Synthesis of 2'-C-beta-fluoromethyluridine.

Article abstract of DOI:10.1021/ol027364b

[reaction: see text] 2'-C-beta-Fluoromethyluridine (17) represents both a potentially important biological agent and a tool for biochemical analysis. Here we describe the first synthesis of this compound starting from uridine. The key steps include protec

Full text of DOI:10.1021/ol027364b

ORGANIC
LETTERS
2003
Vol. 5, No. 6
807-810
Synthesis of 2′-C-â-Fluoromethyluridine
Qing Dai† and Joseph A. Piccirilli*
Howard Hughes Medical Institute, Department of Biochemistry & Molecular Biology
and Department of Chemistry, The UniVersity of Chicago, 5841 S. Maryland AVe.,
MC 1028, Chicago, Illinois 60637
jpicciri@midway.uchicago.edu
Received November 27, 2002
ABSTRACT
2′-C-â-Fluoromethyluridine (17) represents both a potentially important biological agent and a tool for biochemical analysis. Here we describe
the first synthesis of this compound starting from uridine. The key steps include protection of the uracil base with methoxyethoxymethyl
(MEM) chloride, conversion to the corresponding 2′-C-r-epoxide, and regioselective opening of the oxirane ring with potassium fluoride/
hydrogen fluoride. Subsequent acetylation of the 3′- and 5′-hydroxyl groups enables MEM removal using B-bromocatecholborane. Deacetylation
generates the parent nucleoside, 2′-C-â-flurormethyluridine.
2′-Fluoromethylnucleosides represent important targets for
synthesis due to their potential value as clinically useful
medicinal agents and as biochemical probes. As medicinal
agents, 2′-C-â-methylnucleosides possess anticancer and
antiviral properties and function as inhibitors of enzymes.1
Additionally, fluorine substitution within a nucleoside may
enhance clinical efficacy by altering drug metabolism,
lipophilicity, and reactivity.2 As biochemical probes, 2′-C-
â-fluoromethyl nucleosides may provide important tools for
functional analysis of biologically significant RNA mol-
ecules. A series of ribonucloside analogues containing 2′-
C-â-methyl groups of increasing fluorine substitution (CH3,
CH2F, CHF2, or CF3) might allow systematic variation of
the pKa of the 2′-hydroxyl group over a broad range while
maintaining a similar structural context. Such a series of
nucleosides could unleash the power of physical organic
approaches to study the critical biological role played by the
