New PEG-ylated Mn(iii) porphyrins approaching catalytic activity of SOD enzyme

Article abstract of DOI:10.1039/b513761f

Two new tri(ethyleneglycol)-derivatized Mn(iii) porphyrins were synthesized with the aim of increasing their bioavailability, and blood-circulating half-life. These are Mn(iii) tetrakis(N-(1-(2-(2-(2-methoxyethoxy)ethoxy)ethyl) pyridinium-2-yl)porphyrin,

Full text of DOI:10.1039/b513761f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
New PEG-ylated Mn(III) porphyrins approaching catalytic activity
of SOD enzyme
a
b
c
a
Ines Batini c´ -Haberle,* Ivan Spasojevi c´ , Robert D. Stevens, Bruce Bondurant,
d
d ˇ
a
a
Ayako Okado-Matsumoto, Irwin Fridovich, Zeljko Vuja sˇ kovi c´ and Mark W. Dewhirst
Received 28th September 2005, Accepted 5th December 2005
First published as an Advance Article on the web 19th December 2005
DOI: 10.1039/b513761f
Two new tri(ethyleneglycol)-derivatized Mn(III) porphyrins were synthesized with the aim of
increasing their bioavailability, and blood-circulating half-life. These are Mn(III) tetrakis(N-(1-(2-(2-(2-
5
+
methoxyethoxy)ethoxy)ethyl)pyridinium-2-yl)porphyrin, MnTTEG-2-PyP and Mn(III)
ꢀ
tetrakis(N,N -di(1-(2-(2-(2-methoxyethoxy)ethoxy)ethyl)imidazolium-2-yl)porphyrin,
5
+
MnTDTEG-2-ImP . Both porphyrins have ortho pyridyl or di-ortho imidazolyl electron-withdrawing
substituents at meso positions of the porphyrin ring that assure highly positive metal centered redox
5
+
potentials, E1/2 = +250 mV vs. NHE for MnTTEG-2-PyP and E1/2 = + 412 mV vs. NHE for
5
+
•−
MnTDTEG-2-ImP . As expected, from established E1/2 vs. log kcat(O
2
) structure–activity
relationships for metalloporphyrins (Batinic-Haberle et al., Inorg. Chem., 1999, 38, 4011), both
compounds exhibit higher SOD-like activity than any meso-substituted Mn(III) porphyrins-based SOD
5
+
5+
mimic thus far, log kcat = 8.11 (MnTTEG-2-PyP ) and log kcat = 8.55 (MnTDTEG-2-ImP ), the
former being only a few-fold less potent in disproportionating O than the SOD enzyme itself. The
new porphyrins are stable to both acid and EDTA, and non toxic to E. coli. Despite elongated
•
−
2
5
+
substituents, which could potentially lower their ability to cross the cell wall, MnTTEG-2-PyP and
5
+
MnTDTEG-2-ImP exhibit similar protection of SOD-deficient E. coli as their much smaller ethyl
5
+
5+
analogues MnTE-2-PyP and MnTDE-2-ImP , respectively. Consequently, with anticipated increased
blood-circulating half-life, these new Mn(III) porphyrins may be more effective in ameliorating oxidative
stress injuries than ethyl analogues that have been already successfully explored in vivo.
1
5–18
19–21
Introduction
in cancer,
and diabetes,
may be attributed to decreasing
oxidative stress through direct scavenging of ROS and RNS
and/or by affecting signaling pathways. The latter effect may
be either a consequence of the attenuation of levels of ROS
and RNS signaling molecules,
interaction of Mn porphyrins with signaling proteins. While the
former is the consequence of their antioxidant ability, the latter
1
Structure–activity relationships have guided the development
of Mn porphyrins with highly positive metal-centered redox
potentials of ≥ + 200 mV vs. NHE (mainly ortho isomers of N-
15–18,21
or it may result from direct
ꢀ
substituted pyridyl- or N,N -disubstituted imidazolylporphyrins),
21
which possess high antioxidant capacity, i.e. ability to scavenge
•
−
reactive oxygen (ROS) and nitrogen species (RNS), such as O
2
,
21
presumably results from pro-oxidant ability of these compounds.
Such divergent abilities of Mn P arise from their biologically
−
•−
•
ONOO , CO
3
and NO. The rate constants are in excess of
III 5+
7
−1
−1
6
−1
−1
1
0 M
s
for the former three species and 10 M
The ability of Mn porphyrins to eliminate O
s
for
parallels
Both, the
compatible metal-centered redox potential, E/12 = ≥ + 200 mV
•
1–12
•−
NO.
2
II 4+
vs. NHE which allows them to be easily reduced (to Mn P ) and
−
•−
•
8,12,13
their ability to scavenge ONOO , CO
compounds with Mn in its +3 and +2 oxidation state are reactive
towards these species.
species by Mn porphyrins may be coupled with cellular reductants
such as ascorbic acid, glutathione and tetrahydrobiopterin.
There is growing evidence that the remarkable protective effects
of Mn porphyrins in animal models of diseases, particularly
3
and NO.
IV 4+
V
5+
equally easily oxidized (to O=Mn P and to O=Mn P ) with
cellular components and ROS and RNS.
1
,5,9,8,12,13
In vivo, the elimination of these
While we have already succeeded in developing compounds
with fairly high antioxidant capacity,
these compounds is limited by their excessive hydrophilicity, their
1,4,11
the in vivo efficacy of
14
22,23
relatively low blood-circulating half-life (≤2 h),
and their tox-
icity, partly due to their accumulation in liver. We have decreased
hydrophilicity of Mn porphyrins up to 10-fold by increasing the
length of the alkyl chains in Mn(III) alkylpyridylporphyrins from
a
Department of Radiation Oncology, Duke University Medical School,
24,25
methyl to n-octyl, thereby increasing their entry into the E. coli.
However, this modification increased their toxicity to E. coli
presumably due to the increased surfactancy and solubilization
of lipid cellular components.
Herein, we have further modified the porphyrin meso sub-
stituents while retaining ortho quaternized nitrogens to assure high
and biologically compatible E1/2 and high antioxidant capacity.
