Combining G-quadruplex targeting motifs on a single peptide nucleic acid scaffold: A hybrid (3 + 1) PNA-DNA bimolecular quadruplex

Article abstract of DOI:10.1002/chem.200800605

We describe the first G-quadruplex targeting approach that combines intercalation and hybridization strategies by investigating the interaction of a G-rich petide nucleic acid (PNA) acridone conjugate 1 with a three-repeat fragment of the human telomere G

Full text of DOI:10.1002/chem.200800605

DOI: 10.1002/chem.200800605
Combining G-QuadruplexTargeting Motifs on a Single Peptide Nucleic Acid
Scaffold: A Hybrid (3+1) PNA–DNA Bimolecular Quadruplex**
Alexis Paul,[a] Poulami Sengupta,[b] Yamuna Krishnan,[b] and Sylvain Ladame*[a]
Abstract: We describe the first G-
quadruplextargeting approach that
combines intercalation and hybridiza-
tion strategies by investigating the in-
teraction of a G-rich petide nucleic
acid (PNA) acridone conjugate 1 with
a three-repeat fragment of the human
telomere G3 to form a hybrid PNA–
DNA quadruplexthat mimicks the bio-
logically relevant (3+1) pure DNA di-
meric telomeric quadruplex. Using a
combination of UV and fluorescence
spectroscopy, circular dichroism (CD),
and mass-spectrometry, we show that
PNA 1 can induce the formation of a
bimolecular hybrid quadruplexeven at
low salt concentration upon interaction
PNA 1 cannot invade a short fragment
of B-DNA even if the latter contains a
CCC motif complementary to the PNA
sequence. These studies could open up
new possibilities for the design of a
novel generation of quadruplexligands
that target not only the external fea-
tures of the quadruplexbut also its
central core constituted by the tetrads
themselves.
with
a single-stranded three-repeat
fragment of telomeric DNA. However,
Keywords: acridone · circular di-
chroism · DNA recognition · eptide
nucleic acid · quadruplex
Introduction
It has been known for several decades that DNA sequences
containing a high density of clustered guanine units are able
to adopt four-stranded secondary structures named guanine
(G) quadruplexes or tetraplexes in the presence of physio-
logical cations, notably K+ and Na+ (Scheme 1).[1] Converg-
ing data collected from computer simulation and in vitro
have revealed the high prevalence of such G-rich DNA se-
quences throughout the human genome.[2] Recently, there
has been considerable focus on quadruplexes, their cellular
functions, and their exploitation for biological intervention
towards therapeutics. The most widely studied DNA quad-
ruplexes are those derived from telomeric repeat sequen-
ces.[3–6] Biophysical studies on the human telomeric quadru-
[a] A. Paul, Dr. S. Ladame
Institut de Science et d’IngØnierie
SupramolØculaires (ISIS), UniversitØ Louis Pasteur
CNRS UMR 7006, 8, AllØe Gaspard Monge
67083 Strasbourg CØdex(France)
Fax: ( +33)3902-45115
E-mail: s.ladame@isis.u-strasbg.fr
[b] P. Sengupta, Dr. Y. Krishnan
Scheme 1. Top: structure of the first generation acridone–PNA conjugate
1. Bottom: planar array of four guanine units or a G tetrad (left) and
schematic representation of a bimolecular (3+1) G-quadruplexstructure
National Centre for Biological Sciences
TIFR, Bellary Road, Bangalore, 560065 (India)
[**] 1 PNA+1 DNA=(3+1) quadruplex
(right) determined by NMR spectroscopy by Patel and co-workers.[16]
.
8682
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8682 – 8689

FULL PAPER
plexhave provided valuable insight into its structural and
dynamic properties.[3]
meric DNA by the formation of a bimolecular (3+1) hybrid
quadruplex.
NMR spectroscopic and X-ray crystallographic studies of
the intramolecular quadruplexformed from four human te-
lomeric repeats (GGGTTA) have revealed that it is highly
polymorphic.[4] Several lines of evidence link G quadruplex-
es with telomere maintenance[5], given that telomeric DNA
in the quadruplexform is not a competent substrate for te-
lomerase.[6] Telomerase is crucial for immortality in most
human cancers, and therefore ligand-induced quadruplex
stabilization in telomeric DNA has potential as an antican-
Results and Discussion
Design and synthesis: We designed PNA–acridone conjugate
1 such that 1) it incorporated a lysine moiety for water solu-
bility, 2) it contained three consecutive guanine units that
could each be part of a different G tetrad, and 3) the acri-
done was linked to PNA by a flexible linker and can there-
fore stack favorably with the top G tetrad of the hybrid
quadruplexonce formed. Compound 1 was synthesized on
rink amide 4-methylbenzhydrylamine (MBHA) resin
(Merck Biosciences) using standard (9H-fluoren-9-yl-
A
quadruplexes include small molecules,[8] complementary oli-
gonucleotides,[9] or engineered DNA-binding proteins.[10] An
original approach was also recently reported that uses an an-
thracene/diethylene triamine conjugate to induce intramo-
lecular parallel quadruplexformation from single-stranded
DNA through stacking of the anthracene moiety onto the
external G tetard and invasion of the central ion channel by
the triamine side chain.[11] Peptide nucleic acid (PNA) is a
synthetic mimic of DNA in which the negatively charged
phosphate backbone is replaced by a neutral polyamide
backbone, thus allowing it to hybridise in a sequence specif-
ic manner and with high affinity to either DNA or RNA.
