Direct measurement of pyrimidine C6-hydrate stability

Article abstract of DOI:10.1016/S0968-0896(01)00137-7

Pyrimidine C6-hydrates are produced via UV-irradiation and undergo dehydration upon standing. The stability of these compounds has a direct bearing on their genotoxicity. The rate constants for elimination from 5' -benzyoylated derivatives of 5,6-dihydro- 5-hydroxythymidine (6) and 5,6-dihydro-5-hydroxy-2' -deoxyuridine (9) were measured directly via HPLC. The rate constants for dehydration increase from pH 6.0 to 8.0. The half-lives for 6 and 9 at pH 7.4 and 37°C are 46.5 and 24.4 h, respectively. Deglycosylation is not observed, even upon heating at 90°C. These observations reinforce proposals that pyrimidine hydrates are sufficiently long-lived that they can exert significant effects on biological systems. Copyright

Full text of DOI:10.1016/S0968-0896(01)00137-7

Bioorganic & Medicinal Chemistry 9 (2001) 2341–2346
Direct Measurement of Pyrimidine C6-hydrate Stability
K. Nolan Carter and Marc M. Greenberg*
Department of Chemistry, Colorado State University, Fort Collins, CO 80523, USA
Received 1 February 2001; accepted 16 April 2001
Dedicated to Professor Peter B. Dervan.
Abstract—Pyrimidine C6-hydrates are produced via UV-irradiation and undergo dehydration upon standing. The stability of these
compounds has a direct bearing on their genotoxicity. The rate constants for elimination from 50-benzyoylated derivatives of 5,6-
dihydro-5-hydroxythymidine (6) and 5,6-dihydro-5-hydroxy-20-deoxyuridine (9) were measured directly via HPLC. The rate con-
stants for dehydration increase from pH 6.0 to 8.0. The half-lives for 6 and 9 at pH 7.4 and 37 ꢀC are 46.5 and 24.4 h, respectively.
Deglycosylation is not observed, even upon heating at 90 ꢀC. These observations reinforce proposals that pyrimidine hydrates are
sufficiently long-lived that they can exert significant effects on biological systems. # 2001 Elsevier Science Ltd. All rights reserved.
Pyrimidine hydrates are produced by a variety of
reaction pathways that are populated when nucleic
acids are exposed to ionizing radiation.1 Thymidine C5-
hydrate (1) is produced by hydroxyl radical addition
under O2 deficient conditions.1,2 The C6-hydrates of
deoxycytidine (2) and thymidine (3) are the major non-
dimeric lesions produced upon UV-irradiation.3,4 5,6-
Dihydro-6-hydroxy-20-deoxyuridine (4) is produced in
DNA from hydrolysis of the C6-hydrate of dC (2).5À7
The C6-hydrates may also be produced upon hydration
of nucleobase cation radicals.8 The biological relevance
of pyrimidine hydrates is underscored by their excision
by base excision repair enzymes, such as endonuclease
III.5À7,9À12 Furthermore, thymidine C5-hydrate (1) is a
potent inhibitor of DNA polymerase I (Klenow exoÀ
fragment) and 2 is a premutagenic lesion in vitro.13 The
pyrimidine C6-hydrates undergo dehydration, and their
chemical instability potentiates their ability to influence
the fidelity of replication and transcription. We wish to
report on our studies concerning the direct measurement
of dehydration in 50-protected pyrimidine C6-hydrates.
necessitated that we understand the lability of 3 and 4.
Direct measurement of dehydration rates of thymine
and uracil C6-hydrates have been reported.3 However,
we were uncertain if one could safely extrapolate from
studies on free bases where N-1 is not alkylated to
nucleosides, and in turn, DNA. Dehydration rates of
thymidine and deoxyuridine C6-hydrates have been
measured in DNA using a combined enzymatic and
HPLC assay.6,7,11 In this assay, the amount of modified
free base that is released by endonuclease III is mea-
sured by HPLC as a function of time. However, this
approach would obscure any deglycosylation, which
may compete with dehydration. Consequently, we set
out to determine if deglycosylation competes with
dehydration in nucleoside C6-hydrates, and to measure
rates of dehydration directly in 3 and 4.
