Synthesis and in vitro antitumor activity of novel ring D analogues of the marine pyridoacridine ascididemin: Structure-activity relationship

Article abstract of DOI:10.1021/jm0208774

Marine compounds with pyridoacridine skeletons are known to exhibit interesting antitumor activities. Ascididemin has already been reported as displaying significant antitumor activities in vitro and has also been found to have a relatively high global to

Full text of DOI:10.1021/jm0208774

J . Med. Chem. 2002, 45, 3765-3771
3765
Syn th esis a n d In Vitr o An titu m or Activity of Novel Rin g D An a logu es of th e
Ma r in e P yr id oa cr id in e Ascid id em in : Str u ctu r e-Activity Rela tion sh ip
Evelyne Delfourne,*,† Francis Darro,‡ Philippe Portefaix,† Chantal Galaup,† Sylvie Bayssade,† Anne Bouteille´,†
Laurent Le Corre,† J ean Bastide,† Franc¸oise Collignon,‡ Brigitte Lesur,‡ Armand Frydman,‡ and Robert Kiss§
Centre de Phytopharmacie, UMR-CNRS 5054, Universite´ de Perpignan, 52 Avenue de Villeneuve,
66860 Perpignan Cedex, France, Laboratoire Louis Lafon, Centre de Recherches, 19 Avenue du Pr Cadiot,
B.P. 22, 94701 Maisons-Alfort Cedex, France, and Laboratoire d’Histopathologie, Faculte´ de Me´decine,
Universite´ Libre de Bruxelles, 808, Route de Lennik, 1070 Bruxelles, Belgique
Received March 14, 2002
Marine compounds with pyridoacridine skeletons are known to exhibit interesting antitumor
activities. Ascididemin has already been reported as displaying significant antitumor activities
in vitro and has also been found to have a relatively high global toxicity in vivo. We synthesized
a series of 16 analogues (among which 11 compounds were different from previously described
ones) with the aim of developing new anticancer agents with significantly improved efficacy/
tolerability ratios. These compounds were obtained either by total synthesis from 5,8-
quinolinedione and substituted 2-aminoacetophenones or by the direct substitution of
ascididemin. The different compounds and ascididemin used as the control compound were
tested at six different concentrations on 12 different human cancer cell lines of various
histopathological types (glioblastomas and breast, colon, lung, prostate, and bladder cancers).
The IC50 value (i.e., the drug concentration inhibiting the mean growth value of the 12 cell
lines by 50%) of these compounds ranged over five log concentrations, i.e., between 10 000 and
0.1 nM. For several new chemical entities, the antitumor activity (determined in vitro) and
tolerability (determined in vivo) were superior to those of the parent alkaloids, i.e., ascididemin
and 2-bromoleptoclinidone.
In tr od u ction
Researchers have long been investigating natural
products in the quest for new anticancer drugs. Marine
organisms have provided a large number of new com-
pounds including bryostatin,1 dolastatin,2 cryptophycin,3
aplidine,4 and ecteinascidin 743,5 all of which are
currently undergoing clinical trials.
F igu r e 1. (a) Numbered according to IUPAC nomenclature
To date, the largest family of marine alkaloids char-
acterized has been based upon the pyrido[2,3,4-kl]-
acridine skeleton.6 Ascididemin 1, isolated by Kobayashi
et al.7 from the ascidian Didemnum sp., is one of the
first examples of these compounds. Some substituted
ascididemins such as bromoleptoclinidone 2,8 11-hy-
droxyascididemin 3,9 and neocalliactine 410 were also
isolated from marine sources, whereas 11-methoxy-
ascididemin 5,11 1- and 3-nitroascididemins (6 and 7),12
and ring E-substituted ascididemins13 were obtained by
total synthesis. Bracher14 and Kitahara et al.11,15 have
described 3-methoxyascididemin 16 as an intermediate
product in two separate total synthesis processes of
neocalliactine 4 (Figure 1).
It emerges from all of these published data that these
different compounds have cytotoxic properties with an
IC50 value in the micromolar range and that the most
active of them remains ascididemin itself. In a recent
paper, Copp and co-workers16 have shown that 1 can
act to damage DNA oxidatively via a thiol-dependent
rules. (b) Compounds 6 and 7, described in Gellerman’s paper
(ref 12) as 1-nitroascididemin and 3-nitroascididemin, are,
respectively, 7-nitroascididemin and 5-nitroascididemin ac-
cording to IUPAC nomenclature rules.
conversion of oxygen to DNA-cleaving radicals. In view
of this original mode of action, ascididemin analogues
appear therefore to have good prospects as anticancer
drug candidates. As part of our work related to marine
alkaloid analogues,17 we are specifically interested in
novel ring D-substituted ascididemins, and in the
present paper, we describe the synthesis of these
compounds and their in vitro antitumor activity and
discuss the influence of the different substituents on this
activity.
Ch em istr y
The different ascididemin analogues (Table 1) were
synthesized according to two main pathways: (i) sub-
stitution of ascididemin 1 for compounds 8-13 and 19
and (ii) total synthesis using Bracher’s methodology18
for the other derivatives. In addition, a scenario for the
design of difunctional compounds was also evaluated
(through the preparation of compound 19).
* To whom correspondence should be addressed. Tel: 33 468 66 22
51. Fax: 33 468 66 22 23. E-mail:delfourn@univ-perp.fr.
† Universite´ de Perpignan.
‡ Centre de Recherches.
§ Universite´ Libre de Bruxelles.
10.1021/jm0208774 CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/10/2002

3766 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Delfourne et al.
Ta ble 1a
demin 5 was prepared according to Kitahara et al.’s
procedure.15 We are also interested in an attempt to
design difunctional compounds. Compound 19 disubsti-
tuted on ring D was thus obtained by the bromination
of compound 10.
R1
R3
R4
R1
R3
Cl
R4
8
9
NH2
14
15
16
17
18
19
Br
NH2
Me
10
11
12
13
OMe
NMe2
NHBn
NH2
NHCH2CH2Cl
N(CH2CH2Cl)2
NHCH2CH2NMe2
P h a r m a cology
Br
In Vitr o Deter m in a tion of th e Dr u g-In d u ced
In h ibition of Hu m a n Ca n cer Cell Lin e Gr ow th . For
each of the 17 compounds under study (including
ascididemin), six concentrations were tested on 12
different human cancer cell lines including various
histopathological types (glioblastomas and breast, colon,
lung, prostate, and bladder cancers). We made use of
the colorimetric 3-(4,5)-dimethylthiazol-2-yl)-2,5-di-
phenyltetrazolium bromide (MTT) assay, which indi-
rectly assesses the effect of potentially anticancer
compounds on the overall growth of adherent cell lines.20
The IC50 values, i.e., the concentration that reduced the
mean growth value of the 12 cell lines by 50%, was
determined for each drug, in comparison with the mean
control growth value. Table 2 illustrates the individual
IC50 values of the different compounds obtained for each
of the 12 cell lines under study; the in vivo global toxic
effects as revealed by the maximum tolerated dose
(MTD) index are reported in Table 3.
a
R2 and R5 ) H.
Sch em e 1a
a
Reagents: (a) CeCl3‚7H2O, EtOH, room temperature. (b)
Concentrated H2SO4/AcOH, reflux. (c) DMF-DEA, DMF, 120 °C.
(d) NH4Cl, AcOH or EtOH, reflux.
These data show that the IC50 values range over five
logarithmic concentrations. As indicated by the values
of the mean IC50, compounds 6, 8, 10, 11, and 17
exhibited significantly greater in vitro antitumor activ-
ity (range, 7-53 nM) than the natural substance asci-
didemin 1 (mean IC50 ) 100 nM) as far as the overall
panel of solid tumors (n ) 12) is concerned. However,
compounds 9, 13, 15, and 16 selectively exhibited higher
cytotoxic potency than natural ascididemin against cell
lines (such as bladder, colon, breast, lung, and glioblas-
toma cancer cell lines). Besides, the in vitro antitumor
activities of compounds 2, 12, 18, and 19 are comparable
to that of ascididemin (compound 1), again with some
degree of specificity (i.e., compounds 12, 18, and 19).
