Preparation of 2-C- and 3-C-cyano-2-enopyranoside derivatives and their epoxidation

Article abstract of DOI:10.1016/j.carres.2007.07.008

Methyl 4,6-O-benzylidene-2-C- and 3-C-cyano-2,3-dideoxy-d-erythro-hex-2-enopyranosides and -2-enitols were prepared and their epoxidation was performed.

Full text of DOI:10.1016/j.carres.2007.07.008

Carbohydrate Research 342 (2007) 2339–2353
Preparation of 2-C- and 3-C-cyano-2-enopyranoside derivatives
and their epoxidation
Tohru Sakakibara,* Shinya Narumi, Ichiro Matsuo, Saeko Okada and
Takanori Nakamura
Graduate School of Integrated Science, Yokohama City University, Seto Kanazawa-ku, Yokohama 236 Japan, Yokohama 236, Japan
Received 2 June 2007; received in revised form 2 July 2007; accepted 6 July 2007
Available online 18 July 2007
Abstract—Methyl 4,6-O-benzylidene-2-C- and 3-C-cyano-2,3-dideoxy-D-erythro-hex-2-enopyranosides and -2-enitols were prepared
and their epoxidation was performed.
Ó 2007 Elsevier Ltd. All rights reserved.
Keywords: Cyanosugar; Epoxidation; Stereochemistry; Nucleophilic addition reaction
1. Introduction
important in the case of a-nitro alkenes. Different from
a nitro group, a linear cyano group gives no A(1,3) strain
in an anionic intermediate; therefore, Factor 3 plays a
relatively unimportant role in the case of cyano
alkenes. To confirm this speculation we prepared 2-
C-cyano-2-enopyranosides 3 and 4, 3-C-cyano-2-eno-
pyranosides 14 and 15, and 3-C-cyano-2-enitol 16 and
performed their epoxidation.
The epoxidation of methyl 3-nitro-b-D-2-enopyran-
oside and its a-anomer exclusively afforded the allo and
mannopyranoside, respectively, in high yields.6 How-
ever, in methyl 2-nitro-a-D-2-enopyranoside, the direc-
tion of approach strongly depended on the peroxides
used. Hydrogen peroxide afforded the allo isomer,
whereas tert-butyl hydroperoxide gave the manno iso-
mer.7 Therefore, we used these two peroxides herewith.
Furthermore, epoxidation of methyl 2-C-nitro-b-D-2-
enopyranoside (36) and 1,5-anhydro-3-C-nitro-hex-2-
enitol (38) have been performed for comparison with
those of the corresponding cyano alkenes.
Nucleophilic addition reactions of a carbon–carbon
double bond activated by an electron-withdrawing
group is one of the most fundamental reactions in
organic chemistry. It is used in a wide range of organic
syntheses.1
The stereochemistry of these reactions has been sub-
jected to considerable research over the past years.2 In
a six-membered substrate one of the authors proposed
that an approaching direction of a nucleophile seems
to be controlled by the following three factors.3 (1) Fac-
tor 1: The nucleophile comes from the opposite side of
substituent(s), especially at the c-position of an acti-
vated carbon–carbon double bond, due to steric and
electrostatic repulsion. (2) Factor 2: It approaches from
the axial side to lead a chair-like intermediate due to ste-
reoelectronic control;4 otherwise a boat like intermedi-
ate is formed. (3) Factor 3: It attacks from the same
side of a substituent at the b0-position (syn addition) if
an intermediate anion suffers from steric (A(1,3) strain5)
and electrostatic repulsion (Fig. 1). Factor 3 should be
Electrostatic repulsion still remains, because distance between the
anionic nitrogen atom of cyano group and the substituent at the b0-
position is longer in the case of syn addition than that of anti
addition.
*
Corresponding author. Tel.: +81 45 787 2183; fax: +81 45 2413;
e-mail: baratoru@yokohama-cu.ac.jp
0008-6215/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2007.07.008

2340
T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
Me
O
O
Nu
Me
Me
N
O
O
O
β
O
α
OMe
β'
axial
equatorial
attack
O
O
Ph
γ
N
O
attack
NO2
O
O
Ph
O
Nu
O
Ph
O
Me
O
Nu
N
C
O
O
O
OMe
axial
equatorial
attack
O
O
Ph
C
O
attack
CN
O
N
Ph
Nu
O
O
Ph
Figure 1.
2. Results and discussion
2.1. Preparation of a-cyano olefins
Methyl 3-C-cyano-a-D-2-enopyranoside 14 was syn-
thesized through diaxial opening of epoxide 5 with
hydrogen cyanide in the presence of either triethyl alu-
minum,10 cyanotrimethylsilane using aluminum chloride
as a catalyst,11 or diethyl aluminum cyanide.12 Epoxide
5 was opened with triethyl aluminum and hydrogen cya-
nide, followed by the epimerization of the cyano group,
mesylation of the hydroxyl group, and elimination of
Methyl 2-C-cyano-a- 38 and -b-D-2-enopyranosides 49
were prepared from the corresponding 3-C-nitro-2-eno-
pyranosides 1 and 2, respectively, by the previously
reported method with slight modification (Scheme 1).
O
O
Ph
Ph
R1
R1
O
O
O
O
2
CN
R2
O2N
R
1 R1 = H, R2 = OMe
2 R1 = OMe, R2 = H
3 R1 = H, R2 = OMe
4 R1 = OMe, R2 = H
OH
O
O
Ph
R1
O
O
O
Et3Al, HCN
R1
MsCl, Et3N
THF, -20 oC
Ph
O
O
THF, rt., 20 h
R2
R2
NC
5 R1 = H, R2 = OMe
6 R1 = OMe, R2 = H
7 R1 = R2 = H
8 R1 = H, R2 = OMe
9 R1 = OMe, R2 = H
10 R1 = R2 = H
OMs
O
O
Ph
O
R1
R1
O
Ph
O
DBU
CHCl3, -20 oC
O
R2
R2
NC
NC
14 R1 = H, R2 = OMe
15 R1 = OMe, R2 = H
16 R1 = R2 = H
11 R1 = H, R2 = OMe
12 R1 = OMe, R2 = H
13 R1 = R2 = H
Scheme 1.

T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
2341
methanesulfonic acid. When the epimerization was
omitted, the yield was improved.
Similar reaction of 2,3-anhydro-b-D-mannopyran-
oside 6 and 1,5:2,3-dianhydro-D-mannitol 7 afforded
3-C-cyano-b-D-2-enopyranoside 15 and 1,5-anhydro-3-
C-cyano-2-enitol 16, respectively (Scheme 1).
conditions, unreacted 3 was recovered in 90% yield,
along with 2-C-cyano-allopyranoside 19 in 7% yield,
suggesting that the cyano group converted into the carb-
amoyl group after epoxidation. In fact cyano epoxide
19 was changed to carbamoyl epoxide 18 under the con-
ditions employed for preparation of 17 and 18. Epoxida-
tion with 70% tert-butyl hydroperoxide afforded
2,3-anhydro-2-C-cyano-mannopyranoside 20, the carb-
amoyl derivatives having manno 17, and allo configura-
tion 18 in 38, 8, and 40% yields, respectively (Scheme 2).
Epoxidation of b-D-anomer 4 with hydrogen peroxide
for 30 min provided a 1:2 mixture of the carbamoyl
derivatives having manno configuration 21 and allo con-
figuration 22 in 89% yield. When the reaction time was
reduced to 5 min, a complicated mixture was obtained
from which a small amount of hydroperoxide 23 having
the gluco configuration was isolated. Hydroperoxide 23
was converted into carbamoyl epoxide 21 by treatment
with tert-butyl hydroperoxide in the presence of M
sodium hydroxide. Similar epoxidation of 4 with tert-
butyl hydroperoxide for 6.5 h afforded manno epoxide
21 in high yield. A complicated mixture was obtained
when the reaction was stopped at 5 min. Besides
2.2. Epoxidation
Thus prepared cyano alkenes 3, 4, and 14–16 were each
separately treated with hydrogen peroxide and bulky
tert-butyl hydroperoxide in the presence of aqueous M
sodium hydroxide.
Treatment of 2-C-cyano-a-D-2-enopyranosides 3 with
30% aqueous hydrogen peroxide for 30 min afforded a
1:6 mixture of manno 17 and allo epoxides 18 having a
carbamoyl group. Under the conditions employed, the
cyano group was hydrated to the carbamoyl group.
From the steric aspect it is very important whether the
hydration of the cyano group occurred after or before
the epoxidation, because the carbamoyl group is not
linear and suffers from A(1,3) strain in the anionic inter-
mediate. When the reaction was performed under mild
Ph
CONH2
O
Ph
O
O
O
O
O
O
+
OMe
CONH2
OMe
O
18 (80%)
17 (13%)
1,4-dioxane, 30 min
O
H2O2, 1M NaOH
3 (~90%)
Ph
O
O
+
H2O2, 1M NaOH
1,4-dioxane, 10 min
OMe
Ph
CN
O
O
O
CN
3
OMe
O
19 (7%)
1,4-dioxane, 3 h
t-BuOOH, 1M NaOH
Ph
O
O
O
O
18 (40%)
+
+
17 (8%)
OMe
CN
20 (38%)
Scheme 2.

