Two complementary enzymes with separated functions in
threonylation of tRNA, crystal structure of putative threonyl-tRNA
synthetase ThrRS-1 from Aeropyrum pernix
Satoru Shimizu1, Ella Czarina Magat Juan1‡, Yu-ichiro Miyashita1, Yoshiteru Sato1§, Md.
Mominul Hoque1, Kaoru Suzuki2, Masataka Yogiashi2, Masaru Tsunoda3,
Anne-Catherine Dock-Bregeon4, Dino Moras4, Takeshi Sekiguchi2 and Akio Takénaka1,3*
1
Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology,
Nagatsuda, Midori-ku, Yokohama, 226-8501, Japan, 2College of Science and
Engineering, 3Faculty of Pharmacy, Iwaki-Meisei University, Chuodai-iino, Iwaki,
Fukushima 970-8551, Japan and 4Departement de Biologie et Génomique Structurales,
Institut de Génétique et de Biologie Moléculaire et Cellulaire, 1, Rue Laurent Fries,
F-67404 Illkirch, France
‡ Present address: Genomic Sciences Center, RIKEN Yokohama Institute, 1-7-22
Suehiro-cho, Tsurumi, Yokohama 230-0045, Japan.
§ Present address: Departement de Biologie et Génomique Structurales, Institut de
Génétique et de Biologie Moléculaire et Cellulaire
*corresponding author. E-mail address:
atakenak@iwakimu.ac.jp
Abbreviations used: Ap, Aeropyrum pernix; St, Sulfolobus tokodaii; Ec, Escherichia coli;
Sa, Staphylococcus aureus; Pa, Pyrococcus abyssi; ARS, aminoacyl-tRNA syntetase;
A-AMP, aminoacyl-AMP; AMS, 5'-O-(N-(L-threonyl)-sulfamoyl)adenosine; ThrRS,
threonyl-tRNA synthetase.
1
Abstract
In protein synthesis, threonyl-tRNA synthetase (ThrRS) must recognize threonine
(Thr) from the twenty kinds of amino acids and the cognate tRNAThr from 47 different
tRNAs in order to generate Thr-tRNAThr. In general, an organism possess one kind of
gene corresponding to ThrRS. However, it has been recently found that some organisms
have two different genes for ThrRS in the genome, suggesting that their proteins
ThrRS-1 and ThrRS-2 function separately to complement with each other in
threonylation of tRNAThr; one for the catalysis and the other for the editing of
mis-acylated Ser-tRNAThr (Acta Crystallogr., F64, 903-910, 2008). In order to clarify
their three-dimensional structures, X-ray analyses of putatively assigned two ThrRSs
from Aeropyrum pernix (Ap) have been subjected. These proteins were overexpressed in
E. coli, purified, and crystallized. The crystal structure of ApThrRS-1 has been
successfully determined at 2.3 Å resolution, as the first example. ApThrRS-1 is a dimeric
enzyme composed of the two identical subunits, each containing two domains for the
catalytic reaction and for the anticodon-binding. The essential editing domain is
completely missing as expected. These structural features reveal that ThrRS-1 catalyze
only the aminoacylation of the cognate tRNA, suggesting the necessity of the second
enzyme ThrRS-2 for editing. Since the N-terminal sequence of Ap-ThrRS-2 is similar to
the sequence of the editing domain of ThrRS from Pyrococcus abyssi, Ap-ThrRS-2 has
been expected to catalyze de-aminoacylation of a misacylated serine moiety at the CCA
terminus.
2
Introduction
In protein synthesis, the proper attachment of an amino acid (hereafter refer to A as a
variable) to the cognate tRNAA has to perform at the highest accuracy by an
aminoacyl-tRNA syntetase (ARS) specific to both molecules, prior to and independent to
the protein synthesis. ARS catalyzes such an aminoacylation by the following two step
reactions. At first, ARS synthesizes an intermediate aminoacyl-AMP (A-AMP) from A
and ATP. Second, the activated aminoacyl group of A-AMP is transferred to the cognate
tRNAA to form A-tRNAA. To maintain the highest fidelity in translation, twenty kinds of
ARSs exist in general for twenty kinds of amino acids, and they are divided into two
classes, class I and class II, each containing equally ten kinds of ARSs. Threonyl-tRNA
sythetase (ThrRS) responsible to threonine (Thr) belongs to the class II, in which the
enzymes generally adopt a dimeric or a tetrameric form in order to approach the major
groove side of the cognate tRNA so that the corresponding amino acid is charged to the
3′-hydroxyl group of the terminal adenosine residue1.
