An efficient method for the synthesis of 2'-O-modified nucleosides via double alkylation using cyclic sulfates

Article abstract of DOI:10.1016/S0040-4039(99)02329-1

The alkylation of N-3-benzyloxymethyl-5-methyluridine with 1,3,2- dioxathiolane 2,2-dioxide or 1,3,2-dioxathiane 2,2-dioxide resulted in a 2'- O versus 3'-O selectivity of 3:1, respectively. The resulting product has a built-in sulfate leaving group at the terminal end of an ethyl or propyl carbon chain, which can be displaced with sulfur and nitrogen nucleophiles to produce modifications at the 2'-O or 3'-O positions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02329-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1523–1526
An efficient method for the synthesis of 20-O-modified
nucleosides via double alkylation using cyclic sulfates
Allister S. Fraser,a,∗ Andrew M. Kawasaki,a Michael E. Jung b and Muthiah Manoharan a
aDepartment of Medicinal Chemistry, Isis Pharmaceuticals, Inc., 2292 Faraday Ave., Carlsbad, CA 92008, USA
bDepartment of Chemistry and Biochemistry, University of California, Los Angeles, CA 90095, USA
Received 1 December 1999; accepted 14 December 1999
Abstract
The alkylation of N-3-benzyloxymethyl-5-methyluridine with 1,3,2-dioxathiolane 2,2-dioxide or 1,3,2-
dioxathiane 2,2-dioxide resulted in a 20-O versus 30-O selectivity of 3:1, respectively. The resulting product has a
built-in sulfate leaving group at the terminal end of an ethyl or propyl carbon chain, which can be displaced with
sulfur and nitrogen nucleophiles to produce modifications at the 20-O or 30-O positions. © 2000 Elsevier Science
Ltd. All rights reserved.
Keywords: cyclic sulfate; alkylation ; 20-modified nucleosides.
20-O Modifications of ribonucleotides have shown the ability to enhance the antisense properties of
oligonucleotides (e.g., nuclease resistance and binding affinity for the target RNA) (Scheme 1).1–3 Several
20-O-aminoalkyl and 20-O-alkylthioalkyl groups have demonstrated very favorable antisense properties.4
In order to create an efficient synthesis of 20-O-modified oligonucleotides, one needs a facile method to
produce the corresponding monomers. Alkylation of the 20-O position using a five- or six-membered
cyclic sulfate followed by nucleophilic displacement would create a modified nucleoside in only two
or three synthetic steps. There has been an appreciable amount of work reported on the use of cyclic
sulfates5–7 in general organic synthetic transformations.8 However, their use in novel nucleoside synthesis
is a totally unexplored area. In view of the fact that the five- and six-membered cyclic sulfates (namely,
1,3,2-dioxathiolane 2,2-dioxide and 1,3,2-dioxathiane 2,2-dioxide) A and B (Scheme 2), respectively, are
commercially available, we wanted to exploit their use in modified nucleoside synthesis. As a test case,
we decided to synthesize 20-O-(N,N-dimethylaminoalkyl) and 20-O-[(methylthio)ethyl] nucleosides (Fig.
1).
Treatment of the nucleoside 19 with NaH in DMF at −45°C followed by addition of the cyclic
sulfates A or B afforded compounds 2a and 2b in 63 and 50% yields, respectively (Scheme 2). The
selectivity for 20 over 30 alkylation was about 3:1 in both cases. Due to the stability of these sodium
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02329-1
tetl 16257

1524
Scheme 1.
Scheme 2. Reagents and conditions: (i) NaH, DMF, A or B, −45°C to rt; (ii) 3a: NaSCH3, DMF, 80°C; 3b: dimethylamine,
THF, autoclave; (iii) Pd(OH), EtOH, AcOH, H2 at 55 psi; (iv) pyridine, dimethoxytritylchloride; (v) CH2Cl2, diisopropylamine
tetrazolide salt, 2-cyanoethyl-N,N,N0-N0-tetraisopropylphosphorodiamidite. *The BOM group was deprotected under the same
conditions as in (iii) before nucleophilic displacement with NaSCH3

