Stereocontrolled transformation of nitrohexofuranoses into cyclopentylamines via 2-oxabicyclo[2.2.1]heptanes. III: synthesis of enantiopure methyl (1S,2S,3R,4S,5R)-2-amino-3,4,5-trihydroxycyclopentanecarboxylate

Article abstract of DOI:10.1016/j.tetasy.2008.12.017

The first total synthesis of enantiopure methyl (1S,2S,3R,4S,5R)-2-amino-3,4,5-trihydroxycyclopentanecarboxylate was carried out according to our recent novel strategy for the transformation of nitrohexofuranoses into cyclopentylamines, which is based on

Full text of DOI:10.1016/j.tetasy.2008.12.017

Tetrahedron: Asymmetry 19 (2008) 2907–2912
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Stereocontrolled transformation of nitrohexofuranoses into
cyclopentylamines via 2-oxabicyclo[2.2.1]heptanes. III: synthesis
of enantiopure methyl (1S,2S,3R,4S,5R)-2-amino-3,4,5-
trihydroxycyclopentanecarboxylate
b,
a,b,
Fernando Fernández a, Juan C. Estévez a, , Fredy Sussman , Ramón J. Estévez
*
*
*
a Carbohydrate Research Group, Departamento de Química Orgánica, Universidade de Santiago de Compostela, 15782 Santiago de Compostela, Spain
b Molecular Modeling Research Group, Departamento de Química Orgánica, Universidade de Santiago de Compostela, 15782 Santiago de Compostela, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The first total synthesis of enantiopure methyl (1S,2S,3R,4S,5R)-2-amino-3,4,5-trihydroxycyclopentane-
carboxylate was carried out according to our recent novel strategy for the transformation of nitrohexof-
uranoses into cyclopentylamines, which is based on an intramolecular cyclization leading to 2-
oxabicyclo[2.2.1]heptane derivatives. Differences in reactivity for this key step were rationalized by using
molecular mechanism-based calculations.
Received 24 October 2008
Accepted 10 December 2008
Available online 20 January 2009
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
11?12, respectively. The results indicated that the latter cycliza-
tion is more efficient than the former, an outcome that could be ex-
plained by molecular mechanics-based calculations.
Carbohydrates1 and ‘naked sugars’2 provide abundant sources
of versatile synthons of great usefulness for the stereoselective
synthesis of natural products. These powerful synthetic strategies
are well-established tools of special interest for the preparation
of richly functionalized carbo- and heterocycles. Two approaches
have been applied to carbohydrate-based cyclizations. One of them
involves cyclization of an open chain carbohydrate derivative, and
the other includes the generation of a bicyclic derivative consisting
of the original sugar ring and the new ring, which is followed by
the opening of the sugar ring. Although several synthetic applica-
tions of this approach have been developed,3 the variant involved
2. Results and discussion
The reaction of the easily prepared nitroglucofuranose deriva-
tive 1 with trifluoroacetic acid and water, followed by anomeric
oxidation of the resulting hydroxy lactol 2 with bromine and bar-
ium carbonate afforded the lactone 3 as a yellow oil (94% yield
from 1). Reaction of 3 with triflic anhydride in pyridine furnished
the corresponding triflate 4, which when treated with TBAF in
THF readily underwent an intramolecular displacement of the tri-
in the intramolecular alkylation of the nitronate of the D-glucose
flate group by the carbanion
a to the nitro group to afford the bicy-
derivative 4 to give the bicyclic lactone 6a had not been explored
until our recent synthesis of b-peptide 7a,4 that includes a poly-
hydroxylated cyclopentane b-amino acid (see Scheme 1). This ap-
proach takes advantage of the sugar pool for the generation of
chemical diversity and the synthetic potential of nitroalkanes to
form carbon–carbon bonds ahead of transformations of the nitro
group into a variety of functionalities, including its reduction to
an amino group.5
Herein, we report two slight modifications of this route that al-
lowed us to prepare enantiomerically pure 3,4,5-trihydroxy-trans-
2-aminocyclopentanecarboxylic acid derivative 8a from D-glucose.
Both modifications allowed us to substantially improve the yield of
the key nitronate based intramolecular cyclization steps 5?6a and
clic b-nitrolactone 6a in 75% yield for the last two steps (previous
yield, 41%). This substantial yield improvement was achieved
when the concentration of triflate 4 in the reaction mixture was
raised from 0.10 M to 0.11 M; this compound was also allowed
to stand in vacuo for 12 h just before its transformation.
The stereochemical outcome of this key step leading to com-
pound 6a was explained by assuming that both bicyclic com-
pounds 6a and 6b could be formed from the nitronate of
compound 5 (Scheme 2). Under these reaction conditions, com-
pounds 6a and 6b should be in equilibrium with their common
nitronate 6c. At equilibrium, the thermodynamically more stable
compound 6a should be favored with respect to compound 6b,
where the NO2 and the OBn substituents are eclipsed. This explains
the remarkable stereoselectivity of the cyclization.
In a modification of our previous synthetic route leading to 7a,
the reaction of this bicyclic lactone with sodium methoxide in
* Corresponding authors.
E-mail address: ramon.estevez@usc.es (R.J. Estévez).
