Development of cobalt(3,4-diarylsalen) complexes as tumor therapeutics

Article abstract of DOI:10.1021/jm040763n

[1,6-Bis(2-hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-1,5-diene]cobalt(II) complexes (cobalt(3,4-diarylsalen)) with 2-, 3-, or 4-OCH₃/OH substituents in the 3,4-standing aryl rings were synthesized and tested for antitumor activity in vitro on the MCF-7, MDA-MB 231, and LNCaP/ FGC cell lines. The cytotoxicity depended on both the configuration of the diene ligand and the kind of substituents in the 3,4-standing aromatic rings. d,1-7 (2-OCH ₃), d,1-8 (3-OCH₃), and d,1-9 (4-OCH₃) were equipotent to cisplatin, while the respective hydroxy-substituted complexes (d,1-10 (2-OH), d,1-11(3-OH), and d,1-12 (2-OH)) as well as all of the meso-configured compounds (m-7 to m-12) did not influence the cell growth. Interestingly, a high catalytic potency and a rapid and high accumulation in MCF-7 cells (15- to 25-fold compared to the cell culture medium (5 μM)) were demonstrated for m-7 (2-OCH₃), m-8 (3-OCH₃), and m-9 (4-OCH₃). Therefore, a mode of action based on a cobalt-catalyzed oxidative damage of the DNA is not very likely.

Full text of DOI:10.1021/jm040763n

J. Med. Chem. 2004, 47, 5837-5846
5837
Articles
Development of Cobalt(3,4-diarylsalen) Complexes as Tumor Therapeutics
Ronald Gust,* Ingo Ott, Diana Posselt, and Klaus Sommer
Institute of Pharmacy, Free University of Berlin, K o¨ nigin-Luise Strasse 2+4, D-14195 Berlin, Germany
Received January 5, 2004
[
1,6-Bis(2-hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-1,5-diene]cobalt(II) complexes (cobalt(3,4-
diarylsalen)) with 2-, 3-, or 4-OCH /OH substituents in the 3,4-standing aryl rings were
3
synthesized and tested for antitumor activity in vitro on the MCF-7, MDA-MB 231, and LNCaP/
FGC cell lines. The cytotoxicity depended on both the configuration of the diene ligand and
the kind of substituents in the 3,4-standing aromatic rings. d,l-7 (2-OCH
3
3
), d,l-8 (3-OCH ),
and d,l-9 (4-OCH ) were equipotent to cisplatin, while the respective hydroxy-substituted
3
complexes (d,l-10 (2-OH), d,l-11 (3-OH), and d,l-12 (2-OH)) as well as all of the meso-configured
compounds (m-7 to m-12) did not influence the cell growth. Interestingly, a high catalytic
potency and a rapid and high accumulation in MCF-7 cells (15- to 25-fold compared to the cell
culture medium (5 µM)) were demonstrated for m-7 (2-OCH
OCH ). Therefore, a mode of action based on a cobalt-catalyzed oxidative damage of the DNA
is not very likely.
3
), m-8 (3-OCH
3
), and m-9 (4-
3
Introduction
The pioneering work of Barnett Rosenberg on the
antitumor activity of cis-diamminedichloroplatinum(II)
studies in the class of [1,2-diamino-1,2-diarylethane]-
platinum(II) complexes. It was demonstrated that the
1
,2-diarylethane moiety has to be OCH3- or OH-
substituted to get complexes with high tumor se-
lectivity.8 Because this pharmacophor can be realized
in a similar spatial arrangement in [1,6-bis(2-hydroxy-
phenyl)-3,4-diaryl-2,5-diazahexa-1,5-diene]cobalt(II), we
synthesized OH-phenyl and OCH3-phenyl substituted
cobalt(3,4-diarylsalen) complexes and studied their anti-
proliferative effects on the MCF-7 and MDA-MB 231
breast cancer as well as the LnCaP/FGC prostate cancer
cell lines.
(
cisplatin) is an important milestone in cancer chemo-
-12
1
,2
therapy. Cisplatin found its first clinical application
in the early 1970s and is currently one of the most
successful cancer chemotherapeutics. This great success
of an “old” drug (cisplatin was first synthesized by
Peyrone in 1844) led to the search for well-known
compounds that were rarely discovered as potential
active agents.
Recently, we identified acetylenehexacarbonyldicobalt
complexes as a new class of antitumor agents.3 We
demonstrated a strong dependence of the antitumor
activity as well as the tumor selectivity on the kind of
alkyne ligand. The most active derivative [(2-propyn-
,4
Synthesis and Structural Characterization
The 1,6-bis(2-hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-
,5-dienes were synthesized by reaction of the corre-
1
1
-yl)acetylsalicylate]hexacarbonyldicobalt showed a high
sponding diastereomerically pure 1,2-diamino-1,2-
selectivity for human breast cancer cells and was even
8,9,12
diarylethanes
with salicylaldehyde (Scheme 1).
3
more active than cisplatin.
The subsequent coordination to cobalt was performed
A further very interesting class of potential cytotoxic
drugs represent cobalt(salen) complexes (salen ) 1,6-
bis(2-hydroxyphenyl)-2,5-diazahexa-1,5-diene) mainly
developed as reversible oxygen carriers and catalysts
for oxidation reactions. However, there are several
indications that these compounds can also catalyze the
building of hydroxy radicals under physiological condi-
13
according to Berkelew and Calvin using cobalt(II)
acetate.
IR spectroscopy, mass spectrometry, and elemental
analysis allowed the characterization of the new com-
plexes. NMR spectroscopy could not be used because of
the marked paramagnetism of the complexes.
The coordination of 3,4-diarylsalenes and the building
of a square-planar chelate complex led to characteristic
changes in the fingerprint region of the IR spectra. The
5
6
tions. Singlet oxygen and superoxides described as
reactive species may inflict damages to the DNA.7
Therefore, we tried to design cobalt(salen) complexes
with high cytotoxicity and selectivity against tumor
cells.
CdN stretching vibration determined for the ligands at
-1
1
623-1627 cm was shifted to lower wavenumbers
-1
between 1601 and 1608 cm . Furthermore, the bending
vibration disappeared in the area at about 1278 cm
While designing these complexes, it was possible to
consider the results of structure-activity relationship
-1
.
A characteristic fragmentation of the cobalt(3,4-
diarylsalen) complexes could be observed in the mass
spectra. Two fragments resulting from the consecutive
*
To whom correspondence should be addressed. Phone: (030) 838
5
3272. Telefax: (030) 838 56906. E-mail: rgust@zedat.fu-berlin.de.
1
0.1021/jm040763n CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/21/2004

5
838 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24
Gust et al.
Scheme 1
Scheme 2
Table 1. Catalytic Effect of [1,6-Bis(2-hydroxyphenyl)-
,4-diphenyl-2,5-diazahexa-1,5-diene]cobalt Complexes,
3
Showing the Conversion of 2,6-di-tert-butylphenol (DP) to
Benzochinone (BQ) and Diphenochinone (DPQ) According to
Scheme 2
m-7, % m-8, % m-9, % d,l-7, % d,l-8, % d,l-9, %
DP
DPQ
BQ
5
1
94
8
1
91
1
2
97
56
38
6
42
54
4
45
51
4
conversion of DP mainly into DPQ. Interestingly, the
catalytic behavior of the complexes was independent of
the position and the kind of substituents in the 3,4-
standing aromatic rings. The hydroxy-substituted com-
pounds m-10 and d,l-10 to m-12 and d,l-12 showed
effects simliar to those of their methoxy derivatives
(data not shown).
elimination of a salicylimine fragment and a benzyl
fragment were found besides the molecular ion peak.
Investigations of the Catalytic Activity
The catalytic activity was evaluated using the cobalt-
catalyzed oxidation of 2,6-di-tert-butylphenol (DP) first
1
4
described by van Dort and Geursen. This reaction
yielded a mixture of benzoquinone (BQ) and dipheno-
quinone (DPQ) (Scheme 2) whereby the formation of BQ
served as a measure of the catalytic potency of the
complexes. The higher the relation between BQ and
DPQ, the stronger the catalytic effect of the cobalt
complex.
Biological Activity
The antitumor activity was determined in vitro using
the hormone-dependent MCF-7 and the hormone-
independent MDA-MB 231 breast cancer and the
LNCaP/FGC prostate cancer cell lines.
In a preliminary study, the influence of Co(salen)
(Figure 1, right) and CoCl2 (Figure 1, left) was evaluated
against MCF-7 cells. Both compounds showed a con-
centration-dependent reduction of cell proliferation.
