Synthesis and biological evaluation of 2,7-Dihydro-3H-dibenzo[de,h]cinnoline-3,7-dione derivatives, a novel group of anticancer agents active on a multidrug resistant cell line

Article abstract of DOI:10.1016/S0968-0896(02)00425-X

A series of anthrapyridazone derivatives with one or two basic side chains at various positions in the tetracyclic chromophore have been synthesized. The key intermediates in the synthesis are 2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-diones 1, 12 and 15 monosubstituted at position 2 (4d, 16a-e), or 6 (2a-f) or disubstituted at positions 2 and 6 (4a-c) or 2 and 8 (17a-e) with appropriate alkylaminoalkylamines. All analogues showed in vitro cytotoxic activity against murine leukemia (L1210) and human leukemia (K562) cell lines. The compounds were also active against human leukemia multidrug resistant (K562/DX) cell line with resistance index (RI) in the range 1-3 depending on the compound's structure. Two of the most active in vitro compounds 4a and 11 were tested in vivo against murine P388 leukemia and displayed antileukemic activity comparable with that of Mitoxantrone. DNA-binding assays were performed and DNA affinity data were correlated with the structures of the compounds. The cytoplasmatic membrane affinity values (log k′_IAM) have also been determined and the correlation with the resistance indexes discussed. The anthrapyridazones constitute a novel group of antitumor compounds that can overcome multidrug resistance.

Full text of DOI:10.1016/S0968-0896(02)00425-X

Bioorganic & Medicinal Chemistry 11 (2003) 561–572
Synthesis and Biological Evaluation of 2,7-Dihydro-3H-
dibenzo[de,h]cinnoline-3,7-dione Derivatives, a Novel Group of
Anticancer Agents Active on a Multidrug Resistant Cell Line
Barbara Stefanska,a Ma•gorzata Arciemiuk,a Maria M. Bontemps-Gracz,a
Maria Dzieduszycka,a Agnieszka Kupiec,a Sante Martellib and Edward Borowskia,*
a
´ ´
Department of Pharmaceutical Technology and Biochemistry, Gdansk University of Technology 80-952 Gdansk, Poland
bDepartment of Chemistry, University of Camerino, 62033 Camerino, Italy
Received 2 April 2002; revised 9 August 2002; accepted 22 August 2002
Abstract—A series of anthrapyridazone derivatives with one or two basic side chains at various positions in the tetracyclic chro-
mophore have been synthesized. The key intermediates in the synthesis are 2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-diones 1, 12
and 15 monosubstituted at position 2 (4d, 16a–e), or 6 (2a–f) or disubstituted at positions 2 and 6 (4a–c) or 2 and 8 (17a–e) with
appropriate alkylaminoalkylamines. All analogues showed in vitro cytotoxic activity against murine leukemia (L1210) and human
leukemia (K562) cell lines. The compounds were also active against human leukemia multidrug resistant (K562/DX) cell line with
resistance index (RI) in the range 1–3 depending on the compound’s structure. Two of the most active in vitro compounds 4a and
11 were tested in vivo against murine P388 leukemia and displayed antileukemic activity comparable with that of Mitoxantrone.
DNA-binding assays were performed0 and DNA affinity data were correlated with the structures of the compounds. The cytoplas-
matic membrane affinity values (log k IAM) have also been determined and the correlation with the resistance indexes discussed. The
anthrapyridazones constitute a novel group of antitumor compounds that can overcome multidrug resistance.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
potency depends on the structure of the side chains and
other substituents.
The prolonged clinical use of chemotherapeutic agents
often causes the appearance of multidrug resistance
(MDR) toward numerous antitumor compounds. This
effect currently constitutes one of the major problems in
clinical chemotherapy and has not yet been successfully
solved.1À3 Many efforts have been directed toward the
search for new antitumor anthracenedione derivatives
with increased effectiveness against MDR tumor cell
lines. This has resulted in the design of anthrapyrazoles4
and their aza-analogues,5 benzoperimidines6 and anthra-
pyridones,7 which fulfill this requirement. Recently we
postulated that the presence of a heterocyclic ring con-
densed with the anthracenedione or with the structurally
related acridone chromophore determines cytotoxic
activity toward the multidrug resistant cell lines.6
Although the presence of a fused heterocyclic ring seems
to be essential for overcoming MDR, the cytotoxic
In this paper, we describe the synthesis and biological
evaluation of a new group of anthracenedione deriva-
tives with a pyridazone ring incorporated into the chro-
mophore, namely 2,7-dihydro-3H-dibenzo[de,h]cinnoline-
3,7-diones (hereafter referred to as anthrapyridazones),
to determine the relationships between structure and
antitumor activity of derivatives of this novel group. A
series of compounds bearing [(alkylamino)-alkyl]amino
side chain attached at positions 2, 6 or 8 to the chro-
mophore moiety has been obtained. The presence of a
second basic side chain could increase not only the
binding to DNA, but also modify the physicochemical
properties and solubility of the above derivatives and
enhances their cytotoxic activity. Therefore, the 2,6- as
well as the 2,8-disubstituted analogues have been syn-
thesized (Fig. 1).
As evidenced earlier, the presence of hydroxyl groups in
related ring systems such as anthracenediones,8 anthra-
pyrazoles,9 benzoperimidines6 and acridine derivatives10
*Corresponding author. Tel.: +48-58-347-2523; fax: +48-58-347-
1893; e-mail: borowski@altis.chem.pg.gda.pl
0968-0896/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00425-X

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
562
Figure 1. Structure of 2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-diones.
increases antitumor activity in these classes of com-
pounds. For this reason the 2,7-dihydro-3H-dibenzo
[de,h]cinnoline-3,7-dione derivatives with hydroxyl
groups at positions 8 and 11 have also been prepared.
desired 6-[(alkylamino)alkyl]amino-anthrapyridazones
(2a–f). These derivatives were transformed into, respec-
tively, the symmetrically and unsymmetrically sub-
stituted 2-[(alkylamino)alkyl]-6-[(alkylamino)alkyl]amino-
anthrapyridazones (4a–c) by heating with the appropriate
alkylaminoalkyl chloride in the presence of Na2CO3.
Another possibility to obtain these derivatives is the
preparation of the 2-[(alkylamino)alkyl]-6-chloro-
anthrapyridazone 3 by refluxing of the acid chloride of 5
with an appropriate alkylaminoalkyl hydrazine followed
by heating of 3 with the respective amine. 2-[(Alkylamino)
alkyl]-6-amino-anthrapyridazone (4d) was obtained in
the reaction of 3 with aminoacetaldehyde dimethylace-
tal and hydrolysis of the product with 10% HCl in tet-
rahydrofurane (Scheme 1).
Chemistry
The synthetic pathways leading to anthrapyridazone
derivatives are shown in Schemes 1–4. The starting
materials 6-chloro- and 8-aminoanthrapyridazone (1
and 15), used in the synthesis of 2a–f, 4a–d, 16a–e and
17a–e, were obtained by previously reported meth-
ods11,12 by treatment of the respective acid chlorides of
5 or 15 with hydrazine. We prepared the 2-methyl-8-
aminoanthrapyridazone (20) in an analogous way in
slightly modified conditions. The reaction of 1 with the
appropriate amines occurred very readily and led to the
The synthetic route to the 8,11-dihydroxyanthrapyr-
idazone derivatives is illustrated in Scheme 2.
Scheme 1. Synthetic route for compounds 2–4. Reagents: (a) NHR2R3, pyridine; (b) R1Cl, Na2CO3, DMA; (c) 1. benzene, PCl5, 2. R1NHNH2; (d)
1. H2NCH2CH(OCH3)2, 2. 10% HCl, THF.