0957-4166/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2008.12.017

2908
F. Fernández et al. / Tetrahedron: Asymmetry 19 (2008) 2907–2912
O2N
BnO
O2N
BnO
O2N
BnO
O2N
BnO
O
i
O
O
iii
O
ii
OH
OH
OBn
O
O
O
O
OH
OTf
BnO
BnO
BnO
BnO
4
1
2
3
O2N
BnO
OBn
OH
O
O
6
5
1
iv
O
H
vi
v
O
HO
BnO
HO
O
X
X
4
NO2
2
3
O
OTf
Y
NHY
BnO
OBn
HO
H
5
6a
7
8
a) X=GlyOMe, Y=NHGlyZ
b) X=OMe, Y=NO2
a) X=OMe, Y=NH2
b) X=OMe, Y=NHCbz
vii
Scheme 1. Reagents and conditions: (i) TFA/H2O (1:1), rt, 19 h; (ii) Br2, BaCO3, dioxane/H2O (2:1), rt, 36 h (94% from 1); (iii) Tf2O, pyridine, CH2Cl2, ꢀ30 °C, 1.5 h; (iv) TBAF,
THF, rt, 6 h (75% from 3); (v) MeONa/MeOH (0.5 M), rt, 1.5 h, 65%; (vi) H2, Pd/C, MeOH, citric acid, 50 h; (vii) CbzCl, NaHCO3, MeOH, rt, 6 h (60% from 7b).
12b in moderate yield (46%). Here, we describe further studies
on this route, which allowed us to clarify and improve the yield
OBn
of this key cyclization, and to apply it to a new, longer but more
efficient preparation of the above-mentioned cyclopentane b-ami-
no acid derivative 8a.
H
NO2
O
O
OBn
6a
When nitroglucofuranose derivative 1 (the common starting
material for routes in Schemes 1 and 3) was reacted with acetyl
chloride in methanol, it provided a 90% yield of a 1.2:1 epimeric
mixture 9a and 9b, from which anomers 9a and 9b were isolated
carefully by column chromatography. Reaction of 9a with triflic
anhydride and pyridine, allowed us to obtain the key compound
10a, which when directly treated with TBAF in THF readily under-
went the expected C-alkylation of the starting nitronate 11a. The
resulting bicylic compound 12a, which was obtained in 91% yield
(not previously isolated), was easily identified by its spectroscopic
and analytical data. Additionally, its structure was confirmed by
X-ray crystallography.6 On the other hand, anomer 9b was simi-
larly converted into bicyclic pyranoside 12b (87% yield, previous
yield 46%), via compound 10b. A remarkable aspect of this route
is that the intramolecular cyclization leading to 12a was slightly
more efficient than that leading to its anomer 12b. On the other
hand, when the anomeric mixture of 9a and 9b was subjected to
this reaction sequence, a 89% yield of a 1.3:1 anomeric mixture
of 12a and 12b was obtained as seen from the 1H NMR spectrum
of this mixture.7 This anomeric ratio, which differs from those of
the starting mixture of 9a and 9b, additionally confirms the differ-
ent cyclization efficiencies of their derivatives 10a and 10b.
According to our plan, hydrolysis of the mixture 12a and 12b
followed by their immediate oxidation with sodium chlorite pro-
vided the desired 5-nitrocyclopentane-carboxylic acid 15, which
OBn
H
O2N
BnO
6
5
O
1
O
H
O
4
2
3
OTf
OBn
BnO
O
O
6c
N
5
BnO
H
H
O
O
O
OBn
NO2
6b
Scheme 2.
MeOH followed by hydrogenation of the resulting methyl 5-nitro-
cyclopentane carboxylate 7b in an acidic medium provided the de-
sired b-amino acid ester 8a, as a result of the reduction of the nitro
group to the amine and simultaneous removal of the benzyl-pro-
tecting groups. Finally, 8a was directly reacted with CbzCl in order
to transform it into its derivative 8b, with its amino group pro-
tected by a Cbz moiety.
Over the course of our previous work in this field, we obtained a
cyclopentylamine-based glycosidase inhibitor from D-glucose via
the bicyclic pyranoside 12b (Scheme 3).4a A salient aspect of this
approach is that the nitronate-based cyclization of the epimeric
mixture 11a and 11b led to the isolation of a single compound
OBn
O2N
O2N
O2N
BnO
6
5
X
Y
1
O
i
X
Y
X
ii
O
O
H
iii
O
4
1
BnO
2
BnO
NO2
OBn
Y
Y
3
OTf
BnO
OH
OTf
a) X=H, Y=OMe
b) X=OMe, Y=H
BnO
BnO
X
H
9
10
11 a) X=H, Y=OMe
a) X=H, Y=OMe
b) X=OMe, Y=H
12 a) X=H, Y=OMe (91%)
b) X=OMe, Y=H
b) X=OMe, Y=H (87%)
OBn
OBn
OBn
iv
CHO
CO2H
HO
HO
BnO
v
vi
H
NO2
OBn
7b
O
BnO
NO2
NO2
HO
H
13
14
15
Scheme 3. Reagents and conditions: (i) AcCl, MeOH, 0 °C to rt, 13 h (49% for 9a, 41% for 9b); (ii) Tf2O, pyridine, CH2Cl2, ꢀ30 °C, 1.5 h; (iii) TBAF, THF, rt, 14 h (91% for 12a from 9a,
87% for 12b from 9b); (iv) TFA/H2O (3:1), rt, 4 h; (v) NaClO2, NaH2PO4, 2-methyl-2-butene, t-BuOH/H2O (1:1); (vi) TMSCHN2, Et2O/MeOH (7:2), rt, 15 min (75% from 12a + 12b).

