Biocatalytic preparation of enantioenriched 3,4-dihydroxypiperidines and theoretical study of Candida antarctica lipase B enantioselectivity

Article abstract of DOI:10.1016/j.tet.2006.01.061

Enzymatic acetylations of N-substituted cis- and trans-3,4- dihydroxypiperidine and hydrolysis of their diacetylated derivatives have been studied. High enantioselectivities are obtained with Pseudomonas cepacia lipase and Candida antarctica lipase B for

Full text of DOI:10.1016/j.tet.2006.01.061

Tetrahedron 62 (2006) 3284–3291
Biocatalytic preparation of enantioenriched
3,4-dihydroxypiperidines and theoretical study of
Candida antarctica lipase B enantioselectivity
*
´ ´
Laura F. Solares, Ivan Lavandera, Vicente Gotor-Fernandez, Rosario Brieva and Vicente Gotor
´
´
´
´
Departamento de Quımica Organica e Inorganica, Facultad de Quımica, Universidad de Oviedo, 33006 Oviedo, Spain
Received 19 November 2005; revised 17 January 2006; accepted 18 January 2006
Available online 17 February 2006
Abstract—Enzymatic acetylations of N-substituted cis- and trans-3,4-dihydroxypiperidine and hydrolysis of their diacetylated derivatives
have been studied. High enantioselectivities are obtained with Pseudomonas cepacia lipase and Candida antarctica lipase B for the
hydrolysis of the trans-derivative, while the cis-derivatives are not adequate substrates in the same biocatalytic conditions. The
enantiopreference of these processes can be rationalized by means of a molecular modelling study.
q 2006 Elsevier Ltd. All rights reserved.
1. Introduction
interesting structure since the (3R,4R)-trans isomer is an
inhibitor of b-D-glucoronidase5 and has been used as
intermediate for the preparation of a xylanase inhibitor
(Fig. 1).6 Also, this structure is a suitable precursor of
the gastroprokinetic agent cisapride.7 On the other hand, the
cis-isomer is an intermediate for the preparation of the
antidepressant ifoxetine.8
Polyhydroxylated piperidines have attracted great interest
due to their biological properties. Some of them are
inhibitors of glycosidases and glycoprotein-processing
enzymes.1 These inhibitors have shown a promising
therapeutical potential for the treatment of various diseases
related to metabolic disorders of carbohydrates such as
diabetes,2 and viral infections including AIDS3 or cancer.4
In this sense, optically pure 3,4-dihydroxypiperidine is an
The optically pure (3R,4R)-trans-isomer has been
synthesized from either D-arabinose5 or D-tartaric acid.6
NH2
Me
Cl
HO
Me
O
HO
Me
NH
OMe
OH
O
OH
HN
O
xylanase inhibitor
MeO
N
H
N
R
Ifoxetine
R = H norcisapride
O
R =
cisapride
F
Figure 1. Examples of different chiral 3,4-dihydroxypiperidine derivatives.
Keywords: Camdida antarctica lipase B; Dihydroxypiperidine; Enzymatic hydrolysis and transesterification; Molecular modeling.
*
Corresponding author. Tel./fax: C34 985 103 448; e-mail: vgs@fq.uniovi.es
0040–4020/$ - see front matter q 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.01.061

L. F. Solares et al. / Tetrahedron 62 (2006) 3284–3291
3285
OAc
O
OAc
Ac2O
CbzCl
m-CPBA
.
NaHCO3
CH2Cl2
CH2Cl2
BF3 Et2O
N
H
N
N
N
AcOH
Cbz
Cbz
Cbz
1
2
( )-3
( )-trans-4
Scheme 1. Synthesis of (G)-trans-4.
Both routes require a large number of steps to afford the
desired compound in low overall yields. Recently, Chang
et al.9 have reported the preparation of the optically pure
trans-isomers using enantioselective trans-dihydroxylations
or epoxidations of N-substituted 1,2,5,6-tetrahydro-
pyridines, catalyzed by the bacterial strain Sphingomonas
sp. HXN-200.
subsequent opening of the resulting epoxide (G)-3 with
acetic anhydride and BF3$Et2O (Scheme 1).
First, we studied the hydrolysis of (G)-trans-4, in aqueous
media (phosphate buffer, pH 7). In these conditions, the
non-enzymatic hydrolysis occurred in competition with
the enzymatic reaction. So, we decided to examine the
process in organic solvents using a small amount of water as
nucleophile (Scheme 2).
On the other hand, lipase-catalyzed processes in organic
solvents are very advantageous alternatives to the conven-
tional chemical reactions, that have been widely used for the
preparation of optically pure cycloalkanediols10 and
carbosugars.11 Nevertheless, to the best of our knowledge,
the use of this methodology for the preparation of
polyhydroxylated piperidines has not been studied until
now. This paper describes the acetylation of N-protected
cis- and trans-3,4-dihydroxypiperidine and the hydrolysis
of their diacetylated derivatives catalyzed by lipases. Some
of the results presented here have been rationalized using a
computational approach.
Since we had previously obtained good results in these
hydrolytic conditions,12 in a first set of experiments at
30 8C, acetonitrile was chosen as the reaction solvent and
different lipases were used as biocatalysts: Candida
antarctica A (CAL-A), C. antarctica B in two forms
[Novozym 435 (CAL-B) and Chirazyme-L2 (CAL-B-L2)],
Candida rugosa (CRL) and Pseudomonas cepacia (PSL-C).
Three of the tested enzymes catalyzed the regioselective
hydrolysis of (G)-trans-4 affording only the 3-acetyl-4-
hydroxy derivative (3R,4R)-5 (Table 1, entries 1–3). In all
cases, the enzymatic reactions were regioselective towards
the acetyl group at O-4 position. This was confirmed by
1
means of a two-dimensional H–1H COSY NMR analysis.
2. Results and discussion
The enantioselectivity of the enzymatic hydrolysis was
also very high. Both immobilized forms of lipase B from
C. antarctica, CAL-B and CAL-B-L2, showed the highest
enantioselectivities (EO200),13 with excellent reaction
rates, after 24 h a 49% conversion was achieved. The lipase
PSL-C also showed a high enantioselectivity under these
conditions, (EZ98), but displayed a lower reaction rate.