More interestingly, even though the 11 new ascididemin
derivatives exhibited more potent in vitro cytotoxic
activity than the natural compounds (1 and 2) on a
majority of cancer cell lines, their systemic in vivo
toxicity (defined by the MTD in vivo index) was not
simultaneously increased. In some cases, this toxicity
even dramatically decreased. Indeed, except for com-
pounds 6, 13, and 15, which showed similar systemic
toxicity (MTD ) 20 mg/kg) as ascididemin, the remain-
ing compounds exhibited an improvement (at least by
a factor of 8) in the in vivo tolerability (MTD > 160 mg/
kg). However, while the improvement of in vitro cyto-
toxic potency was associated with a lower acute in vivo
toxicity for a significant number of compounds from the
present study, no clear-cut relationship emerged for
those two properties in the case of compounds 6 and 15
when compared to the others.
The nitration and bromination reactions of 1 gave
only substitutions on the benzenic ring at two different
positions, R1 for NO2 and R3 for Br. The nitration of 1
with a mixture of fuming HNO3/concentrated H2SO4
(130 °C, 1.5 h) afforded compound 6 exclusively. In
contrast to what was reported by Kashman,12 we
observed no substitution at R3 (formation of compound
7). The iron reduction of the nitro group of 6 in acetic
acid media gave the corresponding amino-derivative 8.
The direct bromination of 1 led to the compound
substituted at R3 (9). The treatment of 9 by sodium azide
in dimethylformamide (DMF) directly gave the amino-
derivative 10. Such a spontaneous reduction had al-
ready been observed by Kubo and co-workers19 in the
synthesis of cystodamine. The reductive aminoalkyla-
tion of compound 10 with chloroethanal allowed both
the mono- and the disubstitution of the amine and so
provided access to substituted N-derivatives 11 and 12,
respectively. The monosubstituted compound 13 was
obtained when the aldehyde was dimethyl aminoetha-
nal.
Bracher’s methodology, which is the more generally
applicable procedure reported to date for the synthesis
of ascididemin derivatives, starts by the condensation
of quinoline-5,8-dione with an ortho-aminoacetophe-
none, as indicated in Scheme 1. After the acid-catalyzed
cyclization of the condensed product, the final ring E
annulation is achieved by using DMF diethylacetal
(DMF-DEA) in DMF under nitrogen to form an en-
amine, which is then cyclized into the pentacyclic
product with ammonium chloride in ethanol or acetic
acid.
Discu ssion
Ascididemin 1 and its natural analogues 2 and 3 have
been tested at the U.S. National Cancer Institute (NCI)
at the stage of primary in vitro screening. Comprehen-
sive testing has demonstrated that all of the compounds
exhibit selective cytotoxicity against certain cancer types
The different precursors (which are substituted ami-
noacetophenone compounds) of compounds 14-18 were
synthesized according to previously described proce-
dures (see Experimental Section). 10-Methoxyascidi-

Marine Pyridoacridine Ascididemin
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3767
Ta ble 2. Characterization of the In Vitro Cytotoxic-Related Antitumor Effects (IC50 Value in nM) of the Compounds Listed in Table
1a
cell lines
compds
U-87MG
U-373MG
SW1088
T24
J 82
HCT-15
LoVo
MCF7
T-47D
A549
A-427
PC-3
1
2
70
500
500
600
600
400
800
60
300
500
6
100
900
600
70
90
600
60
200
400
60
60
8
80
3
4000
6000
6000
4000
6000
4000
5000
2000
3000
5000
1000
5000
4
not available for testing
5
6
500
4
800
7
2000
6
600
5
2000
20
500
40
600
500
600
5
900
400
600
500
600
8
900
200
7
not available for testing
8
9
300
500
40
6
40
30
400
60
70
30
70
80
50
200
8
80
500
80
0,9
50
500
900
400
300
200
700
200
6
40
7
100
400
50
9
90
30
60
40
7
70
80
80
60
50
60
50
300
3000
600
2000
90
4000
2000
800
500
900
400
900
5
50
20
10
300
30
500
30
80
6
400
50
70
2000
700
2000
400
40
600
80
800
1000
90
500
8
10
60
400
700
100
80
90
60
10
50
5
0,9
20
10
400
50
100
9
400
60
300
80
500
80
300
60
10
11
12
13
14
15
16
17
18
19
2
1
60
70
600
100
80
50
90
10
40
10
90
60
60
20
50
70
90
500
400
90
60
400
200
80
100
800
500
40
70
400
8
4000
100
a
The IC50 value constitutes the concentration of the compound that inhibits the growth of the human cancer cells by 50% as compared
to the control value. Results are reported as mean values (n ) 3). The values for the standard errors are not reported here (for the sake
in clarity of the table) because they reached less than 3% of the mean values. Six concentrations ranging from 10 000 to 0.1 nM were
assayed on 12 different human cancer cell lines for each compound under study. The drug-induced effects at cell line growth level were
determined by means of the MTT colorimetric assay.
Ta ble 3. Mean Cytotoxic Potency (In Vitro/12 Cell Lines) IC50
of the ring D-modified ascididemin analogues at least
have similar cytotoxic effect or, more importantly,
(nM)a and MTDb (mg/kg) in Mice
a
a
mean IC50
(nM)
MTDb
(mg/kg) compds
mean IC50
(nM)
MTDb
(mg/kg)
appear to be more cytotoxic than the reference com-
pound ascididemin. In addition, when substituted on
ring D, these ascididemin analogues exhibit a signifi-
cantly greater MTD (160 mg/kg instead of 20 mg/kg for
ascididemin and 40 mg/kg for 2-bromoleptoclinidone),
which anticipates a better therapeutic index (antitumor
efficacy/overall safety ratio) in in vivo models. Besides,
it seems that both the R1 and the R3 positions on ring
D could be modulated without losing the cytotoxic
activity.
compds
1
2
3
5
6
8
9
10
11
100
120
3200
480
10
53
80
21
7
20
40
>160
>160
20
>160
>160
>160
>160
12
13
14
15
16
17
18
19
100
60
270
60
90
37
>160
40
>160
20
>160
>160
>160
>160
140
140
a
Mean IC50 was established as the arithmetic mean of the 12
IC50 values reported in Table 2. b MTD was defined following single
ip injection to B6D2F1/jico mice.
The Copp group, who investigated a series of ascidi-
demin analogues, has shown that ascididemin can act
to damage DNA oxidatively via a thiol-dependent
conversion of oxygen into DNA-cleaving oxygen radi-
cals.16 These authors found a qualitative agreement
between the compound reduction potential observed and
the percentage of DNA cleavage. At the same time, a
trend in the direction of a statistically significant
correlation was observed between the reduction poten-
tial and the in vitro cytotoxicity in the case of the murine
leukemia cell line P388. The same relation had been
evidenced before, in the case of quinones derived from
spectronigrin.21 The results obtained from a study of 22
quinones indicated that there is a quantitative linear
relationship between their reduction potentials and the
rate at which they degrade DNA in vitro under identical
experimental conditions. We also measured the redox
potential of some of the ascididemin analogues that we
prepared, but we failed to evidence any relationship
between the redox potential and the cytotoxicity of these
different compounds (data not shown).
in vitro. In the present work, we show that these three
compounds have rather different levels of in vivo toxicity
(in the 20-160 mg/kg/day dose range). Among these
natural compounds, 11-hydroxyascididemin (compound
3) exhibits a very high MTD (> 160 mg/kg/day) but
weak cytotoxic potency (IC50 ) 3200 nM).
In a recent paper, Copp’s group report various pyri-
dine ring E analogues of ascididemin in an attempt to
determine the pharmaceutical utility and the structure-
activity requirements for the parent alkaloid.13 All of
the compounds were evaluated for selective cytotoxicity
on a wide range of biological screenings. In vitro, some
of these compounds exhibit cytotoxic levels similar to
ascididemin for human solid tumor cell lines, while they
failed to show any antitumor activity in in vivo xe-
nograft assays.
In the present study, we are interested in ring
D-modified ascididemins. The data that were obtained
in vitro show that most of the compounds have cytotoxic
properties, with some of them having levels of antitumor
activity up to 100 times higher than the parent ascidi-
demin (see Table 2). A large set of experiments are now
under way in order to characterize the mechanism(s)
of action of these new potential anticancer agents. All
In conclusion, we have succeeded in synthesizing
ascididemin analogues, some of which have levels of
antitumor cytotoxic activity equivalent to, or up to 100
times greater than, the natural parent product when
directed against a panel of 12 human cancer cell lines
(including solid tumor models of glioblastomas and

3768 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Delfourne et al.