F. Fernández et al. / Tetrahedron: Asymmetry 19 (2008) 2907–2912
2909
was directly converted into the previously obtained methyl ester
derivative 7b when it was allowed to react with trimethylsilyldia-
zomethane. We assumed that the hydrolysis of the bicyclic glyco-
side mixture 12a and 12b produced the expected mixture of
hemiacetals 13, which is in equilibrium with its open aldehyde
form 14. Treatment of this mixture 13 and 14 with sodium chlorite
resulted in the oxidation of 14 to 15, a process that promotes the
displacement of this equilibrium to compound 14. This hypothesis
was supported by the 1H NMR of the mixture 13 and 14.8
and validated by our reaction yields could be explained by the pre-
ponderance of the ring flexibility over the electronic withdrawing
effect brought about by the presence of the OMe group. It would
be expected that bringing the reactant atoms to a value around
the van der Waals contact will distort the ring, a process that is
more favored when a OMe group is present at position C-1. To ver-
ify this hypothesis, we have calculated the fluctuations of the ring
dihedral angles in compounds 4 and 10a. The time evolution of tor-
sion angle around C-2–C-3 bond for these compounds is displayed
in Figure 2. As seen from this figure, the OMe-containing com-
pound 10a has larger amplitude oscillations than the C@O counter-
part, a result that supports our explanation for the difference in
reactivity amongst these compounds.
3. Structural rationale for the reactivity difference
The only structural difference between compound 4 and com-
pounds 10a or 10b is the presence of a C@O group at the C-1
position in the former reactant instead of the OMe group at the
same position in the latter reactants. As aforementioned, com-
pounds 10a and 10b cyclize more efficiently than compound 4.
The opposite behavior would be expected, based on the premise
that the positive charge induced at the C-2 position of lactone 4
by the carbonyl group should promote a nucleophilic attack on that
atom by the negatively charged carbon atom C-6. Therefore, one
would expect that the replacement of the carbonyl group by a
methoxy substituent would decrease the reactivity. Since elec-
tronic effects do not seem to explain these reactivity differences,
we searched for an alternative explanation based on steric allow-
ance afforded by the substituents.
Compound 4
Compound 10a
60
40
20
0
-20
-40
-60
0
100 200 300 400 500 600 700 800 900 1000 1100
Time (ps)
Figure 2. Time evolution of the dihedral angle centered around the C-2–C-3 bond,
as obtained from our MD simulations.
To shed some light into this option, we have carried out Molec-
ular Dynamic calculations aimed at explaining the reaction yields,
assuming that they could be rationalized in terms of the steric
effects that govern the cyclations of nitrosugars 4 and 10.
The main working hypothesis for our calculations was that the
central role of the substituents at C-1 is to act as ‘steering groups’
on the reaction, influencing it through their steric and/or electro-
static effects, rather than substantially altering the electronic
structure of these compounds. These modulators can facilitate or
impede the generation of what has been called ‘near attack confor-
mations’ (or NAC’s),9,10 that is, those in which the reacting centers
C-2 and C-6 in compounds 4, 10a or 10b are brought closer to a dis-
tance equal or below to the sum of their van der Waals radii. The
molecular dynamics (MDs)-based methodology utilized herein is
described in the Section 5.
In order to estimate the effect of replacing the carbonyl at posi-
tion C-1 of nitrolactone 4 by a OMe group, we have compared the
C-2–C-6 distance distribution functions of lactone 4 with those of
compounds 10a and 10b. The results are shown in Figure 1. A study
of this figure indicates that compound 4 has very few reactive con-
formations (around 20 at around 3.5 Å), which increase substan-
tially in the OMe-containing compounds 10 regardless of group
configuration, in qualitative agreement with experimental results.
4. Conclusion
In conclusion, we have adapted our strategy for the transforma-
tion of nitrosugars into carbasugars4 to the development of two
efficient alternative stereocontrolled transformations of D-glucose
into (1S,2S,3R,4S,5R)-2-amino-3,4,5-trihydroxycyclopentanecarb-
oxylic acid derivative 8a. The alternative via nitrosugarlactone 3
(Scheme 1), which consists of 13 steps, allowed us to obtain this
polyhydroxylated cyclopentane b-amino acid 8b in 15% overall
yield. On the other hand, the alternative via nitroglycosides 9a
and 9b is longer but more efficient, because target 8b is obtained
in 20% yield after 14 steps (Scheme 3).
Molecular dynamic simulations performed on the cyclization
substrates 4 and 10 allowed us to elaborate a hypothesis for the in-
creased reactivity of the latter compounds based on the larger flex-
ibility of their rings.
Work that is currently in progress is aimed at extending these
studies to hexoses other than D-glucose, in order to prepare a range
of polyhydroxylated cyclopentane b-amino acids prior the study of
the structural, physical, and biological properties of their b-pep-
tides. We will also apply our simulation protocols to the above-
mentioned hexoses, in order to determine whether the substituent
effects on the reactivity are due to steric effects, like in the case
presented here.
400
350
300
250
Compound 4
Compound 10a
Compound 10b
200
150
100
50
5. Experimental
5.1. Computer-based simulations
0
3.0
3.5
4.0
Distance
4.5
5.0
5.5
To determine the influence of the nature and stereochemical
conformation of the substituents, we subjected all molecules to a
molecular mechanics protocol that started with a 100-step steep-
est descent energy minimization stage to alleviate steric clashes.
In a second phase, these molecules underwent a molecular dynam-
ics (MD) protocol, starting with a 5-ps equilibration stage and end-
ing in a 1 ns production phase. The parameters utilized in our MD
calculations resembled those employed in the synthetic route.
Figure 1. Distribution of distances between reactive atoms C-2 and C-6 for
compounds 4, 10a, and 10b as obtained from MD production trajectories.