The X-ray structures of ThrRSs from bacteria, E. coli (EcThrRS)2 and
Staphylococcus aureus (SaThrRS)3 have been reported. Both are dimeric enzymes of the
two identical subunits. Each subunit is composed of three domains; the N-terminal
editing domain, the catalytic domain and the C-terminal anticodon-binding domain. At
the active site of the catalytic domain, one cysteine and two histidine residues are
consistently conserved to form the zinc finger motif. In the anticodon-binding domain, a
lysine and an argeinine residues are conserved for interactions with the anticodon G35
and U36 bases of tRNAThr, respectively.
The three-domains system have been also identified in the archaeal ThrRS of
Pyrococcus abyssi (PaThrRS), but only the N-terminal domain has no sequence
similarity with those of eukaryote. This domain has been identified to have an editing
function4. The X-ray structure of only the editing domain has revealed that the tertiary
structure is quite different from those of EcThrRS and SaThrRS5.
Recently, two different genes have been assigned for ThrRS in crenarchaea
Sulfolobus solfataricus genome. It has been suggested that one is the catalytic enzyme
(ThrRS-cat) and the other is the editing enzyme (ThrRS-ed)6. Namely, ThrRS-cat
synthesizes both Thr-tRNAThr and Ser-tRNAThr, but lacks the editing activity to
Ser-tRNAThr. While, ThrRS-ed lacks the aminoacylation activity but removes the serine
moiety from the mis-acylated Ser-tRNAThr. Therefore, it has been considered that the
two proteins work complementarily to each other.
Our previous study7 have found that crenarchaea Aeropyrum pernix (Ap), Sulfolobus
tokodaii (St), Sulfolobus acidocaldarius and Metallosphaera sedula genomes contain the
two different ThrRS genes (ThrRS-1 and ThrRS-2). ApThrRS-1 contains a catalytic
3
domain and an anticodon-binding domain but no editing domain. On the other hand,
ApThrRS-2 contains an editing domain and an anticodon-binding domain but no
catalytic domain. Similar architectural features have been found between StThrRS-1 and
StThrRS-2. In order to reveal their structural features and to obtain insights of
structure-based evolutional aspects, X-ray analyses of ApThrRS-1, ApThrRS-2,
StThrRS-1 and StThrRS-2 have been subjected, and the crystal structure of Ap-ThrRS-1
has been successfully determined at 2.3 Å resolution, as the first example.
Results
Quality of X-ray analyses
In the ApThrRS-1 (selenomethionine derivative) crystal, a protein molecule
occupying the asymmetric unit contains twelve selenomethionine residues. However,
Patterson functions of the MAD data showed only seven Se atoms at a significant level,
from which the first solution was derived with the correlation coefficient of 0.357
between Fo and Fc. After several heavy-atom positional refinements, a total of eleven
selenium atoms with occupancies greater than 50% were finally found. An electron
density map calculated with the estimated phase angles was modified by solvent
flattering technique, the solvent content being fixed to be 34% which was kept to be less
than the estimated solvent content 44.8%8 in order to include solvent molecules in the
asymmetric unit. The final correlation coefficient in |E2| was 0.771. After structure model
building, more than 96% residues of the protein were successively assigned.
The atomic parameters of the crystal structure were refined using the program
Refmac9 in combination with Coot10 for assigning further additional atoms. Finally, 465
amino acid residues were assigned except for the N-terminal six residues. 280 water
molecules, one zinc atom and two sulfate ions were also assigned. The final R factor and
Rfree11 values were 16.9% and 23.2%, respectively. Statistics of the structural refinement
is given in Table 1. In a Ramachandran plot, 382 (93.6), 20 (4.9), 3 (0.7) and 3 (0.7%)
residues are in the most favored, in the allowed, in generally allowed and in disallowed
geometries, respectively.
