Carbohydrate-DNA interactions at G-quadruplexes: Folding and stability changes by attaching sugars at the 5′-End

Article abstract of DOI:10.1002/chem.201203902

Quadruplex DNA structures are attracting an enormous interest in many areas of chemistry, ranging from chemical biology, supramolecular chemistry to nanoscience. We have prepared carbohydrate-DNA conjugates containing the oligonucleotide sequences of G-quadruplexes (thrombin binding aptamer (TBA) and human telomere (TEL)), measured their thermal stability and studied their structure in solution by using NMR and molecular dynamics. The solution structure of a fucose-TBA conjugate shows stacking interactions between the carbohydrate and the DNA G-tetrad in addition to hydrogen bonding and hydrophobic contacts. We have also shown that attaching carbohydrates at the 5′-end of a quadruplex telomeric sequence can alter its folding topology. These results suggest the possibility of modulating the folding of the G-quadruplex by linking carbohydrates and have clear implications in molecular recognition and the design of new G-quadruplex ligands. Copyright

Full text of DOI:10.1002/chem.201203902

FULL PAPER
DOI: 10.1002/chem.201203902
Carbohydrate–DNA Interactions at G-Quadruplexes: Folding and Stability
Changes by Attaching Sugars at the 5’-End
Irene Gꢀmez-Pinto,[a] Empar Vengut-Climent,[b] Ricardo Lucas,[b] Anna AviÇꢀ,[c]
Ramꢀn Eritja,[c] Carlos Gonzꢁlez,*[a] and Juan Carlos Morales*[b]
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Abstract: Quadruplex DNA structures
are attracting an enormous interest in
many areas of chemistry, ranging from
solution by using NMR and molecular
dynamics. The solution structure of a
fucose–TBA conjugate shows stacking
interactions between the carbohydrate
and the DNA G-tetrad in addition to
hydrogen bonding and hydrophobic
contacts. We have also shown that at-
taching carbohydrates at the 5’-end of
a quadruplex telomeric sequence can
alter its folding topology. These results
suggest the possibility of modulating
the folding of the G-quadruplex by
linking carbohydrates and have clear
implications in molecular recognition
and the design of new G-quadruplex li-
gands.
chemical
biology,
supramolecular
chemistry to nanoscience. We have pre-
pared carbohydrate–DNA conjugates
containing the oligonucleotide sequen-
ces of G-quadruplexes (thrombin bind-
ing aptamer (TBA) and human telo-
mere (TEL)), measured their thermal
stability and studied their structure in
Keywords: carbohydrates · DNA ·
G-quadruplexes · hydrogen bonds ·
molecular interactions · stacking
Introduction
We have used carbohydrate oligonucleotide conjugates
(COCs) to study carbohydrate-DNA interactions in double
helices adapting a dangling end DNA model system tradi-
tionally used to study p–p stacking interactions.[10] The sugar
moiety is attached to the 5’-end of the DNA sequence in
our COCs. In previous work, we observed that natural
highly polar carbohydrates stack onto the terminal DNA
base pair of a duplex through CH/p interactions.[11] Howev-
er, DNA stability is only increased when the sugars stack on
C–G or G–C base pairs. If COCs contained permethylated
carbohydrates, apolar versions of the natural sugars, a nota-
ble increase in double-stranded DNA stability was found in
comparison with the natural polyhydroxylated mono- and
disaccharide DNA conjugates. Moreover, these apolar car-
bohydrates are also capable of stabilizing duplexes with A–
T or T–A terminal base-pairs.[12] We hypothesized that the
CH–p pseudo hydrogen-bonds in addition to the higher hy-
drophobicity imparted by the methyl groups are responsible
for duplex stabilization.
