Candida antarctica Lipase B in a chemoenzymatic route to cyclic α-quaternary α-amino acid enantiomers

Article abstract of DOI:10.1002/ejoc.201001575

Kinetic resolution of three cyclic quaternary ethyl 1-amino-2,3-dihydro-1H- indene-1-carboxylates and both 1- and 2-amino-1,2,3,4-tetrahydronaphthalene analogues have been studied. Interesterification with butyl butanoate and Candida antarctica lipase B accomplished the task. The enantiomers of all 1-amino analogues reacted with excellent enantioselectivity (enantiomeric ratio er greater than 200), whereas the 2-amino analogue was not enantioselective (er = 4). Amino acid enantiomers were finally obtained as their respective hydrochlorides with almost maximum theoretical yields. For the first time, a lipase enzyme was effectively used in the kinetic resolution of cyclic α-quaternary α-amino esters. Copyright

Full text of DOI:10.1002/ejoc.201001575

FULL PAPER
DOI: 10.1002/ejoc.201001575
Candida antarctica Lipase B in a Chemoenzymatic Route to Cyclic
α-Quaternary α-Amino Acid Enantiomers
Xiang-Guo Li,[a] Maria Rantapaju,[a] and Liisa T. Kanerva*[a]
Keywords: Enzyme catalysis / Esters / Kinetic resolution / Lipase / Quaternary amino acid
Kinetic resolution of three cyclic quaternary ethyl 1-amino-
2,3-dihydro-1H-indene-1-carboxylates and both 1- and 2-
amino-1,2,3,4-tetrahydronaphthalene analogues have been
studied. Interesterification with butyl butanoate and Candida
antarctica lipase B accomplished the task. The enantiomers
of all 1-amino analogues reacted with excellent enantio-
selectivity (enantiomeric ratio er greater than 200), whereas
the 2-amino analogue was not enantioselective (er = 4).
Amino acid enantiomers were finally obtained as their re-
spective hydrochlorides with almost maximum theoretical
yields. For the first time, a lipase enzyme was effectively
used in the kinetic resolution of cyclic α-quaternary α-amino
esters.
Introduction
Cyclic α-quaternary α-amino acids are key building
blocks of many pharmaceutically interesting compounds.[1]
As bulky amino acids, they are also powerful building
blocks of peptides and proteins that can be used to induce
fascinating, rigid and yet predictable secondary structures.[2]
The asymmetric synthesis and resolution procedures of
many α-quaternary α-amino acids and their derivatives are
well-documented in various review articles.[3] The use of
biocatalysis in this connection has been mainly limited to
the preparation of alcohol and acid precursors by kinetic
resolution followed by chemical transformations into the
amino acids or their derivatives.[3a] Two proteases [Alcalase
2.4L (E.C. 3.4.21.62) and papain (E.C. 3.4.22.2)] have been
applied for the enantioselective hydrolysis of cyclic α-qua-
ternary α-amino esters.[1b,4a] In addition, amino esterase
purified from a crude extract of lipase CE from Humicola
lanuginosa has been reported to catalyze the hydrolysis of
ethyl 2-amino-1,2,3,4-tetrahydronaphthalene-2-carboxylate
(rac-2 of this work, Figure 1),[4b] and the whole cell Rhodo-
coccus sp. AJ270 has provided access to some enantiopure
cyclic α-quaternary α-amino acids.[5] In the latter case, an
amidase in the whole cell was suggested to catalyze the hy-
drolysis of the amide substrates studied.
Figure 1. Substrates and potential reaction types for kinetic resolu-
tion.
the corresponding amino acids, as shown for the prepara-
tion of (R)- and (S)-8a–d in Scheme 1. More specifically, we
aim to show that lipase catalysis can be used in the kinetic
resolution of cyclic α-quaternary α-amino acids. Thus far,
to the best of our knowledge, lipase catalysis has not been
reported in connection with such conformationally con-
strained substrates. Indeed, it has been reported that lipases
cannot be used.[4] Potential alternative methods for the ki-
netic resolution of amino esters by lipases include enantio-
selective N-acylation of the amino function and enantiose-
lective hydrolysis, alcoholysis, or interesterification of the
ester function (Figure 1).
Herein, we have studied the possibility of developing a
chemoenzymatic method for the preparation of the enantio-
mers of rac-1 and rac-2 (Figure 1) as hydrochloride salts of
Interesterification, which is still rarely used in kinetic res-
olution, is an alcoholysis reaction in which two esters with
different alkyl parts react to form two new esters. From a
mechanistic point of view, the formation of an acyl–enzyme
intermediate is crucial for lipase catalysis. To explain the
high efficiency of lipase-catalyzed interesterification com-
pared to the corresponding alcoholysis reaction, it was pre-
[a] Institute of Biomedicine/Department of Pharmacology, Drug
Development and Therapeutics/Laboratory of Synthetic Drug
Chemistry and Department of Chemistry, University of Turku,
20014 Turku, Finland
Fax: +358-2-3337955
E-mail: lkanerva@utu.fi
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201001575.
Eur. J. Org. Chem. 2011, 1755–1762
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1755

L. T. Kanerva et al.
FULL PAPER
Scheme 1. Chemoenzymatic synthesis of cyclic α-quaternary α-amino acid enantiomers 8a–d.
