Preparation of enantiomerically pure N-heterocyclic amino alcohols by enzymatic kinetic resolution

Article abstract of DOI:10.1016/j.tetasy.2015.04.009

Abstract The synthesis of both enantiomers of N-benzyl-3-hydroxypyrrolidine and N-benzyl-3-hydroxypiperidine via enzymatic kinetic resolution of the corresponding racemic esters is described. Various commercially available hydrolases were studied as biocatalysts in native and immobilized form. The best results were obtained with lipases PS, AK, CAL-B and with protease Alcalase, which were active and selective for the kinetic resolutions of racemic esters (E > 100). Under optimized reaction conditions, highly enantiomerically enriched (up to 99.5% ee) resolution products were obtained. Lipase and protease showed opposite enantiopreference on the esters, allowing the preparation of both enantiomers of the target compounds. Semi-continuous reactions in column reactors with immobilized biocatalysts were also performed with high enantioselectivities. Inversion of the configuration at C(3) of N-benzyl-3-hydroxypyrrolidine was quantitatively effected in a short number of steps.

Full text of DOI:10.1016/j.tetasy.2015.04.009

Tetrahedron: Asymmetry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Preparation of enantiomerically pure N-heterocyclic amino alcohols
by enzymatic kinetic resolution
b,
Giorgio Tofani a, Antonella Petri a, , Oreste Piccolo
⇑
⇑
a Dipartimento di Chimica e Chimica Industriale, Università di Pisa, Via Giuseppe Moruzzi 3, 56124 Pisa, Italy
b Studio di Consulenza Scientifica (SCSOP), Via Bornò 5, 23896 Sirtori (LC), Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of both enantiomers of N-benzyl-3-hydroxypyrrolidine and N-benzyl-3-hydroxypiperidine
via enzymatic kinetic resolution of the corresponding racemic esters is described. Various commercially
available hydrolases were studied as biocatalysts in native and immobilized form. The best results were
obtained with lipases PS, AK, CAL-B and with protease Alcalase, which were active and selective for the
kinetic resolutions of racemic esters (E > 100). Under optimized reaction conditions, highly enantiomer-
ically enriched (up to 99.5% ee) resolution products were obtained. Lipase and protease showed opposite
enantiopreference on the esters, allowing the preparation of both enantiomers of the target compounds.
Semi-continuous reactions in column reactors with immobilized biocatalysts were also performed with
high enantioselectivities. Inversion of the configuration at C(3) of N-benzyl-3-hydroxypyrrolidine was
quantitatively effected in a short number of steps.
Received 30 March 2015
Accepted 15 April 2015
Available online xxxx
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction
OH
OH
*
*
N-Heterocyclic chiral molecules have been of considerable
interest for industry and academic research.1,2 In particular, enan-
tiomerically pure 3-hydroxypyrrolidine 1 and 3-hydroxypiperidine
2 (Fig. 1) have proven very useful in a wide range of applications
either as valuable intermediates for the synthesis of biologically
active compounds3–6 or as versatile chiral ligands and promoters
in organocatalysis.7,8
N
N
Ph
Ph
1
2
Figure 1. Structure of N-benzyl-3-hydroxypyrrolidine 1 and N-benzyl-3-hydrox-
ypiperidine 2.
A
number of synthetic5,9 as well as biotransformation10
approaches for the preparation of amino alcohols 1 and 2 have
already been described. However, most of the reported procedures
show one or more drawbacks from the viewpoint of economical
efficiency. Among the available biocatalytic routes, enzymatic
kinetic resolution of racemic alcohols via hydrolysis or acylation
has been extensively studied.11 Indeed, biocatalytic processes have
often proved to be greener, less hazardous and less polluting than
conventional chemical transformations.12 Several examples have
been reported in the literature for the kinetic resolution of racemic
amino alcohols 1 and 2 and of their corresponding esters.13–16
Recently, an efficient chemoenzymatic dynamic kinetic resolution
using various lipases in combination with a ruthenium based cat-
alyst has also been described.17 This efficient approach however
involves an expensive catalyst, making it unsuitable for large scale
synthesis.
Herein, the kinetic resolution of derivatives of N-benzyl-3-hy-
droxypyrrolidine and N-benzyl-3-hydroxypiperidine alcohols by
using several commercially available hydrolytic enzymes in their
native and immobilized form is described. The resulting optimized
conditions for the analytical scale reactions were successfully
applied in semi-continuous processes in column reactors with
immobilized biocatalysts to facilitate the enzymatic resolution.
2. Results and discussion
Initially, the kinetic resolution of racemic acetate rac-3 was
studied using different commercially available enzymes
(Scheme 1).
⇑
Corresponding authors. Tel.: +39 050 2219279 (A.P.).
