Synthesis of glucose- and mannose-derived N-acetylamino imidazopyridines and their evaluation as inhibitors of glycosidases

Article abstract of DOI:10.1055/s-1999-3654

The correlation between basicity and inhibitory power of tetrahydropyridoazole type glycosidase inhibitors has been used to determine the effect of substituents at positions 2 and 3. The gluco-and manno- configurated 2-acetamido derivatives 11 and 13 were

Full text of DOI:10.1055/s-1999-3654

PAPER
1459
Synthesis of Glucose- and Mannose-Derived N-Acetylamino Imidazopyridines
and their Evaluation as Inhibitors of Glycosidases
Narendra Panday, Andrea Vasella*
Laboratorium für Organische Chemie, ETH-Zentrum, Universitätstrasse 16, CH-8092 Zürich, Switzerland
Fax +41(1)6321136; E-mail: vasella@sugar.org.chem.ethz.ch
Received 4 April 1999
correlates qualitatively with their basicity expressed by
Abstract: The correlation between basicity and inhibitory power of
pKHA values of –4.0, 2.4, and 6.1 for the gluco-derivatives
tetrahydropyridoazole type glycosidase inhibitors has been used to
1, 2, and 3, and –4.4, 2.0, and 5.7 for the manno-deriva-
determine the effect of substituents at positions 2 and 3. The gluco-
tives 4, 5 and 635). Conversely, correlation between the
and manno-configurated 2-acetamido derivatives 11 and 13 were
prepared from the lactam 15 in four steps via the amidines 17 and pKHA and Ki values of substituted azoles should indicate
18. Similarly, 15 was transformed into the 3-acetamido analogues
12 and 14 in three steps. The acetamido imidazoles are significantly
less basic than the parent pyridoimidazoles, but possess essentially
the effect of favourable or unfavourable interactions of
substituents with glycosidases.
We wished to synthesise more strongly basic imidazoles
related to 3 and 6 and to correlate their Ki with their pKHA
and thereby decide whether substitution of such imida-
zoles at C(2) or C(3) should be pursued in view of obtain-
ing stronger inhibitors. For this purpose, the
aminoimidazoles 7–10 appeared particularly attractive.
Based on an extension of the Hammett equation to five
membered heterocycles,9,10 we expected the pKHA values
of the amino-imidazoles 7–10 to be higher by 0.5–1.5
units as compared to those of 2,3-unsubstituted imida-
zoles 3 or 6. Many 4- and 5-aminoimidazoles36) are unsta-
ble,11-13 but valuable intermediates for the synthesis of N-
acylaminoimidazoles. Although N-acylaminoimidazoles
such as 11–14 should be more stable than 7–10, they are
expected9 to be less basic than the parent imidazoles 3 and
6, and hence weaker inhibitors, unless the effect of the re-
duced basicity is compensated by favourable interactions
between the acylamino moiety and the glycosidases. As
mentioned above, such an effect will be evident from a
pKHA/Ki correlation.
the same conformation. Correlation of the basicity of 11 and 12 and
their inhibition of b-glucosidases from almonds and from Caldocel-
lum saccharolyticum and, similarly, of the basicity of 13 and 14 and
the inhibition of snail b-mannosidase shows a detrimental effect of
the 3-acetamido group. The 2-acetamido group does not interfere
with the inhibition. The inhibition of yeast a-glucosidase by 11 is
surprisingly strong, evidencing a binding interaction of the 2-aceta-
mido group with this enzyme.
Key words: imidazoles, glycosidase, inhibition, cyclization, N-cy-
anomethylamidine, pKHA
The strong inhibition of retaining b-glycosidases by tet-
rahydroazolopyridines has been attributed to the presence,
at the "anomeric" position, of a nitrogen atom which can
accept a hydrogen bond from the catalytic acid of the gly-
cosidase.1 The proton transfer leads to a (partial) positive
charge on the azolopyridine and a concomitant strength-
ening of the interaction with the catalytic nucleophile.2 In
agreement with this, the strength of inhibition by the isos-
teric gluco- and manno-azolopyridines 1–6 (Figure 1) 3-5
OH
N
OH
OH
3
4
6
5
N
N
2
N
N
N
HO
HO
HO
HO
HO
HO
8
N
N
N
N1
8a
7
OH
OH
OH
1
2
3
HO
HO
HO
HO
HO
OH
HO
HO
HO
OH
N
OH
N
N
N
N
N
HO
N
N
4
5
6
OH
OH
NHR
N
HO
HO
HO
NHR
N
HO
HO
HO
OH
N
NHR
OH
N
NHR
N
N
HO
HO
HO
HO
N
N
OH
OH
8 R = H
12 R = Ac
7 R = H
11 R = Ac
9 R = H
13 R = Ac
10 R = H
14 R = Ac
Figure 1
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

1460
N. Panday, A. Vasella
PAPER
OR2
OBn
N
NHR1
OBn
R2O
R2O
N
4
5
N
CN
NH
BnO
BnO
BnO
BnO
N
1
2
NH2
OR2
R
1 = H, R2 = Bn
3
OBn
OBn
21
OBn
NH
c)
d)
17
19
BnO
22 R1 = Ac, R2 = Bn
11 R1 = Ac, R2 = H
a)
b)
BnO
+
+
X
OBn
15 X = O
16 X = S
BnO
BnO
BnO
R2O
OR2
OBn
OBn
N
N
OBn
NH2
NHR1
CN
NH
R2O
R2O
N
BnO
N
R1 = H, R2 = Bn
R1 = Ac, R2 = Bn
OBn
23
24
18
20
c)
d)
e)
13 R1 = Ac, R2 = H
NHAc
2
OH
OBn 4
BnO
3
N
OBn
NH
OBn
NHAc
5
NH2
N
c)
d)
8a
N
N
N
BnO
BnO
BnO
BnO
1
HO
HO
6
7
N
8
N
N
OBn
25
OBn
OH
OBn
OBn
27
29
12
+
+
+
HO
NHAc
OH
N
BnO
N
BnO
NHAc
N
OBn
BnO
OBn
N
c)
d)
NH2
OBn
N
NH
HO
HO
BnO
BnO
BnO
BnO
N
BnO
BnO
N
N
26
28
30
14
a) NH3, THF/MeOH, Hg(OAc)2, 17.HOAc, 42%; 18∑HOAc, 42%.b) ICH2CN, K2CO3, acetone. c) Ac2O, pyridine, 22, 31% (two steps); 24, 31%
(two steps). d) H2, MeOH/ H2O/ HOAc 1:1:1, Pd/C 10%; 11, 78%; 12, 80%; 13, 80%; 14, 73%. e) 1) NH2CH2CN, Hg(OAc)2, THF, 2) Ac2O,
pyridine, 29, 40%; 30, 28%.
Scheme
We planned to prepare the aminoimidazoles 7–10 (Figure after the H–C(2) signal of 17 and 18 had disappeared by
1) via the N-(cyanomethyl)amidines 19, 20, 25 and 26 exchange with a deuterium from the solvent (ca. 720 min).
(Scheme). These amidines are derivatives of the readily The amidinium acetates 17∑HOAc and 18∑HOAc were
available gluconolactam 154,14,15 (Scheme), and expected stored under argon at 23 °C for months without any sign
to cyclise in situ37). Exploratory activation of the cyanom- of alteration, while the amidines 17 and 18 epimerised
ethylation product of 15 by thionation was not satisfacto- within a few days to a ca. 1:1 mixture of 17 and 18 which
ry. We decided to proceed by amination of the were readily hydrolysed to the corresponding lactams.
thionolactam 16.
The position of the NH bands in the IR spectra of 17 and
The thionolactam 16 (Scheme) is readily available from 18 agrees well with the one expected for the tautomer with
15.4,15,23 Treatment of 16 with a soln. of NH3 in MeOH in an endocyclic double bond.24-26,38 The acetate counter ion
the presence of 1 equiv. of Hg(OAc)2 led to a 1:1-mixture of 17∑HOAc and 18∑HOAc is confirmed by a C = O band
of the gluco- and manno-configurated amidinium acetates at 1702 and 1704 cm-1 and by a quartet at d = 22.43 and
17∑HOAc and 18∑HOAc, isolated each in 42% by chro- 22.60, and a singlet at d = 177.59 and 177.67 ppm, respec-
matography. Amination of 16 in the absence of Hg(OAc)2 tively. The assignment of the gluco-configuration to 17
1
(compare 24) was very slow and, as evidenced by the H and 17∑HOAc is based on the J(3,4) values (9.0 Hz);
NMR of the crude, accompanied by b-elimination of the J(3,4) = 3.1 and 2.8 Hz for the manno-configurated 18
C(3)-benzyloxy group. The amidines 17 and 18 were ob- and 18∑HOAc, respectively. In CDCl3, the gluco-amidine
4
tained in 95% yield by treating a CH2Cl2 soln. of 17 adopts a H3-conformation, as reflected by the large
17∑HOAc and 18∑HOAc, respectively, with a ca. aq J(2,3), J(3,4), and J(4,5) values (9.0, 9.3, and 8.1 Hz, re-
0.05M NaOH and ice. At 23 °C, the amidines epimerised spectively). The J(2,3) and J(3,4) values of the gluco-ami-
partially, pure 17∑HOAc leading to a 9:1 and pure dinium acetate 17∑HOAc remain large (9.0 and 8.1 Hz,
18.HOAc to a 1:9 mixture of 17 and 18. This epimeriza- respectively), but the J(4,5) value is significantly de-
tion was much faster in the presence of a ca. 1M aq. soln creased (5.3 Hz), indicating that 17∑HOAc prefers the S3-
of NaOH or of a sat. aq K2CO3, leading within a few min- conformation. The small J(H,H) of the manno-amidine 18
utes at 0 °C to a 1:1 mixture of 17 and 18 from either and its acetate 18∑HOAc agree well with a 3H4-conforma-
17∑HOAc or 18∑HOAc. In CDCl3 containing 1 equiv. of tion.
