Series of Mn complexes based on N-centered ligands and superoxide - Reactivity in an anhydrous medium and SOD-like activity in an aqueous medium correlated to MnII/MnIII redox potentials

Article abstract of DOI:10.1002/ejic.200400835

Two crystal structures are described in this article: (a) the structure of a monomeric Mn^II complex with the tridentate N-centered N₃ ligand tris[(1-methyl-2-imidazolyl)methyl]amine (TMIMA) ([Mn ^II(TMIMA)₂]²⁺); and (b) the structure of a monomeric Mn^III complex with the tridentate N-centered N₂O ligand 2-{[(1-methyl-2-imidazolyl)methyl]amino}phenolate (PI^-) ^[2] ([Mn^III(PI)₂]⁺) (5). The latter was isolated both in the Mn^II and in the Mn^III state, although only Mn^III crystals were successfully grown. They are part of a series of Mn complexes prepared as SOD mimics, namely [Mn(BMPG)(H ₂O)]⁺ (2) {BMPG = N,N-bis[(6-methyl-2-pyridyl)methyl] glycinate}, [Mn(IPG)(MeOH)]⁺ (3) {IPG = N-[(1-methyl-2-imidazolyl) methyl]-N-(2-pyridylmethyl)glycinate}, [Mn(BIG)(H₂O) ₂]⁺ (4) {BIG = N,N-bis[(1-methyl-2-imidazolyl)methyl] glycinate}. The reactivity of Mn^II complexes 1 and 2 in an anhydrous medium is described and compared to that of complexes 3 and 4, the data for which was previously published. The cyclic voltammograms of the whole complex series were recorded in an aqueous medium (collidine buffer). Their SOD-like activities were estimated by the McCord-Fridovich test (IC50 with 22 μM cytc Fe^III: 1.6±0.1 μMol L^-1 for 1, 1.2±0.5 μmol L^-1 for 2, 3.0±0.2 μmol L^-1 for 3, 3.7±0.6 μmol L^-1 for 4, 0.8±0.1 μmol L ^-1 for 5). IC50 values were converted into the corresponding kinetic constant k_McCF values. A linear correlation between E_a and log(k_McCF) was obtained, indicating that in this series the conversion to Mn^III is probably the rate-limiting step. This is of substantial importance for further Mn-SOD mimic design in this series. Wiley-VCH Verlag GmbH & Co. KGaA 2005.

Full text of DOI:10.1002/ejic.200400835

FULL PAPER
Series of Mn Complexes Based on N-Centered Ligands and Superoxide –
Reactivity in an Anhydrous Medium and SOD-Like Activity in an Aqueous
Medium Correlated to MnII/MnIII Redox Potentials[‡][‡‡]
Stéphanie Durot,[a][‡‡‡] Clotilde Policar,*[a][‡‡‡] Federico Cisnetti,[a] François Lambert,[a]
Jean-Philippe Renault,[c] Giorgio Pelosi,[b] Guillaume Blain,[d] Hafsa Korri-Youssoufi,[a] and
Jean-Pierre Mahy[a]
Keywords: N,O ligands / Manganese / Superoxide / SOD mimics / Imidazole / Phenolate
Two crystal structures are described in this article: (a) the
structure of a monomeric MnII complex with the tridentate N-
centered N3 ligand tris[(1-methyl-2-imidazolyl)methyl]amine
medium is described and compared to that of complexes 3
and 4, the data for which was previously published. The cy-
clic voltammograms of the whole complex series were re-
(TMIMA) ([MnII(TMIMA)2]2+); and (b) the structure of a mo- corded in an aqueous medium (collidine buffer). Their SOD-
nomeric MnIII complex with the tridentate N-centered N2O
ligand 2-{[(1-methyl-2-imidazolyl)methyl]amino}phenolate
(PI–)[2] ([MnIII(PI)2]+) (5). The latter was isolated both in the
MnII and in the MnIII state, although only MnIII crystals were
successfully grown. They are part of a series of Mn com-
plexes prepared as SOD mimics, namely [Mn(BMPG)(H2O)]+
(2) {BMPG = N,N-bis[(6-methyl-2-pyridyl)methyl]glycinate},
[Mn(IPG)(MeOH)]+ (3) {IPG = N-[(1-methyl-2-imidazolyl)
methyl]-N-(2-pyridylmethyl)glycinate}, [Mn(BIG)(H2O)2]+ (4)
like activities were estimated by the McCord–Fridovich test
(IC50 with 22 µ
M
cytc FeIII: 1.6 0.1 µ olL–1 for 1,
M
1.2 0.5 µmolL–1 for 2, 3.0 0.2 µmolL–1 for 3, 3.7 0.6 µmolL–1
for 4, 0.8 0.1 µmolL–1 for 5). IC50 values were converted into
the corresponding kinetic constant kMcCF values. A linear
correlation between Ea and log(kMcCF) was obtained, indicat-
ing that in this series the conversion to MnIII is probably the
rate-limiting step. This is of substantial importance for fur-
ther Mn–SOD mimic design in this series.
{BIG
=
N,N-bis[(1-methyl-2-imidazolyl)methyl]glycinate}.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
The reactivity of MnII complexes 1 and 2 in an anhydrous
ter, catalytically dismutate superoxide.[6–8] The design of low
molecular weight complex SOD mimics requires greater un-
derstanding of the effect of the metal environment on the
SOD-like activity. In the field of superoxide dismutase mim-
ics, copper, iron, manganese and cobalt complexes have
been described.[7–12] Mn is the least toxic of these metal
ions.[8] In our laboratory, we design Mn complexes and test
them as SOD mimics. In this article, the crystal structure of
a monomeric MnII complex (1) derived from L1 {tris[(1-
methyl-2-imidazolyl)methyl]amine or TMIMA} is pre-
sented {[MnII(TMIMA)2](PF6)2}. The crystal structure of
a monomeric MnIII complex derived from L2 (2-{[(1-
methyl-2-imidazolyl)methyl]amino}phenolate or PI–) is also
described {[MnIII(PI)2](PF6)}. In the latter case, both MnII
(5Ј) and MnIII states (5) could be isolated, although only
Introduction
Superoxide (O2·–), as the first species in the dioxygen re-
duction cascade, is a toxic species involved in oxidative
stress.[3–5] Its concentration is controlled in vivo by superox-
ide dismutases, metalloenzymes that catalyze its dismu-
tation. Superoxide is considered to be responsible for se-
veral pathological situations. There is a pharmaceutical
requirement for molecules that could scavenge or, even bet-
[‡] Part 2 of a series. Part 1: Ref.[1]
[‡‡] Abbreviations: Ref.[2]
[a] Laboratoire de Chimie Bio-organique et Bio-inorganique,
UMR8124, Bâtiment 420, Université Paris XI,
91405 Orsay Cedex, France
Fax: +33-1-69157231
E-mail: cpolicar@icmo.u-psud.fr
[b] Dipartimento di Chimica Generale ed Inorganica, Chimica MnIII crystals were successfully obtained. Complexes 1 and
Analitica, Chimica Fisica, Università di Parma,
5 are part of a series of Mn complexes prepared as SOD
Parco Area delle Scienze 17A, 43100 Parma, Italy
mimics, previously published, namely [Mn(BMPG)(H2O)]+
[c]
Laboratoire de Radiolyse CEA Saclay (DSM/DRECAM/
SCM-URA331),
(2) {BMPG
=
N,N-bis[(6-methyl-2-pyridyl)methyl]-
N-[(1-
91405 Orsay Cedex, France
glycinate},[13] [Mn(IPG)(MeOH)]+ (3) {IPG
=
[d] Laboratoire de Chimie Inorganique, UMR8613, Bâtiment
420, Université Paris XI,
methyl-2-imidazolyl)methyl]-N-(2-pyridylmethyl)glycinate},
[Mn(BIG)(H2O)2]+ (4) {BIG = N,N-bis[(1-methyl-2-imid-
azolyl)methyl]glycinate}.[1] A general approach to reaction
with superoxide has already been published.[1] Firstly, this
91405 Orsay Cedex, France
[‡‡‡]
These two authors contributed equally to this work.
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2005, 3513–3523
DOI: 10.1002/ejic.200400835
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3513

C. Policar et al.
FULL PAPER
article consists of a study of the reactivity of MnII com-
plexes with superoxide in an anhydrous medium (DMSO).