E-mail addresses: antonella.petri@unipi.it (A. Petri), orestepiccolo@tin.it
(O. Piccolo).
http://dx.doi.org/10.1016/j.tetasy.2015.04.009
0957-4166/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Tofani, G.; et al. Tetrahedron: Asymmetry (2015), http://dx.doi.org/10.1016/j.tetasy.2015.04.009

2
G. Tofani et al. / Tetrahedron: Asymmetry xxx (2015) xxx–xxx
OH
OCOCH3
OCOCH3
Using the aforementioned optimal experimental conditions for
*
*
enzyme
the resolution of rac-3, the same screening procedure was further
extended to the racemic 3-hydroxypiperidine derivative rac-4
(Scheme 2, Table 2).
+
N
N
N
buffer-MTBE (9:1)
r.t.
Ph
rac-3
Ph
Ph
1
3
OH
OCOR
OCOR
*
*
Scheme 1. Enzymatic kinetic resolution of rac-3.
enzyme
+
buffer-MTBE (9:1)
r.t.
N
N
N
Since the addition of an organic solvent was reported to favour
Ph
Ph
Ph
the biocatalysed hydrolysis and to improve enantioselectivity in
some cases, we decided to perform the reaction in a 9:1 v/v mix-
ture of aqueous phosphate buffer (50 mM, pH 7) and MTBE.
Among the previously examined co-solvents, the choice of MTBE
was determined by the following considerations: (i) good ability
to solubilize substrates and extract the amino alcohols compounds
from the buffer solution; (ii) limited miscibility with the aqueous
phase to prevent the inactivation of the enzyme and facilitate the
work-up of the reaction mixture; (iii) suitability for industrial
use and low cost. In addition, unlike dioxane which has been pre-
viously used in this reaction by Tomori et al.,13 MTBE has not been
classified as a carcinogen. n-Butanol, which has a logP value simi-
lar to that of MTBE, has been considered as a possible alternative
but in this solvent the enzyme catalysed hydrolysis proceeded with
low activity. The reactions were monitored by HPLC. Data on reac-
tion times, conversions and ee values are presented in Table 1.
It was found that the best results in terms of activity and selec-
tivity were obtained using lipase enzymes PS and AK and with pro-
tease Alcalase. All reactions were carried out by using an
enzyme/substrate ratio of 1:5. It is interesting to note that Lipase
PS was already reported13 as an efficient enzyme but the described
reaction required a large amount of the biocatalyst, making it
unsuitable for a large scale synthesis. Lipase AK and Alcalase were
also used in the present research in 100% buffer as the reaction sol-
vent (entries 6 and 9). It was found that under these conditions, the
rate of hydrolysis was higher but the enantioselectivity was lower.
In order to optimize the productivity, the kinetic resolution was
carried out at a higher concentration of the substrate. It was found
that lipase AK was quite effective up to 70 g/L. The selectivity
remained high (E > 100); however, as expected, the rate of
hydrolysis decreased with higher concentrations (25 h vs 8.5 h).
This result is particularly promising since it was possible to carry
out the kinetic resolution in 25 h, with excellent enantio-
selectivities both of alcohol 1 and acetate 3, by using a lower
amount of biocatalyst (<1 wt %) with respect to the values reported
in literature.13–17
4
5
2
, R=CH3
rac-4, R=CH3
rac-5, R=C3H7
, R=C3H7
Scheme 2. Enzymatic kinetic resolution of rac-4 and rac-5.
The best results were obtained with the enzymes PS and CAL-B.
However it is apparent from data in Table 2 that six-member ring
compounds are less reactive than the corresponding compounds
with a five-member ring (Table 1). This behaviour was already
reported in a previous study.13 The butanoate ester rac-5 was also
tested as substrate (Scheme 2). As presented in Table 3, Palatase
and Alcalase were found to be the most active and selective
enzymes. However, these data indicated that the enantioselectivity
was only slightly affected by changing the acyl chain length.
We next investigated the use of immobilized enzymes in
view of a possible improvement of the process. Indeed, immobi-
lized biocatalysts are easily separable from the reaction mixture
and potentially recyclable. The screening process involved
immobilized CAL-B, PS, AK, Palatase and Alcalase which proved
to be most efficient enzymes in their native form for the kinetic
resolution of rac-3–5. The results are reported in Table 4.
AK-IM and PS-IM allowed to obtain (R)-1 with high enantiose-
lectivity (ee >99.5%, E > 100). Both enantiomers of amino alcohol
2 were obtained with 98% ee from the hydrolysis of the corre-
sponding acetate and butanoate.
