Methionine residue acts as a prooxidant in the ?oH-induced oxidation of enkephalins

Article abstract of DOI:10.1021/jp307043q

Enkephalins are bioactive pentapeptides (Tyr-Gly-Gly-Phe-Leu (Leu-enk) and Tyr-Gly-Gly-Phe-Met (Met-enk)) produced while an organism is under mental and/or physical stress. In the course of their biological action they are exposed to reactive oxygen and nitrogen species. We have reinvestigated the reactions of ^?OH radicals toward these peptides in order to elucidate the oxidation mechanisms and the final products. Nanosecond pulse radiolysis was used to obtain the spectra of the reaction intermediates and their kinetics. Additional insight into details of the oxidation mechanism was gained by identification of main final products by means of UV-vis spectrophotometry, HPLC coupled with fluorescence spectroscopy, and mass spectrometry. The key processes are different in both peptides. In Leu-enk, the first step is an ^?OH radical addition to the aromatic rings of Tyr and Phe residues that leads to hydroxylated residues, dihydroxyphenylalanine (DOPA) from Tyr and tyrosine isomers from Phe, respectively. In Met-enk, these processes are less important, an additional target being the sulfur atom of the methionine residue. Depending on pH either an OH-adduct (hydroxysulfuranyl radical) or a sulfur radical cation undergo intramolecular electron transfer with Tyr residue resulting in a repair of Met and oxidation of Tyr to tyrosyl radicals and a final formation of dityrosine. At low pH, the OH-adducts to Tyr residue are precursors of tyrosyl radicals and dityrosine. Thus, the final products coming from oxidation of the Tyr residue depend strongly on the neighboring residues and the pH.

Full text of DOI:10.1021/jp307043q

Article
pubs.acs.org/JPCB
•
Methionine Residue Acts as a Prooxidant in the OH-Induced
Oxidation of Enkephalins
Olivier Mozziconacci,†,‡,§ Jacek Mirkowski,‡ Filippo Rusconi,∥,† Gabriel Kciuk,‡ Pawel B. Wisniowski,‡
Krzysztof Bobrowski,*,‡ and Chantal Houee-Levin*,†
́
†Laboratory of Physical Chemistry and CNRS Bldg 350, Centre Universitaire, F-91405 Orsay, F-91405 Orsay, France
‡Institute of Nuclear Chemistry and Technology, Dorodna, 16, 03-195 Warsaw, Poland
§Department of Pharmaceutical Chemistry, University of Kansas, Lawrence, Kansas 66047, United States
∥Museu
́
m National d’Histoire Naturelle, CNRS, UMR7196 - INSERM, U565 - MNHN USM0503, 57 rue Cuvier, F-75231 Paris
Cedex-05, France
S
* Supporting Information
ABSTRACT: Enkephalins are bioactive pentapeptides (Tyr-
Gly-Gly-Phe-Leu (Leu-enk) and Tyr-Gly-Gly-Phe-Met (Met-
enk)) produced while an organism is under mental and/or
physical stress. In the course of their biological action they are
exposed to reactive oxygen and nitrogen species. We have
reinvestigated the reactions of •OH radicals toward these
peptides in order to elucidate the oxidation mechanisms and
the final products. Nanosecond pulse radiolysis was used to obtain the spectra of the reaction intermediates and their kinetics.
Additional insight into details of the oxidation mechanism was gained by identification of main final products by means of UV−
vis spectrophotometry, HPLC coupled with fluorescence spectroscopy, and mass spectrometry. The key processes are different in
•
both peptides. In Leu-enk, the first step is an OH radical addition to the aromatic rings of Tyr and Phe residues that leads to
hydroxylated residues, dihydroxyphenylalanine (DOPA) from Tyr and tyrosine isomers from Phe, respectively. In Met-enk, these
processes are less important, an additional target being the sulfur atom of the methionine residue. Depending on pH either an
OH-adduct (hydroxysulfuranyl radical) or a sulfur radical cation undergo intramolecular electron transfer with Tyr residue
resulting in a repair of Met and oxidation of Tyr to tyrosyl radicals and a final formation of dityrosine. At low pH, the OH-
adducts to Tyr residue are precursors of tyrosyl radicals and dityrosine. Thus, the final products coming from oxidation of the
Tyr residue depend strongly on the neighboring residues and the pH.
activity.5 They are distributed as a mixture of two
pentapeptides, Tyr-Gly-Gly-Phe-Leu (Leu-enk) and Tyr-Gly-
Gly-Phe-Met (Met-enk), in the central nervous system, specific
INTRODUCTION
■
There is substantial evidence that oxidative stress is a causative
factor in the pathogenesis of several major diseases including
both neurodegenerative Parkinson’s1 and Alzheimer’s diseases.2
areas of the brain, and the spinal cord, as well as in nerve plexi
of intestines.6 These pentapeptides are opiates and have the
Protein modifications are known to play major roles in these
same receptors as morphine and thebaine. Their distribution
diseases. The consequences of protein oxidation are not well-
and the distribution of the opioid receptors ensure a highly
known because they depend on the possibility either to destroy
the damaged ones or to repair the oxidized residues by
effective pain management system. They play the role of
neurotransmitters and neurohormones in regions of the brain
reduction. Tyrosine is one of the main targets of reactive
oxygen and nitrogen species (RONS, i.e., OH, O2•−, NO•,
•
and spine associated with diffuse pain pathways and inhibition
of pain signals.7 Moreover, enkephalins are involved in all kinds
of stresses,8 in red blood cells,9 as well as in liver10 and brain.11
They play diverse roles in biological systems such as controlling
respiration, aiding human immune response,12 producing
dependence, and tolerance to stroke.13 One of their important
roles is the control of pain by binding to nociceptive
receptors.14 In the processes occurring during development of
illnesses, enkephalins are subjected to oxidation by reactive
hydrogen peroxide, peroxynitrite, etc.). When its oxidation
leads to polymerization, toxic protein aggregates are formed.3
On the other hand, methionine, another main target of
oxidative stress, is usually considered as a protective residue
against RONS since one major final product, methionine
sulfoxide, can be “repaired“ by reduction to the initial residue
by methionine sulfoxide reductases.
Enkephalins are usually produced while the organism is
under mental and/or physical stress. These bioactive peptides
originate from the larger protein, β-lipotropin,4 and are
exhibiting, as a common feature, the presence of a tyrosine
residue at the N-terminus, which is essential for their biological
Received: July 16, 2012
Revised: September 19, 2012
Published: September 22, 2012
© 2012 American Chemical Society
12460
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
oxygen species (ROS), i.e., hydroxyl radicals (•OH), superoxide
radical anions (O2•−), and molecular products like H2O2.
The degradations induced by •OH radicals in Met-enk and in
Leu-enk are relevant to the disorders in inflammatory
processes. Because of the importance of these reactions in
vivo, several studies were devoted to their mechanism. In vitro
experiments showed that Met-enk and Leu-enk, in the presence
of hydrogen peroxide, can be oxidized by horseradish
peroxidase to the enkephalin dimers with a dityrosyl bond
between the N-terminal tyrosine residues.15 The same product
was found when Met-enk or its precursor, pre-Met-enk, were
exposed to activated oxygen species produced by stimulated
polymorphonuclear leucocytes.16 Enkephalins scavenge oxygen
free radicals generated by the Fenton reaction. The N-terminal
Aldrich and nitrous oxide (N2O) 99% from the Air Liquide
AlphaGaz.
All solutions were made with water triply distilled from
permanganate or delivered by Elga Maxima system. The pH
was adjusted by the addition of perchloric acid (HClO4).
Typical concentration of solutions (3 mL) for γ irradiation was
0.5 mM (enkephalins) at natural pH (6.5) and at pH 1.5, unless
otherwise specified. Prior to irradiation, the samples were
purged gently with N2O for 20 min. Typical concentration of
solutions in pulse radiolysis experiments was 0.1 mM of
enkephalins at natural pH (6.5) and 1.5 unless otherwise
specified. They were purged with N2O for 30 min per 50 mL
volume before experiments.
Steady-State γ-Radiolysis. γ irradiation was carried out
using the 60Co γ-rays source Issledovatel (USSR) in the
Institute of Nuclear Chemistry and Technology (Warsaw,
Poland) and the panoramic 60Co γ-source IL60PL Cis-Bio
International (France) in the University Paris-Sud (Orsay). The
dose rates were equal to 1.05 Gy min−1 and 7.5 Gy min−1,
respectively as determined by Fricke dosimetry.22 All
irradiations were performed at room temperature. The doses
applied ranged between 80 and 500 Gy.
•
tyrosine residues were found to be the main target of OH
radicals, which convert further into DOPA derivatives.17 In the
presence of an excess of thiols the oxidation of enkephalins
gives rise to S-cysteinyldopa-enkephalins.18 Nitrogen radicals
(peroxynitrite and peroxidase-generated reactive nitrogen
species) also react with the N-terminal Tyr residue, producing
nitrotyrosine and enkephalin dimers linked by dityrosine.19
We have already investigated the oxidation mechanism of
Met-enk induced by a one-electron oxidant species, the azide
Pulse Radiolysis. Pulse radiolysis experiments were
performed with the INCT LAE 10 MeV linear electron
accelerator with typical pulse length of 8 ns. A detailed
description of the experimental setup has been given elsewhere
along with the basic details of the equipment and the data
collection system.23 Radiolytic yields are given in the
Supporting Information (SI) with units as μmol J−1. Absorbed
doses per pulse were on the order of 5.5 Gy (1 Gy = 1 J kg−1).
