Synthesis of N-acetylglucosamine-derived nagstatin analogues and their evaluation as glycosidase inhibitors

Article abstract of DOI:10.1002/hlca.200490286

The gluco-configured analogue 15 of nagstatin (1) and the methyl ester 14 were synthesized via condensation of the thionolactams 17 or 18 with the β-amino ester 19. The silyl ethers 20 and 21 resulting from 17 were desilylated to 22 and 23; these alcohols

Full text of DOI:10.1002/hlca.200490286

10
Helvetica Chimica Acta ± Vol. 88 (2005)
Synthesis of N-Acetylglucosamine-Derived Nagstatin Analogues and Their
Evaluation as Glycosidase Inhibitors
by Miroslav Terinek and Andrea Vasella*
Laboratorium für Organische Chemie, ETH-Hönggerberg, Wolfgang-Pauli-Strasse 10, CH-8093 Zürich
The gluco-configured analogue 15 of nagstatin (1) and the methyl ester 14 were synthesized via
condensation of the thionolactams 17 or 18 with the b-amino ester 19. The silyl ethers 20 and 21 resulting from 17
were desilylated to 22 and 23; these alcohols were directly obtained by condensing 18 and 19. The attempted
substitution of the C(8)ÀOH group of 22 by azide under Mitsunobu conditions led unexpectedly to the
deoxygenated a-azido esters 24. The desired azide 25 was obtained by treating the manno-configured alcohol 23
with diphenyl phosphorazidate. The azide was transformed to the debenzylated acetamido ester 14 that was
hydrolyzed to the nagstatin analogue 15. The imidazole-2-acetates 14 and 15 are nanomolar inhibitors of the N-
acetyl-b-glucosaminidases from Jack beans and from bovine kidney, submicromolar to micromolar inhibitors of
the b-glucosidase from Caldocellum saccharolyticum, and rather weak inhibitors of the snail b-mannosidase. In
all cases, the ester was a stronger inhibitor than the corresponding acid. As expected from their gluco-
configuration, both imidazopyridines 14 and 15 are stronger inhibitors of the b-N-acetylglucosaminidase from
bovine kidney than nagstatin.
Introduction. ± Nagstatin (1), a strong inhibitor of several hexosaminidases [1± 4],
is a N-acetylgalactosamine-derived tetrahydropyridoimidazole-2-acetic acid [5]. Its
inhibitory activity is essentially associated with the imidazole ring and not with the
carboxymethyl substituent [2], although substituents on the imidazole ring may
strongly affect the inhibition of b (and a-)-glycosidases [6 ± 8]. In the preceding paper
[9], we described the influence of the hydrophobic character of C(2)-methyl ester and
carboxylic acid substituents on the inhibition of b-glycosidases. Imidazole-2-propio-
nates 10 ± 13 are stronger inhibitors of the b-glucosidase from Caldocellum saccha-
rolyticum and of the b-mannosidase from snail than imidazole-2-acetates 6 ± 9, and
these are stronger than imidazole-2-carboxylates 2 ± 5. There is a parallel sequence of
inhibitory activity for the methyl esters and the corresponding acids, with the esters
being stronger inhibitors.
ꢀ 2005 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta ± Vol. 88 (2005)
11
We wanted to know if the relationship between the inhibitory activity of esters and
acids is also valid for the inhibition of the N-acetyl-b-glucosaminidase from bovine
kidney1) by the N-acetylglucosamine-derived imidazole-2-acetate 14 and the corre-
sponding acid 15. These compounds will also allow testing the effect of the galacto- vs.
gluco-configuration on the inhibition.
We report the synthesis of 14 and 15, their evaluation as inhibitors of the
hexosaminidases from bovine kidney (unknown family) and from Jack beans (family 18
[11][12]), and a comparison of the inhibition to that of the unsubstituted GlcNAc-
imidazole 16 [2][13].
Synthesis. ± We planned to synthesize the gluco-analogue 15 of nagstatin (1) via the
gluco- and manno-configured hydroxy-imidazoles 22 and 23. These imidazoles should
be obtained by condensing the selectively protected thionolactams 17 [14] or 18 [15]
with the b-amino ester 19 [9], followed by selective deprotection of the products. The
resulting alcohols 22 and 23 should be transformed into 15 by a Mitsunobu substitution,
similarly as described for the galacto- and talo-analogues in the synthesis of nagstatin
[3][4] (Scheme 1).
Condensation of the O-silylated thionolactam 172) with the b-amino ester 19 [9]
followed by treatment of the crude with TsOH ´ H2O provided the gluco- and manno-
configured imidazoles 20 and 21 (53%; 55 :45), which were readily separated by
chromatography to give 29% of 20 and 23% of 21. Standard desilylation [16][17] of 20
and 21 led to the hydroxy-imidazoles 22 (93%) and 23 (93%). The alcohols 22 and 23
were obtained in higher yields by condensing 19 with the acetoxy-thionolactam 18 [15]
followed by acid treatment of the crude. This yielded 81% of a 45 :55 mixture 22/23;
chromatography provided 32% of 22, 33% of 23 and 16% of a 28 :72 mixture 22/23.
Substitution of the OH by an N3 group under Mitsunobu conditions [18 ± 20] using
PBu3 was reported to proceed well for the galacto-analogue of the imidazopyridine 22,
but the same conditions did not affect 22. Replacing Bu3P by Ph3P, however, led to a
mixture of the diastereoisomeric a-azido esters 24 (47%; (2R*)/(2S*) 2 :1); 32% of 22
were recovered. The unexpected and non-trivial formation of 24 is rationalized
(Scheme 2) by assuming that 22 is transformed via a betaine formed from Ph3P and
azodicarboxylate [21] into the phosphonium cation A. Elimination of Ph3PO from A
generates the diazafulvenium cation B. A 1,5-H shift transforms B into the (E)- and/or
(Z)-diazafulvenium cations C and is followed by addition of azide leading to the a-
azido esters 24.
The gluco-azide 25 was finally obtained in a yield of 69% by treating the manno-
configured hydroxy-imidazole 23 with diphenyl phosphorazidate (DPPA) and DBU
[22]. Under otherwise identical conditions, the gluco-imidazole 22 decomposed to a
complex mixture from which the gluco- and manno-configured azides 25 and 26 were
isolated in 6 and 3% yield, respectively. In contradistinction, the substitution of
1
)
)
The inhibition of the N-acetyl-b-glucosaminidase from bovine kidney and the b-glucosidase from C.
saccharolyticum was correlated with the basicity of GlcNAc- and glucose-derived inhibitors of the
azolopyridine type [10].
The thionolactam 17 will be described in a forthcoming paper on the synthesis of the manno-configured 8-
amino-8-deoxy-tetrahydroimidazopyridine [14].
2

12
Helvetica Chimica Acta ± Vol. 88 (2005)
Scheme 1
a) 1. HgCl2, THF; 2. TsOH ´ H2O, toluene, 608; 53% of 20/21 55 :45 (from 17). b) As a; 81 % of 22/23 45 :55
(from 18). c) Bu4NF, THF, 08; 93% of 22; 93% of 23. d) HN3, Ph3P, diethyl azodicarboxylate (DEAD), THF,
0 ! 238; 47% of 24 (2R*)/(2S*) 2 :1. e) Diphenyl phosporazidate (DPPA), 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU), toluene; 9% of 25/26 2 :1(from 22); 69% of 25 (from 23). f) 1 . H, 10% Pd/C, AcOH; 2. Ac2O,
2
MeOH; 3. NH3, MeOH, 408; 60%. g) HS(CH2)3SH, Et3N, MeOH; 85%. h) Ac2O, pyridine; 81%. i) H2, 1 0%
Pd/C, AcOH; 98%. j) KOH, EtOH/H2O, 08; 76%.
C(8)ÀOH of the galacto-analogue of 22 (HN3, Bu3P, DEAD) provided 77% of the
galacto-azide [3][4]. The inversion of configuration observed for the transformation of
23 to 25 marks only a formal contrast to the retention of configuration observed by
Tatsuta et al. for the substitution of the galacto-configured analogue of 23 [4] (cf.
[15][23]), considering the preferred pseudoaxial addition of azide to an intermediate
azoniafulvenium cation of type B (Scheme 2) presumably formed in both cases.
Hydrogenation of the azido-imidazole 25 (10% Pd/C, AcOH), followed by
acetylation (Ac2O, MeOH) and O-deacetylation (NH3, MeOH), led to the debenzy-
lated methyl ester 14 in 60% yield. In a longer but higher-yielding route to 14, we
reduced the azide 25 with propane-1,3-dithiol and Et3N [24] and acetylated the
resulting amino-imidazole 27 (85%) to 28 (81%). Hydrogenolysis of 28 yielded 98% of

