Extended Naphthalene Diimides with Donor/Acceptor Hydrogen-Bonding Properties Targeting G-Quadruplex Nucleic Acids

Article abstract of DOI:10.1002/ejoc.201600757

Naphthalene diimides with one or two centrosymmetric arylethynyl moieties capable of synergic donor and acceptor hydrogen bonding exhibit promising binding properties and selectivity towards parallel G-quadruplex (G4) nucleic acids (c-myc, bcl-2 and parallel hTel22). The hydrogen-bonding network involving the phosphate backbone and outside rim of the G-quartet represents an opportunity to exploit G4 selectivity for extended aromatics.

Full text of DOI:10.1002/ejoc.201600757

DOI: 10.1002/ejoc.201600757
Full Paper
Nucleic Acid Recognition
Extended Naphthalene Diimides with Donor/Acceptor
Hydrogen-Bonding Properties Targeting G-Quadruplex Nucleic
Acids
Filippo Doria,[a][‡] Matteo Nadai,[b][‡] Giosuè Costa,[c] Giovanna Sattin,[b] Caroline Gallati,[a]
Greta Bergamaschi,[a] Federica Moraca,[c] Stefano Alcaro,[c] Mauro Freccero*[a] and
Sara N. Richter*[b]
Abstract: Naphthalene diimides with one or two centrosym- (c-myc, bcl-2 and parallel hTel22). The hydrogen-bonding net-
metric arylethynyl moieties capable of synergic donor and ac- work involving the phosphate backbone and outside rim of the
ceptor hydrogen bonding exhibit promising binding properties G-quartet represents an opportunity to exploit G4 selectivity
and selectivity towards parallel G-quadruplex (G4) nucleic acids for extended aromatics.
Introduction
has been recently provided by an antibody-based ap-
proach[13,14] and by the identification of G4-stabilizing and de-
stabilizing proteins.[15–17] Very recently it was shown that telo-
meric G4-forming sequences represent natural replication fork
barriers.[18] G4-stabilizing molecules can further enhance such a
barrier. In addition, G4s have also been reported in pathogenic
microorganisms such as bacteria[19–22] and viruses,[23–27] which
has broadened the interest in small molecules that target G4s
and could thus exhibit both antibiotic and antiviral potential.[28]
G-quadruplexes (G4s) are three-dimensional four-stranded
structures of nucleic acids that are rich in guanine. G4s consist
of a square arrangement of guanine (G-tetrad) stabilized by
both hydrogen bonding and monovalent cations (preferably K+)
at the centre of the tetrads. They can vary in a number of ways,
including strand stoichiometry and strand orientation.[1]
G4s are involved in several key biological functions. Initially
they were shown to form in vitro[2] and in vivo[3] in the G-rich
sequence of telomeres: stabilization of the G4 folded structure
has been proposed as an effective approach to inhibiting telo-
merase activity in tumour cells.[4] Subsequently, it was shown
that G-rich sequences are tightly clustered immediately up- and
downstream of the transcription start site in proliferation-asso-
ciated genes. Therefore, G4s have been suggested as regulatory
elements in gene expression,[5,6] in particular, in genes involved
in cell signalling, which have representatives from the six hall-
marks of cancer.[7] Indeed, the stabilization of G4-folded struc-
tures by small molecules or selective antibodies is responsible
for oncogene down-regulation in vitro (i.e., bcl-2, c-myc, c-kit,
hTert).[8–12] Strong evidence of G4 formation inside human cells
This wide array of therapeutic potential has propelled the
synthesis and development of new G4 selective ligands.[29,30]
The general features of these G4-recognising ligands include a
large, flat aromatic surface and cationic charges.[31] These struc-
tural features can be modulated to improve the specific recog-
nition of quadruplex over duplex DNA and between different
G4 sequences and topologies. Particularly promising from this
point of view are the synthetic molecules Braco-19[32] and re-
lated acridines,[33] pyridine-2,6-dicarboxylic acid bis-quinolin-2-
ylamides (PDCs)[34] and quinolinium analogues,[35,36] as well as
the natural product Telomestatin[37] and the fluoroquinolone
derivative Quarfloxin.[38] Our group and others have also de-
scribed a series of naphthalene diimide (NDI) derivatives. In
particular, tri- and tetrasubstituted NDIs exhibit high G4
affinity[39,40] with moderate selectivity for the parallel topol-
ogy.[41–43] All the substituents introduced on to the NDI core,
according to the literature, are conformationally flexible moie-
ties with limited steric demand. In this work we further modi-
fied the NDI to obtain core-substituted alkynyl derivatives with
extended planar and rigid structures to exploit π-stacking inter-
actions with the G-tetrad, retaining red and NIR absorption and
remarkable solvatochromic properties.[44] We tested these com-
pounds against the telomeric sequence and a series of G4-form-
ing sequences present in the promoter regions of oncogenes.
We showed that the new NDI derivatives display good affinity
and stabilization towards G4 sequences. By using fluorescence
[a] Department of Chemistry, University of Pavia,
v.le Taramelli 10, 27100 Pavia, Italy
E-mail: mauro.freccero@unipv.it
http://phdscchim.unipv.eu/site/en/home/coordinator.html
[b] Department of Molecular Medicine, University of Padua,
via Gabelli 63, 35121 Padua, Italy
E-mail: sara.richter@unipd.it
https://sites.google.com/site/saranrichter/home
[c] Dipartimento di Scienze della Salute Università degli Studi
“Magna Græcia” di Catanzaro Campus “Salvatore Venuta”,
Viale Europa, 88100 Catanzaro, Italy
[‡] These authors contributed equally to this work.
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under http://dx.doi.org/10.1002/ejoc.201600757.
Eur. J. Org. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
resonance energy transfer (FRET) and circular dichroism (CD),
we were able to show that both dimethylamino and hydroxy
moieties augment the affinity for G4. Moreover, we showed that
these derivatives preferred c-myc and bcl-2 G4s to those
formed by the telomeric and c-kit sequences.
Results and Discussion
Recently we designed a small family of acidic alkynyl-substi-
tuted NDIs (1 and 2, Scheme 1) with intense NIR absorption.[44]
In the present work, we exploited their water solubility, obtain-
ing mono- and di-core-substituted NDIs (M-cNDIs and B-cNDIs,
respectively, 3–13; see Scheme 1).
Scheme 2. Synthesis of the arylethynyl cNDIs 1–13 by a) Sonogashira cross-
coupling and b) subsequent acid (16: 10 % HCl, r.t., 30 min, THF) or base
(m17 or p17: K2CO3, r.t., 1 h, MeOH/H2O, 4:1) hydrolytic deprotections.
acetylation, further Sonogashira cross-coupling with trimethyl-
silylacetylene and finally fluoride-induced desilylation.
Scheme 1. Mono- and di-core-substituted NDIs (M-cNDIs and B-cNDIs, respec-
tively) synthesized and investigated as G4 ligands.
Synthesis of Water-Soluble cNDIs
Both M-cNDIs and B-cNDIs were synthesized starting from 2,6-
dibromonaphthalene diimide, which was easily prepared from
the commercially available anhydride. The latter was bromin-
ated with dibromocyanuric acid and directly subjected to classi-
cal imidation in acetic acid as solvent to yield the bromo-substi-
tuted NDIs 14 and 15 (Scheme 2).[45–48]
To introduce substituents at the aromatic core, we per-
formed Sonogashira cross-coupling reactions between the
hydro-soluble mono- or dibromo-NDI 14 and 15, respectively,
and the substituted ethynylphenols 16, m17, p17, m18, p18,
m19 and p19 using tetrakis(triphenylphosphine)palladium(0)
and CuI in THF/Et3N (1:1) at 40 °C (Scheme 2).
