Asymmetric aza-Michael addition under ultra-high pressure: Short bias to polyhydroxylated piperidines

Article abstract of DOI:10.1039/c1gc15097a

Two polyhydroxylated piperidines have been prepared in short sequences from diacetone gluco- and allofuranose. The key step is a piezo-aza-Michael addition of diphenylmethanamine to enoates bearing a sugar moiety in the γ-position. The combination of ultra-high pressure associated to the presence of readily available sugars chiral pool led to the expected chiral amines in good yields and excellent stereoselectivities.

Full text of DOI:10.1039/c1gc15097a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 1812
www.rsc.org/greenchem
PAPER
Asymmetric aza-Michael addition under ultra-high pressure: short bias to
polyhydroxylated piperidines†
Sidnei Moura,‡a Christine Thomassigny,*b Caroline Ligeour,b Christine Greck,b Delphine Joseph,a
Emmanuelle Dre`gea and Franc¸oise Dumas*a
Received 24th January 2011, Accepted 29th March 2011
DOI: 10.1039/c1gc15097a
Two polyhydroxylated piperidines have been prepared in short sequences from diacetone gluco-
and allofuranose. The key step is a piezo-aza-Michael addition of diphenylmethanamine to
enoates bearing a sugar moiety in the g-position. The combination of ultra-high pressure
associated to the presence of readily available sugars chiral pool led to the expected chiral amines
in good yields and excellent stereoselectivities.
Introduction
Chiral amines are present in numerous and important molecules
with therapeutic applications.1 Thus, significant efforts have
been devoted to the development of transformations of in-
Fig. 1 Piperidines 1–3 prepared from commercial sugars.3
expensive prochiral precursors into optically active amines.
While the creation of a C–N bond is well-documented in
the literature,2 controlling the diastereoselectivity of the newly
formed stereocentre still remains a challenge.
One of the supplier team described years ago the total
synthesis of the polyhydroxylated piperidines 1, 2 and 3 from
commercial sugars (Fig. 1).3 In this previous work, the stereocen-
ter bearing the nitrogen atom was generated by an asymmetric
catalytic hydrogenation of a b-ketoester, leading to the two
possible b-hydroxyesters with a diastereomeric ratio (d.r.) up
to 90% depending on the catalyst used. Then substitution of the
activated hydroxyl group by an azide afforded the precursor of
the desired amine. The azacycles 1, 2 and 3 were obtained from
diacetone D-allofuranose (9) or D-glucofuranose (10) in 14 steps
in overall yields of 10, 13 and 6% respectively.
Scheme 1 Retrosynthetic analysis of piperidines 3 and 4.
In the present investigation (Scheme 1), the azacycles 3 and 4
could be obtained from the same sugars through the amines 5
and 6 respectively. The N-bearing stereocenter could be directly
introduced to the corresponding sugar-based enoate 7 and 8
via an aza-Michael reaction under ultra-high pressure (UHP)4
which could shorten considerably the synthesis. However, the
level of stereocontrol remains to be established in this case.
A very highly stereocontrolled conjugate addition of primary
amines to crotonate bearing a chiral auxiliary at the ester group
has been previously reported, producing the corresponding b-
aminoesters in good chemical yield and excellent diastereomeric
excess (d.e.) up to 99% thanks to the UHP activation.5 In this
context, we thought to study the UHP-induced stereocontrol of
conjugate addition of amine from sugar based-enoate 7 and 8,
in which now the chiral element is set at the g-position of the
electron withdrawing group.
aLaboratoire de Pharmacognosie, BioCIS, CNRS UMR 8076, Universite´
Paris Sud, Faculte´ de Pharmacie, 5, rue Jean-Baptiste Cle´ment, F-92296,
Chaˆtenay-Malabry Cedex, France. E-mail: francoise.dumas@u-psud.fr
bInstitut Lavoisier de Versailles, UMR CNRS 8180, Universite´ de
Versailles-St-Quentin-en-Yvelines, 45, Avenue des Etats Unis, 78035,
Versailles Cedex, France. E-mail: thomassi@chimie.uvsq.fr
From a general point of view, the asymmetric aza-Michael
addition is well-used in organic synthesis, as proven by the recent
review of Krishna et al.6 Stereoselectivity can be introduced
via a chiral amine or lithium amide, a chiral catalyst, a chiral
† Electronic supplementary information (ESI) available: See DOI:
10.1039/c1gc15097a
‡ Present address: University of Caxias do Sul, Institute of Biotechnol-
ogy, R. Francisco G. Vargas 1130, Caxias do Sul, RS 95001-970, Brazil.
1812 | Green Chem., 2011, 13, 1812–1818
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
auxiliary or a chiral substrate. In the last case, it was reported
that the chirality in the g position of the enoate induced generally
good syn-selectivities.7 Enoates bearing sugars have been used as
aza-Michael acceptors,8,9 leading to the corresponding amines
with good reported d.e. Notably, Dhavale et al.8 studied well the
reaction conditions of unsaturated ethyl enoate esters bearing
a 3-benzylated glucose-derived sugar: when the reaction was
Results and discussion
The aza-Michael addition under high-pressure conditions has
been tested recently. It has been demonstrated that weakly
reactive amides, ureas and Michael acceptors could react by
combination between Brønsted acid in pressures of 0.6 GPa
with good results.14 The combined effects of high pressures
and alcohols in Aza-Michael addition reactions allow the
synthesis of dense adducts bearing a nitrogen-tetrasubstituted
centre adjacent to a quaternary carbon centre (QCC).15 On the
contrary, to our knowledge, no reports on the high-pressure
promoted asymmetric aza-Michael addition appeared recently.
