both HypF and TobZ, a
carbamoyl adenylate intermediate is formed by initial
dephosphorylation of carbamoyl phosphate, followed by attack
of the carbamate anion on ATP. These structures are highly
relevant to t6A biosynthesis because this enzyme is a fusion of
both YrdC and YgjD type proteins. The structure reveals that
the carbamoyladenylate intermediate formed by the YrdC-
domain is likely channeled to the YgjD domain where it is
8951
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
BL21(DE3) cells. Protein was overproduced by growing cells at
37 °C to A595 = 0.6, then adding IPTG to 1 mM and harvesting
cells after 5 h additional growth at the same temperature. Cells
from 100 mL cultures were collected by centrifugation and
resuspended in 2 mL of lysis buffer (10 mM Tris pH 7.5, 1 mM
MgCl2) containing 0.3 mg/mL lysozyme. After incubation at rt
for 20 min, cells were cooled on ice and disrupted by sonication
(3 × 20 s), and cell debris was pelleted by centrifugation
(12800g for 30 min). Supernatant was loaded onto a 1 mL
column of HisBind resin (EMD Biosciences) equilibrated in
binding buffer (50 mM NaPi pH 8, 300 mM NaCl, 25 mM
imidazole). Columns were washed with 10 volumes binding
buffer, 4 volumes wash buffer (binding buffer with 40 mM
imidazole) and fusion proteins eluted with 3 column volumes
of elution buffer (binding buffer with 250 mM imidazole). The
main elution fraction (1 mL) was loaded onto a 10 mL
Sephadex G-50 column equilibrated and eluted with 50 mM
sodium phosphate buffer, pH 7.5, 300 mM NaCl, 5% glycerol.
Fractions containing protein were pooled, aliquotted and stored
at −78 °C.
Because of solubility issues of the His6-fusion proteins,
expression plasmids for both ydiC and ydiE (pCL242 and pCL
245) were constructed as the N-terminal thioredoxin (trxA)
fusions using the ligation independent cloning vector pET-32
Xa/LIC (Novagen/EMD Biosciences) according to the
manufacturer protocol. Oligonucleotides used for PCR
amplification for ydiC were 5′-GGTATTGAGGGT CGCAT-
GACAATATTAGCAATTGAT-3′(N-terminal), and 5′-AGA
GGAGAGTTAGAGCCCTACTTTTGACTTTCAATCCA-
3′(C terminal), and for ydiE 5′-GGT ATTGAGGGTCGCAT-
GAGTGAGCAAAAAGACATG-3′ (N-terminal) and 5′-
AGAGGA GAGTTAGAGCCTTATCTCGTGAGACTTTGA-
TA-3′ (C-terminal). To increase expression level and solubility,
the fusion proteins were expressed by growing at 37 °C to A595
= 0.6, then cooling to 21 °C over 20 min, followed by addition
of IPTG to 1 mM and incubating 12 h at the reduced
temperature. Cells were collected by centrifugation (5000g 20
min at 4 °C), disrupted by sonication at a concentration of 1
mL in lysis buffer (10 mM Tris pH 7.5) per 50 mL culture.
Fusion protein was purified on the basis of the His6-tag using
HisBind NTA resin similar to the procedure described above
for YwlC and YdiB. The trx-YdiE fusion protein was soluble
above 1 mg/mL, whereas the concentration of the trx-YdiC
protein was kept below 0.7 mg/mL to prevent precipitation and
aggregation.
RNA Substrates. RNA substrates were obtained by runoff
transcription using T7 RNA polymerase as described
previously.38 DNA templates for the preparation of short
stem-loop RNAs were single stranded oligodeoxynucleotides
comprising the antisense of the desired RNA sequence and a
17-nt region that is the antisense of the T7 promoter. For
example, the DNA template for B. subtilis tRNAThr anticodon
stem loop was 5′- ACTGATTACAAGTCAGTC CTATAGT-
GA GTC GTATTA-3′. These templates were used in
transcription reactions with the 17 nt primer (5- TAATAC-
GACTCACTATAG −3′) containing the T7 promoter
sequence. To prepare unmodified tRNAs, two overlapping
DNA oligonucleotides were used as templates after conversion
to double stranded DNA by reverse transcriptase as previously
described.39 For example, DNA oligonucleotides used for B.
subtilis tRNAThr were 5′-TAATAC GACTCA CTATAG CCG
GTGTAG CTC AATTGGTAG AGC AACTGA CTT GTA
ATC −3′ and 5-TGGTGC CGG CAA GAG GACTTG AAC
CCC CAA CCT ACT GAT TAC AAGTCA −3′. Tran-
scription reactions contained 50 mM Tris, pH 7.8, 50 mM KCl,
10 mM DTT, 1 mM spermidine, 5 mM each of ATP,
UTP,CTP, GTP, 35 mM MgCl2, 1 μM template DNA, RNasin
(0.02 U/μL), and T7 RNA Polymerase 0.1 U/μL. Reactions
were incubated at 37 °C for 12−16 h. Template DNA was
digested by the addition of RQ1 DNase (0.1 U/mL) for 1 h at
37 °C, and an equal volume of 7.5 M urea containing 0.5%
bromophenol blue was added. This mixture was heated to 90
°C for 2 min and the RNA was purified on a 8% polyacrylamide
slab gel. Bands containing RNA of interest were cut from the
gel and eluted in 0.5 M NaCl at RT for 4 h. The gel slices were
removed by filtration, the RNA was precipitated with 2.5
volumes of cold ethanol and after centrifugation (10000g, 30
min), the RNA was dissolved in water and its concentration
measured by absorbance at 260 nm.
HPLC Assay for t6A formation. Complete reaction
mixtures (50 μL) contained 50 mM Tris pH 7.2, 20 mM
MgCl2, 25 mM KCl, 5 mM L-threonine, 2 mM ATP, 20 mM
NaHCO3, 20 μM tRNA substrate, and 1 μM of each enzyme
when present. Reactions were typically incubated for 20 min at
37 °C, then stopped by addition of an equal volume of phenol/
chloroform/isoamyl alcohol (25:24:1). The mixture was
vortexed for 30 s and layers were separated by centrifugation
at 14000g for 4 min. The top aqueous layer was added to a 1
mL column of Sephadex G-50 which had been drained by
centrifugation at 7000g for 1 min. After addition of the aqueous
layer to the column, the column was eluted by centrifugation as
before. The resulting purified RNA was then digested to
nucleosides with nuclease P1 and bacterial alkaline phosphatase
and analyzed by HPLC according to the procedure of Gerhke
et al.40
HPLC Assays for YwlC and YrdC Activity. Reaction
mixtures typically contained 50 mM MOPS pH 7.5, 20 mM
MgCl2, 25 mM KCl, 2 mM ATP, 10 mM L-threonine, and 20
mM NaHCO3 with varying amounts of YwlC or YrdC. After
incubating at 25 °C, the reaction mixture was analyzed by
HPLC on a Shimadzu HPLC with using a Supelco LC-18T
column (25 mm × 4.6 mm) and UV detection at 260 nm. For
analytical measurements of product concentrations, a phos-
phate buffer system was employed consisting of 20 mM
potassium phosphate pH 6 (Buffer A) and Buffer A with 20%
methanol (Buffer B). The gradient used was 100% A from 0 to
12 min followed by a linear gradient of 0−100% B from 12 to
30 min. Product concentrations were measured by peak
integration and compared to a standard curve of ATP or
AMP to convert to the number of moles. For kinetic assays,
product concentrations were measured during the linear part of
the progress curve (typically 30 s) during which less than 10%
product has formed. Plots of rate vs substrate concentration
were plotted and fit to Michaelis−Menten kinetics using
Kaleidagraph graphics software (Synergy). Each rate measure-
ment was the average of two or more measurements.
