Potentiation of BCNU cytotoxicity by molecules targeting abasic lesions in DNA

Article abstract of DOI:10.1016/S0968-0896(01)00097-9

We describe the synthesis, DNA binding measurements and pharmacological properties of a series of new heterodimeric molecules, in which a 2,6-diaminopurine is linked to a 9-aminoacridine chrimophore. The linking chain contains a central N,N′-disubstituted guanidine, connected to the two chromophores by polymethylenic units of variable length. Copyright

Full text of DOI:10.1016/S0968-0896(01)00097-9

Bioorganic & Medicinal Chemistry 9 (2001) 1901–1910
Potentiation of BCNU Cytotoxicity by Molecules
Targeting Abasic Lesions in DNA
a
a,
a
Karine Alarcon, Martine Demeunynck, * Jean Lhomme,
b
b
Daniele Carrez and Alain Croisy
`
aLEDSS, Chimie Bioorganique, UMR CNRS 5616, Universite´ Joseph Fourier, BP53, 38041 Grenoble cedex 9, France
bINSERM U350, Institut Curie Recherche, Laboratoire Raymond Latarjet, Centre Universitaire, 91405 Orsay cedex, France
Received 6 February 2001; accepted 26 March 2001
Abstract—We describe the synthesis, DNA binding measurements and pharmacological properties of a series of new heterodimeric
0
molecules, in which a 2,6-diaminopurine is linked to a 9-aminoacridine chromophore. The linking chain contains a central N,N -
disubstituted guanidine, connected to the two chromophores by polymethylenic units of variable length. # 2001 Elsevier Science
Ltd. All rights reserved.
7
8
The pharmacological properties of many antitumor
drugs are due to their ability to alter DNA. In partic-
ular, alkylating agents that covalently bind to the
nucleic bases, induce major changes in the macro-
agents. Simple molecules such as methoxyamine, 9-ami-
9
10
noellipticine or isopropyloxazolocarbazole (Ipr-OPC)
have been proposed as AP-endonuclease inhibitors.
Their hypothetical mechanism of action is based on
covalent binding to the aldehydic form of the abasic
site.
1
molecule. Most of these lesions are recognized and
repaired by specific enzymes. This repair activity is cri-
tical for cells as the persistence of such lesions could be
mutagenic and even lethal. The therapeutic effect of
high doses of antitumor drugs is mainly due to the pro-
duction of multiple lesions and to the saturation of the
We have previously reported molecules that specifically
1
1À14
interact at AP-sites through another process.
These
molecules are constituted of three units: (1) an inter-
calator for targeting DNA, (2) a nucleic acid base for
the recognition of the abasic site, and (3) a polyamino
linker able to stabilize the drug/DNA complex by elec-
trostatic interactions with the phosphate backbone. The
biochemical and biophysical properties of these hetero-
dimers are strongly dependent on the nature of the lin-
ker. Molecules 1 and 2 that contain secondary amines
display strong DNA cleavage activity at abasic sites and
act as artificial endonucleases while compound 3, that
contains two guanidines in the linker, shows a strong
affinity for abasic site containing duplexes but does not
cleave DNA. In molecules 1 and 2, the amines of the
linker play a dual action, one amino group being essen-
2
repair machinery. Therefore, the design of drugs able
to interfere with DNA repair constitutes an attractive
strategy to potentiate the action of known anticancer
agents, allowing the use of lower doses to achieve com-
3
parable therapeutic result. A number of enzymes are
involved in DNA repair. The subject has been recently
4
reviewed. The major repair process of alkylated nucleic
bases (Base excision repair or BER) begins with the
excision of the modified bases by specific glycosylases,
1
1
5
thus creating abasic sites. Abasic sites or AP-sites (AP
6
for apurinic or apyrimidinic), which also form sponta-
1
4
neously or under the action of physical agents such as
UV or g-radiation, are recognized by AP-endonucleases
that are specialized enzymes able to cleave DNA strands
at AP-sites. Their action constitutes the first stage of the
repair of such lesions by a cascade of specific enzymes.
By their strategic position in the repair process and their
biological importance, AP-sites appear to be attractive
target candidates for potentiation of alkylating antitumor
tially protonated (pK =8.5 and 8.1 for 1 and 2 respec-
a
tively) increases affinity for DNA, while the second
amine being essentially unprotonated (pK =6.5 and 6.7
for 1 and 2) cleaves the phosphodiester backbone fol-
lowing a b elimination mechanism. The guanidine
residues of 3 participate only to the stabilization of the
drug/DNA complex. Compound 3 was shown to
potentiate, both in vitro and in vivo, the activity of the
alkylating antitumor agent, bis(chloroethyl)nitrosourea
a
1
5
*
4
Corresponding author. Tel.: +33-4-7651-4429; Fax: +33-4-7651-
382; e-mail: martine.demeunynck@ujf-grenoble.fr
1
4
(BCNU), while molecules 1 and 2 exhibit only little
0
968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00097-9

1902
K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
synergistic action in certain tests.7,14 These results sug-
gest that targeting the abasic site might be a realistic
strategy to potentiate the action of alkylating drugs.
connecting the guanidine to the two heterocycles were
varied in this study, and their influence on DNA bind-
ing and pharmacological properties was evaluated.
One limitation with compound 3 is a curare-like acute
toxicity in vivo, which was tentatively attributed to the
two guanidine groups. With the aim of modulating both
pharmacological activity and toxicity, and further to
ascertain this abasic site targeting strategy, we designed
new compounds containing a single guanidine in the lin-
ker. In this new series, all molecules contain the 9-amino-
Results andDiscussion
Synthesis
As shown in the retrosynthetic pathway depicted in
Figure 1, the molecules were prepared by step by step
construction of the guanidine linker. The acridine moi-
ety was introduced in the last step. Various procedures
6
-chloro-2-methoxyacridine intercalating chromophore,
and 2,6-diaminopurine was preferred to adenine as
abasic site recognition unit, as it was shown in previous
studies to confer higher binding affinity for DNA to the
0
are reported in the literature to prepare N,N -dis-
1
6À24
ubstituted guanidines. Different methods were tes-
ted and the best result was achieved using the method
1
1
heterodimers. The length of the polymethylene units
Figure 1. Retrosynthetic strategy.

K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
1903
reported by Poss,23 in which a N-protected thiourea was
reacted with a primary amine in the presence of the
coupling reagent EDC (N-diethylaminopropyl-N-ethyl-
carbodiimide, hydrochloride). As described in Scheme 1
the key thiourea intermediates 6 and 7 were prepared
one-pot by reaction of potassium thiocyanate with tri-
of the phthalimido group in acidic conditions (HCl/
AcOH/H O) affording compounds 13 and 14 in 72 and
2
74% yields, respectively. The aminoethyl analogue 12
was prepared in a different way. As observed previously
2
6
for the analogue containing adenine instead of DAP,
the reaction with 2-bromoethylphthalimide gave very
poor yield of 9-alkylated DAP 9. Compound 9 was
therefore prepared by reacting 9-(2-bromoethyl)-2,6-
1
6
chloroethyl-chloroformate, followed by addition of
the Boc-protected diaminoalkane (4 or 5). The Boc and
Troc protected thiourea synthons 6 and 7 were thus
obtained in good yields (83 and 86%, respectively). Two
methods were used to prepare the 9-aminoalkyl-2,6-
diaminopurines [DAP–(CH ) –NH , 12–14]. The propyl
1
1
diaminopurine 8 with tert-butyl-dimethylsilyl phthali-
2
7
mide in the presence of fluoride ion. Deprotection of
the amino group was achieved by acid hydrolysis (HCl/
AcOH/H O) and gave the amine 12 in 72% yield.