DBU, 17 as well as 18 equilibrated to a 55:45-mixture 17/
According to TLC, alkylation39 in acetone of 17 and 18, or
18 within 180 min. The epimerization was much slower in
of a 1:1 mixture 17/18 with iodoacetonitrile in the pres-
the absence of DBU and the equilibrium was reached only
ence of K2CO3 led to two new compounds, assumed to be
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of Glucose- and Mannose-Derived N-Acetylamino Imidazopyridines
1461
the aminoimidazoles 21 and 23.40 These compounds
proved highly unstable and transformed into tars during
their attempted isolation. Formation of 21 and 23 is evi-
denced by the 1H NMR of the crude obtained by workup
with a sat. aq NH4Cl. Two singlets of identical intensity at
d = 6.21 and 6.27 ppm compare well with the H–C(5) sig-
nal at d = 6.20 ppm of an aminoimidazole described by
Julia and Tam.12 The analogous signals of the hydrochlo-
ride in the somewhat less coloured crude resulting from
workup with 1M HCl appeared at d = 6.47 and 6.51 ppm.
In view of these results, the reaction mixture was treated
with Ac2O and pyridine to yield the acetamidoimidazoles
22 (31%) and 24 (31%). Hydrogenolytic debenzylation of
22 and 24 and purification of the products by reversed
phase HPLC gave the highly hygroscopic acetamidoimi-
dazoles 11 (78%) and 13 (80%) that had to be stored under
argon to avoid discolouration.
Table 1 Vicinal Coupling Constants J (H,H) in Hz of Protected and
Unprotected 2- and 3-Acetamidoimidazoles in Various Solvents and
deduced Conformationsa
Compound J (7,8) J (6,7)
J (5,6)
9.0
Conformation
7H6
3b
8.7
9.0
8.4
8.7
9.0
9.0
6.5
5.6
2.5
3.5
3.5
3.7
3.7
3.7
4.1
3.9
4.0
4.0
3.1
2.8
3.1
9.7
10.0
9.5
10.0
8.7
8.7
5.9
7.5
4.1
4.4
4.0
10.0
8.1
9.7
8.7
8.4
7.0
6.3
9.2
8.7
8.1
3.HClb
8.7
7H6
11b
9.5
7H6
11b f
8.4
7H6
,
12b
12b,f
12d
22
)
7S or7H6
7S
g
7.0
g
)
7H6/6H7
7H6/6H7
6H7
7.5
2.2
3.1
2.5
8.4
6.1
8.7
6.2
5.6
4.7
2.5
6.9
4.7
3.7
29
The regioisomeric, gluco- and manno- acetamidoimida-
zoles 29 (40%) and 30 (28%) were obtained by treatment
of the thionolactam 16 with aminoacetonitrile in the pres-
ence of Hg(OAc)2, followed by in situ acetylation. Again,
attempts to isolate the nitriles 25 and 26, and the aminoim-
idazoles 27 and 28 failed on account of their instability.
Similarly as for 21 and 23, formation of 27 and 28 was ev-
idenced by the singlets at d = 6.38 and 6.51 ppm in the 1H
NMR spectrum of the crude obtained after workup with
sat. aq. NH4Cl. The corresponding signals of the bona fide
hydrochlorides appeared at d = 6.67 and 6.70 ppm. Hy-
drogenolytic debenzylation of 29 and 30 gave the highly
29c
29d
6b
6H7
6H7
7H6
6.HClb
7H6/6H7
7H6
13b
13b,f
14b
7H6/6H7
7S /B5,8
B5,8/7S
B5,8
hygroscopic acetamidoimidazoles 12 (80%) and 14 (73%) 14b,e
after purification by ion exchange chromatography. In
contrast to the regioisomeric acetamides 11 and 13, these
acetamides did not require storage under argon.
14b,f
24
7H6/6H7
B5,8/7S
B5,8/7S
The acetamido substituent may directly affect the inhibi-
tion by interaction with the glycosidase, and/or indirectly
by affecting the conformation of the inhibitor. To differ-
entiate between these possibilities, we studied the confor-
30
30d
a 300 MHz in CDCl3, unless indicated otherwise.
b D2O.
mation
of
the
protected
and
unprotected
acetamidoimidazoles. The assignment of the gluco-con-
figuration to the C(2)-acetamides 11 and 12 is based on
large J(7,8) (8.4 and 5.9 Hz) and the assignment of the
manno-configuration to 13 and 14 on small J(7,8) (3.7 and
3.1 Hz). As deduced from their J(H,H) (Table 1), the 2-ac-
etamides 11, 13 and 21–24 adopt different conformations,
closely corresponding, however, to those of the analogous
2,3-unsubstituted tetrahydroazolopyridines.1,3,4 Thus, the
c C6D6
.
d DMSO-d6.
e 5% HOAc.
f 5% CF3COOH.
g Not determined.
J(5,6) (7.0 Hz) for the protonated41 12 in D2O is slightly
smaller than for 3 (9.0 Hz) and the C(2)-acetamides 11
7
unprotected 11 and 13 adopt a H6 conformation that is
very close to the one of 3 and 6. The protected gluco-iso-
mer 22 is a 2:1 mixture of the 7H6 and the 6H7 conformers,
while the protected manno-configurated 24 adopts the 7S
conformation.
7
(8.4 Hz) and 13 (8.7 Hz), indicating a predominant S-
conformation. The effect of the 1,5 interaction on the con-
formation of the unprotected manno acetamide 14 in D2O
is somewhat stronger, the smaller J(5,6) (5.6 Hz) indicat-
7
ing an equilibrium between the S and B5,8 conformers.
For the 3-acetamidoimidazoles 12, 14, 29, and 30 one ex-
pects a conformational effect of the 1,5-interaction be-
tween the acetamido and the benzyloxymethyl groups
(compare the similar effect for C(1), C(8) disubstituted
tetrahydropyrrolopyridines 27,28). The effect on the confor-
mation of the unprotected gluco- and manno acetamides
12 and 14 in D2O is, however, surprisingly small. Thus,
The tetrabenzylated manno-acetamide 30 in CDCl3 and
7
DMSO-d6 also exists as a mixture of the S and the B5,8
conformers. The tetrabenzylated gluco-acetamide 29
adopts almost exclusively the 6H7 conformation in CDCl3,
C6D6 and DMSO-d6, as reflected by the small J(5,6),
J(6,7) and J(7,8) values (2.2–4.4 Hz). J(6,7) and J(7,8) of
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

1462
N. Panday, A. Vasella
PAPER
Table 2 Selected Chemical shifts and Coupling constants of the Major and Minor Conformers of 3-Acetamido imidazoles
a
comp.
solvent
H–C(2)major
H–C(2)minor
NHmajor
NHminor
Memajor
Meminor
J(5,6)major
ratio
29
30
29
29
30
12
12
14
CDCl3
CDCl3
C6D6
DMSO-d6
DMSO-d6
DMSO-d6
D2O
7.16–7.43
7.19–7.61
7.94
6.84
6.84
6.72
7.15
7.10
6.88
6.88
6.90
6.89
6.89
6.78
–
8.67
8.71
8.01
9.60
9.65
9.66
–
7.16–7.43
7.19–7.61
7.29b
9.03
9.05
8.83
1.55
1.56
1.38
1.92
1.93
1.95
2.18
2.11
1.95
1.98
1.64
1.69
1.72
2.02
–
2.0, 8.5
4.4, 8.4
3.1, 7.5
5.0, 7.5
c
c
c
c
90:10
92:8
83:17
95:5
94:6
97:3
–
–
–
D2O
–
–
–
–
a conventional carbohydrate numbering used.
b determined by saturation transfer from H–N of the main conformer.
c not determined.
the unprotected epimeric C(3)-acetamides 12 and 14 in the temperature (ca. 90:10 at 25 °C, ca. 93:7 at –20 °C,
D2O are very similar to those of the regioisomeric 11 and and ca. 96:4 at –60 °C). In CDCl3 and C6D6, H–C(2) of the
13, and of 3 and 6, so that J(7,8) is a valid criterium for the main conformer of 29 and 30 resonates at lower fields
configurational assignment of 12 and 14.
than H–C(2) of the minor conformer, suggesting that H–
C(2) of the major conformer is deshielded by the acetami-
do carbonyl group. This is supported by the NOEs be-
tween H–C(5) and the H–N, between one of the HC–C(5)
and the H–N (but not between H–C(2) and the H–N) for
the major conformer of 29 in C6D6. According to force
field calculations (MM3*, gas phase31), the acetamido
group of the major conformer is rotated by ca 30° around
the N–C(3) bond, H–N pointing towards the oxygen of the
(benzyloxy)methyl group, to which it forms a hydrogen
bond (d (N, O) = 2.62 Å; A in Figure 2). In agreement
with this, H–N of the major conformer of 29 and 30 in
CDCl3 and C6D6 resonates at lower fields than H–N of the
minor conformer. The coupling constants of the benzy-
loxymethyl group agree well with a preferred gt orienta-
tion. Signal overlap and disturbing saturation transfers in
the NOE-experiment did not allow the determination of
the orientation of the acetamido group in the minor con-
former. An (E)-conformation of the acetamide should be
strongly disfavoured. In DMSO-d6, the H–C(2) shifts of
the major and the minor conformer of 12, 29, and 30 are
very similar (cf. Table 2), in contradistinction to the
chemical shifts of the major and the minor conformers in
CDCl3 and C6D6. The orientation of the acetamido group
As observed for 3,4 the J(H, H) values of the 2- and 3-ac-
etamido gluco- tetrahydroimidazopyridines 11 and 12 in
D2O change only insignificantly upon protonation. This is
different for the manno-epimers. Protonation of the 2-ac-
etamido isomer 13 leads to an increased population of the
6H7 conformation, similar to what is known for the 2,3-un-
substituted manno-imidazole 6.4,42 Protonation of the 3-
acetamido isomer 14 shifts the 7S/B5,8 equilibrium towards
B5,8. The protonation induced conformational change of
the manno-imidazoles may explain their lower pKHA as
compared to the one of the analogous gluco-compounds
(Table 3).
