Enzyme-catalyzed kinetic resolution of N-Boc-trans-3-hydroxy-4- phenylpyrrolidine

Article abstract of DOI:10.2478/s11532-013-0347-8

The first enzyme-catalyzed kinetic resolution of tert-butyl-3-hydroxy-4- phenylpyrrolidine-1-carboxylate is presented. Enzyme, solvent and temperature optimization resulted in a new resolution method with E = 40 enantioselectivity. The acetate derivative of the (+)-(3S,4R) enantiomer formed while the (-)-(3R,4S) isomer remained intact. Very good enantioselectivities (E > 200) were achieved in the enzyme-catalyzed alcoholysis of the racemic acetate in i-propanol and t-butanol where the (+)-(3S,4R) enantiomer was prepared in pure form (ee > 99.7%). Absolute configuration of the (-)-(3R,4S)-enantiomer was determined by single crystal X-ray diffraction method. [Figure not available: see fulltext.]

Full text of DOI:10.2478/s11532-013-0347-8

&HQWꢀꢁ(XUꢀꢁ-ꢀꢁ&KHPꢀꢁľꢁꢂꢃꢄꢂꢅꢁľꢁꢃꢆꢂꢇꢁľꢁꢃꢈꢉꢊꢃ
'2,ꢋꢁꢂꢆꢀꢃꢇꢌꢍꢎVꢂꢂꢈꢊꢃꢉꢆꢂꢊꢉꢆꢊꢇꢌꢉꢍ
Central European Journal of Chemistry
Enzyme-catalyzed kinetic resolution
of N-Boc-trans-3-hydroxy-4-phenylpyrrolidine
Research Article
Ferenc Faigl1,2*, Ervin Kovács2, Dóra Balogh1,
Tamás Holczbauer3, Mátyás Czugler3, Béla Simándi4
1Department of Organic Chemistry and Technology,
%XGDSHVWꢁ8QLYHUVLW\ꢁRIꢁ7HFKQRORJ\ꢁDQGꢁ(FRQRPLFVꢄꢁ
H-1111 Budapest, Hungary
2Research Group for Organic Chemical Technology,
Hungarian Academy of Sciences,
H-1111 Budapest, Hungary
ꢂInstitute of Structural Chemistry,
Hungarian Academy of Sciences,
H-1025 Budapest, Hungary
ꢃ'HSDUWPHQWꢁRIꢁ&KHPLFDOꢁDQGꢁ(QYLURQPHQWDOꢁ3URFHVVꢁ(QJLQHHULQJꢄꢁ
%XGDSHVWꢁ8QLYHUVLW\ꢁRIꢁ7HFKQRORJ\ꢁDQGꢁ(FRQRPLFVꢄꢁ
H-1111 Budapest, Hungary
Received 19 July 2013; Accepted 18 August 2013
Abstract:ꢀꢀ
7KHꢀğUVWꢀHQ]\PHꢁFDWDO\]HGꢀNLQHWLFꢀUHVROXWLRQꢀRIꢀtertꢁEXW\OꢁꢂꢁK\GUR[\ꢁꢃꢁSKHQ\OS\UUROLGLQHꢁꢄꢁFDUER[\ODWHꢀLVꢀSUHVHQWHGꢅꢀ(Q]\PHꢆꢀVROYHQWꢀ
DQGꢀ WHPSHUDWXUHꢀ RSWLPL]DWLRQꢀ UHVXOWHGꢀ LQꢀ Dꢀ QHZꢀ UHVROXWLRQꢀ PHWKRGꢀ ZLWKꢀ (ꢀ ꢀ ꢃꢇꢀ HQDQWLRVHOHFWLYLW\ꢅꢀ 7KHꢀ DFHWDWHꢀ GHULYDWLYHꢀ RIꢀ WKHꢀ
ꢈꢉꢊꢁꢈꢂSꢆꢃRꢊꢀHQDQWLRPHUꢀIRUPHGꢀZKLOHꢀWKHꢀꢈꢁꢊꢁꢈꢂ5ꢆꢃ6ꢊꢀLVRPHUꢀUHPDLQHGꢀLQWDFWꢅꢀ9HU\ꢀJRRGꢀHQDQWLRVHOHFWLYLWLHVꢀꢈ(ꢀ!ꢀꢋꢇꢇꢊꢀZHUHꢀDFKLHYHGꢀLQꢀ
WKHꢀHQ]\PHꢁFDWDO\]HGꢀDOFRKRO\VLVꢀRIꢀWKHꢀUDFHPLFꢀDFHWDWHꢀLQꢀiꢁSURSDQROꢀDQGꢀtꢁEXWDQROꢀZKHUHꢀWKHꢀꢈꢉꢊꢁꢈꢂSꢆꢃRꢊꢀHQDQWLRPHUꢀZDVꢀSUHSDUHGꢀLQꢀ
SXUHꢀIRUPꢀꢈHHꢀ!ꢀꢌꢌꢅꢍꢎꢊꢅꢀ$EVROXWHꢀFRQğJXUDWLRQꢀRIꢀWKHꢀꢈꢁꢊꢁꢈꢂRꢆꢃSꢊꢁHQDQWLRPHUꢀZDVꢀGHWHUPLQHGꢀE\ꢀVLQJOHꢀFU\VWDOꢀ;ꢁUD\ꢀGLIIUDFWLRQꢀPHWKRGꢅ
Keywords: (Q]\PHꢀFDWDO\]HGꢁUHVROXWLRQꢁľꢁ/LSDVHꢁHQ]\PHVꢁľꢁꢂꢀK\GUR[\ꢀꢃꢀSKHQ\OS\UUROLGLQHꢁľꢁ(Q]\PDWLFꢁK\GURO\VLVꢁľꢁ(QDQWLRVHOHFWLYHꢁDOFRKRO\VLV
© Versita Sp. z o.o.
