Fine tuning the reactivity of corrole-based catalytic antioxidants

Article abstract of DOI:10.1021/ic300408s

In order to determine the electronic factors that may affect the catalytic antioxidant activity of water-soluble metallocorroles a series of 10-aryl-5,15-pyridinium manganese(III) corroles was prepared. These complexes were examined regarding the effect of the C₁₀ substituent on the Mn^IV/Mn^III redox potentials, catalytic rate constants for decomposition of HOONO, prevention of tyrosine nitration, and superoxide dismutase activity. This structure-activity relationship investigation provides new insight regarding the mechanism by which manganese(III) corroles act as catalytic antioxidants. It also discloses the superiority of the C ₁₀-anysil-substituted complex in all examined aspects.

Full text of DOI:10.1021/ic300408s

Article
pubs.acs.org/IC
Fine Tuning the Reactivity of Corrole-Based Catalytic Antioxidants
Zoya Okun and Zeev Gross*
Schulich Faculty of Chemistry, Technion-Israel Institute of Technology, Haifa 32000, Israel
ABSTRACT: In order to determine the electronic factors that
may affect the catalytic antioxidant activity of water-soluble
metallocorroles a series of 10-aryl-5,15-pyridinium manganese-
(
III) corroles was prepared. These complexes were examined
IV
III
regarding the effect of the C10 substituent on the Mn /Mn
redox potentials, catalytic rate constants for decomposition of
HOONO, prevention of tyrosine nitration, and superoxide
dismutase activity. This structure−activity relationship inves-
tigation provides new insight regarding the mechanism by
which manganese(III) corroles act as catalytic antioxidants. It
also discloses the superiority of the C -anysil-substituted
1
0
complex in all examined aspects.
INTRODUCTION
the M(III)/M(II) redox potentials are anodically shifted, thus
increasing the rate of O2− oxidation by the chelated trivalent
metal ions. Dramatic effects were achieved by introducing
electron-withdrawing groups onto the β-pyrrolic positions of
the porphyrin and imidazolium instead of pyridilium meso
■
•
Oxidative stress, the situation occurring due to an imperfect
balance between the required and excessive amounts of reactive
oxygen and nitrogen species (ROS and RNS), is nowadays well
recognized as a key contributor to numerous diseases. This
may be exemplified by the diffusion-controlled coupling
1
1
8,19
substituents.
The best-performing catalyst to date was
reported by Batinic-Haberle’s group: Mn(II)−β-octabromo-
between superoxide and nitric oxide (both are free radicals of
meso-tetrakis(N-methylpyridinium-3-yl)porphyrin, whose SOD-
vital importance), whose combination leads to peroxynitrite
20
−
like activity approaches that of the enzyme itself. Research
regarding decomposition of PN by metalloporhyrins began in
1996 by M. K. Stern, who revealed that water-soluble iron(III)
porphyrins function as “PN isomerases” by catalyzing the rapid
(
ONOO ) and subsequent transformation of its conjugated
•
acid (HOONO, PN) into the most toxic ROS and RNS: OH
•
2−4
and NO , respectively.
Both radicals are responsible for
2
modifications of lipoproteins and cellular components, but
nitration of tyrosine, cysteine, and tryptophan residues and
subsequent inactivation of numerous vital enzymes that
eventually may cause cell dysfunction and death is considered
21
isomerization of PN to nitrate. This discovery was followed
by intense research on manganese(III)-based water-soluble
porphyrins by J. T. Groves and others, who realized that while
the first step of PN reduction is very fast full catalysis is not
achieved because of the slow reduction of the thus produced
oxo−Mn(IV) intermediate back to the Mn(III) oxidation
5
the main biomarker for involvement of PN. The particularly
severe cytotoxicity of PN may safely be attributed to the fact
that there is no biological defense system against it, as all
natural antioxidants (including diet-supplied and enzymes, such
as superoxide dismutases and catalases) do not react with it
22
state. Combination of Mn(III) porphyrin with biologically
relevant reducing agents is, nevertheless, a powerful approach.
While these naturally occurring antioxidants do not react fast
enough with PN itself, they are very efficient for reducing oxo−
6
faster than the vital biomolecules.
A large variety of porphyrins and other strongly chelating
ligands has been prepared for testing their potential regarding
decomposition of ROS and RNS by the corresponding iron and
23
Mn(IV) to Mn(III) and thus enable catalysis.
The most recently introduced catalytic antioxidants are based
7
−12
on water-soluble corroles, which have already been revealed to
manganese complexes.
The earliest synthetic antioxidants
2
4,25
1
3
be highly potent for decomposition of PN,
superoxide
were developed as superoxide dismutase (SOD) mimics,
which may be attributed to the fact that involvement of PN and
its decomposition products in biological systems was
appreciated only much later.
metalloporphyrins as SOD mimics was started by R. F.
26
27
radical, and hydrogen peroxide. The largest novelty of the
iron(III) corroles is in their efficiency for H O decomposition,
2
2
1
4,15
which is not matched by other complexes (most are extensively
Pioneering research on
28
bleached by it). Manganese(III) corroles were identified as
the first manganese complexes to display catalytic capability for
PN decomposition in the absence of reducing agents as well as
the only known catalyst for its disproportionation to benign
16
Pasternack in 1979 and continued by many others.
A
particularly important contribution came from the Fridovich
group, who established a structure−activity relationship (SAR)
between the redox potential of manganese and iron porphyrins
and their superoxide-dismutating activities in vitro and in
Received: February 22, 2012
Published: July 18, 2012
1
7
vivo. That research revealed that SOD activity increases when
©
2012 American Chemical Society
8083
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090

Inorganic Chemistry
Article
Scheme 1. Synthesis of Water-Soluble Corroles with Different C Aryls To Affect the Electron Richness of the Macrocycle and
1
0
a
Chelated Manganese(III) Ion
a
Inset: Structure of MnS with negatively charged head groups.
