X-ray and solution structures of CuIIGHK and CuIIDAHK complexes: Influence on their redox properties

Article abstract of DOI:10.1002/chem.201100751

The Gly-His-Lys (GHK) peptide and the Asp-Ala-His-Lys (DAHK) sequences are naturally occurring high-affinity copper(II) chelators found in the blood plasma and are hence of biological interest. A structural study of the copper complexes of these peptides was conducted in the solid state and in solution by determining their X-ray structures, and by using a large range of spectroscopies, including EPR and HYSCORE (hyperfine sub-level correlation), X-ray absorption and ¹H and ¹³C NMR spectroscopy. The results indicate that the structures of [Cu^II(DAHK)] in the solid state and in solution are similar and confirm the equatorial coordination sphere of NH₂, two amidyl N and one imidazole N. Additionally, a water molecule is bound apically to Cu^II as revealed by the X-ray structure. As reported previously in the literature, [Cu^II(GHK)], which exhibits a dimeric structure in the solid state, forms a monomeric complex in solution with three nitrogen ligands: NH₂, amidyl and imidazole. The fourth equatorial site is occupied by a labile oxygen atom from a carboxylate ligand in the solid state. We probe that fourth position and study ternary complexes of [Cu^II(GHK)] with glycine or histidine. The Cu^II exchange reaction between different DAHK peptides is very slow, in contrast to [Cu^II(GHK)], in which the fast exchange was attributed to the presence of a [Cu^II(GHK)₂] complex. The redox properties of [Cu^II(GHK)] and [Cu^II(DAHK)] were investigated by cyclic voltammetry and by measuring the ascorbate oxidation in the presence of molecular oxygen. The measurements indicate that both Cu ^II complexes are inert under moderate redox potentials. In contrast to [Cu^II(DAHK)], [Cu^II(GHK)] could be reduced to Cu ^I around -0.62 V (versus AgCl/Ag) with subsequent release of the Cu ion. These complete analyses of structure and redox activity of those complexes gave new insights with biological impact and can serve as models for other more complicated Cu^II-peptide interactions. Copyright

Full text of DOI:10.1002/chem.201100751

FULL PAPER
DOI: 10.1002/chem.201100751
X-ray and Solution Structures of CuIIGHK and CuIIDAHK Complexes:
Influence on Their Redox Properties**
Christelle Hureau,*[b] Hꢀlꢁne Eury,[b] Rꢀgis Guillot,[c] Christian Bijani,[b]
Stꢀphanie Sayen,[d] Pier-Lorenzo Solari,[e] Emmanuel Guillon,[d] Peter Faller,[b] and
Pierre Dorlet*[a]
Abstract: The Gly-His-Lys (GHK)
peptide and the Asp-Ala-His-Lys
(DAHK) sequences are naturally oc-
curring high-affinity copper(II) chela-
tors found in the blood plasma and are
hence of biological interest. A structur-
al study of the copper complexes of
these peptides was conducted in the
solid state and in solution by determin-
ing their X-ray structures, and by using
a large range of spectroscopies, includ-
ing EPR and HYSCORE (hyperfine
sub-level correlation), X-ray absorption
and 1H and 13C NMR spectroscopy.
The results indicate that the structures
bound apically to CuII as revealed by
the X-ray structure. As reported previ-
ACHTUNGTRENNUNG
ously in the literature, [CuII
(GHK)],
ACHTUNGTRENNUNG
which exhibits a dimeric structure in
the solid state, forms a monomeric
complex in solution with three nitrogen
ligands: NH2, amidyl and imidazole.
The fourth equatorial site is occupied
by a labile oxygen atom from a carbox-
ylate ligand in the solid state. We
probe that fourth position and study
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
voltammetry and by measuring the as-
corbate oxidation in the presence of
molecular oxygen. The measurements
indicate that both CuII complexes are
inert under moderate redox potentials.
In contrast to [CuII
ACHTUNGTRENNUNG
(DAHK)], [CuII-
ACHTUNGTRENNUNG
ternary complexes of [CuII
N
around ꢀ0.62 V (versus AgCl/Ag) with
subsequent release of the Cu ion.
These complete analyses of structure
and redox activity of those complexes
gave new insights with biological
impact and can serve as models for
other more complicated CuII–peptide
interactions.
of [CuII
ACHTUNGTRENNUNG(DAHK)] in the solid state and
in solution are similar and confirm the
equatorial coordination sphere of NH2,
two amidyl N and one imidazole N.
Keywords: bioinorganic chemistry ·
copper · ligand interactions · pep-
tides · structural characterisation
Additionally,
a
water molecule is
Introduction
generative disorders such as Alzheimerꢀs or Prion diseases.
In the latter, the redox activity of copper seems to play a
crucial role.[1] The Gly-His-Lys (GHK) peptide and the Asp-
Ala-His-Lys (DAHK) sequence are naturally occurring
high-affinity copper chelators found in the blood plasma.[2]
GHK was isolated by Pickart et al.[3] and first described as a
Copper is an essential trace element involved in the activi-
ties of many enzymes and proteins but can become toxic if
its homeostasis is not tightly controlled. Pathologies related
to copper include Menkes and Wilson diseases and neurode-
[a] Dr. P. Dorlet
[d] Dr. S. Sayen, Prof. E. Guillon
Groupe de Chimie de Coordination, CNRS
ICMR—UMR 6229
Laboratoire du Stress Oxydant et Dꢁtoxication
CNRS, URA 2096, 91191 Gif-sur-Yvette (France)
and
Universitꢁ de Reims Champagne-Ardenne
51687 Reims Cedex 2 (France)
CEA, iBiTec-S, SB2SM, 91191 Gif-sur-Yvette (France)
Fax : (+33)1-69-08-87-17
[e] Dr. P.-L. Solari
E-mail: pierre.dorlet@cea.fr
Synchrotron SOLEIL, Saint-Aubin
91192 Gif-Sur-Yvette Cedex (France)
[b] Dr. C. Hureau, H. Eury, Dr. C. Bijani, Prof. P. Faller
CNRS, LCC (Laboratoire de Chimie de Coordination)
205, route de Narbonne, 31077 Toulouse (France)
and
[**] GHK and DAHK refers to the peptide chains Gly-His-Lys and Asp-
Ala-His-Lys, respectively.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100751.
Universitꢁ de Toulouse; UPS, INPT
LCC, 31077 Toulouse (France)
Fax : (+33)5-61-55-30-03
E-mail: hureau@lcc-toulouse.fr
[c] Dr. R. Guillot
CNRS, ICMMO—UMR 8182, Bꢂt. 420
Universitꢁ Paris-Sud 11, 91405 Orsay Cedex (France)
Chem. Eur. J. 2011, 17, 10151 – 10160
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10151

Table 1. Crystal data and structure refinement for [CuII
[CuII
(DAHK)].
ACHTUNGTRENN(UNG GHK)] and
liver cell growth factor. Further studies showed that GHK
possesses a wide range of biological activities, including ac-
celeration of wound healing[4] and tissue remodelling.[5]
AHCTUNGTRENNUNG
A
E
[CuII
ACHUTGTNRNEUNG AHCUTNGTREN(NUGN GHK)]
When complexed to this peptide, CuII is bound by the
formula
Mr [gmolꢀ1
T [K]
p
]
693.61
100(1)
550.92
100(1)
ꢀ
NH2, the N of the histidine residues and one amidyl
group from the backbone. The equatorial coordination
plane is completed in solution by a labile ligand, thus allow-
ing formation of ternary species with exogenous amino acid
residues in a biological medium.[6] DAHK is the N-terminal
fragment of the human serum albumin (HSA). HSA is the
most abundant blood serum protein involved in CuII trans-
port[7] and also a major component of cerebrospinal fluid.