In the case of ApThrRS-1 crystal, the molecular replacement method applied to the
15-2.3 Å resolution data showed a unique solution which exhibited a high correlation
coefficient (0.70) and a low R value (38 %), the other solutions being 0.10~0.19 and
61~62%, respectively. The ApThrRS-1 crystal was isomolphous with the ApThrRS-1*
crystal containing one protein molecule in the asymmetric unit. The atomic parameters
of the crystal structure were refined using the programs Refmac 59 in combination with

The symbol asterisk indicates selenomethionine derivative of ApThrRS-1
4
Coot10 for assigning further additional atoms. Finally, 465 amino acid residues were
assigned except for the N-terminal six residues. 268 water molecules, one zinc atom and
two sulfate ions were also assigned. The final R factor and Rfree11 were 18.2% and 24.8%,
respectively. Statistics of the structural refinement is given in Table 1. The refined
structure is well fitted on the final 2|Fo|-|Fc| map, as shown in Fig. 1. In a Ramachandran
plot, 378 (92.6%), 23 (5.6%), 2 (0.5%) and 5 (1.2%) residues were in the most favored,
in the allowed, in generally allowed, and in disallowed geometries, respectively.
In order to detect the structural differences between the two structures of ApThrRS-1
and ApThrRS-1*, an r.m.s. deviation between the corresponding C atoms when the two
structures were superimposed onto each other was calculated to be 0.14 Å except for
Gly85 and His86 which were deviated more than 1.0 Å. This shows that the two
structures are essentially the same. Therefore, only the structure of ApThrRS-1 is
described and discussed below.
Overall structure
The two ApThrRS-1 molecules, related by the crystallographic two-fold symmetry,
associate with each other to form a dimer. This is a typical feature as a member of the
class II ARS and similar to those of EcThrRS and SaThrRS. Superimposition of the
ApThrRS-1 dimer onto that of EcThrRS is shown in Fig. 2. The dimer formations are
similar to each other, but different in the peripheral regions. In EcThrRS, the two central
catalytic domains are associated, and the two N-terminal editing domains are extended in
the opposite directions away from the catalytic domains. In the case of ApThrRS-1,
however, the N-terminal domains are interacted to each other to form a dimer. Therefore,
the N-terminal domains of ApThrRS-1 correspond to the catalytic domains of EcThrRS.
In addition, the C-terminal domains of ApThrRS-1 are also well superimposed onto the
anticodon binding domains of EcThrRS. It is easy to see that the editing domains of
EcThrRS are completely missing in ApThrRS-1, as predicted from sequence alignments7.
Hereafter, the N-terminal and the C-terminal domains of ApThrRS-1 are referred to the
catalytic and the anticodon-binding domains, respectively, and their validities are
discussed with the detailed structures.
Discussion
Folding features of the subunit
As the two subunits of a dimmer are related by the crystallographic symmetry, only
the subunit structure will be discussed below. A ribbon drawing of the subunit and a
5
topological diagram of the secondary structures are shown in Fig. 3. The subunit is
composed of the two domains. The N-terminal domain (residue numbers 7-333) contains
twelve -helices (1~12) and ten -strands; 1~3 in a small anti-parallel -sheet and
4~10 in a large -sheet mixed with parallel and anti-parallel strands. The and
helices cover the back side of the large -sheet, the other side of the 9-8-6-7 strands
being opened. On the front side of the 4-5-10 strands, the small domain
(and1-3-2) and additional four helices (and
 form a large pocket, suggesting a catalytic site. This main-chain folding is similar
to the catalytic domains of the class II ARS. 
The catalytic domains of the class II ARS must contain the conserved three
fragmental sequence motifs$: gxxxxP in the motif 1, (F/Y/H)Rx(E/D) and
(R/H)xxxFxxx(D/E) in the motif 2 and xggeR in the motif 3. As shown in
Fig. 3, these motifs are consistently observed in the N-terminal domain of ApThrRS-1.
Fig. 4 shows the three motifs aligned between ApThrRS-1 and EcThrRS and Fig. 5(a)
shows their locations on the structures by superimposing the two catalytic domains. The
secondary structures are very similar to each other though the peripheral main-chain
conformations are different. In addition, the three motifs essential to the catalytic
domains are situated almost the same positions in the tertiary structures. In the motif 1,
Pro74, Ile75, Ile76, Val71 and Tyr68, which are located on a long loop extended from the
 helix of ApThrRS-1, are conserved in the sequence and in the structure (Fig. 4 (b)),
the corresponding residues being Pro294, Phe295, Met296, Val291 and Tyr288 which are
also on a long loop extended from an -helix in EcThrRS. The motif 2 containing the
two fragments HRxE and RxxxFxxxD are found from His143 to Ile167. They are located
at the end of 4 and from an extended long loop containing 7 helix to a half of 5 in
ApThrRS-1. The corresponding residues in EcThrRS are highly conserved to maintain
the partial structure similar to that of ApThrRS-1. The motif 3 from Arg311 to Tyr324
starts from the end of 10 to 12 including a short loop. This motif is also highly
conserved between ApThrRS-1, EcThrRS and SaTheRS. In addition, a zinc atom is
found at almost the same position found in the catalytic domain of EcThrRS. These
features suggest that ApThrRS-1 belongs to the class II ARS and the N-terminal first
domain must be the catalytic domain.