2342
T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
carbamoyl derivative 21, peroxides 24 and 25 were
detected, from which peroxide 25 having the manno con-
figuration was isolated, but its 2-epimer 24 was not.
Treatment of peroxide 25 with tert-butyl hydroperoxide
for 6.5 h gave carbamoyl epoxide 21 in high yield. Cya-
no epoxide 26, together with its allo isomer 27, became
the major product from the treatment of 4 with m-chloro-
peroxybenzoic acid in the presence of M sodium
hydroxide. Although cyano epoxide 27 was not isolated,
its 2-epimer 26 was isolated (Scheme 3).
Reaction of 3-C-cyano-a-D-2-enopyranoside 14 with
hydrogen peroxide afforded 2,3-anhydro-3-C-carbam-
oyl-a-D-mannopyranoside 28 in 74% yield. When
compound 14 was treated with tert-butylperoxide,
2,3-anhydro-3-C-cyano-a-D-mannopyranoside 29 was
formed in 94% yield. Epoxide 29 was converted into car-
bamoyl derivative 28 by treatment with hydrogen perox-
ide in the presence of M sodium hydroxide (Scheme 4).
Treatment of 3-C-cyano-b-D-2-enopyranoside 15 with
hydrogen peroxide for 1 h afforded the carbamoyl deriv-
ative 30 having the b-D-allo structure in almost quanti-
tative yield. When the reaction was stopped at 3 min,
3-cyano-2-hydroperoxy-b-D-glucopyranoside 31 (20%)
was isolated besides the starting material 15 (62%) and
the carbamoyl derivative 30 (18%). Compound 30 was
also obtained in 89% yield by treatment of 15 with
tert-butyl hydroperoxide for 40 min (Scheme 5).
Similar treatment of 1,5-anhydro-3-C-cyano-2-enitol
16 with hydrogen peroxide for 1 h at room temperature
caused double-bond migration instead of epoxidation,
giving glucal 32 (30%) and allal 33 (60% yield). When
the reaction temperature was lowered to 0 °C, the epox-
O
Ph
O
O
21
+
26
+
+
OMe
27
+
+
22
4
HOO
CN
23
1,4-dioxane, 5 min
H2O2, 1M NaOH
Ph
Ph
CONH2
O
O
O
OMe
O
O
OMe
O
O
+
Ph
O
O
O
CONH2
21 (30%)
O
OMe
22 (59%)
CN
4
21 (98%)
t-BuOOH, 1M NaOH
1,4-dioxane, 5 min
Ph
Ph
O
CN
O
O
OMe
+
O
O
Ph
O
O
O
O
+
4 (3%)
21 (36%)
+
O
+
OMe
OMe
tBuOO
tBuOO
CN
CN
26 (4%)
25 (34%)
24 (22%)
Ph
CN
O
OMe
O
MCPBA, 1M NaOH
1,4-dioxane, 4 h
O
26
4
4
+
+
O
27
:
:
2.3
1
4.8
Scheme 3.

T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
2343
Ph
O
O
Ph
O
O
O
O
H2O2, 1M NaOH
O
OMe
H2NCO
28 (73%)
1,4-dioxane, 2 h, 74%
NC
OMe
14
1,4-dioxane
30 min, 94%
H2O2, 1M NaOH
Ph
O
O
O
t-BuOOH, 1M NaOH
O
1,4-dioxane, 20 min
OMe
NC
29 (94%)
Scheme 4.
Ph
O
Ph
O
OMe
O
O
O
OMe
H2O2, 1M NaOH
1,4-dioxane,1 h
O
H2NC(O)
NC
O
15
30 (98%)
H2O2, 1M NaOH
1,4-dioxane, 3 min
30 (89%)
O
Ph
O
+
15 (62%)
30 (18%)
O
+
OMe
NC
OOH
31 (20%)
Scheme 5.
idation smoothly proceeded to yield amides 34 and 35
having the manno and allo configurations in 73% and
10% yields, respectively. However, the double-bond
migration exclusively occurred by treatment of 16 with
tert-butyl hydroperoxide even at 0 °C (Scheme 6).
For comparison with these results, epoxidations of
methyl 2-C-nitro-b-D-2-enopyranoside 36 and 1,5-anhy-
dro-3-C-nitro-D-2-enitol 38 were performed. Treatment
of 36 with hydrogen peroxide or tert-butyl hydroperox-
ide exclusively gave epoxide 37 having the manno config-
uration. Epoxidation of 38 became slightly complicated,
because the hydroperoxide was detected at short reac-
tion time, and epoxide 40 was almost completely decom-
posed on silica gel column chromatography. When the
reaction was performed at room temperature and imme-
diately stopped (30 s), epoxides 39 and 40 with the man-
no and allo configurations as well as hydroperoxides 41
and 42 having the gluco and manno configurations were
formed in a ratio of 1:7.6:4:6. This ratio changed to
1:2.3:0.8:1 after 5 min. When hydroperoxide 42 having
the manno configuration was similarly treated for
2 min, a 1:1.3 mixture of epoxide 39 and the starting per-
oxide 42 was obtained, but alternative epoxide 40 and
3-epimer 41 were not detected, indicating that the per-
oxidation is irreversible, that is, the present reaction
was kinetically controlled (Scheme 7).
2.3. Configurational determination of the epoxides
The axial and equatorial anomeric protons of 1,5:2,3-
dianhydro sugars as well as the anomeric proton (a
singlet) and the benzylidene methine proton (a singlet)
of 2-substituted 2,3-anhydro sugars isolated were also
assigned by a NOESY spectrum. All chemical shifts

2344
T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
Ph
O
Ph
O
O
O
O
O
O
Ph
t-BuOOH, 1M NaOH
THF, 0 oC, 2 d
O
NC
O
H2NCO
O
H2O2, 1M NaOH
THF, 0 oC, 4 d
32 (53%)
34 (73%)
NC
+
+
16
Ph
O
Ph
O
O
O
O
O
H2NC(O)
O
CN
33 (27%)
35 (10%)
Scheme 6.
Ph
Ph
O
O
OMe
OMe
O
O
O
O
O
t-BuOOH, 1M NaOH
H2O2, 1M NaOH
37 (92%)
1,4-dioxane, 20 min
NO2
1,4-dioxane, 10 min
NO2
36
37 (75%)
Ph
O
O
O
t-BuOOH, 1M NaOH
1,4-dioxane, 2 min
39
40
+
:
O2N
1.8
1
38
1,4-dioxane, 30 s
H2O2, 1M NaOH
OOH
O
Ph
+
O
O
O
Ph
O
O
Ph
+
Ph
+
O
O
O
O
O
O
O
O2N
O2N
O2N
OOH
O
O2N
42
40
41
39
:
4
:
:
1
7.6
6
H2O2, 1M NaOH
+
:
42
39
42
1,4-dioxane, 2 min
1.3
1
Scheme 7.
described herewith were recorded in chloroform-d with
tetramethylsilane as an internal standard (Tables 1–3).
Large coupling constants, J4,5 8.9–9.6 Hz, suggested
that all of the 2,3-anhydro sugars thus prepared expect-
and H-2 or H-3 and H-4 is almost 90°; therefore, the
coupling constant is close to 0 Hz.6,13,14 This rule could
be applied to 1,5-anhydro- and methyl 2,3-anhydro-
hexopyranosides except 2,3-anhydro-3-C-carbamoyl-b-
D-allopyanosides 22, in which the J3,4 value was 0 Hz.
Instead of the coupling constants, therefore, the chemi-
cal shift of H-1 (d 5.62 for 21 and d 5.05 for 22) was used
for assignment of the manno and allo configurations for
0
edly adopt the H5 conformation.13 The configurations
of the epoxides thus prepared were determined as fol-
lows. In the 2,3-anhydro compounds the dihedral angle
of trans relationship of hydrogen atoms between H-1

