Towards novel amide-modified oligonucleotides: Asymmetric synthesis of 5′-(S)-methyl-3′-carboxymethylene-3′-deoxythymidine

Article abstract of DOI:10.1016/S0040-4039(02)01053-5

The synthesis of novel 5′-(S)-methyl-3′-carboxymethylene-3′-deoxythymidine is reported. Key steps involve diastereoselective lactonization, enantioselective enzymatic ester hydrolysis and diastereoselective glycosidation of a key intermediate with thymine with 100% β-selectivity via Lewis acid mediated cleavage of a [3.2.1] oxabicyclic lactone.

Full text of DOI:10.1016/S0040-4039(02)01053-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 5461–5464
Towards novel amide-modified oligonucleotides: asymmetric
synthesis of 5%-(S)-methyl-3%-carboxymethylene-3%-deoxythymidine
Sebastian Wendeborn,* Hannes Nussbaumer, Fre´de´ric Robert, Mario Jo¨rg and
Johannes Paul Pachlatko
Syngenta Crop Protection AG, CH-4002 Basel, Switzerland
Received 22 April 2002; accepted 2 June 2002
Abstract—The synthesis of novel 5%-(S)-methyl-3%-carboxymethylene-3%-deoxythymidine is reported. Key steps involve diastereo-
selective lactonization, enantioselective enzymatic ester hydrolysis and diastereoselective glycosidation of a key intermediate with
thymine with 100% b-selectivity via Lewis acid mediated cleavage of a [3.2.1] oxabicyclic lactone. © 2002 Elsevier Science Ltd. All
rights reserved.
Previously we have reported that the replacement of the
phosphoric acid diester in oligonucleotides with an
amide I can lead to increased affinities towards comple-
mentary RNA and DNA targets and higher nuclease
resistance. We have also demonstrated that substituents
X and Y, as well as R% (R=Me) significantly further
preorganize the amide backbone and lead to substantial
improvements in their binding affinities towards com-
plementary oligonucleotide targets.1 We were question-
ing whether the properties of amide containing
oligonucleotides could be further improved by addi-
tional substituents (R%%) in the 5%-position of the car-
boxylic acid portion of the amide dimer. Here, we
report the stereoselective synthesis of II (R%%=Me) from
achiral starting materials.
The retrosynthetic analysis of the desired 5%-(S)-methyl-
3%-carboxymethylene-3%-deoxythymidine is outlined in
Scheme 1. We envisioned selective introduction of the
nucleobase at a late stage facilitating the synthesis of all
four base analogs (T, C, A, G). Therefore, a key
challenge would be the selective introduction of the
nucleobase from the b-face of an intermediate precur-
sor. We anticipated that cyclizations of 3 to 2 would
control the stereochemistry at the 3% position, placing
the carboxethoxymethylene group trans to the C(5%)
group. Intermediate 3 should be readily available from
sorbic acid ethyl ester.
* Corresponding author. Fax: +41 61 3235500; e-mail:
sebastian.wendeborn@snygenta.com
Scheme 1. Retrosynthetic analysis.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01053-5

5462
S. Wendeborn et al. / Tetrahedron Letters 43 (2002) 5461–5464
The realization of this retrosynthetic analysis is outlined
in Scheme 2. Sorbic acid ethyl ester 4 was submitted to
Sharpless asymmetric bishydroxylation reaction accord-
ing to published procedures,2 yielding, upon ketaliza-
tion, compound 6 (45%, 80% ee), which was reacted
with sodium salt of diethylmalonate to give 7. Remark-
ably, 7 could not be decarboxylated under thermal
conditions (NaCl, H2O, DMSO, 160°C, 16 h)3 but
readily reacted to 8 when the reaction was conducted
under microwave irradiation (NaCl, H2O, DMSO, MW
(800 W), 0.3 h). Treatment of 8 with trifluoroacetic acid
in ethanol and dichloroethane resulted in acetonide
cleavage and subsequent lactonization gave compounds
tiomeric excess above the 80% ee derived from the
asymmetric Sharpless bishydroxylation reaction.
Screening of a number of different hydrolases led us to
select the Bacillus licheniformis protease Novozyme 243
from Novo, which exclusively hydrolyzed the desired
enantiomer of 10 yielding 11 in 63% yield and 100% ee
on multigram scale in a biphasic reaction system.7
Treatment of crude 12 with pyridinium toluenesulfonic
acid in toluene lead to bicyclic 13 (Scheme 3).
We next focused our attention on the introduction of
thymine. To this end we worked with a model system
1
2 and 9 (ratio 2:9: 5.5:1, H NMR) in 42% and 11%
isolated yield, respectively.4 Our expectation that
cyclization would occur with concomitant stereo-con-
trol at C(3%) was fully met. The minor product 9
presumably derives from the 3% epimer of 2 which
readily forms the bicyclic 6,5 cis-fused dilactone. We
have demonstrated that isolated 2 does not isomerize to
9 when resubmitted to the acidic reaction conditions
employed for its formation.
Under basic conditions, however (Scheme 2), 2 reacted
to 9, presumably through isomerization to the more
stable 6-membered ring lactone by attack of the C(5%)-
hydroxyl group onto the C(1%) followed by cyclization
of the liberated C(4%) hydroxyl group onto the ethyl
ester. Taking these results into consideration we opted
to protect the 5%-alcohol using benzyl trichloro-
acetamide and a catalytic amount of trifluoromethane-
sulfonic acid by the method of Bundle to give 10 in 76%
yield.5 Attempts to reduce the lactone 10 to the corre-
sponding lactol in the presence of the ethyl ester were
unsuccessful.6 We therefore lyophilized 10 with aqueous
NaOH in THF to carboxylic acid 11, which could be
selectively reduced to the lactol by DIBAL to give 12.
The required hydrolysis of ethyl ester 10 to 11 was also
seen as a welcome opportunity to increase the enan-
Scheme 3. (a) 2 equiv. BnO(NH)CCl3, 0.1 equiv. MeSO3H,
cyclohexane:CH2Cl2 (2:1), rt, 4 h. (b) Bacillus licheniformis
protease Novozyme 243, DIPE/K-buffer pH 6.8, 100% ee. (c)
2.4 equiv. NaOH, THF, H2O, 0°C“rt, 1.5 h. (d) 2.2 equiv.
DIBAL, Tol, −78°C, 0.5 h, 10 min, 0°C. (e) 0.1 equiv. PPTS,
,
3 A MS, PhMe, 70°C.
t
Scheme 2. (a) 1.5 equiv. a-AD-mix, 1 equiv. MeSO2NH2, BuOH:H2O (1:1), 4°C, 18 h. (b) (MeO)2CMe2, cat TsOH, acetone. (c)
1.1 equiv. (EtOOC)2CH2, 1.1 equiv. NaOEt, EtOH, rt, 4 h. (d) 9 equiv. H2O, 0.5 equiv. NaCl, DMSO, microwave, 800 W, 0.25
h. (e) CF3COOH:EtOH:CH2Cl2 (1:1:8). (f) NaH, THF.