1
The H NMR spectra of the 3-acetamides 12, 29, and 30
in various solvents are characterised by two sets of
signals43 (cf. Table 2), corresponding to two conformers
arising from different orientations of the acetamido
group.44 One of the conformers strongly dominates at
26 °C. The two sets of signals begin to coalesce at 50 °C
(CDCl3), are replaced by a single set of broad signals at
90 °C (DMSO-d6), and separate again upon cooling. As
expected, the ratio of conformers of 29 in CDCl3 changes
in favour of the major conformer upon further lowering of
Table 3 pKHA-Values and Inhibition Constants of 2- and 3-Acetamidoimidazoles 11–14 compared to those of 3 and 6
pKHA
b-Glucosidases
from
b-Glucosidase
from
a-Glucosidase
from
b-Mannosidase
from
a-Mannosidase
from
sweet almondsa
Caldocellum sacch.b
brewer's yeasta
snailsc
Jack Beanc
11
12
3
13
14
6
3.2
5.1
6.1
<3.0
4.7
5.7
3 mMd
1.5 mMd
71 mMd
–
–
–
–
–
–
250 mM
91 mM
136 mMd
0.1 mM5
0.02 mM
59 mM
–
–
–
–
–
–
–
–
–
14 mMd
275 mMd
0.055 mM5
85 mMd
130 mMd
0.6 mM5
a pH 6.8, 37°C.
b pH 6.8, 55°C.
c pH 4.0, 25°C.
d IC50.
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of Glucose- and Mannose-Derived N-Acetylamino Imidazopyridines
1463
Figure 2 Low energy conformations of 29 according to MM3* force field calculations. For convenience, the benyloxy groups at C(6), C(7)
and C(8) are replaced by methoxy groups
in this solvent is probably determined by hydrogen bond- and 13 relative to 3 (2.9 and >2.745 to those for 12 and 14
ing of H–N to a DMSO-d6 molecule. To avoid steric inter- relative to 6 (DpKHA = 1), as predicted by the extension of
actions between the (benzyloxy)methyl group or H–C(2) the Hammett equation to five membered heterocycles.10
and DMSO-d6, the acetamido group adopts an almost per- In keeping with their reduced basicity, the gluco-configu-
pendicular orientation relative to the imidazole plane. rated 11 and 12 inhibit the b-glucosidases from almonds
This interpretation is strongly supported by force field cal- and Caldocellum saccharolyticum much less strongly
culations for the analogue of 29 possessing an N–methy- than the parent imidazole 3. Similarly, the manno-config-
lated acetamido group (to imitate the bulk due to urated acetamides 13 and 14 are significantly weaker in-
interaction with DMSO). As suggested by the strong hibitors of b-mannosidase from snails than 6. The Ki
NOEs between H–C(5) and H–N of the major conformer values of the 3-acetamido imidazoles 12 and 14 are much
of 12 and 14 in DMSO-d6, H–N points below the plane of larger than those expected from their pKHA (cf. pKHA/Ki
the imidazole ring of the major conformer (B in Figure 2) correlation in Figure 3), reflecting the unfavourable posi-
and, hence, above it in the minor conformer (C in Figure tion of the C(3)–acetamido substituent and possibly the
7
2). In agreement with the perpendicular orientation of the slight deviation of 12 and 14 from the H6 conformation.
acetamido group in DMSO-d6, a weak NOE is also ob- The Ki of the 2-acetamidoimidazoles 11 and 13 agree rath-
served between H–N and H–C(2) signal of the major con- er well with that expected from its pKHA, indicating that
former. The stronger deshielding of the H–N of the major the C(2)-acetamido group only has a minor influence on
conformer may reflect a stronger hydrogen bond to the inhibition.
DMSO-d6. In the minor conformer, this hydrogen bond is
Since inhibitors of b-glycosidases possessing a basic ni-
probably weakened by the steric interaction of DMSO-d6
trogen atom in the ''anomeric" position also inhibit a-gly-
with the axial benzyloxymethyl group.
cosidases,32 11 and 12 were tested against a-glucosidase
In contrast to the 1H NMR spectra of the 3-acetamidoimi- from brewer's yeast, and 13 and 14 against a-mannosidase
dazoles 12, 29, and 30, the 1H NMR spectra of the 2-ace- from jack beans. Although the 2-acetamido imidazole 11
tamidoimidazoles 11, 13, 22, and 24 are characterised by is significantly less basic than 3, it inhibits a-glucosidase
only one set of signals. Force field calculations indicate from brewer's yeast about as well, indicating a binding in-
that the syn conformer is favoured by 3.8 kcal/mol over teraction with the C(2)-acetamido group. The manno-ace-
the anti conformer. In agreement with this, there is no tamidoimidazoles 13 and 14, however, are considerably
NOE between H–N and H–C(3).
weaker inhibitors of jack beans a-mannosidase than 6.
The 2- and 3-acetamido imidazoles 11–14 are indeed sig-
nificantly less basic than the parent imidazoles 3 and 6
(Table 3). The base weakening effect of the 2-acetamido
Solvents were distilled before use. Normal workup implies distribu-
tion of the crude product between Et2O and sat. aq. NH4Cl and ice,
group is much stronger than that of the 3-acetamido unless indicated otherwise, drying of the organic layer (MgSO4), fil-
tration, and evaporation of the filtrate. TLC: Merck silica gel 60F-
group, as shown by comparing the DpKHA values for 11
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

1464
N. Panday, A. Vasella
PAPER
3
6
*
+5
+5
+3
+1
Ð1
Ð3
12
*
14
*
+3
+1
Ð1
Ð3
11
13
*
5
2
1
4
50
100
150
200
250
50
100
150
200
250
Ki
Ki
(against β-glucosidases from almonds)
(against β-mannosidase from snails)
Figure 3 Qualitative pKHA/Ki correlation of Gluco- and Manno-configurated Acetamidoimidazoles. The straights are those obtained by linear
regression to the data of the 2,3-unsubstituted azoles 1, 2, and 3, and 4, 5, and 6. IC50 values have been divided by two to obtain an approximation
of the Ki.
254 plates; detection by heating with mostain (10% H2SO4 (400
mL), (NH4)6Mo7O24.6H2O (20 g), Ce(SO4)2 (0.4 g)). Flash chroma-
tography (FC): silica gel Fluka 60 (0.04–0.063 mm).IR spectra:
KBr or 3% CHCl3 soln. 1H NMR (300 MHz, if not indicated other-
wise) and 13C NMR (75 MHz, if not indicated otherwise) were mea-
sured at 25 °C: chemical shifts d in ppm and coupling constants J in
Hz. FAB- and CI-MS:3-nitrobenzyl alcohol and NH3 as matrix, re-
sp., unless indicated otherwise.
IR (CHCl3): n = 3449m, 3340–2300m (br), 3007m, 2872m, 1756w,
1704s, 1556w, 1501w, 1454m, 1409m, 1365w, 1099s, 1012w,
913w, 652w, 616w cm-1.
1H NMR (CDCl3): d = 2.02 (s, AcO); 3.57 (t, J = 10.3 Hz, irrad. at
3.70 → s, H–C(6)); 3.68–3.73 (m, H–C(5), H–C(6)); 3.82 (t, J = 3.1
Hz, irrad. at 4.38 → d, J ≈ 3.0, irrad. at 3.98 → d, J ≈ 3.0, H–C(3));
3.98 (dd, J = 3.4 Hz, 1.6, irrad. at 3.70 → d, J ≈ 3.2, irrad. at 3.82
→ d, J ≈ 1.5, H–C(4)); 4.31 (d, J = 12.1 Hz, PhCH); 4.33 (d,
J = 12.5 Hz, PhCH); 4.38 (d, J = 2.9 Hz, irrad. at 3.82 → s, H–
C(2)); 4.39 (d, J = 12.5 Hz, PhCH); 4.42 (d, J = 11.8 Hz, PhCH);
4.44 (d, J = 12.1 Hz, PhCH); 4.49 (d, J = 12.1 Hz, PhCH); 4.53 (d,
J = 11.8 Hz, PhCH); 4.60 (d, J = 11.8 Hz, PhCH); 6.63 (br s. ex-
change with CD3OD, NH); 7.12–7.15 (m, 4 arom. H); 7.22–7.36 (m,
16 arom. H); 9.71 (br s, exchange with CD3OD, 2 NH).
