Photosensitized Thymine Dimerization via Delocalized Triplet Excited States

Article abstract of DOI:10.1002/chem.201502719

A new mechanism of photosensitized formation of thymine (Thy) dimers is proposed, which involves generation of a delocalized triplet excited state as the key step. This is supported by chemical evidence obtained by combining one benzophenone and two Thy u

Full text of DOI:10.1002/chem.201502719

DOI: 10.1002/chem.201502719
Full Paper
&
Dimerization
Photosensitized Thymine Dimerization via Delocalized Triplet
Excited States
[
a]
Paula Miro, Virginie Lhiaubet-Vallet, M. Luisa Marin,* and Miguel A. Miranda*
Abstract: A new mechanism of photosensitized formation of
thymine (Thy) dimers is proposed, which involves generation
of a delocalized triplet excited state as the key step. This is
supported by chemical evidence obtained by combining
one benzophenone and two Thy units with different degrees
of freedom, whereby the photoreactivity is switched from
a clean Paternò–Büchi reaction to a fully chemo-, regio-, and
stereoselective [2 ++ 2] cycloaddition.
Introduction
favored process in solution (triplet energies of 70 and 74 kcal
À1
[5b]
mol for BP and Thy, respectively), it is still feasible at room
temperature through population of the upper vibrational
states. In view of the thermodynamics, the process would be
uphill and therefore expectedly slow, and this allows for alter-
native pathways such as the abovementioned Paternò–Büchi
reaction to give oxetanes. This reaction requires a np* elec-
tronic configuration of the excited state, a requirement that is
fulfilled in the case of BP.
Ultraviolet solar radiation reaching the Earth’s surface has been
[1]
widely reported as a potential mutagenic agent. Nucleobases
are able to absorb UVB light, which causes direct photoreac-
tions from their singlet excited states, such as formation of cy-
clobutane thymine dimers (Thy< >Thy) or (6-4) photoprod-
[
2]
ucts, which are among the most mutagenic alterations. Nano-
second IR spectroscopy has revealed that intrinsically populat-
ed Thy triplets decay in DNA single strands via formation of
biradical intermediates, which can result in cyclobutane
dimers, although they predominantly decay through nonreac-
In fact, previous studies have demonstrated that in solution
3
BP* is efficiently quenched by Thy derivatives (k in the range
q
8
9
À1 À1 [7a,9]
of 10 –10 m s ).
Since the expected rate constant for
[
3]
tive pathways. Although solar UVA light is less efficiently ab-
sorbed by nucleobases, the presence of endogenous or exoge-
nous photosensitizers that can be excited in that region may
greatly enhance UVA-mediated photochemical disorders. More-
over, CASPT2//CASSCF theoretical calculations on the photo-
sensitized process indicated formation of a stabilized triplet ex-
TTET in solution is at least one order of magnitude lower (ac-
cording to the Sandros equation), formation of oxetanes essen-
tially accounts for the determined quenching rate constant.
The intramolecular version of this reaction, in which one mole-
cule of the chromophore is directly attached to the sugar
moiety of thymidine, has also been investigated. Irradiation of
the dyad leads to a mixture of photoproducts, in which the ox-
[
4]
cimer from a parallel-stacked triplet pair of nucleobases.
[
10]
In this context, benzophenone (BP) and drugs containing
the BP chromophore have been widely reported as DNA pho-
etanes are the major ones.
A simplified mechanistic explanation of the reaction be-
[
5]
3
[11]
tosensitizers. It is assumed that BP-photosensitized DNA
damage includes triplet–triplet energy transfer (TTET) from
tween BP* and Thy is summarized in Equations (1)–(3):
3
3
BP* to Thy, followed by [2+2] cycloaddition yielding Thy< >
BP* þ Thy ! oxetanes
ð1Þ
ð2Þ
ð3Þ
[5,6]
Thy.
Formation of these dimers occurs along with the Pater-
3
3
3
3
nò–Büchi reaction between BP* and Thy resulting in the for-
mation of oxetanes, which have been claimed as models to
mimic the highly unstable intermediates involved in the repair
BP* þ Thy ! BP þ Thy*
Thy
*
þ Thy ! Thy <> Thy
[5b,7]
of (6-4) photoproducts.
The rate constant for TTET should be related to the energy
Interestingly, irradiation of BP in the presence of Thy deriva-
tives (2:1 ratio) gives mainly oxetanes, whereas dimers pre-
3
3
[8]
gap between BP* and Thy*. Although it seems a slightly dis-
[
7a,9b]
dominate in the presence of a large excess of Thy.
Al-
though formation of dimers should be dependent on Thy con-
centration according to process (2), the observed changes in
the oxetane-to-dimer ratio are clearly more marked than ex-
pected from kinetic analysis. Moreover, the unambiguous de-
tection of cytosine dimers (Cyt< >Cyt) in photosensitized
DNA (in spite of the high triplet energy of Cyt of ca. 77 kcal
[
a] P. Miro, V. Lhiaubet-Vallet, M. L. Marin, M. A. Miranda
Instituto Universitario Mixto de Tecnología Química (UPV-CSIC)
Universitat Polit›cnica de Val›ncia
Avenida de los Naranjos s/n, 46022 Valencia (Spain)
E-mail: marmarin@qim.upv.es
mmiranda@qim.upv.es
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201502719.
