Measurement of local DNA reorganization on the picosecond and nanosecond time scales

Article abstract of DOI:10.1021/ja992456q

Picosecond and nanosecond dynamics in the interior of DNA are observed for the first time as a dynamic Stokes shift in the fluorescence of a specially designed base-pair analogue. A recently synthesized coumarin C-riboside has been incorporated into an ol

Full text of DOI:10.1021/ja992456q

11644
J. Am. Chem. Soc. 1999, 121, 11644-11649
Measurement of Local DNA Reorganization on the Picosecond and
Nanosecond Time Scales
Eric B. Brauns,† Mihaela L. Madaras,‡ Robert S. Coleman,*,‡ Catherine J. Murphy,*,† and
Mark A. Berg*,†
Contribution from the Department of Chemistry and Biochemistry, UniVersity of South Carolina,
Columbia, South Carolina 29208, and the Department of Chemistry, The Ohio State UniVersity,
100 West 18th AVenue, Columbus, Ohio 43210
ReceiVed July 12, 1999. ReVised Manuscript ReceiVed October 1, 1999
Abstract: Picosecond and nanosecond dynamics in the interior of DNA are observed for the first time as a
dynamic Stokes shift in the fluorescence of a specially designed base-pair analogue. A recently synthesized
coumarin C-riboside has been incorporated into an oligonucleotide in place of a normal purine‚pyrimidine
base pair. The dynamic Stokes shift of this probe occurs with components near 300 ps and 13 ns. These times
are too long for solvent relaxation or simple vibrational motion of the DNA framework. They imply that the
interior of DNA is a unique dynamic environment unlike either a fluid or a molecular crystal.
Introduction
of chemical reactions occurring within the DNA, because they
determine how quickly the DNA can adapt to the product’s new
geometry.15-18 We have directly time-resolved the dynamic
fluctuations within the interior of DNA using ultrafast spec-
troscopy in concert with a specifically designed photophysical
probe. The results show that DNA provides a unique dynamical
environment that is distinctly different from either a simple-
fluid or a rigid-crystalline environment.
Although there have been many observations of DNA motion
over longer lengths or longer times,19,20 experiments measuring
motions at the level of individual bases and in the picosecond
time range have been difficult. The inherent time scales of
nuclear and electron magnetic resonance limit their ability to
quantify motion on the picosecond time scale.20-24 Nonetheless,
modeling of indirect effects often indicates the presence of
significant localized, internal motion within DNA in the
picosecond time range.22,24-30 Fluorescence anisotropy decays
The detail and precision of DNA crystal structures can divert
attention away from the importance of thermal fluctuations
around these structures. Recent computer simulations have raised
the possibility that there are relatively large fluctuations away
from the average structure with lifetimes in the picosecond
range.1-5 The existence of such fluctuations has profound
implications for a variety of biologically important processes.
The sequence dependence of static deviations in DNA structure
has been implicated in gene recognition and DNA/protein
interactions,6-8 and a similar role has been hypothesized for
sequence-dependent variations in flexibility and relaxation
rates.9,10 The possibility of rapid electron transfer through DNA
is being intensely debated.11-14 Strong fluctuations in the DNA
structure would dramatically affect electron-transfer rates, both
by modulating the interbase coupling and by providing a
reorganization coordinate capable of localizing charge. The
dynamics of structural fluctuations are also critical to the rates
(15) Berne, B. J.; Borkovec, M.; Straub, J. E. J. Phys. Chem. 1988, 92,
3711-3725.
(16) Hynes, J. T. Theory of Chemical Reactions; CRC Press: Boca Raton,
FL, 1985; Vol. 4.
(17) Rossky, P. J.; Simon, J. D. Nature 1994, 370, 263-269.
(18) Hanggi, P.; Talkner, P.; Borkovec, M. ReV. Mod. Phys. 1990, 62,
251-341.
(19) McCammon, J. A.; Harvey, S. C. Dynamics of Proteins and Nucleic
Acids; Cambridge University Press: Cambridge, 1987.
(20) Robinson, B. H.; Drobny, G. P. Methods Enzymol. 1995, 261, 451-
509.
* Author to whom correspondence should be addressed.
† University of South Carolina.
‡ The Ohio State University.
(1) Beveridge, D. L.; Ravishanker, G. Curr. Opin. Struct. Biol. 1994, 4,
246-255.
(2) Young, M. A.; Ravishanker, G.; Beveridge, D. L. Biophys. J. 1997,
73, 2313-2336.
(3) Cheatham, T. E.; Brooks, B. R. Theor. Chem. Acc. 1998, 99, 279-
288.
