Synthesis and evaluation of two mannosamine-derived lactone-type inhibitors of snail β-mannosidase

Article abstract of DOI:10.1016/j.tetasy.2004.11.068

The inhibition of snail β-mannosidase by the manno-configured amino- and hydroxy-lactams and -imidazoles 7-10 was compared to the inhibition of the β-glucosidases from Caldocellum saccharolyticum and from sweet almonds by the gluco-configured amino- and hydroxy-lactams and -imidazoles 1, 2, 5 and 6 [ΔΔG_diss.(OH → NH₃⁺)]. Substitution in the gluco-configured 1, 3 and 5, of C(2)-OH by an ammonium group strengthens the interaction of the inhibitor with the catalytic nucleophile of retaining β-glucosidases, and weakens the interaction with the catalytic acid. The analogous substitution in the manno-configured inhibitors 7 and 9, leading to 8 and 10, respectively, was expected to only reflect the impaired interaction of the inhibitor with the catalytic acid, as the catalytic nucleophile and the C(2) substituent are located on opposite sides of the average ring plane. The mannonolactam 10 was synthesized from the known hydroxy-lactam 11 by O-mesylation followed by azidation and hydrogenation. Sultone 13 was formed as side product upon mesylation of 11. The imidazole 8 was obtained from 11, similarly to the synthesis of the known gluco-isomer 2, via the hydroxy-imidazoles 22 and 23; best results were obtained by protecting 11 as the triisopropylsilyl ether 29. The resulting inhibition by the imidazoles 7 and 8 was interpreted as reflecting an improved binding of the catalytic nucleophile of snail β-mannosidase with the protonated imidazole ring of 8 and an impaired interaction with the catalytic acid, while a comparison of the inhibition by the lactams 9 and 10 is in keeping with the results that are expected if there is no significant interaction between the catalytic nucleophile of snail β-mannosidase and the C(2)-OH group of β-mannosides. The amino-imidazole 8 is a surprisingly strong inhibitor of the α-mannosidase from Jack beans [K_i = 1.22 μM; mixed-type (α = 2.3)].

Full text of DOI:10.1016/j.tetasy.2004.11.068

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 449–469
Synthesis and evaluation of two mannosamine-derived
lactone-type inhibitors of snail b-mannosidase
Miroslav Terinek and Andrea Vasella*
Laboratorium fu¨r Organische Chemie, ETH-Ho¨nggerberg, Wolfgang-Pauli Strasse 10, CH-8093 Zu¨rich, Switzerland
Received 8 November 2004; accepted 25 November 2004
Abstract—The inhibition of snail b-mannosidase by the manno-configured amino- and hydroxy-lactams and -imidazoles 7–10 was
compared to the inhibition of the b-glucosidases from Caldocellum saccharolyticum and from sweet almonds by the gluco-configured
amino- and hydroxy-lactams and -imidazoles 1, 2, 5 and 6 [DDGdiss.(OH ! NH3+)]. Substitution in the gluco-configured 1, 3 and 5,
of C(2)–OH by an ammonium group strengthens the interaction of the inhibitor with the catalytic nucleophile of retaining b-glu-
cosidases, and weakens the interaction with the catalytic acid. The analogous substitution in the manno-configured inhibitors 7 and
9, leading to 8 and 10, respectively, was expected to only reflect the impaired interaction of the inhibitor with the catalytic acid, as
the catalytic nucleophile and the C(2) substituent are located on opposite sides of the average ring plane.
The mannonolactam 10 was synthesized from the known hydroxy-lactam 11 by O-mesylation followed by azidation and hydro-
genation. Sultone 13 was formed as side product upon mesylation of 11. The imidazole 8 was obtained from 11, similarly to the
synthesis of the known gluco-isomer 2, via the hydroxy-imidazoles 22 and 23; best results were obtained by protecting 11 as the
triisopropylsilyl ether 29.
The resulting inhibition by the imidazoles 7 and 8 was interpreted as reflecting an improved binding of the catalytic nucleophile of
snail b-mannosidase with the protonated imidazole ring of 8 and an impaired interaction with the catalytic acid, while a comparison
of the inhibition by the lactams 9 and 10 is in keeping with the results that are expected if there is no significant interaction between
the catalytic nucleophile of snail b-mannosidase and the C(2)–OH group of b-mannosides. The amino-imidazole 8 is a surprisingly
strong inhibitor of the a-mannosidase from Jack beans [Ki = 1.22 lM; mixed-type (a = 2.3)].
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
imidazole. The effect on the hydrogen bond from the
catalytic acid is not relevant for the inhibition by the
8-hydroxypyrrole 3 and the corresponding amine 4,
and the stronger hydrogen bond from the ammonium
group to the catalytic nucleophile leads to an stronger
inhibition by 4. In agreement with this interpretation,
substitution of the C(2)–OH group of the gluconolactam
5 by an amino group led to an slightly increased inhibi-
tion by 6 of the b-glucosidases from C. saccharolyticum
(DDGdiss. = À1.1 kcal/mol) and from sweet almonds
(DDGdiss. = À1.1 kcal/mol) evidencing that the carbonyl
group of 6 is a relatively weak H-bond acceptor for
the catalytic acid since the overall effect on the inhibition
is dominated by the interaction of the ammonium group
with the catalytic nucleophile.
Substitution of the C(8)–OH group of the gluco-config-
ured tetrahydropyridoimidazole 1 (Fig. 1) by an amino
group, as in 2, leads to a significantly weaker inhibition
of the b-glucosidases from Caldocellum saccharolyticum
(DDGdiss. = +4.7 kcal/mol) and from sweet almonds
(DDGdiss. = +4.4 kcal/mol). The analogous substitution
of the tetrahydropyrrolopyridine 3 into 4, leads, how-
ever, to a stronger inhibition (DDGdiss. = À2.8 to
À2.9 kcal/mol).1 These observations were interpreted
as the result of opposite influences. While the ammo-
nium group forms a stronger hydrogen bond to the cata-
lytic nucleophile of the b-glucosidases than the hydroxy
group, this effect is overcompensated by the lowered
basicity of the imidazole ring of 2 and the impairment
of the hydrogen bond from the catalytic acid to the
We wished to explore the effect on the inhibition of snail
b-mannosidase of the analogous substitution of C(8)–
OH of the manno-configured tetrahydroimidazole 7
and of the corresponding lactam 9 by an amino group.
As the catalytic nucleophile of this mannosidase and
*
Corresponding author. Tel.: +41 01 632 45 10; e-mail:
vasella@org.chem.ethz.ch
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.11.068

450
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
OH
N
8O
H
OH
N
OH
NH
3
6
HO
HO
HO
HO
HO
HO
N 1
OMe
O
O
O
H
O
H
O
H
H
H
O
O
O
–
O
–
O
–
O
O
O
O
O
O
E
E
E
3
1
5
E
E
E
OH
NH
OH
N
OH
N
HO
HO
HO
HO
HO
HO
O
OMe
O
N
+
+
+
H
H
H
H
H
H
N
N
N
O
H
H
H
O
O
O
H
H
H
–
–
–
O
O
O
O
O
O
O
O
E
E
E
4
6
2
E
E
E
OH
HO
HO
OH
N
H
O
OH
N
OH
HO
N
HO
OH
7
9
OH
OH
HO
HO
HO
HO
HO
HO
H
NH3
NH2
N
+
O
H3N
OH
N
N
N
OH
+
N
N
O
H
H3N
H
8.H+
OH
3H4
10.H+
B2,5
Figure 1.
the C(8) substituent of 7 and the corresponding amine 8
(and similarly the C(2) substituents of the lactams 9 and
10) are located on opposite sites of the average plane of
the tetrahydropyridine ring,ꢀ one does not expect an
interaction of the C(8) substituent with the catalytic
nucleophile. As snail b-mannosidase has a lower pH-
optimum (pH 4.0–4.53) than the C. saccharolyticum
and sweet almond b-glucosidases (pH 6.2 and 5.6,
resp.4,5) 7 and 8 will be extensively protonated at the
low pH of the assay. Protonation of 8 should lead to
an equilibrating mixture of ammonium and imidazolium
salts 8ÆH+; the interaction of both species with the cata-
lytic acid will be impaired.ꢁ Since protonation may
result in a compensating binding interaction of the cat-
alytic nucleophile with the imidazolium (and imidazole)
ring7,8 we also wished to test the essentially neutral hy-
droxy-lactam 9 and the corresponding aminomannono-
lactam 10. A B2,5 conformation for the (protonated)
amine 10 is expected by analogy to the mannono-1,5-
lactam 9 where the B2,5-conformer is found in the solid
state and dominates in solution.9
2. Synthesis
The manno-configured C(2)-amino lactam 10 was pre-
pared in three steps from the known gluco-hydroxy-lac-
tam 111 (Scheme 1). Treatment of 11 with MsCl and
Et3N in CH2Cl2 at 0 ꢁC gave a 18:1 mixture of the mes-
ylate 12 and the sultone 13.§ which were isolated by
chromatography in 91% and 5% yield, respectively.
These results were hardly affected by varying the reac-
tion temperature (À78 to 23 ꢁC) and/or the amount of
MsCl (1.2–2.4 equiv) and Et3N (2–6 equiv). Replacing
Et3N/CH2Cl2 by pyridine did not lead to complete con-
version, and 11 was invariably recovered (18%) besides a
15:1 mixture of 12 (73%) and 13 (5%). The yield of the
sultone 13 increased to 35–50% upon treating 11 with
MsCl and i-Pr2NH. The sultone was always obtained to-
gether with the mesylate 12 (48–60%). Mesylation in the
presence of 2,6-lutidine, NaH or DBU led to poor con-
version of the lactam 11 to 12; TLC indicated that sul-
tone 13 was not formed under these conditions.
Treating the mesylate 12 with excess Et3N, i-Pr2NH or
DBU did not lead to 13.
The formation of 13 may be rationalized by assuming
that the iminomesylate A is generated first. Condensa-
tion of A with the carbanion derived from the reaction
ꢀ The location of the C(2) substituent and the catalytic nucleophile on
opposite sites of the mannopyranoside ring was shown by the crystal
structure of the complex between 2,4-dinitrophenyl 2-deoxy-2-fluoro-
b-mannotrioside and the E212A mutant of the Man2A
mannosidase.2
ꢁ If pseudoequatorially oriented, the ammonium group of 8ÆH+ could
also form a hydrogen bond to the imidazole ring, but the geometry of
this hydrogen bond does not appear favourable (modelling by
Macromodel MM3*6).
§ There are only a few examples of the transformation, under standard
mesylation conditions, of an a-hydroxy ketone to a dehydro-c-
sultone10–13 and/or a d-hydroxy-c-sultone.11–16 One sugar-derived
sultone was prepared in this way.17 For a recent review on carbanion-
mediated intermolecular coupling reactions of sulfonates (or sulfon-
amides), see Ref. 18.

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
451
OBn
NH
OBn
NH
7
a), b)
or c)
e)
BnO
BnO
BnO
+
R2O
R2O
2
BnO
1
3
O
OBn
NH
OMs
12
O
13
NH
SO2
2 R1
O
BnO
BnO
OR
OBn
O
OH
)
f
d)
BnO
BnO
NH
11
15 R1 = N3, R2 = Bn
10 R1 = NH2, R2 = H
g)
O
OTf
14
OBn
NH
OBn
N
OBn
N
OBn
BnO
BnO
BnO
BnO
BnO
BnO
BnO
BnO
N
OMs
OMs
SO2
O
O
O
SO2
O
H
H
H
SO2 Me
X
A
B
D
C
BnO
BnO
OBn
NH
OH
NH
NH
OR
NH
OH
)
a) or d)
e) or
g)
O
f
BnO
BnO
HO
HO
OBn
O
OBn
O
O
BnO
BnO
N3
NH2
16
17 R = Ms
18 R = Tf
19
6
Ms = MeSO2, Tf = CF3SO2
Scheme 1. Reagents and conditions: (a) MsCl, Et3N, CH2Cl2, 0 ꢁC; 91% of 12 and 5% of 13; 91% of 17. (b) MsCl, pyridine, 0 ! 23 ꢁC; 73% of 12, 5%
of 13, and 18% of 11. (c) MsCl, i-Pr2NH, CH2Cl2, 0 ꢁC; 48% of 12 and 50% of 13. (d) Tf2O, pyridine, CH2Cl2, À78 ! À10 ꢁC; 73% of 14; 60% of 18.
(e) NaN3, DMF, 70 ꢁC; 77% of 15 from 12; 62% of 19 from 17. (f) As (e), but 23 ꢁC; 52% of 15 from 14; 26% of 19 from 18. (g) H2, Pd/C, MeOH/
AcOH; 92% of 10; 96% of 6.
of the sulfene resulting from the elimination of HCl
from MsCl19,20 and the amine may form B. Elimination
transforms B to C, and further to 13. Alternatively, the
imino ether A may be mesylated to D. Deprotonation of
D followed by cyclization also provides 13.
synthesis of 6 over the known synthesis, as judged by
yields, number of steps and price of starting materials.
We planned to prepare the manno-configured amino-
imidazole 8 by a similar strategy as described for the
synthesis of the gluco analogue 2.1 In the course of this
synthesis, the lactam 11 was transformed in four steps
(O-acetylation, thionation, Hg(OAc)-promoted conden-
sation with aminoacetaldehyde dimethyl acetal and
treatment with p-TsOHÆH2O) and in an overall yield
of 44% to a 1:1 mixture of the known gluco-and man-
no-alcohols 22 and 231,24,25 (Scheme 2). Depending on
the concentration of HN3, Mitsunobu substitution26–28
of the hydroxy-imidazoles 22 and 23 led either exclu-
sively to the pure gluco-azide 40 or to a ca. 1:1 mixture
of 40 and its manno-isomer 41 (Scheme 3).1,29 We exam-
ined the influence of the nature of the protecting group
at O–C(2) on the yield and ratio of the alcohols 22 and
23 intending to find a more readily reproducible alterna-
tive for their transformation into the manno-configured
azido-imidazole 41 (Schemes 2 and 3).
The mesylate 12 reacted with NaN3 at 70 ꢁC to provide
the manno-configured azido-lactam 15 in 77% yield.
Hydrogenation of 15 in the presence of 10% Pd/C
yielded 92% of a nonseparable 90:10 mixture of the manno-
amino-lactam 10 and its gluco-isomer 6 (64% from 11).
The azide 15 was obtained in lower yields (38%) by
triflation of 11 to 14 (73%) and its reaction with NaN3
at 23 ꢁC to afford 52% of 15.
In parallel to the synthesis of the 2-amino-2-deoxy-
mannonolactam 10 from the gluconolactam 11, we
examined the preparation of the known 2-amino-2-
deoxy-gluconolactam 6 from the known mannonolac-
tam 16,1 hoping to find an advantageous alternative to
the known synthesis.– Mesylation of 16 cleanly afforded
91% of 17. Treatment of 17 with NaN3 led to the gluco-
azide 19 (62%), which was hydrogenated to 6 (96%; 54%
from 16). Proceeding via the labile triflate 18 led to lower
yields. There is no significant advantage of this new
The silyl ethers 20 (86%), 24 (91%) and 29 (92%) were
readily obtained by treating the hydroxy-gluconolactam
11 with the corresponding silyl triflates (pyridine,
CH2Cl2, 0 ꢁC). Under these conditions, the (tert-
butyl)dimethylsilyl ether 24 was accompanied by 3% of
the manno-isomer 25. The silyl ethers 24 and 29 were ob-
tained in slightly improved yields by silylating 11 with
the corresponding chlorosilanes; no C(2)-epimerized
by-products were then observed. Thionation of the
– The lactam 6 was obtained in 92% by hydrogenation of tri-O-benzyl-
glucosamino-1,5-lactam,1 which was prepared in five steps from tri-
O-benzyl-N-acetyl-glucosamine21,22 and its N-benzyloxycarbonyl
analogue23 in overall yields of 54% and 35%, respectively.

452
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
OBn
NH
OBn
NH
BnO
BnO
R2
N
a)
b)
c)
BnO
BnO
BnO
BnO
BnO
N
O
S
R1
OSiEt3
20
OSiEt3
21
22 R1 = OH, R2 = H
23 R1 = H, R2 = OH
OBn
NH
OBn
BnO
R2
d) or
e)
R2
b)
c)
BnO
BnO
BnO
BnO
BnO
N
NH
+
11
BnO
N
S
O
R1
R1
OBn
NH
OTBDMS
26
24 R1 = OTBDMS, R2 = H
25 R1 = H, R2 = OTBDMS
27 R1 = OTBDMS, R2 = H
28 R1 = H, R2 = OTBDMS
22 R1 = OH, R2 = H
BnO
BnO
O
OH
OBn
OBn
NH
OBn
11
f) or
g)
b)
c)
BnO
BnO
BnO
BnO
BnO
BnO
NH
N
N
O
S
OTIPS
31
OTIPS
29
OTIPS
30
h), i), j)
or k)
OBn
NH
OBn
NH
OBn
N
R
b)
BnO
BnO
BnO
BnO
BnO
BnO
+
O
S
O
OR
OR
OR
32 R = Piv
35 R = MOM
33 R = Piv
36 R = MOM
34 R = Piv
TBDMS = t-BuMe2Si, TIPS = i-Pr3Si, Piv = Me3CCO, MOM = MeOCH2
Scheme 2. Reagents and conditions: (a) Et3SiOTf, pyridine, CH2Cl2, 0 ꢁC; 86%. (b) Lawessonꢂs reagent, toluene, 23 ꢁC; 73% of 21; 91% of 26; 96% of
30; 58% of 34. (c) 1. Aminoacetaldehyde dimethyl acetal, Hg(OAc)2, THF, 0 ꢁC; 2. p-TsOHÆH2O, toluene/H2O, 70 ꢁC; 40% of 22/23 60:40 from 21;
8% of 27, 1% of 28, 7% of 11, and 83% of 22 from 26; 88% of 31 from 30. (d) TBDMSCl, imidazole, DMF, 23 ꢁC; 97% of 24. (e) TBDMSOTf,
pyridine, CH2Cl2, 0 ꢁC; 91% of 24 and 3% of 25. (f) TIPSCl, imidazole, DMF, 23 ꢁC; 97%. (g) TIPSOTf, pyridine, CH2Cl2, 0 ꢁC; 92%. (h) Pivaloyl
chloride, pyridine, 5 ! 50 ꢁC; 68% of 32 and 7% of 33. (i) As (h), but 50 ꢁC; 17% of 32 and 76% of 33. (j) CH2(OMe)2, P2O5, CHCl3, 23 ꢁC; 62% of 35
and 9% of 36. (k) P2O5, CH2(OMe)2, 23 ꢁC; 46% of 35 and 23% of 36.
OBn
N
OBn
OBn
N
6
3
5
8
a)
2
b)
BnO
BnO
BnO
BnO
BnO
BnO
N
8a
N
N
N
7
OTBDMS
27
OH
22
OTIPS
31
BnO
BnO
BnO
BnO
OTBDMS
OH
N
a)
BnO
BnO
N
N
N
28
23
OBn
OBn
N
BnO
BnO
R2
N
c) or d)
e)
BnO
BnO
BnO
BnO
N
BnO
N
N
N
R1
OH
22
O
39
37 R1 = OCONi-Pr2, R2 = H
38 R1 = H, R2 = OCONi-Pr2
f), g)
or h)
BnO
R2
BnO
HO
HO
HO
R2
NH2
N
i)
BnO
BnO
BnO
BnO
N
N
N
N
N
R1
R1
22 R1 = OH, R2 = H
23 R1 = H, R2 = OH
40 R1 = N3, R2 = H
41 R1 = H, R2 = N3
8
TBDMS = t-BuMe2Si, TIPS = i-Pr3Si
Scheme 3. Reagents and conditions: (a) n-Bu4NF, THF, 23 ꢁC; 91% of 22; 77% of 23. (b) As (a), but 0 ꢁC; 94%. (c) i-Pr2NCOCl, DMAP, pyridine,
120 ꢁC; 41% of 37 and 36% of 38. (d) i-Pr2NCOCl, NaH, THF, 0 ! 23 ꢁC; 96% of 37. (e) Dess–Martin periodinane, CH2Cl2, 0 ꢁC; 72%. (f) Diphenyl
phosphorazidate (DPPA), DBU, toluene, 23 ꢁC; 65% of 40 and 15% of 41 from 22; 62% of 40 and 11% of 41 from 22/23 55:45; 66% of 40/41 85:15
from 23. (g) Di-p-nitrophenyl phosphorazidate, DBU, toluene, 23 ꢁC; 53% of 40 and 11% of 41 from 22. (h) DPPA, NaH, THF, 0 ! 23 ꢁC; 38% of
40/41 85:15 from 22. (i) H2, Pd/C, AcOH; 65% from 41.

