Preparation of C-3,5-acyl furanoses via highly selective intramolecular acyl migration

Article abstract of DOI:10.1021/jo049121y

A practical synthesis of C-3,5-acyl furanose via a base-catalyzed, highly selective intramolecular acyl migration in alcohol solvents is reported.

Full text of DOI:10.1021/jo049121y

still be achieved5b via a selective intramolecular acyl
migration on open-chain substrates by achieving reaction
conditions in favor of pathway A over B or C. As shown
in Scheme 1, only the C-4 or C-5 OH in substrates II and
III is readily available to form furanose or pyranose while
the remaining OH’s are protected.7 The 1,2 or 1,3 acyl
migration in the pathway A, B, or C could potentially
occur via five- and six-membered ring transition inter-
mediates Va /Vb or even bicyclic ortho ester intermedi-
ates IVa /IVb as shown in Scheme 1.
Starting from the literature known diols 1,3d,8 which
are easily prepared by several methods from inexpensive
commercially available starting materials, the precursors
4 for acyl migration studies can be quickly accessed via
a one-pot through process (Scheme 2).
Esterification of 1 with an acyl chloride in the presence
of 3 equiv of pyridine and 2.2 equiv of RCOCl in 7
volumes of MeCN at 60 °C gave a homogeneous reaction
with 99% selectivity of diester vs monoester and near full
conversion (>98% assay yield) within 10 h. Use of MeCN
allowed carrying out the three-step preparation in one
pot. Thus, direct addition of 3 equiv of aqueous 20%
P r ep a r a tion of C-3,5-Acyl F u r a n oses via
High ly Selective In tr a m olecu la r Acyl
Migr a tion
Feng Xu,* Bryon Simmons, Kimberly Savary,
Chunhua Yang, and Robert A. Reamer
Department of Process Research, Merck Research
Laboratories, Rahway, New J ersey 07065
feng_xu@merck.com
Received May 25, 2004
Abstr a ct: A practical synthesis of C-3,5-acyl furanose via
a base-catalyzed, highly selective intramolecular acyl migra-
tion in alcohol solvents is reported.
C-3,5- and C-2,3,5-protected furanoses are useful build-
ing blocks in carbohydrate chemistry.1,2 These synthons,
including C-branched or nonbranched, have been widely
used in glycosylation2 to prepare important biologically
active compounds such as saccharides, nucleosides, an-
tibiotics, etc. In conjunction with an antiviral nucleoside
drug development program, we became interested in
developing a general protocol for synthesis of C-3,5-acyl
C-2-substituted furanose derivatives. Protected furanoses
are often prepared by a series of protection/deprotection
operations on a suitable furanose or pyranose in order
to achieve selective protection on OH groups. Preparation
of C-3,5 acyl-protected furanoses in the literature is
limited by choices of protection and deprotection proto-
cols, which are often substrate specific.3 In particular,
few methods have been reported for the preparation of
C-2-branched furanoses.4
We report here a practical preparation of C-3,5-acyl
furanoses via selective intramolecular 1,2 vs 1,3 acyl
migration in the presence of base. Intramolecular acyl
migration has been reported in carbohydrate chemistry;
however, it is often used as an efficient method for
selective protection of OH(s) for a specific substrate.5 It
is also well-known the ratio of furanose vs pyranose can
be significantly affected by the strerochemistry of the
substrates if both the C-4 and C-5 OH’s are not capped;6
however, we reasoned that preparation of furanose could
(3) For examples, see: (a) Callam, C. S.; Gadikota, R. R.; Lowery,
T. L. J . Org. Chem. 2001, 66, 4549. (b) McCormick, J .; Li, Y.;
McCormick, K.; Duynstee, H. I.; Van Engen, A. K.; Van der Marel, G.
A.; Ganem, B.; Van Boom, J . H.; Meinwald, J . J . Am. Chem. Soc. 1999,
121, 5661. (c) Desire, J .; Prandi, J . Carbohydr. Res. 1999, 317, 110.
