Covalent modification of actin by 4-hydroxy-trans-2-nonenal (HNE): LC-ESI-MS/MS evidence for Cys374 Michael adduction

Article abstract of DOI:10.1002/jms.872

We demonstrate for the first time, by a combined mass spectrometric and computational approach, that G- and F-actin can be covalently modified by the lipid-derived aldehyde, 4-hydroxy-trans-2-nonenal, providing information on the molecular mass of modified protein and the mechanism and site of adduction. ESI-MS analysis of actin treated with different molar ratios of HNE (1:1 to 1:20) showed the formation of a protein derivative in which there was an increase of 156 Da (42028 Da) over native actin (41872 Da), consistent with the adduction of one HNE residue through Michael addition. To identify the site of HNE adduction, G- and F-actin were stabilized by NaBH₄ reduction and digested with trypsin. LC-ESI-MS/MS analysis in data-dependent scan mode of the resulting peptides unequivocally indicated that Cys374 is the site of HNE adduction. Computational studies showed that the reactivity of Cys374 residue is due to a significant accessible surface and substantial thiol acidity due to the particular microenvironment surrounding Cys374. Copyright

Full text of DOI:10.1002/jms.872

JOURNAL OF MASS SPECTROMETRY
J. Mass Spectrom. 2005; 40: 946–954
Published online 2 June 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jms.872
Covalent modification of actin by 4-hydroxy-trans-
2
-nonenal (HNE): LC-ESI-MS/MS evidence for
Cys374 Michael adduction
Giancarlo Aldini,1 Isabella Dalle-Donne, Giulio Vistoli, Roberto Maffei Facino
∗
2
1
1
and Marina Carini1
1
Istituto Chimico Farmaceutico Tossicologico, University of Milan, Viale Abruzzi 42, I-20131 Milan, Italy
Dipartimento di Biologia, University of Milan, via Celoria 26, I-20133 Milan, Italy
2
Received 14 February 2005; Accepted 22 March 2005
We demonstrate for the first time, by a combined mass spectrometric and computational approach,
that G- and F-actin can be covalently modified by the lipid-derived aldehyde, 4-hydroxy-trans-2-nonenal,
providing information on the molecular mass of modified protein and the mechanism and site of adduction.
ESI-MS analysis of actin treated with different molar ratios of HNE (1 : 1 to 1 : 20) showed the formation
of a protein derivative in which there was an increase of 156 Da (42028 Da) over native actin (41872 Da),
consistent with the adduction of one HNE residue through Michael addition. To identify the site of HNE
4
adduction, G- and F-actin were stabilized by NaBH reduction and digested with trypsin. LC-ESI-MS/MS
analysis in data-dependent scan mode of the resulting peptides unequivocally indicated that Cys374 is
the site of HNE adduction. Computational studies showed that the reactivity of Cys374 residue is due
to a significant accessible surface and substantial thiol acidity due to the particular microenvironment
surrounding Cys374. Copyright  2005 John Wiley & Sons, Ltd.
KEYWORDS: actin; 4-hydroxy-trans-2-nonenal; Michael adduction; LC-ESI-MS/MS analysis; computational analysis; Cys374
as target residue
INTRODUCTION
hydrogen peroxide and hypochlorous acid, used as models
2
,3
of oxidant insult. A significant specific increase in the actin
carbonyl content has been reported in human brain regions
Actin is a protein directly involved in the conversion of
chemical energy into mechanical work, and the molecule
contains ATP as an essential component. Although muscle
actin (˛ isoform) has been the most studied, the protein
is widely found in nature and is a structural component
of the cytoskeleton (ˇ and ꢀ isoforms) of perhaps of all cell
types. The actin cytoskeleton exists in a dynamic steady state,
where monomeric actin (G-actin) polymerizes onto the fast-
growing ends of actin filaments (F-actin), and depolymerizes
off their slow-growing ends.
4
severely affected by Alzheimer’s disease, in colonic mucosa
5
from patients suffering from inflammatory bowel disease,
in post-ischemic isolated rat hearts, associated with signifi-
6,7
cant depression of post-ischemic contractile function, and
8
in septic rat diaphragms.
The introduction of carbonyl groups (aldehydes and
ketones) into proteins can arise at different sites and by
9
different mechanisms. Carbonyl derivatives are produced
by direct reaction of ROS with the side chains of arginine,
lysine, proline, and threonine residues. Carbonyls can
also be generated through oxidative cleavage of proteins
by either the ˛-amidation pathway or by oxidation of
glutamyl side chains, by secondary reaction of the primary
amino group of Lys residues with reducing sugars or
their advanced glycation end-products (AGEs), and by
Michael-addition reaction of cysteine, lysine or histidine
residues with intermediate lipoxidation products generated
during peroxidation of polyunsaturated fatty acids, such
as 2-propenal (acrolein), malondialdehyde (MDA), and 4-
hydroxynonenal (HNE).