Secondly, the reactivity in an aqueous solution is evaluated
with the conventional McCord–Fridovich assay and/or
pulse radiolysis. Presented is a brief study of the MnII com-
plexes (1 and 2) in the anhydrous medium and an evalu-
ation of the SOD activity of complexes 1, 2, and 5 with
the McCord–Fridovich assay. These activities are presented
here in terms of a recalculated kinetic constant, kMcCF. For
two of the complexes, namely, 3 and 4, kMcCF values are
compared with the kinetic constants measured by pulse ra-
diolysis. For the whole series the relationship between redox
potential and kMcCF has been investigated.
Results
Figure 1. Structure of the ligands.
Syntheses
presence of dioxygen, a dark violet precipitate of an MnIII
complex (5) was directly obtained.
Ligand Synthesis
The synthesis of the L1 ligand is already known.[14,15]
A
new synthetic pathway is presented in the Exp. Sect., involv-
ing an amine tertiarization by reductive amination instead
of a nucleophilic substitution. The new method displayed a
higher yield (90% vs. 62% from bis[2-(1-methyl-2-imid-
azolyl)methyl]amine.
Description of the Structure of [Mn(TMIMA)2](PF6)2 (1)
An X-ray diffraction study was carried out on the
[Mn(TMIMA)2](PF6)2 complex 1 and the experimental de-
tails are reported in Table 6 (Exp. Sect.). The asymmetric
unit consists of a cationic moiety, formed by two neutral
tridentate ligands coordinated to a central MnII ion, lying
on a center of symmetry, and two hexafluorophosphate
anions (see Figure 2). Selected distances and angles are re-
ported in Table 1.
Ligand L5 was synthesized as shown in Scheme 1 by a
two-step synthesis. The yield was moderate but the pro-
cedure was very easy. 2-(Methylamino)phenol was obtained
by catalytic hydrogenation (Ni) of 2-cyanophenol. It was
reductively coupled to N-methyl-2-imidazolcarboxaldehyde
obtained according to a published procedure.[16]
The hexafluorophosphate anions are found in two posi-
tions with an occupancy of 0.5. The manganese() ion is
Preparation of the Complexes
MnII complex 1 was obtained from L1 (see Figure 1) with coordinated to six imidazole nitrogen atoms with distances
–
PF6 as the counter-anion in MeOH/H2O (1:1). X-ray suit- of 2.364(2), 2.366(3), and 2.307(2) Å (by symmetry, the
able crystals were successfully grown providing the mono- other Mn–N distances are identical). These distances are
meric [MnII(TMIMA)2](PF6)2 compound 1. Experimental larger than the average value found using the Cambridge
details are described in the Exp. Sect.
Crystallographic Database (CCDC)[17] for manganese com-
MnII complex 5Ј was obtained from L5 (see Figure 1) plexes with ligands containing the imidazole ring [av.
under strictly controlled anaerobic conditions. A white pre- 2.226(12) Å on the basis of 42 fragments; see the list in the
cipitate was obtained, but no crystal has been successfully Supporting information].
grown so far. The powder could be recrystallized from
The Mn–Ntripod distance is 2.916(2) Å, which is longer
CH3CN by diffusion of tBuOMe as [MnIII(PI)2]- than the sum of their van der Waals radii (2.82 Å). This is
(PF6)(CH3CN). When the synthesis was carried out in the due both to the complete metal coordination sphere and to
Scheme 1. Synthetic pathway to PIH.
3514
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3513–3523

Reactivity and SOD-Like Activity Correlated to MnII/MnIII Redox Potentials
FULL PAPER
Table 6 (Exp. Sect.). The structure consists of two cationic
moieties made up of manganese atoms lying on centers of
symmetry surrounded by two tridentate monoanionic li-
gands. Manganese() ions present a slightly distorted octa-
hedral coordination geometry as can be seen from the coor-
dination distances and angles (Table 2), with a compression
along the Ophenolato–Mn–Ophenolato
.
Figure 2. Representation of cationic complex 1.
Table 1. Selected bonds and angles (symmetry operator A: 1 – x,
–y, 1 – z).
Figure 3. Representation of cationic complex 5.
Table 2. Selected distances and angles for 5.
Bonds
[Å]
Angles
[°]
Mn1–N2
Mn1–N4
Mn1–N6
Mn1–N2A
Mn1–N4A
Mn1–N6A
2.364(2)
2.307(2)
2.366(3)
2.364(2)
2.307(2)
2.366(3)
N2–Mn1–N4
N2–Mn1–N6
N4–Mn1–N6
N2–Mn1–N4A
N2–Mn1–N6A
N4–Mn1–N6A
98.73(7)
99.91(8)
100.48(7)
81.28(7)
80.09(7)
79.52(7)
Distances
[Å]
Angles
[°]
Mn1–O1
Mn1–N1
Mn1–N2
Mn2–O2
Mn2–N4
Mn2–N5
1.82(1)
2.20(2)
2.17(2)
1.85(1)
2.19(2)
2.18(2)
O1–Mn1–N1
O1–Mn1–N2
N1–Mn1–N2
O2–Mn2–N4
O2–Mn2–N5
N4–Mn2–N5
91.6(7)
92.3(7)
78.9(8)
91.7(6)
93.3(6)
79.0(7)
the steric constraints imposed by the presence of two bulky
ligands. In comparable structures, with a single tripodal li-
gand and a monodentate donor that completes the octahe-
dron, the tripodal nitrogen atom is much closer to the cen-
tral ion and forms a coordination bond: the average dis-
tance in these cases is 2.53(4) Å (on the basis of the same 42
fragments listed in the Supporting information). Deviation
from the octahedral structure is also noticeable in the angle
values (see Table 1).
Sheets of complex molecules, interacting through van der
Waals contacts, pile up running parallel to the xz plane and
are held together by a complex system of hydrogen bonds
formed by the fluorine atoms of the hexafluorophosphate
anions with the methyl and methylene groups of the ligand.
The shortest distance between manganese ions is Mn···Mn
(x, –y – 1/2, +z – 1/2) and is 9.413(1) Å. The magnetic prop-
erties of complex 1 are that of a Curie-law MnII ion, with
a χmolT value of 4.4 cm3 mol–1 K from ambient temperature
down to 2 K (results not shown), with no efficient MnII–
MnII intermolecular through-space interaction at such dis-
tances.[13]
It is noteworthy that a hydrogen bond exists in the two
independent molecules between the amino NH group and
the phenol oxygen atom of the centrosymmetrically related
ligand molecule [N1···O1 2.81(2) Å and N4···O2 2.83(2) Å].
A hexafluorophosphate anion compensates the charge of
the cationic complex. Hydrogen bonds between fluorine
atoms of PF6– and N1 and N4 are responsible for the over-
all packing. The structure also contains an acetonitrile
crystallization molecule that occupies the remaining inter-
stices.
Reactivity of the MnII Complexes with Superoxide in an
Anhydrous Medium
In the series 1, 2, 3, and 4, MnII ions bear 0, 1 or 2 H2O
molecules, and show different bulkiness around Mn. The
reactivity of both [MnII(BMPG)(H2O)]+ (2) and [MnII-
(TMIMA)2]2+ (1) was studied as described previously in the
case of both 3 and 4.[1]
Description of the Structure of [Mn(PI)2](PF6)(CH3CN) (5)
Successive amounts of superoxide were added to a
An X-ray diffraction study has been carried out on the 10–3 molL–1 solution of the complex in anhydrous DMSO.
complex [Mn(PI)2](PF6)(CH3CN) (5). An ORTEP view of EPR spectra were recorded at 10 K. For 2, the initial MnII
the manganese complex is shown in Figure 3 and the exper- signal[18] decreased from 0 to 1.5 equiv. superoxide. From
imental details regarding data collection are reported in 1.5 to 3 equiv., the solution was EPR-silent; and above
Eur. J. Inorg. Chem. 2005, 3513–3523
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3515

C. Policar et al.
FULL PAPER
Table 3. Reactivity of superoxide in anhydrous DMSO with MnII complexes from the N-tripodal series: amount of superoxide consumed
deduced from the EPR spectrum.
2+
+
MnII(TMIMA)2
0
MnII(IPG)(MeOH)+
0
MnII(BMPG)(H2O)+
1
MnII(BIG)(H2O)2
2
Number of water bound to
Mn
·–
O2 consumed (equiv./MnII)
1 equiv.