Our interest in the chemical stability of pyrimidine C6-
hydrates was cultivated by our studies on 5,6-dihy-
dropyrimidin-6-yl radicals (e.g., 5) from which pyrimdine
C6-hydrates are formed under aerobic conditions.
Investigations of radical mediated DNA damage
*Corresponding author. Tel.:+1-970-491-7972; fax; +1-970-491-
1801; e-mail: mgreenbe@lamar.colostate.edu
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00137-7

2342
K. Nolan Carter, M. M. Greenberg / Bioorg. Med. Chem. 9 (2001) 2341–2346
a source of Br2 and CaCO3 to buffer the reactions
(Scheme 1). Following separation of the diastereomeric
trans-bromohydrins, the respective benzoylated C6-
hydrates were obtained via Zn0/AcOH reduction. Iso-
lated yields of 5,6-dihydro-6-hydroxy-20-deoxyuridines
(8, 9) were compromised due to facile elimination under
the reaction and purification conditions. Careful chro-
matography provided the 6S-diastereomer (8) without
any contamination of 50-benzoyl-20-deoxyuridine (11);
whereas samples of 9 were contaminated by ꢁ5% of 11.
Results and Discussion
Chemical stability studies of single diastereomers of 3
and 4 were carried out using 50-benzoyl derivatives. The
50-benzoyl group was incorporated in order to enhance
detection by UV-absorption, to facilitate purification,
and to remove any possible role that a free 50-hydroxyl
group might play in the chemical stability of the C6-
hydrates. There is precedent for involvement of the pri-
mary hydroxyl group in the rearrangement of for-
mamidopyrimidine deoxynucleosides at pH 7 and
dihydropyrimidines under acidic conditions.14,15 Such
reactions diminish the relevance of monomer studies to
DNA where the presence of the 50-phosphate diester
group prohibits rearrangements of this type.
Stereochemistry in the dihydropyrimidine rings of the
bromohydrins of both pyrimidines and C6-hydrates of
1
thymidine were determined by H NMR spectroscopy
using analogous compounds as a frame of reference.
1
Examination of the H NMR spectra of the bromohy-
drins and C6-hydrates of deoxyuridine, as well as the
6-hydroperoxy-5-hydroxydihydrothymidines and 5-
hydroperoxy-6-hydroxydihydrothymidines consistently
indicated a greater difference in the chemical shifts of
0
the diastereotopic H2 -protons in the 6S-diastereomers
0
than the 6R-diastereomers.17,18 The pro-S H2 -proton is
consistently shifted further downfield relative to the
respective pro-R proton in the 6S-diastereomers than in
the 6R-diastereomers. Stereochemical assignments of
the modified thymidines mentioned above were corro-
borated by their reduction to the thymidine glycols, for
which single crystal X-ray diffraction data is available.19
This chemical shift pattern was identified in diaster-
eomeric pairs of bromohydrins (12–15) and 50-benzoyl
C6-hydrates (6–9) (Table 1). NOE experiments were
carried out on 6, 7, and 12–15 in order to provide addi-
tional evidence for these stereochemical assignments.
The results of these experiments were consistent with
the interpretation of the chemical shift pattern. When
comparing pairs of diastereomers (e.g., 6 and 7) irradiation
0
of H6 produced a larger NOE at the pro-S H2 -proton in
the compound believed to contain the R-configuration at
0
the C6-position (based upon chemical shifts of H2 -pro-
0
tons) and a smaller NOE at the H5 -protons. These quali-
tative trends were also observed in the other two pairs of
diastereomers that were examined (Table 1).
Single diastereomers of benzoylated C6-hydrates of
thymidine (6 and 7) and 20-deoxyuridine (8 and 9) were
synthesized via modification of procedures reported for
the free nucleosides.16 High yields of the bromohydrin
precursors were obtained using N-bromosuccinimide as
Scheme 1. (a) NBS, CaCO3, THF/H2O; (b) Zn/HOAc, THF/H2O.