the enantioselective synthesis of 3,4-disubstituted
pyrrolidines are noticeably rare. Recently, tert-butyl-
1. Introduction
3-hydroxy-4-phenylpyrrolidine-1-carboxylate
(N-Boc-
Substituted pyrrolidines are common structural subunits
found in a variety of natural [1] and synthetic [2] bioactive
compounds. Depending on the substitution pattern
and functionality, pyrrolidines have been shown to be
effective antibacterial [3], antiviral and anticancer agents
[4], protein kinase C inhibitors [5], etc. In addition,
enantiomerically pure pyrrolidines are known as useful
chiral auxiliaries [6,7]. Hence, preparation of enantiopure
trans-3-hydroxy-4-phenylpyrrolidine), 1) was described
as an intermediate of a potential neurokinin-1 (NK1)
receptor antagonist, which could be effective for
treatment of emesis, depression and anxiety [9].
However, the enantiomers of 1 could only be separated
by high performance liquid chromatography using chiral
stationary phase containing columns [9,10].
Since the enantiomers of 1 are of particular interest,
a search for improved methods for their preparation is
warranted. There are numerous publications dealing
with enzymatic resolution of secondary alcohols
[11-13] in the presence of lipases. Compound 1 contains
polysubstituted pyrrolidine derivatives is quite
challenging subject.
a
A
number of stereoselective methods for the
synthesis of 2,5-disubstituted pyrrolidines have been
reported [8], but general and effective methods for
* E-mail: ffaigl@mail.bme.hu
25

Enzyme-catalyzed kinetic resolution
of N-Boc-trans-3-hydroxy-4-phenylpyrrolidine
a hydroxylic group connected to one of the asymmetric into it and the phases were separated. The organic phase
centers; therefore, we used it as a function for resolution was washed with brine (3×80 mL), dried over sodium
viaꢀHQ]\PHꢁFDWDO\]HGꢀWUDQVHVWHUL¿FDWLRQꢂ
sulfate, and it was concentrated in vacuum. The residue
contained the crude N-Boc-pyrrolidine (3, 59.0 g). It was
XVHGꢀLQꢀWKHꢀQH[WꢀVWHSꢀZLWKRXWꢀDQ\ꢀSXUL¿FDWLRQꢂꢀ
1H NMR (500 MHz, CDCl3), dH (ppm): 5.77 (d, 2H,
2. Experimental procedure
J = 17.0 Hz), 4.11 (d, 4H, J = 22.5 Hz), 1.48 (s, 9H). 13
C
NMR (125 MHz, CDCl3) dC (ppm): 154.3, 125.9, 125.8,
ꢈꢉꢂꢊꢄꢀꢋꢊꢂꢌꢄꢀꢋꢆꢂꢍꢀꢆꢍꢂꢋꢂꢀ,5ꢀꢃ¿OPꢅꢀQmax (cm-1): 2978, 1742,
1699, 1341, 1317.
2.1. General
All commercial starting materials were purchased from
Sigma-Aldrich Hungary Ltd. and Merck Hungary Ltd.
DQGꢀZHUHꢀXVHGꢀZLWKRXWꢀIXUWKHUꢀSXUL¿FDWLRQꢂ
N-Boc-3,4-epoxypyrrolidine (4) [10]
Routine 1H NMR and 13C NMR spectra were recorded
in CDCl3 solution on a BRUKER AV 300 or DRX 500
spectrometer. Proton and carbon chemical shifts are
reported in ppm relative to tetramethylsilane (dTMS
= 0.00 ppm) or to the solvent (dCDCl3 = 77.00 ppm),
respectively. IR spectra were recorded on an appliance
type PERKIN ELMER 1600 with a Fourier Transformer.
Data are given in cm–1ꢂꢀ6SHFL¿FꢀURWDWLRQꢀRIꢀWKHꢀRSWLFDOO\ꢀ
active samples were determined on a PERKIN ELMER
245 MC polarimeter using sodium lamp (589 nm).
Alufolien Kieselgel 60 F254 (Merck) plates were used
for TLC investigations and the spots were visualized with
uv light and/or an aqueous solution of (NH4)6Mo7O24,
Ce(SO4)2 and sulfuric acid. Flash chromatography
ZDVꢀ DSSOLHGꢀ IRUꢀ WKHꢀ VHSDUDWLRQꢀ DQGꢀ SXUL¿FDWLRQꢀ RIꢀ WKHꢀ
products using normal phase RediFlash Rf columns.
Eluents are given for each experiment.
A dichloromethane solution (15 mL) of the crude
N-Boc-pyrroline (3, 2.0 g, 11.6 mmol) was cooled
down to 0°C and the dichloromethane solution
(45 mL) of m-chloroperbenzoic acid (70% content,
5.73 g, 23.2 mmol) was added into it within 2.5 hours.
The reaction mixture was stirred at ambient temperature
for 4 days,it was cooled down again to 0°C, and the
precipitated mꢁFKORUREHQ]RLFꢀDFLGꢀZDVꢀ¿OWHUHGꢀRIIꢂꢀ7KHꢀ
¿OWUDWHꢀ ZDVꢀ ZDVKHGꢀ ZLWKꢀ VDWXUDWHGꢀ DTXHRXVꢀ VRGLXPꢀ
thiosulfate solution (100 mL, then 3×30 mL), with
saturated aqueous sodium hydrogencarbonate solution
(60 mL) and with saturated potassium carbonate
solution (3×50 mL) until the peroxide test became
negative (Merckoquant 1.10011.0001 peroxide test).
The solution was washed with brine (40 mL) and dried
over sodium sulfate before concentration in vacuum.
The crude product (4ꢄꢀꢆꢂꢎꢋꢀJꢅꢀZDVꢀSXUL¿HGꢀE\ꢀFROXPQꢀ
chromatography (eluent: hexane/ethyl acetate=4/1 then
2/1) to yield pure 4 (1.30 g, 72%).
