Chemoenzymatic preparation of the enantiomers of β-tryptophan ethyl ester and the β-amino nitrile analogue

Article abstract of DOI:10.1016/j.tetasy.2005.03.016

Racemic α-tryptophan was chemoselectively transformed into the enantiomers of β-tryptophan ethyl ester. The key step in achieving enantiopurity was the N-acylation of the 3-amino-4-(3-indolyl)butanenitrile intermediate with Candida antarctica lipase A (CAL-A). The enzymatic N-acylation of racemic β-tryptophan ethyl ester was also studied. CAL-A was highly (R)-enantioselective in the present kinetic resolutions, leading to a mixture of the butanamide product with an (R)-configuration and the unreacted starting material with an (S)-configuration at 50% conversion.

Full text of DOI:10.1016/j.tetasy.2005.03.016

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 1709–1714
Chemoenzymatic preparation of the enantiomers of
b-tryptophan ethyl ester and the b-amino nitrile analogue
Xiang-Guo Li and Liisa T. Kanerva*
Laboratory of Synthetic Drug Chemistry and Department of Chemistry, University of Turku,
Lemminka¨isenkatu 5 C, FIN-20520 Turku, Finland
Received 21 February 2005; accepted 15 March 2005
Available online 14 April 2005
Abstract—Racemic a-tryptophan was chemoselectively transformed into the enantiomers of b-tryptophan ethyl ester. The key step
in achieving enantiopurity was the N-acylation of the 3-amino-4-(3-indolyl)butanenitrile intermediate with Candida antarctica lipase
A (CAL-A). The enzymatic N-acylation of racemic b-tryptophan ethyl ester was also studied. CAL-A was highly (R)-enantioselec-
tive in the present kinetic resolutions, leading to a mixture of the butanamide product with an (R)-configuration and the unreacted
starting material with an (S)-configuration at 50% conversion.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
through kinetic and dynamic kinetic resolution, indicat-
ing that cyano-substituted substrates are well accepted
by the lipase.13,14 Herein we report chemoenzymatic
and safe ways for the preparation of ethyl 3-amino-4-
(1H-3-indolyl)butanoate [(R)- and (S)-5, Scheme 1]. In
order to introduce enantiopurity in the molecule the
focus was on the kinetic resolution of nitrile rac-4. Just
as cyanohydrins are important intermediates for various
applications, 3-aminoalkylnitriles serve as intermediates
for various kinds of synthetic purposes, for example,
(S)-4 has been used in the preparation of bridged indole
alkaloids of ajmaline–sarpagine family, dregamine and
epidregamine.8 The CAL-A-catalyzed acylation of rac-
5 as N-heterocyclic b-amino ester was also found to be
attractive in order to learn more about the behaviour
of CAL-A. Furthermore, the catalytic behaviour of C.
antarctica lipase B (CAL-B) has also been studied.
In recent years, there has been increased interest in b-
amino acids, as a result of their unique pharmacological
effects, their ability to modify biologically active pep-
tides and their importance in the preparation of pepti-
domimetics and natural products.1–3 Synthetic
methods leading to both racemic and enantiomeric b3-
amino acids and esters include the classical Rodionov
reaction from the corresponding aldehyde through con-
densation with malonic acid in the presence of ammo-
nium acetate, the Arndt–Eistert homologation from
the corresponding N-protected a-amino acid through
diazoketone intermediates and the stereoselective reduc-
tion of the corresponding enamines.4–7 Moreover, the
preparation of N-protected b-amino nitriles has been
described.8,9
Our interest has long been in the preparation of the
enantiomers of b-amino esters by using lipase catalysis
for enantioselective N-acylation. In these studies, it has
become clear that Candida antarctica lipase A (CAL-
A), bearing a GGGX motif in the oxyanion hole, is
highly applicable in the case of b3-amino esters.10–12
CAL-A has also proven to be an excellent catalyst for
the preparation of highly enantiopure cyanohydrin
esters (precursors for non-natural a-amino acids)
2. Results and discussion
As the basis to achieve the present work, we found two
interesting transformations of a-amino acid, L-trypto-
phan: the Arndt–Eistert methodology leading to the