À1 [12]
mol ) suggests that formation of locally excited pyrimidine
Chem. Eur. J. 2015, 21, 17051 – 17056
17051
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
triplet states may not be the only mechanism of dimerization
and that cooperative excited states may play a significant
of the Thy bases. Linking of the carboxylic acid of the lateral
chain with a BP moiety would lead to a fully intramolecular
system in which binary BP/Thy and Thy/Thy interactions (re-
quired in the currently accepted mechanism) are possible; by
contrast, ternary BP/Thy/Thy interactions (involved in the novel
proposal) would be prevented due to geometrical strain. Con-
versely, increasing the degrees of freedom by replacement of
the intramolecular with an intermolecular BP chromophore
should allow assembly of a reactive triplet triplex providing ex-
perimental proof of the concept.
[
13]
role. In this context, excited termolecular complexes (triplex-
es) have previously been proposed to explain the results ob-
tained on photosensitization of some [4+2] and [2+2] cycload-
[
14]
ditions. Although the involvement of triplet triplexes or, in
general, delocalized triplet excited states, has never been
proven in the case of DNA, they could account for most of the
experimental observations on the BP-photosensitized pyrimi-
dine dimerization. This novel hypothesis is illustrated in Equa-
tions (4)–(7).
Results and Discussion
3
3
3
3
3
BP* þ Thy ! ½BP Á Á Á ThyŠ*
ð4Þ
ð5Þ
ð6Þ
ð7Þ
Two analogues with DCA skeleton were synthesized (see
Scheme 1). In one the two Thy units are covalently attached at
positions 3a and 12a and the BP chromophore is esterified at
the lateral chain, and in the other BP was replaced with
a benzyl group, which does not absorb in the UVA region.
Briefly, the developed synthetic strategy outlined in
Scheme 1 started with esterification of the carboxyl group of
DCA with the S enantiomer of reduced ketoprofen (KP-OH) to
give DCAKP. Then, in the presence of an excess of Thy-
CH CO H, the two hydroxyl groups at the 3a and 12a positions
½
½
½
BP Á Á Á ThyŠ* ! oxetanes
3
BP Á Á Á ThyŠ* þ Thy ! ½BP Á Á Á Thy Á Á Á ThyŠ*
BP Á Á Á Thy Á Á Á ThyŠ* ! BP þ Thy <> Thy
3
Thus, quenching of BP* by Thy would lead to oxetanes via
triplet exciplexes, while interaction of the latter with a second
Thy unit would afford triplet triplexes, the immediate precur-
sors of Thy< >Thy.
2
2
were simultaneously esterified to provide 3a,12a-2Thy-DCAKP
1). To prepare the derivative without the BP unit, the carboxyl
To provide experimental evidence supporting the proposed
mechanism, two Thy units have been now tethered to an ap-
propriate bile acid scaffold. For this purpose, deoxycholic acid
(
group at the lateral chain of DCA was treated with benzyl bro-
mide to give DCABn. Subsequent treatment with Thy-CH CO H
yielded 3a,12a-2Thy-DCABn (2).
Laser flash photolysis (LFP) was employed to monitor the
triplet state obtained on 355 nm excitation of the BP chromo-
phore in the presence of two Thy units in the following sys-
tems: 1) the fully intramolecular 1 and 2) the intermolecular
2
2
(
DCA, Scheme 1) was selected as an appropriate skeleton,
since it offers a rigid structure with two hydroxyl groups on
the same face of the molecule allowing covalent attachment
1
:1 mixture of the parent BP and the non-absorbing 2.
Thus, the triplet of 1 was monitored at 520 nm (Figure 1,
top) and compared to that of DCAKP, as a reference com-
pound with BP bound to the bile acid, but lacking the two Thy
units. In this way, intramolecular hydrogen-atom abstraction,
3
a known deactivation pathway for BP* was also considered.
Decays were fitted to a first-order exponential equation and
their lifetimes t were 0.08 and 1.01 ms, respectively. More accu-
rate determination was achieved by means of energy transfer
[
15]
to naphthalene (NPH)
(Figure 1, top, inset), which led to
a value of 0.097 ms. This indicates a fast intramolecular deacti-
6
À1
vation of the triplet, with a rate constant of 9.3”10 s (k=1/
3
t3a,12a-2Thy-DCAKPÀ1/t
). Then, the lifetime of BP* was moni-
DCAKP
tored on addition of increasing concentrations of 2 (Figure 1,
bottom). Stern–Volmer plots of the reciprocal lifetimes ob-
tained in each case allowed determination of the quenching
8
À1 À1
rate constant, which was found to be 3.7”10 m s . For com-
3
parison, quenching of BP* by Thy occurred with k=1.4”
8
À1 À1
1
0 m s , which is reasonably similar when considering the
two units of Thy per molecule in the case of 2.
After completing the photophysical studies, the same sys-
tems were subjected to steady-state photolysis, in order to in-
vestigate their photoreactivity and the nature of the photo-
products. Thus, selective irradiation of the BP chromophore at
lmax =350 nm was monitored through the changes in the UV/
Scheme 1. Developed synthetic strategy to prepare DCA derivatives incorpo-
rating BP or Bn at the lateral chain and two Thy units. Reagents and condi-
tions: i) KP-OH, 4-DMAP, EDC, pyridine; ii) benzyl bromide, DBU, DMF;
iii) Thy-CH
2
COOH, Et
3
N, 2,4,6-trichlorobenzoyl chloride, 4-DMAP, THF.