Durham, NC 27710, USA. E-mail: ibatinic@duke.edu; Fax: +1-919-684-
8
885; Tel: +1-919-684-2101
b
Department of Medicine, Duke University Medical School, Durham, NC
2
7710, USA
25
c
Department of Pediatrics, Duke University Medical School, Durham, NC
7710, USA
2
d
Department of Biochemistry, Duke University Medical School, Durham,
NC 27710, USA
This journal is © The Royal Society of Chemistry 2006
Dalton Trans., 2006, 617–624 | 617

View Article Online
We synthesized and characterized two new Mn(III) porphyrins as
shown in Scheme 1. With 1-(2-(2-(2-methoxyethoxy)ethoxy)ethyls
in ortho positions of meso pyridyl and imidazolyl substituents
60.51 mmoles) was combined with triethyl amine (9.10 g, 12.7 ml,
90.8 mmoles) in dry, alcohol-free chloroform (200 mL) under
nitrogen. The reaction mixture was cooled to 0 C in an ice
brine bath. The 4-toluene sulfonyl chloride (15.0 g, 78.7 mmoles)
was added, and the reaction was stirred on ice for 20 min. The
ice was removed, and the reaction was allowed to reach room
temperature and stirred overnight under nitrogen. 100 mL of
toluene was added to the reaction mixture, and the solvent was
removed under reduced pressure. The residue was re-dissolved in
◦
26–28
we aimed at significantly increasing blood-circulating half-life,
and decreasing unfavorable interactions with biological molecules.
The presence of oxygen atoms in substituents on pyridyls and imi-
dazolyls eliminated their surfactant-like properties. Consequently,
they were not toxic in our simple model of oxidative stress, SOD-
deficient E. coli.
3
00 mL of ethyl acetate and the triethylammonium chloride was
removed by vacuum filtration. The solid was washed twice with
0 mL of ethyl acetate, and the organic filtrates were combined.
The ethyl acetate solution was washed once with 100 mL of 1 M
KHSO followed by 100 mL of saturated aqueous NaHCO , and
5
4
3
twice with 100 mL of water. The organic phase was dried over
magnesium sulfate and the solvent was removed under reduced
pressure to yield 25 g of crude amber oil. The crude product was
purified by chromatography as follows: Fine silica gel (300 mL)
was packed in an 8 cm diameter chromatography column in 30%
benzene and 70% hexane. The crude product was dissolved in
1
00 mL of 40% benzene, and 60% hexane, and loaded onto the
column. The column was eluted with 300 ml of 30% benzene and
7
1
0% hexane, followed by a gradient over 2400 mL starting with
0% ethyl acetate, 90% hexane and ending with 100% ethyl acetate.
Evaporation of product containing fractions produces pure 4-
toluenesulfonic acid 1-(2-(2(-2-methoxyethoxy)ethoxy)ethyl ester
1
(
7
18 g, 90% yield). H NMR of TEGTs (300 MHz, CDCl
3
): d 7.812–
) 7.36–7.33 (d, J = 8.31 Hz, 2H,
) 4.18–4.15 (t, J = 6.6 Hz, 2H, SO CH ) 3.70–3.67 (t,
J = 4.8, 2H, SO CH CH ) 3.59–3.53 (m, 8H, CH (OCH CH
3.71 (s, 3H, CH (OCH CH ) ) 2.45 (s 3H, CH C(CH) ) ppm.
Scheme 1 Structures of Mn(III) porphyrins studied.
.785 (d, J = 8.24 Hz, 2H, CS(CH)
2
CH C(CH)
3
2
3
2
3
2
2
3
2
2
2
) )
Experimental
3
2
2
3
3
2
General
Mn(III) porphyrins. The quaternization of H
2
T-2-PyP and
◦
H
2
T-2-ImP was achieved at 105 C under nitrogen (or argon)
The H
2
T-2-PyP and H
2
T-2-ImP were supplied by MidCentury
O was purchased
and in the presence of anhydrous K
2
CO
3
to avoid undesired
Chemicals (Chicago, IL). MnCl
2
× 4 H
2
11,30
side products.
A major side reaction that can occur is an
from J. T. Baker. N,N-dimethylformamide (DMF), NaCl, sodium
L-ascorbate, monomethoxy tri(ethylene glycol), and silica gel
polyester plates (#805 013) were from Aldrich, while xanthine,
attack on b-hydrogen resulting in elimination (E2). The reaction
thus may produce the corresponding vinyl ether as well as one
equivalent of acid which can deactivate the nucleophile toward
further reaction. Potassium carbonate is an insoluble base, and
would scavenge acid thereby maintaining the reactivity of pyridine
or imidazole nucleophile throughout the course of reaction. With
K
2
CO
were from Sigma. Acetone, ethyl ether (anhydrous), chloroform,
NaCl, KOH, KH PO , methanol, EDTA and KNO were from
Mallinckrodt and acetonitrile was from Fisher Scientific. H
0% solution was from Mallinkrodt. Tris (ultra pure) was from
3
, albumin, and equine ferricytochrome c (lot # 7752)
2
4
3
2
2
O ,
pyridylporphyrin (imidazolylporphyrin) 5.2 mg (8.2 mg) of H
2
T-2-
3
◦
PyP in 1.5 mL (2.0 mL) of DMF, was heated to ∼105 C, purged
ICN Biomedicals, Inc., and ultrapure argon and nitrogen from
National Welders Supply Co. Anion exchange resin BIO-RAD
AG 21 K (50–100 mesh, chloride form) was from Bio-Rad Labo-
ratories (Richmond, CA). Amicon ultrafiltration cellulose acetate
membrane, 25 mm diameter (YC05) was purchased from Millipore
Corporation (Bedford, MA) and used along with Amicon ultrafil-
tration device. Xanthine oxidase was prepared by R. Wiley and was
with argon or nitrogen, and followed by the addition of 50 mg
(
62 mg) of anhydrous K
course of reaction was followed by thin-layer chromatography on
silica gel plates using acetonitrile/H O/KNO -saturated H
O =
/1/1. The completion of quaternization, as observed by a single
2
CO
3
and 0.7 mL (1 mL) of TEGTs. The
2
3
2
8
spot on a silica plate, occurred within 2 days for pyridyl and
5 days for imidazolyl compound, which is in agreement with
bigger steric hindrance in the case of imidazolyl compound
bearing 4 more substituents than pyridyl analogue. The porphyrins
were purified from the excess tosylate and DMF by the chlo-
29
supplied by K. V. Rajagopalan. The 4-toluenesulfonic acid 1-(2-
2(-2-methoxyethoxy)ethoxy)ethyl ester (TEGTs) was synthesized
(
1
herein and analyzed on purity by H NMR. The M9CA medium
was prepared as described previously.
25
1
roform/water extraction as described previously. We were not
1,11,24
able to metallate porphyrins in the way we did previously.
Thus, the aqueous solution of metal-free quaternized pyridyl- and
Synthesis
4
-Toluenesulfonic acid 1-(2-(2(-2-methoxyethoxy)ethoxy)ethyl
imidazolylporphyrin was acidified with HCl to pH ∼2, added
◦
ester (TEGTs). Monomethoxy tri(ethylene glycol) (9.9 g,
∼20-fold excess of MnCl
2
and heated at ∼100 C for ∼24 and
6
18 | Dalton Trans., 2006, 617–624
This journal is © The Royal Society of Chemistry 2006

View Article Online
•
−
∼
72 h, respectively. Steric hindrance of elongated substituents
Catalysis of O
2
disproportionation
accounted for a longer metallation when compared to n-hexyl-
•
−
24
The catalytic rate constants for the O
determined by cytochrome c assay.