1H NMR (500 MHz, CDCl3) dH (ppm): 3.77 (dd, 2H,
J1 = 36.5, J2 = 13.0 Hz), 3.66 (d, 2H, J = 5.0 Hz), 3.31
(dd, 2H, J1 = 13.0, J2 = 5.0 Hz), 1.44 (s, 9H). 13C NMR
(125 MHz, CDCl3) dC (ppm): 154.8, 79.8, 55.6, 55.1,
ꢎꢈꢂꢊꢄꢀꢎꢏꢂꢉꢄꢀꢆꢍꢂꢎꢂꢀ,5ꢀꢃ¿OPꢅꢀQmax (cm-1): 3051, 2976, 2876,
1698, 1423, 1389, 1338, 1174, 1116, 1027, 964.
Racemic N-Boc-3-hydroxy-4-phenylpyrrolidine
(1) [10,11]
Compound 4 (1.40 g, 7.56 mmol) was dissolved in
dry tetrahydrofuran (20 mL) under nitrogen atmosphere,
cuprous iodide (0.108 g, 0.567 mmol) was added into
it and the solution was cooled to 0°C. Tetrahydrofuran
solution of phenylmagnesium chloride (2 mol L-1
solution, 4.35 mL, 8.69 mmol), was added dropwise and
the mixture was stirred at 0°C for two hours then it was
allowed to warm up to ambient temperature and stirred
for overnight. The dark reaction mixture was poured
into ice water and aqueous hydrochloric acid (1.5 molar
solution, 3.8 mL) was added into it followed by the
addition of diethyl ether (38 mL). The organic phase was
separated and the aqueous phase was washed with
diethyl ether (3×30 mL). The tetrahydrofuran content of
the aqueous solution was distilled off in vacuum then it
GC analysis were performed on an Agilent
4890 D equipment using ALPHA DEXTM 120 (30 m,
0.25 mm/0.25 mm) capillary column at 190°C.
HRMS analyses were performed on a LTQ FT
8OWUDꢀ ꢃ7KHUPRꢀ )LVKHUꢀ 6FLHQWL¿Fꢄꢀ %UHPHQꢄꢀ *HUPDQ\ꢅꢀ
system. The ionization method was ESI and operated
in positive ion mode. For CID, experiment helium was
used as the collision gas, and normalized collision
energy (expressed in percentage) was used to bring
about fragmentation. The protonated molecular ion
peaks were fragmented by CID at a normalized collision
energy of 35%. The samples were solved in methanol.
Data acquisition and analysis were accomplished with
;FDOLEXUꢀVRIWZDUHꢀYHUVLRQꢀꢆꢂꢇꢀꢃ7KHUPRꢀ)LVKHUꢀ6FLHQWL¿Fꢅꢂ
2.2. Preparation of rac-1 and rac-5
N-Boc-pyrroline (3) [10]
A 65/35 mixture of pyrroline and pyrrolidine (25.0 g,
0.235 mol) was dissolved in dichloromethane (500 mL).
Di-tert-butyl-dicarbonate (75.8 g, 0.347 mol) was also
dissolved in dichloromethane (250 mL) and the two
precooled solutions were mixed under continuous
cooling at 0-5°C. The reaction mixture was stirred
for two hours at ambient temperature then aqueous
hydrochloric acid solution (1 molar, 200 mL) was poured
26

F. Faigl et al.
was extracted with ethyl acetate ( 2×38 mL). This organic
Kinetic resolution of rac-1 under optimized
solution was dried over sodium sulfate and concentrated conditions
in vacuum to yield the crude product (1, 2.35 g). The pure
Racemic
N-Boc-3-hydroxy-4-phenylpyrrolidine
product was obtained in 88% yield (1.75 g) after column (1, 0.80 g, 3.04 mmol) was dissolved in the mixture
chromatography (eluent hexane/ethyl acetate=4/1). Mp: of methyl tert-butyl ether (40.0 mL) and vinyl acetate
72-74°C (from hexane/ethyl acetate).
(15.0 mL, 163 mmol), and Novozym 435 enzyme
1H NMR (300 MHz, CDCl3) dH (ppm): 7.33 (m, 2H), (1.60 g) was added into it. The reaction mixture was
7.25 (m, 3H), 4.28 (q, 1H, J = 6.0 Hz), 3.83 (m, 1H), 3.71 stirred at 45°C and samples were withdrawn time to
(m, 1H), 3.49 (m, 1H), 3.25 (m, 2H), 2.91 (br, 1H), 1.47 time. Samples were analysed by gas chromatography.
(s, 9H). 13C NMR (75 MHz, CDCl3) dC (ppm): 154.5, When the half of the starting material consumed (at
139.3, 128.8, 127.4, 127.2, 79.7, 75.9, 55.3, 52.1, 51.4, ꢋꢎꢐꢀFRQYHUVLRQꢅꢄꢀWKHꢀHQ]\PHꢀZDVꢀ¿OWHUHGꢀRIIꢄꢀZDVKHGꢀ
ꢆꢍꢂꢋꢂꢀ,5ꢀꢃ¿OPꢅꢀQmax (cm-1): 3420, 2975, 1678, 1422, 1134, ZLWKꢀDFHWRQHꢀꢃꢊîꢋꢀP/ꢅꢀDQGꢀWKHꢀ¿OWUDWHꢀZDVꢀFRQFHQWUDWHGꢀ
770, 699, 585.
in vacuum. The residue (1.06 g) contained the formed
Racemic N-Boc-3-acetoxy-4-phenylpyrrolidine (5) (+)-(3S,4R)-5 ester and the unreacted (-)-(3R,4S)-1.