(4) Auffinger, P.; Westhof, E. Curr. Opin. Struct. Biol. 1998, 8, 227-
(21) Lane, A. N. Methods Enzymol. 1995, 261, 413-435.
(22) Gorenstein, D. G. Chem. ReV. 1994, 94, 1315-1338.
(23) Palmer, A. G.; Williams, J.; McDermott, A. J. Phys. Chem. 1996,
100, 13293-13310.
236.
(5) Leontis, N. B.; Santa Lucia, J. Molecular Modeling of Nucleic Acids;
American Chemical Society: San Francisco, 1998; Vol. 1.
(6) Parvin, J. D.; McCormick, R. J.; Sharp, P. A.; Fisher, D. E. Nature
1995, 373, 724.
(7) Carver, T. E.; Millar, D. P. Biochemistry 1998, 37, 1898-1904.
(8) Crothers, D. M. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 15163-
15165.
(9) Flatters, D.; Lavery, R. Biophys. J. 1998, 75, 372-381.
(10) de Souza, O. N.; Ornstein, R. L. Biopolymers 1998, 46, 403-415.
(11) Murphy, C. J.; Arkin, M. R.; Jenkins, Y.; Chatlia, N. D.; Bossman,
S. H.; Turro, N. J.; Barton, J. K. Science 1993, 262, 1025-1029.
(12) Ratner, M. Nature 1999, 397, 480-481.
(13) Lewis, F. D.; Wu, T.; Zhang, Y.; Letsinger, R. L.; Greenfield, S.
R.; Wasielewski, M. R. Science 1997, 277, 673-676.
(14) Kelley, S. O.; Barton, J. K. Science 1999, 283, 375-381.
(24) Robinson, B. H.; Mailer, C.; Drobny, G. Annu. ReV. Biophys. Biomol.
Struct. 1997, 26, 629-658.
(25) Kintanar, A.; Huang, W. C.; Schindele, D. C.; Wemmer, D. E.;
Drobny, G. Biochemistry 1989, 28, 282-293.
(26) Nuutero, S.; Fujimoto, B. S.; Flynn, P. F.; Reid, B. R.; Ribeiro, N.
S.; Schurr, J. M. Biopolymers 1994, 34, 463-480.
(27) Fujimoto, B. S.; Willie, S. T.; Reid, B. R.; Schurr, J. M. J. Magn.
Reson. B 1995, 106, 64-67.
(28) Alam, T. M.; Drobny, G. Biochemistry 1990, 29, 3421-3430.
(29) Borer, P. N.; Laplante, S. R.; Kumar, A.; Zanatta, N.; Martin, A.;
Hakkinen, A.; Levy, G. C. Biochemistry 1994, 33, 2441-2450.
(30) Michalczyk, R.; Silks, L. A.; Russu, I. M. Magn. Reson. Chem.
1996, 34, S97-S104.
10.1021/ja992456q CCC: $18.00 © 1999 American Chemical Society
Published on Web 12/04/1999

Measurement of Local DNA Reorganization
J. Am. Chem. Soc., Vol. 121, No. 50, 1999 11645
of intercalated fluorophores have picosecond time resolution,
but are dominated by overall tumbling or collective bending
and twisting motions.31,32 Nonetheless, small components in
these decays have been attributed to local motion.33 Anisotropy
decay of intrinsic DNA-base fluorescence has been attributed
to localized motions on the picosecond time scale,34 but the
interpretation is obscured by the possibility of excited-state
mixing and the complex photophysics of the bases.35 Vibrational
spectroscopy provides evidence for changing or multiple
conformations within DNA, but gives no direct information on
the associated time scales.36-40
Figure 1. Structures of the coumarin C riboside 1 and abasic-site
analogue 2 that form a base-pair analogue with good photophysical
properties.
We introduce time-resolved Stokes shifts (TRSS) as a method
of directly observing the local dynamics of DNA. The TRSS
technique has been used extensively to measure the dynamics
of simple liquids by using fluorophores free in solution.41-43
When a fluorophore is excited, its properties, e.g., polarity or
polarizability, change significantly, causing the local solvent
structure to reorganize. This reorganization reduces the energy
of the system and causes the fluorescence to shift to longer
wavelengths, i.e., a Stokes shift develops. The rate of shifting
of the fluorescence spectrum directly reflects the rate of
reorganization of the fluorophore’s environment. By the fluctua-
tion-dissipation theorem,44 this rate also gives the lifetime of
thermal fluctuations in the unperturbed system.
conditions, the intrinsic properties of the DNA would determine
its reorganization time.
