Enantioselective synthesis of both enantiomers of etodolac via lipase-catalyzed kenetic resolution

Article abstract of DOI:10.3987/COM-08-S(N)118

The efficient preparation of both enantiomers of etodolac 1 was achieved by the lipase-catalyzed kinetic resolution of the racemic primary alcohol (±)-2 followed by chemoselective oxidation.

Full text of DOI:10.3987/COM-08-S(N)118

HETEROCYCLES, Vol. 76, No. 2, 2008
1537
HETEROCYCLES, Vol. 76, No. 2, 2008, pp. 1537 - 1547. © The Japan Institute of Heterocyclic Chemistry
Received, 9th May, 2008, Accepted, 19th June, 2008, Published online, 23rd June, 2008. COM-08-S(N)118
ENANTIOSELECTIVE SYNTHESIS OF BOTH ENANTIOMERS OF
ETODOLAC VIA A LIPASE-CATALYZED KINETIC RESOLUTION
Shuji Akai,* Hiroyuki Nemoto, and Masahiro Egi
School of Pharmaceutical Sciences, University of Shizuoka
52-1, Yada, Suruga-ku, Shizuoka, Shizuoka 422-8526, Japan
E-mail: akai@u-shizuoka-ken.ac.jp
Abstract — The efficient preparation of both enantiomers of etodolac 1 was
achieved by the lipase-catalyzed kinetic resolution of the racemic primary alcohol
(±)-2 followed by chemoselective oxidation.
INTRODUCTION
Etodolac 1, a nonsteroidal antiinflammatory drug with analgesic and antipyretic activities, has
therapeutically been used as a racemate, while these pharmacological activities are known to be due to the
(S)-isomer.1 Recently, (R)-1 was discovered to possess antitumor activities, and its clinical application to
B-cell chronic lymphocytic leukemia2 and multiple myeloma3 is now under investigation. In order to
further evaluate the differences in their biological profiles between the enantiomers, an effective
production of the optically pure (S)- and (R)-1 is required.
Various developed methods have mainly provided (S)-1, which included the fractional crystallization of
the diastereomeric mixtures of the salts generated from (±)-1 and optically pure amines, such as
1-phenylethylamine,4 N-methyl glucamine,5 and (–)-cinchonidine,6 and that of the ester derived from
optically pure (–)-isopinocamphenol.6 The enrichment crystallization of (±)-1 was also reported.7 The
HPLC separation of the diastereomeric mixture of the esters prepared from (–)-borneol8 and the synthesis
of (S)-1 starting from either (R)-2,3-isopropylideneglyceraldehyde9 or (–)-β-pinene derivatives10 were
developed as alternative approaches. Some of them afforded the optically pure (S)-1 in fairly good
yields,4,5 while others were not very satisfactory because of their low yields and/or lengthy synthetic
sequences. In addition, there are still few methods for the preparation of the optically pure (R)-1.4,8
The lipase-catalyzed kinetic resolution of racemic carboxylic esters and alcohols has been well
acknowledged as a powerful means of producing optically pure compounds.11 However, it has not yet

1538
HETEROCYCLES, Vol. 76, No. 2, 2008
been successfully applied to the preparation of the optically pure 1. Thus, Achiwa et al. investigated the
hydrolase-catalyzed hydrolysis of a variety of esters derived from (±)-1; however, the best result obtained
from the resolution of its pivaloyloxymethyl ester using Alcaligenes sp. lipase gave (R)-1 (48% ee) in
58% yield, whose enantioselectivity, the E value,12 was 6.3.13 On the other hand, Brenna et al. reported
the kinetic resolution of a racemic precursor 2 using Candida cylindracea lipase followed by oxidation to
give (S)-1.4 However, the resolution was not very efficient, whose E value was 5.3, therefore, they needed
to repeat the resolution along with the recrystallization to improve the optical purity of the initial
resolution product, (S)-2 (33% ee).
O
R
*
Et
N
H
Et
Etodolac (1), R = CO2H
2, R = CH2OH
We now describe an improved preparation of both enantiomers of 1, each in the optical pure form, which
consists of the highly enantioselective resolution of (±)-2 (E value = 89) using lipase R (Penicillium
roqueforti, Amano) in cyclopentyl methyl ether (CPME) and the chemoselective oxidation of the primary
hydroxyl group of the obtained (R)- and (S)-2 without protecting the indole NH-group.