In these conditions, CAL-A and CRL did not show any
activity.
Our initial experiments were designed to find the most
suitable lipase for catalyzing the hydrolysis of the
diacetylated derivative (G)-trans-4. This substrate has
been prepared almost in quantitative yield by treatment of
1-benzyloxycarbonyl-1,2,5,6-tetrahydropyridine (2) with
m-chloroperbenzoic acid (m-CPBA) in CH2Cl2 and
OAc
OH
OAc
OAc
OAc
OAc
H2O / Lipase
The absolute configurations of the product and the
remaining substrate were assigned as follows. Hydrolysis
of (K)-trans-5 using NaOMe in MeOH, afforded the
dihydroxy derivative (C)-trans-6, whose specific rotation
sign was in agreement with that reported for (3R,4R)-(C)-6,
[a]2D0 C3.51 (c 0.76, CHCl3).9
+
N
N
N
Organic solvent
Cbz
Cbz
Cbz
( )-trans-4
(3R,4R)-5
(3S,4S)-4
Scheme 2. Enzymatic hydrolysis of (G)-trans-4 in organic solvents.
Table 1. Lipase catalyzed hydrolysis of (G)-trans-4 in organic solvents at 30 8C
Entry
Lipase
Solvent
H2O (equiv)
t (h)
c (%)a
ees (%)b
eep (%)b
Ec
1
2
3
4
5
6
CAL-B
CAL-B-L2
PSL-C
CAL-B
CAL-B
CAL-B
Acetonitrile
Acetonitrile
Acetonitrile
1,4-Dioxane
Toluene
5
5
5
5
5
10
24
24
24
48
24
24
49
49
8
47
10
36
97
96
9
89
9
O99
O99
98
O99
83
O200
O200
98
O200
12
Acetonitrile
54
O99
O200
a Conversion, cZees/(eesCeep).
b Determined by chiral HPLC.
c Enantiomeric ratio, EZln[(1Kc)(1Kees)]/ln[(1Kc)(1Cees)].13

3286
L. F. Solares et al. / Tetrahedron 62 (2006) 3284–3291
OAc
OH
OAc
OH
OAc
OH
OH
+
OH
NaOMe
MeOH
Vinyl acetate
Lipase
Organic solvent
N
N
N
N
Cbz
Cbz
Cbz
Cbz
( )-trans-4
Scheme 3. Enzymatic acetylation of (G)-trans-6.
( )-trans-6
(3R,4R)-7
(3S,4S)-6
Table 2. Lipase catalyzed acetylation of (G)-trans-6 with vinyl acetate (VA) in organic solvents at 30 8C
Entry
Lipase
Solvent
VA (equiv)
t (h)
c (%)a
ees (%)b
eep (%)b
Ec
1
2
3
4
5
6
CAL-B-L2
PSL-C
CAL-B
CAL-B
PSL-C
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
Toluene
5
5
5
50
5
5
15
16
72
22
15
15
59
30
—
30
62
68
20
74
—
19
39
25
14
61
—
13
24
11
2
2
—
9
2
PSL-C
tBuOMe
2
a Conversion, cZees/(eesCeep).
b Determined by chiral HPLC.
c Enantiomeric ratio, EZln[(1Kc)(1Kees)]/ln[(1Kc)(1Cees)].13
In order to improve the performance of the process, we
studied the influence of the organic solvent on the
enantioselectivity of the hydrolysis catalyzed by CAL-B.
Thus, under the same reaction conditions, similar results
were obtained in 1,4-dioxane than in acetonitrile (Table 1,
entry 4), even though it is apparent that in acetonitrile were
slightly higher. Lower reaction rate and enantioselectivity
were achieved in toluene (Table 1, entry 5). Finally, an
increase of the nucleophile concentration up to 10 equiv
furnished lower conversion (Table 1, entry 6).
Attempts to improve the results obtained in the reaction
catalyzed by PSL-C, using different solvents such as toluene
(entry 5) or tert-butyl methyl ether (entry 6) afforded also
poor enantioselectivities, although these processes were
faster.
In order to establish the scope of this enantiomer separation
methodology, similar biocatalytic conditions were applied
for the hydrolysis of (G)-cis-N-benzyloxycarbonyl-3,4-
diacetoxypiperidine derivative. The substrate was syn-
thesized according to the Scheme 4. Treatment of
1-benzyloxycarbonyl-1,2,5,6-tetrahydropyridine (2) with
OsO4 and N-methylmorpholine N-oxide (NMMO) yielded
the diol (G)-cis-8, that was acetylated to the corresponding
diacetate derivative (G)-cis-9.
Another common approach for the resolution of racemic
polyalcohols is the enzymatic transesterification reaction.14
So, we decided to synthesize the dihydroxypiperidine (G)-
trans-6, starting from the diacetate (G)-trans-4 and using
NaOMe in MeOH. Taking into account that the best enzymes
for the hydrolysis are both CAL-B preparations and PSL-C,
we examined the acetylation of the dihydroxy derivative (G)-
trans-6 catalyzed by these lipases (Scheme 3, Table 2).
OH
OAc
OsO4
NMMO
Ac2O
DMAP
OH
OAc
Acetone/H2O
CH2Cl2/Et3N
N
N
N
First we carried out the process using 5 equiv of vinyl
acetate as acyl donor and acetonitrile as solvent. In these
conditions, CAL-B-L2 and PSL-C catalyzed the processes
with good reaction rates (entries 1 and 2), however, both
showed very low enantioselectivities. Surprisingly, in the
process catalyzed by CAL-B, after 72 h of reaction only
the unaltered substrate was recovered (entry 3). A 50-fold
excess of the acylating agent was necessary to obtain a
moderate reaction rate with this catalyst (entry 4). Also a
slight improvement in the enantioselectivity was
observed.
Cbz
Cbz
Cbz
2
( )-cis-8
( )-cis-9
Scheme 4. Synthesis of (G)-cis-8 and (G)-cis-9.
Unfortunately, the enzymatic hydrolysis carried out with
this substrate showed, in all cases, very low reaction rates.
