Synthesis, electrochemistry, and bioactivity of the cyanobacterial calothrixins and related quinones

Article abstract of DOI:10.1021/jm049625o

The calothrixins are quinone-based natural products isolated from Calothrix cyanobacteria which show potent antiproliferative properties against several cancer cell lines. Preliminary mechanistic studies suggest that the biological mode of action of the c

Full text of DOI:10.1021/jm049625o

4958
J . Med. Chem. 2004, 47, 4958-4963
Syn th esis, Electr och em istr y, a n d Bioa ctivity of th e Cya n oba cter ia l Ca loth r ixin s
a n d Rela ted Qu in on es
Paul H. Bernardo,† Christina L. L. Chai,*,† Graham A. Heath,‡ Peter J . Mahon,‡ Geoffrey D. Smith,§
Paul Waring,† and Bronwyn A. Wilkes†
Department of Chemistry and School of Biochemistry and Molecular Biology, Centre for the Study of Bioactive Molecules,
Australian National University, ACT 0200, Australia, and Research School of Chemistry, Australian National University,
ACT 0200, Australia
Received May 21, 2004
The calothrixins are quinone-based natural products isolated from Calothrix cyanobacteria
which show potent antiproliferative properties against several cancer cell lines. Preliminary
mechanistic studies suggest that the biological mode of action of the calothrixins may be linked
to their ability to undergo redox cycling. In this study we compare the bioactivities of the
calothrixins with those of structurally related quinones in order to identify the structural
features in the calothrixins essential for biological activity. In particular, the reduction potentials
of the calothrixins and some related quinones were measured electrochemically. Our studies
indicate that while there is no direct correlation between the reduction potentials and biological
activities of the studied compounds, in all cases quinones with EC50 values <1.6 µM undergo
reduction to their respective semiquinones readily, with their E1/2 values being more positive
than -0.5 V versus the standard hydrogen electrode (SHE).
In tr od u ction
species.12 The present study was designed to investigate
structure-activity relationships of the calothrixins and
closely related quinones. In view of the apparent redox-
active propensities of calothrixin A, we were interested
in assessing whether there was a correlation between
the redox potentials and the biological activities. To
assess the contribution of the overall structure of the
calothrixins, the reduction potential and biological
activities of a number of structural analogues were also
compared. This is the first reported structure-activity
study of calothrixins and their analogues. In addition,
to our knowledge, this paper also reports the first
electrochemical study of the calothrixins 1-3 as well
as that of their tetracyclic analogues, the ellipticine
quinones 4 and 5.
The quinones are one of the most important classes
of antitumor agents.1,2 Their biological modes of action
differ according to their particular structure, and they
can act as DNA intercalators, bioreductive alkylators
of biomolecules, and/or generators of reactive oxygen
species through redox cycling.3-5 Biological redox activ-
ity is recognized as playing a key role in a number of
processes, including the triggering of cellular events that
can be exploited for therapeutic uses.6 For a number of
quinone systems, a relationship between biological
activity and redox potential has been suggested,7 as, for
example, between the redox potentials of azaanthraquino-
nes and their inhibitory effects on Epstein-Barr virus
early antigen activation,8 or the cytotoxicity of naturally
occurring quinones against human breast adenocarci-
noma.9 Other examples include the higher cytotoxic
potency of mitomycin A versus mitomycin C, which has
been tentatively correlated with the large difference in
the quinone redox potentials of the two drugs.10
In 1999, a new class of quinones known as calothrix-
ins A (1) and B (2) was isolated from the cyanobacteria
Calothrix.11 Calothrixin A (1) and B (2) both inhibited
the in vitro growth of the chloroquine-resistant strain
of the human parasite, Plasmodium falciparum in a
dose-dependent manner.11 They were also found to kill
the human HeLa cancer cell line, with IC50 values at
nanomolar levels.11 Interestingly, subsequent investiga-
tions have shown that calothrixin A is redox-active and
induces the intracellular formation of reactive oxygen
* To whom correspondence should be addressed. Tel: 61-2-6125
4398. Fax: 61-2-6125 0760. E-mail: Christina.chai@anu.edu.au.
† Department of Chemistry, Centre for the Study of Bioactive
Molecules.
Resu lts a n d Discu ssion
Ch em istr y. For structure-activity studies, calothrix-
ins A (1) and B (2) and related derivative 3, as well as
structurally related analogues 4-8, were investigated.
‡ Research School of Chemistry.
§ School of Biochemistry and Molecular Biology, Centre for the Study
of Bioactive Molecules.
10.1021/jm049625o CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/27/2004

Cyanobacterial Calothrixins and Related Quinones
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20 4959
Sch em e 1
Ta ble 1. Electrochemical Reduction Potentials and ln K
reversible one-electron reduction to its semiquinone
radical anion, followed by a separate one-electron reduc-
tion to its dianion. In DMF, the first reduction process
has been observed to occur at a potential of -1.14 V
relative to Fc+/Fc, while in DMSO this reduction was
seen to occur at -1.05 V. This behavior was observed
as expected in the current work, and a reversible wave
was evident in its CV trace at an E1/2 of -1.09 V.
The calothrixins and synthetic tetracyclic analogues
exhibited behavior similar to that observed for mena-
dione. All of the quinones that were studied displayed
electrochemically reversible one-electron reductions to
their respective semiquinones, at potentials within 0.3
V of that of menadione. The observed CV of benzocar-
bazoledione (6) is consistent with that recently reported
in the literature (E1/2 ) -1.18 V relative to ferrocene)
using a glassy carbon working electrode.18
Values of Quinonesa
E1/2 (V) vs Fc+/Fc
∆E1/2
compound
calothrixin A (1)
wave I wave II (V)b ln Kc
-0.80
-0.87
-0.82
-1.09
-0.95
-1.17
-1.46
-1.54
-1.54
-1.67
-1.65
-1.84
-1.86
-1.83
0.66 26.1
0.67 26.5
0.72 28.5
0.58 23.0
0.70 27.7
0.67 26.5
0.71 28.1
0.74 29.3
calothrixin B (2)
N-MOM-calothrixin B (3)
ellipticine quinone (4)
N-MOM-ellipticine quinone (5)
benzocarbazoledione (6)
N-MOM-benzocarbazoledione (7) -1.15
menadione (8)
-1.09
a
Determined by ac and/or cyclic voltammetry at scan rates of
20 mV/s for ac and 50 mV/s for cyclic voltammetry experiments.
