Combining [13C6]-phenylisothiocyanate and the edman degradation reaction: A possible breakthrough for absolute quantitative proteomics together with protein identification

Article abstract of DOI:10.1002/rcm.4372

This manuscript describes the results of a preliminary experiment performed as 'proof of concept' of a novel approach to absolute quantitation of proteins without the use of standard proteins. Absolute quantitation remains a challenging issue in the proteomics field. Therefore, we propose a combination of [¹³C₆]-phenylisothiocyanate (PITC) and the Edman degradation reaction as a possible breakthrough. [¹³C₆]-PITC was synthesized from [¹³C₆]-aniline with O,O(-di-2-pyridyl thiocarbonate to prepare [¹³C₆]-phenylthiohydantoin (PTH)-amino acids as internal standards. Upon the Edman degradation reaction, it has been confirmed that a model protein, bovine serum albumin (BSA), releases the N-terminal amino acid quantitatively as PTH-Asp. The standard curve of PTH-Asp against [¹³C₆]-PTH-Asp showed good linearity (r² 1/40. 9977). BSA could be quantified as PTH-Asp using the standard curve. In addition, the residual des-Asp¹-BSA provided sufficient information for further protein identification.

Full text of DOI:10.1002/rcm.4372

RAPID COMMUNICATIONS IN MASS SPECTROMETRY
Rapid Commun. Mass Spectrom. 2010; 24: 173–179
Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.4372
Combining [13C6]-phenylisothiocyanate and the Edman
degradation reaction: a possible breakthrough for
absolute quantitative proteomics together with
protein identification
*
Tomoyuki Oe , Masamitsu Maekawa, Ryo Satoh, Seon Hwa Lee and Takaaki Goto
Department of Bio-analytical Chemistry, Graduate School of Pharmaceutical Sciences, Tohoku University, Aobayama, Aoba-ku, Sendai 980-
8578, Japan
Received 28 April 2009; Revised 6 November 2009; Accepted 8 November 2009
This manuscript describes the results of a preliminary experiment performed as ‘proof of concept’ of
a novel approach to absolute quantitation of proteins without the use of standard proteins. Absolute
quantitation remains a challenging issue in the proteomics field. Therefore, we propose a combi-
nation of [13C6]-phenylisothiocyanate (PITC) and the Edman degradation reaction as a possible
breakthrough. [13C6]-PITC was synthesized from [13C6]-aniline with O,O(-di-2-pyridyl thiocarbonate
to prepare [13C6]-phenylthiohydantoin (PTH)-amino acids as internal standards. Upon the Edman
degradation reaction, it has been confirmed that a model protein, bovine serum albumin (BSA),
releases the N-terminal amino acid quantitatively as PTH-Asp. The standard curve of PTH-Asp
against [13C6]-PTH-Asp showed good linearity (r2 ¼ 0.9977). BSA could be quantified as PTH-Asp
using the standard curve. In addition, the residual des-Asp1-BSA provided sufficient information for
further protein identification. Copyright # 2009 John Wiley & Sons, Ltd.
It has been more than a decade since the word ‘proteome’
was introduced.1 The mass spectrometry (MS)-based protein
identification methodology named ‘proteomics’ immediately
spread with the high expectation2 that it could express
phenotypes under various physiological states as the next era
of genomics.3 However, researchers have noticed that
quantitation based on this methodology was not as simple
as hoped. For MS-based analyses, the stable isotope dilution
technique is the most powerful technique for quantitation.4
Therefore, several stable isotope labeling methods have been
applied for quantitative proteomics. Oda et al. introduced the
technique for the first quantitative proteomics.5 They used
15N-labeled media for the cell cultures and achieved accurate
quantitation of protein expression and site-specific phos-
phorylation. This concept was further improved by Mann
and colleagues in a method named SILAC (Stable Isotope
Labeling with Amino acids in Cell culture).6 They used [D3]-
leucine instead of 15N-labeled media to facilitate further
applications for mammalian systems and tandem mass
spectrometry (MS/MS)-based protein identifications. These
methodologies enabled the introduction of stable isotopes for
the entire proteome; however, it is limited to cell exper-
iments. On the other hand, Fenselau and colleagues have
reported the use of tryptic digestion in 18O-water as a method
to introduce 18O into whole digested peptides.7 In a different
approach to introducing stable isotopes, several stable
isotope labeled tags have been reported. Gygi et al. have
reported quantitative proteomics attaching a deuterated
biotin tag to cysteines in the technique named Isotope-Coded
