Synthesis and photooxidation of oligodeoxynucleotides containing 5-dimethylaminocytosine as an efficient hole-trapping site in the positive-charge transfer through DNA duplexes

Article abstract of DOI:10.1039/c2ob06642d

We have designed and synthesized DNA duplexes containing 5-dimethylaminocytosine (^DMAC) to investigate the effects of C(5)-substituted cytosine bases on the transfer and trapping of positive charge (holes) in DNA duplexes. Fluorescence quenching experiments revealed that a ^DMAC base is more readily one-electron oxidized into a radical cation intermediate as compared with other natural nucleobases. Upon photoirradiation of the duplexes containing ^DMAC, the photosensitizer-injected hole migrated through the DNA bases and was trapped efficiently at the ^DMAC sites, where an enhanced oxidative strand cleavage occurred by hot piperidine treatment. The ^DMAC radical cation formed by hole transfer may undergo specific hydration and subsequent addition of molecular oxygen, thereby leading to its decomposition followed by a predominant strand cleavage at the ^DMAC site. This remarkable property suggests that the modified cytosine ^DMAC can function as an efficient hole-trapping site in the positive-charge transfer in DNA duplexes.

Full text of DOI:10.1039/c2ob06642d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 2035
www.rsc.org/obc
PAPER
Synthesis and photooxidation of oligodeoxynucleotides containing
5-dimethylaminocytosine as an efficient hole-trapping site in the
positive-charge transfer through DNA duplexes†
Hisatsugu Yamada,*a Masayuki Kurata,b Kazuhito Tanabe,b Takeo Itob and Sei-ichi Nishimoto*b
Received 27th September 2011, Accepted 12th December 2011
DOI: 10.1039/c2ob06642d
We have designed and synthesized DNA duplexes containing 5-dimethylaminocytosine (DMAC) to
investigate the effects of C(5)-substituted cytosine bases on the transfer and trapping of positive charge
(holes) in DNA duplexes. Fluorescence quenching experiments revealed that a DMAC base is more readily
one-electron oxidized into a radical cation intermediate as compared with other natural nucleobases. Upon
photoirradiation of the duplexes containing DMAC, the photosensitizer-injected hole migrated through the
DNA bases and was trapped efficiently at the DMAC sites, where an enhanced oxidative strand cleavage
occurred by hot piperidine treatment. The DMAC radical cation formed by hole transfer may undergo
specific hydration and subsequent addition of molecular oxygen, thereby leading to its decomposition
followed by a predominant strand cleavage at the DMAC site. This remarkable property suggests that the
modified cytosine DMAC can function as an efficient hole-trapping site in the positive-charge transfer in
DNA duplexes.
modified purine bases into DNA can dramatically promote the
Introduction
photosensitizer injected hole transfer efficiency.7,8 These strat-
Photosensitized oxidation of DNA has been studied extensively
in relation to the positive-charge (hole) transfer through DNA.1–3
In general, photosensitized one-electron oxidation of DNA can
form a primary base radical cation, the positive charge of which
migrates through the π-stack of DNA base pairs to produce oxi-
dative cleavage of a DNA strand at specific base with the lowest
oxidation potential such as consecutive guanine (G) sites.3 It has
been demonstrated that a hole generated in DNA can migrate
over a 20–30 nm range under certain conditions.1,4 Hole transfer
through DNA has attracted particular interest in view of potential
application of its properties to gene analysis of mutation and
molecular wires in nanoscale electronic devices.5,6
egies provide useful information for better mechanistic under-
standing as well as the application of hole-transfer to DNA-
based nanomaterials. On the other hand, modified purine bases
such as 7-deazaguanine,9 8-methoxyguanine,10 and 2-amino-7-
deazaadenine11 can thermally trap the hole that was injected and
migrated through the DNA duplex. These modified nucleobases
have also been used as effective tools for investigating long-
range hole transfer through DNA duplexes. Other examples are
N-cyclopropyl-modified purine bases that kinetically trap the
hole migrating along the DNA duplex.12,13 In contrast to various
types of modified purine bases reported so far, there have been
limited examples of chemically modified pyrimidine bases pos-
sessing a hole trapping ability, e.g., N(4)-cyclopropyl-modified
cytosine that kinetically traps holes migrating along the DNA
duplex14 and trimethoxyphenyl-substituted uracil that thermally
traps holes induced by ionizing radiation.15 In this context,
development of an efficient hole-trapping pyrimidine base is an
intriguing area of research for further understanding of the mech-
anism by which hole migrates through a duplex of DNA contain-
ing diverse sequences.
A wide variety of modified nucleobases have been developed
for investigating and promoting the hole transfer reaction in
DNA duplexes.7–15 Recent studies on hole transfer in photosen-
sitizer-tethered DNA duplexes showed that introduction of
aAdvanced Biomedical Engineering Research Unit, Kyoto University,
Katsura, Nishikyo-ku, Kyoto 615-8510, Japan. E-mail: yamada.hisatsu-
gu.3r@kyoto-u.ac.jp; Fax: +81-75-383-2805; Tel: +81-75-383-7554
bDepartment of Energy and Hydrocarbon Chemistry, Graduate School
of Engineering, Kyoto University, Katsura, Kyoto 615-8510, Japan.
E-mail: nishimot@scl.kyoto-u.ac.jp; Fax: +81-75-383-2501;
Tel: +81-75-383-2500
†Electronic supplementary information (ESI) available: synthesis,
Stern–Volmer plots of fluorescence quenching, PAGE analysis, and
ESI-TOF mass analysis. See DOI: 10.1039/c2ob06642d
In this study, we have developed DNA duplexes containing 5-
dimethylaminocytosine (DMAC) as a novel electron-donating
nucleobase and characterized the hole-transfer reactions in DNA
possessing modified base sites of DMAC. Fluorescence quenching
experiments suggested that incorporation of the dimethylamino
group into the C(5) position of cytosine significantly decreases
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2035–2043 | 2035

View Article Online
Synthesis and thermal stability of DMAC-containing ODNs
duplex
the oxidation potential of DMAC because of a strong electron-
donating effect of a dimethylamino group. Gel electrophoresis
analysis revealed that a photoinjected hole migrates through the
DNA bases to be trapped efficiently at the DMAC site, where
enhanced oxidative strand cleavage was induced. Thus, DMAC
can function as an effective hole-trapping site, thereby probing
the hole transfer reaction in DNA.
Scheme 1 shows the synthesis of oligodeoxynucleotides (ODNs)
containing DMAC. 5-Bromo-2′-deoxyuridine protected by a 4,4′-
dimethoxytrityl group (1) was converted to 2 by treatment with
50% dimethylamine. The 3′-hydroxyl group of 2 was then pro-
tected to afford 5-dimethylamino-2′-deoxyuridine derivative (3).