N-protected b-tryptophan through the tryptophan-de-
rived diazoketone3 and the formation of the N-tosylated
nitrile through the tryptophan-derived ditosylated 2-
amino-1-propanol (S)-8 (Scheme 1).8 We concentrated
on the latter method in order to avoid using hazardous
diazomethane, and also because nitriles are known to
*
Corresponding author. Tel.: +358 2333 6773; fax: +358 2333 7955;
e-mail: lkanerva@utu.fi
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.03.016

1710
X.-G. Li, L. T. Kanerva / Tetrahedron: Asymmetry 16 (2005) 1709–1714
CO2H
NH2
CH2OH
CH2O-oNs
NH-oNS
CH2CN
NH-oNs
ii
i
NH2
N
H
N
H
N
H
N
H
rac-1
2
rac-3
vi
iii
CH2OTs
CH2CO2Et
NH2
CH2CN
iv
NHTs
NH2
N
H
N
H
N
H
rac-4
rac-8
v
rac-5
v
CH2CO2Et
CH2CO2Et
Pr
CH2CN
CH2CN
Pr
HN
+
+
NH2
HN
NH2
N
H
N
N
N
H
O
O
H
H
(S)-4
iv
(R)-7
(R)-6
(S)-5
iv
CH2CO2Et
NH2
CH2CO2Et
SO2--
NO2
SO2--
NH2
N
H
N
H
oNs=
Ts =
(S)-5
(R)-5
Scheme 1. Reagents and conditions: (i) 2-nitrobenzenesulfonyl chloride (2-NsCl), pyridine, CH2Cl2; (ii) KCN, methanol, reflux; (iii) thiophenol,
K2CO3 and CH3CN, room temperature; (iv) HCl (0.7 M) in ethanol, then NH4OH (1%); (v) an achiral ester and CAL-A or CAL-B in a solvent; (vi)
4-toluenesulfonyl chloride (TsCl), pyridine, CH2Cl2.
yield carboxylic acids in aqueous acids and esters in
acidic alcohols.
acted rac-4 and produced rac-5 were 30% and 64%,
respectively. The above indicates that rac-tryptophan
can be effectively transformed into rac-5.
For the preparation of b-tryptophan ethyl ester rac-5, a-
tryptophan was first reduced to the corresponding ami-
no alcohol rac-1 in tetrahydrofuran (THF) using lithium
aluminium hydride as previously described (Scheme 1).8
The next step concerned the evaluation of a suitable N-
protecting group. Tosylation has been used earlier, and
accordingly the ditosylate rac-8 was obtained easily.8
However, the cleavage of the N-tosyl group later re-
quired harsh conditions, such as the use of sodium in
liquid ammonia. As a result, the formation of two
unidentified products took place at the expense of the
desired compound. For these reasons, tosylation was re-
placed by nosylation using excess 2-nitrobenzenesulfo-
nyl chloride (2-NsCl) and pyridine. Due to the
possible anchimeric assistance of the amino nitrogen,
the presence of an N-2-nitrobenzene sulfonylated aziri-
dine in the reaction mixture containing the highly unsta-
ble product 2 cannot be excluded. As a support, N-4-
nitrobenzenesulfonyl-2-benzylaziridine was separated
when the amino alcohol was protected using 4-nitro-
benzenesulfonyl chloride.9 We used 2-NsCl rather than
4-NsCl because it was reported to cleave more readily
than the latter.15 Without separation, 2 was treated with
potassium cyanide yielding rac-3 at 81% isolated yield
from rac-1. The following transformation into rac-4
with thiophenol and potassium carbonate in acetonitrile
was accomplished without difficulty as described in the
Experimental section. Finally, rac-5 was produced when
rac-4 was treated with anhydrous HCl–EtOH under
gentle reflux. This transformation was very slow. Thus,
after the reaction of 2 days the isolated yields of unre-
The benefits of successful lipase-catalyzed kinetic resolu-
tions are that chiral induction is introduced by the cata-
lyst under mild reaction conditions and that both
enantiomers are simultaneously obtained indicating that
the reaction is highly enantioselective. Inspired by our
previous success in the kinetic resolution of b3-amino
esters,10–12 rac-4 was allowed to react with butanoate
or acetate esters in mixed solvents, containing tert-butyl-
methyl ether (TBME) or diisopropyl ether (DIPE) in the
presence of CAL-A. Due to the low solubility of polar
rac-4 in ether, acetonitrile (AN), butyl butanoate or
butylacetate was added. In rac-4, the indole nitrogen
did not react under the enzymatic reaction conditions.
Thus, the amino group was the only reactive functional
group. As shown in Table 1, the reaction proceeded with
low enantioselectivity in the presence of acetonitrile
(rows
1 and 2) while excellent enantioselectivity
(E >200) was observed when butyl butanoate acted as
an acyl donor and as a co-solvent in DIPE (row 4).