Compound 1 (0.200 g, 0.698 mmol) was dissolved These compounds were separated from their mixture
in dry pyridine (3 mL) under nitrogen atmosphere and E\ꢀÀDVKꢀFKURPDWRJUDSK\ꢀꢃHOXHQWꢀKH[DQHꢑHWK\OꢀDFHWDWHꢀ
the solution was cooled down to -10°C. Acetic anhydride ꢀ ꢌꢑꢇꢀ ĺꢀ ꢊꢑꢆꢅꢂꢀ7KXVꢀ SXUHꢀ ꢃꢁꢅꢁꢃꢊR,4S)-1 (0.377 g, 94%
0.400 mL, 4.20 mmol) was added and the reaction yield, calculated to the half of the starting racemate) and
mixture was stirred at -10°C for two hours and at 25°C (+)-(3S,4R)-5 (0.542 g, 103% yield, calculated to the
for one day. Then it was poured into a mixture of ice half of the starting racemate) were isolated. Retention
(about 5 g), aqueous sulfuric acid (1 mol L-1, 30 mL) WLPHVꢀLQꢀ*&ꢄꢀVSHFL¿FꢀURWDWLRQꢀSRZHUVꢀDQGꢀHHꢀYDOXHVꢀRIꢀ
and diethyl ether (20 mL). The phases were separated the products are given below.
and the aqueous phase was extracted with diethyl ether
(-)-(3R,4S)-1:
t
R(GC)=97.7 min (for comparison
(3×20 mL), then the ethereal solution was washed (+)-1: tR(GC)=98.3 min);
-35.5 (c=1.94; CHCl3);
with brine (20 mL), dried over sodium sulfate and ee=97.3%.
concentrated in vacuum. The crude product (0.240 g,
(+)-(3S,4R)-5: tR(GC)=69.0 min. +16.1 (c=2.33;
\HOORZLVKꢀ RLOꢅꢀ ZDVꢀ SXUL¿HGꢀ E\ꢀ FROXPQꢀ FKURPDWRJUDSK\ꢀ CHCl3); ee=78.0%.
(eluent hexane/ethyl acetate=4/1) to yield pure 5
2.4. Hydrolysis/alcoholysis of optically active
N-Boc-3-acetoxy-4-phenylpyrrolidine
((+)-(3S,4R)-5)
(0.225 g, 97%).
1H NMR (300 MHz, CDCl3) dH (ppm): 7.31 (m, 5H),
5.15 (m, 1H), 3.79 (m, 3H), 3.46 (m, 1H), 3.36 (m, 1H),
2.08 (s, 3H), 1.50 (s, 9H). 13C NMR (75 MHz, CDCl3) dC Hydrolysis of (+)-(3S,4R)-5
(ppm): 170.8, 154.5, 139.1, 129.1, 127.6, 127.4, 80.1, Aqueous sodium hydroxide (10% solution, 1.0 mL)
ꢈꢈꢂꢋꢄꢀꢋꢌꢂꢊꢄꢀꢎꢉꢂꢉꢄꢀꢎꢍꢂꢊꢄꢀꢆꢍꢂꢈꢄꢀꢆꢌꢂꢊꢂꢀ,5ꢀꢃ¿OPꢅꢀQmax (cm-1): was added to the solution of (+)-(3S,4R)-5 (0.20 g,
2972, 1737, 1699, 1403, 701. The adduct of protonated ee = 72.6%) in methanol (5.0 mL) and the mixture was
molecular ion peak [2M+H]+ can be detected at m/z stirred for 72 hours. Then it was diluted with water
611. HRMS: 611.33281 (C34H47O8N2; calc. 611.33269; (20 mL) and extracted with ethyl acetate (4×10 mL). The
delta=0.2ppm). ESI-MS-MS (611.33@cid35) (rel. int. ethyl acetate solution was dried over sodium sulfate and
%, element composition): 555(2, [2M+H]-C4H8); 511(2, concentrated in vacuum to yield (+)-(3S,4R)-1 (0.15 g,
[2M+H]-C5H8O2); 306(70, C17H24NO4); 250(100, [2M+H]- 87%, ee=72.6% (GC)).
M-C4H8); 206(5, [2M+H]-M-C5H8O2); 190(18, C11H12O2N).
Enzyme catalyzed alcoholysis of rac-5. General
procedure
2.3. Kinetic resolution of N-Boc-3-hydroxy-4-
phenylpyrrolidine (rac-1)
Small-scale kinetic resolution of rac-1
Racemic 5 (0.010 g, 0.033 mmol) was dissolved in the
corresponding solvent (see in Table 4, 0.30 mL) and
Novozym 435 (0.040 g) was added into it.Then the
mixture was shaked for 8 days. Compositions of the
reaction mixtures were determined by GC. The results
are summarized in Tables 4 and 5.
Enzyme catalyzed hydrolysis of rac-5. General
procedure
Racemic 5 (0.010 g, 0.033 mmol) was dissolved in
the corresponding solvent (see in Table 4, 0.12 mL) and
Racemic
N-Boc-3-hydroxy-4-phenylpyrrolidine
(1,
10 mg, 38 μmol) was dissolved in the mixture of the
solvent (0.5 mL) and vinyl acetate (0.2 mL, 2.2 mmol)
and the selected enzyme (20 mg) was added into it. The
reaction mixture was shaked at 30°C and samples were
withdrawn time to time. Samples were analysed by gas
chromatography.