As pointed out above, one would expect that the replacement of
the carbonyl group by a methoxy substituent would decrease the
reactivity. The opposite behavior predicted by our calculations

2910
F. Fernández et al. / Tetrahedron: Asymmetry 19 (2008) 2907–2912
Hence, we use a temperature value of 300 K and a dielectric con-
stant of 7.6, the value that corresponds to THF, the solvent used
in the synthesis. The molecular mechanic simulations were per-
formed with the DISCOVER module in the INSIGHTII suite of pro-
grams.11 The force field used throughout these calculations was
CVFF,12 in a NVT ensemble regime. The present algorithm is far less
time consuming than quantum mechanical calculations, and does
not require parameter calibration that is needed for the transition
state molecular mechanics-based calculations.13
with dichloromethane (100 mL), washed with dilute hydrochloric
acid (100 mL) and brine (100 mL). The organic layer was dried
(anhydrous sodium sulfate) and concentrated to dryness to give
3,5-di-O-benzyl-6-deoxy-6-nitro-2-O-trifluoro-methanesulfonyl-
D-glucono-1,4-lactone 4 as a clear gum, which was used in the next
step without further purification after keeping it in vacuo
overnight.
5.5. (1S,4S,5S,6R,7R)-6,7-Dibenzyloxy-5-nitro-2-oxabicyclo [2.2.1]-
heptan-3-one 6a
5.2. Synthetic procedures
A
1 M solution of tetrabutylammonium fluoride in THF
Melting points were determined using a Kofler Thermogerate
apparatus and are uncorrected. Specific rotations were recorded
on a JASCO DIP-370 optical polarimeter. Infrared spectra were re-
corded on a MIDAC Prospect-IR spectrophotometer. Nuclear mag-
netic resonance spectra were recorded on a Bruker DPX-250
apparatus. Mass spectra were obtained on a Hewlett Packard
5988A mass spectrometer. Elemental analyses were obtained from
the Elemental Analysis Service at the University of Santiago de
Compostela. Thin layer chromatography (TLC) was performed
using Merck GF-254 type 60 silica gel and ethyl acetate/hexane
mixtures as eluents; the TLC spots were visualized with Hanessian
mixture. Column chromatography was carried out using Merck
type 9385 silica gel.
(4.93 mL) was added to a solution of triflate 4 in THF (44 mL),
and the resulting mixture was stirred under argon for 6 h. The sol-
vent was evaporated, and the residue dissolved in dichlorometh-
ane (100 mL). The solution was washed with water (3 ꢁ 100 mL),
and the organic layer was dried (anhydrous sodium sulfate) and
concentrated in vacuo. The resulting gum was purified by flash col-
umn chromatography (ethyl acetate/hexane 1:4) to give the
(1S,4S,5S,6R,7R)-6,7-dibenzyloxy-5-nitro-2-oxabicyclo[2.2.1]hep-
tan-3-one 6a (1.36 g, 3.68 mmol, 75% from 3) as a yellow oil.
½
a 2D1
ꢂ
¼ ꢀ35:0 (c 0.8, CHCl3). 1H NMR (CDCl3, ppm): 4.03 (s, 1H, H-
1); 4.28–4.30, 4.51–4.54 (2 ꢁ m, 4H, CH2Ph, H-6, H-7); 4.67 (dd,
1H, J5,6 = 4.0 Hz, J5,4 = 2.3 Hz, H-5); 4.79 (d, 1H, JH,H = 11.9 Hz,
´
CH2Ph); 4.81 (d, 1H, JH,H = 11.9 Hz, CH2Ph); 5.05 (d, 1H,
´
J4,5 = 2.3 Hz, H-4); 7.17–7.40 (m, 10H, 10 ꢁ CHar). 13C NMR (CDCl3,
ppm): 50.29, 72.57, 72.66, 80.22, 81.42, 82.19, 84.78, 127.97,
128.21, 128.31, 128.63, 128.71, 128.79, 135.39, 136.60, 169.33. IR
5.3. 3,5-Di-O-benzyl-6-deoxy-6-nitro-D-glucono-1,4-lactone 3
3,5-Di-O-benzyl-6-deoxy-1,2-O-isopropylidene-6-nitro-
a-
D-glu
(NaCl, m
max, cmꢀ1): 1804 (st, CO); 1554, 1361 (st, NO2). MS (CI,
cofuranose 1 (2.04 g, 4.76 mmol) was dissolved in a mixture of tri-
fluoroacetic acid/water (1:1, 120 mL), and the reaction mixture
was stirred at room temperature until the starting material had
been consumed (TLC, ethyl acetate/hexane 1:5) (19 h). The solvent
was evaporated in vacuo, and the residue was coevaporated with
m/z,%): 370 (6, [M+H]+); 91 (100, [CH2Ph]+). Anal. Calcd for
C20H19NO6: C, 65.03; H, 5.18; N, 3.79. Found: C, 65.07; H, 5.36; N,
3.51.
5.6. Methyl (1S,2R,3S,4R,5S)-2,4-Dibenzyloxy-3-hydroxy-5-
toluene (3 ꢁ 50 mL) to give 3,5-di-O-benzyl-6-deoxy-6-nitro-
D
-
nitrocyclopentanecarboxylate 7b
glucofuranose 2 as a clear gum, which was used in the next step
without further purification.
