Quinone methides tethered to naphthalene diimides as selective G-quadruplex alkylating agents

Article abstract of DOI:10.1021/ja904876q

We have developed novel G-quadruplex (G-4) ligand/alkylating hybrid structures, tethering the naphthalene diimide moiety to quaternary ammonium salts of Mannich bases, as quinone-methide precursors, activatable by mild thermal digestion (40°C). The bis-substituted naphthalene diimides were efficiently synthesized, and their reactivity as activatable bis-alkylating agents was investigated in the presence of thiols and amines in aqueous buffered solutions. The electrophilic intermediate, quinonemethide, involved in the alkylation process was trapped, in the presence of ethyl vinyl ether, in a hetero Diels-Alder [4 + 2] cycloaddition reaction, yielding a substituted 2-ethoxychroman. The DNA recognition and alkylation properties of these new derivatives were investigated by gel electrophoresis, circular dichroism, and enzymatic assays. The alkylation process occurred preferentially on the G-4 structure in comparison to other DNA conformations. By dissecting reversible recognition and alkylation events, we found that the reversible process is a prerequisite to DNA alkylation, which in turn reinforces the G-quadruplex structural rearrangement.

Full text of DOI:10.1021/ja904876q

Published on Web 08/20/2009
Quinone Methides Tethered to Naphthalene Diimides as
Selective G-Quadruplex Alkylating Agents
Marco Di Antonio,† Filippo Doria,‡ Sara N. Richter,§ Carolina Bertipaglia,†
Mariella Mella,‡ Claudia Sissi,† Manlio Palumbo,*,† and Mauro Freccero*,‡
Dipartimento di Chimica Organica, UniVersita` di PaVia, V.le Taramelli 10, 27100 PaVia, Italy,
Dipartimento di Scienze Farmaceutiche, UniVersita` di PadoVa, Via Marzolo, 5, 35131 PadoVa,
Italy, and Dipartimento di Istologia, Microbiologia e Biotecnologie Mediche, UniVersita` di
PadoVa, Via Gabelli, 63, 35121, PadoVa, Italy
Received June 15, 2009; E-mail: mauro.freccero@unipv.it; manlio.palumbo@unipd.it
Abstract: We have developed novel G-quadruplex (G-4) ligand/alkylating hybrid structures, tethering the
naphthalene diimide moiety to quaternary ammonium salts of Mannich bases, as quinone-methide
precursors, activatable by mild thermal digestion (40 °C). The bis-substituted naphthalene diimides were
efficiently synthesized, and their reactivity as activatable bis-alkylating agents was investigated in the
presence of thiols and amines in aqueous buffered solutions. The electrophilic intermediate, quinone-
methide, involved in the alkylation process was trapped, in the presence of ethyl vinyl ether, in a hetero
Diels-Alder [4 + 2] cycloaddition reaction, yielding a substituted 2-ethoxychroman. The DNA recognition
and alkylation properties of these new derivatives were investigated by gel electrophoresis, circular dichroism,
and enzymatic assays. The alkylation process occurred preferentially on the G-4 structure in comparison
to other DNA conformations. By dissecting reversible recognition and alkylation events, we found that the
reversible process is a prerequisite to DNA alkylation, which in turn reinforces the G-quadruplex structural
rearrangement.
of small molecules, acting as G-4 selective ligands.3 Because
of their recognition properties toward nucleic acids, derivati-
Introduction
Guanine-rich regions, such as the 5′-TTAGGG repetitive
sequence contained in the human telomeres, can form tetraplex
assemblies known as G-quadruplexes (G-4).1 A recent approach
to telomerase inhibition is based on the conformational remodel-
ing of the DNA repetitive telomeric structures, which involves
sequestering its substrate, the single-stranded telomeric DNA,
by inducing it to fold into G-4 structure(s).2 Stabilization of
G-4 conformations has been achieved by using a great variety
ves of 1,8-naphthalimide, such as 1,4,5,8-naphthalene tetracar-
boxylic diimides (NDIs) and perylene analogues (PDIs) have
been extensively studied in recent decades and shown to act by
intercalation or end-stacking.4 In more detail, a few naphthalene
imides and diimides including dimeric derivatives bind selec-
tively duplex DNA.5 Several naphthalimide-containing com-
(3) (a) Yang, P.; De Cian, A.; Teulade-Fichou, M.-P.; Mergny, J.-L.;
Monchaud, D. Angew. Chem., Int. Ed. 2009, 48, 2188–2191. (b) Dash,
J.; Shirude, P. S.; Hsu, S.-Te D.; Balasubramanian, S. J. Am. Chem.
Soc. 2008, 130, 15950–15956. (c) Rodriguez, R.; Muller, S.; Yeoman,
J. A.; Trentesaux, C.; Riou, J.-F.; Balasubramanian, S. J. Am. Chem.
Soc. 2008, 130, 15758–15759. (d) Campbell, N. H.; Parkinson, G. N.;
Reszka, A. P.; Neidle, S. J. Am. Chem. Soc. 2008, 130, 6722–6724.
(e) Tera, M.; Ishizuka, H.; Takagi, M.; Suganuma, M.; Shin-ya, K.;
Nagasawa, K. Angew. Chem., Int. Ed. 2008, 47, 5557–5560. (f) Zhou,
Q.; Li, Lin, X.; Junfeng, T.; Yalin, Z.; Hong; Yang, S.; Li, Qian, Y.;
Qianfan; Xu, G. Angew. Chem., Int. Ed. 2008, 47, 5590–5592. (g)
Monchaud, D.; Yang, P.; Lacroix, L.; Teulade-Fichou, M.-P.; Mergny,
J.-L. Angew. Chem., Int. Ed. 2008, 47, 4858–4861. (h) Bugaut, A.;
Jantos, K.; Wietor, J.-L.; Rodriguez, R.; Sanders, J. K. M.; Balasubra-
manian, S. Angew. Chem., Int. Ed. 2008, 47, 2677–2680. (i) Bejugam,
M.; Sewitz, S.; Shirude, P. S.; Rodriguez, R.; Shahid, R.; Balasubra-
manian, S. J. Am. Chem. Soc. 2007, 129, 12926–12927. (j) Shirude,
P. S.; Gillies, E. R.; Ladame, S.; Godde, F.; Shin-ya, K.; Huc, I.;
Balasubramanian, S. J. Am. Chem. Soc. 2007, 129, 11890–11891. (k)
Searle, M. S.; Balkwill, G. D. In Quadruplex Nucleic Acids; Neidle,
S.; Balasubramanian, S. Eds.; RSC Publishing: Cambridge, U.K., 2006;
pp 131-153.
† Dipartimento di Scienze Farmaceutiche, Universita` di Padova.
‡ Dipartimento di Chimica Organica, Universita` di Pavia.
§ Dipartimento di Istologia, Microbiologia e Biotecnologie Mediche,
Universita` di Padova.
(1) (a) Parkinson, G. N. In Quadruplex Nucleic Acids; Neidle, S.,
Balasubramanian, S., Eds.; RSC Publishing: Cambridge, U.K., 2006;
pp 1-30. (b) Davis, J. T. Angew. Chem., Int. Ed. 2004, 43, 668–698.
(c) Monchaud, D.; Teulade-Fichou, M.-P. Org. Biomol. Chem. 2008,
6, 627–636. (d) Zahler, A. M.; Williamson, J. R.; Cech, T. R.; Prescott,
D. M. Nature 1991, 350, 718–720. (e) Lim, K. W.; Amrane, S.;
Bouaziz, S.; Xu, W.; Mu, Y.; Patel, D. J.; Luu, K. N.; Phan, A. T.
J. Am. Chem. Soc. 2009, 131, 4301–4309. (f) Xu, Y.; Suzuki, Y.;
Lonnberg, T.; Komiyama, M. J. Am. Chem. Soc. 2009, 131, 2871–
2874. (g) Martadinata, H.; Phan, A. T. J. Am. Chem. Soc. 2009, 131,
2570–2578. (h) Petraccone, L.; Trent, J. O.; Chaires, J. B. J. Am. Chem.
Soc. 2008, 130, 16530–16532.
(2) (a) Mergny, J.-L.; Gros, J.; De Cian, A.; Bourdoncle, A.; Rosu, F.;
Sacca`B.; Guittat, L.; Amrane, S.; Mills, M.; Alberti, P.; TakasugyM.;
Lacroix., L. In Quadruplex Nucleic Acids; Neidle, S., Balasubramanian,
S., Eds.; RSC Publishing: Cambridge, U.K., 2006; pp 31-99. (b)
Rodriguez, R.; Pantos, G. D.; Goncalves, D. P. N.; Sanders, J. K. M.;
Balasubramanian, S. Angew. Chem., Int. Ed. 2007, 46, 5405–5407.
