Generation of thiyl radicals by the photolysis of 5-iodo-4-thiouridine

Article abstract of DOI:10.1021/jo048496g

(Chemical Equation Presented) The photochemistry of 2′,3′, 5′-tri-O-acetyl-5-iodo-4-thiouridine (3) in deoxygenated 1:1 CH ₃CN-H₂O pH 5.8 (phosphate buffer) solution has been studied by means of steady-state and nanosecond laser flash photolysis methods. Under steady-state irradiation (λ ≥ 334 nm), the stable photoproducts were iodide ion, 2′,3′,5′-tri-O-acetyl-4-thiouridine (4), and two disulfides. The disulfides were the symmetrical bis-(2′,3′, 5′-tri-O-acetyl-5-iodo-4-thiouridine) (5) and unsymmetrical 6, which contains both 4-thiouridine and 5-iodo-4-thiouridine residues. The formation of the dehalogenated photoproduct suggests that C(S)-I bond cleavage is a primary photochemical step. Attempts to scavenge the resulting C(5)-centered radical by suitable addends, bis-(N-α-acetyl)cystine-bis-N-ethylamide or benzene, were unsuccessful. Analysis of the photoproducts formed under these conditions showed that the S-atom is the reactive center. The photoproduct 4, obtained by irradiation of 3 in CD₃-CN-H₂O, followed by reversed-phase HPLC isolation using nonlabeled eluents, did not contain deuterium. An analogous experiment performed in CH₃CN-D₂O gave deuterated product 4-d with 88% of the deuterium incorporated at C(5). Transient absorption observed upon laser excitation (λ = 308 nm) of 3 was assigned to the 4-uridinylthiyl radical on the basis of the similarity of this spectrum with that obtained upon laser photolysis of the disulfide: bis-(2′, 3′,5′-tri-O-acetyl-4-thiouridine) 14. On the basis of the results of steady-state and laser photolysis studies, a mechanism of the photochemical reaction of 3 is proposed. The key mechanistic step is a transformation of the C(5)-centered radical formed initially by C(5)-I bond cleavage into a long-lived S-centered radical via a 1,3-hydrogen shift. Theoretical calculations confirmed that the long-lived S-centered radical is the most stable radical derived from the 4-thiouracil residue.

Full text of DOI:10.1021/jo048496g

Generation of Thiyl Radicals by the Photolysis of
5
-Iodo-4-thiouridine
,†
†
†
‡
Gra z˘ yna Wenska,* Katarzyna Taras-Go s´ li n´ ska, Bohdan Skalski, Gordon L. Hug,
‡
†
Ian Carmichael, and Bronislaw Marciniak
Faculty of Chemistry, A. Mickiewicz University, Grunwaldzka 6, 60-780 Pozna n´ , Poland, and
Radiation Laboratory, University of Notre Dame, Notre Dame, Indiana 46556
gwenska@amu.edu.pl
Received August 27, 2004
The photochemistry of 2′,3′,5′-tri-O-acetyl-5-iodo-4-thiouridine (3) in deoxygenated 1:1 CH CN-
3
2
H O pH 5.8 (phosphate buffer) solution has been studied by means of steady-state and nanosecond
laser flash photolysis methods. Under steady-state irradiation (λ g 334 nm), the stable photoproducts
were iodide ion, 2′,3′,5′-tri-O-acetyl-4-thiouridine (4), and two disulfides. The disulfides were the
symmetrical bis-(2′,3′,5′-tri-O-acetyl-5-iodo-4-thiouridine) (5) and unsymmetrical 6, which contains
both 4-thiouridine and 5-iodo-4-thiouridine residues. The formation of the dehalogenated photo-
product suggests that C(5)-I bond cleavage is a primary photochemical step. Attempts to scavenge
the resulting C(5)-centered radical by suitable addends, bis-(N-R-acetyl)cystine-bis-N-ethylamide
or benzene, were unsuccessful. Analysis of the photoproducts formed under these conditions showed
that the S-atom is the reactive center. The photoproduct 4, obtained by irradiation of 3 in CD
CN-H O, followed by reversed-phase HPLC isolation using nonlabeled eluents, did not contain
deuterium. An analogous experiment performed in CH CN-D O gave deuterated product 4-d with
8% of the deuterium incorporated at C(5). Transient absorption observed upon laser excitation (λ
3
-
2
3
2
8
)
308 nm) of 3 was assigned to the 4-uridinylthiyl radical on the basis of the similarity of this
spectrum with that obtained upon laser photolysis of the disulfide: bis-(2′,3′,5′-tri-O-acetyl-4-
thiouridine) 14. On the basis of the results of steady-state and laser photolysis studies, a mechanism
of the photochemical reaction of 3 is proposed. The key mechanistic step is a transformation of the
C(5)-centered radical formed initially by C(5)-I bond cleavage into a long-lived S-centered radical
via a 1,3-hydrogen shift. Theoretical calculations confirmed that the long-lived S-centered radical
is the most stable radical derived from the 4-thiouracil residue.
Introduction
CHART 1
Photochemical reactions of 4-thiouridine (1, Chart 1)1,2
and 5-iodouridine (2)3,4 have been extensively studied by
*
Corresponding author. Phone: +48-61-829-13-51. Fax: +48-61-
8
65-80-08.
†
A. Mickiewicz University.
University of Notre Dame.
