Oligodeoxynucleotides containing multiple thiophene-modified isomorphic fluorescent nucleosides

Article abstract of DOI:10.1021/jo4008964

5-(Thien-2-yl)-2′-deoxyuridine, an isomorphic fluorescent nucleoside analogue, was incorporated into multiple positions within single stranded oligodeoxynucleotides. With minimal impact on duplex stability and overall structure, oligonucleotides containin

Full text of DOI:10.1021/jo4008964

Note
pubs.acs.org/joc
Oligodeoxynucleotides Containing Multiple Thiophene-Modified
Isomorphic Fluorescent Nucleosides
Mary S. Noe, Renatus W. Sinkeldam, and Yitzhak Tor*
́
Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla, California 92093-0358, United States
S
* Supporting Information
ABSTRACT: 5-(Thien-2-yl)-2′-deoxyuridine, an isomorphic
fluorescent nucleoside analogue, was incorporated into
multiple positions within single stranded oligodeoxynucleo-
tides. With minimal impact on duplex stability and overall
structure, oligonucleotides containing three identical isomor-
phic fluorescent nucleosides in alternating or neighboring
positions display enhanced, sequence-dependent on-signals for
either duplex formation or dissociation.
he biological roles of nucleic acids have been demon-
strated to be highly dependent on transient secondary and
and the resulting excitonic and exciplex states were analyzed,
providing insight into the excited state electronic behavior of a
duplex.
T
tertiary structures. Although structural studies offer a window
into local structural changes, only techniques with the
appropriate time scales, such as fluorescence spectroscopy,
offer a convenient real-time monitoring of such processes.
While the native nucleosides in DNA and RNA are virtually
nonemissive,1 synthetic isomorphic fluorescent nucleoside
analogues have enabled the monitoring of nucleic acid
dynamics, conformational changes and recognition by diverse
ligands.2,3 With little to no impact on native behavior, such
minimally perturbing nucleoside analogues can enable the
visualization of local structural changes.4,5
Isomorphic fluorescent nucleoside analogues have been
successfully employed in biophysical assays to detect abasic
and oxidized sites,6,7 as well as facilitate the detection of single
nucleotide polymorphisms (SNPs),8 and nucleic acid−drug
interactions.4,5,9,10 Typically a single fluorescent nucleoside
analogue is used. Even with an appreciable response to changes
in their microenvironment, many of these small fluorophores
suffer from a relatively low quantum yield or significantly
diminished brightness upon incorporation into oligonucleo-
tides. Since fluorescent probes may display sensitivity to other
chromophores and fluorophores through ground-state and
excited-state interactions, one can take advantage of multi-
chromophoric systems to elicit an enhanced fluorescence signal
response. In this contribution, we report the preparation as well
as the biophysical and photophysical investigation of
oligonucleotides in which multiple positions are replaced with
isomorphic fluorescent nucleoside analogues.
We have previously developed a series of nucleoside
analogues based upon conjugating an aromatic residue to the
5-position of pyrimidines.14−16 This nonperturbing modifica-
tion extends the π-system, rendering these modified nucleo-
sides emissive. A single bond attaching a 5-membered ring, such
as a thiophene or a furan, to the 5-position imparts molecular
rotor behavior.17 In viscous media, free rotation about this
single bond between the 5-membered ring and the pyrimidine
residue is hampered, resulting in an enhanced fluorescence
quantum yield. Oligonucleotides, singly modified with such 5-
modified fluorescent nucleosides have been used in a number of
biophysical assays.6,18 Here we evaluate the biophysical and
photophysical features of oligonucleotides, which contain
multiple incorporations of 5-(thien-2-yl)-2′-deoxyuridine (1,
Scheme 1), an isomorphic fluorescent T analogue, and compare
the behavior of oligonucleotides with distinct nucleotide
substitution patterns.
Nucleoside 1 was prepared through a standard palladium-
mediated cross-coupling reaction between 5-iodo-2′-deoxyur-
idine and 2-(tributylstannyl)thiophene as reported.6,19 The 5′-
OH was protected with 4,4′-dimethoxytrityl chloride
(DMTrCl). Phosphitylation with 2-cyanoethyl N,N,N′,N′-
tetraisopropylphospordiamidite in the presence of 1H-tetrazole
afforded the 5′-DMT-protected phosphoramidite 2 (Scheme
1). Modified oligonucleotides 3−5 were prepared using
standard cycles following the design criteria articulated below,
along the complementary oligonucleotides (6, 7), which
required only commercially available building blocks. The
resulting oligonucleotides were cleaved, deprotected, gel-
purified, desalted and characterized by MALDI-TOF mass
spectrometry (Table 1) and detailed photophysical analysis.
