Ultrafast dynamics of 1-ethynylpyrene-modified RNA: A photophysical probe of intercalation

Article abstract of DOI:10.1021/jp103176q

The photophysics of pyrene attached to an adenine base within RNA single strands and duplexes is examined with respect to the position of the pyrene within the strand and the number of pyrenes attached to one duplex. Compounds with pyrenes intercalating sequence specifically are examined, as well as a doubly modified compound, where the two pyrenes are located close enough to each other for significant excimer interaction. Femtosecond transient absorption measurements and time correlated single photon counting measurements allow a thorough examination of the local influences on the pyrene photophysics. Our results suggest that optical excitation establishes an equilibration between two molecular states of different spectroscopic properties and lifetimes that are coupled only via the excited state as a gateway. One of them is a neutral pyrene-adenine excited state, S*, while the second one is connected to an excited charge transfer state, S*_CT. In all compounds, an ultrafast sub-ps decay from a higher excited state into the lowest excited state S* occurs, and an excited charge transfer species S*_CT is formed within picoseconds. The fluorescence behavior of the pyrene-modified adenine, however, is strongly dependent on RNA conformation. Both S* and S*_CT states are fluorescent, and decay within hundreds of picoseconds and ～2 ns, respectively. The ratio between S* and S*_CT fluorescence depends strongly on pyrene intercalation, and it is found that the S* state is quenched selectively upon intercalation of the pyrene into RNA. The doubly modified duplex exhibits an additional fluorescent state with a lifetime of 18.7 ns, which is associated with the pyrene excimer state. This state coexists with a significant population of the pyrene monomer, since the characteristic features of the latter can still be observed. Formation of the excimer occurs on femtosecond time scales. The pyrene label thus provides a sensitive tool to monitor the local structural dynamics of RNA with the chromophore acting as a molecular beacon.

Full text of DOI:10.1021/jp103176q

11638
J. Phys. Chem. B 2010, 114, 11638–11645
Ultrafast Dynamics of 1-Ethynylpyrene-Modified RNA: A Photophysical Probe of
Intercalation
Ute Fo¨rster,† Christian Gru¨newald,‡ Joachim W. Engels,‡ and Josef Wachtveitl*,†
Institute of Biophysics and Institute of Physical and Theoretical Chemistry, Goethe-UniVersity Frankfurt,
Max-Von-Laue-Str. 1, 60438 Frankfurt, Germany, and Institute for Organic Chemistry and Chemical Biology,
Goethe-UniVersity Frankfurt, Max-Von-Laue-Str. 7, 60439 Frankfurt, Germany
ReceiVed: April 8, 2010; ReVised Manuscript ReceiVed: July 9, 2010
The photophysics of pyrene attached to an adenine base within RNA single strands and duplexes is examined
with respect to the position of the pyrene within the strand and the number of pyrenes attached to one duplex.
Compounds with pyrenes intercalating sequence specifically are examined, as well as a doubly modified
compound, where the two pyrenes are located close enough to each other for significant excimer interaction.
Femtosecond transient absorption measurements and time correlated single photon counting measurements
allow a thorough examination of the local influences on the pyrene photophysics. Our results suggest that
optical excitation establishes an equilibration between two molecular states of different spectroscopic properties
and lifetimes that are coupled only via the excited state as a gateway. One of them is a neutral pyrene-adenine
excited state, S*, while the second one is connected to an excited charge transfer state, S*CT. In all compounds,
an ultrafast sub-ps decay from a higher excited state into the lowest excited state S* occurs, and an excited
charge transfer species S*CT is formed within picoseconds. The fluorescence behavior of the pyrene-modified
adenine, however, is strongly dependent on RNA conformation. Both S* and S*CT states are fluorescent, and
decay within hundreds of picoseconds and ∼2 ns, respectively. The ratio between S* and S*CT fluorescence
depends strongly on pyrene intercalation, and it is found that the S* state is quenched selectively upon
intercalation of the pyrene into RNA. The doubly modified duplex exhibits an additional fluorescent state
with a lifetime of 18.7 ns, which is associated with the pyrene excimer state. This state coexists with a
significant population of the pyrene monomer, since the characteristic features of the latter can still be observed.
Formation of the excimer occurs on femtosecond time scales. The pyrene label thus provides a sensitive tool
to monitor the local structural dynamics of RNA with the chromophore acting as a molecular beacon.
Introduction
Fluorescent molecules have proven useful in a wide variety
of biomolecular assays both in ViVo and in Vitro.1,2 They are
established over the whole visible spectral range and can be
applied site-specifically.3 Fluorescence is generally highly
dependent on environmental factors4 and is therefore a well
suited parameter for molecular properties of the fluorophore and
its direct surroundings. Additionally, the sensitivity of the
detection level reaches down to single photon events.
Figure 1. Chemical structure of the parent 2-(1-ethynylpyrene)-
adenosine (PyrA).
Pyrene and its derivatives are among the most prominent
fluorescent labels, especially in the context of nucleic acids. The
high sensitivity of pyrene to solvent effects5-8 and its capability to
form excited state complexes both homogeneously9 and heterogene-
ously10,11 has opened up many possibilities in the fields of RNA
recognition,12,13 electron injection,14,15 bioengineering,16,17 structure
stabilization,18 structure determination,19-21 as well as nucleobase
photophysics.22-25 The photophysics of the pyrene or pyrene
derivatives is extremely sensitive to the extension of the pyrene
aromatic system.26-28 Therefore, if pyrene and the corresponding
base are closely connected, strong interactions between the dye
and the base are to be expected. Presently, pyrene residues are
attached to the 8-position of purines, the 5-position of pyrim-
idines, or the sugar. In contrast, we attached the pyrene to the
2-position of a purine (adenosine) for our investigation.