2
Scheme 2. Catalytic Cycle for PN Decomposition by Manganese(III) Corroles
2
4
products. These inventions were followed by studies on the
cellular level for testing the metallocorrole’s antioxidative
properties in vitro. The results revealed cytoprotection by
these catalysts on various cell lines used as models for
as oxidant), yielded 8−14% of the desired compounds. The
free-base corroles were characterized by NMR, UV−vis, and
mass spectroscopies. Insertion of manganese was achieved by
refluxing pyridine solutions of the corroles with manganese
diacetate tetrahydrate, followed by alkylation with methyl
2
9
30,31
diabetes, atherosclerosis,
and neurodegenerative dis-
including even some neurorescue effects. Impor-
3
2,33
−
−
eases,
iodide and I /Cl ion exchange. This procedure led to the
water-soluble manganese(III) corroles (1−5) bearing two
positively charged substituents. This order of synthetic steps
is better than alkylation before metalation, because the possible
tantly, the corrole complexes performed better than structurally
related porphyrins in the examined applications. We now report
on studies that focused on ways to increase the catalytic
antioxidative performance of water-soluble manganese(III)
corroles via establishment of a structure−activity relationship.
Toward this goal we prepared a series of compounds similar to
complex 1, the most efficient manganese(III) corrole regarding
catalytic decomposition of PN and superoxide radical. The new
complexes, positively charged corroles that differ only in the
C10 meso substituent, were examined regarding PN decom-
position rates and prevention of nitration by it as well as for the
ability to decompose superoxide radical. The results provide
new insight regarding the mechanism by which manganese(III)
corroles act as catalytic antioxidants. It also disclosed the
34
N-alkylation of the inner nitrogen atoms is avoided. It is also
easier to remove excess inorganic manganese salts before rather
than after alkylation.
The motivation for preparing this particular series is based on
previous research regarding PN decomposition by the
negatively charged manganese(III) corrole MnS2 and the
better performance of the positively charged 1. One important
observation was that attempts to follow photodiode array
(PDA) spectral changes during PN decomposition by either 1
or MnS revealed no changes in the visible part of the spectrum
2
during catalysis. This phenomenon and a variety of other
indications were used to conclude that formation of (oxo)-
manganese(V) corrole Mn(V)O is the rate-limiting step in the
superiority of the C -anysil-substituted complex in all
examined aspects.
1
0
overall catalytic cycle (k , Scheme 2). We have now decided to
1
RESULTS AND DISCUSSION
A series of five corroles, with different C meso aryl
substituents to affect electronic effects in a predictable fashion,
was prepared using the synthetic route described in Scheme 1.
Condensation of para-pyridyl-substituted dipyrromethane with
aromatic aldehydes in refluxing propionic acid, without the aid
of an oxidizing agent such as DDQ/p-chloranil (relying on air
■
investigate the effect of the corrole electron richness on the
catalytic activity of the manganese complexes, based on the
following hypothesis: Mn(III) will be more reducing when
chelated by an electron-rich ligand, and as long as the Mn(V)O
state will not be too stabilized, the overall catalytic constant for
PN decomposition will increase. The predictions for each of the
crucial steps in the catalytic cycle are outlined in Scheme 2b.
1
0
8
084
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090

Inorganic Chemistry
Article
+
Figure 1. (a) Oxidation potentials (vs Ag/AgCl) of 0.8 mM 1−5 in acetonitrile with TBAP as electrolyte; scan rate 100 mV/s (the Fc/Fc couple
was 0.45 V under identical conditions), and (b) cyclic voltammograms of 1 and 5.
+
Table 1. Oxidation Potentials (vs Ag/AgCl) of 1−5 in Acetonitrile with TBAP as electrolyte (the Fc/Fc couple was measured
as 0.45 V under identical conditions), PN Decomposition Rates, Amount of Nitrite Produced in the Decomposition Reaction,
Attenuation of Tyrosine Nitration, and SOD Activity of the Examined Compounds
%
of nitration
attenuation
IV
kcat (10 M s−1) for PN
5
−1
IC50 (nm) for O2−•
dismutation
E1/2 (V) for Mn /
% of produced nitrite from
PN
1 mM
PN
2 mM
PN
III
complex
Mn
decomposition
1
2
3
4
5
0.62
0.57
0.51
0.50
0.47
1.58
1.67
2.0
62
64
68
92
76
42.3
45.2
55.1
57.3
56.3
39.1
42.2
51.6
53.1
48.6
181
160
25
2.73
1.82
20
26
An experimental evaluation of how the C10 meso aryl
substituents affect the electron richness of the chelated metal
was obtained via examination of their Mn(IV)/Mn(III) redox
couple. Cyclic voltammetry revealed that all manganese(III)
corroles are oxidized at relatively low potentials to manganese-
(
IV) derivatives and that the redox potential gradually
decreases from 0.62 V for the most electron-withdrawing 1 to
.47 V for the most electron-releasing 5 (Figure 1 and Table 1).
0
It may hence be concluded that the C10 substituents stabilize
the higher valent oxidation state increasingly more as a function
of decreased electron-withdrawing ability. This in turn predicts
that the manganese(III) corrole should become more reactive
in the order of 1 to 5 (fastest acting) as long as k of Scheme 2
1
is considered.