Bicyclolactone 6a (0.20 g, 0.54 mmol) was added over a solution
of MeONa in MeOH (1 mL, 0.5M), and the mixture was stirred un-
der argon at rt for 1.5 h. This mixture was then concentrated to
dryness and purified by flash column chromatography (ethyl ace-
tate/hexane 1:2) to give methyl ester 7b (0.13 g, 0.32 mmol, 65%
This crude reaction (1.85 g, 4.75 mmol) was dissolved in diox-
ane/water (2:1, 90 mL). Barium carbonate (1.03 g, 5.23 mmol)
and then bromine (0.6 mL, 11.88 mmol) were added, and the reac-
tion mixture was stirred for 36 h at room temperature with the
exclusion of light. The reaction mixture was quenched with satu-
rated aqueous sodium thiosulfate solution (until the mixture was
colorless), and the mixture was then extracted with ethyl acetate
(3 ꢁ 200 mL). The combined ethyl acetate extracts were dried over
anhydrous sodium sulfate and evaporated to dryness. The crude
residue was purified by flash column chromatography (ethyl ace-
yield) as a yellow oil. ½a D21
ꢂ
¼ þ67:0 (c 1.4, CHCl3). 1H NMR (CDCl3,
ppm): 2.50 (br s, 1H, H-1); 3.64–3.68 (m, 1H, H-4); 3.75 (s, 3H,
OCH3); 4.07–4.09 (m, 1H, H-2); 4.18–4.20 (m, 1H, H-3); 4.53–
4.79 (m, 5H, 2 ꢁ CH2Ph, OH); 5.41 (t, 1H, J = 6.6 Hz, H-5); 7.25–
7.36 (m, 10H, 10 ꢁ CHar). 13C NMR (CDCl3, ppm): 51.34, 52.87,
71.77, 73.10, 74.00, 82.89, 83.19, 88.68, 127.73, 127.81, 127.85,
tate/hexane 1:2) to give 3,5-di-O-benzyl-6-deoxy-6-nitro-
ono-1,4-lactone 3 (1.74 g, 4.49 mmol, 94% from 1) as a clear oil:
D
-gluc-
128.31, 128.42, 128.64, 136.46, 136.92, 170.17. IR (NaCl, mmax,
cmꢀ1): 3381 (br, OH); 1796 (st, CO); 1578, 1386 (st, NO2). MS (CI,
m/z, %): 402 (2, [M+H]+); 91 (100, [CH2Ph]+). Anal. Calcd for
C21H23NO7: C, 62.83; H, 5.78; N, 3.49. Found: C, 63.02; H, 5.94;
N, 3.61.
½
a 2D5
ꢂ
¼ þ17:9 (c 2.8, CHCl3). 1H NMR (CDCl3, ppm): 4.10–4.17 (m,
1H); 4.38–4.87 (m, 10H); 7.08–7.36 (m, 10H, 10 ꢁ CHar). 13C NMR
(CDCl3, ppm): 71.68, 72.52, 74.39, 75.12, 75.92, 78.18, 79.82,
127.88, 128.08, 128.27, 128.51, 128.62, 136.39, 136.47, 174.83. IR
(NaCl,
m
max, cmꢀ1): 3442 (br, OH); 1791 (st, CO) 1555, 1380 (st,
5.7. Methyl (1S,2S,3R,4S,5R)-2-Benzyloxycarbonyl-amino-3,4,5-
trihydroxycyclopentanecarboxylate 8b
NO2). MS (CI, m/z, %): 386 (2, [MꢀH]+); 91 (100, [CH2Ph]+). Anal.
Calcd for C20H21NO7: C, 62.01; H, 5.46; N, 3.62. Found: C, 61.77;
H, 5.53; N, 3.52.
Compound 7b (0.35 g, 0.87 mmol) was dissolved in methanol
(7.2 mL), and the solution was deoxygenated by bubbling argon
through it. Next, Pd–C (0.10 g, 10% w/w) and citric acid (0.20 g,
1.04 mmol) were added, and the suspension was stirred at rt under
a hydrogen atmosphere of 1 atm for 50 h. The reaction mixture was
filtered through Celite, eluted with methanol, and the filtrate evap-
orated in vacuo to give the corresponding amine 8a (0.16 g,
0.87 mmol) that was directly dissolved in MeOH (7.2 mL). After
saturated aqueous sodium bicarbonate solution (4.5 mL) and ben-
zyl chloroformate (0.14 mL, 1.01 mmol) were added, the resulting
5.4. 3,5-Di-O-benzyl-6-deoxy-6-nitro-2-O-trifluoro-methanesul-
fonyl-D-glucono-1,4-lactone 4
Compound 3 (1.91 g, 4.93 mmol) was dissolved in dry dichloro-
methane (40 mL), and the solution was then cooled down to
ꢀ30 °C under argon. Pyridine (1.5 mL) and trifluoromethanesulfo-
nic anhydride (1.20 mL, 7.40 mmol) were added, and the mixture
was stirred at ꢀ30 °C for 1.5 h. The reaction mixture was diluted

F. Fernández et al. / Tetrahedron: Asymmetry 19 (2008) 2907–2912
2911
mixture was stirred at rt for 6 h, and then concentrated to dryness.
The residue was partitioned between ethyl acetate and water, and
the aqueous portion extracted with ethyl acetate (4 ꢁ 20 mL). The
combined organic layers were dried over anhydrous sodium sul-
fate, filtered and the solvent was evaporated in vacuo. The residue
was purified by flash column chromatography (dichloromethane/
methanol 9:1) to give methyl (1S,2S,3R,4S,5R)-2-benzyloxycarbon-
ylamino-3,4,5-trihydroxycyclopentane-carboxylate 8b (0.17 g,
and brine (10 mL). The organic layer was dried (anhydrous sodium
sulfate) and concentrated to dryness to give the corresponding tri-
flate 10a as a clear gum, which was used in the next step without
further purification.
A 1 M solution of tetrabutylammonium fluoride in THF (0.5 mL)
was added to a solution of the previously obtained triflate in THF
(4.5 mL), and the resulting mixture was stirred under argon for
14 h. The solvent was evaporated, and the residue dissolved in
dichloromethane (10 mL). The solution was washed with water
(3 ꢁ 10 mL), and the organic layer was dried (anhydrous sodium
sulfate) and concentrated in vacuo. The resulting gum was purified
by flash column chromatography (ethyl acetate/hexane 1:4) to give
compound 12a (0.17 g, 0.45 mmol, 91% from 9a) as a yellow oil.
