Synthesis and cytotoxic activity of a new potential DNA bisintercalator: 1,4-Bis{3-[N-(4-chlorobenzo[g]phthalazin-1-yl)aminopropyl]}piperazine

Article abstract of DOI:10.1016/j.bmc.2010.05.053

The synthesis of new 1,4-bisalkylamino (2-4) and 1-alkylamino-4-chloro (5-6) substituted benzo[g]phthalazines is reported. Compounds 2-4 and 6 were prepared both in the free and heteroaromatic ring protonated forms. Bifunctional 6 contains the 1,4-bisaminopropylpiperazine chain as a linker between the two heteroaromatic units, whereas 5 is its monofunctional analogue. The in vitro antitumour activity of the synthesized compounds has been tested against human colon, breast and lung carcinoma cells, and also against human glioblastoma cells. Results obtained show that all of them are active in all cases, but bifunctional 6·2HCl is remarkably effective against the four cell lines tested, exhibiting IC₅₀ values in the range of 10^-7 M, similar to those found for doxorubicin. The bifunctional structure of 6·2HCl enhances activity with respect to the monofunctional related compounds 5 and 7, leading to the highest activity among all the compounds tested. Molecular modelling of 6 suggests that those results could be indicative of DNA bisintercalation, which should be specially favoured in the diprotonated form 6·2HCl, a compound suitable for being studied more in depth in further biological tests. Measure of the DNA thermal melting curves show that the linear rise in T_m for bifunctional 6·2HCl is nearly twice than that one obtained for monofunctional 5, and supports the DNA-binding hypothesis.

Full text of DOI:10.1016/j.bmc.2010.05.053

Bioorganic & Medicinal Chemistry 18 (2010) 5301–5309
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and cytotoxic activity of a new potential DNA bisintercalator:
1,4-Bis{3-[N-(4-chlorobenzo[g]phthalazin-1-yl)aminopropyl]}piperazine
b
b
b
Juan Galisteo a, Pilar Navarro a, , Lucrecia Campayo , María J. R. Yunta , Fernando Gómez-Contreras ,
*
Janny A. Villa-Pulgarin c, Beatriz G. Sierra c, Faustino Mollinedo c, Jorge Gonzalez d, Enrique Garcia-España d
a Instituto de Química Médica, Centro de Química Orgánica Manuel Lora-Tamayo, CSIC, c/Juan de la Cierva 3, E-28006 Madrid, Spain
b Departamento de Química Orgánica, Facultad de Química, Universidad Complutense, E-28040 Madrid, Spain
c Instituto de Biología Molecular y Celular del Cáncer, Centro de Investigación del Cáncer, C.S.I.C.-Universidad de Salamanca, E-37007 Salamanca, Spain
d Departamento de Química Inorgánica, Instituto de Ciencia Molecular, Universidad de Valencia, E-46980 Paterna (Valencia), Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of new 1,4-bisalkylamino (2–4) and 1-alkylamino-4-chloro (5–6) substituted
benzo[g]phthalazines is reported. Compounds 2–4 and 6 were prepared both in the free and heteroaro-
matic ring protonated forms. Bifunctional 6 contains the 1,4-bisaminopropylpiperazine chain as a linker
between the two heteroaromatic units, whereas 5 is its monofunctional analogue. The in vitro antitu-
mour activity of the synthesized compounds has been tested against human colon, breast and lung car-
cinoma cells, and also against human glioblastoma cells. Results obtained show that all of them are active
in all cases, but bifunctional 6ꢀ2HCl is remarkably effective against the four cell lines tested, exhibiting
IC50 values in the range of 10ꢁ7 M, similar to those found for doxorubicin. The bifunctional structure of
6ꢀ2HCl enhances activity with respect to the monofunctional related compounds 5 and 7, leading to
the highest activity among all the compounds tested. Molecular modelling of 6 suggests that those results
could be indicative of DNA bisintercalation, which should be specially favoured in the diprotonated form
6ꢀ2HCl, a compound suitable for being studied more in depth in further biological tests. Measure of the
DNA thermal melting curves show that the linear rise in Tm for bifunctional 6ꢀ2HCl is nearly twice than
that one obtained for monofunctional 5, and supports the DNA-binding hypothesis.
Received 5 February 2010
Revised 14 May 2010
Accepted 18 May 2010
Available online 24 May 2010
Keywords:
Bisintercalating agents
Antitumour activity
1,4-Bisalkylaminobenzo[g]phthalazines
1-Bisalkylamino-4-chloro-
benzo[g]phthalazines
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
thermore, the simultaneous presence of two aromatic systems con-
nected by an appropriate linker is considered to enhance the
DNA as a carrier of genetic information is a major target for drug
interaction due to the ability of DNA-bound drugs to interfere with
transcription and/or replication, an essential step in cell growth
and division. Nucleic acids interact reversibly with a broad range
of chemical species that include water, metal ions and their com-
plexes, small organic molecules, and proteins. Intercalating agents
constitute the most important group of small organic molecules
that interact reversibly with the DNA double helix. Some of them
(i.e., daunorubicin, doxorubicin, idarubicin or mitoxantrone) are
valuable drugs currently used for the treatment of ovarian and
breast cancers and acute leukaemias.1 Major features for achieving
remarkable therapeutical results are: (i) an usually tricyclic planar
chromophore able to insert into the base pairs of the nucleotides
and (ii) a positive charge favouring electrostatic interactions with
nucleotides.2 That charge may be located in flexible side chains,
as occurs in mitoxantrone,3 or in an heteroaromatic ring containing
easily protonable sp2 nitrogens, as it is the case of amsacrine.4 Fur-
antitumour activity, since bisintercalation is usually associated
with higher DNA-binding affinities and slower dissociation rates
than those observed for the corresponding monomers.5,6
As a consequence of the above mentioned points, bifunctional
heteroaromatic molecules with the two chromophores linked by
flexible heteroaliphatic chains can be considered as targets in the
design of new drugs with antitumour properties. Therefore, a high
number of bifunctional structures derived form acridines,7 phena-
zines,8 indolocarbazoles,9 naphthalimides,10 anthracenyl isoxazol-
yl amides11 or indenoisoquinolines,12 are under study, and many
of them have shown powerful cytotoxic activities.
In connection with this matter, our research group has been
working in the preparation of two related series of nitrogenated
heterocyclic compounds with the general structures A and B sche-
matized in Figure 1. 1,4-Bisalkylaminobenzo[g]phthalazine deriva-
tives of the type
A
were synthesized by heating 1,4-
dichlorobenzo[g]phthalazine in the presence of an excess of the re-
quired amine in an autoclave without any solvent.13 One of those
compounds (Fig. 1A, X = OMe) was isolated as the monohydrochlo-
ride form, and its reaction with cobalt(II) dichloride in neutral
* Corresponding author. Tel.: +34 912587572; fax: +34 915644853.
E-mail address: iqmnt38@iqm.csic.es (P. Navarro).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.05.053

5302
J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
Finally, it has been reported that the introduction of a pipera-
zine unit in the flexible linker of some bifunctional heteroaromatic
molecules seems to favour biological activity. That is the case of
imidazoacridones linked to a nitronaphthalimide moiety by a
1,4-dipropanepiperazine unit, that are selective and potent com-
pounds against colon cancer and leukaemia,19 and also of other
symmetric bifunctional compounds with two heteroaromatic units
connected by the same linker20 that have shown to be effective
against HIV-1 infections.
HN
+
X
N
N
H
HN
X
X=OH,OR,N(R)2, Me
Taking into account all these previous findings, in this work we
want to report the synthesis and in vitro antitumour activity data
of two new monofunctional compounds with the schematic struc-
ture A: 3 and 4, and of the related mono- and bifunctional com-
pounds 5 and 6, containing one or two benzo[g]phthalazine units
linked to a 1,4-dipropanepiperazine chain (Fig. 2). Molecular mod-
elling studies and DNA thermal melting experiments are also in-
cluded in order to get some information regarding the potential
intercalating features of the most active compound 6, isolated as
the double hydrochloride salt.
NH
Cl
HN
N
N
N
N
Cl
NH, NMe
Figure 1.