Recently, it has been shown that carefully designed G-rich
PNA oligomers were also capable of forming PNA4 quadru-
methoxy)carbonyl (Fmoc) chemistry. The structure of the
acridone was based on the bis(aminoalkylamido)acridone
quadruplexligands [17] and was attached to the PNA by a gly-
cine linker. The acridone monomer 6 was synthesized in so-
lution in sixsteps that required no purification on silica gel
(Scheme 2).
plexes[12]
(PNA–DNA chimeras)4
quadruplexes.
,
hybrid PNA2–DNA2,[13] PNA2—DNA,[14] or
[15]
tetramolecular and trimolecular
A recent NMR spectroscopic study reported an unprece-
dented structure of a DNA fragment of the human telomer-
ic region containing three G-rich repeats, namely,
d(GGGTTAGGGTTAGGGT) in a solution of sodium cat-
ions.[16] An asymmetric dimeric quadruplexwas formed in
which the G-tetrad core involved all three G tracts of one of
the strands and the G tract at the 3’-end of the second
strand. The three-repeat sequence could also associate with
Scheme 2. Synthesis of the acridone carboxylic acid building block 6. Re-
agents and conditions: a) Cu, Cu2O, Cu(OAc)2, DMF, 1608C, 1 h (50%);
A
the solitary G-rich unit d(GGGTTA) to form a bimolecular
A
b) PPA, 1308C, 45 min (85%); c) SnCl2·2H2O, AcOH, HCl, 15 min, RT,
then 45 min, 808C (66%); d) 3-CPC, 608C, 24 h (85%); e) pyrrolidine,
DMF, 608C, 30 min (92%); f) NaOH, H2O, DMF, 508C, 40 min (89%).
DMF=N,N-dimethylformamide, PPA=polyphosphonic acid, 3-CPC=3-
chloropropionyl chloride.
complexcalled the (3 +1) assembly. We, therefore, reasoned
that this bimolecular (3+1) structure could represent the
basis of a promising new design strategy for G-quadruplex
stabilizing ligands. These ligands could combine on a single
scaffold, binding elements that target simultaneously the
central core of the G-quadruplexstructure and key external
features, that is, a G-rich PNA strand attached to a quadru-
plexbinding platform. The G-rich PNA region could hybrid-
ize with the G-rich DNA clusters to form three hybrid G
tetrads, whereas the quadruplexbinding platform could
stack on the resultant terminal G tetrad of the hybrid quad-
ruplexcore. We chose an acridone as our DNA-binding
platform as it is a well-characterized G-tetrad end-stacker[17]
for which high quadruplexversus duplexDNA discrimina-
tion capabilities have been demonstrated when appending
appropriate side chains (e.g., 3-pyrrolidine propionamide).
We report herein the synthesis of ligand 1, a G-rich PNA–
acridone conjugate (Scheme 1), and demonstrate the suc-
cessful targeting of a three-repeat fragment of human telo-
The Ullman reaction between commercially available
methyl anthranilate and 2-chloro-5-nitrobenzoic acid by
using a modified version of a previously reported protocol[18]
afforded the bis(aryl)amine, which was almost quantitatively
A
converted into the corresponding acridone 2 after treatment
in polyphosphoric acid and precipitation in water. Reduc-
tion of the nitro group with tin(II) chloride under acidic
conditions afforded the amino acridone 3, which was treated
with 3-chloropropionyl chloride to form the corresponding
acridone amide 4. Nucleophilic addition of pyrrolidine on
the aliphatic halide generated the desired 3-pyrrolidine pro-
prionamide-substituted acridone methyl ester 5, which was
Chem. Eur. J. 2008, 14, 8682 – 8689
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8683

S. Ladame et al.
finally hydrolyzed under basic conditions to afford the de-
sired acridone carboxylate 6.
Characterization of a DNA1/PNA1 quadruplex: By using a
combination of UV, circular dichroism (CD), and fluores-
cence spectroscopic and mass-spectrometric analysis, we
have established the interactions between a three-repeat
fragment of human telomeric DNA d(GGGTTAGGGT-
TAGGG) G3 and the PNA–acridone 1. Fragment G3
formed a stable quadruplex( Tm =59.58C) in potassium
phosphate buffer (100 mm, pH 7.4 also containing 100 mm
KCl) at a strand concentration of 20 mm. Fragment G3 ex-
hibited a CD signature with two maxima at 260 and 295 nm
characteristic of a mixture of parallel and antiparallel G-
quadruplexconformations, respectively. When annealed in
the presence of an equimolar amount of PNA 1, the UV
thermal denaturation curve showed a melting transition at
Tm =60.58C, only one degree higher than for the DNA
alone (Table 1). However the CD profile was significantly
Table 1. Melting temperatures of DNA (20 mm G3) and DNA/PNA
(20 mm G3+20 mm PNA 1) in buffers containing different potassium con-
centrations.
Figure 1. CD spectra of solutions of 20 mm G3 and 20 mm G3+20 mm
PNA 1 annealed in either potassium phosphate buffer 100 mm at pH 7.4
containing 100 mm KCl (empty triangles and squares), potassium phos-
phate buffer 100 mm at pH 7.4 (full triangles and squares) or neat water
at pH 7.4 (half-full triangles and squares).
Buffer composition
Complex
Tm [8C]
DTm [8C][a]
100mm KH2PO4
pH 7.4+100mm KCl
G3
G3+PNA 1
G3
G3+PNA 1
G3
G3+PNA 1
59.5
60.5
51.0
55.0
<10
23.0
+1
100mm KH2PO4 pH 7.4
+4
functional groups on the guanine units between the tet-
rads.[19] Contrary to this behavior, noncovalent PNA com-
plexes were shown to be slightly destabilized by cations[20]
and the previously reported PNA4 quadruplexshowed com-
parable stability when annealed either in neat water or
100 mm sodium phosphate buffer.[12] Herein, we demonstrate
that the PNA/DNA complexis stabilized by potassium cat-
ions, although not as strongly as a pure DNA quadruplex. It
is also noteworthy that in experiments carried out with stoi-
chiometric mixtures of G3 and acridone 6 that lack the
PNA strand, no thermal stabilization or structural changes
of the quadruplexwere observed (data not shown). This
finding rules out the possibility of a conformational change
of the G3 dimeric quadruplexinduced by the acridone only
and reinforces the initial hypothesis of an active participa-
tion of the PNA in PNA–DNA hybrid quadruplexforma-
tion.