By introducing the TRSS experiment, we have tested that
hypothesis. In a room temperature, aqueous solution, the DNA’s
dynamics are several orders-of-magnitude slower than the
solvent’s. The reorganization is multiexponential, showing that
at least two reorganization processes are involved. Moreover,
a large fraction of the dynamics is still unresolved with the
current 100 ps time resolution. However, straightforward
extensions of the current techniques can explore those faster
processes as well.
This method has not been applied to DNA primarily because
of the poor photophysical properties of either native DNA or
currently available forms of modified DNA. To address this
problem, we synthetically replaced a normal DNA purine‚
pyrimidine base pair with a fluorophore specifically designed
to optimize TRSS measurements. The DNA becomes the
fluorophore’s “solvent”, and the TRSS measures the dynamic
motion of the DNA. Because the TRSS is caused by intermo-
lecular interactions between the fluorophore and nearby groups
in its environment, the TRSS is insensitive to either global
motion of the oligomer or long-range twisting and bending
motions. Instead, it results almost entirely from movement of
groups proximal to the fluorophore.
Results
Incorporating a Photophysical Probe into DNA. Two of
us have recently reported the synthesis of the coumarin
C-riboside 1 (Figure 1), a fluorescent molecule designed
specifically for the present studies.46 Several considerations went
into the design of this molecule. It incorporates the well-
characterized Coumarin 102 molecule as its fluorophore. Work
on solution-phase TRSS has demonstrated that Coumarin 102
and related dyes are nearly ideal probes.41-43,47,48 Coumarins
have simple photophysics uncomplicated by competing pro-
cesses: they have no low-lying electronic states that interact
with their first excited states48 (although see ref 49) and
nonradiative relaxation and intersystem crossing are weak,
resulting in near unity quantum yields.50
The amine group of 1 is rigidified, preventing intramolecular
twisting or pyramidalization in the excited state, as occurs in
other coumarins.50 Extensive measurements by Seidel, Schulz,
and Sauer show that Coumarin 102 is unlikely to either donate
or accept electrons from DNA bases,51 processes which would
quench fluorescence. The absorption maximum of 1 is near 400
nm, a wavelength long enough to prevent energy transfer to or
electronic mixing with the transitions of the normal DNA bases.
In earlier work, we used a high viscosity, cryogenic solvent
to slow the DNA dynamics, so they could be measured by
steady-state techniques.45 We showed that DNA does reorganize
through a slow, diffusive process. However, under those
conditions, the DNA’s reorganization rate matched the solvent’s
reorganization rate. We hypothesized that the solvent was
exerting a rate-limiting effect, and that under physiological
(31) Schurr, J.; Fujimoto, B. S.; Wu, P.; Song, L. Topics in Fluorescence
Spectroscopy, Volume 3: Biochemical Applications; Lakowicz, J. R., Ed.;
Plenum Press: New York and London, 1992; pp 137-222.
(32) Millar, D. P. Curr. Opin. Struct. Biol. 1996, 8, 637-642.
(33) Magde, D.; Zappala, M.; Knox, W. H.; Nordlund, T. M. J. Phys.
Chem. 1983, 87, 3286-3288.
(34) Georghiou, S.; Bradrick, T. D.; Philippetis, A.; Beecham, J. M.
Biophys. J. 1996, 70, 1909-1922.
(35) Georghiou, S.; Nordlund, T. M.; Saim, A. M. Photochem. Photobiol.
1985, 41, 209-212.
Molecular modeling of compound 1 in duplex B-DNA (Figure
2, details in the Experimental Section) shows that when it is
positioned opposite the abasic-site analogue 2 in helical B-DNA,
the coumarin portion effectively replaces a normal purine‚
pyrimidine base pair. There is no significant distortion of the
(36) Thomas, G. J.; Tsuboi, M. AdV. Biophys. Chem. 1993, 3, 1-70.
(37) Peticolas, W. L. Methods Enzymol. 1995, 246, 389.
(38) Brahms, S.; Fritsch, V.; Brahms, J. G.; Westhof, E. J. Mol. Biol.
1992, 223, 455-476.
(46) Coleman, R. S.; Madaras, M. L. J. Org. Chem. 1998, 63, 5700-
5703.
(47) Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M. J. Phys.
Chem. 1995, 99, 17311-17337.
(48) Lewis, J. E.; Maroncelli, M. Chem. Phys. Lett. 1998, 282, 197-
203.
(39) Rudisser, S.; Hallbrucker, A.; Mayer, E. J. Am. Chem. Soc. 1997,
119, 12251-12256.
(40) Terpstra, P. A.; Otto, C.; Greve, J. Biopolymers 1997, 41, 751-
763.
(41) Barbara, P. F.; Jarzeba, W. AdV. Chem. Phys. 1990, 15, 1.
