Improved synthesis of 9-(2,3-dideoxy-2-fluoro-β-D-threo-pentofuranosyl)adenine (FddA) using triethylamine trihydrofluoride

Article abstract of DOI:10.1016/S0040-4039(01)00135-6

An improved synthesis of 9-(2,3-dideoxy-2-fluoro-β-D-threo-pentofuranosyl)adenine (1, FddA) via a fluorination of 3′-O-benzoyl-5′-O-tritylriboside (4a) using noncorrosive triethylamine trihydrofluoride (Et₃N·3HF) is described. The method is suitable for large-scale synthesis. In particular, the synthesis of the pivotal intermediate 4a was much improved in avoidance of the use of toxic tin reagent. Radical deoxygenation with several silanes was also studied. The total yield of FddA from 6-chloropurine riboside (2) in this study was greater than that we reported previously.

Full text of DOI:10.1016/S0040-4039(01)00135-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2321–2324
Improved synthesis of
9-(2,3-dideoxy-2-fluoro-b- -threo-pentofuranosyl)adenine (FddA)
D
using triethylamine trihydrofluoride
Satoshi Takamatsu,a Tokumi Maruyama,b Satoshi Katayama,a Naoko Hirosea and
Kunisuke Izawaa,*
aAminoScience Laboratories, Ajinomoto Co., Inc., 1-1, Suzuki-cho, Kawasaki-ku, Kawasaki, Kanagawa 210-8681, Japan
bFaculty of Pharmaceutical Sciences, Tokushima Bunri University, Yamashiro-cho, Tokushima 770-8514, Japan
Received 22 January 2001; accepted 24 January 2001
Abstract—An improved synthesis of 9-(2,3-dideoxy-2-fluoro-b-D-threo-pentofuranosyl)adenine (1, FddA) via a fluorination of
3%-O-benzoyl-5%-O-tritylriboside (4a) using noncorrosive triethylamine trihydrofluoride (Et3N·3HF) is described. The method is
suitable for large-scale synthesis. In particular, the synthesis of the pivotal intermediate 4a was much improved in avoidance of
the use of toxic tin reagent. Radical deoxygenation with several silanes was also studied. The total yield of FddA from
6-chloropurine riboside (2) in this study was greater than that we reported previously. © 2001 Elsevier Science Ltd. All rights
reserved.
9-(2,3-Dideoxy-2-fluoro-b-
D
-threo-pentofuranosyl)ade-
desired fluorinated compound.1a Recently, we reported
the fluorination of 3%-O-benzoyl-5%-O-tritylriboside (4a)
with diethylaminosulfur trifluoride (DAST).7 The intro-
duction of an electron-withdrawing 6-chloro group to
the purine moiety was essential for improving the yield
of fluorination, probably because the 6-chloro group
decreases the nucleophilicity of purine base and pre-
vents a purine base migration during the fluorination.
Although the reaction yield was excellent and neither
migration nor elimination occurred, it would be prefer-
able to avoid the use of corrosive DAST in large-scale
industrial production.
nine (1, FddA, lodenosine) is a potent anti-HIV agent,1
which has significant advantages over non-fluorinated
dideoxynucleoside analogs such as dideoxyinosine (ddI,
didanosine). One of its most promising advantages is
that 1 has been shown to be effective against otherwise
resistant HIV strains.2 Compound 1 is currently being
tested in clinical trials for the treatment of AIDS.1d,3
FddA has been synthesized by several different meth-
ods.1d,4 A problem in each of these methods is how to
introduce a fluorine at C2%-b of the sugar moiety. Most
attempts to fluorinate the C2%-b position of nucleoside
derivatives lead to a migration or elimination product5
and give a poor yield,6 or sometimes even none of the
Chou et al. reported8a the large-scale synthesis of C-2
(arabino) fluorinated sugar derivatives with noncorro-
sive triethylamine trihydrofluoride (Et3N·3HF).8b,c
Since the condensation of fluorinated sugar with a
nucleic base still presents the problem of the undesired
a-anomer formation,1d,9 we became interested in using
Et3N·3HF for fluorination of nucleoside 4a.10
NH2
N
N
N
N
HO
O
In our previous report,7 one of the major problems for
obtaining the pivotal intermediate 4a from 6-chloro-
purine riboside (2)11 was the use of highly toxic
dibutyltin oxide during the formation of 3%-O-benzoate.
