L-Idoseptanosides: Substrates of D-glucosidases?

Article abstract of DOI:10.1016/j.tetasy.2005.12.007

Based on the reported glycosidase inhibitory activities of 1,6-dideoxy-1,6-imino-l-iditol, the corresponding aryl glycosides 4-nitrophenyl α- and β-l-idoseptanoside 7 and 8 were synthesised as possible glycosidase substrates. Despite their inherently larg

Full text of DOI:10.1016/j.tetasy.2005.12.007

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 17 (2006) 234–239
L-Idoseptanosides: substrates of D-glucosidases?
Andreas Tauss,a Andreas J. Steiner,a Arnold E. Stutz,a,* Chris A. Tarling,b
¨
Stephen G. Withersb and Tanja M. Wrodnigga
aGlycogroup, Institut fu¨r Organische Chemie der Technischen Universita¨t Graz, Stremayrgasse 16, A-8010 Graz, Austria
bDepartment of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, BC, Canada V6T 1Z1
Received 9 November 2005; accepted 8 December 2005
Available online 23 January 2006
Abstract—Based on the reported glycosidase inhibitory activities of 1,6-dideoxy-1,6-imino-L-iditol, the corresponding aryl glyco-
sides 4-nitrophenyl a- and b-L-idoseptanoside 7 and 8 were synthesised as possible glycosidase substrates. Despite their inherently
larger size, these septanosides were indeed shown to be glycosidase substrates, albeit weak ones. In addition, these two substrate
analogues 7 and 8 also demonstrated a remarkable degree of selectivity for b- and a-glucosidases, respectively.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
H
N
H
N
N
HO
HO
HO
Iminosugars are well known and often potent inhibitors
of glycosidases. Their efficacy as inhibitors has often
been ascribed to their mimicry of bound protonated
substrates or of reaction transition states. However, at
least in the case of deoxynojirimycin and castanosper-
mine analogues, more careful kinetic analysis with Agro-
bacterium sp. b-glucosidase has cast doubt on the
hypothesis of transition state mimicry.1 Rather, it was
suggested that they act as ‘fortuitous’ inhibitors, binding
tightly to the enzyme by virtue of the electrostatic inter-
action of a positively charged ammonium centre with
one or more active site carboxylates as well as hydrogen
bonds between active site residues and the ligand. None-
theless, in cases where an analogy is clear, it is often the
case that inhibitor configurations match that of the opti-
mal substrate for the enzyme, consistent at least with
some mimicry of ground state protonated structures.
Thus, for example (Scheme 1), 1-deoxynojirimycin 1
and castanospermine 2, which have D-gluco-configured
hydroxyl groups around a piperidine scaffold, selectively
inhibit D-glucopyranosidases. Similarly, 1-deoxymanno-
jirimycin 3 inhibits D-mannosidases while the corre-
sponding D-galacto-configured iminoalditol 4 is a
strong inhibitor of D-galactosidases. Interestingly, over
the past two decades, it has also been recognised that
OH
HO
HO
OH
HO
OH
OH
OH
OH
1
2
3
H
N
H
N
H
N
HO
HO
OH
H
HO
H
HO
OH
HO
OH
OH
OH
HO
OH
4
5
6
Scheme 1.
many glycosidases are susceptible to inhibition by five-
and seven-membered ring iminoalditols, as well as suit-
ably substituted aminocyclopentanes whose structures
are not as obviously suitable for the purpose as the
pyranoid systems mentioned above. Such compounds
include the well-known 2,5-dideoxy-2,5-imino-D-manni-
tol 5, a very powerful glucosidase inhibitor2 and 1,6-
dideoxy-1,6-imino-L-iditol 6, reported to be a good
inhibitor of a wide range of various glycosidases, just
to mention two examples.3–5
Based on the aforementioned notion that close struc-
tural relatives of good substrates are frequently good
inhibitors, we envisaged that glycosides, which are struc-
turally related to such ‘unusually shaped’ but efficient
*
Corresponding author. Tel.: +43 316 873 8744; fax: +43 316 873
8740; e-mail: stuetz@tugraz.at
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.12.007

A. Tauss et al. / Tetrahedron: Asymmetry 17 (2006) 234–239
235
inhibitors might turn out to be suitable substrates for
the respective glycosidases.
propylidene-b-L-idofuranurono-6,3-lactone,9 or from
1,2-O-isopropylidene-3,5,6-tri-O-methanesulfonyl-a-D-
glucofuranose10) with EtSH gave known dithioacetal
9.11 Its 6-O-tritylation furnished the 1,6-diprotected
open-chain L-idose 10 in 87% overall yield from L-idose.
