Cationic metal-corrole complexes: Design, synthesis, and properties of guanine-quadruplex stabilizers

Article abstract of DOI:10.1002/chem.200800835

A series of pyridinium and quaternary ammonium copper corroies has been designed and synthesized. All new compounds have been fully characterized by NMR spectroscopy, high-resolution mass spectrometry, UV/Vis spectrscopy, and elemental analysis. Biochemical studies have indicated that all of these corroie derivatives can stabilize G-quadruplex structures, with corroie 4 being the most effective according to the results of circular dichroism (CD) melting experiments, polymerase chain reaction (PCR) stop assays, and surface plasmon resonance (SPR) experiments. Moreover, both corroies 3 and 4 tend to induce the human telomeric sequence to form hybrid G-quadruplex structures, whereas corroies 8 and 9 are more inclined to induce the human telomeric sequence to form antiparallel G-quadruplex structures.

Full text of DOI:10.1002/chem.200800835

FULL PAPER
DOI: 10.1002/chem.200800835
Cationic Metal–Corrole Complexes: Design, Synthesis, and Properties of
Guanine-Quadruplex Stabilizers
Boqiao Fu, Dan Zhang, Xiaocheng Weng, Ming Zhang, Heng Ma, Yuzhi Ma, and
Xiang Zhou*[a]
Abstract: A series of pyridinium and
quaternary ammonium copper corroles
has been designed and synthesized. All
new compounds have been fully char-
acterized by NMR spectroscopy, high-
resolution mass spectrometry, UV/Vis
spectrscopy, and elemental analysis.
Biochemical studies have indicated
that all of these corrole derivatives can
stabilize G-quadruplex structures, with
corrole 4 beingthe most effective ac-
cordingto the results of circular di-
chroism (CD) meltingexperiments,
polymerase chain reaction (PCR) stop
assays, and surface plasmon resonance
(SPR) experiments. Moreover, both
corroles 3 and 4 tend to induce the
human telomeric sequence to form
hybrid G-quadruplex structures, where-
as corroles 8 and 9 are more inclined
to induce the human telomeric se-
quence to form antiparallel G-quadru-
plex structures.
Keywords: chelation
· corroles ·
DNA · guanine · quadruplexes
Introduction
tect them from deleterious processes duringreplication
steps. Further investigation has indicated that these repeats
may fold into intramolecular G-quadruplex structures.[4] The
enzyme telomerase, which mediates the maintenance of te-
lomeres, is overexpressed in tumor cells. When the 3’ end of
the telomere folds into a G-quaruplex, the activities of telo-
merase are inhibited, thus killingthe tumor cells without de-
stroyingthe normal cells. It has been reported that telo-
meres can form G-quadruplex structures in the presence of
Na+ or K+ ions.[5]
Corroles are tetrapyrrolic macrocycles, akin to porphyrins,
but with one less carbon atom in their outer periphery and
one additional NH proton in their inner core. The two types
of macrocycles have the same 18 p-electron conjugated
system.[1] It has been reported that the NH proton in the
inner core of corroles is strongly acidic, whereas the NH
proton in the porphyrins is weakly basic.[2] Corroles are tri-
ACHTREUNG
A
Thus, devisingways of stabilizingG-quadruplex structures
has become a key strategy in the design of antitumor agents.
Recently, many groups have synthesized series of G-quadru-
plex stabilizers, which may block the extension of telomeres
by telomerase, thereby conferingantitumor activities with
potential medicinal applications.[6] The porphyrin family was
one of the first that was identified as G-quadruplex stabiliz-
ers. Cationic porphyrins, such as TMPyP4, display good in-
hibitory activities toward telomerase.[6e,f] Gross and co-work-
ers were the first to synthesize cationic corroles and find
that they can interact with DNA,[7] and we reported that
some cationic corroles are capable of stabilizingG-quadru-
lize high-valent metals, whereas porphyrins cannot because
the strong s donation, which in turn greatly increases the
energy of the d orbitals of the metal, is a more dominant
effect than the high energy of the corrole p system relative
to that of the porphyrins.[3]
Telomeres consist of tandem repeats of guanine-rich
TTAGGG sequences at the terminal of all chromosomes.
These sequences can cap the ends of chromosomes and pro-
[a] B. Fu,+ D. Zhang,+ Dr. X. Weng, M. Zhang, H. Ma, Y. Ma,
Prof. X. Zhou
[8]
College of Chemistry and Molecular Sciences
Wuhan University, Hubei
plex structures and inhibitingthe activity of telomerase.
To study the structure–activity relationships of corroles,
we attached pyridinium, piperidine, dimethylamino, and
amide moieties to the corrole skeleton because these func-
tional groups have shown a good affinity for DNA.[9–16]
Metals also play important roles in biological systems, and
metal complexes can influence the activities of telomer-
Wuhan 430072 (China)
Fax : (+86)27-8733-6380
E-mail: xzhou@whu.edu.cn
[+] Both authors contributed equally to this study.
Supportinginformation for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200800835.
Chem. Eur. J. 2008, 14, 9431 – 9441
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
9431

ase.[17] Some studies have reported special chemical proper-
ties of metallated corroles. Metal ions with high-oxidation
states in the core of the corrole may strengthen the interac-
tion between the corroles and G-quadruplexes. X-ray crys-
tallographic studies of corrolatocopper and -manganese
complexes show that they are typically saddle-type, which
are very similar to a plane.[7,18] These structures make cat-
ionic corroles prone to matchingthe G-quartet, which could
strengthen the interaction of the molecular ligand and G-
quadruplex. In view of these points, some metal ions with
high-oxidation states, such as copper(II) and manganese(II),
were inserted into corroles to form trivalent complexes.
However, there are few reports on the interaction be-
tween the metal–corrole complexes and G-quadruplexes.
The strategy of synthesis of these metal-chelating corroles,
in which the metal centers were inserted into the corroles
and then different groups were attached, was based on the
stronger acidity of the inner hydrogen atoms. This property
makes these corroles very active under basic conditions and
the inner nitrogen atom is easily alkylated, thus leading to
reactions that produce more by-products. It is very difficult
to use an aldehyde with a basic group to synthesize the cor-
roles directly, partially as a result of the ready protonation
of the basic group, thus making the aldehyde prone to react-
ingwith pyrrole to form porphyrins instead of corroles.
The corrole derivatives were synthesized in metal-assisted
reactions. For the synthesis of corroles 8, 9, and 12–15, the
key step was the alkylation of phenol in the presence of a
[7,18]
[20]
Consideringthe specific structures of metalated corroles,
strongbase.
To avoid alkylation of the inner nitrogen
we anticipated these compounds would show some interest-
ingproperties through interaction with G-quadruplex DNA.
Thus, a metal–corrole complex may possibly play an impor-
tant role in replacingone of the metal ions that is normally
coordinated in and at the ends of the central ion channel of
the quadruplex.[19]
Herein, we report some new findings concerning the syn-
thesis and biochemical studies of cationic corrole deriva-
tives. Biochemical studies include a polymerase chain reac-
tion (PCR) stop assay, circular dichroism (CD) studies, CD
meltingassays, surface plasmon resonance (SPR) experi-
ments, and UV titration experiments. The results have indi-
cated that the synthesized cationic corrole derivatives are
capable of stabilizingG-quadruplex structures.
atom of the corrole, a metal-protected core is necessary.
Furthermore, it has been reported that manganese and
copper derivatives are good at stabilizing corroles and have
good biological activities.[21] Therefore, manganese and
copper were chosen as our studied metals.
In our initial synthesis of cationic corroles, we mixed 4-
pyridine corrole with metal acetates in pyridine and ob-
tained complexes 1 and 2 (Scheme 1). Methylation of cor-
roles 1 and 2 with methyl iodide in THF/CH3OH then af-
forded the water-soluble pyridinium corroles 3 and 4 in
good yield. The copper or manganese centers in the com-
plexes became CuIII or MnIII, which has been detailed in
many reports.[22] The free 5,10,15-tris(4-aminophenyl)corrole
and 5,10,15-tris(4-pyridyl)corrole are neutral, whereas these
two corroles are basic on metal chelation because the
strongly acidic hydrogen atoms in the inner core are re-
placed by the metal centers. The structures of the free cor-
roles have the stronger coplane character rather than the
saddle-type structure of the metal–corroles.
Results and Discussion
Synthesis and physicochemical characterization: We recently
reported that some water-soluble free cationic corrole deriv-
atives showed good inhibitory activity toward telomerase
and induced the formation of a G-quadruplex structure.[8]
These compounds also showed good selectivities for G-
quadruplex DNA over double-stranded DNA (dsDNA). We
have now designed and synthesized more functionalized cor-
role derivatives to investigate their structure–activity rela-
tionships further.
Quaternary ammonium corroles were synthesized accord-
ingto the procedures shown in Scheme 2. Using5,10,15-
tris(4-hydroxyphenyl)corrole as
a startingmaterial, we
formed the correspondingcopper complex by treatment
with copper(II) acetate in pyridine at room temperature.
Further alkylation of the phenol groups with alkyl chlorides
in the presence of K2CO3 furnished corroles 6 and 7. Finally,
methylation of 6 and 7 with methyl iodide gave the quater-
nary ammonium corroles 8 and 9 in good yield. The low
Scheme 1. i) Cu(Ac)2, pyridine, 30 min, RT; Mn(Ac)2, pyridine, 30 min, 608C; ii) THF/CH3OH, 408C, 12 h.