Chem. Eur. J. 2015, 21, 17051 – 17056
www.chemeurj.org
17052
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. UV/Vis spectra of 1 (top) and BP:2 (bottom) monitored at different
Figure 1. Top: LFP decays monitored at 520 nm of DCAKP (black) and
(gray) in CH CN:H O (4:1). Inset: double-reciprocal plot for energy transfer
from 1 to NPH, revealed through the absorption of NPH* at 415 nm imme-
diately after the laser pulse. Bottom: LFP decays of BP monitored at 520 nm
with increasing concentrations of 2. Inset: Stern–Volmer plot for the quench-
ing of BP by 2 (*) or by Thy (^).
À5
irradiation times (C
i
=4.4”10 m in CH
3
CN:H
2
O (4:1), inert atmosphere,
) recorded at 4.4”10 m.
1
3
2
À5
max =350 nm); UV spectra of BP (*)and Thy (~
Upper insets: photoreaction kinetics of 1 (*) and BP:2 (
3
l
&
) compared to
BP:Thy (1:2) (^). Lower insets: zoom of 320–390 nm region after prolonged
irradiation times of 1 and BP:2 compared to BP (*).
[
5c]
Vis spectra. The results for 1 are shown in Figure 2 (top), and
those for the intermolecular mixture BP:2 in Figure 2 (bottom).
A progressive decrease of the absorbance was observed in
both systems, whereas the control BP:Thy mixture (1:2 ratio)
was almost unreactive (Figure 2, insets). Furthermore, when
the residual UV spectra of 1 and BP:2 after completion of the
reaction were compared to the hypothetical absorption due to
unconsumed BP or Thy (Figure 2), a different reaction pattern
was revealed. In fact, the UV spectrum of 1 at the end of the
reaction was comparable to that of Thy and thus indicated for-
mation of an oxetane; conversely, the UV spectrum of the BP:2
system after irradiation resembled that of BP and thus indicat-
ed formation of Thy< >Thy (see also the difference on zoom-
ing the 300–400 nm region in the insets of Figure 2, top and
bottom). From the initial rates in the linear region, at less than
be 0.03. Indeed, this was confirmed by analysis of the result-
ing photoproduct mixtures.
Thus, 1 was independently irradiated in acetonitrile under
inert atmosphere, and the resulting crude product was purified
by reverse-phase column chromatography. Only one photo-
product was obtained (Figure 3, left), structural characteriza-
1
13
tion of which was achieved on the basis of H and C NMR
13
1
1
1
13
spectroscopy, including C DEPT-135, H– H COSY, and H– C
[
10,16]
HSQC experiments, and also on exact mass determination.
The photoproduct turned out to be a single oxetane derived
from the Paternò–Büchi reaction between the KP moiety and
the Thy unit at the 12a position. This was assessed by the ap-
1
pearance of a new singlet in the H NMR spectrum at
4.72 ppm, with concomitant disappearance of one of the ole-
finic proton signals and with the upfield displacement of the
aromatic signals. Furthermore, on formation of the new ring
1
0% conversion, the reaction quantum yields for Thy< >Thy
13
formation in BP:2 was about three times higher than that of
BP:Thy (1:2). Taking into account the reported data for the
latter system, an upper limit for the value in the former would
the C NMR signal corresponding to the original carbonyl
group at 196.9 ppm shifted to 91.7 ppm, in accordance with
the values reported for analogous oxetanes. The regiochemical
Chem. Eur. J. 2015, 21, 17051 – 17056
www.chemeurj.org
17053
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
production of the oxetane but not for triplex formation. By
contrast, in the BP:2 mixture, the free arrangement of the
ketone relative to the nucleobases allows assembling of the
3
partners in the triplet triplex [BP···Thy···Thy]* with the appro-
priate geometry to afford Thy dimers as the final products.
Conclusion
A new mechanistic pathway leading to photosensitized forma-
tion of cyclobutane pyrimidine dimers is proposed, in which
the key step involves generation of a delocalized triplet excited
state. The concept has been illustrated with systems combin-
ing one BP and two Thy units with different degrees of free-
dom, whereby the photoreactivity can be switched from
a clean Paternò–Büchi to a fully chemo-, regio-, and stereose-
lective [2+2] dimerization. This finding underlines the impor-
tance of cooperative triplet excited states in DNA photodam-
age. Such delocalized chemical entities may predominate over
locally excited triplet states when the thermodynamic require-
ment for energy transfer is not fulfilled.
Figure 3. Photoproducts obtained on irradiation of 1 (left) and BP:2 (right)
in CH CN under N
3
2
.
assignment depicted in Figure 3 was based on the comparison
of the chemical shifts of the two carbon atoms formerly be-
longing to the Thy moiety with those found for related oxe-
[
6,9a]
tanes.
In fact, the obtained values of d =76.3 ppm and
C
dCH =68.4 ppm are in complete agreement with the expected
Experimental Section
values and far from those reported for the alternative re-
gioisomers (d =52.7–57.8 and dCH =80.0–90.0 ppm).