Optimized reaction conditions were established for the synthe-
sis of a,b-systems based in sugars with configuration cis or trans.
With the aim to study the influence of the enoate stereochemistry
in the Michael addition under high pressure, we decided to
prepare the cis and trans-isomers of each acceptor. One of the
easier ways of a,b-unsaturated esters preparation is the well-
known Wittig olefination. Concerning carbohydrate-derived
aldehydes with stabilized ylides, the reaction has been studied
and it is possible to reach preferentially the E or the Z isomer
by changing the reaction conditions.16 In particular, the shortest
way giving the acceptor bearing a xylofuranose moiety from the
commercial diacetone glucose has been described by a sequential
hydrolysis-oxydation-Wittig olefination (SHOWO):16d,f the two
products of the reaction were the expected trans ester and
the lactone proceeding from the cis derivative. We introduced
diacetone allose (9) or diacetone glucose (10) in a similar
sequence, consisting in the hydrolysis in aqueous acetic acid,
the periodic oxidation followed by the olefination in acetonitrile
at reflux during 2 h. In the case of the gluco-derivative (OH-2 and
OH-3 in anti), we obtained the two expected derivatives trans 12
and the lactone 13 (30% and 12% respectively in 3 steps). It is
possible to obtain the cis isomer by making the Wittig olefination
in acetonitrile under reflux during 30 min: we obtained in this
case a cis/trans mixture of 29/71. In the case of the epimer in
C-3 9, the two cis/trans isomers 11 (60/40) were formed with
the good yield of 73% for the overall sequence (Scheme 2).
◦
run in the presence of lithium benzylamide in THF at -40 C,
the conjugate addition occurred in a 85% yield and a complete
stereocontrol (d.e. 100%). Later, Sharma9 applied the same
conditions to the analogous 3-methoxy sugar and obtained as a
major product the transamidated one. He proposed the efficient
action of tetra-n-butylammonium fluoride (TBAF) which avoid
the 1,2-addition in favor of the 1,4-one. The authors described
an efficient method (around 75% of yield) with d.e. up to
90%. However, when we tested these conditions with a similar
substrate (methyl ester instead of ethyl ester, and 3-OBn instead
of 3-OMe), we obtained only 5% of selectivity for a yield of
35%. Unfortunately, changing the reaction time or solvent didn’t
enhance the results. We thus turned to a method based on the use
of ultra-high pressure, which could circumvent these problems of
substrate-dependency for the asymmetric aza-Michael addition.
Recently, alternative energies like microwave10 and
ultrasound11 associated with the use of catalysts and/or
solvent-free conditions have been developed, in agreement with
the tendency and necessity of environmental friendly approach
in organic chemistry. In this context, high-pressure techniques
(0.1–2 GPa) have also been used in green organic synthesis
with good results.12 A great number of reactions that do not
happen in standard conditions can be induced under these
conditions.13 It is a valuable and efficient tool, generally used
for stimulating reluctant organic reactions like those that take
place too slowly, require substantially elevated temperatures
or are hindered by steric or electronic factors. As a result the
application of high pressure favours, for example, reactions
in which the molecularity decreases in the products (e.g.
cycloadditions and condensations) and, in general, reactions
characterized by a negative activation volume. Moreover, as a
thermodynamical parameter, pressure modifies the physical and
physicochemical properties of liquids considered as media for
organic synthesis and most importantly affects solute-medium
interactions associated with volume changes through two main
pressure-sensitive interactions: electrostatic and solvophobic
ones.
Scheme 2 Synthesis of 3-OH sugar-based a,b-systems.
In our case, we assumed that the sugar substituent in g-
position of the a,b-unsaturated ester could induce an important
diastereoselectivity in the piezo-aza-Michael reaction. Advan-
tages are numerous: no need for the introduction of a chiral
auxiliary meaning a gain of steps, energy and materials; waste
reduction; use of encumbered amines to avoid transamidation
side-reaction; no tedious purification needed as one only product
is formed.
We described in this paper the preparation of several un-
saturated methyl esters derived from sugars, then their intro-
duction in asymmetric piezo-aza-Michael additions. Two of
the resulting chiral amines have been shortly transformed into
their corresponding polyhydroxylated piperidines, proving by
the way the absolute configuration of the newly created stereo-
center.
We introduced first the cis/trans mixture of the derivatives
11 or 12 in the piezo aza-Michael reaction in the presence
of 1.2 equivalents of diphenylmethanamine, 3.0 equivalents of
methanol in tetrahydrofuran, during 24 h at room temperature
and over 1.2 GPa (Table 1; Scheme 3). In both cases, the reaction
occurred and a major product was formed, whose NMR spectral
data were in accordance with the ones of the aminated 14 (63%)
or 15 (73%) respectively, and their epimers in C-5 as minor
product, with the d.r. of 80/20 for 14 and 78/22 for 15. We
assumed that the major diastereoisomer was in the configuration
5S,17 based in the works of Sharma9 and Dhavale8 with the
same substrate. We later proved this configuration. Satisfyingly,
applying the same conditions to the cis isomer 11 led to the single
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1812–1818 | 1813
©

View Article Online
Table 1
RT
40 ◦
C
Substrate
Product
d.r. (yield %)
d.r. (yield %)
11cis/trans
11cis
12cis/trans
12cis
14/(5-epi)-14
14/(5-epi)-14
15/(5-epi)-15
15/(5-epi)-15
80/20 (63)
95/5 (nd)
78/22 (73)
—
—
95/5 (82)
95/5 (92)
nd: yield not determined
Scheme 5 Conformations for the Piezo-Aza-Michael exemplified with
trans enoates derivatives 11.
conformations E and F, by its approach near the less hindered
side, leading thus to the corresponding Michael adducts with
lower d.r.