Among the two CuII binding sites present in HSA, the
l [ꢄ]
0.71073
0.18ꢅ0.06ꢅ0.01
purple
trigonal
P32
0.18ꢅ0.06ꢅ0.01
14.4934(3)
14.4934(3)
12.1728(5)
90
90
120
2214.43(11)
3
1.560
0.71073
0.32ꢅ0.27ꢅ0.21
blue
tetragonal
P41212
0.32ꢅ0.27ꢅ0.21
14.4793(2)
14.4793(2)
26.3306(7)
90
crystal size [mm3]
crystal colour
crystal system
space group
unit cell dimensions
a [ꢄ]
b [ꢄ]
c [ꢄ]
a [8]
b [8]
N-terminal motif possesses the higher affinity.[8] CuII is
90
90
p
ꢀ
bound to the NH2, the N of the histidine residues and two
g [8]
amidyl groups from the peptide backbone, thus being arche-
V [ꢄ3]
5520.21(18)
8
1.326
ꢀ
typal of the so-called ATCUN binding motif (H2N XXH
Z
1calcd [Mgmꢀ3
]
beginning sequence in a peptide/protein) also found in other
proteins, that is, neuromedins C and K, human sperm prota-
mine P2a and histatins.[2a] DAHK has also recently been
shown to restore cell survival of cells exposed to Cu and
Cu-Ab (Ab is the amyloid-b peptide involved in Alzheim-
erꢀs disease) and this effect has been attributed to redox si-
lencing of Cu when bound to DAHK.
absorption coefficient 0.907
[mmꢀ1
(000)
0.856
]
F
A
1095
8978
5001 (0.0631)
2272
115905
8484 (0.0391)
7094
R1=0.0666,
wR2=0.1899
R1=0.0854,
wR2=0.2267
0.022(19)
1.060
reflns collected
indep reflns (Rint
obsd reflns (I>2s(I)) 7423
final R indices
(I>2s(I))
)
R1=0.0457,
wR2=0.1017
R1=0.0645,
wR2=0.1110
0.005(10)
1.069
0.000
0.981/ꢀ0.728
As a consequence, CuII binding to these two peptides is of
paramount importance. In addition, the present study was
also motivated by the need to obtain a set of reference spec-
troscopic signatures coupled to redox behaviour to model in
detail CuII sites with 3N+1O or 4N equatorial binding
modes in other peptides and proteins. In this paper, we give
new insights into the CuII coordination to the DAHK and
GHK peptides and discuss the results in comparison with
studies previously performed on these species. The X-ray
structures of both complexes are described and their solu-
tion structures are characterised by advanced spectroscopic
methods. The redox properties are then analysed and de-
tailed in relation to the structural observations.
R indices (all data)
Flack parameter
S
T
0.009
0.780/ꢀ0.137
]
Table 2. Selected bond lengths [ꢄ] and angles [8].
(DAHK)] (100 K) ACTHNGUETRNNU(G GHK)] (100 K)
[CuII [CuII
T
E
R
ꢀ
ꢀ
Cu N(1)
2.029(3)
1.906(3)
1.967(2)
1.969(3)
2.565(3)
83.70(11)
166.74(10)
98.81(11)
90.28(11)
83.11(11)
173.55(10)
89.98(11)
94.45(11)
88.34(11)
95.89(11)
1
Cu N(1)
Cu N(2)
Cu N(3)
Cu O(3)
Cu O(3)
N(1)-Cu-N(2)
1.957(4)
1.945(3)
2.003(3)
1.967(2)
2.513(3)
93.58(15)
172.79(16)
90.40(15)
98.50(14)
83.53(12)
175.82(12)
99.34(12)
92.67(12)
88.51(15)
78.75(16)
1
ꢀ
ꢀ
Cu N(2)
ꢀ
ꢀ
Cu N(3)
[a]
[b]
ꢀ
ꢀ
Cu N(4)
ꢀ
ꢀ
Cu O(8)
N(1)-Cu-N(2)
N(1)-Cu-N(3)
N(1)-Cu-N(4)
N(1)-Cu-O(8)
N(2)-Cu-N(3)
N(2)-Cu-N(4)
N(2)-Cu-O(8)
N(3)-Cu-N(4)
N(3)-Cu-O(8)
N(4)-Cu-O(8)
Results and Discussion
N(1)-Cu-N(3)
N(1)-Cu-O(3)[a]
N(1)-Cu-O(3)[b]
N(2)-Cu-N(3)
X-ray structures: The crystal structures of the [CuII
and [CuII
(DAHK)] complexes were determined by single-
ACHTUNGTRNE(NUNG GHK)]
ACHTUNGTRENNUNG
N(2)-Cu-O(3)[a]
N(2)-Cu-O(3)[b]
N(3)-Cu-O(3)[a]
N(3)-Cu-O(3)[b]
O(3)[a]-Cu-O(3)[b]
crystal X-ray diffraction. Table 1 and Table 2 summarise
crystal data and selected bond lengths and angles. In both
cases, the CuII ion is penta-coordinated in a distorted
square-planar pyramid environment with an oxygen in the
apical position.
1
1
1
[a] Symmetry code: = ꢀx, y, = ꢀz. [b] Symmetry code: = ꢀy, = +x, = +
2
2
2
2
4
In the case of [CuII
ACTHUNTGRENNG(U GHK)] (see Figure 1), the structure is
z.
essentially identical to that previously reported by Perkins
and co-workers,[9] and close to that of [CuII(GH)].[10] The
binding mode in the equatorial plane is 3N+1O: the CuII
ion is ligated by the glycine amino nitrogen, the deprotonat-
ed amide nitrogen of the glycine–histidine peptide bond, the
Np nitrogen of the imidazole side chain of the histidine and
the lysine carboxylate of a neighbouring complex. The
apical oxygen is provided also by the lysine carboxylate of
another neighbouring complex. This is in contrast with the
structure of [CuII(GH)][10] for which the apical oxygen is a
water molecule. Two [CuII
ACTHNUTRGNE(NUG GHK)] monomeric units are thus
arranged in a dimeric diamond-core type structure in which
the two CuII centres are spaced by 3.458 ꢄ and the two
bridging oxygen atoms by 2.874 ꢄ. A total of 16 sites for a
10152
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10151 – 10160

CuIIGHK and CuIIDAHK Complexes
FULL PAPER
unit cell. The main difference with the X-ray structure of
CuII bound to GGH–NH2, the simplest peptide to form an
ATCUN motif, is the presence of an apical water only in
one out of two complexes in the [CuII
ACHTNUTRGNE(NUG GGH–NH2)] struc-
ture.[11] Contrary to what has been conjectured previous-
ly,[11,12] the carboxyl oxygen from Asp1 does not occupy the
apical position in the [CuII
ACTHNUTRGNE(NUG DAHK)] complex.