1 equiv.
3 equiv.
Ͼ5 equiv.
·–
4 equiv., a signal from superoxide was finally obtained and
The superoxide signal was recovered at a different O2 /
quantified (data not shown, see Exp. Sect. and Table 3). For Mn stoichiometry for these complexes and quantified. The
·–
1, the 6-line hyperfine signal from MnII decreased upon su- amount of O2 consumed was determined (see Table 3). It
peroxide addition; and at 2 equiv., the superoxide EPR sig- appears that the more water molecules are coordinated to
nal was recovered and quantified (see Exp. Sect., Table 3 the MnII, the more equivalents of superoxide are consumed.
and Figure 4).
The presence of the Mn ion is necessary since water alone,
at the same concentration and under the same experimental
conditions, did not switch off the superoxide signal.
Tuning Redox Potentials to Tune the SOD-Like Activity
Redox Potential and SOD Activity
The catalysis of the superoxide dismutation is a redox
process, involving both MnII and MnIII oxidation states. To
make this redox catalysis efficient, the redox potential of the
MnIII/MnII couple should be, as encountered in SODs,[20]
·–
·–
between the potential of the two couples O2/O2 and O2
/
H2O2−that is, at pH = 7, –0.40 V and +0.65 V/SCE, respec-
tively (see Figure 5).[9,21–24] Theoretically, the closer the po-
tential of the two couples, the faster the reaction between
them. Hitherto, the value optimizing the kinetics of both
oxidation and reduction is the midway potential, i.e. 0.12 V/
SCE.[9,20]
Figure 4. Reactivity of [Mn(TMIMA)2]2+ (1) with superoxide in
anhydrous DMSO. [Mn(TMIMA)2](PF6)2 (10–3 molL–1), 100 K, ν
= 9.38 GHz, att. 20 dB, mod. amp. 5 G, tc = 41 ms.
The MnIII/MnII redox potential is controlled by the coor-
dination sphere around the metal center, either through the
These results show that both 1 and 2 react with superox- nature of the Lewis bases coordinated to the Mn center, by
ide. No multiline signal was observed, showing that the for- electronic effect on the Lewis bases,[25] or by the geometry
mation of MnIIIMnIV di-µ-oxo dimers, previously observed around the Mn center. Increasing the number of the O–
with 3 and 4,[1,19]was not obtained in the case of 1 and 2. donor will lower the potential by increasing the electron
This can be tentatively assigned to the steric crowding due density onto the metal center. MnIII, as a d4 ion, will prefer
the methyl groups in 2 and to the complete coordination a geometric distortion from the octahedric structure. The
sphere in 1.
redox potential of the MnIII/MnII couple can be lowered by
Figure 5. (A) Apparent standard redox potential for superoxide couples at pH = 7. (B) Anodic and cathodic half-wave potentials for the
complexes in collidine pH 7.5 buffer.
3516
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3513–3523

Reactivity and SOD-Like Activity Correlated to MnII/MnIII Redox Potentials
FULL PAPER
controlling the geometry around the metal center, that is by etic competition for superoxide reduction between the puta-
imposing some deformation with regards to the octahedral tive superoxide scavenger and ferricytochrome c.[31] The re-
environment.[26]
duction of ferricytochrome c was monitored spectrophoto-
No electrochemical wave was observed for the MnII com- metrically at 550 nm. This test requires several preliminary
plexes of this series in dimethyl sulfoxide, acetonitrile or checks in order to ensure its reliability (see Exp.
phosphate buffer. This is not uncommon for MnII com- Sect.).[1,8,32,33]
plexes.[27–29] In collidine buffer (pH = 7.5)[30] however,
Complexes 1–5 inhibited the reduction of ferricyto-
irreversible waves were recorded (see Figure 6 in the case of chrome c. The inhibition percentage was measured for
4). Anodic and cathodic potentials (Table 4 and Figure 5) several complex concentrations and the IC50 value was
are within (2, 5, and 1) or at the limit (3 and 4) of the then deduced by two graphical methods (see Figure 7 and
required region for effective superoxide dismutation. Poten- Exp. Sect. for details).[1,31,32,34,35] They are reported in
tials for 3 and 4 are close, which is consistent with the sim- Table 5.
ilar environment around the metal cations. They are close
to the potential previously reported for a MnII coordinated
with N,N-bis(2-pyridylmethyl)-(S)-histidine, which shows a
similar coordination sphere (2 imidazoles, 1 pyridine, 1 car-
boxylate).[30] For 1 and 2, the anodic wave potentials are
less positive. The conversion of MnII into MnIII is thus eas-
ier. This can be assigned to the steric crowding encountered
in 1 and 2, which may favor a distorted environment around
the metal center. In 5, the N4O2 environment favors the
MnIII by comparison with the nitrogen-richer environment
for the other complexes.
Figure 6. Voltammogram of 4 [2×10–3 molL–1 (on the basis of Mn
content)], collidine buffer (50 m), pH = 7.5, 100 mVs–1. Working
electrode: vitrous carbon disk.
Figure 7. A. Kinetics of the reduction of ferricytochrome c
(550 nm) without and with the putative SOD mimic 4. B. Inhibition
percentage as a function of 4: IC50 determination.
Table 4. Redox potential for the anodic and cathodic waves for
complexes 1–5.
It should be noted that reported IC50 values are depend-
ent both on the detector used [usually cytochrome c FeIII
or nitroblue tetrazolium (NBT)] and on its concentration.
For a given putative SOD mimic, the smaller the detector
concentration, the smaller the IC50 value. Therefore, IC50
values are not appropriate for comparisons with the litera-
ture. From the measured IC50 values, it is possible to calcu-
late a kinetic constant value (kMcCF), which is independent
of both detector concentration and nature. At the IC50
concentration, superoxide reacts at the same speed with the
[a]
[a]
Complex
Ea
Ec
1
2
3
4
5
0.64
0.44
0.76
0.76
0.34
0.28
0.17
0.27
0.33
0.20
[a] V vs. SCE.
detector and the putative SOD mimic. Then, kMcCF
=
Evaluation of the SOD-Like Activity: McCord–Fridovich
Assay
kdetector·[detector]/IC50.[36] In the case of cytochrome c FeIII
as the detector, kCyt c (pH
=
7.8; 21 °C)
=
Reactivity towards the superoxide was investigated in an 2.6×105 mol–1 Ls–1.[37] In the case of NBT, kNBT (pH = 7.8)
aqueous phosphate buffer (pH = 7.8) with the McCord– = 5.94×104 mol–1 Ls–1.[38]
Fridovich assay, using the xanthine-xanthine oxidase sys-
Table 5 reports some values from the literature. The
tem to produce the superoxide. The test is based on a kin- tested compounds 1–4 are within the range of the activities
Eur. J. Inorg. Chem. 2005, 3513–3523
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3517

C. Policar et al.
Ref.
FULL PAPER
Table 5. IC50 values and kinetic catalytic constants: comparison with the literature.