RESULTS AND DISCUSSION
Our first trial for the hydrolase-catalyzed kinetic resolution of (±)-2 (3 mg) was carried out using vinyl
acetate 3a in diisopropyl ether (IPE) at 35 °C. Among the commercially available 46 enzymes, lipase R
(Penicillium roqueforti, Amano) and lipase A12 (Aspergillus niger, Amano) were moderately effective
for producing (R)-2 and (S)-4a after 3 days [E = 6 by lipase R (Entry 1 in Table 1) and E = 4 by lipase
A12 (Entry 6)]. Some lipases derived from the Alcaligenes, Burkholderia, Candida, Mucor, and
Pseudomonas species catalyzed the reaction with poor enantioselectivities (E = <2), while others were not
active at all.
The acyl moiety of 3 was found to have a significant effect on both the enantioselectivity and the
reactivity.14 Thus, the use of vinyl butyrate 3b with lipase R gave a mixture of (R)-2 and (S)-4b (E = 28)
after 1 day (Entry 2), and that of vinyl hexanoate 3c gave the products (E = 26) after 3 days (Entry 3),
while that of vinyl decanoate 3d and vinyl benzoate 3e was not very effective (Entries 4 and 5). A similar
increase in the enantioselectivities by changing 3 was also observed when lipase A12 was used (Entries
7–10). Among these results, the combination of lipase R and 3b was thought to be the best based on the
selectivity and the reaction rate.

HETEROCYCLES, Vol. 76, No. 2, 2008
1539
Table 1. Lipase-catalyzed kinetic resolution of (±)-2 using various kinds of acyl donors 3.
O
O
O
O
O
*
*
H
O
R
R
OH
OH
O
3
N
H
N
H
+
Et
Et
N
Et
H
lipase,
Et
Et
Et
IPE, 35 °C
(±)-2
(R)-2
(S)-4
(R)-2,
% eea
Time, Conv.,
E
value
Entry
Lipase 3, R =
(S)-4
day
%
R =
% eea
49
1
2
R
Me (3a)
3
60
52
27
20
10
48
13
29
29
29
6
74
89
34
18
10
45
10
37
27
37
Me (4a)
R
nC3H7 (3b)
nC5H11 (3c)
nC9H19 (3d)
Ph (3e)
1
28
26
7
nC3H7 (4b)
nC5H11 (4c)
nC9H19 (4d)
Ph (4e)
81
3
R
3
90
4
R
1
70
5
R
4
19
4
89
6
A12
A12
A12
A12
A12
Me (3a)
3
Me (4a)
48
7
nC3H7 (3b)
nC5H11 (3c)
nC9H19 (3d)
Ph (3e)
1
5
nC3H7 (4b)
nC5H11 (4c)
nC9H19 (4d)
Ph (4e)
66
8
10
7
35
7
92
9
67
10
4
30
91
a) The optical purity was determined by HPLC analysis using Daicel CHIRALCEL OD-H.
Besides IPE, some ethereal solvents, such as tBuOMe (TBME) and cyclopentyl methyl ether (CPME),
were equally useful (Entries 4 and 6 in Table 2), whereas the use of a cyclic ether, THF, caused no
reaction (Entry 9). Toluene was another potent solvent that produced a good enantioselectivity (Entry 10);
however, side reactions took place on a larger scale. Lowering the reaction temperature brought about a
significant increase in the E value although the reaction rate decreased (Entries 3, 5, and 8).15 Thus, the
reaction using lipase R and 3b in CPME at 5 °C (Entry 8) was determined to be the best (E = 131) among
all the examined conditions.
The kinetic resolution of (±)-2 (50 mg or 1.0 g) under the same conditions as Entry 8 in Table 2 suffered a
low reproducibility of the reaction rate and the enantioselectivity, which was later found to be due to the
fluctuation of the water content in the reaction media.16 The trials under three different conditions by
changing the water content between 0 to 0.3 w/w % revealed that the use of the solvent containing 0.1
w/w % was the most effective (Entries 1–3 in Table 3). The reproducibility of this method was confirmed
by the resolution of 1.0 g of (±)-2, whose E value was 89, and (R)-2 (99% ee, 43% isolated yield) and
(S)-4b (89% ee, 52% isolated yield) were obtained (Entry 4). The absolute stereochemistries of the
products were determined by the comparison of the optical rotation of the recovered alcohol 2 (99%
ee){[α]D20 +42.5 (c 0.17, CHCl3)} to that of the reported value {[α]D20 +42.4 (c 1.0, CHCl3) for (R)-2}.4

1540
HETEROCYCLES, Vol. 76, No. 2, 2008
Table 2. The effects of the solvent and the reaction temperature on the lipase-catalyzed kinetic resolution
of (±)-2 with 3b.
