Absorption Characteristics and Quantum Yields of Singlet Oxygen Generation of Thioguanosine Derivatives

Article abstract of DOI:10.1111/php.12900

6-Thioguanine (1a) is considered to be photochemotherapeutic due to its specific characteristics of photosensitivity to UVA light and singlet molecular oxygen generation. To extend its phototherapeutic ability, two related thioguanines, 8-thioguanine (2a) and 6,8-dithioguanine (3a), have been designed and explored. Since the solubility of these thioguanines in dehydrated organic solvents is too poor to study, their triacetyl-protected ribonucleosides, that is, 2′,3′,5′-tri-O-acetyl-6-thioguanosine (1c), 2′,3′,5′-tri-O-acetyl-8-thioguanosine (2c) and 2′,3′,5′-tri-O-acetyl-6,8-dithioguanosine (3c) were prepared and investigated. The absorption maxima of 1c, 2c and 3c in acetonitrile were found at longer wavelengths than that of unthiolated guanosine (4c). Especially, 3c has the longest wavelength for absorption maximum and the highest value in terms of molar absorption coefficient among all thionucleobases and thionucleosides reported. These absorption properties were also well reproduced by quantum chemical calculations. Quantum yields of singlet oxygen generation of 2c and 3c were determined by near-infrared emission measurements to be as large as that of 1c. These results suggest that the newly synthesized thioguanosines, in particular 3c, can be further developed as a potential photosensitive agent for light-induced therapies.

Full text of DOI:10.1111/php.12900

Photochemistry and Photobiology, 20**, **: *–*
Absorption Characteristics and Quantum Yields of Singlet Oxygen
Generation of Thioguanosine Derivatives
1
1
1
1
2
Shoma Miyata , Takeshi Yamada , Tasuku Isozaki , Hideyuki Sugimura , Yao–Zhong Xu and
1
Tadashi Suzuki *
1
Department of Chemistry and Biological Science, Aoyama Gakuin University, Sagamihara, Kanagawa, Japan
School of Life, Health and Chemical Sciences, the Open University, Milton Keynes, UK
2
Received 24 November 2017, accepted 17 January 2018, DOI: 10.1111/php.12900
ABSTRACT
As a photochemotherapeutic agent, the photosensitizer must
be sensitive to the light penetrating into deep hypodermal tissue
6-Thioguanine (1a) is considered to be photochemotherapeutic
1
and can also effectively generate singlet oxygen ( O
2
*), one type
due to its specific characteristics of photosensitivity to UVA light
and singlet molecular oxygen generation. To extend its pho-
totherapeutic ability, two related thioguanines, 8-thioguanine
of ROS. In the ultraviolet and visible light region, the longer
wavelength light has higher permeability to subcutaneous tissues
14). However, the absorption maxima of 1a and 1b have been
observed at around 350 nm, which is not long enough to allow
the light to penetrate into deep subcutaneous tissues (2,3,15–19).
Thus, it is well worth designing and developing alternative
thioguanines and their nucleosides with an absorption band at
longer wavelengths.
(
(2a) and 6,8-dithioguanine (3a), have been designed and
explored. Since the solubility of these thioguanines in dehydrated
organic solvents is too poor to study, their triacetyl-protected
0
0
0
ribonucleosides, that is, 2 ,3 ,5 -tri-O-acetyl-6-thioguanosine
0
0
0
0
0
0
(
1c), 2 ,3 ,5 -tri-O-acetyl-8-thioguanosine (2c) and 2 ,3 ,5 -tri-
O-acetyl-6,8-dithioguanosine (3c) were prepared and investi-
gated. The absorption maxima of 1c, 2c and 3c in acetonitrile
were found at longer wavelengths than that of unthiolated
guanosine (4c). Especially, 3c has the longest wavelength for
absorption maximum and the highest value in terms of molar
absorption coefficient among all thionucleobases and thionu-
cleosides reported. These absorption properties were also well
reproduced by quantum chemical calculations. Quantum
yields of singlet oxygen generation of 2c and 3c were deter-
mined by near-infrared emission measurements to be as large
as that of 1c. These results suggest that the newly synthesized
thioguanosines, in particular 3c, can be further developed as
a potential photosensitive agent for light-induced therapies.
Photophysical and photochemical properties of 2a, 3a and
their nucleosides (2b and 3b) have not been documented
although their synthetic studies were reported (20–23). 8-Oxo-
guanine (with a carbonyl group at 8-position of the purine ring
and also known as an oxidation photoproduct of guanine (4a))
was reported to exhibit a redshifted absorption band relative to
its 6-oxo-analogs (4a and 4b) (24–26). In addition to 1a and 1b,
thiocarbonyl-modified pyrimidine bases (such as thiolated-uracil
and thiolated-thymine) have a strong absorption maximum at
longer wavelength than their respective unthiolated nucleobases
(27–31). Thus, the newly designed thioguanines, especially 3a,
could outstrip the thioanalogs of nucleobases examined so far in
terms of their absorption properties. Photochemical experiments
should be carried out in rigidly dehydrated organic solvents to
clarify the intrinsic property of the excited states for the thionu-
cleobases; however, purine nucleobases have generally low solu-
bility in most of the organic solvents (32). Thus, we prepared
triacetyl-protected derivatives (1c–4c) for better solubility and
easier handlings.
INTRODUCTION
6
-Thioguanine (1a, see Scheme 1) and other thioanalogs of natu-
ral nucleobases have high affinity to proliferating cells, and some
of them have been prescribed for the treatment of cancers, leuke-
mia and angina among others (1–9). 1a can be converted into
In this article, we report our work on chemical synthesis and
photochemical investigation of thiolated guanosine derivatives
6
-thioguanosine (1b) through cellular metabolism, followed by
being incorporated into RNA and DNA (1–6,10,11). O’Donovan
et al. (8) reported that 1a localizing in tumor cell generated reac-
tive oxygen species (ROS) by its exposure to UVA light and
thus cellular apoptosis was induced. These findings indicate that
(
1c, 2c and 3c). We also present out results on their structural
characterizations by NMR, absorption properties by steady-state
1
absorption spectra and quantum yields of O * generation by
2
time-resolved near-infrared emission measurement. It is our view
that these thioguanosine derivatives can be further developed as
potential photochemotherapeutic agents.