Enantioselectivity dropped when butylacetate replaced
the butanoate (row 4 compared to row 6). TBME gave
both lower reactivity (measured as conversion after
2 h) and enantioselectivity than DIPE (row 4 compared
to row 5). To this end, the gram-scale resolution of rac-4
with CAL-A in butyl butanoate/DIPE (1:1) was success-
fully accomplished in 2 h as described in the Experimen-
tal. The separated nitrile enantiomers (R)-7 (ee = 95%)
and (S)-4 (ee = 99%) were transformed into the amino
esters (R)- and (S)-5 by refluxing in 0.7 MHCl/EtOH
without any change in the original enantiopurities. It

X.-G. Li, L. T. Kanerva / Tetrahedron: Asymmetry 16 (2005) 1709–1714
1711
Table 1. N-Acylation of rac-4 (0.05 M) by CAL-A preparation (20 mg/mL) at room temperature
Entry
Acyl donor
Solvent
Time (h)
c (%)
Ee(S)-4
Ee(R)-7
E
1
2
3
4
5
6
7
8
PrCO2CH2CF3
PrCO2CH2CF3
TBME/ANa (2:1)
DIPE/ANa (2:1)
2.5
1
51
49
48
49
34
40
45
17
75
84
86
96
51
62
30
13
72
89
94
98
96
92
37
62
19
46
90
PrCO2Bu
3.5
2
PrCO2Bu/DIPE (1:1)
PrCO2Bu/TBME (1:1)
MeCO2Bu/DIPE (1:1)
>200
81
44
3
2
22
68
91
PrCO2Bub
b
PrCO2CH2CF3
TBME/ANa (2:1)
5
a AN is acetonitrile.
b CAL-B (50 mg/mL) as a lipase, temperature 47 ꢁC.
was interesting to recognize that C. antarctica lipase B
(CAL-B), as one of the most generally used lipase in
enantioselective reactions, non-enantioselectively ac-
cepted rac-4 as a substrate (rows 7 and 8).
accomplished, allowing the preparation of enantiopure
amino nitrile synthons (R)-7 and (S)-4 to further syn-
thetic purposes (Scheme 1).
CAL-A has been used for the kinetic resolution of O-
and S-heterocyclic b3-amino esters before,11 but this
work is the first time when N-heterocyclic b3-amino
esters were studied. For this purpose, rac-5 was sub-
jected to acylation in PrCO2Bu/DIPE (1:1) at room tem-
perature (Table 2). The reaction proceeded at excellent
enantioselectivity and practically stopped at 50% con-
version when the resolution products (R)-6 and (S)-5
were both at ee >99% (the other enantiomer was not ob-
served by the chiral HPLC method). However, the time
needed to reach the 50% conversion was considerably
longer for the acylation of rac-5 than for rac-4. The
reaction in the presence of CAL-B under otherwise the
same conditions was extremely slow (36% conversion
was reached after one week) and proceeded at low enan-
tio- and chemoselectivity in the reaction where the N-
acylation of the amino group and the exchange of the
ester group of rac-4 both took place.
4. Experimental part
4.1. Materials and methods
Tryptophan, 2-nitrobenzenesulfonyl chloride, thiophe-
nol, potassium cyanide, butyl butanoate and the sol-
vents were products of Aldrich or Fluka. 2,2,2-
Trifluoroethyl butanoate was prepared from butanoyl
chloride and the corresponding alcohol. All solvents
were of the highest analytical grade and were dried by
standard methods when necessary. CAL-A (lipase A
from C. antarctica, Chirazyme L5, lyo.) was purchased
from Boehringer–Mannheim. Before use, CAL-A was
adsorbed on celite by dissolving the enzyme (5 g) and
sucrose (3 g) in Tris–HCl buffer (250 mL, 20 mM,
pH = 7.9) followed by the addition of celite (17 g). The
mixture was dried by allowing the water to evaporate.
The preparation containing 20% (w/w) of the enzyme
was thus obtained. The enzyme preparation gave the ini-
tial rate 0.086 0.001 mmol/min/g for the acylation of
racemic valine methyl ester (0.1 M) with 2,2,2-trifluoro-
ethyl butanoate (0.2 M) in TBME (E = 13) as a standard
reaction. C. antarctica lipase B (CAL-B, Novozym 435)
was a generous gift from Novozyme. Preparative chro-
matographic separations were performed by column
chromatography on Merck Kieselgel 60 (0.063–
0.200 lm). TLC was carried out with Merck Kieselgel
60F254 sheets. If not otherwise stated, all enzymatic
reactions were performed at room temperature
(23 ꢁC).