In many mammalian cells, including hepatocytes, human
fibroblasts and platelets, the actin cytoskeleton is one of the
earliest targets for attack by reactive oxygen species (ROS)
1
including protein carbonylation. Carbonylation of actin has
been shown in vivo to affect actin filament dynamics and,
consequently, the morphology, contractility, and motility of
1
–3
cells. Enhancement of actin carbonylation, causing disrup-
tion of the actin cytoskeleton and loss of the barrier function,
has been seen in human intestinal cells after exposure to
Ł
Correspondence to: Giancarlo Aldini, Istituto Chimico
Farmaceutico Tossicologico, University of Milan, Viale Abruzzi 42,
2
Although actin carbonylation has been clearly evi-
denced by slot or Western immunoblot with anti-(2,4-
dinitrophenylhydrazone) (DNP) primary antibody, and after
0131 Milan, Italy. E-mail: giancarlo.aldini@unimi.it
Contract/grant sponsor: MURST.
Contract/grant sponsor: Cofinanziamento Programma Nazionale
1
–7
2004-2005.
1D- or 2D-electrophoresis,
the carbonylation mechanism,
Copyright  2005 John Wiley & Sons, Ltd.

LC-MS/MS evidence for actin modification by HNE 947
as well as the sites of carbonyl insertion, still remains
unknown.
was dissolved in 200 µl of water, mixed with 200 µl of
3 2
CH CN-H O-HCOOH (90 : 10 : 0.4, v/v/v), and infused into
ꢀ
1
Since actin contains several nucleophilic sites particularly
susceptible to aldehyde adduction (5 Cys, 9 His, and 19
Lys residues), to achieve a better understanding of the
molecular basis of carbonylation we investigated whether
actin is a target protein for modification by lipid-derived
aldehydes. This was done by incubating actin with HNE, one
of the most reactive and cytotoxic ˛,ˇ-unsaturated aldehydes
generated by the lipid peroxidation of polyunsaturated fatty
acids, and investigating the carbonylation mechanism and
the HNE modification sites of the protein by a combined
computational and mass spectrometric approach never
previously reported for actin.
the mass spectrometer at a flow-rate of 20 µl min under
the following instrumental conditions: positive-ion mode,
mass range m/z 700–2000; capillary temperature 220 °C;
spray voltage applied to the needle 5 kV; capillary voltage
46 V. The flow-rate of the nebulizer gas (nitrogen) was
ꢀ
1
0.5 L min . ESI-MS spectra were deconvoluted using the
software package Bioworks 3.1 (Thermoquest, Milan, Italy).
LC-ESI-MS/MS – Tryptic peptide maps were generated
by LC-ESI-MS/MS analysis in data-dependent scan mode.
For sample preparation, lyophilized target protein ꢂ250 µgꢃ
was dissolved in 100 µl 50 mM Tris-HCl (pH 8) containing
4
6 M guanidine HCl and 0.5 M NaBH . After 60 min at
3
(
7 °C, the sample was spiked with ˇ-mercaptoethanol
3 mM final concentration) and heated at 95 °C for 20 min.
After denaturation, the sample was diluted to 1 ml with
50 mM Tris-HCl (pH 7.6), 1 mM CaCl . A 500 µl-aliquot
was then digested with trypsin at a protease : protein
ratio of 1 : 20 (w/w) for 24 h at 37 °C. To terminate the
EXPERIMENTAL
Chemicals
2
All chemicals and reagents were of analytical grade and pur-
chased from Sigma-Fluka-Aldrich Chemical (Milan, Italy).
LC-grade and analytical-grade organic solvents were pur-
chased from Merck (Bracco, Milan, Italy). LC-grade water
was prepared with a Milli-Q water purification system (Mil-
lipore, Bedford, MA, USA). 4-Hydroxy-non-2-enal diethy-
lacetal (HNE-DEA) was prepared as reported by Rees
protease activity, samples were spiked with TCA (10%
final concentration), centrifuged at 18 000 rpm for 10 min
and 5 µl of the supernatant was injected into a quaternary
pump HPLC system (Surveyor LC system, Thermoquest,
Milan, Italy). Separations were done by reversed-phase
elution with an Agilent Zorbax SB-C18 column (4 mm i.d.,
particle size 3.5 µm) (CPS Analitica, Milan, Italy) protected
by an Agilent Zorbax R-P guard column, thermostated at
25 C. An 80-min linear gradient from 0 to 60% B was
used to elute the digested peptides with phase A (0.1%
1
0
et al. and 4-hydroxy-trans-2-nonenal (HNE) by the acid
treatment of HNE-DEA (1 mM HCl for 1 h at room temper-
ature) and quantitated by UV spectroscopy (ꢁ max 224 nm;
ꢀ
1
ꢀ1
°
ε13 750 L mol cm ). Sequence-grade modified trypsin was
obtained from Promega (Milan, Italy).
HCOOH in water) and B (CH
3 2
CN-H O-HCOOH; 90 : 10 : 0.1,
v/v/v); the mobile phase was delivered at a flow-rate of
Preparation of skeletal muscle actin and HNE exposure
ꢀ
1
0
.2 ml min . The ESI-MS source was set in positive-ion
G-actin was prepared from rabbit skeletal muscles according
1
1
mode, the capillary temperature at 250 °C, spray voltage
to the method of Spudich and Watt with an additional
1
2
4.5 kV, source current 10 µA, and capillary voltage 10.0 V.