‡
(
1) Favre, A. In Bioorganic Photochemistry; Morrison, H., Ed.;
Wiley: New York, 1990; Vol. 1, pp 379-425.
(
2) Favre, A.; Saintome, C.; Fourrey, J.-L.; Clivio, P.; Laugaa, P. J.
Photochem. Photobiol. B: Biol. 1998, 42, 109-124.
3) Norris, C. L.; Meisenheimer, P. L.; Koch, T. H. J. Am. Chem.
Soc. 1996, 118, 5796-5803.
4) G o¨ rner, H. J. Photochem. Photobiol. A: Chem. 1993, 72, 197-
08.
(
steady-state photolysis methods. These compounds are
of interest because they have wide application in various
studies of biological systems. In particular, both are used
(
2
10.1021/jo048496g CCC: $30.25 © 2005 American Chemical Society
982
J. Org. Chem. 2005, 70, 982-988
Published on Web 01/07/2005

Photolysis of 5-Iodo-4-thiouridine
as photo-cross-linking agents in nucleic acids or in nucleic
acid-protein complexes.5,6
The main photoreactions of 1 are photocycloaddition
(
forming thiethane) and H-abstraction. The results of
3
,4
photochemical studies of the nucleoside 2, 5-iodour-
7
-12
13,14
acil,
and other derivatives of pyrimidine
indicate
that the primary photochemical event, following elec-
tronic excitation of the chromophore, is C-I bond ho-
molysis leading to the 5-uracilyl radical.3
,13,14
Subsequent
reactions of this radical and eventually the structures of
isolated, stable photoproducts depend on the presence of
various reactive agents such as H atom donors, radical
scavengers, and olefinic or aromatic compounds present
in the irradiated systems. In the absence of reactive
addends in deoxygenated aqueous or acetonitrile-water
solutions, the major product obtained from 2 or 5-iodo-
FIGURE 1. UV absorption spectra of 1 (dotted line), 2 (broken
line), and 3 (solid line) in a 1:1 mixture of CH CN-H O pH
.8 (phosphate buffer).
3,4,12
3
2
uracil was uridine or uracil, respectively.
Irradiation
5
15
of 2′,3′-O-isopropylidene-5-iodouridine or 1,3-dimethyl-
1
6
5
-iodouracil in the presence of benzene, pyrene, or
study the photochemistry of this latter compound in more
detail. In this paper, we report the results of the steady
state and laser flash photolysis of 2′,3′,5′-tri-O-acetyl-5-
phenantrene gave uracil derivatives with an appropriate
aromatic substituent at the C-5 position of the uracil ring
in addition to the dehalogenated pyrimidine or its nucleo-
side. The (C-5)-uridinyl radical is trapped by bis(2-amino-
iodo-4-thiouridine (3) in deoxygenated 1:1 CH
3
2
CN-H O
pH 5.8 (phosphate buffer) solution. In the nucleoside 3,
the sugar hydroxyl groups are converted into O-acetyl
groups to improve the solubility of the compound.
1
-thioethane) to give 5-(2-aminoethylthio)-uracil9,11,17 upon
irradiation of 5-iodouracil in the presence of the disulfide.
Although several nucleic acid-protein photo-cross-
linking agents have been developed,1 there is still a need
for new, long-wavelength-absorbing, highly reactive, and
selective probes. Recently, we have started systematic
studies of the photophysical and photochemical properties
of uridine derivatives that contain both CdS and CdC-X
fragments, where X ) halogen. In the series of 4-thio-
uridine derivatives with 5-fluoro-, 5-chloro-, and 5-bromo-
substituents, it was found that the presence of the CdS
function in the molecules dominated their photochemistry
in aqueous solution. These compounds give photoproducts
resulting from cycloaddition of the CdS group of the
excited molecule to an ethylenic fragment of the substrate
in its ground state.18 The results of preliminary photo-
chemical studies of the remaining member of the family,
,5
Results and Discussion
Steady-State Photochemistry. The electronic ab-
sorption spectra of nucleosides 1-3 in a 1:1 mixture of
CH
Figure 1. Among these uridine derivatives, the nucleoside
absorbs at the longest wavelengths. The shape of the
3 2
CN-H O pH 5.8 (phosphate buffer) are presented in
3
absorption bands of 3 resembles that of 1. The similarity
of the spectra suggests that 3, like 1, exists in solution
in the thione tautomeric form. The absorption band with
its maximum at λmax ) 345 nm is due to the strong π f
π* (S
transitions in molecule 3.
In aqueous media at pH ≈ 6, nucleoside 3 exists as a
neutral molecule. The pK value for the neutral molecule-
anion equilibrium was determined to be pK ) 7.7 ( 0.1
see Supporting Information) by spectrophotometric ti-
tration. The nucleoside 3 is a stronger acid than 1 (pK
0 2 0 1
f S ) and weak n f π* (S f S ) electronic
19
4
i.e., 5-iodo-substituted derivative, in CCl solution indi-
a
cated that the nucleoside behaves differently and under-
goes dehalogenation.1
9,20
This finding prompted us to
a
(
a
(
5) Meisenheimer, K. M.; Koch, T. H. Crit. Rev. Biochem. Mol. Biol.
1
1
997, 32, 101-140.
) 8.2), which does not contain an iodo substituent, and
(
(
(
6) Favre, A.; Fourrey, J.-L. Acc. Chem. Res. 1995, 28, 375-382.
7) Gilbert, E.; Wagner, G. Z. Naturforsch., B 1972, 27, 644-648.