9432
www.chemeurj.org
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9431 – 9441

Stabilization of Guanine Quadruplexes by Metal–Corrole Complexes
FULL PAPER
yields of 6 and 7 might be at-
tributed to the high-oxidation
state of the copper center,
stronginductive effect of which
leads to low electron density at
the periphery. The electron
density on the oxygen atom of
the hydroxy group was thereby
significantly lowered, thus ren-
deringit only weakly nucleo-
philic.
Compounds 12–15 were syn-
thesized from 5,10,15-tris(4-
aminophenyl)corrole
(Scheme 3). Treatment with
copper(II) acetate afforded
5,10,15-tris(4-aminophenyl)cor-
rolate copper (10) in good
yield. The important intermedi-
ate 11 was prepared by mixing
10 and 4-(chloromethyl)benzoyl
chloride at room temperature.
Finally, the quaternary ammoni-
um copper corrole derivatives
12–15 were obtained in good
yield by treating 11 with tri-
Scheme 2. i) Cu(AC)2, pyridine, RT, 30 min; ii) 1-(2-chloroethyl)piperidine hydrochloride or 3-chloro-N,N-di-
methylpropan-1-amine hydrochloride, K2CO3, 608C, 72 h; iii) CH2Cl2, CH3I, 408C.
Scheme 3. i) Cu(Ac)2, pyridine, 30 min, RT; ii) 4-(chloromethyl)benzoyl chloride, NEt3, THF, RT, 2 h; iii) THF, 33% trimethylamine, alcohol solution,
RT, 12 h; NEt3, reflux 13 h; pyridine, reflux 14 h; 2-(dimethylamino)ethanol, reflux 15 h.
Chem. Eur. J. 2008, 14, 9431 – 9441
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
9433

X. Zhou et al.
methylamine, triethylamine, and pyridine, respectively. All
new compounds were fully characterized by NMR spectros-
copy, high-resolution mass spectrometry, UV/Vis spectrosco-
py, and elemental analysis.
Stabilizing the G-quadruplex structure in the telomeric se-
quence and c-MYC region G-quadruplexes sequence by
PCR stop assay:[23,26] Nondenaturingpolyacrylamide egl
electrophoresis (PAGE) was used to ascertain whether the
corrole derivatives bonded to the test oligomer 21G (5’-
GGGTTAGGGTTAGGGTTAGGG-3’) and stabilized the
G-quadruplex structure.[24,25] The sequences of the tested oli-
gonucleotides (21G and 21Gmu) and the corresponding
complementary sequence (Rev 21G), which is a complemen-
tary oligonucleotide that partially hybridized to the last G
repeat of the test oligonucleotide used here, are presented
in Figure 1. Taking the result of the PCR stop assay of cor-
role 13 as an example, we can find that the inhibitory effect
of 13 is enhanced dramatically as the concentration is in-
creased from 1.0 to 15.0 mm (see gel (a) shown in Figure 2a
with the correspondingdose–response curve). The
IC50 value, which indicates the concentration of 13 required
to achieve 50% inhibition of the reaction, was found to be
7.07 mm (the values shown in Table 1 were obtained at least
from three repeated experiments)
Figure 2. Effect of corrole 13 on the formation of the PCR stop assay
with G-quadruplex forming 21G oligomer (gel a) or on the parellel assay
with the corresponding mutated oligomer 21Gmu (gel b). Increasing con-
centrations of 13 were added to the G-quadruplex forming21G or
21Gmu oligomers. The assay used a buffer containing 10 mm K+ and
1.5 mm Mg2+ ions. The quantification of the final 30-bp double-stranded
PCR product is shown by the lighteness of fluorescence determined by
usingChemilmager 5500 (Alpha Innotech, San Leandro, CA) and are
displayed by dose–response curves.
Figure 1. a) The principle behind the PCR stop assay: The test oligomer
(21G or 21Gmu) was amplified with
a complementary oligomer
(Rev21G) overlappingthe last six bases TTAGGG. A final 30-base-pair
(bp) double-stranded PCR product was available through a successfully
carried out PCR process in the absence of corroles. On the addition of a
corrole, 21G could be induced into and stabilized as a G-quadruplex
structure. At this time, no PCR product will be obtained because of the
PCR inhibition or because Taq polymerase extension is inhibited. b) If
the two underlined bases GG in 21G were replaced by adenine (A), a
mutative sequence 21Gmu could not form a G-quadruplex structure in
the presence of G ligands.
A nonspecific PCR inhibition experiment usinga mutated
oligomer 21Gmu (5’-GGGTTA GAATTA GGGTTA GGG-
3’) instead of the oligomer 21G, which could not form the
G-quaduplex structure, was performed (Figure 1b). The gel
9434
www.chemeurj.org
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9431 – 9441

Stabilization of Guanine Quadruplexes by Metal–Corrole Complexes
FULL PAPER
Table 1. Effects of the corrole derivatives on telomeric quadruplexes and
C-Myc measured by the PCR stop assay.
ly, 4 also possessed the best selectivity of the two systems,
which can be attributed in a large degree to its special struc-
ture. We suggest that an electron-withdrawing group, such
as the pyridinium moiety in 3 and 4, decreased the electron
density of the corrole ring, thus leading to interaction with
the more electron-rich G-quaduplex DNA. Alternatively, a
trivalent manganese–corrole complex could strengthen the
interaction of the molecular ligand and G-quadruplex and
present a relatively higher affinity toward the G-quadruplex
because of its higher electron deficiency and favorable
shape.[28]
Compound
IC50 [mm]
Telo
21G
21Gmu
selectivity
3
4
8
3.51Æ2.14
2.37Æ1.01
7.70Æ0.40
12.35Æ1.74
10.87Æ0.14
7.07Æ0.47
5.99Æ2.27
6.30Æ3.30
7.09Æ1.70
6.46Æ1.92
10.95Æ2.95
14.03Æ1.62
>30.00
11.62Æ0.64
6.28Æ1.43
5.19Æ2.07
2.02
2.72
1.42
1.14
>2.76
1.64
9
12
13
14
15
1.05
0.82
Compound
IC50 [mm]
C-Myc
Pu27
Pu27mu
selectivity
Structure induction or transition of the G-quadruplex in the
presence of corrole derivatives tested by CD spectroscopy:
CD spectroscopy was used to determine the structure induc-
tion and structure transition of the G-quadruplex in the
presence of cationic corrole derivatives. It has been reported
3
4
8
2.74Æ0.40
1.52Æ0.31
15.34Æ1.97
9.59Æ2.28
7.62Æ2.57
10.31Æ2.51
5.39Æ0.78
4.88Æ0.40
4.19Æ2.68
3.23Æ1.70
4.52Æ0.44
9.67Æ0.27
4.44Æ1.72
7.43Æ0.92
5.12Æ0.41
3.48Æ1.33
1.53
2.13
0.29
1.01
0.58
0.72
0.95
0.71
9
12
13
14
15
that the human telomeric dCAHTRE(UGN T2AG3)4 sequence forms a typi-
cal antiparallel G-quadruplex structure in the presence of
Na+ ions and has a positive band near l=295 nm and a neg-
ative band at l=265 nm in CD spectra.[29] On the other
hand, in the presence of K+ ions, there is a small positive
band at l=265 nm, a negative band close to l=240 nm, and
a stronger positive peak at l=295 nm, which may be charac-
teristic of a hybrid of parallel/antiparallel G-quadruplex
structures.[30]
in Figure 2b with the corresponding dose–response curve
shows the inhibitory properties of 13 to a PCR process with
similar concentration gradients. Compared with the corre-
spondingconcentration of 13 in the gel in Figure 2a, it is
clear that whereas 13 totally stabilizes the G-quadruplex for-
mation of the 21G oligomer at 15 mm, it cannot totally inhib-
it the extension of Taq polymerase or the PCR process. The
IC50 value of this reaction between 13 and the 21Gmu oligo-
mer was 11.62 mM relative to the previous IC50 value of
7.07 mm; thus, we can conclude that corrrole 13 has a good
selectivity for the human telomeric sequence.
Figure 3 shows CD spectra of the structures obtained by
addingdifferent concentrations of corrole 4 to the human
telomeric sequence Telo24 in the absence of a high concen-
The results of the PCR stop assay of the eight cationic
corroles 3, 4, 8, 9, and 12–15 toward the human telomeric se-
quence 21G and mutated sequence 21Gmu are presented in
Table 1. Relative to other corrrole derivatives, corrole 4 dis-
played the highest selectivity of 2.72. This outcome clearly
indicates that corrole 4, which is modified with pyridine at
the meso position of the corrole plane and chelated with a
manganese ion, is better than other corroles in this assay be-
cause it can both stabilize a G-quadruplex at a lower con-
centration and has better selectivity.
Another sequence Pu27 (5’-TGGGGAGGGTGGG-
GAGGGTGGGGAAGG-3’), which existed in the NHE III1
element of the C-Myc gene promoter[26,27] and could form a
parallel G-quadruplex structure, was chosen to replace the
oligomer 21G to further demonstrate the conclusion pre-
sented above by utilizingthe same principle. Similarly, oligo-
mer Pu27mu (5’-TGGGGAGGGTGGAAAGGGTGGG-
GAAGG-3’) was chosen to substitute 21Gmu. Detailed re-
sults obtained for these corroles are summarized in Table 1;
all compounds were found to be active at the micromolar
range both in the Telo and C-Myc systems.
