A R T I C L E S
Xiao and Weisz
structures as often observed for nonionic oligonucleotides in
aqueous solution.
ASiT and TSiA spectra (not shown). Obviously, the favored
formation of antiparallel structures as typically found in natural
systems persists with the backbone-modified nucleic acids in
aprotic solvents, indicating that neither solvation effects nor
steric or electrostatic effects of the backbone have a major
impact on this widely observed preference. Yet, parallel-stranded
duplexes are known to also form under specific conditions and
can be stabilized at neutral pH by reverse Watson-Crick A-T
base pairs.23 Alternatively, parallel-stranded structures can also
be formed through Hoogsteen A-T base pairing.24 However,
reported duplexes with A-T sequences only form under
conditions where a parallel strand orientation is enforced or
extensive mismatches in the antiparallel orientation limit
competition with a regular antiparallel structure.
Hydrogen Bond Interactions. Comparison of proton chemical
shifts in the slow exchange regime at 133 K can be used as a
sensitive measure for the relative strength of hydrogen bonds.
A more deshielded proton in the hydrogen bridge originates
from a displacement of the proton toward the acceptor atom
and is associated with a lengthening of the covalent A-H bond
with an attenuation of shielding effects by the sigma electron
pair and a shortening of the H · · ·B hydrogen bond in the
A-H· · ·B hydrogen bridge.5e,25 With an imino proton chemical
shift of 15 ppm in the Watson-Crick duplex, ASiT exhibits the
most downfield shifted proton indicating the strongest hydrogen
bond interaction among all dinucleotide duplexes. In contrast,
the T imino signal of the Watson-Crick duplex formed by the
TSiA dinucleotide is shifted upfield by 0.5 ppm compared to
the imino proton in ASiT. Interestingly, these data again suggest
the presence of a more flexible TSiA structure held together by
weaker hydrogen bonds when compared to ASiT. The TSiA imino
resonance of the additional Hoogsteen duplex resonating at 13.5
ppm is further upfield shifted indicating a stronger Watson-Crick
NH· · ·N1 hydrogen bond when compared to the corresponding
Hoogsteen NH · · ·N7 hydrogen bond. Such a relative strength
of AT Watson-Crick and Hoogsteen hydrogen bonds is also
suggested by previous scalar coupling analyses for Hoogsteen-
Watson-Crick T-A-T triplets in an intramolecular DNA
triplex26 and has also been found as a result of corresponding
low temperature NMR studies in Freon for a mixture of free
uridine and adenosine nucleosides that have shown the formation
of minor amounts of Hoogsteen base pairs in addition to
predominant Watson-Crick base pairs.25 Finally, the most
shielded imino proton with a chemical shift of 12.8 ppm
assigned to the third structure of TSiA supports its tentative
assignment to an intramolecularly folded structure with methanol-
mediated AT base pairing.
It has to be noted that, being less stabilized by stacking
interactions, TA steps in nucleic acids are generally found to
exhibit more flexibility and are more deformable compared to
more rigid AT steps.18 Obviously, the formation of additional
TSiA structures at low temperatures again points to the higher
structural variability of this dinucleotide even under nonaqueous
apolar solution conditions. Thus, a second symmetric duplex
structure with antiparallel strands and AT Hoogsteen pairing is
formed through a syn glycosidic torsion angle of the adenosine
nucleoside. The Hoogsteen dinucleotide duplex is increasingly
populated at very low temperatures and is not significantly less
populated compared to the regular Watson-Crick duplex at 113
K. Isolated Hoogsteen base pairs have frequently been observed
in aqueous solution, such as in the closing loop of DNA hairpins,
in circular oligonucleotides, in chemically modified nucleic acid
structures, or upon protein binding and intercalation by drugs
like echinomycin.19 Most noticeably, a fully Hoogsteen base-
paired antiparallel duplex DNA with adenosines in the syn
conformation has been found for a d(ATATAT) sequence by
single-crystal X-ray crystallography demonstrating its ability to
constitute an alternative secondary structure to the classical
B-DNA double helix.20 It might be argued that AT sequences
are strongly polymorphic and special packing effects may be
responsible for the Hoogsteen structures seen in the crystal.
