Selective Pairing of Polyfluorinated DNA Bases

Article abstract of DOI:10.1021/ja039571s

Novel selective non-hydrogen-bonding DNA base pairs utilizing fluorinated nucleoside analogues have been investigated. Melting studies of DNA duplexes containing 2,3,4,5-tetrafluorobenzene and 4,5,6,7-tetrafluoroindole bases on opposite strands show great

Full text of DOI:10.1021/ja039571s

Published on Web 02/21/2004
Selective Pairing of Polyfluorinated DNA Bases
Jacob S. Lai and Eric T. Kool*
Department of Chemistry, Stanford UniVersity, Stanford, California 94305-5080
Received November 13, 2003; E-mail: kool@stanford.edu
Recent studies of polyfluorinated organic compounds have
revealed useful selective interactions between such “fluorous”
species relative to their interactions with water or the parent
hydrocarbons.1 Although the origins of this selective interaction
are not fully understood, it has been established that perfluorinated
hydrocarbons are significantly more hydrophobic than the analogous
hydrocarbons.2 The shielding of strongly hydrophobic surfaces from
solvent may explain selective polyfluorinated side-chain pairing,
recently employed successfully in peptide-peptide interactions,3
and in enhanced stability of proteins with fluorinated amino acids.4
Fluorocarbon interactions have been gaining widespread utility and
interest, especially in aiding separations of catalysts, reagents,
substrates, and products.1
We sought to establish whether such fluorocarbon selectivity
could be harnessed in pairing of DNA bases, as an alternative to
other known modes of pairing. A number of laboratories have
investigated molecular strategies for selective base pairing, using
approaches other than that of nature itself, namely Watson-Crick
hydrogen bonding. Such work has yielded fundamental information
on DNA biophysics and biology, and applied utility in expansion
of the genetic alphabet.5 Benner proposed over a decade ago that
hydrogen-bonding schemes beyond the standard Watson-Crick
arrangement could be used in selective pairing.6 In more recent
years, strategies that avoid hydrogen bonding altogether have been
introduced, employing selective pairing of hydrocarbon “bases” with
one another.7 Such selectivity can arise from the avoidance of the
energetic cost of desolvation of polar natural bases and from the
advantageous burying of hydrophobic surface area.
Figure 1. Structures of fluorinated and hydrocarbon DNA base replace-
ments. (a) Chemical structures of the four nucleosides in the study. (b)
Electrostatic surface potentials of bases with methyl groups at the point of
attachment to deoxyribose. Calculated with Spartan ‘02 (Wavefunction, Inc.)
using the AM1 Hamiltonian.
For comparison we also prepared the non-fluorinated hydrocarbon
analogues, phenyl (B)11 and indole (I)12 glycoside derivatives
(Figure 1).
To test pairing preferences of the four unnatural base replace-
ments, we placed them in short oligonucleotides and paired these
strands with complementary partners containing either natural bases
or unnatural analogues at single or double positions. Stabilities of
the duplexes were evaluated by thermal denaturation monitored by
UV absorbance, in a pH 7.0 buffer containing 1.0 M NaCl, 10 mM
Na‚phosphate, and 0.1 mM EDTA. Melting temperatures were
determined from the inflection points in the curves, and free energies
were obtained by van’t Hoff plots of the data at multiple DNA
concentrations.
Although nucleic acid base analogues with fluorine substituents
have been reported recently,8-10 there is as yet no report on whether
highly fluorinated DNA bases might display a selective interaction
with one another, analogous to the pairing of polyfluorinated
peptides. A recent report described pentafluorobenzene as a DNA
base replacement,9 but no selective pairing was observed in
oligonucleotides. Subsequent studies have revealed, however, that
pentafluorobenzene is strongly destabilizing to helical DNA because
of an unfavorable effect of two ortho fluorine substituents.10 This
suggested that selective pairing might still be possible if other highly
fluorinated DNA base analogues were to be examined, as long as
this bis-ortho effect were avoided.
Examination of DNA models suggested a number of polyflu-
orinated DNA base replacements as possible candidates for selective
fluorous pairing. The modeling suggested that 2,3,4,5-tetrafluo-
robenzene (abbreviated FB) might pair opposite itself without
distorting the helical geometry. Moreover, this compound has
recently been shown to stack quite strongly at the ends of DNA
helices.10 A second candidate was the previously unknown 4,5,6,7-
tetrafluoroindole (FI), which was readily synthesized as an N-
nucleoside species (see Supporting Information (SI)). These two
deoxyribosides were prepared as their 5′-trityl, 3′-phosphoramidite
derivatives for incorporation into DNA by automated synthesizer.