6-(2-Acet yl-4-m et h yl-p h en yla m in o)q u in olin e-5,8-d i-
on e (A15). Method A was employed using quinoline-5,8-dione
(0.215 g, 1.35 mmol), cerium chloride (0.67 g, 2.70 mmol),
5-methyl-2-aminoacetophenone24 (0.42 g, 2.70 mmol), ethanol
(12 + 5 mL), and acetic acid (30 mL). Flash chromatography
(CH2Cl2/MeOH 95:5) gave compound A15 as a red solid (0.41
bladder, colon, breast, and lung cancers). Moreover, the
11 modified ring D analogues of ascididemin that we
constructed exhibit a significantly better cytotoxic po-
tency/systemic tolerability ratio, as indicated by a
dramatic improvement of the MTD, increasing from 20
to 40 mg/kg (natural ascididemin/2-bromoleptoclinidone)
to at least 160 mg/kg.
1
g, 98%); mp 252 °C. H NMR (CDCl3): δ 2.42 (s, 3H); 2.77 (s,
3H); 6.86 (s, 1H); 7.38 (dd, 1H, J ) 8 et 1.6 Hz); 7.52 (d, 1H,
J ) 8.0 Hz); 7.61 (dd, 1H, J ) 5.2 and 7.6 Hz); 7.74 (d, 1H, J
) 1.6 Hz); 8.46 (dd, 1H, J ) 7.6 and 5.2 Hz); 9.02 (dd, 1H, J
) 2 and 5.2 Hz); 11.18 (s, 1H). 13C NMR (CDCl3): δ 20.91,
28.50, 106.40, 120.88, 126.63, 127.53, 132.69, 133.35, 134.66,
134.83, 137.19, 143.72, 148.58, 155.08, 181.57, 182.64, 201.53.
6-(2-Acetyl-4-m eth oxy-p h en yla m in o)qu in olin e-5,8-d i-
on e (A16). Method A was employed using quinoline-5,8-dione
(3.51 g, 22.08 mmol), cerium chloride (10.9 g, 44.03 mmol),
5-methoxy-2-aminoacetophenone14 (7.29 g, 44.18 mmol), etha-
nol (200 + 90 mL), and acetic acid (500 mL). Flash chroma-
tography (CH2Cl2/MeOH 95:5) gave compound A16 as a red
Exp er im en ta l Section
Ch em istr y. Ch em ica l Syn th esis. 1H and 13C nuclear
magnetic resonance (NMR) spectra were obtained with a J EOL
400 MHz spectrometer, with the chemical shifts in the
remaining protons of the deuterated solvents serving as
internal standards. IR spectra were obtained with a Perkin-
Elmer (1600 series FTIR) spectrometer. Mass spectra (MS)
were recorded on an automass Unicam spectrometer. Reagents
were purchased from commercial sources and used as received.
Chromatography was performed on silicagel (15-40 µm) by
means of the solvent systems indicated below. The purity of
the different ascididemin analogues was evaluated on two
analytical chromatographic systems. System I consisted of a
Luna phenylhexyl, 5 mm column (150 mm × 4.6 mm), H2O/
CH3CN 65:35 at 1 mL/min flow rate, 250 nm, and system II
consisted of a Zorbax-NH2, 5 mm column (150 mm × 4.6 mm),
isooctane/EtOH/MeOH 80:10:10 at 2 mL/min flow rate, 250
nm.
N-(2-Acetyl-4-d iben zyla m in op h en yl)a ceta m id e. A mix-
ture of N-(2-acetyl-4-aminophenyl)acetamide22 (1 g, 5.21 mmol),
benzyl bromide (1.96 g, 11.46 mmol), and K2CO3 (1.58 g) in
DMF (8 mL) was refluxed for 2 h. After DMF was removed
over a vacuum, water (10 mL) was added and the reaction
media was extracted with CH2Cl2 (3 × 20 mL). The organic
layers were washed with brine, dried over MgSO4, and
concentrated to dryness. The crude product was purified by
flash chromatography (CH2Cl2/MeOH 99:1) to give the dibenzyl
derivative (1.68 g, 87%) as a yellow solid; mp 153 °C. 1H NMR
(CDCl3): δ 2.20 (s, 3H); 2.39 (s, 3H); 4.72 (s, 4H); 7.05 (dd, 1H,
J ) 9.6 and 3 Hz); 7.12 (d, 1H, J ) 3 Hz); 7.34 (m, 10H); 8.51
(d, 1H, J ) 9.6 Hz); 11.16 (s, 1H).
1
solid (4.25 g, 60%). H NMR (CDCl3): δ 2.65 (s, 3H); 3.87 (s,
3H); 6.76 (s, 1H); 7.12 (dd, 1H, J ) 2.8 and 8.8 Hz); 7.42 (d,
1H, J ) 2.8 Hz); 7.55 (d, 1H, J ) 8.8 Hz); 7.61 (dd, 1H, J )
7.6 and 4.4 Hz); 8.45 (dd, 1H, J ) 1.6 and 7.6 Hz); 9.01 (dd,
1H, J ) 1.6 and 4.4 Hz); 10.80 (s, 1H).
6-(2-Acetyl-4-d im eth yla m in o-p h en yla m in o)qu in olin e-
5,8-d ion e (A17). Method A was employed using quinoline-
5,8-dione (0.36 g, 2.26 mmol), cerium chloride (1.1 g, 4.49
mmol), 5-(dimethylamino)-2-aminoacetophenone25 (0.8 g, 4.49
mmol), ethanol (20 + 10 mL), and acetic acid (50 mL). Flash
chromatography (CH2Cl2/MeOH 95:5) gave compound A17 as
a red solid (0.65 g, 84%). 1H NMR (CDCl3): δ 2.85 (s, 3H);
3.12 (s, 6H); 6.72 (s, 1H); 6.90 (dd, 1H, J ) 2.8 and 9.2 Hz);
7.15 (d, 1H, J ) 2.8 Hz); 7.49 (d, 1H, J ) 9.2 Hz); 7.58 (dd,
1H, J ) 8.0 and 4.4 Hz); 8.43 (dd, 1H, J ) 1.6 and 8.0 Hz);
9.00 (dd, 1H, J ) 1.6 and 4.4 Hz); 10.69 (s, 1H).
6-(2-Acetyl-4-ben zyla m in o-p h en yla m in o)qu in olin e-5,8-
d ion e (A18). Method A was employed using quinoline-5,8-
dione (0.25 g, 1.57 mmol), cerium chloride (0.77 g, 3.14 mmol),
5-benzylamino-2-aminoacetophenone (0.60 g, 3.14 mmol), eth-
anol (15 + 7 mL), and acetic acid (35 mL). Flash chromatog-
raphy (CH2Cl2/MeOH 95:5) gave compound A18 as a red solid
(0.56 g, 91%), which decomposed before melting. 1H NMR
(CDCl3): δ 2.54 (s, 3H); 4.38 (s, 2H); 6.70 (s, 1H); 6.83 (dd,
1H, J ) 9.6 and 3.2 Hz); 7.08 (d, 1H, J ) 3.2 Hz); 7.30-7.37
(m, 5H); 7.43 (d, 1H, J ) 9.6 Hz); 7.58 (dd, 1H, J ) 7.6 and
4.8 Hz); 8.43 (dd, 1H, J ) 7.6 and 2.0 Hz); 9.03 (dd, 1H, J )
2.0 and 4.8 Hz); 10.67 (s, 1H).
F or m a t ion of Tet r a cyclic In t er m ed ia t es. Gen er a l
Meth od B. A solution of concentrated sulfuric acid in acetic
acid was dropped into a solution of tricyclic compound obtained
according to general procedure A. The reaction media was
refluxed for 30 min, and after it was cooled, it was poured into
ice. The mixture was neutralized with NH4OH and extracted
four times with CH2Cl2. The organic layers were dried over
MgSO4 and concentrated to dryness. The crude product was
purified by flash chromatography to give the expected tetra-
cyclic compound.
5-Ben zylam in o-2-am in oacetoph en on e. A mixture of N-(2-
acetyl-4-dibenzylaminophenyl)acetamide (1.48 g, 3.98 mmol)
in concentrated HCl (30 mL) was refluxed for 3 h. After it was
cooled, the reaction media was pourred into ice-water (100
mL), neutralized with 5 N NaOH, and extracted with CH2Cl2
(4 × 100 mL). The organic layers were dried over MgSO4 and
concentrated to dryness. The crude product was purified by
flash chromatography (CH2Cl2/MeOH 99:1) to give the benzyl
1
derivative as a yellow solid (0.63 g, 66%). H NMR (CDCl3): δ
2.45 (s, 3H); 4.27 (s, 2H); 4.75 (br. s, 1H); 5.77 (br. s, 2H); 6.55
(d, 1H, J ) 8.4 Hz); 6.76 (dd, 1H, J ) 8.4 and 2.4 Hz); 6.92 (d,
1H, J ) 2.4 Hz); 7.22-7.38 (m, 5H). 13C NMR (CDCl3): δ 28.28,
49.92, 114.60, 118.97, 123.25, 127.59, 127.90, 128.81, 138.81,
139.75, 143.59, 200.90.