The substituted arylethynyl building blocks (16, m17, p17,
m18, p18, m19 and p19) were synthesized according to two
different protocols, as shown in Scheme 3. The Sonogashira
cross-coupling of phenol derivatives did not afford the ex-
Scheme 3. Synthesis of the substituted arylethynyl building blocks (16, m17,
p17, m18, p18, m19 and p19) for the Sonogashira cross-coupling reactions
in Scheme 2. Reagents and conditions: a) NaBH4, THF/H2O (15:0.1), 5 min, r.t.;
b) dimethoxypropane, acetone, PTSA,
[PdCl2(PPh3)2], PPh3, TMS-acetylene, 10 h, 90 °C; d) 1
1
h, r.t.; c) triethylamine, CuI,
TBAF in THF solution,
M
THF, 30 min, 0 °C; e) (CH3CO)2O, I2, microwave 300 W, 10 min; f) dry toluene,
CuI, [PdCl2(PPh3)2], TMS-acetylene, tert-butylamine, 24 h, 40 °C; g) abs. EtOH,
paraformaldehyde, dimethylamine or morpholine, 2 h, 78 °C; h) triethylamine,
CuI, [Pd(PPh3)4], TMS-acetylene, 2 h, 25 °C.
Similarly, the substituted arylacetylene 16 was synthesized
pected products in acceptable yields, which indicates that the by sodium borohydride reduction of the commercially available
phenol moiety needed to be protected. For this reason, the 5-bromosalicylaldehyde, protected as the acetonide and subse-
final compounds m17 and p17 were synthesized by phenol quently subjected to Sonogashira reaction. In contrast, the pro-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tection step was not required for the synthesis of the Mannich
Because the cNDIs investigated display three or four amine
base derivatives m18,19 and p18,19. In fact, the strong groups, which, under our experimental conditions, may confer
hydrogen bonding between the amine and the ortho-phenol- a differential cationic nature, we performed FRET in low ionic
OH likely prevents this moiety interfering in the reaction, strength buffer by decreasing the K+ concentration to 20 m
M.
achieving a protecting-group-free cross-coupling process. The Under these conditions, compounds whose binding involves
cross-coupling reactions were performed with trimethylsilyl- with an important electrostatic contribution were expected to
acetylene in the presence of [Pd(PPh3)2Cl2], PPh3 (step c, exhibit higher ΔTm values. The ΔTm values of the NMe2 com-
Scheme 3) or [Pd(PPh3)4] (step h) and CuI in Et3N (steps c and pounds 7–9 at 20 and 100 m
h) or toluene (step f). The different protocols quantitatively the morpholine series of compounds displayed increased ΔTm
yielded the final terminal alkynes 16, m17, p17, m18, p18, m19 at 20 m
K+, which suggests a higher electrostatic contribution.
and p19 after desilylation by 1 TBAF in THF solution at 0 °C
M
K+ were very similar. In contrast,
M
M
under anhydrous conditions (Scheme 3).
Acid/Base Properties of Representative cNDIs
To rationalize the binding behaviour observed in FRET in differ-
ent ionic strength solutions, the acid/base properties of the
most promising G4 ligands 7 and 8 were investigated by UV/
Vis pH and potentiometric titrations. In addition, we compared
their acidity with the most soluble and least effective G4 ligand
10 (see Figure S1 in the Supporting Information). Unfortunately,
the low solubility of 11 compromised the measurements (see
Figure S2). The titrations were performed in H2O, in the pres-
FRET Melting Data
We initially evaluated the binding of the cNDIs 1, 3, and 5–13
to G4 by FRET melting using the fluorescence-labelled human
telomeric sequence F21T as a model for the telomeric G4 DNA.
Unfortunately, poor solubility compromised the measurements
with 2 and 4. We measured the melting temperatures (Tm) of
F21T by FRET in the absence and presence of the cNDIs in
ence of an ionic buffer (0.1
M NaNO3), by the addition of small
100 m
M
K+ solutions (K+ concentration close to that found in
amounts of NaOH to progressively change the pH. The spectra
were subjected to multivariate non-linear regression to calcu-
late the pKa values. An increase in the pH in the range 5.5–7.5
of a solution of 7 induced significant spectral changes (Figure 1,
A). In particular, the absorption band at 468 nm disappeared
with concomitant formation of a new redshifted band at
569 nm. This change was due to the formation of the dipolar
form of the Mannich base moiety within 7 (Figure 1, B), the
cells, Table 1). The net Tm increment (ΔTm) suggested that the
cNDIs could be classified into three groups of ligands of differ-
ent G4 stabilizing efficiency (Table 1): a) cNDIs incorporating
the morpholine moiety (10–13) were poor stabilizers with
ΔTm ≤ 10 °C at 100 mM
K+, b) compounds with no substituents
or containing o-hydroxybenzyl alcohol moieties (1, 3, 5, 6) were
intermediate binders with 13 °C < ΔTm < 18 °C and c) cNDIs
containing a Mannich base with dimethylaminomethyl moieties
(7–9) were good stabilizers with Tm > 82 °C and ΔTm > 19 °C
(Table 1). Of these, the structurally extended bifunctionalized 8
was the most powerful ligand with ΔTm values of 34.6 and
23.8 °C in 20 and 100 mM
K+ solutions, respectively.
Table 1. Stabilization of the fluorescence-labelled human telomeric sequence
(F21T, 0.25 μ ) in the presence of cNDIs in diluted (20 m ) and concentrated
KCl solution (100 m ), evaluated by FRET-melting and CD, together with the
dissociation constants (KD), measured by surface plasmon resonance (SPR).
M
M
M
cNDI[a]
FRET, ΔTm [°C][b]
20 m
K+ 100 m
CD, Tm [°C][c]
SPR, KD [M]
M
M
K+ 290/265 nm 100 m K+
M
10
11
12
13
1
3
5
6
7
22.5
18.0
20.4
27.9
19.9
21.9
21.9
27.8
31.7
34.6
30.8
10.1
3.9
2.4
65.1/75.9
65.7/72.1
66.3/77.5
66.3/70.1
67.9/81.0
67.7/81.0
68.4/79.5
66.8/78.1
72.2/87.5
65.8/84.2
76.3/89.0
–
–
–
–
9.8
Figure 1. A) UV/Vis spectra recorded upon pH spectrophotometric titration
of 7 (30 μM, T = 25 °C) in 0.1 M NaNO3. Red line: spectrum of 7 at pH 2;
blue line: final spectrum taken at pH 9. B) Speciation analysis following the
13.8
15.7
13.8
17.8
20.9
23.8
19.3
2.1 × 10–6
–
–
–
potentiometric titration of 7. Lines: variation in the distribution of species
2+
with pH; tricationic 7H33+, dicationic 7H2 and monocationic 7H+ distribu-
8.7 × 10–7
3.8 × 10–8
4.4 × 10–7
tions are reported. Red and blue triangles show the change in absorption
monitored at 468 and 569 nm, respectively. C) UV/Vis spectra recorded upon
pH spectrophotometric titration of 8 (30 μM, T = 25 °C) in 0.1 M NaNO3. Red
8
9
[a] The cNDIs have been classified as poor (10–13), fairly good (1, 3, 5, 6)
and good ligands (7–9). [b] 1.0 μ cNDIs in 20 and 100 m
K+ solution. [c]
Tm measured by CD at 290 and 265 nm (100 m
K+ solution) relative to the
unfolding of the hybrid G4 [3+1] topology (monitored at 290 nm) and parallel
topologies (monitored at 265 nm), respectively.
line: spectrum of 8 at pH 2; green line: spectrum at pH 7; blue line: final
spectrum taken at pH 11. D) Speciation analysis following the potentiometric
titration of 8. Lines: variation in the distribution of species with pH. Red and
blue triangles show the change in absorption monitored at 516 and 705 nm,
respectively.
