Multifunctional and highly sensitive precolumn reagents for amino acids in liquid chromatography/tandem mass spectrometry

Article abstract of DOI:10.1021/ac900470w

We have developed novel precolumn derivatization reagent, p-N,N,N-trimethylammonioanilyl N′-hydroxysuccinimidyl carbamate iodide (TAHS), for sensitive analyses of amino acids using high-performance liquid chromatography/electrospray ionization tandem mass

Full text of DOI:10.1021/ac900470w

Anal. Chem. 2009, 81, 5172–5179
Multifunctional and Highly Sensitive Precolumn
Reagents for Amino Acids in Liquid
Chromatography/Tandem Mass Spectrometry
Kazutaka Shimbo,† Akihisa Yahashi,† Kazuo Hirayama,†,§ Masakazu Nakazawa,‡ and
Hiroshi Miyano*,†
Institute of Life Sciences, and AminoScience Laboratories, Ajinomoto Company,
Incorporated, 1-1 Suzukicho Kawasaki-ku, Kawasaki 210-8681 Japan
We have developed novel precolumn derivatization re-
agent, p-N,N,N-trimethylammonioanilyl N′-hydroxysuc-
cinimidyl carbamate iodide (TAHS), for sensitive analyses
of amino acids using high-performance liquid chroma-
tography/electrospray ionization tandem mass spectrom-
etry (LC/ESI-MS/MS). TAHS, an activated carbamate, was
reacted briefly with the amino group to form a ureide bond
under mild condition. The derivatives provided selective
cleavage at the binding site between the reagent and the
amino acid in the collision cell of the mass spectrometer
and produced a characteristic fragment derived from the
reagent moiety. Using the precursor ion scan mode of the
tandem mass spectrometry, amino acids derivatized with
the reagents were simultaneously measured on the chro-
matogram. Selective cleavage also enabled the straight-
forward isotope ratio analysis of amino acids by the
selected reaction monitoring mode, which was applicable
in 13C metabolic flux analysis. TAHS, which contains
a cationic quaternary amine, achieved subfemtomole
to attomole levels of amino acids detection by mea-
surement in the selected reaction monitoring mode.
We also synthesized trideuteriummethyl-substituted
TAHS, TAHS-d3, and demonstrated that the combina-
tion of TAHS and TAHS-d3 is useful in comparing
amino acid concentrations between two different
samples using a single LC/MS/MS measurement.
of HPLC pumping performance and the development of more
effective column resins, although the sensitivity is not very high
and the detection limits are a few picomoles.
In addition to the ninhydrin method, a variety of other
techniques for converting amino acids to sensitive analyzable
fluorescent derivatives have been developed since the late 1970s.
The reagents used to produce the derivatives include o-phthala-
ldehyde (OPA),2 9-fluorenylmethylchloroformate (FMOC-Cl),3
5-(dimethylamino) naphthalene-1-sulfonyl chloride (dansyl-Cl),4
7-fluoro-4-nitorobenzo-2oxa-1,3-diazole (NBD-F),5 and 6-amino-
quinolyl-N-hydroxysuccinimidyl carbamate (AQC).6 The sensitivity
of detection is within the range of subpicomoles to femtomoles.
The mass spectrometer (MS) plays an important role as the
detector of HPLC, not only in protein analysis (proteome) but also
metabolite analysis (metabolome). One advantage of using an
HPLC/mass spectrometer (LC/MS) system is its selectivity,
which can distinguish analytes not only by their retention times
but also by their m/z (mass-to-charge ratio) values. HPLC/tandem
mass spectrometry (LC/MS/MS) has increased selectivity and
sensitivity. Tandem mass spectrometry enables specific detection
of an analyte using both the precursor ion and the product ion in
a collision-induced dissociation, and decreases the noise level,
which improves the limits of detection.
Recently, amino acid analysis by LC/MS has been reported.
Piraud et al. separated approximately 80 underivatized amino acids
by using ion-paring reversed-phase LC/MS/MS for the diagnosis
of inherited disorders of amino acid metabolism.7 Precolumn
derivatization reagents for amino acid analyses using LC/MS or
LC/MS/MS were also developed, mainly to achieve greater
sensitivity and selectivity. The reagents used include N-alkylni-
Several high-performance liquid chromatography (HPLC)
methods to analyze amino acids have been developed over 50
years. In 1958, Moore et al. developed an automatic amino acid
analyzer system for the colorimetric detection of amino acids; a
purple color was produced after reaction of amino acids with
ninhydrin reagent, and a cation-exchange resin was used to
separate the amino acids in a stepwise procedure by gradually
raising the pH of the citric acid buffer solution.1 The analyzer is
now capable of assaying not only protein hydrolysates but also
amino acids in biological fluids, with the progressive improvement
(2) Roth, M. Anal. Chem. 1971, 43, 880–882
(3) Einarsson, S.; Josefsson, B.; Lagerkvist, S. J. Chromatogr. 1983, 282, 609–
618
(4) Zanetta, J. P.; Vincendon, G.; Combos, G. J. Chromatogr. 1970, 51, 441–
458
(5) Watanabe, Y.; Imai, K. J. Chromatogr. 1982, 239, 723–732
(6) Cohen, S.; Michaud, D. Anal. Biochem. 1993, 211, 279–287
(7) Piraud, M.; Vianey-Saban, C.; Petritis, K.; Elfakir, C.; Steghens, J.; Bouchu,
D. Rapid Commun. Mass Spectrom. 2005, 19, 1587–1602
(8) Yang, W.-C.; Mirzaei, H.; Liu, X.; Regnier, F. Anal. Chem. 2006, 78, 4702–
4708
.
.
.
.
.
* To whom correspondence should be addressed. Phone: +81-44-245-5067.
Fax: +81-44-211-7609. E-mail: hiroshi_miyano@ajinomoto.com.
† Institute of Life Sciences.
.
.
‡ AminoScience Laboratories.
