60
Screening of drug-mediated modifications on HSA / F. Osaki et al. / Anal. Biochem. 449 (2014) 59–67
Table 1
Reported modifications derived from activated drugs on HSA.
Drug
Modification sites
Authors
Year
In vitro/vivo
Refs.
195, 199, 351, 525, and 541
Tolmetin
Tolmetin
Benoxaprofen
Felbamate
Acetaminophen
Zileuton
Flucloxacillin
Nimesulide
Sulfamethoxazole
Nevirapine
HKI-272 (Neratinib)
Benzylpenicillin
Piperacillin
K137,
K195,
K73,
H242
C34
Ding et al.
Ding et al.
Qiu et al.
Roller et al.
Damsten et al.
Li et al.
1993
1995
1998
2002
2007
2007
2009
2009
2009
2010
2010
2011
2011
2011
2012
In vitro
In vitro
In vitro
In vitro
In vitro/vivo
In vitro
In vitro/vivo
In vitro
In vitro
In vitro
In vitro/vivo
In vitro/vivo
In vitro/vivo
In vitro/vivo
In vitro
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
199, 525, and
and
232, and 480
541; R222
521; S220,
159, 190, 195, 205, 212, 413, 436, 444, 466, 519, and
and 247
470, 480, and 489
541; R222; S312,
C34
162, 190, 195, 199, 212, 351, 432, 525, 541, and 545
K137,
C34
Jenkins et al.
Li et al.
C34
Callan et al.
Antunes et al.
Wang et al.
Meng et al.
Whitaker et al.
Jinno et al.
Ariza et al.
or 525
W214; H338; K190; K524
K190
137, 159, 162, 190, 195, 199, 212, 351, 372, 432, 436, 475, 525, 541, and 545
K20,
12, 137, 162, 190, 195, 199, 212, 351, 432, 525, 541, and 545
K4,
TAK-242
Amoxicillin
K (sites not determined)
199, 351, 432, 541, and 545
K190,
[34]. It contains many nucleophilic amino acid residues: 59 Lys, 16
His, and 1 free Cys (Cys34). Briefly, Lys and Cys34 residues are the
most likely sites to undergo modification, in addition to other
nucleophilic amino acid residues, such as Arg, His, Ser, and Trp.
However, many of these sites have been found in in vitro model
studies (only 6 in vivo studies of 15 articles, Table 1) in which
screening was performed using a nontargeted approach that relies
on a data-dependent scanning mode. However, the strategy is not
practical, because the small amount of chemically modified pep-
tides in proteolytic samples must be manually mined from the
noisy total ion chromatograms (TICs) derived from large amounts
of intact peptides.
Here, we introduce a practical way to screen drug-mediated
modification sites on proteins using HSA that could give deeper
information than one from common trapping methods, such as
GSH trapping, used for the first stage to assess reactive metabolite
formation. Our proposed strategy, referred to as ‘‘predicted multi-
ple selected reaction monitoring’’ (pSRM), consists of the following
steps: (i) first, common trapping methods and/or structural data
Japan). Water-soluble carbodiimide (1-(3-dimethylaminopropyl)-
3-ethylcarbodiimide hydrochloride; WSCD) was purchased from
the Peptide Institute (Osaka, Japan). HSA (lyophilized powder, fatty
acid free, globulin free, P99% (agarose gel electrophoresis)), GSH,
NAPQI, and KP were purchased from Sigma–Aldrich (St Louis,
MO, USA). Sequencing-grade trypsin was purchased from Promega
(Madison, WI, USA). The water used was purified with a Milli-Q
Integral 10 water purification system from Millipore (Bedford,
MA, USA). Kieselgel silica gel (for flash chromatography, 0.2–
0.5 mm, 30–70 mesh ASTM) was purchased from Merck (Darms-
tadt, Germany). AmiconUltra-0.5 centrifugal filter devices (0.5 ml,
cutoff 30 kDa) were purchased from EMD Millipore (Billerica,
MA, USA).