In these models, little or no influence of the 3-OR group
can be expected, in contrast to related Michael addition
conducted at atmospheric pressure,8,9 and in concordance with
our experimental results.
Scheme 3 Piezo-Aza-Michael with derivatives 11 and 12.
isomer 14 (from NMR spectral data; d.r. > 95/5). By heating
◦
at 40 C, the reactions were complete after 24 h for 11cis and
12cis, with the excellent d.r. of 95/5 and excellent yields after
purification.
The introduction of the lactone 13 under the same conditions
led to a unique product (d.r. > 95/5) and some initial material
remaining (<10%). NMR studies of the product clearly showed
the presence of two benzhydrylamino groups, and no more signal
of the methyl ester. These facts were in concordance with the
structure of derivative 16, belonging from the opening of the
lactone into the amide, and the expected aza-Michael addition
(Scheme 6). The yield after purification by flash chromatography
was 58%, which was the maximum possible with the introduction
of the 1.2 equiv. of amine. Using 2 equiv. of amine allowed to
obtain the amine 16 in a yield of 90% and a similar d.r.
The high stereocontrol observed with the cis isomer could be
explained with a Felkin–Ahn model of the transition state as
described by Dhavale.8 It has been demonstrated that, under
UHP, the ester oxygen atoms of the chiral crotonate are H-
bonded to the methanol molecules and that the amine addition
occurs in an anti-fashion relative to the electron withdrawing
group, through an extended s-trans conformation.5c It seems
likely that the sugar-based enoates also are H-bonded with
methanol under pressure, at the both carbomethoxyl group
and sugar part. The cis enoate isomers should exist mostly
in the form of their s-trans (extended) conformations due to
steric hindrance (Scheme 4). Thus, in the Felkin-Ahn models
exemplified for compound 11cis in its s-trans conformations (A
and B), it appears that the more favourable transition structure
will resemble A: a Si approach of the amine, in which steric
interactions would be minimized, would lead almost exclusively
(d.r. 95/5) to the diastereomer having the 5S configuration (talo
configuration 14).
Scheme 6 Piezo-aza-Michael with lactone 13.
Whatever the order of events in this tandem “aza-Michael
addition/lactone opening” or vice versa, examination of the
amine pseudo-axial attack onto the two a,b-unsaturated lactone
13 twisted conformations G and H provide a convergent
explanation of the preferred Si attack leading preferentially to
the 5S isomers (Scheme 7).
Scheme 7 Preferred axial attacks for the Piezo-Aza-Michael on the two
conformations of lactone 13.
Scheme 4 Conformations for the Piezo-Aza-Michael exemplified with
cis enoates derivatives 11.
The diacetone allofuranose (9) and glucofuranose (10) are
also the precursors for the ethylenic esters protected in C-3 with
a benzyloxy or a tert-butyldimethylsilyloxy group 19–22. It is
known that the nature of solvent in the Wittig reaction influenced
the cis/trans ratio of the final product: protic polar solvent
In comparison, the trans sugar-based enoates could present
both s-trans (extended, C and D) and s-cis (compact, E and
F) conformations (Scheme 5). The amine attack appears more
favourable on conformations C, as precedently, and also on
1814 | Green Chem., 2011, 13, 1812–1818
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 2
Table 4
Solvent
T/◦C
Time (h)
19 cis/trans
Yield % (from 17)
Substrate (%) MeOH
T ◦
C
Products
d.r.
Yield
MeOH
MeOH
MeOH
MeOH
CH3CN
CH3CN
Toluene
Toluene
Toluene
-10
0
24
1
24
24
3
2
2
5.0/1
9.0/1
6.5/1
2.7/1
1.5/1
2.2/1
0.6/1
0.8/1
1.2/1
47
55
52
59
46
53
73
69
48
19trans
19cis
3.0 equiv.
3.0 equiv.
—
3.0 equiv.
3.0 equiv.
3.0 equiv.
3.0 equiv.
3.0 equiv.
25
25
40
40
40
25
40
25
23/(5-epi)-23 65/35
23/(5-epi)-23 87/13
23/(5-epi)-23 70/30
23/(5-epi)-23 80/20
24/(5-epi)-24 71/29
24/(5-epi)-24 93/7
25/(5-epi)-25 55/45
26/(5-epi)-26 79/21
90
80
65
>99
95
92
>99
58
25
80
50
80
25
50
110
19cis
19cis
20cis/trans
20cis
21cis
22cis
3
3
60 h under 1.2 GPa at room temperature in the presence of
diphenylmethanamine (1.2 equiv.) and methanol (3.0 equiv.)
in THF. We observed a complete conversion with loss of the
tert-butyldimethylsilyl moiety when this one was used as the
O-protecting group (substrates 20 or 22). After 18 h at room
temperature the reaction was not complete as around 15%
of starting material remained. The best results were generally
obtained at 40 ◦C for 24 h. In this case, we observed the
complete addition reaction and no undesired transformations.17
Best results are described in the Table 4 (Scheme 9).
gives preferentially the cis derivative. Anyway, the reaction
stereospecificity was sometimes to be very substrate dependant.
We tested first the reaction starting from the diacetone glucose
protected at position 3 by a benzyl group 17 (Scheme 8). A
SHOWO using the method of Sartillo–Piscil16d,f or the previously
described one (AcOH–H2O 75/25 then NaIO4 then ylide) led to
the xylose derivative 19 cis or trans (Table 2).