C
Additional experimental procedures, spectra of the new com-
pounds, and details about steady-state and laser flash photolysis
experiments are given in the Supporting Information.
Next, the intermolecular system BP:2 was irradiated in
CH CN under nitrogen. The main photoproduct was isolated
3
by column chromatography (80%) and was found to be
a trans–syn Thy< >Thy (Figure 3, right), the structure of which
was unambiguously established by crystal data (Figure 4).
Synthesis of (S)-KP-OH
BH ·THF (6.600 mL of a 1m solution, 6.60 mmol) was added drop-
3
wise on a stirred solution of (S)-ketoprofen (1.400 g, 5.50 mmol) in
anhydrous THF (14 mL) at À208C, and the reaction mixture was al-
lowed to warm overnight to room temperature. Then, the solution
was cooled to 58C and treated with MeOH:H O (15:85, 17 mL). Af-
2
terwards, the solvents were concentrated in vacuum, the crude
product was redissolved in CH Cl2 and the solution was poured
2
into brine, extracted with CH Cl , washed with NaHCO (5%), dried
2
2
3
over MgSO , and evaporated under reduced pressure. Purification
4
by column chromatography (SiO , CH Cl :MeOH 95:5) gave the (S)-
2
2
2
1
KP-OH as an oil (0.728 g, 55%). H NMR (300 MHz, CDCl ): d=1.31
3
(
(
d, J=7.2 Hz, 3H, CH ), 1.49 (brs, 1H, OH), 3.04 (m, 1H, CH), 3.75
3
13
m, 2H, CH ), 7.40–7.84 ppm (m, 9H, arom); C NMR (75 MHz,
2
CDCl ): d=197.0 (C), 144.5 (C), 137.7 (C), 137.6 (C), 132.5 (CH),
3
1
31.8 (CH), 130.1 (2”CH), 128.9 (CH), 128.5 (CH), 128.4 (CH), 128.3
(
2
2”CH), 68.3 (CH ), 42.3 (CH), 17.6 ppm (CH ); MS: m/z found:
41.1219, calcd for C H O [M+H] : 241.1229.
16 17 2
2
3
+
Figure 4. X-ray crystal structure of the Thy< >Thy 4 resulting from irradia-
tion (lmax =350 nm) of BP:2 in CH CN under N . CCDC 1031373 contains the
3 2
supplementary crystallographic data for this paper. These data are provided
free of charge by The Cambridge Crystallographic Data Centre.
Synthesis of DCAKP
1
1
-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC; 0.348 mL,
.97 mmol) was added dropwise to a stirred mixture of DCA
The obtained results revealed that the photochemical be-
havior of the inter- and intramolecular systems is dramatically
divergent. In principle, the arrangement of the two Thy bases
and a BP unit covalently attached to the same scaffold was ex-
pected to disfavor formation of Thy< >Thy by the triplex
mechanism; in fact, the predominant process was by far pro-
duction of an oxetane. This is clearly related to the evolution
(0.772 g, 1.97 mmol), KP-OH (0.394 g, 1.64 mmol), and 4-dimethyl-
aminopyridine (4-DMAP; 0.200 g, 1.64 mmol) in anhydrous pyridine
(
1
8 mL) at 08C, and the reaction mixture was allowed to react for
h at 08C and then overnight at room temperature. Afterwards,
the suspension was poured into 1m HCl and extracted with
CH Cl ; the combined organic layers were washed with brine, dried
2
2
over MgSO , and concentrated under reduced pressure. Purification
4
by column chromatography (SiO , AcOEt:hexane 60:40) gave
2
1
of the exciplexes depending on the degrees of freedom. In
DCAKP as a colorless solid (0.666 g, 55%). H NMR (300 MHz,
3
1
the geometry of the [BP···Thy]* exciplex is appropriate for
CDCl ): d=0.64 (s, 3H, CH ), 0.90 (s, 3H, CH ), 0.92 (d, J=6.3 Hz,
3
3
3
Chem. Eur. J. 2015, 21, 17051 – 17056
www.chemeurj.org
17054
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3
3
6
7
H, 21-CH ), 1.33 (d, J=7.2 Hz, 3H, KP-CH ), 3.17 (m, 1H, KP-CH),
12.8 ppm (CH ); MS: m/z found: 483.3480, calcd for C H O
4
[M+H] : 483.3474.
