Design and synthesis of glycosidase inhibitor 5-amino-1,2,3,4- cyclohexanetetrol derivatives from (-)-vibo-quercitol

Article abstract of DOI:10.1016/j.bmc.2005.04.003

In continuation of development of bioactive inositol derivatives, a 1-O-methyl derivative of 5-amino-5-deoxy-l-talo-quercitol was designed and synthesized as an analogue of the strong α-fucosidase inhibitor, 5a-carba-α-l-fucopyranosylamine, the methyl bra

Full text of DOI:10.1016/j.bmc.2005.04.003

Bioorganic & Medicinal Chemistry 13 (2005) 4306–4314
Design and synthesis of glycosidase inhibitor
5-amino-1,2,3,4-cyclohexanetetrol derivatives
from (ꢀ)-vibo-quercitolI
Seiichiro Ogawa,a,* Miwako Asada,a Yoriko Ooki,b Midori Mori,b
Masayoshi Itohb and Takashi Korenagab
aDepartment of Biosciences and Informatics, Faculty of Science and Technology, Keio University, Hiyoshi,
Kohoku-ku, Yokohama 223-8522, Japan
bDepartment of Chemistry, Tokyo Metropolitan University, Minami-Ohsawa, Hachioji, Tokyo 192-0397, Japan
Received 10 February 2005; revised 5 April 2005; accepted 5 April 2005
Available online 5 May 2005
Abstract—In continuation of development of bioactive inositol derivatives, a 1-O-methyl derivative of 5-amino-5-deoxy-L-talo-
quercitol was designed and synthesized as an analogue of the strong a-fucosidase inhibitor, 5a-carba-a-L-fucopyranosylamine,
the methyl branch being replaced with methoxyl, and demonstrated to be a moderate a-fucosidase inhibitor. The present approach
provides a possible route to apply alkyl ethers of aminodeoxyinositols as hexopyranose mimics of biological interest.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
On chemical modification of a potent a-mannosidase
inhibitor, mannostatin A1,2 (1), the methyloxy analogue
1L-(1,2,3,5/4)-5-amino-4-O-methyl-1,2,3,4-cyclopentan-
etetrol3 (2) was found to also possess very strong inhib-
itory activity, comparable with that of the parent
compound. Furthermore, previous results3,4 for chemi-
cal modification of 1D-(1,2,3,4/0)-4-amino-1,2,3-cyclo-
pentanetriol, the core structure of 1, suggested the
presence of a methyloxy group, possibly an equivalent
of hydroxymethyl, to render a certain hydrophobic re-
gion of space around C-5, effectively contributing to
inhibitory activity. In the development of a-mannosi-
dase inhibitors, design of novel candidates should essen-
tially be based on mimicking structural features relating
to the postulated transition-state mannopyranosyl cat-
ion.5–7 The methylthio and methoxyl functions of 1
and 2 may match those of the 5-hydroxymethyl in the
carbocation model.
q In this paper, the nomenclature of aminocyclitols follows the
IUPAC–IUB 1973 Recommendation for Cyclitols (Pure Appl. Chem.
1974, 37, 285–297).
*
Figure 1. Methyloxy analogues 2 and 5 of respective mannostatin A
(1) and 5a-carba-a-L-fucopyranosylamine (3).
Corresponding author. Tel.: +81 45 566 1788; fax: +81 45 566
1789; e-mail: sogawa379@ybb.ne.jp
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.04.003

S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
4307
These considerations stimulated us to design, as
glycosidase inhibitors, a series of methyl ethers of 5-ami-
no-1,2,3,4-cyclohexanetetrols structurally related to
5a-carba-hexopyranosylamines (Fig. 1). Among all 5a-
carba-glycosylamines, the ground-state mimicking
glycosidase inhibitors, synthesized so far, 5a-carba-a-L-
fucopyranosylamine8,9 (3) was demonstrated to possess
the strongest inhibitory activity against a-fucosidase.
Therefore, it seemed desirable to choose it as a lead as
well as mimetic compound, and new derivatives were
generated by replacement of the methyl group with
methyloxy elements: the methyloxy analogue, namely
the 1-O-methyl derivative 5 of 5-amino-5-deoxy-L-talo-
quercitol (4) was synthesized, and its enzyme-inhibitory
activity was evaluated.
2. Results and discussion
Reaction of (ꢀ)-vibo-quercitol10 (6) with large excess of
2,2-dimethoxypropane (10 M equiv) in DMF in the pres-
ence of p-toluenesulfonic acid for 3 h at room tempera-
ture gave an inseparable mixture of the 1,2:3,4- and
1,2:4,5-di-O-isopropylidene derivatives11 (7 and 8) in
86% yield (Scheme 1). Treatment of the mixture with sul-
furyl chloride (3 M equiv) in the presence of DMAP in
pyridine at room temperature gave, after fractionation
over a silica gel column, two chloro compounds 9
(40%) and 10 (58%), the structures of which were readily
established on the basis of their 1H NMR spectra, reveal-
ing a doublet of doublets of doublets (d = 4.50, J = 2.3,
3.1, and 5.2 Hz) and a doublet of doublets (d = 4.62,
J = 2.9 and 5.6 Hz), respectively, due to the protons at-
tached to the carbon atoms bonding to the chlorine
atoms. These data indicated that halogen atoms were
incorporated through direct SN2 reactions with inversion
of the configuration.
Azidolysis of the desired chloride 9 with an excess of so-
dium azide in DMF at 100 ꢁC gave selectively the azide
11 (50%), accompanied by some elimination products.12
Attempted selective O-deisopropylidenation of 11 was
conducted under the influence of a trace of p-toluene-
sulfonic acid in methanol at 0 ꢁC, formation of mono-
O-isopropylidene derivative 12 being monitored by
TLC. The mixture of products was successfully sepa-
rated by a silica gel column to give 12 (71%), along with
11 (ꢁ10%) and the tetrol 13 (ꢁ7%). Selective tosylation
of 12 was carried out by treatment with excess p-toluene-
sulfonyl chloride (5 M equiv) in pyridine at ꢀ18 ꢁC.
After 4 days, two mono-tosylates 14 (43%) and 15
(29%), and the ditosylate 16 (15%) were afforded.13
Compounds 14 and 15 isolated by silica gel chromato-
graphy could be readily differentiated, their structures
being firmly established from their NMR spectra as 3-
and 4-tosylates, respectively, combined with data ob-
tained for the acetyl derivative 17 of 14.
