Metal complexes with functionalised 2,2′-dipicolylamine ligand containing an antioxidant 2,6-di-tert-butylphenol moiety: Synthesis and biological studies

Article abstract of DOI:10.1039/c3dt50160d

A series of Zn, Mn, Fe, Co, and Ni complexes ([MX₂L], X = Cl, OAc) of the novel di-(2-picolyl)amine ligand L with an antioxidant 2,6-di-tert-butylphenol pendant were synthesized and characterized by elemental analysis, IR, multinuclear NMR spectroscopy, MALDI-TOF mass spectrometry and the molecular structures of [ZnCl₂L] and [MnCl₂L] were established by X-ray crystallography. The chemical oxidation of complexes with a 2,6-di-tert-butylphenol fragment to the phenoxyl radicals was studied by means of ESR method. The antioxidant radical scavenging activity of the complexes was measured spectrophotometrically using a DPPH-test and linoleic acid peroxidation. The electron transfer reactions were examined in CUPRAC tests and as the inhibition of an enzymatic reaction involving the generation of a reactive oxygen species (superoxide radical-anion) by xanthine oxidase. The lipoxygenase (LOX) inhibition activity of the studied compounds was evaluated. The in vitro biological experiments were performed by using rat brain homogenates. The role of the phenol fragment and metal was found to be essential in antioxidant activity. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3dt50160d

Dalton
Transactions
View Article Online
View Journal
PAPER
Metal complexes with functionalised 2,2’-
dipicolylamine ligand containing an antioxidant
2,6-di-tert-butylphenol moiety: synthesis and
biological studies†
Cite this: DOI: 10.1039/c3dt50160d
Elena R. Milaeva,*a,b Dmitry B. Shpakovsky,a Yulia A. Gracheva,a Sophia I. Orlova,a
Viktoryia V. Maduar,a Boris N. Tarasevich,a Natalia N. Meleshonkova,a
Ludmila G. Dubovab and Elena F. Shevtsovab
A series of Zn, Mn, Fe, Co, and Ni complexes ([MX2L], X = Cl, OAc) of the novel di-(2-picolyl)amine ligand
L with an antioxidant 2,6-di-tert-butylphenol pendant were synthesized and characterized by elemental
analysis, IR, multinuclear NMR spectroscopy, MALDI-TOF mass spectrometry and the molecular structures
of [ZnCl2L] and [MnCl2L] were established by X-ray crystallography. The chemical oxidation of complexes
with a 2,6-di-tert-butylphenol fragment to the phenoxyl radicals was studied by means of ESR method.
The antioxidant radical scavenging activity of the complexes was measured spectrophotometrically using
a DPPH-test and linoleic acid peroxidation. The electron transfer reactions were examined in CUPRAC
tests and as the inhibition of an enzymatic reaction involving the generation of a reactive oxygen species
(superoxide radical-anion) by xanthine oxidase. The lipoxygenase (LOX) inhibition activity of the studied
compounds was evaluated. The in vitro biological experiments were performed by using rat brain homo-
genates. The role of the phenol fragment and metal was found to be essential in antioxidant activity.
Received 16th January 2013,
Accepted 19th February 2013
DOI: 10.1039/c3dt50160d
www.rsc.org/dalton
sense adenosine triphosphate.9 Later, the fluorophore-linked
zinc complexes were used as sensors for phosphatidylserine-
Introduction
The design of metal-based pharmaceuticals depends on either containing membranes.10 Iron(II) and cobalt(II) complexes con-
the ligand scaffold or the choice of metal ion and its oxidation taining polydentate DPA ligands and their derivatives have
state.1–3 Polydentate ligands containing 2-pyridylmethyl frag- been the subject of considerable interest due to their use as
ments attached to a nitrogen atom contain both s-donor and model systems for biologically important molecules.11,12
p-acceptor binding sites.4–6 The dipicolylamine (DPA) as a
Metal complexes of a sterically hindered N,N,O bis-
ligand was first reported in 1964.7 This ligand is known to (2-picolyl)(2-hydroxy-3,5-di-tert-butylbenzyl)amine were syn-
form stable complexes with numerous metal cations. Zn–DPA thesized as models of several hydrolytic enzymes.6,13,14 The
complexes are popular for molecular recognition.8 The three latest investigations of Cu(II) complexes with ethyl 2-[bis(2-
nitrogens of DPA can coordinate strongly to a Zn2+ cation, with pyridylmethyl)amino]propionate interaction with DNA have
an association constant in water around 107 M−1, leaving one demonstrated the cytotoxicity.15
or perhaps two vacant coordination sites for an anionic guest.
On the other hand, metal complexes containing antioxida-
Furthermore, it is straightforward to incorporate multiple DPA tive 2,6-di-tert-butylphenol groups have shown activity as anti-
units into a single organic scaffold. One of the first examples oxidants, inflammatory agents, and scavengers for reactive
of fluorescent chemosensors consisting of a zinc–dipicolyl- oxygen species, which makes them promising agents in
amine appended anthracene derivative can efficiently bind and complex therapy.16–19 The approach based on the modification
of phenolic antioxidants via the incorporation of metal in their
molecules seems to be a promising one. Previously, for the
series of metalloporphyrins with 2,6-di-tert-butylphenol groups
aDepartment of Chemistry, Lomonosov Moscow State University, Lenin Hill 1-3,
we demonstrated that the antioxidative activity depends
dramatically on the nature of metal ion.20,21
Moscow, 119991, Russian Federation. E-mail: milaeva@org.chem.msu.ru
bInstitute of Physiologically Active Compounds of Russian Academy of Sciences,
Metal complexes containing 2,6-di-tert-butylphenol ligands
Chernogolovka, Moscow Region, 142432, Russian Federation
are also of interest because they might provide a synergetic
†CCDC 888418 and 888419. For crystallographic data in CIF or other electronic
antioxidant effect, acting both as essential organic radical
format see DOI: 10.1039/c3dt50160d
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Scheme 1
scavengers and redox-active agents due to the presence of the
metal binding site.22
Fig. 1 The general view of 2 [ZnCl2L] and 3 [MnCl2L] in representation of
atoms by thermal ellipsoids (p = 50 %). Selected bond lengths (Å): Zn(1)–N(2)
2.1581(14), Zn(1)–N(3) 2.1619(13), Zn(1)–N(1) 2.1955(13), Zn(1)–Cl(2)
2.2658(6), Zn(1)–Cl(1) 2.2712(7), Mn(1)–N(3) 2.2388(11), Mn(1)–N(2) 2.2442(12),
Mn(1)–N(1) 2.2809(11), Mn(1)–Cl(2) 2.3559(4), Mn(1)–Cl(1) 2.3722(4).
Herein, we report the synthesis, structures, redox properties
and in vitro antioxidant effects of a series of new Zn, Mn, Fe,
Co, and Ni complexes containing the [N-(3,5-di-tert-butyl-4-
hydroxybenzyl)-N,N-di-(2-pyridylmethyl)]amine ligand (L or 1)
(Scheme 1). The metal complexes 2–11 could be described as
polytopic agents since their molecules possess both organic
antioxidant phenol groups capable of hydrogen atom transfer-
ring and a transition metal center that is capable of electron
transfer.
characterised by IR, UV-vis, 1H and 13C NMR spectroscopy
(2, 6), MALDI TOF mass spectrometry, and elemental analysis.
Crystal structures
The formation and stability of phenoxyl radicals was illus-
trated by ESR. The biological properties of complexes 2–11
have been studied in model radical reactions (DPPH-test, lino-
leic acid peroxidation), electron transfer processes (CUPRAC
method, generation of superoxide radical-anion in the enzy-
matic reaction of xanthine–xanthine oxidase). In addition to
measuring the antioxidant properties in model reactions, their
in vitro activities in experiments in the lipid peroxidation
process in intact mitochondria and cortex neurons isolated
from rat brain correspondingly were also examined.
The Zn and Mn complexes 2 and 3 crystallized as colorless
prisms in a few minutes after the mixing of reagent solutions
and they were used for crystallographic analysis. The X-ray
diffraction investigation of 2 and 3 has revealed that the com-
plexes are isostructural and crystallize as ethanol solvates. In
both complexes, the metal atom is coordinated by two pyridyl
nitrogen atoms, one amine nitrogen atom and two chloride
ions (Fig. 1). The resulting coordination polyhedron in 2 and 3
can be described as a distorted trigonal bipyramid with N(1),
Cl(1) and Cl(2) atoms in the equatorial plane. The N(2)–M–N(3)
angle is equal to 151.03(5) and 146.88(4)° in [ZnCl2L] and
[MnCl2L], respectively, the conformation of the ligand L in the
both complexes is almost identical.