Dosimetry was based on N2O-saturated solutions containing
10−2 M KSCN, taking a radiation chemical yield of G = 0.635
μmol J−1 and a molar absorption coefficient of 7580 M−1 cm−1
•
radicals (N3 ) in N2O and O2-saturated solutions. The N-
terminal Tyr residue was found to be the main target for
oxidation. Moreover, it was demonstrated that, when O2•− was
present, it added to tyrosyl radical, giving a hydroperoxide
which further cyclized by reaction with the terminal amine.20 In
two recent papers the mechanism of oxidation of enkephalins
by peroxidase in the presence of superoxide was investigated.21
The authors proposed a mechanism in which the first step of
oxidation was the formation of tyrosyl radicals followed by
addition of superoxide that led to tyrosine hydroperoxide.21a
However, in Met-enk, an intramolecular two-electron process
was invoked: the reaction of the tyrosine hydroperoxide with
methionine leading to methionine sulfoxide and a deoxy-
genated derivative of tyrosine.
In this study we reinvestigated the one-electron oxidation of
both Leu-enk and Met-enk by •OH radicals produced by pulse
and gamma radiolysis at two pH values (6.5 and 1), to unveil
the mechanisms of degradation of the tyrosine residue in both
peptides. Whereas at natural pH (ca. 6.5), one gets
hydroxylated tyrosine (DOPA) or phenylalanine (o- or p-
tyrosine) in Leu-enk, a peptide dimer linked by dityrosine was
observed in Met-enk where the major target is the methionine
residue. This latter observation has been rationalized by an
intramolecular one-electron transfer from the tyrosine residue
to the methionine radical cation. Methionine sulfoxide, which is
most commonly the final product of methionine oxidation, can
be repaired in vivo by methionine sulfoxide reductases, which
means that oxidation of methionine is not deleterious.
However, the conversion of a methionine radical cation to a
tyrosyl radical leads to a nonrepairable damage. Thus, the final
products coming from tyrosine oxidation and possibly useful as
biomarkers of oxidative stress are numerous. They do not only
depend on the oxidant and the presence of oxygen but also on
the reaction pathways.
at 472 nm for the (SCN)2 radical.24 Experiments were
•−
performed with a continuous flow of sample solutions at room
temperature (∼23 °C).
Radiolysis of Water. γ or pulse irradiation of water leads to
the formation of the primary reactive species shown in reaction
1. In N2O-saturated solutions ([N2O]sat ≈ 2 × 10−2 M),
hydrated electrons, eaq−, are converted into •OH radicals
according to reaction 2 (k2 = 9.1 × 109 M−1 s−1).25 Reaction 2
nearly doubles the amount of •OH radicals available for
reactions with substrates.
•
•
H2O → OH, eaq−, H
(1)
−
•
eaq + N2O → O•− + N2(+H2O) → OH + OH− + N2
(2)
Thus, the radiation-chemical yields (G values) of these
primary species in aqueous N2O-saturated solution at neutral
pH are G(•H) = 0.06 μmol J−1 and G(•OH) = 0.56 μmol J−1
(reactions 1 and 2). At pH <4 the diffusion-controlled reaction
of eaq− with protons becomes important (reaction 3, k3 = 2.3 ×
1010 M−1 s−1)26 resulting in a pH-dependent lowered yield of
•OH radicals and a corresponding increased yield of •H atoms.
−
•
eaq + H+ → H
(3)
EXPERIMENTAL AND THEORETICAL METHODS
■
Materials. Leu-enk (Tyr-Gly-Gly-Phe-Leu) and Met-enk
(Tyr-Gly-Gly-Phe-Met) pentapeptides were purchased from
Bachem (Switzerland). The other chemicals were obtained as
follows: perchloric acid (HClO4) was purchased from Sigma-
Therefore, in N2O-saturated aqueous solutions at pH 1.5,
taking into account the known rate constants for reactions 2
and 3, one can calculate the radiation-chemical yields for G(•H)
= 0.34 μmol J−1 and G(•OH) = 0.28 μmol J−1.22
12461
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
Spectral Analysis and Resolution of Time-Resolved
Optical Spectra. In any time window, following the electron
pulse, the absorbance of the solution is related to the radiation-
chemical yield and the molar absorption coefficients by the
formula i
reference solutions. An aliquot of 1 mL was taken to measure
the absorption spectrum.
High Performance Liquid Chromatography (HPLC).
The HPLC analyses were carried out with a Shimadzu
chromatograph consisting of a pump, a 20 μL manual injector,
a polymer PLRP-S 100-μm column 4.6 × 250 mm, packed with
5-μm particles. The detectors were either a photodiode array
detector or a spectrofluorimetric at several settings. The first
setting with the excitation wavelength of 240 nm and with the
emission wavelength of 285 nm was used to detect mainly a
phenylalanine residue and also a tyrosine residue. The second
setting with the excitation wavelength of 284 nm and with the
emission wavelength of 305 nm was used to detect mainly a
tyrosine residue. The third setting with the excitation
wavelength of 284 nm and with the emission wavelength of
425 nm was used to detect dityrosine. The mobile phase was
delivered with a flow rate of 0.9 mL min−1. It consisted of two
solvents, the first with 95% of water, 5% of acetonitrile and
0.01% of trifluoroacetic acid (TFA) and the second with 80% of
acetonitrile, 20% of water, 0.01% trifluoroacetic acid (TFA).
TFA was used to fix the pH along the separation analysis and to
increase the sensitivity of the analysis (shape peaks).
•−
ε472nmG(SCN)2
Gε(λj) = ΔA(λj)
ΔA472nm
(i)
which are how the spectra are displayed in Figures 1 and 5.
Equation i is a convenient way to normalize the absorbance
ΔA(λj) to the radiation dose, the inverse of which is
proportional to ε472G((SCN)2•−)/ΔA472 from thiocyanate
dosimetry, where ε472 is the molar absorption coefficient of
•−
(SCN)2 at 472 nm (7580 M−1 cm−1), G((SCN)2•−) is the
•−
radiation chemical yield of the (SCN)2 radicals (0.635 μmol
J−1), and ΔA472 represents the observed absorbance change at
472 nm in the thiocyanate dosimeter. The Gε(λj) are in turn
related to the underlying transients via Beer’s Law since ΔA =
log(I0/I) is the absorbance of the sample.
A multiregression analysis was done on the optical transient
spectra monitored at various time delays following the electron
pulse. They were resolved into specific components (represent-
ing individual transients) by linear regression according to
equation ii
All chromatography data acquisition and processing were
performed using the program supplied with the Shimadzu
package system.
n
Mass Spectrometry. All mass spectrometric experiments
were performed in the positive-ion mode using a hybrid
quadrupole time-of-flight mass spectrometer equipped with an
electro-spray ion source (ESI MS, Q-Star Pulsar i, MDS Sciex-
Applied Biosystems). Data acquisition and storage were
performed using the Analyst QS software package shipped
with the mass spectrometer. Sequence editing and mass
calculations were typically performed using either GNU
polyxmass28 or massXpert.29 MS experiments were performed
as follows: injections in the flow of an HPLC system were
performed at 30 μL/min (carrier solvent was acetonitrile/
water/formic acid, 49:50:1); the typical peptide concentration
was 1.15 × 10−2 μg/μL; the ion spray needle was set to 5200 V;
the declustering potential was set to 60 V; the m/z scan range
was from m/z 100 to m/z 2000; the scan cycle was of 1 s. MS/
MS experiments were performed at the same conditions as
above, except for the following: ions were selected based on
their [M+H]+ m/z value, in the mass unit mode, using the
Analyst QS software package and fragmentations were initiated
with nitrogen as the collision gas; the declustering potential was
set to 60 V; the maximum collision energy was typically of 35−
40 eV.
Computations. In order to show that structures with small
methionine sulfur−tyrosine (S−O) distances could be stable in
water, seven different initial conformations were fully optimized
using the B3LYP/6-31G* level of theory and G09 software.30
The starting points were seven conformations from the Protein
Data Bank (PDB 1PLW) chosen because of their different
levels of folding. It should be noted that the structures
deposited in the PDB were recorded not in aqueous solutions
but in micelles thus the stable structures might be different
from those in water. Solvation effects were accounted with the
COSMO option for the polarized continuum model CPCM
considering an aqueous environment. Molecular graphics
images were produced using the UCSF Chimera package
from the Resource for Biocomputing, Visualization, and
Informatics at the University of California, San Francisco
(supported by NIH P41 RR001081).31
G × ε(λ ) =
G × ε(λ )
∑
j
i
i
j
j=1
(ii)
Here Gε(λj) is related to the observed absorbance change
ΔA(λj) of the composite spectrum according to eq 1, Gi is the
linear regression coefficient of the ith species, and εi(λj) is the
molar absorption coefficient of the ith species at the jth
wavelength. Further details of this method have been described
elsewhere.27
The radiation chemical yields of the respective transient
species in the optical spectra were calculated taking the
following representative molar absorption coefficients: ε290
=
4300 M−1 cm−1, ε390 = 2400 M−1 cm−1, and ε405 = 3000 M−1
cm−1 for tyrosyl radicals (TyrO•); ε310 = 2900 M−1 cm−1 and
ε330 = 3000 M−1 cm−1 for hydroxycyclohexadienyl radicals of
tyrosine TyrOH(•OH); ε340 = 5600 M−1 cm−1 for cyclo-
hexadienyl radicals of tyrosine TyrOH(•H); ε320 = 4100 M−1
cm−1 for hydroxycyclohexadienyl radicals of phenylalanine
Phe(•OH); ε320 = 5000 M−1 cm−1 for cyclohexadienyl radicals
of phenylalanine Phe(•H), ε330 = 3400 M−1 cm−1 for
hydroxysulfuranyl radicals of methionine Met(S•−OH), and
ε290 = 3000 M−1 cm−1 for α-(alkylthio)alkyl radicals of
methionine MetS(•CH)/MetS(•CH2).
