(R,S)-azolides as novel substrates for lipase-catalyzed hydrolytic resolution in organic solvents

Article abstract of DOI:10.1002/adsc.200900391

Azolides, that is, N-acylazoles, as versatile acylation reagents are well characterized in the literature, in which the azole structure can not only act as a better leaving group but also make the carbonyl carbon more electrophilic and susceptible to nucleophilic attack. It is therefore desirable to combine this unique property and lipase resolution ability in the development of a new resolution process for preparing optically pure carboxylic acids. With the Candida antarctica lipase B (CALB)-catalyzed hydrolysis of (R,S)-N- profenylazoles in organic solvents as the model system, (R,S)-N-profenyl-l,2,4- triazoles instead of their corresponding ester analogues were exploited as the best substrates for preparing optically pure profens, i.e., 2-arylpropionic acids. The structure-reactivity correlations for the (R,S)-azolides in water-saturated methyl tert-butyl ether (MTBE) at 45°C coupled with a thorough kinetic analysis were further employed for elucidating the rate-limiting formation of a tetrahedral adduct without C-N bond breaking or with moderate C-N bond breaking concerted with C-O bond formation in the acylation step. The advantages of easy substrate preparation, high enzyme reactivity and enantioselectivity, and easy recovery of the product and remaining substrate by aqueous extraction demonstrate the potential of using (R,S)-azolides as novel substrates for the enzymatic resolution process.

Full text of DOI:10.1002/adsc.200900391

FULL PAPERS
DOI: 10.1002/adsc.200900391
ACHTUNGTRENNUNG
Pei-Yun Wang,a Ying-Ju Chen,a An-Chi Wu,a Yi-Sheng Lin,a Min-Fang Kao,a
Jin-Ru Chen,a Jyun-Fen Ciou,a and Shau-Wei Tsaia,*
a
Institute of Biochemical and Biomedical Engineering, Chang Gung University, Kwei-Shan Tao-Yuan, 33302, Taiwan,
R.O.C.
Fax : (+886)-3-211-8668; phone: (+886)-3-211-8800 ext: 3415; e-mail: tsai@mail.cgu.edu.tw
Received: June 4, 2009; Revised: July 24, 2009; Published online: September 23, 2009
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200900391.
Abstract: Azolides, that is, N-acylazoles, as versatile water-saturated methyl tert-butyl ether (MTBE) at
acylation reagents are well characterized in the liter- 458C coupled with a thorough kinetic analysis were
ature, in which the azole structure can not only act further employed for elucidating the rate-limiting
À
as a better leaving group but also make the carbonyl formation of a tetrahedral adduct without C N bond
À
carbon more electrophilic and susceptible to nucleo- breaking or with moderate C N bond breaking con-
À
philic attack. It is therefore desirable to combine this certed with C O bond formation in the acylation
unique property and lipase resolution ability in the step. The advantages of easy substrate preparation,
development of a new resolution process for prepar- high enzyme reactivity and enantioselectivity, and
ing optically pure carboxylic acids. With the Candida easy recovery of the product and remaining substrate
antarctica lipase B (CALB)-catalyzed hydrolysis of by aqueous extraction demonstrate the potential of
(R,S)-N-profenylazoles in organic solvents as the using (R,S)-azolides as novel substrates for the enzy-
model system, (R,S)-N-profenyl-1,2,4-triazoles in- matic resolution process.