Preparative Scale Reactions for the Production of TC-
AMP. For the preparation of TC-AMP, 400 μL reaction
mixtures described above containing 90 μg of YwlC and 15
units of inorganic pyrophosphatase were incubated at 25 °C for
11 min and purified by a single injection using the HPLC
system described for the assays. Collected samples were
immediately frozen on dry ice and could be stored at −78
°C . When samples devoid of salt were required (e.g., for mass
spectrometry), repurification of carefully concentrated samples
was carried out by HPLC using volatile buffers. Buffer A
8952
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
contained 2 mM triethylammonium bicarbonate (TEAB), pH 7
while Buffer B consisted of Buffer A containing 20% methanol.
Typical yields of isolated TC-AMP were ca. 40 nmol for each
400 μL reaction.
NCA was prepared as described42 by refluxing L-Thr (250 mg,
1.42 mmol) and triphosgene (338 mg, 1.14 mmol) in THF
(12.5 mL) for 3 h and then precipitating the concentrated
residue with hexane. This procedure produced a 5:1 mixture of
the desired L-Thr-NCA tert-butyl ether (major) and L-Thr-ox
(minor) via loss of the tert-butyl ether. A portion of this
material (86 mg) was dissolved in 10% trifluoroacetic acid in
dichloromethane (3 mL) and stirred for 1 h at room
temperature to give a similar mixture of L-Thr-NCA and L-
Characterization of TCAMP by Mass Spectrometry.
Samples of TC-AMP were obtained as follows. Product
collections from HPLC purification of four separate 400 μL
reactions using TEAB buffers were pooled to give approximate
8 mL of 20 nmol of product. This solution was transferred to a
50 mL round-bottom flask on ice and solvent was removed by
lyophilization on ice over 4 h. Cold water (0.5 mL) was added
to the flask to dissolve the residue followed by another round of
lyophilization on ice. A final 100 μL of cold water was added
and the solution was stored at −78 °C for analysis by mass
spectrometry. A portion of the final sample was analyzed by
HPLC and found to be 80% TC-AMP with 20% AMP. This
material was diluted 20-fold with a 1:1 mixture of acetonitrile/
water and analyzed by mass spectrometry using a MaXim ultra
high resolution QTOF mass spectrometer in negative ion
mode.
1
Thr-ox that could be used as a standard. H NMR data for L-
Thr-NCA (DMSO-d6): 1.10 (d, 3H, J = 6.4 Hz), 3.93 (m, 1 H,
overlaps with L-Thr-ox signal), 4.29 (dd, 1H, J = 1.0, 2.0 Hz),
9.02 (bs, 1 H).
Measurement of TC-AMP stability. Reaction mixtures
(400 μL) containing 50 mM Tris pH 7.3, 20 mM MgCl2, 25
mM KCl, 1.5 mM ATP, 8 mM L-Thr, and 28 μg YwlC were
incubated at 25 °C for 20 min then injected onto HPLC using
the phosphate buffer system described above. TC-AMP was
collected in a typical volume of ca. 0.7 mL (ca. 5 μM) and was
immediately incubated in a water bath at the desired
temperature. The composition of the HPLC buffer eluent
containing the purified TC-AMP was 20 mM potassium
phosphate pH 6.5 containing approximately 20% methanol.
Aliquots (50 μL) of the incubation mixture were withdrawn at
various time points and frozen immediately on dry ice for later
quantitation by HPLC. Concentrations of TC-AMP and AMP
were measured by integration of peaks in the HPLC
chromatograms detected by absorbance at 260 nm and
calibrated using known AMP standards. The value of A260 for
TC-AMP solutions was found to change by less than 1% after
hydrolysis to AMP. Plots of ln [TC-AMP] vs time were linear,
and (pseudo) first order rate constants were obtained from the
slope of the linear fit. Subsequent stability studies used purified
and concentrated TC-AMP in which the pH was adjusted with
1 N KOH and/or buffer and/or enzyme added to the desired
concentrations. In these latter experiments, reaction mixtures
(final volume 150 μL) without added TC-AMP were
equilibrated to the required temperature in a circulating water
bath and decomposition reactions were initiated by addition of
8.5 μL of an 88 μM solution of TC-AMP. Aliquots of 25 μL
were removed and frozen as described above.
1H NMR Characterization of TC-AMP. For NMR
characterization, TC-AMP isolates (12 × 400 μL reactions)
from HPLC using the phosphate buffer system described above
were pooled (approximately 35 mL) in a 100 mL pear shaped
flask and concentrated by careful rotary evaporation using a
vacuum pump. The solution of TC-AMP was kept cold but not
frozen during the concentration. After removal of the solvent,
cold deuterium oxide (D2O, 4 mL) was added to dissolve the
resulting solid and the evaporation repeated. Finally the solid
was dissolved in 0.4 mL of cold D2O. The final solution
contained 0.4 mL of a 0.64 mM solution that also contained ca.
0.5 M KPi at pH 6. HPLC analysis showed this material to be
1
96% TC-AMP and 4% AMP. An initial H NMR spectrum
(128 transients) of this solution was obtained using a Varian
Inova-500 instrument at 500 MHz using a HCX probe running
at 10 °C. This material gave the following 1H NMR signals that
have been attributed to TC-AMP (H-4′ of the ribose is
obscured by the solvent peak): 1.072 (d, 3H, CH3, J = 6.5 Hz),
3.659 (d, 1H, H-β, J = 4.0 Hz), 4.191 (m, H-5′, 2H), 4.120 (m,
H-α, 1H), 4.334 (m, H-3′, 1H), 4.441 (app t, H-2′, 1H), 6.089
(d, 1H, H-1′, J = 6 Hz), 8.226 (s, 1H, H-2), 8.409 (s, 1H, H-8).
Characterization of TC-AMP Decomposition. The
NMR sample described above was allowed to warm to 25 °C
and the decomposition was monitored by periodic analysis by
Conversion of TC-AMP into t6A by YdiBCE. Reaction
mixtures contained 50 mM Tris pH 7.5, 20 mM MgCl2, 25 mM
KCl, 20 μM TC-AMP and 16 μM unmodified B. subtilis
tRNAThr transcript in a final volume of 30 μL. When included
in the reaction, ATP was 2 mM and YdiB, YdiC, and YdiE each
were 5 μM. Reactions were incubated at 37 °C for 20 min, then
stopped by addition of 1 volume of phenol:chloroform:isoamyl
alcohol (25:24:1) and vortexed for 30 s. RNA was isolated and
analyzed for the presence of t6A as described above for the t6A
HPLC assay.
1
both HPLC and H NMR. After 44 h, decomposition was
complete and final products were identified by the addition of
authentic samples. Separate solutions of L-threonine and (4S,
5R)-5-methyl-2-oxazolidinone-4-carboxylic acid (4 μL of 50
mM solutions in 0.5 M potassium phosphate pH 6 in D2O)
were added to the decomposed TC-AMP sample in the NMR
tube. After the addition of each authentic sample, a spectrum
was acquired for comparison with the original decomposed
sample. The rate of TC-AMP decomposition was measured by
integration of peaks for TC-AMP using HPLC conditions
described above.
Preparation of Putative TC-AMP Decomposition
Products. trans-(4S,5R)-5-Methyl-2-oxazoldinone-4 carboxylic
acid (L-Thr-ox) was prepared as previously described41 and the
1H NMR data (DMSO-d6) matched reported values:34 1.33 (d,
3H, J = 6.4 Hz), 3.92 (dd, 1 H, J = 0.8, 5.2 Hz), 4.53 (app
hextet, 1 H), 8.04 (bs, 1 H).
RESULTS
■
Cloning and Purification of B. subtilis Proteins
Required for t6A Biosynthesis. Cloning and expression of
ywlC and ydiB as the His6 N-fusion proteins were efficient, and
these proteins are highly expressed and soluble in E. coli
BL21(DE3). Unfortunately, expression of ydiC and ydiE gave
protein that was found largely in inclusion bodies. After much
effort, solubility problems were finally solved by the
construction of N-terminal thioredoxin (trxA) fusions, which
allowed good expression and sufficient solubility for in vitro
studies. Even with this tag, expression had to be performed at
L-Threonine N-carboxyanhydride (L-Thr-NCA) was pre-
pared in two steps. In step one, the t-butyl ether of L-Thr-
8953
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
Figure 1. HPLC analysis of t6A formation in vitro. (A) complete reaction mixture with unmodified tRNAThr substrate (B) authentic sample of 0.5
nmol synthetic t6A (D) coinjection of full reaction mixture with authentic t6A. (D) complete reaction mixture with A37C mutant tRNAThr. The peak
labeled “P” corresponds to a variable amount of phenol which passes through the G-50 spin column.