8) Rupp, W. D.; Prusoff, W. H. Biochem. Biophys. Res. Commun.
21
2
(deoxynucleoside, pK
a
) 8.34), an oxygen analogue of
the thione 3.
1
1
4
965, 18, 145-151.
9) Rupp, W. D.; Prusoff, W. H. Biochem. Biophys. Res. Commun.
965, 18, 158-164.
10) Gilbert, E.; Schulte-Frohlinde, D. Z. Naturforsch., B 1970, 25,
92-495.
11) Gilbert, E.; Wagner, G.; Schulte-Frohlinde, D. Z. Naturforsch.,
B 1971, 26, 209-213.
12) Gilbert, E.; Wagner, G.; Schulte-Frohlinde, D. Z. Naturforsch.,
B 1972, 27, 501-504.
13) Wang, S. Y. In Photochemistry and Photobiology of Nucleic
Acids; Wang, S. Y., Ed.; Academic Press: New York, 1976; Vol. 1, pp
Irradiation (λ > 334 nm) of a deoxygenated solution of
in the concentration range [3] ) 0.1-1 mM in a 1:1
(
3
mixture of CH
3
2
CN-H O pH 5.8 (phosphate buffer) leads
(
to three organic photoproducts, 4-6, and iodide ions; see
Scheme 1. The structures of the photoproducts were
established by the spectral data and confirmed by
independent synthesis. Disulfides 5 and 6 are new
compounds. They are quite stable (several days) as solids
at 5 °C. However, the unsymmetrical disulfide 6, in
contrast to symmetrical compound 5, is unstable in
(
(
(
2
6
2
2
95-356.
(14) Rahn, R. O.; Sellin, H. G. Photochem. Photobiol. 1983, 37, 661-
64.
(15) Saito, I.; Ito, S.; Shinmura, T. Tetrahedron Lett. 1980, 21, 2813-
816.
(19) Taras-Goslinska, K.; Wenska, G.; Skalski, B.; Maciejewski, A.;
Burdzinski, G.; Karolczak, J. J. Photochem. Photobiol. A: Chem. 2004,
168, 227-233.
(20) Preliminary results of the work were presented at Annual
Conference of Polish Chemical Society in Katowice, Poland, 2001.
(21) Meisenheimer, K. M.; Meisenheimer, P. L.; Willis, M. C.; Koch,
T. H. Nucl. Acid Res. 1996, 24, 981-982.
(16) Seki, K.; Matsuda, K. Nucl. Acid Res. Symp. Series 1988, 20,
5-26.
(
17) Gilbert, E. Z. Naturforsch., B 1973, 28, 805-809.
(
18) Wenska, G.; Taras-Goslinska, K.; Lamparska-Kupsik, K.; Skal-
ski, B.; Gdaniec, M.; Gdaniec, Z. J. Chem. Soc., Perkin Trans. 1 2002,
5
3-57.
J. Org. Chem, Vol. 70, No. 3, 2005 983

Wenska et al.
CHART 2
FIGURE 2. Production of 4 (full circles), disulfides 5 (open
circles), and 6 (triangles) during the irradiation λ g 334 nm
3 2
of 3 (0.5 mM) in a 1:1 mixture CH CN-H O pH 5.8 (phosphate
buffer) in the absence of oxygen.
SCHEME 1
common reaction of organic halides, e.g., aromatic and
heteroaromatic compounds, containing a variety of func-
3
,22,23
tional groups in addition to iodine substituents.
However, to the best of our knowledge, there are no
reports in the literature on the photochemistry of iodine-
substituted aromatic thiones. Photoinduced homolysis of
C(5)-I in 3, in analogy to the photochemistry of its
oxygen congener 2,3
,4,7-9,11-16
should lead to the formation
the C(5)-centered radical 4-thiouridine-5-yl radical 7
Chart 2) and an iodine atom. However, attempts to trap
(
radical 7 by suitable reagents were unsuccessful. The
trapping reagents were chosen as suitable on the basis
that the chemistry of radical 7 should be similar to the
uridine-5-yl radical 8 (Chart 2), an analogue of 7 that
has a C(4)dO group instead of a C(4)dS group.
solution even at room temperature as indicated by NMR
spectroscopy. Therefore, its complete spectral character-
In particular, irradiation of 3 in the presence of bis-
(
N-R-acetyl)cystine-bis-N-ethylamide in aqueous media
1
3
ization ( C NMR spectrum) could not be performed.
Iodide anions, in the irradiated solutions, were detected
by HPLC. In addition, they were identified on the basis
of their absorption spectrum measured with an HPLC-
and irradiation of 3 in benzene solution did not lead to
C(5)-substituted derivatives of 4-thiouridine, i.e., photo-
products 9 or 10 (Chart 2), both of which are expected
on the basis of the above analogy. In the presence of an
equimolar amount of bis-(N-R-acetyl)cystine-bis-N-ethyl-
amide, the disulfides 11 and 12 were formed (Chart 2)
in addition to the photoproducts identified when the
nucleoside 3 was irradiated without addends (Scheme 1).
The resulting disulfides 11 and 12 were unstable during
the preparative-scale isolation procedure. However, the
UV and MS spectral data (Experimental Section) pro-
vided sufficient information about their structure. The
structural assignment was confirmed by the quantitative
splitting of disulfides 11 and 12 to N-R-acetyl-cysteine
N-ethylamide and equimolar amounts of the respective
nucleosides 4 and 3 by treatment with dithiotreitol,
Cleland’s disulfide reducing agent.