(42) Maroncelli, M. J. Mol. Liq. 1993, 57, 1-37.
(43) Stratt, R. M.; Maroncelli, M. J. Phys. Chem. 1996, 100, 12981.
(44) Chandler, D. Introduction to Modern Statistical Mechanics; Oxford
University Press: New York, 1987.
(45) Brauns, E. B.; Murphy, C. J.; Berg, M. A. J. Am. Chem. Soc. 1998,
120, 2449-2456.
(49) Kovalenko, S. A.; Ruthmann, J.; Ernsting, N. P. Chem. Phys. Lett.
1997, 271, 40-50.
(50) Jones, G.; Jackson, W. R.; Choi, C.; Bergmark, W. R. J. Phys. Chem.
1985, 89, 294-300.
(51) Seidel, C. A. M.; Schulz, A.; Sauer, M. H. M. J. Phys. Chem. 1996,
100, 5541-5553.

11646 J. Am. Chem. Soc., Vol. 121, No. 50, 1999
Brauns et al.
Scheme 1
Scheme 2
formed using a palladium-promoted Heck coupling between
glycal 3 and coumarin triflate 4 (retrosynthetic Scheme 1). This
arylation reaction sequence occurred with complete control of
stereochemistry.53
Incorporation of this fluorophore into an oligonucleotide
proceeded by protection of the 5′-hydroxyl group as the
dimethoxytrityl (DMT) ether followed by phosphitylation of
the 3′-hydroxyl group to afford phosphoramidite 5 (Scheme 2).
This reagent was incorporated directly into a 17 base-pair
DNA oligomer using standard automated oligonucleotide
synthesis protocols. For the photophysical experiments described
below, we prepared the modified T7-RNA-polymerase-
promoter sequence 5′-d(TAATACGA1CCACTATA)-3′ as a
“typical” B-DNA sequence. The complementary sequence 5′-
d(TATAGTGG2TCGTATTA)-3′ was prepared with abasic
model 2 in the position complementary to coumarin 1.52
The two sequences anneal to form a normal DNA duplex.
The circular dichroism spectrum of the oligomer indicates a
B-form double helix. Melting studies show a decrease in stability
compared to the native duplex containing a G‚C base pair. In
pH 7.2 phosphate buffer (100 mM) containing 100 mM NaCl,
the native duplex (1fC, 2fG) has a Tm of 58.5 °C, whereas
the modified duplex containing 1‚2 melts at Tm ) 45.5 °C. This
destabilization is typical for a non-native base-pair substitution.54
The coumarin remains strongly fluorescent in the double
helix. In fact, the fluorescence lifetime is increased relative to
aqueous solution (7.4 vs 4.2 ns). Because the coumarin in this
sequence is surrounded by one of each of the normal DNA
bases, we infer that electron transfer or other quenching
processes will not be a problem for this fluorophore in other
sequences as well.
Figure 2. Molecular modeling structure of the modified 17-mer duplex
showing the synthetic chromophore occupying the site of a base pair.
The yellow fluorophore 1 is covalently attached to the blue DNA strand.
The purple DNA strand has the abasic site 2 complementary to the
fluorophore.
double helix, as measured by C1′ to C1′ distances.52 Coumarin
102 itself has a low solubility in water relative to organic
solvents, so hydrophobic forces will also favor the coumarin
unit occupying the DNA interior as opposed to extending into
the aqueous medium.
Coumarins have a large increase in dipole moment in their
excited states, and as a result, the energy of their excited states
is primarily sensitive to the local electric field. Thus as a TRSS
probe, compound 1 will respond specifically to movement of
charged groups in the dye’s local environment.
Spectroscopic measurements (Figure 3) support the molecular
modeling prediction that 1 adopts a position in the interior of
the DNA. The excitation spectrum of 1 (Figure 3) is red-shifted
in the duplex (c) relative to the spectrum free in aqueous buffer
(a), and the spectrum in the melted oligonucleotide is intermedi-
ate between the two (b). These results are consistent with the
probe being shielded from the solvent in the duplex and partially
exposed in the single-stranded oligomer. Melting experiments
performed with detection in the coumarin absorbance band at
In the synthesis of fluorophore 1, a Coumarin-102 unit is
covalently attached to a 2-deoxyribose via a â-C-glycosidic bond
(53) Packer, M. J.; Hunter, C. A. J. Mol. Biol. 1998, 280, 407-420.
(54) Moran, S.; Ren, X. F.; Rumney, S.; Kool, E. T. J. Am. Chem. Soc.
1997, 119, 2056-2057.
