Differential coordination demands in Fe versus Mn water-soluble cationic metalloporphyrins translate into remarkably different aqueous redox chemistry and biology

Article abstract of DOI:10.1021/ic3012519

The different biological behavior of cationic Fe and Mn pyridylporphyrins in Escherichia coli and mouse studies prompted us to revisit and compare their chemistry. For that purpose, the series of ortho and meta isomers of Fe(III) meso-tetrakis-N-alkylpyridylporphyrins, alkyl being methyl to n-octyl, were synthesized and characterized by elemental analysis, UV/vis spectroscopy, mass spectrometry, lipophilicity, protonation equilibria of axial waters, metal-centered reduction potential, E_1/2 for M^IIIP/M ^IIP redox couple (M = Fe, Mn, P = porphyrin), k_cat for the catalysis of O₂^?- dismutation, stability toward peroxide-driven porphyrin oxidative degradation (produced in the catalysis of ascorbate oxidation by MP), ability to affect growth of SOD-deficient E. coli, and toxicity to mice. Electron-deficiency of the metal site is modulated by the porphyrin ligand, which renders Fe(III) porphyrins ≥5 orders of magnitude more acidic than the analogous Mn(III) porphyrins, as revealed by the pK _a1 of axially coordinated waters. The 5 log units difference in the acidity between the Mn and Fe sites in porphyrin translates into the predominance of tetracationic (OH)(H₂O)FeP complexes relative to pentacationic (H₂O)₂MnP species at pH ～7.8. This is additionally evidenced in large differences in the E_1/2 values of M^IIIP/M^IIP redox couples. The presence of hydroxo ligand labilizes trans-axial water which results in higher reactivity of Fe relative to Mn center. The differences in the catalysis of O₂^?- dismutation (log k_cat) between Fe and Mn porphyrins is modest, 2.5-5-fold, due to predominantly outer-sphere, with partial inner-sphere character of two reaction steps. However, the rate constant for the inner-sphere H₂O₂-based porphyrin oxidative degradation is 18-fold larger for (OH)(H₂O)FeP than for (H₂O)₂MnP. The in vivo consequences of the differences between the Fe and Mn porphyrins were best demonstrated in SOD-deficient E. coli growth. On the basis of fairly similar log k_cat(O₂^?-) values, a very similar effect on the growth of SOD-deficient E. coli was anticipated by both metalloporphyrins. Yet, while (H₂O)₂MnTE-2-PyP ⁵⁺ was fully efficacious at ≥20 μM, the Fe analogue (OH)(H ₂O)FeTE-2-PyP⁴⁺ supported SOD-deficient E. coli growth at as much as 200-fold lower doses in the range of 0.1-1 μM. Moreover the pattern of SOD-deficient E. coli growth was different with Mn and Fe porphyrins. Such results suggested a different mode of action of these metalloporphyrins. Further exploration demonstrated that (1) 0.1 μM (OH)(H₂O)FeTE-2- PyP⁴⁺ provided similar growth stimulation as the 0.1 μM Fe salt, while the 20 μM Mn salt provides no protection to E. coli; and (2) 1 μM Fe porphyrin is fully degraded by 12 h in E. coli cytosol and growth medium, while Mn porphyrin is not. Stimulation of the aerobic growth of SOD-deficient E. coli by the Fe porphyrin is therefore due to iron acquisition. Our data suggest that in vivo, redox-driven degradation of Fe porphyrins resulting in Fe release plays a major role in their biological action. Possibly, iron reconstitutes enzymes bearing [4Fe-4S] clusters as active sites. Under the same experimental conditions, (OH)(H₂O)FePs do not cause mouse arterial hypotension, whereas (H₂O)₂MnPs do, which greatly limits the application of Mn porphyrins in vivo.

Full text of DOI:10.1021/ic3012519

Article
pubs.acs.org/IC
Differential Coordination Demands in Fe versus Mn Water-Soluble
Cationic Metalloporphyrins Translate into Remarkably Different
Aqueous Redox Chemistry and Biology
Artak Tovmasyan,† Tin Weitner,† Huaxin Sheng,‡,∥ MiaoMiao Lu,‡,∥,§ Zrinka Rajic,† David S. Warner,‡,∥
Ivan Spasojevic,⊥ Julio S. Reboucas,# Ludmil Benov,*,○ and Ines Batinic-Haberle*,†
‡
⊥
∥
†Departments of Radiation Oncology, Anesthesiology, Medicine, and Multidisciplinary Neuroprotection Laboratories,
Duke University Medical Center, Durham, North Carolina 27710, United States
§Department of Anesthesiology, Second Affiliated Hospital, Zhengzhou University, Henan, China
#Departamento de Química, CCEN, Universidade Federal da Paraíba, Joao Pessoa, PB 58051-900, Brazil
̃
○Department of Biochemistry, Kuwait University, Faculty of Medicine, 13110 Safat, Kuwait
ABSTRACT: The different biological behavior of cationic Fe and Mn pyridylporphyrins in Escherichia coli and mouse studies
prompted us to revisit and compare their chemistry. For that purpose, the series of ortho and meta isomers of Fe(III) meso-
tetrakis-N-alkylpyridylporphyrins, alkyl being methyl to n-octyl, were synthesized and characterized by elemental analysis, UV/vis
spectroscopy, mass spectrometry, lipophilicity, protonation equilibria of axial waters, metal-centered reduction potential, E1/2 for
MIIIP/MIIP redox couple (M = Fe, Mn, P = porphyrin), kcat for the catalysis of O2•− dismutation, stability toward peroxide-driven
porphyrin oxidative degradation (produced in the catalysis of ascorbate oxidation by MP), ability to affect growth of SOD-
deficient E. coli, and toxicity to mice. Electron-deficiency of the metal site is modulated by the porphyrin ligand, which renders Fe(III)
porphyrins ≥5 orders of magnitude more acidic than the analogous Mn(III) porphyrins, as revealed by the pKa1 of axially coordinated
waters. The 5 log units difference in the acidity between the Mn and Fe sites in porphyrin translates into the predominance of tetracationic
(OH)(H2O)FeP complexes relative to pentacationic (H2O)2MnP species at pH ∼7.8. This is additionally evidenced in large differences in
the E1/2 values of MIIIP/MIIP redox couples. The presence of hydroxo ligand labilizes trans-axial water which results in higher reactivity of
Fe relative to Mn center. The differences in the catalysis of O2•− dismutation (log kcat) between Fe and Mn porphyrins is modest, 2.5−5-
fold, due to predominantly outer-sphere, with partial inner-sphere character of two reaction steps. However, the rate constant for the inner-
sphere H2O2-based porphyrin oxidative degradation is 18-fold larger for (OH)(H2O)FeP than for (H2O)2MnP. The in vivo consequences
of the differences between the Fe and Mn porphyrins were best demonstrated in SOD-deficient E. coli growth. On the basis of fairly
similar log kcat(O2•−) values, a very similar effect on the growth of SOD-deficient E. coli was anticipated by both metalloporphyrins. Yet,
while (H2O)2MnTE-2-PyP5+ was fully efficacious at ≥20 μM, the Fe analogue (OH)(H2O)FeTE-2-PyP4+ supported SOD-deficient
E. coli growth at as much as 200-fold lower doses in the range of 0.1−1 μM. Moreover the pattern of SOD-deficient E. coli growth was
different with Mn and Fe porphyrins. Such results suggested a different mode of action of these metalloporphyrins. Further exploration
demonstrated that (1) 0.1 μM (OH)(H2O)FeTE-2-PyP4+ provided similar growth stimulation as the 0.1 μM Fe salt, while the 20 μM Mn
salt provides no protection to E. coli; and (2) 1 μM Fe porphyrin is fully degraded by 12 h in E. coli cytosol and growth medium, while
Mn porphyrin is not. Stimulation of the aerobic growth of SOD-deficient E. coli by the Fe porphyrin is therefore due to iron acquisition.
Our data suggest that in vivo, redox-driven degradation of Fe porphyrins resulting in Fe release plays a major role in their biological action.
Possibly, iron reconstitutes enzymes bearing [4Fe-4S] clusters as active sites. Under the same experimental conditions, (OH)(H2O)FePs
do not cause mouse arterial hypotension, whereas (H2O)2MnPs do, which greatly limits the application of Mn porphyrins in vivo.
Received: June 12, 2012
Published: May 6, 2013
© 2013 American Chemical Society
5677
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
and meta Fe(III) N-alkylpyridylporphyrins (Figure 1), which
were used to probe the effect of electron-deficiency of Fe and
Mn sites upon their dismuting ability, kcat(O2•−), metal(III)/
metal(II) reduction potential, E1/2, compound lipophilicity, and
stability toward oxidative degradation. Finally we compared
side-by-side the ability of Fe and Mn complexes to protect
SOD-deficient E. coli against endogenously generated O2•−, and
evaluated their systemic toxicity to mice.
INTRODUCTION
■
While the general and biomimetic chemistry of the para isomer
FeTE-2-PyP5+ (the first metalloporphyrin-based SOD mimic
characterized) has been extensively explored, the data on the
redox and coordination chemistry of its meta and ortho ana-
logues and longer alkyl chain derivatives have been con-
siderably neglected. On the redox modulation front, in vivo
studies on Fe porphyrins as SOD mimics and/or peroxynitrite
decomposition catalysts have been limited to compounds from
our own laboratory, and to FP-15, INO-4885, and WW-85.1−8
We have explored a series of anionic and cationic Mn and Fe
porphyrins a decade ago and established the structure−activity
relationship between their metal-centered reduction potential
for (X)MIIIP/(X)MIIP redox couple (M = metal, X = H2O or OH−),
EXPERIMENTAL SECTION
■
General. Meso-tetrakis(2-N-pyridyl)porphyrin (H2T-2-PyP) and
meso-tetrakis(3-N-pyridyl)porphyrin (H2T-3-PyP) were purchased
from Frontier Scientific. Ethyl p-toluenesulfonate (98%) and methyl
p-toluenesulfonate were from Sigma-Aldrich. n-Butyl p-toluenesulfonate,
n-hexyl p-toluenesulfonate, n-octyl p-toluenesulfonate, and methyltri-n-
octylammonium chloride (>95%) were from TCI America. MnCl2 ×
4H2O (99.7%) was supplied by J. T. Baker, FeCl2 (98%) was from
Aldrich, and NH4PF6 (99.99%) was from GFS chemicals. Anhydrous
diethyl ether and acetone were from EMD chemicals, while dichloro-
methane, chloroform, acetonitrile, EDTA, and KNO3 were purchased
from Mallinckrodt. Anhydrous N,N-dimethylformamide (DMF) of 99.8%
purity (kept over 4-Å molecular sieves) and plastic-backed silica gel TLC
plates (Z122777-25EA) were from Sigma-Aldrich. Xanthine, equine
ferricytochrome c (Lot 7752), and (+)-sodium L-ascorbate (>98%) were
from Sigma, whereas xanthine oxidase was prepared by R. Wiley.9 FeSO4,
MnSO4, and sodium citrate was from Sigma Aldrich. To prepare Mn and
Fe citrates, the Na citrate was added to a 10 mM stock solution of FeSO4
or MnSO4. Fresh metal citrate solution was prepared for each experiment.
All chemicals were used as received without further purification.
Synthesis. The general synthetic procedure for meso-tetra-
(N-alkylpyridinium-2 or 3-yl)porphyrins and their iron complexes is
shown in Figure 2. Method A was used for the synthesis of all meta
and shorter-chain ortho analogues, while for longer-chain n-hexyl and
n-octyl ortho species, method B was applied.