T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
2345
Table 1. 1H NMR (400 MHz, CDCl3) chemical shift (d in ppm) of 2,3-anhydrohexopyranosides
17
18
19a
20
21
22
26
27
28
29a
30
37
H-1
H-2
H-3
H-4
H-5
H-6a
H-6e
5.05(s)
—
3.56(s)
3.77(d)
3.85(td) 4.04(td)
3.77(t) 3.72(t)
4.28(dd) 4.28(dd)
5.60(s)
—
4.58(s)
—
4.97(s)
—
5.62(s)
—
5.05(s)
—
3.76(s)
4.37(d)
5.10(br s) 4.96(s)
4.93(s)
3.44(s)
—
4.18(s)
3.03(s)
—
4.88(s)
3.43(s)
—
5.90(s)
—
3.98(s)
3.76(d)
—
—
3.62(br s)b 3.15(br s)
3.90(br s) 3.61(s)
3.65(d) 3.75(d)
3.91(br dt) 3.82(td)
3.92(s)
3.71(d)
3.96(d)
4.17(dd) 3.97(d)
3.98(dd)
3.06(dd)
3.27(d)
4.72(d)
3.49(m) 4.00(ddd) 3.44(ddd) 3.43(dt) 3.88(td) 3.50(td) 3.64(ddd) 3.59(ddd)
3.80(t)
4.32(dd) 4.28(dd)
5.55(s)
3.61(s)
3.20(t)
3.84(dd)
5.14(s)
2.92(s)
3.72(t)
4.27(dd)
5.55(s)
3.56(s)
3.71(t)
3.80(t)
4.32(dd)
5.56(s)
3.60(s)
3.80(t)
3.75(t)
3.19(t)
3.80(t)
3.82(t)
4.35(dd)
5.56(s)
3.66(s)
4.29(dd) 4.32(dd) 3.84(dd) 4.29(dd)
5.56(s)
3.66(s)
PhCH 5.58(s)
OMe 3.56(s)
5.56(s)
3.48(s)
5.57(s)
3.57(s)
5.60(s)
3.48(s)
5.08(s)
2.81(s)
5.66(s)
3.52(s)
a In C6D6, data in CDCl3 see Section 3.
b br—broad.
oside 15 gave only one isomer, that is, manno isomer 28
from 14 and allo isomer 30 from 15. Not only the cou-
pling constants (J1,2 0 Hz), but also chemical shift of
H-4 close to the corresponding 1,5-anhydro sugar sup-
ports the respective configuration; d 3.97 for 28 versus
d 3.95 for 34 and d 4.72 for 30 versus d 4.67 for 35.
For cyano epoxides J3,4 values and H-1 chemical shifts
of the allo isomer 19 is 1.2 Hz and d 5.07, whereas that
of manno isomer 20 is 0 Hz and d 4.97. Both compounds
19 and 20 were hydrated to the carbamoyl derivatives 18
and 17, respectively. Although isolation of 2-C-cyano
epoxide 27 having the allo configuration is not accom-
plished, its configuration was determined by comparison
of 1H NMR data with those of the alternative isomer 26;
d 4.96 (H-1) and J3,4 1.2 Hz for 27 and d 5.10 (H-1) and
J3,4 0 Hz for 26. Both cases, 19 versus 20 and 26 versus
27, the hydrogen atom occupied the same side of cyano
group again appeared slightly lower field than that of
alternative one. 3-C-Cyano epoxide 29 having J1,2
0 Hz was hydrolyzed to the carbamoyl derivative 28.
The difference of chemical shift was again observed in
nitro epoxides. The configuration of 1,5:2,3-dianhydro-
3-C-nitro derivatives 39 and 40 was determined by the
coupling constant and chemical shift of H-4 (J1a,2 1.1,
J1e,2 0 Hz, d 4.04 for 39 and J1a,2 0, J1e,2 3.5 Hz, d 5.01
for 40) and confirmed by chemical transformation of
hydroperoxide 42 having the manno structure into epox-
Table 2. 1H NMR (400 MHz, in CDCl3) chemical shift (d in ppm) of
1,5:2,3-dianhydrohexitols
34
35
39
40
H-1a
H-1e
H-2
4.02(dd)
4.26(d)
3.50(br s)a
—
4.02(br d)
4.14(dd)
3.52(d)
—
4.09(dd)
4.22(d)
3.85(br s)
—
3.99(d)
4.23(dd)
3.91(d)
—
H-3
H-4
H-5
H-6a
H-6e
PhCH
3.95(d)
3.33(td)
3.72(t)
4.33(dd)
5.60(s)
4.67(d)
3.63(ddd)
3.83(t)
4.25(dd)
5.56(s)
4.04(d)
3.47(td)
3.76(t)
4.35(dd)
5.61(s)
5.01(d)
3.59(td)
3.80(t)
4.30(dd)
5.72(s)
a br—broad.
21 and 22, respectively. The hydrogen atom occupied
the same side of a carbamoyl group appears in a lower
field than that occupied the opposite side because of
the anisotropy of the carbamoyl group. In fact, for
example, the anomeric proton of 2-carbamoyl-a-D-
mannopyranoside 17 (d 5.05; J3,4 0 Hz) appeared in a
higher field than that of 2-carbamoyl-a-D-allopyran-
oside 18 (d 5.60; J3,4 0.8 Hz). The manno configuration
for 21 was confirmed by transformation of 3-hydroper-
oxy-b-D-glucopyranoside 23 into 21. The configuration
of 1,5:2,3-dianhydro-3-C-carbamoyl derivatives 34 and
35 could be assigned by the coupling constant (J1a,2
1.0 and J1e,2 0 Hz for 34, and J1a,2 0 and J1e,2 3.5 Hz
for 35) and chemical shift of H-4 (d 3.95 for 34 and d
4.67 for 35). Methyl 3-cyano-a- 14 and -b-2-enopyran-
1
ide 39. In the previous paper a satisfactory H NMR
spectrum of methyl 2,3-anhydro-2-C-nitro-a-D-allopyr-
anoside (46) was not obtained because of its low solubil-
Table 3. Coupling constants (Hz) of 2,3-anhydro sugars (400 MHz, CDCl3)
17
18
19
20
21
22
26
27a
28
29
30
34
35
37
39
40
J1a,1e
J1a,2
J1e,2
J2,3
—
—
—
—
0
9.4
10.0
4.0
—
—
—
—
—
—
—
—
—
—
—
—
0
9.5
10.3
4.5
—
—
—
—
0
9.4
10.3
4.7
—
—
—
—
0
9.4
10.4
4.7
—
—
—
—
0
9.6
9.8
4.6
—
—
—
—
—
—
0
—
—
9.6
10.3
4.0
—
—
0
—
—
9.5
10.2
4.6
—
0
—
—
—
9.3
10.3
4.9
13.8
1.0
0
—
—
9.5
9.7
4.8
10.5
13.9
0
3.5
—
—
—
—
—
0
9.4
10.4
4.6
14.3
1.1
0
—
—
9.4
10.7
4.7
10.2
14.0
0
3.5
—
J3,4
J4,5
0.8
1.2
1.2
—
—
9.2
10.2
4.9
8.9
10.1
5.1
7.6
10.3
4.7
9.0
8.9
4.8
10.2
9.0
10.4
4.9
10.4
J5,6a
J5,6e
J6a,6e
10.0
10.2
10.1
10.3
10.3
10.4
10.5
10.6
10.3
10.3
10.3
10.6
a J1,4 1.2 Hz.

2346
T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
Less bulky hydrogen peroxide predominantly added
Ph
O
O
R
O
O
O
from the axial side; however, the more bulky tert-butyl
hydroperoxide approached from both sides in almost
equal ratio. In spite of the presence of unfavorable Fac-
tor 3, tert-butyl hydroperoxide almost exclusively added
from the equatorial side of the a-anomer of nitro alkene
36 as had been mentioned in Section 1.7 Therefore, we
expected that tert-butyl hydroperoxide should exclu-
sively attack from the equatorial side of cyanoalkene
3, because Factor 3 is unimportant in the case of cyano
alkene, but this is not the case. This unexpected result is
probably explained by additional steric hindrance due to
the deviation of the nitro group from the C-2–C-3 dou-
ble bond plane as suggested by ab initio calculations.à
Equatorial attack is favorable for 2-cyano-b-D-2-eno-
pyranoside 4 due to the Factors 1 and 4, but alternative
attack by the Factor 2. Bulky tert-butyl hydroperoxide
exclusively attacked from the equatorial side of 4; how-
ever, the stereoselectivity decreased in the case of hydro-
gen peroxide. On the other hand, both the reagents
attacked only from the equatorial side of the corre-
sponding 2-C-nitro alkene 36. The difference observed
in the epoxidation of 4 and 36 with hydrogen peroxide
should be caused by a different effect of Factor 3 (the
presence or absence of A(1,3) strain). As expected from
the Factor 3, axial attack of hydrogen peroxide is higher
in 1,5-anhydro cyano alkene 16 than in 1,5-anhydro ni-
tro alkene 38; the ratio of axial and equatorial attack be-
came 7:1 and ꢀ1:1.7 in 16 and 38, respectively. Thus the
latter two examples should indicate an important role of
the A(1,3) strain (Factor 3) to determine the approaching
direction of hydroperoxide anion.
O
O
Ph
O
O
OMe
NO2
OH
OMe
R
43
46
44 R=H (84%)
45 R=D
NO2
Ph
R
O
O
O
Ph
O
R
OMe
OH
OMe
O
47 R=H (89%)
48 R=D
Scheme 8.
ity in CDCl3 and decomposition in hot DMSO-d6;7
however, an acceptable spectrum was recorded at
400 MHz in acetone-d6, which revealed a J3,4 value of
1.2 Hz. The chemical shift of H-1 again supported the
assignment (d 5.46 for 43 and d 6.03 for 46). Adequacy
of the assignment was proved by reduction of 43 and its
manno isomer 46 with lithium aluminum deuteride, in
which a deuteride ion attacked from the opposite side
of the epoxide ring15 to give 3-deoxy-2,3-deuterio-a-D-
allo 45 and -a–D-glucopyranoside 48, respectively
(Scheme 8). The manno structure for nitro epoxide 37,
the only product generated from methyl 2-C-nitro-b-2-
enopyranoside 36, was assigned on the basis of J3,4
0 Hz. It is noteworthy that H-1 chemical shift (d 5.90)
of 37 is close to that of 46 (d 5.98) rather than that of
43 (d 5.32).
3. Experimental
2.4. Stereoselectivity of epoxidation
3.1. General methods
We discuss the stereochemistry assuming that a stereo-
electronic effect inevitably operates. The anomeric effect4
(Factor 4) is in operation in addition to the three factors
mentioned in Section 1. Furthermore, Factor 3 plays an
unimportant role in the case of linear cyano alkene, and
the influence of Factor 1 increases with increasing bulk-
iness of nucleophile. Both hydrogen peroxide and tert-
butyl hydroperoxide approached from the opposite side
of anomeric methoxyl group of 3-C-cyano-2-enopyrano-
sides 14 and 15. The highly stereoselective axial attack to
3-C-cyano-a-D-enopyranoside 14 should be caused by
the Factors 1, 2, and 4. In the case of b-anomer 15,
equatorial attack is favored by Factors 1 and 4, but
disfavored by Factor 2. Equatorial attack exclusively
occurred, suggesting that besides the Factors 1 and 4,
synergetic electrostatic repulsion due to O-1 and O-5
prevents axial attack. In the case of 2-C-cyano-a-D-2-
enopyranoside 3, axial attack is favored by Factors 2
and 4, and equatorial attack is supported by Factor 1.
Melting points are uncorrected. Optical rotations were
determined with a Horiba High Sensitive Polarimeter
(SEPA-200). 1H NMR spectra were recorded at
270 MHz (JNM-EX270) or at 400 MHz (Bruker Ad-
vance 400) in CDCl3 with Me4Si as the internal stan-
dard. The integration values in the NOE difference
spectra are roughly estimated, because measurement
conditions were not completely optimized. IR spectra
were recorded for the compounds in KBr pellets. The
reaction mixtures were dried over MgSO4 and evapo-
rated under diminished pressure. Column chromatogra-
phy was conducted on silica gel (Wakogel C-300).
à Ab initio calculations (B3LYP/6-31+G*)16 indicate that the nitro
group deviates from the double-bond plane by 13°, probably to
reduce repulsion between O-1 and the oxygen atom of the nitro group
[A(1,2) strain].5 As a result the nitro group should suppress the axial
attack of a bulky tert-butyl hydroperoxide.

