JOURNAL OF MASS SPECTROMETRY

J. Mass Spectrom. 2001; 36: 616–625

Investigation of tyrosine nitration in proteins by mass

spectrometry
Ann-Sofi Petersson,1,2 Hanno Steen,2 Dário E. Kalume,2 Kenneth Caidahl1 and
Peter Roepstorff2∗
1
Department of Clinical Physiology, Sahlgrenska University Hospital, Göteborg, Sweden
2
Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense University, Odense M, Denmark

Received 18 December 2000; Revised 17 January 2001; Accepted 20 February 2001

In vivo nitration of tyrosine residues is a post-translational modification mediated by peroxynitrite that

may be involved in a number of diseases. The aim of this study was to evaluate possibilities for site-specific
detection of tyrosine nitration by mass spectrometry. Angiotensin II and bovine serum albumin (BSA)
nitrated with tetranitromethane (TNM) were used as model compounds. Three strategies were investigated:
(i) analysis of single peptides and protein digests by matrix-assisted laser desorption/ionization (MALDI)
peptide mass mapping, (ii) peptide mass mapping by electrospray ionization (ESI) mass spectrometry and
(iii) screening for nitration by selective detection of the immonium ion of nitrotyrosine by precursor ion
scanning with subsequent sequencing of the modified peptides. The MALDI time-of-flight mass spectrum
of nitrated angiotensin II showed an unexpected prompt fragmentation involving the nitro group, in
contrast to ESI-MS, where no fragmentation of nitrated angiotensin II was observed. The ESI mass spectra
showed that mono- and dinitrated angiotensin II were obtained after treatment with TNM. ESI-MS/MS
revealed that the mononitrated angiotensin II was nitrated on the side-chain of tyrosine. The dinitrated
angiotensin II contained two nitro groups on the tyrosine residue. Nitration of BSA was confirmed by
Western blotting with an antibody against nitrotyrosine and the sites for nitration were investigated by
peptide mass mapping after in-gel digestion. Direct mass mapping by ESI revealed that two peptides were
nitrated. Precursor ion scanning for the immonium ion for nitrotyrosine revealed two additional partially
nitrated peptides. Based on the studies with the two model compounds, we suggest that the investigation
of in vivo nitration of tyrosine and identification of nitrated peptides might be performed by precursor ion
scanning for the specific immonium ion at m/z 181.06 combined with ESI-MS/MS for identification of the
specific nitration sites. Copyright  2001 John Wiley & Sons, Ltd.

KEYWORDS: mass spectrometry; nitrotyrosine; peptides; protein; immonium ion; precursor ion scanning

INTRODUCTION or generated by protein hydrolysis.4 – 11 The nitrotyrosine lev-

els observed in these studies varies considerably and, lately,
Nitration of free or protein-incorporated tyrosine residues
it has been shown that nitration may be an artifact obtained
is a post-translational modification, which can occur in
during derivatization.5,12 It is, still unclear however, whether
cells during oxidative stress and inflammation, most likely
tyrosine nitration is random or whether it is protein and site-
through generation of peroxynitrite (ONOO
), a potent oxi-
specific. Therefore, methods that can detect nitrated proteins
dizing and nitrating agent produced by the reaction of
and also determine the nitration site are required.
superoxide anion in the presence of excess nitric oxide.1,2
Mass spectrometry (MS) has become an increasingly
Protein nitration has been observed in connection with more
important technique in the determination of protein mod-
than 60 human disorders, mostly detected by immunodetec-
ifications, either of natural or of artificial nature. The
tion with antibodies directed against nitrotyrosine.3 Most of
two ionization techniques widely used in protein studies
the studies published so far were focused on the detection
and/or quantification of free nitrotyrosine either in plasma are matrix-assisted laser desorption/ionization (MALDI) in
combination with time-of-flight (TOF) mass analyzers13,14
and electrospray ionization (ESI) in combination with a vari-
Ł Correspondence to: P. Roepstorff, Department of Biochemistry
ety of mass analyzers.15
and Molecular Biology, University of Southern Denmark, Odense
University, DK-5230 Odense M, Denmark. E-mail: roe@bmb.sdu.dk A number of studies to investigate nitration of tyrosine
Contract/grant sponsor: Swedish Research Council; residue(s) in peptides by ESI-MS have been reported.16 – 20
Contract/grant number: K99-04RM-13192-01. Greis et al.16 treated surfactant protein A with tetrani-
Contract/grant sponsor: Swedish Heart and Lung Foundation.
Contract/grant sponsor: Göteborg University, Sweden. tromethane (TNM), sequenced the tryptic peptides by liquid
Contract/grant sponsor: Jubileumsfonden, Göteborg, Sweden. chromatography (LC) ESI-MS/MS to determine the specific

DOI: 10.1002/jms.161 Copyright  2001 John Wiley & Sons, Ltd.