Scheme 8 Synthesis of 3-O-protected xylose based a,b-systems.
As expected, the use of methanol was in favour of the
cis isomer. Nevertheless, the best ratio was obtained at 0 ◦C
(cis/trans : 9/1), and a decrease of the temperature at -10 ◦C
following the protocol of Valverde16b didn’t ameliorate the ratio.
In acetonitrile or toluene, the ratio of cis product increased
with the temperature, but the best one remained very modest
(in refluxing acetonitrile: cis/trans 2.2/1). The use of toluene
at room temperature was preferred for obtain the trans isomer
(cis/trans 0.6/1). The cis/trans isomers were easily separated by
flash chromatography. Then, we prepared the acceptors 20–22
(Fig. 2) with similar methods, respectively from 18, (3-epi)-17
and (3-epi)-18 (Table 3). According to the solvent and reaction
temperature, we reached preferentially one isomer or the other.
Scheme 9 Piezo-aza-Michael with substrates 19–22.
The benzylated derivatives 19trans and 19cis have been
submitted separately to the piezo aza-Michael addition. At room
temperature, reactions were not complete but the same product
was obtained as the major stereoisomer (23/(5-epi)-23: 65/35
and 87/13, respectively).18 The best excess was observed from
the cis derivative, as previously (Tables 1 and 4). To establish
the importance of the 3.0 equivalents of methanol,5b we ran
the reaction in absence of◦this solvent: the reaction was then
complete after 24 h at 40 C under 1.2 GPa, but only 65% of
the product was isolated with a 23/(5-epi)-23 ratio lower than
the one obtained in the presence of methanol (70/30). The other
product isolated (35%) was the 19trans isomer, whose formation
could be explained by a reaction of aza-Michael followed by a
retro aza-Michael, giving the more stable isomer. The methanol
is then proved to be very important for the reaction: the reaction
was complete at 40 ◦C with 3.0 equivalents of methanol, with a
d.r. of 80/20.
Fig. 2 3-a-OH enoates 21 and 22.
In order to obtain a complete aza-Michael addition at
1.2 GPa, we analyzed the optimized time and temperature
required. In our first experiments the reactions were kept for
The 3-epimer 21cis and the silylated derivatives 20cis and 22cis
have been introduced in reaction: the corresponding amines were
obtained in a range of d.e. from 10 to 86%.
Table 3
Finally, two of these amines, namely 23 and 14, have been
transformed into the corresponding piperidines. The reduction
of 23 by lithium aluminium tetrahydride led to the primary
alcohol 27 in a yield of 64% (Scheme 10). A selective hydrolysis
of the cyclic ketal followed by the amine deprotection led to the
benzylated azacycle intermediate, which has been directely hy-
drogenolysed in acidic media. A purification through amberlyst
(OH-) afforded the expected polyhydroxylated piperidine 4 (70%
MeOH, 0 ◦C, 1 h
Toluene, 25 ◦C, 2 h
Substrate
Product
cis/trans
cis/trans
18
20
21
22
4.9/1
3.0/1
4.9/1
0.6/1
0.7/1
0.6/1
(3-epi)-17
(3-epi)-18
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1812–1818 | 1815
©

View Article Online
NMR spectra were recorded on a Bruker AC 200, Avance
300 or ARX 400 apparatus with chemical shift values (d)
in ppm downfield from tetramethylsilane. Mass spectra were
recorded on a XEVO Q-TOF Waters apparatus. All products
gave consistent physicochemical data.
Typical procedure for the piezo aza-Michael: To a solution
of enoate (0.25 mmol) in THF (0.5 mL) was added diphenyl-
methanamine (0.25 mmol, 46 mg, 1.0 equiv.) and methanol
(23 mL; 3.0 equiv.). The solution was introduced with a syringe
in a teflon or a pyrex glass cell. The cell was immersed into
hexane which was contained in the high pressure apparatus.
Then the mobile piston was inserted and the whole assembly
was placed between the pistons of a hydraulic press. The
reaction was performed at 1.2 GPa and 40 ◦C for 24 h.
After decompression the solvent was removed and the crude
was analyzed by NMR spectroscopy. The crude was analyzed
by NMR spectroscopy to check the diastereomeric ratio and
was then quickly purified by silica gel chromatography using
mixtures of cyclohexane/AcOEt and 5% of triethylamine to
afford the expected product. The amines were not stable enough
to obtain satisfactory elemental analysis.
Scheme 10 Synthesis of the piperidine 4.
in three steps). The physical data of the final product were in
accordance with the literature ones.19
We obtained the piperidine 3 in a very short way from 14
(Scheme 11): the reduction of the ester function gave 28 with a
satisfying yield (57%). A two-step sequence, namely the acidic
hydrolysis of the cyclic ketal followed by cyclising hydrogenolysis
by treatment with PtO2 under hydrogen atmosphere at 6 bars,
ending with a purification over amberlyst (OH-) gave 3 (65% in
2 steps).
Methyl 5-(N-diphenylmethylamino)-5,6-dideoxy-1,2-O-
isopropylidene-b-L-talo-heptofuranoronate (14)
Scheme 11 Synthesis of the piperidine 3.