1
a and its nucleosides could be used as an effective medical tool
for cancer treatment due to their unique properties as pho-
tochemotherapeutic drugs like 4-thiothymidine, including a pho-
toactivatable genotoxic agent and
photodynamic therapy (PDT), in addition to the hitherto known
use as an anticancer medicine (5,9,12,13).
a
photosensitizer for
MATERIALS AND METHODS
General. Reagents were purchased from standard suppliers and used
without further purification. Solvents were used after distillation.
Reactions were monitored with thin-layer chromatograph (TLC) plate
*Corresponding author email: suzuki@chem.aoyama.ac.jp (Tadashi Suzuki)
© 2018 The American Society of Photobiology
(
Silica gel 60, F254). Spots on the TLC plate were monitored with UV,
1

2
Shoma Miyata et al.
Scheme 1. Structures of guanine, thioguanines and their nucleosides.
Scheme 2. The synthesis routes for thioguanosines and their acetylated derivatives.
ninhydrin or anisaldehyde. A C–200 silica gel was used for silica gel
standard, and coupling constants (J values) were represented in hertz.
Mass spectrum was measured using FAB–MS (JEOL, JMS–700
MStation).
Synthesis of 2 ,3 ,5 -tri-O-acetylguanosine (4c). Acetic anhydride
(9 mL, 95.2 mmol) was added to a solution of guanosine (4b, 2.83 g,
10.0 mmol) in pyridine (27 mL), and the mixture was kept for 2 h at
1
13
flash chromatography. H NMR and C NMR spectra were measured
1
with 500 MHz NMR (JEOL, JNM-ECX 500 MHz), and typical
H
1
3
0
0
0
NMR and C NMR spectra are shown in the Supporting Information
Figures S1–S8). The multiplicity was expressed as follows: s = singlet,
(
d = doublet, t = triplet, m = multiplet and br = broad. The chemical
shifts are expressed in ppm relative to residual solvent as an internal
2
80°C. The reaction was quenched by addition of H O at 0°C after

Photochemistry and Photobiology
3
0
0
0
Synthesis of 2 ,3 ,5 -tri-O-acetyl-6,8-dithioguanosine (3c). Lawesson’s
0
0
0
4
reagent (8.75 g, 21.6 mmol) was added to a solution of 2 ,3 ,5 -tri-O-
acetyl-8-bromoguanosine (6, 5.00 g, 10.3 mmol) in dioxane (100 mL),
and the mixture was kept for 4 h at +100°C. The reaction mixture was
2
2
/
0
0
0
dissolved in ethyl acetate and washed with saturated NaHCO
organic layer was dried with anhydrous magnesium sulfate. After
evaporation of the solvent and column chromatography (CH Cl
MeOH = 97:3), 2 ,3 ,5 -tri-O-acetyl-6,8-thioguanosine (3c, 2.33 g,
5.10 mmol, 50%) was obtained as yellow solid. Rf = 0.54 (CH Cl
) 13.11 (1H,
3
, and the
1
2
2
/
0
0
0
2
2
/
1
MeOH = 9:1): H NMR (500 MHz, dimethylsulfoxide–d
6
br), 12.21 (1H, br), 6.93 (2H, br), 6.40 (1H, d, J = 4.1 Hz), 6.08 (1H, t,
J = 5.5 Hz), 5.60 (1H, t, J = 6.2 Hz), 4.38 (1H, dd, J = 11.7, 6.9 Hz),
1
36.2, 117.3, 84.9, 80.1, 72.6, 70.8, 63.6, 21.1, 20.9 and 20.7.
4.22 (1H, m), 4.16 (1H, dd, J = 11.7, 6.9 Hz), 2.06 (3H, s), 2.02 (3H, s)
and 1.97 (3H, s); C NMR (125 MHz, dimethylsulfoxide–d ) 170.6,
6
0
0
0
13
Synthesis of 2 ,3 ,5 -tri-O-acetyl-6-thioguanosine (1c). Lawesson’s
0
0
0
reagent (1.19 g, 2.93 mmol) was added to a solution of 2 ,3 ,5 -tri-O-
acetylguanosine (4c, 2.01 g, 4.91 mmol) in dioxane (40 mL), and the
mixture was kept for 5 h at +100°C. The reaction mixture was dissolved
169.88, 169.84, 168.1, 164.1, 154.1, 146.7, 117.8, 86.8, 79.6, 71.1, 70.7,
+
63.5, 21.1, 20.8, 20.7: MS (FAB+) m/z 458 (MH ).
Ultraviolet–visible (UV–vis) absorption spectroscopy. The UV–vis
in ethyl acetate and washed with saturated NaHCO
layer was dried with anhydrous magnesium sulfate. After evaporation of
the solvent and column chromatography (CH Cl /MeOH = 97:3),
,3 ,5 -tri-O-acetyl-6-thioguanosine (1c, 0.999 g, 2.35 mmol, 48%) was
3
, and the organic
absorption spectra were recorded at room temperature on
a
spectrophotometer (JASCO, U–best V550) using a quartz cuvette of
1 cm optical path length. The sample solution was prepared with
acetonitrile as a solvent.
2
2
0
0
0
2
1
obtained as white solid. Rf = 0.51 (CH
500 MHz, dimethylsulfoxide–d ) (d, ppm) 12.03 (1H, s), 8.09 (1H, s),
.84 (2H, s), 5.95 (1H, d, J = 6.2 Hz), 5.76 (1H, t, J = 6.2 Hz), 5.45
1H, dd, J = 6.2, 4.1 Hz), 4.34 (1H, dd, J = 11.3, 3.7 Hz), 4.30 (1H, dd,
J = 9.6, 4.1 Hz), 4.23 (1H, dd, J = 11.0, 5.5 Hz), 2.07 (3H, s), 2.00
2
Cl
2
/MeOH = 9:1): H NMR
Time-resolved near-infrared emission spectroscopy. Time-resolved
near-infrared emission measurement was carried out with a thermoelectric
cooled near-infrared photomultiplier tube (Hamamatsu Photonics, H10330–
45; InP/InGaAsP, spectral response 950–1400 nm) combined with a long-
pass filter (Thorlabs, FEL1250; cut-on wavelength 1250 nm) and a
bandpass filter (Edmund, Hard-coated bandpass filter; 1275 ꢀ50 nm)
(
6
6
(
13
(
3H, s) and 1.99 (3H, s); C NMR (125 MHz, dimethylsulfoxide–d
6
) (d,
3
+
ppm) 176.0, 170.0, 169.9, 169.8, 153.7, 148.2, 139.0, 128.9, 85.1, 80.2,
(Figure S9). A forth-harmonic of a Nd :YAG laser (Continuum, Surelite
II–10, 5 ns pulse duration, 10 Hz, 266 nm) was used as an excitation light
source. The sample solution was prepared with acetonitrile as a solvent.