Due to the relevance of the G-quadruplex DNA and
RNA structures we decided to explore the possible carbohy-
drate–DNA interactions in a G-quadruplex context. Herein
we describe the preparation of COCs containing the oligo-
nucleotide sequence of a G-quadruplex and sugars attached
to the 5’-end. The idea was to examine the possibility of car-
bohydrate interactions on top of the nearby G-tetrad and in-
vestigate how these contacts could affect quadruplex stabili-
ty and structure. We have prepared COCs containing the
thrombin-binding aptamer (TBA) and the human telomere
(TEL) sequences in which the carbohydrates were covalent-
ly linked through an ethylene glycol spacer (Figure 1). Sac-
charides conjugated to a G-quadruplex-forming sequence
(5’-TGGGAG) have been reported previously as anti-HIV
agents.[13] The TBA sequence (Figure 1A) was selected due
to the detailed knowledge of its 3D structure[14] and because
of the high number of chemically modified TBA sequences
(containing reverse natural bases,[15] extra natural bases,[16]
modified bases,[17] and unnatural nucleosides[18]) and TBA
conjugates (linked to small molecules[19] and to nanostruc-
tures).[20]
G-quadruplexes consist of a square arrangement of guanines
(G-tetrad), stabilized by Hoogsteen hydrogen-bonding and
by monovalent cations (especially potassium) coordinated in
the center, located between two G-tetrads. The in vivo evi-
dence of the existence of G-quadruplexes at telomeres[1] and
oncogene promoters,[2] as well as their role in controlling dif-
ferent biological processes,[3] have converted them in pri-
mary research targets for therapeutical applications.[4] Con-
sequently, a variety of ligands has been designed and pre-
pared to bind G-quadruplexes as a new class of anticancer
drugs. Some examples of quadruplex binders have been re-
ported to bind through groove specific contacts of the quad-
ruplex structure.[5] However, most quadruplex ligands descri-
bed to date use aromatic p-stacking as the main driving
force for binding on the G-tetrad platform.[6]
In addition, G-quadruplexes are a very attractive motif in
the development of structural and functional supramolecular
assemblies,[7] and in DNA-based nanodevices.[8] The possibil-
ity of modulating the physicochemical properties of a G-
quadruplex by its conjugation to small molecules is also
being explored, especially through attaching large aromatic
groups.[9]
[a] Dr. I. Gꢁmez-Pinto, Prof. Dr. C. Gonzꢂlez
Instituto de Quꢃmica Fꢃsica “Rocasolano”, CSIC
Serrano 119, 28006 Madrid (Spain)
E-mail: cgonzalez@iqfr.csic.es
[b] E. Vengut-Climent, Dr. R. Lucas, Dr. J. C. Morales
Department of Bioorganic Chemistry
Instituto de Investigaciones Quꢃmicas
CSIC–Universidad de Sevilla
Americo Vespucio 49, 41092 Sevilla (Spain)
E-mail: jcmorales@iiq.csic.es
[c] Dr. A. AviÇꢁ, Prof. Dr. R. Eritja
Instituto de Investigaciꢁn Biomꢄdica de Barcelona
IQAC, CSIC, CIBER-BBN
Networking Centre on Bioengineering
Biomaterials and Nanomedicine
Baldiri Reixac 10, 08028 Barcelona (Spain)
Supporting information for this article (including full experimental
procedures and spectroscopic characterization data) is available on
the WWW under http://dx.doi.org/10.1002/chem.201203902.
&
2
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Carbohydrate–DNA Interactions at G-Quadruplexes
FULL PAPER
(compound 2) provokes a destabilization of 4.58C with re-
spect to unmodified TBA 1. Furthermore, when the spacer
is attached to the TBA sequence (compound 3) the decrease
of Tm in comparison with TBA is 7.58C. Therefore, com-
pound 3 must be considered the proper control to evaluate
the carbohydrate contribution to the thermal stability of the
different conjugates. Thus, DTm values of the natural carbo-
hydrate TBA conjugates (4–9) with respect to the spacer–
TBA conjugate (3) indicate an increase in stability between
0.7 and 1.88C (see right column on Table 1). Thermodynam-
ic analysis of the UV melting-curves were obtained by
curve-fitting as described in the literature.[22] Free energies
confirmed the trend observed in the Tm values. Moreover, a
greater enthalpic contribution to the stability of their folded
structure is observed in all carbohydrate–TBA conjugates
with respect to the spacer-TBA conjugate (3) and to natural
TBA (1), which could be attributed to the new molecular in-
teractions between carbohydrates and the G-quadruplex.
A similar scenario is observed when apolar carbohydrates
are bound to the 5’-end of the TBA oligonucleotide. COCs
containing permethylated glucose and maltose (10 and 11)
also show a decrease in Tm values in comparison with TBA
(4.7 and 10.18C, respectively). Nevertheless, the TBA conju-
gate with the apolar version of
Figure 1. A) Sequence of the thrombin-binding aptamer (TBA). B) Sche-
matic model of the carbohydrate–TBA conjugates. C) Enlarged top-view
of monosaccharide–TBA conjugate.
Results and Discussion
The COCs were synthesized by using standard solid-phase
oligonucleotide automatic synthesis. Mono- and disacchar-
ides were covalently bound to the TBA sequence by using
the corresponding carbohydrate phosphoramidite derivative
(Figure 1 and Figure 2).[11,21] b-d-glucose, b-d-galactose and
glucose (10) is 2.88C more
stable than the control spacer-
TBA (3) contributing to moder-
ately improved TBA stability.