3
3
3
31 47
+
.61 (m, 1H, 3b-H), 3.95 (brs, 1H, 12b-H), 4.16 (dd, J=10.8 and
.6 Hz, 1H, KP-CH ), 4.21 (dd, J=10.8 and 7.2 Hz, 1H, KP-CH ),
2
2
1
3
.39–7.82 ppm (m, 9H, arom); C NMR (75 MHz, CDCl ): d=196.9
3
(
C), 174.2 (C), 143.8 (C), 137.9 (C), 137.7 (C), 132.6 (CH), 131.5 (CH),
1
7
30.2 (2”CH), 129.1 (CH), 128.8 (CH), 128.5 (CH), 128.4 (2”CH),
Synthesis of 3a,12a-2Thy-DCABn (2)
3.2 (CH), 72.0 (CH), 69.0 (CH ), 48.4 (CH), 47.4 (CH), 46.6 (C), 42.2
2
(
CH), 39.0 (CH), 36.6 (CH ), 36.2 (CH), 35.3 (CH ), 35.2 (CH), 34.2 (C),
A stirred suspension of Thy-CH COOH (1.720 g, 9.36 mmol) in anhy-
2
2
2
3
2
1
3.8 (CH), 31.4 (CH ), 31.0 (CH ), 30.7 (CH ), 28.8 (CH ), 27.5 (CH ),
drous THF (60 mL) was treated with Et N (2.620 mL, 18.80 mmol)
2
2
2
2
2
3
7.3 (CH ), 26.3 (CH ), 23.8 (CH ), 23.3 (CH ), 18.1 (CH ), 17.4 (CH ),
and 2,4,6-trichlorobenzoyl chloride (1.750 mL, 11.23 mmol), and the
resulting mixture was allowed to react for 1.5 h. Then a solution of
4-DMAP (0.232 g, 1.91 mmol) and DCABn (0.752 g, 1.56 mmol) in
anhydrous THF (25 mL) was added and the mixture stirred, over-
night. Afterwards, the reaction mixture was poured into NaHCO3
2
2
2
3
3
3
2.9 ppm (CH ); MS: m/z found: 615.4038, calcd for C H O
5
3
40 55
+
[M+H] : 615.4050.
Synthesis of 3a,12a-2Thy-DCAKP (1)
(
5%), extracted with CH Cl , and the combined extracts were
2 2
A stirred suspension of Thy-CH COOH (1.192 g, 6.47 mmol) in anhy-
washed with brine, dried over MgSO , and concentrated under
2
4
drous THF (40 mL) was treated with Et N (1.810 mL, 12.99 mmol)
and 2,4,6-trichlorobenzoyl chloride (1.230 mL, 7.85 mmol) and the
resulting mixture was allowed to react for 1.5 h. Then, a solution of
vacuum. Purification by column chromatography (SiO2,
CH Cl :MeOH 98:2) and Li Chroprep RP-18, CH CN:H O 80:20), gave
3
2
2
3
2
1
2 as a white solid (1.210 g, 95%). H NMR (300 MHz, CDCl ): d=
3
4
-DMAP (0.527 g, 1.32 mmol) and DCAKP (0.663 g, 1.08 mmol) in an-
0.65 (s, 3H, CH ), 0.77 (d, J=5.7 Hz, 3H, 21-CH ), 0.86 (s, 3H, CH ),
3
3
3
hydrous THF (25 mL) was added and the resulting reaction mixture
1.87 (brs, 3H, Thy-CH ), 1.89 (brs, 3H, Thy-CH ), 4.38 (d, J=17.4 Hz,
3 3
was stirred overnight. Afterwards, it was poured into NaHCO (5%),
1H, Thy-CH ), 4.40 (d, J=17.4 Hz, 1H, Thy-CH ), 4.49 (d, J=17.4 Hz,
2 2
3
extracted with CH Cl , and the combined extracts were washed with
1H, Thy-CH ), 4.57 (d, J=17.4 Hz, 1H, Thy-CH ), 4.70 (m, 1H, 3b-H),
2 2
2
2
brine, dried over MgSO , and concentrated under vacuum. Purifica-
5.08 (brs, 2H, CH ), 5.11 (brs, 1H, 12b-H), 6.98 (brs, 1H, Thy-CH),
4
2
tion by column chromatography (SiO , CH Cl :MeOH 97:3, followed
7.05 (brs, 1H, Thy-CH), 7.26–7.37 (m, 5H, arom), 10.12 (s, 1H, Thy-
2
2
2
13
by Li Chroprep RP-18, CH CN:H O 80:20) gave 1 as a yellow solid
NH), 10.21 ppm (s, 1H, Thy-NH); C NMR (75 MHz, CDCl ): d=
3
2
3
1
(
(
0.959 g, 94%). H NMR (300 MHz, CDCl ): d=0.68 (s, 3H, CH ), 0.74
174.0 (C), 167.3 (C), 166.7 (C), 164.7 (C), 164.6 (C), 151.5 (C), 150.9
(C), 140.8 (CH), 140.5 (CH), 136.0 (C), 128.6 (2”CH), 128.3 (3xCH),
3
3
d, J=6.3 Hz, 3H, 21-CH ), 0.88 (s, 3H, CH ), 1.33 (d, J=6.9 Hz, 3H,
3
3
KP-CH ), 1.89 (brs, 3H, Thy-CH ), 1.90 (brs, 3H, Thy-CH ), 3.17 (m,
111.2 (C), 110.7 (C), 77.9 (CH), 76.5 (CH), 66.2 (CH ), 49.3 (CH+2”
3
3
3
2
1
H, KP-CH), 4.17 (d, J=6.9 Hz, 2H, KP-CH ), 4.41 (d, J=17.1 Hz, 1H,
CH ), 47.5 (CH), 45.2 (C), 41.7 (CH), 35.5 (CH), 34.7 (CH+CH ), 34.2
2
2
2
Thy-CH ), 4.45 (brs, 2H, Thy-CH ), 4.58 (d, J=17.1 Hz, 1H, Thy-CH ),
(CH), 34.1 (C), 31.9 (CH ), 31.3 (CH ), 30.8 (CH ), 27.3 (CH ), 26.8
2 2 2 2
2
2
2
4
7
.71 (brs, 1H, 3b-H), 5.12 (brs, 1H ,12b-H), 6.97 (brs, 1H, Thy-CH),
.11 (brs, 1H, Thy-CH), 7.39–7.85 ppm (m, 9H, arom); C NMR
(CH ), 26.0 (CH ), 25.9 (CH ), 25.3 (CH ), 23.4 (CH ), 22.9 (CH ), 17.7
2 2 2 2 2 3
13
(CH ), 12.4 (2”CH ), 12.2 ppm (CH ); MS: m/z found: 815.4200,
3 3 3
+
(
(
100 MHz, CDCl ): d=196.9 (C), 174.1 (C), 167.2 (C), 166.8 (C), 164.7
C), 164.3 (C), 151.6 (C), 150.9 (C), 143.8 (C), 140.9 (CH), 140.3 (CH),
calcd for C H N O [M+H] : 815.4231.