Scheme 1. Synthesis of tosyl derivative 17 of 5-azido-5-deoxy-L-vibo-
quercitol. Reagents and conditions: (a) Me2C(OMe)2 (ꢁ6 M equiv),
DMF, p-TsOHÆH2O, rt; (b) SOCl2 (3 M equiv), DMAP, pyridine, 0 ꢁC;
(c) NaN3 (10 M equiv), 15-crown-5 ether, DMF, 100 ꢁC; (d) MeOH, p-
TsOHÆH2O, ꢁpH 4; (e) TsCl (5 M equiv), pyridine, ꢀ18 ꢁC.
sired aminodeoxyquercitol with a talo-configuration
(Scheme 2). Thus, 17 was hydrogenolyzed in ethanol
containing acetic anhydride in the presence of Raney
nickel to give crystalline amide tosylate 18, quantita-
tively. Treatment of 18 with sodium acetate (5 M equiv)
in 90% aqueous 2-methoxyethanol for three days at
120 ꢁC produced an approximately 10:1 mixture of the
two diols 19 and 22 with talo- and allo-configurations.
Alternatively, a similar reaction was carried out using
aqueous DMF to give an about 1:10 mixture of 19
and 22. The mixture was conventionally acetylated and
The azido tosylate 17 was first transformed into the N-
acetyl derivative 18, in the expectation that, under
acetolysis conditions, the amide may participate as a
neighboring group to give rise preferentially to the de-

4308
S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
Scheme 2. Synthesis of methyl ethers 5, 30, and 31 of 5-amino-5-deoxy-L-talo-quercitol (4). Reagents and conditions: (a) H2, Raney Ni, EtOH, Ac2O;
(b) anhydrous NaOAc (5 M equiv), 90% aqueous MeOCH2CH2OH, 110 ꢁC, 2 days; (c) Ac2O, pyridine; (d) benzyl bromide (3 M equiv), NaH,
(4 M equiv), DMF, 0 ꢁC; (e) 80% aqueous AcOH, 50 ꢁC; (f) CH3I, Ag2O, CH3CN, reflux temp, silica gel chromatography; (g) H2, 10% Pd/C, EtOH,
rt, aqueous 1 M Ba(OH)2, 80 ꢁC, 2 h, column of Dowex-50 W · 2 (H+) resin, 1% aqueous NH3.
the di-O-acetyl derivatives 20 and 23 were isolated pure,
respectively. Mechanistically, two compounds were pre-
dicted to be produced mainly by the acetolysis of 18
(Fig. 2). Thus, in an aprotic solvent DMF, direct SN2
reaction with an acetate ion would undergo preferen-
tially to afford the product with allo-configuration,
and, on the other hand, in a protic solvent aqueous 2-
methoxyethanol, the 4-acetoxyl would participate at
C-3 to form an intermediate acetoxonium ion between
C-3 and 4, which was cleaved to give rise to two prod-
ucts with allo- and talo-configurations. The latter was
obtained as major product through subsequent neigh-
boring participation of the amide. Thus, the two com-
pounds 20 and 23 were likely to be provided by direct
nucleophilic substitution and by preferential opening
of the acetoxonium ion, respectively.
deshielded by the axial 3-acetoxyl and the ketal oxygens.
The H NMR data thus provided clear support for the
assigned talo- and allo-configurations of 20 and 23.
1
Compound 20 was O-deacetylated with methanolic so-
dium methoxide (!19) and the diol obtained was again
protected with a benzyl group (!21). Then, removal of
the isopropylidene group generated the 1,2-unprotected
derivative 24. Removal of the benzyl groups of 24 by
hydrogenolysis in ethanol in the presence of Pd/C and
subsequent hydrolysis with aqueous barium hydroxide
gave, after purification over a column of Dowex-
50 W · 2 (H+) resin with aqueous 1% ammonia, 5-ami-
no-5-deoxy-L-talo-quercitol
4
quantitatively. Then,
attempted selective methylation of 24 was carried out
with 2 M equiv of iodomethane in acetonitrile in the
presence of silver oxide at reflux temperature. The mix-
ture of products was fractionated over a silica gel col-
umn to give an inseparable mixture (52%) of two
monomethyl derivatives 25 and 26, and one dimethyl
form 27 (33%). The former mixture was conventionally
acetylated to give the acetyl derivatives, which were
found to be separable on a silica gel column to give 28
(34%) and 29 (49%). Compounds 25, 26, and 27 were
deprotected followed by base hydrolysis, affording the
corresponding free bases 5, 30, and 31.
1
In the H NMR spectra of 20 and 23, adopting the pre-
ferred conformations (Fig. 3), the signals due to axially
oriented H-4 appeared as doublets of doublets (J = 5.1
and 7.4 Hz) and (J = 3.8 and 9.3 Hz) at d 5.30 and
4.95, respectively (Fig. 3). In the former, the signal
was deshielded by close proximity to the ketal oxygen
atoms. Signals due to H-5 attached to the carbon atoms
bonding to the amido groups resonated at d 4.69 and
4.53. The latter axial hydrogen seemed to be appreciably

S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
4309
Figure 2. Postulated reaction mechanism for formation of compounds 20 and 23 by treatment of 18 with sodium acetate in appropriate solvents,
followed by acetylation.
Table 1. Data for inhibitory activity of 5-amino-5-deoxy-L-talo-
quercitol (4) and its methyl ethers 5, 30, and 31 against three
glycosidase, compared with 5a-carba-L-fucopyranosylamine (3)
Compounda
IC50 (M)
a-Fucosidase a-Galactosidase b-Glucosaminidase
(bovine
kidney)
(green coffee
beans)
(bovine
kidney)
3b
4
9.3 · 10ꢀ6
NI
2.3 · 10ꢀ5
1.0 · 10ꢀ4
NI
NI
4.9 · 10ꢀ4
NI
2.0 · 10ꢀ5
NI
NI
NI
5
NI
4.0 · 10ꢀ4
NI
30
31
NI: no inhibition < 10ꢀ3 M.
a None of the compounds showed any notable inhibitory activity
against b-galactosidase (bovine liver), a-glucosidase (sucrase from
rat), b-glucosidase (almond), and a-mannosidase (Jack beans).
b Compound
3
has been observed9 to exhibit strong activity
(Ki = 1.2 · 10ꢀ8 M) against a-fucosidase (bovine kidney).
equivalents of the hydrophobic 5-methyl branching in
1, allowing us to design deoxyinosamine-type glycosi-
dase inhibitors other than a-fucosidase. From the pres-
ent results we therefore propose incorporation of
methyl ether groups as one effective route for applica-
tion of chemically modified inositols as pseudo-sugar
(carba-sugar) derivatives of biological interest.