M
M
M
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
remarkable bathochromic shift being due to the greater stabili-
zation achieved by deprotonation of the S1 state (exhibiting
charge-transfer character) compared with the S0 state. Similarly,
the absorption band at 510 nm for cNDI 8 disappeared with
the parallel formation of the redshifted band at 620 nm with
increasing pH (Figure 1, C). The band at pH 7 is slightly broader
and redshifted (650 nm) in comparison with that recorded at
pH 11 (Figure 1, C, green vs. blue line). This effect may be due
to the self-aggregation of 8 in water at pH 7, as suggested by
a barely detectable turbidity of the solution at a concentration
of 30 μM.
The pKa values of the conjugate acid of 7 (Table 2) and speci-
ation analysis (Figure 1, B) suggest that 7 exists mainly as the
dicationic species (7H22+) at pH 7.2 (see 7H22+, blue line in Fig-
ure 1, B). The potentiometric titrations of 8 (Table 2) together
with the resulting evaluation of the charged species as a func-
tion of pH (Figure 1, D) suggest that the tetracationic cNDI
(8H44+) is very acidic, with 8 reaching a maximum concentration
as the dicationic species (8H22+) at pH 7.0.
Table 2. pKa values for the ligands 7, 8 and 10.[a]
cNDI
pKa1
pKa2
pKa3
pKa4
7
8
10
6.60 0.02
2.93 0.08
6.84 0.05
7.81 0.02
6.45 0.02
8.79 0.02
10.65 0.03
7.48 0.01
10.16 0.10
–
10.31 0.02
–
Figure 2. CD spectra of the complexes formed between the representative
cNDIs 5, 7 and 11 (A, B and C, respectively) in the presence of hTel under
thermal denaturing conditions from 20 to 85 °C (spectra recorded every 5 °C).
In general, the molar ellipticity decreases at 290 nm and increases at 260 nm
upon raising T up to the melting point. The arrow length is a qualitative
indicator of the intensity of the spectral change at the indicated wavelength.
[a] The titrations were performed in aqueous solution in ionic buffer (0.1
NaNO3) at 25 °C.
M
Thus, the electrostatic features of 7 and 8 are very similar.
In contrast, the pKa values of 10 measured by potentiometric
titrations (Table 2) and the resulting speciation analysis (see
Figure S1 in the Supporting Information) suggest that 10 exists
3+
This induced hybrid→parallel folding was remarkable for 7–
9 and was also observed to a lesser extent for 1, 3, 5 and 6
(Figure 2, A and B), but was negligible for the less efficient
ligands 10–13 containing the morpholine moiety (Figure 2, C).
The Tm values were measured for both the hybrid (at 290 nm)
and parallel (at 265 nm) conformations (Table 1). The poor sta-
bilizers 10–13 display low Tm values at both wavelengths (65 °C
at 290 nm and 70–76 °C at 265 nm), the intermediate binders
1, 3, 5 and 6 exhibit low Tm at 290 nm (67 °C) with a higher Tm
at 265 nm (78–81 °C), and the good binders exhibiting a single
dimethylaminomethyl moiety (7 and 9) show Tm values signifi-
cantly higher at both wavelengths (72–76 °C at 290 nm and
88–89 °C at 265 nm). Remarkably, of the good binders, only the
bifunctional 8 shows a low Tm at 290 nm (65.8 °C) and its Tm is
very high at 265 nm (84.2 °C). These data are in good agree-
ment with those obtained by FRET and show that the initial
as an equimolar mixture of tri- and dicationic species (10H3
and 10H22+) at pH 7.0. These data explain the different G4 bind-
ing behaviour of 10 compared with 7 and 8 under different
ionic strength solutions, 10 being more positively charged than
7 and 8 at pH 7.0. However, the remarkable difference in the
binding properties, evaluated by FRET-melting (Table 1), of 7
and 10 at constant ionic strength cannot be ascribed to electro-
static factors, as the stronger cationic nature of the latter should
induce stabilization. A similar consideration holds for the ration-
alization of the difference in binding properties of 7 and its
larger and bifunctional analogue 8.
Circular Dichroism Data
The binding of the cNDIs 1, 3 and 5–13 to the human telomeric hybrid conformation is, in general, poorly stabilized and that
sequence (hTel) was next evaluated by circular dichroism (CD) the greatest degree of stability is conferred on the parallel
analysis in 100 m
M
K+ solution. The extensively studied hTel structure. This statement is particularly true for 8, which among
oligonucleotide[49–51] displays a hybrid G4 [3+1] topology, the cNDIs investigated seems the most potent and selective
which all the cNDIs stabilize to different extents with a maxi- towards the parallel topology. The affinity values (KD) of the
mum peak at 290 nm. However, during thermal denaturation, compounds that best stabilize hTel G4, measured by surface
the initial [3+1] topology is converted in all cases into a parallel plasmon resonance (SPR) analysis, are in accord with the FRET
conformation with a maximum at 265 nm at around 70 °C (Fig- and CD data. Compound 8 displays the highest affinity, with a
ure 2).
KD value of 38 nM. The second and third best binding com-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
pounds 9 and 7, each of which contain a dimethylaminomethyl
moiety, display KD values in the range 440–870 n . In contrast,
M
the KD of intermediate binding ligand 1, which lacks the di-
methylaminomethyl group, is remarkably lower (Table 1).
Selectivity Evaluated by Competition Experiments
To investigate G4 sequence and conformation selectivity, the
binding of the cNDIs to the human telomeric oligonucleotide
was studied in the presence of competitor G4-forming oligo-
nucleotides and analysed by FRET. Competitors displaying well-
defined parallel (c-myc, bcl-2, c-kit1) or a mixture of predomi-
nantly parallel with some anti-parallel-like folding topologies (c-
kit2; in the presence of 100 m
M
K+)[52,53] were chosen. c-myc
and bcl-2 efficiently competed with F21T for binding to the
cNDIs, whereas c-kit2 and c-kit1 were less effective competitors.
In addition, competition experiments with ds- and ss-DNA
scrambled sequences did not affect cNDI binding to the G4 oligo-
nucleotide (Figure 3).
Figure 3. Competition experiments of cNDI binding to the fluorophore-la-
belled human telomeric sequence (F21T) in the presence of 1 μ
M of onco-
gene promoter G4 folded sequences and ds- and ss-DNAs in FRET assays.
Competition was calculated as the percentage of cNDI binding to F21T in
the presence of the specified competitor oligonucleotide.
Because c-myc effectively competes with F21T for binding to
the ligands 5, 6 and 8, we decided to investigate the increased
stability of c-myc induced by 8. It is well known that c-myc
exhibits a very stable G4 structure at 100 m
M
K+. Therefore, we
) to
parallel topologies to an even greater extent than the telomeric
sequence.
performed a CD analysis at low K+ concentrations (20 m
M
lower the initial high stability of the G4 structure (Figure 4, A,B).
Under these conditions, Tm was 49.5 0.2 °C (Figure 4, C). When
compound 8 was added at a concentration of 1 μ
increased to above 90 °C. The resulting ΔTm > 40.5 °C indicates
M, the Tm
Molecular Modeling Studies: Binding Mode of 6, 8 and 11
an excellent stabilization of the c-myc G4 (Figure 4, D). These To qualitatively elucidate the most important requirements of
data altogether support the ability of compound 8 to strongly the cNDI/target recognition, docking calculations were per-
bind and stabilize the G4-forming promoter sequences with formed on the “good” compound 8, the “fairly good” com-
Figure 4. CD analysis of c-myc G4 (4 μM) thermal unfolding A) in the absence of 8 and B) in the presence of 8 (16 μM, 4:1 cNDI/DNA). CD spectra were
recorded in the temperature range 20–90 °C at intervals of 5 °C. Molar ellipticity values at 263 nm were plotted against temperature C) in the absence of 8
or D) in the presence of 8. Experimental data (●) were fitted to the van't Hoff equation (line) to obtain values of Tm.