2
disproportionation were
The xanthine/xanthine
and n-octylpyridylporphyrins. The completion of metallation
was detected spectrophotometrically. Mn porphyrins could not
be precipitated out of water either as PF
salt. Thus, we exchanged the tosylate anion by a chloride anion
exchange resin. We then purified the Cl salt solution of the Mn
porphyrins using a 500 molecular weight cut-off ultrafiltration
membrane. There was up to 10% loss of the porphyrins during
ultrafiltration purification. Routinely, a few mL of the ∼1 mM
solution was washed with ∼200 mL of water. Electrospray mass
spectroscopy was used for the identity and purity analysis of
the porphyrins. The aqueous solutions of pure compounds were
lyophilized to allow for elemental analysis.
1
,21,33,34
−
−
oxidase (40 lM xanthine, ∼2 nM xanthine oxidase) was the source
•
−
2
6
salt, or acetone as Cl
of O
of O
and ferricytochrome c was used as the indicating scavenger
•
−
−
2
. The reduction of cyt c was followed at 550 nm. Assays
◦
were conducted at 25 ± 1 C in 0.05 M phosphate buffer, pH 7.8,
−
1
0
.1 mM EDTA in the presence and absence of 15 lg mL catalase.
•
−
Rate constants for the reaction of metalloporphyrins with O
2
were based upon the competition with 10 lM cyt c, kcytc = 2.6 ×
5
−1 −1
34
•−
1
0 M
s
as described elsewhere. The O
2
was produced at
−
1
1
.2 lM min . Any possible interference of Mn porphyrins with
production of O
95 nm in the absence of cyt c. No reoxidation of cyt c with
•
−
2
was examined following urate formation at
2
Mn(III) porphyrins was observed.
Elemental analysis.
MnTTEG-2-PyPCl
2.26; N, 7.13; H, 5.83. Calculated: C, 52.23; N, 7.16; H, 6.31.
5
× 7H
2
O (C68
H
98
N
8
O
19MnCl ). Found: C,
5
Aerobic growth of E. coli
5
The wild type AB1157 and SOD-deficient JI132 E. coli were
grown aerobically in M9CA medium as already described with
MnTDTEG-2-ImPCl × 13H O × HCONH
MnCl ). Found: C, 46.61; N, 8.01; H, 6.84. Calculated: C, 46.61;
N, 7.94; H, 7.37. The elemental analysis and electrospray mass
5
2
2
. (C89
H
167
N
13
O
38
-
5
0
–30 lM Mn(III) porphyrins in 5 mL volume in 16 × 150 mm
◦
11,14
test tubes in a thermostated-orbital shaker (45 angle, 200 rpm)
at 37 C.
spectroscopy have proven before
and herein that compounds
◦
11,25
Rates of growth were followed turbidimetrically at
that have long substituents on either pyridyl or imidazolyl groups
31
650 nm (using Thermo-max UV/vis plate-reader-Molecular De-
vices) to minimize interference from the absorbance of porphyrins
and were compared to the growth of E. coli in the presence
tend to hold solvent (formamide) or other molecules within well-
11,24
defined cavities.
UV/vis spectra of porphyrin ligands and their Mn complexes
were taken on a Shimadzu UV-2501-PC spectrophotometer.
Thin-layer chromatography of ligands and Mn complexes was
5
+
of reference/standard MnTE-2PyP . The published data on
5
+
11
5+
MnTDE-2-ImP , corrected to MnTE-2-PyP , are also shown.
performed using Aldrich silica gel polyester plates with KNO
O/H
O/acetonitrile = 1/1/8.
3
sat.
H
2
2
Results
UV/vis spectroscopy
Electrochemistry
−
7
The porphyrins obeyed the Beer–Lambert law from 10 M to
Measurements were performed on a CH Instruments Model 600
Voltammetric Analyzer. A three-electrode system in a small
volume cell (0.5 mL to 3 mL), with a 3 mm-diameter glassy
carbon button working electrode (Bioanalytical Systems), plus the
Ag/AgCl reference and Pt auxilliary electrodes was used. Argon-
purged solutions contained 0.05 M phosphate buffer, pH 7.8,
−5
5+
1
0
M. A generally observed red shift relative to MnTE-2-PyP
1
5+1,11
and MnTDE-2-ImP
was the result of the increased porphyrin
macrocycle distortion due to the enhanced crowding upon the
replacement of ethyl with 1-(2-(2-(2-methoxyethoxy)ethoxy)ethyl
groups (Table 1). Molar absorption coefficients along with absorp-
tion maxima are given in Table 1.
0
0
.1 M NaCl, and 0.5 mM metalloporphyrin. The scan rates were
−
1
−1
.01–0.5 V s , typically 0.1 V s . The potentials were standardized
Stability towards oxidative degradation. Both porphyrins are
stable in aqueous solution with Mn in its +3 oxidation state as
evidenced by the splitting of the Soret band in their spectra. The
32
against the potassium ferrocyanide/ferricyanide, and/or against
MnTE-2-PyP .
5
+ 1
3
+
2+ 6
splitting is absent once Mn is reduced to Mn . In the presence
of 70-fold excess ascorbic acid, at pH 7.8 (0.05 M tris buffer),
under aerobic conditions, Mn(III) porphyrins (7 lM) under-
went reduction, followed by oxidative degradation. As discussed
Electrospray mass spectrometry
ESMS measurements were performed on a Micromass Quattro LC
triple quadrupole mass spectrometer with 0.5 mM 50% aqueous
11,24
previously,
the rate of degradation depends upon the value of
−
acetonitrile Cl salt solution of Mn(III) complexes, as well as of
the metal-centered redox potential and upon the steric hindrance
11,24
ꢀ
MnCl
2
at 20 and 30 V cone voltages as described previously.
of the metal site by ortho N-pyridyl or di-ortho N,N -imidazolyl
5+
5+
Table 1 Molar absorption coefficients of MnTTEG-2-PyP and MnTDTEG-2-ImP
a
Mn porphyrin
kmax/nm, (log e)
5
+
MnTTEG-2-PyP
262 (4.56); 364.5 (4.72); 455 (5.21); 497 (3.85); 560 (4.14); 782.5 (3.34)
347 (3.60); 448 (5.03); 502 (4.70); 557(4.06); 590(3.95); 799 (3.31)
5+
MnTDTEG-2-ImP
a
Molar absorptivities (M cm−1) (± 5%) were determined in water at room temperature. kmax were determined with errors inside ± 0.5 nm.