Figure 3. Preferred conformations of compounds 20 and 23.
3. Biological assay
Data for inhibitory activity14 of compounds 4, 5, 30, and
31 toward three glycosidases are summarized in Table 1.
Compounds 5 and 30 possessed moderate inhibitory
activity toward a-fucosidase (bovine liver), as expected,
while, the parent free base 4 and the dimethyl ether 31
did not show any potential. Interestingly, compound
30 showed cross-inhibitory action toward b-galactosi-
dase (Jack beans). Thus, the present study has demon-
strated that methyloxy groups are likely to function as
4. Experimental
4.1. General methods
Optical rotations were measured with a JASCO DIP-370
polarimeter, and [a]D values are given in
1
10ꢀ1 degcm2 gꢀ1. H NMR spectra were recorded for

4310
S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
solutions in deuteriochloroform and deuteriomethanol
with internal tetramethylsilane (TMS) as a reference
with a JEOL JNM Lambda-300 (300 MHz) instrument.
Mass spectra were determined with Hitachi M-8000 ion
trap mass spectrometer. TLC was performed on silica
gel 60 F-254 (E. Merck, Darmstadt). The silica gel used
for a column chromatography was Wakogel C-300
(Wako Junyaku Kogyo Co., Osaka, 200–300 mesh) or
silica gel 60 KO (Katayama Kagaku Kogyo Co., Osaka,
70–230 mesh). Organic solutions were dried over anhy-
drous Na2SO4 and concentrated at >45 ꢁC under dimin-
ished pressure.
4.3. 5-Azido-5-deoxy-1,2:3,4-di-O-isopropylidene-L-vibo-
quercitol (11)
A mixture of 9 (605 mg, 2.3 mmol), sodium azide (1.5 g,
23 mmol), 15-crown-5 ether (4.6 mL, 23 mmol), and
DMF (12 mL) was stirred for 15 h at 100 ꢁC, and then
co-evaporated with butanol and toluene. The residue
was suspended with ethyl acetate (120 mL) and the mix-
ture was washed with water and saline, dried, and evap-
orated. The residue was chromatographed on a column
of silica gel (60 g, acetone–hexane 1:80) to give the azide
11 (310 mg, 50%) as a crystalline solid, Rf = 0.38 (ace-
20
D
1
tone–hexane 1:5); ½aꢂ ꢀ43 (c 5.2, CHCl3); H NMR
4.2. 5-Chloro-5-deoxy-1,2:3,4-di-O-isopropylidene-L-
muco-quercitol (9) and 3-chloro-3-deoxy-1,2:4,5-di-O-
isopropylidene-L-allo-quercitol (10)
(300 MHz, CDCl3): d 4.40 (ddd, 1H, J1,6eq = 2.4 Hz,
J1,6ax = 5.0 Hz, J1,2 = 5.1 Hz, H-1), 4.19 (dd, 1H,
J1,2 = 5.1 Hz, J2,3 = 8.8 Hz, H-2), 3.84 (ddd, 1H,
J5,6eq = 5.6 Hz, J4,5 = 10.0 Hz, J5,6ax = 10.2 Hz, H-5),
3.58 (dd, 1H, J2,3 = 8.8 Hz, J3,4 = 9.7 Hz, H-3), 3.26
(dd, 1H, J3,4 = 9.7 Hz, J4,5 = 10.0 Hz, H-4), 2.69 (ddd,
1H, J1,6eq = 2.4 Hz, J5,6eq = 5.6 Hz, J6gem = 15.4 Hz, H-
6eq), 1.71 (ddd, 1H, J1,6ax = 5.0 Hz, J5,6ax = 10.2 Hz,
J6gem = 15.4 Hz, H-6ax), 1.51, 1.45, 1.44, and 1.35 (4s,
each 3H, 2 · CMe2).
To a solution of (ꢀ)-vibo-quercitol10 (7.97 g, 49 mmol)
in DMF (120 mL) were added 2,2-dimethoxypropane
(80 mL, 0.77 mol) and p-toluenesulfonic acid monohy-
drate (0.93 g, 4.9 mmol), and the mixture was stirred
for 90 min at room temperature. TLC showed a forma-
tion of two components with similar Rf values
(Rf = ꢁ0.5, 2-butanone–toluene 1:2). After neutraliza-
tion with triethylamine, the mixture was evaporated
and the residue was chromatographed on a column
of silica gel (300 g, acetone–hexane 1:2) to give an
inseparable mixture (10.2 g, 86.3%) of 1,2:3,4- and
1,2:4,5-di-O-isopropylidene-L-vibo-quercitol (7 and 8)
as a syrup.
4.4. 5-Azido-5-deoxy-1,2-O-isopropylidene-L-vibo-querci-
tol (12) and 5-azido-5-deoxy-L-vibo-quercitol (13)
To a stirred solution of the azide 11 (5.35 g, 19.9 mmol) in
methanol (100 mL) was added gradually TsOHÆH2O un-
til pH of the mixture being adjusted to ꢁ4. After neutral-
ization with triethylamine, the mixture was evaporated,
and the residue was chromatographed on a column of sil-
ica gel (500 g, acetone–hexane 1:3 ! 1:1, methanol) to
give 12 (3.22 g, 71%), as a crystalline solid, and 13
(0.71 g, 7%), together with 11 (0.55 g, 10%) recovered.
To a 113 mg (0.46 mmol) portion of the mixture in pyr-
idine (2.3 mL) was added DMAP (5.6 mg, 0.05 mmol)
and sulfuryl chloride (112 lL, 1.4 mmol) at 0 ꢁC, and
the mixture was stirred for 21 h at room temperature.
The reaction mixture was roughly co-evaporated with
toluene to dryness, and the residue was diluted with
chloroform (24 mL). The solution was washed with sat-
urated aqueous sodium thiosulfate, water, and saline,
dried, and evaporated. The residue was chromato-
graphed on a column of silica gel (14 g, acetone–hexane
1:60 ! 1:30) to give the chlorides 9 (49 mg, 40%) and 10
(70 mg, 58%) as crystalline solid.
For 12: Rf = 0.23 (acetone–hexane 1:2); [a]D ꢀ74 (c 0.91,
CHCl3); 1H NMR (300 MHz, CDCl3): d 4.39 (m, 1H, H-
1), 3.96 (dd, 1H, J1,2 = 5.1 Hz, J2,3 = 7.7 Hz, H-2), 3.61
(m, 1H, H-5), 3.59 (dd, 1H, J2,3 = 7.7 Hz,
J3,4 = 9.8 Hz, H-3), 3.30 (t, 1H, J3,4 = J4,5 = 9.8 Hz, H-
4), 2.40 (ddd, 1H, J1,6eq = 2.1 Hz, J5,6eq = 4.8 Hz,
J6gem = 15.0 Hz, H-6eq), 1.76 (ddd, 1H, J1,6ax = 3.7 Hz,
J5,6ax = 11.7 Hz, J6gem = 15.0 Hz, H-6ax), 1.58 and 1.42
(2s, each 3H, CMe2); ITMS-APCI (negative mode): m/
z 228 [MꢀH]ꢀ.
