DNA binding and catalytic properties of positively charged corroles

Article abstract of DOI:10.1002/anie.200700757

DNA likes corroles and peroxynitrite doesn't: The water-soluble manganese corrole (crystal structure shown) is a much better catalyst than the analogous porphyrin for decomposition of peroxynitrite (HOONO). The interactions of the two complexes with DNA are also significantly different. These findings suggest that positively charged corroles may be useful for therapeutic approaches. (Figure Presented)

Full text of DOI:10.1002/anie.200700757

Communications
DOI: 10.1002/anie.200700757
Corroles
DNA Binding and Catalytic Properties of Positively Charged
Corroles**
Zoya Gershman, Israel Goldberg,* and Zeev Gross*
The recent disclosure of practical approaches for the synthesis
of triarylcorroles and their selective modifications[1,2] allowed
for the utilization of the corresponding metal complexes as
key components in catalysis, solar cells, sensors, and medicinal
applications.[3,4] Corroles with sulfonate head groups,[2] which
induce water-solubility owing to their negative charge, played
an important role in many of these cases.[4] On the other hand,
there is only a single report about a corrole with positively
charged substituents. This particular derivative was used in
what remains the only reported medicinal investigation
in vivo with corroles.[5] The corrole with three remote positive
charges displayed significantly better efficacy in inhibiting
tumor progression and metastasis in animal models than
analogous porphyrins. The main and probably sole reason for
no other reports regarding positively charged corroles
Scheme 1. Positively charged porphyrins (left) and analogous corroles
appears to be the synthetic challenge. The trispyridylium-
substituted corrole ((3’)3+ in Scheme 1) has not yet been
reported, whereas the corresponding 5,10,15,20-tetrakis(4-N-
methylpyridyl)porphyrin (TMPyP; (1’)4+ in Scheme 1) is the
most intensively investigated porphyrin for medicinal and
biological applications. Specific examples include the inter-
actions between (1’)4+ and DNA and utilization of the metal
complexes of (1’)4+ for site-specific cleavage of DNA and
decomposition of reactive oxygen species (ROS).[6,7]
(right).
Herein we report a facile route to the tris- and bispyridyl
corroles 3 and 4, respectively. Metalation by manganese and
subsequent N-methylation led to the water-soluble
manganese(III) corrole [(4’)Mn]2+ (Scheme 2) whose molec-
ular structure was resolved by X-ray crystallography. The
comparison of [(4’)Mn]2+ and the manganese(III) complex of
the analogous porphyrin (2’)2+ ([(2’)Mn(Cl)]2+) revealed
unique differences regarding both DNA binding and perox-
ynitrite decomposition.
Scheme 2. Synthesis of the water-soluble manganese(III) corrole with
Targeting the synthesis of pyridyl-substituted corroles, we
have first found that the solvent-free method (best suited for
two pyridinium groups.
electron-poor aldehydes)[1] failed when applied to condensa-
tion of pyrrole with 4-pyridinecarboxaldehyde. Modifications
introduced by Gryko and Jadach (trifluoroacetic acid)[8] and
Collman and Decreau (microwaves)[9] also did not yield
appreciable amounts of the desired compound. Large effort
was devoted to further development of the stepwise-approach
technique advanced by Gryko and Piechota: condensation of
preprepared dipyrromethane with aldehyde followed by
oxidation with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ).[10] The utilization of pyridine-substituted dipyrro-
methane 5[11,12] and 4-pyridinecarboxaldehyde was not fruit-
ful, but the combination of 5 with the very reactive
pentafluorobenzaldehyde was. The bispyridyl corrole 4 was
isolated in yields of 2.1–2.4%, accompanied by porphyrin 2
[*] Z. Gershman, Prof. Dr. Z. Gross
Schulich Faculty of Chemistry
Technion—Israel Institute of Technology
Haifa 32,000 (Israel)
Fax: (+972)4829-5703
E-mail: chr10zg@tx.technion.ac.il
Prof. Dr. I. Goldberg
School of Chemistry
Tel Aviv University
Ramat Aviv 69978 (Israel)
Fax: (+972)3640-9293
E-mail: goldberg@post.tau.ac.il
[**] This work was supported by the German Israel Cooperation (Z.G.)
and by the Israel Science Foundation under Grants No. 254/04
(I.G.) and 330/04 (Z.G.).
