Benzoyl and cinnamoyl nitrogen mustard derivatives of benzoheterocyclic analogues of the tallimustine: Synthesis and antitumour activity

Article abstract of DOI:10.1016/S0968-0896(01)00425-4

A series of benzoyl and cinnamoyl nitrogen mustards tethered to different benzoheterocycles and to oligopyrroles structurally related to netropsin consisting of two pyrrole-amide units and terminating with an amidine moiety have been synthesised and a str

Full text of DOI:10.1016/S0968-0896(01)00425-4

Bioorganic & Medicinal Chemistry 10 (2002) 1611–1618
Benzoyl and Cinnamoyl Nitrogen Mustard Derivatives of
Benzoheterocyclic Analogues of the Tallimustine:
Synthesis and Antitumour Activity
Pier Giovanni Baraldi,a,* Romeo Romagnoli,a Maria Giovanna Pavani,a
Maria del Carmen Nunez,a John P. Binghamb and John A. Hartleyb
a
`
Dipartimento di Scienze Farmaceutiche, Universita di Ferrara, Via Fossato di Mortara 17/19, 44100 Ferrara, Italy
bCRC Drug-DNA Interactions Research Group, Department of Oncology, Royal Free and University College Medical School,
91 Riding House Street, London W1W 7BS, UK
Received 9October 2001; accepted 30 November 2001
Abstract—A series of benzoyl and cinnamoyl nitrogen mustards tethered to different benzoheterocycles and to oligopyrroles
structurally related to netropsin consisting of two pyrrole-amide units and terminating with an amidine moiety have been synthe-
sised and a structure–activity relationship determined. Derivatives 3–10 have been evaluated for their sequence selective alkylating
properties and cytotoxicity against human K562 leukaemia cells. They are 2- to 50-fold less cytotoxic than tallimustine, with com-
pound 8 being the most potent member of this series. Among tallimustine isosters, the compounds with an indole 3 or benzothio-
phene 6 are 4-fold less cytotoxic than tallimustine, while the compounds with an N-methyl indole or benzofuran showed a 7- and
14-fold reduced cytotoxic potency, respectively. Our preliminary results indicate that these derivatives preferentially bind to AT-
rich sequence with a sequence selectivity similar to tallimustine. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
yet fully elucidated, may be due to its ability to inhibit
Minor groove alkylating agents are a relatively new
class of anticancer agents reported to possess high
cytotoxic activity in in vitro and in vivo preclinical
models.1 Tallimustine 1,2 a benzoic acid nitrogen mus-
tard (BAM) derivative of distamycin A, is a minor
groove alkylating agent which binds selectively to AT-
rich sequence of DNA, showing excellent activity
against a series of murine transplanted solid tumours
and human xenographs3 and it has proven active in a
wide range of tumour cell lines resistant to alkylating
agents.2,4 Tallimustine retains the AT preference of dis-
tamycin A and appears to possess a high preference for
alkylation of 30-adenine-N3 located in 50-TTTTPuA
sequences in the minor groove of DNA.5 The mechan-
ism by which the alkylation induced by tallimustine
leads to cytotoxicity remains unclear. While the cyto-
toxicity of BAM is related to the ability to form inter-
strand cross-links in DNA6 with consequent inhibition
of DNA replication and transcription; the mechanism
of antitumour action of tallimustine, although it is not
the binding of some transcription factors to their con-
sensus sequences in DNA, thereby preventing tran-
scription.5,7
Scheme 1. H2, 10% Pd/C, MeOH; b.p-[bis-(2-chloroethyl)amino]-
benzoyl chloride, TEA, dioxane; c. KOH 2 N, dioxane, rt; d. SOCl2,
dioxane; e. 15, NaHCO3, dioxane/water (1/4, v/v).
*Corresponding author. Tel.:+39-0532-291293; fax:+39-0532-
291296; e-mail: pgb@dns.unife.it
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00425-4

1612
P. G. Baraldi et al. / Bioorg. Med. Chem. 10 (2002) 1611–1618
Scheme 2. SOCl2, 1,4-dioxane, reflux, 2 h; b. 15, dioxane/water (1/4, v/v) 18 h, rt; c. H2, 10% Pd/C, 2–3 drops of 10% HCl, MeOH/water; d. 24,
NaHCO3, DMF, 18 h, r.t.; e. HOBT, DCC, DMF, 70 ꢀC, 4 h.
Unfortunately, tallimustine showed a severe myelotox-
icity that probably impaired the reaching of effective
therapeutic doses and its Phase II clinical development
was discontinued.8
Following these results, in the present study we report
the synthesis and the biological evaluation of novel iso-
sters of tallimustine and 2, showing the substitution of
the N-terminal pyrrole ring tethered to benzoyl and
cinnamoyl nitrogen mustard alkylating moiety by dif-
ferent benzoheterocycles such as indole, N-methyl
indole, benzothiopene and benzofuran, to give the cor-
responding derivatives 3–6 and 7–10, respectively. The
compounds 7–10 are vinylogues of 3–6, respectively,
and are characterised by an increased distance between
the alkylating moiety and the DNA–polypyrrolic bind-
ing frame via the vinylic double bond. Since pyrrole is
susceptible to oxidative breakdown,10 these new deriv-
atives have been prepared as potentially more stable
minor groove binders aimed to improved the relative
instability of the polypyrrolic skeleton. The choice of
these benzoheterocycles can be also justified due to their
importance in increasing the binding affinity to DNA
and the selectivity of alkylation of CC-1065 analogues
such as U-71, 184, adozelesin and bizelesin.11
Nevertheless, tallimustine has represented an important
model for the design of new cytotoxic minor groove
binders derived from distamycin, where the formyl
group has been replaced by moieties of mild chemical
reactivity with the DNA. Among these, a cinnamoyl
nitrogen mustard derivatives of distamycin A (PNU-
157911, 2)9 a vinylogue of tallimustine, has been dis-
closed by Pharmacia & Upjohn and shows very good
antileukemic activity, significantly superior to that of
tallimustine (IC50=7.2 ng/mL compared with 50.3 ng/
mL for tallimustine against L1210 leukemia). Com-
pound 2 appears significantly more cytotoxic than talli-
mustine in accordance with its increased chemical
reactivity.9 For 2, as in the case of tallimustine, the
cytotoxic activity is the result of the combination of
two moieties, which are per se substantially inactive as
cytotoxics.