Helvetica Chimica Acta ± Vol. 88 (2005)
13
Scheme 2
the desired methyl ester 14 (67% over three steps). Finally, hydrolysis of 14 at 08
provided 76% of the nagstatin analogue 15 that was thus prepared from 18 in five steps
and an overall yield of 14%, or, via the protected amino- and acetamido-imidazoles 27
and 28, in seven steps and an overall yield of 16%. The methyl ester 14 was prepared in
19 and 21% yield over four and six steps, respectively.
Similarly to the other C(2)-substituted imidazoles [6 ± 8][25], the protected gluco-
and manno-tetrahydroimidazopyridines 21 ± 23 and 25 ± 27 exist in solution as 2 :1
mixtures of the 7H6 and the 6H7 conformers3), while the unprotected gluco-
7
imidazopyridines 14 and 15 adopt a H6 conformation (see Tables 3 and 5 in Exper.
Part). The gluco-configured imidazole 20 adopts in CDCl3 a 5,8B conformation with the
C(8)ÀO bond in the pseudo-axial position, as evidenced by J(5,6), J(6,7), and J(7,8)
6
values of 8.4, 3.4, and 3.4 Hz, respectively. The coupling constants indicate a H7
conformation for the gluco-acetamide 28 which is stabilized by an intramolecular H-
bond between the NÀH and OÀC(6)4). The 13C signals of C(5)ÀC(8) of all imidazoles
(Table 4 in Exper. Part) were assigned by analogy to the gluco- and manno-analogues
described in the preceding paper [9]. The formation of the methyl imidazole-2-acetates
20 and 21 is confirmed by the disappearance of the NH signal of 17, 13C singlets at 133.43
(20) and 134.32 ppm (21), 13C doublets at 115.09 (20) and 116.70 ppm (21), and strong
CˆO bands at 1736 cmÀ1. The configuration of the alcohols 22 and 23 (and thus also of
20 and 21) was assigned on the basis of J(7,8) values (7.5 and 3.4 Hz, resp.). The
structure the a-azido esters 24 was deduced on the basis of the replacement of a broad s
at 3.67 ppm for CH2ÀC(2) by two broad ss at 4.87 and 4.88 ppm in the ratio 2 :1, and of
a d at 4.87 ppm for HÀC(8) for 22 by two dds at 3.00 and 3.25 ppm with Jgem ˆ 16.8 Hz
for CH2(8). The structure of 24 is supported by a strong IR band at 2109 cmÀ1 for the
‡
N3 group and by the [M ‡ H] peak at m/z 568.2550.
3
)
)
The direction of numbering of imidazopyridines (cf. 20 in Scheme 1) is opposite to that of pyranosides.
Thus, the sides above and below the plane of the imidazoles, as defined by the clockwise and
counterclockwise numbering, are interchanged relative to those defined by carbohydrate nomenclature.
A similar stabilization of the 6H7 conformation by an intramolecular H-bond was observed for the
corresponding gluco-configured tetrahydrotetrazolopyridine [26] and tetrahydroindolizines [23].
4

14
Helvetica Chimica Acta ± Vol. 88 (2005)
Enzymatic Tests and Discussion. ± The nagstatin analogue 15 and the corresponding
methyl ester 14 were tested as inhibitors of the N-acetyl-b-glucosaminidases from Jack
beans (citrate buffer, pH 5.0, 258) and from bovine kidney (citrate buffer, pH 4.1, 378),
the b-glucosidase from C. saccharolyticum (phosphate buffer, pH 6.8, 558), and the b-
mannosidase from snail (acetate buffer, pH 4.5, 258), using the corresponding 4-
nitrophenyl glycopyranosides as substrates. The inhibition data of 14 and 15 for the
hexosaminidases are summarized in Table 1 and compared to the inhibition by the
parent acetamido-imidazole 16 and by nagstatin [2]. A comparison of the inhibition of
the b-glucosidase and b-mannosidase by 14 and 15 to the inhibition by the glucose- and
mannose-related analogues 6 ± 9 [9] is shown in Table 2.
Table 1. Data (IC50 and Ki Values in [nM]) for the Inhibition of the N-Acetyl-b-glucosaminidase from Jack Beans
and from Bovine Kidney by the Acetamido-imidazoles 1 and 14 ± 16
Inhibitor
pKha
N-Acetyl-b-glucosaminidase
from Jack beansa)
N-Acetyl-b-glucosaminidase
from bovine kidneyb)
IC50
Ki
IC50
Ki
1c)
16c)
14
±
±
±
±
2.0
7.8
±
±
13.4
7.0
4.5
±
±
4.62
1.5 (a ˆ 1.5)
4.0 (a ˆ 1.1)
15
5.32d)
5.9 (a ˆ 1.5)
8.1
5.2 (a ˆ 3.0)
a) At 258 and pH 5.0. b) At 378 and pH 4.1. c) Data taken from [2]. d) No additional pKha was observed between
pH 2.3 ± 5.7.
Table 2. Comparison of the Inhibition (Ki Values in [nM]) of the b-Glucosidase from C. saccharolyticum and b-
Mannosidase from Snail by the Acetamido-imidazoles 14 and 15, and by Their Glucose- and Mannose-Derived
Analogues 6 ± 9
Inhibitor
pKha
b-Glucosidase from
C. saccharolyticuma)
b-Mannosidase from snailb)
14
15
4.62
5.32c)
5.03
6.40
4.61
5.08
167 (a ˆ 1.3)
20 ´ 103 (competitive)
110 ´ 103 (a ˆ 1.6)
145 ´ 103 (competitive)
6d)
7d)
8d)
9d)
5.0 (non-competitive)
±
±
95 (a ˆ 1.6)
±
±
117 (a ˆ 3.4)
810 (competitive)
a) At 558 and pH 6.8. b) At 258 and pH 4.5. c) No additional pKha was observed between pH 2.3 ± 5.7. d) Data
taken from [9].
The acetamido-imidazoles 14 and 15 are nanomolar, albeit essentially non-
competitive inhibitors of the N-acetyl-b-glucosaminidases (a-values between 1.1 and
3.0 [27]; Table 1). As expected from the gluco-configuration of 14 and 15, these
acetates are 2 ± 3 times stronger inhibitors of the N-acetyl-b-glucosaminidase from
bovine kidney than the galacto-configured nagstatin. The ester 14 is a stronger inhibitor
of this enzyme than the C(2)-unsubstituted acetamide 16, which is slightly but not
significantly stronger than the acid 15. That both glucosaminidases are more strongly

Helvetica Chimica Acta ± Vol. 88 (2005)
15
inhibited by the ester 14 is in agreement with the inhibition of other b-glycosidases by
the gluco- and manno-analogues of 14 and 15 [9].
The methyl ester 14 is also a (mixed-type) inhibitor (Ki ˆ 167 nm) of the b-
glucosidase from C. saccharolyticum. Although 14 is 30 ± 35 times weaker than the
corresponding C(8)ÀOH analogue 6, the submicromolar inhibition constant for 14
may imply that there is still a binding interaction of the enzyme with the acetamido
group. The acid 15 (essentially dissociated at the pH of the assay) is a weaker inhibitor
by about one order of magnitude than the ester 14, evidencing a disrupting influence of
the negatively charged carboxylate group. As expected, the imidazoles 14 and 15, which
bear an equatorial acetamido group, are only weak, albeit competitive, inhibitors of the
b-mannosidase from snail.
We thank M. Schneider for the pKha determinations, Dr. B. Bernet for checking the experimental part, and
the Swiss National Science Foundation and Oxford Glycosciences Ltd., Abingdon (UK), for generous support.
Experimental Part
General. Solvents were distilled: THF and toluene from Na and benzophenone, MeOH, Et3N, and DBU
from CaH2. Reactions were carried out under Ar, unless stated otherwise. Qual. TLC: precoated silica-gel plates
(Merck silica gel 60 F254); detection by heating with ꢁmostainꢂ (400 ml of 10% H2SO4 soln., 20 g of
(NH4)6Mo7O24 ´ 6 H2O, 0.4 g of Ce(SO4)2) or ninhydrin. Flash chromatography (FC): silica gel Fluka 60 (0.04 ±
0.063 mm). M.p.ꢂs uncorrected. Optical rotations: 1-dm cell at 258, 589 nm. UV spectra (ca. 0.2 mm solns.) were
taken in 1-cm cell at 258 in the range of 190 to 500 nm (log e values in parenthesis). FT-IR spectra: KBr or ca.
2% soln. in CHCl3, absorption in cmÀ1 1H- and 13C-NMR spectra: chemical shifts d in ppm rel. to TMS as
.
external standard, and coupling constants J in Hz. HR-MALDI-MS: in gentisic acid (ˆ2,5-dihydroxybenzoic
acid, DHB) matrix. The pKHA values were determined in H2O by potentiometric titration with HCl at 258. The
b-N-acetylglucosaminidase from Jack beans (EC 3.2.1.52, as a suspension in 2.5m (NH4)2SO4, pH 7.0, Sigma A-
2264), b-N-acetylglucosaminidase from bovine kidney (EC 3.2.1.52, as a suspension in 3.2m (NH4)2SO4, pH ca.
6.0, Sigma A-2415), b-glucosidase from C. saccharolyticum (EC 3.2.1.21, as a lyophilized powder, Sigma G-
6906), b-mannosidase from snail acetone powder (EC 3.2.1.25, as a suspension in 3.0m (NH4)2SO4 containing
1 0 mm AcONa, pH ca. 4.0, Sigma M-9400), 4-nitrophenyl N-acetyl-b-d-glucosaminide (Sigma N-9376), 4-
nitrophenyl b-d-glucopyranoside (Fluka 73676), and 4-nitrophenyl b-d-mannopyranoside (Sigma N-1268) were
used without further purification.
Preparation of 20 and 21. A suspension of 17 [14] (90 mg, 0.145 mmol) and HgCl2 (56 mg, 0.206 mmol) in
THF (1.7 ml) was treated with 19 [9] (162 mg, 0.914 mmol), stirred for 12 h at 228, and filtered over Celite. The
combined filtrate and washing (50 ml of AcOEt) were washed with brine (30 ml), dried (MgSO4), filtered, and
evaporated. A soln. of the residue (180 mg) in toluene (3.5 ml) was treated with TsOH ´ H2O (129 mg,
0.678 mmol), stirred for 40 h at 608, cooled to 228, diluted with AcOEt (40 ml), and washed with sat. NaHCO3
soln. (3 Â 20 ml). The combined aq. layers were extracted with AcOEt (2 Â 20 ml). The combined org. layers
were washed with H2O (40 ml) and brine (40 ml), dried (MgSO4), filtered, and evaporated. FC (hexane/AcOEt
5 :1 ! 3 :1) gave 20 (29 mg, 29%) and 21 (24 mg, 24%). Repetition of this experiment with 350 mg of 17
(0.565 mmol) gave 20 (97 mg, 25%), 20/21 55 :45 (6 mg, 2%) and 21 (99 mg, 25%).
Methyl (5R,6R,7S,8S)-6,7-Bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-8-(triisopropylsilyloxy)-
imidazo[1,2-a]pyridine-2-acetate (20). Colourless oil. Rf (hexane/AcOEt 1:1) 0.65. [ a]2D5 ˆ ‡29.3 (c ˆ 1.00,
CHCl3). UV (CHCl3): 285 (2.38), 264 (2.77), 259 (2.82), 240 (3.33). IR (CHCl3): 3090w, 3067w, 3032w, 3012m,
2946s, 2867s, 1952w, 1877w, 1808w, 1736s, 1604w, 1528w, 1497w, 1455m, 1438m, 1363m, 1263m, 1223m, 1096s,
1018m, 91 3w, 883m. 1H-NMR (CDCl3, 300 MHz): see Table 3; additionally, 0.97, 1.02 (2d, J ˆ 7.2, (Me2CH)3Si);
1.08 ± 1.22 (m, (Me2CH)3Si); 3.57 (d, J ˆ 16.5, CHÀC(2)); 3.65 (d, J ˆ 16.5, CH'ÀC(2)); 3.68 (s, MeO); 3.73
(irrad. at 4.07 ! d, J ˆ 8.1, irrad. at 4.31! d, J ˆ 3.1); 3.79 (irrad. at 4.31 ! d, J ˆ 10.9); 3.89 (irrad. at 3.79 !
change, irrad. at 4.31 ! d, J ˆ 10.3); 4.07 (irrad. at 3.73 ! d, J ˆ 3.1, irrad. at 5.05! d, J ˆ 3.4); 4.31(irrad. at
3.89 ! dd, J ˆ 4.4, 8.4); 4.36 (d, J ˆ 11.5, PhCH); 4.50 (br. s, PhCH2); 4.55 (d, J ˆ 11.5, PhCH); 4.58 (d, J ˆ 11.2,
PhCH); 4.71( d, J ˆ 11.5, PhCH); 5.05 (irrad. at 4.07 ! s); 7.15 ± 7.19 (m, 2 arom. H); 7.26 ± 7.38 (m, 13 arom. H).
13C-NMR (CDCl3, 75 MHz): see Table 4; additionally, 12.43 (d, (Me2CH)3Si); 17.99, 18.18 (2q, (Me2CH)3Si);