viously suggested that an alcohol product formed from one antarctica lipase B, Novozym 435) for reactions at the ester
acyl donor in the formation of the intermediate (for in- function, have been commonly used in the kinetic resolu-
stance 1-butanol formed from butyl butanoate in the pres- tion of amino esters.[6d] Thus, sterically hindered cyclic β-
ent work) does not leave the active site.[6] Rather, this amino esters with the cis configuration were among the first
alcohol will stay in the active site and remain ready to react cases to be successfully resolved using CAL-A-catalyzed N-
as a nucleophile to form an acyl enzyme–intermediate when acylation.[9] The kinetic resolution of tertiary alcohols has
another ester substrate (for instance the S-enantiomer of also been successful with CAL-A.[10] On the other hand,
rac-1 or rac-2) enters the active site. Interesterification with CAL-B has been used in the kinetic resolution of chiral
various lipases in organic solvents was first recognized as a aminoesters with large acyl parts.[4a,11]
side reaction in the N-acylation of acyclic β-amino esters
more than a decade ago.[6a] Later it was possible to make
use of interesterification. Good examples are the activation
Results and Discussion
of β-amino acids as 2,2,2-trifluoroethyl esters with 2,2,2-
trifluoroethyl butanoate in lipase-catalyzed β-peptide syn-
thesis,[6b] and the kinetic resolution of piperidine-based β-
amino esters.[6c]
Synthesis of Hydantoins and Racemic Amino Esters
In the context of enzyme catalysis in organic synthesis,
Spirohydantoins rac-4 and 3Ј,4Ј-dihydro-2ЈH-spiro[imid-
lipases (E.C. 3.1.1.3) are of central significance in producing azolidine-4,1Ј-naphthalene]-2,5-dione were conveniently
enantiopurity in chiral molecules. This is due to their gen- prepared from the corresponding cyclic ketones in the pres-
erally excellent stability in various non-aqueous media in ence of potassium cyanide and ammonium carbonate in
addition to aqueous environments, good availability in free ethanol/water (1:1, v/v) (Scheme 1). To avoid massive de-
and immobilized forms, high enantio- and diastereoselectiv- composition of ammonium carbonate, the reaction tem-
ity, regioselectivity, and broad substrate scope. Four groups perature was kept below 50 °C. The subsequent hydrolysis
of lipases, namely Rml (large alcohol binding cleft but nar- of the hydantoins obtained was carried out in a sealed steel
row cleft for acyl binding), CAL-A (strong restrictions on reactor, which enabled the use of high reaction temperature
the acyl part due to a narrow tunnel to accommodate the (150 °C). The crude amino acids, thus obtained, were used
acyl group but wide alcohol binding cleft), CAL-B (large without further purification in a routine esterification with
acyl binding cleft but narrow alcohol binding cleft), and ethanol, affording the corresponding amino esters rac-1 and
Cutinase (wide alcohol and acyl binding clefts) type lipases, rac-2 with 72–94% overall yields, the low yield (only 15%)
were previously recognized according to their structural of rac-1d being an exception. The synthetic protocol was
space for acceptance of substrates of different shapes.[7] modified from the pathway previously applied for the prep-
CAL-A and CAL-B type enzymes, in particular CAL-A aration of racemic methyl 2-amino-5,8-dimethoxy-1,2,3,4-
(Candida antarctica lipase A as adsorbed on Celite in the tetrahydro-2-naphthoate, a structural analogue of rac-
presence sucrose[8]) for N-acylation and CAL-B (Candida 2.[1a,1e]
1756
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1755–1762

Cyclic α-Quaternary α-Amino Acid Enantiomers
Enzymatic Kinetic Resolution of Amino Esters
tions. As CAL-B, with the large acyl-binding cleft, cata-
lyzed ester hydrolysis, this enzyme was chosen for
For the kinetic resolution of rac-1 and rac-2 by N-acylation, alcoholysis with 1-butanol and for interesterification with
the more reactive conformationally constrained substrate butyl butanoate using rac-1a in DIPE. Both types of reac-
enantiomer must be bound correctly at the nucleophile tions proceeded with excellent enantioselectivity [er (enan-
binding cleft of the enzyme so that acylation by the acyl– tiomeric ratio) Ͼ 200] at 23 °C (Table 1).
enzyme intermediate can occur. When the same amino es-
Water (often called residual water) is always present in
ters are considered as constrained ester substrates, the acyl- seemingly dry lipase preparations, and this water is known
binding cleft accommodates the reactive enantiomer to to lead to undesired hydrolyses of ester substrates and prod-
form the acyl–enzyme intermediate, which then reacts with ucts as side reactions in organic solvents. This is especially
a suitable achiral nucleophile and leads to the product. This the case with CAL-B because of the Novozym 435 prepara-
consideration led us examine CAL-A and CAL-B in par- tion (CAL-B adsorbed on a divinylbenzene-crosslinked,
ticular. In spite of expectations, N-acylation of rac-1a, rac- hydrophobic macroporous polymer based on methyl and
1d or rac-2 did not proceed with 2,2,2-trifluoroethyl but- butyl methacrylic esters). This lipophilic support material
anoate and CAL-A (adsorbed on Celite in the presence of will readily release any water present into the anhydrous
sucrose[8]) in diisopropyl ether (DIPE). Accordingly, we fo- reaction mixture.[12] Moreover, a water tunnel allowing di-
cused on enzymatic reactions involving the ester function rect access of water from the medium to the active site was
of the substrates (Figure 1). Extensive lipase screening for previously shown to exist in CAL-B.[13] Together, these fac-
the ester hydrolysis of rac-1a as a model substrate revealed tors expose the active site to residual water that can lead
that only CAL-B catalyzes the reaction, accompanied by to acyl–enzyme intermediate hydrolysis in addition to the
significant chemical background hydrolysis. Moreover, the desired lysis with another nucleophile. The competitive na-
poor solubility of the racemates in water was not encourag- ture of the residual water is clearly seen in the alcoholysis
ing for continued optimization studies with aqueous solu- and interesterification of rac-1a (0.05 m) with CAL-B in
Table 1. Alcoholysis and interesterification of rac-1a (0.05 m) with 1-butanol and PrCO2Bu, respectively, in the presence of CAL-B
(50 mgmL–1) in DIPE at 23 °C; reaction time 24 h.
Entry
Reagent
Conversion[a] (x)[b] [%]
er
1
2
3
4
5
6
7
8
residual water
BuOH (5 equiv.)
– (54)
38 (10)
30 (8)
39 (0)
43 (7)
49 (0)
47 (5)
37 (0)
11
Ͼ200
Ͼ200
54
Ͼ200
Ͼ200
Ͼ200
Ͼ200
BuOH (10 equiv.)
BuOH (10 eqiuv. and MS)[c]
PrCO2Bu (10 equiv.)
PrCO2Bu (10 equiv. and MS)[c]
PrCO2Bu (5 equiv. and MS)[c]
PrCO2Bu (neat and MS)[c]
[a] Total conversion (dihexyl ether used as internal standard in GC analysis). [b] Proportion of (S)-9 [obtained by subtracting the sum
of the amounts of (R)-1a and (S)-5a from the initial amount of rac-1a]. [c] Molecular sieves (4 Å; 30 mgmL–1) were added.
Table 2. Interesterification of rac-1a (0.05 m) with PrCO2Bu (10 equiv.) in the presence of CAL-B (50 mgmL–1) and molecular sieves
(30 mgmL–1) in organic solvents at 23 °C.
Entry
rac-1a [M]
Solvent
Time [h]
Conv.[a] [%]
ee [%]
(R)-1a
er
(S)–5a
1
2
3
4
5
6
0.05
0.05
0.05
0.05
0.1
DIPE
PhMe
Et2O
TBME
DIPE
DIPE
24
24
24
24
24/96
24/96
49
48
48
49
46/48
43/45
94
94
93
96
87/90
76/80
Ͼ99
Ͼ99
Ͼ99
Ͼ99
Ͼ99/97
Ͼ99/98
Ͼ200
Ͼ200
Ͼ200
Ͼ200
Ͼ200
Ͼ200
0.2
[a] Total conversion (dihexyl ether used as an internal standard in GC analysis).
Eur. J. Org. Chem. 2011, 1755–1762
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1757

L. T. Kanerva et al.
FULL PAPER
dried DIPE (see values in parentheses for conversions, butanol is a competitive inhibitor of many lipases.[14] For
Table 1 and Table 2). Thus, the enzymatic hydrolysis of rac- reactions in DIPE, increases in butyl butanoate content
1a proceeded to 54% conversion in 24 h when residual hardly affects the reactivity (compare entries 6 and 7),
water was the only nucleophile present (entry 1, Table 1). whereas the reaction in neat butyl butanoate is considerable
Enantioselectivity of the hydrolysis was negligible (er = 11). reduced (entry 8) due to solvent effects.