Sephadex G-150 gel filtration step. G-actin concentration
was determined by measuring the absorbance at 290 nm,
ꢀ
1
The flow-rate of the nebulizer gas (nitrogen) was 5 L min
.
ꢀ
1
ꢀ1
Peptide sequences were identified using the database search
software turboSEQUEST (Bioworks 3.1, Thermoquest, Milan,
Italy), using a database containing only the protein of interest
and assuming trypsin digestion. The protein sequence of
using an extinction coefficient of 0.63 ml mg cm . Before
use, G-actin solutions were clarified by 60-min centrifugation
at 100 000 g; protein solutions and buffers were then filtered
through 0.20 µm disposable filters and degassed. The exper-
iments were done in buffer containing 2 mM Tris-HCl, pH
˛-actin from skeletal muscle was obtained from the SWISS-
PROT database (primary accession number P68136), and
7
.5, 0.2 mM ATP, 0.2 mM CaCl
2 3
, 2 mM NaN , unless stated
1
3
PEPTIDEMASS was used to calculate theoretical digested
masses based on tryptic digest.
otherwise.
F-actin was prepared by polymerization of G-actin
ꢂ23.6 µM) at 25 °C by the addition of 100 mM KCl C 2 mM
MgCl
2
for 18 h. G- and F-actin (23.6 µM as G-actin) were
Computation methods
treated with increasing molar ratios of HNE (1 : 1, 1 : 10,
: 20) for 120 min at 37 °C, with continuous shaking. Excess
of HNE was removed by exhaustive dialysis against Milli-Q.
G-actin and F-actin samples were lyophilized under vacuum
and stored at ꢀ20 °C until MS analysis.
The structure of monomeric actin in ATP state was retrieved
1
4
1
from the PDB archive (PDB Id: 1NWK). The structure
was completed by adding three missing fragments (Asp1 to
Thr5, His40 to Asp51 and Arg372 to Phe375) and hydrogen
atoms (Arg, Lys, Asp and Glu residues were ionized, while
His residues were preserved neutral by default). More
specifically, the first fragment (Asp1 to Thr5) was manually
Protein analysis by mass spectrometry
ESI-MS and LC-ESI-MS/MS analyses were done on a Thermo
1
5
constructed in ˛-helix using the VEGA fragment builder,
Finnigan LCQ Advantage ion trap mass spectrometer
the second fragment (His40 to Asp51) was added using the
corresponding one of the actin in ADP state (PDB code
(Thermoquest, Milan Italy).
1
6
ESI-MS – Undigested native and HNE-treated G- and
F-actin were analyzed by ESI-MS infusion experiments to
detect changes in protein mass. Lyophilized protein ꢂ250 µgꢃ
1J6Z), and the third fragment (Arg372 to Phe375) was
added using the corresponding one of the rabbit actin in
1
7
complex with kabiramide C (PDB code 1QZ5). The AMP
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954

948
G. Aldini et al.
nucleotide was manually transformed in the ATP molecule.
These preliminary modifications were made using the VEGA
from the corresponding electrostatic potential (keywords,
‘AM1’, ‘PRECISE’, ‘1SCF’, ‘ESP’ and ‘MMOK’).
1
5
18
program. The resulting structure was minimized in a two-
phase procedure: first without any constraints (apart from
The simulations were performed with Namd 2.51, using
the CHARMM force field and Gasteiger’s atomic charges,
CC
CC
and the Ca
ions were always kept fixed. The MD sim-
the fixed Ca ions) until root mean square ꢂRMSꢃ D 1 to
ulation was done with the following characteristics: initial
heating from 0 to 300 K over 3000 iterations (3 ps, i.e. 1 K/10
iterations), 1 ns of equilibrating MD with constant tem-
perature of 300 K by means of Langevin’s algorithm, and
integration of Newton’s equation each 1 fs according to Ver-
let’s algorithm. Minimizations were based on the conjugate
gradients algorithm. The SAS values were calculated with a
adjust the conformation of the inserted fragments avoiding
bad intramolecular interactions, and then with the backbone
fixed until RMS D 0.01 to preserve the experimental folding.
A 1-ns molecular dynamics (MD) simulation was carried
out with backbone fixed to equilibrate the conformations
of the side chains, and the lowest energy structure of
this simulation was used to calculate the solvent accessible
surface (SAS) per residue and to extract the peptides around
the Cys residues. In detail, these fragments were obtained by
˚
solvent probe of radius 1.4 A.
RESULTS AND DISCUSSION
˚
selecting the residues inside a 2-A sphere around a cysteine
for example, for Cys374 the selected fragment corresponds
(
Target proteins were first analyzed by ESI-MS (infusion
experiments) to unequivocally establish the stoichiometry of
HNE adduction and to measure the increase in protein
to the tetra-peptide Arg372-Lys373-Cys374-Phe375). Their
precise atomic charges were calculated with MOPAC6.0
Ł
Figure 1. Positive-ion ESI-MS spectra of (a) untreated and (b) HNE-treated actin at a molar ratio of 1 : 10. In spectrum (b), denotes
multiple charged peaks relative to native actin.
