Reduction of glycosyl bromides to anhydroalditols by titanocene borohydride

Full text of DOI:10.1021/jo951880g

J . Org. Chem. 1996, 61, 3863-3864
3863
Notes
Ta ble 1. Red u ction of Glycosyl Br om id es to
An h yd r oa ld itols by 1
Red u ction of Glycosyl Br om id es to
An h yd r oa ld itols by Tita n ocen e
Bor oh yd r id e
Cullen L. Cavallaro and J effrey Schwartz*
Department of Chemistry, Princeton University,
Princeton, New J ersey 08544-1009
Received October 20, 1995. (Revised Manuscript Received
March 4, 1996 )
Anhydroalditols are a class of reduced sugars which
are present in many natural systems and which have also
been employed as synthesis intermediates. For example,
1
,5-anhydro-D-glucitol has been isolated from several
1
plant and animal sources, discovered in human cere-
2
brospinal fluid and blood plasma, and found to be an
inhibitor of several enzymes.3 1,5-Anhydrohexitols have
4
been used in the synthesis of carbohydrates and nucleo-
sides which have antiviral activity.5 Anhydroalditols
have been synthesized variously, including by dehydra-
6
tion of the glycitol, but the most commonly used path-
ways involve reduction of glycoside intermediates. Such
7
reductive routes include desulfurization of 1-thio- or
8
isothiocyanate derivatives, reaction of silyl alkyl glyco-
9
sides with TMSOTf/triethylsilane, or treating a glycosyl
halide with LAH.10 The most commonly used methodol-
ogy involves reduction of a glycosyl halide with tin
hydride and a radical promoter.11 Yields for this latter
process are good, but the toxicity of organotin byproducts
and the difficulty of their complete separation from
high-yielding conversion of glycosyl bromides to anhy-
droalditols using a Ti(III) borohydride complex.
desired materials can be a drawback in pharmaceutical
applications. We have recently shown12 that simple
Reaction between Cp
2
TiCl
2
and NaBH
4
yields ti-
tanocene borohydride13 (Cp
TiBH
; 1), which we have
2
4
Ti(III) complexes can readily activate glycosyl halides by
halogen atom abstraction, and we now report a simple,
shown reduces activated alkyl halides via a radical
intermediate.14 Since glycosyl radicals are rapidly
formed by reaction between glycosyl bromides and
15
(1) Yamanouchi, T.; Tachibana, Y.; Akanuma, H.; Minoda, S.;
1
6
Shinohara, T.; Moromizato, H.; Miyashita, H.; Akaoka, I. Am. J .
Physiol. 1992, 263, E268.
(Cp
2
TiCl)
2
it was of interest to learn if 1 could effect
reduction of glycosyl halides analogously. Indeed, reduc-
tion of the glycosyl halides by 1 can be accomplished
rapidly and at room temperature (Table 1).
(
2) (a) Yoshioka, S.; Saitoh, S.; Negishi, C.; Fujisawa, T.; Fujimori,
A.; Takatani, O.; Imura, M.; Funabashi, M. Clin. Chem. 1983, 29, 1396.
b) Yoshioka, S.; Saitoh, S.; Seki, S.; Seki, K. Clin. Chem. 1984, 30,
88.
3) (a) Glucosidase: Field, R. A.; Haines, A. H.; Chrystal, E. J . T.
(
1
The mechanism of glycosyl halide reduction likely
involves halogen atom abstraction by Ti(III) to give the
(
Bioorg. Med. Chem. Lett. 1991, 1, 667. (b) Phosphorylase a: Bollen,
M.; Malaisse-Lagae, F.; Malaisse, W.; Stalmans, W. Biochem. Biophys.
Acta 1990, 1038, 141.
•
glycosyl radical (2), which then can abstract H from
17,18
borohydride (Scheme 1).
In support of this sugges-
(4) (a) Barili, P. L.; Berti, G.; Catelani, G.; D’Andrea, F.; Gaudiosi,
tion, we note the formation of 1-deuterio-2,3,4,6-tetra-
O-acetyl-1,5-anhydro-D-glucitol (89% d ) when Cp TiBD
1 2 4
is used as the reducing agent.
A. Gazz. Chim. Ital. 1994, 124, 57. (b) Barili, P. L.; Berti, G.; D’Andrea,
F.; Bussolo, V. D.; Granucci, I. Tetrahedron 1992, 48, 6273. (c) Barili,
P. L.; Berti, G.; D’Andrea, F.; Gaudiosi, A. Carbohydr. Res. 1991, 212,
c5.