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
563
Scheme 2. Synthetic route for compounds 11–14. Reagents: (a) 1. CuCN, DMA, 2. 3N HNO3; (b) 15% NaOH, MeOH, DMA; (c) 1. benzene, PCl5,
2. R1NHNH2; (d) NHR2R3, DMA; (e) TFA; (f) 1. benzene, PCl5, 2. NH2NH2.
Scheme 3. Synthetic route for compounds 16 and 17. Reagents: (a) R1Cl, Na2CO3, DMA; (b) R1Cl¼R2Cl, NaH, DMA; (c) R2Cl, NaH, DMA.
Scheme 4. Synthetic route for compound 21. Reagents: (a) EtOH saturated with HCl; (b) CH3NHNH2, pyridine; (c) R1Cl, NaH, DMA.
Treatment of 6 with cupric cyanide in DMA resulted in
the desired monosubstituted derivative 7. The basic
hydrolysis of nitrile group in a 15% solution of NaOH
led to 5,8-di(benzyloxy)-4-chloro-9,10-dioxo-9,10-dihy-
dro-1-anthracenecarboxylic acid (8). The reaction of the
acid chloride of 8 with hydrazine or 2-(N,N-dimethyl-
amino)ethylhydrazine followed by heating 9 or 12 with
2-(dimethylamino)ethylamine in DMA gave compounds
substituted with one or two basic side chains, respec-
tively (13, 10). The removal of the protecting benzyl

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
564
groups in the presence of trifluoroacetic acid yielded the
required 8,11-dihydroxyanthrapyridazones 11 and 14.
For the biological and physicochemical evaluations, all
the obtained compounds were converted into their
hydrochloride or dihydrochloride salts by routine
methods.
Compounds 16a–e were prepared starting from 15 using
similar conditions as described above for the synthesis
of 4a–c (Scheme 3).
Results and Discussion
Cytotoxic activity
Derivatives 17a–c were obtained in the reaction of 15
with sodium hydride in DMA followed by the treatment
of obtained sodium salts with an appropriate (alkyla-
mino)alkyl chloride or, if the side chains at positions 2
and 8 should be different, under similar conditions,
using 16 as starting material (17d–e, Scheme 3).
The anthrapyridazone derivatives were evaluated for
their cytotoxic potency in vitro using the sensitive leu-
kemia cell lines: murine L1210 and human K562 as well
as human leukemia multidrug resistant subline K562/
DX.
For structure–activity relationship, the anthrapyr-
idazone derivative 21 with one side chain at position 8
has been synthesized. Compound 19 was obtained in the
usual way by heating 18 in hydrogen chloride saturated
ethanol. After that 19 refluxed with methylhydrazine led
to 20, which was transformed into 21 in an analogous
way as described for 17d–e (Scheme 4).
Mitoxantrone (MIT) and Doxorubicin (DX) were used
as reference compounds. The cytotoxicity data are pre-
sented in Table 2 as IC50 values.
Most of the studied anthrapyridazones demonstrated
high cytotoxic activity against L1210 leukemia. The 2,6-
disubstituted derivatives (4a–e) are more active than
their 6-monosubstituted analogues (2a–e). Generally
derivatives with side chains at positions 2 and 8 are less
The structures, melting points, yields and formulas of
compounds 2a–f, 4a–d, 11, 14, 16a–e, 17a–e and 21 are
presented in Table 1.
Table 1. Substituents, melting points, yields, and formulas of anthrapyridazone derivatives
Compd Fm
X
R1
R2
R3
Yielda
Mp (ꢀC)
Molecular formulab
2a
2b
2c
2d
2e
2f
4a
4b
4c
4d
11
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
B
B
B
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
H
H
H
H
H
H
H
H
H
H
CH2CH2N(CH3)2
CH2CH2NHCH2CH2OH
CH2CH2–c-N(CH2)5
CH2CH2–c-N(CH2)4NH
CH2CH2–c-N(CH2)4O
CH2CH2N(CH3)2
CH2CH2N(CH3)2
CH2CH2–c-N(CH2)5
CH2CH2–c-N(CH2)5
H
85
50
65
75
75
70
54c
52d
56c
27
69c
42
54
60
50
45
48
35
43
38
25
20
55
274–276
243–245
254–256
260–261
281–283
188–190
179–181
134–136
181–182
225–227
196–198
>300
230–232
233–235
231–233
238–240
243–245
134–135
>290
C19H18N4O2
C19H18N4O3
C22H22N4O2
C21H21N5O2
C21H20N4O3
C20H20N4O2
C23H27N5O2
C26H31N5O2
C29H35N5O2
C19H18N4O2
C23H27N5O4
C19H18N4O4
C19H18N4O2
C21H22N4O2
C20H20N4O2
C22H22N4O2
C21H20N4O2
C23H27N5O2
C27H37Cl2N5O2
C25H33Cl2N5O2
C26H31N5O2
C25H31Cl2N5O2
C20H20N4O2
CH2CH2N(CH3)2
CH2CH2N(CH3)2
CH2CH2–c-N(CH2)5
CH2CH2N(CH3)2
CH2CH2N(CH3)2
H
CH2CH2N(CH3)2
CH2CH2N(C2H5)2
CH2CH2CH2N(CH3)2
CH2CH2–c-N(CH2)5
CH2CH2–c-N(CH2)4
CH2CH2N(CH3)2
CH2CH2N(C2H5)2
CH2CH2CH2N(CH3)2
CH2CH2N(CH3)2
CH2CH2–c-N(CH2)4
CH3
H
OH
OH
CH2CH2N(CH3)2
CH2CH2N(CH3)2
14
16a
16b
16c
16d
16e
17a
17b
17c
17d
17e
21
H
CH2CH2N(CH3)2
CH2CH2N(C2H5)2
CH2CH2CH2N(CH3)2
CH2CH2–c-N(CH2)5
CH2CH2N(CH3)2
CH2CH2N(CH3)2
>290
136–138
>290
280–281
aThe yields are not optimized.
bThe structures of compounds presented in Table 1 were confirmed by their spectral data (1H NMR, IR, UV–vis, MS-FAB) and by elemental
analyses (C, H, N).
cMethod A.
dMethod B.