The rate by which the complexes decompose PN (5−20 μM
catalyst, 400 μM PN, pH 7.4) was investigated by following the
decay of PN at 302 nm using stopped flow kinetics. The
second-order catalytic rate constants were deduced from the
dependence of the pseudo-first-order rate constants on the
catalyst’s concentration (Figure 2, inset). A plot of the second-
order rate constants versus the Mn(IV)/Mn(III) redox couples
Figure 2. PN decomposition rates in aqueous pH 7.4 solutions
catalyzed by the examined manganese(III) corroles in relation to their
Mn(IV)/Mn(III) redox potentials (Inset) Plot used for determination
of the catalytic rate constant of 4.
donating p-morpholinophenyl group in 5. Consequential
reduction of k may lead to a situation of k ≈ k or k > k ,
2
1
2
1
2
5
−1 −1
(
2
Figure 2) reveals an increase from 1.6 × 10 M
s
for 1 to
resulting in an overall smaller rate for decomposition of PN.
Supporting evidence for this proposal was obtained from diode
array spectra of catalytic PN decomposition by 5 and its
comparison with the spectra of the manganese(III) corrole 5
and the independently prepared (via oxidation with iodosyl-
benzene) (oxo)manganese(V) complex of the same corrole (5-
Mn(V)O). As may be appreciated from Figure 3a, the main
differences are as follows: (a) 5 has strong bands at 490 and
686 nm, absent for 5-Mn(V)O; (b) the 420 nm band of 5-
Mn(V)O is much more pronounced than in 5; (c) the 490/420
nm ratio of 5 is 1.4. In Figure 3b, the disappearance of PN (λmax
= 302 nm) is reflected in the spectral reduction at 280−350 nm
and the 490/420 nm ratio is ∼1:1. This suggests that the
5
−1
−1
.7 × 10
M
s
for 4. The larger potency for PN
decomposition by 4 relative to 1−3 is consistent with the
predicted effects of the C substituents on k of Scheme 2: the
1
0
1
complexes with less electron-withdrawing aryls are more
reducing, their k ’s are hence larger, and this effect on the
1
rate-determining step is reflected in a larger kcat in agreement
with the proposed hypothesis. On the other hand, complex 5
does not fit within the above scenario, as its catalytic rate
constant is lower than that of 4 and even 3.
On the basis of the earlier introduced hypothesis (Scheme 2)
the “anomaly” of 5 may be attributed to the increased stability
of the Mn(V)O intermediate due to the highly electron-
8
085
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090

Inorganic Chemistry
Article
Figure 3. (a) Photodiode array spectra of PN decomposition by 5, and (b) spectrum of independently prepared 5-Mn(V)O from Mn(III), all in pH
.4 phosphate buffer solutions.
7
36−39
“steady state” of the catalyst during catalysis is a mixture of
aromatic amino acids.
In particular, excessive amounts of
Mn(III) (i.e., 5) and Mn(V)O (i.e., 5-Mn(V)O), consistent
with a k ≈ k situation. It is important to emphasize that this
nitrotyrosine relative to healthy cells/tissues/organs are the first
4
0,41
1
2
sign for a role of PN in numerous diseases.
On the basis of
finding is unique of 5: similar investigations on 1−4 revealed
only Mn(III) during catalysis.
the above, the ability of complexes 1−5 to attenuate tyrosine
nitration by PN was examined. Fixed concentrations of freshly
prepared PN were added to 0.8 mM tyrosine solutions, with
and without 50 μM of the complexes 1−5, and incubated for 15
min at 37 °C. The amounts of unreacted tyrosine and produced
Previous research has revealed that manganese(III) corrole
complexes decompose PN via disproportionation, yielding
molecular oxygen and nitrite. Accordingly, we also examined
the percentage of nitrite formation from PN (400 μM) via its
decomposition in the presence of a fixed amount (10 μM) of
the various catalysts. Spontaneous decay of PN is composed of
homolytic bond cleavage and isomerization, both of which lead
3
-nitrotyrosine were quantified by HPLC analysis (Figure 5a),
and the results for two different concentrations of PN are
presented in Figure 5b and Table 1. The percentage of tyrosine
nitration attenuation steadily increased from the most electron-
withdrawing complex 1 to the anysil-substituted complex 4, but
the capability of 5 was lower than that of 4.
To summarize the effects of the C10 aryls, determination of
the Mn(III)/Mn(IV) redox potentials confirmed that increased
electron richness leads to increased reducing power of the
chelated manganese(III) in the order of 1 to 5. It is however
important to realize that changes in the electron-donating/
withdrawing capabilities of the corrole macrocyclic could have
conflicting effects on the fundamental steps of the overall
catalytic cycle for decomposition of PN (Scheme 2). While k1
might be anticipated to increase for more electron-rich
−
−
to nitrate and not nitrite. Accordingly, the [NO ]/[NO ]
2
3
ratio may be used for evaluating the competition between
catalyzed and noncatalyzed reaction pathways. Nitrite concen-
tration was quantified by the Griess reagent, which relies on
development of a deep purple color (λmax = 540 nm) whose
intensity is directly correlated with nitrite concentration. The
outcome of this investigation (Figure 4 and Table 1) resembled
manganese(III) corroles, k should be affected in an opposite
2
fashion. Since these rate constants are too fast to be determined
independently, the evaluation is limited to the effect on the
overall catalytic rate, k . The results obtained by focus on three
cat
different phenomena show that electron-releasing aryls do
increase kcat as long as k remains rate limiting. This situation
1
changes in the most electron-rich complexes, for which decay of
the stabilized (oxo)Mn(V) intermediate (k ) is slower than its
2
formation.