All measurements were carried out at 20 °C in high grade DMSO,
in the presence of tetrabutylammonium hexafluorophosphate (0.1
molar) as supporting electrolyte. Values vs the SHE may be
estimated by addition of +0.77 to the Fc/Fc+ values quoted. ∆E1/2
b
) E1/2(wave I) - E1/2 (wave II). c ln K ) (F/RT)(∆E1/2).
As a group, calothrixin A (1), calothrixin B (2), and
N-MOM-calothrixin B (3) were found to have similar
redox potentials, displaying relatively early reductions.
Calothrixin A was the easiest to reduce, while calo-
thrixin B was less facile by 70 mV. N-MOM-calothrixin
B (3) was easier to reduce than calothrixin B (2) by
approximately 50 mV. This may reflect the weakly
electron-withdrawing nature of the indolic methoxym-
ethyl substituent. Comparison of the first reduction
potential of N-MOM ellipticine quinone (5) with that of
ellipticine quinone (4), and also those of N-MOM-
benzocarbazoledione (7) and benzocarbazoledione (6),
showed similar trends in that the N-MOM derivative
was easier to reduce than the corresponding N-H
compound.
Comparisons of the E1/2 values of the first wave
potentials of N-MOM-calothrixin B (3) and N-MOM-
ellipticine quinone (5) showed that the former was
easier to reduce by 130 mV. The relative ease of
reduction of N-MOM-calothrixin B, compared with
N-MOM-ellipticine quinone, is interesting and can be
taken to reflect measurable lowering of the redox-active
Although calothrixin A and B can be sourced from
cyanobacteria, access to sufficient quantities of the
calothrixins 1 and 2 and derivative 3 was conveniently
achieved through chemical synthesis using a route
reported by us (Scheme 1).13,14 The related indoloquino-
nes, ellipticine quinone (4) and benzocarbazoledione (6)
and their derivatives, were also conveniently synthe-
sized following the route described in Scheme 1.
Electrochemical studies were performed on the quino-
nes dissolved in anhydrous dimethyl sulfoxide as sol-
vent, with tetra-n-butylammonium hexafluorophosphate
as the supporting electrolyte and a platinum disk
electrode as the working electrode. In these studies, both
cyclic voltammetry (CV) and alternating current volta-
mmetry (acV) were utilized. The E1/2 values for the first
reduction (wave I) and the second reduction (wave II)
are tabulated with reference to ferrocene (Fc) in Table
1.
The electrochemical behavior of menadione (8) has
been well studied in both organic and aqueous media.15-17
In aprotic solvents menadione is observed to undergo

4960 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20
Bernardo et al.
Ta ble 2. Cytotoxicity of Compounds toward HeLa Cell Lines
LUMO. The extended conjugated π-system of N-MOM-
calothrixin B may provide for greater delocalization of
the unpaired electron of its semiquinone radical anion
and consequently have a stabilizing effect compared to
that of N-MOM-ellipticine quinone.
compound
EC50 (µM)
calothrixin A (1)
calothrixin B (2)
N-MOM-calothrixin B (3)
ellipticine quinone (4)
N-MOM-ellipticine quinone (5)
benzocarbazoledione (6)
N-MOM-benzocarbazoledione (7)
menadione (8)
0.12 ( 0.01
0.24 ( 0.04
0.42 ( 0.02
0.15 ( 0.09
0.37 ( 0.08
0.43 ( 0.1
1.6 ( 1.0
N-MOM-ellipticine quinone (5) was found to be sig-
nificantly easier to reduce than its benzo analogue,
N-MOM-benzocarbazoledione (7), by 200 mV. The puta-
tive semiquinone radical anion of N-MOM-ellipticine
quinone can be better stabilized through delocalization
on to the nitrogen atom of the pyridine ring system than
can the radical anion of N-MOM-benzocarbazoledione.
Thus the influence of introducing the pyrido-nitrogen
atom into the benzo ring system of benzocarbazoledione
is to dramatically facilitate reduction. Similar trends
in the ease of reduction are also observed in the parent
quinones, calothrixin B (2), ellipticine quinone (4) and
benzocarbazoledione (6).
The first one-electron reductions of benzocarbazole-
diones 6 and 7 were both seen to be more negative (i.e.
harder to reduce) than for menadione. All of the other
quinones studied, however, were significantly easier to
reduce than menadione (8), whose clinically relevant
anticancer properties are believed to be dependent to a
certain extent on its redox potential.
The reduction of the semiquinones to their respective
dianions occurs at much more negative potentials (rang-
ing from -1.46 to -1.86 V). The effect of N-MOM
substitution on the second reduction of the quinones was
much less pronounced. In general, the reduction of the
semiquinone to the dianion occurred more readily for
the calothrixin semiquinone than for the semiquinone
of ellipticine quinone, which in turn was more readily
reduced than the benzocarbazoledione semiquinone. The
stabilities of the semiquinone radical anions Q-• toward
disproportionation to Q0 and Q2- can be readily assessed
from the ln K values,18 directly related to the gap in
the electrode potentials so long as the reductions are
free of complication.19 From Table 1, we see that the
three N-MOM derivatives, the N-MOM-calothrixin B
(3), N-MOM-ellipticine quinone (5), and N-MOM-ben-
zocarbazoledione (7), are essentially indistinguishable
in this respect with ∆E ∼ 0.7 V.