O
O
O
O
*
*
H
O
nC3H7
nC3H7
OH
O
3b
lipase R
N
H
+
Et
N
Et
(±)-2
H
Et
Et
(R)-2
(S)-4b
Entry
Solventa Temp.,
°C
Time,
day
Conv.,
%
E value (R)-2, % eeb
(S)-4b, % eeb
1c
2
IPE
35
23
5
1
1
3
1
1
1
1
3
7
1
52
26
9
28
33
72
29
65
31
31
131
––
54
89
32
10
67
29
40
13
29
––
34
81
92
97
87
96
91
93
98
––
95
IPE
3
IPE
4
TBME
TBME
CPME
CPME
CPME
THF
35
5
44
23
31
12
23
<5
26
5
6
35
23
5
7
8
9
35
35
10
toluene
a) IPE: iPr2O, TBME: tBuOMe, CPME: Cyclopentyl methyl ether. b) The optical purity was determined by
HPLC analysis using Daicel CHIRALCEL OD-H. c) Cited from Entry 2 in Table 1.
Table 3. The effect of the water content on the lipase-catalyzed kinetic resolution of (±)-2.a
Entry
Amount Water content,
Conv.,
%
E value
(R)-2, % eeb
(S)-4b, % eeb
of (±)-2
50 mg
50 mg
50 mg
1.0 g
w/w %
1
2
3
4
0
49
35
<5
53
53
82
––
89
86
90
0.1
51
96
0.3
––
––
0.1
99 (43%)c
89 (52%)c
a) Each reaction was carried out using lipase R in CPME at 5 °C for 3 days. b) The optical purity was determined
by HPLC analysis using Daicel CHIRALCEL OD-H. c) Isolated yield is shown in parenthesis.
The other critical issue in this project was the oxidation of the primary hydroxyl group to the carboxyl
group in the presence of the indole moiety. It is well known that the indoles without protection of the
NH-group suffer easy oxidation under various conditions,4 and only a few successful examples of the
oxidation of alcohols or the aldehydes having a NH-free indole moiety were limited to the indoles having
an electron-withdrawing substituent17,18 and/or the oxidation at the reactive benzylic position.18,19

HETEROCYCLES, Vol. 76, No. 2, 2008
1541
Because the development of an effective method for the protection-free synthesis has been requested from
the standpoint of environmental consciousness,20 we thoroughly investigated the direct oxidation of (±)-2
as a preliminary substrate. Brenna et al. conducted the oxidation of (S)-2 using DMSO and Ac2O,4 and
their method was reproduced by our hand to give (±)-5. However, the methylthiomethyl ether (±)-6 (27%
yield) was also obtained as a side product just like they had mentioned (Entry 1 in Table 4). We found
that the Parikh oxidation using DMSO and SO3-pyridine dramatically depressed the formation of (±)-6
and afforded (±)-5 in 80% yield (Entry 2), while the Swern oxidation gave multiple products (Entry 3).
The Ley oxidation using Pr4NRuO4 (TPAP) and N-methylmorpholine N-oxide (NMO) was relatively
effective (Entry 4), and the Margarita oxidation using 2,2,6,6-tetramethyl-1-piperidinyloxy free radical
21
(TEMPO) and PhI(OAc)2 gave (±)-5 in 25% yield (Entry 5). However, other methods were not useful
thus resulting in either the formation of multiple products or no reaction (Entries 6–10).
Table 4. Oxidation of (±)-2 to (±)-5.
O
O
O
CHO
SMe
OH
O
N
H
N
H
+
Et
Et
N
Et
H
Et
Et
Et
(±)-2
(±)-5
(±)-6
Entry Reagents
Temp.
Reaction
time
Productsa
1
2
DMSO, Ac2O
RT
8 h
5 (56%), 6 (27%)
5 (80%), 6 (5%)
multiple products
5 (66%)
DMSO, SO3·pyridine, Et3N
DMSO, (COCl)2, Et3N
TPAP, NMO
RT
40 min
1 h
3
–78 °C
RT
4
1 h
21
5
TEMPO, PhI(OAc)2
0 °C
0 °C
0 °C
RT
1 d
5 (25%)
6
TEMPO, NaOCl, KBr22
Dess–Martin periodinane
PhI(OAc)2, KBr
1 h
multiple products
multiple products
multiple products
multiple products
no reaction
7
40 min
4 h
8
9
PhIO, KBr, MeOH
Ag2CO3, Celite
RT
5 min
4 d
10
RT
a) Isolated yield is shown in parenthesis.