P. G. Baraldi et al. / Bioorg. Med. Chem. 10 (2002) 1611–1618
1613
amines in high yield and purity, which were unstable
and therefore were immediately coupled with the active
ester 24 of cynnamic acid mustard 259 to give the
derivatives 7–10. The compound 24 is readily synthe-
sised from equimolar amounts of the cynnamic acid
mustard 25 and 1-hydroxy-1,2,3-benzotriazole (HOBT)
by employing N,N0-dicyclohexylcarbodiimide (DCC) as
coupling reagent.
Chemistry
The synthetic route followed for the synthesis of deri-
vatives 3–6 is outlined in Scheme 1. The key step was
the coupling between the acid chlorides of benzohetero-
cyclic carboxylic acids 11–1312 and 14 bearing the BAM
moiety and the known amino-amidine 15.9,13 These
condensations were performed at room temperature and
with identical reaction times (18 h), in a water/dioxane
mixture containing sodium bicarbonate (NaHCO3) as
base. Compounds 3–6 were obtained in acceptable
yields (50–65%), after purification by silica gel flash-
chromatography.
Biological Evaluation
Tallimustine and all synthesised compounds 3–10 were
evaluated in vitro for their cytotoxic activities against
human K562 leukemia cells using the MTT assay. The
results, expressed as IC50 values (the concentration of
test agent inhibiting growth by 50%), are reported in
Table 1.
The synthetic approach employed for the preparation of
derivatives 7–10 is shown in Scheme 2. Starting from the
amino-amidine 15,13 it was condensed with the acid
chlorides of the well-known nitro-acids 16–19.12 The
corresponding nitro-amidines 20–23 so obtained, after
catalytic hydrogenation with 10% Pd/C at room tem-
perature, were transformed into the corresponding
None of the compounds 3–10 was more active than tal-
limustine, although some had similar activity. Among
the derivatives, compounds 3, 6, 8 and 9 proved to be
only slightly less potent than tallimustine, with IC50
values of 8.23, 9.00, 3.88 and 8.58 mM, respectively,
compared to 2.38 mM for tallimustine. In contrast,
compound 7, which contains the benzofuran tethered to
the cinnamoyl nitrogen mustard, is clearly much less
active and showed a very low level of cytotoxic potency
(IC50>100 mM).
Table 1. In vitro activities against K562 human leukemia of the tal-
limustine and compounds 3–10
Compd
In vitro IC50 (mM)
Tallimustine
3
4
5
6
7
8
9
10
2.38ꢁ0.5
8.23ꢁ4.08
15.5ꢁ0.5
31.96ꢁ15.08
9.0ꢁ4.51
>100
3.88ꢁ2.1
8.58ꢁ4.41
27.0ꢁ13
Comparing tallimustine with the derivatives 3–6 which
share the BAM moiety, the derivatives 3 and 6, in which
the pyrrole closest to the BAM moiety has been
replaced by indole and benzofuran, respectively, are
approximately 4 times less cytotoxic than tallimustine,
while compounds containing a N-methyl indole (4) or
benzothiopene (5) are 7 and 15 times less potent.
IC50=50% inhibitory concentration represents the mean from dose–
response curves of at least three experiments.
Figure 1. (a) DNAse I footprinting of compounds 4, 6, and 8 on a fragment of plasmid pUC18 DNA. Lane C is control, non-DNAse I treated
DNA. Lane 1 is DNAse I alone; lanes 2–4 are DNA treated with 8 at 0.01, 0.1 and 1 mM, respectively; lanes 5–7 are treated with 6 at 0.01, 0.1 and 1
mM, respectively; lanes 8–10 are treated with 4 at 0.01, 0.1 and 1 mM, respectively. The two major sites of footprinting are indicated. (b) Covalent
modification by compounds 4, 6, and 8 on a fragment of plasmid pUC18 DNA measured using the Taq polimerase stop assay. Lane C is control
DNA; lanes 1–4 are DNA treated with tallimustine, 4, 6 and 8 at 10 mM. The major site of alkylation at the sequence 50-TTTTGA is indicated.

1614
P. G. Baraldi et al. / Bioorg. Med. Chem. 10 (2002) 1611–1618
1
Comparing compounds with the same oligopeptidic
frame, for the compounds 3 versus 7 (containing indole)
and 6 versus 10 (containing benzofuran), the introduc-
tion of a vinylic double bond between the phenyl nitro-
gen mustard and the carboxamido tethered to the
benzoheterocycle, a decrease in cytotoxicity of >12-
and 3-fold, respectively, was observed. In contrast, as
was observed with tallimustine and 2, for derivatives 4
versus 8 (containing N-methyl indole) and 5 versus 9
(containing benzothiophene), the substitution with cin-
namoyl nitrogen mustard led to an increase in the cyto-
toxicity of 4-fold.
plates) and visualized with aqueous KMnO4. HNMR
spectra were obtained in DMSO solutions with a Bruker
AC 200 spectrometer. Chemical shifts (d) are given in
parts per million (ppm) upfield from tetramethylsilane.
All products reported showed 1H NMR spectra in
agreement with the assigned structures. Matrix-assisted
laser desorbtion ionization time-of-flight (MALDI-
TOF) mass spectrometry of all synthesised compounds
was conducted using a Hewlett Packard G 2025 A LD-
TOF instrument. The samples were analysed in the lin-
ear mode with 28 kV accelerating voltage, mixing them
with a saturated solution of a-cyano-4-idroxycinnamic
acid matrix. Melting points (mp) were determined on a
Buchi–Tottoli apparatus and are uncorrected. Ele-
mental analyses were conducted by the Mycroanalytical
Laboratory of the Chemistry Department of the Uni-
versity of Ferrara. Column chromatography was carried
out using Merck silica gel (230–240 mesh). All com-
pounds obtained commercially were used without fur-
ther purification. Organic solutions were dried over
anhydrous Na2SO4. Dioxane was distilled from calcium
hydride and anhydrous DMF was distilled from calcium
chloride and stored over molecular sieves (3 A). In high-
pressure hydrogenation experiments, a Parr shaker on a
high-pressure autoclave was used.
DNA sequence-specific non-covalent interaction of
compounds 3–10 was evaluated by Dnase I footprinting
experiments. Figure 1a shows a representative of such
experiments.