2
n
2
2
and butyl analogues 13 and 14 were prepared following
the method reported by Leonard for the alkylation of
Reaction of the aminoalkyl-DAP 12–14 with the pro-
tected thioureas 6 or 7 in the presence of coupling agent
EDC gave the Boc,Troc protected guanidines 15–20 as
oily products (60–80% yields, purity >95%). If neces-
sary, purification was achieved by column chromato-
graphy. Due to the high polarity of the products, yields
in pure compounds dramatically drop to 12% after
chromatography. Orthogonal deprotections of the two
protecting groups were tested, but as it appeared that
aminoacridines decomposed in the conditions required
to cleave the Troc protecting groups (zinc in acetic
acid), the two groups were cleaved prior to introduction
of the acridine nucleus. One-pot deprotection was thus
achieved by treatment of the protected guanidines 15–20
with a suspension of zinc in acetic acid, followed by
acidic hydrolysis (1 N solution of hydrochloric acid in
acetic acid). The purity of the resulting oils was checked
by hplc (purity ꢀ95%) and compounds 21–26 were
used without further purification. NMR analysis con-
firmed the complete removal of the two protecting
groups. Final reaction of the primary aliphatic amines
present in 21–26 with 6-chloro-2-methoxy-9-phenox-
yacridine (PhOAcr) was performed in phenol as pre-
2
5
adenine, that’s by treating 2,6-diaminopurine (DAP)
with the o-bromo-alkylphthalimides in the presence of
sodium hydride. The reaction took place regioselectively
at position 9, no alkylation at N-7 was observed.
Deprotection of the amine was achieved by hydrolysis
1
1
viously described. Compounds 27–32 were isolated in
–54% yields depending on the purification steps.
9
Interaction with calf-thymus DNA
The binding constants of the different molecules for
calf-thymus DNA (CT-DNA) were measured by com-
petition with ethidium bromide (EB). Studies on
depurinated DNA are precluded because of inherent
instability of the abasic sites that easily undergo strand
scission. Data are collected in Table 1.
The molecules bind strongly to native DNA (K =2–
a
5
À1
8
Â10 M ) with the exception of 29 that displays a
5
À1
). The
lower binding constant (K =0.6Â10
M
a
number of covered sites is two, in agreement with the
rule of the exclusion site for mono-intercalators. Com-
pounds 27 and 28 with two methylene units connecting
the guanidine to the purine show the highest affinities.
Interaction with AP-site containing oligonucleotides
In preceding paper,13 we showed that denaturation
study (Tm experiments) was a method of choice to
compare the relative affinity of the series of hetero-
dimers for abasic site containing oligonucleotides.
Studies were performed on synthetic oligonucleotides
containing a chemically stable analogue of the abasic
Scheme 1. Syntheses.

1904
K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
drug to the duplexes. To compare the relative affinities
of the drugs, we used ÁT values [ÁT =T (duplex in
m
m
m
the presence of the drug)ÀTm (duplex alone)]. Results
are given in Figure 2.
If we first consider the results observed with the
‘
unmodified’ TA duplex, the five compounds 27, 28, 30,
1 and 32 stabilize the duplex as shown by the regular
increase of the Tm values when the drug to DNA ratio
r) increases. The shapes and amplitudes of the curves
are very similar to what was previously observed for
3
(
1
3,14
compounds 1–3.
These data are interpreted in terms
of a non-specific interaction as expected with a drug
containing an acridine intercalator. With the abasic site
containing TX duplex however, the behavior of the
same five compounds (27, 28, 30, 31, 32) is quite differ-
Scheme 2. Sequences of the synthetic oligonucleotides used in the T
experiments.
m
lesion, 3-hydroxy-2-(hydroxymethyl)-tetrahydrofuran
designated as X). We used the sequence previously
studied in our group (molecular modeling and high field
ent. First, the increase in the T values on drug addition
m
(
is much higher. Second, the slopes of the curves are
steeper at low drug to DNA ratios (r<1) than for
higher ratios (r>1). The shapes of the curves suggest
preferred formation of 1:1 complexes between the drugs
and the oligonucleotide. A non-specific interaction is
observed at drug to DNA ratios r>1. This behavior
again is quite similar to what was observed for com-
pounds 1–3 in terms of comparable shapes and ampli-
1
2À14
nmr),
and containing a thymine opposite the lesion
in the complementary strand for possible pairing with
the DAP moiety of the drug. The sequences of the TX
duplex containing the stable abasic site, and the parent
TA duplex in which X is replaced by adenine, are shown
in Scheme 2. The thermal denaturation studies were
monitored by adding increased concentrations of each
1
3,14
tudes of the curves.
In these melting temperature
experiments monitored both with the abasic and the
parent duplexes the curves obtained for compound 29
differ considerably from those observed for the five
analogues (27, 28, 30–32). Apparently, 29 does not sta-
bilize the parent TA duplex, which correlates the lower
affinity constant measured with CT-DNA, while it cau-
ses stabilization of the abasic TX duplex to an extent
that culminates at a 1:1 ratio. These data suggest very
2
8
specific interaction of the drug at the abasic site.
Pharmacological data
In vitro pharmacological evaluation of the target com-
4
1
pounds 27–32 was carried out as previously reported.
Intrinsic cytostatic activity was estimated by measuring
the dose-dependent growth inhibition of L1210 mur-
ine leukemia cells. Cytostatic/cytotoxic properties as
well as synergistic potency with BCNU were deter-
mined by measuring the clonogenicity of A549
human pulmonary carcinoma cells. Results are given
in Table 1.
Cytostatic activity on L1210 cells appeared rather weak
and cytotoxicity on A549 moderate; no apparent rela-
tionship was found between the length of the linker and
toxicity. Compared to the parent compound 3, cyto-
toxicities are within the same scale. Apparent synergy
was measured by simultaneous exposition of the cells to
10 mM of the tested compound and 10 mM BCNU and
expressed as the % of toxicity increase with regard to
the addition of the toxic effects of both drugs alone
taken as 100% (Fig. 3).
Figure 2. Influence of the drugs on the T
and TX (below) duplexes. Experiments were conducted at pH 7
10 mM sodium phosphate, 1 mM EDTA, 20 mM NaCl). The drugs
tested were (*) 27, (&) 28, (^) 29, (X) 30, (Á) 31, and (!) 32. ÁT is
the difference between the T of the duplex in the presence of the drug
and the T of the duplex alone. T values were measured for various
ratios r of drugs (r=[drug]/[Duplex]).
m m
and ÁT of the TA (above)
(
Three compounds, 29, 30 and 32, displayed significant
synergy with BCNU. Interestingly, 29 and 32, which
are the most synergistic molecules, are among the weakest
compounds in terms of toxicity on both cell lines.
m
m
m
m

K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
1905
Table 1. Binding constants and cytostatic/cytotoxic properties (alone and in association with BCNU) of compounds 27–32: comparison with
compounds 1–3 used as references
a
(Â10À5
À1
M )
Compound (mn)
IC50 L1210 (mM)
IC50 A549 (mM)
(clonogenicity inhibition)
Synergy (%)
K
a
(
growth inhibition)
2
2
2
3
3
3
3
1
2
7 (2,3)
8 (2,4)
9 (3,3)
0 (3,4)
1 (4,3)
74
26
>100
60
3
6
28
7
31
43
10
0.1
0.05
—
—
+74
+16
—
6
8
0.6
3.7
2.1
4
20
2
11
>100
>100
33
2 (4,4)
b
+49
+94
d
c
>10
+
—
d
>10
a
Apparent synergy was measured by simultaneous exposition of A549 cells to 10 mM of the tested compound and 10 mM BCNU and expressed as the
% of toxicity increase with regard to the addition of the toxic effects of both drugs alone (theoretical combined toxicity) taken as 100%.
b
data from ref. 14.
+25% synergy observed after exposition to 0.05 mM of 1 and 10 mM BCNU. Cytotoxicity was too strong at highest doses.
c
d
IC50 of, respectively, 1.6 and 2.2 mM have been previously described after 48 h incubation with L1210 cells. Both compounds were shown synergistic
with DNA damaging agents on this cell line.
7
Although BCNU potentiation is slightly weaker with 29
than with 3, its low cytotoxic properties make of 29 a
good candidate for in vivo pharmacological and toxi-
cological studies and compel further investigations both
at the biophysical and biochemical levels.
‘Service Central de Microanalyses du CNRS’, Lyon.
Purifications were generally performed by chromato-
graphy on silica gel or alumina. Purifications on Sepha-
dex LH-20 and ion exchange resins were also tested for
the most polar compounds. The purity of the com-
pounds was assessed by reversed-phase HPLC, per-
formed on a m-bondapak C18 analytical column
(
Waters Associates). The chromatographic system is
Experimental
equipped with two M-510 pumps and a photodiode
array detector Waters 996 using Millenium 32 software.
A linear gradient from 0 to 100% methanol in H O pH
2.5 (phosphoric acid), 2 mL/min flow rate, was used.