Hydrolysis was dramatically suppressed when 1-butanol
At this point, the more efficient interesterification reac-
was present either as an added nucleophile (entries 2–4) or tion in the presence of molecular sieves was chosen to con-
when it was formed in situ at the active site from butyl tinue optimization. Solvent effects and substrate concentra-
butanoate (Table 1, entries 5–8), in this case, (S)-5a was tion, in particular, were examined (Table 2). Interesterifica-
formed as the major product. For interesterification, the tion of rac-1a (0.05 m) with butyl butanoate (10 equiv.) and
presence of molecular sieves (4 Å; 30 mgmL–1) diminished CAL-B (50 mgmL–1) in DIPE, toluene, diethyl ether, and
the amount of residual water in the reaction mixture and, tert-butyl methyl ether (TBME) proceeded in very similar
accordingly, practically eliminated the hydrolysis (entries 4 yields with excellent enantioselectivity (erϾ200), except
and 6 vs. entries 3 and 5, respectively). The reduction in the that minor hydrolysis occurred as a side reaction for reac-
extent of hydrolysis was less prominent with lower amounts tions in solvents other than DIPE (entry 1 vs. entries 1–4).
of molecular sieves (2% hydrolysis was observed with The good solubility of rac-1a in DIPE allowed interesterifi-
20 mgmL–1 and 5% with 10 mgmL–1 of MS vs. 0% with cation studies to be conducted at elevated substrate concen-
30 mgmL–1, entry 6). As expected, reduced butyl butanoate trations (0.1 and 0.2 m). However, conversions reached after
content in DIPE increased the proportion of hydrolysis (en- 24 h were somewhat reduced at high substrate concentra-
try 7 vs. entry 6), evidently because the number of available tions (entries 1 vs. 5 and 6). Moreover, the reactions pro-
active sites was reduced when 1-butanol was able to block ceeded very little at concentrations over 0.05 m after 24 h,
the entrance of the water tunnel to the active site or, by which could be due to complicated equilibria [the mixture
some other mechanism, was able to prevent the hydrolysis. contains almost equal amounts of PrCO2Et and (S)-5a in
It is worth emphasizing that the absolute values for small addition to PrCO2Bu] that the system establishes. Taken to-
proportions of hydrolysis in Table 1 are subject to some un- gether, these results suggest that it is almost impossible to
certainty because the amount of acid (S)-9 reported was obtain both enantiomers pure in a single kinetic resolution
simply obtained by subtracting the sum of (R)-1a and (S)- procedure at the highest substrate concentrations studied.
5a from the initial amount of rac-1a as determined by the
GC method, rather than detecting the amount of acid di- interesterification with butyl butanoate (10 equiv.) and
rectly.
CAL-B (50 mgmL–1) in DIPE in the presence of molecular
Compounds rac-1a–d and rac-2 were then subjected to
It is clear from the conversions observed after 24 h that sieves (4 Å; 30 mgmL–1) at 23 °C. As shown in Table 3, 1-
the enzymatic interesterification of rac-1a in DIPE is fa- amino-1-indancarboxylates rac-1a–c all reacted rapidly
vored over alcoholysis (Table 1, entries 2–4 vs. 5–7). A pos- with excellent enantioselectivities, indicating low substituent
sible explanation is that, as in interesterification, because 1- effects caused by the halogen atoms in the benzene ring
butanol is formed in situ at the active site, its concentration (entries 1–3). The kinetic resolution of rac-1d was highly
is equimolar (or higher) to that of the acyl–enzyme interme- enantioselective (er Ͼ200) although very slow, however,
diate, whereas 1-butanol for alcoholysis needs to enter the considerable rate enhancement was observed at elevated
active site normally from the medium before it can be temperature (entries 4 and 5). On the other hand, rac-2 re-
bound in the nucleophile binding site. Moreover, acted rapidly with negligible enantioselectivity (entry 6).
alcoholysis slows down with increasing alcohol content in Accordingly, preparative-scale kinetic resolution of rac-1a–
the medium (compare entries 2 and 3, Table 1) because 1- d (0.05 m) was effectively performed using interesterification
Table 3. Interesterification of rac-1a–d and rac-2 (0.05 m) with butyl butanoate (10 equiv.) in the presence of CAL-B (50 mgmL–1) and
molecular sieves (30 mgmL–1) in DIPE at 23 °C.
Entry
Substrate
Time [h]
Conv. [%]
ee (R) [%]
ee (S) [%]
er
1
2
3
rac-1a
rac-1b
rac-1c
rac-1d
rac-1d
rac-2
24
24
24
504
144
5
49
51
49
45
49
44
94
99
96
80
Ͼ99
94
Ͼ99
Ͼ99
Ͼ99
45[b]
Ͼ200
Ͼ200
Ͼ200
Ͼ200
Ͼ200
4
4
5[a]
6
95
35[b]
[a] Reaction performed at 47 °C. [b] Absolute configuration was not assigned.
1758
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1755–1762

Cyclic α-Quaternary α-Amino Acid Enantiomers
with PrCO2Bu (10 equiv.) and CAL-B (50 mg/mL, Novo- anoate as the precursor of butanol in DIPE. The inclusion
zym 435 preparation) in the presence of molecular sieves of molecular sieves (4 Å) in the reaction mixture was essen-
(4 Å; 30 mg/mL) in DIPE. Each reaction was stopped at tial to remove residual water and to prevent ester hydrolysis
almost 50% conversion, at which time one of the respective occurring as a side reaction. This hydrolysis side reaction
enantiomer pairs of (S)-5a–d and (R)-1a–d were detected in in organic solvents and main reaction in water both pro-
the resolution mixture with ee values of 94–99%. To facili- ceeded with moderate enantioselectivity. Interesterification
tate subsequent purification by column chromatography, of ethyl 2-amino-1,2,3,4-tetrahydronaphthalene-2-carboxyl-
the enantiomers were transformed into the corresponding ate (rac-2) under the same conditions was fast, but pro-
trifluoroacetamides (R)-6 and (S)-7 simply by adding ceeded with low enantioselectivity. This study also shows
(CF3CO)2O (1.2 equiv.) (see Exp. Sect.). The products were that alcoholysis with 1-butanol is less favorable than inter-
finally isolated in almost maximum theoretical yields of esterification with butyl butanoate under otherwise iden-
50%.
Due to the low enantioselectivity, the kinetic resolution
tical conditions in DIPE.
of rac-2 was not performed on a preparative scale (Table 3,
entry 6). Using Humicola amino esterase, the hydrolysis of
rac-2 as a methyl ester also previously showed low enantio-
selectivity (er = 19), affording only the unreacted ester in
enantiopure form (ee = 99%) at 64% conversion.[4b]
Experimental Section
Materials and Methods: All solvents were of the highest analytical
grade and were dried by standard methods. Ketones 3 were pur-
chased from Aldrich. Candida antarctica lipase B (CAL-B, Novo-
zym 435) was purchased from Novozymes (initial rate
1.97 mmolmin–1 g–1 based on the acylation of 1-phenylethanol with
isopropenyl acetate in toluene at room temperature) and Candida
antarctica lipase A (CAL-A) was purchased as a lyophilized powder
from Biocatalytics. CAL-A (20%, w/w) was adsorbed in Celite in
the presence of sucrose (3%, w/w) before use as described pre-
viously.[8] NMR spectra were recorded with a Bruker Avance
Determination of Absolute Configurations
The value for the specific rotation of (S)-9 has been re-
ported {[α]2D4 = +103.7 (c = 0.96, MeOH)}.[15] This knowl-
edge was used to determine which enantiomer of the pres-
ent substrates reacts faster in the kinetic resolution with
CAL-B. For this purpose, enantiopure 8a, which was ob-
tained from enantiopure 7a, was transformed into the cor-
responding compound 9 by known methods, as described
in the Exp. Section (Scheme 2).[15] The S absolute configu-
ration of the obtained acid was assigned based on its optical
rotation {[α]2D4 = +107.1 (c = 0.96, MeOH)}. The S absolute
configuration is also in accordance, for instance, with the R
enantiopreference previously observed in the CAL-B-cata-
lyzed hydrolysis of 1,2,3,4-tetrahydroisoquinoline-1-carbox-
ylic esters as determined by the Cahn–Ingold–Prelog pri-
ority rules.[11]
1
1
1
500 MHz spectrometer. H-1H COSY, H-13C HQSC, and H-13C
HMBC spectra were used for the assignment of the chemical shifts
when necessary. HRMS were measured in ESI+ mode with a
Bruker micrOTOF-Q quadrupole-TOF spectrometer. Melting
points were recorded with a Sanyo instrument. Optical rotations
were determined with a PerkinElmer 241 polarimeter, and [α]D val-
ues are given in units of 10–1 deg cm2 g–1. Enzymatic reactions were
performed at room temperature (23 °C) unless indicated otherwise.