(c) Gray, R. D.; Chaires, J. B. Nucleic Acids Res. 2008, 36, 4191–
4203.
(4) (a) Guelev, V.; Sorey, S.; Hoffman, D. W.; Iverson, B. L. J. Am. Chem.
Soc. 2002, 124, 2864–265. (b) Bevers, S.; Schutte, S.; McLaughlin,
L. W. J. Am. Chem. Soc. 2000, 122, 5905–5915. (c) Pivetta, C.;
Lucatello, L.; Krapcho, P., A.; Gatto, B.; Palumbo, M.; Sissi, C.
Bioorg. Med. Chem. 2008, 16, 9331–9339.
9
13132 J. AM. CHEM. SOC. 2009, 131, 13132–13141
10.1021/ja904876q CCC: $40.75  2009 American Chemical Society

Selective G-Quadruplex Alkylating Agents
A R T I C L E S
Scheme 1. Generation of QMs from Phenol (1a, 2a,b) and Binol
(3a-c) Derivatives
pounds have also been described as photoinduced DNA cleaving
agents,6 and some of them have been entered into clinical trials
owing to their strong anticancer activity.7 Only very recently,
a series of disubstituted NDIs have also been screened as G-4
ligands, showing lower affinity in comparison to the more
extended perylene analogues.8 More promising G-4 binding
properties have been achieved by Neidle’s group using tri- and
tetra-substituted NDIs.9
Very recently, covalent modifications and, more generally,
chemical reactions have been suggested as effective strategies
for developing methods aimed at probing G-4 structures.10
Possible G-4 alkylating species thus far reported include Pt-
complexes11 and quaternary ammonium porphyrins.12 As a
matter of fact, the first example refers to coordination of the
metal ion to a purine N7 in the tetraplex structure (G-quadruplex
platination),13 and the second infers the possibility of photoin-
duced G-4 cross-linking, by a carbon electrophile (alkylation),
considering that phenol quaternary ammonium derivatives can
cross-link duplex DNA efficiently upon irradiation.14
Scheme 2. Substituent Effects on the Reactivity of QM-Precursors
(QMP)
Chemical reactions to be exploited for G-4 recognition and
covalent modification should be very mild, selective, and readily
initiated. To achieve this aim with single organic compounds,
quinone methide (QM) reactivity has been exploited. QMs are
transient Michael acceptors which can be generated from
suitable and hydrosoluble precursors by mild activation.14
Several groups, including our own, have shown that o-
hydroxy benzyl alcohols (1),15 Mannich base derivatives of
(5) Chu, Y.; Sorey, S.; Hoffman, D. W.; Iverson, B. L. J. Am. Chem.
Soc. 2007, 129, 1304–1311.
(6) (a) Matsugo, S.; Kawanishi, S.; Yamamoto, K.; Sugiyama, H.;
Matsuura, T.; Saito, I. Angew. Chem., Int. Ed. Engl. 1991, 30, 1351–
1353. (b) Saito, I.; Takayama, M.; Sakurai, T. J. Am. Chem. Soc. 1994,
116, 2653–2654.
phenols (2)16,17 and binols (3),14b,c are capable of undergoing
mono- and bis-alkylation in water and DNA cross-linking by
photoactivation,14b mild base catalysis,14c thermal digestion
under physiological conditions or in the presence of fluoride
anion,18 and mild monoelectronic reduction (Scheme 1).19
When studying a model QM in the presence of biological
nucleophiles (including deoxynucleosides), formation of QMs
was demonstrated to depend upon the leaving group at the
benzylic position of its precursor (QMP, in Scheme 2).18
Furthermore, both the generation of QMs under physiological
conditions and the thermodynamic stability of the resulting
adducts were shown to be highly responsive to both the presence
of electron-withdrawing and -donating substituents on the
aromatic ring (Y, in Scheme 2)20 and on the nature of the leaving
group (X, in Scheme 2).
(7) Bran˜a, M. F.; Castellano, J. M.; Mora´n, M.; Pe´rez de Vega, M. J.;
Qian, X. D.; Romerdahl, C. A.; Keilhauer, G. Eur. J. Med. Chem.
1995, 30, 325–329.
(8) Sissi, C.; Lucatello, L.; Krapcho, P. A.; Maloney, D. J.; Boxer, M. B.;
Camarasa, M. V.; Pezzoni, G.; Menta, E.; Palumbo, M. Bioorg. Med.
Chem. 2007, 17, 555–562.
(9) (a) Cuenca, F.; Greciano, O.; Gunaratnam, M.; Haider, S.; Munnur,
D.; Nanjunda, R.; Wilson, W. D.; Neidle, S. Bioorg. Med. Chem. Lett.
2008, 18, 1668–1673. (b) Parkinson, Gary, N.; Cuenca, F.; Neidle, S.
J. Mol. Biol. 2008, 381, 1145–1156.
(10) Xu, Y.; Suzuki, Y; Komiyama, M. Angew. Chem., Int. Ed. 2009, 48,
3281–3284.
(11) (a) Ma, D.-L.; Che, C.-M.; Yan, S.-C. J. Am. Chem. Soc. 2009, 131,
1835–1846. (b) Kieltyka, R.; Englebienne, P.; Fakhoury, J.; Autexier,
C.; Moitessier, N.; Sleiman, H. F. J. Am. Chem. Soc. 2008, 130, 10040–
10041. (c) Kieltyka, R.; Fakhoury, J.; Moitessier, N.; Sleiman, H. F.
Chemistry 2008, 14, 1145–1154. Rao, L.; Bierbach, U. J. Am. Chem.
Soc. 2007, 129, 15764–15765. (d) Bertrand, H.; Bombard, S.;
Monchaud, D.; Teulade-Fichou, M.-P. Nucleic Acids Symp. Ser. 2008,
52, 163–164. (e) Ourliac-Garnier, I.; Elizondo-Riojas, M.-A.; Redon,
S.; Farrell, N. P; Bombard, S. Biochemistry 2005, 44, 10620–10634.
(12) (a) He, H. P.; Liang, F.; Li, D. Q.; Wu, J. J.; Yang, L.; Zhou, X.;
Zhang, X. L.; Cao, X. P. Bioorg. Med. Chem. 2006, 14, 1068–1077.
(b) Wang, P.; Ren, L.; He, H.; Liang, F.; Zhou, X.; Tan, Z. Chem.
Bio. Chem. 2006, 7, 1155–1159.
Electron-donating groups, including a p-alkyl substituent,
facilitate QM generation, under mild conditions (Scheme 2, Y
) CH3), while electron-withdrawing groups strongly suppress
QM formation, making the resulting adducts more stable.20,21
In more detail, quaternary ammonium salts of Mannich bases
(16) Okamoto, A.; Nakamura, T.; Yoshida, K.; Nakatani, K.; Saito, I. Org.
Lett. 2000, 2, 3249–3251.
(13) Bertrand, H.; Bombard, S.; Monchaud, D.; Talbot, E.; Guedin, A.;
Mergny, J.-L.; Gruenert, R.; Bednarski, P. J.; Teulade-Fichou, M.-P.
Org. Biomol. Chem. 2009, 7, 2864–2871.
(17) Doria, F.; Richter, S. N.; Nadai, M.; Colloredo-Mels, S.; Mella, M.;
Palumbo, M.; Freccero, M. J. Med. Chem. 2007, 50, 6570–6579.
(18) (a) Veldhuyzen, W. F.; Praveen, P.; Rokita, S. E. J. Am. Chem. Soc.
2003, 125, 14005–14013. (b) Veldhuyzen, W. F.; Shallop, A. J.; Jones,
R. A.; Rokita, S. E. J. Am. Chem. Soc. 2001, 123, 11126–11132.
(19) Di Antonio, M.; Doria, F.; Mella, M.; Merli, D.; Profumo, A.; Freccero,
M. J. Org. Chem. 2007, 72, 8354–8360.
(14) (a) Richter, S.; Maggi, S.; Colloredo-Mels, S.; Palumbo, M.; Freccero,
M. J. Am. Chem. Soc. 2004, 126, 13973–13979. (b) Colloredo-Mels,
S.; Doria, F.; Verga, D.; Freccero, M. J. Org. Chem. 2006, 71, 3889–
3895. (c) Modica, E.; Zanaletti, R.; Freccero, M.; Mella, M. J. Org.
Chem. 2001, 66, 41–52. (d) Wang, P.; Liu, R.; Wu, X. J.; Ma, H. J.;
Cao, X. P.; Zhou, P.; Zhang, J. Y.; Weng, X. C.; Zhang, X. L.; Qi, J.;
Zhou, X.; Weng, L. H. J. Am. Chem. Soc. 2003, 125, 1116–1117.