−1
This journal is © The Royal Society of Chemistry 2006
Dalton Trans., 2006, 617–624 | 619

View Article Online
•
−
substituents. With higher positive E1/2, and protective effects of
elongated side chains, the new compounds are more stable in
the reduced state and thus less prone to oxidation than their
Table 2 Metal-centered redox potentials E1/2, log kcat for O
2
dismuta-
for MnTTEG-2-PyP and MnTDTEG-2-
ImP and their ethyl analogues
5+
tion and retention factors, R
f
5+
5
+
ethyl analogues. Hence, the MnTTEG-2-PyP and MnTDTEG-
a
b
c
Porphyrin
E
/12, mV vs NHE
log kcat
R
f
5
+
2
-ImP underwent 8.0% and 8.5% degradation within 190 min
5
+
5
+
MnTTEG-2-PyP
+250
+412
+346
+228
8.11
8.55
7.83
0.041
0.082
0.095
0.041
and 207 min, respectively, while 35% of MnTE-2-PyP and 17%
of MnTDE-2-ImP were degraded within 161 and 216 min in the
presence of aerobic ascorbate, respectively.
5+
5
+
MnTDTEG-2-ImP
5+
d
e
MnTDE-2-ImP
5+
e
f
MnTE-2-PyP
7.76
a
Acid stability. Within the 2 week-duration of experiments, only
10% of MnTTEG-2-PyP was demetallated in the presence of
E
1/2 (± 3 mV) determined in 0.1 M NaCl, 0.05 M phosphate buffer,
b
◦
c
5
+
pH 7.8,
k
cat (± 20%) determined by cyt c assay, pH 7.8, (25 ± 1) C,
f
R ,
∼
compound path/solvent path on silica gel TLC plates in KNO -saturated
H
ref. 11. Data from ref. 1.
3
3
6% HCl, while no demetallation occurred in 1 M HCl. Despite
d
e
2
O/H
2
f
O/acetonitrile = 1/1/8, Data from ref. 53 and 11. Data from
2
+
the high metal-centered redox potential (which stabilizes the Mn
oxidation state and thus decreases metal/ligand stability) the
imidazolyl compound was very stable in acid. No demetallation
of MnTDTEG-2-ImP in 1 M HCl was observed within 2 weeks.
In 36% HCl the compound loses metal, yet very slowly with no
significant loss within the first 2 h.
5
+
5+
phyrin, MnTDTEG-2-ImP is 162 mV more positive than of the
5
+
ortho-substituted pyridylporphyrin, MnTTEG-2-PyP (+228 mV
vs. NHE). This may be partly due to charge delocalization
over the imidazolyl ring. However, because it is significantly
higher than of its ethyl, MnTDE-2-ImP , and monomethoxyethyl
derivative, MnTDMOE-2-ImP (+346 and +365 mV vs. NHE,
Stability towards EDTA. . No demetallation of 7 lM
5+
5
+
5+
MnTTEG-2-PyP and MnTDTEG-2-ImP occurred within 2
weeks at 12 000-fold excess EDTA.
5+
11
respectively), another effect may also be responsible.
Thin-layer chromatography
Electrospray mass spectrometry
The retention factors, R (porphyrin path/solvent path) for
f
Similar patterns of fragmentation shown in Fig. 1 and Table 3
were observed as for other cationic pyridyl- and imidazolyl-
Mn(III) porphyrins are given in Table 2. Both compounds are
11
of nearly identical hydrophilicities as their ethyl analogues.
11,24
5
+
porphyrins.
The singly and doubly-reduced porphyrins, species
Surprisingly, the 12 and 24 oxygen atoms in MnTTEG-2-PyP
5
+
with one or two chlorides associated with a metal center, sol-
and MnTDTEG-2-ImP do not increase their hydrophilicity
relative to their ethyl analogues. It may be that the effects of
oxygens are offset by the intervening ethylene moieties.
vated porphyrins and species with fragmented side chains were
•
observed. Species consistent with the loss of CH
2
–O–CH
2
CH –
2
O–CH
2
CH
2
–O–CH were also observed in the spectra of both por-
3
+
phyrins, presumably resulting from fission within N –CH
O–CH
2
CH
2
–
Electrochemistry
11
2
CH
2
–O–CH
2
CH
2
–O–CH .
3
Similar findings were previ-
Cyclic voltammograms, ascribed to the Mn(III)/Mn(II) redox
ously reported for mesoporphyrin-IX dimethyl ester, where the
•
couple, were reversible. The E1/2 is +250 mV vs. NHE for
so-called “benzylic” fission of CH
pionate was detected. Species holding formamide, that originates
2
–COO–CH from methyl pro-
3
5
+
35
MnTTEG-2-PyP and +412 mV vs. NHE for MnTDTEG-2-
5
+
31
ImP (Table 2). The E1/2 of di-ortho substituted imidazolylpor-
from prolonged heating with DMF, were also seen.
5+
5+
2 2
Fig. 1 The ESMS spectra of MnTDTEG-2-ImP , MnTTEG-2-PyP and MnCl . The mass spectra of Mn(III) complexes and MnCl in 50% aqueous
−
1
acetonitrile were scanned at 100 a.m.u. s and at 30 V cone voltage.
6
20 | Dalton Trans., 2006, 617–624
This journal is © The Royal Society of Chemistry 2006

View Article Online
5+
5+a
Table 3 ESMS of MnTTEG-2-PyP and MnTDTEG-2-ImP
5+
5+
Species
MnTTEG-2-PyP
MnTDTEG-2-ImP
III 5+
Mn P /5
361.4
458.9
623.4
III 5+
−
Mn P + Cl /4
323.7
443.3
382.6
290.7
399.7
III 5+
−
Mn P + 2Cl /3
III 4+
+
−
+
Mn P − a + Cl /3
III 5+
−
+
Mn P + Cl − a + CH
3
/4
425.9
579.9
492.9
450.2
611.8
550.6
417.2
567.7
934.8
594.6
668.9
480.4
428.5
III 5+
−
+
+
Mn P + 2Cl − a + CH
3
/3
O/4
III 5+
−
Mn P + Cl + 2AN + 3H
2
II 4+
Mn P /4
315.0
431.6
371.0
281.9
387.6
665.6
420.1
II 4+
−
Mn P + Cl /3
II 4+
+
Mn P − a /3
II 4+
+
+
Mn P − a + CH
3
/4
II 4+
−
+
+
Mn P + Cl − a + CH
3
I
/3
II 4+
−
Mn P + 2Cl /2
II •3+
I
3+
Mn P /3 or Mn P /3
II •3+
Mn P + 6AN/3 or Mn P + 6AN/3
III 5+
−
+
Mn P + AN + Cl + HCONH
2
/4
/4
II 5+
+
Mn P − a + CH
3 2
+ HCONH
a
∼
0.5 mM solutions of porphyrins in 1 : 1 = acetonitrile : H
2
O, 30 V cone voltage. The a is 1-(2-(2-(2-methoxyethoxy)ethoxy)ethyl and AN is acetonitrile.