•
Since the radiation chemical yield of H atoms at pH >3 is
roughly 10-fold lower than that of •OH radicals, the
contribution of cyclohexadienyl radicals during multiregression
analysis procedure of the experimental spectra at natural pH
was neglected. Therefore, one has to be aware that the radiation
chemical yields of hydroxycyclohexadienyl radicals at natural
pH might be slightly overestimated. Moreover, the error limit
(within 15%) validates the spectra resolution and eliminates
unreasonable fits. The error limit ( 15%) allows for 5%
variation in the experimental data and a 10% combined error
in the reported molar absorption coefficients for the spectra of
components under consideration.
UV−vis. The absorption spectra were recorded with the
Perkin-Elmer UV−visible spectrophotometer using a 1 cm
optical path length cell. Nonirradiated solutions were used as
12462
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
•
Scheme 1. First Steps of Oxidation of Leu-enk by OH
Radicals
Figure S1, panels A and B, in the SI). It is noteworthy that, for
this pH, an absorption band at 405 nm started to develop
slowly within 150 μs after the pulse (Figure 1A, insert).
However, its intensity was very weak. At 70 μs after the electron
pulse, the spectrum in the range of 300−360 nm becomes less
intense and much broader, with a more distinct shoulder
located in the 300−315 nm range (Figure 1A, curve b). With
the further elapse of time, the spectrum observed 140 μs after
the pulse (Figure 1A, curve c) was blue-shifted and less intense
characterized by a broad absorption band with weakly
pronounced maximum around 320 nm. On the other hand,
the absorption band with λmax = 405 nm reached the plateau at
140 μs after the pulse, with G × ε405 = 1.1 × 10−4 dm3 J−1 cm−1
(Figure 1A, curve c).
Resolution of Transient Absorption Spectra. It was not
possible to follow the transients directly from the raw data
because of the significant overlap of their absorption spectra.
The experimental spectrum recorded 5 μs after the pulse
(Figure 1A, curve a) could be resolved into contributions from
two components: 1a and 1b (Scheme 1, Figure S1 panel A, in
the SI). The G values are reported in Table 1.
Figure 1. Transient absorption spectra recorded after the pulse
irradiation in N2O-saturated solutions containing Leu-enk (0.2 mM).
A: pH 6.4 with the following time delays: 5 μs (a, red circle), 70 μs (b,
green triangle), and 140 μs (c, blue square). Insets: Kinetic traces
recorded at 305, 325, 350, and 405 nm. B: pH 1.0 with the following
time delays: 5 μs (a, red circle), 70 μs (b, green triangle), and 140 μs
(c, blue square). Insets: Kinetic traces recorded at 305, 325, 350, and
405 nm.
RESULTS
■
Leu-Enkephalin: Transient Species. The reaction of •OH
radicals with Leu-enk was investigated in N2O-saturated
solutions over the concentration range of 0.1−0.6 mM, at
two pH values, natural pH (no buffer, ca. 6.5) and pH 1.
Natural pH. UV−vis Transient Absorption Spectra. A
transient spectrum obtained 5 μs after the electron pulse in
N2O-saturated solutions containing 0.2 mM of Leu-enk at
natural pH (no buffer, pH 6.4−6.5) exhibits a broad absorption
band in the 280−380 nm range with maximum around 325 nm
(G × ε325 = 1.4 × 10−3 dm3 J−1 cm−1; Figure 1A, curve a) and
blue- and red-edge shoulders in 305−315 and 340−350 nm
ranges, respectively. The 325-nm band formed within 5 μs after
the pulse with kobs = (1.6 0.1) × 106 s−1. This apparent rate
constant was proportional to the initial amount of peptide,
which led to k4 = (8 1) × 109 M−1 s−1 (Scheme 1). The 325-
nm band decayed in biexponential mode within 1.2 ms after the
pulse.
The sum over two component spectra with their respective
yields (Gtotal = 0.49 μmol J−1) is in reasonably good agreement
with the expected yield of •OH radicals (GOH = 0.56 μmol J−1).
It is lower than the initial yield of their precursor because the
radicals derived from leucine (Leu) have not been taken into
account. These C-centered radicals (reaction 4c, Scheme 1) do
not absorb substantially at wavelengths >290 nm. The
calculated G(1c) = 0.07 μmol J−1 (Table 1), taken as a
difference between GOH and G(1a + 1b), is in line with the
•
lower rate constant of OH radicals with Leu in comparison
with the rate constants with Tyr and Phe.
This spectrum should represent the combined absorption
bands of hydroxycyclohexadienyl radicals of phenylalanine
Phe(•OH) (1b) and of tyrosine Tyr(OH) (1a) (Scheme 1,
The experimental spectrum recorded 140 μs after the pulse
(Figure 1A, curve c) was best resolved into contributions from
the same components as for the previous spectrum with G(1a)
12463
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
efficiency of transformation of 1a radicals into 1d was of
particular concern. Taking the respective G values extracted at 5
and 140 μs, one could calculate that within 135 μs time
window, the yield of 1a radicals decreases by 0.14 μmol J−1,
whereas that of 1d radicals increases only by 0.03 μmol J−1. It
shows that decay of 1a radicals did not lead quantitatively to
formation of 1d radicals and strongly pointed out for another
reaction pathway of decay of 1a at pH 6.5, probably the
disproportionation of the hydroxylated radicals (1a) leading to
the final product DOPA-Leu-enk (reaction 5c, Scheme 2).
pH 1.0. UV−vis Transient Absorption Spectra. A transient
spectrum obtained 5 μs after the electron pulse in N2O-
saturated solutions containing 0.2 mM of Leu-enk at pH 1.0,
showed a broad absorption band in the 280−380 nm range
with two weakly pronounced maxima at λmax = 305 and 325 nm
with Gε305 = 1.7 × 10−3 dm3 J−1 cm−1 and Gε325 = 1.8 × 10−3
dm3 J−1 cm−1 (Figure 1B, curve a) and some less defined
absorptions below 300 nm toward UV. Contrary to what was
observed at pH 6.5 (Figure 1A), the spectrum exhibits a third
absorption band with two maxima at λmax = 390 and 405 nm
with Gε390 = 4.5 × 10−4 dm3 J−1 cm−1 and Gε405 = 5.1 × 10−4
dm3 J−1 cm−1 (Figure 1B, curve a).
The spectrum in the range of 305−350 nm became less
intense and much broader, 70 μs after the electron pulse, with
two pronounced maxima at 295 and 325 nm with Gε295 = 1.7 ×
10−3 dm3 J−1 cm−1 and Gε325 = 1.6 × 10−3 dm3 J−1 cm−1, more
intense less defined absorptions below 300 nm toward UV, and
a more distinct shoulder located in the 345−355 nm range
(Figure 1B, curve b). Moreover, the absorption bands at 390
and 405 nm continued further slow growth (Figure 1B, insert).
140 μs after the pulse, it became less intense in the range of
285−370 nm (Figure 1B, curve c). On the other hand, the
absorption bands at 390 and 405 nm reached the plateau with
Gε390 = 5.3 × 10−4 dm3 J−1 cm−1 and Gε405 = 6.5 × 10−4 dm3
J−1 cm−1 (Figure 1B, curve c).
Table 1. Primary Yields of the Free Radicals Resulting from
•OH Radical Attack on the Main Residue Targets in Leu
Enkephalin
a
residue
tyrosine
calc [a] (%) exptl [b] (natural pH) exptl [b] (pH 1)
61
31
8
0.39
(70)
0.10
(18)
0.07
(12)
(30)
0.19
(68)
[c]
phenylalanine
leucine
[d]
Phe + Leu
39
(32)
a
[a]: Percentages based on the rate constants with individual amino
acids;26 [b] yields (μmol J−1), 5 μs after the pulse (in parentheses: %
distribution of the OH yield); [c] determination not possible since the
transient absorption spectra of 1b and Phe(•H) are very similar; [d]
determination not possible since both OH and H are precursors of
1c and 1c does not absorb at wavelengths >290 nm. It is important to
note that contribution of tyrosyl radicals (1d, Scheme 2) was
negligible.
•
•
= 0.25 μmol J−1 and G(1b) = 0.10 μmol J−1. Importantly, in
contrast to the picture obtained at 5 μs, the fit of experimental
results required contribution of tyrosyl radicals with G(1d) =
0.03 μmol J−1 (Scheme 2, Figure S1 panel B, in the SI).
