Directed (R)- or (S)-selective dynamic kinetic enzymatic hydrolysis of 1,2,3,4-tetrahydroisoquinoline-1-carboxylic esters

Article abstract of DOI:10.1002/ejoc.200800789

The first synthesis of both enantiomers of 6,7-dimethoxy-1,2,3,4- tetrahydroisoquinoline-1-carboxylic acid was accomplished through dynamic kinetic resolution in procedures based on CAL-B- or subtilisin Carlsberg-catalysed enantioselective hydrolysis of the corresponding ethyl esters in aqueous NH₄OAc buffer at pH 8.5. The products were obtained with high enantiopurity (92-93%ee) in good yields (85-92%). (R)-1,2,3,4-Tetrahydroisoquinoline-1-carboxylic acid was obtained with high enantiopurity (98%ee) and in good yield (85%) in a CAL-B-catalysed process, under similar conditions. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200800789

FULL PAPER
DOI: 10.1002/ejoc.200800789
Directed (R)- or (S)-Selective Dynamic Kinetic Enzymatic Hydrolysis of
1,2,3,4-Tetrahydroisoquinoline-1-carboxylic Esters
Tihamér A. Paál,[a,b] Arto Liljeblad,[b] Liisa T. Kanerva,*[b] Eniko˝ Forró,[a] and
Ferenc Fülöp*[a]
Keywords: Dynamic kinetic resolution / Enzyme catalysis / Nitrogen heterocycles
The first synthesis of both enantiomers of 6,7-dimethoxy-
1,2,3,4-tetrahydroisoquinoline-1-carboxylic acid was ac-
complished through dynamic kinetic resolution in pro-
cedures based on CAL-B- or subtilisin Carlsberg-catalysed
enantioselective hydrolysis of the corresponding ethyl esters
(85–92%). (R)-1,2,3,4-Tetrahydroisoquinoline-1-carboxylic
acid was obtained with high enantiopurity (98%ee) and in
good yield (85%) in a CAL-B-catalysed process, under sim-
ilar conditions.
in aqueous NH4OAc buffer at pH 8.5. The products were ob- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
tained with high enantiopurity (92–93%ee) in good yields
Germany, 2008)
A small number of chemical methods are available for the
preparation of the enantiomers of 1-TIC.[10] 6,7-Dimeth-
oxy-1-TIC enantiomers have not yet been isolated and char-
acterised, but some enantiopure derivatives have been pre-
pared as intermediates in the synthesis of calycotomine.[11]
Methods have been reported for the preparation of 1-TIC
and 6-hydroxy-1-TIC enantiomers by chemoenzymatic ki-
netic resolution through ester hydrolysis.[12] When the aim
is the synthesis of a single enantiomer, dynamic kinetic res-
olution (DKR) methods are more effective than simple ki-
netic resolutions, as they provide 100% theoretical yield for
the desired enantiomer. The importance of DKR methods
is demonstrated by the large number of recent publications
in this area.[13] General biocatalytic DKR processes have
been developed for some groups of acyclic α-aryl-α-amino
acids.[14] A CAL-B-catalysed (R)-selective DKR of 1-TIC
ethyl ester by using enzymatic ester hydrolysis in organic
solvents was recently published by our group.[12a] Although
the method provided (R)-1-TIC in 80% yield (96%ee) the
space-time yield was relatively low (a reaction mixture con-
taining 0.05  substrate concentration required 6 d to attain
conversions Ն99%). In the present work, we set out to de-
velop an efficient chemoenzymatic DKR method for the
preparation of enantiopure 6,7-dimethoxy-1-TIC on the ba-
sis of enzymatic hydrolysis in water. A similar DKR process
with reversed selectivity has also been studied in the cases
of 6,7-dimethoxy-1-TIC and 1-TIC.
Introduction
Isoquinolines of either natural or synthetic origin have
wide-ranging therapeutic applications. Some 1,2,3,4-tetra-
hydroisoquinoline antibiotics exert potent antitumour ac-
tivity,[1] and trabectedin (formerly ecteinascidin-743) was re-
cently introduced into the therapy of cancer.[2] A few deriv-
atives of 1,2,3,4-tetrahydroisoquinoline-1-carboxylic acid
(1-TIC) occur naturally [e.g., peyoxylic acid,[3] identified in
a plant, and (–)-salsolinol-1-carboxylic acid,[4] in the human
brain]. As a building block, 1-TIC can be used in the syn-
thesis of promising matrix metalloproteinase inhibitors.[5] 1-
TIC is also important as a conformationally constrained
analogue of phenylglycine, present in the structures of sev-
eral biologically active synthetic peptides.[6] The replace-
ment of amino acids with conformationally constrained an-
alogues allows elucidation of the minimum conformational
requirements for biological activity and the selectivity for
different targets can be increased.[7] Most isoquinoline alka-
loids with broad therapeutic applications contain 6,7-di-
methoxy or 6,7-methylenedioxy moieties (e.g., papaverine,
noscapine, emetine, atracurium and mivacurium),[8] which
make the molecules less hydrophobic and therefore more
drug-like. 6,7-Dimethoxy-1-TIC has been applied in the
synthesis of β-site amyloid precursor protein-cleaving en-
zyme inhibitors, which are useful in Alzheimer’s disease.[9]
[a] Institute of Pharmaceutical Chemistry, University of Szeged,
Eötvös utca 6, 6720 Szeged, Hungary
Fax: +36-62-545-705
Results and Discussion
E-mail: fulop@pharm.u-szeged.hu
[b] Department of Pharmacology, Drug Development and Thera-
peutics/Laboratory of Synthetic Drug Chemistry, University of
Turku,
Syntheses of Substrates (؎)-1b and (؎)-2b
The amino ester substrates were synthesised according to
literature methods: 1-TIC ethyl ester (Ϯ)-1b was synthesised
Lemminkäisenkatu 5C, 20250 Turku, Finland
E-mail: lkanerva@utu.fi
Eur. J. Org. Chem. 2008, 5269–5276
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5269

T. A. Paál, A. Liljeblad, L. T. Kanerva, E. Forró, F. Fülöp
FULL PAPER
by partial hydrogenation of isoquinoline-1-carboxylic acid pH profiles of the immobilised CAL-B preparations have
and esterification with EtOH/SOCl2.[12a] 6,7-Dimethoxy-1- not been thoroughly studied, according to Table 1 it can
TIC ethyl ester (Ϯ)-2b was synthesised by the Bischler– be concluded that the conversions attained after 4 or 8 h
Napieralsky cyclisation of homoveratrylamine hemioxal- decreased persistently as the pH was lowered from 8.0 (in
amide ethyl ester (formed by heating homoveratrylamine the case of H2O, the pH decreased during the reaction).
and diethyl oxalate at reflux) and subsequent partial hydro- The conversion depends on the rate of the enzymatic hy-
genation.[15]
drolysis and indirectly on the rate of racemisation of 2b
(this latter increases with the pH during a base-catalysed
process),[14] whereas, the effect of pH on the enantiopurities
of (R)-2a is moderate. The minor variations of eep with the
pH and implicitly with the overall reaction rate prompted
us to investigate the chemical hydrolysis of (Ϯ)-2b in dif-
ferent buffer systems (Figure 1). It turned out that the
chemical hydrolysis of (Ϯ)-2b is significant except for the
hydrolysis in malate buffer at pH 5.0, and the reaction is
generally faster at higher pH values (after 12 h, 28–32%
hydrolysed in phosphate and Tris buffer at pH = 7–8,
whereas only 4% hydrolysed at pH = 5.0).