(52) Eritja, R.; Walker, P. A.; Randall, S. K.; Goodman, M. F.; Kaplan,
B. E. Nucleosides Nucleotides 1987, 6, 803-814.

Measurement of Local DNA Reorganization
J. Am. Chem. Soc., Vol. 121, No. 50, 1999 11647
Figure 3. Steady-state fluorescence excitation (left) and emission
(right) spectra of 1 (a) free in solution, (b) in the melted, single-stranded
oligonucleotide (90 °C), and (c) in the double-stranded oligonucleotide
(25 °C).
380 nm show that the coumarin spectral changes occur in con-
cert with the melting of the DNA, as monitored by the absorp-
tion at 260 nm. Additionally, the Stokes shift of 1 in the duplex
is smaller than that in solution (Figure 3), indicating that its
environment cannot reorganize as effectively as in water. Again,
the melted oligomer has an intermediate Stokes shift, consistent
with partial exposure to the solvent. With an appropriate
fluorophore positioned in the interior of the DNA duplex, TRSS
measurements of the DNA dynamics become possible.
Time-Resolved Measurements of DNA Dynamics. In earlier
studies, we measured steady-state Stokes shifts of the interca-
lated fluorophore acridine orange.45 In the current work, we have
not only replaced the structure-perturbing intercalator acridine
orange with the structure-preserving fluorophore 1, we have also
turned to fluorescence measurements with picosecond time
resolution.
Figure 4. Wavelength resolved fluorescence decays of the modified
oligomer: (a) full decays collected at 460 nm (upper points) and 550
nm (lower points) (the decays have been normalized to match at long
time) and (b) subtraction of these two decays (data (points) and
instrument response function (solid line)). The difference is caused by
a dynamic Stokes shift induced by DNA motion.
Fluorescence decay measurements of 1 in the oligomer were
made at 10 wavelengths, but the qualitative features are evident
by examining just two decays. Figure 4 shows decays on the
blue (460 nm) and red (550 nm) sides of the fluorescence
spectrum. The long exponential decay common to both wave-
lengths corresponds to the fluorescence lifetime (7.4 ns). At
short times, the long wavelength measurement shows a rise in
intensity, whereas the short wavelength measurement shows an
additional fast decay. These effects are caused by the fluores-
cence spectrum shifting from short to long wavelengths.
The difference between the blue and red decays is shown in
Figure 4. The TRSS contribution to the signal is much larger
than the noise and has components much longer than the
instrument response function. The presence of at least two
relaxation times is noticeable.
Figure 5. Representative time-resolved fluorescence spectra of the
modified oligomer. Left to right on the short wavelength side: 100 ps,
300 ps, 3.0 ns, and 30 ns. Reconstructed data (points) and log-normal
fits (curves). Spectra are normalized to constant area.
A more quantitative analysis comes from using the full set
of wavelength-dependent decays to reconstruct the time-
dependent fluorescence spectra (Figure 5).55 The spectra show
a continuous shift with time, as expected for a continuous
evolution of structure. In contrast, a transfer between two
discrete conformers or substates would have given an isosbestic
point in the spectra.
The time evolution of the spectral peak gives the standard
solvation response function for the DNA (Figure 6).41-43 The
function is well fit by a biexponential with time constants of
300 ps and 13 ns. (The fit function for the peak fluorescence
frequency is 19856 cm-1 + 312 cm-1[0.47 exp(-t/300 ps) +
0.53 exp(-t/13.4 ns)].) The dynamics clearly occur on two
distinct time scales and at least two distinct mechanisms must
be involved.
Figure 6. Solvation response function for the interior of DNA derived
from the relaxation of the peak of the time-dependent fluorescence
spectrum (Figure 5). The curve is a biexponential fit with time constants
of 300 ps and 13 ns.
We note that there are likely to be other fast relaxation
processes in addition to those observed here. We only observe
a Stokes shift of 312 cm-1 within our time range, whereas the
(55) Maroncelli, M.; Fleming, G. R. J. Chem. Phys. 1987, 86, 6221-
6239.

11648 J. Am. Chem. Soc., Vol. 121, No. 50, 1999
Brauns et al.
total Stokes shift is expected to be roughly 1500 cm-1. (The
Stokes shift in cyclohexane is a bound on the t ) 0 Stokes
shift. In cyclohexane, the emission and excitation spectra cross
at 65% of their maxima.56 Our estimate assumes the same
crossing point occurs in our samples at t ) 0.) Presumably, the
missing Stokes shift occurs faster than we can resolve with our
current instrumentation. This hypothesis is certainly consistent
with molecular dynamics simulations, which see the majority
of the relaxation occurring before 10 ps.2,57,58
lifetimes extending into the 100’s of picoseconds range.1-4,64
Thus, it is very plausible that our 300 ps time scale is connected
with relaxation of the DNA structure. Because of limitations
on the length of computer trajectories, it is not known if these
fluctuations also extend out to the 13 ns time seen in our
experiment.