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
pound 6 and the “poor” ligand 11 (Table 1) with both G4 hTel by monitoring the RMSD progression of the G-tetrad core as a
(PDB ID: 4DA3)[54] and dsDNA using the Extra Precision (XP) function of time (analysis of the RMSD of the G-tetrad for the
Glide software (v. 6.7).[55] The ligands were kept flexible and the 8/G4 hTel and 11/G4 hTel complexes compared with that of the
G4 was considered as a rigid macromolecule. The best-ranked G4 hTel alone are presented in Figure S3A,B, respectively, in the
XP Glide score solution of 6, 8 and 11 shows a π–π end-stack- Supporting Information). Analysing the behaviour of hydrogen-
ing interaction at the 3′-end of the G4 hTel for all three ligands bonding contacts during the MD simulations, it turned out that
(Figure 5, A–C, respectively). The hydroxymethyl- and dimethyl- the best binder 8 retains between one and three hydrogen
aminomethylphenol moieties in ligands 6 and 8, respectively, bonds in the first 42 ns of the simulation, and between one
are oriented similarly towards the phosphate backbone and two hydrogen bonds until the end of the simulation (Fig-
(G8:G9:T4:T10 residues on the one side and T16:G19:G20 resi- ure 6, A). Conversely, the number of hydrogen-bonding con-
dues on the other), engaging in four hydrogen-bonding interac- tacts is significantly reduced for compound 11 after about 10 ns
tions with the G9:T16 and T16:G15:G8:G3 residues, respectively and is maintained at between zero and one until the end of
(Figure 5, D,E). In contrast, the binding orientation of compound the simulation (Figure 6, B). It is worth noting that, in the first
11 is different, with the morpholinomethylphenol group facing 42 ns of the MD simulation, most of the hydrogen bonds of 8
towards the phosphate groups of the G2:G3:T4 residues on one simultaneously involve the hydroxy hydrogen and the residues
side and the G15:T16:T17 residues on the other (Figure 5F). Con- T16:T4 of the G4 hTel, as highlighted in Figure 6 (A), although
versely, the positively charged dimethylaminoethyl portions of even the dimethylaminoethyl groups are involved in hydrogen-
6, 8 and 11 are involved in cation-π and salt-bridge/hydrogen- bonding/salt-bridge interactions, especially with the T4 residue
bonding interactions.
(data not shown). These observations clearly indicate that the
stabilization of the G4 hTel is mainly due to π–π stacking of the
NDI core and salt-bridge interactions of the positively charged
dimethylaminoethyl moieties. In the most interesting com-
pound 8, the hydroxy groups act as hydrogen-bonding donors,
whereas the nitrogen atoms of the dimethylaminoethyl moie-
ties act as hydrogen-bonding acceptors.
Figure 5. Top views of the best-ranked docking conformations of A) 6, B) 8
and C) 11 and lateral views of D) 6, E) 8 and F) 11 with a particular focus on
the hydrogen bonding to the residues. G4 hTel is represented in grey cartoon,
the G-tetrad plane involved in π–π-stacking interactions is displayed as lines
coloured by atom with grey carbon atoms. cNDIs are shown as green, violet
and light-orange sticks for 6, 8 and 11, respectively. Hydrogen bonds are
depicted as dashed yellow lines.
Figure 6. Trends in the hydrogen-bonding interactions occurring during MD
simulations of the interactions between A) 8 and B) 11 and G4 hTel.
Molecular Dynamics Simulations
To take into account the flexibility of the receptor and investi-
gate the main interactions involved in the geometrical stabiliza-
Conclusions
tion of G4 hTel, the best-docked poses obtained for compounds We have synthesized 11 core-substituted NDIs with extended
8 and 11 (Figure 5, B,C) were subjected to molecular dynamics planar and rigid arylethynyl appendages by Sonogashira cross-
(MD) simulations for 100 ns. Details of the MD parameters are coupling reactions and evaluated their G4 binding properties.
reported in the Exp. Sect. The stability of G4 hTel was evaluated The most potent and promising G4 ligands contain a Mannich
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
N,N′-Bis[(dimethylamino)ethyl]-2-{2-[3-(dimethylaminomethyl)-
4-hydroxyphenyl]ethynyl}-1,4,5,8-naphthalenetetracarboxylic
Diimide 3 HCl (7): Orange solid, yield 81 %, m.p. >200 °C (decomp.).
1H NMR (300 MHz, CD3OD): δ = 8.81–8.77 (m, 2 H), 8.72 (s, 1 H),
base with a dipolar dimethylaminomethyl group on the aryl-
ethynyl moieties (7–9). They exhibited good selectivity for par-
allel G4, evaluated by FRET melting, CD melting, SPR and com-
petition experiments. Surprisingly, the structurally most ex-
tended bifunctional cNDI 8 was the most powerful ligand to-
wards c-myc, bcl-2 and parallel-iduced hTel. Molecular model-
ling demonstrated that favorable interactions to stabilize the
parallel G4 hTel, in addition to the π–π stacking of the NDI core,
are the selective hydrogen bonds formed by the hydroxy
groups and dimethylaminoethyl moieties, together with salt
bridges, especially for the best binder 8. Summing up, our data
demonstrate that extended cNDIs with two centrosymmetric
arylethynyl moieties capable of synergic donor and acceptor
hydrogen-bonding interactions exhibit good binding properties
and moderate selectivity towards parallel G4s. The rigid Man-
nich base moieties are capable of engaging both the phosphate
backbones and the outside rim of the G-quartet in a hydrogen-
bonding network. These preliminary results reveal the evidence
of a promising structural feature implementing G4 selectivity in
large aromatics and water-soluble G4 ligands.
3
3
4
7.83 (d, JH,H = 1.8 Hz, 1 H), 7.68 (dd, JH,H = 8.4, JH,H = 1.8 Hz, 1
3
H), 7.09 (d, JH,H = 8.4 Hz, 1 H), 4.64–4.60 (m, 4 H), 4.41 (s, 2 H),
3.63–3.57 (m, 4 H), 3.06 (s, 6 H), 3.04 (s, 6 H) ppm. 13C NMR (75 MHz,
CD3OD): δ = 165.1, 164.9, 164.7, 163.9, 160.2, 138.3, 137.8, 137.2,
134.0, 132.8, 131.8, 128.9, 128.3, 127.9, 127.6, 127.4, 126.7, 119.2,
117.5, 115.5, 103.7, 90.4, 58.1, 57.5, 57.4, 44.6, 44.5, 43.7, 37.3 ppm.
C33H38Cl3N5O5 (691.05): calcd. C 57.36, H 5.54, Cl 15.39, N 10.13, O
11.58; found C 57.38, H 5.53, N 10.11.
N,N′-Bis[(dimethylamino)ethyl]-2,6-bis{2-[3-(dimethylamino-
methyl)-4-hydroxyphenyl]ethynyl}-1,4,5,8-naphthalenetetra-
carboxylic Diimide 4HCl (8): Red-violet solid, yield 89 %, m.p.
>200 °C (decomp.). 1H NMR (300 MHz, CD3OD): δ = 8.78 (s, 2 H),
3
3
7.84 (s, 1 H), 7.78 (d, JH,H = 8.4 Hz, 1 H), 7.01 (d, JH,H = 8.4 Hz, 1
H), 4.82 (br. s, 4 H), 4.66 (s, 2 H), 3.70–3.61 (m, 4 H), 3.1 (s, 12 H),
2.94 (s, 12 H) ppm. 13C NMR (75 MHz, CD3OD): δ = 164.5, 163.9,
160.1, 138.3, 137.7, 128.4, 128.3, 127.2, 126.7, 119.4, 119.2, 117.4,
115.6, 103.7, 90.5, 58.2, 57.4, 44.6, 43.7, 37.4, 35.7 ppm.
