Synthesis of new C(2)-substituted gluco-configured tetrahydroimidazopyridines and their evaluation as glucosidase inhibitors

Article abstract of DOI:10.1002/hlca.200590199

The gluco-configured C(2)-substituted tetrahydroimidazopyridines 8-14 were prepared and tested as inhibitors of the β-glucosidases from Caldocellum saccharolyticum and from sweet almonds, and of the α-glucosidase from brewer's yeast. All new imidazopyridi

Full text of DOI:10.1002/hlca.200590199

H
ELVETICA
CHIMICA
ACTA – Vol. 88 (2005)
2593
Synthesis of New C(2)-Substituted gluco-Configured
Tetrahydroimidazopyridines and Their Evaluation as Glucosidase Inhibitors
by Bhagavathy Shanmugasundaram and Andrea Vasella*
Laboratorium für Organische Chemie, ETH-Hönggerberg, Wolfgang Pauli-Strasse 10, CH-8093, Zürich
Albert Eschenmoser zum 80. Geburtstag mit herzlichen Glückwünschen zugeeignet.
The gluco-configured C(2)-substituted tetrahydroimidazopyridines 8–14 were prepared and tested as
inhibitors of the b-glucosidases from Caldocellum saccharolyticum and from sweet almonds, and of the a-glu-
cosidase from brewer’s yeast. All new imidazopyridines are nanomolar inhibitors of the b-glucosidases and
micromolar inhibitors of the a-glucosidase. The 3-phenylpropyl derivative 14 proved the strongest inhibitor
of the Caldocellum b-glucosidase (Ki =0.9 n
7, and the propyl derivative 13 is the strongest inhibitor of the sweet almond b-glucosidases (Ki =3.2 n
M
), only slightly weaker than the known 2-phenylethyl analogue
),
M
again slightly weaker than 7. There is no strong dependence of the inhibition on the nature of the C(2)-substitu-
ent and no clear correlation between the inhibitory strength of the known manno-configured imidazopyridines
2–6 and the gluco-analogues 8–12. While most manno-imidazopyridines are competitive inhibitors, the gluco-
analogues proved non-competitive inhibitors of the Caldocellum b-glucosidase and mixed-type or partial mixed-
type inhibitors of the sweet almond b-glucosidases.
Introduction. – A recent comparison of the inhibition of retaining b-glucosidases
and snail b-mannosidase [1–3] by C(2)-substituted manno- and gluco-configured tetra-
hydroimidazopyridines [4–11] and by appropriately configured isoquinuclidines [12–
14] strongly suggested that b-glucosidases and b-mannosidases use a different confor-
mational itinerary to attain a similar transition state. In the course of this work, we
noticed that the nature of the C(2)-substituent of the gluco-, manno-, and GlcNAc-
derived tetrahydroimidazopyridines had a analogous effect on the strength of the inhib-
ition of the b-glucosidases from Caldocellum saccharolyticum [15], b-mannosidase
from snail [8], and N-acetyl b-glucosaminidase from bovine kidney [16], respectively.
We also found that the manno-configured inhibitors 2–6 [17] for which there were
no gluco-configured counterparts are particularly strong inhibitors of snail b-mannosi-
dase, ca. 3–5 times stronger than 1 (Ki =20 n
configured 2-phenylethyl imidazopyridine 7 that is so far the strongest known inhibitor
of a retaining b-glucosidase (Ki =0.1 n , C. saccharolyticum) [15]. The parallel effect of
M
) [17], the manno-analogue of the gluco-
M
the C(2)-substituents on the strength of the inhibition suggested that the gluco-config-
ured imidazopyridines 8–12 corresponding to 2–6 should be stronger inhibitors than 7.
However, a comparison of the gluco- and manno-configured C(2)-substituted imidazo-
pyridines shows that the analogous effect of the C(2)-substituent on the strength of the
inhibition is not reflected by a similarity of the type of inhibition (mixed, competitive,
or non-competitive), casting some doubt on the validity of the above extrapolation. In
view of these considerations and of our interest in very strong glucosidase inhibitors, we
decided to prepare the gluco-analogues 8–12. Considering the strong inhibition by the
© 2005 Verlag Helvetica Chimica Acta AG, Zürich

2594
HELVETICA CHIMICA ACTA – Vol. 88 (2005)
2-phenylethyl derivative 7, we also wished to synthesise the propyl derivative 13 and
the 3-phenylpropyl derivative 14.
Synthesis. – We prepared the gluco-analogues 8–14 via the 2-iodo-imidazopyridine
15, according to the established synthesis of the corresponding manno-imidazopyri-
dines [17]. The compound 15 is available in five steps and in an overall yield of 50–
53% from 5-amino-2,3,4,6-tetra-O-benzyl-5-deoxy- -gluconolactam [18–20].
D
Sonogashira coupling [21] of 15 (Scheme) with phenylacetylene yielded 76% of the
known phenylethynyl derivative 16 [15] that was debenzylated (BCl3) to the desired
phenylethynyl imidazopyridine 8 (59%). Similarly, the Sonogashira coupling of 15
with 3-phenylprop-1-yne yielded 17 that was hydrogenated to the 3-phenylpropyl-imi-
dazopyridine 14 (31% from 15).
For the synthesis of the remaining inhibitors, we transformed 15 into the known car-
baldehyde 18 and the known alcohol 23 [15]. Wittig–Horner reaction of 18 with diethyl
benzylphosphonate gave the known 2-phenylethenyl derivative 19 [15] that was deben-
zylated with AlCl3 in the presence of N,N-dimethylaniline to yield 60% of 9. Wittig ole-
fination of 18 with (ethylidene)(triphenyl)phosphoraneyielded 88% of 20 as 45 :55 (E)/
(Z) mixture that was hydrogenated to provide the propyl-imidazopyridine 13 (82%).
The (phenylamino)methyl derivative 10 was obtained by reductive amination of 18
(77%). Treatment of 18 with aniline provided the (phenylimino)methyl derivative 21
that was reduced with NaBH4 and then debenzylated with BCl3 (61%). The (3-
nitrophenoxy)methyl derivative 11 was obtained in a yield of 34% by treatment of
the hydroxymethyl-imidazopyridine 23 with 1-fluoro-3-nitrobenzene and NaH, fol-
lowed by debenzylation of the resulting 25 with AlCl3 in the presence of anisole,
while the phenoxymethyl derivative 26 was prepared in a yield of 77% by alkylation
of phenol with the (chloromethyl)-imidazopyridine 24, followed by hydrogenolysis
(63%). The compound 24 was readily obtained in a yield of 91% from 23 and SOCl2.