The subsequent oxidation of the formyl group of (±)-5 into the carboxyl group in the presence of the
NH-unprotected indole moiety was more laborious. Although Brenna et al. reported the oxidation of (S)-5
using nickel peroxide,4 our trials always gave multiple products (Entry 1 in Table 5). The Pinnick
oxidation (NaClO2, 2-methyl-2-butene) was not available in this case (Entry 2), and 30% aqueous H2O2
did not cause any oxidation (Entry 3). On the contrary, the addition of 6 equiv. of K2CO3 to this reaction

1542
HETEROCYCLES, Vol. 76, No. 2, 2008
gave the desired (±)-1 in 7% yield (Entry 4). The improvement of the yield of (±)-1 by the combination of
H2O2 with various metal catalysts resulted in the formation of multiple products (Entries 5–8). After
many unfruitful trials (some of them are shown in entries 9–11), we finally found that the Ley oxidation
using TPAP and NMO with K2CO3 (1.0 mol equiv.) gave (±)-1 in 56% yield (Entry 12). The addition of
23
K2CO3 was inevitable for the chemoselective oxidation because a similar reaction in the absence of
K2CO3 gave multiple products (Entry 11), while the co-oxidant, NMO, was not always necessary (Entry
13).
Table 5. Oxidation of (±)-5 to (±)-1.
O
O
CHO
CO2H
N
H
N
H
Et
Et
Et
Et
(±)-5
(±)-1
Entry Reagents
Temp. Reaction time Productsa
1
2
3
4
5
Nickel peroxide, aq. NaOH
RT
RT
RT
RT
RT
6 h
1 h
1 d
1 d
1 d
multiple products
multiple products
no reaction
NaClO2, 2-methyl-2-butene
30% aq. H2O2
30% aq. H2O2, K2CO3
1 (7%)b
30% aq. H2O2, K2CO3, CeCl3·7H2O
multiple products
6
7
30% aq. H2O2, K2CO3, (NH4)6Mo7O24·H2O
RT
RT
0.75 h
4 h
multiple products
multiple products
30% aq. H2O2, K2CO3,
cat. (NH4)6Mo7O24·H2O, CeCl3·7H2O
8
9
30% aq. H2O2, SeO2
RT
RT
0 °C
RT
RT
RT
1.5 h
4 h
multiple products
multiple products
multiple products
multiple products
1 (56%)
Ag2O
I2, KOH24
10
11
12
13
0.5 h
1 d
TPAP, NMO, H2O
TPAP, NMO, H2O, K2CO3
TPAP, H2O, K2CO3
1 d
1 d
1 (58%)
a) Isolated yield is shown in parenthesis. b) An unidentified compound was obtained as a major product.
With these promising oxidation protocols in hand, we achieved the synthesis of both (R)- and (S)-1 in
optically pure forms. Thus, (R)-2 (99% ee) was subjected to the Parikh oxidation followed by the Ley
oxidation to give (R)-1 (>98% ee) in 32% overall yield. Its enantiomer, (S)-1 (>98% ee), was similarly
produced from (S)-2 (99% ee), obtained by the alkaline hydrolysis of (S)-4b (89% ee) followed by the
second lipase-catalyzed kinetic resolution under the same reaction conditions (Scheme 1).
·

HETEROCYCLES, Vol. 76, No. 2, 2008
1543
O
O
b
CO2H
OH
N
N
Et
Et
H
H
Et
Et
(R)-1
a
(R)-2,
99% ee, 43% yield
O
>98% ee, 32% yield
O
(±)-2
see
Entry 4 in
Table 3
b
CO2H
OR
N
N
Et
Et
H
H
Et
Et
(S)-4b (R = CO-nC3H7), 89% ee
(S)-2 (R = H), 89% ee
(S)-1
>98% ee, 27% yield
c
a,c
(S)-2 (R = H), 99% ee, 45% yield from (±)-2
Scheme 1. Reagents and conditions: (a) 3b, lipase R, CPME–H2O (1000:1), 5 °C; (b) 1) DMSO,
SO3·pyridine, Et3N, MeCN, RT, 2) TPAP, K2CO3, H2O, MeCN, RT; (c) K2CO3, MeOH, RT.