Table 2. N-Acylation of a substrate (0.05 M) in PrCO2Bu/DIPE (1:1)
by CAL-A preparation (20 mg/mL) at room temperature
Substrate
Time (h)
c (%)
EeS
EeR
rac-4
rac-5
2
16
49
50
96
>99
98
>99
The absolute configuration of the less reactive nitrile
enantiomer was confirmed to be (S)-4 by the transfor-
mation of L-tryptophan to (S)-5 through the nitrile
intermediate (Scheme 1) and by comparing the chro-
matographic behaviour of the compounds. According
to the results, CAL-A furnishes N-acylation with high
(R)-enantiopreference for both the nitrile and ester
substrates.
The H and 13C NMR spectra were recorded on a Bru-
ker 500 spectrometer with tetramethylsilane (TMS) as
1
an internal standard. 1H–1H COSY, 1H–13C HQSC
and H–13C HMBC spectra were used for the assign-
1
ments of the chemical shifts. Mass spectra were taken
on a VG 7070E mass spectrometer. Optical rotations
were determined with a Perkin–Elemer polarimeter,
and [a]D values given in units of 10ꢀ1 deg cm2 gꢀ1. Melt-
ing points were measured on a Mettler FP80 instrument.
The determination of E was based on the equation
E = ln[(1 ꢀ c)(1 ꢀ ees)]/ln[(1 ꢀ c)(1 + ees)] with the use
of linear regression, E being the slope of a line
ln[(1 ꢀ c)(1 ꢀ ees)] versus ln[(1 ꢀ c)(1 + ees)].16
3. Conclusion
In conclusion we have reported an efficient chemoenzy-
matic method for the preparation of the enantiomers of
5 from racemic a-amino acid, tryptophan. More impor-
tantly, the capacity of CAL-A to resolve b-amino nitrile
rac-4 with excellent (R)-enantiopreference has been

1712
X.-G. Li, L. T. Kanerva / Tetrahedron: Asymmetry 16 (2005) 1709–1714
In a typical small-scale experiment, the lipase prepara-
tion was added to the solution of rac-4 or rac-5
(0.1 mmol) and an acyl donor in 2 mL of a solvent.
The progress of the reactions and the ee values were
followed by taking samples (0.1 mL) at intervals
and analyzing them by HPLC. The analyses were
conducted with the HP 1090 instrument at the wave-
length of 263 nm using a CHIRACEL-OD column
(0.46 · 25 cm) and a mixture of hexane and isopropyl
alcohol (9:1) as an eluent at the flow rate of 1 mL/min.
Before analysis, unreacted amine was derivatized with
propionic anhydride in order to achieve good baseline
separation.
m/z (relative intensity) = 384 (10), 186 (4), 130 (100),
103 (4), 77 (6); 1H NMR (acetone-d6, 500 MHz):
d = 2.92–2.97 (dd, J = 4.20, 15.76 Hz, 1H, CH2CN);
2.97–3.01 (dd, J = 5.50, 16.86 Hz, 1H, CH2CN); 3.08–
3.13 (dd, J = 9.61, 14.56 Hz, 1H, indole-CH2);
3.15–3.19 (dd, J = 5.09, 14.54 Hz, 1H, indole-CH2);
4.05–4.10 (m, 1H, CHCH2CN); 6.83 (br s, 1H, NH);
6.88–6.91 (t, J = 7.19 Hz, 1 arom. H); 7.00–7.04 (t,
J = 7.14 Hz, 1 arom. H); 7.19–7.23 (m, 2 arom. H);
7.32–7.38 (m, 2 arom. H); 7.52–7.63 (m, 3 arom. H);
9.85 (br s, 1H, NH). 13C NMR (acetone-d6, 126 MHz):
d = 25.83 (CH2CN), 30.60 (indole-CH2), 52.84
(CHCH2CN), 109.92, 112.41, 118.15, 118.95, 119.71,
122.07, 125.51, 125.53, 127.71, 130.37, 133.06, 133.52,
134.29, 137.51, 147.60.