4320
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 4320 –4324

Angewandte
Chemie
(Scheme 1). Much better results were obtained by combining
the stepwise approach with Adler–Longo conditions
(Scheme 2): the heating of dipyrromethane 5 with penta-
fluorobenzaldehyde in propionic acid yielded 8% of corrole 4
without the aid of DDQ (relying on air as oxidant).[13]
Encouraged by this result, the same procedure was also
performed with 4-pyridinecarboxaldehyde; a 6% yield of the
desired corrole 3 was obtained after separation from 1.[14]
Manganese was inserted into the bispyridyl-substituted
corrole and porphyrin derivatives 4 and 2, respectively,[15]
which were subsequently N-methylated to afford the water-
soluble manganese(III) complexes [(4’)Mn]2+ (Scheme 2) and
[(2’)Mn(Cl)]2+ (Scheme 1).[16] Crystallization of the bispyridi-
nium-substituted derivative [(4’)Mn]2+ from a diethyl ether/
hexane/methanol mixture ( ꢀ 2:3:1) yielded crystals suitable
for X-ray crystallography analysis.[17] It represents the first
crystal structure determination of such a corrole; notably,
only two structures of positively charged manganese porphyr-
ins have been reported in the literature.[18,19]
positively charged N-methyl pyridylium groups bending
outward to minimize the repulsive interactions between
them (Figure 1). The latter observation supports the notion
that there is specific attraction between the two corrole
complexes. The crystal structure of this material contains
additional molecules of methanol and water solvent in the
interface between these dimers, which hydrogen bond to the
iodine counterions as well as to the axial ligands.
The corrole and porphyrin complexes with two pyridin-
ium substituents, [(4’)Mn]2+ and [(2’)Mn(Cl)]2+, respectively,
were investigated with respect to two important aspects:
interaction with DNA and the catalytic decomposition of
peroxynitrite. The method chosen for investigating the former
aspect was induced circular dichroism (ICD), which is the
most straightforward indication for noncovalent binding of
such molecules to chiral hosts.[6] The ICD spectra of
[(4’)Mn]2+ and [(2’)Mn(Cl)]2+ shown in Figure 2a,b provide
The molecular structure of [(4’)Mn]2+ in the respective
crystal (Figure 1) is slightly domed in shape owing to an
additional molecule of methanol (from the crystallization
Figure 2. The visible CD spectra obtained before (gray line) and after
(black line) addition of 27 mm calf-thymus DNA to 6 mm aqueous
solutions of the manganese complexes at two ionic strengths (I).
Figure 1. Molecular structure of [(4’)Mn(MeOH]2+·2Iꢁ. The thermal
ellipsoids are shown at 50% probability.
a) [(4’)Mn]2+, I=0.01m; b) [(2’)Mn(Cl)]2+, I=0.01m; c) [(4’)Mn]2+
I=0.2m; d) [(2’)Mn(Cl)]2+, I=0.2m. Note: Identical results were
obtained at a 200:6 DNA/chromophore ratio.
,
solvent) that ligates axially to the central MnIII ion at
Mn···O(methanol) = 2.19 Š. The five-coordinate MnIII ion
deviates slightly from the mean plane of the inner pyrrole
N atoms towards the axial ligand by 0.20–0.21 Š, and all
equatorial Mn···N bonds are within 1.90–1.92 Š. This is
characteristic of manganese(III) corrole complexes, whereas
in analogous porphyrin complexes, these bond distances are
considerably longer, 1.99–2.02 Š, owing to the larger size of
the porphyrin macrocycle. Also noteworthy is that the
[(4’)Mn]2+ moieties form dimers in the crystal with relatively
close contacts between the concave sides of the two species,
which face one another (an entire dimer also comprises the
asymmetric unit of the structure). Thus, the nonbonding
distances between the nearly overlapping C atoms of the
macrocyclic core in the two units are within 3.4–3.6 Š, and the
distance between the corresponding Mn ion centers is 4.17 Š.