Products were characterized by their retention time and
UV/Visible absorption.
Melting points were determined using a Reicher Ther-
movar apparatus and are uncorrected. Except when
otherwise mentioned NMR spectra were recorded at
2
2
00 MHz on Bruker AC 200 spectrometer using solvent
as the internal reference (dimethylsulphoxide-d6 at
.49 ppm); the chemical shifts are reported in ppm, in d
units. The mass spectra were recorded on Varian Mat
11 and AET MS 30 instruments. High resolution mass
spectra were obtained from Centre Regional de Mesures
Physiques de l’Ouest, Universite de Rennes. Absorption
2
3-(tert-Butoxycarbonylamino)propylamine 4 has been
2
9
3
already described 4-(tert-Butoxycarbonylamino)butyl-
amine 5. A solution of di-tert-butyldicarbonate (10 mL,
44 mmol) in chloroform (250 mL) was added dropwise
to a solution of 1,3-diaminobutane (22 mL, 220 mmol)
´
spectra were obtained on a Perkin-Elmer Lambda UV/
Vis spectrometer. Microanalyses were performed by the
ꢁ
in chloroform (500 mL) cooled at 0 C. The solution was
stirred overnight. The unsoluble was filtered off and the
solvent was removed under reduced pressure. The oily
residue thus formed was diluted with cold brine (80 mL)
0
and the white precipitate of N,N -di(tert-butox-
ycarbonyl)diaminobutane was filtered off. Extraction of
the aqueous phase with diethylether afforded after eva-
poration of the organic solvent the desired mono pro-
tected compound 5 as an oil in 95% yield (7.88 g,
41.8 mmol).
1
H NMR (200 MHz, CDCl ): d ppm 4.74 (1H, m, NH);
3
3
.04 (2H, q, CH –NH); 2.63 (2H, t, CH –NH ); 1.36
3
2 2 2
0
(15H, m, CH –CH –CH –CH NH and C(CH ) )
2 2 2 2, 2 3 3
0
N-(3-(tert-Butoxycarbonylamino)propyl)-N -trichloro-
ethoxycarbonyl thiourea 6. Trichloroethyl chloro-
formate (1.25 mL, 9.26 mmol) was added dropwise to a
solution of potassium thiocyanate (1 g, 10.3 mmol) in
ꢁ
acetone (20 mL) cooled at 0 C. After 2 h of stirring,
Figure 3. Apparent synergy of compounds 29, 30 and 32 with BCNU.
Apparent synergy was measured by simultaneous exposition of A549
cells to 10 mM of the tested compound and 10 mM BCNU and expres-
sed as the % of toxicity increase with regard to the addition of the
toxic effects of both drugs alone taken as 100% (theoretical toxicity).
compound 4 was added (1.79 g, 10.3 mmol) to the solu-
tion and the resulting mixture was stirred overnight at
room temperature. After evaporation of the solvent, the
residue was diluted with water, the pH was adjusted to

1906
K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
pH 4 and compound 6 was extracted with dichloro-
methane. The organic layer was washed with water,
dried over sodium sulfate and evaporated. Compound 6
was obtained as a yellow solid in 83% yield (3.13 g,
resulting mixture was stirred 5 days at room tempera-
ture. The solvent was evaporated to dryness and the
residue was triturated in methanol–water mixture (1/1).
The solid thus formed, was filtered and washed with
methanol. Compound 10 was obtained in 53% yield
ꢁ
1
7
.66 mmol). Mp 160 C. H NMR (200 MHz, CDCl ): d
3
ꢁ
1
ppm=9.61 (1H, m, NH); 8.24 (1H, m, NH); 4.76 (2H, s,
CH –CCl ); 3.73 (2H, q, CH –NH); 3.18 (2H, q, CH –
NH); 1.82 (2H, quint, CH –CH –CH ); 1.61 (1H, m,
(2.39 g, 7.1 mmol). Mp 194 C. H NMR (200 MHz,
DMSO-d ): d 7.83 (4H, m, Phth-H), 7.72 (1H, s, H-8);
2
3
2
2
6
6.60 (2H, m, NH ); 5.72 (2H, m, NH ); 3.99 (2H, t,
DAP-CH ); 3.57 (2H, t, CH -NPht); 2.06 (2H, quint,
2
2
2
2
2
1
3
NH); 1.43 (9H, s, C(CH ) ). C NMR (50 MHz,
3
3
2
2
1
3
CDCl ): d 178.6 (C¼S); 156.2 (C¼O); 150.6 (C¼O);
CH CH CH ).
2
C NMR (75 MHz, DMSO-d ): d
3
2
2
6
9
3.9 [CH -CCl ); 79.4 (C(CH ) ]; 74.7 (CH –CCl ); 42.8
3
167.8; 160.1; 156.0; 151.6; 137.4; 134.2 (2 CH);
131.6; 122.9 (2 CH); 113.2; 40.2 (N–CH ); 35.0 (N–
2
3
3 3
2
(
N–CH ); 37.3 (N–CH ); 29.0 (CH -CH -CH ); 28.3
2
2
2
2
2
2
(
C(CH ) ). ms (CI, ammoniac+isobutane): M=408,
3
CH ); 28.3 (CH ). ms (CI, ammoniac+isobutane):
2
3
2
+
+
+
m/z: 408 (27, M ); 352 (100, M ÀC(CH ) ); 293 (72,
M=337, m/z 338 (97, (M+H) ); 206 (100). Anal.
found: C, 44.87; H, 4.64; N, 22.49%; calcd for
C H N O , 1 HBr, 0.5 H O, 0.5 MeOH: C, 44.71; H,
3
3
+
+
M ÀCCl ); 234 [24, M ÀC(CH ) –CCl ]. Anal.
3
3 3
3
found: C, 35.44; H, 4.81; N, 10.26%; calcd for
C Cl H N O S: C, 35.26; H, 4.93; N, 10.28%.
1
6
15
7
2
2
4.32; N, 22.12%.
1
2
3
20
3
4
0
N - (4 - (tert - Butoxycarbonylamino)butyl) - N - trichloro -
ethoxycarbonyl thiourea 7. The procedure described
above was used starting from potassium thiocyanate
(
(
2,6-Diamino-9-(4-phthalimido-butyl)-9H-purine 11: DAp-
c4-phth. The procedure described above was used to
prepare compound 11, starting from 2,6-diaminopurine
(2 g, 13.3 mmol), sodium hydride (60%, 0.6 g, 15 mmol)
and 4-bromobutylphthalimide (4.14 g, 15 mmol). Com-
pound 11 was thus obtained in 52% yield (2.42 g,
0.5 g, 5.14 mmol), trichloroethyl chloroformate
0.625 mL, 4.62 mmol) and 4-(tert-butoxycarbonylamino)
butylamine 5 (0.968 g, 5.14 mmol). Compound 7 was iso-
ꢁ
1
ꢁ
1
lated in 86% yield (1.69 g, 3.99 mmol). Mp 140 C. H
NMR (200 MHz, CDCl ): d 9.44 (1H, m, NH); 8.28
6,9 mmol). Mp 269 C. H NMR (200 MHz, DMSO-d ):
6
d 7.83 (4H, s, Pht-H); 7.67 (1H, s, H-8); 6.59 (2H, s,
NH ); 5.72 (2H, s, NH ); 3.95 (2H, t, DAP–CH ); 3.58
3
(
1H, m, NH); 4.75 (2H, s, CH –CCl ); 4.56 (1H, s, NH);
2 3
2
2
2
3
.67 (2H, q, CH –NH); 3.14 (2H, q, CH –NH); 1.69
2
(2H, t, CH –NPht); 1.73 (2H, m, DAP-CH –CH ); 1.56
2 2 2
2
1
3
(2H, m, CH –CH –CH –CH ); 1.53 (2H, m, CH –CH –
1
CH –CH ); 1.42 (9H, s, C(CH ) ). C NMR (50 MHz,
(2H, m, CH –CH -NPht). C (50 MHz, DMSO d ): d
2
2
2
2
2
2
2
2
6
3
167.7; 159.9; 155.8; 137.3; 134.2 (CH); 131.3; 122.8
(CH); 112.9; 104.5 (CH); 41.6 (N–CH ); 36.7 (N-CH );
2
2
3 3
CDCl ): d 179.0 (C¼S); 156.4 (C¼O); 151.4 (C¼O);
3
2
2
9
4.4 (CH –CCl ); 79.6 [C(CH ) ]; 75.2 (CH –CCl ); 45.7
2
26.6 (CH ); 25.0 (CH ). ms [FAB(+), glycerol]:
2 2
3
3 3
2
3
+
(
N–CH ); 40.4 (N–CH ); 28.7 [C(CH ) ]; 27.8 (CH );
2
M=351, m/z 352 [100, (M+H) ]. Anal. found: C,
57.49; H, 4.82; N, 27.29%; calcd for C H N O , 0.2
2
3 3
2
2
5.9 (CH ). ms (CI, ammoniac+isobutane): M=423,
2
17 17
7
2
+
+
m/z 424 [49, (M+H) ]; 366 [100, M ÀC(CH ) ]; 248
H O: C, 57.52; H, 4.94; N, 27.62%.