-
-1
PDD detector (I : λmax ) 226 nm, ꢀmax ) 13 500 M
-
1 14
3
cm ) and by their characteristic reaction with AgNO .
All of the organic photoproducts absorb the incident
light and are photolabile. However, as indicated by the
linear plots presented in Figure 2, they are all primary
photoproducts and did not undergo secondary reactions
when the irradiation of 3 was performed at low (<40%)
conversion of the substrate. Under these conditions, they
account for 80-95% of the substrate that reacted.
The quantum yield of disappearance of 3 was es-
sentially independent of the excitation wavelength (λ )
3
13 or 366 nm) and concentration in the 0.1-1 mM
range.
In an attempt to rationalize the formation of the
When 3 in benzene solution was irradiated to 33%
conversion, the major organic photoproducts were the
photoproducts presented in Scheme 1, it appeared rea-
sonable to assume that the primary reaction of the
excited molecule 3 is photodissociation of its C-I bond.
Photoinduced homolysis of carbon-iodide bonds is a
(
22) Kropp, P. J. Acc. Chem. Res. 1984, 17, 131-137.
(23) Marconi, G. J. Photochem. 1979, 11, 385-391.
984 J. Org. Chem., Vol. 70, No. 3, 2005

Photolysis of 5-Iodo-4-thiouridine
TABLE 1. Quantum Yielda of Substrate Decomposition,
O(3), and Formation of Photoproducts, O(4), O(5), and O(6),
upon Irradiation of 3 at λ ) 366 nm in a Deoxygenated
1
:1 Mixture of CH3CN-H2O pH 5.8 (Phosphate Buffer)
solvent
[3] mM
φ(3)
φ(4)
φ(5)
φ(6)
CH3CN-H2O pH 5.8
CH3CN-H2O pH 5.8
C6H6
0.12
1.20
0.12
0.0039 0.0017 0.0006 0.0007
0.0037 0.0018 0.0006 0.0005
0.0028 0.0012 0.0003 0.0002
a
mean of three independent measurements (error ( 20%).
SCHEME 2
FIGURE 3. (a) Transient absorption spectra following laser
flash photolysis (λexc ) 308 nm) of 3 ([3] ) 0.043 mM) in a
deoxygenated 1:1 mixture of CH CN-H O pH 5.8 (0.01 M
3 2
phosphate buffer); triangles 5-20 µs, circles 20-40 µs after
the laser pulse. (b) Time profile of absorption at 450 nm.
same as in the 1:1 mixture of CH
3
CN-H
2
O pH 5.8
(phosphate buffer), although they were formed with
smaller quantum yields (Table 1). None of the several
minor photoproducts were identified as 2′,3′,5′-tri-O-
acetyl-5-phenyl-4-thiouridine (10, Chart 2), as would
normally be expected when the solvent is benzene. A
sample of the latter compound was synthesized indepen-
hydrogen (N-H) from the molecule 3 rather than a
hydrogen from the solvent. This leads to the formation
of the C(5)-H bond in the photoproduct 4.
Laser Flash Photolysis. Solutions of 3 ([3] ) 0.043
dently (Experimental Section). Upon irradiation of 3 in
mM) in 1:1 mixtures of CH
3
2
CN-H O pH 5.8 (phosphate
-
benzene, the inorganic photoproducts were I
2
and I . The
buffer) were irradiated at λ ) 308 nm using an excimer
laser. Transient absorptions were observed in the region
350-600 nm. The resulting spectrum is presented in
Figure 3a. It exhibits a broad band centered at λ ) 430
nm. The absorption decays monoexponentially, and the
lifetime τ ) 95.4 µs was determined from the time profile
of the 450 nm absorbance (Figure 3b). The long lifetime
of the species responsible for the observed absorption
precludes it from being the lowest excited triplet state
of 3. This conclusion follows because the room-tempera-
presence of the former was suggested by the pink color
of the irradiated solution. Iodide anions were detected
by a reversed-phase HPLC analysis of the irradiated
sample after the volatile I was removed in the process
2
of evaporating the sample to dryness.
Finally, to identify the source of hydrogen at the C(5)
position of the photoproduct 4, irradiations of 3 were
performed in deuterium-labeled solvents. The analysis
was limited to the photoproduct 4 (Scheme 2). The
procedure involved irradiation in CD
3
CN-H
2
O pH 5.8
ture phosphorescence lifetime of compound 3 in CCl
4
was
(
phosphate buffer) and isolation of the photoproduct 4
determined to be 0.024 µs, and its triplet lifetime is
1
9,24
by HPLC, which was performed in nonlabeled aqueous
expected to be even shorter in aqueous solution.
media. The result was 2′,3′,5′-tri-O-acetyl-4-thiouridine
Identification of the species responsible for the ob-
served transient absorption was hampered by the lack
of any reported spectrum of C-centered radicals derived
from heterocyclic thione, which could serve as a model
for C(5)-centered radical 7 (Chart 2). On the other hand,
the spectrum of the thiyl radical 13 (Scheme 3) was easy
to obtain. Laser flash photolysis of the readily available
disulfide 14 (Experimental Section) is expected to give
thiyl radicals 13. Disulfides are known to undergo
(
4), which did not contain any deuterium as determined
by mass spectrometric analysis. In contrast, an analogous
experiment performed in CH CN-D O pH 5.8 (phosphate
3
2
buffer), with the identical isolation procedure as the other
deuterated solvent, gave 2′,3′,5′-tri-O-acetyl-4-thiouridine
with 88% incorporation of deuterium (4-d, Scheme 2). The
deuterium was introduced into the C(5) position of the
1
uracil ring of 4-d as shown by H NMR spectroscopy. In
a test experiment, it was confirmed that, in this solvent
system, no dark (C-5)-H/D exchange in 4 molecules takes
place in the absence of light. The results of the labeling
experiments point to the involvement of an exchangeable
(
24) Taras-Goslinska, K.; Wenska, G.; Skalski, B.; Maciejewski, A.;
Burdzinski, G.; Karolczak, J. Photochem. Photobiol. 2002, 75, 448-
456.