Cl
medium afforded di-{1,4-bis-(3-methoxypropylamino)-3(2)H-
benzo[g]phthalazinium}2Co(II)Cl4. The crystalline structure of this
salt together with relevant spectroscopic data obtained from the
starting monohydrochloride salt, demonstrated that protonation
of the 1,4-bis(alkylamino)benzo-[g]phthalazine nucleus occurs at
the ring nitrogen atoms, being the positive charge shared by the
endo- and exocyclic nitrogen atoms.14
N
N
Cl
1
α
R
H
H
H
R
R
It was also verified that this tricyclic heteroaromatic nucleus
may be protonated at physiological pH depending on the nature
of the side chains. Thus, the pKa values of 1,4-bis(propylami-
no)benzo-[g]phthalazines functionalized with cationic groups
(Fig. 1A, X = NMe2) are close to 6, while functionalization with neu-
tral groups (Fig. 1A, X = OMe, OH) gives pKa values higher than 8.
By taking the 3-methoxypropyl derivative as a model, it was
shown that the ionizable 1,4-bis(alkylamino)-benzo[g]phthalazine
system interacts with DNA by intercalation at physiological pH.
Intercalation induces a positive superhelical torsion effect on circu-
lar DNA related to that one found for ethidium bromide.15
On the other hand, we have found that derivatives of the type A in
which the heteroatoms located at the end of the side chains are re-
placed by carbon atoms exhibit both cytotoxic and trypanosomicide
activities.16 Furthermore, the action of 1,4-bis(butylami-
no)benzo[g]phthalazine (ABP) on the cell cycle and differentiation
of U-937 human promonocytic leukaemia cells has been studied
and compared with that one of amsacrine (mAMSA).17 At subcyto-
toxic concentrations, mAMSA and ABP reduced the proliferation
activity to a similar extension and caused little mortality. The mAM-
SA induced the cell to accumulate at the G2 phase of the cycle, while
cycle inhibition originated by ABP was not phase specific. In addi-
tion, mAMSA caused an increase in the cell mass while ABP provoked
cell shrinkage. This was consistent with the fact that ABP consider-
ably inhibited protein synthesis, while mAMSA did not significantly
affect this activity. Finally, both drugs inhibited RNA synthesis, but
the inhibition was more prominent in the case of ABP.
In our previous work about the reactivity of 1,4-dic-
lorobenzo[g]phthalazine with nucleophiles, we had found that the
use of solvents of different nature allows handling the type of substi-
tution at the pyridazine ring: acetonitrile only provides the 1-alkyl-
amino-4-chlorosubstituted mono- and bifunctional products. This
finding allowed us to prepare also bifunctional derivatives of the
type B (Fig. 1), in which the two 1-aminopropylen-4-chloro-
benzo[g]phthalazine fragments are linked by NH or NMe groups.
Those compounds have shown in vitro cytotoxic activity against
HT-29 human cells, that is enhanced by methyl substitution at the
nitrogen atom.18
N
N
+
β
9
10
5
1
9a
5a
10a
8
N
N
N
N
Cl-
H
7
4a
4
6
β
N
N
H
α
R
2, R=CH2Me
[2]·HCl, R = CH
[3]·HCl, R = (CH
[4]·HCl, R = C6H5
2Me
3, R=(CH2)2-Me
4, R=C6H5
2
)
2
-Me
δ
´
γ
α´
γ
α
N
NH2
HN
N
β´
δ
β
9
6
10
5
1
9a
10a
8
7
N
N
5
6
5a
4a
4
Cl
δ
γ
α
N
NH
1
N
HN
δ
β
9
10
5
1
9a
10a
8
N
N
N
N
7
5a
4a
4
4
Cl
6
9
Cl
δ
γ
α
2Cl-
N
NH
HN
N
δ
β
10
5
1
+
+
9a
5a
10a
8
H
NH
N
N
N
7
4a
4
6
6·2HCl
Cl
Cl
γ
γ
α
α
Me
Me
HN
HN
β
β
10
5
10
9
6
9
1
1
+
9a
5a
9a
5a
10a
10a
8
8
7
N
N
NH
N
7
7
7·HCl
4a
4a
4
4
5
6
Cl
Cl
Figure 2.

J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
5303
2. Results and discussion
2.1. Chemistry
Cl
Cl
N
N
1
The preparation of compounds 2–4 in their hydrochloride form
has been performed by nucleophilic substitution of 1,4-dic-
hlorobenzo[g]-phthalazine according to a procedure previously
patented by us.13 The starting compound 1 was reacted in an auto-
clave with a great excess of butylamine, pentylamine or 2-phenyl-
ethylamine under conditions described in Scheme 1 (method A). In
this way, the corresponding amine acts simultaneously as solvent,
reactant, and acceptor of the hydrogen chloride excess. The hydro-
chlorides were obtained as yellow solids in yields ranging from
55% to 80%. Further treatment with aqueous NaOH led to the free
compounds 2–4, which were purified by column chromatography
and isolated in the solid state with yields from 74% to 82%. The free
compounds resulted to be very hygroscopic, and required filtration
through basic Al2O3 in order to obtain them in anhydrous form.
The monoalkylamino substituted compounds 5 and 6 were ob-
tained by refluxing a 1:1 mixture of 1 and 1,4-bis(aminopro-
pyl)piperazine in acetonitrile, in the presence of potassium
carbonate (method B in Scheme 2). Chromatographic work up of
the reaction crude in the presence of ammonium hydroxide al-
lowed the isolation of 5 and 6 as the free compounds with 46%
and 20% yields, respectively. When mononuclear compound 5
was reacted with 1 under the same conditions (method B), a sup-
plementary amount of 6 could be obtained in a 16% yield. On the
other hand, 6 was exclusively formed as the double hydrochloride
salt by treating free 6 with HCl generated in situ in a MeOH/Cl3CH
solution and further insolubilization of the hydrochloride by ethyl
ether addition.
Method B
NH2-(CH2)3-N
N (CH2)3-NH2
(i)
5 (M+1. 413)
6
(M+1, 625)
(46%)
Cl
(i) (20%)
(ii) (16%)
N
N
(ii)
Cl
Method B
Method B: Amine, K2CO3, MeCN / reflux, 7 days
º
HCl / MeOH-Cl3CD / 0 C, 1.5 h
6
6·2HCl
(M·2HCl-1, 698)
(88%)
Scheme 2.
tions of 14.3 and 9.64 ppm in, respectively, 5 and 6. The carbon
pairs 4a/10a and 5/10 follow also the same pattern, although
chemical shifts differences are substantially lower. In a similar
way, carbons and protons of the two aliphatic chains exhibit a
symmetric pattern in 6, but are differentiated as two distinct pro-
pane units in 5.
To establish the protonation sites of 6, a comparison between
the NMR spectra of free 6 and of 6ꢀ2HCl (Table 1) is relevant. The
spectra of both compounds, registered in DMSO-d6, show that
All the new compounds synthesized were unequivocally identi-
fied by their elemental microanalysis, IR, 1H NMR, and 13C NMR
spectroscopic data, as shown in Section 3. Mass spectra of the free
compounds were recorded as ESI in positive (M+1) mode, and
those of the hydrochlorides in negative (MꢀHClꢁ1) mode. Schemes
1 and 2 display the molecular ions obtained from the mass spectra
of all of them, which agree in all cases with the proposed
structures.
H10
0.78 ppm in the hydrochloride salt, whereas H5 is only shifted by
0.20 ppm. Furthermore, the hydrogen atoms in respect to the
, neighbouring the alkylamino group, is deshielded by
a
For an accurate assignment of the 1H and 13C NMR spectra sig-
nals, gHMBC and gHMQC experiments were especially useful. The
mono- and dialkylamino substitution products are easily differen-
tiated in both 1H and 13C NMR spectra on the basis of signals cor-
responding to rings A and B of the polyheterocyclic moiety. Protons
H5 and H10, which appear as an unique singlet in the 1H NMR spec-
tra of 2–4, lose their equivalence in the monosubstituted com-
pounds 5 and 6, with their chemical shifts differing in 0.14 and
0.21 ppm, respectively (Table 1). Accordingly, the carbon atom at-
tached to the alkylamino substituent (C1) is always deshielded
with respect to that one neighbouring the chlorine atom in the
13C NMR spectra: carbons C1 and C4, identical in 2–4, show varia-
same group experiment a shielding of 0.27 ppm in the protonated
form. Those facts point out towards protonation of the amino-imi-
no moiety, and not of the isolated piperazine ring. The protons of
the ammonium group appear in the spectrum of the hydrochloride
salt as a broad signal centred at 10.85 ppm, that is absent in the
free compound. In the 13C spectrum, protonation affects mainly
to the chemical shifts of the NH neighbouring C-1 carbon, that is
Table 1
Comparison of the most significant 1H and 13C NMR data (d, DMSO-d6) for compounds
6 and 7 and their hydrochlorides
α
α
HN
HN
β
β
10
5
10
5
1
1
+
NH
N
N
N
2·HCl (M·HCl-1, 357)(55%)
3·HCl (M·HCl-1, 385)(80%)
R(CH2)2-NH2
4
4
Cl
Cl
4·HCl (M·HCl-1, 453)(65%)
Method A
Cl
N
Compound
6
6ꢀ2HCl
10.85
8.95
9.84
3.36
151.31
144.61
127.04
127.61
39.93
7
7ꢀHCl
11.04
8.90
9.88
3.67
151.26
144.76
127.57
128.24
43.00
NH+
H-5
H-10
—
—
N
8.75
9.06
3.63
8.69
9.08
3.54
2, (M+1, 323)(82%)
3, (M+1, 351)(74%)
4, (M+1, 419)(78%)
Cl
1
H-a
5% aqueous NaOH
C-1
C-4
C-5
C-10
153.83
144.19
124.69
123.37
39.16
152.89
143.59
124.52
123.55
41.11
Method A: Amine, autoclave / 136 ºC, 24 h
C-a
Scheme 1.