Enzymatic Tests and Discussion. – The gluco-imidazopyridines 8–14 were tested as
inhibitors of the b-glucosidase from Caldocellum saccharolyticum (0.08
buffer, pH 6.8, 558), the b-glucosidases from sweet almonds (0.08 phosphate buffer,
pH 6.8, 378), and the a-glucosidase from brewer’s yeast (0.08 phosphate buffer, pH
M
phosphate
M
M
6.8, 378). Table 1 summarises the results, listing the manno- and the corresponding
gluco-configured imidazopyridines in order of decreasing inhibitory strength of the
manno-isomers. Although all gluco-imidazopyridines are nanomolar inhibitors of the

H
ELVETICA
CHIMICA
ACTA – Vol. 88 (2005)
2595
Scheme
a) Phenylacetylene, [Pd(PPh3)4], CuI, Et3N; 76%. b) BCl3, CH2Cl2 ; 59%. c) 3-Phenylprop-1-yne,
[Pd(PPh3)4], CuI, Et3N; 58%. d) H2, Pd/C, AcOH; 53%. e) Diethyl benzylphosphonate, t-BuOK, THF;
81%. f) AlCl3, CH2Cl2, N,N-dimethylaniline; 75%. g) Ethyl(triphenyl)phosphonium bromide, BuLi, THF;
88%. h) Pd/C, 6 bar of H2, AcOH, 82%. i) Aniline, 4-Å mol. sieves, CH2Cl2. j) NaBH4, EtOH; 77% from
18. k) BCl3, CH2Cl2, ꢀ788 to 158; 61%. l) SOCl2, CH2Cl2 ; 91%. m) 1-Fluoro-3-nitrobenzene, NaH, DMF;
59% of 25. n) PhOH, t-BuOK, DMF; 77% of 26. o) AlCl3, anisole, CH2Cl2; 58% of 11. p) H2, Pd(OH)2/C,
AcOEt/MeOH/H2O/AcOH; 63% of 12.

2596
HELVETICA CHIMICA ACTA – Vol. 88 (2005)
Table 1. Inhibition of the b-Glucosidases from C. saccharolyticum and Sweet Almonds, the a-Glucosidase from Brewer’s
Yeast, and the b-Mannosidase from Snail by the the C(2)-Substituted gluco- and manno-Imidazoles: Ki and pKHA Values
Inhibition of
b-mannosidasea)b)
Inhibition of b-glucosidases
Inhibition of
a-glucosidase
R
manno-
Imidazoles
pKHA Ki
[n
gluco-
]c) Imidazoles
pKHA
from C.
sacch.d)e)
from sweet
almondsf)g)
from brewer’s
yeastf)h)
M
Ki [nM]
Ki [nM](a)
Ki [nM]
PhCH=CH
PhCꢁC
3
2
4
5
6
4.77
6
7
8
9
8
4.83
)
5.62/9.51 2.3
4.46
4.69
6.42
6.63
6.03
2.6
7.4
7.6 (1.3)
2.4 (1.1)
5.4 (4.4)
6.6 (2.0)
9.7 (2.0)
3.2 (1.1)
8.5 (0.9)
1.2
3230
7300
1920
1770
3220
1890
450
i
i
)
PhCH2NH
5.09j)
10
11
12
13
14
7 k)
3-NO2ꢀC6H4OCH2
4.36 12
4.39 12
2.2
2.4
1.6
0.9
PhOCH2
MeCH2CH2
PhCH2CH2CH2
PhCH2CH2
1
6.04 20
0.11l)
500
a) At 258 and pH 4.5. b) Data taken from [17]. c) Competitive-type inhibition except for 3. d) At 558 and pH 6.8. e) Non-
competitive-type inhibition. f) At 378 and pH 6.8. g) Mixed-type inhibition except for 14 (partial mixed type) [22]. h) IC50/
2. i) No inflection of the titration curve was observed within pH 2.1–5.4. j) Second pK value not reported. k) Data taken
l
from [15]. ) Mixed-type inhibition (a=15).
b-glucosidases, none was stronger than the 2-phenylethyl derivative 7. The strength of
the inhibition of the C. saccharolyticum b-glucosidase does not vary widely, with Ki val-
ues of 0.9 and 7.4 n
M
for the strongest and weakest inhibitors 14 and 8, respectively, that
correspond to 2–6.
This is also observed for the inhibition of the sweet almond b-glucosidases that were
inhibited less strongly than the Caldocellum enzyme, as noticed earlier for related imi-
dazopyridines [15]. The extreme Ki values for the gluco-analogues of the manno-imida-
zopyridines 2–6 are 24 n
pyl derivative 13 with a Ki of 1.6 n
the sweet almond enzymes. The 3-phenylpropyl derivative 14 proved the strongest
inhibitor of the Caldocellum b-glucosidase (Ki =0.9 n ). It is, however, a weaker inhib-
M
for 8 and 5.4 n
M
for 10. An even stronger inhibitor is the pro-
M
for the Caldocellum b-glucosidase and 3.2 n
M
for
M
itor of the sweet almond enzyme than 13. In contradistinction to the inhibition by the
manno-configured imidazopyridines that is competitive (with an exception for 3), the
gluco-analogues showed a non-competitive-type inhibition of the Caldocellum and a
mixed-type inhibition of the sweet almond b-glucosidases characterised by small a-val-
ues with the exception of 14 that proved a partial mixed-type inhibitor. With micromo-
lar Ki values characterising the inhibition of brewer’s yeast a-glucosidase the gluco-imi-
dazopyridines 7–14 are rather selective inhibitors of b-glucosidases. There is no obvi-
ous correlation of the strength of the inhibition with basicity. In view of the narrow
range of the strength of the inhibition, one may interpret the absence of a clear corre-
spondence of the influence of the nature of the C(2)-substituent on the inhibition by the
manno- and gluco-imidazopyridines as denoting the limits of the effect of these aglycon
mimics on the strength of the inhibition rather than as indicating a different nature of
the reactive intermediates.

H
ELVETICA
CHIMICA
ACTA – Vol. 88 (2005)
2597
We thank Dr. B. Bernet for checking the experimental part, M. Schneider and D. Manser for the pKHA
determination, and the Swiss National Science Foundation and F. Hoffmann-La Roche, Basel, for generous
financial support.
Experimental Part
General. See [16].
(5R,6R,7S,8S)-5,6,7,8-Tetrahydro-5-(hydroxymethyl)-2-(2-phenylethynyl)imidazo[1,2-a]pyridine-6,7,8-triol
(8). A soln. of 16 [15] (38 mg, 57.5 mmol) in CH2Cl2 (1.5 ml) was cooled to ꢀ788, treated with 1
M BCl3 in CH2Cl2
(0.9 ml, 0.923 mmol), and stirred until the mixture had reached 58 (ca. 5 h). The mixture was cooled to ꢀ788,
treated with H2O (1 ml), and neutralised with aq. NH3 (1 ml). Evaporation and FC (AcOEt/MeOH/H2O
1 :0 :0 ! 20 :1 :1 ! 15 :1 :1) gave 8 (10 mg, 59%). White hygroscopic solid. Rf (AcOEt/MeOH/H2O 10 :1 :1)
1
0.15. H-NMR (CD3OD, 300 MHz): see Table 2 ; additionally, 7.32–7.36 (m, HꢀC(3), HꢀC(4), and HꢀC(5)
of Ph); 7.45–7.48 (m, HꢀC(2) and HꢀC(6) of Ph). 13C-NMR (CD3OD, 75 MHz): see Table 3, additionally,
83.39 (s, CꢁCꢀC(2)); 89.76 (s, CꢁCꢀC(2)); 124.06 (d, C(4) of Ph); 124.60 (s, C(1) of Ph, C(2)); 129.06 (2d);
131.80 (2d);. HR-MALDI-MS: 323.1000 (49, [M+Na]+, C16H16N2NaO4þ ; calc. 323.1008), 301.1183 (58,
[M+H]+, C16H17N2Oþ4 ; calc. 301.1188), 283.1071 (100, [MꢀOH]+, C16H15N2Oþ3 ; calc. 283.1083). Anal. calc.
for C16H16N2O4 ·H2O (318.12): C 60.37, H 5.70, N 8.80; found: C 60.24, H 5.64, N 8.78.
