Rapid determination of amino acid incorporation by stable isotope labeling with amino acids in cell...
-
Upload
frank-schmidt -
Category
Documents
-
view
215 -
download
1
Transcript of Rapid determination of amino acid incorporation by stable isotope labeling with amino acids in cell...
RAPID COMMUNICATIONS IN MASS SPECTROMETRY
Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
) DOI: 10.1002/rcm.3290
Published online in Wiley InterScience (www.interscience.wiley.comRapid determination of amino acid incorporation
by stable isotope labeling with amino acids in
cell culture (SILAC)
Frank Schmidt, Margarita Strozynski, Sandra S. Salus, Hilde Nilsen and Bernd Thiede*The Biotechnology Centre of Oslo, University of Oslo, Gaustadalleen 21, 0349 Oslo, Norway
Received 24 May 2007; Revised 13 September 2007; Accepted 1 October 2007
*CorrespoUniversit0317 OsloE-mail: b
Stable isotope labeling with amino acids in cell culture (SILAC) has evolved to be a major technique
for quantitative proteomics using cell cultures. We developed a rapid method to follow and
determine the incorporation of arginine and lysine. Analysis of the heavy state is required to avoid
quantification errors. Moreover, the mixture of light and heavy states can be exploited to normalize
the protein amount for subsequent relative quantification experiments. Therefore, peptides from
different cell lines were extracted with 0.1% trifluoroacetic acid and analyzed bymatrix-assisted laser
desorption/ionization tandem time-of-flight (MALDI-TOF/TOF) mass spectrometry (MS). This
analysis was highly reproducible and was performed in less than 2h, significantly faster than other
methods for the same purpose. Similar peptide mass profiles were obtained for human EBV-
transformed B, Jurkat T, and HeLa cells as well as for mouse embryonic fibroblasts. Proteolytic
fragments of 27 human proteins were identifiedwith 56 peptides byMALDI-MS/MS and can be used
as a database for these kinds of experiments. Sequencing revealed that the peptides were predo-
minantly amino- and carboxy-terminal protein fragments displaying a specificity characteristic of the
acidic proteases cathepsin D and E.Many of the identified peptides contained arginine and/or lysine,
allowing determination of the incorporation rate of these amino acids. Furthermore, the rate of con-
version of arginine into proline could bemonitored easily. Copyright# 2007 JohnWiley& Sons, Ltd.
Mass spectrometry (MS)-based methods that utilize stable
isotopes for quantitative proteome analyses have undergone
significant development in recent years.1 Stable isotope label-
ing with amino acids in cell culture (SILAC) is a simple and
straightforward approach for in vivo incorporation of a label
into proteins for quantitative proteomics by mass spectro-
metry.2 Consequently, SILAC has been applied to many
quantitative proteomics analyses, including functional and
clinical studies.3–21 SILAC involves culturing of one cell
population in normalmedium (light) and another inmedium
containing specific heavy isotopic amino acids, making their
respective peptides distinguishable by mass. These cell
populations can then be subjected to different treatments and
combined prior to further processing. During analysis byMS,
the relative peak intensities between the unlabeled and
labeled peptides are quantitative, making it possible to
evaluate the effects of specific treatment on a large number of
proteins within a single experiment.
The majority of proteins are typically unchanged during
an experiment. To find regulated proteins using stable
isotopes, it is generally recommended to normalize the
relative protein input from controls and sample of interest to
a 1:1 ratio. However, the accuracy is unsatisfactory for
ndence to: B. Thiede, The Biotechnology Centre of Oslo,y of Oslo, Gaustadalleen 21, P.O. Box 1125 Blindern,, [email protected]
protein quantification according to Bradford22 or Lowry23
assay. Analysis of tryptic peptides by MS can be performed
for relative quantificationwith orwithout prior separation by
sodium dodecyl sulfate polyacrylamide gel electrophoresis
(SDS-PAGE) of mixed proteins labeled with light and heavy
amino acids, but this method is relatively slow and requires
approximately 2 days.24 Furthermore, SILAC experiments
depend on readout by MS. Thus, it must be ensured that
proteins are fully labeled to avoid quantification errors.24
For SILAC experiments, relative quantification is required
for normalization of the protein amount of the light and
heavy state. Furthermore, full incorporation of heavy amino
acids is required to exclude quantification errors. In this
study, a simple and rapid methodology, based on peptide
analysis with matrix-assisted laser desorption/ionization
(MALDI)-MS of different cell lines, was established to follow
and determine the incorporation of amino acids by SILAC.