4.2. Preparation of rac-2-amino-3-(1H-3-indolyl)propan-
1-ol 1
4.4. Preparation of rac-3-amino-4-(1H-3-indolyl)butane-
nitrile 4
The known method was followed.8 Lithium aluminum
hydride (3.75 g, 98.7 mmol) and racemic tryptophan
(5.00 g, 24.5 mmol) were suspended in dry tetrahydrofu-
ran (250 mL) under reflux for 15 h. The mixture was
cooled to 0 ꢁC followed by the treatment with saturated
sodium sulfate solution, filtration and evaporation to af-
Thiophenol (0.49 g, 4.45 mmol) was added to a mixture
of rac-3 (0.34 g, 0.88 mmol) and anhydrous K2CO3
(1.22 g, 8.85 mmol) in dry CH3CN (4 mL). After half
an hour, the starting material was totally consumed as
indicated by TLC. The solvent was evaporated and the
residue purified on silica gel by elution with dichloro-
methane/methanol (19:1), affording rac-4 (0.17 g,
0.87 mmol) at 99% yield. HRMS: M+ found (M+ calcu-
lated for C12H13N3): 199.11100 (199.11095); MS: m/z
(relative intensity) = 199 (M+, 10), 130 (100), 103 (7),
77 (10); 1H NMR (CDCl3, 500 MHz): d = 2.34–2.39
(dd, J = 6.42, 16.57 Hz, 1H, CH2CN); 2.46–2.50 (dd,
J = 5.06, 16.57 Hz, 1H, CH2CN); 2.88–2.93 (dd, J =
7.23, 14.34 Hz, 1H, indole-CH2); 2.96–3.00 (dd,
J = 6.27, 14.36 Hz, 1H, indole-CH2); 3.46–3.51 (m, 1H,
CHNH2); 7.07 (d, J = 2.08 Hz, 1 arom. H); 7.12–7.15
(t, J = 7.58 Hz, 1 arom. H); 7.20–7.23 (t, J = 7.29 Hz,
1 arom. H); 7.36–7.38 (d, J = 8.14 Hz, 1 arom. H);
7.60 (d, J = 7.92 Hz, 1 arom. H); 8.34 (br s, 1H, NH).
13C NMR (CDCl3, 126 MHz): d = 26.00 (CH2CN),
32.93 (indole-CH2), 48.81 (CHNH2), 111.32, 111.40,
118.36, 118.65, 119.68, 122.33, 123.02, 127.24, 136.38.
1
ford alcohol rac-1 (4.56 g, 24.0 mmol) at 98% yield. H
NMR (DMSO-d6, 500 MHz): d = 2.53–2.57 (dd,
J = 7.28, 14.10 Hz, 1H, indole-CH2); 2.76–2.80 (dd,
J = 5.82, 14.09 Hz, 1H, indole-CH2); 2.94–2.99 (m, 1H,
indole-CH2CH); 3.20–3.23 (dd, J = 6.73, 10.29 Hz,
1H, CH2OH); 3.33–3.36 (dd, J = 4.70, 10.30 Hz, 1H,
CH2OH); 6.95–6.98 (t, J = 7.42 Hz, 1 arom. H); 7.04–
7.07 (t, J = 7.48 Hz, 1 arom. H); 7.14 (s, 1 arom. H);
7.33–7.34 (d, J = 8.05 Hz, 1 arom. H); 7.53–7.54 (d,
J = 7.84 Hz, 1 arom. H); 10.82 (br s, indole-NH). 13C
NMR (DMSO-d6, 126 MHz): d = 29.65 (indole-CH2),
53.50 (indole-CH2CH), 66.09 (CH2OH), 111.19,
111.71, 118.00, 118.37, 120.65, 123.14, 127.51, 136.12.
4.3. Preparation of rac-3-(2-nitrobenzenesulfonyl)amino-
4-(3-indolyl)butanenitrile 3
2-Nitrobenzenesulfonyl chloride (2-NsCl) (10.0 g,
45.1 mmol) was added in one portion to a solution of
rac-1 (2.85 g, 15.0 mmol) in dry CH2Cl2 and pyridine
(15 mL, 2:1, v/v) at 0 ꢁC. The mixture was stirred over-
night at room temperature, diluted with CH2Cl2
(220 mL) and treated with saturated sodium chloride
solution (100 mL). The organic layer was separated
and the aqueous layer was washed with CH2Cl2
(3 · 100 mL). The combined organic layers were dried
over anhydrous Na2SO4, filtered and evaporated. The
resulted crude product 2 was used for next step without
further purification.