Several studies have been reported with oligonucleotides that
contain multiple fluorophores.11−13 Only a few have been
designed to minimize significant structural perturbations. In
one relevant study the isomorphic fluorescent nucleoside
analogue 2-aminopurine (2-AP) was used to probe the
electronic properties of DNA duplexes.13 The emissive 2-AP
was placed in several strategic positions throughout the duplex,
Received: May 1, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo4008964 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
to be responsive to solvent viscosity.17,22,23 Even though
mixtures of dioxane and water are commonly used to control
sample polarity,24,25 they also impart small nonlinear changes in
viscosity.26 Both changes in emission intensity of 1 and medium
viscosity as a function of the water fraction in dioxane are
plotted (Figure 2b). The two maxima almost concur, strongly
indicating that the observed fluorescence intensity changes are
indeed the result of viscosity changes.
Scheme 1. Synthesis of Phosphoramidite 2 and
Oligonucleotides
a
To assess the impact of replacing native residues with the
thiophene-containing pyrimidine analogue, the basic biophys-
ical features of the modified oligonucleotides were evaluated.
All oligonucleotides form stable DNA duplexes as determined
by thermal denaturation studies (Figure 3a). The measured Tm
values for 3·6, 4·6, and 5·7 (59, 57, and 61 °C, respectively)
closely resemble the Tm values for the native strands 6·8 and
7·9 (59 and 62 °C, respectively). The adjacent modified
nucleosides appear to impart an increase in duplex stability
compared to the other modified oligonucleotides, likely due to
favorable (albeit weak) interactions of the thiophene moieties.
To confirm proper double helix formation of modified
oligonucleotides 3·6, 4·6, and 5·7, circular dichroism spectros-
copy was used to examine their secondary structure. All
modified oligonucleotides displayed spectral characteristics of a
B-form DNA helix (Figure 4a). Of note is the additional
Cotton effect seen in duplex 5·7 centered around 320 nm, near
the red-shifted band in the absorption spectrum (Figure 4b).
This feature, which is much less apparent in the spectrum of
duplex 4·6 where the fluorescent nucleosides are separated,
may further indicate an alignment of the adjacent thiophene
moieties.
The fluorescence emission spectra of each oligonucleotide as
a single strand and when hybridized to its perfect complement
show distinct photophysical behavior (Figure 3b). All
oligonucleotides display an emission maximum near 438 nm,
with no significant wavelength shift compared to the emission
of nucleoside 1 in water. The fluorescence emission of singly
modified oligonucleotide 3 is slightly quenched by duplex
formation (3·6), a phenomenon commonly observed with
fluorescent nucleoside analogues.2
a
Reagents and conditions: (a) 4,4′-dimethoxytrityl chloride, TEA,
pyridine, 85%; (b) (iPr2N)2POCH2CH2CN, 1H-tetrazole, MeCN,
63%; (c) standard solid phase synthesis.
Each 17-mer sequence includes a single incorporation or
three incorporations of the fluorescent nucleoside. The highest
possible sequence homology was maintained across the
oligonucleotide series. Oligonucleotide 3 contains a single
incorporation of 1 in a central position and serves as a reference
for emission intensity and behavior of a single fluorescent
nucleotide within a medium length oligomer. Oligonucleotide 4
contains three incorporations of 1 in central positions, each
alternating with dA residues. Finally, oligonucleotide 5 contains
three incorporations of 1 in adjacent positions. All sequences
contain short dG- and dC-rich stretches near their termini to
maintain adequate thermodynamic stability of the duplexes
formed with their perfect complements (6 and 7).