21 °C overnight for good yields and the YdiC fusion protein
could not be stored above 0.7 mg/mL without precipitation.
SDS gel analysis of the purified proteins loaded separately
compared to a mixture that was incubated in the t6A assay
buffer at 37 °C for 4 h (Figure S2) showed that the B. subtilis
enzymes appear to be stable in the reaction buffer for this
length of time. Attempts to purify the native forms of YdiC and
YdiE by specific cleavage of the fusion proteins with thrombin
or Factor Xa protease led to multiple products, perhaps because
the proteins are not completely stable in the cleavage reaction
conditions. El Yacoubi et al.18 found that E. coli ygjD mutants
could only be complemented by expression of B. subtilis ydiE in
combination with ydiC. However, attempts to express and/or
purify combinations of the proteins without fusions as a soluble
complex have been unsatisfactory thus far. Since the TrxA-
fusion proteins were found to be catalytically active and showed
the predicted RNA substrate specificity (vide infra) for t6A
formation, all experiments in this report are for the purified
fusion proteins.
Characterization of Requirements and RNA Substrate
Specificity for t6A Biosynthesis in vitro. With purified
enzymes in hand, reconstitution experiments were conducted
using unmodified B. subtilis tRNAThr as substrate. Reactions
included Mg-ATP, L-threonine, bicarbonate and Tris buffer in
addition to various combinations of proteins. Following
reaction, the RNA was isolated, then enzymatically digested
to nucleosides and fractionated on HPLC to identify t6A levels.
Figure 1 shows HPLC traces of the representative reaction
mixtures. The full reaction mixture (Figure 1A) shows a peak
present at 46 min in the chromatogram, which is the expected
location of t6A in results in approximately 90% modification of
the tRNA transcript in 20 min at 37 °C. The peak at 46 min in
the chromatogram was confirmed as t6A by coinjection (Figure
1B,C) with an authentic sample prepared synthetically.43 The
high level of activity is somewhat surprising since E. coli YrdC
was found to have poor affinity for unmodified tRNAThr
transcript.17 Similar to the recently reported data for t6A
synthesis in E. coli, omission of any one of the four Bacillus
proteins results in complete absence of t6A formation (Table
1). Modification was also found to require L-threonine, Mg-
ATP and NaHCO3, although more extreme measures
(degassing and addition of PEP carboxylase/PEP) were
required to reduce the bicarbonate/CO2 levels to show a
significant decrease in t6A modification activity.
A brief study of the RNA substrate specificity (Table 1)
shows that the expected substrates tRNAThr(GGU) and
tRNALys(UUU) are modified, whereas tRNAPhe(GAA) is not
modified. Mutation of the normally modified A37 of substrate
tRNAThr to a C leads to complete loss of activity (Figure 1D),
whereas a mutation of A36 to U in tRNAPhe(GAA) to give
tRNAPhe(GAU) leads to full activity as a substrate. Small stem
loop RNAs based on tRNALys or tRNAThr sequences showed no
activity in the in vitro assay even though YrdC has been shown
to bind tightly to these constructs.44 These results are
consistent with the work of both Ames et al.,45 who proposed
that a major substrate determinant is a U at the last codon
position 36 and Morin, et al.,46 who found in oocytes that in
addition to U36, the overall L-shaped architecture is required.
8954
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
Table 1. Requirements for in vitro t6A Formation in B.
subtilis
briefly as follows: YwlC catalyzes L-threonine-dependent
hydrolysis of ATP to give AMP. The previously documented
YdiB-catalyzed hydrolysis of ATP to give ADP is significantly
activated by addition of both YdiC and YdiE. When the full
reaction mixture containing all components was analyzed,
similar amounts of AMP and ADP were formed by the action of
YwlC and YdiBCE, respectively. However, neither of the
ATPase reactions appeared to be dependent on tRNA or on
each other. This made formulation of a coherent initial
biochemical model difficult.
YwlC Catalyzes the Formation and Release of L-
Threonylcarbamoyl-AMP (TC-AMP). To better quantify the
ATPase reactions catalyzed by YwlC and YdiBCE, HPLC
assays were developed. An analytical HPLC system was
employed that could separate ATP, AMP and ADP sufficiently
for quantitation using phosphate buffer with a methanol
gradient. Consistent with the TLC results, reactions of YwlC
with millimolar concentrations of Mg-ATP, bicarbonate and L-
threonine showed clear production of AMP; but reproducible
formation of an unidentified product with a longer retention
time was also observed. Figure 2A shows a representative
chromatogram of the YwlC reaction indicating an unidentified
peak at 21 min under the stated HPLC conditions. Formation
of this product was dependent on the presence of YwlC, L-
threonine and ATP, and its formation was proportional to the
amount of YwlC used. Initial observations suggested this
material was an adenylate species. First, when α-32P-labeled
ATP was used in the reaction mixture, the product collected at
21 min was radiolabeled with specific activity consistent with
the amount of UV absorbance (data not shown). Second, as
shown in Figure 2B, reinjection of isolated material that was
stored at room temperature for varying amounts of time
showed degradation to AMP.
reaction components or RNA substrate
t6A formation
a
complete reaction mixture
+++
(−)
(−)
+
−L-threonine
−ATP
−
−HCO3
−tRNA
−YwlC
−YdiB
−YdiC
(−)
(−)
(−)
(−)
(−)
+++
(−)
(−)
+++
(−)
(−)
−YdiE
tRNALys
tRNAThrA37C
tRNAPhe
tRNAPheA36U
tRNAThr ACSL
tRNALys ACSL
a
Complete reaction mixture contains Tris pH 7.2, 20 mM MgCl2, 25
mM KCl, 2 mM ATP, 20 mM NaHCO3, 20 μM tRNAThr, and 1 μM
each of YwlC, YdiB, YdiC and YdiE. ACSL, anticodon stem loop.
Thus, the in vitro reconstitution experiments contained the
minimum components required to mimic known in vivo
characteristics of t6A biosynthesis.
Characterization of the ATPase Activity of YwlC and
YdiBCE. Initial experiments used TLC analysis of reactions
containing α-32P-labeled ATP to characterize the ATPase
activity of the four individual B. subtilis proteins. These results
corroborate similar experiments recently reported by Deutsch,
et al.29 for the biosynthesis of t6A in E. coli. The results of
experiments with the B. subtilis enzymes can be summarized
Figure 2. (A) Typical HPLC trace of YwlC catalyzed formation of AMP and TC-AMP. (B) Example time course of decomposition of freshly
isolated TC-AMP to give AMP.
8955
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
Figure 3. Mass spectrum of purified TC-AMP. Peaks are visible that are consistent for monoanion (M−H) at m/z = 491, the dianion (M−2H) at m/
z = 245 and the dianion monosodium salt (M−2H Na+) at 513. Also visible is AMP (M−H) at m/z = 346 and a degradation product at m/z = 144.
1
Figure 4. 500 MHz H NMR spectrum of purified TC-AMP and its decomposition at 25 °C. The “t = 0” spectrum (A) is of the initial purified,
concentrated TC-AMP sample with resonances labeled that are consistent with the proposed structure. The t = 44 h spectrum (E) shows the final
decomposition products and AMP. The doublet labeled “I” at 1.1 ppm in spectrum C is a possible intermediate in the decomposition. Resonances
downfield are omitted for clarity.
8956
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
Figure 5. Identification of TC-AMP decomposition products by addition of known standards. (A) Final spectrum of decomposition of TC-AMP,
(B) addition of mol of L-Thr-ox, (C) addition of mol of L-Thr.
After scaling up reactions to isolate more material, further
characterization was achieved by mass spectrometry and H
AMP in ca. 0.5 M KPi, pH 6 that was approximately 90% pure.