T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
2347
3.2. Preparation of methyl 4,6-O-benzylidene-2-C-cyano-
2,3-dideoxy-a-D-erythro-hex-2-enopyranoside (3)
To a solution of a crude 8 (50 mg, 0.17 mmol) in THF
(4 mL) was added Et3N (0.02 mL, 0.15 mmol) at
ꢁ20 °C. To the mixture was added dropwise a solution
of MsCl (0.016 mL, 0.21 mmol). The mixture was stirred
for 1.5 h, diluted with CHCl3 and washed successively
with dil aq HCl, aq NaCl, aq NaHCO3, and aq NaCl.
The organic phase was dried, and evaporated. The
residue was dissolved in CHCl3 (5 mL) and cooled
to ꢁ30 °C, to which was added DBU (0.04 mL, 0.26
mmol). The mixture was stirred for 2 h at ꢁ30 °C,
quenched with dil HCl at ꢁ30 °C and then extracted
with CHCl3. The extracts were successively washed with
dil aq HCl, aq NaCl, aq NaHCO3, and aq NaCl, dried,
and evaporated. The residue was chromatographed with
toluene to give 45.5 mg (97%) of 14 mp 214–214.5 °C,
To a solution of 3-nitro olefin 1 (100 mg, 0.34 mmol)17
in toluene (10 mL) and H2O (5 mL) was successively
added KCN (33 mg, 0.51 mmol), hexadecyltributyl-
phosphonium bromide (4 mg), and AcOH (0.02 mL,
0.35 mmol). After stirring for 4 h at room temperature
(ꢀ23 °C), 1 was consumed. The mixture was diluted
with EtOAc, washed with aq dil HCl, satd aq NaCl,
aq NaHCO3, and satd aq NaCl, dried, and evaporated.
The resulting crystalline residue was dissolved in CH2Cl2
(20 mL), to which Et3N (0.06 mL, 0.44 mmol) was
added. After stirring for 2 h at room temperature, the
mixture was diluted with CH2Cl2, washed with dil aq
HCl, satd aq NaCl, aq NaHCO3, and satd aq NaCl,
dried, and evaporated. The residue was chromato-
graphed with toluene to give 67 mg (72%) of 3, which
1
lit.10 213–214 °C. The H NMR data for the product
was identical with those reported.10
1
was identified by H NMR data.8
3.5. Preparation of methyl 4,6-O-benzylidene-3-C-cyano-
3-deoxy-b-D-altropyranoside (9)
3.3. Preparation of methyl 4,6-O-benzylidene-2-C-cyano-
2,3-dideoxy-b-D-erythro-hex-2-enopyranoside (4)
Similar treatment of 6 (300 mg, 1.14 mmol)19 with Et3Al
(15% in toluene, 4.1 mL, 4.55 mmol) and HCN (0.64 M
in THF, 7.1 mL) gave unreacted starting material 6
(120 mg, 40%) and then altro product 9 (135 mg, 41%),
after column chromatography eluting with a gradient
of 50:1 to 30:1 (v/v), toluene–EtOAc.
Similar treatment of an ice-cooled solution (0–6 °C) of
3-nitroolefin 2 (30 mg, 0.10 mmol)18 with KCN (8 mg,
0.12 mmol) for 2 h, followed by similar treatment with
1
Et3N, gave 24 mg (86%) of 4, the H NMR of which
matched those reported.9
Physicochemical and spectral data for 9: mp 154–
25
156 °C (i-PrOH); ½aꢂD +6.2 (c 0.7, CHCl3); m 3480
1
3.4. Preparation of 4,6-O-benzylidene-3-C-cyano-2,3-di-
deoxy-a-D-erythro-hex-2-enopyranoside (14)
(OH) and 2245 cmꢁ1 (CN); H NMR: d 4.83 (d, 1H,
J1,2 1.3 Hz, H-1), 4.16 (br t, 1H, J2,3 3.0, J2,OH 1.3 Hz,
H-2), 3.51 (dd, 1H, J3,4 5.0 Hz, H-3), 4.11 (dd, 1H, J4,5
9.2 Hz, H-4), 3.90 (td, J5,6a 9.6, J5,6e 4.3 Hz, H-5), 3.83
(t, 1H, J6a,6e 9.6 Hz, H-6a), 4.37 (dd, 1H, H-6e), 5.57
(s, 1H, PhCH), 2.66 (d, 1H, OH), and 3.60 (s, 3H,
OMe). Anal. Calcd for C15H17NO5: C, 61.85; H, 5.88;
N, 4.81. Found: C, 61.84; H, 6.01; N, 4.75.
3.4.1. Methyl 4,6-O-benzylidene-3-C-cyano-3-deoxy-a-D-
altropyranoside (8). To a solution of Et3Al (15% in tolu-
ene, 7.0 mL, 7.68 mmol) in THF (31 mL) was added
HCN (0.94 M in THF, 8.1 mL) under an N2 atmo-
sphere. To the mixture was slowly added a solution of
5 (1 g, 3.78 mmol) in THF (15 mL). After stirring for
20 h, the mixture was slowly added to ice-cooled 2 M
aq NaOH (5 mL) and then extracted with CHCl3. The
extracts were washed with aq NaCl, dil aq HCl, aq
NaCl, aq NaHCO3, and aq NaCl, dried, and evapo-
rated. The residue was chromatographed with 50:1
(v/v), CHCl3–MeOH to give unreacted starting mate-
rial 5 (400 mg, 40%) and then the altro product 8
(350 mg, 32%). The product was used directly in the next
step.
3.6. Methyl 4,6-O-benzylidene-3-C-cyano-3-deoxy-2-O-
methanesulfonyl-b-D-altropyranoside (12)
To a solution of 9 (20 mg, 0.069 mmol) in THF (2 mL)
was added Et3N (0.02 mL, 0.14 mmol) at ꢁ20 °C, fol-
lowed by the dropwise addition of a solution of MsCl
(0.008 mL, 0.10 mmol) in THF (0.1 mL). After 3 min,
the mixture was diluted with CHCl3, and the organic
phase was washed with dil aq HCl, aq NaCl, aq NaH-
CO3, and aq NaCl, dried, and evaporated. The residue
was chromatographed to give the mesylate 12 (25 mg,
25
3.4.2. Preparation of 4,6-O-benzylidene-3-C-cyano-3-
dideoxy-a-D-erythro-hex-2-enopyranoside (14). Davi-
son and Guthrie10 prepared the corresponding cyano
olefin 14 after epimerization at C-3, followed by p-bromo-
phenylsulfonylation and elimination in 27% yield
(overall yield from 5). However, the following modified
method gave 14 in higher yield.
99%): mp 160–161 °C (i-PrOH); ½aꢂD ꢁ21.1 (c 0.7,
1
CHCl3); m 2255 cmꢁ1 (CN); H NMR: d 4.92 (d, 1H,
J1,2 1.0 Hz, H-1), 4.98 (dd, 1H, J2,3 3.0 Hz, H-2), 3.67
(dd, 1H, J3,4 5.0 Hz, H-3), 4.09 (dd, 1H, J4,5 9.2 Hz,
H-4), 3.94 (td, 1H, J5,6a 9.9, J5,6e 4.3 Hz, H-5), 3.86 (t,
1H, J6a,6e 9.5 Hz, H-6a), 4.39 (dd, 1H, H-6e), 5.59 (s,
1H, PhCH), 3.60 (s, 3H, OMe), and 3.14 (s, 3H, Ms).