For m ation of Tr icyclic In ter m ediates. Gen er al Meth od
A. A solution of quinoline-5,8-dione in ethanol was dropped
into a solution of cerium(III) chloride and aminoacetophenone
in ethanol. The reaction media was stirred at room tempera-
ture overnight, hydrolyzed with 10% acetic acid, and extracted
with CHCl3. The organic layers were dried over MgSO4 and
concentrated to dryness. The crude product was purified by
flash chromatography to give the expected product.
9-Ch lor o-11-m e t h yl-1,6-d ia za -n a p h t h a ce n e -5,12-d i-
on e (B14). Method B was employed using tricyclic compound
A14 (0.289 g, 0.88 mmol), H2SO4 cc (1.3 mL), and acetic acid
(8 + 6.5 mL). Flash chromatography (CH2Cl2/MeOH 95:5) gave
the tetracyclic compound B14 as a brown solid (0.26 g, 95%);
1
mp > 260 °C. H NMR (CDCl3): δ 3.25 (s, 3H); 7.76 (dd, 1H,
J ) 8.0 and 4.8 Hz); 7.85 (dd, 1H, J ) 8.8 and 2.0 Hz); 8.33 (d,
1H, J ) 2.0 Hz); 8.38 (d, 1H, J ) 8.8); 8.71 (dd, 1H, J ) 1.6
and 8.0 Hz); 9.15 (dd, 1H, J ) 1.6 and 4.8 Hz). 13C NMR
(CDCl3): δ 16.75, 124.67, 126.10, 128.05, 130.10, 130.65,
133.79, 133.84, 135.86, 136.50, 147.11, 147.68, 150.15, 151.76,
155.84, 181.59, 183.26.
6-(2-Ace t yl-4-ch lor o-p h e n yla m in o)q u in olin e -5,8-d i-
on e (A14). Method A was employed using quinoline-5,8-dione
(0.188 g, 1.18 mmol), cerium chloride (0.58 g, 2.36 mmol),
5-chloro-2-aminoacetophenone23 (0.4 g, 3.14 mmol), ethanol (10
+ 4 mL), and acetic acid (25 mL). Flash chromatography (CH2-
Cl2/MeOH 95:5) gave compound A14 as a red solid (0.3 g, 78%);
mp 210 °C. 1H NMR (CDCl3): δ 2. 65 (s, 3H); 6.84 (s, 1H);
9,11-Dim eth yl-1,6-d ia za -n a p h th a cen e-5,12-d ion e (B15).
Method B was employed using tricyclic compound A15 (0.4 g,
1.3 mmol), H2SO4 cc (1.9 mL), and acetic acid (12 + 9.6 mL).
Flash chromatography (CH2Cl2/MeOH 95:5) gave the tetracy-
clic compound B15 as a dark-yellow solid (0.33 g, 86%); mp >
260 °C. 1H NMR (CDCl3): δ 2.64 (s, 3H); 3.29 (s, 3H); 7.74
(dd, 1H, J ) 7.6 and 4.8 Hz); 7.75 (dd, 1H, J ) 8.4 and 1.6
Hz); 8.12 (dd, J ) 1.6 Hz); 8.33 (d, 1H, J ) 8.4); 8.71 (dd, 1H,
7.52 (dd, 1H, J 8.8 Hz); 7.57 (d, 1H, J ) 8.8 Hz); 7.63 (dd,
)
1H, J ) 8.0 and 4.4 Hz); 7.89 (d, 1H, J ) 2.4 Hz); 8.46 (dd,
1H, J ) 0.8 and 8.0 Hz); 9.02 (dd, 1H, J ) 2.0 and 5.2 Hz);
11.18 (s, 1H). 13C NMR (CDCl3): δ 28.5, 107.36, 121.86, 126.69,
126.85, 127.49, 128.39, 132.10, 134.06, 134.75, 138.36, 143.29,
148.32, 155.22, 181.28, 182.63, 200.39.

Marine Pyridoacridine Ascididemin
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3769
J ) 2.0 and 7.6 Hz); 9.13 (dd, 1H, J ) 2.0 and 4.8 Hz). 13C
NMR (CDCl3): δ 16.63, 22.37, 124.48, 125.68, 127.85, 129.91,
130.10, 132.08, 135.17, 135.76, 140.58, 146.91, 147.32, 150.31,
151.69, 155.60, 181.95, 183.61.
alkaline by the addition of concentrated NaOH. The mixture
was extracted with CH2Cl2, and the organic layer was dried
over MgSO4 and concentrated to dryness to give the expected
amino compound as a blue solid (0.32 g, 88%); mp > 260 °C.
1H NMR (CDCl3): δ 5.68 (s, 2H); 7.16 (d, 1H, J ) 7.8 Hz);
7.66 (dd, 1H, J ) 7.6 and 4.8 Hz); 7.69 (dd, 1H, J ) 7.8 and
7.8 Hz); 7.91 (d, 1H, J ) 7.8 Hz); 8.46 (d, 1H, J ) 5.2 Hz);
8.77 (dd, 1H, J ) 1.6 and 7.6 Hz); 9.17 (dd, 1H, J ) 1.6 and
4.8 Hz); 9.21 (d, 1H, J ) 5.2 Hz). 13C NMR (CDCl3): δ 109.42,
112.71, 117.70, 118.43, 124.29, 125.64, 129.12, 132.63, 132.81,
135.53, 137.27, 141.68, 148.68, 148.89, 149.03, 151.96, 154.68,
180.71. IR (CHCl3) 3510, 3400, 1676 cm-1. MS (m/z): 298 (35);
297 (100); 269 (11); 268 (8). tR was 3.47 min (99% purity) using
system I and 4.99 min (97% purity) using system II.
9-Met h oxy-11-m et h yl-1,6-d ia za -n a p h t h a cen e-5,12-d i-
on e (B16). Method B was employed using tricyclic compound
A16 (4.25 g, 13.18 mmol), H2SO4 cc (20 mL), and acetic acid
(110 + 100 mL). Flash chromatography (CH2Cl2/MeOH 100:
3) gave the tetracyclic compound B16 (which was washed with
1
diethyl ether) as a yellow solid (2.9 g, 72%); mp > 260 °C. H
NMR (CDCl3): δ 3.25 (s, 3H); 4.02 (s, 3H); 7.49 (d, 1H, J )
3.3 Hz); 7.56 (dd, 1H, J ) 3.3 and 9.3 Hz); 7.74 (dd, 1H, J )
8.3 and 4.3 Hz); 8.34 (d, 1H, J ) 9.3 Hz); 8.71 (dd, 1H, J ) 2.5
and 8.3 Hz); 9.12 (dd, 1H, J ) 2.5 and 4.3 Hz). 13C NMR
(CDCl3): δ 16.65, 55.81, 103.02, 125.73, 125.83, 127.83, 130.03,
131.47, 133.90, 135.69, 144.87, 145.46, 149.89, 150.19, 155.47,
160.54, 181.74, 183.77.
5-Br om o-9H-qu in o[4,3,2-de][1,10]ph en an tr olin -9-on e (9).
A solution of bromine (0.2 mL, 3.88 mmol) in acetic acid (5
mL) was dropped into a solution of ascididemin (0.5 g, 1.77
mmol) in acetic acid (20 mL). The reaction was refluxed for
24 h. After it was cooled, the mixture was neutralized with a
NaHCO3 saturated solution and extracted four times by CH2-
Cl2. The organic extracts were dried over MgSO4 and concen-
trated. The crude product was purified by flash chromatogra-
phy (CH2Cl2/MeOH 96:4) to give the bromo-derivative as a
yellow solid (0.548 g, 86%); mp 208 °C. 1H NMR (CDCl3): δ
7.68 (dd, 1H, J ) 4.4 and 8 Hz); 8.09 (dd, 1H, J ) 8.8 and 2.0
Hz); 8.48 (d, 1H, J ) 8.8 Hz); 8.49 (d, 1H, J ) 6.0 Hz); 8.79
(dd, 1H, J ) 2.0 and 8.0 Hz); 8.82 (d, 1H, J ) 2.0 Hz); 9.18
(dd, 1H, J ) 2.0 and 4.4 Hz); 9.30 (d, 1H, J ) 6.0 Hz). 13C
NMR (CDCl3): δ 116.76, 117.04, 118.26, 124.76, 125.81,
125.93, 129.05, 134.52, 135.43, 136.72, 137.01, 144.41, 146.24,
9-Dim eth yla m in o-11-m eth yl-1,6-d ia za -n a p h th a cen e-5,-
12-d ion e (B17). Method B was employed using tricyclic
compound A17 (0.76 g, 2.27 mmol), H2SO4 cc (3.5 mL), and
acetic acid (20 + 18 mL). Tetracyclic compound B17 was
washed with diethylic ether after workup without further
purification; it was brown solid (0.76 g, 93%); mp > 260 °C.