Results and discussion
Analyzing the crystal packing in 2 and 3 we have found that
the hydroxyl group participates in the weak O–H⋯Cl hydrogen
bond (O⋯Cl is 3.340(1) and 3.340(1) Å in Mn and Zn com-
plexes, respectively) assembling the molecules into centro-
symmetric dimers (Fig. 2). Although in the above dimer the two
C(2)–C(7) aromatic planes of the adjust molecules are parallel,
the smallest C⋯C separation (C(3)⋯C(7) is 3.638(3) and
3.610(3) for Zn and Mn complexes, respectively) is too large to
expect a significant contribution of the stacking interaction.
Syntheses
Ligand. [N-(3,5-Di-tert-butyl-4-hydroxybenzyl)-N,N-di-(2-pyri-
dylmethyl)]amine (L, 1) was first reported in 1981 by Grinstead
to apply during metals extraction, although, the synthesis of
the compound of interest was not described.23 In this study,
the ligand L (1) was synthesized in 67% overall yield by amino-
methylation reaction of 2,6-di-tert-butylphenol by bis[(2-
pyridyl)methyl]amine in ethanol under reflux condition. Two
series of complexes (chlorides and acetates: [MX2L] (X = Cl,
OAc; M = Zn, Mn, Fe, Co, Ni, 2–11) were readily prepared by
mixing the appropriate metal salts and 1 in equimolar ratio in
Vibrational spectroscopy
relevant solvents (EtOH, MeOH, Me2CO, MeCN) at room temp- The solid IR spectrum of ligand 1 shows strong vibrational
erature or under reflux condition (Scheme 1).
bands at 1431, 1592 cm−1, which are assigned to δ vibrations
Compounds 1–11 are crystalline substances stable in the air of aromatic system (pyridine and phenol groups), and
and in solutions. The Fe2+ complexes 4 and 9 were quite 2825–2958 cm−1 ν(CH). The IR spectra of the above complexes
stable, although they were stored under argon atmosphere to in the 2000–400 cm−1 region have mainly a function of finger-
prevent possible iron oxidation. The compounds were fully prints, the absorption bands being too numerous for a
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 1 UV-vis data of the ligand 1 and metal complexes 2–11 (MeOH, 25 °C)
Compound
M
λmax (ε, M−1 cm−1
)
1
2
—
263 (7296), 268 sh; 283 sh
Zn 262.5 (9556); 269.5 sh
7
3
8
Zn 231; 256; 262.5 (10 355); 269.5 sh; 281.5 sh
Mn 232.5 sh; 257.5 sh; 262.5 (8908); 268.5 sh
Mn 232.5 sh; 57.5 sh; 262.5 (8919); 268.5 sh
4
9
Fe
Fe
233 sh; 258 (9401); 281 sh; 385 (1422), 419 sh
262.5 (13 860); 419.5 (1645); 495 (371) sh, 680 sh
5
10
6
Co 262.5 (6956); 269 sh; 281.5 sh; 487, 546 (41), 611 sh
Co 230 (8232); 262 (6236); 403 (127); 520 (60)
Ni
Ni
233.5; 256; 261 (7842); 269 sh; 281 sh; 387; 634 (10)
233.5; 256.5; 261 (6810); 268.5 sh; 385, 600 (11)
11
sh – shoulder.
Fig. 2 The general view of centrosymmetric O–H⋯Cl bonded dimers in 2
[ZnCl2L] and 3 [MnCl2L].
reasonably correct assignment. The broad absorption band is
observed in the IR spectrum of 1 in the 3300–3400 cm−1
region corresponding to the stretching vibrations of the O–H
bond of the hindered phenol associated OH group. After
complex formation in the IR spectra of 2–11 the characteristic
ν(OH) absorption band of the non-associated phenol group
has appeared in the 3600–3680 cm−1 region. The shifts of the
characteristic ν(OH) absorption bands for complexes 2–11 at
2249–2246 cm−1 to a region of higher frequencies when com-
pared with non-coordinated ligands confirm the N-coordi-
nation of the ligand and the redistribution of electron density
in the di-picolylamine complexes. As it is well known for
hindered phenols with donor substituents in the para-position
in non-coordinating and non-polar solvents the phenol group
is not involved in hydrogen bond formation and in the IR
spectra the OH stretching are registered in the region of
Fig. 3 UV spectra for: ligand 1 0.155 mM (1); [NiCl2L] 6 0.265 mM (2);
[Ni(OAc)2L] 11 0.406 mM (3) in CH3OH. The long wave bands are sketched out
in the insertion.
similar phenomenon was observed for the substituted tris(2-
pyridylmethyl)-amine Zn complex.26
3600–3650 cm−1 24
According to elemental analysis data and
.
Electronic spectra
IR spectra of the Co and Ni acetates 10 and 11, in which the
absorption band of the non-associated phenol group appeared
in the 3638–3676 cm−1 region and broad bands at
3200–3500 cm−1 were found as well, the complexes exist as
dimers with one coordinated H2O molecule.
The spectral data (λ and ε) of all the complexes and ligand 1
are presented in Table 1. Metal complexes, except those of Zn
and Mn, are colored in solid state and in solutions. The elec-
tronic absorption spectra of the Ni complexes 6 and 11 are
shown in Fig. 3 as examples. Intensive short-wave absorption
is arising below 230 nm because of π–π transitions in aromatic
systems. The visible bands of the other metal complexes are
NMR spectroscopy
Of the complexes, only Zn complexes 2 and 6 are diamagnetic. assigned to d–d transitions and give information about the
Similar Mn, Fe, Co and Ni complexes were found to be para- stereochemistry in solution.
magnetic and gave broad signals as demonstrated by the NMR
resonances. The spin S in similar transition metal complexes ligand 1 in the region of 263 nm (see Table 1). This band has a
was estimated to be equal to 1.25,26
complicated contour due to the superposition of a few bands
The characteristic signals of the tert-butyl protons and hin- of electron transitions in the phenolic part of a molecule
An intensive band is observed in the electron spectra of
1
1
1
dered phenol group are shifted in the spectra of the Zn com- 1B2u ← A1g and allowed on symmetry transition B1 ← A1 in
plexes in comparison with those of ligand 1, confirming the pyridine fragments. These bands remain in the spectra of
ligand–metal coordination. At room temperature in the 1H all the complexes, however, the molar absorption coefficient
NMR spectrum of chloride 2 the methylene protons attached ε changes from 6244 to 13 860 M−1 cm−1 for different
to pyridyl rings were equivalent and displayed as singlets at compounds.
4.14 ppm. In the case of the acetate complex 7 a pair of broad-
ened signals were seen at δH 4.18 and 4.28 ppm, which Zn complexes
could be doublets from diastereotopic methylene protons. The 257–269 nm no long-wave bands were observed. One could
In solution, for the spectra of Mn complexes 3 and 8 and
2
and 7, in addition to absorption at
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 2 Parameters of ESR spectra of phenoxyl radicals 1’–11’ (a, G; 290 K)
a2H
a2H
Number of lines
in spectrum
Radical (M)
Solvent
g-factor
aN
Phenoxyl fragment
Methylene fragment
1′
Toluene
THF
2.0030
2.0046
2.0035
2.0046
2.0049
2.0046
2.0047
2.0047
2.0045
2.0043
2.0042
1.90
2.99
—
1.90
—
1.82
1.80
—
10.9
8.9
—
11.0
11.0
11.2
11.1
11.1
11.0
—
15
2′ (Zn)
3′ (Mn)
4′ (Fe)
5′ (Co)
6′ (Ni)
7′ (Zn)
8′ (Mn)
9′ (Fe)
10′ (Co)
11′ (Ni)
13
a
Toluene/THF
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
1.82
1.55
1.62
1.82
1.82
1.80
3.11
0.70
15
12
12
15
15
15
3
—
1.90
1.90
1.89
—
3.10
—
3
a Weak broad signal was registered.
suggest that the coordination number for the divalent Zn–
chloride complex is five with distorted trigonal bipyramidal
geometry, as it was found by X-ray crystallography (Fig. 1).
The spectrum of Co complex 5 displays low intensive long
wave bands at 487, 546, 611 nm usually attributable to charac-
4
teristic 4E′′(P) ← A′2(F), 4A′2 (P) ← 4A′2(P) transitions of high
spin d7 Co(II) complexes.26,27 The spectrum of Co complex 10
indicates different absorptions at 403 and 520 nm that demon-
strate the influence of a ligand X on the molecular orbitals of
the complex. The electron spectrum of the Ni complex 6 shows
d–d bands at 381 and 634 nm (3E ← 3B1). The spectrum points
out a five-coordinated metal having a N3Cl2 donor set in a dis-
torted bipyramidal complex. The bands corresponding to d–d
transitions in Ni acetate complex 11 were found at 385 and
596 nm.
The electron spectra of the Fe complexes 4 and 9 show long
waves d–d bands at 385, 419 (4), and 495, 680 nm (9), respecti-
vely. The spectra confirm a distorted octahedral geometry.