In a previous study, we exploited 2-(1-ethynylpyrene)-
adenosine (PyrA, Figure 1) as a general RNA folding probe.29
We recently reported a sequence dependence of the intercalation
of this label,30 where we could distinguish between duplexes
with intercalated pyrene and duplexes with the pyrene located
outside of the double strand via temperature dependent pyrene
fluorescence. In the same study, we tested the suitability of
excimers as local probes for RNA dynamics. The photophysics
of the 2-(1-ethynylpyrene)-adenosine in methanol provides clear
evidence for the formation of a charge transfer state and a dual
fluorescence decay of the excited state.31
We now aim to investigate the connection between of the
above-mentioned structural differences and the photophysics of
the PyrA label. We thus use a combination of femtosecond
transient absorption spectroscopy and time correlated single
photon counting to shed light on the ultrafast spectroscopic
* To whom correspondence should be addressed. Phone: 0049 69 798
29351. Fax: 0049 69 798 29709. E-mail: wveitl@theochem.uni-frankfurt.de.
† Institute of Biophysics and Institute of Physical and Theoretical
Chemistry.
‡ Institute for Organic Chemistry and Chemical Biology.
10.1021/jp103176q  2010 American Chemical Society
Published on Web 08/13/2010

Ultrafast Dynamics of 1-Ethynylpyrene-Modified RNA
J. Phys. Chem. B, Vol. 114, No. 35, 2010 11639
TABLE 1: Overview of Synthesized Compounds, Where 1
and 2 Represent the Two Complementary RNA Strands and
the Letters P and X Denote the Pyrene-Modified or
Unmodified Compounds, Respectively
Waltham, Massachusetts) at an excitation wavelength of 380
nm. The concentration was adjusted to ∼10 µM for all samples.
Vis-Pump/Vis-Probe Spectroscopy. A detailed description
of the vis-pump/vis-probe spectroscopy setup has been given
elsewhere.37 In short, the source for the ultrashort laser pulses
was a CLARK CPA 2001 (Clark-MXR, Dexter, MI). It provided
laser pulses of a pulse energy of 800 µJ at 775 nm with a pulse
duration of 170 fs. The excitation pulses of 180 µJ at 387 nm
were generated by frequency doubling of the laser fundamental
in a BBO crystal cut to 29.5° for optimal phase matching. Single
filament supercontinuum white light pulses were generated in
a CaF2 plate and then split into two beams for probe and
reference detection. Two identical 42-segment diode arrays were
used for spectrally broad detection of the absorption and
reference signal.
strand
sequence
Tm
1P
1X
2P
2X
5′-CUUUUCAPyUUCUU-3′
5′-CUUUUCAUUCUU-3′
3′-GAAAACUAPyAGAA-5′
3′-GAAAAGUAAGAA-5′
5′-CUUUUCAPyUUCUU-3′
3′-GAAAAGUAAGAA-5′
5′-CUUUUCAUUCUU-3′
3′-GAAAAGUAPyAGAA-5′
5′-CUUUUCAPyUUCUU-3′
3′-GAAAAGUAPyAGAA-5′
1P2X
35.9 °C
30.0 °C
45.5 °C
1X2P
1P2P
A fused silica cuvette with 1 mm optical path length was
moved perpendicular to the incident beam to avoid excessive
photobleaching.
properties of the pyrene-modified adenine within RNA strands
and thus the dynamics of the RNA itself.
UV-vis absorption spectra were taken before and after the
measurement. These spectra could give assurance that the
photodamage to the chromophores was low and also indicated
that little degradation of the modified RNA occurred during
measurement.
Experimental Methods
Sample Preparation. Synthesis of oligonucleotides followed
standard procedures using the well established ACE-Chemis-
try.32 Site specific modification with 1-ethynyl-pyrene was
accomplished by Sonogashira cross-coupling to an iodinated
nucleobase (adenine) within the strand while still on solid
support33 exploiting the general advantages of solid-phase
synthesis (high yield, ease of purification, etc.). 2-(1-Ethyn-
ylpyrenyl)-3′,5′-O-(tetraisopropyldisiloxano)-N6-(N′,N′-dimethy-
lamino-methyleno)-adenosine (a lipophilic pyrene-adenosine
derivative as model compound) was also synthesized by
Sonogashira cross-coupling of the protected, iodinated nucleo-
side and 1-ethynyl-pyrene.
As a prerequisite, first, 2-iodo-adenosine had to be synthesized
starting with guanosine using literature procedures.34-36 Sub-
sequent conversion to the ACE-phosphoramidite and automated
RNA synthesis yielded the solid-support bound oligonucleotide.
This iodinated oligonucleotide was then reacted two times with
1-ethynyl-pyrene in a Pd-catalyzed Sonogashira cross-coupling
reaction under exclusion of oxygen at room temperature for 3 h.
After deprotection and cleavage from the solid support, the
oligonucleotide was purified by anion-exchange HPLC. After
desalting, the oligonucleotides were analyzed using MALDI-
TOF mass spectrometry (see ref 29 for experimental details).
Two complementary strands were synthesized in modified
and unmodified form and later hybridized to form the duplexes.
An overview of the complexes and the terminology used can
be found in Table 1 together with the melting temperatures.
The single RNA strands are numbered; the suffix P indicates
the attachment of pyrene, and the suffix X indicates the absence
of pyrene. The specific sequence was chosen to provide good
helix formation. It was found in a previous study30 that the two
duplexes 1P2X and 1X2P show significant structural differences.
In 1P2X, the pyrene was at least partially intercalated into the
RNA duplex, while, in 1X2P, the dye was located mostly outside
of the strand.