The 500 MHz 1H NMR spectrum of TC-AMP at 10 °C shows
signals for both the threonine and AMP groups consistent with
the proposed structure. All peaks have been assigned and are
consistent with the TC-AMP structure except for the ribose 2′
proton, which is obscured by the solvent peak.
1
NMR. Figure 3 shows the MS data for purified desalted
material that was approximately 80% pure. Along with AMP
(M-H m/z = 346), a known degradation product, there is
observation of a negatively charged species with a measured
mass consistent with the molecular formula C15H21NO6
(theoretical 491.093316; found 491.09169; difference 3.3
ppm). This is the molecular formula for L-threonylcarbamoyl-
AMP (TC-AMP). In addition to the M-H peak at m/z = 491,
peaks are observed that correspond to the dianion M-2H as
well as the monosodium dianion M-2H+Na. Interestingly, the
mass spectrum also shows a relatively strong peak at m/z =
144.02975. This mass is consistent with a number of possible
decomposition products that lack AMP. These include L-
threonine isocyanate (L-Thr-NCO), the isomeric oxazolidi-
none, (4S,5R)-5-methyl-2-oxazoldinone-4-carboxylic acid (L-
Thr-ox), and L-threonine-N-carboxyanhydride (L-Thr NCA), all
of which have a calculated m/z = 144.02968. L-Threonine itself
was not observed in the mass spectrum under these conditions
perhaps because it exists mainly as the neutral zwitterion.
NMR Characterization of TC-AMP and Its Decom-
position Products. Final characterization of TC-AMP and its
degradation products by 1H NMR was achieved by purification
of the product from optimized large-scale enzymatic reactions.
The addition of inorganic pyrophosphatase to the YwlC
reaction mixtures resulted in a several-fold increase in the final
concentration of TC-AMP. The optimum YwlC concentration
was 5.75 μM and the minimum amount of pyrophosphatase
required was 15 units in a 400 μL reaction. Further increases in
the concentration of either enzyme did not increase the yield of
TC-AMP. Careful concentration of pooled collections from
HPLC purification gave 0.5 mL of a 640 μM solution of TC-
To identify the products of TC-AMP decomposition, the
NMR sample described above was allowed to decay at 25 °C.
Periodically, aliquots were removed for HPLC analysis and a
1H NMR spectrum was recorded. Representative spectra of the
decomposition are shown in Figure 4. The downfield region of
the spectrum, which includes peaks for the ribose H1′ and
adenine H2 and H8 protons are omitted for clarity. The spectra
are consistent with the formation of three final products: AMP
and two additional threonine-like products. The final
decomposition products were identified by doping the NMR
solution with authentic samples of known compounds. As
mentioned above, the MS spectra of a partially degraded TC-
AMP solution showed a peak at m/z = 144 suggesting the
known oxazolidinone L-Thr-ox as a possible breakdown
product. This compound was synthesized,41 purified and
characterized, and added to the degraded TC-AMP NMR
sample. Figure 5A shows a spectrum of the final decomposition
mixture, while Figure 5B shows a spectrum of the same mixture
after the addition of L-Thr-ox, showing matched and increased
signals of one of the major decomposition products. In a similar
manner, the other product was identified as L-threonine (Figure
5C). The signals downfield in the spectrum matched an
authentic sample of AMP (not shown), which is consistent with
the identification of AMP by HPLC.
An additional species in the decomposition mixture,
identified by a doublet at ca. 1.1 ppm (indicated by I in Figure
4C), is visible from the start of the decomposition and present
8957
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
at a nearly constant ratio to TC-AMP throughout the reaction
before decaying completely. It is possible that this species is an
intermediate in the decomposition. Possible candidates for I are
the isocyanate L-Thr-NCO, and the N-carboxyanhydride, L-
Thr-NCA. The isocyanate has not been described to date. L-
Thr-NCA was prepared using a literature method.42 In DMSO,
L-Thr-NCA is relatively stable, decaying directly to the
oxazolidinone with a half-life of about 2 days. However it was
found to decompose rapidly (t1/2 ca. 5 min) in 0.5 M KPi pH 6
to give an approximately 60:40 mixture of the L-Thr-ox and L-
threonine. Because of the short half-life, it would not be
expected to accumulate in the reaction. Moreover, the 1H NMR
spectrum of the NCA in 0.5 M KPi does not appear to match
that of I (data not shown). It is possible that I is L-Thr-NCO,
but this has yet to be confirmed. Together the data suggest that
TC-AMP in phosphate buffer at pH 6 degrades to a mixture of
AMP, L-threonine, and the oxazolidinone, which are formed
directly or through one or more intermediates.
Figure 6. AMP formation during synthesis of TC-AMP by YwlC and
YrdC. (A) YwlC reaction (t = 2 min) (B) YrdC reaction (t = 15 s) (C)
YwlC reaction (t = 2 min) in the presence of inorganic
pyrophosphatase. Concentrations of the other substrates were
saturating.
Kinetic Stability of TC-AMP. The rate of decomposition of
TC-AMP to AMP was initially measured directly after HPLC
purification from scaled up reaction mixtures. Eluted TC-AMP
was incubated at the desired temperature and aliquots taken at
specific time points were frozen and subsequently analyzed by
HPLC (cf., Figure 2B). The HPLC data fit well to (pseudo)
first order kinetics. In subsequent experiments, purified
concentrated TC-AMP solution was added to temperature
equilibrated reaction mixtures under a variety of conditions.
Data and reaction conditions are summarized in Table 2, in
because of the expected formation of an activated bicarbonate
intermediate, such as carboxy-AMP, which would be easily
hydrolyzed if released or exposed to water in the absence of the
other t6A enzymes. However, the reaction catalyzed by YrdC
was found to form almost no AMP at short reaction times
(Figure 6B). Moreover, addition of inorganic pyrophosphatase
to the YwlC reactions drastically reduces AMP production to
significantly below one molar equivalent relative to TC-AMP at
reaction times of a few minutes or less (Figure 6C).
A plot of the progress curves for YwlC formation of TC-
AMP in the absence and presence of inorganic pyrophospha-
tase illustrates the magnitude of the effect further. In the
absence of pyrophosphatase, the production of TC-AMP levels
off while AMP continues to be produced at an approximately
linear rate (Figure 7A). Addition of pyrophosphatase (Figure
7B) inhibits AMP production to levels below that of TC-AMP
for a few minutes, upon which TC-AMP production begins to
level off and AMP rises. The reason for the dependence of
AMP formation on pyrophosphate concentration is not clear at
present.
However, the observation of substoichiometric amounts of
AMP produced in the reaction by both YrdC and YwlC
suggests that the formation of TC-AMP does not require an
ATP-activated bicarbonate intermediate. This implies that the
enzymes utilize ATP only for the formation of TC-AMP from
L-threonine carbamate, the latter of which can be formed
directly from the amino acid and CO2. The very low levels of
AMP produced by YrdC at short reaction times allowed an
estimation of the kinetic parameters for ATP as the variable
substrate. With concentrations of L-Thr at 10 mM and
bicarbonate at 20 mM, values of Km = 93 15 μM and kcat
= 0.10 0.02 s−1 for ATP were found.
YwlC and YrdC Catalyze Efficient Formation of ATP
from TC-AMP and PPi. Purification of TC-AMP allowed the
study of the reverse reaction with PPi to form ATP. Figure 8
shows HPLC analysis of reaction mixtures containing TC-
AMP, PPi, MgCl2 containing buffer and either YrdC or YwlC.