1525
Fig. 1.
sulfate salts as leaving groups, subsequent displacement with nucleophiles required somewhat vigorous
reaction conditions. In the case of dimethylamine as a nucleophile an autoclave was used, while in
the case of sodium methylmercaptide as a nucleophile, reflux temperatures were needed to obtain
nucleophilic displacement of the sulfates to give compounds 3a and 3b in moderate to good yields.
Intramolecular displacement of sulfate salts by nucleophiles has been reported, but to our knowledge
intermolecular displacement has not been reported.7,8 Reductive cleavage of the N-3-benzyloxymethyl
(BOM) protecting group using catalytic hydrogenation over a palladium hydroxide catalyst proceeded
normally for compound 3b to give the 20-modified nucleoside 4b10 in 67% yield. This compound was
then protected at the 50-hydroxyl with a dimethoxytrityl (DMT) group to give compound 5b (56%) which
was then converted to the phosphoroamidite 6b (65%).
In the case of methylthio substitution of the cyclic sulfate, the N-3-BOM group of compound 3a
proved to be more difficult to deprotect due to poisoning of the palladium hydroxide catalyst by
the sulfide functionality. Therefore, the BOM group of the sulfate derivative 2a was removed via
catalytic hydrogenation over a palladium hydroxide catalyst (product not purified) before nucleophilic
displacement with sodium methylmercaptide to give compound 3c (product not purified). The 50-
hydroxyl group of the crude compound 3c was converted to the DMT-protected compound (5a)11 in
34% yield from 2a. Conversion of 5a into the 30-phosphoramidite 6a proceeded in 63% yield.
In conclusion, we have found that alkylation of N-3-benzyloxymethyl-5-methyluridine with the five-
and six-membered cyclic sulfates and subsequent nucleophilic displacement of the sulfate salt leaving
group is an efficient, facile method to produce 20-O-modified 5-methyluridine nucleoside monomers
for incorporation into oligonucleotides. We have used this methodology to synthesize nucleosides
and their phosphoramidites containing 20-O-(N,N-dimethylaminopropyl) and 20-O-[(methylthio)ethyl]
substituents. The synthesis and evaluation of the properties of the modified oligonucleotides containing
these modifications are in progress and will be reported in due course.
Acknowledgements
We are grateful to Robert H. Springer for largescale synthesis of Compound 1.
References
1. Cook, P. D. In Annu. Rep. Med. Chem.; Bristol, J. A., Ed. Second generation antisense oligonucleotides: 20-modifications.
Academic Press: New York, 1998; Vol. 33, p. 313.
2. Martin, P. Helv. Chim. Acta 1995, 78, 486.
3. Monia, B. P.; Lesnik, E. A.; Gonzalez, C.; Lima, W. F.; McGee, D.; Guinosso, C. J.; Kawasaki, A. M.; Cook, P. D.; Freier,
S. M. J. Biol. Chem. 1993, 268, 14514.
4. Manoharan, M. Biochemica et Biophysica Acta 1999, 1489, 117.

1526
5. (a) Kaiser, E. T.; Panar, M.; Westheimer, F. H. J. Am. Chem. Soc. 1963, 85, 602; (b) Kaiser, E. T.; Katz, I. R.; Wulfers, T. F.
J. Am. Chem. Soc. 1965, 87, 3781; (c) Kaiser, E. T. Acc. Chem. Res. 1970, 3, 145.
6. Tomalia, D. A.; Falk, J. C. J. Heterocycl. Chem. 1972, 9, 891.
7. (a) Lohray, B. B.; Gao, Y.; Sharpless, K. B. Tetrahedron Lett. 1989, 30, 2623; (b) Kim, B. M.; Sharpless, K. B. Tetrahedron
Lett. 1989, 30, 655; (c) Gao, Y.; Sharpless, K. B. J. Am. Chem. Soc. 1988, 110, 7538.
8. Lohray, B. B. Synthesis 1992, 11, 1035.
9. Cook, P. D.; Manoharan, M.; Guinosso, C. J. PCT Int. Appl., 1995. Application: WO 9506659 A1 19950309.
10. Compound 4b: 1H NMR (DMSO-d6) δ 1.65 (bs, 2H), 1.78 (s, 3H), 2.17 (s, H), 2.29 (m, 2H), 3.56 (m, 4H), 3.88 (m, 2H),
4.17 (t, 1H), 4.86 (bs, 1H), 5.19 (bs, 1H), 5.84 (d, 1H), 7.80 (s, 1H), 10.24 (bs, 1H). Anal. calcd for C15H25N3O6+0.25 mol
H2O: C, 51.79; H, 7.39; N, 12.08. Found: C, 51.94; H, 7.40; N, 11.80. LRMS (ES) [MH+] m/z calcd: 344. Found: 344.
11. Compound 5a: 1H NMR (CDCl3) δ 1.42 (s, 3H), 2.03 (s, 3H), 2.74 (m, 2H), 3.13 (d, 1H), 3.46 (dd, 2H), 3.81 (bs, 7H), 4.13
(m, 2H), 4.48 (m, 1H), 5.97 (s, 1H), 6.83 (d, 2H), 7.29 (m, 13H), 7.68 (s, 1H), 8.l9 (bs, 1H). Anal. calcd for C34H38N2O8S+1.0
mol H2O: C, 62.56; H, 6.18; N, 4.29. Found: C, 62.87; H, 6.00; N, 4.10. LRMS (ES) [MH] m/z calcd: 633. Found: 633.