(d) Ono, A. M.; Shiina, T.; Ono, A.; Kainosho, M. Tetrahedron Lett.
1998, 3, 2793. (e) Chen, S.-H.; Lin, S.; King, I.; Spinka, T.; Dutschman,
G. E.; Gullen, E. A.; Cheng, Y.-C.; Doyle, T. W. Bioorg. Med. Chem.
Lett. 1998, 8, 3245. (f) Agrofoglio, L. A.; J acquinet, J .-C.; Lancelot, G.
Tetrahedron Lett. 1997, 38, 1411. (g) Robles, R.; Rodriguez, C.;
Izquierdo, I.; Plaza, M. T.; Mota, A. Tetrahedron: Asymmetry 1997, 8,
2959. (h) Kraus, G. A.; Maeda, H. J . Org. Chem. 1995, 60, 2. (i) Genu-
Dellac, C.; Gosselin, G.; Imbach, J . L. Carbohydr. Res. 1991, 216, 249.
(j) J eronic, L. O.; Sznaidman, M. L.; Cirelli, A. F.; De Lederkremer, R.
M. Carbohydr. Res. 1989, 191, 130. (k) Schmidt, R. R.; Gohl, A. Chem.
Ber. 1979, 112, 1689. (l) Ritzmann, G.; Klein, R. S.; Hollenberg, D. H.;
Fox, J . J . Carbohydr. Res. 1975, 39, 227.
(4) (a) Li, N.-S.; Tang, X.-Q.; Piccirilli, J . A. Org. Lett. 2001, 3, 1025.
(b) Schmit, C. Synlett 1994, 2. (c) Beigelman, L. N.; Ermolinskii, B.
S.; Gurskaya, G. V.; Tsapkina, E. N.; Karpeiskii, M. Y.; Mikhailov, S.
N. Carbohydr. Res. 1987, 166, 219. (d) Novak, J . J . K.; Smejkal, J .;
Sorm, F. Collect. Czech. Chem. Commun. 1971, 36, 3670; Tetrahedron
Lett. 1969, 1627. (e) J enkins, S. R.; Arison, B.; Walton, E. J . Org. Chem.
1968, 33, 2490.
(5) (a) Cipolla, L.; La Ferla, B.; Lay, L.; Peri, F.; Nicotra, F.
Tetrahedron: Asymmetry 2000, 11, 295. (b) Horrobin, T.; Tran, C. H.;
Crout, D. J . Chem. Soc., Perkin Trans. 1 1998, 1069. (c) Shing, T. K.
M.; Wan, L. H. J . Org. Chem. 1996, 61, 8468. (d) Goueth, P. Y.; Gogalis,
P.; Bikenga, R.; Gode, P.; Postel, D.; Ronco, G.; Villa, P. J . Carbohydr.
Chem. 1994, 13, 249. (e) Chaplin, D.; Crout, D. H.; Bornemann, S.;
Hutchinson, D. W.; Khan, R. Chem. Soc., Perkin Trans. 1 1992, 235.
(6) Hough L.; Richardso, A. G. Mono Saccharide Chemistry. In
Comprehensive Organic Chemistry; Barton, D., Ollis, W. D., Eds.;
Pergamon Press: London, 1979; Vol. 5, p 687.
(7) Apparently, multiple equilibria are involved in the process. It is
not clear whether the ratio of furanoses vs pyranoses is the kinetic or
thermodynamic controlled results. However, it was found later on that
exposure of 7a or 6a under acyl migration conditions resulted in
formation of less than 10% of 6a or 7a , respectively. This provided
some evidence that (a) formation of furanose through pathways B +
C (I f II f III) or C is not the main pathway; (b) formation of furanose/
pyranose could be dominated via pathways A and B from I.
(1) For recent reviews, see: (a) Nicolaou, K. C.; Mitchell, H. J .