Decrease of the reaction yield due to acyl migration
between the 2%- and 3%-O-benzoyl groups followed by
subsequent dibenzoylation12 was another important
F
FddA
1
* Corresponding author. Tel.: +81/44/245.5066; fax: +81/44/211.7610;
e-mail: kunisuke izawa@ajinomoto.com
–
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00135-6

2322
S. Takamatsu et al. / Tetrahedron Letters 42 (2001) 2321–2324
Cl
N
Cl
N
Cl
N
Cl
N
N
N
N
N
N
N
N
N
N
N
N
N
TrCl / iPr2NH
BzCl / pyridine
toluene
TrO
TrO
HO
TrO
+
O
O
O
O
DMF
BzO OH
HO OBz
HO OH
HO OH
79%
2
3
4a (70%)
4b
Scheme 1.
problem. In order to improve the process, the selective
tritylation at the 5%-hydroxyl group was performed
prior to benzoylation. When we used pyridine as a base
for the benzoylation (Scheme 1), undesired dibenzoyla-
tion was remarkably suppressed and 4a was obtained in
a moderate yield without using the toxic tin reagent.
Compound 4a was fluorinated with Et3N·3HF in two
steps, as shown in Scheme 2. Prior to the reaction with
Et3N·3HF, the 2%-hydroxyl group of 4a was converted
to imidazolesulfonate (5a) or trifluoromethanesulfonate
(5b). We tried to fluorinate the compound having other
leaving group, but without success. It was revealed that
the 6-chloro group was effective again in preventing a
purine base migration and improving the yield of
fluorination.
Thus, compound 2 was mono-tritylated with 2 equiv. of
trityl chloride (TrCl) in the presence of diisopropy-
lamine (iPr2NH) in N,N-dimethylformamide (DMF) at
40°C. Small amounts of the 2%,5%- and 3%,5%-di-O-trity-
lated compounds were also formed. The resulting solu-
tion was subjected to conventional work-up and the
tritylated compound (3) was extracted by toluene.
Without further purification, compound 3 was benzoy-
lated with an excess amount of benzoyl chloride (BzCl)
in the presence of pyridine at 0°C. After conventional
work-up and purification by silica gel chromatography,
the mixture of 4a and 4b was subjected to crystalliza-
tion to afford pure 4a as white crystals in 70% yield.
Compound 4b remained in the mother liquor, and was
recycled, after the acyl migration from 4b to 4a, by
treating with triethylamine (Et3N).
Compound 5a was reacted with Et3N·3HF in several
solvents at 50°C or 70°C (Table 1). High-performance
liquid chromatography (HPLC) analysis showed that
the yield of fluorinated compound 6 varied from 54 to
78% including imidazolesulfonation.13 The best result
was obtained when toluene was used as the solvent. It
has been reported14 that the addition of 0.5 equiv. of
Et3N to Et3N·3HF enhances the nucleophilicity of the
reagent. We also studied the addition of Et3N (run 2),
but the yield of product decreased.
The trifluoromethanesulfonate (5b), which was quanti-
tatively obtained from 4a, was also reacted with 6
Cl
Cl
Cl
NH2
N
N
N
N
N
N
N
N
N
N
N
N
1) SO2Cl2 / pyridine
CH2Cl2
2) imidazole
N
N
N
N
Et3N•3HF
/ MeOH
NH3
TrO
TrO
TrO
TrO
O
O
O
F
O
or
toluene
F
78% through 5a
88% through 5b
CF3SO2Cl / DMAP
toluene
98%
BzO OH
BzO OR1
BzO
HO
4a
5a; R1=SO2-Imd
5b; R1=SO2CF3
6
7
Scheme 2.