CAL-B-Catalysed (R)-Selective DKR of (؎)-2b
In view of the previously reported CAL-B-catalysed ki-
netic resolution of (Ϯ)-1b,[12a] we first attempted the hydrol-
ysis of (Ϯ)-2b in organic solvents. CAL-B provided excel-
lent (R) selectivity in diisopropyl ether (DIPE) at 25 °C.
However, slow spontaneous racemisation of unreacted (S)-
2b was observed, as indicated by the observed enantiopuri-
ties [ees = 88%, eep Ն99% (ees and eep stand for the enan-
tiomeric excess of the substrates and the products, respec-
tively)] at 79% conversion after 14 d. Reactions with basic
additives were performed in toluene/MeCN (4:1) with
added water (1 equiv.) and CAL-B (20 mgmL–1) at 25 °C.
The use of mild bases (iPr2NH or Amberlite IRA-904 resin)
resulted in a slow enzymatic reaction and slow racemisation
[conv. = 54%, ees = 64% and eep Ն99% with Amberlite
IRA-904 (15 mgmL–1) after 6 d]. Strong organic bases such
as 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU, 1 equiv.) gave
almost racemic products (due to chemical hydrolysis or ra-
cemisation). After the failure of our attempts to develop a
real-time DKR in organic solvent, we turned to the CAL-
B-catalysed hydrolysis of (Ϯ)-2b in aqueous medium
(Table 1, Scheme 1).
The enzymatic reaction was associated with fast racemis-
ation of the unreacted substrate, leading to conv. Ն99%
and eep = 75–86% in 1–4 d. Both the nature of the buffer
and pH usually have an impact on enzymatic activity and
enantioselectivity. CAL-B typically works over a wide pH
Figure 1. The chemical hydrolysis of (Ϯ)-2b (0.02 ) in different
aqueous buffers (0.1 ) at 25 °C.
In Tris-HCl buffer, under the conditions given in Table 1,
range in such a way that the hydrolysis rate increases when eep increased along with the amount of enzyme from eep =
the pH is raised from 5 to 7 and decreases between pH 7 62% at 10 mgmL–1 enzyme to eep = 86% at 40 mgmL–1
and 9[16] (the optimum pH = 7 for CAL-B).[17] Although after 24 h. Accordingly, we concluded that the main de-
Table 1. Effects of different aqueous media on the conversion and enantiopurity of (R)-2a in the CAL-B-catalysed hydrolysis of (Ϯ)-2b.[a]
Solvent or buffer salt/pH
H2O/7.3Ǟ6.0
Tris/8.0
Phosphate/7.0
Malonate/6.0
Citrate/5.5
Malate/5.0
Conv. after 4 h [%][b]
Conv. after 8 h [%][b]
eep [%][c]
77
89
69
99
57
93
42
72
46
59
54
61
79[d]
83[d]
75[d]
82[d]
83[e]
86[e]
[a] 0.02  (Ϯ)-2b in the tested solvent (0.1  buffer salt in H2O for buffered reactions), with 20 mgmL–1 CAL-B at 25 °C. [b] Determined
by GC after extraction of the substrate. [c] Determined by HPLC. [d] Samples taken after 24 h (conv. Ն99%). [e] Samples taken after 4 d
(conv. Ն99%).
Scheme 1. (R)-Selective DKR of (Ϯ)-2b by CAL-B-catalysed hydrolysis in the aqueous phase.
5270
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 5269–5276

Directed (R)- or (S)-Selective Dynamic Kinetic Enzymatic Hydrolysis
ficiency of these reactions is the fast chemical hydrolysis,
One reason for the effectiveness of NH4OAc is the rela-
and the enantioselectivity of the enzymatic reaction is tively slow chemical hydrolysis of (Ϯ)-2b in this medium
otherwise satisfactory. As may be expected from the oppo- (relative to H2O and phosphate buffer at pH = 7.0 in Fig-
site effects of the enzymatic activity and the chemical hy- ure 1). Another important benefit of NH4OAc is its volatil-
drolysis upon eep, the enantiopurities obtained with dif- ity: it can be removed from the product by evaporation,
ferent buffers (Table 1) present little variation.
which simplifies the purification. It is likely that NH4OAc
We tested the reactions at different temperatures (3, 25 (at 0.1–0.3 ) inhibits the enzyme activity (Table 2, entry 8
and 47 °C). Hydrolysis at 25 °C provided the highest eep for vs. 9); this could also be observed with Tris-HCl and Na-
each tested buffer (malate at pH = 5.0, phosphate at pH = phosphate buffers as compared with H2O (Table 1, conv.
7.0 and Tris at pH = 8.0).
after 4 h). The inhibition potencies of some hydrolases
The chemical natures of the buffer salts significantly in- (trypsin, α-chymotrypsin, renal acylase, β-amylase, etc.) in
fluenced the chemical hydrolysis, which is slightly faster in response to the action of anions follow the Hofmeister
phosphate buffer (pH = 7.0) than in water or NH4OAc at series (AcO– Ͻ Cl– Ͻ Br–, NO3– Ͻ I– Ͻ ClO4– Ͻ SCN–).[19]
the same pH (Figure 1).
We subsequently used 0.1  NH4OAc as the optimum
Accordingly, we compared the eep obtained with dif- buffer system and studied further the effects of the pH and
ferent buffer salts in the same pH ranges (Table 2). The sub- the enzyme and substrate concentrations on the conversion
strate concentration was increased from 0.02 to 0.1 .
and eep (Table 3).
Table 3. Optimisation of the amounts of enzyme and solvent in the
Table 2. Effects of buffers in different pH ranges on the conversion
and enantiopurity of (R)-2a in the CAL-B-catalysed hydrolysis of
(Ϯ)-2b.[a]
(R)-selective CAL-B-catalysed DKR of (Ϯ)-2b.[a]
Entry Initial conc.
