Quantitative DNA interstrand cross-link formation by coumarin and thymine: Structure determination, sequence effect, and fluorescence detection

Article abstract of DOI:10.1021/jo5014756

The coumarin analogues have been widely utilized in medicine, biology, biochemistry, and material sciences. Here, we report a detailed study on the reactivity of coumarins toward DNA. A series of coumarin analogues were synthesized and incorporated into o

Full text of DOI:10.1021/jo5014756

Article
pubs.acs.org/joc
Quantitative DNA Interstrand Cross-Link Formation by Coumarin and
Thymine: Structure Determination, Sequence Effect, and
Fluorescence Detection
Huabing Sun, Heli Fan, and Xiaohua Peng*
Department of Chemistry and Biochemistry, University of Wisconsin-Milwaukee, 3210 North Cramer Street, Milwaukee, Wisconsin
53211, United States
*
S Supporting Information
ABSTRACT: The coumarin analogues have been widely utilized in medicine, biology, biochemistry, and material sciences. Here,
we report a detailed study on the reactivity of coumarins toward DNA. A series of coumarin analogues were synthesized and
incorporated into oligodeoxynucleotides. A photoinduced [2 + 2] cycloaddition occurs between the coumarin moiety and the
thymidine upon 350 nm irradiation forming both syn- and anti-cyclobutane adducts (17 and 18), which are photoreversible by
2
54/350 nm irradiation in DNA. Quantitative DNA interstrand cross-link (ICL) formation was observed with the coumarin
moieties containing a flexible two-carbon or longer chain. DNA cross-linking by coumarins shows a kinetic preference when
flanked by an A:T base pair as opposed to a G:C pair. An efficient photoinduced electron transfer between coumarin and dG
slows down ICL formation. ICL formation quenches the fluorescence of coumarin, which, for the first time, enables fast, easy,
and real-time monitoring of DNA cross-linking and photoreversibility via fluorescence spectroscopy. It can be used to detect the
transversion mutation between pyrimidines and purines. Overall, this work provides new insights into the biochemical properties
and possible toxicity of coumarins. A quantitative, fluorescence-detectable, and photoswitchable DNA cross-linking reaction of
the coumarin moieties can potentially serve as mechanistic probes and tools for bioresearch without disrupting native biological
environment.
INTRODUCTION
Many photoactive molecules allowing reversible photo-
chemical reactions are widely used in polymer sciences, such
■
DNA interstrand cross-links (ICLs) inhibit DNA replication
1
7,18
as light switchable coatings and drug delivery.
Photoactive
and gene expression by preventing strand separation. Cross1-
molecules have also been studied for reversible photo-cross-
linking agents have been widely applied in cancer therapy,
19−25
2
−6
linking and photomanipulation of DNA.
For instance,
study of DNA repair and gene regulation,
protein−DNA
and DNA nanotechnology. Most DNA cross-
linking processes are irreversible. Reversible reactions provide
7
−9
10
psoralens, a class of naturally occurring photoreactive products,
form interstrand cross-links (ICLs) in duplex and triplex DNA,
which have been utilized for DNA damage and repair study, for
therapeutic gene modulation, and as photoactive probes of
binding,
11−16
opportunities for the development of new applications.
For example, the reversible alkylation reactions induced by
anticancer drug ecteinascidin 743 allow migration of the
alkylating sites to achieve sequence selectivity and anticancer
2
0−23,26,27
nucleic acid structure and function.
Coumarin
molecules, with similar structure scaffold as psoralens, are
capable of photoreversible dimerization via a photoinduced [2
1
1
activity. The reversibility of DNA-drug adducts was also
correlated with the cytotoxicity of antitumor drugs based on
+
2] cycloaddition reaction and subsequent photocleavage at a
28
12
different wavelength. They have wide polymeric applications,
cyclopropylpyrroloindole. Most recently, Rokita and co-
workers discovered the reversible nature of quinone methide
which offers a novel strategy for cell-compatible target selective
alkylation that allows for strand exchange and cross-link
such as electroluminescence studies, light and energy harvest-
Received: July 2, 2014
13−16
migration leading to dynamic cross-linking.
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Synthesis of the Phosphoramidites 8−13
ing, construction of liquid crystalline polymers and photoactive
surfaces, and photoreversible polymerization, chain extension,
photoswitchable DNA ICL formation via coumarin moieties.
Most traditional methods for DNA ICL analysis require
radiolabeling with P, C, or H which are considered to be
health and safety hazards. In addition, they are offline methods
and are time-consuming. Our methodology is capable of a safe
real-time detection of ICL formation, which will undoubtedly
facilitate its bioapplication.
2
8
32
14
3
and cleavage. In addition to polymer science, coumarin
analogues are also used in the field of biological science as
medicines or as fluorescent tags and fluoroprobes for labeling
2
8,29
protein and nucleic acids.
However, to the best of our
knowledge, there is no detailed information on the photo-
reactivity of coumarin with biomolecules such as DNA, RNA,
and proteins.
RESULTS AND DISCUSSION
■
Recently, we have shown that a coumarin-modified
thymidine is capable of cross-linking dT upon 350 nm
Design and Synthesis of the Cross-Linking Precursor
and Their Incorporation into Oligodeoxynucleotides.
We designed and synthesized seven coumarin analogues 1−7
with various modifications at position 4 (2) or 7 (3−7), which
allowed us to study the reaction sites and linker effects on the
cross-linking efficiency (Scheme 1). Compounds 2−7 contain a
hydroxyl group which can be converted to the corresponding
phosphoramidites 8−13 used for solid-phase DNA synthesis.
Compounds 2 and 3 were designed to study the effect of the
phosphorus position on the cross-linking efficiency, while 3−6
were used to investigate the effect of the linker length and 7
allows for flexible incorporation into DNA at any positions.
Compound 1 or 2 was prepared via methylation of 7-hydroxy-
30
irradiation with about 69% cross-linking yield. Such a process
is photoswitchable and leads to fluorescence variation.
However, the cross-linking efficiency by a coumarin-modified
thymidine is too low to allow online fluorescence detection of
DNA cross-linking. In addition, coumarins with suitable linkers
can be directly incorporated into DNA at any position instead
of attaching to thymidine in our previous study, which could
avoid the intramolecular reaction between coumarin and
thymidine. We envisage that the photochemistry of coumarin
or coumarin-containing ODN would enable a method for real-
time monitoring of the cross-linking process via fluorescence
spectroscopy if quantitative DNA cross-linking can be achieved,
which is indispensable for direct measurement of the cross-link
formation and expanding its bioapplication. These predictions
have been experimentally borne out by detailed investigations
on the photoactivity of coumarin-derived compounds toward
DNA. However, the chemistry is more complex than we expect.
Here, we report the structure of coumarin-dT dimers and a
quantitative, fluorescence-detectable, sequence-dependent, and
2
1
H-chromen-2-one derivative 3 or 14, respectively (Scheme
B). Three different linker units were introduced to compound
3 via Williamson ether synthesis yielding 4, 5, and 7 (Scheme
1C and E). Compound 6 was synthesized by “click” reaction
between 3-azidopropan-1-ol and 7-ethynyl-4-methyl-2H-chro-
men-2-one (15) (Scheme 1D). All coumarin monomers with
the hydroxyl group (2−7) were converted to the corresponding
phosphoramidites (8−13) using standard methods (Scheme
B
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 2. Double-Stranded DNAs Used in This Study and Their Melting Temperatures
3
0
1
A−E). All new compounds and phosphoramidites were
NMR analysis. Here, we studied the reactivity of coumarin
moiety with canonical nucleosides dA, dG, dT, and dC using
compound 1 as a representative. Initially, LC-MS was used to
monitor the reaction. A new product with a mass of 455.4
(corresponding to syn- and anti-cyclobutane 1-dT dimers 17
and 18) was detected in the presence of dT after 350 nm
irradiation for 2 days, while no new products were formed
between 1 and dC, dA, or dG (SI, Figure S31a−d). In addition
to 17 and 18, a coumarin dimer 19 was also isolated. All
confirmed by NMR and HRMS (Supporting Information
(SI), Figures S23a−30d).
ODNs containing 2−7 were prepared using commercially
available β-cyanoethyl phosphoramidites containing phenox-
yacetyl protecting groups on the exocyclic amines of dA and
dG, which allowed a very mild deprotection condition avoiding
the decomposition of the functionalized ODNs (2b−10b, 11,
and 13b) (Scheme 2). All ODNs were purified by 20%
denaturing polyacrylamide gel electrophoresis (PAGE) and
characterized by MALDI-TOF-MS (SI, Table S2 and Figures
S35−45).