0.52 mmol, 60% from 7b) as a colorless oil. ½a D26
¼ þ125:5 (c 2.0,
ꢂ
CHCl3). 1H NMR (acetone-d6, ppm): 2.65–2.71 (br s, 1H, H-1);
2.97 (br s, 1H, OH); 3.60 (s, 3H, OCH3); 3.81–3.86 (m, 1H, H-2);
4.00–4.06 (m, 2H, H-4, H-5); 4.15–4.21 (m, 2H, H-3, -OH); 4.55
(br s, 1H, OH); 5.03 (d, 1H, JH,H = 12.6 Hz, CH2Ph); 5.09 (d, 1H,
´
JH,H = 12.6 Hz, CH2Ph); 6.75 (br s, 1H, NH); 7.27–7.37 (m, 5H,
½
a 2D2
ꢂ
¼ ꢀ23:7 (c 1.2, CHCl3). 1H NMR (CDCl3, ppm): 3.41 (s, 3H,
´
5 ꢁ CHar). 13C NMR (acetone-d6, ppm): 53.04, 55.40, 60.69, 67.56,
OCH3); 3.52 (m, 1H, H-4); 4.20 (m, 1H, H-7); 4.40 (m, 1H, H-1);
76.43, 77.96, 79.43, 129.56, 129.59, 130.12, 139.13, 158.07,
4.48 (2 ꢁ d, 1H, JH,H = 11.5 Hz, CH2Ph); 4.74–4.78 (m, 3H, CH2Ph,
´
174.87. IR (NaCl,
m
max, cmꢀ1): 3465 (br, NH, OH); 1731 (st, CO);
H-6); 5.03 (s, 1H, H-1); 5.26 (d, 1H, J3,4 = 3.0 Hz, H-3); 7.25–7.36
(m, 10H, 10 ꢁ CHar). 13C NMR (CDCl3, ppm): 48.26, 55.15, 72.06,
72.23, 76.49, 81.20, 83.01, 87.27, 103.18, 126.77, 127.36, 127.62,
1683 (st, NCO). MS (CI, m/z, %): 326 (1, [M+H]+); 315 (62,
[MꢀCH3]+); 91 (100, [CH2Ph]+). Anal. Calcd for C15H19NO7: C,
55.38; H, 5.89; N, 4.31. Found: C, 55.42; H, 5.77; N, 4.21.
127.68, 136.48, 137.17. IR (NaCl,
m
max, cmꢀ1): 1550, 1378 (st,
NO2). MS (CI, m/z, %): 386 (3, [M+H]+); 91 (100, [CH2Ph]+). Anal.
Calcd for C21H23NO6: C, 65.44; H, 6.02; N, 3.63. Found: C, 65.58;
H, 6.09; N, 3.78.
5.8. Methyl 3,5-di-O-benzyl-6-deoxy-6-nitro-D-glucofurano-
sides 9a and 9b
Acetyl chloride (0.74 mL, 10.34 mmol) was added to a cooled
(0 °C) solution of 3,5-di-O-benzyl-6-deoxy-6-nitro- -glucofura-
5.10. (1S,3R,4S,5R,6R,7R)-6,7-Dibenzyloxy-3-methoxy-5-nitro-
2-oxabicyclo[2.2.1]heptane 12b
D
nose 1 (0.74 g, 1.72 mmol) in dry methanol (11.2 mL) and the
resulting mixture was allowed to warm to rt and stirred for 13 h.
The mixture was concentrated to dryness, to give a (1.2:1) mixture
of epimers 9a and 9b. The residue was subjected to flash column
Compound 9b (0.15 g, 0.37 mmol) was dissolved in dry dichlo-
romethane (3.0 mL), and the solution cooled to ꢀ30 °C under nitro-
gen. Pyridine (0.11 mL) and trifluoromethanesulfonic anhydride
(0.09 mL, 0.56 mmol) were added, and the mixture was stirred
for 1.5 h at ꢀ30 °C. The reaction mixture was diluted with dichlo-
romethane (8 mL), washed with dilute hydrochloric acid (8 mL)
and brine (8 mL). The organic layer was dried (anhydrous sodium
sulfate) and concentrated to dryness to give the corresponding tri-
flate 10b as a clear gum, which was used in the next step without
further purification.
chromatography (ethyl acetate/hexane 1:3) to give the
a-anomer
9a (0.34 g, 0.84 mmol, 49%) as a colorless oil and the b-anomer
9b (0.29 g, 0.72 mmol, 41%) as a crystalline solid.
Spectroscopic data for 9a: ½a D22
ꢂ
¼ ꢀ6:0 (c 1.0, CHCl3). 1H NMR
(CDCl3, ppm): 2.84 (d, 1 H, JOH,2 = 5.9 Hz, OH); 3.48 (s, 3H, OCH3);
4.04–4.06 (m, 1H, H-5); 4.21–4.31 (m, 3H, H-2, H-3, H-4); 4.53–
4.82 (m, 6H, 2 ꢁ CH2Ph, H-6, H-60); 5.01 (d, 1H, J1,2 = 4.4 Hz, H-1);
7.18–7.35 (m, 10H, 10 ꢁ CHar). 13C NMR (CDCl3, ppm): 55.66,
71.35, 72.73, 74.23, 75.94, 76.78, 77.29, 82.94, 101.97, 127.40,
A
1 M solution of tetrabutylammonium fluoride in THF
(0.37 mL) was added to a solution of the previously obtained tri-
flate in THF (3.3 mL), and the resulting mixture was stirred under
argon for 14 h. The solvent was evaporated, and the residue dis-
solved in dichloromethane (8 mL). The solution was washed with
water (3 ꢁ 8 mL), and the organic layer was dried (anhydrous so-
dium sulfate) and concentrated in vacuo. The resulting gum was
purified by flash column chromatography (ethyl acetate/hexane
1:4) to give compound 12b (0.12 g, 0.32 mmol, 87% yield from
127.50, 128.22, 128.27, 137.05, 137.08. IR (NaCl,
m
max, cmꢀ1):
3455 (br, OH); 1558, 1376 (st, NO2). MS (CI, m/z,%) = 403 (8,
[M]+); 402 (23, [MꢀH]+); 91 (100, [CH2Ph]+). Anal. Calcd for
C21H25NO7: C, 62.52; H, 6.25; N, 3.47. Found: C, 62.51; H, 6.24;
N, 3.50.