Electronic and ESI-MS data: In solution and at room tem-
perature, the characteristic blue and purple colours for
ACTHNUTRGNEUNG
Figure 1. X-ray structure of the [CuII(GHK)] complex. Each CuII ion is
[CuII(GHK)] and [CuII
ACHTNUTRGENNUG HCATUNGTRENN(UNG DAHK)] are retained (see Figure S1
penta-coordinated. Considering the ion on the right, the equatorial posi-
tions are occupied by atoms N(1) to N(3) and O(3)[a], whereas the apical
position is occupied by O(3)[b]. These last two oxygen atoms are shared
with the other CuII ion for which their position is interchanged (O(3)[a]
being in the apical position, whereas O(3)[b] is in the equatorial position
for the ion on the left).
in the Supporting Information for UV/Vis and circular di-
chroism data). The ESI-MS spectra (positive detection
mode) of [CuII(GHK)] and [CuII
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
water molecule were present in the unit cell of our structure
of [CuII
ACHTUNGTRENNUNG(GHK)] with an occupancy of 9.25 compared with 14
sites for the structure of Perkins et al. with an occupancy of
9.2.
The complex [CuII
ACTHNUTRGNEU(GN DAHK)] corresponds to the high-af-
finity binding site of human serum albumin and its crystal
structure is reported here for the first time (see Figure 2).
ACTHNUTRGNEUNG
Figure 2. X-ray structure of the [CuII(DAHK)] complex. The CuII ion is
penta-coordinated with four nitrogen ligands in equatorial positions
(N(1) to N(4)) and one water molecule in the apical position (O(8)).
Figure 3. ESI-MS spectra of [CuII
N
ACHTUNGTREN(NUGN DAHK)] complexes
The CuII ion is bound by 4 nitrogen ligands in the equatorial
plane: the aspartate amino group, two deprotonated amide
functions from the first two peptide bonds and the Np nitro-
gen of the imidazole side chain of the histidine residue. The
apical oxygen along the Jahn–Teller distortion axis is provid-
ed by a water molecule in hydrogen-bond interactions with
other solvent molecules in the unit cell. By contrast with the
ACHTUNGTRENNUNG
of CuII-containing species are recognizable by their isotopic
pattern (63Cu=65%, 65Cu=35%). The main peak at m/z
531.5 in the [CuII
ACHTUNGTRENNUNG
structure of [CuII
ACHTUNGTRENNUNG(GHK)], all protons could be resolved and
refined in this structure. The peptide is in its fully protonat-
ed form and one chloride counterion per complex is present.
In addition, the crystal structure of this complex is mono-
meric rather than the dinuclear motif observed for [CuII-
ACHTUNGTRENNUNG
firmed by the detection of a peak at m/z 529.5 in the nega-
tive detection mode (data not shown). Another peak detect-
ed at m/z 385.3 (30% of relative intensity) is attributed to
(GHK)]. A total of 7 water molecules are present in the
the [CuII
ACTHUNGTERNNU(G DAH)] species issued from the fragmentation of
Chem. Eur. J. 2011, 17, 10151 – 10160
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10153

P. Dorlet, C. Hureau et al.
the parent peak at m/z 531.5. Departure of the Lys residues
from the [CuII(DAHK)] complex is in line with a CuII bind-
ing site in which only the first three residues are involved.
The main peak detected in the [CuII
(GHK)] spectrum corre-
sponds to the monocationic [CuII
(GHK)] species. A peak at
m/z 803.5 that corresponds to the [Cu2A(GHK)2] species (Fig-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
CTHUNGTRENNUNG
ure S2) is observed but with a very low relative intensity
(less than 5%), thereby suggesting that the binuclear unit
observed in the solid state is mainly disrupted in solution.
Fragmentation peaks observed at m/z 372.1 (65% of relative
intensity) and 229.3 (22% of relative intensity) are attribut-
Figure 4. The XANES regions of the normalized absorption amplitude
ꢀ
ed to the departure of the [NH2 CH2] N-terminal motif and
versus energy for [CuII(DAHK)] and [CuII
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
the additional removal of the Lys residues from the parent
species and were confirmed by MS/MS experiments. Contra-
[CuII(GHK)]=50 mm and [CuII
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
ry to what was observed in the case of [CuII
ACTHNUGRTENUNG(DAHK)], the
main fragmentation peak observed corresponds to the re-
moval of the [NH2 CH2] N-terminal motif suggesting that
tion characteristic of the CuII valence state. The lower inten-
sity of the [CuII
(DAHK)] pre-edge feature indicates that the
CuII centre is in a more centro-symmetric environment than
in [CuII(GHK)].[13] The XANES signatures differ between
the two CuII species, as indicated by more intense B and D
peaks and less-pronounced A and C peaks for [CuII
(GHK)]
than [CuII
(DAHK)]. In related systems, these peaks have
ꢀ
the principal CuII anchor site is the His residue. Only in a
second time, the Lys residue is lost. The very low intensity
ACHTUNGTRENNUNG
of the peaks observed at m/z 470.1 ([CuII
N
ACHTUNGTRENNUNG
trum) and at m/z 341.1 ([CuII
ACHTUNGTRENNUNG
respond to the DAHK and GHK ligands, respectively,
ACHTUNGTRENNUNG
agrees with a CuII ion strongly bound by both peptides.
AHCTUNGTRENNUNG
been proposed to be sensitive to second- and third-shell per-
turbations.[14] More precisely, an increase of the B peak and
a decrease of the C peak were related to a more important
contribution of external shells. In the present case, the more
X-ray absorption spectroscopy (XAS): To gain further in-
sight into CuII coordination in solution, X-ray absorption
spectra were analysed in the near-edge structure (XANES)
and in the extended region (EXAFS). EXAFS data (Fig-
ure S3 in the Supporting Information) are best fitted with
intense B peak (or less intense C peak) for [CuII
ative to [CuII
having an exchangeable fourth equatorial ligand in the
[CuII
(GHK)] species, also consistent with a greater Debye–
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
five N/O neighbours ([CuII
six N/O neighbours ([CuII
(DAHK)]), four of them being sig-
nificantly closer (Table 3). In the case of [CuII
(DAHK)], it is
ACHTUNGTNER(NUNG GHK)]) and with either five or
G
ACHTUNGTRENNUNG
A
Waller factor. These results show that the square pyramidal
structure observed in the solid state for the two complexes
is retained in solution.
difficult to determine precisely whether one or two axial
oxygen atoms are involved in the copper coordination in so-
lution, given both fits were very similar as shown by the
very close value of the goodness-of-fit (0.44 and 0.53% for
(4+2) and (4+1) neighbours, respectively).