16
Helvetica Chimica Acta ± Vol. 88 (2005)
Table 3. Selected 1H-NMRChemical Shifts [ppm] and Coupling Constants [Hz] of the Protected and
Deprotected gluco-Imidazoles 14, 15, 20, 22, 25, 27, and 28
14
14
15
20
22
22
25
27
28
CD3OD
D2O
D2O
CDCl3 CDCl3 (D6)DMSO CDCl3 CDCl3 CDCl3
HÀC(3)
HÀC(5)
HÀC(6)
HÀC(7)
HÀC(8)
CHÀC(5)
7.22
7.20
7.23
6.97
6.98
4.12
3.93
3.98
4.87
3.70
3.79
7.5
8.4
7.5
5.0
3.4
7.07
4.15
3.82 ± 3.90
6.98
4.11
3.97
3.90
4.67
3.68
3.77
7.2
8.4
7.2
4.4
3.13.7
6.93
4.18
3.98
3.84
4.12
3.69
3.76
6.2
7.00
4.40 ± 4.47
3.84 ± 3.91 3.94 ± 4.06 4.02 ± 4.12 4.31
3.84 ± 3.913.94 ± 4.06 4.03 3.73
4.07
5.05
a
4.02
)
4.06b)
5.38d)
3.72
3.76
2.5
3.73
4.95
3.94
3.90
4.89
4.08
4.22
3.93
4.98
3.82 ± 3.90
4.59c)
3.71
3.90
7.5
4.02 ± 4.12 3.79
CH'ÀC(5) 4.16
4.23
9.3
9.3
9.0
3.89
8.4
3.4
3.4
5.6
2.8
e
e
J(5,6)
J(6,7)
J(7,8)
J(5,CH)
J(5,CH')
)
)
e
9.0
8.7
3.7
1
9.7
9.7
)
7.5
6.5
5.6
5.0
2.5
6.9
6.5
4.7
2.8
e
e
)
)
)
e
.9
2.2
12.8
5.6
10.0 10.0
J(CH,CH') 12.1
12.5
10.6
10.3
10.3
10.0
a) J(6,8) ꢀ 0.6 Hz. b) J(5,7) ꢀ 0.6 Hz. c) J(8,OH) ˆ 5.6 Hz. d) J(8,NH) ˆ 8.1Hz. e) Not assigned.
Table 4. Selected 13C-NMRChemical Shifts [ppm] of the Protected and Deprotected gluco-Imidazoles 14, 15, 20,
22, 25, 27, and 28 and of the Protected manno-Imidazoles 21 and 23
Solvent
C(2)
C(3)
C(5)
CH2ÀC(5)
C(6)
C(7)
C(8)
C(8a)
gluco
14
15
20
22
25
27
28
D2O
D2O
133.93
134.11
133.43
134.80
135.55
134.75
134.67
116.26
116.16
115.09
115.19
115.87
115.44
117.12
60.18
60.87
56.60
58.66
58.73a)
58.81
58.81
58.65
58.34
67.66
68.63
68.18
69.49
71.33
67.87
67.59
78.57
75.02
75.22
75.65
74.19a)
72.01
71.35
82.73
81.69
81.05
82.27
75.87a)
49.87
49.11
66.25
67.63
59.09a)
50.39
46.10
143.68
142.67
144.97
145.95
140.21
146.25
140.42
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
manno
21
23
CDCl3
CDCl3
134.32
134.75
116.70
116.28
59.74
59.03
70.82
70.36
73.62
73.25
81.63
78.77
64.26
62.92
144.76
144.00
a) Assignments may be interchanged.
34.73 (t, CH2ÀC(2)); 51.88 (q, MeO); 72.10, 72.41, 73.16 (3t, 3 PhCH2); 127.64 ± 128.37 (several d); 137.26,
‡
‡
137.51, 137.56 (3s); 171.60 (s, CˆO). HR-MALDI-MS: 721.3647 (29, [M ‡ Na] , C41H54N2NaO6Si ; calc.
‡
‡
‡
‡
721.3649), 699.3822 (24, [M ‡ H] , C41H55N2O6Si ; calc. 699.3829), 655.3203 (4, [M À i-Pr] , C38H47N2O6Si ;
‡
‡
calc. 655.3203), 525.2386 (100, [M À (i-Pr)3SiO] , C32H33N2O5 ; calc. 525.2389). Anal. calc. for C41H54N2O6Si
(698.97): C 70.45, H 7.79, N 4.01; found: C 70.31, H 7.80, N 4.08.
Methyl (5R,6R,7S,8R)-6,7-Bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-8-(triisopropylsilyloxy)-
imidazo[1,2-a]pyridine-2-acetate (21). Colourless oil. Rf (hexane/AcOEt 1:1) 0.52. [ a]2D5 ˆ À20.3 (c ˆ 0.69,
CHCl3). UV (CHCl3): 264 (2.83), 259 (2.86), 240 (3.27). IR (CHCl3): 3090w, 3067w, 3031m, 3013m, 2946s,
2893m, 2867s, 1952w, 1875w, 1810w, 1736s, 1604w, 1561w, 1497w, 1455s, 1438m, 1408w, 1363m, 1266m, 1222m,
1
1102s, 1017s, 952w, 91 5w, 883m. H-NMR (CDCl3, 300 MHz): see Table 5; additionally, 0.94, 1.03 (2d, J ꢀ 7.2,
(Me2CH)3Si); 1.08 ± 1.22 (m, (Me2CH)3Si); 3.55 (irrad. at 3.65 ! d, J ꢀ 5.0, irrad. at 4.05 ! change); 3.59 (br. s,
CH2ÀC(2)); 3.65 (irrad. at 4.05 ! d, J ꢀ 9.3); 3.68 (s, MeO); 3.76 (irrad. at 5.17 ! d, J ˆ 9.3); 4.05 (irrad. at
3.65 ! t, J ˆ 6.9); 4.30 (irrad. at 3.76 ! d, J ˆ 7.5, irrad. at 4.05 ! d, J ˆ 9.3); 4.43 (br. s, PhCH2); 4.61( d, J ˆ 11.5,
PhCH); 4.65 (d, J ˆ 11.8, PhCH); 4.82 (d, J ˆ 11.5, PhCH); 4.93 (d, J ˆ 11.2, PhCH); 5.17 (irrad. at 3.76 ! s);