NMR 1H: dH (200 MHz, CDCl3) 7.41–7.24 (10 H, m, Ar), 5.75
(1 H, d, J 3.8, 1-H), 5.07 (1 H, s, CHPh2), 4.57 (1 H, t, J 3.9,
2-H), 4.23 (1 H, dd, J 5.0, J 8.6, 4-H), 3.91 (1 H, dd, J 3.9, J
8.6, 3-H), 3.64 (3 H, s, OMe), 3.24 (1 H, m, 5-H), 2.67 (2 H, m,
6-H and 6¢-H), 2.34 (2H, bs, OH+NH), 1.56 (3 H, s, Me), 1.38
(3 H, s, Me). 13C: dC (100 MHz, CDCl3) 172.6 (C O), 144.7
and 144.4 (Ar), 128.9–127.1 (Ar), 112.8 (CMe2), 103.9 (1-C),
82.1 (4-C), 79.2 (2-C), 72.0 (3-C), 64.3 (CHPh2), 51.9 and 51.7
(5-C and OCH3), 36.0 (6-C), 26.8 (Me), 26.7 (Me); HRMS: m/z
We have prepared the piperidines 3 and 4 from the commercial
sugars with short sequences, with very good d.e. and global
yields of 18% and 22% respectively. We showed that the aza-
Michael addition under high pressure could induce very good
diastereoselectivities for their formation. At last, a comparison
of the physical data of the obtained piperidines with the ones
of the literature allowed us to confirm the configuration of the
major amine as 5S as proposed on examination of Michael
adducts NMR data.
[M+H]+ calctd for C24H30NO6 : 428.199; found: 428.2070.
+
Methyl 5-(N-diphenylmethylamino)-5,6-dideoxy-1,2-O-
isopropylidene-b-L-ido-heptofuranoronate (15)
Conclusion
NMR 1H: dH (300 MHz, CDCl3) 7.39–7.19 (10 H, m, Ar), 5.92
(1H, d, J 3.6, 1-H), 5.05 (1 H, s, CHPh2), 4.48 (1H, t, J 3.6,
2-H), 4.22 (1H, d, J 2.6, 3-H), 4.12 (1H, t, J 2.6, 4-H), 3.65
(3H, s, OMe), 3.54 (1H, m, 5-H), 2.66 (2 H, m, 6-H and 6¢-H)),
2.12 (2H, bs, OH+NH), 1.33 (3 H, s, Me), 1.30 (3 H, s, Me).
13C: dC (100 MHz, CDCl3) 172.2 (C O), 142.8 and 142.7 (Ar),
128.9–127.3 (Ar), 111.6 (CMe2), 104.9 (1-C), 85.3 (2-C), 81.4
(4-C), 75.9 (3-C), 64.0 (CHPh2), 52.4 and 51.8 (5-C and OMe),
36.8 (6-C), 26.9 (Me), 26.3 (Me); HRMS: m/z [M+H]+ calctd
The use of ultra high pressure applied to aza-Michael additions
is an efficient method for the asymmetric formation of a new C–
N bond. With enoates bearing chiral encumbered g-substituents
such as sugars, d.r. are up to 95/5 whatever the 3-OR substituent
in the pyranose derivative. This step is a very good alternative
way to larger syntheses, as has been shown with the application
of the synthesis of two polyhydroxylated piperidines in very short
ways from commercial cheap sugars.
+
for C24H30NO6 : 428.199; found: 428.2096.
Experimental
Solvents were distilled according to Purification of Laboratory
Chemicals, 4th Ed., W. L. F. Aramarego and D. D. Perrin,
Butterwoth Heinemann, 1996. Flash chromatography was per-
formed on silica gel chromagel 60 ACC 35–70 mm. Analytical
TLC was performed using aluminium-backed silica gel Merck
60 F254 and visualised by UV radiation (254 nm) and/or a
solution of phosphomolybdic acid in MeOH and heating and/or
an ethanolic solution of H2SO4 and heating. Optical rotations
were measured at 25 ◦C on a Perkin–Elmer 241 polarimeter
(1 dm cell) using a sodium lamp as the light source (589 nm).
Methyl 3-O-benzyl-5-(N-diphenylmethylamino)-5,6-dideoxy-
1,2-O-isopropylidene-b-L-ido-heptofuranoronate (23)
NMR 1H: dH (200 MHz, CDCl3) 7.35–7.20 (15 H, m, Ar), 5.96
(1 H, d, J 3.9, 1-H), 5.14 (1 H, s, CHPh2), 4.65 (2 H, m, 2-H,
OCHPh), 4.39 (1 H, d, J 12.0, OCHPh), 4.28 (1 H, dd, J 3.2, J
8.4, 4-H), 3.95 (1 H, d, J 3.2, 3-H), 3.58 (3 H, s, OMe), 3.47 (1 H,
m, 5-H), 2.50 (1 H, dd, J 4.8, J 15.0, 6-H), 2.28 (1 H, dd, J 6.2, J
15.0, 6¢-H), 1.49 (3 H, s, Me), 1.33 (3 H, s, Me); NMR 13C: dC (75
MHz, CDCl3) 172.5 (C O), 144.4 and 144.0 (Ar), 128.5–126.6
1816 | Green Chem., 2011, 13, 1812–1818
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
(Ar), 111.5 (CMe2), 104.9 (1-C), 82.6, 81.9 and 81.7 (2-C, 3-C
and 4-C), 71.3 (CH2Ph), 64.6 (5-C), 51.8 and 51.5 (OMe and
CHPh2), 36.8 (6-C), 26.8 (Me), 26.3 (Me); HRMS: m/z [M+H]+
flash chromatography (AcOEt/cyclohexane: 1/5 then 1/1) gave
pure 27 (124 mg); yield 64%; [a]2D5 -80 (c 0.02, CH2Cl2); NMR
1H: dH (200 MHz, CDCl3) 7.31–7.22 (15 H, m, Ar), 7.07 (1 H, s,
NH), 5.99 (1 H, d, J 3.9, 1-H), 5.38 (1 H, s, CHPh2), 4.67 (1 H,
m, OCH), 4.61 (1 H, m, 2-H), 4.31 (1 H, d, J 11.9, OCH’), 4.19
(1 H, dd, J 2.8, J 9.1, 4-H), 3.81 (1 H, d, J 2.9, 3-H), 3.76 (1 H,
m, 7-H), 3.63 (1 H, m, 7¢-H), 3.33 (1 H, m, 5-H), 3.25 (1 H, s,
OH), 1.56 (1 H, m, 6-H), 1.51 (3 H, s, Me), 1.44 (1 H, m, 6¢-H),
1.34 (3 H, s, Me); NMR 13C: dC(100 MHz, CDCl3) 136.2–135.4
(Ar), 129.4–126.7 (Ar), 112.9 (CMe2), 105.3 (1-C), 80.7 and 80.3
(2-C and 4-C), 71.3 (CH2Ph), 68.0 (5-C), 65.1 (5-C), 57.3 (7-C),
57.3 (CHPh2), 28.5 (6-C), 26.4 (Me), 26.2 (Me); HRMS: m/z
+
calctd for C31H36NO6 : 518.246; found: 518.2552.