[c]
Complex
IC50[a]
kcat [mol–1Ls–1
2.1×104 IT
]
[39]
[1]
[FeIII(IPG)Cl2]
IC5022 µm Cyt c = 2.7×10–4 molL–1
IC5022 µm Cyt c = 1.67 3×10–4 molL–1
IC5022 µm Cyt c = 1.4×10–4 molL–1
IC506.7 µm Cyt c = 1.04×10–5 molL–1
IC5010 µm Cyt c = 9.2×10–6 molL–1
IC5050 µm NBT = 6.5×10–6 molL–1
IC50XTT = (0.33 0.014)×10–5 molL–1 [b]
IC5022 µm cyt c = 4.3×10–6 molL–1
IC5022 µm cyt c = 4.5 0.5×10–6 molL–1
IC5022 µm cyt c = 3.7 0.6×10–6 molL–1
IC5022 µm cyt c = 3.0 0.6×10–6 molL–1
IC5022 µm cyt c = 1.6 0.1×10–6 molL–1
IC5046 µm NBT = 0.7×10–6 molL–1
IC5050 µm NBT = 0.75×10–6 molL–1
IC5046 µm NBT = 0.64×10–6 molL–1
IC5022 µm cyt c = 1.2 0.5×10–6 molL–1
IC5022 µm cyt c = 8.7 0.6×10–7 molL–1
IC5010 µm Cyt c = 2.9×10–7 molL–1
–
[MnII(ClO4)2] + EDTA
3.4 0.1×104 IT
4.1×104 IT
[39,40]
[32]
[FeIII(BIG)Cl2]
[MnIII(DFB)]
1.0×106 IT
[41]
[MnIII(TBAP)]+
2.8×105 IT
[42,43]
[44]
[MnII(PA)2(PAH)(H2O)]
4.6×105 IT
[MnIII(TPAC)]
(9.3 0.4)×105 IT
1.3×106 IT; 1.05×107 PR
1.3 0.2×106 IT
[28]
[MnII(TPAA)](PF6)2
[1]
[MnII(ClO4)2]
4
3
1
1.5 0.3×106 IT; 1.4 0.1×106 PR [1]; this work
1.9 0.5×106 IT; 3.8 0.2×106 PR [1]; this work
3.6 0.2×106 IT
3.9×106 IT
this work
[45]
[MnII(NTB)(salicylate)]ClO4
[42]
[46]
MnII(Obz)(3,5-iPr2pzH)(HB(3,5-iPr2pz)3)
4.0×106 IT
4.3×106 IT
[MnII(NTB)(terephthalate)]
2
4.8 1.4×106 IT
6.6 0.5×106 IT
9.0×106 IT
this work
5
this work
[30]
[MnII{N,N-bis(2-pyridylmethyl)-(S)-histidinate}(H2O)]2(ClO4)2
[MnII(EDTB)(Ac)]Ac·EtOH
M40403
[38]
[47]
[48]
[23]
[49]
[50]
1.0×107 IT
–
1.6×107 SF
5.0×107 IT
[MnIII(BVDME)]2
[MnIIITE-2-PyP(OH2)]5+
[MnIIICl4 TE-2-PyP]5+
IC5010 µm Cyt c = 5.0×10–8 molL–1
IC5010 µm Cyt c = 4.5×10–8 molL–1
IC5010 µm Cyt c = 6.5×10–9 molL–1
–
5.8×107 IT
4.0×108 IT
S,S-Dimethyl-M40403
1.6×109 SF
[a] 50% inhibition concentration. Indices specify the UV/Vis probe (cytc FeIII or NBT) concentration used for the test. kcat is the catalytic
constant directly measured by pulsed radiolysis or stopped-flow techniques and kMcCF are recalculated kinetic constants from the IC50
[51]
value. [b] Measured with XTT as UV/Vis probe.
kcat was directly provided in the article, by comparison with SOD.[52] [c] IT: indirect
test; SF: stopped flow; PR: pulsed radiolysis. DFB: desferrioxamine B; TBAP: 5,10,15,20-tetrakis(4-benzoic acid)porphyrin; PA: picolin-
ate; TPAC: cis,cis-1,3,5-tri{(5-trimethylammonium)-2-hydroxypicolyl)amino}cyclohexane; salen: N,NЈ-bis(salicylideneamino)ethane;
TPAA: tris{2-[N-(2-pyridylmethyl)amino]ethyl}amine; NTB: tris(2-benzimidazolylmethyl)amine; XTT: 3Ј-{1-[(phenylamino)carbonyl]-
3,4-tetrazolium}bis(4-methoxy-6-nitrobenzenesulfonic acid) hydrate; Obz: benzoate; HB(3,5-iPr2pz)3: hydrotris(3,5-diisopropyl-1-pyrazol-
yl)borate; EDTB: N,N,NЈ,NЈ-tetrakis(2Ј-benzimidazolylmethyl)-1,2-ethanediamine; M40403 and S,S-dimethyl-M40403: MnII complexes
from modified 1,4,7,10,13-pentaazacyclopentadecane ligands; BVDME: biliverdine IX dimethyl ester; TE-2-PyP: 5,10,15,20-tetrakis(N-
ethylpyridinium-2-yl)porphyrin.
reported for MnII complexes. Compound 5 is the best lytic activity, the kMcCF value derived from the IC50 value
within this series. IC50 values for MnII complexes 3 and 4 (see above) corresponds to the kcat
.
are far better than those reported for the corresponding
To test the catalytic behavior within the series, we studied
FeIII complexes. Mn seems to be a most suitable ion for the reactivity of 3 and 4 by pulse radiolysis. They were cho-
SOD mimics, as it is the least toxic of the Fe, Co, Ni, Cu, sen because they have the highest IC50 values of the series,
Mn series. Moreover, the best SOD mimic reported so far i.e. they were the slowest of the series.
is a MnII complex (see Table 5).
Pulse radiolysis was used to produce the superoxide radi-
cal by irradiation of oxygenated aqueous solutions with
high energy ionizing radiation.[53] Primary radicals (H·, e–
,
aq
Pulse Radiolysis (for 3 and 4)
HO·) are rapidly and quantitatively converted into either
·–
·
superoxide O2 or its protonated form hydroperoxyl HO2 ,
in the presence of either formate or alcohols.[54,55] Propan-
2-ol was chosen as formate might coordinate to Mn. The
aqueous solution, driven at a constant rate through the UV/
Vis microcell, was irradiated for a short time period with
an irradiation frequency (1 Hz) chosen so that between two
pulses the solution was totally renewed. Absorbances were
averaged over 100 pulses. This setup provides an improve-
ment of the signal to noise ratio, by comparison with gen-
erally used single-pulse experiments.[56–59]
This McCord–Fridovich assay, as a determination of
catalytic SOD-like activity is controversial. It has been
shown, using stopped-flow techniques,[8,33] or pulse radioly-
sis,[28] that such an indirect test can be positive in the case
of both a catalytic and a stoichiometric superoxide scaven-
ger. As a matter of fact, in the McCord–Fridovich assay,
superoxide is produced in low amounts. For compounds
displaying high IC50 values (over 1 µmolL–1), the condi-
tions required to probe a catalytic behavior may not be ful-
filled if ncomplex ϾϾ nsuperoxide in the experiment. However,
this test remains relevant as the X/XO system provides a
Superoxide decay was directly monitored at 270 nm, in
stationary state of superoxide of low concentration, which aqueous phosphate buffer solutions (5 mmolL–1, pH = 7.8).
is closer to what is usually encountered in vivo.[11,24] More- Spontaneous decay of O2·– followed a second order rate law,
over, in the case of a small IC50 value indicative of a cata- as expected for autodismutation of superoxide. The ob-
3518
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3513–3523

Reactivity and SOD-Like Activity Correlated to MnII/MnIII Redox Potentials
FULL PAPER
served rate constant had a value of 1.3×105 mol–1 Ls–1, con- were studied as SOD mimics. Reactivity in an anhydrous
sistent with that reported in the literature[54] at pH = 7.8 medium was described for 1 and 2 and compared with that
(1.0×105 mol–1 Ls–1). The effect of either 3 or 4 on the su- of 3 and 4, previously published.[1] The SOD activity was
peroxide decay was tested for different concentrations of evaluated through the McCord–Fridovich assay and pulse
complex (2×10–7, 10–5 molL–1 for
3
and 2.14×10–7, radiolysis. IC50 values directly obtained from the McCord–
2.14×10–6, and 2.14×10–5 molL–1 for 4, with an initial su- Fridovich assay, are dependent on the experimental condi-
·–
peroxide concentration [O2
]
= 7.3×10–5 molL–1). From tions used−that is on the concentration of the detector and
0
the absorbance decay at 270 nm, kinetic constants were de- its concentration. The results are thus provided and dis-
duced (see Exp. Sect.). The catalytic rate constants kcat were cussed here as derived values, kMcCF. We suggest that kMcCF
obtained (see Exp. Sect.). They are reported in Table 5.
should be more commonly used, as IC50 values are not
The apparent first-order kinetics observed for the super- comparable from one experiment to the other if different
oxide decay in the presence of either complex, under condi- detectors or different concentrations of the same detector
tions where ncomplex Ͻ nsuperoxide, showed that both 3 and 4 were used. The SOD activity in this series is found to be
act as true catalysts for superoxide dismutation. The derived within the reported range for MnII compounds (see
catalytic rate constant for 3 is of the same order of magni- Table 5). The redox properties of the complexes from this
tude than that calculated from the IC50 values obtained by series were studied by cyclic voltammetry in a collidine
the McCord–Fridovich assay. For 4, rate constants ob- buffer. The potentials of the complexes of the series were
tained from both methods are in good agreement.[60]
tuned both by inducing steric constraints and by increasing
This pulse radiolysis experiment is strong evidence for the Lewis donor character of the ligands. The present arti-
the catalytic behavior of 3 and 4. This conclusion concern- cle shows a linear trend between the anodic half-wave po-
ing 3 is reinforced by further pulse radiolysis experiments tential and log(kMcCF) with a negative slope, indicating that
that is described elsewhere.[61]
the SOD activity in such a series can be improved by lower-
ing the redox potential. Some new compounds are now un-
der investigation to improve SOD activity.