For instance, the conversion of the reaction catalyzed by
CAL-B in acetonitrile at 30 8C, using 5 equiv of water was
less than 5% after 7 days of reaction. An increment of the
temperature at 60 8C or an excess of the nucleophile up to
50-fold did not increase the reaction rate. Other lipases and
different organic solvents were tested but no improvement
on the reaction rate was obtained. Even more, when we
analyzed the products using gas chromatography, we
observed a mixture 1:1 of the two possible regioisomers.
As expected according to the principle of microscopic
reversibility, the regio- and stereochemical preferences of
these lipases were the same as for the hydrolytic processes:
the hydroxyl group at C-4 position of the piperidine ring was
preferentially acetylated to afford the product (3R,4R)-7. In
all cases, the formation of only one of the two possible
regioisomers was observed. Unfortunately, the enantio-
selectivities of the enzymatic acylations were very low.
Similar results were achieved in the acetylation of the
dihydroxyderivative (G)-cis-8. Very low reactivities and

L. F. Solares et al. / Tetrahedron 62 (2006) 3284–3291
3287
regioselectivities were found in all the tested conditions
(data not shown).
[one between Asp187 and Nd-His224; one between
N3-His224 and Og-Ser105; one between N3-His224 and
the proton-acceptor oxygen of the acetate; and three
between the oxygen of the oxyanion formed with Gln106
(one) and Thr40 (two)]; and (b) those that minimized the
steric clashes between the phosphonate and the amino acids,
which surrounded the active site of CAL-B. In general, the
nucleophile pocket restricted the binding of the substrate to
such an extent that only a few conformations were obtained
with the desired properties.
Since all these lipase-catalyzed reactions were dramatically
influenced by the stereochemistry of the chiral centres, we
were very interested to rationalize the experimental facts
observed in the hydrolysis of the trans-derivative (G)-4
using a computational approach. Since the X-ray crystal
structure of CAL-B was resolved,15 several molecular
modeling16 and molecular dynamics studies17 have been
made in order to explain its excellent reactivity and
selectivity with chiral alcohols and amines. These reports
have been very useful to make rational design of this
biocatalyst18 with improved selectivities,17b,19 or catalyzing
through novel reaction pathways.20 Thus, engineered
CAL-B has been utilized to perform aldol,21 Michael-type
additions,22 or Baeyer–Villiger reactions.23
We started with the minimization of the intermediate for the
hydrolysis in the 4-position of (3R,4R)-4. This enantiomer
was experimentally the fast-reacting one. We used the X-ray
structure 1LBS16a obtained from the Protein Data Bank
(http://www.rcsb.org/pdb/), which presented an inhibitor
phosphonate that was manually deleted, and then the
substrate was built.16g
The restricted volume of the active site of this lipase has
been used to rationalize its special properties. So, viewed
with the catalytic triad Asp187-His224-Ser105 oriented
from left to right, CAL-B presents two well-differed
subsites: (a) the large hydrophobic pocket (or acyl pocket)
above the catalytic triad, which is lined by Leu140 and
Leu144 at the top, Ile189 and Val190 on the left, as well as
Val154 on the far right of the pocket. Deep in this subsite,
Asp134 is on the left and Gln157 on the right; and (b) the
medium-sized pocket (so called nucleophile or stereospe-
cificity pocket) below it, which is crowded by Trp104 at the
bottom and the Leu278-Ala287 helix to the right. The acyl
chains lie in the large subsite, while nucleophiles bind to the
medium-sized pocket. This smaller subsite has been found
as the key factor to the enantioselectivity of secondary
alcohols.24
When the bulky benzyloxycarbonyl moiety was manually
added to the phosphonate and minimized, two families of
intermediates were usually acquired. In both cases, this
group was accommodated into the narrow tunnel created by
Ile189, Leu278, Ala281, Ala282, Ile285, and Val286, but
with different orientations. Some of them (usually more
stable), presented this moiety on the left, with the phenyl
group close to Ile189 and Leu278, and the other one had the
phenyl ring on the right side near Ala282, Ile285, and
Val286 pointing to the solvent.
The best structure obtained for (3R,4R)-4 is shown in
Figure 2. As described previously,16b,g the cycle was placed
into the stereospecificity pocket, with the acetyl group in the
3-position fitted into the acyl pocket, generating appropriate
interactions between the methyl group with Val154 and the
carbonyl moiety with the hydrophilic side chains of Gln157
and Thr40. The benzyloxycarbonyl group was on the left of
the tunnel with the phenyl ring making stabilizing
hydrophobic interactions with the side chains of Ile189,
Thus, the fast-reacting enantiomer (following Kazlauskas’
rule,25 usually the R), places the medium-sized substituent
into this subsite and the big one into the acyl pocket, while
the slow enantiomer has to introduce them in the opposite
orientation, existing more unfavorable steric interactions.
This general rule for acyclic alcohols, is not so clear for
the cyclic one, although R-acylated or hydrolyzed
enantiomers are usually obtained, in several cases it has
been observed that for 1,2-cis-cyclohexanols, lipase-
catalyzed processes are not achieved or are not selective,
in spite of the excellent enantioselectivity for their trans
counterparts.26 There are no many examples of the use of
molecular modeling to study CAL-B-catalyzed acylations
or hydrolysis of cyclic alcohols.16c,d,g Here, we will use this
tool to explain qualitatively the experimental results
obtained for the hydrolysis at the 4-position in the cyclic
diacetate (G)-4.
We have modeled the first tetrahedral intermediate (TI-1)
since in the hydrolysis processes this is the structure where
the ester is bound to the lipase, and is responsible for the
enantioselection. Phosphonate analogues were used for
mimicking this intermediate using the AMBER force
field,27 and several local energy minima were obtained for
each enantiomer by means of a systematical search, as
described elsewhere.16g The selection of the best structures
was based in two main criteria: (a) those that presented
the most possible of the six essential hydrogen-bonds
Figure 2. Best intermediate obtained for the CAL-B-catalyzed acylation
of (3R,4R)-4. Amino acids, which make stabilizing interactions with the
3-O-acetyl group of the substrate are depicted in green, and those with the
N-benzyloxycarbonyl moiety are shown in dark blue. The above image
displays Leu144 in a line illustration and Ile189 in a stick representation to
allow a better view of the large pocket of the lipase.

