Local disruption of DNA-base stacking by bulky base surrogates

Article abstract of DOI:10.1039/b110842e

A novel biphenyl base surrogate disrupts 2-aminopurine base stacking while maintaining duplex integrity.

Full text of DOI:10.1039/b110842e

Local disruption of DNA-base stacking by bulky base surrogates
Ishwar Singh,a Walburga Hecker,b Ashok K. Prasad,a Virinder S. Parmara and Oliver Seitz*b
a Department of Chemistry, University of Delhi, Delhi 110007, India
b Max-Planck-Institut für Molekulare Physiologie, Otto-Hahn-Str. 11, D-44227 Dortmund, Germanyand
Institut für Organische Chemie, Universität Dortmund, D-44227 Dortmund, Germany.
E-mail: oliver.seitz@mpi-dortmund.mpg.de; Tel: 0049 231 1332424; Fax: 0049 231 1332499
Received (in Cambridge, UK) 28th November 2001, Accepted 21st January 2002
First published as an Advance Article on the web 8th February 2002
A novel biphenyl base surrogate disrupts 2-aminopurine
base stacking while maintaining duplex integrity.
Cellular DNA is the subject of a multitude of chemical and
enzymatic modification reactions. In order to gain access to
their target structure, many DNA-processing enzymes require a
local disruption of base stacking. An extreme form of base
stacking disruption is the so-called ‘base-flipping’ whereby a
nucleobase is rotated out of the DNA-helix and positioned in an
extrahelical conformation.1 This targeted disruption of base
stacking has been observed for DNA-methyltransferases, DNA-
glycosylases, endonucleases and is likely to occur in many other
cases.2 We became interested in mimicking the enzyme-
mediated ‘base-unstacking’ and started a research program
devoted to the conformational preorganisation of locally base-
unstacked DNA, an endeavour ultimately aiming for the
construction of enhanced binders of DNA-modifying en-
zymes.
It has been shown that the removal of one base of a Watson–
Crick base pair can promote an unstacking of the remaining
partner base.3 However, these abasic sites confer a dramatic
Scheme 1 Synthesis of phosphoamidites containing bulky base surrogates.
duplex destabilisation.4 We hypothesised that a bulky base
CNE = b-cyanoethyl, DMTr = 4,4A-dimethoxytrityl, Tol = toluoyl.
surrogate, incorporated in a DNA duplex should allow for a
local unstacking of the opposing base without significant
detriment to the global duplex stability. In order to test this
hypothesis we introduced aromatic base analogues that lacked
First, it was explored whether the C-glycosides 3a–e were
tolerated in a duplex. The melting curves showed sigmoidal
any hydrogen-bonding potential and were therefore devoid of
behaviour which suggests that the binding of the modified
maintaining Watson–Crick base pairing.5
Recently, the pyrenyl nucleotide 3a and the naphthyl
nucleotides 3b and 3c were synthesised and used in the
oligonucleotides 4a–e still followed a co-operative base-paring
mechanism. Duplexes that contained the artificial base surro-
gates b–e showed similar melting temperatures (TM
=
synthesis of modified oligodeoxynucleotides (ODNs).5 For the
introduction of tetrahedral structural elements we chose to
incorporate the acenaphthyl- and the biphenyl-nucleotides 3d
and 3e. The synthesis of the known phosphoamidites 3a–c and
the new building blocks 3d and 3e required the formation of a
C-glycosidic linkage and was accomplished according to a
published protocol (Scheme 1).5a In the event of the C-
glycosylation of 1, it was essential to convert the aryl-Grignard
reagents into less basic cadmium organic species in order to
avoid the formation of glycals and furans. The removal of the
toluoyl-protecting groups from 2a–e and the subsequent
regioselective introduction of the trityl-protecting group was
followed by the reaction with a phosphitylating agent, which
furnished the phosphoramidites 3a–e in satisfying yields. The
building blocks 3a–e were used in the automated synthesis of
oligonucleotides 4a–e (Table 1). The couplings of the modified
phosphoramidites proceeded with 98–99% yield. HPLC- and
MALDI-TOF/MS analysis confirmed the identity and purity of
the modified ODNs 5a–e.
For an examination of the influence of the base surrogates in
ODNs 4a–e hybridisation studies were performed. To this end,
modified oligonucleotides 4a–e were hybridised with the ODN
5 which contained the base analogue 2-aminopurine (2AP).6
2AP forms a base-pair with thymidine and is virtually non-
fluorescent when stacked within a DNA duplex but fluoresces
upon disruption of base stacking.7 In the duplexes 4·5 2AP is
placed opposite to the base surrogates and serves as a probe that
can help in the analysis of stacking interactions.
58.0–59.1 °C). This even holds true for the biphenyl system in
4e, which in contrast to the other modifications is by no means
an entity that would be expected to stack. The pyrene residue
has the largest hydrophobic surface area, and therefore yielded
duplexes of high stability (TM
= 63.5 °C). A marked
destabilisation was observed when the 2AP probe was paired
with the mismatched G in 4g (DTM = 27.1 °C) or an abasic site
Table 1 Thermal denaturation and fluorescence data of 4·5-duplexes
X =
TMa (DTM b)/ °C
Relative fluorescencec
T, 4T
63.2
63.5 (0.3)
0.09 (1.0)
0.11 (1.2)
0.15 (1.6)
0.20 (2.0)
0.13 (1.4)
0.54 (5.8)
0.42 (4.6)
0.40 (4.3)
1-pyrenyl, 4a
1-naphthyl, 4b
2-naphthyl, 4c
acenaphthyl, 4d
4-biphenyl, 4e
abasic site, 4f
G, 4g
58.0 (25.2)
58.2 (25.0)
58.5 (24.7)
59.1 (24.1)
51.9 (211.3)
56.1 (27.1)
a Measured as denaturation curves at 1 mM DNA concentration in a buffered
solution (100 mM NaCl, 10 mM NaH2PO4, pH7). b Relative to duplex 4T·5.
c Ratio between fluorescence of duplex 4·5 and the fluorescence of single
stranded 5 (measured at 369 nm, conditions as specified in Fig. 1). The
increase of the fluorescence of duplex 4·5 relative to the duplex 4T·5 is
given in brackets.
500
CHEM. COMMUN., 2002, 500–501
This journal is © The Royal Society of Chemistry 2002