Affinity Tags (ICAT).8 This methodology has been further
improved as cleavable ICAT, which has a cleavable site on the
reagent to overcome the low recovery from the avidin
column and is labeled with 13C instead of deuterium to
reduce the isotope effect on chromatography.9 In addition,
resins attached with deuterated tags have been designed by
Zhang et al. to facilitate the clean-up step.10 Isobaric Tag for
Relative and Absolute Quantitation (iTRAQTM) was devel-
oped next.11 The iTRAQTM reagents are a set of isobaric
reagents that have four different mass combinations to
enable simultaneous identification and quantitation of up to
four different experiments. Because the reagents can react
through the lysine residue and the N-terminal amino group,
more peptides can be labeled than with ICAT. All of the
above methodologies have made major contributions to the
proteomics field; however, absolute quantitation remains a
challenging issue.12
Absolute quantitation can be attained if limited numbers
of proteins are targeted, such as amyloid betas in cerebro-
spinal fluid,13 membrane transporter proteins,14 and phos-
phorylation on the kinase activation loop of cellular focal
adhesion kinase.15 These approaches are possible when
sufficient amounts of standards and stable isotope labeled
standards (or tryptic peptides) are available to prepare standard
solutions. However, it is virtually impossible to obtain large
enough amounts of standards and the corresponding stable
isotope labeled internal standards (ISs) for the entire
*Correspondence to: T. Oe, Department of Bio-analytical Chem-
istry, Graduate School of Pharmaceutical Sciences, Tohoku Uni-
versity, Aobayama, Aoba-ku, Sendai 980-8578, Japan.
E-mail: t-oe@mail.pharm.tohoku.ac.jp
Copyright # 2009 John Wiley & Sons, Ltd.

174 T. Oe et al.
proteome. Therefore, a novel approach is needed to attain
absolute quantitation in global proteomics.
quantitation. We assumed that proteins could be quantified
as PTH-amino acids released from the proteins if the
following degradation is quantitative for proteins.
A traditional and well-known method for amino acid
sequencing in a peptide, reported in 1950, is the Edman
degradation reaction; it was used almost exclusively for
protein identification before MS-based proteomics appeared.16
The degradation reaction consists of three steps: (i) Edman
reagent (phenylisothiocyanate, PITC, Fig. 1(a)) is allowed to
react with the N-terminal a-amino group of the peptide; (ii)
the resulting phenylthiocarbamoyl derivative (PTC-peptide,
Fig. 1(b)) is cleaved at the first amide bond to produce the
anilinothiazolinone derivative (ATZ-amino acid, Fig. 1(c))
and des-1 peptide (Fig. 1(d)); and (iii) the relatively labile
ATZ-amino acid is converted into the more stable phe-
nylthiohydantoin derivative (PTH-amino acid, Fig. 1(e)). The
PTH-amino acids can be identified by their retention times
using high-performance liquid chromatography–ultraviolet
detection (HPLC-UV), and the remaining des-1 peptide can
be returned to the first step to identify the next amino acid.
Automated,17 miniaturized,18 and fluorescence19 methods
have also been used to improve the methodology. In
addition, the Edman degradation reaction from gel samples
has been reported recently.20
There are three major potential advantages of the original
Edman degradation reaction for absolute quantitative
proteomics (Fig. 2): (i) all PTH-amino acids are commercially
available as standards; (ii) the reaction is well studied and
automation can be used; and (iii) des-1 protein can be used
for further protein identification after protease digestion.
Recently, Brune et al. have reported the great possibility of
quantitation of Edman degradation with detailed data as a
study by the Edman Sequencing Research Group.26 There-
fore, we believe that the Edman degradation reaction as used
in combination with stable isotope labeled PTH derivatives
of selected amino acids could formulate the basis of a
possible breakthrough for quantitative proteomics and
concomitant protein identification.
To prove the concept, we have demonstrated here the
quantitation of bovine serum albumin (BSA) without a
protein standard by analyzing the PTH-Asp released from
the N-terminus. The standard curve was prepared using
commercially available PTH-Asp together with [13C6]-PTH-
Asp as the IS prepared from novel [13C6]-PITC (Fig. 3).
The protein identification from des-Asp1-BSA was also
demonstrated by the peptide mass fingerprinting (PMF)
method.
There have also been some attempts reported to use a
combination of Edman reagent and MS as proteomics tools.