S. Wendeborn et al. / Tetrahedron Letters 43 (2002) 5461–5464
5463
sterically very demanding Lewis acid would lead to
intermediate 18 (or 17) in which the attack from the
a-face would be blocked by the bulk of the Lewis acid.
lacking the 5%(S) methyl group, which was readily avail-
able from previously described intermediates.8 We envi-
sioned two different alternatives in order to favor
introduction of thymine from the b-face: Lewis acid
mediated glycosidation and side group participation of
the 3%-methylene carboxylic acid in 14,9–11 or reaction
with bicyclic intermediate 15. Reaction of bis-silylated
thymine with 14 proceeded well in the presence of a
number of different Lewis acids and solvents (Scheme
4, entries 1–4).12 Unfortunately, b/a-selectivity could
not be improved to a ratio of more than 3:1, and the
two epimers could not be separated by chromatogra-
phy. We next turned our attention to bicyclic 15 but
unfortunately all commercially available Lewis acids
tested gave results not superior to those obtained with
14 (data not shown). In the absence of any Lewis acid
15 did react with bis-silylated thymine, however only at
elevated temperatures, and, to our disappointment with
unsatisfactory b/a-selectivity (Scheme 4, entry 5).
Scheme 5. Lewis acid coordination to 15.
We concluded that Lewis acid activation of 15 led to an
oxonium ion which was little or not stabilized by the
internal, metal-coordinated carboxylate and that the
ratio of 16b/a is controlled mainly by steric factors and
not by side group participation (Scheme 5).
Therefore, we turned our attention to the sterically very
Scheme 6. 1 equiv. MAD, CH2Cl2, −78°C, then 2 equiv.
T(SiMe3)2, −78“−20°C.
demanding Lewis acids developed by Yamamoto.13
A
Scheme 4. a/b selectivity of thymine introduction under modified Vorbru¨ggen conditions.