Time Correlated Single Photon Counting Fluorescence
Spectroscopy (TCSPC). A detailed description of the combined
TCSPC/upconversion setup has been given elsewhere.31 In short,
a Spectra Physics Tsunami-Spitfire-system (Newport Spectra
Physics, Irvine, California) served as the fs-pulse source,
yielding 100 fs pulses with a pulse energy of 1.2 mJ at 800
nm. The excitation pulses of 120 µJ at 400 nm were generated
in a manner similar to the transient absorption experiment by
frequency doubling of the laser fundamental. Emission was
detected after spectral selection via a Jobin-Yvon D3-180
Gemini double monochromator (Horiba Jobin-Yvon, Unter-
haching, Germany) by a Becker & Hickl MSA 1000 counting
system (Becker & Hickl, Berlin, Germany) using a cooled single
photon counting detector head PMC 100-4 (Becker & Hickl,
Berlin, Germany). The time resolution of the counting card was
1 ns. Like in the absorption experiment, a fused silica cuvette
with 1 mm optical path length was used and moved perpen-
dicular to the incident beam. Fluorescence spectra were taken
before and after measurement to estimate the photoinduced long-
term bleach.
Data Analysis. Transient absorption measurements: Several
corrections of the raw data set are necessary prior to the
application of a fitting procedure to the absorption data: solvent
signals have to be subtracted and the transients have to be
corrected for group velocity dispersion using a procedure
introduced by Kovalenko et al.,38 which takes the temporal
evolution of the coherent signal of pure buffer solution into
account.
For the quantitative analysis of the absorption data, we used
a kinetic model that describes the data as a sum of n exponential
decays with the associated lifetimes τi, convoluted with the
system response function. The model assumes Gaussian pump
and probe pulses with a 1/e cross correlation width tcc.
All spectroscopic measurements of oligonucleotides were
performed in phosphate buffer containing 10 mM NaH2PO4,
10 mM Na2HPO4, and 140 mM NaCl adjusted to pH 7.
Steady-State Spectroscopy. Absorption spectra were taken
on a Jasco V670 (Jasco GmbH, Gross-Umstadt, Germany)
absorption spectrometer using a fused silica cuvette with 1 mm
optical path length. All absorption spectra were corrected for
solvent absorption. Fluorescence spectra measurements were
performed on a Perkin-Elmer LS 50 fluorimeter (Perkin-Elmer,
A Levenberg-Marquardt algorithm optimizes a number of
time constants simultaneously to the whole spectrum, yielding
wavelength dependent amplitudes Ai(λ), so-called decay associ-
ated spectra (DAS) corresponding to the different time constants
applied.

11640 J. Phys. Chem. B, Vol. 114, No. 35, 2010
Fo¨rster et al.
2
n
relative quantum efficiency. All compounds exhibit broad, nearly
featureless spectra that can be distinguished mostly by their
shape and central wavelength. Compared to the 1-ethynylpyrene
fluorescence, which shows a variety of characteristic bands
located around 400 nm,40 the observed fluorescence is strongly
red-shifted by 50 nm, losing almost all vibrational fine structure,
as already observed in several similar systems.11,14,41,42 This
strongly indicates the presence of a partial charge transfer
between the pyrene and the adenine into an excited charge
transfer state S*CT and the formation of intramolecular
exciplexes.15,25,26,43 The quantum efficiencies of the singly
modified duplexes vary strongly. In previous studies, similar
variations have been attributed to different conformations of
the pyrene within the RNA.11,41,44 Previous studies conducted
in our group allowed us to conclude via temperature dependent
fluorescence measurements that, in the 1P2X duplex, pyrene is
intercalated into the RNA base stack, while, in 1X2P, pyrene
is located outside of the double strand.30 Intercalation into the
strand leads to substantial quenching of the pyrene fluorescence
due to stacking interactions,11,44 resulting in an overall lower
quantum efficiency of 1P2X compared to 1X2P.30
tcc
t
τi
1
2
∆A(λ, t) )
A (λ) exp
-
·
·
∑
i
(
)
4τi
i)1
tcc
t
tcc
1 + erf
-
(
(
))
2τi
TCSPC Measurements. TCSPC measurements were cor-
rected for stray light from the excitation beam. Then, the same
global fit analysis method as described above for the transient
absorption scheme was also applied for the TCSPC measure-
ments. A cross correlation of 1.2 ns, resulting from a Gaussian
fit of the instrument response function of the system, was used
for deconvolution.31
Results and Discussion
Steady-State Characterization. Absorption spectra were
taken for all five pyrene-modified compounds (Figure 2A). The
spectra exhibit similar shapes including two absorption maxima
of the pyrene-adenine moiety located roughly at 380 and 410
nm, respectively, and a shoulder at 360 nm. An additional
shoulder associated with pyrene is located around 320 nm but
is partially covered by the RNA absorption centered at 260 nm.
The overall spectra are very similar to the ones of 1-ethy-
nylpyrene linked to the 5-position of DNA bases as examined
by Rist et al.39 The shape of the pyrene absorption band is similar
to that of 1-ethynylpyrene in solution;40 however, the band is
red-shifted by almost 50 nm with respect to both pyrene and
1-ethynylpyrene and spectrally broadened significantly. This
leads to the conclusion that the electronic coupling of the
1-ethynylpyrene dye with the adenine base is quite strong.31
The region between 330 and 440 nm can therefore be associated
with the transition into the first allowed excited state, while the
band situated around 320 nm is associated with the absorption
into higher excited states.
The doubly modified RNA duplex exhibits an additional
fluorescence red-shift of 35 nm with respect to the singly
modified duplexes. This is due to the pyrene-pyrene interaction
in the excited state, which leads to excimer-like fluorescence,
as observed in similar systems before.16,29,45,46 A pyrene-pyrene
ground state interaction could already be observed in the
absorption spectra. The effect on the absorption spectrum,
however, is relatively small compared to the 35 nm shift of the
pyrene fluorescence. This leads to the conclusion that, in addition
to the ground state dimer, excited state complexes are formed.