(15) (a) Diao, L.; Cheng, Y.; Wan, P. J. Am. Chem. Soc. 1995, 117, 5369–
5370. (b) Chiang, Y.; Kresge, A. J.; Zhu, Y. J. Am. Chem. Soc. 2002,
123, 6349–6356.
(20) Weinert, E. E.; Dondi, R.; Colloredo-Mels, S.; Frankenfield, K. N.;
Mitchell, C. H.; Freccero, M.; Rokita, S. E. J. Am. Chem. Soc. 2006,
128, 11940–11947.
(21) Wang, H.; Wahi, M. S.; Rokita, S. E. Angew. Chem., Int. Ed. 2008,
47, 1291–1293.
9
J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009 13133

A R T I C L E S
Di Antonio et al.
Chart 1. Bis-Substituted NDIs 4-6 Investigated as G-4 Ligand-Alkylating Hybrid Compounds
Scheme 3. Synthesis of 4-Aminomethyl-, 4-(2-Aminoethyl)-, and 4-(3-Aminopropyl)-2-dimethylaminomethylphenols (n ) 1, 2, and 3,
respectively)a
a Reagents and conditions: (i) acrylonitrile as solvent, AlCl3; (ii) LiAlH4, THF, ∆, 4 h; (iii) Ac2O, NaHCO3, H2O, rt; (iv) paraformaldehyde, dimethylamine
in anhydrous EtOH, ∆, 2 h; (v) HCl aq 10%, 2 h, ∆.
+
(X ) NR3 ) bearing a methoxy and a methyl group (Y ) OCH3,
CH3) have been shown to act both as QM-precursors (QMPs)
under physiological conditions (25 °C < T < 40 °C, 6.0 < pH e
7.8), but only the latter generate fairly stable alkylation
adducts.20 When using nucleosides as nucleophiles, only the
conjugate adducts with deoxycytidine at the N3 (dC-N3),
deoxyadenosine at both N1 and NH2 (dA-N1 and dA-NH2), and
deoxyguanosine at NH2 (dG-NH2) are stable in water solution.
Contrary, the alkylation adducts at N7 of both dG and dA are
thermodynamically unstable, since the reversal of the alkylation
becomes a competitive process (Scheme 2).22
by exhaustive methylation of the diamines 4a-6a (Chart 1).
The NDIs 4a-6a were synthesized in a one-pot synthesis from
the commercially available 1,4,5,8-naphthalenetetracarboxylic
acid dianhydride and 4-aminomethyl-, 4-(2-aminoethyl)-, and
4-(3-aminopropyl)-2-dimethylaminomethylphenols, respectively.
The above intermediates have been prepared according to the
general synthetic procedure depicted in Scheme 3 and described
in detail in Supporting Information.
Activation and Reactivity of Naphthalene Diimides 4-6. Our
study began with the investigation of the reactivity of the NDIs
4-6, in order to define the most convenient protocol for their
activation as bis-alkylating agents. We performed digestion of
4-6 at both 25 and 40 °C in the presence of several simple
nucleophiles (thiols and amines), in aqueous buffered and
unbuffered solutions. The results are collected in the Table 1.
Therefore, taking advantage of such a latent alkylating
reactivity and the G-4 recognizing properties of the NDI moiety,
we decided to investigate the activation conditions and the
reactivity and the G-4 binding properties of three new bis-
alkylating QMPs tethered to the NDI core, by conformationally
flexible (CH2)n spacers (n ) 1-3; 4-6, Chart 1). The aim of
this work is to demonstrate that the G-4 binding properties of
fairly weak G-4 reversible ligands, such as disubstituted NDIs,
can be strengthened by alkylation of the target, using two QMP
moieties as activatable alkylating agents. These compounds
clearly differ from the reported Pt-complexes11,13 as the alky-
lation step by a QM needs activation to be performed, which
in principle allows spatial and temporal control of the reaction
process. In addition a large number of QM-generating species
can be devised, which again allows modulation in reactivity
and hence in the conditions at which alkylation is performed.
This can be particularly relevant in view of therapeutic applica-
tions of the QMP-NDI conjugates.
Activation by base catalysis in the presence of amines under
unbuffered conditions takes advantage of the low pKa (∼7.8)
of the quaternary ammonium salts of the three NDIs 4-6, which
in the presence of basic nucleophiles allows the generation of
reactive zwitterionic forms of 4-6 at pH g 8. The latter
promptly release the alkylating QM (Scheme 1) at 25 °C. In
the presence of nonbasic nucleophiles such as thiols, a longer
reaction time (6-12 h) is required at 40 °C under neutral
conditions. Specific base catalysis in buffered solutions at pH
8.5 triggered the alkylation of the less basic nucleophiles at
lower temperature (25 °C), affording the alkylated adducts in
good yields (g90%).
In the absence of a nucleophile, hydration reaction occurred,
affording diols 7, 14, and 17 in low yield at 25 °C. Higher yields
in a shorter reaction time were achieved at 40 °C. Data in Table
1 suggest that the NDIs 4-6 exhibit a very similar reactivity,
since they can be activated as bis-alkylating agents under the
very same mild conditions (pH 7, at 40 °C, for less than 12 h).
Amines 4a-6a, unlike their quaternary ammonium salts 4-6,
are stable for at least 24 h at 40 °C at both pH 7.0 and 8.5.
Therefore, they cannot generate alkylating QMs under these mild
conditions.
Results and Discussion
Synthesis of Naphthalene Diimides Tethered to
Quinone-methide Precursors. The quaternary ammonium salts
4-6 used as ligand-alkylating hybrid compounds were prepared
(22) (a) Freccero, M.; Di Valentin, C.; Sarzi-Amade, M. J. Am. Chem.
Soc. 2003, 125, 3544–3553. (b) Freccero, M.; Gandolfi, R.; Sarzi-
Amade, M. J. Org. Chem. 2003, 68, 6411–6423.
9
13134 J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009

Selective G-Quadruplex Alkylating Agents
A R T I C L E S
Table 1. Reactivity of 4-6 in the Presence of Nucleophiles
Scheme 4. Generation and Reactivity of QM5 (Tethered to the NDI
Moiety) as Heterodiene
in a G-4 conformation was achieved by reacting increasing
concentrations (0.25-16 µM) of the compounds at 40 °C for
24 h in buffer Li3PO4 10 mM, KCl 50 mM, pH 7.4, with a 32
P
5′-end labeled synthetic oligonucleotide formed by four human
telomeric repeats, 5′-AGGGTTAGGGTTAGGGTTAGGG-3′
(4GGG). In these conditions, 4GGG can fold into an intramo-
lecular G-4 preferentially assuming a hybrid-type (mixed
parallel/antiparallel) conformation.23 The alkylated oligo was
separated from the nonreacted DNA by denaturing polyacry-
lamide gel. As shown in Figure 1A, compound 5 exhibited
alkylating properties. The covalent adduct was detectable at a
concentration as low as 0.25 µM (lane 2, Figure 1A) and reached
15% over total DNA at compound 5 concentration of 2 µM
(lane 5, Figure 1A). At higher ligand concentrations no further
increment in the alkylated adduct was observed. This is likely
related to the NDI-DNA complex precipitation (lanes 6-8,
Figure 1A) occurring at high molar ratios. Reactivity of 5 lead
to maximal 4GGG adduct formation in 24 h at 40 °C whereas
alkylation was substantially absent at 20 °C. In addition, the
identified alkylation product was stable upon incubation at
20-70 °C. Degradation was appreciable at higher temperatures
(Figure 1S, Supporting. Information).
When the reactivity of NDIs with different linkers was
monitored, compound 4 was able to alkylate only 5% of the
total DNA at the highest tested concentration (16 µM), whereas
compound 6 gave no appreciable adduct with 4GGG (Figure
1C). These data indicate that n ) 2 is the best linker length
among bis-substituted NDIs. Additionally, compound 5a (amine
analogue of 5) did not form any alkylation product, confirming
that only the quaternary ammonium salt derivative is active in
the test conditions (Figure 1C).
Drug Alkylation as a Function of DNA Folding. Selective
recognition of G-quadruplex DNA by 5 was assessed by
comparing its reactivity toward 4GGG with (i) single-stranded
DNA exhibiting the same base composition of 4GGG DNA but
unable to fold into a G-4 conformation (ss-scrambled-4GGG)
and (ii) ss-scrambled-4GGG annealed to its complementary
oligo (ds-scrambled-4GGG) (Figure 1A). Double-stranded DNA
turned out to be the poorest substrate for 5 adduct formation.