Quantum chemical calculation. Ground- and excited-state calculations
for corresponding purine bases (1a–4a) were performed using the
Gaussian 09W program package (33). Ground-state geometries of the
purine bases were optimized by the density functional theory (DFT) at
the B3LYP/6ꢁ311 + G(d,p) level. Vertical excitation energies were
estimated by the time-dependent DFT (TD-DFT) at the TDꢁB3LYP/
6ꢁ311 + G(d,p) level. Solvent effects were modeled with the polarizable
continuum model (PCM) for the ground and excited states.
7
2.5, 70.8, 63.5, 21.1, 20.9 and 20.7.
Synthesis of 8-bromoguanosine (5). Br (3.0 mL) was added to a
2
suspension liquid of guanosine (4b, 5.0 g, 17.7 mmol) in
2
H O
(
100 mL), and the vigorous stirring was kept for 24 h at room
2
temperature. Excess Br was quenched by addition of saturated sodium
thiosulfate solution (3.0 mL); the precipitate was collected by filtration
and washed by H O on a Buchner funnel. After evaporation of the
2
solvent, 8-bromoguanosine (5, 6.3 g, 17.5 mmol, 99%) was obtained as
white solid.
Synthesis of 8-thioguanosine (2b). Thiourea was added to
suspension liquid of 8-bromoguanosine (5, 4.01 g, 11.1 mmol) in
ethanol (40 mL). A small quantity of H O (5 mL) was added until the
a
2
RESULTS AND DISCUSSION
suspension liquid being dissolved and kept heated to reflux for 24 h.
The reaction mixture was cooled at room temperature, excess ethanol
was evaporated, and the precipitated solid was filtered. The solid was
Synthesis of thioguanosine derivatives
2
washed with H O on a Buchner funnel, and after evaporation of the
solvent, 8-thioguanosine (2b, 2.39 g, 7.59 mmol, 68%) was obtained as
Scheme 2 outlines the synthetic routes to 1c–4c. The syntheses of
white solid.
1
c, 2c and 4c have been reported previously (34–43). To the best
of our knowledge, this is the first report on chemical synthesis of
c. The structures of all synthesized products were characterized
0
0
0
Synthesis of 2 ,3 ,5 -tri-O-acetyl-8-thioguanosine (2c). Acetic
anhydride (2 mL, 21.2 mmol) was added to solution of
-thioguanosine (2b, 2.10 g, 6.69 mmol) in pyridine (35 mL), and the
mixture was kept for 8 h at room temperature. The reaction was
quenched by addition of H O at 0°C after checking the completion of the
reaction by TLC. The reaction mixture was dissolved in ethyl acetate and
washed with saturated NH Cl, and the organic layer was dried with
anhydrous magnesium sulfate. After evaporation of the solvent and
a
3
8
1
by H NMR, and their purities were estimated to be 99% for 1c,
9% for 2c, 99% for 3c and 97% for 4c with a minor amount of
impurity being H O. The concentration of H O was determined
2
9
2
2
4
by subtracting the peak area deriving from H
2
O in dimethylsul-
0
0
0
column chromatography (CH
2
Cl
2
/MeOH = 95:5), 2 ,3 ,5 -tri-O-acetyl-8-
foxide–d solvent from that in 1c–4c solutions to remove the
6
thioguanosine (2c, 1.76 g, 3.98 mmol, 59%) was obtained as white
intrinsic moisture content of the deuterated solvent. No other
impurity was detected by HPLC spectra as shown in Figure S10.
1
solid. Rf = 0.46 (CH
dimethylsulfoxide–d ) 13.02 (1H, br), 11.09 (1H, br), 6.62 (2H, br), 6.38
1H, s), 6.10 (1H, br), 5.63 (1H, t, J = 6.0 Hz), 4.38 (1H, dd, J = 11.7,
2
Cl
2
/MeOH = 9:1):
H
NMR (500 MHz,
6
1
c was prepared in a two-step process for the first time. First,
(
4
b was quantitatively converted to 4c, which then was trans-
3
2
d
7
.4 Hz), 4.20 (1H, m), 4.17 (1H, dd, J = 11.3, 6.5 Hz) 2.06 (3H, s),
13
.02 (3H, s) and 1.97 (3H, s); C NMR (125 MHz, dimethylsulfoxide–
) 170.7, 169.88, 169.82, 165.5, 154.4, 151.4, 149.7, 104.4, 86.8, 79.5,
1.3, 70.7, 63.5, 21.0, 20.8 and 20.7.
formed to 1c with Lawesson’s reagent. Following the reported
procedure (20,42,43), 2c was synthesized in a three-step process
from 4b. To prepare 3c, 6 is the key intermediate which can be
obtained from bromination of 4b followed by acetylation of the
resultant 5 in an excellent yield (22,43). In an early report (22),
6 had been treated with phosphoryl chloride and hydrolyzed to
afford 2-amino-6,8-dichloropurine, followed by nucleophilic sub-
stitution reaction with thiourea to yield 3b. In order to overcome
the difficulty in handling the phosphoryl chloride and its low
yield of 3b (33%) (22), thus, we developed another synthetic
route to 3c. By a simple treatment of 6 with Lawesson’s reagent,
6
0
0
0
Synthesis of 2 ,3 ,5 -tri-O-acetyl-8-bromoguanosine (6). Acetic
anhydride (5 mL, 52.9 mmol) was added to solution of 8-
a
bromoguanosine (5, 4.76 g, 13.2 mmol) in pyridine (29 mL), and the
mixture was kept for 6 h at room temperature. The reaction was
quenched by addition of H
reaction by TLC. The reaction mixture was dissolved in ethyl acetate and
washed with saturated NH Cl, and the organic layer was dried with
anhydrous magnesium sulfate. After evaporation of the solvent and
2
O at 0°C after checking the completion of the
4
0 0 0
2 2
Cl /MeOH = 95:5), 2 ,3 ,5 -tri-O-acetyl-8-
column chromatography (CH
bromoguanosine (6, 4.79 g, 9.84 mmol, 75%) was obtained as white
solid. Rf = 0.48 (CH Cl /MeOH = 9:1).