The improved insight into the mechanism of catalytic
decomposition of PN raised our interest in investigating
another recently reported activity of manganese(III) corroles:
Figure 4. Nitrite/nitrate ratio produced in the decomposition reaction
of PN at 10 μM catalyst and 400 μM PN as a function of the Mn(IV)/
Mn(III) redox potentials.
26
they catalyze also decomposition of superoxide radical. PN
decomposition and O2− dismutation share similarities by
virtue of one metal being responsible for both the reduction
and the oxidation of the ROS. The difference is that PN
decomposition by the manganese(III) corroles proceeds via
•
the results regarding the PN decomposition rate: the anysil-
substituted complex 4 provided 92% nitrite, the more electron-
poor catalysts provided less nitrite, and the most electron-rich 5
was less efficient than 4.
−•
two-electron processes (Scheme 2), while O
dismutation by
2
The ability to prevent nitration of vital proteins and enzymes
is one of the most desirable features required for potent
the enzyme and all its synthetic mimics relies on one-electron
−
•
redox reactions: oxidation and reduction of O
, as depicted in
2
35
catalytic antioxidants. In this context, it is important to note
that the main biomarker for involvement of PN is nitration of
eqs 1 and 2 (note that n = 3 for corroles but n = 2 for SOD and
porphyrin-based SOD mimics), respectively. Accordingly, we
8
086
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090

Inorganic Chemistry
Article
Figure 5. (a) Representative chromatograms used for evaluation of the ability of the manganese(III) corroles to prevent tyrosine nitration. (b)
Attenuation of PN-induced tyrosine nitration by the examined compounds.
decided to examine the effect of the C aryls in 1−5 on the
With redox potentials of 0.47−0.62 V (vs Ag/AgCl), the only
1
0
−•
−•
catalytic rates for O2 dismutation thereby.
feasible reaction of the Mn(III) corroles with O
is that
2
depicted in eq 2, i.e., transformation to Mn(IV). The
thermodynamic driving force hence increases in the order of
k
n
−•
3
n−1
M + O
→ M
+ O2
(1)
(2)
2
1
> 2 > 3−5. The predicted effect of the redox potentials on k
4
k
n−1
+ O2−• + 2H → M + H O
+
4
n
is in the opposite order, but the E values of 3−5 are similar,
which is the apparent reason for the similar catalytic rates that
M
1/2
2
2
The well-known enzymatic assay, which is based on
are obtained once k becomes fully or partially rate limiting.
4
4
2
cytochrome c reduction attenuation, was used for exploring
the ability of 1−5 to decompose superoxide radical. To verify
that the examined complexes do not inhibit production of
superoxide by interaction with xanthine oxidase, we also
monitored the formation rate of uric acid from xanthine with
and without the examined compounds. Identical results were
obtained by the two methods. Regarding the cytochrome c
assay, very significant decreases in IC values were obtained as
SUMMARY AND CONCLUSIONS
■
The main goal of the investigations was to obtain insight into
the factors that determine catalytic antioxidant activity and to
use the acquired mechanistic knowledge for design of catalysts
with superior performance. We synthesized complexes 1−5,
five water-soluble manganese corroles that differ significantly in
the electron-withdrawing capability of their C aryl substituent.
5
0
10
a function of the C10 aryl substituents: from 181 and 160 nM
for the electron-poor 1 and 2 to 20 ± 3 nM for the more
electron rich 3−5 (Figure 6 and Table 1).
Cyclic voltammetry revealed that the C10 substituents
destabilize the manganese(III) oxidation state increasingly
more as a function of decreased electron-withdrawing ability.
This has been demonstrated to have a direct influence on the
rate of PN decomposition, the percentage of PN disproportio-
nation, attenuation of tyrosine nitration, and the effectiveness of
superoxide radical dismutation. Evaluation of the overall order
catalytic rate constants revealed a larger potency for PN
decomposition by 4 relative to 1−3, attributed to an effect on
the first and rate-limiting step of catalysis. The most electron-
donating C10 substituent present in complex 5 increased the
stability of the Mn(V)O intermediate to an extent that led to a
change in the rate-limiting step and a lower overall catalytic
rate. Regarding tyrosine nitration, the most established
biomarker for involvement of PN in diseases, the trend for
PN decomposition fully came into account: the efficiency for
preventing this reaction was in the order of 1 < 2 < 3 < 4 > 5.
Similar phenomena were seen in the examination of superoxide
radical decomposition: the less electron-rich 1 and 2 displayed
much larger IC50 values (lower efficiency) than the electron-
poor 3−5. Even though the elementary step in the catalytic
decomposition of superoxide and peroxynitrite is significantly
different, the similarity is that reduction by manganese(III) is
the rate-limiting step for 1. The rate of that elementary reaction
and consequently the overall catalytic rate is increased for the
more electron-rich complexes, but there is no more acceleration
for 3−5 because the second step becomes rate limiting.
Figure 6. SOD activity of the examined manganese(III) complexes in
relation to their Mn(IV)/Mn(III) redox potentials.
Analysis of these results with respect to eqs 1 and 2 suggests
that (a) the rate-limiting step for 1 is k , (b) k is increased in
3
3
the order of 1 > 2 > 3−5, (c) for 3−5 the rate-limiting step
becomes k , and (d) the effect of the C aryls on k is less
4
10
4
pronounced than on k . Supporting evidence for these
3
conclusions comes from the extensive knowledge of the
17
correlation between metal redox potentials and SOD activity.
8
087
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090

Inorganic Chemistry
Article
In summary, these structure−activity investigations have
identified the anysil-substituted complex 4 as the best-
performing manganese(III) corrole-based catalytic antioxidant
regarding all examined aspects. We trust that the insight
obtained in this work will be of great asset in the design of
optimal catalysts for treating the many diseases that are affected
by reactive oxygen and nitrogen species.