2
3
3
+
[
5
0
47, M ÀC(CH ) –CCl ]. Anal. found: C, 35.35; H,
3
3
3
.02; N, 9.76%; calcd for C H N O Cl S, 0.7 H O,
1
2,6-Diamino-9-(!-aminoalkyl)-9H-purine: general pro-
cedure. A solution of DAP-C -Phth (1.8 mmol) in 12N
3
22
3
4
3
2
.05 CH Cl C, 35.65; H, 5.39; N, 9.56%.
2
2
n
hydrochloric acid/water/acetic acid mixture (1/1/1)
ꢁ
2,6-Diamino-9-(2-phthalimidoethyl)-9H-purine 9: DAP-C -
Phth. N-(tert-Butyl-dimethylsilyl)phthalimide
(6 mL) was stirred at 100 C for 24 h. After cooling to
2
2
7
(0.2 g,
.78 mmol) was added to a suspension of 9-(2-bromo-
room temperature, the solution was concentrated under
reduced pressure. The phthalic acid that precipitated
was filtered off. The solvent was evaporated to dryness
and the residue triturated in diethylether. The solid thus
formed was filtered, and (DAP-C –NH ) were obtained
0
1
1
ethyl)-2,6-diaminopurine (0.2 g, 0.78 mmol) in THF
20 mL). Tetrabutylammonium fluoride (0.4 g,
.1 mmol) was added in portions at room temperature
(
1
n
2
ꢁ
and the mixture was heated at 50 C under nitrogen for
4 h. After cooling to room temperature, the solution
as the trihydrochlorides in 70–75% yield. Products were
purified by crystallization from ethanol.
2
was filtered and the solid residue was washed several
times with THF. Purification was achieved by column
chromatography (silica gel, from 0 to 2% methanol in
ethylacetate), 9 was then obtained in 40% yield (0.1 g,
12. (DAP-C –NH ). Mp 202–205 C. 1H NMR
ꢁ
2
2
(200 MHz, D O): d 8.01 (1H, s, H-8); 4.49 (2H, t, DAP-
CH ); 3.53 (2H, m, CH –NH ). ms [FAB (+), glycerol]:
+
2
2
2
2
ꢁ
1
0
.3 mmol). Mp 299–300 C. H NMR (200 MHz,
M=193, m/z 194 (100, M+H) .
DMSO-d ): d 7.79 (4H, m, Phth-H), 7.61 (1H, s, H-8);
6
ꢁ
1
6
DAP-CH ); 3.91 (2H, t, CH -NPhth). ms [FAB (+),
.55 (2H, m, NH ); 5.37 (2H, m, NH ); 4.21 (2H, t,
13. (DAP-C -NH ). Mp 290 C. H NMR (200 MHz,
2
2
2
3
DMSO-d ): d 8.81 (2H, m, NH ); 8.20 (3H, m, NH and
2
2
6
2
2
+
glycerol]: M=323, m/z 324 (M+H) .
H-8); 7.59 (2H, m, NH ); 4.15 (2H, t, DAP-CH ); 2.75
(2H, m, CH –NH ); 2.08 (2H, quint, CH CH CH ); ms
(CI, ammoniac+isobutane): M=208, m/z 208 (100,
2 2
2 2 2 2 2
2,6-Diamino-9-(3-phthalimidopropyl)-9H-purine 10: DAP-
C -Phth. Sodium hydride (600 mg, 15 mmol) was
+
M ).
3
washed with pentane and then added to a suspension of
2
ꢁ
14. (DAP-C –NH ): mp 277 C. H NMR (200 MHz,
4 2
1
,6-diaminopurine (2 g, 13.3 mmol) in DMF(60 mL)
kept under nitrogen. After 2 h of stirring, 3-bromopro-
pylphthalimide (4 g, 15 mmol) was added and the
DMSO-d ): d 8.11 (1H, s, H-8); 7.91 (2H, m, NH ); 7.52
6
2
(2H, m, NH ); 4.04 (2H, m, DAP-CH ); 2.80 (2H, m,
2
2

K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
1907
0
ethoxycarbonyl)-N -[3-(2,6-diamino-purin-9-yl)propyl]-
CH –NH ); 1.80 (2H, m, DAP-CH –CH ); 1.49 (2H, m,
2
N-[4-(tert-Butoxycarbonylamino)-butyl]-N -(trichloro-
00
2
2
2
13
CH –CH –NH ). C NMR (75 MHz, DMSO-d ): d
2
2
2
6
1
52.5; 150.5; 150.2; 140.9; 110.4; 42.5 (N–CH ); 37.9
guanidine 18, was prepared from thiourea 7 (0.72 g,
1.70 mmol) and DAP-C –NH 13 (0.70 g, 2.22 mmol).
Compound 18 was obtained as an oil in 79% yield
2
(N–CH ); 26.1 (CH –CH –CH –CH ); 23.7 (CH –
2 2 2 2 2 2
CH –CH –CH ). ms [FAB(+), glycerol]: M=221, m/z
3
2
2
2
2
+
1
2
22 (100, (M+H) ).
(0.804 g, 1.35 mmol). H NMR (200 MHz, CDCl ): d
3
7.53 (1H, s, H-8); 5.61 (1H, m, NH); 4.92 (1H, m, NH);
4
.75 (2H, s, CH –CCl ); 4.13 (2H, t, DAP-CH ); 3.26
2
3
2
Preparation of the protectedgua ni ni de s. Typical
procedure
(4H, m, 2 CH ); 3.11 (2H, m, CH ); 2.33 (4H, m,
2 2
2
NH ); 2.00 (2H, m, CH –CH –CH ); 1.57 (4H, m,
2 2 2 2
0
1
N-[3-(tert-Butoxycarbonylamino)-propyl]-N -(trichloro-
0
CH –CH –CH –CH ); 1.39 (9H, s, C(CH ) ). H NMR
2 2 2 2 3 3
(200 MHz, DMSO-d ): d 7.72 (1H, s, H-8); 6.80 (1H, m,
NH); 6.64 (1H, m, NH); 6.39 (1H, m, NH); 5.76 (1H, m,
NH); 4.74 (2H, s, CH –CCl ); 3.96 (2H, t, DAP-CH );
0
ethoxycarbonyl)-N -[2-(2,6-diamino-purin-9-yl)ethyl]gua-
nidine 15. A mixture of thiourea 6 (0.5 g, 1.22 mmol),
6
0
DAP-C –NH 12 (0.48 g, 1.58 mmol) and EDC [N -(3-
2
2
2
3
2
dimethylaminopropyl)-N-ethyl-carbodiimide
chloride] (0.352 g, 1.83 mmol) was solubilized in DMF
14 mL). Triethylamine (0.663 mL, 4.76 mmol) was
hydro-
3.15 (4H, m, 2 CH ); 2.91 (2H, m, CH ); 1.93 (2H, m,
2 2
CH –CH –CH ); 1.35 [13H, m, CH –CH –CH –CH
2
2
2
2
2
2
2
1
3
(
and C(CH ) ]. C NMR (75 MHz, DMSO-d ): d 160.2;
3
3
6
added to the solution and the mixture was stirred at
room temperature for 4 h. The solvent was evaporated
and the oily residue was diluted with water and com-
pound 15 was extracted with dichloromethane. After
159.6; 156.8; 156.1; 155.6; 153.5; 137.4; 96.7 (CH2–
CCl ); 77.4; 75.3; 37.7; 28.2; 26.9; 22.9. ms (FAB(+),
NBA): M=595, m/z 596 [(M+H) ]; 447 (M ÀDAP).