Methyl 3-O-tertbutyldimethylsilyl-5-(N-diphenylmethylamino)-
5,6-dideoxy-1,2-O-isopropylidene-b-L-ido-
heptofuranoronate (24)
NMR 1H: dH (300 MHz, CDCl3) 7.11-7.36 (10 H, m, Ar), 5.86
(1 H, d, J 3.6, 1-H), 5.17 (1 H, s, CHPh2), 4.29 (1 H, dd, J 3.6,
2-H), 4.21 (1 H, dd, J 2.7, J 9.0, 4-H), 4.13 (1 H, dd, J 2.7, 3-H),
3.58 (3 H, s, OMe), 3.30 (1 H, m, 5-H), 2.50 (1 H, dd, J 4.5, J
14.7, 6-H), 2.22 (1 H, dd, J 12.5, J 20.7, 6¢-H), 1.44 (3 H, s, Me),
1.28 (3 H, s, Me), 0.69 (9 H, s, tBuSi), -0.06 (3 H, s, SiMe), - 0.07
(3 H, s, SiMe); NMR 13C: dC (75 MHz, CDCl3) 172.1 (C O),
144.2 and 143.8 (Ar), 128.3–126.3 (Ar), 111.5 (CMe2), 104.7 (1-
C), 85.2 and 83.4 (2-C and 4-C), 75.8 (3-C), 64.5 (5-C), 51.4 and
50.9 (OMe and CHPh2), 36.6 (6-C), 26.8 (Me), 26.3 (Me), 25.5
(CMe3), -4.2 (SiMe), -5.3 (SiMe).
[M+H]+ calctd for C30H36NO5 : 490.252; found: 490.2587.
+
1,5,6-Trideoxy-1,5-imino-L-ido-heptitol (4)
The derivative 27 (100 mg) was stirred in
a mixture
TFA/H2O : 3/2 (1.5 mL) at 40 ◦C during 1.5 h. After concentra-
tion under reduced pressure, the overall was dissolved in MeOH
(4.5 mL) and PtO2 (10%) was added. A vigorous stirring under
hydrogen atmosphere at 600 kPa during 12 h, followed by a
filtration over celite and an evaporation of the solvent under
reduced pressure gave the intermediate benzylated piperidine.
The later was retaken in EtOH (3 mL) and 6 drops of aqueous
HCl (37%) and catalytic Pd/C were added. The overall was
stirred under hydrogen atmosphere at 600 kPa during 48 h.
After filtration over celite and concentration, a purification over
amberlyst (OH-) gave the pure 4 (24 mg); yield 70%; [a]2D5 -10.1
(c 1.8, MeOH), lit.19 [a]D -13.8 (c 0.8, H2O).
Methyl 3-O-benzyl-5-(N-diphenylmethylamino)-5,6-dideoxy-
1,2-O-isopropylidene-b-L-talo-heptofuranoronate (25)
NMR 1H: dH (300 MHz, CDCl3) 7.27–7.05 (15H, m, Ar), 5.89
(1H, d, J 3.9, 1-H), 5.05 (1H, s, CHPh2), 4;51 (2H, CH2Ph),
4.28 (1H, t, J 4.2, 3-H), 4.20 (1H, dd, J 3.0, J 8.1, 2-H), 3;84
(1H, dd, J 3.3, J 8.7, 4-H), 3.47 (3H, s, CO2CH3); 3.40 (1H, m,
5-H), 2.40 (1H, dd, J 4.8, J 15.0, 6-H), 2.23 (1H, dd, J 6.0, J
15.0, 6-H) 1.39 (3H, s), 1.21 (3H, s). NMR 13C: dC (75 MHz,
CDCl3) 172.3 (C O), 144.2 (C Ar), 143.0 (2C Ar), 128.7 to 127.0
(15CAr), 111.4 (CMe2), 104.9 (1-C), 81.6 (2-C), 82.4 (4-C), 71.2
(3-C), 64.5 (CHPh2), 51.7 (OMe), 51.2 (5-C), 36.7 (6-C), 26.8
5-(N-diphenylmethylamino)-5,6-dideoxy-1,2-O-isopro-
pylidene-b-L-talo-hepto-1,4-furanose (28)
(Me), 26.7 (Me); HRMS: m/z [M+H]+ calctd for C31H36NO6 :
+
518.246; found: 518.2562.