Concentration Profiles of Transients. Resolutions of
absorption spectra at any desired time delay following the
electron pulse were performed in order to extract concentration
profiles of 1a and 1b (Figure S1, panel C in the SI). The
Scheme 2. Reaction Steps of Dihydroxycyclohexadienyl
Radicals of Tyr 1a Residue in Leu-enk Leading to the Final
Products
Resolution of Transient Absorption Spectra. At pH 1, the
hydrated electrons are converted into H atoms (reaction 3; see
the Radiolysis of Water section), thus increasing G(•H) from
0.06 to 0.34 μmol J−1 and simultaneously decreasing G(•OH)
from 0.56 to 0.28 μmol J−1. Therefore, the contribution of the
cyclohexadienyl radicals Phe•H and Tyr•H to the absorption
spectrum could not be neglected, in particular, in the 320−370
nm range (vide infra). The spectrum recorded represents the
combined absorption bands of 1a, 1b and 1d (Scheme 1,
Figure S2, panel A in SI). It was only possible to estimate the
sum of the radiation chemical yields of 1b and cyclohehadienyl
Phe(•H) radicals in Phe since their spectral parameters (λmax
and ε) are very similar.32 Thus the experimental spectrum
recorded 5 μs after the pulse (Figure 1B, curve a) could be
resolved into contributions from four components: 1b and
Phe(•H), with G(1b) + G(Phe(•H)) = 0.24 μmol J−1, the
cyclohexadienyl radical in tyrosine Tyr(•H), with G(Tyr(•H))
= 0.16 μmol J−1, and 1d, with G(1d) = 0.14 μmol J−1 (Figure
S2, panel A, in the SI). Interestingly, the contribution of 1a was
found negligible. The sum over four component spectra with
their respective yields (Gtotal = 0.54 μmol J−1) is in reasonably
good agreement with the sum (Gtotal = 0.62 μmol J−1) of the
•
•
expected yield of OH radicals (GOH = 0.28 μmol J−1) and H
atoms (GH = 0.34 μmol J−1) at pH 1. The fact that combined
yields of transients are lower than the initial yields of their
precursors might be again due to the fact that in the resolution
procedure the radicals derived from leucine (1c) have not been
taken into account. The calculated G(1c) = 0.08 μmol J−1,
12464
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
taken as a difference between GOH+H and a sum of G((1d) +
Tyr(•H) + (1b) + Phe(•H)) is in an excellent agreement with
G(Leu•) calculated for pH 6.4 (vide supra).
The experimental spectrum recorded 140 μs after the pulse
(Figure 1B, curve c) was best resolved into contributions from
the same components as for the previous spectrum with G(1b)
+ G(Phe(•H)) = 0.19 μmol J−1, the cyclohexadienyl radical in
tyrosine (Tyr(•H), with G(Tyr(•H)) = 0.12 μmol J−1, and 1d,
with G(1d) = 0.19 μmol J−1 (Figure S2, panel B, in the SI). The
sum of radiation chemical yields of cyclohexadienyl radicals
G(Phe(•H) and Tyr(•H)) ≥ 0.31 μmol J−1 is in a very good
agreement with the expected radiation chemical yield of their
precursor G(•H) = 0.34 μmol J−1.
Indirect determination of the initial yield of 1a was possible.
Indeed, one might expect that the final radiation chemical yield
of 1d radicals (0.19 μmol J−1, measured 140 μs after the pulse)
is equal to the primary radiation chemical yield of 1a. Taking
that value, one might estimate the primary radiation chemical
yield of 1b as G(1b) ≤ 0.09 μmol J−1 and simultaneously the
primary radiation chemical yield of cyclohexadienyl radicals in
phenylalanine Phe(•H) as G(Phe(•H)) ≥ 0.15 μmol J−1
(Figure S2B, in the SI).
Concentration Profiles of Transients. Resolutions of
absorption spectra at any desired time delay following the
electron pulse were performed in order to extract concentration
profiles of 1b, (Phe(•H)) derived from Phe, (Tyr((•H)), and
1d (Figure S2, panel C in the SI). Taking the respective G
values of 1d radicals extracted at 5 and 140 μs, one can calculate
that within 135 μs time window, the yield of 1d radicals
increased by 0.05 μmol J−1.
Figure 2. UV−visible difference absorption spectra (reference = non
irradiated solution) recorded in γ-irradiated N2O-saturated aqueous
solutions. A: Leu-enk (0.5 mM) at pH 6.5. Insert: Change of the
intensity of the absorption band at 290 nm with the dose. B: Met Enk
(0.5 mM) at pH 6.5. Insert: Change of the intensity of the absorption
band at 290 nm with the dose.
Kinetics of TyrO• Formation. The formation of the 405-nm
absorption band showed pH-dependent biexponential kinetics
after the electron pulse. From the mechanistic point of view,
this points out that formation of 1d radicals occurs via two
independent processes. The plot of apparent rate constants
against [H+] shows reasonably good straight lines with the
slopes representing the respective bimolecular rate constants
(k5a = 1.7 × 106 M−1 s−1; k5b = 1.6 × 107 M−1 s−1; Figure S3 in
the SI) for the formation of 1d via proton-catalyzed
dehydration (reactions 5a and 5b, Scheme 2) of the respective
two isomers 1′a and 1″a.
underwent hydroxylation. Indeed, oxidation products of Tyr
were detected as part of the b series fragments (B1 and B4).28
Similarly, oxidation products of Phe were detected as part of
the y series (fragments y2 to y4). Increasing the dose allowed
us to oxidize the peptide as much as 5 times, with detected
species corresponding to the initial peptide plus up to five OH
groups added to it. As first seen with HPLC, peptide dimers
were detected. (Figure S4, panel D, in the SI). The relative
amount of dimer with respect to the Leu-enk monomer was
plotted vs dose (Figure 4). Interestingly, the amount of dimer
was doubled by decreasing the pH of solutions from 6.5 to 1.
Met-Enkephalin: Transient Species. The reaction of
•OH radicals with Met-enk was investigated in N2O-saturated
solutions over the concentration range of 0.1−0.6 mM, at
various pH values.
Natural pH. UV−vis Transient Absorption Spectra. A
transient absorption spectrum obtained 4 μs after electron
pulse in N2O-saturated aqueous solutions containing 0.1 mM of
Met-enk at natural pH (pH 6.0) exhibits broad absorption
bands in the 260−450 nm range with four distinct absorption
maxima located at 295, 320, 390, and 405 nm, a shoulder
around 350 nm, and some less defined absorption below 300
nm toward UV (Figure 5A, curve a). The respective Gε values
are Gε295 = 1.5 × 10−3, Gε320 = 1.25 × 10−3, Gε390 = 4.3 × 10−4,
and Gε405 = 5.5 × 10−4 dm3 mol−1 cm−1 (Figure 5A, curve a). It
is noteworthy that the 405-nm band was fully developed 4 μs
after the pulse (Figure 5A, insert), which is different from that
observed at the same conditions for Leu-enk (vide Figure 1A,
and insert).
Leu-Enkephalin: Final Products. UV−vis. γ-irradiated
N2O-saturated aqueous solutions containing Leu-enk (0.5 mM)
at pH 6.5 exhibit three absorption bands located at 230, 277,
and 290 nm and a weak shoulder in the range of 310−330 nm
(Figure 2A). The intensity of these absorption bands increased
linearly with the dose applied. The small peaks around 277 nm
are reminiscent of 2,3-DOPA-Leu-enkephalin.33 The absorp-
tion band at 290 nm and a shoulder in the range of 310−330
nm could be assigned to bityrosine of Leu-enkephaline.34
HPLC. Leu-enk was irradiated with a dose of 90 Gy in N2O-
saturated aqueous solutions at pH 6.5. There was no
fluorescence at 425 nm in the nonirradiated peptide, as
expected (Figure 3A). The peak that was eluted at 17.5 min
belongs to the native Leu-enk as shown in Figure 3A. Six
products were formed after γ irradiation (Figure 3B,C). Only
the compound with the elution time t = 20.0 min, fluoresced at
425 nm. It was assigned to a Leu-enk dimer linked by
bityrosine.
Mass Spectrometry. Mass spectrometry analysis of irradi-
ated samples showed that the major compounds corresponded
to hydroxylated peptides. Gas phase fragmentation sequencing
of the peptides revealed that both residues Tyr and Phe
The 320-nm band formed within 4 μs after the pulse with
kobs = (1.5 0.1) × 106 s−1 (Figure 5A, insert). The spectra
12465
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
Figure 4. Ratio of the intensities of the m/z of the bityrosine derivative
enkephalin on the m/z of the enkephalin versus the dose. Met-enk at
pH 1.5 (red square), Met-enk at natural pH (blue square), Leu-enk at
pH 1.5 (red circle), Leu-enk at natural pH (blue circle).
Figure 3. HPLC chromatograms of nonirradiated (panel A) and
irradiated with 90 Gy (panels B and C) aqueous solutions containing
Leu-enk (0.25 mM) at pH 6.5 saturated with N2O. Absorption
detection at 284 nm (blue line, left axis). Fluorescence detections with
excitation at 284 nm and emission at 305 nm (magenta line, right axis)
and 425 nm (orange line, right axis). For the panel C, the units of right
y axis have been multiplied by 10.
recorded at 70 and 140 μs after the pulse exhibited similar
features to that recorded at 4 μs except that absorption
intensities in the range of 260−370 nm were lower whereas that
located at 390 and 405 nm remained almost constant (Figure 5,
curves b and c). Moreover, a peak with absorption maximum at
320 nm was now as a shoulder, and in the opposite, a shoulder
around 350 nm has been developed as a weak peak.