CD spectroscopic experiments were also carried out on
G3 and G3/PNA 1 complexes annealed either in potassium
phosphate buffer (with or without 100 mm KCl) or neat
water at pH 7.4 (Figure 1). Fragment G3 formed a mixed
parallel/antiparallel quadruplexin potassium phosphate
buffer but remained unstructured when annealed in neat
water, which is consistent with the reported data.[3] When
annealed in the presence of an equimolar amount of PNA 1,
G3 formed a mixed parallel/antiparallel quadruplex in po-
tassium phosphate buffer and neat water, which is consistent
with the UV/melting experiments that demonstrated that
PNA 1 could induce the formation of a quadruplexeven
under conditions in which the dimeric DNA quadruplex
>+13
H20 pH 7.4
[a] Difference in Tm values obtained under identical conditions for the
solutions of 20 mm G3+20 mm PNA 1 and 20 mm G3.
different with a main maximum at l=260 nm and a smaller
one at l=295 nm (Figure 1), thus indicting either 1) a
ligand-induced conformational switch of the dimeric DNA
quadruplexor 2) the formation of an alternative hybrid
PNA–DNA quadruplexstructure that accounts for the new
CD signature.
Similar CD and UV spectroscopic experiments were car-
ried out but with decreasing concentrations of salt (either in
100 mm potassium phosphate buffer with no added KCl first
or in neat water at controlled pH 7.4). As expected for the
20 mm solution of G3, the Tm value decreased significantly
on lowering the potassium concentration: the Tm value went
from 59.5 to 518C on removing the additional 100 mm KCl
of the initial potassium phosphate buffer, and Tm <108C
when annealed in neat water (Table 1). A salt-dependent
shift in the Tm values was also observed for the equimolar
PNA/DNA mixture: the Tm value went from 60.5 to 558C
with no added KCl and Tm =238C in neat water (Table 1).
Interestingly, the difference in the Tm values between G3
alone and G3+1 equiv PNA 1 gradually increases on de-
creasing the salt concentration.
Cation-mediated stabilization of DNA quadruplexes is
mainly through 1) electrostatic screening of the negatively
charged DNA strands and 2) coordination of the eight O6
8684
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8682 – 8689

A Hybrid Quadruplexon a Peptide Nucleic Acid Scaffold
FULL PAPER
(nano ESI-MS) on a mixture of DNA G3 and PNA 1. The
nano ESI mass spectrum showed multiple equidistant peaks
in the regime m/z 1540–1560 (Figure 2). The peak centred at
m/z 1546.2 showed fine structure (Figure 2, inset) with a
constant separation of 0.25ꢀ0.01, which indicates that the
peaks in this regime correspond to a quadruply charged spe-
cies. The associated molecular weight of this species (MW =
6181.5ꢀ2.5 Da) corresponds to a bimolecular complexthat
is composed of one DNA strand and one PNA strand (cal-
culated molecular weight=6181.6 Da). Moreover, the family
of peaks in this regime centred at m/z 1550.59 and 1555.57
corresponded to the same complexassociated with one
+
NH4 and one K+ cation, respectively. These results confirm
the formation of a (3+1) hybrid PNA1–DNA1 quadruplex.
To delineate the ability of acridone-bearing PNA 1 to effi-
ciently target such three G repeat sequences, its binding
with the former was further investigated. Quadruplexli-
gands such as quinacridines[22] or carbazoles[23] exhibit
changes in their fluorescent properties upon quadruplex
binding that efficiently report on their DNA binding affinity
and/or binding mode. Fluorescently labeled nucleic acid
probes that are capable of detecting DNA single-base mis-
matches were recently reported that exploit the fact that the
fluorescence of the intercalator is directly responsive to
local perturbations that arise upon hybridization of the
probe to the DNA target.[24] In these systems, interaction of
the fluorophore with G bases or G–C base pairs often result
in a significant quenching of fluorescence.
The emission of fluorescence of a solution of PNA 1 was
recorded under different conditions to ascertain whether
indeed the acridone moiety behaved as a quadruplexbind-
ing ligand (Figure 3). When 3 mm PNA 1 was annealed with
3 mm DNA G3 under quadruplexforming conditions, a pal-
pable increase in the fluorescence intensity (ꢁ40%) was
observed. This fluorescence enhancement was also accompa-
nied by a moderate red shift from l=431 to 436 nm. How-
ever, when 3 mm PNA 1 was incubated with 3 mm ssDNA (of
sequence similar to that of G3 but carrying three G!T mu-
tations to prevent quadruplexformation), it did not lead to
any significant changes in the fluorescence properties of the
acridone moiety. This result is consistent with a specific in-
teraction of PNA 1 with G3 that additionally incorporates a
stacking mode of the acridone moiety on a quadruplexcore.
Indeed, whereas in the case of PNA 1 the fluorescence of
the acridone is likely to be quenched by free neighboring G
bases, a structuration of PNA 1 within a quadruplexconfor-
mation, and of the three free guanine units in G-tertad for-
mation could prevent such quenching and account for the
observed fluorescence exhaltation.