ESR spectra
The chemical oxidation of the phenol containing compounds
1–11 was carried out in toluene using lead dioxide in
vacuumed tubes yielding phenoxyl radicals 1′–11′ and it was
monitored by ESR method. The X-band ESR spectra measured
at 293 K show the spin density distribution in the organic
ligand. The isotropic g-value for the radicals 1′–11′ are varied
in the range of 2.0030–2.0049. The parameters of the ESR
spectra of radicals are listed in Table 2.
Fig. 4 Experimental (toluene, PbO2, 290 K) (a) and simulated (b) ESR spectra
of radical 1’ with parameters g = 2.0030; aN = 1.90 (1N), aH = 10.89 (2H), aH
1.82 (2H); Γ = 0.60 G, L/G = 0.30 (b).
=
The ESR spectra of radicals are multiplets corresponding to
the coupling of the unpaired electron with two equivalent
meta-protons of the phenoxyl ring (1H, S = 1/2), protons of CH2
and amine nitrogen atom (14N, S = 1). The experimental and
simulated ESR spectra (Fig. 4) of radical 1′ exhibit 15 lines
corresponding to the coupling of the unpaired electron with
paramagnetic nuclei 1H and 14N.
The stability of phenoxyl radicals is influenced by the
nature of the metal in the complex. The most intensive signals
were found for more stable radicals formed from ligand 1, Zn
and Fe complexes. The radical 1′ is stable at room temperature
in the absence of dioxygen for several days. The presence of
Co, Mn and Ni ions in complexes leads to other radical pro-
ducts during oxidation by PbO2, probably due to the ligand
destruction.
Previously a number of analogues of phenoxyl radicals gen-
erated in the oxidation of the corresponding substituted benzyl-
amines have been studied.28,29 We observed that most of
the organometallic and coordination compounds based on
2,6-di-tert-butylphenol undergo intramolecular transform-
ations or destruction upon the change of ligand magnetic
state. The stability and the values of the hyperfine splitting
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
constants of these radicals are influenced by the electron-with-
drawing or electron-donating character of the para-substituent
in the phenyl ring. In general, the coordination group between
M and the phenoxyl moiety serves as a mediator of electronic
interactions. The main factors that influence intramolecular
electron transfer (IET) are: (1) conjugation between the phe-
noxyl and metal; (2) the distance between the radical center
and metal; (3) the nature of the paramagnetic metal nucleus;
(4) temperature. The introduction of a transition metal in most
cases increases the stability of the phenoxyl radicals. There are
few examples of stable organometallic phenoxyls with strong
magnetic interaction, among them the strongest interaction
aM 94 G (M = 195Pt) was observed for transition metals (Pt, Pd)
linked through the σ bond to the para-position of phenoxyl.30
The analogues of DPA complexes are porphyrins. The ESR
spectra of the phenoxyl radicals attached to metalloporphyrins
(M = Zn, Fe, Co, Cu, Ni, Pt, Pd) were registered as triplets indi-
cating only coupling between the unpaired electron and the
meta-protons of the phenoxyl ring a2H 1.5–1.8 G.20
Fig. 5 Changes in UV-vis spectrum of DPPH (0.1 M) in the presence of ligand 1
(0.05 M) during 30 min in MeOH, 293 K.
Table 3 The values of EC50 and k in DPPH test (MeOH, 20 °C) and inhibition
data in linoleic acid peroxidation, lipoxygenase (LOX) and Cu2+ reducing activity
in Trolox-equivalents in the presence of 1–11
DPPH test
Inhibition LOX
In the present work the absence of IET between radical and
metal could be explained by the absence of phenol ring conju-
gation with the bis(2-pyridylmethyl) fragment and metal.
The ferrocene unit, for instance, with an Fe(II) center is an
electron-donating group and increases the stability of radicals
as has been shown by ESR.31
EC50
(μM)
k (L mol−1 of linoleic inhibition
N (M)
s−1
)
acida (%)
IC50 (μM)
TEACCUPRAC
1 (L)
2 (Zn)
3 (Mn) 70
4 (Fe)
5 (Co)
6 (Ni)
7 (Zn)
8 (Mn) 40
9 (Fe)
50
>500
1
2
13.5
0.65
54
49
60
29
27
20
46
61
38
48
46
79.8 1.9 0.24 0.01
41.1 3.2 0.07 0.03
33.6 2.1 0.17 0.01
39.0 2.0 0.77 0.03
31.6 1.6 0.07 0.01
29.7 1.8 0.09 0.01
35.2 3.1 0.07 0.03
24.9 2.2 0.18 0.02
26.6 2.0 0.31 0.01
28.7 1.9 0.09 0.01
32.2 2.0 0.08 0.01
15.3
b
100
230
450 10
350
5
5
2.1
0.3
0.7
17.7
4.6
2.3
DPPH radical scavenging activity
6
2
The presence of a 3,5-di-tert-butyl-4-hydroxyphenyl substituent
linked to the DPA ligand in the metal complexes allows the
suggestion that these compounds might display antioxidant
properties. Since the compounds act by the mechanism of
hydrogen atom abstraction from the phenol moiety and
produce quite stable phenoxyl radicals one could expect they
would show a high antioxidant capacity. The scavenging
activity has been studied in the process of hydrogen atom
transfer to the stable free radical 2,2-diphenyl-1-picrylhydrazyl
(DPPH).32 This radical is a useful reagent for the investigation
of the radical scavenging activity of phenols, catecholes, etc.33
The accepted reaction mechanism is abstracting a hydrogen
atom from a phenol donor to give diphenylpicrylhydrazine and
phenoxyl radical. The reaction involves a color change from
violet to yellow, which can be monitored spectrophotometri-
cally by measuring the decrease in absorbance at 517 nm
(Fig. 5).
>500
10 (Co) >500
11 (Ni) >500
0.5
a Concentration of test compound was 2 mM, 20 °C. b The rate of
reaction was too high to be determined distinctly.
curve (Fig. 6, Table 3). For ligand 1 at equimolar ratio with
DPPH 1 : 1 and initial concentration of 0.1 mM k = 13.5 L
mol−1 s−1 was found.
Thus, the rate constants were found to depend on the metal
incorporated in the complex. The most active reducing agent
was complex 4, probably due to the presence of readily oxidiz-
able Fe2+. The acetate anion in 9 complicated the DPPH
reduction. In other metal complexes the influence of the anion
is not well pronounced.
The results of the DPPH test at 20 °C shows that the activity
depends strongly upon the metal in the compounds and in the
Inhibitory activity in linoleic acid peroxidation
series of compounds 2–11 the activity was found to be highest In order to expand the study of the radical scavenging activity
for Mn complexes 3 and 8 (EC50 = 70 2 and 40 2 μM, of compounds 1–11 the radical chain process of peroxidation
respectively). The EC50 value of ligand 1 was found to be of unsaturated fatty acids has been used with the example of
similar (50 1 μM) whereas the activity of other complexes was linoleic acid as a model substrate for lipid peroxidation. The
at least one order lower (Table 3). The EC50 parameter indi- peroxidation has been induced by Fe2+ ions and carried out at
cates the reactivity of a compound toward DPPH, giving 37 °C in the presence of 2 mM of the compounds under inves-
restricted information on the mechanistic outcome of the reac- tigation and dioxygen as an oxidant. The monitoring of
tion. The reaction followed second-order kinetics, the rate con- linoleic acid peroxidation level was performed by the determi-
stants k were obtained from a plot of 1/[DPPH] vs. time. Linear nation of the total concentration of thiobarbituric acid reactive
regression (r2 > 0.99) gave the parameter k as a slope of the substances (TBARS).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
inhibition activity of the studied compounds was estimated in
order to determine the capacity of compounds of decreasing
oxidative stress level. IC50 values for each compound (Table 3)
were calculated graphically using data from experiments in the
presence of different concentrations of compounds 1–11.
All compounds demonstrate IC50 values lower than 50 μM
and can be considered as potential lipoxygenase inhibitors.
There was no significant difference between the inhibitory
activities of chloride or acetate metal complexes. Ligand 1 is
the weakest inhibitor, which proves the key role of the metal in
inhibitory activity.
Superoxide radical anion scavenging activity
Fig. 6 Reaction of DPPH (0.1 M) with 1 (0.1 M), monitored at 517 nm in
MeOH, 293 K. The insertion: plot of 1/[DPPH] vs. time for the reaction of DPPH
and 1 in MeOH.
Reactive oxygen species (ROS) are produced during normal cel-
lular metabolism and assist in many biochemical processes.