J. Org. Chem, Vol. 70, No. 3, 2005 985

Wenska et al.
SCHEME 4
FIGURE 4. Transient absorption spectrum following laser
flash photolysis (λexc ) 308 nm) of 14 ([14] ) 0.05 mM) in a
3 2
deoxygenated 1:1 mixture CH CN-H O pH 5.8 (0.01 M
phosphate buffer); 5-10 µs after the laser pulse.
SCHEME 3
lived thiyl radical 13 by the 1,3-hydrogen shift: N(3)-H
f C(5)-H. The deuterium-labeling evidence that the
exchangeable hydrogen (N-H) of 3 appears as nonex-
changeable C-H in photoproduct 4 supports the proposed
1
,3-hydrogen shift in Scheme 4.
Presumably this shift is not a one-step process because
of the unfavorable spatial arrangement of the orbitals.
An alternative explanation would be the tautomerization
of the initially formed C(5)-centered radical 7 to the thiol
form, followed by a 1,2-H shift from the C-SH to the
neighboring C(5) radical center. This would lead to the
more stable thiyl radical 13. Furthermore, the DFT
calculations support this mechanistic step. The DFT
calculation was performed on simplified models of radi-
cals 7 and 13. The simplified models replaced the ribosyl
substituent with a methyl group. The results show that
the S-centered (thiyl) radical is more stable than the
thione C(5)-centered radical or its tautomeric thiol form
homolytic scission of their S-S bond upon laser excitation
as well as in steady-state photolysis.2
5-27
The transient absorption spectrum observed after laser
photolysis of the disulfide 14 is presented in Figure 4.
The shape of the spectrum and the location of the
maximum closely resemble that observed after laser flash
photolysis of 3. Thus, we conclude that the only observ-
able species in the transient spectrum of 3 is the thiyl
radical 13. The failure to observe C(5)-centered radical
7
could be due to it being short-lived and/or to its
-1
by 80 and 135 kJ mol , respectively, in vacuo. In
spectrum being located in a shorter wavelength region.
By analogy to the properties of the radicals formed from
N-hydroxypyridine 2(1H)-thione, the former possibility
-1
acetonitrile, the thiyl radical is about 70 kJ mol more
stable than the thione C(5)-centered radical and about
-1
1
35 kJ mol more stable than the tautomeric thiol form
(radical 7 being short-lived) appears likely. In particular,
of the C(5)-centered radical.
even though the O-centered N-oxypyridine-2(1H)-thione
and S-centered N-oxy-2-pyridylthiyl radicals absorb in
similar spectral regions, the lifetimes of the latter (25
Although thiyl radicals are known to undergo inef-
26
ficient, reversible addition to olefins, they add readily
to the carbon-sulfur double bonds of thiones, with large
28
µs) are 2 orders of magnitude longer than the former.
28-31
rate constants approaching the diffusion limit.
The
Mechanism of Photoreaction. The suggested mech-
anism of the photoreaction of 3, which takes into account
the results of steady-state and laser flash photolysis
studies of 3, is presented in Scheme 4. The initially
formed radical 7 is short-lived (estimated lifetime τ < 0.1
µs). It subsequently transforms into the more stable, long-
rapid reaction between thiyl radical 13 and ground-state
molecule 3 (route a, Scheme 4), combined with the long
lifetime of 13, is consistent with the proposed mechanism.
These two kinetic features account for the observation
that the photoreaction quantum yields (Table 1) remain
unchanged as the substrate concentration was varied
between 0.1 and 1 mM. Another way of interpreting this
is that almost all of the radicals that escape geminate
(
25) von Sonntag, C. In Sulfur-Centered Reactive Intermediates in
Chemistry and Biology; Chatgilialoglu, C., Asmus, K.-D., Eds.; Ple-
num: New York, 1990; pp 359-366.
(
(
26) Alam, M. M.; Ito, O. J. Org. Chem. 1999, 64, 1285-1290.
27) Osamu, I.; Matsuda, M. J. Am. Chem. Soc. 1979, 101, 1815-
(29) Bruhlmann, U.; Huber, J. R. J. Photochem. 1979, 10, 205-213.
(30) Boivin, J.; Crepon, E.; Zard, S. Z. Tetrahedron Lett. 1990, 31,
1
819.
6869-6872.
(
28) Aveline, B. M.; Kochevar, I. E.; Redmond, R. W. J. Am. Chem.
(31) Scaiano, J. C.; Ingold, K. U. J. Am. Chem. Soc. 1976, 98, 4727-
Soc. 1996, 118, 10113-10123.
4732.