(9) Ross, P.; Huang, Y.; Marchese, J.; Williamson, B.; Parker, K.; Hatten, S.;
Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.;
Martin, S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. Mol. Cell.
§ Current address: Science Education Co., Inc., 2-13-16, Kamata, Ohtaku,
Tokyo, 144-0052 Japan.
(1) Moore, S.; Spackman, D.; Stein, W. Anal. Chem. 1958, 30, 1185–1190
.
Proteomics 2004, 3, 1154–1169.
5172 Analytical Chemistry, Vol. 81, No. 13, July 1, 2009
10.1021/ac900470w CCC: $40.75  2009 American Chemical Society
Published on Web 05/29/2009

cotinic acid, N-hydroxysuccinimide ester,8 and iTRAQ (isobaric
tag for relative and absolute quantitation).9
Scheme 1. Synthesis of TAHS
We have recently focused on developing new types of precol-
umn derivatization reagents of the amino group, for use in LC/
MS/MS. One of our most successful reagents, p-N,N,N-trimeth-
ylammonioanilyl N′-hydroxysuccinimidyl carbamate iodide (TAHS),
achieved subfemtomole to attomole levels of amino acids detec-
tion. Amino acids derivatized with TAHS were selectively cleaved
at the binding site between TAHS and the amino acid in the
collision cell of the triple-stage quadrupole mass spectrometer.
Amino acid analyses, such as simultaneous analysis of compounds
with amino groups and biosynthetic pathway studies, including
13C metabolic flux analysis, have been performed.10,11
Scheme 2. Reaction of Amino Acids with TAHS
Reagent
EXPERIMENTAL SECTION
Chemicals. The amino acid standard mixture solution, type
H (mixture of
L
-lysine,
-serine, -glutamic acid,
-valine, -methionine, -isoleucine,
-phenylalanine, with each amino acid at a 2.5 mM concentra-
tion) was purchased from Wako Pure Chemical Industries, Ltd.
(Osaka, Japan). -Tryptophan, -glutamine, and -asparagine were
L
-histidine,
L
-arginine,
-proline, -glycine,
-leucine,
L
-aspartic acid,
-alanine,
-tyrosine,
L
-threonine,
-cystine,
L
L
L
L
L
L
L
L
L
L
L
and
L
1H NMR, 400 MHz (JEOL R400, Japan) (CD3CN, ppm,
tetramethylsilane as an internal standard) δ7.22 (d, 9.0 Hz, 2
H), δ6.72 (d, 9.0 Hz, 2 H), δ2.88 (s, 6 H), δ2.76 (s, 4 H); MS
(micromass Q-Tof-2, Manchester, U.K.), m/z 278.1141 [M +
H]+, molecular formula, C13H16N3O4 (∆ -4.4 ppm).
p-N,N-Dimethylaminoanilyl N′-hydroxysuccinimidyl carbamate
(264 mg, 1.1 mmol) was dissolved in 10 mL of acetonitrile/
dichloromethane (4:1) at room temperature. Iodomethane (0.4 mL,
8 equiv) was added to the solution, which was then stirred for
23 h at room temperature. After the reaction mixture was filtered,
354 mg of TAHS was obtained (76% yield) (Scheme 1).
1H NMR, 400 MHz (JEOL R400, Japan) (DMSO-d6, ppm,
tetramethylsilane as an internal standard) δ7.97 (d, 8.4 Hz, 2
H), δ7.62 (d, 8.4 Hz, 2 H), δ3.58 (s, 9 H), δ2.83 (s, 4 H); MS
(micromass Q-Tof-2, Manchester, U.K.), m/z 292.1297 [M]+,
molecular formula, C14H18N3O4 (∆ -2.9 ppm).
Synthesis of p-N,N-Dimethyl,N-trideuteromethylammo-
nioanilyl N′-Hydroxysuccinimidyl Carbamate Iodide (TAHS-
d3). p-N,N-Dimethylaminoanilyl N′-hydroxysuccinimidyl carbam-
ate (264 mg, 1.1 mmol) was dissolved in 10 mL of acetonitrile/
dichloromethane (4:1) at room temperature. Iodomethane-d3 (0.4
mL, 8 equiv) was added to the solution, which was then stirred
for 50 h at room temperature to yield TAHS-d3.
L
L
L
purchased from Sigma-Aldrich Japan (Tokyo, Japan). Stable
isotope-labeled amino acids, including 13C, 15N uniformly labeled
L-glutamine (Gln-IS), 15N uniformly labeled L-arginine (Arg-IS),
13C, 15N uniformly labeled L-histidine (His-IS), 13C, 15N uniformly
labeled L-glutamic acid (Glu-IS), 13C, 15N uniformly labeled
L-serine (Ser-IS), and 13C, 15N uniformly labeled L-tryptophan
(Trp-IS) were obtained from Ajinomoto Co., Inc. (Tokyo,
Japan). Glycine-2,2-d2 (Gly-IS) and proline-d7 (Pro-IS) were
purchased from Cambridge Isotope Laboratories (Andover,
U.S.A.). L-Alanine-3,3,3-d3 (Ala-IS), L-leucine-5,5,5-d3 (Leu-IS),
DL-lysine-4,4,5,5-d4 (Lys-IS), DL-valine-d8 (Val-IS), L-phenyl-d5-
alanine (Phe-IS), and L-methionine-d3 (methyl d3; Met-IS) were
purchased from Isotec (Tokyo, Japan). N,N-Dimethylamino-p-
phenylenediamine and N,N′-dihydroxysuccinimidyl carbonate
(DSC) were purchased from Wako Pure Chemical Industries,
Ltd. (Osaka, Japan). Iodomethane and iodomethane-d3 were
purchased from Nacalai Tesque (Kyoto, Japan) and Taiyo
Nippon Sanso Corporation (Tokyo, Japan), respectively. Acetic
acid and acetonitrile were purchased from Junsei Chemical Co.,
Ltd. (Tokyo, Japan). Deionized water was used after purification
through a Milli-Q gradient A10 System (Millipore, Bedford,
MA).