Helvetica Chimica Acta ± Vol. 88 (2005)
17
7.22 ± 7.38 (m, 15 arom. H). 13C-NMR (CDCl3, 75 MHz): see Table 4; additionally, 12.50 (d, (Me2CH)3Si); 18.02,
18.22 (2q, (Me2CH)3Si); 34.56 (t, CH2ÀC(2)); 51.92 (q, MeO); 71.97, 73.06, 74.68 (3t, 3 PhCH2); 127.48 ± 128.32
‡
(several d); 137.46 (s); 137.88 (2s); 171.47 (s, CˆO). HR-MALDI-MS: 721.3641 (28, [M ‡ Na] ,
‡
‡
‡
C41H54N2NaO6Si ; calc. 721.3649), 699.3817 (82, [M ‡ H] , C41H55N2O6Si ; calc. 699.3829), 655.3196 (6,
‡
‡
‡
‡
[M À i-Pr] , C38H47N2O6Si ; calc. 655.3203), 525.2380 (100, [M À (i-Pr)3SiO] , C32H33N2O5 ; calc. 525.2389).
Anal. calc. for C41H54N2O6Si (698.97): C 70.45, H 7.79, N 4.01; found: C 70.32, H 7.79, N 4.15.
Table 5. Selected 1H-NMRChemical Shifts [ppm] and Coupling Constants [Hz] of the Protected manno-
Imidazoles 21, 23, and 26
21
23
23
26
CDCl3
CDCl3
(D6)DMSO
CDCl3
HÀC(3)
HÀC(5)
HÀC(6)
HÀC(7)
HÀC(8)
CHÀC(5)
CH'ÀC(5)
J(5,6)
6.99
4.05
4.30
3.76
5.17
3.55
3.65
7.5
7.04
4.12
4.19
3.95
5.07
3.70
3.77
6.2
7.07
4.10
4.17
3.845
4.86a)
3.67
3.847
6.8
7.04
4.06
4.12
3.99
4.875
3.648
3.74
7.2
J(6,7)
9.3
7.8
8.7
8.1
J(7,8)
2.2
3.4
3.4
3.7
J(5,CH)
J(5,CH')
J(CH,CH')
6.9
3.13.7
10.0
6.2
5.6
5.6
3.4
3.4
10.0
10.0
10.0
a) J(8,OH) ˆ 5.0 Hz
Preparation of 22 and 23. a) Treatment of 18 with the b-Amino Ester 19 and HgCl2. A suspension of 18 [15]
(300 mg, 0.593 mmol) and HgCl2 (225 mg, 0.829 mmol) in THF (6.9 ml) was treated with 19 [9] (635 mg,
3.58 mmol), stirred for 19 h at 238, and filtered over Celite. The combined filtrate and washing (80 ml of AcOEt)
were washed with brine (50 ml), dried (MgSO4), filtered, and evaporated. A soln. of the residue (660 mg) in
toluene (15 ml) was treated with TsOH ´ H2O (525 mg, 2.76 mmol), stirred for 40 h at 608, cooled to 228, diluted
with AcOEt (100 ml), and washed with sat. NaHCO3 soln. (3 Â 70 ml). The combined aq. layers were extracted
with AcOEt (2 Â 60 ml). The combined org. layers were washed with H2O (120 ml) and brine (120 ml), dried
(MgSO4), filtered, and evaporated. FC (hexane/AcOEt/MeOH 1:3 :0 ! 0 :1 :0! 0 :20 :1) gave 22 (102 mg,
32%), 22/23 28 :72 (53 mg, 16%) and 23 (107 mg, 33%). Repetition of this experiment with 980 mg of 18
(1.94 mmol) gave 22 (299 mg, 28%), 22/23 1:4 (236 mg, 22%), and 23 (299 mg, 28%).
b) Desilylation of 20. At 08, a soln. of 20 (68 mg, 97.3 mmol) in THF (1.3 ml) was treated with 1m Bu4NF in
THF (0.20 ml, 0.20 mmol), stirred at 08 for 2 h, and treated with sat. NH4Cl soln. (5 ml). The mixture was
diluted with Et2O (30 ml) and washed with sat. NH4Cl soln. (3 Â 15 ml). The combined aq. layers were extracted
with Et2O (2 Â 15 ml). The combined org. layers were washed with H2O (30 ml) and brine (30 ml), dried
(MgSO4), filtered, and evaporated. FC (AcOEt) gave 22 (49 mg, 93%).
c) Desilylation of 21. At 08, a soln. of 21 (80 mg, 0.115 mmol) in THF (1.5 ml) was treated with 1m Bu4NF in
THF (0.22 ml, 0.22 mmol), stirred at 08 for 2 h, and treated with sat. NH4Cl soln. (5 ml). Workup and FC (as
described in a) gave 23 (55 mg, 93%).
Methyl (5R,6R,7S,8S)-6,7-Bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-8-hydroxyimid-
azo[1,2-a]pyridine-2-acetate (22). Colourless oil. Rf (AcOEt) 0.15. Rf (AcOEt/MeOH 10 :1) 0.44. [a]2D5 ˆ À3.1
(c ˆ 0.74, CHCl3). UV (CHCl3): 258 (3.07), 240 (3.44). IR (CHCl3): 3305w, 3163w, 3090w, 3067w, 3031m, 3011m,
2954m, 2918m, 2869m, 1952w, 1877w, 1810w, 1736s, 1604w, 1565w, 1497m, 1454s, 1438m, 1408w, 1361m, 1308m,
1269m, 1144s, 1111s, 1027s, 91 2w. 1H-NMR (CDCl3, 300 MHz): see Table 3; additionally, 3.67 (br. s, CH2ÀC(2));
3.68 (s, MeO); 3.70 (irrad. at 4.12 ! d, J ˆ 10.2); 3.79 (irrad. at 4.12 ! d, J ˆ 10.2); 3.93 (irrad. at 4.12 ! d, J ˆ
8.2); 3.98 (irrad. at 4.87 ! d, J ˆ 8.2); 4.42 (d, J ˆ 12.1, PhCH); 4.46 (d, J ˆ 12.1, PhCH); 4.53 (d, J ˆ 11.2,
PhCH); 4.84 (d, J ˆ 11.5, PhCH); 4.87 (irrad. at 3.98 ! br. s); 4.91( d, J ˆ 11.2, PhCH); 5.07 (d, J ˆ 11.2, PhCH);