13C NMR (CDCl3): d = 22.60 (q, Me); 54.93 (d, C(5)); 69.31 (t,
CH2C(5)); 71.41, 71.66, 71.92 (3d, C(2), C(3), C(4)); 71.53 (t,
PhCH2); 72.29 (t, PhCH2); 72.63 (t, PhCH2); 73.37 (t, PhCH2);
128.05–129.15 (several d); 135.88 (s); 137.19 (s); 137.32 (s);
138.12 (s); 165.59 (s, C(1)); 177.67 (s, C = O).
1-Amino-2,3,4,6-tetra-O-benzyl-1,5-dideoxy-1,1,5-nitrilo-D-
glucitol Acetate (17.HOAc) and 1-Amino-2,3,4,6-tetra-O-ben-
zyl-1,5-dideoxy-1,1,5-nitrilo-D-mannitol Acetate (18◊HOAc)
At 5 °C, a solution of 16 (100 mg, 0.181 mmol) and Hg(OAc)2 (58
mg, 0.181 mmol) in THF (5 mL) was treated with 2 M NH3 in
MeOH (1 mL) and stirred at 23 °C for 10 h. Filtration of the black
mixture through Celite and evaporation gave 17∑HOAc/18∑HOAc
1:1 (96 mg, 1H NMR). FC (Et2O/CH2Cl2/HOAc 5:5:1) gave
17∑HOAc (45 mg, 42%) und 18∑HOAc (45 mg, 42%).
Data of 17.HOAc: Rf (Et2O/CH2Cl2/HOAc 5:5:1) 0.56.
CI-MS:537 (5, [M–OAc]+), 429 (22, [M–OAc–BnOH]+), 321 (29),
231 (13), 201 (19), 123 (14), 108 (100), 91 (59).
IR (CHCl3): n = 3445m, 3340–2300m (br), 3090w, 2971m, 2870m,
1702s, 1559m, 1497m, 1455m, 1409m, 1362m, 1028m, 914m,
658m, 600m cm-1.
1H NMR (CDCl3): d = 2.08 (s, AcO); 3.46 (dd, J = 10.3 Hz, 1.5, H–
C(6)); 3.61 (dd, J = 10.3 Hz, 2.5, H–C(6)); 3.74 (br m, irrad. at 4.00
→ change, H–C(5)); 3.85 (dd, J = 9.3 Hz, 8.1, irrad. at 4.00 → d, J
≈ 9.0, irrad. at 4.26 → d, J ≈ 8.0, H–C(3)); 3.99 (dd, J = 8.1 Hz, 5.3,
H–C(4)); 4.26 (d, J = 9.0 Hz, H–C(2)); 4.36 (d, J = 11.2 Hz, PhCH);
4.48 (d, J = 11.5 Hz, PhCH); 4.55 (d, J = 11.2 Hz, PhCH); 4.57 (d,
J = 11.2 Hz, PhCH); 4.68 (d, J = 11.5 Hz, PhCH); 4.79 (d, J = 11.2
Hz, PhCH); 4.85 (d, J = 10.9 Hz, PhCH); 4.91 (d, J = 11.2 Hz,
PhCH); 6.39 (br s, exchange with CD3OD, NH); 7.22–7.59 (m, 20
arom. H); 10.49 (br s, exchange with CD3OD, NH).
13C NMR (CDCl3): d = 22.43 (q, Me); 57.19 (d, C(5)); 69.63 (t,
C(6)); 73.63 (t, PhCH2); 73.86 (t, PhCH2); 74.49, 77.21, 81.99 (3d,
C(3), C(4), C(5)); 75.02 (t, PhCH2); 75.88 (t, PhCH2); 128.21–
129.25 (several d); 136.40 (s); 137.40 (s); 137.73 (s); 137.79 (s);
166.69 (s, C(1)); 177.59 (s, C = O). CI-MS:537 (7, [M – OAc]+),
429 (41, [M – OAc – BnOH]+), 321 (100), 231 (29), 201 (36), 125
(77), 108 (71), 91 (47).
1-Amino-2,3,4,6-tetra-O-benzyl-1,5-dideoxy-1,1,5-nitrilo-D-
glucitol (17) and 1-Amino-2,3,4,6-tetra-O-benzyl-1,5-dideoxy-
1,1,5-nitrilo-D-mannitol (18)
a) At 0 °C, 17∑HOAc (50 mg, 0.084 mmol) was distributed between
CH2Cl2 and a 0.05M aq NaOH and ice within 3 min. Drying of the
organic layer (MgSO4) and evaporation gave 17 (45 mg, 98%).
b) As described in a), but with 18∑HOAc (50 mg, 0.084 mmol):
gave 18 (45 mg, 98%).
c) As described in a) but without ice at 23 °C: gave 17/18 9:1 (45
mg, 98%, 1H NMR).
d) As described in b) but without ice at 23 °C: gave 17/18 1:9 (45
mg, 98%, 1H NMR).
e) As described in a) but with a 1M aq NaOH as the aq layer: gave
17/18 1:1 (45 mg, 98%, 1H NMR).
f) As described in b) but with a 1M aq NaOH as the aq layer: gave
17/18 1:1 (45 mg, 98%, 1H NMR).
Data of 18∑HOAc: Rf(Et2O/CH2Cl2/HOAc 5:5:1) 0.47.
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of Glucose- and Mannose-Derived N-Acetylamino Imidazopyridines
1465
g) As described in a) but with a sat. aq K2CO3 as the aqueous layer:
gave 17/18 1:1 (45 mg, 98%, 1H NMR).
13C NMR (CDCl3): d = 23.49 (q, Me); 58.55 (d, C(5)); 67.63 (t,
CH2C(5)); 72.85 (t, PhCH2); 73.53 (t, PhCH2); 74.44 (br t, 2
PhCH2); 74.58, 75.68, 82.01 (3d, C(6), C(7), C(8)); 105.14 (d,
C(3)); 127.97–128.78 (several d, s of C(2)); 137.65 (br s); 137.69
(s); 138.23 (s); 140.19 (s, C(8a)); 167.22 (s, C=O).
Data of 17: Rf (AcOEt/MeOH 10:1) 0.14.
IR (CHCl3): n = 3514m, 3405m, 3090m, 3067m, 3008s, 2914m,
2868m, 1951w, 1876w, 1818w, 1663m, 1577m, 1497m, 1454m,
1406w, 1362m, 1159w, 1091s, 1028m, 909m, 651w cm-1. 1H NMR
(CDCl3): d = 3.53–3.68 (m, H–C(5)), 2 H–C(6)); 3.79 (dd, J = 9.3
Hz, 8.1, irrad. at 3.93 → d, J ≈ 8.2, H–C(4)); 3.93 (t, J = 9.3 Hz, ir-
rad. at 4.12 → d, J ≈ 9.0, H–C(3)); 4.17 (dd, J = 9.0 Hz, 0.9, irrad.
at 3.93 → br s, H–C(2)); 4.45 (d, J = 11.2 Hz, PhCH); 4.51 (d,
J = 11.2 Hz, PhCH); 4.57 (d, J = 11.2 Hz, PhCH); 4.64 (d, J = 11.5
Hz, PhCH); 4.82 (d, J = 11.0 Hz, PhCH); 4.84 (d, J = 11.0 Hz,
PhCH); 4.90 (d, J = 11.2 Hz, PhCH); 4.94 (d, J = 11.0 Hz, PhCH);
4.90–5.10 (br s, exchange with CD3OD, H2N); 7.14–7.41 (m, 20
arom. H).
CI-MS (NH3):618 (100, [M+1]+), 91 (24).
Anal. calc. for C38H39N3O5 (617.74): C 73.88, H 6.36, N 6.80;
found: C 73.91, H 6.46, N 6.66.
Data of 24
Colourless oil. Rf (AcOEt):0.67.
IR (CHCl3): n = 3436m, 3067m, 3007s, 2870m, 1953w, 1877w,
1684s, 1540s, 1455s, 1262m, 1096s, 1028m, 909w cm-1.
1H NMR (CDCl3): d = 2.14 (s, AcN); 3.71 (dd, J = 10.3 Hz, 5.3,
CH–C(5)); 3.81 (dd, J = 10.8 Hz, 3.1, CH–C(5)); 3.84 (dd, J = 9.2
Hz, 3.1, irrad. at 4.49 → d, J = 3.1 Hz, irrad. at 4.65 → d, J = 9.0
Hz, H–C(7)); 4.04–4.10 (m, irrad. at 4.49 → change, H–C(5)); 4.44
(d, J = 12.5 Hz, PhCH); 4.49 (dd, J = 9.3 Hz, 6.9, irrad. at 3.84 →
d, J = 4.3 Hz, H–C(6)); 4.49 (d, J = 12.1 Hz, PhCH); 4.57 (d,
J = 10.9 Hz, PhCH); 4.59 (d, J = 11.8 Hz, PhCH); 4.65 (d, J = 3.1
Hz, irrad. at 3.84 → s, H–C(8)); 4.66 (d, J = 11.8 Hz, PhCH); 4.66
(d, J = 11.7 Hz, PhCH); 4.75 (d, J = 12.5 Hz, PhCH); 4.95 (d,
J = 11.2 Hz, PhCH); 7.19–7.42 (m, 20 arom. H); 7.46 (s, H–C(3));
8.34 (br s, exchange with CD3OD, NH).