2348
T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
Anal. Calcd for C16H19NO7S: C, 52.03; H, 5.18; N, 3.79;
S, 8.68. Found: C, 51.85; H, 5.40; N, 3.75; S. 8.60.
(ddd, 1H, J6,6a 10.6, J5,6e 4.6 Hz, H-5), 3.80 (t, 1H,
J6a,6e 10.6 Hz, H-6a), 4.34 (dd, 1H, H-6e), and 5.65 (s,
1H, PhCH). Anal. Calcd for C14H13NO3: C, 69.12; H,
5.39; N, 5.76. Found: C, 69.40; H, 5.10; N, 5.72.
3.7. Methyl 4,6-O-benzylidene-3-C-cyano-2,3-dideoxy-
b-D-erythro-hex-2-enopyranoside (15)
3.10. Typical procedure for epoxidation of cyano olefin
with hydrogen peroxide; epoxidation of 3
Similar treatment of mesylate 12 (20 mg, 0.054 mmol)
with DBU (0.012 mL, 0.08 mmol) employed for prepa-
ration of 14 afforded cyano olefin 15 (15 mg, ꢀ100%)
after column chromatography eluting 50:1 to 30:1 tolu-
3.10.1. Method A. To a solution of cyano olefin 3
(20 mg, 0.10 mmol) in 1,4-dioxane (1 mL) was added
30% aq H2O2 (0.2 mL) and M NaOH (0.2 mL). After
stirring for 30 min at room temperature, the mixture
was diluted with EtOAc, and the organic phase was
washed with dil aq HCl, aq NaCl, aq NaHCO3, and
aq sodium thiosulfate, dried, and evaporated. The crys-
talline residue was chromatographed with 2:1 (v/v), hex-
ane–EtOAc to give 4.4 mg (13%) of amide 17 with the
manno configuration and 27 mg (80%) of 18 with the allo
configuration. Physicochemical and spectral data for
25
ene–EtOAc; mp 167–167.5 °C (i-PrOH); ½aꢂD ꢁ34.7 (c
0.8, CHCl3); m 2230 cmꢁ1 (CN); 1H NMR (C6D6): d
4.49 (dd, 1H, J1,2 1.7, J1,4 3.0 Hz, H-1), 5.72 (dd, 1H,
J2,4 2.3 Hz, H-2), 3.62 (td, 1H, J4,5 8.3 Hz, H-4), 3.27
(m, 1H, J5,6a 9.9, J5,6e 4.3 Hz, H-5), 3.38 (t, 1H, J6a,6e
9.9 Hz, H-6a), 3.97 (dd, 1H, H-6e), 5.16 (s, 1H, PhCH),
and 3.06 (s, 3H, OMe). Anal. Calcd for C15H15NO4: C,
65.93; H, 5.53; N, 5.13. Found: C, 65.81; H, 5.40; N,
4.95.
25
17: mp 222–223°C (i-PrOH); ½aꢂD +61.2 (c 0.5, CHCl3);
m
3350, 1680 cmꢁ1 (CONH2); 1H NMR(CDCl3,
1
3.8. 1,5-Anhydro-4,6-O-benzylidene-3-C-cyano-3-deoxy-
D-altritol (10)
400 MHz, see Table 1): H NMR (C6D6, 400 MHz): d
4.71 (s, 1H, H-1), 3.75 (s, 1H, H-3), 3.49 (d, 1H, J4,5
9.6 Hz, H-4), 3.69 (td, 1H, H-5), 3.37 (t, 1H,
J5,6a = J6a,6e 10.3 Hz, H-6a), 3.97 (dd, 1 H, J5,6e
4.4 Hz, H-6e), 5.14 (s, 1H, PhCH), 2.86 (s, 3H, OMe).
Correlation was observed between H-1 and OMe and
between PhCH and H-4 as well as H-6a in the NOESY
spectrum. Anal. Calcd for C15H17NO6: C, 58.63; H,
5.58; N, 4.56. Found: C, 58.73; H, 5.41; N, 4.72. Physi-
cochemical and spectral data for 18: mp 244–246 °C
Similar treatment of 7 (500 mg, 2.13 mmol)20 with Et3Al
(15% in toluene, 4.0 mL, 4.28 mmol) and HCN (0.94 M
in THF, 4.6 mL, 4.28 mmol) for 20 h, followed by col-
umn chromatography eluting with 20:1 (v/v), toluene–
EtOAc to give unreacted starting material 7 (120 mg,
24%) and then altritol 10 (335 mg, 60%): mp 117–
25
118 °C (toluene–hexane); ½aꢂD +47.5 (c 0.8, CHCl3); m
3480 (OH) and 2245 cmꢁ1 (CN); 1H NMR (C6D6,
400 MHz): d 3.44 (dd, 1H, J1a,1e 13.0, J1a,2 1.4 Hz, H-
1a), 3.25 (dt, 1H, J1e,2 = J1e,3 1.6 Hz, H-1e), 3.31 (br s,
1H, H-2), 2.76 (m, 1H, H-3), 3.71 (dd, 1H, J3,4 4.9,
J4,5 9.5 Hz, H-4), 3.63 (ddd, J5,6a 9.7, J5,6e 4.9 Hz, H-
5), 3.39 (br t, 1H, J6a,6e 10.0 Hz, H-6a), 4.05 (dd, 1H,
H-6e), 5.20 (s, 1H, PhCH), 1.43 (d, 1H, J2,OH 4.9 Hz,
OH). Anal. Calcd for C14H15NO4: C, 64.36; H, 5.79;
N, 5.36. Found: C, 64.58; H, 5.64; N, 5.40.
25
(i-PrOH), ½aꢂD +124 (c 0.6, CHCl3); m 3350, 1660 cmꢁ1
1
(CONH2). For H NMR data see Table 1. Correlation
was observed between H-1 and OMe and between
PhCH and H-4 as well as H-6a in the NOESY spectrum.
Anal. Calcd for C15H17NO6: C, 58.63; H, 5.58; N, 4.56.
Found: C, 58.62; H, 5.49; N, 4.28.
3.10.2. Method B. To a solution of 3 (100 mg,
0.37 mmol) in 1,4-dioxane (5 mL) was added 30% aq
H2O2 (0.3 mL) and M NaOH (0.3 mL). After stirring
for 10 min at room temperature, the mixture was simi-
larly worked up to give a crystalline residue that was
chromatographed with 100:1 (v/v), toluene–EtOAc to
afford 7 mg (7%) of the cyano epoxide having the allo
configuration 19 and starting material 3 (ꢀ90%). Physi-
cochemical and spectral data for 19: mp 168–169 °C (i-
3.9. 1,5-Anhydro-4,6-O-benzylidene-3-C-cyano-2,3-di-
deoxy-b-D-erythro-hex-2-enitol (16)
Similar mesylation of 10 (70 mg, 0.27 mmol) with MsCl
(0.031 mL, 0.40 mmol) and Et3N (0.094 mL, 0.68 mmol)
for 30 min, as employed for the preparation of 12 gave
mesylate 13, which was found to be labile on silica gel.
Compound 13 was converted to cyano olefin 16 by treat-
ment with DBU (0.05 mL, 0.33 mmol) under the same
conditions employed for 14. Compound 16 (52 mg,
80% yield) was obtained after column chromatography:
25
PrOH); ½aꢂD +102.2 (c 0.5, CHCl3); m 2240 cmꢁ1 (CN);
in 1H NMR (C6D6, 400 MHz, Table 1) correlation
was observed between H-1 and OMe and between
PhCH and H-4 as well as H-6a in the NOESY spectrum.
The signals of H-1 (s, 1H, 5.07) and PhCH (s, 1H, 5.56)
in CDCl3 (400 MHz) were assigned by the 1D-NOESY
spectrum by irradiation at OMe (s, 3H, 3.53). Anal.
Calcd for C15H15NO5: C, 62.28; H, 5.23; N, 4.84.
Found: C, 62.29; H, 5.01; N, 4.92.
25
mp 92–93 °C (i-PrOH); ½aꢂD ꢁ47.3 (c 0.9, CH2Cl2); m
1
2220 cmꢁ1 (CN); H NMR: d 4.41–4.44 (m, 2H, H-1a,
H-1e), 6.66 (q, 1H, J1a,2 = J1e,2 = J2,4 2.6 Hz, H-2),
4.24 (m, 1H, J1a,4 = J1e,4 3.0, J4,5 7.9 Hz, H-4), 5.52

T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
2349
3.10.3. Method C. To a dispersion of 19 (5 mg) in 1,4-
dioxane (0.25 mL) with H2O2 (0.05 mL), was added M
NaOH (0.05 mL) and the solution was stirred for
30 min at room temperature. After the work-up com-
pound 18 was exclusively formed as judged from the
1H NMR spectrum.
PhCH and H-4 as well as H-6a in the NOESY spectrum.
Anal. Calcd for C15H17NO6: C, 58.63; H, 5.58; N, 4.56.
Found: C, 58.42; H, 5.71; N, 4.66.
3.12.2. Method B. Similar reaction of 4 (100 mg,
0.36 mmol) for 5 min and the same workup afforded a
residue, which was chromatographed with 5:1, 2:1, and
1:2 (v/v), hexane–EtOAc to give a mixture of 4, 26,
and 27 and that of 21, 22, 23, and unknown compounds
(not completely purified). Thus the fraction contami-
nated the unknown compound, therefore, estimation
of yields was abandoned. Peroxide 23 crystallized from
the above mixture by the use of EtOAc was purified
by column chromatography with 2:1 (v/v), hexane–
3.11. Typical procedure for epoxidation of cyano olefins
with tert-butyl hydroperoxide; epoxidation of 3
To a solution of 3 (54 mg, 0.20 mmol) in 1,4-dioxane
(3.5 mL) was added ꢀ70% aq tert-BuOOH (0.2 mL)
and M NaOH (0.2 mL). After stirring for 3 h at room
temperature, the mixture was diluted with EtOAc, and
the organic phase was washed with dil aq HCl, aq NaCl,
aq NaHCO3, aq sodium thiosulfate, and satd aq NaCl,
dried, and evaporated. The crystalline residue was chro-
matographed with 2:1 (v/v), hexane–EtOAc to give
22 mg (38%) of the cyano epoxide with manno configu-
ration 20 and the amide with manno configuration 17
(5 mg, 8%) and allo configuration 18 (24 mg, 40%).
24
EtOAc.; mp 175–178.5 °C (EtOAc); ½aꢂD ꢁ175 (c 0.03,
CHCl3); m 3326 (OH), 2253 cmꢁ1 (CN); 1H NMR
(400 MHz): d 4.66 (d, 1H, J1,2 8.5 Hz, H-1), 3.13 (dd,
1H, J2,3 10.9 Hz, H-2), 4.37 (dd, 1H, J3,4 9.9 Hz, H-3),
3.83 (dd, 1H, J4,5 9.4 Hz, H-4), 3.49 (ddd, 1H, H-5),
4.39 (dd, 1H, J5,6e 4.9, J6a,6e 10.3 Hz, H-6e), 3.82 (dd,
1H, J5,6a 10.0 Hz, H-6a), 5.58 (s, 1H, PhCH), 8.43 (s,
1H, –OOH), 3.61 (s, 3H, OMe). Anal. Calcd for
C15H17NO6: C, 58.63; H, 5.58; N, 4.56. Found: C,
58.33; H, 5.68; N, 4.34.
Physicochemical and spectral data for 20: mp 228–
25
230 °C (i-PrOH); ½aꢂD +0.5 (c 0.6, CH2Cl2);
m
2240 cmꢁ1 (CN). For 1H NMR data see Table 1. Corre-
lation was observed between H-1 and OMe and between
PhCH and H-4 as well as H-6a in the NOESY spectrum.
Anal. Calcd for C15H15NO5: C, 62.28; H, 5.23; N, 4.84.
Found: C, 62.39; H, 4.99; N, 4.65.
To hydroperoxide 23 (5 mg, 0.016 mmol) in 1,4-diox-
ane (0.22 mL) was added 70% of tert-BuOOH (0.03 mL)
and M NaOH (0.03 mL), and the mixture was moni-
tored by TLC. Even after 10 h starting material still
remained; therefore, the same amount of tert-BuOOH
and M NaOH was added. After 24.5 h, similar workup
Similar treatment of 20 (17 mg, mmol) with 30% aq
H2O2 (0.2 mL) and M NaOH (0.2 mL) for 30 min at
room temperature afforded in almost quantitative yield
amide 17 having the manno configuration.
1
gave a residue, that was shown by H NMR spectro-
scopy to be compound 21.
3.13. The reaction of methyl 4,6-O-benzylidene-2-C-
cyano-2,3-dideoxy-b-D-erythro-hex-2-enopyranoside (4)
with tert-butylperoxide and related reactions
3.12. The reaction of methyl 4,6-O-benzylidene-2-C-
cyano-2,3-dideoxy-b-D-erythro-hex-2-enopyranoside (4)
with hydrogen peroxide
3.13.1. Method A. Similar epoxidation of 4 (30 mg,
0.11 mmol) in 1,4-dioxane (1.5 mL) with ꢀ70% aq
tert-BuOOH (0.2 mL) in the presence of M NaOH
(0.2 mL) for 6.5 h gave a crystalline residue that was
chromatographed with 5:1 (v/v), toluene–EtOAc to give
33 mg (98%) of 21.
3.12.1. Method A. Similar epoxidation of 4 (30 mg,
0.11 mmol) with 30% aq H2O2 (0.2 mL) in the presence
of M NaOH (0.2 mL) for 30 min gave a mixture, from
which 10 mg (30%) of amide 21 with the manno config-
uration and 20 mg (59%) of 22 with the allo configura-
tion were separated by column chromatography
eluting with 5:1 (v/v), toluene–EtOAc. Physicochemical
and spectral data for 21: mp 236–237 °C (i-PrOH),
3.13.2. Method B. Similar reaction of 4 (100 mg,
0.36 mmol) for 5 min, followed by the same workup,
gave a mixture. The same reaction was repeated and
the residues were combined and then chromatographed
with 4:1, 2:1 (v/v), hexane–EtOAc to give a mixture of 4,
24, and 26 as the first-running fraction and then 21
(36%) and 25 (34%). The yields of 4 (3%), 24 (22%),
26 (4%) were calculated by integration of their charac-
teristic signals in the 1H NMR spectrum of the mixture.
Physicochemical and spectral data for 25: amorphous,
25
½aꢂD ꢁ49.5 (c 0.5, CH2Cl2);
m
3370, 1638 cmꢁ1
1
(CONH2). For H NMR data see Table 1. The signals
of H-1 and PhCH were confirmed by a 1D-NOESY
spectra irradiated at the PhCH signal. Anal. Calcd for
C15H17NO6: C, 58.63; H, 5.58; N, 4.56. Found: C,
58.80; H, 5.39; N, 4.30. Physicochemical and spectral
25
data for 22: mp 175–176.5 °C (i-PrOH), ½aꢂD ꢁ32.7 (c
0.6, CHCl3); m 3420, 1670 cmꢁ1 (CONH2). For 1H
NMR data see Table 1. Correlation was observed
between H-1 and OMe as well as H-5 and between
24
½aꢂD ꢁ240 (c 0.07, CHCl3); m 2252 cmꢁ1 (CN); 1H