2′-hydroxyl group of RNA. In previous work, we reported
the syntheses of 2′-C-â-methyl and 2′-C-â-trifluoromethyl
ribonucleosides.3 Here we describe the synthesis of the 2′-
C-â-monofluoromethyl nucleoside, 2′-C-â-fluoromethyl-uri-
dine.
Yoshimura et al. successfully prepared a derivative of
1-(2′-C-R-monofluoromethyl-â-D-arabinofuranosyl)uridine
from the corresponding 2′-â-spiroepoxy-uridine by regio-
selective opening of the oxirane ring with potassium fluoride/
hydrogen fluoride.4 We envisioned that the analogous
reaction with an appropriate 2′-R-spiroepoxy derivative could
† Present address: The Ben May Institute for Cancer Research, The
University of Chicago, 5841 S. Maryland Ave. MC 6027, Chicago, IL,
60637.
(1) (a) Walton, E.; Holly, F. W.; Boxer, G. E.; Nutt, R. F. J. Org. Chem.
1966, 31, 1163-9. (b) Walton, E.; Jenkins, S. R.; Nutt, R. F.; Holly, F. W.
J. Med. Chem. 1969, 12, 306-9. (c) Yoshimura, Y.; Satoh, H.; Sakata, S.;
Ashida, N.; Miyazaki, S.; Matsuda, A. Nucleosides Nucleotides 1995, 14,
427-429. (d) Cory, A. H.; Samano, V.; Robins, M. J.; Cory, J. G. Biochem.
Pharmacol. 1994, 47, 365-371. (e) Awano, H.; Shuto, S.; Baba, M.; Kira,
T.; Shigeta, S.; Matsuda, A. Bioorg. Med. Chem. Lett. 1994, 4, 367-370.
(f) Baker, C. H.; Banzon, J.; Bollinger, J. M.; Stubbe, J.; Samano, V.;
Robins, M. J.; Lippert, B.; Jarvi, E.; Resvick, R. J. Med. Chem. 1991, 34,
1879-84. (g) Ong, S. P.; McFarlan, S. C.; Hogenkamp, H. P. Biochemistry
1993, 32, 11397-404. (h) Van der Donk, W. A.; Yu, G.; Silva, D. J.; Stubbe,
J.; McCarthy, J. R.; Jarvi, E. T.; Matthews, D. P.; Resvick, R. J.; Wagner,
E. Biochemistry 1996, 35, 8381-91.
(2) (a) Wright, J. A.; Wilson, D. P.; Fox, J. J. J. Med. Chem. 1970, 13,
269-72. (b) Fanucchi, M. P.; Leyland-Jones, B.; Young, C. W.; Burchenal,
J. H.; Watanabe, K. A.; Fox, J. J. Cancer Treat. Rep. 1985, 69, 55-9. (c)
Bergstrom, D. E.; Shum, P. W. J. Org. Chem. 1988, 53, 3953-3958. (d)
Fox, J. J.; Watanabe, K. A.; Schinazi, R. F.; Lopez, C. Int. Congr. Ser.
1985, 667, 53-56.
(3) (a) Tang, X. Q.; Liao, X.; Piccirilli, J. A. J. Org. Chem. 1999, 64,
747-754. (b) Li, N.-S.; Tang, X.-Q.; Piccirilli, J. A. Org. Lett. 2001, 3,
1025-1028.
(4) Yoshimura, Y.; Saitoh, K.; Ashida, N.; Sakata, S.; Matsuda, A.
Bioorg. Med. Chem. Lett. 1994, 4, 721-724.
10.1021/ol027364b CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/14/2003