stead of their corresponding ester analogues were ex-
ploited as the best substrates for preparing optically Keywords:
pure profens, i.e., 2-arylpropionic acids. The struc- B; hydrolytic resolution; improved activity and enan-
ture-reactivity correlations for the (R,S)-azolides in tioselectivity; (R,S)-profens
ACHTUNGTRENUN(NG R,S)-azolides; Candida antarctica lipase
AHCTUNGTRENNUNG
Introduction
Aside from the enzyme engineering approach[6] of
using site-directed mutations, evolutions or chemical
Serine-type hydrolases such as esterases, lipases and modifications as well as the medium engineering
serine proteases are widely applied as versatile biocat- strategy[7] of varying temperature, reaction media or
alysts for preparing a variety of pharmaceuticals and additives for improving the activity and enantioselec-
fine chemicals containing a chiral center.[1] Their tivity, the substrate engineering approach[8] of em-
active site consisting of a Ser-His-Asp/Glu catalytic ploying activated or irreversible acyl donors or ac-
triad catalyzes the hydrolysis or organic synthesis by ceptors has been proposed for the fine-tuning of
following an acylation-deacylation displacement enzyme active-site structures to enhance the enantio-
mechanism.[2] Frequently, the acylation step on the mer discrimination towards targeted racemates. Re-
formation of an acyl-enzyme intermediate for the hy- cently, in order to rationalize the optimal enantiose-
drolysis of esters[3] or amides[4] was found to be the lectivity of esterase or lipase-catalyzed hydrolysis of
rate-limiting step. Moreover, since the catalytic histi- (R,S)-2-substituted or (R,S)-3-substituted carboxylic
dine was highlighted as a general acid-base catalyst in acid esters in the presence of an organic solvent, a
the whole acylation step, an expanded Michaelis– two-stage Brønsted slope for the fast-reacting enan-
Menten kinetics was further employed for interpret- tiomers but not their antipodes was determined.[3d,6g,9]
ing the structure-reactivity correlations in terms of This indicated that, by altering the leaving alcohol
rate-limiting formation or breakdown of the tetrahe- moiety, the rate-limiting acylation step might shift
dral adduct.[5]
from breakdown to formation of the tetrahedral
Adv. Synth. Catal. 2009, 351, 2333 – 2341
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2333

Pei-Yun Wang et al.
FULL PAPERS
adduct and lead to a maximum enantioselectivity at values are found to be less than 20 for the resolution
the breaking point. If the rationale of obtaining the of (R,S)-profens via esterification, hydrolysis or trans-
optimal enantioselectivity is valid for serine-type hy- esterification.[13] The low enantioselectivity towards
drolases catalyzing the hydrolytic resolution of (R,S)- stereocenters in the acyl part of an ester may be at-
carboxylic acid esters, the question is then how one tributed to fewer and/or weaker contacts between the
can further stabilize the transition state for the forma- acyl part and its binding site, in contrast to the excel-
tion of the tetrahedral adduct of fast-reacting enantio- lent enantiopreference towards the nucleophiles such
mers and improve the enzyme performance. Appa- as secondary alcohols and primary amines in the re-
rently, the enzyme engineering approach of changing stricted active pocket.[14]
or chemically modifying the amino acid residues
In order to expand CALB as an efficient biocata-
around the catalytic triad to enhance the nucleophilic lyst for resolving optically pure carboxylic acids, the
power of catalytic serine oxygen atom is a possible, hydrolytic resolution of (R,S)-N-profenylazoles in or-
but difficult to obtain, answer.[10] On the other hand, ganic solvents was selected as the model system
the substrate engineering strategy of using (R,S)-azo- (Scheme 1). Effects of solvent, water content, temper-
lides but not their corresponding ester analogues as ature, and substrate structure on varying the activity
the extreme substrates is another feasible and easy to and enantioselectivity were first studied. The Brønst-
reach solution.
ed slopes estimated from the structure-reactivity cor-
Azolides are heterocyclic amides in which the relations for (R,S)-N-2-phenylpropionylazoles (1–3),
amide nitrogen is part of an azole ring. Their synthet- (R,S)-2-phenylpropionic acid esters (4–7) and (R,S)-
ic properties, structural aspects and hydrolysis kinetics N-2-naproxenazoles (8–10) in water-saturated MTBE
in aqueous solutions are well characterized in the lit- were then employed for elucidating the rate-limiting
erature.[11] Despite the wide spectrum of reactivity dis- formation or breakdown of the tetrahedral adduct of
played in nucleophilic reactions, azolides as versatile the acylation step. Moreover, comparisons of the ki-
acylation reagents can be easily prepared from the netic constants obtained at various reacting conditions
acylation of azoles with acid chlorides or via direct ac- as well as the separation of remaining optical (S)-3
tivation of carboxylic acids by N,N-carbonylbisazoles. using pH 7 phosphate buffers as the extract were ad-
In contrast to normal amides showing particularly low dressed.