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
453
gluconolactams 20, 24 and 29 with Lawessonꢂs reagent30
at 23 ꢁC provided the thionolactams 21 (73%), 26 (91%)
and 30 (96%). According to TLC analysis, thionation at
70–90 ꢁC led to epimerization at C(2), in contrast to
what was observed by Panday et al.1 Condensation of
the triethylsilylated thionolactam 21 with aminoacetal-
dehyde dimethyl acetal followed by acid-promoted cycli-
zation in the presence of small amounts of H2O1,31 was
accompanied by desilylation and yielded 40% of a 60:40
mixture of the gluco-and manno-hydroxy-imidazoles 22
and 23.k Under the same conditions, the (tert-butyl)di-
methylsilyloxy-thionolactam 26 provided 83% of the
desilylated gluco-imidazole 22 besides a mixture of the
silylated gluco-and manno-imidazoles 27 and 28 (8%
and 1%, respectively) and the hydroxy-lactam 11 (7%).
Lowering the temperature of the cyclization to 55 ꢁC in-
creased the yield of the protected imidazole 27 to 45%
and that of 28 to 2%. However, the cyclization did not
go to completion, and an intricate mixture of the hydr-
oxy-imidazole 22 and the intermediary amidine was also
isolated (ca. 37%; 9:1). No epimerization or deprotec-
tion was observed during the condensation of the triiso-
propylsilylated thionolactam 30. The same conditions as
those used for the condensation of 21 transformed 30
cleanly into the gluco-imidazole 31 (88%).
22 with N,N-diisopropylcarbamoyl chloride (DMAP,
pyridine, 120 ꢁC35) provided a 55:45 mixture of carba-
mates, which were separated by chromatography to
yield 41% of 37 and 36% of the manno-isomer 38.
Replacing DMAP/pyridine by NaH/THF36 and lower-
ing the temperature from 120 to 0–23 ꢁC led exclusively
to the gluco-isomer 37 (96%). Unfortunately, neither of
the carbamates 37 and 38 reacted with NaN3 under
the reported (or under slightly harsher) conditions,
and starting material was recovered.
Attempts to introduce the C(8) amino group by reduc-
tive amination of the ketone 39 failed. The ketone was
obtained in 72% yield by oxidation of 22 with Dess–
Martinꢂs periodinane,37–40 but did not react with BnNH2
in the presence of MgSO4 at 23 ꢁC and decomposed at
70 ꢁC. Decomposition was also observed upon at-
tempted oximation of 39 (NH2OHÆHCl, pyridine,
MeOH). Reduction of 39 with NaBH4 or L-Select-
ride41,42 led to the gluco-and manno-configured hydro-
xy-imidazoles 22 and 23 in ratios between 3:1 and 5:1
1
(according to H NMR spectra of the crude) as the re-
sult of a preferred pseudoaxial hydride addition. Finally,
a 80:20 mixture of the gluco-and manno-azido-imidaz-
oles 40 and 41 was obtained by treating the gluco-alco-
hol 22 with diphenyl phosphorazidate (DPPA) and
DBU in toluene.43 Chromatographic separation afford-
ed 65% of 40 and 15% of 41. Similar yields and ratios
of 40/41 resulted from the DPPA azidation of the
manno-alcohol 23 and of a mixture 22/23 (55:45).**
The azides 40 (53%) and 41 (11%) were also obtained
by replacing DPPA with di-p-nitrophenyl phosphorazi-
date,45 which was prepared in 62% yield46 (see Experi-
mental). A considerably lower yield of 40/41 (38%;
85:15) resulted from treating 22 with DPPA and NaH.
Hydrogenation of the manno-azido-imidazole 41 re-
sulted in 65% of the desired amine 8.
Protection of 11 by a pivaloyl or (methoxy)methyl group
proved less attractive. Treatment of 11 with pivaloyl
chloride in pyridine led consistently to a mixture of
the pivaloate 32 and the imide 33. Yields depended on
the reaction temperature; higher temperatures favouring
33 (32: 17–68% and 33: 7–76%). The pivaloate 32 re-
acted slowly with Lawessonꢂs reagent to provide, after
168 h, the thionolactam 34 (58%); 37% of 32 were recov-
ered. Methoxymethylation of 11 (dimethoxymethane/
P2O5) provided a mixture of the O-and N,O-protected
lactams 35 (46–62%) and 36 (9–23%), respectively.
Thionation of 35 gave only traces of the corresponding
thionolactam and was accompanied by extensive epi-
merization at C(2) as indicated by TLC analysis. We
did not pursue these routes to 22 and 23.
3
The amino-mannonolactam 10 in D2O adopts the H4-
conformation, as evidenced by J(2,3), J(3,4) and J(4,5)
of 3.7, 3.7 and 4.0 Hz, respectively (Table 9). A down-
field shift of ca. 0.7 ppm for H–C(2) of 10 was observed
upon protonation by CF3CO2H. Unfortunately, signal
overlap prevented an unambiguous assignment of the
conformation of 10ÆH+. The coupling constants of the
amino-imidazole 8 agree best with a ca. 1:1 mixture
The synthesis of 8 was continued (Scheme 3) by stan-
dard desilylation32,33 of the gluco-imidazopyridines 27
and 31, to provide 22 in 91% and 94%, respectively.
Similarly, desilylation of the analogue 28 gave the
manno-configured hydroxy-imidazole 23 (77%). The tri-
isopropylsilyl group proved most advantageous for the
synthesis of the gluco-alcohol 22 or its manno-isomer
23. It allowed a clean transformation of the gluconolac-
tam 11 into 22 in five steps and 77% overall yield.
of the 7H6-and
E7-conformersꢀꢀ (modelling with
MM3*6). Protonation of 8 led to a downfield shift for
H–C(8) (Dd = 0.96 ppm) and to slightly increased vicinal
constants, indicating a slightly larger contribution of the
7H6 conformer, excluding a H-bond between the ammo-
nium group and the imidazole ring. Lowering the pH
value further did not significantly affect the chemical
shift and J(H,H) values.
We originally intended to prepare the gluco-and manno-
azido-imidazoles 40 and 41 via the carbamates 37 and
38. The formation of azides by heating similar a-imidaz-
olyl-N,N-diisopropylcarbamates with NaN3 in the pres-
ence of BF3ÆOEt2 and trifluoroacetic acid has recently
been reported.34 Carbamoylation of the gluco-imidazole
** In agreement with earlier reports5,29,44 these results suggest that the
formal substitution of C(8)–OH proceeds via an azafulvenium
intermediate.
ꢀꢀ The direction of numbering of imidazopyridines (cf. 22 in Scheme 3)
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.
k Under otherwise identical conditions, the tetrabenzylated analogue of
21 led exclusively to the corresponding analogue of the gluco-
imidazopyridine 22.31

454
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
Table 1. Mannose-and mannosamine-derived lactams and imidazoles 7–10: pKHA values and comparison of their inhibition of snail b-mannosidase
and of the a-mannosidase from Jack beans to the inhibition of the b-glucosidases from Caldocellum saccharolyticum and from sweet almonds by the
glucose-and glucosamine-derived analogues 1, 2, 5 and 6 (Ki values in [lM])
OH
N
OH
NH
HO
HO
HO
HO
O
N
R
R
manno (R)
pH
7 (OH)
8 (NH2)
DDGdiss. [kcal/mol]
9 (OH)
10 (NH2)
DDGdiss. [kcal/mol]
pKHA
b-Mannosidase (snail)
5.7a
7.02b
138
—
7.46
4.5
5.5
4.5
0.115c
0.045
0.75c
+4.2
+3.8
+0.3
3.01d,e
5.12
43h
2000f
350f
+3.8
+2.6
+2.9
28
1.22g
a-Mannosidase (Jack beans)
gluco (R)
4850f
pH
1i (OH)
2i (NH2)
DDGdiss. [kcal/mol]
5i (OH)
6i (NH2)
DDGdiss. [kcal/mol]
pKHA
b-Glucosidase (C. saccharolyticum)
6.12a
0.41
6.33
241
77
—
7
7.04
21
4.6
5.4
6.8
4.6
5.4
6.8
+4.1
—
+0.7
+0.04
À1.3
+0.3
—
j
7.5
10
79
80
138
8
1.4
0.05
0.71
0.31
0.15
85
+4.8
+4.1
+4.0
+4.5
b-Glucosidase (sweet almonds)
593
200
213
129
j
8
À1.8
a Data taken from Ref. 47.
b No additional pKHA value was observed in the pH range 2.9–10.3.
c Data taken from Ref. 48.
d Mixed-type inhibition (a = 8.4).
e Ki = 9.0 lM (at 37 ꢁC and pH 4.0).9
f IC50/2.
g Mixed-type inhibition (a = 2.3).
h Ki = 68 lM (at 37 ꢁC and pH 4.5).9
i Data taken from Ref. 1 (IC50 values in [lM]).
j Not determined.
Similarly to the gluco-and manno-configured lactams 11
and 16,1 the protected gluco-lactams 12, 14, 19, 32 and
35 and gluco-thiolactam 34 (in solution) adopt the C1-
1H–C(1) d at 5.60 ppm with J(1,3) = 1.9 Hz (see num-
bering of 13 in Scheme 1), a strong IR OSO2 band at
1335 cmÀ1 and the [M+Na]+ peak at m/z 530.1631. For-
mation of the ketone 39 is confirmed by the disappear-
4
conformation, while the protected manno-lactams 15,
17, 18 and 25 form 2:1 mixtures of the 1C4-and 4C1-con-
formers (see Experimental, Tables 2, 3 and 5). A steri-
cally demanding C(2) substituent leads to an increased
population of the B2,5-conformation of the gluco-lac-
tams 20, 24 and 29. This conformation is dominating
in the gluco-thiolactams 21, 26 and 30, as evidenced by
their J(H,H) values. The large J(2,3) (7.2–9.0 Hz) and
rather small J(3,4) and J(4,5) (3.1–5.0 Hz) evidence the
2,5B-conformation of the N-substituted gluco-lactams
33 and 36. The 13C(2)–C(6) signals of the protected glu-
co-lactams 19 and 24 and of their manno-isomers 15 and
25 were assigned on the basis of HSQC-GRASP spectra;
those of the other lactams were assigned by analogy (see
Table 4 in Experimental). Similarly to their tetra-O-benz-
ylated analogues,31,47 the gluco-and manno-configured
imidazopyridines 22, 23, 27, 28, 38, 40 and 41 exist in
CDCl3 as 2:1 mixtures of 7H6-and 6H7-conformers,
while the gluco-imidazoles 31 and 37 adopt a conforma-
tion close to 5,8B and 6H7, respectively (see Tables 6 and
8 in Experimental). The 13C signals of C(5)–C(8) of all
imidazoles were assigned in analogy to Refs. 48–50
(see Table 7 in Experimental).
1
ance of the H–C(8) signal for 39, the replacement of
the 13C(8) d at 67.87 ppm for 22 by a 13C s at
181.93 ppm, and by a strong IR C@O band at
1696 cmÀ1
.
3. Enzymatic tests and discussion
The manno-configured hydroxy- and amino-imidazoles
7 and 8 and the corresponding lactams 9 and 10 were
tested as inhibitors of snail b-mannosidase (pH opti-
mum 4.0–4.53) at 25 ꢁC and pH 4.5 and 5.5 and of the
a-mannosidase from Jack beans (pH optimum 4.0–
5.051,52) at 37 ꢁC and pH 4.5, using the corresponding
4-nitrophenyl mannopyranosides as substrates. The
inhibition data are summarized in Table 1 and com-
pared to the inhibition of the b-glucosidases from C.
saccharolyticum and from almonds by the gluco-config-
ured hydroxy-and aminoi-midazoles 1 and 2 and the
corresponding lactams 5 and 6.1
Inspection of Table 1 shows a clear difference between
the effect of substitution of the hydroxy by the amino
group for the manno-imidazoles 7 and 8 as compared
to the effect on the gluco-imidazoles 1 and 2 versus the
The structure of the sultone 13 was deduced on the basis
of a 13C(1) d at 92.35 ppm, a 13C(2) s at 151.94 ppm, a

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
455
effect of the analogous substitution of the manno-lactams
9 and 10 as compared to the gluco-lactams 5 and 6.ꢁꢁ The
amino-imidazole 8 (Ki = 138 lM at pH 4.5; competitive)
is a 1200 times weaker inhibitor of snail b-mannosidase
than the hydroxy-imidazole 7. A similar, slightly stron-
ger influence of the C(8) substituent is observed for the
inhibition of the two b-glucosidases by the gluco-config-
ured imidazoles 1 and 2. The difference for 7 and 8 is
slightly smaller at higher pH. The pKHA values of 8
and the pH of the enzymatic assay mean that the inhibi-
tor is bound to the enzyme as imidazolium/ammonium
salt 8ÆH+. The effect of the amino group on the inhibition
appears to be dominated, for both gluco and manno com-
pounds, by the (partial or complete) disruption of the
interaction with the catalytic acid. However, one expects
this effect to manifest itself significantly more strongly in
the manno series, as there should be no compensating ef-
fect from an improved interaction of the ammonium
group with the catalytic nucleophile. That this is not ob-
served points to a stronger compensation, in the manno
than in the gluco series, from the interaction of the cat.
nucleophile with the imidazolium (and/or the imidazole)
ring. This implies a different positioning of the catalytic
nucleophile, depending on whether the C(8) ammonium
group is on the opposite or same face of the tetrahydro-
pyridine ring. One has to assume that the interaction of
the catalytic nucleophile of the b-glucosidases with the
C(8) ammonium group of the inhibitor entails a position
of the catalytic nucleophile that is not favourable to a
simultaneous interaction with the imidazolium ring; no
such detraction of the cat. nucleophile form its optimal
position for such an interaction is expected for the b-
mannosidase.§§
protonation of 10 reduces its inhibitory activity is in
keeping with the weakening effect of the ammonium
group (a strong r-acceptor) on the interaction of the lac-
tam moiety with the catalytic acid; this effect is not (par-
tially) compensated by a stronger interaction with the
catalytic nucleophile, requiring a significant proton
transfer to the lactam moiety of 10, a much weaker pro-
ton acceptor than the imidazole ring of 8. While the cata-
lytic nucleophile can strongly interact with the
imidazolium ring of 8,7,8 partially compensating for the
impaired interaction with the catalytic acid, it cannot
likewise interact significantly more strongly with the
essentially neutral lactam moiety of 10 than of 9. These
results are thus in keeping with the assumption that the
catalytic nucleophile of the retaining snail b-mannosi-
dase does not interact with the C(2) hydroxy group.
The lactams 9 and 10 and the amino-imidazopyridine 8
were also tested as inhibitors of the retaining a-mannosi-
dase from Jack beans (family 38;53,54 Table 1). Substitu-
tion of the C(2)–OH group of 9 (Ki = 43 lM; compare
Ref. 9) by the amino group, as in 10, led to a 110 times
weaker inhibition of the a-mannosidase. The imidazole
8 is, however, only a 1.6 times weaker (mixed-type)
inhibitor (Ki = 1.22 lM; a = 2.3) of this enzyme than
the hydroxy analogue 7 (Ki = 0.75 lM; competitive).
That 8 is a better inhibitor of the a-than of the b-man-
nosidase may be due to the conformational change of
the inhibitor at the pH of the assay and/or to a particu-
larly favourable interaction of the cat. nucleophile with
the imidazolium ring, or with the ammonium group. A
stronger interaction with the catalytic nucleophile is,
however, not in agreement with the observation that
10 is a weaker inhibitor of the a-mannosidase than 9.
The situation for the lactams is quite different. A com-
parison of the b-mannosidase inhibition by the hydroxy-
and amino-mannonolactams 9 and 10 to the inhibition of
the b-glucosidases by the hydroxy- and amino-glucono-
lactams 5 and 6, respectively, shows that the introduction
of the C(2) amino group weakens the inhibition of the b-
mannosidase (at pH 4.5 and 5.5) 68–664 times, while the
inhibition of the b-glucosidases (at similar pH values) is
only weakened 1.1–3.0 times (it is increased at pH 6.8).
The hydroxy-imidazole 7 is a much stronger inhibitor
than the hydroxy-lactam 9. Similarly as observed for
the amino-imidazole 8, the amino-lactam 10 is a six times
better inhibitor of the b-mannosidase at pH 5.5
(IC50 = 700 lM) than at pH 4.5 (IC50 = 4000 lM). The
hydroxy-lactam 9 is a slightly weaker inhibitor at
pH 5.5 (Ki = 5.12 lM) than at pH 4.5 (Ki = 3.01 lM;
compare Ref. 9) while the hydroxy-imidazole 7 is a
slightly better inhibitor at pH 5.5. That a more extensive
4. Experimental
4.1. General
Solvents were distilled: THF and toluene from Na and
benzophenone, CH2Cl2 from P2O5, DMF 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 ꢃmostainꢂ
(400 mL of 10% H2SO4 soln, 20 g of (NH4)6Mo7O24Æ
6H2O, 0.4 g of Ce(SO4)2. Flash chromatography (FC):
silica-gel Fluka 60 (0.04–0.063 mm). Mpꢂs uncorrected.
Optical rotations: 1-dm cell at 25 ꢁC, 589 nm. UV spec-
tra (ca. 0.2 mM solutions) were taken in 1-cm cell at
25 ꢁC in the range of 190–500 nm (loge values in paren-
thesis). FT-IR spectra: KBr or ca. 2% solution in
1
CHCl3, absorption in cmÀ1. H and 13C NMR spectra:
chemical shifts d in ppm rel. to TMS as external stan-
dard, and coupling constants J in hertz. FAB-MS: in
3-nitrobenzyl alcohol (NOBA) matrix. MALDI-MS
and HR-MALDI-MS: in gentisic acid (= 2,5-dihydroxy-
benzoic acid, DHB) matrix. The pKHA values were
determined in H2O by potentiometric titration with
HCl at 25 ꢁC. The b-mannosidase from snail acetone
powder (EC 3.2.1.25, as a suspension in 3.0 M
(NH4)2SO4 containing 10 mM AcONa, pH ꢀ 4.0, Sigma
M-9400), a-mannosidase from Jack beans (EC 3.2.1.24,
ꢁꢁ The different pH optima of the enzymes and the different pH of the
assay mean that only differences of differences can be interpreted,
that is, the relative inhibition by 7 and 8, as compared to the one by 1
and 2 can be compared to the relative inhibition by 9 and 10, as
compared to the one by 5 and 6.
§§ An alternative rationalization postulates that the interaction with the
catalytic acid of a b-mannosidase at a late stage of the reaction is less
important for a manno-configured inhibitor mimicking the putative
reactive intermediate than it is for the analogous interaction of the
two b-glucosidases with an analogous gluco-configured inhibitor.