Angew. Chem., Int. Ed. 2001, 40, 1624. (b) Vorbru¨ggen, H. Ruh-
Pohlenx, C. Org. React. 2000, 55, 1.
(2) For recent examples, see: (a) Eldrup, A. B.; Allerson, C. R.;
Bennett, C. F.; Bera, S.; Bhat, B.; Bhat, N.; Bosserman, M. R.; Brooks,
J .; Burlein, C.; Carroll, S. S.; Cook, P. D.; Getty, K. L.; MacCoss, M.;
McMasters, D. R.; Olsen, D. B.; Prakash, T. P.; Prhavc, M.; Song, Q.;
Tomassini, J . E.; Xie, J . J . Med. Chem. 2004, 47, 2283. (b) Li, N.,
Piccirilli, J . A. J . Org. Chem. 2003, 68, 6799. (c) Griffon, J .-F.; Mathe,
G.; Faraj, A.; Aubertin, A.-M.; De Clercq, E.; Balzarini, J .; Sommadossi,
J .-P.; Gosselin, G. Eur. J . Med. Chem. 2001, 36, 447. (d) Yin, H.;
Lowary, T. L. Tetrahedron Lett. 2001, 42, 5829. (e) D’Souza, F. W.;
Ayers, J . D.; McCarren, P. R.; Lowary, T. L. J . Am. Chem. Soc. 2000,
122, 1251. (f) Wang, Z.; Prudhomme, D. R.; Buck, J . R.; Park, M.; Rizzo,
C. J . J . Org. Chem. 2000, 65, 5969. (g) Girardet J .; Gunic, E.; Esler,
C.; Cieslak, D.; Pietrzkowski, Z.; Wang, G. J . Med. Chem. 2000, 43,
3704. (h) Sabchez, S.; Bamhaoud, T.; Prandi, J . Tetrahedron Lett. 2000,
41, 7447. (i) Tang, X.; Liao, X.; Piccirilli, J . A. J . Org. Chem. 1999, 64,
747. (j) Harry-O’kuru, R. E.; Smith, J . M. Wolfe, M. S. J . Org. Chem.
1997, 62, 1754.
(8) (a) Shing, T. K. M.; Wong, C.-H.; Yip, T. Tetrahedron: Asymmetry
1996, 7, 1323-1340. (b) Xie, M.; Berges, D. A.; Robins, M. J . Org.
Chem. 1996, 61, 5178. (c) Shiraki, R.; Sumino, A.; Tadano, K.; Ogawa,
S. J . Org. Chem. 1996, 61, 2845. (d) Yoshikawa, M.; Okaichi, Y.; Cha,
B. C.; Kitagawa, I. Tetrahedron 1990, 47, 7459. (e) Weber, J . F.;
Talhouk, J . W.; Nachman, R. J .; You, T.-P.; Halaska, R. C.; Williams,
T. M.; Mosher, H. S. J . Org. Chem. 1986, 51, 2702-2706. (f) Brima-
combe, J . S.; Ching, O. A. Carbohydr. Res. 1968, 8, 82.