Fig. 5(b) shows the main-chain folding of the C-terminal domain (residue numbers
334-471) superimposed on the anti-codon binding domain of EcThrRS. In these domains,
too, their secondary-structure folding are very similar to each other though the peripheral
$
One-letter codes are used for amino acid residues in sequence expression, and their
lower case letters indicate less conservative amino acid residues. The letter x stands for
any residues,  for small amino acids P, G, S and T;  for hydrophobic residues F, Y, W, I,
L, V, M and A1.
6
conformations are different. The -sheet composed of the five -strands, 11~15, is
surrounded by the three principal -helices. The long 13 and 16 helices are on the
back side of the -sheet and another long helix 16 covers above the front side of 12
and 11. An additional short helix 15 is located between the two ends of the 13 and
14 strands. The corresponding part in EcThrRS is a long loop. A short helix 17
appears at the C-terminus of ApThrRS-1 after a long extension from 16 which goes
around the back side of the -sheet. As a result, the 13, 14 and 15 strands expose the
front side of the -sheet. This folding feature is very similar to those of the
anticodon-binding domains of EcThrRS and SaThrRS.
Interactions between the two subunits of the dimer
Fig. 5(c) shows the interface between the two catalytic domains. The two 2 strands,
which are at the one ends of the catalytic domains, are associated to each other between
the two domains related by the crystallographic two-fold symmetry. An extended sheet
formed by jointing the two small -sheets (2, 3 and 1) is largely curved like a saddle,
inside of which several hydrophobic amino acid residues are packed, as seen in the
figure.
Catalytic site of ApThrRS-1
In the catalytic domain of ApThrRS-1, the zinc atom occupies a position in the large
pocket surrounded by the three motifs. The Cys112, His166 and His310 residues are
coordinate to the Zn2+ atom at the distances 2.3 Å (Zn-S of Cys112), 2.0 Å (Zn-N of
His166) and 2.1 Å (Zn-N atom of His310) in a tetrahedral configuration, as seen in Fig.
6. The fourth position of the tetrahedron is occupied by a water molecule which is further
hydrogen-bonded to Asp164. In the proximity, a sulfate ion is bound through hydrogen
bonds with them. In order to confirm the validity of the above speculations, it is
necessary to compare the present structure with those of EcThrRS reported already in
details. Fig. 7 shows a superimposition of the catalytic domain onto that of EcThRS in
complex with Thr-AMS [5'-O-(N-(L-threonyl)-sulfamoyl)adenosine, an analog of
Thr-AMP]12. The zinc finger is conserved in both structures at almost the same position,
The three corresponding ligands in EcThrRS are Cys112, His166 and His310. At the
forth position, the hydroxyl group and the amino group of the threonine moiety of
Thr-AMS are bound so that the zinc atom adopts a distorted square-pyramidal
pentacoordination. Furthermore, the bound threonine in EcThrRS is trapped by hydrogen
bonds between the threonine carbonyl group and the side chains of Arg363, Gln381 and
Gln484, and between the threonine amino group and the hydroxyl group of Tyr462.
7
These interacting residues are also conserved in ApThrRS-1. They are Arg144, Gln162,
Gln282 and Tyr257, respectively. The Tyr257 is positioned at slightly distant position in
the apo-form of ApThrRS-1, but it may come closer when the substrate is bound. In
ApThrRS-1, the hydroxyl group of the threonine moiety might form a hydrogen bond
with the side chain of Asp164 as well as Asp383 does in EcThrRS. This interaction and
the zinc atom coordination are essential for recognition of threonine. These specific
features are also able to conserve in ApThrRS-1. The methyl group of threonine is placed
in a hydrophobic environment by Thr482 and Ala513 in EcThrRS. This situation can be
formed by Thr280 and Ala312 in ApThrRS-1. These common features suggest that this
domain contains a threonine specific site.