18
Helvetica Chimica Acta ± Vol. 88 (2005)
6.26 ± 6.36 (br. s, OH); 7.18 ± 7.21 (m, 2 arom. H); 7.24 ± 7.40 (m, 13 arom. H). 1H-NMR ((D6)DMSO, 300 MHz):
see Table 3; additionally, 3.52 (br. s, CH2ÀC(2)); 3.59 (s, MeO); 3.71(irrad. at 4.15 ! d, J ˆ 10.3); 3.82 ± 3.90
(irrad. at 4.15 ! change); 3.90 (irrad. at 4.15 ! d, J ˆ 10.6); 4.44 (d, J ˆ 12.1, PhCH); 4.45 (d, J ˆ 10.9, PhCH);
4.50 (d, J ˆ 12.1, PhCH); 4.59 (irrad. at 5.84 ! d, J ˆ 6.5); 4.73 (d, J ˆ 11.5, PhCH); 4.77 (d, J ˆ 11.5, PhCH); 4.88
(d, J ˆ 11.8, PhCH); 5.84 (d, J ˆ 5.6, exchange with CD3OD, OH); 7.15 ± 7.20 (m, 2 arom. H); 7.23 ± 7.37 (m, 1 3
arom. H). 13C-NMR (CDCl3, 75 MHz): see Table 4; additionally, 34.06 (t, CH2ÀC(2)); 51.95 (q, MeO); 73.20,
74.60, 74.67 (3t, 3 PhCH2); 127.57 ± 128.37 (several d); 137.20, 137.26, 138.10 (3s); 171.34 (s, CˆO). HR-
‡
‡
‡
‡
MALDI-MS: 581.2053 (2, [M ‡ K] , C32H34KN2O6 ; calc. 581.2054), 565.2304 (32, [M ‡ Na] , C32H34N2NaO6
‡
;
;
‡
‡
‡
calc. 565.2314), 543.2490 (100, [M ‡ H] , C32H35N2O6 ; calc. 543.2495), 525.2366 (18, [M À OH] , C32H33N2O5
calc. 525.2389). Anal. calc. for C32H34N2O6 (542.63): C 70.83, H 6.32, N 5.16; found: C 70.56, H 6.52, N 5.24.
Methyl (5R,6R,7S,8R)-6,7-Bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-8-hydroxyimid-
azo[1,2-a]pyridine-2-acetate (23). White solid. M. p. 109 ± 1118 (AcOEt/hexane 1:1). Rf (AcOEt) 0.11. Rf
(AcOEt/MeOH 10 :1) 0.39. [a]2D5 ˆ ‡0.7 (c ˆ 1.01, CHCl3). UV (CHCl3): 259 (2.91), 240 (3.33). IR (CHCl3):
3304w, 3163w, 3089w, 3067w, 3031m, 3011m, 2954m, 2925m, 2869m, 1953w, 1877w, 1810w, 1736s, 1603w, 1564w,
1496w, 1454m, 1438m, 1407w, 1363m, 1306m, 1269m, 1099s, 1018m, 91 1w. 1H-NMR (CDCl3, 300 MHz): see
Table 5; additionally, 3.57 (d, J ˆ 16.5, CHÀC(2)); 3.64 (d, J ˆ 17.1, CH 'ÀC(2)); 3.68 (s, MeO); 4.45 (d, J ˆ 12.1,
PhCH); 4.50 (d, J ˆ 12.1, PhCH); 4.57 (d, J ˆ 11.5, PhCH); 4.71( d, J ˆ 12.1, PhCH); 4.79 (d, J ˆ 11.8, PhCH);
4.87 (d, J ˆ 11.2, PhCH); 7.22 ± 7.38 (m, 15 arom. H). 1H-NMR ((D6)DMSO, 300 MHz): see Table 5;
additionally, 3.47 (d, J ˆ 16.2, CHÀC(2)); 3.53 (d, J ˆ 16.8, CH'ÀC(2)); 3.58 (s, MeO); 3.67 (irrad. at 4.10 ! d,
J ˆ 10.3); 3.845 (irrad. at 4.17 ! d, J ˆ 3.1, irrad. at 4.86! d, J ˆ 8.4); 3.847 (irrad. at 3.67 ! change, irrad. at
4.10 ! change); 4.10 (irrad. at 3.67 ! change); 4.44 (d, J ˆ 12.1, PhCH); 4.51(br. d, J ˆ 11.2, 2 PhCH); 4.59 (d,
J ˆ 12.1, PhCH); 4.77 (d, J ˆ 11.8, PhCH); 4.79 (d, J ˆ 11.2, PhCH); 4.86 (irrad. at 5.60 ! d, J ˆ 3.4); 5.60 (d, J ˆ
5.0, irrad. at 4.86 ! s, exchange with CD3OD, OH); 7.19 ± 7.39 (m, 15 arom. H). 13C-NMR (CDCl3, 75 MHz): see
Table 4; additionally, 34.22 (t, CH2ÀC(2)); 51.97 (q, MeO); 72.66, 73.19, 74.21 (3t, 3 PhCH2); 127.69 ± 128.37
‡
(several d); 137.27, 137.43, 137.46 (3s); 171.42 (s, CˆO). HR-MALDI-MS: 581.2054 (7, [M ‡ K] ,
‡
‡
‡
C32H34KN2O6 ; calc. 581.2054), 565.2297 (37, [M ‡ Na] , C32H34N2NaO6 ; calc. 565.2314), 543.2484 (100,
‡
‡
‡
‡
[M ‡ H] , C32H35N2O6 ; calc. 543.2495), 525.2357 (18, [M À OH] , C32H33N2O5 ; calc. 525.2389). Anal. calc. for
C32H34N2O6 (542.63): C 70.83, H 6.32, N 5.16; found: C 70.84, H 6.45, N 5.08.
Methyl (R*)- and (S*)-2-Azido-2-{(5R,6R,7S)-6,7-bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetra-
hydroimidazo[1,2-a]pyridin-2-yl}acetate (24). At 08, a soln. of 22 (35 mg, 64.5 mmol) and Ph3P (25 mg,
95.3 mmol) in THF (3 ml) was successively treated with 0.67m HN3 in toluene5) (0.19 ml, 0.127 mmol) and ca.
40% DEAD in toluene (44 ml, 96 mmol), stirred for 5 min at 08 and 3 h at 238, diluted with Et2O (40 ml), and
washed with sat. NH4Cl soln. (3 Â 15 ml). The combined aq. layers were extracted with Et2O (2 Â 15 ml). The
combined org. layers were washed with H2O (30 ml) and brine (30 ml), dried (MgSO4), filtered, and
evaporated. FC (hexane/AcOEt/MeOH 1:1:0 ! 1:3 :0 ! 0 :1 :0! 0 :10 :1) gave 24 (17.3 mg, 47%, (2R*)/
(2S*) 2 :1) and 22 (11.2 mg, 32%).
Data of 24: Colourless oil. Rf (hexane/AcOEt 1:1) 0.16. IR (CHCl3): 3090w, 3067w, 3032w, 3012m, 2956m,
2926m, 2868m, 2109s, 1951w, 1879w, 1810w, 1747s, 1673w, 1603w, 1521w, 1496w, 1454m, 1437m, 1364m, 1334m,
1266m, 1163m, 1097s, 1076s, 1027m, 91 0w. 1H-NMR (CDCl3, 300 MHz, (2R*)/(2S*) 2 :1): 3.00 (dd, J ˆ 5.9, 16.8,
irrad. at 3.25 ! d, J ꢀ 7.2, HÀC(8)); 3.25 (dd, J ˆ 4.7, 16.8, irrad. at 3.00 ! change, H'ÀC(8)); 3.69 (dd, J ˆ 6.2,
10.0, irrad. at 4.24 ! d, J ˆ 9.7, CHÀC(5)); 3.75 (dd, J ˆ 4.7, 10.0, irrad. at 3.69 ! d, J ꢀ 7.5, irrad. at 4.24 ! d, J ˆ
9.7, 0.7 H), 3.76 (dd, J ˆ 4.1, 10.0, 0.3 H) (CH'ÀC(5)); 3.78 (s, 2.1H), 3.79 ( s, 0.9 H) (MeO); 3.92 (dd, J ˆ 4.7,
6.2, irrad. at 4.24 ! d, J ˆ 6.5, 0.7 H), 3.93 (dd, J ꢀ 4.4, 6.5, 0.3 H) (HÀC(6)); 4.00 (dt, J ˆ 4.7, 5.9, irrad. at
3.00 ! change, irrad. at 3.25 ! dd, J ˆ 4.7, 6.2, HÀC(7)); 4.22 (td, J ˆ 4.7, 6.2, 0.3 H), 4.24 (td, J ˆ 4.7, 6.2, irrad. at
3.69 ! t, J ˆ 4.4, 0.7 H) (HÀC(5)); 4.41( d, J ˆ 12.1, 0.3 H), 4.42 (d, J ˆ 12.1, 0.7 H) (PhCH); 4.47 (d, J ˆ 11.8,
0.7 H), 4.49 (d, J ˆ 12.5, 0.3 H) (PhCH); 4.54 (d, J ꢀ 11.8, 0.3 H), 4.57 (d, J ˆ 11.5, 0.7 H) (PhCH); 4.61( d, J ˆ
11.5, PhCH); 4.642 (d, J ˆ 11.8, 0.7 H), 4.644 (d, J ˆ 12.1, 0.3 H) (PhCH); 4.77 (d, J ˆ 11.5, PhCH); 4.87 (br. s,
w1/2 ˆ 1.4, 0.7 H), 4.88 (br. s, w1/2 ˆ 1.4, 0.3 H) (CHÀC(2)); 7.108 (d, J ˆ 0.6, irrad. at 4.875 ! s, 0.7 H), 7.112 (d,
J ˆ 0.6, irrad. at 4.875 ! s, 0.3 H) (HÀC(3)); 7.22 ± 7.38 (m, 15 arom. H). 13C-NMR (CDCl3, 75 MHz, (2R*)/
(2S*) 2 :1): 27.29 (t, C(8)); 52.88 (q, MeO); 58.72 (d, 0.7 C), 58.80 (d, 0.3 C), 59.94 (d, 0.7 C), 60.08 (d, 0.3 C)
(C(5), CHÀC(2)); 70.36 (t, CH2ÀC(5)); 71.55 (t, PhCH2); 73.25, 73.30, 73.35 (3t, 2 PhCH2); 74.52, 74.62 (2d,
C(6), C(7)); 116.53 (d, 0.3 C), 116.65 (d, 0.7 C) (C(3)); 127.47 ± 128.40 (several d); 134.16 (s, 0.3 C), 134.20 (s,
5
)
Prepared according to [28]. The molar concentration was determined by titration with 0.92m NaOH using
phenolphthalein as indicator.