13C NMR (CDCl3): d = 60.78 (d, C(5)); 70.78 (t, C(6)); 73.48 (t,
PhCH2); 74.73 (t, PhCH2); 74.83 (t, PhCH2); 75.01 (t, PhCH2);
78.01, 78.50, 83.14 (3d, C(2), C(3), C(4)); 127.87–128.99 (several
d); 158.60 (s, C(1)).
Data of 18:Rf (EtOAc/MeOH 10:1) 0.13.
IR (CHCl3): n = 3514m, 3405m, 3066m, 3067m, 3008m, 2668m,
1951w, 1811w, 1663w, 1575w, 1496m, 1454m, 1406w, 1363m,
1092s, 1028m, 910m cm-1.
1H NMR (CDCl3): d = 3.45–3.67 (m, H–C(5)), 2 H–C(6)); 3.77 (dd,
J = 3.6 Hz, 2.8, irrad. at 3.91 → d, J ≈ 5.6, H–C(4)); 3.91 (t, J = 3.1
Hz, irrad. at 4.25 → d, J ≈ 3.1, H–C(3)); 4.25 (d, J = 3.1 Hz, irrad.
at 3.93 → s, H–C(2)); 4.43 (s, PhCH2); 4.50–4.61 (m, 6 PhCH);
4.85–5.05 (br s, exchange with CD3OD, H2N); 7.15–7.41 (m, 20
arom. H).
13C NMR (CDCl3): d = 57.81 (d, C(5)); 71.70 (t, C(6)); 72.01 (t,
PhCH2); 72.34 (t, PhCH2); 72.37 (t, PhCH2); 73.29 (t, PhCH2);
73.82, 74.55, 76.02 (3d, C(2), C(3), C(4)); 127.81–128.96 (several
d); 137.44 (s); 138.23 (s); 138.36 (s); 138.70 (s); 160.57 (s, C(1)).
13C NMR (CDCl3): d = 23.38 (q, Me); 60.12 (d, C(5)); 68.15 (d,
C(8)); 69.68 (t, CH2–C(5)); 70.67 (t, PhCH2); 72.17 (t, PhCH2);
73.37 (t, PhCH2); 73.92 (d), 79.73 (2d, C(6), C(7)); 75.02 (t,
PhCH2); 106.79 (d, C(3)); 127.94–128.70 (several d, s of C(2));
137.87 (s); 138.02 (s); 138.08 (s); 138.28 (s); 139.00(s, C(8a));
167.24 (s, C = O).
CI MS (NH3):618 (100, [M+1]+), 526 (7), 404 (7), 298 (11), 193 (8)
108 (10), 91(25).
(5R,6R,7S,8S)-N-[6,7,8-Trihydroxy-5-(hydroxymethyl)-5,6,7,8-
tetrahydroimidazo[1,2-a]pyridin-2-yl]acetamide (11)
(5R,6R,7S,8S)- and (5R,6R,7S,8R)-N-{6,7,8-Tris(benzyloxy)-5-
[(benzyloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]pyridin-
2-yl}acetamide (22 and 24)
a) At 23 °C, a mixture of 17/18 1:1 (260 mg, 0.478 mmol) and
K2CO3 (130 mg, 1.063 mmol) in acetone (1 mL) was treated with
iodoacetonitrile (1.2 mL, 11.32 mmol) and stirred until the starting
material was no longer detected on TLC (30 min). After treatment
with Ac2O (2 mL) and pyridine (2 mL), the mixture was stirred for
7 h. Normal workup and FC (hexane/EtOAc 2:1 → 1:2) gave 22 (94
mg, 32%) and 24 (94 mg, 32%).
A solution of 22 (40 mg, 0.065 mmol) in HOAc (1 mL) was treated
with 10% Pd/C (20 mg) and hydrogenated at 6.5 bar for 24 h. Fil-
tration through Celite, evaporation and HPLC (H2O/MeO Merck
LiChrosorb RP-8) gave 11 (13 mg, 78%). Highly hygroscopic white
solid. Rf (EtOAc/EtOH/HOAc 5:1:1) 0.23.
1H NMR (D2O): d = 1.92 (s, AcN); 3.78 (dd, J = 9.7 Hz, 8.7, irrad.
at 4.57 → d, J ≈ 9.5, H–C(7)); 3.93 (t, J = 9.3 Hz, irrad. at 3.78 →
d, J ≈ 9.0, H–C(6)); 3.99–4.03 (m, H–C(5)); 4.06 (dd, J = 12.8 Hz,
2.2, CH–C(5)); 4.11 (dd, J = 12.8 Hz, 2.2, CH–C(5)); 4.57 (d,
J = 8.4 Hz, irrad. at 3.78 → s, H–C(8)); 7.32 (s, H–C(3)).
1H NMR (D2O+5% CF3COOH): d = 2.02 (s, AcN); 3.86 (dd,
J = 10.0 Hz, 8.7, H–C(7)); 3.97 (dd, J = 10.0 Hz, 8.4, H–C(6)); 4.04
(dd, J = 12.8 Hz, 3.1, CH–C(5)); 4.13–4.19 (m, H–C(5)); 4.72 (dd,
J = 12.8 Hz, 2.5, CH–C(5)); 4.76–4.86 (hidden by HDO, H–C(8)).
7.52 (s, H–C(3)).
b) As described in a), but with 17 (20 mg, 0.037 mmol), K2CO3 (10
mg, 0.082), acetone (700 µl), iodoacetonitrile (0.1 ml):22 (7 mg,
31%) and 24 (7 mg, 31%).
c) As described in b) but with 18 gave 22 (7 mg, 31%) and 24 (7 mg,
31%).
Data of 22: Colourless oil. Rf (AcOEt) 0.75.
13C NMR (D2O): d = 24.72 (q, Me); 61.22 (t, CH2–C(5)); 63.29 (d,
C(5)); 69.83, 70.56, 77.32 (3d, C(6), C(7), C(8)); 109.82 (d, C(3));
138.86 (s, C(2)); 146.01 (s, C(8a)); 175.18 (s, C = O). CI MS
(NH3):258 (63, [M+1]+), 102 (100), 86 (55), 60 (64).
IR (CHCl3): n = 3436m, 3090w, 3007s, 2940m, 2868m, 1953w,
1877w, 1670s, 1622s, 1580m, 1533s, 1496m, 1454s, 1405m,
1364m, 1319s, 1094s, 1028m, 1096s, 990w, 909m, 658w, 602w
cm-1.
1H NMR (CDCl3): d = 2.10 (s, AcN); 3.81 (dd, J = 10.4 Hz, 4.4,
CH–C(5)) 3.89 (dd, J = 10.6 Hz, 3.4, CH–C(5)); 3.98 (t, J = 7.5 Hz,
H–C(6)); 4.06 (dd, J = 7.5 Hz, 5.6, irrad. at 4.66 → d, J ≈ 7.5, H–
C(7)); 4.13–4.18 (m, H–C(5)); 4.48 (s, PhCH2); 4.53 (d, J = 11.2
Hz, PhCH); 4.66 (d, J = 5.9 Hz, H–C(8)); 4.71 (d, J = 11.2 Hz,
PhCH); 4.80 (d, J = 11.2 Hz, PhCH); 4.81 (d, J = 11.8 Hz, PhCH);
4.84 (d, J = 11.2 Hz, PhCH); 5.09 (d, J = 11.5 Hz, PhCH); 7.18–
7.23 (m, 2 arom. H); 7.23–7.41 (m, 18 arom. H); 7.42 (s, H–C(3));
8.15 (br s, exchange with CD3OD, NH).
Anal. calc. for C10H15N3O5. 1.5 H2O (284.27): C 42.25, H 6.38, N
14.78; found: C 42.07, H 6.59, N 14.53.
(5R,6R,7S,8R)-N-[6,7,8-Trihydroxy-5-(hydroxymethyl)-5,6,7,8-
tetrahydroimidazo[1,2-a]pyridin-2-yl]acetamide (13)
As described for 22, with 24 (30 mg, 0.049 mmol), HOAc (0.8 mL)
and 10% Pd/C (15 mg; 24 h). HPLC (H2O/MeO Merck LiChrosorb
RP-8 ) gave 13 (10 mg, 80%). Highly hygroscopic off-white solid.
Rf (EtOAc/EtOH/HOAc 5:1:1) 0.23.
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

1466
N. Panday, A. Vasella
PAPER
1H NMR (D2O): d = 2.17 (s, AcN); 3.99 (dd, J = 10.0 Hz, 3.7.