45 59 4 10
3
1
1
7
37.8 (C), 137.7 (C), 132.7 (CH), 131.5 (CH), 130.2 (2”CH), 129.2 (CH),
28.9 (CH), 128.5 (CH), 128.4 (2”CH), 111.5 (C), 110.8 (C), 78.1 (CH),
6.9 (CH), 69.1 (CH ), 49.7 (CH ), 49.5 (CH), 49.4 (CH ), 47.6 (CH), 45.3
2
2
2
Irradiation of 1
(
C), 41.9 (CH), 39.1 (CH), 35.7 (CH), 34.8 (CH), 34.7 (CH ), 34.5 (CH),
2
3
4.2 (C), 32.1 (CH ), 31.4 (CH ), 31.0 (CH ), 27.5 (CH ), 27.0 (CH ), 26.1
A solution of 1 (0.148 g, 0.16 mmol) in CH CN (250 mL) in a Pyrex
2
2
2
2
2
3
(
(
CH ), 25.9 (CH ), 25.6 (CH ), 23.6 (CH ), 23.1 (CH ), 18.1 (CH ), 17.8
round-bottom flask was purged with N and irradiated in a photo-
2
2
2
2
3
3
2
CH ), 12.5 (2”CH ), 12.4 ppm (CH ); MS: m/z found: 947.4810, calcd
reactor with eight lamps (lmax =350 nm) for 2 h. Then, the solvent
was concentrated under vacuum and the crude product was puri-
fied by column chromatography (SiO , CH Cl :MeOH 99:1) and (Li
3
3
3
+
for C H N O [M+H] : 947.4807.
5
4
67
4
11
2
2
2
Chroprep RP-18, CH CN:H O, 80:20) to give 3 (3a-Thy-DCA-12a-
3
2
Synthesis of DCABn
A stirred solution of DCA (2.510 g, 6.39 mmol) in anhydrous DMF
1
ThyhOiKP; 0.095 g, 64%). H NMR (300 MHz, CDCl ): d=0.75 (s,
3
3H, CH ), 0.86 (m, 6H, CH +21-CH ), 1.35 (d, J=7.2 Hz, 3H, KP-
3
3
3
(
8 mL) was treated with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU;
CH ), 1.82 (brs, 3H, Thy-CH ), 1.92 (brs, 3H, Thy-CH ), 2.23 (m, 2H,
3 3 3
1
7
.060 mL, 7.19 mmol). Ten minutes later, benzyl bromide (0.850 mL,
.18 mmol) was added dropwise and the solution was allowed to
CH ), 3.10 (m, 1H, KP-CH), 3.57 (d, J=17.7 Hz, 1H, Thy-CH ), 4.02 (t,
2 2
J=10.5 Hz, 1H, KP-CH ), 4.21 (d, J=17.1 Hz, 1H, Thy-CH ), 4.51 (d,
2
2
react overnight at room temperature. Then, the solvent was evapo-
rated and the crude product was redissolved in AcOEt, washed
with NaHCO (5%), 1m HCl, and brine, dried over MgSO , and con-
J=17.1 Hz, 1H, Thy-CH ), 4.53 (dd, J=3.9 and 10.5 Hz, 1H, KP-CH ),
2 2
4.61 (m, 1H, 3b-H), 4.62 (d, J=17.7 Hz, 1H, Thy-CH ), 4.72 (brs, 1H,
2
oxetane-CH), 5.27 (s, 1H, 12b-H), 6.91 (s, 1H, Thy-CH), 7.10–7.46 (m,
3
4
1
3
centrated under reduced pressure. Purification by column chroma-
9H, arom), 8.96 ppm (s, 1H, Thy-NH), 9.38 (s, 1H, Thy-NH). C NMR
tography (SiO , AcOEt:hexane 90:10) gave DCABn as a colorless
(75 MHz, CDCl ) d=173.5 (C), 170.5 (C), 167.0 (C), 166.7 (C), 164.0
2
3
1
solid (1.770 g, 57%). H NMR (300 MHz, CDCl ): d=0.64 (s, 3H,
(C), 151.6 (C), 150.7 (C), 144.3 (C), 143.3 (C), 140.0 (CH), 138.2 (C),
129.6 (CH), 128.9 (2”CH), 128.8 (CH), 126.2 (CH), 126.1 (2”CH),
124.9 (CH), 124.6 (CH), 111.8 (C), 91.7 (C), 77.6 (CH), 77.2 (CH), 76.3
(C), 69.1 (CH ), 68.4 (CH), 50.1 (CH), 50.0 (CH ), 49.2 (CH ), 47.7 (CH),
3
CH ), 0.90 (s, 3H, CH ), 0.95 (d, J=6.0 Hz, 3H, 21-CH ), 3.60 (m, 1H,
3
3
3
3
b-H), 3.95 (brs, 1H, 12b-H)), 5.08 (d, J=12.3 Hz, 1H, CH ), 5.13 (d,
2
13
J=12.3 Hz, 1H, CH ), 7.38–7.28 ppm (m, 5H, arom); C NMR