We found that this approach allowed identification and
relative quantification of a number of peptides that were
derived from larger proteins in less than 2h. Further-
more, the rate of conversion of arginine into proline could be
monitored easily. Notably, the determined proteolytic
fragments can be useful as a database for these kinds of
experiments.
Copyright # 2007 John Wiley & Sons, Ltd.
3920 F. Schmidt et al.
EXPERIMENTAL
Isolation of peptidesThe Jurkat T cell line E6, EBV-transformed human B-cells,
HeLa cells, and mouse embryonic fibroblasts were main-
tained in RPMI tissue culture medium without arginine and
lysine (Invitrogen, Karlsruhe, Germany) supplemented
with 10% dialyzed fetal calf serum (Invitrogen), penicillin
(100U/mL)/streptomycin (100mg/mL) (Invitrogen) at 378Cin 5.0% CO2, and 100mg/L of arginine-12C6 monohydro-
chloride and lysine-12C6 monohydrochloride (light) or
arginine-13C6 monohydrochloride and lysine-13C6 monohydro-
chloride (heavy) (Sigma-Aldrich, Oslo, Norway). Only
50mg/L arginine monohydrochloride was used for the
experiments with reduced arginine. The cells were washed
twice with phosphate-buffered saline (PBS) at room tem-
perature (RT). Subsequently, 1� 106 cells were treated with
100mL of 0.1% trifluoroacetic acid (TFA) for 5min at 48C.Then, the cells were homogenized with a pellet pestle
(Sigma-Aldrich), incubated for 5min and centrifuged for
10min at 10 000 rpm. The supernatant was purified using
m-C18 ZipTips (Millipore, Eschborn, Germany) and eluted
directly onto a MALDI sample plate with 0.4mL a-cyano-
4-hydroxycinnamic acid (CHCA, Bruker Daltonics, Bremen,
Germany; 5mg/mL in 0.2% aqueous TFA/acetonitrile (3:2)).
Liquid chromatographyThe purified and dried peptides were dissolved in 5mL
sample solution (0.1% TFA in water) and loaded onto
an UltiMateTM 3000 nanoLC (Dionex, Idstein, Germany)
equipped with a FLM 3X00 (integrated flow manager) and
micro-vacuum degasser. Peptide separation was performed
on a C18 PepMap 100 column (3.0mm, 150mm� 75mm) for
gradient elution (eluent A: 0.05% TFA in water and 5%
acetonitrile; eluent B: 0.05% TFA in 95% acetonitrile) from
0 to 50% eluent B within 40min and for further 5min from
50% to 90% eluent B. After 15min, the eluted sample
fractions were continuously diluted with 0.5mL/min CHCA
and spotted onto a MALDI target using a Probot (LC-
Packings/Dionex, California, USA) with an interval of 1min
resulting in 30 fractions.
MALDI-TOF/TOF MSThe mass spectra were recorded with the software FlexCon-
trol 2.4 (Bruker Daltonics, Bremen, Germany) using a
MALDI-tandem time-of-flight (TOF/TOF) mass spectro-
meter (Ultraflex II, Bruker Daltonics, Bremen, Germany).
External calibration was performed with the peptides kemp-
tide, bradykinin, substance P free acid, glu-fibrinopeptide B,
and dynorphin A (Sigma-Aldrich, Steinheim, Germany). The
peak lists were created from the raw data by FlexAnalysis 2.4
(Bruker Daltonics, Bremen, Germany) using mass-range
700–4000Da (MS), and minimal signal-to-noise (S/N) ratio 6
(MS) as peak filter settings. MS/MS spectra of pre-selected
peaks were generated manually. The MS/MS peptide
identifications were achieved by database comparisons
(SwissProt database, 20050621, human, 12359 sequences)
using Biotools 3.0 (Bruker Daltonics, Bremen, Germany)
in conjunction with the Mascot in-house version 2.0.0.25 A
tolerance of 50 ppm was used for the precursor mass and
Copyright # 2007 John Wiley & Sons, Ltd.