Chait et al. have reported a novel protein ladder sequencing
based on partial Edman degradation.21 The methodology has
been used for combinatorial chemistry because of its high
throughput and good cost performance.22 Aebersold et al.
have reported several Edman-type reagents to enhance the
ionization of peptides.23 Recently, Marekov and Steinert
have reported that charge derivatization using 4-sulfophe-
nylisothiocyanate (SPITC) had high ability to enhance
peptide sequencing by post-source decay matrix-assisted
laser desorption/ionization time-of-flight (MALDI-TOF)
MS.24 Guillaume et al. have reported that the combination
of [13C6]-SPITC and SPITC can be further applied, not only to
protein identification, but also to quantitation.25 Therefore, a
combination of Edman reagent and MS has the potential
ability in the proteomics field for both sequencing and
quantitation. However, no one has tried it for absolute
EXPERIMENTAL
Chemicals and materials
Aniline, ethylmercaptan, triethylamine, arginine (Arg),
tryptophan (Trp), CDCl3 (containing 0.03% tetramethylsilane
(TMS)), ethyl acetate, n-hexane, ammonium bicarbonate,
iodoacetamide (IAA), dithiothreitol (DTT), Na2EDTA, and
HCl were purchased from Nacalai Tesque, Inc. (Kyoto,
Japan). Phenylisothiocyanate (PITC), O,O0-di-2-pyridyl thio-
carbonate, and 2,5-dihydroxybenzoic acid (DHB) were
purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo,
Japan). N2 gas and [13C6]-aniline were purchased from Taiyo
Nippon Sanso Co. (Tokyo, Japan). PTH-Asp, PTH-nor-Val,
PTH-amino acids mixture standard, trifluoroacetic acid
Figure 1. Edman degradation reaction. (a) Edman reagent (phenylisothiocyanate, PITC), (b) phenylthiocarbamoyl (PTC)-
peptide, (c) anilinothiazolinone (ATZ)-amino acid, (d) des-1-peptide, (e) phenylthiohydantoin (PTH)-amino acid.
Copyright # 2009 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 173–179
DOI: 10.1002/rcm

Possible breakthrough for absolute quantitative proteomics 175
Figure 2. Concept of quantitative proteomics together with protein identification.
Figure 3. Structure and synthesis of [13C6]-Edman reagent.
(TFA), ammonium acetate, chloroethyl carbonate, phos-
Japan) that consisted of two model 2001 semimicropumps, a
model 2002 UV-Vis detector, and a model 7125 six-port
syringe loading sample injector valve (Rheodyne Co., Cotati,
CA, USA). A Wakosil II 3C18 RS column (150 ꢀ 2.0 mm i.d.,
phoryl chloride, formic acid, Wakogel1 FC-40 silica gel (for
flash chromatography, 20–40 mm), and tris(hydroxyme-
thyl)aminomethane (Tris) were purchased from Wako Pure
Chemical Industries, Ltd. (Osaka, Japan). HPLC grade
acetonitrile and ethanol were obtained from Kanto Chemical
Co. Inc. (Tokyo, Japan). Purified water was purchased from
Daiwa-Yakuhin Co. Ltd. (Sendai, Japan) and further filtered
using a CPW-100 Ultrapure water system (Advantec Toyo
Kaisha, Ltd., Tokyo, Japan). Bovine serum albumin (BSA), a-
cyano-4-hydroxycinnamic acid (CHCA), ACTH (18–39), and
bovine insulin oxidized b-chain were purchased from Sigma-
Aldrich Co. (St. Louis, MO, USA). Angiotensin (Ang) I was
purchased from American Peptide Company, Inc. (Sunnyvale,
CA, USA). Human Ang II was obtained from Calbiochem/
EMD Chemicals, Inc. (San Diego, CA, USA). Ang IV was
purchased from the Peptide Institute, Inc. (Osaka, Japan).
Oasis1 HLB cartridges (1 cc, 10 mg) were obtained from
Waters Co. (Milford, MA, USA). Slide-A-lyzer1 Mini-Dialysis
units (3500 molecular weight cutoff (MWCO)) were pur-
chased from Thermo Fisher Scientific Inc. (Rockford, IL,
USA). Sequencing-grade trypsin and guanidine-HCl were
purchased from Promega Co. (Madison, WI, USA).
˚
3 mm, 300 A; Wako Pure Chemicals Co., Osaka, Japan) and a
˚
YMC-Pack ODS-AM column (150 ꢀ 6.0 mm i.d., 5 mm, 120 A,
YMC Co. Ltd., Kyoto, Japan) were used for systems 1 and 2,
respectively. Solvent A was 20 mM ammonium acetate buffer
(pH 4.88)/acetonitrile (95:5, v/v), and solvent B was 20 mM
ammonium acetate buffer (pH 4.88)/acetonitrile (5:95, v/v).