21
D
For 9: Rf = 0.37 (acetone–hexane 1:5); ½aꢂ ꢀ8.5 (c 0.57,
1
CHCl3); H NMR (300 MHz, CDCl3): d 4.50 (ddd, 1H,
J5,6eq = 2.3 Hz, J4,5 = 3.1 Hz, J5,6ax = 5.2 Hz, H-5), 4.38
(ddd, 1H, J1,6eq = 1.8 Hz, J1,6ax = 4.9 Hz, J1,2 = 5.1 Hz,
H-1), 4.22 (m, 2H, H-2, H-3), 3.48 (dd, 1H,
J4,5 = 3.1 Hz, J3,4 = 9.2 Hz, H-4), 2.69 (ddd, 1H,
J1,6eq = 1.8 Hz, J5,6eq = 2.3 Hz, J6gem = 16.6 Hz, H-
6eq), 2.30 (ddd, 1H, J1,6ax = 4.9 Hz, J5,6ax = 5.2 Hz,
J6gem = 16.6 Hz, H-6ax), 1.56, 1.49, 1.47, and 1.36 (4 s,
each 3H, 2 · CMe2).
20
For 13: Rf = 0.01 (acetone–hexane 1:2); ½aꢂ ꢀ13 (c
D
0.44, MeOH); 1H NMR (300 MHz, CD3OD): d 4.05
(m, 1H, H-1), 3.62 (m, 1H, H-5), 3.58 (t, 1H,
J3,4 = J4,5 = 9.5 Hz, H-4), 3.43 (dd, 1H, J1,2 = 2.5 Hz,
J2,3 = 9.7 Hz, H-2), 3.32 (dd, 1H, J3,4 = 9.5 Hz, J2,3
9.7 Hz, H-3), 2.09 (ddd, 1H, J1,6eq = 3.8 Hz, J5,6eq = 4.1
Hz, J6gem = 14.2 Hz, H-6eq), 1.53 (ddd, 1H, J5,6ax
=
=
ꢁ1.5 Hz, J1,6ax = J6gem = 14.2 Hz, H-6ax); ITMS-APCI
19
D
For 10: Rf = 0.40 (acetone–hexane 1:5); ½aꢂ +74 (c 0.61,
(negative mode): m/z 188 [MꢀH]ꢀ.
1
CHCl3); H NMR (300 MHz, CDCl3): d 4.62 (dd, 1H,
J3,4 = 2.9 Hz, J2,3 = 5.6 Hz, H-3), 4.49 (m, 1H,
J1,6ax = 5.5 Hz, H-1), 4.35 (m, 2H, H-2, H-5), 3.46 (dd,
1H, J3,4 = 2.9 Hz, J4,5 = 9.2 Hz, H-4), 2.65 (m, 1H,
J5,6eq = 5.0 Hz, J6gem = 14.2 Hz, H-6eq), 1.87 (ddd, 1H,
J1,6ax = 5.5 Hz, J5,6ax = 11.8 Hz, J6gem = 14.2 Hz, H-
6ax), 1.64, 1.47, 1.45, and 1.36 (4s, each 3H, 2 · CMe2).
4.5. 5-Azido-5-deoxy-1,2-O-isopropylidene-3-O-tosyl-
(14), 4-O-tosyl- (15), and 3,4-di-O-tosyl-L-vibo-quercitols
(16)
A mixture of 12 (423 mg, 1.84 mmol) in pyridine
(8.5 mL) was added tosyl chloride (1.76 g, 9.22 mmol)

S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
4311
at ꢀ18 ꢁC, and the mixture was stirred for 4 days at the
similar temperature. The reaction was quenched by addi-
tion of methanol (1 mL) and then the mixture was co-
evaporated with toluene. The residue was diluted with
ethyl acetate (90 mL), and the solution was washed with
water and saline, dried, and evaporated. The residue was
chromatographed on a silica gel column (70 g, EtOAc–
toluene 1:30 ! 1:10) to give 14 (305 mg, 43%), 15
(208 mg, 29%), and 16 (150 mg, 15%) as crystalline solid.
J2,3 = 7.3 Hz, H-2), 3.88 (ddd, 1H, J5,6eq = 4.6 Hz,
J4,5 = 9.8 Hz, J5,6ax = 12.0 Hz, H-5), 2.48 (ddd, 1H,
J1,6eq = 2.1 Hz, J5,6eq = 4.6 Hz, J6gem = 15.4 Hz, H-6eq),
2.43 (s, 3H, PhCH3), 2.14 (s, 3H, Ac), 1.72 (ddd, 1H,
J1,6ax = 3.7 Hz, J5,6ax = 12.0 Hz, J6gem = 15.4 Hz, H-
6ax), 1.57 and 1.33 (2s, each 3H, CMe2); ITMS-ESI (po-
sitive mode): m/z 448 [M+Na]+.
4.7. 5-Acetamido-4-O-acetyl-5-deoxy-1,2-O-isopropylid-
ene-3-O-tosyl-L-vibo-quercitol (18)
20
For 14: Rf = 0.50 (acetone–hexane 1:6); ½aꢂ ꢀ76.5 (c
D
1.71, CHCl3); 1H NMR (300 MHz, CDCl3): d 7.88–
7.33 (m, 4H, 2 · Ph), 4.58 (dd, 1H, J2,3 = 7.4 Hz,
J3,4 = 9.8 Hz, H-3), 4.27 (m, 1H, H-1), 4.00 (dd, 1H,
J1,2 = 5.2 Hz, J2,3 = 7.4 Hz, H-2), 3.74 (ddd, 1H,
J5,6eq = 4.6 Hz, J4,5 = 9.5 Hz, J5,6ax = 11.8 Hz, H-5),
3.52 (dd, 1H, J4,5 = 9.5 Hz, J3,4 = 9.8 Hz, H-4), 2.45 (s,
3H, PhCH3), 2.29 (ddd, 1H, J1,6eq = 2.2 Hz,
J5,6eq = 4.6 Hz, J6gem = 15.5 Hz, H-6eq), 1.62 (ddd, 1H,
J1,6ax = 3.8 Hz, J5,6ax = 11.8 Hz, J6gem = 15.5 Hz, H-
6ax), 1.39 and 1.29 (2s, each 3H, CMe2); ITMS-ESI (po-
sitive mode): m/z 406 [M+Na]+.