1
13
compounds were characterized by H NMR, C NMR, and
HRMS (SI, Figures S32a−34c). They were formed by a [2 + 2]
cycloaddition reaction. Two isomers 17 and 18 were formed
between 1 and dT, while syn-adduct 17 was the predominant
product (yield: 15% for 17 and 9% for 18) (Scheme 3). The
final structure of 17 and 18 was determined by two-
dimensional nuclear magnetic resonance spectroscopy (H−H
COSY). Formation of 17−19 quenches the fluorescence of
coumarin due to the disruption of conjugated π-system (Figure
1).
After successfully synthesizing the ODNs containing
coumarin moieties, we studied the effect of coumarin on
duplex thermal stability by measuring the UV-melting temper-
atures (T ) (Scheme 2). The coumarin moieties increased the
M
T (1.1−3.1 °C) relative to when 2′-deoxyadenosine (dA) was
M
present, possibly due to the increased π−π stacking and/or
31−33
increased lipophilicity.
The T increase was more obvious
M
when coumarin was incorporated into ODNs via position 4 (ds
Effect of Linker Position and Length on DNA Cross-
Linking. The monomer reactions indicated that the coumarin
moiety can react with dT. To study its reactivity in DNA, we
designed six double stranded (ds) DNAs (1−6) and
investigated the DNA interstrand cross-linking efficiency
(Scheme 2). The ds DNA-2−6 contain a coumarin moiety
opposing dT, while ds DNA-1 has an A:T base pair which is
used as a control. All DNA duplexes were photoirradiated at
350 nm. A wavelength of 350 nm was chosen because near-UV
light (>300 nm) is compatible with living cells and is almost
not absorbed by most biological molecules other than the
DNA-2, ΔT = 2.0 °C) than via position 7 (ds DNA-3, ΔT =
M
M
1.1 °C). A flexible linker unit at position 7 further enhances the
duplex stability by about 3.0 °C (ds DNA-4, ΔT = 2.9 °C; ds
M
DNA-5, ΔT = 3.1 °C). A triazole moiety with less flexibility in
M
stacking and decreased lipophilicity seems less favorable for
duplex stability (ds DNA-6) than an alkyl chain (ds DNA-5).
Monomer Reaction. Previously, we have utilized LC-MS
to analyze the cross-linked adduct formed between coumarin
and dT, while the detailed structure was not elucidated due to
the difficulty isolating the adducts in amounts sufficient for
C
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 3. Photo-Induced [2 + 2] Cycloaddition between 1
and dT
steric hindrance for the ICL formation, while compounds 3−6
may adopt a more favorable conformation for producing a
highly efficient transition state between coumarin moieties and
dT.
The coumarin reactivity in DNA can be further tuned by
altering the linker units between the phosphate group and the
coumarin moiety. For example, compound 3 without a linker
led to a relatively low ICL yield (86%, lane 3) because of the
lack of flexibility, while quantitative ICL yields were observed
with ds DNA-4 and 5 with an adaptable chain between the
phosphate group and coumarin moiety (Figure 2, lanes 4 and
5
). The rate constants for ds DNA-4 and 5 were 5−6 times the
rate constant for ds DNA-3 (Table 1 and SI, Figure S2−3).
Table 1. ICL Formation Reaction Rates for DNA Duplexes
with Different Coumarin Moieties
k (10− s−1
4
)
t1/2 (min)
entry
a
ds DNA-2
ds DNA-3
ds DNA-4
ds DNA-5
ds DNA-6
n.d.
n.d.
9.5 ± 1.6
50.4 ± 3.8
53.9 ± 4.1
9.8 ± 1.1
12.4 ± 2.0
2.3 ± 0.2
2.3 ± 0.2
11.9 ± 1.2
a
n.d.: not determined.
However, installation of a triazole ring in 6 decreases the
reactivity possibly due to an electron transfer in the excitation
state of coumarin. In summary, compounds 4 and 5 enable
highly efficient and fast DNA interstrand cross-linking with
quantitative ICL yields and a 2.3 min half-life (Table 1 and SI,
Figure S1a,b). Finally, compound 4 was chosen for further
study.
Flanking Sequence Effect on DNA Cross-Linking and
the Fluorescence Intensity of the Coumarin Moiety. In
addition to the linker units, flanking bases also have a great
effect on cross-linking. Although 4 efficiently forms DNA ICLs
when flanked by all four bases, lower ICL yields were observed
when 4 was flanked by dC (ds DNA-9: 88% vs quantitative ICL
formation for ds DNA-4: 98%, 7: 100%, or 8: 97%) (Figure 3).
Figure 1. Fluorescence emission spectra of coumarin 1 and dimers
1
7−19 (10 μM, λ = 325 nm, slit width = 15 nm; λem = 385 nm, slit
ex
width = 2.5 nm).
coumarin-modified ODNs. Irradiation of ds DNA-2−6 resulted
in a new band which migrates more slowly than unreacted
ODN, indicative of the formation of interstrand cross-linked
material (SI, Figures S46−49), while this was not observed with
native ds DNA-1 (Figure 2). A higher ICL yield was observed
with ds DNA-3 (86%, lane 3) compared to ds DNA-2 (3%, lane
2
). As cycloaddition reaction occurs at positions 3 and 4,
coumarin moiety 2 attached to ODN via position-4 may induce
Figure 3. Phosphorimage autoradiogram of denaturing PAGE analysis
of ICL formation (100 nM ds DNAs-1, 4, and 7−9 were irradiated at
32
3
50 nm for 2 h; ODNs without coumarin were 5′-[ P]-labeled).
Moreover, the rate constant for cross-link formation in ds
DNA-8 or 9, where 4/dT was flanked by a G:C base pair, was
∼
10 times slower than when 4/dT was flanked by a T:A pair
ds DNA-4 and 7) (Table 2 and SI, Figures S2b, S3b, S4, and
(
S5). ICL formation with ds DNA-7 and ds DNA-4 upon 350
nm irradiation was complete within 10 min with more than
95% cross-linking yield, while it took more than 120 min with
ds DNA-8 and 9 (SI, Figure S1b).
Figure 2. Phosphorimage autoradiogram of denaturing PAGE analysis
of ICL formation (100 nM ds DNA-1−6 was irradiated at 350 nm for
32
2
h; ODN 1a was 5′-[ P]-labeled).
D
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 2. Rate of ICL Formation or Cleavage of DNAs Containing 4 with Different Flanking Bases
(
ds DNA-4 and 7−9: 3′-dTYTACGGCGGGT ● 5′-d4XATGCCGCCCA)
k (ICL Formation, 10− s−1
4
)
t1/2 (min)
a
kc (Cleavage, 10−3 s−1
entry
X:Y
p.d.
f.d.
p.d.
f.d.
)
t1/2 (min)
ds DNA-7
ds DNA-4
ds DNA-8
ds DNA-9
A:T
T:A
G:C
C:G
42.0 ± 4.0
50.4 ± 3.8
5.8 ± 0.2
4.1 ± 0.1
31.7 ± 0.4
46.6 ± 2.7
6.8 ± 0.2
5.1 ± 1.2
2.7 ± 0.2
2.3 ± 0.2
20.0 ± 0.8
28.1 ± 1.0
3.6 ± 0.1
2.5 ± 0.2
16.9 ± 0.6
22.8 ± 5.5
4.6 ± 0.2
4.0 ± 0.5
17.1 ± 0.1
12.2 ± 1.2
2.5 ± 0.1
2.9 ± 0.4
0.7 ± 0.0
1.0 ± 0.2
ap.d.: determined by radioactive 32P-labeling method; f.d.: determined by fluorescence assay.
Figure 4. Comparison of fluorescence intensity of 4 when flanked by different nucleotides: (a) fluorescence emission spectra of 10 μM single-
stranded ODNs; (b) fluorescence emission spectra of 10 μM duplexes (measured in 100 mM NaCl and 10 mM pH 7 phosphate buffer solution with
λex = 350 nm, slit width = 10 nm; λem = 390 nm, slit width = 10 nm).
We also observed that the flanking sequences greatly affect
the fluorescence intensity of the coumarin moiety. The ODNs
Scheme 4. Mechanism of ICL Formation
(7b and 8b) with the coumarin moiety adjacent to a purine
nucleoside showed a greatly decreased fluorescence intensity in
comparison with those (4b and 9b) containing 4 flanking to a
pyrimidine nucleoside (Figure 4a). It is well-known that the
photoinduced electron transfer leads to nucleobase-specific
3
4
fluorescence quenching. We propose that the excited
coumarin 4 is able to oxidize other nucleobases due to its
large excited state reduction potentials (2.10 V determined by
cyclic voltammetry, SI, Figure S6). The sequence of the
fluorescence quenching efficiency is on the order of dG > dA >
dC ≈ dT, which is equivalent to the order of nucleobase
oxidation potentials (E : G (1.25), A (1.72), C (1.87), and T
O,CV
formation via 350 nm irradiation (Figure 5a). Almost no
fluorescence was detected when 98% ICL was produced with
3
5
34
(
1.90)). This observation is consistent with Seidel’s report.
