Aminocyclopentanols as sugar mimics. Synthesis from unsaturated bicyclic lactones by Overman rearrangement

Article abstract of DOI:10.1039/b710232a

Bicyclic cyclopentane lactones, prepared from bromodeoxyaldonolactones, were transformed into aminocyclopentanols with an Overman rearrangement as the key step. Two of the compounds prepared, 7 and 19, were found to be good inhibitors of jack bean α-mannosidase and β-d-N- acetylglucosaminidase, respectively. The Royal Society of Chemistry.

Full text of DOI:10.1039/b710232a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Aminocyclopentanols as sugar mimics. Synthesis from unsaturated bicyclic
lactones by Overman rearrangement
Marie Bøjstrup, Mette Fanefjord and Inge Lundt*
Received 5th July 2007, Accepted 3rd August 2007
First published as an Advance Article on the web 22nd August 2007
DOI: 10.1039/b710232a
Bicyclic cyclopentane lactones, prepared from bromodeoxyaldonolactones, were transformed into
aminocyclopentanols with an Overman rearrangement as the key step. Two of the compounds
prepared, 7 and 19, were found to be good inhibitors of jack bean a-mannosidase and
b-D-N-acetylglucosaminidase, respectively.
group synthesized an aminocyclopentanol with a b-D-galacto
substitution pattern, III (Fig. 1), and showed that this was
Introduction
selective towards b-galactosidases over a-galactosidases.7 Based on
these findings, they suggested that aminocyclopentanols could be
considered as mimics of the protonated glycosides governed by the
configuration of the amino group corresponding to the anomeric
configuration. Thus, it should be possible to synthesize selective
inhibitors of a- or b-glycosidases by matching the stereochemistry
of the aminocyclopentanol with the relevant carbohydrate. An
investigation of the structure–activity relationship has just recently
been published by Reymond and co-workers.8
Work in our group has recently been focused on the synthesis
of carbohydrate mimics, especially the synthesis of carbasugars
from easily available carbohydrate-based starting materials.9 In
particular, the building block 1, synthesized from a bromod-
eoxyheptonolactone derived from carbohydrates, has been shown
to be well suited for the synthesis of several cyclopentanols10,11
(Scheme 1) and aminocyclopentanols.11
Interest in obtaining access to polyhydroxylated aminocyclopen-
tanes has emerged in recent years, since they have been shown to
exhibit a wide range of biological activities.1 An important feature
is their ability to inhibit glycosidases, since it is assumed that they
are transition state mimics of the glycosyl cation formed during the
hydrolysis of glycosides.2 The first synthetic aminocyclopentanol
was prepared by Farr and co-workers in 1990, the so-called
Merrell–Dow cyclopentylamine (MDC) I (Fig. 1).3 The three
hydroxyl groups in MDC have configurations corresponding to
those of the hydroxy groups at the 2, 3, and 4 positions in D-
mannopyranose. This aminocyclopentanol can be viewed as a
ring-contracted analogue of a-D-mannose where the ring oxygen
is missing but having an external amino group for protonation
by the glycosidase. MDC was shown to be a potent inhibitor
of a-mannosidase (jack beans) with an IC50 value of 62 nM.3 A
computational study3,5 showed good overlap of two of the three
hydroxy groups of I with the 2-OH and 3-OH groups of the “flap-
up” mannopyranosyl cation.4,5 Furthermore, the N-methyl group
appeared to be located on the a-face of the ring close to C-1,
suggesting that I could mimic the oxocarbenium ion as well as the
protonated substrate.
Scheme 1
In the present work we further investigate the use of building
block 1 and its enantiomer ent-1 for the synthesis of aminocy-
clopentanols with stereochemistry mimicking common sugars in
order to find new selective and potent glycosidase inhibitors.
Fig. 1
Farr and co-workers never expanded the concept to other
analogues of common carbohydrates. This was done years later
by Reymond and co-workers in the synthesis of an aminocy-
clopentanol II (Fig. 1) in which the substitution pattern matches
that of the parent carbohydrate, L-fucose. However, since the
configuration of the carbon simulating C-1 in the sugar has
the amino group in the “b”-configuration, it was shown to be
only a weak inhibitor of a-L-fucosidase.6 Soon after, the same
Results and discussion
In order to obtain an amino functionality adjacent to the carbon
side chain, either C-1 or C-6 of the building block 1 must be
functionalized with a nitrogen. Recently we have published a
method for the incorporation of nitrogen at C-1 giving access
to mimics of D-sugars,12 while introduction of nitrogen at C-6 will
give access to mimics corresponding to L-sugars (Fig. 2). To pursue
the latter strategy we took advantage of the allylic alcohol motif
in 1 by investigation of the Overman rearrangement, which is a
[3,3]-sigmatropic rearrangement of allylic trichloroacetimidates.13
Department of Chemistry, Technical University of Denmark, Building 201,
DK-2800, Kgs. Lyngby, Denmark. E-mail: il@kemi.dtu.dk; Fax: +45
45933968; Tel: +45 45252133
3164 | Org. Biomol. Chem., 2007, 5, 3164–3171
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Fig. 2
This rearrangement would result in introduction of the desired
amino function at C-6.
Scheme 3 Synthesis of aminocyclopentanol 7. Reagents and conditions:
a) OsO4, NMO, CH2Cl2, rt, overnight; b) 2,2-dimethoxypropane, acetone,
H2SO4, rt, 20 min (80%); c) aq. HCl (6 M), reflux, 2.5 h (79%); d) Ca(BH4)2,
ethanol, −20 ◦C, overnight; e) Ac2O, pyridine, rt, overnight (61%); f) aq.
HCl (4 M), 60 ◦C, 4.5 h (quant.).
Thus, 1 was treated with NaH and trichloroacetonitrile to give
the crude trichloroacetimidate 2, which was shown by NMR to
be sufficiently pure to be used directly in the subsequent synthesis
(Scheme 2). The imidate 2 was therefore subjected to a thermally
mediated Overman rearrangement by reflux in xylene to give the
allylic trichloroacetamide 3 in good yield.
Scheme 4 Epoxidation of 3. Reagents and conditions: a) m-CPBA,
CH2Cl2, reflux, 4 d (98%); b) aq. HCl (4 M), 90 ◦C, 4 h.
The bicyclic cyclopentane lactone 1 was prepared from a
2,3-unsaturated 6-bromo-6-deoxyheptonolactone by a radical-
induced carbocyclisation.9 For the synthesis of ent-1 by a similar
carbacyclisation procedure, the precursor should be the 2,3-
unsaturated heptonolactone 9 (Scheme 5). 2,3-Unsaturated hep-
tonolactones are available by diastereoselective coupling of an
enantiomerically pure aldehyde with 2-(trimethylsilyloxy)furan
following the concept developed by Casiraghi and co-workers.15
For the synthesis of 9, 2,3-isopropylidene-L-glyceraldehyde and
2-(trimethylsilyloxy)furan were used.15,16 Treatment of 9 with HBr
in HOAc gave the 7-bromo-7-deoxyheptonolactone 10, which by
treatment with tributyltin hydride gave the bicyclic cyclopentane-
lactone 11 in quantitative yield. Treatment of the diol 11 with HBr–
HOAc introduced bromine in a regio- and stereoselective reaction
to 12, which by treatment with 1,8-diazabicyclo[5.4.0]undec-5-
ene (DBU) gave the unsaturated compound 13. Finally, acidic
Scheme 2 Overman rearrangement. Reagents and conditions: a) Cl3CCN,
NaH, THF, 0 ◦C, 1 h; b) xylene, reflux, 2 h (75%, 2 steps).