0.4Da for MS/MS fragments. Moreover, taxonomy human,
none enzyme, andN-acetyl (protein) as variablemodification
were selected. Peptides were considered to be identified
when the probability based individual ion scores were at
least 40with rank 1. For relative quantification, the intensities
of labeled and unlabeled peptides with a shift of 6.02Da for
arginine and lysine and 5.02Da for proline-containing
peptides (as a side product of the arginine biosynthesis)
were determined with the program FlexAnalysis 2.4 (Bruker
Daltonics, Bremen, Germany).
RESULTS
Peptide mass profiling of differentcell lines by MALDI-MSThe objective was to establish a fast and sensitive procedure
to determine the incorporation of arginine and lysine in
peptides from cells cultured in SILAC medium. After SILAC
treatment, 1� 106 cells were incubated with 100mL 0.1% TFA
for 5min, homogenized, and centrifuged. The peptides in the
supernatant were directly purified and concentrated with
m-C18 ZipTips, spotted onto aMALDI target and the peptide
mass profile was recorded. Representative mass spectra of
Jurkat T cells, HeLa cells, EBV-transformed human B cells,
and mouse embryonic fibroblasts (MEF) are shown in Fig. 1.
The peptide mass profiles of the three human cell lines were
comparable, although differences in intensity of the detected
masses were observed. The mouse cell line showed similar
peptide mass profiles in comparison to the human cells.
However, a peptide with significant intensity at m/z 2092.98
was detected in MEF cells, whereas a peptide was missing
which was detected with high intensity at m/z 2076.97 in
human cells. MS/MS analysis of both peptides confirmed the
known sequence variation of glyceraldehyde-3-phosphate
dehydrogenase from phenylalanine in human to tyrosine in
mouse within the same sequence. The most prominent
mass peaks of the peptide mass profiles were subjected to
fragmentation by MALDI-MS/MS in order to identify the
proteins these peptides were derived from (Fig. 1). All
peptides were fragments of proteins, predominantly derived
from, or close to, the amino- or carboxy-termini (Fig. 1,
Table 1). The most intense mass peaks corresponded
to 10 kDa heat shock protein (CH10; m/z 1551.84, and
1397.71), glyceraldehyde-3-phosphate dehydrogenase (G3P2;
m/z 2077.01, 1690.81, and 1169.61), macrophage migration
inhibitory factor (MIF; m/z 1785.98), peptidyl-prolyl cis-trans
isomerase A (PPIA; m/z 1716.93, 1864.00, 1474.78), actin
(ACTB; m/z 2409.23, 2223.13), GTP-binding nuclear protein
Ran (RAN; m/z 1372.78), and nucleophosmin (NPM;
m/z 1003.57). In addition, fragmentation patterns with
polyglutamic acid stretches were detected for the mass
peaks at m/z 1101.42 and 1230.46, which could correspond to
different proteins.
Peptide mass profiling of JurkatT cells by LC/MALDI-MS/MSThe peptides derived from Jurkat T cells were further
separated by nano-liquid chromatography (LC). The LC
fractions were analyzed by MALDI-MS to detect the
same peptides as in the peptide mass profiles. In summary,
Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
DOI: 10.1002/rcm
Figure 1. Peptide mass profiles of different cell lines. Peptide mass profiles of Jurkat T cells (A), HeLa cells (B),
EBV-transformed human B cells (C), and mouse embryonic mouse cells (D) are shown. Proteins that give rise to
peptides causing themajor mass peaks are shownwith their human Swiss-Prot entry name. The significant mass peak
at m/z 2092.98 in MEF cells (D) corresponds to glyceraldehyde-3-phosphate dehydrogenase of mouse
(G3P_MOUSE).
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
DOI: 10.1002/rcm
Rapid determination of SILAC incorporation 3921
Table 1. Proteins identified by peptide mass profiling of human cells. The human entry names, protein names, and accession
numbers according to Swiss-Prot/TrEmbl are displayed. Amino- and carboxy-terminal peptides are highlighted in gray at protein
range. Proteins identified bymanual de novo sequencing are depicted at Mascot score and the identified amino acids are shown at
peptide sequence as bold letters. Furthermore, two different sequenceswithin the same protein entry were found for actin (#), most
likely due to an exchange of cysteine to tryptophan. Furthermore, the mitochondrial precursor of 60 kDa heat shock protein was
deleted from the sequence to define the protein range (�)
Entry name Protein name
Accession
no.