10.1021/jo049121y CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/06/2004
J . Org. Chem. 2004, 69, 7783-7786
7783

TABLE 1. Selected Scr een in g Resu lts for Acyl
Migr a tion
SCHEME 1. Exp ected Acyl Migr a tion P a th w a ys
Lea d in g to F u r a n oses/P yr a n osesa
conditions
(T (°C),
conv
entry
base
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
solvents
time (days)) (%) 6a /7a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
10%H2O-MeOH
10%H2O-MeOH
MeOH
EtOH
i-PrOH
n-PrOH
THF-H2O
THF
22, 2
60, 5 h
60, 10 h
60, 2
60, 2
60, 2
60, 2
60, 2
0, 1
99
97
98
97
52
90
54
4
99
98
99
80
45
97
92:8
90:10
90:10
94:6
94:6
85:15
88:12
63:37
96:4
Et3N
i-Pr2NH
i-Pr2NH
i-Pr2NH
DMAP
imidazole MeOH
NaHCO3 MeOH
MeOH
MeOH-i-PrOAc
MeOH
0, 2
96:4
22, 1
22, 2
45, 3
45, 2
88:12
93:7
94:6
MeOH
88:12
conversion within 1 h. Then, the mixture was partitioned
between i-PrOAc and water. The organic phase was
washed with water, NaHCO3, 5% sodium thiosulfate (to
remove color), and H2O. Partial deformylation (R3 ) H)
was observed for some substrates.9
a
Our studies on selective acyl migration were initially
focused on substrate 4a . A variety of solvents and bases
were examined to improve/optimize the selectivity. Table
1 illustrates selected results. Under certain conditions,
hydrolysis and methanolysis of toluoyl esters become
significant. Use of polar protic solvents such as MeOH
or water greatly accelerates the reaction rate. The
reaction rate decreases as MeOH is replaced with EtOH,
n-PrOH, and i-PrOH (entries 3-6). In the absence of
alcohol, solvent acyl migration suffers either low selectiv-
ity or low conversion (entries 7 and 8). Bases used to
catalyze acyl migration affect the selectivity dramatically.
i-Pr2NH is one of the best in terms of selectivity and yield.
Lower reaction temperature helps to improve the selec-
tivity (entries 9-11). Use of i-Pr2NH in 20% i-PrOAc/
MeOH or MeOH at 0 °C improves the selectivity to 96%
with >98% conversion (entry 9).10 Adding other solvents
such as i-PrOAc slows down the reaction rate (entry 10).
Acyl migration in Et3N/MeOH/H2O or Et3N/EtOH was
found to work well, too. However, use of i-Pr2NH/i-PrOAc/
MeOH in the through process9 avoids the need for a
complete solvent switch11 from i-PrOAc to MeOH between
the oxidation and acyl migration steps and streamlines
the process for large-scale preparation.
Possible five- and six-membered ring or bicyclic ortho ester
intermediates for pathway C are not shown here for simplicity.
SCHEME 2. P r ep a r a tion of C-4,5-Dia cyl
Ald eh yd esa
(9) For example, about 10% of 5 was formed during the oxidation.
Although the aldehyde 4a can be isolated as a crystalline solid, the
deformylated product 5 (oil) would be lost leading to lower overall
yields. In practice, the crude oxidation product is used for through
process without loss of acyl migration selectivity and yield, because
acyl migration goes through the deformylated alcohol (Table 2, entry
1), which can be observed and isolated during acyl migration. Unless
otherwise mentioned, the results listed in Tables 1 and 2 were obtained
by using purified material to minimize possible peak overlap caused
by other byproducts with pyranose or furanose on HPLC.
a
Reagents and conditions: (a) RCOCl, Py, MeCN, 60 °C, 10 h;
(b) HClO4, MeCN-H2O, 55 °C; (c) H5IO6, MeCN-H2O, 0 °C.
HClO4 to the crude reaction mixture of 2 at 55 °C led to
about 95% conversion to 3. Periodic acid (1.2 equiv) was
charged at 0 °C to oxidatively cleave lactol 3 with full
(10) The ratio of furanose vs pyranose was almost unchanged during
the entire course of the acyl migration.
7784 J . Org. Chem., Vol. 69, No. 22, 2004

TABLE 2. Syn th esis of F u r a n ose via Selective Acyl
Migr a tion
crude acyl migration products could be directly hydro-
genated to give the crystalline diol 8 in five-step through
process. Thus, after workup and charcoal treatment, the
crude i-PrOAc reaction mixture of 6a and 7a (typical ratio
96:4) is subjected to hydrogenolysis, which has typically
been performed using 20 wt % of 10% Pd/C, 45 psi, 60
°C, 24 h. The product 8, in 66% isolated overall yield of
five-step through process, is directly crystallized from
20% i-PrOAc/heptane as white solid with excellent rejec-
tion of pyranose 9.