The 4-oxo group of the resulting 3 was transformed to an amino
group to give the DMAC derivative (4) in two steps. N-Acetyl-
protected DMAC (6) was prepared from 4 by standard acetyl pro-
tection of the exocyclic amine and subsequent desilylation of the
hydroxy group, then converted into DMAC phosphoramidite
derivative (7) which was incorporated into DNA using a DNA
synthesizer and conventional phosphoramidite chemistry. Our
choice of the photosensitizer was 9,10-anthraquinone (AQ),
which is well known to induce hole injection to the adjacent
DNA bases and thereafter positive-charge transfer through the
DNA duplex.18 ODNs with AQ sensitizer were similarly pre-
pared from a prescribed phosphoramidite derivative as in pre-
viously reported methods.18b The ODNs used in this study are
summarized in Fig. 1a.
Results and discussion
Electron transfer oxidation property of DMA
C
We evaluated the one-electron oxidation property of DMAC using
a Stern–Volmer analysis of the fluorescence quenching by mono-
meric DMAC (5-dimethylamino-2′-deoxycytidine, dDMAC), which
was prepared from 5-bromo-2′-deoxycytidine with dimethyl-
amine (see ESI†). 9,10-Dicyanoanthracene (DCA) was employed
as a photooxidizing fluorophore for this analysis. It has sufficient
reduction potential (1.95 V versus NHE in the singlet excited
state) to oxidize all of the naturally occurring nucleobases.16a
Similar analysis has been widely used for investigation of the
one-electron oxidation process of DNA bases.16
The thermal stability of DMAC-containing duplexes was evalu-
ated by monitoring the melting temperatures (Tm) in 10 mM
sodium cacodylate buffer (pH 7.0) containing 100 mM NaCl
(Fig. 1b). Introduction of the AQ moiety slightly enhanced the
thermal stability of the duplex because of the π-stacking effect
between the hydrophobic planar ring of AQ and the flanking
Upon electronic excitation of DCA at 390 nm in deoxygenated
buffer solution, the intensity of the fluorescence emission
decreased with increasing concentrations of dDMAC. The fluor-
escence quenching rate constant (kq) of the singlet excited state,
1DCA*, was determined by the Stern–Volmer plots of the fluor-
escence spectral data (see Fig. S1†) in conjunction with the
bases. The Tm of the AQ-ODN1(G)/ODN1(DMAC) duplex (Tm
=
57.0 °C) was only slightly different from that of a reference
duplex AQ-ODN1(G)/ODN1(C) (Tm = 56.9 °C). The circular
dichroism (CD) spectra for each duplex showed a positive peak
at 274 nm and a negative peak at 249 nm (Fig. 1c), indicating
that the global structure of the AQ-ODN1(G)/ODN1(DMAC)
duplex was retained as a characteristic B-form. These results
suggest that the dimethylamino group of DMAC oriented toward
the major groove does not significantly perturb the duplex
structure.
1
fluorescence lifetime of DCA*.17 As summarized in Table 1,
dDMAC showed the highest kq value for the fluorescence quench-
ing of 1DCA* (kq = 6.5 × 109 M−1 s−1) among the 2′-deoxyribo-
nucleosides evaluated: the kq values of 2′-deoxycytidine (dC),
thymidine (dT), 2′-deoxyadenosine (dA), and 2′-deoxyguanosine
(dG) were 2.9, 3.2, 4.6, and 5.6 × 109 M−1 s−1, respectively.
These results strongly suggest that DMAC is more readily one-
electron oxidized into a radical cation intermediate than any
other natural nucleobases. Therefore, DMAC can function as an
electron-donating site, where a hole (radical cation) migrating
through DNA is localized.
Hole trapping at DMAC in the positive-charge transfer through
DNA duplex
We also investigated the hole-transfer and hole-trapping proper-
ties of AQ-tethered DNA duplexes possessing a DMAC base in
the middle spot of the sequence. Hole injection into the duplex
was induced upon photoexcitation of AQ at the 5′-end of the
duplex. The duplex consisting of AQ-ODN1(G) with 32P-labeled
ODN1(DMAC) in sodium cacodylate buffer solution (pH = 7.0)
was photoirradiated (365 nm) at 20 °C. The photoirradiated
samples were treated with hot piperidine and subjected to poly-
acrylamide gel electrophoresis (PAGE). The results are shown in
Fig. 2a and 2b.
Table 1 Fluorescence quenching rate constants (kq) of DCA with 2′-
deoxyribonucleosidesa
2′-Deoxyribonucleosides
dDMA
dG
dA
dT
dC
kq (× 109 M−1 −1 b
s )
C
6.5
5.6
4.6
3.2
2.9
a The excitation wavelength for DCA (25 μM) was 390 nm. Relative
intensity of the emission band at 487 nm was measured in the presence
of 25 μM DCA, in deoxygenated solution of 20 mM phosphate buffer
(pH 7.0). b The dynamic quenching rate constant kq was determined by
the Stern–Volmer equation: I0/I = 1 + kqτ0[2′-deoxyribonucleoside],
where I0/I is the ratio of the emission intensities in the absence and
presence of 2′-deoxyribonucleoside and τ0 is the lifetime of the singlet
excited state of DCA (15.1 ns)17 in the absence of quencher.
A predominant strand cleavage at the DMAC site was observed
in the ODN1(DMAC)/AQ-ODN1(G) duplex, whereas relatively
weak strand cleavage occurred at the 5′-G14 site of a G-doublet,
which is well known as an efficient hole-trapping site3 (Fig. 2a,
lanes 2–4). The cleavage yields at the DMAC site and G14 site
were 21% and 10%, respectively, after 20-min photoirradiation
(Fig. 2b). No strand cleavage band was observed at the DMAC
2036 | Org. Biomol. Chem., 2012, 10, 2035–2043
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 1 Reagents and conditions: a) 50% aqueous dimethylamine, 60 °C, 60 h, 52%; b) tert-butyldimethylsilylchloride, imidazole, DMF, r.t., 6 h,
93%; c) 2,4,6-triisopropylbenzenesulfonyl chloride, DMAP, Et3N, CH3CN, r.t., 16 h, and then 28% aqueous NH3, 0 °C to r.t., 3 h, 57%; d) acetic
anhydride, DMAP, pyridine, r.t., 12 h, 92%; e) TBAF, THF, r.t., 1 h, 80%; f) N,N-diisopropylmethylphosphonamidic chloride, DIPEA, CH3CN, r.t.,
4 h, quant. g) automated DNA synthesis.