The large shift of the fluorescence wavelength shows that the
excited state interaction is the dominating contribution of the
pyrene dimer formation.
UV-vis Transient Absorption Measurements. Figure 3
shows the time-resolved transient absorption of all five pyrene-
modified compounds. The general spectral features are very
similar. A broad positive absorbance change is observed between
530 nm and the red end of the spectral observation window,
which can be interpreted as the excited state absorption band
(ESA). It is blue-shifted by ∼20 nm in 1P2P with respect to
the other compounds. As discussed above, this effect can be
attributed to the presence of an exciplex-like state in the doubly
modified compound.
A small red-shift can be seen in 1P2X compared to 1P; this
effect however is absent in 1X2P (and the corresponding pyrene-
modified single strand 2P). The ratio between the two major
pyrene bands changes upon hybridization of the RNA; however,
these effects have already been discussed in ref 30 and are not
a subject of this study.
The pyrene absorption spectrum of the doubly modified
duplex is slightly red-shifted with respect to the singly modified
duplexes, indicating that a pyrene-pyrene complex is partially
present already in the ground state of the compound. Such a
complex would commonly lower the energies of both the ground
state and the excited state. Due to the usually higher polarity
of the excited state, the effect of dimerization is often more
prominent in the excited state, leading to an overall red shift of
the absorption band. The observation of a red-shifted absorption
is in contrast to the common notion of an excimer complex,
which is usually encountered in pyrenes at close proximity,
where the interaction between the two pyrenes is limited to the
excited state.9 The reason for this substantial difference may
be found in the fact that the two pyrenes are connected directly
to the RNA and are thus fixed at a distance of roughly 4 Å. A
classic excimer, on the other hand, is usually found in solution,
where the mean distance of two pyrene dyes is considerably
larger. As a result, in solution, even at high concentration,
ground-state interaction is not seen.
A ground state bleach (GSB) exhibiting two maxima can be
seen up to 420 nm, corresponding well to the bands of the
steady-state absorption spectra. Around 450 nm, stimulated
emission appears. This band is red-shifted in 1P2P with respect
to the singly modified compounds, as already mentioned in the
previous paragraph. Therefore, it strongly overlaps with the
excited state absorption and cannot be resolved. It can be
concluded from the simultaneously occurring shifts of ESA and
fluorescence in 1P2P that the higher excited states are less
affected by the presence of the second pyrene than the first
excited state.
In all compounds, a dynamic blue shift of the ESA of about
10 nm can be observed, which occurs on the order of a few
picoseconds and corresponds to a similar red-shift in the
stimulated emission region. Prior to quantitative analysis, it can
already be seen that all compounds exhibit a relatively long
general recovery time of >1 ns, with the notable exception of
1P2X, which shows a significantly faster overall recovery. On
the other hand, the temporal characteristics of 1P2P suggest a
recovery time which clearly exceeds the experimental observa-
Figure 2B shows the fluorescence spectra of the pyrene-
modified single and double strands. The spectra were taken at
equal concentrations and corrected for absorption at the excita-
tion wavelength. They therefore represent an estimate of the

Ultrafast Dynamics of 1-Ethynylpyrene-Modified RNA
J. Phys. Chem. B, Vol. 114, No. 35, 2010 11641
Figure 2. (A) Absorption spectra of modified single and double stranded RNA. The lowest pyrene transition ranges from 350 to 430 nm; pyrene
S0-Sn absorption is located around 300-330 nm and is significantly weaker than the lowest transition. Nucleobase absorption can be found around
260 nm. All spectra were taken at a concentration of ∼10 µmol in buffer solution at room temperature. (B) Fluorescence spectra of modified single
and double stranded RNA. Spectra were corrected for absorption at an excitation wavelength of 385 nm and are therefore allowing a relative
estimate of the samples’ quantum efficiency. All spectra were taken at a concentration of ∼10 µmol in buffer solution at room temperature.
TABLE 2: Time Constants of Pyrene Labeled RNA Model
Compounds Obtained by Global Fit Analysis of the
Transient Absorption Measurements
τ1
τ2
τ3
τ4
1P
2P
1P2X
1X2P
1P2P
120 fs
85 fs
114 fs
160 fs
210 fs
2.7 ps
2.8 ps
3.7 ps
3.6 ps
5 ps
270 ps
295 ps
350 ps
550 ps
275 ps
2 ns
2.1 ns
2.1 ns
2.4 ns
5 ns
contribution in the red wing and a negative contribution at the
blue wing of the excited state absorption. In addition to this, a
positive contribution in the range of the stimulated emission
can be seen for the singly modified compounds. While, as
already mentioned, the DAS of τ1 for all compounds are
certainly perturbed by coherent effects (such as wavepacket
motion),47 they show some general characteristics of relaxation
into the lowest level of the S* state. It is a commonly accepted
feature of the photophysics of pyrene and many of its derivatives
that excitation mostly occurs into the S2 state, followed by a
fast S1 population.26,27,48-50 However, the comparatively fast
overall recovery time of the system observed in this study
suggests an allowed S1-S0 transition, and thus the directly
excited state is presumed to be the S1 state. The deexcitation
process seems to be slightly faster if the pyrene modification is
located on strand 2.
For the doubly modified compound 1P2P, the DAS look
slightly different. This can partially be explained by the fact
that the ESA is blue-shifted for the pyrene-pyrene dimer. In
addition to this, non diffusion controlled excimer formation
should occur on this time scale, and is superimposed to the other
processes, obscuring the dynamics of the decay into S*.