6,8−11
•−
E
1/2, and ability to catalyze O2 dismutation, kcat(O2•−).
We have further shown that in addition to these thermody-
namic parameters, as it is with natural SODs, electrostatics
plays a major role in modulating the kinetic parameter kcat.12,13
Although both Fe and Mn cationic N-alkylpyridylporphyrins
are potent SOD mimics, as inferred by their kcat values, earlier
side-by-side in vivo studies at doses of ∼20 μM revealed that
only Mn porphyrins offered the highest protection to SOD-
deficient E. coli, while Fe porphyrins were toxic.8 Therefore, in
an attempt to avoid any possible toxicity arising from Fe-driven
Fenton chemistry, we have since explored predominantly Mn
porphyrins.6,10,11,14−16
Recent dose-dependence studies on SOD-deficient E. coli,
however, suggested that Fe porphyrins may be efficacious at
concentrations orders of magnitude lower than Mn porphyrins.
This prompted us to undertake a systematic comparison of Fe
versus Mn porphyrin-based compounds. Specifically for this
work, we synthesized and fully characterized a series of ortho
Figure 1. Structures of ortho and meta isomeric Fe(III) and Mn(III) meso-tetrakis N-alkylpyridylporphyrins. Axial ligands indicated on the schemes
are not related to the complexes isolated in solid state but to the species present at physiological pH in aqueous systems (see Protonation Equilibria
section).
5678
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
Figure 2. The synthesis of metalloporphyrins. In method A, the ligand was first alkylated and then metalated. The procedure was adapted from that
previously described for the methyl analogue.8 In method B, the ligand was first metalated and then alkylated (described in brief under Synthesis).
After alkylation was completed, the rest of the workup procedure was analogous to that described previously.8
Method A. Synthesis of meta Fe Porphyrins. Alkylation. The
synthesis, isolation, purification, and characterization of H2TM-3-
PyPCl4, H2TE-3-PyPCl4, H2TnBu-3-PyPCl4 and H2TnHex-3-PyPCl4
were performed as described earlier.6,8,17 H2TnOct-3-PyPCl4: H2T-3-
PyP (300 mg; 0.485 mmol) was dissolved in 25 mL of DMF at
112 °C, preheated for 10 min, and to the resulting solution 25 g (0.088
mol) of n-octyl p-toluenesulfonate was added. The course of N-octylation
was followed by thin-layer chromatography (TLC) on silica gel plates
using acetonitrile/KNO3(sat)/water = 8:1:1 as the mobile phase. The
reaction was completed within 5 h. Porphyrin was precipitated from
the reaction mixture by diethyl ether, filtrated, and washed with diethyl
ether (5 × 50 mL). It was then dissolved in methanol and purified as
described previously for their shorter alkyl-chain analogues.31,37,43
Yield (calculated based on formulation given by elemental analysis):
650 mg (96%). Metalation: The pH of an aqueous solution of
H2Talkyl-3-PyPCl4 (∼2 mM) was adjusted to 2 (with 1 M HCl), and a
40-fold molar excess FeCl2 was added into the solution and stirred at
reflux. The course of metalation was followed on silica gel TLC plates
using acetonitrile/KNO3(sat)/water = 8:1:1 as a mobile phase (pH = 2,
adjusted by 1 M HCl). Additionally, the loss of metal-free porphyrin
fluorescence under UV light at ∼350 nm was determined. The metala-
tion proceeded relatively fast with meta series of porphyrins: after 5 h
the reaction was completed. The solution was filtered first through
coarse then through fine filter papers. The Fe porphyrin was pre-
cipitated as the PF6 salt with saturated aqueous solution of NH4PF6.
−
The precipitate was thoroughly washed with diethyl ether. The dried
precipitate was then dissolved in acetone, filtered, and precipitated as
the chloride salt with saturated acetone solution of methyl-tri-n-
octylammonium chloride. The precipitate was washed with acetone
and dissolved in water. The whole precipitation procedure was re-
peated once again to ensure high purity of preparation. The meta Fe
porphyrins were isolated in quantitative yields.
Synthesis of ortho Fe Porphyrins. Alkylation. The synthesis,
isolation, purification, and characterization of metal-free H2TM-2-
PyPCl4, H2TE-2-PyPCl4, and H2TnBu-2-PyPCl4 were earlier
described.6,8 Metalation: The metalation and purification of ortho iso-
meric porphyrins were similar to their meta analogues except that
FeCl2 was added in a 100-fold molar excess. The metalation was
slower due to the steric hindrance of the vertically stuck alkyl chains.
Method B. Fe(III) meso-tetrakis(2-pyridyl)porphyrin, FeT-2-PyP+
(Figure 2B). The 150 mg of H2T-2-PyP was dissolved in ∼20 mL of
chloroform and 3 mL of methanol, and preheated for about 10 min to
fully dissolve the compound. Then, the 150 mg of FeCl2 was dissolved in
10 mL of warm ethanol and added to the porphyrin mixture. The course
of metalation was followed on silica gel TLC plates using 1:5 methanol/
chloroform mixture as a mobile phase. The reaction was completed in
5679
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
an hour (no fluorescence observed on the TLC plate). The solvent
was then evaporated to ∼5 mL volume. Ammonium hydroxide was
subsequently added to the reaction mixture to precipitate the
porphyrin. The precipitate was then filtered through fritted disc and
washed with water until neutral pH. The precipitate was washed
to dryness with diethyl ether and dissolved on fritted disc with
dichloromethane. A purple solid was obtained after solvent evapora-
tion. Yield was 165 mg. FeTnHex-2-PyPCl5: 50 mg (0.071 mmol) of
Fe(III) meso-tetrakis(2-N-pyridyl)porphyrin was dissolved in 3 mL of
DMF at 112 °C and preheated for 10 min under nitrogen. To the
resulting solution, 4.2 mL (0.018 mol) of n-hexyl p-toluenesulfonate
was added. The course of N-hexylation was followed by thin-layer
with 0.5 nm resolution using a 1 cm quartz cuvette. The data at pH 2
relate to diaqua Fe porphyrins (Table 1).
Table 1. The Spectral Properties of H2TnOct-3-PyPCl4 and
Metal Complexes
a
Porphyrin
λmax nm (log ε)
b
FeTM-2-PyPCl5
FeTM-3-PyPCl5
FeTE-2-PyPCl5
FeTE-3-PyPCl5
FeTnBu-2-PyPCl5
FeTnBu-3-PyPCl5
395.0 (5.11), 500.0 (4.15), 610.0 (3.91)
b
397.8 (5.01)
b
396.8 (5.16)
262.2 (4.47), 333.3 (4.52), 396.4 (5.11), 509.2 (4.05)
263.0 (4.44), 395.6 (5.03), 500.7 (4.03), 616.0 (3.90)
262.5 (4.48), 333.5 (4.53), 396.8 (5.10), 506.6 (4.07)
chromatography on silica gel TLC plates using acetonitrile/KNO3(sat)
/
water = 8:1:1 as a mobile phase (pH = 2, adjusted by 1 M HCl). The
reaction was completed after a day of stirring at 112 °C. The isolation
and purification of FeTnHex-2-PyPCl5 were done as described for
meta Fe porphyrins via method A. FeTnOct-2-PyPCl5: The synthesis
and purification were similar to those described for FeTnHex-2-
PyPCl5. The yields of FeTnHex-2-PyPCl5 and FeTnOct-2-PyPCl5
were quantitative.
FeTnHex-2-PyPCl5 262.6 (4.47), 395.7 (5.06), 501.0 (4.08), 615.1 (3.95)
FeTnHex-3-PyPCl5 263.3 (4.47), 334.0 (4.51), 397.4 (5.06), 506.8 (4.05)
FeTnOct-2-PyPCl5 263.1 (4.49), 415.0 (5.03), 500.2 (4.11), 616.1 (3.95)
FeTnOct-3-PyPCl5 263.0 (4.47), 333.7 (4.47), 397.5 (5.08), 507.7 (4.06),
577.6 (3.88)
H2TnOct-3-PyPCl4 263.1 (4.41), 417.8 (5.55), 514.2 (4.29), 581.6 (3.85),
c
Synthesis of Mn Porphyrins. MnTM-2-PyPCl5, MnTE-2-
PyPCl5, MnTnPr-2-PyPCl5, MnTnBu-2-PyPCl5, MnTnHex-2-
PyPCl5, MnTnHep-2-PyPCl5, MnTnOct-2-PyPCl5, MnTM-3-PyPCl5,
MnTE-3-PyPCl5, MnTnBu-3-PyPCl5, MnTnHex-3-PyPCl5: Synthesis
of these Mn porphyrins was performed as described elsewhere.6,8,17
MnTnOct-3-PyPCl5: 150 mg of H2TnOct-3-PyPCl4 (0.124 mmol)
was dissolved in 60 mL of water and the pH of the resulting solution
was adjusted to 11.5. A 20-fold excess MnCl2 (2.47 mmol, 0.49 g) was
added into the solution at 25 °C while stirring, resulting in a pH drop
to ∼8.0. The course of metalation was followed on silica gel TLC
plates using acetonitrile/KNO3(sat)/water = 8:1:1 as a mobile phase.
After 12 h of stirring, the reaction was worked-up by filtration and
636.1 (3.01)
MnTnOct-3-PyPCl5 215.2 (4.77), 261.6 (4.58), 373.9 (4.72), 395.0 (4.69),
460.0 (5.18), 501.6 (3.88), 557.8 (4.16), 675.1 (3.37),
c
765.6 (3.40)
a
Spectra were measured in 0.01 M HCl at room temperature, unless
noted otherwise. Thus, all metalloporphyrins have two H2O molecules as
axial ligands. Molar absorption coefficients (M−1 cm−1) were determined
within 5% errors considering the formulation given by elemental analysis.
b
λmax (nm) were determined with errors inside 0.5 nm. Data are taken
c
from ref 6. Spectra were taken in water at room temperature.
−
PF6 /Cl− sequential precipitation procedures: the isolation and
purification of the MnTnOct-3-PyPCl5 were done as described for the
meta Fe porphyrins (see above). The isolated yield was quantitative.
Elemental Analysis. Elemental analyses of H2TnOct-3-PyPCl4
and metal complexes were performed by Atlantic MicroLab (Norcross,
GA, USA). All analyses were done in duplicates. Formulations account-
ing for 9−13 hydration waters are consistent with related MnTE-2-
PyPCl5·11H2O sample isolated under similar conditions.18
FeTE-3-PyPCl5·11H2O: Anal. Calcd for C48H66Cl5FeN8O11:
H, 5.71; C, 49.52; N, 9.63; Cl, 15.23%.
Found: H, 5.51; C, 49.35; N, 9.56; Cl, 14.87%.
FeTnBut-2-PyPCl5·11H2O: Anal. Calcd for C56H82Cl5FeN8O11:
H, 6.48; C, 52.69; N, 8.78; Cl, 13.89%.
Found: H, 6.43; C, 52.72; N, 8.85; Cl, 13.56%.
FeTnBut-3-PyPCl5·10H2O: Anal. Calcd for C56H80Cl5FeN8O10:
H, 6.41; C, 53.45; N, 8.9; Cl, 14.09%.
Found: H, 6.38; C, 53.22; N, 9.01; Cl, 13.8%.
FeTnHex-2-PyPCl5·11H2O: Anal. Calcd for C64H116Cl5FeN8O11:
H, 7.11; C, 55.36; N, 8.07; Cl, 12.77%.