K. Nolan Carter, M. M. Greenberg / Bioorg. Med. Chem. 9 (2001) 2341–2346
2343
Epimerization of 5,6-dihydro-6-hydroxy-20-deoxyuridine
glycosidic bond of the modified nucleotide, followed by
HPLC quantitation of the amount of damaged nucleo-
bases released. In some instances these experiments were
carried out at high temperatures (up to 80 ꢀC) in order
to decrease the time span of the investigation.5À7,11,12
We considered the possibility that the absolute rate
constants determined in these experiments may be
affected by deglycosylation. If the rate of dehydration
differs in DNA from that of the free base, deglycosyla-
tion would result in the measurement of a rate constant
for dehydration that is a weighted-average of the two
processes. In order to check for deglycosylation from
the 50-benzoylated nucleosides, 5-benzoyl-2-deoxyribose
(16) was independently prepared as a mixture of a,b-
anomers via reduction of the respective 2-deoxyr-
ibonolactone. The lactol (16) was analyzed for in sam-
ples of 6 and 8 that were heated at 90 ꢀC for 1 h in
phosphate buffer (20 mM, pH 7.4). Thermolysis of 6
and 8 produces the dehydrated derivatized nucleosides
as the sole observed products within 1 h, in 71 and 72%
yield, respectively (data not shown). Spiking samples
with 16 prior to thermolysis indicates that as little as 4%
of the deglycosylation product could have been
observed. A lower detection limit is not possible as 16
undergoes significant decomposition under these condi-
tions. Furthermore, based upon the mass balances
observed, we cannot exclude the formation of other
deoxyribose-derived products.
At 37 ꢀC, in the absence of a general base catalyst,
hydrates 8 and 9 underwent epimerization with negli-
gible dehydration to form an equilibrium mixture
(9:8=1.5) of diastereomers. The approach to equili-
1
brium was determined by H NMR using the C6-pro-
tons to measure the concentrations of each epimer. The
mixture of diastereomers was independent of which side
of the equilibrium one approaches from, and the abso-
lute rate constants were within 25% of each other
(Table 2). Thymidine C6-hydrates (6 and 7) also epi-
merized much more rapidly than they underwent dehy-
dration in the presence or absence of catalyst.
Epimerization at C6 was accompanied by isomerization
at the C5-position, resulting in four diastereomers from
one original isomer. Separation of these isomers by
HPLC and resolution of signals in 1H NMR were
insufficient to allow analysis of the kinetics of these
processes.
Deglycosylation and dehydration in pyrimidine C6-
hydrates.
Previous measurements of dehydration in pyrimidine
C6-hydrates utilized endonuclease III to hydrolyze the
Table 1. NMR determination of stereochemistry in 50-benzoyl dihy-
dropyrimidinesa
Dehydration of 50-benzoylated thymidine C6-hydrate
(6) was followed by HPLC as a function of pH, water
content, and buffer concentration (Table 3, Fig. 1).
Epimerization was faster than dehydration, but it was
not possible to determine rate constants for elimination
from individual stereoisomers. The effects of pH are
consistent with previous studies on the stability of pyr-
imidine hydrates. Maximal stability was observed at pH
6.0, where essentially no decomposition was observed
Compd
NOE
0
0
0
C6-Configuration ÁH2 -Chemical H6!pro-S H2 H6!H5
Shiftb
Table 3. Rate constants for dehydration of 50-benzoylated pyr-
imidine C6-hydrates at 37 ꢀC
6
7
12
13
14
15
S
R
S
R
S
0.34
0.10
0.33
0.26
0.26
0.11
2.76
3.18
2.28
3.55
2.27
2.86
0.83
0.59
0.60
0.87c
0.96
0.63
R
aFirst proton listed is irradiated and second is where effect is mea-
sured.
0
bÁH2 =dpro-S H2 - dpro-R H2
0
0
.
0
cThe enhancement could not be distinguished from that at H3 due to
poor chemical shift dispersion.
Compound pH [KH2PO4] % CH3CNa kDehy ꢂ106
t1/2
(h)
(mM)
(sÀ1 a
)
Table 2. The rate of epimerization of 50-O-benzoyl-5,6-dihydro-6-
hydroxy-20-deoxyuridines (8 and 9)
6
6
6
6
6
6
6
9
9
9
6.0
7.4
8.0
7.4
6.0
7.4
8.0
6.0
7.4
8.0
20
20
20
40
20
20
20
20
20
20
5
5
5
<0.4
4.1
11.3
4.7
<0.2
2.9
5.9
>785
46.5
17.0
40.6
5
50
50
50
5
5
5
>1046
66.3
32.6
Starting
diastereomer
k1ꢂ106 (sÀ1 a
)
k
À1ꢂ106 (sÀ1
)
ꢁ0.5
ꢁ385
7.9
26.8
24.4
7.2
8
9
11.2
8.3
7.6
5.7
aRate constants are the composite of four separate experiments.
aRate constants are an average of two separate experiments.