Encouraged by these results, we next examined the possibility
of using the immobilized enzymes for semi-continuous kinetic res-
olution reactions. For this purpose, a microreactor consisting of a
PEEK column filled with AK-IM was prepared and the reaction mix-
ture containing rac-3 was re-circulated inside by using a suitable
pump. The reaction was monitored by HPLC and stopped at an
approx. 50% conversion, removing the reaction mixture. The col-
umn was washed with phosphate buffer (50 mM, pH 7). The eluted
solution and washings were combined to quantitatively recover
Table 1
Screening of different enzymes for the hydrolysis of rac-3
a
a
Entry
Enzyme
Time (h)
conva(%)
eeP (%)
eeS (%)
Eb
1
2
3
4
AS
23
1
5
6
1
62
68
39
52
49
42
44
50
48
50
50
12 (S)
24 (R)
62 (R)
82 (R)
96 (R)
90 (S)
>99.5 (S)
>99.5 (R)
98 (R)
>99.5 (R)
>99.5 (R)
20 (R)
50 (S)
40 (S)
90 (S)
92 (S)
66 (R)
78 (R)
>99.5 (S)
90 (S)
>99.5 (S)
98 (S)
2
3
6
AY30
CAL-A
Palatase
CAL-B
Alcalase
Alcalase
PS
AK
AK
AK
31
>100
38
>100
>100
>100
>100
>100
5
6c
7
5
15
0.5
2
8.5
25
8
9c
10
11d
All reactions were carried out at room temperature. Enzyme/Substrate 1:5 (w/w). 10% MTBE (v/v) co-solvent used
with respect to buffer.
a
Conversion and ee were determined from HPLC. The absolute configurations of the substrate and the product
were assigned by a comparison of the elution orders on chiral HPLC (cfr. Experimental).
b
E was evaluated by using E = ln[1- conv(1 + eeP)]/ln[1-conv(1 - eeP)].18
100% buffer as reaction solvent.
Substrate concentration 70 g/L.
c
d
Please cite this article in press as: Tofani, G.; et al. Tetrahedron: Asymmetry (2015), http://dx.doi.org/10.1016/j.tetasy.2015.04.009

G. Tofani et al. / Tetrahedron: Asymmetry xxx (2015) xxx–xxx
3
Table 2
Screening of different enzymes for the hydrolysis of rac-4
a
a
Entry
Enzyme
Time (h)
conva(%)
eeP (%)
eeS (%)
Eb
1
2
3
4
5
6
7
8
AS
32
30
0.25
33
72
16
2
33
42
53
20
33
35
50
33
20 (S)
38 (R)
56 (S)
88 (S)
94 (R)
94 (R)
96 (R)
>99.5 (R)
19 (R)
28 (S)
64 (R)
22 (R)
46 (S)
50 (S)
96 (S)
50 (S)
2
3
7
CAL-A
AY30
Alcalase
AK
Palatase
PS
19
51
53
>100
>100
CAL-B
6
All reactions were carried out at room temperature. Enzyme/Substrate 1:5 (w/w). 10% MTBE co-solvent used with
respect to buffer.
a
Conversion and ee were determined using HPLC. The absolute configurations of the substrate and the product
were assigned by a comparison of the elution orders on chiral HPLC (cfr. Experimental).
b
E was evaluated by using E = ln[1- conv(1 + eeP)]/ln[1-conv(1 - eeP)].18
Table 3
Screening of different enzymes for the hydrolysis of rac-5
conva(%)
eeP (%)
eeS (%)
Eb
a
a
Entry
Enzyme
Time (h)
1
2c
3
4
5
6
7
8
AK
72
0.2
0.5
25
4
4
9
3
36
39
33
52
33
49
40
42
18 (R)
20 (S)
36 (R)
66 (S)
94 (R)
92 (R)
96 (R)
98 (S)
10 (S)
13 (R)
18 (S)
72 (R)
46 (S)
88 (S)
64 (S)
72 (R)
2
2
3
10
51
70
95
AY30
CAL-A
AS
CAL-B
PS
Palatase
Alcalase
>100
All reactions were carried out at room temperature. Enzyme/Substrate 1:5 (w/w). 10% MTBE co-solvent used with
respect to buffer.
a
Conversion and ee were determined using HPLC. The absolute configurations of the substrate and the product
were assigned by a comparison of the elution orders on chiral HPLC (cfr. Experimental).
b
E was evaluated by using E = ln[1- conv(1 + eeP)]/ln[1-conv(1 - eeP)].18
Enzyme/Substrate 1:50 (w/w).
c
Table 4
Hydrolysis of rac-3–5 with immobilized enzymes
a
a
Entry
Substrate
Enzyme
Time (h)
conva(%)
eeP (%)
eeS (%)
Eb
1
2
3c
4
5
6
7
8
rac-3
CAL-B-IM
Alcalase-IM
AK-IM
PS-IM
CAL-B-IM
PS-IM
5
7
7
24
72
48
48
5
49
53
46
50
32
42
49
49
90 (R)
82 (S)
>99.5 (R)
>99.5 (R)
92 (R)
98 (R)
86 (R)
98 (S)
86 (S)
94 (R)
84 (S)
>99.5 (S)
44 (S)
70 (S)
82 (S)
94 (R)
53
35
>100
>100
37
>100
33
>100
rac-4
rac-5
Palatase-IM
Alcalase-IM
All reactions were carried out at room temperature. Substrate/Immobilized enzyme1:20 (w/w). 10% MTBE co-solvent
used with respect to buffer.
a
Conversion and ee were determined using HPLC. The absolute configurations of the substrate and the product
were assigned by a comparison of the elution orders on chiral HPLC (cfr. Experimental).
b
E was evaluated by using E = ln[1- conv(1 + eeP)]/ln[1-conv(1 - eeP)].18
Substrate/Immobilized enzyme 1:10 (w/w).
c
the amino alcohol (R)-1 and the unreacted acetate (S)-3 which
exhibited the same ee values reported for the batch reaction.