Mascot
score
Peptide
mass
Protein
range Peptide sequence P1/P1́ P1/P1́
ACTB Actin, cytoplasmic 1 P60709 82 2223.09 357–375(375) ISKQEYDESGPSIVHRKCF WI
de novo 2306.16 357–375(375)# ISKQEYDESGPSIVHRKWF WI
96 2409.17 356–375(375) WISKQEYDESGPSIVHRKCF MW
ALDOA Fructose-bisphosphate aldolase A P04075 de novo 837.39 357–363(363) FVSNHAY LF
40 1655.90 63–78(363) LTADDRVNPCIGGVIL LF LF
BACH Cytosolic acyl coenzyme A
thioester hydrolase
O00154 57 1963.98 310–327(380) VDADPVVDSSQKRYRAAS LV SA
CH10 10 kDa heat shock protein, mitochondrial P61604 41 1176.64 1–10(101) Ac-AGQAFRKFLP PL
57 1289.73 1–11(101) Ac-AGQAFRKFLPL LF
73 1397.71 90–101(101) FRDGDILGKYVD LF
61 1551.82 1–13(101) Ac-AGQAFRKFLPLFD DR
CH60 60 kDa heat shock protein, mitochondrial P10809 50 2118.14 1–20(547)� AKDVKFGADARALMLQGVDL LL
89 2926.33 518–547(547)� VVVTEIPKEEKDPGMGAMGGMGGGMGGGMF EV
COF1 Cofilin-1 P23528 95 1819.86 84–98(165) YDATYETKESKKEDL LY LV
DUT Deoxyuridine 5́-triphosphate
nucleotidohydrolase, mitochondrial
P33316 54 1995.94 233–252(252) VQALDDTERGSGGFGSTGKN EV
EF1A1 Elongation factor 1-alpha 1 P68104 40 1198.59 415–424(462) FSDYPPLGRF SF FA
ENOA Enolase alpha P06733 75 1764.87 283–297(443) IKDYPVVSIEDPFDQ FI QD
46 2159.11 24–44(433) FTSKGLFRAAVPSGASTGIYE LF EA
63 2324.23 1–20(433) Ac-SILKIHAREIFDSRGNPTVE EV
53 2651.41 1–23(433) Ac-SILKIHAREIFDSRGNPTVEVDL LF
G3P2 Glyceraldehyde-3-phosphate P04406 60 1264.52 313–322(334) YDNEFGYSNR WY RV
dehydrogenase, liver 69 1577.69 313–325(334) YDNEFGYSNRVVD WY DL
57 1586.79 77–90(334) QERDPSKIKWGDAG FQ GA
52 1690.77 313–326(334) YDNEFGYSNRVVDL WY LM
47 1949.94 77–93(334) QERDPSKIKWGDAGAEY FQ YV
78 1963.88 310–325(334) ISWYDNEFGYSNRVVD LI DL
76 2076.97 310–326(334) ISWYDNEFGYSNRVVDL LI LM
H2AL Histone H2A.1 Q93077 90 1801.08 114–129(129) VLLPKKTESHHKAKGK AV
H2AX Histone H2A.x P16104 46 2650.53 114–140(142) VLLPKKTSATVGPKAPSGGKKATQASQ AV QE
IPYR Inorganic pyrophosphatase Q15181 57 1427.63 1–13(278) Ac-SGFSTEERAAPFS SL
KPYM Pyruvate kinase isozymes M1/M2 P14618 de novo 2093.03 1–19(530) Ac-SKPHSEAGTAFIQTQQLHA AA
de novo 2295.11 1–21(530) Ac-SKPHSEAGTAFIQTQQLHAAM MA
METK S-adenosylmethionine synthetase gamma P31153 71 1406.61 1–12(386) Ac-MNGQLNGFHEAF FI
MIF Macrophage migration inhibitory factor P14174 66 1438.77 4–17(114) IVNTNVPRASVPDG FI GF
52 1585.83 3–17(114) FIVNTNVPRASVPDG MF GF
53 1785.95 4–20(114) IVNTNVPRASVPDGFLS FI SE
47 1933.02 3–20(114) FIVNTNVPRASVPDGFLS MF SE
44 2028.08 3–22(114) IVNTNVPRASVPDGFLSEL FI LT
79 2489.32 23–45(114) TQQLAQATGKPPQYIAVHVVPDQ LT QL
NPM Nucleophosmin P06748 48 1003.55 288–294(294) WQWRKSL LW
PAIRB Plasminogen activator inhibitor 1
RNA-binding protein
Q8NC51 54 2003.94 1–18(392) PGHLQEGFGCVVTNRFDQ QL
PPIA Peptidyl-prolyl cis-trans isomerase A P62937 63 1141.59 8–18(164) DIAVDGEPLGR FD RV
78 1288.65 7–18(164) FDIAVDGEPLGR FF RV
41 1716.88 8–23(164) DIAVDGEPLGRVSFEL FD LF