The addition of the natural and
apolar sugars to the conjugate 3
is able to recover, to some
extent, the TBA G-quadruplex
stability. These results indicate
that favorable carbohydrate–
DNA interactions are taking
place.
We also measured the ther-
mal stability of oligonucleotide
TBA sequences with an extra
base at the 5’-end (thymidine,
sequence 12 and adenine, se-
quence 13). A decrease in Tm
Figure 2. Carbohydrate–TBA conjugates and oligonucleotide controls prepared for this study.
value was observed for both se-
quences (6.0 and 6.38C, respec-
b-l-fucose were the selected monosaccharides, whereas b-d-
maltose, b-d-lactose and b-d-cellobiose were the chosen dis-
accharides. Also, two apolar sugars, permethylated glucose
and maltose, were attached to the TBA sequence by using a
similar synthetic strategy.[12]
The thermal stability of the carbohydrate TBA conjugates
was measured by UV-monitored thermal denaturation ex-
periments in a pH 7.0 cacodylate buffer containing KCl
(100 mm) (Table 1). All COCs containing natural carbohy-
drates (4–9) showed a lower thermal stability than the natu-
ral TBA (1). The reduction in Tm values ranges from 5.7 to
6.88C. However, it is important to note that the addition of
a phosphate group at the 5’-end of the TBA quadruplex
tively) with respect to the unmodified TBA (1). Comparable
results were obtained by Smirnov et al.;[23] these values are
similar to those obtained for COCs with the natural carbo-
hydrates and a similar behavior could be envisioned. It is
important to note that here the “spacer” of the natural aro-
matic base is deoxyribose, which is conformationally more
restricted than ethylene glycol, the spacer used for the
mono- and disaccharide TBA conjugates.
To confirm that the observed increased stability in COCs
is due to carbohydrate–G-tetrad interaction, the structure of
TBA–carbohydrate conjugates containing two mono- and
three disaccharides (compounds 5, 6, 7, 8, and 9) was stud-
ied by NMR spectroscopy, circular dichroism (CD), and gel
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

C. Gonzꢂlez, J. C. Morales et al.
Table 1. Melting temperatures Tm values (8C) for the carbohydrate–oligonucleotide conjugates containing the thrombin-binding aptamer sequence and
the corresponding oligonucleotide controls.[a]
5’-X-GGTTGGTGTGGTTGG
Tm
[8C]
DTm
DTm*
[8C][c]
ÀDS8
ÀDH8
DGꢀ37
DDGꢀ37
DDGꢀ37*
[d]
[e]
[8C][b]
[calkmolÀ1
]
[kcalmolÀ1
]
[kcalmolÀ1
]
[kcalmolÀ1
]
[kcalmolÀ1
]
A
1
2
3
4
5
6
7
8
48.2Æ0.5
43.7Æ0.2
40.7Æ0.2
42.5Æ0.2
41.4Æ0.2
42.3Æ0.2
41.7Æ0.2
41.5Æ0.3
41.8Æ0.3
43.5Æ0.7
38.1Æ0.4
42.2Æ0.1
41.9Æ0.1
–
À7.5
À3.0
–
117.5Æ2
134.1Æ2
119.8Æ1
132.3Æ1
135.1Æ2
135.7Æ2
133.7Æ2
139.7Æ2
136.9Æ2
134.1Æ1
127.2Æ2
141.5Æ3
139.2Æ2
37.7Æ0.5
42.3Æ0.5
37.5Æ0.4
41.7Æ0.4
42.3Æ0.5
42.6Æ0.5
42.0Æ0.5
43.8Æ0.7
42.9Æ0.6
42.4Æ0.5
39.7Æ0.5
44.4Æ0.8
43.8Æ0.6
À2.8
À2.3
À1.8
À2.2
À2.0
À2.2
À2.1
À2.2
À2.1
À2.4
À1.8
À2.2
À2.3
–
À1.0
À0.5
–
À
PO3
HO-C2-OPO2
b-d-glucose-C2-OPO2
b-d-galactose-C2-OPO2
b-l-fucose-C2-OPO2
b-d-maltose-C2-OPO2
b-d-lactose-C2-OPO2
b-d-cellobiose-C2-OPO2
-
À4.5
À7.5
À5.7
À6.8
À5.9
À6.5
À6.7
À6.4
À4.7
À10.1
À6.0
À6.3
0.5
1.0
0.6
0.8
0.6
0.7
0.6
0.7
0.4
1.0
0.6
0.5
À
[f]
-
À
[f]
-
+1.8
+0.7
+1.6
+1.0
+0.8
+1.1
+2.8
À1.6
+1.5
+1.2
À0.4
À0.2
À0.4
À0.3
À0.4
À0.3
À0.6
0.0
À
[f]
-
À
[f]
-
À
[f]
-
À
[f]
-
À
[f]
-
9
À
[f]
b-d-glucose(Me)-C2-OPO2 À
-
10
11
12
13
[f]
b-d-maltose(Me)-C2-OPO2
-
À
2’-deoxythymidine-OPO2À
2’-deoxyadenosine-OPO2
-
-
À0.4
À0.5
[a] Buffer: Na·cacodylate (10 mm), KCl (100 mm), pH 7.0; Tm values are the average of three experiments measured at 5 mm. [b] DTm =Tm (5’-modified
TBA sequence)ÀTm (TBA control sequence). [c] DTm*=Tm (5’-modified TBA sequence)ÀTm (5’-spacer-OPO2À-TBA sequence). [d] DDGꢀ37 =DG (5’-
modified TBA sequence)ÀDG (TBA control sequence). [e] DDGꢀ37*=DG (5’-modified TBA sequence)ÀDG (5’-spacer-OPO2À-TBA sequence). [f] C2
À
represents -CH2 CH2-.