Helvetica Chimica Acta ± Vol. 88 (2005)
19
0.7 C) (C(2)); 137.26 (2s); 137.40 (s); 142.82 (br. s, C(8a)); 169.37 (s, 0.7 C), 169.47 (s, 0.3 C) (CˆO). HR-
‡
‡
‡
MALDI-MS: 568.2550 (59, [M ‡ H] , C32H34N5O5
;
calc. 568.2560), 562.2319 (22, [M ‡ Na À N2] ,
‡
‡
‡
C32H33N3NaO5 ; calc. 562.2318), 540.2489 (100, [M ‡ H À N2] , C32H34N3O5 ; calc. 540.2498), 527.2535 (27,
‡
‡
‡
‡
[M ‡ 2 H À N3] , C32H35N2O5 ; calc. 527.2546), 525.2375 (15, [M À N3] , C32H33N2O5 ; calc. 525.2389), 480.2281
‡
‡
‡
‡
(27, [M À N2 À COOMe] , C30H30N3O3 ; calc. 480.2287), 419.1958 (32, [M ‡ H À N3 À BnO] , C25H27N2O4
;
calc. 419.1971).
Methyl (5R,6R,7S,8S)-8-Azido-6,7-bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-
a]pyridine-2-acetate (25). a) A soln. of 23 (51mg, 94.0 mmol) in toluene (1ml) was successively treated with
DPPA (0.1ml, 0.464 mmol) and DBU (70 ml, 0.469 mmol), stirred for 5 h at 238, diluted with CH2Cl2 (50 ml),
and washed with sat. NH4Cl soln. (3 Â 25 ml). The combined aq. layers were extracted with CH2Cl2 (2 Â 30 ml).
The combined org. layers were washed with H2O (60 ml) and brine (60 ml), dried (MgSO4), filtered, and
evaporated. FC (hexane/AcOEt 1:0 ! 3 :1 ! 1:1) gave 25 (37 mg, 69%).
b) Repetition of the experiment a): a mixture of 295 mg 23 (0.544 mmol), DPPA (0.59 ml, 2.74 mmol), and
DBU (0.41ml, 2.75 mmol) in toluene (5.5 ml) was stirred for 4 h at 23 8. Workup and FC gave 25 (180 mg, 58%).
Data of 25: Yellowish oil. Rf (hexane/AcOEt 1:1) 0.25. [ a]2D5 ˆ ‡28.5 (c ˆ 1.04, CHCl3). UV (CHCl3): 241
(3.53). IR (CHCl3): 3089w, 3067w, 3032w, 3011w, 2954w, 2921w, 2870w, 2109s, 1951w, 1877w, 1810w, 1736m,
1589w, 1564w, 1496w, 1454m, 1438w, 1407w, 1362m, 1332w, 1269m, 1145m, 1098m, 1018m, 91 7w. 1H-NMR
(CDCl3, 300 MHz): see Table 3; additionally, 3.66 (br. s, CH2ÀC(2)); 3.68 (irrad. at 3.77 ! change, irrad. at
4.11 ! d, J ˆ 10.0); 3.71 (s, MeO); 3.77 (irrad. at 4.11 ! d, J ˆ 10.0); 3.90 (irrad. at 3.97 ! change, irrad. at
4.67 ! d, J ˆ 8.4); 3.97 (irrad. at 3.90 ! change, irrad. at 4.11 ! change); 4.11 (irrad. at 3.77 ! dd, J ˆ 2.8, 7.2,
irrad. at 3.97 ! change); 4.39 (d, J ˆ 11.8, PhCH); 4.45 (d, J ˆ 12.1, PhCH); 4.54 (d, J ˆ 11.5, PhCH); 4.67 (irrad.
at 3.90 ! s); 4.82 (d, J ˆ 11.2, PhCH); 4.87 (d, J ˆ 11.2, PhCH); 4.88 (d, J ˆ 11.2, PhCH); 7.18 ± 7.25 (m, 4 arom.
H); 7.27 ± 7.37 (m, 11 arom. H). 13C-NMR (CDCl3, 75 MHz): see Table 4; additionally, 34.39 (t, CH2ÀC(2));
51.98 (q, MeO); 73.19, 74.59, 74.91 (3t, 3 PhCH2); 127.75 ± 128.42 (several d); 136.95 (s); 137.04 (2s); 171.34 (s,
‡
‡
‡
CˆO). HR-MALDI-MS: 590.2363 (25, [M ‡ Na] , C32H33N5NaO5 ; calc. 590.2379), 568.2549 (86, [M ‡ H] ,
‡
‡
‡
C32H34N5O5 ; calc. 568.2560), 562.2317 (82, [M ‡ Na À N2] , C32H33N3NaO5 ; calc. 562.2318), 540.2486 (57, [M ‡
‡
‡
‡
‡
H À N2] , C32H34N3O5 ; calc. 540.2498), 527.2537 (64, [M ‡ 2 H À N3] , C32H35N2O5 ; calc. 527.2546), 525.2374
‡
‡
(100, [M À N3] , C32H33N2O5 ; calc. 525.2389), 434.2066 (18), 324.1337 (18), 310.1182 (24). Anal. calc. for
C32H33N5O5 (567.64): C 67.71, H 5.86, N 12.34; found: C 67.85, H 5.83, N 12.33.
Treatment of 22 with DPPA and DBU. A soln. of 22 (50 mg, 92.1 mmol) in toluene (1ml) was successively
treated with DPPA (99 ml, 0.459 mmol) and DBU (69 ml, 0.462 mmol) and stirred for 22 h at 238. Workup (as
described for the preparation of 25) and FC (hexane/AcOEt 1:0 ! 3 :1 ! 1:1 ! 1:3) of the mixture gave 25
(3.2 mg, 6%) and 26 (1.6 mg, 3%).
Methyl (5R,6R,7S,8R)-8-Azido-6,7-bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-
1
a]pyridine-2-acetate (26). Yellow oil. Rf (hexane/AcOEt 1 :1) 0.13. H-NMR (CDCl3, 300 MHz): see Table 5;
additionally, 3.59 (d, J ˆ 14.6, CHÀC(2)); 3.646 (d, J ˆ 14.6, CH'ÀC(2)); 3.70 (s, MeO); 4.42 (d, J ˆ 12.1,
PhCH); 4.47 (d, J ˆ 11.8, PhCH); 4.56 (d, J ˆ 11.2, PhCH); 4.71( d, J ˆ 11.8, PhCH); 4.76 (d, J ˆ 11.8, PhCH);
4.875 (d, J ˆ 10.9, PhCH); 7.18 ± 7.39 (m, 15 arom. H).
Methyl (5R,6R,7S,8S)-8-Amino-6,7-bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-
a]pyridine-2-acetate (27). A soln. of 25 (31mg, 54.6 mmol) in MeOH (0.5 ml) was successively treated with
Et3N (0.15 ml, 1.08 mmol) and propane-1,3-dithiol (0.15 ml, 1.49 mmol) and stirred for 23 h at 238. The mixture
was diluted with CH2Cl2 (40 ml) and poured onto 0.1m NaOH/ice (40 ml). The layers were separated, and the aq.
layer was extracted with CH2Cl2 (3 Â 20 ml). The combined org. layers were dried (MgSO4). Evaporation and
FC (AcOEt/MeOH 1:0 ! 1 0 :1! 5 :1) gave 27 (25.3 mg, 85%). Yellowish oil. Rf (AcOEt/MeOH 10 :1) 0.11.
[a]2D5 ˆ ‡12.8 (c ˆ 0.98, CHCl3). UV (CHCl3): 258 (2.86), 240 (3.40). IR (CHCl3): 3378w, 3308w, 3167w, 3090w,
3067w, 3032m, 3009m, 2954m, 2929m, 2869m, 1952w, 1875w, 1810w, 1736s, 1587w, 1497m, 1454s, 1438m, 1407w,
1363m, 1269m, 1137s, 1098s, 1027m, 1020m, 940w, 91 2w. 1H-NMR (CDCl3, 300 MHz): see Table 3; additionally,
2.00 (br. s, exchange with CD3OD, NH2); 3.62 (br. s, CH2ÀC(2)); 3.69 (irrad. at 3.76 ! d, J ꢀ 5.0, irrad. at 4.18 !
d, J ꢀ 10.0); 3.70 (s, MeO); 3.76 (irrad. at 4.18 ! d, J ˆ 10.0); 3.84 (irrad. at 3.98 ! d, J ꢀ 5.9, irrad. at 4.12 ! d,
J ˆ 7.5); 3.98 (irrad. at 3.84 ! change, irrad. at 4.18 ! d, J ˆ 7.5); 4.12 (irrad. at 3.84 ! s); 4.18 (irrad. at 3.76 !
br. t, J ꢀ 5.6, irrad. at 3.98 ! change); 4.41( d, J ˆ 12.1, PhCH); 4.46 (d, J ˆ 12.1, PhCH); 4.56 (d, J ˆ 11.2,
PhCH); 4.80 (br. s, PhCH2); 4.82 (d, J ˆ 11.2, PhCH); 7.20 ± 7.36 (m, 15 arom. H). 13C-NMR (CDCl3, 75 MHz):
see Table 4; additionally, 34.38 (t, CH2ÀC(2)); 51.92 (q, MeO); 73.24, 74.11, 74.30 (3t, 3 PhCH2); 127.80 ± 128.53
‡
(several d); 137.33, 137.42, 137.88 (3s); 171.69 (s, CˆO). HR-MALDI-MS: 564.2461(51, [ M ‡ Na] ,
‡
‡
‡
C32H35N3NaO5 ; calc. 564.2474), 542.2639 (14, [M ‡ H] , C32H36N3O5 ; calc. 542.2655), 525.2368 (100,