Figure 2. Nano ESI mass spectrum of an equimolar solution of DNA G3
and PNA 1 annealed in ammonium acetate buffer.
cannot form. Interestingly, 1) the absolute amount of quad-
ruplexformed decreased on decreasing the salt concentra-
tion, 2) the ratio of emissions at l=260/295 nm in the CD
spectra remains almost constant for either the G3 or G3/
PNA 1 complexwhatever the buffer, and 3) this ratio re-
mains significantly higher for the PNA/DNA quadruplex
than for the pure DNA quadruplex. Taken all together,
these CD and UV spectroscopic data confirm that G3
forms a stable mixed parallel/antiparallel quadruplex struc-
ture, the stability of which is highly dependent on the salt
concentration. When annealed with a stoichiometric amount
of PNA 1, G3 also folds into a quadruplex, but with a differ-
ent conformation (more parallel-like), and the stability of
this new quadruplexis significantly less influenced by the
salt concentration. This outcome is consistent with the for-
mation of a hybrid PNA/DNA quadruplexwith a participa-
tion of the PNA guanine units into mixed tetrads with the
guanine moieties of the G3 DNA strand. The difference in
the CD spectra between the pure DNA and hybrid PNA/
DNA quadruplexcould be epxlained by an acridone-in-
duced orientation of the PNA strand. Indeed, whereas in
the dimeric DNA quadruplexthe lagging strand that partici-
pates in only a quarter of each G tetrad can freely and
equally orientate in either way (all three guanine units are
in a syn or anti conformation), it is anticipated that the acri-
done will preferentially bind to the most polar face of a
quadruplex, and by this means will direct the positioning of
the PNA with respect to the DNA strand. Indeed, it was
previously reported that as a consequence of the position of
the loops, quadruplexes often have one face rich in nega-
tive-charge potential whereas the other is more hydropho-
bic.[21] Given that the acridone possesses a positively charged
side chain it is likely that it will preferentially interact with
the most polar face.
To further investigate the nature of the binding interac-
tion between PNA 1 and the DNA fragment G3, we first ti-
trated a solution of 1 (3 mm) with increasing amounts of G3
prefolded into the quadruplexform as confirmed by CD
spectroscopic analysis. This procedure resulted in a gradual
decrease in the fluorescence intensity of the acridone
moiety. Interestingly, when the same experiment was per-
formed with increasing amounts of G3 in the unfolded
To determine the stoichiometry of this quadruplex, we
performed nano-electospray ionization mass spectrometry
Chem. Eur. J. 2008, 14, 8682 – 8689
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8685

S. Ladame et al.
Scheme 3. Proposed modes of interaction of PNA 1 with G3 DNA.
Figure 3. Top: fluorescence emission spectra (lex =410 nm) of 3 mm PNA
(dark-grey triangles), 3 mm PNA+3 mm ssDNA (pale grey squares), and
3 mm PNA+3 mm G3 DNA (black circles) in 100 mm potassium phos-
phate buffer at pH 7.4 containing 100 mm KCl. Bottom: plots of F/F0
(F=fluorescence intensity) versus the concentration of DNA G3 for the
titration of a 3 mm solution of PNA 1 in potassium phosphate buffer
100 mm at pH 7.4 containing 100 mm KCl with a solution of annealed G3
(black squares) or unstructured G3 (black triangles). Excitation was at
l=410 nm and emission was recorded at l=436 nm.
single-stranded state it led to a significant increase in acri-
done fluorescence (Figure 3). Additionally, when a fixed
amount of PNA 1 was annealed with increasing amounts of
DNA G3 an increase in fluorescence intensity was observed
(data not shown) similar to that seen for the unfolded DNA
G3 (Figure 3). These results reflect two different binding
modes for 1 on interacting with either a preformed quadru-
plexor a quadruplexforming sequence in the single-strand-
ed state. Unlike nonfunctionalized G-rich PNA, PNA 1 is
not capable of invading the dimeric quadruplexof G3 once
formed within the timescales investigated. However, it is ca-
pable of inducing G-rich single-stranded DNA folding into a
quadruplexconformation even at room temperature and in
neat water, thus leading to a thermodynamically favored
hybrid structure (Scheme 3).
To demonstrate the influence of the acridone moiety on
both the stability and conformation of the complexformed
between PNA 1 and G3, an analogue of 1 that lacks the ac-
ridone platform, PNA 1b (of general sequence Lys-GGG-
NH2), was synthesized (Figure 4). The ability of 1b to form
a stable bimolecular quadruplexupon binding to DNA G3
was investigated using both UV and CD spectroscopic anal-
Figure 4. Structure of PNA 1b and CD spectra of a stoichiometric mix-
ture of (20 mm G3+20 mm PNA 1b) annealed in either in potassium
phosphate buffer 100 mm at pH 7.4 (full line) or water at pH 7.4 (dotted
line).
ysis in either 100 mm potassium phosphate buffer (pH 7.4)
with no added KCl or neat water (pH 7.4). A stoichiometric
mixture of PNA 1b and DNA G3 formed a stable quadru-
plex( Tm =52.58C) in 100 mm potassium phosphate buffer
that was slightly more stable than the G3 dimeric quadru-
plex( +1.58C) but less stable than the bimolecular hybrid
quadruplexformed from PNA 1 and G3 (ꢂ2.58C) under
similar conditions. The CD spectrum of this new hybrid 1b/
G3 quadruplexrevealed a mixed parallel/antiparallel con-
formation with a main maximum at l=260 nm (Figure 4),
as previously observed with PNA 1. More striking is the ob-
servation that PNA 1b (which lacks the acridone platform)
proved unable to form a hybrid PNA/DNA quadruplexin
8686
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8682 – 8689

A Hybrid Quadruplexon a Peptide Nucleic Acid Scaffold
FULL PAPER
the absence of any salt, unlike that observed with PNA acri-
done conjugate 1 (Figure 4). This observation along with the
differences in Tm values measured in phosphate buffer dem-
onstrates the influence of the acridone moiety on the stabili-
ty of the hybrid quadruplexformed.