electrophoresis. CD spectra clearly show that all conjugates
adopt antiparallel G-quadruplex structures (see Figure S1 in
the Supporting Information). Gel electrophoresis experi-
ments conducted under non-denaturating conditions show
single bands for monosaccharide–TBA conjugates, running
as the native TBA. However, two bands are observed for
the disaccharide–TBA conjugates (Figure S2 in the Support-
ing Information). Since the melting temperatures of these
conjugates do not depend on oligonucleotide concentration
(Table S1 in the Supporting Information), we conclude that
the two bands observed in the gels correspond to two mono-
meric species with different conformations.
The imino proton region of the NMR spectra clearly indi-
cates that all conjugates adopt a quadruplex structure with
two guanine tetrads (Figure S3 in the Supporting Informa-
tion). Exchangeable proton signals could be assigned on the
basis of the strong similarity between the spectra of the con-
jugates and the unmodified aptamer. Assignment of most
non-exchangeable protons could be carried out following
standard bidimensional techniques, and were confirmed with
previously reported assignment of native TBA (see Figures 3
Figure 3. Region of the NOESY spectra of the maltose–TBA conjugate
(7) superimposed onto the native TBA (1, pink) spectra. Sequential as-
signment pathways show the two different species for residues G6, T7
and G8 (labeled in different colors).
and 4 and the Supporting Information for assignment
tables).
In the case of the disaccharide conjugates, spectral assign-
ment was complicated due to the presence of additional
spin-systems. After sequential assignment, we concluded
that these systems correspond to two species, affecting pro-
tons in residues G6, T7, and G8 (see Figure 3), being both
species in slow equilibrium on the NMR timescale. The
chemical shifts of one of the conformations are almost coin-
cident with those of the native aptamer.
This effect is not observed in the monosaccharide–TBA
conjugates, in which single species are observed. The differ-
ent behavior between mono- and disaccharide conjugates is
consistent with the non-denaturating gel experiments.
Chemical shift differences between equivalent protons in
modified- and unmodified TBA are shown in Figure S4 (the
Supporting Information). Most protons in COCs exhibit
very similar chemical shifts to the natural TBA except those
in residue G1 and in the neighborhood of the TGT loop
(mainly, protons in residues G6, T7, and G8). Interestingly,
the profile of chemical shift differences is very similar in all
the conjugates.
NOE data also indicate that all conjugates maintain a sim-
ilar structure as the native TBA. Exchangeable NOE pat-
terns are consistent with formation of two G-tetrads. The
strong intensities of intraresidue NOE cross-peaks between
guanine H8 and H1’ protons confirmed the syn conforma-
&
4
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Carbohydrate–DNA Interactions at G-Quadruplexes
FULL PAPER
tion for the glycosydic angle of G1, G5, G10, and G14.
Stacking of the lateral loop thymines, T4 and T13, on top of
the adjacent guanines, G2 and G11, is corroborated by sev-
eral strong NOEs. However, some differences in NOE
cross-peaks are observed in the TGT loop suggesting that
the carbohydrate affects its structure and alters the interac-
tion of residues G8 and T9 with the nearby tetrad
(G1:G6:G10:G15; see Figure 4 and Table S2 in the Support-
ing Information).