Interestingly, comparison of reactivity toward G-4 folded and
ss-scrambled oligos showed a concentration window where
alkylation occurs more efficiently on the G-4 folded DNA
(Figure 1B). Indeed, at concentration below 1 µM, 5 was able
to form a covalent adduct with 4GGG DNA only (lanes 2-4,
a Reaction time and temperature; CH3CN/H2O
) 1:1, [4-6] )
10-3M, [HNu] ) 10-2 M. b Unbuffered. c Buffered at pH 7.0. d Buffered
at pH 8.5.
To unequivocally clarify the nature of the electrophilic
intermediate involved in the alkylation process, we ran additional
trapping experiments in the presence of EVE (ethyl vinyl ether),
in buffered aqueous acetonitrile (1:1). Compound 5 gave 20
(58%, yield) as the main product in the presence of the hydration
adduct 7 (38%). The formation of the substituted 2-ethoxychro-
man 20 is only possible through the intermediacy of two 1,4-
dipolar species trapped by EVE in hetero-Diels-Alder [4 + 2]
cycloaddition reactions (Scheme 4). Therefore, the efficient
generation of 20 in aqueous acetonitrile provides conclusive
evidence for the involvement of two sequential quinone methides
such as QM5.
Alkylation of DNA Folded into a G-Quadruplex. The initial
(23) Ambrus, A.; Chen, D.; Dai, J.; Bialis, T.; Jones, R. A.; Yang, D.
assessment of the ability of compounds 4-6 to alkylate DNA
Nucleic Acids Res. 2006, 34, 2723–2735.
9
J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009 13135

A R T I C L E S
Di Antonio et al.
Figure 1. Specificity of NDI-mediated alkylation. (A) Increasing concentrations of compound 5 (0.25, 0.5, 1.0, 2.0, 4.0, 8.0, 16.0 µM) were incubated with
labeled 4GGG (G-4), ss-scrambled-4GGG (ss), or ds-scrambled-4GGG (ds) DNA for 24 h at 40 °C in Li3PO4 10 mM, KCl 50 mM, pH 7.4. Alkylated DNA
(ALK) was separated from nonreacted oligo (NR) in 20% denaturing polyacrylamide gel. “C” refers to labeled 4GGG DNA without compound 5. (B)
Quantification of the specificity of the alkylation experiment reported in panel A. (C) Quantification of alkylation of G-4 DNA by compounds 5a, 4, 5, 6.
Increasing compound concentrations (0.25, 0.5, 1.0, 2.0, 4.0, 8.0, 16.0 µM) were reacted with 4GGG DNA for 24 h at 40 °C in Li3PO4 10 mM, KCl 50 mM,
pH 7.4. Adducts (ALK) were separated from nonreacted DNA (NR) by 20% denaturing polyacrylamide gel.
Figure 1A and B). Conversely, ss-stranded 4GGG was alkylated
at higher concentrations (2-16 µM) (lanes 13-16, Figure 1A
and B). This clearly suggests a modulation of drug reactivity
upon DNA folding.
To evaluate if this behavior derives from a selective
drug-DNA interaction, competition experiments were per-
formed. Thus, a constant concentration of compound 5 was
incubated with a fixed amount of 32P 5′-end labeled 4GGG DNA
in the presence of increasing concentrations of cold 4GGG, ss-
scrambled-4GGG, or ds-scrambled-4GGG DNA. As shown in
Figure 2, cold 4GGG DNA was able to effectively compete for
adduct formation (adduct amounts decreased to less than 2%
of total labeled DNA in the presence of a 10-fold excess of
4GGG DNA competitor) (lane 5, Figure 2A and B). On the
contrary, the alkylated adduct was only modestly affected with
ss and ds-scrambled-4GGG competitor DNAs (Figure 2). These
data clearly demonstrate that compound 5 can specifically
alkylate G-4 conformed DNA.
The apparent selectivity of compounds 5 toward differently
conformed DNAs depends on both reversible and irreversible
target recognition. Whereas experiments below (see G-quadru-
Figure 2. Competition of NDI-mediated alkylation. (A) Compound 5 (2
µM) was incubated with labeled 4GGG (3.9 pmol) in the presence of
increasing molar ratios (0.5, 1, 2, 5, 10) of cold 4GGG (G-4), ss-scrambled-
plex folding induction versus alkylation chapter) clarify their
interdependency, different techniques other than electrophoretic
methods are warranted to define reversible binding constants
(Kd) and the orientation of the electrophilic center with respect
to the alkylation site.
4GGG (ss), or ds-scrambled-4GGG (ds) DNA for 24 h at 40 °C. Alkylated
DNA (ALK) was separated from nonreacted oligo (NR) in 20% denaturing
gel. “C” refers to labeled 4GGG (3.9 pmol) treated at 40 °C for 24 h, and
“R” refers to the 4GGG-compound 5 reaction product in the absence of
any competitor DNA. (B) Quantification of the competition experiment in
panel A. Adduct amount obtained in sample R was taken as 100%, and
quantification of adduct amount in lanes 1-15 was calculated as relative
percent.
G-Quadruplex Folding Induction versus Alkylation. To
clarify how this selectivity is related to the DNA recognition
step, we used a 32P 5′-end labeled synthetic oligonucleotide
containing only two repeats of the human telomeric sequence,
5′-TACAGATAGTTAGGGTTAGGGTTA-3′ (2GGG). This
oligo can fold into G-4 arrangements only by intermolecular
pairing leading to dimeric or tetrameric species resolvable by
native gel electrophoresis. This process is poorly efficient in
the absence of a G-4 stabilizer, thus allowing us to compare
the ability of compound 5 to induce G-4 folding and alkylation
on the same DNA sequence. DNA-ligand incubation was
performed using alkylating (40 °C) and nonalkylating (20 °C)
conditions in the presence/absence of 100 mM K+, to promote
G-4 conformation. At variable time, samples were loaded onto
9
13136 J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009

Selective G-Quadruplex Alkylating Agents
A R T I C L E S
Figure 4. CD spectra of 4 µM 4GGG incubated for 24 h alone (solid line)
and in the presence of 26 µM 5 at 25 °C (dotted line) or 40 °C (dashed
line) in Li3PO4 10 mM, KCl 50 mM, pH 7.4.
in the presence of 5 (Figure 4). In our experimental conditions,
4GGG preferentially folds into a mixed type G-4 structure.1a,24,25
Upon 24 h incubation with 5 at 20 °C, hence in the absence of
alkylation events, modest changes occurred in the DNA dichroic
spectra that represent the sum of DNA folding rearrangements
and ligand-induced CD. It is interesting to note that after 24 h
of incubation at 40 °C, a larger modulation of the dichroic
response occurred. In particular, a significant increment of the
265 nm band associated with a reduction of the 295 nm one
emerged. This suggests that DNA can undergo a conformational
rearrangement in the presence of 5. In line with the gel shift
data, the modest level of alkylation obtained in these conditions
should not justify the extensive conformational change presented
in Figure 4. Hence, reversible interaction between derivative 5
and G-4 represents the main responsible for the observed
conformational changes.
Alkylated G-Quadruplex Properties. With the aim to identify
a potential selective alkylation site along the G-quadruplex
folded 4GGG, we used a exonuclease I assay.26 Adduct
instability at temperatures above 70 °C prevented us from using
standard techniques, such as Taq polymerase stop assay or
thermal depurination at alkylated sites. The 32P 5′-end labeled
alkylated G-4 DNA was gel purified and treated with increasing
amounts of exonuclease I at 50 °C for 30 min, alongside the
nonalkylated G-4 DNA treated in the same conditions. As shown
in Figure 5, the enzyme could not process the alkylated G-4
DNA even at the highest tested enzyme amounts, whereas it
was able to digest the nonalkylated oligo, with an efficiency
linearly related to enzyme concentrations.
Figure 3. Comparison of induction of G-4 conformation and alkylation
by NDI 5. 2GGG DNA was incubated in the presence (+) or absence (-)
of compound 5 (G-4/5 molar ratio ) 0.04) for variable times at 20 or 40
°C. (A) The monomeric DNA (M) was separated from the dimeric G-4
DNA (D) by 16% native polyacrylamide gel. (B) Quantification data. Dimer
formation is reported as a function of DNA alkylation. Values were taken
at 0 (circles), 3 (rectangles), 6 (squares), and 24 (diamonds) h time intervals
after incubation at 20 °C (open symbols) or 40 °C (filled symbols). Dimer
formation and DNA alkylation were calculated as percent over total DNA
obtained as integration values of electrophoretic bands from native (A) and
denaturing (not shown) gels, respectively.
16% native and 20% denaturing polyacrylamide gels, to check
for folding stabilization and alkylation, respectively.