Table 1. Fluorination of imidazolesulfonate (5a)
Run
Solvent
Et3N·3HF (equiv.)
Et3N (equiv.)
Temp. (°C)
Time (h)
Yield (%)a
1
2
3
4
5
AcOEt
AcOEt
Toluene
CH3CN
CH2Cl2
6.0
6.0
6.0
6.0
6.0
–
3.0
–
–
–
70
70
70
70
50
24
24
24
24
24
67
54
78
62
60
a The yield includes imidazolesulfonation of 4a.

S. Takamatsu et al. / Tetrahedron Letters 42 (2001) 2321–2324
2323
NH2
N
NH2
N
NH2
NH2
N
N
N
N
N
N
N
N
N
N
N
N
ClC(S)(OR)
DMAP
N
N
Silane
AIBN
80% AcOH
TrO
HO
TrO
TrO
O
O
O
O
CH3CN
F
F
F
F
85%
92% (R=Ph)
HO
O
OR2
64% (R=4-F-Ph)
8a; R2=Ph
8b; R2=4-F-Ph
7
9
1
S
Scheme 3.
Table 2. Deoxygenation of 8a and 8b with silanes
Acknowledgements
Run R2
Silane
Solvent
Yield (%)
The authors thank Mr. Hideo Takeda and Mr. Mit-
suhiko Kojima of Ajinomoto Co., Inc. for their help in
the research.
1
2
3
4
5
6
Ph
Ph
Ph
Ph
4-F-Ph
4-F-Ph
[(Me)3Si]3SiH
Ph2SiH2
Ph2SiH2
Ph2MeSiH
Et3SiH
Et3SiH
Toluene
Dioxane
Toluene
Dioxane
Dioxane
Acetone
78
81
74
69
55
56
References
1. (a) Marquez, V. E.; Tseng, C. K.-H.; Kelley, J. A.;
Mitsuya, H.; Broder, S.; Roth, J. S.; Driscoll, J. S.
Biochem. Pharmacol. 1987, 36, 2719–2722; (b) Marquez,
V. E.; Tseng, C. K.-H.; Mitsuya, H.; Aoki, S.; Kelley, J.
A.; Ford, Jr., H.; Roth, J. S.; Broder, S.; Johns, D. G.;
Driscoll, J. S. J. Med. Chem. 1990, 33, 978–985; (c)
Ruxrungtham, K.; Boone, E.; Ford, Jr., H.; Driscoll, J.
S.; Davey, Jr., R. T.; Lane, H. C. Antimicrob. Agents
Chemother. 1996, 40, 2369–2374 and references cited
therein; (d) Graul, A.; Silvestre, J. Drugs Future 1998, 23,
1176–1189 and references cited therein.
2. (a) Driscoll, J. S.; Mayers, D. L.; Bader, J. P.; Weislow,
O. S.; Johns, D. G.; Buckheit, Jr., R. W. Antivir. Chem.
Chemother. 1997, 8, 107–111; (b) Tanaka, M.; Srinivas,
R. V.; Ueno, T.; Kavlick, M. F.; Hui, F. K.; Fridland,
A.; Driscoll, J. S.; Mitsuya, H. Antimicrob. Agents
Chemother. 1997, 41, 1313–1318.
3. During the preparation of this manuscript, a clinical trial
of FddA was suspended. US Bioscience, Inc.: West Con-
shohocken, PA, Press Release, Oct. 14, 1999.
4. (a) Maqbool, A. S.; Driscoll, J. S.; Abushanab, E.; Kel-
ley, J. A.; Barchi, Jr., J. J.; Marquez, V. E. Nucleosides
Nucleotides Nucleic Acids 2000, 19, 1–12; (b) Maqbool,
A. S.; Driscoll, J. S.; Marquez, V. E. Tetrahedron Lett.
1998, 39, 1657–1660.
5. (a) Pankiewicz, K. W.; Krzeminski, J.; Ciszewski, L. A.;
Ren, W.-Y.; Watanabe, K. A. J. Org. Chem. 1992, 57,
553–559; (b) Krzeminski, J.; Barbara, N.; Pankiewicz, K.