Introduction of hydroxyl groups at C-7 and C-8 might be
performed by either dihydroxylation or epoxidation of the double
bond in 3, leading to mimics of different sugars.
An OsO4-catalysed dihydroxylation was expected to be influ-
enced with respect to the exo–endo selectivity by the conformation
of the bicyclic system. Actually, it turned out that the allylic amide
3 gave only one product when subjected to the OsO4-catalysed
dihydroxylation using NMO·H2O as a re-oxidant. The product
was directly transformed into the acetonide 4 for purification.
Deprotection under acidic conditions gave the amino-hydroxy-
substituted lactone 5 which by reduction using Ca(BH4)2 in THF
and subsequent acetylation gave 6. Deprotection by acidic hydrol-
ysis gave the aminocyclopentanol 7 (Scheme 3). The configurations
were proven by an NOE-experiment on 6, which showed a strong
NOE from H-3 to H-4 and a medium NOE from H-2 to H-4,
indicating that the cis-dihydroxylation had occurred from the exo-
side of the bicyclic system.
The allylic amide 3 was also subjected to epoxidation using m-
CPBA to give exclusively the exo-epoxide 8 in a good yield. The
stereochemistry was confirmed by a NOESY experiment.
Not surprisingly, hydrolysis of the epoxide ring in 8 required
rather harsh conditions due to the adjacent electron-withdrawing
groups.14 Attempts to open the epoxide with aqueous TFA at room
temperature resulted in no conversion even after several days of
reaction. In contrast, heating with aqueous HCl resulted in slow
opening of the epoxide together with hydrolysis of the trichlo-
racetamide. NMR analysis of the crude mixture showed that
several products had formed, and since they were inseparable by
chromatography this strategy was not pursued further (Scheme 4).
Scheme 5 Synthesis of ent-1. Reagents and conditions: a) HBr/AcOH, rt,
20 min, then MeOH, rt, overnight, 81%; b) Bu3SnH, AIBN, EtOAc, rt,
4 h, quant.; c) HBr/AcOH, rt, 2 h, then Ac2O, rt, 2 h, 88%; d) DBU, THF,
reflux, overnight, 92%; e) HCl/MeOH, rt, 2 d, 94%.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3164–3171 | 3165
©

View Article Online
deacetylation provided the target ent-1 in a convenient high-
yielding reaction sequence from 9.
For the synthesis of the aminocyclopentanol ent-6, a reaction se-
quence similar to the one described from 1 was performed starting
with ent-1. The Overman rearrangement of the trichloroimidate
was followed by dihydroxylation and protection to give ent-4, and
finally reduction to give the aminocyclopentanol ent-7 (Scheme 6).
Scheme 8 Retrosynthesis to the N-acetylglucosamine mimic.
groups except for the carbon side chain correspond to those of
N-acetylglucosamine.
Following the retrosynthesis in Scheme 8, epoxide 8 was treated
with NaN3 in DMF at elevated temperatures. As was apparent
from the hydrolysis of 8, the epoxide is highly stabilised by
the adjacent electron-withdrawing groups. Meanwhile, azide is
a far better nucleophile compared to water, and the opening
gave primarily one azide after a short reaction time at elevated
temperature. The major product 14 could be isolated in a moderate
yield by flash chromatography (Scheme 9). The minor isomer
was not isolated, as this could not be separated from other
impurities. The regioselectivity of the nucleophilic attack could
not be determined at this stage. but the major product was later
shown by 2D-NMR analysis of 18 to arise from nucleophilic attack
at C-7.
Scheme 6 Synthesis of aminocyclopentanol ent-7. Reagents and condi-
tions: a) Cl3CCN, NaH, THF, 0 ◦C, 2 h; b) xylene, reflux, 3 h (43%, 2
steps). c) OsO4, NMO, CH2Cl2, rt, overnight; d) 2,2-dimethoxypropane,
acetone, conc. H2SO4, rt, 2 h (72%); e) aq. HCl (6 M), reflux, 2.5 h (79%);
f) Ca(BH4)2, ethanol, −20 ◦C, overnight; g) Ac2O, pyridine, rt, overnight
(42%); h) aq. HCl (4 M), 60 ◦C, 4.5 h (quant.).
Furthermore, ent-3 was transformed into the epoxide ent-8
as the only product following the procedure for preparation
of 8. The aim was to open ent-8 regioselectively to give an
aminocyclopentanol, IV, which would mimic a “a-D-galacto”-
configured carbocation (Scheme 7). An “a-D-galacto”-configured
cyclopentanol was considered as the missing link in a review of the
configuration of the amino group as an anomer-selective mimic.17
We have now published the synthesis of IV12 together with the
corresponding aminocyclopentanol having a one-carbon shorter
side chain directly mimicking the parent sugar.8,12
Scheme 9 Synthesis of the potential b-D-N-acetylglucosaminidase in-
hibitor 19. Reagents and conditions: a) NaN3, NH4Cl, DMF, 80 ◦C, 1.5 h
(65%); b) CaCl2, NaBH4, THF, −20 ◦C, overnight; c) Boc2O, NaHCO3,
acetone, H2O, rt, overnight (40%); d) H2, 5% Pd/C, ethanol, rt, 2.5 h; e)
H2O–MeOH (1 : 2), Ac2O, rt, 1 h (80%); f) aq. HCl (4 M), rt, 15 min
(quant.).
Compound 14 was reduced with Ca(BH4)2 in THF, which
resulted in reduction of both the lactone and the trichloroacetyl
group but left the azide group untouched to give 15. The free
amine could be selectively Boc-protected under basic conditions
to give 16 in a moderate yield. Reduction of the azide and selective
acetylation of the free amine in 17 gave 18 in a good yield. The
Boc-group was finally removed by acidic hydrolysis to give the
target aminocyclopentanol 19 in a quantitative yield (Scheme 9).
Scheme 7 Epoxidation of ent-3. Reagents and conditions: a) m-CPBA,
CH2Cl2, reflux, 4 d (98%).