2
2
3
c was successfully afforded with a higher yield of 50%.

4
Shoma Miyata et al.
1
Table 1. H NMR chemical shifts of 1c–4c.
1
7
2
0
0
0
0
0
0
5
N H
N H
N H
2
1
2
3
4
5
1c
2c
3c
4c
12.03
11.09 13.02
12.21 13.11
ꢁ
6.84
6.62
6.93
6.55
5.95 5.76 5.45 4.30 4.34 4.23
6.38 6.10 5.63 4.20 4.38 4.17
6.40 6.08 5.60 4.22 4.38 4.16
5.95 5.75 5.46 4.27 4.34 4.22
10.76
ꢁ
Bold values are the values that are noticeably high due to the effect of
the thiocarbonyl modification in these compounds.
1
13
H and C NMR analysis of thioguanosine derivatives
To ascertain the correct structures of the synthesized products,
1
NMR spectroscopy was used. The H NMR chemical shifts of
1
c–4c are listed in Table 1. The chemical shift values for the
proton at 1-position (the imide group) of thioguanosine deriva-
tives were observed at lower magnetic field than that of unthio-
lated guanosine. The peaks for the imide group, especially in 1c
and 3c, were significantly shifted to a lower magnetic field. This
shift can be ascribed to the thiocarbonyl substitution at 6-position
of the purine ring. The thiocarbonyl modification at 8-position
also causes a significant shift of the peak for the imide group to
a lower magnetic field, as observed for 2c. The peaks for the
amine protons at 2-position of the purine ring for 1c and 3c also
exhibited a slight shift to the low field in comparison with the
corresponding peak for 2c and 4c. The chemical shift values
deriving from ribose sugar protons, especially for anomeric pro-
Figure 1. (a) Absorption spectra of 1c–4c in acetonitrile solution, and
(b) computational vertical transition energy and oscillator strength of 1a–
4
a at PCM/TD–B3LYP/6–311G+(d,p) level.
0
ton (1 H), were also shifted to a lower magnetic field in both 2c
and 3c. These shifts can be ascribed to the thiocarbonyl modifi-
cation at 8-position, consistent with a recent publication in which
the peak for the proton at 3-position (the imide group) in 4-thio-
pyrimidines was also found to be at lower magnetic field than
that of unthiolated pyrimidine bases (44).
Table 2 lists C NMR chemical shifts of all the carbon atoms
in compounds 1c–4c. The peak at 157.2 ppm in unthiolated 4c
was found to shift to 176.0 ppm (18.8 ppm lower magnetic field)
in the 6-thiolated analog (1c). Similarly, the peak at 136.2 ppm
in the unthiolated 4c was also found to shift substantially to
than 295 nm, was almost identical to that of guanosine (4b),
revealing that the acetylation of three hydroxyl groups in the
sugar component had little effect on electronic states concerning
with transitions in this spectral range. Since the solubility of 4c
(triacetyl-protected guanosine) in acetonitrile was over 300 times
larger than that of 4b (unprotected guanosine) in units of molar-
ity (16.5 mM for 4c and 41.6 lM for 4b), thus corresponding tri-
acetyl-protected analogs instead of the unprotected analog were
used to study their UV properties. The absorption spectrum for
2c (8-thiolated analogy) exhibited an intense absorption band
centered at 302 nm with a high molar absorption coefficient
13
1
65.5 ppm in the 8-thiolated analog (2c), resulting 29.3 ppm
3
2
02nm
4
ꢁ1
ꢁ1
lower magnetic field shift. Compound 3c is a doubly thiolated
analog, both of the thiocarbonyl carbons are shifted to a lower
field (i.e. 168.1 and 164.1 ppm) compared with those of unthio-
lated 4c. These shifts should be due to the thiolation at their
respective positions (6-position and 8-position). The similar
effect was also reported for the carbon atoms of thiocarbonyl
carbons in 2-thiouracil and 4-thiouracil (44,45).
[e
= (2.39 ꢀ0.01) 9 10
M
cm ]. In comparison with
c
4c, this redshift of the band (48 nm) observed from 2c is likely
to result from the extension of the p-conjugation due to the thio-
carbonyl modification at 8-position of the purine ring. 1c (6-thio-
lated analogy) has a much large redshifted band (91 nm) with a
346nm
1c
higher molar absorption coefficient [e
= (2.82 ꢀ0.01) 9
4
ꢁ1
ꢁ1
10
M
cm ]. The absorption maximum of 1c appeared in the
longer wavelength region than that of 2c, indicating that thiocar-
bonyl modification at 6-position has more contribution to its
electronic transition than that at 8-position. Our findings offer a
solid support to an early report (17) that the replacement of an
oxygen atom by a sulfur atom in a carbonyl group is expected to
UV–vis absorption spectroscopy
The absorption spectra of 1c, 2c, 3c and 4c are shown in Fig. 1a.
Absorption spectrum of 4c, appeared in the spectral range less
1
3
Table 2. C NMR chemical shifts of 1c–4c.
2
4
5
6
8
0
0
0
0
0
5
C
C
C
C
C
1
2
3
4
1c
2c
3c
4c
153.7
151.4
146.7
154.5
148.2
154.4
154.1
151.6
128.9
104.4
117.8
117.3
176.0
149.7
168.1
157.2
139.0
165.5
164.1
136.2
85.1
86.8
86.8
84.9
80.2
79.5
79.6
80.1
72.5
71.3
71.1
72.6
70.8
70.7
70.7
70.8
63.5
63.5
63.5
63.6
Bold values are the values that are noticeably high due to the thiolation at the respective positions in these compounds.

Photochemistry and Photobiology
5
Table 3. Photophysical properties of 1c, 2c and 3c in acetonitrile solution.