LC conditions
Chromatography for LC/MS analyses was carried out using three
different systems. Each system used 0.1% (v/v) formic acid in water
and 0.1% (v/v) formic acid in ACN as solvent A and solvent B,
respectively. An Agilent 1100 HPLC system (Agilent Technologies,
Santa Clara, CA, USA) used for systems 1 and 2 consisted of an
1100 G1312A pump, 1100 G1367A autosampler, 1100 G1314A
photodiode array detector, and 1100 G1316A column oven. A Ther-
mo Surveyor HPLC system (Thermo Fisher Scientific, Waltham, MA,
USA) with a pump, autosampler, and column oven was used for
system 3. The chromatographic conditions for each LC system were
are used to predict the delta value (D) of the MS shift (Fig. 1A);
(ii) in silico tryptic digestion is performed to detect the candidate
target peptides containing nucleophilic amino acids (Fig. 1B); (iii)
the first list of multiple SRM is created using predicted precursor
ions ([M+D
+nH]n+) and corresponding product ions (b2 and y2 ions)
(Fig. 1C); (iv) the first SRM is performed to find the modified pep-
tides (Fig. 1D); (v) the second list of multiple SRM for adducted
peptides is found in step (iv) using the fixed precursor ion and
as follows: for system 1, column, ODS 100 V column (5 lm, 100 Å,
the multiple product ions (series of b, y, b+D, and y+D ions)
50 ꢀ 2.0 mm i.d.; Tosoh, Tokyo, Japan); column temperature, 40 °C;
linear gradient, 0% B at 0 min, 80% B at 8 min, 80% B at 12 min, 0% B
at 12.1 min, 0% B at 15 min; and flow rate, 0.3 ml/min; for system
(Fig. 1E); and finally (vi) the second SRM is performed to identify
the modification sites (Fig. 1F). To examine the strategy, a phase
II metabolite of ketoprofen (2-(3-benzoylphenyl)propanoic acid;
KP), KP-N-hydroxysuccinimidyl (NHS) ester (equivalent to KP-glu-
curonide, Fig. 2A), and a phase I metabolite of acetaminophen (N-
(4-hydroxyphenyl)acetamide), N-acetyl-p-benzoquinone imine
2, column, Accucore-150-C4 (2.6
l
m, 150 Å, 100 ꢀ 2.1 mm i.d.;
Thermo Fisher Scientific); column temperature, 40 °C; linear gradi-
ent, 0% B at 0 min, 0% B at 3 min, 60% B at 21 min, 90% B at 22 min,
90% B at 26.9 min, 0% B at 27 min, 0% B at 30 min; and flow rate,
0.3 ml/min; and for system 3, column, Sunfire C18 column
(N-(4-oxo-1-cyclohexa-2,5-dienylidene)
acetamide;
NAPQI)
(Fig. 2B), were allowed to react with HSA as model experiments.
(3.5
l
m, 100 Å, 150 ꢀ 2.1 mm i.d.; Waters, Milford, MA, USA); col-
umn temperature, 25 °C; linear gradient, 0% B at 0 min, 0% B at
10 min, 30% B at 40 min, 50% B at 50 min, 70% B at 51 min, 70% B
at 70 min, 0% B at 71 min, 0% B at 90 min; and flow rate, 0.2 ml/
min.
Materials and methods
Materials
Ammonium bicarbonate, iodoacetamide (IAA), dithiothreitol
(DTT), dimethyl sulfoxide (Me2SO), NHS, sodium chloride (NaCl),
disodium hydrogen phosphate dodecahydrate, dipotassium hydro-
gen phosphate, toluene, formic acid, and urea were purchased from
Nacalai Tesque (Kyoto, Japan). Ethyl acetate and HPLC-grade aceto-
nitrile (ACN) were obtained from Kanto Chemical Co. (Tokyo,
MS conditions
An API2000 triple-quadrupole mass spectrometer (AB Sciex,
Framingham, MA, USA) equipped with electrospray ionization
(ESI) interface was used in positive ionization mode for pSRM
experiments of system 1 or 2. The typical ESI parameters were as