•
The fission of CH
2
–O–CH
2
CH
2
–O–CH
2
CH
2
–O–CH
3
radical from 1-(2-(2-(2-methoxyethoxy)ethoxy)ethyl is noted for simplicity as the loss of 1-(2-(2(-
+
+
2+
2
-methoxyethoxy)ethoxy)ethyl (a ) and the gain of CH
3
. In the ESMS of MnCl
2
the peaks arising from 4, 5 and 6 acetonitriles coordinated to Mn were
, formamide
observed, m/z being 109.4, 129.9 and 150.6, respectively, while the m/z 102.0, 142.9, 122.6, and 134.9 are the background peaks. HCONH
originates from prolonged heating of porphyrin with DMF.
2
3
1
36–38
Previously published data
support our observation that free
manganese impurity in Mn porphyrin preparations affects the
outcome of E. coli studies, and may thus affect in vivo studies
as well. In the ESMS of MnCl
2
, in 50% aqueous acetonitrile,
complexes with four to six acetonitriles (AN) were observed. The
2
+
most prominent peak comes from m/z = 129.9 (Mn(AN)
5
/2).
The lack of m/z = 129.9 in the spectra of porphyrins supports the
purity of the preparations (Fig. 1, Table 3).
•
−
Catalysis of O
2
disproportionation
The catalytic rate constants of the new porphyrins were found to
be higher than those of the previously described meso-substituted
Mn(III) porphyrins. This is in agreement with E1/2 vs. log kcat
structure–activity relationships (Fig. 2—the rising limb of the
1
,11
5+
bell shape curve), where MnTTEG-2-PyP and MnTDTEG-
2
5
+
-ImP are indicated as #18 and #19, respectively.
Aerobic growth of E. coli
None of the compounds were toxic to wild type E. coli in the
concentration range tested, up to 30 lM (Fig. 3D). The protection
of SOD-deficient E. coli was concentration-dependent. The ability
Fig. 2 Reactivity of manganese complexes expressed in terms of log
cat vs. metal-centered redox potential, E1/2. Only water-soluble Mn(III)
porphyrins are given in the left, linear section of the curve that
k
5
+
5+
of MnTTEG-2-PyP and MnTDTEG-2-ImP to protect SOD-
1,39–41
5
+
obeys the Marcus equation
Mn TCPP , (#2) Mn T(TMAP) , (#3) Mn T(2,6-F
and data (᭹) are from ref. 1: (#1)
deficient E. coli was similar to their ethyl analogues MnTE-2-PyP
and MnTDE-2-PyP , respectively (Fig. 3).
III
3−
III
5+
III
3−
5
+
2
-3-SO
-P)P
3
-P)P
,
III
5+
III
3−
(
#4) Mn T(TFTMAP)P , (#5) Mn T(2,6-Cl
2
-3-SO
3
,
(#6)
III
3+
III
5+
III
5+
Mn BM-2-PyP , (#7) Mn TM-3-PyP , (#8) Mn TM-4-PyP , (#9)
III
4+
III
5+
III
5+
Discussion
Mn Tr-2-PyP , (#10) Mn TM-2-PyP , (#11) Mn TE-2-PyP . Data
for EUK-8 is from ref. 32 and for MnCl from ref. 2 and 33, data for
Mn(II) cyclic polyamine M40403 from ref. 52. Data for SOD is from
2
We have previously established the structure–activity relationships
•
−
III
5+
between E1/2 and log kcat for the O
2
disproportionation by Mn(III)
ref. 47. Data for Mn Cl1–4TE-2-PyP (#12–15) are from ref. 4, data for
1
II
4+
II
4+
porphyrins. (The porphyrin with Mn in its +3 state is assigned as
a neutral species).
Mn Br
8 5
TM-4-PyP (#16) are from ref. 7, and for Mn Cl TE-2-PyP
III
5+
(#17) from ref. 43 (ꢀ). Data for Mn TnBu-2-PyP (a) is from ref. 24;
III
5+
III
5+
Mn TMOE-2-PyP (b) from ref. 11; Mn TD(M)E-2-ImP (c,d) from
III
•−
II
−
Mn P + O
2
ꢀ Mn P + O
2
, kred
(1)
(2)
III
5+
III
5+
ref. 11 and 53; Mn TM,MOE-2-PyP (e), and for Mn TDMOE-2-ImP
III
5+
(
f) from ref. 11 (ꢁ). Data point #18 and #19 (᭛) are Mn TTEG-2-PyP
II
−
•−
+
III
III 5+
Mn P + O
2
+ 2H ꢀ Mn P + H
2
O
2
, kox
and Mn TDTEG-2-ImP , respectively from this work.
This journal is © The Royal Society of Chemistry 2006
Dalton Trans., 2006, 617–624 | 621

View Article Online
5
+
5+
Fig. 3 Aerobic growth of SOD-deficient E. coli JI132 (C, ᭹) in the absence and presence of MnTTEG-2-PyP (TEGPy), MnTDTEG-2-ImP (TEGIm),
5+
5+
and their ethyl analogues MnTE-2-PyP (EPy) and MnTDE-2-ImP (DEIm) in M9CA medium at 3 lM (A), 10 lM (B) and 30 lM concentrations (C).
Aerobic growth of SOD-proficient E. coli AB1157 in M9CA medium in the presence of 30 lM MnTTEG-2-PyP (TEGPy), and MnTDTEG-2-ImP
TEGIm) (D). The published data on MnTDE-2-ImP , corrected to reference/standard MnTE-2-PyP , are shown.
5+
5+
5+
11
5+
(
3
9–41
The relationships obey the Marcus equation,
whereby kcat
lipophilicity was enhanced ∼10-fold going from methyl to n-
24
increases 10-fold with each 120 mV increase in E1/2 (the rising
limb of the bell shape curve in Fig. 2). As we approach the midway
potential (+360 mV vs. NHE) for the reduction and oxidation of
octyl porphyrin. However, increasing the length of alkyl chains
increased the surfactant properties of compounds; i.e. n-hexyl
5
+
5+
(MnTnHex-2-PyP ) and n-octyl (MnTnOct-2-PyP ) porphyrins
•
−
42
25
O
2
,
there is an equal thermodynamic driving force for both
became significantly toxic to E. coli at concentrations ≥3 lM.
•
−
half-reactions of the catalytic cycle (eqn (1) and (2)). Thus, on
thermodynamic grounds, we are in the region of maximal values
for kcat. Any further increase in E1/2 stabilizes the +2 oxidation
state of Mn, so that the oxidation of Mn(II) compounds becomes
rate-limiting, accompanied by a decrease in log kcat (falling limb of
Fig. 2, for discussion see also ref. 11). The two b-substituted Mn(II)
porphyrins of fairly high SOD-like activity have Mn in its +2
We managed to eliminate toxicity while maintaining the same O
2
dismuting ability by introducing an oxygen atom into each pyridyl
11
and imidazolyl substituent.