3288
L. F. Solares et al. / Tetrahedron 62 (2006) 3284–3291
Glu188, and Leu278, with the carbonyl group making a
˚
H-bond with the NH-backbone of the Ala282 (3.03 A), and
the methylene moiety interacting with the hydrophobic side
chain of Ala282. All key-hydrogen bonds were present in
this structure.
difference between the first tetrahedral intermediate in the
hydrolysis process, and the second tetrahedral intermediate
in the acetylation reaction, is that in the first case at the
3-position there is an acetoxy group, and in the second one
there is a hydroxyl moiety at this position. Some of the
main structural differences obtained between both enantio-
mers were based on the altered interactions of the 3-O-
acetyl group with the amino acids close to the lipase
active-site. Thus, while for the (R,R) structure they were
stabilizing (see green interactions in Fig. 2), in the (S,S)
enantiomer they destabilized the intermediate (see red
interactions in Fig. 3). Since this moiety is not present in
the acetylation intermediates both structures would present
less differences.
When the intermediate for (3S,4S)-4 was built and
minimized, the best structure obtained presented several
structural differences with regards to the (R,R)-enantiomer
(Fig. 3). Although all essential-hydrogen bonds existed too,
due to the acetyl moiety in 3-position bound into the
medium nucleophile subsite, the carbocycle did not fit in
this pocket, and the Cbz was not placed as stable as in the
case of (3R,4R)-4. On the one hand, many destabilizing
interactions appeared between the acetyl group and Trp104,
Ala281 (both with the carbonyl), Gly39 (with the methyl),
Leu278 and more importantly, with the catalytic His224
(both with the oxygen of the alcohol); on the other hand, the
benzyloxycarbonyl moiety was placed at the top of the
tunnel making unfavorable clashes between the carbonyl
group and the hydrophobic chain of Ile285, and the phenyl
ring with the carbonyl-backbone of Glu188.
3. Conclusions
This paper describes an easy methodology for the
preparation of both enantiomers of trans-1-benzyloxy-
carbonyl-3,4-dihydroxypiperidine and the (R,R)-3-O-mono-
acetylated derivative via a lipase catalyzed hydrolysis.
Thus, among all the lipases tested, CAL-B presented the
best results, allowing us to obtain one of the six possible
products of this reaction (four monoalcohols and two diols)
with high yield, showing this process total regio- and
enantioselectivity at the same time. Attempts to acetylate
the trans-diol (G)-6 afforded good regioselectivity but
lower reaction rates and enantioselectivities.
Molecular modeling has been used in order to rationalize
these experimental facts. Thus, phosphonate analogues of
both enantiomers of diacetate 4 were built and minimized,
obtaining several structural divergences that can explain the
excellent enantioselectivity obtained in the hydrolysis
reaction with this substrate.
On the other hand, from the results shown in this paper, it is
apparent that the cis-isomer is not a suitable substrate for the
tested lipases, neither through hydrolysis nor acylation
processes.
Figure 3. Best intermediate obtained for the CAL-B-catalyzed acylation
of (3S,4S)-4. Amino acids, which make destabilizing interactions with
the 3-O-acetyl group of the substrate are depicted in red and with the
N-benzyloxycarbonyl moiety are shown in violet. The above image
displays Ile189 in a line illustration and Glu188 and Leu278 in a stick
representation to allow a better view of the phosphonate structure.
Taking into account the simplicity and easy scale-up of
lipase catalyzed reactions,28 it is noteworthy the applica-
bility of this method for the preparation of the gastro-
prokinetic agent cisapride and other optically pure
structures with highly interesting biological properties.
Molecular modeling studies agree with the experimental
results, explaining the excellent enantioselectivity showed
by CAL-B towards trans-diacetate 4. Thus, although all
essential hydrogen-bonds were present for both intermedi-
ates, better fitting of the (R,R)-enantiomer into the active site
of the lipase, and more stabilizing interactions with the
amino acids, which are surrounding it can be the key
explanation for this resolution. Furthermore, in the (S,S)-
structure, the acetyl in 3-position seemed to be very close
from the catalytic His224, what might destabilize its
position for the necessary transfer of the proton to the
reactive carbonyl.
4. Experimental
4.1. General remarks
Enzymatic reactions were carried out in a Gallenkamp
incubatory orbital shaker. Immobilized C. antarctica lipase
B, CAL-B Novozym 435 (7300 PLU/g), was a gift from Novo
Nordisk co. C. antarctica lipase B (CAL-B-L2, CHIRA-
ZYME L-2, c-f, C3, R400 U/g), Candida rugosa lipase
(CRL, CHIRAZYME L3, O250 U/mg) and C. antarctica
lipase A (CAL-A, CHIRAZYME L5, 1 kU/g) were supplied
by Roche Molecular Biochemicals. P. cepacia lipase
lyophilized (PSL, 30000 U/g) and immobilized (PSL-C,
783 U/g) are commercialized by Amano Pharmaceuticals.
Looking at these structures, a plausible explanation for
the bad enantioselectivity in the acetylation of the diol
derivative trans-(G)-6 could be established. Thus, the only

L. F. Solares et al. / Tetrahedron 62 (2006) 3284–3291
3289
Chemical reagents were commercialized by Aldrich, Fluka,
Lancaster or Prolabo. Solvents were distilled over an
appropriate desiccant under nitrogen. Flash chromatography
was performed using Merck silica gel 60 (230–400 mesh).
Optical rotations were measured using a Perkin-Elmer 241
Na2SO4, the solvent was removed under reduced pressure
and the crude residue was purified by flash chromatography
on silica gel (hexane/EtOAc 1:1) to afford the product (G)-
3 as a yellow oil (3.8 g, 90%). 1H NMR (CDCl3,
300.13 MHz): d 7.36 (m, 5H), 5.10 (s, 2H), 3.85–3.72 (m,
2H), 3.44 (m, 2H), 3.23 (m, 2H), 2.01 (m, 2H); 13C NMR
(CDCl3, 75.5 MHz): d 155.2 (CO), 136.3 (C), 128.4 (CH),
128.0 (CH), 127.8 (2 CH), 67.4 (CH2), 55.3 (2 CH), 47.5
(CH2), 41.3 (CH2), 21.1 (CH2); IR (CH2Cl2): n 1715, 1428,
1212 cmK1; MS (ESIC, m/z): 234 [(MCH)C, 10%], 256
[(MC Na)C, 100%].
polarimeter and are quoted in units of 10K1 deg cm2 gK1
.