5304
J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
shielded in 2.52 ppm, whereas C-5 an C-10 are deshielded about
2.5–4.5 ppm. Concerning the aliphatic chains, carbons are
a
deshielded by 0.8 ppm in the hydrochloride, a fact also indicative
of protonation at the same site. A final confirmation has been ob-
tained by synthesizing the monobutylamino derivative 7 [Fig. 2,
mp 158–160 °C, 89%, mass spectrum positive mode: 286 (M+1)]
in which the piperazine ring is not present, and its corresponding
monohydrochloride 7ꢀHCl [mp 274–276 °C, 88%, mass spectrum
negative mode: 320 (MꢀHClꢁ1), %]. It was found that chemical shift
variations for protons H5 and H10 in the free and hydrochloride
forms are closely related to those commented above for 6 and
6ꢀ2HCl (Table 1). The same happens with carbons C-4, C-5, C-10
and C-a, that are all of them deshielded in the protonated form,
whereas C-1 is shielded, in perfect agreement with results ob-
tained for 6 and 6ꢀ2HCl.
2.2. Functional antitumour activity
The in vitro antitumour activity for compounds 2–7 was deter-
mined by measurement of their cytostatic and cytotoxic properties
in human tumour cell lines by XTT assays (Table 2). The human cell
lines used were HT-29 (human colon adenocarcinoma), MCF-7 (hu-
man breast adenocarcinoma), NCI-H460 (human large cell carci-
noma of the lung), and SF-268 (human glioblastoma). Data
displayed in Table 2 show that the 1,4-bis(alkylamino) derivatives
2–4 present significant antiproliferative activity (10ꢁ6 Mꢁ1).
Bifunctional compound 6ꢀ2HCl, which exhibits both hydrophobic
and positively charged regions, is the most effective in inhibiting
the proliferation of all the human tumour cell lines studied with
an IC50 in the range of 10ꢁ7 M, with values similar to those of doxo-
rubicin, a well established drug used in clinic. Interestingly, the
mononuclear compound 7ꢀHCl, lacking the piperazine ring, was
rather inactive. The introduction of the piperazine ring in com-
pound 5 increased significantly the antitumour activity, and this
antiproliferative activity was even further increased by the pres-
ence of both the piperazine ring and another phthalazine group
in bifunctional 6ꢀ2HCl, thus warranting more specific biological
studies with this compound, that could be an interesting target
for further in vitro and in vivo cytotoxic tests.
Figure 3. Image of the monoprotonated 3–DNA complex. In this and the next
drawing, DNA carbon atoms are coloured in black, nitrogens in deep blue, oxygens
in red, hydrogens in pale grey, phosphorus in yellow and counterions in violet. All
the atoms of the intercalating molecule are coloured in light blue.
The heteroaromatic ring of protonated 3 intercalates between
cytosine-guanine base pairs while lateral chains are oriented up-
wards and downwards following the minor groove (Fig. 3). Theo-
retical stabilization for this complex is
higher than that one calculated
H = ꢁ15.67 kcal/mol) and smaller than that one of daunomici-
none (
H = ꢁ88.12 kcal/mol). A similar behaviour is found for 4,
with
D
H = ꢁ35.27 kcal/mol,
for rhodomicinone
(D
D
D
H = ꢁ29.90 kcal/mol. In this case, although a similar interca-
lation mode occurs, stabilisation is lower due to the bigger volume
of phenetylamine chains which difficult minor groove interaction.
With respect to the interaction of 6 with DNA, several consider-
ations can be made. First, the geometrical features of 6 seem to be
adequate for bisintercalation. Usual chain lengths for this interac-
0
tion mode are known to be about 11, 14 and 18 ÅA for bisintercala-
tion between two, three and four base pairs, 0 respectively.
Therefore, the calculated chain length for 6 of 12 ÅA, is adequate
for interaction between two base pairs.21 Since theoretical calcula-
tions of binding mode not conforming to the neighbour exclusion
principle have been proposed by different authors,22 the possibility
of bisintercalation with one base pair in between was taken into
consideration. Treatment of bisintercalation was done as stated
previously on duplexes of five and six base pairs, obtaining much
better results in the case of five base pairs. The protonation state
of compound 6 at physiological pH is a crucial factor for under-
standing its intercalating abilities. Preliminary calculations pointed
2.3. Molecular modelling
In order to gather some information about the ability of com-
pounds 3–6 for intercalating into DNA, we have performed theo-
retical modelling studies following the methodology described in
Section 3. The models obtained for the interaction of ligands 3
and 4 with DNA suggest that intercalation could be contributing
to the interaction.
Table 2
Inhibition of proliferation of human tumour cell lines by compounds 2–5, 6ꢀ2HCl and 7ꢀHCl
Compound
IC50 (M)
HT-29
MCF-7
NCI-H460
SF-268
2
3
4
5
2.35 0.21 ꢂ 10ꢁ6
2.70 0.14 ꢂ 10ꢁ6
2.95 0.10 ꢂ 10ꢁ6
1.05 0.32 ꢂ 10ꢁ5
4.97 0.10 ꢂ 10ꢁ7
>10ꢁ4
2.47 0.31 ꢂ 10ꢁ6
2.57 0.37 ꢂ 10ꢁ6
2.57 0.37 ꢂ 10ꢁ6
0.92 0.24 ꢂ 10ꢁ5
8.63 0.80 ꢂ 10ꢁ7
>10ꢁ4
2.03 0.65 ꢂ 10ꢁ6
2.37 0.58 ꢂ 10ꢁ6
1.83 0.52 ꢂ 10ꢁ6
0.88 0.20 ꢂ 10ꢁ5
7.42 0.60 ꢂ 10ꢁ7
8.55 0.64 ꢂ 10ꢁ5
2.64 0.1 ꢂ 10ꢁ7
2.63 0.40 ꢂ 10ꢁ6
2.70 0.35 ꢂ 10ꢁ6
2.63 0.40 ꢂ 10ꢁ6
1.26 0.28 ꢂ 10ꢁ5
6.65 0.20 ꢂ 10ꢁ7
8.80 0.56 ꢂ 10ꢁ5
2.20 0.1 ꢂ 10ꢁ7
6ꢀ2HCl
7ꢀHCl
Doxorubucin
2.80 0.2 ꢂ 10ꢁ7
3.24 0.1 ꢂ 10ꢁ7
Data are shown as mean values S.D. of three independent determinations.
HT-29: human colon carcinoma cell line.
MCF-7: human breast adenocarcinoma cell line.
NCI-H460: human non-small cell lung cancer cell line.
SF.268: human glioblastoma cell line.

J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
5305
Figure 5. Melting temperatures values (°C) of (ct)-DNA upon addition of studied
compounds 5 and 6ꢀ2HCl at different ratios r at pH 7 (r = [drug]/[base-pair], [base-
pair] = 7.5 ꢂ 10ꢁ6 M) buffer sodium cacodylate, I = 0.05 mol dmꢁ3).
pair). The increase in the slope is directly proportional to the num-
ber of occupied intercalation sites in DNA, suggesting that bisinter-
calation might occur for 6ꢀ2HCl.23 This result is in accordance with
the theoretical studies commented above.
3. Experimental
Figure 4. Image of the diprotonated 6–DNA complex.
The starting materials were purchased from commercial
sources (Aldrich or Fluka) and used without further purification.