C44H50Cl4N6O6 (900.73): calcd. C 58.67, H 5.60, Cl 15.74, N 9.33, O
10.66; found C 58.65, H 5.61, N 9.31.
N,N′-Bis[(dimethylamino)ethyl]-2-{2-[4-(dimethylaminomethyl)-
3-hydroxyphenyl]ethynyl}-1,4,5,8-naphthalenetetracarboxylic
Diimide 3HCl (9): Yellow solid, yield 86 %, m.p. >200 °C (decomp.).
1H NMR (300 MHz, CD3OD): δ = 8.88–8.81 (m, 2 H), 8.90 (s, 1 H),
8.62 (d, JH,H = 8.4 Hz, 1 H), 8.48 (d, JH,H = 8.4 Hz, 1 H), 7.43 (s, 1
H), 4.71 (m, 4 H), 4.49 (s, 2 H), 3.74 (m, 4 H), 3.20 (s, 6 H), 3.17 (s, 6
H), 3.02 (s, 6 H) ppm. 13C NMR (75 MHz, CD3OD): δ = 165.1, 164.9,
164.7, 163.9, 158.2, 137.3, 134.2, 132.9, 132.2, 129.4, 128.3, 128.1,
127.7, 127.4, 125.4, 120.2, 119.9, 102.1, 91.2, 58.3, 57.5, 57.4, 44.5,
44.4, 43.7, 37.2 ppm. C33H38Cl3N5O5 (691.05): calcd. C 57.36, H 5.54,
Cl 15.39, N 10.13, O 11.58; found C 57.37, H 5.55, N 10.11.
Experimental Section
Synthesis of cNDIs 1–13: We synthesized the NDI precursors 14
and 15 according to a published procedure.[46] Ligands 1, 2, 5 and
6 were synthesized according to our previous work.[44] The cNDIs 3,
4 and 7–13 required a modified and optimized Sonogashira cross-
coupling protocol, which is described below.
3
3
General Procedure for the Synthesis of 3, 4 and 7–13 by the
Modified Sonogashira Cross-Coupling Reaction: A mixture of 14
and 15 (0.21 mmol + 0.32 mmol, respectively) was suspended in dry
THF (30 mL) and kept under Ar. [Pd(PPh3)4] (30 mg, 0.026 mmol),
CuI (4.9 mg, 0.026 mmol) and triethylamine (30 mL) were sequen-
tially added to the stirred mixture at room temperature. After a few
minutes, 4-arylethynyl derivative 18 or 19 (1.06 mmol) was added
to the solution, stirring at 40 °C. After 1.5 h the resulting dark-brown
solution was evaporated under vacuum and the products were sep-
arated and purified by preparative HPLC. The purified HPLC frac-
N,N′-Bis[(dimethylamino)ethyl]-2-[2-(3-morpholinomethyl-4-
hydroxyphenyl)ethynyl]-1,4,5,8-naphthalenetetracarboxylic Di-
imide 3 HCl (10): Orange solid, yield 80 %, m.p. >200 °C (decomp.).
1H NMR (300 MHz, CD3OD): δ = 8.77 (m, 2 H), 8.69 (s, 1 H), 7.84 (d,
3
4
3JH,H = 1.9 Hz, 1 H), 7.77 (dd, JH,H = 8.4, JH,H = 1.9 Hz, 1 H), 7.09
3
(d, JH,H = 8.4 Hz, 1 H), 4.67–4.60 (m, 4 H), 4.46 (s, 2 H), 4.0–3.8 (br.
s, 4 H), 3.66–3.61 (m, 4 H), 3.55–3.3 (br. s, 4 H), 3.1 (s, 6 H), 3.07 (s,
6 H) ppm. 13C NMR (75 MHz, CD3OD): δ = 165.1, 164.9, 164.7, 163.9,
160.5, 138.9, 137.9, 137.3, 132.8, 131.8, 129.4, 129.0, 128.2, 127.8,
127.4, 127.3, 126.6, 117.8, 117.6, 115.5, 103.8, 90.4, 65.2, 57.5, 57.4,
56.8, 50.1, 44.5, 44.4, 37.3, 37.2 ppm. C35H40Cl3N5O6 (733.09): calcd.
C 57.34, H 5.50, Cl 14.51, N 9.55, O 13.09; found C 57.32, H 5.51, N
9.54.
tions were added to 1
in order to collect the samples as crystalline hydrochlorides.
M HCl (0.5 mL) before evaporating the eluent
N,N′-Bis[(dimethylamino)ethyl]-2-[2-(3-hydroxyphenyl)ethynyl]-
1,4,5,8-naphthalenetetracarboxylic Diimide 2HCl (3): Green
solid, yield 85 %, m.p. >200 °C (decomp.). 1H NMR (300 MHz,
[D6]DMSO): δ = 10.1 (br. s, OH), 9.82 (br. s, NH), 8.78–8.69 (m, 2 H),
3
3
8.65 (s, 1 H), 7.33 (t, JH,H = 7.8 Hz, 1 H), 7.20 (d, JH,H = 7.8 Hz, 1
3
4
N,N′-Bis[(dimethylamino)ethyl]-2,6-bis[2-(3-morpholinomethyl-
4-hydroxyphenyl)ethynyl]-1,4,5,8-naphthalenetetracarboxylic
Diimide 4HCl (11): Red solid, yield 71 %, m.p. >200 °C (decomp.).
H), 7.15 (s, 1 H), 6.96 (t, JH,H = 8.1, JH,H = 2.0 Hz, 1 H), 4.46–4-44
(m, 4 H), 3.51 (br. s, 4 H), 2.95 (s, 12 H) ppm. 13C NMR (75 MHz,
[D6]DMSO): δ = 163.0, 162.6, 161.7, 157.6, 134.9, 131.1, 130.3, 130.1,
127.1, 126.4, 126.1, 126.0, 125.7, 125.6, 125.3, 123.1, 122.6, 118.4,
117.8, 101.5, 88.9, 54.6, 54.5, 42.6, 35.5 ppm. C30H30Cl2N4O5 (597.50):
calcd. C 60.31, H 5.06, Cl 11.87, N 9.38, O 13.39; found C 60.33, H
5.07, N 9.36.
3
1H NMR (300 MHz, CD3OD): δ = 8.76 (s, 2 H), 7.84 (d, JH,H = 1.9 Hz,
1 H), 7.78 (dd, 3JH,H = 8.4, 4JH,H = 1.9 Hz, 1 H), 7.09 (d, 3JH,H = 8.4 Hz,
1 H), 4.68–4.62 (m, 4 H), 4.45 (s, 4 H), 4.0–3.8 (br. s, 8 H), 3.91–3.87
(m, 4 H), 3.55–3.3 (br. s, 8 H), 3.09 (s, 12 H) ppm. 13C NMR (75 MHz,
CD3OD): δ = 164.4, 163.8, 160.5, 138.9, 137.9, 128.5, 128.4, 127.0,
126.5, 117.8, 117.6, 115.6, 103.8, 90.5, 65.2, 57.4, 56.9, 53.4, 44.5,
37.3 ppm. C50H58Cl4N6O6 (980.86): calcd. C 61.23, H 5.96, Cl 14.46,
N 8.57, O 9.79; found C 61.20, H 5.95, N 8.59.