According to previous results with NDI derivatives,8 using
reversible conditions, 5 was poorly effective in inducing G-4
structures. Indeed, incubation of 2GGG with 25 µM 5 at 20 °C
induced only 18% of G-4 dimer at 24 h (lanes 1-6, Figure
3A). In these incubation conditions, the presence of KCl was
demanding; indeed, at 20 °C no quadruplex stabilization was
observed in the absence of KCl. Correspondingly, negligible
alkylation occurred at this incubating temperature (data not
shown).
Remarkably, the G-4 folded dimeric conformation was
induced by 5 upon incubation at 40 °C (lanes 7-12, Figure
3A). Denaturing gel showed that in these conditions 2GGG
alkylation products were detectable after 6 and 24 h of
incubation (not shown). Direct comparison of G-4 folding
induction and alkylation indicates that the amount of DNA
adduct is linearly related to the extent of G-4 folding. However,
since compound 5 induces a larger amount of G-4 structure than
it undergoes alkylation, we infer that G-4 folded DNA stabiliza-
tion occurs prior to and is a prerequisite for DNA alkylation,
which in turn irreversibly stabilizes the NDI-G-4 complex.
CD Spectroscopic Study. The above conclusion is supported
by an analysis of the chiroptical properties of 4GGG alone and
The fact that no enzyme stop positions were observed likely
reflects a lack of DNA-enzyme recognition. On the basis of the
data obtained by circular dichroism studies, this behavior is
probably related to the structural rearrangement occurring on
the 4GGG bound by compound 5, which prevents anchoring/
processing by exonuclease I. However, we cannot exclude a
3′-end DNA alkylation event which would result in a similar
experimental outcome.
(24) (a) Smargiasso, N.; Rosu, F.; Hsia, W.; Colson, P.; Baker, E. S.;
Bowers, M. T.; De Pauw, E.; Gabelica, V. J. Am. Chem. Soc. 2008,
130, 10208–10216. (b) Prislan, I.; Lah, J.; Vesnaver, G. J. Am. Chem.
Soc. 2008, 130, 14161–14169. (c) Wei, C.; Wang, L.; Jia, G.; Zhou,
J.; Han, G.; Li, C. Biophys Chem 2009, 143, 79–84. (d) Li, H.; Liu,
Y.; Lin, S.; Yuan, G. Chemistry 2009, 15, 2445–2452. (e) Gray, D. M.;
Wen, J.-D.; Gray, C. W.; Repges, R.; Repges, C.; Raabe, G.;
Fleischhauer, J. Chirality 2008, 20, 431–440. (f) Zhou, J.; Yuan, G.
Chemistry 2007, 13, 5018–5023.
Stabilization of G-4 conformation upon compound 5 alky-
lation was further supported by CD melting analysis, according
(25) Paramasivan, S.; Rujan, I.; Bolton, P. H. Methods 2007, 43, 324–
331.
(26) Yao, Y.; Wang, Q.; Hao, Y.-H.; Tan, Z. Nucleic Acids Res. 2007, 35,
e68/1-e68/9.
9
J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009 13137

A R T I C L E S
Di Antonio et al.
Figure 5. Digestion of alkylated and nonalkylated G-4 DNA by exonuclease I. Labeled 4GGG DNA (G4) and the purified labeled 5-alkylated 4GGG DNA
(ALK G4) were incubated with increasing amounts of Exonuclease I (ExoI) (10, 20, and 40 units) for 30 min at 50 °C. Cleavage products were analyzed
in 20% denaturing polyacrylamide gel. “M” indicates purine markers run aside samples. Densitometry profiles referring to lanes 5 and 9 are shown on the
right.
to which a Tm above 75 °C was observed (data not shown). As
already mentioned, however, above 70 °C the alkylated adduct
is not stable (see Figure 1S, Supporting Information), and hence
the observed Tm value reflects a number of processes (i.e., G-4
folding/unfolding, reversible NDI association/dissociation to/
from DNA, and covalent bond formation/disruption) which take
place concurrently at high temperatures.
NDIs Cytotoxicity. Cytotoxic effects of NDI derivatives 4,
5, and 6 were investigated using the human embryonic kidney
293T cell line. Cells were exposed to increasing concentra-
tions of the tested compounds (12 nM-40 µM) at 37 °C for
48 h, after which cell damage was assessed by MTT assay.
The effective drug concentration able to kill 50% of cell
population after drug exposure (EC50) was 4.5 ( 0.3 µM,
10.5 ( 3.5 µM, and >40 µM for compounds 5, 4, and 6,
respectively. Compound 5b, which was also included as a
control ligand that completely lacks the ability to generate
QM, displayed EC50 > 40 µM. These data indicate that the
cytotoxic potency of these compounds depends on and
parallels their DNA alkylating ability.
alkylating moieties by flexible spacers. We have shown that
this strategy is effective in the enhancement of the G-4
folding induction and stabilization, appearing to be a promis-
ing tool for probing G-4 structures.
Experimental Section
N,N′-Bis(3-dimethylaminomethyl-4-hydroxybenzyl)-1,4,5,8-naph-
thalenetetracarboxylic Acid Diimide (4a). A 3.2 g portion of
4-aminomethyl-2-dimethylaminomethyl phenol dihydrochloride
(12.6 mmol) and 1.61 g (6.0 mmol) of 1,4,5,8 tetracarboxylic
dianhydride were suspended in 20 mL of a mixture dioxane/DMF
9:1. Under nitrogen atmosphere, 1 mL of TEA was added to the
suspension, and the mixture was allowed to reflux with vigorously
stirring for 5 h. The reaction mixture was cooled at room
temperature and poured into 30 mL of water. The resulting
suspension was filtered and washed with water and cold EtOH
anhydrous. The 4a was obtained as a pale yellow solid. Yield 91%.
1
Mp > 350 °C. H NMR (300 MHz, CDCl3, 25 °C, TMS): δ 8.80
(s, 4H), 7.40 (d, 2H, J ) 6.9 Hz), 7.20 (s, 2H), 6.85 (d, 2H, J )
6.9 Hz), 5.20 (s, 4H), 3.60 (s, 4H), 2.30 (s, 12H). 13C NMR (300
MHz, CDCl3, 25 °C, TMS): δ 162.74; 157.76; 131.03; 130.95;
131.00; 129.55; 126.87; 126.61; 121.68; 115.91; 62.60; 44.37;
43.41. Anal. Calcd for C34H32N4O6: C, 68.91; H, 5.44; N, 9.45; O,
16.20. Found: C, 68.95; H, 5.40; N, 9.47.
Conclusion
In summary, we have described the first example of
thermally induced non-metal-based G-quadruplex adduct
formation, where a reversible binding process is associated
with selective alkylation and stabilization of the G-4. This
goal has been achieved using novel ligand/alkylating hybrid
structures exhibiting a NDI core tethered to activatable
4. A 1.7 g (2.9 mmol) portion of the bisimide 4a was suspended
in 50 mL of CH3CN and 1.2 g (8.5 mmol) of CH3I was added.
This suspension was refluxed under nitrogen atmosphere and in a
few minutes the reaction mixture turned dark red. After heating
3 h at 80 °C, the reaction was cooled at room temperature. Red
solid formation was observed after addition of Et2O (50 mL) to
9
13138 J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009

Selective G-Quadruplex Alkylating Agents
A R T I C L E S
the reaction mixture. The suspension was filtered and washed with
CH3CN to give 2.7 g (3.0 mmol) of 4 as a red-brownish solid. Mp
> 350 °C. Yield 92%. 1H NMR (300 MHz, DMSO-d6, 25 °C, TMS):
δ 10.30 (s, 2H), 8.70 (s, 4H), 7.50 (s, 2H), 7.40 (d, 2H, J ) 8.4
Hz), 6.90 (d, 2H, J ) 8.4 Hz), 5.15 (s, 4H), 4.45 (s, 4H), 3.05 (s,
18H). 13C NMR (DMSO-d6, 25 °C, TMS): 162.72; 156.59; 134.88;
132.07; 130.71; 127.63; 126.38; 126.28; 116.09; 114.39; 62.91;
51.90; 42.7. Anal. Calcd for C36H38I2N4O6: C, 49.33; H, 4.37; I,
28.96; N, 6.39; O, 10.95. Found: C, 49.35; H, 4.41; I, 28.93; N,
6.42.
C40H46I2N4O6: C, 51.51; H, 4.97; I, 27.21; N, 6.01; O, 10.29 Found:
C, 51.46; H, 4.93; I, 27.19; N, 6.06
Activation of the Bis-alkylating Properties by Mild Thermal
Digestion. General Procedures. Base Catalysis General
Method. To trap the nucleophiles by mild thermal digestion the
following procedures were used:
Procedure A (for basic nucleophiles: pKa > 8). To 50 mL of
a solution 5 × 10-3 M of the bisimides (4-6) in 1:1 CH3CN/H2O,
outgassed with a N2 flux, was added the nucleophile (5 × 10-2
M). This solution is allowed to stand at 40 °C and after 0.5 h a
pale yellow precipitate starts to form. The reaction mixture was
kept at 40 °C for 4 h. After this time the CH3CN was evaporated
under vacuum, and the suspension was cooled at 4 °C for 12 h.