The cells protection from ROS at the normal metabolism is
realized by antioxidative defense enzymes such as superoxide
dismutase (SOD) and catalase. The superoxide anion O2˙− may
cause different heavy effects, leading to tissue injury and
inflammation.36 SOD represents the first line of defense
against O2˙−. SOD catalyzes the dismutation of superoxide
anion to O2 and hydrogen peroxide. It plays a protective effect
in the models of several diseases. However, the use of the
native SOD has several restrictions related with low cell per-
meability and short half-life.37 Synthetic therapeutic anti-
oxidants (catalytic oxidants) such as SOD mimics with a broad
range of antioxidant activity can solve the problem of cell
defense system from reactive biological oxidants.38 The super-
oxide radical-anion scavenging activity of the complexes was
determined by NBT assay39 In this method O2˙− generated by
the xanthine–xanthine oxidase (X–XO system) reduces yellow
nitro blue tetrazolium (NBT) to the blue formazan, which is
measured spectrophotometrically at 560 nm. Therefore, com-
pounds can compete with NBT for oxidation of the generated
superoxide radical-anion. Scavengers are able to inhibit the
formation of blue formazan. The kinetic curves of blue forma-
zan formation in the presence of complex 4 are shown in
Fig. 7.
Comparing with the data of the control experiment without
any additives the inhibition percentage was calculated
(Table 3). The complexes of Mn 3 and 8 and ligand 1 show the
activity in the inhibition of radical chain oxidation >50%. It
should be pointed out that the iron complexes 4 and 9 possess
low activity. It may be that the iron site in the molecules of 4
and 9, which is capable of one-electron transfer, acts as a pro-
oxidant under experimental conditions and it can promote
free radical chain oxidation of the unsaturated fatty acid as a
model substrate of lipid peroxidation.
Cu2+ ion reducing (CUPRAC) assay. The one-electron trans-
fer capacity of compounds 1–11 has been examined as the ion
Cu2+ ion reducing ability, the method proposed by Apak et al.
that was used with slight modification.34 The kinetics of
copper(II) reduction by metal complexes was not detected spectro-
photometrically. The results show that there is significant
activity among the studied compounds that have iron com-
plexes only (Table 3). Thus, the iron complexes exhibit strong
electron transfer capacity. The main role of the metal ion is
experimentally proved. To confirm this hypothesis, additional
experiments were provided with the participation of ferrous
salts – chloride and acetate. Results were presented in Trolox-
equivalents (TEAC CUPRAC – Trolox equivalent antioxidant
capacity). The comparison of the obtained TEAC CUPRAC
values allows the conclusion that the ligand does not play any
significant role in electron transfer to the metal ion. Compar-
ing the ability of ferrous salts to transfer electrons (TEAC
CUPRAC for FeCl2 and Fe(OAc)2 are 0.78 and 0.06, respecti-
vely), it is obvious that in the form of chloride the iron
complex reduces the copper ion as effectively as iron salt does
it. Comparison of the other metal complexes shows overall low
capacity for electron transfer and a counter-ion has no signifi-
cant effect on the activity.
All complexes 2–11 demonstrate the ability to scavenge the
O2˙−, with IC50 values in the micromolar range (Table 4).
Lipoxygenase inhibitory assessment. During oxidative stress
lipoxygenase (LOX) can exhibit uncontrolled activity and may
cause cell membrane destruction by phospholipids oxidation.
Thus, lipoxygenase inhibition under pathologic conditions
Fig. 7 Kinetic curves of formazan formation in the system NBT–xanthine–
decreases the total level of oxidative stress.35 The LOX xanthine oxidase in the presence of different concentrations of Fe complex 4.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 4 Superoxide radical-anion scavenging activity of compounds 1–11 in
the xanthine–xanthine oxidase test
Table 5 Effect of complexes (10 μM) on induced LP of RBH, estimated as
TBARS
Inhibition (%)
Level of RBH LP in the
presence of 10 μM of
compound (%)
Inhibition of LP,
induced by Fe3+
N (M)
10 μM
100 μM
IC50 (μM)
Fe3+-induced
LP
tBHP-induced
LP
a
1 (L)
0.70 0.05
13.0 2.2
83.0 0.5
40.0 0.9
2.5 1.4
20.0 1.3
1.7 0.5
85.0 0.7
1.4 1.9
19.0 1.1
2.0 0.5
6.0 1.9
52.0 2.1
98.0 1.1
82.0 0.7
61.0 0.8
31.0 1.7
5.0 2.2
99.0 0.5
2.5 2.4
24.0 1.5
7.0 1.9
Compound
IC50 (μM)
2 (Zn)
3 (Mn)
4 (Fe)
5 (Co)
6 (Ni)
7 (Zn)
8 (Mn)
9 (Fe)
10 (Co)
11 (Ni)
91
2
15
1 (L)
0.4 0.1
3.6 0.1
ndb
13.6 1.2
10.2 0.5
14.1 0.5
12.9 0.9
17.6 1.4
10.2 0.3
10.7 0.5
13.5 0.6
14.1 0.7
17.9 0.6
10.8 0.6
94.8 5.3
86.6 0.7
81
2 (Zn)
3 (Mn)
4 (Fe)
5 (Co)
6 (Ni)
7 (Zn)
8 (Mn)
9 (Fe)
10 (Co)
11 (Ni)
a
73.2 5.9
a
a
ndb
0.5
2.5 0.1
5.1 0.1
1.9 0.1
ndb
65.8 4.8
78.0 12.0
84.1 4.2
a
a
a
74.3 4.0
a
ndb
a Inhibition activity of compound is less than 50%.
2.6 0.1
8.7 0.1
67.4 10.4
85.9 5.4
a Complex with pro-oxidant activity – the peroxidation level is higher
than that of control experiment. b nd – not determined.
processes implicated in oxidative stress include free radical
stages of membrane damage. They include intensification of
lipid peroxidation (LP) of biological membranes, which can be
determined as an increase in the content of TBARS. The
increase of TBARS including lipid hydroperoxides and alde-
hydes in biological samples is a result of oxidative stress. In
practice, TBARS are expressed in terms of malondialdehyde
(MDA) equivalents. The comparative study of activity of a
series of complexes in lipid peroxidation has been carried out
using rat brain homogenates (RBH). We studied lipid peroxi-
dation induced by Fe3+-ions (FeNH4(SO4)2·12H2O) or by tert-
butylhydroperoxide (tBHP). All the experiments were repeated
using four different preparates of RBH. Data are normalized
between the control probe with oxidant (100%) and blank
probe with only diluent (0%). The values are expressed as
mean % SD in Table 5.
Fig. 8 Scavenging activity of the ligand 1 and complexes 2–6 toward the
superoxide radical-anion generated in the X–XO system.
However, the most active inhibitors were found to be chloride
3 (Mn) followed by 4 (Fe) > 5 (Co) > 2 (Zn) > 6 (Ni) > 1 (L) at the
concentration of 100 μM (Fig. 8). The scavenging effect of
other metal acetates on the superoxide radical-anion decreases
in the order: 8 (Mn) ≫ 10 (Co) > 11 (Ni) > 7 (Zn) > 9 (Fe) at the
concentration of 100 μM. In the series of acetates, Mn complex
8 was the best inhibitor.
These results indicate that manganese complexes possess
the highest superoxide radical-anion scavenging activity in a
series of other metal complexes. Moreover, the ligand X also
influences the activity. The chloride complexes are more
effective inhibitors than acetates and have lower IC50 values.
On the contrary, the value of inhibition percentage in the
presence of 100 μM ligand 1 (6.0 1.9%) proves that the ligand
does not possess any significant activity in the interaction with
superoxide radical-anion. Thus, it was found that the nature of
the metal plays a critical role in superoxide scavenging activity
of the complexes.
One can see from Table 5 that all the compounds are highly
effective inhibitors of Fe3+ induced LP. However, they have
shown low activity against tBHP induced LP. It should be
t
noted that Fe complexes significantly increased BHP induced
LP. It seems to be related to oxidation of Fe2+ in the complex to
Fe3+, which could be additional promoters of LP.
Conclusions
The synthesis, structures, redox properties and in vitro antioxi-
dant effects of a series of new Mn, Fe, Co, Ni, Zn, Cu com-
plexes with [N-(3,5-di-tert-butyl-4-hydroxybenzyl)-N,N-di-(2-
pyridylmethyl)]amine is presented. The metal complexes 2–11
are polytopic agents since their molecules possess both
organic antioxidant phenol groups capable of hydrogen atom
Evaluation of complexes influence on lipids peroxidation in rat
brain homogenates
Oxidative stress is one of the major mechanisms implicated in transferring and a transition metal center that is capable of
a variety of diseases; in particular, neurodegenerative diseases, electron transfer.
including stroke, Alzheimer’s disease, and Parkinson’s
The oxidation of the compounds leads to the formation of
disease.40,41 The molecular mechanisms of pathogenetic phenoxyl radicals in the ligand environment and was
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
investigated by ESR. The relatively stable paramagnetic species the 1 M KOH and 10 mL of ether was added, white product
formed in the redox transformations are ligand-centered phe- was repeatedly extracted by a new portion of KOH and ether.
noxyl radicals.