Biological tests
Unfortunately, ent-8 (like 8) was very sluggish in acid-catalysed
epoxide opening. Since the epoxide 8 was readily available, the
opening of the epoxide with a more efficient nucleophile than
water (as described above) was explored. Furthermore, if an
amino function could be introduced regioselectively, as shown
in Scheme 8, a possible inhibitor for N-acetylglucosaminidase
could be synthesized. The position and stereochemistry of the
The aminocyclopentanols 7 and ent-7, together with the cy-
clopentane lactones 5 and ent-5, were assayed for inhibition of a
range of commercial enzymes: a-D-glucosidase (bakers’ yeast), b-
D-glucosidase (almonds), a-D-galactosidase (green coffee beans),
b-D-galatosidase (E. coli), b-D-galactosidase (bovine liver), a-D-
mannosidase (jack beans) and a-D-mannosidase (almonds) using
the procedure described previously.12
3166 | Org. Biomol. Chem., 2007, 5, 3164–3171
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
The aminocyclopentanol 7 showed potent inhibition of a-
D-mannosidase from jack beans (Ki = 0.3 lM). This activity
is almost identical to the inhibition data obtained for the
aminocyclopentanol 20 synthesized by Ja¨ger and co-workers.18
This is an interesting result since 20 has the correct a-D-manno
configuration (like DMC, I), while 7 has the amino group in
the b-configuration and an extended side chain corresponding
to an L-sugar. The non-importance of the anomeric configuration
indicates that this aminocyclopentanol mimics the oxocarbenium
ion rather than the protonated substrate. For DMC I, having
the a-configuration, it was suggested that mimicking of both
the “flap-up” mannopyranosyl cation as well as the protonated
substrate might occur (see Introduction). Thus no anomer-
selective inhibition from manno-configured inhibitors has been
found, which also has been observed by Ja¨ger and co-workers for
the two anomers 20 and 2118 (Fig. 3). The motif for inhibition of
a mannosidase is the cis-relationship of the two hydroxyl groups
at C-2 and C-3 (see Fig. 1) corresponding to the 2- and 3-hydroxyl
groups in the substrate, as has also been found by synthesis of a
range of mannostatin analogues.19
Fig.
4
Ki values for inhibition of b-N-acetyl-D-gluco-aminocyclo-
pentanols towards b-D-N-acetylglucosaminidase from jack beans.
and amino groups were in accordance with the corresponding
configurations in the respective substrates for the enzymes. The
a-galactosidase could not tolerate the wrong configuration at “C-
3” in ent-7, and the b-glucosidase was not inhibited by 7 with the
wrong stereochemistry at “C-2”.
Experimental
1H NMR spectra were recorded on a Bruker AM 500 instrument
and 13C NMR spectra on Varian Mercury 300 instrument. Chemi-
cal shifts were measured in d (ppm) and coupling constants J in Hz.
For NMR spectra in deuterated solvents, the solvent peak was used
as the reference (CDCl3: d = 7.26 for H, 76.93 for 13C; MeOH-
1
d4: d = 3.31 for H, 49.00 for 13C). When necessary, NMR data
1
were assigned using H–H- and C–H-correlated spectra. Melting
points are uncorrected. Specific rotations were measured on a
Perkin-Elmer 241 polarimeter. Elemental analyses were performed
by the Institute of Physical Chemistry, Vienna. HR-MS was
performed by BioCentrum, DTU. TLC was performed on Merck
60 F254 precoated silica plates, and spots were generally detected
by spraying with a solution of 1.5% NH4Mo2O2, 1% Ce(SO4)2
and 10% H2SO4, followed by charring. Flash chromatography was
performed with silica gel 60 (Merck, 40–63 lm). Concentrations
were performed on a rotary evaporator at a temperature below
40 ◦C. All solvents were distilled before use. Celite refers to Filter
Aid from Celite Corporation.
Fig. 3 Ki values for inhibition of manno-configured aminocyclopentanols
towards a-D-mannosidase from jack beans.
Enzymatic assays
The corresponding lactones 5 and ent-5 did not show any
activity towards the a-D-mannosidases or any of the other enzymes
assayed.
Inhibitory activities of the synthesized compounds were deter-
mined on a Labsystem iEMS reader MF following the procedure
described previously.12 The following commercial enzymes were
tested: a-D-glucosidase (bakers’ yeast), b-D-glucosidase (almonds),
a-D-galactosidase (green coffee beans), b-D-galatosidase (E. coli),
b-D-galactosidase (bovine liver), a-D-mannosidase (jack beans)
and a-D-mannosidase (almonds).
The aminocyclopentanol 19 was tested towards b-D-N-
acetylglucosaminidase from jack beans and showed inhibitory
activity with Ki 8.9 lM. This confirmed our expectations, as
only the carbon side chain is of opposite stereochemistry to the
correct b-N-acetyl-D-gluco configuration. The activity was about
ten-fold less than for the aminocyclopentanol 22,18 which has the
correct configuration. Thus, for this enzyme the correct identity
and configuration of the carbon side chain seems to be crucial for
optimal activity (Fig. 4).
(1S,5R,6S)-6-Trichloroacetamido-2-oxabicyclo[3.3.0]oct-
7-en-3-one (3)
The allylic alcohol 111 (4.02 g, 28.7 mmol) was dissolved in
dry THF (45 mL) and added to NaH (1.56 g, 55–65% in
mineral oil), which prior to the addition was washed with dry
THF and subsequently suspended in dry THF (10 mL). The
reaction mixture was stirred under nitrogen at rt for 30 min, after
which the slurry was added via syringe to trichloroacetonitrile
(4 mL, 40 mmol) over 15 min at 0 ◦C. The mixture was stirred
at this temperature for an additional 1 h, warmed to rt and
filtered though a layer of MgSO4. The solvents were removed by
Conclusion
The unsaturated bicyclic lactones 1 and ent-1 were shown to be
valuable starting materials for synthesising densely functionalised
aminocyclopentanols where the key step was an Overman rear-
rangement. The possible glycosidase inhibitors 7, ent-7 and 19
were synthesised. Both 7 and 19 were identified as good glycosidase
inhibitors. In both compounds the configuration of the hydroxy
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3164–3171 | 3167
©

View Article Online
evaporation in vacuo to give (1R,5R,8R)-8-O-trichloroacetimido-
2-oxabicyclo[3.3.0]oct-6-en-3-one (2) as slightly yellow crystals
(quant.). The crude product was used immediately without further
purification.
in MeOH–acetonitrile to give the title◦compound as colourless
crystals (209 mg, 79%); mp 220–221 C (decomposition); [a]D20
−15.9 (c 1, H2O); Found; C, 39.81; H, 5.66; N, 6.68; Cl, 16.60.
1
Calc. for C7H12O4NCl; C, 40.11; H, 5.77; N, 6.68; Cl, 16.91; H
The imidate 2 (8.16 g, 28.7 mmol) was dissolved in xylene
(150 mL) and heated to reflux for 2 h. The solvent was removed
by evaporation in vacuo to give a brown oil. The residue was
dissolved in EtOAc (250 mL), treated with activated charcoal,
filtered through Celite and evaporated in vacuo to give a yellow
oil (7.47 g, 91%). Purification by flash chromatography (EtOAc–
hexane, 1 : 1) gave the title compound 3 as colourless crystals
NMR (500 MHz, D2O): dH 5.05 (1H, dd, J = 4.4, 9.1, H-1), 4.40
(1H, dd, J = 3.5, 4.5, H-7), 4.36 (1H, dd, J = 3.5, 4.5, H-8),
3.78 (1H, dd, J = 4.5, 6.5, H-6), 3.38 (1H, m, H-5), 3.14 (1H, dd,
J = 11.1, 19, H-4), 2.76 (1H, dd, J = 4.1, 19.0, H-4ꢀ); 13C NMR
(125 MHz, D2O): dC 180.4 (CO), 88.6 (C-1), 77.0, 71.9 (C-7, C-8),
57.7 (C-6), 39.6, 33.6 (C-4, C-5).