Enzyme
pH
Time until Ն99%
eep
of (Ϯ)-2b [M] [mgmL–1
]
conv. [h][b]
[%][c]
Entry Buffer salt Starting Final Conv. after Conv. after
eep
pH[b]
pH
24 h [%][c]
48 h [%][c]
[%][d]
1
2
0.1
0.1
0.1
0.1
0.2
50
50
50
50
100
7.0
16
16
6
7
7
87
84
87
89
92
7.0[d]
8.5[d]
8.5[d]
8.5[d]
1
2
3
4
5
6
7
8
9
phosphate
malonate
malate
malate
malate
7.0
6.0
5.0
6.0
7.0
5.0
6.0
7.0
7.0
8.0
6.7
5.6
4.7
4.9
5.1
4.9
5.3
5.6
5.2
6.4
Ն99
Ն99
92
94
96
53
65
88
Ն99
Ն99
Ն99
Ն99
94
94
Ն99
86
77
86
86
81
90
83
92
92
92
88
3[e]
4
5
[a] (Ϯ)-2b in aqueous NH4OAc (0.1 ) in the given concentration
at 25 °C. [b] The conversions were determined by GC after extrac-
tion of the unreacted substrate. [c] Determined by HPLC (after
conv. Ն99%). [d] The pH of the reaction mixture was adjusted from
time to time to the given value with NH4OH (0.5 ) in H2O. [e] In
H2O, without added buffer salt.
NH4OAc[e]
NH4OAc[e]
NH4OAc[e]
NH4OAc
NH4OAc
98
Ն99
Ն99
Ն99
10
[a] 0.1  (Ϯ)-2b in the tested buffer (0.1  buffer salt) with
75 mgmL–1 CAL-B at 25 °C. [b] The pH of the reaction mixture
was adjusted to the given value before the enzyme was added.
[c] Determined by GC after extraction of the unreacted substrate.
[d] Determined by HPLC (after conv. Ն99%). [e] The reaction mix-
ture contained NH4OAc (0.3 ); the initial pH was adjusted with
AcOH.
The decreased enzyme/substrate ratio (2:1, m/m) fur-
nished a worse eep (Table 3, entry 1), as expected because
of the marked chemical hydrolysis of (Ϯ)-2b. For reasons
of economy we subsequently used this ratio. Keeping the
pH constant at 7.0 decreased the value of eep (Table 3, en-
try 2). The good reactivity and relatively high eep obtained
at pH = 8.0 (Table 1) encouraged us to test the reaction at
pH Ն 8.0 in NH4OAc. The reactivity was improved to a
noteworthy extent and the value of eep was found to be
higher at pH = 8.5 (Table 3, entry 4). With a doubled sub-
strate concentration (0.2 ) and an identical enzyme/sub-
strate ratio the reaction required the same time until com-
pletion and furnished a better eep (Table 3, entry 5 vs. 4).
Preparative-scale DKR with (Ϯ)-2b (0.2 ) under the opti-
mum conditions (in 0.1  aqueous NH4OAc solution, with
100 mgmL–1 CAL-B at constant pH = 8.5 at 25 °C;
Scheme 1) furnished (R)-2a with 92% enantiopurity in 85%
yield after 7 h (Table 6, entry 1).
Ions can also act on the rate of the enzymatic hydrolysis
by specific binding (allosteric modification) or by stabilis-
ing/destabilising the enzyme structure in the aqueous
phase.[18] This latter effect (salting-out/salting-in) is the
most general; the influence on the enzyme activity follows
the Hofmeister series (with a few exceptions).[18a]
For practical reasons, our aim was to identify the circum-
stances under which the salt effect on the two competing
reactions (enzymatic and chemical hydrolysis) has the most
favourable influence on eep. The reactions performed in
solutions containing malate (0.1 ) and especially NH4OAc
(0.1 or 0.3 ) in the pH range of 7.0–5.1 or 7.0–5.6 (Table 2,
entries 5, 8 and 9) gave products with enantiopurities supe-
rior to those obtained with phosphate and malonate buffers
(Table 2, entries 1 and 2). NH4OAc generally has favourable
effects on enzyme activities (compared to other salts), where
Subtilisin-Catalysed (S)-Selective DKR of (؎)-2b
Enzyme-catalysed DKR methods with opposite enantio-
AcO– is a kosmotropic anion (preceded in this respect by selectivities for the same substrate allow the directed syn-
F–, SO42– or HPO42–) and NH4+ is a chaotropic cation (pre- thesis of both enantiomers in high yields. Although very
ceded by tetraalkyl ammonium).[18a]
useful, there are only a few compounds for which such pro-
Eur. J. Org. Chem. 2008, 5269–5276
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5271

T. A. Paál, A. Liljeblad, L. T. Kanerva, E. Forró, F. Fülöp
FULL PAPER
Scheme 2. (S)-Selective DKR of (Ϯ)-2b by subtilisin Carlsberg-catalysed hydrolysis in the aqueous phase.
cesses have been developed: the DKRs of some secondary (Table 4, entries 1 and 2). This demonstrates better enantio-
alcohols (1-phenylethanol and related compounds) were ac- selectivity at lower temperature. A doubled substrate con-
complished with opposite enantioselectivities by CAL-B- centration (with maintained enzyme/substrate ratio) fav-
and subtilisin-catalysed acylation, and Ru complexes were oured the space-time yield (the reaction required the same
used as hydrogen-transfer catalysts for racemisation.[20] En- interval to proceed), and the value of eep also improved
zyme screening for the hydrolysis of (Ϯ)-2b (0.05 ) in (Table 4, entry 7 vs. 6 and 9 vs. 8). A gram-scale resolution
DIPE saturated with water was performed with 20 mgmL–1 under the optimum conditions [0.2  (Ϯ)-2b in 0.1  aque-
enzymes at 25 °C. Of the eight enzymes tested only Alcalase ous NH4OAc solution with 80 mgmL–1 Alcalase at con-
(a cross-linked enzyme aggregate from subtilisin Carlsberg) stant pH = 8.5 at 3 °C; Scheme 2] afforded (S)-2a with
displayed modest (S) selectivity (E = 6). The selectivity was 93%ee in 92% yield (Table 6, entry 2).
slightly better in tert-butyl methyl ether (TBME; E = 13).
As to the given E values, they may contain contributions of
chemical hydrolysis and spontaneous racemisation, al-
though they were calculated at an early stage of hydrolysis.
