A concise synthesis of methyl 2,6-dideoxy-2-fluoro-β-L-talopyranoside

Article abstract of DOI:10.1016/S0008-6215(98)00325-5

The 4,6-benzylidene acetal of methyl 2-deoxy-2-fluoro-α,β-D-glucopyranoside underwent inversion at C-3 via an oxidation-reduction sequence, and treatment of the derived 3-acetate with N-bromosuccinimide in carbon tetrachloride gave methyl 3-O-acetyl-4-O-b

Full text of DOI:10.1016/S0008-6215(98)00325-5

Carbohydrate Research 315 (1999) 187–191
Note
A concise synthesis of methyl
2,6-dideoxy-2-fluoro-b-
L
-talopyranosideꢀ
S. Todd Deal 1, Derek Horton *
Department of Chemistry, The Ohio State Uni6ersity, Columbus, OH 43210, USA
Received 15 October 1998; accepted 7 December 1998
Abstract
The 4,6-benzylidene acetal of methyl 2-deoxy-2-fluoro-a,b-
oxidation–reduction sequence, and treatment of the derived 3-acetate with N-bromosuccinimide in carbon tetrachlo-
-allopyranoside (6). Dehydrobromination
D-glucopyranoside underwent inversion at C-3 via an
ride gave methyl 3-O-acetyl-4-O-benzoyl-6-bromo-2,6-dideoxy-2-fluoro-a-
D
of 6 and reduction of the resultant 5,6-ene gave the 5-epimer of 6, which after removal of the ester substituents,
afforded the title compound in good overall yield. © 1999 Elsevier Science Ltd. All rights reserved.
Keywords: Fluoro sugars; 2,6-Dideoxy-2-fluoro-
L
-talose; Configurational inversion; Anthracycline glycons
construct the required 2,6-dideoxy-2-fluoro- -
L
1. Introduction
talopyranose precursor sugar.
Here we present a concise synthesis of this
Semisynthetic
anthracycline
glycosides
fluoro sugar, as its methyl b-glycoside 9, utiliz-
based on the 3%-deamino-3%-hydroxydoxoru-
bicin structure [2] and having an axial halogen
atom at the C-2% position have shown [3]
promise as antitumor agents. The 2%-iodo
structures are readily generated [3] by alkoxy-
halogenation from appropriate glycal precur-
sors, and the 2%-bromo and 2%-chloro analogs
can be conveniently produced [4] via halo-
genation of glycals and subsequent glycosidic
coupling. The corresponding 2%-fluoro struc-
tures are not similarly accessible from glycals,
and alternative procedures are necessary [5] to
ing as precursor 2-deoxy-2-fluoro-a- -glu-
D
copyranose tetraacetate, itself accessible [6] in
two steps from -mannose. The sequence is
D
closely modeled after the methodology suc-
cessfully utilized in this laboratory [7] for syn-
thesis of daunosamine from
D
-mannose, and
affords 9 in 27% net yield on all steps.
2. Results and discussion
The synthesis begins from 1,3,4,6-tetra-
O-acetyl-2-deoxy-2-fluoro-b-
which was prepared by treating b-
ranose-1,3,4,6-tetraacetate with diethylamino-
sulfur trifluoride by the method of Kova´c [6].
Glycosidation and deacetylation of this com-
pound was accomplished in a single step with
D
-glucopyranose,
ꢀ For a preliminary report see Ref [1].
* Corresponding author. Present address: Department of
Chemistry, The American University, 4400 Massachusetts
Avenue NW, Washington, DC 20016, USA.
E-mail address: carbchm@american.edu (D. Horton)
1 Present address: Department of Chemistry, Georgia
Southern University, Statesboro, GA 30460-8064, USA.
D
-mannopy-
'
0008-6215/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(98)00325-5

188
S.T. Deal, D. Horton / Carbohydrate Research 315 (1999) 187–191
Dowex-50W (H+) resin in refluxing
methanol. The products of the reaction
proved to be an inseparable mixture (1) of the
methyl a- and b-glycosides, with an anomeric
composition of 58% a and 42% b as deter-
freshly prepared according to the procedure of
Corey and Suggs [10], and the molecular
˚
sieves (3 A) were dried overnight at 160 °C in
vacuo. The product was only slightly soluble
in dichloromethane, the reaction solvent, and
it precipitated out as a fluffy white solid upon
cooling or partial evaporation of the solvent.