◦
(6.11 g, 75%), mp 100–102 C. Recrystallisation of an analytical
(1R,2S,3R,4R,5R)-4-Acetamido-1,2,3-tri-O-acetyl-5-
(2-acetoxyethyl)cyclopentane-1,2,3-triol (6)
◦
sample (EtOAc–hexane) gave mp 105–106 C. [a]2D0+ 182.9 (c
1.6, CHCl3); found; C, 38.13; H, 2.84; N, 4.80; Cl, 37.34. Calc.
for C9H8O3NCl3; C, 37.99; H, 2.83; N, 4.92; Cl, 37.38;1H NMR
(500 MHz, CDCl3): dH 2.63 (1H, m, H-4ꢀ), 2.94–3.05 (2H, m, H-4,
H-5), 4.76 (1H, m, H-6), 5.66 (1H, m, H-1), 6.11 (1H, ddd, J =
1.0, 2.2, 5.5, H-7), 6.25 (1H, ddd, J = 2.2, 2.2, 6.1, H-8), 6.96 (1H,
d, J = 6.1, NH); 13C NMR (125 MHz, CDCl3): dC 34.0 (C-4), 43.6
(C-5), 64.2 (C-6), 87.3 (C-1), 92.0 (CCl3CO) 134.5, 135.3 (C-7,
C-8), 161.8 (Cl3CCO), 175.6 (C-3).
CaCl2 (551 mg, 5.0 mmol) and NaBH4 (405 mg, 10.7 mmol)
were suspended in dry ethanol (3 mL) and stirred at −20 C for
◦
2 h under N2 to ensure formation of Ca(BH4)2. The lactone 5
(210 mg, 1.0 mmol) was dissolved in EtOH (2 mL) and added at
−20 ◦C under N2, and the mixture was stirred at rt overnight. The
reaction was then quenched with aq. HCl (4 M, 8 mL), stirred for
30 min and evaporated in vacuo followed by co-evaporation with
HCl/MeOH (0.1 M). The residue was dissolved in H2O (10 mL)
and loaded onto a column of ion-exchange resin (Amberlite IR-
120, H+, 150 mL). The column was eluted with H2O (250 mL)
to neutral pH and then with 12.5% NH3 (250 mL). The alkaline
phases were concentrated to give a crude residue, which was co-
evaporated three times with toluene. The residue was dissolved
in dry pyridine (4 mL) and Ac2O (1.5 mL) and stirred at rt
overnight. The mixture was evaporated in vacuo and purified
by flash chromatography (EtOAc) to give the title compound
as slightly yellow crystals (267 mg, 68%). The compound was
then recrystallised (EtOAc–hexane) to give the title compound as
colourless crystals (235 mg, 61%); mp 110–111 ◦C; [a]D20 +28.04 (c
1, EtOAc); found; C, 52.48; H, 6.34; N, 3.47. Calc. for C17H25O9N;
C, 52.70; H, 6.50; N, 3.62; 1H NMR (500 MHz, CDCl3): dH 1.73
(1H, m, CH2CH2OAc), 1.86 (1H, m, CH2CH2OAc), 2.01, 2.03,
2.04, 2.11, 2.12 (15 H, 5 × CH3), 2.30 (1H, m, H-5), 4.02 (1H, m,
CH2CH2OAc), 4.12 (1H, m, CH2CH2OAc), 4.52 (1H, ddd, J =
5.1, 9.8, 10.7, H-4), 5.20 (1H, dd, J = 4.3, 4.7, H-2), 5.29 (1H,
dd, J = 4.3, 8.1, H-1), 5.37 (1H, dd, J = 4.7, 5.1, H-3), 5.63 (1H,
d, J = 8.9, NH); 13C NMR (75 MHz, CDCl3): dC 171.0, 169.9,
169.6, 169.4, 169.1 (5 × CO), 75.9 (C-2), 74.5 (C-1), 71.9 (C-3),
62.5 (CH2CH2OAc), 52.4 (C-4), 41.8 (C-5), 26.1 (CH2CH2OAc),
23.2 (CH3CONH), 20.9, 20.7, 20.6, 20.4 (4 × CH3COO).
(1R,5R,6R,7R,8S)-6-Trichloroacetamido-7,8-
isopropylidenedioxy-2-oxabicyclo[3.3.0]oct-3-one (4)
The allylic amide 3 (19.8 g, 69 mmol) was dissolved in CH2Cl2
(200 mL), and NMO·H2O (11.4 g, 84 mmol) was added together
with a catalytic amount of OsO4(s). The mixture was stirred
under N2 at rt overnight. Na2SO3(s) was added and the mixture
stirred for 30 min followed by evaporation in vacuo and co-
evaporation four times with toluene. The crude residue was
dissolved in acetone (200 mL), 2,2-dimethoxypropane (100 mL)
and conc. H2SO4 (4 mL). The mixture was stirred for 20 min
at rt. NaHCO3 was then added until neutral pH, after which
the mixture was filtered and concentrated in vacuo. The residue
was dissolved in CH2Cl2, washed with H2O (100 mL) and brine
(100 mL), dried (MgSO4), filtered and evaporated in vacuo to give
colourless crystals (quant). The compound was purified by flash
chromatography (EtOAc–hexane, 1 : 3) to give the title compound
as colourless crystals (19.95 g, 80%); mp 132–133 ◦C. An analytical
sample was crystallised from EtOAc–hexane; mp 161–162 ◦C; [a]D20
−12.6 (c 1, CHCl3); found; C, 39.94; H, 3.75; N, 4.20; Cl, 29.47.
Calc. for C12H14O5NCl3; C, 40.19; H, 3.93; N, 3.91; Cl, 29.66; 1H
NMR (500 MHz, CDCl3): dH 7.27 (1H, d, J = 9.0, NH), 4.81 (1H,
d, J = 5.4, H-8), 4.77 (1H, dd, J = 5.3, 5.4, H-7), 4.66 (1H, d,
J = 5.4, H-1), 4.20 (1H, ddd, J = 5.2, 8.0, 9.0, H-6), 3.00 (1H, d,
J = 18.5, H-4), 2.94 (1H, ddd, J = 5.5, 8.2, 8.2, H-5), 2.77 (1H,
(1R,2S,3R,4R,5R)-4-Amino-5-(2-hydroxyethyl)-
cyclopentane-1,2,3-triol hydrochloride (7)
dd, J = 8.2, 18.2, H-4ꢀ); 1.50 (3H, s, CH3), 1,36 (3H, s, CH3); 13
C
NMR (125 MHz, CDCl3): dC 174.6 (CO), 162.1 (COCCl3), 112.3
((CH3)2C), 86.5 (C-1), 82.1, 80.2 (C-7, C-8), 58.3 (C-6), 44.8 (C-5),
33.7 (C-4), 26.3, 24.2 (2 × CH3).