Found: H, 6.81; C, 55.27; N, 8.22; Cl, 12.37%.
FeTnHex-3-PyPCl5·9H2O: Anal. Calcd for C64H94Cl5FeN8O9:
H, 7.00; C, 56.83; N, 8.28; Cl, 13.12%.
Found: H, 6.79; C, 56.67; N, 8.35; Cl, 12.76%.
FeTnOct-2-PyPCl5·13H2O: Anal. Calcd for C72H118Cl5FeN8O13:
H, 7.74; C, 56.27; N, 7.29; Cl, 11.53%.
Found: H, 7.01; C, 55.81; N, 7.32; Cl, 11.32%.
FeTnOct-3-PyPCl5·10H2O: Anal. Calcd for C72H112Cl5FeN8O10:
H, 7.61; C, 58.32; N, 7.56; Cl, 11.95%.
Found: H, 7.40; C, 58.00; N, 7.54; Cl, 11.58%.
H2TnOct-3-PyPCl4 10H2O: Anal. Calcd for C72H114Cl4N8O10:
H, 8.25; C, 62.06; N, 8.04; Cl, 10.18%. Found: H, 7.96; C,
62.05; N, 8.04; Cl, 10.4%.
MnTnOct-3-PyPCl5·10H2O: Anal. Calcd for C72H112Cl5MnN8O10:
H, 7.62; C, 58.36; N, 7.56; Cl, 11.96%. Found: H, 7.43; C, 57.99;
N, 7.55; Cl, 11.87%.
UV/vis Spectroscopy. UV/vis spectra were recorded in 0.01 M HCl
at room temperature on a UV-2550 PC Shimadzu spectrophotometer
5680
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
b
Table 2. Electrospray Ionization Mass Spectrometry (ESI-MS) Data for Fe Porphyrins
m/z [found (calculated)]
a
species
FeTE-3-PyP5+ FeTnBu-2-PyP5+ FeTnBu-3-PyP5+ FeTnHex-2-PyP5+ FeTnHex-3-PyP5+ FeTnOct-2-PyP5+ FeTnOct-3-PyP5+
[FeP5+ + HFBA−]4+/4
[FeP5+ + 2HFBA−]3+/3
[FeP5+ + 3HFBA−]2+/2
250.4 (250.3)
404.8 (404.8)
713.5 (713.6)
278.5 (278.4)
442.3 (442.1)
769.5 (769.7)
278.4 (278.4)
442.1 (442.1)
769.5 (769.7)
306.5 (306.4)
497.5 (497.5)
825.6 (825.7)
306.6 (306.4)
479.5 (479.5)
825.5 (825.7)
334.7 (334.4)
517 (516.9)
881.5 (881.8)
334.5 (334.4)
516.8 (516.9)
881.5 (881.8)
a
∼1 μM solution of porphyrins and metalloporphyrins in 1:1 v/v acetonitrile/H2O (containing 0.01% v/v heptafluorobutyric acid (HFBA)) mixture,
b
20 V cone voltage. In the presence of heptafluorobutyric acid, in the solvent mixture no monohydroxo (OH)(H2O)FePs species was detected.
calculate the pKa1 values (Table 4) for other FeIIIP analogues described
in this work:
Table 3. Electrospray Ionization Mass Spectrometry
a
(ESI-MS) Data for H2TnOct-3-PyP4+ and MnTnOct-3-PyP5+
pKa = 7.8 − ΔE/0.059
(1)
m/z [found (calculated)]
species
H2TnOct-3-PyP4+ MnTnOct-3-PyP5+
where ΔE = E1/2 (pH = 2) − E1/2 (pH = 7.8).
[Pn+ + HFBA−](n−1)+/(n − 1)
[Pn+ + 2HFBA−](n−2)+/(n − 2)
[Pn+ + 3HFBA−](n−3)+/(n − 3)
Pn+/4
428.1 (427.9)
748.3 (748.4)
334.3 (334.2)
516.7 (516.5)
881.0 (881.3)
Stability toward Oxidative Degradation. The stability of
metalloporphyrins toward ascorbate-mediated oxidative degradation
with regard to alkyl chain length (ethyl vs hexyl), its position on the
pyridyl ring (ortho vs meta), and the type of metal center (Fe vs Mn)
was explored. Metalloporphyrins catalyze the oxidation of ascorbate
and thiols.20,32 Herein we explored their reactivity toward ascorbate
(Figure 3, process I). The subsequent peroxide formation degrades
porphyrin (Figure 3, process II).20,33,34 The stability of porphyrins
toward H2O2 was studied also, and data are in agreement with those
previously reported.8 The conditions were 6 μM metalloporphyrin,
0.42 mM sodium ascorbate at pH 7.8 maintained with 0.05 M Tris
buffer.
Lipophilicity. We have shown that both the TLC retention factor,
Rf, and the partition between n-octanol and water, log POW, are equally
valid parameters in assessing Mn porphyrin lipophilicity.6,17 Herein we
assessed lipophilicity via Rf, in acetonitrile/KNO3(sat)/water = 8:1:1 as
previously described.17 The Rf value for Fe and Mn porphyrins was
determined at pH 2, adjusted with the addition of 1 M HCl solution to
the chromatographic solvent system. At this pH, the predominant
species of either Fe or Mn porphyrins is diaqua, which is consistent
with the same profile of the distribution of atropoisomers on the TLC
plate both for Fe and Mn porphyrins. Because of the different protona-
tion equilibria of Fe vs Mn porphyrins, the partition between water
and n-octanol, log POW, was not assessed, as that would not allow the
comparison of the same FeP and MnP species. The data are sum-
marized in Table 5.
267.9 (267.7)
356.6 (356.6)
[P4+ − H+]3+/3
a
∼1 μM solution of porphyrins and metalloporphyrins in 1: 1 v/v
acetonitrile/H2O (containing 0.01% v/v heptafluorobutyric acid
(HFBA)) mixture, 20 V cone voltage.
Protonation Equilibria. The pKa values of the axial waters for MnPs
have been reported previously.25,26 The existence of diaqua (i.e., fully
protonated) species of either ortho, meta, or para N-methylpyridyl
Fe(III) porphyrins at pH = 2 has been documented in the litera-
ture,8,24,25,27−31 as determined by spectroscopic and electro-
chemical methods. The relationships between pKa values for Fe +3
and Fe +2 oxidation states of the metal sites of these isomers have
been documented as well. Kobayashi et al. reported values of pKa1
=
5.3−5.9 and pKa2 = 10.1−12.3 for (H2O)2FeIIITM-2(3 and 4)-PyP5+
and pK′a1 = 10.8−12.2 for (H2O)FeIITM-2(3 and 4)-PyP4+.29,30 The
data indicate significantly decreased acidity of FeIIPs compared
to FeIIIPs, due to a large decrease in electron-deficiency upon FeIIIP
reduction.
Consequently, in the region pH = 2−8, any shift in the redox poten-
tial depends solely on the deprotonation of FeIIIP, assuming that the
differences in the pyridyl substituents should not affect the established
relationships significantly (i.e., Ka1 ≫ Ka2 and Ka1 ≫ K′a1), already
demonstrated for the corresponding MnP isomers.25,26 Specifically, the
lengthening of the alkyl chain does not affect axial protonation equi-
libria: the pKa1 and pKa2 for Mn(III) methyl to n-octyl ortho analogues
differ insignificantly, by only 0.5 and 0.3 units, respectively.25
Therefore, a simple expression given in eq 1 could be applied to
SOD-like Activity of Metalloporphyrins. SOD activity was
examined using the cytochrome c (cyt c) assay.20,22,23 The experiments
were conducted at room temperature in 0.05 M potassium phosphate
buffer, pH 7.8, and 0.1 mM EDTA, as reported.20,22,23 Data are
summarized in Table 4.
Superoxide-Specific Biological Model - Aerobic Growth of
SOD-Deficient E. coli. E. coli mutants lacking the two cytosolic
Table 4. The Reduction Potentials E1/2 for MIII/MII Redox Couple (M = Fe, Mn), Deprotonation Constant, pKa1 for the Axial
•−
Water of (H2O)2FePs, and SOD Activity, Given As log kcat(O2
)
(H2O)2FeP5+
(OH)(H2O)FeP4+
(H2O)2MnP5+
a
b
a
ac
,
c
•−
•−
compound
ortho isomers
E1/2 (pH = 2)
pKa1
E1/2 (pH = 7.8)
log kcat(O2
)
E1/2 (pH = 7.8)
log kcat(O2 )
methyl
ethyl
0.381
0.369
0.348
0.370
0.406
0.232
0.229
0.232
0.230
0.236
5.02
5.12
5.95
5.75
5.34
5.66
5.68
5.63
5.77
5.95
0.217
0.211
0.239
0.249
0.261
0.106
0.104
0.104
0.110
0.127
7.89
8.05
7.82
7.53
7.09
6.99
6.98
6.99
6.86
6.93
0.220
0.228
0.254
0.314
0.367
0.052
0.054
0.064
0.066
0.074
7.79
7.76
7.25
7.48
7.71
6.61
6.65
6.69
6.64
6.53
n-butyl
n-hexyl
n-octyl
methyl
ethyl
meta isomers
n-butyl
n-hexyl
n-octyl
a
b
c
Redox potentials are given in volts vs NHE. Deprotonation constants are estimated according to eq 1. Redox potentials and rate constants
(in M−1 s−1) are from refs 6 and 11 except for (H2O)2MnTnOct-3-PyP5+, this work.
5681
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
Figure 3. The electrochemistry of Fe and Mn porphyrins as exemplified with ethyl analogues. The cyclic voltammetry of Fe porphyrins (0.2 mM)
was performed at pH = 2 and at pH = 7.8 where different species exist, (H2O)2FeTE-2(or 3)-PyP5+ and (H2O)(OH)FeTE-2(or 3)-PyP4+,
respectively. The increase of pH from 2 to 7.8, associated with the change in a type of species present, resulted in a shift of E1/2 from +369 to
+211 mV (left panel). Under both pH conditions the same diaqua Mn porphyrin ((H2O)2MnTE-2-PyP5+) exists.8 The shift of E1/2 from +369 with
(H2O)2FeTE-2-PyP5+ to +228 with (H2O)2MnTE-2-PyP5+ reflects much higher electron-deficiency of Fe than the Mn site. Similar to Mn
porphyrins, the E1/2 of ortho Fe porphyrins is higher than that of meta isomers (right panel) due to stronger electron-withdrawing properties of the
pyridyl nitrogens situated closer to the metal center. (H2O)2MnTE-2-PyP5+ was herein used as a standard. No μ-oxo dimers were observed under
given experimental conditions (0.2 mM concentration at pH = 2 and 7.8). No axial ligands are specified in the figure titles, as FeP species studied
bear different axial ligands.
No significant differences between the two sets of strains were
observed.
Table 5. Lipophilicity of Mn and Fe Porphyrin-Based SOD
Mimics Expressed in Terms of Rf Values Taken in
Acetonitrile/KNO3(sat)/Water = 8:1:1 Solvent System at pH 2
a
The effect of (OH)(H2O)FePs (0.01−25 μM) on the growth of the
SOD-deficient mutant in 5AA medium was compared to the effect of
optimal 20 μM Mn ethyl analogues ortho (H2O)2MnTE-2-PyP5+ and
meta (H2O)2MnTE-3-PyP5+. For comparison, the effect of different
Mn and Fe salts was studies. The corresponding anions had no effect
on cell growth: citrate, sulfate, and chloride salts of Fe and Mn had
identical effect.