A solution of 17 (1.47 g, 3.46 mmol) in ethanol (25 mL)
was hydrogenated in the presence of Raney Ni (two
spoonful) and acetic anhydride (0.66 mL) under atmo-
spheric pressure of hydrogen for 4.5 h at room temper-
ature. A catalyst was removed by filtration and the
filtrate was evaporated to dryness. The residue was chro-
matographed on a column of silica gel (160 g, acetone–
hexane 2:3) to give 18 (1.57 g, 99%) as a crystalline solid,
19.5
Rf = 0.14 (acetone–toluene 1:4); ½aꢂ
ꢀ45.5 (c 1.5,
D
1
CHCl3); H NMR (300 MHz, CDCl3): d 5.58 (d, 1H,
J5,NH = 8.5 Hz, NH), 4.92 (dd, 1H, J2,3 = 7.3 Hz,
J3,4 = 10.3 Hz, H-3), 4.79 (t, 1H, J3,4 = J4,5 = 10.3 Hz,
H-4), 4.42 (m, 1H, H-5), 4.27 (m, 1H, H-1), 4.02 (dd,
1H, J1,2 = 5.1 Hz, J2,3 = 7.3 Hz, H-2), 2.55 (ddd, 1H,
J1,6eq = 2.0 Hz, J5,6eq = 4.6 Hz, J6gem = 15.1 Hz, H-6eq),
1.63 (m, 1H, H-6ax), 2.45 (s, 3H, PhCH3), 2.05
and 1.98 (2s, each 3H, 2 · Ac), 1.51 and 1.40 (2s,
each 3H, CMe2); ITMS-ESI (positive mode): m/z 464
[M+Na]+.
20
For 15: Rf = 0.34 (acetone–hexane 1:6); ½aꢂ +5.5 (c
D
6.52, CHCl3); 1H NMR (300 MHz, CDCl3): d 7.89–
7.36 (m, 4H, Ph), 4.35 (dd, 1H, J3,4 = 9.5 Hz,
J4,5 = 10.0 Hz, H-4), 4.31 (m, 1H, H-1), 4.04 (dd, 1H,
J1,2 = J2,3 = 6.3 Hz, H-2), 3.81 (dd, 1H, J2,3 = 6.3 Hz,
J3,4 = 9.5 Hz, H-3), 3.74 (m, 1H, H-5), 2.46 (s, 3H,
PhCH3), 2.45 (m, 1H, H-6eq), 1.74 (ddd, 1H,
J1,6ax = 3.7 Hz, J5,6ax = 12.2 Hz, J6gem = 15.6 Hz, H-
6ax), 1.49 and 1.35 (2s, each 3H, CMe2); ITMS-ESI (po-
sitive mode): m/z 406 [M+Na]+.
4.8. 5-Acetamido-5-deoxy-1,2-O-isopropylidene-L-talo-
quercitol (19) and 5-acetamido-5-deoxy-1,2-O-isopropyl-
idene-L-allo-quercitol (22)
20
D
For 16: Rf = 0.59 (acetone–hexane 1:6); ½aꢂ +0.9 (c 1.5,
1
CHCl3); H NMR (300 MHz, CDCl3): d 7.84–7.32 (m,
8H, 2 · Ph), 4.76 (dd, 1H, J2,3 = 6.2 Hz, J3,4 = 7.4 Hz,
H-3), 4.44 (dd, 1H, J3,4 = 7.5 Hz, J4,5 = 8.5 Hz, H-4),
4.32 (m, 1H, H-1), 4.17 (dd, 1H, J1,2 = 6.1 Hz,
J2,3 = 6.2 Hz, H-2), 3.78 (ddd, 1H, J5,6eq = 4.4 Hz,
J4,5 = 8.5 Hz, J5,6ax = 12.3 Hz, H-5), 2.44 (s, 6H,
(a) A mixture of 18 (1.25 g, 2.74 mmol), anhydrous
NaOAc (1.12 g, 13.7 mmol), and 90% aqueous 2-meth-
oxyethanol (25 mL) was stirred for 2 days at 110 ꢁC,
and then co-evaporated with ethanol and toluene. The
residue was triturated with acetone (50 mL) and an
insoluble material was removed by filtration. The filtrate
was evaporated and the residue was chromatographed
on a column of silica gel (60 g, MeOH–CHCl3 1:10) to
give ca. 10:1 mixture (0.56 g, 83%) of the diols 19 and
22 as a syrup.
2 · PhCH3),
2.29
(ddd,
1H,
J1,6eq = 2.4 Hz,
J5,6eq = 4.4 Hz, J6gem = 14.9 Hz, H-6eq), 1.77 (ddd, 1H,
J1,6ax = 3.9 Hz, J5,6ax = 12.3 Hz, J6gem = 14.9 Hz, H-
6ax), 1.42 and 1.24 (2s, each 3H, CMe2); ITMS-ESI (po-
sitive mode): m/z 560 [M+Na]+.
(b) A mixture of 18 (49 mg, 0.11 mmol), anhydrous
NaOAc (44 mg, 0.53 mmol), and 90% aqueous DMF
(1.0 mL) was stirred for 3 days at 115 ꢁC. The reaction
mixture was processed similarly and the products were
chromatographed on a column of silica gel to give to
give ca. 1:10 mixture of 19 and 22 (12 mg, 35%), together
with 18 (10 mg, 20%) recovered.