3.7 ( 0.3
uncyclized precursors to benzocarbazoledione and >50
N-MOM benzocarbazoledione (10c) and (11c)
presence of the methoxymethyl (MOM) substituent on
the indolic nitrogen of N-MOM-calothrixin B (3) (EC50
) 0.42 µM) had the effect of decreasing the potency of
the calothrixin B derivative 2-fold.
Interestingly, the biological activity of the tetracyclic
quinone, ellipticine quinone (4), was comparable to that
of calothrixin B (2). These two compounds differ struc-
turally only in the presence of the fifth fused benzene
ring in the calothrixin molecule. The equipotency of the
pentacyclic calothrixin B and the tetracyclic ellipticine
quinone suggests that the E-ring in the calothrixins is
not crucial for activity per se, although it might of course
have subtle effects on its activity. Comparison of ellip-
ticine quinone (4) (EC50 ) 0.15 µM) and benzocarba-
zoledione (6) (EC50 ) 0.43 µM) revealed a small decrease
in potency upon replacement of the heteroaromatic
D-ring of ellipticine quinone with its carbocyclic aro-
matic analogue. The presence of the MOM substituent
on the indolic nitrogen atom of N-MOM-benzocarba-
zoledione was observed to cause a decrease in the
activity, compared with the unsubstituted benzocarba-
zoledione, by a factor of 4. A similar trend is observed
with calothrixin B, ellipticine quinone, and their N-
MOM derivatives.
In summary the quinones showed micromolar EC50
values against HeLa cells. The most active of these
quinones were the calothrixins and the ellipticine
quinone, followed by the benzocarbazolediones. In con-
trast, menadione (8) also showed micromolar EC50
values against HeLa cells (EC50 ) 3.7 µM) compared to
the value of 0.43 µM for benzocarbazoledione 6.
The uncyclized precursors, 10c and 11c, lack the
connectivity of the central quinone ring system. Their
lack of activity, even at 10 times the EC50 values of their
cyclized products, suggests that the tetracyclic core
structure is an important aspect of bioactivity. This
perhaps suggests that the rigid geometry of the tetra-
cyclic compounds is a defining factor; for example, their
planar geometry may confer an ability to intercalate into
DNA or to bind with high affinity to a particular cleft
in a protein or a DNA-enzyme complex. Alternatively,
it might simply reflect the loss of the quinone function-
ality.
Biology. In this study, the MTT colorimetric assay20
was used to measure cell viability of HeLa cells. The
antiproliferative assays performed in this work involved
a three-day period of incubation of the cells with the
test compound before determination of the proportion
of metabolically active cells in each sample.
The EC50 value of each compound was determined
from a plot of the log of concentration against absor-
bance at 590 nm (A590), in which nonlinear regression
analysis was used to fit a sigmoidal concentration-
response curve through the data points. The GraphPad
Prism 2.0 software package was used to fit the curves
from quadruplicate sets of data and to calculate the
corresponding EC50 values.
Cytotoxicity a n d Red u ction P oten tia ls. A com-
parison of the E1/2 values with the EC50 of the quinones
(1-8) reveals no direct correlation. It is, however, clear
that all these quinones (especially 1-3) are distinctively
easily reduced to the semiquinones, with E1/2 values
more positive than -0.5 V versus the standard hydrogen
electrode (SHE). It has previously been documented that
this threshold is significant for physiologically active
substances to permit electron acceptance from biological
The value obtained for calothrixin B (2) (EC50 ) 0.24
µM; Table 2) was in agreement with that from earlier
work, whereas the value obtained for calothrixin A (1)
differed from the earlier report.11 In our hands, calo-
thrixin A was approximately twice as active (EC50
0.12 µM) as calothrixin B, whereas the earlier report
)
found an almost 10-fold difference in activity. The

Cyanobacterial Calothrixins and Related Quinones
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20 4961
donors.7 The ease of reduction and the pattern for
bioactivity of these quinones are consistent with pre-
liminary mechanistic studies on calothrixins which
suggest that the production of reactive oxygen species
maybe an important factor in the manifestation of their
biological activities.12 Further studies are in progress.
pearance of starting material (Rf ) 0.48 in this eluent). The
reaction mixture was stirred at room temperature for 16 h,
and the THF was removed in vacuo before the addition of
saturated aqueous NaHCO3 (25 mL), followed by the addition
of CH2Cl2 (25 mL). The organic layer was separated, dried over
Na2SO4, filtered, and concentrated to give a pale yellow solid
(420 mg). Trituration with ether and n-hexane, followed by
recrystallization from hot acetone and n-hexane, gave 11c as
a fine white solid (380 mg, 1.2 mmol, 93% yield) in excellent
purity. The product was thoroughly dried in vacuo before
further use.
Rf (1:1 EtOAc/light petroleum): 0.52. Mp: 128 °C. 1H NMR
(CDCl3): δ 3.23 (s, 3H), 3.62 (s, 3H), 5.41 (s, 2H), 7.31 (m, 3H),
7.50-7.65 (m, 4H), 8.01 (d, J ) 6 Hz, 1H), 8.34 (t, J ) 4 Hz,
1H). 13C NMR (CDCl3): 52.2, 56.2, 78.1, 110.4, 118.0, 122.6,
123.2, 124.1, 126.8, 127.8, 129.3, 129.4, 130.1, 132.0, 136.3,
137.0, 142.8, 167.1, 191.3. IR (KBr disk) cm-1: 1712, 1640.
EIMS m/z (relative abundance): 323 (M+•, 100%), 292 (M -
CH3O, 55), 247 (15), 188 (79), 163 (44), 158 (23), 77 (15).
HRMS: calcd 323.1158 for C19H17NO4, found 323.1153. Anal.