In EcThrRS, the adenine moiety of AMS is surrounded by the residues of Val376,
Phe379 and Arg520, which belong to the motifs 2 and 3 (see Fig. 5). The corresponding
residues in ApThrRS-1 are Val157, Phe160 and Arg319. The main chains of Gln479 and
Cys480, and the side chain of Gln479 in EcThrRS interact with the hydroxyl groups of
the ribose. The corresponding residues in ApThrRS-1 are Gln277 and Met278. Although
they shift away in ApThrRS-1, they can move back to interact with the ribose when ATP
is bound. These structural consistencies also support that this domain of ApThrRS-1 has
the catalytic features.
Furthermore, another possibility that the CCA terminus of tRNAThr binds the
catalytic site of ApThrRS-1 was examined by comparing to EcThrRS (PDBID 1QF62 in
complex with tRNAThr. Fig. 8 shows the catalytic site of ApThrRS-1 superimposed on
that of EcThrRS with the bound tRNAThr. In EcThrRS, the terminal adenine (A76
residue) moiety of tRNAThr is intercalated between Tyr313 and Arg363, and its N6 atom
forms a hydrogen bond with the main-chain carbonyl group of Ala316. These three
residues correspond to Tyr91, Arg144 and Asn94 in ApThrRS-1, respectively. As shown
in Fig. 8, the two hydroxyl oxygen atoms of A76 of tRNAThr are involved in
hydrogen-bonded interactions; the 2′-hydroxyl group with His309 and Tyr462 and the
3′-hydroxyl group with Gln484. The corresponding residues in ApThrRS-1 are His87,
Tyr257 and Gln282, respectively. These structural consistencies also support that
tRNAThr also can be bound in the catalytic site of ApThrRS-1.
Anticodon-binding site of ApThrRS-1
A superimposition of the anticodon-binging domain of ApThrRS-1 onto that of
EcThrRS (PDBID 1QF6)2 is shown in Fig. 9. The Glu600 and Arg609 residues in
EcThrRS, which interact with the anticodon G35 and U36 residues, respectively,
correspond to Glu404 and Arg416 in ApThrRS-1, respectively. Although their side
chains are pointed away in ApThrRS-1 with no interactions with tRNA, they might
8
interact to the anticodon bases when tRNAThr is bound in this domain of ApThrRS-1. In
EcThrRS, the wobble base C34 is positioned at the end of the anticodon binding domain,
and has no specific interaction with the proximal residue (Gln575); it interacts with C34,
but it is not always (see PDBID 1QF6). The corresponding residue in ApThrRS-1 is
Thr382, the O atom of which may be possible for joining such a non-specific interaction
with C34 of tRNAThr. These structural features suggest that the anticodon-binding
domain of ApThRS-1 is able to recognize the anticodon of tRNAThr.
Plausible structure of ApThrRS-2
The structure of ApThrRS-2 is not yet determined. However, it is worthwhile to
examine a possibility of the editing and the anticodon-binding functions. The sequence
of ApThrRS-2 can be devided into three domains; the editing domain, an additional
(unknown) domain of 109 residues and the anticodon-binding domain7. The amino acid
sequence in the editing domain of ApThrRS-2 is similar to that of PaThrRS with the
identity of 37%. It has been reported that the two editing domains of PaThrRS form a
dimer5. The structure of the editing domain of ApThrRS-2 might be similar to that of
PaThrRS which is different from that of EcThrRS. The crystal structure of the editing
domain of PaThrRS in complex with seryl-3′-aminoadenosine, an analog of Ser-AMP, is
reported (PDBID 2HL05). The proposed catalytic residues and the interacting residues
with the adenine base in PaThrRS are conserved in ApThrRS-2 (data not shown). These
features suggest that the editing domain of ApThrRS-2 hydrolyzes the misacylated
Ser-tRNAThr in a manner similar to PaThrRS.
Conclutions
X-Ray analysis of ApThrRS-1 has shown that ApThrRS-1 adopts a dimeric form of the
two identical subunits, each subunit containing the three characteristic motifs 1, 2 and 3
in the amino acid sequence as the typical features of the class II aminoacyl-tRNA
synthetase. In comparison of the structure with those of EcThrRS, it is easy to identify
that ApThrRS-1 is composed of the catalytic and the anticodon-binding domains but it
has no editing domain as expected. Superimpositions of ApThrRS-1 onto EcThrRS in
complex with several substrates have shown that the residues interacting with threonine,
ATP and tRNAThr in EcThrRS are highly conserved in ApThrRS-1, and that the residues
interacting to the anticodon of tRNAThr are also highly conserved. These structural
features suggest that ApThrRS-1 generate Thr-tRNAThr but cannot hydrolyze a
misacylated product Ser-tRNAThr. ApThrRS-2 might be a dimeric enzyme that each
subunit consists of an archaeal editing domain and an anticodon-binding domain, and
9
that dimerization occurs between the two editing domains. If so, the two enzymes might
be developed from different ancestor enzymes. X-ray analysis of ApThrRS-2 in progress
will provide insights of structure-based evolutional aspects for two ThrRS gene
organisms.