2344
K. Nolan Carter, M. M. Greenberg / Bioorg. Med. Chem. 9 (2001) 2341–2346
over the course of 5 days. The rate constant for dehy-
dration increased significantly as the pH was increased
to 8.0. The dehydration reaction also showed a depen-
dence on buffer concentration, which is consistent
with general base catalysis. Finally, the decrease in
half-life for 6 as the solvent polarity increased from
50% H2O to 95% H2O is also consistent with the
anticipated negative charge buildup in the transition
state.
triethylamine were distilled from CaH2. Acetonitrile was
passed through CuSO4 and then distilled from CaH2. N-
Bromosuccinimide (NBS) was recrystallized from H2O.
5,6-Dihydro-5-bromo-6-hydroxy-20-deoxyuridine (14 and
15). CaCO3 (407 mg, 4.06 mmol) and NBS (578 mg,
3.25 mmol) were added to solution of 50-benzoyl-20-
deoxyuridine (11, 900 mg, 2.71 mmol) in 3:1 THF/H2O
(21 mL) at 0 ꢀC. The solution was allowed to warm to
ambient temperature and stirred for 6 h. The reaction
mixture was filtered through Celite and concentrated.
The crude mixture was subjected to silica gel flash
chromatography (2–10% MeOH/CHCl3) to yield 15
(489 mg, 42%) and 14 (407 mg, 35%) as white foams.
15: 1H NMR (MeOH-d4) d 8.07–8.03 (m, 2H), 7.67–7.61
(m, 1H), 7.53–7.48 (m, 2H), 6.25 (t, J=6.6 Hz, 1H), 5.31
(d, J=2.4 Hz, 1H), 4.53 (d, J=4.2 Hz, 2H), 4.46–4.42
(m, 1H), 4.21 (d, J=2.4 Hz, 1H), 4.14 (dd, J=8.1,
4.2 Hz), 2.31–2.27 (m, 1H), 2.20–2.16 (m, 1H); 13C
NMR (MeOH-d4) d 168.1, 167.9, 152.4, 134.6, 131.2,
130.7, 129.9, 86.1, 85.4, 77.9, 72.5, 66.0, 41.8, 40.2; IR
(film) 3431, 1700, 1465, 1273, 1087 cmÀ1; HR-MS
Similar behavior was observed for the C6-hydrate of 20-
deoxyuridine (9). As predicted based upon previous
experiments in biopolymers, dehydration from 9 was
significantly faster than from the thymidine analogue
(Table 3). The higher rate of dehydration observed in
the 20-deoxyuridine analogue compared to 6 is con-
sistent with the anticipated higher acidity of the C5-
proton in these compounds. The half-life for elimination
at 37 ꢀC was 24.4 and 7.2 h at pH 7.4 and 8.0, respec-
tively. This is very close to that measured for 5,6-dihy-
dro-6-hydroxy-20-deoxyuridine in DNA when it is base-
paired to 20-deoxyadenosine.6
1
(FAB) calcd (M+H) 431.0277, found 431.0285. 14: H
NMR (MeOH-d4) d 8.07–8.05 (m, 2H), 7.65–7.60 (m,
1H), 7.52–7.47 (m, 2H), 6.24 (t, J=6.9 Hz, 1H), 5.28 (d,
J=2.1 Hz, 1H) 4.54–4.41 (m, 3H), 4.17 (d, J=2.1 Hz,
1H), 4.14–4.10 (m, 1H), 2.47–2.38 (m, 1H), 2.19–2.14
(m, 1H); 13C NMR (MeOH-d4) d 168.2, 167.9, 152.3,
134.6, 131.3, 130.8, 129.8, 85.6, 84.7, 77.6, 72.4, 65.8,
41.4, 38.9; IR (film) 3396, 1698, 1452, 1276, 1090
cmÀ1; HR-MS (FAB) calcd (M+H) 431.0277, found
431.0279.