J = 4.0 Hz, 2.5, irrad. at 4.12 → d, J ≈ 2.5, H–C(6)); 4.40 (s,
PhCH2); 4.52 (d, J = 11.5 Hz, PhCH); 4.55–4.58 (m, irrad at 3.52 →
change, irrad. at 4.24 → change, H–C(5)); 4.58 (d, J = 3.5 Hz, irrad.
at 4.12 → s, H–C(8)); 4.60 (d, J = 11.5 Hz, PhCH); 4.61 (d,
J = 11.5 Hz, PhCH); 4.66 (d, J = 11.8 Hz, PhCH); 4.74 (d, J = 12.1
Hz, PhCH); 4.96 (d, J = 12.1 Hz, PhCH); 6.84 (s, H–C(2)); 7.19–
7.34 (m, 20 arom. H); 9.60 (br s, NH). Signals of minor diastereoi-
somer: 1.69 (s, AcN); 4.08–4.09 (m, H–C(7)); 4.89 (d, J = 12.0 Hz,
PhCH); 6.89 (s, H–C(2)); 9.03 (s, exchange with CD3OD, NH).
13C NMR (CDCl3): d = 23.05 (q, Me); 58.10 (d, C(5)); 70.63 (d,
C(8)); 71.80 (t, CH2C(5)); 72.36 (t, PhCH2); 72.78 (t, PhCH2);
74.18 (t, PhCH2); 74.35 (t, PhCH2); 74.37, 75.70 (2d, C(6), C(7));
119.86 (d, C(2)); 127.51–128.88 (several d, s of C(3)); 136.43 (s,
C(8a)); 137.47 (s); 137.51 (s); 137.65 (s); 138.77 (s); 167.60 (s,
C = O). FAB-MS:618 (100, [M+1]+), 510 (17), 307 (12), 154 (24),
91 (34).
irrad. at 4.89 → d, J ≈ 9.5, H–C(7)); 3.96–4.01 (m, C(5)); 4.06 (dd,
J = 12.8 Hz, 3.7, CH–C(5)); 4.24 (dd, J = 9.7 Hz, 8.7, H–C(6)); 4.25
(dd, J = 12.8 Hz, 3.1, CH–C(5)); 4.89 (d, J = 3.7 Hz, H–C(8)); 7.37
(s, H–C(3)).
1H NMR (D2O+5% CF3COOH): d = 2.21 (s, AcN); 4.07 (dd,
J = 13.7 Hz, 5.0, CH–C(5)); 4.12 ( dd, J = 8.7 Hz, 4.1, irrad at 5.16
→ d, J ≈ 8.7, H–C(7)); 4.18–4.24 (m, H–C(5), CH–C(5)); 4.32 (dd,
J = 8.7 Hz, 6.2, H–C(6)); 5.16 (d, J = 4.1 Hz, H–C(8)); 7.53 (s, H–
C(3)).
13C NMR (125MHz, D2O): d = 24.74 (q, Me); 62.01 (t, CH2–C(5));
63.90 (d, C(5)); 66.40, 67.31, 73.47 (3d, C(6), C(7), C(8)); 110.34
(d, C(3)); 138.84 (s, C(2)); 144.65 (s, C(8a)); 174.97 (s, C = O). CI-
MS (NH3):258 (3, [M+1]+), 102 (63), 86 (55), 60 (100).
Anal. calc. for C10H15N3O5. 1.8 H2O (289.67): C 41.46, H 6.47: N
14.50; found: C 41.29, H 6.63, N 14.43.
Anal. calc. for C38H39N3O5 (617.74): C 73.88, H 6.36, N 6.80;
found: C 73.89, H 6.40, N 6.73.
(5R,6R,7S,8S)- and (5R,6R,7S,8R)-N-{6,7,8-Tris(benzyloxy)-5-
[(benzyloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]pyridin-
3-yl}acetamide (29 and 30)
Data of 30
Colourless oil. Rf (EtOAc) 0.30.
At 5 °C, a soln. of 16 (200 mg, 0.36 mmol) and Hg(OAc)2 (100 mg,
0.60 mmol) in THF (6 ml) was treated with aminoacetonitrile46 (70
mg, 1.25 mmol) and stirred for 8 h at 23 °C. After treatment of the
black mixture with Ac2O (2 mL) and pyridine (2 mL), stirring was
continued for 7 h. Filtration through Celite and normal workup gave
a yellow oil containing 29 and 30 in a ratio of 3:2 (1H NMR). FC
(EtOAc/hexane 1:1→ 2:1) gave 24 (93 mg, 42%) and 25 (62 mg,
28%).
IR (CHCl3): n = 3308w, 3068w, 3008m, 2928m, 2870w, 1689s,
1514w, 1497w, 1454m, 1370w, 1261m, 1095s, 1028 w, 909m
cm-1.
1H NMR (CDCl3, 92:8-mixture of 2 diastereoisomers). Signals of
major disastereoisomer: d = 1.56 (s, AcN); 3.55–3.63 (m, irrad. at
4.19 → change, CH2–C(5)); 3.79 (dd, J = 8.7 Hz, 2.8, irrad. at 4.04
→ d, J ≈ 2.5, irrad. at 4.77 → d, J ≈ 8.5, H–C(7)); 4.04 (dd, J = 8.4
Hz, 4.7, irrad. at 4.19 → d, J ≈ 8.5, H–C(6)); 4.19 (dt, J = 8.4 Hz,
4.4, H–C(5)); 4.36 (d, J = 10.3 Hz, PhCH); 4.41 (d, J = 11.8 Hz,
PhCH); 4.54 (d, J = 10.3 Hz, PhCH); 4.30 (d, J = 10.3 Hz, PhCH);
4.37 (d, J = 10.6 Hz, PhCH); 4.65 (d, J = 11.8 Hz, PhCH); 4.68 (d,
J = 11.5 Hz, PhCH); 4.77 (d, J = 2.5 Hz, H–C(8)); 4.97 (d, J = 11.5
Hz, PhCH); 7.19–7.61 (m, 20 arom. H, H–C(2)); 8.71 (s, exchange
with CD3OD, NH). Signals of minor diastereoisomer: 1.98 (s,
AcN); 6.88 (s, H–C(2)); 3.50–3.55 (m, CH2–C(5)); 3.88–3.92 (dd, J
≈ 5.0, 2.5, H–C(6));4.95 (d, J = 11.5 Hz, PhCH).
1H NMR (DMSO-d6, 94:6-mixture of 2 diastereoisomers). Signals
of major diastereoisomer: d = 1.93 (s, AcN); 3.54 (dd, J = 9.7 Hz,
4.7, irrad. at 4.40 → d, J ≈ 9.5, CH–C(5)); 3.72 (dd, J = 9.6 Hz, 6.8,
irrad. at 4.40 → d, J ≈ 9.5, CH–C(5)); 3.83 (dd, J = 8.1 Hz, 3.1, ir-
rad. at 4.72 → d, J ≈ 8.0, H–C(7)); 4.26 (dd, J = 8.1 Hz, 3.7, irrad.
at 3.83 → d, J ≈ 3.5, irrad. at 4.40 → d, J ≈ 8.0, H–C(6)); 4.35 (d,
J = 12.1 Hz, PhCH); 4.41 (d, J = 12.1 Hz, PhCH); 4.39–4.42 (m, ir-
rad. at 4.26 → change, H–C(5)); 4.52–4.63 (m, 4 PhCH); 4.65 (d,
J = 11.8 Hz, PhCH); 4.71 (d, J = 11.5 Hz, PhCH); 4.72 (d, J = 3.1
Hz, irrad. at 3.83 → s, H–C(8)); 6.84 (s, H–C(2)); 7.19–7.30 (m, 20
arom. H); 9.65 (br s, NH). Signals of minor diastereoisomer: 1.72
(s, AcN); 4.10–4.14 (m, H–C(6)); 6.89 (s, H–C(2)); 9.05 (s, NH).
Data of 29
Colourless oil. Rf (EtOAc) 0.34.
IR (CHCl3): n = 3311m, 3008m, 2929m, 2872m, 1685s, 1577m,
1540m, 1497m, 1455m, 1367m, 1071s, 1028m, 980w, 909m cm-1.
1H NMR (CDCl3, 90:10-mixture of 2 diastereoisomers). Signals of
main diastereoisomer: d = 1.55 (s, AcN); 3.73 (dd, J = 10.1 Hz, 2.0,
CH–C(5)); 3.83 (dd, J = 10.1 Hz, 8.5, CH–C(5)); 3.89 (dd, J = 4.1
Hz, 2.2, irrad. at 4.14 → d, J ≈ 2.2, H–C(6)); 4.14 (dd, J = 4.1 Hz,
2.5, H–C(7)); 4.34–4.46 (m, H–C(5)); 4.36 (d, J = 10.3 Hz, PhCH);
4.41 (d, J = 11.8 Hz, PhCH); 4.43 (d, J = 10.3 Hz, PhCH); 4.44 (d,
J = 11.8 Hz, PhCH); 4.60 (d, J = 11.8 Hz, PhCH); 4.70 (d, J = 2.5
Hz, irrad. at 4.14 → s, H–C(8)); 4.70 (d, J = 11.8 Hz, PhCH); 4.86
(d, J = 11.8 Hz, PhCH); 5.14 (d, J = 11.8 Hz, PhCH); 7.16–7.43 (m,
20 arom. H, H–C(3)); 8.67 (s, exchange with CD3OD, NH). Signals
of minor diastereoisomer: 1.95 (s, AcN); 3.57 (d, J = 6.2 Hz, CH2–
C(5)); 4.02 (dd, J ≈ 4.0, 2.5, H–C(7)); 4.56 (d, J = 12.1 Hz, PhCH);
4.90 (d, J = 12.1 Hz, PhCH); 5.12 (d, J = 12.1 Hz, PhCH); 6.88 (s,
H–C(2)).