Materials and Methods
X-ray experiments and data collection
As described in the previous paper7, ApThrRS-1 encoded on the APE0809.1 gene
and its selenomethionine derivative (ApThrRS-1*) were overexpressed, purified and
crystallized. Prior to data collection for MAD (Multiple Anomalous Dispersion) phasing,
an XAFS (X-ray Absorption Fine Structure) spectrum of the ApThrRS-1* crystal was
measured on the synchrotron radiation beamline BL-17A of PF (Photon Factory, Ibaraki,
Japan). Using three wavelengths selected as the peak, the edge and the remote, X-ray
diffraction patterns were recorded on a Quantum 4R CCD detector (Area Detector
Systems Co., San Diego, California, USA) at 95 K in the full range (1-360˚) of
oscillation angle. Exposure time was 5 s for 1° oscillation per image, and the detector
was positioned at 167.6, 167.4 and 163.2 mm away from the crystal for the three
wavelengths, respectively.
X-ray data collection of the ApThrRS-1 crystal was carried out at 95 K in a range of
1-180˚ oscillation angle using synchrotron radiation (=1.00 Å) on the beamline BL-6A
of PF (Photon Factory, Ibaraki, Japan). The diffraction patterns were recorded on a
Quantum 210 CCD detector (Area Detector Systems Co., San Diego, California, USA)
positioned at 148.9 mm away from the crystal. Each frame was taken with a 2 s for a 1°
oscillation step.
For both crystals, Bragg spots recorded on every image were indexed and their
intensities were estimated by integrating around them. Intensity data were then merged
and scaled between the frames with the HKL2000 package13. These data were further
converted to structure-factor amplitudes with TRUNCATE from the CCP4 suite14.
Structure determination
The MAD method was applied to solve the phase problems of the ApThrRS-1*
crystal with the program SHARP/autoSHARP15. The program autoSHARP was the
automated structure solution pipeline, built around the program SHELXD16 for location
of heavy atoms, the program SHARP for the heavy-atom refinement, the program
SOLOMON17 for density modification and the program ARP/wARP18 for automated
structural model building and the program REFMAC 59 for atomic refinement.
10
The initial structures of the ApThrRS-1 crystal was derived by the molecular
replacement method with the program Molrep19 using the ApThrRS-1* structure as a
probe. All the atomic parameters were refined using the program REFMAC 59. The
structural modification was performed with the program COOT10. These two calculations
were alternately repeated. The refined structure was validated on a Ramachandran plot
using the program PROCHECK20.
Accession numbers
Protein Data Bank accession numbers
Coordinates and structure factors have been deposited in the Protein Data Bank21 with
accession codes 3A31 for the selenomethionine derivative (ApThrRS-1*) and 3A32 for
the native type (ApThrRS).
Acknowledgements
This work was supported in part by the Grants-in-Aid for Scientific Research
(No.20570104) and Protein3000 Research Program from the Ministry of Education,
Culture, Sports, Science and Technology of Japan. We thank S. Kuramitsu for organizing
the research group in the program, and N. Igarashi and S. Wakatsuki for facilities and
help during data collection.
11
References
1. Eriani G., Delarue M., Poch O., Gangloff J. & Moras D. (1990). Partition of tRNA
synthetases into two classes based on mutually exclusive sets of sequence motif.
Nature, 347, 203-206.
2. Sankaranarayanan R., Dock-Bregeon A. C., Romby P., Caillet J., Springer M., Rees
B., Ehresmann C., Ehresmann B. & Moras D. (1999). The structure of
threonyl-tRNA synthetase-tRNAThr complex enlightens its repressor activity and
reveals an essential zinc ion in the active site. Cell, 97, 371-381.
3. Torres-Larios A., Sankaranarayanan R., Rees B., Dock-Bregeon A. C. & Moras D.
(2003). Conformational movements and cooperativity upon amino acid, ATP and
tRNA binding in threonyl-tRNA synthetase. J. Mol. Biol. 331, 201-211.