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
565
Table 2. In vitro cytotoxic activity against L1210, K562 and K562/DX tumor cell lines, DNA binding and membrane affinity of target compounds
2a–f, 4a–d, 11, 14, 16a–e, 17a–e, 21, and of reference compounds Mitoxantrone (MIT) and Doxorubicin (DX)
log k0IAM
c
d
Compd
Cell linea/IC50 (nM)ꢁSEM
RIb
Kapp ꢂ10À7 CT-DNA
L1210
K562
K562/DX
2a
2b
2c
2d
2e
2f
4a
4b
4c
4d
11
43.4ꢁ1.4
91.6ꢁ3.6
446.0ꢁ60.2
338.6ꢁ29.3
1536.3ꢁ243.3
ND
1217.6ꢁ208.3
7604.9ꢁ1394.0
2009.0ꢁ306.5
ND
2.73
22.5
1.31
0.43 (2.93)
0.32 (3.89)
0.55 (2.30)
1.649
1.793
2.226
1.415
2.008
1.308
1.754
2.077
2.622
1.688
1.982
4.575
1.733
532.3ꢁ31.1
165.9ꢁ26.6
1781.6ꢁ123.5
191.1ꢁ11.0
2.1ꢁ0.1
2101.6ꢁ825.3
792.5ꢁ125.2
14.0ꢁ1.4
4944.9ꢁ852.8
2389.0ꢁ317.5
31.6ꢁ8.2
2.35
3.01
2.26
1.00
2.11
1.79
3.86
2.28 (0.55)
0.88 (1.49)
1.12 (1.15)
0.46 (2.76)
3.32 (0.39)
31.9ꢁ1.7
138.1ꢁ18.6
286.9ꢁ11.2
185.2ꢁ7.4
3.53ꢁ0.51
ND
138.5ꢁ9.8
75.1ꢁ1.2
605.9ꢁ33.8
331.6ꢁ67.5
13.61ꢁ2.51
ND
66.0ꢁ2.8
0.43ꢁ0.06
19.0ꢁ4.7
14
16a
16b
16c
16d
16e
17a
17b
17c
17d
17e
21
30.3ꢁ3.5
133.8ꢁ16.2
330.0ꢁ36.8
963.7ꢁ220.0
573.4ꢁ26.5
148.4ꢁ6.8
591.3ꢁ48.4
672.4ꢁ100.7
996.7ꢁ114.6
335.6ꢁ45.1
267.2ꢁ49.8
701.3ꢁ47.5
21.0ꢁ4.0
263.0ꢁ34.7
582.8ꢁ140.9
1750.5ꢁ65.8
1125.5ꢁ55.3
379.2ꢁ86.0
716.4ꢁ39.1
631.8ꢁ75.8
2655.0ꢁ52.3
687.2ꢁ115.3
716.6ꢁ164.1
1417.6ꢁ226.6
554.0ꢁ49.5
7031.5ꢁ542.0
1.96
1.77
1.82
1.96
2.56
1.21
0.94
2.66
2.04
2.68
2.02
124.7ꢁ15.0
300.8ꢁ35.5
239.7ꢁ14.3
38.5ꢁ2.1
1.521
137.8ꢁ9.2
122.5ꢁ31.5
254.4ꢁ47.3
44.1ꢁ2.7
4.58 (0.28)
2.235
1.745
40.6ꢁ7.1
2.230
242.1ꢁ17.9
1.4ꢁ0.1
MIT
DX
26.4
167.0
19 (0.053)e
1.841
1.434
19.4ꢁ1.3
42.1ꢁ3.3
ND, not determined.
aL1210, murine lymphocytic leukemia; K562, human myelogenous leukemia and Doxorubicin resistant (MDR type) subline K562/DX.
bRI, resistance index: the ratio of IC50 value for resistant cell line to IC50 value for sensitive cell line.
cKapp=1.26/C50ꢂ107, in which 1.26 is the concentration (mM) of ethidium–DNA complex, C50 is drug concentration (mM) to effect 50% drop in
fluorescence of bound ethidium and 107 is the value of Kapp assumed for ethidium in the complex. The C50 values are in parentheses.
dLogarithm of HPLC capacity factor determined in an immobilized artificial membrane column with acetonitrile buffer, pH 7.0.
eAccording to ref 13.
active than their 2,6-disubstituted analogues. However,
anthrapyridazone derivatives with two side arms at
positions 2 and 8 (17a–e) exhibit comparable or lower
activity than their analogues with one side chain at
position 2 (16a–e). A conclusion could be drawn that
not only the presence of the second side chain but also
its respective position in the chromophore of anthra-
pyridazones is important for cytotoxic activity. A simi-
lar structure–activity relationship was found in the
anthrapyrazole and anthracenedione systems.8,9 The
alkylaminoalkylamino side chains at positions 2 and 6
are indispensable for the high cytotoxicity of anthra-
pyridazone derivatives. As expected,6 for the activity
was a optimal substitution of the ring with a 2-[(dimethyl-
amino)ethyl]amino side arm regardless of its position.
Derivatives with a piperidyl-, morpholinyl-, or piper-
azyl-terminal residue in the aminoethyl side arm are
about tenfold less active than derivatives with a 2-
[(dimethylamino)ethyl]amino side chain, independently,
if they are situated at position 2 or 6. The cytotoxic
activity of monosubstituted 2-(alkylamino)alkyl-
anthrapyridazones is similar to that of 6-[(alkylami-
no)alkyl]amino-8-aminoderivatives.
The introduction of hydroxyl groups at positions 8 and
11 of compounds 2a and 4a increased the cytotoxic
activity of these derivatives, similarly as was found for
other anthracenedione and acridine derivatives. It
should be noted that the compound 11 is more active
against sensitive cell lines than Mitoxantrone.
The anthrapyridazones also exhibit significant cytotoxic
activity against human leukemia sensitive cell line K562
and the multidrug resistant subline K562/DX. A good
resistance index (RI) in the range 1–3 depending on the
compound’s structure was evidenced for the evaluated
compounds regardless of their cytotoxic potency. A
much higher RI was exhibited by the derivative 2b with
the hydroxyethylaminoethylamino side arm at position
6.
We assumed that the possible hydrogen bond between
the carbonyl at position 7 and the hydrogen of the
amino group at position 6 of 2a could mimic an addi-
tional fifth ring, which may be important for MDR
activity as has been shown for the pyrimidoacridines.13
To examine this effect, we synthesized compound 2f (the
N-methyl derivative of 2a) in which the formation of the
hydrogen bond is not possible. However, 2f is 2-fold less
active than 2a and has a similar resistance index.
Compound 4a with two dimethylaminoethylamino side
chains at positions 2 and 6 is the most active, while the
6-monosubstituted analogue, as well as the 2-substituted,
6-amino or 6-chloro ones exhibit tenfold less activity than
4a.
The best effectiveness in overcoming the multidrug
resistance was exhibited by derivatives bearing a

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
566
2-[(dimethylamino)ethyl]amino side chain, which corre-
lates well with data obtained for the benzoperimidines,
anthrapyridones, aza-anthrapyrazoles and pyrimido-
acridines.
affinity rather then as commonly used lipophilicity
characteristics.
Most studied anthrapyridazone derivatives maintain
cytotoxic activity against human leukemia multidrug
resistant subline K562/DX with P-gp efflux pump
(MDR), and exhibit low RI values regardless of their
cytotoxic potency. There is a qualitative correlation
tendency between resistance indexes of examined com-
pounds and their log k0IAM values (Table 2). One should
not expect a quantitative correlation, as the ability to
overcome multidrug resistance is a rather complex phe-
nomenon, influenced by the changes in the structure of
the examined compounds. Besides the velocity of drug
uptake, other factors like its affinity to the drug export-
ing protein (P-gp) and intracellular distribution char-
acteristics also play a defined role. In the latter case, the
nuclear DNA binding characteristics and the ability to
be entrapped in other organelles, like lysosomes, influ-
ence the final response of the cell to the drug action.18
However, the data obtained for anthrapyridazones
point to the uptake characteristics, expressed by
logk0IAM values, as essential ones. The logk0IAM values
for anthrapyridazones with low RI are located in the
range optimal for good uptake. Values near or below 1
would slow down the drug diffusion into cells, while
values near 3 or more would cause the retention of the
compounds in cytoplasmatic membrane. The obtained
data are similar to the earlier ones found for the benzo-
perimidine derivatives.19 The exception is compound 2b
with a very high RI. However, due to the presence of the
hydroxyl group in the side chain, the compound is
expected to acquire a specific conformation that might
influence its properties. We have earlier evidenced that
anthraquinone derivatives with this type of side chain
form a very stable hydrogen bond network essentially
influencing the conformation of the molecule.20 Other
authors also observed that aza-anthrapyrazole5 deriva-
tives as well as pyrimidoacridines13 with the same type
of side chains exhibit high resistant indexes.
Among the synthesized anthrapyridazones, the most
active against the MDR cell line is the derivative with a
dimethylaminoethylamino arm at positions 2 and 6 (4a),
as well as its 8,11-dihydroxy analogue 11. Compounds
4a–c with two basic side chains at positions 2 and 6 were
much more active than compounds bearing only one
chain at position 6. However, a second side chain at
position 8 is rather unfavorable.
The above data demonstrate that both the presence of a
condensed heterocyclic ring and the structure, number
and position in the ring system of the side chains are of
importance for the overcoming of multidrug resistance
by anthrapyridazone derivatives.
DNA binding
Competitive displacement (C50) in fluorometric assays
with DNA-bound ethidium was applied to determine
‘apparent’ equilibrium constants (Kapp) for drug bind-
ing, as the C50 value is approximately inversely propor-
tional to the binding constant.14 In the present work,
fluorescence displacement assays were performed at pH
7 to enable comparison with biological conditions. The
Kapp values for the representative anthrapyridazones
(Table 2) indicate that: (a) the disubstituted derivatives
4a, 4c, 11 and 17b are stronger DNA-binding ligands
than ethidium, with the exception of 4b, (b) the presence
of 8,11-hydroxy groups in the chromophore moiety
markedly increased the Kapp value (11), (c) the side
chain at position 8 (17a) seems to be more important for
binding than that in position 6 (4a), which does not
correlate with the cytotoxicity data.
Membrane affinity
We have reported earlier that the ability of the anthra-
cenedione analogues to overcome multidrug resistance
is dependent on the velocity of their uptake, exceeding
the ABC proteins mediated efflux. This is ensured by the
presence in the molecule of a fused heterocyclic ring;
however, various substituents quantitatively influence
this effect.6,15
Antileukemic activity
On the basis of results of the cytotoxicity studies, com-
pounds 4a and 11 were selected for preliminary evalua-
tion of their antitumor activity. Compounds 4a and 11
were examined with intraperitoneal (ip) administration
on P388 murine leukemia. The data, in comparison with
reference Mitoxantrone, are shown in Table 3.
The essential factor governing the diffusion of a drug to
the cells is its cytoplasmatic membrane affinity. The
determinations have been performed on an HPLC-
reverse-phase chromatographic IAM (column immobi-
lized artificial membrane) constituting chromatographic
surfaces prepared by covalently immobilized cell mem-
brane phospholipids to solid surfaces at monolayer
densities. IAM surfaces mimic the fluid cell mem-
brane.16,17 The retention time for a given compound,
expressed as log k0IAM values, comprises not only lipo-
philicity characteristics but also other interactions with
membrane like hydrogen bond formation, electrostatic
interactions, and so on. Therefore we prefer to interpret
the log k0IAM values as characterizing the membrane
Both compounds 4a and 11 exhibit significant antileu-
kemic efficacy (T/C% >200). The in vivo activity of 4a
with maximum T/C%=240 (at the dose 0.8 mg/kg) is
comparable with that of Mitoxantrone (T/C%=211, at
the same dose). It should be noticed that the antileu-
kemic activity of 11, which is the 8,11-dihydroxy analo-
gue of 4a, is lower than that for 4a (T/C%=200 at 0.8
mg/kg and T/C%=222 at a 2-fold higher dose). These
data are rather surprising because as noted above the
introduction of hydroxyl groups into the chromophore
of different antitumor compounds of anthracenedione
and acridine groups increases their antitumor activity.
On the other hand, 11 is less toxic than its dehydroxy