Next, the scope of the selective acyl migration with
other substrates was explored. Selective migration of
aromatic or alkyl esters works well when alcoholysis/
hydrolysis of ester is minimized under reaction condi-
tions. For aromatic esters, having electron-withdrawing
or electron-donating substituents on aromatic rings does
not greatly influence the selectivity of acyl migration. For
example, high selectivity of acyl migration can be still
obtained by replacing the toluoyl ester group in 4a with
p-chlorobenzoyl or benzoyl esters (Table 1, entries 9 and
10; Table 2, entries 8 and 9). However, the acyl migration
reaction rate, as expected, does increase when an aro-
matic ester has an electron-withdrawing group and is
more labile under basic reaction conditions. For example,
6c was obtained in about 45% yield and 4:1 selectivity
due to methanolysis of the p-chlorbenzoyl group if 4c was
exposed to i-Pr2NH in MeOH for 20 h. However, 6c can
be obtained in 85% yield with 94:6 selectivity if the
reaction was quenched within 5 h. This methanolysis
issue could be easily overcome by replacing MeOH with
EtOH, while the selectivity remains almost unaffected.12
Exposure of the product in EtOH/i-Pr2NH for days does
affect the yield and selectivity. Migration of bulky piv-
aloyl ester 4d 8c can also be achieved highly selectively
although the reaction needs to be carried out at 30-35
°C.
a
1.5 equiv of i-Pr2NH was used for all reactions. Key: (A)
MeOH, 0 °C; (B) EtOH, 0 °C; (C) MeOH, 35 °C; (D) EtOH, 30 °C.
b
Selectivity ) furanose vs pyranose. The ratio is determined by
HPLC analysis (Zorbox RX-C8, 5 µm particle size, 250 × 4.6 mm,
mobile phase: 10 mM pH 6.5 NaKHPO4-NaH2PO4/MeCN, 35 °C).
While intramolecular acyl migration works well with
C-2-branched substrates, our initial selectivity for pre-
paring C-2 nonbranched furanoses in MeOH are as low
as ∼3:1 (Table 2, entries 3 and 4). By carefully monitoring
c Yield of furanoses. Isolated yield. e Assay yields calculated by
d
HPLC using an external reference standard. f Four-step through-
process yield.
1
the reaction with HPLC and H NMR, it was found that
In a typical experiment, a solution of 4a or a crude
mixture of 4a and 5 (5 works equally well as 4 under
the same reaction conditions; Table 2, entry 1) in MeOH
or MeOH/i-PrOAc in the presence of 1.5-0.5 equiv of
i-Pr2NH is aged at 0-5 °C. The desired product can be
isolated by column chromatography in 82% overall yield
(four steps from 1a ). The ratio of R,â-lactols (by 1H NMR)
usually changes after column chromatography. Although
a mixture of R,â-lactols 6a or 7a is obtained, both 6a and
7a can be unambiguously characterized by NMR tech-
niques. In addition, we further demonstrated that the
lower selectivity is partially due to significant competition
reactions between migration vs hydrolysis/methanolysis
of benzoyl esters and cyclized lactols. Thus, treatment
of 4e with i-Pr2NH in EtOH at 0-5 °C for 30 h gave the
desired furanose in 94:6 selectivity and 82% yield.
Although no epimerization on C-2 was observed under
reaction conditions, the stereochemistry of C-2 in this
case does influence the reaction rate. Acyl migration of
4e is significantly faster than that of 4f at 0 °C. It almost
required a week to reach 95% conversion. However, 4f
is converted to furanose 6f at 30 °C within 1.5 days in
65% yield and 84:16 selectivity (vide infra).7,10
(11) MeCN was removed during the concentration of the organic
extracts, obtained from the oxidation cleavage step, to give a solution
of 4 and 5 in i-PrOAc, which can be directly used for acyl migration.