This close interaction is associated with further significant
“saddle”-type distortion of the corrole rings owing to the
clear-cut evidence for the chromophores residing within the
chiral pocket provided by the macromolecule:[20] both DNA
(no absorbance in the visible) and the chromophore (achiral)
otherwise display no CD signals in that part of the spectrum
(Figure 2, gray lines). The difference is that although the
spectrum of [(4’)Mn]2+ is composed of both negative and
positive components (Figure 2a), that of the analogous
porphyrin derivative [(2’)Mn(Cl)]2+ displays only a positive
CD band that is also of much lower ellipticity (Figure 2b).
These phenomena may be analyzed with the aid of empirical
rules that were developed regarding porphyrin binding modes
to DNA: intercalation is characterized by a negative CD
band, whereas a positive CD band is indicative of outside
(groove) binding.[21,22] Assuming that the same rules hold for
corroles as well, the results suggest that the corrole is capable
of intercalation into DNA, whereas the analogous porphyrin
is only capable of outside binding. This may be attributed to
Angew. Chem. Int. Ed. 2007, 46, 4320 –4324
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4321

Communications
differences in the coordination chemistry of the manganese-
products. We have recently demonstrated that manganese and
iron complexes of negatively charged corroles are potent
catalysts for that purpose.[31] The former was identified as the
first manganese complex displaying catalytic activity without
the aid of sacrificial agents and it did so by a genuine
(III) ion chelated by dianionic porphyrins and trianionic
corroles. Four nitrogen atoms define the equatorial plane in
both complexes, but a full charge balance exists in the latter
case only. Manganese porphyrin complexese require at least
one strongly bound axial ligand,[18,19] which is considered the
main obstacle for intercalation.[21] On the other hand, the
molecular structure of [(4’)Mn]2+ reveals only one axially
coordinated solvent molecule; four-coordinated manganese-
(III) corroles have been reported as well.[23] Accordingly, the
probability of forming a four-coordinated manganese(III)
complex, which is required for DNA intercalation, may safely
be concluded to be much larger for corroles than for
porphyrins. We further note that the earlier discussed
intermolecular interactions evident in the solid-state structure
of [(4’)Mn]2+ hint at p-stacking interactions with DNA base
pairs as an additional driving force for intercalation. The
importance of this aspect will be addressed in future inves-
tigations with the metal-free derivatives.
Another difference is evident in the effect of ionic
strength (I) on the ICD of the two complexes (Figure 2c,d):
an increase in I from 0.01 to 0.2m completely eliminated the
signal for the porphyrin complex, whereas that of the corrole
analogue only diminished in intensity (to about one third).
Both the positive and the negative CD bands remained quite
intense, suggesting that both intercalation (the negative band)
and groove binding (the positive band) play an important role
for [(4’)Mn]2+ even at I = 0.2m. High ionic strength may be
expected to affect all binding modes of positively charged
macrocycles to DNA owing to the competition of chromo-
phore molecules and electrolyte cations for the negatively
charged moieties on the DNA superstructure.[6] The larger
ratio of [positive charges]/[molecular size] of the corrole
(owing to the “missing” C6F5 group) could well be the reason
for what appears as an overall larger affinity for DNA of the
[(4’)Mn]2+ ion relative to the [(2’)Mn(Cl)]2+ ion.