1H, 13C NMR and DEPT spectra were recorded in a Bruker
AC-300, Bruker AC-300 DPX and Bruker NAV-400
spectrometer using CDCl3 as solvent. The chemical shift
values (d) are given in parts per million. Positive
electrospray ionisation (ESIC) was used to record mass
spectra on an Hewlett-Packard 110 LC/MSD Series
spectrometers. The enantiomeric excesses were determined
by chiral HPLC analysis on a Hewlett-Packard 1100, LC
liquid chromatograph, using a CHIRALCEL OD for the
diacetylated derivative 4. The diol 6 and the monoacetylated
products 5 and 7 were first acetylated to convert them into
the diacetate derivative 4 and then analyzed.
4.1.3. Preparation of (G)-trans-3,4-diacetoxy-1-benzyl-
oxycarbonylpiperidine, [(G)-trans-4]. To a solution of
(G)-3 (0.95 g, 4.1 mmol) in acetic acid (35 mL), acetic
anhydride was added (1.6 mL, 17 mmol). Boron trifluoride
etherate (507 mL, 4.1 mmol) was slowly added, and the
mixture was stirred at room temperature for 4 h. An aqueous
saturated solution of NaHCO3 (10 mL) was added, and the
mixture extracted with EtOAc (20 mL). The organic phase
was washed with saturated aqueous Na2CO3 solution
(20 mL) followed by saturated aqueous NaCl solution
(20 mL), and then was dried over Na2SO4. The solvent
was removed under reduced pressure and the crude residue
was purified by flash chromatography on silica gel (hexane/
EtOAc 1:1) to afford the product (G)-trans-4 as a yellow oil
(1.2 g, 90%). 1H NMR (CDCl3, 400.13 MHz, 40 8C): d 7.38
(m, 5H), 5.18 (d, 2H), 4.98 (s, 1H), 4.83 (m, 1H), 3.84 (dd,
Molecular modeling. The program Insight II, version
2000.1, was used for viewing the structures. The geometric
optimizations were performed using Discover, version 2.9.7
(Accelrys, San Diego CA, USA), using the AMBER force
field.27 The distance dependent dielectric constant was set to
4.0 to mimic the electrostatic shielding of the solvent and
the 1–4 van der Waals interactions were scaled to 50%. The
crystal structure for the CAL-B (1LBS16a) was obtained
from the Protein Data Bank (www.rcsb.org/pdb/). Protein
structures in Figures 2 and 3 were generated using PyMOL
0.97.29 The methodology followed for localization of
energy minima structures by means of a systematically
search has been described elsewhere.16g
3
1H, JHHZ3.45, 13.80 Hz), 3.71–3.47 (m, 3H), 2.11–2.01
(m, 1H), 2.10 (s, 3H), 1.97 (s, 3H), 1.70 (m, 1H); 13C NMR
(CDCl3, 100.6 MHz, 40 8C): d 169.7 (CO), 169.5 (CO),
155.1 (CO), 136.2 (C), 128.3 (CH), 127.8 (CH), 127.6 (CH),
68.0 (2 CH), 67.1 (CH2), 44.1 (CH2), 40.1 (CH2), 26.1
(CH2), 20.8 (CH3); IR (CH2Cl2): n 1737, 1712, 1425,
1210 cmK1; MS (ESIC, m/z): 358 [(MC Na)C, 100%].
4.1.1. Synthesis of 1-benzyloxycarbonyl-1,2,5,6-tetra-
hydropyridine, (2). Benzyl chloroformate (3.8 mL,
26.3 mmol) was added dropwise to a stirred mixture of
1,2,5,6-tetrahydropyridine 1 (2.0 g, 21.9 mmol) and
NaHCO3 (2.8 g, 26.3 mmol) in H2O (50 mL) at 0 8C, and
the mixture was stirred at room temperature for 4 h. After
this time CH2Cl2 (50 mL) and 5% aqueous Na2CO3 (20 mL)
were added, the organic phase was separated, and the
aqueous phase was extracted with CH2Cl2 (50 mL). The
combined organic phases were dried over Na2SO4,
the solvent was removed under reduced pressure and the
crude residue was purified by flash chromatography on
silica gel (hexane/EtOAc 3:1) to afford the Cbz-protected
4.1.4. Enzymatic hydrolysis of (G)-trans-4. The reaction
mixture containing (G)-trans-4 (50 mg, 0.15 mmol), the
appropriate amount of H2O (see Table 1) and the lipase
(50 mg) in the corresponding organic solvent (50 mL) was
shaken at 30 8C and 250 rpm in an orbital shaker. The
progress of the reaction was monitored by TLC (hexane/
EtOAc 1:1) until the achievement of the required
conversion. Then, the enzyme was removed by filtration
and washed with the corresponding organic solvent. The
solvent was evaporated under reduced pressure and the
crude residue was purified by flash chromatography on
silica gel (hexane/EtOAc 1:1) to afford the monoacetylated
product (3R,4R)-5 and the remaining substrate (3S,4S)-4.
1
product 2 as a yellow oil (4.4 g, 85%). H NMR (CDCl3,
300.13 MHz): d 7.36 (m, 5H), 5.81 (br s, 1H), 5.64 (br s,
1H), 5.16 (s, 2H), 3.96 (m, 2H), 3.57 (m, 2H), 2.14 (m, 2H);
13C NMR (CDCl3, 75.5 MHz): d 155.2 (CO), 136.6 (C),
128.2 (CH), 127.7 (CH), 127.6 (CH), 125.1 (CH), 124.8
(CH), 66.7 (CH2), 43.1 (CH2), 40.1 (CH2), 24.7 (CH2); IR
(CH2Cl2): n 1713, 1630, 1425 cmK1; MS (ESIC, m/z): 218
[(MCH)C, 100%], 240 [(MC Na)C, 30].