The determination of er (enantiomeric ratio) was based on the
equation er = ln[(1 – c)(1 – ees)]/ln[(1 – c)(1 + ees)] with c =
ees/(ees + eep) using linear regression {er as the slope of the line
ln[(1 – c)(1 – ees)] vs. ln[(1 – c)(1 + ees)]}.[16] Flash chromatography
was performed using silica gel (60 Å, Merck, 230–400 mesh, en-
riched with 0.1% Ca).
Racemic
2Ј,3Ј-Dihydrospiro[imidazolidine-4,1Ј-indene]-2,5-dione
(rac-4a): 1-Indanone 3a (3.30 g, 25.00 mmol), KCN (2.11 g,
32.46 mmol), and (NH4)2CO3 (18.24 g, 0.19 mol) were dissolved in
EtOH/H2O (90 mL, 1:1, v/v) and the mixture was stirred at 50 °C
for 144 h. The reaction mixture was cooled in an ice-bath and
brought to pH 3 with HCl (3 n). The resulting precipitate was col-
lected by filtration and washed with copious amounts of EtOH/
H2O (1:1, v/v), affording the desired product rac-4a as a powdery
solid (4.06 g, 20.10 mmol, 80%); m.p. 239–241 °C (ref.[1a,1e]
239 °C). 1H NMR (500 MHz, [D6]DMSO): δ = 10.74 (s, 1 H), 8.43
(s, 1 H), 7.34–7.32 (m, 2 H), 7.31–7.24 (m, 1 H), 7.16 (d, J = 7.5 Hz,
1 H), 3.01 (m, 2 H), 2.54 (m, 1 H), 2.17 (m, 1 H) ppm. 13C NMR
(126 MHz, [D6]DMSO): δ = 177.73, 157.01, 144.36, 141.80, 129.32,
127.47, 125.50, 123.25, 72.24, 36.72, 30.18 ppm. HRMS: calcd. for
C11H10N2O2Na [M + Na]+ 225.0634; found 225.0631.
Scheme 2. Preparation of (S)-8a for determining the enantioprefer-
ence of CAL-B.
Conclusions
This work describes the preparative-scale kinetic resolu-
tion of cyclic α-quaternary α-amino esters – a new class
of important compounds in the context of lipase catalysis.
Although N-acylation of the amino group at the quaternary
carbon was not successfully achieved by lipase catalysis, the
ester group at the same carbon did react. The application
of CAL-B (as the Novozym 435 preparation) in the interes-
terification of ethyl 1-amino-1,2,3,4-tetrahydronaphthalene-
1-carboxylate (rac-1d) and ethyl 1-amino-2,3-dihydro-1H-
indene-1-carboxylates rac-1a–c all gave excellent enantio-
5Ј-Chloro-2Ј,3Ј-dihydrospiro[imidazolidine-4,1Ј-indene]-2,5-dione
(rac-4b): According to a similar procedure to the synthesis of rac-
4a, compound 3b (1.39 g, 8.33 mmol) was transformed into rac-4b
(1.47 g, 6.23 mmol, 75%); m.p. 251–252 °C. 1H NMR (500 MHz,
CD3OD): δ = 7.36 (s, 1 H), 7.29 (d, J = 8.0 Hz, 1 H), 7.19 (d, J =
8.0 Hz, 1 H), 3.17 (m, 1 H), 3.08 (m, 1 H), 2.69 (m, 1 H), 2.32
selectivity (er Ͼ200) and chemoselectivity with butyl but- (m, 1 H) ppm. 13C NMR (126 MHz, CD3OD): δ = 177.80, 157.63,
Eur. J. Org. Chem. 2011, 1755–1762
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1759

L. T. Kanerva et al.
FULL PAPER
146.47, 139.46, 134.84, 127.14, 125.02, 123.88, 72.08, 36.48,
29.45 ppm. HRMS: calcd. for C11H8ClN2O2Na [M + Na]+
259.0245; found 259.0233.
125.22, 124.09, 68.00, 61.64, 39.45, 30.42, 14.06 ppm. HRMS:
calcd. for C12H14ClNO2 [M + H]+ 240.0786; found 240.0774.
Racemic Ethyl 1-Amino-5-fluoro-2,3-dihydro-1H-indene-1-carboxyl-
ate (rac-1c): As above, rac-4c (0.75 g, 3.41 mmol) was transformed
into rac-1c (0.71 g, 3.21 mmol, 94%). 1H NMR (500 MHz, CDCl3):
δ = 7.22 (dd, J = 8.3, 5.3 Hz, 1 H), 6.93–6.88 (m, 2 H), 4.15 (m, 2
H), 3.04 (m, 2 H), 2.81 (m, 1 H), 2.07 (m, 1 H), 1.21 (t, J = 7.3 Hz,
3 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 174.99, 163.21 (d, J
= 245.7 Hz), 146.14 (d, J = 8.8 Hz), 141.41, 124.12 (d, J = 8.8 Hz),
113.95 (d, J = 22.7 Hz), 111.87 (d, J = 21.4 Hz), 67.85, 61.44, 39.98,
30.50, 14.06 ppm. HRMS: calcd. for C12H15FNO2 [M + H]+
224.1081; found 224.1071.
5Ј-Fluoro-2Ј,3Ј-dihydrospiro[imidazolidine-4,1Ј-indene]-2,5-dione
(rac-4c): According to a similar procedure to the synthesis of rac-
4a, compound 3c (2.50 g, 16.67 mmol) was transformed into rac-
4c (2.97 g, 13.50 mmol, 81%); m.p. 285–286 °C (dec.). 1H NMR
(500 MHz, [D6]DMSO): δ = 10.79 (s, 1 H), 8.43 (s, 1 H), 7.21–7.16
(m, 2 H), 7.07 (dt, J = 9.0, 2.5 Hz, 1 H), 3.01 (m, 2 H), 2.55 (m, 1
H), 2.20 (m, 1 H) ppm. 13C NMR (126 MHz, [D6]DMSO): δ =
177.57, 163.27 (d, J = 244.4 Hz), 156.92, 147.18 (d, J = 8.8 Hz),
137.85, 125.03 (d, J = 8.8 Hz), 114.58 (d, J = 22.7 Hz), 112.38 (d,
J = 22.7 Hz), 71.52, 37.17, 30.10 ppm. HRMS: calcd. for
C11H9FN2O2Na [M + Na]+ 243.0540; found 243.0531.