Co(salen) was more active than CoCl2 and caused
Nishinaga and Drago16 already elucidated the steps
involved in this reaction, which starts with the building
of a cobalt/O2 complex. The reaction of this superoxo
compound with DP yields a phenoxy and an H2O
radical. While the latter disproportionates into H2O2 and
O2, the phenoxy radical reacts with the cobalt/O2
complex to give a cobalt peroxocyclohexa-2,5-dienone,
which decomposes into BQ. If the reaction with the
activated cobalt complex is disturbed, the concentration
of the phenoxy radical correspondingly increases and
the coupling to DPQ is observed.
15
The quantification of the reaction was performed by
H NMR spectroscopy using the integrals of the tert-
1
butyl groups as a parameter (DP, δ ) 1.28; DPQ, δ )
1
.37). The reaction conditions were chosen (see Materi-
als and Methods) to get a quantitative conversion (100%
BQ) with cobalt(3,4-diarylsalen).
The results of the diastereomeric complexes m-7 and
d,l-7 to m-9 and d,l-9 are listed in Table 1. Complexes
m-7 to m-9 oxidized DP nearly quantitatively (> 90%)
into BQ (see Table 1), while the d,l-configured com-
pounds d,l-7 to d,l-9 caused only an approximately 50%
Figure 1. Antiproliferative effects of CoCl
2
and Co(salen) on
concentrations
the MCF-7 breast cancer cell line: (left) CoCl
2
of 20, 40, 60, 80, 100, 120, 150, 200 µM; (right) Co(salen)
concentrations of 10, 20, 30, 40, 50, 60, 70 µM.

Cobalt(3,4-diarylsalen) Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24 5839
3-OCH3) < m-9 (4-OCH3) < d,l-7 (2-OCH3) < d,l-8 (3-
(
OCH3) ) d,l-9 (4-OCH3).
All complexes reached their maximum effects at the
beginning of the test. While m-7 showed only marginal
effects (T/C ) 80.4% at 10 µM), m-8 and m-9 reduced
the cell proliferation by more than 50% (T/C ) 45.2%
(
m-8) and T/C ) 33.6% (m-9) at 10 µM). Even cytocidal
effects were achieved with d,l-7 (τ ) -15.9% (10 µM)),
d,l-8 (τ ) -8.5% (5 µM) and -31.5% (10 µM)), and d,l-9
(τ ) -17.7% (5 µM) and -26.6% (10 µM)).
The time-activity curves of d,l-7 to d,l-9 presented
in Figure 3 show a recuperation of the tumor cells after
a prolonged exposition. Because exponential cell growth
is guaranteed for at least 250 h of incubation, the rise
of the growth curve can be explained by the development
Figure 2. Cytotoxicity (maximal effects) of cobalt(3,4-diaryl-
salen) complexes against MCF-7 cells.
17
of drug resistance, which depends on the position of
the OCH3 group in the aromatic ring. While the cells
treated with d,l-7 showed at 5 and 10 µM a strong
development of resistance, the shift of the methoxy
group into the 3- (d,l-8) and the 4-positions (d,l-9)
reduced this effect significantly. d,l-8 and d,l-9 were
comparably active to cisplatin and even caused cytocidal
effects (τ ≈ -20% to -40%) at 10 mM concentration
during the entire incubation period (244 h).
antiproliferative effects in concentrations higher than
1
0 µM. Cytocidal effects were achieved in concentrations
of about 60-70 µM. Cisplatin possessed a considerably
higher activity and completely blocked the growth of the
MCF-7 cells at a concentration of 5 µM (Figure 7).
Introduction of aryl rings into the 3,4-positions of
the salen ligand enhanced the antitumor activity of
Co(salen), which is dependent on the configuration at
the asymmetric C atoms and the substitution pattern
of the aromatic rings. The growth inhibition increased
in the following series (Figure 2): m-7 (2-OCH3) < m-8
In contrast to cisplatin, which reached its maximum
effect toward the end of the test (see Figure 7), the onset
of antiproliferative effects was observed much earlier.
Figure 3. Antiproliferative effects of the cobalt(3,4-bis(methoxyphenyl)salen) complexes d,l-7 (up, left), d,l-8 (up, middle), and
d,l-9 (up, right) as well as the respective hydroxyl derivative d,l-10 (down, left), d,l-11 (down, middle), and d,l-12 (down, right).

5
840 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24
Gust et al.
Figure 5. Accumulation of cobalt(3,4-diarylsalen) complexes
into MCF-7 cells.
diarylsalen) complexes. Only the incubation with m-11
Figure 4. Cytotoxicity of d,l-9 on the MCF-7 cell line,
(6.1 µM) and m-12 (7.4 µM) led to a slightly higher
dependent on the time of drug exposure.
cobalt concentration in the tumor cells compared to the
medium (5 µM). This finding correlates very well with
the missing cytotoxicity of the OH-substituted com-
plexes (see Figure 3).
Therefore, we investigated the minimal time of drug
exposure to achieve cytotoxic effects on the example of
d,l-9.
The MCF-7 cells were treated with a drug concentra-
tion of 5 µM for 1, 2, 3, 4, 6, or 12 h. The medium was
subsequently replaced by a drug-free one, and the test
was continued for about 200 h.
As shown in Figure 4, a 1 h drug exposure already
led to a cell growth inhibition of about 60%. This effect
was continuously enhanced with the elongation of the
incubation time. The maximal effect was achieved after
Furthermore, the antiproliferative properties of d,l-
configured cobalt(3,4-diarylsalen) complexes depended
on an intact metal complex. The free ligands d,l-1, d,l-
2
, and d,l-3 reduced the cell growth at 5 µM only
minimally to T/C ) 83%, 75%, and 71%, respectively.
The same dependence on the salen ligand was deter-
mined on the MDA-MB 231 cell line. The OH-substi-
tuted and meso-configured complexes were completely
inactive, while the O-methyl derivatives showed a
concentration-dependent reduction of the cell growth
6
h, and the time-activity curve showed only marginal
differences compared to the one obtained in the regular
experiment (see Figure 3). Therefore, it can be concluded
that the complexes were taken up in MCF-7 cells during
the incubation time of 6 h in cytotoxic amounts.
(Figure 6). They reduced the cell growth at a concentra-
tion of 5 µM by 70-90% after an incubation time of 48
h. At higher concentration, the complexes caused cyto-
cidal effects. A comparison of the time-activity curves
indicates that the antiproliferative properties decreased
slightly with the shift of the O-methyl group from the
4-position to the 2-position. Generally, the effect on the
MDA-MB 231 cell line was somewhat lower than on the
MCF-7 cell line.
The complexes d,l-7, d,l-8, and d,l-9 were also tested
against LNCaP/FGC cells (Figure 6). They were highly
active and reduced the cell growth, comparable to
cisplatin (Figure 7). Interestingly, the complexes showed
nearly identical cell growth characteristics on the three
cell lines. A very fast onset of antitumor activity during
the first 48 h of incubation was followed by a develop-
ment of resistance.
To quantify this process, we used electrothermic
atomic absorption spectroscopy according to the method
already described for cellular studies on [1,2-diamino-
,2-diarylethane]platinum(II) complexes.18
1
The intracellular cobalt concentration increased in a
fast rate (see Figure 5) and already reached a saturation
point after 100 min. Figure 5 demonstrates that the
meso-configured complexes m-7 to m-9 were taken up
in MCF-7 cells to a much higher level than their
diastereomers. The highest intracellular level was
achieved with m-9 (128 µM), which means a 25-fold
enrichment compared to the cell culture medium (5 µM).
The shift of the O-methyl group into the 3- or the
2
-position distinctly reduced the accumulation. MCF-7
cells treated with m-7 and m-8 showed a cobalt content
of 83.5 and 73.0 µM, respectively. The uptake of their
d,l-configured congeners was comparable to that of the
lead compound Co(salen) (26.8 µM) and nearly inde-
pendent of the position of the O-methyl substituent
Chemoresistance is a multifunctional phenomenon;
however, reduced uptake and increased efflux are the
first evidence of resistance for metal complexes. There-
fore, we incubated MDA-MB 231 cells with m-9 and
d,l-9 at a concentration of 5 µM, respectively, and
quantified the cobalt content over a period of 120 h. On
one hand, m-9 showed a 21-fold accumulation in the
tumor cells after 24 h, which decreased to about 13-fold
(
d,l-7 (23.8 µM), d,l-8 (22.7 µM), and d,l-9 (30.7 µM)).
Hydroxy groups in the aromatic rings prevented the
uptake of both the d,l- and meso-configured cobalt(3,4-

Cobalt(3,4-diarylsalen) Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24 5841
Figure 6. Cytotoxic properties of cobalt(3,4-diarylsalen) complexes d,l-7 (left), d,l-8 (middle), and d,l-9 (right) on the MDA-MB
31 (up) and the LNCaP/FGC cell lines (down).