site in a control duplex without an AQ (Fig. 2a, lanes 6–9), indi-
cating that strand cleavage at the DMAC due to possible direct
photolysis is negligible under the present conditions.19 In the
absence of piperidine treatment after photoirradiation, the strand
cleavage was significantly suppressed (Fig. 2a, lane 5). These
results clearly indicate that a given DNA base radical cation
(hole) induced by the excited AQ migrates through the DNA
duplex to be trapped efficiently at the DMAC base rather than the
5′-G site of a G-doublet. Similarly, separate experiments using
the ODN2(X1/X2/X3)/AQ-ODN2(Y1/Y2/Y3) duplexes showed
that the cleavage efficiency of the DMAC sites in ODN2(T/DMAC/
T)/AQ-ODN2(A/G/A) were significantly higher than that of the
corresponding 5′-G site in ODN2(T/G/G)/AQ-ODN2(C/C/A)
(see Fig. S2†) and that of the middle of the G triplet in ODN2
(G/G/G)/AQ-ODN2(C/C/C) (see Fig. S3†) arranged at the same
distance from AQ.
relative to the AQ-ODN1(G) duplex. As shown in Fig. 2c, clea-
vage efficiencies at DMAC sites in AQ-ODN1(I)/ODN1(DMAC)
were virtually identical to that in the AQ-ODN1(G)/ODN1
(
DMAC), indicating that the oxidation potentials of complemen-
tary bases have little effect on the strand cleavage at DMAC. In
light of the Stern–Volmer analysis data (Table 1), this suggests
that a photoinjected and migrated hole may be localized at the
DMAC site rather than its complementary base, which further sup-
ports the idea that DMAC can function as an efficient electron
donor for hole transfer through DNA duplexes.
LC/ESI-TOF mass analysis in the one-electron photooxidation
of dDMAC to get a mechanistic insight into photooxidation and
strand cleavage of DMA
C
It has been suggested that the hole transfer through DNA
involves a short-range charge-transfer process between G or A
bases, which are known to have the highest HOMO (highest
occupied molecular orbital) energy levels among the naturally
occurring nucleobases.1–3 In this context, competitive localiz-
ation of the migrating hole at the complementary base is possibly
involved in the hole trapping at DMAC. In addition, Kawai and
co-workers have reported that the oxidation potential of G can
be regulated by introducing a substituent to cytosine on the
complementary strand.20 In this view, we examined the effects of
complementary bases on the strand cleavage efficiency at
DMAC. Hypoxanthine (I), which has a higher oxidation potential
(Eox = 1.50 V versus NHE)21 than that of G (Eox = 1.29 V
versus NHE),22 was introduced into the complementary strand
(AQ-ODN1(I)). The cleavage efficiency at DMAC was compared
For a mechanistic understanding of the one-electron oxidation
and piperidine-induced strand cleavage at DMAC, we carried out
the LC/ESI-TOF mass analysis in the one-electron oxidation of
dDMAC induced by photoexcited riboflavin, which has been
reported to predominantly sensitize a Type I photooxidation.23
Aerobic solutions of dDMAC (200 μM) and riboflavin (200 μM)
in 2 mM sodium cacodylate buffer (pH 7.0) containing 10 mM
NaCl were photoirradiated with 365 nm UV light, and the reac-
tion was analyzed by LC/ESI–MS. The results are shown in
Fig. 3. After 2 min irradiation, the HPLC profile exhibited two
major peaks at 1.9 and 3.4 min along with the degradation of
dDMAC (Fig. 3a). Upon prolonged photoirradiation (ca. 10 min),
dDMAC was completely degraded, while the peak intensity at
3.4 min decreased with an increase in the peak at 1.9 min. The
ESI–MS analyses of these characteristic LC peaks indicates
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2035–2043 | 2037

View Article Online
Fig. 2 (a) PAGE image of photoirradiated AQ-ODN1(G)/ODN1
(
DMAC) (lanes 2–5) and c-ODN1/ODN1(DMAC) (lanes 6–9). ODN
duplexes in 10 mM sodium cacodylate buffer (pH 7.0) containing
100 mM NaCl were photoirradiated (365 nm, 0–20 min) at 20 °C, fol-
lowed by piperidine treatment (90 °C, 20 min). The samples in lanes 5
and 9 were not treated with hot piperidine. G+A indicates Maxam–
Gilbert sequencing lane (lane 1). (b) Strand cleavage yields at the DMAC
site (closed circle) and the G14 site (closed square) in photoirradiated
AQ-ODN1(G)/ODN1(DMAC) estimated by densitometric analysis of the
gel image. (c) Effects of the complementary base (Y = G or I in
AQ-ODN1(Y)) on the strand cleavage yields of DMAC were determined
by PAGE. AQ-ODN1(G)/ODN1(DMAC) (open circle) and AQ-ODN1(I)/
ODN1(DMAC) (open square) were photoirradiated under conditions that
otherwise were identical to those described above. Cleavage yields were
calculated from the corresponding band intensities relative to the total
intensity. Error bars represent the standard deviation of three independent
measurements.
Fig. 1 (a) Sequences and structures of ODNs used in this study. DMAC,
AmC, MC, BrC, and I denote 5-dimethylaminocytosine, 5-aminocytosine,
5-methylcytosine, 5-bromocytosine, and hypoxanthine, respectively. (b)
Normalized UV melting curve of the duplexes (1 μM strand concen-
tration) measured at 260 nm in 10 mM sodium cacodylate buffer (pH
7.0) containing 100 mM NaCl: AQ-ODN1(G)/ODN1(DMAC) (solid
line), AQ-ODN1(G)/ODN1(C) (dotted line), and c-ODN1/ODN1
(
DMAC) (dashed line). (c) CD spectra (5 μM strand concentration) of
Cadet and co-workers have reported that 2′-deoxyribofuranosyl-
5-hydroxy-5-methylhydantoin was formed from the hydrolytic
decomposition of 6-hydroxy-5-hydroperoxy-5,6-dihydrothymi-
dine (one of the oxidation products of dT) through the opening
of the 1,6-pyrimidine ring.26 In this light, the resulting C(5)-
hydroperoxide intermediate of DMAC may undergo hydrolytic
decomposition to give an imidazolone analogue (Imz), a related
degradation product to 5-hydroxy-5-methylhydantoin, as in the
case of dT photooxidation. Another pathway could also be con-
sidered: an N-formylurea analogue (Nfu) is likely to be generated
by the dispropotionation of C(5)-peroxyl radical intermediate of
DMAC.29 The resulting Nfu may undergo hydrolysis of the
formyl group and intramolecular cyclization to give an Imz ana-
logue. However, the occurrence of the bi-molecular processes
may be ruled out in the AQ-photosensitized oxidation of the
DMAC site in the restricted structure of the DNA duplex.