Figure 5. A. Transient absorption spectra recorded after electron pulse
irradiation of an N2O-saturated solution containing Met-enk (0.2 mM)
at pH 6.0 with time delay: 4 μs (a, red circle), 70 μs (b, green triangle),
and 140 μs (c, blue square). Insets: Kinetic traces recorded at 305, 320,
350, and 405 nm. B: Transient absorption spectra recorded after
electron pulse irradiation of an N2O-saturated solution containing
Met-enk (0.2 mM) at pH 1.0 with time delay: 6 μs (a, red circle), 70
μs (b, green triangle), and 140 μs (c, blue square). Insets: Kinetic
traces recorded at 305, 320, 350, and 405 nm.
The characteristic peaks at 390−405 nm resemble those of
the tyrosyl radical35 (2d). It suggests that, with Met-enk, 2d is
formed through a process that is not possible in Leu-enk. As the
only difference with Leu-enk is the presence of a methionine
instead of a leucine residue, it led us to assume that this new
process for the formation of 2d should involve a precursor
36
•
resulted from a reaction between Met and the OH radical.
Resolution of Transient Absorption Spectra. For the same
reason as for Leu-enk at both pH’s, the interpretation of the
raw absorption spectra in the region of 280−450 nm is difficult.
Moreover, in addition to five transients that could potentially
contribute to this part of the spectrum [2a and 2b (Schemes 3
and 4), two cyclohexadienyl radicals (Phe(•H), Tyr (•H)), and
tyrosyl radical 2d (Scheme 4)], one might expect transients
12466
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
•
derived from Met residue. There are six potential species which
might be formed upon oxidation of methionine by •OH
radicals: hydroxysulfuranyl radicals (Met(>S•−OH)) (2c),
intramolecularly three-electron-bonded S∴O and S∴N radicals,
intermolecularly three-electron-bonded S∴S dimeric radical
cations, monomeric radical cations (MetS•+) (2e), and α-
(alkylthio)-alkyl radicals (Met(αS)).36a,37 All of them absorb in
the region of 280−450 nm and could contribute to the
experimental spectrum. The contribution of radicals coming
from the reactions of H atoms were neglected as explained
above.
Scheme 3. First Steps of Oxidation of Met-enk by OH
Radicals
The experimental spectrum recorded 4 μs after the pulse
(Figure 5A, curve a) could be resolved into contributions from
three components: 2a and 2d from Tyr and 2b from Phe
(Schemes 3 and 4, Figure S5, panel A, in the SI). The respective
yields of each transient are collected in Table 2.
Table 2. Primary Yields of the Free Radicals Resulting from
•OH Radical Attack on the Main Residue Targets in Met
a
Enkephalin
calc [a]
(%)
exptl [b] (pH
1)
residue
tyrosine
exptl [b] (natural pH)
41
21
0.31(2a) + 0.21 (2d) (57)
[c]
phenylalanine
methionine
Tyr+Met
0.09
(17)
0.14
(26)
(83)
0.04
(14)
[c]
38
79
Scheme 4. Reaction Steps of Dihydroxycyclohexadienyl
Radicals of Tyr 2a Residue and Hydroxysulfuranyl Radicals
of Met Residue 2c in Met-enk Leading to the Final Products
(86)
a
[a] Percentages based on the rate constants with individual amino
acids.; [b] yields in μmol J−1; [c] both Tyr(OH) and MetS•+ are
precursors.
The sum over three component spectra (Gtotal = 0.54 μmol
J−1) is in very good agreement with the expected yield of •OH
radicals (GOH = 0.56 μmol J−1). Moreover, there is no
contribution of Met-derived radicals (2c and derivatives). The
initial concentration of the transient species (2c; 0.14 μmol J−1)
was calculated indirectly and taken as the concentration of
tyrosyl radicals (2d) measured 4 μs after the pulse (Figure S5A,
in the SI).
The experimental spectrum recorded 140 μs after the pulse
(Figure 5A, curve c) was best resolved into contributions from
the same components as for the previous spectrum with G(2a)
= 0.18 μmol J−1, G(2b) = 0.08 μmol J−1, and G(2d) = 0.14
μmol J−1.
Concentration Profiles of Transients at Natural pH. Panel
C in Figure S4 shows radiation chemical yield (G) versus time
plots for the three intermediates 2a, 2b, and 2d. On the basis of
extracted concentration profiles, it was possible to conclude
whether the apparent decay of 2a occurs with accompanying
formation of 2d (reaction 7a, Scheme 4). Taking the respective
G values extracted at 4 and 140 μs, one can simply calculate
that, within the 136 μs time window, the yield of 2a decreases
by 0.13 μmol J−1, whereas the yield of 2d remains constant.
From the mechanistic point of view, this shows that decay of 2a
does not lead quantitatively to formation of 2d radicals and
strongly points out for another reaction pathway of decay of 2a,
possibly leading to DOPA-Met-enk (reaction 7b, Scheme 4).
pH 1. UV−vis Transient Absorption Spectra. A transient
absorption spectrum obtained 4 μs after electron pulse in N2O-
saturated aqueous solutions containing 0.2 mM of Met-enk at
pH 1.0, exhibited broad absorption bands in the 270−450 nm
12467
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
range with five distinct maxima (295, 325, 350, 390, and 405
nm; Figure 5B, curve a; Gε295 = 1.8 × 10−3, Gε325 = 1.4 × 10−3,
Gε350 = 6.8 × 10−4, Gε390 = 6.2 × 10−4, Gε405 = 7.2 × 10−4 dm3
mol−1 cm−1; Figure 5B, curve a). The 320-nm band formed
within 4 μs after the pulse with kobs = (1.5 0.1) × 106 s−1
(Figure 5B, insert). Again the bands at 390−405 nm were
developed within 4 μs after the pulse (Figure 5B curve a,
insert). The spectra recorded at 70 and 140 μs after the pulse
exhibited similar features to those recorded at 4 μs except that
absorption intensities in the range of 260−370 nm were lower
whereas those located at 390−405 nm slightly increased
(Figure 5B, curves b and c).
Resolution of Transient Absorption Spectra. The con-
tribution of the cyclohexadienyl radicals Phe•H and Tyr•H to
the absorption spectrum cannot be neglected, in particular, in
the 320−370 nm range (vide supra). The spectrum should thus
represent the combined absorption bands of 2b (Scheme 3),
Tyr(•H) and phenylalanine Phe(•H), and 2d (Scheme 4)
(Figure S6, panels A and B, in the SI).
radicals increased by 0.03 μmol J−1. It points out that formation
of 2d occurs via two independent processes (vide infra).
The apparent first order rate constants at 325 and 405 nm
were proportional to the initial concentration of Met-enk,
hence rate constants k6 equal to (1.16 0.1) × 1010 M−1 s−1
and (1.36 0.2) × 1010 M−1 s−1 were deduced (Scheme 3).
Their similarity points out that formation of TyrO• radicals is
not controlled by the intramolecular transformation involving
radical transients derived from Met residue.
Met-Enkephalin: Final Products. UV−vis. Met-enk (0.5
mM) was irradiated in N2O-saturated aqueous solutions at
natural pH (ca. 6.0) with several doses in the range of 20−120
Gy (Figure 2B). The absorption band at 290 nm increased
linearly with the dose. (Figure 2B, insert). Contrary to the
N2O-saturated solution containing Leu-enk (Figure 2A) no
small peak with λmax around 277 nm was observed. Conversely,
the absorbance around 320 nm and up to the visible region
increased with the dose. Taken together, it indicates the
formation of dityrosines.
HPLC. Three peaks with elution times 16.8 min, 18.6 and
19.6 min were observed after γ-irradiation (Figure S7, in the
SI). The peak eluting at 16.8 min observed in the nonirradiated
sample was assigned to the native Met-enk. Two products
represented by the peaks with elution times 18.6, and 19.6 min
exhibited fluorescence at 425 nm. The peak eluted after 18.6
min and detected by the same fluorescence setting (ex: 284 nm,
em: 425 nm) was characterized by the highest intensity
emission at λ = 425 nm. Therefore its presence could be
assigned to the formation of dityrosine of Met-enk,34 in line
with our earlier observations in solutions containing Met-enk
From the spectral resolutions, it was only possible to estimate
the sum of the radiation chemical yields of 2b and Phe(•H)
radicals since their spectral parameters (λmax, and ε) are very
similar (vide supra). Taking this into account, the experimental
spectrum recorded 4 μs after the pulse (Figure 5B, curve a)
could be resolved into contributions from four components: 2b
and Phe(•H), with G(2b) + G(Phe(•H)) = 0.18 μmol J−1,
(Tyr(•H), with G(Tyr(•H)) = 0.11 μmol J−1, and 2d with
G(2d) = 0.21 μmol J−1 (Figure S6, panel A, in the SI). Again
the sum over four component spectra with their respective
yields (Gtotal = 0.52 μmol J−1) is lower than the sum (of the
expected yield of •OH radicals and •H atoms (GOH + GH = 0.28
+ 0.34 = 0.62 μmol J−1). It might be due to the fact that the
absorption spectrum of CH3S• from the H-atom reaction with
Met has not been taken into account.
The experimental spectrum recorded 140 μs after the pulse
(Figure 5B, curve c) was best resolved into contributions from
the same components as for the previous spectrum with G(2b)
+ G(Phe(•H)) = 0.11 μmol J−1, the cyclohexadienyl radical in
Tyr (Tyr(•H), with G(Tyr(•H)) = 0.08 μmol J−1, and the
tyrosyl radical 2d, with G(2d) = 0.24 μmol J−1 (Figure S6,
panel B, in the SI). For pH 1, one might expect that the final
radiation chemical yield of 2d (0.24 μmol J−1) is equal to the
sum of the primary radiation chemical yields of 2a and 2c.