All the compounds 3–10 showed evidence of non-cova-
lent binding by footprinting to two AT-rich sequences
in the minor groove (AAATAA and TATAT), which
were also sites of non covalent binding for distamycin
and tallimustine.
Sequence-specific DNA alkylation produced by the
compounds 3–10 tested was evaluated using the Taq
polymerase stop assay (Fig. 1b). As can be seen from
Figure 1b, at the concentration of 10 mM tallimustine
showed sequence specific alkylation, with preferential
alkyation at a 50-TTTTGA sequence. Similar alkylation
was also observed for several of the compounds, but
with lower intensity compared to tallimustine. The least
active compound 7 did not give any evidence of alkyl-
ation in the same DNA sequence at 100 mM, and the
fact that this compound is 50 times less cytotoxic than
tallimustine is likely to result from its reduced capacity
to interact with DNA. For the other compounds in the
series, alkylation was evident within this DNA sequence
and the sequence specificity observed with tallimustine
was preserved.
Synthesis of 5-[4-N,N-bis(2-chloroethyl)aminobenzene-1-
carboxamido]benzothiophene-2-carboxylic acid (14). A
solution of methyl 5-nitrobenzothiophene-2-carboxy-
late14 (3 mmol, 711 mg) in methanol (15 mL) was
hydrogenated over 10% Pd/C at 50 psi. After 2 h, the
catalyst was removed by filtration through a bed of
Celite and the filtrate concentrated in vacuum. The
resulting foam solid (615 mg, 95% of yield) was used
without any purification for the next reaction. 1H NMR
(CDCl3) d 3.44 (bs, 2H), 4.00 (s, 3H), 7.98 (d, J=9Hz,
1H), 8.38 (dd, J=9and 2 Hz, 2H), 8.77 (d, J=1.6 Hz,
1H). FAB-MS (MALDI-TOF): 208.4 [M+1]+.
p-N,N-bis(2-Chloroethyl)aminobenzoyl chloride15 (603
mg, 2.3 mmol) dissolved in dry dioxane (5 mL) was
added dropwise at room temperature to a stirring solu-
tion of methyl 5-aminobenzothiophene-2-carboxylate
(435 mg, 2.1 mmol) and dry triethylamine (0.32 mL, 2.3
mmol) in dry dioxane (10 mL). The reaction mixture
was allowed to stir overnight. After TLC analysis indi-
cated that all of the starting material had disappeared,
the reaction mixture was concentrated under reduced
pressure to a brown solid, which was dissolved in
EtOAc and washed with water twice (10 mL each). The
organic layer was dried (Na2SO4), concentrated and the
resulting pure product was precipitated from EtOAc/
hexane, to give methyl 5-[4-N,N-bis(2-chloro-
ethyl)aminobenzene-1-carboxamido]benzothiophene-2-
carboxylate as a brown powder which was dried in
vacuum at room temperature. Yield: 430 mg (85%); mp
In conclusion, for both tallimustine and 2, replacement
of the N-terminal pyrrole by a benzoheterocycle does
not impart any improvement in terms of cytotoxicity
activity and DNA-binding capability. No clear-cut
structure–activity relationship can be observed and it is
difficult to find common physico-chemical features both
among active and inactive reported compounds. It is
noted that other factors such as low penetration into
cell, cellular distribution and metabolic deactivation
may also influence the cytoxicity result, but they are not
assessed in the present study.
Experimental
Chemistry
200 ꢀC; H NMR (CDCl3) d 3.78 (m, 8H), 3.94 (s, 3H),
1
6.72 (d, J=7.8 Hz, 2H), 7.59(d, J=9Hz, 1H), 7.82 (t,
J=8 Hz, 3H), 7.99 (s, 1H), 8.35 (s, 1H), 10.1 (s, 1H).
FAB-MS (MALDI-TOF): 451.2 [M+1]+.
All reactions were carried out under Argon atmosphere,
unless otherwise described. Standard syringe techniques
were applied for transferring anhydrous solvents. Reac-
tion courses and product mixtures were routinely mon-
itored by TLC on silica gel (precoated F254 Merck
To a well-stirred solution of methyl 5-[4-N,N-bis(2-chlor-
oethyl)aminobenzene-1-carboxamido]benzothiophene-2-

P. G. Baraldi et al. / Bioorg. Med. Chem. 10 (2002) 1611–1618
1615
carboxylate (370 mg, 0.82 mmol) in 5 mL of dioxane was
treated with 2 N KOH in water (2 mL) and stirred at
room temperature for 1 h. The clear solution was eva-
porated to remove dioxane, diluted with water (5 mL),
cooled on an ice-water bath and acidified with aqueous
hydrochloric acid to pH 2. The aqueous suspension was
extracted with ethyl acetate (2ꢂ10 mL) and the organic
layers were combined, dried (Na2SO4) and con-
centrated. The resulting residue was precipitated from
ethyl acetate with hexane to give the product 14 as a
added to the amino-amidine 15 (200 mg, 0.5 mmol),
after work-up, the residue was purified by flash chro-
matography using methanol/methylene chloride 1/9
(v/v) as eluant. The resulting oil, precipitated by a mix-
ture of methanol/ethyl ether 1/20 (v/v), yielded the
compound 4 as a brown solid (256 mg, 65% yield); mp
193–195 ꢀC; H NMR (DMSO-d6) d 2.66 (m, 2H), 3.66
1
(m, 2H), 3.78 (m, 8H), 3.81 (s, 3H), 3.88 (s, 3H), 4.02 (s,
3H), 6.86 (d, J=8.8 Hz, 2H), 6.96 (s, 1H), 7.12 (s, 1H),
7.22 (s, 2H), 7.35 (s, 1H), 7.54 (d, J=6 Hz, 2H), 7.91 (d,
J=8.8 Hz, 2H), 8.17 (s, 1H), 8.23 (t, J=5.6 Hz, 1H),
8.70 (bs., 2H), 9.04 (bs., 2H), 9.93 (s, 1H), 10.01 (s, 1H),
10.45 (s, 1H). FAB-MS (MALDI-TOF): 748.6
[M+1]+. Anal. calcd for C36H41Cl3N10O4: C, 55.14; H,
5.25; Cl, 13.56; N, 17.86. Found: C, 55.03; H, 5.11; Cl,
13.45; N, 17.77.
brown powder. Yield: 330 mg (92%). mp>300 ꢀC. H
1
NMR (DMSO-d6): d 1H NMR (CDCl3) d 3.80 (m, 8H),
6.86 (d, J=8.6 Hz, 2H), 7.94 (m, 4H), 8.09 (s, 1H), 8.51
(s, 1H), 10.17 (s, 1H), 13.03 (bs, 1H). FAB-MS
(MALDI-TOF): 436.2 [M+1]+.