2
Figure 7. Cytotoxicity of cisplatin on the MCF-7, MDA-MB 231, and LNCaP/FGC cell lines.
during the following 96 h. On the other hand, d,l-9 holds
have to be considered for the observed phenomenon.
its 16-fold accumulation during the incubation time of
Metal complexes such as cisplatin enter the tumor cells
by passive diffusion, facilitated diffusion, or active
transport. The loss of activity and development of
resistance are often associated with elevated glutathione
(GSH) levels, enhanced repair of drug-induced DNA
1
20 h.
These values did not correlate with the growth curves
depicted in Figure 6 and contradict the above-mentioned
resistance mechanism. Therefore, other mechanisms

5
842 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24
Gust et al.
lesions, or overexpression of the P-170 glycoprotein.19
On the basis of our accumulation studies, the decrease
of activity observed for cobalt(salen) complexes after a
longer drug exposure may be attributed to these phe-
nomena. However, other cellular defense mechanisms
become more likely because cisplatin caused its anti-
tumor effect on the breast cancer cell lines slowly and
reached its maximal effect at the end of the test (Figure
7
).
Discussion
1,6-Bis(2-hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-
,5-diene]cobalt(II) complexes represent a new class of
[
1
antitumor active metal complexes. They showed high
antiproliferative activity against hormone-sensitive
MCF-7 and hormone-insensitive MDA-MB 231 breast
cancer cells as well as LNCaP/FGC prostate cancer cells.
The cytotoxicity of the complexes depended on an intact
metal complex. The free ligands (data not shown) and
Figure 8. Proposed spatial structure of diastereomeric cobalt-
(
3,4-diarylsalen) complexes.
Cobalt(3,4-diarylsalen) complexes are square-planar
with either bisequatorially (d,l-configured compounds)
or axially/equatorially standing (meso-configured com-
pounds) aromatic rings at the ethane bridge (Figure 8).
The differing arrangements of the 1,2-diarylethane
pharmacophor are of great importance for the pharma-
cological properties of the complexes. In contrast to the
platinum complexes, the orientation towards the cobalt
central atom in meso-configured compounds forced the
transport into the tumor cells and increased the oxida-
tive potency compared to the d,l-configured diastereo-
mers. Nevertheless, the complexes d,l-7 to d,l-9 showed
much higher cytotoxicity. Therefore, we assume that
DNA damage inflicted by superoxide radical mediated
active species only plays a subordinate role.
2+
Co ions did not influence significantly the cell growth
at concentrations less than 50 µM.
The aromatic rings at the 3,4-position in the cobalt-
salen) complexes played an essential role (see Figures
(
2
and 3). Both the substituents in the aromatic rings
and the configuration of the asymmetric C atoms
determined the cytotoxic potency. On one hand, hy-
droxylation prevented the uptake into the tumor cells
and led to completely inactive compounds. On the other
hand, O-methylation resulted in compounds with cyto-
toxic potency comparable to that of cisplatin. In this
case, however, the ligand had to be d,l-configured. The
respective meso-configured diastereomers were enriched
in the tumor cells in higher amounts than the corre-
sponding d,l-configured compounds and showed lower
antiproliferative effects despite their higher catalytic
activity. It indicates the involvement of more specific
processes or interactions inside the tumor cells. Differ-
ences in the intracellular distribution and the uptake
into the nuclei as well as distinct covalent or interca-
lative DNA interactions caused by the different stereo-
chemistry of meso- and d,l-configured derivatives must
be taken into account in future studies on cobalt(salen)
complexes.
24
A mode of action postulated by Liu et al. seems to
be more likely. They investigated the interactive effects
of co(salen) complexes on the DNA and found that the
bonding to DNA decreased the fluorescence intensity
accompanied by a bathochromic shift of the excitation
and emission peaks. A hypochromism in the UV absorp-
tion spectra was also observed. KI quenching and
competitive binding to DNA between Co(salen) and
ethidium bromide showed that the interactive model
between Co(salen) and DNA was an intercalative one.
These studies are quite difficult for cobalt(3,4-diaryl-
salen) complexes because of their relatively low
water solubility. Therefore, we used in an initial
study the [meso-1,6-bis(2-hydroxy-4-trimethylamino-
phenyl)-3,4-bis(4-methoxyphenyl)-2,5-diazahexa-1,5-
The observed discrimination between diastereomers
supports the evidence that the uptake of the complexes
into tumor cells is carrier-mediated, as already con-
firmed for the aqua[1,2-diamino-1,2-bis(4-fluorophenyl)-
ethane]sulfatoplatinum(II) complexes d,l-4F-PtSO4 and
+
dien]cobalt(II) complex m-9-NMe3 as a model. m-9-
20
meso-4F-PtSO4. The accumulation of d,l-4F-PtSO4 was
very rapid, with a 14-fold intracellular platinum con-
centration compared to that of the cell culture medium
+
NMe3 induced a significant change of the melting point
∆Tm) of the DNA. ∆Tm amounted to 2.7 °C at a DNA
(
base pair/ligand ratio of 20 and amounted to 7.2 °C at
a DNA base pair/ligand ratio of 40. Compared to other
metal(salen) complexes, however, these values were
low.25 This seems to be the consequence of the aryl
substituents at the salen ligand, which hinder the
approach to the DNA.
This effect was already discussed for [1,2-diamino-
1,2-diarylethane]platinum(II) complexes. meso-4F-
PtSO4 shows a dynamic interconversion of the five-
membered chelate ring. The orientation of the aryl
substituents during the interconversion of the chelate
ring towards the platinum central atom reduced the
binding tendency to guanine bases of the DNA, resulting
in low cytotoxic effects.
(5 µM) that was reached within 4 h. With meso-4F-
PtSO4, only a 3-fold enrichment was achieved.
The 1,2-diarylethane pharmacophor is also realized
in the 1,6-bis(2-hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-
1
,5-diene. These ligands can be considered to be salicyl-
imine derivatives of the 1,2-diamino-1,2-diarylethane
carrier ligands.
The spatial structure of [1,6-bis(2-hydroxyphenyl)-3,4-
diaryl-2,5-diazahexa-1,5-diene]cobalt(II) complexes could
1
not be evaluated by H NMR spectroscopy because of
the paramagnetism of the complexes. Therefore, we
propose a three-dimensional structure based on already
2
1-23
published bond length and angles.

Cobalt(3,4-diarylsalen) Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24 5843
The d,l-4F-PtSO4 and related hydroxy- or methoxy-
substituted complexes exist in a stable conformation,
with exclusively equatorial arranged aromatic rings at
the puckered five-membered chelate ring. A conversion
into the conformation with axial standing phenyl rings
is not observed. A better approach of d,l-4F-PtSO4 to
the DNA is therefore very likely, resulting in higher
cytotoxicity.
This might also be true for d,l-9 and m-9. Unfortu-
nately, the introduction of trimethylamonium groups
into the 1,6-standing phenol rings did not increase the
solubility of d,l-9 to determine the DNA intercalation
significantly. Nevertheless, these findings demonstrated
clearly that the DNA intercalation of cobalt(3,4-diaryl-
salen) complexes must be taken into consideration as a
possible mode of action.
mmol, 272.0 mg) and salicylaldehyde (2.1 mmol, 256.2 mg).
Yield: 0.93 mmol (448.4 mg), 93%, pale-yellow crystals, mp
-
1
1
98 °C. IR (KBr, cm ): νj ) 2836 m (OCH
3
), 2600 m, br (OH),
1
623 s (CdN), 1533 m, 1489 m, 1465 m, 1363 m, 1322 m, 1260
s (OH), 1226 s, 1165 m, 1123 m, 1040 m, 996 w, 976 m, 847
1
m, 790 m, 747 w, 700 m. H NMR (DMSO-d
6
): δ ) 3.68 (s,
J ) 8.05
4
3
6H, OCH ), 5.03 (s, 2H, CH), 6.79 (dd, J ) 2.0 Hz,
3
3
Hz, 2H, ArH-3), 6.85 (m, 4H, Ar′H-2, Ar′H-6), 7.22 (dd, J )
3
8
.13, J ) 8.05 Hz, 2H, Ar′H-5), 7.31 (m, 4H, ArH-4, ArH-6),
.43 (s, 2H, NCH), 13,19 (s, 2H, OH).