nitration sites. Yi et al.17 used LC/ESI-MS and LC/ESI-
MS/MS to analyze three short peptides nitrated with per-
oxynitrite. They also observed traces of dinitrotyrosine but
no formation of nitrophenylalanine. MacMillan-Crow et al.18
used high-performance liquid chromatography (HPLC) com-
bined off-line with ESI-MS/MS, to analyze the enzymatic
digest of the nitrated mitochondrial protein manganese
superoxide dismutase (MnSOD). Complete inhibition of
MnSOD was achieved when it was treated with perox-
ynitrite and the analysis revealed that only three of the in
total nine tyrosine residues were nitrated. Ducrocq et al.19
used HPLC combined off-line with liquid secondary ion
mass spectrometry for the analysis of angiotensin II treated
with a variety of nitration reagents. They observed nitration
on the tyrosine residue in angiotensin II and that nitration
resulted in total inhibition of the biological activity. Curcu-
ruto et al.20 detected, using LC/ESI-MS/MS, mononitration
of angiotensin II after treatment with peroxynitrite.
Precursor ion scanning is an established tandem MS
method for the detection of post-translational modifications
such as phosphorylation and glycosylation, preceded either
by LC21,22 or by direct analysis of unseparated peptide
mixtures.23,24 Precursor ion scanning requires a fragment
ion highly characteristic for the modified amino acid that
is stable enough to survive low energy collision-induced
dissociation (CID). We have now extended this method to
the application of the selective detection of protein nitration
in unseparated peptide mixtures. This is, to our knowledge,
the first time precursor ion scanning has been used for the
identification of nitrated peptides utilizing the characteristic
nitrotyrosine immonium ion at m/z 181.06. Additionally,
the different mass spectrometric techniques to investigate
tyrosine nitration in proteins were evaluated, and a strategy
for rapid screening for nitrated peptides, and also the exact
localization of nitrotyrosine residues, is proposed.
EXPERIMENTAL
In vitro nitration
Angiotensin II, human (Sigma, St Louis, MO, USA) or BSA,
fraction V (Sigma), approximately 1–6 mg ml
1 , dissolved
in 50 mM Tris (pH 8.8), was nitrated by addition of a 50-fold
molar excess of tetranitromethane (TNM) (Sigma-Aldrich, St
Louis, MO, USA) as described.25,26 The reaction was carried
out overnight at 4 ° C, and the modified BSA was rapidly
desalted with a PD-10 column (Pharmacia Biotech, Uppsala,
Sweden). Angiotensin II was, without further purification,
stored at
20 ° C.
Immunoaffinity control of nitrated BSA and
non-modified BSA
Dilution series of the model protein BSA and nitrated BSA
resulting in the application of 1–50 µg were analyzed by
electrophoresis on 7% Tris–acetate gels (NuPAGE, Novex,
San Diego, CA, USA). For visualization of proteins half
of the gel was stained with Coomassie Brilliant Blue and
for identification of the protein nitration the other half
was visualized by Western blotting, using monoclonal anti-
nitrotyrosine antibody (Upstate Biotechnology, New York,
USA).
Copyright  2001 John Wiley & Sons, Ltd.
MS detection of tyrosine nitration in proteins 617
Tryptic in-gel digestion
Protein bands were excised from the Coomassie Brilliant-
Blue-stained gel and subsequently reduced, alkylated and
in-gel digested with trypsin according to our in-house
procedure,27 which is a slightly modified version of the
procedure described by Rosenfeld et al.28
Purification of samples for ESI and MALDI-MS
Sample micro-purification and preparation for ESI-MS and
MALDI-MS were performed as described previously.23
Microcolumns were prepared with Poros R2 (Perceptive
Biosystems, Framingham, MA, USA) in GELoader tips
(Eppendorf, Hamburg, Germany).29 The peptide mixture
was dissolved in 5% formic acid and loaded on to the
column, and thereafter washed with 20 µl of 5% formic
acid. For MALDI experiments, the peptides were eluted
from the microcolumns directly on to the target with ˛-
cyano-4-hydroxycinnamic acid (4-HCCA) matrix solution.
For ESI experiments the peptides were eluted from the
microcolumns with 50% methanol–5% formic acid directly
into the nanoelectrospray needles (MDS PROTANA, Odense,
Denmark).
Sample preparation for MALDI-MS
Dihydroxybenzoic acid (DHB)
DHB, obtained from Hewlett-Packard (Palo Alto, CA, USA),
was dissolved in 0.1% trifluoroacetic acid (TFA)–acetonitrile
(1 : 1, v/v) to a concentration of 20 µg µl
1 . A 0.5 µl volume
of the matrix solution was mixed on the MALDI target
with 0.5 µl of 20 mM diammonium hydrogencitrate solution
(Sigma) and 0.5 µl of sample solution. The sample was
allowed to dry at ambient temperature.
2,4,6-Trihydroxyacetophenone (THAP)–nitrocellulose
THAP (Aldrich, Gillingham, Dorset, UK) was dissolved in
methanol to a concentration of 200 µg µl
1 . Nitrocellulose
(NC) was dissolved in acetone to a concentration of 30 µg µl
1
and diluted with propan-2-ol to a final concentration of
15 µg µl
1 . THAP and NC solutions were mixed (4 : 1) and
0.2 µl of the mixture was placed on the target. The solution
spread out fast and formed a thin matrix layer. Equal
volumes of 20 mM diammonium hydrogencitrate solution
and micropurified sample solution (0.5 µl each, pH 4.5) were
mixed directly on the thin layer of THAP–NC.
˛-Cyano-4-hydroxycinnamic acid (4-HCCA)
4-HCCA dissolved as 20 µg µl
1 in methanol–0.5% formic
acid ( 1: 1,v/v; pH 1) was used for elution of samples from
the microcolumns directly on to the MALDI target.
Protein identification, MALDI-MS
Peptide mass maps were acquired in reflector mode on a
Reflex II MALDI-TOF mass spectrometer, equipped with
delayed ion extraction and reflector (Bruker Daltonik,
Bremen, Germany).
Peptide sequencing, ESI
NanoESI mass spectra were acquired on a Q-TOF mass
spectrometer (Micromass ,Manchester, UK) equipped with