4.6. 4-O-Acetyl-5-azido-5-deoxy-1,2-O-isopropylidene-3-
O-tosyl-L-vibo-quercitol (17)
Compound 14 (1.33 g, 3.47 mmol) was treated with ace-
tic anhydride (6.7 mL) in pyridine (13 mL) for 14 h at
room temperature, and then co-evaporated with tolu-
ene. The residue was dissolved in ethyl acetate
(300 mL) and the solution was washed with water and
saline, dried, and evaporated. The product was chro-
matographed on a column of silica gel (150 g, ace-
Compounds 19 and 22 were further characterized by
converting into the corresponding di-O-acetyl deriva-
tives 20 and 23 in the conventional manner.
tone–hexane 1:15 ! 1:5) to give 17 (1.44 g, 97%) as
For 20: Rf = 0.52 (MeOH–CHCl3 1:5); 1H NMR
(300 MHz, CDCl3): d 5.66 (d, 1H, J5,NH = 8.3 Hz,
NH), 5.30 (dd, 1H, J4,5 = 5.1 Hz, J3,4 = 7.4 Hz, H-4),
5.19 (dd, 1H, J2,3 = 3.5 Hz, J3,4 = 7.4 Hz, H-3), 4.69
(m, 1H, H-5), 4.42 (m, 2H, H-1, H-2), 2.06 (m, 1H, H-
20
D
a syrup, Rf = 0.61 (acetone–toluene 1:4); ½aꢂ ꢀ63
1
(c 1.7, CHCl3); H NMR (300 MHz, CDCl3): d 7.88–
7.33 (m, 4H, Ph), 4.96 (t, 1H, J4,5 = 9.8 Hz, J3,4 = 10.5
Hz, H-4), 4.84 (dd, 1H, J2,3 = 7.3 Hz, J3,4 = 10.5 Hz,
H-3), 4.31 (m, 1H, H-1), 4.05 (dd, 1H, J1,2 = 5.1 Hz,

4312
S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
6eq), 1.95 (m, 1H, H-6ax), 2.16, 2.06 and 1.95 (3 s, each
3H, 3 · Ac), 1.57 and 1.30 (2s, each 3H, CMe2); ITMS-
ESI (positive mode): m/z 352 [M+Na]+.
4.11. 5-Amino-5-deoxy-L-talo-quercitol (4)
A solution of 24 (30 mg, 0.08 mmol) in ethanol (1 mL)
containing a drop of 1 M hydrochloric acid was hydro-
genated in the presence of 10% Pd/C (a spoonful) in
atmospheric pressure for 14 h at room temperature. A
catalyst was removed by filtration and the filtrate was
evaporated. The residual product was treated with 6 M
hydrochloric acid for 15 h at 80 ꢁC, and then evapo-
rated. The product was purified by a column of Dow-
ex-50 W · 2 (1.2 g, 1% aqueous ammonia) to give the
For 23: Rf = 0.52 (MeOH–CHCl3 1:5); 1H NMR
(300 MHz, CDCl3): d 5.59 (d, 1H, J5,NH = 8.3 Hz,
NH), 5.36 (dd, 1H, J3,4 = 3.8 Hz, J2,3 = 4.3 Hz, H-3),
4.95 (dd, 1H, J3,4 = 3.8 Hz, J4,5 = 9.3 Hz, H-4), 4.53
(m, 1H, H-5), 4.41 (m, 1H, H-1), 4.31 (dd, 1H,
J2,3 = 4.3 Hz, J1,2 = 6.2 Hz, H-2), 2.42 (ddd, 1H,
J1,6eq = 2.4 Hz, J5,6eq = 4.9 Hz, J6gem = 14.9 Hz, H-6eq),
2.16, 2.06 and 1.95 (3s, each 3H, 3 · Ac), 1.71 (ddd,
1H, J1,6ax = 4.6 Hz, J5,6ax = 11.2 Hz, J6gem = 14.9 Hz,
H-6ax), 1.54 and 1.34 (2s, each 3H, CMe2); ITMS-ESI
(positive mode): m/z 352 [M+Na]+.
free bose 4 (13.4 mg, ꢁ100%) as a syrup, Rf = 0.53
20
(H2O–AcOH–n-BuOH 1:1:2); ½aꢂ ꢀ35 (c 0.53, H2O);
D
1H NMR (300 MHz, D2O): d 3.86 (m, 2H, H-1, H-2),
3.70 (dd, 1H, J4,5 = 3.5 Hz, J3,4 = 8.3 Hz, H-4), 3.64
(dd, 1H, J2,3 = ꢁ3 Hz, J3,4 = 8.3 Hz, H-3), 3.32 (m,
1H, H-5), 1.77 (ddd, 1H, J5,6ax = 4.0 Hz, J1,6ax = 9.9 Hz,
J6gem = 13.8 Hz, H-6ax), 1.64 (ddd, 1H, J5,6eq = 4.6 Hz,
J1,6eq = 5.1 Hz, J6gem = 13.8 Hz, H-6eq); ITMS-ESI (po-
sitive mode): m/z 164 [M+H]+.
4.9. 5-Acetamido-3,4-di-O-benzyl-5-deoxy-1,2-O-isopro-
pylidene-L-talo-quercitol (21)
To a 390 mg (1.60 mmol) portion of the ca. 10:1 mixture
of 19 and 22 dissolved in DMF (1.0 mL) was added
NaH (60% in oil, 0.26 g, 6.4 mmol), and the mixture
was stirred for 30 min at 0 ꢁC. Benzyl bromide
(570 lL, 4.8 mmol) was added to the mixture and it
was stirred for 3 h at room temperature. After being
quenched by addition of MeOH, the reaction mixture
was evaporated and further co-evaporated with n-BuOH
and toluene. The residue was diluted with ethyl acetate
and the solution was washed with water and saline,
dried, and evaporated. The residual product was chro-
matographed on a column of silica gel (70 g, acetone–
hexane 1:4 ! 1:3) to give mainly the dibenzyl ether 21
4.12. 5-Acetamido-3,4-di-O-benzyl-5-deoxy-1-O-methyl-
(25), 2-O-methyl- (26), and 1,2-di-O-methyl-L-talo-
quercitols (27)
To a mixture of 24 (81 mg, 0.21 mmol) and acetonitrile
(1.6 mL), were added iodomethane (2.6 mL, 42 mmol)
and silver oxide (146 mg, 0.63 mmol), and the mixture
was refluxed for 4 h, and then evaporated to dryness.
The residue was chromatographed on a column on silica
gel (11 g, acetone–toluene 1:6) to give the dimethyl ether
27 (28.4 mL, 33%) as a crystalline solid, and an insepa-
rable mixture (44 mg, 52%) of the methyl ethers 25 and
26. The latter mixture was treated with acetic anhydride
(0.5 mL) in pyridine (1 mL) for 13 h at room tempera-
ture. The reaction mixture was processed in the usual
manner and the products were chromatographed on a
column of silica gel (5 g, acetone–toluene 1:8) to give
the respective O-acetyl methyl ethers 28 (17 mg, 34%)
and 29 (24 mg, 49%) as a syrup.