Per-O-benzoylation of the remaining secondary posi-
tions was found to be superior to O-acetylation in terms
of stabilities of intermediates in the following steps
yielding compound 11 (97%). Chemo- and regioselective
deprotection of the aldehyde function in the presence of
the trityl ether to give compound 12 could be achieved
with HgO/HgCl2 in an ice cold mixture of acetone/
water/acetonitrile 2:1:2 (v/v/v) and was found to be
superior to simultaneous 1,6-deprotection in terms of
overall yields and reproducibility. Conventional re-
moval of the trityl group from free aldehyde 12 gave
the anomeric mixture of 2,3,4,5-O-protected L-idosepta-
noses 13a/b in 88% yield. Reaction with DAST led
smoothly to a mixture of the corresponding glycosyl fluo-
rides 14a/b in 97% combined yield. These intermedi-
ates, in our hands, were found to be the most reliable
starting materials for the following step. Acid-catalysed
reaction with 4-nitrophenol gave the desired protected
septanosides 15 and 16, which were separated by chro-
Septanoses are usually only transient and elusive tauto-
meric forms of hexoses and higher carbon sugars, dis-
advantaged in the equilibria by their inherent high free
energies, which exceed those of the pyranose and fura-
nose forms by 3–8 kcal (8–33 kJ) per mole.6 In the case
of idose, a sugar whose configuration intrinsically dis-
favours the pyranose form, NMR experiments revealed
that the septanose was present as only 1.6% of the popu-
lation.7 Even in the case when the 5-hydroxyl was
methylated, thereby preventing the formation of a pyra-
nose tautomer, only 5.6% of the septanose tautomer was
observed. Not unexpectedly in the light of these findings,
O-glycosides of septanoses have not yet been found in
Nature. Nonetheless, based on the fact that compound
6 was reported to be an excellent glycosidase inhibi-
tor,3–5 the display of the functional groups they present
may allow for some suitable diastereomers of septano-
sides to mimic the substrates of various glycosidases.
Encouragingly, molecular modelling (Fig. 1) supported
this hypothesis, revealing that for a selection of low-
energy conformers the structure of the b-L-idoseptano-
side can be readily superimposed on the corresponding
a-D-glucopyranoside. The same was true for the ring
oxygen and the secondary hydroxyl groups in the corre-
sponding a-L-idoseptanoside, which all mapped well
onto the functional groups in a b-D-glucopyranoside.
In both anomers, O-5 of the septanoside was found to
be in close proximity to the position of O-6 of the natu-
ral substrates.
´
matography. Zemplen transesterification of the benzo-
ate esters gave free septanosides 7 (60%) and 8 (51%)
as final products. As previously predicted by the model-
ling study (Fig. 1), the pNP group of septanoside 7 was
observed to occupy an equatorial position, with an
indicative J1,2 of 6.8 Hz. In contrast, an axial orientation
of the pNP group in septanoside 8 was implied by the
considerably smaller J1,2 of 2.0 Hz.
To test this concept, 4-nitrophenyl (pNP) a-7 and b-L-
idoseptanosides 8 were prepared (Scheme 2) by standard
methods and probed as substrates of a range of
glycosidases.
2.2. Enzymatic studies
Compounds 7 and 8 were tested as substrates for a range
of a- and b-glycosidases (Table 1). Where an enzyme-
catalysed hydrolysis rate was observed, Michaelis–
Menten parameters were determined. The parameters
obtained were compared to those for hydrolysis of the
corresponding preferred pNP-pyranoside substrate for
that enzyme in order to estimate the catalytic efficiency
of turnover. The absence of an observed hydrolysis rate
for the septanoide substrates could be interpreted in one
of the two ways, either the septanoside can bind to the
2. Results and discussion
2.1. Syntheses
2.1.1. 4-Nitrophenyl a- and b-L-idoseptanosides. Con-
ventional reaction of L-idose,8 (prepared from 1,2-O-iso-
Figure 1. Left: Superposition of 4-nitrophenyl a-L-idoseptanoside 7 (grey) with 4-nitrophenyl b-D-glucopyranoside (green). Right: Superposition of
4-nitrophenyl b-L-idoseptanoside 8 (grey) with 4-nitrophenyl a-D-glucopyranoside (green).