Herein we addressed the insufficient blood circulating half-
life by synthesizing tri(ethyleneglycol)-derivatized pyridyl- and
imidazolylporphyrins (Scheme I). We retained positively charged
electron-withdrawing ortho nitrogens which resulted in high E1/2
.
7,43
oxidation state and thus suffer from low metal/ligand stability.
As expected, based on E1/2 vs. log kcat dependence, the compounds
proved to be of high SOD-like activity, fitting the relationships
shown in Fig. 2 (Table 2). With log kcat = 8.55, the MnTDTEG-2-
4
+
Those are MnBr
8
TM-4-PyP (E1/2 = + 480 mV vs. NHE, log kcat
=
8
.08
−1
8
.34, metal/ligand stability constant being 10
M , #16 in
7
4+
5+
•−
Fig. 2), and MnCl
cat = 8.41 with presumably even lower stability, #17 in Fig. 2).
Based on such relationships the Mn(III) porphyrins that
bear electron-withdrawing quaternized ortho pyridyl nitrogens
5
TE-2-PyP (E1/2 = + 560 mV vs. NHE, log
ImP was only 2-fold or 5.6-fold less potent in O
2
disproportion-
43
k
ation than the SOD enzyme itself, given the log kcat of SOD is 8.85
47
48–50
(recently reported by Michel et al.), or 9.3,
respectively. It is
the most potent meso-substituted Mn(III) porphyrin antioxidant
reported. Although fairly reducible, as indicated by high E1/2 of
5
+
(
MnTM(E)-2-PyP ) were tested, and proved effective in differ-
1
5–23,44,45
5+
ent animal models of oxidative stress injuries.
The di-
+412 mV vs. NHE (Table 2), the MnTDTEG-2-ImP was still
5
+
+
ortho substituted imidazolylporphyrin, MnTDE-2-ImP was also
remarkably stable towards H and EDTA. (A very potent SOD-
11,21,22,46
5+
synthesized and shown effective in vivo.
Both pyridyl and
mimic, b-substituted MnCl TE-2-PyP , indicated as #15 in Fig. 2,
4
imidazolyl compounds suffer from excessive hydrophilicity and
whose log kcat is 8.60, is insufficiently stable towards both EDTA
and proton, as implied by its E1/2 of +448 mV vs. NHE, to be of
practical use). Such high stability of MnTDTEG-2-ImP likely
results from the shielding of the Mn site by long side chains.
22,23
low blood-circulating half-life.
To lower the hydrophilicity
24
4
5+
we synthesized longer alkyl-chain pyridylporphyrins. The abil-
•
−
−
24
ity to scavenge O
2
and ONOO were maintained, and the
6
22 | Dalton Trans., 2006, 617–624
This journal is © The Royal Society of Chemistry 2006

View Article Online
For this reason, and the higher E1/2, the new compounds are
also significantly more stable towards oxidative degradation than
their ethyl analogues. Furthermore, such elongated chains would
decrease the interactions of porphyrins with biological molecules
that are presumably responsible for neuro- and liver toxicity of
presently being successfully employed in studying oxidative stress
injuries in vivo.
Abbreviations
5
+
22,23,51
III/II 5+/4+
MnTE-2-PyP .
We have already observed that alkylpyridyl
PEG, poly(ethyleneglycol), PEG; Mn
P
, any Mn porphyrin
5
+
porphyrins with longer chains (MnTnHex-2-PyP and MnTnOct-
in oxidized and reduced state (superscripts indicating Mn
oxidation state are removed for simplicity from the formulae
in the text, other than in the legend of Fig. 2; meso indicates
substituents at the 5, 10, 15 and 20 positions of the porphyrin ring;
b indicates substituents at pyrrolic carbons; Mn(III) tetrakis(N-(1-
(2-(2-(2-methoxyethoxy)ethoxy)ethyl)pyridinium-2-yl)porphyrin,
5
+
2
-PyP ) exhibited significantly lower neurotoxicity when given
51
intracerebroventricularly.
MnTTEG-2-PyP was more protective to SOD-deficient E. coli
than MnTDTEG-2-ImP , presumably because it is less bulky
and able to enter the cell more easily. The efficacy is similar to
5
+
5
+
5
+
5+
ꢀ
its ethyl analogue, MnTE-2-PyP despite the significantly larger
MnTTEG-2-PyP (TEGPy) and Mn(III) tetrakis(N,N -di(1-(2-
size (Fig. 3). So, the 2.2-fold higher kcat (Table 2) of MnTTEG-2-
PyP , at least in part, balances out its higher molecular weight.
(2-(2-methoxyethoxy)ethoxy)ethyl)imidazolium-2-yl)porphyrin,
MnTDTEG-2-ImP (TEGIm); because 1-(2-(2(-2-methoxyethoxy)-
5
+
5+
Also, despite the larger size, at 3 lM and 10 lM, the MnTDTEG-
ethoxy)ethyl is derived from tri(ethyleneglycol) we often described
the porphyrin modification as being tri(ethyleneglycol)ated and
5
+
2
-ImP is even better in protecting E. coli than the much smaller
5
+
III
5+
ethyl analogue MnTDE-2-ImP (Fig. 3). This may again be partly
with TEG abbreviation; Mn T(alkyl)-2-PyP , Mn(III) meso
tetrakis(N-alkylpyridinium-2-yl)porphyrin; alkyl being methyl,
ethyl (EPy) (AEOL-10113), n-butyl, n-hexyl and n-octyl;
5
+
attributed to the much higher kcat of MnTDTEG-2-ImP . No
toxicity to the wild type E. coli was observed at the 30 lM
concentrations studied (Fig. 3D). Whereas no growth of wild
type E. coli was observed with 30 lM MnTnHex-2-PyP and
MnTnOct-2-PyP .
III
5+
ꢀ
Mn TDE-2-ImP (DEIm), Mn(III) meso tetrakis[N,N -diethyl-
5
+
III
3+
imidazolium-2-yl]porphyrin (AEOL-10150); Mn BM-2-PyP ,
5
+ 25
Mn(III) 5,10-bis(2-pyridyl)-15,20-bis(N-methylpyridinium-2-yl)-
12
•−
III
4+
Ferrer-Sueta et al. have shown that O
2
elimination by
porphyrin; Mn TrM-2-PyP , manganese(III) 5-mono-(2-pyridyl)-
−
•−
•
III
5+
porphyrins parallels their ONOO and CO
ity. Nitrosylation of Mn(III) porphyrins by NO involves their
reduction, and would thus parallel their SOD activity as well.