Conclusion
These studies support previous investigations regarding
the stability of pyrimidine C6-hydrates in DNA, indi-
cating that the repair of such lesions is important in
order to avoid their potential deleterious biological
effects. In addition, the facile and quantitative dehydra-
tion of 6–9 at higher temperatures indicates that this
reaction will be a useful tool for manipulating tandem
nucleic acids containing these lesions.
5,6-Dihydro-5-bromo-6-hydroxythymidine (12 and 13).
To a solution of 50-benzoyl thymidine (10, 1.00 g,
2.89 mmol) in 3:1 THF/H2O (28 mL) at 0 ꢀC was added
CaCO3 (434 mg, 4.33 mmol) and NBS (616 mg,
3.46 mmol) as described above for the preparation of 14
and 15. The reaction mixture was stirred for 0.5 h at
0 ꢀC and then allowed to warm to ambient temperature
where it was stirred for an additional 2 h. The reaction
mixture was filtered through Celite and concentrated.
The crude reaction mixture was purified via silica gel
flash chromatography (2–10% MeOH/CHCl3) to afford
12 (701 mg, 55%) and 13 (247 mg, 19%) as a partially
separable mixture of diastereomers. 12: 1H NMR
(MeOH-d4) d 8.09–8.07 (m, 2H), 7.65–7.61 (m, 1H),
7.52–7.48 (m, 2H), 6.25 (t, J=5.5 Hz, 1H), 5.15 (s, 1H),
4.56 (dd, J=9.0, 3.0 Hz, 1H), 4.13–4.10 (m, 1H), 2.48–
2.42 (m, 1H), 2.19–2.14 (m, 1H), 1.69 (s, 3H); 13C NMR
(DMSO-d6) d 167.7, 165.7, 150.7, 133.5, 129.6, 129.2,
128.8, 84.0, 83.1, 78.9, 70.6, 65.3, 55.3, 38.1, 22.5; IR
(film) 3446, 1701, 1466, 1273, 1070 cmÀ1; HR-MS
Experimental
All reactions were carried out in oven-dried glassware
under an atmosphere of argon or nitrogen unless
otherwise noted. THF was freshly distilled from Na/
benzophenone ketyl. Dichloromethane, DMF, and
1
(FAB) calcd 445.0433 (M+H), found 445.0447. 13: H
NMR (MeOH-d4) d 8.09–8.07 (m, 2H), 7.67–7.63 (m,
1H), 7.54–7.50 (m, 2H), 6.29 (t, J=5.2 Hz, 1H), 5.17 (s,
1H), 4.62 (dd, J=9.0, 2.4 Hz, 1H), 4.51–4.47 (m, 2H),
4.18–4.15 (m, 1H), 2.45–2.39 (m, 1H), 2.22–2.16 (m,
1H), 1.64 (s, 3H); 13C NMR (DMSO-d6) d 167.8, 165.6,
150.4, 133.5, 129.6, 129.4, 128.8, 83.2, 82.5, 78.8, 70.2,
64.3, 54.3, 37.1, 22.5; IR (film) 3396, 1700, 1452, 1275,
1087 cmÀ1, HR-MS (FAB) calcd 445.0433 (M+H),
found 445.0438.
Figure 1. Disappearance of 6 as a function of time at 37 ꢀC.

K. Nolan Carter, M. M. Greenberg / Bioorg. Med. Chem. 9 (2001) 2341–2346
2345
1
5R,6S-Dihydro-6-hydroxy thymidine (6). To a solution
of 5,6-dihydro-5-bromo-6-hydroxy thymidine 12
(100 mg, 0.226 mmol) in 3:1 THF/H2O at 0 ꢀC was
added zinc dust (46 mg, 1.31 mmol) in acetic acid
(24 mg, 0.41 mmol). The reaction was stirred at 0 ꢀC for
20 min, filtered through Celite and concentrated at
ambient temperature. The crude reaction mixture was
purified via silica gel flash chromatography (2%
MeOH/CHCl3) to afford 6 (23 mg, 28%) as a white
residue; 1H NMR (MeOH-d4) d 8.08–8.05 (m, 2H),
7.66–7.61 (m, 1H), 7.53–7.48 (m, 2H), 6.27 (t,
J=6.3 Hz, 1H), 5.10 (d, J=3.0 Hz, 1H), 4.56–4.43 (m,
3H), 4.06 (dd, J=8.7, 3.9 Hz, 1H), 2.73–2.65 (m, 1H),
2.53–2.43 (m, 1H), 2.18–2.10 (m, 1H), 1.07 (d,
J=6.9 Hz, 3H); 13C NMR (MeOH-d4) d 173.8, 167.9,
154.0, 134.6, 131.2, 130.8, 129.9, 85.5, 84.5, 76.8, 72.5,
65.7, 43.1, 38.5, 10.5; IR (film) 3445, 3231, 2929, 2857,
1709, 1462, 1256, 1215, 1125, 1028, 835; HR-MS (FAB)
calcd (M+H) 365.1349, found 365.1352.