1H NMR (C6D6, 83:17-mixture of 2 diastereoisomers). Signals of
main diastereoisomer: d = 1.38 (s, AcN); 3.43 (dd, J = 10.3 Hz, 3.1,
CH–C(5)); 3.59 (dd, J = 10.0 Hz, 7.5, CH–C(5)); 3.80 (dd, J = 4.4
Hz, 3.1, irrad. at 4.15 → d, J ≈ 3.1, H–C(6)); 3.87 (s, PhCH2); 4.15
(dd, J = 4.1 Hz, 3.7, irrad. at 3.80 → d, J ≈ 3.5, H–C(7)); 4.22 (d,
J = 11.8 Hz, PhCH); 4.26 (d, J = 11.8 Hz, PhCH); 4.37–4.40 (m, H–
C(5)); 4.45 (d, J = 11.8 Hz, PhCH); 4.64 (d, J = 11.5 Hz, PhCH);
4.90 (d, J = 3.5 Hz, irrad. at 4.15 → s, H–C(8)); 5.19 (d, J = 11.8
Hz, PhCH); 5.48 (d, J = 11.5 Hz, PhCH) 6.97–7.15 (m, 18 arom.
H); 7.51–7.53 (m, 2 arom. H); 7.94 (s, H–C(2)); 8.01 (br s, exchange
with CD3OD, NH). Signals of minor diastereoisomer: 1.64 (s,
AcN); 3.37–3.41 (m, CH–C(5)); 5.16 (d, J = 12.0 Hz, PhCH); 5.41
(d, J = 12.0 Hz, PhCH); 6.90 (s, H–C(2).
13C NMR (CDCl3): d = 23.04 (q, Me); 59.34 (d, C(5)); 68.91 (t,
CH2–C(5)); 70.57 (d, C(8)); 71.79 (t, PhCH2); 74.23 (t, PhCH2);
74.52 (t, PhCH2); 74.86 (t, PhCH2); 75.89, 80.54 (2d, C(6), C(7));
119.51 (d, C(2)); 127.56–129.12 (several d, s of C(3)); 136.45 (s,
C(8a)); 137.79 (s); 138.12 (s); 138.33 (s); 138.47 (s); 167.64 (s,
C = O). CI-MS (NH3):618 (100, [M+1]+), 512 (51), 404 (55), 298
(30), 193 (17) 108 (35), 91(44).
(5R,6R,7S,8S)-N-[6,7,8-Trihydroxy-5-(hydroxymethyl)-5,6,7,8-
tetrahydroimidazo[1,2-a]pyridin-3-yl]acetamide (12)
As described for 22, with 29 (50 mg, 0.081 mmol), AcOH (1 mL)
and 10% Pd/C (7 mg; 24 h). Ion-exchange chromatography (Am-
berlite CG-120, H+-form, elution with 0.05 M aq NH3) gave 12
(16.7 mg, 80%). Highly hygroscopic off-white solid. Rf (MeCN/
MeOH/HOAc 5:2:1) 0.30.
1H NMR (DMSO-d6, 95:5-mixture of 2 diastereoisomers). Signals
of main diastereoisomer: d = 1.92 (s, AcN); 3.52 (dd, J = 9.6 Hz,
5.0, CH–C(5)); 3.83 (dd, J = 9.7 Hz, 7.5, CH–C(5)); 4.12 (dd,
J = 4.1 Hz, 3.7, irrad. at 4.24 → d, J ≈ 4.0, H–C(7)); 4.24 (dd,
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of Glucose- and Mannose-Derived N-Acetylamino Imidazopyridines
1467
1H NMR (D2O): d = 2.18 (s, AcN); 3.76 (t, J = 8.7 Hz, irrad. at 4.62
→ d, J ≈ 8.5, H–C(7)); 3.89 ( dd, J = 13.2 Hz, 2.8, CH–C(5)); 4.03–
4.11(m, H–C(5), H–C(6), CH–C(5)); 4.62 (d, J = 9.0 Hz, H–C(8));
7.15 (s, H–C(2)).
1H NMR (D2O+5% CF3OH): d = 2.22 (s, AcN); 3.85 (t, J = 8.7 Hz,
irrad. at 4.62 Æ d, J ≈ 8.7, H–C(7)); 3.95 (dd, J = 12.8 Hz, 2.5, CH–
C(5)); 4.12 (dd, J = 8.7 Hz, 6.9, irrad. at 3.85 → d, J ≈ 7.0, H–C(6));
4.15 (dd, J = 12.6 Hz, 2.5, CH–C(5)); 4.32 (dt, J = 6.9 Hz, 2.5, H–
C(5)); 4.85 (d, J = 9.0 Hz, H–C(8)); 7.53 (s, H–C(2)).
1H NMR (DMSO-d6, 97:3-mixture of 2 diastereoisomers). Signals
of main diastereoisomer: d = 1.95 (s, AcN); 3.54 (dd, J = 6.7 Hz,
5.9, irrad. at 4.23 → d, J ≈ 6.0, H–C(7)); 3.57 (dd, J = 11.2 Hz, 3.4,
CH–C(5)); 3.77 (dd, J = 11.2 Hz, 4.4, CH–C(5)); 3.86–3.92 (m, H–
C(5), H–C(6)); 4.23 (d, J = 6.5 Hz, H–C(8)); 6.72 (s, H–C(2)); 9.66
(br s, NH). Signals of minor diastereoisomer:2.02 (s, AcN); 6.78 (s,
H–C(2)); 8.83 (s, NH).
13C NMR (D2O): d = 24.75 (q, Me); 61.93 (t, CH2C(5)); 65.43 (d,
C(5)); 68.90, 71.38, 75.57 (3d, C(6), C(7), C(8)); 117.87 (d, C(2));
130.12 (s, C(3)); 147.14 (s, C(8a)); 177.80 (s, C = O). CI-MS
(NH3):258 (100, [M+1]+), 176 (38).
absence of the inhibitor gave the appropriate IC50 value. Ki values
were determined by taking the slopes from the Lineweaver-Burk
plots33 and plotting them vs. the inhibtor concentrations.34 After fit-
ting a straight line to the data by linear regression, the negative [I]-
intercept of this plot gave the appropriate Ki.
b) Inhibition of Caldocellum Saccharolyticum b-Glucosidase
Similarly as a). The inhibition constants and IC50-values were deter-
mined at 55 °C.
c) Inhibition of Brewer’s Yeast a-Glucosidase
Similarly as a). The inhibition constants were determined using
0.025M KH2PO4/K2HPO4/NaCl buffer (pH 6.8) and 4-nitrophenyl
a-D-glucopyranoside as substrate.
d) Inhibition of Snail b-Mannosidase
Similarly as a) The inhibition constants were determined using
0.05M sodium citrate buffer (pH 4.0, 25 °C) and 4-nitrophenyl b-D-
mannopyranoside as substrate. The velocity of substrate hydrolysis
was determined by quenching the reaction after 5 min using 0.02M
Borate buffer (pH 9.2) and measuring the absorption at 400 nm.
e) Inhibition of Jack Bean a-Mannosidase
Similarly as d) The inhibition constants were determined using
0.05M/0.01M sodium citrate/ZnCl2 buffer (pH 4.0, 25 °C) and 4-ni-
trophenyl a-D-mannopyranoside as substrate.
Anal. calc. for C10H15N3O5. 2 H2O (293.27): C 40.95, H 6.53, N
14.33; found: C 40.72, H 6.79, N 14.21.
(5R,6R,7S,8R)-N-[6,7,8-Trihydroxy-5-(hydroxymethyl)-5,6,7,8-
tetrahydroimidazo[1,2-a]pyridin-3-yl]acetamide (14)
As described for 22, with 30 (30 mg, 0.049 mmol), HOAc (0.8 ml)
and 10% Pd/C (5 mg; 24 h). Ion-exchange chromatography (Am-
berlite CG-120, H+-form, elution with 0.05M aq NH3) gave 14 (9.2
mg, 73%). Highly hygroscopic off-white solid. Rf (MeCN/MeOH/
HOAc 5:2:1) 0.29.
1H NMR (D2O): d = 2.11 (s, AcN); 3.88 (dd, J = 11.9 Hz, 3.0, CH–
C(5)); 3.95 (dd, J = 8.4 Hz, 3.9, H–C(7)); 3.99–4.12 (m, H–C(5),
CH–C(5)); 4.32 (dd, J = 8.4 Hz, 6.2, H–C(6)); 4.90 (hidden by
HDO signal, H–C(8)); 7.10 (s, H–C(2)).
1H NMR (D2O+5% AcOH): d = 2.15 (s, AcN); 3.91 (dd, J = 12.4
Hz, 2.9, CH–C(5)); 4.06–4.19 (m, H–C(5), H–C(7), CH–C(5)); 4.39
(dd, J = 6.7 Hz, 4.7 H–C(6)); 5.00 (d, J = 4.0 Hz, H–C(8)); 7.25 (s,
H–C(2)).
1H NMR (D2O+5% CF3COOH): d = 2.17 (s, AcN); 3.92 (dd,
J = 12.3 Hz, 4.4, CH–C(5)); 4.07 (dd, J = 12.5 Hz, 4.4, CH–C(5));
4.19 (dd, J = 6.3 Hz, 3.8, irrad. at 5.14 → d, J ≈ 6.0, H–C(7)); 4.34–
4.35 (m, H–C(5)); 4.46 (dd, J = 6.2 Hz, 2.5, H–C(6)); 5.14 (d,
J = 4.0 Hz, H–C(8)); 7.48 (s, H–C(2)).