However, an antiparallel Hoogsteen duplex in polyd(AT)
sequences was found to be stable in an aqueous solution during
molecular dynamics simulations with an energy similar to that
of the B-type duplex.21 Also, a dynamic conversion of the fully
Watson-Crick to a partly Hoogsteen base-paired structure was
reported as a result of a single-point substitution of the O4′
oxygen by CH2 at an adenosine sugar residue in a self-
complementary DNA double helix.22 Overall these data suggest
that an antiparallel Hoogsteen duplex exhibiting a smaller strand
separation is an accessible structure for an AT-alternating
sequence and might compete with normal B-DNA for natural
nucleic acids and even more so for nonionic nucleic acid
analogues with no electrostatic repulsions between the two
backbones.
For the homoassociation of the self-complementary TSiA and
ASiT dinucleotides, duplexes formed through two AT base pairs
are only possible with an antiparallel strand orientation. To also
test the potential formation of parallel-stranded structures with
the backbone-modified analogues in an aprotic environment, a
1:1 mixture of ASiT and TSiA were dissolved in the freonic
solvent and subjected to additional low temperature NMR
experiments. In case of heteroassociation, a parallel duplex
stabilized by two AT base pairs is expected to form. However,
even when the temperature is decreased to 123 K, no additional
signals of a heteroassociate are observed for the mixture and
spectra are always a mere superposition of the two individual
Interaction of Polycyclic Mutagens with AsiT in Aprotic
Solvents. Since ASiT has been shown to form a single well-
defined Watson-Crick duplex structure in a freonic solvent, it
was employed as a potential target for DNA binding ligands.
Thus, the polycyclic aromatic hydrocarbon 9-(methylaminom-
(18) (a) Packer, M. J.; Dauncey, M. P.; Hunter, C. A. J. Mol. Biol. 2000,
295, 71–83. (b) Mack, D. R.; Chiu, T. K.; Dickerson, R. E. J. Mol.
Biol. 2001, 312, 1037–1049.
(23) (a) van de Sande, J. H.; Ramsing, N. B.; Germann, M. W.; Elhorst,
W.; Kalisch, B. W.; v. Kitzing, E.; Pon, R. T.; Clegg, R. C.; Jovin,
T. M. Science 1988, 241, 551–557. (b) Otto, C.; Rippe, K.; Thomas,
K.; Ramsing, N. B.; Jovin, T. M. Biochemistry 1991, 30, 3062–3069.
(c) Parvathy, V. R.; Bhaumik, S. R.; Chary, K. V. R.; Govil, G.; Liu,
K.; Howard, F. B.; Miles, H. T. Nucleic Acids Res. 2002, 30, 1500–
1511.
(19) Mendel, D.; Dervan, P. B. Proc. Natl. Acad. Sci. U.S.A. 1987, 84,
910–914.
(20) (a) Abrescia, N. G. A.; Thompson, A.; Huynh-Dinh, T.; Subirana,
J. A. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 2806–2811. (b) Abrescia,
N. G. A.; Gonza´lez, C.; Gouyette, C.; Subirana, J. A. Biochemistry
2004, 43, 4092–4100.
(24) Raghunathan, G.; Miles, H. T.; Sasisekharan, V. Biopolymers 1994,
34, 1573–1581.
(21) Cubero, E.; Abrescia, N. G. A.; Subirana, J. A.; Luque, F. J.; Orozco,
M. J. Am. Chem. Soc. 2003, 125, 14603–14612.
(25) Dunger, A.; Limbach, H.-H.; Weisz, K. J. Am. Chem. Soc. 2000, 122,
10109–10114.
(22) Isaksson, J.; Zamaratski, E.; Maltseva, T. V.; Agback, P.; Kumar, A.;
Chattopadhyaya, J. J. Biomol. Struct. Dyn. 2001, 18, 783–806.
(26) Dingley, A. J.; Masse, J. E.; Peterson, R. D.; Barfield, M.; Feigon, J.;
Grzesiek, S. J. Am. Chem. Soc. 1999, 121, 6019–6027.
9
3868 J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010