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
567
Table 3. Antitumor activity of selected compounds 4a and 11 against
P388 murine leukemia in comparison with Mitoxantrone
Typical procedure for the preparation of hydrochloride
salts of 2a–f, 4a–d, 11, 14, 16a–e, 17a–e, 21. To a stir-
red solution of the compound (as free base) in CHCl3 or
CHCl3/MeOH a slightly molar excess of hydrogen chlo-
ride in absolute ethanol was added dropwise at 5 ꢀC. The
hydrochloride salt was precipitated by adding dry diethyl
ether and collected by filtration. The solid was purified
by column chromatography (Sephadex LH-20) eluting
with MeOH and crystallized from MeOH/Et2O.
Compd
Total dose (mg/kg)a
% T/Cb
4a
0.1
0.2
0.4
0.8
1.6
2.4
108
165
130
240
216
tox
11
0.1
0.2
0.4
0.8
1.6
2.4
130
162
General procedure for the synthesis of compounds 2a–f
170
200 (1/6 LTS)c
219
6-[[2-(Dimethylamino)ethyl]amino]-2,7-dihydro-3H-diben-
zo[de,h]cinnoline-3,7-dione (2a). A sample of 400 mg (1.4
mmol) of 6-chloro-2,7-dihydro-3H-dibenzo[de,h]cinno-
line-3,7-dione (1)12 with 2 mL of 2-(dimethylami-
no)ethylamine in pyridine was stirred at 100 ꢀC under a
nitrogen atmosphere for 7 h. The progress of the reac-
tion was followed by TLC in the solvent system CHCl3/
MeOH (10:1). The reaction mixture was diluted with
CHCl3, washed several times with dilute HCl to remove
an excess of amines and then washed with water. The
organic layer was dried over Na2SO4. The solvent was
evaporated and the residue was purified by column flash
chromatography (silica gel) eluting with CHCl3/MeOH
mixture (10:1, 5:1). The chromatographic fractions were
concentrated to afford 2a (300 mg, 85%) as a yellow
powder. Mp 274–276 ꢀC. 1H NMR (DMSO-d6) d 2.52 (s,
6H, 2ꢂCH3), 2.64 (t, 2H, J=5.9 Hz), 3.6 (q, 2H, J=5.9
Hz), 7.2 (d, 1H, J=9.3 Hz), 7.58–7.76 (m, 2H), 8.4–8.58
(m, 3H), 10.8 (t, 1H, NH), 13.2 (br s, 1H, NH, ex).
Found: C, 67.87; H, 5.56; N, 16.42; calcd for
C19H18N4O2: C, 68.25; H, 5.43; N, 16.76.
222
MIT
0.1
0.2
0.4
0.8
163
165
179
211
aSingle dose ip administration.
bThe average value of two separate experiments, %T/C, the ratio of
medium survival time of treated to control mice, expressed as a per-
centage. %T/C ꢃ125 are considered to be a significant activity. Long-
term survivors were not included in the calculation.
cLTS, number of long-term survivors (cures)/total number of mice.
analogue 4a.
In conclusion
Anthrapyridazones constitute a novel group of anthra-
cenedione analogues able to overcome the multidrug
resistance of tumor cells. The presence of respective
heterocyclic ring, here the pyridazone one, fused with
the anthracenedione chromophore determines cytotoxic
activity toward the multidrug resistant cell lines. The
positive effect of this structural factor in overcoming the
multidrug resistance of tumor cells is quantitatively
influenced, regarding the cytotoxic potency, by the type
and location of substituents.
6-Chloro-2-[2-(dimethylamino)ethyl]-2,7-dihydro-3H-di-
benzo[de,h]cinnoline-3,7-dione (3). A sample of 500 mg
(1.75 mmol) of 4-chloro-9,10-dioxo-9,10-dihydro-1-
anthracenecarboxylic acid (5)11 and 500 mg of PCl5 was
suspended in 6 mL benzene at room temperature to
obtain an intimate mixture. Additionally, 16 mL of
benzene was added and then dropwise 500 mg (4.85
mmol) of 2-(N,N-dimethylamino)ethylhydrazine during
10 min. After stirring for 1 h, the resulting suspension
was diluted with CHCl3. The mixture was extracted
with a 5% water solution of Na2CO3 (2ꢂ50 mL) and
with water (2ꢂ50 mL). The organic layer was dried over
Na2SO4 and the solvent was evaporated. The residue
was chromatographed on silica gel column (eluent
CHCl3/MeOH from 20:1 to 5:1) to yield 3 (312 mg,
51%) as a yellow solid. Mp 163–164 ꢀC. 1H NMR
(CDCl3) d 2.36 (s, 6H, 2ꢂCH3), 2.9 (t, 2H, J=6.6 Hz),
4.5 (t, 2H, J=6.6 Hz), 7.58–7.78 (m, 2H), 7.9 (d, 1H,
J=8.5 Hz), 8.34 (dd, 1H, J=1.4 Hz, J=7.8 Hz), 8.46
(dd, 1H, J=1.3 Hz, J=9.0 Hz).
Experimental
Chemistry
Melting points, determined with a Boeticus PHMK05
apparatus, are uncorrected. Analyses are withinꢁ0.4%
of the theoretical values and were carried out on a Carlo
Erba CHNS-O-EA1108 instrument for C, H, N. A
Beckman 3600 spectrophotometer was used for UV
spectral determination. IR spectra were recorded on a
UR 10 Zeiss spectrometer in KBr pellets; 1H NMR
spectra were taken on a Varian 300-MHz or 500-MHz
spectrometer using tetramethylsilane as an internal
standard. The following NMR abbreviations are used:
ex (exchangeable with D2O), d ex (exchangeable with
D2O, but with difficulty). Mass spectra were recorded
on a Quadrupolic Mass Spectrophotometer Trio-3
(FAB technique). Thin layer chromatography (TLC)
was carried out on Kieselgel 60 plates (Merck), column
chromatography on Kieselgel Merck (70–230 mesh) and
on Sephadex LH-20 (Pharmacia).
General procedure for the synthesis of compounds 4a–c
(Method A)
2-[2-(Dimethylamino)ethyl]-6-[[2-(dimethylamino)ethyl]-
amino]-2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-dione
(4a). A mixture of 50 mg (0.14 mmol) of 3 and 1 mL of
N,N-dimethylethylenediamine in 2 mL of DMA was
stirred for 30 min at 60 ꢀC. The reaction mixture was