To investigate the basic photophysical characteristics of 1, its
absorption and emission spectra were recorded in water,
dioxane, and mixtures thereof (Figure 1). Both the absorption
Interestingly, as a single-strand, modified oligonucleotide 4,
with three alternating incorporations of nucleoside 1, has nearly
identical fluorescence intensity as the singly modified
oligonucleotide 3. Upon hybridization with its native perfect
complement (4·6), the fluorescence emission dramatically
increases. Both sequences in 3 and 4 have adenine residues
surrounding the modified nucleoside. However, upon hybrid-
ization, the fluorescence emission behaves in opposite fashion
for each strand (3·6 vs 4·6). This behavior suggests that
chromophore−chromophore self-quenching interactions in the
single strand 4, facilitated by the relatively high conformational
flexibility of the single stranded oligomer, are disrupted upon
formation of a highly ordered duplex, where the chromophores
are forced to be spatially separated.
In stark contrast, oligonucleotide 5, with three adjacent
incorporations of nucleoside 1, has significantly lower
fluorescence emission in the single strand. This low intensity
is diminished even further upon hybridization (5·7), which
suggests an even more effective self-quenching mechanism
between the adjacent modified nucleosides (and perhaps their
neighbors). While the underlying mechanism responsible for
the observed quenching in 4 and 5 is likely to be the same, their
distinct behavior upon duplex formation reflects the effective
Figure 1. Absorption (dashed) and emission (solid) spectra of 1 in
water (blue), dioxane (red), and mixtures thereof (gray).
and emission spectra of 1 respond to changes in polarity. Going
from dioxane to water the absorption maximum undergoes a
hypsochromic shift from 320 to 314 nm, while the emission
maximum shifts bathochromically from 416 to 439 nm. A
nearly linear relationship between the sample’s ET(30) value, a
microenvironmental polarity parameter,20 and the Stokes shift
(vabs−vem) is obtained (Figure 2a).21 Interestingly, marked
nonlinear changes in intensity are observed in the emission
spectra. Fluorophores like 1, consisting of a molecular rotor
element, two π-systems separated by a single bond, are known
B
dx.doi.org/10.1021/jo4008964 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
Figure 2. (a) Stokes shift vs sample ET(30) values with a linear fit (orange line, slope = 0.197, Y intercept = 13.9, R2 = 0.899). (b) Fluorescence
intensity at 440 nm (solid orange circles) and the change in sample viscosity (solid blue circles) as a function of the volume fraction of water in
dioxane.
Figure 3. (a) Thermal denaturation curves (solid circles) and (b) fluorescence spectra (solid lines) of duplexes 3·6 (blue), 4·6 (red), and 5·7 (green)
and single strands 3, 4, and 5 (dashed lines). Samples consist of 5 μM ssDNA or dsDNA in 10 mM phosphate buffer containing 100 mM NaCl at
pH 7. Fluorimeter settings: λex = 332 and 5 nm excitation and emission slit widths.
fluorescence spectra of the two systems were taken between
90 and 18 °C with 4 °C intervals (Figure 5a and b).
Corroborating our room temperature findings (Figure 3b),
annealing 4 to 6, by gradually lowering the temperature, results
Figure 4. (a) CD spectra and (b) absorption spectra of 3·6 (blue), 4·6
(red), 5·7 (green), and CD spectra of native control duplexes (dashed
lines). Samples consist of 25 μM dsDNA in 10 mM phosphate buffer
containing 100 mM NaCl at pH 7.
Figure 5. Temperature dependent fluorescence spectra of duplex (blue
solid lines) and single strand (orange dashed lines) for (a) 4·6 and 4,
chromophore−chromophore separation enforced by hybrid-
(b) 5·7 and 5; and average fluorescence intensity at 433 nm vs
ization, which is apparent in 4·6.
temperature for duplex (solid blue circles), single strand (orange
Recognizing the sensitivity of nucleoside 1 to both
circles), and the difference (solid black circles) with piecewise linear
environmental and local dynamic effects, we next investigated
fits (gray lines) for (c) 4·6 and 4, (d) 5·7 and 5. Samples consist of 5
the impact of temperature on the photophysical features of the
two distinctly behaving multichromophoric DNA duplexes: 4 vs
its duplex 4·6, and 5 vs 5·7. Temperature dependent
μM ssDNA or dsDNA in 10 mM phosphate buffer containing 100
mM NaCl at pH 7. Fluorimeter settings: λex = 332 and 3.6 nm
excitation and emission slit widths.
C
dx.doi.org/10.1021/jo4008964 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
in an intensification of the fluorescence emission. Likewise,
annealing 5 to 7 results in an intensification, albeit more
modest, of the fluorescence response. The latter finding
challenges the observations made at room temperature and
suggests temperature effects are partly responsible for the
observed differences (Figure 3b).
complexity and multitude of effects that may impact the overall
temperature-dependent fluorescence changes, careful data
analysis allows for estimation of the melting temperature that
correlates well with classically determined Tm values.