(5R,6R,7S,8S)-6,7,8-Tris(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-2-[(3-phenylprop-1-ynyl)-
imidazo[1,2-a]pyridine (17). A soln. of 15 (172 mg, 0.25 mmol), CuI (5 mg, 0.026 mmol), and Et3N (174 ml, 1.25
mmol) in degassed DMF (7 ml) was treated with Pd(PPh3)4 (14 mg, 0.012 mmol), degassed with Ar, treated with
PhCH2CꢁCH (93 ml, 0.75 mmol), heated to 808, stirred for 2 h, cooled to r.t., diluted with Et2O, washed with sat.
NH4Cl soln. and H2O, dried (Na2SO4), and evaporated. FC (hexane/AcOEt 3 :2) gave 17 (98 mg, 58%). Pale
brown oil. Rf (hexane/AcOEt 2 :3) 0.52. [a]2D5 =+28.7 (c=0.88, CHCl3). UV (CHCl3): 295 (3.3), 245 (4.2). IR
(CHCl3): 3067w, 3023s, 3015s, 2869m, 2123w, 1602m, 1496m, 1454s, 1361m, 1336m, 1095s, 1028s, 911w. 1H-
NMR (CDCl3, 300 MHz): see Table 4 ; additionally, 3.86 (br. s, PhCH2CꢁC); 4.44 (d, J=12.1, PhCH); 4.48
(d, J=12.3, PhCH); 4.50 (d, J=11.8, PhCH); 4.64 (d, J=11.3, PhCH); 4.80 (d, J=11.3, PhCH); 4.81 (d,
J=12.3, PhCH); 4.87 (d, J=12.1, PhCH); 5.19 (d, J=11.5, PhCH); 7.16–7.18 (m, 3 arom. H); 7.25–7.37 (m,
18 arom. H); 7.43–7.45 (m, 4 arom. H). 13C-NMR (CDCl3, 75 MHz): see Table 3 ; additionally, 26.00 (t,
PhCH2CꢁC); 72.75, 73.28, 73.80, 73.98 (4t, 4 PhCH2); 126.46 (2d); 127.49 (2d); 127.80–128.44 (several d);
136.54, 137.02, 137.32, 137.53, 137.97 (5s). HR-MALDI-MS: 697.2989 (23, [M+Na]+, C45H42N2NaO4þ ; calc.
697.3042), 676.3246 (20), 675.3209 (39, [M+H]+, C45H43N2Oþ4 ; calc. 675.3223), 568.2649 (42), 567.2614 (100,
[MꢀBnO]+, C38H35N2Oþ3 ; calc. 567.2648).
Table 2. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Deprotected C(2)-Sub-
stituted gluco-Imidazoles 8–14 in CD3ODa)
Compound
8
9
10
7.16
11
7.43
12
7.37
13
7.12
14
6.99
HꢀC(3)
HꢀC(5)
HꢀC(6)
HꢀC(7)
HꢀC(8)
CHꢀC(5)
CH’ꢀC(5)
J(5,6)
7.59
3.89–3.91
3.82
3.70
4.49
3.96
4.18
8.7
7.38
3.85–3.90
3.81
3.71
4.52
3.95
4.19
8.4
3.77–3.82
3.78
3.66
4.47
3.91
4.13
8.4
3.87–3.89
3.81
3.68
4.49
3.94
4.17
8.6
3.86–3.89
3.81
3.68
4.49
3.93
4.17
8.7
3.87–3.89
3.81
3.68
4.52
3.94
4.17
8.4
3.80–3.84
3.78
3.66
4.46
3.92
4.15
8.1
J(6,7)
8.7
8.4
8.1
8.9
8.7
8.7
8.4
J(7,8)
8.1
7.8
7.8
8.0
7.8
7.5
7.8
J(5,CH)
J(5,CH’)
4.0
2.0
3.4
1.5
3.9
2.1
4.3
)
4.3
1.9
3.9
1.5
3.9
2.1
b
J(CH,CH’) 10.8
11.8
12.0
11.0
11.5
11.5
12.0
a) Assignment based on homodecoupling experiments. b) Not assigned.

2598
HELVETICA CHIMICA ACTA – Vol. 88 (2005)
Table 3. Selected 13C-NMR Chemical Shifts [ppm] of the C(2)-Substituted gluco-Imidazoles 8–14, 16, 17, 19, 20,
22, 25, and 26
Compd.
Solvent
C(3)
C(5)
CH2ꢀC(5)
C(6)
C(7)
C(8)
C(8a)
16
17
19
CDCl3
CDCl3
CDCl3
C6D6
122.13
121.12
115.95
113.52/116.17
114.42
117.05
116.27
122.45
116.50
115.38
117.88
116.57
114.16
113.67
58.42
58.34
58.07
58.03/58.11
58.07
58.27
58.12
62.79
62.53
62.44
62.59
61.70
61.97
61.60
68.49
68.28
68.47
68.46
68.23
68.42
68.25
61.02
61.08
61.09
61.15
60.28
59.98
60.17
74.16
75.78
76.30
76.27
75.93
75.86
75.92
69.37
69.47
69.48
69.46
68.63
68.16
68.50
81.72
81.27
82.03
82.50
81.85
81.62
81.85
76.06
76.23
76.21
76.25
75.44
75.15
75.44
73.84
73.99
74.24
76.17
74.29
73.98
73.96
68.79
68.87
69.05
68.90
68.09
67.96
68.07
144.47
143.53
144.73
143.79/144.41
143.56
144.11
143.71
148.02
147.96
147.09
147.21
146.88
145.69
145.79
20a)
22b)
25b)
26b)
8
CDCl3
CDCl3
CDCl3
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
9
10
11
12
13
14
a) (Z)/(E) 55 : 45. b) Assignment based on a HSQC spectrum.
Table 4. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Protected C(2)-Substituted
gluco-Imidazoles 16, 17, 19, 20, 22, 25, and 26
Compound
16a)
17
19a)
20b)
22c)
25c)
26c)
d
HꢀC(3)
HꢀC(5)
HꢀC(6)
HꢀC(7)
HꢀC(8)
CHꢀC(5)
CH’ꢀC(5)
J(5,6)
)
7.19
7.06
6.82/6.94
6.93
7.15
7.11
4.23
4.15–4.20
3.80–3.84
4.10
4.74
3.72
4.19
3.86
4.11
4.78
3.75
3.86
7.8
3.89–3.92
3.66/3.67
4.02/4.03
4.77
4.12–4.16
3.88
4.09
4.74
3.72
3.82
7.6
4.19
3.85
4.11
4.75
3.74
3.85
7.2
4.15–4.20
3.86
4.11
4.76
3.74
3.85
7.8
3.88
4.14
4.76
3.76
3.87
6.9
3.40/3.41
3.49
3.82
e
e
)
)
J(6,7)
J(7,8)
J(5,CH)
J(5,CH’)
J(CH,CH’)
6.9
5.0
5.3
2.8
7.1
5.2
5.2
7.2
5.3
5.6
2.2
8.1
5.9
5.1
2.8
7.5
5.6
5.3
2.8
7.2
5.3
5.3
3.4
7.5
5.3
5.3
3.1
e
)
10.3
10.5
10.6
10.3
10.3
10.9
11.5
a) Data taken from [15]. b) (Z)/(E) 55 : 45 and in solvent C6D6. c) Assignment based on a HSQC spectrum.
d) Hidden by aromatic signals. e) Not assigned.