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
568
partitioned between CHCl3 with Et2O and water. The
organic layer was worked up to give a residue which was
flash chromatographed on silica gel column eluted with
CHCl3/MeOH (from 20:1 to 5:1) to afford 31.5 mg
mmol) of 1,4-di(benzyloxy)-5,8-dichloro-9,10-dihydro-
9,10-anthracenedione (6)9 and 108 mg (1.22 mmol)
CuCN in 5 mL of DMA was stirred and heated at
125 ꢀC for 4 h. The course of the reaction was followed
by TLC in the solvent system toluene/acetone (30:1).
The reaction mixture was allowed to cool to room tem-
perature and poured into cold water. The brown solid
was collected by filtration and washed well with water.
The material was suspended in 3 N HNO3 and stirred at
70ꢀ C for 4 h. The resulting suspension was filtered, and
the precipitate was washed several times with cold
water. The deep-orange product was air dried overnight
and then flash chromatographed eluting with benzene
and benzene/acetone (50:1). The chromatographic frac-
tions containing the product were concentrated giving
300 mg (61%) of product (7) as an orange powder. Mp
(54%) of pure 4a. Mp 179–181 ꢀC. H NMR (as dihy-
1
drochloride, DMSO-d6) d 2.84 (s, 6H, 2ꢂCH3), 2.89 (s,
6H, 2ꢂCH3), 3.39 (t, 2H, J=6.3 Hz), 3.65 (t, 2H, J=5.7
Hz), 4.0 (q, 2H, J=6.3), 4.68 (t, 2H, J=5.6), 7.68 (d,
1H, J=9.8 Hz), 7.76 (t, 1H, J=7.6 Hz), 7.87 (t, 1H,
J=7.6 Hz), 8.31–8.53 (m, 2H), 10.06 (br s, 1H, NH+),
10.62 (br s, 1H, NH+), 10.78 (t, 1H, NH). Found: C,
68.10; H, 6.80; N, 17.01; calcd for C23H27N5O2: C,
68.13; H, 6.71; N, 17.27.
Method B
199–201 ꢀC. H NMR (CDCl3) d 5.34 (s, 2H), 5.35 (s,
1
2-[2-(Dimethylamino)ethyl]-6-[[2-(piperidin-1-yl)ethyl]-
amino]-2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-dione
(4b). A sample of 185 mg (0.5 mmol) of 2c, 50 mg (0.5
mmol) of Na2CO3 and 0.15 mL (1.5 mmol) of 2-(dime-
thylamino)ethyl chloride in 5 mL of DMA was stirred
for 1 h at 80 ꢀC. The course of reaction was monitored
by TLC in the solvent system CHCl3/MeOH (10:1). The
reaction mixture was diluted with CHCl3, washed twice
with water and the organic layer dried over Na2SO4.
After evaporation under reduced pressure, the residue
was chromatographed on silica gel column eluting with
CHCl3/MeOH (10:1) to afford 4b. Yield 115 mg (52%).
2H), 7.36–7.60 (m, 12H), 7.95 (d, 1H, J=8.2 Hz), 8.02
(d, 1H, J=8.2 Hz).
5,8-Di(benzyloxy)-4-chloro-9,10-dioxo-9,10-dihydro-1-an-
thracenecarboxylic acid (8). To a stirred solution of 300
mg (0.62 mmol) of 7 in 8 mL MeOH and 2 mL of DMA
was added dropwise 1 mL of 20% NaOH. The mixture
was refluxed for 4 h and then allowed to cool to room
temperature. To the reaction mixture dilute HCl was
added giving an orange suspension. The solid was col-
lected by filtration, washed with cold water, dried and
then purified by column chromatography using as elu-
ent CHCl3/MeOH (20:1 to 5:1) to afford 258 mg
(82.7%) of pure acid 8 as a yellow powder. Mp 212–
ꢀ
1
Mp 134–136 C. H NMR (CDCl3) d 1.6–1.7 (m, 6H),
2.4 (s, 6H, 2ꢂCH3), 2.56 (t, 4H, J=4.8 Hz), 2.8 (t, 2H,
J=6.6 Hz), 2.87 (t, 2H, J=6.6 Hz), 3.64 (q, 2H, J=4.8
Hz), 4.48 (t, 2H, J=6.6 Hz), 7.24 (d, 1H, J=10.3 Hz),
7.58–7.76 (m, 2H), 8.38–8.48 (m, 2H), 8.56 (d, 1H, J=7.7
Hz), 11.0 (t, 1H, NH). Found: C, 70.10; H, 7.28; N,
15.54; calcd for C26H31N5O2: C, 70.09; H, 7.01; N, 15.72.
214 ꢀC. H NMR (DMSO-d6) d 5.34 (s, 2H), 5.36 (s,
1
2H), 7.4–7.6 (m, 12H), 8.0 (d, 1H, J=8.5 Hz), 8.4 (d,
1H, J=8.5 Hz), 13.3 (s, 1H, ex).
8,11-Di(benzyloxy)-6-chloro-2-[2-(dimethylamino)ethyl]-
2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-dione (9). A
6-Amino-2-[2-(dimethylamino)ethyl]-2,7-dihydro-3H-di-
benzo[de,h]cinnoline-3,7-dione (4d). Step (i). A suspen-
sion of 70 mg (0.2 mmol) of 3 and aminoacetaldehyde
dimethylacetal (0.5 mL) in 2 mL of DMA was heated at
110 ꢀC with stirring, until all starting material was con-
sumed. After cooling, the reaction mixture was diluted
with CHCl3 and extracted with water. The organic layer
was dried over Na2SO4. Removal of the solvent in
vacuo gave a residue, which was chromatographed
eluting with CHCl3. Step (ii). The aminoacetal, obtained
by the above procedure, was treated with 10% HCl. The
reaction mixture was stirred at 30–40 ꢀC for 30 min and
then the suspension was poured into ice water and neu-
tralized with NaHCO3. The precipitate was collected by
filtration, washed with THF, then with water and dried
in vacuo. The solid was purified by crystallization from
DMA to give 4d (18 mg, 27%). Mp 225–227 ꢀC. 1H
NMR (CDCl3) d 2.4 (s, 6H, 2ꢂCH3), 2.9 (t, 2H, J=6.6
Hz), 4.5 (t, 2H, J=6.6 Hz), 7.0 (d, 1H, J=9.0 Hz),
7.58–7.78 (m, 2H), 8.24–8.32 (m, 2H), 8.5 (d, 1H, J=8.0
Hz), 9.8 (br s, 2H, NH2). MS m/z (relative intensity, %):
334 ([M]+, 100); 335 ([M+1]+, 41). Found: C, 67.95;
H, 5.65; N, 16.48; calcd for C19H18N4O2: C, 68.25; H,
5.43; N, 16.76.
sample of 200 mg (0.4 mmol) of 5,8-di(benzyloxy)-4-
chloro-9,10-dioxo-9,10-dihydro-1-anthracenecarboxylic
acid and 200 mg of PCl5 was stirred in 5 mL of benzene
at room temperature to obtain an intimate mixture.
Additional 5 mL of benzene was added and then drop-
wise 500 mg (4.85 mmol) of 2-(N,N-dimethylami-
no)ethylhydrazine during
5
min. The resulting
suspension was kept in these conditions for additional 1 h.
After this the mixture was partitioned between CHCl3 and
aqueous 5% Na2CO3. The organic layer was worked up
to give a residue, which was flash chromatographed (elu-
ent CHCl3/MeOH 20:1) to yield pure 9 (114 mg, 50%).
Mp 157–160 ꢀC. 1H NMR (CDCl3) d 2.19 (s, 6H,
2ꢂCH3), 2.61 (t, 2H, J=6.6 Hz), 4.31 (t, 2H, J=6.6 Hz),
5.2 (s, 2H), 5.32 (s, 2H), 7.1 (d, 1H, J=9.2 Hz), 7.29–7.61
(m, 11H), 7.87 (d, 1H, J=8.6 Hz), 8.52 (d, 1H, J=8.6 Hz).
8,11-Di(benzyloxy)-2-[2-(dimethylamino)ethyl]-6-[[2-(di-
methylamino)ethyl]amino]-2,7-dihydro-3H-dibenzo[de,h]-
cinnoline-3,7-dione (10). A sample of 9 (100 mg, 0.17
mmol) with 2 mL of 2-(dimethylamino)ethylamine in 2
mL of DMA was stirred at 60 ꢀC for 1 h. The progress
of the reaction was followed by TLC in the solvent sys-
tem CHCl3/MeOH (10:1). The reaction mixture was
diluted with CHCl3 and Et2O and then washed several
times with water. The organic layer was dried over
5,8-Di(benzyloxy)-4-chloro-9,10-dioxo-9,10-dihydro-1-an-
thracenecarbonitrile (7). A sample of 500 mg (1.02