3
+
+
0
ethoxycarbonyl)-N -[4-(2,6-diamino-purin-9-yl)butyl]-
evaporation of the solvent, 15 was obtained as an oil in
1
N-[3-(tert-Butoxycarbonylamino)-propyl]-N -(trichloro-
00
7
1% yield (0.494 g, 0.87 mmol). H NMR (200 MHz,
CDCl ): d 7.50 (1H, s, H-8); 5.43 (1H, m, NH); 4.94
guanidine 19, was prepared from thiourea 6 (0.76 g,
1.86 mmol) and DAP-C -NH 14 (0.80 g, 2.41 mmol).
3
(
CH ); 3.74 (2H, m, CH ); 3.17 (4H, m, 2 CH ); 1.97–
2H, s, NH ); 4.74 (2H, s, CH -CCl ); 4.25 (2H, t, DAP-
2 2 3
4
2
After the treatment described above, the oily residue
was purified by column chromatography on silica gel
(elution dichloromethane/methanol, 9/1 mixture con-
taining 1% ammonium hydroxide). Compound 19 was
obtained as an oil in 12% yield (0.132 g, 0.22 mmol).
2
2
2
1
C(CH ) ). ms [FAB(+), glycerol]: M=568, m/z 569
.69 (4H, m, CH –CH –CH and NH ); 1.41 (9H, s,
2 2 2 2
3
3
+
+
(
(M+H) ); 419 (M ÀDAP); 276 (100).
0
ꢁ
1
N-[4-(tert-Butoxycarbonylamino)-butyl]-N -(trichloro-
Mp 79–80 C. H NMR (200 MHz, CDCl ): d 8.71 (1H,
3
00
ethoxycarbonyl)-N -[2-(2,6-diamino-purin-9-yl)ethyl]-
guanidine 16, was prepared from thiourea 7 (0.250 g,
m, NH); 7.44 (1H, s, H-8); 6.40 (2H, broad m, NH2);
5.27 (3H, m, NH and NH); 4.98 (1H, m, NH); 4.67
2
0
It was obtained in 60% yield (0.207 g, 0.35 mmol). H
.59 mmol) and DAP-C –NH 12 (0.232 g, 0,77 mmol).
1
(2H, s, CH –CCl ); 3.93 (2H, t, DAP-CH ); 3.28 (2H,
2
m, CH ); 3.19 (2H, m, CH ); 3.07 (2H, m, CH ); 1.80
2 2 2
2
2
3
2
NMR (200 MHz, CDCl ): d 7.48 (1H, s, H-8); 5.50 (1H,
(2H, m, CH –CH –CH ); 1.54 (4H, m, CH –CH –CH –
2 2 2 2 2 2
CH ); 1.32 [9H, s, C(CH ) ]. C NMR (75 MHz,
2 3 3
3
1
3
m, NH); 4.82 (2H, m, NH ); 4.74 (2H, s, CH –CCl );
4
2
2
3
.25 (2H, t, DAP-CH ); 3.78 (2H, m, CH ); 3.11
2
CD OD): d 163.1; 161.7; 161.6; 157.5; 152.7; 139.8 (CH-
2
3
(
CH ); 1.40 [9H, s, C(CH ) ]. ms [FAB(+), glycerol]:
M=582, m/z 583 [100, (M+H) ].
4H, m, 2 CH ); 1.51 (4H, m, CH –CH –CH –
8); 114.2 (C-5); 97.6 (CH –CCl ); 80.1 [C(CH ) ]; 76.4
(CH –CCl ); 44.0 (N–CH ); 41.4 (N–CH ); 39.4 (N–
2 3 2 2
CH ); 38.5 (N–CH ); 30.8 (CH ); 28.8 (C(CH ) ); 28.2
2 2 2 3 3
2
2
2
2
2
3
3 3
2
3 3
+
(
CH ). ms [FAB(+), glycerol]: M=596, m/z 597
2
0
+
+
N-[3-(tert-Butoxycarbonylamino)-propyl]-N -(trichloro-
0
((M+H) ); 447 (M ÀDAP). UV (ethanol 95%): l
max
0
ethoxycarbonyl)-N -[3-(2,6-diamino-purin-9-yl)propyl]-
guanidine 17, was prepared from thiourea 6 (0.8 g,
.96 mmol) and DAP-C –NH 13 (0.805 g, 2.55 mmol).
3
After the treatment described above, the oily residue
was purified by column chromatography on silica gel
(e): 280 (13,000), 257 (11,000), 218 (43,900) nm.
0
ethoxycarbonyl)-N -[4-(2,6-diamino-purin-9-yl)butyl]-
1
N-[4-(tert-Butoxycarbonylamino)-butyl]-N -(trichloro-
0
2
guanidine 20, was prepared from thiourea 7 (0.7 g,
1.65 mmol) and DAP-C -NH 14 (0.71 g, 2.14 mmol).
(elution dichloromethane/methanol, 9/1, v/v mixture
containing 1% ammonium hydroxide). Compound 17
4
2
After the treatment described above, the oily residue
was purified by column chromatography on silica gel
(elution dichloromethane/methanol, 9/1 mixture con-
taining 1% ammonium hydroxide). Compound 20 was
obtained as an oil in 22% yield (0.225 g, 0.36 mmol).
was obtained as an oil in 12% yield (0.14 g, 0.24 mmol).
1
H NMR (200 MHz, CDCl ): d 9.51 (1H, m, NH); 7.48
3
(
1H, s, H-8); 6.43 (1H, m, NH); 5.95 (2H, m, NH2);
5.37–5.07 (3H, m, NH and NH); 4.74 (2H, s, CH –
CCl ); 4.09 (2H, t, DAP-CH ); 3.39 (2H, m, CH ); 3.19
2
2
ꢁ
1
Mp 92–95 C. H NMR (200 MHz, CDCl ): d 7.47 (1H,
3
2
2
3
(
(
4H, m, 2 CH ); 1.99 (2H, m, CH –CH –CH ); 1.64
2
s, H-8); 6.30 (2H, broad m, NH ); 5.24–5.00 (3H, m,
NH and NH); 4.69 (2H, s, CH –CCl ); 3.96 (2H, t,
2
2
2
2
1
3
2H, m, CH –CH –CH ); 1.38 (9H, s, C(CH ) ).
2
C
2
2
3 3
2
2
3
NMR (50 MHz, CD OD): d ppm 162.9; 161.9; 161.5;
DAP-CH ); 3.19 (4H, m, 2 CH ); 3.05 (2H, m, CH );
2 2 2
1.80 (2H, m, CH –CH –CH –CH ); 1.46 (6H, m,
2 2 2 2
3
1
9
4
59.9; 158.7; 157.5; 153.2; 139.8 (CH-8); 114.1 (C-5);
7.6 (CH –CCl ); 80.1 (C(CH ) ); 76.4 (CH –CCl );
CH –CH –CH –CH and CH –CH –CH –CH ); 1.34
2
2
3
3 3
2
3
2
2
2
2
2
2
2
1
3
1.5 (N–CH ); 39.4 (N–CH ); 38.9 (N–CH ); 38.5 (N–
2
(9H, s, C(CH3)3). C NMR (50 MHz, CD OD): d
2
2
3
CH ); 30.8 (CH ); 28.8 [C(CH ) ]. ms [FAB(+), gly-
2
cerol]: M=582, m/z 583 [(M+H) ].
163.1; 161.7; 161.5; 158.6; 152.7; 139.9 (CH-8); 114.2;
100.0; 97.6 (CH –CCl ); 79.9 [C(CH ) ]; 76.5 (CH –
2
3 3
+
2
3
3 3
2

1908
K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
CCl ); 44.0 (N–CH ); 41.8 (N–CH ); 41.4 (N–CH );
3
Final coupling with the acridine nucleus. Typical procedure
2
2
2
4
2
0.8 (N–CH ); 28.8 (C(CH ) ); 28.3 (CH ); 28.1 (CH );
2 3 3 2 2
7.5 (CH ). ms [FAB(+), NBA]: M=609, m/z 609
2
N-[3-(6-Chloro-2-methoxy-acridin-9-yl-amino)propyl]-
N -[2-(2,6-diamino-purin-9-yl)ethyl]-guanidine DAP-C2-
+
+
00
gua-C3-NHAcr 27.