5,15-Bispyridyl-10-phenylcorrole. Column chromatography: silica,
hexane followed by gradual addition of EtOAc. Fluorescent, green
fractions were released from the column at hexane/EtOAc 1:2. This
was followed by crystallization (hexane/CH Cl 2:1) to afford the
2
2
−3
desired corrole in 11% yield. UV−vis (CH Cl ): λ (ε × 10 ) = 418
2
2
max
(
5
30.4), 575 (3.4), 623 (2.4), 647 (2.3). MS (MALDI-TOF): m/z (%)
+
1
27 (100) [M ]. H NMR (400 MHz, C D ), δ = 8.93 (unresolved
6 6
doublet, 4H), 8.81 (d, J = 4.4 Hz, 2H), 8.73 (d, J = 3.6 Hz, 2H), 8.55
unresolved doublet 2H), 8.37 (d, J = 6.8 Hz, 2H), 7.88 (unresolved
doublet, 2H), 7.63 (t, J = 7.2 Hz, 4H), 7.53 (t, J = 7.2 Hz, 2H).
,15-Bispyridyl-10-(4-methoxyphenyl)corrole. Column chroma-
(
EXPERIMENTAL SECTION
■
5
Materials. The materials used for synthesis and workup procedures
were purchased from Aldrich, Merck, Fluka, Riedel-de Haen
tography: silica, hexane/EtOac 1:1 followed by gradual increase of
EtOAc until 100%, followed by another column (silica, DCM/EtOH
̈
, Biolab,
and Frutarom and used as received unless otherwise stated. Deuterated
solvents (Aldrich isotopes products) with a 99.5% minimum
deuteration were used as received. Silica gel for column chromatog-
raphy (Silica Gel 60, 63−200 μm mesh) was obtained from E. Merck
Ltd. Most starting materials for syntheses were from Sigma-Aldrich
and used without further purification. Pyrrole was run through a short
basic alumina column, and aldehydes were purified by vacuum
distillation before use. Tetrabutylammonium perchlorate as a
supporting electrolyte in the CV experiments was obtained from
Fluka and used after three recrystallizations from absolute ethanol.
Electrodes for CV were obtained from CH Instruments.
1
50:1 followed by a gradual increase of EtOH until 50:1). Collection
of fluorescent dark-green-colored fractions afforded the desired corrole
−3
in 9% yield. UV−vis (CH Cl ): λ (ε × 10 ) = 420 (48.9), 574
2
2
max
(6.6), 623 (4.8), 649 (4.7). MS (MALDI-TOF): m/z (%) 558 (100)
+
1
6
6
[
M ]. H NMR (400 MHz, C D ): δ = 8.98 (d, J = 4.8 Hz, 4H), 8.80
(d, J = 3.6 Hz, 2H), 8.69 (d, J = 4.4 Hz, 2H), 8.64 (d, J = 4.4 Hz, 2H),
8
7
.47 ppm (br s, 2H), 8.15 (d, J = 8 Hz, 2H), 8.02 (d, J = 5.2 Hz, 4H),
.27 (d, J = 8 Hz, 2H), 3.60 (s, 3H).
5
,15-Bispyridyl-10-(4-morpholinophenyl)corrole. Column chro-
matography: silica, EtOAc followed by crystallization (hexane/
CH Cl2 1:1), affording the desired corrole in 14% yield. UV−vis
2
Physical Methods. Nuclear Magnetic Resonance Spectroscopy.
H NMR and F NMR spectra were recorded at room temperature on
−3
(
(
EtOAc): λmax (ε × 10 ) = 421 (49.6), 577 (11.4), 620 (8.6). MS
1
19
+
1
MALDI-TOF): m/z (%) 613 (100) [M ]. H NMR (400 MHz,
a Bruker Avance 400 III spectrometer equipped with automated tuning
and matching on a broad band (BBFO) probe with gradients and 5
mm probe head. Chemical shifts are reported in ppm relative to
C D ): δ = 8.96 (d, J = 5.0 Hz, 4H), 8.79 (d, J = 4.2 Hz, 2H), 8.77 (d, J
=
6
6
4.7 Hz, 2H), 8.70 (d, J = 4.7 Hz, 2H), 8.45 ppm (d, J = 4.1 Hz, 2H),
8
.24 (d, J = 8.3 Hz, 2H), 8.01 (d, J = 5.5 Hz, 4H), 7.12 (unresolved
residual hydrogen atoms of TMS (δ = 0.00) or relative to CFCl (δ =
3
F
doublet), 3.71 (t, J = 4.6 Hz, 4H), 3.07 (t, J = 4.6 Hz, 4 H).
Manganese Insertion and Alkylation. Manganese insertion,
alkylation, and ion exchange: free-base corrole 21−32 mg was refluxed
in 5 mL of pyridine with 15 equiv of Mn(OAc) ·4H O for 20 min,
0.00).
Mass Spectrometry. Measurements were performed using a
Micromass MS Technologies Maldi micro.
2
2
Cyclic Voltammetry. CV measurements were performed with a
WaveNow USB potentiostat/galvanostat (Pine Research Instrumenta-
tion) using Pine AfterMath Data Organizer software. A three-electrode
system consisting of a platinum wire working electrode, a platinum
wire counter electrode, and an Ag/AgCl reference electrode was
employed. CV measurements were performed using acetonitrile
solutions of 0.3 M tetrabutylammonium perchlorate under nitrogen
atmosphere at ambient temperature. The scan rate was 100 mV/s, and
the E1/2 value for the ferrocene/ferrocenium couple under these
conditions was 0.45 V.