On the other hand, the TRSS may be sensing motion in the
perturbed environment outside the DNA proper. For example,
the counterion atmosphere of the DNA may affect the electric
field in the interior of the DNA. The δ-relaxation seen in
dielectric measurements of DNA solutions is attributed to
diffusion of the counterions along the DNA chain.65 It occurs
on the tens of nanoseconds time scale and could be related to
the long time seen in our TRSS measurements.
The water next to and in the grooves in the DNA is strongly
perturbed, up to the point of forming a crystallographically
structured “spine of hydration”.66 The constrained motion of
this water67 or exchange of this water with ions68 may also
contribute to the TRSS.
These initial results establish that the interior of DNA
experiences an unusual dynamic environment. They also raise
numerous questions about the origin of the dynamics, their
dependence on sequence and external conditions, and the
possibility of even faster processes. The techniques introduced
here are easily extended to systematically modified DNA’s and
to shorter times.59 These future studies will provide an extensive
and detailed exploration of the unique dynamics of DNA.
Discussion
The relaxation times seen in the TRSS experiments are too
slow for many of the most obvious processes. They are too slow
to represent even low-frequency vibrational motion of the DNA
or quasiharmonic oscillation of the fluorophore within a static
DNA structure. The relaxation of water itself is subpicosecond,59
so we are not directly detecting bulk solvent motion. Times
this slow must result from complex, cooperative reorganization
of the DNA structure and its immediate surroundings. These
motions are more similar to the diffusive reorganization of a
liquid than to the vibrational motions of a crystal.
A similar situation occurs with proteins. Proteins have
crystallographically well-defined average structures, but suffer
dynamic fluctuations about this average.60-62 These fluctuations
are often described in terms of diffusion on a rugged potential
energy landscape,63 and related ideas may be useful in describing
DNA dynamics.
In previous work, we found that in a high-viscosity, cryogenic
medium, the average DNA relaxation rate exactly equals the
solvent relaxation rate.45 At that time, we hypothesized that
movement of portions of the DNA also involves displacement
of solvent. In a high-viscosity solvent, the solvent reorganization
becomes rate limiting. In the current work, the aqueous solvent
relaxes very quickly, in under 1 ps.59 Under these conditions,
the DNA relaxation rates are much slower than the solvent
relaxation rates. These results support the conclusion that the
DNA relaxation requires significant motion of the external
solvent, but also show that there are significant internal
constraints on the motion as well.
A more specific assignment of the DNA motions being
observed is still somewhat speculative. The coumarin probe we
used is specifically sensitive to motions of charges. A variety
of charged groups exist in the DNA proper. The hydrogen bonds
in the base pairs have strong dipole and quadrupole moments.
These moments produce strong, but short ranged electric fields.
Thus distortions of the base pairs are likely to contribute to the
observed dynamics, but the contribution from the pairs neigh-
boring the fluorophore will dominate over more distant base
pairs. The phosphate groups on the DNA backbone have full
charges, which produce long-range monopole fields. Motion of
the phosphates is also likely to contribute to the TRSS.
Molecular dynamics (MD) simulations indicate that there are
a variety of relatively large fluctuations in DNA structure with
Experimental Section
Coumarin Dimethoxytrityl Ether. A solution of 1 (100 mg, 0.28
mmol), coevaporated with pyridine (2 × 3 mL), dissolved in dry
pyridine (4 mL) under N2 at 25 °C was treated with bis(4-methoxy-
phenyl)phenylmethyl chloride (114 mg, 0.34 mmol, 1.2 equiv). After
3 h at 25 °C, the reaction mixture was diluted with saturated aqueous
NaHCO3 (10 mL) and extracted with CH2Cl2 (3 × 10 mL). The
combined organic extracts were dried (Na2SO4) and concentrated, and
the residue was purified by flash chromatography (2.5 × 15 cm Et3N
deactivated silica, Et2O) to afford the corresponding DMT ether (164
mg, 89%) as a yellow foam: 1H NMR (300 MHz, CDCl3) δ 7.25 (m,
9H, ArH), 6.91 (s, 1H, ArH), 6.83 (m, 4H, ArH), 6.26 (s, 1H, ArH),
5.34 (dd, J ) 6.4, 9.5 Hz, 1H, C1′-H), 4.33 (m, 1H, C3′-H), 4.11 (m,
1H, C4′-H), 3.71 (s, 6H, 2 OCH3), 3.35 (m, 1H, C5′-H), 3.15 (m, 5H,
2 CH2 + C5′-H), 2.85 (m, 2H, CH2), 2.71 (m, 2H, CH2), 2.40 (m, 1H,
C2′-H), 1.85 (m, 5H, 2 CH2 + C2′-H); 13C NMR (75 MHz, CDCl3) δ
162.8, 158.5, 156.3, 151.2, 145.6, 144.7, 135.7, 130.0, 129.9, 128.1,
127.8, 126.8, 120.9, 117.9, 113.1, 106.9, 106.2, 104.0, 103.4, 86.3,
85.9, 75.2, 64.3, 58.5, 55.2, 49.8, 49.4, 21.5, 20.6, 20.4; FABMS, m/z
659 (M+), 303, 118.