Figure 3. CD spectroscopic titration of the G-quadruplex (T2AG3)4 with
corrole 4 in the absence of 100 mm NaCl. Arrows from bottom to up indi-
cated increments in r values from 0 to 4. (r=corrole 4/ACTHREU(NG T2AG3)4).
tration of free univalent metal ions.[6b,31] The CD profile of
single-strand Telo24 (5’-TTAGGGTTAGGGTTAGGGT-
TAGGG-3’) shows a positive band at l=257 nm attributa-
ble to the absorbance of individual bases. On addingin-
creasingamounts of 4, there is a sharp decrease in the band
at l=257 nm accompanied by a sharp increase in the band
around l=295 nm, thus showingthe formation of hybrid of
parallel/antiparallel G-quadruplex structures.[6b,8]
Amongthe eihgt corroles, 4 was found to be the most
active in this assay, with an IC50 value of 1.52 mM in the C-
Myc system (IC50 =2.37 mm in the Telo system). Additional-
Chem. Eur. J. 2008, 14, 9431 – 9441
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
9435

X. Zhou et al.
Table 2. Values of Tm and DTm for the antiparallel G-quadruplex of the human telomeric sequence stabilized
by corrole derivatives.
From other CD spectra (see
Figure S1 in the Supporting In-
formation), we could conclude
Compound
control
37.1
3
4
8
9
12
13
14
15
that both
3 and 4 tend to
Tm [8C]
DTm [8C]
54.5
17.4
58.4
21.3
62.8
25.7
63.8
26.7
42.8
5.7
45.6
8.5
46.1
9.0
46.1
9.0
induce the human telomeric se-
quence to form hybrid G-quad-
ruplex structures, whereas cor-
roles 8 and 9 tend to induce the
human telomeric sequence to form antiparallel G-quadru-
plex structures. However, the other corroles, such as 12–15,
do not induce the same sequences in any G-quadruplex
structure formation. Therefore, this finding suggests that
pyridinium and piperidinium quaternary ammonium sub-
stituents at the meso positions might interact with the loops
and grooves of a G-quadruplex,[32] whereas phenyl amide
substituents at the meso positions do not have this ability to
interact with DNA.
4, 8, and 9 are higher than those of the other cationic cor-
roles. As explained before, four of them interact well with
the G-quadruplex, thus suggesting that 3, 4, 8, and 9 are
able to stabilize the G-quadruplex DNA well.
The DTm values of 8 and 9 are 25.7 and 26.78C, respec-
tively, which suggests that these corroles are more able to
stabilize the G-quadruplex structure. However, this outcome
seem to conflict with the results of the PCR stop assay,
which suggested that 3 and 4 are more able to stabilize the
G-quadruplex structure. We assume that these conflicting
results are a result of the different buffers used in the two
experimental systems. In the PCR stop assay, the samples
contained 10 mm K+ and 1.5 mm Mg2+ ions, which could
induce the Telo sequence to form a parallel/antiparallel G-
quadruplex structure.[6b] Under these conditions, 3 and 4
may favor this structure because 3 and 4 could induce a
hybrid of parallel/antiparallel G-quadruplexes, thus leading
to improved stabilization of the G-quadruplex. In the CD
meltingexperiment, the tested samples included 100 m m
Na+ ions that induced the Telo sequence to form an antipar-
allel G-quadruplex structure, which may be favorable for 8
and 9 because they could induce the Telo sequence to form
the same structure.
Thermodynamic stability of G-quadruplex structures in the
presence of corrole derivatives: The bindingcapacities of
the corrole derivatives to the G-quadruplex structure, thus
reflectingthe degree of stabilization, were studied by CD
[8]
meltingexperiments.
CD spectroscopy was also used to
measure the thermodynamic stability profile of the G4C oli-
gomer (sequence 5’-CATGGTGGTTTGGGTTAGGGT-
TAGGGTTAGGGTTACCAC-3’) incubated with the cor-
role derivatives in the presence of Na+ ions. In these experi-
ments, thermal CD transitions were monitored at l=
295 nm, which is diagnostic of the antiparallel G-quadruplex
structure in the human telomeric sequence.[8]
Selectivities of corroles comparing the G-quadruplex to
double-stranded DNA (dsDNA) tested by SPR: The bind-
ingabilities and constants of corroles 3, 4, 8, and 9 to the G-
quadruplex were quantitatively determined by usingSPR
(Table 3). The experimental data are shown in Table 3 (see
Figure S3 in the Supporting Information for sensorgrams),
and we can take the ratio of KA(G4)/KBACHTRE(UGN dsDNA) to evalu-
ate the selectivities of the corroles.[6b,28,31] We found that 3
and 4 had better selectivities than other corroles by compar-
ingthe four ratios of KA(G4)/KB dsDNA). Corrole 4 was
the most selective because its bindingconstant to the G4 se-
quence was about 64 times greater than binding to the
dsDNA sequence. Relative to the 4-pyridinium corrole
ligand in the absence of copper or manganese, the selectivi-
ties of 3 or 4 to the G-quadruplex were sligtly decreased.[8]
It is probable that the formation of copper or manganese
corroles did not result in suitable shapes to interact with the
G-quadruplex DNA compared with the 4-pyridinium cor-
role. Further SRA experiments on this series of compounds
are under investigation.
Figure 4. CD melting curve of 12.5 mm DNA oligonucleotide (G4C) and
corrole 4 in 100 mm NaCl at pH 7.4, 10 mm Tris–HCl, and 1 mm EDTA
buffer. The CD spectra were monitored at l=295 nm. The concentration
ratio of corrole/DNA (r) is 5:1.
Figure 4 shows the melting curve of 4 in the presence of
100 mm Na+ ions, 10 mm 2-amino-2-(hydroxymethyl)-1,3-
propanediol hydrochloride (Tris—HCl), and 1 mm ethylene-
diaminetetraacetic acid (EDTA) at pH 7.4. The concentra-
tion of 4 was five times greater than the G4C oligomer.
Table 2 displays the DTm values of the eight corroles (R=5)
(see Figure S2 in the Supporting Information for the melting
curves). All the corrole derivatives possess the ability to sta-
bilize the G-quadruplex structure in the presence of Na+
ions. Amongthe eight cationic corroles, the DTm values of 3,
Binding modes of corroles/G-quadruplex and corroles/calf
thymus DNA (CT-DNA) by UV/Vis titration: To find the
bindingmodes of the corroles to G-quadruplex DNA and
dsDNA, we measured their maximium absorbance in the
9436
www.chemeurj.org
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9431 – 9441

Stabilization of Guanine Quadruplexes by Metal–Corrole Complexes
FULL PAPER
Table 3. Equilibrium bindingconstants determined by SPR and rate constants for quadruplex interactions, de-
termined from BIAcore analysis of 1:1 bindingwith mass-transfer fitting.
Conclusion
[a]
[a]
[b]
[c]
Compound
DNA
ka
kd
KA
KD
KA(G4)/
AA(dsDNA)
A series of new cationic copper
and manganese corroles has
been designed and synthesized
by usingdifferent strateiegs.
Biochemical studies indicate
that all of the eight cationic cor-
roles 3, 4, 8, 9, and 12–15 can
stablize G-quadruplex struc-
tures. Furthermore, corrole de-
rivatives 3 and 4 can induce the
hybrid formation of a parallel/
K CHTREUNG
G4
dsDNA
G4
dsDNA
G4
dsDNA
G4
2.16”104
4.13”102
4.34”103
7.75”101
3.70”102
3.05”102
1.80”102
1.61”102
2.26”10À3
2.14”10À3
2.24”10À3
2.54”10À3
1.33”10À3
1.86”10À3
6.24”10À3
6.84”10À3
9.53”106
1.93”105
1.94”106
3.05”104
2.78”105
1.64”105
2.88”104
2.35”104
1.05”10À7
5.18”10À6
5.17”10À7
3.28”10À5
3.60”10À6
6.10”10À6
3.47”10À5
4.26”10À5
3
4
8
9
49.4
63.6
1.7
1.2
dsDNA
[a] Kinetic constants (ka and kd are the correspondingassociation and dissociation rate constants) determined
from BIAcore analysis by using 1:1 Langmuir global fitting of 240-s association and 240 s disassociation.
[b] Determined from ka/kd. [c] Determined from kd/ka.
presence of G4 DNA or duplex DNA by UV/Vis titration
(see Figures S4 and S5 in the Supporting Information). We
found that the absorbance was red shifted by less than 8 nm
and that the hypochromicity was between 10 and 35% for
the interaction between the corroles and G4 DNA
(Table 4). It is suggested that corroles 8, 9, and 12–15 have
Table 5. Effect of CT-DNA on the absorption maximium of the Soret
band for corroles.
Compound
lmax [nm]
CT-DNA
lmax [nm]
Red shift [nm]
H [%][a]
3
4
8
435
500
423
421
419
424
420
420
462
505
424
424
425
429
426
426
37
22
39
23
12
23
15
11
27
5
1
3
6
5
6
6
9
12
13
14
15
Table 4. Effect of G4A on the absorption maximum of the Soret band
for corroles.
Compound
lmax [nm]
G4A
Red shift
[nm]
lmax [nm]
H [%][a]
[a] H=[Abs freeÀAbs corrole (bound)]/Abs free; Abs free is the absorb-
ance of each corrole at the concentration of 2.5 mm without any DNA;
Abs corrole (bound) is the absorbance of the fully bound corroles as
measured at the Soret maxium of free corrole.