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954

LC-MS/MS evidence for actin modification by HNE 949
mass adduct(s). Figure 1 shows representative ESI-mass
spectra of native (a) and HNE-modified (b) G-actin (1 : 10
molar ratio), which were recorded by signal averaging for
The pattern of adduction and component molecular
weights is clearer in the mass deconvoluted spectra shown
in Fig. 2: the native G- and F-actin show a MW of 41 872
Da, which corresponds to the theoretical isotope-averaged
molecular weight of N-terminal acetylated ˛-actin with one
methylated His residue, as recently determined by FTICR-
MS analysis of skeletal muscle actin. When actin (G or F)
was incubated with HNE (1 : 1), a distinct peak at 42 028 Da
appeared in the deconvoluted spectra, whose intensity sig-
nificantly increased with the protein : HNE molar ratio, con-
comitant to a decrease in the peak for native protein (Fig. 2).
These results clearly indicate that: (i) only a mono-HNE
adduct is formed under these experimental conditions (2 h
1
3
0 min, corresponding to the consumption of approximately
nmoles protein. The mass spectrum of native actin shows
a range of protonation states (a typical feature of ESI for
proteins) with multiple charged peaks ranging from m/z
1
9
8
39 to m/z 1995, corresponding to the addition of 50 to 21
protons. The mass spectrum reported in Fig. 1(b) for HNE-
treated actin shows, besides the multiple charged peaks of
unmodified actin (peaks labeled with an asterisk), a second
series of multiple charged peaks with higher abundance,
which clearly indicates modification of the actin.
Figure 2. Deconvoluted ESI-mass spectra relative to native and HNE-treated G- and F-actin at molar ratios of 1 : 1, 1 : 10, 1 : 20.
ŁMW D 41 872 Da;
°MW D 42 028 Da.
Figure 3. LC-ESI-MS analysis of tryptic peptides from actin. The elution positions of identified tryptic peptides are indicated in the
total ion current and summarized in Table 2.
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954

950
G. Aldini et al.
incubation); (ii) the increase of 156 Da in MW confirms
the insertion of HNE through a Michael adduction (the
theoretical mass increment for the HNE-Michael adduct is
Table 1. SAS values for the
nucleophilic reactive residues of actin
˚
2
SAS %a
Residue
SAS (A )
1
56.22) (iii) both the monomeric and the polymerized forms
of actin show similar reactivity and adduction mechanism
towards HNE.
Cys free
Cys10
Cys217
Cys257
Cys285
Cys374b
256.4
0
0.7
0
0.4
106.2
–
0.0
0.3
0.0
0.2
41.5
As regards the site of HNE adduction, two parameters
mainly regulate the adduction chemistry: the nucleophilic
reactivity of amino acids and their accessible surface. Actin
contains 32 nucleophilic residues (5 Cys, 8 His and 19 Lys)
and several of these, as calculated by computational analysis
His free
His40b
His87
320.1
122.2
85.6
38.3
4.5
–
(
Table 1), have an accessible surface that is at least 30% of
38.2
26.7
12.0
1.4
0.0
36.7
9.6
the free amino acid, making them potential sites for HNE
adduction (namely, Cys374, His40, His173, Lys113, Lys191,
Lys215, Lys238, and Lys326).
To characterize the structural modification of actin
by HNE, the native and the HNE-treated actin were
digested with trypsin, then analyzed by LC-ESI-MS/MS
His88
His101
His161
His173b
His275
His371
0
117.4
30.7
45.5
after NaBH
stabilization.
4
reduction, an established procedure for adduct
14.2
2
0,21
Peptide mass mapping (Fig. 3) of native
Lys free
Lys18
Lys50
Lys61
Lys68
341.6
0.5
–
0.1
actin provided identification of the peptides accounting
for approximately 78% of the protein sequence (Table 2).
This sequence analysis confirms the N-terminal acetylation
in Asp1 and His73 as a methylated residue (peak 7). Cys
residues are present as disulfides with the reducing agent
ˇ-mercaptoethanol.
76.9
60.8
47.5
53.6
123.6
83.6
108.9
0.2
111.5
106.8
82.2
101.3
64.8
132.1
96.6
37.7
71.4
70.1
22.5
17.8
13.9
15.7
36.2
24.5
31.9
0.1
32.6
31.3
24.1
29.7
19.0
38.7
28.3
11.0
20.9
20.6
Lys84
Lys113b
Lys118
Lys191b
Lys213
Lys215b
Lys238b
Lys284
Lys291
Lys315
Lys326b
Lys328
Lys336
Lys359
Lys373
Since the order of reactivity of the nucleophilic amino
2
2
acids towards HNE is Cys × His > Lys and the most
accessible residue is Cys374, as a first step we searched
for HNE-modified Cys-containing peptides. This was done
by neutral loss scan analysis of the HNE-adducted tryptic
mixture by scanning for the neutral loss of m/z 158, 79 and
5
3, corresponding to the loss of a neutral reduced HNE
moiety from singly, doubly and triply charged precursors.