Synthesis of p-N,N,N-Trimethylammonioanilyl N′-Hydroxy-
succinimidyl Carbamate Iodide. DSC (1.28 g, 5 mmol) was
dissolved in 25 mL of acetonitrile at room temperature. N,N-
Dimethylamino-p-phenylenediamine (535 mg, 5 mmol), dissolved
in 25 mL of acetonitrile, was added dropwise to the carbonate
solution over a period of approximately 2 h. After an additional
22 h of stirring, the reaction mixture was concentrated by rotary
evaporation. The residue was resuspended in 5 mL of acetonitrile
and then filtered, to obtain p-N,N-dimethylaminoanilyl N′-hydrox-
ysuccinimidyl carbamate (597 mg, 48% yield) (Scheme 1).
Derivatization of the Amino Acids Standard Solution with
TAHS. The synthesized TAHS reagent was dissolved in dry
acetonitrile to a concentration of 20 mg/mL. The mixed amino
acids solution was prepared from the dilution of the commercial
amino acid standard mixture solution (type H) and the
L-
tryptophan, -glutamine, and -asparagine solutions. A 10 µL
L
L
aliquot of the mixed amino acid solution was mixed with 30 µL of
0.2 M sodium borate buffer at pH 8.8, in a sealed 1 mL reaction
vial or 1.5 mL polypropylene micro test tube. A 10 µL aliquot of
TAHS was added to the above solution, and the mixture was
heated at 55 °C for 10 min. A 200 µL aliquot of 0.2% aqueous acetic
acid was then added to the reaction mixture before storage in a
tightly sealed container at 4 °C until LC/MS/MS analysis
(Scheme 2).
(10) Miyano, H.; Yahashi, A.; Shimbo, K.; Nakazawa, M.; Hirayama, K. United
States Patent 7,148,069 B2; Dec 12, 2006.
(11) Iwatani, S.; Van Dien, S.; Shimbo, K.; Kubota, K.; Kageyama, N.; Iwahata,
D.; Miyano, H.; Hirayama, K.; Usuda, Y.; Shimizu, K.; Matsui, K. J. Bio-
technol. 2007, 128, 93–111
.
Analytical Chemistry, Vol. 81, No. 13, July 1, 2009 5173

Preparation of Plasma Samples. Blood samples were
obtained from Sprague-Dawley (SD) rats (8 week old, male). The
samples were collected in tubes containing EDTA as an antico-
agulant (NONCLOT-D; Daiichi Pure Chemicals, Tokyo, Japan)
and centrifuged for 10 min at 3000 rpm, at 4 °C (model H-103N;
Kokusan Centrifuge Ltd., Tokyo, Japan). Plasma samples were
collected, pooled, and then stored at -80 °C until analysis.
A 10 µL aliquot of the pooled plasma was added to the 10 µL
aliquot of internal standard solution (isotope-labeled amino acids
mixture solution) and diluted with 80 µL of Milli-Q water and
deproteinized to a final concentration of 50% acetonitrile. The
mixture was vortexed and centrifuged for 10 min at 15 000 rpm.
A 10 µL aliquot of the deproteinized supernatant, in a sealed 1
mL reaction vial or 1.5 mL polypropylene micro test tube, was
mixed with 30 µL of 0.2 M sodium borate buffer at pH 8.8. A 10
µL aliquot of the reagent solution was added to the solution, and
the mixture was heated at 55 °C for 10 min. A 200 µL aliquot of
0.2% aqueous acetic acid was added to the reaction mixture. The
mixture was diluted 10-fold with 0.2% aqueous acetic acid before
storage in a tightly closed container at 4 °C until LC/MS/MS
analysis.
E. coli Extract Sample. The E. coli extract sample, cultivated
with 13C-labeled glucose, was used for isotope ratio analysis of
amino acid. The cultivation and the extraction conditions are
described in Iwatani et al.11 A 10 µL aliquot of the TAHS
reagent solution was added to the isotope-labeled extract
solution; the mixture was then heated at 55 °C for 10 min. A
200 µL aliquot of 0.2% aqueous acetic acid was added to the
reaction mixture before storage in a tightly closed container
at 4 °C until LC/MS/MS analysis.
Instrumentation. An Agilent 1100 series liquid chromatog-
raphy system (Agilent Technologies, Waldbrunn, Germany) with
a binary pump, degasser, autosampler, column compartment, and
UV detector was used.
The system was coupled to a triple-quadrupole mass spec-
trometer, Applied Biosystems Sciex API 4000 (Applied Biosystems-
MDS Sciex, Concord, Canada) equipped with a TurboIonSpray
interface. The Applied Biosystems-MDS Sciex Analyst 1.4.2
software was used to control these instruments.
LC/MS/MS of the Precursor Ion Scan by Simultaneous
Analysis of TAHS-Tagged Amino Acids. The TAHS-tagged
amino acids were injected into a column of Inertsil ODS-3, 3 µm
particle size, 100 mm × 2.1 mm i.d. (GL Sciences, Tokyo, Japan).
The column temperature was maintained at 40 °C throughout the
analysis. Mobile phase A consisted of 0.2% acetic acid in Milli-Q-
water (v/v), and mobile phase B consisted of 0.2% acetic acid in
acetonitrile. Each mobile phase was filtered through a 0.2 µm
membrane (Nalge Nunc International, Rochester, NY) before use.
The gradient conditions (B %) were 0-0.10 min ) 0%, 0.11-15
min ) 5-12% (linear), 15-18 min ) 12%, and 18-22 min ) 70%.
Re-equilibration was performed for 8 min at 100% of mobile phase
A. The flow rate was 0.2 mL/min. Mobile phase B was introduced
just after the column outlet as the sheath solution, at a flow rate
of 50 µL/min. The injection volume was 3 µL.