in 4f (DTM = 211.3 °C), which is known to support an
unstacking of the opposing base.3
pairing for all cases. It was surprising to note that the
incorporation of the bulky biphenyl residue led to a small but
significant increase of duplex stability when compared with
duplexes 4b·5 and 4c·5 which contained intercalatable naphthyl
rings. Recently, the incorporation of self-pairing or metal-
coordinating bipyridyl residues was described.9 It has to be
noted that the biphenyl modification reported in here differs
from the bipyridyl system since it spans two tilted planes rather
than one.
The fluorescence of ODN 5 was determined before and after
hybridisation with the complementary strands 4T and 4a–g. Fig.
1A shows the steady-state fluorescence of oligonucleotide 5.
Upon hybridisation with the complementary strand 4T a 2AP–T
base-pair is formed. Expectedly, base stacking within the
interior of duplex 4T·5 decreased the 2AP-fluorescence by a
factor of 10.6. The fluorescence of duplexes 4a–e·5 was
measured, in order to determine whether one of the base
surrogates was able to increase the 2AP-fluorescence by base-
unstacking (Fig. 1A). The emission profiles were characteristic
for the 2AP fluorophore, with the exception of the pyrene
containing duplex 4a·5 which was dominated by the pyrene
fluorophore. The strongest fluorescence enhancement was
observed with the biphenyl-ODN 4e. The 2AP-emission was
enhanced by a factor of 5.8 when compared to the A–T-
containing duplex 4T·5. Remarkably, the biphenyl-induced
fluorescence enhancement exceeded the 4.6 fold and 4.3 fold
increases that were observed when 2AP was paired against an
abasic site or the mismatched base G (Fig. 1B). This suggests
that the biphenyl residue is more efficient in promoting a local
base stacking disruption than abasic residues or mismatched
bases despite the fact that the latter modifications led to a larger
global destabilisation as judged by the thermal stability of the
duplexes (DTM = 24.1 °C for 4e·5 and DTM = 211.3 °C and
27.1 °C for 4f·5 and 4g·5).
The biphenyl residue in duplex 4e·5 proved to be most
efficient in enhancing the fluorescence of an opposing 2AP-
probe. Measurements with duplexes in which the 2AP-base was
replaced by adenine (data not shown) excluded any hybrid-
isation-induced biphenyl fluorescence. It is therefore justified to
conclude that the increased fluorescence reflects 2AP-base
unstacking. The studies outlined above represent the first
example in which a local unstacking of the 2AP-base has been
attempted by introducing non-natural base surrogates. In-
tuitively, a disruption of base stacking would be expected to
correlate with duplex destabilisation. For example, it was shown
that abasic sites destabilise stacking interactions and confer a
dramatic duplex destabilisation.3,4 However, the disruption of
2AP-base stacking was more effective in the biphenyl duplex
4e·5 than in abasic site and G containing duplexes 4f·5 and 4g·5.
Since the latter displayed lower duplex stabilities, it has to be
concluded that a disruption of local base stacking can be
achieved without destabilising the duplex significantly. When
assessing the nature of the unstacked state it has to be noted that
unstacked 2AP can adopt innerhelical and extrahelical con-
formations. The different susceptibilities to quenching reagents
suggest that a control of the conformational equilibrium
between innerhelical and extrahelical states could be possible.
For example, the biphenyl base appears to shift the equilibrium
in favour of innerhelically unstacked 2AP when compared with
duplexes that contain an abasic site or the mismatched G-
base.