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
569
Na2SO4, evaporated in vacuo and the residue was pur-
ified by column flash chromatography (silica gel) eluting
with CHCl3/MeOH (10:1, then 5:1). The chromato-
graphic fractions were concentrated to afford 10 (300
2H), 7.2 (d, 1H, J=9.2 Hz), 7.27–7.70 (m, 12H), 8.42 (d,
1H, J=9.2 Hz), 10.42 (s, 1H, NH ex), 11.14 (s, 1H, NH).
6-[[(Dimethylamino)ethyl]amino]-8,11-dihydroxy-2,7-di-
hydro-3H-dibenzo[de,h]cinnoline-3,7-dione (14). A solu-
tion of 50 mg (0.1 mmol) of 13 in 1 mL of trifluoroacetic
acid was left at room temperature for 12 h. The excess of
TFA was removed under reduced pressure by coevapora-
tion with benzene to afford a salt of 6-[2-(dimethylami-
no)ethyl]amino-8,11-dihydroxy-2,7-dihydro-3H-dibenzo-
[de,h]cinnoline-3,7-dione. The residue was dissolved in
aqueous 5% Na2CO3 and extracted several times with
chloroform. The organic layer was washed with water,
then collected and dried over Na2SO4. The solvent was
evaporated and the crude product was purified using
silica gel column chromatography (eluent CHCl3/
MeOH 10:1, 5:1) to give 14 as a dark red solid (14.2 mg,
42%). Mp >300 ꢀC. 1H NMR (as hydrochloride,
DMSO-d6) d 2.85 (s, 6H, 2ꢂCH3), 3.68 (t, 2H, J=7.1
Hz), 4.06 (q, 2H, J=7.1 Hz), 7.17–7.32 (m, 2H), 7.61 (d,
1H, J=9.5 Hz), 8.30 (d, 1H, J=9.5 Hz), 9.85 (br s, 1H,
NH), 10.40 (br s, 1H, NH ex), 11.49 (s, 1H, NH+),
13.25 (s, 1H, ex), 13.59 (s, 1H, ex). MS m/z (relative
intensity, %): 365 ([MÀ1]+, 100); 366 ([M]+, 50).
Found: C, 62.03; H, 5.10; N, 15.04; calcd for
C19H18N4O4: C, 62.29; H, 4.95; N, 15.29.
mg, 84.6%) as a yellow powder. Mp 186—188 ꢀC. H
1
NMR (CDCl3) d 2.21 (s, 6H, 2ꢂCH3), 2.41 (s, 6H,
2ꢂCH3), 2.66 (t, 2H, J=6.7 Hz), 2.74 (t, 2H, J=6.7
Hz), 3.56 (q, 2H, J=5.1 Hz), 4.36 (t, 2H, J=6.7 Hz),
5.23 (s, 2H), 5.28 (s, 2H), 7.1 (d, 1H, J=9.2 Hz), 7.2 (d,
1H, J=9.2 Hz), 7.27–7.64 (m, 10H), 7.68 (d, 1H, J=6.7
Hz), 8.44 (d, 1H, J=9.3 Hz).
2-[2-(Dimethylamino)ethyl]-6-[[2-(dimethylamino)ethyl]ami-
no]-8,11-dihydroxy-2,7-dihydro-3H-dibenzo[de,h]cinno-
line-3,7-dione (11). A sample of 10 (100 mg, 0.16 mmol)
was left in 1 mL of trifluoroacetic acid at room tem-
perature for 18 h. The TFA was removed under reduced
pressure by coevaporation with benzene. The residue
was suspended in CHCl3 and the solution was carefully
washed with NaHCO3 water solution and then with
water. The organic layer was dried over Na2SO4, eva-
porated under reduced pressure and the product was
solidified under Et2O to yield 11 (48.8 mg, 69%) as red
powder. Mp 196–198 ꢀC. IR (KBr, major peaks cmÀ1):
1
1209, 1468, 1570, 1631, 1656. H NMR (CDCl3) d 2.37
(s, 6H, 2ꢂCH3), 2.41 (s, 6H, 2ꢂCH3), 2.75 (t, 2H,
J=6.2 Hz), 2.86 (t, 2H, J=6.3 Hz), 3.54 (q, 2H, J=6.2
Hz), 4.48 (t, 2H, J=6.3 Hz), 7.1 (d, 1H, J=9.0 Hz),
7.19–7.41 (m, 2H), 8.43 (d, 1H, J=9.3 Hz), 10.57 (s, 1H,
NH), 11.48 (s, 1H, ex), 13.21 (s, 1H, ex). Found: C,
63.01; H, 6.50; N, 15.84; calcd for C23H27N5O4: C,
63.14; H, 6.22; N, 16.01.
General procedure for the synthesis of compounds 16a–e
2-[2-(Dimethylamino)ethyl]-8-amino-2,7-dihydro-3H-di-
benzo[de,h]cinnoline-3,7-dione (16a). A sample of 130
mg (0.5 mmol) of 15,12 50 mg (0.5 mmol) of Na2CO3
and 0.15 mL (1.5 mmol) of 2-(dimethylamino)ethyl
chloride in 5 mL of DMA was stirred for 1 h at 60–
70 ꢀC. The course of reaction was monitored by TLC in
the solvent system CHCl3/MeOH (10:1). The reaction
mixture was diluted with CHCl3, washed twice with
water and the organic layer dried over Na2SO4. After
evaporation under reduced pressure, the residue was
chromatographed on silica gel column eluting with
CHCl3/MeOH (10:1) to afford 16a as a red powder.
Yield 110 mg (54%). Mp. 230–232 ꢀC. IR (KBr, major
8,11-Di(benzyloxy)-6-chloro-2,7-dihydro-3H-dibenzo[de,h]-
cinnoline-3,7-dione (12). A sample of 200 mg (0.4 mmol)
of 5,8-di(benzyloxy)-4-chloro-9,10-dioxo-9,10-dihydro-
1-anthracene-carboxylic acid (8) and 200 mg of PCl5
was stirred in 2 mL of benzene at room temperature to
obtain an intimate mixture. Then, the mixture was
diluted with 5 mL of benzene and 0.5 mL of hydrazine
hydrate 98% was added dropwise during 5 min. The
resulting suspension was kept in these conditions for
additional 1 h. The isolation and purification proce-
dures of the reaction mixture to obtain the product 12
were the same as described for compound 9. Yield 176
mg (89%). Mp 277–280ꢀ C. 1H NMR (DMSO-d6) d 5.24
(s, 2H), 5.27 (s, 2H), 7.3–7.6 (m, 12H), 8.0 (d, 1H, J=8.5
Hz), 8.4 (d, 1H, J=8.5 Hz), 13.5 (s, 1H, NH, ex).
1
peaks cmÀ1): 1395, 1590, 1630, and 1650. H NMR (as
hydrochloride, DMSO-d6) d 2.8 (s, 6H, 2ꢂCH3); 3.62 (t,
2H, J=5.5 Hz); 4.64 (t, 2H, J=5.5 Hz); 6.99 (d, 1H,
J=7.4 Hz); 7.46 (t, 1H, J=7.7 Hz); 7.59 (d, 1H, J=7.1
Hz); 8.11 (t, 1H, J=7.7 Hz); 8.48 (t, 1H, J=6.9 Hz);
8.65 (d, 1H, J=7.4 Hz); 10.22 (br s, 1H, NH ex). MS m/
z (relative intensity, %): 333 ([MÀ1]+, 100); 334 ([M]+,
70). Found: C, 67.89; H, 5.70; N, 16.54; calcd for
C19H18N4O2: C, 68.25; H, 5.43; N, 16.76.
8,11-Di(benzyloxy)-6-[[2-(dimethylamino)ethyl]amino]-2,7-di-
hydro-3H-dibenzo[de,h]cinnoline-3,7-dione (13). To
a
stirred suspension of 12 (100 mg, 0.2 mmol) in 2 mL of
DMA, 1 mL of N,N-dimethylethylenediamine was
added. The mixture was stirred for 1 h at 60 ꢀC, diluted
with CHCl3 and Et2O and extracted with water. The
organic layer was dried over Na2SO4, and concentrated
to dryness. The residue was purified by flash chromato-
graphy (silica gel) eluting with CHCl3/MeOH (10:1).
The fractions containing the product were pooled and
concentrated to dryness. The product 13 (65 mg, 59%)
General procedure for the synthesis of compounds 17a–c
2-[2-(Dimethylamino)ethyl]-8-[[2-(dimethylamino)ethyl]-
amino]-2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-dione
(17a). A suspension of NaH (60% in mineral oil, 30
mg) in 3 mL DMA and 100 mg (0.4 mmol) of 15 was
stirred at room temperature for 15 min. After that 0.15
mL (1.5 mmol) of 2-(dimethylamino)ethyl chloride in 1
mL of benzene was added dropwise and the reaction
mixture heated for 1.5 h at 70 ꢀC. The mixture was
cooled, diluted with CHCl3, washed several times with
was obtained as a yellow powder. Mp 224–226ꢀ C. H
1
NMR (CDCl3) d 2.41 (s, 6H, 2ꢂCH3), 2.75 (t, 2H,
J=6.5 Hz), 3.56 (q, 2H, J=6.5 Hz), 5.21 (s, 2H), 5.29 (s,