Interaction of PNA 1 with duplexDNA : The ability of the
PNA acridone conjugate to stabilize double-stranded DNA
was also examined using UV and fluorescence spectroscopic
analysis. PNA binding to double-stranded DNA can happen
through different binding modes. For example, one PNA
strand can interact with two complementary DNA strands
to form a PNA–DNA triplexor can also lead to duplexin-
vasion with the formation of a PNA–DNA heteroduplex.
Such an invasion has been demonstrated using a homopur-
ine PNA decamer that only forms a duplexwith the comple-
mentary oligonucleotides.[25] Herein, we investigated the ca-
pacity of our PNA 1 to either invade or associate with two
B-DNA undecamers containing either a CCC motif comple-
mentary to the GGG PNA sequence (duplexB) or no se-
quence complementarity with PNA 1 (duplexA).
UV thermal denaturation experiments were carried out
on duplexes A and B alone and with a stoichiometric
amount of PNA 1. Melting temperatures (Tm) of 538C were
obtained for both duplexes A and B. Interestingly, no shift
in the Tm values was observed on incubating either duplex
A or B with PNA 1, thus suggesting that our GGG PNA ac-
ridone conjugate cannot invade and stabilize B-DNA, even
if the latter contains a CCC complementary sequence
(Figure 5). To further investigate the specificity of PNA 1
toward quadruplexversus double-stranded DNA, additional
UV thermal denaturation studies were carried out under
low salt conditions (i.e., 10 mm potassium phosphate buffer,
pH 7.4 and no added salt). Conditions employing low salt
Figure 5. Top: Virtually superposed UV thermal denaturation curves
(heating curve) of 4 mm duplexA, 4 mm duplexA +4 mm PNA 1, 4 mm
duplexB, and 4 mm duplexB +4 mm PNA 1. Buffer was potassium phos-
phate buffer 100 mm at pH 7.4 containing 100 mm KCl. Bottom: plots of
F/F0 versus B-DNA concentration for the titration of a 3 mm solution of
PNA 1 in potassium phosphate buffer 100 mm at pH 7.4 containing
100 mm KCl, with a solution of either noncomplementary duplexA
(squares) or complementary duplexB (circles). Ecxitation was at
l=
410 nm and emission was recorded at l=436 nm.
concentrations were previously reported to favor PNA bind-
Conclusion
[26]
ing to double-helixDNA.
Under those conditions, duplex
B proved moderately stable (Tm =368C). When incubated
with a stoichiometric amount of PNA 1, duplexB showed a
similar melting temperature of 368C, indicative again of an
absence of PNA-induced B-DNA stabilization. It is also
noteworthy that 1) most PNA moieties commonly designed
to invade duplexDNA and therefore interfere with DNA
transcription are significantly longer than PNA 1 and 2) hy-
bridization of 1 with a single strand of DNA containing a
complementary CCC motif (i.e., ACGCCCTAAGC) did not
lead to the formation of stable hybrid PNA/DNA duplexes
or triplexes.
On titrating in a solution of 1 (3 mm) with increasing
amounts of B-DNA, a decrease in the fluorescence intensity
of the acridone was observed, which is comparable to that
observed for the titration with a solution of folded G3 quad-
ruplex(Figure 5). This finding is consistent with our previ-
ous model that the nonspecific interaction of PNA 1 with
either prefolded quadruplexor duplexDNA results in fluo-
rescence quenching, whereas only the specific involvement
of the PNA guanine units in G-tetrad formation leads to a
significant fluorescence enhancement.
We have reported a novel approach to target the human te-
lomeric DNA sequence that combines a hybridization and
an intercalation motif within a single scaffold. We have
shown that this strategy is successful and have characterized
the resultant (3+1) hybrid PNA1/DNA1 quadruplex, which
is the first of a new genre of hybrid quadruplexes that also
mimics one of the biologically relevant structures of the
human telomeric quadruplex. The hybrid PNA–DNA bimo-
lecular quadruplexshowed increased stability at low salt
concentration relative to the corresponding DNA2 dimeric
quadruplex. These studies open up new possibilities for the
next generation of quadruplex ligands that target not only
the external features of the quadruplex but also its central
core constituted by the tetrads themselves. The absence of
PNA-induced B-DNA stabilization also reinforces the po-
tential of such conjugates as quadruplexspecific ligands.
However, because of the inability of PNA 1 to invade a di-
meric quadruplexby displacement of one DNA strand,
PNAs with increased affinity will be necessary before such
quadruplextargeting strategy can be used in vivo. The
design of second generation ligands containing two quadru-
Chem. Eur. J. 2008, 14, 8682 – 8689
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8687

S. Ladame et al.
ethyl ether, thus affording the desired product pure as a pale-brown solid
(0.6 g, 85%). 1H NMR ([D6]DMSO, 400 MHz): d=7.61 (d, J=8.4 Hz,
1H), 7.47 (d, J=8 Hz, 1H), 7.32 (s, 3H), 7.26 (d, J=8.4 Hz, 1H), 6.44 (t,
J=8 Hz, 1H), 6.40 (d, J=8 Hz, 1H), 3.08 (s, 3H), 3.02 (t, J=6 Hz, 4H),
2.03 ppm (t, J=6 Hz, 2H); 13C NMR (DMSO, 100 MHz): d=176, 169.5,
plexbinding platforms at both ends of the PNA strand that
could significantly improve the stability of the hybrid quad-
ruplexare currently being developed. Considering the in-
creasing interest in DNA quadruplexes as either therapeutic
targets in biology or supramolecular objects in nanosciences,
ligands that can target DNA sequences containing only
three clusters of guanine units, even under conditions in
which DNA quadruplexes cannot readily form (low salt con-
centration), could be highly valuable.