‡
‡
Experimental
Computational
†
4
Mꢁ1
ꢁ1
‡
§
D
¶
s
††
T
kmax*/nm
e
max /10
cm
E
T
/eV
Φ
E
/nm
f **
E
/eV
1c
2c
3c
346
302
381
2.82 ꢀ0.01
2.39 ꢀ0.01
3.76 ꢀ0.02
2.71
2.93
2.50
0.37 ꢀ0.01
0.28 ꢀ0.01
0.33 ꢀ0.01
338
312
383
0.571
0.497
0.669
2.70
3.04
2.43
†
‡
*
Wavelength at absorption maximum. Molar absorption coefficient at absorption maximum. Triplet state energy obtained from emission peak of phos-
§
¶
††
‡‡
phorescence spectrum. Quantum yield of singlet oxygen generation. Vertical transition energy. **Oscillator strength. Triplet state energy. Calculated
at the PCM/TD–B3LYP/6–311 + G(d,p) level. Bold values are the values that are substantially high due to the thiolation at both 6-position and 8-posi-
tions of the compound.
shift the absorption band to the red. The absorption maximum of
c (dithiolated analog) was observed at 381 nm, which surpris-
ingly results in a redshift up to 126 nm in comparison with
emission spectrum of 1c was identical to the reported phospho-
rescence spectrum of 1b (3,48). Although the quantum yields
of triplet formation for these thioguanosines (1c, 2c and 3c)
have not been obtained yet, their triplet–triplet absorption spec-
tra were observed at room temperature, which will be described
in detail in the next paper. In addition, these thioguanosines
have relatively high quantum yields of singlet molecular oxy-
gen generation (see below), indicating high triplet formation of
these compounds. Therefore, those emissions can be confidently
assigned to their respective phosphorescence of 1c–3c. Thus, 2c
and 3c as well as 1a–1c will form the excited triplet manifold
through intersystem crossing from the singlet excited states.
3
4
c. In addition, the absorption intensity at the maximum
3
3
81nm
4
ꢁ1
ꢁ1
[
e
= (3.76 ꢀ0.02) 9 10
M
cm
]
was
remarkably
c
higher than any other thioanalogs of nucleobases examined so
far (2,3,15–18,27–31), revealing 3c can be activated with a very
low dose of UV light. These desirable UV properties suggest that
3
c would be much sensitive to the light penetrating into the
human skin and could be used as a powerful photosensitive
agent for light-induced therapies, including PDT. Thioguanosines
(
1c, 2c and 3c) are not regarded as suitable agents for PDT
because of the absence of one-photon absorption at visible light.
However, in our recent works (46,47), 1b and 3c were success-
fully excited at red light by multiphoton excitation. Thus, with
the multiphoton approach, thioguanosines have offered a poten-
tial as a PDT sensitizer.
Quantum chemical calculations
Optimized ground-state geometries of 1a, 2a and 3a are shown in
Figure S12 and Tables S1–S3. These molecules belong to the Cs
symmetry, and all atoms lie in a plane of the purine ring. The
bond lengths and angles were comparable to each other except
Steady-state emission measurements were also carried out on
1
c–3c, but no emission was observed at room temperature, indi-
6
6
6
cating a fluorescence quantum yield of virtually zero. On the
other hand, an emission was clearly observed at 77 K in glassy
ethanol matrix, as described in the supporting materials (Fig-
ure S11). Those emission bands exhibited significantly large
Stokes shift from those absorption maxima (beyond 100 nm, as
listed in Table 3) with longer lifetimes, over a microsecond. The
for the C =S and C =O bond lengths. The C =S bond lengths of
6
ꢀ
1a and 3a (1.69 A) were significantly larger than the C =O bond
ꢀ
lengths of 2a and 4a (1.23 A). This clearly reveals that the
6
6
strength of the C =S bond is weaker than that of the C =O bond.
Computational vertical transition energies and oscillator
strengths of 1a–3a are shown in Fig. 1b and listed in Table 3.
Figure 2. Molecular orbitals involved in transitions to the first and second excited single states of 1a, 2a and 3a.

6
Shoma Miyata et al.
The calculated vertical transition energies and oscillator strengths
of 1a–3a well reproduce the redshifted absorptions of 1c–3c in
comparison with 4c.
Molecular orbitals involved in the transitions to first and second
excited singlet states of 1a–3a are shown in Fig. 2. In all com-
pounds, HOMO–1 has n characters with electronic density local-
ized around the sulfur atom perpendicularly to the molecular
*
plane, whereas both HOMO and LUMO have p and p characters
with extended electronic density throughout the molecular plane,
1
respectively. For 1a, the first excited singlet (S ) state arises from
the transition from the n orbital localized on the sulfur atom
*
(
HOMO–1) to the p orbital (LUMO), and the calculated small
*
1
oscillator strength (f < 0.0001) indicates the forbidden S (np )
S0 transition. On the other hand, the transition to the second
*
excited singlet (S ) state is allowed pp transition (HOMO ?
2
LUMO) (f = 0.571). Thus, the intense absorption band of 1c
*
around 346 nm would be attributed to the S
2
(pp )
0
S transi-
tion. For 2a and 3a, the transition to the S and S states arises
1
2
*
from the allowed pp transition (HOMO ? LUMO) and forbid-
den np transition (HOMO–1 ? LUMO). Thus, the absorption
*
peak at the longest wavelength of 2c and 3c would be attributed to
*
the S
1
(pp )
0
S transition. The p and p* orbitals of 2a and 3a
were widely extended to the 8-position of the purine ring as shown
in Fig. 2, resulting in the redshifted absorption of 2c and 3c with
respect to 4c and 1c, respectively.
For 1a–3a, the T state was also optimized and shown in Fig-
1
ure S13 and Tables S4–S6. The optimized structures for 1a–3a at
the T state are also comparable to that at ground-state except for
1
6
6
the S atom of 1a at the T
1
state (only the S atom is out of molec-
state can
be assigned to the pp state, as shown in Figure S14. The T state
ular plane). The optimized structures for 1a–3a at the T
1
*
1
1
energies of 1a–3a are listed in Table 3. The calculated T energies
well agree with the experimental values, estimated from the maxi-
mum emission wavelength in the phosphorescence spectra of 1c–
3
c.
Time-resolved near-infrared emission spectroscopy
The quantum yield of singlet oxygen generation for the
thioguanosines was determined for exploration of these potential
drugs in photochemotherapy. Figure 3a shows the decay profiles
of singlet oxygen phosphorescence measured at around 1275 nm
Figure 3. (a) Decay profiles of singlet oxygen phosphorescence mea-
sured at around 1275 nm of thioguanosines and PN in acetonitrile solu-
tion. Signals are corrected for absorptance at excitation wavelength
by photosensitization with 1c, 2c and 3c in O
2
-saturated acetoni-
trile solutions. All signals decayed mono-exponentially, and their
(266 nm) and incident laser power. (b) Plots of the emission intensity
1
0
maxima (I
S
) immediately after laser irradiation in 3c solutions against
lifetimes were about 65 ls, well agreeing to the lifetime of O
2
*
0
incident laser power (I
and PN against the absorptance (1–10 ) at excitation wavelength
266 nm).