Table 1. Selected Monitor Ions and Correlation
Coefficients and Detection Limits for Each
TAHS-Derivatized Amino Acid
amino
acid
linearity
range
LOD
Q1
Q3
[R2]
[nmol/mL] (S/N ) 3) fmola
glycine
252.1 177.1 0.993
266.1 177.1 0.991
282.1 177.1 0.994
292.1 177.1 0.990
294.1 177.1 0.995
296.1 177.1 0.997
308.1 177.1 0.997
308.1 177.1 0.998
309.1 177.1 0.995
0.25-10
0.05-10
0.01-10
0.01-10
0.01-25
0.01-10
0.01-25
0.01-10
0.01-10
0.01-250
0.01-10
0.01-50
0.025-100
0.01-100
0.01-10
0.01-10
0.01-10
0.01-25
0.01-25
0.025-250
0.10
0.05
0.10
0.05
0.11
0.07
0.10
0.10
0.09
0.25
0.06
0.07
0.15
0.30
0.16
0.05
0.14
0.06
0.11
0.34
alanine
serine
proline
valine
threonine
isoleucine
leucine
asparagine
aspartic acid 310.1 177.1 0.998
glutamine
lysine
323.2 177.1 0.995
250.3 177.1 1.000
glutamic acid 324.1 177.1 0.997
methionine
histidine
326.1 177.1 0.999
166.8 177.1 0.998
phenylalanine 342.2 177.1 0.999
arginine
tyrosine
tryptophan
cystine
176.1 177.1 0.998
358.2 177.1 0.997
381.2 177.1 0.993
297.8 177.1 0.999
a LOD: limit of detection was calculated from a signal-to-noise ratio
of 3 after the Savitzky-Golay smoothing.
177 amu, which was derived from the TAHS moiety, and the scan
range for the precursor ion scan was set to 150-500 amu, with
2.5 s for the scan time.
LC/MS/MS Selected Reaction Monitoring for Quantitative
Analysis of Amino Acid Derivatives. Amino acid derivatives
were injected into a column of Capcell Pack MGIII, 3 µm particle
size, 100 mm × 2.0 mm i.d. (Shiseido Co., Ltd., Japan). The column
temperature was maintained at 40 °C throughout the analysis.
Other conditions, including injection volume, gradient conditions,
flow rate, and composition of the sheath solution for the analysis
of amino acids derivatized with TAHS were the same as those of
the precursor ion scan by simultaneous analysis of the TAHS-
tagged amino acids.
The TurboIonSpray interface was operated in the positive mode
at 5000 V and 700 °C. Other MS parameters for TAHS-tagged
amino acids, CUR, GS1, GS2, CAD, DP, EP, CE, and CXP, were
set to 10, 30, 30, 5, 50, 10, 20, and 15, respectively. Quantitative
analysis of the amino acids was performed by selected reaction
monitoring (SRM). The monitored SRM channels for the TAHS-
tagged amino acids are summarized in Table 1.
LC/MS/MS for Isotope Ratio Analysis of Amino Acids by
TAHS. The HPLC conditions were the same as those of the
precursor ion scan by simultaneous analysis of TAHS-tagged
amino acids.
The TurboIonSpray interface was operated in the positive mode
at 5000 V and 700 °C. Other MS parameters, CUR, GS1, GS2,
CAD, DP, EP, CE, and CXP, were set to 15, 50, 30, 5, 71, 10, 29,
and 15, respectively. Isotope ratio analysis of amino acids was
performed by SRM. The monitored SRM channels are sum-
marized in Table 2.
Comparative Analysis of Amino Acids between Samples
The TurboIonSpray interface was operated in the positive mode
at 5000 V and 700 °C. Other MS parameters, CUR, GS1, GS2,
CAD, DP, EP, CE, and CXP, were set to 10, 70, 30, 5, 50, 10, 20,
and 14, respectively. The precursor ion monitoring was set to m/z
Using TAHS and TAHS-d3. Two aliquots of the same sample,
10 µM amino acid mixture, were derivatized with the TAHS
reagent and the TAHS-d3 reagent, respectively, and then mixed
at a 1:1 ratio.