AQ-ODN1(G)/ODN1(DMAC) (solid line) and AQ-ODN1(G)/ODN1(C)
(dashed line) observed at 25 °C in 10 mM sodium cacodylate buffer (pH
7.0) containing 100 mM NaCl.
transient formation of unstable N-formylurea derivative (dNfu)
(m/z 301.1172; calcd. for [M − H]− = 301.1148) (Fig. 3b,
bottom) and its intramolecularly cyclized imidazolone derivative
(dImz) (m/z 271.1223, 309.0989; calcd. for [M − H]−
=
273.1199, [M + Cl]− = 309.0971) (Fig. 3b, top).24,25
Scheme 2 shows a plausible mechanism for one-electron oxi-
dation and strand cleavage at DMAC. The primary step involves
one-electron oxidation of DMAC into the corresponding radical
cation (DMAC˙+) intermediate by way of AQ-injected hole trans-
fer through DNA and subsequent hole trapping at the DMAC site.
By reference to the proposed reaction mechanisms for one-elec-
tron oxidation of thymidine26 and 2′-deoxycytidine,27 the result-
ing DMAC˙+ may predominantly undergo specific hydration at
C(6) to give a C(5)-yl radical intermediate, followed by addition
of molecular oxygen at the C(5)-yl radical to afford C(5)-peroxyl
radical intermediate that is converted to a C(5)-hydroperoxide.28
An attempt was also made to investigate the photooxidation of
the AQ-ODN1(G)/ODN1(DMAC) duplex using ESI-TOF mass
spectroscopy. The duplex (10 μM) in sodium cacodylate buffer
solution (pH = 7.0) was photoirradiated (365 nm) at 20 °C for
20 min, and then subjected to ESI-TOF mass analysis. As shown
in Fig. S4†, the mass spectral peak of m/z 1554.5 was observed,
2038 | Org. Biomol. Chem., 2012, 10, 2035–2043
This journal is © The Royal Society of Chemistry 2012

View Article Online
(BrC) according to previously reported methods.31 Duplexes con-
sisting of AQ-ODN1(G) with 32P-labeled ODN1(X) (X = DMAC,
AmC, 5-methylcytosine (MC), normal C, and BrC) were photoirra-
diated, treated with hot piperidine, and then subjected to PAGE
under the conditions described above. As shown in Fig. 4, strand
cleavage at the G14 site was observed in each photoirradiated
ODN, whereas the band intensities in ODN1(DMAC) and ODN1
(
AmC) were slightly lower than those observed in other ODNs.
Intense strand cleavages at DMAC and AmC in ODN1(DMAC) and
ODN1(AmC), respectively, were observed after 20-min
irradiation; the corresponding cleavage band intensities were of a
background level at the MC in ODN1(MC), normal C in ODN1
(C), and BrC in ODN(BrC) (Fig. 4, lanes 5, 9, 13, 17, and 21).
These results clearly indicate that DMAC as well as AmC can
efficiently trap the AQ-photoinjected holes in the positive-charge
transfer through DNA duplexes. Incorporation of a strongly elec-
tron-donating group such as dimethylamino and amino groups,
but not the methyl group, at the C(5) position of cytosine may
substantially reduce oxidation potentials of the modified cyto-
sines,32 thereby leading to efficient hole trapping and subsequent
strand cleavage at these sites. Considering the evidence that Am
C
itself was slightly degraded under the conditions of hot piper-
idine treatment (a basic condition at high temperature) (Fig. 4,
lanes 2 and 6), the higher yield of strand cleavage at DMAC rela-
tive toAmC is most likely to originate from hole-transfer reaction
and hole-trapping process in DNA. Thus, the DMAC is potentially
applicable to an efficient molecular probe for investigation of the
positive-charge transfer though DNA using gel electrophoresis.
Conclusions
In summary, we have characterized the AQ photosensitizer-
injected hole transfer through DNA duplexes containing a DMAC
base that is more readily one-electron oxidized into a radical
cation intermediate than the other natural nucleobases. The AQ-
sensitized photooxidation of duplexes containing DMAC demon-
strated that photoinjected holes can migrate through the DNA
bases to be trapped efficiently at the DMAC site, resulting in an
enhanced oxidative strand cleavage at the DMAC site rather than
at consecutive G sites, which would normally be a hotspot for
the photosensitized oxidation. The DMAC radical cation inter-
mediate formed by hole trapping may undergo specific hydration
at C(6) and subsequent addition of molecular oxygen at C(5),
leading to its degradation followed by strand cleavage at the
DMAC site. This remarkable hole trapping and selective strand
cleavage property of DMAC has prompted us to investigate hole-
transfer mechanisms in more detail for higher-order DNA struc-
tures such as i-motifs consisting of cytosine-rich sequences.
Further work along these lines is currently under way in our
laboratories.
Fig. 3 LC/ESI–MS (negative mode) profiles of the photooxidation of
DMAC (200 μM) sensitized by riboflavin (200 μM) in 2 mM sodium
d
cacodylate buffer (pH 7.0) containing 10 mM NaCl upon 365 nm photo-
irradiation at 0 °C. (a) Representative HPLC profiles of the reaction
mixture after 2 min photoirradiation. The column eluents were moni-
tored by UV absorbance at 260 nm. (b) LC/MS traces of the peaks at
1.9 min (top) and 3.4 min (bottom).
indicating the formation of oxidized ODN1(Imz) (calcd. for
[M − 4H]4− = 1554.7). Irrespective of the reaction mechanisms,
it is likely that a hole-trapped intermediate DMAC˙+ may be con-
verted into an alkali-labile final product such as the Imz ana-
logue,30 via the hydrolytic decomposition of C(5)-hydroperoxide
intermediate and/or the intramolecular cyclization of a transient
Nfu species, resulting in a predominant oxidative strand cleavage
at the DMAC site. Further detailed product analysis of the hole
trapping reaction of DMAC in single-stranded DNA would be
necessary for better understanding of the reactivity and detailed
reaction mechanisms.
C(5)-substituent effects on the hole-trapping efficiency
Experimental section
We examined the electronic substituent effects on the hole-trap-
ping ability of C(5)-modified cytosines. As an electronic substi-
tuent incorporated into cytosine at C(5), we focused on four
typical types of dimethylamino, amino, methyl, and bromo
groups. The ODN containing 5-aminocytosine (AmC) was syn-
thesized from the corresponding ODN bearing 5-bromocytosine
General
The reagents and solvents used were purchased from standard
suppliers and used without further purification. The reagents for
the DNA synthesizer were purchased from Glen Research.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2035–2043 | 2039

View Article Online
Scheme 2 A plausible photochemical reaction mechanism for exclusive strand cleavage at DMAC.
trihydroxyacetophenone as the matrix. ESI–TOF mass spectra
were obtained on a Bruker Daltonics micrOTOF focus-KE.
