First Preparation of Enantiopure Indane Monomer, (S)-(-)- and (R)-(+)-2,3-dihydro-3-(4′-hydroxyphenyl)-1,1,3-trimethyl-1H-inden-5-ol, via a Unique Enantio- and Regioselective Enzymatic Kinetic Resolution

Article abstract of DOI:10.1021/jo990795w

Compound 1, (2,3-dihydro-3- (4′-hydroxyphenyl)-1,1,3-trimethyl-1H-inden-5-ol), a highly valuable monomer was prepared for the first time into its two enantiomerically pure forms via enzyme catalyzed kinetic resolution of corresponding 1-diesters. Hydrolysis of 1-dipentanoate catalyzed by lipase from Chromobacterium viscosum (CVL) is highly regioselective (38:1) between two phenyl groups and highly enantioselective (E = 48) toward a remote quaternary chiral center (five bonds), yielding (S)-(-)-1-4′-monopentanoate and unreacted (R)-(+)-1-dipentanoate in high yield and excellent ee. Unlike other hydrolase-catalyzed reactions, the CVL-catalyzed reaction does not proceed to sequential hydrolysis of (S)-(-)-1-4′-monopentanoate to (S)-C-)-1, showing that the reaction is also highly chemical selective between 1-diester and 1-monoester. The structural preference of the reaction was clearly determined by ¹H and COSY NMR. The absolute configuration of the nonreacted (R)-(+)-1-dipentanoate was determined consistently by circular dichroism and X-ray crystallography after being chemically transformed to (R)-(+)-1 and derivatives. Surprisingly, CVL favors the carbonyl group on the more substituted phenol, which has more steric hindrance, shorter bond length (1.39 A compared to 1.41 A on less substituted phenyl), and was believed to be the less favored group. In brief, in this reaction, the more substituted phenol group is preferred. (S)-enantiomer is preferred. 1-Diesters are substrates while corresponding 1-monoesters are not. The unique feature of CVL provides a simple access to enantiopure diol 1, corresponding 1-monoestesrs, 1-homodiesters, as well as 1-mixed diesters.

Full text of DOI:10.1021/jo990795w

7498
J . Org. Chem. 1999, 64, 7498-7503
F ir st P r ep a r a tion of En a n tiop u r e In d a n e Mon om er , (S)-(-)- a n d
(R)-(+)-2,3-d ih yd r o-3-
(4′-h yd r oxyp h en yl)-1,1,3-tr im eth yl-1H-in d en -5-ol, via a Un iqu e
En a n tio- a n d Regioselective En zym a tic Kin etic Resolu tion
Michael Zhang* and Romas Kazlauskas
McGill University, Department of Chemistry, 801 Sherbrooke Street West,
Montreal, Quebec H3A 2K6 Canada
Received May 14, 1999
Compound 1, (2,3-dihydro-3- (4′-hydroxyphenyl)-1,1,3-trimethyl-1H-inden-5-ol), a highly valuable
monomer was prepared for the first time into its two enantiomerically pure forms via enzyme
catalyzed kinetic resolution of corresponding 1-diesters. Hydrolysis of 1-dipentanoate catalyzed by
lipase from Chromobacterium viscosum (CVL) is highly regioselective (38:1) between two phenyl
groups and highly enantioselective (E ) 48) toward a remote quaternary chiral center (five bonds),
yielding (S)-(-)-1-4′-monopentanoate and unreacted (R)-(+)-1-dipentanoate in high yield and
excellent ee. Unlike other hydrolase-catalyzed reactions, the CVL-catalyzed reaction does not proceed
to sequential hydrolysis of (S)-(-)-1-4′-monopentanoate to (S)-(-)-1, showing that the reaction is
also highly chemical selective between 1-diester and 1-monoester. The structural preference of the
reaction was clearly determined by 1H and COSY NMR. The absolute configuration of the nonreacted
(R)-(+)-1-dipentanoate was determined consistently by circular dichroism and X-ray crystallography
after being chemically transformed to (R)-(+)-1 and derivatives. Surprisingly, CVL favors the
carbonyl group on the more substituted phenol, which has more steric hindrance, shorter bond
length (1.39 Å compared to 1.41 Å on less substituted phenyl), and was believed to be the less
favored group. In brief, in this reaction, the more substituted phenol group is preferred.
(S)-enantiomer is preferred. 1-Diesters are substrates while corresponding 1-monoesters are not.
The unique feature of CVL provides a simple access to enantiopure diol 1, corresponding
1-monoestesrs, 1-homodiesters, as well as 1-mixed diesters.
In tr od u ction
Chirality is another factor to consider in order to
enhance NLO. It is expected that the use of enantiomeri-
cally pure monomer (one handed) in the polymer back-
bone might avoid NLO fading effects, because of the
uniformity of the bond orientation compared to its
racemic counterpart. Compound 1, 2,3-dihydro-3- (4′-
hydroxyphenyl)-1,1,3-trimethyl-1H-inden-5-ol, is a mono-
mer bearing two asymmetrically located chromophores
with hyperpolarizability. It has been widely used in
polymeric materials. These materials are known for their
inherent toughness, clarity, thermal and mechanical
stability, and good optical properties and have been the
subject of hundreds of patents.3 The structure of com-
pound 1 contains a quaternary chiral center. It is
therefore a chiral molecule and readily available in
racemic form. It is of great interest for both industry and
academia to prepare this compound in its enantiopure
forms. However, the preparation of enantiomers of this
compound has been a challenge. Because of the lack of
an efficient method for the formation of a quaternary
chiral center, an asymmetric synthesis, if possible, needs
a multistep sequence and an efficient method to introduce
chirality and transfer it onto the quaternary carbon
Optical communication systems of the next generation
of information technologies require a high quality of
second-order nonlinear optical (NLO) materials for high-
speed information transfer vehicles, namely, photons.