(5R,6R,7S,8S)-5,6,7,8-Tetrahydro-5-(hydroxymethyl)-2-(3-phenylpropyl)imidazo[1,2-a]pyridine-6,7,8-triol
(14). A soln. of 17 (41 mg, 0.061 mmol) in AcOH (3 ml) was treated with 10% Pd/C (40 mg), hydrogenated for
72 h under 6 bar of H2, diluted with MeOH, and filtered over Celite. Evaporation, co-evaporation with toluene,
and FC (AcOEt/MeOH/H2O 10 :0 :0 ! 10 :1 :0 ! 10 :3 :1) gave 14 (10 mg, 53%). White solid. A sample for
microanalysis was dried for 4 d at 10ꢀ4 Torr. Rf (AcOEt/MeOH/H2O 10 :1 :1) 0.22. [a]2þ5 =ꢀ28.4 (c=0.53,
MeOH). UV (MeOH): 209 (4.0). IR (KBr): 3500–3180s (br.), 2922s, 2851m, 1631w, 1495w, 1452m, 1179w,
1104m, 1018m, 870w, 747w, 699m. 1H-NMR (CD3OD, 300 MHz): see Table 2 ; additionally, 1.12–1.28 (1 H),
1.61–1.71 (1 H), 1.88–1.98 (1 H), 2.48–2.67 (3 H) (4m, PhCH2CH2CH2); 3.66 (t, Jꢂ8.4, irrad. 4.15 ! d,
Jꢂ8.1, HꢀC(7)); 4.46 (d, J=7.8, irrad. 3.66 ! s, HꢀC(8)); 7.12–7.26 (m, 5 arom. H). 13C-NMR (CD3OD,
75 MHz): see Table 3 ; additionally, 31.12, 33.36, 35.26 (3t, PhCH2CH2CH2); 125.55 (d); 128.11 (2d); 128.30
(2d); 141.59, 142.35 (2s, C(2), C(1) of Ph). HR-MALDI-MS: 341.1474 (27, [M+Na]+, C17H22N2NaO4þ ; calc.
341.1477), 325.2114 (55), 320.1690 (25), 319.1649 (100, [M+H]+, C17H23N2O4þ ; calc. 319.1658), 301.1546 (15,

H
ELVETICA
CHIMICA
ACTA – Vol. 88 (2005)
2599
[MꢀOH]+, C17H21N2Oþ3 ; calc. 301.1552). Anal. calc. for C17H22N2O4 ·H2O (327.37): C 60.88, H 6.91, N 8.35;
found: C 60.66, H 6.66, N 8.35.
(5R,6R,7S,8S)-5,6,7,8-Tetrahydro-5-(hydroxymethyl)-2-[(E)-2-phenylethenyl]imidazo[1,2-a]pyridine-6,7,8-
triol (9). A soln. of 19 (29 mg, 43.6 mmol) in CH2Cl2 (1 ml) was treated with AlCl3 (93 mg, 0.69 mmol) and N,N-
dimethylaniline (66 ml, 0.52 mmol), stirred at 238 for 12 h, diluted with AcOEt, and extracted with H2O. Evap-
oration of the aq. layer and FC (AcOEt/MeOH/H2O 1 :0 :0 ! 20 :1:1 ! 15 :1:1) gave 9 (10 mg, 75%). White
hygroscopic solid. A sample for microanalysis was dried for 4 d at 10ꢀ4 Torr. Rf (AcOEt/MeOH/H2O 10 :1:1)
0.25. 1H-NMR (CD3OD, 300 MHz): see Table 2 ; additionally, 3.71 (br. t, Jꢂ8.4, HꢀC(7)); 7.00 (d, J=17.1,
CH=CH); 7.14–7.21 (m, HꢀC(4) of Ph, CH=CH); 7.30 (dd, J=7.8, 7.1, HꢀC(3) and HꢀC(5) of Ph); 7.45–
7.48 (m, HꢀC(2) and HꢀC(6) of Ph). 13C-NMR (CD3OD, 75 MHz): see Table 3 ; additionally, 120.65 (d,
PhCH=CH); 126.69 (d, C(3) and C(5) of Ph); 127.68, 127.72 (2d, PhCH=CH, C(4) of Ph); 129.13 (d, C(2)
and C(6) of Ph); 138.58 (s, C(1) of Ph); 141.10 (s, C(2)). HR-MALDI-MS: 303.1336 (100, [M+H]+, C16H19N2
O4þ ; calc. 303.1345), 285.1230 (24, [MꢀOH]+, C16H17N2Oþ3 ; calc. 285.1239). Anal. calc. for C16H18N2O4 ·0.5
MeOH (318.34): C 62.25, H 6.33, N 8.80; found: C 62.18, H 6.31, N 8.84.
(5R,6R,7S,8S)-6,7,8-Tris(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-2-[(E/Z)-2-prop-1-enyl]imi-
dazo[1,2-a]pyridine (20). A suspension of EtPPh3Br (137 mg, 0.36 mmol) in Et2O (5 ml) was cooled to ꢀ408,
treated slowly with 1.5M BuLi in hexane (195 ml, 0.29 mmol), stirred for 10 min, warmed to r.t, stirred for
15 min, cooled to ꢀ788, treated with a soln. of 18 (144 mg, 0.25 mmol) in THF (5 ml), warmed to r.t. within
20 min, cooled to ꢀ788, and treated with sat. NH4Cl soln. Normal workup gave (Z)/(E)-20 55 :45 (129 mg,
88%). Rf (hexane/AcOEt 3 :2) 0.60. IR (CHCl3): 3089m, 3066s, 3011s, 2918m, 2869m, 1951w, 1603s, 1504s,
1454s, 1430m, 1311m, 1179m, 1070s, 1028m, 911m. 1H-NMR (C6D6, 300 MHz; (Z)/(E) 55 :45): 1.81 (dd,
J=1.5, 6.7, 1.35 H), 2.23 (dd, J=1.7, 7.0, 1.65 H) (Me); 3.40 (dd, J=5.3, 10.0, 0.45 H), 3.41 (dd, J=5.1, 10.2,
0.55 H) (CHꢀC(5)); 3.49 (dd, J=2.8, 10.3, CH’ꢀC(5)); 3.66 (t, J=8.1, 0.45 H), 3.67 (t, J=8.1, 0.55 H) (Hꢀ
C(6)); 3.85–3.92 (m, HꢀC(5)); 4.02, 4.03 (2dd, J=5.9, 8.1, HꢀC(7)); 4.07, 4.12 (2d, J=11.5, PhCH); 4.33,
4.35 (2d, J=11.5, PhCH); 4.58 (d, J=11.5, PhCH); 4.77 (d, J=5.9, HꢀC(8)); 4.78, 4.80 (2d, J=11.2, 2
PhCH); 5.12 (d, J=11.8, PhCH); 5.48 (d, J=11.7, 0.45 H), 5.51 (d, J=11.7, 0.55 H) (PhCH); 5.75 (qd, J=7.8,
11.5, 0.55 H), 6.77 (qd, J=7.8, 15.5, 0.45 H) (CH=CHMe); 6.50 (qd, J=1.5, 15.5, 0.45 H), 6.70 (qd, J=1.5,
11.5, 0.55 H) (CH=CHMe); 6.82 (s, 0.45 H), 6.94 (s, 0.55 H) (HꢀC(3)); 7.06–7.26 (m, 16 arom. H); 7.24 (d,
J=8.1, 2 arom. H); 7.53 (d, J=7.8, 2 arom. H). 13C-NMR (C6D6, 75 MHz; (Z)/(E) 55:45): 15.41, 18.31 (2q,
Me); 58.03, 58.11 (2d, C(5)); 68.46 (t, CH2ꢀC(5)); 72.56, 73.00, 74.01, 74.05 (4t, 4 PhCH2); 76.17 (d, C(8));
76.27 (d, C(6)); 82.50 (d, C(7)); 113.52, 116.17 (2d, C(3)); 123.43, 123.85, 124.05, 124.10 (4d, CH=CH);
127.58–128.77 (several d); 137.83, 137.88, 138.49, 138.71, 139.01 (5s); 141.08, 141.56 (2s); 143.79, 144.41 (2s,
C(8a)). HR-MALDI-MS: 623.2881 (9, [M+Na]+, C39H40N2NaO4þ ; calc. 623.2886), 601.3058 (100, [M+H]+,
C39H41N2Oþ4 ; calc. 601.3066), 509.2427 (10, [MꢀBn]+, C32H33N2O4þ ; calc. 509.2440), 493.2483 (99,
[MꢀBnO]+, C32H33N2Oþ3 ; calc. 493.2491), 387.2059 (11, [Mꢀ2 BnO+H]+, C25H27N2Oþ2 ; calc. 387.2073).