EPR spectroscopy: EPR spectra (9 GHz) were recorded for
[CuII(GHK)] and [CuII
ACHTNUTRGENNUG HCATUNGTRENN(NGU DAHK)] complexes in solution and
The EXAFS data are thus consistent with the X-ray struc-
tures. Qualitative XANES analysis agrees with geometrical
are shown in Figure 5. Characteristic parameters are listed
in Table 4. Both complexes display a classical mononuclear
square-planar-type EPR spectrum with some resolved super-
hyperfine interaction in the g? region, as previously report-
ed.[15] The gk and Ak values measured on the spectra are
consistent with the 3N+1O and 4N binding mode, respec-
tively.[16] These binding modes are thus retained from the
ACTHNUTRGNEUNG
data determined by EXAFS. The [CuII(DAHK)] and [CuII-
ACHTUNGTRENNUNG(GHK)] XANES data (Figure 4) show the presence of the
pre-edge P peak at 8979.7 eV, assigned to the 1s!3d transi-
Table 3. First coordination shell structural data obtained from R space
fits of EXAFS spectra: N is the number of neighbours, R is the absorb-
er–neighbour distance, s is the Debye–Waller factor; uncertainties are es-
timated in coordination numbers to ꢁ10%, in R to ꢁ0.02 ꢄ, and to
solid state to solution. In the case of [CuII
ACTHNUTRGNE(NUG GHK)], however,
the dinuclear structure observed in the crystal state is not
present in solution.[15a] Therefore it is likely that the equato-
rial oxygen ligand, provided by the C-terminus carboxylate
ꢁ0.001 ꢄ2 in s2. For [CuII
ACHTUNGTRENUN(NG DAHK)], results of fits with 4+2 neighbours
and 4+1 neighbours as starting parameters are given.
Scattered/Backscat-
tered
N
R
[ꢄ]
s2
[ꢄ2]
R factor
Table 4. EPR parameters obtained from the spectra of the different com-
plexes in Figure 5.
[%][a]
ꢀ
G
E
Cu N/O
4.49 1.97 0.0057 0.12
0.79 2.61 0.01[b]
gk
g?
Ak (63Cu) [MHz]
ꢀ
Cu O
[CuII
N
(DAHK)] Cu N/O
4.49 1.95 0.0046 0.44
A
(DAHK)]
2.19
2.23
2.22
2.22
2.21
2.04
2.05
2.05
2.05
2.05
596
560
595
568
590
ꢀ
Cu O
1.52 2.50 0.01[b]
R
ꢀ
G
ꢀ
T
(DAHK)] Cu N/O
4.42 1.95 0.0045 0.53
1.49 2.49 0.01[b]
N
ꢀ
G
Cu O
A
[a] R factor represents the overall goodness-of-fit. [b] Fixed parameter.
10154
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10151 – 10160

CuIIGHK and CuIIDAHK Complexes
FULL PAPER
Figure 6. Six-pulse HYSCORE contour plot recorded for [CuII
(13COO)] (left panel) and [CuII (13Ca)] (right panel). The labels
(GHK)G
ACHTUNGTRENNUNG(GHK)G-
A
R
ACHTUNGTRENNUNG
for the cross-peaks are indicated on the left panel. Experimental condi-
tions: microwave frequency 9.71 GHz, B=337 mT, t1 =104 ns, t1 =136 ns,
T=4.2 K. Samples prepared in HEPES buffer 100 mm, pH 7.4.
Figure 5. cw EPR spectra recorded for different Cu complexes. The right
panel is an enlargement of the g? region for some spectra to compare
their superhyperfine patterns. Experimental conditions: microwave fre-
quency 9.42 GHz, microwave power 0.13 mW, modulation amplitude
0.5 mT, modulation frequency 100 kHz, time constant 164 ms, T=50 K.
Samples prepared in HEPES buffer 100 mm, pH 7.4.
pattern consistent with a 4N binding mode for [CuII-
AHCTUNGTREG(NNNU GHK)H]. This latter pattern is affected when histidine la-
belled with 15N only on the imidazole ring is used to prepare
the complex, thereby indicating that the additional histidine
binds to the Cu ion through the imidazole ring rather than
its N terminus. This is confirmed by HYSCORE measure-
ments performed on the labelled sample for which a charac-
teristic 15N cross-peak was detected (see Figure S4 in the
Supporting Information).
of a neighbouring molecule in the crystal, is replaced by a
solvent molecule in solution, as previously proposed,[15a] or
by an acetate anion present from the peptide batch, sold as
an acetate salt. This is in line with the XANES results,
which propose an exchangeable equatorial ligand. This
labile position has been investigated by addition of either
glycine or histidine to the [CuII
The EPR spectrum obtained for [CuII
slightly from that of [CuII
(GHK)], particularly in the gk
G
In the case of the [CuII
to that observed for the [CuII
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
N
N
ed in the superhyperfine coupling pattern as well as in the
gk and Ak parameters. This indicates that the second GHK
ligand should bind the CuII centre through its imidazole
ring.
region. The superhyperfine structure on the g? region is
also affected. To better probe the coordination mode of the
additional glycine amino acid, we have used different isotop-
ically labelled glycine (15N on the amine group or 13C either
on the carboxylate group or on the Ca). The superhyperfine
structure observed for the 15N-labelled glycine differs suffi-
ciently from that of the unlabelled sample to support bind-
ing through the amine function. The weak 13C couplings
were probed with pulse EPR. Figure 6 shows the 6-pulse hy-
perfine sub-level correlation (HYSCORE) contour plots ob-
tained for both carboxylate and Ca-labelled samples. This
technique is best adapted in this type of multinuclear cou-
pled systems and was chosen in particular to avoid intensity
suppression effects.[17] Ridges centred at the 13C Larmor fre-
quency were detected in both samples. The signal was more
intense and the coupling stronger for the Ca-labelled sample
relative to the carboxylate-labelled sample, thereby support-
ing the fact that the additional glycine binds predominantly
by its amine function in equatorial position (we note that
based on these data we cannot fully exclude that a fraction
of the additional glycine binds through its carboxylate func-
tion). This binding mode is in agreement with the observa-
tions made on the continuous-wave (cw) EPR spectra.
NMR spectroscopy: In conjunction with other spectroscopic
techniques, NMR spectroscopy can yield information about
both CuII coordination to peptides and the dynamic/mecha-
nism of CuII exchange between peptides. Due to its para-
magnetism, CuII induces the broadening of the binding
groups in its close vicinity.[18] The impact of CuII on the spec-
trum will vary considerably depending on the CuII exchange
rate (between holo and apo forms) relative to the timescale
of the NMR spectroscopic experiment performed. The two
examples studied here are representative of extreme cases:
the CuII exchange is very slow when bound to DAHK and
fast when bound to GHK. 13C spectra of the GHK and
DAHK peptides in presence of 0.1 and 0.5 equiv of CuII, re-
spectively, are shown in Figure 7 and Figure 8. The corre-
1
sponding H spectra are shown in the Supporting Informa-
tion. Based on these data, a mapping of the residues most
affected is proposed in Scheme 1.
In the case of the [CuII
ACHTUNGTRENNUNG
In the case of the [CuII
N
whereas in the [CuII
ACTHUNTGRENNG(U GHK)] complex, despite being record-
change was observed in the superhyperfine coupling pattern.
ed in the presence of a fivefold lesser CuII content, almost
all residues are strongly broadened with the exception of
the Lys side chain. This suggests that in the case of DAHK,
The seven superhyperfine lines, consistent with a 3N+1O
binding mode for [CuII
ACTHNUTRGNEU(GN GHK)], were replaced by a nine-line
Chem. Eur. J. 2011, 17, 10151 – 10160
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10155

P. Dorlet, C. Hureau et al.
Figure 7. 13C{1H} NMR spectra of 10 mm GHK peptide in D2O in the ab-
sence (bottom spectrum) or in the presence of 0.1 equiv CuII (top spec-
trum) at pH 7.2. Note that the shift of some peaks is due to a slight modi-
fication of the pH value after CuII addition. T=298 K, n=125.8 MHz.