foam; H NMR (MeOH-d4) d 8.05–8.01 (m, 2H), 7.65–
7.59 (m, 1H), 7.51–7.47 (m, 2H), 6.17 (t, J=6.9 Hz, 1H),
3.52 (dd, J=4.1, 2.1 Hz, 1H), 4.53–4.41 (m, 3H), 4.10
(dd, J=9.0, 4.5 Hz), 2.82 (dd, J=17, 4.1 Hz, 1H), 2.53
(dd, J=17, 2.1 Hz, 1H) 2.35–2.28 (m, 1H), 2.21–2.16
(m, 1H); 13C NMR (MeOH-d4) d 171.4, 168.0, 153.9,
142.1, 134.6, 131.2, 130.7, 129.8, 86.9, 85.3, 74.5, 72.6,
65.9, 39.8; IR (film) 3446, 1700, 1472, 1275, 1069,
757 cmÀ1
.
5-Benzoyl-2-deoxyribonolactone (17). To a solution of 2-
deoxyribonolactone20 (230 mg, 1.74 mmol) and Et3N
(229 mg, 2.26 mmol) in DMF (15 mL) at À40 ꢀC was
added benzoyl cyanide (274 mg, 2.09 mmol). The reac-
tion was stirred overnight and allowed to warm to
ambient temperature. The reaction was quenched with
H2O (4 mL), diluted with EtOAc (50 mL), washed with
H2O (3ꢂ50 mL), brine (50 mL) dried over Na2SO4, and
concentrated. The crude product was purified via silica
gel flash chromatography (1:2–1:1, EtOAc/hexanes) to
afford 17 (127 mg, 31%) as a clear oil. 1H NMR
(CDCl3) d 7.99–7.96 (m, 2H), 7.62–7.57 (m, 1H), 7.48–
7.45 (m, 2H), 4.70–4.68 (m, 1H), 4.61–4.55 (m, 3H), 3.08
(s, J=3 Hz, 1H), 2.95 (dd, J=18, 6.9 Hz, 1H), 2.62 (dd,
J=18 Hz, 3.9 Hz, 1H), IR (film) 3461, 2950, 1601, 1451,
5S,6R-Dihydro-6-hydroxy thymidine (7). Using the same
general procedure for the preparation of 7, 13 (100 mg,
0.226 mmol) was treated with zinc dust (46 mg,
1.31 mmol) and acetic acid (24 mg, 0.41 mmol). Pur-
ification of the crude product by silica gel flash chro-
matography (2% MeOH/CHCl3) afforded 7 (16 mg,
20%) as a white solid; 1H NMR (MeOH-d4) d 8.07–8.04
(m, 2H), 7.66–7.61 (m, 1H), 7.52–7.48 (m, 2H), 6.19 (t,
J=6.6 Hz, 1H), 5.10 (d, J=3.3 Hz, 1H), 4.62–4.44 (m,
3H), 4.13–4.11 (m, 1H), 2.76–2.72 (m, 1H), 2.32–2.28
(m, 1H), 2.22–2.19 (m, 1H), 1.04 (d, J=7.2 Hz, 3H); 13C
NMR (MeOH-d4) 173.9, 168.0, 154.4, 134.7, 131.3,
130.8, 129.9, 86.7, 85.3, 78.0, 72.6, 65.9, 42.6, 40.2, 10.6;
IR (film) 3417, 2924, 1714, 1472, 1276, 1251, 711; HR-
MS (FAB) calcd (M+H) 365.1349, found 365.1347.