168, 144.5, 141.4, 141.3, 136.5, 133, 127.5, 122, 120.5, 117, 115.2, 114.8,
+
106.4, 53.2, 41, 24 ppm; HRMS: m/z calcd for C18H15ClLiN2O4
:
365.0875; found: 365.0861.
Methyl-3-[3-(pyrrolidin-1-yl)propanamido]-acridone-5-carboxylate
(5):
Pyrrolidine (500 mL, 5.6 mmol) was added to a solution of 4 (0.5 g,
1.4 mmol) in DMF (2.5 mL). The reaction mixture was stirred at 608C
for 30 min until the reaction had gone to completion. After cooling to
room temperature, diethyl ether was added to precipitate a yellow solid.
After recrystallization from diethyl ether, the desired product was ob-
Experimental Section
tained pure as a
yellow solid (0.5 g, 92%). 1H NMR ([D6]DMSO,
400 MHz): d=8.54 ( d, J=7.8 Hz, 1H), 8.42 (d, J=7.8 Hz, 1H), 8.20 (s,
1H), 8.20 (d, J=7.8 Hz, 1H), 7.39 (t, J=7.8 Hz, 1H), 7.38 (d, J=7.8 Hz,
1H), 4.00 (s, 3H), 3.47 (t, J=6 Hz, 2H), 1.85 (t, J=6 Hz, 2H), 3.10 (brm,
4H), 1.97 ppm (brm, 4H); 13C NMR ([D6]DMSO, 100 MHz): d=176,
169.5, 168, 144.5, 141.4, 141.3, 136.5, 133, 127.5, 122, 120.5, 117, 115.2,
114.8, 106, (53.5, 2C), 53.2, 50, 33 ppm (23, 2C); HRMS: m/z calcd for
C22H24N3O4+: 394.1761; found: 394.1789.
1H and 13C NMR spectra were recorded by using a Bruker Avance DRX
400 spectrometer at 400 and 100.6 MHz, respectively. Chemical shifts are
reported as d values (ppm) with reference to the residual solvent peaks.
All the reagents and solvents were obtained from commercial sources
and used without further purification. Purification of PNA acridone con-
jugate 1 was performed using a Gilson HPLC using a reverse-phase C18
column with gradients combining buffers A and B: buffer A=H2O+
0.1% trifluoroacetic acid (TFA)); buffer B=(MeCN+0.1% TFA).
Sodium-3-[3-(pyrrolidin-1-yl)propanamido]-acridone-5-carboxylate (6):
An aqueous solution of NaOH (0.96 mL, 1m, 1.5 equiv) was added to a
solution of 5 (0.25 g, 0.64 mmol) in DMF (10 mL). The reaction mixture
was stirred at 508C for 40 min. After removal of the water through evap-
oration, diethyl ether was added. The resulting precipitate was collected
by filtration and triturated with diethyl ether until a yellow solid was ob-
tained. The solid was washed with ethyl acetate and diethyl ether, thus
leading to the desired carboxylate acridone pure as a light-yellow solid
(0.23 g, 89%). 1H NMR (D2O, 400 MHz): d=7.88 (d, J=7.2 Hz, 1H),
7.72 (d, J=7.2 Hz, 1H), 7.18 (d, J=8.8 Hz, 1H), 6.85 (t, J=7.2 Hz, 1H),
6.51 (s, 3H), 6.17 (d, J=8.8 Hz, 1H), 2.64 (t, J=7.2 Hz, 2H), 2.55 (brm,
4H, brm), 2.20 (t, J=7.2 Hz, 2H), 1.73 ppm (brm, 4H); 13C NMR (D2O,
100 MHz): d=177, 173, 171, 141.5, 140, 139, 136, 129, 126, 120.5, 119.9,
119.8, 115, 113.5, 104, (53.5, 2C), 50, 35 ppm (23, 2C); HRMS:
m/z calcd for C21H22N3O4+: 380.1605; found: 380.1616.
2-{[2-(Methoxycarbonyl)phenyl]amino}-para-nitrobenzoic acid:[17] A mix-
ture of 2-chloro-para-nitrobenzoic acid (5 g, 25 mmol), methyl anthrani-
late (5 g, 33 mmol), copper powder (190 mg, 3 mmol), Cu2O (140 mg,
1 mmol), and Cu(OAc)2 (360 mg, 2 mmol) were stirred at 1608C in DMF
R
(120 mL) for 3 h and then allowed to slowly warm to room temperature.
After filtration through a thin layer of silica between two layers of celite,
the addition of aqueous 0.1m HCl (120 mL) and vigorous stirring led to
the formation of an orange precipitate. The desired product was obtained
as an orange solid (4 g) by filtration and used for the subsequent step
1
without further purification (50%). H NMR ([D6]DMSO, 400 MHz): d=
8.15 (brm, 2H), 7.98 (d, J=7.6 Hz, 1H), 7.67 (d, J=7.8 Hz, 1H), 7.66 (s,
1H), 7.60 (t, J=7.8 Hz, 1H), 7.17 (t, J=7.6 Hz, 1H), 3.88 ppm (s, 3H).
Methyl-3-nitro-acridone-5-carboxylate (2):[17] A mixture of 2-{[2-(meth-
MALDI-TOF spectra of PNA 1 and 1b: PNA 1 and 1b were synthesized
on a solid support using Fmoc chemistry. The PNA strand was cleaved
from the resin with a 95% TFA solution and purified by HPLC. The
final compounds were characterized by HRMS (MALDI-TOF): PNA 1:
m/z: calcd for C62H77N28O14: 1437.617; found: 1437.675 [M+H]+. PNA
1b: m/z: calcd for C39H55N24O10: 1019.446; found: 1019.542 [M+H]+.