The ether extract was dried over CaCl2. The solvent was par-
The antioxidant radical scavenging activity of complexes tially removed in vacuum and the solution left for crystalliza-
was measured using DPPH-test and linoleic acid peroxidation. tion on the air. White crystals of 1 were washed by a small
The highest activity was found for Mn complexes in both tests. amount of ether and dried in the air. Yield 1.11 g (67%). Mp
In electron transfer reactions (CUPRAC test and inhibition 129–130 °C. C27H35N3O: Anal. calcd: C, 77.66; H 8.45; N 10.06.
of enzymatic generation of superoxide radical-anion by Found: C 77.74; H 8.52; N 9.91%. IR (KBr pellet, cm−1): 1431
xanthine oxidase) Mn complexes possess an activity higher (s), 1593 (s), ν(CH) 2825–3010 (s) ν(OH associated) 3293 (w),
than that of other metal complexes. The LOX inhibition 3511 (w) ν(OH). MS (MALDI TOF), m/z: 440 [C27H35N3ONa]+ (M
activity of the studied compounds was evaluated. All com- + Na)+.
pounds demonstrate IC50 values lower than 50 μM and can be
1H-NMR (δ (ppm), CDCl3): 1.45 (s, 18 H, C(CH3)3); 3.62 (s,
2H, CH2–Ar); 3.85 (s, 4H, N(CH2)2)); 5.14 (s, 1 H, OH); 7.14 (t,
considered as potential lipoxygenase inhibitors.
The in vitro biological experiments, which were performed 2H, 3JHH = 6.7 Hz, 2 Py); 7.21 (s, 2H, C6H2); 7.61–7.70. (m, 4 H,
by using rat brain homogenate, indicate high antioxidant 2 Py); 8.52 (d, 2H, JHH = 5.1 Hz, 2 Py). 13C-NMR: 30.35
3
activity of all the compounds studied. This result opens up the (C(CH3)3); 34.31 (CC(CH3)3); 58.73 (CH2–Ar); 60.12 ((ArCH2)2N–);
possibility for the design of novel biomimetic systems that 121.85; 122.65; 125.63; 129.37; 135.63; 136.33; 148.90; 152.74;
might serve as electron and proton transfer agents in biologi- 160.19 (C-arom). UV-vis spectrum (MeOH), λmax/nm (ε, M−1
cal systems.
cm−1): 263 (7296), 268 sh; 283 sh.
[ZnCl2L] (2). To a solution of 1 (209 mg, 0.5 mmol) in 3 mL
of ethanol, ZnCl2·2H2O (86 mg, 0.5 mmol) in 0.6 mL of
ethanol was added under stirring. The white crystals that
appeared in several seconds were left to crystallize for 1 h. The
crystals were filtered, washed with EtOH and dried in the air.
Yield 220 mg (79%). Mp. >250 °C. C27H35N3OZnCl2: Anal.
calcd: C, 58.55; H 6.37; N 7.59. Found: C 58.29; H 6.33; N
7.40%. IR (KBr pellet, cm−1): 1446 (s), 1604 (s), ν(CH)
2869–2956 (s), ν(OH non-associated) 3618 (w). MS (MALDI
TOF), m/z: 516 [C27H35N3OZnCl]+ (M − Cl)+. 1H-NMR (δ (ppm),
CDCl3): 1.41 (s, 18 H, C(CH3)3); 3.62 (s, 2H, CH2–Ar); 4.14 (s,
4H, N(CH2)2); 5.26 (s, 1 H, OH); 6.80 (s, 2H, Ar); 7.26 (d, 2H,
3JHH = 7.8 Hz, 2 Py); 7.46–7.53 (m, 2 H, Py); 7.86–7.94 (m, 2H,
Py); 9.26 (d, 2H, 3JHH = 5.0 Hz, 2 Py). 13C-NMR: 30.22 (C(CH3)3);
34.23 (CC(CH3)3); 55.32 (CH2–Ar); 55.74 ((ArCH2)2N–); 122.08;
123.22; 124.51; 127.88; 135.95; 136.64; 149.91; 153.68; 153.98
(C-arom). UV-vis spectrum (MeOH), λmax/nm (ε, M−1 cm−1):
262.5 (9556); 269.5 sh. The crystals of 2 were suitable for
single-crystal X-ray structure determination.
[MnCl2L] (3). To a solution of 1 (250 mg, 0.6 mmol) in 1 mL
of MeOH MnCl2·4H2O (118 mg, 0.6 mmol) was added under
stirring. The white crystals that appeared in several seconds
were left to crystallize for 15 min. The crystals were filtered,
washed with MeOH and dried in the air. Yield 279 mg (80%).
Mp. (dec.) 246–250 °C. C27H39N3O3MnCl2: Anal. calcd: C,
55.95; H 6.73; N 7.25. Found: C 56.60; H 6.79; N 7.37%. IR
(KBr pellet, cm−1): 1446 (s), 1604 (s), ν(CH) 2869–2956 (s),
ν(OH associated) 3467–3523 (s), ν(OH non-associated) 3617
(w). MS (MALDI TOF), m/z: 507 [C27H35N3OMnCl]+ (M − Cl −
Experimental
Materials and equipment
The starting 2,6-di-tert-butylphenol, di-(2-picolyl)amine (97%
Aldrich) and metal salts were commercially available and used
as supplied. CDCl3 (Merck) was used without further purifi-
cation; other solvents were routinely distilled prior to use.
Elemental analyses were performed in the Laboratory of Micro-
analysis at the Moscow State Lomonosov University (Moscow),
Russia.
Infrared spectra were recorded in KBr pellets or in CH2Cl2
solution with a “IR200 Thermo Nicolet” spectrometer. Elec-
tronic absorption spectra were measured on a “Evolution 300
Termo Scientific” spectrophotometer.
The NMR measurements were performed with a Bruker
Avance-400 spectrometer operating on 400.1 (1H) and
100.6 MHz (13C) in CDCl3.
The MALDI TOF spectra were acquired using a Autoflex II
(Bruker daltonics) mass spectrometer. The samples were dis-
solved in CHCl3 and put at target.
ESR spectra were recorded with Bruker EMX spectrometer
at X-band frequency (9.8 GHz). The measurements were
carried out after pre-evacuation (10−2 Torr) of tubes with solu-
tions of samples (concentration 1 × 10−4 M). The oxidant PbO2
(Aldrich) was taken in a tenfold excess.
Syntheses
Ligand L (1). To a solution of CH2O in H2O (37%, 0.32 mL, 2H2O)+. UV-vis spectrum (MeOH), λmax/nm (ε, M−1 cm−1):
4.0 mmol) in 1 mL of EtOH, the solution of di-2-picolylamine 232.5 sh; 257.5 sh; 262.5 (8908); 268.5 sh. The crystals of 3
(0.72 mL, 4.0 mmol) in 0.5 mL of EtOH was added, the suspen- were suitable for single-crystal X-ray structure determination.
sion was stirred at 0 °C for 30 min, then 2,6-di-tert-butylphenol
[FeCl2L] (4). To a solution of 1 (210 mg, 0.5 mmol) in 1 mL
(0.82 g, 4.0 mmol) in 2 mL of EtOH was added and stirred at of EtOH, FeCl2 (64 mg, 0.5 mmol) in Ar atmosphere was added
20 °C for 1 h. The resulting mixture was refluxed for 4 h and under stirring. The mixture was stirred for 5 min. The solvent
left overnight. The solution was added to 100 mL of 0.01 M was removed under reduced pressure, the yellow powder was
HCl, and the mixture was washed by ether. To water solution washed with hexane and dried in the air. Yield 246 mg (90%).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Mp. (dec.) 200 °C. C27H35N3OFeCl2: Anal. calcd: C, 59.58; H ν(CH) 2868–3002 (s), ν(OH associated), 3631 (w)ν(OH non-
6.48; N 7.72. Found: C 59.40; H 6.48; N 7.79%. IR (KBr pellet, associated). MS (MALDI TOF), m/z: 514 [C29H38N3O3Mn]+
cm−1): 1434 (s), 1606 (s), ν(CH) 2844–3000 (s), ν(OH non- (M − OAc)+. UV-vis spectrum (MeOH), λmax/nm (ε, M−1 cm−1):
associated) 3667 (s). MS (MALDI TOF), m/z: 508 232.5 sh; 57.5 sh; 262.5 (8919); 268.5 sh.