These materials can be used for optoelectronic modula-
tion or frequency doubling, information storage, optical
signal processing and display, polarizing coating, optic
fiber, and optical switches and devices. Research has been
focused on four types of materials including inorganic
crystals, semiconductors, glass optical fibers, and poly-
mers.1 Among these, polymers are exceptionally promis-
ing materials. Unlike other three types of materials,
which are generally fragile, polymers are suitable for
microfabrication and can be made inexpensively. To be
practically useful, the polymers should possess large
second-order nonlinear optical response, that is, non-
vanishing first molecular hyperpolarizability and good
thermal and mechanical stability for microfabrications.
At the molecular level it is necessary that chromophores
be incorporated into a polymer with rigid noncentrosym-
metric macroscopic structures. Several approaches have
been used in the development of such polymers, namely,
embedding the chromophores in the backbone or align-
ment of NLO side chains as pendants.2
(2) (a) Kauranen, M.; Verbiest, T.; Boutton, C.; Teerenstra, M. N.;
Clays, K.; Schouten, A. J .; Nolte, R. J . M.; Persoons, A. Science 1995,
270, 966. (b) Sattigeri, J . A.; Shiau, C.-W.; Hsu, C. C.; Yeh, F.-F.; Liou
S.; J in, B.-Y.; Luh, T.-Y. J . Am. Chem. Soc. 1999, 121, 1007. (c) Yu,
D.; Gharavi, A.; Yu, L. J . Am. Chem. Soc. 1995, 117, 11680. (d) Yu, L.
P.; Chan, W. K.; Bao, Z. N. Macromolecules 1992, 25, 5609.
(3) (a) J P 11001587 (1999). (b) WO9901415 (1999). (c) WO 9901413
(1999). (d) J P 11012416 (1999). (e) EP 881271 (1998). (f) U.S. 5, 703,
197 (1997). (g) U.S. 5, 830, 988 (1998).
* To whom correspondence should be addressed. Altus Biologics Inc.
625 Putnam Avenue, Cambridge, MA 02139-4807. Email: zhangm@
altus.com.
(1) (a) Angel, C. A. Science 1995, 267, 1924. (b) Stillinger, F. H.
Science 1995, 267, 1935. (c) Frick, B.; Richter, D. Science 1995, 267,
1939. Hodge, I. M. Science 1995, 267, 1945.
10.1021/jo990795w CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/16/1999

First Preparation of Enantiopure Indane Monomer
J . Org. Chem., Vol. 64, No. 20, 1999 7499
Sch em e 1. Hyd r olysis of Ra cem ic 1-Dip en ta n oa te Ma y Yield Tw o Regioisom er s of Mon op en ta n oa tes
(ea ch of w h ich con sists of a p a ir of en a n tiom er s) a n d Tw o En a n tiom er s of Diol 1
Ta ble 1. Regioselectivity a n d En a n tioselectivity of th e Best Hyd r ola ses tow a r d 1-Dip en ta n oa te a n d 1-Did eca n oa tea
lipase
1-diesters
rate (U/mg)
(de/me/diol)b
ee (ac) of 1
ee (1-4′-me)
Ec
bacterial lipase
CVL (Sigma)
CVL (Sigma)
CVL (Asahi)
CVL (ATCC 6918)
fungal lipases
RDL (Amano D)
RDL (Amano D)
ROL (Amano F)
ROL (Amano F)
CAL-A
dipentanoate
didecanoate
dipentanoate
dipentanoate
5.1
6.8
4.8
nd
55/45/0
87/12/0
64/36/0
96/4/0
na
na
na
na
92 (S)
60 (S)
86 (S)
81 (S)
38 ( 18d
4.3
22 ( 4
11
dipentanoate
didecanoate
dipentanoate
didecanoate
dipentanoate
didecanoate
0.1
0.12
0.5
0.5
0.05
0.05
33/14/51
45/12/43
37/16/47
84/7/9
18/40/24
62/22/16
64 (S)
67 (S)
74 (S)
71 (S)
46 (R)
52 (R)
nd
nd
43
nd
nd
nd
9e
8e
13e
6e
3e
CAL-A
8e
a
de: diester; me: monoester; nd: not determined; na: not applicable; ac: absolute configuration; Amano D: lipase from Rhizopus
delemar supplied by Amano; Amano F: lipase from Rhizopus oryzae supplied by Amano; CAL-A: lipase from Candida antarctica, fraction
b
A supplied by Boehringer Mannheim Biochemicals as Chirazyme L-5. Molar ratios of 1-dipentanoate, 1-monopentanoates, and diol 1
measured by HPLC. Peak areas were corrected for differences in extinction coefficient. c Enantiomeric ratio as defined by Chen et al. (ref
d
6). Average and standard deviation of eight measurements. The data listed are for one of these measurements. e Approximate value of
the overall enantioselectivity calculated from the amount and enantiomeric purity of diol using equations for a single-step resolution.
without racemization. Enzymatic kinetic resolution ap-
pears to be the simplest strategy. The technology uses
enzymes to catalyze a reaction enantioselectively, leading
two enantiomers in a racemic mixture of starting mate-
rial into two different chemical forms, which are then
easily separated. In this paper, we report the first
preparation of enantiopure 1 and its derivatives via a
unique lipase-catalyzed enantioselective and regioselec-
tive hydrolysis of corresponding 1-diesters. This work has
led to successful preparation of a series of chiral polymers
with unparalleled optical properties.4
As shown in Scheme 1, there are two distinguishable
ways that hydrolysis can proceed. Hydrolysis of the acyl
group A or B may yield two different regioisomers of
1-monoester, 1-4′-monoester, and 1-5-monoester, respec-
tively. Each regioisomer consists of a pair of enantiomers.