1H NMR (CDCl3): δ 3.17 (s, 3H); 3.21 (s, 6H); 7.04 (d, 1H, J )
3.2 Hz); 7.51 (dd, 1H, J ) 3.2 and 9.2 Hz); 7.71 (dd, 1H, J )
8 and 4.4 Hz); 8.26 (d, 1H, J ) 9.2 Hz); 8.7 (dd, 1H, J ) 1.6
and 8.0 Hz); 9.09 (dd, 1H, J ) 1.6 and 4.4 Hz). 13C NMR
(CDCl3): δ 16.40, 40.47 (2C), 101.39, 121.73, 126.11, 127.66,
130.32, 131.95, 133.50, 135.60, 142.37, 147.84, 150.40, 150.52,
155.10, 181.84, 184.38.
149.93, 150.12, 152.27, 155.67, 181.69. IR (CHCl3) 1684 cm-1
.
9-Ben zyla m in o-11-m eth yl-1,6-d ia za -n a p h th a cen e-5,12-
d ion e (B18). Method B was employed using tricyclic com-
pound A18 (4.0 g, 10 mmol), H2SO4 cc (15.1 mL), and acetic
acid (92 + 75 mL). The tetracyclic compound B18 was just
washed with diethyl ether after workup without further
MS (m/z): 363 (99); 362 (83); 361 (100); 360 (27); 255 (9); 254
(51). tR was 10.73 min (97% purity) using system I and 3.52
min (99% purity) using system II.
5-Am in o-9H -q u in o[4,3,2-d e][1,10]p h e n a n t h r olin -9-
on e (10). A solution of bromoascididemin 9 (9.0 g, 24.9 mmol)
and NaN3 (9.0 g, 120 mmol) in DMF (800 mL) was refluxed
for 9 h. DMF was removed over vacuum, and 1 N KOH and
CHCl3/MeOH 95:5 (500 mL) were added. The mixture was
filtered to recover the precipitate, and the organic layer was
extracted by CHCl3/MeOH 95:5 (5 × 100 mL). The organic
layers were dried over MgSO4 and concentrated to give a solid,
which was added to the previous precipitate. These solids were
washed by CHCl3 and recrystallized in MeOH to give the
amino compound as a black solid (6.5 g, 88%); mp > 260 °C.
1H NMR (CDCl3): δ 7.43 (dd, 1H, J ) 8.8 and 2.4 Hz); 7.74
(dd, 1H, J ) 4.8 and 8.0 Hz); 7.81 (d, 1H, J ) 2.4 Hz); 8.48 (d,
1H, J ) 6.0 Hz); 8.50 (d, 1H, J ) 8.8 Hz); 8.90 (dd, 1H, J )
2.0 and 8.0 Hz); 9.25 (dd, 1H, J ) 2.0 and 4.8 Hz); 9.29 (d, 1H,
1
purification; black solid (3.58 g, 98%); mp > 260 °C. H NMR
(CDCl3): δ 3.09 (s, 3H); 4.52 (d, 2H, J ) 5.4 Hz); 4.86 (t, 1H,
J ) 5.4 Hz); 7.06 (d, 1H, J ) 2.8 Hz); 7.29 (dd, 1H, J ) 9.2
and 2.8 Hz); 7.3-7.43 (m, 5H); 7,71 (dd, 1H, J ) 4.8 and 8
Hz); 8.20 (d, 1H, J ) 9.8 Hz); 8.69 (dd,1H, J ) 1.6 and 8 Hz);
9.09 (dd, 1H, J ) 1.6 and 4.8 Hz).
F or m a tion of P en ta cycles. Gen er a l Meth od C. A solu-
tion of tetracyclic intermediate obtained according to general
method B and DMF-DEA in DMF was refluxed for 1 h. After
it was concentrated to dryness, NH4Cl and ethanol were added
and the mixture was refluxed for an additional 30 min. The
solvent was removed over vacuum, water was added, and the
mixture was extracted four times with CH2Cl2. The organic
layers were dried over MgSO4 and concentrated. The crude
product was purified to give the expected pentacyclic deriva-
tive.
7-Nitr o-9H-qu in o[4,3,2-de][1,10]ph en an th r olin -9-on e (6).
Ascididemin (2 g, 7.06 mmol) was dropped at 0 °C into a
solution of sulfuric acid (45 mL) and nitric acid (45 mL). The
reaction mixture was warmed at 130 °C for 2 h, and after it
was cooled, it was poured into ice (400 g). The precipitate was
recovered by filtration, and a mixture CH2Cl2/NH4OH/H2O
(600:1:300) was added. The organic layer was separated, and
the aqueous layer was extracted with CH2Cl2. The organic
extracts were dried on MgSO4 and concentrated over a vacuum
to give the pentacyclic compound 6 as a yellow solid (1.62 g,
70%); mp 224 °C. 1H NMR (CDCl3): δ 7.69 (dd, 1H, J ) 4.4
and 8 Hz); 8.04 (dd, 1H, J ) 8.0 and 8.0 Hz); 8.28 (d, 1H, J )
8 Hz); 8.56 (d, 1H, J ) 5.2 Hz); 8.75 (dd, 1H, J ) 2.0 and 8.0
Hz); 8.89 (dd, 1H, J ) 1.2 and 8 Hz); 9.18 (dd, 1H, J ) 4.4 and
2.0 Hz); 9.37 (d, 1H, J ) 5.6 Hz). 13C NMR (CDCl3): δ 79.20,
117.61, 118.39, 124.21, 124.89, 125.98, 127.54, 129.04, 130.14,
135.62, 136.63, 148.17, 149.76, 149.94, 150.12, 151.66, 154.88,
180.56. IR (CHCl3): 1689 cm-1. MS (m/z): 328 (18); 327 (100);
299 (22); 297 (9); 269 (10); 253 (24); 242 (11); 241 (33). tR is
8.51 min (97% purity), using system I, and tR is 11.99 min
(100% purity), using system II.
J
) 6.0 Hz).13C NMR (DMSO): δ 102.26, 117.13, 118.54,
121.62, 123.20, 125.34, 126.11, 129.18, 133.80, 134.83, 135.47,
138.42, 147.65, 148.29, 151.63, 152.39, 154.32, 180.35. IR (KBr)
3312, 3200, 1692 cm-1. MS (m/z): 298 (32); 297 (100); 269 (4);
268 (1). tR was 2.56 min (100% purity) using system I and 10.6
min (97% purity) using system II.
5-(2-Ch lor oeth yla m in o)-9H-qu in o[4,3,2-d e][1,10]p h en -
a n th r olin -9-on e (11) a n d 5-Bis(2-ch lor oeth yla m in o)-9H-
qu in o[4,3,2-d e][1,10]p h en a n th r olin -9-on e (12). NaBH3CN
(0.63 g, 10 mmol) was added portionwise at 0 °C into a solution
of aminoascididemin 10 (1 g, 3.95 mmol) and chloroacetalde-
hyde (50% in water, 2.6 mL, 16.8 mmol) in acetic acid (30 mL).
The reaction was maintained at 0 °C for 5 min and at room
temperature for a further 30 min. The mixture was made
alkaline by the addition of an NaHCO3-saturated solution and
extracted by CHCl3/MeOH 95:5. The organic layers were dried
over MgSO4 and concentrated. The crude product was purified
by flash chromatography (CHCl3 and then CHCl3/MeOH 99:
1) to give compounds 11 and 12.