Compound 6 (73 mg, 0.19 mmol) was dissolved in aq. HCl (4 M,
5 mL), the mixture was heated to 60 ◦C for 4.5 h, and then
evaporated in vacuo and co-evaporated several times with toluene
to give the title compound as a slightly yellow oil (40 mg, quant.);
[a]2D0 +37.94 (c 1, MeOH); HR-EI-MS; C7H15O4N·HCl, Calc. for
(1R,5R,6R,7R,8R)-6-Amino-7,8-dihydroxy-2-
oxabicyclo[3.3.0]oct-3-one hydrochloride (5)
1
[M − 1]; m/z 212.0690. Found; 212.0682; H NMR (500 MHz,
MeOD): dH 4.34 (1H, dd, J = 6.5, 4.5, H-3), 4.03 (1H, dd, J =
2.5, 6.3, H-1), 3.94 (1H, dd, J = 2.5, 4.5, H-2), 3.79 (1H, ddd,
J = 5.5, 5.5, 10.5, CH2CH2OH), 3.63 (1H, ddd, J = 4.3, 10.1,
10.1, CH2CH2OH), 3.30 (1H, dd, J = 8.5, 8.5, H-4), 2.32 (1H,
m, H-5), 1.89 (1H, m, CH2CH2OH), 1.76 (1H, m, CH2CH2OH);
Compound 4 (455 mg, 1.26 mmol) was suspended in aq. HCl (6 M,
10 mL) and heated to reflux for 2.5 h. The mixture was cooled to
rt, evaporated in vacuo and co-evaporated 4 times with toluene
to give 5 as colourless crystals (quant.). These were recrystallised
3168 | Org. Biomol. Chem., 2007, 5, 3164–3171
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
13C NMR (75 MHz, MeOD): dC 78.6 (C-2), 76.6 (C-1), 70.6 (C-3),
61.7 (CH2CH2OH), 57.5 (C-4), 45.0 (C-5), 31.4 (CH2CH2OH).
was dissolved in acetonitrile (30.0 mL) and extracted five times
with hexane. The acetonitrile phase was then concentrated to
form a crystalline crude product, which was further purified by
flash chromatography (EtOAc) to give 11 as colourless crystals
(1.33 g, quant.); mp 76–77 ◦C; [a]2D0 +70.6 (c 1.0, MeOH); found;
(1R,5R,6R,7R,8S)-6-Trichloroacetamido-7,8-epoxy-2-
oxabicyclo[3.3.0]oct-3-one (8)
1
C, 52.94 H, 6.23. Calculated for C7H10O4; C, 53.16; H, 6.37; H
The allylic amide 3 (2.01 g, 7.1 mmol) was dissolved in CH2Cl2
(50 mL), and m-CPBA (50–90% in H2O, 4.824 g, 14 mmol) was
added. The mixture was heated to reflux for 4 d, after which
CH2Cl2 (30 mL) was added and the organic phase washed with aq.
Na2SO3 (10%, 40 mL), half-saturated aq. NaHCO3 (2 × 30 mL),
H2O (25 mL) and brine (25 mL). The organic phase was dried
(MgSO4), filtered and evaporated in vacuo to give 8 as colourless
crystals; mp 152–162 ◦C (520 mg, 98%). These were recrystallised
from EtOAc–hexane to give the title compound as white crystals
NMR (500 MHz, D2O) d 4.96 (1H, ddd, J = 8.4, 3.4, 1.8, H-1),
4.31 (1H, ddd, J = 9.3, 5.1, 1.4, H-7), 3.29 (1H, m, H-5), 4.23 (1H,
dd, J = 7.3, 3.9, H-8), 3.02 (1H, ddd, J = 19.4, 11.0, 2.0, H-4ꢀ),
2.52 (1H, ddd, J = 19.5, 4.1, 1.9, H-4), 2.22 (1H, m, H-6), 1.87
(1H, ddd, J = 14.3, 7.4, 2.2, H-6ꢀ); 13C NMR (125 MHz, D2O)
d 182.5 (C-3), 90.4 (C-1), 77.9 (C-8), 72.9 (C-7), 37.1, 35.8 (C-4,
C-6), 34.3 (C-5).
◦
(1S,5S,7R,8R)-8-Acetoxy-7-bromo-7-deoxy-2-
oxabicyclo[3.3.0]oct-3-one 12
(1.45 g, 68%); mp 161–163 C; [a]2D0 +38.74 (c 1, EtOAc); found;
C, 35.88; H, 2.67; N, 4.55; Cl, 35.27. Calc. for C9H8O4NCl3; C,
1
35.97; H, 2.68; N, 4.66; Cl, 35.39; H NMR (500 MHz, CDCl3):
Compound 11 (729 mg, 4.6 mmol) was suspended in HBr/AcOH
(33.6%, 25.0 mL). The system was closed with a glass stopper
and stirred at room temperature for 2 h, followed by addition of
acetic anhydride (50.0 mL), and stirring for an additional 2 h.
The reaction mixture was concentrated, and then co-concentrated
with MeOH to give a syrupy residue, which was dissolved in
dichloromethane (15.0 mL), washed with saturated NaCl (aq)
(10.0 mL) and concentrated in vacuo. The crude product was
purified further by flash chromatography (EtOAc–heptane, 1 : 1)
giving 12 as fine yellow crystals (1.07 g, 88%); mp 71.5–73◦◦C; [a]D20
+21.5 (c 1.0, CHCl3); [reported for ent-12:10 mp 72–73 C; [a]D20
−24.6 (c 1.0, CHCl3)]; ES-MS; C9H11BrO4Na, Calc. for [M + Na];
dH 2.51 (1H, m, H-5), 2.88 (1H, d, J = 2.5, H-4), 2.87 (1H, s, H-4ꢀ),
3.80 (1H, dd, J = 1.7, 2.5, H-7), 3.90 (1H, d, J = 2.5, H-8), 4.33
(1H, ddd, J = 1.7, 5.5, 7.3, H-6), 5.03 (1H, d, J = 6.4, H-1), 7.05
(1H, d, J = 7.3, NH); 13C NMR (75 MHz, CDCl3): dC 33.5 (C-
4), 42.8 (C-5), 56.8, 59.0 (C-7, C-8), 59.3 (C-6), 81.5 (C-1), 162.4
(Cl3CCO), 174.8 (C-3).
7-Bromo-2,3,7-trideoxy-L-arabino-hept-2-enono-1,4-lactone (10)
6,7-O-Isopropylidene-2,3-dideoxy-L-arabino-hept-2-enono-1,4-
lactone 9 (2.90 g, 13.5 mmol) was suspended in HBr/AcOH
(33.4%, 25 mL). The system was closed with a glass stopper and
the reaction mixture was stirred vigorously at room temperature
for 20 min, followed by addition of MeOH (50 mL), and stirring
was continued at room temperature overnight. The mixture was
concentrated, and then co-concentrated 5 times with MeOH,
3 times with H2O and 3 times with toluene to give a brown
syrupy residue. This was dissolved in H2O (50 mL), extracted
3 times with CH2Cl2 (50 mL) which was re-extracted 3 times
with H2O. The combined aqueous phases were extracted 6 times
with EtOAc (50 mL) and the combined EtOAc phases were
then dried (MgSO4) and concentrated to give yellow crystals.