Spectroscopic data for 9b: mp 91–93 °C. ½a D22
¼ ꢀ77:0 (c 1.0,
ꢂ
CHCl3). 1H NMR (CDCl3, ppm): 2.84 (br s, 1H, OH); 3.39 (s, 3H,
OCH3); 3.93 (d, 1H, J3,4 = 4.2 Hz, H-3); 4.22 (s, 1H, H-2); 4.40–
4.55 (m, 1H, H-4); 4.50 (m, 1H, CH2Ph); 4.55–4.58 (m, 2H, CH2Ph);
4.59–4.69 (m, 3H, CH2Ph, H-5, H-60); 4.80 (s, 1H, H-1); 4.89 (dd, 1H,
9b) as a yellow solid. Mp 99–101 °C. ½a D19
¼ ꢀ47:2 (c 1.0, CHCl3).
ꢂ
1H NMR (CDCl3, ppm): 3.49 (s, 3 H, OCH3); 3.67 (br s, 1H, H-4);
4.12 (s, 1H, H-7); 4.22–4.24 (m, 1H, H-1); 4.29 (d, 1H,
JH,H = 11.5 Hz, CH2Ph); 4.48 (d, 1H, JH,H = 11.5 Hz, CH2Ph); 4.78
J6,6 = 13.2 Hz, J6,5 = 2.8 Hz, H-6); 7.20–7.38 (m, 10H, 10 ꢁ CHar). 13
C
´
´
´
NMR (CDCl3, ppm): 55.89, 71.83, 72.67, 75.02, 77.19, 77.29, 80.53,
82.35, 109.92, 127.55, 127.58, 128.09, 128.12, 136.97. IR (NaCl,
(d, 1H, JH,H = 12.1 Hz, CH2Ph); 4.82 (m, 1H, H-6); 4.86 (d, 1H,
´
JH,H = 12.1 Hz, CH2Ph); 4.99 (s, 1H, H-5); 5.16 (d, 1H, J3,4 = 3.0
´
m
max, cmꢀ1): 3448 (br, OH); 1551, 1379 (st, NO2). MS (CI, m/z,
Hz, H-3); 7.27–7.47 (m, 10H, 10 ꢁ CHar). 13C NMR (CDCl3, ppm):
47.97, 56.18, 72.02, 72.16, 79.10, 81.45, 83.19, 83.57, 101.93,
127.81, 127.85, 127.89, 128.18, 128.39, 136.44, 137.83. IR (NaCl,
%) = 403 (5, [M]+); 402 (11, [MꢀH]+); 91 (100, [CH2Ph]+). Anal.
Calcd for C21H25NO7: C, 62.52; H, 6.25; N, 3.47. Found: C, 62.55;
H, 6.30; N, 3.47.
m
max, cmꢀ1): 1552, 1371 (st, NO2). MS (CI, m/z, %): 386 (2,
[M+H]+); 91 (100, [CH2Ph]+). Anal. Calcd for C21H23NO6: C, 65.44;
H, 6.02; N, 3.63. Found: C, 65.49; H, 6.05; N, 3.67.
5.9. (1S,3S,4S,5R,6R,7R)-6,7-Dibenzyloxy-3-methoxy-5-nitro-2-
oxabicyclo[2.2.1]heptane 12a
5.11. Methyl (1S,2R,3S,4R,5S)-2,4-dibenzyloxy-3-hydroxy-5-
Compound 9a (0.20 g, 0.50 mmol) was dissolved in dry dichlo-
romethane (4.0 mL), and the solution cooled to ꢀ30 °C under nitro-
gen. Pyridine (0.15 mL) and trifluoromethanesulfonic anhydride
(0.12 mL, 0.74 mmol) were added, and the mixture was stirred
for 1.5 h at ꢀ30 °C. The reaction mixture was diluted with dichlo-
romethane (10 mL), washed with dilute hydrochloric acid (10 mL)
nitrocyclopentanecarboxylate 7b
A
mixture (1.3:1) of compounds 12a and 12b (0.52 g,
1.35 mmol) was dissolved in trifluoroacetic acid/water (3:1,
12 mL) and stirred at room temperature until the starting material
had been consumed (TLC, ethyl acetate/hexane 1:2) (4 h). The sol-

2912
F. Fernández et al. / Tetrahedron: Asymmetry 19 (2008) 2907–2912
Mitchell, H. J.; Nicolau, K. C. Angew. Chem., Int. Ed. 2001, 40, 1576–1624; (d)
Hollingsworth, R. I.; Wang, G. Chem. Rev. 2000, 100, 4267–4282; (e) Knapp, S.
Chem. Rev. 1995, 95, 1859–1876; (f) Hanessian, S. Aldrichim. Acta 1989, 22, 3–14.