3
4
8
437
500
423
409
417
424
419
419
453
520
436
436
421
427
423
423
37
37
26
29
13
22
16
17
16
20
13
27
4
3
4
4
9
12
13
14
15
antiparallel G-quadruplex structure of the human telomeric
sequence and corroles 8 and 9 could induce the formation of
an antiparallel G-quadruplex structure, as examined by CD
spectroscopy.
[a] H=[Abs freeÀAbs corrole (bound)]/Abs free; Abs free is the absorb-
ance of each corrole at the concentration of 2.5 mm without any DNA;
Abs corrole (bound) is the absorbance of the fully bound corroles as
measured at the Soret maxium of free corrole.
Amongthese corroles, manganese–corrole 4 with pyridini-
um substituents at the meso position of the corrole unit
exerts a better stabilization effect toward the G-quadruplex
structure. The strongelectron-withdrawinggroups (i.e., pyri-
dinium substituents on the periphery of corrole and the tri-
valent manganese) decrease the electron density of the cor-
role ring, thus allowing ready interaction with the electron-
rich guanine G-quadruplex. SPR data suggested that cor-
roles 3 and 4 showed good selectivity between dsDNA and
G-quadruplexes.
Interestingly, these corroles could also stabilize the G-
quadruplex structure of Pu27 of NHE III1, as determined by
PCR stop assays. The experimental results suggested that
corrole 4 had the best ability to stabilize the G-quadruplex
structure of Pu27 of NHE III1 amongall the tested corroles.
This series of corroles has potential in antitumor applica-
tions.
different bindingmodes compared with 3 and 4,[33] because
of the increased bulkiness of the G4 DNA relative to cor-
roles.
For the interaction between corroles and dsDNA (i.e.,
CT-DNA), we found that all the corroles, except 3, might
have the same bindingmode with CT-DNA because the ab-
sorbance was red shifted by less than 8 nm and the hypo-
chromicity was between 10 and 35% in the UV/Vis spectra
(see Table 5 and Figure S5 in the Supporting Informa-
tion).[34] We explained that corrole 3 with pyridinium arms
has a planar structure, as determined by X-ray crystallo-
graphic analysis,[22b] and can intercalate into CT-DNA. Cor-
role 4 has axis coordination to the manganese complex, and
this steric effect prohibits it from intercalatingwith CT-
DNA.[7] Other corroles with steric side arm groups might
not interact with CT-DNA in the same way as 4.
Chem. Eur. J. 2008, 14, 9431 – 9441
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
9437

X. Zhou et al.
5,10,15-Tris(N-methyl-4-pyridyl)corrolatomanganeseACHTREU(NG III) (4): Compound
Experimental Section
4 was obtained by usingthe same procedures described for 3. The start-
ingmaterials was 5,10,15-tris(4-pyridyl)corrolatomanagnese (26 m, g
0.045 mmol) and 4 was obtained in 73% yield (33 mg, 0.033 mmol).
1H NMR (300 MHz, [D7]DMF/CD3OD (1:1), 258C, TMS): d=4.56 (s,
9H, CH3), 8.00–8.40 (m, 9H; b-pyrrole H and pyridine H), 8.95–9.20 ppm
(m, 11H; b-pyrrole H and pyridine H); UV/Vis (CH3OH): lmax (e”
10À3 LmolÀ1 cmÀ1)=427 (16.4), 506 (35.76), 665 (10.52) nm; HRMS
(ESI): m/z: calcd for (C37H29I3MnN7À3I)/3: 208.7288; found: 208.7285
[M+À3I]/3; elemental analysis (%) calcd for C37H29I3MnN7·3H2O: C
41.87, H 3.32, N 9.24; found: C 41.77, H 3.14, N 8.82.
Physical methods: The NMR spectra were recorded on a Varian Mercury
VX-300 MHz spectrometer operatingat 300 MHz for 1H NMR. Chemical
1
shifts in the H NMR spectra are reported in ppm relative to the residual
hydrogen atoms in the deuterated solvents: d=2.50 and 7.25 ppm for
[D6]DMSO and CDCl3, respectively. The couplingconstants J are report-
ed in Hz. Mass-spectroscopic analysis was performed on a VG ZAB-HS
instrument. UV/Vis spectra were collected with BeijingPurkinje General
Instrument TU-1900 spectrophotometer. The elemental analysis was per-
formed on a Elementary Vario ElIII CHNSO machine.
5,10,15-TrisACHTREU(NG 4-phenol)corrolatocopperAHCRTE(UNG III) (5): Compound 5 was ob-
Materials: N,N-Dimethylformamide (DMF) and pyrrole were distilled
over CaH2 before use. Absolute anhydrous THF was distilled over Na. 1-
(2-Chloroethyl)piperidine hydrochloride was obtained from Alfa Aesar.
The other reagents and solvents were bought from Sinopharm Chemical
Reagent Co., Ltd.
tained by usingthe same procedure described for 1. The startingmateri-
als were 5,10,15-tris(4-hydroxylphenyl)corrole (460 mg, 0.8 mmol) and
cupric acetate hydrate (4.60 g, 23 mmol). The resulting solid was isolated
by chromatography on silical gel (200—300 mesh) and eluted with ethyl
acetate/cyclohexane (1:2). The first brown fraction was collected. After
evaporatation of the solvent, the crude product was recrystallized from
acetone/petroleum ether (1:5). Compound 5 was obtained as a brown
Synthetic methods: The synthetic details of the preparation 5,10,15-tris(4-
hydroxylphenyl)corrole, 5,10,15-tris(4-aminophenyl)corrole, and 5,10,15-
tris(4-pyridyl)corrole are provided in our previous report.[8]
1
solid in 59% yield (300 mg, 0.47 mmol). H NMR (300 MHz, [D6]DMSO,
258C, TMS): d=6.84–6.92 (m, 6H; Ph H), 7.22 (d, 3J
A
À
5,10,15-Tris(4-pyridyl)corrolatocopperACHTRE(UGN III) (1): A solution of 5,10,15-
4H; Ph H), 7.38 (d, 3J
A
À
À
tris(4-pyridyl)corrole (95 mg, 0.18 mmol) and cupric acetate hydrate
(538 mg, 5.4 mmol) in pyridine (20 mL) was stirred at room temperature
for 30 min. The solvent was evaporated and the crude solid was purified
by chromatography on alumina (200–300 mesh) by elution with
CH3COOEt/CH3OH=10:1 and 5:1 to collect the first yellow–green frac-
tion. After removal of the solvents, the solid was recrystallized from
THF/petroleum ether (1:5). Compound 1 was obtained as a black solid in
43% yield (46 mg, 0.078 mmol). 1H NMR (300 MHz, CDCl3, 258C,
role H), 7.83 (s, 2H; b-pyrrole H), 9.93 (s, 1H; OH), 9.99 ppm (s, 2H;
OH); 13C NMR (75 MHz, [D6]DMSO, 258C, TMS): d=115.67, 116.01,
121.47, 126.99, 129.49, 130.28, 132.20, 131.11, 143.93, 145.08, 150.34,
152.54,
158.58,
158.95 ppm;
UV/Vis
(acetone):
lmax
(e”
10À3 LmolÀ1 cmÀ1)=431 (104.7), 637 (6.2), 547 (12.8) nm; HRMS (ESI):
m/z: calcd for C37H23CuN4O3: 634.1066; found: 634.1052; elemental anal-
ysis (%) calcd for C37H23CuN4O3·H2O: C 68.04, H 3.86, N 8.58; found: C
67.61, H 3.82, N 8.38.
TMS): d=7.13 (d, 3J
ACHTREUNG(H,H)=3.6 Hz, 4H; pyridine H), 7.27 (s, 2H; pyri-
5,10,15-Tris(4-{[1-ACHTRENGU(2-ethyl)piperidine]oxy}phenyl)corrolatocopperACHTREUNG(III)
dine H), 7.47–7.56 (m, 7H; b-pyrrole H and pyridine H), 7.90 (s, 2H; b-
pyrrole H), 8.74 ppm (s, 5H; b-pyrrole H); 13C NMR (75 MHz, CDCl3,
258C, TMS): d=122.78, 125.21, 130.27, 133.03, 144.87, 146.42, 150,
150.6 ppm; UV/Vis (THF): lmax (e”10À3 LmolÀ1 cmÀ1)=407 (106.5), 542
(9.5), 617 (6.1) nm; HRMS (ESI): m/z: calcd for C34H20N7Cu+H:
590.1149; found: 590.1145 [M+H+]; elemental analysis (%) calcd for
C34H20N7Cu·H2O: C 67.15, H 3.65, N 16.12; found: C 67.14, H 3.41, N
16.16.
(6): A suspension of 5 (300 mg, 0.47 mmol), 1-(2-chloroethyl)piperidine
hydrochloride (533 mg, 2.84 mmol), and K2CO3 (685 mg, 4.96 mmol) in
dry DMF (50 mL) was stirred under N2 and heated by usingan oil bath
at 608C for 72 h. The solvent was evaporated under reduced pressure.