CAL-B-Catalysed (R)-Selective DKR of (؎)-1b
Alcalase generally preferentially hydrolyses -amino acid es-
The relatively fast racemisation of the enantiomers of 1b
ters.[21] We attempted the hydrolysis by using a basic pH in aprotic polar organic solvents is already known.[12a] Also
range (Table 4) favourable for subtilisin (pH optimum 6– known is that the CAL-B-catalysed hydrolysis of (Ϯ)-1b in
9).[21] The racemisation of (R)-2b took place when the Alca- organic solvents is enantioselective at 25 °C.[12a] After the
lase-catalysed hydrolysis was performed in Tris buffer (pH success with the CAL-B-catalysed DKR of (Ϯ)-2b in water,
= 8.0) at 25 °C, furnishing (S)-2a with eep = 61% after total (Ϯ)-1b was subjected to hydrolysis with CAL-B in aqueous
conversion (Table 4, entry 1). It was not possible to enhance media, buffered with NH4OAc (0.1 , Table 5). Perform-
the enantiopurity of (S)-2a by increasing the enzyme con- ance of the reaction at 25 °C provided (R)-1a with eep
=
tent (Table 4, entry 2). NH4OAc buffer (0.1 ) was once 91% after total conversion in less than 20 h (Table 5, en-
again preferred over Tris-HCl, and it was used during fur- try 1).
ther optimisations (Table 4, entry 3). Acetate salts pre-
viously proved to be the most potent activators of lyophi-
lised subtilisin Carlsberg in transesterifications carried out
in hexane.[22] Increase of the pH to 8.5 improved the enantio-
purity of the product (Table 4, entry 4). When the reaction
was performed at 3 °C, eep increased significantly (Table 4,
entry 6), and it could be improved further by increasing the
enzyme amount (Table 4, entries 5, 6 and 8), unlike at 25 °C
Table 5. Effects of the initial pH, temperature and enzyme concen-
tration on the conversion and enantiopurity of (R)-1a in the (R)-
selective CAL-B-catalysed DKR of (Ϯ)-1b.[a]
Entry Starting
pH[b]
Enzyme
Temp. Conv. after Conv. after
[°C]
eep
[mgmL–1
]
20 h [%][c] 48 h [%][c]
[%][d]
1
2
3
4
5
8
8
8
8
7
20
10
20
30
20
25
3
3
3
3
Ն99
86
97
Ն99
50
Ն99
Ն99
Ն99
Ն99
78
91
93
98
98
93
Table 4. Effects of buffers, pH, temperature and enzyme concentra-
tion on the conversion and enantiopurity of (S)-2a in the (S)-selec-
tive Alcalase-catalysed DKR of (Ϯ)-2b.
[a] 0.1  (Ϯ)-1b in aqueous NH4OAc (0.1 ). [b] The pH of the
reaction mixture was adjusted to the given value before the enzyme
was added. [c] Determined by GC after extraction of the unreacted
substrate. [d] Determined by HPLC (after conv. Ն99%).
Entry Conc. of
Enzyme
Buffer/
pH[a]
Temp. Time until
eep
[%][c]
(Ϯ)-2b [] [mgmL–1
]
[°C]
Ն99% conv.
[d][b]
1
2
3
4
5
6
7
8
9
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.1
0.2
10
20
20
20
10
20
40
40
80
Tris-HCl/8.0
Tris-HCl/8.0
NH4OAc/8.0 25
NH4OAc/8.5 25
NH4OAc/8.5
NH4OAc/8.5
NH4OAc/8.5
NH4OAc/8.5
NH4OAc/8.5
25
25
1
0.5
0.5
0.5
6
4
4
3
3
61
60
64
69
70
81
84
86
93
In order to improve the enantioselectivity, the reaction
was attempted at 3 °C. The value of eep increased to 98%
(Table 5, entry 3). Modification of the starting pH to 7.0
decreased the conversion and the enantiopurity of (R)-1a
(Table 5, entry 5).
3
3
3
3
3
Enzyme contents over 20 mgmL–1 did not influence the
value of eep. A gram-scale DKR with (Ϯ)-1b (0.1 ) under
the optimum conditions (in 0.1  aqueous NH4OAc solu-
tion with starting pH = 8.0 at 3 °C; Scheme 3) furnished
(R)-1a in isolated 85% yield (eep = 98%) after 24 h (Table 6,
[a] Buffers at 0.1 . The pH of the reaction mixture was adjusted
from time to time to the given value with NH4OH (0.5 ) in H2O.
[b] The conversions were determined by GC after extraction of the
unreacted substrate. [c] Determined by HPLC (after conv. Ն99%). entry 3).
5272
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 5269–5276

Directed (R)- or (S)-Selective Dynamic Kinetic Enzymatic Hydrolysis
Table 6. Preparative scale DKRs on the studied amino ester substrates.
Entry Substrate/conc. [][a]
Enzyme
pH
Temperature
Time
[h][b]
Product
Yield [%]
eep [%][c]
[mg mL–1]
[°C]
1
2
3
(Ϯ)-2b/0.2
(Ϯ)-2b/0.2
(Ϯ)-1b/0.1
CAL-B/100
Alcalase/80
CAL-B/20
Const.8.5
Const.8.5
8.0Ǟ6.5
25
3
3
7
72
24
(R)-2a
(S)-2a
(R)-1a
85
92
85
92
93
98
[a] The substrates were dissolved in aqueous NH4OAc (0.1 ) solution. [b] The time required until conv. Ն99%. [c] Determined by HPLC.
edge, this is the first procedure for the preparation of both
enantiomers of an amino acid through enzyme-catalysed
DKR. The methods are based on the CAL-B- or subtilisin
Carlsberg-catalysed enantioselective hydrolysis of 6,7-di-
methoxy-1-TIC ethyl ester in the aqueous phase and the
simultaneous base-catalysed racemisation of the unreacted
Scheme 3. (R)-Selective DKR of (Ϯ)-1b by CAL-B-catalysed hy-
amino ester enantiomer. 0.1  NH4OAc proved useful in
drolysis in the aqueous phase.
both cases: it stabilises the pH of the reaction, keeps the
chemical hydrolysis at a relatively low level and can be elim-
inated easily due to its volatility. The gram-scale resolutions
In comparison to the similar DKR in organic sol-
vent,[12a] the most evident difference is the 12-fold increase
provided the enantiomers with good enantiopurities (92–
in the space-time yield (0.1  substrate instead of 0.05 
93%) in high yields (85–92%). Followed by the LiAlH4-
required 1 d until completion, in contrast with 6 d in the
mediated reduction of the amino acid products, the method
organic medium). Moreover, H2O as solvent is considerably
serves for the biocatalytic preparation of both enantiomers
more environmentally friendly than toluene/MeCN (4:1).
of calycotomine. The CAL-B-catalysed DKR of 1-TIC
The process is enzyme economic (1:1 enzyme/substrate
mass ratio instead of 4:1) and both the yield (85%) and
ethyl ester was also optimised in an aqueous solution of
NH4OAc (0.1 ): the preparative-scale resolution at 3 °C
eep (98%) were slightly improved relative to the published
resulted in (R)-1-TIC with high enantiopurity (98%ee) and
method.[12a] The gram-scale DKRs are listed in Table 6.
in good yield (85%).