27

Enzyme-catalyzed kinetic resolution
of N-Boc-trans-3-hydroxy-4-phenylpyrrolidine
RIꢀ SDUDPHWHUVꢀ ꢀ ꢌꢈꢉꢄꢀ JRRGQHVVꢁRIꢁ¿Wꢀ ꢀ ꢌꢂꢇꢊꢊꢄꢀ WKHꢀ
maximum and mean shift/esd is 0.000 and 0.000). The
maximum and minimum residual electron density in the
H
O
N
O
oc
m
3
4
1
N
B
2O
PBA
PhM
l
C
g
C
2
5
u
C
I
N
H
N
-3
¿QDOꢀGLIIHUHQFHꢀPDSꢀZDVꢀꢇꢂꢌꢆꢇꢀDQGꢀꢁꢇꢂꢇꢉꢉHꢂc . Absolute
O
O
O
O
O
O
FRQ¿JXUDWLRQꢀFU\VWDOꢀGDWDꢓꢀ)ULHGHOꢀ3DLUꢀ&RYHUDJHꢀ ꢀꢉꢋꢐꢄꢀ
Flack parameter: 0.2(2), Hooft parameter: 0.19(9).
CCDC 921454 contains the supplementary
crystallographic data for this Sructure. These data
can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
rac-
2
1
3
4
Scheme 1. Synthesis of rac-1.
O
H
O
O
Ph
Ph
c
A
2O
ne
r
N
idi
N
py
O
O
O
O
3. Results and discussion
rac-
5
rac-
1
3.1. Synthesis of rac-1 and rac-5
The starting material was prepared according to
Scheme 2. Synthesis of rac-5.
a phosphate buffer (pH=7.0, 0.12 mL) then Novozym the literature method [10] from pyrroline (2) via Boc
435 (0.040 g) was added into it. The mixture was protection (intermediate 3), epoxidation (intermediate
shaken at room temperature for 8 days. The mixture 4) and copper catalyzed phenylmagnesium bromide
ZDVꢀ ¿OWHUHGꢄꢀ GLOXWHGꢀ ZLWKꢀ GLVWLOOHGꢀ ZDWHUꢀ ꢃꢌꢂꢇꢀ P/ꢅꢀ DQGꢀ addition (Scheme 1). It is worth mentioning that
extracted with ethyl acetate (4×2.5 mL). Compositions commercial pyrroline (2) contained pyrrolidine, which
of the ethyl acetate solutions were determined by GC. can easily be separated from the main product after the
The results are summarized in Table 4.
epoxidation step. The epoxidation step provided cis-4
and addition of the Grignard reagent was completely
diastereoselective, thus racemic trans-1 has formed in
2.5. Crystal structure determination
Single crystals of (-)-(3R,4S)-tert-butyl-3-hydroxy-4- the reaction as sole product, according to the results of
phenylpyrrolidine-1-carboxylate ((-)-1) were obtained gas chromatographic and spectroscopic investigations.
from ethyl acetate, mp= 88-90 °C.
The racemic acetate of 1 was prepared in good
Crystal data: C15H21NO3, Fwt.: 263.33, colourless, yield using acetic anhydride as acylating agent (5,
bladed, size: 0.75×0.25×0.01 mm, monoclinic, Scheme 2). Ester 5 was used as a standard for analytical
space group P 21, a = 9.0787(6) Å, b = 5.8006(4) Å, measurements and as a starting material of enzyme-
c = 14.1226(9)Å, D = 90.00°, E = 94.054(3)°, J = 90.00°, catalyzed enantioselective alcoholysis and hydrolysis
V = 741.86(9) Å3, T = 295(2) K, Z= 2, F(000) = 284, experiments.
Dx = 1.179 Mg m-3, P = 0.660 mm-1.
A crystal was mounted on a loop. Cell parameters 3.2. Enzyme catalyzed kinetic resolution of rac-1
were determined by least-squares using 8392 Enantioselectivity
of
the
enzyme-catalyzed
(7.64dTd71.82oꢅꢀUHÀHFWLRQVꢂ
WUDQVHVWHUL¿FDWLRQꢀ RIꢀ UDFHPLFꢀ 1 was screened with a
Intensity data were collected on a Rigaku R-AXIS- wide selection of enzymes [AK, A, F-AP15, M and PS
RAPID diffractometer (graphite monochromator Cu- lipases from Amano; Novozym 435, Liposime TL IM
KDꢀradiation, O = 1.54187 Å) at 295(2) K in the range and CAL-A from Novozymes; porcine liver esterase
7.69dTdꢋꢍꢂꢉꢌꢂꢀ$ꢀWRWDOꢀRIꢀꢍꢍꢌꢎꢀUHÀHFWLRQVꢀZHUHꢀFROOHFWHGꢀ (liver acetone powder), Candida cylindracea lipase
of which 2043 were unique [R(int) =0.0243, R(V) and 3VHXGRPRQDVꢀ ÀXRUHVFHQV lipases from Fluka;
ꢇꢂꢇꢆꢆꢎ@ꢒꢀ LQWHQVLWLHVꢀ RIꢀ ꢌꢍꢆꢏꢀ UHÀHFWLRQVꢀ ZHUHꢀ JUHDWHUꢀ papain, Candida cylindracae (Type VII, powder) and
than 2V(I). Completeness to T = 0.972.
porcine pancreatic lipases(Type II, crude powder) from
A numerical absorption correction was applied to the Sigma and BUTE-3 lipase (crude, precipitated with
data (the minimum and maximum transmission factors acetone), isolated at our university from thermophilic
were 0.715 and 0.973).
¿ODPHQWRXVꢀ fungi [12]. Vinyl acetate was used as
The structure was solved by direct methods [17]. DQꢀ DF\ODWLQJꢀ UHDJHQWꢀ DQGꢀ WKHꢀ WUDQVHVWHUL¿FDWLRQꢀ ZDVꢀ
$QLVRWURSLFꢀ IXOOꢁPDWUL[ꢀ OHDVWꢁVTXDUHVꢀ UH¿QHPHQWꢀ >17] carried out in tetrahydrofuran in the test reactions
on F2 for all non-hydrogen atoms yielded R1 = 0.0328 (Scheme 3).
and wR2 = 0.0800 for 1332 [I>2s(I)] and R1 = 0.0364
and wR2 = 0.0823 for all (2043) intensity data, (number
Gas chromatographic analyses performed on a
chiral stationary phase containing capillary column
28

F. Faigl et al.
Table 1. Solvent dependent kinetic resolution of rac-1 in the presence BUTE-3 enzyme.