Scheme 1. Schematic representation of the C and H atoms for which the
NMR spectroscopic signal is broadened upon addition of CuII to GHK
(top) and DAHK (bottom) at pH 7.2. Black=highly broadened, dark
grey=broadened, grey=moderately broadened, light grey=slightly
broadened.
Figure S7 in the Supporting Information and ref. [2b]), with
the second GHK occupying the fourth equatorial position
through the imidazole ring of the His side chain (see the
EPR spectrum in Figure 5 and ref. [21]). Formation of such
a ternary complex will help Cu exchange from one peptide
to another. As previously described,[14e] CuII broadening of
peptide residues can also originate from involvement in Cu
transfer (even if the Cu transfer is very slow, as in the case
of DAHK). This is one possible reason why Asp and His lat-
eral chains and the C- and N-terminal of DAHK are the res-
idues mostly affected. The second possibility is that these
four functions (from the apo peptide) may be involved in
chemical exchange for CuII apical binding (in the holo pep-
tide). Related to that, the carboxylate groups from Asp1 or
C-terminal are equivalently broadened, which strongly sug-
gests that these functions are involved in Cu transfer but not
in direct Cu binding in solution. Hence, this data is in disa-
greement with the previously proposed coordination of
Asp1 carboxylate function in the CuII apical position.[12]
Figure 8. 13C{1H} NMR spectra of 10 mm DAHK peptide in D2O in the
absence (bottom spectrum) or in the presence of 0.5 equiv CuII (top spec-
trum) at pH 7.2. Note that the shift of some peaks is due to slight modifi-
cation of the pH value after CuII addition. T=298 K, n=125.8 MHz.
the CuII exchange is very slow and that the peaks observed
belong to the apo form of the peptide, with those of the CuII
complex being fully broadened and thus difficult to detect,
which is in line with what was proposed on longer HSA N-
terminal fragments.[19] This hypothesis is confirmed by quan-
Electrochemistry: Cyclic voltammetry (CV) traces of [CuII-
1
tification of the H NMR spectroscopic signal, which shows
(GHK)] and [CuII
ACHUTNGTRENNUG AHCTUNGTRENN(GUN DAHK)] complexes are shown in
that in presence of 0.5 equiv of CuII, the intensity of the
signal recorded is about half that of the apopeptide (Fig-
ure S6 in the Supporting Information). Knowing that the
CuII binding affinities for both peptides are close,[2b,20] this
result was unanticipated. Actually, the reason for such a dis-
crepancy may be connected to the fact that, in a substoichio-
Figure 9. The main difference with previous data published
in the literature is the nature of the working electrode.[22] In
the present study, we used vitreous carbon electrode, which
has been shown to be a well-suited surface for the study of
CuII–peptide complexes.[23]
The CV trace of the [CuII
cathodic process and a reversible anodic process at E =
ACHTUNGNRTE(UNG DAHK)] complex shows no
1
=
2
metric CuII ratio, GHK can form [CuII
ACTHNUTRGNE(UNG GHK)2] species (see
10156
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10151 – 10160

CuIIGHK and CuIIDAHK Complexes
FULL PAPER
Figure 9. Cyclovoltammogram recorded for [CuII], [CuII
(GHK)2], [CuII(GHK)H] (left panel) and for [CuII
(DAHK)] (right
panel). Experimental conditions: scan rate 100 mVsꢀ1, T=298 K, 0.1m
50 mm, Tris buffer pH 7.4, 0.1m NaClO4, [CuII(DAHK)]=2 mm; [CuII-
(GHK)]=1.6 mm.
(GHK)], [CuII-
ACHTUNGTRENNUNG
A
R
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
0.77 V versus AgCl/Ag (DEp =130 mV) that corresponds to
the reversible oxidation of the [CuII
(DAHK)] complex into
[CuIII
Scheme 2. Electrochemical processes associated with the reduction of
A) [CuII], B) of the [CuII(GHK)] complex and C) of the [CuII
ACTHNUTRGENNUG ACHTUNGTRENNUNG(GHK)]
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
(DAHK)]+, which is in line with what was previously
1
2
complex in presence of an imidazole (Im) donor species. Potentials refer
to Figure 9.
observed for related complexes.[24] More precisely, the E =
oxidation process abides by the following trend with respect
to the equatorial binding plane: {NH2, 3Nꢀ}[24a] (0.41 V
versus AgCl/Ag)<{Im
Ag)<{NH2, Im
(His), 2Nꢀ} (0.77 V versus AgCl/Ag).
The CV trace of the [CuII
(His), 3Nꢀ}[24b] (0.66 V versus AgCl/
tected in the CV trace of the [CuII
marised in Scheme 2B.
ACHTUNGERTN(NUNG GHK)] complex are sum-
ACHTUNGTRENNUNG
A
In the presence of an excess amount of GHK or His, the
cathodic peak is not modified significantly, which means
that the nature (N or O) of the fourth ligand has little
impact on the reduction process. On the contrary, the de-
sorption peak (p2 in Figure 9) is greatly reduced and the ox-
idation peak of CuI to CuII (Ep2’) is slightly shifted. This is in
line with the coordination of CuI by the imidazole (Im)
from the side chain of GHK or His. We propose that in the
presence of an excess amount of GHK or His, the CuI re-
anodic process and an irreversible cathodic process at Ep1 =
ꢀ0.49 V versus AgCl/Ag that corresponds to the reduction
of the [CuII
ACHTUNGTRENNUNG(GHK)] complex. This is in drastic contrast to
the CV trace obtained for CuII with no peptide in the same
buffer, which shows a reduction of CuII to CuI at Ep0 =
ꢀ0.07 V versus AgCl/Ag followed by reduction of CuI to
Cu0 at Ep1 =ꢀ0.54 V versus AgCl/Ag (redox processes are
summarised in Scheme 2A). On the reverse scan for [CuII-
A
lease after reduction of the [CuII
ACTHNGUTERNN(UG GHK)] species is coordi-
versus AgCl/Ag. Its shape and position are characteristic of
the oxidation–solubilisation process of adsorbed Cu0 that
nated by the excess ligand, thus leading to the formation of
a new CuI complex. This new CuI complex is oxidised at
Ep2’ =0.04 V versus AgCl/Ag (in the case of an excess
amount of GHK) and Ep2’ =ꢀ0.01 V versus AgCl/Ag (in the
case of an excess amount of His). In this latter case, we pro-
pose that the motif Im-CuI-Im, which is quite usual in CuI
coordination,[26a,b] is formed (illustrated in Scheme 2C). In
the former case, the GHK ligand may remain bound by
lead to CuI. A second anodic peak is detected at Ep2’
0.02 V versus AgCl/Ag and corresponds to oxidation of CuI
to CuII. This indicates that reduction of [CuII
lowed by decoordination of the peptide ligand. Due to the
difference in Lewis acidity of CuI and CuII, protonation of
the amidyl function of the Gly–His peptide bond involved in
the CuII binding is likely concomitant with the reduction
ꢀ
both the terminal NH2 and the His.
process. At the [CuII
(GHK)] reduction potential (ꢀ0.54 V
Note that we also performed the same series of experi-
ments in phosphate buffer (see Figure S8 and Scheme S2 in
the Supporting Information). In the latter case, it is not pos-
sible to record the CV trace of unbound CuII because of its
rapid precipitation in phosphate buffer. However, in the
presence of GHK or DAHK peptide, the results obtained in
phosphate buffer were fully consistent with those obtained
versus AgCl/Ag), unbound CuI is reduced to Cu0[25] and ad-
sorbed onto the electrode. When scanning back towards
cathodic potential, the reduction of unbound CuII to CuI
(expected near Ep0 =ꢀ0.07 V versus AgCl/Ag) is not ob-
served, thus indicating that formation of the [CuII
ACTHNUTRGNE(NUG GHK)]
complex is faster than 1 s. The different redox processes de-
Chem. Eur. J. 2011, 17, 10151 – 10160
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10157

P. Dorlet, C. Hureau et al.
in tris(hydroxymethyl)aminomethane (Tris) buffer except
for small modifications in peak potentials. This suggests
that, even if the tris buffer interferes with CuII binding to
DAHK and GHK peptides, this doesn’t impact significantly
the electrochemical signatures.
new light on CuII exchange dynamics and redox behaviour
in particular. This is important in the biological context.