8
d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-methoxyphenyl)-
,5-diazahexa-1,5-diene (d,l-2). d,l-2 was obtained from d,l-
2
1
,2-diamino-1,2-bis(3-methoxyphenyl)ethane (1.11 mmol, 302.8
mg) and salicylaldehyde (2.75 mmol, 335.8 mg). Yield: 0.93
mmol (449.7 mg), 84%, yellow powder, mp 116 °C. IR (KBr,
-
1
cm ): νj ) 2835 m (OCH
3
), 2600 m, br (OH), 1627 s (CdN),
1601 s, 1584 s, 1491 m, 1276 s (OH), 1152 m, 1117 w, 1041 m,
1
9
(
24 w, 785 m, 757 m, 706 m. H NMR (DMSO-d ): δ ) 3.66
6
4
3
s, 6H, OCH
3
), 5.06 (s, 2H, CH), 6.75 (dd, J ) 2.3 Hz, J ) 8.1
To confirm this assumption, it is necessary to study
the interaction of the cobalt(3,4-diarylsalen) complexes
with the DNA in detail. Therefore, we intend to optimize
the physical properties of cobalt(3,4-diarylsalen) com-
plexes to perform DNA binding studies.
Hz, 2H, ArH-3), 6.87 (m, 8H, ArH-5, Ar′H-2, Ar′H-4, Ar′H-6),
3
4
7
6
.17 (dd, J ) 8.0 Hz, J ) 2.3 Hz, ArH-4), 7.31 (m, 4H, ArH-
, Ar′H-5), 8.50 (s, 2H, NCH), 13.30 (s, 2H, OH).
meso-1,6-Bis(2-hydroxyphenyl-3,4-bis(4-methoxy-
phenyl)-2,5-diazahexa-1,5-diene (m-3). m-3 was obtained
from meso-1,2-diamino-1,2-bis(4-methoxyphenyl)ethane (1.00
mmol, 272.0 mg) and salicylaldehyde (2.10 mmol, 256.2 mg).
Yield: 0.89 mmol (427.2 mg), 89%, pale-yellow crystals, mp
Materials and Methods
General Procedures. The following instrumentation was
-
1
1
1
1
3
79 °C. IR (KBr, cm ): νj ) 2837 m (OCH ), 2600 m, br (OH),
used: IR spectra (KBr pellets), Perkin-Elmer model 580 A;
628 vs (CdN), 1582 m, 1511 s, 1461 m, 1414 m, 1278 s (OH),
1
H NMR, Bruker ADX 400 spectrometer at 400 MHz (internal
1
249 vs, 1177 m, 1152 m, 1115 m, 1032 s, 830 m, 760 s. H
standard, TMS); EI-MS spectra, CH-7A-Varian MAT (70 eV)
and Kratos MS 25 RF (80 eV). Elemental analyses were carried
out at the Microlaboratory of the Free University of Berlin.
Computational graphics were built using SYBYL 6.7 (Tripos
Inc. 1699 South Hanley Road, St. Louis, MO, 63144). Geometry
optimization was carried out using the MM3 force field within
the program, running on an INDY work station.
NMR (DMSO-d
6
): δ ) 3.68 (s, 6H, OCH
3
), 4.98 (s, 2H, CH),
6
6
.86 (m, 8H, ArH-3, ArH-5, Ar′H-3), 7.27 (m, 8H, ArH-4, ArH-
, Ar′H-2, Ar′H-6), 8.42 (s, 2H, NCH), 13.23 (s, 2H, OH).
d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(4-methoxyphenyl)-
,5-diazahexa-1,5-diene (d,l-3). d,l-3 was obtained from d,l-
2
1
,2-diamino-1,2-bis(4-methoxyphenyl)ethane (1.26 mmol, 311.6
mg) and salicylaldehyde (2.74 mmol, 333.7 mg). Yield: 0.93
mmol (444.6 mg), 74%, yellow needles, mp 189 °C. IR (KBr,
Syntheses. The 1,2-diamino-1,2-diarylethanes were syn-
8
,9,12
thesized as described earlier.
-
1
cm ): νj ) 2836 m (OCH
3
), 2600 m, br (OH), 1625 s (CdN),
581 s, 1512 s, 1460 s, 1415 m, 1606 s, 1277 s (OH), 1248 s,
177 s, 1151 m, 1117 m, 1059 m, 1033 s, 935 w, 905 m, 833 s,
General Procedure for the Synthesis of 1,6-Bis(2-
hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-1,5-dienes. An
amount of 1 mmol of the respective 1,2-diamino-1,2-diaryl-
ethane was suspended in 50 mL of acetonitrile and reacted
with 2.1 mmol of salicylaldehyde. The reaction mixture was
heated to reflux for 6 h. The solvent was reduced by half, and
the diimine was subsequently allowed to crystallize. The
crystals were filtered off, washed with diethyl ether, and dried
1
1
7
5
7
1
1
6 3
55 s, 666 m. H NMR (DMSO-d ): δ ) 3.68 (s, 6H, OCH ),
.01 (s, 2H, CH), 6.83 (m, 8H, ArH-3, ArH-5, Ar′H-2, Ar′H-4),
.28 (m, 8H, ArH-4, ArH-6, Ar′H-2, Ar′H-6), 8.52 (s, 2H, NCH),
3.33 (s, 2H, OH).
meso-1,3,4,6-Tetrakis(2-hydroxyphenyl)-2,5-diazahexa-
,5-diene (m-4). m-4 was obtained from meso-1,2-diamino-
,2-bis(2-hydroxyphenyl)ethane (1.00 mmol, 244.3 mg) and
1
4
over CaSO .
1
meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(2-methoxy-
phenyl)-2,5-diazahexa-1,5-diene (m-1). m-1 was obtained
from meso-1,2-diamino-1,2-bis(2-methoxyphenyl)ethane (1.00
mmol, 272.0 mg) and salicylaldehyde (2.10 mmol, 256.1 mg).
Yield: 1.97 mmol (947.5 mg), 94%, pale-yellow crystals, mp
salicylaldehyde (2.10 mmol, 256.4 mg). Yield: 0.85 mmol (384.6
-
1
mg), 85%, pale-yellow powder, mp 232 °C. IR (KBr, cm ): νj
2600 m, br (OH), 1627 s (CdN), 1528 m, 1490 s, 1458 s,
393 m, 1337 m, 1278 s (OH), 1223 m, 1152 m, 1117 m, 1038
)
1
1
08 °C. IR (KBr, cm-1): νj ) 2837 m (OCH
3
), 2600 m, br (OH),
m, 853 w, 755 s. H NMR (DMSO-d
.79 (m, 4H, ArH-3, ArH-5, Ar′H-3, Ar′H-5), 7.02 (d, J ) 7.7
Hz, 2H, Ar′H-6), 7.28 (m, 6H, ArH-4, ArH-6, Ar′H-4), 8.35 (s,
6
): δ ) 5.51 (s, 2H, CH),
2
1
1
3
6
627 s (CdN), 1583 s, 1493 s, 1461 s, 1402 m, 1278 s (OH),
1
246 s, 1153 m, 1111 m, 1048 m, 1026 m, 849 m, 754 s. H
2
H, NCH), 9.59 (s, br, 2H, Ar′-OH), 13.34 (s, br, 2H, Ar-OH).
6 3
NMR (DMSO-d ): δ ) 3.67 (s, 6H, OCH ), 5.42 (s, 2H, CH),
6
.77 (m, 4H, ArH-3, Ar′H-3), 6.91 (m, 4H, ArH-5, Ar′H-5), 7.26
d,l-1,3,4,6-Tetrakis(2-hydroxyphenyl)-2,5-diazahexa-
(
m, 8H, ArH-4, ArH-6, Ar′H-4, Ar′H-6), 8.40 (s, 2H, NCH),
1,5-diene (d,l-4). d,l-4 was obtained from d,l-1,2-diamino-1,2-
bis(2-hydroxyphenyl)ethane (0.76 mmol, 186.1 mg) and salicyl-
aldehyde (1.57 mmol, 191.7 mg). Yield: 0.30 mmol (132.5 mg),
1
3.36 (s, 2H, OH).
d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(2-methoxyphenyl)-
,5-diazahexa-1,5-diene (d,l-1). d,l-1 was obtained from d,l-
-
1
2
39%, yellow powder, mp 217 °C. IR (KBr, cm ): νj ) 1627 s
(CdN), 1530 m, 1489 m, 1456 s, 1397 m, 1335 m, 1277 s (OH),
1
,2-diamino-1,2-bis(2-methoxyphenyl)ethane (0.491 mmol, 133.8
1
mg) and salicylaldehyde (1.13 mmol, 126.0 mg). Yield: 0.34
mmol (145.9 mg), 62%, yellow crystals, mp 199 °C. IR (KBr,
1221 m, 1151 m, 1119 w, 1040 m, 854 w, 753 s. H NMR
3
3
6
(DMSO-d ): δ ) 5.53 (s, 2H, CH), 6.66 (dd, J ) 7.5 Hz, J )
-
1
8.4 Hz, 2H, ArH-5), 6.90 (m, 6H, ArH-3, Ar′H-3, Ar′H-5), 6.98
cm ): νj ) 2937 m (OCH
3
), 2600 m, br (OH), 1627 s (CdN),
4
3
3
1
1
600 m, 1594 m, 1492 s, 1460 m, 1277 s (OH), 1246 s, 1150 m,
(ddd, J ) 1.4 Hz, J ) 8.4 Hz, J ) 7.5 Hz, 2H, ArH-4), 7.28
(m, 6H, ArH-5, Ar′H-4, Ar′H-6), 8.43 (s, 2H, NCH), 9.63 (s,
2H, Ar′-OH), 13.61 (s, 2H, Ar-OH).