generate the 2′-fluoromethyl group on the â-face of the ribose
ring to give 1-(2′-C-â-monofluoromethyl-â-D-ribofuranosyl)-
uridine. Scheme 1 shows our strategy for the synthesis of
tion, we peracylated the crude product by treatment with
acetic anhydride in pyridine. Analytical characterization (1H
NMR, 13C NMR, MS, and elemental analysis) of this product
showed no fluoromethyl group and was consistent with two
possible products: 3′,5′-di-O-acetyl-2′-C-R-acetyloxymethyl-
2,2′-anhydro-1-(â-D-arabinofuranosyl)uracil 6a or 2′,3′,5′-
O-acetyl-2′-C-â-hydroxylmethyl-2,2′-anhydro-1-(â-D-ribo-
furanosyl)uracil 6b (Scheme 1). To distinguish between these
two possibilities, we carried out single-crystal X-ray dif-
fraction analysis. The structural data established the product
identity as 3′,5′-di-O-acetyl-2′-C-R-acetyloxymethyl-2,2′-
anhydro-1-(â-D-arabinofuranosyl)uracil 6a,8 suggesting that
under the reaction conditions, the 2-keto oxygen of the uracil
heterocycle attacks the R-epoxide at C-2′ (Scheme 2). The
Scheme 1a
Scheme 2
a Reaction conditions: (a) dichloro-di-tert-butylsilane, AgNO3;
(b) Dess-Martin periodinane, CH2Cl2, 24 h; (c) methyltriphenyl-
phosphonium bromide, potassium tert-pentoxide, ether, 48 h; (d)
MCPBA, CH2Cl2, 48 h; (e) KHF2, 2-methoxyethanol; (f) Ac2O,
pyridine.
formation of 6a supports our assignment of 5 as the
R-epoxide. Yoshimura et al. reported no such reaction in
the case of the â-epoxide,4 presumably because the cis
relationship between the epoxide and the nucleophile pro-
hibits backside attack at C-2′.
the 2′-R-spiro epoxide 5. Reaction of uridine with di-tert-
butylsilyl chloride and silver nitrate in DMF at room-
temperature protected the 3′- and 5′-hydroxyl groups as the
silyl ether 2 in 70% yield.5 Dess-Martin periodinane in
dichloromethane at room-temperature oxidized 2 in 88%
yield to the corresponding 2′-keto nucleoside 3,6 which was
dehydrated by azeotropic distillation with toluene before
further use. Treatment of 3 with methylenetriphenylphos-
phine, produced in situ from methyltriphenylphosphonium
bromide and potassium tert-pentoxide in ether, installed the
2′-methylene group to give 4.7 Epoxidation of the 2′-
methylene group with MCPBA in methylene chloride gener-
ated the diastereomerically pure R-spiro-epoxide 5 in 40%
yield. NOE experiments supported the stereochemical as-
signment at C-2′: irradiation at one of the oxirane protons
enhanced only the H-3′ resonance; irradiation at the other
oxirane proton weakly enhanced H-1′ only.
Protection of uracil at N-3 might eliminate the formation
of 6a and consequently enhance fluoride attack on the
epoxide. We tested the feasibility of four different uracil
protecting groups (Scheme 3): benzoyl (Bz), triphenyl-
methane-sulfenyl (TPMS),9 2-methoxyethoxyl-methyl
(MEM),10 and p-methoxybenzyl (PMB).11 The MEM deriva-
tive proved to be the most effective. Treatment of 7c with
MCPBA in dichloromethane for 2 days produced the
corresponding R-epoxide 8c in 30% yield (50% of the
starting material was also recovered). Longer reaction times
resulted in significantly lower yields. NOE experiments
analogous to those for 5 confirmed the stereochemistry at
C-2′.
Exposure of the R-epoxide 5 with KF/HF in 2-methoxy-
ethanol at 130 °C generated a highly polar product due to
removal of the di-tert-butylsilyl group. To facilitate purifica-
(8) Dai, Q.; Piccirilli, J. A.; Zheng, C.; Li, S. J. Z. Kristallogr. NCS
2001, 216, 343-346.
(9) Takaku, H.; Imai, K.; Nagai, M. Chem. Lett. 1988, 857-860.
(10) (a) Ito, T.; Ueda, S.; Takaku, H. J. Org. Chem. 1986, 51, 931-
933. (b) Takaku, H.; Ueda, S.; Ito, T. Tetrahedron Lett. 1983, 24, 5363-
5366.
(5) Furusawa, K.; Ueno, K.; Katsura, T. Chem. Lett. 1990, 97-100.
(6) Samano, V.; Robins, M. J. J. Org. Chem. 1990, 55, 5186-5188.
(7) Samano, V.; Robins, M. Synthesis 1991, 283.
(11) Aerschot, A. V.; Jie, L.; Akiyama, T.; Herdewijn, P.; Jumegawa,
M. Tetrahedron Lett. 1991, 32, 1905.
808
Org. Lett., Vol. 5, No. 6, 2003