Again, the anomers could be separated by
careful chromatography, but the product was
used without further purification in the next
step. Reduction of the resulting ketone 3 was
accomplished with sodium borohydride in
ethanol. Because of the large difference in Rf
values (DRf =0.1) of the reduction products of
the a and b anomers, the desired a product 4
was easily separated from the product mix-
ture. In a separate experiment using the pure a
anomer of 3, the reduction was shown to be
stereospecific, that is, the reaction product was
1
mined by H NMR spectroscopy. Benzylide-
nation of this anomeric mixture was carried
out using a modification of the procedure of
Evans [8]. The anomers could be separated at
this point by careful chromatography; how-
ever, in view of the difficulty involved in the
chromatography and the resulting decrease in
yield, the anomeric mixture 2 was used in the
next step without further purification (Scheme
1).
Inversion of the stereochemistry at C-3 of 2
was accomplished via a stereospecific oxida-
tion–reduction procedure. The oxidation was
carried out using the methodology developed
by Herscovici and Antonakis [9] for the
molecular sieve-assisted oxidation of carbohy-
drates. Pyridinium dichromate (PDC) was
the desired
D
-allo isomer with no trace of the
D
-gluco isomer being detected. This fact was
1
established by comparing the H NMR spec-
Scheme 1.

S.T. Deal, D. Horton / Carbohydrate Research 315 (1999) 187–191
189
tra of 2 and 4, which show J2,3=8.9 Hz for 2,
as would be expected for the trans-diaxial
mended [13] with Silica Gel 60 (230–400
mesh, E. Merck). Melting points were deter-
mined with a Thomas–Hoover apparatus and
are uncorrected. Optical rotations were mea-
sured with a Perkin–Elmer 141 polarimeter at
relationship of these protons in a
D
-glucose
derivative and J2,3 =3.5 Hz for 4, as would be
expected for the axial–equatorial relationship
1
13
of these protons in a
D
-allose derivative.
ꢀ25 °C. H and C NMR spectra at 500 and
125 MHz, respectively, were recorded by Dr
C.E. Cottrell with a Bruker AM 500 instru-
Compound 4 was acetylated at OH-3 using
acetic anhydride and pyridine to give 5 in 91%
yield. Opening of the benzylidene acetal ring
of 5 was performed with N-bromosuccinimide
(NBS) using standard Hanessian–Hullar [11]
conditions for this type of reaction, to give 6
as colorless needles from ethanol. The dehy-
drobromination of 6 was accomplished with
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in
refluxing benzene to give the 5,6-unsaturated
derivative 7 as a low-melting solid that was
unstable at room temperature, but which
could be stored at −20 °C for short periods.
Inversion of the stereocenter at C-5 in 6 was
completed by hydrogenation of the alkene
product 7 in dry ethyl acetate over 10% palla-
dium–barium sulfate. Previous studies in this
laboratory [7] indicated that hydrogenation of
a 5,6-unsaturated sugar derivative with a
stereochemistry comparable with 7 would lead
to stereospecific addition of hydrogen from
19
ment (unless otherwise noted). The F spectra
at 235 MHz were recorded by C. Engelman
with a Bruker AM 250 instrument. Chemical
shifts are in ppm from Me4Si, measured from
the CHCl3 lock signal at l 7.26. Mass spectra
were obtained with a VG 70-250S mass spec-
trometer in the FAB mode in a ‘magic bullet’
matrix and were recorded by D. Chang. Ele-
mental analyses were performed by Atlantic
Microlab, Inc., Norcross, GA.
Methyl 2-deoxy-2-fluoro-h,i-
oside (1).—To a solution of 1,3,4,6-tetra-O-
acetyl-2-deoxy-2-fluoro-b- -glucopyranose [6]
D
-glucopyran-
D
(4.28 g, 12.2 mmol) in 60 mL of dry MeOH,
Dowex-50W (H+) resin (12.8 g), which had
been exhaustively washed with dry MeOH and
dried overnight in vacuo at 50 °C, was added.
The resulting mixture was boiled under reflux
for 48 h at 75–78 °C, at which time TLC (4:1
CHCl3–MeOH) revealed a single spot, Rf
0.57. The mixture was diluted with CHCl3
(240 mL) and filtered through a sintered glass
funnel containing a small pad of silica gel with
10% MgSO4. Removal of the solvent gave a
colorless oil, which by NMR spectroscopy,
was a mixture of anomers (58% a, 42% b);
yield 2.27 g (95%). This oil was used in the
next step without further purification. An ana-
lytical sample of the anomeric mixture was
obtained by flash chromatography with 4:1
CHCl3–MeOH; [h]D +76.6° (c 1.1, H2O);
MS: m/z 197 (M+1), 165 (M+1−MeOH),
177 (M+1−HF); 1H NMR (500 MHz,
Me2SO-d6): l 4.83 (d, J1,2=3.8 Hz, H-1a),
4.41 (dd, J1,2 =7.7 Hz, J1,F=2.2 Hz, H-1b).