N,N′-Bis[(dimethylamino)ethyl]-2,6-bis[2-(3-hydroxyphenyl)-
ethynyl]-1,4,5,8-naphthalenetetracarboxylic Diimide (4):
Orange solid, yield 79 %, m.p. >200 °C (decomp.). 1H NMR (300 MHz,
[D6]DMSO): δ = 8.72 (s, 2 H), 7.60–7.56 (m, 2 H), 7.45–7.38 (m, 4 H),
7.27–7.19 (m, 2 H), 4.35 (br. s, 4 H), 2.80 (br. s, 4 H), 2.42 (s, 12 N,N′-Bis[(dimethylamino)ethyl]-2-[2-(4-morpholinomethyl-3-
H) ppm. C38H32N4O6 (640.69): calcd. C 71.24, H 5.03, N 8.74, O 14.98;
found C 71.23, H 5.05, N 8.75.
hydroxyphenyl)ethynyl]-1,4,5,8-naphthalenetetracarboxylic Di-
imide 3HCl (12): Orange solid, yield 62 %, m.p. >200 °C (decomp.).
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1H NMR (300 MHz, [D6]DMSO): δ = 11.0 (br. s, NH), 10.28 (br. s, OH),
deg cm2 dmol–1 (molar ellipticity). The oligonucleotides hTel (5′-
AGGGTTAGGGTTAGGGTTAGGG-3′) and c-myc were diluted from
3
8.86–8.82 (m, 2 H), 8.76 (s, 1 H), 7.76 (d, JH,H = 8.4 Hz, 1 H), 7.51 (s,
3
1 H), 7.32 (d, JH,H = 8.4 Hz, 1 H), 4.57–4.51 (m, 4 H), 4.43 (s, 2 H),
stock to the final concentration (4 μ
M) in Li cacodylate buffer
4.06–3.93 (br. s, 4 H), 3.48–3.42 (m, 4 H), 3.27 (br. s, 4 H), 2.98 (br. s,
(10 m , pH 7.4) with 100 m KCl and then annealed by heating at
M
M
12 H) ppm. 13C NMR (75 MHz, [D6]DMSO): δ = 163.1, 162.7, 161.9, 95 °C for 5 min and gradually cooled to room temperature. The
157.0, 134.7, 134.0, 131.0, 130.4, 127.2, 126.6, 126.3, 126.2, 125.6, NDIs were added to give final concentrations of 16 μ
125.0, 124.2, 122.9, 122.8, 118.5, 117.9, 100.1, 89.75, 63.0, 54.4, 53.5, were recorded after 24 h in the temperature range 20–95 °C with
50.8, 42.6, 42.5, 35.5 ppm. C35H40Cl3N5O6 (733.09): calcd. C 57.34, H temperature increases of 5 °C. Tm values were calculated according
M. CD spectra
5.50, Cl 14.51, N 9.55, O 13.09; found C 57.33, H 5.49, N 9.54.
to the van′t Hoff equation, applied for a two state transition from
a folded to unfolded state, assuming that the heat capacities of the
folded and unfolded states are equal.[56]
N,N′-Bis[(dimethylamino)ethyl]-2,6-bis[2-(4-morpholinomethyl-
3-hydroxyphenyl)ethynyl]-1,4,5,8-naphthalenetetracarboxylic
Diimide 4HCl (13): Red solid, yield 91 %, m.p. >200 °C (decomp.).
1H NMR (300 MHz, [D6]DMSO): δ = 10.93 (br. s, 2 H, 2 NH), 10.34 (s,
Surface Plasmon Resonance: SPR measurements of the 5′-biotinyl-
ated hTel sequence (5′-AGGGTTAGGGTTAGGGTTAGGG-3′) were per-
formed with the BIAcore T100 (GE Healthcare) system using strepta-
vidin-coated sensor chips (Series S Sensor chip SA, GE Healthcare),
as described previously.[57]
3
2 H, 2 OH), 8.70 (s, 2 H), 7.68 (d, JH,H = 7.7 Hz, 2 H), 7.41 (s, 2 H),
3
7.30 (d, JH,H = 7.7 Hz, 2 H), 4.48 (br. s, 4 H), 4.32 (br. s, 4 H), 3.93
(br. s, 4 H), 3.86 (br. s, 4 H), 3.50 (br. s, 4 H), 3.35–3.30 (m, 8 H), 3.16
(br. s, 4 H), 2.91 (s, 6 H), 2.90 (s, 6 H) ppm. 13C NMR (75 MHz,
[D6]DMSO): δ = 162.2, 161.7, 157.0, 135.4, 126.5, 126.1, 125.9, 124.9,
123.0, 118.5, 100.6, 89.8, 63.1, 54.3, 53.5, 50.9, 42.6, 35.6 ppm.
C50H58Cl4N6O6 (980.86): calcd. C 61.23, H 5.96, Cl 14.46, N 8.57, O
9.79; found C 61.22, H 5.96, N 8.56.
Molecular Modelling: The crystal structure of the intramolecular
parallel human telomeric DNA G-quadruplex 21-mer sequence 5′-
GGGTTAGGGTTAGGGTTAGGG-3′ (G4-hTel) complexed with an NDI
derivative was downloaded from the Protein Data Bank
(www.rcsb.org; PDB ID: 4DA3).[54] After removal of the NDI-com-
plexed ligand and water of crystallization, the obtained G4-hTel
structure was prepared by using the “Protein Preparation Wizard”
panel implemented in the Schrödinger 2015 package[58] with the
force field OPLS 2005 in order to assign the correct bond orders
and adding all the hydrogen atoms. Afterwards, the prepared G4-
hTel was energetically minimized by using 2500 steps of the Polak–
Ribiere conjugate gradient minimization and the force field OPLS
2005. The 3D structures of the NDI derivatives 6, 8 and 11 were
constructed by using the Maestro Build Panel[59] v. 10.2. The proto-
nation states of the 6, 8 and 11 NDI derivatives are stated in accord
with those determined experimentally at pH 7.0. Finally, the 6, 8
and 11 NDI ligands were submitted to 5000 step Polak-Ribiere con-
jugate gradient minimization (0.05 kJ/Å mol convergence threshold)
with MacroModel[60] v. 10.8 by using the force field OPLS 2005. To
compare the energy of the selectivity between G4-hTel and the
double helix (dsDNA), an homologous model of dsDNA having the
same experimental sequence 5′-CAATCGGATCGAATTCGATCCGATTG-
3′ was constructed by using the Maestro Build Panel v. 10.2. It was
energetically minimized with the OPLS force field using 20000 steps
of the Polak–Ribiere conjugate gradient minimization (0.05 kJ/Å mol
convergence threshold) with MacroModel v. 10.8. The minimized
dsDNA thus-obtained was then submitted to 100 ns of molecular
dynamics (MD) simulation performed in the NPT ensemble at 300 K
and 1 atm.
Materials and Methods
Oligonucleotides: All oligonucleotides were purchased from
Sigma–Aldrich. After an initial dilution at 1 m
M in purified water,
further dilutions were carried out in the relevant buffer.