ꢁ
disproportionation process forming only NO2 and O2
(Scheme 3, route c). One limitation is that the catalytic rate
(kcat = 4.0 ” 104 mꢁ1 sꢁ1 at 258C) is quite close to that of the
oxidation of biological targets by peroxynitrite (103–
104 mꢁ1 sꢁ1)[29b] and is significantly slower than that obtained
by the combination of manganese(III) porphyrins and
reducing agents.[32–34] Our motivation for examining the new
compounds relied on literature data that showed that metal
complexes of positively charged porphyrins are faster-per-
forming catalysts than negatively charged analogues.[34]
The rate constants for decomposition of peroxynitrite at
pH 7.4and 25 8C were determined with and without the
presence of the potential catalyst by following the decay of
385 mm peroxynitrite at its characteristic lmax (302 nm) by
using a stopped-flow spectrophotometer.[36] Importantly, the
iodide counterions in the complexes were first exchanged by
The structurally related corrole and porphyrin manganese
complexes were also investigated as catalysts for decompo-
sition of peroxynitrite (HOONO),[7,24] a cytotoxic agent that
allegedly plays a major role in many pathological condi-
tions.[25–29] HOONO, produced by the combination of nitric
oxide and the superoxide anion, decomposes at physiological
pH values by two distinct pathways (Scheme 3): isomerization
to nitrate (route a) and homolytic bond cleavage (route b)
into the most-reactive oxygen and nitrogen species COH and
CNO2.[30] The urgent need to develop synthetic catalysts as a
means for protecting cells and tissues from peroxynitrite
formed in situ is highlighted by the absence of a specific
enzyme capable of transforming it into biologically benign
Figure 3. a) Decomposition of 385 mm peroxynitrite at pH 7.4 and
258C monitored at l=302 nm as a function of the concentration of
[(4’)Mn]2+. Inset: Plot used for determination of the catalytic rate
constant. b) The same experiments with catalytic amounts of
[(2’)Mn(Cl)]2+, with and without 300 mm ascorbate (asc).
Scheme 3. Spontaneous (a and b) and desired (c) pathways for
decomposition of peroxynitrite.
4322
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 4320 –4324

Angewandte
Chemie
[4] a) A. Mahammed, Z. Gross, J. Am. Chem. Soc. 2005, 127, 2883;
chlorides to eliminate possible reduction of peroxynitrite by
the former ions.[35]
b) H. Agadjanian, J. J. Weaver, A. Mahammed, A. Rentsendorj,
S. Bass, J. Kim, I. J. Dmochowski, R. Margalit, H. B. Gray, Z.
Gross, L. K. Medina-Kauwe, Pharm. Res. 2006, 23, 367; c) D.
Walker, S. Chappel, A. Mahammed, B. S. Brunschwig, J. R.
Winkler, H. B. Gray, A. Zaban, Z. Gross, J. Porphyrins
Phthalocyanines 2006, 10, 1259.
The results uncovered large differences between the two
complexes: the [(4’)Mn]2+ ion displayed the characteristics of
a true catalyst (Figure 3a), but the [(2’)Mn(Cl)]2+ ion had
absolutely no effect on peroxynitrite unless used in combina-
tion with ascorbate as a coreductant (Figure 3b). A catalytic
rate constant of 4.0 ” 105 mꢁ1 sꢁ1 for decomposition of perox-
ynitrite by [(4’)Mn]2+ was elucidated from the linear relation-
ship shown in the inset of Figure 3a; one order of magnitude
larger than that of negatively charged manganese corrole.[31]
What is more, it is as large as the one we obtained for
[(2’)Mn(Cl)]2+ (kcat = 4.3 ” 105 mꢁ1 sꢁ1) in the presence of
ascorbate. Without a sacrificial agent, this derivative (similar
to all other non-corrole manganese complexes) displayed no
catalytic activity.