reactivity in such nucleophilic reactions as hydrolysis,
alcoholysis and aminolysis under mild conditions, the
high reactivity of azolides can be attributed to the un- Results and Discussion
shared electron pairs on the acyl-substituted nitrogen
N(1) that are part of the cyclic p-system of the azole Effects of Solvent, Water Content, and Temperature
units. This unique structure has led to a partial posi-
tive charge on N(1) that interfaces with the normal Table 1 demonstrates the effects of water-saturated
carboxamide resonance and exerts an electron-with- organic solvents on the specific activity [i.e., VR/(Et)
drawing effect on the carbonyl groups, making the and VS/(Et)] and enantioselectivity in terms of the
carbonyl carbon more electrophilic and susceptible to ratio of initial rates (i.e., VR/VS) for (R,S)-N-2-phenyl-
nucleophilic attack.[11f] Moreover, with increasing propionyl-1,2,4-triazole (3). In general, the specific ac-
number of nitrogens, the azole units become better tivity for each enantiomer increases with the solvent’s
leaving groups, especially when accepting an proton hydrophobicity in terms of the log P value, and leads
from proton-donating environments, for example, the to an optimal enantioselectivity of VR/VS =95.6 for
imidazolium of catalytic histidine of serine-type hy- water-saturated MTBE. In comparison with the low E
drolases. It is therefore aimed to realize the combina- values of 3.5 for the hydrolysis of (R,S)-2-phenylpro-
tion of azolide chemistry and enzymatic resolution pionyl ethyl ester in pH 7 buffers as well as 26 or 17
ability for developing a new resolution process for for the transesterification of (R,S)-2-phenylpropionyl
preparing optically pure carboxylic acids.
vinyl or ethyl ester with n-hexanol in toluene at
Candida antarctica lipase B (CALB) is well recog- 408C,[13e] the high enantioselectivity can apparently be
nized for tolerating a great variation in experimental attributed to the unique leaving 1,2,4-triazole moiety
conditions such as polar organic solvents, alkaline pH that exerts chiral discrimination ability for the enan-
and elevated temperature.[1] It exhibits a very high tioselective hydrolysis of 3 but not 4–7. This can be
degree of activity and substrate selectivity both with elucidated later from the linear structure-reactivity
respect to regioselectivity and enantioselectivity, espe- correlations of the variation of log
ACHTUGNRTNE(NUNG k2R/KmR) and
cially for the resolution of alcohols and amines, allow- log(k2S/KmS) with pKa of the leaving azole moiety in
AHCTUNGTRENNUNG
ing the industrial preparation of an important variety Figure 1.
of optically active hydroxy and amino compounds.[12]
Several models in terms of solvent characteristics,
However, CALB also shows low to modest enantiose- desolvation of substrates in the enzyme-bound transi-
lectivity towards carboxylic acids, for example, the E tion state, flexibility of enzyme, and exchange of
2334
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2333 – 2341

ACHTUNGTRENNUNG(R,S)-Azolides as Novel Substrates for Lipase-Catalyzed Hydrolytic Resolution
FULL PAPERS
Scheme 1. CALB-catalyzed hydrolytic resolution of (R,S)-N-profenylazoles and (R,S)-profen esters in organic solvents.
enzyme-bound water have been proposed for inter- for water-saturated organic solvents, the hydration
preting solvent influences on the enzyme perfor- water level on the enzyme surface and active site
mance.[15] Although the overall enzyme structures and region in different solvents may change.[15b,e] There-
individual residue flexibility do not show significant fore, the low water content in water-saturated non-
differences in water, for organic solvents and hence polar solvents should result in a low hydration water
Table 1. Effects of solvent on specific initial rates VR/(Et) and VS/(Et), VR/VS, racemate conversion Xt, enantiomeric excess
for the substrate ees for a given lipase concentration (Et).[a]
Solvent (log P)
VR/(Et) [mmol/h·g] VS/(Et) [mmol/h·g] VR/VS (Et) [mgmLÀ1] Time [h] Xt [%] ees [%]
Isooctane (4.50)
Hexane (3.90)
Cyclohexane (3.20)
Diisopropyl ether (1.90) 4.31
MTBE (1.43)
9.17
7.52
5.20
4.76E-1
2.16E-1
1.23E-1
6.78E-2
4.58E-2
1.27E-2
19.3
34.8
42.3
63.5
95.6
44.6
2.0
2.0
2.0
2.0
2.0
6.0
0.38
0.50
0.67
1.00
0.67
5.00
58.8
59.5
54.6
52.0
55.6
66.6
100.0
100.0
100.0
100.0
100.0
100.0
4.38
5.67E-1
Cyclohexanone (0.96)
[a]
Conditions: 10 mL water-saturated solvent containing 3 mM of 3 at 458C and 400 rpm. Symbol of E-1 as 10À1.