Racemic Ethyl 1-Amino-1,2,3,4-tetrahydronaphthalene-1-carboxyl-
ate (rac-1d): As above, rac-4d (0.65 g, 3.00 mmol) was transformed
into rac-1d (98.55 mg, 0.45 mmol, yield 15%). 1H NMR (500 MHz,
CDCl3): δ = 7.26–7.10 (m, 4 H), 4.18 (m, 2 H), 2.83 (t, J = 7.0 Hz,
2 H), 2.29 (m, 1 H), 1.92 (m, 3 H), 1.20 (t, J = 7.0 Hz, 3 H) ppm.
13C NMR (126 MHz, CDCl3): δ = 177.06, 136.70, 130.97, 129.51,
128.87, 127.44, 126.41, 61.65, 59.75, 35.86, 30.01, 19.01, 14.24 ppm.
HRMS: calcd. for C13H18NO2 [M + H]+ 220.1332; found 220.1343.
3Ј,4Ј-Dihydro-2ЈH-spiro[imidazolidine-4,1Ј-naphthalene]-2,5-dione
(rac-4d): According to a similar procedure to the synthesis of rac-
4a, α-tetralone 3d (3.65 g, 25.00 mmol) was transformed into the
title compound 4d (3.02 g, 14.01 mmol, 56%); m.p. 241–242 °C (ref.
[1a,1e] 242 °C). 1H NMR (500 MHz, [D6]DMSO): δ = 10.81 (s, 1 H),
8.53 (s, 1 H), 7.25–7.15 (m, 3 H), 7.07 (d, J = 7.5 Hz, 1 H), 2.76
(t, J = 5.5 Hz, 2 H), 2.07 (m, 2 H), 1.91 (m, 1 H), 1.81 (m, 1
H) ppm. 13C NMR (126 MHz, [D6]DMSO): δ = 177.95, 156.48,
137.73, 134.37, 129.27, 127.96, 126.62, 126.53, 63.09, 33.52, 28.45,
18.42 ppm. HRMS: calcd. for C12H12N2O2Na [M + Na]+ 239.0791;
found 239.0781.
Preparation of Racemic Ethyl 2-Amino-1,2,3,4-tetrahydronaphth-
alene-1-carboxylate (rac-2): As above, 3Ј,4Ј-dihydro-2ЈH-spiro-
[imidazolidine-4,1Ј-naphthalene]-2,5-dione (0.65 g, 3.00 mmol) was
transformed into rac-2 (0.53 g, 2.43 mmol, 81%). 1H NMR
(500 MHz, CDCl3): δ = 7.14–7.06 (m, 4 H), 4.19 (q, J = 7.0 Hz, 2
H), 3.30 (d, J = 16.0 Hz, 1 H), 3.00 (m, 1 H), 2.84 (m, 1 H), 2.76
(d, J = 16.0 Hz, 1 H), 2.17 (m, 1 H), 1.91 (m, 1 H), 1.26 (t, J =
7.0 Hz, 3 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 176.58,
134.75, 133.57, 129.42, 128.73, 126.06, 125.98, 61.19, 56.42, 39.31,
31.91, 25.41, 14.17 ppm. HRMS: calcd. for C13H18NO2 [M + H]+
220.1332; found 220.1340.
3Ј,4Ј-Dihydro-2ЈH-spiro[imidazolidine-4,1Ј-naphthalene]-2,5-dione:
According to a similar procedure to the synthesis of rac-4a, β-tet-
ralone (3.65 g, 25.00 mmol) was transformed into the title com-
pound (4.64 g, 21.50 mmol, 86 %); m.p. 268–269 °C (ref.[1a,1e]
267 °C). 1H NMR (500 MHz, [D6]DMSO): δ = 10.69 (s, 1 H), 8.31
(s, 1 H), 7.15–7.07 (m, 4 H), 3.11 (d, J = 17.0 Hz, 1 H), 2.91 (m, 2
H), 2.77 (d, J = 17.0 Hz, 1 H), 1.94 (m, 1 H), 1.81 (m, 1 H) ppm.
13C NMR (126 MHz, [D6]DMSO): δ = 178.22, 156.35, 134.86,
132.62, 128.99, 128.60, 125.97, 125.88, 60.75, 36.84, 30.04,
24.70 ppm. HRMS: calcd. for C12H12N2O2Na [M + Na]+ 239.0791;
found 239.0783.
General Procedure for Small-Scale Enzymatic Kinetic Resolution: In
a typical small-scale experiment, CAL-B (50 mgmL–1) and molecu-
lar sieves (4 Å, 30 mgmL–1) were added to one of the substrates
rac-1a–d or rac-2 (0.05 m) and butyl butanoate (10 equiv.) or 1-
butanol (10 equiv.) in DIPE (1 mL). Reaction vessels were shaken
at 170 rpm and the progress of the reactions were followed by tak-
ing samples at intervals, filtering off the enzyme and analyzing the
samples by GC on a Chrompack CP-Chirasil-DEX CB FAST col-
umn. Dihexyl ether was used as an internal standard for quantita-
tive analyses of the resolution products (the unreacted ethyl ester
and the obtained butyl ester). The sum of the analyzed products
was subtracted from the initial amount of the racemate to obtain
the proportion of hydrolysis occurring during the enzymatic reac-
tion.
Racemic Ethyl 1-Amino-2,3-dihydro-1H-indene-1-carboxylate (rac-
1a): A suspension of rac-4a (1.50 g, 7.42 mmol) and Ba(OH)2·
8H2O (3.54 g, 11.2 mmol) in distilled water (30 mL) was heated at
150 °C in a sealed steel reactor for 72 h. The reaction mixture was
cooled to room temperature and acidified to pH 1.6 with H2SO4
(1 m). The precipitate was filtered off and the filtrate was evapo-
rated to dryness. The residue was suspended in EtOH (50 mL) and
SOCl2 (0.65 mL, 8.90 mmol) was added at 0 °C. The reaction mix-
ture was heated to reflux for 24 h before being evaporated to dry-
ness. Water (20 mL) was then added and the pH was adjusted with
aqueous NH4OH (25%) to pH 9–10 (indicated by pH paper). The
aqueous solution was extracted with CH2Cl2 (2ϫ50 mL) and the
combined organic layers were dried with anhydrous Na2SO4, fil-
tered, and evaporated, affording rac-1a (1.10 g, 5.36 mmol, 72%).
1H NMR (500 MHz, CDCl3): δ = 7.29–7.19 (m, 4 H), 4.14 (m, 2
H), 3.07 (m, 2 H), 2.80 (m, 1 H), 2.04 (m, 1 H), 1.21 (t, J = 7.0 Hz,
3 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 175.26, 145.91,
143.73, 128.35, 126.85, 124.99, 122.79, 68.54, 61.35, 39.61, 30.61,
14.09 ppm. HRMS: calcd. for C12H16NO2 [M + H]+ 206.1176;
found 206.1167.