Correlation between SOD Activity and Potentials
As shown above, all redox potentials of the complexes
of this series are in the expected range for the catalysis of
superoxide dismutation. A linear correlation can be found
between anodic half-wave potentials for the MnIII/MnII
Experimental Section
General: IR spectra (KBr) were recorded with a Bruker IFS 66 FT-
couple and log(kMcCF) with a negative slope, as shown in IR spectrometer. 1H NMR spectra were recorded with a Bruker
AM 250 spectrometer (1H, 250 MHz) or a Bruker AC 360 (1H,
Figure 8. Within this series, the activity can be improved by
360 MHz). Cyclic voltammetry experiments were performed with
a stabilization of the + oxidation state: the rate-limiting
step is the oxidation into MnIII. Similar correlations have
an AUTOLAB potentiostat. Electronic absorption spectra were re-
corded with a Safas 190 DES double-mode spectrophotometer.
been described previously: (a) in the case of different series
Chemical reagents were purchased either from Aldrich or Acros
of MnIII–porphyrins,[9,23,49] with a positive slope, indicating
and used without further purification. Xanthine oxidase and ferric-
ytochrome c were purchased from Sigma.
that the reduction to MnII was the rate-limiting step and;
(b) in the case of FeII complexes with a negative slope, indi-
Syntheses and Characterization
cating that the oxidation to iron() was the rate-limiting
step.[62]
N,N,N-Tris[(1-methyl-2-imidazolyl)methyl]amine (TMIMA):
A
solution of N,N-bis[(1-methyl-2-imidazolyl)methyl]amine[40] (see
also refs.[14,63,64]) (4.1 g, 0.02 mol) with 1-methyl-2-imidazolcarbal-
dehyde (2.5 g, 0.02 mol)[16] in non-anhydrous MeOH (150 mL) was
hydrogenated in the presence of Pd/C (10%) for 3 d. Pd/C was re-
moved by filtration and the supernatant was concentrated to dry-
ness. The white solid was purified by precipitation in CH3Cl/Et2O.
1
Yield: 90%. M.p. 200 °C. H NMR (250 MHz, CD3OD): δ = 3.0
(s, 9 H, CH3N), 3.8 (s, 6 H, 3 im-CH2N), 6.7 (d, J = 2.3 Hz, 3 H,
Him), 6.9 (d, J = 2.3 Hz, 3 H, Him) ppm. 13C NMR (62 MHz,
CD3OD): δ = 31.7 (CH3N), 48.7 (NCH2-im), 121.3 (C4im), 127.3
(C5im), 145.4 (C2im) ppm. IR (KBr; strong bands only): ν = 1495.8
˜
(νim), 749.3 (δim) cm–1. ESI-MS: m/z (%) = 300.2 (30) [M + H],
322.2,(10) [M + Na].
[2-(Methylamino)phenol]: This protocol was inspired from a pub-
lished aliphatic nitrile reduction procedure.[65] 2-Cyanophenol (4 g,
33.6 mmol) was dissolved in EtOH (120 mL). An aqueous solution
(20 mL) of NaOH (5 g, 125 mmol) and Raney nickel (about 3 g)
were then added. The resulting suspension was stirred under H2
(40 bars) for 4 h. After decantation of the nickel, the solution was
filtered and EtOH was evaporated. CH2Cl2 (30 mL) was added and
Figure 8. Correlation between Ea and log(kMcCF).
Conclusions
The new Mn complexes 1 and 5, whose structures are
presented here, are part of a series of Mn complexes that the pH of the aqueous phase was brought to 9.3 by addition of 2 
Eur. J. Inorg. Chem. 2005, 3513–3523
www.eurjic.org © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 3519

C. Policar et al.
FULL PAPER
HCl. The organic phase was decanted and the aqueous phase was
further extracted 3 times with small portions of CH2Cl2. The or-
ganic phase was dried (Na2SO4) and the solvents were evaporated.
2-(Methylamino)phenol (2.42 g, yield 59%) was obtained as a white
solid. The nonquantitative yield is probably due to the formation
of the NiII complex which gives a blue-green coloration to the
aqueous phase. 1H NMR (250 MHz, CDCl3): δ = 1.25 (br. s, NH2),
4.14 (s, 2 H, CH2), 6.81 (m, 2 H, Ar), 6.98 (d, J = 5 Hz, 1 H, Ar),
7.18 (t, J = 7.5 Hz, 1 H, Ar) ppm.
pressure. A deoxygenated solution (2.5 mL H2O/2.5 mL MeOH) of
NH4PF6 (0.25 g, 1.53 mmol, 2.6 equiv.) was then added. A white
powder was then precipitated. The solid was filtered off and redis-
solved in CH3CN. The solution turned violet and crystals of
[Mn(PI)2](PF6)·CH3CN suitable for X ray diffraction were ob-
tained in low yield (50% from the powder) from acetonitrile by
diffusion of tBuOMe as tiny dark crystals (see below). When the
preparation was carried out at a higher temperature (50 °C) and
under dioxygen, a purple black solution was obtained and a purple
black powder was precipitated upon NH4PF6 addition.
C24H28F6N6O2MnP(CH3CN)0.5(H2O)1.5 (679.66): calcd. C 44.14,
H 4.82, Mn 8.08, N 13.39, P 4.56; found C 44.30, H 4.51, Mn 7.99,
2-({[(1-Methyl-2-imidazolyl)methyl]amino}methyl)phenol (PIH): 2-
(Methylamino)phenol (1 g, 8.1 mmol) and 1-methyl-2-imidazole-
carbaldehyde (0.89 g, 8.1 mmol), synthesized according to a pub-
lished procedure[16] were dissolved in anhydrous EtOH (80 mL).
This solution was heated to reflux over 3 h and thencooled to ambi-
ent temperature. NaBH4 (0.384 g, 10.2 mmol) was then added in
small portions over 15 min and the resulting solution was refluxed
for 5 min. 2 N aqueous HCl was then added until the end of gas-
eous release. EtOH was evaporated and CH2Cl2 (30 mL) was
added. The pH of the aqueous phase was brought to 9.5 by ad-
dition of a 2  aqueous NaOH solution. The organic phase was
decanted and the aqueous phase was further extracted 3 times with
small portions of CH2Cl2. The joint organic phases were dried
(Na2SO4) and the solvents evaporated. PIH (1.2 g, yield 68%) was
N 13.37, P 4.66. IR (KBr; strong bands only): ν = 3438 and 3339
˜
(νNH), 1595, 1507, 1477, 1268 (νCOphenol), 841 (PF6), 779, 767 (δim),
557 (PF6) cm–1. Among the crystals obtained a few specimens were
chosen and tested for diffraction. All of them were very small and
badly twinned. The best crystal found (0.1×0.1×0.3 mm) was
mounted on a Philips diffractometer with Mo-Kα radiation (λ =
0.7107 Å). Tabulated atomic scattering factors were taken from
ref.[66] The structure was solved by direct methods using SIR97[67]
and refined using SHELXL-97[68] on F2 by full-matrix least squares
with anisotropic displacement parameters for all atoms but the ace-
tonitrile molecule and the hydrogen atoms. All hydrogen atoms
were calculated in ideal geometrical positions. Because of its very
small size the crystal used for structure characterization gave ex-
tremely weak diffraction. The final refinement suffers therefore
from a very low ratio between the parameters to refine and the
number of observed reflections. Final refinement gave R1 = 0.117
and wR2 = 0.36. Experimental details for the X-ray data collection
are reported in Table 6. The refinement procedure was carried out
using the WinGX package[69] with the program PARST[70] for the
geometrical description of the structures and ORTEP[71] and
PLUTO[72] for the structure drawings.
