Mass spectrometric screening of ligands with lower off-rate from a clicked-based pooled library

Article abstract of DOI:10.1021/co300028n

This paper describes a convenient screening method using ion trap electrospray ionization mass spectrometry to classify ligands to a target molecule in terms of kinetic parameters. We demonstrate this method in the screening of ligands to a hexahistidine tag from a pooled library synthesized by click chemistry. The ion trap mass spectrometry analysis revealed that higher stabilities of ligand-target complexes in the gas phase were related to lower dissociation rate constants, i.e., off-rates in solution. Finally, we prepared a fluorescent probe utilizing the ligand with lowest off-rate and succeeded in performing single molecule observations of hexahistidine-tagged myosin V walking on actin filaments.

Full text of DOI:10.1021/co300028n

Letter
pubs.acs.org/acscombsci
Mass Spectrometric Screening of Ligands with Lower Off-Rate from a
Clicked-Based Pooled Library
Satoshi Arai,† Shota Hirosawa,‡ Yusuke Oguchi,§ Madoka Suzuki,∥ Atsushi Murata,‡ Shin’ichi Ishiwata,§,∥
and Shinji Takeoka*,‡,∥
†Consolidated Research Institute for Advanced Science and Medical Care, Waseda University, 513 Wasedatsurumaki-cho, Tokyo
162-0041, Japan
‡Department of Life Science and Medical Bioscience, Graduate School of Advanced Science and Engineering, Waseda University
(TWIns), 2-2 Wakamatsu-cho, Shinjuku-ku, Tokyo 162-8480, Japan
§Department of Physics, Faculty of Science and Engineering, Waseda University, 3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
∥Waseda Bioscience Research Institute in Singapore, Waseda University, 11 Biopolis Way, #05-01/02 Helios, Singapore 138667,
Republic of Singapore.
S
* Supporting Information
ABSTRACT: This paper describes a convenient screening method using ion trap
electrospray ionization mass spectrometry to classify ligands to a target molecule in
terms of kinetic parameters. We demonstrate this method in the screening of ligands to
a hexahistidine tag from a pooled library synthesized by click chemistry. The ion trap
mass spectrometry analysis revealed that higher stabilities of ligand-target complexes in
the gas phase were related to lower dissociation rate constants, i.e., off-rates in solution.
Finally, we prepared a fluorescent probe utilizing the ligand with lowest off-rate and
succeeded in performing single molecule observations of hexahistidine-tagged myosin
V walking on actin filaments.
KEYWORDS: click chemistry, fluorescent probes, high-throughput screening, mass spectrometry
arious methods to screen chemical libraries with huge
diversity have been developed in the fields of catalysis,
nickel ion complexed by the ligand. To develop ligands with
strong affinity for a His-tag, scaffolds bearing two or three NTA
moieties, so-called bis- or tris-NTA ligands, have been studied.7
By estimating the structure of hexahistidine, those researchers
designed and synthesized several multiple NTA ligands.
However, the three-dimensional structure of the His-tag
peptide in solution is difficult to determine due to the flexibility
of the peptide.8 Hence, it is not easy to computationally design
the scaffold and spacers required for multiple NTA ligands to
achieve optimum binding to the target. In contrast, very few
reports have appeared describing the systematic screening of
multiple NTA ligands, even bis-NTA.7
A pooled library containing various bis-NTA ligands in
solution was prepared based on the azide−alkyne Huisgen
cycloaddition click reaction.9 NTA derivatives bearing an azide
group on the end of four different aliphatic chain lengths (1−4
methylene units) were combined with a selection of 20 bis-
terminal alkyne spacer groups to generate a total of 200 bis-
NTA ligands with substantial structural diversity (Figure 1).
V
material science and medicinal chemistry.1 Recently, electro-
spray ionization mass spectrometry (ESI-MS) has attracted
attention as a potential tool for label-free screening in a high-
throughput format.2 This strategy has already been imple-
mented in catalyst screening through the investigation of
reaction intermediate species.3 Such an approach allows a
readout of ligand-target noncovalent interactions from a pooled
library without the need to purify each ligand.4 We have also
investigated the relationship between the stability of the ligand-
target noncovalent complex in gas phase and the dissociation
constant (KD) in solution by using a collision activated
dissociation method based on ion trap ESI-MS.5 However,
the relationship between the stability in gas phase and kinetic
parameters (kon and koff) in solution remains ambiguous.