3
scavenging activ-
10,15,20-tris(N-methylpyridinium-2-yl)porphyrin; Mn T(TMA)P ,
manganese(III) meso tetrakis(N,N,N-trimethylanilinium-4-yl)-
8
III
5+
porphyrin; Mn T(TFTMA)P , manganese(III) meso tetrakis-
(2,3,5,6-tetrafluoro-N,N,N -trimethylanilinium-4-yl)porphyrin;
Thus we can safely assume that these porphyrins will be efficient
in attenuating levels of diverse ROS and RNS.
Generally, the drugs are modified with poly(ethyleneglycol)
III
3−
Mn TCPP , manganese meso tetrakis(4-carboxylatophenyl)-
porphyrin; Mn TSPP , manganese(III) meso tetrakis(4-
sulfonatophenyl)porphyrin; Mn T(2,6-Cl
III
3−
III
3−
(
PEG) in order to increase their blood-circulating half-life
2
-3-SO –P)P , manga-
3
26–28
and/or hydrophilicity.
Much to our surprise, and luckily so
nese(III) meso tetrakis(2,6-dichloro-3-sulfonatophenyl)porphyrin;
III
3−
(
as our compounds are already hydrophilic enough), no gain
Mn T(2,6-F
2
-3-SO
3
–P)P , manganese(III) meso tetrakis(2,6-
5
+
III
5+
in hydrophilicity was observed (Table 2). MnTE-2-PyP and
MnTDE-2-ImP thus far used in vivo had reportedly half-lives
not greater than 2 h.
difluoro-3-sulfonatophenyl)porphyrin; Mn Cl1–4TE-2-PyP , Mn(II)
5
+
b-mono-, di-, tri-, and tetrachloro meso tetrakis(N-ethylpyridi-
2,23
II
4+
It is likely that PEG-ylation would
nium-2-yl)porphyrin; Mn Cl
5
TE-2-PyP , Mn(II) b-pentachloro
II
significantly increase the half-life of these compounds, while
preventing unfavorable interactions with biological molecules.
Further studies will show whether the increase in blood-
circulating time, along with high antioxidant capacity and ability
to protect SOD-deficient E. coli, will significantly ameliorate
oxidative stress injuries and establish these porphyrins as more
meso tetrakis(N-ethylpyridinium-2-yl)porphyrin Mn Br
PyP , Mn(II) b-octabromo meso tetrakis(N-methylpyridinium-
4-yl)porphyrin; Mn TMOE-2-PyP , Mn(III) meso tetrakis[N-
(2-methoxyethyl)pyridinium-2-yl]porphyrin; Mn TM,MOE-
2-ImP , Mn(III) meso tetrakis[N-methyl-N -(2-methoxyethyl)-
imidazolium-2-yl]porphyrin; Mn TDMOE-2-ImP , Mn(III)
meso tetrakis[N,N -di(2-methoxyethyl)imidazolium-2-yl]porphyrin;
DMF, N,N -dimethylformamide, AN, acetonitrile; 4-toluenesul-
8
TM-4-
4
+
III
5+
III
5
+
ꢀ
III
5+
5
+
ꢀ
prospective drugs than their ethyl analogues, MnTE-2-PyP and
5
+
ꢀ
MnTDE-2-ImP that have already been successfully explored
15–20,22,23,44–46
in vivo.
fonic acid 1-(2-(2-(2-methoxyethoxy)ethoxy)ethyl ester, TEGTs;
NHE, normal hydrogen electrode; SOD, superoxide dismutase;
SOD-deficient, JI132 E. coli; SOD-proficient wild type, AB1157
E. coli; M9CA, M9 casamino acids medium.
Conclusions
5
+
The two new stable Mn(III) porphyrins, MnTTEG-2-PyP and
Acknowledgements
5
+
MnTDTEG-2-ImP , were synthesized. They possess the highest
•
−
ability to disproportionate O
2
among meso-substituted por-
IBH is grateful for the financial support to DOD CDMRP grant #
BC024326, Institute for the Study of Aging, and to Aeolus/Incara
Pharmaceuticals, RTP, NC. IBH and ZV acknowledge support
from to NIH U19 AI67798-01, and IF and AOM acknowledge
support of the grant from National Institute on Diabetes, Digestive
and Kidney Diseases of the NIH (RO1 DK 59868). MWD
acknowledges support from NIH CA40355 and the Duke SPORE
for breast cancer.
phyrins, the latter being only a few-fold less potent than the
SOD enzyme itself. Despite their significantly larger size, these
porphyrins are able to protect SOD-deficient E. coli at levels similar
to their ethyl analogues, MnTE-2-PyP and MnTDE-2-ImP ,
respectively. That being said, it is likely that, with increased blood-
circulating half-life, these compounds may be more effective in
ameliorating oxidative stress injuries than ethyl analogues that are
5
+
5+
This journal is © The Royal Society of Chemistry 2006
Dalton Trans., 2006, 617–624 | 623

View Article Online
References
24 I. Batini c´ -Haberle, I. Spasojevi c´ , R. D. Stevens, P. Hambright and I.
Fridovich, I., J. Chem. Soc., Dalton Trans., 2002, 2689.
25 A. Okado-Matsumoto, I. Batini c´ -Haberle and I. Fridovich, Free
Radical Biol. Med., 2004, 37, 401.
26 O. C. Farokhzad, S. Jon, A. Khademhosseini, T. N. Tran, D. A. Lavan
and R. Langer, Cancer Res., 2004, 64, 7668.
27 G. Kaul and M. J. Amiji, Drug Target., 2004, 12, 585.
28 K. Hyoudou, M. Nishikawa, Y. Umeyama, Y. Kobayashi, F. Yamashita
and M. Hashida, Clin. Cancer Res., 2004, 10, 7685.
29 W. R. Waud, F. O. Brady, R. D. Wiley and K. V. Rajagopalan, Arch.
Biochem. Biophys., 1975, 19, 695.
30 L. R. Milgrom, P. J. F. Dempsey and G. Yahiouglu, Tetrahedron, 1996,
52, 9877.
31 P. G. Cooksey, E. N. Gate, A. Gescher, J. A. Hickman, S. P. Langdon
and M. F. G. Stevens, Biochem. Pharmacol., 1983, 32, 3037.
32 I. M. Kolthof and W. J. Tomsicek, J. Phys. Chem., 1935, 39, 945.
33 I. Spasojevi c´ , I. Batini c´ -Haberle, R. D. Stevens, P. Hambright, A. N.
Thorpe, J. Grodkowski, P. Neta and I. Fridovich, Inorg. Chem., 2001,
40, 726.
34 J. M. McCord and I. Fridovich, J. Biol. Chem., 1969, 244, 6049.
35 K. M. Smith, Mass Spectrometry of Porphyrins and Metalloporphyrins
in Porphyrins and Metalloporphyrins, ed. K. M. Smith, Elsevier
Scientific Publishing Company, Amsterdam, 1975, p. 388.