We report the synthesis and molecular structure of a novel
positively charged water-soluble manganese(III) corrole as
well as its utilization for DNA binding and for catalyzing
peroxynitrite decomposition. Relying on the characteristics
that were developed for CD analysis of porphyrin/DNA
binding, we conclude that the interaction of the manganese
corrole with DNA is strong enough to be observed even at
high ionic strengths with the possibility of intercalation into
DNA. Both phenomena are unique and not observed for the
analogous porphyrin. The positively charged [(4’)Mn]2+ com-
plex also appears to be a ten-times-faster catalyst for
decomposition of peroxynitrite than a negatively charged
manganese corrole: it actually reacts with peroxynitrite as fast
as ascorbate-aided manganese porphyrin and faster than
biological targets. Taken together, positively charged corroles
may be very useful for therapeutic approaches that rely on
specific interactions with DNA and as decomposition cata-
lysts of reactive oxygen and nitrogen species (the manganese
or other transition metal complexes).
[5] D. Aviezer, S. Cotton, M. David, A. Segev, N. Khaselev, N. Galili,
Z. Gross, A. Yayon, Cancer Res. 2000, 60, 2793.
[6] R. F. Pasternack, Chirality 2003, 15, 329, and references therein.
[7] J. T. Groves, Curr. Opin. Chem. Biol. 1999, 3, 226.
[8] D. T. Gryko, K. Jadach, J. Org. Chem. 2001, 66, 4267.
[9] J. P. Collman, R. A. Decreau, Tetrahedron Lett. 2003, 44, 1207.
[10] D. T. Gryko and K. E. Piechota, J. Porphyrins Phthalocyanines
2002, 6, 81.
[11] D. Gryko, J. S. Lindsey, J. Org. Chem. 2000, 65, 2249.
[12] Synthesis of 5: The required dipyrromethane (1.03 g) was
prepared as in Ref. 15 (yield = 49%), but purified by two
chromatographic treatments (alumina, dichloromethane/ethyl
acetate 5:1, followed by silica gel, dichloromethane/ethyl acetate
10:0!10:1).
[13] Synthesis of 4: Pentafluorobenzaldehyde (50 mL, 0,4mmol) was
added to a 10-mL solution of 5 (178 mg, 0.8 mmol) in propionic
acid and the mixture was heated to reflux for 50 min. The residue
obtained after solvent evaporation was washed with hot water,
neutralized with ammonium hydroxide (25%), and washed
again with hot water. The solid material was dissolved in
methanol, basic alumina was added, and the solvent was
evaporated. Separation between 4 and the analogous porphyrin
2 was achieved by column chromatography (silica, CH2Cl2
followed by 0.5% methanol) followed by separation by prepa-
rative thin-layer chromatography (silica plate, CHCl3/MeOH
50:1) affording pure 4 (18 mg, 8%). 4: Rf = 0.15 (CH2Cl2/ethyl
acetate 1:1). UV/Vis (CH2Cl2/MeOH (2:1)): lmax (e10ꢁ3) = 416
(104.99), 576 (16.14), 610 (9.40), 640 (5.34). MS (MALDI-TOF):
1
m/z (%): 619 (100) [M+]. H NMR (200 MHz, C6D6): d = 8.93
(br s, 4H), 8.67 (d, J = 4Hz, 4H), 8.26 (m, 4H), 7.90 ppm (br. s,
4H). 19F (188 MHz): d = ꢁ138.79 (d, J = 23.5 Hz, 2F), ꢁ153.22
(t, J = 21.9 Hz, 1F), ꢁ162.37 ppm (t, J = 22.5 Hz, 1F). 2: Rf =
0.37 (CH2Cl2/ethyl acetate 1:1). UV/Vis (CH2Cl2): lmax (e10ꢁ3) =
414 (233.34), 510 (15.51), 542 (3.17), 586 (4.69). MS (MALDI-
TOF LD+): m/z (%): 796 (100) [M+]. 1H NMR (200 MHz,
CDCl3): d = 9.05 (d, J = 5.2 Hz, 4H), 8.87 (m, 8H), 8.15 (d, J =
5.4Hz, 4H), ꢁ2.94ppm (s, 2H). 19F (188 MHz): d = ꢁ137.22 (dd,
3J = 23.3 Hz, 4J = 7.9 Hz, 4F), 151.96 (t, J = 20.9 Hz, 2F),
ꢁ161.88 ppm (td, 3J = 22.6 Hz, 4J = 8.3 Hz, 4F).