Anal. Calcd for C7H13FO5 (196.19): C, 42.85;
H, 6.69. Found: C, 42.70; H, 6.70.
above the sugar ring to yield the desired -talo
L
1
product 8. This was confirmed by H NMR
spectroscopy, which demonstrated a J1,F value
of 18.5 Hz characteristic of a trans-diaxial
relationship between fluorine and the C-1 pro-
1
ton, and indicative of the C4 conformation of
this compound, which is to be expected for
this
L
sugar derivative. The synthesis was
completed by deacylation of 8 with methano-
lic sodium methoxide to yield 9 as a crystalline
solid, in 27% overall yield from 2-deoxy-2-
fluoro-b- -glucopyranose tetraacetate.
D
3. Experimental
General methods.—Solvents were purified
and dried as recommended [12] and were
evaporated under diminished pressure below
50 °C. TLC was performed on precoated glass
plates (0.25 mm) of Silica Gel 60 F254 (E.
Merck); detection was performed by spraying
with H2SO4 and subsequent heating. Flash
chromatography was performed as recom-
Methyl 4,6-O-benzylidene-2-deoxy-2-fluoro-
h,i-glucopyranoside (2).—To a suspension of
1 (2.27 g, 11.6 mmol) in dry MeCN (50 mL),
a,a-dimethoxytoluene (2.78 mL, 18.5 mmol)
and p-toluenesulfonic acid (26 mg) were
added. After 2 h, TLC (1:1 hexane–EtOAc)

190
S.T. Deal, D. Horton / Carbohydrate Research 315 (1999) 187–191
resulting solution was washed with satd aq
NaHCO3 and brine. The organic layer was
dried (MgSO4), and the solvent was evapo-
rated to yield an oil, which was adsorbed onto
a small amount of silica gel and purified by
flash chromatography with 3:2 hexane–
EtOAc to furnish first a solid (0.38 g, 42%)
that was shown by NMR spectroscopy to be a
2:1 mixture of the b anomer of 4 and its
3-epimer. Continued elution of the column
gave 4 (0.52 g, 57%) as a white, crystalline
solid; mp 109–110 °C, [h]D +126° (c 1,
CHCl3); MS: m/z 285 (M+1), 253 (M+1−
MeOH), 179 (M+1−PhCHO). Anal. Calcd
for C14H17FO5 (284.3): C, 59.14; H, 6.04.
Found: C, 59.22; H, 6.08.
indicated only a trace of 1 remaining. The
solvent was removed at 40 °C to yield a
slightly opaque oil that showed no 1 by TLC.
The oil was dissolved in boiling ether, and the
resulting solution was cooled to −20 °C. This
gave a white solid, which by TLC was an
anomeric mixture. The total yield of product
after processing of mother liquors was 2.96 g
(90%). This solid was used in the next step
without further purification. An analytical
sample of the a anomer was obtained by flash
chromatography with 2:1 hexane–EtOAc; mp
155–156 °C, [h]D +106° (c 1, CHCl3), (lit.
[14] mp 163–164 °C, [h]D +116.3°); MS: m/z
285 (M+1), 253 (M+1−MeOH), 207 (M+
1−Ph), 179 (M+1−PhCHO). Anal. Calcd
for C14H17FO5 (284.30): C, 59.14; H, 6.04.
Found: C, 59.06; H, 6.06.
Methyl 3-O-acetyl-4,6-O-benzylidene-2-de-
oxy-2-fluoro-h- -allopyranoside (5).—To a
D
Methyl 4,6-O-benzylidene-2-deoxy-2-fluoro-
solution of 4 (0.52 g, 1.83 mmol) in dry
pyridine (20 mL) at 0 °C, Ac2O (10 mL) and a
catalytic amount of 4-dimethylaminopyridine
were added and the solution was left overnight
to warm to room temperature. Removal of the
solvent using toluene for azeotropic removal
of pyridine gave a yellow oil that was filtered
through a short column of silica gel with 2:1
hexane–EtOAc to yield 5 (0.544 g, 91%) as a
crystalline solid after drying in vacuo at 50 °C;
mp 79–80 °C, [h]D +91° (c 1.4, CHCl3); MS:
m/z 327 (M+1), 295 (M+1−MeOH), 267
(M+1−HOAc), 221 (M+1−PhCHO).