As shown in Figure 5, the resulting structures are well-de-
fined. The solution structure of the fucose–TBA conjugate
resembles the general fold of the unmodified aptamer. The
main differences are in the TGT loop-region in which the
fucose moiety displaces the nucleobases of residues G8 and
T9 and stacks on top of G1 and G15 (see Figure 5, bottom).
In fact, fucose interaction with the guanine tetrad seems to
occur by an induced-fit mechanism. In addition to the stack-
ing interaction with the adjacent tetrad, fucose O5 is at a hy-
drogen bond distance of one amino proton of G8 (see
Figure 6). Also, a favorable hydrophobic interaction occurs
between fucose and T9 methyl groups (Figure 6). The T7
base tends to insert into the quadruplex groove. The carbo-
4
hydrate maintains the usual C1 chair conformation.
The structure of the conjugate 6 is totally consistent with
the observed NOEs and the profiles of chemical shift differ-
ences between this conjugate and the native aptamer.
Chemical shifts are very sensitive to structural variations.
Therefore, the strong similarity between DNA proton reso-
nances in the two monosaccharide–DNA conjugates implies
that fucose–TBA and galactose–TBA adopt very similar
conformations (see Figure S5 in the Supporting Informa-
tion). Although in the latter the number of carbohydrate–
DNA NOEs is lower, this is not due to a different confor-
mation of the galactose in the conjugate, but to the lack of
the fucose methyl-group that facilitates the detection of car-
bohydrate–DNA NOEs.
Inspection of the structure of fucose–TBA shows that the
site occupied by the fucose sugar cannot be easily occupied
by a disaccharide without further disruption of the TGT
loop (see Figure 5). Most probably, this is the reason of the
two species observed in the NMR spectra of the disacchar-
ide conjugates, and confirmed by electrophoretic gels. 1H
chemical shifts and NOEs suggest that in one species the
disaccharide interacts with the G-tetrad disrupting the TGT
loop, and in the other, the TGT loop adopts a native-like
conformation with the disaccharide mainly disordered
(Figure 7). Presumably, this second species has a different
electrophoretic mobility (see Figure S2 in the Supporting In-
formation). At the same time, this equilibrium must be af-
fecting the stability of the carbohydrate–TBA conjugate.
It is worth mentioning that the disaccharide–TBA conju-
gates are not more stable than the monosaccharide–TBA
conjugates, as observed in the case of sugar-capping DNA
duplexes.[11–12] The stabilization conferred to the TBA by the
different carbohydrates is more difficult to interpret than in
the case of carbohydrates stacking at the terminal base of
double helices. In the latter, the increase in thermal stability
is a direct measurement of a favorable interaction of the car-
bohydrate with the terminal base pair. However, in TBA
conjugates the carbohydrate G-tetrad interaction competes
with native nucleobase p-stacking interactions. This event is
difficult to circumvent because G-tetrads are rarely exposed
to the solvent. In fact, the capping interactions in the termi-
nal G-tetrads affect dramatically the folding topology of the
quadruplex.
Figure 4. Regions of the NOESY experiments of fucose–TBA conjugate
(6). Contour plots of the carbohydrate–DNA NOEs are shown in red.
Most interestingly, several carbohydrate–DNA NOEs
could be detected in all conjugates. The number of these
NOEs is particularly high in the case of conjugate 6
(fucose–TBA), most probably due to the presence of a
methyl group in this carbohydrate; this resonance is easily
identified and does not overlap with other signals of the
conjugate (see Figure 4, top).
The solution structure of conjugate 6 was determined on
the basis of experimental NOE constraints. Restrained mo-
lecular-dynamics calculations were carried out with the
AMBER package as explained in the Methods section in
the Supporting Information. A total of 142 distance con-
straints (10 between carbohydrate and DNA) were used in
the calculation (Table S2, the Supporting Information).
Structures were deposited in the Protein Data Bank (2lyg).
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

C. Gonzꢂlez, J. C. Morales et al.
tween terminal G-tetrads and
nucleotides in the loops. In
many cases, terminal residues
are also involved in these inter-
actions, as observed in human
telomeric sequences. As a proof
of concept, we have explored
the effect of substituting the 5’-
terminal thymine of the oligo-
nucleotide
5’-TAGGGT-
TAGGGT-3’ (TEL) with a car-
bohydrate. TEL folds as a G-
quadruplex containing two re-
peats of the human telomeric
sequence, and its structure has
been studied by NMR spectro-
scopy and X-ray crystallogra-
phy. In the crystallographic
structure TEL forms a dimeric
parallel propeller-like quadru-
plex.[24] However, in solution,
TEL exists in an equilibrium
between dimeric parallel and
antiparallel structures, being
the latter predominant in Na+
buffer.[25]
As it can be observed in
Figure 8, the NMR spectra of
Figure 5. A) Stereoview of the superposition of 10 refined structures of the fucose–TBA, conjugate 6. The
average structure is shown in bold bonds. B) Top view of the fucose–TBA conjugate 6. C) Top view of the
native TBA G-quadruplex 1 (arrows indicate the direction where G8 and T9 move in the fucose–TBA conju-
TEL
and
b-d-glucose-C2-
À
OPO2 -AGGGTTAGGGT 14
À
À
À
À
gate 6). The fucose moiety is shown in red, guanine core in blue, TT lateral loops in cyan, and the TGT
loop in green.