This procedure was of limited analytical utility in the case of
2
3
HNE-modified His peptides, but proved particularly suited
for HNE-modified Cys peptides, since their typical MS/MS
fragmentation pattern is characterized by a retro-Michael
1
0,24
reaction, as for GSH-HNE adducts.
Figure 4 shows the map for the selected neutral loss
mass of 158 Da, relative to the tryptic peptides of native
a
Calculated in relation to the free
amino acid.
Denotes residues with SAS values
(
a) and HNE-treated actin (b). These reconstructed spectra
b
consist of all the parent ions that gave products within
the neutral loss of 158 Da (š1 tolerance). Both spectra
are characterized by several parent ions and differ only
in the presence of the precursor ion at m/z 427 in digested,
modified actin, corresponding to the [M C H]C of the HNE-
adducted Cys374-Phe375 dipeptide. The MS/MS spectrum
½
30% of the corresponding free amino
acid.
peak at 32.6 min relative to the Cys374-Phe375 dipep-
tide is present only in digested native actin, not in
HNE-treated actin (Fig. 6). Conversely, the SIC of the
ion at m/z 427 shows four distinct peaks with reten-
tion times (RT) between 60 and 62 min, which are
present only in the digested HNE-treated actin, and
are attributed to the four diasteroisomers of the HNE-
Cys374-Phe375 adduct. Formation of these isomers is
explained by considering that nucleophilic addition of the
thiol group of Cys374 to C3 of a racemic mixture of
C
of the [M C H] ion at m/z 427 (Fig. 5) contains, besides the
C
[
M C H ꢀ H
2
O] ion at m/z 409, the typical retro-Michael ion
C
fragment at m/z 269 [M C H ꢀ 158] and the corresponding
3
fragment ion due to loss of NH (m/z 252). The presence of
the HNE-adducted b ion at m/z 262 unequivocally indicates
that the HNE-Michael adduct is associated with Cys374.
Confirmation is obtained by recording the selected
ion current (SIC) for the [M C H]C ion of the unmod-
ified Cys374-Phe375 dipeptide at m/z 345 (as disul-
fide with the reducing agent ˇ-mercaptoethanol): the
HNE, followed by NaBH reduction, generates a Michael
adduct diol derivative (Fig. 6), bearing two chiral carbons
beside the two fixed centers (R)Cys and (S)Phe (four
4
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954

LC-MS/MS evidence for actin modification by HNE 951
Table 2. Peptides identified by ESI-LC-MS/MS from native actin
Mass
obs.)
Mass
(calc.)
Sequence
position
Peak
no.
a
(
Sequence
c
b
1
983.9
1984.0
976.0
945.1
1–18
19–28
29–37
D EDETTALVC DNGSGLVK
AGFAGDDAPR
AVFPSIVGR
1
2
3
9
9
75.4
45.8
1
1
171.0
197.8
1171.4
1198.2
643.8
1975.2
1515.7
1956.2
2726.0
643.7
40–50
51–61
63–68
69–84
HQGVMVGMGQK
DSYVGDEAQSK
GILTLK
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
6
43.4
d
1
1
1
2
975.7
515.4
956.2
726.3
YPIEH GIITNWDDMEK
85–95
IWHHTFYNELR
VAPEEHPTLLTEAPLNPK
TTGIVLDSGDGVTHNVPIYEGYALPH
LDLAGR
DLTDYLMK
ILTER
96–113
148–173
178–183
184–191
192–196
197–206
216–238
239–254
257–284
285–290
292–312
316–326
329–335
360–372
374–375
6
9
6
43.5
97.6
30.5
998.1
630.7
1
2
1
3
129.9
555.1
790.8
208.1
1130.2
2555.7
1790.9
3208.5
809.8
2246.5
1161.4
794.9
GYSFVTTAER
b
LC YVALDFENEMATAASSSSLEK
SYELPDGQVITIGNER
b
C PETLFQPSFIGMESAGIHETTYNSIMK
b
8
09.3
C DIDIR
2
1
246.0
161.0
DLYANNVMSGGTTMYPGIADR
EITALAPSTMK
IIAPPER
7
94.5
500.4
44.0
1
1500.6
344.3
QEYDEAGPSIVHR
b
3
C F
a
b
c
Average MW.
Disulfide with the reducing agent ˇ-mercaptoethanol.
N-acetyl derivative.
d
Methyl derivative; sequence coverage 78%.
Figure 4. Reconstructed neutral loss spectra showing all the parent ions that produced products within the neutral loss of 158 Da
(
š1 tolerance) relative to the tryptic peptides from (a) untreated and (b) HNE-treated actin (1 : 20 molar ratio). The spectra differ only
C
in the presence of the precursor ion at m/z 427 (arrow) in digested modified actin, corresponding to the [M C H] of the
HNE-adducted Cys374-Phe375 dipeptide.