General procedure for the synthesis of the acyl chlorides
of the nitro-acids (11–14). A suspension of 11–14 (1
mmol) in dry dioxane (3 mL) containing SOCl2 (0.5
mL) was boiled under reflux for 2 h. The solvent was
evaporated and the residue was used without any pur-
ification for the next reaction.
3-[1-Methyl-4[1-methyl-4[5-[4-N,N-bis(2-chloroethyl)ami-
nobenzene-1-carboxamido] benzofuran-2-carboxamido]-
pyrrole-2-carboxamido]pyrrole-2-carboxamido]propion-
amidine hydrochloride (5). Following the general
procedure, starting from 13 (105 mg, 0.25 mmol), it was
transformed in the corresponding acyl chloride and
added to the amino-amidine 15 (100 mg, 0.25 mmol),
after workup, the residue was purified by flash chroma-
tography using methanol/methylene chloride 2/8 (v/v)
as eluant. The resulting oil, precipitated by a mixture
methanol/ethyl ether 1/20 (v/v), yielded the compound 5
as a white solid (100 mg, 52% yield); mp 205–210 ꢀC; 1H
NMR (DMSO-d6) d 2.63 (m, 2H), 3.48 (m, 2H), 3.82
(m, 8H), 3.81 (s, 3H), 3.82 (s, 3H), 6.86 (d, J=9Hz,
2H), 6.96 (s, 1H), 7.15 (s, 1H), 7.21 (s, 2H), 7.34 (s, 1H),
7.68 (m, 3H), 7.90 (d, J=8.6 Hz, 2H), 8.29(s, 1H), 8.61
(bs, 2H), 8.98 (bs, 2H), 10.03 (s, 1H), 10.09 (s, 1H),
10.73 (s, 1H). FAB-MS (MALDI-TOF): 735.6
[M+1]+. Anal. calcd for C35H38Cl3N9O5: C, 54.52; H,
4.97; Cl, 13.79; N, 16.35. Found: C, 54.41; H, 4.78; Cl,
13.64; N, 16.25.
General procedure for the synthesis of compounds (3–6).
To a cooled solution of the amino-amidine 15 (1 mmol),
in a mixture of water–dioxane (10 mL, 1/4 v/v) con-
taining NaHCO3 (2 mmol), the acyl chloride of the
compounds 11–14 (1 mmol) in dioxane (3 mL) was
added dropwise. After 2 h at 0 ꢀC, the reaction mixture
was allowed to rise to room temperature and stirred for
18 h. After this time, the solution was carefully acidified
(pH 1–2) with 10% HCl in water, evaporated, and the
residue purified by silica gel column chromatography
[eluant dichloromethane (DCM)/MeOH 9.5/0.5 then 9/1)].
3-[1-Methyl-4[1-methyl-4[5-[4- N,N-bis(2-chloroethyl)-
aminobenzene - 1- carboxamido] indole-2-carboxamido]-
pyrrole-2-carboxamido]pyrrole-2-carboxamido]propiona-
midine hydrochloride (3). Following the general
procedure, starting from 11 (100 mg, 0.24 mmol), it was
transformed in the corresponding acyl chloride and
added to the amino-amidine 15 (100 mg, 0.24 mmol),
after workup, the residue was purified by flash chroma-
tography using methanol–methylene chloride 1/9(v/v) as
eluant. The resulting oil, precipitated by a mixture
methanol/ethyl ether 1/20 (v/v), yielded the compound 3
as a brown solid (143 mg, 77% yield); mp 200–203 ꢀC; 1H
NMR (DMSO-d6) d 2.64 (m, 2H), 3.49(m, 2H), 3.82 (m,
8H), 3.81 (s, 3H), 3.88 (s, 3H), 6.85 (d, J=9Hz, 2H), 6.96
(s, 1H), 7.10 (s, 1H), 7.42 (m, 5H), 7.91 (d, J=8.6 Hz,
2H), 8.11 (s, 1H), 8.29(t, J=5.6 Hz, 1H), 8.77 (bs, 2H),
9.08 (bs, 2H), 9.89 (s, 1H), 10.05 (s, 1H), 10.49 (s, 1H),
11.70 (s, 1H). FAB-MS (MALDI-TOF): 734.6 [M+1]+.
Anal. calcd. for C35H39Cl3N10O4: C, 54.59; H, 5.10; Cl,
13.81; N, 18.19. Found: C, 54.48; H, 5.02; Cl, 13.68; N,
18.04.
3-[1-Methyl-4[1-methyl-4[5-[4-N,N-bis(2-chloroethyl)ami-
nobenzene-1-carboxamido] benzothiophene-2-carboxami-
do]pyrrole-2-carboxamido]pyrrole-2-carboxamido] pro-
pionamidine hydrochloride (6). Following the general
procedure, starting from 14 (180 mg, 0.41 mmol), it was
transformed in the corresponding acyl chloride and
added to the amino-amidine 15 (160 mg, 0.41 mmol),
after workup, the residue was purified by flash chroma-
tography using methanol/methylene chloride 2/8 (v/v)
as eluant. The resulting oil, precipitated by a mixture
methanol/ethyl ether 1/15 (v/v), yielded the compound 6
as a brown solid (186 mg, 58% yield); mp 188–190 ꢀC;
1H NMR (DMSO-d6) d 2.66 (m, 2H), 3.36 (m, 2H), 3.78
(m, 8H), 3.81 (s, 3H), 3.87 (s, 3H), 6.86 (d, J=9Hz, 2H),
6.96 (s, 1H), 7.13 (s, 1H), 7.23 (s, 1H), 7.35 (s, 1H), 7.94 (m,
4H), 8.29(s, 1H), 8.50 (s, 1H), 8.73 (bs, 2H), 9.05 (bs, 2H),
10.04 (s, 1H), 10.17 (s, 1H), 10.78 (s, 1H). FAB-MS
(MALDI-TOF): 751.5 [M+1]+. Anal. calcd for
C35H38Cl3N9O4S: C, 53.40; H, 4.87; Cl, 13.51; N, 16.01; S,
4.07: Found: C, 53.26; H, 4.75; Cl, 13.38; N, 15.89 ; S, 3.9 5.