A
polyphosphoric acid (10 g) was heated at 1308C for 45 min. Water was
added after cooling to room temperature, which led to the formation of a
bright-yellow solid. The solid was removed by filtration, washed with
water, and dried under vacuum. The acridone compound was obtained as
a yellow solid (0.8 g, 85%). 1H NMR ([D6]DMSO, 400 MHz): d=9.02 (s,
1H), 8.55 (d, J=7.2 Hz, 1H), 8.47 (d, J=7.2 Hz, 1H), 8.41 (d, J=8.4 Hz,
1H), 8.02 (d, J=8.4 Hz, 1H), 7.44 (t, J=7.2 Hz, 1H), 4.03 ppm (s, 3H).
DNA oligonucleotide preparation: All oligonucleotides were purchased
from Sigma Genosis. All concentrations were expressed in strand molari-
Methyl-3-aminoacridone-5-carboxylate
(3):[17] A mixture of 2 (1 g, 3.4 mmol),
tin(II) chloride (3 g), and concentrated
HCl (5 mL) was stirred vigorously for
20 min in glacial AcOH (20 mL). The
reaction mixture was heated at 808C
for 40 min and cooled to room temper-
ature before water (80 mL) was added
to allow the amino acridone to precipi-
tate. After filtration and subsequent
washes with ethyl acetate and diethyl
ether, the desired product was ob-
tained pure as a yellow powder (0.6 g,
66%). 1H NMR ([D6]DMSO, 400 MHz): d=8.46 (dd, 2J=8 Hz, 3J=
1.8 Hz, 1H), 8.33 (dd, 2J=8 Hz, 3J=1.8 Hz, 1H), 7.93 (d, J=8 Hz, 1H),
ty with a nearest-neighbor approximation for the absorption concentra-
tions of the unfolded species. The DNA and hybrid PNA: DNA quadru-
plexes were prepared in 100 mm potassium phosphate buffer containing
100 mm KCl at pH 7.4. After heating to 958C for 5 min, the solution was
slowly cooled for 6 h to room temperature and the oligonucleotide solu-
tions were finally stored at 48C.
7.28 (t, J=8 Hz, 1H), 6.62 (dd, 2J=8 Hz, 3J=1.8 Hz, 1H), 6.54 (d, J=
+
1.8 Hz, 1H), 3.98 ppm (s, 3H); HRMS: m/z calcd for C15H13N2O3
:
269.0921; found: 269.0914.
Methyl-3-(3-chloropropionamide)-acridone-5-carboxylate (4): A suspen-
sion of 3 (0.5 g, 1.9 mmol) in 3-chloropropionyl chloride (20 mL) was
heated at 608C for 24 h. The product precipitated at room temperature
by the addition of diethyl ether and was collected by filtration. The re-
sulting dark-yellow powder was washed with ethyl acetate, water, and di-
Oligonucleotide sequences: G3 d(GGGTTAGGGTTAGGG), ssDNA
d(GTGTTAGTGTTAGTG), duplexA d(GCATAGTGCGT) hybridized
with its complementary sequence, and duplexB d(GCTTAGGGCGT)
hybridized with its complementary sequence.
8688
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8682 – 8689

A Hybrid Quadruplexon a Peptide Nucleic Acid Scaffold
FULL PAPER
Fluorescence spectroscopy: Fluorescence emission spectra were recorded
in quartz cells at 208C by means of a Jobin Yvon Fluorolog 3.22 instru-
ment. The excitation and emission bandwidths were fixed to 5 and 5 nm,
respectively. Fluorescence titration experiments were carried out by
using a 500-mL quartz cuvette containing a solution of 3 mm PNA–acri-
done conjugate 1 in 100 mm potassium phosphate buffer containing
100 mm KCl (pH 7.4). Concentrated DNA aliquots (from a 150 mm stock
solution) were directly added to the PNA solution. The spectra were re-
corded between l=420 and 520 nm while exciting at 410 nm.
Res. 2005, 33, 2901–2907; c) J. Eddy, N. Maizels, Nucleic Acids Res.
2008, 36, 1321–1333.
[3] I. M. Pedroso, L. F. Duarte, G. Yanez, K. Burkewitz, T. M. Fletcher,
Biopolymers 2007, 87, 74–84.
[4] D. J. Patel, A. T. Phan, V. Kuryavyi, Nucleic Acids Res. 2007, 35,
7429–7455.
[5] K. Paeschke, T. Simonsson, J. Postberg, D. Rhodes, H. J. Lipps, Nat.
Struct. Mol. Biol. 2005, 12, 847–854.
[6] A. M. Zahler, J. R. Williamson, T. R. Cech, D. M. Prescott, Nature
1991, 350, 718–720.
[7] W. C. Hahn, S. A. Stewart, M. W. Brooks, S. G. York, E. Eaton, A.
Kurachi, R. L. Beijersbergen, J. H. Knoll, M. Meyerson, R. A. Wein-
berg, Nature Med. 1999, 5, 1164–1170.
UV spectroscopy: Absorption spectra were recorded by using a Uvicon
XL spectrophotometer. Quadruplexand duplexDNA melting curves
were recorded at wavelengths of l=295 and 260 nm, respectively, be-
tween 15 and 908C at 0.58Cminꢂ1. For quadruplexmelting, the DNA
(and PNA when appropriate) concentration was 20 mm. For duplexmelt-
ing, the DNA (and PNA when appropriate) concentration was 4 mm.
[8] For a recent review, see: D. Monchaud, M.-P. Teulade-Fichou, Org.