DAHK is the high-affinity CuII binding site of human serum
albumin; this region is 5% occupied with CuII. This is the
first time this copper-loaded sequence has been crystallised.
It has been suggested that albumin is involved in CuII trans-
port in the blood towards other targets. The slow exchange
rate of CuII in DAHK indicates that albumin is not a fast
CuII supplier. The observation of an axial water molecule in-
dicates that also other ligands can bind on this labile posi-
tion, which might be the reason why CuII is not completely
inert and can be exchanged, albeit very slowly. Thus differ-
ences in the coordination of the apical position might ex-
plain differences seen between the reactivity of [CuII-
Reactive oxygen species (ROS) production: The cyclic vol-
tammetry data of the [CuII
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
unique reversible oxidation process at 0.72 V versus AgCl/
Ag (but no reduction process); hence these data suggest the
impossibility for both complexes to be directly reduced by
ascorbate (Asc). This is why catalytic oxidation of Asc
under aerobic conditions by both complexes was compared
to oxidation by unbound CuII. Consumption of Asc was
monitored by the decrease of its absorption band at 265 nm.
Drastic deceleration of Asc consumption was observed in
the presence of both peptides (Figure S9 in the Supporting
Information), which is in line with a redox silencing of CuII
when bound to the peptides and with the redox potential
AHCTUNGTRENNUNG
(DAHK)] and Cu–albumin.[8,28] Although the affinity of
GHK for CuII is on the same order of magnitude as that of
DAHK,[21] it exchanges much faster and hence could serve
as a fast CuII supplier. The structural reason for that is that
the fourth equatorial binding site on CuII is available to bind
a transfer partner. Another way to transfer CuII from
DAHK and GHK is by reducing CuII to CuI, which might
occur in the reducing intracellular environment. The CV re-
sults indicate that CuII in DAHK is stable and not reduced
to CuI. By contrast, it is possible in the case of GHK with a
concomitant release of CuI. Thus CuII in GHK could be re-
leased and transferred to other targets by reduction (with
reductants stronger than Asc), as it could be expected intra-
cellularly.
values of the [CuII(GHK)] and [CuII
ACHTUNGTRENNUNG ACHTUNTGREN(NUGN DAHK)] complexes
and that of Asc (ꢂ0.1 V versus AgCl/Ag).
Previous results showed a propensity of the [CuII
ACTHNUTRGNE(NUG GHK)]
C
complex to catalyze the HO formation using Asc as reduc-
tant.[22b] As HO production from dioxygen requires the
C
redox cycling of the copper complex (Scheme S3 in the Sup-
porting Information), and as our data show that Asc is
unable to reduce [CuII
ACHTUNGTRENNUNG
C
coumarin-3-carboxylic acid. We found very weak HO pro-
Experimental Section
C
duction (in the range of residual HO production in the ab-
sence of added CuII ion) for both [CuII(GHK)] and [CuII-
AHCTUNGTRENNUNG
Chemicals: All solutions were prepared with Milli-Q (18 MW) water.
GHK peptide was bought from Bachem (Switzerland); DAHK peptide
was bought from Bachem (Switzerland) or GeneCust (Dudelange, Lux-
embourg). pH was controlled using a 744 pH meter equipped with a bio-
trode electrode (Metrohm SA, Switzerland). Stock solutions of peptides
were prepared by dissolving the peptides in D2O (ꢂ50 mg in 1 mL) and
concentrations were determined by titration followed by UV/Vis absorp-
tion spectroscopy (using an Agilent 8453 spectrometer at 258C in 1 cm
path-length quartz cuvette). In a typical experiment, the stock solution
was diluted tenfold and amounts of a CuII solution of known concentra-
tion were added until no increase in the d–d band of the CuII–peptide
complex and turbidimetry due to unbound CuII precipitation in the
buffer were observed. Measurements were performed in 0.1m phosphate
buffer (pH 7.4). This experimental determination led to concentration
values that are 10–20% off compared to those obtained using the molec-
ular mass of the peptide and its counterions, thereby suggesting that
counterion salts co-precipitate during peptide synthesis. Stock solutions
were stored at ꢀ208C. Unless specified, the CuII ion source was hydrated
C
ACHTUNGTRENNUNG(DAHK)] complexes relative to the HO production in simi-
lar concentrations of unbound CuII (Figure S10 in the Sup-
porting Information). Similar results were obtained with
EPR using the a-(4-pyridyl-1-oxide)-N-tert-butylnitrone
(POBN) spin trap (data not shown). Error in the previous
results[22b] with regards to the [CuII
ACTHNUTRGNEUNG(GHK)] complex is at-
tributed either to 1) a miscalculation of the GHK concentra-
tion (see the Experimental Section), thus leading to un-
bound CuII in solution or 2) an inversion with data of the
C [22b]
[CuII
a result further questioned.[27]
ACHTUNGTRENNUNG(Gly)2] species, which was found to produce no HO ,
Conclusion
CuACHTNUGTRENNUG ACHTUNGTRENNUNG(NO3)2
(SO4). For EPR measurements, a 0.1m stock solution of 63Cu
obtained by nitric acid treatment of isotopically pure 63Cu (Eurisotop,
Saclay, France) was used and 10% glycerol was added to the samples.
[CuACHTUGNRTEN(NNUG GHK)2], [CuACHTUNGTREG(NNUN GHK)H] and [CuACHTUNGTREN(NUGN GHK)G] were prepared by adding
In the present work we gained new insight into the structur-
al aspects of [CuII(GHK)] and [CuII
(DAHK)] and reported
A
ACHTUNGTRENNUNG
GHK (1 equiv), His (1 equiv) and Gly (10 equiv) to the [Cu
plex, respectively.
ACHTUNGTRENUN(NG GHK)] com-
data on their redox behaviour. The use of several spectro-
scopic techniques together with the determination of the
crystal structure gave a complete set of information, which
will be important as reference in the analysis of other, more
complicated CuII–peptide interactions with peptides such as
amyloid-b or a-synuclein sequences, for example, for which
no crystal structure is available. Moreover, the analysis shed
For ascorbate, a stock solution (20 mm) of ascorbate was prepared in
milli-Q water at room temperature just before beginning the experiment
and was used immediately. Because ascorbate degrades very quickly, a
new solution was prepared for each experiment.