4.1.4.1. (3S,4S)-3,4-Diacetoxy-1-benzyloxycarbonyl-
piperidine, [(3S,4S)-4]. Yellow oil [a]2D5 C30.6 (c 1,
CHCl3), eeO99%.
4.1.4.2. (3R,4R)-3-Acetoxy-1-benzyloxycarbonyl-4-
hydroxypiperidine, [(3R,4R)-5]. Yellow oil [a]2D5 K20.2
4.1.2. Synthesis of (G)-1-benzyloxycarbonyl-3,4-epoxy-
piperidine, [(G)-3]. A solution of m-CPBA (4.3 g,
25.0 mmol) in dry CH2Cl2 (30 mL) was added dropwise
to a stirred solution of 2 (3.9 g, 18.0 mmol) in dry CH2Cl2
(20 mL) at 0 8C, and the mixture was stirred at room
temperature for 4 h, then washed with 5% aqueous K2CO3
solution (20 mL), followed by a saturated aqueous NaCl
solution (20 mL). The organic phase was dried over
1
(c 1, CHCl3), eeO99%. H NMR (CDCl3, 400.13 MHz,
40 8C): d 7.38 (m, 5H), 5.16 (2s, 2H), 4.66 (m, 1H), 4.03 (dd,
3
1H, JHHZ3.32, 13.32 Hz), 3.88–3.80 (m, 3H), 3.25 (m,
1H), 2.22 (br s, 1H, OH), 2.06 (s, 3H), 2.04–2.00 (m, 1H),
1.61 (m, 1H); 13C NMR (CDCl3, 100.6 MHz, 40 8C): d
170.6 (CO), 155.3 (CO), 136.6 (C), 128.5 (CH), 128.0 (CH),
127.9 (CH), 72.5 (CH), 69.4 (CH), 67.3 (CH2), 44.4 (CH2),

3290
L. F. Solares et al. / Tetrahedron 62 (2006) 3284–3291
40.8 (CH2), 30.8 (CH2), 20.9 (CH3); IR (CH2Cl2): n
3621, 1728, 1715, 1430, 1213 cmK1; MS (ESIC, m/z):
316 [(MC Na)C, 100%].
by flash chromatography on silica gel (hexane/EtOAc 1:2)
to afford the product (G)-cis-8 as a yellow oil (0.88 g,
76%). 1H NMR (CDCl3, 400.13 MHz, 0 8C): d 7.35 (m, 5H),
5.12 (2s, 2H), 3.84–3.30 (m, 6H), 1.82 (m, 1H), 1.65 (m,
1H); 13C NMR (CDCl3, 100.6 MHz, 0 8C): d 155.8 (CO),
136.4 (C), 128.4 (CH), 128.0 (CH), 127.7 (CH), 68.4 (CH),
67.8 (CH), 67.2 (CH2), 46.1 (CH2), 40.1 (CH2), 29.6 (CH2);
IR (CH2Cl2): n 3534, 3427, 1715, 1423, 1210 cmK1; MS
(ESIC, m/z): 252 [(MCH)C, 100%].
4.1.5. Preparation of (G)-trans-1-benzyloxycarbonyl-
3,4-dihydroxypiperidine, [(G)-trans-6]. To a solution of
(G)-trans-4 (1.2 g, 3.7 mmol) in MeOH (35 mL), a solution
of a catalytic amount of NaOMe in MeOH (1 mL) was
added. The mixture was stirred for 2 h, and then the solvent
was evaporated under reduced pressure and the product was
purified by flash chromatography on silica gel (hexane/
EtOAc 1:1) to afford the product (G)-trans-6 as a yellow oil
4.1.8. Synthesis of (G)-cis-3,4-diacetoxy-1-benzyloxy-
carbonylpiperidine, [(G)-cis-9]. To a solution at 0 8C of
(G)-cis-8 (0.9 g, 3.6 mmol) in CH2Cl2 (8 mL), Et3N
(2.0 mL, 14.3 mmol) and a catalytic amount of DMAP,
acetic anhydride (1.4 mL, 14.3 mmol) was added dropwise.
The resulting mixture was stirred at room temperature for
8 h. Then the solvent was removed under reduced pressure
and the crude residue was purified by flash chromatography
on silica gel (hexane/EtOAc 2:1) to afford the product
1
(0.85 g, 94%). H NMR (CDCl3, 400.13 MHz, 40 8C): d
3
7.35 (m, 5H), 5.13 (s, 2H), 4.25 (dd, 1H, JHHZ3.23,
13.10 Hz), 4.11–4.08 (m, 3H), 3.55–3.47 (m, 2H), 1.98 (m,
1H), 1.50 (m, 1H); 13C NMR (CDCl3, 100.6 MHz, 40 8C): d
155.3 (CO), 136.5 (C), 128.5 (CH), 128.1 (CH), 127.9 (CH),
71.8 (2 CH), 67.4 (CH2), 47.8 (CH2), 42.1 (CH2), 31.4
(CH2); IR (CH2Cl2): n 3618, 3447, 1710, 1425, 1204 cmK1
;
MS (ESIC, m/z): 252 [(MCH)C, 30%], 274 [(MC Na)C,
100%].
1
(G)-cis-9 as a yellow oil (1.1 g, 90%). H NMR (CDCl3,
400.13 MHz, 0 8C): d 7.38 (m, 5H), 5.22–5.03 (m, 4H),
4.07–3.95 (m, 2H), 3.89–3.84 (m, 2H), 2.08 (s, 3H),
2.02–1.81 (m, 2H), 1.95 (s, 3H); 13C NMR (CDCl3,
100.6 MHz, 0 8C): d 170.5 (2CO), 155.5 (CO), 136.4 (C),
128.7 (CH), 128.3 (CH), 128.0 (CH), 69.5 (2CH), 67.5
(CH2), 44.8 (CH2), 41.2 (CH2), 26.5 (CH2), 21.1 (CH3);
IR (CH2Cl2): n 1740, 1716, 1432, 1207 cmK1; MS (ESIC,
m/z): 358 [(MC Na)C, 100%].