(494 mg, 73%) as a syrup, Rf = 0.53 (acetone–toluene
22
1:2); ½aꢂ +41 (c 0.63, CHCl3); 1H NMR (300 MHz,
D
CDCl3): d 7.41–7.23 (m, 10H, Ph), 5.65 (d, 1H,
J5,NH = 7.8 Hz, NH), 4.79 and 4.66 (ABq, Jgem
=
12.2 Hz), and 4.66 and 4.44 (ABq, Jgem = 11.5 Hz)
(2 · CH2Ph), 4.58 (m, 1H, H-5), 4.41 (dd, 1H,
J2,3 = 3.8 Hz, J1,2 = 6.3 Hz, H-2), 4.32 (m, 1H, H-1),
3.85 (dd, 1H, J4,5 = 5.3 Hz, J3,4 = 5.8 Hz, H-4), 3.67
(dd, 1H, J2,3 = 3.8 Hz, J3,4 = 5.8 Hz, H-3), 1.88 (m,
1H, H-6eq), 1.85 (s, 3H, Ac), 1.70 (m, 1H, H-6ax),
1.56 and 1.34 (2s, each 3H, CMe2); ITMS-ESI (positive
mode): m/z 448 [M+Na]+.
21
D
For 27: Rf = 0.56 (acetone–toluene 1:1); ½aꢂ +4.7 (c
1
0.37, CHCl3); H NMR (300 MHz, CDCl3): d 7.21 (m,
10H, 2 · Ph), 5.51 (br s, 1H, NH), 4.79 and 4.72
(ABq, Jgem = 12.2 Hz), and 4.55 and 4.44 (ABq,
Jgem = 11.3 Hz) (2 · CH2Ph), 4.39 (m, 1H, H-5), 3.81
(m, 1H, H-4), 3.68 (br s, 1H, H-2), 3.59 (m, 1H, H-3),
3.52 (s, 3H, OMe), 3.43 (m, 1H, H-1), 3.37 (s, 3H,
OMe), 2.05 (m, 2H, 2 · H-6), 1.90 (s, 3H, Ac); ITMS-
ESI (positive mode): m/z 414 [M+H]+.
4.10. 5-Acetamido-3,4-di-O-benzyl-5-deoxy-L-talo-querc-
itol (24)
A mixture of 21 (13 mg, 0.03 mmol) and 80% aqueous
acetic acid (0.4 mL) was stirred for 1.5 h at 50 ꢁC, and
then co-evaporated with ethanol and toluene. The resi-
due was chromatographed on a column of silica gel
(1 g, MeOH–CHCl3 1:20) to give the diol 24 (12 mg,
ꢁ100%) as a syrup, Rf = 0.65 (MeOH–CHCl3 1:20);
21
D
For 28: Rf = 0.37 (acetone–toluene 1:1); ½aꢂ ꢀ9.4 (c
1
0.24, CHCl3); H NMR (300 MHz, CDCl3): d 7.32 (m,
10H, Ph), 5.58 (br s, 1H, H-2), 5.51 (d, 1H,
J5,NH = 5.9 Hz, NH), 4.77 and 4.58 (ABq,
22
1
½aꢂ +69 (c 0.51, CHCl3); H NMR (300 MHz, CDCl3):
D
d 7.42–7.20 (m, 10H, 2 · Ph), 5.45 (d, 1H, J5,NH
=
8.5 Hz, NH), 4.71 and 4.60 (ABq, Jgem = 11.5 Hz), and
4.56 and 4.31 (ABq, Jgem = 11.8 Hz) (2 · CH2Ph), 4.48
(m, 1H, H-5), 3.93–3.77 (m, 4H, H-1, H-2, H-3, H-4),
3.08 and 2.82 (2br s, each 1H, 2 · OH), 2.00 (ddd, 1H,
J5,6eq = 3.3 Hz, J1,6eq = 3.8 Hz, J6gem = 13.5 Hz, H-6eq),
1.83 (s, 3H, Ac), 1.79 (m, 1H, H-6ax); ITMS-ESI (posi-
tive mode): m/z 408 [M+Na]+.
Jgem = 12.0 Hz),
and
4.60
and
4.45
(ABq,
Jgem = 11.2 Hz) (2 · CH2Ph), 4.43 (m, 1H, H-5), 3.77
(dd, 1H, J4,5 = 4.4 Hz, J3,4 = 7.8 Hz, H-4), 3.65 (dd,
1H, J2,3 = 2.9 Hz, J3,4 = 7.8 Hz, H-3), 3.44 (m, 1H, H-
1), 3.36 (s, 3H, OMe), 2.20 (m, 1H, H-6eq), 2.13 and
1.92 (2s, each 3H, 2 · Ac), 1.86 (m, 1H, H-6ax);
ITMS-ESI (positive mode): m/z 464 [M+Na]+.

S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
4313
21
D
For 29: ½aꢂ +23.5 (c 0.19, CHCl3); 1H NMR (300 MHz,
in our experimental conditions. a-Fucosidase (bovine
kidney) was assayed using p-nitrophenyl a-L-fucopyr-
anoside (4 mM) as a substrate at pH 5.5 (30 mM
CH3COONa) at 37 ꢁC. a-Galactosidase (green coffee
beans) was assayed using p-nitrophenyl a-D-galactopy-
ranoside as a substrate at pH 6.8 (20 mM, phosphate
buffer) at 27 ꢁC. b-Galactosidase (bovine liver) assayed
similarly using p-nitrophenyl b-D-galactopyranoside at
37 ꢁC. a-Glucosidase (Bakerꢀs yeast) was assayed simi-
larly using p-nitrophenyl a-D-glucopyranoside (4 mM)
as a substrate at pH 5.5 (30 mM CH3COONa) at
27 ꢁC. b-Glucosidase (almond) was assayed similarly
using p-nitrophenyl b-D-glucopyranoside (4 mM) as a
substrate at pH 5.5 (30 mM CH3COONa) at 27 ꢁC. b-
Glucosaminidase (bovine kidney) was assayed using p-
nitrophenyl N-acetyl-b-D-glucosaminide (4 mM) as a
substrate at pH 5.5 (30 mM CH3COONa) at 37 ꢁC. a-
Mannosidase was assayed using p-nitrophenyl a-D-
mannopyranoside (4 mM) as a substrate at pH 5.5
(30 mM CH3COONa) at 27 ꢁC.
CDCl3): d 7.33 (m, 10H, 2 · Ph), 5.49 (d, 1H,
J5,NH = 6.1 Hz, NH), 5.17 (m, 1H, H-1), 4.79 and 4.68
(ABq, Jgem = 12.2 Hz), and 4.60 and 4.45 (ABq,
Jgem = 11.7 Hz)
(2 · CH2Ph),
3.82
(dd,
1H,
J4,5 = 3.7 Hz, J3,4 = 6.4 Hz, H-4), 3.77 (m, 1H, H-2),
3.66 (br s, 1H, H-3), 3.45 (s, 3H, OMe), 1.96 (m, 2H,
2 · H-6), 2.06 and 1.87 (2s, each 3H, 2 · Ac); ITMS-
ESI (positive mode): m/z 464 [M+Na]+.