At different time points, the cells were harvested, rapidly washed
with cold buffered saline, and disrupted by French pressing. Cytosolic
fractions were isolated as described earlier,41 and the content of
metalloporphyrins was determined by UV/vis spectroscopy. Aliquots
of the growth media were analyzed at same time points.
Mouse Study. All animal procedures were conducted in
accordance with IUCAC guidelines and approval from Duke
University Medical Center, as described in the Guide for Care and
Use of Laboratory Animals published by the National Institutes of
Health (publication #NIH 85-23, revised 1985). Animals were housed
on a 12 h light cycle (lights on at 6:00 a.m.) with standard laboratory
diets and water available ad libitum. 8−10 weeks old male C57BL/6J
mice were used for studies.
Both Fe- and Mn porphyrins were administered by intraperitoneal
(i.p.) injection. Maximal tolerable doses (MTDs) were considered
those which caused mild and transient toxicity manifested by quiet-
ness, lethargy, and reluctance to ambulate. Then, doses of 10 mg/kg
of (OH)(H2O)FeTE-2-PyP4+, 10 mg/kg of (H2O)2MnTE-2-PyP5+,
5 mg/kg (OH)(H2O)FeTnHex-2-PyP4+, and 1 mg/kg of (H2O)2-
MnTnHex-2-PyP5+ were selected for the toxicity study. The com-
pounds were administered i.p. twice per day for one week. Animal
behavior, weight and performance on rotarod were followed for one
week as described previously.23
Rf
metalloporphyrin
(H2O)2FeP5+
(H2O)2MnP5+
ortho isomers
meta isomers
methyl
0.131
0.186
0.443
0.597
0.640
0.160
0.246
0.554
0.646
0.674
0.125
0.156
0.394
0.600
0.644
0.125
0.188
0.538
0.631
0.656
ethyl
n-butyl
n-hexyl
n-octyl
methyl
ethyl
n-butyl
n-hexyl
n-octyl
a
The Rf values relate to the diaqua Fe and diaqua Mn porphyrin
species.
superoxide dismutases (FeSOD and MnSOD, sodA‑sodB‑) suffer
phenotypic deficiencies including poor aerobic growth35 and auxo-
trophies for branched-chain,36 aromatic,37 and sulfur-containing amino
acids.38 The failure of the sodA‑sodB‑ mutants to grow aerobically in the
absence of the listed amino acids can be relieved by compounds
scavenging superoxide. Therefore, aerobic growth of sodA‑sodB‑ E. coli
in a restricted, five amino-acid (5AA) medium can be used to assess
the capacity of SOD mimics to substitute for the natural enzymes. The
5AA medium consisted of M9 salts supplemented with filter-sterilized
L-leucine, L-threonine, L-proline, L-arginine, and L-histidine at a final
concentration of 0.5 mM each, 0.2% glucose and 3 mg/L of panto-
thenic acid and thiamine. M9 salts were prepared by autoclaving 0.6 g
of Na2HPO4, 0.3 g of KH2PO4, 0.05 g of NaCl, and 0.1 g of NH4Cl per
liter in distilled water. After cooling, separately autoclaved solutions of
MgSO4 and CaCl2 were added to a final concentration of 1.0 mM.
Growth of the SOD-proficient AB1157 (F-thr-1; leuB6; proA2; his-4;
thi-1; argE2; lacY1; galK2; rpsL; supE44; ara-14; xyl-15; mtl-1; tsx-33)
and the SOD-deficient (sodA−sodB−) JI132 (same as AB1157 plus
(sodA::mudPR13)25 (sodB-kan)1-Δ2) E. coli strains in 5AA medium
was done as already described.20 Both strains were obtained from J. A.
Imlay.39 The effect of the compounds was tested in parallel on parental
and SOD-deficient strains with different genetic background: GC4468 =
parental strain; SOD-deficient, QC1799 = GC4468 ΔsodA3, ΔsodB-kan40
(D. Touati, Institute Jacques Monod, CNRS, Universite Paris, France).
The effect of porphyrins on arterial blood pressure was also assessed
through right femoral artery in isoflurane anesthetized mice. After
15 min of baseline recording, 10 mg/kg porphyrin was slowly infused
in 5 min and arterial blood pressure was continuously monitored
for 30 min. (Procedure: Mice were anesthetized with 5% isoflurane in
30% oxygen balanced with nitrogen and intubated. Both lungs were
mechanically controlled by a rodent ventilator. Body temperature was
controlled at 37 °C. Both right femoral artery and vein were can-
nulated by PE-10 tubes for blood pressure measurement and por-
phyrin slow infusion, respectively.
5682
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
bond is reasonably longer (2.221 Å), which implies that the
axial water molecules in the Fe system are more tightly held as a
response to the higher electron-deficiency of the FeP moiety,
whose acidity is thus transferred to the coordinated water
molecules and deprotonation is favored.
RESULTS AND DISCUSSION
■
Redox Property of Metal Site and Protonation Equilibria
of Axial Waters of Metalloporphyrins. We have already
reported that the major difference between the speciation
of Fe and Mn cationic N-alkylpyridylporphyrins in aqueous
systems at physiological conditions (at pH 7.8) is the
presence of axial electron-rich hydroxo ligand with Fe- vs
electron-poor axial aqua ligand with Mn porphyrins.8 All ortho
and meta Fe porphyrins exhibited an acid−base behavior
governed by the two proton-transfer equilibria represented by
eqs 2 and 3, whose first proton dissociation constants (pKa1)
fall in the 5−6 range (Table 4) followed by the second proton
dissociation constant (pKa2) of ∼11. The pKa data for
(H2O)2FeTM-2(or 3 or 4)-PyP5+ are in agreement with
those previously reported.8,24,27,29−31 These pKa sets indicate
that all FePs have lost a proton from the coordinated axial
water to yield the hydroxo-aqua complex as the major species at
pH 7.8. Conversely, the acid−base equilibrium of the Mn
Another major difference in the biology of “free” Fe vs “free”
Mn is related to their reduction potentials. Because of a vastly
higher reduction potential of “free” MnIII/MnII (Eo = +1.51 V
vs NHE) than of “free” FeIII/FeII (Eo = +0.77 V vs NHE), Mn
•
cannot be easily oxidized with H2O2 to produce OH radical;
consequently Mn does not undergo, while Fe is capable of
performing “Fenton chemistry” (Fe2+ + H2O2 ⇌ Fe3+ + OH− +
·OH).47 When Mn or Fe binds to SOD protein, the reduction
•−
potentials became identical and are within limits of O2
oxidation similar to Mn or Fe binding to porphyrin ligand.
With water as axial ligands (at pH = 2),8 (H2O)2FeTE-2-
PyP5+ has much higher E1/2 (+369 mV vs NHE) relative to
(H2O)2MnTE-2-PyP5+ (+228 mV vs NHE). This is consistent
with a stronger electron-deficiency and in turn acidity of the
FeP moiety, which is thus more prone to being reduced than
the corresponding MnP system. The replacement of an axial
aqua with an axial hydroxo ligand in FeP at pH 7.8 shifts the
E1/2 negatively from +369 to +211 mV vs NHE. Therefore, at
pH 7.8 (H2O)2MnTE-2-PyP5+ and (OH)(H2O)FeTE-2-PyP4+
have nearly identical E1/2 for MIII/MII redox couple (+228 and
+211 mV vs NHE, respectively) and in turn are equally
efficacious SOD mimics (Figure 3, Table 4). The magnitude of
the electron donating effect of hydroxo vs aqua ligand is similar
to the effect caused by moving the N-alkyl chains from ortho to
meta positions at pyridyl rings (Table 4).
analogues described by the same eqs 2 and 3, with pKa1 and
8,25
pKa2 values for the (H2O)2MnTM-2(or 3)-PyP5+
and
26
(H2O)2MnTE-2(or 3)-PyP5+ in the region of ∼10.5−11.4
(ortho isomers) and ∼11.5−13.5 (meta isomers), indicate that
the Mn porphyrins exist as the diaqua species at pH 7.8.
[(H2O)2MIIIP]5+⇌ [(OH)(H2O)MIIIP]4+ + H+, Ka1
(M = FeIII or MnIII)
(2)
[(OH)(H2O)MIIIP]4+⇌ [(OH)2MIIIP]3+ + H+, Ka2
(M = FeIII or MnIII)
(3)
Redox Property and SOD-like Activity of Metal-
loporphyrins. Figures 4 and 5 depict the differences in the
relationships between the thermodynamic and kinetic param-
The acidity of the metalloporphyrin axial waters with respect
to free water (pKw ∼ 14) is increased by ∼8.5 and ∼3 log units
upon coordination to the FeP and MnP moieties, respectively.
Such a large difference between the pKa1 values shows that the
axially coordinated water in Fe porphyrins is about 5 log units
more acidic than in the corresponding Mn porphyrins. This is
in direct contrast with the acidity of simple “free ions”, hexaaqua
species [FeIII(H2O)6]3+ (pKa1 ∼ 2.2)42 and [MnIII(H2O)6]3+(pKa1
∼ 0.1),43 which indicates that the Mn3+ species is more acidic than
the Fe3+ complex by ∼2 log units. Thus, substitution of four
neutral H2O ligands in the coordination sphere of free ions
by a [N4]2− coordination system of the porphyrinato ligand
(ignoring the charge of the peripheral pyridinium moieties)
increases the electronic density around the metal center, which
translates into a decrease in the coordinated water acidity; this
effect is, however, much more pronounced in the MnP case,
where the ligand substitution is followed by an increase of
∼10 log Ka1 units, whereas the FeP series has pKa1 values only
∼3 orders of magnitude greater than those of [FeIII(H2O)6]3+.
On electrostatic grounds, these results suggest that the positive
charge on the “[Fe(H2O)2]+” and “[Mn(H2O)2]+” moieties in
the porphyrin series is considerably more delocalized with
MnP, which results in their lower acidity relative to their Fe
counterparts. Diaqua and hydroxo-aqua MnIIIPs and FeIIIPs are
all high spin species;6,44 therefore this difference in axial water
acidity is not related to a spin state change. Although a full
description of aqueous behavior of FeP vs. MnP series based on
the metalloporphyrin electronic structure is not yet available,
the structural single crystal X-ray data for diaqua complexes of
•−
eters for the O2 dismutation (eqs 4 and 5 for Fe porphyrins,
eqs 6 and 7 for Mn porphyrins) and the length of the alkyl
chains between Fe and Mn porphyrins at pH 7.8.
(OH)(H2O)FeIIIP4+ + O2•− ⇌ (OH)FeIIP3+
+ O2 + H2O
(4a)
(OH)FeIIP3+ + H+ ⇌ (H2O)FeIIP4+
(4b)
•−
(OH)(H2O)FeIIIP4+ + O2 + H+ ⇌ (H2O)FeIIP4+
+ O2 + H2O
(4)
(H2O)FeIIP4+ + O2 + 2H+ ⇌ (H2O)FeIIIP5+ + H2O2
•−
(5a)
(H2O)FeIIIP5+ + H2O ⇌ (OH)(H2O)FeIIIP4+ + H+
(5b)
•−
(H2O)FeIIP4+ + O2 + H+ + H2O ⇌
(OH)(H2O)FeIIIP4+ + H2O2
(5)
(H2O)2MnIIIP5+ + O2•− ⇌ (H2O)2MnIIP4+ + O2
(6a)
(H2O)2MnIIP4+ ⇌ (H2O)MnIIP4+ + H2O
(6b)
45
the para isomers (H2O)2FeTM-4-PyP5+ and (H2O)2MnTM-
(H2O)2MnIIIP5+ + O2•− ⇌ (H2O)MnIIP4+ + O2 + H2O
46
4-PyP5+ are consistent with the lower acidity of the MnPs.