Taking that value, one might estimate the primary radiation
chemical yield of 2b: G(2b) = 0.04 μmol J−1. This value allows
calculation of the primary radiation chemical yield of
cyclohexadienyl radicals in phenylalanine (Phe(•H)): G(Phe-
(•H)) = 0.14 μmol J−1. The sum of radiation chemical yields of
cyclohexadienyl radicals G(Phe(•H) and G(Tyr(•H)) = 0.25
μmol J−1 is lower than the radiation chemical yield of their
precursor G(•H) = 0.34 μmol J−1. Indeed the absorption
spectrum of thiyl radicals (CH3S•) derived from Met and
formed with G(CH3S•) = 0.09 μmol J−1 has not been taken
into account.38
and the N3 radical as one-electron oxidant.20 The product
•
eluted after 19.6 min also exhibited a characteristic fluorescence
at 425 nm and was tentatively assigned to another isomer of
dityrosine of Met-enk.
Mass Spectrometry. Solutions of Met-enk were irradiated up
to 900 Gy. Irradiation of the samples yielded a number of
distinct molecular species: (i) oxidized species detected as mass
increments of 16 Da on the monocharged ion, (ii) dimeric
biprotonated at m/z = 577.22, and (iii) a molecular species
corresponding to a loss of 47 Da from the YGGFM peptide. In
all cases the mass peaks increased with the γ−irradiation dose
(Figure 6 and Figures S8A and B, in the SI).
Each ion described above was fragmented in a gas phase. The
ion at m/z = 528.33 (loss of 47 Da) corresponded to to
desulfurated methionine, also called aminobutyric acid (Aba).
Its formation is attributed to the reactions of H atoms with
methionine.38 As expected this reaction is minor in solutions at
natural pH but quite important in acidic medium. Fragmenta-
tion in a gas phase of the ions at m/z = 590.22 (increment of 16
Da) revealed that they correspond to a mixture of two
hydroxylated peptides: at methionine (i.e., methionine
sulfoxide) and at tyrosine (i.e., DOPA-Met-enk).
Interestingly, we found a very low amount of hydroxylated
phenylalanine, contrary to what was observed in Leu-enk. In
addition, for the highest dose, we also found doubly
hydroxylated forms of Met enk corresponding to the presence
of either methionine sulfone or a mixture of methionine
sulfoxide and hydroxylated tyrosine (DOPA-Met-enk). As for
the dimer, the ratio of intensities (dimer)/(Met-enk) did not
vary with the pH from 1.5 to 6.5 (Figure 4).
Concentration Profiles of Transients. Resolutions of
absorption spectra at any desired time delay following the
electron pulse were performed in order to extract concentration
profiles of 2b and Phe(•H) derived from Phe, Tyr((•H) and
2d. Radiation chemical yields (G) versus time plot for these
transients are shown in Figure S6, panel C (in the SI). Taking
the respective G values of 2d extracted at 6 and 140 μs, one can
calculate that within 134 μs time window, the yield of 2d
Computational Analysis. The aim was to verify whether
conformations of Met-enk exhibiting proximity between the
sulfur atom of Met residue and the OH group of Tyr residue in
12468
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
Figure 7. (A) Experimental conformation of Met-enk in micelles
(PDB, 1PLW, structure 51). (B) Structure of lowest energy optimized
in water (B3LYP/6)31G*°). Both would be in agreement with a very
fast intramolecular electron transfer between MetS•OH/Met+• and
Tyr.
Figure 6. Variation of the relative intensity of the different final
compounds formed after radiolysis of Met-enkephalin in N2O-
saturated solutions at natural pH and at pH 1.5.
water are stable enough, to allow very fast intramolecular
electron transfer from Tyr residue to Met residue. A complete
conformational analysis of this peptide was not performed since
several papers dealing with this subject were published.39
However, the question of the distance between the sulfur atom
of Met and the −OH group of Tyr (S−O) has never been
addressed before.
We used a simple DFT approach (B3LYP/6-31G*) and the
starting points were seven conformations among the ca. 80
ones deposited in the PDB (1PLW). It should be noted that
these experimental structures were recorded in micelles and not
in water, thus their stability in water can be questioned. The
starting points were chosen for their different levels of folding.
Among the fully optimized structures, two exhibited small S−O
distances (lower than 6 Å) in agreement with the very fast
electron transfer observed by pulse radiolysis. Their energies
were 3.83 and 6.90 kJ mol−1 higher than the most stable
structure, indicating that both could be attained in solutions at
room temperature. The structure of lowest energy is shown in
Figure 7B with an experimental conformation in micellar
environment (Figure 7A).
mechanisms of oxidation, the final products, and the influence
of the sequence and pH on their nature.
Oxidation Mechanism in Leu-enk. Fast formation of OH
adducts on both aromatic residues Tyr (1a) (reaction 4a,
Scheme 1) and Phe (1b) in Leu-enk (reaction 4b, Scheme 1)
(k4 = (8 1) 109 M−1 s−1) was confirmed unquestionably by
pulse radiolysis. Resolution of transient absorption spectra at
natural pH (Figure S1A, in the SI) allowed direct determination
of their initial yields (Table 1) and thus primary distribution of
•
the OH attack (in %) on the main residue targets (Tyr and
Phe) (Table 1, column 2). They follow the trend of those
•
calculated by taking the rate constants of the OH radical
reactions with the respective individual amino acids (Table 1,
column 1). However the one of 1b (18%) is substantially lower
to that calculated (31%). This observation is consistent with the
already reported low amount of o- and p-tyrosine formed from
•
•
the oxidation of Phe by OH radicals (10% of the total OH
yield).40
•
At pH 1 the primary distribution yield of the OH radical
attack on both Phe and Leu residues (32%) is similar to that for
natural pH (30%) (Table 1, columns 2 and 3). The primary
distribution yield of 1a (68%) normalized to the total yield of
•OH radicals (0.28 μmol J−1) available at pH 1 is practically the
same as that measured at natural pH (70%) (Table 1, columns
2 and 3).
DISCUSSION
■
We have investigated the one-electron oxidation of both Leu-
and Met-enkephalins by OH radicals. These peptides contain
•
•
major targets of reactive oxygen species, two aromatic residues
(tyrosine and phenylalanine) and in Met-enk additionally one
methionine residue. Our aims were to determine the kinetic
process, to evaluate the proton-assisted steps in the
It is well-known that OH radicals form ortho and meta
isomers of the dihydroxy-cyclohexadienyl radical adducts on the
tyrosine ring, and an addition to the ortho position of the
tyrosine ring is the most favorable.41 These isomers undergo
12469
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
•OH attack probability on Phe residue equal to 14%, i.e.
practically the same as that measured at natural pH (16%)
(Table 2, columns 2 and 3).
Intramolecular Electron Transfer in Met-enk. Three
questions arise which are directly connected with the postulated
intramolecular electron transfer in Met-enk: (i) Can
intermolecular electron transfer (ET) be neglected? (ii) What
is the rate constant for intramolecular ET? (iii) What is the
mechanism of intramolecular ET, does it occurs through space
(solvent) or through the peptide bonds?
The answer for the first question is the following: under
experimental conditions (0.2 mM of Met-enk), intermolecular
ET could be neglected. The rate constant for intermolecular
electron transfer between Met(S∴S)+ and Tyr (k = 3.8 × 107
M−1 s−1),48 can only compete with intramolecular ET for very
high concentrations of Met-enk (>0.1 M, vide infra).
proton-catalyzed dehydration with different rate constants.