42 1863.95 7–23(164) FDIAVDGEPLGRVSFEL FF LF
PROF1 Profilin-1 P07737 90 1988.08 40–58(139) VNITPAEVGVLVGKDRSSF FV FY
142 2759.51 32–58(139) AAVPGKTFVNITPAEVGVLVGKDRSSF WA FY
PSA7 Proteasome subunit alpha type 7 O14818 52 1994.95 1–18(248) Ac-SYDRAITVFSPDGHLFQ QV
RAN GTP-binding nuclear protein Ran P62826 58 1357.70 1–12(205) Ac-AAQGEPQVQFKL LV
48 1372.76 165–176(205) ARKLIGDPNLEF LA FV
RLA2 60S acidic ribosomal protein P2 P05387 40 1115.59 1–9(115) MRYVASYLL LA
ROA1 Heterogeneous nuclear P09651 40 1147.53 30–38(371) RSHFEQWGT LR TL
ribonucleoprotein A1 43 2172.16 1–19(371) Ac-SKSESPKEPEQLRKLFIGG GL
RS17 40S ribosomal protein S17 P08708 79 1957.03 117–134(134) QVTQPTVGMNFKTPRGPV LQ
SODC Superoxide dismutase P00441 145 2072.11 1–20(153) Ac-ATKAVCVLKGDGPVQGIINF FE
STMN1 Stathmin P16949 73 2106.07 1–19(148) Ac-ASSDIQVKELEKRASGQAF FE
65 2348.20 1–21(148) Ac-ASSDIQVKELEKRASGQAFEL LI
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
DOI: 10.1002/rcm
3922 F. Schmidt et al.
Rapid determination of SILAC incorporation 3923
56 peptides of 27 proteins could be identified by MALDI-
MS/MS (Table 1). Eighteen amino-terminal sequences,
including 15 acetylated forms, 10 carboxy-terminal sequen-
ces, and 14 sequences originating less than ten amino acids
from the terminal ends, were detected (Table 1). Several
proteins yielded more than one peptide, usually with
overlapping sequences. Peptides from the amino- and the
carboxy-terminal ends were identified for the 10 kDa and
Figure 2. Peptide mass profiles of Jurkat Tcells. Peptide mas
heavy-13C6 (B); arginines and/or lysines are shown. In additi
detected peptides contained arginines and lysines that can be
mixture (C).
Copyright # 2007 John Wiley & Sons, Ltd.
60 kDa heat shock proteins (Table 1). Cleavage tended to
occur predominantly at hydrophobic amino acid residues.
Leucine (35 times), phenylalanine (34 times), valine (15 times),
alanine (12 times) and isoleucine (10 times) were the most
frequent amino acids N-terminal to the cleavage site, but
valine and isoleucine were not found at position P1.
Furthermore, tryptophan (8 times) was found relatively
often, considering the low occurrence of this amino acid in
s profiles of Jurkat Tcells cultured with light-12C6 (A) and
on, the 1:1 mixture is depicted (C). Notably, most of the
seen by the occurrence of numerous mass pairs in the
Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
DOI: 10.1002/rcm
3924 F. Schmidt et al.
the human proteome. On the other hand, asparagine,
cysteine, histidine, and lysine were never observed to be
involved in cleavage and proline was found only once.
The observed cleavage specificity is typical for the acidic
proteases cathepsin D and E.26 Accordingly, the addition of
pepstatin A to the 0.1% TFA solution, a known inhibitor of
acidic proteases, abolished the production of the peptides.