The linear gradient for system 1 was as follows: 10% B at
0 min, 50% B at 10 min, 50% B at 20 min, 95% B at 21 min, 95%
B at 30 min, 10% B at 31 min, and 10% B at 45 min with a flow
rate of 0.2 mL/min. The linear gradient for system 2 was as
follows: 10% B at 0 min, 50% B at 10 min, 95% B at 11 min, 95%
B at 25 min, 10% B at 26 min, and 10% B at 40 min with a flow
rate of 2.0 mL/min. All separations were performed at
ambient temperature. HPLC-UV was performed at 269 nm
and the data were processed using Chromato-PRO (Runtime
Instrument Co., Sagamihara, Japan).
MALDI-TOF MS analyses of PTH-amino acids
and tryptic peptides
MALDI-TOF MS experiments were carried out using a
Voyager-DE STR MALDI-TOF mass spectrometer (Applied
Biosystems, Framingham, MA, USA) located in the Biome-
dical Research Core, School of Medicine, Tohoku University.
HPLC analyses of PTH-amino acids
Chromatography was carried out using a Nanospace SI-1
semimicrocolumn HPLC system (Shiseido Co. Ltd., Tokyo,
Copyright # 2009 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 173–179
DOI: 10.1002/rcm

176 T. Oe et al.
The mass spectrometer was equipped with a nitrogen laser
(337 nm, 3 ns pulse width, 20 Hz repetition rate) and had a
flight path of 200 cm. All spectra presented were acquired in
positive-ion mode in reflection mode with the following
parameters. For PTH-amino acids: accelerating voltage,
10 kV; grid voltage, 70%; extraction delay time, 100 ns; low
mass gate, 100 Da; shots in the spectrum, 100–150. For tryptic
peptides: accelerating voltage, 20 kV; grid voltage, 64%;
extraction delay time, 100 ns; low mass gate, off; shots in the
spectrum, 100. TOF MS experiments were performed in the
range of m/z 100–1500 for PTH-amino acids and 500–5000 for
tryptic peptides, respectively. All acquired data were
processed using Data Explorer version 4.0.0.0.
following criteria: curve, linear; origin, ignored. The above
experiment was performed in triplicate (n ¼ 3).
Synthesis of [13C6]-PITC
The isothiocyanate group was constructed by the method
described by Bures et al.27 with minor modifications. [13C6]-
Aniline (250 mg, 2.5 mmol) in CHCl3 (2 mL) was allowed to
react with O,O0-di-2-pyridyl thiocarbonate (1.25 g, 5 mmol) at
room temperature for 3 h then at 508C for 1 h. After the
reaction, the solvent was evaporated and the residue
was applied to a flash silica gel chromatography column
with n-hexane to remove excess reagent. The eluate was
then evaporated to give [13C6]-PITC as a colorless oil
(392 mg, quant.). High-resolution FABMS calculated for
13C162C1H5N1S1: theoretical, m/z 141.0344; found, m/z 141.0324
Instrumental analyses
1H-NMR
.
(Mþ ). 1H-NMR (500 MHz, CDCl3, d): 7.05–7.25 and 7.35–7.55
(doublet of multiplets, J(13CꢁH) ¼ 167 Hz, HAr, 5H). IR (KBr,
NMR spectra were recorded at 258C at 500 MHz using an
ECP-500 (JEOL Ltd., Tokyo, Japan) equipped with a 3 mm
indirect detection gradient probe (Wilmad Glass Co. Inc.,
Buena, NJ, USA). [13C6]-PITC (ca. 1 mg) was dissolved in
150 mL of CDCl3 (containing 0.03% TMS). Data processing
was conducted directly on the NMR spectrometer. Chemical
shifts are reported on the d scale (ppm) by assigning the TMS
peak.
cmꢁ1): 2078 (vs, NCS).
Synthesis of [13C6]-PTH-Asp
Ang II (7 mM in 0.1 M HCl, 10 mL ꢀ 10) solvent was
evaporated under a N2 stream. The residue was allowed
to react with 10 mL of [13C6]-PITC solution ([13C6]-PITC/
triethylamine/water/ethanol 1:1:1:7, v/v/v/v) at 558C for
15 min on a heating block. The reaction mixture was dried
under a N2 stream and the residue was then washed with n-
hexane/ethyl acetate (15:1, v/v, 200 mL) and ethyl acetate
(200 mL). Ethylmercaptan in TFA (0.1%, v/v, 20 mL) was
added to the tube at 558C for 15 min on a heating block. After
evaporating the solvent under a N2 stream, the residue was
redissolved in water (20 mL) at 558C for 5 min and 50% (v/v)
aqueous TFA (20 mL) at 558C for 60 min. The reaction mixture
was dried under a N2 stream and the residue was then
dissolved in 50% (v/v) ethanol (50 mL). The solution from
each tube was then applied to HPLC system 2, the fraction
between tR 4.3 and 5.3 min was collected, dried under a N2
stream, and redissolved in 50% (v/v) aqueous ethanol (7 mL)
as [13C6]-PTH-Asp stock solution (100 mM). MALDI-TOF MS
(DHB): m/z 257.07 (MHþ).