(iodoacetamido)-tetramethylrhodamine,26 or fluorescent
diasteroisomers, namely: (3R,4R)-(R)Cys-(S)Phe; (3S,4S)-
R)Cys-(S)Phe; (3R,4S)-(R)Cys-(S)Phe; (3S,4R)-(R)Cys-(S)
Phe).
The data obtained agree well with those in the litera-
(
dyes,27,28 and it is specifically involved in the S-nitrosylation29
and S-glutathionylation reaction.30
The specific carbonylation at Cys374, as well as the struc-
tural elements that favor the adduction and make the other
nucleophilic sites less reactive, can be explained by con-
sidering both steric and electronic factors. Table 3 indicates
that Cys374 is the unique cysteine that simultaneously pos-
sesses both a significant accessible surface and a noteworthy
ture reporting that the sulfhydryl group of Cys374 in actin
is highly accessible and reactive and that Cys374 is the
most preferred target of oxidation. In particular, it has been
found that Cys374 is the site of the oxidative modifica-
2
5
tion induced by tert-butyl hydroperoxide, it is chemically
2
6
modified by different agent such as N-ethylmaleimide,
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954

952
G. Aldini et al.
acidity of S-H bond, as indicated by both the polarity of S-H
bond and the stability of corresponding sulphur anion. The
atomic charges of the thiol group range from ꢀ0.02 (S) and
Table 3. SAS and polarity of the S–H bond of
actin Cys residues
˚
2
Residue
SAS (A )
q H
q S
0
.04 (H) in free cysteine to ꢀ0.14 (S) and 0.10 (H) in Cys374
(
Table 3) and this significant acidity can be explained con-
Cys Free
Cys10
Cys217
Cys257
Cys285
Cys374
263.2
0.0
0.7
0.0
0.4
C0.04
C0.05
C0.03
C0.10
C0.02
C0.10
ꢀ0.02
ꢀ0.07
ꢀ0.01
ꢀ0.18
ꢀ0.05
ꢀ0.14
sidering that side chain of Cys374 is closely surrounded by
polar residues (such as Arg372). Arg372 and Phe375 raise
the acidity of the cysteine thiol group by stabilizing the cor-
responding sulfur anion, arginine by means of a salt bridge,
and phenylalanine by means of an anion-ꢄ interaction that
106.2
3
1
involves the sulfur lone pairs. The thiol group of Cys257
also shows significant polarity, but its accessibility is virtually
nil.
CONCLUSIONS
Carbonylation of actin has been detected in several patho-
logical situations by conventional immunochemical and
immunohistochemical techniques, which, although furnish-
ing qualitative and quantitative evidence of the process,
do not provide information on the mechanism and site.
MS-based techniques play a key role in the study of
the kinetics and mechanisms of reactive aldehyde-protein
3
2,33
interactions,
providing information on the molecular
mass of modified protein and helping estimate the stoi-
chiometry of the reaction (complete structure assignment of
protein-HNE adducts is generally obtained by trypsin diges-
tion followed by peptide mass mapping, mainly by HPLC-
ESI-MS/MS techniques). The emerging potential of these
techniques notwithstanding, no studies have yet investigated
whether actin can be covalently modified by lipid-derived
aldehydes, such as HNE.
Figure 5. CID mass spectra of the ion m/z 427 originating from
the tryptic digestion of HNE-adducted actin. Modified fragment
ions are labeled with an asterisk and according to the
nomenclature of Biemann.36 The peptide sequence is
represented in the one-letter code nomenclature.
The present findings provide the first evidence that both
monomeric and polymeric actin are extremely sensitive to
HNE-Michael adduction because of the high reactivity of
Cys374, suggesting that this mechanism may be responsible
for actin carbonylation detected in vivo. In addition, the
Figure 6. LC-ESI-MS analysis (SICs) of native (a) and HNE-modified (b) actin. SICs are relative to the ions at m/z 345 (upper panels)
and m/z 427 (lower panels).
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954

LC-MS/MS evidence for actin modification by HNE 953
computational analysis suggests it may be possible to
rationalize (or predict) the carbonylation adducts on the
basis of both structural and electronic properties of possible
reactive residues. It is important to stress that with 2 h
incubation (longer incubation times were not explored to
5. Keshavarzian A, Banan A, Farhadi A, Komanduri S, Mutlu E,
Zhang Y, Fields JZ. Increases in free radicals and cytoskeletal
protein oxidation and nitration in the colon of patients with
inflammatory bowel disease. Gut 2003; 52: 720.
6. Powell SA, Gurzenda EM, Wahezi SE. Actin is oxidized during
myocardial ischemia. Free Radic. Biol. Med. 2001; 30: 1171.
avoid excessive actin aggregation, favored by incubation
at 37 °C in the absence of reducing agents) the protein was
7. Canton M, Neverova I, Menabo R, Van Eyk J, Di Lisa F. Evidence
of myofibrillar protein oxidation induced by postischemic
reperfusion in isolated rat hearts. Am. J. Physiol. Heart Circ.