1
050 m, 1028 m, 849 m, 751 s, 664 w. H NMR (DMSO-d
6
): δ
)
3.79 (s, 6H, OCH ), 5.44 (s, 2H, CH), 6.83 (m, 6H, ArH-3,
3
4
3
ArH-5, Ar′H-5), 6.94 (dd, J ) 1.4 Hz, J ) 7.5 Hz, 2H, ArH-
meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-hydroxy-
phenyl)-2,5-diazahexa-1,5-diene (m-5). m-5 was obtained
from meso-1,2-diamino-1,2-bis(3-hydroxyphenyl)ethane (4.08
mmol, 987.6 mg) and salicylaldehyde (8.24 mmol, 1004.7 mg).
Yield: 2.24 mmol (1010.3 mg), 55%, yellow powder, mp 232
4
1
), 7.27 (m, 6H, ArH-6, Ar′H-4, Ar′H-6), 8.38 (s, 2H, NCH),
3.62 (s, 2H, OH).
meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-methoxy-
phenyl)-2,5-diazahexa-1,5-diene (m-2). m-2 was obtained
from meso-1,2-diamino-1,2-bis(3-methoxyphenyl)ethane (1.00
-
1
°C. IR (KBr, cm ): νj ) 2600 m, br (OH), 1627 s(CdN), 1491

5
844 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24
Gust et al.
m, 1456 m, 1043 w, 1276 m (OH), 1216 m, 1154 m, 997 w, 758
[meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-methoxy-
phenyl)-2,5-diazahexa-1,5-diene]cobalt(II) (m-8). m-8 was
obtained from meso-1,6-bis(2-hydroxyphenyl)-3,4-bis(3-meth-
oxyphenyl)-2,5-diazahexa-1,5-diene (m-2) (0.32 mmol, 144.3
1
6
m, 700 w. H NMR (DMSO-d ): δ ) 4.89 (s, 2H, CH), 6.58
4
3
(
dd, J ) 1.7 Hz, J ) 7.4 Hz, 2H, ArH-3), 6.81 (m, 8H, ArH-5,
3 4
Ar′H-2, Ar′H-4, Ar′H-6), 7.05 (dd, J ) 7.8 Hz, J ) 1.7 Hz,
ArH-6), 7.30 (m, 4H, ArH-4, Ar′H-5), 8.37 (s, 2H, Ar-OH),
3.22(s, 2H, Ar-OH).
mg) and (cobalt(II) acetate)‚4H
2
O (0.36 mmol, 88.7 mg).
1
Yield: 0.23 mmol (124.9 mg), 72%, orange powder, mp 272
-
1
d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-hydroxyphenyl)-
,5-diazahexa-1,5-diene (d,l-5). d,l-5 was obtained from d,l-
3
°C. IR (KBr, cm ): νj ) 2982 w (OCH ), 1603 s (CdN), 1528
2
m, 1489 m, 1444 m, 1316 m, 1262 m, 1209 w, 1153 m, 1042
1
,2-diamino-1,2-bis(3-hydroxyphenyl)ethane (0.59 mmol, 143.2
m, 911 w, 760 m, 717 m. MS (EI, 215 °C): m/z (%) ) 537 (100)
+
•
1
+
mg) and salicylaldehyde (1.33 mmol, 162.8 mg). Yield: 0.51
mmol (231.5 mg), 87%, yellow powder, mp 218 °C. IR (KBr,
[M] , 298 (50) [( /
H, N.
2 26 2 4 2
Lig)Co] . Anal. (C30H N O Co‚0.5H O) C,
-
1
cm ): νj ) 1627 s (CdN), 1532 m, 1489 s, 1457 s, 1399 m,
[d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-methoxyphenyl)-
2,5-diazahexa-1,5-diene]cobalt(II) (d,l-8). d,l-8 was ob-
tained from d,l-1,6-bis(2-hydroxyphenyl)-3,4-bis(3-methoxy-
phenyl)-2,5-diazahexa-1,5-diene (d,l-2) (0.28 mmol, 73.9 mg)
1
275 s (OH), 1218 s, 1152 s, 1118 m, 1053 m, 1030 m, 758 s,
1
7
6
03 m. H NMR (DMSO-d ): δ ) 4.92 (s, 2H, CH), 6.57 (dd,
4
3
J ) 2.0 Hz, J ) 8.1 Hz, 2H, ArH-3), 6.75 (m, 4H, Ar′H-2,
Ar′H-4), 6.84 (m, 4H, ArH-5, Ar′H-6), 7.01 (m, 2H, ArH-6), 7.30
and (cobalt(II) acetate)‚4H
2
O (0.30 mmol, 73.9 mg). Yield: 0.22
(
m, 4H, ArH-4, Ar′H-5), 8.51 (s, 2H, NCH), 9.33 (s, 3H, Ar′-
mmol (118.8 mg), 79%, pale-red powder, mp 260 °C. IR (KBr,
-
1
OH), 13.34 (s, 2H, Ar-OH).
cm ): νj ) 2834 w (OCH
3
), 1604 s (CdN), 1527 m, 1491 m,
1
447 m, 1320 m, 1263 m, 1209 w, 1151 m, 1043 m, 915 w, 760
m. MS (EI, 250 °C): m/z (%) ) 537 (100) [M] , 298 (57)
meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(4-hydroxy-
phenyl)-2,5-diazahexa-1,5-diene (m-6). m-6 was obtained
from meso-1,2-diamino-1,2-bis(4-hydroxyphenyl)ethane (4.05
mmol, 980.8 mg) and salicylaldehyde (8.14 mmol, 993.0 mg).
Yield: 2.41 mmol (1090.6 mg), 60%, orange powder, mp 224
+
•
1
+
[
( /
2 26 2 4
Lig)Co] . Anal. (C30H N O Co) C, H, N.
[meso-1,6-Bis(2-hydroxyphenyl-3,4-bis(4-methoxy-
phenyl)-2,5-diazahexa-1,5-diene]cobalt(II) (m-9). m-9 was
obtained from meso-1,6-bis(2-hydroxyphenyl-3,4-bis(4-meth-
oxyphenyl)-2,5-diazahexa-1,5-diene (m-3) (0.69 mmol, 332.8
-
1
°
C. IR (KBr, cm ): νj ) 2600 w, br (OH), 1625 s (CdN), 1513
m, 1459 m, 1274 m (OH), 1221 m, 1173 m, 1115 w, 1051 w,
1
1
017 w, 837 m, 759 m. H NMR (DMSO-d
6
): δ ) 4.85 (s, 2H,
mg) and (cobalt(II) acetate)‚4H
2
O (0.72 mmol, 180.2 mg).
3
CH), 6.67 (d, J ) 8.4 Hz, 4H, Ar′H-3, Ar′H-5), 6.84 (m, 4H,
Yield: 0.47 mmol (255.1 mg), 68%, red powder, mp 295 °C. IR
3
-1
ArH-3, ArH-5), 7.09 (d, J ) 8.4 Hz, 4H, Ar′H-2, Ar′H-6), 7.30
3
(KBr, cm ): νj ) 2835 w (OCH ), 1606 s (CdN), 1514 s, 1442
(
m, 4H, ArH-4, ArH-6), 8.38 (s, 2H, NCH), 9.31 (s, 2H, Ar′-
s, 1382 w, 1308 m, 1252 s, 1208 m, 1179 s, 1150 m, 1033 m,
989 w, 908 m, 826 m, 807 m, 759 m, 629 w. MS (EI, 280 °C):
OH), 13.41 (s, 2H, Ar-OH).
+
•
1
+
m/z (%) ) 537 (100) [M] , 298 (59) [( /
Co‚0.5H O) C, H, N.