Received: February 19, 2007
Published online: May 4, 2007
Keywords: corroles · DNA · manganese · peroxynitrite ·
.
porphyrinoids · structure elucidation
[14] 3: 4-Pyridinecarboxaldehyde (38 mL, 0.40 mmol) was added to a
10-mL solution of 5 (178 mg, 0.8 mmol) in propionic acid, and
the mixture was heated at reflux for 70 min. The residue
obtained after solvent evaporation was washed with hot water,
neutralized with ammonium hydroxide (25% ammonia), and
washed again with hot water. The solid material was dissolved in
methanol, basic alumina was added, and the solvent was
evaporated. Separation of 3 was achieved by column chroma-
tography (silica, CH2Cl2 followed by 0.5% methanol) followed
by separation on PTLC (silica plate, CHCl3/MeOH 100:1). The
faster eluting fraction was comprised of the brownish red
porphyrin (1) and the next slightly fluorescent dark-green-
colored fraction afforded the desired corrole 3 (6%, 12 mg).
Alternatively, good chromatographic separation could be ach-
ieved by eluting with ethyl acetate to which methanol was
gradually added. Rf = 0.73 (CH2Cl2/MeOH 5:1). UV/Vis
(CH2Cl2): lmax (e10ꢁ3) = 418 (61.15), 576 (9.85), 614 (5.62); MS
(MALDI-TOF LD+): m/z (%): 530 (100) [M+H]. 1H NMR
(300 MHz), (C6D6): d = 7.87 (d, J = 5.1 Hz, 2H), 7.99 (d, J =
5.1 Hz, 4H), 8.35 (d, J = 4.8 Hz, 2H), 8.43 (d, J = 4.2 Hz, 2H),
[1] For the original syntheses of triarylcorroles, see: a) Z. Gross, N.
Galili, I. Saltsman, Angew. Chem. 1999, 111, 1530; Angew. Chem.
Int. Ed. 1999, 38, 1427; b) R. Paolesse, L. Jaquinod, D. J. Nurco,
S. Mini, F. Sagone, T. Boschi, K. M. Smith, Chem. Commun.
1999, 1307; c) Z. Gross, N. Galili, L. Simkhovich, I. Saltsman, M.
Botoshansky, D. Blaeser, R. Boese, I. Goldberg, Org. Lett. 1999,
1, 599; For reviews about corrole syntheses, see: d) D. T. Gryko,
Eur. J. Org. Chem. 2002, 1735; e) S. Nardis, D. Monti, R.
Paolesse, Mini-Rev. Org. Chem. 2005, 2, 355.
[2] For modifications of triarylcorroles, see: a) A. Mahammed, I.
Goldberg, Z. Gross, Org. Lett. 2001, 3, 3443; b) I. Saltsman, A.
Mahammed, I. Goldberg, E. Tkachenko, M. Botoshansky, Z.
Gross, J. Am. Chem. Soc. 2002, 124, 7411; c) Z. Gross, A.
Mahammed, J. Porphyrins Phthalocyanines 2002, 6, 553.
[3] a) Z. Gross, H. B. Gray, Adv. Synth. Catal. 2004, 346, 165;
b) D. T. Gryko, J. P. Fox, D. P. Goldberg, J. Porphyrins Phthalo-
cyanines 2004, 8, 1091; c) I. Aviv, Z. Gross, Chem. Commun.
2007, DOI: 10.1039/b618482k.
Angew. Chem. Int. Ed. 2007, 46, 4320 –4324
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4323

Communications
8.66 (d, J = 4.8 Hz, 2H), 8.78 (d, J = 4.2 Hz, 2H), 9.02 ppm (brs,
6H).