In this paper, we examine this relationship through the
screening of ligands which favor a hexahistidine tag (His-tag) as
a target molecule.6 The best screened ligand was then applied
to the creation of a new fluorescent probe to label His-tagged
proteins. His-tag ligands are widely used in affinity chromatog-
raphy to purify His-tagged proteins. Nitrilotriacetic acid
(NTA), the most common such moiety, is capable of
recognizing two imidazolyl rings of histidine units with one
Received: March 7, 2012
Revised: July 17, 2012
Published: July 18, 2012
© 2012 American Chemical Society
451
dx.doi.org/10.1021/co300028n | ACS Comb. Sci. 2012, 14, 451−455

ACS Combinatorial Science
Letter
The bis-alkyne scaffolds included functional groups such as five-
or six-membered aromatic rings (structures a−j) and aliphatic
chains (k−t).
{2,4}−{3,3}, were not produced in the same solutions (see
the Supporting Information). This strategy avoids the
generation of complexes that cannot be distinguished by ESI-
MS.
Each mixture of the library was screened by the collision
activated dissociation method in an ion trap ESI-MS instru-
ment. The bis-NTA ligands were mixed with NiCl2 in a 1:1
molar ratio (Ni:NTA), and the complexes were treated with
hexahistidine (His). The resulting noncovalent ternary
complexes were detected as divalent negative ions in ESI-MS.
For example, when His was added to the library pools
containing t{1,1}, t{1,2}, t{1,4}, t{2,2}, t{2,4}, and t{4,4}, six
species, all ternary complexes with His and two nickel ions,
were detected in a single scan as shown in Figure 2a.
Figure 1. Synthesis of 200 compounds for the chemical library by click
chemistry. (a) To synthesize branched structures, bis (alkynes) from a
to t were used as clickable scaffolds. (b) As a recognition moiety,
nitrilotriacetic acid (NTA) with different carbon numbers having an
azide group at the terminus were synthesized as a tert-butyl ester
protected form. Each NTA with an azide group was termed as 1, 2, 3
and 4. (c) The “Click” conjugation of one bis (alkynes) scaffold to four
different azide NTAs generated ten species. The nomenclature R is
from a to t and bis-NTA ligands were named as follows;
(alphabet){m, n}.
The bis-terminal alkynes were commercially available or
synthesized by a simple Sonogashira reaction (see the
Supporting Information).10 As a recognition moiety, the four
azidoalkyl-functionalized NTA components (1−4, Figure 1b)
were synthesized in tert-butyl protected form (see the
Supporting Information). A typical ESI-MS sample was
prepared as follows: one scaffold (t) was reacted with an
equimolar mixture of 1, 2, and 4 in a single flask in the presence
of copper sulfate and sodium ascorbate in THF/water(v/v = 1/
1) at room temperature. After the products were separated
from the catalyst by extraction with CH2Cl2, tert-butyl groups
were deprotected with TFA. The resulting library contained a
total of six compounds, both symmetric and unsymmetric:
t{1,1}, t{1,2}, t{1,4}, t{2,2}, t{2,4}, and t{4,4}. Because of the
reliability of the click reaction and its insensitivity to the steric
nature of the azide group, we expected these candidate ligands
to be formed in the statistical ratio of 1:2:2:1:2:1. The relative
intensities observed by ESI-MS supported this assumption (see
Supporting Information). The same result was observed for
each of the other scaffolds, indicating that the reactivity was
independent of the scaffold used. A major advantage of the click
reaction approach is that a simple extraction can be used to
separate the catalyst components from the library, which may
otherwise interfere with the ionization process. Library mixtures
were designed so that different structures with the same
molecular weight, such as {1,3}−{2,2}, {1,4}−{2,3}, and
Figure 2. Screening of the mixed ligands library by ion trap ESI-MS.