1,4-Dichlorobenzo[g]-phtalazine was obtained from 2,3-naphtha-
lene-dicarboxylic acid as reported previously by our group.13 Sol-
vents were dried using standard techniques.24 All the reactions
were monitored using thin layer chromatography (TLC) on pre-
coated aluminium sheets of Silica Gel 60F254. Compounds were
detected with UV light (245 nm). Chromatografic separations were
performed on columns in the indicated solvent system using flash
chromatography on silica gel (particle size 0.040–0.063 mm or
standard techniques on basic aluminium oxide. Melting points
were determined in a Reichert Jung hot-stage microscope. The 1H
and 13C NMR spectra were recorded with Varian Unity XL-300, Var-
ian Unity Inova-400, or Varian Unity 500 spectrometers at room
temperature employing CDCl3, CD3OD or DMSO-d6 as solvents.
Chemical shifts are reported in ppm from tetramethylsilane (d
scale). All assignments were performed on the basis of 1H–13C het-
eronuclear multiple quantum coherence experiments (gHSQC and
gHMQC). IR spectra were recorded on a Perkin Elmer 681 spec-
trometer. Electrospray mass spectra were recorded with a Hew-
lett–Packard 1100 MSD apparatus and fast atomic bombardment
mass spectra with a VG Autospec spectrometer using m-nitroben-
zyl alcohol (NBA) matrix. Elemental analyses were provided by the
Departamento de Análisis, Centro de Química Orgánica ‘Manuel
Lora-Tamayo’, CSIC, Madrid, Spain.
out that protonation of the pyridazine ring nitrogens should be fa-
voured in all cases with respect to the piperazine ring by a differ-
ence of 5.18–7.09 kcal/mol for monoprotonatión and 12.17 kcal/
mol for diprotonation. If we consider the Nꢀ ꢀ ꢀH interactions in suc-
cessive protonations of compound 6, it is shown that both the first
and second proton bind respectively to the two pyridazine nitro-
gens of the first and second benzo[g]phthalazine systems, creating
a symmetric structure in accordance with the NMR spectra ob-
tained for this compound. The above results were considered for
calculations in the theoretical intercalation studies performed. Fig-
ure 4 displays the most stable geometries obtained for complexes
of the diprotonated ligand 6 through bisintercalation on duplexes
of five base pairs.
The preferred geometry found for the diprotonated 6–DNA
complex, with a
D
H value of ꢁ122.21 kcal/mol, corresponds to
an arrangement of the two heterocyclic units in a ‘head on’ orien-
tation that could account for the stability value found. Stabilizing
hydrogen bondings between both exocyclic NH groups and their
upper heterocyclic systems have been detected. Finally, the consid-
eration of six base pairs duplexes leads to
D
H value of ꢁ42 kcal/
mol, with hydrogen bondings between the base pairs and the cyclic
nitrogens. All remarks pointed above are compatible with the
hypothesis that compound 6ꢀ2HCl could have DNA bisintercalating
properties by interaction with base pairs in between.
3.1. Synthesis of the 1,4-bis(alkylamino)benzo[g]-phthalazine
2.4. DNA melting studies
derivatives 2–4
After pondering the results obtained in the molecular modelling
of the DNA–6ꢀ2HCl complex, it could be relevant to obtain some
experimental information about the bisintercalating ability of
bifunctional 6ꢀ2HCl with respect to monofunctional 5. Therefore,
we have performed a DNA melting study in the presence of both
ligands.
Following the method indicated in Section 3, we looked at the
effect caused by the presence of 5 and 6ꢀ2HCl on DNA melting
curves measured with ct-DNA. Figure 5 plots Tm as a function of
the average number of ligand molecules per ct-DNA.
General procedure A: A mixture of 1,4-dichloro-benzo[g]phthal-
azine (248 mg, 1 mmol) and the corresponding amine (5 mL) was
heated in an autoclave at 130 °C for 24 h. After cooling to room
temperature the excess of the amine was evaporated to dryness
under vacuum. The residue was treated with acetone and the solid
was filtered off and dried to give the corresponding hydrochloride,
which was transformed into the free compound by treatment with
5% aqueous sodium hydroxide (5 mL) and further extraction with
chloroform. The organic layer was washed with water and dried.
Removal of the solvent under vacuum afforded the free amine,
which was purified on silica gel by flash column chromatography,
using hexane/ethyl acetate/methanol (v/v 1:1:0.3) as the eluent.
The linear rise in Tm is 20.6 °C/(drug/base-pair) for 6ꢀ2HCl,
which is nearly twice the one obtained for 5, 10.5 °C/(drug/base-

5306
J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
3.1.1. 1,4-Bis(butylamino)benzo[g]phthalazine monohydrochlo-
ride (2ꢀHCl)
H-8), 4.75 (br s, 2H, NH), 3.62 (t, 4H, H-
a), 1.78 (m, 4H, H-b),
1.41 (m, 8H, H-
c,
H-d) 0.91 (t, 6H, Me); 13C NMR (Cl3CD,
Following the general procedure A, the reaction of 1,4-dichloro-
benzo[g]phthalazine 1 and butylamine afforded a yellow solid
identified as 2ꢀHCl (212 mg, 55%, mp 195–197 °C). Anal. Calcd for
75 MHz): d 149.26 (C-1, C-4), 133.61 (C-5a, C-9a), 128.57 (C-7, C-
8), 127.66 (C-6, C-9), 121.01 (C-5, C-10), 118.49 (C-4a, C-10a),
42.32 (C-a), 29.53 (C-b), 29.39 (C-c), 22.57 (C-d), 14.09 (Me);
MS-ESI (positive mode, Cl3CH) m/z (%): 351 (M+1) (100), 307
C
20H26N4ꢀHClꢀ0.5H2O: C, 63.74; H, 7.81; N, 14.68. Found: C, 64.87,
H, 7.72; N, 14.92; IR (KBr, cmꢁ1) 3320–2660 (3240, 3100, 3000,
2940, 2920, 2850), 1625, 1575, 1545, 1425, 1360, 1325, 1155,
1025; 1H NMR (CD3OD, 300 MHz): d 9.10 (s, 2H, H-5, H-10), 8.33
(60), 290 (45), 210 (83), 210(12), 185 (15).
3.1.5. 1,4-Bis(phenetylamino)benzo[g]phthalazine monohydro-
chloride (4ꢀHCl)
(m, 2H, H-6, H-9), 7.97 (m, 2H, H-7, H-8), 3.67, (t, 4H, H-
a), 3.49
(m, 1H, NH), 2.00 (m, 4H, H-b), 1.72, (m, 4H, H-
c
), 1.22 (t, 6H,
Following the general procedure A, the reaction of 1,4-dic-
hlorobenzo[g]phthalazine 1 and phenetylamine afforded a yellow
solid which after being crystallized from acetonitrile was identified
as 4ꢀHCl (295 mg, 65%) [mp 143–145 °C (dec)]. Anal. Calcd for
Me); 13C NMR (CD3OD, 75 MHz): d 150.21 (C-1, C-4), 136.24 (C-
5a, C-9a), 131.23 (C-7, C-8), 130.40 (C-6, C-9), 126.60 (C-5, C-10),
119.26 (C-4a, C-10a), 43.40 (C-a), 31.77 (C-b), 21.68, (C-c), 14.51
(Me); MS-ESI (negative mode, MeOH), m/z (%): 357.1(MꢀHClꢁ1)
C
28H26N4ꢀHClꢀ0.5H2O: C, 72.49; H, 6.04; N, 12.08. Found: C,
(100).