(5R,6R,7S,8S)-5,6,7,8-Tetrahydro-5-(hydroxymethyl)-2-propylimidazo[1,2-a]pyridine-6,7,8-triol (13). A
soln. of 20 (60 mg, 0.1 mmol) in AcOH (5 ml) was treated with 10% Pd/C (60 mg) and hydrogenated at 6 bar
for 68 h, diluted with MeOH, and filtered over Celite. Evaporation of the filtrate, co-evaporation with toluene,
and FC (AcOEt/MeOH/H2O 10 :0 :0 ! 10 :1 :0 ! 10 :2 :1) gave 13 (20 mg, 82%). White hygroscopic solid. A
sample for microanalysis was dried for 4 d at 10ꢀ4 Torr. Rf (AcOEt/MeOH/H2O 10 :1 :1) 0.20. [a]D25 =ꢀ30.8
(c=0.35, MeOH). UV (MeOH): 224 (3.7). IR (ATR): 3500–3100s (br.), 2959s, 2926s, 2851s, 1505m, 1455s,
1318s, 1254m, 1198m, 1115s, 1026s, 867m. 1H-NMR (CD3OD, 300 MHz): see Table 2 ; additionally, 0.96 (t,
J=7.5, Me); 1.65 (sext., J=7.5, MeCH2CH2); 2.53 (t, J=7.5, MeCH2CH2); 4.52 (d, J=8.0, irrad. at 3.68 ! s,
HꢀC(8)). 13C-NMR (CD3OD, 75 MHz): see Table 3; additionally, 12.89 (q, Me); 22.30 (t, MeCH2CH2); 28.96
(t, MeCH2CH2); 140.38 (s, C(2)). HR-MALDI-MS: 265.1156 (5, [M+Na]+, C11H18N2NaO4þ ; calc. 265.1164),
243.1336 (100, [M+H]+, C11H19N2O4þ ; calc. 243.1345). Anal. calc. for C11H18N2O4 ·1/3 H2O (248.27): C 53.21,
H 7.58, N 11.28; found: C 53.12, H 7.28, N 11.28.
(5R,6R,7S,8S)-6,7,8-Tris(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-2-[(phenylimino)methyl]-
imidazo[1,2-a]pyridine (21). A suspension of 18 (84 mg, 0.142 mmol) and 4-Å activated mol. sieves in CH2Cl2
(2 ml) was treated with freshly distilled aniline (16 ml, 0.17 mmol), stirred at 238 for 5 h, and filtered. The filtrate
was diluted with Et2O, washed with sat. NaHCO3 soln. and brine, dried (Na2SO4), and evaporated. The 1H-NMR
spectrum of the residue (105 mg) evidenced 21. 1H-NMR (CDCl3, 300 MHz): 3.81 (dd, J=5.0, 10.6, CHꢀC(5));
3.88–3.93 (m, CH’ꢀC(5), HꢀC(6)); 4.16 (dd, J=5.0, 6.8, HꢀC(7)); 4.25–4.30 (m, HꢀC(5)); 4.48 (d, J=12.0,
PhCH); 4.50 (s, PhCH2); 4.67 (d, J=11.0, PhCH); 4.81 (d, J=10.5, PhCH); 4.82 (d, J=4.8, HꢀC(8)); 4.83,
4.87 (d, J=11.7, 2 PhCH); 5.15 (d, J=11.2, PhCH); 7.14–7.45 (m, 25 arom. H); 7.72 (s, HꢀC(3)); 8.48 (s,
CH=NPh).

2600
HELVETICA CHIMICA ACTA – Vol. 88 (2005)
(5R,6R,7S,8S)-6,7,8-Tris(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-2-[(phenylamino)methyl]-
imidazo[1,2-a]pyridine (22). A soln. of crude 21 (105 mg) in EtOH (10 ml) was treated with NaBH4 (18 mg,
0.5 mmol), stirred at 238 for 3 h, and evaporated. The residue was treated with sat. NH4Cl soln. and extracted
with Et2O. The org. layer was dried (Na2SO4) and evaporated. FC (hexane/AcOEt 1 :1) gave 22 (80 mg, 84%
from 18). White solid. M.p. 948. Rf (hexane/AcOEt 2 :3) 0.40. [a]2D5 =+45.9 (c=0.96, CHCl3). UV (CHCl3):
295 (3.3), 245 (4.2). IR (CHCl3): 3319m, 3059m, 3030s, 2870s, 1954w, 1819w, 1602s, 1508s, 1497m, 1454m,
1
1435m, 1309s, 1256m, 1069s, 989m. H-NMR (CDCl3, 300 MHz; assignment based on a HSQC spectrum): see
Table 4 ; additionally, 4.21 (br. s, NH); 4.28 (s, CH2ꢀC(2)); 4.41 (d, J=12.1, PhCH); 4.46 (d, J=12.1, PhCH);
4.50 (d, J=11.2, PhCH); 4.70 (d, J=11.2, PhCH); 4.81 (d, J=11.2, PhCH); 4.83 (d, J=11.2, PhCH); 4.84 (d,
J=11.2, PhCH); 5.16 (d, J=11.5, PhCH); 6.70–6.75 (m, 3 arom. H); 7.17–7.34 (m, 20 arom. H); 7.43 (dd,
J=1.5, 7.6, 2 arom. H). 13C-NMR (CDCl3, 75 MHz; assignment based on a HSQC spectrum): see Table 3; addi-
tionally, 42.52 (t, CH2ꢀC(2)); 72.69, 73.25, 74.01, 74.18 (4t, 4 PhCH2); 113.05 (2d); 117.31 (d); 127.48–129.03
(several d); 137.17, 137.48, 137.71, 138.11 (4s); 140.18 (s, C(2)); 148.23 (s, C(1) of NHPh). HR-MALDI-MS:
688.3123 (26, [M+Na]+, C43H43N3NaO4þ ; calc. 688.3151), 666.3319 (29, [M+H]+, C43H44N3Oþ4 ; calc.