Anal. Calcd for C16H19FO6 (326.34): C, 58.88;
H, 5.88. Found: C, 58.94; H, 5.91.
h,i- -ribo-hexopyranosid-3-ulose (3).—To a
D
solution of 2 (1.16 g, 4.08 mmol) in dry
CH2Cl2 (50 mL), pyridinium dichromate (2.30
˚
g, 6.12 mmol) and crushed 3 A molecular
sieves (4.0 g, dried overnight at 160 °C in
vacuo) were added. After 4 h, TLC (2:1 hex-
ane–EtOAc) indicated no starting material.
The mixture was diluted with ether (250 mL)
and filtered through a sintered glass funnel
containing a small pad of silica gel with 10%
MgSO4. Evaporation of the solvent gave the
product as a white solid; yield 1.10 g (96%).
This solid was used in the next step without
further purification. An analytical sample of
the a anomer was obtained by flash chro-
matography with 1:1 hexane–EtOAc; mp
216–217 °C (dec.), [h]D +114° (c 1, CHCl3),
wmax 1754 cm−1 (CO); MS: m/z 283 (M+1),
251 (M+1−MeOH), 177 (M+1−PhCHO).
Anal. Calcd for C14H15FO5 (282.28): C, 59.56;
H, 5.37. Found: C, 59.34; H, 5.42.
Methyl 3-O-acetyl-4-O-benzoyl-6-bromo-2,
6-dideoxy-2-fluoro-h- -allopyranoside (6).—
D
Dry CCl4 (50 mL) was filtered over alumina
into a round-bottom flask containing a mix-
ture of 5 (0.888 g, 2.72 mmol), NBS (0.533 g,
2.99 mmol), and BaCO3 (0.537 g, 2.72 mmol).
(Note: The NBS and BaCO3 had been dried
in vacuo for 3 h at 80 °C.) The resulting
mixture was stirred vigorously at 75–80 °C
under reflux until the usual color changes
were complete (ca. 20 min), at which time
TLC (2:1 hexane–EtOAc) indicated that no
starting material was present. The mixture
was filtered through Celite, and the filter cake
was washed with hot CCl4 (200 mL). The
solvent was removed, and the residue was
dissolved in CH2Cl2 (100 mL). The resulting
solution was washed with satd aq NaHCO3
Methyl 4,6-O-benzylidene-2-deoxy-2-fluoro-
h- -allopyranoside (4).—A solution of 3 (0.90
D
g, 3.17 mmol) in EtOH (50 mL) was cooled to
−20 °C and NaBH4 (0.362 g, 9.57 mmol) was
added. The mixture was allowed to warm
slowly to room temperature, at which point it
became homogeneous. The excess borohy-
dride was decomposed by dropwise addition
of 1.0 M HOAc until effervescence ceased.
The solvent was removed in vacuo, and the
residue was dissolved in CH2Cl2 (50 mL). The

S.T. Deal, D. Horton / Carbohydrate Research 315 (1999) 187–191
191
and brine, dried (Na2SO4), and the solvent
removed to yield a light-yellow solid. This
solid was recrystallized from 95% EtOH to
yield 6 (0.961 g, 87%) as colorless needles; mp
149–150 °C, [h]D +105° (c 1, CHCl3); MS:
m/z 405 (M+), 373 (M+ −MeOH), 327
(M+1−HBr). Anal. Calcd for C16H18BrFO6
(405.23): C, 47.42; H, 4.49; Br, 19.72. Found:
C, 47.49; H, 4.52, Br, 19.66.
Methyl 2,6-dideoxy-2-fluoro-i-
L
-talopyran-
oside (9).—To a solution of 8 (0.455 g, 1.39
mmol) in dry MeOH (20 mL), a 25%
methanolic NaOMe solution (0.160 mL) was
added. The reaction was complete in 10 min
by TLC (1:1 hexane–EtOAc). An excess of
solid CO2 was added, and the solvent was
removed. The residue was taken up in EtOAc
and filtered through a short column of silica
gel with EtOAc to yield 9 (0.220 g, 88%) as a
white solid; mp 136–137 °C, [h]D +63.7° (c
1.45, CHCl3); MS: m/z 181 (M+1), 161
(M+1−HF), 149 (M+1−MeOH). Anal.