Sequential hydrolysis of the second acyl group on these
monoesters yields 1. Depending on the selectivity of each
step and the reaction extension, the result might be a
mixture of all possible compositions of the four different
compounds and their enantiomers. To achieve a practi-
cally useful kinetic resolution of this complex system, at
least one step of high enantioslectivity (E) is needed.5 As
molecular modeling is still in its infancy in the selection
of biocatalysts, a screening was carried out in order to
find the right enzyme.
Resu lts
Selection of Hyd r ola ses. Twenty-one mammalian,
fungal, and bacterial hydrolases were screened for their
ability to catalyze the hydrolysis of (()-1-dipentanoate.
Fourteen hydrolases showed catalytic activity. Of these,
four gave moderate to high enantiomeric purity for the
products. These four enzymes were tested again using
(()-1-didecanoate as substrate. The results are shown in
Table 1.6
Unlike other hydrolases, which gave a mixture of diol
1, 1-monoesters and unreacted 1-diester at the end of
reaction, lipase from Chromobacterium viscosum (CVL)
(4) (a) U.S. 5, 856, 422 (1999). (b) U.S. 5, 777, 063 (1998). (c) U.S.
5, 883, 218 (1999).
(5) A rational analysis of possibilities of resolution is available as
Supporting Information.

7500 J . Org. Chem., Vol. 64, No. 20, 1999
Zhang and Kazlauskas
(Sigma) gave a single 1-monopentanoate as product and
unreacted 1-dipentanoate. The product was determined
to be (S)-(-)-1-4′-monopentanoate (see below for details
on structure determination). The regioselectivity was
higher than 30:1 and the enantioselectivity E ) 38 ( 12.7
A remarkable drop of the E value was observed using
(()-1-didecanoate as substrate (E ) 4.3), indicating that
the chain length of acyl is an influential factor on E.
CVL from two other sources, Asahi and the culture
supernatant from a lipase-producing strain of CVL
(ATCC 6918), were also investigated, both of which were
less enantioselective toward (()-1-dipentanoate. The
former demonstrated an E value of 22 and the latter a
value of 11.
Three fungal lipases catalyzed the hydrolysis of (()-
1-dipentanoate or (()-1-didecanoate to a mixture of 1,
corresponding 1-monoesters, and unreacted 1-diester.
Both Rhizopus delemar lipases (RDL, Amano D) and
Rhizopus oryzae lipase (ROL, Amano F) yielded (S)-
(-)-1 with similar enantiomeric purity. Candida antarc-
tica lipase (CAL-A) yielded the (R)-(+)-1 with lower
enantiomeric purity. Hydrolysis of (()-1-diesters to 1
requires removal of two ester groups and each step may
be enantioselective. We did not determine the enantio-
selectivity of each step; instead, we estimated an overall
enantioselectivity for both steps from the amount and ee
of 1 using equations for a single-step hydrolysis. This
estimate shows that the overall E for hydrolysis of (()-
1-dipentanoate or (()-1-didecanoate to 1 by RDL, ROL,
and CAL-A is substantially lower than the single step
CVL (Sigma)-catalyzed hydrolysis of (()-1-dipentanoate
to (S)-(-)-1-4′-monopentanoate. Consistent with this
estimate, the ee of 1 from these reactions was signifi-
cantly lower than the ee of (S)-(-)-1-4′-monopentanoate
from the CVL-catalyzed hydrolysis.
F igu r e 1. Variation in the enantioselectivity of CVL-catalyzed
hydrolysis of 1-diesters with different acyl chain length. CVL
from Sigma (diagonal lines) showed the highest enantioselec-
tivity, E ) 48 ( 20, for the dibutanoate. CVL from Asahi
(white) showed lower enantioselectivity. The error bars rep-
resent standard deviations for three to eight separate mea-
surements, some with different lots of lipase.
Ta ble 2. En a n tioselectivity of Ch r om oba cter iu m
vicosu m Lip a se tow a r d 1-Diester s w ith Differ en t Acyl
Ch a in Len gth
rate
ratioa
ee of
ester
(U/mg)
(de/me)
me (%)
E
diacetate
15
11.5
8
(4.6)
11
(6.3)
0.4
10
34.0/63.4b
62.6/37.4
63.9/36.1
(59.5/40.5)
61.2/38.8
(64.5/33.5)
48.8/51.2
96.6/3.4
42.3
78.5
93.2
(76.4)
(86.5)
92.4
78.4
86.4
63.8
51.1
60.1
5
dipropionate
dibutanoate
(Asahi)d
14 ( 3
48 ( 20c
13 (22 ( 13)c
dipentanoate
21 (19 ( 4)
(Asahi)d
45 ( 18c
dihexanoate
diheptanoate
dioctanoate
dinonanoate
didecanoate
21
12
5
5
5
2.8
6.4
6.8
91.2/8.8
53.2/46.8
79.6/20.4
An attempted acylation of 1 with vinyl butanoate using
CVL in tert-butyl methyl ether gave no detectable acy-
lated product after 3 days incubation at room tempera-
ture.
a
Measured by HPLC monitoring at 272 nm. Baseline resolution
of two enantiomers of diol, four isomers of monoester, and
overlapped two enantiomers of diesters were obtained on the same
chromatogram using Chiralcel OD eluted with ethanol-hexane
(8/92), see Supporting Information (1). Peak areas for the mo-
noester were divided by 1.07 to account for the difference in
Op tim iza tion of En a n tioselectivity w ith Differ en t
Acyl Ch a in Len gth . The variation of the E value versus
acyl chain lengths was examined. In the case of Amano
D, the difference of chain length was less influential. A
similar composition of products and almost the same ee
of 1 were obtained using (()-1-dipentanoate and (()-1-
didecanoate (E ) 9 and 8, respectively). No remarkable
improvement was observed. In the case of Amano F, a
similar ee was obtained at a lower conversion using (()-
1-didecanoate. This indicates a decrease of enantioselec-
tivity, which was reflected in the decrease on overall E
b
extinction coefficient for diester and monester. 2.6% diol noted
in the hydrolysis of the diacetate. c Average and standard deviation
d
for six different experiments. Use of CVL supplied by Asahi.