L S L
), and (c) plots of the I /I value of 1c, 2c, 3c
in acetonitrile (49,50). Since this signal was not detectable in Ar-
ꢁA
1
2
saturated solutions, the emission should be due to O *, gener-
(
ated by photosensitization with thioguanosines.
1
The quantum yields of O
2
* generation of thioguanosines were
determined in O
cally matched phenalenone (PN) solution (ΦD = 1.00 ꢀ0.03)
51). Individual phosphorescence traces were fitted using a single-
exponential function to estimate the emission intensity maxima
2
-saturated acetonitrile solutions relative to opti-
thioguanosines with that of PN, we were able to determine Φ
ues, with a high degree of accuracy, as 0.37 ꢀ0.01 for 1c,
value for 1c
D
val-
(
0.28 ꢀ0.01 for 2c and 0.33 ꢀ0.01 for 3c. The Φ
D
was close to those for 1a and 1b in the previous report (18,19).
0
0
immediately after laser irradiation (I ). The I value was plotted
These results further confirm that those thioguanosines generate
s
s
1
against the laser fluence (I
was observed between Is and I . This finding reveals that O *
L 2
L
) (Fig. 3b). A good linear relationship
O
2
* effectively through photosensitization.
It was noted that there are only very small differences in the
0
1
1
was generated by one-photon process through photosensitization
Φ
D
values among the thioguanosines (see Table 3). O
2
* is con-
by thioguanosines. The values of the slope obtained from these
sidered to generate through energy transfer from the T
1
state of
3
ꢁ
0
plots (I /I
L
) were plotted against the ground-state absorptance at
donor molecule to an oxygen molecule (X Σ
acceptor by collision each other, thus the Φ
depend on the following factors: the intersystem crossing
g
) as an energy
s
ꢁA
excitation wavelength (1–10 ), as shown in Fig. 3c. These plots
D
value should
also show good linear relationships. By comparing the slopes of

Photochemistry and Photobiology
7
quantum yield, the triplet lifetime of the sensitizer and the SD
value (a fraction of the triplet states quenched by dissolved oxy-
gen which gives rise to singlet oxygen formation). Generally, tri-
Figure S11. Phosphorescence spectra in optically matched
2
66 nm
(kex = 266 nm, A
= 0.4) 1c, 2c, and 3c glassy ethanol
matrix measured at 77 K. The spectrum of 2c was 10 times
plet states of pp* have been reported to give a S
range of 0.7–1.0, whereas it is ~0.3 for np* triplet states (52).
All the T state of 1c, 2c and 3c have pp* character in the
D
value of a
multiplied.
Figure S12. Optimized structures at the ground state of 1a,
2a, and 3a.
1
Franck–Condon region obtained by the TD–DFT calculation, and
Figure S13. Optimized structures at the triplet state of 1a, 2a,
and 3a.
its T1 energies are large enough to surpass vertical transition
1
3
ꢁ
energy of oxygen molecules (0.97 eV; a D
cussed above. Therefore, the differences in Φ
g
X Σ
g
), as dis-
Figure S14. Molecular orbitals at the T
3a.
1
state of 1a, 2a and
D
will depend on
the lifetime of each T state and/or quantum yields of intersys-
Table S1. Cartesian coordinates for optimized structure at the
ground state of 1a.
Table S2. Cartesian coordinates for optimized structure at the
ground state of 2a.
1
tem crossing to triplet manifolds. To gain the more detailed
information on the triplet state such as lifetime and quantum
yield, time-resolved spectroscopy is under way.
Table S3. Cartesian coordinates for optimized structure at the
ground state of 3a.
CONCLUSION
Table S4. Cartesian coordinates for optimized structure at the
Three novel thioguanosine derivatives (1c–3c) have been suc-
cessfully synthesized and characterized by various spectro-
scopies. The absorption bands of these thioguanosines are
found to be at longer wavelengths than those of unthiolated
guanosines (4b and 4c). Especially, 3c has the most redshifted
band with a large molar absorption coefficient, indicating that
T
T
T
1
1
1
state of 1a.
Table S5. Cartesian coordinates for optimized structure at the
state of 2a.
Table S6. Cartesian coordinates for optimized structure at the
state of 3a.
3
c is much more sensitive to the light penetrating into the
human skin. The redshifted spectra for thioanalogs were well
reproduced with the quantum chemical calculations. The T
REFERENCES
1
character was found to be a pp* character. In addition, the
1. Lepage, G. A. (1960) Incorporation of 6–thioguanine into nucleic
acids. Cancer Res. 20, 403–408.
1
thioguanosines generated O * effectively through photosensiti-
2
2
. Karran, P. and N. Attard (2008) Thiopurines in current medical prac-
tice: Molecular mechanisms and contributions to therapy–related can-
cer. Nature 8, 24–36.
D
zation (Φ = 0.28–0.37). Taken together, these results clearly
show that our reported thioguanosines have some potential for
photochemotherapy, as a photoactivatable genotoxic agent and/or
3. Rubin, Y. V., Y. P. Blagoi and V. A. Bokovoy (1995) 6–Thiogua-
nine luminescence probe to study DNA and low–molecular–weight
systems. J. Fluoresc. 5, 263–272.
a photosensitizer for PDT. Although the S
D
value for the T
values
1
(
pp*) was known to be high, the small difference in Φ
D
4
. Qu, P., H. Lu, X. Ding, Y. Tao and Z. Lu (2009) Influences of urea
and guanidine hydrochloride on the interaction of 6–thioguanine with
bovine serum albumin. Spectrochim. Acta. A Mol. Biomol. Spectrosc.
74, 1224–1228.
among the thioguanosines is likely to be dependent on the inter-
system crossing quantum yield and/or T1 state lifetime of each
individual thioguanosine.
5
6
. Massey, A., Y.-Z. Xu and P. Karran (2001) Photoactivation of DNA
thiobases as a potential novel therapeutic option. Curr. Biol. 11,
1142–1146.
. Brem, R., F. Li and P. Karran (2009) Reactive oxygen species gener-
ated by thiopurine/UVA cause irreparable transcription–blocking
DNA lesions. Nucleic Acids Res. 37, 1951–1961.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online
version of this article:
1
7. Gay, L., M. R. Miller, P. B. Ventura, V. Devasthali, Z. Vue, H. L.
Thompson, S. Temple, H. Zong, M. D. Cleary, K. Stankunas and C.