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
570
water, dried over Na2SO4 and evaporated in vacuo. The
residue was purified on column chromatography (deac-
tivated silica gel) using the solvent system CHCl3/
MeOH/25% NH4OH (5:1:0.1). The purple product was
purified as described for 2a–f. Yield 65 mg (35%). Mp
134–135 ꢀC dec. IR (KBr, major peaks cmÀ1): 1390,
1480, 1590, 1620, 1650. 1H NMR (CDCl3) d 2.35 (s, 6H,
2ꢂCH3); 2.45 (s, 6H, 2ꢂCH3); 2.75 (t, 2H, J=3.2 Hz);
2.95 (t, 2H, J=6.6 Hz); 3.45 (q, 2H, J=3.1 Hz); 4.5 (t,
2H, J=6.6 Hz); 6.83 (d, 1H, J=7.5 Hz); 7.54 (t, 1H,
J=7.6 Hz); 7.75 (d, 1H, J=7.1 Hz); 7.94 (t, 1H, J=7.8
Hz); 8.53 (d, 1H, J=6.8 Hz); 8.62 (d, 1H, J=7.4 Hz);
9.95 (t, 1H, NH d ex). MS m/z (relative intensity, %):
403 ([MÀ2]+, 100); 405 ([M]+, 30); 406 ([M+1]+, 70).
Found: C, 67.89; H, 6.98; N, 17.10; calcd for
C23H27N5O2: C, 68.13; H, 6.71; N, 17.27.
8-[[2-(Dimethylamino)ethyl]amino]-2-methyl-2,7-dihydro-
3H-dibenzo[de,h]cinnoline-3,7-dione (21). A suspension
of 110 mg (0.4 mmol) of 20 and NaH (60% in mineral
oil, 30 mg) in 3 mL DMA was stirred at 60 ꢀC for 5 min.
Then 2-(dimethylamino)ethyl chloride in benzene solu-
tion (four times molar excess) was dropped in and the
reaction mixture was stirred at 60ꢀ C for 1 h. After
cooling the reaction mixture was diluted with CHCl3
and washed with water. The organic layer was dried
over Na2SO4 and evaporated under diminished pres-
sure. The residue was chromatographed on a silica gel
column (eluent CHCl3/MeOH 5:1) to afford 21 as a red
powder (yield 55%). Mp 280–281 ꢀC. 1H NMR (CDCl3)
d 2.4 (s, 6H, 2ꢂCH3); 2.7 (t, 2H, J=3.2 Hz); 3.4 (q, 2H,
J=3.0 Hz); 4.0 (s, 3H); 6.8 (d, 1H, J=7.4 Hz); 7.5 (t,
1H, J=7.7 Hz); 7.7 (d, 1H, J=7.4 Hz); 7.9 (t, 1H,
J=7.8 Hz); 8.5 (d, 1H, J=6.7 Hz); 8.6 (d, 1H, J=7.4
Hz); 9.9 (t, 1H, NH d ex). Found: C, 68.57; H, 5.89; N,
15.83; calcd for C20H20N4O2: C, 68.95; H, 5.79; N,
16.08.
General procedure for the synthesis of compounds 17d–e
2-[2-(Dimethylamino)ethyl]-8-[[2-(piperidin-1-yl)ethyl]-
amino]-2,7-dihydro-3H-dibenzo[de,h]cinnoline-3,7-dione
(17d). A suspension of NaH (60% in mineral oil, 30
mg) in 2.5 mL DMA and 130 mg (0.4 mmol) of 16a was
stirred at room temperature for 15 min. After that 0.2
mL (1.5 mmol) of 1-(2-chloroethyl)piperidine in 1 mL of
benzene was added dropwise and the reaction mixture
heated for 1 h at 70 ꢀC. The product was isolated and
purified similarly asꢀ described for 17a. Yield 38 mg
Membrane affinity measurements
HPLC column containing as stationary phase 1-myristoyl
-2-[13-carbonylimidazolide-tridecanoyl]sn -3-glycero-
phospholine(lecithin-imidazolide) bonded to silica-pro-
pylamine with the unreacted propylamine moieties end-
capped with C10 and C3 alkyl chains (IAM.PC.DD2)
was purchased from Regis Technologies Inc. (Morton
Grove, IL, USA). The IAM.PC.DD2 column was 3
cmꢂ4.6 mm; particle diameter 12 mm; pore diameter
300 A. For all studies the injection volume was ca. 10
mL of a solute aqueous solution. Acetonitrile/0.1 M
Sorensen buffer (K2HPO4/KH2PO4) pH 7.2 eluent was
used in proportions 50:50, 40:60, 35:65, 30:70 and
25:75% (v/v). The flow rate was 1 mL/min and solute
detection was at 495 nm. The chromatographic system
consisted of a Model L-6200 A pump, Model L-4250
UV–vis detector and a Model D-2500 chromatointe-
grator (all from Merck-Hitachi, Vienna, Austria).
Capacity factors, k0IAM, were calculated assuming that
the dead volume of the column was the signal given by
50 mg/mL citric acid solution. A standard, commercially
available statistical package for regression analysis was
employed on a personal computer.
1
(25%). Mp 136–138 C. H NMR (as dihydrochloride,
DMSO-d6) d 1.4–1.8 (m, 6H, 2ꢂCH3); 2.8 (s, 6H,
2ꢂCH3); 2.85 (t, 4H, J=7.3 Hz); 3.33 (t, 2H, J=6.5
Hz); 3.58 (t, 2H, J=6.4 Hz); 4.05 (q, 2H, J=6.4 Hz);
4.65 (t, 2H, J=6.4 Hz); 6.97 (d, 1H, J=7.35 Hz); 7.45
(t, 1H, J=7.6 Hz); 7.6 (d, 1H, J=7.0 Hz); 8.10 (t, 1H,
J=7.7 Hz); 8.46 (t, 1H, J=6.9 Hz); 8.50 (d, 1H, J=7.4
Hz); 9.8 (t, 1H, NH d ex). Found: C, 69.83; H, 6.74; N,
15.45; calcd for C26H31N5O2: C, 70.09; H, 7.01; N,
15.72.
Ethyl 5-amino-9,10-dioxo-9,10-dihydro-1-anthracenecar-
boxylate (19). 5-Amino-9,10-dioxo-9,10-dihydro-1-
anthracenecarboxylic acid (18) was refluxed for 10 h in
absolute ethanol saturated with hydrogen chloride. The
crude product was crystallized from CHCl3/Et2O to
afford 19 as a purple solid. Mp 158–160 ꢀC. H NMR
1
(CDCl3) d 1.4 (t, 3H, J=7.12 Hz); 4.5 (q, 2H, J=7.13
Hz); 6.0–6.2 (br s, ex); 7.0 (d, 1H, J=8.4 Hz); 7.46 (t,
1H, J=6.5 Hz); 7.6–7.72 (m, 2H); 7.8 (t, 1H, J=7.6
Hz); 8.38 (dd, 1H, J=7.6 Hz, J=1.4 Hz).
Fluorescence binding studies
The C50 values for ethidium displacement from CT-
DNA were determined using aqueous buffer (10 mM
Na2HPO4, 10 mM NaH2PO4, 1 mM EDTA, pH 7.0)
containing 1.26 mM ethidium bromide and 1 mM CT-
DNA respectively.
8-Amino-2-methyl-2,7-dihydro-3H-dibenzo[de,h]cinno-
line-3,7-dione (20). A suspension of 300 mg (1 mmol) of
19 and 0.53 mL (10 mmol) of methylhydrazine was
stirred at room temperature for 18 h. The progress of
the reaction was monitored by TLC in the solvent sys-
tem CHCl3/MeOH (50:1). To the reaction mixture some
amount of water was added and the collected precipitate
was washed with dilute HCl and water. The solid was
purified by crystallization from DMF to give 20 (160
All measurements were made in 10-mm quartz cuvettes
at 20 ꢀC using a Perkin–Elmer LS5 instrument (excita-
tion at 546 nm; emission at 595 nm) following a serial
addition of aliquots of a stock drug solution (ꢄ1 mM in
H2O). The C50 values are defined as the drug con-
centrations that reduce the fluorescence of the DNA-
bound ethidium by 50% and are reported as the mean
from three determinations. Apparent equilibrium bind-
ing constants were calculated from the C50 values (in
mg, 52%). Mp 287–289 ꢀC. H NMR (DMSO-d6) d 3.8
1
(s, 3H, 1ꢂCH3); 6.85 (t, 1H, J=4.1 Hz); 7.4 (m, 1H); 7.5
(t, 1H, J=4.2 Hz); 8.0 (t, 1H, J=4.0 Hz); 8.38–8.58 (m,
2H), 13.2 (s, 2H, NH2 ex).