986 J. Org. Chem., Vol. 70, No. 3, 2005

Photolysis of 5-Iodo-4-thiouridine
recombination react with the substrate. This is why the
quantum yield of degradation of the substrate matches
the yield of photoproducts. The radical adduct 15, the
product of this bimolecular reaction, could be a precursor
of the stable photoproducts identified in steady-state
photochemistry. Oxidation of radical 15 by iodine gives
iodide and disulfide 6 (route b, Scheme 4). Decomposition
of the radical 15 gives 4 and radical 17 (route c, Scheme
169.5, 169.4, 152.5, 142.2, 100.5, 89.5, 80.1, 73.8, 69.6, 62.6,
-
1
-1
2
0.8, 20.4; UV λmax nm (ꢀmax
13 400); HR-LSIMS calcd for C30
71.14868, obsd 771.14879.
M
cm ) 309 (18 200), 264
+
(
7
H
35
N
4
O
16
S
2
(M + H )
2
′,3′,5′-Tri-O-acetyl-5-phenyl-4-thiouridine (10). Com-
pound 10 was obtained as a yellow oil (65.6 mg, 33% yield) by
treatment of 2′,3′,5′-tri-O-acetyl-5-phenyluridine (192 mg, 0.43
18,34 1
mmol) with P
2
S
5
according to the literature procedure:
H
NMR δ 9.81 (s, 1H), 7.48 (s, 1H), 7.44-7.43 (m, 5H), 6.15 (d,
J ) 5.4 Hz, 1H), 5.45 (t, J ) 5.4 Hz, 1H), 5.36 (t, J ) 5.4 Hz,
4
). The latter species finally gives symmetrical sulfide
. Iodine formed in the primary photochemical step in
1
H), 4.44-4.42 (m, 1H), 4.36-4.34 (m, 2H), 2.17 (s, 3H), 2.16
5
13
(
s, 3H), 1.75 (s, 3H); C NMR δ 188.7, 170.0, 169.6, 147.1,
benzene solution recombines to give I
2
(route d, Scheme
). In an aqueous solution, the iodine atoms may undergo
a reaction with the solvent to give iodide and other
1
6
34.6, 132.5, 129.8, 128.6, 128.3, 125.4, 87.4, 80.2, 72.8, 70.1,
4
-
1
-1
2.9, 20.5, 20.4, 20.1; UV λmax nm (ꢀmax
M
cm ) 341 nm, 260
+
nm; HR-LSIMS calcd for C21
H
23
N
2
O
8
S (M + H ) 463.11751,
1
3,32
species.
Moreover, in aqueous solutions of 3, dark
obsd 463.11936.
iodine chemistry may also be responsible for the oxidation
of the thione 7 to the disulfide 5 (route e, Scheme 4). It
has been established that 4-thiouracil and its nucleosides
are instantaneously and quantitatively converted into
disulfides upon treatment with iodine/iodide in aqueous
Analytical Scale Irradiation and Quantum Yield De-
-
3
termination. Solutions (2.4 mL) of 3 (0.1-1 × 10 M) in 1:1
3 2
(v/v) CH CN-H O pH 5.8 (0.05 M phosphate buffer) or
benzene were placed in a 1 × 1 cm UV cell, deoxygenated by
bubbling argon, and irradiated on an optical bench using a
2
00 W high-pressure mercury lamp equipped with a glass (λ
33
solution. In a control experiment, we demonstrated that
addition of an equimolar amount of I to CH CN-H
g 334 nm) or an interference filter to isolate the λ ) 366 or
13 nm lines. The progress of the reaction was monitored by
2
3
2
O
3
solution of 3 ([3] ) 0.5 mM) leads instantaneously to the
disulfide 5 as the sole product. No dark oxidation of 3 to
disulfide was observed in benzene. The dark oxidation
may be responsible for the ca. 2 times greater quantum
yields for the production of disulfides upon irradiation
of 3 in aqueous solution as compared to that in benzene
solution (Table 1).
UV spectroscopy and HPLC. For the quantum-yield determi-
nation, the samples were irradiated to <15% conversion of
substrate, and the concentrations of substrate 3 and photo-
products 4-6 were determined by HPLC analysis. Benzophe-
3
6
none-benzhydrol actinometry was used.
Preparative Irradiation for Product Analysis. A solu-
tion of 3 (0.2 mmol) in 400 mL of the above solvents was
irradiated, in portions, in an 80 mL photoreactor with a 150
W high-pressure mercury lamp through a Pyrex filter under
an argon atmosphere. Irradiation was continued to ∼40%
conversion of the substrate as checked by HPLC. Irradiated
solutions were collected and concentrated under reduced
pressure. The photoproducts were isolated from the residue
by preparative HPLC and identified by a comparison of their
HPLC retention times (coinjections) and spectral data (UV,
HR-LSIMS) with those of synthesized samples.
Experimental Section
Preparation of Disulfides 5, 6, and 14. The disulfides
were prepared by iodine oxidation of the corresponding thiones.
3
3
A standard published procedure was accordingly modified.
See Supporting Information.
5
-Iodo-4-{[1-(2′,3′,5′-tri-O-acetyl-â-D-ribofuranosyl)-5-
iodo-2-oxo-1,2-dihydropyrimidin-4-yl]dithio}-1-(2′,3′,5′-
tri-O-acetyl-â-D-ribofuranosyl)-pyrimidin-2(1H)-one (5).