4.5. Preparation of rac-3-amino-4-(1H-3-indolyl)butanoic
acid ethyl ester 5
A solution of rac-4 (0.10 g, 0.50 mmol) in 0.7 MHCl in
ethanol was stirred for 48 h at 90 ꢁC followed by the
evaporation of the solvent. The residue was dissolved
in 10 mL chloroform/aqueous NH4OH (1%, 1:1). The
organic layer was separated and the aqueous layer
washed with chloroform (2 · 8 mL). The combined
organic layers were dried over anhydrous Na2SO4 and
the solvent evaporated. After column chromatography
eluting with dichloromethane/methanol (19:1), the unre-
acted rac-4 (0.03 g, 0.15 mmol) was recovered and rac-5
(0.08 g, 0.32 mmol) obtained at 65.0% yield as a semi-
solid product. HRMS: M+ found (M+ calculated for
C14H18N2O2): 246.13710 (246.13683); MS: m/z (relative
intensity) = 246 (9), 232 (33), 131 (24), 123 (100), 116
Potassium cyanide (1.46 g, 22.5 mmol) was added to a
solution of the above crude product 2 in MeOH
(50 mL) at room temperature. The mixture was gently
refluxed until TLC indicated the disappearance of the
starting material. The reaction mixture was treated with
silica gel (2 g) before the solvent was evaporated. rac-3
(4.67 g, 12.2 mmol, yield 81%, solid, mp 152–154 ꢁC)
was purified on silica gel by elution with petroleum
ether/ethyl acetate (6:2). HRMS: M+ found (M+ calcu-
lated for C18H16N4O4S): 384.08960 (384.08923); MS:
1
(14), 109 (11), 77 (16); H NMR (CDCl3, 500 MHz):
d = 1.24–1.27 (t, J = 7.09 Hz, 3H, CH3); 2.36–2.41 (dd,
J = 8.64, 16.00 Hz, 1H, CH2CO2); 2.55–2.59 (dd, J =
4.15, 15.99 Hz, 1H, CH2CO2); 2.77–2.81 (dd, J = 8.06,

X.-G. Li, L. T. Kanerva / Tetrahedron: Asymmetry 16 (2005) 1709–1714
1713
14.17 Hz, 1H, indole-CH2); 2.93–2.97 (dd, J = 5.37,
14.18 Hz, 1H, indole-CH2); 3.60–3.62 (m, 1H, CHNH2);
4.12–4.16 (q, J = 7.13 Hz, 2H, CO2CH2CH3); 7.07 (s, 1
arom. H); 7.11–7.14 (t, J = 7.49 Hz, 1 arom. H); 7.19–
1.24–1.27 (t, J = 7.12 Hz, 3H, CO2CH2CH3); 1.57–1.64
(m, 2H, COCH2CH2CH3); 2.10–2.13 (t, J = 7.34 Hz,
2H, COCH2CH2CH3); 2.47–2.51 (dd, J = 4.58,
16.00 Hz, 1H, CHCH2CO2); 2.52–2.56 (dd, J = 4.37,
16.58 Hz, 1H, CHCH2CO2); 2.97–3.02 (dd, J = 7.83,
14.54 Hz, 1H, indole-CH2); 3.08–3.12 (dd, J = 5.87,
14.54 Hz, 1H, indole-CH2); 4.12–4.16 (q, J = 7.11 Hz,
2H, CO2CH2CH3); 4.60–4.67 (m, 1H, CHNHCO);
7.22 (t, J = 7.65 Hz,
1 arom. H); 7.36–7.38 (d,
J = 8.10 Hz, 1 arom. H); 7.62–7.64 (d, J = 7.88 Hz, 1
arom. H); 8.14 (br s, 1H, NH). 13C NMR (CDCl3,
126 MHz): d = 14.18 (CH3), 32.41 (indole-CH2), 40.99
(CH2CO2), 48.70 (CHNH2), 60.45 (CO2CH2CH3),
111.18, 112.53, 119.00, 119.50, 122.16, 122.82, 127.57,
136.38, 172.62 (CO2).
7.03 (d, J = 2.20 Hz,
1 arom. H); 7.10–7.13 (t,
J = 7.78 Hz, 1 arom. H); 7.17–7.20 (t, J = 8.03 Hz, 1
arom. H); 7.35–7.37 (d, J = 8.12 Hz, 1 arom. H); 7.64–
7.66 (d, J = 7.87 Hz, 1 arom. H); 8.29 (br s, 1 H, NH).
13C NMR (CDCl3, 126 MHz): d = 13.65 (COCH2-
CH2CH3), 14.18 (CO2CH2CH3), 19.07 (COCH2-
CH2CH3), 29.48 (indole-CH2), 37.40 (CHCH2CO2),
38.83 (COCH2CH2CH3); 46.70 (CHCH2CO2), 60.67
(CO2CH2CH3); 111.18, 111.62, 118.89, 119.56, 122.11,
122.82, 127.72, 136.23, 172.21 (CO2CH2CH3), 172.76
(COCH2CH2CH3).