Interestingly, hybridization of DNAs further enhances the
fluorescence of ds DNA-4 and 7 with an A:T base pair next to
1
0 μM ds DNA-4 upon 60 min irradiation at 350 nm (Figures 3
and 5a). As the ICL product is nonfluorescent, the fluorescence
intensity observed in the reaction mixture results from the
unreacted DNA duplexes. It is well-known that the fluorescence
intensity is linearly correlated with the fluorophore concen-
tration at the μM level (self-quenching could occur at high
4:T, and quenches the fluorescence of ds DNA-9 and 8 with a
G:C pair adjacent to 4:T (Figure 4b). Overall, the sequence of
the fluorescence quenching efficiency is consistent with the
order of the DNA cross-linking rate constant kds‑4, kds‑7 ≫ kds‑8,
kds‑9. The correlation between the quenching efficiency and the
ICL formation constant indicates that the photoinduced
electron transfer plays an important role in decreased ICL
formation when 4 is adjacent to a G:C pair.
Fluorescence Detection of DNA Cross-Linking For-
mation. Similar to the monomer study, no fluorescence was
observed with the DNA cross-linked products (20) possibly
due to its disrupted conjugation system (Scheme 4 and SI,
Figure S7). The fast and quantitative formation of a
nonfluorescent ICL product from a highly fluorescent coumarin
moiety encourages us to develop a novel method for detecting
DNA cross-linking using a fluorescence assay. We observed that
the fluorescence of ds DNA-4 decreased steadily upon ICL
3
6
concentration of mM level). Based on these facts, a
reasonable equation ICL yield = [I − I/I − I ]* 98% was
0
0
60
proposed, where I is the fluorescence intensity of unreacted ds
0
DNA-4 at 0 min, I is the fluorescence intensity of ds DNA-4
upon photoirradiation for the desired time, I60 is the
fluorescence intensity of ds DNA-4 upon photoirradiation for
60 min. The rate constants determined by fluorescence assay
(kcal.‑4 (4.66 ± 0.27) × 10− s ) were within the experimental
3
−1
32
error of those measured using P-labeling method ((5.04 ±
0.38) × 10− s ) (Figure 5b and Table 2). The reliability of
this method was further examined by measuring the rate
constants of ICL formation with ds DNA-7−9 using
fluorescence assay (Table 2 and SI, Figures S8−10). The data
3
−1
E
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 5. Determination of ICL formation rate constant via fluorescence assay: (a) fluorescence emission spectra of two independent samples
containing 10 μM ds DNA-4 which were irradiated at 350 nm for the desired time (λ = 325 nm, slit width = 15 nm; λ = 385 nm, slit width = 10
ex
em
nm); (b) rate of DNA cross-linking formation.
Figure 6. Photoreversible process for ds DNA-4 over three cycles of photoirradiation at 350 nm for 50 min (UV1) and 254 nm (UV2) for 15 min:
32
(
a) ICL yields obtained from phosphorimage autoradiogram of denaturing PAGE analysis of 100 nM ds DNA-4 (ODN 1a was 5′-[ P]-labeled); (b)
fluorescence intensity at 390 nm of 10 μM ds DNA-4 (λ = 325 nm, slit width = 15 nm; λ = 390 nm, slit width = 10 nm).
ex
em
obtained from the fluorescence assay are consistent with those
obtained by 32P-labeling method (Table 2). This is the first
example that allows rapid, direct, and efficient monitoring of
the DNA cross-linking process on time via a noninvasive
method instead of employing a traditionally used harmful 32P-
labeling method in biochemical research. The novel method
could be introduced as an effective way in biology for DNA
cross-linking study and allowing real-time detection without
disrupting native cell environment.
Photoreversibility. The ICL formation induced by
coumarin moiety is photoreversible. The DNA cross-linked
products generated by 350 nm irradiation were split into single-
stranded ODNs upon irradiation at 254 nm for 15 min.
Initially, we observed the decomposition of dT-coumarin
dimers (17 and 18) to coumarin 1 and thymidine during
HRMS mass measurement, which indicated that [2 + 2]
cycloaddition products were cleaved into starting materials
upon high voltage electrons. Later, we found that 17 and 18
were cleaved by 254 nm irradiation, which was confirmed by
NMR analysis (SI, Figure S11). Cleavage reaction was also
observed with DNA cross-linked products formed with ds
DNA-4 and 7−9 upon irradiation at 254 nm. The cleavage of
the ICL products followed first-order kinetics and the rate for
cleavage reaction is sequence-dependent. The G:C base pair
flanking to the cross-linking site led to a faster cleavage reaction
than an A:T base pair (ds DNA-8 and 9 vs ds DNA-4 and 7)
observed over three cycles of photo irradiation at 350 nm for 50
min and 254 nm for 15 min (Figure 6). The efficient ICL
formation (around 80% ICL yield) was still achieved after three
cycles of 350 nm/254 nm irradiation (Figure 6a and SI, Figure
S14b). The photoswitchable ICL formation can be monitored
by fluorescence spectroscopy. Irradiation of ds DNA-4 at 350
nm quenches the fluorescence due to formation of cyclobutane
product 20, while 254 nm irradiation cleaves 20 to a single-
stranded ODN which greatly enhanced the fluorescence
intensity (Figure 6b). A similar result was observed for ds
DNA-7 indicating its generality (SI, Figures S14a and S15).
Fluorescence Detection of Single Nucleotide Discrim-
ination. Different reactivities of coumarin toward canonical
nucleosides and fluorescence quenching induced by ICL
formation encouraged us to use this reaction for detecting
single nucleotide polymorphisms (SNPs) via fluorescence
assay. As purines decrease the activity and fluorescence
intensity of coumarin, we expect that a method based on
coumarin cross-linking can be developed for detecting
transversion of pyrimidines to purines and vice versa. A
probe ODN 11 containing coumarin moiety 4 and templates
ODNs 11a−c were designed for this study. The templates
contain sequences of codon 248 in exon 7 of the p53 gene,
which are often mutated in human cancers. ODN 11a contains
the wild-type genetic sequence with dT at the 3′-termius, while
ODN 11b and 11c contain a T → A or G transversion
mutation. Upon photoirradiation at 350 nm for 1 h, a much
(Table 2 and SI, Figures S12−13). This reversible behavior was
F
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
higher cross-linking yield was observed for ds DNA-11a (92%)
than ds DNA-11b (8%) and 11c (6%) (SI, Figure S16a). The
different cross-linking efficiency was also detected via
fluorescence assay. The fluorescence intensity was significantly
decreased with DNA-11a upon 350 nm irritation for 1 h, while
a much lower decrease was observed for DNA-11b and DNA-
formation and cleavage reactions were very fast and finished
with 10 min (Table 3, and SI, Figures S19a, 20a, 21a, and 22a).
Table 3. Rate of ICL Formation or Cleavage of ds DNA-10,
12, and 13
k (ICL
1
1c (SI, Figure S16b). The decreased fluorescence intensity
Formation,
k (Cleavage,
c
−2 −1
after UV irradiation for 1 h is shown in Figure 7. Obviously,
−3 −1
entry
10
s
)
t1/2 (min)
10
s
)
t1/2 (min)
ds
DNA-10
3.99 ± 0.32
6.27 ± 0.62
1.55 ± 0.14
2.91 ± 0.23
0.57 ± 0.04
1.67 ± 0.19
1.42 ± 0.03
2.02 ± 0.13
ds
DNA-12
1.86 ± 0.18
7.51 ± 0.66
0.70 ± 0.08
0.82 ± 0.02
ds
DNA-13
After having established the chemistry and the generality of
the photoswitchable DNA interstrand cross-linking by coumar-
in moieties, we investigated their reactivities toward a longer
DNA template which provides a more realistic illustration of
how these molecules might perform in cellular DNA. We
studied the reactivity of a terminal coumarin (4 in ODN 7b) as
well as an internal coumarin (7 in ODN 13b) toward dT in a
5
0-mer ODN template 12a. Almost quantitative DNA cross-
link formation was observed for 4 in ds DNA-12 (98%, Figure
, lane 5) upon 350 nm irradiation and compound 7 in ds
8
Figure 7. Decreased fluorescence intensity at 385 nm after UV
DNA-13 also lead to a reasonable ICL yield (84%, Figure 8,
lane 8). The cross-linked products formed with ds DNA-12 and
irritation at 350 nm for 1 h with 10 μM ds DNA-11a−c (λ = 325 nm,
ex
slit width = 6 nm; λ = 385 nm, slit width = 14 nm).
em
13 efficiently reverted to the single-stranded ODNs upon 254
nm irradiation (Figure 8, lanes 6 and 9). The cleavage reactions
proceeded at a faster rate compared with when a shorter ODN
template was used, such as ds DNA-10 (Table 3, and SI,
Figures S21−22). The generality and template versatility of
photoswitchable coumarin-dT cross-linking provide a powerful
and versatile tool for expanding its bioapplication.
such a reaction can be directly used to monitor a T → A or G
transversion and vice versa. In addition, ICL formation with ds
−4
−1
DNA-11a ((9.17 ± 1.01) × 10 s ) is about 8−11 times
−5 −1
faster than those for ds DNA-11b ((8.3 ± 0.9) × 10 s ) and
−
5 −1
1
1
1c ((9.7 ± 0.8) × 10 s ) (SI, Table S1 and Figures S17−
8). In this way, coumarin-containing DNAs may have the
CONCLUSION
potential to detect T↔A or G transversion via fluorescence
assay.