The studies illustrated above successfully demonstrate the
utility of multiply modified oligonucleotides in two model
systems. The enhanced fluorescence signal obtained from these
oligonucleotides may prove useful in biophysical assays. Instead
of adding a large, structurally perturbing fluorophore to gain
brightness, several modified isomorphic analogues may be
employed. Not only do these nucleosides allow for more
biologically accurate secondary structures, they can be easily
prepared by standard solid-phase oligonucleotide synthesizers
and, in certain cases, by enzymatic means.
The influence of temperature, in addition to the denatura-
tion/annealing effect, on the fluorescence signal may be
accounted for by measuring the modified single strand in a
separate, simplified “control” experiment. Single strand 4
fluoresces with higher efficiency upon lowering the temper-
ature, but to a much smaller extent than duplex 4·6 (Figure 5a).
Single strand 5, however, shows a significantly larger increase
upon cooling than the intensification in fluorescence observed
for the duplex 5·7 (Figure 5b). The fluorescence intensity at
433 nm was plotted as a function of temperature illustrating a
stark contrast in the degree of fluorescence intensification upon
annealing for each modified oligonucleotide (Figure 5c and d).
Both graphs suggest that the inception of the growing deviation
in fluorescence intensity of single strand and duplex
corresponds closely to the melting transition temperature
determined by monitoring absorbance at 260 nm (Figure 3a).
To correct, at least partially, for temperature effects on the
chromophores’ emission, the difference between the fluores-
cence of the single strand and the corresponding duplex were
plotted (Figure 5c and d). The inflection of this “difference
plot” underlines that the stark changes in fluorescence signal
coincide with the melting of the duplex. Recognizing the two
possibly linear components in the difference plot, a piecewise
linear fit analysis was performed (Figure 5c and d). The
intersection of the two linear segments for the difference plot of
duplex 4·6, and single strand 4 is 55.5 °C, which approximates
the 57 °C melting transition temperature (Tm) observed
(Figure 3a). The 5·7 duplex is characterized by a Tm of 61 °C
(Figure 3a), which correlates nicely to 64.1 °C, the intersection
of the two linear segments that make up the difference plot
(Figure 5d).
To put these observations in perspective, one recognizes that
both increases and decreases in fluorescence intensity may be
effectively used as signals to monitor hybridization, conforma-
tional changes and ligand binding events.27 As systems become
increasingly complex, however, diminished fluorescence signals
are a less reliable outcome, since fluorescence quenching might
be caused by any number of molecules and phenomena leading
to incorrect interpretation. In contrast, increases in fluorescence
emission, referred to as an “on-signal”, are rarely observed
because of such unrelated phenomena.
As seen in Figure 5a, upon hybridization of 4, duplex 4·6
exhibits a large increase in fluorescence intensity. This increase
in fluorescence can be used to follow duplex formation in real
time. Additionally, at room temperature, the fluorescence
intensity of the single strand 5 is significantly higher than
duplex 5·7, allowing for monitoring of duplex denaturation
(Figure 5b).28
EXPERIMENTAL SECTION
■
General Procedures. Reagents were purchased and used without
further purification unless otherwise specified. Anhydrous N,N-
dimethylformamide was obtained using a two-column purification
system. NMR solvents were purchased from Cambridge Isotope
Laboratories (Andover, MA). Reactions were monitored with
analytical thin-layer chromatography (TLC) performed on precoated
silica gel aluminum-backed plates. All experiments involving air and/or
moisture sensitive compounds were carried out under an argon
atmosphere. Column chromatography was performed with silica gel
particle size 40−63 μm. All absorption measurements were obtained at
21 °C using a quartz cuvette with a 1.0 cm path length. Steady state
fluorescence measurements were obtained using a quartz fluorescence
cell with a 1.0 cm path length. All CD spectra were recorded in a 350
μL quartz cell with a path length of 0.1 cm at 25 °C.