456
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
as a suspension in 3.0 M (NH4)2SO4 and 0.1 mM zinc
acetate, pH 7.5, Sigma M-7257), p-nitrophenyl b-D-
mannopyranoside (Sigma N-1268) and p-nitrophenyl
a-D-mannopyranoside (Sigma N-2127) were used with-
out further purification.
40.68; H, 6.26; N, 7.91. Found: C, 40.42; H, 6.08; N,
7.81.
4.4. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-methane-
sulfonyl-D-glucono-1,5-lactam 12 and 11,3-anhydro-4,5,7-
tri-O-benzyl-1,2,6-trideoxy-2,6-imino-D-gluco-hept-1-eni-
tol-1-sulfonic acid 13
4.2. Di-(p-nitrophenyl) phosphorazidate46
Rf (AcOEt) 0.62. Mp 108–110 ꢁC (lit.:46 108–110 ꢁC). IR
(CHCl3): 3437w, 3117w, 3087w, 3027w, 2867w, 2458w,
2179s, 1915w, 1775w, 1671w, 1617m, 1591s, 1530s,
1490s, 1349s, 1305s, 1271m, 1189m, 1160s, 1111w,
1014w, 965s, 859s, 829w. 1H NMR (CDCl3,
300 MHz): 7.43–7.48 (m, H–C(2), H–C(6)); 8.28–8.34
(m, H–C(3), H–C(5)). 13C NMR (CDCl3, 75 MHz):
(a) At 0 ꢁC, a soln of 11 (250 mg, 0.559 mmol) in
CH2Cl2 (5 mL) was successively treated with Et3N
(0.15 mL, 1.08 mmol) and MsCl (55 lL, 0.708 mmol),
stirred at 0 ꢁC for 30 min and treated with satd NH4Cl
soln (1 mL). The mixture was diluted with Et2O
(15 mL) and washed with satd NH4Cl soln (3 ·
15 mL). The combined aq layers were extracted with
Et2O (2 · 15 mL). The combined org. layers were
washed with H2O (20 mL) and brine (20 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
2:1 ! 1:1) gave 13 (15 mg, 5%) and 12 (266 mg, 91%).
3
120.87 (dd, J(C,P) = 5.5, C(2), C(6)); 125.93 (d, C(3),
2
C(5)); 145.56 (s, C(4)); 153.62 (d, J(C,P) = 7.3, C(1)).
31P NMR (CDCl3, 121 MHz): À10.39. HR-ESI-MS:
388.0055 (100, [M+Na]+, C12H8N5NaO7P+; calcd
388.0059), 377.0148 (56), 369.2002 (28), 281.0055 (51).
Anal. Calcd for C12H8N5O7P (365.20): C, 39.47; H,
2.21; N, 19.18. Found: C, 39.51; H, 2.33; N, 19.07.
(b) At 0 ꢁC, a soln of 11 (50 mg, 0.112 mmol) in pyridine
(5 mL) was treated with MsCl (17 lL, 0.219 mmol), stir-
red at 0 ꢁC for 1 h and at 23 ꢁC for 19 h and treated with
satd NH4Cl soln (1 mL). Workup and FC as described
in (a) gave 13 (3 mg, 5%), 12 (43 mg, 73%) and 11
(9 mg, 18%).
4.3. 5-Amino-5-deoxy-D-mannono-1,5-lactam 955
FC (AcOEt/MeOH/H2O 5:2:1) gave 9 (171 mg, 86%) as
an amorphous solid, which was recrystallized from a
mixture of EtOH and H2O. Colourless crystals. Rf
(AcOEt/MeOH/H2O 5:2:1) 0.29. Mp 174–175 ꢁC (lit.:56
(c) At 0 ꢁC, a soln of 11 (25 mg, 55.9 lmol) in CH2Cl2
(1 mL) was successively treated with (i-Pr)2NH (32 lL,
0.226 mmol) and MsCl (10 lL, 0.129 mmol), stirred at
0 ꢁC for 30 min and treated with satd NH4Cl soln
(1 mL). Workup and FC as described in (a) gave 13
(14 mg, 50%) and 12 (14 mg, 48%).
164–167 ꢁC, lit.:9 169–171 ꢁC, lit.:57 169–170 ꢁC, lit.:58
25
165–169 ꢁC, lit.:59 170–172 ꢁC). ½aŠ ¼ þ0:2 (c 1.07,
D
H2O) [lit.:56 +1.2 (H2O), lit.:9 +1.0 (H2O), lit.:59 +2.0
(H2O), lit.:55 +1.6 (H2O), lit.:58 +0.9 (H2O), lit.:57 +1.6
(H2O)]. IR (0.5% in KBr): 3600–3000s (br), 3324s,
3201s, 2949m, 2938m, 2888m, 2830m, 2747w, 2663w,
1655s, 1461w, 1418m, 1385m, 1357m, 1335s, 1306w,
1267m, 1240m, 1215w, 1159w, 1131s, 1111m, 1089m,
1056s, 1030s, 1004m, 962w, 928w, 847m, 746m. 1H
NMR (D2O, 300 MHz): 3.22 (dt, J = 3.6, 6.0, H–C(5));
3.55 (dd, J = 6.0, 11.8, irrad. at 3.22 ! d, J = 12.4, H–
C(6)); 3.67 (dd, J = 3.6, 11.8, irrad. at 3.22 ! d,
J = 12.4, H0–C(6)); 3.72 (t, J = 6.0, irrad. at 3.22 ! d,
J = 6.0, irrad. at 3.89 ! d, J = 6.0, H–C(4)); 3.89 (dd,
J = 3.8, 6.0, irrad. at 4.19 ! d, J = 6.0, H–C(3)); 4.19
(d, J = 3.8, irrad. at 3.89 ! s, H–C(2)). 1H NMR
(DMSO-d6, 300 MHz): 3.11 (q, J = 5.0, H–C(5)); 3.40
(td, J = 5.6, 10.6, H–C(6)); 3.57 (td, J = 5.3, 10.9, H0–
C(6)); 3.61 (br q, J ꢀ 5.3, H–C(4)); 3.79 (dt, J = 3.7,
5.0, H–C(3)); 4.03 (br t, J ꢀ 3.7, H–C(2)); 4.67 (d,
J = 4.7, irrad. at 4.03 ! s, HO–C(2)); 4.78 (t, J = 5.6,
irrad. at 3.40 ! change, HO–C(6)); 4.98 (d, J = 3.7,
irrad. at 3.79 ! s, HO–C(3)); 5.24 (d, J = 4.7, irrad. at
3.61 ! s, HO–C(4)); 7.18 (br s, NH). 13C NMR (D2O,
75 MHz, assignment based on HSQC-GRASP spec-
trum): 57.63 (d, C(5)); 61.37 (t, C(6)); 67.40 (d, C(4));
68.48 (d, C(2)); 72.22 (d, C(3)); 173.53 (s, C(1)). 13C
NMR (DMSO-d6, 75 MHz): 59.07 (d, C(5)); 62.24 (t,
C(6)); 67.98 (d, C(4)); 68.47 (d, C(2)); 73.16 (d, C(3));
4.4.1. Data of 12. Colourless oil. Rf (hexane/AcOEt
25
D
1:1) 0.32. ½aŠ ¼ þ73:3 (c 1.12, CHCl3). IR (CHCl3):
3387w, 3089w, 3067w, 3032w, 3013w, 2914w, 2869w,
1953w, 1877w, 1810w, 1695s, 1603w, 1497w, 1454m,
1366s, 1331m, 1174m, 1109s, 1052m, 1026m, 971m,
1
910w. H NMR (CDCl3, 300 MHz): see Table 2; addi-
tionally, 3.31 (s, MsO); 3.48–3.58 (irrad. at
3.24 ! change); 3.64 (irrad. at 4.00 ! d, J ꢀ 9.0); 4.00
(irrad. at 3.64 ! d, J = 9.3, irrad. at 5.10 ! d, J =
9.3); 4.42 (d, J = 12.5, PhCH); 4.46 (d, J = 12.5, PhCH);
4.52 (d, J = 11.2, PhCH); 4.79 (d, J = 10.3, PhCH); 4.90
(d, J = 11.2, PhCH); 5.00 (d, J = 10.6, PhCH); 5.10 (ir-
rad. at 4.00 ! s); 6.38 (exchange with CD3OD); 7.19–
7.23 (m, 2 arom. H); 7.26–7.45 (m, 13 arom. H). 13C
NMR (CDCl3, 75 MHz): see Table 4; additionally,
39.79 (q, MsO); 73.24, 75.01, 75.38 (3t, 3PhCH2);
127.80–128.61 (several d); 136.99 (s); 137.13 (2s). HR-
MALDI-MS: 564.1447 (18, [M+K]+, C28H31KNO7S+;
calcd
564.1458),
548.1706
(100,
[M+Na]+,
C28H31NNaO7S+; calcd 548.1719), 526.1882 (11,
[M+H]+, C28H32NO7S+; calcd 526.1899), 452.1828 (8,
[M+NaÀMsOH]+, C27H27NNaO4+; calcd 452.1838),
362.1363 (82, [M+NaÀMsOÀBn]+, C20H21NNaO4
;
+
calcd 362.1368), 346.0953 (11). Anal. Calcd for
C28H31NO7S (525.62): C, 63.98; H, 5.94; N, 2.66; S,
6.10. Found: C, 63.83; H, 6.11; N, 2.48; S, 5.92.
171.37 (s, C(1)). HR-ESI-MS: 554.1720 (15,
[3M+Na]+,
C18H33N3NaO15
;
calcd
554.1809),
+
377.1134 (100, [2M+Na]+, C12H22N2NaO10+; calcd
377.1172), 200.0533 (8, [M+Na]+, C6H11NNaO5+; calcd
200.0535). Anal. Calcd for C6H11NO5 (177.16): C,
4.4.2. Data of 13. Colourless oil. Rf (hexane/AcOEt
1:1) 0.56. IR (CHCl3): 3391w, 3034w, 2927m, 2866m,

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
457
Table 2. Selected 1H NMR chemical shifts [ppm] and coupling constants [Hz] of the protected gluco-lactams 12, 14, 19, 20 and 24 and the gluco-
thiolactams 21 and 26 in CDCl3
12
14
1920
21
24
8.01a
26
NH
6.38
5.10
7.03
5.15
6.11
4.06
3.70
3.55
3.45–3.53
3.22
3.57
9.3
6.04
4.18
3.82
3.54
3.57–3.65
3.29
3.61
7.8
6.02
4.15
7.99b
4.719d
3.78
3.53d
3.99b
3.41
3.67
2.5
H–C(2)
H–C(3)
H–C(4)
H–C(5)
H–C(6)
H0–C(6)
J(2,3)
4.70c
3.80
3.54c
4.02a
3.41
3.66
2.5
4.00
3.64
4.03
3.80
3.81
3.58
3.48–3.58
3.24
3.50–3.58
3.35
3.53
3.52–3.64
3.28
3.48–3.58
9.3
9.3
3.52–3.64
8.1
7.8
10.0
J(3,4)
J(4,5)
9.3
8.1
9.0
9.0
7.8
8.1
2.8
9.3
2.8
9.3
9.3
7.5
7.8
7.8
J(5,6)
J(5,60)
J(6,60)
6.5
2.8
7.5
2.8
8.1
2.8
7.5
3.4
7.5
3.4
e
e
9.3
10.3
9.0
10.0
10.0
10.0
10.0
a J(5,NH) = 2.2 Hz.
b J(5,NH) = 1.9 Hz.
c J(2,4) = 1.2 Hz.
d J(2,4) = 0.9 Hz.
e Not determined.
Table 3. Selected 1H NMR chemical shifts [ppm] and coupling constants [Hz] of the protected gluco-lactams 29, 32, 33, 35 and 36 and the gluco-
thiolactams 30 and 34 in CDCl3
2930
32
8.06a
33
34
35
36
NH
5.87
4.34
3.85
3.53
3.80–3.92
3.34
3.63
5.3
6.05
5.28
—
8.11
5.60
6.01
4.19
—
H–C(2)
H–C(3)
H–C(4)
H–C(5)
H–C(6)
H0–C(6)
J(2,3)
4.91b
3.91
3.55b
4.14a
3.45
3.69
2.8
5.56
4.35
3.86
3.96
3.79–3.83
3.52
3.63
9.0
4.07
3.95c
3.98
4.69c
3.57
3.63
7.2
3.96
3.90
3.56–3.68
3.56–3.68
3.22–3.33
3.56–3.68
9.3
3.56–3.70
3.56–3.70
3.29
3.50–3.62
3.50–3.62
3.20–3.30
3.50–3.62
8.7
3.56–3.70
8.1
8.1
J(3,4)
J(4,5)
5.0
9.0
1.6
9.3
9.0
d
4.1
4.1
8.7
d
5.0
3.1
d
d
d
d
d
d
d
J(5,6)
J(5,60)
J(6,60)
8.1
3.1
7.5
3.1
5.3
5.9
7.5
d
4.7
5.9
9.7
10.0
10.0
9.3
9.7
a J(5,NH) = 2.2 Hz.
b J(2,4) = 1.3 Hz.
c J(3,5) = 1.3 Hz.
d Not determined.
1951w, 1886w, 1806w, 1733w, 1648s, 1497w, 1455m,
1367m, 1335s, 1262m, 1159m, 1133s, 1098s, 1054m,
1015m, 911w. 1H NMR (CDCl3, 300 MHz): 3.24 (t,
J = 9.0, irrad. at 3.63 ! d, J ꢀ 7.8, H–C(7)); 3.38–3.46
(m, irrad. at 3.63 ! br t, J ꢀ 9.0, H–C(6)); 3.50 (t,
J = 8.4, irrad. at 3.89 ! d, J = 8.7, H–C(5)); 3.63 (dd,
J = 2.8, 9.0, irrad. at 3.24 ! d, J ꢀ 2.5, H0–C(7)); 3.89
(dd, J = 8.4, 9.3, irrad. at 3.50 ! d, J = 9.3, irrad. at
4.98 ! d, J = 8.1, H–C(4)); 4.46 (br s, PhCH2); 4.51
(d, J = 11.2, PhCH); 4.74 (d, J = 10.9, PhCH); 4.85 (br
s, exchange with CD3OD, NH); 4.89 (d, J = 11.2,
PhCH); 4.97 (d, J = 11.2, PhCH); 4.98 (dd, J = 1.9,
9.3, irrad. at 3.89 ! d, J ꢀ 1.3, irrad. at 5.60 ! d,
J = 9.7, addition of CD3OD ! d, J = 9.7, H–C(3));
5.60 (d, J = 1.9, irrad. at 4.98 ! s, exchange with
CD3OD, H–C(1)); 7.17–7.20 (m, 2 arom. H); 7.28–7.41
(m, 13 arom. H). 13C NMR (CDCl3, 75 MHz): 57.78
(d, C(6)); 70.46 (t, C(7)); 73.57, 74.51, 75.22 (3t,
3PhCH2); 76.35 (d, C(5)); 80.64, 81.35 (2d, C(3), C(4));
92.35 (d, C(1)); 128.09–128.72 (several d); 137.21 (2s);
137.31 (s); 151.94 (s, C(2)). HR-MALDI-MS: 530.1631
(49, [M+Na]+, C28H29NNaO6S+; calcd 530.1613),
450.2055 (100, [M+NaÀSO3]+, C28H29NNaO3+; calcd
450.2045).
4.5. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-(trifluoro-
methanesulfonyl)-D-glucono-1,5-lactam 14
At À78 ꢁC, a soln of 11 (96 mg, 0.215 mmol) in CH2Cl2
(2 mL) was treated successively with pyridine (36 lL,
0.446 mmol) and Tf2O (48 lL, 0.291 mmol), stirred for
3 h at À78 to À10 ꢁC and treated with satd NH4Cl soln
(5 mL). The mixture was diluted with Et2O (20 mL),
washed with satd NH4Cl soln (15 mL) and brine
(15 mL), dried (MgSO4) and evaporated. FC (hexane/
AcOEt 2:1) gave 14 (91 mg, 73%). Colourless oil.

458
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
Table 4. Selected 13C NMR chemical shifts [ppm] of the protected
gluco-lactams and thiolactams 12, 14, 19, 20, 21, 24, 26, 29, 30 and 32–
36 and of the protected manno-lactams 15, 17, 18 and 25 in CDCl3
Table 5. Selected 1H NMR chemical shifts [ppm] and coupling
constants [Hz] of the protected manno-lactams 15, 17, 18 and 25 in
CDCl3
Compound C(1)
C(2)
C(3)
C(4)
C(5)
C(6)
15
17
18
25
gluco
NH
6.12a
4.12
3.98
3.59
3.64a
3.42
3.48
3.4
6.18
5.39
6.81
5.44
5.97
4.450
3.90
3.68
3.53–3.62
3.40
3.53
2.8
12
14
19a
20
21
24a
26
29
30
32
33
34
35
36
166.18 79.10
164.19 79.28
167.22 63.78
170.59 73.40
201.61 79.02
170.58 73.14
201.44 79.06
170.60 72.93
201.18 79.31
166.63 72.39
168.58 72.90
80.26
82.15
81.22
83.48
81.55
83.22
81.25
82.57
80.67
80.72
81.11
75.89
76.44
76.53
77.20
77.03
77.43
77.05
78.64
77.05
76.77
74.43
54.27 69.44
54.89 68.90
54.15 69.82
53.89 70.04
55.35 68.43
53.97 70.04
55.40 68.50
53.15 69.54
55.11 68.31
54.08 69.98
57.52 68.36
H–C(2)
H–C(3)
H–C(4)
H–C(5)
H–C(6)
H0–C(6)
J(2,3)
4.15
3.57
4.08
3.68–3.78
3.68–3.78
3.46–3.56
3.46–3.56
3.1
3.62–3.70
3.44–3.49
3.44–3.49
3.3
J(3,4)
J(4,5)
4.7
4.4
4.1
4.1
3.4
b
5.0
5.0
b
b
b
b
J(5,6)
J(5,60)
J(6,60)
8.7
4.4
8.7
3.7
b
b
197.72 75.94b 80.11
76.44b 58.08 68.87
8.7
8.7
169.89 74.63
170.25 74.20
81.90
82.44
77.11
76.86
54.14 70.13
57.85 68.61
a J(5,NH) ꢀ 1.3 Hz.
b Not determined.
manno
15a
17
166.77 59.57
77.59
72.80
55.51 70.96
55.77 70.94
55.81 70.43
55.35 71.40
165.63 76.82b 76.20b 73.43
18
25a
163.34 80.34
170.00 70.60
75.68
79.11
72.11
74.63
with Et2O (2 · 20 mL). The combined org. layers were
washed with H2O (50 mL) and brine (50 mL), dried
(MgSO4) and evaporated. FC (hexane/AcOEt 2:1) gave
19 (3 mg, 2%) and 15 (94 mg, 77%).
a Assignments based on a HSQC-GRASP spectrum.
b Assignment may be interchanged.
(b) A suspension of 14 (75 mg, 0.129 mmol) and NaN3
(17 mg, 0.261 mmol) in DMF (1 mL) was stirred for
3 h at 23 ꢁC. Workup as described in (a) and FC (hex-
ane/AcOEt 2:1 ! 1:1 ! 1:3) gave 15 (32 mg, 52%) and
161 (9 mg, 16%).
25
D
Rf (hexane/AcOEt 2:1) 0.21. ½aŠ ¼ þ56:9 (c 2.01,
CHCl3). IR (CHCl3): 3385w, 3234w (br), 3089w,
3068w, 3033w, 3013w, 2914w, 2869w, 1951w, 1877w,
1810w, 1704s, 1603w, 1497w, 1454m, 1419s, 1363m,
1317w, 1285w, 1230m, 1170m, 1141s, 1111s, 1028m,
1
4.6.1. Data of 15. Colourless oil. Rf (hexane/AcOEt
1004m, 935w, 910w. H NMR (CDCl3, 300 MHz): see
Table
25
D
2:1) 0.11. ½aŠ ¼ À16:5 (c 1.00, CHCl3). UV (CHCl3):
2;
additionally,
3.50–3.58
(irrad.
at
259 (2.89). IR (CHCl3): 3389m, 3089w, 3061w, 2920m,
2868m, 2116s, 1953w, 1877w, 1811w, 1681s, 1604w,
1496m, 1454m, 1393w, 1362m, 1311m, 1075s, 1028m,
3.80 ! change); 3.80 (irrad. at 4.03 ! d, J = 8.1); 4.03
(irrad. at 3.80 ! d, J ꢀ 10.0, irrad. at 5.15 ! d,
J = 9.3); 4.43 (d, J = 11.8, PhCH); 4.51 (d, J = 11.8,
PhCH); 4.53 (d, J = 11.2, PhCH); 4.84 (d, J = 10.6,
PhCH); 4.873 (d, J = 11.8, PhCH); 4.875 (d, J = 10.6,
PhCH); 5.15 (irrad. at 4.03 ! s); 7.16–7.21 (m, 2 arom.
H); 7.27–7.40 (m, 13 arom. H). 13C NMR (CDCl3,
75 MHz): see Table 4; additionally, 73.36, 75.09, 75.46
(3t, 3PhCH2); 118.38 (q, 1J(C,F) = 319.2, CF3);
127.82–128.40 (several d); 136.65, 136.80, 136.98 (3s).
1
911w. H NMR (CDCl3, 300 MHz): see Table 5; addi-
tionally, 4.42 (d, J = 12.1, PhCH); 4.46 (d, J = 12.5,
PhCH); 4.49 (d, J = 11.8, PhCH); 4.55 (d, J = 11.5,
PhCH); 4.57 (d, J = 11.8, PhCH); 4.70 (d, J = 11.8,
PhCH); 6.12 (exchange with CD3OD); 7.19–7.22 (m, 2
arom. H); 7.25–7.39 (m, 13 arom. H). 13C NMR
(CDCl3, 75 MHz): see Table 4; additionally, 72.40,
73.15, 73.33 (3t, 3PhCH2); 127.81–128.57 (several d);
19F NMR (CDCl3, 282 MHz): À74.05. HR-MALDI-
+
MS: 899.3843 (8, [2M+NaÀTf2O]+, C54H56N2NaO9
;
136.94, 136.96, 137.25 (3s). HR-MALDI-MS: 495.1997
; calcd 495.2008),
(100, [M+Na]+, C27H28N4NaO4
+
calcd 899.3884), 877.3989 (43, [2M+HÀTf2O]+,
+
469.2101 (16, [M+2H+NaÀN2]+, C27H30N2NaO4
;
+
C54H57N2O9
;
calcd 877.4064), 859.3908 (28,
calcd 469.2103), 467.1938 (39, [M+NaÀN2]+,
[2MÀHÀ2TfO]+, C54H55N2O8
;
769.3438 (10), 620.2632 (11), 470.1936 (14,
calcd 859.3958),
+
+
C27H28N2NaO4
;
calcd 467.1947), 447.2283 (34,
447.2284),
[M+3HÀN2]+,
C27H31N2O4
;
calcd
+
[M+H+NaÀTf]+, C27H29NNaO5
;
calcd 470.1943),
;
+
+
445.2121 (85, [M+HÀN2]+, C27H29N2O4
;
calcd
448.2110 (36, [M+2HÀTf]+, C27H30NO5
calcd
+
445.2127). Anal. Calcd for C27H28N4O4 (472.54): C,
68.63; H, 5.97; N, 11.86. Found: C, 68.82; H, 6.08; N,
11.70.
448.2124), 402.2056 (100).
4.6. 5-Amino-2-azido-3,4,6-tri-O-benzyl-2,5-dideoxy-D-
mannono-1,5-lactam 15
4.7. 2,5-Diamino-2,5-dideoxy-D-mannono-1,5-lactam 10
(a) A suspension of 12 (135 mg, 0.257 mmol) and NaN3
(165 mg, 2.54 mmol) in DMF (3.2 mL) was stirred for
2 h at 70 ꢁC, cooled to 22 ꢁC, diluted with Et2O
(40 mL) and washed with satd NH4Cl soln
(3 · 20 mL). The combined aq layers were extracted
A soln of 15 (100 mg, 0.212 mmol) in MeOH (4 mL) was
treated with AcOH (0.8 mL) and 10% Pd/C (90 mg),
hydrogenated at 6 bar for 22 h and filtered through Cel-
ite. Evaporation of the combined filtrate and washing