High-resolution FABMS
High-resolution mass spectra were recorded using a JMS-700
double-focusing magnetic sector mass spectrometer (JEOL
Ltd., Tokyo, Japan).
IR
IR spectra were determined at ambient temperature using an
FTIR-400 (JASCO Co., Tokyo, Japan). [13C6]-PITC was
diluted with CHCl3. An aliquot of the solution was dropped
on a KBr disk and analyzed after drying.
Edman degradation reaction for BSA
Each BSA solution (20, 50, 100, 200, 500, 1000, 2000, 5000,
10000, 20000, and 50000 pmol/10 mL in 0.1 M HCl) was
transferred to a microcentrifuge tube and evaporated under
a N2 stream using a Jet Air vaporizer (Ishii Shoten Co., Tokyo,
Japan). Each residue was allowed to react with 10 mL of PITC
solution (PITC/triethylamine/water/ethanol 1:1:1:7, v/v/
v/v) at 558C for 15 min on a heating block (Yamato Scientific
Co., Ltd., Tokyo, Japan). The reaction mixture was evapor-
ated under a N2 stream and the residue was then washed
with n-hexane/ethyl acetate (15:1, v/v, 200 mL), ethyl acetate
(200 mL), and acetone (200 mL ꢀ 3). Ethylmercaptan in TFA
(0.1%, v/v, 20 mL) was added to the tube at 558C for 15 min on
a heating block. After evaporation under a N2 stream, the
residue was redissolved in 25% (v/v) aqueous TFA (20 mL)
and heated at 658C for 30 min. At that stage, an aliquot of this
solution can be used for further protein identification if
required. PTH-nor-Val (1 nmol/100 mL of 50% aqueous
ethanol, v/v) was added to each tube as an IS. An aliquot
(5 mL) of the final solution was analyzed using HPLC system
1. The peak area ratio (PTH-Asp/PTH-nor-Val) was then
plotted against BSA concentration. The calculations were
performed using Microsoft Office Excel 97–2003 with the
Calibration curve using PTH-Asp and
[13C6]-PTH-Asp for BSA quantitation
The calibration curve was prepared with PTH-Asp standard
and [13C6]-PTH-Asp. PTH-Asp solution (10, 20, 50, 100, 200,
500, or 1000 pmol/10 mL of 50% aqueous ethanol, v/v) and
[13C6]-PTH-Asp solution (1000 pmol/10 mL of 50% aqueous
ethanol, v/v) were mixed. BSA (1 nmol/10 mL in 0.1 M HCl)
was allowed to react with Edman reagent as described in the
HPLC-UV experiment and [13C6]-PTH-Asp (1000 pmol/
10 mL of 50% aqueous ethanol, v/v) was added as an IS.
For MALDI-TOF MS analyses, aliquots (0.5 mL) were loaded
on the MALDI sample plate followed by 0.5 mL of 100 mM
DHB in a mixture of water/acetonitrile/TFA (50:50:0.1, v/v/
v) containing internal calibrants (Arg, 10 pmol, MHþ
175.1195; Trp, 10 pmol, MHþ 205.0977; and Ang IV 1 pmol,
MHþ 775.4143) and allowed to dry at room temperature.
After the MALDI-TOF MS analyses, the peak area ratios of
PTH-Asp/[13C6]-PTH-Asp (monoisotopic MHþ; intensity @
251/intensity @ 257) were plotted against PTH-Asp concen-
trations. The calculations were performed using Microsoft
Copyright # 2009 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 173–179
DOI: 10.1002/rcm

Possible breakthrough for absolute quantitative proteomics 177
Office Excel 97–2003 with the following criteria: curve, linear;
origin, ignored. The above experiment was performed in
triplicate (n ¼ 3).
Tryptic digestion for des-Asp1-BSA
After Edman degradation, the solution containing des-Asp1-
BSA coexisting with PTH-Asp and [13C6]-PTH-Asp in 12.5%
(v/v) aqueous TFA (40 mL) was evaporated and redissolved
in reduction and carboxymethylation (RCM) buffer (1.65 M
Tris-HCl, 7.3 M guanidine-HCl, 0.03 M EDTA, pH 8.3, 50 mL).