[20] Stock solutions were prepared by dissolving calf-thymus DNA
(1 mg, Sigma Aldrich) in 1 mL of doubly distilled water and were
left overnight at 48C. Experiments were carried out at 248C and
pH 6.8 (5 mm NaH2PO4, 2.5 mm Na2HPO4, and either 0.01m or
0.2m NaCl); DNA concentrations were calculated by using
[15] Manganese insertion: 2-Mn and 4-Mn were prepared by heating
the porphyrin/corrole pyridine solution at reflux with 15 equiv-
alents of Mn(OAc)2·4H2O followed by chromatographic sepa-
ration (silica, starting with CH2Cl2 and gradually adding
methanol), affording 92%, and 81% yield, respectively. 2-Mn:
UV/Vis (MeOH): lmax (e10ꢁ3) = 328 (12.6), 370 (22.5), 392 (19.5),
458 (49.7), 554 (5.2), 766 (0.84). MS (MALDI-TOF LD+): m/z
(%): 849 (100) [M+]. 19F (MeOD) (188 MHz): d = ꢁ134.49 (br s,
4F), ꢁ148.20 (s, 2F), ꢁ157.08 ppm (s, 4F). 4-Mn: UV/Vis
(MeOH): lmax (e10ꢁ3) = 368 (16.7), 402 (26.7), 420 (4.8), 458
(18.4), 484 (16.2), 634 (9.4). MS (MALDI-TOF LD+): m/z (%):
670 (100) [M+]. 19F (C5D4N) (188 MHz): d = ꢁ136.58 (br s, 2F),
ꢁ155.16 (s, 1F), ꢁ161.23 ppm (s, 2F)
e
262nm = 1.32 ” 104 mꢁ1 cm1.[21]
[21] R. F. Pasternack, E. J. Gibbs, J. J. Villafranca, Biochemistry 1983,
22, 2406.
[22] R. Kuroda, H. Tanaka, J. Chem. Soc. Chem. Commun. 1994,
1575.
[23] S. Licoccia, E. Morgante, R. Paolesse, F. Polizio, M. Senge, E.
Tondello, T. Boschi, Inorg. Chem. 1997, 36, 1564.
[24] a) J. T. Groves, S. S. Marla, J. Am. Chem. Soc. 1995, 117, 9578;
b) C. Szabꢁ, J. G. Mabley, S. M. Moeller, R. Shimanovich, P.
Pacher, L. Virꢂg, F. G. Soriano, J. H. Van Duzer, W. Williams,
A. L. Salzman, J. T. Groves, Mol. Med. 2002, 8, 571.
[25] S. A. Lipton, Y. B. Choi, Z. H. Pan, S. Z. Lei, H.-S. Vincent
Chen, N. J. Sucher, J. Loscalzo, D. J. Singel, J. S. Stamler, Nature
1993, 364, 626.
[26] J. Zhang, V. L. Dawson, T. M. Dawson, S. H. Snyder, Science
1994, 263, 687.
[27] J. S. Beckman, M. Carson, C. D. Smith, W. H. Koppenol, Nature
1993, 364, 584.
[28] M. Wiedau-Pazos, J. J. Goto, S. Rabizadeh, E. B. Gralla, J. A.
Roe, M. K. Lee, J. S. Valentine, D. E. Bredsen, Science 1996, 271,
515.
[29] a) W. R. Markesbery, Free Radical Biol. Med. 1997, 23, 134; b) B.
Alvarez, G. Ferrer-Sueta, B. A. Freeman, R. Radi, J. Biol. Chem.
1999, 274, 842.