The column was used for five reaction cycles to evaluate the
recycling of the biocatalyst. The activity and enantioselectivity
values remained comparable with the data obtained in the first
reaction.
The enzyme Alcalase-IM was then used for the kinetic resolu-
tion of substrate rac-5 in a microreactor prepared as previously
described for the hydrolysis of rac-3. (S)-2 and the unreacted (R)-
5 were quantitatively recovered from the eluted solution and
washings, obtaining the same ee values reported for the batch
reaction. The column was used five times, without loss of activity
or enantioselectivity.
The inversion of configuration at C(3) of alcohol 1 was finally
investigated in order to evaluate the possibility of reusing the
unwanted enantiomer, thus obtaining a quantitative yield after
the kinetic resolution. One of the possible approaches is the trans-
formation of the hydroxyl group into a suitable leaving group
which allows the attack by a nucleophile and the resulting inver-
sion of the configuration of the alcohol. If this reaction is carried
out on the alcohol/acetate mixture obtained at the end of the enzy-
matic reaction, it would be possible to obtain the enantiomerically
pure acetate with a theoretical yield of 100%. Inversion of configu-
ration of alcohol 1 via the corresponding mesylate has already been
reported,16 but with partial loss of ee. We therefore envisaged
using the tosylate in the same reaction sequence (Scheme 3).
Please cite this article in press as: Tofani, G.; et al. Tetrahedron: Asymmetry (2015), http://dx.doi.org/10.1016/j.tetasy.2015.04.009

4
G. Tofani et al. / Tetrahedron: Asymmetry xxx (2015) xxx–xxx
OTs
OH
OCOCH3
OH
a)
b)
c)
N
N
N
N
Ph
(S)-1
Ph
(S)-6
Ph
(R)-3
Ph
(R)-1
Scheme 3. Inversion of configuration at C(3) of (S)-1. Reagents and conditions: (a) TsCl, DABCO, MTBE, rt; (b) AcONa, DMSO, 120 °C; (c) NaOH 2 M, rt.
Tosylation of the alcohol (S)-1 (ee >99.5%) afforded the key
Chemical shifts on the d scale are expressed in ppm values. Thin-
layer chromatography (TLC) was carried out using with
Macherey-Nagel AlugramÒ SIL G/UV254 plates by using ethyl
acetate/triethylamine (100:1, v/v) as eluent.
intermediate 6 with quantitative yield. Subsequent displacement
of the tosylate with sodium acetate gave the R enantiomer of acet-
ate 3 (ee 95%). Finally, amino alcohol (R)-1 was obtained with
quantitative yield and 95% ee by alkaline hydrolysis. This result
illustrates the viability of this procedure for the preparation of
the enantiopure alcohols with quantitative yield starting from
alcohol/acetate mixtures obtained at the end of the enzymatic
reaction without any purification. The same reaction sequence
was performed on the substrate (S)-2. The tosylate was obtained
in good yield, but subsequent reaction with sodium acetate
afforded only the elimination product.
High performance liquid chromatography (HPLC) analyses were
conducted with a Jasco instrument using a Gemini C18 column
(150 Â 4.6 mm) and a mixture of acetonitrile/water/diethylamine
65:35:0.1 as eluent [flow rate: 1 ml/min; detection wavelength:
256 nm; retention times: 2.0 min for 1, 3.2 min for 3, 2.3 min for
2, 4.5 min for 4 and 8.5 min for 5]. The identification of peaks in
the chromatograms was made by spiking with commercially avail-
able substrates or references prepared during this investigation.