J. Mass Spectrom. 2001; 36: 616–625
618 A.-S. Petersson et al.
a Z-spray nanoelectrospray ion source. The samples were
sprayed from a gold–palladium-coated glass capillary
supplied by MDS PROTANA. All spectra were obtained
in the positive ion mode. The potential voltage applied to the
nanoflow tip in the ion source and the nitrogen back-pressure
(<1 psi) was adjusted to produce a flow-rate (of the sample
solution) of about 30 nl min
1 , which allowed a long analysis
time (¾30 min on a sample of 1 µl). A potential of 800–1200 V
was applied to the nanoflow tip in the ion source. The cone
voltage was set to 30–55 V and the MCP (microchannel
plate detector) voltage was set to 2600–2800 V. The source
temperature was 40 ° C and nitrogen was used as a drying gas
(flow-rate ¾30 l h
1 ). The acquisition and the deconvolution
of data were performed on a Mass Lynx PC data system
(software version 3.2 for Windows NT). For all experiments
a sodium iodide solution (¾1 µg µl
1 in propan-2-ol–1%
formic acid (1 : 1, v/v)) was used for the TOF calibration.
For the CID experiments the precursor ion was selected in
the first quadrupole (Q1) and fragmented in the hexapole
collision cell with a collision energy varying from 20 to 40 eV.
Argon was used as the collision gas at a recorded pressure
of 6.0 ð 10
5 mbar. The fragment ions were analyzed in the
reflector TOF mass analyzer at a resolution (FWHM) of
¾5000.
Precursor ion scan, mass spectrometry
The precursor ion scanning experiments of the digest
of nitrated BSA followed by peptide sequencing were
carried out on a QSTAR Pulsar quadrupole time-of-flight
hybrid tandem mass spectrometer (AB/MDS Sciex, Toronto,
Canada). Precursor ion scans were acquired in the positive
ion mode with a dwell time of 50 ms at a step size of
0.5 thomson (Th) and with the Q2 -pulsing function turned
on. Nitrogen was used as the collision gas at a recorded
pressure of 6.3 ð 10
5 Torr (1 Torr D 133.3 pa). As a rough
rule, the Q0 voltage, which determines the collision energy
on the QSTAR, was ramped over the m/z range in a manner
corresponding to one tenth of the m/z-value of the precursor
ion. It should be noted that optimal collision energies might
need to be determined empirically for different instruments.
RESULTS AND DISCUSSION
In this study, angiotensin II (DRVYIHPF) and bovine serum
albumin (BSA) were used as model compounds. Nitrated
angiotensin II was analyzed directly by the different mass
spectrometric techniques whereas nitrated BSA was first
characterized by Western blotting using an anti-nitrotyrosine
antibody (Fig. 1). The Coomassie Brilliant Blue-stained gel
and also the Western blot showed the presence of the mono-,
di- and trimers of BSA. The specificity of the antibody for
nitrotyrosine was demonstrated by lack of reaction with non-
nitrated BSA and by the fact that the monoclonal antibody
immunoreacted with the nitrated mono-, di- and trimers in
a dose-dependent manner (Fig. 1).
Mass spectrometric analysis of nitrated
angiotensin II
The MALDI spectra of nitrated angiotensin II, showed
an unexpected prompt fragmentation involving the nitro
Copyright  2001 John Wiley & Sons, Ltd.
Figure 1. Gel electrophoresis of native and nitrated BSA on
Tris–acetate gels, 7%. Half of the gel was Coomassie Brilliant
Blue stained (A) and the other half was used for Western blot
analysis with a monoclonal anti-nitrotyrosine antibody,
followed by chemiluminiscence detection utilizing a second
antibody (B). CM D Color marker (molecular mass standard).
group (Fig. 2). In addition to MHC , peaks were observed
at [M C H
16]C , [M C H
30]C and [M C H
32]C . The
presence of these ions suggests that prompt fragmentation
and/or chemical reactions take place during the desorp-
tion/ionization process. A possible explanation is provided
by examining the UV spectra of nitrotyrosine at different
pH values.30 The absorption maximum is at 350 nm in acetic
medium (pH 3.5), i.e. very close to the laser wavelength
(337 nm), and at 440 nm at pH 10. The effect of the pH on
the formation of these ions was investigated using a vari-
ety of matrices the acetic matrices, i.e. 4-HCCA, [Fig. 2(A)]
and DHB, and the neutral THAP [Fig. 2(B)]. Differences in
the abundances of the ions were observed for the various
matrices, with clearly less fragmentation in THAP, but the
general pattern appeared to be similar. In agreement with
Sarver,31 we propose that the observed ions are the prod-
ucts of photoinduced chemical reactions involving the nitro
group. Thus the fragmentation pattern represents loss of one
and two oxygen atoms, i.e. NO2 ! NO and NO2 ! N,
respectively. [M C H
30]C most likely represents reduction
of the nitrogroup to amine NO2 ! NH2 . No peak corre-
sponding to the loss of the nitro group was observed at m/z
1045.5.
Upon nanoESI-MS, no fragmentation of nitrated angio-
tensin II was observed, as shown in the mass spectrum
(Fig. 3), in which peaks corresponding to the doubly charged
ions for mono- and dinitrated angiotensin II were observed
at m/z 546.38 and 568.85, respectively. In all spectra (MALDI
and ESI) of nitrated angiotensin II, peaks corresponding the
to loss of one and two protons from the molecular ion were
observed. Similar losses were not observed for unmodified
angiotensin II. Attempts to isolate these peaks and perform
MS/MS did not provide conclusive results. However, the fact
that they have approximately the same relative abundance in
the MALDI and ESI spectra suggests that they are caused by
a side-reaction during nitration rather than a fragmentation
in the mass spectrometer.
To establish the specific sites of nitration, nanoESI-
MS/MS was performed after selection of the ions for the
mono- and dinitrated angiotensin II. The CID spectrum of
J. Mass Spectrom. 2001; 36: 616–625