5′-(4,4′-Dimethoxytrityl)-5-dimethylamino-2′-deoxyuridine (2)
5′-(Dimethoxytrityl)-5-bromo-2′-deoxyuridine (1) was prepared
as described elsewhere. A solution of 1 (3.66 g, 6.00 mmol) in
10 mL of 50% aqueous dimethylamine was sealed in a 10 mL
vial and heated for 60 h at 60 °C. The residue was evaporated
under reduced pressure. The crude product was purified by
column chromatography (SiO2, 100% ethyl acetate) to give 2
1
(1.79 g, 52%) as a white foam: H NMR (CDCl3, 400 MHz) δ
8.34 (s, 1H), 7.34–7.09 (9H), 6.76–6.75 (4H), 6.30 (t, 1H, J =
6.6 Hz), 4.42 (t, 1H, J = 2.9 Hz), 3.97 (d, 1H, J = 3.4 Hz), 3.71
(s, 6H), 3.36 (dd, 1H, J = 4.4, 10.4 Hz), 3.28 (dd, 1H, J = 3.2,
10 Hz), 2.34 (s, 6H), 2.29–2.14 (2H); 13C NMR (CDCl3,
400 MHz) δ 158.8, 149.0, 144.4, 139.5, 131.0, 130.2, 129.2,
128.9, 128.2, 127.9, 127.1, 113.3, 113.2, 85.5, 84.6, 72.6, 63.6,
55.3, 42.2, 38.8; FABMS m/z 574 [(M + H)+]; HRMS calcd. for
C32H36N3O7 574.2553, found 574.2557.
Fig. 4 Substituent effects on the hole-trapping induced strand cleavage
at C(5)-modified cytosine. PAGE image of photoirradiated AQ-ODN1
(G)/ODN1(X) [X = DMAC (lanes 2–5), AmC (lanes 6–9), MC (lanes
10–13), C (lanes 14–17), and BrC (lanes 18–21)]. ODN duplexes were
photoirradiated (0–20 min), followed by piperidine treatment under con-
ditions identical to Fig. 2a. G+A indicates Maxam–Gilbert sequencing
lane (lane 1).
3′-O-(tert-Butyldimethylsilyl)-5′-O-(4,4′-dimethoxytrityl)-5-
dimethylamino-2′-deoxyuridine (3)
Complementary ODNs were purchased from Invitrogen unless
otherwise noted. Monomeric DMAC (dDMAC) was prepared from
5-bromo-2′-deoxycytidine as detailed in ESI.† The NMR spectra
were obtained on a JEOL JMN-AL-300 (300 MHz), a JEOL
JMN-EX-400 (400 MHz), or a JEOL JMN-ECX-400 (400 MHz)
spectrometer. J values are given in Hz. The FAB mass spectra
were recorded on a JEOL JMS-SX102A spectrometer, using a
nitrobenzyl alcohol or glycerol matrix. Matrix-assisted laser des-
orption ionization time-of-flight (MALDI–TOF) mass spec-
trometry of ODNs was performed on a JEOL LMS-ELITE
A solution of 2 (1.70 g, 3.00 mmol), tert-butyldimethylsi-
lylchloride (900 mg, 5.97 mmol), and imidazole (820 mg,
12.0 mmol) in DMF (7 mL) was stirred for 6 h at room tempera-
ture. The reaction mixture was quenched by water and extracted
with ethyl acetate. The organic layer was washed with brine,
dried over Na2SO4, filtered, and concentrated. The crude product
was purified by column chromatography (SiO2, 100% ethyl
acetate) to give 3 (1.95 g, 94%) as a white solid: 1H NMR
(CDCl3, 400 MHz) δ 8.45 (s, 1H), 7.33–7.13 (9H), 6.74–6.71
MALDI–TOF
mass
spectrometer
with
2′,3′,4′-
2040 | Org. Biomol. Chem., 2012, 10, 2035–2043
This journal is © The Royal Society of Chemistry 2012

View Article Online
5′-O-(4,4′-Dimethoxytrityl)-N-acetyl-5-dimethylamino-2′-
deoxycytidine (6)
(4H), 6.26 (dd, 1H, J = 5.9, 6.0 Hz), 4.34 (t, 1H, J = 3.2 Hz),
3.88 (dd, 1H, J = 3.2, 6.8 Hz), 3.69 (s, 6H), 3.32 (dd, 1H, J =
3.4, 10.2 Hz), 3.12 (dd, 1H, J = 3,8, 10.6 Hz), 2.35 (s, 6H),
2.20–2.15 (1H), 2.05–1.98 (1H), 0.77 (s, 9H), −0.04 (6H); 13C
NMR (CDCl3, 400 MHz) δ 158.6, 149.0, 144.4, 135.5, 130.0,
129.1, 128.1, 127.9, 127.0, 113.2, 113.1, 86.4, 72.2, 62.9, 55.2,
42.3, 40.7, 25.6, 17.9, −4.7, −4.9; FABMS m/z 688 [(M+H)+];
HRMS calcd. for C38H50N3O7Si 688.3418, found 688.3412.
Tetrabutyl ammonium fluoride (1.0 M, 3.5 mL, 3.5 mmol) was
added to a solution of 5 (1.28 g, 1.76 mmol) in dry THF
(24 mL) and the mixture was stirred at room temperature for 1 h.
The reaction mixture was diluted with water and extracted with
ethyl acetate. The organic layer was washed with brine, dried
over Na2SO4, filtered, and concentrated. The crude product was
purified by column chromatography (SiO2, 5% methanol–chloro-
form) to give 6 (887 mg, 82%) as a yellow solid: 1H NMR
((CD3)2O, 400 MHz) δ 7.68 (s, 1H), 7.37–7.13 (9H), 6.74–6.71
(4H), 6.28 (t, 1H, J = 7.8 Hz), 4.49 (1H), 4.12 (1H), 3.79 (s,
6H), 3.43 (dd, 1H, J = 3.6, 10.4 Hz), 3.24 (dd, 1H, J = 3.4, 10.2
Hz), 2.53 (s, 3H), 2.44–2.41 (1H), 2.38 (s, 6H), 2.23–2.20 (1H);
13C NMR (CDCl3, 400 MHz) δ 158.7, 144.3, 135.5, 135.3,
130.1, 130.0, 128.1, 127.9, 127.1, 113.2, 113.1, 86.9, 86.5, 86.4,
72.3, 63.4, 55.2, 44.8, 41.9, 26.3, 25.6; FABMS m/z 615
[(M+H)+]; HRMS calcd. for C34H39N4O7 615.2819, found
615.2821.