4.6. Gram-scale resolution of rac-4
CAL-A preparation (20 mg/mL) was added to a solu-
tion of racemic 4 (1.00 g, 5.02 mmol) in PrCO2Bu/DIPE
(1:1) (100 mL). The reaction was stopped at 51% conver-
sion by filtering off the enzyme. The filtrate was evapo-
rated and the residue purified on silica gel eluting with
dichloromethane/methanol (19:1), affording (S)-4,
20
D
CHCl3) and semi-solid (R)-7, 0.67 g, 2.49 mmol,
0.47 g, 2.38 mmol, ee = 99% and ½aŠ ¼ þ15:0 (c 0.9,
20
4.8. Preparation of rac-3-(1H-3-indolyl)-2-(tosylamino)-
propyl toluene-4-sulfonate 8
ee = 95%, ½aŠ ¼ þ32:5 (c 0.9, CHCl3). Spectral data
D
for (R)-7: HRMS: M+ found (M+ calculated for
C16H19N3O): 269.15260 (269.15281); MS: m/z (relative
intensity) = 269 (12), 182 (90), 130 (100), 103 (7), 77
(9); 1H NMR (CDCl3, 500 MHz): d = 0.89–0.92 (t,
J = 7.39 Hz, 3H, CH3); 1.59–1.66 (m, 2H, CH2CH3);
2.14–2.17 (t, J = 7.34 Hz, 2H, COCH2); 2.47–2.51 (dd,
J = 4.20, 16.77 Hz, 1H, CH2CN); 2.73–2.77 (dd, J =
5.48, 16.77 Hz, 1H, CH2CN); 3.05–3.09 (dd, J = 8.26,
14.61 Hz, 1H, indole-CH2); 3.15–3.19 (dd, J = 6.20,
14.61 Hz, 1H, indole-CH2); 4.45–4.54 (m, 1H,
CHNHCO); 5.80 (br s, 1H, NH); 7.11–7.16 (m, 2 arom.
H); 7.21–7.24 (t, J = 7.21 Hz, 1 arom. H); 7.34–7.39 (d,
J = 8.13 Hz, 1 arom. H); 7.62–7.63 (d, J = 7.91 Hz, 1
arom. H); 8.34 (br s, 1 H). 13C NMR (CDCl3,
126 MHz): d = 13.65 (CH3), 19.00 (CH2CH3), 22.26
(CH2CN), 28.92 (indole-CH2), 38.53 (COCH2), 46.29
(CHNHCO), 110.09, 111.48, 117.63, 118.50, 119.99,
122.57, 123.06, 127.14, 136.38, 173.20 (NHCO).
Compound 8 was prepared at 85% yield by the known
method.8 1H NMR (CDCl3, 500 MHz): d = 2.32 (s,
3H, CH3); 2.44 (s, 3H, CH3); 2.77–2.82 (dd, J = 6.98,
14.63 Hz, 1H, indole-CH2); 2.98–3.02 (dd, J = 7.00,
14.63 Hz, 1H, indole-CH2); 3.57–3.64 (m, 1H, indole-
CH2CH); 3.87–3.90 (dd, J = 5.65, 9.97 Hz, 1H, CH2O-
SO2); 4.08–4.11 (dd, J = 4.42, 9.98 Hz, 1H, CH2OSO2);
6.88–6.89 (d, J = 2.32 Hz, 1 arom. H); 6.95–6.98 (dt,
J = 0.84, 7.87 Hz,
1
arom. H); 7.01–7.03 (d,
J = 8.07 Hz, 2 arom. H); 7.12–7.18 (m, 2 arom. H);
7.27–7.29 (d, J = 8.14 Hz, 1 arom. H); 7.31–7.33 (d,
J = 8.01 Hz, 2 arom. H); 7.43–7.45 (d, J = 8.29 Hz, 2
arom. H); 7.74–7.76 (d, J = 8.31 Hz, 2 arom. H); 8.16
(br s, 1H, NH). 13C NMR (CDCl3, 126 MHz):
d = 21.52 (CH3), 21.69 (CH3), 27.32 (indole-CH2),
52.13 (indole-CH2CH), 70.60 (CH2OSO2), 109.39,
111.30, 118.23, 119.62, 122.15, 123.45, 126.75, 128.01,
129.45, 130.00, 132.31, 136.21, 136.29, 143.38, 145.19.