(C19H17NO4) C, H, N.
Exp er im en ta l Section
1H NMR and 13C NMR spectra were acquired on a Varian
Gemini 2 spectrometer operating at 300 MHz and 75 MHz,
respectively. The spectra were obtained in deuterated solvents
with chemical shifts (δ), reported in parts per million, refer-
enced to the respective trace solvent peak.
Electron impact mass spectra (EIMS) and high-resolution
mass spectra (HRMS) were obtained by the Research School
of Chemistry Mass Spectrometry Unit, ANU, using a VG
Micromass 7070F double focusing mass spectrometer and
positive ion impact techniques.
Infrared (IR) spectra were measured on a Perkin-Elmer
Spectrum One Fourier transform infrared spectrometer, with
samples analyzed as KBr disks.
Melting points were measured using a Reichert hot-stage
microscope and are reported uncalibrated. Flash column
chromatography was performed using Merck silica gel (Kie-
selgel 60, 0.040-0.063 mm), with positive pressure applied by
nitrogen flow.
Menadione (2-methyl-1,4-naphthoquinone, 8) and 3-(4,5-
dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide
(MTT), used in the bioassays, were purchased from Sigma.
Natural calothrixin A (1) was kindly donated by Professor Rod
Rickards. Calothrixin A (1), calothrixin B (2), and N-MOM
calothrixin B (3) were synthesized as described previously and
are identical to the authentic product (for the former) and
reported data (for the latter).
1-P h en ylsu lfon yl-3-(1-m eth oxyca r bon ylben zen e-2-yl)-
ca r bon ylin d ole (12c). To a stirred solution of 10c (260 mg,
0.94 mmol) in dry THF (40 mL) at -78 °C under nitrogen was
added NaH (0.055 g of a 60% dispersion in mineral oil, 1.4
mmol, 1.5 equiv). The reaction mixture was stirred for 30 min
at -78 °C before being warmed to room temperature, upon
which the previously clear, colorless solution became trans-
lucent orange in color. The reaction mixture was stirred for a
further 20 min at room temperature before being cooled to -78
°C, upon which PhSO2Cl (0.24 mL, 1.87 mmol, 2 equiv) was
added and the color dissipated. The reaction mixture was
warmed to room temperature and stirred for 16 h before
removal of the solvent in vacuo. To the residue was added
saturated aqueous NaHCO3 (25 mL), followed by CH2Cl2 (40
mL). The organic layer was separated, dried over Na2SO4, and
filtered, and the filtrate was concentrated in vacuo to leave a
yellow oil. The residue was then subjected to flash column
chromatography with 2:3 EtOAc/light petroleum to give 12c
in 73% yield, as a fine white solid (268 mg, 0.69 mmol). The
product was thoroughly dried in vacuo before further use.
The syntheses of quinones 4-7 via Scheme 1 have not been
reported previously, and experimental procedures and data are
presented below.
3-(1-Meth oxycar bon ylben zen e-2-yl)car bon ylin dole (10c).
To a stirred mixture of indole (0.86 g, 7.3 mmol) and anhydrous
zinc chloride (2.0 g, 14.6 mmol, 2 equiv) in dry CH2Cl2 (40 mL)
at room temperature under nitrogen was added dropwise
MeMgCl (2.6 mL of 3.0 M solution, 1 equiv). After stirring for
1 h, the reaction mixture was added in one portion to a solution
of acid chloride 9c (1.21 g, 6 mmol, 0.82 equiv) in dry CH2Cl2
(20 mL). The reaction mixture was stirred overnight at room
temperature under nitrogen, before being quenched by the
addition of saturated aqueous NH4Cl (20 mL). The aqueous
phase was extracted with CHCl3 (3 × 20 mL), and the organic
layers were combined and washed with saturated aqueous
NaHCO3 (25 mL) and saturated brine (25 mL). The organic
layer was then dried over MgSO4 and filtered and the solvent
removed to give an orange solid. The crude product was
triturated with hot ether to yield 10c as a fine white powder
(1.30 g, 4.65 mmol, 76% yield).
1
Rf (2:3 EtOAc/light petroleum): 0.52. Mp: 106-108 °C. H
NMR (CDCl3): δ 3.51 (s, 3H), 7.40-8.39 (m, 14H). 13C NMR
(CDCl3): δ 52.2, 113.1, 122.2, 123.0, 125.0, 126.0, 127.1, 127.2,
127.61, 127.64, 129.2, 129.6, 129.6, 130.0, 130.3, 132.4, 133.3,
134.5, 135.0, 137.3, 141.6, 166.6, 191.6. IR (KBr disk) cm-1
:
1720, 1664. EIMS m/z (relative abundance): 419 (M+•, 100%),
388 (4), 283 (22), 250 (13), 235 (25), 234 (18), 219 (24), 190
(12), 163 (16), 144 (22), 115 (13), 77 (57). HRMS: calcd
419.0830 for C23H17NO5S, found 419.0827.
N-Meth oxym eth yl-5H-ben zo[b]car bazole-6,11-dion e (7).
To a stirred solution of 11c (110 mg, 0.34 mmol) and TMEDA
(52 µL, 0.34 mmol, 1.1 equiv) in dry THF (25 mL) at -78 °C
under nitrogen was added dropwise a solution of LHMDS (0.68
mmol, 2.2 equiv) in dry THF (2 mL) under N2 at -78 °C. The
reaction mixture was stirred for 2 h at -78 °C before being
allowed to warm to room temperature and stirred overnight.
The reaction mixture was concentrated under reduced pres-
sure, following which saturated aqueous NaHCO3 (10 mL) was
added. The product was extracted with CH2Cl2 (3 × 25 mL).