A 1092.5
20
Micropurification/4-HCCA
Intensity
16.0 Da
1062.5
14.0 Da
0
1060.0
B 1092.5
Micropurification/THAP/NC
10
Intensity
0
1060.0
Figure 2. MALDI spectra of nitrated angiotensin II, using two different matrices 4HCCA (A) and THAP-NC (B).
the ion at m/z 546.30 contained a nearly complete A and
B ion series and a smaller number of Y00 ions, demonstrat-
ing that the mononitrated angiotensin II was nitrated on
the tyrosine residue [Fig. 4(A)]. The CID spectrum obtained
with untreated angiotensin II showed a comparable frag-
mentation pattern indicating that the nitro group did not
influence the ion fragmentation (results not shown). The
CID spectra obtained after selection of the ion correspond-
ing to the dinitrated angiotensin II (m/z 568.80) showed the
presence of a doubly nitrated tyrosine residue [Fig. 4(B)].
No evidence was found for the presence of a dinitrated
angiotensin II that contained mononitrotyrosine and nitro-
phenylalanine. Mononitration of tyrosine in angiotensin II
but not dinitration has previously been reported using per-
oxynitrite nitration.20 Tyrosine nitration is likely to take place
in the 3- and 5- positions in the aromatic ring. No attempts
were made to determine which site was preferred in the
mononitrated form.
The studies of nitrated angiotensin II indicated that the
molecular ions for nitrated angiotensin II could be observed
with both MALDI-TOF and ESI. The characteristic ion pattern
observed in the MALDI spectra might be used as a diagnostic
pattern for identification of nitrated peptides. However, the
lability of the nitro group will result in distribution of
the signals for nitrated peptides on several peaks with a
concomitant reduction in sensitivity for observation of such
peptides. In contrast to this, the nitro group is fully stable
Copyright  2001 John Wiley & Sons, Ltd.
MS detection of tyrosine nitration in proteins 619
1076.5
16.0 Da
m/z 1095.0
1076.5
m/z 1095.0
under ESI and ESI-CID conditions, allowing site-specific
assignment of nitrated residues by MS/MS.
Analysis of nitrated BSA
To investigate the possibility of the observation of nitrated
peptides in peptide mass maps derived from gel bands,
parallel bands for BSA and nitrated BSA were excised
from the electrophoretic gel and subjected to tryptic in-
gel digestion prior to mass spectrometric analysis of the
resulting mixture.
Attempts to identify nitrated peptides by comparing
the MALDI peptide mass maps obtained with BSA and
nitrated BSA did not allow the identification of any signal
which might be derived from nitrated peptides in the latter.
This may partially be due to suppression of the signals
for the nitrated peptides, and partially because the signals
for the nitrated peptides are distributed on a number of
signals due to fragmentation of the nitro group. Only after
identification of nitrated peptides in the ESI spectra (see
below) was it possible to assign retrospectively some of
the weaker signals in the MALDI spectrum to nitrated
peptides. The ESI mass spectra of the mixture derived
by tryptic digestion of BSA and nitrated BSA contained
numerous peaks and identification of potentially nitrated
peptides could be performed by comparing the peptide
mass maps obtained from native and nitrated BSA [Fig. 5(A)
and (B)]. Two doubly charged peptide signals separated by
J. Mass Spectrom. 2001; 36: 616–625
620 A.-S. Petersson et al.