(
M ); 461 (M +1-DAP).
A mixture of 21 (0.095 g,
.22 mmol), triethylamine (0.12 mL, 0.88 mmol) and 6-
0
chloro-2-methoxy-9-phenoxyacridine (0.145 g, 0.44 mmol)
One-pot deprotection of the guanidine and the primary
amine. General procedure
ꢁ
was stirred in phenol (2 mL) at 80 C under nitrogen for
A suspension of zinc powder (1.4 g) in acetic acid/water
ꢁ
5 h. The solution was slowly poured into acetone
(25 mL) under vigorous stirring. The yellow solid was
filtered and washed carefully with acetone. The solid
was solubilized in water and addition of 1N sodium
hydroxide allowed compound 27 to precipitate as a free
base. After filtration, the solid was washed several times
with water. Crystallization from methanol containing 12
N hydrochloric acid and acetone afforded 27 as the tet-
rahydrochloride in 9% yield (0.013 g, 0.02 mmol). Mp
220–230 C. H NMR (200 MHz, CD3OD) (free base):
d 8.16 (1H, d, J=9.6 Hz, Acr-H); 7.76 (1H, s, H-8); 7.71
(2H, m, Acr-H); 7.44 (1H, d, J=2.4 Hz, Acr-H); 7.33
(1H, dd, J=9.2 and 2.4 Hz, Acr-H); 7.20 (1H, dd,
(
5/1, v/v) mixture (3.5 mL) was cooled at 0 C. Boc-,
Troc-protected compound (15–20) was added and the
resulting mixture was stirred for 4 h. The suspension
was then filtered onto Celite to remove zinc and salts.
The solvent was evaporated and the oily residue was
diluted in hydrochloric acid 1 N in acetic acid. After 1 h
of stirring at room temperature, the solvent was evapo-
rated under reduced pressure. The fully deprotected
compounds (21–26) were obtained as tetra-
hydrochlorides. They appeared as very hygroscopic
gums and they were used without further purification.
The purity was checked by hplc (purity 595%), and the
total disappearance of Boc and Troc protecting groups
ꢁ
1
J=9.2 and 2.1 Hz, Acr-H); 4.07 (2H, t, DAP-CH ); 3.89
2
1
was assessed by H NMR.
(3H, s, O–CH ); 3.80 (2H, t, CH ); 3.42 (2H, t, CH );
3
2
2
3.12 (2H, t, CH ); 1.91 (2H, m, CH –CH –CH ). ms
(CI, ammoniac+isobutane): M=533, m/z 534 (2,
2 2 2 2
0
0
N-(3-Aminopropyl)-N -[2-(2,6-diamino-purin-9-yl)ethyl]-
1
+
+
guanidine 21. 31% yield. H NMR (200 MHz, DMSO-
d ): d 7.98 (1H, d, H-8); 7.68 (7H, m, 2 NH and NH);
4.15 (2H, m, DAP-CH ); 3.55 (2H, m, CH ); 3.15 (2H, m,
CH ); 2.79 (2H, m, CH ); 1.71 (2H, m, CH –CH –CH ).
(M+H) ); 244 (76, (Acr+1) ). UV (H O): l
(e):
2
max
444 (7100), 423 (7400), 343 (3800), 279 (44,500), 217
(36,500) nm. hrms [FAB(+), NBA]: found: 534.2248;
calcd for C ClH N O+H: 534.2245.
6
2
2 2
2
2
2
2
2
25
28 11
0
0
00
[2-(2,6-diamino-purin-9-yl)ethyl]-guanidine DAP-C -gua-
C –NHAcr 28. Compound 22 (0.113 g, 0.25 mmol),
4
N-(4-Aminobutyl)-N -[2-(2,6-diamino-purin-9-yl)ethyl]-
1
N-[4-(6-Chloro-2-methoxy-acridin-9-yl-amino)butyl]-N -
guanidine 22. 30% yield. H NMR (200 MHz, DMSO-
d ): d 7.81 (1H, s, H-8); 7.66 (7H, m, NH and NH);
2
6
2
7
CH ); 3.10 (2H, m, CH ); 2.77 (2H, m, CH ); 1.49 (4H,
.53 (1H, m, NH); 4.12 (2H, t, DAP-CH ); 3.55 (2H, m,
triethylamine (0.14 mL, 1.0 mmol) and 6-chloro-2-
methoxy-9-phenoxyacridine (0.167 g, 0.50 mmol) were
mixed in phenol (3 mL). The procedure was the same as
2
2
2
2
1
m, CH –CH –CH –CH ). H NMR (200 MHz, D O): d
2
2
2
2
2
ꢁ
1
7
CH ); 2.84 (4H, m, 2 CH ); 1.41 (2H, m, CH –CH –
.71 (1H, s, H-8); 4.11 (2H, t, DAP-CH ); 3.49 (2H, t,
described above. 51% yield. mp: 210–215 C. H NMR
2
(200 MHz, D O): d 8.03 (1H, d, J=9.6 Hz, Acr-H); 7.60
2
2
2
2
2
CH –CH ); 1.26 (2H, m, CH –CH –CH –CH ).
2
(1H, s, H-8); 7.30 (5H, m, Acr-H); 3.94 (2H, t, DAP-
CH ); 3.77 (3H, s, O–CH ); 3.55 (2H, m, CH ); 3.21
(2H, m, CH ); 3.12 (2H, t, CH ); 1.84 (2H, m, CH –
2 2 2
2
2
2
2
2
2
3
2
00
N-(3-Aminopropyl)-N -[3-(2,6-diamino-purin-9-yl)pro-
1
pyl]guanidine 23. 54% yield. H NMR (300 MHz,
D O): d 7.96 (1H, s, H-8); 3.99 (2H, t, DAP-CH ); 3.20
CH –CH –CH ); 1.62 (2H, m, CH –CH –CH –CH ).
2 2 2 2 2 2 2
1
3
C NMR (50 MHz, D O): d 155.8; 155.5; 154.3; 151.6;
2
2
2
(
CH –CH ); 1.77 (2H, m, CH –CH –CH ).
4H, t, 2 CH ); 2.84 (2H, m, CH ); 1.97 (2H, m, CH –
2
151.3; 148.9; 142.0; 140.6; 138.4; 127.1; 124.4; 124.3;
119.8; 116.7; 112.9; 108.8; 102.3; 56.3 (O–CH ); 47.4
2
2
2
2
2
2
2
3
(
NH–CH ); 41.7 (NH–CH ); 40.6 (NH–CH ); 40.5
2 2 2
0
0
N-(4-Aminobutyl)-N -[3-(2,6-diamino-purin-9-yl)propyl]-
1
(NH-CH ); 26.5 (CH –CH –CH –CH ); 25.6 (CH –
2 2 2 2 2 2
CH –CH –CH ). ms (FAB(+), NBA): M=547, m/z:
2 2 2
548 ((M+H) ). UV (H O): l (e): 423 (6000), 279
(38500), 217 (31300) nm. hrms [FAB(+), NBA]: found:
guanidine 24. H NMR (200 MHz, D O): d 7.67 (1H, d,
2
+
H-8); 4.03 (2H, m, DAP-CH ); 3.07 (2H, m, CH ); 2.84
(
2
2
2
max
4H, m, 2 CH ); 2.00 (2H, m, CH –CH –CH ); 1.41
2
2
2
2
(
4H, m, CH –CH –CH –CH ).
2
548.2420; calcd for C ClH N O+H: 548.2402.
26 30 11
2
2
2
0
0
N-(3-Aminopropyl)-N -[4-(2,6-diamino-purin-9-yl)butyl]-
1
N-[3-(6-Chloro-2-methoxy-acridin-9-yl-amino)propyl]-
N -[3-(2,6-diamino-purin-9-yl)propyl]guanidine DAP-C -
00
gua-C NHAcr 29. The procedure was the same as the
guanidine 25. 41% yield. H NMR (200 MHz, D O): d
7.79 (1H, s, H-8); 3.90 (2H, t, DAP-CH ); 3.06 (2H, t,
CH ); 2.98 (2H, t, CH ); 2.84 (2H, t, CH ); 1.71 (4H,
2
3
2
3
typical procedure described above, except the purifica-
tion step. After filtration, the solid obtained from the
precipitation in acetone was dissolved in water. The pH
was adjusted to pH 12 sodium hydroxide. As 29 did not
precipitate from the alkaline solution, the aqueous
phase was washed several times with ethyl acetate, and
extracted with butanol. The organic layer was washed
with water, dried on sodium sulfate and evaporated
2
2
2
m, CH –CH –CH –CH ); 1.38 (2H, m, CH –CH –
2
2
2
2
2
2
CH2).