Initial experiments pairing the two fluorinated nucleosides
opposite natural DNA bases in a 12-bp duplex confirmed they pair
with low stability opposite the hydrophilic nucleobases (see SI).
However, when paired opposite themselves, a significant degree
of stability was regained for both compounds. This confirms that
the pairing of the polyfluorinated bases operates selectively in the
context of natural DNA, thus displaying significant levels of
orthogonality. The FI-FI pair in this context is nearly as stable as
the natural T-A pair. The mild-to-moderate destabilization of the
duplex by these pairs (as compared to natural base pairs) is
consistent with several previous nonpolar DNA base-pair analogues
and is most likely due to the energetic cost of desolvation.13
We then began more detailed studies with a new sequence,
increasing the nonpolar pair content to 2/12 (17%) to emphasize
differences among various combinations of the nonnatural bases.
The data confirm that all the self-complementary sequences give
the expected concentration dependence, confirming two-stranded
duplexes (as opposed to self-folded hairpins (see SI)). Overall, the
results show (Table 1, Figure 2) that the fluorinated bases pair
selectively with each other, as compared to the hydrocarbon-
hydrocarbon pairing or the mixed fluorocarbon-hydrocarbon
pairing. For example, the tetrafluorobenzene-tetrafluoroindole pair
(FB-FI) is more stable than the similar benzene-indole pair lacking
9
3040
J. AM. CHEM. SOC. 2004, 126, 3040-3041
10.1021/ja039571s CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
0.10); FI (1.66 ( 0.10). The results confirm what has been
previously reported for polyfluorinated saturated hydrocarbons:
namely, that they are more hydrophobic than their hydrocarbon
variants.14 For the present compounds, the order of hydrophobicity
is FI > FB > I > B. This correlates well with the stabilities of
their self-pairs as well as with their relative stacking abilities.10
Taken together, the data suggest that this selective pairing may
be due to solvent avoidance of these specially hydrophobic
structures on formation of a duplex relative to the more exposed
single strands. Placing them in pairs opposite one another buries
large fractions of the flat π surfaces and significant parts of the
edges facing one another as well (see Figure S2, SI). Thus, the
basic physical origins of the selective interaction appear to be similar
to those seen recently in selective fluorinated peptide interactions.3
Our findings suggest that polyfluoroaromatic base pairing might
be employed as a new, selective approach to pairing in DNA that
is orthogonal to that of the natural genetic system. Future structural
studies could shed light on the orientations of the base analogues
in DNA. Also of interest is whether such fluorocarbon pairing
selectivity could exert significant effects in the enzymatic replication
of DNA.
Figure 2. Histogram of base-pair stabilities as measured for double
substitution of the pair into a 12-bp duplex (see Table 1).
Table 1. Thermodynamic Data for Duplexes Containing Fluorous
and Hydrocarbon Basesa
c
d
e
base pairb
Tm
∆G°37
∆∆G°37
(X•Y)
(°C)
(kcal/mol)
(kcal/mol)
B•B
FB•FB
I•I
29.8
34.6
31.5
45.2
27.8
41.8
58.1
20.3
-6.7 ( 0.2
-7.3 ( 0.1
-7.2 ( 0.2
-8.8 ( 0.1
-6.7 ( 0.1
-8.2 ( 0.1
-11.9 ( 0.3
-5.7 ( 0.2
-1.0 ( 0.4
-1.6 ( 0.3
-1.5 ( 0.4
-3.1 ( 0.3
-1.0 ( 0.3
-2.5 ( 0.3
-6.2 ( 0.5
-
Acknowledgment. This work was supported by the U.S.
National Institutes of Health (GM52956/EB002059).
FI•FI
B•I
FB•FI
T•A
T•C
Supporting Information Available: Details of synthesis and
thermodynamics methods (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
a Conditions: 1 M NaCl, 10 mM phosphage (pH 7.0) with 0.1 mM
EDTA. b Sequence is 5′- CGGXAGCTYCCG (self-complementary). c Tm
values are at 5.0 µM. d Averages of values from van’t Hoff and curve fitting
methods. e Values resolve to the least stable duplex (the T-C mismatch).
References
(1) (a) Curran, D. P. Pure Appl. Chem. 2000, 72, 1649-1653. (b) Luo, Z.
Y.; Zhang, Q. S.; Oderaotoshi, Y.; Curran, D. P. Science 2001, 291, 1766-
1769. (c) Yoder, N. C.; Kumar, K. Chem. Soc. ReV. 2002, 31, 335-341.