FRET Analysis: FRET assays were performed with FAM (6-carb-
oxyfluorescein) 5′-end- and Tamra (6-carboxytetramethyl-
rhodamine) 3′-end-labelled F21T oligonucleotide [5′-d(FAM-
GGGTTAGGGTTAGGGTTAGGG-Tamra-3′)]. Fluorescence melting
curves were determined with a LightCycler II (Roche) real-time PCR
machine using a total reaction volume of 20 μL with 0.25 μ
tagged oligonucleotide in a buffer containing 10 m lithium cacod-
ylate (pH 7.4) with 20 or 100 m KCl in the presence or absence of
1.0 μ NDI. After a first equilibration step at 30 °C during 2 min, a
M of
M
M
M
stepwise increase of 1 °C every minute for 65 cycles to reach 95 °C
was performed and measurements were made after each cycle with
excitation at 470 nm and detection at 530 nm. For the FRET compe-
tition assay, the oligonucleotide F21T was treated as above in the
presence of 1.0
μ
M
NDI and G-quadruplex competitors,
that is, c-myc (5′-TGGGGAGGGTGGGGAGGGTGGGGAAGG-3′),
bcl-2 (5′-AGGGGCGGGCGCGGGAGGAAGGGGGCGGGAGCGGGGCTG
-3′), c-kit1 (5′-AGGGAGGGCGCTGGGAGGAGGG-3′), c-kit2 (5′-
CGGGCGGGCGCGAGGGAGGGG-3′), as well as ssDNA (5′-
GGATGTGAGTGTGAGTGTGAGG-3′)
and
dsDNA
(5′-
Docking simulations of NDIs 6, 8 and 11 with both G4-hTel and
dsDNA were performed by using the Extra Precision mode (XP) of
Glide v. 6.7 using the force field OPLS 2005. The grid box was cen-
tred on the G-tetrad residues and made wide enough to include
the whole G4 hTel. Thus, 10 binding conformations were generated
for each ligand and the Glide XP score function was used to rank
the obtained binding conformations for each ligand. The binding
free energies (ΔGbind) were predicted by using the generalized-
Born/surface area (MM-GBSA) continuum solvent model of the
Prime module v. 4.0.[61] To further confirm the predicted binding
modes of 8 and 11 and demonstrate the geometrical stability of
G4 hTel, their best-ranked conformation was submitted to MD simu-
lations with the Desmond[63] program v. 4.2. The complexes were
first minimized with the force field OPLS 2005 and afterwards
CAATCGGATCGAATTCGATCCGATTG-3′) as controls for single- and
double-stranded DNA, respectively. The melting of F21T was moni-
tored alone as well as in the presence of the NDIs and of the com-
petitor DNAs. Final analysis of the data was carried out by using
Microsoft Excel® and SigmaPlot[62] software. Tm is defined as the
temperature corresponding to the maximum of the first derivative
of the fluorescence versus temperature curve. Tm values are means
of 2–3 experiments and ΔTm was calculated as the difference in Tm
in the presence and absence of the competitor oligonucleotides.
Circular Dichroism: CD experiments were performed with a PC100
spectrometer. A quartz cuvette with 5 mm path length was used
for spectra recorded from 230–320 nm at 2 nm bandwidth, 0.1 nm
step size and 4 s per point. The spectrum reported for each sample
represents the average of two scans. The observed ellipticities were placed in a cubic water box using the TIP4P explicit water model
converted to mean residue ellipticity (θ) with units of
by using the System Builder tool implemented in Desmond v. 4.2.
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
8 © 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Further K+ ions were added to neutralize the net charge of the
system and a KCl salt concentration equal to 0.02 m was also
considered to reproduce the experimental ionic conditions. Finally,
each of the obtained systems was equilibrated by using the Des-
mond Setup protocol and submitted to 100 ns of MD simulation
performed in the NPT ensemble at 300 K and 1 atm.
[24]
[25]
[26]
[27]
R. Perrone, M. Nadai, J. A. Poe, I. Frasson, M. Palumbo, G. Palu, T. E. Smith-
gall, S. N. Richter, Plos One 2013, 8, e73121.
S. Artusi, M. Nadai, R. Perrone, M. A. Biasolo, G. Palu, L. Flamand, A.
Calistri, S. N. Richter, Antiviral Res. 2015, 118, 123–131.
D. Piekna-Przybylska, M. A. Sullivan, G. Sharma, R. A. Bambara, Biochemis-
try 2014, 53, 2581–2593.
P. Murat, J. Zhong, L. Lekieffre, N. P. Cowieson, J. L. Clancy, T. Preiss, S.
Balasubramanian, R. Khanna, J. Tellam, Nat. Chem. Biol. 2014, 10, 358–
364.
R. Perrone, F. Doria, E. Butovskaya, I. Frasson, S. Botti, M. Scalabrin, S.
Lago, V. Grande, M. Nadai, M. Freccero, S. N. Richter, J. Med. Chem. 2015,
58, 9639–9652.
M
Acknowledgments
[28]
Financial support from the Ministero dell'Università e della Ric-
erca (MIUR), (FIRB-Ideas RBID082ATK_003) and from the Italian
Association for Cancer Research (AIRC, IG2013-14708) is grate-
fully acknowledged.
[29]
[30]
A. Arola, R. Vilar, Curr. Top. Med. Chem. 2008, 8, 1405–1415.
D. Monchaud, M. P. Teulade-Fichou, Org. Biomol. Chem. 2008, 6, 627–
636.
[31]
[32]
[33]
[34]
[35]
[36]
T. M. Ou, Y. J. Lu, J. H. Tan, Z. S. Huang, K. Y. Wong, L. Q. Gu, ChemMed-
Chem 2008, 3, 690–713.
N. H. Campbell, G. N. Parkinson, A. P. Reszka, S. Neidle, J. Am. Chem. Soc.
2008, 130, 6722–6724.
S. Sparapani, S. M. Haider, F. Doria, M. Gunaratnam, S. Neidle, J. Am.
Chem. Soc. 2010, 132, 12263–12272.
R. Rodriguez, S. Muller, J. A. Yeoman, C. Trentesaux, J. F. Riou, S. Balasu-
bramanian, J. Am. Chem. Soc. 2008, 130, 15758–15759.
A. De Cian, E. Delemos, J. L. Mergny, M. P. Teulade-Fichou, D. Monchaud,
J. Am. Chem. Soc. 2007, 129, 1856–1857.
Keywords: Nucleic acids · Cross-coupling · Hydrogen
bonds · Carboximides · G-Quadruplexes
[1] J. L. Huppert, FEBS J. 2010, 277, 3452–3458.
[2] T. A. Brooks, S. Kendrick, L. Hurley, FEBS J. 2010, 277, 3459–3469.
[3] K. Paeschke, T. Simonsson, J. Postberg, D. Rhodes, H. J. Lipps, Nat. Struct.
Mol. Biol. 2005, 12, 847–854.
[4] M. Folini, L. Venturini, G. Cimino-Reale, N. Zaffaroni, Expert Opin. Ther.
Targets 2011, 15, 579–593.
P. Yang, A. De Cian, M. P. Teulade-Fichou, J. L. Mergny, D. Monchaud,
Angew. Chem. Int. Ed. 2009, 48, 2188; Angew. Chem. 2009, 121, 2222–
2191.
[37]
[38]
M. Y. Kim, H. Vankayalapati, K. Shin-Ya, K. Wierzba, L. H. Hurley, J. Am.
Chem. Soc. 2002, 124, 2098–2099.
[5] J. Eddy, N. Maizels, Nucleic Acids Res. 2008, 36, 1321–1333.
[6] J. L. Huppert, S. Balasubramanian, Nucleic Acids Res. 2007, 35, 406–413.
[7] S. Balasubramanian, L. H. Hurley, S. Neidle, Nat. Rev. Drug Discovery 2011,
10, 261–275.
[8] T. Taka, L. Huang, A. Wongnoppavich, S. W. Tam-Chang, T. R. Lee, W.
Tuntiwechapikul, Bioorg. Med. Chem. 2013, 21, 883–890.
[9] A. Rangan, O. Y. Fedoroff, L. H. Hurley, J. Biol. Chem. 2001, 276, 4640–
4646.
D. Drygin, A. Siddiqui-Jain, S. O'Brien, M. Schwaebe, A. Lin, J. Bliesath,
C. B. Ho, C. Proffitt, K. Trent, J. P. Whitten, J. K. Lim, D. Von Hoff, K. An-
deres, W. G. Rice, Cancer Res. 2009, 69, 7653–7661.
F. Cuenca, O. Greciano, M. Gunaratnam, S. Haider, D. Munnur, R. Nan-
junda, W. D. Wilson, S. Neidle, Bioorg. Med. Chem. Lett. 2008, 18, 1668–
1673.