0
0
N-(4-Aminobutyl)-N -[4-(2,6-diamino-purin-9-yl)butyl]-
guanidine 26. The presence of large amount of salts in
the final product and its high hygroscopic character
precluded accurate NMR analysis.

K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
1909
1
3
under reduced pressure. The oily residue was dissolved
in methanol, the solution was filtered, and 12 N hydro-
chloric acid was added to it. Compound 29 precipitated
by adding acetone/diethylether (1/1 mixture) to the
solution and was isolated as tetrahydrochloride in 42%
C NMR (75 MHz, D O): d 156.4; 155.9; 155.7; 153.3;
2
150.9; 150.5; 141.8; 141.0; 139.3; 133.9; 127.4; 127.3;
124.5; 120.4; 117.4; 113.6; 111.0; 109.3; 102.8; 56.5 (O–
CH ); 46.5 (NH–CH ); 43.5 (NH–CH ); 40.7 (NH–
CH ); 38.6 (NH–CH ); 28.6 (CH –CH –CH ); 26.6
2
(CH –CH –CH ); 25.1 (CH –CH –CH ). ms (FAB(+),
2 2 2 2 2 2
3
2
2
2
2
2
2
ꢁ
1
yield (0.111 g, 0.16 mmol). mp: 210–220 C. H NMR
+
(
(
(
300 MHz, D O): d 7.97 (1H, d, J=9.3 Hz, Acr-H); 7.64
1H, s, Acr-H); 7.34–7.24 (4H, m, Acr-H and H-8); 7.17
1H, d, J=9.3 Hz, Acr-H); 4.01 (2H, m, DAP-CH2);
glycerol): M=561, m/z 562 [(M+H) ]. UV (H O): l
2
2
max
(e): 423 (6200), 279 (41,800), 217 (36,000) nm. hrms
[FAB(+), NBA]: found: 562.2556; calcd for
C ClH N O+H: 562.2558.
3
CH ); 2.95 (2H, m, CH ); 2.14 (2H, m, CH –CH –
.77 (3H, s, O–CH ); 3.57 (2H, m, CH ); 3.34 (2H, m,
3 2
27
32 11
2
2
2
2
1
3
00
[4-(2,6-diamino-purin-9-yl)butyl]guanidine DAP-C -gua-
CH ); 1.67 (2H, m, CH –CH –CH ).
2
C NMR
N-[4-(6-Chloro-2-methoxy-acridin-9-yl-amino)butyl]-N -
2
2
2
(
75 MHz, D O): d 156.5; 155.9; 154.4; 152.3; 151.6;
2
4
1
(
1
49.6; 142.3 (CH); 141.2; 139.4; 134.0; 127.6 (CH); 124.8
CH); 120.5 (CH); 117.4 (CH); 113.7; 110.9; 109.6;
02.9; 102.8 (CH); 56.6 (O–CH ); 46.5 (N–CH ); 40.7
C -NHAcr 32. Compound 32 was prepared and pur-
4
ified with the methodology described for 29. It was iso-
lated as the tetrahydrochloride in 37% (0.011 g,
3
2
ꢁ
1
(
N–CH ); 39.7 (N–CH ); 39.1 (N–CH ); 28.2 (CH );
2
0.015 mmol). Mp 210–215 C. H NMR (300 MHz,
2
2
2
2
8.0 (CH ). ms (CI, ammoniac+isobutane): M=547,
2
D O): d 7.81 (1H, d, J=9.3 Hz, Acr-H); 7.50 (1H, s,
2
+
m/z 548 [(M+H) ]. UV (H O): l
(e): 424 (7100),
Acr-H); 7.30–7.25 (3H, m, Acr-H and H-8); 7.14 (1H,
dd, J=9.3 and 2.0 Hz, Acr-H); 7.09 (1H, d, J=2.0 Hz,
Acr-H); 3.78 (2H, m, DAP-CH ); 3.74 (3H, s, O–CH );
2
max
3
43 (3800), 281 (46,800), 217 (53,200) nm.
2
3
0
0
N-[4-(6-Chloro-2-methoxy-acridin-9-yl-amino)butyl]-N -
3-(2,6-diamino-purin-9-yl)propyl]guanidine DAP-C -gua-
C -NHAcr 30. Compound 30 was prepared as
3.56 (2H, t, CH ); 3.17 (2H, t, CH ); 3.03 (2H, t, CH );
2 2 2
1.79 (2H, m, CH –CH –CH –CH ); 1.60 (2H, m, CH –
2 2 2 2 2
[
3
CH –CH –CH ); 1.49 (2H, m, CH –CH –CH –CH );
2
4
2
2
2
2
2
2
1
3
described above for 27. The free base was purified by
chromatography on acidic alumina (elution dichloro-
methane/methanol, 8/2 followed by 6/4 mixtures). The
fractions containing 30 were collected and evaporated
to dryness. The residue was contaminated with alumina.
It was dissolved in water and 30 was extracted with
butanol. The solvent was evaporated under reduced
pressure. 30 was crystallized from methanol containing
1.37 (2H, m, CH –CH –CH –CH ).
2
C
NMR
2
2
2
(75 MHz, D O): d ppm 156.2; 156.1; 155.3; 153.1; 150.9;
2
150.3; 141.6; 140.9; 139.2; 134.0; 128.9; 127.3; 124.7;
120.4; 117.4; 113.3; 110.9; 109.4; 56.5 (O–CH ); 48.3
3
(NH–CH ); 43.3 (NH–CH ); 40.8 (NH–CH ); 40.5
2
2
2
(NH–CH ); 26.9 (CH –CH –CH –CH ); 26.3 (CH –
2
2
2
2
2
2
CH –CH –CH ); 25.7 (CH –CH –CH –CH ); 25.0
2
2
2
2
2
2
2
(CH –CH –CH –CH ). ms [FAB(+), glycerol]:
2
2
2
2
+
12 N hydrochloric acid by adding diethylether. It was
thus isolated as the tetrahydrochloride in 24% yield
M=575, m/z 576 [(M+H) ]. UV (H O): l
(e): 423
2
max
(5000), 278 (32,100), 217 (29,900) nm.
ꢁ
1
(
0.215 g, 0.32 mmol). Mp: 210–220 C.
H NMR
(
200 MHz, CD OD) (Free base): d 8.20 (1H, d,
3
DNA binding studies
J=9.2 Hz, Acr-H); 7.77 (1H, s, H-8); 7.72 (2H, m, Acr-
H); 7.46 (1H, d, J=1,7 Hz, Acr-H); 7.32 (1H, dd,
J=9.2 Hz and 2.4 Hz, Acr-H); 7.20 (1H, dd, J=9.2 Hz
DNA binding experiments were performed with Perkin
Elmer MPF-44A and LS50 spectrofluorimeters in a
thermostated quartz cell (25 C). Excitation and emis-
ꢁ
and 2.1 Hz, Acr-H); 3.99 (2H, t, DAP-CH ); 3.89 (3H, s,
2
O–CH ); 3.79 (2H, t, CH ); 3.01 (4H, m, 2 CH ); 1.90
3
sion were monitored at the ethidium bromide bands
(520 and 600 nm). All solutions were made in 25 mM
Tris–HCl, pH 7.0, 0.1 M NaCl and 0.2 mM EDTA buf-
fer. Calf-thymus DNA was used. DNA concentration
2
2
1
3
(
4H, m, 2 CH ); 1.45 (2H, m, CH ). C NMR (50 MHz,
2 2
D O): d 155.8; 155.6; 154.0; 151.4; 150.8; 148.7; 141.8;
2
1
1
4
40.6; 132.9; 127.5; 127.0; 124.5; 119.8; 116.8; 112.6;
10.0; 56.2 (O–CH ); 47.8 (NH–CH ); 40.5 (NH–CH );
À5
was 1.6Â10 M (in base pairs) and that of tested drug
3
2
2
À5
0.4 (NH–CH ); 38.0; 26.5; 25.6. ms [FAB(+), gly-
2
2Â10 M.