UV−Vis Absorption Spectroscopy. Absorption spectra were
recorded with an Agilent 8453 diode array spectrophotometer.
Stopped-Flow Spectrophotometry. Fast kinetics experiments were
performed on the AppliedPhotophysics RX.2000 rapid-mixing
stopped-flow unit.
followed by chromatographic separation without solvent evaporation
dry silica, starting with CH Cl and gradually adding methanol, unless
(
2
2
otherwise stated). The product was dissolved in a minimal amount of
hot THF, and excess methyl iodide 500 μL was added to the solution,
which was then left at 40 °C until complete precipitation. Solid
material was collected by centrifugation and washed with THF and
diethyl ether until the solvent was colorless, affording pure 1−5. The
product was dissolved in a minimal amount of water, 500 mg of freshly
HCl regenerated ion-exchange resin (Dowex 1:8 chloride form) was
added, and the vessel was slowly shaken overnight. The resin was
filtrated, and the solvent was lyophilized.
Manganese(III) 5,15-Bis(N-methylpyridinium)-10-pentafluoro-
phenylcorrole (1). This compound was prepared as previously
2
5
published.
Manganese(III) 5,15-bis(N-methylpyridinium)-10-(2,6
difluorophenyl)corrole (2). Manganese insertion was performed as
described by the general procedure starting with 21 mg of free base,
affording 20 mg of the manganese(III) complex (89% yield). Product
was dissolved in hot THF and alkylated as described by the general
HPLC Analysis. Analyses were performed using a Merck Hitachi
HPLC with a UV−vis detector system, equipped with a Luna 5u C18
Phenomenex column 250 × 4.6 mm, 5 μm. Commercially available
tyrosine and 3-nitrotyrosine were used as standards.
Syntheses of Free-Base A B (A = pyridyl) Corroles: General
procedure, affording 27 mg of pure 2 (95% yield 2). UV−vis (H O):
2
2
−3
Procedure. All of the examined compounds were synthesized by the
previously developed method. In brief, aldehyde (42−76 mg, 0.4
mmol) was added to a 10 mL solution of 5-(4-pyridyl)dipyrromethane
λ
(ε × 10 ) = 420 (13.2), 491 (28.0), 668 (9.7). MS (MALDI-
max
TOF LD+): m/z (%) 646 (20) [M − 2Cl], 631 (100) [M − Me −
+
+
2
6
Cl]. HRMS (m/z): [C H N F MnCl] ([M − Cl] ) calcd (found)
37
25
6 2
(
178 mg, 0.8 mmol) in propionic acid, and the mixture was refluxed
81.1178 (681.1177).
Manganese(III) 5,15-Bis(N-methylpyridinium)-10-phenylcorrole
for 1 h. The residue obtained after solvent evaporation was washed
with hot water, neutralized with ammonium hydroxide (25%), and
washed again with hot water. Product purification and chemical yield
are described for each compound.
(3). Manganese insertion was performed as described by the general
procedure starting with 29 mg of free base, affording 29 mg of the
manganese(III) complex (91% yield). The product was dissolved in
hot THF and alkylated as described by the general procedure,
affording 39 mg of pure 3 (96% yield). UV−vis (H O): λ (ε ×
5
,15-Bis(pyridyl)-10-pentafluorophenylcorrole. This compound
25
was prepared as previously published.
2
max
−
3
5
,15-Bis(pyridyl)-10-(2,6 difluorophenyl)corrole. Column chroma-
10 ) = 355 (21.7), 424 (21.6), 490 (28.6), 680 (18.4). MS (MALDI-
TOF LD+): m/z (%) 680 (10) [M], 645 (100) [M − Cl], 610 (10)
[M − 2Cl], 580 (30) [M − 2Me − 2Cl]. HRMS (m/z):
[C H N MnCl] ([M − Cl] ) calcd (found) 645.1366 (645.1347).
Manganese(III) 5,15-Bis(N-methylpyridinium)-10-(4-
methoxyphenyl)corrole (4). Manganese insertion was performed as
described by the general procedure starting with 28 mg of free base,
tography: silica, hexane/EtOAc 1:1 followed by another column (silica,
DCM/EtOH 150:1 and gradual increase of EtOH until 50:1.
Fluorescent, dark-green-colored fractions afforded the desired corrole
in 9% yield. All physical data recorded for the product was identical to
that reported for the same compound obtained by a different synthetic
+
+
37
27
6
43
route.
8
088
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090

Inorganic Chemistry
Article
affording 27 mg of the manganese(III) complex (88% yield). Product
was dissolved in hot THF and alkylated as described by the general
the rate of accumulation of urate at 290 nm in the absence of
cytochrome c.
procedure, affording 36 mg of pure 4 (94% yield). UV−vis (H O):
2
−3
λ
(ε × 10 ) = 354 (15.9), 492 (32.1), 601 (8.54), 684 (16.3). MS
max
AUTHOR INFORMATION
■
(
MALDI-TOF LD+) :m/z (%) 640 (100) [M − 2Cl], 625 (10) [M −
Me − 2Cl], 610 (50) [M − 2Me − 2Cl]. HRMS (m/z):
Corresponding Author
*E-mail: chr10zg@tx.technion.ac.il.
+
+
[
C H N OMnCl] ([M − Cl] ) calcd (found) 675.1475 (645.1472).
38
29
6
Manganese(III) 5,15-Bis(N-methylpyridinium)-10-(4-
Notes
morpholinophenyl)corrole (5). Manganese insertion was performed
as described by the general procedure starting with 32 mg of free base,
followed by chromatographic separation without solvent evaporation
The authors declare no competing financial interest.