While the Fe−Owater bond length is 2.086 Å, the Mn−Owater
(6)
5683
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
Figure 4. The relationships between each of log kcat (O2•−) (A and C) and E1/2 (V vs NHE) (B and D) and the length of the alkyl chains (number of
•−
carbon atoms in chains, nC) for ortho and meta Mn and Fe porphyrins at pH 7.8. Numerical values and the experimental conditions for kcat (O2
)
and E1/2 (V vs NHE) are given in Table 5. The effects shown in these plots translate into the E1/2 vs log kcat (O2•−) relationships shown in Figure 5.
The experimental points are connected to better visualize the trends − no mathematical model was applied.
Figure 5. Structure−activity relationships between log kcat(O2•−) and E/1/2 for MIII/MII (M = metal) describe the profound difference between Fe
and Mn porphyrins dominated by the differences in the thermodynamic properties and steric and electrostatic factors which all affect porphyrin
•−
solvation and approach of anionic superoxide to cationic porphyrin and thus log kcat for the catalysis of O2 dismutation. Numbers in the plots
correspond to the number of carbon atoms present in each of the N-alkyl chains. The experimental points are connected to better visualize the
trends  no mathematical model was applied.
•−
(H2O)MnIIP4+ + O2 + 2H+ ⇌ (H2O)MnIIIP5+
(OH)(H2O)FeP species relative to the pentacationic (H2O)2MnP
species.6,8,10−13,24 Yet, labilization of the axial water in the trans
position relative to OH− increases largely the reactivity of the
Fe site. Consequently, (OH)(H2O)FePs (with the exception of
excessively sterically hindered n-octyl ortho analogue) have
higher kcat(O2•−) than (H2O)2MnPs (Figure 4A,C, Table 4).
The interaction between metalloporphyrins and superoxide is
outer-sphere with partial inner-sphere character.12,49 Therefore,
the labilization of trans-axial water due to the presence of
axial hydroxo ligand has only a modest effect; the difference
in the kcat(O2•−) between Fe- and Mn porphyrins is in the
range of 2.5−5-fold (0.4−0.7 log units). Yet, with reactions
of inner-sphere character which involves ligand binding, such
as H2O2-based degradation, the difference between Fe- and Mn
porphyrins is more than an order of magnitude larger (see
under Stability toward Oxidative Degradation).
+ H2O2
(7a)
(7b)
(H2O)MnIIIP5+ + H2O ⇌ (H2O)2MnIIIP5+
•−
(H2O)MnIIP4+ + O2 + 2H+ + H2O ⇌
(H2O)2MnIIIP5+ + H2O2
(7)
As detailed under the Redox Property of Metal Site and
Protonation Equilibria, the differences are the consequence of
the vast difference in the electron-deficiency of the metal site,
which results in different species present in solution under
physiological conditions: (OH)(H2O)FePs vs (H2O)2MnP.8,24,48
The OH− axial ligand affects both thermodynamics and kinetics
•−
of O2 dismutation. It neutralizes the single charge on the Fe
site and in turn reduces favorable electrostatics with the tetracationic
5684
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
The axially ligated hydroxo ligand affects more the thermo-
dynamic properties of both ortho and meta (OH)(H2O)FePs
than the polarity/apolarity of alkyl chains; consequently E1/2
varies little between ortho methyl and ortho n-octyl (OH)-
(H2O)FeP species, while a significant increase in E1/2 of
∼160 mV was seen from ortho methyl to ortho n-octyl
(H2O)2MnPs (Figure 4B). Further, E1/2 varies much less
between ortho and meta (OH)(H2O)FePs (116−170 mV) than
ortho and meta (H2O)2MnPs (168−293 mV) (Figure 4B,D,
Table 4). Consequently meta (OH)(H2O)FePs with higher
E1/2 and kcat are more potent SOD mimics than meta
(H2O)2MnPs (Table 4, Figure 4C). In meta series of both
porphyrins though, the E1/2 does not change as dramatically
with lengthening of alkyl chains as it does with ortho analogues;
the same is true for log kcat (Figure 5). The reversed trend is
seen between E1/2 vs nC and log kcat(O2•−) vs nC relationships
(Figure 4C,D). The small increase in E1/2 of (OH)(H2O)FePs
from methyl to n-octyl could not have compensated for steric
hindrance; consequently, the log kcat of ortho (OH)(H2O)FePs
with shorter methyl and ethyl chains first increases as the trans
axial position is activated and more reactive to O2•−, but it
decreases then due to the steric hindrance of the longer alkyl
chains (Figure 4A). Minor effects are seen with meta isomers
(Figure 4C) as are further reflected in structure−activity
relationships (Figure 5).
Figure 6. Metalloporphyrin-catalyzed ascorbate (HA−) oxidation at
pH 7.8, leading to the production of O2 and H2O2 (process I) and
•−
subsequent metalloporphyrin oxidative degradation (process II).
Peroxide can also be produced in oxidation of metalloporphyrin and
•−
ascorbyl radical (A•−) with O2 (see ref 33; also eqs 5 and 7), A•−
•−
with O2 (producing O2•−) and self-dismutation of O2
.
Herein we have addressed the H2O2-based degradation of
ortho and meta Mn and Fe porphyrins, bearing short ethyl
and long n-hexyl chains. Metal center/ascorbate was a producer
of peroxide. The lower E1/2 of meta metalloporphyrins relative
to ortho analogues indicates their preference of higher +3
relative to lower +2 Mn oxidation state. Thus, once reduced
with ascorbate, meta Mn porphyrins readily reoxidize with
oxygen producing more H2O2 and are in turn more degradable than
ortho species. The k(H2O2) is 1.3 M−1 s−1 for (H2O)2MnTM(E)-2-
PyP5+ and 4.9 M−1 s−1 for (H2O)2MnTM(E)-3-PyP5+.8 The exten-
sive degradation of porphyrins coincides with spectrophotometric
evidence of complete consumption of ascorbate. Ascorbyl radical,
A•−, formed during reduction of Mn porphyrin, readily oxidizes
further to dehydroascorbate producing O2•−, which upon
dismutation gives rise to H2O2. The Fe porphyrins are much
more potent catalysts of ascorbate oxidation than analogous
Mn porphyrins due to the labilization of trans-axial water which
greatly increases their reactivity in reactions of inner-sphere
character (Figure 7). During catalysis, peroxide is formed and
its binding and subsequent porphyrin oxidation leads to porphyrin
degradation and Fe release. The k(H2O2) is 23 M−1 s−1 for
(OH)(H2O)FeTM(E)-2-PyP4+ and is ∼20-fold higher relative to
(H2O)2MnTM(E)-2-PyP5+.8
Lipophilicity. We have previously reported that meta Mn
porphyrins are more lipophilic than their ortho isomers and
showed here that the same is valid for Fe porphyrins (Figure 8).
When Fe and Mn porphyrins bear shorter methyl and ethyl
chains, the overall charge of the complex is largely exposed to
the solvent, making them highly hydrophilic. As the chains
lengthen, the impact of the metal site charge is surpassed by the
effect of the alkyl chains, and both Fe and Mn porphyrins
become similarly lipophilic (Table 5, Figure 8). It is worth
noting that the difference in lipophilicity between Fe and Mn
porphyrins among the shorter alkyl side-chains is more
pronounced in the meta than in the ortho isomers; this is
likely related to the higher exposure of the metal center in the
meta isomers when compared to their ortho analogues, which
allows a better differentiation and greater impact of the
“[(H2O)2Fe]+” and “[(H2O)2Mn]+” moieties on the overall
lipophilicity of the complexes. Under physiological conditions,
Fe porphyrins exist predominantly as tetracationic (OH)(H2O)-
FeP4+ species which makes them more lipophilic than the corre-
sponding Mn porphyrins which, having water as axial ligands,
bear an overall 5+ charge, (H2O)2MnP5+. The impact of the 1 unit
reduction from 5+ to 4+ in overall charge of the complexes on
their lipophilicity and biological efficiency has been demon-
strated in the MnIIIP5+ and MnIIP4+ systems.69
Differences in the effect of alkyl substituents upon thermo-
•−
dynamics and kinetics of O2 dismutation result in entirely
different structure activity relationships (SAR) between the log
kcat(O2•−) and E/1/2 for isomeric Fe and Mn porphyrins (Figure 5).
The large increases in E1/2 with the alkyl chain length in ortho
Mn porphyrins (Figure 4B) compensate for the sterically
unfavorable effects of longer alkyl-chain analogues on super-
oxide approach; after initial kcat drop from methyl to n-butyl
analogue, kcat increases from n-butyl to n-octyl (Figure 5, left).
With Fe porphyrins, there is essentially no major increase in
E1/2 as the alkyl chains lengthen (Figure 4B); consequently
steric effects predominate and determine the shape of SAR
(Figure 5, left). While the total change in log kcat is ∼1 log unit
with ortho porphyrins, with meta analogues, the total change in
kcat is only ∼0.2 log units within each of the methyl to n-octyl
series of Fe and Mn porphyrins (Figures 4A,C and 5); yet the
behavior of meta MnPs is reversed from the one seen with ortho
analogues as a consequence of interplay of thermodynamic and
kinetics factors. With meta Mn porphyrins, the E1/2 decreases
from n-butyl to n-octyl much more than with meta Fe por-
phyrins (Figure 4C), and this in turn dominates the SAR.
Stability toward Oxidative Degradation. Given high
millimolar intracellular levels of ascorbate and the significance
of H2O2 as a major in vivo signaling (nM) and cytotoxic species
(mM), we routinely address the interactions of metal-
loporphyrins with these species.21,34,50,51 Cationic Mn and Fe
N-substituted pyridylporphyrins readily couple with ascorbate
catalyzing its oxidation (Figure 6). The catalysis results in
H2O2 production and subsequent porphyrin degradation. The
therapeutic potential of such a system for anticancer therapy via
production of long-lasting, highly cytotoxic peroxide was
demonstrated.34,50 Metalloporphyrins have widely been ex-
plored in cyt P450 biomimetic models.52−67 The metal-
loporphyrin/ascorbate system was proposed by us as a
potential in situ cyt P450 mimicking system for hydroxylation
of cyclophosphamide to its active metabolite 4-hydroxycyclo-
phosphamide.33
5685
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
Effect of Fe Porphyrins on the Aerobic Growth of
SOD-Deficient E. coli. The aerobic growth of E. coli in the
presence of ortho (OH)(H2O)FeTE-2-PyP4+ and meta (OH)-
(H2O)FeTE-3-PyP4+ and our standard SOD mimics ortho
(H2O)2MnTE-2-PyP5+ is shown in Figure 9. The more
lipophilic long-alkyl chain (n-hexyl and n-octyl) Mn and Fe
porphyrins are much more toxic to E. coli than the hydrophilic
analogues. The fairly lipophilic n-butyl porphyrin, (OH)(H2O)-
FeTnBu-2-PyP4+, already starts exerting toxicity at 1 μM.