The enantiomeric excess (ee %) was determined by HPLC with chi-
ral columns. For separation of rac-1 and rac-3: Chiralcel OJ (Daicel),
250 Â 4.6 mm; eluent hexane/ethanol/diethylamine, 100:1:0.1;
flow rate: 1 ml/min; detection wavelength: 254 nm; retention
times: 27.5 min for (R)-1, 30.5 min for (S)-1, 12.4 min for (R)-3
and 13.3 min for (S)-3. For separation of rac-2 and rac-4:
Chiralcel OJ (Daicel), 250 Â 4.6 mm; eluent hexane/ethanol/diethy-
lamine, 98:2:0.1; flow rate: 0.5 ml/min; detection wavelength:
254 nm; retention times: 26.8 min for (R)-2, 29.4 min for (S)-2,
14.4 min for (R)-4 and 22 min for (S)-4. For separation of
rac-2 and rac-5: Lux Cellulose-3 (Phenomenex), 250 Â 4.6 mm;
eluent hexane/isopropanol/diethylamine, 95:5:0.1; flow rate:
0.5 ml/min; detection wavelength: 254 nm; retention times:
16.9 min for (R)-2, 19.0 min for (S)-2, 9.8 min for (R)-5, 11.5 min
for (S)-5. The absolute configuration of substrates and products
was assigned by a comparison of the elution orders with those
of authentic samples of known configuration. Optical rotations
were determined on a Jasco Dip-360 polarimeter at 20 °C using
sodium D light.
3. Conclusions
In conclusion, an efficient enzymatic kinetic resolution of race-
mic esters rac-3–5 has been achieved to provide a practical prepa-
ration of both enantiomers of N-benzyl-3-hydroxypyrrolidine 1
and N-benzyl-3-hydroxypiperidine 2. The target compounds were
obtained with high enantiomeric excess under mild reaction condi-
tions by the proper selection of the biocatalyst and the use of a
suitable co-solvent. The use of immobilized enzymes was per-
formed with excellent enantioselectivity in both batch and column
reactions. Finally, inversion of the stereochemistry at C(3) of alco-
hol 1 was achieved without racemization.
4. Experimental
4.1. Reagents and solvents
N-Benzyl-3-hydroxypyrrolidine, N-benzyl-3-hydroxypiperidine,
(S)-N-benzyl-3-hydroxypyrrolidine, (R)-N-benzyl-3-hydroxypiperidine
and all other inorganic and organic reagents and solvents
were purchased from Sigma–Aldrich. All solvents were dried and
purified by standard methods as required. Lipase enzymes CAL-B
from Candida antarctica B (from Novozymes, protein concentration
3.5 mg/ml, activity 8000 TBU/ml), Palatase from Rhizomucor miehei
recombinant in Aspergillus oryzae (from Novozymes, 5.5 mg/ml,
4.3. Synthesis of racemic esters rac-3–5
4.3.1. Synthesis of racemic N-benzyl-3-pyrrolidinyl acetate rac-3
To a solution of 0.50 g (2.82 mmol) of rac-1 in 1 ml of dry pyr-
idine cooled to 0 °C, 0.35 ml (3.70 mmol) of acetic anhydride was
added under nitrogen. The resulting mixture was stirred overnight
at room temperature. The reaction mixture was diluted with
toluene (3 Â 3 ml) and then concentrated under reduced pressure
to give 0.61 g (2.79 mmol) of rac-3 as a reddish oil (99% yield).
1H NMR (300 MHz, CDCl3, d ppm): 1.80 (s, 1H); 2.03 (s, 3H);
2.21–2.43 (m, 2H); 2.65–2.83 (m, 3H); 3.58 (d, 1H, J = 12.9 Hz);
3.68 (d, 1H, J = 12.9 Hz); 5.14–5.20 (m, 1H); 7.22–7.40 (m, 5H);
TLC Rf = 0.52.
19,000 TBU/ml), CAL-A from Candida antarctica
A (from
Novozymes, 13.5 mg/ml, 10,000 TBU/ml), PS from Burkholderia
cepacia (from Amano, 5.4 mg/g, 24,000 TBU/g), AK from
Pseudomonas fluorescens (from Amano, 142 mg/g, 20,000 TBU/g),
AY30 from Candida rugosa (from Amano, 5.3 mg/g, 6500 TBU/g),
AS from Aspergillus niger (from Amano, 71.9 mg/g, 12,000 TBU/g)
and protease Alcalase from Bacillus licheniformis (from
Novozymes, 50 mg/ml, 4500 ELU/ml) were all obtained from BiCT
srl (Lodi, Italy). Immobilized enzymes CALB-IM from Candida
antarctica B (3200 TBU/g), PS-IM Burkholderia cepacia (3211 TBU/g),
Palatase-IM from Rhizomucor miehei recombinant in Aspergillus
oryzae (5100 TBU/g), AK-IM from Pseudomonas fluorescens
(2583 TBU/g) and Alcalase-IM from Bacillus licheniformis (850 ELU/g)
were also purchased from BiCT srl.
4.3.2. Synthesis of racemic N-benzyl-3-piperidinyl acetate rac-4
To a solution of 0.50 g (2.62 mmol) of rac-2 in 1 ml of dry pyr-
idine cooled to 0 °C, 0.35 ml (3.70 mmol) of acetic anhydride was
added under nitrogen. The resulting mixture was stirred overnight
at room temperature. The reaction mixture was diluted with
toluene (3 Â 3 ml) and concentrated under reduced pressure to
give 0.57 g (2.45 mmol) of rac-4 as a pale yellow oil (94% yield).