1
Com p ou n d 11. Purple solid (0.22 g, 18%); mp 196 °C. H
NMR (CDCl3): δ 3.81 (t, 2H, J ) 5.5 Hz); 3.88 (t, 2H, J ) 5.5
Hz); 5.01 (br.s, 1H); 7.34 (dd, 1H, J ) 8.8 and 2.5 Hz); 7.60 (d,
1H, J ) 2.5 Hz); 7.65 (dd, 1H, J ) 7.5 and 4.4 Hz); 8.41 (d,
1H, J ) 5.8 Hz); 8.43 (d, 1H, J ) 8.8 Hz); 8.82 (dd, 1H, J )
7.5 and 1.5 Hz); 9.15 (dd, 1H, J ) 4.4 and 1.5 Hz); 9.21 (dd,
1H, J ) 5.8 Hz). 13C NMR (CDCl3): δ 42.83, 45.01, 100.76,
116.81, 118.78, 120.85, 125.38, 126.35, 129.35, 135.04, 136.04,
136.43, 140.22, 141.56, 148. IR (CHCl3): 3427, 1672, 1620
7-Am in o-9H -q u in o[4,3,2-d e][1,10]p h e n a n t h r olin -9-
on e (8). A suspension of nitro-derivative 6 (0.4 g, 1.22 mmol)
and iron (0.37 g, 6.59 mmol) in AcOH/H2O (10 mL:10 mL) was
refluxed for 1 h. Ethylenediaminetetraacetic acid (EDTA, 1.94
g, 6.59 mmol) was added, and the reaction media was made

3770 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17
Delfourne et al.
cm-1. tR was 5.22 min (98% purity) using system I and 5.83
min (99% purity) using system II.
as a green solid (170 mg, 66%); mp > 260 °C. 1H NMR
(CDCl3): δ 4.10 (s, 3H); 7.62 (dd, 1H, J ) 9.2 Hz, J ) 2.4 Hz);
7.66 (dd, 1H, J ) 4.4 and 8.0 Hz); 7.96 (d, 1H, J ) 2.4 Hz);
8.48 (d, 1H, J ) 2.4 Hz); 8.54 (d, 1H, J ) 9.2 Hz); 8.80 (dd,
1H, J ) 2.4 and 8.0 Hz); 9.16 (dd, 1H, J ) 4.4 and 2.4 Hz);
9.25 (d, 1H, J ) 5.2 Hz). 13C NMR (CDCl3): δ 30.93, 116.86,
118.41, 122.44, 125.56, 129.25, 134.96, 136.55, 137.13, 141.52,
143.67, 149.11, 149.77, 152.37, 155.38, 161.71, 181.93, 207.00.
IR (CHCl3): 1677, 1616 cm-1. MS (m/z): 313 (26); 312 (100);
285 (2); 284 (15); 269 (15); 242 (33). tR was 4.93 min (99%
purity) using system I and 5.09 min (100% purity) using
system II.
5-(D im e t h y la m in o )-9H -q u in o [4,3,2-d e][1,10]p h e n -
a n th r olin -9-on e (17). Method C was employed using tetra-
cyclic compound B17 (0.25 g, 0.79 mmol) and dimethylforma-
mide diethylacetal (1.5 mL, 8.75 mmol) in DMF (5 mL), NH4Cl
(1 g, 18.7 mmol), and ethanol (16 mL). The crude product was
purified by flash chromatography (CH2Cl2/MeOH 100: 5) to
give compound 17 as a purple solid (170 mg, 66%); mp > 260
°C. 1H NMR (CDCl3): δ 3.25 (s, 6H); 7.45 (dd, 1H, J ) 9.2 and
3.0 Hz); 7.57 (d, 1H, J ) 3.0 Hz); 7.63 (dd, 1H, J ) 4.4 and 8.0
Hz); 8.41 (d, 1H, J ) 9.2 Hz); 8.43 (d, 1H, J ) 5.6 Hz); 8.81
(dd, 1H, J ) 2.0 and 7.6 Hz); 9.13 (dd, 1H, J ) 4.4 and 2.0
Hz); 9.17 (d, 1H, J ) 5.6 Hz). 13C NMR (CDCl3): δ 40.45,
100.84, 116.81, 118.69, 118.99, 125.19, 126.10, 129.46, 134.62,
136.03, 136.30, 139.00, 140.69, 148.16, 149.15, 151.53, 152.47,
154.83, 181.65. IR (CHCl3): 1666, 1615 cm-1. MS (m/z): 326
(35); 325 (100); 324 (100); 254 (16); 253 (14). tR was 4.38 min
(98% purity) using system I and 5.4 min (98% purity) using
system II.
5-(Ben zylam in o)-9H-qu in o[4,3,2-de][1,10]ph en an th r olin -
9-on e (18). Method C was employed using tetracyclic com-
pound B18 (3.58 g, 9.45 mmol), dimethylformamide diethyl-
acetal (5.7 mL, 33.26 mmol) in DMF (19 mL), NH4Cl (2.95 g,
55 mmol), and ethanol (50 mL). The crude product was purified
by flash chromatography (CH2Cl2/MeOH 96:4) to give com-
pound 18 as a pink-purple solid (2 g, 55%); mp 219 ° C. 1H
NMR (CDCl3): δ 4.61 (d, 2H); 5.10 (t, 1H); 7.31 (dd, 1H, J )
8.8 and 2.4 Hz); 7.452-7.327 (m, 5H); 7.55 (d, 1H, J ) 2,4
Hz); 7.63 (dd, 1H, J ) 4.4 and 8.4 Hz); 8.29 (d, 1H, J ) 5.2
Hz); 8.36 (d, 1H, J ) 8.8 Hz); 8.79 (dd, 1H, J ) 1.2 and 8.4
Hz); 9.13 (dd, 1H, J ) 4.4 and 1.2 Hz); 9.14 (d, 1H, J ) 5.2
Hz). IR (CHCl3): 3428, 1668, 1620 cm-1. MS (m/z): 388 (7);
387 (100); 386 (85); 385 (25); 369 (99); 368(44). tR was 11.03
min (97% purity) using system I and 5.01 min (99% purity)
using system II.
Com p ou n d 12. Pink solid (0.14 g,10%); mp 220 °C. 1H NMR
(CDCl3): δ 3.83 (t, 4H, J ) 7.0 Hz); 4.04 (t, 4H, J ) 7.0 Hz);
7.47 (dd, 1H, J ) 9.5 and 2.9 Hz); 7.66 (dd, 1H, J ) 8.0 and
4.4 Hz); 7.70 (d, 1H, J ) 2.9 Hz); 8.42 (d, 1H, J ) 5.6 Hz);
8.50 (d, 1H, J ) 9.5 Hz); 8.81 (dd, 1H, J ) 8.0 and 1.8 Hz);
9.16 (dd, 1H, J ) 4.4 and 1.8 Hz); 9.23 (d, 1H, J ) 5.6 Hz). 13
C
NMR (CDCl3): δ 40.16, 53.60, 101.70, 116.60, 118.37, 118.68,
125.39, 125.91, 129.25, 135.13, 136.12, 136.38, 139.42, 141.93,
148.24, 148.73, 149.34, 152.22, 155.08, 181.43. IR (KBr): 1666;
1650 cm-1. tR was 3.39 min (97% purity) using system I and
4.82 min (100% purity) using system II.
5-(Dim et h yla m in o-2-et h yla m in o)-9H -q u in o[4,3,2-d e]-
[1,10]p h en a n th r olin -9-on e (13). Trifluoroacetic acid (25 mL,
166 mmol) was added dropwise at 0 °C into a mixture of
compound 10 (2.56 g, 8.59 mmol) and dimethylaminoacetal-
dehyde diethyl acetal (7.7 mL, 43.3 mmol). The reaction was
stirred for 5 min, and sodium cyanoborohydride (8.2 g, 130
mmol) was added portionwise. The reaction was warmed at
95 °C for 18 h. An NaHCO3-saturated solution (600 mL) was
added, and the mixture was extracted by CHCl3/MeOH 95:5
(3 × 800 mL). The organic layers were washed with water and
dried over MgSO4. The solvent was removed over a vacuum,
and the crude product was purified by filtration on alumina
(CHCl3, CHCl3/MeOH 95:5) to give compound 13 as a black
solid (1.15 g, 36%), which decomposes before melting. 1H NMR
(CDCl3): δ 2.37 (s, 6H); 2.62 (t, 2H, J ) 7.32 Hz); 3.70 (t, 2H,
J ) 7.32 Hz); 7.39 (dd, 1H, J ) 9.2 and 3.0 Hz); 7.62 (dd, 1H,
J ) 8.0 and 4.5 Hz); 7.66 (d, 1H, J ) 3.0 Hz); 8.35 (d, 1H, J )
9.2 Hz); 8.38 (d, 1H, J ) 5.7 Hz); 8.79 (dd, 1H, J ) 8.0 and 1.8
Hz); 9.12 (dd, 1H, J ) 4.5 and 1.8 Hz); 9.15 (d, 1H, J ) 5.7
Hz). 13C NMR (CDCl3): δ 45.97, 50.31, 56.40, 101.05, 116.81,
118.48, 118.89, 125.22, 126.30, 129.35, 134.87, 135.97, 136.32,
138.91, 140.55, 148.25, 148.98, 149.69, 152.23, 154.82, 181.37.