36 M. A. Al-Maghrebi, I. Fridovich and L. Benov, Arch. Biochem.
Biophys., 2002, 402, 104.
1
I. Batini c´ -Haberle, L. Benov, I. Spasojevi c´ , P. Hambright, A. L.
Crumbliss and I. Fridovich, Inorg. Chem., 1999, 38, 4011.
2
I. Spasojevi c´ and I. Batini c´ -Haberle, Inorg. Chim. Acta, 2001, 317,
2
30.
3
4
I. Batini c´ -Haberle, Methods Enzymol., 2001, 349, 223.
R. Kachadourian, I. Batini c´ -Haberle and I. Fridovich, Inorg. Chem.,
1
999, 38, 391.
5
6
7
8
9
G. Ferrer-Sueta, I. Batini c´ -Haberle, I. Spasojevi c´ , I. Fridovich and R.
Radi, Chem. Res. Toxicol., 1999, 12, 442.
I. Batini c´ -Haberle, L. Benov, I. Spasojevi c´ and I. Fridovich, J. Biol.
Chem., 1998, 273, 24521.
I. Batini c´ -Haberle, S. Liochev, I. Spasojevi c´ and I. Fridovich, Arch.
Biochem. Biophys., 1997, 343, 225.
I. Spasojevi c´ , I. Batini c´ -Haberle and I. Fridovich, I., Nitric Oxide: Biol.
Chem., 2000, 4, 526.
I. Spasojevi c´ , I. Batini c´ -Haberle, I. J. S. Julio, Y. M. Idemori and I.
Fridovich, J. Biol. Chem., 278, 2003, 6831.
0 A. Trostchansky, G. Ferrer-Sueta, C. Batthyany, H. Botti, I. Batini c´ -
1
1
Haberle, R. Radi and H. Rubbo, Free Radical Biol. Med., 2003, 35,
1
293.
1 I. Batini c´ -Haberle, I. Spasojevi c´ , R. D. Stevens, P. Hambright, P.
Neta, A. Okado-Matsumoto and I. Fridovich, Dalton Trans., 2004,
1
696.
1
1
1
1
2 G. Ferrer-Sueta, D. Vitturi, I. Batini c´ -Haberle, I. Fridovich, S. Gold-
stein, G. Czapski and R. Radi, J. Biol. Chem., 2003, 278, 27432.
3 G. Ferrer-Sueta, C. Quijano, B. Alvarez and R. Radi, Methods
Enzymol., 2002, 349, 23.
4 I. Batini c´ -Haberle, I. Spasojevi c´ and I. Fridovich, Free Radical Biol.
Med., 2004, 37, 367.
5 B. J. Moeller, I. Batini c´ -Haberle, I. Spasojevi c´ , Z. N. Rabbani, M. S.
Anscher, Z. Vuja sˇ kovi c´ and M. W. Dewhirst, Int. J. Radiat. Oncol. Biol.
Phys., 2005, 63, 545.
37 F. S. Archibald and I. Fridovich, J. Bacteriol., 1981, 145, 442.
38 F. S. Archibald and I. Fridovich, J. Bacteriol., 1981, 146, 928.
39 R. A. Marcus, Annu. Rev. Phys. Chem., 1964, 15, 155.
40 R. B. Jordan, Reaction Mechanisms of Inorganic and Organometallic
Systems, Oxford University Press, New York, 2nd edn, 1998.
41 L. E. Bennet, Prog. Inorg. Chem., 1973, 18, 1.
42 P. M. Wood, J. Biochem., 1988, 253, 287.
43 R. Kachadourian, I. Batini c´ -Haberle and I. Fridovich, Free Radical
Biol. Med., 1998, 25(Suppl. 1), S17.
1
1
1
1
6 Y. Zhao, L. Chaiswing, J. M. Velez, I. Batini c´ -Haberle, N. H. Colburn,
T. D. Oberley and D. K. St. Clair, Cancer Res., 2005, 65, 3745.
7 Y. Zhao, L. Chaiswing, T. D. Oberley, I. Batini c´ -Haberle, W. St. Clair,
C. J. Epstein and D. St. Clair, Cancer Res., 2005, 65, 1401.
8 B. J. Moeller, Y. Cao, Y., C. Y. Li and M. W. Dewhirst, Cancer Cell,
44 J. H. Lee and J.-W. Park, Free Radical Biol. Med., 2004, 37, 272.
45 J. H. Lee, Y. M. Lee and J.-W. Park, Biochem. Biophys. Res. Commun.,
2005, 334, 298.
46 J. P. Crow, N. Y. Calingasan, J. Chen, J. L. Hill and M. F. Beal, Ann.
Neurol., 2005, 58, 258.
47 E. Michel, T. Nauser, B. Sutter, P. L. Bounds and W. H. Koppenol,
Arch. Biochem. Biophys., 2005, 439, 234.
48 R. M. Ellerby, D. E. Cabelli, J. A. Graden and J. S. Valentine, J. Am.
Chem. Soc., 1996, 118, 6556.
2
004, 5, 429.
9 J. D. Piganelli, S. C. Flores, C. Cruz, J. Koepp, R. Young, B. Bradley, R.
Kachadourian, R., I. Batini c´ -Haberle and K. Haskins, Diabetes, 2002,
5
1, 347.
2
2
0 L. Benov and I. Batini c´ -Haberle, Free Radical Res., 2005, 38, 81.
1 H. Tse, M. J. Milton and J. D. Piganelli, Free Radical Biol. Med., 2004,
49 C. K. Vance and A.-F. Miller, J. Am. Chem. Soc., 1998, 120, 461.
50 D. Klug-Roth, I. Fridovich and J. Rabani, J., J. Am. Chem. Soc., 1973,
95, 2786.
51 H. Sheng, M. Panni, I. Batini c´ -Haberle and D. S. Warner, unpublished
data.
52 K. Aston, N. Rath, A. Naik, U. Slomczynska, O. F. Schall and D. P.
Riley, Inorg. Chem., 2001, 40, 1779.
53 R. Kachadourian, C. A. Johnson, E. Min, I. Spasojevi c´ and B. J. Day,
Biochem. Pharmacol., 2004, 67, 77.
3
6, 233.
2
2 H. Sheng, J. Enghild, R. Bowler, M. Patel, C. L. Calvi, I. Batini c´ -
Haberle, B. J. Day, R. D. Pearlstein, J. D. Crapo and D. S. Warner, Free
Radical Biol. Med., 2002, 33, 947.
2
3 G. B. Mackensen, M. Patel, H. Sheng, C. L. Calvi, I. Batini c´ -Haberle,
B. J. Day, L. P. Liang, I. Fridovich, J. D. Crapo, R. D. Pearlstein and
D. S. Warner, J. Neurosci., 2001, 21, 4582.
6
24 | Dalton Trans., 2006, 617–624
This journal is © The Royal Society of Chemistry 2006