(13 to 6. In the case of CAL-A, the enantioselectivity was
slightly improved (E ) 3 to 8). These data were shown
in Table 1.
However, the E value of CVL-catalyzed hydrolysis of
(()-1-diesters varies dramatically with the chain length
of esters. This phenomenon was examined using the same
source of CVL (Sigma). The highest enantioselectivity
was obtained with (()-1-dibutanoate. Both shorter and
longer ester chains render the reaction less enantio-
selective. Some undesired sequential hydrolysis to the
diol 1 was observed using (()-1-diacetate. CVL from
Asahi was less enantioselective than CVL from Sigma
for both (()-1-dibutanoate and (()-1-dipentanoate. The
rate of reaction remained constant within a factor of 2
for all chain lengths except (()-1-dihexanoate (slower by
a factor of 20) and (()-1-dioctanoate (slower by a factor
of 4). See Figure 1 and Table 2.
(6) Hydrolases with low selectivity toward (()-1-dipentanoate: pen
acylase (12% ee (R) diol, no monoesters, very slow), porcine
G
pancreatic cholesterol esterase (∼5% ee diol, 2:2:6:1 ratio of mono-
esters), porcine liver esterase (22% ee (S) diol, 1:1:1:1 ratio of
monoesters), lipase from Candida antarctica B (8% ee (R) diol, trace
monesters, very slow), lipase from Candida rugosa (21% ee (R) diol,
0:1:1:0 ratio of monoesters), lipase from Humicola lanuginosa (29%
ee (S) diol, 1:12:16:1 ratio of monoesters), lipase from Pseudomonas
cepacia (0% ee diol, 1:3:3:9 ratio of monoesters), lipase from Pseudo-
monas sp. (Boehringer Mannheim) (35% ee (R) diol, 1:3:2:10 ratio of
monoesters). Lipases from Rhizopus niveus and Rhizomucor javanicus
(Amano lipase
N and M) were very slow catalysts and yielded
approximately equimolar mixtures of all four monoesters. The following
lipases showed no detectable hydrolysis: Amano lipases K-10 (Pseudo-
monas sp.), AK (Pseudomonas fluorescens), G (Penicillium camem-
bertii), R (Penicillium roqueforti), L (Candida lipolytica), Sigma wheat
germ lipase.
(7) Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J . J . Am. Chem.
Soc. 1982, 104, 7294-7299.
Str u ctu r e of 1-Mon op en ta n oa te a n d Regioselec-
tivity. The regioselectivity of the CVL-catalyzed hydroly-
1
sis was clearly determined using a simple method. H
NMR chemical shifts unambiguously show that the acyl

First Preparation of Enantiopure Indane Monomer
J . Org. Chem., Vol. 64, No. 20, 1999 7501
F igu r e 2. 1H NMR chemical shifts of the aromatic protons in 1, 1-dipentanoate, and the 1-monopentanoate isolated from a
CVL-catalyzed hydrolysis of 1-dipentanoate. All of the aromatic protons in 1 resonate 0.2-0.3 ppm upfield of those in 1-dipentanoate.
In the monopentanoate, the resonances of the more substituted ring lie 0.15-0.27 ppm upfield from those in the dipentanoate,
while those in the less substituted ring lie within 0.01 ppm of the dipentanoate. Thus, the ester group lies on the less substituted
ring. Resonances were assigned from 1H-1H COSY spectra.
Sch em e 2. P r ep a r a tive Sca le Resolu tion of
Com p ou n d 1 via CVL-Ca ta lyzed Hyd r olysis of
Cor r esp on d in g (()-1-Dibu ta n oa te
F igu r e 3. Assignment of absolute configuration using the
exiton coupling method. Circular dichroism spectra of (R)-(+)-1
(dotted line) and (R)-1-di-4-dimethylaminobenzoate (solid line).
The bisgnate signal for (R)-1-di-4-dimethylaminobenzoate
shows a positive peak at lower energy, which indicates positive
helicity for the orientation of chromophores. Positive helicity
corresponds to an (R)- absolute configuration.
group of the monoester isolated from CVL-catalyzed
hydrolysis of (()-1-pentanoate is on the less substituted
phenol. This means, surprisingly, that CVL preferably
hydrolyzes the acyl group on the more substituted phenol,
which was believed to be the more hindered and less
favored one. See Figure 2.
sponds to the (R)-configuration.8 The helicity model of
(+)-(R)-1-di-(4-dimethylamino)benzoate is shown also in
Figure 3.
Absolu te Con figu r a tion . The slow-reacting enanti-
omer (+)-1-dipentanoate was chemically transformed to
(+)-1. The absolute configuration of the latter was
unambiguously established to be (R)-(+)-1 using two
independent methods, circular dichroism (CD) and X-ray
crystallography. The exiton chirality method assigns the
absolute configuration of two chromophores in space
when the electronic absorption couples, thereby causing
a bisignate CD spectrum. One expects two split CD
signals of opposite sign around UV maximum, one is
slightly higher and the other slightly lower in energy.
Compound (+)-1 contains two chromophores - phenols
- but does not show any split CD signals. Only the lower
energy signal appears in the CD spectrum (dotted line)
in Figure 3. The higher energy signal is presumably
obscured by other electronic transitions. To overcome this
obstacle, we added two new chromophores (4-dimethyl-
aminobenzoyl), which has an UV absorption maximum
at 308 nm, in the molecule by making (+)-1-di-(4-
dimethylamino)benzoate. This moves the electronic tran-
sitions to lower energy. The (+)-1-di-(4-dimethyl-
amino)benzoate shows, indeed, two weak but clear bisig-
nate CD signals: positive at 329 nm and negative at 302
nm as expected for splitting due to exiton coupling. Since
the lower energy signal is positive, the benzoate chro-
mophores orient with a positive helicity, which corre-
This (R)-configuration was confirmed independently by
X-ray crystallography method (see the Supporting Infor-
mation for X-ray crystallography data and an ORTEP
drawing of (R)-(+)-1).