2
2
2
2
(
75 MHz, CDCl ): d=174.1 (C), 136.2 (C), 128.7 (2”CH), 128.4 (2”
45.2 (C), 41.9 (CH), 39.8 (CH), 36.4 (CH), 35.7 (CH), 35.1 (CH), 34.7
(CH ), 34.5 (C), 32.2 (CH ), 31.7 (CH ), 30.0 (CH ), 28.0 (CH ), 27.0
3
CH), 128.3 (CH), 73.2 (CH), 71.9 (CH), 66.2 (CH ), 48.4 (CH), 47.4
2
2
2
2
2
2
(
CH), 46.6 (C), 42.2 (CH), 36.6 (CH ), 36.1 (CH), 35.3 (CH ), 35.2 (CH),
(CH ), 26.2 (CH ), 25.9 (CH ), 25.7 (CH ), 24.5 (CH ), 23.5 (CH ), 23.3
2 2 2 2 3 2
2
2
3
4.2 (C), 33.8 (CH), 31.5 (CH ), 31.0 (CH ), 30.6 (CH ), 28.8 (CH ), 27.6
(CH ), 17.9 (CH ), 17.8 (CH ), 12.5 (CH ), 12.4 ppm (CH ); MS: m/z
3 3 3 3 3
found: 947.4805, calcd for C H N O [M+H] : 947.4807.
54 67 4 11
2
2
2
2
+
(
CH ), 27.3 (CH ), 26.2 (CH ), 23.8 (CH ), 23.3 (CH ), 17.4 (CH ),
2
2
2
2
3
3
Chem. Eur. J. 2015, 21, 17051 – 17056
www.chemeurj.org
17055
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Irradiation of the mixture BP:2
[2] a) C. A. Smith, M. Wang, N. Jiang, L. Che, X. Zhao, J.-S. Taylor, Biochem-
istry 1996, 35, 4146–4154; b) A. Gentil, F. Le Page, A. Margot, C. W. Law-
rence, A. Borden, A. Sarasin, Nucl. Acids Res. 1996, 24, 1837–1840; c) K.
Heil, D. Pearson, T. Carell, Chem. Soc. Rev. 2011, 40, 4271–4278.
A solution of 2 (0.102 g, 0.12 mmol) and BP (0.022 g, 0.12 mmol) in
CH CN (150 mL) in a Pyrex round-bottom flask was purged with N2
3
and irradiated in a photoreactor with eight lamps (lmax =350 nm)
for 6 h. Then, the solvent was concentrated under vacuum and the
crude was purified by column chromatography (SiO2,
CH Cl :acetone 90:10) to give 3a,12a-Thy< >Thy-DCABn (4)
[3] a) W. J. Schreier, T. E. Schrader, F. O. Koller, P. Gilch, C. E. Crespo-Hernan-
dez, V. N. Swaminathan, T. Carell, W. Zinth, B. Kohler, Science 2007, 315,
625–629; b) B. M. Pilles, D. B. Bucher, L. Liu, P. Clivio, P. Gilch, W. Zinth,
W. J. Schreier, J. Phys. Chem. Lett. 2014, 5, 1616–1622.
4] T. Climent, I. Gonzalez-Ramirez, R. Gonzalez-Luque, M. Merchan, L. Serra-
no-Andres, J. Phys. Chem. Lett. 2010, 1, 2072–2076.
5] a) T. Delatour, T. Douki, C. D’Ham, J. Cadet, J. Photochem. Photobiol. B
1998, 44, 191–198; b) M. C. Cuquerella, V. Lhiaubet-Vallet, J. Cadet,
M. A. Miranda, Acc. Chem. Res. 2012, 45, 1558–1570; c) M. C. Cuquerella,
V. Lhiaubet-Vallet, F. Bosca, M. A. Miranda, Chem. Sci. 2011, 2, 1219–
1232.
2
2
[
[
1
(
0.084 g, 82%). H NMR (300 MHz, CDCl ): d=0.70 (s, 3H, CH ),
3 3
0
.83 (d, J=4.5 Hz, 3H, 21-CH ), 0.88 (s, 3H, CH ), 1.59 (s, 3H, Thy-
3 3
CH ), 1.68 (s, 3H, Thy-CH ), 3.40–3.56 (m, 2H, Thy-CH ), 3.60 (brs,
3
3
2
1
2
7
H, Thy< >Thy-CH), 4.16 (brs, 1H, Thy< >Thy-CH), 4.35–4.56 (m,
H, Thy-CH ), 4.92 (m, 1H, 3b-H), 5.12 (brs, 3H, CH +12b-H), 7.24–
.38 (m, 5H, arom), 8.40 (s, 1H, Thy-NH), 8.58 ppm (s, 1H, Thy-NH);
2
2
1
3
[
[
6] S. L. Murov, I. Carmichael, G. L. Hug, 2nd ed. (Ed.: M. Dekker), New York,
C NMR (75 MHz, CDCl ): d=174.3 (C), 168.4 (C), 168.2 (C), 166.6
3
2009.