1H NMR (300 MHz, CDCl3, d ppm): 1.46–1.94 (m, 4H); 2.04 (s,
3H); 2.10–2.21 (m, 2H); 2.45–2.79 (m, 2H); 3.57 (s, 2H); 4.84 (s,
1H); 7.22–7.45 (m, 5H); TLC Rf = 0.59.
4.2. Analytical methods
The 1H NMR spectra were recorded on a Varian 300 spectrom-
eter at 300 MHz. Spectra were recorded at 25 °C in CDCl3.
Please cite this article in press as: Tofani, G.; et al. Tetrahedron: Asymmetry (2015), http://dx.doi.org/10.1016/j.tetasy.2015.04.009

G. Tofani et al. / Tetrahedron: Asymmetry xxx (2015) xxx–xxx
5
4.3.3. Synthesis of racemic N-benzyl-3-piperidinyl butanoate
rac-5
To a mixture of 0.50 g (2.62 mmol) of rac-2 and 1 ml of dry pyr-
(R)-1: colourless oil; [
(c 0.5, MeOH)}. The spectroscopic data for this compound were in
agreement with those of its racemate.
a
]
20 = +3.9 (c 0.5, MeOH) {lit.13
[
a
]
20 = +3.9
D
D
idine 0.57 ml (3.50 mmol) of butanoic anhydride was added at 0 °C
under nitrogen. The resulting mixture was stirred over night at
room temperature. The reaction mixture was diluted with dichlor-
omethane and then extracted with a saturated sodium carbonate
solution (2 Â 4 ml) and with 4 ml of a saturated sodium chloride
solution. The organic phase was dried with sodium sulfate and
then concentrated at reduced pressure to give 0.65 g (2.49 mmol)
of rac-5 as a pale yellow oil (95% yield). 1H NMR (300 MHz,
CDCl3, d ppm): 0.98 (t, 3H, J = 5.9 Hz); 1.43–1.45 (m, 1H); 1.63–
1.90 (m, 3H); 2.31–2.40 (m, 4H); 2.46 (m, 2H), 2.67–2.83 (m,
2H); 3.58–3.64 (m, 2H); 4.85–4.89 (m, 1H); 7.22–7.31 (m, 5H);
TLC Rf = 0.60.
The column was reused in five subsequent reactions under
identical experimental conditions without loss in activity and
enantioselectivity.
4.5.3. Preparative scale hydrolysis of rac-5 by using an
immobilized enzyme in a semi-continuous flow system
0.65 g of protease Alcalase-IM was packed into a PEEK column
which was then connected to a suitable pump. Phosphate buffer
solution (50 mM, pH 7) was circulated through the column at a
flow rate of 4 mL/min prior to use. A solution of rac-5 (0.20 g,
0.77 mmol) in a 9:1 mixture (4 mL) of phosphate buffer (50 mM,
pH 7) and MTBE was passed through the column reactor at
25 °C. The reaction was monitored by HPLC by taking samples at
different intervals of time and stopped at an approximately 50%
conversion. After 8 h, the column was washed with 5 ml of a
9:1 mixture of potassium phosphate buffer (50 mM, pH 7) and
MTBE and then with 5 ml of potassium phosphate buffer. The
eluted solution and washings were extracted with dichloro-
methane. The organic layer was dried over sodium sulfate,
concentrated at reduced pressure and analysed by HPLC as
described in Section 4.2. Purification of the residue by silica gel
chromatography gave (S)-2 (43% yield) with 98% ee and (R)-5
4.4. Analytical scale enzymatic hydrolysis of rac-3–5
The reactions were conducted at room temperature in suitable
vessels, using a vortex mixer. In a typical small scale experiment, to
a solution of racemic ester (10 mg) in potassium phosphate buffer
(50 mM, pH 7) or in a potassium phosphate buffer/organic solvent
mixture (1 mL), native (2 mg) or immobilized (200 mg) enzymes
were added. The reactions were monitored by HPLC by taking sam-
ples at different intervals of time. The reaction mixture was
extracted with dichloromethane (2 Â 0.5 mL). The organic layer
was dried over sodium sulfate, concentrated at reduced pressure
and analysed by HPLC as described in Section 4.2. Data on reaction
times, conversion and enantiomeric excess of the substrates and
the products are presented in Tables 1–4.
(47% yield) with 94% ee.
20
(S)-2: colourless oil; [
a]
20 = +13.5 (c 0.5, MeOH) {lit.13
[
a]
of
D
D
(R)-2 = À13.3 (c 0.22, MeOH)}. The spectral data for this compound
were in agreement with those of its racemate.
The column was reused in five subsequent reactions under
identical experimental conditions without loss in activity and
enantioselectivity.