The residue was dissolved in dichloromethane (300 mL) and washed with
water. The organic layer was evaporated under reduced pressure. The re-
sultingsolid was isolated by chromatorgaphy on alumina (200—
300 mesh) with ethyl acetate/methanol (100:1). The first fraction was col-
lected and the solvent was evaporated. Further purification was carried
out by recrystallization from methanol/dichloromethane (5:1). Com-
pound 6 was obtained as a red–brown solid in 27% yield (123 mg,
0.127 mmol). 1H NMR (300 MHz, CDCl3, 258C, TMS): d=1.42 (s, 6H;
5,10,15-Tris(4-pyridyl)corrolatomanganeseACHTREU(NG III) (2): A solution of 5,10,15-
tris(4-pyridyl)corrole (100 mg, 0.19 mmol) and manganese acetate tetra-
hydrate (463 mg, 1.9 mmol) in pyridine (20 mL) was heated by using an
oil bath at 608C for 30 min. The solvent was evaporated and the resulting
solid was isolated by chromatography on alumina (200–300 mesh) by gra-
dient elution with CH3COOEt/CH3OH (10:1, 5:1, 2:1) to collect the first
yellow–green fraction. After evaporation of the solvents, the crude prod-
uct was recrystallized from THF/petroleum ether (1:5). Compound 2 was
obtained as a green solid in 31% yield (34 mg, 0.059 mmol). 1H NMR
(300 MHz, [D6]DMSO/[D4]CH3OH (1:1), 258C, TMS): d=8.87 (brs, 5H;
b-pyrrole H), 7.00–7.80 ppm (brs, 15H; b-pyrrole H and pyridine H);
UV/Vis (THF): lmax (e”10À3 LmolÀ1 cmÀ1)=403 (55.4), 438 (63.95), 494
(44.5), 554 (12.55), 583 (15.55), 648 (21.25) nm; HRMS (ESI): m/z: calcd
for C34H20N7Mn+H: 582.1233; found: 582.1234 [M+H+] (100%); ele-
mental analysis (%) calcd for C34H20N7Mn·H2O: C 68.91, H 3.70, N
16.35; found: C 68.95, H 3.91, N 16.59.
piperidine H), 1.59 (s, 12H; piperidine H), 2.50 (s, 12H; piperidine H),
3
2.79 (s, 6H; NCH2), 4.15 (d, J
A
6H; Ph-H), 7.29 (d, 3J(H,H)=7.5 Hz; 4H, Ph-H), 7.57 (d, 3J
ACTHERUGN ACHTREUNG
7.5 Hz, 2H; Ph-H), 7.63–7.68 (m, 6H; b-pyrrole H), 7.79 ppm (s, 2H; b-
pyrrole H); 13C NMR (75 MHz, CDCl3, 258C, TMS): d=24.43, 26.19,
55.32, 58.19, 66.29, 114.32, 114.54, 121.31, 129.52, 131.78, 132.34, 133.02,
144.29, 145.54, 150.77, 159.29, 159.53 ppm; UV/Vis (CH2Cl2): lmax (e”
10À3 LmolÀ1 cmÀ1)=435 (99), 548 (10.4), 627 (5) nm; HRMS (ESI): m/z:
calcd for C58H62N7CuO3 +H+: 968.4283; found: 968.4276 [M+H+]; ele-
mental analysis (%) calcd for C58H62N7CuO3·H2O: C 70.60, H 6.54, N
9.94; found: C71.05, H 6.40, N 10.15.
5,10,15-Tris{4-[(3-dimethylaminopropionylamino)oxy]phenyl}corrolato-
ACHRTEcGNU opperACHRTE(UGN III) (7): A suspension of 5 (91 mg, 0.47 mmol), 3-chloro-N,N-di-
5,10,15-Tris(N-methyl-4-pyridyl)corrolatocopperACHTRE(UGN III) (3): A solution of
5,10,15-tris(4-pyridyl)corrolatocopper (45 mg, 0.076 mmol) and CH3I
(10 mL, 0.16 mmol) in a mixture of acetone (20 mL) and methanol
(2 mL) was heated at 408C overnight. The solvent was evaporated and
the crude product was recystallized from methanol/petroleum ether (1:2).
methylpropan-1-amine hydrochloride (136 mg, 0.86 mmol), and K2CO3
(208 mg, 1.5 mmol) in dry DMF (50 mL) was stirred under N2 and heated
by usingoil bath at 60 8C for 72 h. The solvent was evaporated under re-
duced pressure. The residue was dissolved in dichloromethane (300 mL)
and washed with water. The organic layer was evaporated under reduced
pressure. The resultingsolid was isolated by chromatography on alumina
(200–300 mesh) with ethyl acetate/methanol (50:3). The first fraction was
collected and the solvent was evaporated. The crude product was recrys-
tallized from petroleum ether. Compound 7 was obtained as a red–brown
solid in 20% yield (26 mg, 0.029 mmol). 1H NMR (300 MHz, CDCl3,
Compound
3 was obtained as a black solid in 52% yield (43 mg,
0.042 mmol). 1H NMR (300 MHz, [D6]DMSO, 258C, TMS): d=4.52 (s,
9H; CH3), 9.14 ppm (brs, 20H; b-pyrrole H and pyridine H); UV/Vis
(CH3OH): lmax (e”10À3 LmolÀ1 cmÀ1)=427 (16.7), 506 (35.76), 665
(10.52) nm; HRMS (ESI): m/z: calcd for (C37H29I3CuN7À3I)/3: 211.3927;
found: 211.3929 [M+À3I]/3; elemental analysis (%) calcd for
C37H29I3CuN7·3H2O: C 41.53, H 3.30, N 9.16; found: C41.36, H3.52,
N9.16.
258C, TMS): d=2.04 (s, 6H; CH2), 2.30 (s, 18H; NCH3), 2.52 (s, 6H;
3
NCH2), 4.13 (s, 6H; OCH2), 6.96–7.03 (m, 6H; Ph-H), 7.35 (d, J
A
9438
www.chemeurj.org
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9431 – 9441

Stabilization of Guanine Quadruplexes by Metal–Corrole Complexes
FULL PAPER
6.6 Hz, 4H; Ph-H), 7.64–7.75 (m, 8H; b-pyrrole H and pyridine H),
7.87 ppm (s, 2H; b-pyrrole H); 13C NMR (75 MHz, CDCl3, 258C, TMS):
d=27.77, 45.73, 56.65, 66.54, 114.26, 114.49, 121.29, 127.00, 129.46,
131.80, 133.02, 144.29, 145.57, 150.77, 159.72 ppm; UV/Vis (CH2Cl2): lmax
(e”10À3 LmolÀ1 cmÀ1)=434 (79.75), 541 (7.75), 636 (2.4) nm; HRMS
(ESI): m/z: calcd for C52H26CuN7O3 +H: 890.3813; found: 890.3825
[M+H]+; elemental analysis (%) calcd for C52H26CuN7O3·H2O: C 68.74,
H 6.43, N 10.79; found: C 68.79, H 6.32, N 10.87.
by chromatography on alumina (200—300 mesh) with THF/methanol
(200:1). The first fraction was collected. After evaporation of the sol-
vents, the solid was recrystallized from THF/petroleum ether. Compound
11 was obtained as a brown solid in 74% yield (76 mg, 0.07 mmol).
1H NMR (300 MHz, [D6]DMSO, 258C, TMS): d=4.85 (s, 6H; CH2), 7.33
(s, 4H; Ph), 7.40 (s, 2H; Ph), 7.59–7.77 (m, 13H; Ph-H and b-pyrrole H),
7.98–8.08 (m, 13H; Ph-H and b-pyrrole H), 10.56 ppm (d, 3J (H,H)=
6.6 Hz, 3H; NH); 13C NMR (75 MHz, [D6]DMSO, 258C, TMS): d=
46.12, 120.43, 120.68, 128.85, 129.45, 131.37, 135.45, 140.70, 141.83,
166.01 ppm; UV/Vis (THF): lmax (e”10À3 LmolÀ1 cmÀ1)=421 (232.2), 541
(25.2) nm; HRMS (ESI): m/z: calcd for C61H41Cl3N7CuO3: 327.4189;
found: 327.4159 [M+À3Cl]/3; elemental analysis (%) calcd for
C61H41Cl3N7CuO3: C 67.22, H 3.79, N 9.00; found: C 67.51, H 3.56, N
8.99.
5,10,15-Tris(4-{[1-ACHTREUNG(2-ethyl)-N-methyl-piperidine]oxy}phenyl)corrolato-
A
ACHTREUNG
0.16 mmol) in dichloromethane (30 mL) was stirred under N2 and heated
by usingan oil bath at 40 8C for 24 h. The solvent was evaporated and
the resultingsolid was washed with CH 2Cl2 and Et2O in turn. Compound
8 was obtained as a brown solid in 82% yield (59 mg, 0.04 mmol).