A stock solution (1 mm) of desferrioxamine (DFO, Sigma) was prepared
by dissolving the appropriate mass in milli-Q water.
10158
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10151 – 10160

CuIIGHK and CuIIDAHK Complexes
FULL PAPER
A stock solution of coumarin-3-carboxylic acid (5 mm) was prepared in
phosphate (20 mm), NaCl (100 mm) buffer at pH 9 at room temperature.
The stock solution was stored at ꢀ208C.
chemical shifts are relative to tetramethylsilane. 1D NMR and 2D NMR
spectra were collected at 298 K in pure D2O.
Cyclic voltammetry (CV): CV measurements were recorded under argon
using a 620C electrochemical analyzer (CH Instruments, Inc). The work-
ing electrode was a glassy carbon disk and a Pt wire was used as counter-
electrode. The reference electrode was an AgCl/Ag electrode (0.223 V
versus NHE) isolated in a fritted bridge. Immediately before the mea-
surement of each voltammogram, the working electrode was carefully
polished with alumina suspensions (1, 0.3 and 0.05 mm, successively), so-
nicated in an ethanol bath and then washed carefully with ethanol. The
electrochemical cell medium used was milli-Q water with 0.1m sodium
phosphate (pH 7.4) added as supporting electrolyte.
Measurements of HO· production: These were performed in phosphate
(20 mm), NaCl (100 mm) buffer at pH 7.4. The reaction was started by
the addition of ascorbate (500 mm). In all these experiments, DFO (at a
final concentration of 1 or 2 mm depending on buffer concentration) was
15N-labelled glycine, 13C-(COOꢀ)-labelled glycine and 15N(Im)-labelled
histidine were purchased from Eurisotop (Saclay, France).
X-ray diffraction: Crystals suitable for X-ray diffraction were obtained
by slow ethanol diffusion and evaporation of a 0.2m solution in milli-Q
water of the corresponding complex. Data were collected using a Kappa
X8 APPEX II Bruker diffractometer with graphite-monochromated
MoKa radiation (l=0.71073 ꢄ). The temperature of the crystal was main-
tained at the selected value (100 K) by using a 700 series Cryostream
cooling device within an accuracy of ꢁ1 K. Intensity data were corrected
for Lorentz polarisation and absorption factors. The structures were
solved by direct methods using SHELXS-97[29] and refined against F2 by
full-matrix least-squares techniques using SHELXL-97[30] with anisotropic
displacement parameters for all non-hydrogen atoms.
C
added to avoid non-specific OH production by metallic ion impurities
Treatment of H: H atoms of the ligand were added from the difference
from the buffer.
Fourier map and refined by the riding model. For the [CuII
ACTHNUTRGNEU(GN DAHK)]
C [22b]
Coumarin-3-carboxylic acid (3-CCA) (Sigma) was used to detect HO .
complex only, the H atoms of the water molecules were subsequently in-
C
HO reacts with 3-CCA to form 7-hydroxycoumarin-3-carboxylic acid (7-
ꢀ
cluded in the refinement in geometrically idealized positions, with C H=
OH-CCA), which is fluorescent at 452 nm upon excitation at 395 nm.
The intensity of the fluorescence signal is proportional to the number of
ꢀ
0.96(3) ꢄ and H H=1.52(3) ꢄ, and refined using the riding model with
isotropic displacement parameters of Uiso(H)=1.4Ueq (parent atom). For
C
7-OH-CCA molecules formed, which in turn is proportional to the HO
[CuII
ACHTUNGTRENNUNG(GHK)], the lateral chain of lysine is disordered over two sites with
radicals generated. [CCA]=500 mm.
occupancies of 0.5:0.5. Occupancy parameters for the 16 water oxygen
atoms were refined. The net occupancy of the ordered water is 9.25. All
calculations were performed by using the Crystal Structure crystallo-
graphic software package WINGX.[31] The absolute configuration was de-
termined by refining the Flack parameter[32] using a large number of Frie-
del pairs.
Ascorbate consumption was monitored by UV/Vis spectroscopy. The in-
tensity of the Asc absorption band at l=265 nm (e=14500mꢀ1 cmꢀ1) was
monitored as a function of time, in 100 mm phosphate buffer (pH 7.4)
that contained 100 mm of Asc, 2 mm of DFO, 5 mm of CuII and after addi-
tion of 6 mm of GHK or DAHK peptide.
CCDC-809108 ([CuII(GHK)]) and 809109 ([CuII
ACHTUNGTRENNUNG ACHTUNGTERN(NUGN DAHK)]) contain the
supplementary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
ESI-MS measurements: ESI mass spectra were recorded using an API
This work was supported by a grant from the Agence Nationale de la Re-
cherche, Programme Blanc NT09-488591, “NEUROMETALS” (P.F.,
C.H. and P.D.) and 05-JCJC-0010-01, “NEUROARPE” (P.D.). We thank
the staff of the SAMBA beamline at SOLEIL and more particularly Dr.
Emiliano Fonda for help in performing the XAS experiments (SOLEIL
Project 20080324). Dr. Yannick Coppel and Valꢁrie Bourdon are ac-
knowledged for their help in NMR spectroscopic experiments and analy-
sis and in ESI-MS experiments, respectively. We thank Fanny Leroux for
365 MS mass spectrometer at a flow rate of 5 mLminꢀ1
.
XAS measurements: Cu K-edge XAS spectra were recorded using the
SAMBA bending magnet beamline at Synchrotron SOLEIL (Saint-
Aubin, France). The main optical elements consisted of a double crystal
SiACHTUNGTRENNUNG(111) dynamical focusing monochromator between two palladium-
coated mirrors, one collimating the beam in the vertical direction and
one focusing it on the sample. X-ray harmonic rejection was obtained by
setting the energy cutoff to change the incidence of the mirrors. The
measurements were performed on solution samples at room temperature
in the fluorescence mode using a single-element Silicon Drift Detector
(SDD). The energy was calibrated by the simultaneous measurement of a
Cu foil spectrum in transmission (first inflection point set at 8980.3 eV).
earlier results obtained on [CuII
ACTHNUTRGNEU(GN GHK)], Bertrand Badei for the ascor-
bate consumption experiments and Bruno Alies for his participation in
the NMR spectroscopic experiments.
Possible X-ray photo-reduction of the [CuII
ACTHNUTRGNE(NUG peptide)] samples was moni-
[1] a) H. Kozlowski, A. Janicka-Klos, J. Brasun, E. Gaggelli, D. Valen-
sin, G. Valensin, Coord. Chem. Rev. 2009, 253, 2665–2685; b) E.
Gaggelli, H. Kozlowski, D. Valensin, G. Valensin, Chem. Rev. 2006,
106, 1995–2044.
[2] a) C. Harford, B. Sarkar, Acc. Chem. Res. 1997, 30, 123–130; b) C.
Conato, R. Gavioli, R. Guerrini, H. Kozlowski, P. Mlynarz, C. Pasti,
F. Pulidori, M. Remelli, Biochim. Biophys. Acta Gen. Subj. 2001,
1526, 199–210; c) M. Rꢆzga, M. Sokolowska, A. M. Protas, W. Bal,
J. Biol. Inorg. Chem. 2007, 12, 913–918.
[3] L. Pickart, M. M. Thaler, Nature 1973, 243, 85–87.