20
Helvetica Chimica Acta ± Vol. 88 (2005)
‡
‡
[M À NH2] , C32H33N2O5 ; calc. 525.2389). Anal. calc. for C32H35N3O5 ´ 0.5 H2O (550.65): C 69.80, H 6.59, N 7.63;
found: C 69.86, H 6.63, N 7.67.
Methyl (5R,6R,7S,8S)-8-Acetamido-6,7-bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydroimid-
azo[1,2-a]pyridine-2-acetate (28). A soln. of 27 (10 mg, 18.5 mmol) in pyridine/Ac2O 3 :1(0.4 ml) was stirred
for 3 h at 238, evaporated, and co-evaporated with toluene (3 Â 5 ml). FC (AcOEt/MeOH 1:0 ! 20 :1) gave 28
(8.8 mg, 81%). Colourless oil. Rf (AcOEt/MeOH 10 :1) 0.35. [a]2D5 ˆ ‡8.0 (c ˆ 0.40, CHCl3). UV (CHCl3): 258
(2.84), 240 (3.39). IR (CHCl3): 3441w, 3308w (br.), 3089w, 3066w, 3019m, 3009m, 2954m, 2929m, 2869m, 1951w,
1875w, 1810w, 1736s, 1669s, 1603w, 1497s, 1455s, 1438m, 1408w, 1368m, 1268m, 1153m, 1097s, 1028m, 1018m,
911w. 1H-NMR (CDCl3, 300 MHz): see Table 3; additionally, 1.89 (s, AcN); 3.63 (br. s, CH2ÀC(2)); 3.70 (s,
MeO); 3.72 (irrad. at 4.44 ! d, J ꢀ 10.3); 3.76 (irrad. at 4.44 ! d, J ꢀ 11.5); 4.02 (irrad. at 4.44 ! br. d, J ˆ 5.0,
irrad. at 5.38 ! dd, J ˆ 2.5, 4.7); 4.06 (irrad. at 4.44 ! dd, J ˆ 2.8, 4.7, irrad. at 5.38 ! br. d, J ˆ 4.7); 4.42 (d, J ˆ
11.8, PhCH); 4.48 (d, J ˆ 11.8, PhCH); 4.52 (d, J ˆ 12.5, PhCH); 4.56 (d, J ˆ 11.8, PhCH); 4.68 (d, J ˆ 12.1,
PhCH); 4.84 (d, J ˆ 11.8, PhCH); 5.38 (irrad. at 6.02 ! br. d, J ˆ 2.5); 6.02 (d, J ˆ 8.1, irrad. at 5.38! s, NH);
7.15 ± 7.20 (m, 2 arom. H); 7.23 ± 7.38 (m, 13 arom. H). 13C-NMR (CDCl3, 75 MHz): see Table 4; additionally,
23.38 (q, MeCˆO); 34.38 (t, CH2ÀC(2)); 52.00 (q, MeO); 72.50, 72.55, 73.39 (3t, 3 PhCH2); 127.59 ± 128.52
(several d); 136.81, 137.40, 137.49 (3s); 168.92 (s, NHCˆO); 171.54 (s, CO2Me). HR-MALDI-MS: 622.2312 (2,
‡
‡
‡
‡
[M ‡ K] , C34H37KN3O6 ; calc. 622.2319), 606.2564 (95, [M ‡ Na] , C34H37N3NaO6 ; calc. 606.2580), 584.2748
‡
‡
‡
‡
(100, [M ‡ H] , C34H38N3O6 ; calc. 584.2761), 525.2361 (4, [M À AcNH] , C32H33N2O5 ; calc. 525.2389),
‡
‡
‡
‡
476.2168 (73, [M À BnO] , C27H30N3O5 ; calc. 476.2185), 434.2067 (14, [M ‡ H À BnO À Ac] , C25H28N3O4
;
calc. 434.2080), 338.1499 (12), 260.1023 (15). Anal. calc. for C34H37N3O6 (583.68): C 69.97, H 6.39, N 7.20; found:
C 69.69, H 6.60, N 7.00.
Methyl (5R,6R,7S,8S)-8-Acetamido-5,6,7,8-tetrahydro-6,7-dihydroxy-5-(hydroxymethyl)imidazo[1,2-a]py-
ridine-2-acetate (14). a) A mixture of 25 (142 mg, 0.250 mmol) and 10% Pd/C (140 mg) in AcOH (6 ml) was
hydrogenated for 22 h at 6 bar and filtered over Celite (washing with 40 ml MeOH). The combined filtrates were
evaporated and co-evaporated with toluene (5 Â 5 ml). The residue (91mg) was dissolved in MeOH (9 ml),
treated with Ac2O (4.5 ml), and stirred for 4 h at 238. Evaporation of the mixture and co-evaporation with
toluene (4 Â 5 ml) yielded a yellow oil (122 mg), which was dissolved in 2m NH3 in MeOH (3 ml) and heated for
1 h at 408. Evaporation and FC (AcOEt/MeOH/H2O 1:0 :0 ! 5:1:1) gave 14 (47.2 mg, 60%).
b) A soln. of 28 (47 mg, 80.5 mmol) in AcOH (2 ml) was treated with 10% Pd/C (27 mg) and hydrogenated
for 22 h at 6 bar. Filtration over Celite (washing with 30 ml MeOH), evaporation of the combined filtrates, co-
evaporation with toluene (3 Â 5 ml), and FC (AcOEt/MeOH/H2O 1:0 :0 ! 5 :1:1) gave 14 (24.7 mg, 98%).
Data of 14: White solid. M. p. 208 ± 2128. Rf (AcOEt/MeOH/H2O 5 :1 :1) 0.12. [a]2D5 ˆ ‡24.0 (c ˆ 0.89,
H2O). pKha ˆ 4.62. UV (H2O): 216 (3.77), 192 (3.91). IR (0.4% in KBr): 3600 ± 2600s (br.), 3585s, 3415s, 3261s,
3098s, 2956m, 2918m, 2887m, 1725s, 1650s, 1566s, 1508m, 1440s, 1408m, 1375m, 1349m, 1317m, 1288s, 1228m,
1205m, 1164m, 1128m, 1087m, 1070m, 1055m, 1018m, 990m, 951w, 905w, 867w, 760w, 700m, 635m, 594m.
1H-NMR (D2O, 300 MHz): see Table 3; additionally, 2.12 (s, AcN); 3.68 (br. s, CH2ÀC(2)); 3.73 (s, MeO); 3.90
(with virtual coupling, irrad. at 4.89 ! d, J ˆ 9.7, with virtual coupling); 3.94 ± 4.06 (irrad. at 3.90 ! change, irrad.
at 4.22 ! change); 4.08 (irrad. at 4.22 ! br. s); 4.89 (irrad. at 3.90 ! s). 1H-NMR (CD3OD, 300 MHz): see
Table 3; additionally, 2.06 (s, AcN); 3.59 (br. s, CH2ÀC(2)); 3.68 (s, MeO); 3.73 (with virtual coupling, irrad. at
4.95 ! br. d, J ˆ 8.7); 3.84 ± 3.91(irrad. at 3.73 ! change, irrad. at 4.16 ! change); 3.94 (irrad. at 4.16 ! d, J ꢀ
4.4); 4.95 (irrad. at 3.73 ! s). 13C-NMR (D2O, 75 MHz): see Table 4; additionally, 22.18 (q, MeCˆO); 32.84 (t,
CH2ÀC(2)); 52.56 (q, MeO); 174.10, 174.18 (2s, NHCˆO, CO2Me). HR-MALDI-MS: 336.1163 (55, [M ‡
‡
‡
‡
‡
Na] , C13H19N3NaO6 ; calc. 336.1172), 314.1343 (99, [M ‡ H] , C13H20N3O6 ; calc. 314.1352), 296.1240 (100,
‡
‡
‡
‡
[M À OH] , C13H18N3O5 ; calc. 296.1246), 254.1131 (42, [M À CO2Me] , C11H16N3O4 ; calc. 254.1141).
(5R,6R,7S,8S)-8-Acetamido-5,6,7,8-tetrahydro-6,7-dihydroxy-5-(hydroxymethyl)imidazo[1,2-a]pyridine-2-
acetic Acid (15). At 08, a soln. of 14 (11.2 mg, 35.7 mmol) in H2O (0.3 ml) was treated with 0.4m KOH in EtOH/
H2O 4 :1(0.5 ml), stirred for 45 min at 0 8, treated with 0.1m HCl (5 ml), and evaporated. The residue was taken
‡
up in H2O (2 ml) and subjected to ion-exchange chromatography (Amberlite CG-120, H form, elution with 0.1m
aq. NH3). Evaporation and lyophilisation gave 15 (8.1mg, 76%). Colourless hygroscopic resin containing
substantial amounts of H2O. The sample for microanalysis was dried for 8 d at 10À4 Torr. Rf (AcOEt/MeOH/H2O
3 :1:1) 0.09. [ a]D25 ˆ ‡28.0 (c ˆ 0.39, H2O). pKha ˆ 5.32 (no additional pK value was observed in the pH range
2.3 ± 5.7). UV (H2O): 222 (3.81). IR (0.4% in KBr): 3600 ± 2400s (br.), 2923m, 2852m, 1647s, 1580s, 1460m,
1
1414m, 1383s, 1319m, 1273m, 1202w, 1167w, 1101m, 1075m, 1042m, 909w, 873w, 704m, 660m, 586m. H-NMR
(D2O, 300 MHz): see Table 3; additionally, 2.10 (s, AcN); 3.49 (br. s, CH2ÀC(2)). 13C-NMR (D2O, 75 MHz): see
Table 4; additionally, 22.09 (q, MeCˆO); 34.68 (t, CH2ÀC(2)); 174.39 (s, NHCˆO); 177.62 (s, CO2H). HR-
‡
‡
‡
MALDI-MS: 344.0817 (9, [M À H ‡ 2 Na] , C12H16N3Na2O6 ; calc. 344.0834), 322.1005 (83, [M ‡ Na] ,