72.06; H, 6.53; N, 12.24; IR (KBr, cmꢁ1): 3054, 3025, 2925, 2862,
1625, 1549, 1453, 1351, 1197, 1114, 1031, 893, 750, 699; 1H
NMR (CD3OD, 300 MHz): d 9.12 (br s, 2H, H-5, H-10), 8.31 (br s,
2H, H-6, H-9), 7.90 (br s, 2H, H-7, H-8), 7.47 (m, 8H, H-20, H-30),
3.1.2. 1,4-Bis(butylamino)benzo[g]phthalazine (2)
The hydrochloride 2ꢀHCl (100 mg, 0.28 mmol) was treated with
5% NaOH as stated in the general procedure and purified on silica-
gel by flash column chromatography (hexane/ethyl acetate/metha-
nol, v/v 1:1:0.3). The fraction with Rf = 0.11 afforded a hygroscopic
solid, which was filtered through a small column of basic alumin-
ium oxide using chloroform as the eluent. Removal of the solvent
under vacuum yielded 74 mg (82%) of anhydrous compound 2 with
mp 177–179 °C. Anal. Calcd for C20H26N4: C, 74.53; H, 8.07; N,
17.39. Found: C, 74.78, H, 7.96; N, 17.31; IR (KBr, cmꢁ1), 3300,
3050, 2950, 2910, 2850, 1620, 1495, 1425, 1360, 1220, 1140; 1H
NMR (Cl3CD, 300 MHz): d 8.08 (s, 2H, H-5, H-10), 7.87 (m, H-6,
H-9), 7.47 (m, 2H, H-7, H-8), 4.65 (br s, 2H, NH), 3.54 (t, 4H, H-
7.38 (2H, H-40), 3.93 (br s, 4H, H-
a), 3.50 (m, 2H, NH), 3.30 (t, 4H,
H-b); 13C NMR (CD3OD, 75 MHz): d 150.52 (C-1, C-4), 141.5 (C-
10), 136.00 (C-5a,C-9a), 131.03 (C-7, C-8), 130.12 (C-6, C-9),
129.94 (C-20), 129.59 (C-30), 127.80 (C-40), 127.57 (C-5, C-10),
118.86 (C-4a, C-10a), 44.77 (C-a), 35.52 (C-b); MS-ESI (negative
mode, MeOH), m/z (%): 453.1 (MꢀHCl-1) (100).
3.1.6. 1,4-Bis(phenetylamino)benzo[g]phthalazine (4)
The hydrochloride 4ꢀHCl (100 mg, 0.22 mmol) was treated with
0.5% NaOH as stated in the general procedure and purified on sili-
cagel column by flash chromatography (hexane/ethyl acetate/
methanol, v/v 1:1:0.3). The fraction with Rf = 0.46 (MeOH) afforded
a solid, which was filtered through a small column of basic alumin-
ium oxide using chloroform as the eluent. Removal of the solvent
under vacuum yielded 72 mg (78%) of an orange compound 4 with
mp 65–67 °C. Anal. Calcd for C28H26N4ꢀ0.25H2O: C, 79.52; H, 6.27;
N, 13.25. Found: C, 79.60, H, 6.22; N, 13.47; IR (KBr, cmꢁ1): 3435,
1622, 1496, 1347, 748, 699. 1H NMR (Cl3CD, 300 MHz): d 8.15 (s,
2H, H-5. H-10), 7.96 (m, 2H, H-6, H-9), 7.80 (m, 8H, H-20, H-30),
7.59 (m, 2H, H-7, H-8), 7.22 (m, 2H, H-40), 4.78 (br s, 2H, NH),
a
), 1.66 (m, 4H, H-b), 1.40 (m, 4H, H-c
), 0.88 (t, 6H, Me); 13C
NMR (Cl3CD, 75 MHz): d 149.74 (C-1, C-4), 133.65 (C-5a, C-9a),
128.57 (C-7, C-8), 127.70 (C-6, C-9), 121.73 (C-5, C-10), 118.45
(C-4a, C-10a), 42.09 (C-a), 31.73 (C-b), 20.49 (C-c), 13.93 (Me);
MS-EI m/z (%) 323.0 (M+1) (11), 322.0 (M) (51), 279.1 (100),
178.9 (81), 43.2 (24), 41.2 (27).
3.1.3. 1,4-Bis(pentylamino)benzo[g]phthalazine monohydro-
chloride (3ꢀHCl)
Following the general procedure A, the reaction of 1,4-dichloro-
benzo[g]phthalazine 1 and pentylamine afforded a yellow solid
which was identified as 3ꢀHCl after crystallization from acetonitrile
[309 mg, 80%, mp 185–187 °C (dec)]. Anal. Calcd for
3.96 (c, 4H, H-a
), 3.12 (t, 4H, H-b); 13C NMR (Cl3CD, 75 MHz): d
149.13 (C-1, C-4), 140.00 (C-10), 133,62 (C-5a, C-9a), 129.00 (C-
20), 128.59 (C-7, C-8), 128.53 (C-30), 127.74 (C-6, C-9), 126.25 (C-
40), 120.97 (C-5, C-10), 118.44 (C-4a, C-10a), 43.00 (C-
a), 35.44
C
22H30N4ꢀHClꢀH2O: C, 65.26; H, 7.66; N, 13.84. Found: C, 65.87, H,
(C-b); MS-ESI (positive mode, Cl3CH): m/z (%) 419 (M+1) (100).
7.94; N, 13.77; IR (KBr, cmꢁ1): 3429, 3269, 3054, 2956, 2929,
2859, 1625, 1551, 1466, 1354, 1148, 912, 753, 670; 1H NMR
(CD3OD, 300 MHz): d 9.12 (s, 2H, H-5, H-10), 8.34 (m, 2H, H-6,
3.2. Synthesis of the 1-alkylamino-4-chlorobenzo[g]-phthal-
azine derivatives
H-9), 7.98 (m, 2H, H-7, H-8), 3.67 (t, 4H, H-
), 1.16 (t, 3H, Me); 13C NMR (CD3OD, 75 MHz): d 150.23 (C-1,
C-4), 135.91 (C-5a, C-9a), 130.93 (C-7, C-8), 130.10 (C-6, C-9),
126.27 (C-5, C-10), 118.67 (C-4a, C-10a), 43.36 (C- ), 30.46 (C-b),
29.10 (C- ), 23,54 (C-d), 14.39 (Me); MS-ESI (negative mode,
MeOH), m/z (%): 385.1 (100) (MꢀHCl-1).
a), 1.65 (m, 8H, H-b,
Hc
General procedure B: To a refluxed suspension of 1,4-dic-
hlorobenzo[g]phthalazine (248 mg, 1 mmol) and anhydrous potas-
sium carbonate (1.38 g, 10 mmol) in anhydrous acetonitrile
(130 mL) a solution of the corresponding amine in anhydrous ace-
tonitrile (50 mL) was slowly added. The reaction was further re-
fluxed and monitored by TLC plates until completion (7 days).
a
c
3.1.4. 1,4-Bis(pentylamino)benzo[g]phthalazine (3)
The hydrochloride 3ꢀHCl (100 mg, 0.28 mmol) was treated with
0.5% NaOH as stated in the general procedure and purified on sili-
cagel by flash column chromatography (hexane/ethyl acetate/
methanol, v/v 1:1:0.3). The fraction with Rf = 0.35 (MeOH) afforded
a solid, which was filtered through a small column of basic alumin-
ium oxide using chloroform as the eluent. Removal of the solvent
under vacuum yielded 68 mg (74%) of an orange compound 3 with
mp 134–136 °C. Anal. Calcd for C22H30N4ꢀ0.5H2O: C, 73.53; H, 8.35;
N, 15.59. Found: C, 73.43, H, 8,76; N, 15.41; IR (KBr, cmꢁ1): 3435,
2929, 1626, 1547, 1356, 750; 1H NMR (Cl3CD, 300 MHz): d
8.24(s, 2H, H-5, H-10), 8.02 (m, 2H, H-6, H-9), 7.61 (m, 2H, H-7,
3.2.1. Synthesis of compounds 5 and 6
Following method B, 1,4-dichlorobenzo[g]phthalazine was
made react with 1,4-bis(propylamino)piperazine (200 mg,
1 mmol). After cooling to room temperature, a yellow solid con-
taining the reaction products and potassium carbonate was sepa-
rated by filtration and extracted with chloroform. The organic
solution was evaporated under reduced pressure to give a brown
residue which was purified on silica gel by flash column chroma-
tography (chloroform/ethanol/25% ammonium hydroxide v/v
3:1:0.1). Following monitorization by TLC, the appropriate frac-
tions were combined to give two products with Rf 0.12 and 0.79.

J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
5307
3.2.1.1.
1-[3-(4-Aminopropylpiperazin-1-yl)propyl-amino]-4-
The most retained fraction
1)(55), 505.2 (30), 475.2 (100). ESI (positive mode, MeOH) m/z
(%): 625.5 [(M+1)ꢁ2HCl] (60), 270.1 (100).
chlorobenzo[g]phthalazine (5).