Ortho (H2O)2MnTE-2-PyP5+ is commonly used as a positive
control at 20 μM. At such concentration, it allowed SOD-
deficient E. coli to grow aerobically in restricted, 5 AA medium
at a rate comparable to that of the parental strain.41 Whereas at
concentrations >1 μM (OH)(H2O)FePs start to exert toxicity,
at very low concentrations they supported the growth of the
SOD-deficient E. coli, and at 0.1 and 1 μM it reached the
growth rate of the wild type strain. It is worth noting, however,
that the growth pattern of the SOD-deficient strain in medium
supplemented with (OH)(H2O)FeP was different from the
growth of wild type E. coli or the growth of the sodAsodB strain
in (H2O)2MnP-supplemented medium (Figure 9B). Growth
Figure 7. The stability of Fe vs Mn porphyrins toward ascorbate-
mediated oxidative degradation at pH 7.8 as shown in Figure 6.20,33,34,68
The conditions were 6 μM metalloporphyrin, 0.42 mM sodium ascorbate
at pH 7.8 maintained with 0.05 M Tris buffer.
Figure 8. Lipophilicity of Fe and Mn porphyrins as described by chromatographic retention factor, Rf. Data relate to pH 2 and thus to identical
diaqua species of both metalloporphyrins (see Table 5). Shorter-chain (H2O)2FePs are slightly more lipophilic than isomeric (H2O)2MnPs; as the
chains lengthen, their influence on the lipophilicity prevails over charge/ligation at the metal site.
Figure 9. Growth of SOD-deficient JI132 and wild type AB1157 E. coli in restricted, five amino-acid medium in the presence or absence of isomeric
Fe and Mn porphyrins. In the concentration range of 0.01−1 μM (1 μM is shown only), (OH)(H2O)FePs stimulated the growth of the SOD-
deficient E. coli. MnP stimulates E. coli growth at concentrations >5 μM (A). Growth curves, demonstrating differences in growth pattern (B); in the
presence of (H2O)2MnPs JI132 starts to growth simultaneously with the SOD-proficient strain. In contrast, (OH)(H2O)FePs does not shorten the
lag period but helps the growth of JI132 after it had already overcome the growth lag.
5686
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
curves show that in the presence of (H2O)2MnPs, the SOD-
deficient mutant started growth simultaneously with the SOD-
proficient parent. In a medium supplemented with (OH)-
(H2O)FeP, however, growth was delayed and did not start
earlier than in a nonsupplemented 5AA medium. Only after a
prolonged lag period was (OH)(H2O)FeP capable of stimulat-
ing the growth of the mutant to a rate comparable to that of the
SOD-proficient parent. Such data suggest that (OH)(H2O)FeP
stimulates the growth of the SOD-deficient E. coli by a
mechanism different than that of (H2O)2MnP despite the fact
that the two compounds have similar kcat(O2•−), E1/2 and
lipophilicity (Tables 4 and 5).
In a cell-free test tube Fe porphyrins are rapidly degraded by
hydrogen peroxide with release of Fe2+. A similar process can
occur in vivo. Organisms contain millimolar concentrations of
cellular reductants (ascorbate, glutathione and other thiols, etc.)
which are readily oxidized by Fe and Mn porphyrins acting as
catalysts.21,34 A product of such reactions is hydrogen peroxide,
which rapidly degrades FePs but not MnPs.
To further explore the reasons behind growth stimulation by
low concentration of Fe porphyrins, we compared it with the
effect of Fe(II) and Mn(II) salts under the same condi-
tions. The Mn porphyrin at 20 μM served as a positive control.
Figure 10 shows that at 1 μM Fe(II) citrate stimulated the
growth of the SOD-deficient E. coli similar to 1 μM Fe
porphyrin. Under the conditions of this experiment, Mn-citrate
did not improve the growth of any of the SOD-deficient strains
(Figure 10). Importantly, the pattern in E. coli growth with Mn
porphyrins and Fe salts is essentially the same (Figure 10).
To check if indeed the Fe porphyrin is degraded during
E. coli growth, cytosolic fractions of sodAsodB E. coli grown
in a medium supplemented with either Fe or Mn porphyrin
were analyzed over a 12 h period. Judging by the disappearance
of the Soret band, no loss of Mn porphyrin occurred during the
examination period, while the Fe porphyrin almost completely
disappeared from the E. coli cytosol (Figure 11) and from the
medium (data not shown). The time interval needed for SOD-
deficient E. coli to start growing coincides with the Fe porphyrin
almost complete degradation period (Figure 10 and 11).
Therefore, growth stimulation by the Fe porphyrin should be
Scheme 1. The Differential Aspects of the in Vivo
a
Mechanisms of Action of Fe and Mn Porphyrins
Figure 10. Comparison of the effect of Fe(II) citrate and Mn(II) citrate on
growth of two different SOD-deficient strains, JI1132 and QC1799, in 5AA
medium. The metal salts were used at concentrations of 1 and 20 μM. The
same effect was observed with other Mn and Fe salts (sulfates and
chlorides). The growth of SOD-proficient strains (AB1157 and
GC4468) and their SOD-deficient analogues (JI132 and QC1799) in
the absence of metal salts and metalloporphyrins is shown also.
a
The proposed mechanisms are based on the aerobic growth of SOD-
deficient E. coli. Mn and Fe porphyrins are presumably uptaken by a
heme transport mechanism. Subsequently FePs undergo rapid
degradation due to fast redox cycling with cellular reductants whereby
“free” Fe2+ is released. Fe2+ is readily chelated to iron-transporting/
sequestering siderophores (e.g., [Fch] - ferrochelatase; [Dps] - Fe-
storage protein; indicated as green circles). At very low levels, Fe2+
could reconstitute [4Fe-4S] clusters of enzymes, such as aconitases.
These enzymes undergo superoxide-driven oxidative degradation and
subsequent, yet reversible release of Fe2+. At high concentration of
Fe2+, the deleterious effects of Fenton chemistry ·OH radical
production prevail. The MnPs, however, due to their different
chemistry when compared to FePs are more resistant towards
oxidative degradation and protect E. coli by scavenging superoxide
whereby preventing its damaging effect upon different [4Fe-4S] clusters-
containing enzymes.
Figure 11. The disappearance of FeTE-2-PyP5+ from the cytosol of
E. coli during the 12-h growth in 5AA medium. Absorbances of Soret
band were given and relate to the different initial concentrations of Fe
porphyrin. For comparison the data on Mn porphyrin, MnTE-2-PyP5+
were given also.
5687
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
Figure 12. Comparison of the effects of Fe- and Mn porphyrins on mouse performance and weight. The behavior, weights and rotarod performances were
followed daily. The compounds were tested at doses which cause no or marginal, transient toxicity: (OH)(H2O)FeTE-2-PyP4+ (2 × 10 mg/kg/day),
(H2O)2MnTE-2-PyP5+ (2 × 10 mg/kg/day), (OH)(H2O)FeTnHex-2-PyP4+ (2 × 5 mg/kg/day) and (H2O)2MnTnHex-2-PyP5+ (2 × 1 mg/kg/day).
Figure 13. The comparison of the mouse effects of Fe- and Mn porphyrins. At 5 mg/kg single ip injection all 4 tested mice died within 2 h with
(H2O)2MnTnHex-2-PyP5+ while two daily ip injections of 5 mg/kg of (OH)(H2O)FeTnHex-2-PyP4+ for a week was well tolerated. Changes in
systolic blood pressure were monitored for an hour after intravenous injection of 10 mg/kg of either metalloporphyrin. No arterial hypotension was
observed with either (OH)(H2O)FeTnHex-2-PyP4+ or (OH)(H2O)FeTE-2-PyP4+ (data for ethyl (E) analogue not shown).
attributed to facilitation of microbial iron acquisition, rather
than to Fe porphyrin acting as an SOD mimic.
Scheme 1 summarizes the differential aspects of in vivo
mechanisms of action of Mn and Fe porphyrins as exemplified
by the protection of aerobic growth of SOD-deficient E. coli.
Mouse Study. We conducted a mouse study for initial
comparison of Fe- and Mn porphyrins in mammalian systems.
Four porphyrins were tested for toxicity to mice: (OH)(H2O)-
FeTE-2-PyP4+ (2 × 10 mg/kg/day), (H2O)2MnTE-2-PyP5+
(2 × 10 mg/kg/day), (OH)(H2O)FeTnHex-2-PyP4+ (2 ×
5 mg/kg/day) and (H2O)2MnTnHex-2-PyP5+ (2 × 1 mg/kg/day).
In most in vivo studies, several fold lower doses of Fe por-
phyrins than Mn porphyrins were used.3,4,86−88 To obtain stronger
response, we tested both Fe- and Mn porphyrins at com-
parable doses (in E. coli the nontoxic doses of Fe porphyrins
are 10−1000-fold lower than those of Mn porphyrins).
While not anticipated, significant toxic effects were observed
with (OH)(H2O)FeTE-2-PyP4+. The nature of those effects
is not yet understood.
Several enzymes, which include dihydroxy acid dehydratase,
aconitase, 6-phosphogluconate dehydratase, and fumarases A
and B, are oxidatively damaged by superoxide.70−74 As a result
Fe2+ is released from the [4Fe-4S]2+ clusters and the enzymes
are inactivated. The Fe2+ added exogenously may reconstitute
these enzymes and restore their activity.75 Our data demon-
strate a remarkable difference in the behavior of Fe vs Mn
porphyrin as a direct consequence of large difference in their
reactivities (as exemplified with ∼18-fold faster peroxide-based
degradation of Fe porphyrin) which in turn is a consequence of
5 log units difference in the electron-deficiency of the metal
site. Therefore, the rapid degradation of Fe porphyrins in vivo
with peroxide per se or peroxide produced while redox-cycling
with cellular reductants21,25,32 may have a major effect on their
mechanism of action.
Although efficacious at low concentrations (0.01−1 μM) Fe
porphyrins become toxic at >1 μM (Figure 11). In contrast,
Fe(II) salts promote the SOD-deficient E. coli growth even at
20 μM. This phenomenon is likely related to different transport
mechanisms that E. coli employs to control the uptake of Fe
from the medium. The uptake of “free” Fe is tightly regulated via
specific compounds  siderophores  which microorganism
excretes into the medium.76−83 Consequently, high amounts of
“free” Fe2+ (Fe salts) may not be taken up even if present in the
medium. Fe porphyrin bypasses such cellular control
mechanisms and presumably accumulates within the cell via
heme-uptake systems.84,85
Unexpectedly, (OH)(H2O)FeTnHex-2-PyP4+ appears less
toxic than its Mn counterpart, (H2O)2MnTnHex-2-PyP5+
(Figures 12 and 13). While with 5 mg/kg single injection of
(H2O)2MnTnHex-2-PyP5+ all four mice died,23 with (OH)-
(H2O)FeTnHex-2-PyP4+ no significant toxicity was seen in
4 out of 5 mice (one mouse died). These compounds have
similar lipophilicity. It is tempting to speculate that axial
hydroxo vs axial aqua ligand affects differentially the interaction
of Fe- and Mn porphyrins with biological targets which in turn
leads to differences in toxicity. It should be noted, however, that
long-alkyl chains hinder the impact of the type of axial ligand
analogues in all atropoisomers, except αααα. Another major
difference between the Mn and Fe porphyrins is the absence of
5688
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
a net positive charge on Fe site, as it is neutralized by the
presence of the axial hydroxo ligand.