Similarly, 1′a and 1″a underwent dehydration (reactions 5a and
5b, Scheme 2). Thus, the decay pathway of 1a is controlled by a
competition between disproportionation (reaction 5c, Scheme
2) and acid-catalyzed dehydration mechanisms (reactions 5a
and 5b, Scheme 2).42 Consequently the amount of dimers
linked by dityrosines was low in neutral medium and higher in
acidic one (Figure 4). The final products resulting from the
•OH attack on tyrosine in Leu-enk are mostly either DOPA-
Leu-enk (reaction 5c, Scheme 2) in neutral and slightly acidic
medium or dimers (3,3′-dityrosine-Leu-enk) in acidic medium
(reaction 5d, Scheme 2)
Oxidation Mechanism in Met-enk. Fast formation of OH
adducts on both aromatic residues Tyr (2a) (reaction 6a in
Scheme 3) and Phe (2b) in Met-enk (reaction 6b, Scheme 3)
(k6 = (1.3 0.1) 1010 M−1 s−1) was also confirmed by pulse
radiolysis. However, no transient coming from methionine was
observed (2c and/or 2e).36a,43,37d,44
The intramolecular ET step was not observed using the
ELYSE picosecond pulse radiolysis set up.49 Using the Met-enk
initial concentration of 1.1 mM, the formation of 2d (observed
at 405 nm) was found to occur with τ1/2 ≈ 60 ns which
corresponds to the pseudo-first order rate constant equal to 1.2
× 107 s−1. This observation increased the lower limit for the
rate constant of intramolecular ET by nearly 1 order of
magnitude. Dividing this value by a concentration of Met-enk
used, one can arrive with the value of ∼1 × 1010 M−1 s−1, which
is very close to the k6 value. In another words, formation of 2d
is controlled in principle by reaction 6 (Scheme 3), and not by
reaction 7c (Scheme 4). A high value of the rate constant for
intramolecular ET in Met-enk is consistent with the lack of α-
(alkylthio)alkyl radicals in the resolution of transient absorption
spectra and with the lack of CO2 formation.50 The former
radical might be potentially produced by the competitive
deprotonation of 2c and 2e. Formation of CO2 would have
resulted from the competitive oxidation of the carboxylate
function in Met via intramolecular ET in 2c and 2e transients.50
The difficulty in elucidating whether the intramolecular ET
occurs through space or through the peptide bonds is due to
the large variety of conformations in which the enkephalins can
exist in aqueous environment. In the Protein Data Bank (PDB
1PLW) ca. 80 conformations of Met-enk in micelles were
observed by NMR. Among them several show close proximity
between these two residues. A typical structure (structure 51,
PDB 1PLW) is shown in Figure 7A. Optimization with DFT
methods taking into account solvation in water showed that
conformations with a close proximity of the sulfur atom and the
−OH group of tyrosine were stable in water at room
temperature (Figure 7B). Calculations showed that in the
radical cation of Met-enk the odd electron orbital was
delocalized on the sulfur atom and on the tyrosine aromatic
ring.51 Previous conformational analysis using Empirical
Conformational Energy Program for Peptides (ECEPP) has
shown existence of hydrogen bonds in the lowest energy
structures of Met-enk. One of them having the length of 1.90 Å
involves Tyr and Met residues.39c A more recent molecular
mechanics optimization (BIOS-CHARMM) has shown the
existence of 4 hydrogen bonds in the optimized structure of
Met-enk (type I′- β-bend). One of them involves the proton of
the N-terminal amine group (in Tyr residue) and the oxygen
atom in the carbonyl group located between Phe and Met
residues.49
Interestingly, a comparison of the absorption spectra of
transients derived from Met-enk and Leu-enk recorded 4 μs
after the pulse and at natural pH revealed a substantial
difference (Figures 5A and 1A, respectively). The characteristic
absorption band with λmax = 405 nm of tyrosyl radical is clearly
seen very fast with Met-enk (within 4 μs, Figure 5A, insert),
and absent with Leu-enk albeit rate constants of reactions with
•OH radicals are very close ((8 1) 109 and (1.3 0.1) 1010
M−1 s−1 for Leu-enk and Met-enk, respectively). These
observations led us to the hypothesis that TyrO•radical (2d)
should be formed through a new process not present in Leu-
enk, i.e an electron transfer involving a hydroxysulfuranyl
radical on Met residue (2c) and a Tyr residue (reaction 7c,
Scheme 4). The reduction potential E0(MetS•OH/MetS) is not
known. Assuming a value similar to that of dimethylsufide
(DMS) (+1.41 V (vs NHE)45) and E0(TyrO•, H+/TyrOH) =
+1.0 V (vs. NHE)46 at pH 6, it would be thermodynamically
feasible.
Resolution of transient absorption spectra at natural pH
(Figure S5A, in the SI) after pulse radiolysis of Met-enk
solutions allowed direct determination of the initial concen-
tration of the transient species (Table 2, column 2). As it was
found for Leu-enk, the yields follow the trend of the rate
constants of the •OH radical reactions with the respective
individual amino acids (Table 2, column 1). However, the
primary distribution yields of 2b (17%) and 2c (26%)
normalized to the total yield of •OH radicals were substantially
lower than those calculated (21%, 38%, respectively) taking the
•
respective rate constants of OH radicals with Phe and Met.
Transients 2a and 2c underwent at pH 1 proton-catalyzed
dehydrations (reactions 7a and 7d, Scheme 4). Both processes
led to 2d within the same time window (less than 4 μs). The
intramolecular electron transfer from Tyr to Met radical would
be thermodynamically feasible on the basis of the reduction
potential of DMS45 and E0(TyrO•, H+/TyrOH) = 1.3 V (vs
NHE)46,36b at pH 1. Similar process was observed in Met-enk
during photosensitized oxidation of Tyr and Met by the triplet
state of 4-carboxybenzophenone.47
Since both 2a and 2c were precursors of 2d, a resolution of
absorption spectra at pH 1 (Figure S6B, in the SI) after pulse
radiolysis of Met-enk solutions did not allow even indirect
determination of their initial concentrations. Thus, the primary
•
distribution yield of the OH attack (in %) could be only
Earlier oxidation studies of peptides containing Tyr and Met
•−
calculated for both residues (Met + Tyr) and was equal to 86%
(Table 2, column 3). The last value allowed calculation of the
residues and Met-enk using Br2 as a selective one-electron
oxidant allowed determination of first-order rate constants of
12470
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
intramolecular ET involving Met(S∴Br) and Tyr residues (1.1
REFERENCES
■
52
•
× 105 s−1). Similar feature was observed when OH radical
was used for oxidation of Tyr-Met dipeptides. The lower limit
for the rate constant of intramolecular ET was found to be 2.5
× 106 s−1.53
(1) Henry, B.; Duty, S.; Fox, S. H.; Crossman, A. R.; Brotchie, J. M.
Exp. Neurol. 2003, 183, 458−468.
(2) (a) Mathieu-Kia, A. M.; Fan, L. Q.; Kreek, M. J.; Simon, E. J.;
Hiller, J. M. Brain Res. 2001, 893, 121−134. (b) Yew, D. T.; Li, W. P.;
Webb, S. E.; Lai, H. W.; Zhang, L. Exp. Gerontol. 1999, 34, 117−133.
(c) Rissser, D.; You, Z. B.; Cairns, N.; Herrera-Marschitz, M.; Seidl, R.;
Schneider, C.; Terenius, L.; Lubec, G. Neurosci. Lett. 1996, 203, 111−
114.
On the other hand, the rate constant for intramolecular ET
between Met(S∴Br) and Tyr residues measured in Tyr-(Pro)3-
Met (a peptide having the same number (3) of amino acid
residues between Tyr and Met residues as Met-enk) was found
to decrease by 1 order of magnitude and equal to 1.1 × 104
s−1.54 In this case intramolecular ET is likely partitioning along
the peptide backbone and direct water mediated contacts
between groups bearing radical sites.
Taking all these structural and experimental features above
into account, one can suspect that most probably in Met-enk an
intramolecular ET occurs rather through space (water) pathway
with a possible involvement of the hydrogen bond shortcuts.
(3) Grune, T.; Jung, T.; Merker, K.; Davies, K. J. A. Int. J. Biochem.
Cell Biol. 2004, 36, 2519−2530.
(4) Stryer, L. Biochemistry; W. H. Freeman and Company: New York,
1988.
(5) Terenius, L.; Wahlstrom, A.; Lindeberg, G.; Karlsson, S.;
Ragnarsson, U. Biochem. Biophys. Res. Commun. 1976, 71, 175−179.
(6) Hughes, J.; Smith, T. W.; Kosterlitz, H. W.; Fothergill, L. A.;
Morgan, B. A.; Morris, H. R. Nature 1975, 258, 577−579.
(7) (a) Thanavala, V.; Kadam, V. J.; Ghosh, R. Curr. Drug Targets
2008, 9, 887−894. (b) Smith, H. S. Pain Physician 2008, 11 (2 Suppl),
S121−132. (c) Machelska, H. Neuropeptides 2007, 41, 355−363.
(8) Papathanassoglou, E. D.; Giannakopoulou, M.; Mpouzika, M.;
Bozas, E.; Karabinis, A. Nurs. Crit. Care 2010, 15, 204−216.
(9) Sobocanec, S.; Balog, T.; Sverko, V.; Marotti, T. Physiol. Res.
2005, 54, 97−104.
(10) Sobocanec, S.; Kusic, B.; Sverko, V.; Balog, T.; Marotti, T.
Biogerontology 2006, 7, 53−62.
(11) Balog, T.; Sobocanec, S.; Sverko, V.; Marotti, T. Neuropeptides
2004, 38, 298−303.
(12) Plotnikoff, N. P.; Faith, R. E.; Murgo, A. J.; Herberman, R. B.;
Good, R. A. Clin. Immunol. Immunopathol. 1997, 82, 93−101.
(13) (a) Bolling, S. F.; Badhavar, V.; Schwartz, C. F.; Oeltgen, P. R.;
Kilgore, K.; P., S. J. Thorac. Cardiovasc. Surg. 2001, 122, 476−481.
(b) Plotnikoff, N. P.; Faith, R. E.; Murgo, A. J.; Good, R. A.
Enkephalins and Endorphins. Stress and Immune System; Plenum Press:
New York, 1986; pp 1−446.
CONCLUSIONS
■
The oxidation of proteins is known to be a major event in many
problems related to oxidative stress. After oxidation, the protein
can be totally destabilized, and lose55 or increase its activity.56
Although a lot is known about protein oxidation still many facts
are not understood.
In this work we have compared the oxidation processes of
two pentapeptides differing only by their C-terminal residues.
This difference had consequences in the oxidation process. In
Leu-enk, the N-terminal residue became DOPA instead of
tyrosine. In Met-enk, tyrosine dimers are created as a
consequence of intramolecular electron transfer to methionine.
It means that the fate of tyrosine in oxidative stress and thus the
biomarkers coming from this residue can differ according to the
protein sequence and the neighboring residues and the
character of the oxidant.
(14) Reichling, D. B.; Levine, J. D. Ann. Neurol. 2011, 69, 13−21.
(15) Rosei, M. A.; Blarzino, C.; Foppoli, C.; Coccia, R.; De Marco, C.
Biochem. Biophys. Res. Comms 1991, 179, 147−152.
(16) Rabgaoui, N.; Slaoui-Hasnaoui, A.; Torreilles, J. Free Radic. Biol.