1
obtained as an orange oil. H NMR (360 MHz, CDCl3): δ = 3.45
(s, 3 H, NCH3), 3.72 (s, 2 H, CH2), 3.87 (s, 2 H, CH2), 7.01 (t, J
= 8 Hz, 1 H, Ar), 6.94 (d, J = 7.3 Hz, 1 H, Ar), 6.87 (br s, 1 H,
Ar), 6.74 (br. m, 3 H, Ar) ppm. 13C NMR (90 MHz, CDCl3): δ =
32.3 (NimCH3), 43.3 and 51.0 (NimCH2 and NArCH2), 116.0, 118.9,
121.1, 122.8, 126.7, 128.5, 128.9, 145.2 (phenol and imidazole),
157.5 (CphenolOH) ppm. IR (KBr; strong bands only): ν = 1456 and
˜
1489 (νim), 1255 (νphenol), 756 (δim) cm–1. ESI-MS: m/z (%) = 218.1
(100) [M + H].
Mn(TMIMA)2(PF6)2 (1): An MnBr2 solution in distilled water
(125 mg in 7 mL) was added after deoxygenation (argon) to a hot
(40 °C) deoxygenated methanolic solution of TMIMA (150 mg in
7 mL). After 1.5 h at 40 °C, it was allowed to cool to room temp.
A deoxygenated solution of ammonium hexafluorophosphate
(288 mg) in MeOH/H2O (5 mL, 1:1) was then added. Crystals suit-
able for X-ray diffraction were grown by slow concentration over
24 h. Yield: 50%. C30H42F12MnN14P2 (943.6): calcd. C 38.19, H
4.49, F 24.16, Mn 5.82, N 20.78, P 6.56; found C 38.00, H 4.39, F
CCDC-245983 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Reactivity in Anhydrous DMSO with Superoxide: The protocol for
studying the reactivity with MnII complexes was the same as that
previously published (preparation of superoxide solutions, spectro-
photometric titration).[1] Several aliquots were prepared with a final
volume of 1 mL, containing the complex and n equiv. of superox-
ide. After the superoxide was added, the aliquot was manually
stirred. 100 µL was introduced in an EPR tube which was immedi-
ately frozen in an EtOH/N2(liq) bath and then kept in N2(liq). EPR
spectra were then recorded. For each experiment, the superoxide
was quantified by EPR from the intensity of the gxy = 2.00 peak
and a linear correlation with concentration was found (see Sup-
porting information). This was used to quantify the remaining su-
peroxide within each EPR tube.
24.60, Mn 5.83, N 20.70, P 6.48. IR (KBr): ν = 1503 (νim), 839
˜
(PF6 ), 741 (δim), 556 (PF6 ) cm–1. Crystals of [Mn(TMIMA)2]-
(PF6)2 (1) were obtained from a water/methanol solution as pris-
matic crystals. One of these, the size of which was
0.3×0.6×0.6 mm, was mounted on a SMART 1000 Bruker AXS
diffractometer with Mo-Kα radiation (λ = 0.7107 Å). Atomic scat-
tering factors were taken from ref.[66] The structure was solved by
direct methods using SIR97[67] and refined using SHELXL-97[68]
on F2 by full-matrix least squares with anisotropic displacement
parameters for all non-hydrogen atoms. All hydrogen atoms were
calculated in ideal geometrical positions. Final refinement gave R1
= 0.0348 and wR2 = 0.0984. Experimental details for the X-ray
data collection are reported in Table 6. The refinement procedure
was carried out using the WinGX package[69] with the program
PARST[70] for the geometrical description of the structures and
ORTEP[71] and PLUTO[72] for the structure drawings.
–
–
Cyclic Voltammetry Experiments: Cyclic voltammetry experiments
were performed at room temperature with an argon stream in a
collidine buffer (pH = 7.5; 50 mmolL–1). The working electrode
was a vitreous carbon disk. The reference electrode was an SCE,
saturated with KCl. Concentrations of the complexes were
2×10–3 molL–1 and 6×10–4 molL–1 on the basis of MnII content.
McCord–Fridovich Assay or Xanthine-Xanthine Oxidase-Cyto-
chrome c Assay: The superoxide anion was supplied to the system
from the xanthine-xanthine oxidase reaction.
[Mn(PI)2](CH3CN)(PF6) (5): A deoxygenated aqueous solution
(8 mL) of MnBr2 (124 mg, 0.58 mmol) was added to a deoxy-
genated methanolic solution (8 mL) of PIH (0.70 mmol). Et3N
(0.025 g, 0.25 mmol) was added. The resulting solution was allowed
to stand for 1.5 h at ambient temperature under positive argon
Reliability of the McCord–Fridovich Assay: To check that the tested
compounds do not inhibit the production of superoxide by xan-
thine oxidase, the rate of conversion of xanthine to urate (see be-
3520
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3513–3523

Reactivity and SOD-Like Activity Correlated to MnII/MnIII Redox Potentials
FULL PAPER
Table 6. Crystal data for complexes [Mn(TMIMA)2](PF6)2 (1) and [Mn(PI)2](PF6)(CH3CN) (5).
1
5
Empirical formula
Formula mass
Crystal system
Space group
C30H42F12MnN14P2
C26H31MF6n1N7O2P
673.5
943.622
monoclinic
P21/c
triclinic
¯
P1
a [Å]
b [Å]
c [Å]
α [°]
12.295(2)
13.879(3)
12.720(2)
90
109.43(2)
90
11.815(11)
13.318(13)
10.770(9)
94.91(5)
112.79(7)
80.84(7)
1542(3)
2
β [°]
γ [°]
V [Å3]
2129.4(8)
2
Z
F(000)
966
1.531
0.4987
0.4×0.6×0.6
293
692
1.45
0.71069
0.1×0.1×0.3
293
Dcalcd. [mg m–3]
µ(Mo-Kα) [mm–1]
Crystal size [mm]
T [K]
No. of reflections measured
θ range [°]
hkl ranges
18649
5409
2.00 Ͻ 2θ Ͻ 50.00
–13, 13; –15, 15; –14, 14
2950
2381
323
0.0348
0.0523, 0.54
0.0984
3.00 Ͻ 2θ Ͻ 60.00
–14,12; –15, 15; 0, 12
5409
887
372
0.117
0.1122, 0.00
0.368
No. of unique reflections
No. of reflections observed [I Ͼ 2σ(I)]
No. of parameters
[a]
R1
Weight (A, B)[b]
[c]
wR2
Max. Fourier difference [e/Å3]
0.28, –0.16
0.56, –0.64
2
2
2 2
[a] R1 = Σ||Fo| – |Fc||/Σ|Fo|. [b] 1/σ2(Fo2) + (A·P)2 + B·P] where P = [Max(Fo2,0) + 2·Fc ]/3. [c] wR2 = {Σ[w(Fo – Fc ) /Σ[w(Fo2)2]}1/2
.
low) was determined by measuring the increase in absorbance at
290 nm over a 2-min period with and without the tested com-
pounds. At concentrations higher than the IC50 value, no inhibi-
tion of this conversion was recorded. We also checked that ferric-
ytochrome c was stable in the presence of the putative SOD mimics.
Pulse Radiolysis Experiments: The pulse radiolysis experiment was
carried out using a 10 MeV Titan β linear accelerator at the Labo-
ratory of Radiolysis, CEA Saclay. Pulses can last between 5 ns and
1 µs, and the frequency between two pulses is adjustable. A 10 ns
pulse corresponds to a 20 Gray dose. Solutions saturated with oxy-
gen are driven in a quartz circulating microcell (spectral path length
1 cm, volume 400 µL) at a constant flux (1 mLmin–1). The spectro-
scopic and temporal analysis is mediated by an optical fiber, linked
to a monochromator, then to a photomultiplier and lastly to a Tek-
tronix numerical oscilloscope allowing the observation of the tem-
poral evolution of absorbance. Absorbance was averaged over 100
pulses. It provides an improvement of the signal/noise ratio, by
comparison with generally used single-pulse experiments. The
amount of superoxide generated during pulse radiolysis was estab-
Xanthine to Urate Assay: To measure the rate of conversion of xan-
thine to urate, xanthine oxidase (30 µL of 0.77 UmL–1 XO) was
added to a solution of potassium phosphate buffer (pH = 7.8;
50 mmolL–1) containing xanthine (150 µmolL–1) at a final volume
of 1.0 mL at 25 °C. Urate production was monitored at 290 nm.