[4] A. Simꢁon, F. Monier, H. Emonard, Y. Wegrowski, G. Bellon, J. C.
Monboisse, P. Gillery, W. Hornebeck, F.-X. Maquart, Curr. Top.
Pathol. 1999, 93, 95–101.
[5] L. Pickart, J. Biomater. Sci. Polym. Ed. 2008, 19, 969–988.
[6] P. M. May, J. Whittaker, D. R. Williams, Inorg. Chim. Acta 1983, 80,
L5–L7.
tored by checking, on consecutive scans, the appearance of the XANES
feature at 8984 eV, which is typical of CuI formation. After 5 scans, which
lasted around 10 min each, the feature was still small, thus indicating a
poor reduction effect. Two series of around 10 to 20 scans were recorded.
For the EXAFS analysis, the photo-reduction of scans that can be ne-
glected was averaged. For the XANES study, only the two first scans of
each series were added. Points were measured every 0.25 eV in the
XANES region and steps were gradually increased from 0.5 eV (at E=
9000 eV) to 5 eV (at E=9600 eV) in the EXAFS region.
EPR spectroscopy: EPR spectra (9.4 GHz) were recorded using
a
Bruker ELEXSYS 500 spectrometer equipped with a continuous-flow He
cryostat (Oxford). The field modulation frequency was 100 kHz. Pulsed
EPR experiments were recorded using a Bruker ELEXSYS 580 spec-
trometer at liquid helium temperatures. Pulsed EPR data were processed
by using routines locally written with Matlab (R2008b, The Mathworks,
Inc.).
NMR spectroscopy: 1D 1H and 13C experiments and 2D experiments
were recorded using a Bruker Avance 500 spectrometer equipped with a
5 mm triple-resonance inverse Z-gradient probe (TBI 1H, 31P, BB). All
[7] D. C. Carter, J. X. Ho, Adv. Protein Chem. 1994, 45, 153–203.
[8] W. Bal, J. Christodoulou, P. J. Sadler, A. Tucker, J. Inorg. Biochem.
1998, 70, 33–39.
[9] C. M. Perkins, N. J. Rose, B. Weinstein, R. E. Stenkamp, L. H.
Jensen, L. Pickart, Inorg. Chim. Acta 1984, 82, 93–99.
Chem. Eur. J. 2011, 17, 10151 – 10160
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10159

P. Dorlet, C. Hureau et al.
[10] J. F. Blount, K. A. Fraser, J. T. Freeman, J. T. Szymanski, C.-H.
Wang, Acta Crystallogr. 1967, 22, 396–405.
[21] A. Trapaidze, C. Hureau, W. Bal, M. Winterhalter, P. Faller, unpub-
lished results.
[22] a) K. Takehara, Y. Ide, Inorg. Chim. Acta 1991, 183, 195–202; b) L.
Guilloreau, S. Combalbert, A. Sournia-Saquet, H. Marzaguil, P.
Faller, ChemBioChem 2007, 8, 1317–1325.
[23] V. Balland, C. Hureau, J.-M. Savꢁant, Proc. Natl. Acad. Sci. USA
2010, 107, 3367–3372.
[11] N. Camerman, A. Camerman, B. Sarkar, Can. J. Chem. 1976, 54,
1309–1316.
[12] J. P. Laussac, B. Sarkar, J. Biol. Chem. 1980, 255, 7563–7568.
[13] M. Sano, S. Komorita, H. Yamatera, Inorg. Chem. 1992, 31, 459–
463.
[24] a) D. W. Margerum, K. L. Chellappa, F. P. Bossu, G. L. Burce, J.
Am. Chem. Soc. 1975, 97, 6874–6876; b) C. Hureau, L. Charlet, P.
Dorlet, F. Gonnet, L. Spadini, E. Anxolabꢁhꢇre-Mallart, J.-J. Girerd,
J. Biol. Inorg. Chem. 2006, 11, 735–744.
[25] M. Brzyska, K. Trzesniewska, A. Wieckowska, A. Szczepankiewicz,
D. Elbaum, ChemBioChem 2009, 10, 1–12.
[26] a) S. Furlan, C. Hureau, P. Faller, G. La Penna, J. Phys. Chem. B
2010, 114, 15119–15133; b) C. Hureau, V. Balland, Y. Coppel, P. L.
Solari, E. Fonda, P. Faller, J. Biol. Inorg. Chem. 2009, 14, 995–1000.
[27] R. C. Nadal, S. E. Rigby, J. H. Viles, Biochemistry 2008, 47, 11653–
11664.
[28] L. Perrone, E. Mothes, M. Vignes, A. Mockel, C. Figueroa, M. C.
Miquel, M. L. Maddelein, P. Faller, ChemBioChem 2010, 11, 110–
118.
[14] a) V. A. Streltsov, S. J. Titmuss, V. C. Epa, K. J. Barnham, C. L. Mas-
ters, J. N. Varghese, Biophys. J. 2008, 95, 3447–3456; b) R. W.
Strange, L. Alagna, P. Durham, S. S. Hasnain, J. Am. Chem. Soc.
1990, 112, 4265–4268; c) P. Frank, M. Benfatto, B. Hedman, K. O.
Hodgson, Inorg. Chem. 2008, 47, 4126–4139; d) E. Borghi, P. L.
Solari, J. Synchrotron Radiat. 2005, 12, 102–110; e) C. Hureau, Y.
Coppel, P. Dorlet, P. L. Solari, S. Sayen, E. Guillon, L. Sabater, P.
Faller, Angew. Chem. 2009, 121, 9686–9689; Angew. Chem. Int. Ed.
2009, 48, 9522–9525.
[15] a) J. H. Freedman, L. Pickart, B. Weinstein, W. B. Mims, J. Peisach,
Biochemistry 1982, 21, 4540–4544; b) G. Rakhit, B. Sarkar, J. Inorg.
Biochem. 1981, 15, 233–241.
[16] J. Peisach, W. E. Blumberg, Arch. Biochem. Biophys. 1974, 165,
691–708.
[17] a) S. Stoll, C. Calle, G. Mitrikas, A. Schweiger, J. Magn. Reson.
2005, 177, 93–101; b) B. Kasumaj, S. Stoll, J. Magn. Reson. 2008,
190, 233–247; c) P. Dorlet, S. Gambarelli, P. Faller, C. Hureau,
Angew. Chem. 2009, 121, 9437–9440; Angew. Chem. Int. Ed. 2009,
48, 9273–9276.
[18] I. Bertini, C. Luchinat, Coord. Chem. Rev. 1996, 150, 77–110.
[19] J. P. Laussac, B. Sarkar, Biochemistry 1984, 23, 2832–2838.
[20] M. Sokolowska, A. Krezel, M. Dyba, Z. Szewczuk, W. Bal, Eur. J.
Biochem. 2002, 269, 1323–1331.
[29] SHELXS-97, Program for Crystal Structure Solution, G. M. Shel-
drick, University of Gçttingen, Gçttingen (Germany), 1997.
[30] SHELXL-97, Program for the refinement of crystal structures from
diffraction data, G. M. Sheldrick, University of Gçttingen, Gçttin-
gen (Germany), 1997.
[31] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[32] H. D. Flack, Acta Crystallogr. Sect. A 1983, 39, 876–881.
Received: March 11, 2011
Published online: July 20, 2011
10160
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10151 – 10160