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
571
mM) using: Kapp=(1.26/C50)ꢂKethidium, with a value of
Kethidium=107 MÀ1 for ethidium bromide.13,14
ogy, Poland. We also thank Dr. H. Kusnierczyk for the
in vivo tests.
Biological tests
References and Notes
Cell lines. Murine L1210 lymphocytic leukemia cells
were grown in RPMI 1640 medium supplemented
with 5% FBS (fetal bovine serum), penicillin G
(100,000 units/L), and streptomycin (100 mg/L).
Human myelogenous leukemia sensitive cell line K562
and Doxorubicin resistant subline K562/DX (ICIG,
Villejuif, France) were grown in RPMI 1640 medium
supplemented with 10% FBS, penicillin G (100,000
units/L), streptomycin (100 mg/L), and 2 mM l-gluta-
mine. K562/DX cells were exposed to Doxorubicin
(500 nM) for 1 week each month. Cells were not
exposed to the drug during the week prior to experi-
mental protocol. Cell lines were grown in a con-
trolled (air–5% CO2) humidified atmosphere at 37 ꢀC
and were transplanted three times a week. For the
experiments the cells in logarithmic growth were sus-
pended in the growth medium to give the final required
density.
1. Bradley, G.; Juranca, P. F.; Ling, V. Biochim. Biophys.
Acta 1988, 948, 87.
2. Dicato, M.; Duhem, C.; Pauly, M.; Ries, F. Cytokines Cell.
Mol. Ther. 1997, 3, 91.
3. Selassie, C. D.; Hansch, C.; Khwaja, T. J. Med. Chem.
1990, 33, 1914.
4. Showalter, H. D. W.; Weber, L. W.; Leopold, W. R.; Fry,
D. W.; Klohs, W. D.; Jackson, R. C.; Design, Tumor, Biology
and Biochemical Pharmacology of Anthrapyrazoles. In
Anthracycline and Anthracenodione-based Anticancer Agents;
Lown, J. W., Ed.; Elsevier Science: Amsterdam, The Nether-
lands, 1988; p 201.
5. Krapcho, A. P.; Menta, E.; Oliva, A.; Di Domenico, R.;
Fiocchi, L.; Maresch, M. E.; Gallagher, C. E.; Hacker, M. P.;
Beggiolin, G.; Giuliani, F. C.; Pezzoni, G.; Spinelli, S. J. Med.
Chem. 1998, 41, 5429.
6. Stefanska, B.; Dzieduszycka, M.; Bontemps-Gracz, M. M.;
Borowski, E.; Martelli, S.; Supino, R.; Pratesi, G.; De Cesare,
MA.; Zunino, F.; Kusnierczyk, H.; Radzikowski, C. J. Med.
Chem. 1999, 42, 3494.
In vitro cytotoxic evaluation. Cells of required density
were seeded and different concentrations of the drugs
were added. The experiments were carried out in a con-
trolled (air–5% CO2) humidified atmosphere at 37 ꢀC.
The exposure times were 48 h for L1210 cells and 72 h
for other cell lines. The cytotoxic activity (IC50 values)
of the compounds was defined as their in vitro con-
centrations causing 50% inhibition of cell growth after
continuous exposure to the drug, as measured by cell
counting with ZbI Coulter Counter (Coulter Electro-
nics, Ltd., UK). Results are given as the mean of at least
three independent experimentsꢁstandard error of the
mean (SEM). The resistance index was defined as the
ratio of IC50 value for resistant cell line to IC50 value for
sensitive cell line.
7. (a) Tarasiuk, J.; Stefanska, B.; P•odzich, I.; Tkaczyk-
Gobis, K.; Bontemps-Gracz, M. M.; Garnier-Suillerot, A.;
Borowski E. Abstract Book, 7th International Symposium
on Molecular Aspects of Chemotherapy; Sept. 8–11, 1999,
Gdansk, Poland; Abstract 20. (b) Bontemps-Gracz, M. M.;
Stefanska, B.; Dzieduszycka, M.; Borowski, E.; Antonini, I.;
Martelli, S.; Supino, R.; Zunino, F. Abstract Book, Multi-
dyscyplinarna Konferencja Nauki o Leku (Multidisciplinary
Conference of Medicinal Science), March 3–5, 1999, Mus-
zyna, Poland; Abstract P-80. (c) Tarasiuk, J.; Stefanska, B.;
P•odzich, I.; Tkaczyk-Gdois, K.; Seksek, O.; Martelli, S.;
Garnier-Suillerot, A.; Borowski, E. Brit J. Pharmacol. 2002,
135, 1513.
8. Murdock, K. C.; Child, R. G.; Fabio, P. F.; Angier, R. B.;
Wallace, R. E.; Durr, F. E.; Citarella, R. V. J. Med. Chem.
1979, 22, 1024.
Antitumor activity evaluation. Murine leukemia P388
was maintained by ip passages of tumor cells in DBA/2
mice. Studies were conducted according to standard
protocols of the US National Cancer Institute.21 For
experimental purposes, 106 cells were injected ip to
CD2F1 (Balb/cꢂDBA/2) mice on day 0. Twenty-four h
after tumor implantation, solutions of compounds in
physiological saline were administrated ip in a single
dose. The treated group consisted of six and control
group of 14 animals. The medium survival time (MST)
of the treated (T) and control (C) groups was deter-
mined, and the percent of T/C was calculated by using
the following formula: %T/C=[(MST treated)/(MST
control)]ꢂ100.
9. Showalter, H. D. H.; Johnson, J. L.; Hoftiezer, J. M.;
Turner, W. R.; Werbel, L. M.; Leopold, W. R.; Shillis, J.
L.; Jackson, R. C.; Elslager, E. F. J. Med. Chem. 1987, 30,
121.
10. (a) Cho•ody, W. M.; Martelli, S.; Konopa, J. J. Med.
Chem. 1992, 30, 493. (b) Kato, N.; Sugaya, T.; Mimura, T.;
Ikuta, M.; Kato, S.; Kuge, Y.; Tomioka, S.; Kasai, M. Synth-
esis 1997, 625.
11. (a) Ullmann, F.; van der Schalk, W. Liebigs Ann. 1912,
388, 199. (b) Ullmann, F.; Minaew, W. Liebigs Ann. 1912, 388,
217.
12. Dokunikhin, N. S.; Fain, V. A. I. Zh. Obshch. Khim. RU
1964, 34, 2372.
13. Antonini, I.; Polucci, P.; Kelland, L. R.; Menta, E.;
Pescalli, N.; Martelli, S. J. Med. Chem. 1999, 42, 2535.
14. Morgan, A. R.; Lee, J. S.; Pulleyblank, D. E.; Murray,
N. L.; Evans, D. H. Nucleic Acids Res. 1979, 7, 547.
15. Tkaczyk-Gobis, K.; Tarasiuk, J.; Stefanska, B.; Haber, P.;
Garnier-Suillerot, A.; Borowski, E. Abstract Book, 6th Inter-
national Symposium on Molecular Aspects of Chemotherapy;
July 9–12, 1997, Gdansk, Poland; Abstract 28.
16. Kaliszan, R.; Nasal, A.; Bucinski, A. Eur. J. Med. Chem.
1994, 29, 163.
Acknowledgements
The authors acknowledge the financial support of these
studies by the State Committee for Scientific Research
(KBN), Warsaw, Poland (Grant No. 4 PO5F 035 19),
and in part by Italian Ministry of Foreign Affairs, and
the Chemical Faculty Gdansk University of Technol-
17. Ong, S.; Liu, H.; Pidgeon, Ch. J. Chromatogr. A 1996,
728, 113.

´
B. Stefanska et al. / Bioorg. Med. Chem. 11 (2003) 561–572
572
18. Tkaczyk-Gobis, K.; Tarasiuk, J.; Seksek, O.; Stefanska,
B.; Borowski, E.; Garnier-Suillerot, A. Eur. J. Pharmacol.
2001, 413, 131.
19. Dzieduszycka, M.; Martelli, S.; Arciemiuk, M.; Bon-
temps-Gracz, M. M.; Kupiec, A.; Borowski, E. Bioorg. Med.
Chem. 2002, 10, 1025.
20. Mazerski, J.; Martelli, S.; Borowski, E. Polish J. Chem.
1997, 71, 338.
21. (a) Geran, R. I.; Greenberg, N. H.; Macdonald, M. M.;
Schumacher, A. M.; Abbott, B. J. Cancer Chemother. Rep.
1972, 3, 1. (b) In Vivo Cancer Models; NIH Publication No.
84–2635, 1984; p 27.