5174 Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

Table 2. Selected Monitor Ions for Isotope Ratio Analysis and Calculated Theoretical Natural Isotope Ratio and
Measured Isotope Ratio of TAHS-Derivatized Amino Acids of the Standard Amino Acid Solution and the E. coli
Extract Samplea b
,
amino acid
Q1
Q3 calcd [%] measd [%] E. coli extract [%] amino acid Q1
Q3 calcd [%] measd [%] E. coli extract [%]
Glycine
M
M + 1
252.1 177.1
253.1 177.1
96.9
2.7
97.0
2.6
65.5
12.0
M + 2
M + 3
254.1 177.1
255.1 177.1
0.4
0.0
0.5
0.0
21.6
0.8
Alanine
M
M + 1
M + 2
266.1 177.1
267.1 177.1
268.1 177.1
95.8
3.7
0.4
95.4
4.0
0.6
32.0
20.9
4.9
M + 3
M + 4
269.1 177.1
270.1 177.1
0.0
0.0
0.0
0.0
42.0
0.3
Serine
M
M + 1
M + 2
282.1 177.1
283.1 177.1
284.1 177.1
95.6
3.8
0.6
95.2
3.9
0.7
35.6
22.2
9.6
M + 3
M + 4
285.1 177.1
286.1 177.1
0.0
0.0
0.1
0.0
32.0
0.7
Proline
M
M + 1
M + 2
292.1 177.1
293.1 177.1
294.1 177.1
93.7
5.8
0.5
93.0
6.2
0.6
54.1
9.3
9.8
M + 3
M + 4
M + 5
295.1 177.1
296.1 177.1
297.1 177.1
0.0
0.0
0.0
0.1
0.0
0.0
11.3
9.6
5.5
Valine
M
M + 1
M + 2
294.1 177.1
295.1 177.1
296.1 177.1
93.6
5.8
0.5
92.9
6.5
0.6
84.1
7.0
2.7
M + 3
M + 4
M + 5
297.1 177.1
298.1 177.1
299.1 177.1
0.0
0.0
0.0
0.0
0.0
0.0
2.7
1.3
2.3
Threonine
M
M + 1
M + 2
296.1 177.1
297.1 177.1
298.1 177.1
94.5
4.8
0.7
93.9
5.3
0.8
88.7
6.0
2.0
M + 3
M + 4
M + 5
299.1 177.1
300.1 177.1
301.1 177.1
0.0
0.0
0.0
0.0
0.0
0.0
1.7
1.5
0.1
Isoleucine
M
M + 1
M + 2
308.1 177.1
309.1 177.1
310.1 177.1
92.6
6.8
0.6
92.3
7.0
0.6
87.6
7.3
3.9
M + 3
M + 4
M + 5
311.1 177.1
312.1 177.1
313.1 177.1
0.0
0.0
0.0
0.0
0.0
0.0
0.4
0.4
0.2
Leucine
M
M + 1
M + 2
308.1 177.1
309.1 177.1
310.1 177.1
92.6
6.8
0.6
92.1
7.2
0.7
90.3
7.2
1.7
M + 3
M + 4
M + 5
311.1 177.1
312.1 177.1
313.1 177.1
0.0
0.0
0.0
0.0
0.0
0.0
0.2
0.2
0.1
Aspartic Acid
M
310.1 177.1
311.1 177.1
312.1 177.1
94.3
4.8
0.9
94.9
4.2
0.8
39.3
15.7
11.5
M + 3
M + 4
313.1 177.1
314.1 177.1
0.0
0.0
0.1
0.0
16.7
16.8
M + 1
M + 2
Glutamic Acid
M
324.1 177.1
325.1 177.1
326.1 177.1
93.2
5.8
0.9
91.6
6.8
1.1
7.3
11.6
19.3
M + 3
M + 4
M + 5
327.1 177.1
328.1 177.1
329.1 177.1
0.1
0.0
0.0
0.4
0.1
0.0
30.7
17.4
13.7
M + 1
M + 2
Methionine
M
M + 1
M + 2
326.1 177.1
327.1 177.1
328.1 177.1
89.0
6.2
4.5
89.1
6.3
4.2
30.5
18.0
14.7
M + 3
M + 4
M + 5
329.1 177.1
330.1 177.1
331.1 177.1
0.3
0.0
0.0
0.3
0.1
0.0
16.5
10.2
10.1
Phenylalanine
M
M + 1
M + 2
M + 3
M + 4
342.1 177.1
343.1 177.1
344.1 177.1
345.1 177.1
346.1 177.1
89.6
9.6
0.8
0.1
0.0
89.3
9.8
0.9
0.0
0.0
88.1
9.0
1.0
0.3
0.3
M + 5
M + 6
M + 7
M + 8
M + 9
347.1 177.1
348.1 177.1
349.1 177.1
350.1 177.1
351.1 177.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.4
0.3
0.3
0.2
0.2
Tyrosine
M
M + 1
M + 2
M + 3
M + 4
358.2 177.1
359.2 177.1
360.2 177.1
361.2 177.1
362.2 177.1
89.3
9.6
1.0
0.1
0.0
89.1
9.8
1.1
0.1
0.0
tr
tr
tr
tr
tr
M + 5
M + 6
M + 7
M + 8
M + 9
363.2 177.1
364.2 177.1
365.2 177.1
366.2 177.1
367.2 177.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
tr
tr
tr
tr
tr
a The theoretical natural isotope ratio of each amino acids was calculated by using MS-Isotope in ProteinProspector (http://prospector.ucsf.edu/)
site, University of California, San Francisco. b tr: trace.
HPLC conditions were the same as those of the precursor ion
scan for simultaneous analysis of the TAHS-tagged amino acids.
The TurboIonSpray interface was operated in the positive
mode at 5000 V and 700 °C. The other MS parameters, CUR,
GS1, GS2, CAD, DP, EP, CE, and CXP, were set to 10, 50, 30,
5, 50, 10, 25, and 14, respectively. The monitored SRM channel
for TAHS-d3 of the Q3 channel was 180.1, and the Q1
channels for glycine, alanine, serine, proline, valine, threo-
nine, isoleucine, leucine, asparagine, aspartic acid, glutamine,
lysine, glutamic acid, methionine, histidine, phenylalanine,
Analytical Chemistry, Vol. 81, No. 13, July 1, 2009 5175

arginine, tyrosine, tryptophan, and cystine were set to 255.1,
269.1, 285.1, 295.1, 297.1, 299.1, 311.1, 311.1, 313.1, 323.1,
253.2, 324.1, 326.1, 168.2, 345.2, 177.7, 361.2, 384.1, and 300.1,
respectively.
RESULTS AND DISCUSSION
Design and Synthesis of Precolumn Derivatization Re-
agents for LC/MS/MS. Precolumn derivatization techniques for
amino acid analysis have been developed because they increase
detection sensitivity and selectivity of analytes. Mass spectrometry,
especially in the tandem mode, has been eagerly developed as a
powerful tool in proteome, metabolome, and pharmacokinetics
studies due to considerable improvements in its analytical sensitiv-
ity and selectivity. In this study, we developed new precolumn
derivatization reagents for amino acid analysis by liquid chroma-
tography/electrospray ionization tandem mass spectrometry (LC/
ESI-MS/MS).
The derivatization of an amino acid reacted with the reagent
was designed to provide characteristic and selective cleavage at
the bonding site between the reagent moiety and the amino acid
in the collision cell of the triple-stage quadrupole mass spectrom-
eter. Subsequently, each derivative generated the same product
ion, which was derived from the reagent moiety. We discovered
the formation of a ureide bond (R-NH-(CdO)-NH-R′), in
which R is the component of the amino compound and R′ is the
reagent moiety, facilitates analysis. To form the ureide bond with
the amino groups of the analytes, we selected an activated
carbamate-containing reagent (e.g., succinimidyl carbamate), since
carbamates are well-known for their rapid and selective reactions
with amino groups under mild conditions.