■
A series of coumarin analogues (2−7) with varied modifications
at the position-4 or -7 have been synthesized and successfully
incorporated into oligonucleotides. Coumarin analogues
increase the DNA duplex stability. Coumarin scaffolds undergo
photoinduced [2 + 2] cycloaddition with thymidine and form
syn- and anti-cyclobutane adducts (17 and 18) which were
confirmed by NMR as well as mass analysis. The photo-
cyclization reaction between coumarin and thymidine leads to
fast and quantitative DNA interstrand cross-link formation with
a 2.3 min half-life. A two-carbon or longer chain linker at the
position-7 of coumarin moiety (4 and 5) is optimal for
quantitative DNA cross-linking formation (>98%). Flanking
sequences greatly affect the cross-linking efficiency and rate
constant, as well as the fluorescence intensity of the coumarin
Effect of the Modification Site and the Template
Length on Coumarin-dT Cross-Linking. To explore the
generality and versatility of the photoswitchable DNA cross-
linking induced by coumarin moieties, we performed the same
type of experiments using a diol-based coumarin 7 which was
incorporated into ODNs at two different internal positions
(
10b and 13b). Similar to a terminal coumarin, compound 7 at
the internal position (ds DNA-10) lead to almost quantitative
ICL formation (97%) upon photoirradiation at 350 nm (Figure
8
, lane 2). The DNA cross-linking product formed with ds
DNA-10 was split into single-stranded ODNs upon 254 nm
irradiation with good efficiency (cleavage percentage 84%)
(Figure 8, lane 3). Kinetic study showed that both ICL
Figure 8. Phosphorimage autoradiogram of denaturing PAGE analysis of ICL formation and cleavage of ds DNAs-10, 12, and 13 (100 nM ds DNA
32
were irradiated at 350 or 254 nm; ODNs without coumarin were 5′-[ P]-labeled).
G
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
using 10 μM coumarin-containing oligonucleotides, which were
annealed with 1.0 equiv of the complementary strands by heating to
8
0 °C for 3 min in a buffer containing 10 mM pH 7 potassium
phosphate buffer and 100 mM NaCl, followed by slowly cooling to
room temperature overnight. The DNA duplexes were irradiated with
the UV light at 350 nm for the desired time to make sure that the
reactions were complete. Then, 2 M NaCl solution (200 μL) and cold
ethanol (1.2 mL) were added to the reaction mixture which was
incubated at −80 °C for 30 min. ODNs were precipitated by
centrifugation at 15 000 g for 6 min at room temperature. The crude
obtained even after 3 cycles. Most importantly, ICL formation
completely quenches the fluorescence of coumarin, which
allows for the monitoring of DNA cross-linking process on time
via fluorescence spectroscopy. To the best of our knowledge,
this is the first example that DNA interstrand cross-linking can
be directly quantitated using fluorescence spectroscopy. This
methodology also allows us to detect DNA transversion
mutation between pyrimidines and purines with a good
discriminating effect. Overall, coumarin analogues developed
in this work provide a means of fast, quantitative, directly
detectable, and photoswitchable DNA cross-linking, which
could serve as mechanistic probes and tools for bioresearch.
DNA was further purified via C18 column eluting with H O (3 × 1.0
2
mL) followed by MeOH:H O (3:2, 1.0 mL) to remove salts. The
2
afforded ODNs were dried and characterized by MALDI-TOF-MS at
the University of California-Riverside Mass Spectrometry Lab.
Determination of DNA Duplexes’ Thermal Stability. All
measurements were carried out in 10 mM potassium phosphate buffer
(pH 7), 100 μM ethylenediaminetetraacetic acid (EDTA), and 100
mM NaCl, with 4 μM + 4 μM single-strand concentration. Samples
−1
were heated at 1 °C min from 20 to 80 °C and the absorbance of
DNA at 260 nm was measured at 1.0 °C steps. At least two
independent samples have been tested to get the melting temperatures
of DNA duplexes.
PAGE Analysis of Interstrand Cross-Link Formation and
Kinetic Study using 32P-Labeling Method. ODNs (0.1 μM)
32
without coumarin were 5′- P-labeled and hybridized with 1.5 equiv of
the complementary strands in 100 mM NaCl and 10 mM potassium
phosphate (pH 7). The DNA duplexes were irradiated at 350 nm to
form ICLs (a control reaction was carried out without photo-
irradiation). Aliquots were taken at the prescribed time and
immediately quenched with the equal volume of 95% formamide
loading buffer, and stored at −20 °C until subjecting to 20%
denaturing PAGE analysis. For kinetics study, three independent
samples were used with the same procedures mentioned above.
Kinetic Study of DNA Interstrand Cross-Link Formation
using Fluorescence Assay. Coumarin-modified ODNs (10 μM)
were hybridized with 1.0 equiv of the complementary strands in 100
mM NaCl and 10 mM potassium phosphate (pH 7). The DNA
duplexes were irradiated at 350 nm and the fluorescence was measured
at the prescribed time. The equations used for calculation of the ICL
yields were shown in each figure. For kinetics study, two independent
samples were used.
EXPERIMENTAL SECTION
■
General Methods. Unless otherwise specified, reagents were used
as received without further purification. T4 polynucleotide kinase was
purchased from New England Biolabs. [γ- P]-ATP was obtained from
PerkinElmer Life Sciences. ODNs were synthesized via standard
automated DNA synthesis techniques in a 1.0 μmol scale using
32
i
commercial 1000 Å CPG-succinyl-nucleoside supports. The Pr-Pac-
i
dA and Pr-Pac-dG phosphoramidites were employed for the synthesis
of coumarin-containing ODNs. Deprotection of the nucleobases and
phosphate moieties as well as cleavage of the linker for normal ODNs
were carried out with a mixture of 40% aq. MeNH and 28% aq. NH3
1:1) at room temperature for 2 h. Functional ODNs were
2
(
deprotected and cleaved using 28% aq. NH at room temperature
3
for 2 h. Crude ODNs were purified by 20% denaturing polyacrylamide
gel electrophoresis (PAGE) analysis, and quantified in water by UV−
vis spectrophotometer at 260 nm. Thermal denaturation temperatures
PAGE Analysis of Photo-Induced (254 nm) Cleavage
Reactions of DNA ICL Products and Kinetic Study. The reactions
were carried out with photoirradiation at 254 nm using two or three
independent samples. Aliquots were taken at the prescribed time and
immediately quenched with the equal volume of 95% formamide
loading buffer, and stored at −20 °C until subjected to 20% denaturing
PAGE analysis.
(
Tm) were measured via changing temperature of DNA duplex at a rate
of 1 °C/min on a UV/vis sepectrometer equipped with a
thermoelectrical temperature controller. Radiolabeling was carried
out according to the standard protocols. The 32P-labeled ODN (100
nM) was annealed with 1.5 equiv of the complementary strand by
heating to 80 °C for 3 min in a buffer of 10 mM potassium phosphate
(
pH 7.0) and 100 mM NaCl, followed by slow-cooling to room
7
-Hydroxy-4-methyl-2H-chromen-2-O-(2-cyanoethyl-N,N-
temperature overnight. Quantification of radiolabeled ODNs was
carried out using a Molecular Dynamics Phosphorimager equipped
with ImageQuant v 5.2 software.
diisopropyl)-phosphoramidite (8). To a solution of 7-hydroxy-4-
methyl-2H-chromen-2-one (3, 88 mg, 0.5 mmol) in dichloromethane
(
10 mL), N,N-diisopropylethylamine (DIPEA) (156 μL, 0.9 mmol),
and 2-cyanoethyl-N,N-diisopropylchlorophosphoramidite (167 μL,
.75 mmol) were added under an atmosphere of argon. The reaction
Photoirradiation at 350 nm was conducted in Rayonet Photo-
chemical Chamber Reactor. Fluorescence spectra were recorded using
the quartz cell with 10 mm path lengths at room temperature.
Compounds 1, 2, 4, 5, and 7 were synthesized with modified
0
mixture was stirred at room temperature for 3 h and diluted with
dichloromethane (50 mL). The organic layer was washed with
3
7,38
1
13
31
procedures.