236
A. Tauss et al. / Tetrahedron: Asymmetry 17 (2006) 234–239
SEt SEt
HO
HO
SEt
SEt
OH
OH
SEt
OH
OH
SEt
HO
HO
BzO
BzO
c
a
b
OBz
OBz
OH
OTrit
10
TritO
11
9
O
OH
OBz
F
O
O
BzO
BzO
OBz
OBz
d
e
f
BzO
BzO
BzO
BzO
OBz
OBz
OBz
TritO
12
13
14
OpNP
OBz
OpNP
OBz
O
O
BzO
BzO
BzO
BzO
+
OBz
OBz
15
16
g
g
OpNP
OH
OpNP
OH
O
O
HO
HO
+
HO
HO
OH
OH
7
8
Scheme 2. Reagents and conditions: (a) tritylCl, pyridine; (b) BzCl, pyridine; (c) HgO, HgCl2, MeCN/acetone/H2O; (d) BF3ÆEt2O, Et2O; (e) DAST,
CH2Cl2; (f) (4-NO2)C6H4OH, NEt3, BF3ÆEt2O, CH2Cl2; (g) NaOMe, MeOH.
active site, but is unable to be turned over by the
enzyme, or, the substrate may have little or no affinity
for the enzyme active site. In order to distinguish
between these two possibilities, each septanoside was
tested as an inhibitor of the enzyme. If the outcome of
this experiment were to demonstrate that septanosides
were competitive inhibitors, this would imply that these
compounds were bound to the active site, but were
catalytically incompetent substrates. The absence of
any inhibition in the presence of each septanoside would
confirm that these compounds were unable to bind to
the enzyme active site.
resembles pNP-a-D-glucopyranoside (Fig. 1), was not
observed to undergo enzyme-catalysed hydrolysis. No
inhibition of the hydrolysis of pNP b-D-glucopyranoside
was observed in the presence of 2.4 mM septanoside 8,
confirming that this anomer is not accommodated by
the enzyme active site to any significant extent. When
the equivalent study was performed with S. cerevisiae
a-glucosidase, the complementary result was observed;
septanoside 8 was a slow substrate with an overall 103-
fold lower specificity constant than that for pNP a-D-
glucopyranoside, whereas septanoside 7 was found not
to bind to the enzyme at a concentration of 1.6 mM.
While these are clearly not good substrates, the observa-
tion of significant turnover would appear to be vindica-
tion of the notions best represented in Figure 1.
However, it was quite possible that these septanosides
might not show any specificity, but in fact be hydrolysed
by a range of glycosidases via non-specific association.
In order to probe this further, studies with a wide selec-
tion of other glycosidases (a- and b-galactosidases, a-
and b-xylosidases and a-mannosidase) were performed.
No detectable hydrolysis of either septanoside was seen
As can be seen from the molecular modelling overlay
(Fig. 1), in terms of the three-dimensional spatial
arrangement of hydroxyl groups, septanoside 7 has
many features in common with pNP b-D-glucopyrano-
side. Indeed, septanoside 7 was observed to be a slow
substrate for Agrobacterium sp. b-glucosidase (Table
1). The kcat/KM value for this substrate is approximately
105-fold lower than that of the equivalent pyranoside
substrate. In contrast, septanoside 8, which more closely

A. Tauss et al. / Tetrahedron: Asymmetry 17 (2006) 234–239
237
Table 1.
Enzyme
membered ring iminoalditols are indeed binding in a
‘productive’ conformation. However, the low activities
measured imply that those inhibitors likely are not tran-
sition state mimics. The findings also expand the known
catalytic potential of glucopyranosidases in demonstrat-
ing that they are also capable of effecting the transfer of
septanosides and may be useful in this regard for syn-
thetic applications.
Substrate
kcat
(sꢀ1 a
KM
kcat/KM
)
(mM) (sꢀ1 Mꢀ1
)
b-Glucosidase
Agrobacterium sp.
pH 7.0
pNP-b-Glc13
7
170
0.024 1.1
<0.001 ndb
0.078 2.2 · 106
22
nd
8
a-Glucosidase
S. cerevisiae
pH 6.8
pNP-a-Glcc
7
7.0
<0.001 nd
0.07 2.9
0.27
2.6 · 104
nd
8
24
b-Galactosidase
E. coli
pNP-b-Gal17
7
90
<0.1
<0.1
0.040 2.3 · 103
4. Experimental
nd
nd
pH 7.0
8
nd
nd
Optical rotations were measured on a JASCO Digital
Polarimeter or with a Perkin Elmer 341 with a path
length of 10 cm. NMR spectra were recorded at 200 as
well as 500 MHz (1H), and at 50 and 125 MHz (13C).
CDCl3 or acetone-d6 were employed for protected com-
pounds and MeOH-d4 for free glycosides. Chemical
shifts are listed in delta employing residual, not deuter-
ated, solvent as the internal standard. The signals of aro-
matic substituents as well as of protecting groups were
found in the expected regions and are not listed explic-
itly. Structures of crucial intermediates were unambigu-
ously assigned by 1D-TOCSY and HSQC experiments.