2. (a) Vogel, P. Curr. Org. Chem. 2000, 4, 455–480; (b) Vogel, P.; Sevin, A.-F.;
Kernen, P.; Bialecki, M. Pure Appl. Chem. 1996, 68, 719–722.
vent was evaporated in vacuo and the residue was coevaporated
with toluene (3 ꢁ 20 mL) to give an unstable mixture of isomers
13 and 14 (0.32 g, 0.86 mmol), which was directly dissolved
in tert-butanol/water (1:1, 20 mL). After 2-methyl-2-butene
(6.6 mL), sodium chlorite (1.76 g, 19.43 mmol), and sodium dihy-
drogenphosphate dihydrate (1.76 g, 9.72 mmol) were added, the
mixture was stirred for 1 h. until the starting material had been
consumed (TLC, ethyl acetate/hexane 1:1). Water (15 mL) was
added, and the solution extracted with ethyl acetate (4 ꢁ 20mL).
The combined ethyl acetate extracts were dried over anhydrous so-
dium sulfate and evaporated to dryness. The resulting crude car-
boxylic acid 15 (0.33 g, 0.85 mmol) was directly dissolved in
ethyl ether/methanol (7:2, 9 mL). A 2 M solution of trimethylsilyl-
diazomethane in ethyl ether (0.5 mL, 1.1 mmol) was added, and
the mixture was stirred at rt for 15 min (TLC, ethyl acetate/hexane
3:1). The mixture was concentrated to dryness and purified by
flash column chromatography (ethyl acetate/hexane 1:2) to give
the methyl ester 7b (0.35 g, 0.87 mmol, 75% from 12a + 12b) as a
yellow oil, with the same data as described before.
3. (a) Estevez, J. C.; Fairbanks, A. J.; Fleet, G. W. J. Tetrahedron 2001, 54, 13591–
13620; (b) Long, D.; Frederiksen, S. M.; Marquess, D. G.; Lane, A.; Watkin, D. J.;
Winkler, D. A.; Fleet, G. W. J. Tetrahedron Lett. 1998, 39, 6091–6094; (c) Krulle,
T. M.; De la Fuente, C.; Watson, K. A.; Gregoriou, M.; Jonhson, L. N.; Tsitsanou, K.
E.; Zographos, S. E.; Oikonomakos, N. G.; Fleet, G. W. J. Synlett 1997, 211–213;
(d) Fairbanks, A. J.; Fleet, G. W. J. Tetrahedron 1995, 51, 3881–3894; (e) Hsia, K.
Y.; Ward, P.; Lamont, R. B.; Lilley, P. M. deQ.; Watkin, D. J.; Fleet, G. W. J.
Tetrahedron Lett. 1994, 35, 4823–4826; (f) Estevez, J. C.; Fairbanks, A. J.; Hsia, K.
Y.; Ward, P.; Fleet, G. W. J. Tetrahedron Lett. 1994, 35, 3361–3364; (g) Fairbanks,
A. J.; Elliot, R. P.; Smith, C.; Hui, A.; Way, G.; Storer, R.; Taylor, H.; Watkin, D. J.;
Wincherter, B. G.; Fleet, G. W. J. Tetrahedron Lett. 1993, 34, 7953–7996.
4. For the first total synthesis of polyhydroxylated cyclopentane b-amino acids,
see: (a) Soengas, R. G.; Estevez, J. C.; Estevez, R. J. Org. Lett. 2003, 5, 1423–1425;
(b) Soengas, R. G.; Pampin, M. B.; Estevez, J. C.; Estevez, R. J. Tetrahedron:
Asymmetry 2005, 16, 205–211.
5. Noboru, O. In The Nitro Group in Organic Synthesis; Feuer, H., Ed.; Organic Nitro
Chemistry Series; Wiley-VCH, 2001.
6. Crystallographic data for the structure of compound 12b have been deposited
with the Cambridge Crystallographic Data Centre as supplementary
publication no. CCDC-205610. Copies of the data can be obtained free of
charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax
(+44)1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
Acknowledgments
7. The 1H NMR spectrum of the epimeric mixture 12a + 12b displayed two
singlets at 3.41 ppm and 3.49, both due to the anomeric OMe substituents,
respectively. A ratio 1.1:1.0 for both anomers was deduced from these signals.
8. The 1H NMR spectrum of the mixture 13+14 includes two signals at 4.70 ppm
and at 4.85 ppm, corresponding to the H-3 protons of anomers 13. The signal
displayed at 9.45 ppm was attributed to the aldehyde proton of compound 14.
9. (a) Bruice, T. C.; Lightstone, F. C. Acc. Chem. Res. 1999, 32, 127; (b) Lightstone, F.
C.; Bruice, T. C. J. Am. Chem. Soc. 1996, 118, 2595.
We thank the Spanish Ministry of Science and Innovation for
financial support (CTQ2005-00555) and for F.P.U. grants to Fer-
nando Fernández Nieto, and the Xunta de Galicia for financial sup-
port and for a grant to Fredy Sussman.
10. Perakyla, M.; Kollman, P. A. J. Phys. Chem. A 1999, 103, 8067.
11. InsightII, and Discover are trademarks of Accelrys Corp., 2002, San Diego, CA
92121, 2002.
12. Dauber-Osguthorpe, P.; Roberts, V. A.; Osguthorpe, D. J.; Wolf, J.; Genest, M.;
Hagler, A. T. Proteins 1988, 4, 31.
References
1. For a book on this topic, see: Hanessian, S. Total Synthesis of Natural Products; The
Chiron Approach; Pergamon: Oxford, 1993; For reviews, see: (a) Jarosz, S. Chim.
Oggi 2006, 24, 58–61; (b) Timmer, M. S. M.; Verhelst, S. H. L.; Grotenbreg, G. M.;
Overhand, M. O.; Overkleeft, H. S. Pure Appl. Chem. 2005, 77, 1173–1181; (c)
13. Eksterowicz, J. B.; Houk, K. N. Chem. Rev. 1993, 93, 2409.