The combined aqueous phases from the above procedure were
continuously extracted with EtOAc overnight. The organic phase
was concentrated and purified by flash chromatography using
EtOAc as eluent to give colourless crystals (total 2.60 g, 81%); mp
1
m/z 283. Found; 285; H NMR (300 MHz, CDCl3) d 5.37 (1H,
dd, J = 3.9, 0.6, H-8), 4.76 (1H, d, J = 8.1, H-1), 4.10 (1H, ddd,
J = 5.1, 4.5, 0.9, H-7), 3.02–3.18 (1H, m, H-5), 2.81 (1H, dd, J =
18.6, 10.8, H-4ꢀ), 2.71 (1H, ddd, J = 14.7, 6.3, 2.1, H-6), 2.51 (1H,
dd, J = 18.9, 3.6, H-4), 2.09 (1H, ddd, J = 14.7, 5.7, 5.7, H-6),
2.05 (3H, s, CH3); 13C NMR (75.0 MHz, CDCl3) d 176.0 (C-3),
=
169.3 (CH3–C O), 87.0 (C-1), 83.5 (C-8), 46.6 (C-7), 41.0 (C-6),
36.6 (C-4), 35.5 (C-5), 20.9 (CH3). All NMR data are identical to
those reported for the enantiomer.10
(1S,5S,8S)-8-Acetoxy-2-oxabicyclo[3.3.0]oct-6-ene-3-one 13
Compound 12 (1.26 g, 4.8 mmol) was dissolved in dry THF
(40.0 mL), and DBU (1.40 mL, 9.4 mmol) was added. The
reaction mixture was heated to reflux overnight under N2. The
reaction mixture was then cooled to room temperature, filtered
and concentrated. The residue was dissolved in CH2Cl2 (10.0 mL),
washed with H2O (15.0 mL) and HCl (1 M, 15.0 mL), and the
combined aqueous phases were re-extracted twice with CH2Cl2
(15.0 mL). The combined organic phases were washed once with
saturated aq. NaCl (15.0 mL), dried (MgSO4) and concentrated
to give a yellow syrup. The crude product was purified by
flash chromatography (EtOAc–heptane, 1 : 2), to give the title
compound as a transparent oil (0.80 g, 92%) (99%, based on
recovered starting material). Crystallization from ether gave fine
◦
102–104 C; [a]2D0 −83.9 (c 1, MeOH); found; C, 35.83; H, 3.78;
Br, 32.86. Calculated for C7H9O4Br; C, 35.47; H, 3.83; Br, 33.71;
1H NMR (300 MHz, D2O) d 7.84 (1H, dd, J = 5.5, 1.3, H-3),
6.36 (1H, dd, J = 6.0, 2.1, H-2), 5.67 (1H, dd, J = 3.8, 1.7, H-4),
4.08 (1H, dd, J = 9.0, 1.7, H-5), 4.03 (1H, ddd, J = 8.9, 4.7, 2.6,
H-6), 3.87 (1H, dd, J = 11.1, 2.5, H-7), 3.80 (1H, dd, J = 11.5,
4.3, H-7ꢀ); 13C NMR (75 MHz, D2O) d 176.8 (C-1), 157.4 (C-3),
121.7 (C-2), 84.2 (C-4), 70.2, 69.7 (C-5, C-6), 37.4 (C-7).
(1S,5R,7S,8S)-7,8-Dihydroxy-2-oxabicyclo[3.3.0]oct-3-one (11)
◦
Compound 10 (2.01 g, 8.4 mmol) was dissolved in dry EtOAc
(25.0 mL) and heated to reflux under N2. A solution of AIBN
(145 mg, 0.9 mmol) and Bu3SnH (2.5 mL, 9.4 mmol) in dry EtOAc
was added slowly over 2 h. The solution was kept under reflux
for an additional 2 h. Concentration left an oily residue which
colourless crystals; mp 44.5–46 C; [a]2D0 +283.3 (c 1.0, CHCl3);
[reported for ent-13:11 mp 49–50 ◦C; [a]D20 −285.0 (c 1.0, CHCl3)];
ES-MS; C9H10O4Na, Calc. for [M + Na]; m/z 205. Found; 205; 1H
NMR (300 MHz, CDCl3) d 5.97 (1H, dd, J = 1.3, 5.7, H-8), 5.91
(1H, dt, J = 2.2, 4.9 H-6), 5.62 (1H, ddd, J = 0.6, 1.4, 2.8, H-7),
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3164–3171 | 3169
©

View Article Online
above to give the product (44 mg, 42%); [a]2D5 −28.0 (c 1, EtOAc).
4.90 (1H, ddd, J = 0.6, 1.3, 5.9, H-1), 3.70 (1H, m, H-5), 2.78 (1H,
dd, J = 10.3, 18.3, H-4ꢀ), 2.40 (1H, dd, J = 2.3, 18.3, H-4), 2.04
(3H, s, CH3); 13C NMR (75.0 MHz, CDCl3) d 175.4 (C-3), 170.1
1
The H and 13C NMR spectra were identical to those described
above for 6.
=
(CH3–C O), 139.0 (C-6), 129.0 (C-7), 85.5 (C-1), 82.1 (C-8), 43.8
(C-4), 32.3 (C-5), 21.0 (CH3). All NMR data are indetical to those
(1S,2R,3S,4S,5S)-4-Amino-5-(2-hydroxyethyl)cyclopentane-
1,2,3-triol hydrochloride (ent-7)
reported for the enantiomer.11
(1S,5S,8S)-8-Hydroxy-2-oxabicyclo[3.3.0]oct-6-en-3-one (ent-1)
The acetylated compound ent-6 (32.6 mg, 0.08 mmol) was
dissolved in aq. HCl (4 M, 15.0 mL) heated to 60 ◦C for 4.5 h.
Concentration in vacuo and co-evaporation 3 times with toluene
gave the title compound as an oil (17.0 mg, quant); [a]2D5 −43.3
Compound 13 (280 mg, 1.5 mmol) was dissolved in a solution
of dry 1% HCl/MeOH (15.0 mL), and the solution was stirred
at room temperature for 2 days. The reaction mixture was
concentrated, co-concentrated with toluene, and the residue was
purified further by flash chromatography (EtOAc), to give the
product as an oil (204 mg, 94%), which wa◦s crystallized and
recrystallised from hexane–EtOAc; mp 60–66 C; [a]2D0 +169.4 (c
1
(c 0.5, MeOH); The H and 13C NMR spectra were identical to
those described above for 7.