Electrospray mass spectra were recorded on an HP
1100 series MSD, Hewlett Packard. Samples were dis-
solved in acetonitrile/MeOH mixtures. The scan mode
for negative ions (mass range 100–1000 D) was
employed varying the fragmentation voltage from 30
to 130 V. TLC was performed on precoated aluminium
sheets (E. Merck 5554). Compounds were detected by
staining with concd H2SO4 containing 5% vanillin.
a-Galactosidase
Green coffee bean
pH 6.5
pNP-a-Gal18
7
8
66
<0.01
<0.01
0.45
nd
nd
1.5 · 105
nd
nd
b-Xylosidase
T. saccharolyticum
pH 6.5
pNP-b-Xyl14
7
8
9.7
<0.01
<0.01
0.036 2.7 · 105
nd
nd
nd
nd
a-Xylosidase
E. coli K12
pH 7.0
pNP-a-Xyl15
7
8
0.13
<0.001 nd
<0.001 nd
0.71
180
nd
nd
a-Mannosidase
C. ensiformis
pH 4.5
pNP-a-Man19 130
7
8
2.0
nd
nd
6.5 · 104
nd
<0.1
<0.1
nd
a Lower rate limit calculated at 1.6 mM 7 and 2.4 mM 8.
b nd = not determined.
c This work.
in any case. These substrate mimetics were again tested
for their ability to bind to, and inhibit, each of the
enzymes tested. However, with the exception of E. coli
a-xylosidase Yic1, no inhibition was observed for any
of the enzymes studied. This clearly demonstrates that,
not only do 7 and 8 not bind productively to the active
sites of these other enzymes, but rather they do not bind
at all. In the case of Yic1, both septanosides inhibited
the hydrolysis of pNP a-D-xylopyranoside (95% at
1.6 mM and 70% at 2.4 mM for 7 and 8, respectively).
These data are in accordance with previous (unpub-
lished) work from this laboratory, which has shown that
Yic1 has a very high affinity (low lM) for the binding of
b-linked glucosides in the aglycone subsites. In the case
of Thermoanaerobacterium saccharolyticum b-xylosidase
XynB, an increase in the rate of enzyme-catalysed cleav-
age of pNP b-D-xylopyranoside was observed in the
presence of both septanosides (1.6 and 2.4 mM for 7
and 8, respectively). Since neither of these compounds
are substrates for XynB, this increase in rate was
attributed to their ability to act as glycosyl acceptors
in a transglycosylation reaction, a phenomenon previ-
ously described for this enzyme when studied with
substrates possessing activated leaving groups such as
p-nitrophenol.
For column chromatography Silica Gel 60 (E. Merck)
was used.
Molecular modelling was performed as previously
described12 employing Sybyl versions on an Octane work-
station by Silicon Graphics using the BGFS-minimiser
and the Tripos Force Field.
The known L-idose and its diethyldithioacetal 9 were
prepared by available methods as mentioned in Section
2.
4.1. 6-O-Trityl-L-idose diethyldithioacetal 10
To a solution of L-idose diethyl dithioacetal 9 (1.38 g,
4.82 mmol) in pyridine (15 mL), chlorotriphenyl-
methane (1.5 equiv, 2 g) was added and the mixture
was kept at 40 ꢁC until TLC indicated quantitative reac-
tion of the starting material. The reaction mixture was
concentrated under reduced pressure and the remaining
residue dissolved in CH2Cl2. The organic layer was
sequentially washed with 5% aqueous HCl, satd aque-
ous sodium bicarbonate and dried over Na2SO4. Filtra-
tion, removal of the solvent and purification on silica gel
gave trityl ether 10 (2.40 g, 94%) as a faintly yellow oil.
Found: C, 65.84; H, 6.91; C29H36O5S2 (528.7353)
3. Conclusion
In conclusion, the finding that the septanosides 7 and 8
studied are indeed substrates, albeit weak ones, for the
predicted glycosidases, but not for enzymes of other
specificities, supports the contention that the seven-
requires: C, 65.88; H, 6.86; Mass spectrum (API-ES):
20
m/z: 527.70 [MꢀH]; ½aꢁD ¼ ꢀ4:6 ðc 1:9; MeOHÞ; dH
(MeOH-d4) 4.06 (d, 1H, J1,2 6.8 Hz, H-1), 4.04 (dd,
1H, J2,3 3.9 Hz, J3,4 4.4 Hz, H-3), 3.89 (dd, 1H, J4,5

238
A. Tauss et al. / Tetrahedron: Asymmetry 17 (2006) 234–239
6.8 Hz, H-4), 3.87 (m, 1H, H-5), 3.82 (dd, 1H, H-2), 3.30
protected septanose 13 (2.25 g, 88%). Found: C, 68.38;
H, 4.80; C34H28O10 (596.5963) requires: C, 68.45; H,
4.73; Mass spectrum (API-ES): m/z: 595.60 [MꢀH]; a-
anomer: dH (CDCl3) 6.13 (dd, 1H, J3,4 9.3 Hz, J4,5
9.3 Hz, H-4), 5.96 (dd, 1H, J2,3 9.8 Hz, H-3), 5.82 (dd,
0
0
(dd, 1H, J5,6 4.9 Hz, J6;6 9.5 Hz, H-6), 3.19 (dd, 1H, J5;6
4.9 Hz, H-60); dC (MeOH-d4) 73.9, 72.2, 71.8, 71.1 (C-2,
C-3, C-4, C-5), 65.2 (C-6), 54.6 (C-1).