CONCLUSION
Both enantiomers of the optically pure etodolac 1 were synthesized from the racemic precursor 2. The
following two accomplishments are worth noting: (1) The lipase-catalyzed kinetic resolution was found to
be effective even for the primary alcohol 2 whose kinetic resolution has been thought to be difficult due to
its unique structure; viz., the quaternary carbon as the sole stereogenic center and the highly flexible
β-hydroxy ethyl group as the reactive site. (2) The proper choice of the reaction conditions achieved the
chemoselective oxidation of the primary hydroxyl group into the carboxyl group without any protection
of the indole NH-group. The application of this method to the preparation of the optically pure derivatives
of 1, such as SDX-308 and SDX-309,25 is now under investigation in our laboratory.
EXPERIMENTAL
Typical Procedure for Hydrolase-Catalyzed Kinetic Resolution of (±)-2 Using 3a–e.
To a resealable vessel were added a hydrolase (3 mg), (±)-24 (3 mg, 0.01 mmol), an organic solvent (0.1
mL), and 3 (0.1 mL, 0.07 mmol). The reaction mixture was stirred at the designated temperature by
monitoring the reaction with TLC. After the designated period of time, the reaction mixture was filtered
through a Celite pad, and the filtrate was concentrated in vacuo. The residue was purified by preparative
TLC (hexanes–EtOAc, 3:1) to give (R)-2 and (S)-4. The reaction conditions, the conversion of the
reaction, and the optical purity of the products are listed in Tables 1–3.
(R)-2-(1,8-Diethyl-1,3,4,9-tetrahydropyrano[3,4-b]indol-1-yl)ethanol ((R)-2).
Under an argon atmosphere, a suspension of lipase R (1.0 g), (±)-24 (1.0 g, 3.7 mmol), 3b (3.3 mL, 26
mmol) and distilled water (100 µL) in CPME (100 mL) was stirred at 5 °C for 3 days. The reaction
mixture was filtered through a Celite pad, and the filtrate was concentrated in vacuo. The residue was

1544
HETEROCYCLES, Vol. 76, No. 2, 2008
purified by column chromatography (hexanes–EtOAc, 5:1→3:1 then EtOAc only) to give (R)-2 (0.44 g,
43% yield, 99% ee) and (S)-4b (0.65 g, 52% yield, 89% ee). The optical purity of (R)-2 was determined
by HPLC analysis using Daicel CHIRALCEL OD-H (hexanes–iPrOH 90:10, flow rate 0.8 mL/min,
20 °C). Retention time: 7.0 min for (R)-2 and 14.3 min for (S)-2. The optical purity of (S)-4b was
determined by HPLC analysis using Daicel CHIRALCEL OD-H (hexanes–iPrOH 90:10, flow rate 0.8
mL/min, 20 °C). Retention time: 5.5 min for (R)-4b and 7.9 min for (S)-4b.
20
20
(R)-2: A colorless amorphous solid; [α]D +42.5 (c 0.17, CHCl3) {lit.,4 [α]D +42.4 (c 1.0, CHCl3) for
(R)-2}. 1H NMR (500 MHz, CDCl3) δ: 0.95 (3H, t, J = 7.5 Hz), 1.36 (3H, t, J = 7.5 Hz), 1.87–2.04 (2H,
m), 2.05–2.24 (2H, m), 2.74–2.92 (4H, m), 3.62–3.72 (2H, m), 3.97–4.09 (2H, m), 7.03 (1H, d, J = 7.5
Hz), 7.10 (1H, t, J = 7.0 Hz), 7.38 (1H, d, J = 7.5 Hz), 7.97 (1H, br s). 13C NMR (125 MHz, CDCl3) δ: 7.9,
13.7, 22.4, 24.0, 31.3, 39.1, 59.6, 60.4, 78.1, 109.0, 115.9, 119.8, 120.4, 126.3, 126.5, 134.6, 136.0. IR
(KBr): 3470 cm-1. High resolution FAB-MS Calcd for C17H23NO2 (M+) m/z: 273.1729, found: 273.1754.
(S)-2-(1,8-Diethyl-1,3,4,9-tetrahydropyrano[3,4-b]indol-1-yl)ethyl butyrate ((S)-4b).