1781, 1720, 1176, 1164, 1026, 944 cmÀ1
.
50-Benzoyl-20-deoxyribose (16). A toluene solution of
DIBAL (2.0 mL, 1.0 M.) was added via syringe pump
overnight to a solution of 17 (496 mg, 2.09 mmol) in
CH2Cl2 (20 mL) maintained at À78 ꢀC. The reaction
was quenched while cold with MeOH (1 mL) and
allowed to warm to ambient temperature. The reaction
mixture was diluted with CH2Cl2 (80 mL) and washed
with saturated Rochelle salt solution (100 mL), brine
(100 mL), and concentrated in vacuo. The crude residue
was purified via silica gel flash chromatography (1:3–1:1
EtOAc/CH2Cl2) to afford both anomers of 16 (22 mg,
6S-5,6-Dihydro-6-hydroxy-20-deoxyuridine (8). Using the
same procedure for the reduction of 15 (see below), 5,6-
dihydro-5-bromo-6-hydroxy-20-deoxyuridine (14, 75 mg,
0.17 mmol) was treated with zinc dust (66 mg, 1.0 mmol)
in acetic acid (19 mg, 0.31 mmol). The crude product
was purified via silica gel flash chromatography (1–6%
MeOH/CHCl3) to afford 8 (42 mg, 63% overall) as a
1
4.4%) as a clear oil. H NMR (CDCl3) d 8.07–8.00 (m,
2H), 7.60–7.54 (m, 1H), 7.46–7.41 (m, 2H), 5.64 (t,
J=4.5 Hz, 1H), 4.58–4.54 (m, 1H), 4.49 (d, J=5.1 Hz,
0.6H), 4.37–4.32 (m, 2H), 4.20 (q, J=5.1 Hz, 0.4H),
3.71 (d, J=5.1 Hz, 0.7H), 3.48 (s, 0.3 Hz), 3.11 (d,
J=8.1 Hz, 0.7H), 2.47 (s, 0.3H), 2.98–2.14 (m, 2H); 13C
NMR (CDCl3) d 166.9, 166.6, 133.5, 129.9, 129.8, 128.7,
99.6, 98.9, 85.3, 73.5, 72.6, 65.8, 64.8, 42.5, 41.5; IR
(film) 3418, 2926, 1715, 1468, 1385, 1277, 1070,
1026 cmÀ1; ESI-MS, 237.1 (MÀH).
1
white foam; H NMR (MeOH-d4) d 8.06–8.04 (m, 2H),
7.65–7.60 (m, 1H), 7.52–7.47 (m, 2H), 6.23 (t,
J=6.0 Hz, 1H), 5.31 (dd, J=3.9, 2.1 Hz, 1H), 4.56–4.41
(m, 3H), 4.07 (dd, J=9.3, 4.2 Hz, 1H), 2.77 (dd,
J=16.8, 3.9 Hz, 1H), 2.54–2.44 (m, 2H), 2.18–2.10 (m,
1H); 13C NMR (MeOH-d4) d 174.3, 171.3, 167.9, 153.7,
134.6, 131.2, 130.8, 129.9, 85.7, 84.5, 73.1, 72.4, 65.7,
40.7, 38.7; IR (film) cmÀ1 3392, 1700, 1472, 1274, 1067,
Analysis of pyrimidine hydrate decomposition. Aliquots
were taken at appropriate times from a 1 mM solution
(95/5 20 mM KH2PO4/CH3CN) of the respective pyr-
imidine hydrate maintained at 37 or 90 ꢀC. An internal
standard (10 mL of a 1 mM solution of 10 for 20-deox-
yuridine hydrates, p-methoxybenzyl alcohol for thymi-
dine hydrates) was added and the samples were diluted
with 1:1 40 mM KH2PO4, pH 6.0/CH3CN to inhibit
further decomposition.
711 cmÀ1
.