DTT was added to produce a final concentration of 10 mM
and incubated at 608C for 1 h to reduce the disulfide bonds.
The reduced cysteines were then alkylated upon the addition
of IAA at a final concentration of 55 mM and incubated at
room temperature in the dark for 45 min. The reduced and
alkylated proteins were dialyzed with Slide-A-lyzer1 Mini
Dialysis units (3500 MWCO) against water (1 L, 1 h ꢀ 3).
After evaporating the solution under a N2 stream, the residue
was redissolved in 100 mM ammonium bicarbonate (40 mL)
and digested using sequencing-grade trypsin (enzyme/
protein 1:165, w/w) with overnight incubation at 378C.
The pH of the digested protein sample was then lowered to 3
with 1% (v/v) aqueous TFA solution. The sample was
desalted using an Oasis1 HLB cartridge by washing with
water (1 mL) and eluted with 75% (v/v) aqueous acetonitrile
(1 mL). After evaporating the eluate under a N2 stream, the
residue was redissolved in the mixture of water/aceto-
nitrile/TFA (50:50:0.1, v/v/v, 20 mL). The aliquots (0.5 mL)
were loaded on the MALDI sample plate followed by
addition of 0.5 mL of matrix solution (saturated CHCA in
water/acetonitrile/TFA 50:50:0.1, v/v/v) and allowed to dry
at room temperature. The calibrant solution (1 pmol Ang II,
MHþ 1046.5418; 1 pmol Ang I, MHþ 1296.6853; 1 pmol ACTH
(18–39), MHþ 2465.1989; and 10 pmol bovine insulin
oxidized b-chain, MHþ 3494.6513) in saturated CHCA
solution (water/acetonitrile/TFA 50:50:0.1, v/v/v) was used
for the external calibration. Protein identification was carried
out using a PMF tool, MS-Fit in ProteinProspector Version
5.2.1 Basic, developed in the UCSF Mass Spectrometry
Facility,28 with the following criteria: database, SwissProt.
2008.12.16; digest used, trypsin; maximum number of missed
cleavages, 2; constant modifications, carbamidomethyl (C),
minimum matches, 4; sort type, score sort; considered
modifications, oxidation of M; min precursor ion matches, 1;
MOWSE On, 1; MOWSE P factor, 0.4; masses, monoisotopic;
mass tolerance, 0.2 Da. Peaks for which the signal/noise ratio
was greater than 14.0 were used for PMF.
Figure 4. Production of PTH-Asp released from BSA by the
Edman degradation reaction.
encouraged us to investigate the yield of PTH-amino acids
from larger proteins to see if this reaction can be used for the
key reaction in our strategy. Therefore, the degradation yield
of BSA (67 kDa) was examined to optimize the reaction
conditions. In our conditions, the released PTH-Asp and des-
1 protein were not separated from each other to avoid
reducing the recovery and further protein digestion in the
same tube. The response of released PTH-Asp from BSA
showed a good linearity (y ¼ 0.0006x – 0.0559, r2 ¼ 0.9998)
throughout the range of 20–50000 pmol/tube (Fig. 4).
Synthesis of [13C6]-Edman reagent
An attempt to introduce 13C on PITC has already been
reported by Ares et al.29 However, only one 13C was
introduced on the isothiocyanate group (PheꢁN¼13C¼S)
for the purpose of protein probing by 13C-NMR. For our
purpose, a 1 Da shift is not enough as an IS, so we have
prepared [13C6]-PITC labeled on the phenyl group because
[13C6]-aniline is commercially available. As the preliminary
experiment, three different methods were tried to synthesize
PITC from [12C]-aniline. The reaction between aniline and
carbon disulfide followed by E1 elimination using chloro-
ethyl carbonate30 and phosphoryl chloride29 gave 40% and
90% yields, respectively. The reaction with O,O0-di-2-pyridyl
thiocarbonate27 gave 90% yield. Therefore, a one-step
reaction with O,O0-di-2-pyridyl thiocarbonate was chosen
for [13C6]-PITC synthesis (Fig. 3). The yield from [13C6]-
aniline was quantitative and the product was sufficiently
isotopically pure (>99.9%) for further preparations of [13C6]-
PTH-amino acids as ISs.