H NMR, C NMR, and P NMR analysis were
NaHCO (20 mL) and saturated aqueous NaCl (20 mL), and dried
performed on a 300 MHz spectrophotometer. Chemical shifts are
reported in ppm relative to Me Si ( H and C) or H PO ( P).
Coupling constants (J) are reported in Hz. High resolution MALDI-
TOF-MS spectrometry was performed at the University of California-
Riverside Mass Spectrometry Lab.
LC-MS Analysis of Photoinduced Reaction between
Coumarin 1 and Canonical Nucleosides. The reaction mixtures
containing 30 nmol deoxyribonucleosides (dA: 7.5 mg, dT: 7.2 mg,
dG: 8.0 mg, or dC: 6.8 mg) and coumarin 1 (5.7 mg, 30 nmol) in
methanol (1−6 mL, depending on solubility) were placed in four vials,
respectively. After UV irradiation at 350 nm for 2 days, the mixtures
were diluted to the desired concentration for LC-MS analysis.
Preparation of DNA Cross-Linked Products for MALDI-TOF-
MS Analysis. Interstrand cross-linking reactions were performed
3
1
13
31
over anhydrous sodium sulfate. The solvent was removed under
reduced pressure. Upon purification by column chromatography
4
3
4
(
EtOAc:hexane:Et N = 59:40:1), the product was isolated as a white
3
31
1
solid (8, 170 mg, 90%). P NMR (300 MHz, CDCl -d): δ 147.48. H
3
NMR (300 MHz, CDCl -d): δ 7.50−7.53 (m, 1H), 7.00−7.02 (dd, J =
3
1.5 Hz, 2H), 6.18 (s, 1H), 3.93−4.01 (m, 2H), 3.71−3.79 (m, 2H),
1
3
2.69−2.73 (t, J = 6.3 Hz, 3H), 1.18−1.27 (dd, J = 6.6 Hz, 1H).
C
NMR (300 MHz, CDCl -d): δ 161.1, 157.7, 157.6, 154.8, 152.4, 125.5,
3
117.3, 116.5, 116.4, 115.0, 112.7, 107.5, 107.3, 59.3, 59.0, 44.0, 43.8,
24.7, 24.6, 24.5, 24.4, 20.4, 20.3, 18.7. HRMS-ESI (+) (m/z): [M +
+
H] calcd. for C19
H
N
26
O
4
P, 377.1625; found: 377.1633.
2
4-Methyl-7-methoxy-2H-chromen-2-one (1). Iodomethane
(1.3 mL, 20 mmol) was added to a suspension of 7-hydroxy-4-
H
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
methyl-2H-chromen-2-one (3, 704 mg, 4 mmol) and K CO (600 mg,
7-(2-Hydroxyethoxy)-4-methyl-2H-chromen-2-O-(2-cya-
noethyl-N,N-diisopropyl)-phosphoramidite (10). To a solution
of 4 (110 mg, 0.5 mmol) in dichloromethane (10 mL), N,N-
diisopropylethylamine (DIPEA) (156 μL, 0.9 mmol) and 2-
cyanoethyl-N,N-diisopropylchlorophosphoramidite (167 μL, 0.75
mmol) were added under an atmosphere of argon. The reaction
mixture was stirred at room temperature for 3 h and diluted with
2
3
3
5
.34 mmol) in acetone (50 mL). The reaction mixture was stirred at
0 °C overnight. After cooling to room temperature, the solvent was
removed under reduced pressure. The residue was diluted with ethyl
acetate (50 mL), washed with 1 M HCl (30 mL), H O (30 mL), and
2
brine (20 mL), and then dried over anhydrous sodium sulfate. The
solvent was removed under vacuum providing the product 1 as a white
1
EtOAc (50 mL). The organic layer was washed with NaHCO (20
solid (699 mg, 92%). H NMR (300 MHz, CDCl -d): δ 7.60−7.65 (m,
3
3
mL) and saturated aqueous NaCl (20 mL), and dried over anhydrous
sodium sulfate. The solvent was removed under reduced pressure. The
crude product was purified by column chromatography (EtOAc:hex-
1
H), 7.01−7.02 (t, J = 1.8 Hz, 1H), 6.93−6.98 (m, 2H), 6.25, 6.28 (t, J
=
1.8 Hz, 1H), 3.86 (s, 3H), 2.37 (s, 3H).
-(Hydroxymethyl)-7-methoxy-2H-chromen-2-one (2). To a
suspension of 7-hydroxy-4-(hydroxylmethyl)-2H-chromen-2-one (14,
84 mg, 2 mmol) and K CO (300 mg, 2.17 mmol) in acetone (30
4
3
1
ane:Et N = 49:50:1) yielding 10 as a white solid (162 mg, 77%).
P
3
1
NMR (CDCl -d, 300 MHz): δ 149.28. H NMR (CDCl -d, 300
3
3
3
2
3
MHz): δ 7.41, 7.44 (d, J = 8.7 Hz, 1H), 6.75−6.82 (t, J = 8.7 Hz, 2H),
mL), iodomethane (0.65 mL, 10 mmol) was added. The reaction
mixture was stirred at 50 °C overnight. After cooling to room
temperature, the solvent was removed under reduced pressure. The
residue was diluted with ethyl acetate (50 mL), washed with 1 M HCl
6
.06 (s, 1H), 4.12−4.15 (t, J = 4.5 Hz, 2H), 3.88−4.00 (m, 2H), 3.75−
.80 (m, 2H), 3.52−3.60 (m, 2H), 2.55−2.59 (t, J = 6.3 Hz, 2H), 2.32
3
(
s, 3H), 1.11, 1.13 (d, J = 6.6 Hz, 12H). 13C NMR (300 MHz, CDCl -
3
d): δ 161.8, 161.2, 155.2, 152.6, 125.6, 117.6, 113.7, 112.6, 112.0,
01.6, 68.6, 68.5, 61.9, 61.6, 58.6, 43.2, 43.1, 24.7, 24.6, 20.4, 20.3,
(
30 mL), and brine (20 mL), and then dried over anhydrous sodium
1
sulfate. The solvent was removed under vacuum providing 2 as a white
+
1
18.7. HRMS-ESI (+) (m/z): [M + H] calcd. for C H N O P,
solid (394 mg, 96%). H NMR (300 MHz, CDCl -d): δ 7.61−7.64 (m,
21 30 2 5
3
4
21.1887; found: 421.1899.
-(3-Hydroxypropoxy)-4-methyl-2H-chromen-2-O-(2-cya-
1
H), 7.01−7.02 (t, J = 1.8 Hz, 1H), 6.92−6.97 (m, 1H), 6.30, 6.31 (t, J
7
=
3
1.5 Hz, 1H), 5.60−5.64 (m, 1H), 4.73−4.75 (t, J = 1.5 Hz, 2H),
.85, 3.86 (d, J = 1.8 Hz, 3H).
-(Hydroxymethyl)-7-methoxy-2H-chromen-2-O-(2-cya-
noethyl-N,N-diisopropyl)-phosphor-amidite (9). Compound 2
103 mg, 0.5 mmol) was dissolved in dichloromethane (8 mL) under
noethyl-N,N-diisopropyl)-phosphoramidite (11). To a solution
of 5 (117 mg, 0.5 mmol) in dichloromethane (10 mL), N,N-
diisopropylethylamine (DIPEA) (156 μL, 0.9 mmol) and 2-
cyanoethyl-N,N-diisopropylchlorophosphoramidite (167 μL, 0.75
mmol) were added under an atmosphere of argon. The reaction
mixture was stirred at room temperature for 3 h and diluted with
dichloromethane (30 mL). The organic layer was washed with
4
(
an atmosphere of argon. N,N-Diisopropylethylamine (DIPEA) (156
μL, 0.9 mmol) was then added dropwise, followed by 2-cyanoethyl-
N,N-diisopropylchlorophosphoramidite (167 μL, 0.75 mmol). The
reaction mixture was stirred at room temperature for 3 h, then diluted
NaHCO (20 mL) and saturated aqueous NaCl (20 mL), and dried
3
over anhydrous sodium sulfate. The solvent was removed under
reduced pressure. The residue was purified by column chromatog-
by dichloromethane (30 mL), washed with NaHCO (20 mL), and
3
brine (20 mL), and dried over anhydrous sodium sulfate. The solvent
was removed under reduced pressure. Upon purification by column
raphy (EtOAc:hexane:Et N = 49:50:1) providing 11 as a white solid
3
1
(
120 mg, 55%). 31P NMR (CDCl -d, 300 MHz): δ 147.89. H NMR
3
chromatography (EtOAc:hexane:Et N = 40:59:1), the product 9 was
3
31
(CDCl -d, 300 MHz): δ 7.40, 7.43 (d, J = 8.7 Hz, 1H), 6.75−6.81 (m,
3
isolated as a white solid (148 mg, 73%). P NMR (300 MHz, CDCl3-
2
3
1
1
5
H), 6.06 (s, 1H), 4.06−4.10 (t, J = 6.0 Hz, 2H), 3.72−3.80 (m, 2H),
1
d): δ 149.96. H NMR (300 MHz, CDCl -d): δ 7.44−7.47 (m, 1H),
3
.49−3.57 (m, 2H), 2.33 (s, 3H), 2.03−2.07 (t, J = 6.0 Hz, 2H), 1.10,
6
4
2
.84−6.88 (dd, J = 3.0 Hz, 2H), 6.44−6.45 (d, J = 1.5 Hz, 1H), 4.84−
13
.11 (t, J = 2.1 Hz, 12H). C NMR (300 MHz, CDCl -d): δ 162.1,
3
.90 (m, 2H), 3.85−3.95 (m, 5H), 3.68−3.71 (dd, J = 3.6 Hz, 2H),
61.3, 155.3, 152.5, 125.5, 117.6, 113.6, 112.6, 111.9, 101.5, 65.1, 60.0,
13
.66−2.71 (t, J = 6.3 Hz, 2H), 1.22−1.25 (dd, J = 2.4 Hz, 12H).