3-[1-Methyl-4[1-methyl-4[1-methyl-5-[4-N,N-bis(2-chloro-
ethyl)aminobenzene-1-carboxamido]indole-2-carboxami-
do]pyrrole-2-carboxamido]pyrrole-2-carboxamido] pro-
pionamidine hydrochloride (4). Following the general
procedure, starting from 12 (260 mg, 0.6 mmol), it was
transformed in the corresponding acyl chloride and
General procedure for the synthesis of compounds (20–
23). The nitroacid 16–19 (2 mmol) was refluxed in a
mixture of dioxane (5 mL) and thionyl chloride (4 mL)

1616
P. G. Baraldi et al. / Bioorg. Med. Chem. 10 (2002) 1611–1618
of for 1 h. The reaction mixture was concentrated in
vacuo and the resulting residue was coevaporated with
toluene. The resulting acid chloride was used without
further purification.
6.37; N, 20.12. Found: C, 51.58; H, 4.42; Cl, 6.21; N,
22.01.
3-[1-Methyl-4[1-methyl-4[5-nitro-benzothiophene-2-car-
boxamido]pyrrole-2-carboxamido] pyrrole-2-carboxami-
do]propionamidine hydrochloride (23). Following the
general procedure, starting from the nitro-acid 19 (223
mg, 1 mmol), it was transformed into the corresponding
acyl chloride and coupled with the amino-amidine 15
(404 mg, 1 mmol). After work-up, the nitro-amidine 23
was obtained as a yellow solid (311 mg, 54% yield); mp
To a cooled solution of 15 (2 mmol) and NaHCO3 (336
mg, 4 mmol) in a mixture of water/dioxane (10 mL, 1/4
v/v), the above acid chloride in dioxane (5 mL) was
added dropwise. After 2 h at 0 ꢀC, the reaction mixture
was allowed to rise to room temperature and stirred for
18 h. Solvent was evaporated and the residue purified by
column chromatography using methylene chloride/
methanol as eluant (9/1).
210–212 ꢀC; H NMR (DMSO-d6) d 2.74 (m, 2H), 3.50
1
(m, 2H), 3.82 (s, 3H), 3.88 (s, 3H), 6.97 (s, 1H), 7.15 (s,
1H), 7.23 (s, 1H), 7.36 (s, 1H), 8.31 (m, 2H), 8.52 (s, 1H),
8.11 (t, J=7.2 Hz, 1H), 8.69 (bs, 2H), 8.95 (s, 1H), 9.03
(bs, 2H), 10.04 (s, 1H), 11.05 (s, 1H). FAB-MS (MALDI-
TOF): 538.6 [M+1]+. Anal. calcd for C24H25ClN8O5S:
C, 50.30; H, 4.40; Cl, 6.19; N, 19.55; S, 5.60. Found: C,
50.14; H, 4.28; Cl, 6.02; N, 22.01; S, 5.35.
3-[1-Methyl-4[1-methyl-4[5-nitro-indole-2-carboxamido]-
pyrrole-2-carboxamido]pyrrole-2-carboxamido]propiona-
midine hydrochloride (20). Following the general
procedure, starting from the nitro-acid 16 (206 mg, 1
mmol), it was transformed into the corresponding acyl
chloride and coupled with the amino-amidine 15 (404
mg, 1 mmol). After workup, the nitro-amidine 20 was
obtained ꢀ as a yellow solid (358 mg, 67% yield);
mp>300 C; 1H NMR (DMSO-d6) d 2.63 (m, 2H), 3.49
(m, 2H), 3.82 (s, 3H), 3.88 (s, 3H), 6.96 (s, 1H), 7.11 (s,
1H), 7.22 (s, 1H), 7.38 (s, 1H), 7.62 (s, 2H), 8.08 (d,
J=8.8 Hz, 1H), 8.26 (s, 1H), 8.67 (bs, 2H), 8.75 (s, 1H),
9.02 (bs, 2H), 10.04 (s, 1H), 10.75 (s, 1H), 12.48 (s, 1H).
FAB-MS (MALDI-TOF): 521.5 [M+1]+. Anal. calcd
for C24H26ClN9O5: C, 51.85; H, 4.71; Cl, 6.38; N, 22.67;
Found: C, 51.74; H, 4.61; Cl, 6.23; N, 22.54.
Synthesis of 1-hydroxybenzotriazolyl-N,N-bis(2-chloro-
ethyl)aminocinnamoate (24). A solution of 25 (2 mmol,
596 mg), HOBt (2 mmol, 270 mg), DCC (2 mmol, 412
mg) in 10 mL of dry DMF was heated at 70 ꢀC for 4 h.
After this time, the mixture was filtered, removing the
DCU formed. The filtered containing 24 was added
dropwise to the appropriate amine.
General procedure for the synthesis of compounds (7–
10). A solution of 20–23 (2 mmol) in 15 mL of a mix-
ture MeOH/water (4:1 v/v) containing a few drops of
aqueous 10% HCl was hydrogenated over 100 mg of
10% Pd/C at 50 psi for 3 h. The catalyst was removed by
filtration, the filtrate was concentrated to give a residue
which was used without purification for the next step.
3-[1-Methyl-4[1-methyl-4[1-methyl-5-nitro-indole-2-car-
boxamido]pyrrole-2-carboxamido] pyrrole-2-carboxami-
do]propionamidine hydrochloride (21). Following the
general procedure, starting from the nitro-acid 17 (216
mg, 1 mmol), it was transformed into the corresponding
acyl chloride and coupled with the amino-amidine 15
(404 mg, 1 mmol). After workup, the nitro-amidine 21
was obtained as a yellow solid (283 mg, 53% yield); mp
The crude amine above was prepared was dissolved in
anhydrous DMF (7 mL) containing NaHCO3 (2 mmol,
168 mg) and at this solution was added 24 (808 mg, 2
mmol) dissolved in 5 mL of dry DMF. The reaction
mixture was stirred at room temperature for 12 h, and
the evaporation of the DMF under reduced pressure
gave a solid which was purified by column chromato-
graphy using methylene chloride/MeOH as eluant (9/1
and then 8/2 v/v).