0
4.2. 2,3,4,5-Tetra-O-benzoyl-6-O-trityl-L-idose diethyl-
dithioacetal 11
1H, J1,2 6.8 Hz, H-2), 5.59 (ddd, 1H, J5,6 9.8 Hz, J5;6
0
4.4 Hz, H-5), 5.38 (d, 1H, H-1), 4.30 (dd, 1H, J6;6
13.2 Hz, H-6), 4.04 (dd, 1H, H-60); dC (CDCl3) 96.2
(C-1), 74.0, 73.6, 72.3, 69.0 (C-2, C-3, C-4, C-5), 60.9
(C-6). b-Anomer: dH (CDCl3) 5.91 (d, 1H, J1,2 3.4 Hz,
H-1); dC (CDCl3) 92.9 (C-1), 75.2, 73.6, 72.7, 70.8 (C-
2, C-3, C-4, C-5), 61.2 (C-6).
To a solution of compound 10 (15.1 g, 28.6 mmol) in
pyridine (100 mL), benzoyl chloride (16.2 mL,
140 mmol) was added and the mixture stirred at ambient
temp for 16 h. Precipitated pyridinium hydrochloride
was removed by filtration and the solution was concen-
trated under reduced pressure. The remaining residue
was dissolved in CH2Cl2 and the organic layer sequen-
tially washed with 5% aqueous HCl, 5% aqueous
sodium bicarbonate and dried over Na2SO4. Filtration,
removal of the solvent and purification on silica gel gave
tetrabenzoate 11 (26.1 g, 97%) as a faintly yellow oil.
Found: C, 72.35; H, 5.60; C57H52O9S2 (945.1726)
4.5. 2,3,4,5-Tetra-O-benzoyl-a/b-L-idoseptanosyl fluoride
14
To a 3% solution of septanose 13 (1.73 g, 2.90 mmol) in
dry CH2Cl2, DAST (760 lL, 2 equiv) was added at
ꢀ50 ꢁC and the mixture stirred at ꢀ30 ꢁC until all the
starting materials had reacted. The reaction mixture
was carefully washed with 5% aqueous bicarbonate,
dried over Na2SO4 and filtered. Removal of the solvent
under reduced pressure furnished an anomeric mixture
(a/b ꢂ 1:10) of the septanosyl fluorides 14, which was
pure enough for the next step (1.69 g, 97%). Found: C,
68.14; H, 4.60; C34H27FO9 (598.5873) requires: C,
68.22; H, 4.55; Mass spectrum (API-ES): m/z: 597.50
[MꢀH]; b-anomer (main product) dH (CDCl3) 6.26
(dd, 1H, J3,4 9.8 Hz, J4,5 6.8 Hz, H-4), 6.02–5.87 (m,
3H, H-1, H-2, H-3), 5.60 (m, 1H, H-5), 4.62 (dd, 1H,
requires: C, 72.43; H, 5.55; Mass spectrum (API-ES):
20
m/z: 944.10 [MꢀH]; ½aꢁD ¼ þ6:2 ðc 1:6; CH2Cl2Þ; dH
(acetone-d6) 6.39 (dd, 1H, J3,4 4.4 Hz, J4,5 5.9 Hz,
H-4), 6.32 (dd, 1H, J2,3 5.9 Hz, H-3), 5.93 (dd, 1H, J1,2
5.4 Hz, H-2), 5.90 (m, 1H, H-5), 4.43 (d, 1H, H-1),
0
3.51 (dd, 1H, J5,6 5.6 Hz, J6;6 10.0 Hz, H-6), 3.39 (dd,
1H, J5;6 4.6 Hz, H-60); dC (acetone-d6) 73.0, 71.8, 70.9,
0
70.0 (C-2, C-3, C-4, C-5), 61.8 (C-6), 51.7 (C-1).