C
9.8, 58.4, 58.2, 43.2, 43.0, 30.8, 30.7, 24.7, 24.6, 20.4, 20.3, 18.7.
NMR (300 MHz, CDCl -d): δ 162.6, 161.5, 155.4, 152.5, 152.4, 124.5,
+
3
HRMS-ESI (+) (m/z): [M + H] calcd. for C H N O P, 435.2043;
22
32
2
5
1
2
17.6, 112.4, 110.9, 109.5, 101.1, 61.7, 61.5, 58.5, 58.2, 55.8, 43.5, 43.3,
4.8, 24.7, 24.6, 20.5, 20.4. HRMS-ESI (+) (m/z): [M + H] calcd. for
found: 435.2055.
+
7
-(1-(3-Hydroxypropyl)-1H-1,2,3-triazol-5-yl)-4-methyl-2H-
C H N O P, 407.1730; found: 407.1745.
2
0
28
2
5
chromen-2-one (6). To a suspension of 7-ethynyl-4-methyl-2H-
chromen-2-one (15, 184 mg, 1 mmol), 3-azidopropan-1-ol (152 mg,
7
-(2-Hydroxyethoxy)-4-methyl-2H-chromen-2-one (4). To a
suspension of 7-hydroxy-4-methyl-2H-chromen-2-one (3, 704 mg, 4
1
.5 mmol) and copper(II) sulfate (pentahydrate) (50 mg, 0.2 mmol)
mmol), KI (1.65 g, 10 mmol), and K CO (900 mg, 6.51 mmol) in
2
3
in a mixture of water (10 mL) and tert-butyl alcohol (10 mL), sodium
ascorbate (119 mg, 0.6 mmol) was added and stirred at room
temperature overnight. After addition of water (20 mL) to mixture,
extract with ethyl acetate (20 mL × 3). The organic phases were
collected, washed with water (20 mL × 2) and brine (20 mL), and
then dried over anhydrous Na SO . The solvent was removed under
acetone (60 mL), 2-bromoethanol (1.42 mL, 20 mmol) was added.
The reaction mixture was stirred at 50 °C for 48 h. After cooling to
room temperature, the solvents were removed under reduced pressure.
The residue was diluted with ethyl acetate (80 mL), washed with 1 M
HCl (40 mL), and brine (20 mL), and then dried over anhydrous
2
4
Na SO . The solvent was removed under reduced pressure. The crude
2
4
reduced pressure and the white residue was purified by column
product was purified by column chromatography (EtOAc) yielding 4
chromatography (EtOAc) to afford 6 as a white solid (186 mg, 65%).
1
1
as a white solid (642 mg, 73%). H NMR (300 MHz, CDCl -d): δ
3
H NMR (300 MHz, CDCl -d): δ 8.79 (s, 1H), 7.81−7.86 (m, 3H),
3
7
1
3
.67, 7.70 (d, J = 9.3 Hz, 1H), 6.96, 6.99 (d, J = 6.6 Hz, 2H), 6.21 (s,
H), 4.90−4.93 (t, J = 5.4 Hz, 1H), 4.09−4.12 (t, J = 5.4 Hz, 2H),
.72−3.77 (dd, J = 5.1 Hz, 2H), 2.40 (s, 3H).
6
2
2
1
.39 (s, 1H), 4.70−4.73 (t, J = 5.1 Hz, 1H), 4.47−4.52 (t, J = 6.9 Hz,
H), 3.43−3.49 (dd, J = 6.0 Hz, 2H), 2.45 (s, 3H), 2.00−2.08 (m,
13
H). C NMR (300 MHz, CDCl -d): δ 160.2, 154.0, 153.5, 145.3,
3
7
-(3-Hydroxypropoxy)-4-methyl-2H-chromen-2-one (5). To
34.8, 126.6, 123.4, 121.4, 119.4, 114.5, 57.9, 47.5, 33.2, 18.5. HRMS-
a suspension of 3 (704 mg, 4 mmol), KI (1.65 g, 10 mmol), and
+
ESI (+) (m/z): [M + H] calcd. for C H N O , 286.1192; found:
15
16
3
3
K CO3 (900 mg, 6.51 mmol) in acetone (60 mL), 3-bromo-1-
2
286.1186.
-Cyanoethyl (3-(4-(4-methyl-2-oxo-2H-chromen-7-yl)-1H-
propanol (0.54 mL, 6 mmol) was added. The reaction mixture was
stirred at 50 °C for 36 h. After cooling to room temperature, the
solvent was removed under reduced pressure. The residue was diluted
with ethyl acetate (80 mL), washed with 1 M HCl (40 mL), and brine
2
1,2,3-triazol-1-yl)propyl) diisopropyl-phosphoramidite (12).
To a suspension of 6 (143 mg, 0.5 mmol) in dichloromethane (10
mL), N,N-diisopropylethylamine (DIPEA) (156 μL, 0.9 mmol) and 2-
cyanoethyl-N,N-diisopropylchlorophosphoramidite (167 μL, 0.75
mmol) were added under an atmosphere of argon. The reaction
mixture was protected from light by aluminum foil, stirred at room
temperature for 3 h, and diluted with dichloromethane (30 mL). The
(
20 mL), and then dried over anhydrous Na SO . The solvent was
2 4
removed under reduced pressure to afford 5 as a white solid (760 mg,
1
8
1
3
1%). H NMR (300 MHz, CDCl -d): δ 7.49, 7.52 (d, J = 8.7 Hz,
3
H), 6.83−6.90 (m, 2H), 6.14 (s, 1H), 4.18−4.22 (t, J = 6.0 Hz, 2H),
.87−3.92 (dd, J = 5.4 Hz, 2H), 2.41 (s, 3H), 2.06−2.14 (m, 2H).
organic layer was washed with 1 M NaHCO (20 mL) and saturated
3
I
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
aqueous NaCl (20 mL), and dried over anhydrous sodium sulfate in
the dark room. The solvent was removed under reduced pressure. The
crude product was purified by column chromatography (EtOAc:hex-
3.75, 3.16−3.31 (m, 2H), 2.49−2.60 (m, 1H), 2.37−2.41 (m, 1H),
1
3
2.32 (s, 3H), 1.07−1.13 (m, 9H), 0.96−0.98 (d, J = 6.6 Hz, 3H).
C
NMR (300 MHz, CDCl -d): δ 160.8, 160.2, 157.5, 154.2, 151.5, 143.6,
3
ane:Et N = 49:50:1) in the dark room yielding 12 as a white solid (73
134.9, 129.1, 127.2, 126.8, 125.8, 124.6, 116.6, 112.7, 112.1, 111.5,
111.4, 111.0, 100.7, 85.2, 70.9, 70.7, 70.5, 70.3, 68.3, 68.1, 62.4, 57.5,
3
31
1
mg, 30%). P (CDCl -d, 300 MHz): δ 148.06. H NMR (CDCl -d,
3
3
3
1
00 MHz): δ 7.92 (s, 1H), 7.76−7.78 (d, J = 8.1 Hz, 1H), 7.63 (s,
57.2, 54.2, 42.3, 42.1, 29.7, 23.4, 19.3, 19.1, 17.7. HRMS-ESI (+) (m/
+
H), 7.56−7.58 (d, J = 8.4 Hz, 1H), 6.19 (s, 1H), 4.49−4.54 (t, J = 6.9
z): [M + H] calcd. for C43
H
50
N
O
2
8
P, 753.3305; found: 753.3295.
Hz, 2H), 3.48−3.86 (m, 6H), 2.57−2.61 (t, J = 6.0 Hz, 2H), 2.38 (s,
Synthesis of Coumarin-dT and Coumarin−Coumarin
3
H), 2.20−2.24 (t, J = 6.0 Hz, 2H), 1.09−1.13 (dd, J = 4.2 Hz, 12H).