32 g Sca le Resolu tion of (()-1-Dibu ta n oa te. The
selection of the biocatalysts was stopped on the CVL
(Sigma) as it gave high enantioselectivity, high regio-
selectivity, and unique one-step resolution. The reaction
was well characterized and optimized. A large-scale
resolution was carried out using 32 g of (()-1-di-
butanoate. At 43% conversion, unreacted (R)-(+)-1-di-
butanoate was isolated as a colorless oil (54%, 17.3 g, 52%
ee) and (S)-(-)-1-4′-monobutanoate as a light yellow oil
(35%, 9.3 g, 92% ee), corresponding E ) 50. The (R)-(+)-
1-dibutanoate was cleaved with sodium methoxide and
the resulting (R)-(+)-1 was recrystallized from methanol-
dichloromethane, yielding pure enantiomer as white
crystals (2.5 g, >99% ee), Scheme 2.
Con clu sion
Compound 1, a highly valuable monomer for NLO
polymeric materials, was successfully prepared for the
(8) Harada, N., Nakanishi, K. Circular Dichroic Spectroscopy;
University Science Books: Mill Valley, CA, 1983.

7502 J . Org. Chem., Vol. 64, No. 20, 1999
Zhang and Kazlauskas
1
(()-1-Dihexanoate. H NMR δ 0.8-0.9 (m, 6H), 1.0 (d, 3H),
1.2-1.5 (m, 13H), 1.5-1.8 (m, 6H), 2.2 (m, 1H), 2.4 (m, 1H),
2.5 (m, 3H), 6.7 (d, 1H, J ) 2 Hz), 6.9-7.0 (m, 3H), 7.1-7.2
(m, 3H); MS (FAB, M+ H+) 465.
first time in its enantiomerically pure forms via an
enzyme-mediated kinetic resolution method. The work
has made it possible to prepare a series of polymers with
improved NOL properties.4 The method described in this
work also provides simple access to monoesters, homo-
and mixed diesters, and all possible derivatives of 1 in
pure enantiomeric forms. The CVL-catalyzed hydrolysis
of 1-dibutanoate (1-pentanoate) is highly enantio- and
regioselective toward the remote quaternary chiral center
and it shows a certain unusual quality of the enzyme
beyond our understanding. It favors the carbonyl group
on the more substituted phenol, which is more hindered,
the bond is shorter and is believed to be the less favored
group. Nevertheless, the reaction does not proceed to
sequential hydrolysis of monoesters to diol, excluding the
monoesters as substrates. The origin of these enzyme
features remains unknown. Revealing the fundamentals
beneath these phenomena will certainly enrich current
scientific knowledge.
(()-1-Diheptanoate. 1H NMR δ 0.8-0.9 (m, 12H), 1.0 (s, 3H),
1.2-1.4 (m, 26H), 1.6 (s, 3H), 1.7 (m, 4H), 2.2 (s, 2H), 2.5 (m,
4H), 6.8 (d, 1H, J ) 2 Hz), 6.9-7.0 (m, 3H). 7.1-7.2 (m, 3H);
MS (FAB, M+ H+) 493.
1
(()-1-Dioctanoate. H NMR δ 0.9 (m, 6H), 1.1 (s, 3H), 1.2-
1.5 (m, 20H), 1.7 (s, 3H), 1.7-1.9 (m, 3H), 2.2 (d, 2 H, J ) 13
Hz), 2.4 (d, 1H, J ) 13 Hz), 2.5-2.6 (m, 4H), 6.8 (d, 1H, J )
2 Hz), 6.9-7.0 (m, 3H), 7.1-7.2 (m, 3H); MS (FAB, M+ H+)
521.
(()-1-Dinonanoate. 1H NMR δ 0.9 (m, 7H), 1.1 (s, 3H), 1.2-
1.5 (m, 22H), 1.7-1.9 (m, 8H), 2.2 (d, 2 H, J ) 13 Hz), 2.4 (d,
1H, J ) 13 Hz), 2.5-2.6 (m, 4H), 6.8 (d, 1H, J ) 2 Hz), 6.9-
7.0 (m, 3H), 7.1-7.2 (m, 3H); MS (FAB, M+H+) 549.
1
(()-1-Didecanoate. H NMR δ 0.9 (m, 6H), 1.1 (s, 3H), 1.2-
1.5 (m, 30H), 1.6-1.8 (m, 8H), 2.2 (d, 2 H, J ) 13 Hz), 2.4 (d,
1H, J ) 13 Hz), 2.5-2.6 (m, 4H), 6.8 (d, 1H, J ) 2 Hz), 6.9-
7.0 (m, 2H), 7.0 (d, 1H, J ) 2 Hz), 7.1-7.2 (m, 3H); MS (FAB,
M+H+) 577.
Mixtu r e of (()-1-4′-Mon op en ta n oa te a n d (()-1-5-
Mon op en ta n oa te. The same procedure as for 1-diesters,
except using one equivalent of pentanoyl chloride and of
triethylamine yielded a mixture of diol 1, 1-monopentanoates
and 1-dipentanoate: Rf ) 0.1, 0.28, 0.73 (CH2Cl2). The mixture
of 1-monopentanoates was isolated by flash chromatography
on silica gel eluted with dichloromethane: oil, 1.30 g, 50%.
The samples were used as the standard mixture for develop-
ment of HPLC analytical method.