666.3332), 574.2773 (39), 573.2733 (100, [MꢀPhNH]+, C37H37N2O4þ ; calc. 573.2753). Anal. calc. for C43H43N3
O4 (665.83): C 77.57, H 6.51, N 6.31; found: C 77.53, H 6.73, N 6.24.
(5R,6R,7S,8S)-5,6,7,8-Tetrahydro-5-(hydroxymethyl)-2-[(phenylamino)methyl]imidazo[1,2-a]pyridine-6,7,
8-triol (10). A soln. of 22 (38 mg, 57.1 mmol) in CH2Cl2 (3 ml) was cooled to ꢀ788, treated with 1
M BCl3 in CH2Cl2
(0.68 ml, 0.685 mmol), stirred until the mixture had reached a temp. of 108, cooled to ꢀ788, diluted with H2O,
and washed with AcOEt. The aq. layer was evaporated. The residue was dissolved in H2O and applied to ion-
exchange chromatography (Amberlite CG-120, H+ form, elution with 0.1
M NH3). Evaporation, lyophilisation,
and drying gave 10 (15 mg, 86%). Hygroscopic white solid. A sample for microanalysis was dried for 4 d at
10ꢀ4 Torr. Rf (AcOEt/MeOH/H2O 10 :1 :1 0.12. [a]2D5 =ꢀ25.7 (c=0.44, MeOH). UV (MeOH): 293 (3.2), 245
(4.1), 205 (4.3). IR (ATR): 3300–3200s (br.), 3026m, 2882m, 1600s, 1496s, 1318m, 1256m, 1177m, 1087s,
1025s, 907m. 1H-NMR (CD3OD, 300 MHz): see Table 2 ; additionally, 4.19 (s, CH2ꢀC(2)); 4.47 (d, J=7.8,
irrad. at 3.66 ! s, HꢀC(8)); 6.60 (tt, J=7.3, 0.9, HꢀC(4) of Ph); 6.67 (dt, J=7.5, 0.9, HꢀC(2) and HꢀC(6)
of Ph); 7.05–7.11 (m, HꢀC(3) and HꢀC(5) of Ph). 13C-NMR (CD3OD, 75 MHz): see Table 3 ; additionally,
42.45 (t, CH2ꢀC(2)); 113.90 (2d); 117.84 (d); 129.48 (2d); 141.03 (s, C(2)); 149.41 (s, C(1) of Ph). HR-
MALDI-MS: 328.1273 (33, [M+Na]+, C15H19N3NaO4þ ; calc. 328.1273), 306.1450 (100, [M+H]+, C15H20N3Oþ4 ;
calc. 306.1454), 213.0871 (43, [MꢀPhNH]+, C9H13N2O4þ ; calc. 213.0875). Anal. calc. for C15H19N3O4 ·1/2 H2O
(314.33): C 57.31, H 6.31, N 13.37; found: C 57.01, H 6.06, N 13.19.
(5R,6R,7S,8S)-6,7,8-Tris(benzyloxy)-5-[(benzyloxy)methyl]-2-(chloromethyl)-5,6,7,8-tetrahydroimidazo[1,
2-a]pyridine (24). A soln. of 23 [15] (170 mg, 0.288 mmol) in CH2Cl2 (7 ml) was treated with SOCl2 (42 ml,
0.58 mmol), stirred at 228 for 25 min, treated with sat. NaHCO3 soln., and extracted with Et2O. The organic
layer was washed with sat. NaHCO3 soln. and brine, dried (Na2SO4), and evaporated affording crude 24 (161
mg, sufficiently pure for the next reaction). FC (hexane/AcOEt 1 :1) gave pure 24 (136 mg, 77%). White
solid. Rf (hexane/AcOEt 3 :2) 0.52. M.p. 80.98. [a]2D5 =+50.4 (c=0.94, CHCl3). UV (CHCl3): 240 (3.6). IR
(CHCl3): 3067w, 3032m, 3020s, 3013s, 2869m, 1497m, 1454s, 1362m, 1259m, 1216m, 1094s, 1028m. 1H-NMR
(CDCl3, 300 MHz): 3.76 (dd, J=5.3, 10.4, CHꢀC(5)); 3.86 (dd, J=3.1, 10.1, CH’ꢀC(5)); 3.88 (t, J=7.5, Hꢀ
C(6)); 4.14 (dd, J=5.6, 7.3, HꢀC(7)); 4.18–4.21 (m, HꢀC(5)); 4.47 (d, J=12.1, PhCH); 4.53 (d, J=12.1,
PhCH); 4.63 (d, J=12.1, PhCH); 4.65 (s, CH2ꢀC(2)); 4.71 (d, J=11.5, PhCH); 4.78 (d, J=5.4, HꢀC(8));
4.84 (d, J=10.9, PhCH); 4.86 (d, J=11.5, PhCH); 4.88 (d, J=11.2, PhCH); 5.20 (d, J=11.5, PhCH); 7.09 (s,
HꢀC(3)); 7.22–7.24 (m, 2 arom. H); 7.30–7.37 (m, 16 arom. H); 7.39–7.47 (m, 2 arom. H). 13C-NMR
(CDCl3, 75 MHz): 40.36 (t, CH2ꢀC(2)); 58.35 (d, C(5)); 68.45 (t, CH2ꢀC(5)); 72.81, 73.46, 74.10 (3t, 3
PhCH2); 74.18 (d, C(8)); 74.30 (t, PhCH2); 76.09 (d, C(6)); 81.85 (d, C(7)); 116.78 (d, C(3)); 127.82–128.79 (sev-
eral d); 137.57, 137.75, 137.97, 138.39, 138.84 (5s); 144.33 (s, C(8a)). HR-MALDI-MS: 631.2337 (31, [M+Na]+,
C37H37ClN2NaOþ4 ; calc. 631.2340), 609.2509 (90, [M+H]+, C37H38ClN2O4þ ; calc. 609.2520), 575.2901 (36,
[MꢀCl+2 H]+, C37H39N2O4þ ; calc. 575.2910), 501.1942 (100, [MꢀBnO]+, C30H30ClN2Oþ3 ; calc. 501.1945).
Anal. calc. for C37H37ClN2O4 (609.16): C 72.95, H 6.12, N 4.60; found: C 73.07, H 6.32, N 4.62.
(5R,6R,7S,8S)-6,7,8-Tris(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-2-[(3-nitrophenoxy)methyl]-
imidazo[1,2-a]pyridine (25). A suspension of 23 (190 mg, 0.32 mmol) and NaH (60% washed with dry hexane,
23 mg, 0.96 mmol) in degassed DMF (3 ml) was treated with 1-fluoro-3-nitrobenzene (68 ml, 0.64 mmol), stirred
for 5 h at 658, cooled to r.t., diluted with Et2O (50 ml), and washed with sat. NH4Cl soln. The combined org. lay-
ers were washed with brine, dried (Na2SO4), filtered, and evaporated. FC (hexane/AcOEt 1:0 ! 3 :1 ! 2 :1 !
1:1) gave 25 (130 mg, 57%) and 23 (27 mg, 14%).