Dimers. A solution of 4-methyl-7-methoxy-2H-chromen-2-one (1,
1
3
3
8 mg, 0.2 mmol) and dT (146 mg, 0.6 mmol) in methanol (200 mL)
C NMR (300 MHz, CDCl -d): δ 159.8, 152.9, 151.2, 144.9, 133.4,
3
was irritated with UV light at 350 nm for 3 days at room temperature.
After removing solvent under reduced pressure, products were isolated
upon purification by column chromatography.
1
4
24.2, 120.5, 118.5, 116.8, 113.8, 112.5, 59.0, 58.8, 57.4, 57.1, 46.3,
2.2, 42.0, 30.6, 30.5, 23.7, 23.6, 19.6, 19.5, 17.6. HRMS-ESI (+) (m/
+
z): [M + H] calcd. for C H N O P, 486.2278; found: 486.2285.
24
33
5
4
7
-(4-Hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-3-
7
-(2,3-Dihydroxypropoxy)-4-methyl-2H-chromen-2-one (7).
To a suspension of 7-hydroxy-4-methyl-2H-chromen-2-one (3, 1.76 g,
0 mmol) and KOH (1.12 g, 20 mmol) in ethanol (50 mL),
methoxy-10a,10b-dimethyl-6b,7,10a,10b-tetrahydro-6H-
chromeno[4′,3′:3,4]cyclobuta[1,2-d]pyrimidine-6,8,10-
1
(
6aH,9H)-trione (17). (CH Cl :MeOH = 95:5, a yellowish oil, 13 mg,
5%). H NMR (300 MHz, DMSO-d ): δ 7.18, 7.21 (d, J = 8.7 Hz,
2
2
epichlorohydrin (8.0 mL, 102 mmol) and KI (166 mg, 1 mmol) were
added. The reaction mixture was stirred at 70 °C overnight. After
cooling to room temperature, the solvent was removed under vacuum.
The residue was submitted to flash chromatography (CH Cl :MeOH
1
1
1
4
3
1
1
7
6
H), 6.70−6.73 (m, 1H), 6.51−6.53 (t, J = 2.4 Hz, 1H), 5.88 (m, 1H),
.78, 4.82 (d, J = 9.3 Hz, 1H), 4.19−4.21 (t, J = 2.4 Hz, 1H), 4.04 (s,
2
2
H), 3.64 (s, 2H), 3.28, 3.31 (d, J = 9.3 Hz, 1H), 1.99−2.06 (m, 2H),
=
98:2) to afford 4-methyl-7-(oxiran-2-ylmethoxy)-2H-chromen-2-one
13
.56 (s, 3H), 1.27 (s, 3H). C NMR (300 MHz, DMSO-d ): δ 172.4,
1
6
as a white solid (1.83 g, 79%). H NMR (300 MHz, CDCl -d): δ
3
64.7, 159.8, 151.5, 151.3, 128.7, 116.5, 111.2, 101.7, 87.3, 85.7, 83.3,
0.8, 62.0, 55.8, 52.3, 52.4, 46.9, 22.8, 22.6, 18.3, 17.9. HRMS-ESI (+)
m/z): [M + H] calcd. for C H N O , 433.1605; found: 433.1612.
7
.42−7.45 (d, J = 9.0 Hz, 1H), 6.81−6.85 (dd, J = 2.4 Hz, 1H), 6.76
(
3
1
d, J = 2.4 Hz, 1H), 6.08 (s, 1H), 4.25−4.29 (d, J = 2.7 Hz, 1H), 3.88−
+
(
21
25
2
8
.93 (q, J = 6.0 Hz, 1H), 3.30−3.35 (m, 1H), 2.86−2.89 (t, J = 4.5 Hz,
H), 2.71−2.74 (q, J = 2.7 Hz, 1H), 2.33 (s, 3H). The solution of 4-
1
0-(4-Hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-3-
methoxy-6b,10b-dimethyl-8,10,10a,10b-tetrahydro-6H-
chromeno[3′,4′:3,4]cyclobuta[1,2-d]pyrimidine-6,7,9-
(6aH,6bH)-trione (18). (CH Cl :MeOH = 95:5, a yellowish oil, 8
methyl-7-(oxiran-2-ylmethoxy)-2H-chromen-2-one (1.16 g, 5 mmol)
in 6% perchloric acid (40 mL) was stirred at room temperature
overnight. Then, the pH of the solution was adjusted to 8 via addition
of Na CO . After stirring for 2 h, ethyl acetate (80 mL) was added and
2
2
1
mg, 9.2%). H NMR (300 MHz, DMSO-d ): δ 10.65 (s, 1H), 7.20,
6
2
3
7.23 (d, J = 8.7 Hz, 1H), 6.79−6.82 (m, 1H), 6.65, 6.66 (d, J = 2.4 Hz,
1H), 5.97−6.02 (m, 1H), 5.06, 5.08 (d, J = 3.6 Hz, 1H), 4.50 (s, 1H),
4.03−4.10 (m, 3H), 3.78 (s, 3H), 3.16−3.18 (m, 3H), 1.77 (m, 2H),
stirred for 20 h. The organic layer was washed with water (20 mL) and
saturated aqueous NaCl (20 mL), and dried over anhydrous sodium
sulfate. The solvent was removed under reduced pressure to afford 7
1.36 (s, 3H), 1.01 (s, 3H). 13C NMR (300 MHz, DMSO-d ): δ 172.5,
6
(
1.05 g, 84%) as a viscous liquid which was easily solidified into a
163.8, 160.0, 151.7, 130.9, 114.3, 111.2, 102.6, 87.3, 84.1, 71.7, 62.1,
55.9, 52.7, 52.2, 50.3, 49.1, 42.7, 42.0, 37.9, 27.1, 20.4. HRMS-ESI (+)
1
white solid at room temperature. H NMR (DMSO-d ): δ 7.66−7.68
6
+
(
d, J = 8.4 Hz, 1H), 6.95−6.99 (m, 2H), 6.20 (s, 1H), 5.04−5.06 (d, J
5.1 Hz, 1H), 4.72−4.76 (t, J = 5.4 Hz, 1H), 4.10−4.14 (dd, J = 3.9
Hz, 1H), 3.96−4.01 (m, 1H), 3.80−3.85 (m, 1H), 3.44−3.48 (t, J =
.4 Hz, 2H), 3.04 (s, 3H).
-(3-(Bis(4-methoxyphenyl)(phenyl)methoxy)-2-hydroxy-
(m/z): [M + H] calcd. for C H N O , 433.1605; found: 433.1615.
21
25
2
8
=
3,9-Dimethoxy-6b,12b-dimethyl-12a,12b-dihydrocyclobuta-
[1,2-c:3,4-c′]dichromene-6,12(6aH,6bH)-dione (19).
1
5
(CH
MHz, CDCl
2H), 6.56−6.00 (m, 2H), 3.76, 3.60 (s, 6H), 3.35, 3.31 (s, 2H), 1.60,
1.18 (s, 6H). 13C NMR (300 MHz, CDCl
-d): δ 165.0, 163.7, 159.3,
2
Cl
2
:MeOH = 99:1, a white solid, 5.6 mg, 15%). H NMR (300
7
-d): δ 6.97−7.04 (m, 2H), 6.70−6.73, 5.98−5.99 (m,
3
propoxy)-4-methyl-2H-chromen-2-one (16). To a solution of
compound 7 (0.50 g, 2.0 mmol) in pyridine (8 mL), 4,4′-
dimethoxytriphenylmethyl chloride (0.81 g, 2.4 mmol) was added.
The mixture was stirred at room temperature overnight. The reaction
mixture was quenched with MeOH (5 mL) and concentrated under
reduced pressure. Upon purification by column chromatography
3
150.7, 149.3, 127.0, 126.5, 114.0, 112.8, 111.0, 110.7, 101.4, 100.8,
54.6, 54.2, 45.6, 44.0, 40.0, 30.7, 25.4. HRMS-ESI (+) (m/z): [M +
+
H] calcd. for C22
H
O
21
, 381.1338; found: 381.1346.
6
(
EtOAc:Et N = 99:1), 16 was isolated as a white foam (0.76 g, 69%).
ASSOCIATED CONTENT
Supporting Information
NMR and HRMS-ESI spectra for all new compounds, MALDI-
TOF-MS spectra for functional ODNs, ICL formation or
cleavage study via 32P-labeling method or fluorescence assay,
and related kinetic study. This material is available free of
charge via the Internet at http://pubs.acs.org.