are dramatically different. As
Figure 7. Side view of the two possible conformations of maltose–TBA
conjugate 7: Left: maltose stacked on the G-tetrad with T9 shifted; right:
maltose pointing to the bulk aqueous solution with T9 and G8 stacked
on the G-tetrad.
Figure 6. Detail on the structure of the fucose–TBA conjugate 6. Con-
tacts of fucose with the bases of G8 and T9 in the TGT loop of the G-
quadruplex can be observed (methyl–methyl interactions and a hydrogen
bond shown in yellow).
previously reported, the TEL spectra exhibit many broad
imino signals, indicating the presence of multiple conforma-
tions. A similar complexity is observed in the control spacer-
OPO2 -AGGGTTAGGGT conjugate 15 (see Figure S9 in
the Supporting Information).
Interfering with capping interactions by conjugating car-
bohydrates makes it possible to modulate G-quadruplex
folding, since the particular topology adopted by a G-rich
oligonucleotide depends strongly on capping interactions be-
À
&
6
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Carbohydrate–DNA Interactions at G-Quadruplexes
FULL PAPER
engage the possibility of developing carbohydrate-based li-
gands that could bind through CH–p noncovalent forces to-
gether with other molecular interactions.
Finally, we have also shown that by conjugating carbohy-
drates at the 5’-end of G-quadruplex structures, we can alter
their folding topology. The success of the modulation is
based on manipulating the interaction between the terminal
G-tetrads and other residues. The intrinsic versatility of car-
bohydrates introduces the possibility of modulating G-quad-
ruplex systems in nanomaterials and biomedical applica-
tions.
Acknowledgements
We thank the MICINN (grants CTQ2009–13705, CTQ2010–20451,
CTQ2010–21567-C02–02), EU COST project MP0802, Generalitat de
Catalunya (2009/SGR/208), and Instituto de Salud Carlos III (CIBER-
BNN, CB06 01 0019) for financial support. IGP and RL thank CSIC for
a JAE contract. EVC thanks the Ministry of Education for an FPU doc-
toral fellowship.
Figure 8. Imino region of the NMR spectra of A) 14 and B) TEL and
schemes of the two species coexistent in solution; C) H1’-aromatic region
of the NOESY spectra of 14 in D2O (tm =300 ms) (Experimental condi-
tions: oligo concentration=1 mm, potassium phosphate buffer (10 mm),
pH 7, T=58C).
[1] K. Paeschke, T. Simonsson, J. Postberg, D. Rhodes, H. J. Lipps, Nat
Struct Mol Biol 2005, 12, 847–854.
[2] a) T. A. Brooks, L. H. Hurley, Nat Rev Cancer 2009, 9, 849–861;
b) D. Sun, K. Guo, Y.-J. Shin, Nucleic Acids Res. 2011, 39, 1256–
1265.
[3] a) S. Balasubramanian, L. H. Hurley, S. Neidle, Nat Rev Drug
Discov 2011, 10, 261–275; b) H. J. Lipps, D. Rhodes, Trends Cell
Biol. 2009, 19, 414–422.
[4] a) G. W. Collie, G. N. Parkinson, Chem. Lett. Chem. Soc Rev 2011,
40, 5867–5892; b) Y. Qin, L. H. Hurley, Biochimie 2008, 90, 1149–
1171.
[5] S. Cosconati, L. Marinelli, R. Trotta, A. Virno, S. De Tito, R. Ro-
magnoli, B. Pagano, V. Limongelli, C. Giancola, P. G. Baraldi, L.
Mayol, E. Novellino, A. Randazzo, J Am Chem. Soc 2010, 132,
6425–6433.
[6] a) S. M. Haider, S. Neidle, G. N. Parkinson, Biochimie 2011, 93,
1239–1251; b) S. Neidle, FEBS J. 2010, 277, 1118–1125.
[7] M. Nikan, J. C. Sherman, Angew. Chem. Int. Ed. 2008, 47, 4900–
4902.