(a) Hexahistidine (His) as target was added to the solution containing
various bis-NTA ligands (for example, t{1,1}, t{1,2}, t{2,2}, t{1,4},
t{2,4}, and t{4,4}) preloaded with nickel ions (Ni2+), to form the
ligand-target complex. ESI-MS detects peaks corresponding to the
ternary complex with His and each bis-NTA ligand-Ni × 2 as double
negative ion charge. (b) Each complex is isolated within the ion trap
and the stability of the complex is evaluated by collision activated
dissociation. E1/2 is defined as the index of the stability, where the
relative abundance of the complex (Irel) reaches 0.5. (c) E1/2 from 200
bis-NTA ligands were divided into six categories (i.e., <16%, 16−17%,
17−18%, 18−19%, 19−20% and >20%), which were color coded. (d)
Top 20 bis-NTA ligands from among the 200 library members.
Complexes with other ratios of His to bis-NTA ligand were
not observed. Subsequently, each of the six desired ternary
complexes species was isolated in the ion trap and subjected to
collision with helium gas at increasing energies. Each complex
was observed to dissociate into the ligand and target, such as
[t{1,1}+Ni]− and [His+Ni]−, without any problematic
decomposition of the compounds. The decrease of the relative
abundance of the complex as a function of collision energy was
displayed as a collision profile (Figure 2b).
An index of the stability of ligand-target complexes, E1/2, as
described by Gross et al., was defined as the relative energy
452
dx.doi.org/10.1021/co300028n | ACS Comb. Sci. 2012, 14, 451−455

ACS Combinatorial Science
Letter
value at which the abundance of the complex reached 0.5.11
These E1/2 values represent the relative stabilities of the
complexes in the gas phase, and show substantial differences in
magnitude among the 200 bis-NTA compounds in the library
(Figure 2c and Table S1 in Supporting Information). The rigid
para-phenylene dialkyne f gave rise to several of the most stable
structures, including the one displaying the highest E1/2 value
(f{2,4}). None of the flexible dialkynes of approximately the
same alkyne−alkyne distance (m, n, o), nor the regioisomeric
dialkynylbenzenes g and h, performed as well. Interestingly, the
relatively flexible dialkyne j did provide several good ligands
with the longer NTA-azides ({3,3}, {3,4}, and {4,4}). In several
cases, single carbon differences in spacer length of the azide
NTA changed the E1/2 substantially even for the same scaffold
(c.f. f{2,4} vs f{3,4} and f{3,3} vs f{3,4}), showing that overall
affinities for histidine are a sensitive combination of multiple
structural and dynamic factors.
We expected that a greater E1/2 value would reflect a higher
stability of the ligand-target complex in solution, that is, better
ligand binding to His and hexahistidine tags. Previous work has
address the question of whether ligand-target affinities in the
gas phase measured with ESI-MS can be correlated with
solution phase properties.2,4 While such correlations have not
been reliable for noncovalent complexes formed principally via
hydrophobic interactions,12 good agreement has been found in
many cases of noncovalent complexes which are easily ionized
by virtue of being charged.13 Among these are noncovalent
complexes formed via hydrogen bonding, coordination
bonding, and electrostatic interactions.5,13
Table 1. Kinetic parameters derived from Biacore
measurements
a
entry
E1/2 (%) kon (103 M−1s−1
)
koff (103 s−1
)
KD (μM)
f{2,4}
f{3,3}
g{3,3}
f{3,4}
m{3,3}
h{3,3}
t{3,3}
20.2
18.6
17.4
16.9
16.3
16.0
15.4
2.11 0.03
2.99 0.04
4.80 0.06
1.39 0.11
1.61 0.03
4.59 0.11
2.24 0.06
1.31 0.03
1.48 0.02
2.29 0.02
3.58 0.23
7.73 0.05
6.76 0.07
8.10 0.45
0.62 0.03
0.49 0.01
0.48 0.01
2.57 0.37
4.79 0.12
1.47 0.05
3.62 0.31
a
KD = koff /kon.