Synthetic Procedures. 5-(Thien-2-yl)-3′-(2-cyanoethyldiisopro-
pylphosphoramidite)-5′-O-(4,4′-dimethoxytrityl)-2′-deoxythymi-
dine (2). (a). 5-(Thien-2-yl)-5′-O-(4,4′-dimethoxytrityl)-2′-deoxythy-
midine. This compound was synthesized and purified by reported
procedures.6,19,29
(b). 5-(Thien-2-yl)-3′-(2-cyanoethyldiisopropylphosphoramidite)-
5′-O-(4,4′-dimethoxytrityl)-2′-deoxythymidine (2). DMT-protected
1 (727 mg, 1.19 mmol) was dissolved in anhydrous acetonitrile (21.2
mL) and stirred at room temperature. To this mixture, 2-cyanoethyl
N,N,N′,N′-tetraisopropylphospordiamidite (415 μL, 1.31 mmol) was
added via syringe and allowed to stir under argon. To this mixture, 1H-
tetrazole (2.64 mL, 0.45 M in acetonitrile, 1.19 mmol) was added
dropwise via syringe over thirty minutes. The reaction mixture was
stirred at room temperature for 4 h, monitoring closely by TLC
(DCM/MeOH 9/1). The reaction was quenched with a few drops of
methanol and dried to a white foam under reduced pressure. The
product was purified by flash silica column chromatography (98/1/1
chloroform/MeOH/triethylamine) to produce a mixture of diaster-
1
eomers as a bright white foam (610 mg, 0.75 mmol, 63% yield): H
NMR (CDCl3, 400 MHz) δ 7.88 (s, 1H), 7.40−7.38 (d, J = 7.9 Hz,
2H), 7.28−7.18 (m, 7H), 7.15−7.13 (m, 1H), 6.85−6.84 (m, 1H),
6.77−6.75 (m, 1H), 6.77−6.75 (d, J = 8.9, 4H), 6.71−6.69 (m, 1H),
6.41−6.38 (m, 1H), 4.63−4.58 (m, 1H), 4.25−4.21 (m, 1H), 3.86−
3.79 (m, 1H), 3.74 (s, 6H), 3.61−3.52 (m, 2H), 3.43−3.40 (m, 1H),
3.34−3.30 (m, 1H), 2.80−2.75 (m, 1H), 2.64−2.61 (t, J = 6.3, 2H),
2.32−2.25 (m, 1H), 1.28−1.25 (m, 1H), 1.18−1.16 (d, J = 6.8 Hz,
6H), 1.08−1.06 (d, J = 6.8 Hz, 6H); 13C NMR (CDCl3, 100 MHz) δ
161.5, 158.8, 149.7, 144.5, 135.6, 134.6, 133.2, 130.3, 128.4, 128.2,
127.3, 125.9, 124.6, 117.6, 113.4, 110.7, 94.5, 87.0, 86.0, 85.5, 63.3,
58.3, 55.4, 43.6, 40.4, 24.8, 20.4; 31P NMR (CDCl3, 162 MHz) δ
150.0, 149.6; HRMS [M + Na]+ calculated for C43H49N4O8PSNa+
835.2901, found 835.2911.
Oligonucleotide Synthesis and Purification. All modified and
native oligonucleotides were synthesized on a DNA synthesizer at 1.0
μmol scale (500 Å CPG). Phosphoramidite 2 was dissolved in
anhydrous acetonitrile at a concentration of 100 mg/mL and placed
directly onto a port on the DNA synthesizer. All standard
phosphoramidites (5′-dimethoxytrityl-N-benzoyl-2′-deoxyadenosine-
We further note that determination of the melting transition
temperature (Tm) of a duplex using the fluorescence signal of a
singly incorporated fluorescent nucleoside has been previously
reported.6 Therein the Tm curve obtained by fluorescence
spectroscopy was, like the absorption derived curves, sigmoidal
in shape with closely matching values for the Tm. Herein, using
a far more complex system with triply incorporated fluorescent
nucleosides analogues in two different contexts, we do not
observe sigmoidal melting curves. Even with the greater
D
dx.doi.org/10.1021/jo4008964 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
3′-[(2-cyanoethyl)-(N,N-diisopropyl)] phosphoramidite, 5′-dimethox-
ytrityl-N-benzoyl-2′-deoxycytidine-3′-[(2-cyanoethyl)-(N,N-diisoprop-
yl)] phosphoramidite, 5′-dimethoxytrityl-N-isobutyryl-2′-deoxyguano-
sine-3′-[(2-cyanoethyl)-(N,N-diisopropyl)] phosphoramidite, and 5′-
dimethoxytrityl-2′-deoxythymidine-3′-[(2-cyanoethyl)-(N,N-diiso-
propyl)] phosphoramidite) were purchased from Glen Research
(Sterling, VA) and dissolved in anhydrous acetonitrile to prepare
solutions at 25 mg/mL. The coupling time for phosphoramidite 2 was
increased to 253 s, and all other couplings were performed via
standard, trityl-off procedures, utilizing standard capping, coupling,
oxidizing, and deprotection solutions.