3′-O-(tert-Butyldimethylsilyl)-5′-O-(4,4′-dimethoxytrityl)-5-
dimethylamino-2′-deoxycytidine (4)
A solution of 3 (1.94 g, 2.82 mmol), triethylamine (1.2 mL,
8.62 mmol), 4-dimethylaminopyridine (1.04 g, 8.46 mmol), and
2,4,6-triisopropylbenzenesulfonyl chloride (2.56 g, 8.46 mmol)
in anhydrous acetonitrile (24 mL) was stirred at room tempera-
ture for 16 h. The mixture was cooled in an ice bath. Aqueous
ammonia (28%, 35 mL) was added, and the mixture was stirred
for 3 h at ambient temperature. The reaction mixture was concen-
trated under reduced pressure and was taken up in ethyl acetate.
The organic layer was washed with brine, dried over Na2SO4,
filtered, and concentrated. The crude product was purified by
column chromatography (SiO2, 4% methanol–chloroform) to
give 4 (1.87 g, 97%) as a white solid: 1H NMR (CDCl3,
400 MHz) δ 7.55 (s, 1H), 7.54–7.23 (9H), 6.92–6.89 (4H), 6.46
(t, 1H, J = 6.8 Hz), 4.46 (dd, 1H, J = 6.6, 7.9 Hz), 4.05 (dd, 1H,
J = 3.0, 7.0 Hz), 3.87 (s, 6H), 3.57 (dd, 1H, J = 3.2, 10.8 Hz),
3.26 (dd, 1H, J = 3.2, 10.8 Hz), 2.55–2.49 (1H), 2.38 (s, 6H),
2.17–2.10 (1H), 0.89 (s, 9H), 0.03 (6H); 13C NMR (CDCl3,
400 MHz) δ 163.4, 158.6, 144.4, 135.7, 135.5, 130.1, 129.8,
128.2, 127.9, 127.0, 113.2, 113.1, 86.3, 85.9, 72.0, 62.7, 55.2,
44.2, 42.0, 25.7, 17.9, −4.7, −5.0; FABMS m/z 687 [(M+H)+];
HRMS calcd. for C38H51N4O6Si 687.3578, found 687.3580.
3′-(N,N-Diisopropylmethylphosphonamidite)-5′-O-(4,4′-
dimethoxytrityl)-N-acetyl-5-dimethylamino-2′-deoxycytidine (7)
Compound 6 was coevaporated twice with anhydrous acetonitrile
(1 mL). A solution of 6 (20 mg, 0.033 mmol), anhydrous diiso-
propylethylamine (16 μL, 0.10 mmol), and N,N-diisopropyl-
methylphosphonamidic chloride (7 μL, 0.033 mmol) in
anhydrous acetonitrile (350 μL) was stirred for 4 h at room temp-
erature. The reaction mixture was filtered and then placed on a
DNA synthesizer.
Synthesis of ODNs
ODNs containing 5-dimethylaminocytosine were synthesized on
an Applied Biosystems Model 3400 DNA/RNA synthesizer
using standard phosphoramidite chemistry. ODNs containing the
AQ sensitizer were prepared as in previously reported meth-
ods.18b ODN containing 5-aminocytosine was synthesized from
ODN1(BrC) (including 5-bromocytosine) as previously
reported.31 Synthesized ODNs were purified by reversed phase
HPLC on an Inertsil ODS-3 (ϕ 10 mm × 250 mm, elution with a
solvent mixture of 0.1 M triethylammonium acetate (TEAA) pH
7.0, linear gradient over 60 min from 0% to 30% acetonitrile at a
flow rate 3.0 mL min−1). The purity and concentration of
modified ODN were determined by complete digestion with calf
intestinal alkaline phosphatase, nuclease P1, and phosphodiester-
ase I. The synthesized ODNs were identified by MALDI–TOF
mass spectrometry (for ODN1(DMAC), AQ-ODN1(G),
AQ-ODN1(I), and ODN2(T/DMAC/T)) or ESI–TOF mass spec-
trometry (for ODN1(AmC), ODN1(MC), ODN1(BrC), AQ-ODN2
(A/G/A), AQ-ODN2(C/C/A), and AQ-ODN2(C/C/C)); ODN1
3′-O-(tert-Butyldimethylsilyl)-5′-O-(4,4′-dimethoxytrityl)-N-
acetyl-5-dimethylamino-2′-deoxycytidine (5)
A solution of 4 (1.87 g, 2.73 mmol), acetic anhydride (645 μL,
6.83 mmol), and 4-dimethylaminopyridine (167 mg, 1.37 mmol)
in anhydrous pyridine (27 mL) was stirred for 12 h at room
temperature. The reaction mixture was concentrated under
reduced pressure, diluted with water, and extracted with ethyl
acetate. The organic layer was washed with brine, dried over
Na2SO4, filtered, and concentrated. The crude product was
purified by column chromatography (SiO2, 50% ethyl acetate–
1
hexane) to give 5 (1.29 g, 65%) as a yellow solid: H NMR
(CDCl3, 400 MHz) δ 7.82 (s, 1H), 7.43–7.24 (9H), 6.84–6.81
(4H), 6.26 (t, 1H, J = 6.6 Hz), 4.40 (dd, 1H, J = 3,4, 7.0 Hz),
4.02 (dd, 1H, J = 3.6, 6.8 Hz), 3.79 (s, 6H), 3.54 (dd, 1H, J =
3.2, 11.6 Hz), 3.22 (dd, 1H, J = 3.4, 10.6 Hz), 2.71 (s, 3H),
2.58–2.53 (1H), 2.29 (s, 6H), 2.12–2.05 (1H), 0.82 (s, 9H),
−0.03 (6H); 13C NMR (CDCl3, 400 MHz) δ 158.7, 144.3,
135.5, 135.3, 130.1, 130.0, 129.1, 128.2, 127.9, 127.1, 113.2,
113.1, 86.7, 86.4, 71.7, 62.4, 44.8, 42.1, 25.7, 17.9, 4.7, −5.0;
FABMS m/z 729 [(M+H)+]; HRMS calcd. for C40H53N4O7Si
729.3684, found 729.3701.