4. Beebe K., Merriman E., Ribas De Pouplana L. & Schimmel P. (2004). A domain for
editing by an archaebacterial tRNA synthetase. Proc. Natl. Acad. Sci. USA, 101.
5958-5963.
5. Hussain T., Kruparani S. P., Pal B., Dock-Bregeon A. C., Dwivedi S., Shekar M. R.,
Sureshbabu K. & Sankaranarayanan R. (2006). Post-transfer editing mechanism of a
D-aminoacyl-tRNA deacylase-like domain in threonyl-tRNA synthetase from
archaea. EMBO J. 25, 4152-4162.
6. Korencic D., Ahel I., Schelert J., Sacher M., Ruan B., Stathopoulos C., Blum P., Ibba
M. & Söll D. (2004). A freestanding proofreading domain is required for protein
synthesis quality control in archaea. Proc. Natl. Acad. Sci. USA, 101 10260-10265.
7. Shimizu S., Juan E. C. M., Miyashita Y., Sato Y., Suzuki K., Yogiashi M., Tsunoda
M., Dock-Bregeon A. C., Moras D., Sekiguchi T. & Takénaka A., (2008).
Crystallographic study on two types of putative threonyl-tRNA synthetases from
crenarchaea Aeropyrum pernix and those from Sulfolobus tokodaii. Acta Crystallogr.
Sect. F, 64, 903-910.
8. Matthews B. W., (1968). Solvent content of protein crystals. J. Mol. Biol. 33,
491-497.
9. Murshudov G. N., Vagin A. A. & Dodson E. J., (1997). Refinement of
Macromolecular Structures by the Maximum-Likelihood Method. Acta Crystallogr.
Sect. D, 53, 240-255.
10. Emsley P. & Cowtan K. (2004). Coot: model-building tools for molecular graphics.
Acta Crystallogr. Sect. D, 60, 2126-2132.
11. Brünger, A. T. (1992). Free R value: a novel statistical quantity for assessing the
accuracy of crystal structures. Nature, 355, 472–475.
12. Sankaranarayanan, R., Dock-Bregeon, A.C., Rees, B., Bovee, M., Caillet, J., Romby,
P., Francklyn, C.S. & Moras, D. (2000). Zinc ion mediated amino acid
12
discrimination by threonyl-tRNA synthetase. Nature Struct. Biol. 7, 461-465.
13. Otwinowski Z. & Minor W., (1997). Processing of X-ray Diffraction Data Collected
in Oscillation Mode. Methods in Enzymology, 276, 307-326.
14. Collaborative Computational Project, Number 4, 1994.
15. Vonrhein C., Blanc E., Roversi P. & Bricogne G., (2007). Automated structure
solution with autoSHARP. Methods Mol. Biol. 364, 215-230.
16. Schneider T. R. & Sheldrick G. M., (2002). Substructure solution with SHELXD 2002.
Acta Crystallogr. Sect. D, 58, 1772-1779.
17. Abrahams J. P. & Leslie A. G. W., (1996). Methods used in the structure
determination of bovine mitochondrial F1 ATPase. Acta Crystallogr. Sect. D, 52,
30-42.
18. Perrakis A., Morris R. M. & Lamzin V. S., (1999). Automated protein model building
combined with iterative structure refinement. Nature Struct. Biol. 6, 458-463.
19. Vagin A. & Teplyakov A., (1997). MOLREP: an automated program for molecular
replacement. J. Appl. Crystallogr. 30, 1022-1025.
20. Laskowski R. A., MacArthur M. W., Moss D. S. & Thornton J. M., (1993).
PROCHECK: a program to check the stereochemical quality of protein structures. J.
Appl. Crystallogr. 26, 283-291.
21. Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H. et al.
(2000). The Protein Data Bank. Nucleic Acids Res. 28, 235–242.
13
Table 1. Statistics of data processing and structural refinement.