167–170 ꢀC; H NMR (DMSO-d6) d 2.66 (m, 2H), 3.42
1
(m, 2H), 3.54 (s, 3H), 3.76 (s, 3H), 4.08 (s, 3H), 6.97 (s,
1H), 7.01 (s, 1H), 7.13 (s, 1H), 7.22 (s, 1H), 7.36 (s, 1H),
7.51 (s, 1H), 7.97 (m, 1H), 8.12 (m, 1H), 8.28 (t, J=7.2
Hz, 1H), 8.72 (bs, 2H), 9.06 (bs, 2H), 10.04 (s, 1H), 10.78
(s, 1H). FAB-MS (MALDI-TOF): 535.5 [M+1]+. Anal.
calcd for C25H28ClN9O5: C, 52.68; H, 4.95; Cl, 6.22; N,
22.12. Found: C, 52.53; H, 4.78; Cl, 6.11; N, 22.01.
3-[1-Methyl-4[1-methyl-4[5-[4- N,N-bis(2-chloroethyl)-
aminocinnamoyl]aminoindole-2-carboxamido]pyrrole-2-car-
boxamido]pyrrole-2-carboxamido]propionamidine hydro-
chloride (7). Following the general procedure, starting
from 20 (273 mg, 0.5 mmol), after catalytic hydrogen-
ation, this compound was coupled-to 24 (202 mg, 0.5
mmol). The usual workup furnished the compound 7 as
3-[1-Methyl-4[1-methyl-4[5-nitro-benzofuran-2-carboxa-
mido]pyrrole-2-carboxamido]
pyrrole-2-carboxamido]-
propionamidine hydrochloride (22). Following the
general procedure, starting from the nitro-acid 18 (207
mg, 1 mmol), it was transformed into the corresponding
acyl chloride and coupled with the amino-amidine 15
(404 mg, 1 mmol). After work-up, the nitro-amidine 22
was obtained as a yellow solid (243 mg, 46% yield); mp
a brown powder (386 mg, 48.6%). mp 202–205 ꢀC; H
1
NMR (DMSO-d6) d 2.68 (m, 2H), 3.43 (m, 2H), 3.76 (m,
8H), 3.77 (s, 3H), 3.83 (s, 3H), 6.36 (d, J=15.4 Hz, 1H),
6.65 (s, 1H), 6.78 (d, J=8.4 Hz, 2H), 6.91 (s, 1H), 7.05 (s,
1H), 7.12 (s, 1H), 7.28 (s, 1H), 7.48 (m, 5H), 7.75 (d,
J=15.4 Hz, 1H), 8.02 (s, 1H), 8.08 (s, 1H), 8.34 (bs, 2H),
8.68 (bs, 2H), 10.04 (s, 1H), 10.75 (s, 1H), 12.48 (s, 1H).
FAB-MS (MALDI-TOF): 760.7 [M+1]+. Anal. calcd
for C37H41Cl3N10O4: C, 55.82; H, 5.19; Cl, 13.36; N,
17.59. Found: C, 55.67; H, 5.04; Cl, 13.17; N, 17.43.
188–190 ꢀC; H NMR (DMSO-d6) d 2.62 (m, 2H), 3.49
1
(m, 2H), 3.82 (s, 3H), 3.89(s, 3H), 6.96 (s, 1H), 7.11 (s,
1H), 7.22 (s, 1H), 7.38 (s, 1H), 7.62 (s, 1H), 8.07 (m,
1H), 8.11 (t, J=7.2 Hz, 1H), 8.26 (s, 1H), 8.68 (bs,
2H), 8.75 (s, 1H), 9.02 (bs, 2H), 10.04 (s, 1H), 10.78
(s, 1H). FAB-MS (MALDI-TOF): 522.5 [M+1]+.
Anal. calcd for C24H25ClN8O6: C, 51.76; H, 4.52; Cl,

P. G. Baraldi et al. / Bioorg. Med. Chem. 10 (2002) 1611–1618
1617
3-[1-Methyl-4[1-methyl-4[1-methyl-5-[4-N,N-bis(2-chloro-
ethyl)aminocinnamoyl]amino indole-2-carboxamido]pyrrole
- 2 - carboxamido] pyrrole - 2 - carboxamido]propionamidine
hydrochloride (8). Following the general procedure,
starting from 21 (267 mg, 0.5 mmol), after catalytic
hydrogenation, this compound was coupled to 24 (202
mg, 0.5 mmol). The usual workup furnished the com-
pound 8 as a yellow powder (310 mg, 48%); mp>217–
produce the target 447bp Human Topoisomerase IIa
promotor region for footprinting. Amplification was
carried out in a final volume of 50 ml containing 4 pmols
of both the labelled primer and ‘complementary’ primer
(50 ctgtccagaaagccggcactcag), 10 ng pCAT557 plasmid
(Promega) containing the topo2a promotor (10 ng/ml)
(courtesy of Dr. B. Tolner), 120 mM of each dNTP, 2.5
mM MgCl2, 10 mM Tris–HCl (pH 9), 50 mM KCl,
0.1% Triton X-100 and 10 u Taq polymerase. After an
initial denaturation at 94 ꢀC for 3 min, the cycling con-
ditions were as follows: 94 ꢀC for 1 min, 58 ꢀC for 1 min
and 72 ꢀC for 1 min for a total of 30 cycles. This was
followed by a final step of 72 ꢀC for 5 min. Amplified
samples were then run down a 2% agarose gel and
purified.
220 ꢀC; H NMR (DMSO-d6) d 2.72 (m, 2H), 3.34 (m,
1
2H), 3.71 (m, 8H), 3.82 (s, 3H), 3.88 (s, 3H), 4.03 (s,
3H), 6.33 (d, J=15.4 Hz, 1H), 6.65 (s, 1H), 6.80 (d,
J=8.4 Hz, 2H), 6.91 (s, 1H), 7.03 (s, 1H), 7.10 (s, 1H),
7.48 (m, 6H), 7.70 (d, J=15.4 Hz, 1H), 8.01 (s, 1H),
8.12 (s, 1H), 8.52 (bs, 2H), 9.12 (bs, 2H), 10.04 (s,
1H), 10.78 (s, 1H). FAB-MS (MALDI-TOF): 774.7
[M+1]+. Anal. calcd. for C38H43Cl3N10O4: C, 56.33;
H, 5.35; Cl, 13.13; N, 17.29; Found: C, 56.18; H, 5.18;
Cl, 13.02; N, 17.05.