Compared with the control reaction without enzyme, both
YwlC and YrdC catalyze formation of ATP with no hydrolysis
to AMP above background levels. This was surprising, since
pyrophosphate is clearly shown to promote formation of AMP
by YwlC in the presence of saturating ATP and L-threonine. In
addition, the concentration of YrdC used in the reactions
shown in Figure is over 600-fold less than that for YwlC,
indicating a relatively large value of kcat for YrdC in the reverse
Table 2. Conditions and Reported Half-Life for
Decomposition of TC-AMP
e
buffer
MgCl2 (mM)
pH
temp (°C)
t1/2
a
0.5 M KPi
0
0
6.0
6.5
6.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
25
25
37
37
25
25
25
37
37
37
8.5 h
b
b
b
20 mM KPi
20 mM KPi
20 mM KPi
4.0 h
0
1.16 h
0
16.5 min
11.8 min
14.4 min
19.3 min
3.5 min
1.7 min
21 s
c
50 mM MOPS
20
20
20
2
d
MOPS/YwlC
d
MOPS/YrdC
50 mM MOPS
50 mM MOPS
50 mM MOPS
5
20
a
b
In D2O, single measurement. Contains ca. 20% methanol.
c
d
Reactions with MOPS contained 25 mM KCl. 0.5 μM enzyme.
e
Average of at least two experiments.
which the rates are given as the calculated half-life. Not
surprisingly, TC-AMP is destabilized by increases in pH,
temperature or Mg2+ concentration. Incubation of TC-AMP
with either YwlC or YrdC resulted in no increase in the rate of
decomposition, and in fact showed a small but statistically
significant protective effect that could be due to sequestration
and/or binding in a stable conformation. Thus, alone these
enzymes do not appear to catalyze further transformation of
TC-AMP into a more reactive product, such as the isocyanate.
Production of TC-AMP By Both YwlC and E. coli YrdC
Does Not Proceed via An Activated Bicarbonate
Intermediate. The progress curves of TC-AMP formation
by YwlC and its E. coli ortholog YrdC were studied. As shown
in Figure 6A, the reaction catalyzed by YwlC produces a
significant molar excess of AMP relative to TC-AMP, even at
reaction times as short as 20 s. At first, this was not surprising
8958
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
Figure 7. Progress curve plots for YwlC catalyzed formation of TC-AMP in the (A) absence of inorganic pyrophosphatase and (B) presence of
■
●
pyrophosphatase. For both plots the number of pmol of AMP ( ) and TC-AMP ( ) are shown vs time.
Table 3. Requirements for the Synthesis of t6A from TC-
AMP
a
reaction components
t6A formation (pmol)
YdiBCE + ATP + TC-AMP + tRNAThr (complete)
276 88
353 58
complete − ATP
complete − TC-AMP
complete − YdiB
b
nd
nd
nd
nd
complete − YdiC
complete − YdiE
a
b
15 min reaction at 37 °C. nd, none detected.
consistent with the overall requirements in the initial
characterization experiments. Interestingly, ATP is not required
for this step of the reaction, and in fact, seems to suppress the
modification somewhat, although the data are not statistically
significant. HPLC data confirm earlier TLC analysis that
showed ATP hydrolysis to ADP occurs at a roughly similar rate
to AMP production during t6A formation in the full reaction
mixture (Figure 9). An estimate of the turnover rate for
YdiBCE-catalyzed conversion of TC-AMP to give t6A gave a
value of 0.19 0.06 min−1 using 10 μM TC-AMP. This can be
Figure 8. Formation of ATP from TC-AMP catalyzed by YwlC and
YrdC. Reactions (50 μL) contained 5 μM TC-AMP and 1 mM PPi in
buffer with either (A) 0.11 pmol YrdC; (B) 76 pmol YwlC; or (C) no
enzyme. Reactions were incubated for 2 min at 25 °C and analyzed by
HPLC.
compared to the measured rate of 0.40
overall reaction starting with L-Thr, bicarbonate and ATP. The
observed slightly slower turnover could be evidence for
0.1 min−1 for the
direction. With TC-AMP as the variable substrate and 1 mM
PPi with 30 s reaction times, the following kinetic constants
were measured for YrdC: kcat = 20 2 s−1; Km = 0.68 0.15
μM; kcat/Km = 3.3 1.1 × 107 M−1 s−1. The apparent low value
for Km for TC-AMP prevented accurate measurement of kinetic
constants for YwlC.
YdiBCE Catalyzes Formation of t6A from TC-AMP in
the Absence of Both YwlC and ATP. To answer the
question of whether TC-AMP is a true productive intermediate
for t6A biosynthesis, we incubated this product with the
YdiBCE complex and unmodified tRNAThr. Reaction of 12.5
μM tRNAThr with 20 μM purified TC-AMP and 5 μM YdiBCE
for 20 min at 25 °C gave efficient t6A formation as measured by
HPLC analysis of the recovered and digested tRNA. Since these
reactions did not contain L-threonine, this experiment is
consistent with the assigned structure of TC-AMP and also
supports the hypothesis that it is a productive intermediate in
the biosynthetic pathway. As shown in Table 3, formation of
t6A required all three proteins of the YdiBCE complex,
Figure 9. HPLC analysis of TC-AMP production in the full reaction
mixture with and without substrate tRNA. (A) Full reaction
components with tRNAThrA37C mutant (B) same as A, but with
tRNAThr wt substrate.
8959
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
Scheme 2. Decomposition Pathways for TC-AMP in Aqueous Media
channeling, although the data must be interpreted as
preliminary because of the large fusion constructs used in
these experiments.
10−5 is found in favor of ATP formation. This corresponds to a
value of ΔG = −25.6 kJ/mol. By adding to this the reported
updated value of ΔG′° = −45.6 kJ/mol for the hydrolysis of
ATP to give AMP and PPi,48 an estimated value of ΔG′° for
TC-AMP hydrolysis to AMP of −71 kJ/mol is obtained. This
number is quite large, but in the correct relative range,
considering the corresponding values for ΔG′° for acetyl
phosphate (−43 kJ/mol), acetyl adenylate (−55 kJ/mol) and
carbamoyl phosphate (−51 kJ/mol).49
NMR analysis of the decomposition of TC-AMP showed
that in phosphate buffer at pH 6, the final decomposition
products are L-threonine and the oxazolidinone (L-Thr-ox),
which are formed in approximately equal amounts. A third
species observed in the NMR spectra at 5−10% of TC-AMP
could be an intermediate in the degradation pathway, such as
the isocyanate (L-Thr-NCO). As shown in Scheme 2, both
products as well as the putative intermediates can be formed
directly from TC-AMP, via either an intramolecular addition or
elimination reaction or by hydrolysis. The isocyanate and the
NCA would each be predicted to give both the oxazolidinone
and L-threonine. This has been confirmed for the NCA, which
interestingly degrades in phosphate buffer rapidly to give nearly
the same ratio of the two products as in the NMR study. Thus,
it is not expected to accumulate during the NMR experiment.
The rearrangement of the NCA to the oxazolidinone has been
previously documented under anhydrous basic conditions.51
However, basic conditions do not appear to be required since
this rearrangement occurs in DMSO, although slowly.
Analysis of the complete reaction mixture by HPLC (Figure
8) shows a marked effect of the presence of tRNA substrate on
the appearance of TC-AMP in the reaction mixture. In the
absence of tRNA (or in the presence of a tRNAThr A37C
mutant), there is a significant buildup and release of the
intermediate (Figure 8A). Presence of substrate tRNA in the
reaction suppresses this buildup and gives a noticeable increase
in AMP production (Figure 8B). This is further evidence that
TC-AMP is an intermediate in the modification reaction.
DISCUSSION
■
The experiments described in this work focus on both the
chemical and biochemical characterization of L-threonylcarba-
moyl-AMP (TC-AMP), an intermediate in the biosynthesis of
t6A. Evidence has been presented in this work for the role of
YwlC/TsaC (and its E. coli ortholog YrdC) as a TC-AMP
synthetase. TC-AMP has been isolated, fully characterized and
its stability measured under physiological conditions. The
measured half-life for TC-AMP of 3.5 min in phosphate buffer
at 37 °C, pH 7.5, 2 mM MgCl2 is in the lower range of kinetic
stability for other biologically relevant phosphoric anhydrides.