[M+2H] 2+
100
546.38

[NO2-TYR]-AII
Relative intensity %

[M+2H] 2+

568.85

[(NO2)2-TYR]-AII
0
546 m/z 570

Figure 3. Nanoelectrospray spectrum of nitrated angiotensin II showing that mono- and dinitrated angiotensin II were obtained by
treatment with TNM.

[M+2H]2+
546.30 Y”2 Y”1
100
A
D R V YNT I H P F
Relative abundance (%)

Y”2
263.16 B2 B3 B4 B5 B6
A3 A4 A5 A6
50
B2 A5 A6
A3 A4
664.40 801.48
Y” 1 272.15 B3 551.30
343.23 829.47
166.09 371.24 579.31 692.39
B5 B6
B4
0
200 300 400 500 600 700 800 900
Y”2 m/z
263.16 Y”2 Y”1
100
B
D R V Y2NT I H P F
Relative abundance (%)

B2 B4 B5 B6
A4 A5 A6
50 A4 A5
B2 A6
[M+2H]2+ 596.30 709.39
Y”1 272.16 568.80 846.47
B4 B5 B6
166.10
624.28 737.39 874.47

0
200 300 400 500 600 700 800
m/z
Figure 4. CID spectra of nitrated angiotensin II peptides after selection of the doubly charged ions for the mononitrated
angiotensin II at m/z 546.30 (A) and for the dinitrated angiotensin II at m/z 568.80 (B).

Copyright  2001 John Wiley & Sons, Ltd. J. Mass Spectrom. 2001; 36: 616–625
 740.4
100 582.3 784.4
636.7
(A)
700.4
464.3 653.4 820.4
547.3 831.7
% 682.4
590.3

0
582.3
100
653.4
(B)

Copyright  2001 John Wiley & Sons, Ltd.

464.3 636.7 700.4 740.4 820.4

Relative intensity %
547.3 784.4
% 486.8 763.4
590.4
∗ ∗ 846.9
∗ ∗
0
500 600 m/z 700 800

(C) (D)
[M+2H]2+ Y”12 Y”11 Y”10 Y”9 Y”8 Y”7 Y”6 Y”5 Y”4 Y”3 Y”2 Y”1
Y”2
Y6∗ Y5∗ Y”4 Y”3 Y”2 Y”1 486.79 Y”6 Y”5 Y”4 Y”3 Y”2 Y”1
100 100 274.19 L G E YNO2 G F Q N A L II V R
Y”3
Y L YNO2 E I A R YNO2 L Y E I A R
387.27 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12
B2 B3∗ B4∗ B5∗ B6∗ B2∗ B3∗ B4∗ B5∗ B6∗ Y”6
A1 A2 ∗
A1 A2∗ 685.43 Y”7
294.18 813.49
651.41 [M+2H]2+
A2∗ Y”4
Y5” 763.39
500.36
Y”5
A1 Y”9
50 B2∗ B5∗ 50 300.16 571.39
136.10 Y”1 1017.57
322.18 727.40 B3
175.13 954.39
Y”3 B6∗ 508.21 B8
359.28 171.12 TYR-NO2 B4 1025.42

Relative intensity (%)

Y”2 798.45 Y”8 Y”10
B2 B9
Relative intensity (%)

Y”1 246.19 Y6” 712.30 960.56 1225.62

B 3∗ Y”
4 B6 Y”12
175.14 764.50 Y6∗ 565.22 TYR-NO2
1251.61
485.26 488.33 B4∗ Y5∗ 809.49 B5
A∗1 A2B 614.31 696.41 840.35 B11 Y” 1411.68
11
86.12 181.09 2 B7 1138.49 1354.67
B10
0 0
100 200 300 400 500 600 700 800 200 400 600 800 1000 1200 1400
m/z
m/z

Figure 5. ESI mass spectra of the peptide mixtures derived by in-gel digestion with trypsin of native BSA (A) and TNM-nitrated BSA (B), ESI-CID spectrum of the nitrated peptide at m/z