The bis-alkylated adduct formed was filtered and washed with cold
anhydrous EtOH to provide the product in good yield (60-85%).
Procedure B (for non basic nucleophiles: pKb < 8). To 50 mL
of a solution 5 × 10-3 M of the bisimides (4-6) in 1:1 CH3CN/
N,N′-Bis[2-(3-dimethylaminomethyl-4-hydroxyphenyl)-
ethyl]-1,4,5,8-naphthalenetetracarboxylic Acid Diimide (5a).
The same protocol used for the synthesis of 4a was followed; 2.1 g
of N-(3-((dimethylamino)methyl)-4-hydroxyphenylethyl)amine di-
hydrochloride (8.3 mmol) and 1.07 g (4.0 mmol) of 1,4,5,8
tetracarboxylic dianhydride were used. The 5a was obtained as a
pale yellow solid, 2.20 g (yield 89%). Mp > 350 °C 1H NMR (300
MHz, CDCl3, 25 °C, TMS): δ 8.70 (s, 4H), 7.20 (d, 2H, J ) 7.4
Hz), 7.00 (s, 2H), 6.80 (d, 2H, J ) 7.4 Hz), 4.40 (t, 4H, J ) 8.1
Hz), 3.75 (s, 4H), 2.90 (t, 4H, J ) 8.1 Hz), 2.35 (s, 12H). 13C
NMR (DMSO-d6, 25 °C, TMS): δ 162.46; 155.08; 132.92; 131.29;
130.48; 129.00; 126.27; 126.12; 116.34; 115.71; 54.54; 45.15;
32.43; 30.78. Anal. Calcd for C36H36N4O6: C, 69.66; H, 5.85; N,
9.03; O, 15.47. Found: C, 69.62; H, 5.88; N, 9.00.
2-
H2O buffered to pH 8.5 with a H2PO4-/HPO4 buffer, outgassed
with a N2 flux, was added the nucleophile (5 × 10-2 M). This
solution was heated at 40 °C for 12 h. After this time the CH3CN
was evaporated under vacuum and the suspension was kept at 4
°C for 12 h. The bis-alkylated adduct precipitated was filtered and
washed with cold water and anhydrous EtOH to provide the product
in fairly good yield (55-70%)
5. The same protocol used for the synthesis of 4 was followed;
2.0 g (3.2 mmol) of the bisimide 5a was used and 2.7 g (3.0 mmol)
of 5 as a red-brownish solid was obtained. Mp > 350 °C. Yield
1
Adduct 7. Pale yellow needles. Mp > 350 °C. H NMR (300
MHz, DMSO-d6, 25 °C, TMS): δ 10.20 (s, 2H), 8.70 (s, 4H), 7.30
(s, 2H), 7.20 (d, 2H, J ) 7.9 Hz), 6.90 (d, 2H, J ) 7.9 Hz), 4.30
(s, 4H), 4.20 (t, 4H); 2.9 (t, 4H).13C NMR (DMSO-d6, 25 °C, TMS):
δ 162.48; 155.80; 134.71; 132.37; 130.47; 129.30; 126.28; 126.13;
116.16; 114.63; 63.11; 51.99; 41.60; 32.41. Anal. Calcd for
C32H26N2O8: C, 67.84; H, 4.63; N, 4.94; O, 22.59. Found: C, 67.81;
H, 4.66; N, 4.95.
1
84%. H NMR (300 MHz, DMSO-d6, 25 °C, TMS): δ 10.20 (s,
2H), 8.80 (s, 4H), 7.30 (s, 2H), 7.20 (d, 2H, J ) 7.6 Hz), 6.90 (d,
2H, J ) 7.6 Hz), 4.40 (s, 4H), 4.20 (t, 4H, J ) 8.1 Hz), 3.00 (s,
18H), 2.85 (t, 4H, J ) 8.1 Hz). 13C NMR (DMSO-d6, 25 °C, TMS):
δ 162.50; 155.80; 134.72; 132.38; 130.47; 129.28; 126.30; 126.14;
116.15; 114.63; 63.06; 51.95; 41.60; 32.41. Anal. Calcd for
C38H42I2N4O6: C, 50.46; H, 4.68; I, 28.06; N, 6.19; O, 10.61. Found:
C, 50.41; H, 4.71; I, 28.01; N, 6.16.
Adduct 8. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.80 (s, 4H), 7.10 (d, 2H, J ) 8.1 Hz),
7.00 (s, 2H), 6.75 (d, 2H, J ) 8.1 Hz), 4.40 (t, 4H, J ) 7.8 Hz),
3.75 (s, 4H); 2.95 (t, 4H, J ) 7.8 Hz); 2.60 (q, 8H, J ) 7.1 Hz);
1.10 (t, 12H, J ) 7.1 Hz). 13C NMR (300 MHz, CDCl3, 25 °C,
TMS): δ 162.58; 156.88; 132.90; 130.82; 128.78; 128.21; 126.52;
126.42; 122.11; 115.93; 56.80; 46.18; 42.38; 33.24; 11.12. Anal.
Calcd for C40H44N4O6: C, 70.99; H, 6.55; N, 8.28; O, 14.18. Found:
C, 71.03; H, 6.51; N, 8.33.
N,N′-Bis[2-(4-hydroxyphenyl)ethyl]-1,4,5,8-naphthalene-
tetracarboxylic Acid Diimide (5b). The same protocol used for
the synthesis of 4a was followed; 1.0 g of Tyramine (7.3 mmol)
and 0.84 g (3.0 mmol) of 1,4,5,8-tetracarboxylic dianhydride was
used. The 5b was obtained as a yellow solid, 1.50 g (yield 99%).
1
Mp > 350 °C. H NMR (300 MHz, DMSO-d6, 25 °C, TMS) δ
1
9.20 (s, 2H), 8.70 (s, 4H), 7.10 (d, 4H, J ) 7.9 Hz), 6.80 (d, 2H,
J ) 7.9 Hz), 4.20 (t, 4H, J ) 8.1 Hz), 2.80 (s, 4H). 13C NMR
(DMSO-d6, 25 °C, TMS): δ 162.63; 156.62; 130.55; 129.51; 127.15;
126.31; 115.05; 45.35; 32.01. Anal. Calcd for C30H22N2O6 C, 71.14;
H, 4.38; N, 5.53; O, 18.95 Found: C, 71.55; H, 4.58; N, 5.31.
Adduct 9. Pale yellow needles. Mp > 350 °C. H NMR (300
MHz, CDCl3, 25 °C, TMS): δ 8.90 (s, 4H), 7.15 (d, 2H, J ) 7.9
Hz), 7.00 (s, 2H), 6.90 (d, 2H, J ) 7.9 Hz), 4.40 (t, 4H, J ) 7.4
Hz), 3.75 (s, 4H); 2.95 (t, 4H, J ) 7.4 Hz); 2.60 (m, 8H); 1.50 (q,
4H, J ) 6.5 Hz); 1.25 (q, 4H, J ) 6.1 Hz); 1.1 (t, 6H, J ) 6.1 Hz);
0.90 (t, 6H, J ) 6.5 Hz). 13C NMR (300 MHz, CDCl3, 25 °C, TMS):
δ 162.58; 156.79; 130.83; 128.78; 128.20; 126.52; 122.17; 115.88;
57.39; 52.49; 46.57; 44.38; 42.40; 33.26; 28.54; 20.46; 13.86; 10.84.
Anal. Calcd for C44H52N4O6: C, 72.11; H, 7.15; N, 7.64; O, 13.10.
Found: C, 72.13; H, 7.11; N, 7.68.
N,N′-Bis[3-(3-dimethylaminomethyl-4-hydroxyphenyl)
propyl]-1,4,5,8-naphthalenetetracarboxylic Acid Diimide (6a).