Solvent
Acylating agenta
Conversion (%)b
Ee of (3R,4S)-1 (%)c
Ed
Tetrahydrofuran
Methyl tert-butyl ether
Diisopropyl ether
Toluene
vinyl acetate
vinyl acetate
vinyl acetate
vinyl acetate
vinyl acetate
vinyl acetate
vinyl acetate
vinyl acetate
ethyl acetate
16
65
69
54
68
48
22
46
8
10.0
94.4
>99.0
75.3
96.4
50.5
16.7
53.0
5.1
4
10
14
10
9
Hexane
Hexane/tetrahydrofuran=2/1
Acetone
6
5
Ethyl acetate
8
Ethyl acetate
4
a Details of the small scale experiments are given in the experimental part.
b Conversions were determined after 72 hours reactions by GC analysis (details are given in the experimental section) and calculated to the whole amount
of rac-1 as follows c: conversion, x1 and x5: mol% of 1 and 5.
c Ee values were determined by GC on chiral stationary phase containing capillary column.
d Enantioselectivity (E) of the resolution was calculated according to the literature standards [14,16]
Table 2. Solvent dependent kinetic resolution of rac-1 in the
O
presence Novozym 435 enzymea.
O
Ph
H
HO
Ph
O
O
O
Ph
O
Solvent
Conversion
(%)b
Ee of
(3R,4S)-1
(%)c
Ed
N
N
N
enz me
y
,
THF
O
O
O
O
O
-
H
C
H
C O
(
)
3
Tetrahydrofuran
35
60
42.6
94.3
12
15
Methyl tert-butyl
ether
rac-
- -
-
1
+ -
-
5
4R
S,
)
1
3R 4S
3
,
( ) (
)
( ) (
Scheme 3. Enzyme-catalyzed kinetic resolution of rac-1.
Diisopropyl ether
Toluene
59
42
60
91.1
51.6
89.3
13
10
11
Catalytic activities and enantioselectivities of the
enzymes strongly depend on the solvent therefore a
series of experiments were carried out with the above
mentioned two enzymes in different solvents and
solvent mixtures. Each reaction was accomplished at
30°C, under the same conditions (for details, see the
experimental section). Samples were withdrawn time
to time and the conversion (acetate formation) together
with the ee values of the residual alcohol 1 were
determined by gas chromatography. The results are
summarized in Tables 1 and 2.
Hexane
Hexane/
tetrahydrofuran=2/1
44
60.2
13
Acetone
35
55
38.1
82.8
9
Ethyl acetate
13
a Details of the small scale experiments are given in the experimental part.
b Conversions were determined after 72 hours reactions by GC analysis.
Calculation is explained in the footnote of Table 1.
c
Ee values were determined by GC on chiral stationary phase containing
capillary column.
d Calculation method is given in Table 1 foot-note d.
We achieved the best enantioselectivity in
helped us in monitoring the enzymatic reactions. The diisopropyl ether with BUTE-3 lipase. In this reaction,
starting racemate was completely pure trans-1, and the unreacted (3R,4S)-1 alcohol could be isolated in
ZHꢀGLGꢀQRWꢀ¿QGꢀDQ\ꢀLVRPHUL]DWLRQꢀSURGXFWꢀꢃcis-1 or cis- practically enantiopure form (31% yield) after a 72 hour
5 isomer) during the resolution processes according to reaction. Comparison of the results obtained in ethyl
the GC investigations.
acetate/vinyl acetate mixture and pure ethyl acetate,
Consequently, the diastereoisomeric excess (de) demonstrate the importance of the quasy irreversible
was 100% in each case. Comparison of the yields of the WUDQVHVWHUL¿FDWLRQꢀ ꢃZLWKꢀ YLQ\Oꢀ DFHWDWHꢅꢂꢀ 0XFKꢀ ORZHUꢀ
produced acetate 5 and the residual alcohol 1 and ee yield and enantioselectivity could be achieved during
values of 1 lead us to conclude that the Novozyme 435 WUDQVHVWHUL¿FDWLRQꢀLQꢀSXUHꢀHWK\OꢀDFHWDWHꢀ>13].
and BUTE-3 were the most active and enantioselective
catalysts. Therefore, further optimization experiments UHVROXWLRQꢀ ZDVꢀ PRUHꢀ HI¿FLHQWꢀ LQꢀ DOPRVWꢀ HYHU\ꢀ VROYHQWꢂꢀ
were carried out with these two enzymes.
The best enantioselectivity was observed in methyl tert-
In the presence of Novozym 435, the kinetic
29

Enzyme-catalyzed kinetic resolution
of N-Boc-trans-3-hydroxy-4-phenylpyrrolidine
Table 3. Temperature dependence of the conversion and
enantioselectivity in kinetic resolution of 1.
butyl ether (E=15). The higher enantioselectivity of this
enzyme let us to produce pure (3R,4S)-1 with better
yield than in the previous resolution.
Temperature
(°C)
Conversion
(%)a
E
Temperature dependence of the enzyme activity and
enantioselectivity was also investigated in the presence
of Novozyme 435 enzyme. In these experiments,
methyl tert-butyl ether was used as a solvent and vinyl
acetate was the acylating agent. Yields of (3S,4R)-5
were determined after 24 hours reactions at different
temperatures between 30°C and 55°C. The results are
summarized in Table 3.
30
40
45
50
55
46
55
55
58
58
15
27
27
28
19
a Conversions were determined after 72 hours reactions by GC analysis.