Unfortunately, the Alcalase-catalysed hydrolysis of (Ϯ)-
1b did not provide the desired (S)-1a enantiomer. Instead,
Alcalase displayed weak (R) selectivity. The reaction with
(Ϯ)-1b under the conditions described for (Ϯ)-2b (Table 4,
entry 8) furnished (R)-1a (eep = 20%) after conv. Ն99%
Experimental Section
Materials and Methods: CAL-B (lipase B from Candida antarctica),
produced by submerged fermentation of a genetically modified As-
pergillus oryzae microorganism and adsorbed on a macroporous
resin, was purchased from Sigma–Aldrich. Alcalase, a cross-linked
enzyme aggregate from subtilisin Carlsberg (a serine-protease from
Bacillus licheniformis) was a product of Fluka. Isoquinoline-1-car-
boxylic acid, applied in the synthesis of (Ϯ)-1b, was purchased from
Sigma–Aldrich, whereas homoveratrylamine [2-(3,4-dimeth-
oxyphenyl)ethylamine, used in the synthesis of (Ϯ)-2b] was pur-
chased from Chemos. Octadecane (used as a standard for GC) was
obtained from Fluka. Solvents were of the highest analytical grade.
DIPE was a product of Fluka, and TBME was purchased from
Sigma–Aldrich.
achieved in 3 d.
The Absolute Configurations of the Enantiomeric Products
Compound (R)-3, the (R) enantiomer of the alkaloid ca-
lycotomine [(R)-(6,7-dimethoxy-1,2,3,4-tetrahydroisoquin-
olin-1-yl)methanol] was synthesised by the reduction of (R)-
2a (95%ee) with LiAlH4 (Scheme 4). The optical rotation
of (R)-3 (91%ee) corresponds to that of (R)-calycotomine:
[α]2D5 = –33 (c = 0.2, H2O), in agreement with the literature
data [α]2D5 = +37.9 (c = 0.2, H2O) for (S)-calycotomine.[23]
In small-scale experiments carried out in organic solvents with
CAL-B (20, 40 or 50 mgmL–1), the enzyme was added to (Ϯ)-2b
(0.05 ) in DIPE or toluene/MeCN (4:1). Additives such as H2O
(0 or 1 equiv.), organic bases [iPr2NH, 1,8-bis(dimethylamino)-
naphthalene or DBU, 0 or 1 equiv.] or a basic resin (Amberlite
IRA-904, 0 or 15 mgmL–1), and an internal standard (octadecane,
0.012 ) were applied. The Alcalase (10 mgmL–1) reaction mixtures
contained (Ϯ)-2b (0.05 ) in DIPE or TBME saturated with H2O,
with octadecane (0.012 ) as internal standard.
Scheme 4. The synthesis of (R)-calycotomine.
Small-scale reactions in buffered aqueous media contained (Ϯ)-1b
(0.1 ) or (Ϯ)-2b (0.02, 0.1 or 0.2 ) as substrate. Compound (Ϯ)-
2b was dissolved in liquid (oily) form. The pH of the reaction mix-
tures was measured with a Metrohm digital pH meter with a glass
Conclusions
Directed DKR processes were developed for the efficient,
independent syntheses of both enantiomers of 6,7-dimeth-
oxy-1-TIC in enantiopure form. To the best of our knowl- membrane electrode. Various buffer systems were used: Tris buffer
Eur. J. Org. Chem. 2008, 5269–5276
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5273

T. A. Paál, A. Liljeblad, L. T. Kanerva, E. Forró, F. Fülöp
FULL PAPER
[tris(hydroxymethyl)aminomethane hydrochloride (0.1 ) + NaOH,
pH = 8.0], phosphate buffer [NaH2PO4 (0.1 ) + NaOH, pH =
7.0], malonate buffer [malonic acid (0.1 ) + NaOH, pH = 6.0],
citrate buffer [citric acid (0.1 ) + NaOH, pH = 5.5], malate buffer
[malic acid (0.1 ) + NaOH, tried under many pH conditions], and
ammonium acetate buffer [NH4OAc (0.1 or 0.3 ) + acetic acid or
NH4OH, tried under many pH conditions]. The pH of the reaction
mixtures was adjusted appropriately (with NH4OH in the case of
NH4OAc buffer; otherwise with NaOH) to the desired value before
the enzyme was added. In the case of reactions performed at con-
stant pH, the pH was adjusted manually (with aqueous NH4OH
or NaOH), from time to time.
hydrolysis) was quantified by HPLC on the basis of a calibration
with salicylic acid (used as external standard).
Optical rotations were measured with a Perkin–Elmer 341 polari-
meter. 1H and 13C NMR spectra were recorded with a Bruker
Avance DRX 400 spectrometer. Melting points were determined
with a Kofler apparatus.
(؎)-Ethyl 1,2,3,4-Tetrahydroisoquinoline-1-carboxylate [(؎)-1b]:
Compound (Ϯ)-1b was prepared by a method similar to that de-
scribed previously for this compound.[12a] Starting from isoquino-
line-1-carboxylic acid (5.65 g, 32.63 mmol), the procedure yielded
(Ϯ)-1b·HCl (4.81 g, 19.90 mmol, 61% overall yield). Compound
(Ϯ)-1b was released from (Ϯ)-1b·HCl with an aqueous solution of
KOH under cooling with ice. Compound (Ϯ)-1b·HCl (2.00 g,
8.27 mmol) resulted in (Ϯ)-1b (1.51 g, 7.36 mmol, 89% yield) after
extraction with Et2O, drying and evaporation under vacuum. Com-
pound (Ϯ)-1b was obtained as a pale-orange oil. 1H NMR
(400 MHz, CDCl3): δ = 1.19–1.37 (t, J = 7.13 Hz, 3 H, CH2–CH3),
1.70–2.35 (br. s, 1 H, NH), 2.61–2.80 (m, 2 H, CH2–CH2–NH),
2.97–3.08 (m, 1 H, CH2–CHH–NH), 3.22–3.35 (m, 1 H, CH2–
CHH–NH), 4.12–4.33 (m, 2 H, CH2–CH3), 4.64–4.80 (s, 1 H, Ar–
CH–NH), 7.02–7.43 (m, 4 H, Ar) ppm. 13C NMR (100.62 MHz,
CDCl3) δ = 14.8, 29.8, 41.5, 59.7, 62.0, 126.4, 127.7, 128.3, 130.0,
132.7, 136.1, 173.7 ppm. C12H15NO2 (205.25): calcd. C 70.22, H
7.37, N 6.82; found C 70.08, H 7.45, N 6.74.