No difference in the slope was recorded with or without the puta-
tive SOD mimics.
Reduction of Ferricytochrome c: Activities were measured using fer-
ricytochrome c reduction.[1,31,33,73] The assay was performed at
25 °C in 3 mL of reaction buffer (50 mmolL–1 potassium phos-
phate buffer, pH = 7.8) containing ferricytochrome c (22 µmolL–1),
xanthine (200 µ), and an amount of xanthine oxidase such as to
give a rate of ∆OD550nm = 0.025 min–1 (about 0.01 UmL–1) in the
absence of a putative SOD mimic. A reduction of ferricytochrome
c was monitored at 550 nm. After 2 min, different amounts of the
putative SOD mimic were added. Rates were linear for at least
8 min. Both rates in the absence and in the presence of the complex
were determined for each concentration of complex added. The
IC50 value represents the concentration of putative-SOD mimic
that induces a 50% inhibition of the reduction of cytochrome c. If
s1 is the slope before addition of the putative SOD mimic and s2 is
the slope after addition of the putative SOD mimic, the inhibition
percentage is given by I (%) = (s1 – s2)/s1×100. For an IC50 value
determination, I is measured for several concentrations in the range
of the IC50. IC50 is obtained for I = 50%; (s1 – s2)/s2 was also
plotted against the complex concentration, providing a linear corre-
lation. IC50 is then obtained for (s1 – s2)/s2 = 1.[34]
lished by the absorbance value at 260 nm, assuming that ε260
=
1940 molL–1 cm–1[54] All measurements were carried out in propan-
2-ol (100 mmolL–1) in phosphate buffer (5 mmolL–1) at pH = 7.8
and 25 °C. Various amounts of complexes MnIPG or MnBIG were
dissolved in that buffer. Solutions were oxygen-saturated by bub-
bling O2 for at least 60 min. The electron pulse lasted 1 µs, generat-
ing a 7.3×10–5 molL–1 superoxide solution. The irradiation fre-
quency was 1 Hz, so that the cell content was totally renewed be-
tween two pulses. The disappearance of superoxide was monitored
at 270 nm (ε270 = 1479 molL–1 cm–1 at pH = 7.8 and 23 °C[53]).
Rates were derived assuming, either second-order kinetics corre-
sponding to the auto-dismutation of superoxide [OD = OD0/(1 +
kobs OD0t)], or a pseudo-first order kinetics in the presence of com-
plexes {OD = OD0[exp(–kobst)] + B, kobs = kcat[complex] + A}.
Simulations of kinetic traces are performed using Gepasi3 (auto-
dismutation) or Kaleidagraph (complex catalyzed dismutation).
Supporting Information: See also the footnote on the first page of
this article. A. Reference for statistical analysis of the Mn–N dis-
Eur. J. Inorg. Chem. 2005, 3513–3523
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3521

C. Policar et al.
FULL PAPER
[22] J. Stein, J. P. Fackler, G. J. Mcclune, J. A. Fee, L. T. Chan, In-
org. Chem. 1979, 18, 3511–3518.
tances; B. additional crystallographic data for 1; C. EPR quantifi-
cation of the superoxide content.
[23] I. Batinic-Haberle, I. Spasojevic, P. Hambright, L. Benov, A. L.
Crumbliss, I. Fridovich, Inorg. Chem. 1999, 38, 4011–4022.
[24] J.-L. Pierre, Regard sur la Biochimie 2000, 17–20.
[25] E. A. Lewis, J. R. Lindsay-Smith, P. H. Walton, S. J. Archibald,
S. P. Foxo, G. M. P. Giblin, J. Chem. Soc., Dalton Trans. 2001,
1159_1161.
[26] M. G. B. Drew, C. J. Harding, V. Mckee, G. G. Morgan, J. Nel-
son, J. Chem. Soc. Chem. Commun. 1995, 1035–1038.
[27] F. C. Frederick, W. M. Coleman, L. T. Taylor, Inorg. Chem.
1983, 22, 792–795.
[28] A. Deroche, I. Morgenstern-Badarau, M. Cesario, J. Guilhem,
B. Keita, L. Nadjo, C. Houée-Levin, J. Am. Chem. Soc. 1996,
118, 4567–4573.
[29] Y. Nishida, N. Tanaka, A. Yamazaki, T. Tokii, N. Hashimoto,
K. Ide, K. Iwasawa, Inorg. Chem. 1995, 34, 3616–3620.
[30] K. Yamato, I. Miyahara, A. Ichimura, K. Hirotsu, Y. Kojima,
H. Sakurai, D. Shiomi, K. Sato, T. Takui, Chem. Lett. 1999,
295–296.
Acknowledgments
We would like to acknowledge the European Community for finan-
cial support (TMR contract FMRX-CT980174).
[1] C. Policar, S. Durot, F. Lambert, M. Cesario, F. Ramiandrasoa,
I. Morgenstern-Badarau, Eur. J. Inorg. Chem. 2001, 1807–1818.
[2] Abbreviations: IPG = N-[(1-methyl-2-imidazolyl)methyl]-N-(2-
pyridylmethyl)glycinate; BIG
azolyl)methyl]glycinate; TMIMA
azolyl)methyl]amine; BMPG = N,N-bis[(6-methyl-2-pyridyl)-
methyl]glycinate; PI 2-{[(1-methyl-2-imidazolyl)methyl]-
amino}phenolate; EPR = electron paramagnetic resonance;
IC50 = concentration that induces a diminution of 50% in the
kinetic of reduction of ferricytochrome c in the McCord–Fri-
dovich assay; SOD = superoxide dismutase; SCE: saturated
calomel electrode; cyt c = Cytochrome c FeIII; NBT: nitro-blue
tetrazolium; X/XO = xanthine/xanthine oxidase. See also
Table 5.
=
N,N-bis[(1-methyl-2-imid-
=
tris[(1-methyl-2-imid-
=
[31] J. M. McCord, I. Fridovich, J. Biol. Chem. 1969, 244, 6049–
6055.
[32] K. M. Faulkner, R. D. Stevens, I. Fridovich, Arch. Biochem.
Biophys. 1994, 310, 341–346.
[33] R. H. Weiss, A. G. Flickinger, W. J. Rivers, M. M. Hardy, K. W.
Aston, U. S. Ryan, D. P. Riley, J. Biol. Chem. 1993, 268, 23049–
23054.
[3] I. Fridovich, Arch. Biochem. Biophys. 1986, 247, 1–11.
[4] I. Fridovich, Ann. Rev. Biochem. 1995, 64, 97–112.
[5] I. Fridovich, J. Biol. Chem. 1997, 272, 18515–18517.
[6] C. X. Zhang, S. J. Lippard, Curr. Opin. Chem. Biol. 2003, 7,
481–489.
[7] D. Salvemini, C. Muscoli, D. P. Riley, S. Cuzzocrea, Pulm.
Pharmacol. Ther. 2002, 15, 439–447.
[8] D. P. Riley, Chem. Rev. 1999, 99, 2573–2587.
[9] I. Batinic-Haberle, I. Spasojevic, R. D. Stevens, P. Hambright,
P. Neta, A. Okado-Matsumoto, I. Fridovich, J. Chem. Soc.,
Dalton Trans. 2004, 1696–1702.
[10] J. R. Anacona, M. Azocar, O. Nusetti, C. Rodriguez-Barbarin,
Transition Met. Chem. 2003, 28, 24–28.
[11] J.-L. Pierre, P. Chautemps, S. Refaif, C. Beguin, A. El Marzuki,
G. Serratrice, E. Saint-Aman, P. Rey, J. Am. Chem. Soc. 1995,
117, 1965–1973.
[12] M. A. Bailey, M. J. Ingram, D. P. Naughton, Biochem. Biophys.
Res. Commun. 2004, 317, 1155–1158.
[13] S. Durot, C. Policar, F. Bisceglie, G. Pelosi, T. Mallah, J.-P.
Mahy, Inorg. Chem. 2003, 42, 8072–8080.
[14] K. J. Oberhausen, R. J. O’Brien, J. F. Richardson, R. M. Bu-
chanan, R. Costa, J.-M. Latour, H.-L. Tsai, D. N. Hendrick-
son, Inorg. Chim. Acta 1990, 173, 145.