Med. 1993, 14, 519−529.
Similar difficulties may concern other residues thus the story
of protein oxidation is far from being completed.
(17) Fontana, M.; Mosca, L.; Rosei, M. A. Biochem. Pharmacol. 2001,
61, 1253−1257.
ASSOCIATED CONTENT
■
S
* Supporting Information
(18) Coccia, R.; Foppoli, C.; Blarzino, C.; De Marco, C.; Rosei, M. A.
Biochim. Biophys. Acta 2001, 1525, 43−49.
Resolution of the spectral components and their concentration
profiles (Figures S1, S2, S5, and S6), variation of the rate
constant vs pH for Leu enk (Figure S3), ES-MS spectra
(Figures S4 and S8), and HPLC chromatograms (Figure S7)
are included. This material is available free of charge via the
Internet at http://pubs.acs.org.
(19) Fontana, M.; Pecci, L.; Schinina, M. E.; Montefoschi, G.; Rosei,
M. A. Free Radic. Res. 2006, 40, 697−706.
(20) Mozziconacci, O.; Mirkowski, J.; Rusconi, F.; Pernot, P.;
́
Bobrowski, K.; Houee-Levin, C. Free Radical Biol. Med. 2007, 43, 229−
240.
(21) (a) Nagy, P.; Kettle, A. J.; Winterbourn, C. C. Free Radic. Biol.
Med. 2010, 49, 792−799. (b) Nagy, P.; Kettle, A. J.; Winterbourn, C.
C. J. Biol. Chem. 2009, 284, 14723−14733.
(22) Spinks, J. W. T.; Woods, R. J. Introduction to Radiation
Chemistry, 3rd ed.; Wiley: New York, 1990.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: k.bobrowski@ichtj.waw.pl; Chantal.houee@u-psud.fr.
Notes
The authors declare no competing financial interest.
(23) Bobrowski, K. Nukleonika 2005, 50 (Supplement3), S67−S76.
(24) (a) Buxton, G. V.; Stuart, C. R. J. Chem. Soc. Faraday Trans.
1995, 91, 279−281. (b) Schuler, R. H.; Patterson, L. K.; Janata, E. J.
Phys. Chem. 1980, 84, 2088−2090.
ACKNOWLEDGMENTS
(25) Janata, E.; Schuler, R. H. J. Phys. Chem. 1982, 86, 2078−2084.
(26) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J.
Phys. Chem. Ref. Data 1988, 17, 513−886.
■
We thank Dr. J.-L. Marignier for the use of the ELYSE ultra fast
pulse radiolysis set up. The support and sponsorship of the
COST Action CM0603 on “Free Radicals in Chemical Biology
“(CHEMBIORADICAL), of the COST Action CM1001 “Non-
enzymatic protein oxidation”, and the financial help of the Non-
Cofinancing International Project of the Polish Ministry of
Science and Higher Education (455/N-COST/2009/0) are
kindly acknowledged.
(27) Marciniak, B.; Bobrowski, K.; Hug, G. L. J. Phys. Chem. 1993,
97, 11937−11943.
(28) Rusconi, F. BMC Bioinformat. 2006, 7, 226.
(29) (a) Rusconi, F.; Belghazi, M. Bioinformatics 2002, 18, 644.
(b) Rusconi, F. Bioinformatics 2009, 25, 2741−2742.
(30) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
12471
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472

The Journal of Physical Chemistry B
Article
B.; Petersson, G. A. et al. Gaussian 09, revision A.02; Gaussian Inc.:
Wallingford, CT, 2009.
(31) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. J. Comput. Chem. 2004,
25, 1605−1612.
(56) (a) Sicard-Roselli, C.; Lemaire, S.; Jacquot, J.-P.; Favaudon, V.;
́
Marchand, C.; Houee-Levin, C. Eur. J. Biochem. 2004, 271, 3481−
3487. (b) DalleDonne, I.; Milzani, A.; Colombo, R. Biochemistry 1999,
38, 12471−12480.
(32) (a) Chrysochoos, J. Radiat. Res. 1968, 33, 465−479. (b) Solar, S.
Radiat. Phys. Chem. 1985, 26, 103−108. (c) Dizdaroglu, M.; Simic, M.
G. Radiat. Res. 1980, 93, 437. (d) Lichtin, N. N.; Shafferman, A.
Radiat. Res. 1974, 60, 432−440.
(33) Zegota, H.; Kolodziejczyk, K.; Krol
Chem. 2005, 72, 25−33.
(34) Malencik, D. A.; Anderson, S. R. Amino Acids 2003, 25, 233−
247.
(35) Bansal, K. M.; Fessenden, R. W. Radiat. Res. 1976, 67, 1−8.
(36) (a) Hiller, K.-O.; Masloch, B.; Gobl, M.; Asmus, K.-D. J. Am.
́ ́
, M.; Krol, B. Radiat. Phys.
̈
Chem. Soc. 1981, 103, 2734−2743. (b) Prutz, W. A.; Butler, J.; Land, E.
̈
J.; Swallow, A. J. Int. J. Radiat. Biol. 1989, 55, 539−556. (c) Prutz, W.
̈
A. In Sulfur-centered reactive intermediates in chemistry and biology;
Chatgilialoglu, C., Asmus, K.-D., Eds.; Plenum Press: New York, 1990;
pp 389. (d) Prutz, W. A.; Butler, J.; Land, E. J. Free Rad. Res. Comms.
̈
1986, 2, 69−75.
(37) (a) Hiller, K.-O.; Asmus, K.-D. J. Phys. Chem. 1983, 87, 3682−
3688. (b) Asmus, K.-D. In Sulfur-Centered Reactive Intermediates in
Chemistry and Biology; Chatgilialoglu, C., Asmus, K.-D., Eds.; Plenum
Press: New York, 1990; Vol. 197, pp 155−172. (c) Asmus, K.-D.;
Gobl, M.; Hiller, K.-O.; Mahling, S.; Monig, J. J. Chem. Soc., Perkin
̈
̈
Trans. 2 1985, 641−646. (d) Bobrowski, K.; Holcman, J. Int. J. Radiat.
Biol. Relat. Stud. Phys., Chem. Med. 1987, 52, 139−144.
(38) Mozziconacci, O.; Bobrowski, K.; Ferreri, C.; Chatgilialoglu, C.
Chem.Eur. J. 2007, 13, 2029−2033.
(39) (a) Vengadesan, K.; Gautham, N. Biopolymers 2004, 74, 476−
494. (b) Ramya, L.; Gautham, N. Biopolymers 2012, 97, 165−176.
(c) Zhan, L.; Chen, J. Z. Y.; Liu, W. K. Biophys. J. 2006, 91, 2399−
2404. (d) Shen, M. I.; Freed, K. F. Biophys. J. 2002, 82, 1791−1808.
(e) Hansman, U. H. E.; Okamoto, Y.; Onuchic, J. N. Proteins: Struct.,
Funct., Genet. 1999, 34, 472−483.
(40) Goshe, M. B.; Chen, Y. H.; Anderson, V. E. Biochemistry 2000,
22, 1761−1770.
(41) Solar, S.; Solar, W.; Getoff, N. J. Phys. Chem. 1984, 88, 2091−
2095.
(42) von Sonntag, C. The Chemical Basis of Radiation Biology; Taylor
and Francis: New York, 1987; pp 1−515.
(43) Prutz, W. A.; Butler, J.; Land, E. J. Int. J. Radiat. Biol. 1985, 47,
̈
146−159.
(44) Schoneich, C.; Pogocki, D.; Hug, G. L.; Bobrowski, K. J. Am.
̈
Chem. Soc. 2003, 125, 13700−13713.
(45) Merenyi, G.; Lind, J.; Engman, L. J. Phys. Chem. 1996, 100,
8875−8881.
(46) Harriman, A. J. Phys. Chem. 1987, 91, 6102−6104.
(47) Wojcik, A.; Lukaszewicz, A.; Brede, O.; Marciniak, B. J.
Photochem. Photobiol. A: Chem. 2008, 198, 111−118.
(48) Bobrowski, K.; Lubis, R. Int. J. Radiat. Biol. 1986, 50, 1039−
1050.
(49) Mozziconacci, O. In One-electron transfers in the enkephalins;
Universite De Paris-Sud: Orsay, 2007.
(50) Bobrowski, K.; Schoneich, C.; Holcman, J.; Asmus, K.-D. J.
̈
Chem. Soc., Perkin Trans. 2 1991, 353−362.
(51) Berges
261, 012003.
̀ ́
, J.; Trouillas, P.; Houee-Levin, C. J. Phys. Conf. Ser. 2011,
(52) Bobrowski, K.; Wierzchowski, K. L.; Holcman, J.; Ciurak, M. Int.
J. Radiat. Biol. 1990, 57, 919−932.
(53) Kciuk, G. In Influence of functional groups of aminoacids on the
mechanisms of radical reactions in peptides containing tyrosine; Institute
of Nuclear Chemistry and Technology: Warsaw, 2011.
(54) Bobrowski, K.; Wierzchowski, K. L.; Holcman, J.; Ciurak, M. Int.
J. Radiat. Biol. 1992, 62, 507−516.
(55) Ostuni, M. A.; Gelinotte, M.; Bizouarn, T.; Baciou, L.; Houee
́
-
Levin, C. Free Radic. Biol. Med. 2010, 49, 900−907.
12472
dx.doi.org/10.1021/jp307043q | J. Phys. Chem. B 2012, 116, 12460−12472