2′,3′-Anhydrouridine

Article abstract of DOI:10.1039/a608179g

The previously unreported title compound 4 is readily obtained by treating 2,2′-anhydro-1-β-D-arabinofuranosyluracil 1 with sodium hydride in dry dimethyl sulfoxide at room temperature.

Full text of DOI:10.1039/a608179g

View Article Online / Journal Homepage / Table of Contents for this issue
2
A,3A-Anhydrouridine
Anwar Miah, Colin B. Reese* and Quanlai Song
Department of Chemistry, King’s College London, Strand, London, UK WC2R 2LS
The previously unreported title compound 4 is readily
obtained by treating 2,2A-anhydro-1-b- -arabinofuranosyl-
uracil 1 with sodium hydride in dry dimethyl sulfoxide at
resonance signals (d 59.5 and 58.9 for the ribo-compound 4; d
D
55.9 and 55.6 for the lyxo-compound 5) in the 13C NMR spectra
of these compounds are particularly noteworthy.
room temperature.
When 2A,3A-anhydrouridine 4 was allowed to stand in
triethylamine–methanol (1:9 v/v) solution at room temperat-
ure, it was quantitatively converted back into the 2,2A-anhy-
dronucleoside 1 within 1 h. The interconversion of compounds
1 and 4 is reminiscent of the previously reported inter-
conversion6 of 2A,3A-anhydro-7,8-dihydro-8-oxoadenosine 6
and the isomeric 8,2A-anhydronucleoside 7. Perhaps the reason
why the preparation of 2A,3A-anhydrouridine 4 has eluded
previous workers is that stringently anhydrous conditions are
required if the irreversible conversion of 1 into 1-b-d-
arabinofuranosyluracil 8 is to be avoided. It should be added
that the preparation of 2A,3A-anhydro-3-N-methyluridine 9 has
been reported.7
In the course of their early and pioneering studies on the
chemistry of pyrimidine anhydronucleosides, Brown, Todd and
co-workers reported1 that when 2,2A-anhydro-1-b-d-arab-
inofuranosyluracil 1 washeated at 100 °C in N,N-dimethylform-
amide (DMF) solution with a very large excess of sodium
ethanethiolate, the 3A-ethylsulfanyl derivative 2 was obtained
and isolated as a colourless glass in ca. 55% yield. We
subsequently showed that when the anhydronucleoside 1 was
heated in DMF solution at 60 °C with an excess both of
1
1
3
3
ethanethiol and N , N , N , N -tetramethylguanidine (TMG),
A-deoxy-2A-(ethylsulfanyl)uridine 3 was obtained and isolated
2
as a crystalline solid in 93% yield.
Brown et al. rationalized their observations by suggesting
that, under their reaction conditions, the 2,2A-anhydronucleoside
When 2,2A-anhydro-1-b-d-arabinofuranosyluracil 1 (1.0
1
mmol) was allowed to react with sodium ethanethiolate
[prepared from ethanethiol (4.0 mmol) and sodium hydride (2.0
3
1
first isomerized to 2A,3A-anhydrouridine 4 which then
mmol)] in dry DMF (10 cm ) solution at room temperature for
underwent nucleophilic attack at C-3A to give the observed
2
product 2. At the time of our earlier study, we were unable to
explain why the sodium and N , N , N , N -tetramethylguani-
1
1
3
3
(a)
(b)
dinium salts of ethanethiol should react in a different manner
3
with the 2,2A-anhydronucleoside 1, and previous attempts to
prepare 2A,3A-anhydrouridine 4 have been unsuccessful. We now
4
report that when 2,2A-anhydro-1-b-d-arabinofuranosyluracil 1
was treated with 1.5 equiv. of sodium hydride in dry dimethyl
sulfoxide (DMSO) at room temperature for 20 min, it was
virtually quantitatively converted into 2A,3A-anhydrouridine 4
which was isolated as a colourless crystalline solid† in 69%
yield. The characterization of this epoxy compound 4 is based
on microanalytical data (found: C, 47.63; H, 4.34; N, 12.36.
1
C
9
H
10
N
2
O
5
requires: C, 47.79; H, 4.46; N, 12.38%) and on H
1
3
and C NMR spectroscopic evidence. It can be seen from Fig. 1
SO] of 2A,3A-anhydrouridine 4
Fig. 1(a)] and the previously reported and much more stable
180
140
100
δ / ppm
60
20
that the 13C NMR spectra [(CD
[
3 2
)
5
Fig. 1 13C NMR spectra in (CD
)
SO of (a) 2A,3A-anhydrouridine 4 and (b)
3
2
2
A,3A-anhydro-1-b-d-lyxofuranosyluracil 5 [Fig. 1(b)] are
2
A,3A-anhydro-1-b-d-lyxofuranosyluracil 5
closely similar. The comparatively high field C-2A and C-3A
NH2
NH2
O
N
O
N
O
N
H
N
N
N
N
N
8
N
NH
NH
O
O
2
HO
HO
N
N
N
HO
HO
HO
O
O
O
O
O
O
O
SEt
O
3′
2′
2′
2′
O
O
2′
3′
HO
6
7
HO
OH
HO
SEt
1
2
3
O
N
O
O
O
NH
NMe
NH
NH
NH
HO
HO
O
5′
N
O
N
O
HO
HO
N
O
N
O
O
HO
O
O
O O
O
O
O
O
3′
2′
3′
HO
OH
4
5
8
9
10
Chem. Commun., 1997
407

View Article Online
1
8 h, 2A-deoxy-2A-(ethylsulfanyl)uridine 3 (mp 184–185°C, lit.,2
mp 183.5 °C) was obtained as the sole nucleoside product in
2% isolated yield. The course of this reaction corresponds with
anosyluracil 1 before it melted. Compounds 1 and 4 have Rf = 0.12 and
0
‡
5
7
.44, respectively, in chloroform–methanol (85:15 v/v).
Found: C, 45.80; H, 5.58; N, 9.65. Calc. for C11
.59; N, 9.72%; d [(CD SO] 1.19 (3 H, t, J 7.4 Hz), 2.60 (2 H, quart, J
.4 Hz), 3.36 (1 H, m), 3.64 (2 H, m), 4.11 (1 H, m), 4.29 (1 H, m), 5.06 (1
16 2 5
H N O S: C, 45.82; H,
9
3 2
)
2
H
that of the previously reported reaction between compound 1,
ethanethiol and TMG. However, when the 2,2A-anhydronucleo-
side 1 (1.0 mmol) was first treated with sodium hydride (4.0
H, t, J 4.4 Hz), 5.69 (2 H, m), 5.82 (1 H, d, J 5.6 Hz), 7.93 (1 H, d, J 8.1 Hz),
1
1
1.34 (1 H, br s); d
C
[(CD
3
)
2
SO] 15.2, 26.1, 50.4, 61.4, 78.5, 80.3, 87.8,
0.50 in chloroform–methanol (85:15 v/v)
3
mmol) in DMF (10 cm ) for 2 h at room temperature and then,
01.9, 140.9, 150.9, 163.1; R
f
following the addition of ethanethiol (2.0 mmol), the reactants
were stirred at room temperature for a further period of 18 h, the
(the Rf of the 2A-ethylsulfanyl isomer 3 is 0.43 in the same solvent
system).
3
9
A-ethylsulfanyl derivative 2 (mp 130–132 °C)‡ was obtained in
0% isolated yield. It would therefore seem likely that the
discrepancy between the results obtained by us in our previous
study and by Brown et al. was due to the latter workers’
sodium ethanethiolate being contaminated with a strong base
References
2
1
1
D. M. Brown, D. B. Parihar, A. Todd and S. Varadarajan, J. Chem. Soc.,
958, 3028.
1
(e.g. sodium methoxide). Finally, when the 2,2A-anhydronucleo-
2 K. J. Divakar and C. B. Reese, J. Chem. Soc., Perkin Trans. 1, 1982,
side 1 was heated at 60 °C for 18 h with a five-fold excess of
sodium hydride in dry N,N-dimethylacetamide in the absence of
an additional nucleophile, 3A,5A-anhydro-1-b-d-xylofuranosyl-
1625.
3 J. G. Buchanan and D. R. Clark, Carbohydr. Res., 1979, 68, 331.
4 J. P. H. Verheyden, D. Wagner and J. G. Moffatt, J. Org. Chem., 1971,
36, 250.
J. F. Codington, R. Fecher and J. J. Fox, J. Org. Chem., 1962, 27, 163.
J. B. Chattopadhyaya and C. B. Reese, J. Chem. Soc., Chem. Commun.,
8
uracil 10 [mp 214–217 °C (decomp.), lit., mp 214–216 °C] was
5
6
obtained and isolated in 62% yield. The fact that 2,2A-anhydro-
1
-b-d-arabinofuranosyluracil 1 can so easily be converted in
1
976, 860.
situ into the isomeric 2A,3A-epoxide 4 makes it a particularly
7
8
M. M a` rton-Mer e´ sz, J. Kuszmann, I. Pelczer, L. P e` rk a` nyi, T.
Korits a` nszky and A. K a` lm a` n, Tetrahedron, 1983, 39, 275.
I. L. Doerr, J. F. Codington and J. J. Fox, J. Org. Chem., 1965, 30,
versatile synthetic intermediate.
4
67.
Footnotes
†
4
It was not possible to determine the melting point of 2A,3A-anhydrouridine
as, on heating, it isomerized back to 2,2A-anhydro-1-b-d-arabinofur-
Received, 4th December 1996; Com. 6/08179G
408
Chem. Commun., 1997