For this purpose, we individually synthesized several bis-
NTA ligands covering a broad range of E1/2 values: f{2,4},
f{3,3}, f{3,4}, g{3,3}, h{3,3}, m{3,3}, and t{3,3}. The kinetic
parameters of each bis-NTA ligand in solution were determined
by Biacore measurement of the interaction of the candidate bis-
NTA species with biotinylated-hexahistidine immobilized onto
an avidin-conjugated substrate (Table 1). Figure 3 shows good
agreement between relative koff and E1/2 values: the higher the
stability of a histidine complex in the gas phase, the lower the
kinetic dissociation rate of the corresponding hexahistidine
complex in the solution phase. On the other hand, kon did not
correlate with E1/2 (Figure 3), presumably because the ion trap
experiment does not probe the complex formation step. The
correlation of E1/2 with KD was therefore not as strong as with
off-rate. This observation indicates that care must be taken
when analyzing the stability of a noncovalent complex in the
gas phase evaluated by collision activated dissociation because
the results do not always reflect the overall dissociation
constant KD.
The ligand possessing the slowest off-rate, f{2,4}, was
fabricated into a fluorescent probe to target a His-tagged
protein. Biotin-modified f{2,4} (Figure 4a) was synthesized and
its binding parameters to His-tag determined as above (see
Supporting Information). The resulting koff value was 1.00 ×
10−3 s−1, slightly lower than that of the parent f{2,4} (1.31 ×
10−3 s−1). The biotinylated ligand was then immobilized onto a
straptavidin-coated quantum dot. The resulting fluorescent
probe was tested in a single-molecule myosin V movement
assay because off-rate is a crucial factor for labeling his-tagged
proteins under dilute conditions. Myosin V is a dimeric motor
protein that transports organelles along actin filaments in
cells.14 Myosin V with a His-tag at the N-terminus was treated
with the f{2,4}-based probe and observed under an optical
microscope in the presence of ATP. As shown in Figure 4b, we
Figure 3. The relationship between E1/2 and the kinetic parameters
(kon and koff). The kinetic parameters were determined using a Biacore
X-100. f{2,4}(the highest E1/2), f{3,3}, g{3,3}, f{3,4}, m{3,3}, h{3,3}
and t{3,3}, were plotted from right to left in decreasing order of the
E1/2 values.
successfully observed single Q-dots moving unidirectionally
along an actin filament (see Movie S1 in the Supporting
Information). The average velocity (453 nm s−1, n = 72) and
the average run length (741 nm, n = 72) are comparable to
previous observations (Figure 4c and 4d).15 These single-
molecule observations, which are usually performed at low
concentrations of protein (i.e., 1−10 nM), indicate that f{2,4}
possesses a koff value that is low enough to label the active
motor protein.
In summary, the mass spectroscopic screening method
described in this paper identified an asymmetric ligand with
high affinity for a hexahistidine tag from a pooled library of
mixed ligands synthesized by click chemistry. The isolation and
453
dx.doi.org/10.1021/co300028n | ACS Comb. Sci. 2012, 14, 451−455

ACS Combinatorial Science
Letter
various levels of hierarchy”, a Grant-in-Aid for Scientific
Research on Innovative Areas and “High-Tech Research
Center” Project for Waseda University from MEXT, Japan.
We thank Prof. H. Kurumizaka for providing access to Biacore
X-100. We also thank Dr. Y. Lu, National University of
Singapore and Dr. T. Shibue, Waseda University for helpful
discussions.
REFERENCES
■
(1) (a) Miyazaki, I.; Simizu, S.; Okumura, H.; Takagi, S.; Osada, H. A
small-molecule inhibitor shows that pirin regulates migration of
melanoma cells. Nat. Chem. Biol. 2010, 6, 667−673. (b) Ahn, Y. H.;
Chang, Y. T. Tagged small molecule library approach for facilitated
chemical genetics. Acc. Chem. Res. 2007, 40, 1025−1033.
(2) Hofstadler, S. A.; Sannes-Lowery, K. A. Applications of ESI-MS in
drug discovery: Interrogation of noncovalent complexes. Nat. Rev.
Drug Discovery 2006, 585−595.
(3) (a) Chen, P. Electrospray ionization tandem mass spectrometry
in high-throughput screening of homogeneous catalysts. Angew. Chem.,
Int. Ed. 2003, 42, 2832−2847. (b) Wassenaar, J.; Jansen, E.; van Zeist,
W. J.; Bickelhaupt, F. M.; Siegler, M. A.; Spek, A. L.; Reek, J. N. H.
Catalyst selection based on intermediate stability measured by mass
spectrometry. Nat. Chem. 2010, 2, 417−421.