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
459
(25 mL of MeOH), co-evaporation with toluene
4.9. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-(trifluoro-
methanesulfonyl)-D-mannono-1,5-lactam 18
(4 · 5 mL), ion-exchange chromatography (Amberlite
CG-120, NH4 form, elution with 0.05 M aq NH3)
and lyophilization gave 10/6 9:1 (34.1 mg, 92%).
+
At À78 ꢁC, a soln of 16 (105 mg, 0.235 mmol) in CH2Cl2
(2 mL) was treated successively with pyridine (36 lL,
0.446 mmol) and Tf2O (48 lL, 0.291 mmol), stirred for
3 h at À78 to À10 ꢁC and treated with satd NH4Cl soln
(5 mL). The mixture was diluted with Et2O (20 mL),
washed with satd NH4Cl soln (15 mL) and brine
(15 mL), dried (MgSO4), filtered and evaporated. FC
25
Rf (CHCl3/MeOH/NH4OH 5:4:1) 0.19. ½aŠ ¼ þ9:7 (c
D
1.04, H2O). pKHA = 7.46. IR (0.5% in KBr): 3600–
2600s (br), 2925m, 1659s, 1588s, 1459m, 1416m,
1374m, 1056m, 895w. H NMR (D2O, 300 MHz, only
1
signals of 10 listed): see Table 9; additionally, 3.64 (ir-
rad. at 3.44 ! d, J = 12.5); 3.72 (irrad. at 3.44 ! d,
J = 12.1); 3.87 (irrad. at 3.44 ! d, J = 4.4); 3.98 (irrad.
at 3.58 ! d, J = 4.0). 1H NMR (D2O, 300 MHz, 1 equiv
of CF3CO2H, only signals of 10ÆCF3CO2H listed):
see Table 9. 13C NMR (D2O, 75 MHz, only signals
of 10 listed): 51.49 (d, C(2)); 58.81 (d, C(5)); 61.73 (t,
C(6)); 69.37 (d, C(4)); 71.66 (d, C(3)); 175.11 (s, C(1)).
(hexane/AcOEt 3:2) gave 18 (81 mg, 60%). Colourless
25
D
oil. Rf (hexane/AcOEt 3:2) 0.14. ½aŠ ¼ þ6:4 (c 2.02,
CHCl3). IR (CHCl3): 3389w, 3241w (br), 3089w,
3067w, 3032m, 3013w, 2925w, 2870w, 1953w, 1877w,
1810w, 1702s, 1602w, 1496w, 1454m, 1420s, 1362m,
1287m, 1236s, 1169m, 1143s, 1089s, 1027s, 908w. 1H
NMR (CDCl3, 300 MHz): see Table 5; additionally,
4.43 (d, J = 11.8, PhCH); 4.48 (d, J = 12.1, PhCH);
4.51 (br s, PhCH2); 4.52 (d, J = 11.2, PhCH); 4.62 (d,
J = 12.1, PhCH); 7.19–7.24 (m, 4 arom. H); 7.27–7.40
(m, 11 arom. H). 13C NMR (CDCl3, 75 MHz): see Table
4; additionally, 72.19 (t, PhCH2); 73.30 (t, 2PhCH2);
118.37 (q, 1J(C,F) = 318.6, CF3); 127.73–128.51 (several
d); 136.21, 136.40, 137.15 (3s). 19F NMR (CDCl3,
282 MHz): À74.64. HR-MALDI-MS: 899.3867 (2,
[2M+NaÀTf2O]+, C54H56N2NaO9+; calcd 899.3884),
HR-ESI-MS: 511.1296 (17), 375.1507 (100, [2M+Na]+,
+
C12H24N4NaO8
;
calcd 375.1492), 353.1687 (23,
[2M+H]+, C12H25N4O8+; calcd 353.1672), 273.1817
(41), 199.0694 (4, [M+Na]+, C6H12N2NaO4+; calcd
199.0695), 177.0869 (8, [M+H]+, C6H13N2O4+; calcd
177.0875).
4.8. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-methane-
sulfonyl-D-mannono-1,5-lactam 17
At 0 ꢁC, a soln of 16 (270 mg, 0.603 mmol) in CH2Cl2
(5 mL) was successively treated with Et3N (0.17 mL,
1.22 mmol) and MsCl (60 lL, 0.772 mmol), stirred at
0 ꢁC for 30 min and treated with satd NH4Cl soln
(1 mL). The mixture was diluted with Et2O (25 mL)
and washed with satd NH4Cl soln (3 · 15 mL). The
combined aq layers were extracted with Et2O
(2 · 15 mL). The combined org. layers were washed with
H2O (20 mL) and brine (20 mL), dried (MgSO4), filtered
and evaporated. FC (hexane/AcOEt 2:1 ! 1:1) gave 17
877.4002 (31), 859.3915 (36, [2MÀHÀ2TfO]+,
+
C54H55N2O8
;
calcd 859.3958), 751.3357 (100,
calcd
+
[2MÀHÀBnOHÀ2TfO]+,
C47H47N2O7
;
751.3383), 620.2634 (15), 602.1423 (5, [M+Na]+,
C28H28F3NNaO7S+; calcd 602.1436), 580.1610 (2,
[M+H]+, C28H29F3NO7S+; calcd 580.1617), 470.1936
+
(13,
[M+H+NaÀTf]+,
C27H29NNaO5
;
calcd
;
470.1943), 448.2112 (20, [M+2HÀTf]+, C27H30NO5
+
calcd 448.2124), 402.2055 (57).
(288 mg, 91%). Colourless oil. Rf (hexane/AcOEt 1:1)
4.10. 5-Amino-2-azido-3,4,6-tri-O-benzyl-2,5-dideoxy-D-
glucono-1,5-lactam 19
25
0.24. ½aŠ ¼ À2:4 (c 1.32, CHCl3). IR (CHCl3):
D
3392w, 3089w, 3067w, 3032m, 3013w, 2924w, 2868w,
1953w, 1877w, 1810w, 1693s, 1603w, 1496w, 1455m,
1361s, 1333m, 1268w, 1174m, 1087s, 1048m, 1028m,
(a) A suspension of 17 (100 mg, 0.190 mmol) and NaN3
(124 mg, 1.91 mmol) in DMF (2.4 mL) was stirred for
2 h at 70 ꢁC, cooled to 23 ꢁC, diluted with Et2O
(40 mL) and washed with satd NH4Cl soln
(3 · 20 mL). The combined aq layers were extracted
with Et2O (2 · 20 mL). The combined org. layers were
washed with H2O (50 mL) and brine (50 mL), dried
(MgSO4) and filtered. Evaporation and FC (hexane/
AcOEt 2:1) gave 19 (56 mg, 62%) and 15 (5 mg, 6%).
1
971s, 927w. H NMR (CDCl3, 300 MHz): see Table 5;
additionally, 3.30 (s, MsO); 3.44–3.49 (irrad. at
3.66 ! change); 3.57 (irrad. at 3.66 ! d, J ꢀ 2.9, irrad.
at 4.15 ! d, J = 3.3); 4.15 (irrad. at 3.57 ! d, J = 3.3,
irrad. at 5.39 ! d, J = 4.4); 4.41 (d, J = 11.8, PhCH);
4.42 (d, J = 11.8, PhCH); 4.48 (d, J = 11.8,
PhCH); 4.54 (d, J = 11.8, PhCH); 4.55 (d, J = 11.8,
PhCH); 4.80 (d, J = 12.1, PhCH); 5.39 (irrad. at
4.15 ! s); 6.18 (exchange with CD3OD); 7.18–7.22 (m,
2 arom. H); 7.25–7.39 (m, 13 arom. H). 13C NMR
(CDCl3, 75 MHz): see Table 4; additionally, 39.35 (q,
MsO); 72.17, 73.29, 73.68 (3t, 3PhCH2); 127.78–128.56
(several d); 136.65, 137.11, 137.20 (3s). HR-MALDI-
MS: 564.1446 (7, [M+K]+, C28H31KNO7S+; calcd
564.1458), 548.1705 (100, [M+Na]+, C28H31NNaO7S+;
calcd 548.1719), 526.1879 (7, [M+H]+, C28H32NO7S+;
(b) A suspension of 18 (65 mg, 0.112 mmol) and NaN3
(15 mg, 0.231 mmol) in DMF (1 mL) was stirred for
3 h at 23 ꢁC. Workup (as described in a) and FC (hex-
ane/AcOEt 2:1 ! 1:1 ! 1:3) gave 19 (14 mg, 26%) and
111 (13 mg, 26%).
4.10.1. Data of 19. Colourless solid. Rf (hexane/AcOEt
25
D
2:1) 0.18. Mp 90–91 ꢁC. ½aŠ ¼ þ105:1 (c 0.79, CHCl3).
calcd 526.1899), 452.1828 (17, [M+NaÀMsOH]+,
UV (CHCl3): 259 (2.94). IR (CHCl3): 3385w, 3090w,
3061w, 2960w, 2869w, 2114s, 1952w, 1877w, 1810w,
1683s, 1605w, 1497w, 1454m, 1398w, 1362m, 1313m,
+
C27H27NNaO4
;
calcd 452.1838), 362.1359 (4,
[M+NaÀMsOÀBn]+, C20H21NNaO4+; calcd 362.1368),
346.0955 (11). Anal. Calcd for C28H31NO7S (525.62):
C, 63.98; H, 5.94; N, 2.66; S, 6.10. Found: C, 63.90;
H, 5.98; N, 2.55; S, 5.87.
1
1148m, 1110s, 1051m, 1028m, 912w. H NMR (CDCl3,
300 MHz): see Table 2; additionally, 3.70 (irrad. at
4.06 ! d, J = 9.0); 4.06 (irrad. at 3.70 ! s); 4.42 (d,

460
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
J = 12.5, PhCH); 4.46 (d, J = 12.1, PhCH); 4.53
(d, J = 11.2, PhCH); 4.83 (d, J = 10.9, PhCH); 4.89 (d,
J = 10.9, PhCH); 4.90 (d, J = 10.6, PhCH); 6.11
(exchange with CD3OD); 7.19–7.22 (m, 2 arom. H);
7.25–7.28 (m, 3 arom. H); 7.30–7.40 (m, 10 arom. H).
13C NMR (CDCl3, 75 MHz): see Table 4; additionally,
73.37, 75.08, 75.32 (3t, 3PhCH2); 127.77–128.46 (several
C33H43NO5Si (561.79): C, 70.55; H, 7.71; N, 2.49.
Found: C, 70.38; H, 7.69; N, 2.59.
4.13. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-(triethyl-
silyl)-D-glucono-1,5-thiolactam 21
A soln of 20 (40 mg, 71.2 lmol) in toluene (1.3 mL) was
treated with Lawessonꢂs reagent (22 mg, 54.4 lmol), stir-
red at 23 ꢁC for 24 h, diluted with Et2O (30 mL) and
washed with satd NaHCO3 soln (3 · 20 mL). The com-
bined aq layers were extracted with Et2O (2 · 20 mL).
The combined org. layers were washed with H2O
(50 mL) and brine (50 mL), dried (MgSO4), filtered
and evaporated. FC (hexane/AcOEt 6:1) gave 21
d); 136.93, 137.09, 137.31 (3s). HR-MALDI-MS:
;
+
495.1996 (100, [M+Na]+, C27H28N4NaO4
calcd
495.2008),
C27H30N2NaO4
469.2100
;
(23,
calcd 469.2103), 467.1934 (64,
[M+2H+NaÀN2]+,
+
[M+NaÀN2]+, C27H28N2NaO4
;
calcd 467.1947),
; calcd
+
+
447.2275 (51, [M+3HÀN2]+, C27H31N2O4
447.2284), 445.2116 (55, [M+HÀN2]+, C27H29N2O4
;
+
calcd 445.2127). Anal. Calcd for C27H28N4O4 (472.54):
C, 68.63; H, 5.97; N, 11.86. Found: C, 68.55; H, 6.09;
N, 11.71.
(30 mg, 73%). Colourless oil. Rf (hexane/AcOEt 6:1)
25
D
0.39. ½aŠ ¼ þ78:0 (c 1.00, CHCl3). IR (CHCl3):
3376w, 3089w, 3068w, 3020s, 2957m, 2936m, 2914m,
2876m, 1953w, 1877w, 1810w, 1753w, 1682w, 1601w,
1513s, 1455m, 1412w, 1363w, 1313w, 1233w, 1159m,
1092s, 1028m, 1005m, 911w. 1H NMR (CDCl3,
300 MHz): see Table 2; additionally, 0.60–0.79 (m,
(MeCH2)3Si); 0.88–1.00 (m, (MeCH2)3Si); 4.33 (d,
J = 11.5, PhCH); 4.45 (d, J = 11.8, PhCH); 4.50
(d, J = 11.8, PhCH); 4.52 (d, J = 11.5, PhCH); 4.53 (d,
J = 11.5, PhCH); 4.73 (d, J = 11.5, PhCH); 7.13–7.19
(m, 2 arom. H); 7.27–7.39 (m, 13 arom. H); 8.01
(exchange with CD3OD). 13C NMR (CDCl3, 75 MHz):
see Table 4; additionally, 4.96 (t, (MeCH2)3Si); 6.91
(q, (MeCH2)3Si); 71.97, 71.98, 73.30 (3t, 3PhCH2);
127.77–128.43 (several d); 136.97, 137.28, 137.32
(3s). HR-MALDI-MS: 600.2577 (100, [M+Na]+,
C33H43NNaO4SSi+; calcd 600.2580), 578.2756 (21,
[M+H]+, C33H44NO4SSi+; calcd 578.2760), 440.1701
(32). Anal. Calcd for C33H43NO4SSi (577.86): C,
68.59; H, 7.50; N, 2.42. Found: C, 68.59; H, 7.63; N,
2.60.
4.11. 2,5-Diamino-2,5-dideoxy-D-glucono-1,5-lactam 6
A soln of 19 (50 mg, 0.106 mmol) in MeOH (1 mL) was
treated with AcOH (1 mL) and 10% Pd/C (50 mg),
hydrogenated at 6 bar for 22 h and filtered through Cel-
ite. Evaporation of the combined filtrate and washing
(20 mL of MeOH), co-evaporation with toluene
(4 · 5 mL), ion-exchange chromatography (Amberlite
+
CG-120, NH4 form, elution with 0.05 M aq NH3)
and lyophilization gave 61 (17.9 mg, 96%).
4.12. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-(triethyl-
silyl)-D-glucono-1,5-lactam 20
At 0 ꢁC, a soln of 11 (50 mg, 0.112 mmol) in CH2Cl2
(0.5 mL) was successively treated with pyridine (25 lL,
0.309 mmol) and Et3SiOTf (34 lL, 0.149 mmol), stirred
at 0 ꢁC for 2 h and treated with satd NH4Cl soln (5 mL).
The mixture was diluted with Et2O (30 mL) and washed
with satd NH4Cl soln (3 · 15 mL). The combined aq
layers were extracted with Et2O (2 · 15 mL). The com-
bined org. layers were extracted with H2O (30 mL)
and brine (30 mL), dried (MgSO4), filtered and evapo-
rated. FC (hexane/AcOEt 4:1 ! 2:1) gave 20 (54 mg,
4.13.1. Condensation of 21 with aminoacetaldehyde
dimethyl acetal in the presence of Hg(OAc)2. At 0 ꢁC,
a suspension of 21 (24 mg, 41.5 lmol) and Hg(OAc)2
(24 mg, 75.3 lmol) in THF (1.3 mL) was treated with
aminoacetaldehyde dimethyl acetal (70 lL, 0.649 mmol)
and stirred at 0 ꢁC for 8 h. The black mixture was di-
luted with AcOEt (5 mL), filtered over Celite (the solid
was washed with 30 mL of AcOEt). The combined fil-
trates were washed with brine (20 mL), dried (MgSO4),
filtered and evaporated. A soln of the residue (25 mg)
in toluene (2 mL) and H2O (0.2 mL) was treated with
p-TsOHÆH2O (26 mg, 0.137 mmol), stirred for 20 h at
70 ꢁC, cooled to 22 ꢁC, diluted with AcOEt (20 mL)
and washed with satd NaHCO3 soln (3 · 10 mL). The
combined aq layers were extracted with AcOEt
(2 · 10 mL). The combined org. layers were washed with
H2O (20 mL) and brine (20 mL), dried (MgSO4), filtered
and evaporated. FC (AcOEt) gave 22/23 60:40 (7.8 mg,
40%).
86%). Colourless oil. Rf (hexane/AcOEt 2:1) 0.40.
25
½aŠ ¼ þ64:6 (c 1.03, CHCl3). IR (CHCl3): 3390w,
D
3089w, 3067w, 3032w, 3010m, 2956m, 2912m, 2877m,
1952w, 1875w, 1810w, 1684s, 1603w, 1497w, 1454m,
1412w, 1363m, 1315m, 1258w, 1161m, 1099s, 1070s,
1028m, 1007m, 912w. 1H NMR (CDCl3, 300 MHz):
see Table 2; additionally, 0.64–0.84 (m, (MeCH2)3Si);
0.95–1.04 (m, (MeCH2)3Si); 4.44 (d, J = 12.1, PhCH);
4.48 (d, J = 11.8, PhCH); 4.50 (d, J = 11.8, PhCH);
4.78 (d, J = 11.2, PhCH); 4.82 (d, J = 11.2,
PhCH); 4.94 (d, J = 11.2, PhCH); 6.04 (exchange with
CD3OD); 7.16–7.21 (m, 2 arom. H); 7.26–7.40 (m, 13
arom. H). 13C NMR (CDCl3, 75 MHz): see Table 4;
additionally, 5.16 (t, (MeCH2)3Si); 7.02 (q, (MeCH2)3-
Si); 73.25, 74.44, 74.61 (3t, 3PhCH2); 127.53–128.38
(several d); 137.19, 137.50, 138.06 (3s). HR-MALDI-
MS: 584.2797 (100, [M+Na]+, C33H43NNaO5Si+; calcd
584.2808), 562.2981 (4, [M+H]+, C33H44NO5Si+; calcd
562.2989), 532.2514 (40, [MÀEt]+, C31H38NO5Si+; calcd
532.2519), 424.1936 (40, [MÀBnOHÀEt]+, C24H30NO4-
Si+; calcd 424.1944), 286.1256 (25). Anal. Calcd for
4.14. Data of (5R,6R,7S,8S)-6,7-bis(benzyloxy)-5-[(benz-
yloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]pyridin-
8-ol 221,24,25
Colourless oil. Rf (AcOEt/MeOH 20:1) 0.15.
25
D
½aŠ ¼ þ12:5 (c 1.02, CHCl3). UV (CHCl3): 259 (2.78),