4.13. 5-Amino-5-deoxy-1-O-methyl-L-talo-quercitol (5)
Compound 25 (17 mg, 0.04 mmol) was hydrogenated as
in the preparation of 4, and the product was treated with
1 M aqueous Ba(OH)2 (1 mL) for 20 h at 80 ꢁC. After
neutralization with CO2, an insoluble material was re-
moved by filtration and the filtrate was evaporated.
The residue was similarly purified by a column of Dow-
ex 50 W · 2 (H+) resin (1 g, 1% aqueous ammonia) to
give 5 (7 mg, ꢁ100%) as a syrup, Rf = 0.26 (H2O–
20
AcOH–n-BuOH 1:1:4); ½aꢂ ꢀ10 (c 0.24, H2O); 1H
NMR (300 MHz, D2O): d D4.00 (br s, 1H, H-2), 3.75
(dd, 1H, J2,3 = 3.7 Hz, J3,4 = 7.8 Hz, H-3), 3.62 (dd,
1H, J4,5 = ꢁ3 Hz, J3,4 = 7.8 Hz, H-4), 3.52 (m, 1H, H-
1), 3.29 (m, 1H, H-5), 3.22 (s, 3H, OMe), 1.85 and
1.71 (2m, each 1H, 2 · H-6); ITMS-ESI (positive mode):
m/z 178 [M+H]+.
Test compound was added to each reaction mixture de-
scribed above and it was incubated for 1 h at 27 or
37 ꢁC. After 1 h, the reaction was quenched by addition
of three volumes of aqueous 0.2 M sodium carbonate,
and the absorbance of the liberated p-nitrophenol was
measured at 410 nm. The percentage inhibition was cal-
culated by the formula BA · 100, where A is the p-nitro-
phenol liberated by the enzyme without an inhibitor and
B is that with an inhibitor.
4.14. 5-Amino-5-deoxy-2-O-methyl-L-talo-quercitol (30)
A solution of 26 (17 mg, 0.04 mmol) in ethanol (1 mL)
containing a drop of 1 M hydrochloric acid was hydro-
genated in the presence of 10% Pd/C (a spoonful) under
an atmospheric pressure of hydrogen for 16 h at room
temperature. The product was purified by a column of
Dowex-50 W · 2 (1 g, 1% aqueous ammonia) to give
Acknowledgements
The authors sincerely thank Drs. Atsushi Takahashi
and Akihiro Tomoda (Hokko Chemical Industry Co.
Ltd, Atsugi, Japan) for the biological assays, and
Ms. Miki Kanto for her assistance in preparing this
manuscript.
the free base 30 (7 mg, ꢁ100%) as a syrup: Rf = 0.26
20
(H2O–AcOH–n-BuOH 1:1:4); ½aꢂ ꢀ38.5 (c 0.12, H2O);
D
1H NMR (300 MHz, D2O): d 4.01 (br s, 1H, H-5), 3.79
(br s, 1H, H-3), 3.74 (br s, 1H, H-2), 3.44 (br s, 1H, H-
4), 3.35 (s, 3H, OMe), 3.29 (m, 1H, H-1), 1.72 (m, 2H,
2 · H-6); ITMS-ESI (positive mode): m/z 178 [M+H]+.
References and notes
4.15. 5-Amino-5-deoxy-1,2-di-O-methyl-L-talo-quercitol
(31)
1. Aoyagi, T.; Yamamoto, T.; Kojiri, K.; Morishima, H.;
Nagai, M.; Hamada, M.; Takeuchi, T.; Umezawa, H. J.
Antibiot. 1989, 42, 883–889.
2. Morishima, H.; Kojiri, K.; Yamamoto, T.; Aoyagi, T.;
Nakamura, H.; Iitaka, Y. J. Antibiot. 1989, 42, 1008–
1011.
3. A part of this work was previously reported: Ogawa, S.;
Morikawa, T. Bioorg. Med. Chem. Lett. 2000, 10, 1047–
1050.
4. S. Ogawa, T. Morikawa, Eur. J. Org. Chem., submitted
for publication.
5. Legler, G. Adv. Carbohydr. Chem. Biochem. 1990, 48, 319–
384.
6. Legler, G. In Carbohydrate Mimics; Chapleur, Y., Ed.;
Wiley-VCH: Weinheim, 1998, pp 463–490.
7. Lock, G. C.; Fotsch, C. H.; Wong, C. H. Acc. Chem. Res.
1993, 26, 182–190.
The dibenzyl ether 27 (20 mg, 0.05 mmol) was hydro-
genolyzed and subsequently hydrolyzed as in the prepa-
ration of 4. The product was also purified by a column
of Dowex-50 W · 2 resin to give the dimethyl ether 31
(4.3 mg, 46%) as a syrup, Rf = 0.36 (H2O–AcOH–n-
20
D
BuOH 1:1:4); ½aꢂ ꢀ85 (c 0.04, H2O); 1H NMR
(300 MHz, D2O): d 3.67 (m, 4H, H-2, H-3, H-4, H-5),
3.24 and 3.19 (2s, each 3H, 2 · OMe), ꢁ3.2 (m, 1H,
H-1), 1.79 and 1.64 (m, each 1H, 2 · H-6); ITMS-ESI
(positive mode): m/z 192 [M+H]+.
5. Biological assay
8. Ogawa, S.; Sekura, R.; Maruyama, A.; Yuasa, H.;
Hashimoto, H. Eur. J. Org. Chem. 2000, 2089–2093.
9. Ogawa, S.; Maruyama, A.; Odagiri, T.; Yuasa, H.;
Hashimoto, H. Eur. J. Org. Chem. 2001, 967–974.
Determination of IC50: The IC50 values were determined
for all inhibitors and correspond to the inhibition con-
centration required for 50% inhibition of the enzyme

4314
S. Ogawa et al. / Bioorg. Med. Chem. 13 (2005) 4306–4314
10. Takahashi, A.; Kanbe, K.; Tamamura, T.; Sato, K.
Anticancer Res. 1999, 3807.
12. The reaction conditions have not been optimized yet in
order to reduce the elimination products.
11. Ogawa, S.; Uetsuki, S.; Tezuka, Y.; Morikawa, T.;
Takahashi, A.; Sato, K. Bioorg. Med. Chem. Lett. 1999,
9, 1493–1498.
13. The reaction conditions have not been optimized yet.
14. All biological assays were carried out in a standard
manner by Dr. A. Tomoda.