Preparative-Scale Enzymatic Kinetic Resolution of rac-1a: CAL-B
preparation (50 mg/mL) was added to a solution of rac-1a (0.30 g,
1.46 mmol) and PrCO2Bu (2.42 mL, 14.60 mmol) in DIPE
(29.2 mL) in the presence of 4 Å MS (30 mgmL–1). The reaction
was stopped at 49% conversion by filtering off the enzyme after
24 h. (CF3CO)2O (368 mg, 1.75 mmol) in CH2Cl2 (20 mL) was
added to the above filtrate and the reaction mixture was stirred for
2 h before evaporating the solvent. The residue was purified on a
silica gel column (petroleum ether/ethyl acetate, 20:1), affording
oily (R)-6a (Rf = 0.10; 0.21 g, 0.71 mmol, 49%, ee 95%) and N-
trifluoroacetylated (S)-7a (Rf = 0.17; 0.23 g, 0.70 mmol, 48%, ee
99%).
Racemic Ethyl 1-Amino-5-chloro-2,3-dihydro-1H-indene-1-carboxyl-
ate (rac-1b): As above, rac-4b (0.81 g, 3.41 mmol) was transformed
into rac-1b (0.74 g, 3.10 mmol, 91 %). 1H NMR (500 MHz,
CDCl3): δ = 7.23–7.18 (m, 3 H), 4.16 (m, 2 H), 3.06 (m, 2 H), 2.79
(m, 1 H), 2.09 (m, 1 H), 1.21 (t, J = 7.0 Hz, 3 H) ppm. 13C NMR
(126 MHz, CDCl3): δ = 174.46, 145.77, 143.80, 134.33, 127.15,
Compound (R)-6a: [α]2D4 = –93.0 (c = 1.0, CHCl3). 1H NMR
(500 MHz, CDCl3): δ = 7.36–7.23 (m, 4 H), 4.20 (m, 2 H), 3.18 (t,
J = 7.3 Hz, 2 H), 3.02 (m, 1 H), 2.53 (m, 1 H), 1.21 (t, J = 7.3 Hz,
3 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 171.14, 156.12 (q, J
1760
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1755–1762

Cyclic α-Quaternary α-Amino Acid Enantiomers
= 37.7 Hz), 144.89, 139.62, 129.80, 127.32, 125.48, 122.59, 115.53
Compound (S)-8b: Compound (S)-7b (62.00 mg, 0.17 mmol) was
(q, J = 128.0 Hz), 69.77, 62.52, 35.41, 31.04, 13.92 ppm. HRMS: transformed into (S)-8b (42.16 mg, 0.17 mmol, 99%) as described
calcd. for C14H14F3NO3Na [M + Na]+ 324.0818; found 324.0807.
above. [α]2D4 = +98.3 (c = 0.58, MeOH).
Compound (S)-7a: [α]2D4 = +146.5 (c = 1.0, CHCl3). 1H NMR
(500 MHz, CDCl3): δ = 7.36–7.31 (m, 2 H), 7.26–7.24 (m, 2 H),
4.14 (t, J = 6.5 Hz, 2 H), 3.18 (t, J = 7.3 Hz, 2 H), 3.00 (m, 1 H),
2.54 (m, 1 H), 1.55 (m, 2 H), 1.23 (m, 2 H), 0.85 (t, J = 7.3 Hz, 3
H) ppm. HRMS: calcd. for C16H18F3NO3 [M + Na]+ 352.1131;
found 352.1123.
Preparative-Scale Enzymatic Kinetic Resolution of rac-1c: As de-
scribed above, rac-1c (0.32 g, 1.44 mmol) was resolved, affording
solid products (R)-6c (Rf = 0.19; 0.21 g, 0.67 mmol, 47%, ee 96%)
and (S)-7c (Rf = 0.29; 0.25 g, 0.71 mmol, 49%, ee 99%).
Compound (R)-6c: M.p. 78–79 °C; [α]2D4 = –93.2 (c = 1.0, CHCl3).
1H NMR (500 MHz, CDCl3): δ = 7.18 (dd, J = 8.5, 5.0 Hz, 1 H),
6.99 (d, J = 8.5 Hz, 1 H), 6.93 (t, J = 8.5 Hz, 1 H), 4.20 (m, 2 H),
3.17 (m, 2 H), 2.98 (m, 1 H), 2.58 (m, 1 H), 1.21 (t, J = 7.3 Hz, 3
H) ppm. 13C NMR (126 MHz, CDCl3): δ = 171.19, 163.94 (d, J =
248.2 Hz), 156.03 (q, J = 37.8 Hz), 147.43 (d, J = 8.8 Hz), 135.31
(d, J = 2.5 Hz), 123.97 (d, J = 10.1 Hz), 115.47 (q, J = 288.5 Hz),
114.63 (d, J = 23.9 Hz), 112.44 (d, J = 22.7 Hz), 69.06, 62.70, 35.74,
30.99, 13.89 ppm. HRMS: calcd. for C14H13F4NO3Na [M + Na]+
342.0724; found 342.0709.
Compound (R)-8a: Hydrolysis of (R)-6a (50.0 mg, 0.17 mmol) in
aqueous HCl (6 m) produced (R)-8a (36.28 mg, 0.17 mmol, 99%);
m.p. 230 °C (dec.). [α]2D4 = –100.1 (c = 0.18, MeOH). 1H NMR
(500 MHz, CD3OD): δ = 7.46–7.40 (m, 3 H), 7.34 (t, J = 7.0 Hz,
1 H), 3.24 (t, J = 7.0 Hz, 2 H), 2.92 (m, 1 H), 2.35 (m, 1 H) ppm.
13C NMR (126 MHz, CD3OD): δ = 171.57, 144.67, 138.19, 130.22,
127.25, 125.33, 122.98, 68.27, 34.72, 30.20 ppm. HRMS: calcd. for
C10H12NO2 [M – Cl]+ 178.0863; found 178.0843.
Compound (S)-8a: Compound (S)-7a (60.0 mg, 0.18 mmol) was
Compound (S)-7c: [α]2D4 = +89.2 (c = 1.0, CHCl3). 1H NMR
(500 MHz, CDCl3): δ = 7.18 (dd, J = 8.5, 5.0 Hz, 1 H), 6.99 (d, J
= 8.5 Hz, 1 H), 6.93 (t, J = 8.5 Hz, 1 H), 4.15 (t, J = 6.5 Hz, 2 H),
3.17 (m, 2 H), 2.96 (m, 1 H), 2.58 (m, 1 H), 1.55 (m, 2 H), 1.24
(m, 2 H), 0.86 (t, J = 7.5 Hz, 3 H) ppm. 13C NMR (126 MHz,
CDCl3): δ = 171.33, 163.92 (d, J = 248.2 Hz), 156.00 (d, J =
37.8 Hz), 147.35 (d, J = 8.8 Hz), 135.37 (d, J = 2.5 Hz), 123.92 (d,
J = 8.8 Hz), 115.62 (q, J = 289.8 Hz), 114.55 (d, J = 23.9 Hz),
112.41 (d, J = 22.7 Hz), 69.09, 66.45, 35.70, 30.98, 30.31, 18.85,
transformed into (S)-8a (38.41 mg, 0.18 mmol, 99%). [α]2D4
+105.8 (c = 0.18, MeOH).