The obtained (S)-4 was transformed to (S)-5 at 60%
20
D
yield, ½aŠ ¼ ꢀ5:4 (c 1.0, CHCl3), ee = 99% and (R)-7
Acknowledgements
20
to (R)-5 at 63% yield, ½aŠ ¼ þ5:2 (c 1.0, CHCl3),
D
ee = 95% as described above.
The authors thank the Academy of Finland for financial
support (grant 75267 to L.K.).
4.7. Gram-scale resolution of rac-5
CAL-A preparation (20 mg/mL) was added to the solu-
tion of racemic 5 (1.00 g, 4.07 mmol) in PrCO2Bu/DIPE
(1:1) (81 mL). The reaction was stopped at 50% conver-
sion by filtering off the enzyme. The filtrate was evapo-
rated and the residue purified on silica gel eluting with
References
1. Juaristi, E. Enantioselective Synthesis of b-Amino Acids;
Wiley-VHC: New York, 1997.
2. Steer, D. L.; Lew, R. A.; Perlmutter, P.; Smith, A. I.;
Aguilar, M.-I. Curr. Med. Chem. 2002, 9, 811–822.
3. Seebach, D.; Kimmerlin, T.; Sebesta, R.; Campo, M. A.;
Beck, A. K. Tetrahedron 2004, 60, 7455–7506.
4. Rodionov, W. M.; Malivinskaya, E. Th. Chem. Ber. 1926,
59, 2952–2958.
5. Cimarelli, C.; Palmieri, G. J. Org. Chem. 1996, 61, 5557–
5563.
6. Cohen, J. H.; Abdel-Magid, A. F.; Almond, H. R., Jr.;
Maryanoff, C. A. Tetrahedron Lett. 2002, 43, 1977–
1981.
dichloromethane/methanol (19:1), affording (S)-5,
20
D
CHCl3), and semi-solid (R)-6, 0.62 g, 1.96 mmol,
0.48 g, 1.95 mmol, ee >99% and ½aŠ ¼ ꢀ5:4 (c 1.0,
20
ee >99%, ½aŠ ¼ þ8:0 (c 1.0, CHCl3). Spectral data for
D
(R)-6: HRMS: M+ found (M+ calculated for
C18H24N2O3): 316.17860 (316.17869); MS: m/z (relative
intensity) = 316 (11), 229 (100), 184 (10), 156 (31), 143
1
(15), 130 (50), 116 (27); H NMR (CDCl3, 500 MHz):
d = 0.88–0.91 (t, J = 7.39 Hz, 3H, COCH2CH2CH3);

1714
X.-G. Li, L. T. Kanerva / Tetrahedron: Asymmetry 16 (2005) 1709–1714
´
7. Cole, D. C. Tetrahedron 1994, 50, 9517–9582.
12. Solymar, M.; Kanerva, L. T.; Fulo¨p, F. Tetrahedron:
¨
Asymmetry 2004, 15, 1893–1897.
8. Kutney, J. P.; Eigendorf, G. K.; Matsue, H.; Murai, A.;
Tanaka, K.; Sung, W. L.; Wada, K.; Worth, B. R. J. Am.
Chem. Soc. 1978, 100, 938–942.
13. Paizs, C.; Toßsa, M.; Majdik, C.; Ta¨htinen, P.; Irimie,
F.-D.; Kanerva, L. T. Tetrahedron: Asymmetry 2003, 14,
619–627.
14. Paizs, C.; Ta¨htinen, P.; Toßsa, M.; Majdik, C.; Irimie,
F.-D.; Kanerva, L. T. Tetrahedron 2004, 60, 10533–10540.
15. Miller, S. C.; Scanlan, T. S. J. Am. Chem. Soc. 1997, 119,
2301–2302.
`
9. Sutton, P. W.; Bradley, A.; Farras, J.; Romea, P.;
Urpı, F.; Vilarrasa, J. Tetrahedron 2000, 56, 7947–
`
7958.
10. Gedey, S.; Liljeblad, A.; Lazar, L.; Fulo¨p, F.; Kanerva, L.
¨
´ ´
T. Tetrahedron: Asymmetry 2001, 12, 105–110.
´
11. Solymar, M.; Fulo¨p, F.; Kanerva, L. T. Tetrahedron:
¨
Asymmetry 2002, 13, 2383–2388.
16. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J.
J. Am. Chem. Soc. 1982, 104, 7294–7299.