RESULTS AND DISCUSSION
Synthesis of [13C6]-PTH-amino acid as IS
Quantitation of the Edman degradation
reaction
All [13C6]-PTH-amino acids (ISs) for 20 amino acids can be
prepared simply using the reaction of [13C6]-PITC and
individual amino acid standards (or an amino acid standard
mixture). In this experiment, only [13C6]-PTH-Asp was
synthesized for the further model experiments with BSA. The
yields of PTH-amino acids from individual amino acids were
lower than that from the peptide in our preliminary
experiments (ꢃ50%). Therefore, [13C6]-PTH-Asp was pre-
pared from Ang II. In the future, reaction of PITC and
carboxyamidated-amino acids (NH2ꢁCHRꢁCONH2) will be
Quantitation of the Edman degradation reaction has mainly
been studied for peptides. Recently, Brune et al. have
reported the important possibility of quantitative Edman
degradation with detailed data as a study by the Edman
Sequencing Research Group.26 They used three peptides,
KAQYARSVLLEKDAEPDILELATGYR
(peptide
B),
RQAKVLLYSGR (peptide C), and RQAK(Ac)VLLYSGR
(peptide Cꢂ), and have shown favorable data. The data
Copyright # 2009 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 173–179
DOI: 10.1002/rcm

178 T. Oe et al.
experiment using BSA. The samples were analyzed using
MALDI-TOF MS. The peak area ratios (PTH-Asp/[13C6]-
PTH-Asp) were plotted against PTH-Asp concentrations.
The calibration curve showed good linearity over the entire
range (y ¼ 0.0015x – 0.0112, r2 ¼ 0.9977) (Fig. 5). The corre-
lation coefficient (r) was slightly lower than that from the
HPLC-UV experiment because of the higher background in
the low mass range and the lower dynamic range based on
the use of MALDI-TOF MS.
Quantitation and qualification of BSA
The method was applied to BSA to prove our proposed
concept. The PTH-Asp released from the N-terminus of BSA
(1.0 nmol) was clearly found together with [13C6]-PTH-Asp
(IS, 1.0 nmol) (Fig. 6). The peak area ratio (PTH-Asp/[13C6]-
PTH-Asp ¼ 1.01) was ideal for the quantitation. This suggests
that the Edman degradation is the key reaction to attain
absolute quantitative proteomics without protein standards.
Then, the protein digestion procedure was optimized for
further protein identification. To reduce disulfide bonds,
RCM buffer was required to denature the aggregated des-
Asp1-BSA resulting from the Edman degradation conditions.
After the following S-alkylation with IAA, des-Asp1-BSA
was successfully digested using trypsin (Fig. 7). Using the
PMF method, des-Asp1-BSA has been assigned to BSA as the
top score (6.97 ꢀ 104) with 25.0% amino acid sequence
coverage (152/607 amino acids). This suggests that des-1
proteins can provide sufficient information to identify the
intact protein by the PMF method, even if the N-terminal
amino acids are missing. MS/MS-based techniques can
therefore give better identification.
Figure 5. Calibration curve for PTH-Asp against [13C6]-PTH-
Asp.
evaluated for efficient preparation of [13C6]-PTH-amino acids
as a general method. The [13C6]-PTH-Asp prepared here was
sufficiently isotopically pure (>99.9%) to be used as an IS,
and was applied for further experiments.
Calibration curve of PTH-Asp over
[13C6]-PTH-Asp
The commercial availability of individual and mixtures of 20
PTH-amino acids is one of the biggest advantages of this
strategy. Therefore, calibration curves can be prepared for
the entire proteome without standard proteins using the
corresponding [13C6]-PTH amino acids as ISs. A calibration
curve of PTH-Asp against [13C6]-PTH-Asp was prepared in
the range from 10 pmol to 1 nmol for the next model
Figure 6. MALDI-TOF MS spectrum of PTH-Asp released from BSA by the Edman
degradation reaction.
Figure 7. MALDI-TOF MS spectrum of des-Asp1-BSA after tryptic digestion.
Copyright # 2009 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 173–179
DOI: 10.1002/rcm

Possible breakthrough for absolute quantitative proteomics 179
CONCLUSIONS
4. De Leenheer AP, Thienpont LM. Mass Spectrom. Rev. 1992;
11: 249.