(
2”C), 150.5 (C), 150.4 (C), 136.0 (C), 128.7 (2”CH), 128.5 (2”CH),
7] a) S. Encinas, N. Belmadoui, M. J. Climent, S. Gil, M. A. Miranda, Chem.
Res. Toxicol. 2004, 17, 857–862; b) J. Trzcionka, V. Lhiaubet-Vallet, C.
Paris, N. Belmadoui, M. J. Climent, M. A. Miranda, ChemBioChem 2007, 8,
1
5
28.3 (CH), 79.0 (CH), 74.1 (CH), 66.5 (CH ), 63.6 (CH), 52.5 (CH),
2
1.3 (CH ), 49.4 (CH ), 46.7 (CH), 45.3 (2”C), 45.0 (C), 39.9 (CH), 34.9
2
2
(
CH), 34.5 (CH), 34.0 (CH ), 33.2 (CH), 32.4 (C), 31.0 (CH), 30.6 (CH ),
2 2
402–407; c) A. Joseph, G. Prakash, D. E. Falvey, J. Am. Chem. Soc. 2000,
3
2
1
0.2 (CH ), 29.6 (CH ), 27.4 (CH ), 25.7 (CH ), 25.6 (CH ), 25.1 (CH ),
2 2 2 2 2 2
122, 11219–11225.
3.4 (CH ), 23.1 (CH ), 22.8 (CH ), 22.4 (CH ), 21.5 (CH ), 17.1 (CH ),
[8] K. Sandros, Acta Chem. Scand. 1964, 18, 2355.
2
2
3
3
3
3
2.5 ppm (CH ); MS: m/z found: 815.4266, calcd for C H N O
10
[9] a) I. G. Gut, P. D. Wood, R. W. Redmond, J. Am. Chem. Soc. 1996, 118,
2366–2373; b) V. Lhiaubet-Vallet, S. Encinas, M. A. Miranda, J. Am.
Chem. Soc. 2005, 127, 12774–12775.
3
45 59
4
+
[M+H] : 815.4231.
[
[
[
10] N. Belmadoui, S. Encinas, M. J. Climent, S. Gil, M. A. Miranda, Chem. Eur.
J. 2006, 12, 553–561.
11] X.-L. Liu, J.-B. Wang, Y. Tong, Q.-H. Song, Chem. Eur. J. 2013, 19, 13216–
Acknowledgements
13223.
Financial support from the Spanish Government (Grants SEV-
012-0267, CTQ2012-38754-C03-03, and CTQ2012-32621), Gen-
12] P. D. Wood, R. W. Redmond, J. Am. Chem. Soc. 1996, 118, 4256–4263.
2
[13] T. Douki, I. Berard, A. Wack, S. Andrae, Chem. Eur. J. 2014, 20, 5787–
794.
[
5
eralitat Valenciana (Prometeo Program), and Technical Universi-
ty of Valencia (Predoctoral FPI fellowship for P.M.) is gratefully
acknowledged.
14] a) J. I. Kim, G. B. Schuster, J. Am. Chem. Soc. 1992, 114, 9309–9317; b) J.
PØrez-Prieto, S.-E. Stiriba, M. Gonzµlez-BØjar, L. R. Domingo, M. A. Miran-
da, Org. Lett. 2004, 6, 3905–3908; c) M. Gonzµlez-BØjar, A. Bentama,
M. A. Miranda, S.-E. Stiriba, J. PØrez-Prieto, Org. Lett. 2007, 9, 2067–
2
7
070; d) I. Wolfle, S. Chan, G. B. Schuster, J. Org. Chem. 1991, 56, 7313–
319.
Keywords: dimerization
· DNA damage · nucleobases ·
photochemistry · reaction mechanisms
[15] F. Bosca, I. Andreu, I. M. Morera, A. Samadi, M. A. Miranda, Chem.
Commun. 2003, 1592–1593.
[
1] a) S. Mouret, C. Baudouin, M. Charveron, A. Favier, J. Cadet, T. Douki,
Proc. Natl. Acad. Sci. USA 2006, 103, 13765–13770; b) R. B. Setlow, E.
Grist, K. Thompson, A. D. Woodhead, Proc. Natl. Acad. Sci. USA 1993, 90,
[16] K. Nakatani, T. Yoshida, I. Saito, J. Am. Chem. Soc. 2002, 124, 2118–2119.
6
666–6670; c) P. J. Rochette, J.-P. Therrien, R. Drouin, D. Perdiz, N. Bas-
Received: July 13, 2015
tien, E. A. Drobetsky, E. Sage, Nucl. Acids Res. 2003, 31, 2786–2794.
Published online on October 14, 2015
Chem. Eur. J. 2015, 21, 17051 – 17056
www.chemeurj.org
17056
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim