Stereoselective organocatalytic approach to α,β-disubstituted- β-amino acids: A short enantioselective synthesis of cispentacin

Article abstract of DOI:10.1002/ejoc.201300197

α-Branched α,β-unsaturated aldehydes have been tested in the organocatalytic tandem Michael addition/cyclization with N-(benzyloxycarbonyl)hydroxylamine, a reaction which until now has been restricted to α-unsubstituted enals. Starting from cyclopentene-2- carbaldehyde, and using diphenylprolinol trimethylsilyl ether as a chiral amine catalyst, this approach has led to the development of a practical, high yielding (93-98 % overall yield, three steps), and highly enantioselective (up to 98:2 er) route to the cyclic β-amino acid cispentacin, which compares favourably with previously described asymmetric syntheses of this biologically active natural product. When using acyclic α-branched α,β-unsaturated aldehydes as substrates, the reaction yields depend on the substitution pattern of the aldehydes, and mixtures of cis- and trans-isomers are obtained. Nevertheless, this strategy has proved to be successful in some instances, and (3R,4R)-benzyl 3-ethyl-4-methyl-5-oxoisoxazolidin-2-carboxylate could be obtained in 70 % overall yield (two steps) from the reaction of 2-ethylcrotonaldehyde and N-(benzyloxycarbonyl)hydroxylamine under catalysis with diphenylprolinol trimethylsilyl ether, and with high enantiomeric purity (99:1 er). The organocatalytic tandem Michael addition/cyclization of cyclopentene-2-carbaldehyde with N-Cbz-hydroxylamine, using diphenylprolinol TMS ether as a chiral amine catalyst, gives a practical, high yielding, and highly enantioselective route to the cyclic β-amino acid cispentacin. Acyclic α-branched enals give variable results in the same reaction, depending on the substitution pattern. Copyright

Full text of DOI:10.1002/ejoc.201300197

FULL PAPER
DOI: 10.1002/ejoc.201300197
Stereoselective Organocatalytic Approach to α,β-Disubstituted-β-amino Acids:
A Short Enantioselective Synthesis of Cispentacin
Ana Pou[a] and Albert Moyano*[a]
Keywords: Asymmetric catalysis / Organocatalysis / Aza-Michael addition / Aldehydes / Amino acids
α-Branched α,β-unsaturated aldehydes have been tested in
the organocatalytic tandem Michael addition/cyclization
with N-(benzyloxycarbonyl)hydroxylamine, a reaction which
until now has been restricted to α-unsubstituted enals. Start-
ing from cyclopentene-2-carbaldehyde, and using di-
phenylprolinol trimethylsilyl ether as a chiral amine catalyst,
this approach has led to the development of a practical, high
yielding (93–98% overall yield, three steps), and highly
enantioselective (up to 98:2 er) route to the cyclic β-amino
acid cispentacin, which compares favourably with previously
described asymmetric syntheses of this biologically active
natural product. When using acyclic α-branched α,β-unsatu-
rated aldehydes as substrates, the reaction yields depend on
the substitution pattern of the aldehydes, and mixtures of cis-
and trans-isomers are obtained. Nevertheless, this strategy
has proved to be successful in some instances, and (3R,4R)-
benzyl
3-ethyl-4-methyl-5-oxoisoxazolidin-2-carboxylate
could be obtained in 70% overall yield (two steps) from the
reaction of 2-ethylcrotonaldehyde and N-(benzyloxycarb-
onyl)hydroxylamine under catalysis with diphenylprolinol
trimethylsilyl ether, and with high enantiomeric purity (99:1
er).
Until now, most approaches to the preparation of 1 in
optically active form have relied on resolution of racemic
cispentacin by crystallization of diastereomeric salts,[5,8] or
by enzymatic hydrolysis, either of cispentacin derivatives or
of synthetic precursors.[9] However, Davies[10] and Aggar-
wal[11] have achieved asymmetric syntheses of cispentacin
using chiral auxiliaries, and Bolm and co-workers have de-
veloped a three-step sequence relying on the quinidine-pro-
moted asymmetric desymmetrization of the meso-anhydride
of cis-1,2-cyclopentanedicarboxylic acid, followed by
Curtius degradation of the acyl azide derived from the op-
tically active hemiester.[12]
A retrosynthetic analysis of cispentacin (Scheme 1)
showed that a potentially attractive synthesis of this com-
pound could be based on the stereocontrolled tandem aza-
Michael/cyclization reaction of cyclopentene-1-carbal-
dehyde (2) with N-(benzyloxycarbonyl)hydroxylamine (3),
followed by oxidation of intermediate 5-hydroxyisoxazolid-
ine 4 to isoxazolidinone 5 and then hydrogenolysis. In fact,
the chiral-secondary-amine-catalysed tandem aza-Michael/
cyclization reaction of α-unsubstituted enals with 3, re-
ported by Córdova and co-workers a few years ago, consti-
tutes a highly enantioselective route for the synthesis of 2-
substituted 5-hydroxyisoxazolidines, which can be con-
verted in two steps into the corresponding acyclic β-amino
Introduction
Acyclic and alicyclic β-amino acids are important com-
pounds from both chemical and biological points of view.[1]
Although less abundant than α-amino acids, they are found
in peptides and in other types of natural products; they are
useful precursors of β-lactams;[2] they can be incorporated
into synthetic peptides to give well-defined three-dimen-
sional structures and with improved stability;[3] and some
of them have significant pharmacological properties.[4]
An especially relevant alicyclic β-amino acid is cis-2-
aminocyclopentane-1-carboxylic acid, cispentacin (1). The
(1R,2S) enantiomer of cispentacin was isolated in 1989 as
an antifungal antibiotic from the culture broth of a Bacillus
cereus strain (L450–B2).[5] The same compound was inde-
pendently isolated from the culture broth of Streptomyces
setonii,[6] and was found to have a potent therapeutic effect
against Candida albicans.[5–7] On the other hand, the non-
natural (1S,2R) enantiomer of 1 is devoid of biological ac-
tivity.[5] The interesting pharmacological activity, together
with the relative structural simplicity of cispentacin, has
stimulated the search for practical enantioselective synthe-
ses of this compound.
[a] Departament de Química Orgànica, Universitat de Barcelona, acids.[13]
Facultat de Química,
It is worth noting that until now, the substrate scope of
c. Martí i Franquès 1–11, 08028 Barcelona, Catalonia, Spain
Fax: +34-933397878
E-mail: amoyano@ub.edu
Homepage: http://www.qo.ub.es/personals/amoyano.htm
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201300197.
asymmetric organocatalytic aza-Michael additions to un-
saturated aldehydes has been mainly restricted to simple β-
substituted enals,[13,14] since the use of α-branched α,β-un-
saturated aldehydes in these reactions is plagued by very
Eur. J. Org. Chem. 2013, 3103–3111
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3103

A. Pou, A. Moyano
FULL PAPER
Table 1. Pyrrolidine-promoted aza-Michael/cyclization reaction of
cyclopentene-1-carbaldehyde (2) with N-Cbz-hydroxylamine (3).
Entry Solvent 2/3 ratio Pyrrolidine [mol-%] Time [d][a] Yield [%][b]
1
2
3
4
5
6
7
8
CHCl3
MeOH
EtOH
toluene
CH3CN
CHCl3
CHCl3
toluene
1.0
1.0
1.0
1.0
1.0
2.0
2.0
2.0
100
100
100
100
100
100
20
2
2
2
2
2
1
4
4
47
24
34
42
32
60
56
80
Scheme 1. Retrosynthetic analysis of cispentacin.
low rates and/or stereoselectivities, due to steric hindrance
for the formation of the enamine intermediate.[15] As a re-
sult of these limitations, there are very few examples de-
scribing successful amine-catalysed Michael additions to
cyclopentene-1-carbaldehyde (2).[16,17] In particular, to the
best of our knowledge, there are no reports concerning aza-
Michael additions with this compound. In light of these
precedents, we decided to investigate in some depth the
amine-catalysed Michael additions of N-(benzyloxycarb-
onyl)hydroxylamine (3) with α-branched α,β-unsaturated
aldehydes, and in particular with cyclopentene-1-carbal-
dehyde (2).
20
[a] Time necessary for consumption of 3, as monitored by 1H NMR
spectroscopy. [b] Yield of isolated rac-4 after chromatographic puri-
fication.
As can be seen from the results shown in Table 1, the
best yields were obtained in relatively non-polar solvents
such as chloroform (Table 1, entry 1) and toluene (Table 1,
entry 4), although these yields are still relatively low due to
the consumption of 3 in the formation of N-Cbz-pyrrolid-
ine. This side-reaction could be minimized by using an ex-
cess of the aldehyde, and by reducing the concentration of
pyrrolidine in the reaction mixture. Thus, using the condi-
tions of Table 1, entry 8, rac-4 could be isolated in 80%
yield after 4 d of reaction at room temperature in toluene.
We then proceeded with the conversion of rac-4 into ra-
Results and Discussion
We initiated our study by choosing as a benchmark reac-
tion the addition of N-(benzyloxycarbonyl)hydroxylamine cemic cispentacin. In contrast to the behaviour of monocy-
(3) to cyclopentene-1-carbaldehyde (2)[18] using pyrrolidine clic 5-hydroxyisoxazolidines, the oxidation of this com-
as the amine promoter. We were confident that although pound to the corresponding isoxazolidinone (i.e., rac-5) did
the primary product of the Michael addition would be pre- not take place with sodium chlorite,[13a] but the transforma-
sumably obtained as a cis/trans diastereomeric mixture, tion could be efficiently brought about using pyridinium
amine-catalysed epimerization of the exocyclic carbal- dichromate (PDC) in the presence of 4 Å molecular
dehyde moiety would lead to the selective formation of the sieves.[20] Finally, medium-pressure catalytic hydrogenation
desired isoxazolidine (i.e., 4), given the greater thermo- of rac-5 effected both the cleavage of the N–O bond and of
dynamic stability of cis-1-aza-2-oxabicyclo[3.3.0]octane sys- the Cbz amine-protecting group to give racemic cispentacin
tems relative to the trans isomers. This expectation turned (1) in quantitative yield (Scheme 2).
out to be right, and we found that in a variety of solvents,
We next examined the pyrrolidine-promoted Michael ad-
racemic bicyclic isoxazolidine 4 was isolated as a single dia- dition/cyclization of 3 to a set of acyclic α,β-unsaturated, α-
stereomer[19] in moderate to good yields (Table 1).
branched aldehydes 6a–d (Table 2).
Scheme 2. Synthesis of racemic cispentacin (1).
3104
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 3103–3111

Short Enantioselective Synthesis of Cispentacin
Table 2. Pyrrolidine-promoted aza-Michael/cyclization reaction of purity of the product was established after its essentially
α-branched enals 6a–d with N-Cbz-hydroxylamine (3).
quantitative conversion into isoxazolidinone 5 under the
conditions of Scheme 2. We initially screened a set of chiral
primary and secondary amine catalysts I–V (Table 3).
Table 3. Catalyst screening for the asymmetric aza-Michael/cycliza-
tion reaction of cyclopentene-1-carbaldehyde (2) with N-Cbz-hy-
droxylamine (3).
Entry[a]
Enal
R1, R2
Product
Yield [%][a]
dr[b]
1
2
3
4
6a
6b
6c
6d
Me, Me
Me, Et
Me, Ph
nPr, nBu
7a
7b
7c
7d
96
82
7:1
3:1
3:1
–
35[c]
0[d]
[a] Yield of isolated product after chromatographic purification.
1
[b] Determined by H NMR of the reaction product before chro-
matographic purification. [c] N-Cbz-pyrrolidine isolated in 44%
yield, together with an 11% of 3. [d] N-Cbz-pyrrolidine isolated in
18% yield, together with unreacted 3.
The reaction yields were heavily dependent on the substi-
tution pattern of the enal. Good yields were obtained both
with tiglic aldehyde (6a; Table 2, entry 1) and with 6b
(Table 2, entry 2). With α-methylcinnamaldehyde (6c), the
reaction was not complete after 6 d at room temperature,
and consumption of N-Cbz-hydroxylamine (3) to form N-
Cbz-pyrrolidine took place (Table 2, entry 3). On the other
hand, enal 6d failed to give any addition product (Table 2,
entry 4). Isoxazolidines 7a–c were obtained as inseparable
mixtures of diastereomers. Subsequent oxidation with PDC
gave racemic isoxazolidinones 8a–c in moderate to good
yields, also as diastereomeric mixtures (Scheme 3). In this
case, however, the two diastereomers could be readily sepa-
rated by column chromatography on silica gel, and the rela-
tive stereochemistry was established by means of NOESY
experiments on compounds 8a-trans, 8b-cis, and 8c-cis (see
Supporting Information). As expected, the trans isomers
were the major isomers in all instances.
Entry
Catalyst
Time [d][a]
Yield [%][b]
er[c]
1
2
3
4
I
10
5[d]
6[f]
4
8
2
73
16
0
73
84
80
91:9
II
III
IV
V
38:62[e]
–
38:62[e]
26:74[e]
33:67[e]
5
6[g]
V
1
[a] Time necessary for consumption of 3, monitored by H NMR
spectroscopy. [b] Yield of isolated 4 after chromatographic purifica-
tion. [c] Determined by chiral HPLC analysis of 5. [d] Reaction
stopped before completion. [e] The major enantiomer had the con-
figuration (3aR,6aS). [f] No reaction was observed after 6 d at
room temp. [g] Reaction performed in toluene in the presence of
water (2.0 equiv.) and acetic acid (0.15 equiv.).
We were pleased to find that the reaction could be cata-
lysed by the commercially available diphenylprolinol tri-
methylsilyl ether (I), giving bicyclic isoxazolidine 4 in 73%
yield and with good enantiomeric purity, although the reac-
Scheme 3. Synthesis of 4,5-disubstituted N-Cbz-isoxazolidinones
8a–c.
We then turned our attention to the possibility of using tion rate was much lower than that observed when pyrrol-
chiral amine catalysts to develop a catalytic asymmetric ver- idine was used as a catalyst under the same conditions
sion of this methodology. As a benchmark reaction, we (Table 3, entry 1). Prolinol (II) showed very poor catalytic
chose again the addition of N-(benzyloxycarbonyl)hydrox- activity (Table 3, entry 2), and in the presence of di-
ylamine (3) to cyclopentene-1-carbaldehyde (2), using chlo- phenylprolinol (III; Table 3, entry 3), no reaction was ob-
roform as solvent at room temperature. The enantiomeric served after 6 d. Primary amine catalyst IV, derived from
Eur. J. Org. Chem. 2013, 3103–3111
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3105

A. Pou, A. Moyano
FULL PAPER
cinchonidine,[21] led to the preferential formation of the high yields of 4, but eroded its enantiomeric purity (Table 4,
(3S,3aR,6aS) enantiomer of 4, in good yield but with poor entries 7 and 8). Lowering the reaction temperature to 4 °C
enantioselectivity (Table 3, entry 4). Better results were ob- (fridge) somewhat reduced the reaction rate, but gave 4 in
tained with Tsogoeva’s amine-thiourea catalyst V,[22] al- high yields and with excellent enantioselectivity (Table 4,
though the enantiomeric purity of 4 was only moderate entry 9); finally, further lowering of the reaction tempera-
(Table 3, entry 5). With this catalyst, the reaction was also ture to –15 °C greatly diminished the speed of the reaction
performed in toluene, with acetic acid as co-catalyst and and did not improve the performance of the process
in the presence of water;[22] these conditions increased the (Table 4, entry 10). The conditions of entry 9 (toluene as a
reaction rate, but 4 was obtained in a lower yield and with solvent, 20 mol-% of both catalyst I and benzoic acid, 4 °C)
a lower enantioselectivity (Table 3, entry 6).
appear to be the optimal conditions for the aza-Michael/
Having thus selected I as the most enantioselective cata- cyclization reaction between 2 and 3.
lyst, we proceeded to optimize the reaction conditions (sol-
Oxidation of the resulting non-racemic product 4 to iso-
vent, temperature, acidic co-catalysts) for the formation of xazolidinone 5, and catalytic hydrogenation under the con-
4 (Table 4).
ditions of Scheme 2 above, gave dextrorotatory cispentacin
(1), with a known (1S,2R) configuration,[11a] establishing
the (3R,3aS,6aR) stereochemistry of the major enantiomer
of 4 obtained with catalyst I, derived from l-proline. This
stereochemical sense of induction is the one expected for
iminium-catalysed Michael additions to enals with I.[23]
Thus, attack of the N-Cbz-hydroxylamine (3) to the less-
hindered Re-face of the β-carbon of the unsaturated imin-
ium ion derived from 2 and I gives rise to an (R) absolute
configuration at the newly created stereogenic centre in the
resulting exocyclic enamine, which ultimately leads to the
(1S,2R) enantiomer of cispentacin (Figure 1).
Table 4. Optimization studies for the asymmetric aza-Michael/
cyclization reaction of cyclopentene-1-carbaldehyde (2) with N-
Cbz-hydroxylamine (3) using catalyst I.
Entry
Solvent Additive
[mol-%]
T
[°C]
Time[a] Yield[b] er[c]
[d]
[%]
1
2
3
4
5
6
7
8
9
CHCl3
toluene
CH3CN –
EtOH
THF
toluene BzOH (20)
toluene BzOH (40)
toluene BzOH (80)
toluene BzOH (20)
toluene BzOH (20)
–
–
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
4
10
6
6
6
6
2
2
2
3
73
73
56
63
0[d]
86
99
98
98
80
91:9
92:8
91:9
86:14
–
91:9
87:13
86:14
98:2
86:14
–
–
Figure 1. Rationalization of the absolute configuration of the
major enantiomer of compound 4 under catalysis by I.
10
–15
10
1
[a] Time necessary for consumption of 3, monitored by H NMR
spectroscopy. [b] Yield of isolated 4 after chromatographic purifica-
tion. [c] Determined by chiral HPLC analysis of 5. [d] No reaction
was observed after 6 d at room temp.
Since the d-proline-derived catalyst ent-I is also commer-
cially available, these findings constitute a practical, high
yielding (93–98% global yield, three steps) and highly enan-
tioselective (up to 98:2 er) route to both enantiomers of
A solvent screening at room temp. (Table 4, entries 1–5) cispentacin (1), which compares favourably with previously
showed that the best results were obtained in toluene described asymmetric syntheses of this compound.[10–12]
(Table 4, entry 2). We therefore investigated the effect of
When we used the bis(3,5-trifluoromethylphenyl)prolinol
acid co-catalysts in this solvent. After some trials with dif- trimethylsilyl ether (VI) as the catalyst, the reaction between
ferent carboxylic acids, we found that an equimolar amount 2 and 3 (toluene, 4 d at 4 °C) unexpectedly gave a new com-
of benzoic acid (with respect to I) reduced the reaction time pound 9 instead of bicyclic isoxazolidine 4. Analysis of the
1
from 6 to 2 d, and increased the yield of 4 to 86%, essen- spectroscopic data (HRMS, IR, H and 13C NMR) of this
tially maintaining the enantioselectivity (Table 4, entry 6). compound strongly suggested that we had in fact obtained,
Further increases in the quantity of benzoic acid gave very as the sole reaction product, the trans Michael adduct 9 in
3106
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 3103–3111

Short Enantioselective Synthesis of Cispentacin
70% yield after chromatographic purification (Scheme 4).
This structural assignment was confirmed when, in the
presence of a 20 mol-% of pyrrolidine, a toluene solution of
9 was almost quantitatively converted into 4, for which an
enantiomeric ratio of 94:6 was found after oxidation to
isoxazolidinone 5.
Scheme 4. Asymmetric aza-Michael addition of N-Cbz-hydroxyl-
amine (3) to cyclopentene-1-carbaldehyde (2) using catalyst VI.
This result clearly shows that the stereochemical outcome
of the aza-Michael addition of 3 to 2 depends on the struc-
ture of the secondary amine catalyst. This behaviour can be
tentatively rationalized by assuming that in the catalytic cy-
cle, intermediate enamine B, arising from the attack of 3 on
unsaturated iminium A, is protonated to give a mixture of
saturated iminium cation cis-C and trans-D, which are sub-
Scheme 5. Rationalization of the stereochemical outcome of the
amine-catalysed aza-Michael addition of N-Cbz-hydroxylamine (3)
to cyclopentene-1-carbaldehyde (2).
sequently hydrolysed to give the corresponding aldehydes yields of racemic isoxazolidines rac-7a and rac-7b when pyr-
(Scheme 5). When using pyrrolidine or diphenylprolinol tri- rolidine was used as the promoter. The reactions were run
methylsilyl ether (I) as catalysts, intermediates cis-C and in toluene at 4 °C in the presence of 20 mol-% of I, and
trans-D are in equilibrium through enamine B, and al- required 10 d for the total consumption of 3. After removal
though presumably trans-D is more thermodynamically of the solvent and filtration through a short pad of silica
stable than cis-C, its hydrolysis is reversible, so that the sub- gel, isoxazolidines 7a and 7b were oxidized with pyridinium
sequent cyclization of the cis aldehyde completely shifts the dichromate, and the resulting products were submitted to
equilibrium to the formation of the isoxazolidine 4. On the column chromatography. The diastereomerically pure trans
other hand, when bis(3,5-trifluoromethylphenyl)prolinol isomers of 8a and 8b were obtained with 85:15 and with
trimethylsilyl ether (VI) is used as the catalyst, trans-D is 99:1 er, respectively (Scheme 6).[24] A (4R) absolute configu-
much more stable thermodynamically than cis-C, and its ration is assumed for the major enantiomers of both com-
hydrolysis is essentially irreversible. In this way, trans alde- pounds, in accordance with the mechanistic rationalization
hyde 9 cannot be equilibrated to 4.
of the absolute configuration of compound 4 under cataly-
Finally, we decided to briefly explore the feasibility of sis by I. These isoxazolidinones are immediate precursors
using I as a catalyst in the addition of 3 to the acyclic α,β- of (2R,3R)-3-amino-2-methylbutanoic acid and of (2R,3R)-
unsaturated aldehydes 6a and 6b, which had given good 3-amino-2-ethylbutanoic acid, respectively.[25]
Eur. J. Org. Chem. 2013, 3103–3111
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3107

A. Pou, A. Moyano
FULL PAPER
bottomed flasks or in loosely stoppered glass vials. Commercially
available reagents, catalysts, and solvents were used as received,
with the exception of catalyst VI, which was chromatographically
purified immediately prior to use, and dichloromethane, which was
distilled from calcium hydride. Aldehydes 2,[18] 6b,[27] and 6d[27]
were prepared according to literature procedures. For thin-layer
chromatography, silica gel plates Merck 60 F254 were used, and
compounds were visualized by irradiation with UV light and/or by
treatment with
a solution of phosphomolybdic acid (2.5 g),
Ce(SO4)2·2H2O (1.0 g), conc. H2SO4 (1.0 mL), and H2O (94 mL)
followed by heating, or by treatment with a solution of p-anisal-
dehyde (2.3 mL), conc. H2SO4 (3.5 mL), acetic acid (1.0 mL), and
ethanol (90 mL) followed by heating. Flash column chromatog-
raphy was performed using silica gel Merck 60 (particle size: 0.040–
0.063 mm). IR spectra were obtained with a Nicolet 6700 FTIR
apparatus, using ATR techniques. 1H (400 MHz) and 13C
(100.6 MHz) NMR spectra were recorded with a Varian Mercury
400 spectrometer. Chemical shifts (δ) are given in ppm relative to
tetramethylsilane (TMS), and coupling constants (J) are given in
Hz. The spectra were recorded in CDCl3 as solvent at room tem-
perature. TMS served as an internal standard (δ = 0.00 ppm) for
1H NMR spectra, and CDCl3 (δ = 77.0 ppm) for 13C NMR spectra.
Data are reported as follows: s, singlet; d, doublet, t, triplet; q,
quartet; m, multiplet; br, broad signal. Where appropriate, 2D tech-
Scheme 6. Enantioselective synthesis of isoxazolidinones 8a and 8b.
Conclusions
In conclusion, we have demonstrated that the organocat- niques (COSY, NOESY) were also used to assist in structure eluci-
dation. High-resolution mass spectra (HRMS) were recorded with
a Bruker MicrOTOF spectrometer by the Unitat d’Espectrometria
de Masses, CCiT-UB. Specific rotations were determined at room
temp. with a Perkin–Elmer 241 MC polarimeter. Chiral HPLC
analyses were performed with a Shimadzu LC Series 20 apparatus
with an M20 diode array UV/Vis detector, using Chiralpak® IA,
IB, and IC columns. The homogeneity of the peaks corresponding
to the two enantiomers of the product was thoroughly checked by
comparison of the UV spectra.
alytic Michael addition of N-(benzyloxycarbonyl)hydroxyl-
amine (3) to cyclopentene-2-carbaldehyde 2 takes place in
good yields to give cyclized isoxazolidine 4 as a dia-
stereomerically pure compound. Racemic 4 can be obtained
by using pyrrolidine, and a variety of chiral amino catalysts
can be used to obtain non-racemic 4 with variable yields
and enantioselectivities. Commercially available (S)-di-
phenylprolinol trimethylsilyl ether (I), in the presence of
benzoic acid as a co-catalyst, provided 4 in 98% yield (after
chromatographic purification) and with an excellent 98:2 er
when the reaction is run in toluene at 4 °C. Oxidation of 4
with pyridinium dichromate, followed by catalytic hydro-
genation, gave (1S,2R)-cispentacin 1 in essentially quantita-
Benzyl
(3R*,3aS*,6aR*)-3-Hydroxyhexahydro-1H-cyclopenta[c]-
isoxazole-1-carboxylate (rac-4): Pyrrolidine (from a 10 mg/mL solu-
tion in toluene; 0.74 mL, 0.10 mmol) and N-Cbz-hydroxylamine (3;
87 mg, 0.52 mmol) were added sequentially to a magnetically
stirred solution of cyclopentene-1-carbaldehyde (2; 100 mg,
tive yield. Since the (R) enantiomer of I is also commer- 1.04 mmol) in toluene (2 mL), and the resulting solution was stirred
at room temp. The progress of the reaction was monitored both by
1H NMR spectroscopy and by TLC. When no starting material 3
was detected (4 d), toluene and pyrrolidine were removed in vacuo,
and the crude residue was purified by column chromatography (sil-
ica gel; 4:1 hexane/ethyl acetate mixtures) to give rac-4 (110 mg,
80%) as a colourless oil. 1H NMR (400 MHz, CDCl3): δ = 7.33
(m, 5 H), 5.35 (s, 1 H), 5.20 (part A of AB system, J = 12.7 Hz, 1
H), 5.17 (part B of AB system, J = 12.7 Hz, 1 H), 4.78–4.73 (m, 1
H), 2.93–2.86 (m, 1 H), 1.51–1.70 (m, 6 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 187.2, 135.8, 128.5, 128.2, 128.0, 104.3,
cially available, the same synthetic sequence can be applied
to the preparation of the (1R,2S) natural enantiomer of cis-
pentacin, and this demonstrates the utility of organocata-
lysis in the preparation of medicinally relevant drugs and
natural products.[26] Unexpectedly, the bis(3,5-trifluoro-
methylphenyl)prolinol trimethylsilyl ether catalyst (VI) pro-
duced the trans Michael adduct (i.e., 9) in 70% yield (94:6
er) after chromatographic purification. When using acyclic
α-branched α,β-unsaturated aldehydes 6a–d as substrates,
the reaction yields depend on the substitution pattern of 67.8, 63.7, 53.0, 34.2, 29.1, 25.0 ppm. HRMS (ESI): calcd. for
C14H17NNaO4 [M + Na]+ 286.1055; found 286.1060.
the aldehydes, and mixtures of cis- and trans-isomers were
obtained. Nevertheless, the present strategy has proved to
be successful in some instances, and oxazolidinone trans-8b
Benzyl
(3aS*,6aR*)-3-Oxohexahydro-1H-cyclopentaisoxazol-1-
carboxylate (rac-5): Racemic isoxazolidine 4 (134 mg, 0.50 mmol)
could be obtained in 70% overall yield from the reaction of and pyridinium dichromate (376 mg, 1.0 mmol) were added se-
quentially to a stirred suspension of molecular sieves (4 Å, pow-
dered; 750 mg) in anhydrous dichloromethane (4 mL), and the re-
action mixture was stirred overnight at room temperature. After
the addition of hexane/ethyl acetate (4:1; 10 mL) to precipitate the
chromium salts, the solution was decanted and filtered through a
short pad of Celite®. After removal of the solvents in vacuo, the
reaction product was purified by column chromatography (silica
gel, hexane/ethyl acetate mixtures) to give racemic isoxazolidinone
6b and 3 under catalysis with I, and with high enantiomeric
purity (99:1 er).
Experimental Section
General Methods: Room-temperature reactions were generally per-
formed with magnetic stirring and open to the air, either in round-
3108
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 3103–3111

Short Enantioselective Synthesis of Cispentacin
1
5 (127 mg, 98%) as a colourless oil. H NMR (400 MHz, CDCl3):
H), 5.26 (part A of AB system, J = 12.3 Hz, 1 H), 5.25 (part B of
AB system, J = 12.3 Hz, 1 H), 4.06–3.99 (m, 1 H), 2.65–2.56 (m, 1
H), 1.51 (d, J = 5.8 Hz, 3 H), 1.28 (d, J = 7.6 Hz, 3 H) ppm. 13C
NMR (100 MHz, CDCl3): δ = 173.8, 156.8, 134.9, 128.6, 128.3,
128.2, 68.7, 43.1, 29.5, 19.4, 9.23 ppm. HRMS (ESI): calcd. for
δ = 7.41–7.31 (m, 5 H), 5.25 (part A of AB system, J = 12.5 Hz, 1
H), 5.24 (part B of AB system, J = 12.5 Hz, 1 H), 4.88 (td, J =
8.2, J = 2.5 Hz, 1 H), 3.35 (td, J = 8.2, J = 2.5 Hz, 1 H), 2.26–2.18
(m, 1 H), 2.14–2.06 (m, 1 H), 2.02–1.85 (m, 2 H), 1.84–1.74 (m, 1
H), 1.72–1.60 (m, 1 H) ppm. 13C NMR (100 MHz, CDCl3): δ = C13H16NO4 [M + H]+ 250.1079; found 250.1075.
175.0, 155.7, 135.0, 128.6, 128.3, 68.6, 65.1, 45.1, 34.8, 31.0, 29.6,
(3S*,4R*)-Benzyl
(rac-cis-8a): H NMR (400 MHz, CDCl3): δ = 7.46–7.42 (m, 2 H),
3,4-Dimethyl-5-oxoisoxazolidin-2-carboxylate
24.1 ppm. HRMS (ESI): calcd. for C14H15NNaO4 [M + Na]+
284.0899; found 284.0897.
1
7.40–7.31 (m, 3 H), 5.34 (s, 2 H), 4.80–4.72 (m, 1 H), 3.02–2.94 (m,
General Procedure for the Pyrrolidine-Promoted Michael Addition/
1 H), 1.32 (d, J = 6.0 Hz, 3 H), 1.22 (d, J = 7.0 Hz, 3 H) ppm. 13C
Cyclization Reaction Between Aldehydes 6a–d and N-Cbz-Hydroxyl- NMR (100 MHz, CDCl3): δ = 170.6, 134.6, 128.6, 128.4, 114.0,
amine 3: Pyrrolidine (0.10 mL, 1.21 mmol) and N-Cbz-hydroxyl-
amine (3; 200 mg, 1.22 mmol) were added sequentially to a stirred
solution of unsaturated aldehyde 6a–d (1.22 mmol) in chloroform
(4 mL). After 6 d at room temperature, the solvent and the pyrroli-
dine were removed in vacuo, and the resulting product was purified
by column chromatography (silica gel, hexane/ethyl acetate mix-
tures) to give racemic oxazolidines 7a–d as mixtures of dia-
stereomers.
68.8, 42.9, 29.7, 13.7, 9.3 ppm. HRMS (ESI): calcd. for C13H16NO4
[M + H]+ 250.1079; found 250.1077.
Benzyl 3-Ethyl-4-methyl-5-oxoisoxazolidin-2-carboxylate (8b): Ra-
cemic isoxazolidine 7b (100 mg, 0.38 mmol, diastereomeric mix-
ture) and pyridinium dichromate (300 mg, 0.8 mmol) were added
sequentially to a stirred suspension of molecular sieves (4 Å, pow-
dered; 900 mg) in anhydrous dichloromethane (3 mL), and the re-
action mixture was stirred overnight at room temperature. After
the addition of hexane/ethyl acetate (4:1; 10 mL) to precipitate the
chromium salts, the solution was decanted and filtered through a
short pad of Celite®. After removal of the solvents in vacuo, 8b
(90 mg, 90%) was obtained as a 7:1 trans/cis diastereomeric mix-
ture. The two diastereomers were separated by column chromatog-
Benzyl 5-Hydroxy-3,4-dimethylisoxazolidin-2-carboxylate (7a): Ob-
tained from 6a as a 1:7 diastereomeric mixture in 96% yield by
the general procedure described above. Colourless oil. 1H NMR
(400 MHz, CDCl3): δ = 7.38–7.29 (m, 5 H, both), 5.46 (s, 1 H,
major), 5.29 (s, 1 H, minor), 5.24–5.18 (m, 2 H, both), 4.39–4.30
(m, 1 H, minor), 3.79–3.70 (m, 1 H, major), 2.60–2.51 (m, 1 H, raphy (silica gel, hexane/ethyl acetate mixtures).
minor), 2.16–2.06 (m, 1 H, major), 1.34 (d, J = 6.2 Hz, 3 H, major),
(3R*,4R*)-Benzyl 3-Ethyl-4-methyl-5-oxoisoxazolidin-2-carboxylate
1.25 (d, J = 7.2 Hz, 3 H, minor), 1.09 (d, J = 6.2 Hz, 3 H, major),
0.99 (d, J = 7.2 Hz, 3 H, minor) ppm. 13C NMR (100 MHz,
CDCl3): δ = 159.3, 135.8, 128.5, 128.2, 127.9, 103.7 (minor), 99.2
(major), 67.9, 60.5 (major), 56.9 (minor), 48.7 (major), 45.7
(minor), 19.7 (major), 15.4 (minor), 12.0 (minor), 10.8 (major)
ppm.
(rac-trans-8b): 1H NMR (400 MHz, CDCl3): δ = 7.41–7.34 (m, 5
H), 5.27 (part A of AB system, J = 12.3 Hz, 1 H), 5.26 (part B of
AB system, J = 12.3 Hz, 1 H), 4.11–4.06 (m, 1 H), 2.67–2.58 (m, 1
H), 1.87–1.69 (m, 2 H), 1.27 (d, J = 8.0 Hz, 3 H), 1.00 (t, J =
7.4 Hz, 3 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 175.3, 157.2,
135.0, 128.7, 128.6, 128.3, 68.8, 68.7, 40.4, 26.5, 15.1, 9.3 ppm.
Benzyl 3-Ethyl-5-hydroxy-4-methylisoxazolidin-2-carboxylate (7b): HRMS (ESI): calcd. for C14H18NO4 [M + H]+ 264.1236; found
Obtained from 6b as a 1:3 diastereomeric mixture in 82% yield by
the general procedure described above. Colourless oil. 1H NMR
(400 MHz, CDCl3): δ = 7.35 (m, 5 H, both), 5.48 (s, 1 H, both),
5.25–5.12 (m, 2 H, both), 4.17–4.01 (m, 1 H, minor), 3.76–3.67 (m,
1 H, major), 2.27–2.14 (m, 1 H, major), 2.02 (s, 1 H, minor), 1.78–
1.43 (m, 2 H, both), 1.11 (d, J = 6.6 Hz, 3 H, both), 0.99–0.91 (m,
3 H, both) ppm.
264.1235.
(3S*,4R*)-Benzyl 3-Ethyl-4-methyl-5-oxoisoxazolidin-2-carboxylate
(rac-cis-8b): H NMR (400 MHz, CDCl3): δ = 7.47–7.42 (m, 3 H),
1
7.41–7.31 (m, 2 H), 5.35 (part A of AB system, J = 12.3 Hz, 1 H),
5.34 (part B of AB system, J = 12.3 Hz, 1 H), 4.51–4.44 (m, 1 H),
3.04–2.95 (m, 1 H), 1.76–1.52 (m, 2 H), 1.23 (d, J = 6.8 Hz, 3 H),
1.03 (t, J = 8.0 Hz, 3 H) ppm. 13C NMR (100 MHz, CDCl3): δ =
170.8, 147.9, 134.7, 128.6, 128.4, 83.8, 68.7, 42.4, 29.7, 21.1, 9.6,
Benzyl 3-Ethyl-5-hydroxy-4-methylisoxazolidin-2-carboxylate (7c):
Obtained from 6c as a 1:3 diastereomeric mixture in 35% yield by 9.2 ppm. HRMS (ESI): calcd. for C14H18NO4 [M + H]+ 264.1236;
the general procedure described above. Colourless oil. 1H NMR
(400 MHz, CDCl3): δ = 7.39–7.08 (m, 10 H, both), 5.65 (s, 1 H,
major), 5.46 (d, J = 3.7 Hz, 1 H, minor), 5.15 (s, 2 H, both), 4.67
(d, J = 9.18 Hz, 1 H, major), 2.92–2.84 (m, 1 H, minor), 2.54–2.44
(m, 1 H, major), 1.76 (br., 1 H, OH), 1.15 (d, J = 6.4 Hz, 3 H,
major), 0.66 (d, J = 7.0 Hz, 3 H, minor) ppm.
found 264.1232.