[16] N-methylation: 2-Mn and 4-Mn were dissolved in hot THF and
excess methyl iodide was added to the solutions, which were then
left at 408C until complete precipitation. The solid material was
collected by centrifugation and washed with THF and diethyl
ether until the solvent was colorless. [(2’)Mn(X)]2+: UV/Vis
(phosphate buffer solution; pH 6.8): lmax (e10ꢁ3) = 372 (22.4),
394 (19.7), 460 (56.4), 558 (5.7). MS (MALDI-TOF LD+) m/z
(%): 879 (10) [M+], 864(100) [ Mꢁ15], 849 (50) [Mꢁ30]. 19F
(MeOD) (188 MHz): d = ꢁ137.87 (br s, 4F), ꢁ151.28 (s, 2F),
ꢁ160.04ppm (s, 4F). [( 4’)Mn]2+
: UV/Vis (KH2PO4 0.3m,
pH 6.8): lmax (e10ꢁ3) = 488 (49.4), 556 (8.4), 598 (10.3), 660
(15.9); MS (MALDI-TOF LD+): m/z (%): 700 (10) [M+], 685
(100) [Mꢁ15]. ESI (CH3CN/H2O, 70/30) m/z (%): 685 (40)
[M+ꢁ15], 370.5 (85) [M++CH3CN]/2, 350.0 (100) [M+]/2. 19F
(MeOD) (188 MHz): d = ꢁ129.08 (br s, 2F), ꢁ152.51 (s, 1F),
ꢁ158.68 ppm (s, 2F). The product was crystallized by slow
diffusion of an n-hexane/diethyl ether mixture into concentrated
methanol solution.
[30] a) J. S. Beckman, T. W. Beckman, J. Chen, P. A. Marshall, B. A.
Freeman, Proc. Natl. Acad. Sci. USA 1990, 87, 1620; b) J. S.
Beckman, Chem. Res. Toxicol. 1996, 9, 836; c) G. Merenyi, J.
Lind, S. Goldstein, G. Szapski, Chem. Res. Toxicol. 1998, 11, 712.
[31] A. Mahammed, Z. Gross, Angew. Chem. 2006, 118, 6694; Angew.
Chem. Int. Ed. 2006, 45, 6544.
[32] a) J. Lee, J. A. Hunt, J. T. Groves, Bioorg. Med. Chem. Lett. 1997,
7, 2913; b) J. Lee, J. A. Hunt, J. T. Groves, J. Am. Chem. Soc.
1998, 120, 6053.
[17] [(4’)Mn]2+ was crystallized as a methanol and water solvate.
Crystal data: 2(C37H22F5MnN6)2+·4Iꢁ·5CH4O·H2O, M = 2086.92,
¯
triclinic, space group P1, a = 16.1423(4), b = 16.5311(4), c =
16.6329(5) Š, a = 115.329(1), b = 90.631(1), g = 94.626(1)8, V=
3993.6(2) Š3, Z = 2, 1calcd = 1.735 gcmꢁ3, 45552 reflections mea-
sured, 15503 unique (Rint = 0.061, 2qmax = 52.08), final R = 0.069
(wR = 0.177) for 10085 reflections with I > 2s(I) and R = 0.112
(wR = 0.203) for all data. CCDC-634668 contains the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[33] J. A. Hunt, J. Lee, J. T. Groves, Chem. Biol. 1997, 4, 845.
[34] G. Ferrer-Sueta, C. Quijano, B. Alvarez, R. Radi, Methods
Enzymol. 2002, 349, 23.
[35] [(2’)Mn(X)]2+ and [(4’)Mn]2+ were dissolved in 10 mL of water
(aided by 1 mL of methanol for the latter), 1.5 g of freshly HCl-
regenerated ion-exchange resin (Dowex 1:8 chloride form) was
added and the vessels were slowly shaken over night. The resin
was filtrated and the solvent was lyophilized
[18] S. Prince, F. Kꢀrber, P. R. Cooke, J. R. Lindsay Smith, M. A.
Mazid, Acta Crystallogr. Sect. C 1993, 49, 1158.
[19] G. Zheng, Q. An, T. Wang, F. Wang, X. Cao, Chem. J. Chin. Uni.
(in Chinese) 1992, 13, 727.
[36] Peroxynitrite was prepared according to published proce-
dures,[33] and the experimental procedures were identical to
those reported in Ref. [31].
4324
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 4320 –4324