J. Mass Spectrom. 2001; 36: 616–625

MS detection of tyrosine nitration in proteins

486.79 2C (C) and ESI-CID spectrum of the nitrated peptide at m/z 763.39 2C (D). Asterisks indicate a nitrated peptide.
621
622 A.-S. Petersson et al.
22.5 Th, at m/z 464.3 and 740.4 in Fig. 5(A) and at m/z 486.8
and 763.4 in Fig. 5(B), were readily identified, indicating
that the corresponding peptides were potentially nitrated.
Selection and MS/MS of the doubly charged ions at m/z
486.79 and 763.39 in the peptide mass maps of the nitrated
BSA [Fig. 5(C) and (D)] verified the nitration site to be on the
tyrosine residues. One of the peptides (YLYEIAR, m/z 486.8)
contained two tyrosine residues. Based on the abundance of
the B2 and Y005 fragment ions relative to the B*2 and Y*5 ions
(* indicating nitrotyrosine), it is seen that the first tyrosine
residue is predominantly nitrated [Fig. 5(C)]. In addition, it
can be seen in the product ion spectrum that the immonium
ion for nitrotyrosine m/z 181.06 was less abundant than
the immonium ion for tyrosine m/z 136.10. The N-terminal
tyrosine residue was found to be predominantly nitrated
with the result that the A1 ion from this peptide is congruent
with the immonium ion for nitrotyrosine. Therefore, an
abundant ion at m/z 181.06 was expected. These results
indicate that formation of immonium ions for nitrotyrosine
is suppressed. No evidence was found in the spectra for a
doubly nitrated YLYEIAR. However, the presence of this
compound cannot be entirely excluded since corresponding
signals may be suppressed. Selection and CID spectra of
the doubly charged ion at m/z 763.4 corresponding to the
second nitrated peptide resulted in an almost complete B and
Y ion series, confirming the sequence LGEY*GFQNALIVR
containing a nitrated tyrosine residue [Fig. 5(D)]. In this
spectrum, the peak corresponding to the immonium ion of
nitrotyrosine was also of very low abundance, indicating
that formation of this ion is unfavorable.
The results show that comparison of the peptide maps
obtained with the unmodified and the nitrated BSA allowed
the detection of the nitrated peptides. The signals for the
nitrated peptides were, however, of rather low abundance.
Therefore, it seems difficult to use peptide mass mapping for
the detection of in vivo nitration where only a minor fraction
of protein molecules are likely to be modified.
Two alternatives seem to be available for the detection
of nitrated peptides derived by proteolytic digestion of
proteins. One is LC-MS, provided that the nitrated peptides
are separated from the corresponding unmodified peptides,
and the other is selective detection of peptides yielding
characteristic fragment ions such as the immonium ion
for nitrotyrosine by precursor ion scanning. Preliminary
experiments demonstrated that separation of angiotensin II
and nitrated angiotensin II was possible by HPLC and that
the nitrotyrosine-containing angiotensin II have a slightly
longer retention time than unmodified angiotensin II (data
not shown). Thus, LC-MS and LC-MS/MS are indeed a
possibility for investigating protein nitration.
Screening for nitration by detection of the
immonium ion of nitrotyrosine by precursor ion
scanning
We decided, however, to investigate the potential of
precursor ion scanning for the detection of mono- and
dinitrated tyrosine residues by making use of the formation
of the immonium ions at m/z 181.06 for nitrotyrosine and at
m/z 226.0 for dinitrotyrosine. Although of low abundance,
these ions might be used to screen for tyrosine-nitrated
Copyright  2001 John Wiley & Sons, Ltd.
A 546.2
32
intensity
m/z
B 9
intensity
464.2 568.8
445.8
450 550 m/z 650 750
Figure 6. Precursor ion scans spectra of nitrated angiotensin
II for formation of the immonium ion at m/z 181.06 for
mononitrated tyrosine (A) and at m/z 226.0 for dinitrated
tyrosine (B).
peptides in the peptide mass maps. The results of precursor
ion scanning for the formation of the m/z 181.06 ion for
the mononitrated and the m/z 226.0 ion for the dinitrated
angiotensin II sample are shown in Fig. 6(A) and (B),
respectively. As expected, the immonium ion of nitrotyrosine
at m/z 181.06 is derived from the doubly charged peak at
m/z 546.2 and the immonium ion of dinitrotyrosine m/z
226.0 from the doubly charged peak at m/z 568.8. Some
additional peaks were found in the precursor ion spectrum
of dinitrotyrosine. These ions most likely represent skimmer-
induced fragment ions or minor impurities of truncated
angiotensin II, which were hidden in the chemical noise in
the mass spectra. Thus the ion at m/z 445.8 was determined
by MS/MS to represent dinitrated angiotensin II lacking the
two C-terminal amino acids.
Precursor ion scans for the formation of m/z 181.06
ion was also performed on the unseparated peptide mix-
tures derived by tryptic digestion of non-modified BSA
and nitrated BSA (Fig. 7(B) and 7(A), respectively). The
former experiment was performed as a control for false pos-
itives. Two peaks could be found which could be assigned
as triply and quadruply charged ions, corresponding to
the non-modified tryptic peptide T9 (45–63) GLVLIAF-
SQYQQPFDEHVK, at m/z 624.08 (4+) and at m/z 831.69
3C. The precursor ion spectrum obtained with the tryp-
tic mixture derived from nitrated BSA shows the presence
of four potentially nitrated peptides [Fig. 7(A)]. Based on
their MS and MS/MS data they were identified as T9 (45–63)
GLVLIAFSQY*QQPFDEHVK, T22 (161–167) Y*LYEIAR, T59
(421–433) LGEY*GFQNALIVR and T66 (469–482) MPCT-
EDY*LSLILNR (Y* indicates a nitrated tyrosine). Two of
these peptides, T22 and T59, were already identified in the
ESI peptide mass maps. Two additional peptides, T9 and
J. Mass Spectrom. 2001; 36: 616–625
 MS detection of tyrosine nitration in proteins 623

590.81
A 3+

MPCTEDYLSLILNR

GLVLIAFSQYLQQPFDEHVK

GLVLIAFSQYLQQPFDEHVK
GLVLIAFSQYLQQPFDEHVK
∗NO
2
Relative intensity (%)

LGEYGFQNALIVR

MPCTEDYLSLILNR
LGEYGFQNALIVR
TVMENFVAFVDK
MPCTEDYLSLIL-H2O
∗
∗
∗
∗
YLYEIAR

∗ 3+
3+ 2+ 763.66 3+ 846.43
486.76 4+ 2+
2+ 2+
2+

450 550 650 750 850

B GLVLIAFSQYLQQPFDEHVK

GLVLIAFSQYLQQPFDEHVK
Relative intensity (%)

4+
3+

450 550 650 750 850

Figure 7. Precursor ion scans for the formation of the ion at m/z 181.06 from tryptic in-gel digests of nitrated BSA (A) and
non-modified BSA (B). The signals observed in (B) represent false positives.