NIH/NCI Duke Comprehensive Cancer Center Core Grant
(5-P30-CA14236-29) (I.S.), NIH U19AI067798, The Robert
Preston Tisch Brain Tumor Center at Duke Developmental
Career Award to IBH, and IBH general research funds (A.T.,
Z.R., T.W.). J.S.R. thanks CNPq and UFPB. I.B.H., D.S.W., and
I.S. are consultants with BioMimetix Pharmaceutical, Inc. L.B.
acknowledges the financial support from Kuwait University,
Grant MB01/09, the excellent technical assistance of Milini
Thomas, and Research Core Facility Grants GM01/01 and
GM01/05.
Fe- and Mn porphyrins had different effects on mice blood
pressure (Figure 13). It is known that Mn porphyrins cause
hypotension,89 and dramatic drop of blood pressure was ob-
served when (H2O)2MnTnHex-2-PyP5+ was injected (Figure 13).
In contrast, none of the tested Fe porphyrins exerted
hypotensive effect on mice. The hypotensive action of the
Mn porphyrins was strongly influenced by the length of the
alkyl chain at pyridyl ring. Thus, (H2O)2MnTnHex-2-PyP5+
was the strongest and (H2O)2MnTE-2-PyP5+ was the weakest
hypotensive agent. The difference in axial coordination may not
play major role either. Metal site is less hindered and more
exposed in short chain ethyl analogs, (H2O)2MnTE-2-PyP5+
and (OH)(H2O)FeTE-2-PyP4+, thus the differential effect of
axial hydroxo vs aqua ligand in their interactions with biological
target should have been mostly pronounced. However, a weak
hypotensive effect was observed with (H2O)2MnTE-2-PyP5+
and no hypotension with (OH)(H2O)FeTE-2-PyP4+. With
n-hexyl compounds, the long chains form a cavity around the
metal center, and axial ligation may not play a role other than in
αααα atropoisomer. With both Fe and Mn n-hexyl analogues,
the interaction with biological targets should be dominated by
similar hydrophobic nature of long alkyl chains. The nature of
systemic toxicity of Fe vs Mn porphyrin is, thus, not yet
understood and requires clearly further exploration.
ABBREVIATIONS
■
ONOO−, peroxynitrite; O2•−, superoxide only porphyrin
ligands are listed here; H2TM-2(or 3)-PyP4+, meso-tetrakis(N-
methylpyridinum-2(or 3)-yl)porphyrin; H2TE-2(or 3)-PyP4+,
meso-tetrakis(N-ethylpyridinum-2(or 3)-yl)porphyrin;
H2TnBu-2(or 3)-PyP4+, meso-tetrakis(N-n-butylpyridinum-
2(or 3)-yl)porphyrin; H2TnHex-2(or 3)-PyP4+, meso-tetrakis-
(N-n-hexylpyridinum-2(or 3)-yl)porphyrin; H2TnOct-2(or 3)-
PyP4+, meso-tetrakis(N-n-octylpyridinum-2(or 3)-yl)porphyrin;
when used, FeP and MnP abbreviations do not imply axial
ligation; (H2O)2MnPs, diaqua Mn(III) N-alkylpyridylporphyrins
and (OH)(H2O)FePs and (H2O)2FePs, monohydroxo-monoaqua
and diaqua Fe(III) N-alkylpyridylporphyrins, respectively, no
charges are indicated for simplicity; MnTE-2-PyP5+,
AEOL10113, Mn(III) meso-tetrakis(N-ethylpyridinum-2-yl)-
porphyrin; FeTM-4-PyP4+, Fe(III) meso-tetrakis(N-methylpyr-
idinum-4-yl)porphyrin; FP-15, Fe(III) meso-tetrakis(N-(1-(2-
(2-(2-methoxyethoxy)ethoxy)ethyl)pyridinium-2-yl)porphyrin;
INO-4885, the structure of WW-85 has not been made available
in J. Pharmacol. Exp. Ther.; DMF, N,N-dimethylformamide;
HFBA, heptafluorobutyric acid; Rf, thin-layer chromatographic
retention factor that presents the ratio between the solvent and
compound path in acetonitrile/KNO3(sat)/H2O = 8:1:1 solvent
system; E1/2, half-wave reduction potential; SOD, superoxide
dismutase; NHE, normal hydrogen electrode; POW, partition
coefficient between n-octanol and water; HA−, monodeproto-
nated ascorbic acid; A•−, ascorbyl radical; MTD, maximal
tolerable dose; charges omitted in some figures.
CONCLUSIONS
■
Water-soluble Mn(III) N-alkylpyridylporphyrins, (H2O)2MnPs
are bis-aqua complexes under physiological conditions, while
the analogous Fe porphyrins (OH)(H2O)FePs have a hydroxo
axial ligand trans axially coordinated to water molecule. As a
consequence the Fe site has 5 log units lower electron density
than its Mn analogue, which determines the differences
between Fe and Mn porphyrins in redox chemistry, reactivity,
and biological actions. The presence of hydroxo ligand labilizes
trans-axial water which increases the reactivity of Fe porphyrin
in comparison with the Mn porphyrin center. This was
demonstrated in vitro and in vivo with E. coli. In the presence of
high millimolar levels of cellular reductants (ascorbate,
glutathione, cysteine, etc.) and/or peroxide, Fe porphyrins are
rapidly degraded releasing “free” Fe2+. In contrast to Mn
porphyrin, delivery of iron but not the Fe porphyrin itself
stimulated the growth of the SOD-deficient E. coli. The
relevance of such intriguing biochemistry of Fe porphyrins to
eukaryotic systems is under further exploration to understand
the favorable effects of Fe porphyrins often reported in animal
models of oxidative stress injuries.2,3,88,90
REFERENCES
■
(1) Genovese, T.; Mazzon, E.; Esposito, E.; Di Paola, R.; Murthy, K.;
Neville, L.; Bramanti, P.; Cuzzocrea, S. Free Radic. Res. 2009, 43 (7),
631−645.
(2) Lange, M.; Szabo, C.; Enkhbaatar, P.; Connelly, R.; Horvath, E.;
Hamahata, A.; Cox, R. A.; Esechie, A.; Nakano, Y.; Traber, L. D.;
Herndon, D. N.; Traber, D. L. Am. J. Physiol. Lung Cell Mol. Physiol.
2011, 300 (2), L167−L175.
(3) Maybauer, D. M.; Maybauer, M. O.; Szabo, C.; Cox, R. A.;
Westphal, M.; Kiss, L.; Horvath, E. M.; Traber, L. D.; Hawkins, H. K.;
Salzman, A. L.; Southan, G. J.; Herndon, D. N.; Traber, D. L. Shock
2011, 35 (2), 148−155.
AUTHOR INFORMATION
Corresponding Authors
*(I.B.-H.) Address: Department of Radiation Oncology, Duke
University Medical Center, Durham, NC 27710. Tel: 919-684-
2101. E-mail: ibatinic@duke.edu. (L.B.) Address: Kuwait University,
Faculty of Medicine, Department of Biochemistry, Jabriya, 4th floor,
Rm 48 13110 Safat, Kuwait. E-mail: lbenov@HSC.EDU.KW.
■
(4) Maybauer, D. M.; Maybauer, M. O.; Szabo, C.; Westphal, M.;
Traber, L. D.; Salzman, A. L.; Herndon, D. N.; Traber, D. L. Burns
2011, 37 (5), 842−850.
(5) Pieper, G. M.; Nilakantan, V.; Chen, M.; Zhou, J.; Khanna, A. K.;
Henderson, J. D., Jr.; Johnson, C. P.; Roza, A. M.; Szabo, C. J.
Pharmacol. Exp. Ther. 2005, 314 (1), 53−60.
(6) Batinic-Haberle, I.; Reboucas, J. S.; Benov, L.; Spasojevic, I.,
Chemistry, biology and medical effects of water soluble metal-
loporphyrins. In Handbook of Porphyrin Science; Kadish, K. M., Smith,
K. M., Guillard, R., Eds.; World Scientific: Singapore, 2011; Vol. 11, pp
291−393.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Authors acknowledge financial help from W.H. Coulter
Translational Partners Grant Program (I.B.H., I.S., A.T.), and
5689
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
(7) Szabo, C.; Mabley, J. G.; Moeller, S. M.; Shimanovich, R.; Pacher,
P.; Virag, L.; Soriano, F. G.; Van Duzer, J. H.; Williams, W.; Salzman,
A. L.; Groves, J. T. Mol. Med. 2002, 8 (10), 571−580.
(35) Al-Maghrebi, M.; Fridovich, I.; Benov, L. Arch. Biochem. Biophys.
2002, 402 (1), 104−109.
(36) Brown, O. R.; Smyk-Randall, E.; Draczynska-Lusiak, B.; Fee, J.
A. Arch. Biochem. Biophys. 1995, 319 (1), 10−22.
(37) Benov, L.; Fridovich, I. J. Biol. Chem. 1999, 274 (7), 4202−
4206.
́ ́
(8) Batinic-Haberle, I.; Spasojevic, I.; Hambright, P.; Benov, L.;
Crumbliss, A. L.; Fridovich, I. Inorg. Chem. 1999, 38 (18), 4011−4022.
(9) Batinic-Haberle, I.; Benov, L.; Spasojevic, I.; Fridovich, I. J. Biol.
Chem. 1998, 273 (38), 24521−24528.
(10) Batinic-Haberle, I.; Rajic, Z.; Tovmasyan, A.; Reboucas, J. S.; Ye,
X.; Leong, K. W.; Dewhirst, M. W.; Vujaskovic, Z.; Benov, L.;
Spasojevic, I. Free Radic. Biol. Med. 2011, 51 (5), 1035−1053.
(11) Batinic-Haberle, I.; Reboucas, J. S.; Spasojevic, I. Antioxid. Redox
Signal 2010, 13 (6), 877−918.
(38) Benov, L.; Kredich, N. M.; Fridovich, I. J. Biol. Chem. 1996, 271
(35), 21037−21040.
(39) Imlay, J. A.; Linn, S. J. Bacteriol. 1987, 169 (7), 2967−2976.
(40) Touati, D.; Jacques, M.; Tardat, B.; Bouchard, L.; Despied, S. J.
Bacteriol. 1995, 177 (9), 2305−2314.
(41) Kos, I.; Benov, L.; Spasojevic, I.; Reboucas, J. S.; Batinic-
Haberle, I. J. Med. Chem. 2009, 52 (23), 7868−7872.
(42) Flynn, C. M. Chem. Rev. 1984, 84 (1), 31−41.
(43) Macartney, D. H.; Sutin, N. Inorg. Chem. 1985, 24 (21), 3403−
3409.
(12) Reboucas, J. S.; DeFreitas-Silva, G.; Spasojevic, I.; Idemori, Y.
M.; Benov, L.; Batinic-Haberle, I. Free Radic. Biol. Med. 2008, 45 (2),
201−210.
(13) Spasojevic, I.; Batinic-Haberle, I.; Reboucas, J. S.; Idemori, Y.
M.; Fridovich, I. J. Biol. Chem. 2003, 278 (9), 6831−6837.
(14) Batinic-Haberle, I.; Spasojevic, I.; Tse, H. M.; Tovmasyan, A.;
Rajic, Z.; St Clair, D. K.; Vujaskovic, Z.; Dewhirst, M. W.; Piganelli, J.
D. Amino Acids 2010, 42, 95−113.
(15) Miriyala, S.; Spasojevic, I.; Tovmasyan, A.; Salvemini, D.;
Vujaskovic, Z.; St Clair, D.; Batinic-Haberle, I. Biochim. Biophys. Acta
2012, 1822 (5), 794−814.