The same procedure for obtaining 27 has been applied to 14. Af-
ter purification by flash chromatography (AcOEt/cyclohexane:
1/2 then AcOEt) the product 28 (111 mg) was obtained; yield
57%; [a]2D5 -5.3 (c 1.3, DCM); NMR 1H: dH (200 MHz, CDCl3)
7.42–7.22 (10 H, m, Ar), 5.76 (1 H, s, 1-H), 5.16 (1 H, s, CHPh2),
4.54 (1 H, s, 2-H), 3.98 (2 H, m, 3-H and 4-H), 3.73 (2 H, m,
7-H and 7¢-H), 3.18 (1 H, s, OH), 2.98 (1 H, m, 5-H), 1.89 (1
H, m, 6-H), 1.78 (1 H, m, 6¢-H), 1.58 (3 H, s, Me), 1.37 (3 H,
s, Me); NMR 13C: dC(75 MHz, CDCl3) 143.5 and 143.0 (Ar),
128.8–127.2 (Ar), 112.7 (CMe2), 103.5 (1-C), 81.3 and 78.8 (2-C
and 4-C), 72.5 (3-C), 63.8 (5-C), 60.7 (7-C), 54.5 (CHPh2), 31.1
(6-C), 26.6 (Me), 26.5 (Me); HRMS: m/z [M+H]+ calctd for
Methyl 3-O-tertbutyldimethylsilyl-5-(N-diphenylmethylamino)-
5,6-dideoxy-1,2-O-isopropylidene-b-L-talo-
heptofuranoronate (26)
NMR 1H: dH (300 MHz, CDCl3) 7.12–7.32 (10H, m, Ar), 5.81
(1H, d, J 3.9, 1-H), 5.11 (1H, s, CHPh2), 4.23 (1H, dd, J 3.6,
2-H), 4.13 (1H, dd, J 2.4, J 4.8, 4-H), 4.07 (1H, dd, J 2.7, 3-H),
3.56 (3H, s, CO2CH3); 3.27 (1H, m, 5-H), 2.44 (1H, dd, J 3.6,
J 14.7, 6-H), 2.17 (1H, dd, J 6.6, J 14.7, 6-H) 1.45 (3H, s), 1.25
(3H, s), 0.85 (s, 9H, C(CH3)3), 0.21 (3H, s, SiCH3), 0.10 (s, 3H
SiCH3). NMR 13C: dC (75 MHz, CDCl3) 173.0 (C O), 144.1
and 142.7 (Ar), 129.2–126.3 (Ar), 121.9 (CMe2), 109.3 (1-C),
90.0 (2-C), 87.5 (4-C), 76.3 (3-C), 62.3 (CHPh2), 51.9 (OMe),
47.4 (5-C), 35.6 (6-C), 26.8 (Me), 26.5 (Me), 25.7 (CMe3), -4.9
(SiMe), -5.1 (SiMe).
+
C23H30NO5 : 400.205; found: 400.2128.
1,5,6-Trideoxy-1,5-imino-L-talo-heptitol (3)
The derivative 28 (100 mg) was stirred in a mixture TFA/H2O:
3/2 (1.5 mL) at 40 ◦C during 1.5 h. After concentration under
reduced pressure, the overall was retaken with MeOH (4.5 mL)
and PtO2 (10%) was added. A vigorous stirring under hydrogen
atmosphere at 600 kPa during 12 h, followed by a filtration over
celite and an evaporation of the solvent under reduced pressure
gave a crude (150 mg) whose purification over amberlyst (OH-)
gave the pure 3 (29 mg); yield 67%; [a]2D5 +23 (c 0.8, MeOH),
lit.3b [a]D +24 (c 0.6, MeOH).
3-O-Benzyl-5-(N-diphenylmethylamino)-5,6-dideoxy-1,2-O-
isopropylidene-b-L-ido-hepto-1,4-furanose (27)
To a solution of 23 (200 mg, 0.4 mmol) in dry THF (2 mL) at 0 ◦C
under argon, was added LiAlH4 (16 mg, 0.4 mmol, 1.1 equiv.).
After stirring 2 h, water was added (5 mL) and the reaction
mixture was extracted with AcOEt (4 ¥ 3 mL). The organic layer
was washed with brine (3 mL), dried, filtered and concentrated
under vacuum for give 176 mg of yellow oil. A purification by
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1812–1818 | 1817
©

View Article Online
Yamada, Tetrahedron Lett., 1983, 24, 3009–3012; (c) P. G. Lima, L. C.
Sequeira and P. R. R. Costa, Tetrahedron Lett., 2001, 42, 3525–3527.
8 N. T. Patil, J. N. Tilekar and D. D. Dhavale, J. Org. Chem., 2001, 66,
1065–1074.
9 G. V. M. Sharma, V. G. Reddy, A. S. Chander and K. R. Reddy,
Tetrahedron: Asymm., 2002, 13, 21–24.
10 (a) S. P. Singh, T. V. Kumar, M. Chandrasekharam, L. Giribabu and
P. Y. Reddy, Synth. Commun., 2009, 39, 3982–3989; (b) M. Saikia,
D. Kakati, M. S. Joseph and J. C. Sarma, Lett. Org. Chem., 2009, 6,
654–658.
Acknowledgements
We thank Universud Paris (PRES) for financial support. SM
thanks University Paris Sud DRI for postdoctoral fellowship.
We greatly acknowledge Estelle Morvan (BioCIS, Chaˆternay-
Malabry) for NMR experiments, Karine Leblanc (BioCIS,
Chaˆtenay-Malabry) for performing chiral HPLC and elemen-
tal analysis, Estelle Galmiche (ILV, Versailles) and Miche`le
Danet (SAMM, Chaˆtenay-Malabry) for mass measurements
and Jorge Ballester (ILV, Versailles) and Shengda Chen (BioCIS,
Chaˆtenay-Malabry) for experimental support.
11 Z. Duan, X. Xuan, T. Li, C. Yang and Y. Wu, Tetrahedron Lett.,
2006, 47, 5433–5436.
12 (a) G. Jenner, Tetrahedron, 2002, 58, 5185–5202; (b) C. Camara,
L. Keller, K. Jean-Charles, D. Joseph and F. Dumas, Int. J. High
Pressure Res., 2004, 24, 149–162; (c) K. Matsumoto, Top. Heterocycl.
Chem., 2007, 8, 1–42; (d) L. Minuti, in Eco-Friendly Synthesis of
Fine chemicals, 2009, pp 237–274, ISBN: 978-1-84755-976-0; (e) F.
Benito-Lo´pez, R. J. M. Egberink, D. N. Reinhoudt and W. Verboom,
Tetrahedron, 2008, 64, 10023–10040.
References
1 (a) V. Gududuru, E. Hurh, G. G. Durgam, S. S. Hong, V. M. Sardar,
H. Xu, J. T. Dalton and D. D. Miller, Bioorg. Med. Chem. Lett.,
2004, 14, 4919–4923; (b) D. X. Sun, L. Liu, B. Heinz, A. Kolykhalov,
J. Lamar, R. B. Johnson, Q. M. Wang, Y. Yip and S.-H. Chen,
Bioorg. Med. Chem. Lett., 2004, 14, 4333–4338; (c) M. Guzma´n,
J. Lo Verme, J. Fu, F. Oveisi, C. Bla´zquez and D. Piomelli, J. Biol.
Chem., 2004, 279, 27849–27854; B. Ernst and J. L. Magnani, Nat.
Rev. Drug Discovery, 2009, 8, 661–677.
13 See for example: C. Me´nard-Moyon, F. Dumas, E. Doris and C.
Mioskowski, J. Am. Chem. Soc., 2006, 128, 14764–14765.
14 S. Azad, T. Kobayashi, K. Nakano, Y. Ichikawa and H. Kotsuki,
Tetrahedron Lett., 2009, 50, 48–50.
15 A. Y. Rulev, S. Azad, H. Kotsuki and J. Maddaluno, Eur. J. Org.
Chem., 2010, 6423–6429.
16 As examples, see: (a) E. J. Corey and G. Goto, Tetrahedron
Lett., 1980, 21, 3463–3466; (b) S. Valverde, M. Martin-Lomas,
B. Herradon and S. Garcia-Ochoa, Tetrahedron, 1987, 43, 1895–
1901; (c) C. Harcken and S. F. Martin, Org. Lett., 2001, 3, 3591–
3593; (d) F. Sartillo-Piscil, M. Vargas, C. A. de Parrodi and L.
Quintero, Tetrahedron Lett., 2003, 44, 3919–3921; (e) L. V. R.
Reddy, G. N. Swamy and A. K. Sham, Tetrahedron: Asymm., 2008,
19, 1372–1375; (f) E. Ramirez, M. Sanchez, R. L. Meza-Leon, L.
Quintero and F. Sartillo-Piscil, Tetrahedron Lett., 2010, 51, 2178–
2180.
17 Chiralcel-OD HPLC (UV 254, isopropanol/hexane (15/85))
was performed for determination of diastereoisomeric
ratio.
18 The reaction of enoate 19 with diphenylmethanamine in the presence
of TBAF at ambient pressure,9 led to the same major diastereomer.
19 N. N. Saha, V. N. Desai and D. D. Dhavale, Tetrahedron, 2001, 57,
39–46.
2 See for example: R. C. Larock, in Comprehensive Organic Transfor-
mations, VCH Publishers, USA, 1989, pp 389–400 and pp 417–435.
3 (a) C. Thomassigny, M. T. Barroso and C. Greck, Lett. Org. Chem.,
2005, 2, 316–138; (b) G. Le Bouc, C. Thomassigny and C. Greck,
Tetrahedron: Asymm., 2006, 17, 2006–2014.
4 High pressure means in general pressure up to 200 bar (hydrogena-
tions, carbonylations, etc.) while useful UHP scale is 0.2 to 2 GPa (2
to 20 kbar) for organic reactions.
5 (a) J. d’Angelo and J. Maddaluno, J. Am. Chem. Soc., 1986, 108,
8112–8114; (b) F. Dumas, B. Mezrhab and J. d’Angelo, J. Org. Chem.,
1996, 61, 2293–2304; (c) F. Dumas, C. Fressigne´, J. Langlet and C.
Giessner-Prettre, J. Org. Chem., 1999, 64, 4725–4732.
6 P. R. Krishna, A. Sreeshailam and R. Srinivas, Tetrahedron, 2009,
65, 9657–9672.
7 See e.g. (a) N. Asao, T. Shimada, T. Sudo, N. Tsukada, K. Yazawa,
Y. S. Gyoung, T. Uyehara and Y. Yamamoto, J. Org. Chem., 1997,
62, 6274–6282; (b) H. Matsunaga, T. Sakamaki, H. Nagaoka and Y.
1818 | Green Chem., 2011, 13, 1812–1818
This journal is
The Royal Society of Chemistry 2011
©