The second decay constant τ2 ranges from 2.7 to 5 ps. It
corresponds to a dynamic shift, which is visible in both the
excited state absorption and the stimulated emission. A similar
behavior has been observed in pyrenyl biphenyl esters,26 5-(1-
ethynylypyrene)-uracil,14 and the modified base used in this
study,31 and has been assigned to a charge-transfer process,
forming an excited pyrene-adenine charge-transfer state
(S*CT).25 The charge-transfer process is slower in the duplexes,
compared to the single strands in solution. Possibly, the
π-stacking interactions within the base stack render charge
transfer into the RNA slightly more unfavorable. The transition
Figure 3. Transient absorption spectra of all pyrene-modified com-
pounds. Spectra are color-coded, with red indicating positive relative
absorption and blue indicating negative relative absorption. The time
axis is linear up to 1 ps and logarithmic for longer delay times. All
spectra were taken at a concentration of ∼5-10 µmol in buffer solution
at room temperature.
tion window. It can already be concluded at this stage that, apart
from the spectral differences between 1P2P and the singly
modified compounds, the changes in the overall recovery time
of the system appear to reflect most directly the structural
differences of the examined compounds.
All five data sets have been analyzed by the global fit routine
as described above. For each of the five pyrene-modified
compounds, four time constants were required for an optimized
fit and are shown in Table 2. The corresponding wavelength-
dependent amplitudes (DAS) are depicted in Figure 4.
The shortest decay constant τ1 ranges from 85 to 210 fs, and
is thus in the range of the temporal resolution of the system. Its
general features are also obstructed by coherent effects,
especially around the excitation wavelength, due to the overlap
of pump and probe pulses.47 Nonetheless, a tentative discussion
of the long wavelength range of the DAS, which shows several
common features, is possible. All DAS contain a positive

11642 J. Phys. Chem. B, Vol. 114, No. 35, 2010
Fo¨rster et al.
Figure 4. Decay associated spectra (DAS) for 1P, 1P2X, 1X2P, and 1P2P as obtained by a global fitting procedure. Four time constants were
sufficient to describe the data in each case.
is even slower in the doubly modified duplex, where the pyrene
excimer is probably less likely to transfer an electron into the
base stack than the monomer.
the relation between the S*CT and S* fluorescence components
is reversed in 1P2X, where the A3/A4 ratio is 2.2. A
quantitative comparison of the fluorescence amplitudes
derived from the DAS and a correlation to the samples’
quantum efficiencies is not possible due to variations in the
experimental conditions and the overlap of the ESA and the
stimulated emission bands. However, connection of the
amplitude ratios to the steady-state quantum efficiencies of
the pyrene-modified RNA strands (as shown via the steady-
state fluorescence spectra, and, in more detail, in ref 30)
allows a qualitative discussion of the lifetime distribution.
1P2X shows a considerably smaller quantum efficiency than
any of the single strands or the other singly modified duplex
1X2P. As mentioned before, the reason for this can be found
in the intercalation of pyrene, and thus efficient quenching
of the pyrene fluorescence.30,44 Comparison of the A3/A4 ratio
of 1P and 1P2X shows that the long-lived fluorescence state,
which is associated with the S*CT state, is quenched
selectively. The single strands in general show a smaller A3/
A4 ratio than the duplexes. Thus, in single strands, the
fluorescence of the S*CT state is generally more pronounced
with respect to the S* state. This might be explained by the
energy distribution along a base-paired stack, which could
render fluorescence from the S*CT state more unfavorable
with respect to other decay paths. The higher fluorescence
quantum efficiency in 1X2P compared to its single strand
1P in solution results from a selective enhancement of the
fluorescence of the S* state.
The third and fourth decay constants τ3 and τ4, ranging from
270 to 550 ps and from 2 to 5 ns, respectively, both show a
strong positive signal in the range of the excited state absorption,
and negative signals in the stimulated emission band and in the
range of the ground state absorption. The spectral characteristics
can therefore be attributed to the S*CT and S* decays. Here,
the largest differences between the individual compounds
become evident. In the singly modified compounds, the central
wavelengths of both the excited state absorption and the
fluorescence differ by about 15 nm, and can thus be attributed
to two different states. Since a S*CT state, which is slightly lower
in energy than the S* state, is likely to be present in the
compound, the lower-energy transition (τ4) can be attributed to
the fluorescence of the S*CT state, while the higher-energy
transition (τ3) corresponds to the fluorescence of the S* state.
It should be noted that both fluorescence contributions are broad
and featureless. While for the S*CT state this was to be expected,
the featureless characteristics of the excited state fluorescence
allow the conclusion that the excitation in S* is delocalized over
both the pyrene and the adenine part, thus broadening the
fluorescence spectrum significantly.
The ratio between the longer and shorter lifetimes varies
from compound to compound. While a direct comparison of
the DAS amplitudes is difficult due to the different spectral
characteristics for τ3 and τ4, the ratio of the ESA maximum
amplitudes can be used for a qualitative discussion. For the
single strands, the A3/A4 ratio was found to be small (0.10
for 1P, 0.18 for 2P) and higher in 1X2P (0.31). However,
For the doubly modified compound, fluorescence lifetimes
of 275 ps and 5.1 ns were found. However, the latter time
constant is well beyond the experimental delay time range and

Ultrafast Dynamics of 1-Ethynylpyrene-Modified RNA
J. Phys. Chem. B, Vol. 114, No. 35, 2010 11643
Figure 5. Left: 2-D fluorescence spectra of strand 1P2P. The spectrum is color-coded, with blue indicating no fluorescence and red indicating
maximum fluorescence. Right: Global fit analysis of the measurement that shows two decay times associated with two separate spectra.
is difficult to determine directly in a pump-probe experiment.
The spectral features of τ3 are nearly identical to those of the
corresponding components in the singly modified compounds.