2350
T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
NMR (400 MHz): d 4.55 (d, 1H, J1,2 2.2 Hz, H-1), 3.85
(dd, 1H, J2,3 5.0 Hz, H-2), 4.37 (dd, 1H, J3,4 10.2 Hz, H-
3), 3.92 (dd, 1H, J4,5 10.1 Hz, H-4), 3.46 (ddd, 1H, H-5),
4.36 (dd, 1H, J5,6e 4.8, J6a,6e 10.6 Hz, H-6e), 3.92 (dd,
1H, J5,6a 9.8 Hz, H-6a), 5.60 (s, 1H, PhCH), 3.58 (s,
3H, OMe), 1.28 (s, 9H, tert-Bu). Anal. Calcd for
C19H25NO6: C, 62.80; H, 6.93; N, 3.85. Found: C,
62.76; H, 7.11; N, 3.66.
Although purification of alternative cyano epoxide 27
was not accomplished, its 1H NMR (400 MHz) see
Tables 1 and 3.
3.14. The reaction of methyl 4,6-O-benzylidene-3-C-
cyano-2,3-dideoxy-a-D-erythro-hex-2-enopyranoside (14)
with tert-butyl hydroperoxide
Similar epoxidation of 14 (28 mg, 0.10 mmol) with tert-
BuOOH (0.1 mL) in the presence of a drop of M NaOH
for 20 min afforded a crystalline residue that was chro-
matographed with toluene to give 28 mg (94%) of cya-
3.13.3. Partial purification of compound 24. Complete
purification of peroxide 24 having the gluco configura-
tion was not accomplished, but the NMR data of almost
25
1
noepoxide 29: mp 185–186.5 °C (i-PrOH); ½aꢂD +87.2
pure 24 was as follows. H NMR (400 MHz): d 4.62 (d,
(c 0.7, CH2Cl2);
m
2310 cmꢁ1 (CN); 1H NMR
1H, J1,2 8.7 Hz, H-1), 3.25 (dd, 1H, J2,3 10.6 Hz, H-2),
4.27 (dd, 1H, J3,4 9.3 Hz, H-3), 3.84 (dd, 1H, J4,5
9.6 Hz, H-4), 3.43 (ddd, 1H, H-5), 4.38 (dd, 1H, J5,6e
4.9, J6a,6e 10.4 Hz, H-6e), 3.81 (dd, 1H, J5,6a 10.3 Hz,
H-6a), 5.55 (s, 1H, PhCH), 3.60 (s, 3H, OMe), 1.29 (s,
9H, tert-Bu).
(400 MHz): d 4.90 (s, 1H, H-1), 3.60 (s, 1H, H-2),
3.70–3.82 (m, 3H, H-4, H-5, H-6a), ꢀ4.28 (m, 1H, H-
6e), 3.48 (s, 3H, OMe), and 5.62 (s, 1H, PhCH). Corre-
lation was observed between H-1 and OMe in the
NOESY spectrum. For 1H NMR data (C6D6,
400 MHz) see Table 1. Correlation was observed be-
tween H-1 and OMe and between PhCH and H-4 as well
as H-6a in the NOESY spectrum. Anal. Calcd for
C15H15NO5: C, 62.28; H, 5.23; N, 4.84. Found: C,
62.55; H, 4.98; N, 4.88.
3.13.4. Epoxidation of 24 to 21. To almost pure 24
(12 mg) (obtained by after several fractional crystalliza-
tions) in 1,4-dioxane (0.5 mL) was added 70% tert-
BuOOH (0.073 mL) and M NaOH (0.073 mL). The
mixture was stirred at room temperature for 6.5 h. Sim-
ilar workup afforded a residue, of which NMR spectro-
scopy showed the presence of a small amount of starting
material 24. Therefore, the same treatment of the residue
was repeated again to give 21 in almost quantitative
yield.
3.15. The reaction of methyl 4,6-O-benzylidene-3-C-
cyano-2,3-dideoxy-a-D-erythro-hex-2-enopyranoside (14)
with hydrogen peroxide
Similar treatment of 14 (30 mg, 0.11 mmol) with 30% aq
H2O2 (0.2 mL) in the presence of a drop of M NaOH for
2 h, followed by addition of H2O, gave a precipitate.
The precipitate was dried over P2O5 to give 25 mg
(74%) of carbamoyl epoxide 28 (amorphous). Com-
pound 28 was barely soluble in common organic sol-
vents, and its purification by recrystallization was not
accomplished. Its 1H NMR spectrum was obtained after
repeated accumulations. The manno structure was
assigned by the transformation of 29 into amide 28 as
follows.
3.13.5. Conversion of 25 to 21. To a solution of 25
(6 mg), the 2-epimer of 24, in 1,4-dioxane (0.25 mL)
were added 70% tert-BuOOH (0.036 mL), and M NaOH
(0.036 mL) and the mixture was stirred at room temper-
ature for 6.5 h. Similar workup afforded a residue, that
was shown by NMR spectroscopy to be 21, together
with traces of 24–26.
3.13.6. Epoxidation of 4 with m-chloroperoxybenzoic
acid. To a solution of 4 (50 mg, 0.18 mmol) in 1,4-
dioxane (2.5 mL) was added 70% m-chloroperoxybenz-
oic acid (MCPBA) (135 mg) and M NaOH (0.64 mL),
and the mixture was stirred at room temperature for
4 h. Similar workup afforded a residue (56 mg), that
was shown by NMR spectroscopy to be a 4.8:2.3:1 mix-
ture of 26:27:4. The mixture was dissolved in EtOAc,
from which the cyano epoxide 26 (12 mg) was crystal-
To a solution of 29 (10 mg, 0.04 mmol) in 1,4-dioxane
(2 mL) was added 30% H2O2 (0.1 mL) and a drop of M
NaOH. After stirring for 30 min, similar workup gave
1
10 mg (94%) of 28, H NMR spectrum of which was
identical with that of an authentic sample. For 1H
NMR spectral data (400 MHz, CDCl3) see Tables 1
and 3.
3.16. The reaction of methyl 4,6-O-benzylidene-3-C-
cyano-2,3-dideoxy-b-D-erythro-hex-2-enopyranoside (15)
with hydrogen peroxide
25
lized: mp 200 °C (dec) (EtOAc); ½aꢂD ꢁ32 (c 0.1, CHCl3).
The absorption band of CN was not detected in the IR
1
spectrum. For H NMR data see Table 1. Correlation
was observed between H-1 and OMe and between H-1
and H-5 in the NOESY spectrum. Anal. Calcd for
C15H15NO5: C, 62.28; H, 5.23; N, 4.84. Found: C,
62.34; H, 5.50; N, 4.63.
Similar epoxidation of 15 (10 mg, 0.037 mmol) with aq
H2O2 (0.2 mL) in presence of M NaOH (0.2 mL) for
1 h afforded a residue that was chromatographed with
5:1 (v/v), toluene–EtOAc to give the amide having the