(
dry silica, starting with EtOAc and gradually adding methanol),
ACKNOWLEDGMENTS
■
affording 29 mg of the manganese(III) complex (84% yield). Product
was dissolved in THF and alkylated as described by the general
procedure, affording 37 mg of pure 5 (91% yield). UV−vis (H O):
λ
This work was supported by a grant from Johnson & Johnson
to Z.G.
2
−3
(ε × 10 ) = 358 (12.2), 421 (11.8), 490 (14.6), 684 (6.7). MS
max
(
MALDI-TOF LD+): m/z (%) 695 (10) [M − 2Cl], 680 (100) [M −
REFERENCES
■
Me − 2Cl], 665 (20) [M − 2Me − 2Cl]. HRMS (m/z):
C H N OMn ([M − 2Cl] ) calcd (found) 695.2205 (695.2202).
2+
(1) Halliwell, B. G. J. Free radicals in biology and medicine, 4th ed.;
Oxford University Press: New York, 2007.
2) Goldstein, S.; Lind, J.; Merenyi, G. Chem. Rev. 2005, 105, 2457.
3) Huie, R. E.; Padmaja, S. Free Radical Res. Commun. 1993, 18, 195.
4) Beckman, J. S.; Koppenol, W. H. Am. J. Physiol. 1996, 271,
41
34
7
PN Decomposition. PN was prepared according to a published
4
4
(
(
(
procedure, and the experimental procedures for studying its
decomposition were identical to those reported in ref25. Decom-
position of 400 μM PN was measured with and without 5, 10, 15, and
C1424.
2
0 μM of the examined compounds.
Nitrite Quantification. The amount of nitrite produced in PN
(
(
2
(
5) Peluffo, G.; Radi, R. Cardiovasc. Res. 2007, 75, 291.
6) Szabo, C.; Ischiropoulos, H.; Radi, R. Nat. Rev. Drug Discovery
decomposition experiments by different catalysts was quantified using
the Griess reagent. A 15 μM concentration of the examined compound
007, 6, 662.
7) Rajic, Z.; Benov, L.; Kos, I.; Tovmasyan, A.; Batinic-Haberle, I.
Free Radical Biol. Med. 2010, 49, S194.
8) Rausaria, S.; Kamadulski, A.; Rath, N. P.; Bryant, L.; Chen, Z. M.;
Salvemini, D.; Neumann, W. L. J. Am. Chem. Soc. 2011, 133, 4200.
9) Spasojevic, I.; Batinic-Haberle, I.; Stevens, R. D.; Hambright, P.;
Thorpe, A. N.; Grodkowski, J.; Neta, P.; Fridovich, I. Inorg. Chem.
001, 40, 726.
(
dissolved in 0.3 M KH PO pH 7 buffer) was reacted with 400 μM of
2 4
freshly prepared PN (dissolved in 0.05 M NaOH). All samples were
diluted by a factor of 10 before addition of the Grees reagent. Ninety-
six well plates were used, and every reaction was examined in triplicate.
A 50 μL amount of the sample was added to each well, followed by
addition of 50 μL of sulfanilamide solution (1% sulfanilamide in 5%
phosphoric acid) and incubation for 10 min (protected from light), 50
μL of 0.1% N-1-napthylethylenediamine dihydrochloride (NED) in
water was added, and the plate was incubated in the dark for 10 min
more. Absorbance at 540 nm was detected using a Spectramax
M2Microplate Reader (Molecular Devices). Nitrite concentrations
were determined relative to a standard curve. The background nitrite
(
(
2
(
(
10) Day, B. J. Drug Discovery Today 2004, 9, 557.
11) Radovits, T.; Seres, L.; Gero, D.; Lin, L. N.; Beller, C. J.; Chen,
S. H.; Zotkina, J.; Berger, I.; Groves, J. T.; Szabo, C.; Szabo, G. Mech.
Ageing Dev. 2007, 128, 173.
(12) Masini, E.; Cuzzocrea, S.; Mazzon, E.; Marzocca, C.; Mannaioni,
P. F.; Salvemini, D. Br. J. Pharmacol. 2002, 136, 905.
(13) Batinic-Haberle, I.; Reboucas, J. S.; Spasojevic, I. Antiox. Redox
Signaling. 2010, 13, 877.
(from PN synthesis) was measured and subtracted from all samples by
performing spontaneous decomposition of PN under acidic conditions
45−47
(pH 3). As stated,
in acidic environment PN decomposes only via
the isomerization mechanism, i.e., only nitrate is formed, so any
measured nitrite originates from preparation of PN. Nitrite
concentrations were determined relative to a standard curve.
Tyrosine Nitration. Tyrosine (0.8 mM) in 0.05 M phosphate
buffer (pH 7.4) was reacted with either 1 or 2 mM PN with and
without 50 μM of the examined compounds 1−5. Samples were
incubated at 37 °C for 15 min and kept on ice. The ability of 1−5 to
inhibit PN-mediated reactions was expressed as the percentage of
inhibition of 3-nitrotyrosine formation relative to the catalyst-free
conditions. The amounts of tyrosine and nitrated tyrosine were
determined by HPLC analysis.
(14) Ferrer-Sueta, G.; Radi, R. Acs. Chem. Biol. 2009, 4, 161.
(15) Blough, N. V.; Zafiriou, O. C. Inorg. Chem. 1985, 24, 3502.
(16) Pasternack, R. F.; Halliwell, B. J. Am. Chem. Soc. 1979, 101,
1026.
(17) Batinic-Haberle, I.; Spasojevic, I.; Hambright, P.; Benov, L.;
Crumbliss, A. L.; Fridovich, I. Inorg. Chem. 1999, 38, 4011.
(18) Kachadourian, R.; Batinic-Haberle, I.; Fridovich, I. Inorg. Chem.