4.1.6. Enzymatic acetylation of (G)-trans-6. The reaction
mixture containing (G)-trans-6 (50 mg, 0.15 mmol), vinyl
acetate (see Table 2) and the lipase (50 mg) in the
corresponding organic solvent (50 mL) was shaken at
30 8C and 250 rpm in an orbital shaker. The progress of
the reaction was monitored by TLC using the solvent system
hexane/EtOAc 1:1 until the required conversion was
achieved. Then, the enzyme was removed by filtration and
washed with the corresponding organic solvent. The solvent
was evaporated under reduced pressure and the crude
residue was purified by flash chromatography on silica gel
(hexane/EtOAc 1:1) to afford the monoacetylated product
(3R,4R)-7 and the remaining substrate (3S,4S)-6.
4.1.9. Enzymatic hydrolysis and acetylation of (G)-cis-8
and (G)-cis-9. These reactions were carried out using the
same procedures as described as for the trans derivatives.
Acknowledgements
4.1.6.1. (3S,4S)-1-Benzyloxycarbonyl-3,4-dihydroxy-
piperidine, [(3S,4S)-6]. Yellow oil [a]2D5 K3.7 (c 1,
CHCl3), eeO99%.
This work was supported by the Ministerio de Ciencia y
´
predoctoral fellowship. V.G.-F. thanks MEC for a personal
grant (Juan de la Cierva Program). We express our
appreciation to Novo Nordisk Co. for the generous gift of
the lipase Novozym 435.
Tecnologıa (PTR1995-0775-OP). L.F.S. thanks MEC for a
4.1.6.2. (3R,4R)-4-Acetoxy-1-benzyloxycarbonyl-3-
hydroxypiperidine, [(3R,4R)-7]. Yellow oil [a]2D5 K12.2
1
(c 1, CHCl3), eeO99%. H NMR (CDCl3, 400.13 MHz,
40 8C): d 7.38 (m, 5H), 5.15 (2s, 2H), 4.77 (m, 1H), 4.13 (dd,
3
1H, JHHZ3.16, 13.52 Hz), 3.97–3.87 (m, 3H), 3.19–2.92
(m, 1H), 2.52 (br s, 1H, OH), 2.13 (s, 3H), 2.12–2.05 (m,
1H), 1.59 (m, 1H); 13C NMR (CDCl3, 100.6 MHz, 40 8C): d
171.0 (CO), 155.4 (CO), 136.5 (C), 128.5 (CH), 128.1 (CH),
127.9 (CH), 74.7 (CH), 68.8 (CH), 67.4 (CH2), 47.7 (CH2),
41.4 (CH2), 28.2 (CH2), 21.1 (CH3); IR (CH2Cl2): n 3520,
1717, 1432, 1218 cmK1; MS (ESIC, m/z): 294.1 [(MCH)C,
40%], 316.1 [(MC Na)C, 100%].
References and notes
1. (a) Butters, T. D.; Dwek, R. A.; Platt, F. M. Chem. Rev. 2000,
100, 4683–4696. (b) Felpin, F.-X.; Bouberkeur, K.; Lebrenton,
J. J. Org. Chem. 2004, 69, 1497–1503 and references cited
therein.
2. (a) Anzeveno, P. B.; Creemer, L. J.; Daniel, J. K.; King, C.-H. R.;
Liu, P. S. J. Org. Chem. 1989, 54, 2539–2542. (b) Balfour, J. A.;
McTavish, D. Drugs 1993, 46, 1025–1054.
4.1.7. Preparation of (G)-cis-1-benzyloxycarbonyl-3,4-
dihydroxypiperidine, [(G)-cis-8]. To a solution of 2
(1.0 g, 4.6 mmol) in acetone–H2O (1/1, 8 mL), N-methyl-
morpholine N-oxide (0.8 g, 6.8 mmol) was added, followed
by a solution of OsO4 in tert-butanol (0.8 mL, 2.5% w/v),
then the mixture was stirred for 5 days at room temperature.
A saturated solution of Na2S2O5 (60 mL) was added and the
mixture was extracted with EtOAc (5!40 mL); the organic
phase was dried over Na2SO4, the solvent was removed
under reduced pressure and the crude residue was purified
3. Ratner, L.; Heyden, N. V.; Dedera, D. Virology 1999, 181,
180–192 and references cited therein.
4. Gross, P. E.; Baker, M. A.; Carver, J. P.; Dennis, J. W. Clin.
Cancer Res. 1995, 1, 935–944.
5. Ichikawa, Y.; Igasashi, Y.; Ichikawa, M.; Suhara, Y. J. Am.
Chem. Soc. 1998, 120, 3007–3018.
6. Williams, S. J.; Hoos, R.; Withers, S. G. J. Am. Chem. Soc.
2000, 122, 2223–2235.

L. F. Solares et al. / Tetrahedron 62 (2006) 3284–3291
3291
7. Cossy, J.; Molina, J. L.; Desmurs, J.-R. Tetrahedron Lett.
2001, 42, 5713–5715.
18. Lutz, S. Tetrahedron: Asymmetry 2004, 15, 2743–2748.
19. Recently, an S-selective CAL-B has been designed using a
rational redesign approach: Magnusson, A. O.; Takwa, M.;
Hamberg, A.; Hult, K. Angew. Chem., Int. Ed. 2005, 44,
4582–4585.
8. Waldmeier, P. C.; Maitre, L.; Baumann, P. A.; Hauser, K.;
Bischoff, S.; Bittiger, H.; Paioni, R. Eur. J. Pharmacol. 1986,
130, 1–10.
9. Chang, D.; Heringa, M. F.; Witholt, B.; Li, Z. J. Org. Chem.
2003, 68, 8599–8606.
20. Magnusson, A.; Hult, K.; Holmquist, M. J. Am. Chem. Soc.
2001, 123, 4354–4355.
10. (a) Crout, D. H. G.; Gaudet, V. S. B.; Laumen, K.; Schneider,
M. P. J. Chem. Soc., Chem. Commun. 1986, 808–810. (b)
Xie, Z.-F.; Nakamura, I.; Suemune, H.; Sakai, K. J. Chem.