The organic layers were combined, washed with saturated
brine, dried over MgSO4, and filtered, and the solvent was
evaporated to leave an orange solid. Column chromatography
in 1:2 EtOAc/light petroleum, followed by recrystallization
from ether/n-hexane, gave the desired compound as a yellow
solid (87 mg, 0.30 mmol, 88% purified yield).
1
Rf (1:1EtOAc/light petroleum): 0.48. Mp: 197-198 °C. H
NMR (d-DMSO): δ 3.59 (s, 3H), 7.24 (m, 2H), 7.49-7.73 (m,
5H), 7.90 (d, J ) 8 Hz, 1H), 8.13 (d, J ) 8 Hz, 1H), 12.01 (s,
1H). 13C NMR (d-DMSO): 52.1, 112.3, 116.4, 121.2, 121.9,
123.1, 125.6, 128.0, 129.43, 129.46, 129.52, 132.0, 135.4, 136.8,
142.6, 166.8, 190.1. IR (KBr disk) cm-1: 3214, 1721, 1604.
EIMS m/z (relative abundance): 279 (M+•, 65%), 248 (M -
OCH3, 22), 220 (M - CO2CH3, 13), 163 (13), 144 (100), 116
(29), 89 (24). HRMS: calcd 279.0895 for C17H13NO3, found
279.0896. Anal. (C17H13NO3) C, H, N.
1-Meth oxym eth yl-3-(1-m eth oxyca r bon ylben zen e-2-yl)-
ca r bon ylin d ole (11c). To a stirred solution of 10c (350 mg,
1.25 mmol) in dry THF (35 mL) at 0 °C under nitrogen was
added NaH (60 mg of a 60% dispersion in mineral oil, 1.5
mmol, 1.2 equiv). The reaction mixture was stirred for 20 min
at 0 °C, upon which MOMCl (0.14 mL, 1.8 mmol, 1.5 equiv)
was added and the color dissipated. After the mixture was
allowed to warm to room temperature, the reaction was
monitored by TLC (1:1 EtOAc/light petroleum) for the disap-
1
Rf (1:2 EtOAc/light petroleum): 0.58. Mp: 143-145 °C. H
NMR (CDCl3): δ 3.39 (s, 3H), 6.18 (s, 2H), 7.43 (t, J ) 7 Hz,
1H), 7.51 (t, J ) 7 Hz, 1H), 7.65 (d, J ) 8 Hz, 1H), 7.73 (m,
2H), 8.20 (m, 2H), 8.49 (d, J ) 8 Hz, 1H). 13C NMR (CDCl3):
56.5, 75.3, 111.9, 120.2, 123.9, 124.0, 124.9, 126.3, 126.6, 127.9,
133.0, 133.5, 133.7, 133.9, 135.1, 139.8, 179.0, 181.6. IR (KBr
disk) cm-1: 1658. EIMS m/z (relative abundance): 291 (M+•
,
100%), 276 (M - CH3, 88), 260 (M - OCH3, 47), 248 (M -

4962 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20
Bernardo et al.
CH2OCH3, 94), 190 (22), 102 (28). HRMS: calcd 291.0895 for
added NaH (0.16 g of a 60% dispersion in mineral oil, 2.2
mmol). The reaction mixture was stirred for 30 min at 0 °C,
after which MOMCl (0.34 mL, 4.4 mmol, 2 equiv) was added
and the reaction mixture became orange in color. The reaction
mixture was warmed to room temperature and stirred over-
night. The THF was removed in vacuo before addition of
saturated aqueous NaHCO3 (20 mL). The product was ex-
tracted into CH2Cl2 (3 × 30 mL), dried over Na2SO4, and
filtered. Concentration of the filtrate gave a pale brown solid,
which upon flash chromatography with EtOAc gave the
N-protected indole 11b as a fine white solid (0.66 g, 2 mmol,
90% purified). The product was thoroughly dried in vacuo
before further use.
Rf (1:1 EtOAc/light petroleum): 0.24. Rf (EtOAc): 0.40.
Mp: 126 °C. 1H NMR (CDCl3): δ 3.26 (s, 3H), 3.71 (s, 3H);
5.41 (s, 2H), 7.38 (m, 3H), 7.51 (m, 1H), 7.80 (d, J ) 5 Hz,
1H), 8.37 (t, J ) 4 Hz, 1H), 8.85 (br s, 2H). 13C NMR (CDCl3):
δ 52.9, 56.3, 78.2, 110.5, 117.7, 122.6, 123.1, 123.5, 124.5, 126.6,
136.1, 136.7, 137.0, 137.3, 148.9, 151.2, 165.7, 188.2. EIMS
m/z (relative abundance): 324 (M+•, 100), 293 (48), 188 (72),
C
18H13NO3, found 291.0897. Anal. (C18H13NO3) C, H, N.
5H-Ben zo[b]ca r ba zole-6,11-d ion e (6). Meth od 1: Cy-
cliza tion of 12c.21 To a stirred solution of 12c (100 mg, 0.26
mmol) and TMEDA (44 µL, 0.29 mmol, 1.1 equiv) in dry THF
(20 mL) at -78 °C under nitrogen was added dropwise a
solution of LHMDS (0.57 mmol, 2.2 equiv). The reaction
mixture was stirred for 2 h at -78 °C under nitrogen, before
being allowed to warm to room temperature and stirred
overnight. The solvent was removed under reduced pressure,
5 mL of saturated aqueous NaHCO3 was added to quench the
reaction, and the product was extracted with CH2Cl2 (4 × 30
mL). The organic layers were combined, washed with satu-
rated brine, dried over Na2SO4, and filtered, and the solvent
was evaporated to leave a dark orange solid (100 mg).
Trituration with cold CHCl3 gave 4 mg (0.02 mmol, 6%) of fine
orange crystals. The filtrate was subjected to flash column
chromatography in 100% CH2Cl2, which afforded a further 8
mg of fine orange crystals. The combined yield of desired
product was 18% (12 mg, 0.06 mmol).