Columns containing CPG-oligonucleotides were removed from the
DNA synthesizer, and residual solvent was removed under reduced
pressure. The plastic columns containing the solid CPG support were
cut open and the contents were placed into 5 mL reaction vessels. To
each vial, 3 mL of 30% aqueous ammonium hydroxide was added, and
the vessel was sealed at 55 °C overnight. After allowing the reaction
vessels to cool to room temperature, excess ammonia was removed
under a steady stream of argon gas. The remaining water was
lyophilized, and the samples were resuspended in 150 μL or less of 7
M urea in 1× TBE buffer. Samples were loaded onto a preheated 20%
polyacrilamide gel and run at roughly 300 mA for 6 h. Resulting UV-
active bands were excised from the gel, and the oligonucleotides were
extracted into 4.0 mL of 0.5 M NaOAc buffer (pH 7.4) at room
temperature overnight. Samples were desalted by loading oligonucleo-
tides onto C-18 Sep-Pak columns, rinsing with water, and eluting with
40% acetonitrile in water. Samples containing oligonucleotides
(determined by UV−vis profile) were lyophilized and dissolved in 1
mL of deionized water. Oligonucleotides were quantified on the basis
of absorption at 260 nm and the calculated (nearest neighbor method)
molar extinction coefficient (http://www.ambion.com/techlib/misc/
oligo_calculator.html). MALDI-TOF spectra were obtained in 3-
hydroxypiccolinic acid matrix with ammonium citrate buffer, after
mixing with ion-exchange resin using a 25mer DNA standard for
calibration. See Table 1 for all calculated and found masses.
cuvette was used. Sample concentrations where used to arrive at an
O.D. ∼ 0.1 at λex. Spectroscopy samples were prepared from a
concentrated DMSO stock-solution; hence, all sample solutions
contain 0.4 vol % DMSO. The polarity dependent steady state
fluorescence studies were performed using an excitation wavelength of
314 nm (slit-widths: 6 nm), using sample concentrations of 10 mM.
Both a Horiba Fluoromax 3 spectrofluorometer and a PTI
luminescence spectrometer were used, and parameters are listed in
the figure captions. Stokes shifts, in cm−1, were calculated after
correction of the emission intensity according the following equation:
intensity [v] = λ2 × intensity [λ].27 The cm−1 units were converted to
kcal/mol by multiplication with 0.0028591 to plot the polarity
sensitivity plot. The study was performed in duplicate, average values
are plotted, and the error bars represent the standard error of mean as
calculated by OriginPro.
Sample ET(30) values (in kcal/mol) were determined by dissolving
a small amount of Reichardt’s dye in the solvent (mixture) used to
prepare the sample.20 The observed long-wavelength absorption
maximum in nm (λabsmax) was converted to the sample ET(30) value
according the following equation:
28591
ET(30) =
λabsmax
Because of the solubility limitations of Reichardt’s dye, the literature
ET(30) value for pure water was used.20
Viscosity Correlation. Viscosity values for mole fractions water in
dioxane mixtures are reported in centipoise (cp).26 A total of 28 points
were taken from the available data set and graphed in a plot of
viscosity as a function of mole fraction water in dioxane. A polynomial
fit of the selected data was performed using OriginPro to establish a
mathematical relationship that allows for calculated interpolation
(Figure 6). Subsequently, the fit was used to calculate the viscosity of
the sample after conversion of volume to mole fractions.
Table 1. MALDI-TOF Determined Masses for All
Synthesized Oligonucleotides
oligonucleotide
calculated mass
found mass
3
4
5
6
7
8
9
5327.5
5463.7
5454.7
5121.4
5130.4
5121.4
5130.4
5327.8
5464.3
5453.4
5121.7
5130.3
5121.7
5130.3
Figure 6. Viscosity as a function of mole fraction water in dioxane
(black circles). A polynomial fit (orange line, R2 = 0.997) allows for
interpolation.