Physiol. 2004; 286: H870.
quantitatively modified by HNE molecule (molar ratio 1 : 20),
suggesting carbonylation is extremely rapid. However,
we cannot exclude that in other conditions (prolonged
incubation times or in biological systems) other reactive
and accessible nucleophilic residues on actin (such as His40,
His173, Lys113, Lys191, Lys215, Lys238, and Lys326) may
react with additional HNE molecules.
8
. Barreiro E, Gea J, Di Falco M, Kriazhev L, James S, Hussain SN.
Protein carbonyl formation in the diaphragm. Am. J. Respir. Cell
Mol. Biol. 2005; 32: 9.
9. Stadtman ER, Levine RL. Free radical-mediated oxidation of free
amino acids and amino acid residues in proteins. Amino Acids
2003; 25: 207.
1
0. Rees MS, van Kuijk FJGM, Siakotos AN, Mundy BP. Improved
synthesis of various isotope labelled 4-hydroxyalkenals and
peroxidation intermediates. Synth. Commun. 1995; 25: 3225.
1. Spudich JA, Watt S. The regulation of rabbit skeletal muscle
contraction. I. Biochemical studies of the interaction of the
tropomyosin-troponin complex with actin and the proteolytic
fragments of myosin. J. Biol. Chem. 1971; 246: 4866.
2. MacLean-Fletcher S, Pollard TD. Identification of a factor in
conventional muscle actin preparations which inhibits actin
filament self-association. Biochem. Biophys. Res. Commun. 1980;
HNE-modified proteins in isolated rat hearts during
ischemia/reperfusion were detected and quantitated using
an antibody to HNE-Cys/His/Lys and densitometry of
1
3
4
Western blots. Most of the HNE adduct signal was in the
cell membrane, but HNE immunofluorescence also showed
longitudinal striations with possible co-localization with the
myofibrils, suggesting modification of contractile proteins.
In addition, using ventricular myocytes isolated from adult
male rats, it was recently demonstrated that HNE inhibits
1
96: 18.
1
3. Wilkins MR, Lindskog I, Gasteiger E, Bairoch A, Sanchez JC,
Hochstrasser DF, Appel RD. Detailed peptide characterisation
using PEPTIDEMASS – a World-Wide Web accessible tool.
Electrophoresis 1997; 18: 403.
14. Graceffa P, Dominguez R. Crystal structure of monomeric actin
in the ATP state. Structural basis of nucleotide-dependent actin
dynamics. J. Biol. Chem. 2003; 278: 34 172.
5. Pedretti A, Villa L, Vistoli G. VEGA: a versatile program to
convert, handle and visualize molecular structure on Windows-
based PCs. J. Mol. Graph. Model. 2002; 21: 47.
3
5
cardiac myocyte contraction.
Because this evidence suggests that the myofilaments
may also be an important target for oxidant stress dur-
ing ischemia-reperfusion and that myofilament dysfunction
may be related to the formation of HNE adducts with key
contractile proteins, studies are in progress in our laborato-
ries to evaluate the effects of HNE carbonylation on actin
function in vitro and in vivo, and to demonstrate actin-HNE
adduct formation by mass spectrometric techniques in cell
lines, isolated tissues/organs exposed to free radicals, or in
the experimental animal under conditions that mimic patho-
logical situations where HNE is massively involved, e.g.
ischemia/reperfusion damage as a model of cardiovascular
diseases. It must, in any event, be borne in mind that actin car-
bonylation in vivo could also result from mechanisms other
than HNE-Michael addition. Our investigations in this field
have really just begun.
1
16. Otterbein LR, Graceffa P, Dominguez R. The crystal structure of
uncomplexed actin in the ADP state. Science 2001; 293: 708.
1
7. Klenchin VA, Allingham JS, King R, Tanaka J, Marriott G,
Rayment I. Trisoxazole macrolide toxins mimic the binding of
actin-capping proteins to actin. Nat. Struct. Biol. 2003; 10: 1058.
18. Kal e` L, Skeel R, Bhandarkar M, Brunner R, Gursoy A,
Krawetz N, Phillips J, Shinozaki A, Varadarajan K, Schulten K.
NAMD2: Greater scalability for parallel molecular dynamics. J.
Comput. Phys. 1999; 151: 283.
9. Bergen HR, Ajtai K, Burghardt TP, Nepomuceno AI, Muddi-
man DC. 3rd. Mass spectral determination of skeletal/cardiac
actin isoform ratios in cardiac muscle. Rapid Commun. Mass
Spectrom. 2003; 17: 1467.
0. Crabb JW, O’Neil J, Miyagi M, West K, Hoff HF. Hydroxynone-
nal inactivates cathepsin B by forming Michael adducts with
active site residues. Protein Sci. 2002; 11: 831.
1
Acknowledgements
This work was supported by MURST grants (Cofinanziamento
Programma Nazionale 2004-2005).
2
2
1. Fenaille F, Guy PA, Tabet JC. Study of protein modification by
REFERENCES
4-hydroxy-2-nonenal and other short-chain aldehydes analyzed
1
. Dalle-Donne I, Rossi R, Milzani A, Di Simplicio P, Colombo R.