Unreacted substrates 1b and 2b were analysed by GC on a CP
Chirasil-DEX CB-coated chiral capillary column (25 mϫ0.25 mm)
containing permethylated β-cyclodextrin as chiral selector (manu-
factured by Varian). Compound 1b {145 °C constant, 140 kPa; car-
rier gas: N2; retention times [min]: 16.26 for (R)-1b and 17.32 for
(S)-1b} and 2b {152 °C constant, 140 kPa; carrier gas: N2; reten-
tion times [min]: 67.96 for (R)-2b and 70.75 for (S)-2b}.
Samples from reactions in organic solvents were injected directly.
Samples from aqueous phases were extracted with DIPE [50 µL
sample was extracted with 2 mL of DIPE after the addition of
NH4OH (0.5 , 10 µL)], the organic phase was separated, the sol-
vents evaporated under vacuum, and the residue was dissolved in
toluene (100 µL) and thereafter injected. The ees values obtained
from reactions in the aqueous phase may not be accurate (the evap-
oration under heating might lead to partial racemisation) and are
therefore not reported (we found ees Յ30% in each case).
(؎)-Ethyl 6,7-Dimethoxy-1,2,3,4-tetrahydroisoquinoline-1-carboxyl-
ate [(؎)-2b]: The synthesis of (Ϯ)-2b was based on a literature
method.[15] Starting from 2-(3,4-dimethoxyphenyl)ethylamine
(24.21 g, 133.59 mmol), the procedure yielded (Ϯ)-2b·HCl (24.86 g,
82.38 mmol, 62% overall yield) as a slightly-yellowish crystalline
powder. Compound (Ϯ)-2b was liberated from (Ϯ)-2b·HCl with an
aqueous solution of KOH under ice cooling. Compound (Ϯ)-
2b·HCl (2.00 g, 6.63 mmol) resulted in (Ϯ)-2b (1.57 g, 5.92 mmol,
89% yield) after extraction with Et2O, drying and evaporation un-
der vacuum. Compound (Ϯ)-2b was obtained as a dense, slightly
yellow oil, which solidified with difficulty into a white crystalline
mass (m.p. 67–71 °C). Compound (Ϯ)-2b was applied in enzymatic
Conversions were determined with the aid of a standard, based
on a calibration, by GC. Octadecane (0.012 ) served as internal
standard for samples in organic solvents. For samples in aqueous
media, octadecane was used as external standard: the samples were
extracted as described previously; the organic phase was evaporated
under vacuum and the residue was dissolved in 100 µL solution of
octadecane (0.02 ) in toluene, and thereafter injected for GC.
Amino acid products (R)-1a, (R)-2a and (S)-2a were analysed by
HPLC by using a Chirobiotic TAG (250ϫ4.6 mm) column with
teicoplanin aglycon as chiral selector (manufactured by Supelco);
eluent: MeOH/H2O (1:1), flow: 0.8 mLmin–1, at 25 °C. 1a
{retention times [min]: 17.80 for (S)-1a and 25.12 for (R)-1a} and
2a {retention times [min]: 21.81 for (S)-2a and 27.84 for (R)-2a}
were detected at 220 and 232 nm, respectively.
1
reactions in liquid form. H NMR (400 MHz, CDCl3): δ = 1.19–
1.38 (t, J = 7.12 Hz, 3 H, CH2–CH3), 1.70–2.35 (br. s, 1 H, NH),
2.59–2.81 (m, 2 H, CH2–CH2–NH), 2.95–3.08 (m, 1 H, CH2–
CHH–NH), 3.20–3.35 (m, 1 H, CH2–CHH–NH), 3.73–3.85 (s, 3
H, CH3O–Ar), 3.85–3.93 (s, 3 H, CH3O–Ar), 4.14–4.30 (m, 2 H,
CH2–CH3), 4.56–4.67 (s, 1 H, Ar–CH–NH), 6.51–6.62 (s, 1 H, Ar),
6.83–6.92 (s, 1 H, Ar) ppm. 13C NMR (100.62 MHz, CDCl3) δ =
14.9, 29.4, 41.5, 56.4, 56.5, 59.2, 61.9, 111.2, 112.4, 124.3, 128.4,
147.8, 148.9, 173.8 ppm. C14H19NO4 (265.31): calcd. C 63.38, H
7.22, N 5.28; found C 63.24, H 7.31, N 5.22.
Compound (R)-3 was analysed after derivatisation (O-trimethylsilyl
derivative) by HPLC by using a Chiralcel OD-H chiral HPLC col-
umn (manufactured by Daicel); eluent: n-hexane (+0.1% Et2NH)/
EtOH (49:1); flow: 1.0 mLmin–1, at 25 °C {retention times [min]:
14.17 for derivatised (R)-3 and 16.77 for derivatised (S)-3}. Com-
pound (R)-3 (2 mg) was dissolved in CH2Cl2 (200 µL). N,O-Bis(tri-
methylsilyl)trifluoroacetamide (BSTFA, 20 µL) was added, the
mixture was kept at 25 °C for 1 h, MeOH (200 µL) was added, and
the mixture was incubated at 60 °C for 30 min. After evaporation,
the residue was dissolved in the eluent (1 mL) and injected.
Preparative-Scale DKRs of Substrates (؎)-1b and (؎)-2b
(R)-Selective Gram-Scale DKR of (؎)-1b: Compound (Ϯ)-1b
(0.50 g, 2.44 mmol) was dissolved in an aqueous solution of
NH4OAc (0.19 g, 2.45 mmol, 23.6 mL) with the aid of EtOH
(0.48 mL) as cosolvent at 3 °C. CAL-B (0.49 g, ca. 20 mgmL–1) was
added. The pH was set to 8.0 (with NH4OH 0.5 ) at the start of
the reaction. The mixture was shaken at 3 °C for 24 h, during which
conv. Ն99%, and (R)-1a partly crystallised as a white solid. The
enzyme was filtered off and washed with warm H2O. The solvent
and NH4OAc were evaporated off under vacuum, and the residue
was washed with Me2CO. Compound (R)-1a (0.37 g, 2.09 mmol,
86% yield) was isolated with 98%ee; it was recrystallised from
H2O/Me2CO. [α]2D5 = –65.5 (c = 1, 1  HCl), ref.[12a] [α]2D5 = –63.1
(c = 1, 1  HCl). M.p. 241–244 °C, ref.[12a] m.p. 239–240 °C.
The chemical hydrolysis of (Ϯ)-2b (0.02  in aqueous buffers) was
determined under the reported pH conditions (Figure 1), and the
pH was maintained constant with different buffer systems (0.1 ).