Although the reaction was completed in a minute at room
temperature, phenolic hydroxyl group was also reacted with TAHS
reagent. Further heating (more than 5 min at 55 °C) was needed
to hydrolysis of the phenolic hydroxyl group of tyrosine. And
optimized reaction pH for TAHS derivatization was pH 8.0-9.0.
In addition, a cationic group was introduced to the reagent
skeleton, to increase the ion efficiency for electrospray ionization.
The sensitivity of TAHS derivatives show 7-320 higher than the
sensitivity of p-N,N-dimethylaminoanilyl N′-hydroxysuccinimidyl
carbamate derivatives. Furthermore, a hydrophobic moiety, such
as a phenyl, naphthyl, quinoly, or pyridyl group, was combined
with the activated carbamate to increase the retention of the
derivatives and to facilitate their separation, in comparison with
the unreacted amino acids, in reversed-phase chromatography.
To achieve effective ionization, the reagents were designed and
have both phenyl group as hydrophobic moiety and cationic group
as ionic moiety.
Figure 1. Specific cleavage performed in a tandem mass spectrom-
eter. Weak collision energy produces the unique fragment ion derived
from the reagent skeleton. All TAHS derivatives of the amino acids
show the same m/z 177 product ions in the product ion scan mode:
(A) TAHS derivative of alanine, (B) TAHS derivative of aspartic acid,
(C)TAHSderivativeoflysine,and(D)TAHSderivativeofphenylalanine.
compounds in tandem mass measurements, without any method
optimization for Q3 ions.
The activated carbamates in the reagent can be substituted
with isocyanates, which also form a ureide bond with amino
groups.
LC/MS/MS Precursor Ion Scan by Simultaneous Analysis
of TAHS-Tagged Amino Acids. Figure 2 shows the total ion
chromatogram for the precursor ion scan of the amino acid
standard mixture solution (type H) derivatized with TAHS; the
product ion was set at m/z 177. All of the amino acids derivatized
with TAHS, that is, all of the analytes with amino groups, were
selectively and simultaneously monitored by the precursor ion
scan. Although the compounds with two amino groups, such as
lysine, were bound with 2 mol of TAHS, the product ions were
also set at a value of m/z 177.
The results indicated that this method, combining LC/MS/
MS and precolumn derivatization with these reagents, was not
only similar to selective amino acid analysis by using the
fluorescent reagents for the amino groups but also superior to
the method, as it includes additional information on the parent
mass number of each peak. Thus, when an unidentified peak is
observed on the chromatogram, its structure with amino group(s)
may be estimated from its mass number.
The fragmentation patterns of the derivatives were evaluated
by MS/MS product ion scans. The fragment ions from the TAHS
adducts were due to regulatory cleavage at the ureide bond of
the carbamate moiety and the amino groups and the loss of amino
acid moieties from the precursor ions under the optimized positive-
ion collision-induced dissociation conditions of [M + H]+. The
results were utilized to develop the SRM conditions, which were
programmed to include the transitions of all of the protonated
molecular ions to the common fragment at m/z 177, derived
from TAHS (Figure 1). All of the derivatives can be monitored
in the positive mode, making it easy to add the monitoring
Figure 3 shows the total ion chromatograms for the precursor
ion scan of deproteinized rat plasma and the TAHS-derivatized
5176 Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

the interaction between the basic part of the TAHS derivative and
silanols of the silica gel particles and providing the sharp peaks.
LC/MS/MS for Isotope Ratio Analysis of Amino Acids by
TAHS. Selective cleavage at the binding site between the reagent
moiety and the amino acid moiety in the derivative facilitated
isotope ratio analysis of the analyte itself, without any complex
correction calculations. When the selected ion monitoring mode
analysis of MS/MS was set at m/z of 177, derived from the TAHS
moiety of the p-N,N,N-trimethylammonioanilyl group (C10H13N2O),
the product ion distribution reflected the isotope ratio of the
amino acids themselves.
Table 2 shows the measured isotope distributions of natural
amino acids, after derivatization with TAHS, by selected ion
monitoring. A good correlation was observed between the
theoretical and observed values. For instance, the theoretical
isotope distributions of natural alanine are 95.8% (m () monoiso-
topic)), 3.7% (m + 1), and 0.5% (m + 2), whereas the observed
isotope distributions were 95.4% (m), 4.0% (m + 1), and 0.6% (m
+ 2). Correction calculations were needed to obtain the isotope
distributions with more than two amino groups, such as lysine.
We also applied this method to the biological sample, E. coli
extract, which was cultivated with 13C-labeled glucose to measure
the non-natural isotope ratio of amino acids. A difference in
isotope ratio patterns was observed between labeled and natural
amino acids with this method. Increased incorporation of 13C
atoms, derived from the 13C-labeled glucose, were observed in
amino acids more readily converted for energy metabolism in
glycolysis and the citric cycle (TCA cycle). For instance, we
observed the isotope ratio (m) for glutamic acid was only 7.3%,
suggesting rapid turnover of glutamic acid was performed in
the intracellular metabolic dynamics.
Figure 2. Total ion chromatogram of a 17 amino acids standard
mixture derivatized with TAHS reagent. Detection was achieved by
a precursor ion scan of the precursor of m/z 177; the range of m/z
was 150-500. Peak identification: 1, histidine; 2, arginine; 3, serine;
4, glycine; 5, aspartic acid; 6, threonine; 7, glutamic acid; 8, alanine;
9, lysine; 10, proline; 11, cystine; 12, valine; 13, methionine; 14,
tyrosine; 15, isoleucine; 16, leucine; 17, phenylalanine.
deproteinized rat plasma. The results show that product ion of
m/z 177 is only present in TAHS-derivatized samples.