Wavelength (Å)
Data processing
Resolution (Å)
Space group
Unit cell (Å)
a
b
c
Observed reflections
Unique reflections
Completeness (%)
Redundancy
I/
Rmerge (%)
Structural refinement
Resolution (Å)
Used reflections
R (%)
Rfree (%)
r.m.s bond length (Å)
r.m.s. angles (˚)
Protein atoms
Water molecules
SO42Zn2+
peak
0.97898
ApThrRS-1*
edge
0.97966
remote
1.00
ApThrRS-1
Native
1.00
50-2.5 (2.59-2.50)
C2221
50-2.5 (2.59-2.50)
C2221
50-2.5 (2.59-2.50)
C2221
50-2.3 (2.38-2.30)
C2221
81.8
103.6
112.5
244 125
16 854
100 (100)
14.5 (14.6)
49.6 (17.5)
8.3 (20.9)
81.9
103.6
112.6
246 544
17 017
100 (100)
14.5 (14.7)
47.9 (16.0)
8.4 (23.3)
82.0
103.7
112.7
240 627
16 636
100 (100)
14.5 (14.6)
47.6 (14.4)
8.0 (25.8)
81.5
103.7
112.5
155 186
21 565
100 (99.9)
7.2 (7.0)
19.9 (6.9)
9.6 (30.0)
41.0-2.50
16 758
16.9
23.2
0.02
2.0
3710
280
2
1
40.7-2.3
21 157
18.2
24.8
0.02
2.0
3713
221
5
1
Rmerge = 100 × ∑hkl∑i|Ii(hkl)-I(hkl)|/∑hkl∑iI(hkl) R = 100 ×∑||Fo|-|Fc||/∑|Fo|, where |Fo| and |Fc| are the observed and calculated
11
structure-factor, respectively .
14
(Figure captions)
Fig. 1. A part of the final 2|Fo|-|Fc| electron density map of the ApThrRS-1 crystal,
contoured at 1.5 level.
Fig. 2. C-chain presentations of the tertiary structures superimposed between the
ApThrRS-1 dimer and the EcThrRS dimer (the two subunits are colored in red and
green) along the two-fold axis (a) and its side view (b). In the case of ApThrRS-1, the
two subunits are colored in the same blue.
Fig. 3. A ribbon drawing of the tertiary structure of ApThrRS-1 subunit (a) and a
topological diagram of the secondary structures. Orange bars and green arrows indicate
-helices and -strands, respectively. Values indicate the residue number. -Helices ()
and -strands () are independently numbered from the N-terminus. In (b), regions
containing the three motifs specific to the class II RS are highlighted with blue lines.
Blue dotted lines indicate the predicted active sites of the enzymes.
Fig. 4. Sequence alignments of the three motifs in EcThrRS and ApThrRS-1. Red letters
indicate the characteristic sequences of the three motifs1. Numbers indicate the residue
numbers at the both ends of sequences.
Fig. 5. Ribbon drawings of the superimposed domain structures between ApThrRS-1
(colored in green) and EcThrRS (colored in gray), (a) the catalytic domains and (b) the
anticodon binding domains. In (a), the bound zinc atoms are colored in brawn and the
three motifs (the specific sequences colored in red in Fig. 4) are indicated with different
colors. The hydrophobic interactions found in the interface between the two subunits of
ApThrRS-1 dimer (c).
Fig. 6. A stereo-pair diagram of the zinc finger found in the catalytic domain of
ApThrRS-1 (colored in dark red) superimposed onto that of EcThrRS (colored in light
blue). Zinc, sulfur, oxygen, nitrogen and carbon atoms are colored in green, yellow, red,
dark blue and gray, respectively.
Fig. 7. A stereoview of the catalytic sites, superimposed between ApThrRS-1 (red) and
EcThrRS (from PDBID 1EVK12, colored in cyan) with a threoninyl-AMP analogue
(green). The gray sphere indicates the zinc ion. Dotted lines indicate possible hydrogen
bonds in EcThrRS. Zinc, sulfur, oxygen, nitrogen and carbon atoms are colored in light
gray, yellow, red, dark blue and dark gray, respectively. The residue numbers are for
15
ApThrRS-1.
Fig. 8. A stereoview of the catalytic sites, superimposed between ApThrRS-1 (red) and
EcThrRS (from PDBID 1QF62, colored in cyan) with the cognate tRNAThr (green).
Dotted lines indicate possible hydrogen bonds in EcThrRS. Zinc, sulfur, oxygen,
nitrogen and carbon atoms are colored in light gray, yellow, red, dark blue and dark gray,
respectively. The residue numbers are for ApThrRS-1.
Fig. 9. A ribbon drawing of the anticodon binding domain of ApThrRS-1 (green)
superimposed onto that of EcThrRS (gray) complexed with tRNAThr (yellow)2. The
tRNA anticodon bases are bound at the bottom of the anticodon binding domain, in
which the two bases, G35 and U36, are specifically interacted with the conserved Glu
and Arg residues. The wobble base C34 has no strong interactions with the protein and it
is extruded into solvent region. The amino acid residue number colored in green are for
ApThrRS-1.
16