The pH was set at the start of the test [immediately after the ad-
dition of (Ϯ)-2b]. The test reactions (V = 5.0 mL) were run at
25 °C. Samples (1.0 mL) taken at the specified intervals were ex-
tracted with DIPE (3ϫ2 mL) and the aqueous phase was partially
evaporated under vacuum. Compound (Ϯ)-2a (formed by chemical
5274
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 5269–5276

Directed (R)- or (S)-Selective Dynamic Kinetic Enzymatic Hydrolysis
Ͼ99%ee. 1H NMR (400 MHz, CDCl3): δ = 2.98–3.19 (m, 2 H,
CH2–CH2–NH), 3.35–3.53 (m, 1 H, CH2–CHH–NH), 3.53–3.70
(m, 1 H, CH2–CHH–NH), 4.88–5.05 (s, 1 H, Ar–CH–NH), 7.21–
7.63 (m, 4 H, Ar) ppm. 13C NMR (100.62 MHz, CDCl3) δ = 25.0,
40.3, 59.0, 127.4, 128.3, 128.6, 128.8, 129.1, 132.2, 172.4 ppm.
C10H11NO2 (177.20): calcd. C 67.78, H 6.26, N 7.90; found C
67.71, H 6.38, N 7.72.
Acknowledgments
The Hungarian authors acknowledge the receipt of grants from
the Hungarian Scientific Research Fund (OTKA) (K 71938 and T
049407). T.A.P. is grateful for a fellowship awarded by the Center
for International Mobility (CIMO) in Finland.
(R)-Selective Gram-Scale DKR of (؎)-2b: Compound (Ϯ)-2b
(0.50 g, 1.88 mmol) was transformed by the procedure described
above in an aqueous solution of NH4OAc (0.072 g, 0.94 mmol,
8.7 mL) with EtOH (0.19 mL), at 25 °C, with the aid of CAL-B
(1.00 g, ca. 100 mgmL–1) at constant pH 8.5 (adjusted from time
to time with NH4OH 0.5 ). After 7 h (conv. Ն99%), the reaction
was worked up. Compound (R)-2a (0.38 g, 1.60 mmol, 85% yield)
was isolated with 92%ee; it was recrystallised from H2O/EtOH
[1] J. D. Scott, R. M. Williams, Chem. Rev. 2002, 102, 1669–1730.
[2] a) C. van Kesteren, M. M. M. de Vooght, L. Lopez-Lazaro,
R. A. A. Mathot, J. H. M. Schellens, J. M. Jimeno, J. H. Be-
ijnen, Anti-Cancer Drugs 2003, 14, 487–502; b) N. J. Carter,
S. J. Keam, Drugs 2007, 67, 2257–2276.
[3] D. R. Dalton in Studies in Organic Chemistry (Ed.: P. G. Gass-
man), Marcel Dekker Inc., New York, 1979, vol. 7, pp. 184–
190.
[4] a) M. Naoi, W. Maruyama, P. Dostert, K. Kohda, T. Kaiya,
Neurosci. Lett. 1996, 212, 183–186; b) B. E. Tóth, I. Bodnár,
K. G. Homicskó, F. Fülöp, M. I. K. Fekete, G. M. Nagy, Neur-
otoxicol. Teratol. 2002, 24, 655–666.
[5] A. Hofmeister, M. Schudok, H. Matter, K. Breitschopf, A. Ug-
olini, PCT Int. Appl. WO 2006002763, 2006.
[6] a) Y. Gao, X. Liu, W. Liu, Y. Qi, X. Liu, Y. Zhou, R. Wang,
Bioorg. Med. Chem. Lett. 2006, 16, 3688–3692; b) Y. Moro-
hashi, T. Kan, Y. Tominari, H. Fuwa, Y. Okamura, N. Watan-
abe, C. Sato, H. Natsugari, T. Fukuyama, T. Iwatsubo, T. Tom-
ita, J. Biol. Chem. 2006, 281, 14670–14676; c) P. Örtqvist, S. D.
Peterson, E. Åkerblom, T. Gossas, Y. A. Sabnis, R. Fransson,
G. Lindeberg, U. H. Danielson, A. Karlén, A. Sandström,
Bioorg. Med. Chem. 2007, 15, 1448–1474.
[7] a) S. E. Gibson, N. Guillo, J. O. Jones, I. M. Buck, S. B. Kal-
indjian, S. Roberts, M. J. Tozer, Eur. J. Med. Chem. 2002, 37,
379–389; b) S. E. Gibson, N. Guillo, M. J. Tozer, Tetrahedron
1999, 55, 585–615.
1
(1:1). [α]2D5 = –63 (c = 0.30, H2O). M.p. 266–268 °C. Ն99%ee. H
NMR (400 MHz, CDCl3): δ = 2.95–3.15 (m, 2 H, CH2–CH2–NH),
3.42–3.57 (m, 1 H, CH2–CHH–NH), 3.57–3.71 (m, 1 H, CH2–
CHH–NH), 3.77–3.92 (s, 3 H, CH3O–Ar), 3.92–4.03 (s, 3 H,
CH3O–Ar), 4.88–4.99 (s, 1 H, Ar–CH–NH), 6.88–7.01 (s, 1 H, Ar),
7.17–7.28 (s, 1 H, Ar) ppm. 13C NMR (100.62 MHz, CDCl3) δ =
24.5, 40.3, 56.1, 56.2, 58.6, 111.1, 112.0, 120.3, 124.9, 146.9, 148.0,
172.1 ppm. C12H15NO4 (237.25): calcd. C 60.75, H 6.37, N 5.90;
found C 60.58, H 6.51, N 5.81.
(S)-Selective Gram-Scale DKR of (؎)-2b: Compound (Ϯ)-2b
(0.50 g, 1.88 mmol) was transformed by the procedure described
above in an aqueous solution of NH4OAc (0.072 g, 0.94 mmol,
8.7 mL) with EtOH (0.19 mL) at 3 °C, with Alcalase (0.76 g, ca.
80 mgmL–1) at constant pH 8.5. After 3 d (conv. Ն99%), the reac-
tion was worked up. Compound (S)-2a (0.41 g, 1.73 mmol, 92%
yield) was isolated with 93%ee; it was recrystallised from H2O/
EtOH (1:1). [α]2D5 = +62 (c = 0.30, H2O). M.p. 265–267 °C.
Ն99%ee. The 1H NMR (400 MHz, H2O) data are similar to those
for (R)-2a. C12H15NO4 (237.25): calcd. C 60.75, H 6.37, N 5.90;
found C 60.62, H 6.46, N 5.83.
[8] K. W. Bentley, Nat. Prod. Rep. 2001, 18, 148–170.