(Rf = 0.12) afforded 190 mg (46%) of an orange solid with mp
250–252 °C. Anal. Calcd for C22H29N6ClꢀH2O: C, 61.32; H, 7.2; N,
19.51. Found: C, 61.82; H, 7.33; N, 19.55; IR (KBr, cmꢁ1), 3435,
2932, 1630, 1555, 1505, 1431, 1156; 1H NMR (CDCl3, 500 MHz), d
8.59 (s, 1H, H-10), 8.45 (s, 1H, H-5), 8.08 (m, 2H. H-6, H-9), 7.63
Alternatively, the double hydrochloride of 6 was also ob-
tained as follows: a solution of 1,4-bis-(propylamino)piperazine
(200 mg, 0.75 mmol) in anhydrous acetonitrile (50 mL) was
slowly added to
a suspension of 1,4-dichlorobenzo[g]phthal-
azine (1 mmol) in anhydrous acetonitrile (130 mL), and one
drop of HCl 6 N was added. The mixture was further refluxed
and monitored by TLC plates until the reaction was completed.
The organic solution was evaporated under reduced pressure
and the residue was purified on silica gel by flash column chro-
matography (chloroform/ethanol/25% ammonium hydroxide v/v
3:1:0.1). After monitorization by TLC, the appropriate fractions
were combined to give a brown solid of Rf = 0.84 which, after
drying under vacuum yielded 104 mg (29%) of 6ꢀ2HCl, with
mp 250 °C (dec). MS-ESI (positive mode, MeOH) m/z (%):
625.2363 [(Mꢀ2HCl +1)ꢁ2HCl] (38), 356.1647 (77), 270.0864
(100), 230.0481 (38).
(m, 2H, H-7, H-8), 3.78 (t, 2H; H-
10H, 2H-
a
), 2.83 (m, 2H, H-a0), 2.63 (m,
c
, 4H-d, 4H-d0), 2.55 (t, 2H, H-c0), 1.94 (m, 2H, H-b), 1.72
(m, 2H, H-b0); 13C NMR (CDCl3, 125 MHz), d 159.83 (C-1), 145.50
(C-4), 134.24, 134.35 (C-5a, C-9a), 129.17 (C-6, C-9), 128.25 (C-7,
C-8), 125.59 (C-5), 122.90 (C-4a), 121.71 (C-10), 118.40 (C-10a),
58.63 (C-c a),
), 56.82 (C-c0), 53.63, 53.58 (C-d, C-d0), 43.35 (C-
40.83 (C-a0), 29.13 (C-b0), 23.77 (C-b); MS-FAB, m/z (%): 413
(M+1, 100), 270 (80), 154 (20).
3.2.1.2. 1,4-Bis-{3-[N-(4-chlorobenzo[g]phtalazin-1-yl)-amino-
propyl]}piperazine (6).
The less retained fraction (Rf = 0.79)
gave 122 mg (20%) of a yellow solid with mp 244–246 °C. Anal.
Calcd for C34H34N8Cl2ꢀ1.5H2O: C, 62.57; H, 5.67; N, 17.17. Found:
C, 62.52; H, 5.95; N, 16.94; IR (KBr, cmꢁ1), 3435, 3277, 2888,
2800, 1622, 1544, 1518, 1417, 1098, 887; 1H NMR (DMSO-d6,
500 MHz), d 9.06 (s, 2H, H-10), 8.75 (s, 2H, H-5), 8.36 (m, 2H, H-
3.2.2. 1-Butylamino-4-chlorobenzo[g]phthalazine 7
Following the general procedure B, 1,4-dichlorobenzo[g]phthal-
azine
(249 mg,
1 mmol),
and
butylamine
(366 mg,
5 mmol)(0.49 mL) were reacted. After cooling to room tempera-
ture, potassium carbonate was eliminated by filtration, and the or-
ganic solution evaporated under reduced pressure. The brown
residue was purified on silicagel column by flash chromatography
(Cl3CH/EtOH, v/v 30:1). The appropriate fractions (Rf 0.37) moni-
tored by TLC were combined to give 7 (253 mg, 89%) as a yellow
solid with mp 158–160 °C. Anal. Calcd for C16H16N3Cl: C, 67.48;
H, 5.27; N, 14.76. Found: C, 67.55; H, 5.55; N,14.64; IR (KBr,
cmꢁ1), 3422, 3271, 2956, 2869, 1564, 1520, 1423, 1379, 1288,
1159, 1119, 1069, 893, 745, 668; 1H NMR (DMSO-d6): d 9.08 (s,
1H, H-10), 8.69 (s, 1H, H-5), 8.32 (br s, H-9), 8.11 (br s, 1H, H-6),
7.66 (m, 2H, H-7, H-8), 5.5 (br signal, 1H, NH), 3.54 (br s, 2H, H-
6), 8.16 (m, 2H, H-9), 7.78 (m, 4H, H-7, H-8), 3.63 (m, 4H, H-
a),
3.1–2.65 (very broad signal, 8H, H-d), 2.62 (m, 4H-
c), 1.98 (m,
4H, H-b); 13C NMR (DMSO-d6, 125 MHz), d 153.83 (C-1), 144.19
(C-4), 134.07, 134.01 (C-5a, C-9a), 129.10 (C-6), 128.75 (C-9),
128.59 (C-7, C-8), 124.69 (C-5), 123.37 (C-10), 122.16 (C-10a),
117.81 (C-4a), 55.45 (C-c), 51.96 (C-d), 39.16 (C-a), 26.04 (C-b);
MS-ESI (positive mode, MeOH) m/z (%): 625 (M+1, 18), 583 (22),
425 (18), 423 (46), 384 (34), 271 (23), 270 (100).
Following the general procedure B, 1,4-dichlorobenzo[g]phthal-
azine (30 mg, 0.12 mmol) was made react with compound 5
(50 mg, 0.12 mmol). After cooling to room temperature, potassium
carbonate was eliminated by filtration, the organic solution evapo-
rated under reduced pressure, and the residue purified on silicagel
by flash column chromatography (chloroform/ethanol/aq 25%
ammonium hydroxide v/v 3:1:0.1). The combined fractions with
Rf = 0.79 gave an additional amount of a yellow solid identified
as compound 6 (12 mg, 16%).
a
), 2.49 (br s, 2H, H-b), 1.36 (br s, 2H, H-c
), 0.90 (t, 3H, Me); 13C
NMR (Cl3CD): d 152.89 (C-1), 143.59 (C-4), 133.35, 133.62 (C-5a.
C-9a), 128.72 (C-9), 128.29 (C-6), 128.55, 128.47 (C-7,C-8),
124.52 (C-5) 123.55 (C-10), 121.67 (C-10a), 117.40 (C-4a), 41.11
(C-a), 27.47 (C-b), 21.58 (C-c), 13.53 (CH3); MS-ESI (positive mode
MeOH) m/z (%): 286 (M+1) (100).
3.2.1.3. 1,4-Bis-{3-[N-(4-chlorobenzo[g]phtalazin-1-yl)-amino-
3.2.3. 1-Butyl-4-chlorobenzo[g]phthalazine hydrochloride 7ꢀHCl
A solution of 1-butyl-4-chlorobenzo[g]phthalazine 7 (86 mg,
0.3 mmol) in 10 mL of methanol and 5 mL of chloroform was pre-
pared, and a solution of 0.5 mL of freshly distiled acetyl chloride in
20 mL of methanol, maintained under argon atmosphere and
cooled down to 0 °C for 30 min, was added dropwise. After finish-
ing addition the solution was stirred for 1 h at 0 °C, and 80 mL of
ether were added. After stirring for 1.5 h more the organic solution
was partially concentrated until a precipitate was formed. The
reaction mixture was filtered under argon to give the titled com-
pound 7ꢀHCl (33 mg, 88%) as a pale brown solid with mp 274–
276 °C. Anal. Calcd for C16H17N3Cl2ꢀH2O: C, 56.48; H, 5.63; N,
12.35; Cl, 20.84. Found: C, 66.66; H, 5.68; N, 12.69; Cl, 20.84. IR
(KBr, cmꢁ1): 2435, 3159, 3016.48, 2958, 2928, 2871, 2747, 1620,
1463, 1378, 1289, 1161, 1122, 971, 893, 753, 661; 1H NMR
(DMSO-d6, 400 MHz): d 11.04 (low s, 2H, +NH), 9.88 (s, 2H, H-10),
8.90 (s, 2H, H-5), 8.46 (m, 2H, H-6), (8.23 (m, 2H, H-9), 7.93 (m,
propyl]}piperazine hydrochloride (6ꢀ2HCl).