The best enantioselectivity was achieved in the 40-
50°C range and, of course, in a much faster reaction
(24 hours) than it was at 30°C. Scale up of the
enzyme catalyzed reaction at 45°C resulted in further
enhancement of the enantioselectivity (E=40). It is
probably due to the slightly different conditions (stirring
instead of shaking). It means that, in larger scale, under
optimum conditions, (3S,4R)-1 could be prepared in
good yield (94%, calculated to half of racemic 1) and
high ee (97.3%) in methyl tert-butyl ether (E=40). It is
probably due to the slightly different conditions (stirring
instead of shaking). It means that, in larger scale under
optimum conditions, (3R,4S)-1 could be prepared in
good yield (47%) and high ee (97.3%) in methyl tert-
butyl ether.
Table 4. Novozyme 435 catalysed deacetylation of rac-5 in different
solventsa. Results were obtained after 8 days reaction at
room temperature.
Solvent
Conversionb Ee of 1c
Ed
(%)
(%)
Ethanol
8
98
92
90
n-Propanol
0.6
36
10
45
2
24
i-Propanol
99.7
>99.7
>99.7
>99.7
>99.7
>99.
75
>200
>200
>200
>200
>200
>200
7.1
n-Butanol
t-Butanol
n-Heptanol
n-Octanol
2
n-Butanol/buffer=1/1
t-Butanol/buffer=1/1
Acetonitrile/buffer=1/1
Buffer
0.2
1.2
0.6
34
ꢀꢁꢀꢁꢂ'HWHUPLQDWLRQꢂRIꢂWKHꢂDEVROXWHꢂFRQğJXUDWLRQꢂꢂ
of (-)-1
45
2.6
,Qꢀ RUGHUꢀ WRꢀ GHWHUPLQHꢀ WKHꢀ DEVROXWHꢀ FRQ¿JXUDWLRQꢀ RIꢀ
(-)-1, single crystal X-ray diffraction studies were
accomplished. The molecular structure of the pure
(-)-1 enantiomer in the crystals is shown in Fig. 1.
From this structure, one can conclude that the absolute
FRQ¿JXUDWLRQꢀRIꢀWKHꢀꢃꢁꢅꢁ1 enantiomer is (3R,4S)-1.
3.8
1.1
a Details of the small scale experiments are given in the experimental part.
b Conversions and enantiomeric compositions were determined by GC.
Conversion was calculated as follows:
(c: conversion, x1, x5: mol% of 1 and 5, respectively).
c Ee values were determined by GC on chiral stationary phase containing
capillary column.
d
Enantioselectivity of the resolution was calculated according to the
literature standards [14,16]
3.4. Hydrolysis/alcoholysis of rac-5
The (+)-(3S,4R)-1ꢀ LVRPHUꢀ ZDVꢀ SUHSDUHGꢀ ¿UVWꢀ E\ꢀ
hydrolysis of the optically active (+)-(3S,4R)-5 ester
(Scheme 4). However, the enantiomeric purity of (+)-
(3S,4R)-1 strongly depends on the conversion of the
enzyme catalyzed resolution. Therefore, enzyme-
catalyzed alcoholysis of (3S,4R>3R,4S)-5 enantiomeric
mixture was also studied with the hope of preparation
(+)-(3S,4R)-1 in high ee. According to the literature
examples [15], lipase enzymes usually prefer the
hydrolysis or alcoholysis of the faster formed ester
enantiomer ((3S,4R)-5 isomer in our case).
The rate of the enzyme-catalyzed alcoholysis
and hydrolysis was one order of magnitude smaller
than the rate of the acylation reactions. However, the
enantioselectivities were much higher (Table 4). The
best results were achieved in i-propanol and t-butanol.
Figure 1. Ortep style diagram [17] of (-)-(3R,4S)-1 in the single
crystal.
30

F. Faigl et al.
Table 5. Temperature dependence of the conversion and enantioselectivity of the Novozyme 435 catalyzed alcoholysis of rac-5a.
Temperature
(°C)
Conversion in
i-PrOH (%)b
Ec in
i-PrOH
Conversion in
t-BuOH (%)b
Ec in
t-BuOH
30
40
50
60
15
31
34
38
>200
>200
>200
>200
42
45
46
47
>200
>200
>200
>200
a Details of the small scale experiments are given in the experimental part.
b Conversions were determined after 3 days reactions by GC analysis. Calculation is explained in Table 1.
c Calculation method is given in Table 4, footnote 4.
O
H
O
4. Conclusions
O
O
Ph
Ph
a
N OH
It has been found that Novzyme 435 and BUTE-3
lipases are suitable catalysts for the kinetic resolution
of racemic tert-butyl-3-hydroxy-4-phenylpyrrolidine-1-
carboxylate (1). The enzymes initiate the acetylation
of (+)-(3S,4R)-1 while (-)-(3R,4S)-1 remains intact.
The best separation could be achieved with Novozyme
435 using vinyl acetate in methyl tert-butyl ether at
50°C. Under these conditions, the enantioselectivity E
= 40 and (-)-(3S,4R)-1 could be isolated in 94% yield
(ee>97.3%).
N
N
e
H H
O 2O
M
/
O
O
O
+ -
-
5
+ -
-
1
3S 4R
3S 4R
,
,
( ) (
)
( ) (
)
Scheme 4. Hydrolysis of (+)-(3S,4R)-5 with sodium hydroxide.
O
O
H
O
O
O
O
O
Ph
Ph
Ph
-
R
H
O
N
N
N
ovoz m
N
435
y
$EVROXWHꢀ FRQ¿JXUDWLRQꢀ RIꢀ WKHꢀ ꢃꢁꢅꢁꢃꢊR,4S)-1
enantiomer was determined by single crystal Xray
diffraction method because such a measurement was
missing from the literature until now.