(4) (a) Loo, J. A. Studying noncovalent protein complexes by
electrospray ionization mass spectrometry. Mass Spectrom. Rev. 1997,
16, 1−23. (b) Jecklin, M. C.; Schauer, S.; Dumelin, C. E.; Zenobi, R.
Label-free determination of protein−ligand binding constants using
mass spectrometry and validation using surface plasmon resonance
and isothermal titration calorimetry. J. Mol. Recognit. 2009, 22, 319−
329.
Figure 4. Observation of single myosin V molecules. (a) Biotin-
modified f{2,4} (b) Top left: actin filaments. Top right: Q-dots
moving along the actin filaments. Video frames were integrated for six
seconds. Four moving Q-dots appeared as traces (white arrows).
Bottom panel: kymograph along an actin filament in the top panel (*).
Horizontal scale bars, 2 μm. Vertical scale bar, 1 sec. Brightness and
contrast of images have been artificially enhanced. (c) Distribution of
the running velocity. The thick line shows the fit to a Gaussian peaked
at 453 nm s−1. (d) Distribution of the run length. The thick line shows
the fit to a single exponential, p(x) = 21 exp(-x/741), where p(x) is the
probability of myosin V traveling the distance x.
(5) (a) Arai, S.; Ohkawa, H.; Ishihara, S.; Shibue, T.; Takeoka, S.;
Nishide, H. Porphyrin capped with calix[4]arene derivative via
hydrogen bonds. Bull. Chem. Soc. Jpn. 2005, 78, 2007. (b) Arai, S.;
Ohkawa, H.; Ishihara, S.; Takeoka, S.; Sibue, T.; Nishide, H. Stability
of porphyrin and calix[4]arene complexes analyzed by electrospray
ionization mass spectrometry. Rapid Commun. Mass Spectrom. 2004,
18, 2065−2068.
(6) Hochuli, E.; Dobeli, H.; Schacher, A. New metal chelate
adsorbent selective for proteins and peptides containing neighbouring
histidine residues. J. Chromatogr. 1987, 411, 177−184.
testing of candidate species in the ion trap mass spectrometer
allows the evaluation of complex mixtures of asymmetric
ligands without having to synthesize the individual species. The
gas-phase stabilities of a set of complexes agreed with their
dissociation rate constants in solution. The mass spectrometry
screening method also has the capability to significantly
improve drug development because off-rate is generally
considered an effective parameter for predicting the pharmaco-
logical effect.16 Furthermore, this is a particularly suitable tool
in cases where the structure of the peptide or protein of interest
is unpredictable. In order to make this method generic, other
combinations of proteins and small molecules that rely on
nonmetal interactions and have broad affinity ranges will be
selected and examined in the future.13
(7) (a) Kapanidis, A. N.; Ebright, Y. W.; Ebright, R. H. Site-specific
incorporation of fluorescent probes into protein: hexahistidine-tag-
mediated fluorescent labeling with (Ni2+:Nitrilotriacetic acid) n-
fluorochrome conjugates. J. Am. Chem. Soc. 2001, 123, 12123−
12125. (b) Lata, S.; Reichel, A.; Brock, R.; Tampe, R.; Piehler, J. High-
affinity adaptors for switchable recognition of histidine-tagged
proteins. J. Am. Chem. Soc. 2005, 127, 10205−10215. (c) Huang, Z.
H.; Park, J. I.; Watson, D. S.; Hwang, P.; Szoka, F. C. Facile synthesis
of multivalent nitrilotriacetic acid (NTA) and NTA conjugates for
analytical and drug delivery applications. Bioconjug. Chem. 2006, 17,
1592−1600. (d) Huang, Z. H.; Hwang, P.; Watson, D. S.; Cao, L.;
Szoka, F. C. Tris-nitrilotriacetic acids of subnanomolar affinity toward
hexahistidine tagged molecules. Bioconjug. Chem 2009, 20, 1667−
1672. (e) Brellier, M.; Barlaam, B.; Mioskowski, C.; Baati, R. Insight
into the complexation mode of Bis-(nitrilotriacetic acid) (NTA)
ligands with Ni2+ involved in the labeling of histidine tagged proteins.