[d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(4-methoxyphenyl)-
2
Lig)Co] . Anal.
d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(4-hydroxyphenyl)-
(
30
C H
26
2
N O
4
2
2
,5-diazahexa-1,5-diene (d,l-6). d,l-6 was obtained from d,l-
1
,2-diamino-1,2-bis(4-hydroxyphenyl)ethane (0.53 mmol, 129.8
mg) and salicylaldehyde (1.16 mmol, 141.4 mg). Yield: 0.18
mmol (82.0 mg), 34%, yellow powder, mp 203 °C. IR (KBr,
2,5-diazahexa-1,5-diene]cobalt(II) (d,l-9). d,l-9 was ob-
tained from d,l-1,6-bis(2-hydroxyphenyl)-3,4-bis(4-methoxy-
phenyl)-2,5-diazahexa-1,5-diene (d,l-3) (277.0 mmol, 133.0 mg)
-
1
cm ): νj ) 2600 m, br (OH), 1625 s (CdN), 1514 s, 1456 s,
1
8
386 m, 1276 s (OH), 1219 s, 1152 s, 1117 m, 1030 m, 904 m,
and (cobalt(II) acetate)‚4H
2
O (0.29 mmol, 72.2 mg). Yield: 0.21
1
6
33 s, 755 s. H NMR (DMSO-d ): δ ) 4.88 (s, 2H, CH), 6.62
mmol (114.1 mg), 77%, red powder, mp 269-275 °C. IR (KBr,
3
-1
(
d, J ) 8.5 Hz, 4H, Ar′H-3, Ar′H-5), 6.83 (m, 4H, ArH-3, ArH-
3
cm ): νj ) 2835 w (OCH ), 1606 s (CdN), 1585 s, 1513 s, 1466
3
5), 7.07 (d, J ) 8.5 Hz, 4H, Ar′H-2, Ar′H-6), 7.29 (m, 4H, ArH-
4, ArH-6), 8.48 (s, 2H, NCH), 9.28 (s, 2H, Ar′-OH), 13.41 (s,
2H, Ar-OH).
m, 1444 s, 1308 m, 1252 s, 1177 m, 1150 m, 1030 m, 912 w,
833 w, 759 m, 617 w. MS (EI, 240 °C): m/z (%) ) 537 (100)
+
•
[M] . Anal. (C30
26 2 4 2
H N O Co‚0.5H O) C, H, N.
General Procedure for the Synthesis of [1,6-Bis(2-
[meso-1,3,4,6-Tetrakis(2-hydroxyphenyl)-2,5-diazahexa-
1,5-diene]cobalt(II) (m-10). m-10 was obtained from meso-
1,3,4,6-tetrakis(2-hydroxyphenyl)-2,5-diazahexa-1,5-diene (m-
hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-1,5-dienes]co-
balt(II). An amount of 0.3 mmol of the respective 1,6-bis(2-
hydroxyphenyl)-3,4-diaryl-2,5-diazahexa-1,5-diene was sus-
pended in 5 mL of ethanol and heated to 60-80 °C under an
argon atmosphere. The suspension of 0.3 mmol of cobalt(II)
acetate in 5 mL of ethanol was heated on a water bath till the
formation of a precipitate was completed. Subsequently, both
mixtures were combined and stirred for 1 h. The colored
4) (0.50 mmol, 227.5 mg) and (cobalt(II) acetate)‚4H
2
O (0.54
mmol, 75.0 mg). Yield: 0.39 mmol (177.2 mg), 78%, brownish
-1
powder, mp 270 °C. IR (KBr, cm ): νj ) 1636 s, 1601 s (CdN),
1540 s, 1474 s, 1450 s, 1408 m, 1286 s, 1240 m, 1152 m, 1129
m, 1038 m, 903 m, 855 m, 754 s, 658 w. MS (FAB, negative
-
•
(CH OH)): m/z (%) ) 509 (100) [M] . Anal. (C28
3
22 2 4
H N O Co‚
crystals were sucked off under nitrogen, dried over CaSO
stored under an argon atmosphere.
4
, and
1.5H O) C, H, N.
2
[d,l-1,3,4,6-Tetrakis(2-hydroxyphenyl)-2,5-diazahexa-
1,5-diene]cobalt(II) (d,l-10). d,l-10 was obtained from d,l-
1,3,4,6-tetrakis(2-hydroxyphenyl)-2,5-diazahexa-1,5-diene (d,l-
[
meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(2-methoxy-
phenyl)-2,5-diazahexa-1,5-diene]cobalt(II) (m-7). m-7 was
obtained from meso-1,6-bis(2-hydroxyphenyl)-3,4-bis(2-meth-
oxyphenyl)-2,5-diazahexa-1,5-diene (m-1) (0.30 mmol, 145.8
4) (0.18 mmol, 79.8 mg) and (cobalt(II) acetate)‚4H
2
O (0.19
mmol, 47.1 mg). Yield: 0.12 mmol (62.7 mg), 69%, red powder,
-
1
mg) and (cobalt(II) acetate)‚4H
2
O (0.33 mmol, 82.0 mg).
mp >300 °C. IR (KBr, cm ): νj ) 2600 m, br (OH), 1641 s,
1601 s (CdN), 1536 m, 1472 s, 1450 s, 1513 s, 1283 s, 1207 m,
1152 m, 1130 m, 1036 w, 903 m, 753 s. MS (EI, 300 °C): m/z
Yield: 0.20 mmol (106.3 mg), 65%, red crystals, mp 275 °C.
-
1
IR (KBr, cm ): νj ) 2837 w (OCH
491 m, 1462 s, 1443 s, 1382 w, 1314 m, 1246 s, 1208 m, 1150
m, 1051 m, 1025 m, 957 w, 910 w, 870 w, 756 w, 756 s, 647 w.
3
), 1602 s (CdN), 1527 s,
+
•
1
+
1
(%)
)
509 (2) [M] , 283 (3) [( /
2
Lig)Co] , 225 (24)
1
+
[( / Lig)] . Anal. (C28H N O Co) C, H, N.
2 22 2 4
+
•
MS (EI, 120 °C): m/z (%) ) 537 (51) [M] . Anal. (C30
Co‚0.5H O) C, H, N.
d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(2-methoxyphenyl)-
H
26
N
2
O
4
-
[meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-hydroxy-
phenyl)-2,5-diazahexa-1,5-diene]cobalt(II) (m-11). m-11
was obtained from meso-1,6-bis(2-hydroxyphenyl)-3,4-bis(3-
hydroxyphenyl)-2,5-diazahexa-1,5-diene (m-5) (0.70 mmol,
2
[
2
,5-diazahexa-1,5-diene]cobalt(II) (d,l-7). d,l-7 was ob-
tained from d,l-1,6-bis(2-hydroxyphenyl)-3,4-bis(2-methoxy-
phenyl)-2,5-diazahexa-1,5-diene (d,l-1) (0.42 mmol, 200.1 mg)
2
315.4 mg) and (cobalt(II) acetate)‚4H O (0.73 mmol, 181.1 mg).
Yield: 0.51 mmol (257.1 mg), 72%, red powder, mp >300°C.
-1
and (cobalt(II) acetate)‚4H
2
O (0.44 mmol, 110.6 mg). Yield:
IR (KBr, cm ): νj ) 1604 s (CdN), 1530 s, 1448 s, 1381 w, 1303
0
.28 mmol (152.2 mg), 66%, red powder, mp >300 °C. IR (KBr,
m, 1272 m, 1212 m, 1153 m, 1049 w, 928 w, 908 w, 760 m,
-
1
719 w, 628 w. MS (EI, 340 °C): m/z (%) ) 509 (23) [M]+•, 284
cm ): νj ) 2836 w (OCH
3
), 1602 s (CdN), 1528 s, 1491 s, 1464
1
+
m, 1444 s, 1315 m, 1246 s, 1150 m, 1051 m, 1025 m, 912 w,
2 22 2 4
(23) [( / Lig)Co] . Anal. (C28H N O Co) C, H, N.
+
•
8
50 w, 755 s. MS (EI, 300 °C): m/z (%) ) 537 (48) [M] . Anal.
[d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(3-hydroxyphenyl)-
2,5-diazahexa-1,5-diene]cobalt(II) (d,l-11). d,l-11 was ob-
(
C
30
H
26
N
2
O
4
‚0.5H O) C, H, N.
2

Cobalt(3,4-diarylsalen) Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24 5845
tained from d,l-1,6-bis(2-hydroxyphenyl)-3,4-bis(3-hydroxy-
phenyl)-2,5-diazahexa-1,5-diene (d,l-5) (0.33 mmol, 151.3 mg)
incubation time. Subsequently, the cells were fixed with a
glutaric dialdehyde solution and stored under phosphate
buffered saline (PBS) at 4 °C. Cell biomass was determined
2
and (cobalt(II) acetate)‚4H O (0.36 mmol, 88.7 mg). Yield: 0.27
mmol (140.8 mg), 81%, orange powder, mp >300 °C. IR (KBr,
by means of a crystal violet staining technique as described
-
1
17,27
cm ): νj ) 1602 s (CdN), 1531 s, 1446 s, 1381 w, 1446 s, 1302
earlier.
The effectiveness of the complexes is expressed
m, 1210 m, 1152 m, 999 w, 916 w, 757 m. MS (EI, 340 °C):
as corrected % T/Ccorr or % τ values according to the following
+
•
1
+
m/z (%) ) 509 (100) [M] , 283 (60) [( /
2
Lig)Co] . Anal.
equations:
(
C
28
H
22
N
2
O
4
Co‚0.5H
2
O) C, H, N.