A suspension of (S)-4b (0.65 g, 1.88 mmol, 89% ee) and K2CO3 (0.53 g, 3.8 mmol) in MeOH (10 mL)
was stirred at room temperature for 3 h. The reaction mixture was filtered through a SiO2 pad and the
filtrate was concentrated in vacuo to give (S)-2 (0.52 g, quant.). Under an argon atmosphere, a suspension
of lipase R (0.50 g), the above obtained (S)-2 (0.52 g, 1.88 mmol), 3b (1.6 mL, 13.0 mmol) and distilled
water (50 µL) in CPME (50 mL) was stirred at 5 °C for 3 days. The reaction mixture was filtered through
a Celite pad, and the filtrate was concentrated in vacuo. The residue was purified by column
chromatography (hexanes–EtOAc, 5:1) to give (S)-4b (0.56 g, 87% yield, 99% ee) as white crystals;
[α]D20 –1.07 (c 0.95, CHCl3), mp 81–83 °C. 1H NMR (500 MHz, CDCl3) δ: 0.86 (3H, t, J = 7.5 Hz), 0.90
(3H, t, J = 7.5 Hz), 1.37 (3H, t, J = 7.5 Hz), 1.50–1.57 (2H, m), 1.82–1.89 (1H, m), 1.91–1.97 (1H, m),
2.11 (2H, q, J = 7.5 Hz), 2.19–2.22 (2H, m), 2.72–2.83 (2H, m), 2.87 (2H, q, J = 7.0 Hz), 4.00 (2H, t, J =
6.0 Hz), 4.09–4.19 (2H, m), 7.02 (1H, d, J = 7.0 Hz), 7.07 (1H, t, J = 7.0 Hz), 7.35 (1H, d, J = 7.0 Hz),
7.72 (1H, br s). 13C NMR (125 MHz, CDCl3) δ: 7.9, 13.6, 13.8, 18.3, 22.3, 24.0, 31.9, 36.1, 36.7, 60.5,
60.8, 75.3, 109.0, 115.9, 120.0, 120.4, 126.3, 126.5, 134.6, 135.9, 173.9. IR (KBr): 3470, 1724 cm-1. Anal.
Calcd for C21H29NO3: C, 73.44; H, 8.51; N, 4.08. Found: C, 73.40; H, 8.34; N, 4.10.
(S)-2-(1,8-Diethyl-1,3,4,9-tetrahydropyrano[3,4-b]indol-1-yl)ethanol ((S)-2).
A suspension of (S)-4b (0.20 g, 0.58 mmol, 99% ee) and K2CO3 (163 mg, 1.18 mmol) in MeOH (3.0 mL)
was stirred at room temperature for 15 h. The reaction mixture was filtered through a SiO2 pad and
concentrated in vacuo to give (S)-2 (169 mg, quant., 99% ee) as a colorless amorphous solid. [α]D20 –42.7
(c 0.87, CHCl3). The NMR and IR data of this product were in full agreement with those of (R)-2.
(R)-2-(1,8-Diethyl-1,3,4,9-tetrahydropyrano[3,4-b]indol-1-yl)acetaldehyde ((R)-5).
A solution of (R)-2 (0.22 g, 0.81 mmol, 99% ee) and SO3·pyridine (0.78 g, 4.9 mmol) in a mixture of

HETEROCYCLES, Vol. 76, No. 2, 2008
1545
MeCN (1.1 mL), DMSO (1.1 mL), and Et3N (1.1 mL) was stirred at room temperature for 40 min. The
reaction mixture was poured into water, and the product was extracted with EtOAc. The combined
organic layer was successively washed with 3% aqueous HCl, saturated aqueous NaHCO3, and brine,
dried over MgSO4, and concentrated in vacuo. The residue was purified by column chromatography
(hexanes–EtOAc, 5:1) to give (R)-5 (176 mg, 80% yield) as a colorless oil. The optical purity of (R)-5
could not be determined by HPLC analysis using chiral columns due to the significant broadening of the
peaks. [α]D20 –28.5 (c 3.1, CHCl3). 1H NMR (500 MHz, CDCl3) δ: 0.86 (3H, t, J = 7.5 Hz), 1.33 (3H, t, J
= 7.5 Hz), 1.89–2.12 (2H, m), 2.78–2.85 (2H, m), 2.83 (2H, q, J = 7.5 Hz), 2.98 (1H, d, J = 18.0 Hz),
3.02 (1H, d, J = 18.0 Hz), 3.96–4.06 (2H, m), 7.00 (1H, d, J = 7.5 Hz), 7.06 (1H, t, J = 7.5 Hz), 7.35 (1H,
d, J = 7.5 Hz), 8.44 (1H, s), 9.71 (1H, s). 13C NMR (125 MHz, CDCl3) δ: 7.7, 13.7, 22.3, 24.0, 31.0, 51.8,
60.6, 74.8, 108.8, 116.0, 119.8, 120.6, 126.2, 126.5, 134.5, 135.1, 202.4. IR (KBr): 3466, 3431, 1720 cm-1.