Acknowledgement
We thank T. Mäder for the HPLC purifications, Dr. B. Bernet for
numerous discussions about the conformational analysis and for
checking the experimental part, B. Brandenberg for the measure-
ments of the NOE's, M. Schneider and D. Manser for the determi-
nation of the pKHA values, and the Swiss National Science
Foundation, Hoffmann-La Roche AG, Basel, and Oxford Glycosci-
ences Ltd., Abingdon UK, for generous support.
References
(1) T. D. Heightman, M. Locatelli, A. Vasella, Helv. Chim. Acta
1996, 79, 2190.
(2) T. D. Heightman, A. Vasella, K. E. Tsitsanou, S. E.
Zographos, V. T. Skamnaki, N. G. Oikonomakos, Helv. Chim.
Acta 1998, 81, 853.
(3) P. Ermert, A. Vasella, Helv. Chim. Acta 1991, 74, 2043.
(4) T. Granier, N. Panday, A. Vasella, Helv. Chim. Acta 1997, 80,
979.
(5) K. Tatsuta, S. Miura, S. Ohta, H. Gunji, J. Antibiot. 1995, 48,
286.
13C NMR (D2O): d = 24.73 (q, Me); 62.01 (t, CH2C(5)); 64.98 (d,
C(5)); 65.97, 68.63, 71.32 (3d, C(6), C(7), C(8)); 117.94 (d, C(2));
130.25 (s, C(3)); 147.03 (s, C(8a)); 177.89 (s, C = O). CI-MS
(NH3):258 (100, [M+1]+), 176 (15).
(6) J. Catalan, J. L. M. Abboud, J. Elguero, Adv. Heterocycl.
Chem. 1986, 41, 187.
(7) K. Schofield, M. R. Grimmett, B. R. T. Keene, 'The Azoles',
Cambridge University Press: London, 1976.
Anal. calc. for C10H15N3O5. 2 H2O (293.27): C 40.95, H 6.53, N
14.33; found: C 40.69, H 6.75, N 14.25.
(8) V. A. Ostrovskii, G. I. Koldobskii, N. P. Shirokiva, V. S.
Poplavski, Khim. Geterosikl. Soedin 1981, 581,
(9) D. D. Perrin, B. Dempsey, E. P. Serjeant, 'pKa Prediction for
Organic Acids and Bases', Chapman and Hall, London, 1981.
(10) H. H. Jaffé, H. L. Jones, Adv. Hetrocyclic Chem. 1964, 3, 209.
(11) A. R. Katritzky, 'Handbook of Heterocyclic Chemistry',
Pergamon: Oxford, 1985.
(12) M. Julia, T. Huynh-Dinh, Bull. Soc. Chim. Fr. 1971, 1303.
(13) K. Hofman, in ‘Chemistry of Heterocyclic Compounds’, Ed.
A. Weissberger, Interscience: New York, 1953.
(14) R. Hoos, A. B. Naughton, A. Vasella, Helv. Chim. Acta 1993,
76, 1802.
Inhibition Studies
Determination of the inhibition constants (Ki) or the IC50 values
were performed with a range of inhibitor concentrations (typically
4–8 concentrations) which bracket the Ki or IC50 value.
a) Inhibition of Sweet Almonds b-Glucosidases
Inhibition constants (Ki) and IC50 were determined at 37 °C, using
a 0.08M KH2PO4/K2HPO4 buffer (pH 6.8) and 4-nitrophenyl b-D-
glucopyranoside as the substrate. The increase of absorption per
min at 400 nm was taken as velocity for the hydrolysis of the sub-
strate. The increase was linear during all measurements (2 min).
IC50 values were determined by plotting the velocity of substrate hy-
drolysis vs. the inhibitor concentration. Determination of the inhib-
itor concentration corresponding to half the velocity measured in
(15) H. S. Overkleeft, J. van Wiltenburg, U. K. Pandit,
Tetrahedron 1994, 50, 4215.
(16) H. Bader, J. D. Downer, P. Driver, J. Chem. Soc. 1950, 2775.
(17) H. Bader, J. D. Downer, J. Chem. Soc. 1953, 1636.
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York

1468
N. Panday, A. Vasella
PAPER
(18) J. Heyes, N. Ward, Chem. Abstr. 1972, 77, 19645a.
(19) A. McKillop, A. Henderson, P. S. Ray, C. Avendaño, E. G.
Molinero, Tetrahedron Lett. 1982, 23, 3357.
(20) E. Gomez, C. Avendaño, A. McKillop, Tetrahedron 1986, 42,
2625.
but not isolated, since they cyclised in situ to
aminoimidazoles. The preparation of N-cyanomethylamidines
by N-cyanomethylation of amidines with bromo- or
chloroacetonitrile has been used for the synthesis of
aminoimidazolium salt.21,22
(21) K. T. Potts, S. Husain, J. Org. Chem. 1971, 36, 3368.
(22) A. Chinone, S. Sato, M. Ohta, Bull. Chem. Soc. Jpn. 1971, 44,
826.
(38) According to reference 25, the NH bands of the tautomer with
an exocyclic double bond are expected at lower wave numbers
(3450 and 3350 cm-1).
(23) R. Hoos, A. B. Naughton, W. Thiel, A. Vasella, W. Weber, K.
Rupitz, S. G. Withers, Helv. Chim. Acta 1993, 76, 2666.
(24) G. Papandreou, M. K. Tong, B. Ganem, J. Am. Chem. Soc
1993, 115, 11682.
(25) H. U. Sieveking, W. Lüttke, Liebigs Ann. Chem. 1977, 2, 189.
(26) G. Legler, M.-T. Finken, Carbohydr. Res. 1996, 292, 103.
(27) T. Granier, F. Gaiser, L. Hintermann, A. Vasella, Helv. Chim.
Acta 1997, 80, 1443.
(28) N. Panday, T. Granier, A. Vasella, Helv. Chim. Acta 1998, 81,
475.
(29) P. Fowler, B. Bernet, A. Vasella, Helv. Chim. Acta 1996, 79,
269.
(30) M. Avalos, R. Babiano, C. J. Duran, J. L. Jimenez, J. C.
Palacios, J. Chem. Soc. Perkin Trans. 2 1992, 2, 2205–2215.
(31) F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M.
Lipton, C. Caufield, G. Chang, T. Hendrickson, W. C. Still, J.
Comput. Chem. 1990, 400.
(39) In view of the endocyclic double bond of the amindine, we
expected alkylation of the endocyclic rather than the exocyclic
nitrogen. This was confirmed later by comparison of the 2-
acetamidoimidazoles 22 and 24 with the 3-acetamidoimid-
azoles 29 and 30.
(40) There was no TLC evidence for the N-cyanomethylamidine1s
19 and 20.
(41) Due to signal overlap, the J(5,6) value of the unprotonated 12
could not be determined, but identical J(6,7) and J(7,8) values
for the protonated and unprotonated 12 suggest that its
conformation is close to 7S or 7H6.
(42) The data in reference 4 are those of the protonated 6.
(43) Only one set of signals is observed in the 1H NMR spectra of
12 and 14 in D2O.
(44) The conformation of the acetamido group is described by two
parameters: the rotation around the N, C(O) bond that gives
rise to (E) and (Z)-isomers, and rotation around the C(2,3), N
bond, leading to syn and anti isomers, as defined by the
dihedral angles N–C–N–H of 0 and 180°. It is known that (E)-
acetamides are much less stable than (Z)-acetamides.29,30
(45) The pKHA of 13 could not be determined, since no inflection of
its titration curve was observed. Probably the inflection starts
at a pH significantly below 3, where the measurement of the
pH is rather inaccurate. Its pKHA may be estimated at ca. 2.8
by substracting the difference between the pKHA of 3 and 6
from the pKHA of 11.
(32) R. Hoos, A. Vasella, K. Rupitz, S. G. Withers, Carbohydr.
Res. 1997, 298, 291.
(33) H. Lineweaver, D. Burk, J. Am. Chem. Soc. 1934, 56, 658.
(34) I. H. Segel, 'Enzyme Kinetics', Wiley: New York, 1975.
(35) The pKHA values of 3 and 6 were determined at 25 °C in H2O
by titration with HCl. The difference between the pKHA of 3
and 1-methylimidazole (7.16) was used to estimate the pKHA
of 1 and 2 from the pKHA values of 4-methyl-1,2,4-triazole
(3.47) and 1-methyl tetrazole (–3.08). The pKHA of 4 and 5
were estimated in a similar fashion using the difference
between the pKHA of 6 and the one of 1-methylimidazole.
(36) The positions 4 and 5 of imidazoles correspond to the
positions 2 and 3 of pyridoazoles.
(46) Prepared from aminoacetonitrile hydrochloride according to
reference 12.
(37) The preparation of aminoimidazoles by cyclization of N-
(cyanomethyl)amidines has been reported.16-20 The N-
(cyanomethyl)amidines were generated by condensation of
either iminoethers or iminothioethers with aminoacetonitrile
Article Identifier:
1437-210X,E;1999,0,SI,1459,1468,ftx,en;C01999SS.pdf
Synthesis 1999, No. SI, 1459–1468 ISSN 0039-7881 © Thieme Stuttgart · New York