[C27H39N3O3FeCl]+ (M − Cl)+. UV-vis spectrum (MeOH), λmax
/
[Fe(OAc)2L] (9). The mixture of 1 (210 mg, 0.5 mmol) and
nm (ε, M−1 cm−1): 233 sh; 258 (9401); 281 sh; 385 (1422), 419 Fe(OAc)2 (87 mg, 0.5 mmol) in 3 mL of acetonitrile in Ar
sh. atmosphere was stirred for 1 h. The solvent was removed
[CoCl2L] (5). To a solution of 1 (210 mg, 0.5 mmol) in 2 mL under reduced pressure; the brown residue was washed with
of MeOH, CoCl2·6H2O (119 mg, 0.5 mmol) was added under hexane and dried in the air. Yield 222 mg (75%). Mp. (dec.)
stirring. The purple crystals that appeared in several seconds 150 °C. C31H41N3O5Fe: Anal. calcd: C, 62.95; H 6.99; N 7.10.
were left to crystallize for 15 min. The crystals were filtered, Found: C 63.14; H 6.96; N 7.25%. IR (KBr pellet, cm−1): 1433
washed with MeOH and dried in the air. Yield 165 mg (60%). (s), 1556 (s), ν(COCH3) 1587 (s), ν(CH) 2856–3002 (w), ν(OH
Mp. (dec.) >250 °C. C27H35N3OCoCl2: Anal. calcd: C, 59.24; H associated) 3428-(wide, w), ν(OH non-associated) 3627 (w). MS
6.44; N 7.68. Found: C 59.12; H 6.67; N 7.51%. IR (KBr pellet, (MALDI TOF), m/z: 532 [C29H38N3O3Fe]+ (M − OAc)+. UV-vis
cm−1): 1434 (s), 1608 (s), ν(CH) 2844–3000 (s), ν(OH non- spectrum (MeOH), λmax/nm (ε, M−1 cm−1): 262.5 (13 860);
associated) 3637 (s). MS (MALDI TOF), m/z: 511 419.5 (1645); 495 (371) sh, 680 sh.
[C27H35N3OCoCl]+ (M − Cl)+. UV-vis spectrum (MeOH), λmax
nm (ε, M−1 cm−1): 262.5 (6956); 269 sh; 281.5 sh; 487, 546 (41), 5 mL of acetone, Co(OAc)2·4H2O (124 mg, 0.5 mmol) was
611. added and refluxed for 1 h. The pink crystals were filtered,
/
[Co(OAc)2L] (10). To a solution of 1 (210 mg, 0.5 mmol) in
[NiCl2L] (6). The mixture of 1 (210 mg, 0.5 mmol) and washed with acetone and dried in the air. Yield 175 mg (59%).
NiCl2·6H2O (119 mg, 0.5 mmol) in 5 mL of acetone was Mp. (dec.) 205 °C. C31H41N3O5Co·0.5 H2O: Anal. calcd: C,
refluxed for 1 h. The light green crystals were left to crystallize 61.69; H 7.01; N 6.96. Found: C 61.39; H 7.19; N 6.74%. IR
for 15 min at room temperature. The crystals were filtered, (KBr pellet, cm−1): 1446 (s), ν(COCH3) 1594 (s), ν(CH)
washed with acetone and dried in the air. Yield 207 mg (75%). 2867–3002 (s), ν(OH associated) 3338–3450 (s), ν(OH non-
Mp. (dec.) >255 °C. C27H35N3ONiCl2: Anal. calcd: C, 59.21; H associated). MS (MALDI TOF), m/z: 535 [C29H38N3O3Co]+
6.39; N 7.67. Found: C 59.32; H 6.64; N 7.42%. IR (KBr pellet, (M − OAc)+. UV-vis spectrum (MeOH), λmax/nm (ε, M−1 cm−1):
cm−1): 1434 (s), 1608 (s), 1702 (s), ν(CH) 2848–2998 (s), ν(OH 230 (8232); 262 (6236); 403 (127); 520 (60).
non-associated) 3635 (w). MS (MALDI TOF), m/z: 510
[C27H35N3ONiCl]+ (M − Cl)+. UV-vis spectrum (MeOH), λmax
[Ni(OAc)2L] (11). To a solution of 1 (210 mg, 0.5 mmol) in
/
6 mL of acetone, Ni(OAc)2·4H2O (125 mg, 0.5 mmol) was
nm (ε, M−1 cm−1): 233.5; 256; 261 (7842); 269 sh; 281 sh; 387; added and refluxed for 1 h. The pale blue crystals were filtered,
634 (10).
washed with acetone and dried in the air. Yield 205 mg (69%).
[Zn(OAc)2L] (7). The mixture of 1 (209 mg, 0.5 mmol) and Mp. (dec.) >250 °C. C31H41N3O5Ni·0.5 H2O: Anal. calcd: C,
Zn(OAc)2·2H2O (110 mg, 0.5 mmol) in 3 mL of acetone was 61.71; H 7.02; N 6.96. Found: C 61.41; H 7.14; N 6.82%. IR
refluxed for 1 h. The white crystals were filtered, washed with (KBr pellet, cm−1): 1448 (s), 1583 (s), 1606 (s), ν(CH) 2869–3004
acetone and dried in the air. Yield 220 mg (73%). Mp. >235 °C. (s), ν(OH associated) 3300–3500 (s), ν(OH non-associated) 3629
C31H41N3O5Zn: Anal. calcd: C, 61.94; H 6.88; N 6.99. Found: C (w), 3677 (w). MS (MALDI TOF), m/z: 534 [C29H38N3O3Ni]+
61.77; H 6.83; N 7.05%. IR (KBr pellet, cm−1): 1432 (s), 1577 (M − OAc)+. UV-vis spectrum (MeOH), λmax/nm (ε, M−1 cm−1):
(s), ν(CH) 2867–3002 (s), ν(OH non-associated) 3631 (w). MS 233.5; 256.5; 261 (6810); 268.5 sh; 385, 600 (11).
(MALDI TOF), m/z: 540 [C29H38N3OZn]+ (M − OAc)+. 1H-NMR
Crystallographic data collection and structure determination
(δ (ppm), CDCl3): 1.42 (s, 18 H, C(CH3)3); 2.00 (s, 6 H, CH3CO);
3.60 (s, 2H, CH2–Ar); 4.18 (s, 2H, NCH2); 4.27 (s, 2H, NCH2); Single crystals were grown from EtOH solution. All diffraction
5.43 (s, 1 H, OH); 6.75 (s, 2H, Ar); 7.28 (m, 2H, Py); 7.48 (m, data were collected on a Bruker SMART APEX II CCD diffracto-
2H, Py); 7.89 (m, 2H, Py); 8.90 (m, 2H, Py). 13C-NMR: 23.31 meter [λ(MoKα) = 0.71072 Å, ω-scans] at 100 K. The substantial
(CH3CO); 30.30 (C(CH3)3); 34.28 (CC(CH3)3); 55.17 (CH2–Ar); redundancy in data allows empirical absorption correction to
55.81 ((ArCH2)2N–); 122.08; 123.07; 124.45; 128.17; 136.06; be performed with SADABS, using multiple measurements of
139.26; 149.45; 154.01, 154.55; (C-arom); 182.22 (CvO). UV-vis equivalent reflections. The structures were solved by direct
spectrum (MeOH), λmax/nm (ε, M−1 cm−1): 231; 256; 262.5 methods and refined by the full-matrix least-squares technique
(10355); 269.5 sh; 281.5 sh.
against F2 in the anisotropic–isotropic approximation. The
[Mn(OAc)2L] (8). To a solution of 1 (210 mg, 0.5 mmol) in hydrogen atoms were located from difference electron density
3 mL of acetone, Mn(OAc)2·4H2O (122 mg, 0.5 mmol) was syntheses and refined in a rigid body model. Analysis of
added under stirring, and the mixture was refluxed for 1 h. Fourier maps in both structures allowed one to reveal about
The colorless crystals, after cooling to room temperature, were 8–10 peaks of residual electron density (1.2–3.1% e × A−3) in
filtered off, washed with acetone and dried in the air. Yield cavities between complexes. These peaks can be considered as
266 mg (85%). Mp. (dec.) 180–185 °C. C31H41N3O5Mn: Anal. the superposition of several C2H5OH molecules with different
calcd: C, 63.05; H 7.00; N 7.12. Found: C 63.16; H 7.12; occupation values. We could not establish an appropriate
N 7.16%. IR (KBr pellet, cm−1): 1433 (s), 1577 (s), 1604 (s), model for refining the above residual electron density. Thus,
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Linoleic acid peroxidation
Table 6 Crystal data and structure refinement parameters for 2, 3
The effect of compounds on the linoleic acid peroxidation was
estimated according to a known procedure.45 Phosphate buffer
(2 mL, pH 7.4) was added to a solution of linoleic acid
(0.034 mL) in ethanol (1.97 mL). The reaction mixture con-
tained a solution of linoleic acid (0.33 mL), a solution of FeCl2
(0.066 mL, 4 mM), phosphate buffer (0.164 mL), and a solution
of the compound under study in ethanol (1 mL, 30 mM). The
resulting solution was incubated for 30 min at 37 °C, then a
solution (0.84 mL) containing 0.25 M HCl, 10% trichloroacetic
acid and 0.6% thiobarbituric acid was added. The mixture was
heated at 95 °C for 15 min, centrifuged and the absorption at
λmax 532 nm was measured on a 96 well plate spectro-
photometer.
