Angewandte
Communications
Chemie
The mechanistic study of 6-4PP photorepair entails an
additional difficulty compared to the CPD case. This relies on
the instability of the four membered ring oxetane/azetidine
intermediates, which prevents not only their isolation and
characterization but also their use as substrates to investigate
the electron-induced cycloreversion. Hence, this step has
been largely studied using oxetane models;[5] however, until
now no analogous report has appeared on model azetidine
sytems. Electron transfer cycloreversion of triphenylazeti-
dines has been achieved following a photooxidative path-
way,[5,6] whereas the photoreductive approach has only been
applied to the ring splitting of structurally unrelated azetidin-
2-ones, where the nitrogen atom belongs to a strained b-
lactam moiety.[5,7]
absorption properties of AZT and T-AU (Figure 1 bottom),
the thermal stability of AZT in acetonitrile was assessed by
UV/Vis measurements at 298 K, no spectral changes were
detected after 48 h (Figure S7).
It is assumed that the electron transfer process occurs
from the singlet excited state of the reduced flavin
(1FADHÀ*) to the azetidine moiety. Thus, direct mechanistic
information should in principle be obtained by monitoring the
changes in the intensity and/or kinetics of the cofactor
emission in the presence of the azetidine by steady-state
and/or time-resolved fluorescence, respectively. However, the
very short lifetime of 1FADHÀ* (in the subnanosecond
timescale) does not provide a time-window compatible with
diffusion-controlled intermolecular reaction.[11] To overcome
this limitation, a series of photosensitizers (Phs) with singlet
lifetime in the nanosecond range and oxidation potential
close to that of 1FADHÀ* (Eox* of ca. À2.9 V)[12] was selected
(see Table 1).[9a,b,13] In a first stage, steady-state fluorescence
Here, a stable azabipyrimidinic azetidine (AZT, Figure 1
top) has been designed as a model for the purported
Table 1: Oxidation potential in the singlet excited state (Eox*) of the
selected Phs, and bimolecular rate constant (kq) for the quenching of the
Phs by AZT determined by time-resolved fluorescence.
[a]
Phs
Eox*
t
kq
[V vs. SCE] [ns] [109 mÀ1 sÀ1
]
N,N,N’,N’-tetramethyl-1,4-
phenylenediamine (TMPD)
N,N,N’,N’-tetramethylbenzidine
(TMB)
À3.3[b]
À3.2[b]
1.5 N.D. [c]
6.0 8.0Æ0.7
2.8 7.3Æ1.0
N,N-dimethylaniline (DMA)
À3.0[b]
À2.9[d]
À2.5[e]
À2.5[b]
À2.5[f]
À2.3[b]
À2.1[b]
FADHÀ
carbazole (CAR)
7.4 4.0Æ0.2
10.6 3.7Æ0.2
6.2 4.0Æ0.3
6.6 2.5Æ0.2
11.9 1.7Æ0.2
acenaphthene (ACE)
1-methoxynaphthalene (1-MN)
2-methoxynaphthalene (2-MN)
Chrysene (CHRY)
Figure 1. Top: Model azetidine AZT and its cycloreversion product
(T-AU), i) Phs+hn (lirr =350 nm), ii) acetone+hn (lirr >320 nm).
Bottom: UV absorption spectra of AZT (110À5 m, left) and T-AU
(510À5 m, right) in acetonitrile.
[a] The experiments were performed twice and the errors correspond to
average deviations. [b] From Ref. [9a]. [c] Not determined because of the
temporal resolution of the setup. [d] From Ref. [12]. [e] From Ref. [13].
[f] From Ref. [9b].
intermediates in the repair of TC dimers by photolyases, in
order to address for the first time spectroscopic and photo-
chemical studies of their photoreductive cycloreversion. From
a different perspective, AZT is also an aza analog of CPD.
This new compound was obtained by the photocycloaddition
between the N3-methyl derivatives of thymine and 6-aza-
uracil linked at N1 by a trimethylene bridge (T-AU, Figure 1
top). This type of bridge has previously been used in
formation[8] and repair[9] studies of model CPD, and it appears
to favor the interaction of the lesion with flavin singlet excited
state.[9c] The synthesis of T-AU was accomplished by reaction
of 1-bromo-3-chloropropane with N3-methylthymine and
subsequent reaction of the resulting 1-(3-chloropropyl)-3-
methylthymine with the N3-methyl derivative of 6-azauracil.
Then, AZT was obtained as the main photoproduct by
irradiation of T-AU at l > 320 nm in the presence of acetone
as photosensitizer, to avoid photocycloreversion of the
product under direct irradiation.[10] The cis-syn configuration
of AZT was secured by NOE experiments through the
interaction of azetidine ring protons with the C5-methyl
group of the dihydrothymine moiety (Figure S6 in the
Supporting Information). In view of the markedly different
experiments were performed. Thus, fluorescence intensity of
the selected sensitizers was measured in the absence or in the
presence of different concentrations of AZT (see Figure 2A
for the case of carbazole). Next, time-resolved fluorescence
experiments were run in order to conclude about the dynamic
character of the quenching process. The singlet excited state
lifetime of all photosensitizers was shortened in the presence
of AZT (see Figure 2B for the case of carbazole). The
bimolecular rate constants kq were determined from the plots
representing the reciprocal of the Phs lifetime as a function of
AZT concentration (Figure 2B, inset). According to the
Rehm-Weller equation, the quenching process was more
efficient as Eox* became increasingly negative (Table 1 and
Figure 3), reaching the diffusion limit near À3.0 V.
By contrast, no clear correlation was obtained between kq
and the Phs singlet excited state energy (Figure S17) ruling
out a singlet–singlet energy transfer process as responsible for
1
the deactivation of Phs*. Altogether, these data point to an
1
electron transfer mechanism between Phs* and AZT. It is
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6037 –6040
Fraga-Timiraos, Ana B.