Rapid determination of arginineand lysine incorporation by SILACPeptide mass profiling was employed to determine the
amino acid incorporation rate during cell culture. Compari-
son of the peptide mass profiles of Jurkat T cells cultured
with light-12C6 (Fig. 2(a)) and heavy-13C6 (Fig. 2(b)) arginine
and lysine showed that the most abundant peptides
displayed mass differences of 6, 12, and 18Da. These mass
differences were in accordance with the number of arginines
and lysines within the peptide sequences (Table 1). The lack
of peptides with light arginines and lysines in the peptide
mass profile of Jurkat T cells cultured with heavy amino
acids confirmed full incorporation of the amino acids and
was observed for all detected peptides (Fig. 2(b)). Moreover,
the ratio of light to heavy (L/H) can be determined bymixing
the samples. A 1:1 mixture of 106 cells from both light and
heavy cultures revealed a peptide mass profile with more
Figure 3. Peptide mass pairs of a 1:1 mixture of Jurkat T cell
Six peptide mass pairs corresponding to six different proteins w
(L/H) ratio was calculated by the intensities of the peptide pairs.
mass peaks at þ5Da to the heavy peptides were considered f
Copyright # 2007 John Wiley & Sons, Ltd.
than 10 mass pairs with high intensity and S/N ratio
(Fig. 2(c)). The L/H ratio of the individual mass pairs was on
average 1.02� 0.07 (Figs. 2(c) and 3). Interestingly, full
incorporation of heavy amino acids in Jurkat T and HeLa
cells was observed for all identified proteins as early as two
cell splittings (96 h) after exposure of the SILAC medium
(Fig. 4). However, the slower growing MEF cells required
four cell splittings for full incorporation of arginine and
lysine.
Rapid determination of conversionof 13C6-arginine into 13C5-prolineSome cell lines convert 13C6-arginine into
13C5-proline, which
results in the formation of two distinct peak clusters for all of
the labeled proline-containing peptides.27 The correct peak
intensity of the heavy state is therefore the sum of both the13C6-arginine and the 13C5-proline mass peaks. Actually,
some of the identified peptides contained prolines (Table 1).
Considering the four cell lines used, significant additional
mass peaks that were þ5Da compared to the peptide with
heavy arginines and/or lysines were observed only in Jurkat
T cells (Fig. 3). The conversion of 13C6-arginine into13C5-proline
was observed as early as one round of sub-culture post-SILAC
addition, which corresponded to 48 h of SILAC exposure
(Figs. 4(a) and 4(b)). The conversion increased only slightly,
s cultured with either 12C6- or13C6-arginines and lysines.
ere selected from those shown in Fig. 2. The light to heavy
Due to the formation of 13C5-proline, the intensities of the
or the calculation of the ratios.
Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
DOI: 10.1002/rcm
Figure 4. Peptide mass profiling to monitor arginine and lysine incorporation efficiency and conversion of proline. Full
incorporation of arginine and lysine was observed in Jurkat T cells after the second round of sub-culture because the light
form (m/z 1551.83) of the peptide Ac-AGQAFRKFLPLFD of the mitochondrial 10 kDa heat shock protein was only detected
after one cell splitting (A, B). The formation of 13C5-proline detected atm/z 1568.89 increased only marginally between split 2
and 8 (A). Halving the amount of added arginine slightly decreased the conversion of 13C6-arginine into 13C5-proline while
preserving the incorporation efficiency of arginine.
Rapid determination of SILAC incorporation 3925
exemplified for the most abundant mass m/z 1563.86, from
15% after two and 19% after eight rounds of sub-culture
(Fig. 4(a)). Similar values were observed for other proline-
containing peptides (Fig. 3). The conversion of 13C6-arginine
into 13C5-proline could be reduced to some extent using
50mg/L 13C6-arginine in RPMI medium (Fig. 4(b)) instead
of 100mg/L (Fig. 4(a)). Notably, the same amount of12C6-arginine must be used for relative quantification
experiments.
DISCUSSION
The incorporation of amino acids by SILACmust be checked
for every cell line. The cell population of the heavy state
must be analyzed to determine that the heavy amino acids
are incorporated and that the light amino acids are not
detectable. This analysis usually requires up to 2 days.24 In
this study, we exploited protein cleavage products of
different human cell lines to follow and determine the
Copyright # 2007 John Wiley & Sons, Ltd.
incorporation rate of amino acids by SILAC within less than
2 h. Similar peptide mass profiles were obtained from four
different cell lines and revealed a number of well-known
abundant proteins. This suggests that the peptide mass list
presented here (Table 1) could be of general use. Interest-
ingly, the protein fragments revealed that cleavage was
mainly due to cathepsin D and E activity and most peptides
were derived from, or close to, either the amino- or
carboxy-termini. Significantly, the observed peptide mass
profile contained fragments derived from several proteins
and was not dominated by one or two extremely abundant
proteins. Hence, we have developed a simple, direct, and
rapid mass spectrometric approach to determine the
incorporation and interconversion of amino acids by SILAC.