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
461
Table 6. Selected 1H NMR chemical shifts [ppm] and coupling constants [Hz] of the protected gluco-imidazoles 22, 27, 31, 37 and 40 and of the
arabino-imidazole 39 in CDCl3
1
221
27
7.02a
31
7.02a
37
3940
7.39
H–C(2)
H–C(3)
H–C(5)
H–C(6)
H–C(7)
H–C(8)
CH–C(5)
CH0–C(5)
J(2,3)
7.05a
7.06a
7.14a
7.03a
7.15a
4.16
3.98
3.90
4.75
3.70
3.80
1.2
7.12a
4.17
3.95
4.08
4.99
3.75
3.87
1.3
7.08a
4.21
3.85
3.98
4.93
3.76
3.87
1.3
7.06a
4.39
3.76
4.11
5.12
3.83
3.93
1.2
7.24
4.46–4.54
4.13
4.22
—
4.35–4.42
3.93
4.21
6.11
3.71
3.78
1.3
3.81
3.87
1.2
J(5,6)
J(6,7)
J(7,8)
7.8
8.7
8.1
6.9
8.4
3.4
5.3
5.3
5.0
6.2
7.5
8.1
7.5
5.0
5.3
5.3
3.1
5.9
3.7
6.5
—
6.9
5.0
J(5,CH)
J(5,CH0)
J(CH,CH0)
6.8
3.7
2.8
10.3
3.1
10.3
2.8
10.6
4.4
10.0
3.1
10.3
10.3
a Assignment may be interchanged.
Table 7. Selected 13C NMR chemical shifts [ppm] of the protected gluco-imidazoles 22, 27, 31, 37 and 40, of the protected manno-imidazoles 23, 28,
38 and 41, and of the arabino-imidazole 39 in CDCl3
Compound
C(2)
C(3)
C(5)
CH2–C(5)
C(6)
C(7)
C(8)
C(8a)
gluco
221
27
128.72
128.95
116.65
116.93
117.11
118.32
122.31
117.79
58.48
57.86
56.58
58.50
59.62
58.72b
68.35
68.37
68.04
70.46
69.87
68.56
75.26
76.81
78.50
74.55
75.68
75.23
82.92
83.49
82.77
77.79
79.84
80.76
67.87
147.29
145.22
145.36
140.90
140.24
140.91
68.26
66.38
66.54
a
31
37
129.33
133.17
129.96
39 (arabino)
401
181.93
59.10b
manno
231
28
128.83
128.86
129.94
129.77
118.20
118.68
118.03
118.72
59.26
59.65
59.15
58.96
70.84
71.00
69.33
70.03
73.59
73.60
73.18
73.14
79.19
81.26
78.60
78.29
62.46
63.75
63.15
55.57
144.93
144.74
141.74
140.26
38
411
a Not assigned.
b Assignment may be interchanged.
+
240 (3.03). IR (CHCl3): 3327w (br), 3163w, 3090w,
3067m, 3032m, 3012m, 2914m, 2869m, 1952w, 1879w,
1810w, 1751w, 1603w, 1525w, 1496m, 1454m, 1362m,
1310m, 1283w, 1167w, 1114s, 1085s, 1049s, 1028m,
493.2094 (22, [M+Na]+, C29H30N2NaO4
;
calcd
493.2103), 471.2277 (100, [M+H]+, C29H31N2O4+; calcd
+
471.2284), 453.2158 (34, [MÀOH]+, C29H29N2O3
;
calcd 453.2178).
1
912w. H NMR (CDCl3, 300 MHz): see Table 6. 13C
NMR (CDCl3, 75 MHz): see Table 7. HR-MALDI-
MS: 493.2094 (19, [M+Na]+, C29H30N2NaO4+; calcd
4.16. 5-Amino-3,4,6-tri-O-benzyl-2-O-[(tert-butyl)dimeth-
ylsilyl]-5-deoxy-D-glucono-1,5-lactam 24 and 5-amino-
3,4,6-tri-O-benzyl-2-O-[(tert-butyl)dimethylsilyl]-5-
deoxy-D-mannono-1,5-lactam 25
493.2103), 471.2279 (100, [M+H]+, C29H31N2O4+; calcd
+
471.2284), 453.2163 (41, [MÀOH]+, C29H29N2O3
;
calcd 453.2178).
(a) A mixture of 11 (150 mg, 0.335 mmol), imidazole
(57 mg, 0.837 mmol) and t-BuMe2SiCl (66 mg,
0.438 mmol) in DMF (1.2 mL) was stirred at 23 ꢁC for
22 h, diluted with Et2O (30 mL) and washed with satd
NH4Cl soln (3 · 15 mL). The combined aq layers were
extracted with Et2O (2 · 20 mL). The combined org. lay-
ers were extracted with H2O (40 mL) and brine (40 mL),
dried (MgSO4), filtered and evaporated. FC (hexane/
AcOEt 4:1) gave 24 (182 mg, 97%).
4.15. Data of (5R,6R,7S,8R)-6,7-bis(benzyloxy)-5-[(benz-
yloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]pyridin-
8-ol 231,24,25
Colourless solid. Rf (AcOEt/MeOH 20:1) 0.14. Mp 112–
25
113 ꢁC (lit.:24,25 116.5–117.5 ꢁC). ½aŠ ¼ À0:3 (c 0.98,
D
CHCl3) [lit.:24,25 À4.0 (CHCl3)]. UV (CHCl3): 259
(2.71), 240 (3.03). IR (CHCl3): 3325w (br), 3163w,
3090w, 3067w, 3032m, 3011m, 2926m, 2868m, 1952w,
1877w, 1810w, 1725w, 1603w, 1496m, 1454m, 1363m,
(b) At 0 ꢁC, a soln of 11 (150 mg, 0.335 mmol) in
CH2Cl2 (1.2 mL) was successively treated with pyridine
(60 lL, 0.742 mmol) and t-BuMe2SiOTf (100 lL,
0.435 mmol) and stirred at 0 ꢁC for 4 h. Workup and
1
1312w, 1269w, 1170w, 1100s, 1028m, 1013m, 912w. H
NMR (CDCl3, 300 MHz): see Table 8. 13C NMR
(CDCl3, 75 MHz): see Table 7. HR-MALDI-MS:

462
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
Table 8. Selected 1H NMR chemical shifts [ppm] and coupling
constants [Hz] of the protected manno-imidazoles 23, 28, 38 and 41
in CDCl3
546.2662 (55), 528.2559 (15), 438.2088 (56), 300.1412
(35). Anal. Calcd for C33H43NO5Si (561.79): C, 70.55;
H, 7.71; N, 2.49. Found: C, 70.34; H, 7.57; N, 2.54.
231
28
7.01
38
7.07
411
H–C(2)a
H–C(3)a
H–C(5)
H–C(6)
H–C(7)
H–C(8)
CH–C(5)
CH0–C(5)
J(2,3)
7.01
7.11
7.13
4.16.2. Data of 25. Colourless oil. Rf (hexane/AcOEt
25
D
2:1) 0.21. ½aŠ ¼ þ0:9 (c 2.01, CHCl3). IR (CHCl3):
7.11
4.16
4.23
3.97
5.16
3.71
3.81
1.2
7.07
4.10
4.25
3.79
5.10
3.56
3.69
1.2
7.13
4.07
4.24
4.05
6.58
3.70
3.80
1.3
3394w, 3089w, 3067w, 3031w, 3012m, 2954m, 2930m,
2885m, 2858m, 1951w, 1875w, 1810w, 1681s, 1604w,
1496w, 1471m, 1462m, 1454m, 1389w, 1362m, 1319w,
1260m, 1098s, 1072s, 1028m, 1006w, 939w, 907w, 838s.
1H NMR (CDCl3, 300 MHz): see Table 5; additionally,
0.17, 0.23 (2s, Me2Si); 0.97 (s, Me3CSi); 4.448 (d,
J = 11.8, PhCH); 4.48 (d, J = 11.8, PhCH); 4.50 (d, J =
11.8, PhCH); 4.62 (d, J = 11.8, PhCH); 4.65 (d,
J = 11.8, PhCH); 4.86 (d, J = 12.1, PhCH); 5.97
(exchange with CD3OD); 7.21–7.24 (m, 2 arom. H);
7.28–7.40 (m, 13 arom. H). 13C NMR (CDCl3,
75 MHz): see Table 4; additionally, À5.33, À4.39 (2q,
Me2Si); 18.51 (s, Me3CSi); 25.91 (q, Me3CSi); 72.60,
73.05, 73.18 (3t, 3PhCH2); 127.51–128.31 (several d);
137.26, 137.30, 138.02 (3s). HR-MALDI-MS: 600.2540
(4, [M+K]+, C33H43KNO5Si+; calcd 600.2548),
584.2797 (100, [M+Na]+, C33H43NNaO5Si+; calcd
584.2808), 562.2964 (15, [M+H]+, C33H44NO5Si+; calcd
562.2989), 546.2666 (97). Anal. Calcd for C33H43NO5Si
(561.79): C, 70.55; H, 7.71; N, 2.49. Found: C, 70.56; H,
7.76; N, 2.64.
4.11–4.17
4.11–4.17
4.02
4.93
3.67
3.77
1.2
b
J(5,6)
J(6,7)
J(7,8)
6.5
8.4
6.9
9.0
7.8
9.3
8.4
3.7
5.9
3.1
3.4
6.5
2.8
6.9
3.7
5.3
J(5,CH)
J(5,CH0)
J(CH,CH0)
3.1
10.0
3.1
10.0
2.8
10.0
10.0
a Assignment may be interchanged.
b Not determined.
Table 9. Selected 1H NMR chemical shifts [ppm] and coupling
constants [Hz] of the unprotonated and protonated mannosamine-
derived imidazole 8 and lactam 10 in D2O (conventional carbohydrate
numbering used)
8
8ÆH+
8Æ2H+
10
10ÆH+
H–C(2)
H–C(3)
H–C(4)
H–C(5)
H–C(6)
H0–C(6)
J(2,3)
4.10
4.009
4.22
4.14
3.88
4.008
4.0
5.06
5.15
4.32
4.43
4.38
4.09
4.18
5.0
3.58
4.24–4.36
4.24–4.36
3.95–4.02
3.48–3.58
3.71
4.32
4.43
4.32–4.39
4.10
4.20
5.3
3.98
3.87
3.44
3.64
3.72
3.7
4.17. 5-Amino-3,4,6-tri-O-benzyl-2-O-[(tert-butyl)di-
methylsilyl]-5-deoxy-D-glucono-1,5-thiolactam 26
A soln of 24 (301 mg, 0.536 mmol) in toluene (9 mL)
was treated with Lawessonꢂs reagent (160 mg,
0.396 mmol), stirred at 23 ꢁC for 22 h, diluted with
Et2O (60 mL) and washed with satd NaHCO3 soln
(3 · 40 mL). The combined aq layers were extracted
with Et2O (2 · 40 mL). The combined org. layers were
washed with H2O (70 mL) and brine (70 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
3.80
a
a
a
J(3,4)
J(4,5)
6.5
4.0
7.2
5.0
7.5
5.0
3.7
4.0
J(5,6)
J(5,60)
J(6,60)
2.8
3.4
5.0
3.1
5.3
3.4
5.0
4.0
6.9
4.4
12.1
12.8
12.8
11.8
10.9
a Not determined.
6:1) gave 26 (283 mg, 91%). Colourless oil. Rf (hexane/
25
D
AcOEt 6:1) 0.20. ½aŠ ¼ þ84:8 (c 0.86, CHCl3). IR
FC (as described in a) gave 24 (172 mg, 91%) and 25
(6 mg, 3%).
(CHCl3): 3376w, 3090w, 3067w, 3020m, 2955m,
2931m, 2884m, 2859m, 1952w, 1875w, 1811w, 1602w,
1513s, 1471m, 1463m, 1454m, 1390w, 1363m, 1313w,
1
4.16.1. Data of 24. Colourless oil. Rf (hexane/AcOEt
1258m, 1160m, 1093s, 1028m, 1006w, 911w, 841s. H
25
D
4:1) 0.11. ½aŠ ¼ þ66:0 (c 1.01, CHCl3). IR (CHCl3):
NMR (CDCl3, 300 MHz): see Table 2; additionally,
0.16, 0.18 (2s, Me2Si); 0.89 (s, Me3CSi); 3.41 (irrad. at
3.67 ! d, J = 7.2, irrad. at 3.99 ! d, J = 10.3); 3.53
(irrad. at 3.78 ! br d, J ꢀ 7.5, irrad. at 3.99 ! br s,
irrad. at 4.72 ! dd, J = 2.8, 9.0); 3.67 (irrad. at
3.41 ! change, irrad. at 3.99 ! d, J = 9.3); 3.78 (irrad.
at 3.53 ! d, J = 3.1, irrad. at 4.72 ! d, J = 2.8); 3.99
(irrad. at 3.41 ! change, irrad. at 3.53 ! change, irrad.
at 3.67 ! change); 4.33 (d, J = 11.5, PhCH); 4.45 (d,
J = 12.1, PhCH); 4.49 (br s, PhCH2); 4.51 (d, J = 11.5,
PhCH); 4.719 (irrad. at 3.53 ! d, J = 3.1, irrad. at
3.78 ! br s); 4.723 (d, J = 11.8, PhCH); 7.15–7.21 (m,
2 arom. H); 7.28–7.40 (m, 13 arom. H); 7.99 (exchange
with CD3OD). 13C NMR (CDCl3, 75 MHz): see Table
4; additionally, À4.91, À4.06 (2q, Me2Si); 18.20 (s,
Me3CSi); 25.76 (q, Me3CSi); 71.88, 71.91, 73.31 (3t,
3PhCH2); 127.76–128.46 (several d); 136.98, 137.24,
137.36 (3s). HR-MALDI-MS: 616.2522 (21, [M+K]+,
3391w, 3090w, 3067w, 3032w, 3012m, 2954m, 2930m,
2886m, 2858m, 1951w, 1875w, 1810w, 1686s, 1604w,
1497w, 1472w, 1454m, 1389w, 1362m, 1314m, 1259m,
1160m, 1100s, 1072s, 1028m, 1005w, 939w, 911w,
1
841m. H NMR (CDCl3, 300 MHz): see Table 2; addi-
tionally, 0.15, 0.22 (2s, Me2Si); 0.94 (s, Me3CSi); 4.43
(d, J = 12.1, PhCH); 4.47 (d, J = 11.8, PhCH); 4.49 (d,
J = 12.1, PhCH); 4.79 (d, J = 11.5, PhCH); 4.80 (d,
J = 11.2, PhCH); 4.92 (d, J = 11.5, PhCH); 6.02 (ex-
change with CD3OD); 7.14–7.18 (m, 2 arom. H); 7.26–
7.36 (m, 13 arom. H). 13C NMR (CDCl3, 75 MHz):
see Table 4; additionally, À5.01, À4.02 (2q, Me2Si);
18.56 (s, Me3CSi); 25.95 (q, Me3CSi); 73.25, 74.47,
74.73 (3t, 3PhCH2); 127.51–128.39 (several d); 137.18,
137.44, 137.92 (3s). HR-MALDI-MS: 584.2797 (100,
[M+Na]+, C33H43NNaO5Si+; calcd 584.2808), 562.2980
(11, [M+H]+, C33H44NO5Si+; calcd 562.2989),