=
Determination of Absolute Configuration: Hydrochloride salt (S)-8a
(19.22 mg, 0.09 mmol) was treated with 2-propanol/propylene ox-
ide (2.5/0.2 mL) at room temperature for 3 h. The white precipitate
(S)-9 was filtered off and dried (10.62 mg, 0.06 mmol, 65%). [α]2D4
= +107.1 (c = 0.96, MeOH) [ref. for (S)-9[15] [α]2D4 = +103.7 (c =
0.96, MeOH)]; m.p. 248–249 °C (ref.[15] 247 °C). 1H NMR
(500 MHz, CD3OD): δ = 7.47 (d, J = 8.0 Hz, 1 H), 7.35 (d, J =
4.0 Hz, 2 H), 7.31–7.23 (m, 1 H), 3.17 (m, 2 H), 2.92 (m, 1 H), 2.22
(m, 1 H) ppm. 13C NMR (126 MHz, CD3OD): δ = 174.33, 144.64,
141.18, 129.10, 126.71, 124.77, 123.16, 70.02, 34.70, 30.36 ppm.
HRMS: calcd. for C10H11NO2 [M + Na]+ 200.0682; found
200.0673.
13.46 ppm. HRMS: calcd. for C16H17F4NO3Na [M
371.1115; found 371.1101.
+
Na]+
Compound (R)-8c: The hydrolysis of (R)-6c (57.4 mg, 0.18 mmol)
in aqueous HCl (6 m) produced (R)-8c (41.67 mg, 0.18 mmol,
1
99%); m.p. 249–250 °C. [α]2D4 = –94.8 (c = 0.57, MeOH). H NMR
(500 MHz, CD3OD): δ = 7.47 (dd, J = 8.5, 4.5 Hz, 1 H), 7.16–7.07
(m, 2 H), 3.24 (t, J = 7.5 Hz, 2 H), 2.94 (m, 1 H), 2.41 (m, 1
H) ppm. 13C NMR (126 MHz, CD3OD): δ = 171.31, 164.41 (d, J
= 248.2 Hz), 147.79 (d, J = 8.8 Hz), 134.17, 124.90 (d, J = 10.1 Hz),
114.44 (d, J = 23.9 Hz), 112.17 (d, J = 22.7 Hz), 67.54, 35.17,
30.18 ppm. HRMS: calcd. for C10H11FNO2 [M – Cl]+ 196.0768;
found 196.0745.
Preparative-Scale Enzymatic Kinetic Resolution of rac-1b: As above,
rac-1b (0.30 g, 1.25 mmol) was resolved, affording solid products
(R)-6b (Rf = 0.11; 0.20 g, 0.61 mmol, 49%, ee 96%) and (S)-7b (Rf
= 0.19; 0.22 g, 0.60 mmol, 48%, ee 99%).
Compound (R)-6b: m.p. 72–73 °C. [α]2D4 = –81.6 (c = 1.0, CHCl3).
1H NMR (500 MHz, CDCl3): δ = 7.30 (s, 1 H), 7.21 (d, J = 8.0 Hz,
1 H), 7.14 (d, J = 8.0 Hz, 1 H), 4.21 (m, 2 H), 3.18 (m, 2 H), 2.93
(m, 1 H), 2.59 (m, 1 H), 1.21 (t, J = 7.3 Hz, 3 H) ppm. 13C NMR
(126 MHz, CDCl3): δ = 171.16, 155.97 (d, J = 37.8 Hz), 146.81,
138.19, 135.71, 127.65, 125.71, 123.63, 115.44 (d, J = 289.8 Hz),
69.20, 62.83, 35.48, 30.91, 13.89 ppm. HRMS: calcd. for
C14H11ClF3NO3Na [M + Na]+ 358.0428; found 358.0421.
Compound (S)-7b: [α]2D4 = +66.5 (c = 1.0, CHCl3). 1H NMR
(500 MHz, CDCl3): δ = 7.29 (s, 1 H), 7.20 (d, J = 8.0 Hz, 1 H),
7.13 (d, J = 8.0 Hz, 1 H), 4.14 (m, 2 H), 3.17 (m, 2 H), 2.92 (m, 1
H), 2.57 (m, 1 H), 1.54 (m, 2 H), 1.24 (m, 2 H), 0.86 (t, J = 7.5 Hz,
3 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 171.24, 155.97 (q, J
= 28.4 Hz), 146.73, 138.25, 135.64, 127.57, 125.65, 123.65, 115.45
(q, J = 216.72 Hz), 69.24, 66.55, 35.47, 30.88, 30.30, 18.86,
13.46 ppm. HRMS: calcd. for C16H16ClF3NO3Na [M + Na]+
386.0746; found 386.0732.
Compound (S)-8c: Compound (S)-7c (50.00 mg, 0.14 mmol) was
transformed into (S)-8c (32.41 mg, 0.14 mmol, 99%). [α]2D4 = +98.5
(c = 0.57, MeOH).
Preparative-Scale Enzymatic Kinetic Resolution of rac-1d: Per-
formed as described above except that the reaction was carried out
at 47 °C for 144 h, rac-1d (0.30 g, 1.37 mmol) was resolved, afford-
ing solid product (R)-6d (Rf = 0.17; 0.21 g, 0.66 mmol, 48%, ee
95%) and oily product (S)-7d (Rf = 0.25; 0.22 g, 0.65 mmol, 47%,
ee 99%).
Compound (R)-6d: M.p. 69–70 °C. [α]2D4 = –60.3 (c = 0.56, CHCl3).
1H NMR (500 MHz, CDCl3): δ = 7.26–7.16 (m, 4 H), 4.22 (m, 2
H), 2.93 (m, 1 H), 2.84 (m, 1 H), 2.59 (m, 1 H), 2.38 (m, 1 H), 2.10
(m, 1 H), 1.89 (m, 1 H), 1.22 (t, J = 7.0 Hz, 3 H) ppm. 13C NMR
(126 MHz, CDCl3): δ = 171.96, 155.36 (d, J = 36.5 Hz), 138.83,
132.89, 129.79, 128.62, 126.98, 125.93, 115.53 (q, J = 289.8 Hz),
62.69, 61.38, 30.83, 29.13, 20.02, 13.90 ppm. HRMS: calcd. for
C15H16F3NO3Na [M + Na]+ 338.0974; found 338.0984.
Compound (R)-8b: The hydrolysis of (R)-6b (60.0 mg, 0.18 mmol)
in aqueous HCl (6 m) produced (R)-8b (44.64 mg, 0.18 mmol,
99%); m.p. 265 °C (dec.). [α]2D4 = –94.9 (c = 0.58, MeOH). 1H NMR
(500 MHz, CD3OD): δ = 7.44–7.35 (m, 3 H), 3.24 (t, J = 7.0 Hz, Compound (S)-7d: [α]2D4 = +73.9 (c = 1.00, CHCl3). 1H NMR
2 H), 2.94 (m, 1 H), 2.38 (m, 1 H) ppm. 13C NMR (126 MHz,
CD3OD): δ = 171.13, 147.07, 136.98, 136.24, 127.51, 125.52, (t, J = 7.5 Hz, 2 H), 4.16 (m, 2 H), 2.95 (m, 1 H), 2.82 (m, 1 H),
(500 MHz, CDCl3): δ = 7.43 (br. s, 1 H), 7.26–7.22 (m, 2 H), 7.17
124.45, 67.65, 34.94, 30.07 ppm. HRMS: calcd. for C10H11ClNO2
2.59 (m, 1 H), 2.36 (m, 1 H), 2.10 (m, 1 H), 1.90 (m, 1 H), 1.56
[M – Cl]+ 212.0473; found 212.0464.