Oligonucleotide Hybridizations. All oligonucleotide samples
were prepared in 5 μM solutions of 100 mM NaCl in 10 mM sodium
phosphate buffer (pH 7.0). Samples were heated to 95 °C for 5 min
and cooled to room temperature over 2−3 h. Samples were then
placed at 0 °C until spectroscopic measurements were obtained. UV−
vis spectra were obtained on an absorption spectrometer using a 125
μL quartz cuvette with a 1.0 cm path length at 21 °C. Steady-state
fluorescence spectra were obtained using a 125 μL quartz fluorescence
cell with a 1.0 cm path length at 21 °C with slit widths of 5 nm (unless
otherwise noted). All samples were excited at a wavelength of 332 nm.
Thermal denaturation curves were obtained with a high performance
temperature controller and a micro-auto-six-sample holder. Samples
were heated from room temperature to 80 °C at a rate of 0.5 °C min−1
with optical monitoring every minute at 260 nm. Software provided on
the spectrometer determined the first derivative from the melting
profile to calculate the inflection point/Tm (estimated error 0.5 °C).
For CD spectra, samples were prepared to contain 25 μM of the
duplex in 100 mM NaCl, 10 mM sodium phosphate buffer (pH 7).
Photophysics in Water, Dioxane, and Mixtures Thereof.
Spectroscopic grade dimethylsulfoxide (DMSO) and dioxane were
used. For all spectroscopic measurements a 1 cm four-sided quartz
Temperature-Dependent Steady State Fluorescence Spec-
troscopy. Single strand and duplex samples were prepared as 5 mM
solutions in 10 mM phosphate buffer containing 100 mM NaCl at pH
7. Aliquots of 500 mL were transferred to a 600 mL fluorescence cell
and capped with a Teflon stopper. Prior to the temperature study,
samples were heated for 5 min at 90 °C in the cuvette holder followed
by the recording of the fluorescence after excitation at 332 nm using
slit widths of 3.6 nm (for 4·6 and 4) or 6.0 nm (for 5·7 and 5).
Subsequently, the temperature was lowered to 18 °C with 4 °C
intervals. The fluorescence was measured at every temperature after a
2 min settle time. All samples are measured in triplicate. Averaged
values for the fluorescence intensity at 433 nm and their standard error
of mean are calculated using OriginPro 9.0. The difference plot is fit to
two linear functions using the “Piecewise Linear” function in
OriginPro.30 The intersection of the piecewise linear fits (gray lines)
are 55.5 °C (R2 = 0.95647), and 64.1 °C (R2 = 0.99875) for Figure 5c
and d, respectively.
E
dx.doi.org/10.1021/jo4008964 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
(25) Numerical Data and Functional Relationships in Science and
Technology. Group IV; Wohlfahrt, C., Ed.; Landolt-Bornstein, Springer-
Verlag: Berlin, Heidelberg, 1991; Vol. 6 Static and Dielectric
Constants of Pure Liquids and Binary Liquid Mixtures.
(26) Ouerfelli, N.; Iulian, O.; Bouaziz, M. Phys. Chem. Liq. 2010, 48,
488−513.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional fluorescence data including thermal melting data
and corresponding plots, and NMR spectra of compound 2.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(27) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.;
Springer: New York, 2006.
(28) Multiple quenching mechanisms might be operative, depending
on the specific oligonucleotide and its sequence context. A static
quenching model can be envisioned, particularly for multiply modified
oligonucleotides and their duplexes, because of the close proximity of
the chromophores. Note, however, that distinguishing between various
quenching mechanisms in such oligonucleotides is complex, since they
are intramolecular in nature. As a result, the photophysical behavior of
the oligonucleotides is unlikely to depend on concentration (at least in
relatively dilute solutions). An additional complicating factor is the
susceptibility of the photphysical characteristics of these chromo-
phores to multiple environmental effects (polarity, viscosity, temper-
ature, etc.).
(29) Emissive thiophene-modified purines have been reported; see:
(a) Kimoto, M.; Mitsui, T.; Harada, Y.; Sato, A.; Yokoyama, S.; Hirao,
I. Nucleic Acids Res. 2007, 35, 5360−5369. (b) Yamashige, R.; Kimoto,
M.; Takezawa, Y.; Sato, A.; Mitsui, T.; Yokoyama, S.; Hirao, I. Nucleic
Acids Res. 2012, 40, 2793−2806.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ytor@ucsd.edu. Fax: +1 858 534 0202. Tel: +1 858
534 6401.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Institutes of Health for their generous
support (GM 069773) and the NSF (Instrumentation Grant
CHE-0741968).