Coumarin Phosphoramidite 5. A solution of the above ether (100
mg, 0.15 mmol) in anhydrous CH2Cl2 (3 mL) under N2 at 25 °C was
treated sequentially with tetrazole (6 mg, 75 µmol, 0.5 equiv), dry
diisopropylamine (11 µL, 75 µmol, 0.5 equiv), and (2-cyanoethyl)-
N,N,N′,N′-tetraisopropyl phosphorodiamidite (55 mg, 0.18 mmol, 1.2
equiv). After 3 h at 25 °C, the reaction mixture was diluted with CH2-
Cl2 (10 mL) and dried (Na2SO4). The solvent was removed in vacuo
and the residue was purified by flash chromatography (2 × 12 cm Et3N
deactivated silica, 5% CH3OH/CH2Cl2) to afford 5 (85 mg, 71%) as a
yellow foam: 1H NMR (300 MHz, CDCl3) δ 7.30 (m, 9H, ArH), 6.9
(56) Fee, R. S.; Maroncelli, M. Chem. Phys. 1994, 183, 235-247.
(57) Winger, R. H.; Liedl, K. R.; Ru¨disser, S.; Pichler, A.; Hallbrucker,
A.; Mayer, E. J. Phys. Chem. 1998, 102, 8934-8940.
(58) Genest, D. Biopolymers 1995, 38, 389-399.
(59) Jimenez, R.; Fleming, G. R.; Kumar, P. V.; Maroncelli, M. Nature
1994, 369, 471-473.
(64) Beveridge, D. L.; Young, M. A.; Sprous, D. Mol. Model. Nucleic
Acids 1998, 682.
(60) Brooks, C. L.; Karplus, M.; Pettitt, B. M. AdV. Chem. Phys. 1988,
71, 1.
(65) Saif, B.; Mohr, R. K.; Montrose, C. J.; Litovitz, T. A. Biopolymers
1991, 31, 1171-1180.
(61) Frauenfelder, H.; Parak, F.; Young, R. D. Annu. ReV. Biophys.
Biophys. Chem. 1988, 17, 451-479.
(66) Shui, X.; McFail-Isom, L.; Hu, G. G.; Williams, L. D. Biochemistry
1998, 37, 8341-8355.
(62) Frauenfelder, H.; Wolynes, P. G.; Austin, R. H. ReV. Mod. Phys.
1999, 71, S419-S430.
(67) Denisov, V. P.; Carlstrom, G.; Venu, K.; Halle, B. J. Mol. Biol.
1997, 268, 118-136.
(63) Frauenfelder, H.; Sligar, S. G.; Wolynes, P. G. Science 1991, 254,
1598-1603.
(68) Young, M. A.; Jayaram, B.; Beveridge, D. L. J. Am. Chem. Soc.
1997, 119, 59-69.

Measurement of Local DNA Reorganization
J. Am. Chem. Soc., Vol. 121, No. 50, 1999 11649
(s, 1H, ArH), 6.8 (m, 4H, ArH), 6.32 (s, 1H, ArH), 5.3 (dd, J ) 6.2,
9.5 Hz, 1H, C1′-H), 4.45 (m, 1H, C3′-H), 4.10 (m, 1H, C4′-H), 3.80
(s, 6H, 2 OCH3), 3.65 (m, 6H, CH2CH2CN, and 2 CH(CH3)2), 3.4 (dd,
J ) 5.2, 9.8 Hz, 1H, C5′-H), 3.2 (m, 5H, CH2 + C5′-H), 2.8 (m, 2H,
CH2), 2.7 (m, 2H, CH2), 2.40 (ddd, J ) 2.4, 6.2, 13.1 Hz, 1H, C2′-H),
1.9 (m, 5H, 2 CH2 + C2′-H), 1.19 (d, J ) 3.1 Hz, 6H, CH(CH3)2, 1.17
(d, J ) 3.1 Hz, 6H, CH(CH3)2,); 13C NMR (75 MHz, CDCl3) δ 162.7,
158.5, 156.1, 151.2, 145.6, 144.7, 135.7, 130.0, 129.9, 128.1, 127.8,
126.8, 120.9, 117.9, 117.6, 113.2, 106.9, 106.2, 103.5, 86.3, 85.8, 75.1,
74.1, 64.3, 59.2, 58.2, 51.2, 49.8, 49.4, 42.9, 42.1, 24.6, 24.5, 21.5,
20.6, 20.4, 20.3, 20.2, 16.9, 16.8.