P as partition coefficient of the solvent bewteen octanol and water.
[b]
Adv. Synth. Catal. 2009, 351, 2333 – 2341
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2335

Pei-Yun Wang et al.
FULL PAPERS
water-saturated MTBE should be lower than those in
water-saturated hexane as demonstrated below in
Table 5.
The inverse dependence of the specific initial rates
on the water content in Table 2 is also perceived and
yields a maximum enantioselectivity of VR/VS =95.6 in
water-saturated MTBE. Similar kinetic behaviors for
the hydrolysis of lutein diesters in hexane via CALB
or dodecyl decanonate in diisopropyl ether via Rhizo-
pus arrhizus lipase have been reported.[16] As expect-
ed for the water-saturated hexane containing less
water content and hence hydration water level on the
active site region, the kinetic constants k2R and KmR in
anhydrous MTBE are lower than those in water-satu-
rated MTBE (see Table 5 below).
Table 3 further shows the temperature effects on
varying the lipase activity and enantioselectivity in
~
&
*
Figure 1. Variations of log
N
,
), log
(VR/VS) ( , , N) with pKa of the leaving
azole and alcohol group; (–) best-fit results. Conditions as
ACHTUGNETRN(NNUG k2S/KmS)
~
!
^
*
( , &, ), and log
ACHTUNGTRENNUNG
water-saturated MTBE. liner relationship of
A
ln(VR/(Et)), ln(VS/(Et)) or lnACTHNGUTERN(NUG VR/VS) with inverse of
A
U
~
~
given in Table 4 for (R,S)-N-2-phenylpropionylazoles ( ,
,
the absolute temperature was found (data not
shown), implying that the lipase was still thermally
stable at 558C. Yet, 458C was selected as the best
temperature by considering a compromise between
!
^
), (R,S)-N-naproxenylazoles ( , &, ), and (R,S)-2-phenyl-
*
*
propionic acid esters ( , , N).
level on the active site region, leading to the easy sub- the enzyme activity and enantioselectivity.
strate affinity to the binding pockets but difficult
proton transfer from the catalytic serine to histidine
for performing the nucleophilic reaction owing to the Effects of Substrate Structure
lack of water molecules acting as the lubricant.[15d]
This implies that the catalytic constant k2R and Mi- The variations of acyl or leaving groups on the
chaelis constant KmR for the fast-reacting (R)-3 in enzyme performance are represented in Table 4. A
change of the leaving azole moiety from imidazole,
pyrazole to 1,2,4-triazole in 1–3 results in 152- and
9.6-fold enhancements of VR/(Et) and VS/(Et) for the
fast- and slow-reacting substrates, respectively, and
then increasing of the enantioselectivity of VR/VS =6.1
for 1 to 95.6 for 3. Similarly, this change also yields
109- and 1.5-fold enhancements of VR/(Et) and VS/(Et)
in 8–10 and hence VR/VS =1.1 for 8 and 81.8 for 10.
However, if the leaving moiety is varied from metha-
nol to 2,2,2-trifluoroethanol in 4–7, only 19.6- and
12.4-fold enhancements of VR/(Et) and VS/(Et) with
minute changes of VR/VS in the range of 4.7 and 7.4
are obtainable. By considering the improved activity
and enantioselectivity of using 3 but not 7 (or 10 but
not 19) as the substrate, the fact that (R,S)-N-profen-
yl-1,2,4-triazoles are the best substrates for CALB-
Table 2. Effects of water content on specific initial rates and
VR/VS.[a]
Saturated-water con- VR/(Et)
VS/(Et)
VR/
tent [%]
[mmol/h·g]
[mmol/h·g]
VS
0
1.08E+1
9.24
6.61
5.56
4.44
7.23E-1
3.01E-1
1.48E-1
8.38E-2
6.14E-2
4.58E-2
14.9
30.6
44.8
66.3
72.2
95.6
20
40
60
80
100
4.38
[a]
Conditions: 10 mL MTBE containing 2 mgmLÀ1 CALB
and 3 mM of 3 at 458C and 400 rpm. Symbol of E-1 as
10À1.