This singly end-labelled DNA was then used at a con-
centration of approx 100 cps per sample and drug-trea-
ted for 2.5 h (room temperature) in 50 mM KCl,1 mM
MgCl2, 500 mM EDTA, 500 mM DTT and 20 mM
Hepes pH7.9in a final volume of 50 mL. After treatment
the samples were immediately cleaved by the addition of
a 3 mL solution containing 0.5 u DNAseI and 83.3 mM
MgCl2 and CaCl2. Cleavage was terminated by the
addition of stop mix 100 mL 3M NaOAC and ethanol
precipitated. Samples were washed, dried, resuspended
in 5ml formamide dye and run down a 6% sequencing
gel (Sequagel, National Diagnostics). The gels were
dried, and X-ray film was exposed to the gels (Hyper-
film, Amersham, UK).
3-[1-Methyl-4[1-methyl-4[5-[4-N,N-bis(2-chloroethyl)-
aminocinnamoyl]amino benzofuran-2-carboxamido]pyr-
role-2-carboxamido]pyrrole-2-carboxamido]propionamidine
hydrochloride (9). Following the general procedure,
starting from 22 (273 mg, 0.5 mmol), after catalytic
hydrogenation, this compound was coupled to 24 (202
mg, 0.5 mmol). The usual workup furnished the com-
pound 9 as a yellow powder (386 mg, 48%); mp 200–
203 ꢀC; H NMR (DMSO-d6) d 2.68 (m, 2H), 3.48 (m,
1
2H), 3.72 (m, 8H), 3.80 (s, 3H), 3.88 (s, 3H), 6.45 (d,
J=15.4 Hz, 1H), 6.54 (s, 1H), 6.62 (d, J=7.2 Hz, 2H),
6.84 (s, 1H), 7.06 (s, 1H), 7.22 (s, 1H), 7.28 (s, 1H), 7.45
(m, 5H), 7.78 (d, J=15.4 Hz, 1H), 8.07 (s, 1H), 8.11 (s,
1H), 8.34 (bs, 2H), 8.46 (bs, 2H), 10.04 (s, 1H), 10.55 (s,
1H). FAB-MS (MALDI-TOF): 761.7 [M+1]+. Anal.
calcd for C37H40Cl3N9O5: C, 55.75; H, 5.06; Cl, 13.34;
N, 15.81. Found: C, 55.62; H, 4.89; Cl, 13.16; N, 15.69.
Taq DNA polymerase stop assay.16 Plasmid pUC18
DNA was linearised with HindIII to provide a stop for the
Taq downstream from the primer. The oligodeoxynucleo-
tide primer (50CTCACTCAAAGGCGGTAATAC) was
50-end labelled prior to amplification using T4 poly-
nucleotide kinase and [g-32P]-ATP (5000Ci/mmol,
Amersham, UK). The labelled primers were purified by
elution through Bio-Rad spin columns. Linear amplifi-
cation of DNA was carried out in a total volume of 100
mL containing 0.5 mg of template DNA, 5 pmol of
labelled primer, 200 mM of each dNTP, 10U Taq poly-
merase, 50 mM KCl, 10 mM Tris–HCl, pH 9.0, 0.1%
Triton X-100, 2.5 mM MgCl2. Where appropriate tem-
plate DNA had been reacted with test agent for 1 h at
37 ꢀC and then precipitated with ethanol. After an initial
denaturation at 94 ꢀC for 4 min, the cycling conditions
were as follows: 94 ꢀC for 1 min, 60 ꢀC for 1 min, and
72 ꢀC for 1 min, for a total of 30 cycles. After being
amplified, the samples were ethanol precipitated and
washed with 70% ethanol. Samples were dissolved in
formamide loading dye, heated for 2 min at 90 ꢀC,
cooled on ice, and electrophoresed at 2500–3000 V for
3 h on a 80 cmꢂ20 cmꢂ0.4 mm 6% acrylamide dena-
turing sequencing gel (Sequagel, National Diagnostics).
The gels were dried, and X-ray film was exposed to the
gels (Hyperfilm, Amersham, UK).
3-[1-Methyl-4[1-methyl-4[5-[4- N,N-bis(2-chloroethyl)-
aminocinnamoyl]amino benzothiophene-2-carboxamido]-
pyrrole-2-carboxamido]pyrrole-2-carboxamido] propion-
amidine hydrochloride (10). Following the general
procedure, starting from 23 (285 mg, 0.5 mmol), after
catalytic hydrogenation, this compound was coupled to
24 (202 mg, 0.5 mmol). The usual workup furnished the
compounꢀd 10 as a yellow powder (412 mg, 51%);
mp>300 C; 1H NMR (DMSO-d6) d 2.56 (m, 2H), 3.33
(m, 2H), 3.68 (m, 8H), 3.81 (s, 3H), 3.88 (s, 3H), 6.32 (d,
J=15.4 Hz, 1H), 6.58 (s, 1H), 6.68 (d, J=7.2 Hz, 2H),
6.92 (s, 1H), 7.05 (s, 1H), 7.20 (s, 1H), 7.28 (s, 1H), 7.36
(m, 5H), 7.75 (d, J=15.4 Hz, 1H), 8.02 (s, 1H), 8.06 (s,
1H), 8.34 (bs, 2H), 8.56 (bs, 2H), 10.04 (s, 1H), 10.75 (s,
1H). FAB-MS (MALDI-TOF): 777.7 [M+1]+. Anal.
calcd for C37H40Cl3N9O4S: C, 54.65; H, 4.96; Cl, 13.08;
N, 15.50; S,3.94. Found: C, 54.61; H, 4.89; Cl, 13.01; N,
15.43; S, 3.85.