4.5. Preparative scale procedure
4.5.1. Preparative scale hydrolysis of rac-3 by using a native
enzyme
4.6. Inversion at C-3 of (3S)-N-benzyl-3-hydroxypyrrolidine (S)-1
Into a stirred solution of 250 mg (1.14 mmol) of rac-3 in a 9:1
mixture (3.5 mL) of potassium phosphate buffer (50 mM, pH 7)
and MTBE, 2 mg of lipase AK were added. The reaction was moni-
tored by HPLC by taking samples at different intervals of time and
stopped at an approximately 50% conversion. The reaction mixture
was extracted with dichloromethane. The organic layer was dried
over sodium sulfate, concentrated at reduced pressure and sub-
jected to preparative TLC to give (R)-1 with >99.5% ee (83 mg;
41% yield) and (S)-3 with 98% ee (122 mg; 49% yield) (conv = 50,
To a solution of 0.50 g (2.82 mmol) of (S)-1 (ee >99.5%) and
0.40 g (3.57 mmol) of 1,4-diazabicyclo[2.2.2]octane (DABCO) in
3 ml of MTBE cooled to 0 °C, 0.65 g (3.14 mmol) of 4-toluenesul-
fonyl chloride was added under nitrogen. The mixture was stirred
overnight at room temperature. Then 2 ml of ethyl acetate and
3 ml of saturated sodium bicarbonate solution were added and
the mixture was stirred for 15 min. The organic layer was sepa-
rated and extracted with 3 ml of saturated sodium bicarbonate
solution, 3 ml of saturated ammonium chloride solution and
3 ml of saturated sodium chloride solution. The organic layer
was dried over sodium sulfate and concentrated at reduced
pressure to give 0.93 g (2.81 mmol) of (S)-6 as a reddish oil (99%
yield). 1H NMR (300 MHz, CDCl3, d ppm): 1.60–2.41 (m, 6H);
2.78–3.10 (m, 3H); 3.65 (s, 2H); 4.9 (m, 1H); 7.32–7.58 (m, 9H);
TLC Rf = 0.54.
To a solution of 0.50 g (1.51 mmol) of (S)-6 in 2 ml of dry
dimethyl sulfoxide, 0.25 g (3.05 mmol) of anhydrous sodium acet-
ate was added. The mixture was stirred at 120 °C for 2 h. The reac-
tion was monitored by HPLC. After cooling to room temperature,
the mixture was diluted with water. The aqueous layer was
extracted with dichloromethane (3 Â 0.5 mL). The organic layers
were combined, dried over sodium sulfate and concentrated at
reduced pressure to give 0.31 g (1.42 mmol) of (R)-3, (94% yield,
95% ee). To this residue, 1 ml (80 mmol) of 2 M sodium hydroxide
solution was added. This mixture was stirred at room temperature
for 1 h and then the aqueous layer was extracted with dichloro-
methane (3 Â 0.5 mL). The organic layers were combined, dried
over sodium sulfate and concentrated at reduced pressure to give
0.20 g of (R)-N-benzyl-3-hydroxypyrrolidine (R)-1 (quantitative
yield, 95% ee).
E > 100). (R)-1: colourless oil; [
a]
20 = +3.9 (c 0.5, MeOH) {lit.13
D
[a]
D
20 = +3.9 (c 0.5, MeOH)}. The spectroscopic data for this com-
pound were in agreement with those of its racemate.
4.5.2. Preparative scale hydrolysis of rac-3 by using an
immobilized enzyme in a semi-continuous flow system
0.5 g of lipase AK-IM was packed into a PEEK column which was
then connected to a pump. Phosphate buffer solution (50 mM, pH
7) was circulated through the column at a flow rate of 4 mL/min
prior to use. A solution of rac-3 (0.20 g, 0.91 mmol) in a 9:1 mixture
(4 mL) of phosphate buffer (50 mM, pH 7) and MTBE was passed
through the column reactor at 25 °C. The reaction was monitored
by HPLC by taking samples at different intervals of time and
stopped at an approximately 50% conversion. After 24 h, the col-
umn was washed with 5 ml of a 9:1 mixture of potassium phos-
phate buffer (50 mM, pH 7) and MTBE and then with 5 ml of
potassium phosphate buffer. The eluted solution and washings
were extracted with dichloromethane. The organic layer was dried
over sodium sulfate, concentrated at reduced pressure and anal-
ysed by HPLC as described in Section 4.2. Purification of the residue
by silica gel chromatography gave (R)-1 (40% yield) with >99.5% ee
and (S)-3 (47% yield) with 98% ee.