1999, 38, 391.
(19) Batinic-Haberle, I.; Spasojevic, I.; Stevens, R. D.; Hambright, P.;
Neta, P.; Okada-Matsumoto, A.; Fridovich, I. Dalton Trans. 2004,
1696.
The separation between tyrosine and 3-nitrotyrosine was achieved
using a Luna 5u C18 Phenomenex column 250 × 4.6 mm. The mobile
phase consisted of 0.05 M KH PO /H PO (pH 3) and 10%
(20) DeFreitas-Silva, G.; Reboucas, J. S.; Spasojevic, I.; Benov, L.;
Idemori, Y. M.; Batinic-Haberle, I. Arch. . Biochem. Biophys. 2008, 477,
105.
2
4
3
4
methanol, the flow rate was 1 mL/min, and the detection wavelength
was 275 nm. The injection volume of the samples was 40 μL. Tyrosine
and 3-nitrotyrosine had a retention time of 5 and 13 min, respectively.
Cytochrome c Assay. Catalysis of dismutation of superoxide was
measured using a xanthine oxidase/xanthine system as a source of
superoxide and cytochrome c as the indicating scavenger of superoxide.
Reduction of cytochrome c was followed at 550 nm. Assays were
conducted at 25 ± 1 °C in 0.05 M phosphate buffer, pH 7.8 and 0.1
mM EDTA. IC50 values were determined by plotting v /v − 1 versus
(21) Stern, M. K.; Jensen, M. P.; Kramer, K. J. Am. Chem. Soc. 1996,
118, 8735.
(22) Lee, J. B.; Hunt, J. A.; Groves, J. T. J. Am. Chem. Soc. 1998, 120,
6053.
(23) Lee, J. B.; Hunt, J. A.; Groves, J. T. Bioorg. Med. Chem. Lett.
1997, 7, 2913.
(24) Mahammed, A.; Gross, Z. Angew. Chem., Int. Ed. 2006, 45, 6544.
(25) Gershman, Z.; Goldberg, I.; Gross, Z. Angew. Chem., Int. Ed.
2007, 46, 4320.
(26) Eckshtain, M.; Zilbermann, I.; Mahammed, A.; Saltsman, I.;
Okun, Z.; Maimon, E.; Cohen, H.; Meyerstein, D.; Gross, Z. Dalton
Trans. 2009, 7879.
0
i
the concentration of corrole metal complex (v is the rate of reduction
0
of cytochrome c by superoxide and v is the rate of reduction of
i
cytochrome c inhibited by corrole metal complex); IC50 is obtained
when v /v − 1 = 1. Possible interference by inhibition of xanthine
(27) Mahammed, A.; Gross, Z. Chem. Commun. 2010, 46, 7040.
(28) Mahammed, A.; Gross, Z. Catal. Sci. Technol. 2011, 1, 535.
0
i
oxidase reaction by the examined compounds was ruled out following
8
089
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090

Inorganic Chemistry
Article
(
29) Okun, Z.; Kupershmidt, L.; Amit, T.; Mandel, S.; Bar-Am, O.;
Youdim, M. B. H.; Gross, Z. ACS Chem. Biol. 2009, 4, 910.
30) Haber, A.; Mahammed, A.; Fuhrman, B.; Volkova, N.; Coleman,
R.; Hayek, T.; Aviram, M.; Gross, Z. Angew. Chem., Int. Ed. 2008, 47,
896.
(
7
(
31) Haber, A.; Aviram, M.; Gross, Z. Chem. Sci. 2011, 2, 295.
(
32) Kupershmidt, L.; Okun, Z.; Amit, T.; Mandel, S.; Saltsman, I.;
Mahammed, A.; Bar-Am, O.; Gross, Z.; Youdim, M. B. H. J.
Neurochem. 2010, 113, 363.
(
33) Kanamori, A.; Catrinescu, M. M.; Mahammed, A.; Gross, Z.;
Levin, L. A. J. Neurochem. 2010, 114, 488.
(
(
(
(
(
(
(
34) Gross, Z.; Galili, N. Angew. Chem., Int. Ed. 1999, 38, 2366.
35) Szabo, C. Br. J. Pharmacol. 2009, 156, 713.
36) Crow, J. P.; Ischiropoulos, H. Nitric Oxide, Pt B 1996, 269, 185.
37) Peluffo, G.; Radi, R. Cardiovasc. Res. 2007, 75, 291.
38) Radi, R. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 4003.
39) Olmos, A.; Giner, R. M.; Manez, S. Med. Res. Rev. 2007, 27, 1.
40) Pacher, P.; Beckman, J. S.; Liaudet, L. Physiol. Rev. 2007, 87,
3
(
15.
41) MacMillan-Crow, L. A.; Crow, J. P.; Thompson, J. A.
Biochemistry 1998, 37, 1613.
(
(
42) McCord, J. M.; Fridovich, I. J. Biol. Chem. 1969, 244, 6049.
43) Gryko, D. T.; Piechota, K. E. J. Porphyrins Phthalocyanines 2002,
81.
(
(
44) Hunt, J. A.; Lee, J.; Groves, J. T. Chem. Biol. 1997, 4, 845.
45) Coddington, J. W.; Hurst, J. K.; Lymar, S. V. J. Am. Chem. Soc.
1
999, 121, 2438.
(
(
46) Edwards, J. O.; Plumb, R. C. Prog. Inorg. Chem. 1994, 41, 599.
47) Lymar, S. V.; Hurst, J. K. Inorg. Chem. 1998, 37, 294.
8
090
dx.doi.org/10.1021/ic300408s | Inorg. Chem. 2012, 51, 8083−8090