Here, we have proved the concept of absolute quantitative
proteomics using the combination of [13C6]-PITC as ISs for
products of the Edman degradation reaction. Also, we have
shown that analysis of tryptic fragments of the residual des-1
protein can provide sufficient information for protein
identification. Because the concept worked, we are now
focusing on ways to solve the potential problems for practical
use. For instance, a spot from a 2-D gel normally contains
multiple proteins, and some of them could have the same N-
terminal amino acid. For this problem, we are trying to use
not only PTH-amino acids from the first cycle, but also those
from the second cycle together for the quantitation. In
addition, use of an LC/MS system can solve the problems of
coexisting proteins that have the same mass N-terminal
amino acids, such as Leu vs. Ile and Gln vs. Lys,
chromatographically. This should further increase the
sensitivity because the current sensitivity with MALDI-
TOF MS is limited by the high background from the matrix in
the low molecular weight region for PTH-amino acids (m/z
193–322). Besides those steps, enzymatic or chemical
digestions prior to the Edman reaction will be tried for N-
protected and N-truncated proteins.
5. Oda Y, Huang K, Cross FR, Cowburn D, Chait BT. Proc. Natl.
Acad. Sci. USA 1999; 96: 6591.
6. Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H,
Pandey A, Mann M. Mol. Cell. Proteomics 2002; 1: 376.
7. Yao X, Freas A, Ramirez J, Demirev PA, Fenselau C. Anal.
Chem. 2001; 73: 2836.
8. Gygi SP, Rist B, Gerber SA, Turecek F, Gelb MH, Aebersold
R. Nat. Biotechnol. 1999; 17: 994.
9. Li J, Steen H, Gygi SP. Mol. Cell. Proteomics 2003; 2: 1198.
10. Zhang L, Guo YL, Liu HQ. J. Mass Spectrom. 2004; 39: 447.
11. Zieske LR. J. Exp. Bot. 2006; 57: 1501.
12. Sechi S, Oda Y. Curr. Opin. Chem. Biol. 2003; 7: 70.
13. Oe T, Ackermann BL, Inoue K, Berna MJ, Garner CO,
Gelfanova V, Dean RA, Siemers ER, Holtzman DM, Farlow
MR, Blair IA. Rapid Commun. Mass Spectrom. 2006; 20: 3723.
14. Kamiie J, Ohtsuki S, Iwase R, Ohmine K, Katsukura Y, Yanai
K, Sekine Y, Uchida Y, Ito S, Terasaki T. Pharm. Res. 2008; 25:
1469.
15. Ciccimaro E, Hanks SK, Yu KH, Blair IA. Anal. Chem. 2009;
81: 3304.
16. Edman P. Acta Chem. Scand. 1950; 4: 283.
17. Edman P, Begg G. Eur. J. Biochem. 1967; 1: 80.
18. Chen W, Yin X, Mu J, Yin Y. Chem. Commun. (Camb.) 2007;
2488.
19. Jin SW, Chen GX, Palacz Z, Wittmann-Liebold B. FEBS Lett.
1986; 198: 150.
20. Komatsu S. Methods Mol. Biol. 2007; 355: 211.
21. Chait BT, Wang R, Beavis RC, Kent SBH. Science 1993; 262:
89.
22. Thakkar A, Wavreille AS, Pei D. Anal. Chem. 2006; 78: 5935.
23. Aebersold R, Bures EJ, Namchuk M, Goghari MH, Shushan
B, Covey TC. Protein Sci. 1992; 1: 494.
24. Marekov LN, Steinert PM. J. Mass Spectrom. 2003; 38: 373.
25. Guillaume E, Panchaud A, Affolter M, Desvergnes V, Kuss-
mann M. Proteomics 2006; 6: 2338.
Acknowledgements
The authors thank the Central Analytical Center and the
Laboratory of Synthetic Chemistry in our school for high-
resolution FABMS and FTIR analyses, respectively.
26. Brune DC, Hampton B, Kobayashi R, Leone JW, Linse KD,
Pohl J, Thoma RS, Denslow ND. J. Biomol. Technol. 2007; 18:
306.
27. Bures EJ, Nika H, Chow DT, Morrison HD, Hess D, Aeber-
sold R. Anal. Biochem. 1995; 224: 364.
REFERENCES
1. Wasinger VC, Cordwell SJ, Cerpa-Poljak A, Yan JX, Gooley
AA, Wilkins MR, Duncan MW, Harris R, Williams KL,
Humphery-Smith I. Electrophoresis 1995; 16: 1090.
2. Persidis A. Nat. Biotechnol. 1998; 16: 393.
28. Available:
http://prospector.ucsf.edu/cgi-bin/msform.
cgi?form=msfitstandard.
29. Ares JJ, Fowble JW, Urchek TG, Miller DD. Biochem. Biophys.
Res. Commun. 1987; 142: 1064.
3. Brower V. EMBO Rep. 2001; 2: 558.
30. Hodgkins JE, Reeves WP. J. Org. Chem. 1964; 29: 3098.
Copyright # 2009 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2010; 24: 173–179
DOI: 10.1002/rcm