[9] A. Mjalli, D. Jones, D. R. Gohimmukkula, PCT Int. Appl. WO
2006099379, 2006.
[10] a) S. Bajusz, L. Mohai, A. Fehér, J. Lavich, G. Széll, B.
Véghelyi, PCT Int. Appl. WO 199312091, 1993; b) O. Sieck, S.
Schaller, S. Grimme, J. Liebscher, Synlett 2003, 337–340.
[11] a) M. Ziółkowski, Z. Czarnoczki, A. Leniewski, J. K. Maurin,
Tetrahedron: Asymmetry 1999, 10, 3371–3380; b) N. Sasamoto,
C. Dubs, Y. Hamashima, M. Sodeoka, J. Am. Chem. Soc. 2006,
128, 14010–14011; c) T. Kanemitsu, Y. Yamashita, K. Nagata,
T. Itoh, Synlett 2006, 10, 1595–1597.
[12] a) T. A. Paál, E. Forró, A. Liljeblad, L. T. Kanerva, F. Fülöp,
Tetrahedron: Asymmetry 2007, 18, 1428–1433; b) I. S. Gill, E.
Kick, K. Richlin-Zack, W. Yang, Y. Wang, R. N. Patel, Tetra-
hedron: Asymmetry 2007, 18, 2147–2154.
[13] H. Pellissier, Tetrahedron 2008, 64, 1563–1601.
[14] a) J. A. Chaplin, M. D. Levin, B. Morgan, N. Farid, J. Li, Z.
Zhu, J. McQuaid, L. W. Nicholson, C. A. Rand, M. J. Burk,
Tetrahedron: Asymmetry 2004, 15, 2793–2796; b) D. Arosio,
A. Caligiuri, P. D’Arrigoa, G. Pedrocchi-Fantoni, C. Rossi, C.
Saraceno, S. Servi, D. Tessaro, Adv. Synth. Catal. 2007, 349,
1345–1348.
Assignment of the Absolute Configurations of (R)-2a and (S)-2a
(R)-(6,7-Dimethoxy-1,2,3,4-tetrahydroisoquinolin-1-yl)methanol
[(R)-Calycotomine, (R)-3]: To a suspension of LiAlH4 (0.19 g,
5.01 mmol) in dry THF (4 mL) at 25 °C was added (R)-2a (0.20 g,
0.84 mmol, 95%ee) in portions. The mixture was heated at reflux
for 6 h under vigorous stirring and thereafter cooled to 25 °C. The
excess amount of LiAlH4 was decomposed by adding H2O
(0.36 mL) in THF (5.0 mL) in portions, and stirring the mixture
for 30 min. The insoluble residue was filtered off and washed re-
peatedly with THF. The filtrate was dried with Na2SO4, and the
solvent was evaporated. The residue was recrystallised from tolu-
ene. Compound (R)-3 (0.075 g, 0.34 mmol, 40% yield) was ob-
tained as white crystalline powder after filtration. [α]2D5 = –33 (c =
0.2, H2O), ref.[23] [α]2D5 = +37.9 (c = 0.2, H2O) (for (S)-calycotom-
ine). M.p. 146–148 °C, ref.[11c] m.p. 138 °C, ref.[23] m.p. 140 °C.
91%ee. 1H NMR (400 MHz, CDCl3): δ = 1.33–2.06 (br. s, 2 H,
OH, NH), 2.58–2.77 (m, 2 H, CH2–CH2–NH), 2.97–3.15 (m, 2 H,
CH2–CH2–NH), 3.53–3.69 (m, 1 H, CH–CH2–OH), 3.69–3.79 (dd,
J = 10.7 Hz, J = 4.3 Hz, 1 H, CH–CHH–OH), 3.79–3.84 (s, 3 H,
CH3O–Ar), 3.84–3.90 (s, 3 H, CH3O–Ar), 3.90–4.04 (dd, J =
9.2 Hz, J = 4.3 Hz, 1 H, CH–CHH–OH), 6.50–6.58 (s, 1 H, Ar),
6.58–6.67 (s, 1 H, Ar) ppm. 13C NMR (100.62 MHz, CDCl3) δ =
29.8, 39.5, 56.5, 56.7, 64.8, 109.9, 112.7, 127.8, 128.4, 148.2,
148.5 ppm. C12H17NO3 (223.27): calcd. C 64.55, H 7.67, N 6.27;
found C 64.37, H 7.81, N 6.11.
[15] Z. Zalán, T. A. Martinek, L. Lázár, R. Sillanpää, F. Fülöp,
Tetrahedron 2006, 62, 2883–2891.
[16] a) J. M. Palomo, G. Fernández-Lorente, C. Mateo, M. Fuentes,
R. Fernández-Lafuente, J. M. Guisán, Tetrahedron: Asymmetry
2002, 13, 1337–1345; b) R. Torres, C. Ortiz, B. C. C. Pessela,
J. M. Palomo, C. Mateo, J. M. Guisán, R. Fernández-Lafuente,
Enzyme Microb. Technol. 2006, 39, 167–171.
[17] O. Kirk, M. W. Christensen, Org. Process Res. Dev. 2002, 6,
446–451.
[18] a) H. Zhao, J. Mol. Catal. B: Enzym. 2005, 37, 16–25; b) E. M.
Bowers, L. O. Ragland, L. D. Byers, Biochim. Biophys. Acta
2007, 1774, 1500–1507.
[19] J. C. Warren, S. G. Cheatum, Biochemistry 1966, 5, 1702.
Eur. J. Org. Chem. 2008, 5269–5276
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5275

T. A. Paál, A. Liljeblad, L. T. Kanerva, E. Forró, F. Fülöp
FULL PAPER
[20] a) B. A. Persson, A. L. E. Larsson, M. L. Ray, J. E. Bäckvall,
J. Am. Chem. Soc. 1999, 121, 1645–1650; b) M. J. Kim, Y. I.
Chung, Y. K. Choi, H. K. Lee, D. Kim, J. Park, J. Am. Chem.
Soc. 2003, 125, 11494–11495.
[21] H. J. Gais, F. Teil in Enzyme Catalysis in Organic Synthesis
(Eds.: K. Drauz, H. Waldmann), Wiley-VCH, Weinheim, 2002,
vol. 2, pp. 407–412.
[22] M. T. Ru, S. Y. Hirokane, A. S. Lo, J. S. Dordick, J. A. Reimer,
D. S. Clark, J. Am. Chem. Soc. 2000, 122, 1565–1571.
[23] T. Pedrosa, C. Andrés, J. M. Iglesias, J. Org. Chem. 2001, 66,
243–250.
Received: August 12, 2008
Published Online: September 30, 2008
5276
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 5269–5276