IR (CHCl3): 1663 cm-1. MS: m/z 369 (100); 354 (15); 236 (37).
tR was 10.37 min (98% purity) using system I and 6.86 min
(97% purity) using system II.
5-Ch lor o-9H -q u in o[4,3,2-d e][1,10]p h en a n t r olin -9-on e
(14). Method C was employed using tetracyclic compound B14
(0.25 g, 0.81 mmol) and dimethylformamide diethylacetal (1.5
mL, 8.75 mmol) in DMF (4.5 mL), NH4Cl (2.95 g, 55 mmol),
and ethanol (50 mL). The crude product was purified by flash
chromatography (CH2Cl2/MeOH 98:2) to give compound 14 as
1
a yellow solid (60 mg, 23%); mp 200 °C. H NMR (CDCl3): δ
7.68 (dd, 1H, J ) 8.4 and 4.8 Hz); 7.94 (dd, 1H, J ) 8.8 and
2.0 Hz); 8.46 (d, 1H, J ) 5.6 Hz); 8.55 (d, 1H, J ) 8.8 Hz);
8.63 (d, 1H, J ) 2.0 Hz); 8.79 (dd, 1H, J ) 2.0 and 8.4 Hz);
4-B r o m o -5-a m in o -9-H -q u in o [4,3,2-d e][1,10]p h e n -
a n th r olin -9-on e (19). Bromine (35 mL, 0.67 mmol) was added
into a suspension of compound 10 (0.2 g, 0.67 mmol) in acetic
acid (8 mL). The reaction was warmed at 50 °C for 6 h. After
it was concentrated, the mixture was made alkaline by the
addition of 5 N NaOH (20 mL) and extracted by CHCl3/MeOH
95:5 (4 × 100 mL). The organic layers were dried over MgSO4
and concentrated to give compound 19, which was recrystal-
lized in CHCl3/pentane 20 mL:15 mL (152 mg, 61%); mp >
9.18 (dd, 1H, J ) 4.8 and 2.0 Hz); 9.30 (d, 1H, J ) 5.6 Hz).13
C
NMR (CDCl3): δ 117.07, 118.46, 122.98, 124.82, 126.12, 129.34,
133.02, 134.81, 137.00, 137.42, 137.79, 144.45, 146.35, 150.24,
150.45, 152.55, 156.02, 181.9. IR (CHCl3): 1684, 1602 cm-1
.
MS (m/z): 319 (43); 318 (15); 317 (100); 291 (14,5); 290 (18);
289 (100). tR was 8.30 min (96% purity) using system I and
3.58 min (97% purity) using system II.
1
260 °C. H NMR (DMSO-d6): δ 7.07 (br. s, 2H); 7.61 (d, 1H, J
5-Me t h yl-9H -q u in o[4,3,2-d e][1,10]p h e n a n t h r olin -9-
on e (15). Method C was employed using tetracyclic compound
B15 (1.0 g, 3.47 mmol) and dimethylformamide diethylacetal
(1.8 mL, 10.41 mmol) in DMF (7 mL), NH4Cl (2.77 g, 52 mmol),
and ethanol (50 mL). The recrystallization of the crude product
gave compound 15 as a yellow solid (0.7 g, 67%); mp 200 °C.
1H NMR (CDCl3): δ 2.69 (s, 3H); 7.65 (dd, 1H, J ) 8.0 and 4.8
Hz); 7.81 (dd, 1H, J ) 8.0 and 1.2 Hz); 8.44 (d, 1H, J ) 1.2
Hz); 8.49 (d, 1H, J ) 8.0 Hz); 8.50 (d, 1H, J ) 5.6 Hz); 8.78
(dd, 1H, J ) 2.0 and 8.0 Hz); 9.15 (dd, 1H, J ) 4.8 and 2.0
Hz); 9.24 (d, 1H, J ) 5.6 Hz). 13C NMR (CDCl3): δ 22.06,
116.54, 117.87, 122.15, 123.12, 125.24, 128.74, 132.58, 133.47,
136.25, 137.19, 141.63, 143.88, 144.79, 149.16, 149.31, 152.09,
155.15, 181.53. IR (CHCl3): 1681, 1622 cm-1. MS (m/z): 297
(18); 296 (34); 268 (25); 149 (50). tR was 5.44 min (99% purity)
using system I and 3.88 min (100% purity) using system II.
5-Met h oxy-9H -q u in o[4,3,2-d e][1,10]p h en a n t h r olin -9-
on e (16). Method C was employed using tetracyclic compound
B16 (2.0 g, 6.57 mmol) and dimethylformamide diethylacetal
(4 mL, 23.34 mmol) in DMF (14 mL), NH4Cl (8 g, 149.5 mmol),
and ethanol (130 mL). The crude product was purified by flash
chromatography (CH2Cl2/MeOH 100: 5) to give compound 16
) 8.8 Hz); 7.77 (dd, 1H, J ) 7.7 and 4.0 Hz); 8.18 (d, 1H, J )
8.8 Hz); 8.61 (d, 1H, J ) 7.7 Hz); 9.10 (d, 1H, J ) 4.0 Hz);
9.14 (d, 1H, J ) 5.9 Hz); 9.91 (d, 1H, J ) 5.9 Hz). IR (CHCl3):
3501; 3400; 1673 cm-1. MS: m/z 378 (42); 377 (100); 376 (48);
375 (27). tR was 5.66 min (97% purity) using system I and 7.03
min (99% purity) using system II.
P h a r m a cology. In Vitr o Ch a r a cter iza tion of th e Dr u g-
In d u ced Effects on Hu m a n Ca n cer Cell Lin e Gr ow th .
Twelve human tumor cell lines were obtained from the
American Type Culture Collection (ATCC, Manassas, VA).
These included three glioblastomas (SW1088, U-373 MG, and
U-87 MG), two colon (HCT-15 and LoVo), two nonsmall-cell-
lung (A549 and A-427), two bladder (J 82 and T24), one
prostate (PC-3), and two breast (T-47D and MCF7) cancer
models. The ATCC numbers of these cell lines are HTB 12
(SW1088), HTB 14 (U-87 MG), HTB 17 (U-373 MG), CCL225
(HCT-15), CCL229 (LoVo), CCL 185 (A549), HBT 53 (A-427),
HTB1 (J 82), HTB4 (T24), HTB133 (T-47D), HTB22 (MCF7),
and CRL1435 (PC-3). The cells were cultured at 37 °C in sealed
(airtight) Falcon plastic dishes (Nunc, Gibco, Belgium) con-
taining Eagle’s minimal essential medium (MEM, Gibco)

Marine Pyridoacridine Ascididemin
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 17 3771
(8) Bloor, S. J .; Schmitz, F. J . A novel pentacyclic aromatic alkaloid
from an ascidian. J . Am. Chem. Soc. 1987, 109, 6134-6136.
(9) This compound is 10-hydroxyascididemin according to IUPAC
rules of nomenclature. Schmitz, F. J .; de Guzman, F. S.; Hossain,
M. B.; van der Helm, D. Cytotoxic aromatic alkaloids from the
ascidian Amphicarpa meridiana and Leptoclinides sp.: Meridine
and 11-hydroxyascididemin. J . Org. Chem. 1991, 56, 804-808.
(10) Cimino, G.; Crispino, A.; De Rosa, S.; De Stefano, S.; Gavagnin,
M.; Sodano, G. Studies on the structure of calliactine, the
zoochrome of the sea anemone Calliactis parasita. Tetrahedron
1987, 43, 4023-4030.
(11) This compound is 10-methoxyxyascididemin according to IUPAC
nomenclature rules. Kitahara, Y.; Nakahara, S.; Yonezawa, T.;
Nagatsu, M.; Kubo, A. Total synthesis of kuanoniamine A, 11-
hydroxyascididemin and neocalliactine acetate. Heterocycles
1993, 36, 943-946.
supplemented with 5% fetal calf serum (FCS). All of the media
were supplemented with a mixture of 0.6 mg/mL glutamine
(Gibco), 200 IU/mL penicillin (Gibco), 200 IU/mL streptomycin
(Gibco), and 0.1 mg/mL gentamycin (Gibco). The FCS was
heat-inactivated for 1 h at 56 °C.