T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
2351
25
allo configuration 30 (11 mg, 98%): mp 211-212 °C (i-
141 °C (i-PrOH); ½aꢂD ꢁ64.9 (c 0.6, CHCl3); m 2240
25
1
PrOH); ½aꢂD ꢁ79.4 (c 0.8, CHCl3); m 3400, 1680 and
(CN) 1640 cmꢁ1 (O–C=C); H NMR: d 6.49 (dd, 1H,
1
1635 cmꢁ1 (CONH2). For H NMR data see Table 1.
J1,2 5.6, J1,3 1.1 Hz, H-1), 4.83 (t, 1H, J2,3 5.6 Hz, H-
3), 3.52 (td, 1H, J3,4 5.6 Hz, H-3), 4.01 (dd, 1H, J4,5
10.1 Hz, H-4), 4.16 (td, 1H, J5,6a 10.4, J5,6e 5.4 Hz, H-
5), 3.84 (t, 1H, J6a,6e 10.4 Hz, H-6a), 4.47 (dd, 1H, H-
6e), and 5.65 (s, 1H, PhCH). Anal. Calcd for
C14H13NO3: C, 69.12; H, 5.39; N, 5.76. Found: C,
68.93; H, 5.18; N, 5.48.
As shown above, the double-bond migration readily
occurred; therefore the epoxidation was carried out as
follows:
Correlation was observed between H-1 and OMe,
between H-1 and H-2, and between PhCH and H-4 as
well as H-6a in the NOESY spectrum. Anal. Calcd
for C15H17NO6: C, 58.63; H, 5.58; N, 4.56. Found: C,
58.74; H, 5.70; N, 4.56.
Compound 15 (50 mg, 0.18 mmol) in 1,4-dioxane
(1 mL) was treated with 30% aq H2O2 (0.2 mL) and M
NaOH (0.2 mL), and the mixture was stirred for 3 min
at room temperature. Similar workup gave a residue
that was chromatographed with a gradient of toluene
and EtOAc, 100:1 to 5:1, to give the starting material
15 (31 mg, 62%), gluco adduct 31 (11 mg, 20%), and
amide 30 (10 mg, 18%). Physicochemical and spectral
3.18.2. Method B. To a solution of 16 (60 mg,
0.25 mmol) in THF (5 mL) were added 30% aq H2O2
(0.2 mL) and M NaOH (0.1 mL) at 0 °C. After stirring
for 24 h at 0 °C, 0.2 mL of 30% aq H2O2 and 0.1 mL
of M NaOH were added to the mixture. This procedure
was repeated once again after 24 h. After 4 days (total
reaction time), the mixture was diluted with EtOAc,
and the organic phase was washed with dil aq HCl, satd
aq NaCl, aq NaHCO3, and aq sodium thiosulfate, dried,
and evaporated. The crystalline residue (20 mg) was
chromatographed with 50:1 (v/v) CHCl3–MeOH to give
50 mg (73%) of epoxide 34 and 7 mg (10%) of epoxide
35. Physicochemical and spectral data for 34: mp 234–
25
data for 31: mp 139.5–141 °C (i-PrOH); ½aꢂD ꢁ85.3 (c
1
0.7, CHCl3); m 2260 (CN) and 3330 cmꢁ1 (OOH); H
NMR: d 4.66 (d, 1H, J1,2 7.6 Hz, H-1), 4.04 (dd, 1H,
J2,3 10.9 Hz, H-2), 3.31 (t, 1H, J3,4 10.9 Hz, H-3), 3.82
(dd, 1H, J4,5 8.9 Hz, H-4), 3.44 (td, 1H, J5,6a 9.9, J5,6e
5.0 Hz, H-5), 3.80 (t, 1H, J6a,6e 10.6 Hz, H-6a), 4.38
(dd, 1H, H-6e), 5.60 (s, 1H, PhCH), 3.62 (s, 3H,
OMe), and 8.94 (br s, 1H, OOH). Anal. Calcd for
C15H17NO6: C, 58.63; H, 5.58; N, 4.56. Found: C,
58.62; H, 5.64; N, 4.71.
25
235.5 °C (i-PrOH–EtOH); ½aꢂD +11.8 (c 0.6, CHCl3); m
3360, 1665 cmꢁ1 (CONH2). For 1H NMR data see
Table 1. Correlation was observed between H-5 and
H-1a in the NOESY spectrum. Anal. Calcd for
C14H15NO5: C, 60.64; H, 5.45; N, 5.05. Found: C,
60.58; H, 5.05; N, 4.78. Physicochemical and spectral
3.17. The reaction of methyl 4,6-O-benzylidene-3-C-
cyano-2,3-dideoxy-b-D-erythro-hex-2-enopyranoside (15)
with tert-butyl hydroperoxide
Similar epoxidation of 15 (10 mg, 0.037 mmol) with tert-
BuOOH (0.2 mL) in the presence of M NaOH (0.2 mL)
for 40 min gave a residue that was chromatographed
with 5:1 (v/v) toluene–EtOAc to give 10 mg (89%) of 30.
25
data for 35: mp 218–220 °C (i-PrOH–EtOH); ½aꢂD
+18.9 (c 0.4, CHCl3); m 3360, 1655 cmꢁ1 (CONH2).
For 1H NMR data see Table 1. Correlation was
observed between H-5 and H-1a in the NOESY spec-
trum. Anal. Calcd for C14H15NO5: C, 60.64; H, 5.45;
N, 5.05. Found: C, 60.64; H, 5.15; N, 4.80.
3.18. The reaction of 1,5-anhydro-4,6-O-benzylidene-3-C-
cyano-2,3-dideoxy-D-erythro-hex-2-enitol (16) with
hydrogen peroxide and related reactions
3.18.3. Reaction of 16 with tert-butylhydroperoxide. To
a solution of 16 (30 mg, 0.12 mmol) in 1,4-dioxane
(2.5 mL) were added tert-BuOOH (0.2 mL) and M
NaOH (0.1 mL) at 0 °C. After stirring at 0 °C for 2 days,
the mixture was similarly worked up to give 16 mg
(53%) of 32 and 8 mg (27%) of 33, but there was no
evidence for the formation of an epoxide.
3.18.1. Method A. Similar epoxidation of 16 (20 mg,
0.08 mmol) with H2O2 (0.1 mL) in the presence of a
drop of M NaOH for 1 h gave a crystalline residue
(20 mg) that was chromatographed with toluene to give
6 mg (30%) of glucal 32 and 12 mg (60%) of allal 33.
Physicochemical and spectral data for 32: mp 127–
25
128.5 °C (i-PrOH), ½aꢂD ꢁ133.3 (c 0.3, CHCl3); m 2250
1
(CN) 1642 cmꢁ1 (O–C@C); H NMR: d 6.47 (dd, 1H,
3.19. Methyl 2,3-anhydro-4,6-O-benzylidene-2-C-nitro-
b-D-mannopyranoside (37)
J1,2 5.9, J1,3 2.8 Hz, H-1), 4.77 (dd, 1H, J2,3 2.8 Hz, H-
2), 3.56 (t, 1H, J3,4 9.5 Hz, H-3), 4.12 (t, 1H, J4,5
9.5 Hz, H-4), 3.76–3.89 (m, 2H, J5,6a 10.4, H-5, -6a),
4.43 (dd, 1H, J5,6e 5.4, J6a,6e 10.4 Hz, H-6e), and 5.68
(s, 1H, PhCH). Anal. Calcd for C14H13NO3: C, 69.12;
H, 5.39; N, 5.76. Found: C, 68.98; H, 5.64; N, 5.65.
Physicochemical and spectral data for 33: mp 140.5–
3.19.1. Method A. To a solution of 36 (47 mg,
0.16 mmol) in 1,4-dioxane (3.0 mL) were added 30%
H2O2 (0.3 mL) and then M NaOH (0.3 mL). After stir-
ring for 10 min at room temperature, the mixture was

2352
T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
similarly worked up as described for the preparation of
18 to give 37 mg (75%) of 37.
drop of M NaOH. The mixture was stirred for 30 s at
room temperature, and then the reaction was quenched
by the addition of dil aq HCl. The mixture was diluted
with EtOAc, and the organic phase was washed with
aq NaCl (three times) and aq sodium thiosulfate, dried,
and evaporated to give a residue (21 mg). The 1H NMR
spectrum of the residue showed that it consisted of the
epoxides having the manno configuration 39 and the allo
configuration 40 and peroxides having gluco 41 and
manno configuration 42 in the ratio of 1:7.6:4:6 (based
on benzylidene methine protons). During the chromato-
graphic separation on silica gel, compound 40 was
almost completely decomposed. The product ratio was
changed to 1:2.3:0.8:1 after 5 min.
3.19.2. Method B. To a solution of 36 (29 mg,
0.10 mmol) in 1,4-dioxane (2 mL) were added 70%
tert-BuOOH (0.04 mL) and then
M aq NaOH
(0.04 mL). The mixture was stirred for 20 min. After
similar workup, compound 37 was isolated in 92% yield
25
(28 mg); mp 193–196 °C (i-PrOH); ½aꢂD ꢁ55.6 (c 1.0,
CHCl3); m 1565 cmꢁ1 (NO2). For 1H NMR data see
Table 1. Correlation was observed between H-1 and
H-5 and between PhCH and H-4 as well as H-6a in
1
the NOESY spectrum. H NMR (CD3COCD3): d 6.03
(s, 1H, H-1), 4.16 (s, 1H, H-3), 3.80 (s, 1H, H-4),
3.82–3.61 (m, 1H, H-5), 3.86 (t, 1H, J5,6a = J6a,6e
9.6 Hz, H-6a), 4.29 (dd, 1H, J5,6e 3.7 Hz, H-6e), 5.72
(s, 1H, PhCH), and 3.61 (s, 3H, OMe). Correlation
was observed between H-1 and OMe and between
PhCH and H-4 as well as H-6a in the NOESY spectrum.
Anal. Calcd for C14H15NO7: C, 54.37; H, 4.89, N, 4.53.
Found: C, 54.39; H, 4.62; N, 4.49.
Compounds 40–42 could be isolated as follows. The
syrup obtained by the reaction for 2 min was dissolved
in CH2Cl2, and three times its weight of silica gel was
added to the solution. The mixture was evaporated until
only a small amount of the solvent remained, and then it
was chromatographed by the use of 20 times its weight
of silica gel, eluting with 50:1 (v/v) toluene–EtOAc to
give successively 40, 41, and 42. However, by this meth-
od compound 39 could not be isolated due to the over-
lap with decomposed products generated (probably
from 40). Physicochemical and spectral data for 40:
3.20. The reaction of 1,5-anhydro-4,6-O-benzylidene-2,3-
dideoxy-3-C-nitro-D-erythro-hex-2-enitol (38) with tert-
butyl hydroperoxide
25
mp 143–145 °C (toluene–EtOAc), ½aꢂD ꢁ76.0 (c 0.6,
To a solution of 38 (20 mg, 0.08 mmol) in 1,4-dioxane
(1 mL) were added ꢀ70% aq tert-BuOOH (0.1 mL)
and then a drop of M NaOH. The mixture was stirred
for 2 min at room temperature, and then the reaction
was quenched by the addition of dil aq HCl. The mix-
ture was diluted with EtOAc, and the organic phase
was washed with aq NaCl (three times), and aq sodium
thiosulfate, dried, and evaporated to give a residue
CHCl3); m 1565 cmꢁ1(NO2). For 1H NMR data see
Table 1. Correlation was observed between H-5 and
H-1a in the NOESY spectrum. Anal. Calcd for
C13H13NO6: C, 55.92; H, 4.69; N, 5.02. Found: C,
55.66; H, 4.56; N, 5.02. Physicochemical and spectral
25
data for 41: mp 87–88 °C (toluene–hexane); ½aꢂD +2.0
(c 0.5, CHCl3); m 3480 (OH), 1558 cmꢁ1(NO2); 1H
NMR: d 3.68 (dd, 1H, J1a,1e 11.6, J1a,2 10.2 Hz, H-1a),
4.32 (dd, 1H, J1e,2 5.6 Hz, H-1e), 4.64 (m, 1H, H-2),
5.04 (t, 1H, J2,3 = J3,4 10.2 Hz, H-3), 4.13 (t, 1H, J4,5
10.2 Hz, H-4), 3.45 (m, 1H, H-5), 3.77 (t, 1H, J5,6a
10.2, J6a,6e 10.6 Hz, H-6a), 4.40 (dd, 1H, J5,6e 5.0 Hz,
H-6e), and 5.55 (s, 1H, PhCH). Anal. Calcd for
C13H15NO7: C, 52.53; H, 5.09; N, 4.71. Found: C,
52.41; H, 5.24; N, 4.77. Physicochemical and spectral
1
(22 mg). The H NMR spectrum of the residue showed
that it consisted of 1:1.8 mixture of the epoxides having
the manno configuration 39 and the allo configuration
40.
When the reaction was prolonged to 1 h, 2,3-anhydro-
mannopyranoside 39 was isolated, because during that
time 40 had been decomposed. After workup as
described above, column chromatography of the residue
with 50:1 (v/v) toluene–EtOAc gave 39 in 15% yield; mp
25
data for 42: mp 81–82 °C (toluene–hexane); ½aꢂD ꢁ86.4
(c 0.9, CHCl3); m 3320 (OH), 1558 cmꢁ1(NO2); 1H
NMR: d 3.66 (dd, 1H, J1a,1e 13.2, J1a,2 1.3 Hz, H-1a),
4.50 (dd, 1H, J1e,2 1.6 Hz, H-1e), 4.82 (dt, 1H, H-2),
4.77 (dd, 1H, J2,3 4.1, J3,4 10.6 Hz, H-3), 4.55 (dd, 1H,
J4,5 9.2 Hz, H-4), 3.44 (m, 1H, H-5), 3.88 (t, 1H, J5,6a
10.2, J6a,6e 10.2 Hz, H-6a), 4.37 (dd, 1H, J5,6e 4.9 Hz,
H-6e), 5.70 (s, 1H, PhCH), and 8.55 (s, 1H, OOH).
Anal. Calcd for C13H15NO7: C, 52.53; H, 5.09; N,
4.71. Found: C, 52.59; H, 5.34; N, 4.67.
25
155–155.5 °C (i-PrOH); ½aꢂD ꢁ53.6 (c 1.0, CHCl3);
m 1558 cmꢁ1 (NO2). For 1H NMR data see Table 1.
Correlation was observed between H-5 and H-1 in the
NOESY spectrum. Anal. Calcd for C13H13NO6: C,
55.92; H, 4.69; N, 5.02. Found: C, 55.93; H, 4.52; N,
4.82.
3.21. The reaction of 1,5-anhydro-4,6-O-benzylidene-2,3-
dideoxy-3-C-nitro-D-erythro-hex-2-enitol (38) with
hydrogen peroxide
3.22. Treatment of 42 with hydrogen peroxide
To a solution of 38 (20 mg, 0.08 mmol) in 1,4-dioxane
(1 mL) were added 30% aq H2O2 (0.1 mL) and then a
To a solution of the peroxide having the manno config-
uration 42 (11 mg, 0.04 mmol) in 1,4-dioxane (0.5 mL)