(
5) (a) Verheggen, I.; Van Aerschot, A.; Toppet, S.; Snoeck, R.;
J anssen, G.; Balzarini, J .; De Clercq, E.; Herdewijn, P. J . Med. Chem.
993, 36, 2033. (b) Van Aerschot, A.; Verheggen, I.; Herdewijn, P.
(13) (a) N o¨ th H.; Hartwimmer, R. Chem. Ber. 1960, 93, 2238. (b)
Lucas, C. R. Inorg. Synth. 1977, 17, 91.
1
Bioorg. Med. Chem. Lett. 1993, 3, 1013. (c) Huryn, D. M.; Sluboski, B.
C.; Tam, S. Y.; Weigele, M.; Sim, I.; Anderson, B. D.; Mitsuya, H.;
Broder, S. J . Med. Chem. 1992, 35, 2347. (d) Nair, V.; Nuesca, Z. M.
J . Am. Chem. Soc. 1992, 114, 7951.
(14) Liu, Y.; Schwartz, J . Tetrahedron 1995, 51, 4471.
(15) (a) Dupuis, J .; Giese, B.; R u¨ egge, D.; Fischer, H.; Korth, H.-G.;
Sustmann, R. Angew. Chem., Int. Ed. Engl. 1984, 23, 896. (b) Korth,
H.-G.; Sustmann, R.; Dupuis, J .; Giese, B. J . Chem. Soc., Perkin Trans.
2 1986, 1453. (c) Korth, H.-G.; Sustmann, R.; Gr o¨ ninger, K. S.; Witzel,
T.; Giese, B. J . Chem. Soc., Perkin Trans. 2 1986, 1461.
(16) Coutts, R. S. P.; Wailes, P. C.; Martin, R. L. J . Organomet.
Chem. 1973, 47, 375.
(
6) Duclos, A.; Fayet, C.; Gelas, J . Synthesis 1994, 1087.
(7) Richtmyer, N. K.; Carr, C. J .; Hudson, C. S. J . Am. Chem. Soc.
1
943, 65, 1477.
(
(
(
8) Witczak, Z. J . Tetrahedron Lett. 1986, 27, 155.
9) Bennek, J . A.; Gray, G. R. J . Org. Chem. 1987, 52, 892.
10) (a) Ness, R. K.; Fletcher, H. G., J r.; Hudson, C. S. J . Am. Chem.
(17) A small amount (1-5%) of the glycal is sometimes observed,
which might be formed by radical trapping by Ti(III) and Ti(IV)-OAc
elimination (see ref 12).
Soc. 1950, 72, 4547. (b) Funabashi, M.; Hasegawa, T. Bull. Chem. Soc.
J pn. 1991, 64, 2528.
(18) For example, see: Liu, Y.; Schwartz, J . J . Org. Chem. 1994,
59, 940. Aryl radicals photochemically generated from aryl halides
(11) (a) Giese, B.; Dupuis, J . Tetrahedron Lett. 1984, 25, 1349. (b)
-
•
•-
Kocienski, P.; Pant, C. Carbohydr. Res. 1982, 110, 330. (c) Auge, J .;
David, S. Carbohydr. Res. 1977, 59, 255.
react with BH
Bradbury, D. J . Am. Chem. Soc. 1973, 95, 5085. Analogous
abstraction from Ti(IV)BH would give a Ti(III)BH species.
4
by H abstraction to give BH . See: Barltrop, J . A.;
3
•
H
(12) Cavallaro, C. L.; Schwartz, J . J . Org. Chem. 1995, 60, 7055.
4
3
S0022-3263(95)01880-9 CCC: $12.00 © 1996 American Chemical Society

3
864 J . Org. Chem., Vol. 61, No. 11, 1996
Notes
Sch em e 1
acetyl-R-D-mannopyranose, 1,2,3,4-tetra-O-acetyl-â-D-xylopyra-
nose, and 2,3,4,6-tetra-O-acetyl-â-D-galactopyranosyl bromide
were purchased from Sigma Chemical Co. All other starting
materials were purchased from Aldrich Chemical Co. and were
used without further purification.
P er a cetyla ted Glycosyl Br om id es. The title compounds
were prepared as described elsewhere.12
Dicyclop en ta d ien yltita n iu m (III) Bor oh yd r id e.13b A so-
lution of Cp
2 2 4
TiCl (900 mg, 3.61 mmol) and NaBH (900 mg,
2
3.79 mmol) in 15 mL of DME was prepared under inert
atmosphere at room temperature. The resulting deep violet
solution was filtered to remove inorganic salts and concentrated
in vacuo. Solids were dissolved in toluene and filtered again.
Crystallization from toluene and pentane under N
of Cp TiBH as fluffy, purple crystals (86%).