The 12 cell lines were incubated for 24 h in 96 microwell
plates (at a concentration of 40 000 cells/mL culture medium)
to ensure adequate plating prior to the determination of the
cell growth. This process was carried out by means of the
colorimetric MTT assay, as detailed previously.26,27 This as-
sessment of cell population growth is based on the capability
of living cells to reduce the yellow product MTT (Sigma, St.
Louis, MO) to a blue product, formazan, by a reduction reaction
occurring in the mitochondria. The number of living cells is
directly proportional to the intensity of the blue, which is
quantitatively measured by spectrophotometry on a DIAS
microplate reader (Dynatech Laboratories, Guyancourt, France)
at a 570 nm wavelength (with a reference of 630 nm). Each
experiment was carried out in sextuplicate. We validated the
MTT-related data using two alternative techniques, namely,
direct cell counting and the genomic incorporation of tritiated
thymidine (data not shown). Six concentrations ranging from
10-5 to 10-9 M were assayed for each of the 17 compounds
under study (see Table 2).
In Vivo Deter m in a tion of Dr u g-In d u ced Toxicity.
Drug-induced toxicity can be monitored in vivo by determining
the MTD. This MTD determination is carried out by defining
the maximum dose of the drug that can be administered
acutely (i.e., in one intraperitoneal single dose) to healthy
animals (B6D2F1 mice, Iffa Credo), i.e., not grafted with
tumors. The survival and weight of the animals are recorded
for up to 28 days postinjection. Six different doses of each drug
(5, 10, 20, 40, 80, and 160 mg/kg) are used for the MTD index
determination, with each experimental group being composed
of three mice for this purpose.
(12) Gellerman, G.; Rudi, A.; Kashman, Y. Biomimetic synthesis of
ascididemin and derivatives. Synthesis 1994, 239-241.
(13) Lindsay, B. S.; Christiansen, H. C.; Copp, B. R. Structural
studies of cytotoxic marine alkaloid: synthesis of novel ring-E
analogues of ascididemin and their in vitro and in vivo biological
evaluation. Tetrahedron 2000, 56, 497-505.
(14) Bracher, F. The structure of neocalliactine acetate: proof by total
synthesis. Liebigs Ann. Chem. 1992, 1205-1208.
(15) Kitahara, Y.; Nakahara, S.; Yonezawa, T.; Nagatsu, M.; Shibano,
Y.; Kubo, A. Synthetic studies on pentacyclic aromatic alkaloids,
kuanoniamine A, 11-hydroxyascididemin and neocalliactine
acetate. Tetrahedron 1997, 53, 17029-17038.
(16) Matsumoto, S. S.; Sidford, M. H.; Holden, J . A.; Barrows, L. R.;
Copp, B. R. Mechanism of action studies of cytotoxic marine
alkaloids: ascididemin exhibits thiol-dependent oxidative DNA
cleavage. Tetrahedron Lett. 2000, 41, 1667-1670.
(17) Delfourne, E.; Darro, F.; Bontemps-Subielos, N.; Decaestecker,
C.; Bastide, J .; Frydman, A.; Kiss, R. Synthesis and character-
ization of the antitumor activities of analogues of meridine a
marine pyridoacridine alkaloid. J . Med. Chem. 2001, 44, 3275-
3282.
(18) Bracher, F. Total synthesis of the pentacyclic alkaloid ascidi-
demin. Heterocycles 1989, 29, 2093-2095.
(19) Kitahara, Y.; Tamura, F.; Kubo, A. Synthesis of cystodamine, a
pentacyclic aza-aromatic alkaloid. Tetrahedron Lett. 1997, 38,
4441-4442.
(20) Weinstein, J . N.; Kohn, K. W.; Grever, M. R.; Viswanadhan, V.
N.; Rubinstein, L. V.; Monks, A. P.; Scudiero, D. A.; Welch, L.;
Koutsoukos, A. D.; Chiausa, A. J .; Paull, K. D. Neural computing
in cancer drug development: predicting mechanism of action.
Science 1992, 258, 447-449.
Sta tistica l An a lysis. The statistical comparisons of the
data were carried out by means of the Fisher F (one-way
variance analysis for more than two groups) or the Student t
(for two groups) tests after a check of the equality of variance
by means of the Levene test and of the normal distribution
fitting of the data by means of the c2 test of goodness-of-fit.
When these parametric conditions were not satisfied, the
nonparametric Kruskall-Wallis (for more than two groups)
or the Mann-Whitney (for two groups) tests were carried out.
All of the statistical analyses were carried out using Statistica
(Statsoft, Tulsa, OK).
(21) Shaikh, I. A.; J onhson, F.; Grollman, A. P. Streptonigrin 1.
Structure-activity relationships among simple bicyclic ana-
logues. Rate dependence of DNA degradation on quinone reduc-
tion potential. J . Med. Chem. 1986, 29, 1329-1340.
(22) McIntyre, J .; Simpson, J . C. E. Synthesis of azocinnoline
derivatives. J . Chem. Soc. 1952, 2606-2611.
(23) Ravi, S.; Saravanan, N.; Shanti, A.; Dharmaraj, N.; Laksha-
manan, A. J . Thermal Fries rearrangement of anilides. Ind. J .
Chem. Sect. B 1991, 30, 443-445.
(24) Basha, A.; Syed, S. A.; Tanveer, A. F. A new method for
rearrangement of anilides. Tetrahedron Lett. 1976, 17, 3217-
3220.
(25) Bandurco, V. T.; Schwender, C. F.; Bell, S. C.; Combs, D. W.;
Kanojia, R. M.; Ramesh, M. Synthesis and cardiotonic activity
of a series of substituted 4-alkyl-2(1H)-quinazolines. J . Med.
Chem. 1987, 30, 1421-1426.
(26) Pauwels, O.; Kiss, R.; Pasteels, J . L.; Atassi, G. Characterization
of alkylating versus intercalating anticancer drug-induced effects
on cell survival, cell cycle kinetic and morphonuclear pattern of
three neoplastic cell lines growing in vitro. Pharm. Res. 1995,
12, 1011-1018.
(27) Camby, I.; Salmon, I.; Danguy, A.; Pasteels, J . L.; Brotchi, J .;
Martinez, J .; Kiss, R. Influence of gastrin on human astrocytic
tumor cell proliferation. J . Natl. Cancer Inst. 1996, 88, 594-600.
Refer en ces
(1) Pettit, G. R.; Kamano, Y.; Herald, C. L. Antineoplastic agents
CXVIII: isolation and structure of bryostatin 9. J . Nat. Prod.
1986, 49, 661-664.
(2) Pettit, G. R.; Kamano, Y.; Fujii, Y.; Herald, C. L.; Inoue, M.;
Brown, P.; Gust, D.; Kitahara, K.; Schmidt, J . M.; Doubek, D.
L.; Michel, C. Marine animal biosynthetic constituents for cancer
chemotherapy. J . Nat. Prod. 1981, 44, 482-485.
(3) Smith, C. D.; Zhang, X.; Mooberry, S. L.; Patterson, G. M.; Moore,
R. E. Cryptophycin: a new antimicrotubule agent against drug-
resistant cells. Cancer Res. 1994, 54, 3779-3784.
(4) Rinehart, K. L.; Lithgow-Bertelloni, A. M. Dehydrodidemnin B.
Patent WO 91/04985. Chem. Abstr. 1991, 115, 248086 q.
(5) Rinehart, K. L.; Holt, T. G.; Frejeau, N. L.; Stroh, J . G.; Keifer,
P. A.; Sun, F.; Li, L. H.; Martin, D. G. Ecteinascidins 729, 743,
745, 759A, 759B and 770: potent antitumor from the caribbean
tunicate Ecteinascidia turbinata. J . Org. Chem. 1990, 55, 4512-
4515.
(6) (a) Molinski, T. F. Marine pyridoacridine alkaloids: structure,
synthesis, and biological chemistry. Chem. Rev. 1993, 93, 1825-
1838. (b) Ding, Q.; Chichak, K.; Lown, J . W. Pyrroloquinoline
and pyridoacridine alkaloids from marine source. Curr. Med.
Chem. 1999, 6, 1-27.
(7) Kobayashi, J .; Cheng, J . F.; Nakamura, H.; Ohizumi, Y.; Hirata,
Y.; Sasaki, T.; Ohta, T.; Nozoa, S. Ascididemin, a novel penta-
cyclic aromatic alkaloid with potent antileukemic activity from
Okinawan tunicate Didemnum sp. Tetrahedron Lett. 1988, 29,
1177-1180.
J M0208774