(m, 2 H), 1.24 (m, 2 H), 0.85 (t, J = 7.5 Hz, 3 H) ppm. 13C NMR
Eur. J. Org. Chem. 2011, 1755–1762
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1761

L. T. Kanerva et al.
FULL PAPER
[3]
[4]
For example, see: a) C. Cativiela, M. Ordónez, Tetrahedron:
Asymmetry 2009, 20, 1–63; b) K. Undheim, Amino Acids 2008,
34, 357–402; c) K.-H. Park, M. J. Kurth, Tetrahedron 2002, 58,
8629–8659.
a) G. Roda, P. Conti, M. De Amici, J. He, P. L. Polavarapu, C.
De Micheli, Tetrahedron: Asymmetry 2004, 15, 3079–3090; b)
W. Liu, P. Ray, S. A. Benezra, J. Chem. Soc. Perkin Trans. 1
1995, 553–559.
(126 MHz, CDCl3): δ = 172.14, 155.33 (d, J = 36.5 Hz), 138.81,
132.95, 129.75, 128.58, 126.95, 125.86, 115.55 (d, J = 288.0 Hz),
66.48, 61.40, 30.89, 30.32, 29.13, 20.11, 18.88, 13.50 ppm. HRMS:
calcd. for C17H20F3NO3Na [M + Na]+ 366.1287; found 366.1274.
Compound (R)-8d: The hydrolysis of (R)-6d (58.00 mg, 0.18 mmol)
in aqueous HCl (6 m) produced (R)-8d (40.95 mg, 0.18 mmol,
99%); m.p. 267 °C (dec.). [α]2D4 = –63.2 (c = 0.33, MeOH). 1H NMR
(500 MHz, CD3OD): δ = 7.36–7.26 (m, 4 H), 2.92 (m, 2 H), 2.51
(m, 1 H), 2.21 (m, 1 H), 2.14 (m, 1 H), 1.94 (m, 1 H) ppm. 13C
NMR (126 MHz, CD3OD): δ = 174.04, 139.46, 132.02, 131.49,
130.86, 128.29, 128.21, 61.90, 33.21, 29.93, 19.64 ppm. HRMS:
calcd. for C11H14NO2 [M – Cl]+ 192.1019; found 192.0996.
[5]
[6]
M.-X. Wang, S.-J. Lin, J. Liu, Q.-Y. Zheng, Adv. Synth. Catal.
2004, 346, 439–445.
a) S. Gedey, A. Liljeblad, F. Fülöp, L. T. Kanerva, Tetrahedron:
Asymmetry 1999, 10, 2573–2581; b) X.-G. Li, L. T. Kanerva,
Org. Lett. 2006, 8, 5593–5596; c) A. Liljeblad, H.-M. Kavenius,
P. Tähtinen, L. T. Kanerva, Tetrahedron: Asymmetry 2007, 18,
181–191; d) A. Liljeblad, L. T. Kanerva, Tetrahedron 2006, 62,
5831–5854; e) L. T. Kanerva, A. Liljeblad, Transesterification –
Biological, Wiley Online Library, Encyclopedia of Catalysis,
2010; DOI: 10.1002/0471227617.eoc197.
S. Naik, A. B. Basu, R. Saikia, B. Madan, P. Paul, R. Chat-
erjee, J. Brask, A. Svendsen, J. Mol. Catal. B 2010, 65, 18–23.
L. T. Kanerva, O. Sundholm, J. Chem. Soc. Perkin Trans. 1
1993, 2407–2410.
Compound (S)-8d: Compound (S)-7d (60.0 mg, 0.17 mmol) was
transformed into (S)-8d (38.68 mg, 0.17 mmol, 99%). [α]2D4 = +66.7
(c = 0.33, MeOH).
Supporting Information (see footnote on the first page of this arti-
cle): 1H and 13C NMR spectra of products 8a–d, and GC chroma-
tograms of rac-1a–d and their kinetic resolution products.
[7]
[8]
[9]
L. T. Kanerva, P. Csomós, O. Sundholm, G. Bernát, F. Fülöp,
Tetrahedron: Asymmetry 1996, 7, 1705–1716.
Acknowledgments
[10] R. Kourist, P. D. de Maria, U. T. Bornscheuer, ChemBioChem
2008, 9, 491–498.
[11] T. A. Paál, A. Liljeblad, L. T. Kanerva, E. Forró, F. Fülöp,
Eur. J. Org. Chem. 2008, 5269–5276.
[12] a) N. W. J. T. Heinsman, C. G. P. H. Schroën, A. van der Padt,
M. C. R. Franssen, R. M. Boom, K. van’t Riet, Tetrahedron:
Asymmetry 2003, 14, 2699–2704; b) L. Veum, L. T. Kanerva,
P. J. Halling, T. Maschmeyer, U. Hanefeld, Adv. Synthesis Ca-
tal. 2005, 347, 1015–1021.
[13] M. W. Larsen, D. F. Zielinska, M. Martinelle, A. Hidalgo, L. J.
Jensen, U. T. Bornscheuer, K. Hult, ChemBioChem 2010, 11,
796–801.
[14] a) M. Martinelle, K. Hult, Biochim. Biophys. Acta 1995, 1251,
191–197; b) L. F. García-Alles, V. Gotor, Biotechnol. Bioeng.
1998, 59, 163–170.
[15] R. Warmuth, T. E. Munsch, R. A. Stalker, B. Li, A. Beatty,
Tetrahedron 2001, 57, 6383–6397.
[16] C.-S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih, J. Am. Chem.
Soc. 1982, 104, 7294–7299.
Financial support from the Finnish Funding Agency for Technol-
ogy and Innovation (Technology Program SYMBIO Project
#40168/07: Developing New Chemoenzymatic Methods and Bio-
catalysts) is gratefully acknowledged. X.-G. Li is also grateful for
the research funding (#133127) from the Academy of Finland.
[1] a) K. Ishizumi, N. Ohashi, N. Tanno, J. Org. Chem. 1987, 52,
4477–4485; b) P. L. Beaulieu, J. Gillard, M. D. Bailey, C. Bou-
cher, J.-S. Duceppe, B. Simoneau, J. Org. Chem. 2005, 70,
5869–5879; c) P. Rovero, M. Pellegrini, A. D. Fenza, S. Meini,
L. Quartara, C. A. Maggi, F. Formaggio, C. Toniolo, D. F. Mi-
erke, J. Med. Chem. 2001, 44, 274–278; d) P. Conti, M.
De Amici, G. Grazioso, G. Roda, A. Pinto, K. B. Hansen, B.
Nielsen, U. Madsen, H. Bräuner-Osborne, J. Egebjerg, V. Ves-
tri, D. E. Pellegrini-Giampietro, P. Sibille, F. C. Acher, C.
De Micheli, J. Med. Chem. 2005, 48, 6315–6325; e) A. Pesquet,
A. Daïch, L. Van Hijfte, J. Org. Chem. 2006, 71, 5303–5311.
[2] For instance, see: C. Toniolo, M. Crisma, F. Formaggio, C.
Peggion, Biopolymers (Pept. Sci.) 2001, 60, 396–419.
Received: November 23, 2010
Published Online: February 9, 2011
1762
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1755–1762