(1S,5S,6S,7S,8R)-6-Trichloroacetamino-7,8-epoxy-
2-oxabicyclo[3.3.0]oct-3-one (ent-8)
◦
1.0, CHCl3); [reported for 1:11 mp 67–68 C; [a]D20 +170.9 (c 1.0,
CHCl3)]; ES-MS; C7H8O3Na, Calc. for [M + Na]; m/z 163. Found;
163; 1H NMR (300.0 MHz, CDCl3) d 5.93 (1H, m, H-8), 5.88 (1H,
dd, J = 1.1, 5.7, H-6 or H-6ꢀ, 4.83 (2H, m, H-1, H-7), 3.73 (1H,
m, H-5), 3.47 (1H, d, J = 0.6, OH), 2.78 (1H, dd, J = 10.1, 18.2,
H-4ꢀ), 2.40 (1H, dd, J = 1.8, 18.2, H-4). 13C NMR (75.0 MHz,
CDCl3) d 176.2 (C-3), 137.0, 132.3 (C-7, C-6), 88.4 (C-1), 80.3
(C-8), 43.6 (C-4), 32.6 (C-5). All NMR data are identical to those
reported for the enantiomer.11
The unsaturated compound ent-3 (37 mg, 0.14 mmol) was
epoxidised with m-CPBA (93 mg, 0.26 mmol, 50–90% in H2O)
as described above. Purification of the crude product by column
chromatography (heptane–EtOAc, 1 : 1) gave ent-8 (36 mg, 92%);
1
[a]2D2 − 40.9 (c 0.9, EtOAc); The H and 13C NMR spectra were
identical to those described above for 8.
(1R,5R,6R,7S,8R)-6-Trichloroacetamido-7-azido-8-hydroxy-
2-oxabicyclo[3.3.0]oct-3-one (14)
(1R,5S,6R)-6-Trichloroacetamino-2-oxabicyclo[3.3.0]oct-
7-ene-3-one (ent-3)
The epoxide 8 (297 mg, 0.99 mmol), NaN3 (697 mg, 10.7 mmol)
and NH4Cl (650 mg, 12.2 mmol) were dissolved in dry DMF
(10 mL) and heated to 80 ◦C for 1.5 h. The mixture was evaporated
in vacuo, dissolved in EtOAc (50 mL) and washed with H2O (mL).
The aqueous phase was re-extracted with EtOAc (2 × 20 mL).
The organic phases were combined, washed with brine (15 mL),
dried (MgSO4), filtered and concentrated in vacuo to give a mixture
of isomeric products as colourless crystals (quant.). The residue
was purified by flash chromatography (EtOAc–hexane, 1 : 1) to
give the title compound as colourless crystals (220 mg, 65%); mp
194–196 ◦C; [a]2D0 −23.6 (c 1.1, acetone); found; H, 2.64; C, 31.68;
N, 16.09; Cl, 30.86. Calc. for C9H9O4N4Cl3; H, 2.64; C, 31.46; N,
16.38; Cl, 30.96; 1H NMR (500 MHz, acetone-d6): dH 5.41 (1H, d,
J = 5.7, NH), 4.75 (1H, dd, J = 4.4, 9.7, H-1), 4.20–4.09 (3H, m,
H-6, H-7, H-8), 3.20 (1H, m, H-5), 2.84 (1H, dd, J = 18.1, 10.6,
H-4), 2.62 (1H, dd, J = 3.5, 18.2, H-4ꢀ); 13C NMR (75.5 MHz,
acetone-d6): dC 176.3 (CO), 163.0, (COCCl3), 86.4 (C-1), 79.8 (C-
8), 68.7, 59.8 (C-6, C-7), 40.6 (C-5), 33.7 (C-4).
The allylic alcohol ent-1 (125 mg, 0.9 mmol) was transformed to
the trichloroacetamide ent-2 and rearranged immediately to ent-
3 as described for 3 to give a crude product, which was purified
by flash chromatography (EtOAc–heptane 1 : 3 → 1 : 1), giving
the title compound as colourless crystals (109 mg, 43%); mp 101–
103 ◦C; [a]2D0 −166.3 (c 1.0, CHCl3); ES-MS; C9H8Cl3NO3Na, Calc.
for [M + Na]; m/z 306. Found; 306. The 1H and 13C NMR spectra
were in accordance with those for 3.
(1S,5S,6S,7S,8S)-6-Trichloroacetamido-7,8-isopropylidene-
2-oxabicyclo[3.3.0]oct-3-one (ent-4)
The allylic amide ent-3 (136 mg, 0.47 mmol) was dihydroxylated
and isolated as the isopropylidene derivative as described for 4
to give the product as colourless crystals (122 mg, 72%); mp
129–130 ◦C. Recrystallisation from EtOAc–hexane gave mp 159–
◦
161 C; [a]2D0 +11.4 (c 1, CHCl3); The H and 13C NMR spectra
1
were identical to those described above for 4.
(1R,2R,3S,4R,5R)-4-(N-tert-Butoxycarbonyl)amino-3-azido-5-
(2-hydroxyethyl)cyclopentane-1,2-diol (16).
(1S,5S,6S,7S,8S)-6-Trichloroacetamido-7,8-dihydroxy-
2-oxabicyclo[3.3.0]oct-3-one (ent-5)
CaCl2 (940 mg, 8.5 mmol) and NaBH4 (599 mg, 15.8 mmol) were
suspended in dry THF (15 mL) and stirred at rt for 1.5 h. The
mixture was cooled to −20 ◦C, 14 (362 mg, 1.05 mmol) was added,
and the mixture stirred at −20 ◦C overnight. The reaction was
quenched using aq. HCl (4 M, 8 mL) and stirred for 0.5 hour,
concentrated in vacuo and co-evaporated twice with 1% HCl in
MeOH. The residue was dissolved in water (10 mL) and loaded
onto a column of ion-exchange resin (Amberlite IR-120, H+,
170 mL). The column was eluted with water (200 mL) to neutral
pH and then with 12.5% aq. ammonia (300 mL). The alkaline
The isopropylidene derivative ent-4 was deprotected as d◦escribed
above to give ent-5 as colourless crystals; mp 218–221 C; [a]D25
+12.1 (c 1 in MeOH). The 1H and 13C NMR spectra were identical
to those described above for 5.
(1S,2R,3S,4S,5S)-4-Acetamido-1,2,3-tri-O-acetyl-5-
(2-acetoxyethyl)cyclopentane-1,2,3-triol (ent-6)
The bicyclic lactone ent-5 (86 mg, 0.27 mmol) was subjected to
reduction with Ca(BH4)2 followed by acetylation as described
3170 | Org. Biomol. Chem., 2007, 5, 3164–3171
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
phases were concentrated in vacuo and then co-evaporated with
aq. HCl (4 M, 10 mL) to give crude 15 (140 mg, 56%).
61.5 (C-7), 58.9 (C-4), 44.7 (C-5), 30.4 (C-6), 28.7 ((CH3)3C), 22.7
(CH3CO).
The above crude 15 was dissolved in acetone (3 mL) and H2O
(3 mL), and the pH was adjusted to pH 8 with NaHCO3(s).