High resolution FAB-MS Calcd for C17H21NO2 (M+) m/z: 271.1572, found: 271.1552.
(S)-2-(1,8-Diethyl-1,3,4,9-tetrahydropyrano[3,4-b]indol-1-yl)acetaldehyde ((S)-5).
Similarly to the preparation of (R)-5, (S)-5 (30 mg, 76% yield) was obtained from (S)-2 (40 mg, 0.143
mmol, 99% ee). A colorless oil. [α]D20 +31.6 (c 2.8, CHCl3). The NMR and IR data of this product were
in full agreement with those of (R)-5. High resolution FAB-MS Calcd for C17H22NO2 [(M+H)+] m/z:
272.1651, found: 272.1630.
(R)-Etodolac ((R)-1).
To a suspension of (R)-5 (50 mg, 0.184 mmol) of MeCN (2.4 mL), K2CO3 (25 mg, 0.184 mmol), distilled
water (two drops) and TPAP (19 mg, total 0.22 mmol) were added every 30 min for 4 times. The reaction
mixture was filtered through a SiO2 pad, the filtrate was concentrated in vacuo, and the residue was
purified by column chromatography (hexanes–EtOAc–AcOH, 50:50:1) to give (R)-1 (21 mg, 40% yield,
>98% ee) as white crystals. The optical purity of this product was determined by HPLC analysis using
Daicel CHIRALCEL OD (hexanes–iPrOH–AcOH 89:10:1, flow rate 0.8 mL/min, 20 °C). Retention time:
20
7.7 min for (R)-1 and 9.6 min for (S)-1. [α]D –67.8 (c 0.72, CHCl3), mp 77–78 °C (138–140 °C after
recrystallization from EtOAc){lit.,4 [α]D20 –66.8 (c 1, CHCl3), lit.,8 mp 139–141 °C}. 1H NMR (500 MHz,
CDCl3) δ: 0.88 (3H, t, J = 7.5 Hz), 1.34 (3H, t, J = 7.5 Hz), 2.01–2.15 (2H, m), 2.79–2.86 (4H, m), 3.03
(1H, d, J = 17.0 Hz), 3.05 (1H, d, J = 17.0 Hz), 4.04–4.13 (2H, m), 7.01 (1H, d, J = 7.5 Hz), 7.08 (1H, t, J
13
= 7.5 Hz), 7.36 (1H, d, J = 7.5 Hz), 8.47 (1H, s), C NMR (125 MHz, CDCl3) δ: 7.7, 13.7, 22.1, 24.0,
30.9, 42.6, 60.9, 75.2, 108.5, 116.0, 119.8, 120.7, 126.0, 126.6, 134.5, 134.6, 175.8. IR (KBr): 3684, 1747
cm-1. High resolution FAB-MS Calcd for C17H21NO3 (M+) m/z: 287.1521, found: 287.1531.
(S)-Etodolac ((S)-1).
Similarly to the preparation of (R)-1, (S)-1 (20 mg, 35% yield, >98% ee) was obtained from (S)-5 (54 mg,
0.20 mmol). [α]D20 +70.1 (c 0.50, CHCl3), mp 78–81 °C (138–141 °C after recrystallization from EtOAc)

1546
HETEROCYCLES, Vol. 76, No. 2, 2008
20
{lit.,4 [α]D +65.6 (c 1, CHCl3), lit.,8 mp 138–140 °C}. High resolution FAB-MS Calcd for C17H23NO2
(M+) m/z: 287.1521, found: 287.1524.
ACKNOWLEDGEMENTS
Amano Enzyme, Japan, and Zeon, Japan, are acknowledged for their generous gifts of the lipases and
CPME, respectively.
REFERENCES
1. L. G. Humber, C. A. Demerson, and P. Swaminathan, J. Med. Chem., 1986, 29, 871 and references
cited therein.
2. M. Jensen, A. Engert, F. Weissinger, W. Knauf, E. Kimby, C. Poynton, I. A. Oliff, M. J. Rummel,
and A. Öesterborg, Invest. New Drugs, 2008, 26, 139 and references cited therein.