P r ep a r a tion of 2,3,4,6-Tetr a -O-a cetyl-1,5-a n h yd r o-D-ga -
la ctitol Usin g Isola ted Cp TiBH Under an inert atmo-
sphere, Cp TiBH (1; 100 mg, 0.518 mmol) was dissolved in 6
2
gave 600 mg
2
4
Sch em e 2
2
4
.
2
4
mL of THF, giving a purple solution. 2,3,4,6-Tetra-O-acetyl-R-
D-galactopyranosyl bromide (100 mg, 0.243 mmol) was added.
The solution was stirred for 45 min at room temperature,
concentrated, and dried in vacuo. The crude mixture was
dissolved in 2:1 ethyl acetate/hexanes and passed through a
short plug of silica gel. Concentration in vacuo gave 2,3,4,6-
tetra-O-acetyl-1,5-anhydro-D-galactitol (80.1 mg; 99%) as a pale
yellow oil. Other anhydroalditols were prepared in analogous
fashion.
P r ep a r a tion of 2,3,4,6-Tetr a -O-a cetyl-1,5-a n h yd r o-D-glu -
2 2
We note that 1 prepared in situ from Cp TiCl and
citol Usin g Cp
170 mg, 4.49 mmol) and Cp
2
TiBH
4
P r ep a r ed in Situ . Sodium borohydride
TiCl (260 mg, 1.04 mmol) were
4
NaBH is less effective for glycosyl halide reduction than
(
2
2
IV
is isolated 1. We also note that the Ti byproduct of
glycosyl halide reduction (3) can be recycled to 1 with
borohydride, thus enabling a catalytic cycle for anhy-
droalditol synthesis (Scheme 2).19 However, in contrast
to reaction with ex situ prepared 1, significant glycal
byproduct is also obtained (anhydroalditol:glycal g 2:1),
likely formed by competitive trapping of 2 by either 1 or
dissolved in 10 mL of dimethoxyethane (DME) under an inert
atmosphere to give a purple solution to which 2,3,4,6-tetra-O-
acetyl-R-D-glucopyranosyl bromide (396 mg, 0.96 mmol) was
added. The reaction mixture was allowed to stir at room
temperature for 2 h and was then concentrated in vacuo to give
2
,3,4,6-tetra-O-acetyl-1,5-anhydroglucitol (83%) and 2,3,4,6-
1
2
tetra-O-acetyl-D-glucal (17%).
1
2
Ca ta lyzed Syn th esis of 2,3,4,6-Tetr a -O-a cetyl-1,5-a n h y-
d r o-D-glu citol. Sodium borohydride (40 mg, 1.06 mmol) and
3
.
2 2
Cp TiCl (6 mg, 0.02 mmol) were dissolved in 5 mL of dimeth-
Exp er im en ta l Section
oxyethane (DME) under an inert atmosphere to give a pale
purple solution to which was added 2,3,4,6-tetra-O-acetyl-R-D-
glucopyranosyl bromide (98 mg, 0.24 mmol). The solution was
allowed to stir at room temperature for 20 h and was then
concentrated in vacuo to give a mixture of 2,3,4,6-tetra-O-ace-
tyl-1,5-anhydro-D-glucitol (80%) and 2,3,4,6-tetra-O-acetyl-D-
Gen er a l Meth od s. Reaction solvents were dried and dis-
tilled prior to use using standard methods. 1,2,3,4,6-Penta-O-
(
19) In a separate experiment, it was found that Cp
converted back to 1 using NaBH
20) Paulsen, H.; Schnell, D.; Stenzel, W. Chem. Ber. 1977, 110,
707.
21) H NMR (270 MHz; C
2H, m, H2, H4); 3.56-3.41 (4H, m, H1, H1′, H5, H5′); 1.72 (3H, s,
2
TiCl is easily
4
.
(
12
glucal (20%).
3
1
(
6 6
D ): δ 5.54 (1H, broad s, H3); 4.95-4.89
(
Ack n ow led gm en t. The authors acknowledge sup-
port for this work provided by the National Science
Foundation.
1
3
OAc); 1.62 (6H, s, OAc). C NMR (75 MHz; C
6
D
6
): δ 164.1, 163.7
CdO); 63.0 (C3); 62.1 (C2, C4); 59.8 (C1, C5); 15.0, 14.9 (CH ). IR
NaCl) 1736 cm- (νCdO). Anal. C, H. Mp: 129-130 °C (cf. 133 °C;
(
(
3
1
Tejima, S.; Maki, T.; Akagi, M. Chem. Pharm. Bull. (Tokyo) 1964, 12,
28.
5
J O951880G