The actin cytoskeleton response to oxidants: from small heat
shock protein phosphorylation to changes in the redox state of
actin itself. Free Radic. Biol. Med. 2001; 31: 1624.
by electrospray ionization tandem mass spectrometry. J. Am.
Soc. Mass Spectrom. 2003; 14: 215.
22. Doorn JA, Petersen DR. Covalent adduction of nucleophilic
amino acids by 4-hydroxynonenal and 4-oxononenal. Chem.
Biol. Interact. 2003; 143–144: 93.
23. Bolgar MS, Gaskell SJ. Determination of the sites of 4-hydroxy-
2-nonenal adduction to protein by electrospray tandem mass
spectrometry. Anal. Chem. 1996; 68: 2325.
24. Aldini G, Granata P, Carini M. Detoxification of cytotoxic
˛,ˇ-unsaturated aldehydes by carnosine: characterization of
conjugated adducts by electrospray ionization tandem mass
spectrometry and detection by liquid chromatography/mass
spectrometry in rat skeletal muscle. J. Mass Spectrom. 2002; 37:
1219.
2
. Banan A, Zhang Y, Losurdo J, Keshavarzian A. Carbonylation
and disassembly of the F-actin cytoskeleton in oxidant-induced
barrier dysfunction and its prevention by epidermal growth
factor and transforming growth factor alpha in a human colonic
cell line. Gut 2000; 46: 830.
3
. Banan A, Fitzpatrick L, Zhang Y, Keshavarzian A. OPC-
compounds prevent oxidant-induced carbonylation and
depolymerization of the F-actin cytoskeleton and intestinal
barrier hyperpermeability. Free Radic. Biol. Med. 2001; 30: 287.
. Aksenov MY, Aksenova MV, Butterfield DA, Geddes JW,
Markesbery WR. Protein oxidation in the brain in Alzheimer’s
disease. Neurosci. 2001; 103: 373.
4
25. Dalle-Donne I, Milzani A, Colombo R. The t-butyl hydropero-
xide-induced oxidation of actin Cys-374 is coupled with
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954

954
G. Aldini et al.
structural changes in distant regions of the protein. Biochemistry
999; 38: 12 471.
6. Tait JF, Frieden C. Chemical modification of actin. Acceleration
of polymerization and reduction of network formation by reac-
tion with N-ethylmaleimide, (iodoacetamido) tetramethylrho-
damine, or 7-chloro-4-nitro-2,1,3-benzoxadiazole. Biochemistry
31. Garau C, Frontera A, Quinonero D, Ballester P, Costa A,
Deya PM. Cation-pi versus anion-pi interactions: energetic,
charge transfer, and aromatic aspects. J. Phys. Chem. A 2004;
108: 9423.
32. Carini M, Aldini G, Maffei Facino R. Mass spectrometry for
detection of 4-hydroxy-trans-2-nonenal (HNE) adducts with
peptides and proteins. Mass Spectrom. Rev. 2004; 23: 281.
33. Dalle-Donne I, Scaloni A, Giustarini D, Cavarra E, Tell G,
Lungarella G, Colombo R, Rossi R, Milzani A. Proteins as
biomarkers of oxidative/nitrosative stress in diseases: The
contribution of redox proteomics. Mass Spectrom. Rev. 2005;
24: 55.
34. Eaton P, Li JM, Hearse DJ, Shattock MJ. Formation of 4-hydroxy-
2-nonenal-modified proteins in ischemic rat heart. Am. J. Physiol.
1999; 276: H935.
35. Aberle NS 2nd Sr., Picklo MJ, Amarnath V, Ren J. Inhibition
of cardiac myocyte contraction by 4-hydroxy-trans-2-nonenal.
Cardiovasc. Toxicol. 2004; 4: 21.
1
2
1982; 21: 6046.
2
7. Yasunaga T, Wakabayashi T. Relocation of Cys374 of actin
induced by labeling with fluorescent dyes. J. Biochem. (Tokyo)
2001; 129: 201.
2
8. Crosbie RH, Miller C, Cheung P, Goodnight T, Muhlrad A,
Reisler E. Structural connectivity in actin: effect of C-terminal
modifications on the properties of actin. Biophys. J. 1994; 67:
1957.
2
9. Dalle-Donne I, Milzani A, Giustarini D, Di Simplicio P,
Colombo R, Rossi R. S-NO-Actin: S-nitrosylation kinetics and
effect on vascular smooth muscle. J. Muscle Res. Cell Motil. 2000;
21: 171.
3
0. Dalle-Donne I, Giustarini D, Rossi R, Colombo R, Milzani A.
Reversible S-glutathionylation of Cys(374) regulates actin
filament formation by inducing structural changes in the actin
molecule. Free Radic. Biol. Med. 2003; 34: 23.
36. Biemann K. Contributions of mass spectrometry to peptide and
protein structure. Biomed. Environ. Mass Spectrom. 1988; 16: 99.
Copyright  2005 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2005; 40: 946–954