3. S. Nakamura, M. Kobayashi, K. Shibata, N. Sahara, K. Shigeno, K. Shinjo, K. Naito, H. Hayashi,
and K. Ohnishi, Leuk. Res., 2006, 30, 123; P. Neri, H. Yasui, T. Hideshima, P. Tassone, N. Raje, C.
P. Laurence, K. Ishitsuka, S. Blotta, T. Kiziltepe, E. M. Ocio, M. Fulciniti, S. Kanekal, G. T. Elliott,
N. C. Munshi, and K. C. Anderson, Br. J. Haematol., 2006, 134, 37.
4. E. Brenna, C. Fuganti, D. Fuganti, P. Grasselli, L. Malpezzi, and G. Pedrocchi-Fantoni, Tetrahedron,
1997, 53, 17769.
5. M. Woods, U. C. Dyer, J. F. Andrews, C. N. Morfitt, R. Valentine, and J. Sanderson, Org. Process
Res. Dev., 2000, 4, 418.
6. S.-Y. Chou, C.-L. Tseng, and L.-S. Chang, J. Chin. Chem. Soc., 2001, 48, 229.
7. C. A. Demerson and L. G. Humber, U.S. Patent 4,515,961, 1985.
8. C. A. Demerson, L. G. Humber, N. A. Abraham, G. Schilling, R. R. Martel, and C. Pace-Asciak, J.
Med. Chem., 1983, 26, 1778.
9. S.-Y. Chou, C.-L. Tseng, and S.-F. Chen, Heterocycles, 1999, 51, 1527.
10. L. M. Cabral, P. R. R. Costa, M. L. A. A. Vasconcellos, E. J. Barreiro, and R. N. Castro, Synth.
Commun., 1996, 26, 3671.
11. For recent reviews: V. Gotor-Fernández, R. Brieva, and V. Gotor, J. Mol. Cat. B: Enzym., 2006, 40,
111; A. Ghanem, Tetrahedron, 2007, 63, 1721.
12. For the definition of the E value, see: C.-S. Chen, Y. Fujimoto, G. Girdaukas, and C. J. Sih, J. Am.
Chem. Soc., 1982, 104, 7294.
13. E. Mizuguchi, M. Itanami, and K. Achiwa, Heterocycles, 1997, 46, 149.
14. For some examples regarding the increase in the enantioselectivity by changing the acyl moiety of
the acyl donors, see: S. Akai, T. Naka, T. Fujita, Y. Takebe, T. Tsujino, and Y. Kita, J. Org. Chem.,
2002, 67, 411; M. Kawasaki, M. Goto, S. Kawabata, and T. Kometani, Tetrahedron: Asymmetry,

HETEROCYCLES, Vol. 76, No. 2, 2008
1547
2001, 12, 585.
15. For some examples regarding the increase in the enantioselectivity at a lower reaction temperature,
see: T. Sakai, H. Mitsutomi, T. Korenaga, and T. Ema, Tetrahedron, 2005, 16, 1535 and references
cited therein.
16. For the importance of the water content in the enzymatic reactions in organic media, see: K. Faber,
‘Biotransformations in Organic Chemistry,’ 5 th Ed., Springer, Berlin, 2004, Chapter 3.
17. K. S. Feldman and A. G. Karatjas, Org. Lett., 2006, 8, 4137.
18. T. Ohta and M. Somei, Heterocycles, 1989, 29, 1663.
19. F. Yamada, A. Kawanishi, A. Tomita, and M. Somei, Arkivoc, 2003, 102.
20. For some examples regarding the protection-free synthesis, see: R. W. Hoffmann, Synthesis, 2006,
3531; P. S. Baran, T. J. Maimone, and J. M. Richter, Nature, 2007, 446, 404.
21. A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, and G. Piancatelli, J. Org. Chem., 1997, 62,
6974.
22. P. L. Anelli, C. Biffi, F. Montanari, and S. Quici, J. Org. Chem., 1987, 52, 2559.
23. T. Hu and J. S. Panek, J. Org. Chem., 1999, 64, 3000.
24. S. Yamada, D. Morizono, and K. Yamamoto, Tetrahedron Lett., 1992, 33, 4329.
25. E. Lindhagen, S. Nissle, L. Leoni, G. Elliott, Q. Chao, R. Larsson, and A. Åleskog, Cancer
Chemother. Pharmacol., 2007, 60, 545; R. Feng, G. Anderson, G. Xiao, G. Elliott, L. Leoni, M. Y.
Mapara, G. D. Roodman, and S. Lentzsch, Blood, 2007, 109, 2130.