Benzyl 4-Methyl-5-oxo-3-phenylisoxazolidin-2-carboxylate (8c): Ra-
cemic isoxazolidine 7c (40 mg, 0.13 mmol, diastereomeric mixture)
and pyridinium dichromate (100 mg, 0.25 mmol) were added se-
quentially to a stirred suspension of molecular sieves (4 Å, pow-
dered; 360 mg) in anhydrous dichloromethane (1 mL), and the re-
action mixture was stirred overnight at room temperature. Follow-
ing the addition of hexane/ethyl acetate (4:1; 5 mL) to precipitate
the chromium salts, the solution was decanted and filtered through
a short pad of Celite®. After removal of the solvents in vacuo, 8c
(22 mg, 55%) was obtained as a 3:1 trans/cis diastereomeric mix-
ture. The two diastereomers were separated by column chromatog-
raphy (silica gel, hexane/ethyl acetate mixtures).
Benzyl 3,4-Dimethyl-5-oxoisoxazolidin-2-carboxylate (8a): Racemic
isoxazolidine 7a (100 mg, 0.40 mmol, diastereomeric mixture) and
pyridinium dichromate (300 mg, 0.8 mmol) were added sequentially
to a stirred suspension of molecular sieves (4 Å, powdered; 900 mg)
in anhydrous dichloromethane (3 mL), and the reaction mixture
was stirred overnight at room temperature. After the addition of
hexane/ethyl acetate (4:1; 10 mL) to precipitate the chromium salts,
the solution was decanted and filtered through a short pad of
Celite®. After removal of the solvents in vacuo, 8a (84 mg, 85%)
was obtained as a 7:1 trans/cis diastereomeric mixture. The two
diastereomers were separated by column chromatography (silica
gel, hexane/ethyl acetate mixtures).
(3S*,4R*)-Benzyl 4-Methyl-5-oxo-3-phenylisoxazolidin-2-carboxyl-
ate (rac-trans-8c): H NMR (400 MHz, CDCl3): δ = 7.48–7.18 (m,
10 H), 5.18 (s, 2 H), 4.90 (d, J = 8.6 Hz, 1 H), 3.02–2.93 (m, 1 H),
1.35 (d, J = 6.6 Hz, 3 H) ppm. 13C NMR (100 MHz, CDCl3): δ =
173.0, 156.4, 137.8, 134.7, 129.1, 128.7, 128.5, 128.1, 126.0, 109.9,
1
(3R*,4R*)-Benzyl
3,4-Dimethyl-5-oxoisoxazolidin-2-carboxylate 71.1, 68.8, 45.2, 13.1 ppm. HRMS (ESI): calcd. for C18H18NO4 [M
(rac-trans-8a): 1H NMR (400 MHz, CDCl3): δ = 7.31–7.14 (m, 5
+ H]+ 312.1236; found 312.1233.
Eur. J. Org. Chem. 2013, 3103–3111
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3109

A. Pou, A. Moyano
FULL PAPER
(3R*,4R*)-Benzyl 4-Methyl-5-oxo-3-phenylisoxazolidin-2-carboxyl-
ate (rac-cis-8c): H NMR (400 MHz, CDCl3): δ = 7.38–7.26 (m, 8
chromatography (silica gel; 4:1 hexane/ethyl acetate mixture) to
give 4 (94 mg, 98%). Oxidation of 4 to the oxazolidinone under the
conditions described above gave (3aS,6aR)-5 in essentially quanti-
tative yield and with 94:6 er.
1
H), 7.23–7.19 (m, 2 H), 5.62 (d, J = 9.2 Hz, 1 H), 5.21 (part A of
AB system, J = 12.1 Hz, 1 H), 5.20 (part B of AB system, J =
12.1 Hz, 1 H), 3.40–3.32 (m, 1 H), 0.92 (d, J = 7.6 Hz, 3 H) ppm.
13C NMR (100 MHz, CDCl3): δ = 174.6, 156.7, 134.7, 129.0, 128.7,
128.6, 128.2, 126.6, 68.9, 67.1, 29.7, 10.6 ppm. HRMS (ESI): calcd.
for C18H18NO4 [M + H]+ 312.1236; found 312.1231.
General Procedure for the Reaction of Aldehydes 6a and 6b with N-
Cbz-hydroxylamine 3 Catalysed by I: Diphenylprolinol trimethyl-
silyl ether (I; 20 mg, 0.12 mmol) and N-Cbz-hydroxylamine (3;
100 mg, 0.60 mmol) were added sequentially to a magnetically
stirred solution of unsaturated aldehyde 6a or 6b (1.18 mmol) in
toluene (2 mL), and the resulting solution was stirred at 4 °C. The
progress of the reaction was monitored both by 1H NMR spec-
troscopy and by TLC. When no starting material 3 was detected
(10 d), toluene was removed in vacuo, and the crude residue was
filtered through a short pad of silica gel, washing with a 4:1 hexane/
ethyl acetate mixture, to give crude isoxazolidines 7a and 7b as
colourless oils. Without further purification, the isoxazolidine was
dissolved in anhydrous dichloromethane (3 mL). After the addition
of molecular sieves (4 Å, powdered; 900 mg) and pyridinium di-
chromate (330 mg, 0.90 mmol), the resulting suspension was mag-
netically stirred overnight at room temperature. Following the ad-
dition of hexane/ethyl acetate (4:1; 10 mL) to precipitate the chro-
mium salts, the solution was decanted and filtered through a short
pad of Celite®. The solvents were removed in vacuo, and the crude
residue was purified by column chromatography (silica gel; hexane/
ethyl acetate mixtures) to give isoxazolidinones 8a and 8b.
Benzyl (3R,3aS,6aR)-3-Hydroxyhexahydro-1H-cyclopenta[c]isox-
azole-1-carboxylate (4): Diphenylprolinol trimethylsilyl ether (I;
26 mg, 0.10 mmol), benzoic acid (12.8 mg, 0.10 mmol), and N-Cbz-
hydroxylamine (3; 87 mg, 0.52 mmol) were added sequentially to
a magnetically stirred solution of cyclopentene-1-carbaldehyde (2;
100 mg, 1.04 mmol) in toluene (2 mL), and the resulting solution
was stirred at 4 °C. The progress of the reaction was monitored
both by 1H NMR spectroscopy and by TLC. When no starting
material 3 was detected (3 d), toluene and pyrrolidine were removed
in vacuo, and the crude residue was purified by column chromatog-
raphy (silica gel; 4:1 hexane/ethyl acetate mixture) to give 4
(183 mg, 98%) as a colourless oil.
Benzyl (3aS,6aR)-3-Oxohexahydro-1H-cyclopentaisoxazole-1-carb-
oxylate (5): Oxidation of (3R,3aS,6aR)-4 with pyridinium dichro-
mate (0.50 mmol scale) under the conditions described above for
the racemic compound gave (3aS,6aR)-5 in essentially quantitative
yield and with 98:2 er. [α]2D5 = –45 (c = 2.3, CH2Cl2). HPLC (Chi-
ralpak® IA column, 90:10 hexane/isopropyl alcohol, 1.0 mL/min, λ
= 220 nm, 25 °C): tR = 11.2 min (minor), 13.5 min (major).
(3R,4R)-Benzyl
3,4-Dimethyl-5-oxoisoxazolidin-2-carboxylate
(trans-8a): 59 mg, 41% yield, 85:15 er. [α]2D5 = –9.1 (c = 0.33,
CH2Cl2). HPLC (Chiralpak® IA column, 95:5 hexane/isopropyl
alcohol, 1.0 mL/min, λ = 220 nm, 25 °C): tR = 9.6 min (major),
11.2 min (minor).
(1S,2R)-Cispentacin (1): A stirred solution of (3aS,6aR)-5 (118 mg,
0.45 mmol, 98:2 er) in MeOH (4 mL) was hydrogenated at 60 atm
using Pd/C (10%; 11.8 mg) as the catalyst in a medium-pressure
reactor. After 24 h at 65 °C, the reaction mixture was filtered
through a short pad of Celite®, which was washed with ethyl acet-
ate (2ϫ 5 mL). Evaporation of the solvents in vacuo gave cispenta-
cin 1 (54 mg, 93%) as a colourless crystalline solid, m.p. 199–
202 °C. [ref.[11a] m.p. 198–200 °C]. [α]2D5 = +6 (c = 1.1, H2O).
(3S,4R)-Benzyl 3,4-Dimethyl-5-oxoisoxazolidin-2-carboxylate (cis-
8a): 48 mg, 34% yield, er not determined. [α]2D5 = –1.9 (c = 0.53,
CH2Cl2).
(3R,4R)-Benzyl 3-Ethyl-4-methyl-5-oxoisoxazolidin-2-carboxylate
(trans-8b): 86 mg, 70% yield, 99:1 er. [α]2D5 = –10.6 (c = 0.47,
CH2Cl2). HPLC (Chiralpak® IC column, 93:7 hexane/isopropyl
alcohol, 0.8 mL/min, λ = 220 nm, 25 °C): tR = 27.4 min (minor),
28.5 min (major).
{ref.[11a] [α]2D2 = +9 (c = 1.1, H2O)}. H and 13C NMR spectra (see
1
Supporting Information) were identical to those described in the
literature.
(1R,2S)-Benzyl 2-Formylcyclopentyl-N-hydroxycarbamate (9): Cat-
alyst VI (62 mg, 0.10 mmol) and N-Cbz-hydroxylamine (3; 87 mg,
0.52 mmol) were added sequentially to a magnetically stirred solu-
tion of cyclopentene-1-carbaldehyde (2; 100 mg, 1.04 mmol) in tol-
uene (2 mL), and the resulting solution was stirred at 4 °C (fridge).
The progress of the reaction was monitored both by 1H NMR spec-
troscopy and by TLC. When no starting material 3 was detected
(4 d), the solvent was removed in vacuo, and the crude residue was
purified by column chromatography (silica gel; 4:1 hexane/ethyl
acetate mixtures) to give uncyclized Michael adduct 9 (96 mg, 70%)
(3S,4R)-Benzyl 3-Ethyl-4-methyl-5-oxoisoxazolidin-2-carboxylate
(cis-8b): 34 mg, 27% yield.
Supporting Information (see footnote on the first page of this arti-
1
cle): Copies of the H and 13C NMR spectra of compounds 1, 4,
5, 7a–c, cis-8a–c, trans-8a–c, and 9. NOESY spectra of compounds
trans-8a, cis-8b, and cis-8c. HPLC traces of compounds 5, trans-
8a, and trans-8b.
as a colourless oil. IR (ATR): ν = 3300 (broad), 3032, 2954, 2873,
˜
Acknowledgments
1
1695, 1425, 1401, 1295, 1075, 735, 690 cm–1. H NMR (400 MHz,
CDCl3): δ = 9.64 (d, J = 2.5 Hz, 1 H), 7.35 (m, 5 H), 5.17 (part A
of AB system, J = 12.4 Hz, 1 H), 5.16 (part B of AB system, J =
12.4 Hz, 1 H), 4.81 (m, 1 H), 3.11 (m, 1 H), 1.95 (m, 4 H), 1.83
(m, 2 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 202.2, 157.1,
135.6, 128.6, 128.5, 128.3, 68.3, 59.8, 53.8, 28.6, 25.8, 23.8 ppm.
HRMS (ESI): calcd. for C14H18NO4 [M + H]+ 264.1236; found
264.1231.
Financial support from Ministerio de Ciencia e Innovación/Minis-
terio de Economía y Competitividad (MICINN/MINECO) (pro-
ject number AYA2009-13920-C02-02) and from ENANTIA, S. L.
(to A. P.) is gratefully acknowledged. The authors also wish to
acknowledge COST Action CM0905 (ORCA), Dr. R. Rios (South-
ampton University) for helpful discussions, Prof. S. B. Tsogoeva
(Erlangen University) for a sample of catalyst V, and Dr. F.
Cárdenas (CCiT-UB) for performing the NOESY experiments.
A solution of this compound (0.70 mmol) in toluene (1 mL) was
treated with pyrrolidine (from a 10 mg/mL solution in toluene;
0.74 mL, 0.10 mmol), and the resulting solution was magnetically
stirred overnight at room temperature. The solvent was distilled off
under reduced pressure, and the residue was purified by column
[1] For reviews on β-amino acids, see: a) O. W. Griffith, Ann. Rev.
Biochem. 1986, 55, 855–978; b) E. Juaristi (Ed.), Enantioselec-
tive Synthesis of β-Amino Acids, Wiley-VCH, New York, 1997;
3110
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 3103–3111

Short Enantioselective Synthesis of Cispentacin
c) F. Fülop, Chem. Rev. 2001, 101, 2181–2204; d) F. Fülöp, in:
Studies in Natural Products Chemistry (Ed.: Atta-ur-Rahman),
Elsevier Science Publishers, New York, 2000, vol. 22, pp. 273–
306; e) M. Liu, M. P. Sibi, Tetrahedron 2002, 58, 7991–8035; f)
G. Lelais, D. Seebach, Biopolymers 2004, 76, 206–243; g) E.
Juaristi, V. A. Soloshonok (Eds.), Enantioselective Synthesis of
β-Amino Acids, 2nd ed., Wiley–VCH, New York, 2005.
[2] a) J. Escalante, M. A. González-Totozin, J. Aviña, O. Muñoz-
Muñiz, E. Juaristi, Tetrahedron 2001, 57, 1883–1890; b) D.
Krishnaswamy, V. V. Govande, V. K. Gumaste, B. M. Whawal,
A. R. A. S. Desmukh, Tetrahedron 2002, 58, 2215–2225; c)
M. C. Sleeman, C. H. MacKinnon, K. S. Hewitson, C. J. Scho-
field, Bioorg. Med. Chem. Lett. 2002, 12, 597–599.
[3] a) D. L. Steer, R. A. Lew, P. Perlmutter, A. I. Smith, M. I. Agu-
ilar, Curr. Med. Chem. 2002, 9, 811–822; b) G. A. Lengyel,
R. C. Frank, W. S. Horne, J. Am. Chem. Soc. 2011, 133, 4246–
4249.
[4] F. Fülöp, Farmaco 2000, 55, 181–183.
[5] M. Konishi, M. Nishio, K. Saitoh, T. Miyaki, T. Oki, H. Kawa-
guchi, J. Antibiot. 1989, 42, 1749–1755.
amines to α,β-unsaturated carbonyls, see: f) Y. K. Chen, M.
Yoshida, D. W. C. MacMillan, J. Am. Chem. Soc. 2006, 128,
9328–9329; g) D. Pettersen, F. Piana, L. Bernardi, F. Fini, M.
Fochi, V. Sgarzani, A. Ricci, Tetrahedron Lett. 2007, 48, 7805–
7808.
[15] See, inter alia: a) L. Deiana, G.-L. Zhao, S. Lin, P. Dziedzic,
Q. Zhang, H. Leijonmarck, A. Córdova, Adv. Synth. Catal.
2010, 352, 3201–3207; b) P. Galzerano, F. Pesciaioli, A. Maz-
zanti, G. Bartoli, P. Melchiorre, Angew. Chem. 2009, 121, 8032–
8034; Angew. Chem. Int. Ed. 2009, 48, 7892–7894; c) A. De-
smarchelier, D. P. de Sant’Ana, V. Terrasson, J. M. Campagne,
X. Moreau, C. Greck, R. M. de Figueiredo, Eur. J. Org. Chem.
2011, 4046–4052.
[16] a) H. D. King, Z. Meng, D. Denhart, R. Mattson, R. Kimura,
D. Wu, Q. Gao, J. E. Macor, Org. Lett. 2005, 7, 3437–3440; b)
E.-M. Tanzer, L. E. Zimmer, W. B. Schweizer, R. Gilmour,
Chem. Eur. J. 2012, 18, 11334–11342.
[17] For the chiral secondary amine catalysis of 1,3-dipolar cycload-
ditions of cyclopentene-1-carbaldehyde with nitrones, see: a) S.
Karlsson, H.-E. Högberg, Eur. J. Org. Chem. 2003, 2782–2791;
b) S. S. Chow, M. Nevalainen, C. A. Evans, C. W. Johannes,
Tetrahedron Lett. 2007, 48, 277–280.
[6] T. Iwamoto, E. Tsujii, M. Ezaki, A. Fujie, S. Hashimoto, M.
Okuhara, M. Kohsaka, H. Imanaka, J. Antibiot. 1990, 43, 1–
7.
[18] Easily prepared in a one-pot process from racemic trans-1,2-
cyclohexanediol, as described in: M. J. Meyers, G. B. Arhancet,
S. L. Hockerman, X. Chen, S. A. Long, M. W. Mahoney, J. R.
Rico, D. J. Garland, J. R. Blinn, J. T. Collins, S. Yang, H. C.
Huang, K. F. McGee, J. M. Wendling, J. D. Dietz, M. A.
Payne, B. L. Homer, M. I. Heron, D. B. Reitz, X. Hu, J. Med.
Chem. 2010, 53, 5979–6002.
[7] a) H. Ohki, Y. Inamoto, K. Kawabata, T. Kamimura, K. Sak-
ane, J. Antibiot. 1991, 44, 546–549; b) K. Ziegelbauer, Anti-
microb. Agents Chemother. 1998, 42, 1581–1586; c) J. Mitten-
dorf, F. Kunisch, M. Matzke, H. C. Militzer, A. Schmidth, W.
Schönfeld, Bioorg. Med. Chem. Lett. 2003, 13, 433–436.
[8] a) K. Kawabata, Y. Inamoto, K. Sakane, J. Antibiot. 1990, 43,
513–518; b) T. Yamazaki, Y.-F. Zhu, A. Probstl, R. K. Chadha,
M. Goodman, J. Org. Chem. 1991, 56, 6644–6655.
[19] The fact that the hemiacetalic C-3 proton of 4 appears as a
1
singlet in the H NMR spectrum strongly suggests that in this
[9] a) C. Evans, R. McCague, S. M. Roberts, A. G. Sutherland, R.
Wisdom, J. Chem. Soc. Perkin Trans. 1 1991, 2276–2277; b) T.
Konosu, S. Oida, Chem. Pharm. Bull. 1993, 41, 1012–1018; c)
L. T. Kanerva, P. Csomós, O. Sundholm, G. Bernáth, F. Fülöp,
Tetrahedron: Asymmetry 1996, 7, 1705–1716; d) F. Theil, S.
Ballschuh, Tetrahedron: Asymmetry 1996, 7, 3565–3572; e) E.
Forró, F. Fülöp, Org. Lett. 2003, 5, 1209–1212; f) E. Forró, F.
Fülöp, Chem. Eur. J. 2007, 13, 6397–6401; g) E. Forró, F.
Fülöp, Tetrahedron: Asymmetry 2008, 19, 1005–1009.
[10] S. G. Davies, O. Ichihara, I. Lenoir, I. A. S. Walters, J. Chem.
Soc. Perkin Trans. 1 1994, 1411–1415.
[11] a) V. K. Aggarwal, S. J. Roseblade, J. K. Barrell, R. Alexander,
Org. Lett. 2002, 4, 1227–1229; b) V. K. Aggarwal, S. Roseblade,
R. Alexander, Org. Biomol. Chem. 2003, 1, 684–691.
[12] a) C. Bolm, I. Schiffers, C. L. Dinter, L. Defrère, A. Gerlach,
G. Raabe, Synthesis 2001, 1719–1730; b) C. Bolm, I. Schiffers,
I. Atodiresei, C. P. R. Hackenberger, Tetrahedron: Asymmetry
2003, 14, 3455–3467.
[13] a) I. Ibrahem, R. Rios, J. Vesely, G.-L. Zhao, A. Córdova,
Chem. Commun. 2007, 849–851; b) G.-L. Zhao, S. Lin, A. Kor-
otvicka, L. Deiana, M. Kullberg, A. Córdova, Adv. Synth. Ca-
tal. 2010, 352, 2291–2298. See also: c) O. V. Maltsev, A. S. Ku-
cherenko, A. L. Chimishkyan, S. G. Zlotin, Tetrahedron: Asym-
metry 2010, 21, 2659–2670.
[14] For reviews on asymmetric aza-Michael reactions, see: a) M.
Liu, M. P. Sibi, Tetrahedron 2002, 58, 7991–8035; b) J. L. Vic-
ario, D. Badía, L. Carrillo, J. Etxebarria, E. Reyes, N. Ruiz,
Org. Prep. Proced. Int. 2005, 37, 513–538; c) L.-W. Xu, C.-G.
Xia, Eur. J. Org. Chem. 2005, 633–639; d) D. Enders, C. Wang,
J. X. Liebich, Chem. Eur. J. 2009, 15, 11058–11076; e) J. Wang,
P. Li, P. Y. Choy, A. S. C. Chan, F. Y. Kwong, ChemCatChem
2012, 4, 917–925. For seminal reports on asymmetric organ-
ocatalytic aza-Michael additions of O-substituted hydroxyl-
compound, the hydroxy group is on the endo face of the cis-2-
aza-3-oxabicyclo[3.3.0]octane system, probably due to the an-
omeric effect.
[20] The relatively difficult oxidation of 4 could be tentatively as-
cribed to the endo configuration of the C-3 hydroxy.
[21] X. Lu, L. Deng, Angew. Chem. Int. Ed. 2008, 47, 7824–7827.
[22] S. B. Tsogoeva, S. Wei, Chem. Commun. 2006, 1451–1453.
[23] For a theoretical study on the mechanism of the diphenylproli-
nol-trimethlsilyl-ether-catalysed aza-Michael reactions of α,β-
unsaturated aldehydes, see: C. T. Wong, Tetrahedron 2010, 66,
8267–8272.
[24] The enantiomeric purity of the cis isomers could not be reliably
determined because of incomplete peak separation for the two
enantiomers in the HPLC trace, but in the case of 8b, a maxi-
mum enantiomeric ratio of 64:36 can be safely assumed.
[25] For asymmetric approaches to these α-alkyl β-amino acids, see,
inter alia: a) G. Cardillo, A. Tolomelli, C. Tomasini, J. Org.
Chem. 1996, 61, 8651–8654; b) M. P. Sibi, N. Prabagaran, S. G.
Ghorpade, C. P. Jasperse, J. Am. Chem. Soc. 2003, 125, 11796–
11797.
[26] For reviews dealing with the organocatalytic synthesis of drugs
and of bioactive natural products, see: a) R. M. de Figueiredo,
M. Christmann, Eur. J. Org. Chem. 2007, 2575–2600; b) C.
Grondal, M. Jeanty, D. Enders, Nature Chem. 2010, 2, 167–
178; c) E. Marqués-López, R. P. Herrera, M. Christmann, Nat.
Prod. Rep. 2010, 27, 1138–1176; d) C. A. Busacca, D. R. Fand-
rick, J. J. Song, C. H. Senanayake, Adv. Synth. Catal. 2011, 353,
1825–1864; e) M. Bhanushali, C.-G. Zhao, Synthesis 2011,
1815–1830; f) O. V. Maltsev, I. P. Beletskaya, S. G. Zlotin, Russ.
Chem. Rev. 2011, 80, 1067–1113.
[27] A. Erkkilä, P. M. Pihko, Eur. J. Org. Chem. 2007, 4205–4216.
Received: February 5, 2013
Published Online: April 4, 2013
Eur. J. Org. Chem. 2013, 3103–3111
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3111