REFERENCES
■
(1) Daniels, M.; Hauswirth, W. Science 1971, 171, 675−677.
(2) Sinkeldam, R. W.; Greco, N. J.; Tor, Y. Chem. Rev. 2010, 110,
2579−2619.
(3) Wilhelmsson, L. M. Q. Rev. Biophys. 2010, 43, 159−183.
(4) Kaul, M.; Barbieri, C. M.; Pilch, D. S. J. Am. Chem. Soc. 2004, 126,
3447−3453.
(30) OriginPro 9.0.0, Copyright 1991−2013 OriginLab Corpo-
ration.http://wiki.originlab.com/∼originla/howto/index.php?title=
Tutorial:Fitting_with_a_Piecewise_Linear_Function (accessed 03/
18/2013).
(5) Shandrick, S.; Zhao, Q.; Han, Q.; Ayida, B. K.; Takahashi, M.;
Winters, G. C.; Simonsen, K. B.; Vourloumis, D.; Hermann, T. Angew.
Chem., Int. Ed. 2004, 43, 3177−3182.
(6) Greco, N. J.; Tor, Y. J. Am. Chem. Soc. 2005, 127, 10784−10785.
(7) Greco, N. J.; Sinkeldam, R. W.; Tor, Y. Org. Lett. 2009, 11, 1115−
1118.
(8) Xie, Y.; Maxson, T.; Tor, Y. Org. Biomol. Chem. 2010, 8, 5053−
5055.
(9) Srivatsan, S. G.; Tor, Y. J. Am. Chem. Soc. 2007, 129, 2044−2053.
(10) Xie, Y.; Dix, A. V.; Tor, Y. J. Am. Chem. Soc. 2009, 131, 17605−
17614.
(11) Teo, Y. N.; Wilson, J. N.; Kool, E. T. J. Am. Chem. Soc. 2009,
131, 3923−3933.
(12) Mayer-Enthart, E.; Wagenknecht, H. A. Angew. Chem., Int. Ed.
2006, 45, 3372−3375.
(13) Rist, M.; Wagenknecht, H. A.; Fiebig, T. ChemPhysChem 2002,
3, 704−707.
(14) Srivatsan, S. G.; Tor, Y. J. Am. Chem. Soc. 2007, 129, 2044−
2053.
(15) Srivatsan, S. G.; Greco, N. J.; Tor, Y. Angew. Chem. 2008, 120,
6763−6767; Angew. Chem., Int. Ed. 2008, 47, 6661−6665.
(16) Shin, D.; Sinkeldam, R. W.; Tor, Y. J. Am. Chem. Soc. 2011, 133,
14912−14915.
(17) Sinkeldam, R. W.; Wheat, A. J.; Boyaci, H.; Tor, Y.
ChemPhysChem 2011, 12, 567−570.
(18) Sinkeldam, R. W.; Greco, N. J.; Tor, Y. ChemBioChem 2008, 9,
706−709.
(19) (a) Wigerinck, P.; Pannecouque, C.; Snoeck, R.; Claes, P.; De
Clercq, E.; Herdewijn, P. J. Med. Chem. 1991, 34, 2383−2389.
(b) Gutierrez, A. J.; Terhorst, T. J.; Matteucci, M. D.; Froehler, B. C. J.
Am. Chem. Soc. 1994, 116, 5540−5544.
(20) Reichardt, C. Chem. Rev. 1994, 94, 2319−2358.
(21) Sinkeldam, R. W.; Tor, Y. Org. Biomol. Chem. 2007, 5, 2523−
2528.
(22) Haidekker, M. A.; Theodorakis, E. A. Org. Biomol. Chem. 2007,
5, 1669−1678.
(23) Foerster, T.; Hoffmann, G. Z. Phys. Chem. 1971, 75, 63−76.
(24) Critchfield, F. E.; Gibson, J. A.; Hall, J. L. J. Am. Chem. Soc.
1953, 75, 1991−1992.
F
dx.doi.org/10.1021/jo4008964 | J. Org. Chem. XXXX, XXX, XXX−XXX