4.3. 2,3,4,5-Tetra-O-benzoyl-6-O-trityl-L-idose 12
J5,6 1.7 Hz, J6;6 13.8 Hz, H-6), 4.28 (m, 1H, H-60); dC
0
To a 3% solution of compound 11 (1.61 g, 1.70 mmol) in
a mixture of acetone/H2O/CH3CN 2:1:2 (v/v/v) HgO
(590 mg, 1.6 equiv) and HgCl2 (740 mg, 1.6 equiv) were
added and the mixture stirred at 60 ꢁC for 18 h until
TLC indicated complete conversion of the starting
material. Salts were removed by filtration and washed
with acetone. The reaction mixture was concentrated
under reduced pressure and the residue was dissolved
in CH2Cl2. This solution was consecutively washed with
1 M aqueous KI solution and saturated aqueous sodium
thiosulfate and dried (Na2SO4). Evaporation of the sol-
vent and purification on silica gel furnished unstable free
aldehyde 12 (1.39 g, 97%) as an oil, which was used
immediately in the next step. Found: C, 75.82; H,
5.11; C53H42O10 (838.9197) requires: C, 75.88; H, 5.05;
(CDCl3) 106.6 (JC,F 230.1 Hz, C-1), 73.5, 71.9 (C-4, C-
5), 71.9 (JC,F 25.4 Hz, C-2), 70.3 (JC,F 2.9 Hz, C-3),
61.6 (JC,F 4.8 Hz, C-6). a-Anomer: dC (CDCl3) 108.8
(JC,F 229.3 Hz, C-1), 74.2, 71.3 (C-4, C-5), 70.8 (JC,F
34.6 Hz, C-2), 67.7 (JC,F 8.0 Hz, C-3), 61.5 (C-6).
4.6. 4-Nitrophenyl 2,3,4,5-tetra-O-benzoyl-a-L-idosepta-
noside 15 and 4-nitrophenyl 2,3,4,5-tetra-O-benzoyl-b-L-
idoseptanoside 16
To a 3% solution of septanosyl fluoride 14 (380 mg,
0.635 mmol) in CH2Cl2 containing NEt3 (110 lL), 4-
nitrophenol (115 mg, 0.83 mmol) and a few drops of
BF3ÆEt2O in Et2O (1 M) were added. After 1 h, the mix-
ture was washed with 5% aqueous bicarbonate, dried
over Na2SO4, filtered and concentrated under reduced
pressure. Chromatography gave pure b-16 (190 mg,
42%) and a-septanosides (15, 175 mg, 38%). Found:
C, 66.87; H, 4.40; C40H31NO12 (717.6926) requires:
Mass spectrum (API-ES): m/z: 837.80 [MꢀH];
20
½aꢁD ¼ ꢀ13:8 ðc 1:7; CH2Cl2Þ; dH (acetone-d6) 9.77 (m,
1H, H-1), 6.62 (m, 1H, H-4), 6.28 (m, 1H, H-3), 6.05
(m, 1H, H-2), 5.91 (m, 1H, H-5), 3.51 (m, 2H, H-6, H-
60); dC (acetone-d6) 194.4 (C-1), 76.5, 71.8, 70.0, 69.8
(C-2, C-3, C-4, C-5), 62.1 (C-6).
C, 66.94; H, 4.35; Mass spectrum (API-ES): m/z:
20
716.60
[MꢀH];
a-anomer:
½aꢁD ¼ ꢀ23:8 ðc 2:1;
CH2Cl2Þ; dH (CDCl3) 6.19 (dd, 1H, J1,2 6.8 Hz, J2,3
9.7 Hz, H-2), 6.17 (dd, 1H, J3,4 9.7 Hz, J4,5 8.8 Hz,
H-4), 6.02 (dd, 1H, H-3), 5.83 (d, 1H, H-1), 5.61 (ddd,
4.4. 2,3,4,5-Tetra-O-benzoyl-L-idoseptanose 13
0
0
To a 3% solution of free aldehyde 12 (3.60 g, 4.29 mmol)
in CH2Cl2, MeOH (25 mL) and BF3ÆEt2O (1.1 equiv) in
Et2O were added at ambient temp. After 15–20 min,
CH2Cl2 (150 mL) was added and the mixture washed
with 5% aqueous bicarbonate until neutral and dried
over Na2SO4. Filtration, removal of the solvents under
reduced pressure and purification of the residue on silica
gel gave an anomeric mixture (a/b ꢂ 6:1) of partially
1H, J5,6 10.7 Hz, J5;6 4.4 Hz, H-5), 4.29 (dd, 1H, J6;6
13.2 Hz, H-6), 4.07 (dd, 1H, H-6); dC (CDCl3) 99.5
(C-1), 74.2, 71.6, 71.1, 68.3 (C-2, C-3, C-4, C-5), 60.6
20
(C-6). b-Anomer: ½aꢁD ¼ þ79:0 ðc 2:3; CH2Cl2Þ; dH
(CDCl3) 6.41 (dd, 1H, J3,4 5.9 Hz, J2,3 9.8 Hz, H-3),
6.08 (dd, 1H, J1,2 2.4 Hz, H-2), 5.94 (dd, 1H, J4,5
5.9 Hz, H-4), 5.91 (d, 1H, H-1), 5.58 (ddd, 1H, J5,6
0
0
2.2 Hz, J5;6 4.9 Hz, H-5), 4.43 (dd, 1H, J6;6 13.7 Hz,

A. Tauss et al. / Tetrahedron: Asymmetry 17 (2006) 234–239
239
H-6), 4.17 (dd, 1H, H-60); dC (CDCl3) 97.2 (C-1), 73.2,
72.7, 72.0, 71.4 (C-2, C-3, C-4, C-5), 60.7 (C-6).