(16) Tovmasyan, A.; Sheng, H.; Weitner, T.; Arulpragasam, A.; Lu,
M.; Warner, D. S.; Spasojevic, I.; Batinic-Haberle, I. Med. Princ. Pract.
2013, 22, 103−130.
(17) Kos, I.; Reboucas, J. S.; DeFreitas-Silva, G.; Salvemini, D.;
Vujaskovic, Z.; Dewhirst, M. W.; Spasojevic, I.; Batinic-Haberle, I. Free
Radic. Biol. Med. 2009, 47 (1), 72−78.
(18) Pinto, V. H.; Carvalhoda-Silva, D.; Santos, J. L.; Weitner, T.;
Fonseca, M. G.; Yoshida, M. I.; Idemori, Y. M.; Batinic-Haberle, I.;
Reboucas, J. S. J. Pharm. Biomed. Anal. 2013, 73, 29−34.
(19) Reboucas, J. S.; Spasojevic, I.; Batinic-Haberle, I. J. Pharm.
Biomed. Anal. 2008, 48 (3), 1046−1049.
(44) Scheidt, W. R.; Cohen, I. A.; Kastner, M. E. Biochemistry 1979,
18 (16), 3546−3552.
(45) Korber, F. C. F.; Smith, J. R. L.; Prince, S.; Rizkallah, P.;
Reynolds, C. D.; Shawcross, D. R. J. Chem. Soc., Dalton Trans. 1991,
12, 3291−3294.
(46) Prince, S.; Korber, F.; Cooke, P. R.; Lindsay Smith, J. R.; Mazid,
M. A. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1993, 49 (6),
1158−1160.
(47) Aguirre, J. D.; Culotta, V. C. J. Biol. Chem. 2012, 287 (17),
13541−8.
(48) Ramirez-Gutierrez, O.; Claret, J.; Ribo, J. M. J. Porphyrins
Phthalocyanines 2005, 9 (6), 436−43.
(49) Batinic-Haberle, I.; Spasojevic, I.; Stevens, R. D.; Hambright, P.;
Neta, P.; Okado-Matsumoto, A.; Fridovich, I. Dalton Trans. 2004, 11,
1696−1702.
(50) Tian, J.; Peehl, D. M.; Knox, S. J. Cancer Biother. Radiopharm.
2010, 25 (4), 439−448.
(51) Tovmasyan, A.; Weitner, T.; Spasojevic, I.; Rajic, Z.; Benov, L.;
Batinic-Haberle, I. Free Radic. Biol. Med. 2011, 51, S99.
(52) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. Rev. 2004, 104 (9),
3947−3980.
(20) Tovmasyan, A. G.; Rajic, Z.; Spasojevic, I.; Reboucas, J. S.;
Chen, X.; Salvemini, D.; Sheng, H.; Warner, D. S.; Benov, L.; Batinic-
Haberle, I. Dalton Trans. 2011, 40 (16), 4111−4121.
(21) Ferrer-Sueta, G.; Batinic-Haberle, I.; Spasojevic, I.; Fridovich, I.;
Radi, R. Chem. Res. Toxicol. 1999, 12 (5), 442−449.
(22) Spasojevic, I.; Batinic-Haberle, I.; Stevens, R. D.; Hambright, P.;
Thorpe, A. N.; Grodkowski, J.; Neta, P.; Fridovich, I. Inorg. Chem.
2001, 40 (4), 726−739.
(53) Almarsson, O.; Bruice, T. C. J. Am. Chem. Soc. 1995, 117 (16),
4533−4544.
(54) Murata, K.; Panicucci, R.; Gopinath, E.; Bruice, T. C. J. Am.
Chem. Soc. 1990, 112 (16), 6072−6083.
(55) Meunier, B.; Bernadou, J. Top. Catal. 2002, 21 (1), 47−54.
(56) Newcomb, M.; Toy, P. H. Acc. Chem. Res. 2000, 33 (7), 449−
455.
(23) Rajic, Z.; Tovmasyan, A.; Spasojevic, I.; Sheng, H.; Lu, M.; Li, A.
M.; Gralla, E. B.; Warner, D. S.; Benov, L.; Batinic-Haberle, I. Free
Radic. Biol. Med. 2012, 52 (9), 1828−1834.
(57) Aguey-Zinsou, K.-F.; Bernhardt, P. V.; De Voss, J. J.; Slessor, K.
(24) Rodgers, K. R.; Reed, R. A.; Su, Y. O.; Spiro, T. G. Inorg. Chem.
1992, 31 (13), 2688−2700.
E. Chem. Commun. 2003, 3, 418−419.
(58) Nam, W.-H.; Rhee, S.-W. ChemInform 2004, 35 (50), 201−205.
(59) Nam, W.; Jin, S. W.; Lim, M. H.; Ryu, J. Y.; Kim, C. Inorg. Chem
2002, 41 (14), 3647−3652.
(25) Ferrer-Sueta, G.; Vitturi, D.; Batinic-Haberle, I.; Fridovich, I.;
Goldstein, S.; Czapski, G.; Radi, R. J. Biol. Chem. 2003, 278 (30),
27432−27438.
(60) Groves, J. T. J. Inorg. Biochem. 2006, 100 (4), 434−447.
(61) Nam, W.; J. Choi, H.; J. Han, H.; H. Cho, S.; J. Lee, H.; Han, S.-
Y. Chem. Commun. 1999, 4, 387−388.
(26) Weitner, T.; Budimir, A.; Kos, I.; Batinic-Haberle, I.; Birus, M.
Dalton Trans. 2010, 39 (48), 11568−11576.
(27) Chen, S.-M.; Sun, P.-J.; Su, Y. O. J. Electroanal. Chem. Interfacial
Electrochem. 1990, 294 (1−2), 151−164.
(28) Forshey, P. A.; Kuwana, T. Inorg. Chem. 1981, 20 (3), 693−700.
(29) Kobayashi, N. Inorg. Chem. 1985, 24 (21), 3324−3330.
(30) Kobayashi, N.; Koshiyama, M.; Osa, T.; Kuwana, T. Inorg. Chem.
1983, 22 (24), 3608−3614.
(31) Tondreau, G. A.; Wilkins, R. G. Inorg. Chem. 1986, 25 (16),
2745−2750.
(32) Tovmasyan, A.; Weitner, T.; Roberts, E.; Jaramillo, M.;
Spasojevic, I.; Leong, K.; Tome, M.; Benov, L.; Batinic-Haberle, I.
Free Radic. Biol. Med. 2012, 53 (2), S120.
(33) Spasojevic, I.; Colvin, O. M.; Warshany, K. R.; Batinic-Haberle,
I. J. Inorg. Biochem. 2006, 100 (11), 1897−1902.
(62) Song, W.; Ryu, Y.; Song, R.; Nam, W. J. Biol. Inorg. Chem. 2005,
10 (3), 294−304.
(63) Beck, M. J.; Gopinath, E.; Bruice, T. C. J. Am. Chem. Soc. 1993,
115 (1), 21−29.
(64) Arasasingham, R. D.; Bruice, T. C. J. Am. Chem. Soc. 1991, 113
(16), 6095−6103.
(65) Zipplies, M. F.; Lee, W. A.; Bruice, T. C. J. Am. Chem. Soc. 1986,
108 (15), 4433−4445.
(66) Panicucci, R.; Bruice, T. C. J. Am. Chem. Soc. 1990, 112 (16),
6063−6071.
(67) Jeon, S.; Bruice, T. C. Inorg. Chem. 1992, 31 (23), 4843−4848.
(68) Chen, Q.; Espey, M. G.; Sun, A. Y.; Pooput, C.; Kirk, K. L.;
Krishna, M. C.; Khosh, D. B.; Drisko, J.; Levine, M. Proc. Natl. Acad.
Sci. U. S. A. 2008, 105 (32), 11105−11109.
(34) Ye, X.; Fels, D.; Tovmasyan, A.; Aird, K. M.; Dedeugd, C.;
Allensworth, J. L.; Kos, I.; Park, W.; Spasojevic, I.; Devi, G. R.;
Dewhirst, M. W.; Leong, K. W.; Batinic-Haberle, I. Free Radic. Res.
2011, 45 (11−12), 1289−1306.
(69) Spasojevic, I.; Kos, I.; Benov, L. T.; Rajic, Z.; Fels, D.; Dedeugd,
C.; Ye, X.; Vujaskovic, Z.; Reboucas, J. S.; Leong, K. W.; Dewhirst, M.
W.; Batinic-Haberle, I. Free Radic. Res. 2011, 45 (2), 188−200.
5690
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691

Inorganic Chemistry
Article
(70) Flint, D. H.; Tuminello, J. F.; Emptage, M. H. J. Biol. Chem.
1993, 268 (30), 22369−22376.
(71) Gardner, P. R.; Fridovich, I. J. Biol. Chem. 1991, 266 (3), 1478−
1483.
(72) Gardner, P. R.; Fridovich, I. J. Biol. Chem. 1991, 266 (29),
19328−33.
(73) Kuo, C. F.; Mashino, T.; Fridovich, I. J. Biol. Chem. 1987, 262
(10), 4724−4727.
(74) Liochev, S. I.; Fridovich, I.; Fumarase, C. Proc. Natl. Acad. Sci. U.
S. A. 1992, 89 (13), 5892−5896.
(75) Benov, L.; Fridovich, I. J. Biol. Chem. 1998, 273 (17), 10313−
10316.
(76) Bagg, A.; Neilands, J. B. Biochemistry 1987, 26 (17), 5471−5477.
(77) Bagg, A.; Neilands, J. B. Microbiol. Rev. 1987, 51 (4), 509−518.
(78) Braun, V. Int. J. Med. Microbiol. 2001, 291 (2), 67−79.
(79) Braun, V. Front Biosci. 2003, 8, s1409−s1421.
(80) Braun, V.; Killmann, H. Trends Biochem. Sci. 1999, 24 (3), 104−
109.
(81) Hantke, K. FEMS Microbiol. Lett. 1982, 15 (2), 83−86.
(82) Kammler, M.; Schon, C.; Hantke, K. J. Bacteriol. 1993, 175 (19),
6212−6219.
(83) Touati, D.; Jacques, M.; Tardat, B.; Bouchard, L.; Despied, S. J.
Bacteriol. 1995, 177 (9), 2305−2314.
(84) Anzaldi, L. L.; Skaar, E. P. Infect. Immun. 2010, 78 (12), 4977−
4989.
(85) Tong, Y.; Guo, M. Arch. Biochem. Biophys. 2009, 481 (1), 1−15.
(86) Radovits, T.; Beller, C. J.; Groves, J. T.; Merkely, B.; Karck, M.;
Szabo, C.; Szabo, G. Eur. J. Cardiothorac. Surg. 2012, 41 (2), 391−396.
(87) 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 (2), 173−181.
(88) Szabo, G.; Loganathan, S.; Merkely, B.; Groves, J. T.; Karck, M.;
Szabo, C.; Radovits, T. J. Thorac. Cardiovasc. Surg. 2012, 143, 1443−
1449.
(89) Ross, A. D.; Sheng, H.; Warner, D. S.; Piantadosi, C. A.; Batinic-
Haberle, I.; Day, B. J.; Crapo, J. D. Free Radic. Biol. Med. 2002, 33 (12),
1657−1669.
(90) Soriano, F. G.; Lorigados, C. B.; Pacher, P.; Szabo, C. Shock
2011, 35 (6), 560−566.
5691
dx.doi.org/10.1021/ic3012519 | Inorg. Chem. 2013, 52, 5677−5691