5464
S. Wendeborn et al. / Tetrahedron Letters 43 (2002) 5461–5464
In full agreement with our analysis, the use of
ATPH13,14 as Lewis acids led to formation 16b and 16a
in a ratio 45:1; even more impressive results were
obtained with MAD,13,15 which led to exclusive forma-
tion of the desired 16b. Interestingly, catalytic amounts
of MAD led to formation of 16 with only slightly
diminished b-selectivity.
2. (a) Becker, H.; Soler, M. A.; Sharpless, K. B. Tetrahedron
1995, 51, 1345; (b) Allevi, P.; Tarocco, G.; Longo, A.;
Anastasia, M.; Cajone, F. Tetrahedron: Asymmetry 1997,
8, 1315.
3. Krapcho, P. Synthesis 1982, 805.
4. Mulzer, J.; Kappert, M.; Huttner, G.; Jibril, I. Angew.
Chem., Int. Ed. Engl. 1984, 23, 704.
5. (a) Wessel, H.-P.; Iverson, T.; Bundle, D. R. J. Chem.
Soc., Perkin Trans. I 1985, 2247; (b) Nakajima, N.;
Horita, K.; Abe, R.; Yonemitsu, O. Tetrahedron Lett.
1988, 29, 4139 The resulting trichloroacetamide was best
removed from the crude product by Kugelrohr destilla-
tion, since it co-eluted on silica gel together with 10.
6. Examples for selective reductions of lactols in the pres-
ence of esters can be found: (a) Binch, H.; Stangier, K.;
Thiem, J. Carbohydrate Res. 1998, 306, 409; (b) Kohn, P.;
Samaritano, R. H.; Lerner, L. M. J. Am. Chem. Soc.
1965, 87, 5475; (c) Tse, A.; Mansour, T. Tetrahedron
Lett. 1995, 36, 7807.
7. ee’s were determined by chiral HPLC; hydrolysis of the
enantiomer of 10 (prepared from 4 by the same route but
using b-AD-mix) with NaOH gave enantiomer of 12,
allowing unambiguous determination of enantiomeric
excess by chiral HPLC.
When these conditions were applied to 13, compound
19 was obtained as a single diastereomer in 61% yield
(Scheme 6).
In conclusion, we have developed a highly stereoselec-
tive synthesis of 5%-(S)-methyl-3%-carboxymethylene-3%-
deoxythymidine. The stereochemistry at C(4%) and C(5%)
has been controlled by asymmetric Sharpless bishydroxy-
lation, while the C(3%)-stereochemistry could be
efficiently controlled during cyclization of the C(3%)-
pro-chiral diester 7 to the lactone 2. Enantiomeric
excess was increased to 100% through enantioselective
enzymatic hydrolysis of ester 10. In a final key step
thymine introduction was accomplished with exclusive
b-face selectivity through the novel use of Yamamoto’s
bulky aluminum based Lewis acid ‘MAD’ in a modified
Vorbru¨ggen-type nucleosidation reaction. We expect
that this methodology may be applied to the synthesis
of related nucleoside and C-nucleoside analogs.
8. See Ref. 1b.
9. Side group participation was reported to be successful in
a related Vorbru¨ggen reaction when the thionoester was
employed: Lavalle´e, J.-F.; Just, G. Tetrahedron Lett.
1991, 32, 3472.
10. For side group participation of methylenephosphonoth-
ioate in a related Vorbru¨ggen reaction see: Yokomatsu,
T.; Sada, T.; Shimizu, T.; Shibuya, S. Tetrahedron Lett.
1998, 39, 6299.
The use of 19 in the synthesis of novel oligonucleotide
backbone modifications will be reported elsewhere.
Acknowledgements
11. While our studies were in progress a similar approach for
the synthesis of C-glycosides from bicyclic lactones has
been reported: Gaertzen, O.; Misske, A. M.; Wolbers, P.;
Hoffmann, H. M. R. Synlett 1999, 1041.
We thank Mr. Hans-Rudolf Baumgartner for perform-
ing chiral HPLC analysis and Dr. Tammo Winkler for
superb NMR support.
12. Niedballa, U.; Vorbru¨ggen, H. J. Org. Chem. 1974, 39,
3654.
References
13. (a) Maruoka, K.; Imoto, H.; Yamamoto, H. Synlett
1994, 6, 441; (b) Maruoka, K.; Imoto, H.; Saito, S.;
Yamamoto, H. J. Am. Chem. Soc. 1994, 116, 4131; (c)
Maruoka, K.; Shimada, I.; Imoto, H.; Yamamoto, H.
Synlett 1994, 7, 519; (d) Maruako, K.; Concepcion, B.;
Murase, N.; Oishi, M.; Hirayama, N.; Yamamoto, H. J.
Am. Chem. Soc. 1993, 115, 3943; (e) Saito, S.; Shiozawa,
M.; Yamamoto, H. Angew. Chem., Int. Ed. 1999, 38,
1769; (f) Saito, S.; Yamamoto, H. J. Chem. Soc., Chem.
Commun. 1997, 1585.
1. (a) Lebreton, J.; De Mesmaeker, A.; Waldner, A.;
Fritsch, V.; Wolf, R. M.; Freier, S. M. Tetrahedron Lett.
1993, 34, 6383; (b) De Mesmaeker, A.; Waldner, A.;
Lebreton, J.; Hoffmann, P.; Fritsch, V.; Wolf, R. M.;
Freier, S. M. Angew. Chem., Int. Ed. Engl. 1994, 33, 226;
(c) Idziak, I.; Just, G.; Damha, M. J.; Giannaris, P. A.
Tetrahedron Lett. 1993, 34, 5417; (d) Wendeborn, S.;
Wolf, R. M.; De Mesmaeker, A. Tetrahedron Lett. 1995,
36, 6879; (e) De Mesmaeker, A.; Haener, R.; Martin, P.;
Moser, H. E. Acc. Chem. Res. 1995, 28, 366; (f) De
Mesmaeker, A.; Altmann, K.-H.; Waldner, A.; Wen-
deborn, S. Curr. Opin. Struct. Biol. 1995, 5, 343.
14. ATPH=aluminum tris(2,6-diphenylphenoxide).
15. MAD=methylaluminum bis(2,6-di-tert-butyl-4-methyl-
phenoxide).