From 3 (51.2 mg, 0.1 mmol) was obtained disulfide 5 as a
Isolation and Identifications of Photoproducts 11 and
1
2. A solution of 3 (0.08 mmol) and bis-(N-R-acetyl)cystine-
bis-N-ethylamide (0.08 mmol) in 160 mL of 1:1 (v/v) CH
3
CN-
1
colorless oil (36.8 mg, 72% yield): H NMR δ 8.08 (s, 2H), 6.03
H
2
O pH 5.8 was irradiated as described above. The photo-
(
d, J ) 3.6 Hz, 2H), 5.38 (m, 2H), 5.31 (m, 2H), 4.43-4.33 (m,
products 11 and 12 were isolated from the concentrated
irradiated solution by preparative HPLC and identified on the
basis of UV and HR-LSIMS spectral data. The samples of
disulfides were treated with 1,4-dimercapto-2,3-butanediol
6
1
6
2
H), 2.23 (s, 6H), 2.11 (s, 6H), 2.08 (s, 6H); 13C NMR δ 175.0,
70.1, 169.3, 169.1, 151.7, 146.3, 121.3, 89.2, 80.1, 73.9, 69.3,
-
1
-1
3.5, 62.4, 21.1, 20.4; UV λmax nm (ꢀmax
M
cm ) 229 (39 000),
33 2 4 16 2
65 (14 700), 338 (16 800); HR-LSIMS calcd for C30H I N O S
(
DTT). Thus, to 11 or 12 (1 mL, c ≈ 0.2 mM) in 50% aq CH
3
-
+
(
M + H ) 1022.94198, obsd 1022.94087.
-(2′,3′,5′-Tri-O-acetyl-â-D-ribofuranosyl)-5-iodo-4-{[1-
CN was added a solution of DTT (50 µL, c ) 11 mM) in H
2
O.
1
The reaction mixture was allowed to stand for 15 min at room
temperature and then analyzed by HPLC using a gradient of
2
(
2′,3′,5′-tri-O-acetyl-â-D-ribofuranosyl)-2-oxo-1,2-dihydro-
pyrimidin-4-yl]dithio}pyrimidin-2(1H)-one (6). From a
mixture of 3 (25.6 mg, 0.05 mmol) and 4 (19.3 mg, 0.05 mmol)
3 3
7% aq CH CN (initial) f 80% aq CH CN (30 min) as an
eluting phase. The products were identified by a comparison
of UV spectra and retention times (coinjection) with that of
authentic samples of N-R-acetyl-L-cysteine N-ethylamide and
1
was obtained a colorless solid of disulfide 6 (10.8 mg): H NMR
δ 8.07 (s, 1H), 7.79 (d, J ) 7.14 Hz, 1H), 6.51 (d, J ) 7.14 Hz,
1
H), 6.08 (d, J ) 4.40 Hz, 1H), 6.04 (d, J ) 3.84 Hz, 1H), 5.39-
.35 (m, 2H), 5.32-5.27 (m, 2H), 4.46-4.34 (m, 6H), 2.24 (s,
H), 2.15 (s, 3H), 2.12 (s, 6H), 2.09 (s, 6H); UV λmax nm (ꢀmax
4
or 3 in the case of 11 and 12, respectively.
-(Acetylamino)-N-ethyl-3{[1-(2′,3′,5′-tri-O-acetyl-â-D-
ribofuranosyl)-2-oxo-1,2-dihydropyrimidin-4-yl]dithio}-
5
2
3
-
1
-1
M
cm ) 220 (26 800), 268 (15 400), 315 (14 400); HR-LSIMS
-
1
-1
propanamide (11): UV λmax (ꢀmax
M
cm ) 270 nm (8600),
(M
+
calcd for C30
H34IN
4
O
16
S
2
(M + H ) 897.04533, obsd 897.04389.
3
06 nm (12 000); HR-LSIMS m/z calcd for C22
31 4 10 2
H N O S
4
-{[1-(2′,3′,5′-Tri-O-acetyl-â-D-ribofuranosyl)-2-oxo-1,2-
+
+
H ) 575.14835, obsd 575.14827.
-(Acetylamino)-N-ethyl-3{[1-(2′,3′,5′-tri-O-acetyl-â-D-
ribofuranosyl)-5-iodo-2-oxo-1,2-dihydropyrimidin-4-yl]-
dihydropyrimidin-4-yl]dithio}-1-(2′,3′,5′-tri-O-acetyl-â-D-
ribofuranosyl)-pyrimidin-2(1H)-one (14). From 4 (38.4 mg,
0
2
.1 mmol) was obtained disulfide 14 (22.7 mg, 59% yield) as a
-
1
-1
dithio}-propanamide (12): UV λmax (ꢀmax
M
cm ) 270 nm
1
colorless oil. H NMR δ 7.88 (d, J ) 7.2 Hz, 2H), 6.70 (d, J )
.2 Hz, 2H), 6.06 (d, J ) 3.9 Hz, 2H), 5.40 (dd, J ) 5.6 Hz, J
3.9 Hz, 2H), 5.29 (t, J ) 5.6 Hz, 2H), 4.43-4.37 (m, 6H),
7
)
(34) Fox, J.; Van Praag, D.; Wempen, I.; Doerr, I. L.; Cheong, L.;
Knoll, J. E.; Eidinoff, M. L.; Bendich, A.; Brown, G. B. J. Am. Chem.
Soc. 1959, 81, 178-187.