The same protocol used for the synthesis of 4a was followed; 1.17 g
of 4-(3-aminopropyl)-2-dimethylaminomethyl phenol dihydrochlo-
ride (6.01 mmol) and 0.8 g (3.0 mmol) of 1,4,5,8 tetracarboxylic
dianhydride were used. The 6a was obtained as a pale yellow solid,
Adduct 10. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS):δ 8.80 (s, 4H), 7.10 (d, 2H, J ) 7.5 Hz),
7.00 (s, 2H), 6.75 (d, 2H, J ) 7.5 Hz), 4.45 (t, 4H, J ) 7.4 Hz),
3.60 (t, 8H, J ) 7.1 Hz); 3.00 (t, 8H, J ) 7.1 Hz), 2.90 (t, 4H, J
) 7.4 Hz). 13C NMR (300 MHz, CDCl3, 25 °C, TMS): δ 163.12;
156.34; 130.36; 129.72; 128.47; 126.98; 126.20; 122.79; 122.17;
115.88; 66.13; 58.01; 52.49; 46.57; 33.26. Anal. Calcd for
C40H40N4O8: C, 68.17; H, 5.72; N, 7.95; O, 18.16. Found: C, 68.13;
H, 5.70; N, 7.91.
1
2.2 g (yield 98%). Mp > 350 °C. H NMR (300 MHz, CDCl3, 25
°C, TMS): δ 8.75 (s, 4H), 7.35 (d, 2H, J ) 7.6 Hz), 6.85 (s, 2H),
6.70 (d, 2H, J ) 7.6 Hz), 4.25 (t, 4H, J ) 6.0 Hz), 3.60 (s, 4H),
2.69 (t, 4H, J ) 6.0 Hz), 2.30 (s,12H), 2.10 (t, 4H, J ) 8.0 Hz).
13C NMR (300 MHz, CDCl3, 25 °C, TMS): δ 162.70; 155.99;
130.77; 128.22; 127.93; 126.48; 121.61; 126.12; 115.71; 62.74;
44.38; 40.70; 32.37; 29.44. Anal. Calcd for C38H40N4O6: C, 70.35;
H, 6.21; N, 8.64; O, 14.80 Found: C, 70.37; H, 6.22; N, 8.86.
6. The same protocol used for the synthesis of 4 was followed;
0.90 g (1.4 mmol) of the bisimide 6a was used and 1.06 g of 6 as
Adduct 11. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.6 (s, 4H); 7.20 (d, 2H, J ) 7.7 Hz); 7.00
(s, 2H); 6.75 (d, 2H, J ) 7.7 Hz); 4.45 (t, 4H, J ) 7.8 Hz); 3.60
(s, 4H); 3.0 (t, 4H, J ) 7.8 Hz); 2.4 (m, 8H); 1.8 (m, 8H).13C NMR
(300 MHz, CDCl3, 25 °C, TMS): δ 163.56; 156.60; 130.82; 128.84;
128.28; 128.17; 126.57; 126.50; 122.45; 15.80; 58.66; 53.35; 42.34;
33.22; 23.53. Anal. Calcd for C40H40N4O6: C, 71.41; H, 5.99; N,
8.33; O, 14.27. Found: C, 71.38; H, 6.04; N, 8.30.
1
a red-brownish solid was obtained. Mp > 350 °C. Yield 81%. H
NMR (300 MHz, DMSO-d6, 25 °C, TMS): δ 10.15 (s, 2H), 8.60
(s, 4H), 7.25 (m, 4H), 6.85 (d, 2H, J ) 8.6 Hz), 4.40 (t, 4H), 4.05
(s, 4H), 3.05 (s, 18H), 2.65 (t, 4H); 1.95 (t, 4H). 13C NMR (DMSO-
d6, 25 °C, TMS): 163.07; 157.35; 134.82; 132.60; 130.53; 126.97;
126.63; 126.34; 116.43; 114.84; 62.81; 52.04. Anal. Calcd for
9
J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009 13139

A R T I C L E S
Di Antonio et al.
Adduct 12. Green crystals. Mp > 350 °C: 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.80 (s, 4H); 7.10 (d, 2H, J ) 7.8 Hz),
7.00 (s, 2H), 6.85 (d, 2H, J ) 7.8 Hz); 4.40 (t, 4H, J ) 7.7 Hz);
4.00 (s, 4H); 2.90(t, 4H, J ) 7.7 Hz); 1.20 (s, 18H). 13C NMR
(300 MHz, CDCl3, 25 °C, TMS): δ 162.58; 157.03; 130.85; 128.81;
128.39; 128.34; 126.53; 123.54; 116.43; 50.98; 45.95; 42.45; 33.29;
28.47. Anal. Calcd for C40H44N4O6:C, 70.99; H, 6.55; N, 8.28; O,
14.18. Found: C, 70.92; H, 6.58; N, 8.22.
4.40 (t, 4H, J ) 8.0 Hz), 3.90 (m, 2H), 3.60 (m, 2H), 3.00 (t, 4H,
J ) 8.0 Hz), 2.60 (m, 2H), 2.00 (m, 2H), 1.30 (t, 6H, J ) 7.0 Hz).
13C NMR (300 MHz, CDCl3, 25 °C, TMS): δ 162.59; 150.8;
130.85; 130.10; 129.60; 127.66; 126.56; 122.50; 116.91; 96.81;
63.55; 42.40; 33.35; 30.80; 26.43; 20.4; 15.02. Anal. Calcd for
C40H38N2O8: C, 71.20; H, 5.68; N, 4.15; O, 18.97. Found: C, 71.25;
H, 5.77; N, 4.11.
G-Quadruplex DNA Alkylation and Folding Induction. For
reaction with DNA, compounds were dissolved and diluted in
DMSO, in order to obtain a final 2.5% DMSO. DNA sequences
(Biosense) were 4GGG, 5′-AGGGTTAGGGTTAGGGTTAGGG-
3′; ss-scrambled-4GGG, 5′-GGATGTGAGTGTGAGTGTGAG-3′
and its complementary strand 5′-CTCACACTCACACTCACATCC-
3′ to form ds-scrambled-4GGG; 2GGG, 5′-TACAGATAGT-
TAGGGTTAGGGTTA-3′. For DNA labeling reactions, 100 pmol
of DNA was incubated with 1 µL (10 µCi/µL) of [γ-32P]ATP and
10 units of T4 polynucleotide kinase (Fermentas) in 50 mM TRIS-
HCl (pH 7.6), 10 mM MgCl2, 5 mM DTT, 0.1 mM spermidine,
and 0.1 mM EDTA at 37 °C for 30 min. After incubation, DNA
was purified through MicroSpin G-25 columns (GE Healthcare).
For the alkylation experiment, 5-10 pmol of 32P-labeled DNA was
incubated with increasing drug amounts in phosphate buffer pH
7.4, 10 mM LiOH, and 50 mM KCl. Samples were incubated at
the indicated temperature and time and reactions were stopped by
precipitation. For competition experiments, annealed-labeled 4GGG
(3.9 pmol) and drug (2 µM 5) were mixed with cold competitor-
DNA (4GGG, ss-scrambled-, and ds-scrambled-4GGG) at increas-
ing concentrations, so that the molar ratios competitor/labeled
4GGG were 0.5, 1, 2, 5, and 10. After 24 h of incubation at 40 °C,
samples were precipitated. Alkylation products were then resolved
in 20% polyacrylamide 7 M urea sequencing gels. The NDI’s ability
to induce the G-quadruplex structure was tested by gel shift
experiments. For G-4 folding induction, 32P-labeled 2GGG and 5
µM cold 2GGG were annealed in phosphate buffer pH 7.4,
containing 10 mM LiOH with or without 50 mM KCl. The annealed
2GGG was mixed with 25 µM of drug for increasing times at the
indicated temperatures. Samples were loaded onto 16% polyacry-
lamide gel containing 20 mM KCl.
Adduct 13. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
DMSO-d6, 25 °C, TMS): δ 10.20 (s, 2H), 8.80 (s, 4H); 7.20 (d,
2H, J ) 7.9 Hz), 7.0 (s, 2H), 6.85 (d, 2H, J ) 7.9 Hz); 4.45 (s,
4H), 4.30 (t, 4H, J ) 7.6 Hz); 3.15 (s, 18H), 3.00 (t, 4H, J ) 7.6
Hz). 13C NMR (DMSO-d6, 25 °C, TMS): δ 162.79; 157.12; 130.82;
128.81; 128.39; 128.34; 126.93; 126.51; 123.54; 116.43; 60.21;
45.95; 42.41; 38.29; 33.47 Anal. Calcd for C40H42N2O6S2: C, 67.58;
H, 5.95; N, 3.94; O, 13.50; S, 9.02. Found: C, 67.51; H, 5.98; N,
3.92; S, 9.00
Adduct 14. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
DMSO-d6, 25 °C, TMS): δ 10.00 (s, 2H), 8.70 (s, 4H), 7.30(s,
2H), 7.20 (d, 2H, J ) 7.8 Hz), 6.90 (d, 2H, J ) 7.8 Hz), 5.1 (s,
4H), 4.30 (s, 4H). 13C NMR (DMSO-d6, 25 °C, TMS): δ 162.48;
155.80; 134.71; 132.37; 130.47; 129.30; 126.28; 126.13; 121.16;
115.93; 63.11; 62.58. Anal. Calcd for C30H22N2O8: C, 66.91; H,
4.12; N, 5.20; O, 23.77. Found: C, 67.01; H, 4.17; N, 5.26.