+
cerol]: M=561, m/z 562 [(M+H) ]. UV (H O): l
2
max
(
[
e): 423 (6000), 278 (36900), 217 (32900) nm. hrms
FAB(+), NBA]: found: 562.2552; calcd for
C ClH N O+H: 562.2558.
In vitro assays. L1210 cells (Mouse leukemia, ATCC
CCL 219) were cultivated in Dulbecco’s MEM supple-
mented with 10% fetal calf serum. L1210 cells were
2
7
32 11
5
seeded at 10 cells/mL in 1 mL microwell plates. After
5
N-[3-(6-Chloro-2-methoxy-acridin-9-yl-amino)propyl]-
N -[4-(2,6-diamino-purin-9-yl)butyl]guanidine DAP-C -
24 h (usually 3 to 4Â10 cells/mL) tested compounds
0
gua-C -NHAcr 31. Compound 31 was prepared and
0
were added in duplicate at various concentrations and
incubated for 24 h. Cells were counted with a Coulter-
Counter ZM (Coultronics Inc.). The dose inhibiting the
growth by 50% (IC ) was interpolated from regression
4
3
purified with the methodology described for 29. It was
isolated as the tetrahydrochloride in 25% yield (0.013 g,
0
5
0
ꢁ
1
.018 mmol). Mp 185–190 C. H NMR (300 MHz,
curves obtained with experimental points without sig-
nificant toxicity.
D O): d 7.80 (1H, d, J=9.3 Hz, Acr-H); 7.59 (1H, s,
2
Acr-H); 7.28–7.25 (3H, m, Acr-H and H-8); 7.13–7.09
2H, m, Acr-H); 3.87 (2H, t, DAP-CH ); 3.74 (3H, s, O–
(
CH ); 3.66 (2H, t, CH ); 3.20 (2H, t, CH ); 2.92 (2H, t,
A549 cells (human pulmonary adenocarcinoma, ATCC
CCL 185) were grown in Kaighn modified Ham F12
medium (F12-K) supplemented with 10% fetal calf
serum. Cells were plated in 3 cm diameter multiwell
2
3
2
2
CH ); 2.01 (2H, m, CH –CH –CH ); 1.49 (2H, m, CH –
2
2
2
2
2
CH –CH –CH ); 1.34 (2H, m, CH –CH –CH –CH ).
2
2
2
2
2
2
2

1910
K. Alarcon et al. / Bioorg. Med. Chem. 9 (2001) 1901–1910
plates (200 cells/well). After 24 h, tested compounds,
alone or with 10 mM BCNU, were added in duplicate at
various concentrations and cells were incubated for
another 24 h. Cells were washed with phosphate buf-
fered saline (PBS) and then reincubated for 14 days with
fresh medium free of drug. Colonies were numbered
after washing with PBS and staining with Giemsa. Sur-
12. Coppel, Y.; Constant, J.-F.; Coulombeau, C.; Demeu-
nynck, M.; Garcia, J.; Lhomme, J. Biochemistry 1997, 36, 4831.
1
3. Berthet, N.; Constant, J.-F.; Demeunynck, M.; Michon,
P.; Lhomme, J. J. Med. Chem. 1997, 40, 3346.
4. Belmont, P.; Jourdan, M.; Demeunynck, M.; Constant, J.-
1
F.; Garcia, J.; Lhomme, J.; Croisy, A.; Carrez, D. J. Med.
Chem. 1999, 42, 5153.
1
5. Belmont, P.; Boudali, A.; Constant, J.-F.; Demeunynck,
vival was expressed as% of untreated controls and IC50
M.; Fkyerat, A.; Michon, P.; Serratrice, G.; Lhomme, J. New
J. Chem. 1997, 21, 47.
16. Atkins, P. R.; Glue, S. E. J.; Kay, I. T. J. Chem. Soc.
Perkin 1 1973, 22, 2644.
was interpolated from regression curves obtained with
experimental points.
ꢁ
jacketed CO2 incubator (5% CO , 100% relative
humidity).
All incubations were carried out at 37 C in water-
17. Atwal, K. S.; Ahmed, S. Z.; O’Reilly, B. C. Tetrahedron
Lett. 1989, 30, 7313.
2
1
1
8. Kim, K. S.; Qian, L. Tetrahedron Lett. 1993, 34, 7677.
9. Lammin, S. G.; Pedgrift, B. L.; Ratcliffe, A. J. Tetra-
hedron Lett. 1996, 37, 6815.
The purity of tested compounds was assessed by HPLC
and was ꢀ98%.
2
5
2
0. Molina, P.; Alajarin, M.; Vidal, A. Tetrahedron 1995, 51,
351.
1. Nagarajan, S.; Ho, T.-L.; DuBois, G. E. Synthetic Com-
mun. 1992, 22, 1191.
Acknowledgements
22. Pannecouque, C.; Wigerinck, P.; Van Aerschot, A.; Her-
dewijn, P. Tetrahedron Lett. 1992, 33, 7609.
23. Poss, M. A.; Iwanowicz, E. J.; Reid, J. A.; Lin, J.; Gu, Z.
Tetrahedron Lett. 1992, 33, 5933.
24. Schneider, S. E.; Bishop, P. A.; Salazar, M. A.; Bishop,
O. A.; Anslyn, E. V. Tetrahedron 1998, 54, 15063.
The authors thank Dr. Catherine Bougault for provid-
ing the abasic site containing oligonucleotide and Dr.
Pierre Michon for assistance in thermal denaturation
studies.
2
3
2
5. Leonard, N. J.; Lambert, R. F. J. Org. Chem. 1969, 34,
240.
6. Belmont, P.; Demeunynck, M.; Constant, J.-F.; Lhomme,
References andNotes
J. Bioorg. Med. Chem. Lett. 2000, 10, 293.
7. Kita, Y.; Haratu, J. I.; Fujii, T.; Segawa, J.; Tamura, Y.
Synthesis 1981, 451–452.
2
1
7
2
3
4
. Hemminki, K.; Ludlum, D. B. J. Natl. Cancer Inst. 1984,
3, 1021.
. Chaney, S. G.; Sancar, A. J. Natl. Cancer Inst. 1996, 88, 1346.
. Barret, J. M.; Hill, B. T. Anti-Cancer Drugs 1998, 9, 105.
. Berthet, N.; Boturyn, D.; Constant, J.-F. Exp. Opin. Ther.
Patents 1999, 9, 401.
. Sancar, A. Annu. Rev. Biochem. 1996, 65, 43.
. Lhomme, J.; Constant, J. F.; Demeunynck, M. Biopolym.,
28. Possible interpretation for low affinity of 29 for
unmodified DNA may reside in strong intramolecular p–p
stacking between purine and acridine rings favored by the
symmetry of the linker that incorporates two identical
and short three methylene units as substituents of the
guanidine. Such an influence of strong intramolecular
stacking on the binding to DNA has been reported in
the past for comparable heterodimeric molecules. The
extend of binding was show to be inversely proportional
5
6
Nucleic Acid Sci. 1999, 52, 65.
. Barret, J. M.; Etievant, C.; Fahy, J.; Lhomme, J.; Hill, B. T.
Anti-Cancer Drugs 1999, 10, 55.
7
ˆ
to the degree of stacking. See: Constant, J.-F.; Laugaa,
8
1
9
1
1
. Liuzzi, M.; Weinfeld, M.; Paterson, M. C. Biochemistry
987, 26, 3315.
. Lefranc¸ ois, M.; Bertrand, J. R.; Malvy, C. Mutation Res.
990, 236, 9.
0. Malvy, C.; Safraoui, H.; Bloch, E.; Bertrand, J. R. Anti-
P.; Roques, B. P.; Lhomme, J. Biochemistry 1988, 27,
3997. The structure of the complex(es) formed between 29
and the DNA containing the abasic residue is currently
being studied.
29. Boturyn, D.; Boudali, A.; Constant, J.-F.; Lhomme, J.
Tetrahedron 1997, 53, 5485.
30. Houssin, R.; Bernier, J.-L.; Henichart, J.-P. Synthesis
´
Cancer Drug Des. 1988, 2, 361.
1. Fkyerat, A.; Demeunynck, M.; Constant, J.-F.; Michon,
P.; Lhomme, J. J. Am. Chem. Soc. 1993, 115, 9952.
1
1988, 259–261.