Exp er im en ta l Section
Gen er a l Meth od s. 1H and 13C NMR spectra were recorded
at 270 MHz in CDCl3 and COSY NMR were recorded at 500
MHz in CDCl3. Mass spectra were obtained in direct-inlet
mode (FAB 6 kV Xe). Circular dichroism spectra were mea-
sured in CHCl3, 0.1 cm path length, 5 accumulations at 100
nm/min scan speed, room temperature, 0.2 nm resolution.
Bond lengths were obtained from simple modeling using CS
Chem3D Pro (CambridgeSoft, Cambridge, MA). All chemicals
were purchased from Aldrich-Sigma Canada unless otherwise
indicated. THF was dried by distillation from sodium ben-
zophenone ketyl under N2. Enzymes were purchased from
Sigma, Boehringer Mannheim, Amano, and Amresco. Lipase
from Chromobacterium viscosum was also obtained as gener-
ous gift from Asahi (J apan) and prepared in our laboratory
from a bacterial strain (ATCC 6918) purchased from American
Type Culture Collection. Racemic 1 was a gift from Molecular
OptoElectronics Corporation (Watervliet, NY).
Syn th esis of 1-Diester s. (Example of 1-dibutanoate; the
same procedure was used for all other diesters). Butyric acid
chloride (8.9 mmol, 2.4 eq) in dry THF (25 mL) was added over
10 min to a solution of 1 (1.0 g, 3.7 mmol) and triethylamine
(0.90 g, 8.9 mmol, 2.4 eq) in dry THF (25 mL). TLC showed
complete consumption of 1 after stirring overnight at room
temperature. HCl (50 mL, 1 M) was added and the mixture
was extracted with ethyl acetate (3 × 20 mL). The combined
extracts were washed with NaHCO3 (10%, 3 × 20 mL) and
water (2 × 20 mL) and dried over magnesium sulfate (10 g).
Chromatography on silica gel eluted with dichloromethane
yielded colorless oils in 90-95% yield. The corresponding
products are characterized as follows.
Scr een in g of Hyd r ola ses. Hydrolase (10 mg for most, 0.7
mg for CVL) was dissolved in 1 mL of phosphate buffer (10
mM, pH 7.5) and added to a mixture of the same buffer (3
mL) and 1-dipentanoate solution (50 mg in 4 mL of tert-
butylmethyl ether). A pH stat regulated the addition of NaOH
(0.1 N) to maintain the pH at 7.5. After 1 h (2 h for reactions
where no base was consumed in the first hour), the mixture
was poured into 20 mL of diethyl ether. The phases were
separated and the aqueous phase was extracted by ether (3 ×
10 mL). The combined organic phases were washed with water
(2 × 20 mL), dried over magnesium sulfate, filtered, and
evaporated. The slurry was dissolved in 2 mL of ethanol for
HPLC analysis. Enantioselectivity of CVL was calculated
according to Sih’s7 equation from the degree of conversion
measured and the enantiomeric purity of the product 1-4′-
monoester. The peak areas were corrected for the relative
extinction coefficients at 272 nm: 1-dipentanoate, 1.00; 1-4′-
monopentanoate, 1.06; 1, 1.60; 1-dibutanoate, 1.00; 1-4′-
monobutanoate, 1.07. 1, 1.70. For other enzymes the hydrolysis
proceeded to the formation of diol. We estimated the overall
enantioselectivity using the same equation from the fraction
of diol and its enantiomeric purity. Note that values for overall
enantioselectivity are only estimates.
(S)-(-)-1-4′-Mon op en t a n oa t e a n d (R)-(+)-1-Dip en t a -
n oa te. Hydrolysis of (()-1-dipentanoate (500 mg) by CVL (3.4
mg, 10000 units, Sigma) using the above-described screening
procedure yielded crude product (435 mg). Chromotography
of a portion (145 mg) on silica gel (10 g) yielded (R)-(+)-1-
dipentanoate upon elution with CH2Cl2-hexane (80/20), fol-
lowed by (S)-(-)-1-4′-monopentanoate upon elution with CH2-
Cl2-Et2O (90/10). (R)-(+)-1-dipentanoate: oil, 70 mg, 48%; 83%
ee, [R]D25 +64.6 (c ) 2.47, MeOH). NMR spectra were identical
with the spectra of the racemate. (S)-(-)-1-4′-monopen-
tanoate: oil, 45 mg, 40%; 92% ee; [R]D25 73.2 (c ) 2.47, MeOH);
1H NMR δ 1.0 (t, 3H, J ) 7.4 Hz), 1.0 (s, 3H), 1.3 (s, 3H), 1.4-
1.5 (m, 2H), 1.6 (s, 3H), 1.7-1.8 (m, 2H), 2.2 (d, 1H, J ) 13
Hz), 2.4 (d, 1H, J ) 13 Hz), 2.5 (t, 2H, J ) 7 Hz), 6.5 (d, 1H,
J ) 2 Hz), 6.7 (dd, 1H, J 1 ) 2, J 2 ) 8 Hz), 6.9 (d, 2H, J ) 9
Hz), 7.0 (d, 1H, J ) 8 Hz), 7.2 (d, 2H, J ) 9 Hz); 13C NMR δ
13.7, 22.3, 27.1, 30.7, 31.1, 34.2, 42.4, 50.4, 59.8, 111.4, 114.7,
120.9, 123.5, 127.7, 144.5, 148.3, 148.6, 150.5, 154.9, 172.6.