164 (26). HRMS: calcd 324.1110 for
324.1112. Anal. (C18H16N2O4) C, H, N.
C18H16N2O4, found
Meth od 2: Dep r otection of 7. To a stirred solution of 7
(42 mg, 0.14 mmol) in CH2Cl2 (10 mL) at -78 °C under
nitrogen was added BBr3 via a disposable syringe (0.28 mL of
1.0 M solution in hexanes, 0.28 mmol, 1 equiv). The reaction
mixture was stirred at room temperature for 1 h. An equal
volume of saturated aqueous NH4Cl (10 mL) was added to the
reaction mixture and stirred vigorously for 1 h at 60 °C. The
aqueous layer was extracted with CH2Cl2 (3 × 15 mL), the
combined organic layers were dried over Na2SO4 and filtered,
and the solvent was removed in vacuo to give an orange solid
(30 mg). Purification by column chromatography (1:2 EtOAc/
light petroleum), followed by recrystallization from ether/n-
hexane, afforded the desired compound as a fine orange solid
(20 mg, 0.08 mmol, 57%).
Rf (1:2 EtOAc/light petroleum): 0.34. Mp (lit.22 307-310
°C): 300-303 °C. 1H NMR (d-DMSO): δ 7.36 (t, J ) 7 Hz,
1H), 7.48 (t, J ) 7 Hz, 1H), 7.58 (d, J ) 8 Hz, 1H), 7.84 (m,
2H), 8.10 (t, J ) 7 Hz, 2H), 8.21 (d, J ) 8 Hz, 1H), 13.09 (s,
1H). 13C NMR (d-DMSO): 113.9, 117.4, 122.4, 123.9, 124.0,
126.0, 126.1, 127.0, 132.6, 133.2, 134.1, 134.3, 137.2, 138.2,
177.6, 180.4. EIMS m/z (relative abundance): 247 (M+•, 64%),
228 (19), 219 (M - CO, 19), 201 (23), 199 (22), 190 (15), 91
(100). HRMS: calcd 247.0633 for C16H9NO2, found 247.0637.
Anal. (C16H9NO2) C, H, N.
3-(4-Methoxycarbonylpyridine-3-yl)carbonylindole (10b).
To a stirred mixture of indole (0.39 g, 3.33 mmol) and
anhydrous zinc chloride (0.91 g, 6.66 mmol, 2 equiv) in dry
CH2Cl2 (50 mL) at 0 °C was added dropwise MeMgCl (1.11
mL of 3.0 M solution, 3.33 mmol, 1 equiv). The ice bath was
removed and the reaction mixture allowed to warm to room
temperature. After being stirred for 1 h, the reaction mixture,
which was opaque pink in color, was transferred in one portion
to the orange solution of the acid chloride 9b (0.55 g, 2.76
mmol, 0.83 equiv) in dry CH2Cl2 (50 mL). The reaction mixture
was stirred overnight at room temperature under nitrogen,
before being quenched by addition of saturated aqueous NH4-
Cl (50 mL). The aqueous phase was extracted with CH2Cl2 (2
× 20 mL), and the organic layers were combined and washed
with saturated aqueous NaHCO3 and saturated brine (50 mL).
The organic layer was then dried over MgSO4 and filtered, and
the solvent was removed to leave a solid residue. Trituration
with ether gave the desired product as a white solid (0.28 g,
2.0 mmol, 72% yield).
Mp: 173-175 °C. 1H NMR (CDCl3): δ 3.44 (s, 3H), 7.2-7.3
(m, 2H), 7.41 (m, 1H), 7.56 (d, J ) 4 Hz, 1H), 7.85 (d, J ) 9
Hz, 1H), 8.79 (s, 1H), 9.16 (s, 1H). 13C NMR (CDCl3): 53.0,
111.7, 118.1, 122.3, 123.2, 124.4, 125.4, 134.2, 136.5, 137.9,
148.3, 150.5, 165.7, 188.0. EIMS m/z (relative abundance): 281
(M+•, 27), 280 (71), 144 (100), 116 (32), 89 (27). HRMS: calcd
280.0848 for C16H12N2O3, found 280.0854. Anal. (C16H12N2O3)
C, H, N.
N-Meth oxym eth yl-6H-p yr id o[4,3-b]ca r ba zole-5,11-d i-
on e (5). To a stirred solution of 11b (0.35 g, 1.1 mmol) and
TMEDA (0.16 mL, 1.1 mmol) in dry THF (50 mL) at -78 °C
under nitrogen was added dropwise a solution of LHMDS (2.2
mmol) at -78 °C. The reaction mixture was stirred for 2 h at
-78 °C, before being allowed to warm to room temperature
and stirred overnight. The solvent was removed under reduced
pressure, following which saturated aqueous NaHCO3 (50 mL)
was added. The product was extracted with EtOAc (4 × 25
mL), and the combined organic layers were washed with
saturated brine (50 mL), dried over Na2SO4, and filtered. The
filtrate was concentrated in vacuo to leave a yellow solid (0.46
g). Column chromatography (1:1 EtOAc/light petroleum fol-
lowed by pure EtOAc) afforded pure N-MOM-ellipticine quino-
ne 5 (0.11 g, 35% purified yield).
Rf (2:1 EtOAc/light petroleum): 0.47. Mp: (lit.23 196-197
°C): 195 °C. 1H NMR (CDCl3): δ 3.39 (s, 3H), 6.15 (s, 2H),
7.47 (t, J ) 6 Hz, 1H), 7.56 (t, J ) 6 Hz, 1H), 7.67 (d, J ) 9
Hz, 1H), 7.99 (d, J ) 9 Hz, 1H), 8.49 (d, J ) 8 Hz, 1H), 9.06
(d, J ) 5 Hz, 1H), 9.46 (s, 1H). 13C NMR (CDCl3): 56.6, 75.4,
112.0, 120.3, 123.7, 124.1, 125.4, 128.7, 134.4, 139.4, 140.2,
148.2, 154.9, 177.8, 180.9. IR (KBr disk) cm-1: 1667, 1651.