It can be expected that a conformationally controlled
disruption of local base stacking will render the target base more
reactive towards enzymic modification reactions. We envision
that the design of thermally stable duplex molecules with
locally unstacked bases could allow for the construction of new
biomolecular tools for interference with DNA-modifying
enzymes. Studies towards the inhibition of DNA methyl-
transferases are in progress.
Fig. 1 Fluorescence emission spectra (calibrated on fluorescence of 5) of A:
5, 4e·5, 4c·5, 4b·5, 4d·5, 4a·5 and 4T·5 (order by fluorescence intensity) and
of B: 4e·5, 4f·5, 4g·5 and 4T·5. Excitation: 305 nm. C: Acrylamide
quenching studies (Stern–Volmer plot) of 2AP in duplexes 4e·5, 4f·5, 4g·5
and 4T·5. Buffer contained 15 mM MgCl2.
This work was supported by the DFG, DAAD and DST. O.S.
is grateful for a Liebig- and DFG-fellowship.
Notes and references
1 S. Klimasauskas, S. Kumar, R. J. Roberts and X. Cheng, Cell, 1994, 76,
357.
2 R. J. Roberts and X. D. Cheng, Annu. Rev. Biochem., 1998, 67.
3 (a) J. T. Stivers, Nucleic Acids Res., 1998, 26, 3837; (b) P. Cuniasse, G.
V. Fazakerley, W. Guschlbauer, B. E. Kaplan and L. C. Sowers, J. Mol.
Biol., 1990, 213, 303.
4 C. A. Gelfand, G. E. Plum, A. P. Grollman, F. Johnson and K. J.
Breslauer, Biochemistry, 1998, 37, 7321.
5 (a) R. X. F. Ren, N. C. Chaudhuri, P. L. Paris, S. Rumney and E. T. Kool,
J. Am. Chem. Soc., 1996, 118, 7671; (b) T. J. Matray and E. T. Kool, J.
Am. Chem. Soc., 1998, 120, 6191; (c) Pyrenyl-LNA was recently
reported; (d) Base-pairs lacking hydrogen binding were reported: A. K.
Ogawa, Y. Q. Wu, M. Berger, P. G. Schultz and F. E. Romesberg, J. Am.
Chem. Soc., 2000, 122, 8803.
6 D. C. Ward, E. Reich and L. Stryer, J. Biol. Chem., 1969, 244, 1228.
7 (a) C. R. Guest, R. A. Hochstrasser, L. C. Sowers and D. P. Millar,
Biochemistry, 1991, 30, 3271; (b) B. W. Allan and N. O. Reich,
Biochemistry, 1996, 35, 14757; (c) B. Holz, S. Klimasauskas, S. Serva
and E. Weinhold, Nucleic Acids Res., 1998, 26, 1076.
8 E. L. Rachofsky, E. Seibert, J. T. Stivers, R. Osman and J. B. A. Ross,
Biochemistry, 2001, 40, 957.
In a minimum model the increase of the 2AP fluorescence
can arise from two modes of base unstacking.8 The 2AP-base
can be unstacked but still reside in the interior of a locally
distorted double helix. Alternatively, the 2AP-base could be
pushed into an extrahelical conformation. Extrahelical 2AP but
not innerhelical unstacked 2AP, would be susceptible to
quenching by external reagents. In order to probe the environ-
ment of the 2AP-base, fluorescence quenching studies were
performed. Fig. 1C shows Stern–Volmer plots in which the ratio
of the fluorescence intensities in the absence (F0) and the
presence (F) of added quencher is plotted against the concentra-
tion of the quencher acrylamide. The slopes are proportional to
the rate constant kQ of bimolecular quenching and are hence a
measure of the accessibility of the 2AP to the quencher.3a It
became apparent that the biphenyl duplex 4e·5 is less sensitive
to external quenching than the other duplexes [kQ(4f·5)/kQ(4e·5)
= 2.8]. This suggests that in 4e·5 the conformational equilib-
rium is in favour of innerhelically unstacked 2AP when
compared with duplexes that contain abasic sites or the
mismatched G.
The bulky base surrogates in 4a–e share no similarities with
canonical nucleobases, neither by shape nor by polarity.
Nevertheless, the melting curves indicated co-operative base
9 (a) H. Weizman and Y. Tor, J. Am. Chem. Soc., 2001, 123, 3375; (b) C.
Brotschi, A. Häberli and C. J. Leumann, Angew. Chem., Int. Ed., 2001,
40, 3012.
CHEM. COMMUN., 2002, 500–501
501