For example, carbamoyl phosphate has a half-life of 45 min at
pH 7.8 at 37 °C.47 Since this measurement is in the absence of
Mg2+, this value should be compared to a half-life of 16.5 min
for TC-AMP under similar conditions (Table 2). The relatively
faster degradation of TC-AMP is not surprising, given its
multiple possible intramolecular degradation pathways, and
greater free energy (ΔG′°) of hydrolysis. Acyl adenylates are
known to have greater ΔG′° values for hydrolysis compared
with the corresponding acyl phosphates.48,49
Support for the isocyanate as an intermediate comes from
the reported decomposition of carbamoyl phosphate that
involves unimolecular elimination of phosphate from the
dianionic species to give isocyanate.47 Although TC-AMP is a
phosphodiester and hence only a monoanion at pH > 2, one
can postulate a similar unimolecular elimination pathway for
TC-AMP at pH > 4, with the ionized carboxylate acting as base
to remove the proton on the carbamate nitrogen to eliminate
AMP. Alternatively, enough inorganic phosphate dianion in the
reaction may exist in 0.5 M KPi at pH 6 to promote the
elimination.
By measuring the values of kcat/Km for both the formation of
TC-AMP and its reverse reaction to give ATP, it is possible to
use the Haldane relation (eq 1):
(kcat/Km)fwd
(kcat/Km)rev
Keq
=
(1)
The results of the kinetics and stability studies of TC-AMP
have implications for the mechanistic role of YwlC and YrdC.
Unlike proposed mechanisms to date, the data presented in this
work do not support a mechanism in which t6A biosynthesis
to estimate the equilibrium constant for this reaction.50 Using
these values (kcat/Km forward = 1.1 × 103 M−1 s−1; kcat/Km
reverse = 3.3 × 107 M−1 s−1), an estimated value for Keq = 3.3 ×
8960
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
proceeds via the formation of an ATP-activated bicarbonate
species. Instead, as shown in Scheme 3, these enzymes most
higher eukaryotes. It is possible that this extra domain contains
an additional binding site for ATP and/or PPi that influences
the kinetics of the forward and reverse reactions. It is clear from
the effect of pyrophosphatase that PPi causes excess AMP
formation in the forward reaction in the presence of ATP, but
not in the reverse reaction. It is presently unclear if the source
of AMP in the forward reaction is via hydrolysis of ATP directly
or via TC-AMP. It is conceivable that the YwlC-TC-AMP
complex may actively hydrolyze ATP to AMP via the C-
terminal domain. Such reactivity has been observed in firefly
luciferase, in which the bound luciferase-luciferyl-AMP complex
appears to hydrolyze ATP to give AMP and PPi.54 In addition,
it was found that both YwlC and YrdC contain anywhere from
1 to 5 mol % of bound TC-AMP after isolation by affinity
chromatography and gel filtration. This is consistent with the
recent reassessment of the crystal structure of S. tadaii SUA5 by
Parthier et al.33 in which significant electron density was found
for bound TC-AMP, which they suspect remained bound
throughout the protein isolation. Interestingly, whereas purified
YrdC also contained bound AMP, YwlC contained only bound
ATP, which is again indicative of possible mechanistic
differences.
Scheme 3. Proposed Mechanism for t6A Biosynthesis in
Bacteria
likely directly catalyze the formation of N-carboxy-L-threonine,
by shifting the equilibrium to favor this product from carbon
dioxide and the amino acid. It is well documented that aliphatic
amines, including amino acids such as threonine, form stable
carbamates at high concentrations of CO2/bicarbonate and
amino acid at alkaline pH (∼8.5−12).52 Thus, the enzyme’s
role in forming N-carboxy-L-threonine (L-threonine carbamate)
is essentially that of (a) approximation to increase the local
concentration of reactants on the enzyme surface, and (b)
increasing the effective pH in the active site. The latter is
accomplished by positioning functional groups to deprotonate
the L-threonine ammonium group or selectively bind the
neutral amine for attack on CO2, deprotonate the same
nitrogen again after attack, and position group(s) to stabilize
the resulting carbamate anion. The source of the carbamate
carbon would then most likely be from the more electrophilic
CO2 rather than bicarbonate.
Once formed, the carbamate could be positioned to form the
adenylate by attack on the alpha phosphate of ATP to give the
product TC-AMP, as hypothesized in recent structural work.31
This mechanism is consistent with recent crystal structure data
for the carbamoylation enzymes HypF and TobZ, which also
catalyze formation of carbamoyl adenylate intermediates. For
both of these enzymes, the proposed substrate is the anion of
carbamic acid, formed by energetically favorable hydrolysis of
carbamoyl phosphate, which then attacks ATP to give the
analogous adenylate species. Unlike this pathway for generating
the carbamate, the YwlC reaction to form TC-AMP is quite
energetically unfavorable. It is similar in some respects to the
formation of aminoacyl adenylates by aminoacyl-tRNA
synthetases, in which the Keq for formation of the adenylate
is 10−7.53 For both reactions, formation of the final products
(t6A and aminoacyl-tRNA) give release of AMP which renders
the overall reaction favorable.
The final step in t6A formation is the addition of TC-AMP to
A37 of substrate tRNAs to give t6A. Previous work suggested
that YrdC binds RNA35 and exerts the binding specificity
required of a protein that modifies tRNA directly.45 Structural
work with YrdC homologues31−33 suggested that this domain is
involved in the activation of carbamate by ATP. The in vitro
experiments presented here show that in B. subtilis, the reaction
of tRNA with TC-AMP can be catalyzed by the YdiBCE
(TsaEBD) complex in the absence of YwlC. Consistent with
the overall reaction, each of the YdiBCE proteins is required
the final step. This result suggests that one or more of the
YdiBCE proteins also binds tRNA. In addition, ATP was not
required, which was surprising because of the YdiBCE complex-
dependent activation in the production of ADP in the complete
reaction mixture. While it is clear that ATP would not be
required for the chemistry of addition of TC-AMP to tRNA to
give t6A, in vivo data has shown that the ATPase activity of YdiB
is required for normal growth.37 The exact role for YdiB(TsaE)
and YdiC(TsaB) are not yet known. It is conceivable that ATP
hydrolysis by YdiBCE may be necessary for the binding of
tRNA in a productive manner or for chaperone activity in vivo
as has been proposed previously. Alternatively, ATP hydrolysis
by YdiBCE may serve an additional role, in a capacity that may
or may not be directly related to the chemistry of t6A
formation. Future work on the kinetics of t6A formation and
RNA binding may elucidate the role of the prokaryotic specific
proteins.
Finally, these enzymes have been proposed as potential
antibiotic targets. Elucidation of the structure and reactivity of
the intermediate provides an initial framework for the rational
design of inhibitors that may be optimized into specific
antibiotic drugs. While the overall t6A modification is required
for all forms of life, there appears to be sufficient differentiation
not only between bacteria and eukaryotes but even within
bacteria to enable the development of specific inhibitors.
Further structural and mechanistic work on these systems will
be necessary to provide such targeted therapeutics.
While the Bacillus results match the recent work in E. coli29 in
the overall requirements for TC-AMP and t6A formation, there
are significant differences between YwlC and YrdC in both the
rate of AMP formation in the forward reaction and the kcat for
ATP formation in the reverse reaction. These differences
between YwlC and YrdC may in part be because YwlC and
many other SUA5 homologues contain an additional “SUA5”
domain (pfam 03481) of about 100 aa at the C-terminus. This
domain is absent in E. coli YrdC and in other organisms such as
8961
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
(10) Na, J. G., Pinto, I., and Hampsey, M. (1992) Isolation and
characterization of SUA5, a novel gene required for normal growth in
Saccharomyces cerevisiae. Genetics 131, 791−801.
(11) Lin, C. A., Ellis, S. R., and True, H. L. (2010) The Sua5 protein
is essential for normal translational regulation in yeast. Mol. Cell. Biol.
30, 354−363.
(12) Chheda, G., Hong, C., Piskorz, C., and Harmon, G. (1972)
Biosynthesis of N-(purin-6-ylcarbamoyl)-L-threonine riboside. Incor-
poration of L-threonine in vivo into modified nucleoside of transfer
ribonucleic acid. Biochem. J. 127, 515−519.