This leads to the conclusion that the S* fluorescence is not
disturbed by the presence of the second pyrene but coexists with
the exciplex state. A TCSPC experiment, which is better suited
for this time regime, was thus carried out to further analyze the
long-lived fluorescent state.
TCSPC Measurements. Figure 5 shows the TCSPC mea-
surements on the doubly modified duplex 1P2P. It can be readily
seen that, in the shorter wavelength range, the fluorescence
decays significantly faster than for longer wavelengths. A global
Figure 6. Reaction scheme for the singly modified RNA strands.
fit, carried out as described above, yields two different decay
constants of 2.5 and 18.7 ns, which are spectrally well separated
from one another. The decay associated spectra (Figure 5, right)
relative to the duplex allows us to draw conclusions on the
underline the spectral differences of the two components. The
interactions between the label and its local environment.
Examination of two singly modified duplexessone with
pyrene intercalated and the other one with pyrene located outside
of the duplexsand a doubly modified compound as well as the
amplitude of the smaller time constant exhibits the spectral
features of the monomer pyrene S*CT fluorescence known from
the singly modified compounds. The additional second com-
ponent of 18.7 ns was not found in the TCSPC measurements
comparison to singly modified single strands allowed us to
of any of the singly modified strands or duplexes (data not
examine the influence of various RNA surroundings on the
shown) and can therefore be unambiguously associated with
photophysics of pyrene.
the pyrene excimer. The doubly modified compound thus
Figure 6 summarizes the conclusions for the photophysics
exhibits three fluorescence decay times in total: 275 ps (which
of the singly modified RNA strands. Excitation probably occurs
is not visible in the TCSPC experiment due to the time resolution
into a hot S1 state, even though an excitation into a higher
of 1 ns), 2.5 ns, both being associated with the monomer
excited state cannot be completely excluded. Within a time
complex, and 18.7 ns, associated with the pyrene excimer. The
pyrene population of the doubly modified compound can
frame of 85-210 fs, the system reaches the lowest excited state
S*. While results in this time regime have to be interpreted
therefore be divided into a monomer component, which exhibits
carefully due to the limited time resolution of the setup, it can
all characteristics of the monomer photophysics, and an excimer
be stated that in the doubly modified duplex this process exhibits
component, which shows a significantly different behavior. The
the longest lifetime of all compounds. Presumably, in 1P2P,
separation into these two populations occurs very early after
the observed spectral characteristics originate not only from an
excitation, since no mutual influences of monomer and excimer
ultrafast decay from a higher excited state to the S* state but
fluorescence can be observed. Another explanation might be
also from excimer formation, which, for densely packed pyrenes,
found in the heterogeneity of the sample, thus a superposition
may occur on the same time scale.
of different populations.
From the S* state, a charge transfer state S*CT can be
reached. The formation time of the charge transfer state is
Conclusions
on the order of a few picoseconds. It is slower in duplexes,
presumably due to the π-stacking between the RNA bases,
and slowest in 1P2P.
From both S*CT and S* states, fluorescence into the ground
state occurs. The S* state exhibits a lifetime of a few hundred
picoseconds, while the decay time of the S*CT state occurs
on the order of 2 ns. The ratio between the two decay
On the basis of the structural conclusions drawn in ref 30
for the same system, we investigated the structural influences
of intercalation and dimerization of pyrene attached to RNA
with femtosecond pump-probe absorption spectroscopy and
time correlated single photon counting. The observation of the
ultrafast photodynamics of the pyrene in various positions

11644 J. Phys. Chem. B, Vol. 114, No. 35, 2010
Fo¨rster et al.
(21) Kawai, K.; Yoshida, H.; Takada, T.; Tojo, S.; Majima, T. Formation
of pyrene dimer radical cation at the internal site of oligodeoxynucleotides.
J. Phys. Chem. B 2004, 108, 13547–13550.
(22) Netzel, T. L.; Zhao, M.; Nafisi, K.; Headrick, J.; Sigman, M. S.;
Eaton, B. E. Photophysics of 2′-Deoxyuridine (Du) Nucleosides Covalently
Substituted with Either 1-Pyrenyl or 1-Pyrenoyl - Observation of Pyrene-
to-Nucleoside Charge-Transfer Emission in 5-(1-Pyrenyl)-Du. J. Am. Chem.
Soc. 1995, 117, 9119–9128.
(23) Manoharan, M.; Tivel, K. L.; Zhao, M.; Nafisi, K.; Netzel, T. L.
Base-Sequence Dependence of Emission Lifetimes for DNA Oligomers
and Duplexes Covalently Labeled with Pyrene - Relative Electron-
Transfer Quenching Efficiencies of a-Nucleoside, G-Nucleoside, C-
Nucleoside, and T-Nucleoside toward Pyrene. J. Phys. Chem. 1995, 99,
17461–17472.
(24) Amann, N.; Pandurski, E.; Fiebig, T.; Wagenknecht, H. A. A model
nucleoside for electron injection into DNA: 5-pyrenyl-2 ′-deoxyribose.
Angew. Chem., Int. Ed. 2002, 41, 2978–2980.
(25) Huber, R.; Fiebig, T.; Wagenknecht, H. A. Pyrene as a fluorescent
probe for DNA base radicals. Chem. Commun. 2003, 1878–1879.
(26) Fiebig, T.; Stock, K.; Lochbrunner, S.; Riedle, E. Femtosecond
charge transfer dynamics in artificial donor/acceptor systems: Switching
from adiabatic to nonadiabatic regimes by small structural changes. Chem.
Phys. Lett. 2001, 345, 81–88.
(27) Pandurski, E.; Fiebig, T. Femtosecond dynamics in directly linked
pyrenyl donor-acceptor systems: orbital control of optical charge transfer
in the excited state. Chem. Phys. Lett. 2002, 357, 272–278.