Calcd for C7H13FO4 (180.19): C, 46.66; H,
7.29. Found: C, 46.40; H, 7.24.
Methyl
3-O-acetyl-4-O-benzoyl-2,6-dide-
-ribo-hex-5-enopyranoside
oxy-2-fluoro-h-
D
(7).—A solution of 6 (0.101 g, 0.249 mmol) in
5 mL of dry benzene was heated to reflux, and
DBU (0.075 mL, 0.499 mmol) was added to
the hot solution under argon. After 2 h, TLC
(2:1 hexane–EtOAc) indicated incomplete re-
action. An additional 0.038 mL of DBU was
added, and the reaction was allowed to con-
tinue for 30 min, at the end of which time
TLC indicated that all of the starting material
had been consumed. The mixture was diluted
with benzene, washed with 5% HCl (2×15
mL), satd aq NaHCO3 (2×15 mL) and H2O
(1×15 mL), dried (Na2SO4), and the solvent
was removed in vacuo to yield 7 (0.075 g,
93%) as a chromatographically homogeneous,
very low melting solid; [h]D +91° (c 1,
CHCl3); MS: m/z 325 (M+1), 293 (M+1−
MeOH), 265 (M+1−HOAc). Anal. Calcd
for C16H17FO6 (324.32): C, 59.25; H, 5.29.
Found: C, 59.18; H, 5.33.
References
[1] For preliminary reports, see (a) S.T. Deal, D. Horton,
Abstr. Papers Am. Chem. Soc. Meet., 200 (1990) CARB-
038. (b) S.T. Deal, Ph.D. Dissertation (D. Horton, pre-
ceptor), The Ohio State University, 1990. (c) Dissert.
Abstr. Int., 51-12B (1990) 5870.
[2] (a) E.-F. Fuchs, D. Horton, W. Weckerle, Carbohydr.
Res., 57 (1977) C36–C39. (b) E.-F. Fuchs, D. Horton,
W. Weckerle, E. Winter-Mihaly, J. Med. Chem., 22
(1979) 406–411. (c) D. Horton, W.R. Turner, Carbohydr.
Res., 77 (1979) C8–C11. (d) D. Horton, W. Priebe, W.R.
Turner, Carbohydr. Res., 94 (1981) 11–25. (e) D. Hor-
ton, W. Priebe, J. Antibiot., 34 (1981) 31–37. (f) D.
Horton, W. Priebe, J. Antibiot., 36 (1983) 1211–1215.
[3] (a) D. Horton, W. Priebe, O. Varela, Carbohydr. Res.,
130 (1984) C1–C3. (b) D. Horton, W. Priebe, US Pat.
4,537,882, August 27, 1985. (c) D. Horton, W. Priebe,
Carbohydr. Res., 136 (1985) 391–396.
Methyl
oxy-2-fluoro-i-
3-O-acetyl-4-O-benzoyl-2,6-dide-
-talopyranoside (8).—To a
[4] D. Horton, W. Priebe, O. Varela, Carbohydr. Res., 144
(1985) 305–315.
L
solution of 7 (0.643 g, 1.98 mmol) in dry
EtOAc (10 mL), 10% Pd/BaSO4 (0.090 g) was
added and hydrogen was gently bubbled into
the mixture. The reaction was complete in 30
min by TLC (2:1 hexane–EtOAc). The mix-
ture was filtered through Celite, and the sol-
vent evaporated. The residue was filtered
through a short column of silica gel with 2:1
hexane–EtOAc to yield 8 (0.573 g, 89%) as a
thick colorless oil; [h]D −1.3° (c 1.88,
CHCl3); MS: m/z 327 (M+1), 307 (M+1−
HF), 295 (M+1−MeOH). Anal. Calcd for
C16H19FO6 (326.34): C, 58.88; H, 5.88. Found:
C, 58.96; H, 5.94.
[5] K. Ok, Y. Takagi, T. Tsuchiya, S. Umezawa, H.
Umezawa, Carbohydr. Res., 169 (1987) 6–81.
[6] P. Kova´c' , Carbohydr. Res., 153 (1986) 168–170.
[7] D. Horton, W. Weckerle, Carbohydr. Res., 44 (1975)
227–240.
[8] M.E. Evans, Carbohydr. Res., 21 (1972) 473–475.
[9] J. Herscovici, K. Antonakis, J. Chem. Soc., Chem. Com-
mun., (1980) 561–562.
[10] E.J. Corey, J.W. Suggs, Tetrahedron Lett., (1975) 2647–
2650.
[11] S. Hanessian, N.R. Plessas, J. Org. Chem., 34 (1969)
1035–1058.
[12] A.J. Gordon, R.A. Ford, The Chemist’s Companion,
Wiley, New York, 1972.
[13] W.C. Still, M. Kahn, A. Mitra, J. Org. Chem., 43 (1978)
2923–2925.
[14] H.H. Baer, F. Herna´ndez Mateo, L. Siemsen, Carbohydr.
Res., 195 (1989) 225–245.
.