Chem.Eur. J. 2009, 15, 12689−12701.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures; synthesis of compounds, the
characterizations, the detailed ESI-MS results and the micro-
scope observation (supporting movies). This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Tel: (+81) 3-5369-7324. Fax: (+81) 3-5369-7324. E-mail:
takeoka@waseda.jp.
■
(8) Knecht, S.; Ricklin, D.; Eberle, A. N.; Ernst, B. Oligohis-tags:
Mechanisms of binding to Ni2+−NTA surfaces. J. Mol. Recognit. 2009,
22, 270−279.
(9) Kolb, H. C.; Sharpless, K. B. The growing impact of click
chemistry on drug discovery. Drug Discovery Today 2003, 8, 1128−
1137.
(10) Tokuhisa, H.; Kanesato, M. Soluble 1D coordination polymers
based on dendron-functionalized bispyridine ligand for linking
between immobilized molecules on substrates. Langmuir 2005, 21,
9728−9732.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This study was partially supported by Special Coordination
Funds for Promoting Science and Technology “Development
of the advanced techniques for imaging molecular dynamics at
454
dx.doi.org/10.1021/co300028n | ACS Comb. Sci. 2012, 14, 451−455

ACS Combinatorial Science
Letter
(11) (a) Wan, K. T.; Gross, M. L.; Shibue, T. Gas-phase stability of
double-stranded oligodeoxynucleotides and their noncovalent com-
plexes with DNA-binding drugs as revealed by collisional activation in
an ion trap. J. Am. Soc. Mass. Spectrom. 2000, 11, 450−457. (b) Wan,
K. T.; Shibue, T.; Gross, M. L. Non-covalent complexes between
DNA-binding drugs and double-stranded oligodeoxynucleotides: a
study by ESI ion-trap mass spectrometry. J. Am. Chem. Soc. 2000, 122,
300−307.
(12) Aplin, R. T.; Robinson, C. V.; Schofield, C. J.; Westwood, N. J.
Does the observation of noncovalent complexes between biomolecules
by electrospray ionisation mass spectrometry necessarily reflect
specific solution interactions? J. Chem. Soc. Chem. Commun. 1994,
2415−2417.
(13) (a) Cheng, X.; Chen, R.; Bruce, J. E.; Schwartz, B. L.; Anderson,
G. A.; Hofstadler, S. A.; Gale, D. C.; Smith, R. D. Using electrospray
ionization FTICR mass spectrometry to study competitive binding of
inhibitors to carbonic anhydrase. J. Am. Chem. Soc. 1995, 117, 8859−
8860. (b) Cheng, X.; Gao, Q.; Smith, R. D.; Simanek, E. E.; Mammen,
M.; Whitesides, G. M. Characterization of hydrogen-bonded
aggregates in chloroform by electrospray ionization mass spectrom-
etry. J. Org. Chem. 1996, 61, 2204−2206.
(14) (a) Cheney, R. E.; O’Shea, M. K.; Heuser, J. E.; Coelho, M. V.;
Wolenski, J. S.; Espreafico, E. M.; Forscher, P.; Larson, R. E.;
Mooseker, M. S. Brain myosin-V is a two-headed unconventional
myosin with motor activity. Cell 1993, 75, 13−23. (b) Mehta, A. D.;
Rock, R. S.; Rief, M.; Spudich, J. A.; Mooseker, M. S.; Cheney, R. E.
Myosin-V is a processive actin-based motor. Nature 1999, 400, 590−
593.
(15) (a) Pierobon, P.; Achouri, S.; Courty, S.; Dunn, A. R.; Spudich,
J. A.; Dahan, M.; Cappello, G. Velocity, processivity, and individual
steps of single myosin V molecules in live cells. Biophys. J. 2009, 96,
4268−4275. (b) Kubota, H.; Ishikawa, R.; Ohki, T.; Ishizuka, J.;
Mikhailenko, S. V.; Ishiwata, S. Modulation of the mechano-chemical
properties of myosin V by drebrin-E. Biochem. Biophys. Res. Commun.
2010, 400, 643−648.
(16) Copeland, R. A.; Pompiano, D. L.; Meek, T. D. Drug−target
residence time and its implications for lead optimization. Nat. Rev.
Drug Discovery 2006, 5, 730−739.
455
dx.doi.org/10.1021/co300028n | ACS Comb. Sci. 2012, 14, 451−455