W.; Watanabe, K. A. Nucleosides Nucleotides 1991, 10,
781–798; (c) Pankiewicz, K. W.; Watanabe, K. A. J.
Fluorine Chem. 1993, 64, 15–36; (d) Pankiewicz, K. W.;
Watanabe, K. A. In Nucleosides and Nucleotides as Anti-
tumor and Antiviral Agents; Chu, C. K.; Baker, D. C.,
Eds.; Plenum Press: New York, 1993; pp. 55–71.
6. (a) Herdewijn, P.; Pauwels, R.; Baba, M.; Balzarini, J.;
De Clercq, E. J. Med. Chem. 1987, 30, 2131–2137; (b)
Shiragami, H.; Tanaka, Y.; Uchida, Y.; Iwagami, H.;
Izawa, K.; Yukawa, T. Nucleosides Nucleotides 1992, 11,
391–400.
equiv. of Et3N·3HF and 3 equiv. of Et3N in ethyl
acetate (AcOEt) at reflux temperature overnight.15
HPLC analysis indicated that the yield of fluorinated
compound 6 was 88%. After conventional work-up, a
toluene solution of 6 was added to ammonia in
methanol and kept in a sealed tube at 100°C for 18
hours. HPLC analysis revealed that the 3%-O-depro-
tected and 6-aminated compound 7 was formed in 98%
yield. After purification by silica gel chromatography,
pure 7 was obtained as white crystals.
Deoxygenation of the 3%-hydroxyl group in 7 was
achieved by the Barton–McCombie method with
silane16 to avoid the use of toxic tin hydride (Scheme 3).
Recently, the same radical reduction of b-fluorinated
substrates using dilauroyl peroxide has been reported.4a
Compound 7 was converted with chlorothionoformate
to 3%-O-thiocarbonyl derivatives 8a and 8b in good
yield. The products were then subjected to deoxygena-
tion with various silanes in the presence of 2,2%-azobi-
sisobutyronitrile (AIBN). The results are shown in
Table 2. Diphenylsilane gave the best results among
various silanes. Finally, separation and purification of
9, followed by acidic hydrolysis of the 5%-O-trityl group,
gave FddA (1) in good yield. Its spectroscopic proper-
ties were identical in all respects to the published data.1
In conclusion, FddA (1) was synthesized in good yield
by fluorination with
a noncorrosive fluorinating
reagent, Et3N·3HF, starting from a readily available
nucleoside. The method proceeds without using toxic
tin reagent for selective benzoylation and deoxygena-
tion. Accordingly, the method is suitable for large-scale
synthesis and is environmentally benign. The total yield
of FddA (1) from 6-chloropurine riboside (2) in this
study (30.2%) was greater than that we reported previ-
ously (13.5%).

2324
S. Takamatsu et al. / Tetrahedron Letters 42 (2001) 2321–2324
7. Maruyama, T.; Takamatsu, S.; Kozai, S.; Satoh, Y.;
overnight at 70°C. After cooling, ethyl acetate and an
aqueous saturated solution of sodium hydrogencarbonate
were added for phase separation. The organic layer was
washed with ethyl acetate. The aqueous layers were com-
bined, dried with sodium sulfate, filtered, and then dis-
tilled. The residue was dissolved in acetonitrile and
analyzed by liquid chromatography, which revealed that
6 was obtained in 81% yield. The analytical sample was
purified by silica gel chromatography. Analytical data for
compound 6: 1H NMR (CDCl3, 300 MHz): l 8.76 (1H, s,
H-2), 8.36 (1H, d, J=3.0 Hz, H-8), 8.07 (2H, d, J=7.2
Hz, C6H5CO), 7.66 (1H, t, J=7.3 Hz, C6H5CO), 7.23–
7.54 (ca. 17H, m, C6H5CO, trityl), 6.66 (1H, dd, J=21.7,
2.8 Hz, H-1%), 5.69 (1H, dd, J=16.9, 3.2 Hz, H-3%), 5.28
(1H, dd, J=50.0, 2.8 Hz, H-2%), 4.41 (1H, m, H-4%), 3.63
(1H, dd, J=10.3, 5.9 Hz, H-5%a), 3.54 (1H, dd, J=10.3,
4.2 Hz, H-5%b); 13C NMR (DMSO-d6, 75 MHz): l 164.9,
152.2, 151.5, 149.8, 146.1 (d, J=5.1 Hz), 143.6, 134.2,
129.8, 129.0, 128.9, 128.4, 128.1, 127.3, 93.4 (d, J=193.2
Hz), 86.6, 82.7 (d, J=16.9 Hz), 79.6, 76.5 (d, J=27.3
Hz), 63.4; IR (KBr): w 3061, 1732, 1592, 1567, 1492, 1450,
Izawa, K. Chem. Pharm. Bull. 1999, 47, 966–970.