(S)-(-)-1-4′-Mon obu ta n oa te a n d (R)-(+)-1. A 4 L Er-
lenmeyer flask containing (()-1 (21.8 g, 81 mmol) and triethyl-
amine (19.7 g, 195 mmol, 2.4 equiv) dissolved in dry THF (500
(()-1-Diacetate. Rf ) 0.74 (CH2Cl2), 0.40 (9/1 hexane/2-
propanol); 1H NMR δ 1.1 (s, 3H), 1.3 (s, 3H), 1.7 (s, 3H), 2.2-
2.4 (m, 8H), 6.8 (s, 1H), 6.9-7.1 (m, 3H), 7.2 (m, 3H); 13C NMR,
δ 21.6, 30.9, 31.2, 31.3, 42.9, 50.6, 59.5, 117.3, 120.1, 120.4,
122.7, 126.9, 121.1, 146.9, 147.1, 147.6, 148.6, 148.9, 149.1,
168.35, 168.5; MS (FAB, M+H+) 353.
(()-1-Dipropanoate. 1H NMR δ 1.0 (s, 3H), 1.1-1.4 (m, 12H),
1.7 (s, 3H), 2.2 (d, 1H, J ) 13 Hz), 2.4 (d, 1H, J ) 13 Hz),
2.5-2.6 (m, 4H), 6.8 (d, 1H, J ) 2 Hz), 6.9-7.0 (m, 3H), 7.1-
7.2 (m, 3H); MS (FAB, M+H+) 381.
(()-1-Dibutanoate. 1H NMR δ 1.1 (m, 10 H), 1.3 (s, 3H), 1.7
(s, 3H), 1.8 (m, 4H), 2.2 (d, 1H), 2.4 (d, 1H), 2.5(dt, 4H), 6.8 (d,
1H), 7.0 (m, 3H), 7.2 (m, 3H); MS (FAB, M+H+) 409.
(()-1-Dipentanoate. 1H NMR δ 0.8-1.1 (s+m, 10H), 1.3-
1.5 (s+m, 6H), 1.6-1.9 (s+m, 6H), 2.1-2.3 (m, 1H), 2.4-2.6
(s, 6H), 6.8 (s, 1H), 6.9-7.2 (m, 3H), 7.1-7.2 (m, 3H); 13C NMR,
δ 14.3, 14.4, 20.4, 22.8, 27.6, 30.1, 31.0, 31.2, 34.6, 43.0, 50.7,
53.8, 59.7, 117.4, 120.08, 120.4, 122.7, 126.9, 147.0, 147.7,
148.5, 148.9, 149.1, 171.2, 171.3; MS (FAB, M+H+) 437.

First Preparation of Enantiopure Indane Monomer
J . Org. Chem., Vol. 64, No. 20, 1999 7503
mL) was stirred magnetically and cooled in an ice-water bath.
Butanoyl chloride (21.0 g, 197 mmol, 2.43 equiv) was added
over 20 min. The flask was removed from the ice-water bath
and stirred for 1 h. The mixture was washed with NaHCO3 (1
M, 2 × 150 mL) and water (1 × 160 mL) yielding a clear,
slightly yellow solution. The solution was dried over MgSO4
and evaporated to a yellow oil (32 g). The oily material was
dissolved in tert-butyl methyl ether (240 mL) and transferred
to a 1 L round-bottomed flask. Phosphate buffer (240 mL, 0.1
M, pH 7.5) was added and the pH was adjusted to 7.20 with
sodium hydroxide (1 M). CVL (20.4 mg, 60 000 units, Sigma)
dissolved in phosphate buffer (15 mL) was added. The mixture
was stirred and the pH was maintained at 7.2 ( 0.3 by the
manual addition of sodium hydroxide (1 M). Approximately
20 mL were required in the first 3 h, and an additional 18 mL
over the next 2 h. HPLC analysis after 5 h showed 43%
conversion and 92% ee of (S)-(-)-1-4′-monobutanoate. The
reaction was stopped by separating the two phases and
extracting the aqueous phase with ethyl acetate (4 × 100 mL)
and dichloromethane (3 × 100). The aqueous phase containing
CVL was stored at -20 °C. The combined organic phases were
concentrated by rotary evaporation to a sticky gum which was
dissolved in dichloromethane (200 mL), washed with sodium
bicarbonate (1 M, 2 × 100 mL) and water (3 × 100 mL), dried
over MgSO4 (∼30 g), and concentrated by rotary evaporator
to a thick oil (30 g). Chromatography in three portions of 10 g
on silica gel column (500 g of 60-200 mesh packed in 7 cm
diameter column) yielded (R)-(+)-1-dibutanoate eluting with
dichloromethane: clear oil, 17.3 g, 54%, 51% ee; [R]D25 ) +59.3
(c ) 3.25, MeOH); Rf ) 0.84; 1H and 13C NMR spectra are
identical with the racemate. 1-Monobutanoate was obtained
by eluting with ethyl acetate (Rf ) 0.24 in dichloromethane,
pletely according to TLC. Sodium bicarbonate solution (20 mL,
10%) was added to the reaction mixture. The mixture was
extracted with ethyl acetate (3 × 10 mL). The combined
extracts were washed by water (3 × 20 mL) and dried over
magnesium sulfate (5 g). Column chromatography on silica
gel eluted with dichloromethane yielded a white solid: 51 mg,
1
48%; H NMR, δ 1.0 (m, 3H) 1.3 (m, 3H), 1.6 (m, 3H), 2.2 (m,
1H), 2.4 (m, 1H), 3.1 (m, 18H), 6.7 (m, 6H), 6.9-7.2 (m, 3H),
7.9-8.1 (m, 6H). The CD spectrum shown in Figure 3 was run
in chloroform, 0.1 cm path length, 5 accumulations at 100 nm/
min scan speed, room temperature, 0.2 nm resolution.
P r ep a r a tion of Lip a se fr om Ch r om oba cter iu m visco-
su m (ATCC 6918). Chromobacterium viscosum ATCC 6918
was grown on nutrient broth (Difco) at 37 °C for 24 h, then at
42 °C for 8 h. The culture was centrifuged at 5K rpm for 10
min. The pellet was discarded and the supernatant was used
to catalyze hydrolysis of 1-dibutanoate, see Table 1 in the text.