(
(
DMAC) m/z 6241.48 (calcd. for [M − H]− 6240.15), ODN1
AmC) m/z 886.5 (calcd. for [M − 7H]7− 886.6) ODN1(MC) m/z
1034.2 (calcd. for [M − 6H]6− 1034.3), ODN1(BrC) m/z 1045.2
(calcd. for [M − 6H]6− 1045.1), AQ-ODN1(G) m/z 6390.21
(calcd. for [M − H]− 6389.17), AQ-ODN1(I) m/z 6375.04
(calcd. for [M − H]− 6374.05), ODN2(T/DMAC/T) m/z 5512.55
(calcd. for [M − H]− 5514.66), AQ-ODN2(A/G/A) m/z 1468.3
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2035–2043 | 2041

View Article Online
(calcd. for [M − 4H]4− 1468.4), AQ-ODN2(C/C/A) m/z 1452.3
(calcd. for [M − 4H]4− 1452.4), AQ-ODN2(C/C/C) m/z 1156.8
(calcd. for [M − 5H]5− 1156.9).
at 260 nm. The solvent mixture of triethylammonium acetate
buffer solution (0.1 M, pH 7.0) containing 5 vol% acetonitrile
was delivered as the mobile phase.
Fluorescence quenching
Melting temperature (Tm) of hybridized ODNs
Quenching experiments of the fluorescence of the photosensiti-
zer were carried out on a Shimadzu RF-5300PC spectro-
photometer. The excitation wavelength for DCA (25 μM) was
390 nm and the monitoring wavelength was that corresponding
to the respective emission band at 487 nm. The dynamic quench-
ing rate constant kq was determined by the Stern–Volmer
equation: I0/I = 1 + kqτ0[2′-deoxyribonucleoside] where I0/I is
the ratio of the emission intensities in the absence and presence
of 2′-deoxyribonucleoside and τ0 is the lifetime of the singlet
excited state of DCA (15.1 ns) in the absence of quencher.
1 μM of appropriate ODNs was dissolved in 10 mM sodium
cacodylate buffer (pH 7.0) containing 100 mM NaCl. UV
melting curves were recorded on a JASCO-V630 spectropho-
tometer equipped with a multicell block and a Peltier tempera-
ture controller. Melting curves were obtained by monitoring the
UV absorbance at 260 nm with the temperature being increased
from 4 °C to 80 °C at a rate of 1 °C min−1
.
CD spectrum of hybridized ODNs
5 μM duplexes were dissolved in 10 mM sodium cacodylate
buffer (pH 7.0) containing 100 mM NaCl. CD spectra of the sol-
utions were recorded at 20 °C on a JASCO J-700 spectropho-
tometer, using a UV cell with 0.1 cm path length.
Photooxidative cleavage reaction and PAGE analysis
ODNs were 5′-32P-labeled by phosphorylation with 4 μL of
[γ-32P]ATP (PerkinElmer) and 4 μL of T4 polynucleotide kinase
(Nippon Gene). The reaction mixtures were purified using QIA-
quick Nucleotide Removal Kit (QIAGEN) to remove excess
unincorporated nucleotide. Complementary ODNs (1 μM) were
annealed in 10 mM sodium cacodylate buffer (pH 7.0) contain-
ing 100 mM NaCl by heating to 90 °C, followed by slow
cooling to room temperature. The samples were irradiated at
365 nm UV light with a transilluminator (FTI-20L, Funakoshi,
Tokyo) at 20 °C on exposure to air. After irradiation, the DNA
samples were precipitated by adding 10 μL of herring sperm
DNA (1 mg mL−1), 10 μL of 3 M sodium acetate and 800 μL of
ethanol. The precipitated DNA was resolved in 50 μL of 10%
piperidine (v/v), heated at 90 °C for 20 min, and concentrated.
The radioactivity of samples was assayed using an Aloka 1000
liquid scintillation counter and the dried DNA pellets were resus-
pended in loading buffer (a solution of 80% formamide (v/v),
1 mM EDTA, 0.1% xylene cyanol, and 0.1% bromophenol
blue). All reactions, along with Maxam–Gilbert G+A sequencing
reactions, were heat denatured at 90 °C for 3 min and quickly
chilled on ice. The samples (3–10 × 103 cpm) were loaded onto
20% polyacrylamide/7 M urea sequencing gels and electrophor-
esed at 1900 V for 60–90 min, transferred to a cassette, and
stored at −80 °C with Fuji X-ray film (RX-U). The gels were
analyzed using autoradiography with ATTO CS Analyzer
(version 3.0). Individual yields were calculated relative to each
total band intensity. The average of three independent measure-
ments for each sample is indicated.
Acknowledgements
This research was supported financially by Grant-in-Aid for
Young Scientists (B) from Japan Society for the Promotion of
Science. We thank Prof. Teruyuki Kondo and Dr Yu Kimura
(Advanced Biomedical Engineering Research Unit, Kyoto Uni-
versity) for ESI-TOF mass measurements.
Notes and references
1 (a) J. C. Genereux and J. K. Barton, Chem. Rev., 2010, 110, 1642–1662;
(b) G. B. Schuster, Acc. Chem. Res., 2000, 33, 253–260; (c) B. Giese,
Acc. Chem. Res., 2000, 33, 631–636; (d) F. D. Lewis, R. L. Letsinger and
M. R. Wasielewski, Acc. Chem. Res., 2001, 34, 159–170.
2 (a) Long-Range Charge Transfer in DNA I (Topics in Current Chemistry
Vol. 236), ed. G. B. Schuster, Springer-Verlag, New York, 2004;
(b) Long-Range Charge Transfer in DNA II (Topics in Current Chemistry
Vol. 237), ed. G. B. Schuster, Springer-Verlag, New York, 2004;
(c) Charge Transfer in DNA, ed. H.-A. Wagenknecht, Wiley-VCH, Wein-
heim, 2005.
3 (a) I. Saito, T. Nakamura, K. Nakatani, Y. Yoshioka, K. Yamaguchi and
H. Sugiyama, J. Am. Chem. Soc., 1998, 120, 12686–12687;
(b) H. Sugiyama and I. Saito, J. Am. Chem. Soc., 1996, 118, 7063–7068;
(c) V. Shafirovich, A. Dourandin and N. E. Geacintov, J. Phys. Chem. B,
2001, 105, 8431–8435; (d) V. Shafirovich, J. Cadet, D. Gasparutto,
A. Dourandin, W. D. Huang and N. E. Geacintov, J. Phys. Chem. B,
2001, 105, 586–592.
4 J. D. Slinker, N. B. Muren, S. E. Renfrew and J. K. Barton, Nat. Chem.,
2011, 3, 228–233.
5 T. G. Drummond, M. G. Hill and J. K. Barton, Nat. Biotechnol., 2003,
21, 1192–1199 and references therein.
6 A. Okamoto, T. Kamei and I. Saito, J. Am. Chem. Soc., 2006, 128, 658–
662.