Soc., Chem. Commun. 1988, 966–967. (c) Seemayer, R.;
Schneider, M. P. J. Chem. Soc., Chem. Commun. 1991, 49–50.
(d) Caron, G.; Kazlauskas, J. J. Org. Chem. 1991, 56,
7251–7256.
21. Branneby, C.; Carlqvist, P.; Magnusson, A.; Hult, K.; Brinck, T.;
Berglund, P. J. Am. Chem. Soc. 2003, 125, 874–875.
22. Carlqvist, P.; Svedendahl, M.; Branneby, C.; Hult, K.;
Brinck, T.; Berglund, P. ChemBioChem 2005, 6, 331–336.
Wild-type CAL-B has also achieved this type of transform-
ation: Torre, O.; Alfonso, I.; Gotor, V. Chem. Commun. 2004,
´
1724–1725. Torre, O.; Gotor-Fernandez, V.; Alfonso, I.;
´
Garcıa-Alles, L. F.; Gotor, V. Adv. Synth. Catal. 2005, 347,
11. Yu, S.-H.; Chung, S.-K. Tetrahedron: Asymmetry 2004,
581–584.
´
12. Solares, L. F.; Brieva, R.; Quiros, M.; Llorente, I.; Bayod, M.;
1007–1014.
23. Carlqvist, P.; Eklund, R.; Hult, K.; Brinck, T. J. Mol. Model.
2003, 9, 164–171.
Gotor, V. Tetrahedron: Asymmetry 2004, 15, 341–345.
13. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294–7299.
´ ´
14. Garcıa, J.; Fernandez, S.; Ferrero, M.; Sanghvi, Y. S.; Gotor,
V. Org. Lett. 2004, 6, 3759–3762.
24. Magnusson et al. have shown that for wild-type CAL-B
there is enough space up to an ethyl moiety into the
stereospecifity pocket: Magnusson, A. O.; Rotticci-Mulder,
J. C.; Santagostino, A.; Hult, K. ChemBioChem 2005, 6,
1051–1056.
15. (a) Uppenberg, J.; Patkar, S. A.; Bergfors, T.; Jones, T. A. J. Mol.
Biol. 1994, 235, 790–792. (b) Uppenberg, J.; Hansen, M. T.;
Patkar, S. A.; Jones, T. A. Structure 1994, 2, 293–298.
25. Kazlauskas, R. J.; Weissfloch, A. N. E.; Rappaport, A. T.;
Cuccia, L. A. J. Org. Chem. 1991, 56, 2656–2665.
¨
16. (a) Uppenberg, J.; Ohrner, N.; Norin, M.; Hult, K.; Kleywegt,
´
26. (a) Forro, E.; Kanerva, L. T.; Fu¨lop, F. Tetrahedron:
Asymmetry 1998, 9, 513–520. (b) Luna, A.; Astorga, C.;
¨
G. J.; Patkar, S.; Waagen, V.; Anthonsen, T.; Jones, T. A.
Biochemistry 1995, 34, 16838–16851. (b) Hæffner, F.;
Norin, T.; Hult, K. Biophys. J. 1998, 74, 1251–1262. (c)
Cuiper, A. D.; Kouwijzer, M. L. C. E.; Grootenhius, P. D. J.;
Kellogg, R. M.; Feringa, B. L. J. Org. Chem. 1999, 64,
9529–9537. (d) Otto, R. T.; Scheib, H.; Bornscheuer, U. T.;
Pleiss, J.; Syldatk, C.; Schmid, R. D. J. Mol. Catal. B 2000, 8,
¨
Fu¨lop, F.; Gotor, V. Tetrahedron: Asymmetry 1998, 9,
4483–4487. (c) Brunet, C.; Zarevucka, M.; Wimmer, Z.;
Legoy, M.-D. Enzyme Microb. Technol. 2002, 31, 609–614.
(d) Levy, L. M.; Dehli, J. R.; Gotor, V. Tetrahedron:
Asymmetry 2003, 14, 2053–2058. (e) Tanikaga, R.;
Matsumoto, Y.; Sakaguchi, M.; Koyama, Y.; Ono, K.
Tetrahedron Lett. 2003, 44, 6781–6783.
´
¨
201–211. (e) Park, S.; Forro, E.; Grewal, H.; Fu¨lop, F.;
Kazlauskas, R. J. Adv. Synth. Catal. 2003, 345, 986–995.
(f) Venkatachalam, T. K.; Samuel, P.; Li, G.; Qazi, S.;
Mao, C.; Pendergrass, S.; Uckun, F. M. Bioorg. Med. Chem.
2004, 12, 3371–3381. (g) Levy, L. M.; Lavandera, I.; Gotor, V.
Org. Biomol. Chem. 2004, 2, 2572–2577. (h) Lavandera, I.;
27. (a) Cornell, W. D.; Cieplak, P.; Bayley, C. I.; Kollmann, P. A.
J. Am. Chem. Soc. 1993, 115, 9620–9631. (b) Weiner, S. J.;
Kollman, P. A.; Case, D. A.; Singh, U. C.; Ghio, C.;
Alagona, G.; Profeta, S.; Weiner, P. J. Am. Chem. Soc. 1984,
106, 765–784.
´
Fernandez, S.; Magdalena, J.; Ferrero, M.; Kazlauskas, R. J.;
Gotor, V. ChemBioChem 2005, 6, 1381–1390.
28. Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kebeler, M.;
Stu¨mer, R.; Zelinski, T. Angew. Chem., Int. Ed. 2004, 43,
788–824.
17. (a) Raza, S.; Fransson, L.; Hult, K. Protein Sci. 2001, 10,
329–338. (b) Ottosson, J.; Rotticci-Mulder, J. C.; Rotticci, D.;
Hult, K. Protein Sci. 2001, 10, 1769–1774. (c) Ottosson, J.;
Fransson, L.; Hult, K. Protein Sci. 2002, 11, 1462–1471.
29. DeLano, W. L. The PyMOL Molecular Graphics System;
DeLano Scientific LLC: San Carlos, CA, USA 2004;
http://www.pymol.org.