However, the imino region of the NMR spectra of the car-
bohydrate–TEL conjugate 14 shows six sharp signals, indi-
cating clearly the presence of a single conformation. The
two-dimensional spectra of 14 shown in Figure 8 indicates
that all guanines are in an anti-conformation. These data are
consistent with the formation of a parallel symmetric quad-
ruplex. A complete structural determination of this and
other COCs containing telomeric sequences is in progress in
our laboratories and will be reported in due time. With the
present data we can already conclude that substitution of
the 5’-terminal thymine by glucose alters the equilibrium be-
tween different species, promoting the formation of a stable
parallel quadruplex (see Figure 8).
[8] a) H.-M. So, K. Won, Y. H. Kim, B.-K. Kim, B. H. Ryu, P. S. Na, H.
Kim, J.-O. Lee, J Am Chem. Soc 2005, 127, 11906–11907; b) C.-L.
Hsu, H.-T. Chang, C.-T. Chen, S.-C. Wei, Y.-C. Shiang, C.-C. Huang,
Chem. Eur. J. 2011, 17, 10994–11000.
[9] a) F. D. Lewis, Y. Wu, L. Zhang, Chem. Commun. 2004, 636–637;
b) J. Jayawickramarajah, D. M. Tagore, L. K. Tsou, A. D. Hamilton,
Angew. Chem. Int. Ed. 2007, 46, 7583–7586.
Conclusion
[10] a) M. Senior, R. A. Jones, K. J. Breslauer, Biochemistry 1988, 27,
3879–3885; b) K. M. Guckian, B. A. Schweitzer, R. X. F. Ren, C. J.
Sheils, D. C. Tahmassebi, E. T. Kool, J. Am. Chem. Soc. 2000, 122,
2213–2222; c) Z. Dogan, R. Paulini, J. A. Rojas Stutz, S. Narayanan,
C. Richert, J Am Chem. Soc 2004, 126, 4762–4763.
[11] R. Lucas, I. Gꢁmez-Pinto, A. AviÇꢁ, J. J. Reina, R. Eritja, C. Gon-
zꢂlez, J. C. Morales, J Am Chem. Soc 2011, 133, 1909–1916.
[12] R. Lucas, E. Vengut-Climent, I. Gꢁmez-Pinto, A. AviÇꢁ, R. Eritja,
C. Gonzꢂlez, J. C. Morales, Chem. Commun. 2012, 48, 2991–2993.
[13] J. D’Onofrio, L. Petraccone, L. Martino, G. Di Fabio, A. Iadonisi, J.
Balzarini, C. Giancola, D. Montesarchio, Bioconjugate Chem. 2008,
19, 607–616.
Our studies on sugar–DNA conjugates show that favorable
carbohydrate–DNA interactions occur in a G-quadruplex
structural context as we have observed in the case of TBA.
Carbohydrates stack on top of guanine tetrads, and also in-
teract with loop DNA bases through hydrogen bonding or
hydrophobic contacts when they are available (like the in-
teraction between fucose and thymine methyl groups in the
fucose–TBA conjugate).
Our results have implications in the design of new G-
quadruplex ligands. Until now only aromatic compounds
binding through p–p stacking on the guanine tetrad or
through van der Waals contacts with the grooves of the G-
quadruplex structures have been reported. Our results
[14] a) R. F. Macaya, P. Schultze, F. W. Smith, J. A. Roe, J. Feigon, Proc
Natl Acad Sci USA 1993, 90, 3745–3749; b) K. Y. Wang, S. McCur-
dy, R. G. Shea, S. Swaminathan, P. H. Bolton, Biochemistry 1993, 32,
1899–1904; c) K. Padmanabhan, K. P. Padmanabhan, J. D. Ferrara,
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

C. Gonzꢂlez, J. C. Morales et al.
J. E. Sadler, A. Tulinsky, J Biol Chem. 1993, 268, 17651–17654;
d) J. A. Kelly, J. Feigon, T. O. Yeates, J Mol Biol 1996, 256, 417–422.
[15] L. Martino, A. Virno, A. Randazzo, A. Virgilio, V. Esposito, C.
Giancola, M. Bucci, G. Cirino, L. Mayol, Nucleic Acids Res. 2006,
34, 6653–6662.
Nucleic Acids Res. 2011, 39, 1155–1164; e) H. Saneyoshi, S. Mazzini,
A. AviÇꢁ, G. Portella, C. Gonzalez, M. Orozco, V. E. Marquez, R.
Eritja, Nucleic Acids Res. 2009, 37, 5589–5601.
[19] a) A. AviÇꢁ, S. Mazzini, R. Ferreira, R. Eritja, Bioorg. Med. Chem.