Complex
[ZnCl2L] 2
29H41Cl2ZnN3O2
599.92
100
Monoclinic
P21/n
14.579(3)
13.032(3)
16.021(3)
106.13(3)
2924.1(10)
4 (1)
1264
1.363
10.53
58
[MnCl2L] 3
Empirical formula
Fw
Temp [K]
Cryst syst
Space group
a (Å)
C
C29H41Cl2MnN3O2
589.49
100
Monoclinic
P21/n
14.9561(15)
13.1472(14)
16.190(2)
105.619(3)
3065.8(6)
4 (1)
b (Å)
c (Å)
β (°)
V (Å3)
Z (Z′)
F(000)
1244
1.277
6.34
58
dcalc (g cm−1
)
μ (cm−1
2θmax (°)
)
Reflns collected
Independent reflns (Rint
Observed reflns [I > 2σ(I)]
R1 (obs. refl.)
34536
7784 (0.0408)
6335
0.0299
0.0777
0.964
0.549/−0.509
34470
8154 (0.0345)
6887
0.0322
0.0860
1.054
0.471/−0.331
Lipoxygenase activity
)
Lipoxygenase (LOX) type 1-B from Glycine max (soybean), boric
acid, linoleic acid, ammonium acetate, copper(II) chloride,
ethanol (96%) and Trolox were purchased from Sigma-Aldrich
and were used with no further purification.
wR2 (all data)
GOF on F2
Δρmax/Δρmin (e Å−3
)
LOX inhibition activity was determined spectrophotometri-
cally by measuring the increase in absorbance at 234 nm for
the oxidation of linoleic acid.46 The reaction mixture contained
the test compounds dissolved in 10 μL of DMSO at initial con-
centrations of 1.5–30 mM (final concentrations 5, 10, 20, 30,
50, 100 μM) or the 10 μL of DMSO (blank) and 2 mL of 0.3 mM
linoleic acid in borate buffer (pH = 9.0) and 0.89 mL of borate
buffer. The total sample volume was 3 mL, the final concen-
tration of DMSO was 0.33% v/v. The reaction was started by
adding 100 μL of lipoxygenase solution (500 units) in borate
buffer. The increase in absorbance was recorded every 1 s
during 10 min under controlled temperature 25 °C.
we had to exclude unresolved solvent by means of SQUEEZE
procedure.42 It is most likely that the sum of the occupation
values of the disordered C2H5OH molecule is equal to 1. Thus,
we add one molecule of ethanol to brutto and moiety for-
mulas. All calculations were performed with the SHELXTL soft-
ware package.43 Crystal data and structure refinement
parameters are listed in Table 6.
DPPH radical scavenging activity
The free radical scavenging activity was evaluated using the
stable radical DPPH, according to the method described by
Brand-Williams et al. with a slight modification.44 For each
compound a 1 : 1 ratio expressed as moles of compound per
mole of DPPH radical was tested. A 1 mL sample of the com-
pound solution in methanol was added to 1 mL of DPPH solu-
tion in methanol so that the initial DPPH concentration in the
cuvettes was 0.1 mM. The samples were incubated for 30 min
at 20 °C in methanol and the decrease in the absorbance
values of the DPPH solution were measured at λmax 517 nm on
the Zenyth 200RT Anthos microplate spectrophotometer.
Results were expressed as scavenging activity calculated as
follows:
The degree of LOX activity (A%) in the presence of the com-
plexes was calculated according to the following.47
A, % = (ν0 in the presence of inhibitor/ν0 in the absence of
inhibitor) × 100, where ν0 is the initial rate.
The value of the initial rate (ν0, μM min−1) was calculated
according to formula:
ν0 ¼ ΔC=Δt ¼ ΔA=Δtε ¼ tgα=Δtε
where C is the product concentration (hydroperoxy-linoleic
acid), t is the reaction time, ε is molar absorbance coefficient
of hydroperoxy-linoleic acid, tgα is the slope of the kinetic
curve plotted as absorbance vs. time.
IC50 values were obtained graphically by plotting a logistic
curve in coordinates inhibition activity (A, %) vs. inhibitor con-
centration ([I], μM). The general curve equation is the follow-
ing:
Scavenging activity; % ¼ ½ðA0 ꢀ A1Þ=A0ꢁ ꢂ 100
The concentration of compound needed to decrease 50% of
the initial substrate concentration (EC50) is a parameter widely
used to measure the antioxidant effect.33 For determination of
the EC50 the values of DPPH decrease after 30 min were used.
The EC50 values were calculated graphically by plotting scaven-
A; % ¼ 100 ꢂ ð1=ð1 þ ½Iꢁ=IC50ÞÞ
CUPRAC assay (Cu2+ reducing)
ging activity against compound concentration. Different Neocuproine (2,9-dimethyl-1,10-phenanthroline) was supplied
sample concentrations (0.01, 0.02, 0.05 and 0.1 mM) were used from Sigma Chemical Co. The method proposed by Apak et al.
in order to obtain kinetic curves and to calculate the EC50 was used with slight modification.34 For this reason 0.05 mL
values.
of CuCl2 solution (0.01 M), 0.05 mL of ethanol neocuproine
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
solution (7.5 mM) and 0.25 mL of ammonium acetate buffer heated at 95 °C for 15 min, cooling at 4 °C. Then probes were
solution (1 M) were added to a test tube, followed by mixing centrifuged (10 min at 10 000 g) and the supernatants trans-
with different concentrations of tested compounds ferred into 96-well plate and absorbance was measured at
(10–100 μM). The mixtures were kept at room temperature. 530–620 nm at the Wallac Victor 1420 Multilabel Counter
Absorbance was measured at 450 nm against a reagent blank (PerkinElmer Wallac). All the experiments were performed
30 min later. The increase of reaction mixture absorbance in using four independent experiments with different brain hom-
comparison with control indicates the reduction capability of ogenate preparations. Data are normalized to a control probe
the test compound.
with oxidant as 100% and a blank probe with diluent but
Results were presented in Trolox-equivalents. TEACCUPRAC without oxidizing agent. Preliminary experiments were done
values were obtained graphically using absorbance data and and the absence of compounds interaction with thiobarbituric
linear calibration curve plotted as absorbance vs. Trolox acid was seen. The values are expressed as mean % SD. The
concentration.
concentrations of ferrocenes giving half-maximal inhibition
(IC50) of lipid peroxidation were determined by dose-effect
analysis in Origin 6.1 (Origin Lab Corporation).
Inhibition of superoxide radical anion formation by xanthine
oxidase (NBT assay)
EDTA, xanthine, bovine serum albumin, nitroblue tetrazolium,
and xanthine oxidase (0.04 MU) were purchased from Sigma-
Aldrich Chemical Co. The superoxide anion was generated
Acknowledgements
enzymatically by the xanthine oxidase system. The reaction We greatly thank Prof. Konstantin A. Lyssenko from Nesmeya-
mixture consisted of 2.70 mL of 40 mM sodium carbonate nov Institute of Organoelement Compounds of RAS for the
buffer containing 0.1 mM EDTA (pH 10.0), 0.06 mL of 10 mM experimental X-ray structural investigations of complexes and
xanthine, 0.03 mL of 0.5% bovine serum albumin, 0.03 mL of valuable discussions. Also the financial support from Russian
2.5 mM nitroblue tetrazolium and 0.06 mL of the sample solu- Foundation for Basic Research (Grants No 11-03-01165, 11-03-
tion in DMSO. To the mixture at 25 °C, 0.12 mL of xanthine 01134, 12-03-00776) is gratefully acknowledged.
oxidase (0.04 units) was added, and the absorbance at 560 nm
(by formation of blue formazan)48 was recorded by microplate
spectrophotometer Zenyth 200rt Anthos for 60 s. A control
experiment was carried out by replacing the sample solution
References
with the same amount of DMSO. Inhibition I (%) = [(1 − Ai/A0)
× 100%], where Ai is the absorbance in the presence of the
testing compound, A0 is the absorbance of the blank solution.
All experiments were performed three times.
1 T. W. Hambley, Dalton Trans., 2007, 4929–4937.
2 S. P. Fricker, Dalton Trans., 2007, 4903–4917.
3 M. A. Jakupec, M. Galanski, V. B. Arion, C. G. Hartinger
and B. K. Keppler, Dalton Trans., 2008, 183–194.
4 J. Bjernemose, A. Hazell, C. J. McKenzie, M. F. Mahon,
L. P. Nielsen, P. R. Raithby, O. Simonsen, H. Toftlund and
J. A. Wolny, Polyhedron, 2003, 22, 875–885.
5 J. P. Wikstrom, A. Y. Nazarenko, W. M. Reiff and
E. V. Rybak-Akimova, Inorg. Chim. Acta, 2007, 360,
3733–3740.
6 W. A. Chomitz, S. G. Manasian, A. D. Sutton and J. Arnold,
Inorg. Chem., 2007, 46, 7199–7209.
7 B. Kabzinska, Ann. Pharm. Fr., 1964, 22, 685–687.
8 E. J. O’Neil and B. D. Smith, Coord. Chem. Rev., 2006, 250,
3068–3080.