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and figures for B. subtilis ywlC mutant
tRNA analysis, protein expression, and TLC analysis of ATPase
reactions. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*Telephone, (608)-262-3083; fax, (608)-262-3397; email,
clauhon@wisc.edu.
Funding
Funding from the University of Wisconsin School of Pharmacy
is gratefully acknowledged.
Notes
■
(13) Powers, D. M., and Peterkofsky, A. (1972) Biosynthesis and
specific labeling of N-(purin-6-ylcarbamoyl)threonine of Escherichia
coli transfer RNA. Biochem. Biophys. Res. Commun. 46, 831−838.
(14) Powers, D. M., and Peterkofsky, A. (1972) The presence of N-
(purin-6-ylcarbamoyl)threonine in transfer ribonucleic acid species
whose codons begin with adenine. J. Biol. Chem. 247, 6394−6401.
(15) Korner, A., and Soll, D. (1974) N-(purin-6-ylcarbamoyl)-
̈
̈
threonine: biosynthesis in vitro in transfer RNA by an enzyme purified
from Escherichia coli. FEBS Lett. 39, 301−306.
The authors declare no competing financial interest.
(16) Elkins, B. N, and Keller, E. B. (1974) The enzymatic synthesis of
N-(purin-6-ylcarbamoyl)threonine, an anticodon-adjacent base in
transfer ribonucleic acid. Biochemistry 13, 4622−4629.
ACKNOWLEDGMENTS
The author thanks Brett J. Kopina for technical assistance with
acquiring NMR spectra.
■
(17) El Yacoubi, B., Lyons, B., Cruz, Y., Reddy, R., Nordin, B.,
Agnelli, F., Williamson, J. R., Schimmel, P., Swairjo, M. A., and de
ABBREVIATIONS USED
■
́
Crecy-Lagard, V. (2009) The universal Sua5/YciO/YrdC family is
PPi, pyrophosphate; TC-AMP, L-threonylcarbamoyl adenosine-
5′-monophosphate; L-Thr, L-threonine; L-Thr-ox, trans-
(4S,5R)-5-methyl-2-oxazoldinone-4 carboxylic acid; L-Thr-
NCO, L-threonine isocyanate; NCA, N-carboxyanhydride
required for the formation of threonylcarbamoyladenosine in tRNA.
Nucleic Acids Res. 37, 2894−2909.
(18) El Yacoubi, B., Hatin, I., Deutsch, C., Kahveci, T., Rousset, J. P.,
́
Iwata-Reuyl, D., Murzin, A. G., and de Crecy-Lagard, V. (2011) A role
for the universal Kae1/Qri7/YgjD (COG0533) family in tRNA
modification. EMBO J. 30, 882−893.
(19) Srinivasan, M., Mehta, P., Yu, Y., Prugar, E., Koonin, E., Karzai,
A., and Sternglanz, R. (2011) The highly conserved KEOPS/EKC
complex is essential for a universal tRNA modification, t6A. EMBO J.
30, 873−881.
(20) Daugeron, M. C., Lenstra, T. L., Frizzarin, M., El Yacoubi, B.,
Liu, X., Baudin-Baillieu, A., Lijnzaad, P., Decourty, L., Saveanu, C.,
REFERENCES
■
(1) Gustilo, E. M., Fendeix, F. A. P., and Agris, P. F. (2008) tRNA’s
modifications bring order to gene expression. Curr. Opin. Microbiol. 11,
134−140.
(2) Bjork, G. R. 1995. Biosynthesis and function of modified
̈
nucleosides. In tRNA: Structure, Biosynthesis, and Function (Soll, D.,
̈
and RajBhandary, U., Eds.) pp 165−205, American Society for
Microbiology, Washington, DC.
́
Jacquier, A., Holstege, F. C., de Crecy-Lagard, V., van Tilbeurgh, H.,
and Libri, D. (2011) Gcn4 misregulation reveals a direct role for the
evolutionary conserved EKC/KEOPS in the t6A modification of
tRNAs. Nucleic Acids Res. 39, 6148−6160.
(21) Hecker, A., Leulliot, N., Gadelle, D., Graille, M., Justome, A.,
Dorlet, P., Brochier, C., Quevillon-Cheruel, S., Le Cam, E., van
Tilbeurgh, H., and Forterre, P. (2007) An archaeal orthologue of the
universal protein Kae1 is an iron metalloprotein which exhibits atypical
DNA binding properties and apurinic-endonuclease activity in vitro.
Nucleic Acids Res. 35, 6042−6051.
(22) Downey, M., Houlsworth, R., Maringele, L, Rollie, A., Brehme,
M., Galicia, S., Guillard, S., Partington, M., Zubko, M. K., Krogan, N. J.,
Emili, A., Greenblatt, J. F., Harrington, L., Lydall, D., and Durocher, D.
(2006) A genome-wide screen identifies the evolutionarily conserved
KEOPS complex as a telomere regulator. Cell 124, 1155−1168.
(23) Meng, F. L., Hu, Y., Shen, N., Tong, X. J., Wang, J., Ding, J., and
Zhou, J. Q. (2009) Sua5p a single-stranded telomeric DNA-binding
protein facilitates telomere replication. EMBO J. 28, 1466−1478.
(24) Kisseleva-Romanova, E., Lopreiato, R., Baudin-Baillieu, A.,
Rousselle, J. C., Ilan, L., Hofmann, K., Namane, A., Mann, C., and
Libri, D. (2006) Yeast homolog of a cancer-testis antigen defines a
new transcription complex. EMBO J. 25, 3576−3585.
(25) Oberto, J ., Breuil, N., Hecker, A., Farina, F., Brochier-Armanet,
C., Culetto, E., and Forterre, P. (2009) Qri7/OSGEPL, the
mitochondrial version of the universal Kae1/YgjD protein, is essential
for mitochondrial genome maintenance. Nucleic Acids Res. 37, 5343−
5352.
(3) Schweizer, M. P., Chheda, G. B., Baczynskyj, L., and Hall, R. H.
(1969) Aminoacyl nucleosides. VII. N-(Purin-6-ylcarbamoyl)-
threonine. A new component of transfer ribonucleic acid. Biochemistry
8, 3283−3289.
(4) Miller, J. P., Hussain, Z., and Schweizer, M. P. (1976) The
involvement of the anticodon adjacent modified nucleoside N-(9-
(beta-D-ribofuranosyl)purine-6-ylcarbamoyl)-threonine in the biolog-
ical function of E. coli tRNAIle. Nucleic Acids Res. 3, 1185−1201.
(5) Weissenbach, J., and Grosjean, H. (1981) Effect of
threonylcarbamoyl modification (t6A) in yeast tRNA Arg III on
codon-anticodon and anticodon-anticodon interactions. A thermody-
namic and kinetic evaluation. Eur. J. Biochem. 116, 207−213.
(6) Murphy, F. V., Ramakrishnan, V., Malkiewicz, A., and Agris, P. F.
(2004) The role of modifications in codon discrimination by
tRNALysUUU. Nat. Struct. Mol. Biol. 11, 1186−1191.
(7) Durant, P. C., Bajji, A. C., Sundaram, M., Kumar, R. K., and
Davis, D. R. (2005) Structural Effects of Hypermodified Nucleosides
in the Escherichia coli and Human tRNALys Anticodon Loop: The
Effect of Nucleosides s2U, mcm5U, mcm5s2U, mnm5s2U, t6A, and
ms2t6A. Biochemistry 44, 8078−8089.
(8) Stuart, J., Gdaniec, Z., Guenther, R., Marszalek, M., Sochacka, E.,
Malkiewicz, A., and Agris, P. (2000) Functional anticodon architecture
of human tRNALys3 includes disruption of intraloop hydrogen
bonding by the naturally occurring amino acid modification, t6A.
Biochemistry 39, 13396−13404.
(9) Lescrinier, E., Nauwelaerts, K., Zanier, K., Poesen, K., Sattler, M.,
and Herdewijn, P. (2006) The naturally occurring N6-threonyl
adenine in anticodon loop of Schizosaccharomyces pombe tRNAi
causes formation of a unique U-turn motif. Nucleic Acids Res. 34,
2878−2886.