Please cite this article in press as: Tofani, G.; et al. Tetrahedron: Asymmetry (2015), http://dx.doi.org/10.1016/j.tetasy.2015.04.009

6
G. Tofani et al. / Tetrahedron: Asymmetry xxx (2015) xxx–xxx
Hayes, G. A.; Birch, P. J. J. Med. Chem. 1994, 37, 2138–2144; (d) Métro, T. X.;
Acknowledgements
Pardo, D. G.; Cossy, J. Synlett 2007, 2888–2890; (e) Pienaar, D. P.; Mitra, R. K.;
Van Deventer, T. I.; Botes, A. L. Tetrahedron Lett. 2008, 49, 6752–6755; (f) Aga,
M. A.; Kumar, B.; Rouf, A.; Shah, B. A.; Andotra, S. S.; Taneja, S. C. Helv. Chim.
Acta 2013, 96, 969–977.
This study was supported by research funding from Università
di Pisa, Pisa, Italy and by Studio di Consulenza Scientifica
(SCSOP), Sirtori (LC), Italy.
10. (a) Yamada-Onodera, K.; Fukui, M.; Tani, Y. J. Biosci. Bioeng. 2007, 103, 174–
178; (b) Kizaki, N.; Yasohara, Y.; Nagashima, N.; Hasegawa, J. J. Mol. Catal. B:
Enzym. 2008, 51, 73–80; (c) Lacheretz, R.; Pardo, D. G.; Cossy, J. Org. Lett. 2009,
11, 1245–1248; (d) Pham, S. Q.; Pompidor, G.; Liu, J.; Li, X. D.; Li, Z. Chem.
Commun. 2012, 4618–4620.
References
11. (a) Bornscheuer, U. T.; Kazlauskas, R. J. Hydrolases in Organic Synthesis: Regio-
and Stereoselective Biotransformations, 2nd ed.; Wiley-VCH: Weinheim, 2006;
(b) Gotor-Fernández, V.; Gotor, V. Asymmetric Organic Synthesis with Enzymes;
Wiley-VCH: Weinheim, 2008; (c) Ahmed, M.; Kelly, T.; Ghanem, A. Tetrahedron
2012, 68, 6781–6802.
12. Liese, A.; Seelbach, K.; Buchholz, A.; Haberland, J. Industrial Biotransformations;
Wiley-VCH: Weinheim, 2006.
13. Tomori, H.; Shibutani, K.; Ogura, K. Bull. Chem. Soc. Jpn. 1996, 69, 207–215.
14. Hasegawa, J.; Sawa, I.; Mori, N.; Kutsuki, H.; Ohashi, T. Enantiomer 1997, 2,
311–314.
15. Brossat, M.; Moody, T. S.; Taylor, S. J. C.; Wiffen, J. W. Tetrahedron: Asymmetry
2009, 20, 2112–2116.
16. Horiguchi, A.; Mochida, K. Biosci. Biotechol. Biochem. 1995, 59, 1287–1290.
17. Lihammar, R.; Millet, R.; Bäckvall, J. E. Adv. Synth. Catal. 2011, 353, 2321–2327.
18. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104,
7294–7299.
1. (a) Lawrence, S. A. Amines: Synthesis Properties and Applications; Cambridge
University Press: Cambridge, 2006; (b) Pyne, S. G.; Tang, M. Curr. Org. Chem.
2005, 9, 1393–1418.
2. Busto, E.; Gotor-Fernández, V.; Gotor, V. Chem. Rev. 2011, 111, 3998–4035.
3. Merli, V.; Canavesi, A.; Daverio, P. WO 2,007,076,158, 2007.
4. Hasseberg, H.-A.; Gerlach, H. Liebigs Ann. Chem. 1989, 255–261.
5. Tamazawa, K.; Arima, H.; Kojima, T.; Isomura, Y.; Okada, M.; Fujita, S.; Furuya,
T.; Takenaka, T.; Inagaki, O.; Terai, M. J. Med. Chem. 1986, 29, 2504–2511.
6. Yoshida, K.; Nakayama, K.; Ohtsuka, M.; Kuru, N.; Yokomizo, Y.; Sakamoto, A.;
Takemura, M.; Hoshino, K.; Kanda, H.; Nitanai, H.; Namba, K.; Yoshida, K.;
Imamura, Y.; Zhang, J. Z.; Ving, J. L.; Watkins, W. J. Bioorg. Med. Chem. 2007, 15,
7087–7097.
7. Nagel, U.; Nedden, H. G. Chem. Ber. 1997, 130, 385–397.
8. Xie, X.; Chen, Y.; Ma, D. J. Am. Chem. Soc. 2006, 128, 16050.
9. (a) Muto, K.; Kuroda, T.; Kawato, H.; Karasawa, A.; Kubo, K.; Nakazimo, N.
Arzneim.-Forsch. 1988, 38, 1662–1665; (b) Morimoto, M.; Sakai, K. Tetrahedron:
Asymmetry 2008, 19, 1465–1469; (c) Naylor, A.; Judd, D. B.; Scopes, D. I. C.;
Please cite this article in press as: Tofani, G.; et al. Tetrahedron: Asymmetry (2015), http://dx.doi.org/10.1016/j.tetasy.2015.04.009