6R-5,6-Dihydro-6-hydroxy-20-deoxyuridine (9). To
a
solution of 5,6-dihydro-5-bromo-6-hydroxy-20-deoxyur-
idine (15, 75 mg, 0.17 mmol) in 3:1 THF/H2O at 0 ꢀC
was added zinc dust (66 mg, 1.0 mmol) and acetic acid
(19 mg, 0.31 mmol). The reaction was stirred 20 min at
0 ꢀC, filtered through Celite and concentrated at ambi-
ent temperature. The crude reaction mixture was pur-
ified via silica gel flash chromatography (1–6% MeOH/
CHCl3) to afford an 11.5:1 mixture of 9 and 50-benzoyl-
20-deoxyuridine (11, 42 mg, 63% overall) as a white
1H NMR analysis of 20-deoxyuridine hydrate epimer-
ization. Hydrate 8 or 9 (10 mg, 0.028 mmol) and internal
standard (i-PrOH, 0.3 mg, 0.005 mmol) were dissolved
in 2.8:1 D2O/CD3CN (0.95 mL). Samples were placed in

2346
K. Nolan Carter, M. M. Greenberg / Bioorg. Med. Chem. 9 (2001) 2341–2346
NMR tubes and incubated in a 37 ꢀC bath. H NMR
spectra were collected at appropriate time intervals. The
rate of epimerization was determined by observing the
change in integration of the C-6 hydrogen peaks (d 5.29
for 9, d 5.22 for 8).
5. Boorstein, R. J.; Hilbert, T. P.; Cadet, J.; Cunningham,
R. P.; Teebor, G. W. Biochemistry 1989, 28, 6164.
6. Boorstein, R. J.; Hilbert, T. P.; Cunningham, R. P.; Tee-
bor, G. W. Biochemistry 1990, 29, 10455.
7. O’Donnell, R. E.; Boorstein, R. J.; Cunningham, R. P.;
Teebor, G. W. Biochemistry 1994, 33, 9875.
8. Shaw, A. A.; Voituriez, L.; Cadet, J.; Gregoli, S.; Symons,
M. C. R.. J. Chem. Soc., Perkin Trans. 1988, 2, 1303.
9. David, S. S.; Williams, S. D. Chem. Rev. 1998, 99, 1221.
10. Dizdaroglu, M.; Bauche, C.; Rodriguez, H.; Laval, J.
Biochemistry 2000, 39, 5586.
11. Ganguly, T.; Duker, N. J. Nucleic Acids Res. 1991, 19,
3319.
12. Ganguly, T.; Weems, K. M.; Duker, N. J. Biochemistry
1990, 29, 7222.
13. Greenberg, M. M.; Matray, T. J. Biochemistry 1997, 36,
14071.
1
Acknowledgements
Financial support of this work from the National Insti-
tutes of Health (GM-54996) is appreciated. Mass spec-
tra were obtained on instruments supported by the
National Institutes of Health shared instrumentation
grant (GM-49631).
14. Raoul, S.; Bardet, M.; Cadet, J. Chem. Res. Toxicol. 1995,
8, 924.
References and Notes
15. Cadet, J.; Teoule, R. Carbohydrate Res. 1973, 29, 345.
16. Cadet, J.; Ducolomb, R.; Teoule, R. Tetrahedron 1977, 33,
1603.
1. von Sonntag, C. The Chemical Basis of Radiation Biology;
Taylor & Francis: New York, 1987.
2. Barvian, M. R.; Barkley, R. M.; Greenberg, M. M. J. Am.
Chem. Soc. 1995, 117, 4894.
17. Cadet, J.; Kan, L.-S.; Wang, S. Y. J. Am. Chem. Soc.
1978, 100, 6715.
3. Fisher, G. J.; Johns, H. E. In Photochemistry & Photo-
biology of Nucleic Acids; Wang, S. Y., Ed.; Academic: New
York, 1976; Vol. 1, pp 169–224.
4. Cadet, J.; Vigny, P. In Bioorganic Photochemistry; Morri-
son, H., Ed.; Wiley Interscience: New York, 1990; Vol. 1, pp
1–272.
18. Wagner, J. R.; van Lier, J. E.; Berger, M.; Cadet, J. J. Am.
Chem. Soc. 1994, 116, 2235.
19. Hruska, F. E.; Sebastian, R.; Grand, A.; Voituriez, L.;
Cadet, J. Can. J. Chem. 1987, 65, 2618.
20. Deriaz, R. E.; Overend, W. G.; Stacey, M.; Teece, E. G.;
Wiggins, L. F. J. Chem. Soc. 1949, 1879.