LC/MS/MS Selected Reaction Monitoring for Quantitative
Analysis of Amino Acid Derivatives. The detection limits,
linearity, and quantitative range of TAHS-tagged amino acids were
evaluated in the selected ion monitoring mode (Table 1). The 20
different amino acids that make up proteins were used. The
detection limits of amino acids derivatized with TAHS, which were
calculated from a signal-to-noise ratio of 3 after Savitzky-Golay
smoothing using Analyst 1.4 software, were 0.05-0.34 fmol.
Ionization of the derivatives with TAHS was very efficient because
it has a trimethylammonium group: the sensitivity of the procedure
was lower than femtomole levels of amino acids. The aromatic
ring moiety of TAHS is also important in increasing the hydro-
phobicity necessary for efficient electrospray ionization.
To increase the ionization efficiency of the derivatives, we used
an exterior pump for the sheath flow. The addition of acidic
acetonitrile as the sheath flow increased the sensitivity of analytes,
especially those eluted with lower organic solvent compositions.
With sheath flow, TAHS derivatives were increased 1-2.2-fold
higher intensity. Gln and Asp derivatives increased more than
2-fold with using the sheath flow.
Table 3 shows quantitation of the rat plasma samples. The
analysis was performed six times. The relative standard deviations
of the precisions were 3-6%. The recovery rates after addition of
the 100 µM amino acids mixture to plasma samples were 94-112%.
Because of the high sensitivity of this method, samples could be
diluted before LC/MS/MS analysis so that the matrix effects were
reduced and better precisions and recovery rates were achieved.
We used a Capcell Pack MGIII column in the SRM, although
an Inertsil ODS-3 column shows similar separation results for
TAHS-tagged amino acids. However, using our method, a Capcell
Pack MGIII column shows better signal-to-noise ratio for basic
amino acids. TAHS derivative of histidine, arginine, and lysine
showed 3.5-7.2-fold higher signal-to-noise ratio. These results
might be from the highly end-capped silica gel particles reducing
Metabolic flux analysis using 13C-labeled substrates is a well-
developed method for investigating intracellular metabolite
behavior. Recently, several 13C metabolic flux analysis experi-
ments were performed, measuring intracellular amino acids,
rather than relying on proteinogenic amino acids, because of
their relatively rapid turnover. For example, Kromer et al.
utilized the labeling of intracellular amino acids, using gas
chromatography/mass spectrometry (GC/MS), to determine
the fluxes in the lysine production in different phases of a batch
cultivation of Corynebacterium glutamicum.12 Although GC/
MS has been one of the most convenient and sensitive tools
in measuring isotope distributions for this purpose, to obtain
the isotope ratio in the target compound by GC/MS, the
observation error derived from the isotope ratio of the reagent
itself has to be considered. Our method, involving selective
cleavage at the binding site between the reagent moiety and
the amino acid in the derivative, allows the isotope ratio of the
analyte to be measured, without any complex calculated
corrections.
The metabolic flux analysis experiment was very expensive
because of the 13C-labeled glucose. The high sensitivity of TAHS
will help to reduce costs as the experiments can be performed
using low-concentration intracellular amino acids. Iwatani et
al. determined the metabolic flux change during the fed-batch
cultivation of a lysine-producing strain of E. coli using mass
(12) Kromer, J. O.; Sorgenfrei, O.; Klopprogge, K.; Heinzle, E.; Wittmann, C. J.
Bacteriol. 2004, 186, 1769–1784
.
Analytical Chemistry, Vol. 81, No. 13, July 1, 2009 5177

Figure 3. Total ion chromatograms of the precursor ion scan of TAHS-derivatized deproteinized rat plasma (A) and deproteinized rat plasma
(B): 1, histidine; 2, arginine; 3, serine; 4, glycine; 5, glutamine; 6, threonine; 7, glutamic acid; 8, alanine; 9, lysine; 10, proline; 11, cystine; 12,
valine; 13, methionine; 14, tyrosine; 15, isoleucine; 16, leucine; 17, phenylalanine; 18, tryptophan.
between samples. For example, isotope-coded affinity tag (ICAT)13
Table 3. Quantitative Data for the Determination of 16
reagents have been used in proteomics. Yang et al. synthesized
isotope-labeled reagents of N-alkylnicotinic acid N-hydroxysuc-
cinimide ester, for the comparative quantification of amino acids.8
In this study, a heavy reagent of TAHS, TAHS-d3, was
synthesized from p-N,N-dimethylaminoanilyl N′-hydroxysuc-
cinimidyl carbamate and trideuterized methyl iodide.
Figure 4 shows the sample preparation scheme. One amino
acid-containing solution was derivatized with TAHS, and another
solution was separately derivatized with TAHS-d3. Equivalent
volumes of each derivatized sample were mixed, and the
mixture was analyzed using the selected ion monitoring mode
of LC/MS/MS. The detection channels of the product ions
were set to the m/z values of TAHS-tagged amino acids and
TAHS-d3-tagged amino acids. The detection channels of the
fragment ions were set to an m/z of 177 for the TAHS
derivatives and an m/z of 180 for the TAHS-d3 derivatives.
Relative concentrations of the amino acids in the two samples
were measured in a single analysis in order to compare the
peak intensities derived from the two reagents.