A solution of
1,4-bis{3-[N-(4-chlorobenzo[g]-phthalazin-1-yl)aminopropyl]}
piperazine, 6, (42 mg, 0.067 mmol) in 10 mL of methanol and 5 mL
of chloroform was prepared, and a solution of 0.4 mL of freshly
distiled acetyl chloride in 15 mL of methanol, maintained under ar-
gon atmosphere and cooled down to 0 °C for 30 min, was added
dropwise. At the end of the addition the solution looked turbid
and, after stirring for 1 h at 0 °C, the presence of a heavy precipitate
was evident. After that, 15 mL of ether were added. One hour more
of stirring and the reaction mixture was filtered under argon to
give 41.5 mg (88%) of a pale brown solid that was identified as
the dihydrochloride of compound 6. Mp 250–252 °C. Anal. Calcd
for C34H36 Cl4N8ꢀ0.8Cl3CHꢀ8H2O: C, 44.02; H, 5.68; N, 12.06; Cl,
24.43. Found: C, 43.70; H, 5.60; N, 12.48, Cl, 24.54. IR (KBr,
cmꢁ1), 3435, 2923, 2889, 2730, 2622, 2510, 1622, 1462, 1379,
1191, 1163, 1124; 974, 758, 664. 1H NMR (DMSO-d6, 500 MHz):
d 10.85 (broad signal (2NH+), 9.84 (s, 2H, H-10), 8.95 (s, 2H, H-5),
8.48 (s, 2H, H-6), 8.24 (m, 2H, H-9), 7.94 (m, 4H, H-7, H-8), 3.83
4H, H-7, H-8), 3.67 (m, 4H, H-
a), 1.76 (m, 4H, H-b), 1.49 (m, 4H,
H-c
), 0.95 (t, 6H, Me); 13C NMR (DMSO-d6, 100 MHz): d 151.26
(br s, 4H, H-
a
), 3.62 (m, 8H, H-d), 3.36 (m, 4H, H-
c) 2,26 (br s,
(C-1), 144.74 (C-4), 135.42 (C-9a), 134.41 (C-5a), 131.10 (C-7),
130.82 (C-8), 129.87 (C-6), 129.42 (C-9), 128.24 (C-10), 127.57
4H, H-b); 13C NMR (DMSO-d6, 125 MHz), d 151.31 (C-1), 144.61
(C-4), 135.04 (C-9a), 133.08 (C-5a), 130.33 (C-7,8), 129.41 (C-6),
128.89 (C-9), 127.61 (C-10), 127.04 (C-5), 121.55 (C-9a), 117.82
(C-5), 122.03 (C-10a), 118.28 (C-4a), 43.00 (C-
a), 29.83 (C-b),
19.94 (C- ), 14.03 (Me). MS-ESI (negative mode MeOH) m/z (%):
c
(C-4a), 52.50 (C-
c
), 47.83 (C-d), 39.93 (C-
a
), 21.98 (C-b); ESI (neg-
320 (Mꢁ1) (100); MS-ESI (positive mode) m/z (%): 286
ative mode, MeOH) m/z (%): 698.3 (Mꢀ2HCl-1, 35), 659.2 (MꢀHCl-
[(M+1)ꢁHCl] (100).

5308
J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
3.3. Cell culture
computing non-bonded interactions between the reactants. Alter-
ations to ligand side chain conformations were performed in order
to minimize close intermolecular contacts with the host.
The complexes were refined by progressively minimizing their
potential energy: only the intercalator drug atoms were allowed
to move in the first place, and finally the whole systems were re-
laxed, using the Polak-Ribiere algorithm (conjugate gradient) until
an energy convergence criterion of 0.1 kcal molꢁ1 Aꢁ1 was reached.
To verify that the placement of the intercalator resulted in a
structure with a low energy minimum on the potential surface
for each complex, the intercalator was translated from its mini-
mized position by 100 pm along the principal axis of the chromo-
phore and the resulting structures were optimized using our
minimization protocol.
HT-29 (human colon adenocarcinoma), MCF-7 (human breast
adenocarcinoma), NCI-H460 (human non-small cell lung carci-
noma), and SF-268 (human glioblastoma) cell lines were cultured
in DMEM culture medium containing 10% (v/v) heat-inactivated
faetal bovine serum (FBS), 2 mM
L-glutamine, 100 U/ml penicillin,
and 100 g/ml streptomycin at 37 °C in a humidified atmosphere
l
of 5% CO2/95% air. Cells were periodically tested for Mycoplasma
infection and found to be negative.
3.4. Cell growth inhibition assay
The effect of the distinct compounds in the proliferation of hu-
man tumour cell lines was determined as previously described25
by using the XTT (sodium 3,3’-{1-[(phenylamino)carbonyl]-3,4-
tetrazolium}-bis(4-methoxy-6-nitro)benzenesulphonic acid hy-
drate) cell proliferation kit (Roche Molecular Biochemicals, Mann-
heim, Germany) according to the manufacturer’s instructions. Cells
The intercalation energies were calculated as the sum of the en-
ergy required for opening of the DNA sequence (
molecular interaction energy between the intercalator and DNA
Ec)
DEi) and the inter-
(D
Ebinding ¼ Ecomplex ꢁ ðEoligomere þ EcompoundÞ ¼
DEc
(1.5 ꢂ 103 in 100
lL) were incubated in DMEM culture medium
containing 10% heat-inactivated FBS, in the absence and in the
presence of the indicated compounds at a concentration range of
10ꢁ4 to 10ꢁ9 M, in 96-well flat-bottomed microtiter plates, and fol-
lowing 72 h of incubation at 37 °C in a humidified atmosphere of
air/CO2 (19/1) the XTT assay was performed. Measurements were
done in triplicate, and the IC50 (50% inhibitory concentration) va-
lue, defined as the drug concentration required to cause 50% inhi-
bition in the cellular proliferation with respect to the untreated
controls, was determined for each compound.
Solvation energy contributions were not included in binding
calculations.
3.6. DNA melting measures
Calf thymus DNA (ct-DNA) was purchased to Aldrich. The poly-
nucleotide was dissolved in sodium cacodylate buffer,
I = 0.05 mol dmꢁ3, pH 7.0. The calf thymus ct-DNA was additionally
sonicated and filtered through a 0.45 l
m filter.31 Polynucleotide
concentration was determined spectroscopically32 as the concen-
tration of base pairs (7.5 ꢂ 10ꢁ6 M). Electronic absorption spectra
were recorded with an Agilent 8453 UV–vis spectrophotometer
using (1 cm) quartz cuvettes.
3.5. Molecular modelling
Models were built by using the Hyperchem 7.5 (Hypercube
Inc.)26 package capabilities. Monointercalation was studied on tet-
ramer duplexes following the Newlin et al. methodology and look-
ing for specificity among the different base pair sequences.18,27
Treatment of bisintercalation was done in a similar way on du-
plexes of five and six base pairs, as the chain length of C and D al-
lows intercalation with one or two base pairs in between. We
selected the AMBER method28 modified by the inclusion of the
appropriate parameters.29 When available, the parameters came
from analogous parameters used in the literature. All others were
developed following Kölman and Hopfinger procedures.29,30 The
equilibrium bond length and angle values came from experimental
ones on reasonable reference compounds. Solvent effects were
simulated by the use of a distance-dependent dielectric function
Thermal melting curves for ct-DNA were determined by follow-
ing the absorption change at 260 nm as a function of temperature
of solutions in which increasing amounts of 5 and 6ꢀ2HCl were
added to ct-DNA.32 The absorbance of the ligands was subtracted
from every curve, and the absorbance scale was normalised. Tm
values are the midpoints of the transition curves, determined
either from the maximum of the first derivative or graphically,
by a tangent method.33
Tm of the free nucleic acid from Tm of the complex. Every
here reported was the average of at least three measurements; the
error in Tm was estimated to be 0.5 °C. In the two studied system
D
Tm values were calculated by subtracting
Tm value
D
D
(5-ct-DNA and 6-ct-DNA) molar ratios r = [drug]/[baseꢁpair] great-
er than 1.0 led to precipitation.
of
e = 4r in all calculations. In order to achieve electrical neutrality
an0 appropriate number of Na+ counterions were included, placed at
6 ÅA distance from each phosphate-oxygen bisector, being initially
energy-minimized using 500 and 2000 cycles of steepest descent
and conjugate gradients.
Acknowledgements
The authors thank the Spanish Comision Interministerial de
Ciencia y Tecnología for financial support given to this research
work. This work was supported by grants from the Spanish Minis-
try of Science and Innovation (SAF2008-02251, and RD06/0020/
1037 from Red Temática de Investigación Cooperativa en Cáncer,
Instituto de Salud Carlos III). We are also grateful to staff of the
C.N.Q.O. Manuel Lora-Tamayo (CSIC) for determining the NMR
spectra and performing elemental analyses.