T66, were found to be nitrated using precursor ion scan- In spite of the low abundance of the immonium ion
ning and their identity was confirmed by MS/MS. Figure 8 and a few false-positive signals (readily identified based
shows the CID spectrum of the triply charged nitrated pep- on their MS and MS/MS data), precursor ion scanning
tide T66 at m/z 590.8 seen in Fig. 7(A), clearly identifying followed by analysis by MS/MS to confirm the peptide
the nitration site on the tyrosine residue. The false-positive identity and nitration was found to be a reliable and fast
signals observed for non-modified T9 were also present in method to identify nitrated peptides in complex peptide
the precursor ion spectrum of the digest of nitrated BSA mass maps derived by tryptic digestion of proteins. In this
showing that T9 is present in both the nitrated and the non- study, precursor ion spectra were generated by the software
modified form. A signal at m/z 700.1 was demonstrated by using a Q-STAR-pulsar quadrupole time-of-flight hybrid
MS/MS to be false positive derived from the non-nitrated mass spectrometer. Traditionally precursor ion scans are
peptide TVMENFVAFVDK. Investigation of the CID spectra performed by selecting the precursor ion scan mode on a
of the false positives showed that the peak at m/z 181.06 triple quadrupole mass spectrometer. In recent study,32 it
was barely visible. We believe that the presence of this ion is was demonstrated that similar sensitivity were obtained for
the result of several successive fragmentations caused by the precursor ion scanning in the two types of instruments.
high collision energy used for precursor ion scanning. Under
normal CID conditions this ion was not observed. Site specificity of tyrosine nitration in BSA
The reason for the failure to observe the nitrated peptide It is noticeable that even treatment with a strong nitrating
T66 at m/z 590.4 3C in the direct peptide mass maps can be reagent such as TNM resulted in nitration of four out of 21
explained by the presence of a signal at m/z 590.3 3C in the tyrosine residues in BSA. The fifth, Y163, the second tyrosine
spectra of the non-modified BSA [Fig. 5(A)], corresponding residue in peptide T22, was nitrated to a very low degree.
to [M C H C 2Na]3C of unmodified T66, resulting in a lack of The site-specific nitration is supported by previous studies
recognition of the nitrated peptide by comparing the spectra. of in vitro nitration of proteins.33,34 . An x-ray crystal structure
The peak at m/z 846.4 3C corresponding to the peptide T9 is available for human serum albumin (HSA), which is 76%
was not visible in the mass spectrum, most likely because the identical with BSA and with all tyrosine residues conserved
peptide is only partially nitrated, the resulting signal being except two. Two of the nitrated tyrosine residues are placed
of too low abundance. on the surface Y161 and Y424. Interestingly, the peptide T22,

Copyright  2001 John Wiley & Sons, Ltd. J. Mass Spectrom. 2001; 36: 616–625
624 A.-S. Petersson et al.

Y”3
402.30 Y”7 Y”6 Y”5 Y”4 Y”3 Y”2 Y”1
100

M P C T E D YNT L S L I L N R

B5 B3 B4 B5 B6 B7 B8 B9 B10 B11
Relative abundance

A2 A8
619.32

50

Y”2 Y”4 Y”6

Y”5
515.40 715.54
Y”1 289.21 B B4 628.49
3
389.21 490.24 B7
175.16 B6 Y”7 942.41 B8 B9 B10
734.35 B11
A2 828.61 1055.531142.56
86.14 1255.64
A8 1368.79
0
200 400 600 800 1000 1200 1400
m/z
Figure 8. ESI-CID spectra of the peak at m/z 590.8 3C in Fig. 7(A), identifying the nitration site on the tyrosine residue in the
peptide MPCTEDYLSLILNR, derived by tryptic in-gel digestion of nitrated BSA.