1H NMR (300 MHz, [D6]DMSO, 258C, TMS): d=1.60 (s, 6H; piperidine
5,10,15-Tris[4-(4-N-trimethylbenzyl)amidephenyl]corrolatocopperACHTREUNG(III)
H), 1.89 (s, 12H; piperidine H), 3.21 (s, 9H; NCH3), 3.51 (d, 3J
A
(12): Trimethylamine in alcohol solution (33%, 10 mL) was added to the
solution of 11 (35 mg, 0.032 mmol) in THF (10 mL). The reaction mixture
was stirred overnight at room temperature. The solvent was evaporated
under reduced pressure and the crude product was recrystallized from
methanol/petroleum ether. Compound 12 was obtained as a black solid
in 98% (40 mg, 0.032 mmol). 1H NMR (300 MHz, [D6]DMSO, 258C,
TMS): d=3.10 (s, 27H; NCH3), 4.67 (s, 6H; CH2), 7.77 (s, 16H; Ph-H
and b-pyrrole H), 8.06 (s, 8H; Ph-H and b-pyrrole H), 8.19 (d, 3J
(H,H)=6.0 Hz, 8H; Ph-H and b-pyrrole H), 10.79 ppm (s, 3H; NH);
13C NMR (75 MHz, [D6]DMSO, 258C, TMS): d=52.65, 67.75, 120.90,
129.02, 132.42, 133.58, 136.98, 165.74 ppm; UV/Vis (CH3OH): lmax (e”
1.5 Hz, 12H; piperidine H), 3.91 (s, 6H; NCH2), 4.61(s, 6H; OCH2),
7.18–7.24 (m, 6H; Ph-H), 7.31(s, 2H; Ph-H), 7.41 (s, 2H; Ph-H), 7.66 (s,
3
2H; Ph-H and b-pyrrole H), 7.76 (d, J
A
8.16 ppm (s, 2H; b-pyrrole H); 13C NMR (75 MHz, [D6]DMSO, 258C,
TMS): d=19.27, 20.47, 47.87, 60.83, 61.22, 114.57, 114.77, 122.12, 131.83,
133.20, 143.08, 149.23, 157.91, 158.17 ppm ; UV/Vis (CH3OH): lmax (e”
10À3 LmolÀ1 cmÀ1)=424 (112.7), 530 (9.95), 627 (5) nm; HRMS (ESI): m/
z: calcd for C61H71CuN7O3I3; found: 337.4960 [M+À3I]/3 (100%),
569.6955 [M+À2I]/2 (14%), 337.4971 [M+À3I]/3, 569.6979 [M+À2I]/2;
elemental analysis (%) calcd for C61H71CuN7O3I3·H2O: C 51.87, H 5.21,
N 6.94; found: C 52.05, H 5.21, N 7.03.
10À3 LmolÀ1 cmÀ1)=417 (115.7), 539 (11) nm; HRMS
ACHTRE(UNG ESI): m/z: calcd
for [C70H68Cl3N10O3CuÀ3Cl]/3: 386.4923; found: 386.4923 [M+À3Cl]/3;
elemental analysis (%) calcd for C70H68Cl3N10O3Cu·3H2O: C 63.63, H
5.64, N 10.60; found: C 63.22, H 5.93, N10.14.
5,10,15-Tris{4-[(3-trimethylaminopropionylamino)oxy]phenyl}corrolato-
ACHTREUNGcopperACHTREUNG(III) (9): A solution of 7 (30 mg, 0.034 mmol) and CH3I (10 mL,
0.16 mmol) in dichloromethane (40 mL) was stirred under N2 and heated
by usingan oil bath at 40 8C for 24 h . The solvent was evaporated and
the resultingsolid was washed with CH 2Cl2 and Et2O in turn. Compound
9 was obtained as a brown solid in 90% yield (40 mg, 0.03 mmol).
1H NMR(300 MHz, [D6]DMSO, 258C, TMS): d=2.28 (s, 6H; CH2), 3.15
(s, 27H; NCH3), 3.56 (s, 6H; NCH2), 4.21 (s, 6H; OCH2), 7.17 (d, 3J-
5,10,15-Tris[4-(4-N-triethylbenzyl)amidephenyl]corrolatocopper
ACHTREUNG(III) (13):
A
solution of 11 (30 mg, 0.028 mmol) and triethylamine (10 mL,
72 mmol) in THF (10 mL) was stirred and heated to reflux for 16 h.
After the solvent and the startingmaterial triethylamine were evaporated
under reduced pressure, the crude product was recrystallized from meth-
anol/ether. Compound 13 was obtained as a black solid in 87% yield
(33 mg, 0.024 mmol). 1H NMR (300 MHz, [D6]DMSO, 258C, TMS): d=
1.35 (s, 27H; CH3), 3.25 (s, 18H; NCH2), 4.61 (s, 6H; CH2 in benzyl),
7.73 (brs, 16H; Ph-H and b-pyrrole H), 8.18 (brs,16H; Ph-H and b-pyr-
role H), 10.79 ppm (brs, 3H; NH); 13C NMR (75 MHz, [D6]DMSO,
258C, TMS): d=52.75, 59.70, 120.75, 129.09, 131.86, 133.33, 136.96,
165.67 ppm; UV/Vis (CH3OH): lmax (e”10À3 LmolÀ1 cmÀ1)=417 (136.3),
539 (13.2) nm; HRMS (ESI): m/z: calcd for [C79H86Cl3N10O3CuÀ3Cl]/3:
428.5393; found: 428.5387 [M+À3Cl]/3; elemental analysis (%) calcd for
C79H86Cl3N10O3Cu·3H2O: C 65.55, H 6.41, N 9.68; found: C 65.63, H
6.82, N 9.13.
ACHTREUNG
AHCTREUNG
H), 8.14 ppm (s, 2H; b-pyrrole H); 13C NMR (75 MHz, [D6]DMSO,
258C, TMS): d=23.38, 53.14, 63.78, 65.76, 115.19, 115.40, 132.80, 159.60,
159.85 ppm; UV/Vis (CH3OH): lmax (e”10À3 LmolÀ1 cmÀ1)=427 (91.35),
544 (8.35), 634 (3.9) nm; HRMS (ESI): m/z: calcd for
[C55H65CuI3N7O3À3I]/3: 311.4819; found: 311.4806 [M+À3I]/3; elemental
analysis (%) calcd for C55H65CuI3N7O3·2H2O: C 48.84, H 5.14, N 7.25;
found: C 48.41, H 5.35, N 6.98.
5,10,15-Tris(4-aminophenyl)corrolatocopperACTHRE(UNG III) (10): A solution of
5,10,15-tris(4-aminophenyl)corrole (524 mg, 0.92 mmol) and cupric ace-
tate hydrate (5.24 g, 26.3 mmol) in pyridine (50 mL) was stirred at room
temperature for 30 min. The solvent was evaporated and the resulting
solid was isolated by chromatography on alumina (200–300 mesh) with
THF/methanol (10:1). The second brown fraction was collected. After re-
moval of the solvents under reduced pressure, the solid was recrystallized
from THF/petroleum ether. Compound 10 was obtained as a black–
brown solid in 83% yield (480 mg, 0.76 mmol). 1H NMR (300 MHz,
5,10,15-Tris{4-[4-(1-pyridyl)benzyl]amidephenyl}corrolatocopperACHTREUNG(III)
(14): This compound was obtained by usingthe same procedure de-
scribed for 13. The startingmaterials were 11 (40 mg, 0.037 mmol) and
pyridine (10 mL, 124 mmol). Compound 14 was obtained as a black solid
in 76% yield (37 mg, 0.028 mmol). 1H NMR (300 MHz, [D6]DMSO,
258C, TMS): d=6.00 (s, 6H; Ph), 7.74 (s, 12H; Ph-H and b-pyrrole H),
8.09 (brs, 13H; Ph-H , pyridine H and b-pyrrole H), 8.24 (s, 8H; pyridine
H and b-pyrrole H), 8.68 (s, 6H; pyridine H and b-pyrrole H), 9.30 (s,
8H; pyridine H and b-pyrrole H), 10.69 ppm (s, 3H; NH); 13C NMR
(75 MHz, [D6]DMSO, 258C, TMS): d=63.36, 120.80, 129.26, 129.42,
129.53, 131.73, 136.27, 138.30, 140.66, 145.74, 146.86, 165.80 ppm; UV/Vis
(CH3OH): lmax (e”10À3 LmolÀ1 cmÀ1)=417 (126.7), 539 (11.8) nm;
HRMS (ESI): m/z: calcd for [C76H56Cl3N10O3CuÀ3Cl]/3: 406.4609;
found: 406.4606 [M+À3Cl]/3; elemental analysis (%) calcd for
C76H56Cl3N10O3Cu·3H2O: C 66.08, H 4.52, N 10.14; found: C 66.02, H
4.89, N 9.75.
[D6]DMSO, 258C, TMS): d=5.74 (d, 3J
ACHTREUNG
6.73 (m, 6H; Ph-H), 7.42–7.48 (m, 6H; Ph-H), 7.54 (d, 3J
AHCTREUNG
4H; b-pyrrole H), 7.74 (s, 2H; b-pyrrole H), 7.95 ppm (s, 2H; b-pyrrole
H); 13C NMR (75 MHz, [D6]DMSO, 258C, TMS): d=113.95, 114.19,
121.44, 126.65, 127.44, 128.83, 130.08, 132.39, 133.67, 143.27, 144.90,
149.95, 150.54, 150.83, 152.16 ppm; UV/Vis (THF): lmax (e”
10À3 LmolÀ1 cmÀ1)=387 (186.5), 450 (30.6), 584 (90.5) nm; HRMS (ESI):
m/z: calcd for C37H26N7Cu: 631.1546; found: 631.1551; elemental analysis
(%) calcd for C37H26N7Cu: C 70.29, H 4.15, N 15.51; found: C 70.08, H
3.78, N15.26.