8. (a) Chou, T. S.; Becke, L. M.; O’Toole, J. C.; Carr, M.
A.; Parker, B. E. Tetrahedron Lett. 1996, 37, 17–20; (b)
McClinton, M. A. Aldrichim. Acta 1995, 28, 31–35; (c)
Haufe, G. J. Prakt. Chem. 1996, 338, 99–113.
9. Barchi, Jr., J. J.; Marquez, V. E.; Driscoll, J. S.; Ford,
Jr., H.; Mitsuya, H.; Shirasaka, T.; Aoki, S.; Kelley, J. A.
J. Med. Chem. 1991, 34, 1647–1655.
10. The only known use of Et3N·3HF in the synthesis of
nucleoside derivatives is in the synthesis of 3%-b fluori-
nated thymidine: von Janta-Lipinski, M.; Langen, P.;
Czech, D. Z. Chem. 1983, 23, 325.
11. Brown, G. B.; Weliky, V. S. J. Biol. Chem. 1953, 204,
1019–1024.
12. Kozai, S.; Takamatsu, S.; Izawa, K.; Maruyama, T.
Tetrahedron Lett. 1999, 40, 4355–4358.
13. A representative experimental procedure is as follows: 4a
(2.0 g, 3.1 mmol) was dissolved in dichloromethane (15.5
ml) and cooled to −35°C. To this solution was added
pyridine (0.45 ml, 5.57 mmol) and sulfuryl chloride (0.30
ml, 3.72 mmol), and the mixture was stirred for 30
minutes. Imidazole (0.96 g, 13.9 mmol) was then added,
and the resulting mixture was allowed to return to room
temperature and stirred overnight. Water (10 ml) was
then added for phase separation. The aqueous layer was
washed with 10 ml of dichloromethane. The organic
layers were combined, dried with sodium sulfate, filtered,
and distilled. The residue was purified by silica gel chro-
matography to obtain the desired compound (oily). The
yield was 2.28 g (FW: 763.22, 2.99 mmol, 96%). This
imidazolesulfonate (377 mg, 0.494 mmol) was dissolved
in toluene (2.4 ml), triethylamine trihydrofluoride (0.49
ml, 3.00 mmol) was added, and the mixture was stirred
1265, 1218, 1092 cm−1
; HRMS (FAB+) calcd for
C36H29ClFN4O4 (M+H)+: 635.1861; found: 635.1846.
Anal. calcd for C36H29ClFN4O4: C, 68.08; H, 4.44; Cl,
5.58; F, 2.99; N, 8.82. Found: C, 68.11; H, 4.54; Cl, 5.82;
F, 2.65; N, 8.69.
14. Giudicelli, M. B.; Picq, D.; Veyron, B. Tetrahedron Lett.
1990, 31, 6527–6530.
15. Since Et3N·3HF is noncorrosive, the glassware did not
show any indications of corrosion even after several days
under these reaction conditions. See also Ref. 8.
16. Chatgilialoglu, C.; Ferreri, C. Res. Chem. Intermed. 1993,
19, 755–775.
.
.