[
meso-1,6-Bis(2-hydroxyphenyl)-3,4-bis(4-hydroxy-
T - C0
phenyl)-2,5-diazahexa-1,5-diene]cobalt(II) (m-12). m-12
was obtained from meso-1,6-bis(2-hydroxyphenyl)-3,4-bis(4-
hydroxyphenyl)-2,5-diazahexa-1,5-diene (m-6) (0.71 mmol,
cytostatic effect: T/Ccorr [%] )
× 100
C - C0
T - C0
3
21.8 mg) and (cobalt(II) acetate)‚4H
2
O (0.75 mmol, 186.0 mg).
cytocidal effect: τ [%] )
× 100
Yield: 0.56 mmol (283.3 mg), 78%, red powder, mp >300 °C.
C0
-
1
IR (KBr, cm ): νj ) 1606 s (CdN), 1516 s, 1443 s, 1381 w,
304 m, 1256 m, 1233 m, 1175 m, 1152 m, 1028 w, 907 w, 814
1
+
•
where T (test) and C (control) are the optical densities at 590
nm of the crystal violet extract of the cells in the wells (i.e.,
the chromatin-bound crystal violet extracted with ethanol
m, 760 m, 658 w. MS (EI, 400 °C): m/z (%) ) 509 (100) [M] ,
1
+
1
+
2
0
84 (95) [( /
.5H O) C, H, N.
d,l-1,6-Bis(2-hydroxyphenyl)-3,4-bis(4-hydroxyphenyl)-
2 2 22 2 4
Lig)Co] , 225 (26) [ / Lig] . Anal. (C28H N O Co‚
2
7
0
0%), and C is the density of the cell extract immediately
[
before treatment.
2
,5-diazahexa-1,5-diene]cobalt(II) (d,l-12). d,l-12 was ob-
For the automatic estimation of the optical density of the
crystal violet extract in the wells, a Microplate EL 309
autoreader was used.
tained from d,l-1,6-bis(2-hydroxyphenyl)-3,4-bis(4-hydroxy-
phenyl)-2,5-diazahexa-1,5-diene (d,l-12) (0.17 mmol, 78.1 mg)
2
and (cobalt(II) acetate)‚4H O (0.18 mmol, 45.2 mg). Yield: 0.14
Accumulation Studies. The accumulation studies were
mmol (73.6 mg), 82%, pale-red powder, mp >300 °C. IR (KBr,
4
,20
-
1
performed according to methods already described by us.
cm ): νj ) 2600 w, br (OH), 1604 s (CdN), 1531 s, 1469 m,
445 s, 1304 m, 1264 m, 1209 s, 1173 m, 1152 m, 997 w, 912
w, 838 m, 758 m, 621 w. MS (EI, 330 °C): m/z (%) ) 509 (100)
Briefly, for the cobalt accumulation studies, cells were grown
1
2
in 75 cm culture flasks (Falcon Plastics 3028) in a humidified
+
•
1
+
atmosphere at 37 °C. When the cells had approached conflu-
ency, the culture medium was aspirated and replaced with a
fresh medium containing 5 µM cobalt complex or 0.1% DMF
[
M] , 284 (83) [( /
H, N.
Assay To Verify the Catalytic Property of Cobalt(3,4-
diarylsalen) Complexes. O was triggered for 2 h in a
methanol solution consisting of 2,6-di-tert-butylphenol and the
cobalt(3,4-diarylsalen) complex. The analysis took place after
having separated the solvent with the help of the tert-butyl
group integrals in the H NMR spectrum (δ(BQ) ) 1.28, δ-
DPQ) ) 1.37, δ(DP) ) 1.45).
Biological Methods. Cell Culture. The human MCF-7
2 22 2 4 2
Lig)Co] . Anal. (C28H N O Co‚0.5H O) C,
(
blank) (50 µL of stock solution was added to 50 mL of culture
2
medium). Twelve culture flasks were used for the blanks as
well as for the incubation range 20-360 min. At appropriate
intervals (0, 20, 40, 60, 80, 100, 120, 180, 240, 300, and 360
min), the flasks were removed from the incubator, the medium
was aspirated, and the cells were washed with PBS and
harvested after treatment with 1-2 mL of trypsin (0.05%) and
EDTA (0.02%) in PBS. For the complete removal of the cells,
the culture flasks were rinsed with an additional 7 mL of PBS.
At each time point, the cells were pooled and centrifuged at
1
(
and MDA-MB 231 breast cancer cell lines as well as the
LNCaP/FGC cell line were obtained from the American Type
Culture Collection (ATCC). Cell line banking and quality
control were performed according to the seed stock concept
reviewed by Hay.2 The MCF-7 cells were maintained in
L-glutamine containing Eagle’s MEM (Sigma, Germany),
2
000 U/min for 5 min at 15 °C. The pellet was twice
resuspended in 8 mL of PBS and centrifuged. Finally, the
washing solution was decanted from the cell pellet which was
treated with 900 µL of TA solution (200 µL of Triton X-100 +
6
3
supplemented with NaHCO (2.2 g/L), sodium pyruvate (110
2
00 µL of Antifoam B (Sigma) diluted in 100 mL of water).
mg/L), gentamycin (50 mg/L), and 10% fetal calf serum (FCS;
After sonification, the homogeneous cell suspensions (about
2
Gibco, Germany) using 75 cm culture flasks in a humidified
8
00 µL) were used for cobalt determination and protein
2
atmosphere (5% CO ) at 37 °C. The MDA-MB 231 cells
measurement.
(
3
McCoy’s 5A medium supplemented with NaHCO (2.2 g/L),
The Co determinations were performed by flameless atomic
absorption spectroscopy (wavelength, 240.7 nm; apparatus,
Philips PU 9100x atomic absorption spectrophotometer with
graphite furnace TJA-Unicam). An amount of 20 µL of an
appropriately diluted sample was introduced into the atomizer.
sodium pyruvate (110 mg/L), gentamycin (50 mg/L), and 5%
FCS) and the LNCaP/FGC cells (L-glutamine containing RPMI
1
3
640, supplemented with NaHCO (2.0 g/L), gentamycin (50
mg/L), and 7.5% FCS) were maintained under the same
conditions. The cell lines were passaged weekly after previous
treatment with trypsin (0.05%)/ethylenediaminetetraacetic
acid (0.02% EDTA; Boehringer, Germany). Mycoplasma con-
tamination was routinely monitored, and only mycoplasma-
free cultures were used.
As Co standard, we used dilutions of Co(NO
3 2 2
) ‚H O in 0.05
M HNO (c(Co) ) 1.00 g/L). The calibration curve showed a
3
linear run in the low-concentration range of 1.0-10.0 µM. The
Co values were calculated by means of this calibration curve.
Each Co analysis was based on no less than three individual
determinations. The detectability was about 0.08 µM Co for a
In Vitro Chemosensitivity Assays. The in vitro testing
of the cobalt complexes for antitumor activity was carried out
on exponentially dividing human cancer cells according to a
1
0 µL probe, and the precision of the method was (3.0%.
Correlation of cell number and protein content and the
calculation of cellular cobalt concentration were performed
according to Reile et al. For this purpose, 5 µL of a cell
suspension and 95 µL of water were combined with 1 mL of
1
7,27
previously published microtiter assay.
Exponential cell
growth is guaranteed during the whole time of incubation.
Briefly, by utilization of 96-well microtiter plates, 100 µL of a
cell suspension were plated into each well at 7700 cells/mL
culture medium (MCF-7), at 3200 cells/mL (MDA-MB 231),
and at 2700 cells/mL (LNCaP/FGC), incubated at 37 °C for 3
or 5 days in the case of LNCaP/FGC cells in a humidified
2
0
2
8
Bradford reagent. After 20 min incubation at room temper-
ature, the extinction was measured in a UV spectrometer at
5
95 nm. For calibration, BSA (0.1, 0.2, ..., 1.0 µg/µL) was
dissolved in PBS. The diameters of trypsinized MCF-7 cells
2
atmosphere (5% CO ). By addition of an adequate volume of a
2
0
(
n ) 50) as determined by Reile et al. were used for the
stock solution of the respective compound (solvent: DMF) to
the medium, the desired test concentration was obtained.
Sixteen wells were used for each test concentration and for
the control, which contained the corresponding amount of
DMF. The medium was removed after reaching the proper
calculation of cellular cobalt concentration.
Acknowledgment. The technical assistance of S.
Bergemann and I. Schnautz is acknowledged.

5
846 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 24
Supporting Information Available: IR-spectra of d,l-3
Gust et al.