(d) Wende, M.; Gladysz, J. A. J. Am. Chem. Soc. 2003, 125, 5861-5872.
(e) Horvath, I. T. Acc. Chem. Res. 1998, 31, 641-650.
(2) Barthel-Rosa, L.; Gladysz, J. A. Coord. Chem. ReV. 1999, 190-192, 587.
(3) Bilgicer, B.; Xing, X. C.; Kumar, K. J. Am. Chem. Soc. 2001, 123, 11815-
11816.
(4) Tang, Y.; Ghirlanda, G.; Vaidehi, N.; Kua, J.; Mainz, D. T.; Goddard,
W. A.; DeGrado, W. F.; Tirrell, D. A. Biochemistry 2001, 40, 2790-
2796.
(5) Kool, E. T. Acc. Chem. Res. 2002, 35, 936-943.
(6) Piccirilli, J. A.; Krauch, T.; Moroney, S. E.; Benner, S. A. Nature 1990,
343, 33-37.
(7) (a) Ogawa, A. K.; Wu, Y.; McMinn, D. L.; Liu, J.; Schultz, P. G.;
Romesberg, F. E. J. Am. Chem. Soc. 2000, 122, 3274-3287. (b) Mitsui,
T.; Kitamura, A.; Kimoto, M.; To, T.; Sato, A.; Hirao, I.; Yokoyama, S.
J. Am. Chem. Soc. 2003, 125, 5298-5307. (c) Schweitzer, B. A.; Kool,
E. T. J. Am. Chem. Soc. 1995, 117, 737-738.
(8) (a) Schweitzer, B. A.; Kool, E. T. J. Org. Chem. 1994, 59, 7238-7242.
(b) Parsch, J.; Engels, J. W. HelV. Chim. Acta 2000, 83, 1791. (c) Shibata,
N.; Das, B. K.; Honjo, H.; Takeuchi, Y. J. Chem. Soc., Perkin Trans. 1
2001, 14, 1605-1611.
(9) Mathis, G.; Hunziker, J. Angew. Chem., Int. Ed. 2002, 41, 3203-3205.
(10) Lai, J. S.; Qu, J.; Kool, E. T. Angew. Chem., Int. Ed. 2003, 42, 5973-
5977.
fluorine; the difference is a significant 14 °C in Tm and 1.5 kcal/
mol in free energy. The mixed versions of these pairs, placing
hydrocarbon opposite fluorocarbon, resulted in pairing stabilities
falling between those of the fully fluorinated and nonfluorinated
pairs. The two other cases also confirm the selective pairing
effect: the FB-FB pair is more stable than the all-hydrocarbon B-B
pair, and FI-FI is more stable than I-I. The fluorinated FI-FI
pair is the most stable of the series, while the hydrocarbon B-B
pair is the least stable. The difference between these extremes is
15.4 °C (2.1 kcal/mol), illustrating the significant degree by which
structure and polyfluorination can affect base-pair stability.
To seek evidence for the origins of this selectivity, we examined
individual properties of the four nonpolar nucleoside analogues.
Stacking was evaluated by the standard dangling end approach using
a 6-bp self-complementary DNA. Results showed (Table S3, SI)
that the two polyfluorinated analogues do, in fact, stack more
strongly than the two parent hydrocarbons. FB stacks 1.0 kcal/mol
(11) Wichai, U.; Woski, S. A. Org. Lett. 1999, 1, 1173-1175.
(12) Girgis, N. S.; Cottam, H. B.; Robins, R. K. J. Heterocycl. Chem. 1988,
25, 361-366.
F
(13) (a) Kool, E. T.; Morales, J. C.; Guckian, K. M. Angew. Chem., Int. Ed.
2000, 39, 990-1009. (b) Sun, Z. H.; Chen, D. L.; Lan, T.; McLaughlin,
L. W. Biopolymers 2002, 65, 211-217.
more favorably than B, and I, 1.1 kcal/mol more favorably than
I.
We explicitly examined hydrophobicity of the four nucleosides
by partitioning between 1-octanol and water. The data are as
(14) Guckian, K. M.; Schweitzer, B. A.; Ren, R. X.-F.; Sheils, C. J.;
Tahmassebi, D. C.; Kool, E. T. J. Am. Chem. Soc. 2000, 122, 2213-2222.
F
follows: B (log P ) 0.77 ( 0.10); B (1.39 ( 0.10); I (0.99 (
JA039571S
9
J. AM. CHEM. SOC. VOL. 126, NO. 10, 2004 3041