3
■
1
H NMR (DMSO-d ): δ 7.67−7.70 (d, J = 8.7 Hz, 1H), 7.39−7.41 (d,
6
S
*
J = 7.5 Hz, 2H), 7.19−7.32 (m, 7H), 6.85−6.96 (m, 6H), 6.22 (s, 1H),
5
3
.24−5.26 (d, J = 5.4 Hz, 1H), 4.00−4.19 (m, 3H), 3.73 (s, 6H),
13
.08−3.10 (d, J = 5.7 Hz, 2H), 2.4 (s, 3H). C NMR (300 MHz,
CDCl -d): δ 162.2, 160.7, 158.5, 155.2, 153.9, 145.4, 136.2, 130.2,
3
1
5
5
28.2, 127.1, 126.9, 113.6, 113.0, 111.6, 101.7, 85.8, 70.6, 68.4, 64.6,
+
5.5, 18.6. HRMS-ESI (+) (m/z): [M + H] calcd. for C H O ,
34
33
7
53.2226; found: 553.2216.
1
-(Bis(4-methoxyphenyl)(phenyl)methoxy)-3-((4-methyl-2-
AUTHOR INFORMATION
Corresponding Author
E-mail: pengx@uwm.edu.
Notes
■
*
oxo-2H-chromen-7-yl)oxy)-propan-2-yl (2-cyanoethyl) diiso-
propylphosphoramidite (13). To a solution of compound 16
(
atmosphere, N,N-diisopropylethylamine (DIPEA) (156 μL, 0.9 mmol)
was then added dropwise, followed by 2-cyanoethyl-N,N-diisopropyl-
chlorophosphoramidite (167 μL, 0.75 mmol). The reaction mixture
was stirred at room temperature for 3 h, then diluted by
dichloromethane (30 mL), washed with NaHCO3 (20 mL), and
brine (20 mL), and dried over anhydrous sodium sulfate. The solvent
was removed under reduced pressure. Upon purification by column
chromatography (EtOAc:hexane:Et N = 40:59:1), the product 13 was
isolated as a white foam (266 mg, 71%). P NMR (CDCl -d, 300
MHz): δ 149.99, 149.67. H NMR (DMSO-d ): δ 7.12−7.43 (m,
1
226 mg, 0.50 mmol) in anhydrous CH Cl (5 mL) under argon
2 2
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful for financial support from the UWM Research
Growth Initiative (RGI101 × 234), the Greater Milwaukee
Foundation (Shaw Scientist Award), and Wisconsin Applied
Research Grant (ARG) Award. We thank UWM graduate
school for providing distinguished graduate student fellowship
for Huabing Sun. We also thank Hye Young Eom from the
3
31
3
1
6
0H), 6.71−6.80 (m, 6H), 6.06 (s, 1H), 4.02−4.26 (m, 3H), 3.42−
J
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
undergraduate research program for preparation and property
study of coumarin 1.
(35) Heinlein, T.; Knemeyer, J. P.; Piestert, O.; Sauer, M. J. Phys.
Chem. B 2003, 107, 7957.
(36) Kendall, D. A.; MacDonald, R. C. J. Biol. Chem. 1982, 257,
1
3892.
REFERENCES
■
(37) Zhang, B.-L.; Wang, F.-D.; Yue, J.-M. Synth. Commun. 2007, 37,
6
(
3.
(
1) Brulikova, L.; Hlavac, J.; Hradil, P. Curr. Med. Chem. 2012, 19,
38) Chen, Y. L.; Lu, C. M.; Lee, S. J.; Kuo, D. H.; Chen, I. L.; Wang,
3
64.
T. C.; Tzen, C. C. Bioorg. Med. Chem. 2005, 13, 5710.
(
2) Noll, D. M.; Mason, T. M.; Miller, P. S. Chem. Rev. 2006, 106,
2
77.
(
(
3) Dronkert, M. L.; Kanaar, R. Mutat. Res. 2001, 486, 217.
4) Peng, X.; Ghosh, A. K.; Van Houten, B.; Greenberg, M. M.
Biochemistry 2010, 49, 11.
5) Peng, X.; Pigli, Y. Z.; Rice, P. A.; Greenberg, M. M. J. Am. Chem.
Soc. 2008, 130, 12890.
6) Weng, M. W.; Zheng, Y.; Jasti, V. P.; Champeil, E.; Tomasz, M.;
Wang, Y.; Basu, A. K.; Tang, M. S. Nucleic Acids Res. 2010, 38, 6976.
7) Banerjee, A.; Santos, W. L.; Verdine, G. L. Science 2006, 311,
153.
(
(
(
1
(
(
(
8) Mishina, Y.; He, C. J. Am. Chem. Soc. 2003, 125, 8730.
9) Duguid, E. M.; Mishina, Y.; He, C. Chem. Biol. 2003, 10, 827.
10) Rusling, D. A.; Nandhakumar, I. S.; Brown, T.; Fox, K. R. Chem.
Commun. 2012, 48, 9592.
11) Zewail-Foote, M.; Hurley, L. H. J. Am. Chem. Soc. 2001, 123,
485.
12) Asai, A.; Nagamura, S.; Saito, H.; Takahashi, I.; Nakano, H.
Nucleic Acids Res. 1994, 22, 88.
(
6
(
(
(
13) Liu, Y.; Rokita, S. E. Biochemistry 2012, 51, 1020.
14) Wang, H.; Wahi, M. S.; Rokita, S. E. Angew. Chem., Int. Ed. 2008,
4
(
7, 1291.
15) Weinert, E. E.; Dondi, R.; Colloredo-Melz, S.; Frankenfield, K.
N.; Mitchell, C. H.; Freccero, M.; Rokita, S. E. J. Am. Chem. Soc. 2006,
28, 11940.
16) Zhou, Q.; Rokita, S. E. Proc. Natl. Acad. Sci. U. S. A. 2003, 100,
5452.
17) Browne, W. R.; Feringa, B. L. Annu. Rev. Phys. Chem. 2009, 60,
07.
18) Chung, J. W.; Lee, K.; Neikirk, C.; Nelson, C. M.; Priestley, R.
D. Small 2012, 8, 1693.
19) Majumdar, A.; Muniandy, P. A.; Liu, J.; Liu, J. L.; Liu, S. T.;
Cuenoud, B.; Seidman, M. M. J. Biol. Chem. 2008, 283, 11244.
20) Shahid, K. A.; Majumdar, A.; Alam, R.; Liu, S. T.; Kuan, J. Y.;
1
(
1
(
4
(
(
(
Sui, X.; Cuenoud, B.; Glazer, P. M.; Miller, P. S.; Seidman, M. M.
Biochemistry 2006, 45, 1970.
(
21) Thazhathveetil, A. K.; Liu, S. T.; Indig, F. E.; Seidman, M. M.
Bioconjugate Chem. 2007, 18, 431.
22) Cimino, G. D.; Gamper, H. B.; Isaacs, S. T.; Hearst, J. E. Annu.
Rev. Biochem. 1985, 54, 1151.
23) Takasugi, M.; Guendouz, A.; Chassignol, M.; Decout, J. L.;
Lhomme, J.; Thuong, N. T.; Helene, C. Proc. Natl. Acad. Sci. U. S. A.
991, 88, 5602.
24) Fujimoto, K.; Yamada, A.; Yoshimura, Y.; Tsukaguch, T.;
Sakamoto, T. J. Am. Chem. Soc. 2013, 135, 16161.
(
(
1
(
(
25) Yoshimura, Y.; Fujimoto, K. Org. Lett. 2008, 10, 3227.
(
26) Li, H.; Broughton-Head, V. J.; Fox, K. R.; Brown, T. Nucleosides,
Nucleotides Nucleic Acids 2007, 26, 1005.
27) Li, H.; Broughton-Head, V. J.; Peng, G. M.; Powers, V. E. C.;
Ovens, M. J.; Fox, K. R.; Brown, T. Bioconjugate Chem. 2006, 17, 1561.
28) Trenor, S. R.; Shultz, A. R.; Love, B. J.; Long, T. E. Chem. Rev.
004, 104, 3059.
29) Musa, M. A.; Cooperwood, J. S.; Khan, M. O. Curr. Med. Chem.
008, 15, 2664.
30) Haque, M. M.; Sun, H.; Liu, S.; Wang, Y.; Peng, X. Angew.
(
(
2
(
2
(
Chem., Int. Ed. 2014, 53, 7001−7005.
(
(
(
31) Kool, E. T. Chem. Rev. 1997, 97, 1473.
32) Sun, H.; Peng, X. Bioconjugate Chem. 2013, 24, 1226.
33) Laing, B. M.; Barrow-Laing, L.; Harrington, M.; Long, E. C.;
Bergstrom, D. E. Bioconjugate Chem. 2010, 21, 1537.
34) Seidel, C. A. M.; Schulz, A.; Sauer, M. H. M. J. Phys. Chem.
996, 100, 5541.
(
1
K
dx.doi.org/10.1021/jo5014756 | J. Org. Chem. XXXX, XXX, XXX−XXX