O
O
O
O
+ -
-
1
- -
-
5
R 4
S
)
rac-
3
4R
S
,
3
,
5
( ) (
)
( ) (
Scheme 5. Alcoholysis/hydrolysis of rac-5. (R-OH: ethanol,
High enantioselectivities could be achieved during
the alcoholysis of the racemic acetate (rac-5). The
Novozyme 435 catalyzed reaction was the most
selective and the fastest in i-propanol and t-butanol (E >
n-propanol, i-propanol, n-butanol, t-butanol,
n-heptanol, n-octanol, water. Ee data are shown in
Table 4).
The enantioselectivities were very high (E > 200) in both 200). Even though the reactions were quite slow at 30°C,
cases and practically pure (+)-(3S,4R)-1 (ee > 99.7%) 47% conversion could be achieved within 70 hours at
could be isolated from the reaction mixture after 8 days 60°C, while the enantioselectivity remained high (E >
of reaction at 30°C.
200). Thus practically enantiopure (+)-(3R,4S)-1 could
3UHVHQFHꢀ RIꢀ DTXHRXVꢀ EXIIHUꢀ FDXVHGꢀ VLJQL¿FDQWꢀ be prepared (ee = 99.7%).
decreasing of both the reaction rate and the
enantioselectivity in alcohol/buffer mixtures and the
$FNQRZOHGJHPHQWV
hydrolysis was not enantioselective in pure aqueous
buffer solution (Table 4). In order to enhance the rate of
7KHꢀ ZRUNꢀ LVꢀ FRQQHFWHGꢀ WRꢀ WKHꢀ VFLHQWL¿Fꢀ SURJUDPꢀ RIꢀ
the alcoholysis, experiments were carried out at higher
temperatures in i-propanol and t-butanol, too. The
results are summarized in Table 5.
Experimental data show that the enantioselectivity
remains very high at elevated temperature while the
rate of the reaction increases. Thus, 38% and 47%
conversion could be achieved in i-propanol and t-butanol
after 70 hours, respectively. Consequently, t-butanol
seems to be the best reactant and solvent for enzyme-
catalyzed resolution of rac-5.
the “Development of quality-oriented and harmonized
R+D+I strategy and functional model at BME” project in
the framework of the New Hungary Development Plan
(TÁMOP-4.2.1/B-09/1/KMR-2010-0002). The authors
DUHꢀDOVRꢀJUDWHIXOꢀWRꢀWKHꢀ+XQJDULDQꢀ6FLHQWL¿Fꢀ5HVHDUFKꢀ
)XQGꢀ ꢃ27.$ꢀ SURMHFWꢀ QXPEHUꢓꢀ ꢈꢆꢍꢏꢌꢅꢀ IRUꢀ ¿QDQFLDOꢀ
support. The authors are grateful to Dr. Miklós Dékány
(Richter Gedeon plc) for MS measurements.
31

Enzyme-catalyzed kinetic resolution
of N-Boc-trans-3-hydroxy-4-phenylpyrrolidine
References
[1] R.K. Dieter, N. Chen, R.T. Watson, Tetrahedron 61, [10] P. Lin, L. Chang, R.J. DeVita, J.R. Young, R. Eid,
3221 (2005)
X. Tong, S. Zheng, R.G. Ball, N.N. Tsou,
G.G. Chicchi, M.M. Kurtz, K.C. Tsao, A. Wheeldon,
E.J. Carlson, W. Eng, H.D. Burns, R.J. Hargreaves,
S.G. Mills, Bioorg. Med. Chem. Lett. 17, 5191
(2007)
[2] D. Rejman, P. Kocalka, M. Budesinsky, R. Pohl,
I. Rosenberg, Tetrahedron 63, 1243 (2007)
[3] G.B. Evans, A.I. Longshaw, V.L. Schramm,
P.C. Tyler, WO 2011/008110 patent application
(CAN 154:158729)
[4] Se. Lee, Su. Lee, H.J. Park, S.K. Lee, S. Kim, Org.
Biomol. Chem. 9, 4580 (2011)
[5] J.M. Defauw, M.M. Murphy, G.E. Jagdmann,
H. Hu, J.W. Lampe, S.P. Hollinshead, T.J. Mitchell,
[11] R.J. Devita, S.G. Mills, J.R. Young, P. Lin, WO
2005/032464 patent application (CAN 142:373675)
[12] V. Bódai, R. Peredi, J. Bálint, G. Egri, L. Novák,
G. Szakács, L. Poppe, Adv. Synth. Catal. 345, 811
(2003)
H.M. Crane, J.M. Heerding, J.S. Mendoza, [13] V. Gotor, I. Alfonso, E. García-Urdiales, Asymmetric
J.E. Davis, J.W. Darges, J. Med. Chem. 39, 5215
(1996)
Organic Synthesis with Enzymes. (Wiley-VCH,
Weinheim, 2008) 150
[6] P.M. Liu, D.R. Magar, K. Chen, Eur. J. Org. Chem. [14] A.G.M. Sheldrick, Acta Cryst. A64, 112 (2008)
29, 5705 (2010)
[7] P.I. Dalko, L. Moisan, Angew. Chem. Int. Ed. 43,
5138 (2004)
[8] M. Pichon, B. Figadere, Tetrahedron: Asymmetry 7,
927 (1996)
[15] C.S. Chen, S.H. Wu, G. Girdaukas, C.J. Sih, J. Am.
Chem. Soc. 109, 2812 (1987)
[16] Ghanem,
Asymmetry 15, 3331 (2004)
[17] A.L. Spek, Acta Crystallogr. D65, 148 (2009)
H.Y.
Aboul-Enein,
Tetrahedron:
[9] R.L. Barreto, M.J.S. Carpes, C.C. Santana,
C.R.D. Correia, Tetrahedron: Asymmetry 18, 435
(2007)
32