Biological evaluation
DNA footprinting. The oligodeoxynucleotide primer (50
gtcggttaggagagctccacttg) was 50-end labelled using T4
polynucleotide kinase and [g-23P]-ATP (5000 Ci/mmol,
Amersham, UK) and then purified down a Bio-Rad
spin column. PCR amplification was carried out to
Cytotoxicity assay. The K562 human chronic myeloid
leukemia cells were maintained in RPM1 1640 medium
supplemented with 10% fetal calf serum and 2 mM
glutamine at 37 ꢀC in a humidified atmosphere contain-
ing 5% CO2 and were incubated with a specified dose of

1618
P. G. Baraldi et al. / Bioorg. Med. Chem. 10 (2002) 1611–1618
drug for 1 h at 37 ꢀC in the dark. The incubation was
terminated by centrifugation (5 min, 300g) and the cells
were washed once with drug-free medium. Following
the appropriate drug treatment, the cells were trans-
ferred to 96-well microtitre plates, 104 cells per well,
eight wells per sample. Plates were then kept in the dark
at 37 ꢀC in a humidified atmosphere containing 5%
CO2. The assay is based in the ability of viable cells to
reduce a yellow soluble tetrazolium salt, 3-(4,5-di-
2000, 6, 891. (c) Cozzi, P.; Mongelli, N. Curr. Pharm. Des.
1998, 4, 181.
2. Arcamone, F. M.; Animati, F.; Barbieri, B.; Configliacchi,
E.; D’Alessio, R.; Geroni, C.; Giuliani, F. C.; Lazzari, E.;
Menozzi, M.; Mongelli, N. J. Med. Chem. 1989, 32, 774.
3. Pezzoni, G.; Grandi, M.; Biasoli, G.; Capolongo, L.; Balli-
nari, D.; Giuliani, F. C.; Barbieri, B.; Pastori, A.; Pesenti, E.;
Mongelli, N.; Spreafico, F. Br. J. Cancer 1991, 64, 1047.
4. Capolongo, L.; Melegaro, G.; Broggini, M.; Mongelli, N.;
Grandi, M. Br. J. Cancer 1993, 86, 916.
methylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium
bro-
5. (a) Broggini, M.; Erba, E.; Ponti, M.; Ballinari, D.; Geroni,
C.; Spreafico, F.; D’Incalci, M. Cancer Res. 1991, 51, 199. (b)
Broggini, M.; Coley, H. M.; Mongelli, N.; Pesenti, E.; Wyatt,
M. D.; Hartley, J. A.; D’Incalci, M. Nucl. Acids Res. 1995, 23,
81.
6. Sunters, A.; Springer, C. J.; Bagshawe, K.; Souhami, R. L.;
Hartley, J. A. Biochem. Pharmacol. 1992, 44, 59 .
7. Bellorini, M.; Moncellin, V.; D’Incalci, M.; Mongelli, N.;
Mantovani, R. Nucl. Acids Res. 1995, 23, 1657.
8. (a) Sessa, C.; Pagani, O.; Zurlo, M. G.; Jong, D. E.; Hoff-
mann, C.; Lassus, M.; Marrari, P.; Strolin-Benedetti, M.; Cavalli,
F. Ann. Oncol. 1994, 5, 901. (b) Ghielmini, M.; Bosshard, G.;
Capolongo, L.; Geroni, C.; Pesenti, E.; Torri, V.; D’Incalci,
M.; Cavalli, F.; Sessa, C. Br. J. Cancer 1997, 75, 878.
9. Cozzi, P.; Beria, I.; Biasoli, G.; Caldarelli, M.; Capolongo,
L.; Geroni, C.; Mongelli, N. Bioorg. Med. Chem. Lett. 1997, 7,
2985.
mide (MTT, Sigma Chemical Co.) to an insoluble
purple formazan precipitate. Following incubation of
the plates for 4 days (to allow control cells to increase in
number by 10-fold) 20 mL of a 5 mg/mL solution of
MTT in phosphate buffered saline was added to each
well and the plates further incubated for 5 h. The plates
were then centrifuged for 5 min at 300g and the bulk of
the medium pipetted from the cell pellet leaving 10–20
mL per well. 200 mL DMSO was added to each well and
the samples agitated to ensure complete mixing. The
optical density was then read at a wavelength of 550 nm
on a Titertek Multiscan ELISA plate reader, and the
dose-response curve was constructed. For each curve,
an IC50 value was read as the dose required to reduce
the final optical density to 50% of the control value.
10. Lee, H. H.; Boyd, M.; Gravatt, G. L.; Denny, W. A. Anti-
Cancer Drug Des. 1991, 6, 501.
11. (a) Boger, D. L.; Boyce, C. W.; Garbaccio, R. M.; Gol-
berg, J. A. Chem. Rev. 1997, 97, 787. (b) Zhou, Q.; Duan, W.;
Simmons, D.; Shayo, Y.; Raymond, M. A.; Dorr, R. T.;
Hurley, L. H. J. Am. Chem. Soc. 2001, 123, 4865. (c) Wang,
Y.; Yuan, H.; Ye, W.; Wright, S. C.; Wang, H.; Larrick, J. W.
J. Med. Chem. 2000, 43, 1541.
12. Baraldi, P. G.; Romagnoli, R.; Beria, I.; Cozzi, P.; Geroni,
G.; Mongelli, N.; Bianchi, N.; Mischiati, C.; Gambari, R. J.
Med. Chem. 2000, 43, 2675.
Acknowledgements
We wish to thank Pharmacia & Upjohn for providing
distamycin A and Ministero Universita e Ricerca Sci-
entifica (MURST) (60%) for generous financial support
of this work. Maria del Carmen Nunez is the recipient
of a Fundacion Ramon Areces fellowship.
13. Xie, G.; Gupta, R.; Lown, J. W. Anti-Cancer Drug Des.
1995, 10, 389.
14. Zambias, R. A.; Hammond, M. L. Synth. Commun. 1991,
21, 959.
References and Notes
15. Elderfield, R. C.; Liao, T. K. J. Org. Chem. 1961, 26,
4996.
16. Ponti, M.; Forrow, S. M.; Souhami, R. L.; D’Incalci, M.;
Hartley, J. A. Nucl. Acids Res. 1991, 19, 2929.
1. (a) Marchini, S.; Broggini, M.; Sessa, C.; D’Incalci, M.
Exp. Opin. Invest. Drugs 2001, 10, 1703. (b) Baraldi, P. G.;
Cacciari, B.; Spalluto, G.; Romagnoli, R. Exp. Op. Ther. Pat.