[15] K. J. Oberhausen, R. J. O’Brien, J. F. Richardson, R. M. Bu-
chanan, R. Costa, J.-M. Latour, H.-L. Tsai, D. N. Hendrick-
son, Inorg. Chem. 1993, 32, 4561–4565.
[16] P. E. Iberson, H. Lund, Acta Chem. Scand. 1966, 20, 2649.
[17] F. H. Allen, Acta Crystallogr., Sect. B 2002, 58, 380–388.
[18] The six hyperfine lines are broadened because of intermo-
lecular interactions.
[34] Y. Sawada, I. Yamazaki, Biochim. Biophys. Acta 1973, 327,
257–265.
[35] M. Baudry, S. Etienne, A. Bruce, M. Palucki, E. Jacobsen, B.
Malfroy, Biochem. Biophys. Res. Commun. 1993, 192, 964–968.
[36] R. F. Pasternack, B. Halliwell, J. Am. Chem. Soc. 1979, 101,
1026–1031.
[37] J. Butler, W. H. Koppenol, E. Margoliash, J. Biol. Chem. 1982,
257, 10747–10750.
[38] Z.-R. Liao, X.-F. Zheng, B.-S. Luo, L.-R. Shen, D.-F. Li, H.-
L. Liu, W. Zhao, Polyhedron 2001, 20, 2813–2821.
[39] M. C. Rodriguez, Thesis, Université Paris-Sud, 1996, p. 249.
[40] M. C. Rodriguez, I. Morgenstern-Badarau, M. Cesario, J. Gu-
ilhem, B. Keita, L. Nadjo, Inorg. Chem. 1996, 35, 7804–7810.
[41] P. J. Gauuan, M. P. Trova, L. Gregor-Boros, S. B. Bocckino,
J. D. Crapo, B. J. Day, Bioorg. Med. Chem. 2002, 10, 3013–
3021.
[42] N. Kitajima, M. Osawa, N. Tamura, Y. Moro-Oka, T. Hirano,
M. Hirobe, T. Nagano, Inorg. Chem. 1993, 32, 1879–1880.
[43] K. S. Yamaguchi, L. Spencer, D. T. Sawyer, FEBS Lett. 1986,
197, 249–252.
[44] E. A. Lewis, H. H. Khodr, R. C. Hider, J. R. Lindsay-Smith,
P. H. Walton, J. Chem. Soc., Dalton Trans. 2004, 187–188.
[45] D. F. Xiang, C. Y. Duan, X. S. Tan, Q. W. Hang, W. X. Tang,
J. Chem. Soc., Dalton Trans. 1998, 1201–1204.
[46] D. F. Xiang, X. S. Tan, Q. W. Hang, W. X. Tang, B.-M. Wu,
T. C. W. Mak, Inorg. Chim. Acta 1998, 277, 21–25.
[47]
D. Salvemini, Z.-Q. Wang, J. L. Zweier, A. Samouilov, H. Mac-
arthur, T. P. Misko, M. G. Currie, S. Cuzzocrea, J. A. Sikorski,
D. P. Riley, Science 1999, 286, 304–305.
[19] In the case of 3, the formation of a di-µ-oxo–MnIIIMnIV spe-
cies was observed after addition 0.5 equiv. of superoxide. The
di-µ-oxo–MnIIIMnIV species has been shown to be further re-
duced to a di-µ-oxo–MnIIIMnIII species (1 equiv. of superox-
ide). In the case of 4, a similar behaviour was recorded upon
addition of superoxide, but with a different stoichiometry. The
MnIIIMnIV–di-µ-oxo complex was obtained after addition of
1.5 equiv. of superoxide and the EPR-silent MnIIIMnIII–di-µ-
oxo species was obtained after addition of 2 equiv. of superox-
ide. A mechanism involving the oxidation of the starting MnII
complex to an MnIIIMnIV–di-µ-oxo and then a reduction to
the MnIIIMnIII–di-µ-oxo species has been proposed.
[20] W. C. J. Barrette, D. T. Sawyer, J. A. Fee, K. Asada, Biochemis-
try 1983, 22, 624–627.
[48]
I. Spasojevic, I. Batinic-Haberle, R. D. Stevens, P. Hambright,
A. N. Thorpe, J. Grodkowski, P. Neta, I. Fridovich, Inorg.
Chem. 2001, 40, 726–739.
[49]
[50]
[51]
[52]
R. Kachadourian, I. Batinic-Haberle, I. Fridovich, Inorg.
Chem. 1999, 38, 391–396.
K. W. Aston, N. Rath, A. Naik, U. Slomczynska, O. F. Schall,
D. P. Riley, Inorg. Chem. 2001, 40, 1779–1789.
H. Ukeda, S. Maeda, T. Ishii, M. Sawamura, Anal. Biochem.
1997, 251, 206–209.
P. H. Walton, Department of Chemistry, University of York,
personal communication.
[53]
[54]
B. H. J. Bielski, Photochem. Photobiol. 1978, 28, 645–649.
B. H. J. Bielski, D. E. Cabelli, R. L. Arudi, A. B. Ross, J. Phys.
Chem. Ref. Data 1985, 14, 1041–1100.
[21] W. H. Koppenol, F. Levine, T. L. Hatmaker, J. Epp, J. D. Rush,
Arch. Biochem. Biophys. 1986, 251, 594–599.
3522
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3513–3523

Reactivity and SOD-Like Activity Correlated to MnII/MnIII Redox Potentials
FULL PAPER
[55] B. H. J. Bielski, R. L. Arudi, Anal. Biochem. 1983, 133, 170–
178.
[63] S. Chen, J. F. Richardson, R. M. Buchanan, Inorg. Chem. 1994,
33, 2376–2382.
[64] K. J. Oberhausen, J. F. Richardson, R. M. Buchanan, W.
Pierce, Polyhedron 1989, 8, 659–668.
[56] D. E. Cabelli, B. H. J. Bielski, J. Phys. Chem. 1984, 88, 3111–
3115.
[57] D. E. Cabelli, B. H. J. Bielski, J. Phys. Chem. 1984, 88, 6291–
6294.
[58] A. Deroche, I. Morgenstern-Badarau, M. Cesario, J. Guilhem,
B. Keita, L. Nadjo, C. Houée-Levin, J. Am. Chem. Soc. 1996,
118, 4567–4573.
[65] H. Geneste, M. Hesse, Tetrahedron 1996, 15, 15199.
[66] International Tables for X-ray Crystallography, Kluwer Aca-
demic Publishers, Dordrecht/Boston/London, 1995.
[67] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C.
Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 1999, 32, 115–119.
[68] G. M. Sheldrick, SHELXL-97, University of Göttingen, 1997.
[69] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[70] M. Nardelli, J. Appl. Crystallogr. 1995, 28, 659.
[71] C. K. Johnson, M. N. Burnett, Report ORNL-6895, Oak Ridge
National Laboratory, Oak Ridge, Tennessee, USA, 1996.
[72] A. L. Spek, PLATON, A Multipurpose Crystallographic Tool,
University of Utrecht, 1999.
[59] S. Goldstein, G. Czapski, D. Meyerstein, J. Am. Chem. Soc.
1990, 112, 6489–6492.
[60] In ref. 1[1], two IC50 values are reported for each complex. One
value corresponds to that reported here. The other one, labeled
specific IC50 or IC50spe, was measured in the presence of an
excess of ligand. This was done to estimate the effect of decoor-
dination which could occur at a low concentration. For 3, IC5-
0spe = 2.0 0.2 µmolL–1 (kcatspe = 2.9 0.5 106 mol–1 Ls–1) and
for
4,
IC50spe
=
1.7 0.2 µmolL–1
(kcatspe
=
[73] K. M. Faulkner, S. I. Liochev, I. Fridovich, J. Biol. Chem. 1994,
269, 23471–23476.
3.4 0.4 106 mol–1 Ls–1). These values are closer to the pulse ra-
diolysis values.
Received: October 6, 2004
[61] S. Durot, F. Lambert, J.-P. Renault, C. Policar, Eur. J. Inorg.
Chem. 2005, 2789–2793.
Published Online: August 2, 2005
[62] M. Tamura, Y. Urano, K. Kikuchi, T. Higuchi, M. Hirobe, T.
Nagano, Chem. Pharm. Bull. 2000, 48, 1514–1518.
Eur. J. Inorg. Chem. 2005, 3513–3523
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3523