Amino Acids in Rat Plasma Samples (N ) 6)
concentration
in plasma [µM] RSD [%] recovery [%] standard
internal
glycine
102.7
195.6
79.6
65.5
79.0
88.8
37.7
68.3
47.8
5.9
81.9
134.0
125.8
20.6
28.8
30.2
41.6
39.2
141.1
0.7
2
2
2
1
3
3
2
3
5
2
4
4
3
2
9
2
6
3
2
8
103
102
104
93
Gly-IS
Ala-IS
Ser-IS
Pro-IS
Val-IS
Ser-IS
Leu-IS
Leu-IS
Ser-IS
Gln-IS
Lys-IS
Gln-IS
Glu-IS
Met-IS
His-IS
Phe-IS
Arg-IS
Pro-IS
Trp-IS
Lys-IS
alanine
serine
proline
valine
103
104
102
102
104
104
99
threonine
isoleucine
leucine
asparagine
aspartic Acid
lysine
glutamine
glutamic acid
methionine
histidine
phenylalanine
arginine
97
101
102
100
109
99
tyrosine
tryptophan
cystine
102
93
107
Figure 4 shows the peak area ratio of the TAHS/TAHS-d3
relative analysis. Each 10 µM amino acid mixture solution was
derivatized with the same volume of the TAHS and TAHS-d3
reagent and analyzed by LC/MS/MS. The mean ratio of six
injections for each amino acid except tyrosine was 0.92-1.18.
The ratio of the TAHS-tagged tyrosine to TAHS-d3-tagged
tyrosine was 0.77. The reason remains unexplained, and further
experiments will be needed to the accuracy of the ratio to obtain
optimized methods for this analysis.
distribution data derived from both intracellular free amino
acids and proteinogenic amino acids measured using the
precolumn LC/MS/MS method.11
Comparative Analysis of Amino Acids between TAHS and
TAHS-d3 Samples. In mass spectrometric studies, the reagents
labeled with isotopes can be used in quantitative comparisons
(13) Gygi, S.; Rist, B.; Gerber, S.; Turecek, F.; Gelb, M.; Aebersold, R. Nat.
Biotechnol. 1999, 17, 994–999
.
5178 Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

instead of 12C in the structure of the reagent, as they do not
show chromatographic isotope effects, which derive from the
interaction between the C-H or C-D bonds and the stationary
,14
phase of the reversed-phase HPLC column.8
CONCLUSIONS
We developed new derivatization reagents for amino acid
analysis using LC/ESI-MS/MS. Subfemtomole to attomole levels
of amino acid analysis were achieved using the precolumn reagent
with cationic groups, TAHS, and selected ion monitoring by LC/
MS/MS. This highly sensitive method for amino acid analysis may
be utilized to determine the intracellular concentration of amino
acids and amino acids present in biological fluids and various
organisms where their concentrations are very low. TAHS may
be utilized for highly sensitive analyses of proteomes, in combina-
tion with its deuterium isomer, TAHS-d3
.
The selective cleavage induced by derivatized reagents was
reportedly useful for simultaneous detection during functional
group-specific LC/MS/MS analyses.15,16
Although other labeling reagents, such as NBD-F or dansyl-
Cl,17,18 can be used in LC/MS/MS analysis to increase sensitivity
and selectivity, NBD-F-labeled amino compounds show different
fragmentation patterns and are difficult to apply to the precursor
ion scan analysis or isotope ratio analysis. With the use of our
reagents, the characteristic and selective cleavage of the ureide
bond in the derivatives facilitated not only the simultaneous
analysis of the compounds with amino groups using LC/MS/MS
but also the measurement of the isotope distribution of amino
acids. These analyses were performed without complex correction
calculations, which was useful in the 13C metabolic flux analysis.
Selective cleavage induced by a precolumn reagent may be
useful for other functional groups.
Comprehensive analysis approaches using LC/MS or NMR
techniques have been widely applicable for the metabolome.
Although these methods are useful in discrimination analysis or
principal component analysis, there are too many compounds to
identify key metabolites. Therefore, the strategy of selective
analysis, featuring the functional group, using a combination of
the precursor ion scan mode of tandem mass spectrometry with
the developed reagent, will be more useful in offering practical
knowledge for biochemical studies.
Figure 4. Peak area ratio of selected TAHS derivatives and TAHS-
d3 derivatives (TAHS derivatives/TAHS-d3 derivatives). Two samples
(10 µM amino acid mixture solution) were reacted with each reagent
and mixed together before analysis.
When the amino acid standard solution was derivatized with
TAHS-d3 and the actual sample was derivatized with TAHS, we
were able to calculate the concentrations of the amino acids in
the sample with few ion suppression effects in the analysis.
Multiple samples, derivatized with different mass numbers of
TAHS reagent, analyzed in the single run will show the same
matrix effects.
ACKNOWLEDGMENT
We thank Dr. Kunisuke Izawa for helpful advice on the
synthesis of reagents and Mr. Akira Nakayama for the collection
of rat plasma and useful discussions on bioanalysis. The authors
also thank Dr. Shintaro Iwatani for the E. coli extract samples.
It would also be easy to compare the concentrations of several
samples if other isotope-labeled reagents were used, such as
TAHS-d6 or TAHS-d9.
We used TAHS-d3 as the heavy reagent in this study because
large isotope effects are known between deuterium and
hydrogen, such as retention time shift. It is better to use 13C
Received for review August 18, 2008. Accepted April 23,
2009.
AC900470W
(16) Santa, T.; Al-Dirbashi, O. Y.; Ichibangase, T.; Fukushima, T.; Rashed, M. S.;
Funatsu, T.; Imai, K. Biomed. Chromatogr. 2007, 21, 1207–1213
(17) Song, Y.; Liang, F.; Liu, Y.-M. Rapid Commun. Mass Spectrom. 2007, 21,
73–77
(18) Timperio, A. M.; Fagioni, M.; Grandinetti, F.; Zolla, L. Biomed. Chromatogr.
2007, 21, 1069–1076
.
(14) Wade, D. Chem.-Biol. Interact. 1999, 117, 191–217
(15) Santa, T.; Al-Dirbashi, O. Y.; Ichibangase, T.; Rashed, M. S.; Fukushima,
T.; Imai, K. Biomed. Chromatogr. 2008, 22, 115–118
.
.
.
.
Analytical Chemistry, Vol. 81, No. 13, July 1, 2009 5179