Reactions were followed in a UV–vis spectrophotometer
by measuring the change in absorbance of light at
400 nm. Data were analysed by direct fit of the rates
observed to the Michaelis–Menten equation using the
programme GraFit.16
4.7. 4-Nitrophenyl a-L-idoseptanoside 7
To a 1% methanolic solution of compound 15 (87 mg,
0.12 mmol), methanolic NaOMe (1 M, two drops) was
added at ꢀ20 ꢁC. After completion of the reaction, the
mixture was neutralised with ion exchange resin Amber-
lite IR 120 [H+], filtered and concentrated under reduced
pressure. Final purification was achieved by preparative
TLC to give the deprotected septanoside 7 (22 mg, 60%).
Found: C, 47.80; H, 5.07; C12H15NO8 (301.2553)
For pyranoid substrates pNP b-D-glucopyranoside,13
pNP b-D-galactopyranoside,17 pNP a-D-galactopyrano-
side,18 pNP b-D-xylopyranoside,14 pNP a-D-xylopyr-
anoside,15 pNP a-D-mannopyranoside,19 the previously
reported Km and kcat values have been used.
Acknowledgements
requires: C, 47.84; H, 5.02; Mass spectrum (API-ES):
20
m/z: 300.20 [MꢀH]; ½aꢁD ¼ ꢀ169:6 ðc 0:87; MeOHÞ;
Financial support by the Austrian Fonds zur Fo¨rderung
der Wissenschaflichen Forschung (FWF), Vienna, Pro-
ject P 15726-N03, as well as from the Protein Engineer-
ing Network of Centres of Excellence of Canada
(PENCE) is gratefully acknowledged.
dH (MeOH-d4) 5.31 (d, 1H, J1,2 6.8 Hz, H-1), 3.86 (dd,
0
1H, J2,3 9.8 Hz, H-2), 3.71 (dd, 1H, J5,6 10.7 Hz, J6;6
0
12.2 Hz, H-6), 3.57 (ddd, 1H, J4,5 8.8 Hz, J5;6 3.9 Hz,
H-5), 3.52 (dd, 1H, J3,4 9.3 Hz, H-3), 3.49 (dd, 1H,
H-60), 3.35 (dd, 1H, H-4); dC (MeOH-d4) 102.5 (C-1),
79.3, 72.4, 71.8, 70.0 (C-2, C-3, C-4, C-5), 63.3 (C-6).
References
4.8. 4-Nitrophenyl b-L-idoseptanoside 8
1. Withers, S. G.; Namchuk, M.; Mosi, R. In Iminosugars as
To a 1% methanolic solution of compound 16 (116 mg,
0.16 mmol), methanolic NaOMe (1 M, two drops) was
added at ꢀ20 ꢁC. After completion of the reaction, the
mixture was neutralised with ion exchange resin Amber-
lite IR 120 [H+], filtered and concentrated under reduced
pressure. Final purification was achieved by preparative
TLC to give the deprotected septanoside 8 (25 mg, 51%).
Found: C, 47.78; H, 5.06; C12H15NO8 (301.2553)
Glycosidase Inhibitors; Stutz, A. E., Ed.; Wiley-VCH:
¨
Weinheim, Germany, 1999; pp 188–206.
2. Wrodnigg, T. M. From liana to a glycotechnology tool:
25 years of 2,5-imino-2,5-dideoxy-2,5-imino-D-mannitol
(DMDP). In Timely Research Perspectives in Carbohy-
drate Chemistry; Schmid, W., Stutz, A. E., Eds.; Springer:
¨
Vienna, New York, 2002; pp 393–426.
3. Moris-Varas, F.; Qian, X.-H.; Wong, C.-H. J. Am. Chem.
Soc. 1996, 118, 7647–7652.