7 K. Kawai, H. Kodera, Y. Osakada and T. Majima, Nat. Chem., 2009, 1,
156–159.
8 A. Okamoto, K. Tanaka and I. Saito, J. Am. Chem. Soc., 2003, 125,
5066–5071.
9 (a) S. O. Kelly and J. K. Barton, Chem. Biol., 1998, 5, 413–425;
(b) K. Nakatani, C. Dohno and I. Saito, J. Am. Chem. Soc., 2000, 122,
5893–5894.
10 H. Ikeda and I. Saito, J. Am. Chem. Soc., 1999, 121, 10836–10837.
11 A. Okamoto, K. Tanaka and I. Saito, Bioorg. Med. Chem. Lett., 2002, 12,
3641–3643.
LC/ESI–MS analysis of one-electron photooxidation of
5-dimethylamino-2′-deoxycytidine sensitized by riboflavin
A
solution of 5-dimethylamino-2′-deoxycytidine (dDMAC)
(200 μM) and riboflavin (200 μM) in 2 mM sodium cacodylate
buffer (pH 7.0) containing 20 mM NaCl (total volume 100 μL)
was exposed to 365 nm UV light at 0 °C and subjected to LC/
ESI–MS analysis. LC/ESI–MS (negative mode) was performed
with a Thermo Exactive equipped with a Thermo Accela LC
system. The column eluents were monitored by UV absorbance
2042 | Org. Biomol. Chem., 2012, 10, 2035–2043
This journal is © The Royal Society of Chemistry 2012

View Article Online
12 (a) K. E. Augustyn, J. C. Genereux and J. K. Barton, Angew. Chem., Int.
Ed., 2007, 46, 5731–5733; (b) F. Shao, K. Augustyn and J. K. Barton, J.
Am. Chem. Soc., 2005, 127, 17445–17452.
means of NMR spectroscopy, dImz may exist as the 5R* and 5S* forms,
as in the case of 2′-deoxyribofuranosyl-5-hydroxy-5-methylhydantoin, a
related degradation product to dImz26
.
13 (a) K. Nakatani, C. Dohno and I. Saito, J. Am. Chem. Soc., 2001, 123,
9681–9682; (b) C. Dohno, A. Ogawa, K. Nakatani and I. Saito, J. Am.
Chem. Soc., 2003, 125, 10154–10155.
14 F. Shao, M. A. O’Neill and J. K. Barton, Proc. Natl. Acad. Sci. U. S. A.,
2004, 101, 17914–17919.
25 We could not detect other photooxidation products under the present LC
conditions. At the present stage, however, another possibility cannot be
ruled out that other minor degradation products could be formed.
26 J. R. Wagner, J. E. van Lier, M. Berger and J. Cadet, J. Am. Chem. Soc.,
1994, 116, 2235–2242.
15 H. Ding and M. M. Greenberg, J. Am. Chem. Soc., 2007, 129, 772–773.
16 (a) S. Fukuzumi, H. Miyao, K. Okubo and T. Suenobu, J. Phys. Chem. A,
2005, 109, 3285–3294; (b) H. Yamada, K. Tanabe, T. Ito and
S. Nishimoto, Chem. Eur. J., 2008, 14, 10453–10461.
17 S. L. Murov, I. Carmichael, G. L. Hug, Handbook of Photochemistry,,
2nd edn, Marcel Dekker, New York, 1993.
27 (a) J. R. Wagner, C. Decarroz, M. Berger and J. Cadet, J. Am. Chem.
Soc., 1999, 121, 4102–4110; (b) J. R. Wagner and J. Cadet, Acc. Chem.
Res., 2010, 43, 564–571 and references therein.
28 In addition to predominant hydration at C(6) of DMAC˙+, the minor
pathway could be considered: the DMAC˙+ may undergo competitive
deprotonation of the exocyclic amine group.27 In this context, other piper-
idine-stable oxidation products may be produced along with the formation
of piperidine-labile Imz.
18 (a) B. Armitage, Chem. Rev., 1998, 98, 1171–1200; (b) S. M. Gasper and
G. B. Schuster, J. Am. Chem. Soc., 1997, 119, 12762–12771;
(c) R. P. Fahlman, R. D. Sharma and D. Sen, J. Am. Chem. Soc., 2002,
124, 12477–12485; (d) T. T. Williams, C. Dohno, E. D. A. Stemp and J.
K. Barton, J. Am. Chem. Soc., 2004, 126, 8148–8158; (e) F. D. Lewis, A.
K. Thazhathveetil, T. A. Zeidan, J. Vura-Weis and M. R. Wasielewski,
J. Am. Chem. Soc., 2010, 132, 444–445.
29 C. von Sonntag, Free-Radical-Induced DNA Damage and Its Repair: A
Chemical Perspective, Springer-Verlag, Heidelberg, 2006.
30 The Imz site would be piperidine-labile in view of the fact that similar
imidazolone analogs have been shown to be alkali-sensitive, resulting in
strand cleavage at these sites. See. K. Kino, I. Saito and H. Sugiyama,
J. Am. Chem. Soc., 1998, 120, 7373–7374.
31 (a) E. Ferrer, M. Wiersma, B. Kazimierczak, C. W. Müller and R. Eritja,
Bioconjugate Chem., 1997, 8, 757–761; (b) C. Dohno, A. Okamoto and
I. Saito, J. Am. Chem. Soc., 2005, 127, 16681–16684.
19 Monomeric DMAC has the absorption band at wavelengths shorter than
350 nm. The photoirradiation at 365 nm can selectively excite AQ
sensitizer.
20 K. Kawai, Y. Wata, M. Hara, S. Tojo and T. Majima, J. Am. Chem. Soc.,
2002, 124, 3586–3590.
21 D. J.-F. Chinnapen and D. Sen, J. Mol. Biol., 2007, 365, 1326–1336.
22 S. Steenken and S. V. Jovanovic, J. Am. Chem. Soc., 1997, 119, 617–
618.
23 T. Douki and J. Cadet, Int. J. Radiat. Biol., 1999, 75, 571–581.
24 Although we have not yet accomplished full assignment in the absence of
comprehensive structural information on the photooxidation products by
32 The ionization potentials of DMAC and AmC were estimated to be 8.58
and 8.64 eV, respectively, using the previously reported methods.16b This
indicates that the oxidation potential of DMAC is not much different to
that of AmC. The ionization potentials of DMAC and AmC were estimated
by ab initio calculations at the B3LYP/6-31G* level. Ab initio calcu-
lations were performed for DMAC and AmC derivatives in which the sugar
units were replaced by methyl groups.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2035–2043 | 2043