EIMS m/z (relative abundance): 292 (M+•, 100); 277 (60); 261
(25); 249 (60); 219 (10); 164 (13). HRMS: calcd 292.0842 for
C
17H12N2O3, found 292.0842. Anal. (C17H12N2O3) C, H, N.
6H-P yr id o[4,3-b]ca r ba zole-5,11-d ion e22 (4). The title
compound was prepared from 5 by removal of the MOM group
as described above and was purified by column chromatogra-
phy (1:2 EtOAc/light petroleum), followed by recrystallization
from diethyl ether and hexane to give the pure compound in
71% yield.
Rf (1:2 EtOAc/light petroleum): 0.29. Mp (lit.22 315-316):
1
g300 °C. H NMR (d6-DMSO): δ 7.40 (t, J ) 7 Hz, 1H), 7.49
(t, J ) 7 Hz, 1H), 7.62 (d, J ) 8 Hz, 1H), 7.95 (d, J ) 5 Hz,
1H), 8.23 (d, J ) 8 Hz, 1H), 9.08 (d, J ) 5 Hz, 1H), 9.27 (s,
1H). Anal. (C17H12N2O3) H, N; C: calcd 72.58, found 71.95.
Volta m m etr ic Exp er im en ts. Voltammetric experiments
were performed using a three-electrode potentiostat, which
consisted of a Princeton Applied Research electrochemistry
system (PAR-170) or a PAR-273 system, both under computer
control through a Maclab interface. A Tacussel platinum disk
electrode was employed as the working electrode, together with
a platinum wire quasi-reference electrode, a platinum wire as
the auxiliary electrode, and tetra-n-butylammonium hexafluo-
rophosphate (0.1 molar) as the supporting electrolyte. The
ferrocene/ferrocenium couple was used as an internal stan-
dard. The dimethyl sulfoxide used as a solvent was of analyti-
cal grade and dried over 4 Å molecular sieves before use. The
electrolyte solution was deoxygenated by vigorous purging with
nitrogen for 30 min prior to the experiments, which were
performed under nitrogen. All experiments were carried out
at 20 °C. Typical scan rates of 20 mV/s were used for acV
1-Meth oxym eth yl-3-(4-m eth oxyca r bon ylp yr id in e-3-yl-
)ca r bon ylin d ole (11b). To a stirred solution of 10b (0.63 g,
2.2 mmol) in dry THF (100 mL) at 0 °C under nitrogen was

Cyanobacterial Calothrixins and Related Quinones
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20 4963
(5) O’Brien, P. J . Molecular Mechanisms of Quinone Cytotoxicity.
Chem. Biol. Interact. 1991, 80, 1-14.
(6) Hrbac, J .; Kohen, R. Biological Redox Activity: Its Importance,
Methods For Its Quantification And Implication For Health And
Disease. Drug Dev. Res. 2000, 50, 516-527.
(7) De Abreu, F. C.; de L. Ferraz, P. A.; Goulart, M. O. F. Some
Applications of Electrochemistry in Biomedical Chemistry.
Emphasis on the Correlation of Electrochemical and Bioactive
Properties. J . Braz. Chem. Soc. 2002, 13, 19-35.
(8) Koyama, J .; Morita, I.; Tagahara, K.; Osakai, T.; Hotta, H.; Yang,
M. X.; Mukainaka, T.; Nishino, H.; Tokuda, H. Correlation with
Redox Potentials and Inhibitory Effects on Epstein-Barr Virus
Activation of Azaanthraquinones. Chem. Pharm. Bull. 2001, 49,
1214-1216.
(9) Inbaraj, J . J .; Krishna, M. C.; Gandhidasan, R.; Murugesan, R.
Cytotoxicity, Redox Cycling and Photodynamic Action of Two
Naturally Occurring Quinones. Biochim. Biophys. Acta 1999,
1472, 462-470.
(10) Paz, M. M.; Das, A.; Palom, Y.; He, Q.-Y.; Tomasz, M. Selective
Activation of Mitomycin A by Thiols To Form DNA Cross-links
and Monoadducts: Biochemical Basis for the Modulation of
Mitomycin Cytoxicity by the Quinone Redox Potential. J . Med.
Chem. 2001, 44, 2834-2842.
(11) Rickards, R. W.; Rothschild, J . M.; Willis, A. J .; de Chazal, N.
M.; Kirk, J .; Kirk, K.; Saliba, K. J .; Smith, G. D. Calothrixins A
and B, Novel Pentacyclic Metabolites from Calothrix Cyanobac-
teria with Potent Activity Against Malaria Parasites and Human
Cancer Cells. Tetrahedron 1999, 55, 13513-13520.
experiments, and 50-1000 mV/s for CV experiments. The
potentials quoted are the convergent results of 50 mV CV data
and the acV data. The acV experiments were performed with
the oscillating voltage modulation having an amplitude of 10
mV and frequency of 205 Hz.
Bioa ssa ys. Human cervical cancer cells (HeLa cell line)
were used for all assays. The HeLa cells were cultured as
monolayers in RPMI 1640 media with 10% fetal calf serum,
at 37 °C with 5% CO2. Progression of confluence was monitored
by observation using an inverted light microscope. Cells were
harvested according to the following protocol: the existing
medium was decanted and the cells were washed twice with
phosphate-buffered saline (PBS), before addition of sufficient
Trypsin-EDTA solution to cover the cell monolayer and
incubation for 5 min at 37 °C. The released cells were
transferred to a centrifuge tube with 30 mL of fresh medium
and spun at 800 rpm for 3 min. The media was decanted and
the pellet resuspended in 5-10 mL of fresh media. The cell
density of this suspension was determined by use of a
haemocytometer, which had a volume of 10-4 mL and was
transposed with a 5 × 5 grid. The number of cells present in
the 25 squares of the grid were counted under
a light
microscope (Olympus, BHB, 20× magnification) and the
concentration thereby determined.