5,10,15-Tris{4-[4-N-dimethyl-N-(2-ethanol]benzyl]amidephenyl}-
5,10,15-Tris[4-(4-benzylchloride)amidephenyl]corrolatocopper
A
ACHRTEUGcN orrolatocopperACHRTE(UGN III) (15): This compound was obtained by usingthe
A solution of 10 (100 mg, 0.16 mmol), 4-(chloromethyl)benzoyl chloride
(180 mg, 0.95 mmol), and triethylamine (96 mg, 0.95 mmol) was stirred at
room temperature for 2 h. After the reaction went to completion, the sol-
vents were evaporated under reduced pressure. The residue was isolated
same procedure described for 13. The startingmaterials were 11 (40 mg,
0.037 mmol) and 2-(dimethylamino)ethanol (10 mL, 100 mmol). Com-
pound 15 was obtained as
a black solid in 84% yield (42 mg,
0.031 mmol). 1H NMR (300 MHz, [D6]DMSO, 258C, TMS): d=3.06 (s,
Chem. Eur. J. 2008, 14, 9431 – 9441
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
9439

X. Zhou et al.
18H; NCH3), 3.94 (s, 6H; NCH2), 4.71 (s, 6H; CH2 in ethanol), 5.50 (s,
6H; CH2), 7.80 (s, 13H; Ph-H), 8.03 (brs, 6H; Ph-H and b-pyrrole H),
8.15 (brs, 13H; Ph-H and b-pyrrole H), 10.72 ppm (brs, 3H; NH);
13C NMR (75 MHz, [D6]DMSO, 258C, TMS): d=50.77, 55.60, 66.02,
67.44, 120.80, 128.93, 132.17, 133.87, 136.99, 165.74 ppm; UV/Vis
(CH3OH): lmax (e ”10À3 LmolÀ1 cmÀ1)=417 (134.7), 539 (12.4) nm;
HRMS (ESI): m/z: calcd for [C73H74Cl3N10O3CuÀ3Cl]/3: 416.5029;
found: 416.5026 [M+À3Cl]/3; elemental analysis (%) calcd for
C73H74Cl3N10O3Cu·3H2O: C 62.12, H 5.71, N 9.92; found: C 61.75, H
6.11, N 9.61.
LiCl) was heated at 958C for 5 min and cooled slowly to room tempera-
ture. Then biotin–DNA (30 mL) was injected at a flow rate of 5 mLminÀ1
.
One of the flow cells was used to immobilize the DNA and another
served as a blank reference.
All the samples were dissolved in dimethylsulfoxide (DMSO; 10 mm)
and prepared in freshly filtered and degassed running buffer (N-(2-hy-
droxyethyl)-piperazine-N’-2-ethanesulfonic acid (HEPES) buffer saline
HBS–EP/KCl, pH 7.4, 0.01m HEPES, 0.2m KCl, 3 mm EDTA, and
0.005% (v/v) surfactant P20) by serial dilutions from stock solutions. This
+
DNA folded in the runningbuffer with K ions (200 mm) and formed a
[31]
Biology
quadruplex duringextended flow in the SPR experiments.
Materials: All the oligomers/primers were synthesized and PAGE puri-
fied by Invitrogen Technology (Shanghai, China). Taq DNA polymerase
was purchased from Toyobo (China). CT-DNA was purchased from Bio-
sharp (China). The oligomers Pu27 (5’-TGGGGAGGGTGGG-
The dsDNA consisted of biotin–oligoDNA hybridized with its comple-
mentary sequence. The sensor chip was regenerated by injection of
NaOH/NaCl (20 mm/1m, 5 mL). All the procedures used repetitive cycles
of the same injection and regeneration.
GAGGGTGGGGAAGG-3’),
GAAAGGGTGGGGAAGG-3’),
Pu27 mu
and
(5’-TGGGGAGGGTG-
RevPu27 (5’-ATC-
DNA bindingexperiments were carried out in the runningbuffer
(pH 7.4, 0.01m HEPES, 0.2m KCl, 3 mm EDTA, and 0.005% (v/v) surfac-
tant P20) at a flow rate of 10 mLminÀ1
.
GATCGCTTCTCGTCCTTCCCCA-3’) were used in the PCR stop assay,
the sequence Telo24 (5’-TTAGGGTTAGGGTTAGGGTTAGGG-3’) was
used in CD spectroscopy, and G4C (5’-CATGG TGGTTT GGGTTA
GGGTTA GGGTTA GGGTTA CCAC-3’) was used in CD melting. The
sequence G4A (5’-AGGGTTAGGGTTAGGGTTAGGG-3’) was used in
the UV/Vis titration experiments.
Fitting of the kinetic data: Generally kinetics parameters are obtained by
global fitting of the kinetic data by using the BIAevaluation program
with 1:1 bindingwith mass transfer. The values ka and kd are the corre-
spondingassociation and dissociation rate constants: KA =ka/kd and KD =
kd/ka.[8,31]
PCR stop assay: The PCR stop assay performed was different from the
previous study. The test oligomers 21G, 21Gmu or Pu27, Pu27 mu and
the correspondingcomplementary sequences Rev21G and RevPu27 were
used in the current study.[23,26] The reactions were performed in 1”PCR
buffer, containinga mixture of 21G, Rev21G, dNTP, Taq polymerase, and
the correspondingcorrole derivatives in solution with water and were
amplified by PCR at a total volume of 12.5 mL. After PCR, 2 mL of load-
ingbuffer (solution (10 mL) included glycerol (1 mL), bromophenol blue
(25 mg), and Ficoll-400 (1.5 g)) was added to each mixture. The samples
were subsequently analyzed by 16% nondenaturingpolyacrylamide gels,
which were carried out in 1”Tris/borate/EDTA buffers at room tempera-
ture. The gels were stained by using ethidium bromide and photos were
taken.[23–26]
UV/Vis titration assay: The corroles were prepared in a solution of
DMSO (10 mm). G4A was dissolved in 10 mm Tris—HCl, 1 mm EDTA,
and 100 mm NaCl to form a 100 mm solution with respect to G4A. The so-
lution was heated at 958C for 5 min, cooled down to room temperature,
and stored in the refrigerator. CT-DNA was dissolved in 10 mm Tris—
HCl and 1 mm EDTA. The concentration of CT-DNA was determined by
measuringthe absorbance at l=260 nm with an extinction coefficient of
13100mÀ1 cmÀ1
.
The absorption spectra were collected at room temperature usinga Shi-
madzu 1901 UV/Vis double-beam spectrophotometer. Quartz cuvettes
with a path length of 1 cm were used for all the absorbance studies. A so-
lution of 10 mm Tris—HCl, 1 mm EDTA, and 100 mm NaCl at pH 7.4 was
used for the G4A DNA and a solution of 10 mm Tris—HCl and 1 mm
EDTA at pH 7.4 was used for the CT-DNA.
CD spectroscopy: The CD experiments, utilizinga Jasco-810 spectropo-
larimeter (Jasco, Easton, MD, USA), were measured at room tempera-
All the absorption titrations were performed by addinga stock solution
of G4A in 10 mm Tris—HCl, 1 mm EDTA, and 100 mm NaCl at pH 7.4 to
a cuvette containingsome corroles in 10 m m Tris—HCl, 1 mm EDTA,
and 100 mm NaCl at pH 7.4 (see Figure S4 in the Supporting Informa-
tion). For the CT-DNA, the titrations were performed by addinga stock
solution of CT-DNA in 10 mm Tris—HCl and 1 mm EDTA at pH 7.4 to a
cuvette containingsome corroles in 10 m m Tris—HCl and 1 mm EDTA at
pH 7.4 (see Figure S5 in the Supporting Information).
ture usinga quartz cell with a 1-mm path length cuvette. The CD spectra
À1
were collected at l=220–350 nm with a scanningspeed of 100 nmmin
The oligomer d
.
A
pended in 10 mm Tris–HCl buffer (1 mm EDTA, pH 7.4) in the absence
or presence of 100 mm NaCl. Various amounts of a stock solution of the
corrole were continually added to increase the concentration ratio of cor-
role/DNA strand. All the CD spectra were baseline corrected for signal
contributions from the buffer.[8,30]
CD thermal melting studies: A solution of G4C oligomer (10 mm; se-
quence 5’-CATGGTGGTTTGGGTTAGGGTTAGGGTTAGGGTTAC-
CAC-3’) was prepared in Na+ buffer (10 mm Tris–HCl, 1 mm EDTA,
100 mm NaCl, pH 7.4) containing50 mm corrole to be tested, the mixture
was first heated to 958C for 5 min, then slowly cooled to room tempera-
ture, and incubated at 48C for at least 6 h. The CD spectra were recorded
on a Jasco-810 spectropolarimeter equipped with a temperature control-
ler with a response time of 1 s and a bandwidth of 1 nm. The CD melting
curves were obtained by monitoringthe absorbance at l=295 nm while
Acknowledgements
This work was supported by the National Science of Foundation of China
(No. 20672084), National Science Fund for Distinguished Young Scholars
(No. 20425206), the Cultivation Fund of the Key Scientific and Technical
Innovation Project, the Ministry of Education of China (No. 706040), the
State Key Laboratory of Natural and Biomimetic Drugs, and the State
Key Laboratory of Applied Organic Chemistry.
the temperature was increased from 20 to 858C at about
[6b,8,30]
1.58CminÀ1
.
SPR testing: The oligonucleotides and materials purchased from Sangon
(Shanghai, China) are G4 quadruplex: 5’-biotin-AG3A(TTAG3)3-3’; DNA
CTHREUNG
duplex: 5’-biotin-GGGCATAGTGCGTGGGCGTTAGC-3’; complemen-
tary strand: (5’-TAACGCCCACGCACTATGCC-3’).