Scheme 3a
Scheme 4a
a Reaction conditions: (a) Ac2O, pyridine; (b) B-bromo-
catecholborane, CH2Cl2, rt, 2 h; (c) 1:4 guanidine/guanidine
hydrochloride, 1:1 MeOH/CH2Cl2.
reaction conditions gave mainly hydrolysis products (cor-
responding to 11 and 12).
Reagents known to remove MEM from uracil (TrBF4 or
NH3 in pyridine, or the Lewis acids TiCl413 and ZnBr214 and
bromodimethylborane15) failed to remove MEM from 9c or
15a. However, B-bromocatecholborane16 in methylene chlo-
ride at room-temperature transformed 15a to 16 in 60% yield.
To prepare the parent nucleoside, 2′-C-â-fluoromethyluridine,
deacetylation of 16 must occur without the loss of fluoride
via epoxide formation. A mixture of guanidine and guanidine
hydrochloride (1:4) in methanol17 proved to be effective,
yielding 17 in 95% yield (Scheme 4).18
In summary, attempts to prepare 2′-C-â-monofluoro-
methyluridine via regioselective oxirane opening of 2′-C-
R-spiroepoxyuridine with fluoride fail because the 2-keto
oxygen attacks nucleophilically at C-2′ to give the corre-
sponding 2,2′-anhydrouridine derivative 6a. MEM protection
of the uracil moiety at N-3 prohibits this reaction such that
exposure of the spiroepoxide to KF/HF generates the MEM-
protected 2′-C-â-fluoromethyl-nucleoside 9c as the major
product. Additional products arise from nucleophilic attack
by solvent, water, or fluoride at either carbon of the oxirane.
Acetylation of the 3′- and 5′-hydroxyl groups of 9c followed
by treatment with B-bromocatecholborane promotes facile
removal of MEM group in high yield. Subsequent exposure
to guanidine/guanidine hydrochloride removes the acetyl
a Reaction conditions: (a) diisopropylethylamine, CH2Cl2, ben-
zoyl chloride (or MEM-Cl or triphenylmethane sulfenyl chloride);
(b) CH2Cl2, MCPBA.
Subsequent treatment of the epoxide 8c with KF/HF in
2-methoxyethanol at 130 °C for 8 h provided 9c as the major
product (35%) together with several byproducts (10-14;
Scheme 3). NMR characterization of 9c confirmed the
presence of the fluoromethyl group -CH2F (2JH-F ) 48.0
Hz, 1JC-F ) 175.3 Hz; 19F NMR δ ) -229 ppm, CFCl3 as
a reference).4,12 To avoid spectral overlap, NOE experiments
were conducted on 15b, the peracetylated form of 9c
(Scheme 4). Irradiation of the fluorine enhanced the reso-
nance at H-3′ strongly but only enhanced the resonance of
H-1′ weakly. These results established that the -CH2F group
of 15b (and by inference 15a and 9c) resides on the â-face
of the ribose ring.
The formation of major product 9c indicated that fluoride
ion attacked predominantly at the less-hindered carbon of
the epoxide, though some attack at the more hindered C-2′
occurred to give a small amount of 10. The byproducts 11-
13 arise as a consequence of hydrolysis and solvolysis,
respectively. Byproduct 14 arises from initial desilylation
of the starting material 8c but remains only in trace amounts
after the reaction. During workup, however, 14 increased at
the expense of 9c until removal of excess KF/HF. Presum-
ably, the 2′-oxygen nucleophilically attacks the fluoromethyl
group to displace fluoride and regenerate the epoxide. In
contrast to the results for 8c, exposure of the PMB-protected
epoxide 8d to KF/HF in 2-methoxyethanol under the same
(12) Muehlbacher, M.; Poulter, C. D. J. Org. Chem. 1988, 53, 1026-
1030.
(13) Corey, E. J.; Gras, J.-L.; Ulrich, P. Tetrahedron Lett. 1976, 17, 809-
812.
(14) Miyata, O.; Shinada, T.; Ninomiya, I.; Naito, T. Tetrahedron Lett.
1991, 32, 3519-3522.
(15) Guindon, Y.; Yoakim, C.; Morton, H. E. J. Org. Chem. 1984, 49,
3912-20.
(16) Boeckman, R. K.; Potenza, J. C. Tetrahedron Lett. 1985, 26, 1411.
(17) Ellervik, U.; Magnusson, G. Tetrahedron Lett. 1997, 38, 1627-
1628.
(18) Starting from uridine, our nine-step synthesis provided 17 with an
overall yield of ∼1.4%.
Org. Lett., Vol. 5, No. 6, 2003
809

groups without loss of fluoride to afford 2′-C-â-fluoro-
methyluridine. This nucleoside may represent a new class
of drugs and/or tools for chemical biology.
Supporting Information Available: Full experimental
and analytical data for compounds 2-5, 6b, 7c, 8c, 9c, 10-
14, 15a,b, 16, and 17; 19F-1H NOE spectra of 15b; 1H and
13C NMR spectra of 9c, 15b, 16, and 17; and further
information on MEM resistance to deprotection. This mate-
rial is available free of charge via the Internet at http://
pubs.acs.org.
Acknowledgment. Q.D. was a Research Associate and
J.A.P. is an Associate Investigator of the Howard Hughes
Medical Institute. We thank Dr. X. Tang, C. Cortez, L. Elias,
A. Jurkiewicz, and J. Ye for technical support and Dr. N.
Li, J. Hougland, S. R. Das, and J. Schwans for helpful
discussion and critical comments on the manuscripts.
OL027364B
810
Org. Lett., Vol. 5, No. 6, 2003