Helvetica Chimica Acta ± Vol. 88 (2005)
21
‡
‡
‡
C12H17N3NaO6 ; calc. 322.1015), 300.1189 (100, [M ‡ H] , C12H18N3O6 ; calc. 300.1196), 282.1084 (43,
‡
‡
‡
‡
[M À OH] , C12H16N3O5 ; calc. 282.1090), 240.0977 (9, [M À CH2CO2H] , C10H14N3O4 ; calc. 240.0984). Anal.
calc. for C12H17N3O6 ´ H2O (317.30): C 45.42, H 6.04, N 13.24; found: C 45.25, H 5.63, N 13.42.
Inhibition Studies. The inhibition constants (Ki) or the IC50 values were determined for a range of inhibitor
concentrations (typically 4 ± 7 concentrations) that bracket the Ki or IC50 value, and the substrate concentrations
that bracket the KM of each enzyme (for Ki, typically 5 ± 7 concentrations), or correspond to it (for IC50).
a) Inhibition of Jack Beans N-Acetyl-b-glucosaminidase. Km ˆ 0.57 ± 0.81m m ([29]: Km ˆ 0.62 mm; [30]:
Km ˆ 0.72 mm). Inhibition constants (Ki) and IC50 values were determined at 258 at an enzyme concentration of
0.2 units/ml, with a 0.04m citric acid/sodium citrate buffer (pH 5.0), containing 0.08m NaCl, and 4-nitrophenyl 2-
acetamido-2-deoxy-b-d-glucopyranoside as the substrate. The enzymatic reaction was started, after incubation
of the enzyme (15 ml) in the presence of the inhibitor (15 ml) and buffer (100 ml) for 1h at 25 8, by addition of the
substrate (20 ml). The enzyme reaction was quenched by the addition of 0.2m borate buffer (pH 9.0, 150 ml)
after 10 min, and the absorption at 405 nm was taken as rate of hydrolysis of the substrate after subtraction of
the absorption of a blank probe (H2O, buffer, substrate). IC50 Values were determined by plotting the reciprocal
value of the rate of substrate hydrolysis vs. the inhibitor concentration. After fitting a straight line to the data by
linear regression, the negative [I]-intercept of this plot provided the appropriate IC50 value. Ki Values were
determined by taking the slopes from the Lineweaver-Burk plots [31] and plotting them vs. the inhibitor
concentrations [27]. After fitting a straight line to the data by linear regression, the negative [I]-intercept of this
plot provided the appropriate Ki value. a Values were determined by plotting the 1/v axis intercepts of the
Lineweaver-Burk plots vs. the inhibitor concentrations [27]. After fitting a straight line to the data by linear
regression, the negative [I]-intercept of this plot provided the appropriate a ´ Ki value.
b) Inhibition of Bovine Kidney N-Acetyl-b-glucosaminidase. Km ˆ 1.17 ± 1.62 mm ([32]: Km ˆ 1.87 mm; [30]:
Km ˆ 1.24 mm). As described in a, inhibition studies were carried out at 378 at an enzyme concentration of 0.5
units/ml, using a 0.05m citric acid/sodium citrate buffer (pH 4.1). The enzymatic reaction was started after the
incubation at 378 for 1h and quenched by the addition of 0.2 m borate buffer (pH 9.8, 150 ml).
c) Inhibition of C. saccharolyticum b-Glucosidase. Km ˆ 0.68 ± 0.77 mm ([33]: Km ˆ 0.51m m; [34]: Km ˆ
0.51± 0.78 m m). Inhibition studies were carried out at 558 at an enzyme concentration of 0.005 units/ml, with a
0.08m KH2PO4/K2HPO4 buffer (pH 6.8) and 4-nitrophenyl b-d-glucopyranoside as the substrate. The enzymatic
reaction was started after incubation of the enzyme (150 ml) in the presence of the inhibitor (40 ml) for 1h at 55 8
by addition of the substrate (10 ml). The enzyme reaction was quenched by addition of 0.2m borate buffer
(pH 9.0, 100 ml) after 30 min, and the absorption at 405 nm was taken as rate of hydrolysis of the substrate after
subtraction of the absorption of a blank probe (H2O, buffer, substrate). IC50, Ki, and a values were determined
by plots as described in a.
d) Inhibition of Snail b-Mannosidase. Km ˆ 0.53 ± 0.58 mm ([8]: Km ˆ 0.42 ± 0.80 mm). Inhibition constants
(Ki) and IC50 values were determined at 258 at an enzyme concentration of 0.048 units/ml, using a 0.04m acetate
buffer (pH 4.5) and 4-nitrophenyl b-d-mannopyranoside as the substrate. The enzymatic reaction was started,
after incubation of the enzyme (100 ml) in the presence of the inhibitor (50 ml) for 1h at 25 8, by addition of the
substrate (50 ml). The enzyme reaction was quenched by addition of 0.2m borate buffer (pH 9.0, 100 ml) after
5 min, and the absorption at 405 nm was taken as rate of the hydrolysis of the substrate after subtraction of the
absorption of a blank probe (H2O, buffer, substrate). IC50, Ki, and a values were determined by plots as
described in a.
REFERENCES
[1] T. Aoyagi, H. Suda, K. Uotani, F. Kojima, T. Aoyama, K. Horiguchi, M. Hamada, T. Takeuchi, J. Antibiot.
1992, 45, 1404.
[2] K. Tatsuta, S. Miura, S. Ohta, H. Gunji, J. Antibiot. 1995, 48, 286.
[3] K. Tatsuta, S. Miura, Tetrahedron Lett. 1995, 36, 6721.
[4] K. Tatsuta, S. Miura, H. Gunji, Bull. Chem. Soc. Jpn. 1997, 70, 427.
[5] T. Aoyama, H. Naganawa, H. Suda, K. Uotani, T. Aoyagi, T. Takeuchi, J. Antibiot. 1992, 45, 1557.
[6] N. Panday, A. Vasella, Synthesis 1999, 1459.
[7] N. Panday, Y. Canac, A. Vasella, Helv. Chim. Acta 2000, 83, 58.
[8] M. Terinek, A. Vasella, Helv. Chim. Acta 2003, 86, 3482.
[9] M. Terinek, A. Vasella, Helv. Chim. Acta, 2004, 87, 3035.
[10] N. Panday, A. Vasella, Helv. Chim. Acta 2000, 83, 1205.

22
Helvetica Chimica Acta ± Vol. 88 (2005)
[11] B. Schlesier, V. H. Nong, C. Horstmann, M. Hennig, J. Plant Physiol. 1996, 147, 665.
[12] M. Hennig, J. N. Jansonius, A. C. Terwisscha van Scheltinga, B. W. Dijkstra, B. Schlesier, J. Mol. Biol. 1995,
254, 237.
[13] K. Tatsuta, S. Miura, S. Ohta, H. Gunji, Tetrahedron Lett. 1995, 36, 1085.
[14] M. Terinek, A. Vasella, Tetrahedron: Asymmetry, in press.
[15] N. Panday, M. Meyyappan, A. Vasella, Helv. Chim. Acta 2000, 83, 513.
[16] E. J. Corey, A. Venkateswarlu, J. Am. Chem. Soc. 1972, 94, 6190.
[17] E. J. Corey, D. H. Hua, B.-C. Pan, S. P. Seitz, J. Am. Chem. Soc. 1982, 104, 6818.
[18] O. Mitsunobu, Synthesis 1981, 1 .
[19] H. Loibner, E. Zbiral, Helv. Chim. Acta 1976, 59, 2100.
[20] H. Loibner, E. Zbiral, Helv. Chim. Acta 1977, 60, 41 7.
[21] E. Brunn, R. Huisgen, Angew. Chem., Int. Ed. 1969, 8, 513.
[22] A. S. Thompson, G. R. Humphrey, A. M. DeMarco, D. J. Mathre, E. J. J. Grabowski, J. Org. Chem. 1993,
58, 5886.
[23] N. Panday, T. Granier, A. Vasella, Helv. Chim. Acta 1998, 81, 475.
[24] H. Bayley, D. N. Standring, J. R. Knowles, Tetrahedron Lett. 1978, 19, 3633.
[25] M. Terinek, A. Vasella, Helv. Chim. Acta 2004, 87, 719.
[26] T. D. Heightman, P. Ermert, D. Klein, A. Vasella, Helv. Chim. Acta 1995, 78, 514.
[27] I. H. Segel, ꢁEnzyme Kineticsꢂ, John Wiley & Sons, New York, 1975.
[28] H. Wolff, Org. React. 1955, 3, 327.
[29] E. C. K. Lai, S. G. Withers, Biochemistry 1994, 33, 14743.
[30] M. Böhm, A. Vasella, Helv. Chim. Acta 2004, 87, 2566.
[31] H. Lineweaver, D. Burk, J. Am. Chem. Soc. 1934, 56, 658.
[32] A. J. Pope, N. Mian, D. G. Herries, FEBS Lett. 1978, 93, 174.
[33] A. R. Plant, J. E. Oliver, M. L. Patchett, R. M. Daniel, H. W. Morgan, Arch. Biochem. Biophys. 1988, 262,
181.
[34] C. Blüchel, C. V. Ramana, A. Vasella, Helv. Chim. Acta 2003, 86, 2998.
Received October 5, 2004