T. Sakakibara et al. / Carbohydrate Research 342 (2007) 2339–2353
2353
were added 0.05 mL of H2O2 and a drop of M NaOH.
The mixture was stirred for 2 min. Similar workup of
the mixture gave a syrup (11 mg), the 1H NMR of which
showed it was a 1:1.3 mixture of 39 and 42, but gave no
evidence for formation of 40 and 41. For their physico-
chemical and spectral data see Ref. 9.
Similar reduction of 2-nitro-a-D-allopyranoside 46
(30 mg, 0.10 mmol) with LiAlD4 (16 mg, mmol) for
30 min gave 23 mg (89%) of 48 (after column chroma-
tography), of which the axial and equatorial positions
at C-2 and C-3, respectively, were deuterated; 1H
NMR: d 1.86 (d, J3,4 11.6 Hz, H-3a).
3.23. Reactions of 43 with lithium aluminum deuteride
References
1
The H NMR of 43 (CD3COCD3, 400 MHz): d 5.46 (s,
1. For example, (a) Morrison, R. T.; Boyd, R. N. Organic
Chemistry, 5th ed.; Allyn and Bacon: Boston, 1987,
Chapter 31; (b) Perlmutter, P. Conjugate Addition Reac-
tions in Organic Synthesis. In Tetrahedron Organic
Chemistry Series; Pergamon Press: Oxford, 1992; Vol. 9.
2. For example, (a) Sibi, M. P.; Manyem, S. Tetrahedron
2000, 56, 8033–8061; (b) Dorigo, A. E.; Morokuma, K.
J. Am. Chem. Soc. 1989, 111, 6524–6536; (c) Lucet, D.;
Toupet, L.; Mioskowski, C. J. Org. Chem. 1997, 62, 2682–
2683.
1H, H-1), 4.49 (s, 1H, H-3), 3.83 (d, 1H, J4,5 9.6 Hz, H-
4), 3.95 (ddd, 1H, H-5), 3.85 (t, 1H, J5,6a = J6a,6e
10.1 Hz, H-6a), 4.30 (dd, 1H, J5,6e 4.4 Hz, H-6e), 5.75
(s, 1H, PhCH), and 3.57 (s, 3H, OMe). Correlation
was observed between H-1 and OMe and between
PhCH and H-4 as well as H-6a in the NOESY spectrum.
To an ice-cooled solution of 2-nitro-a-D-mannopyr-
anoside 43 (29 mg, 0.09 mmol) in THF (2 mL) was
added LiAlH4 (16 mg, 0.42 mmol) with stirring under
N2. After stirring for 30 min at room temperature, the
mixture was partitioned between water and CHCl3,
and the organic layer was evaporated and chromato-
graphed with CHCl3 as an eluent to give 21 mg (84%)
3. Sakakibara, T.; Tachimori, Y.; Sudoh, R. Tetrahedron
1984, 40, 1533–1539.
4. For example, (a) Deslongchamps, P. Stereoelectronic
Effects in Organic Chemistry; Pergamon Press: Oxford,
1983; (b) Kirby, A. J. The Anomeric Effect and Related
Stereoelectronic Effects at Oxygen; Springer: Berlin, 1983.
5. (a) Johnson, F.; Malhotra, S. K. J. Am. Chem. Soc. 1965,
87, 5492–5493; (b) Johnson, F. Chem. Rev. 1968, 68, 375–
413.
6. (a) Baer, H. H.; Rank, W. Can. J. Chem. 1971, 49, 3192–
3196; (b) Sakakibara, T.; Kumazawa, S.; Nakagawa, T.
Bull. Chem. Soc. Jpn. 1970, 43, 2655.
7. Sakakibara, T.; Tachimori, Y.; Sudoh, R. Carbohydr. Res.
1984, 131, 197–208.
8. Sakakibara, T.; Sudoh, R. Carbohydr. Res. 1976, 50, 191–
196.
9. Sakakibara, T.; Sudoh, R. Carbohydr. Res. 1980, 85, 33–
40.
10. Davison, B. E.; Guthrie, R. D. J. Chem. Soc., Perkin
Trans. 1 1972, 658–662.
25
of 44: mp 184–184.5 °C, lit.21 186.5–187.5 °C; ½aꢂD
25
+118.8 (c 1, CHCl3), lit.21 ½aꢂD 126:9 ꢃ 2. 1H NMR data:
d 4.69 (d, 1H, J1,2 3.6 Hz, H-1), 3.70 ꢀ 4.80 (m 3H, H-2,
-5, -6a), 1.86 (q, 1H, J2,3a = J3a,3e = J3a,4 11.5 Hz, H-3a),
2.31 (td, 1H, J2,3e = J3e,4 4.6 Hz, H-3e), 3.54 (m, 1H, H-
4), 3.45 (dt, 1H, J5,6a 10.2, J5,6e 5.0 Hz, H-5), 3.77 (t, 1H,
J6a,6e 10.6 Hz, H-6a), 4.27 (dd, 1H, J6a,6e 10.5, J5,6e
5.6 Hz, H-6e), 5.52 (s, 1H, PhCH), 2.05 (br d, 1H,
J2,OH 2.1 Hz, OH), and 3.48 (s, 3H, OMe).
Similar treatment of 43 (26 mg) with LiAlD4 (20 mg)
in THF (4 mL) afforded the dideuterio derivative 45,
having the deuterium atom at the axial positions of
C-2 and C-3; d 2.31 (br d, 1H, H-3e).
11. Kazmi, S. N.-ul-H.; Ahmed, Z.; Khan, A. Q.; Malik, A.
Synth. Commun. 1988, 18, 151–156.
12. Spanevello, R. A.; Pellegrinet, S. C. Synth. Commun. 1995,
25, 3663–3670, Furthermore yield from 8 to 14 was
improved.
13. Abdel-Malik, M. M.; Peng, Q.-J.; Perlin, A. S. Carbohydr.
Res. 1987, 159, 11–23.
14. For example, Seta, A.; Ito, S.; Tokuda, K.; Tamura, T.;
Konda, Y.; Sakakibara, T. Carbohydr. Res. 1995, 267,
217–226.
15. For example, (a) Nakagawa, T.; Sakakibara, T. Carbo-
hydr. Res. 1987, 163, 227–237; (b) Baer, H. H.; Madumelu,
C. B. Can. J. Chem. 1978, 56, 1177–1182.
3.24. Reactions of 46 with lithium aluminum deuteride
1
The H NMR spectrum of 46 (CD3COCD3, 400 MHz):
d 6.03 (s, 1H, H-1), 4.36 (br s, 1H, H-3), 4.27 (dd, 1H,
J3,4 1.2, J4,5 8.9 Hz, H-4), 3.90 (ddd, 1H, H-5), 3.84 (t,
1H, J5,6a 10.0 Hz, H-6a), 4.32 (dd, 1H, J5,6e 4.2, J6a,6e
9.3 Hz, H-6e), 5.75 (s, 1H, PhCH), and 2.79 (s, 3H,
OMe). Correlation was observed between H-1 and
OMe and between PhCH and H-4 as well as H-6a in
the NOESY spectrum.
16. Sakakibara, T. unpublished data.
17. Baer, H. H.; Kienzle, F. Can. J. Chem. 1967, 45, 983–990.
18. Baer, H. H.; Neilson, T. Can. J. Chem. 1965, 43, 840–846.
19. Abdel-Malik, M. M.; Perlin, A. Carbohydr. Res. 1989,
190, 39–52.
20. Newth, F. H. J. Chem. Soc. 1959, 2717.
21. Vis, E.; Karrer, P. Helv. Chim. Acta 1954, 37, 378–381.
In the 1H NMR spectrum of a ꢀ4:1 mixture of 43 and
46 in CDCl3 (400 MHz), the signal due to H-1 of 43 was
observed at d 5.32 and that of 46 at d 5.98; these signals
correlated with OMe signals, respectively, in the
NOESY spectrum.