DNA Synthesis. Oligonucleotides were synthesized (0.2 µmol scale)
using an Applied Biosystems 392 DNA/RNA synthesizer with com-
mercially available reagents and protocols. Phosphoramidite 5 was
incorporated using standard coupling times for oligomers containing
1. Cleavage from the support and base deprotection was accomplished
by treatment with concentrated NH4OH (16 h, 25 °C). Following HPLC
purification of the 5′-DMT-protected DNA (95 to 50% A over 40 min),
the product fraction was lyophilized and detritylated with 80% acetic
acid (400 µL, 0.5 h, 25 °C). The mixture was concentrated to dryness
and dissolved in 0.2 mL of H2O, and the final oligonucleotide product
was purified by HPLC (92 to 75% A over 30 min). ESIMS for DNA
containing coumarin 1: found 5254.0; calcd 5252.5. ESIMS for DNA
containing tetrahydrofuran 2: found 5108.5; calcd 5105.6.
HPLC purification of the oligonucleotides was performed using a
Perkin-Elmer Model 250 Biocompatible LC pump, LC-290 detector,
and PE Nelson 1022 Plus computing integrator with a 4.1 × 250 mm
Hamilton PRP-1 reverse-phase column (mobile phases: A ) 0.1 M
triethylammonium acetate, pH 6.5; B ) CH3CN).
Molecular Modeling. Energy minimization was performed to
convergence (RMS ) 0.01 Å/mol) with the AMBER* force field and
long-range electrostatic corrections as implemented in MacroModel v
5.5, using explicit sodium counterions and GB/SA water treatment.
Sample Preparation and Spectroscopy. All experiments were
performed in 100 mM phosphate buffer, pH 7.2, at 25 °C. The
concentration of the DNA was ∼1.75 mM (nucleotide). Melting studies
show that the oligomer is double helical under these conditions.
Both time-resolved and steady-state spectroscopy were performed
with magic-angle polarization. Steady-state emission spectra were
collected with excitation at 390 nm and excitation and emission band-
passes of 8 and 4 nm, respectively. Excitation spectra were recorded
with detection at 520 nm and emission and excitation band-passes of
4 and 8 nm, respectively. Melting studies were done on a Beckman
DU 640 UV/VIS spectrophotometer with Peltier temperature control.
A260 values were recorded every 0.5 min from 15 to 80 °C (0.5 deg
C/min).
Fluorescence decays were collected using standard time correlated
single-photon counting techniques.69 Subpicosecond excitation pulses
at 390 nm were generated from the output of a home-built, mode-
locked, Ti:Sapphire laser by using an external acousto-optic pulse
selector, followed by second harmonic generation in a 1 mm BBO
crystal. Fluorescence was collected by a subtractive double monochro-
mator with a band-pass of 3 nm. The instrument response function
typically had a fwhm of 100 ps.
Acknowledgment. We thank Jason L. McCary for perform-
ing molecular modeling studies and Angela Johnson for steady-
state spectroscopy of Coumarin 102. This work was supported
by the National Institutes of Health (GM-47991 to R.S.C. and
GM-55566 to C.J.M.) and by the National Science Foundation
through its Chemistry Division (CHE-9809719 to M.A.B.) and
through the NSF-South Carolina EPSCoR Program (EPS-
9630167 to C.J.M. and M.A.B.). C.J.M. received support from
an NSF CAREER Award, a Cottrell Scholar Award (1996-01),
an Alfred P. Sloan Fellowship (1997-1999), and a Camille
Dreyfus Teacher-Scholar Award (1998-2000). R.S.C. received
support from an Alfred P. Sloan Research Fellowship (1995-
1998).
Supporting Information Available: A circular dichroism
spectrum of the modified DNA oligomer (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
JA992456Q
(69) Birch, D. S.; Imhof, R. E. Topics in Fluorescence Spectroscopy,
Volume 1: Techniques; Lakowicz, J. R., Ed.; Plenum Press: New York,
1991; pp 1-95.