(28) Maeda, H.; Maeda, T.; Mizuno, K.; Fujimoto, K.; Shimizu, H.;
Inouye, M. Alkynylpyrenes as improved pyrene-based biomolecular probes
with the advantages of high fluorescence quantum yields and long
absorption/emission wavelengths. Chem.sEur. J. 2006, 12, 824–831.
(29) Gru¨newald, C.; Kwon, T.; Piton, N.; Fo¨rster, U.; Wachtveitl, J.;
Engels, J. W. RNA as scaffold for pyrene excited complexes. Bioorg. Med.
Chem. 2008, 16, 19–26.
(30) Fo¨rster, U.; Lommel, K.; Sauter, D.; Gru¨newald, C.; Engels, J. W.;
Wachtveitl, J. 2-(1-Ethynylpyrene)-adenosine as a folding probe for
RNA - Pyrene in or out. ChemBioChem 2010, 11, 664–672.
(31) Fo¨rster, U.; Gildenhoff, N.; Gru¨newald, C.; Engels, J. W.; Wacht-
veitl, J. Photophysics of 1-ethynylpyrene-modified RNA base adenine. J.
Lumin. 2009, 129, 1454–1458.
(32) Scaringe, S. A.; Wincott, F. E.; Caruthers, M. H. Novel RNA
synthesis method using 5′-O-silyl-2′-O-orthoester protecting groups. J. Am.
Chem. Soc. 1998, 120, 11820–11821.
(33) Khan, S. I.; Grinstaff, M. W. Palladium(0)-catalyzed modification
of oligonucleotides during automated solid-phase synthesis. J. Am. Chem.
Soc. 1999, 121, 4704–4705.
(34) Robins, M. J.; Uznanski, B. Nucleic-Acid Related-Compounds,
Conversions of Adenosine and Guanosine to 2,6-Dichloro, 2-Amino-6-
Chloro, and Derived Purine Nucleosides. Can. J. Chem. 1981, 59, 2601–
2607.
(35) Matsuda, A.; Shinozaki, M.; Yamaguchi, T.; Homma, H.; Nomoto,
R.; Miyasaka, T.; Watanabe, Y.; Abiru, T. Nucleosides and Nucleotides
0.103. 2-Alkynyladenosines - a Novel Class of Selective Adenosine-A2
Receptor Agonists with Potent Antihypertensive Effects. J. Med. Chem.
1992, 35, 241–252.
(36) Matsuda, A.; Shinozaki, M.; Suzuki, M.; Watanabe, K.; Miyasaka,
T. A Convenient Method for the Selective Acylation of Guanine Nucleo-
sides. Synthesis 1986, 385–386.
(37) Lenz, M. O.; Huber, R.; Schmidt, B.; Gilch, P.; Kalmbach, R.;
Engelhard, M.; Wachtveitl, J. First steps of retinal photoisomerization in
proteorhodopsin. Biophys. J. 2006, 91, 255–262.
(38) Kovalenko, S. A.; Dobryakov, A. L.; Ruthmann, J.; Ernsting, N. P.
Femtosecond spectroscopy of condensed phases with chirped supercon-
tinuum probing. Phys. ReV. A 1999, 59, 2369–2384.
(39) Rist, M.; Amann, N.; Wagenknecht, H. A. Preparation of 1-ethy-
nylpyrene-modified DNA via Sonogashira-type solid-phase couplings and
characterization of the fluorescence properties for electron-transfer studies.
Eur. J. Org. Chem. 2003, 2498–2504.
(40) Seo, Y. J.; Rhee, H.; Joo, T.; Kim, B. H. Self-duplex formation of
an A(Py)-substituted oligodeoxyadenylate and its unique fluorescence. J. Am.
Chem. Soc. 2007, 129, 5244–5247.
channels is dependent on the position of the pyrene within
or outside the RNA strand. For intercalated pyrene, the S*
state fluorescence dominates the emission behavior. This
allows the conclusion that upon intercalation the S*CT state
is quenched selectively, due to the greater overlap of the
π-system of the excited molecule with the π-system of the
surrounding bases.
In the doubly modified compound, an additional fluorescence
lifetime of 18.7 ns could be detected and associated with pyrene
excimer emission. It was found, however, that the ultrafast
dynamics of the pyrene-base compound remains otherwise
nearly unchanged, which leads to the conclusion that formation
of the pyrene complex must occur from higher excited states
and thus within the time resolution of the system.
Acknowledgment. The authors thank Dr. Markus Braun for
helpful discussions and suggestions and the Cluster of Excel-
lence “Macromolecular Complexes” of the University of
Frankfurt for financial support.
References and Notes
(1) Valeur, B. Molecular Fluorescence; Wiley-VCH: Weinheim,
Germany, 2002.
(2) Haugland, R. P. Handbook of Fluorescent Probes and Research
Chemicals; Molecular Probes: Eugene, OR, 2002.
(3) Lakowizc, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Springer: Berlin, 1999.
(4) Birks, J. B. Photophysics of Aromatic Molecules; Wiley-Inter-
science: London and New York, 1970.
(5) Nakajima, A. Intensity Enhancement Induced by Solute-Solvent
Interaction between Pyrene and Polar-Solvents. Spectrochim. Acta, Part A
1982, 38, 693–695.
(6) Nakajima, A. Fluorescence-Spectra of Pyrene in Chlorinated
Aromatic Solvents. J. Lumin. 1976, 11, 429–432.
(7) Nakajima, A. Fluorescence Lifetime of Pyrene in Different Solvents.
Bull. Chem. Soc. Jpn. 1973, 46, 2602–2604.
(8) Hara, K.; Ware, W. R. Influence of Solvent Perturbation on the
Radiative Transition-Probability from the B-1(Iu) State of Pyrene. Chem.