Q. Doe (2013) Mouse TU tagging: A chemical/genetic intersectional
method for purifying cell type–specific nascent RNA. Genes Dev.
27, 98–115.
8. O’Donovan, P., C. M. Perrett, X. Zhang, B. Montaner, Y.-Z. Xu, C.
A. Harwood, F. Hanaoka and P. Karran (2005) Azathioprine and
UVA Light generate mutagenic oxidative DNA damage. Science
309, 1871–1874.
9. Reelfs, O., Y.-Z. Xu, A. Massey, P. Karran and A. Storey (2007)
Thiothymidine plus low–dose UVA kills hyperproliferative human
skin cells independently of their human papilloma virus status. Mol.
Cancer Ther. 6, 2487–2495.
Figure S1. H NMR spectrum of 1c in dimethylsulfoxide–d
6
6
6
solution.
Figure S2. C NMR spectrum of 1c in dimethylsulfoxide–d
solution.
Figure S3. H NMR spectrum of 2c in dimethylsulfoxide–d
solution.
Figure S4. C NMR spectrum of 2c in dimethylsulfoxide–d6
solution.
Figure S5. H NMR spectrum of 3c in dimethylsulfoxide–d
solution.
Figure S6. C NMR spectrum of 3c in dimethylsulfoxide–d
solution.
Figure S7. H NMR spectrum of 4c in dimethylsulfoxide–d
solution.
Figure S8. C NMR spectrum of 4c in dimethylsulfoxide–d
solution.
13
1
13
1
6
6
6
6
13
1
0. Lepage, G. A. and M. Jones (1961) Further studies on the mecha-
nism of action of 6–thioguanine. Cancer Res. 21, 1590–1594.
11. Ren, X., F. Li, G. Jeffs, X. Zhan, Y.-Z. Xu and P. Karran (2010)
Guanine sulphinate is a major stable product of photochemical oxi-
dation of DNA 6–thioguanine by UVA irradiation. Nucleic Acids
Res. 38, 1832–1840.
1
13
1
2. Pridgeon, S. W., R. Heer, G. A. Taylor, D. R. Newell, K. O’Toole,
M. Robinson, Y.-Z. Xu, P. Karran and A. V. Boddy (2011) Thio-
thymidine combined with UVA as a potential novel therapy for blad-
der cancer. Br. J. Cancer 104, 1869–1876.
Figure S9. Schematic diagram of the experimental setup for
the time–resolved near IR emission measurement.
Figure S10. HPLC chart for (a) 1c, (b) 2c, (c) 3c and (d) 4c.

8
Shoma Miyata et al.
1
3. Reelfs, O., P. Karran and A. R. Young (2012) 4–Thiothymidine sen-
sitization of DNA to UVA offers potential for a novel pho-
tochemotherapy. Photochem. Photobiol. Sci. 11, 148–154.
Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Fores-
man, J. V. Ortiz, J. Cioslowski and D. J. Fox. (2013) Gaussian 09,
Revision D.01. Gaussian, Inc., Wallingford, CT.
1
1
4. Boulnois, J. L. (1986) Photophysical processes in recent medical
laser developments: A review. Lasers Med. Sci. 1, 47–66.
5. Sanli, S., Y. Altun and G. Guven (2014) Solvent effects on pK val-
a
ues of some anticancer agents in acetonitrileꢁwater binary mixtures.
J. Chem. Eng. Data 59, 4015–4020.
1
1
1
6. Eccleston, J. F. and P. M. Bayley (1980) Circular dichroic spectra of
34. Gerster, J. F., J. W. Jones and R. K. Robins (1963) Purine nucleo-
sides. IV. The synthesis of 6–halogenated 9–b–D–ribofuranosylpuri-
nes from inosine and guanosine. J. Org. Chem. 28, 945–948.
35. Lewis, L. R., R. K. Robins and C. C. Cheng (1964) The preparation
and antitumor properties of acylated derivatives of 6–thiopurine ribo-
sides. J. Med. Chem. 7, 200–204.
36. Rossi, R. R. and L. M. Lerner (1973) Specificity of inhibition of
adenosine deaminase by trialcohols derived from nucleosides. J.
Med. Chem. 16, 457–460.
6
–thioguanosine nucleotides and their complexes with myosin sub-
fragment. Biochemisty 19, 5050–5056.
7. Reichardt, C., C. Guo and C. E. Crespo-Hern ꢁa ndez (2011) Excited–
state dynamics in 6–thioguanosine from the femtosecond to
microsecond time scale. J. Phys. Chem. B 115, 3263–3270.
8. Pollum, M., L. A. Ortiz-Rodr ꢁı guez, S. Jockusch and C. E. Crespo-
0
Hern ꢁa ndez (2016) The triplet state of 6–thio–2 –deoxyguanosine:
Intrinsic properties and reactivity toward molecular oxygen. Pho-
tochem. Photobiol. 92, 286–292.
37. Naito, T., K. Ueno and F. Ishikawa (1964) Studies on nucleosides
and nucleotides. VII. Synthesis of 2–amino–6–substituted–9–b–D–ri-
bofuranosylpurines. Chem. Pharm. Bull. 12, 951–954.
1
9. Zhang, Y., X. Zhu, J. Smith, M. T. Haygood and R. Gao (2011)
Direct observation and quantitative characterization of singlet oxygen
in aqueous solution upon UVA excitation of 6–thioguanines. J. Phys.
Chem. B 115, 1889–1894.
38. Ikaunieks, M. and M. Madre (2002) Reinvestigation of the reaction
of 8–bromoguanine derivatives with sodium thiosulfate. J. Chem.
Res. 8, 226–227.
2
0. Holmes, R. E. and R. K. Robins (1964) Purine nucleosides. VII.
Direct bromination of adenosine, deoxyadenosine, guanosine, and
related purine nucleosides. J. Am. Chem. Soc. 86, 1242–1245.
1. Michael, M. A., H. B. Cottam, D. F. Smee, R. K. Robins and G. D.
Kini (1993) Alkylpurines as immunopotentiating agents. Synthesis
and antiviral activity of certain alkylguanines. J. Med. Chem. 36,
2
39. Golubeva, N. A. and A. V. Shipitsyn (2007) The synthesis of N ,
6
N –substituted diaminopurine ribosides. Russ. J. Bioorg. Chem. 33,
2
562–568.