Effects on cell proliferation and viability were measured
using a colorimetric MTT assay. To wells 2-12 of a 96-well
flat-bottomed microtiter plate was added 50 µL of fresh
medium (RPMI 1640 with 10% FCS). To well 1 was added 100
µL of a solution of the sample compound (of known concentra-
tion in medium, made up from an initial stock solution in
Hybrimax DMSO; Sigma), and this was serially diluted (1:1)
along the plate, discarding after well 10. To wells 1-11 was
added 50 µL of cell suspension, adjusted to achieve a final
concentration of 10000 cells per well in 100 µL for a 3 day
assay. The cells were incubated at 37 °C for 72 h, after which
the solution was tipped from the plate and the cells were
washed with PBS (2 × 100 µL). To each well were added 100
µL of fresh medium and 10 µL of MTT (as a solution in PBS,
5 mg/mL), followed by incubation at 37 °C for 5 h. After this
time, 100 µL of solubilizing agent (10% sodium dodecyl sulfate
in 0.01 M aqueous HCl) was added to each well before
incubation overnight at 37 °C. The absorbance of each well of
the plate was read at 590 nm by an ELISA automated
spectrophotometric plate reader. Results were expressed as
percentages of control (untreated) cells and were analyzed
using GraphPad Prism 2.0 software. All determinations were
performed in quadruplicate, and results were typical of
multiple independent experiments.
(12) Chen, X.; Smith, G. D.; Waring, P. W. Human Cancer Cell
(J urkat) Killing By The Cyanobacterial Metabolite Calothrixin
A. J . Appl. Phycol. 2003, 15, 269-277.
(13) Bernardo, P. H.; Elix, J . A.; Chai, C. L. L. A Simple and Concise
Route to Calothrixin B. Tetrahedron Lett. 2002, 43, 2939-2940.
(14) Bernardo, P. H.; Chai, C. L. L. Friedel-Crafts Acylation and
Metalation Strategies in the Synthesis of Calothrixins A and B.
J . Org. Chem. 2003, 68, 8906-8909.
(15) Schmid, R.; Goebel, F.; Warnecke, A.; Labahn, A. Synthesis and
Redox Potentials of Methylated Vitamin K Derivatives. J . Chem.
Soc., Perkin Trans. 2 1999, 1199-1202.
(16) Gonzalez, F. Cyclic Voltammetry of Two Analogue K-Group
Vitamin Compounds in Dimethylsulfoxide. J . Electroanal. 1998,
10, 638-642.
(17) Roginsky, V. A.; Leonid, M. P.; Bors, W.; Michel, C. The Kinetics
and Thermodynamics of Quinone-Semiquinone-Hydroquinone
Systems Under Physiological Conditions. J . Chem. Soc., Perkin
Trans. 2 1999, 871-876.
(18) Macias-Ruvalcaba, N.; Cuevas, G.; Gonzalez, I.; Aguilar-Mar-
tinez, M. Relationship between Molecular Structure and Electron
Targets in the Electroreduction of Benzocarbazolediones and
Anilinenaphthoquinones. Experimental and Theoretical Study.
J . Org. Chem. 2002, 67, 3673-3681.
(19) Macias-Ruvalcaba, N.; Gonzalez, I.; Aguilar-Martinez, M. Evolu-
tion from Hydrogen Bonding to Proton-Transfer Pathways in
the Electroreduction of the R-NH Quinones in Acetonitrile. J .
Electrochem. Soc. 2004, 151, E110-E118.
(20) Mossmann, T. F. Rapid Colorimetric Assay For Cellular Growth
And Survival: Application To Proliferation And Cytotoxicity
Assays. J . Immunol. Methods 1983, 65, 55-63.
(21) Ketcha, D. M.; Gribble, G. W. A Convenient Synthesis of
3-Acylindoles via Friedel-Crafts Acylation of 1-(Phenylsul-
fonyl)indole. A New Route to Pyridocarbazole-5,11-quinones and
Ellipticine. J . Org. Chem. 1985, 50, 5451-5457.
(22) Taylor, D. A.; Baradarani, M. M.; Martinez, S. J .; J oule, J . A.
The Synthesis Of 5,11-Didemethylellipticine-5,11-Quinone And
Its Transformation Into Ellipticine And Ellipticine Analogues.
J . Chem. Res. 1979, 4801-4817.
(23) Watanabe, M.; Snieckus, V. Tandem Directed Metalation Reac-
tions. Short Syntheses of Polycyclic Aromatic Hydrocarbons and
Ellipticine Alkaloids. J . Am. Chem. Soc. 1980, 102, 1457-1460.
Su p p or tin g In for m a tion Ava ila ble: An appendix con-
taining purity criteria data. This material is available free of
charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Monks, T. J .; Hanzlik, R. P.; Cohen, G. M.; Ross, D.; Graham,
D. G. Contemporary Issues in Toxicology: Free Radicals in
Toxicology. Toxicol. Appl. Pharmacol. 1992, 112, 2-16.
(2) Halliwell, B.; Gutteridge, J . M. C. Free Radicals in Biology and
Medicine; Oxford University Press, 3rd ed.; pp 715-720.
(3) Bolton, J . L.; Trush, M. A.; Penning, T. M.; Dryhurst, G.; Monks,
T. J . Role of Quinones in Toxicology. Chem. Res. Toxicol. 2000,
13, 135-160.
(4) Powis, G. Metabolism and Reactions of Quinoid Anticancer
Agents. Pharmacol. Ther. 1987, 35, 57-162.
J M049625O