Boc2O (800 mg, 4.6 mmol) was added and the mixture stirred
at rt overnight. The mixture was evaporated in vacuo, dissolved
in EtOAc (10 mL) and washed with brine (10 mL). The aqueous
phase was re-extracted with EtOAc (4 × 10 mL), the organic
phases were combined, dried (Na2SO4), filtered and evaporated
in vacuo. The crude product was purified by flash chromatography
to give th◦e title compound as colourless crystals (123 mg, 40%); mp
117–119 C.; [a]D20 +75.8 (c 0.7, MeOH); HR-EI-MS; C12H22O5N4,
(1R,2R,3S,4R,5R)-4-Amino-3-acetamido-5-
(2-hydroxyethyl)cyclopentane-1,2-diol hydrochloride (19)
Compound 18 (44 mg, 0.14 mmol) was dissolved in aq. HCl (4 M,
5 mL) and stirred at rt for 15 min. The mixture was concentrated
in vacuo and co-evaporated several times with toluene to give the
title compound as a hygroscopic oil (36 mg, quant); [a]2D0 +60.06
(c 1, MeOH); HR-EI-MS; C9H18O4N2·HCl, Calc for [M − 1] m/z
253.0955. Found 253.0914; 1H NMR (500 MHz, MeOD): dH 3.97–
3.95 (2H, m, H-1, H-3), 3.91 (1H, dd, J = 2.6, 4.5, H-2), 3.82 (1H,
ddd, J = 4.5, 4.5, 10.3, H-7), 3.61 (1H, ddd, J = 3.4, 10.3, 10.3,
H-7ꢀ), 3.24 (1H, dd, J = 8.2, 9.3, H-4), 2.31 (1H, m, H-5), 2.04
(3H, s, CH3CO), 1.93 (1H, m, H-6), 1.77 (1H, m, H-6ꢀ); 13C NMR
(75.4 MHz, MeOD): dC 174.9 (CO), 81.5 (C-2), 78.0 (C-1), 62.6
(C-3), 61.7 (C-4), 61.6 (C-7), 45.3 (C-5), 30.9 (C-6), 22.3 (COCH3).
1
Calc. for [M + Na]; m/z 225.1488; found; 225.1476; H NMR
(500 MHz, MeOD) : d 3.85 (1H, dd, J = 3.2, 6.8, H-1), 3.71 (1H,
dd, J = 9.5, 11.2, H-4), 3.67 (1H, dd, J = 3.1, 6.3, H-2), 3.61
(2H, m, CH2CH2OH), 3.38(1H, dd, J = 6.7, 9.3, H-3), 2.00 (1H,
m, H-5), 1.82 (1H, m, CH2CH2OH), 1.59 (1H, m, CH2CH2OH),
1.45 (3H, s, OC(CH3)3); 13C NMR (75 MHz, MeOD): dC 158.0
(CO), 82.5 (OC(CH3)3), 80.3 (C-2), 76.2 (C-1), 72.9 (C-3), 61.4
(CH2CH2OH), 59.0 (C-4), 44.4 (C-5), 30.2 (CH2CH2OH), 28.8
(OC(CH3)3). IR mmax(film)/cm−1 3316 s br (OH); 2106 s (N3); 1690 s
References
1 A. Berecibar, C. Grandjean and A. Siriwardena, Chem. Rev., 1999, 99,
779.
2 V. H. Lillelund, H. H. Jensen, X. Liang and M. Bols, Chem. Rev., 2002,
102, 515.
3 R. A. Farr, N. P. Peet and M. S. Kang, Tetrahedron Lett., 1990, 31,
7109.
=
(C O).
(1R,2R,3S,4R,5R)-4-(N-tert-Butoxycarbonyl)amino-3-acetamido-
5-(2-hydroxyethyl)cyclopentane-1,2-diol (17)
4 D. A. Winkler and G. Holan, J. Med. Chem., 1989, 32, 2084.
5 D. A. Winkler, J. Med. Chem., 1996, 39, 4332.
6 A. Blaser and J.-L. Reymond, Helv. Chim. Acta, 1999, 82, 760.
7 E. Leroy and J.-L. Reymond, Org. Lett., 1999, 1, 775.
8 L. G. Dickson, E. Leroy and J.-L. Reymond, Org. Biomol. Chem., 2004,
2, 1217.
9 A. M. Horneman and I. Lundt, Tetrahedron, 1997, 53, 6879.
10 S. K. Johansen, H. T. Kornø and I. Lundt, Synthesis, 1999, 171.
11 S. K. Johansen and I. Lundt, J. Chem. Soc., Perkin Trans. 1, 1999,
3615.
12 M. Bøjstrup and I. Lundt, Org. Biomol. Chem., 2005, 3, 1738.
13 L. E. Overman, J. Am. Chem. Soc., 1976, 98, 2901.
14 S. B. King and B. Ganem, J. Am. Chem. Soc., 1994, 116, 562.
15 G. Casiraghi and G. Rassu, Synthesis, 1995, 607.
16 G. Casiraghi, L. Colombo, G. Rassu, P. Spanu, G. G. Fava and M. F.
Belicchi, Tetrahedron, 1990, 46, 5807.
The azide 16 (116 mg, 0.38 mmol) was dissolved in ethanol (7 mL),
Pd/C (5%, 77 mg) was added and the mixture was stirred under
H2 for 2.5 h. The mixture was then filtered trough Celite and
evaporated in vacuo. The residue was dissolved in H2O–MeOH
(1 : 2, 6 mL). Ac2O (0.4 mL) was added and the mixture was
stirred at rt for 1 h. The mixture was concentrated in vacuo and co-
evaporated three times with toluene. The residue was then purified
by flash chromatography (EtOAc–hexane, 9 : 1), to give the title
◦
compound as colourless crystals (98 mg, 80%); mp 213–214 C;
[a]2D0 +70.04 (c 1, MeOH); found; H, 8.03; C, 52.74; N, 8.54. Calc.
for C14H26O6N2; H, 8.23; C, 52.82; N, 8.80; 1H NMR (500 MHz,
MeOD): dH 3.88 (1H, dd, J = 2.3, 6.2, H-1), 3.77 (1H, dd, J =
5.9, 8.7, H-3), 3.71 (1H, dd, J = 2.3, 5.5, H-2), 3.69 (H, dd, J =
8.9, 8.9, H-4), 3.63 (2H, t, J = 6.5, H-7, H-7ꢀ), 2.02 (1H, m, H-5),
1.95 (3H, s, CH3CO), 1.84 (1H, m, H-6), 1.64 (1H, m, H-6ꢀ), 1.42
(9H, s, (CH3)3C); 13C NMR (75.4 MHz, MeOD): dC 173.8 (CO),
158.5 (CO), 82.9 (C-2), 80.1 (OCCH3), 76.6 (C-1), 63.4 (C-3),
17 V. H. Lillelund, H. H. Jensen, X. Liang and M. Bols, Chem. Rev., 2002,
102, 515.
18 M. Kleban, P. Hilgers, J. N. Greul, R. D. Kugler, J. Li, S. Picasso, P.
Vogel and V. Ja¨ger, ChemBioChem, 2001, 2, 365.
19 S. Ogawa and T. Morikawa, Eur. J. Org. Chem., 2005, 4065 and
references therein.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3164–3171 | 3171
©