Qu a n tita tive HP LC An a lysis of En a n tiom er ic P u r ity.
Enantiomers of 1 were separated on a Chiralcel OD HPLC
column (Chiral Technologies Inc., Exton, PA) eluted with
hexanes-ethanol (90/10, 1 mL/min): δ ) 1.89, k′S ) 1.94, k′R
) 3.66. Four isomers of 1-monobutanoate and 1-monopen-
tanoate were separated on the Chiralcel OD column eluted
with hexanes-ethanol (96/4, 1 mL/min): 1-4′-monobutanoate,
δ ) 1.71, k′R ) 1.85, k′S ) 3.17; 1-5-monobutanoate, δ ) 3.01,
k′S ) 0.92, k′R ) 2.77; 1-4′-monopentanoate, δ ) 1.63, k′R
)
1.71, k′S ) 2.78; 1-5-monobutanoate, δ ) 2.44, k′S ) 0.95, k′R
) 2.31.9
Absolu te Con figu r a tion by X-r a y Cr ysta llogr a p h y. The
(+)-1-dipentanoate isolated from the CVL-catalyzed hydrolysis
was transformed to (+)-1 by treatment with sodium methoxide.
Crystals of (+)-1 were grown from dichloromethane. Data were
collected over a whole sphere, to 2 of 140ø on a CAD4
diffractometer using copper radiation. Data were processed
using the NRCVAX94 software package.10 Structures were
solved by direct methods (Shelxs-86) and refined against F2.
Rf ) 1 ethyl acetate): oil, 9.3 g, 35%, 92% ee; [R]D -61.6 (c
25
) 3.85, meow); 1H NMR, δ 1.0 (m, 6H) 1.3 (s, 3H), 1.6 (s, 3H),
1.7-1.8 (DT, 2H), 2.2 (d, 1H, J ) 13 Hz), 2.4 (d, 1H, J ) 13
Hz), 2.5 (t, 2H, J ) 7 Hz), 6.5 (d, 1H, J ) 2.5 Hz), 6.7 (DD,
1H, J 1) 2.5, J 2) 8 Hz), 6.9 (d, 2H, J ) 9 Hz), 7.0 (d, 1H, J )
8 Hz), 7.2 (d, 2H, J ) 9 Hz); 13C NMR, δ 13.7, 15.3, 18.5, 30.7,
31.1, 36.3, 42.4, 50.4, 53.5, 59.8, 66.0, 111.4, 114.7, 120.9, 123.5,
127.7, 144.5, 148.3, 148.5, 150.5, 154.8, 172.6. Sodium meth-
oxide (5.1 g of 95% purity, 2.2 equiv) was added to (R)-(+)-1-
dibutanoate (15.7 g, 38 mmol) dissolved in methanol (300 mL)
and the solution was stirred for 2 h at room temperature. The
reaction mixture was neutralized to pH 7 with HCl (1 M) and
diluted water (300 mL) and extracted with ethyl acetate (3 ×
250 mL). Evaporation of the combined extracts yielded (R)-
(+)-1: white powder, 10.2 g, 52% ee. The solid was dissolved
in a minimum amount of hot methanol and water was added
dropwise until the solution became turbid. Upon cooling two
crops of powder formed (2.1 g, 3% ee (R), 200 mg, 13% ee (R)),
leaving the mother liquor with 62% ee (R). The mixture was
then extracted by methylene chloride and the aqueous phase
discarded. The solid was recrystallized from methanol-dichlo-
romethane. The first crop yielded white crystals of (R)-(+)-1:
Ack n ow led gm en t. The authors extend thanks to
Molecular OptoElectronics Corporation for financial
support (M.Z.) and for the gift of racemic compound 1.
Thanks should also be addressed to Professor J . Turn-
bull (Concordia University) for permitting the use of
circular dichroism as well as to Dr. A. M. Lebuis for
X-ray crystallography, Mr. N. Saade for GC-mass
spectrometry, and to Dr. F. Sauriol for recording COSY
spectra.
Su p p or tin g In for m a tion Ava ila ble: The following ma-
terials are available: (1) Figure S1. HPLC chromatograms
showing separation of enantiomers of monoesters and diol. (2)
A case by case rational discussion of possibilities to achieve a
good resolution of the complex system. (3) Figure S2. X-ray
crystallographic data and ORTEP drawing. This material is
available free of charge via the Internet at http://pubs.acs.org.
2.5 g, >99% ee; [R]D ) +105.2 (c ) 3.50, MeOH); mp 152-
25
158 °C (racemic mp 203-206 °C); NMR spectrum was identical
with racemic 1. Circular dichroism spectrum shows a positive
Cotton effect at 285 nm as shown (dotted line) in Figure 3.
(R)-(+)-1-Di-4-d im eth yla m in oben zoa te. A solution of
4-dimethylaminobenzoyl chloride (88 mg, 0.47 mmol, 2.5 equiv)
in dry THF (5 mL) was added over 10 min to a solution of
(R)-(+)-1 (50 mg, 0.19 mmol) and triethylamine (47 mg, 0.46
mmol, 2.5 equiv) in dry THF (10 mL). The reaction was stirred
overnight at room temperature. The diol had reacted com-
J O990795W
(9) At first we found it difficult to separate the four isomeric
1-monopentanoates and used a ternary mixture of hexane-2-propanol-
ethanol (97.5/1/0.25) to elute the Chiracel OD column. Once we started
screening reaction mixtures, the separation improved dramatically and
we used the conditions noted above. We do not know why separation
improved.
(10) Gabe, E. J .; LePage, Y.; Charland, J . P.; Lee, F. L.; White, P.
S. J . Appl. Crystallogr. 1989, 22, 384-389.