Our approach is significantly faster than other methods to
determine and normalize the relative quantities of light and
heavy labeled isotopes. Moreover, it can be used to rapidly
monitor whether full incorporation of labeled amino acids by
SILAC has occurred such that the samples can be employed
in subsequent biological experiments.
Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
DOI: 10.1002/rcm
3926 F. Schmidt et al.
REFERENCES
1. Ong SE, Mann M. Nat. Chem. Biol. 2005; 1: 252.2. Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H,
Pandey A, Mann M. Mol. Cell. Proteomics 2002; 1: 376.3. Olsen JV, Blagoev B, Gnad F, Macek B, Kumar C, Mortensen
P, Mann M. Cell 2006; 127: 635.4. Bose R, Molina H, Patterson AS, Bitok JK, Periaswamy B,
Bader JS, Pandey A, Cole PA. Proc. Natl. Acad. Sci. USA 2006;103: 9773.
5. Cantin GT, Venable JD, Cociorva D, Yates JR III. J. ProteomeRes. 2006; 5: 127.
6. Gronborg M, Kristiansen TZ, Iwahori A, Chang R, Reddy R,Sato N, Molina H, Jensen ON, Hruban RH, Goggins MG,Maitra A, Pandey A. Mol. Cell. Proteomics 2006; 5: 157.
7. Hwang SI, Lundgren DH, Mayya V, Rezaul K, Cowan AE,Eng JK, Han DK. Mol. Cell. Proteomics 2006; 5: 1131.
8. Liang X, Zhao J, HajivandiM,WuR, Tao J, Amshey JW, PopeRM. J. Proteome Res. 2006; 5: 2632.
9. Selbach M, Mann M. Nat. Methods 2006; 3: 981.10. Thiede B, Kretschmer A, Rudel T. Proteomics 2006; 6: 614.11. Vinther J, Hedegaard MM, Gardner PP, Andersen JS, Arc-
tander P. Nucleic Acids Res. 2006; 34: e107.12. Amanchy R, Kalume DE, Iwahori A, Zhong J, Pandey A.
J. Proteome Res. 2005; 4: 1661.13. Andersen JS, Lam YW, Leung AK, Ong SE, Lyon CE,
Lamond AI, Mann M. Nature 2005; 433: 77.
Copyright # 2007 John Wiley & Sons, Ltd.
14. Gruhler A, Olsen JV, Mohammed S, Mortensen P, Faerge-man NJ, Mann M, Jensen ON. Mol. Cell. Proteomics 2005;4: 310.
15. Romijn EP, Christis C, Wieffer M, Gouw JW, Fullaondo A,van der Sluijs P, Braakman I, Heck AJ. Mol. Cell. Proteomics2005; 4: 1297.
16. Blagoev B, Ong SE, Kratchmarova I, MannM.Nat. Biotechnol.2004; 22: 1139.
17. de Hoog CL, Foster LJ, Mann M. Cell 2004; 117: 649.18. Everley PA, Krijgsveld J, Zetter BR, Gygi SP. Mol. Cell.
Proteomics 2004; 3: 729.19. Ibarrola N, Molina H, Iwahori A, Pandey A. J. Biol. Chem.
2004; 279: 15805.20. Ong SE, Mittler G, Mann M. Nat. Methods 2004; 1:
119.21. Blagoev B, Kratchmarova I, Ong SE, Nielsen M, Foster LJ,
Mann M. Nat. Biotechnol. 2003; 21: 315.22. Bradford MM. Anal. Biochem. 1976; 72: 248.23. Lowry OH, Rosbrough NJ, Farr AL, Randall RJ. J. Biol. Chem.
1951; 193: 265.24. Ong SE, Mann M. Nat. Protoc. 2006; 1: 2650.25. Perkins DN, Pappin DJ, Creasy DM, Cottrell JS. Electropho-
resis 1999; 20: 3551.26. Arnold D, Keilholz W, Schild H, Dumrese T, Stevanovic S,
Rammensee HG. Eur. J. Biochem. 1997; 249: 171.27. Ong SE, Kratchmarova I, Mann M. J. Proteome Res. 2003;
2: 173.
Rapid Commun. Mass Spectrom. 2007; 21: 3919–3926
DOI: 10.1002/rcm