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
463
C33H43KNO4SSi+; calcd 616.2319), 600.2575 (100,
[M+Na]+, C33H43NNaO4SSi+;
calcd 600.2580),
(100, [MÀTBDMSO]+, C29H29N2O3+; calcd 453.2178).
Anal. Calcd for C35H44N2O4Si (584.83): C, 71.88; H,
7.58; N, 4.79. Found: C, 71.85; H, 7.53; N, 4.97.
594.2687 (16), 584.2795 (12), 578.2757 (53, [M+H]+,
C33H44NO4SSi+; calcd 578.2760), 560.2652 (18),
470.2176 (13, [MÀBnO]+, C26H36NO3SSi+; calcd
470.2185), 422.1968 (18). Anal. Calcd for C33H43NO4S-
Si (577.86): C, 68.59; H, 7.50; N, 2.42. Found: C, 68.40;
H, 7.52; N, 2.46.
4.19. Data of (5R,6R,7S,8R)-6,7-bis(benzyloxy)-5-[(benz-
yloxy)methyl]-8-[(tert-butyl)dimethylsilyl]-5,6,7,8-tetra-
hydroimidazo[1,2-a]pyridine 28
Colourless oil. Rf (hexane/AcOEt 2:1) 0.43.
25
D
4.17.1. Condensation of 26 with aminoacetaldehyde
dimethyl acetal in the presence of Hg(OAc)2. (a) At
0 ꢁC, a suspension of 26 (258 mg, 0.446 mmol) and
Hg(OAc)2 (200 mg, 0.628 mmol) in THF (2.5 mL) was
treated with aminoacetaldehyde dimethyl acetal
(0.25 mL, 2.28 mmol) and stirred at 0 ꢁC for 5 h. The
black mixture was diluted with AcOEt (4 mL), filtered
over Celite (the solid was washed with 50 mL of
AcOEt). The combined filtrate and washing were
washed with brine (40 mL), dried (MgSO4), filtered
and evaporated. A soln of the residue (340 mg) in tolu-
ene (12.5 mL) and H2O (1.2 mL) was treated with
p-TsOHÆH2O (225 mg, 1.18 mmol), stirred for 20 h at
70ꢁ, cooled to 22 ꢁC, diluted with AcOEt (60 mL) and
washed with satd NaHCO3 soln (3 · 40 mL). The
combined aq layers were extracted with AcOEt
(2 · 40 mL). The combined org. layers were washed with
H2O (80 mL) and brine (80 mL), dried (MgSO4), filtered
and evaporated. FC (hexane/AcOEt 1:0 ! 4:1 !
2:1 ! 1:1 ! 0:1) gave 27 (21 mg, 8%), 28 (3 mg, 1%),
111 (13 mg, 7%) and 221,24,25 (174 mg, 83%).
½aŠ ¼ À17:3 (c 0.96, CHCl3). UV (CHCl3): 259 (2.80),
240 (3.01). IR (CHCl3): 3089w, 3067w, 3032w, 3013m,
2955s, 2929s, 2892m, 2858m, 1951w, 1875w, 1810w,
1725w, 1603w, 1496m, 1471m, 1462m, 1454m, 1362m,
1313w, 1264w, 1170w, 1133s, 1100s, 1028w, 1006w,
954m, 912w, 838s. 1H NMR (CDCl3, 300 MHz): see Ta-
ble 8; additionally, À0.04, 0.22 (2s, Me2Si); 0.87 (s,
Me3CSi); 4.44 (br s, PhCH2); 4.64 (d, J = 11.5, PhCH);
4.65 (d, J = 11.8, PhCH); 4.81 (d, J = 11.8, PhCH);
4.93 (d, J = 11.5, PhCH); 7.24–7.40 (m, 15 arom. H).
13C NMR (CDCl3, 75 MHz): see Table 7; additionally,
À4.90, À4.60 (2q, Me2Si); 18.24 (s, Me3CSi); 25.79 (q,
Me3CSi); 71.59, 73.08, 74.58 (3t, 3PhCH2); 127.51–
128.33 (several d); 137.45, 137.82, 137.92 (3s). FAB-
MS: 585 (100, [M+H]+), 528 (44, [M+HÀt-Bu]+), 478
(6), 453 (16, [MÀTBDMSO]+), 391 (57), 167 (9), 149
(42), 91 (44, C7H7+). HR-MALDI-MS: 607.2947 (17,
[M+Na]+,
C35H44N2NaO4Si+;
calcd
607.2968),
585.3135 (65, [M+H]+, C35H45N2O4Si+; calcd
585.3149), 453.2168 (100, [MÀTBDMSO]+, C29H29-
N2O3+; calcd 453.2178). Anal. Calcd for C35H44N2O4Si
(584.83): C, 71.88; H, 7.58; N, 4.79. Found: C, 72.07; H,
7.76; N, 4.84.
(b) Similarly as described in a, but the mixture of the res-
idue with p-TsOHÆH2O was stirred at 55 ꢁC for 96 h.
After workup and FC, such a transformation of 26
(4.47 g, 7.74 mmol) afforded 27 (2.02 g, 45%), 28
(0.11 g, 2%) and a mixture of 221,24,25 and intermediate
amidine 9:1 (1.34 g, ca. 37%).
4.20. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-(triisoprop-
ylsilyl)-D-glucono-1,5-lactam 29
(a) A mixture of 11 (1.00 g, 2.23 mmol), imidazole
(0.37 g, 5.43 mmol) and TIPSCl (0.63 mL, 2.94 mmol)
in DMF (8 mL) was stirred at 23 ꢁC for 16 h, diluted
with Et2O (250 mL) and washed with satd NH4Cl soln
(3 · 100 mL). The combined aq layers were extracted
with Et2O (2 · 60 mL). The combined org. layers were
washed with H2O (200 mL) and brine (200 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
4:1) gave 29 (1.31 g, 97%).
4.18. Data of (5R,6R,7S,8S)-6,7-bis(benzyloxy)-5-[(benz-
yloxy)methyl]-8-[(tert-butyl)dimethylsilyl]-5,6,7,8-tetra-
hydroimidazo[1,2-a]pyridine 27
25
D
Colourless oil. Rf (hexane/AcOEt 2:1) 0.63. ½aŠ ¼ þ41:9
(c 1.99, CHCl3). UV (CHCl3): 259 (2.77), 240 (3.03). IR
(CHCl3): 3089w, 3067w, 3032w, 3012w, 2954m, 2930m,
2884m, 2859m, 1951w, 1875w, 1810w, 1725w, 1603w,
1496w, 1472w, 1454m, 1389w, 1362m, 1308w, 1283w,
1261m, 1090s, 1028m, 1005w, 940w, 909w, 840s. 1H
NMR (CDCl3, 300 MHz): see Table 6; additionally,
0.17, 0.23 (2s, Me2Si); 0.94 (s, Me3CSi); 4.46 (d,
J = 12.1, PhCH); 4.49 (d, J = 11.2, PhCH); 4.51 (d, J =
12.1, PhCH); 4.77 (d, J = 11.5, PhCH); 4.80 (d,
J = 11.2, PhCH); 4.87 (d, J = 11.5, PhCH); 7.16–7.19
(m, 2 arom. H); 7.27–7.39 (m, 13 arom. H). 13C NMR
(CDCl3, 75 MHz): see Table 7; additionally, À4.80,
À4.23 (2q, Me2Si); 18.36 (s, Me3CSi); 25.92 (q, Me3CSi);
73.20, 73.80, 73.82 (3t, 3PhCH2); 127.53–128.35 (several
d); 137.19, 137.48, 137.78 (3s). FAB-MS: 1170 (11,
[2M+H]+), 585 (100, [M+H]+), 528 (81, [M+HÀt-
Bu]+), 478 (6), 453 (19, [MÀTBDMSO]+), 154 (7), 91
(50, C7H7+). HR-MALDI-MS: 607.2962 (7, [M+Na]+,
C35H44N2NaO4Si+; calcd 607.2968), 585.3149 (12,
[M+H]+, C35H45N2O4Si+; calcd 585.3149), 453.2181
(b) At 0 ꢁC, a soln of 11 (125 mg, 0.279 mmol) and pyr-
idine (50 lL, 0.618 mmol) in CH2Cl2 (1 mL) was treated
with TIPSOTf (100 lL, 0.372 mmol) and stirred at 0 ꢁC
for 3.5 h. Workup and FC (as described in a) gave 29
(156 mg, 92%).
4.20.1. Data of 29. Colourless oil. Rf (hexane/AcOEt
25
D
4:1) 0.17. ½aŠ ¼ þ62:0 (c 0.61, CHCl3). IR (CHCl3):
3396w, 3090w, 3067w, 3033w, 3011m, 2946s, 2893s,
2868s, 1952w, 1877w, 1808w, 1688s, 1497w, 1455m,
1386w, 1363m, 1314w, 1261w, 1213m, 1100s, 1072s,
1
1028m, 913w, 883m. H NMR (CDCl3, 300 MHz): see
Table 3; additionally, 1.00–1.32 (m, (Me2CH)3Si); 4.38
(d, J = 11.8, PhCH); 4.43 (d, J = 11.8, PhCH); 4.50 (d,
J = 11.8, PhCH); 4.63 (d, J = 11.5, PhCH); 4.64 (d,
J = 11.8, PhCH); 4.79 (d, J = 11.5, PhCH); 5.87
(exchange with CD3OD); 7.16–7.19 (m, 2 arom. H);

464
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
7.27–7.39 (m, 13 arom. H). 13C NMR (CDCl3,
75 MHz): see Table 4; additionally, 12.61 (d, (Me2CH)3-
Si); 18.05, 18.12 (2q, (Me2CH)3Si); 72.93, 73.00, 73.06
(3t, 3PhCH2); 127.62–128.29 (several d); 137.23,
137.49, 137.67 (3s). HR-MALDI-MS: 626.3278 (100,
[M+Na]+, C36H49NNaO5Si+; calcd 626.3278), 604.3456
(2, [M+H]+, C36H50NO5Si+; calcd 604.3458), 560.2826
(64, [MÀi-Pr]+, C33H42NO5Si+; calcd 560.2832),
452.2245 (35, [MÀi-PrÀBnOH]+, C26H34NO4Si+; calcd
452.2257), 314.1565 (17). Anal. Calcd for C36H49NO5Si
(603.87): C, 71.60; H, 8.18; N, 2.32. Found: C, 71.77; H,
8.01; N, 2.37.
2.13 mmol), stirred for 20 h at 70 ꢁC, cooled to 22 ꢁC,
diluted with AcOEt (70 mL) and washed with satd
NaHCO3 soln (3 · 40 mL). The combined aq layers
were extracted with AcOEt (2 · 40 mL). The combined
org. layers were washed with H2O (80 mL) and brine
(80 mL), dried (MgSO4), filtered and evaporated. FC
(hexane/AcOEt 1:0 ! 5:1 ! 3:1) gave 31 (445 mg,
88%). Colourless oil.
4.22. Data of (5R,6R,7S,8S)-6,7-bis(benzyloxy)-5-[(benz-
yloxy)methyl]-5,6,7,8-tetrahydro-8-(triisopropylsilyl)-imid-
azo[1,2-a]pyridine 31
25
D
4.21. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-(triisoprop-
ylsilyl)-D-glucono-1,5-thiolactam 30
Rf (hexane/AcOEt 3:1) 0.22. ½aŠ ¼ þ49:9 (c 1.10,
CHCl3). UV (CHCl3): 259 (2.77), 239 (3.09). IR
(CHCl3): 3089w, 3067w, 3033w, 3010w, 2945s, 2892m,
2867s, 1951w, 1877w, 1810w, 1731w, 1686w, 1602w,
1534w, 1496w, 1454m, 1384w, 1363w, 1337w, 1319w,
1289w, 1263w, 1092s, 1028w, 1015w, 909w, 883m,
A soln of 29 (1.28 g, 2.12 mmol) in toluene (35 mL) was
treated with Lawessonꢂs reagent (630 mg, 1.56 mmol),
stirred for 36 h at 23 ꢁC, diluted with Et2O (250 mL)
and washed with satd NaHCO3 soln (3 · 100 mL). The
combined aq layers were extracted with Et2O
(2 · 80 mL). The combined org. layers were washed with
H2O (150 mL) and brine (150 mL), dried (MgSO4), fil-
tered and evaporated. FC (hexane/AcOEt 8:1) gave 30
1
843w. H NMR (CDCl3, 300 MHz): see Table 6; addi-
tionally, 0.96–1.26 (m, (Me2CH)3Si); 4.45 (d, J = 11.8,
PhCH); 4.53 (br s, PhCH2); 4.60 (d, J = 11.8, PhCH);
4.63 (d, J = 11.5, PhCH); 4.73 (d, J = 11.8, PhCH);
7.20–7.40 (m, 15 arom. H). 13C NMR (CDCl3,
75 MHz): see Table 7; additionally, 12.34 (d, (Me2CH)3-
Si); 17.87, 18.05 (2q, (Me2CH)3Si); 72.09, 72.38, 73.13
(3t, 3PhCH2); 127.73–128.45 (several d including C(2));
137.37 (s); 137.65 (2s). HR-MALDI-MS: 649.3438 (6,
(1.26 g, 96%). Colourless oil. Rf (hexane/AcOEt 8:1)
25
D
0.15. ½aŠ ¼ þ73:1 (c 0.67, CHCl3). IR (CHCl3):
3376w, 3090w, 3068w, 3020m, 2946s, 2892m, 2868s,
1952w, 1882w, 1813w, 1604w, 1513s, 1465m, 1455m,
1385w, 1365m, 1313w, 1252w, 1160m, 1098s, 1029m,
[M+Na]+,
C38H50N2NaO4Si+;
calcd
649.3437),
1
910w, 883m. H NMR (CDCl3, 300 MHz): see Table
627.3621 (7, [M+H]+, C38H51N2O4Si+; calcd 627.3618),
3; additionally, 1.02–1.26 (m, (Me2CH)3Si); 4.32 (d,
J = 11.8, PhCH); 4.46 (d, J = 11.8, PhCH); 4.49
(d, J = 11.8, PhCH); 4.50 (d, J = 11.8, PhCH); 4.54 (d,
J = 11.8, PhCH); 4.72 (d, J = 11.8, PhCH); 7.18–7.21
(m, 2 arom. H); 7.29–7.40 (m, 13 arom. H); 8.06
(exchange with CD3OD). 13C NMR (CDCl3, 75 MHz):
see Table 4; additionally, 12.33 (d, (Me2CH)3Si);
18.09, 18.13 (2q, (Me2CH)3Si); 71.53 (t, 2PhCH2);
73.11 (t, PhCH2); 127.67–128.38 (several d); 137.00,
137.19, 137.29 (3s). HR-MALDI-MS: 658.3001 (24,
[M+K]+, C36H49KNO4SSi+; calcd 658.2789), 642.3038
(100, [M+Na]+, C36H49NNaO4SSi+; calcd 642.3049),
636.3159 (23), 626.3273 (25), 620.3257 (18, [M+H]+,
C36H50NO4SSi+; calcd 620.3230), 576.2593 (32, [MÀi-
Pr]+, C33H42NO4SSi+; calcd 576.2604), 468.2011 (73,
[MÀi-PrÀBnOH]+, C26H34NO3SSi+; calcd 468.2029),
462.1731 (73, [MÀi-Pr3Si]+, C27H28NO4S+; calcd
462.1739). Anal. Calcd for C36H49NO4SSi (619.94): C,
69.75; H, 7.97; N, 2.26. Found: C, 69.81; H, 7.73; N,
2.38.
583.2990 (8, [MÀi-Pr]+, C35H43N2O4Si+; calcd
+
583.2992), 453.2175 (100, [MÀTIPSO]+, C29H29N2O3
;
calcd 453.2178). Anal. Calcd for C38H50N2O4Si
(626.91): C, 72.80; H, 8.04; N, 4.47. Found: C, 72.69;
H, 8.02; N, 4.45.
4.23. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-pivaloyl-
D-glucono-1,5-lactam 32 and 3,4,6-tri-O-benzyl-5-deoxy-
2-O-pivaloyl-5-(pivaloylamino)-D-glucono-1,5-lactam 33
(a) At 5 ꢁC, a soln of 11 (83 mg, 0.185 mmol) in pyridine
(2.5 mL) was treated with pivaloyl chloride (40 lL,
0.325 mmol), stirred for 1 h at 5 ꢁC and for 22 h at
23 ꢁC, treated with pivaloyl chloride (40 lL,
0.325 mmol), stirred for 8 h at 23 ꢁC and for 6 h at
50 ꢁC and evaporated. A soln of the residue in CHCl3
(10 mL) was washed with satd NaHCO3 soln
(2 · 10 mL), H2O (10 mL) and brine (15 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
6:1 ! 4:1 ! 2:1) gave 33 (8 mg, 7%) and 32 (67 mg,
68%).
4.21.1. Condensation of 30 with aminoacetaldehyde
dimethyl acetal in the presence of Hg(OAc)2. At 0 ꢁC,
a suspension of 30 (500 mg, 0.807 mmol) and Hg(OAc)2
(360 mg, 1.13 mmol) in THF (4.5 mL) was treated with
(b) At 23 ꢁC, a soln of 11 (75 mg, 0.168 mmol) in pyr-
idine (2.4 mL) was treated with pivaloyl chloride
(43 lL, 0.350 mmol), stirred for 3 h at 50 ꢁC and evapo-
rated. A soln of the residue in Et2O (40 mL) was washed
with satd NH4Cl soln (3 · 15 mL). The combined aq ex-
tracts were extracted with Et2O (2 · 15 mL). The com-
bined org. extracts were washed with H2O (40 mL)
and brine (40 mL), dried (MgSO4), filtered and evapo-
rated. FC (as described in a) gave 33 (78 mg, 76%, col-
ourless oil crystallizing upon drying) and 32 (15 mg,
17%).
aminoacetaldehyde
dimethyl
acetal
(0.45 mL,
4.17 mmol) and stirred at 0 ꢁC for 4 h. The black mix-
ture was diluted with AcOEt (4 mL), filtered over Celite
(the solid was washed with 50 mL of AcOEt). The com-
bined filtrates were washed with brine (50 mL), dried
(MgSO4), filtered and evaporated. A soln of the residue
(600 mg) was dissolved in toluene (22.5 mL) and H2O
(2.2 mL), treated with p-TsOHÆH2O (405 mg,