4. Le Merrer, Y.; Poitout, L.; Depezay, J.-C.; Dosbaa, I.;
Geoffroy, S.; Foglietti, M.-J. Bioorg. Med. Chem. 1997, 5,
519–533.
5. Painter, G. F.; Eldridge, P. J.; Falshaw, A. Bioorg. Med.
Chem. 2004, 12, 225–232.
6. Pakulski, Z. Polish J. Chem. 1996, 70, 667–707.
7. Grindley, T. B.; Gulasekharam, V. J. Chem. Soc., Chem.
Commun. 1978, 1073–1074.
requires: C, 47.84; H, 5.02; Mass spectrum (API-ES):
20
m/z: 300.20 [MꢀH]; ½aꢁD ¼ þ94:5 ðc 0:59; MeOHÞ; dH
(MeOH-d4) 5.65 (d, 1H, J1,2 2.0 Hz, H-1), 3.96 (dd,
0
1H, J5,6 3.4 Hz, J6;6 12.7 Hz, H-6), 3.91 (dd, 1H, J2,3
8.8 Hz, J3,4 8.3 Hz, H-3), 3.87 (dd, 1H, H-2), 3.75
0
(ddd, 1H, J4,5 7.3 Hz, J5;6 6.5 Hz H-5), 3.61 (dd, 1H,
H-60), 3.55 (dd, 1H, H-4); dC (MeOH-d4) 99.4 (C-1),
77.5, 72.7, 72.6, 71.7 (C-2, C-3, C-4, C-5), 65.5 (C-6).
8. Meyer, A. J.; Reichstein, T. Helv. Chim. Acta 1946, 29,
152–159.
4.9. Enzyme kinetics
9. Albert, R.; Dax, K.; Link, R. W.; Stutz, A. E. Carbohydr.
¨
Res. 1983, 118, c5–c6.
All pNP-pyranoside substrates, S. cerevisiae a-glucosi-
dase, E. coli b-galactosidase, Green Coffee Bean a-galac-
tosidase and Canavalia ensiformis a-mannosidase were
purchased from Sigma chemical company. Agrobacte-
rium sp. b-glucosidase, T. saccharolyticum b-xylosidase
and E. coli str. K12 a-xylosidase were expressed and
purified as previously reported.13–15 Kinetic studies were
performed at 37 ꢁC in 50 mM sodium citrate/phosphate
buffer at the pH stated in Table 1. Enzyme concentra-
tions ranging from 0.5 lg mLꢀ1 to 0.1 mg mLꢀ1 were
used, depending on the substrate studied. As a primary
screen, the hydrolysis of septanosides 7 and 8 was
attempted at substrate concentrations of 1.6 and
2.4 mM, respectively. In the absence of any observed
hydrolysis, inhibition of the enzyme-catalysed hydroly-
sis of the natural pNP pyranoside was attempted at
the same septanoside concentrations with the pNP-
pyranoside concentration around its KM value. Sub-
strate concentrations ranging from approximately
0.5 · KM to 5 · KM were employed wherever possible.
10. Helferich, B.; Gnuchtel, A. Chem. Ber. 1938, 71, 712–718.
¨
11. Blanc-Muesser, M.; Defaye, J. Synthesis 1977, 8, 568–569.
12. Andersen, S. M.; Ekhart, C.; Lundt, I.; Stutz, A. E.
¨
Carbohydr. Res. 2000, 326, 22–33; Andersen, S. M.; Ebner,
M.; Ekhart, C. W.; Gradnig, G.; Legler, G.; Lundt, I.;
Stutz, A. E.; Withers, S. G.; Wrodnigg, T. Carbohydr. Res.
¨
1997, 301, 155–166.
13. Kempton, J. B.; Withers, S. G. Biochemistry 1992, 31,
9961–9969.
14. Vocadlo, D. J.; Wicki, J.; Rupitz, K.; Withers, S. G.
Biochemistry 2002, 41, 9727–9735.
15. Lovering, A. L.; Lee, S. S.; Kim, Y. W.; Withers, S. G.;
Strynadka, N. C. J. J. Biol. Chem. 2005, 280, 2105–2115.
16. Leatherbarrow, R. J. Grafit 4.0; Erithacus Software Ltd.:
Staines, UK, 1998.
17. Juers, D. H.; Hakda, S.; Matthews, B. W.; Huber, R. E.
Biochemistry 2003, 42, 13505–13511.
18. Ly, H. D.; Howard, S.; Shum, K.; He, S. M.; Zhu, A.;
Withers, S. G. Carbohydr. Res. 2000, 329, 539–547.
19. Einhoff, W.; Ruediger, H. Biol. Chem. Hoppe-Seyler 1988,
369, 165–169.