YLYEIAR, which is fully conserved in both species, was also
found nitrated in HSA by Mandapati35 in spite of the fact that
several other tyrosine residues seems readily accessible on
the surface. This indicates that tyrosine nitration is a highly
site-specific reaction most likely directed by the surrounding
amino acid residues. Surprisingly, one, Y476, is situated in
the cleft between the two domains and the other Y54 seems
to be buried in the inside of the structure.
CONCLUSION
MALDI and ESI mass spectrometry allow the assignment
of protein nitration. The prompt fragmentation of nitrated
peptides observed by MALDI-MS, however, reduces the
abundance of the signals for nitrated peptides resulting
in failure to observe such signals in complex peptide
mass maps. Differential peptide mass mapping by ESI-
MS allowed the identification of two out of four nitrated
peptides in BSA by comparing the spectra obtained from the
nitrated and unmodified protein. However, this procedure
failed to observe two partially nitrated peptides, and will
not be a reliable method for the detection of in vivo
protein nitration. Precursor ion scanning was found to be
a sensitive and specific method to identify nitrated peptides
in a complex peptide mixture combined with MS/MS. We
therefore suggest that the optimal method for studies of
in vivo protein nitration is analysis of peptide mixture
derived by proteolytic digestion of the protein by precursor
ion scanning, either directly of peptide mixtures or in
combination with LC-MS.
Acknowledgements
The study was supported by the Swedish Research Council
(K99-04RM-13192-01), the Swedish Heart and Lung Foundation,
Stockholm, Sweden, Göteborg University and Jubileumsfonden,
Göteborg, Sweden. The authors are grateful to Ms Lena Hultman for
the performance of immunoblotting studies. This work is part of the
activities of the Center for Experimental Bioinformatics, sponsored
by the Danish National Research Foundation.
REFERENCES
1. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA.
Proc. Natl. Acad. Sci. USA 1990; 87: 1620.
2. Saran M, Michel C, Bors W. Free Radic. Res. Commun. 1990; 10:
221.
3. Ischiropoulos H. Arch. Biochem. Biophys. 1998; 356: 1.
4. Ohshima H, Friesen M, Brouet I, Bartsch H. Food Chem. Toxicol.
1990; 28: 647.
5. Yi D, Ingelse BA, Duncan MW, Smythe GA. J. Am. Soc. Mass
Spectrom. 2000; 11: 578.
6. Kaur H, Lyras L, Jenner P, Halliwell B. J. Neurochem 1998; 70:
2220.
7. Crowley JR, Yarasheski K, Leeuwenburgh C, Turk J, Heinecke
JW. Anal. Biochem. 1998; 259: 127.
8. Herce-Pagliai C, Kotecha S, Shuker DE. Nitric Oxide 1998; 2: 324.
9. Rimbach G, Hohler D, Fischer A, Roy S, Virgili F, Pallauf J,
Packer L. Arch. Tierernahr. 1999; 52: 203.
10. Pennathur S, Jackson-Lewis V, Przedborski S, Heinecke JW. J.
Biol. Chem. 1999; 274: 34621.
11. Frost MT, Halliwell B, Moore KP. Biochem. J. 2000; 345 (Pt 3): 453.
12. Schwedhelm E, Tsikas D, Gutzki FM, Frolich JC. Anal. Biochem.
1999; 276: 195.
13. Karas M, Hillenkamp F. Anal. Chem. 1988; 60: 2299.
14. Hillenkamp F, Karas M, Beavis RC, Chait BT. Anal. Chem. 1991;
63: 1193A.
15. Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM. Science
1989; 246: 64.
16. Greis KD, Zhu S, Matalon S. Arch. Biochem. Biophys. 1996; 335:
396.
17. Yi D, Smythe GA, Blount BC, Duncan MW. Arch. Biochem. Bio-
phys. 1997; 344: 253.
18. MacMillan-Crow LA, Crow JP, Thompson JA. Biochemistry 1998;
37: 1613.
19. Ducrocq C, Dendane M, Laprevote O, Serani L, Das BC, Bou-
chemal-Chibani N, Doan BT, Gillet B, Karim A, Carayon A,
Payen D. Eur. J. Biochem. 1998; 253: 146.

Copyright  2001 John Wiley & Sons, Ltd. J. Mass Spectrom. 2001; 36: 616–625

20. Curcuruto O, Rovatti l, Pastorino A, Hamdan M. Rapid Commun.
Mass Spectrom. 1999; 13: 156.
21. Carr SA, Huddleston MJ, Bean MF. Protein Sci. 1993; 2: 183.
22. Huddleston MJ, Bean MF, Carr SA. Anal. Chem. 1993; 65: 877.
23. Wilm M, Neubauer G, Mann M. Anal. Chem. 1996; 68: 527.
24. Borchers CJ, Parker CE, Deterding LJ, Tomer KB. J. Chromatogr.
A 1999; 854: 119.
25. Sokolovsky M, Riordan JF, Vallee BL. Biochemistry 1966; 5: 3582.
26. Riordan JF, Sokolovsky M, Vallee BL. Biochemistry 1967; 6: 358.
27. Jensen ON, Larsen MR, Roepstorff P. Proteins: Struct. Funct.
Genet. 1998; 999: Suppl.2: 74.
28. Rosenfeld J, Capdevielle J, Guillemot JC, Ferrara P. Anal. Bio-
chem. 1992; 203: 173.
Copyright  2001 John Wiley & Sons, Ltd.
MS detection of tyrosine nitration in proteins 625
29. Gobom J, Nordhoff E, Mirgorodskaya E, Ekman R, Roepstorff P.
J. Mass Spectrom. 1999; 34: 105.
30. Crow JP, Ischiropoulos H. Methods Enzymol. 1996; 269:
185.
31. Sarver A, Scheffler NK, Shetlar D, Gibson BW. J. Am. Soc. Mass
Spectrom. 2001; 12: 439.
32. Steen H, Küster B, Mann M. J. Mass Spectrom. 2001; submitted.
33. Yamakura F, Taka H, Fujimura T, Murayama K. J. Biol. Chem.
1998; 273: 14085.
34. Viner RI, Ferrington DA, Williams TD, Bigelow DJ, Schoneich C.
Biochem. J. 1999; 340: 657.
35. Mandapati S. In Proceedings of the 48th ASMS Conference on Mass
Spectrometry and Allied Topics, 2000; WPF 230 (abstract).
J. Mass Spectrom. 2001; 36: 616–625