Phys. 1980, 51, 61–68.
(9) Fo¨rster, T. Excimers. Angew. Chem., Int. Ed. 1969, 8, 333–343.
(10) Nakajima, A. Intensity Enhancement Caused by Complex-Forma-
tion between Pyrene and Alcohols. Bull. Chem. Soc. Jpn. 1983, 56, 929–
930.
(11) Seo, Y. J.; Ryu, J. H.; Kim, B. H. Quencher-free, end-stacking
oligonucleotides for probing single-base mismatches in DNA. Org. Lett.
2005, 7, 4931–4933.
(12) Okamoto, A.; Kanatani, K.; Saito, I. Pyrene-labeled base-
discriminating fluorescent DNA probes for homogeneous SNP typing. J. Am.
Chem. Soc. 2004, 126, 4820–4827.
(13) Mahara, A.; Iwase, R.; Sakamoto, T.; Yamaoka, T.; Yamana, K.;
Murakami, A. Detection of acceptor sites for antisense oligonucleotides on
native folded RNA by fluorescence spectroscopy. Bioorg. Med. Chem. 2003,
11, 2783–2790.
(14) Trifonov, A.; Raytchev, M.; Buchvarov, I.; Rist, M.; Barbaric, J.;
Wagenknecht, H. A.; Fiebig, T. Ultrafast energy transfer and structural
dynamics in DNA. J. Phys. Chem. B 2005, 109, 19490–19495.
(15) Wanninger-Weiss, C.; Valis, L.; Wagenknecht, H. A. Pyrene-
modified guanosine as fluorescent probe for DNA modulated by charge
transfer. Bioorg. Med. Chem. 2008, 16, 100–106.
(16) Nakamura, M.; Murakami, Y.; Sasa, K.; Hayashi, H.; Yamana, K.
Pyrene-zipper array assembled via RNA duplex formation. J. Am. Chem.
Soc. 2008, 130, 6904–6905.
(17) Maie, K.; Nakamura, M.; Yamana, K. Photocurrent responses from
pyrene-modified RNA duplexes on gold surface. Nucleic Acids Symp Ser.
2007, 319–320.
(18) Filichev, V. V.; Astakhova, I. V.; Malakhov, A. D.; Korshun, V. A.;
Pedersen, E. B. 1-, 2-, and 4-Ethynylpyrenes in the Structure of Twisted
Intercalating Nucleic Acids: Structure, Thermal Stability, and Fluorescence
Relationship. Chem.sEur. J. 2008, 14, 9968–9980.
(41) Hwang, G. T.; Seo, Y. J.; Kim, S. J.; Kim, B. H. Fluorescent
oligonucleotide incorporating 5-(1-ethynylpyrenyl)2 ′-deoxyuridine: sequence-
specific fluorescence changes upon duplex formation. Tetrahedron Lett.
2004, 45, 3543–3546.
(42) Mayer, E.; Valis, L.; Wagner, C.; Rist, M.; Amann, N.; Wagen-
knecht, H. A. 1-ethynylpyrene as a tunable and versatile molecular beacon
for DNA. ChemBioChem 2004, 5, 865–868.
(43) Techert, S.; Schmatz, S.; Wiessner, A.; Staerk, H. Photophysical
characteristics of directly linked pyrene-dimethylaniline derivatives. J. Phys.
Chem. A 2000, 104, 5700–5710.
(19) Wang, G. J.; Bobkov, G. V.; Mikhailov, S. N.; Schepers, G.; Van
Aerschot, A.; Rozenski, J.; Van der Auweraer, M.; Herdewijn, P.; De Feyter,
S. Detection of RNA Hybridization by Pyrene-Labeled Probes. ChemBio-
Chem 2009, 10, 1175–1185.
(20) Silverman, S. K.; Cech, T. R. RNA tertiary folding monitored by
fluorescence of covalently attached pyrene. Biochemistry 1999, 38, 14224–
14237.
(44) Nakamura, M.; Fukunaga, Y.; Sasa, K.; Ohtoshi, Y.; Kanaori, K.;
Hayashi, H.; Nakano, H.; Yamana, K. Pyrene is highly emissive when

Ultrafast Dynamics of 1-Ethynylpyrene-Modified RNA
J. Phys. Chem. B, Vol. 114, No. 35, 2010 11645
attached to the RNA duplex but not to the DNA duplex: the structural basis
of this difference. Nucleic Acids Res. 2005, 33, 5887–5895.
(45) Langenegger, S. M.; Haner, R. Excimer formation by interstrand
stacked pyrenes. Chem. Commun. 2004, 2792–2793.
(46) Hwang, G. T.; Seo, Y. J.; Kim, B. H. Pyrene-labeled deoxyuridine
and deoxyadenosine: fluorescent discriminating phenomena in their oligo-
nucleotides. Tetrahedron Lett. 2005, 46, 1475–1477.
(47) Lorenc, M.; Ziolek, M.; Naskrecki, R.; Karolczak, J.; Kubicki, J.;
Maciejewski, A. Artifacts in femtosecond transient absorption spectroscopy.
Appl. Phys. B 2002, 74, 19–27.
(48) Tanaka, J. Electronic Spectra of Pyrene Chrysene Azulene Coronene
and Tetracene Crystals. Bull. Chem. Soc. Jpn. 1965, 38, 86–102.
(49) Yoshinaga, T.; Hiratsuka, H.; Tanizaki, Y. Electronic Absorption-
Spectra of Pyrene and Hydropyrenes. Bull. Chem. Soc. Jpn. 1977, 50, 3096–
3102.
(50) Neuwahl, F. V. R.; Foggi, P. Direct observation of S-2-S-1 internal
conversion in pyrene by femtosecond transient absorption. Laser Chem.
1999, 19, 375–379.
JP103176Q