Immobilization of the DNA and SPR binding: Biotinylated DNA was
immobilized on a CM5 sensor chip (BIAcore, Switzerland) with strepta-
vidin by usinga BIAcore 3000 optical biosensor. Streptavidin was cou-
pled to the carboxymethylated dextran matrix coveringthe surface of the
sensor chip by usingan amine couplingkit (BIAcore, Switzerland) ac-
cordingto the manufacturerꢁs instruction. Biotinylated oliognucleotide
(0.5 mm) in couplingbuffer (10 m m Tris, pH 7.4, 1 mm EDTA, and 150 mm
[1] a) J. L. Sessler, S. J. Weghorn, Expanded, Contracted, and Isomeric
Porphyrins, Pergamon, Oxford, 1997, pp. 11–125; b) A. Jasat, D.
Dolphin, Chem. Rev. 1997, 97, 2267–2340.
[2] a) R. Grigg, R. J. Hamilton, M. L. Josefowicz, C. H. Rochester, R. T.
Terrell, H. Wickwar, J. Chem. Soc. Perkin Trans. 2 1973, 407–413;
b) A. W. Johnson, I. T. Kay, J. Chem. Soc. 1965, 1620–1629.
[3] Z. Gross, H. B Gray, Comments Inorg. Chem. 2006, 27, 61–72.
9440
www.chemeurj.org
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9431 – 9441

Stabilization of Guanine Quadruplexes by Metal–Corrole Complexes
FULL PAPER
[4] a) L. R. Kelland, Eur. J. Cancer 2005, 41, 971–979; b) J. L. Mergny,
J. F. Riou, P. Mailliet, M. P. Teulade-Fichou, E. Gilson, Nucleic
Acids Res. 2002, 30, 839–865.
[18] I. H. Wasbotten, T. Wondimagegn, A. Ghosh, J. Am. Chem. Soc.
2002, 124, 8104.
[19] M. Read, R. J. Harrison, B. Romagnoli, F. A. Tanious, S. H. Gowan,
A. P. Reszka, W. D. Wilson, L. R. Kelland, S. Neidle, Proc. Natl.
Acad. Sci. USA 2001, 98, 4844–4849.
[5] a) Y. Wang, D. J. Patel, Structure 1993, 1, 263–282; b) A. T. Phan, V.
Kuryavyi, K. N. Luu, D. J. Patel, Proc. Natl. Acad. Sci. USA 2007,
35, 4927–4940; c) J. X. Dai, C. Punchihewa, A. Ambrus, D. Chen,
R. A. Jones, D. Z. Yang, Proc. Natl. Acad. Sci. USA 2007, 104,
2440–2450; d) A. Ambrus, D. Chen, J. X. Dai, T. Bialis, R. A. Jones,
D. Z. Yang, Proc. Natl. Acad. Sci. USA 2006, 103, 2723–2735;
e) E. M. Rezler, J. Seenisamy, S. Bashyam, M. Y. Kim, E. White,
W. D. Wilson, L. H. Hurley, J. Am. Chem. Soc. 2005, 127, 9439–
9447.
[20] S. Nardis, D. Monti, R. Paolesse, Mini-Rev. Org. Chem. 2005, 2,
355–374.
[21] a) D. S. Sigman, Acc. Chem. Res. 1986, 19, 180–186; b) D. S. Sigman,
A. Mazunder, D. M. Perrin, Chem. Rev. 1993, 93, 2295–2316;
c) P. U. Maheswari, S. Roy, H. den Dulk, S. Barends, G. van Wezel,
B. Kozlevcar, P. Gamez, J. Reedijk, J. Am. Chem. Soc. 2006, 128,
710–711.
[6] a) V. Gabelica, E. S. Baker, M. P. Teulade-Fichou, E. De Pauw, M. T.
Bowers, J. Am. Chem. Soc. 2007, 129, 895–904; b) L. G. Ren, A. M.
Zhang, J. Huang, P. Wang, X. C. Weng, L. X. Zhang, F. Liang, Z.
Tan, X. Zhou, ChemBioChem 2007, 8, 775–780; c) E. S. Baker, J. T.
Lee, J. L. Sessler, M. T. Bowers, J. Am. Chem. Soc. 2006, 128, 2641–
2648; d) J. F. Riou, Curr. Cancer Drug Targets 2004, 4, 439–443;
e) R. T. Wheelhouse, D. Sun, H. Han, F. X. Han, L. H. Hurley, J.
Am. Chem. Soc. 1998, 120, 3261–3262; f) N. V. Anantha, M. Azam,
R. D. Sheardy, Biochemistry 1998, 37, 2709–2714.
[7] Z. Gershman, I. Goldberg, Z. Gross, Angew. Chem. 2007, 119,
4398–4402; Angew. Chem. Int. Ed. 2007, 46, 4320–4324.
[8] B. Q. Fu, J. Huang, L. G. Ren, X. C. Weng, Y. Y. Zhou, Y. H. Du,
X. J. Wu, X. Zhou, G. F. Yang, Chem. Commun. 2007, 3264–3266.
[9] J. E. Reed, A. A. Amal, S. Neidle, R. Vilar, J. Am. Chem. Soc. 2006,
128, 5992–5993.
[10] P. J. Perry, M. A. Read, R. T. Davies, S. M. Gowan, A. P. Reszka,
A. A. Wood, L. R. Kelland, S. Neidle, J. Med. Chem. 1999, 42,
2679–2684.
[11] R. J. Harrison, J. Cuesta, G. Chessari, M. A. Read, . S. K. Basra,
A. P. Reszka, J. Morrell, S. M. Gowan, C. M. Incles, F. A. Tanious,
W. D. Wilson, . L. R. Kelland, S. Neidle, J. Med. Chem. 2003, 46,
4463–4476.
[22] a) I. H. Wasbotten, T. Wondimagegn, A. Ghosh, J. Am. Chem. Soc.
2002, 124, 8104–8116; b) C. Bru1ckner, R. P. Brinas, J. A. K. Bauer,
Inorg. Chem. 2003, 42, 4495–4497; c) I. Aviva, Z. Gross, Chem.
Commun. 2007, 1987–1999.
[23] T. Lemarteleur, D. Gomez, R. Paterski, E. Mandine, P. Mailliet, J. F.
Riou, Biochem. Biophys. Res. Commun. 2004, 323, 802–808.
[24] P. Balagurumoorthy, S. K. Brahmachari, J. Biol. Chem. 1994, 269,
21858–21869.
[25] P. Balagurumoorthy, S. K. Brahmachari, D. Mohanty, M. Bansal, V.
Sasisekharan, Nucleic Acids Res. 1992, 20, 4061–4067.
[26] T. M. Ou, Y. J. Lu, C. Zhang, Z. S. Huang, X. D. Wang, J. H. Tang,
Y. Chen, D. L. Ma, K. Y. Wong, J. C. Tang, A. S. Chan, L. Q. Gu, J.
Med. Chem. 2007, 50, 1465–1474.
[27] C. L. Grand, T. J. Powell, R. B. Nagle, D. J. Bearss, D. Tye, M. Glea-
son-Guzman, L. H. Hurley, Proc. Natl. Acad. Sci. USA 2004, 101,
6140–6145.
[28] a) I. M. Dixon, F. Lopez, A. M. Tejera, J.-P. Estꢂ ve, M.A. Blasco,
G. Pratviel, B. Meunier, J. Am. Chem. Soc. 2007, 129, 1502–1503;
b) J. Bendix, H. B. Gray, G. Golubkov, Z. Gross, Chem Commun.
2000, 1957–1958.
[29] R. Giraldo, M. Suzuki, L. Chapman, D. Rhodes, Proc. Natl. Acad.
Sci. USA 1994, 91, 7658–7662.
[12] K. Jantos, R. Rodriguez, S. Ladame, P. S. Shirude, S. Balasubramani-
an, J. Am. Chem. Soc. 2006, 128, 13662- 13663.
[30] D. P. N. Goncalves, R. Rodriguez, S. Balasubramanian, J. K. M.
Sanders, Chem. Commun. 2006, 4685–4687.
[13] P. S. Shirude, E. R. Gillies, S. Ladame, F. Godde, K. Shin-ya, I. Huc,
S. Balasubramanian, J. Am. Chem. Soc. 2007, 129, 11890–11891.
[14] A. D. Moorhouse, A. M. Santos, M. Gunaratnam, M. Moore, S.
Neidle, J. E. Moses, J. Am. Chem. Soc. 2006, 128, 15972–15973.
[15] A. D. Cian, E. DeLemos, J.-L. Mergny, M.-P. Teulade-Fichou, D.
Monchaud, J. Am. Chem. Soc. 2007, 129, 1856–1857.
[16] I. M. Dixon, F. Lopez, A. M. Tejera, J.-P. Estꢂve, M. A. Blasco, G.
Pratviel, B. Meunier, J. Am. Chem. Soc. 2007, 129, 1502–1503.
[17] K. Nemoto, Y. Kondo, S. Himeno, Y. Suzuki, S. Hara, M. Akimoto,
N. Imura, Biochem. Pharmacol. 2000, 59, 401–405.
[31] P. Wang, L. Ren, H. He, F. Liang, X. Zhou, Z. Tan, ChemBioChem
2006, 7, 1155–1159.
[32] J. E. Reed, A. A. Arnal, S. Neidle, R. Vilar, J. Am. Chem. Soc. 2006,
128, 5992–5993.
[33] L. R. Keating, V. A. Szalai, Biochemistry 2004, 43, 15891–15900.
[34] R. F. Pasternack, E. J. Gibbs, J. J. Villafranca, Biochemistry 1983, 22,
2406–2414.
Received: May 2, 2008
Published online: August 27, 2008
Chem. Eur. J. 2008, 14, 9431 – 9441
ꢀ 2008 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
9441