H
ELVETICA
CHIMICA
ACTA – Vol. 88 (2005)
2601
Data of 25: Pale yellow solid. Rf (hexane/AcOEt 2 :3) 0.71. M.p. 98.58. [a]2D5 =+39.4 (c=1.04, CHCl3). UV
(CHCl3): 329 (3.3), 269 (3.8), 240 (3.9). IR (CHCl3): 3019s, 2976w, 2895w, 1884m, 1728w, 1618w, 1580w, 1529s,
1477m, 1423m, 1391m, 1351m, 1335m, 1046s, 928s, 877m, 849m. 1H-NMR (CDCl3, 300 MHz; assignment
based on a HSQC spectrum): see Table 4 ; additionally, 4.43 (d, J=12.1, PhCH); 4.49 (d, J=11.8, PhCH);
4.50 (d, J=11.2, PhCH); 4.68 (d, J=11.5, PhCH); 4.82 (d, J=11.5, PhCH); 4.83 (d, J=11.2, PhCH); 4.86 (d,
J=11.8, PhCH); 5.08 (d, J=11.7, CHꢀC(2)); 5.13 (d, J=11.5, CH’ꢀC(2)); 5.17 (d, J=11.5, PhCH); 7.17–
7.21 (m, 2 arom. H); 7.26–7.43 (m, 18 arom. H, HꢀC(6) of PhNO2, HꢀC(5) of PhNO2); 7.82 (dt, J=7.8, 1.8,
HꢀC(4) of PhNO2); 7.95 (t, J=2.1, HꢀC(2) of PhNO2). 13C-NMR (CDCl3, 75 MHz; assignment based on a
HSQC spectrum): see Table 3 ; additionally, 65.14 (t, CH2ꢀC(2)); 72.69, 73.29 (2t, 2 PhCH2); 74.15 (t, 2
PhCH2); 109.48 (d, C(2) of PhNO2); 115.69 (d, C(4) of PhNO2); 121.95 (d, C(6) of PhNO2); 127.54–128.46 (sev-
eral d); 129.73 (d, C(5) of PhNO2); 136.95, 137.11, 137.38, 137.63, 138.01 (5s); 149.02 (s, C(3) of PhNO2); 159.21
(s, C(1) of PhNO2). HR-MALDI-MS: 712.3008 (9, [M+H]+, C43H42N3O7þ ; calc. 712.3023), 605.2461 (40),
604.2432 (100, [MꢀBnO]+, C36H34N3O6þ ; calc. 604.2448), 573.2727 (5, [MꢀC6H4NO3]+, C37H37N2Oþ4 ; calc.
573.2753), 465.2158 (20, [Mꢀ C6H4NO3 ꢀBnOH]+, C30H29N2O3þ ; calc. 465.2178), 375.1695 (21, C23H23N2Oþ3 ,
[MꢀC6H4NO3 ꢀBnOꢀBn]+; calc. 375.1709), 359.1745 (20, [MꢀC6H4NO3ꢀ2 BnO]+, C23H23N2Oþ2 ; calc.
359.1760). Anal. calc. for C43H41N3O7 (711.81): C 72.56, H 5.81, N 5.90; found: C 72.61, H 5.54, N 6.04.
(5R,6R,7S,8S)-6,7,8-Tris(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-2-(phenoxymethyl)imidazo-
[1,2-a]pyridine (26). A soln. of 24 (49 mg, 0.081 mmol) in DMF (2 ml) was treated with t-BuOK (13 mg, 0.12
mmol) and PhOH (11.0 mg, 0.12 mmol), stirred for 3.5 h at 808, cooled to r.t., diluted with H2O, and extracted
with Et2O. The org. layer was washed with 0.5M NaOH and brine, dried (Na2SO4), and evaporated. FC (hexane/
AcOEt 1 :1) yielded 26 (41 mg, 77%). Colourless oil. Rf (hexane/AcOEt 3 :2) 0.52. [a]2D5 =+37.2 (c=0.865,
CHCl3). UV (CHCl3): 271 (3.3), 240 (3.5). 1H-NMR (CDCl3, 300 MHz, assignment based on a HSQC spectrum):
see Table 4 ; additionally, 4.43 (d, J=12.1, PhCH); 4.48 (d, J=11.8, PhCH); 4.51 (d, J=11.8, PhCH); 4.69 (d,
J=11.5, PhCH); 4.82 (d, J=11.8, PhCH); 4.84 (d, J=12.1, PhCH); 4.86 (d, J=11.8, PhCH); 5.04 (d, J=11.8,
CHꢀC(2)); 5.09 (d, J=11.8, CH’ꢀC(2)); 5.19 (d, J=11.8, PhCH); 6.96 (tt, J=7.2, 0.9, HꢀC(4) of PhO); 7.05
(dt, J=1.2, 8.9, HꢀC(3) and HꢀC(5) of PhO); 7.17–7.24 (m, 2 arom. H); 7.25–7.36 (m, 18 arom. H); 7.43
(dd, J=1.5, 7.8, HꢀC(2) and HꢀC(6) of PhO). 13C-NMR (CDCl3, 75 MHz, assignment based on a HSQC spec-
trum): see Table 3; additionally, 64.63 (t, CH2ꢀC(2)); 72.61, 73.24, 73.94, 74.15 (4t, 4 PhCH2); 114.79 (d, C(2) and
C(6) of PhO); 120.59 (d, C(4) of PhO); 127.73–129.23 (several d); 137.11, 137.43, 137.67, 138.09, 138.20 (5s);
158.69 (s, C(1) of PhO). HR-MALDI-MS: 689.2943 (34, [M+Na]+, C43H42N2NaO5þ ; calc. 689.2991), 667.3159
(100, [M+H]+, C43H43N2O5þ ; calc. 667.3172), 573.2732 (18, [MꢀPhO]+, C37H37N2Oþ4 ; calc. 573.2753),
559.2589 (27, [MꢀBnO]+, C36H35N2O4þ ; calc. 559.2597), 375.1701 (24). Anal. calc. for C43H42N2O5 (666.82): C
77.45, H 6.35, N 4.20; found: C 77.51, H 6.46, N 4.07.