Rat brain homogenates preparation
On the day of the experiment adult male rats fasted overnight
were euthanized in a CO2-chamber followed by decapitation.
The procedure being in compliance with the Guidelines for
Animal Experiments at IPAC RAS. The brain was rapidly
removed and homogenized in 0.12 M HEPES saline buffer
(HBS) at pH 7.4 (10% w/v), centrifuged at 3000 g at 4 °C for
10 minutes and supernatant was used immediately for assay.
Protein concentrations in rat brain homogenates were
determined by the biuret assay using bovine serum albumin
as a standard.49
9 A. Ojida, S.-K. Park, Y. Mito-oka and I. Hamachia, Tetra-
hedron Lett., 2002, 43, 6193–6195.
10 C. Lakshmi, R. G. Hanshaw and B. D. Smith, Tetrahedron,
2004, 60, 11307–11315.
Fe3+ and tBHP induced lipid peroxidation assay
The extent of lipid peroxidation was estimated by the levels of
malondialdehyde measured using the TBARS assay. Study of 11 C. J. Davies, G. A. Solan and J. Fawcett, Polyhedron, 2004,
compounds influence on lipid peroxidation of the rat brain 23, 3105–3114.
homogenates were carried out at 30 °C for 40 min in 0.25 mL 12 E. I. Solomon, T. C. Brunold, M. I. Davis, J. N. Kemsley,
of the homogenates (2 mg of protein ml−1) in HBS in the pres-
ence or absence of compounds or vehicle (DMSO). Lipid per-
S.-K. Lee, N. Lehnert, F. Neese, A. J. Skulan, Y.-S. Yang and
J. Zhou, Chem. Rev., 2000, 100, 235–349.
oxidation was induced by using Fe3+ (0.5 mM FeNH4(SO4)2)50 13 A. Trosch and H. Vahrenkamp, Inorg. Chem., 2001, 40,
t
or BHP as an oxidizing agent. Then, 0.25 mL aliquots were
2305–2311.
mixed with 0.5 mL of thiobarbituric acid medium containing 14 Y. Shimazaki, R. Kabe, S. Huth, F. Tani, Y. Naruta and
250 mM HCl, 15% trichloroacetic acid and 3 mM thiobar,
O. Yamauchi, Inorg. Chem., 2007, 46, 6083–6090.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans.

View Article Online
Paper
Dalton Transactions
15 Q.-Y. Chen, J. Huang, W.-J. Guo and J. Gao, Spectrochim. 33 J. Ancerewicz, E. Migliavacca, P.-A. Carrupt, B. Testa,
Acta, Part A, 2009, 72, 648–653.
16 U.S. Patent, US 6127356
J. D. CrapoI. Fridovich, T. Ouru, B. J. Day, J. Brian,
F. Brée, R. Zini, J.-P. Tillement, S. Labidalle, D. Guyot,
A.-M. Chauvet-Monges, A. Crevat and A. Le Ridant, Free
Radical Biol. Med., 1998, 25, 113–120.
A3
03.10.00.
R. J. Folz, B. A. Freeman, M. P. Trova and I. Batinik- 34 R. Apak, K. Guglu, M. Ozyurek and S. Karademir, J. Agric.
Haberle, Appl. US 1996-663028, 07.06.96. 97 pp. Food Chem., 2004, 52, 7970–7981.
17 E. R. Milaeva, D. Shpakovsky, Yu. Gracheva, O. Gerasimova, 35 E. Novo and M. Parola, Fibrog. Tissue Repair, 2008, 1, 5.
V. Tyurin and V. Petrosyan, J. Porphyrins Phthalocyanines,
2003, 8, 701–706.
18 E. R. Milaeva, V. Yu. Tyurin, D. B. Shpakovsky,
O. A. Gerasimova, Z. Jingwei and Yu. A. Gracheva, Heteroat.
Chem., 2006, 17, 475–480.
58pp.
36 D. Salvemini, E. Mazzon, L. Dugo, D. P. Riley, J. L. Zweier,
C. Thiemermann, Z. Wang and D. Salvemini,
Br. J. Pharmacol., 2001, 132, 815–827.
37 D. Salvemini, D. P. Riley and S. Cuzzocrea, Nat. Rev. Drug
Discovery, 2002, 1, 367–374.
19 E. R. Milaeva, S. I. Filimonova, N. N. Meleshonkova,
L. G. Dubova, E. F. Shevtsova, S. O. Bachurin and N. S. Zefirov, 38 J. B. Day and M. Patel, Drugs Future, 2008, 33, 1025–1031.
Bioinorg. Chem. Appl., 2010, Article ID 165482, 6 p.
20 E. R. Milaeva, O. A. Gerasimova, Z. Jingwei,
39 I. Kubo, N. Masuoka, T. J. Ha and K. Tsujimoto, Food
Chem., 2006, 99, 555–562.
D. B. Shpakovsky, S. A. Syrbu, A. S. Semeykin, 40 J. L. Pierre and M. Fontecave, BioMetals, 1999, 12,
O. I. Koifman, E. G. Kireeva, E. F. Shevtsova, S. O. Bachurin
and N. S. Zefirov, J. Inorg. Biochem., 2008, 102, 1348–1358.
21 V. Yu. Tyurin, J. Zhang, A. Glukhova and E. R. Milaeva,
Macroheterocycles, 2011, 4, 211–212.
195–199.
41 E.-J. Shin, K. Ho Ko, W.-K. Kim, J. S. Chae, T. P. H. Yen,
H. J. Kim, M.-B. Wie and H.-C. Kim, Neurochem. Int., 2008,
52, 1134–1147.
22 E. Milaeva, Curr. Top. Med. Chem., 2011, 11, 2703–2713.
23 R. R. Grinstead, Pat. Fr. Demande 2,462,182 (Cl.B01D11/04).
42 A. L. Spek, PLATON software, J. Appl. Crystallogr., 2003, 36,
7–13.
24 E. R. Milaeva, N. N. Meleshonkova, D. B. Shpakovsky, 43 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
K. V. Uspensky, A. V. Dolganov, T. V. Magdesieva, logr., 2008, 64, 112–122.
A. V. Fionov, A. A. Sidorov, G. G. Aleksandrov and 44 W. Brand-Williams, M. E. Cuvelier and C. Berset, Food Sci.
I. L. Eremenko, Inorg. Chim. Acta, 2010, 363, 1455–1461. Technol., 1995, 28, 25–30.
25 T. Nagataki, K. Ishii, Y. Tachi and S. Itoh, Dalton Trans., 45 J. K. Callaway, P. M. Beart and B. Jarrott, J. Pharmacol.
2007, 1120–1128.
Toxicol. Methods, 1998, 39, 155–162.
Ozturk, S. Filimonova,
N. Kourkoumelis, V. Dokorou, M. J. Manos,
A. J. Tasiopoulos, M. M. Barsan, I. S. Butler, E. R. Milaeva,
J. Balzarini and N. Hadjiliadis, Inorg. Chem., 2010, 49,
488–501.
26 Z. He, D. C. Craig and S. B. Colbran, J. Chem. Soc., Dalton 46 I.
Trans., 2002, 4224–4235.
27 A. B. P. Lever and E. I. Solomon, in Ligand-Field Theory and
the Properties of Transition Metal Complexes, ed. A. B. P.
Lever and E. I. Solomon, J. Wiley and Sons, 1999, vol. 1,
S.
K.
Hadjikakou,
pp. 1–92.
47 M. N. Xanthopoulou, S. K. Hadjikakou, N. Hadjiliadis,
E. R. Milaeva, J. A. Gracheva, V. Yu. Tyurin,
N. Kourkoumelis, K. C. Christoforidis, A. K. Metsios,
S. Karkabounas and K. Charalabopoulos, Eur. J. Med.
Chem., 2008, 43, 327–335.
28 J. Becconsall, S. Clough and G. Scott, Trans. Faraday Soc.,
1960, 56, 459–472.
29 C. R. H. I. de Jonge, H. M. van Dort and L. Vollbracht,
Tetrahedron Lett., 1970, 11, 1881–1884.
30 E. R. Milaeva, Russ. Chem. Bull., 2001, 50, 573–586.
31 N. N. Meleshonkova, D. B. Shpakovsky, A. V. Fionov,
48 S. Toda, M. Kumura and M. Ohnishi, Planta Med., 1991, 57,
8–10.
A. V. Dolganov, T. V. Magdesieva and E. R. Milaeva, J. Orga- 49 A. G. Gornall, C. J. Bardawill and M. M. David, J. Biol.
nomet. Chem., 2007, 692, 5339–5344. Chem., 1949, 177, 751–766.
32 P. Molyneux, Songklanakarin J. Sci. Technol., 2004, 26, 50 J. K. Callaway, P. M. Beart and B. Jarrott, J. Pharmacol.
211–219.
Toxicol. Methods, 1998, 39, 155–162.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2013