[39]
[10] C. Douarre, D. Gomez, H. Morjani, J. M. Zahm, M. F. O'Donohue, L. Edda-
bra, P. Mailliet, J. F. Riou, C. Trentesaux, Nucleic Acids Res. 2005, 33, 2192–
2203.
[11] H. Li, Y. Liu, S. Lin, G. Yuan, Chem. Eur. J. 2009, 15, 2445–2452.
[12] K. I. McLuckie, Z. A. Waller, D. A. Sanders, D. Alves, R. Rodriguez, J. Dash,
G. J. McKenzie, A. R. Venkitaraman, S. Balasubramanian, J. Am. Chem. Soc.
2011, 133, 2658–2663.
[13] G. Biffi, D. Tannahill, J. McCafferty, S. Balasubramanian, Nat. Chem. 2013,
5, 182–186.
[14] A. Henderson, Y. Wu, Y. C. Huang, E. A. Chavez, J. Platt, F. B. Johnson,
R. M. Brosh Jr., D. Sen, P. M. Lansdorp, Nucleic Acids Res. 2014, 42, 860–
869.
[40]
[41]
[42]
[43]
[44]
[45]
M. Gunaratnam, S. Swank, S. M. Haider, K. Galesa, A. P. Reszka, M. Beltran,
F. Cuenca, J. A. Fletcher, S. Neidle, J. Med. Chem. 2009, 52, 3774–3783.
G. Collie, A. P. Reszka, S. M. Haider, V. Gabelica, G. N. Parkinson, S. Neidle,
Chem. Commun. 2009, 7482–7484.
S. M. Hampel, A. Sidibe, M. Gunaratnam, J. F. Riou, S. Neidle, Bioorg. Med.
Chem. Lett. 2010, 20, 6459–6463.
G. W. Collie, R. Promontorio, S. M. Hampel, M. Micco, S. Neidle, G. N.
Parkinson, J. Am. Chem. Soc. 2012, 134, 2723–2731.
F. Doria, C. M. Gallati, M. Freccero, Org. Biomol. Chem. 2013, 11, 7838–
7842.
F. Doria, M. Nadai, M. Folini, M. Scalabrin, L. Germani, G. Sattin, M. Mella,
M. Palumbo, N. Zaffaroni, D. Fabris, M. Freccero, S. N. Richter, Chem. Eur.
J. 2013, 19, 78–81.
M. Nadai, F. Doria, L. Germani, S. N. Richter, M. Freccero, Chem. Eur. J.
2015, 21, 2330–2334.
F. Doria, I. Manet, V. Grande, S. Monti, M. Freccero, J. Org. Chem. 2013,
78, 8065–8073.
F. Doria, M. Nadai, G. Sattin, L. Pasotti, S. N. Richter, M. Freccero, Org.
Biomol. Chem. 2012, 10, 3830–3840.
P. Hazel, J. Huppert, S. Balasubramanian, S. Neidle, J. Am. Chem. Soc.
2004, 126, 16405–16415.
S. Redon, S. Bombard, M. A. Elizondo-Riojas, J. C. Chottard, Nucleic Acids
Res. 2003, 31, 1605–1613.
Y. Wang, D. J. Patel, Structure 1993, 1, 263–282.
L. N. Wen, M. X. Xie, Biochimie 2013, 95, 1185–1195.
H. Fernando, A. P. Reszka, J. Huppert, S. Ladame, S. Rankin, A. R. Venkitar-
aman, S. Neidle, S. Balasubramanian, Biochemistry 2006, 45, 7854–7860.
M. Micco, G. W. Collie, A. G. Dale, S. A. Ohnmacht, I. Pazitna, M. Gunarat-
nam, A. P. Reszka, S. Neidle, J. Med. Chem. 2013, 56, 2959–2974.
Glide, version 6.7, Schrödinger, LLC, New York, NY, 2015.
N. J. Greenfield, Nat. Protoc. 2006, 1, 2527–2535.
[15] S. Cogoi, M. Paramasivam, B. Spolaore, L. E. Xodo, Nucleic Acids Res. 2008,
36, 3765–3780.
[16] V. Gonzalez, K. Guo, L. Hurley, D. Sun, J. Biol. Chem. 2009, 284, 23622–
23635.
[46]
[47]
[48]
[49]
[50]
[17] M. J. Law, K. M. Lower, H. P. Voon, J. R. Hughes, D. Garrick, V. Viprakasit,
M. Mitson, M. De Gobbi, M. Marra, A. Morris, A. Abbott, S. P. Wilder, S.
Taylor, G. M. Santos, J. Cross, H. Ayyub, S. Jones, J. Ragoussis, D. Rhodes,
I. Dunham, D. R. Higgs, R. J. Gibbons, Cell 2011, 143, 367–378.
[18] J. Zimmer, E. M. C. Tacconi, C. Folio, S. Badie, M. Porru, K. Klare, M. Tumiati,
E. Markkanen, S. Halder, A. Ryan, S. P. Jackson, K. Ramadan, S. G. Kuznet-
sov, A. Biroccio, J. E. Sale, M. Tarsounas, Mol. Cell 2016, 61, 449–460.
[19] N. Beaume, R. Pathak, V. K. Yadav, S. Kota, H. S. Misra, H. K. Gautam, S.
Chowdhury, Nucleic Acids Res. 2013, 41, 76–89.
[20] P. Rawal, V. B. Kummarasetti, J. Ravindran, N. Kumar, K. Halder, R. Sharma,
M. Mukerji, S. K. Das, S. Chowdhury, Genome Res. 2006, 16, 644–655.
[21] M. Wieland, J. S. Hartig, Nat. Protoc. 2009, 4, 1632–1640.
[22] L. M. Harris, C. J. Merrick, PLoS Pathog. 2015, 11, e1004562.CCDC
1004562 (for XX1) contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre.
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[23] R. Perrone, M. Nadai, I. Frasson, J. A. Poe, E. Butovskaya, T. E. Smithgall,
M. Palumbo, G. Palu, S. N. Richter, J. Med. Chem. 2013, 56, 6521–6530.
M. Nadai, F. Doria, M. Di Antonio, G. Sattin, L. Germani, C. Percivalle, M.
Palumbo, S. N. Richter, M. Freccero, Biochimie 2011, 93, 1328–1340.
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[58] Schrödinger Suite 2015–2 Protein Preparation Wizard, LLC, New York, NY,
2015.
[63] Desmond Molecular Dynamics System, version 4.2, D. E. Shaw Research,
New York, NY, 2015.
[59] Maestro, version 10.2, Schrödinger, LLC, New York, NY, 2015.
[60] MacroModel, version 10.8, Schrödinger, LLC, New York, NY, 2015.
[61] Prime, version 4.0, Schrödinger, LLC, New York, NY, 2015.
[62] SigmaPlot, version 12.3, Systat Software, San Jose, CA, 2011.
Received: June 21, 2016
Published Online: ■
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
10
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Nucleic Acid Recognition
X-shaped naphthalene diimides with
synergic
donor
and
acceptor
F. Doria, M. Nadai, G. Costa, G. Sattin,
C. Gallati, G. Bergamaschi,
F. Moraca, S. Alcaro, M. Freccero,*
S. N. Richter* .................................... 1–11
hydrogen-bonding moieties selectively
bind parallel G-quadruplex nucleic
acids (c-myc, bcl-2 and parallel hTel22).
The hydrogen-bonding network in-
volving the outside rim of the G-quar-
tet is a key factor controlling the se-
lectivity
Extended Naphthalene Diimides
with Donor/Acceptor Hydrogen-
Bonding Properties Targeting G-
Quadruplex Nucleic Acids
DOI: 10.1002/ejoc.201600757
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
11
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim