Synthesis and conformation of 3′-O,4′-C-methyleneribonucleosides, novel bicyclic nucleoside analogues for 2′,5′-linked oligonucleotide modification

Article abstract of DOI:10.1039/a704376g

Novel bicyclic nucleoside analogues 3′-O,4′-C-methylene-ribonucleosides 1 are conveniently prepared starting from uridine; the sugar puckering of 1 is found to be nearly in the S-conformation by means of PM3 calculations and ¹H NMR studies.

Full text of DOI:10.1039/a704376g

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and conformation of 3A-O,4A-C-methyleneribonucleosides, novel
bicyclic nucleoside analogues for 2A,5A-linked oligonucleotide modification
Satoshi Obika, Ken-ichiro Morio, Daishu Nanbu and Takeshi Imanishi*
Faculty of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka, Suita, Osaka 565, Japan
Novel bicyclic nucleoside analogues 3A-O,4A-C-methylene-
ribonucleosides 1 are conveniently prepared starting from
uridine; the sugar puckering of 1 is found to be nearly in the
S-conformation by means of PM3 calculations and 1H NMR
studies.
pyridine16 gave 13, which was then subjected to ammonolysis
to give 11† (overall 53% yield from 7). Acid catalysed
detritylation of 7 and 11 gave the corresponding 2A,5A-diol
compounds 1 (B = U)† and 1 (B = C)† quantitatively.
In general, the sugar moiety in natural and modified
nucleosides exists in an equilibrium between N- and S-confor-
mations. The conformation of nucleoside analogues in solution
is easily obtainable from coupling constants between the ribose
ring protons in the 1H NMR spectrum.17 The simplest equation
for calculation of percentage of S conformation is shown in
eqn. (1). The calculation results for the obtained nucleoside
Non-genetic 2A,5A-linked oligonucleotides have been attracting
much attention since the syntheses and properties of oligo-
nucleotides with 2A,5A-linkages were reported in 1992.1–3
Using NMR studies4 and molecular modeling,5 it was suggested
that oligoribonucleotides containing 2A,5A-linkages (2A,5A-RNA)
preferentially
adopted
the
C2A-endo
conformation
S(%) = 100 3 (J1A2A 2 1)/6.9
(1)
(S-conformation), while 2A,5A-linked oligodeoxyribonucleotides
(2A,5A-DNA) existed mainly with C3A-endo sugar puckering
(N-conformation). Interestingly, these results are the opposite of
what is observed with the corresponding ‘genetic’ 3A,5A-RNA
and -DNA. There have been some reports concerning successful
application of 2A,5A-linked oligonucleotides to antisense uses.6–9
On the other hand, 2-5A (2A,5A-oligoadenylate 5A-triphosphate)
is known to be formed in interferon-treated cells and to have
antiviral activity with the aid of 2-5A dependent ribonuclease
(RNase L),10 and intensive studies on various 2-5A analogues
have been reported.11
analogues are summarized in Table 1, and show that all of the
3A-O,4-C-methyleneribonucleoside analogues have very large S
values.
A
PM3 calculation18§ also indicates that an
S-conformation (C1A-exo-C2A-endo) is the most stable con-
formation for the ribose ring in 1 (Fig. 1). It is very interesting
and noteworthy that the oxetane ring fused nucleoside deriva-
HO
HO
TsO
U
U
U
O
O
O
We have designed a bicyclic nucleoside analogue, 3A-O,4A-
C-methyleneribonucleoside 1,12 which is a promising precursor
for both novel types of therapeutic nucleoside derivatives and
2A,5A-linked oligonucleotide analogues. The sugar moiety in 1
consists of a 2,6-dioxabicyclo[3.2.0]heptane ring system, with
the tetrahydrofuran part conformationally restricted by the
fused oxetane ring. Here we describe a convenient first
HO
TsO
HO
O
O
O
O
O
O
i
+
ii–iii
2
4 12%
HO
3 69%
RO
DMTrO
5′
U
U
B
synthesis of 1 (B
= U and C), and also discuss its
O
O
O
conformation.
4′
iv
viii
1′
3′
2′
The synthetic route to the target molecule 1 (B = U) is shown
in Scheme 1. A selective toluene-p-sulfonylation of the
diastereotopic hydroxymethyl groups in 2A,3A-O-cyclohexyl-
idene-4A-hydroxymethyluridine13,14 2 afforded monotosylate 3
(69%) along with its isomer 4 (12%). The stereochemistry at
C4A in 3 was confirmed by means of NOE measurements.
Acidic hydrolysis of 3 gave triol 5 (94%) and then selective
protection of the primary hydroxy group in 5 with a
4,4A-dimethoxytrityl (DMTr) group in the usual manner af-
forded 6 (56%).
TsO
OH OH
94%
6 R = DMTr 56%
O
7 63%
OH
O
OH
1
5 R = H
B = U, C
N
N
N
v
N
N
viii
O
DMTrO
C
DMTrO
DMTrO
vi
U
O
O
O
vii
Oxetane ring formation from 6 was accomplished on
treatment with a large excess of sodium hexamethyldisilazide in
THF at room temperature, yielding the desired product 7†
(63%). The exclusive oxetane ring formation is attributable to
the predominant S-conformation of the starting material 6,‡ in
which only the 3A-OH is thought to be located near the
4A-methylene carbon centre; the 2A-OH is too far away to attack
the 4A-methylene carbon. In fact, 1A-methoxy congener 8, which
should exist mainly in the N-conformation due to the anomeric
effect,‡ gives a ca. 1:1 mixture of 2A-O,4A-C-methylene
derivative 9 and 3A-O,4A-C-methylene derivative 10 under the
same conditions.15
53% from 12
O
OH
11
O
OAc
O
OAc
13
12 100%
RO
iv
RO
RO
OMe
OMe
OMe
O
O
O
+
TsO
O
OH
OH OH
O
OH
10 35%
9 43%
8
Scheme 1 Reagents and conditions: i, TsCl (1.7 equiv.), 110 °C; ii,
CF3CO2H–H2O (98:2), room temp.; iii, 4,4A-dimethoxytrityl chloride,
DMAP, pyridine, room temp., iv, (Me3Si)2NNa (10 equiv.), THF, 20 °C; v,
Ac2O, pyridine, room temp.; vi, 1,2,4-triazole, 4-chlorophenyl phosphoro-
dichloridate, pyridine, room temp.; vii, aqueous NH3, 1,4-dioxane, room
temp.; viii, 3% CCl3CO2H in H2O, CH2Cl2, room temp.
Transformation of 7 into the cytidine derivative 11 by
exchange of nucleobase was also undertaken as shown in
Scheme 1. After 2A-O-acetylation of 7, treatment of 12 with
1,2,4-triazole and 4-chlorophenyl phosphorodichloridate in
Chem. Commun., 1997
1643

View Article Online
1
Table 1 Conformational analysis of nucleoside analogues using H NMR
spectroscopy (in CD3OD)
d, J 7), 6.85 (4 H, d, J 9), 7.21–7.42 (9 H, m), 7.67 (1 H, d, J 7); m/z (FAB)
558 (M + H)+. For 1 (B = U): mp 215–216 °C (AcOEt); [a]D 218.2 (c
17
1.2, acetone); 1H NMR (CD3OD): d 3.74, 3.82 (2 H, AB, J 12), 4.21 (1 H,
dd, J 5, 8), 4.42, 4.82 (2 H, AB, J 8), 5.05 (1 H, d, J 5), 5.66 (1 H, d, J 9),
6.38 (1 H, d, J 8), 7.67 (1 H, d, J 9); Calc. for C10H12N2O6·1/2H2O; C,
45.29; H, 4.94; N, 10.56. Found C, 45.07; H, 4.82; N, 10.15%. For 1 (B =
HO
B
HO
B
O
O
X
OH
Y
Y
C): mp 224–225 °C (PriOH); [a]D +32.6 (c 0.49, H2O); 1H NMR
25
X
OH
(CD3OD): d 3.73, 3.81 (2 H, AB, J 12), 4.16 (1 H, dd, J 5, 7), 4.52, 4.81 (2
H, AB, J 8), 5.04 (1 H, d, J 5), 5.94 (1 H, d, J 8), 6.41 (1 H, d, J 7), 7.68 (1
H, d, J 8); Calc. for C10H13N3O5·1/3H2O; C, 45.98; H, 5.27; N, 16.09.
Found 46.04; H, 5.02; N, 15.82%.
N-conformation
S-confirmation
Nucleoside analogues
J1A2A/Hz
S (%)a
‡ J1A2A Values (6.0 Hz for 6 and 0 Hz for 8) provide the conformational
information.
§
Cambridge Soft Corporation) utilizing the MNDO-PM3 Hamiltonian was
used for the semi-empirical MO calculations. Geometry optimization was
carried out with the use of the keyword MMOK implemented in MOPAC.
All initial structures used for the MO calculation were generated by
reference to the X-ray structures of natural nucleosides (ref. 21). Numerical
calculations were performed on a Power Macintosh computer.
HO
B
O
B = U
B = C
7.5
7.3
94
91
The MOPAC93 molecular orbital package (CS MOPAC ProTM
,
O
OH
HO
HO
B
O
B = U
B = C
4.6
2.8
52
26
OH OH
1 R. Kierzek, L. He and D. H. Turner, Nucleic Acids Res., 1992, 20,
1685.
2 J. P. Dougherty, C. J. Rizzo and R. Breslow, J. Am. Chem. Soc., 1992,
114, 6254.
3 H. Hashimoto and C. Switzer, J. Am. Chem. Soc., 1992, 114, 6255.
4 H. Robinson, K.-E. Jung, C. Switzer and A. H.-J. Wang, J. Am. Chem.
Soc., 1995, 117, 837.
5 V. Lalitha and N. Yathindra, Curr. Sci., 1995, 68, 68.
6 P. A. Giannaris and M. J. Damha, Nucleic Acids Res., 1993, 21,
4742.
B
O
B = U
B = C
1.2b
0c
3
0
OH
Calculated from eqn. (1). Ref. 19 (Measured in [2H6]DMSO). Ref.
20.
a
b
c
7 J. Seki, K. Kuroda, H. Ozaki and H. Sawai, Nucleic Acids Symp. Ser.,
1993, 29, 71.
8 A. Maran, R. K. Maitra, A. Kumar, B. Dong, W. Xiao, G. Li, B. R. G.
Williams, P. F. Torrence and R. H. Silverman, Science, 1995, 265,
789.
C(5′)
9 E. R. Kanimalla, A. Manning, Q. Zhao, D. R. Shaw, R. A. Byrn, V.
Sasisekharan and S. Agrawal, Nucleic Acids Res., 1997, 25, 370.
10 H. C. Schro¨der, R. J. Suhadolnik, W. Pfleiderer, R. Charubala and
W. E. G. Muller, Int. J. Biochem., 1992, 24, 55; G. C. Sen and P.
Lengyei, J. Biol. Chem., 1992, 267, 5017.
C(3′)
C(2′)
C(1′)
C(2)
N(1)
C(4′)
O(4′)
11 C. Horndler and W. Pfleiderer, Helv. Chim. Acta, 1996, 79, 718 and
references cited therein.
12 Some other bicyclic nucleoside analogues have appeared in the
literature. For example, M. Tarkoy and C. Leumann, Angew. Chem., Int.
Ed. Engl., 1996, 35, 1968; R. Zou and M. D. Matteucci, Tetrahedron
Lett., 1996, 37, 941; K.-H. Altmann, R. Imwinkelried, R. Kesselring and
G. Rihs, Tetrahedron Lett., 1994, 35, 7625.
Fig. 1 Computer-generated (PM3) representation of 1 (B = U)
13 S. L. Cook and J. A. Secrist, J. Am. Chem. Soc., 1979, 101, 1554.
14 G. H. Jones, M. Taniguchi, D. Tegg and J. G. Moffatt, J. Org. Chem.,
1979, 44, 1309.
15 S. Obika, Y. Hari, K.-i. Morio, D. Nanbu and T. Imanishi, unpublished
work.
16 T.-S. Lin, M.-Z. Luo, M.-C. Liu, Y.-L. Zhu, E. Gullen, G. E. Dutschman
and Y.-C. Cheng, J. Med. Chem., 1996, 39, 693.
17 C. Altona and M. Sundaralingam, J. Am. Chem. Soc., 1973, 95, 2333;
1972, 94, 8205.
18 J. J. P. Stewart, J. Comput. Chem., 1989, 10, 209, 221.
19 T.-S. Lin, J.-H. Yang, M.-C. Liu, Z.-Y. Shen, Y.-C. Cheng, W. H.
Prusoff, G. I. Birnbaum, J. Giziewicz, I. Ghazzouli, V. Brankovan, J.-S.
Feng and G.-D. Hsiung, J. Med. Chem., 1991, 34, 693.
20 T. L. Sheppard, A. T. Rosenblatt and R. Breslow, J. Org. Chem., 1994,
59, 7243.
tives 1 exist predominantly in an S-conformation, in contrast
with the typical monocyclic nucleoside compounds (Table
1).19,20
Further chemical modification of 1 and its incorporation into
oligonucleotides are now in progress.
Footnotes and References
* E-mail address: imanishi@phs.osaka-u.ac.jp
21
† Selected data for 7: mp 120–121 °C (Et2O–hexane); [a]D 237.7 (c 1.1,
acetone); 1H NMR ([2H6]acetone): d 2.92 (1 H, br s), 3.47, 3.51 (2 H, AB,
J 10), 3.85 (6 H, s), 4.36 (1 H, dd, J 4, 4), 4.52, 4.83 (2 H, AB, J 8), 5.11
(1 H, d, J 4), 5.57 (1 H, d, J 8), 6.51 (1 H, d, J 8), 6.96 (4 H, d, J 9), 7.39–7.41
(7 H, m), 7.52 (2 H, d, J 5), 7.71 (1 H, d, J 9); m/z (EI) 558 (M+). For 11:
22
1
mp 139–140 °C (Et2O–hexane); [a]D + 0.60 (c 0.34, MeOH); H NMR
(CD3OD): d 3.40, 3.50 (2 H, AB, J 10), 3.76 (6 H, s), 4.14 (1 H, dd, J 4, 7),
4.48, 4.72 (2 H, AB, J 7), 4.98 (1 H, d, J 4), 5.79 (1 H, d, J 7), 6.56 (1 H,
21 S. Arnott and W. Hukins, Biochem. J., 1972, 130, 453.
Received in Cambridge, UK, 23rd June 1997; 7/04376G
1644
Chem. Commun., 1997