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
465
4.23.1. Data of 32. Colourless crystals. Rf (hexane/
AcOEt 2:1) 0.20. Mp 105–107 ꢁC. ½aŠ ¼ þ176:8 (c
4.24.1. Data of 34. Colourless oil. Rf (hexane/AcOEt
3:1) 0.32. ½aŠ ¼ þ92:3 (c 0.56, CHCl3). IR (CHCl3):
25
D
25
D
0.69, CHCl3). IR (CHCl3): 3390m, 3090w, 3067w,
3031m, 3012m, 2973m, 2933m, 2909m, 2871m, 1952w,
1875w, 1810w, 1737s, 1689s, 1605w, 1497w, 1479m,
1454s, 1398m, 1363m, 1317m, 1278m, 1257m, 1139s,
1113s, 1061s, 1039m, 1028m, 941w, 912w. 1H NMR
(CDCl3, 300 MHz): see Table 3; additionally, 1.27 (s,
Me3C); 4.07 (with virtual coupling); 4.45 (d, J = 12.1,
PhCH); 4.49 (d, J = 12.1, PhCH); 4.53 (d, J = 10.9,
PhCH); 4.77 (br s, PhCH2); 4.94 (d, J = 11.2, PhCH);
6.05 (exchange with CD3OD); 7.17–7.22 (m, 2 arom.
H); 7.27–7.40 (m, 13 arom. H). 13C NMR (CDCl3,
75 MHz): see Table 4; additionally, 27.27 (q, Me3C);
38.83 (s, Me3C); 73.27, 74.79, 74.92 (3t, 3PhCH2);
127.49–128.41 (several d); 137.06, 137.22, 137.58 (3s);
177.35 (s, OC@O). HR-MALDI-MS: 570.2234 (5,
3360w, 3089w, 3067w, 3019m, 2974m, 2931m, 2870m,
1951w, 1875w, 1810w, 1736s, 1603w, 1514s, 1479m,
1455m, 1398w, 1364m, 1313m, 1276m, 1143s, 1122s,
1071s, 1028m, 944w, 911w. 1H NMR (CDCl3,
300 MHz): see Table 3; additionally, 1.26 (s, Me3C);
3.29 (with virtual coupling); 3.56–3.70 (irrad. at
3.29 ! change, irrad. at 3.96 ! change); 3.96 (with vir-
tual coupling, irrad. at 5.60 d, J = 8.1, with virtual cou-
pling); 4.443 (d, J = 12.5, PhCH); 4.444 (d, J = 11.8,
PhCH); 4.48 (d, J = 12.5, PhCH); 4.71 (d, J = 11.2,
PhCH); 4.73 (d, J = 11.2, PhCH); 4.79 (d, J = 11.2,
PhCH); 5.60 (irrad. at 3.96 ! s); 7.12–7.16 (m, 2 arom.
H); 7.26–7.40 (m, 13 arom. H); 8.11 (exchange with
CD3OD). 13C NMR (CDCl3, 75 MHz): see Table 4;
additionally, 27.27 (q, Me3C); 38.82 (s, Me3C); 73.42,
74.40, 74.43 (3t, 3PhCH2); 127.54–128.48 (several d);
136.76, 136.97, 137.39 (3s); 176.90 (s, OC@O). HR-
MALDI-MS: 570.2279 (98, [M+Na]+, C32H37-
NNaO5S+; calcd 570.2290), 554.2512 (42), 548.2462 (5,
[M+H]+, C32H38NO5S+; calcd 548.2471), 448.1939 (69,
[M+2HÀPivO]+, C27H30NO3S+; calcd 448.1946),
446.1787 (17, [MÀPivO]+, C27H28NO3S+; calcd
446.1790), 440.1888 (12, [MÀBnO]+, C25H30NO4S+;
calcd 440.1896), 425.3598 (47), 397.3288 (100). Anal.
Calcd for C32H37NO5S (547.71): C, 70.17; H, 6.81; N,
2.56. Found: C, 69.98; H, 6.74; N, 2.62.
[M+K]+, C32H37KNO6+; calcd 570.2258), 554.2506
+
(100, [M+Na]+, C32H37NNaO6
; calcd 554.2519),
532.2690 (12, [M+H]+, C32H38NO6+; calcd 532.2699),
430.2006 (6, [MÀPivO]+, C27H28NO4+; calcd 430.2018).
Anal. Calcd for C32H37NO6 (531.65): C, 72.29; H, 7.01;
N, 2.63. Found: C, 72.09; H, 6.95; N, 2.66.
4.23.2. Data of 33. Colourless oil. Rf (hexane/AcOEt
25
D
2:1) 0.80. ½aŠ ¼ þ54:2 (c 1.04, CHCl3). IR (CHCl3):
3089w, 3067w, 3031w, 3013w, 2971m, 2932m, 2872m,
1953w, 1875w, 1810w, 1737s, 1723s, 1700s, 1603w,
1496w, 1481m, 1455m, 1397m, 1365m, 1281m, 1262m,
1139s, 1099s, 1074s, 1042m, 1028m, 942w, 909w. 1H
NMR (CDCl3, 300 MHz): see Table 3; additionally,
1.27, 1.29 (2s, 2Me3C); 3.57 (irrad. at 4.69 ! d,
J = 10.3); 3.63 (irrad. at 4.69 ! d, J = 9.7); 3.95 (irrad.
at 4.69 ! dd, J = 4.4, 7.2, irrad. at 5.56 ! dd, J = 1.3,
4.1); 3.98 (irrad. at 4.69 ! d, J = 4.4); 4.42 (d, J = 11.8,
PhCH); 4.48 (d, J = 11.8, PhCH); 4.57 (d, J ꢀ 12.1,
PhCH); 4.58 (d, J = 11.2, PhCH); 4.62 (d, J = 11.5,
PhCH); 4.63 (d, J = 12.1, PhCH); 7.22–7.36 (m, 15 arom.
H). 13C NMR (CDCl3, 75 MHz): see Table 4; addition-
ally, 27.23, 27.78 (2q, 2Me3C); 38.78 (s, Me3CCO2);
43.78 (s, Me3CCON); 71.74, 72.90, 73.25 (3t, 3PhCH2);
127.49–128.34 (several d); 137.15 (2s); 137.18 (s);
4.25. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-(methoxy-
methyl)-D-glucono-1,5-lactam 35 and 3,4,6-tri-O-benzyl-
5-deoxy-2-O-(methoxymethyl)-5-[(methoxymethyl)-
amino]-D-glucono-1,5-lactam 36
(a) A soln of 11 (22 mg, 49.2 lmol) in CHCl3 (1 mL) was
treated successively with MeOCH2OMe (0.3 mL,
3.38 mmol) and P2O5 (163 mg, 1.15 mmol), stirred for
30 min at 23 ꢁC, cooled to 0 ꢁC and treated with satd
NaHCO3 soln (2 mL). The mixture was diluted with
Et2O (20 mL) and washed with satd NaHCO3 soln
(3 · 15 mL). The combined aq layers were extracted
with Et2O (2 · 15 mL). The combined org. layers were
washed with H2O (25 mL) and brine (25 mL), dried over
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
4:1 ! 2:1) gave 36 (2 mg, 9%) and 35 (15 mg, 62%).
177.19 (s, OC@O); 188.17 (s, NC@O). HR-MALDI-
; calcd
MS: 654.2817 (3, [M+K]+, C37H45KNO7
+
654.2833), 638.3080 (100, [M+Na]+, C37H45NNaO7+
;
;
+
calcd 638.3094), 616.3252 (12, [M+H]+, C37H46NO7
calcd 616.3274), 532.2699 (8), 319.1518 (7). Anal. Calcd
for C37H45NO7 (615.77): C, 72.17; H, 7.37; N, 2.27.
Found: C, 72.04; H, 7.21; N, 2.39.
(b) A suspension of 11 (50 mg, 0.112 mmol) and P2O5
(32 mg, 0.225 mmol) in MeOCH2OMe (0.5 mL) was
stirred for 3 h at 23 ꢁC. Workup and FC as described
in (a) gave 36 (14 mg, 23%) and 35 (25 mg, 46%).
4.24. 5-Amino-3,4,6-tri-O-benzyl-5-deoxy-2-O-pivaloyl-
D-glucono-1,5-thiolactam 34
4.25.1. Data of 35. Colourless oil. Rf (hexane/AcOEt
25
D
2:1) 0.15. ½aŠ ¼ þ158:5 (c 0.80, CHCl3). IR (CHCl3):
A soln of 32 (25 mg, 47.02 lmol) in toluene (0.9 mL)
was treated with Lawessonꢂs reagent (14 mg,
34.61 lmol), stirred for 168 h at 23 ꢁC, diluted with
Et2O (30 mL) and washed with satd NaHCO3 soln
(3 · 15 mL). The combined aq layers were extracted
with Et2O (2 · 10 mL). The combined org. layers were
washed with H2O (25 mL) and brine (25 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
5:1 ! 3:1 ! 2:1) gave 34 (15.0 mg, 58%) and 32
(9.3 mg, 37%).
3388w, 3090w, 3067w, 3031w, 3012m, 2903m, 2866w,
1952w, 1875w, 1810w, 1682s, 1603w, 1497w, 1454m,
1400w, 1363m, 1316m, 1282w, 1259w, 1150s, 1117s,
1102s, 1036s, 1029s, 915w. 1H NMR (CDCl3,
300 MHz): see Table 3; additionally, 3.45 (s, MeO);
3.90 (with virtual coupling, irrad. at 4.19 ! d, J = 8.7,
with virtual coupling); 4.19 (irrad. at 3.90 ! s); 4.43
(d, J = 12.1, PhCH); 4.48 (d, J = 12.1, PhCH); 4.51
(d, J = 11.2, PhCH); 4.83 (d, J = 11.2, PhCH); 4.85 (d,
J = 11.2, PhCH); 4.86 (d, J = 6.5, OCHO); 4.92 (d,

466
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
J = 11.2, PhCH); 5.11 (d, J = 6.5, OCH0O); 6.01 (ex-
change with CD3OD); 7.17–7.22 (m, 2 arom. H); 7.25–
7.40 (m, 13 arom. H). 13C NMR (CDCl3, 75 MHz):
see Table 4; additionally, 56.30 (q, MeO); 73.29, 74.68,
74.87 (3t, 3PhCH2); 97.34 (t, OCH2O); 127.66–128.40
(several d); 137.09, 137.29, 137.80 (3s). HR-MALDI-
(MgSO4), filtered and evaporated. FC (hexane/AcOEt/
MeOH 1:3:0 ! 0:1:0 ! 0:20:1) gave 231,24,25 (21 mg,
77%).
4.26. (5R,6R,7S,8S)-6,7-Bis(benzyloxy)-5-[(benzyl-
oxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]pyridin-8-yl
N,N-diisopropylcarbamate 37 and (5R,6R,7S,8R)-6,7-
bis(benzyloxy)-5-[(benzyloxy)methyl]-5,6,7,8-tetrahydro-
imidazo[1,2-a]pyridin-8-yl N,N-diisopropylcarbamate 38
MS: 514.2191 (100, [M+Na]+, C29H33NNaO6+; calcd
514.2206), 460.2111 (32, [MÀMeO]+, C28H30NO5
;
+
calcd 460.2124). Anal. Calcd for C29H33NO6 (491.58):
C, 70.86; H, 6.77; N, 2.85. Found: C, 70.69; H, 6.75;
N, 2.91.
(a) A soln of 221,24,25 (25 mg, 53.1 lmol), i-Pr2NCOCl
(17.3 mg, 0.106 mmol) and DMAP (12.9 mg,
0.106 mmol) in pyridine (1 mL) was stirred at 120 ꢁC
for 7 h, treated with H2O (1 mL) and evaporated. A soln
of the brown residue in Et2O (50 mL) was washed with
2M aq HCl (3 · 30 mL). The combined aq layers were
extracted with Et2O (2 · 30 mL). The combined org. lay-
ers were washed with water (60 mL) and brine (60 mL),
dried (MgSO4), filtered and evaporated. FC (hexane/
AcOEt 1:1 ! 0:1) gave 37 (13.0 mg, 41%) and 38
(11.3 mg, 36%).
4.25.2. Data of 36. Colourless oil. Rf (hexane/AcOEt
25
D
2:1) 0.38. ½aŠ ¼ þ34:1 (c 0.58, CHCl3). IR (CHCl3):
3090w, 3067w, 3030m, 3012m, 2934m, 2900m, 2866m,
1952w, 1873w, 1810w, 1679s, 1603w, 1496w, 1454m,
1387w, 1363m, 1290w, 1261w, 1172m, 1149m, 1099s,
1071s, 1039s, 1028s, 914m. 1H NMR (CDCl3,
300 MHz): see Table 3; additionally, 3.27, 3.43 (2s,
2MeO); 4.44 (d, J = 12.1, PhCH); 4.48 (d, J = 12.1,
PhCH); 4.57 (d, J = 11.8, PhCH); 4.61 (d, J = 10.6,
NCHO); 4.62 (d, J = 11.8, PhCH); 4.67 (d, J = 11.2,
PhCH); 4.80 (d, J = 11.5, PhCH); 4.86 (d, J = 6.8,
OCHO); 5.00 (d, J = 6.5, OCH0O); 5.10 (d, J = 10.6,
NCH0O); 7.24–7.37 (m, 15 arom. H). 13C NMR (CDCl3,
75 MHz): see Table 4; additionally, 56.10, 56.27 (2q,
2MeO); 72.09, 73.31, 73.72 (3t, 3PhCH2); 75.74 (t,
NCH2O); 96.89 (t, OCH2O); 127.53–128.34 (several d);
(b) A suspension of 221,24,25 (25 mg, 53.1 lmol) and 55%
NaH in oil (4.5 mg, 0.103 mmol) in THF (1 mL) was
stirred at 23 ꢁC for 1 h, cooled to 0 ꢁC and treated with
i-Pr2NCOCl (13.8 mg, 84.3 lmol). The mixture was stir-
red at 0 ꢁC for 15 min and at 23 ꢁC for 6 h, treated with
satd NH4Cl soln (3 mL) at 0 ꢁC, diluted with Et2O
(20 mL) and washed with satd NH4Cl soln
(3 · 15 mL). The combined aq layers were extracted
with Et2O (2 · 15 mL). The combined org. layers were
washed with water (30 mL) and brine (30 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
1:1) gave 37 (30.4 mg, 96%).
137.34, 137.48, 137.84 (3s). HR-MALDI-MS: 558.2459
(100, [M+Na]+, C31H37NNaO7
+
; calcd 558.2468),
calcd
504.2383 (15, [MÀMeO]+, C30H34NO6
;
+
504.2386),
C22H24NO5
382.1643
calcd 382.1654). Anal. Calcd for
(8,
[MÀMOMÀBnOH]+,
+
;
C31H37NO7 (535.64): C, 69.51; H, 6.96; N, 2.61. Found:
C, 69.54; H, 6.96; N, 2.71.
25
D
4.26.1. Data of 37. Rf (AcOEt) 0.40. ½aŠ ¼ þ25:5 (c
4.25.3. Desilylation of 27. A soln of 27 (660 mg,
1.13 mmol) in THF (13 mL) was treated with 1 M
Bu4NF in THF (2.2 mL, 2.20 mmol) and stirred at
23 ꢁC for 3.5 h. The mixture was diluted with Et2O
(30 mL) and washed with satd NH4Cl soln
(3 · 20 mL). The combined aq layers were extracted
with Et2O (2 · 20 mL). The combined org. layers were
washed with water (40 mL) and brine (40 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt/
MeOH 1:3:0 ! 0:1:0 ! 0:20:1) gave 221,24,25 (482 mg,
91%).
0.70, CHCl3). UV (CHCl3): 259 (2.78), 240 (3.24). IR
(CHCl3): 3089w, 3067w, 3032w, 3008m, 2970m,
2935m, 2873m, 1951w, 1873w, 1810w, 1684s, 1604w,
1495m, 1478m, 1454s, 1441m, 1369m, 1342m, 1289s,
1135s, 1086s, 1029m, 910w. 1H NMR (CDCl3,
300 MHz): see Table 6; additionally, 1.02–1.36 (m,
(Me2CH)2N); 3.64–3.84 (m, Me2CH); 3.71 (irrad. at
3.78 ! change, irrad. at 4.39 ! d, J = 10.0); 3.78 (irrad.
at 3.71 ! change, irrad. at 4.39 ! d, J = 10.0); 3.93
(irrad. at 4.21 ! d, J ꢀ 4.0, irrad. at 4.39 ! d, J ꢀ 5.6);
3.96–4.12 (br s, Me2CH); 4.21 (irrad. at 3.93 ! d,
J = 3.7, irrad. at 6.11 ! d, J = 5.3); 4.35–4.42 (irrad. at
3.71 ! change, irrad. at 3.78 ! t, J = 4.0, irrad. at
3.93 ! dd, J = 3.7, 5.9); 4.41 (d, J = 11.8, PhCH); 4.43
(d, J = 11.8, PhCH); 4.46 (d, J = 11.8, PhCH); 4.66 (d,
J = 11.8, PhCH); 4.74 (d, J = 11.8, PhCH); 4.88 (d,
J = 11.8, PhCH); 6.11 (irrad. at 4.21 ! s); 7.15–7.18
(m, 2 arom. H); 7.23–7.37 (m, 13 arom. H). 13C NMR
(CDCl3, 75 MHz): see Table 7; additionally, 20.73,
21.47 (2br q, (Me2CH)2N); 45.81, 46.43 (2br d,
(Me2CH)2N); 72.82, 73.02, 73.31 (3t, 3PhCH2);
127.59–128.37 (several d); 137.25, 137.34, 137.68 (3s);
154.45 (s, C@O). HR-MALDI-MS: 636.2789 (<1,
4.25.4. Desilylation of 31. At 0 ꢁC, a soln of 31
(350 mg, 0.558 mmol) in THF (6 mL) was treated with
1 M Bu4NF in THF (1.1 mL, 1.10 mmol), stirred at
0 ꢁC for 1 h and treated with satd NH4Cl soln
(10 mL). Workup and FC as described for the desilyl-
ation of 27 afforded 221,24,25 (248 mg, 94%).
4.25.5. Desilylation of 28. A soln of 28 (34 mg,
58.1 lmol) in THF (0.7 mL) was treated with 1 M
Bu4NF in THF (0.12 mL, 0.12 mmol) and stirred at
23 ꢁC for 3.5 h. The mixture was diluted with Et2O
(30 mL) and washed with satd NH4Cl soln
(3 · 20 mL). The combined aq layers were extracted
with Et2O (2 · 20 mL). The combined org. layers were
washed with water (40 mL) and brine (40 mL), dried
[M+K]+, C36H43KN3O5+; calcd 636.2840), 620.3084
(17, [M+Na]+, C36H43N3NaO5
;
calcd 620.3100),
+
598.3264 (12, [M+H]+, C36H44N3O5+; calcd 598.3281),
475.1981 (6, [M+Na–i-Pr2NHÀCO2]+, C29H28N2-

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
467
+
NaO3
;
calcd 475.1997), 453.2168 (100, [MÀi-
13C NMR (CDCl3, 75 MHz): see Table 7; additionally,
73.39 (t, PhCH2); 73.47 (t, 2PhCH2); 127.69–128.44 (sev-
eral d); 136.67, 136.85, 136.96 (3s). HR-MALDI-MS:
Pr2NÀCO2]+, C29H29N2O3+; calcd 453.2178).
25
4.26.2. Data of 38. Rf (AcOEt) 0.23. ½aŠ ¼ À41:6 (c
469.2122 (2, [M+H]+, C29H29N2O4+; calcd 469.2127),
D
+
0.43, CHCl3). UV (CHCl3): 259 (2.96), 239 (3.30). IR
(CHCl3): 3089w, 3067w, 3031w, 3003m, 2971m,
2934m, 2872m, 1951w, 1879w, 1810w, 1688s, 1603w,
1496m, 1473m, 1454m, 1441m, 1369m, 1346m, 1291s,
1136s, 1107s, 1080s, 1046m, 1028m, 908w. 1H NMR
(CDCl3, 300 MHz): see Table 8; additionally, 1.02–
1.32 (m, (Me2CH)2N); 3.67–3.83 (m, Me2CH); 4.02–
4.16 (m, Me2CH); 4.05 (irrad. at 4.24 ! change, irrad.
at 6.58 ! d, J = 9.3); 4.07 (irrad. at 4.24 ! change);
4.39 (d, J = 12.8, PhCH); 4.43 (d, J = 12.8, PhCH);
4.55 (d, J = 10.9, PhCH); 4.59 (d, J = 10.6,
PhCH); 4.94 (d, J = 11.2, PhCH); 4.97 (d, J = 11.2,
PhCH); 7.20–7.39 (m, 15 arom. H). 13C NMR (CDCl3,
75 MHz): see Table 7; additionally, 20.76, 21.57 (2br
q, (Me2CH)2N); 45.45, 46.76 (2br d, (Me2CH)2N);
71.79, 73.13, 74.84 (3t, 3PhCH2); 127.53–128.34 (several
457.2118 (2), 451.2013 (1, [MÀOH]+, C29H27N2O3
;
calcd 451.2022), 383.1362 (6, [M+NaÀBnOH]+,
+
C22H20N2NaO3
;
calcd 383.1372), 377.1497 (3,
[MÀBn]+, C22H21N2O3+; calcd 377.1501), 361.1549
+
(100, [MÀBnO]+, C22H21N2O3
;
calcd 361.1552),
;
+
292.0818 (5, [M+NaÀBnOHÀBn]+, C15H13N2NaO3
calcd 292.0824), 270.0996 (6, [MÀBnOÀBn]+,
+
C15H14N2O3
;
calcd
270.1004),
253.0968
(4,
[MÀBnOÀBnOH]+, C15H13N2O2+; calcd 253.0977).
Anal. Calcd for C29H28N2O4Æ0.5H2O (477.56): C,
72.94; H, 6.12; N, 5.87. Found: C, 72.84; H, 5.77; N,
5.87.
4.28. (5R,6R,7S,8S)-8-Azido-6,7-bis(benzyloxy)-5-[(benz-
yloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]pyridine
401,24,25 and (5R,6R,7S,8R)-8-azido-6,7-bis(benzyloxy)-
5-[(benzyloxy)methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]-
pyridine 411
d); 137.32, 137.62, 137.72 (3s); 154.23 (s, C@O). HR-
MALDI-MS: 636.2809 (1, [M+K]+, C36H43KN3O5
;
+
calcd
C36H43N3NaO5
[M+H]+, C36H44N3O5
636.2840),
620.3090
(19,
[M+Na]+,
+
;
calcd 620.3100), 598.3273 (10,
(a) A soln of 221,24,25 (25 mg, 53.1 mol) in toluene
(0.5 mL) was treated successively with diphenyl phos-
phorazidate (= DPPA, 57 lL, 0.264 mmol) and DBU
(40 lL, 0.268 mmol), and stirred at 23 ꢁC for 4 h. The
brown mixture was diluted with CH2Cl2 (30 mL) and
washed with satd NH4Cl soln (3 · 15 mL). The com-
bined aq layers were extracted with CH2Cl2
(2 · 20 mL). The combined org. layers were washed with
H2O (35 mL) and brine (35 mL), dried (MgSO4), filtered
and evaporated. FC (hexane/AcOEt 1:0 ! 3:1 ! 1:1)
gave 40 (17 mg, 65%) and 41 (4 mg, 15%).
+
;
calcd 598.3281), 475.1985
(4, [M+Na–i-Pr2NH–CO2]+, C29H28N2NaO3+; calcd
475.1997),
453.2171
(100,
[MÀi-Pr2N–CO2]+,
C29H29N2O3+; calcd 453.2178).
4.27. (5R,6R,7S)-6,7-Bis(benzyloxy)-5-[(benzyloxy)-
methyl]-5,6,7,8-tetrahydroimidazo[1,2-a]pyridin-8-one 39
At 0 ꢁC, a soln of 221,24,25 (35 mg, 74.4 lmol) in CH2Cl2
(1.5 mL) was treated with 15% soln of Dess–Martin
periodinane in CH2Cl2 (0.31 mL, 0.149 mmol), stirred
at 0 ꢁC for 2 h, treated with a mixture of 10% aq
Na2S2O3 soln/satd NaHCO3 soln 1:1 (3 mL) and stirred
at 23 ꢁC for 10 min. The mixture was diluted with Et2O
(40 mL) and washed with a mixture of 10% aq Na2S2O3
soln/satd NaHCO3 soln 1:1 (3 · 25 mL). The combined
aq layers were extracted with Et2O (2 · 20 mL). The
combined org. layers were washed with water (50 mL)
and brine (50 mL), dried (MgSO4), filtered and evapo-
rated. The crude product (34 mg, a single compound
(b) A soln of 221,24,25 (458 mg, 0.973 mmol) in toluene
(9 mL) was treated successively with DPPA (1.05 mL,
4.87 mmol) and DBU (0.73 mL, 4.89 mmol), and stirred
at 23 ꢁC for 4.5 h. Workup and FC as described in (a)
gave 40 (290 mg, 60%) and a mixture containing mainly
41 and traces of DBU (60 mg). FC (RP-C18 silica gel,
MeOH/H2O 4:1) of this mixture afforded pure 41
(56 mg, 12%).
1
according to the H NMR spectrum) was precipitated
(c) As (a), but with 22/23 55:451,24,25 (35 mg, 74.4 lmol)
instead of 22. Workup and FC gave 40 (23 mg, 62%)
and 41 (4 mg, 11%).
from hexane/AcOEt 2:1 (3 mL) at À50 ꢁC to afford after
drying 39 (25 mg, ca. 72%). Colourless solid containing
substantial amounts of H2O. The sample for microanal-
ysis was dried for 5 days at 10À4 Torr. Rf (AcOEt) 0.41.
(d) As (a), but with 231,24,25 (13 mg, 27.6 lmol) instead
of 22. Workup and FC afforded 40/41 85:15 (9 mg,
66%).
25
D
Mp 90–93ꢁ. ½aŠ ¼ À12:1 (c 0.97, CHCl3). UV (CHCl3):
292 (4.11). IR (CHCl3): 3375w (br), 3156w, 3089w,
3067w, 3033m, 3012m, 2920w, 2869m, 1953w, 1877w,
1810w, 1696s, 1603w, 1508w, 1497m, 1455s, 1403s,
1364m, 1326w, 1278w, 1219w, 1149m, 1092s, 1043m,
(e) As (a), but with di-(p-nitrophenyl)phosphorazidate
(97 mg, 0.266 mmol) instead of DPPA. Workup and
FC gave 40 (14 mg, 53%) and 41 (3 mg, 11%).
1
1029m, 922w, 911w. H NMR (CDCl3, 300 MHz): see
Table 6; additionally, 3.81 (irrad. at 4.50 ! d,
J = 10.2); 3.87 (irrad. at 4.50 ! d, J = 10.3); 4.13 (irrad.
at 4.22 ! change, irrad. at 4.50 ! d, J = 6.2); 4.22
(irrad. at 4.13 ! s); 4.43 (d, J = 11.8, PhCH); 4.48 (d,
J = 12.1, PhCH); 4.46–4.54 (irrad. at 4.13 ! change);
4.53 (d, J = 11.8, PhCH); 4.67 (d, J = 11.5, PhCH);
4.75 (d, J = 11.5, PhCH); 4.96 (d, J = 11.8, PhCH);
7.14–7.25 (m, 4 arom. H); 7.28–7.40 (m, 11 arom. H).
(f) A suspension of 221,24,25 (25 mg, 53.1 lmol) and 55%
NaH in oil (4.1 mg, 94.0 lmol) was stirred at 23 ꢁC for
1 h, cooled to 0 ꢁC and treated with DPPA (35 lL,
0.162 mmol). The mixture was stirred at 0 ꢁC for
10 min and at 23 ꢁC for 10 h, diluted with Et2O
(30 mL) and washed with satd NH4Cl soln
(3 · 15 mL). The combined aq layers were extracted