For detailed experimental procedures, see the Supporting Information.
(12) As we noticed in Table 1, acyl migration can also be carried
out in EtOH to achieve 94% selectivity for 4a .
J . Org. Chem, Vol. 69, No. 22, 2004 7785

By taking advantage of the observed ester migration
rate differences on 4a -c, we could achieve higher
selectivity by installing different esters at C-5,6 on
substrates 1 to expand the scope of our methodology.
Actually, when 1013 was subjected to typical acyl migra-
tion conditions, above 98:2 acyl migration selectivity was
achieved. 11 was isolated as a solid in 70% overall yield
over four steps.
To further probe the influence of stereochemistry on
acyl migration, 12, which is the C-4 epimer of 10 and 4f,
was prepared.13 Acyl migration of 12 vs 4e and 4f is much
faster. Exposure of 12 in i-Pr2NH/MeOH at 0 °C for 4 h
afforded the desired furanose 13 in 70% yield and >99:1
selectivity.14
R1 ) OBn and R2 ) H or alkyl (such as Me and n-Bu) on
C-2. However, when 4h and 4i are subjected to typical
reaction conditions, retro-aldol reaction instead of acyl
migration via the corresponding deformylated alcohol
becomes the main reaction pathway when R2 ) vinyl or
phenyl.15
In conclusion, a practical method for preparation of
C-3,5 acyl furanoses via base catalyzed, highly selective
intramolecular acyl migration was demonstrated. In
addition, higher acyl migration selectivity could be
obtained by having different ester groups, whose migra-
tion rates can be differentiated, on substrates. However,
the substituents on C-2 do affect the reaction pathway.
In the cases of both vinyl and aryl substituents are
installed on C-2 position, retro-aldol reaction becomes a
major reaction pathway. By applying this methodology,
a variety of C-2,3,5 or C-3,5 selectively protected fura-
noses can be quickly accessed via a practical through
process in high selectivity and yield.
The variability of the substituents on C-2 was also
studied. High yield and selectivity can be achieved when
(13) 10 and 12 were prepared from its corresponding monoester by
following the same procedures used for making 4a -h . For detailed
experimental procedures, see the Supporting Information.
(14) Although the toluoyl ester migrates for 12 vs benzoyl ester for
4e and 4f, the following migration rate order was observed for other
substrates: toluoyl < benzoyl < p-chlorbenzoyl. Therefore, the observed
acyl migration rate order (12 > 4e > 4f) is conceivable. The detailed
migration mechanism is unclear;7,10 however, the interesting observed
rate and selectivity differences between 4e, 4f, and 12 (4e and 4f are
C-2 epimers; 4f and 12 are C-4 epimers) seems to indicate that the
aldehyde group could possibly participate in formation of a key acyl
migration intermediate such as the bicyclic ortho ester intermediates
(Scheme 1). The conformational/stereochemical differences in the
bicyclic ortho esters intermediates IVa /IVb resulted from the substit-
uents at C-2 and C-4 of 4e, 4f and 12 are significant,7 which could
contribute to the observed rate and selectivity differences. Similarly,
the differences of the stereogenic centers at C-2, C-3, and C-4 on the
five or six member ring intermediates Va /Vb (Scheme 1) would not be
expected to result in such a dramatic difference in terms of the reaction
rate.
Ack n ow led gm en t. We gratefully acknowledge Dr.
E. J . J . Grabowski for his encouragement.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data for compounds 4a -c,f-i,
6a -g, 8, and 10-13. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O049121Y
(15) For example, 14 and 15 were obtained when 4h was treated
with 1.5 equiv of i-Pr2NH in MeOH at 0 °C for 10 h. Formation of 16,
observed by HPLC, was completed within 2 h. The stereochemistry of
15 was determined by NOE studies.
7786 J . Org. Chem., Vol. 69, No. 22, 2004