Biomol. Chem. 2008, 6, 627–636.
[9] B. Datta, B. Armitage, J. Am. Chem. Soc. 2001, 123, 9612–9619.
[10] S. Ladame, J.A. Schouten, J. Roldan, J. E. Redman, S. Neidle, S. Ba-
lasubramanian, Biochemistry 2006, 45, 1393–1399.
[11] R. Rodriguez, G. D. Pantos, D. P. N. GonÅalvez, J.K. M. Sanders, S.
Balasubramanian, Angew. Chem. 2007, 119, 5501–5503; Angew.
Chem. Int. Ed. 2007, 46, 5405–5407.
[12] Y. Krishnan-Ghosh, E. Stephens, S. Balasubramanian, J. Am. Chem.
Soc. 2004, 126, 5944–5945.
[13] B. Datta, C. Schmitt, B. A. Armitage, J. Am. Chem. Soc. 2003, 125,
4111–4118.
[14] S. Roy, F. A. Tanious, W. D. Wilson, D. H. Ly, B. A. Armitage, Bio-
chemistry 2007, 46, 10433–10443.
[15] L. Petraccone, B. Pagano, V. Esposito, A. Randazzo, G. Piccialli, G.
Barone, C. A. Mattia, C. Giancola, J. Am. Chem. Soc. 2005, 127,
16215–16223.
[16] N. Zhang, A. T. Phan, D. J. Patel, J. Am. Chem. Soc. 2005, 127,
17277–17285.
[17] R. J. Harrison, A. P. Reszka, S. M. Haider, B. Romagnoli, J. Morrell,
M.A. Read, S. M. Gowan, C. M. Incles, L. R. Kelland, S. Neidle,
Bioorg. Med. Chem. Lett. 2004, 14, 5845–5849.
CD spectroscopy: CD spectra were recorded by using a Jobin-Yvon spec-
trophotometer in a 2 mm pathlength cuvette (300 mL). Spectra were re-
corded at 258C from l=320 to 220 nm and are presented at an average
of three successive scans. Finally, all the spectra were subtracted from a
baseline corresponding to the buffer alone (100 mm potassium phosphate
buffer containing 100 mm KCl) when appropriate.
Nano ESI-MS: All mass-spectrometric measurements were performed by
using a Micromass ESI-MS Q-TOF Ultima Mass Spectrometer (Man-
chester, UK) with microchannel plate detector. Noncovalent nano ESI-
MS spectra were collected in positive-ion mode for the PNA–DNA com-
plexand the negative-ion mode for the sample containing DNA alone. A
1:1 PNA/DNA solution with a strand concentration of 1 mm, prepared in
150 mm ammonium acetate buffer (pH 7) was heated to 958C, annealed
to room temperature over 3 h, and equilibrated at 48C overnight. A
500 mm solution containing DNA was also prepared similarly. Just prior to
injection, the equilibrated DNA/PNA complexwas diluted fivefold with
isopropanol/water (1:4). The solution of DNA was similarly diluted five-
fold with methanol/water (1:4) The source parameters employed during
acquisition of the mass spectra: source temperature: 208C; capillary volt-
age: 2 and ꢂ2.23 kV for the PNA/DNA and DNA complexes, respective-
ly; cone voltage: 70 V.
[18] W. A. Denny, G. J. Atwell, B. C. Baguley, J. Med. Chem. 1983, 26,
1619–1625.
[19] C. C. Hardin, M. J. Corregan, D. V. Lieberman, B. A. Brown, Bio-
chemistry 1997, 36, 15428–15450.
[20] S. Tomac, M. Sarkar, T. Ratilainen, P. Wittung, P. E. Nielsen, B.
NordØn, A. Grꢁslund, J. Am. Chem. Soc. 1996, 118, 5544–5552.
[21] S. Ladame, J.A. Schouten, J. Stuart, J. Roldan, S. Neidle, S. Balasu-
bramanian, Org. Biomol. Chem. 2004, 2, 2925–2931.
[22] M.-P. Teulade-Fichou, C. Carrasco, L. Guittat, C. Bailly, P. Alberti,
J.-L. Mergny, A. David, J.-M. Lehn, W. D. Wilson, J. Am. Chem.
Soc. 2003, 125, 4732–4740.
[23] C.-C. Chang, J.-Y. Wu, C.-W. Chien, W.-S. Wu, H. Liu, C.-C. Kang,
L.-J. Yu, T.-C. Chang, Anal. Chem. 2003, 75, 6177–6183.
[24] C. Dohno, I. Saito, ChemBioChem 2005, 6, 1075–1081.
[25] E. Nielsen, L. Christensen, J. Am. Chem. Soc. 1996, 118, 2287–2288.
[26] H. J. Larsen, P. E. Nielsen, Nucleic Acids Res. 1996, 24, 458–463.
Acknowledgement
The authors thank the “Centre National pour la Recherche Scientifique”
(CNRS) (S.L.), the “Minist›re de la Recherche et de l’Enseignement Su-
pØrieur” (A.P.), and CSIR, Government of India, (P.S.) for financial sup-
port. Y.K. thanks DBT, Government of India for the Innovative Young
Biotechnologist Award.
[1] G. N. Parkinson in Q uadruplex Nucleic Acids; (Eds.: S. Neidle, S.
Balasubramanian), RSC Biomolecular Sciences Publishing, Cam-
bridge 2006, pp. 1–30.
[2] a) J. L. Huppert, S. Balasubramanian, S. Nucleic Acids Res. 2005,
33, 2908–2916; b) A.K. Todd, M. Johnston, S. Neidle, Nucleic Acids
Received: April 1, 2008
Published online: July 30, 2008
Chem. Eur. J. 2008, 14, 8682 – 8689
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8689