Model intercalation sites were generated by separation of the
bases until 680 pm, and molecular mechanics energy minimization
was then used to relieve any steric distortion in the chains, follow-
ing the same procedure. Counterions positions were modified
accordingly. Distance restraints with
a force constant of
25 kcal molꢁ1 Aꢁ2 were applied to nucleotides affected during this
process, in order to consider the Watson-Crick hydrogen bonds,
and convergence was achieved at 0.1 kcal molꢁ1 Aꢁ1
.
References and notes
Models for each synthesized compound were constructed using
standard geometry and standard bond lengths. Atom potentials
were assigned according to parameters defined within the AMBER
force-field and lowest energy conformers were located by perform-
ing simulated annealing. These conformers were used for docking.
Then, each drug structure was manually docked into the tetramer
DNA duplex intercalation site, looking for optimal orientations by
1. (a) Denny, W. A.; Baguley, B. C.; Cain, B. F.; Waring, M. J. In Mechanism of Action
of Anticancer Drugs; Neidle, S., Waring, M. J., Eds.; McMillan: London, 1983; (b)
Chabner, B. A.; Allegra, C. J.; Curt, G. A.; Calabresi, P. Antineoplastic Agents. In
Goodman
& Gilman´s: The Pharmacological Basis of Therapeutics, 9th ed.;
Hardman, J. G.; Goodman Gilman, A.; Limbird, L. E., Eds.; Mac Graw-Hill:
New York, 1996, Chapter 51, p 1264.; (c) Braña, M. F.; Cacho, M.; Gradillas, A.;
Pascual-Teresa, B.; Ramos, A. Curr. Pharm. Des. 2001, 7, 1745.

J. Galisteo et al. / Bioorg. Med. Chem. 18 (2010) 5301–5309
5309
2. (a) Lerman, L. S. J. Mol. Biol. 1961, 3, 18; (b) Waring, M. J. J. Mol. Biol. 1970, 54, 247.
3. (a) Wiseman, L. R.; Spencer, C. M. Drugs Aging 1997, 10, 473; (b) Lown, J. W.
Anthracyclines and Anthracenedione-Based Anticancer Agents; Elsevier:
Amsterdam, 1988.
4. (a) Denny, W. A. In Cancer Chemotherapeutic Agents; Foye, W. O., Ed.; Am. Chem.
Soc.; Washington, DC, 1995, p 218.; (b) Baguley, B. C.; Wakelin, L. P. G.;
Jacintho, J. D.; Kovacik, P. Curr. Med. Chem. 2003, 10, 2643.
5. Wakelin, L. P. G. Med. Res. Rev. 1986, 6, 276.
19. Cholody, M. W.; Kosakowska-Cholody, T.; Holligshead, M.; Hariprakasha, H. K.;
Michejda, C. J. J. Med. Chem. 2005, 48, 4474.
20. Turpin, J. A.; Buckheit, R. W.; Derse, D.; Holligshead, M. W.; Palamone, C.;
Osterling, M. C.; Hill, S. A.; Graham, L.; Schaeffer, C. A.; Bu, M.; Huang, M.;
Cholody, W. M.; Michejda, C. J.; Rice, W. G. Antimicrob. Agents Chemother. 1998,
42, 487.
21. Molecular Aspects of Anticancer Drug Action; Neidle, S., Waring, M. J., Eds.;
Verlag-Chemie, 1983.
6. Wakelin, L. P. G.; Waring, M. J.. In Polyfunctional Intercalating Agents in
Comprehensive Medicinal Chemistry: The Rational Design, Mecanistic Study and
Therapeutic Applications of Chemical Compounds; Hansch, C., Sammes, P. G.,
Taylor, J. B., Eds.; Pergamon: Oxford, 1990; Vol. II,.
7. Antonini, I.; Polucci, P.; Magnano, A.; Gatto, B.; Palumbo, M.; Menta, E.; Pescalli,
N.; Martelli, S. Med. Chem. 2003, 46, 3109.
22. (a) Veal, J. M.; Li, Y.; Zimmerman, S. C.; Lambenon, C. R.; Cory, M.; Zon, G.;
Wilson, W. D. Biochemistry 1990, 29, 10918. and references cited therein; (b)
There are a variety of X-ray structures in the Protein Data Bank that show
bisintercalation with only one intervening base pair, that is: 1KCI (NDB ID
DD0048) or 1K9G (NBD 1D DD0047).
23. Bjorndal, M. T.; Fygenson, D. K. Biopolymers 2002, 65, 40.
24. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory
Chemicals; Pergamon Press: Oxford, 1980.
8. Dai, J.; Punchihewa, C.; Mistry, P.; Ooi, A. T.; Yang, D. J. Biol. Chem. 2004, 279,
46096.
9. Pindur, U.; Kim, Y. S.; Mehrabani, F. Curr. Med. Chem. 1999, 6, 29.
10. Filosa, R.; Peduto, A.; Di Micco, S.; De Capraris, P.; Festa, M.; Petrella, A.;
Capranico, G.; Bifulco, G. Bioorg. Med. Chem. 2009, 17, 13.
11. Gajewski, M. P.; Beall, H.; Schnieder, M.; Stranahan, S. M.; Mosher, M. D.; Rider,
K. C.; Natale, N. R. Bioorg. Med. Chem. Lett. 2009, 4067.
25. David-Cordonnier, M.-H.; Gajate, C.; Olmea, O.; Laine, W.; de la Iglesia-Vicente,
J.; Perez, C.; Cuevas, C.; Otero, G.; Manzanares, I.; Bailly, C.; Mollinedo, F. Chem.
Biol. 2005, 12, 1201.
26. HyperChem(TM) Professional 7.51, Hypercube, Inc., 1115 NW 4th Street,
Gainesville, Florida 32601, USA.
12. Antony, S.; Agama, K. K.; Miao, Z-H.; Hollingshead, M.; Holbeck, S. L.; Wright,
M. H.; Varticovski, L.; Nagarajan, M.; Morrell, A.; Cushman, M.; Pommier, Y.
Mol. Pharmacol. 2006, 70, 1109.
13. Campayo, L.; Navarro, P. Eur. J. Med. Chem. 1986, 21, 143.
14. Campayo, L.; Navarro, P.; Cano, F. H.; Foces-Foces, C. J. Chem. Soc., Perkin Trans. 2
1987, 569.
15. Pons, M.; Campayo, M.; Martinez-Balbas, M. A.; Azorin, F.; Navarro, P.; Giralt, E.
J. Med. Chem. 1991, 34, 82.
16. (a) Navarro, P.; Campayo, L.; Alvarez, I.; Escario, J. A. U.S. Patent 5,719.282.; (b)
Navarro, P.; Campayo, L.; Alvarez, I.; Escario, J. A. European Patent EP 0 440 787.
17. Perez, C.; Campayo, L.; Navarro, P.; Garcia-Bermejo, L.; Aller, P. Biochem.
Pharmacol. 1994, 48, 75.
27. Newlin, D. D.; Miller, K. J.; Pilch, D. F. Biopolymers 1984, 23, 139.
28. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M.; Ferguson, D. M.;
Spelmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc. 1995,
117, 5179.
29. Miranda, C.; Escartí, F.; Lamarque, L.; Yunta, M. J. R.; Navarro, P.; Garcia-España,
E.; Jimeno, M. L. J. Am. Chem. Soc. 2004, 126, 823. and references cited therein.
30. Hopfinger, A. J.; Pearlstein, R. A. J. Comput. Chem. 1994, 15, 1019.
31. Chaires, J. B.; Dattagupta, N.; Crothers, D. M. Biochemistry 1982, 21, 3933.
32. Palm, B. S.; Piantanida, I.; Zinic, M.; Schneider, H.-J. J. Chem. Soc., Perkin Trans 2
2000, 385.
33. Wilson, W. D.; Tanious, F. A.; Fernandez-Saiz, M.; Rigl, C. T. In Evaluation of
Drug/Nucleic Acid Interactions by Thermal Melting Curves from: Methods in
Molecular Biology; Fox, K. R., Ed.; Drug–DNA Interaction Protocols; Humana
Press: Totowa, NY, 1998; Vol. 90,.
18. Rodriguez-Ciria, M. M.; Sanz, A. M.; Yunta, M. J. R.; Gomez-Contreras, F.;
Navarro, P.; Fernandez, I.; Pardo, M.; Cano, M. C. Bioorg. Med. Chem. 2003, 11,
2143.