468
M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
with Et2O (2 · 15 mL). The combined org. layers were
washed with H2O (30 mL) and brine (30 mL), dried
(MgSO4), filtered and evaporated. FC (hexane/AcOEt
1:0 ! 3:1 ! 1:1) gave 40 (7 mg, 27%) and 40/41 1:1
(3 mg, 11%).
Table 9; additionally, 4.008 (irrad. at 3.88 ! d,
J = 3.1); 4.009 (irrad. at 4.22 ! d, J = 4.0); 4.14 (irrad.
at 3.88 ! br t, J ꢀ 4.1, irrad. at 4.22 ! change); 7.02,
7.16 (2d, J = 1.2, H–C(2), H–C(3)). 1H NMR (D2O,
300 MHz, 1 equiv of CF3CO2H): see Table 9; addition-
1
ally, 7.53, 7.67 (2d, J = 1.9, H–C(2), H–C(3)). H NMR
4.28.1. Data of 40. Colourless oil. Rf (hexane/AcOEt
(D2O, 300 MHz, 2 equiv of CF3CO2H): see Table 9;
additionally, 7.62, 7.75 (2d, J = 2.2, H–C(2), H–C(3)).
13C NMR (D2O, 75 MHz): 46.77 (d, C(8)); 61.42 (t,
CH2–C(5)); 62.10 (d, C(5)); 68.92, 69.56 (2d, C(6),
C(7)); 118.32 (d, C(3)); 127.61 (d, C(2)); 147.06
(s, C(8a)). HR-MALDI-MS: 222.0849 (46, [M+Na]+,
C8H13N3O3++; calcd 222.0855), 200.1030 (100, [M+H]+,
25
D
1:1) 0.29. ½aŠ ¼ þ58:6 (c 1.19, CHCl3) (lit.:24,25 +59.0
(CHCl3)). UV (CHCl3): 258 (2.88), 240 (3.36). IR
(CHCl3): 3159w, 3089w, 3067w, 3033w, 3011w, 2955w,
2920w, 2870w, 2109s, 1951w, 1877w, 1810w, 1589w,
1523w, 1496w, 1488w, 1454m, 1362m, 1331w, 1308w,
1282m, 1259w, 1170w, 1142m, 1086m, 1028w, 1011w,
1
940w, 909w. H NMR (CDCl3, 300 MHz): see Table
C8H14N3O3
;
calcd
200.1035),
183.0757
(53,
6. 13C NMR (CDCl3, 75 MHz): see Table 7. HR-
[MÀNH2]+, C8H11N2O3+; calcd 183.0757). HR-ESI-
MALDI-MS: 607.2436 (16), 518.2166 (5, [M+Na]+,
MS: 421.1785 (95, [2M+Na]+, C16H26N6NaO6+; calcd
C29H29N5NaO3
;
calcd 518.2168), 496.2339 (66,
421.1812), 222.0852 (18, [M+Na]+, C8H13N3O3+; calcd
+
[M+H]+, C29H30N5O3+; calcd 496.2349), 490.2099 (14,
222.0855), 200.1027 (100, [M+H]+, C8H14N3O3
calcd 200.1035).
;
+
+
[M+NaÀN2]+, C29H29N3NaO3
;
calcd 490.2107),
;
470.2446 (20, [M+3HÀN2]+, C29H32N3O3
calcd
;
+
470.2444), 468.2278 (43, [M+HÀN2]+, C29H30N3O3
4.30. Inhibition studies
+
calcd 468.2287), 455.2328 (65, [M+2HÀN3]+,
+
C29H31N2O3
;
calcd 455.2235), 453.2170 (100,
Determination of the inhibition constants (Ki) or the
IC50 values was performed with a range of inhibitor con-
centrations (typically 4–7 concentrations), which bracket
the Ki or IC50 value, and substrate concentrations,
which bracket the KM of each enzyme (for Ki, typically
5–7 concentrations), or correspond to it (for IC50).
[MÀN3]+, C29H29N2O3+; calcd 453.2178), 361.1541 (11).
4.28.2. Data of 41. Colourless oil. Rf (hexane/AcOEt
25
D
1:1) 0.13. ½aŠ ¼ À49:8 (c 1.00, CHCl3). UV (CHCl3):
258 (2.86), 241 (3.43). IR (CHCl3): 3159w, 3090w,
3067w, 3032w, 3011m, 2930w, 2869m, 2104s, 1951w,
1877w, 1810w, 1603w, 1525w, 1496m, 1454m, 1363m,
1310m, 1268m, 1168w, 1131m, 1102s, 1082s, 1028m,
(a) Inhibition of snail b-mannosidase. KM = 0.48–
0.56 mM (Ref. 48: KM = 0.42–0.80 mM). IC50 and Ki
values were determined at 25 ꢁC at an enzyme concen-
tration of 0.048 units/mL, using a 0.04 M acetate buffer
(pH 4.5 or 5.5) and 4-nitrophenyl b-D-mannopyranoside
as the substrate. The enzymatic reaction was started
after incubation of the enzyme (100 lL) in presence of
the inhibitor (50 lL) during 1 h at 25 ꢁC, by the addition
of the substrate (50 lL). The enzyme reaction was
quenched by addition of 0.2 M borate buffer (pH 9.0,
100 lL) after 5 min and the absorption at 405 nm was
taken as rate for the hydrolysis of the substrate after
subtraction of the absorption of a blank probe (H2O,
buffer, substrate). IC50 Values were determined by plot-
ting the reciprocal value of the rate of substrate hydro-
lysis versus 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 plots60 and plotting them
versus the inhibitor concentrations.61 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 from the Lineweaver–Burk plots versus the
inhibitor concentrations.61 After fitting a straight line
to the data by linear regression, the negative [I]-intercept
of this plot provided the appropriate aÆKi value.
1
1013m, 909w. H NMR (CDCl3, 300 MHz): see Table
8. 13C NMR (CDCl3, 75 MHz): see Table 7. HR-
MALDI-MS: 607.2431 (18), 518.2167 (10, [M+Na]+,
+
C29H29N5NaO3
;
calcd 518.2168), 496.2338 (89,
[M+H]+, C29H30N5O3+; calcd 496.2349), 490.2098 (20,
[M+Na–N2]+, C29H29N3NaO3
;
calcd 490.2107),
calcd
+
470.2442 (55, [M+3HÀN2]+, C29H32N3O3
;
+
+
470.2444), 468.2276 (72, [M+HÀN2]+, C29H30N3O3
;
calcd 468.2287), 455.2325 (73, [M+2HÀN3]+,
+
C29H31N2O3
;
calcd 455.2235), 453.2166 (100,
[MÀN3]+, C29H29N2O3+; calcd 453.2178), 361.1547 (18).
4.29. (5R,6R,7S,8R)-8-Amino-5,6,7,8-tetrahydro-5-
(hydroxymethyl)imidazo[1,2-]pyridine-6, 7-diol 8
A soln of 41 (63 mg, 0.127 mmol) in AcOH (3 mL) was
treated with 10% Pd/C (60 mg) and hydrogenated at
6 bar for 60 h. The suspension was filtered through Cel-
ite, and the residue was washed with MeOH (20 mL).
Evaporation of the combined filtrates, co-evaporation
with toluene (4 · 5 mL), ion-exchange chromatography
+
(Amberlite CG-120, NH4 form, elution with 0.05 M
aq NH3) and lyophilization afforded 8 (16.5 mg, 65%).
Colourless hygroscopic solid. Rf (CHCl3/MeOH/
25
D
NH4OH 5:4:1) 0.35. ½aŠ ¼ À30:1 (c 0.85, H2O).
pKHA = 7.02 (no additional pK value was observed in
the pH range 2.9–10.3). UV (H2O): 216 (3.67), 192
(3.61). IR (0.4% in KBr): 3600–2100s (br), 3356s,
3277s, 3162s, 3073s, 2913s, 2841s, 2695s, 1591m,
1484s, 1445s, 1384m, 1336m, 1323m, 1289m, 1277m,
1219w, 1169w, 1154w, 1131w, 1088m, 1061s, 1022s,
959w, 932w, 881m. 1H NMR (D2O, 300 MHz): see
(b) Inhibition of Jack bean a-mannosidase. KM
=
2.27 mM (Ref. 51: KM = 2.5 mM; Ref. 48: KM = 1.8–
2.8 mM). As described in (a), inhibition studies were
carried out at 37 ꢁC at an enzyme concentration of
0.086 units/mL, using a 0.04 M acetate buffer (pH 4.5),

M. Terinek, A. Vasella / Tetrahedron: Asymmetry 16 (2005) 449–469
469
25. Tatsuta, K.; Miura, S.; Ohta, S.; Gunji, H. J. Antibiot.
1995, 48, 286.
26. Mitsunobu, O. Synthesis 1981, 1.
27. Loibner, H.; Zbiral, E. Helv. Chim. Acta 1976, 59, 2100.
28. Loibner, H.; Zbiral, E. Helv. Chim. Acta 1977, 60, 417.
29. Tatsuta, K.; Miura, S.; Gunji, H. Bull. Chem. Soc. Jpn.
1997, 70, 427.
containing 1.5 mmol of ZnCl2 and 4-nitrophenyl a-D-
mannopyranoside as the substrate. The enzymatic
reaction was started after the incubation at 37 ꢁC for
1 h.
Acknowledgements
30. Jesberger, M.; Davis, T. P.; Barner, L. Synthesis 2003,
1929.
We thank M. Schneider for the pKHA determinations,
Dr. M. Linke for a generous sample of 16, Dr. B. Bernet
for modelling and for checking the experimental part
and the Swiss National Science Foundation, Oxford
Glycosciences Ltd., Abingdon (UK) and F. Hoffmann-
La Roche AG, Basel (CH), for generous support.
31. Granier, T.; Panday, N.; Vasella, A. Helv. Chim. Acta
1997, 80, 979.
32. Corey, E. J.; Venkateswarlu, A. J. Am. Chem. Soc. 1972,
94, 6190.
33. Corey, E. J.; Hua, D. H.; Pan, B.-C.; Seitz, S. P. J. Am.
Chem. Soc. 1982, 104, 6818.
34. Deng, Y.; Hlasta, D. J. Tetrahedron Lett. 2002, 43, 189.
35. Hammerschmidt, F.; Hanninger, A.; Simov, B. P.; Vo¨l-
lenkle, H.; Werner, A. Eur. J. Org. Chem. 1999, 3511.
36. Peters, J. G.; Seppi, M.; Fro¨hlich, R.; Wibbeling, B.;
Hoppe, D. Synthesis 2002, 381.
References
1. Panday, N.; Meyyappan, M.; Vasella, A. Helv. Chim. Acta
2000, 83, 513.
2. Ducros, V. M.-A.; Zechel, D. L.; Murshudov, G. N.;
37. Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.
38. Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113,
7277.
´
Gilbert, H. J.; Szabo, L.; Stoll, D.; Withers, S. G.; Davies,
39. Costanzo, M. J.; Maryanoff, B. E.; Hecker, L. R.; Schott,
M. R.; Yabut, S. C.; Zhang, H.-C.; Andrade-Gordon, P.;
Kauffman, J. A.; Lewis, J. M.; Krishnan, R.; Tulinsky, A.
J. Med. Chem. 1996, 39, 3039.
40. Tao, M.; Bihovsky, R.; Kauer, J. C. Bioorg. Med. Chem.
Lett. 1996, 6, 3009.
41. Holden, K. G.; Mattson, M. N.; Cha, K. H.; Rapoport,
H. J. Org. Chem. 2002, 67, 5913.
42. Lichtenthaler, F. W.; Lergenmuller, M.; Peters, S.; Varga,
¨
Z. Tetrahedron: Asymmetry 2003, 14, 727.
43. Thompson, A. S.; Humphrey, G. R.; DeMarco, A. M.;
Mathre, D. J.; Grabowski, E. J. J. J. Org. Chem. 1993, 58,
5886.
44. Panday, N.; Granier, T.; Vasella, A. Helv. Chim. Acta
1998, 81, 475.
45. Mizuno, M.; Shioiri, T. Chem. Commun. 1997, 2165.
46. Shioiri, T.; Yamada, S.-I. Chem. Pharm. Bull. 1974, 22,
855.
47. Panday, N.; Vasella, A. Synthesis 1999, 1459.
48. Terinek, M.; Vasella, A. Helv. Chim. Acta 2003, 86,
3482.
G. J. Angew. Chem., Int. Ed. 2002, 41, 2824.
3. McCleary, B. V. Carbohydr. Res. 1983, 111, 297.
4. Patchett, M. L.; Daniel, R. M.; Morgan, H. W. Biochem.
J. 1987, 243, 779.
5. Dale, M. P.; Ensley, H. E.; Kern, K.; Sastry, K. A. R.;
Byers, L. D. Biochemistry 1985, 24, 3530.
6. Mohamadi, F.; Richards, N. G. J.; Guida, W. C.;
Liskamp, R.; Caufield, C.; Lipton, M.; Chang, G.;
Hendrickson, T.; Still, W. C. J. Comput. Chem. 1990, 11,
440.
7. Heightman, T. D.; Vasella, A.; Tsitsanou, K. E.; Zogra-
phos, S. E.; Skamnaki, V. T.; Oikonomakos, N. G. Helv.
Chim. Acta 1998, 81, 853.
8. Heightman, T. D.; Vasella, A. T. Angew. Chem., Int. Ed.
1999, 38, 750.
9. Nishimura, Y.; Adachi, H.; Satoh, T.; Shitara, E.;
Nakamura, H.; Kojima, F.; Takeuchi, T. J. Org. Chem.
2000, 65, 4871.
10. Jung, M. E.; Lam, P. Y.-S.; Mansuri, M. M.; Speltz, L. M.
J. Org. Chem. 1985, 50, 1087.
11. Takada, D.; Suemune, H.; Sakai, K. Chem. Pharm. Bull.
1990, 38, 234.
12. Barbee, T. R.; Albizati, K. F. J. Org. Chem. 1991, 56,
6764.
13. Draper, R. W.; Hu, B.; McPhail, A. T.; Puar, M. S.; Vater,
E. J.; Weber, L. Tetrahedron 1999, 55, 3355.
14. Wendler, N. L.; Taub, D. J. Org. Chem. 1960, 25, 1828.
49. Terinek, M.; Vasella, A. Helv. Chim. Acta 2004, 87, 719.
50. Terinek, M.; Vasella, A. Helv. Chim. Acta 2004, 87, 3035.
51. Li, Y. T. J. Biol. Chem. 1967, 242, 5474.
52. Shepherd, V.; Montgomery, R. Biochim. Biophys. Acta
1976, 429, 884.
53. Howard, S.; Braun, C.; McCarter, J.; Moremen, K. W.;
Liao, Y.-F.; Withers, S. G. Biochem. Biophys. Res.
Commun. 1997, 238, 896.
54. Howard, S.; He, S.; Withers, S. G. J. Biol. Chem. 1998,
273, 2067.
55. Overkleeft, H. S.; van Wiltenburg, J.; Pandit, U. K.
Tetrahedron 1994, 50, 4215.
56. Knight, J. G.; Tchabanenko, K. Tetrahedron 2003, 59,
281.
57. Niwa, T.; Tsuruoka, T.; Goi, H.; Kodama, Y.; Itoh, J.;
Inouye, S.; Yamada, Y.; Niida, T.; Nobe, M.; Ogawa, Y.
J. Antibiot. 1984, 37, 1579.
´
´
15. Patonay, T.; Patonay-Peli, E.; Litkei, G.; Szilagyi, L.;
Batta, G.; Dinya, Z. J. Heterocycl. Chem. 1988, 25, 343.
16. Konosu, T.; Miyaoka, T.; Tajima, Y.; Oida, S. Chem.
Pharm. Bull. 1991, 39, 2241.
17. Yunker, M. B.; Fraser-Reid, B. J. Chem. Soc., Chem.
Commun. 1978, 325.
18. Postel, D.; Nguyen Van Nhien, A.; Marco, J. L. Eur. J.
Org. Chem. 2003, 3713.
19. Opitz, G. Angew. Chem., Int. Ed. Engl. 1967, 6, 107.
20. King, J. F. Acc. Chem. Res. 1975, 8, 10.
21. Granier, T.; Vasella, A. Helv. Chim. Acta 1998, 81, 865.
58. Fleet, G. W. J.; Ramsden, N. G.; Witty, D. R. Tetrahedron
1989, 45, 319.
22. Granier, T. Dissertation No. 12395, ETH Zurich, 1998.
¨
23. van den Berg, R. J. B. H. N.; Noort, D.; Milder-Enacache,
E. S.; van der Marel, G. A.; van Boom, J. H.; Benschop,
H. P. Eur. J. Org. Chem. 1999, 2593.
24. Tatsuta, K.; Miura, S.; Ohta, S.; Gunji, H. Tetrahedron
Lett. 1995, 36, 1085.
59. Shing, T. K. M. J. Chem. Soc., Chem. Commun. 1988,
1221.
60. Lineweaver, H.; Burk, D. J. Am. Chem. Soc. 1934, 56,
658.
61. Segel, I. H. Enzyme Kinetics; Wiley: New York, 1975.