(5R,6R,7S,8S)-5,6,7,8-Tetrahydro-5-(hydroxymethyl)-2-[(3-nitrophenoxy)methyl]imidazo[1,2-a]pyridine-6,
7,8-triol (11). A soln. of 25 (21.8 mg, 28.1 mmol) in CH2Cl2 (1.5 ml) was treated with AlCl3 (49 mg, 0.33 mmol)
and anisole (53 ml, 0.49 mmol), stirred for 10 h at 238, and treated with H2O (20 ml) and AcOEt (15 ml). After
separation of the layers, the org. layer was extracted with H2O. The combined aq. layers were evaporated. FC
(AcOEt/MeOH/H2O 1:0 :0 ! 20 :1 :1 ! 15 :1 :1) yielded 11 (6.3 mg, 58%). White solid. Rf (AcOEt/MeOH/
H2O 10 :1 :1) 0.25. [a]2D5 =ꢀ23.7 (c=0.435, MeOH). UV (MeOH): 325 (3.2), 268 (3.7), 213 (4.3). IR (KBr):
3675–2800s (br.), 2925m, 2853m, 1618m, 1580m, 1528s, 1481m, 1460m, 1350s, 1285m, 1250m, 1177m, 1103m,
1025m, 1008m, 871m, 829m, 801m, 738s. 1H-NMR (CD3OD, 300 MHz): see Table 2 ; additionally, 4.49 (d,
J=8.1, irrad. at 3.68 ! s, HꢀC(8)); 5.08 (br. s, CH2ꢀC(2)); 7.37–7.41 (m, HꢀC(6) of PhNO2); 7.47–7.53 (m,
HꢀC(5) of PhNO2); 7.81 (m, HꢀC(2) and HꢀC(4) of PhNO2). 13C-NMR (CD3OD, 75 MHz): see Table 3; addi-
tionally, 65.08 (t, CH2ꢀC(2)); 109.97 (d, C(2) of PhNO2); 116.13 (d, C(4) of PhNO2); 122.24 (d, C(6) of PhNO2);
130.87 (d, C(5) of PhNO2); 136.87 (s, C(2)); 149.42 (s, C(3) of PhNO2); 159.55 (s, C(1) of PhNO2). HR-MALDI-
MS: 374.0956 (15, [M+Na]+, C15H17N3NaO7þ ; calc. 374.0964), 358.1011 (27, [M+NaꢀO]+, C15H17N3NaO6
;
+
calc. 358.1015), 352.1139 (28, [M+H]+, C15H18N3O7þ ; calc. 352.1145), 336.1190 (35), 320.1240 (31), 213.0871
(100, [MꢀC6N4NO3]+, C9H13N2O4þ ; calc. 213.0875), 195.0763 (39, [MꢀC6N4NO3 ꢀH2O]+, C9H11N2Oþ3 ; calc.
195.077). Anal. calc. for C15H17N3O7 (351.32): C 51.30, H 5.05, N 11.69; found: C 51.28, H 4.88, N 11.96.
(5R,6R,7S,8S)-5,6,7,8-Tetrahydro-5-(hydroxymethyl)-2-(phenoxymethyl)imidazo[1,2-a]pyridine-6,7,8-triol
(12). A soln. of 26 (40 mg, 60.1 mmol) in AcOEt/MeOH/H2O 3 :1:1 (2 ml) was treated with 20% Pd/C (20 mg),
hydrogenated at amb. pressure for 46 h, and filtered through Celite (washing with MeOH). Evaporation of the
combined filtrates, FC (AcOEt/MeOH/H2O 1 :0 :0 ! 20 :1 :1 ! 15 :1 :1), and drying afforded 12 (11.5 mg,
63%). White hygroscopic solid. Rf (AcOEt/MeOH/H2O 10 :1 :1) 0.10. [a]2D5 =+29.2 (c=0.605, MeOH). UV
(MeOH): 277 (3.1), 271 (3.2), 220 (4.1). IR (KBr): 3600–3000s (br.), 2926m, 1598m, 1586m, 1458m, 1383w,
1299w, 1240s, 1175m, 1103m, 1028s, 874w, 754m, 691m. 1H-NMR (CD3OD, 300 MHz): see Table 2; additionally,

2602
HELVETICA CHIMICA ACTA – Vol. 88 (2005)
4.95 (s, CH2ꢀC(2)); 6.90 (tt, J=0.9, 7.3, HꢀC(4) of PhO); 6.94–6.97 (m, HꢀC(2) and HꢀC(6) of PhO); 7.21–
7.27 (m, HꢀC(3) and HꢀC(5) of PhO). 13C-NMR (CD3OD, 75 MHz): see Table 3 ; additionally, 63.51 (t,
CH2ꢀC(2)); 114.56 (d, C(2) and C(6) of PhO); 120.65 (d, C(4) of PhO); 129.25 (d, C(3) and C(5) of PhO);
137.87 (s, C(2)); 158.96 (s, C(1) of PhO). HR-MALDI-MS: 329.1109 (35, [M+Na]+, C15H18N2NaO5þ ; calc.
329.1113), 307.1290 (88, [M+H]+, C15H19N2O5þ ; calc. 307.1294), 213.0870 (100, [MꢀPhO]+, C9H13N2Oþ4 ; calc.
213.0875). Anal. calc. for C15H18N2O5 ·0.5 H2O (315.32): C 57.14, H 6.07, N 8.88; found: C 56.90, H 6.43, N 8.73.
Inhibition Studies. See [23]. The b-glucosidase from Caldocellum saccharolyticum (EC 3.2.1.21, as lyophi-
lised powder, Sigma G-6906), b-glucosidase from almonds (as lyophilised powder, Fluka 49290), a-glucosidase
from yeast (as lyophilised powder, Fluka 63412). 4-nitrophenyl b-D-glucopyranoside (Fluka 73676), and 4-nitro-
phenyl a-D-glucopyranoside (Fluka 73673) were used without further purification. The inhibition studies were
carried out for b-glucosidases from C. saccharolyticum and sweet almonds, and a-glucosidase from brewer’s
yeast. The results of the inhibition studies are summarised in Table 1.
REFERENCES
[1] S. G. Withers, Carbohydr. Polym. 2001, 44, 325.
[2] C. S. Rye, S. G. Withers, Curr. Opin. Chem. Biol. 2000, 4, 573.
[3] D. L. Zechel, S. G. Withers, Acc. Chem. Res. 2000, 33, 11.
[4] M. Terinek, A. Vasella, Tetrahedron: Asymmetry 2005, 16, 449.
[5] M. Terinek, A. Vasella, Helv. Chim. Acta 2004, 87, 3035.
[6] N. Panday, M. Meyyappan, A. Vasella, Helv. Chim. Acta 2000, 83, 513.
[7] T. D. Heightman, P. Ermert, D. Klein, A. Vasella, Helv. Chim. Acta 1995, 78, 514.
[8] P. Ermert, A. Vasella, M. Weber, K. Rupitz, S. G. Withers, Carbohydr. Res. 1993, 250, 113.
[9] P. Ermert, A. Vasella, Helv. Chim. Acta 1991, 74, 2043.
[10] T. D. Heightman, A. T. Vasella, Angew. Chem., Int. Ed. 1999, 38, 750.
[11] A. Vasella, G. J. Davies, M. Böhm, Curr. Opin. Chem. Biol. 2002, 6, 619.
[12] K. S. E. Tanaka, A. J. Bennet, Can. J. Chem. 1998, 76, 431.
[13] E. Lorthiois, M. Meyyappan, A. Vasella, Chem. Commun. 2000, 1829.
[14] M. Böhm, E. Lorthiois, M. Meyyappan, A. Vasella, Helv. Chim. Acta 2003, 86, 3787.
[15] N. Panday, Y. Canac, A. Vasella, Helv. Chim. Acta 2000, 83, 58.
[16] M. Terinek, A. Vasella, Helv. Chim. Acta 2005, 88, 10.
[17] M. Terinek, A. Vasella, Helv. Chim. Acta 2003, 86, 3482.
[18] K. Tatsuta, S. Miura, S. Ohta, H. Gunji, J. Antibiot. 1995, 48, 286.
[19] T. Granier, N. Panday, A. Vasella, Helv. Chim. Acta 1997, 80, 979.
[20] R. Hoos, A. B. Naughton, A. Vasella, Helv. Chim. Acta 1992, 75, 1802.
[21] Y. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 16, 4467.
[22] I. H. Segel, ‘Enzyme Kinetics’, John Wiley & Sons, New York, 1975.
[23] C. Blüchel, C. V. Ramana, A. Vasella, Helv. Chim. Acta 2003, 86, 2998.
Received July 8, 2005