Adduct 15. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.70 (s, 4H), 7.30 (s, 2H), 7.00 (d, 2H, J
) 7.6 Hz), 6.90 (d, 2H, J ) 7.6 Hz), 5.1 (s, 4H), 4.20 (s, 4H) 2.60
(q, 8H, J ) 7.1 Hz); 1.1 (t, 12H, J ) 7.1 Hz). 13C NMR (300
MHz, CDCl3, 25 °C, TMS): δ 163.70; 155.99; 131.29; 130.77;
128.22; 127.93; 126.48; 121.61; 115.71; 62.74; 44.38; 40.70; 32.37;
29.44. Anal. Calcd for C38H40N4O6: C, 70.35; H, 6.21; N, 8.64; O,
14.80. Found: C, 70.33; H, 6.27; N, 8.62.
Adduct 16. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
DMSO-d6, 25 °C, TMS): δ 10.80 (s, 2H), 8.80 (s, 4H), 7.50(s,
2H), 7.40 (d, 2H, J ) 8.8 Hz), 7.20 (d, 2H, J ) 8.8 Hz), 5.1 (s,
4H), 4.30 (s, 4H), 3.00(s, 18H). 13C NMR (DMSO-d6, 25 °C, TMS):
δ 163.44; 155.98; 132.55; 131.59; 130.99; 126.96; 126.72; 126.21;
116.02; 115.76; 62.18; 60.26; 45.78; 38.30. Anal. Calcd for
C38H38N2O6S2: C, 66.84; H, 5.61; N, 4.10; O, 14.06; S, 9.39. Found:
C, 66.81; H, 5.55; N, 4.06; S, 9.34.
Exonuclease I Digestion. For exonuclease I reaction, the
alkylation product was resolved in 20% polyacrylamide 7 M urea
gel and purified by cutting and crushing the band of interest and
eluting the alkylated oligo in TE buffer for 2 h. The purified
alkyalted oligo was then precipitated and resuspended in exonu-
clease I reaction buffer. Exonuclease I was added at 10, 20, and
40 units and the reaction was incubated for 30 min at 37 or 50 °C.
The reaction was stopped by DNA precipitation and samples were
loaded onto 20% polyacrylamide 7 M urea gel.
Adduct 17. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
DMSO-d6, 25 °C, TMS): δ 10.15 (s, 2H), 8.60 (s, 4H), 7.25 (d,
2H, J ) 8.0 Hz), 7.0 (s, 2H); 6.85 (d, 2H, J ) 8.0 Hz), 4.40 (t, 2H,
J ) 7.5 Hz); 4.30 (s, 4H); 2.65 (t, 2H, J ) 7.5 Hz); 2.00 (m, 4H).
13C NMR (DMSO-d6, 25 °C, TMS): δ 163.23; 158.35; 134.82;
132.60; 130.53; 126.97; 126.63; 126.34; 116.43; 114.84; 62.81;
52.04; 33.25; 29.07. Anal. Calcd for C34H30N2O8:C, 68.68; H, 5.09;
N, 4.71; O, 21.53. Found: C, 68.71; H, 5.11; N, 4.74.
Adduct 18. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.70 (s, 4H), 7.10 (d, 2H, J ) 7.9 Hz),
7.00 (s, 2H), 6.90 (d, 2H, J ) 7.9 Hz), 4.40 (t, 4H, J ) 7.6 Hz),
3.70 (s, 4H); 2.95 (t, 4H, J ) 7.6 Hz); 2.60 (q, 8H, J ) 7.1 Hz);
2,10 (t, 4H, J ) 7.6 Hz) 1.1 (t, 12H, J ) 7.1 Hz). 13C NMR (300
MHz, CDCl3, 25 °C, TMS): δ 162.17; 155.17; 132.91; 130.45;
128.79; 128.31; 126.55; 126.42; 122.09; 116.13; 57.39; 46.57;
44.38; 33.26; 28.54; 20.46. Anal. Calcd for: C42H48N4O6 C, 71.57;
H, 6.86; N, 7.95; O, 13.62. Found: C, 71.52; H, 6.88; N, 7.89.
Adduct 19. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
DMSO-d6, 25 °C, TMS): δ 10.10 (s, 2H), 8.70 (s, 4H), 7.00 (d,
2H, J ) 7.7 Hz), 6.90 (s, 2H), 6.90 (d, 2H, J ) 7.7 Hz), 4.25 (t,
4H, J ) 6.0 Hz), 3,60 (s, 4H), 2.69 (t, 4H, J ) 6.0 Hz), 2.00 (s,
18H), 2.10 (t, 4H, J ) 8.0 Hz). 13C NMR (DMSO-d6, 25 °C, TMS):
δ 163.04; 156.42; 132.46; 131.43; 130.56; 126.92; 126.62; 126.31;
117.75; 115.99; 62.90; 61.46; 32.37; 29.44; 22.47; 18.34. Anal.
Calcd for C42H46N2O6S2: C, 68.27; H, 6.27; N, 3.79; O, 12.99; S,
8.68. Found: C, 68.22; H, 6.21; N, 3.82; S, 8.69.
Circular Dichroism Measurements. ligonucleotide 4GGG
circular dichroism spectra from 230 to 350 nm were recorded
using 10 mm path length cells on a Jasco J 810 spectropola-
rimeter equipped with a NESLAB temperature controller. Before
data acquisition, a 4 µM 4GGG solution in Li3PO4 10 mM, KCl
50 mM, pH 7.4, was prepared, heated at 95 °C for 5 min, and
left to cool at room temperature overnight. This solution was
divided into four fractions: two were incubated for 24 h at 25
and 40 °C, respectively; 5 was added to a final concentration of
26 µM to the remaining fractions and incubated for 24 h at 25
and 40 °C, respectively. The reported spectrum of each sample
represents the average of 3 scans recorded with 1-nm step
resolution at 25 °C. Thermal denaturation experiments were
performed recording the CD signal at 265 nm while increasing
the temperature at 0.8 °C/min and stirring the DNA solution to
allow equilibration.
Cytotoxicity Assay. Human embryonic kidney 293T cells
(293T) were purchased from ATCC (ATCC number CRL-
11268). 293T cells were grown as monolayers in Dulbecco’s
Modified Eagle Medium (Invitrogen, Italy) with 10% fetal bovine
serum (FBS) supplemented with penicillin (100 U/mL) and
Adduct 20. Yellow needles. Mp > 350 °C. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.70 (s, 4H), 7.20 (d, 2H, J ) 8.2 Hz),
7.10 (s, 2H), 6.75 (d, 2H, J ) 8.2 Hz), 5.25 (dd, 2H, J ) 2.7 Hz),
9
13140 J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009

Selective G-Quadruplex Alkylating Agents
A R T I C L E S
streptomycin (100 µg/mL) in a humidified atmosphere with 5%
CO2 at 37 °C.
individual experiments conducted in triplicate. The percentage of
cell survival was calculated as follows: cell survival ) (Awell
-
Cytotoxic effects on cell growth were determined by MTT assay.
NDI compounds were dissolved and diluted into working concen-
trations with DMSO. Cells (1.75 × 104 cells/well) were plated onto
96-microwell plates to a final volume of 100 µL and allowed an
overnight period for attachment. At day 1, 1 µL of each dilution of
tested compounds was added per well in order to get a 1% final
concentration of drug solvent per well; at day 2, medium was
removed, cells were washed with PBS, and fresh medium was
added. Control cells (without any compound but with 1% drug
solvent) were treated in the exact same conditions. Cell survival
was evaluated by MTT assay: 10 µL of a freshly dissolved solution
of MTT (5 mg/mL in PBS) was added to each well, and after 4 h
of incubation, 100 µL of solubilization solution (10% SDS, 0,01
M HCl) was added. After overnight incubation at 37 °C, absorbance
was read at 540 nm. Data were expressed as mean values of three
Ablank)/(Acontrol - Ablank) × 100, where blank denotes medium without
cells.
Acknowledgment. This work was supported by MIUR, Grant
FIRB-Ideas RBID082ATK, and Grant AIRC 5826 (Associazione
Italiana per la Ricerca sul Cancro).
Supporting Information Available: Experimental and char-
1
acterization details, H and 13C NMR spectra for the modified
NDIs 4-6, their related amines 4a-6a and precursors, together
with the NDI-QM alkylated adducts 7-20. This material is
available free of charge via the Internet at http://pubs.acs.org.
JA904876Q
9
J. AM. CHEM. SOC. VOL. 131, NO. 36, 2009 13141