(14) Van Dort, H. M.; Geursen, H. J. Salcomine-catalyzed oxidations
of some phenols. A new method for the preparation of a number
of p-benzoquinones. Recl. Trav. Chim. Pays-Bas 1967, 86, 520-
and d,l-9, typical mass spectra of the cobalt(3,4-diarylsalen)
on d,l-9, and graphite furnace atomic absorption temperature
program. This material is available free of charge via the
Internet at http://pubs.acs.org.
5
26.
(
15) Nishinaga, A.; Tomita, H.; Tarumi, Y.; Matsuura, T. Reaction
of superoxo cobalt(III) complexes with 2,6-di-tert-butyl-p-benzo-
quinone methides. Tetrahedron Lett. 1980, 21, 4849-4852.
16) Zombeck, A.; Drago, R. S.; Corden, B. B.; Gaul, J. H. Activation
of molecular oxygen. Kinetic studies of the oxidation of hindered
phenols with cobalt-dioxygen complexes. J. Am. Chem. Soc.
(
References
(
(
(
(
1) Rosenberg, B.; VanCamp, B.; Krigas, T. Inhibition of cell division
in Escherichia coli by electrolysis products from a platinum
electrode. Nature 1965, 205, 698-699.
1
981, 103, 7580-7585.
(
17) Bernhardt, G.; Reile, H.; Birnb o¨ ck, H.; Spruss, T.; Sch o¨ nen-
berger, H. Standardized kinetic microassay to quantify dif-
ferential chemosensitivity on the basis of proliferative activity.
J. Res. Clin. Oncol. 1992, 118, 35-43.
18) Bernhardt, B.; Koch, M.; Spruss, T.; Gust, R.; Krauser, R.;
Schlemmer, R.; Sch o¨ nenberger H. [meso-1,2-Bis(2,6-dichloro-4-
hydroxyphenyl)ethylenediamine]sulfatoplatinum(II). Pharma-
cokinetic Studies. Arch. Pharm. (Weinheim, Ger.) 1999, 332,
2) Rosenberg, B.; VanCamp, L.; Trosko, J. E.; Mansour, V. H.
Platinum compounds: a new class of potent antitumour agents.
Nature 1969, 385-386.
(
3) Schmidt, K.; Jung, M.; Keilitz, R.; Schnurr, B.; Gust, R. Acetyl-
enehexacarbonyldicobalt complexes, a novel class of antitumor
drugs. Inorg. Chim. Acta 2000, 306, 6-16.
4) Ott, I.; Kirchner, B.; Gust, R. Investigations on the effects of
cobalt-alkyne complexes on leukemia and lymphoma cells:
cytotoxicity and cellular uptake. J. Inorg. Biochem. 2004, 98,
1
95-200.
(
19) Boulikas, T.; Vougiouka, M. Cisplatin and platinum drugs at
the molecular level (review). Oncol. Rep. 2003, 10, 1663-1682.
20) Reile, H.; Bernhardt, G.; Koch, M.; Sch o¨ nenberger, H.; Hollstein,
M.; Lux, F. Chemosensitivity of human MCF-7 breast cancer
cells to diastereomeric diaqua(1,2-diphenylethylenediamine)-
platinum(II) sulfates and specific platinum accumulation. Cancer
Chemother. Pharmacol. 1992, 30, 113-122.
21) Gall, R. S.; Rogers, J. F.; Schaefer, W. P.; Christoph, G. G. The
structure of a monomeric oxygen carrying cobalt complex: di-
oxygen-N,N′-(1,1,2,2-tetramethyl)ethylenebis(3-tert-butylsalicyl-
ideniminato)(1-benzylimidazole)cobalt(II). J. Am. Chem. Soc.
485-489.
(
(
5) Niederhoffer, E. C.; Timmons, J. H.; Martell, A. E. Thermody-
namics of oxygen binding in natural and synthetic dioxygen
complexes. Chem. Rev. 1984, 84, 137-146.
(
6) Nackerdien, Z.; Kasprzak, K. S.; Rao, G.; Halliwell, B.; Dizdaro-
glu, M. Nickel(II)- and cobalt(II)-dependent damage by hydrogen
peroxide to the DNA bases in isolated human chromatin. Cancer
Res. 1991, 51, 5837-5842.
(
(
7) Robison, S. H.; Cantoni, O.; Costa, M. Strand breakage and
decreased molecular weight of DNA induced by specific metal
compounds. Carcinogenesis 1982, 3, 657-662.
1
976, 98, 5135-5138.
(
22) Avdeef, A.; Schaefer, W. P. Reversible oxygen carriers. The
synthesis and low temperature (-171â) structure of an un-
stable monomeric dioxygen adduct of N,N′-(1,1,2,2-tetra-
methyl)ethylenebis(3-fluorosalicylideniminato)(1-methylimid-
azole)cobalt(II), Co(3-F-Saltmen)(1-Me-Imid)(O2).2Me2CO. J.
Am. Chem. Soc. 1976, 98, 5153-5161.
23) Br u¨ ckner, S.; Calligaris, M.; Nardi, G.; Randaccio, L. Crystal
and molecular structure of (1-diacetyl-monoximatoimino-3-di-
acetylmonoximeimino-propane)methylaquocobalt(III) perchlor-
ate. Inorg. Chim. Acta 1969, 3, 308-330.
24) Liu, G.-D.; Yang, X.; Chen, Z.-P.; Shen, G.-L.; Yu, R.-Q. Interac-
tion of metal complexes of bis(salicylidene)ethylenediamine with
DNA. Anal. Sci. 2000, 16, 1255-1259.
(
8) Jennerwein, M.; Wappes, B.; Gust, R.; Sch o¨ nenberger, H.; Engel,
J.; Seeber, S.; Osieka, R. Influence of ring substituents on the
antitumor effect of dichloro(1,2-diphenylethylenediamine)plati-
num(II) complexes. J. Cancer Res. Clin. Oncol. 1998, 114, 347-
358.
(
9) M u¨ ller, R.; Gust, R.; Bernhardt, G.; Keller, C.; Sch o¨ nenberger,
H.; Seeber, S.; Osieka, R.; Eastman, A.; Jennerwein, M. DL-1,2-
bis(2-hydroxyphenyl)ethylenediamine]dichloroplatinum(II), a new
compound for the therapy of ovarian cancer. J. Cancer Res. Clin.
Oncol. 1990, 116, 237-244.
(
(
(
(
(
(
10) Schertl, S.; Gust, R.; M u¨ ller, R.; Spruss, T.; Sch o¨ nenberger, H.
Stereoisomeric [1,2-bis(3-hydroxyphenyl)ethylenediamine]plat-
inum(II) complexes, Part III: Evaluation of the mammary tumor
inhibiting properties. Arch. Pharm. (Weinheim, Ger.) 1992, 325,
(
25) Routier, S.; Joanny, V.; Zaparucha, A.; Vezin, H.; Catteau, J.-
P.; Bernier, J.-L.; Bailly, C. Synthesis of metal complexes of 2,9-
bis(2-hydroxyphenyl)-1,10-phenantroline and their DNA binding
and cleaving activities. J. Chem. Soc., Perkin Trans. 2 1998,
113-118.
11) Spruss, T.; Gust, R.; M u¨ ller, R.; Engel, J.; Sch o¨ nenberger, H.
Mammary tumor inhibiting properties of the (S,S)-config-
urated [1,2-bis(4-hydroxyphenyl)ethylenediamine]dichloroplat-
inum(II) complex. Arch. Pharm. (Weinheim, Ger.) 1990, 323, 99-
8
63-868.
(
26) Hay, R. J. The seed stock concept and quality control for cell
lines. Anal. Biochem. 1988, 171, 225-237.
102.
(27) Reile, H.; Birnb o¨ ck, H.; Bernhardt, G.; Spruss, T.; Sch o¨ nen-
berger, H. Computerized determination of growth kinetic curves
and doubling times from cells in microculture. Anal. Biochem.
1990, 187, 262-267.
(28) Bradford, M. M. A rapid and sensitive method for the quanti-
tation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 1976, 72, 248-254.
12) Jennerwein, M.; Gust, R.; M u¨ ller, R.; Sch o¨ nenberger, H.; Engel,
J.; Berger, M. R.; Schm a¨ hl, D.; Seeber, S.; Osieka, R.; Atassi, G.
Tumor inhibiting properties of stereoisomeric [1,2-bis(3-hydroxy-
phenyl)ethylenediamine]dichloroplatinum(II) complexes, Part
I: Synthesis. Arch. Pharm. (Weinheim, Ger.) 1989, 322, 25-29.
13) Barkelew, C. H.; Calvin, M. The oxygen-carrying synthetic
chelate compounds. II. The rates of oxygenation of the solid
compounds. J. Am. Chem. Soc. 1946, 68, 2257-2262.
JM040763N