(26) Handford, J. I., Ize, B., Buchanan, G., Butland, G. P., Greenblatt,
J., Emili, A., and Palmer, T. (2009) Conserved network of proteins
essential for bacterial viability. J. Bacteriol. 191, 4732−4749.
8962
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963

Biochemistry
Article
(27) Allali-Hassani, A., Campbell, T. L., Ho, A., Schertzer, J. W., and
Brown, E. D. (2004) Probing the active site of YjeE: a vital Escherichia
coli protein of unknown function. Biochem. J. 384, 577−584.
(46) Morin, A., Auxilien, S., Senger, B., Tewari, R., and Grosjean, H.
(1998) Structural requirements for enzymatic formation of threo-
nylcarbamoyladenosine (t6A) in tRNA: an in vivo study with Xenopus
laevis oocytes. RNA 4, 24−37.
́
(28) Gerdes, S. Y., Scholle, M. D., Campbell, J. W., Balazsi, G.,
(47) Allen, C. M., and Jones, M. E. (1964) Decomposition of
Carbamyl Phosphate in Aqueous Solution. Biochemistry 3, 1238−1247.
(48) Frey, P. A., and Arabshahi, A. (1975) Standard Free Energy
Change for the Hydrolysis of the α,β-Phosphoanhydride Bridge of
ATP. Biochemistry 34, 11307−11310.
(49) Jencks, W. P. (1976) Free energies of hydrolysis and
decarboxylation, in Handbook of Biochemistry and Molecular Biology,
3rd ed. (Fasman, G. D., Ed.) Vol. 1, CRC Press, Cleveland.
(50) Haldane, J. B. S. (1930) Enzymes, p 80, Longmans, Green and
Co., London.
(51) Saito, T. (1964) A new rearrangement reaction of β-Hydroxy-α-
Amino acid N-carboxyanhydrides to 2-oxazolidinone derivatives. Bull.
Chem. Soc. Jpn. 37, 624−628.
(52) Yamamoto, Y., Hasegawa, J., and Ito, Y. (2012) Kinetic
investigation on carbamate formation from the reaction of carbon
dioxide with amino acids in homogeneous aqueous solution. J. Phys.
Org. Chem. 25, 239−247.
Ravasz, E., Daugherty, M. D., Somera, A. L., Kyrpides, N. C.,
Anderson, I., Gelfand, M. S., Bhattacharya, A., Kapatral, V., D’Souza,
M., Baev, M. V., Grechkin, Y., Mseeh, F., Fonstein, M. Y., Overbeek, E.
́
R., Barabasi, A.-L., Oltvai, Z. N., and Osterman, A. L. (2003)
Experimental determination and system level analysis of essential
genes in Escherichia coli MG1655. J. Bacteriol. 185, 5673−5684.
(29) Deutsch, C., El Yacoubi, B., de Crecy-Lagard, V., Iwata-Reuyl, D.
(2012) The biosynthesis of threonylcarbamoyl adenosine (t6A), a
universal tRNA nucleoside. J. Biol. Chem., 287, 13666−13673.
(30) Kuratani, M., Kasai, T., Akasaka, R., Higashijima, K., Terada, T.,
Kigawa, T., Shinkai, A., Besshow, Y., and Yokoyama, S. (2011) Crystal
Structure of Sulfolobus tadaii SUA5 Complexed with L-Threonine and
AMPPNP. Proteins 79, 2065−2075.
(31) Petkun, S., Shi, R., Li, Y., Asinas, A., Munger, C., Zhang, L.,
Waclawek, M., Soboh, B., Sawers, R. G., and Cygler, M. (2011)
Structure of hydrogenase maturation protein HypF with intermediates
shows two active sites. Structure 19, 1773−1783.
(32) Shomura, Y., and Higuchi, Y. (2012) Structural Basis for the
Reaction Mechanism of S-Carbamoylation of HypE by HypF in the
Maturation of [Ni-Fe] Hydrogenases. J. Biol. Chem. 287, 28409−
28419.
(33) Parthier, C., Gorlich, S., Jaenecke, F., Breithaupt, C., Brauer, U.,
Fandrich, U., Clausnitzer, D., Wehmeier, U. F., Bottcher, C., Scheel,
D., and Stubbs, M. T. (2012) The O-carbamoyltransferase TobZ
catalyzes an ancient enzymatic reaction. Angew. Chem., Int. Ed. 57, 1−7.
(34) Teplova, M., Tereshko, V., Sanishvilli, R., Joachimiak, A.,
Bushueva, T., Anderson, W. F., and Egli, M. (2000) The structure of
the YrdC gene product reveals a new fold and suggests a role in RNA
binding. Protein Sci. 9, 2557−2566.
(53) Wells, T. N. C., and Fersht, A. R. (1986) Use of binding energy
in catalysis analyzed by mutagenesis of the tyrosyl tRNA synthetase.
Biochemistry 25, 1881−1886.
(54) Rhodes, W. C., and McElroy, W. D. (1958) The synthesis and
function of luciferyl adenylate and oxy-luciferyl adenylate. J. Biol. Chem.
233, 1528−1537.
(35) Kobayashi, K., Ehrlich, S. D., Albertini, A., et al. (2003) Essential
Bacillus subtilis genes. Proc. Natl. Acad. Sci. U. S. A. 100, 4678−4683.
(36) Hunt, A., Rawlins, J. P., Thomaides, H. B., and Errington, J.
(2006) Functional analysis of 11 putative essential genes in Bacillus
subtilis. Microbiology 152, 2895−2907.
(37) Karst, J. C., Foucher, A. E., Campbell, T. L., Di Guilmi, A. M.,
Stroebel, D., Mangat, C. S., Brown, E. D., and Jault, J. M. (2009) The
ATPase activity of an ‘essential’ Bacillus subtilis enzyme YdiB is
required for its cellular function and is modulated by oligomerization.
Microbiology 155, 944−956.
(38) Milligan, J. F., Groebe, D. R., Witherell, G. W., and Uhlenbeck,
O. C. (1987) Oligoribonucleotide synthesis using T7 RNA polymerase
and synthetic DNA templates. Nucleic Acids Res. 15, 8783−8798.
(39) Peterson, E. T., and Uhlenbeck, O. C. (1992) Determination of
recognition nucleotides for Escherichia coli phenylalanyl-tRNA
synthetase. Biochemistry 31, 10380−10389.
(40) Gehrke, C. W., Kuo, K. C., McCune, R. A., Gerhardt, K. O., and
Agris, P. F. (1982) Quantitative enzymatic hydrolysis of tRNAs:
reverse phased high performance liquid chromatography of tRNA
nucleosides. J. Chromatogr. 230, 297−308.
(41) Falb, E., Nudelman, A., and Hassner, A. (1993) A Convenient
Synthesis of Chiral Oxazoldin-2-ones and Thiazolidin-2-ones and an
Improved Preparation of Triphosgene. Syn. Commun. 23, 2839−2844.
(42) Piccariello, T., Kirk, R. J., Olson, L. P. (2004) Active Agent
Delivery Systems and Methods for Protecting and Administering
Active Agents, U.S. 2004/0062628, p. 38.
(43) Hong, C. I., Chheda, G. B. (1978) in Nucleic Acid Chemistry Part
2 (Townsend, L. B. Tipson, R. S., Eds.) p 664, John Wiley & Sons,
New York.
(44) Harris, K. A., Jones, V., Bilbille, Y., Swairjo, M. A., and Agris, P.
F. (2011) YrdC exhibits properties expected of a subunit for a tRNA
threonylcarbamoyl transferase. RNA 17, 1678−1687.
(45) Tsang, H., Buck, M., and Ames, B. N. (1983) Sequence
specificity of tRNA-modifying enzymes. An analysis of 258 tRNA
sequences. Biochim. Biophys. Acta 741, 180−196.
8963
dx.doi.org/10.1021/bi301233d | Biochemistry 2012, 51, 8950−8963