2010, 18, 7348–7356; b) S. Nagatoishi, T. Nojima, B. Juskowiak, S.
Takenaka, Angew. Chem. Int. Ed. 2005, 44, 5067–5070.
[20] a) Y. Wang, F. Yang, X. Yang, Biosens. Bioelectron. 2010, 25, 1994–
1998; b) H. de Puig, S. Federici, S. H. Baxamusa, P. Bergese, K.
Hamad-Schifferli, Small 2011; c) Y. C. Shiang, C. L. Hsu, C. C.
Huang, H. T. Chang, Angew. Chem. Int. Ed. 2011, 50, 7660–7665.
[21] J. C. Morales, J. J. Reina, I. Dꢃaz, A. AviÇꢁ, P. M. Nieto, R. Eritja,
Chem. Eur. J. 2008, 14, 7828–7835.
[16] A. De Rache, I. Kejnovska, M. Vorlickova, C. Buess-Herman,
Chem. Eur. J. 2012, 18, 4392–4400.
[17] a) S. R. Nallagatla, B. Heuberger, A. Haque, C. Switzer, J Comb
Chem. 2009, 11, 364–369; b) J. Gros, A. AviÇꢁ, J. Lopez de La Osa,
C. Gonzalez, L. Lacroix, A. Perez, M. Orozco, R. Eritja, J. L.
Mergny, Chem. Commun. 2008, 2926–2928; c) S. Ogasawara, M.
Maeda, Angew. Chem. Int. Ed. 2009, 48, 6671–6674.
[18] a) T. Coppola, M. Varra, G. Oliviero, A. Galeone, G. D’Isa, L.
Mayol, E. Morelli, M. R. Bucci, V. Vellecco, G. Cirino, N. Borbone,
Bioorg. Med. Chem. 2008, 16, 8244–8253; b) Y. Kasahara, S. Kita-
dume, K. Morihiro, M. Kuwahara, H. Ozaki, H. Sawai, T. Imanishi,
S. Obika, Bioorg. Med. Chem. Lett. 2010, 20, 1626–1629; c) C. G.
Peng, M. J. Damha, Nucleic Acids Res. 2007, 35, 4977–4988; d) A.
Pasternak, F. J. Hernandez, L. M. Rasmussen, B. Vester, J. Wengel,
[22] J. D. Puglisi, I. Tinoco, Jr., Methods Enzymol 1989, 180, 304–325.
[23] I. Smirnov, R. H. Shafer, Biochemistry 2000, 39, 1462–1468.
[24] G. N. Parkinson, M. P. Lee, S. Neidle, Nature 2002, 417, 876–880.
[25] A. T. Phan, D. J. Patel, J Am Chem. Soc 2003, 125, 15021–15027.
Received: October 31, 2012
Published online: && &&, 2012
&
8
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Carbohydrate–DNA Interactions at G-Quadruplexes
FULL PAPER
G-Quadruplexes
I. Gꢀmez-Pinto, E. Vengut-Climent,
R. Lucas, A. AviÇꢀ, R. Eritja,
C. Gonzꢁlez,* J. C. Morales* . &&& — &&&
Carbohydrate–DNA Interactions at G-
Quadruplexes: Folding and Stability
Changes by Attaching Sugars at the 5’-
End
Sweetening G-quadruplexes: Carbohy-
drate–DNA contacts are observed in a
G-quadruplex context. Mono- and dis-
accharides attached to the 5’-end of
the thrombin-binding aptamer (TBA)
are capable of stabilizing the conjugate
with respect to the spacer-TBA con-
trol. A detailed structure elucidation
of the fucose–TBA conjugate shows
sugar stacking on the guanine tetrad
together with methyl–methyl van der
Waals interactions and a possible
hydrogen bond (see figure).
G-Quadruplexes
Molecular interactions between sugars and G-quadruplex
DNA structures are observed in different carbohydrate G-
quadruplex conjugates. Mono- and disaccharides are cova-
lently linked to the 5ꢅ-end of oligonucleotide sequences of
G-quadruplexes (thrombin binding aptamer-TBA and
human telomere-TEL) and their thermal stability and
structure in solution have been studied using UV, NMR
and molecular dynamics. Saccharides and the DNA bases
from the TGT loop of the TBA quadruplex actually
compete for interacting with the guanine tetrad. CH/p
stacking, hydrogen bonding, and hydrophobic contacts
play a role in the interaction between carbohydrates and
the DNA G-tetrad. For more details, see the Full Paper by
C. Gonzalez and J. C. Morales et al. on page &&ff.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!