40. Torii, T., H. Shiragami, K. Yamashita, Y. Suzuki, T. Hijiya, T.
Kashiwagi and K. Izawa (2006) Practical syntheses of penciclovir
and famciclovir from N2–acetyl–7–benzylguanine. Tetrahedron 62,
5709–5716.
41. Gr €u newald, C., T. Kwon, N. Piton, U. F }o rster, J. Wachtveitl and J.
W. Engels (2008) RNA as scaffold for pyrene excited complex.
Bioorg. Med. Chem. 16, 19–26.
3
431–3436.
2. Gerster, J. F., B. C. Hinshaw, R. K. Robins and L. B. Townsed
1968) Purine nucleosides. XIX. The synthesis of certain 8–chlorop-
urine nucleosides and related derivatives. J. Org. Chem. 33, 1070–
073.
2
(
1
2
2
2
3. Bookser, B. C. and N. B. Raffaele (2007) High–throughput five min-
ute microwave accelerated glycosylation approach to the synthesis of
nucleoside libraries. J. Org. Chem. 72, 173–179.
42. Lin, T., J. Cheng, K. Ishiguro and A. C. Sartorelli (1985) 8–Substi-
0
tuted guanosine and 2 –deoxyguanosine derivatives as potential
inducers of the differentiation of friend erythroleukemia cells. J.
Med. Chem. 28, 1194–1198.
4. Jayanth, N., S. Ramachandran and M. Puranik (2009) Solution struc-
ture of the DNA damage lesion 8–oxoguanosine from ultraviolet res-
onance Raman spectroscopy. J. Phys. Chem. A 113, 1459–1471.
5. Zhang, Y., J. Dood, A. Beckstead, J. Chen, X. Li, C. J. Burrows, Z.
Lu, S. Matsika and B. Kohler (2013) Ultrafast excited–state dynam-
43. Reitz, A. B., M. G. Goodman, B. L. Pope, D. C. Argentieri, S. C.
Bell, L. E. Burr, E. Chourmouzis, J. Come, J. H. Goodman, D. H.
Klaubert, B. E. Maryanoff, M. E. McDonnel, M. S. Rampulla, M. R.
Schott and R. Chen (1994) Small–molecule immunostimulants. Syn-
thesis and activity of 7,8–disubstituted guanosines and structurally
related compounds. J. Med. Chem. 37, 3561–3578.
0
ics and vibrational cooling of 8–oxo–7, 8–dihydro–2 –deoxyguano-
2
sine in D O. J. Phys. Chem. A 117, 12851–12857.
2
6. Lu, Z., A. A. Beckstead, B. Kohler and S. Matsika (2015) Excited
44. Zhang, X., J. Wang and Y.-Z. Xu (2013) Systematic assignment of
NMR spectra of 5–substituted–4–thiopyrimidine nucleosides. Magn.
Reson. Chem. 51, 523–529.
state relaxation of neutral and basic 8–oxoguanine. J. Phys. Chem. B
119, 8293–8301.
0
2
7. Harada, Y., T. Suzuki, T. Ichimura and Y.-Z. Xu (2007) Triplet for-
mation of 4–thiothymidine and its photosensitization to oxygen stud-
ied by time–resolved thermal lensing technique. J. Phys. Chem. B
45. Zhang, X. and Y.-Z. Xu (2011) NMR and UV studies of 4–thio–2 –
deoxyuridine and its derivatives. Molecules 16, 5655–5664.
46. Isozaki, T., J. Ikemi and T. Suzuki (2012) Multi–photon excitation
studies of 6–thioguanosine as a potential agent for photodynamic
therapy. Photomed. Photobiol. 34, 67–68.
111, 5518–5524.
2
8. Kuramochi, H., T. Kobayashi, T. Suzuki and T. Ichimura (2010)
Excited–state dynamics of 6–aza–2–thiothymine and 2–thiothymine:
Highly efficient intersystem crossing and singlet oxygen photosensiti-
zation. J. Phys. Chem. B 114, 8782–8789.
9. Suzuki, T., S. Miyata and T. Isozaki (2015) Absorption spectra of 8–
oxoguanosine and 8–thioguanosine. Photomed. Photobiol. 37, 21–22.
0. Pollum, M., S. Jockusch and C. E. Crespo-Hern ꢁa ndez (2014) 2,4–
Dithiothymine as a potent UVA chemotherapeutic agent. J. Am.
Chem. Soc. 136, 17930–17933.
1. Pollum, M., S. Jockuschb and C. E. Crespo-Hern ꢁa ndez (2015)
Increase in the photoreactivity of uracil derivatives by doubling
thionation. Phys. Chem. Chem. Phys. 17, 27851–27861.
2. Haynes, W. M. (2011) Handbook of Chemistry and Physics, 92nd
edn. CRC Press, Boca Raton, FL.
3. Frisch, M. J., G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.
A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A.
F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima,
Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J.
E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N.
Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K.
47. Isozaki, T., S. Ando and T. Suzuki. Simultaneous two–photon
0
0
0
absorption of thioguanosine analogue 2 ,3 ,5 –tri–O–acetyl–6,8–
dithioguanosine with its potential application to photodynamic ther-
apy. In preparation.
2
48. Stewart, M. J., J. Leszczynski, Y. V. Rubin and Y. P. Blagoi (1997)
Tautomerism of thioguanine: From gas phase to DNA. J. Phys.
Chem. A 101, 4753–4760.
49. Rodgers, M. A. J. (1983) Solvent–induced deactivation of singlet
oxygen: Additivity relationships in nonaromatic solvents. J. Am.
Chem. Soc. 105, 6201–6205.
3
3
50. Clennan, E. L., L. J. Noe, T. Wen and E. Szneler (1989) Solvent
effects on the ability of amines to physically quench singlet oxygen
as determined by time–resolved infrared emission studies. J. Org.
Chem. 54, 3584–3587.
3
3
51. Mart ꢁı , C., O. J €u rgens, O. Cuenca, M. Casals and S. Nonell (1996)
1
Aromatic ketones as standards for singlet molecular oxygen O
2
( D
g
)
photosensitization. Time–resolved photoacoustic and near–IR emis-
sion studies. J. Photochem. Photobiol., A 97, 11–18.
52. Schweitzer, C. and R. Schmidt (2003) Physical mechanisms of gen-
eration and deactivation of singlet oxygen. Chem. Rev. 103, 1685–
1758.