13
2
.14 (s, 6H), 2.12 (s, 6H), 2.09 (s, 6H); C NMR δ 177.0, 170.1,
(
35) Dietz, T. M.; Koch, T. H. Photochem. Photobiol. 1989, 49, 121-
(
32) Rahn, R. O. Anal. Chim. Acta 1991, 248, 595-602.
129.
(36) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photo-
chemistry; 2nd ed.; Dekker: New York., 1993.
(
33) Coleman, R. S.; McCary, J. L.; Perez, R. J. Tetrahedron 1999,
5
5, 12009-12022.
J. Org. Chem, Vol. 70, No. 3, 2005 987

Wenska et al.
(
12 000), 332 nm (14 000); HR-LSIMS calcd for C22
H
30IN
4
O
10
S
2
by self-consistent reaction field theory using a polarized cavity
+
45
46
(
M + H ) 701.04500, obsd 701.04432.
model with COSMO boundary conditions. Thermochemical
47
Irradiation in Deuterated Solvents. Solution of 3 (36 mg,
quantities were derived using standard conversion formulas.
0
3 2
.07 mmol) in CD CN-H O pH 5.8 (70 mL) was deoxygenated
Nanosecond Laser Flash Photolysis. The nanosecond
and irradiated in a photoreactor as described above. Irradia-
tion was followed by HPLC and continued to 50% conversion
of the substrate. After irradiation, the solution was concen-
trated to a small volume under reduced pressure. Photoproduct
laser flash photolysis setup and its data acquisition system
4
8
have been described in detail. The excimer laser (Lambda
Physik, Compex) provided 20 ns, 10 mJ pulses at 308 nm. The
transients were monitored with a pulsed 1 kW xenon lamp,
having the monitoring beam perpendicular to the laser beam.
All experiments were carried out within a rectangular quartz
optical cell (0.5 × 1 cm). The monitoring light path length, l,
was 0.5 cm.
4
was isolated by preparative HPLC. The same procedure was
3 2
used for irradiation of 3 in CH CN-D O pH 5.8. Samples of
photoproduct 4 obtained in the two experiments were subjected
to ESI-MS (negative ions) analysis. The percent of deuterium
3
7
incorporation was calculated from the measured intensities
of peaks at m/z 385, 386, 387, and 388 in the spectra of the
two samples of photoproduct and authentic, unlabeled sample
of 4. Photoproduct isolated from irradiated solution of 3 in CH -
3
CN-D
Acknowledgment. This work was supported in part
by the Polish Committee for Scientific Research (KBN)
Grant No. 3T09A 02615 and in part by 5th Frame EC
Program (Research Training Network HPRN CT-2002-
1
2
O was also analyzed by H NMR. In the low-field region
1
(
6-10 ppm), the H NMR spectrum in CDCl shows two
3
0
0184 SULFRAD). The Notre Dame Radiation Labora-
intense singlets at δ 7.2 ppm due to C(6)-H and 7.3 due to
solvent resonances. Only the trace signal at 6.4 ppm attribut-
able to C(5)-H was observed.
tory is supported by Basic Energy Sciences of DOE. This
is Document No. NDRL-4548 from the Notre Dame
Radiation Laboratory.
Theoretical Calculation. Density functional theory (DFT)
calculations were performed with the Gaussian98 electronic
3
8
structure software package employing the popular B3LYP
Supporting Information Available: General experimen-
tal methods, experimental procedure for synthesis of disulfides,
pH-dependent spectra of 3 and spectrophotometric titration
39
40
functional. This hybrid functional comprises both local and
4
1
42,43
nonlocal (gradient-corrected) exchange and correlation
contributions mixed with a piece of exact (Hartree-Fock)
exchange. The mixing parameters were derived from fits to
known thermochemistry of a well-characterized set of small
molecules. For open-shell systems (the transient radicals and
radical cations), spin-unrestricted DFT (UB3LYP) was em-
ployed. For structural studies, the (heavy-atom) polarized split-
1
13
curve, H NMR spectra of 5, 6, 10, and 14, C NMR spectra
of 5, 10, and 14, and total energy calculations for 1-methyl-
4
-thiouracil-5-yl radical, its thiol tautomer, and 1-methyluracil-
-thiyl radical. This material is available free of charge via
the Internet at http://pubs.acs.org.
4
44
valence 6-31G* basis set was used. DFT geometries are often
converged with modest basis sets. Solvent effects were modeled
JO048496G
(
39) Becke, D. A. J. Chem. Phys. 1993, 98, 5648-5652.
(
37) Biemann, K. Mass Spectrometry. Organic Chemical Applica-
tions; McGraw Hill Book Company, Inc.: New York, 1962.
38) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
(40) Slater, J. C. The Self-Consistent Field for Molecules and Solids;
McGraw-Hill: New York, 1974.
(
(41) Becke, A. D. ACS Symposium Series 1989, 394, 165-179.
(42) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200-
1211.
(43) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789.
(44) Hariharan, P. C.; Pople, J. A. Chem. Phys. Lett. 1972, 16, 217-
219.
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 98; Gaussian, Inc.: Pittsburgh, PA, 1998.
(45) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995-2001.
(46) Klamt, A.; Sch u¨ u¨ rmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
799-805.
(47) McQuarrie, D. A. Statistical Thermodynamics; Harper & Row:
New York, 1973.
(48) Thomas, M. D.; Hug, G. L. Comput. Chem. (Oxford) 1998, 22,
491-498.
988 J. Org. Chem., Vol. 70, No. 3, 2005