Table 3. Effects of temperature on specific initial rates, VR/VS, Xt and ees.[a]
Temp. [oC]
VR/(Et) [mmol/h·g]
VS/(Et) [mmol/h·g]
VR/VS
(Et) [mgmLÀ1]
Time [h]
Xt [%]
ees [%]
15
25
35
45
55
1.32
1.91
3.54
4.38
7.78
6.75E-3
1.51E-2
3.50E-2
4.58E-2
1.17E-1
196.8
126.9
101.3
95.6
2
2
2
2
2
6.00
2.25
1.08
0.67
0.50
55.3
50.9
51.6
55.6
53.6
100.0
100.0
100.0
100.0
100.0
66.7
[a]
Conditions: 10 mL water-saturated MTBE containing 3 mM of 3. Symbol of E-1 as 10À1.
2336
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2333 – 2341

ACHTUNGTRENNUNG(R,S)-Azolides as Novel Substrates for Lipase-Catalyzed Hydrolytic Resolution
FULL PAPERS
Table 4. Effect of substrate structure on pKa of leaving moiety, specific initial rates, VR/VS, Xt and ees.[a]
Substrate pKa (of conjugate acid) pKa (of acid) VR/(Et) VS/(Et) VR/VS (Et)
Time Xt
ees
A
E
N
[%] [%]
1
2
3
4
5
6
7
8
7.00
2.52
2.19
14.40
14.21
10.04
15.49
15.12
14.82
12.37
14.40
14.21
10.04
10.04
10.04
10.04
10.04
15.49
15.49
15.49
15.49
15.49
2.88E-2
2.62
4.38
4.76E-3
3.23E-2
4.58E-2
2.34E-2
8.27E-2
1.61E-1
2.92E-1
2.10E-3
2.85E-3
3.18E-3
1.61E-3
1.80E-2
1.87E-3
3.22E-2
2.02E-2
2.30E-3
2.55E-3
2.55E-2
2.18E-3
6.1
20.0
2.0
2.0
2.0
6.0
2.0
2.0
40.0
10.0
6.0
2.0
2.0
10.0
2.0
2.0
10.0
10.0
10.0
10.0
4.3
1.0
0.7
9.0
1.0
1.8
0.6
24.0
5.0
3.0
21.0
4.5
5.0
1.7
10.0
23.4
21.0
2.0
51.8 58.8
53.5 100.0
55.6 100.0
44.5 48.1
56.5 64.4
60.1 69.1
60.4 100.0
62.0 8.6
50.5 82.5
53.3 100.0
51.8 100.0
52.3 100.0
55.8 100.0
53.1 100.0
54.9 68.3
49.2 36.9
56.8 22.3
61.1 84.0
24.2 17.2
81.3
95.6
4.7
6.2
5.7
1.10E-1
5.17E-1
9.15E-1
2.16
2.38E-3
1.00E-1
2.60E-1
3.03E-1
2.55
7.4
1.1
7.00
2.52
2.19
2.19
2.19
2.19
2.19
9
35.2
81.8
188.2
141.6
82.3
92.8
7.5
3.3
3.1
8.3
3.8
10
11
12
13
14
15
16
17
18
19
1.54E-1
2.99
1.52E-1
7.48E-3
7.79E-3
2.12E-1
8.36E-3
8.3
[a]
Conditions: 10 mL water-saturated MTBE containing 3 mM racemate at 458C and 400 rpm. Symbol of E-1 as 10À1.
Table 5. Effects of water content, solvent and substrate structure on kinetic constants and E (i.e., k2RKmS/k2SKmR).[a]
Substrate
3
Water-saturated
MTBE
4
7
Anhydrous
MTBE
Water-saturated
hexane
Water-saturated
MTBE
Water-saturated
MTBE
k2R [mmol hÀ1·g]
KmR [mM]
2.01E+2
2.54E+1
7.91
2.07E+1
3.42E+1
6.05E-1
13.1
1.33E+3
4.59E+2
2.90
1.55E+1
5.31E+2
2.92E-2
99.0
7.83E+1
1.11E+1
7.04
1.70
8.43
6.83E-1
1.53E+1
4.47E-2
1.23E-1
1.49E+1
8.29E-3
5.4
3.05E+1
2.99E+1
1.02
k2R/KmR [LhÀ1·g]
k2S [mmolhÀ1·g]
KmS [mM]
5.29
4.45E+1
1.19E-1
8.6
k2S/KmS [LhÀ1·g]
E
2.01E-1
35.0
[a]
Conditions: 10 mL MTBE or hexane containing 20 mg CALB at 458C and 400 rpm. Symbol of E-1 as 10À1.
catalyzed hydrolytic resolution for preparing optically adduct for the esters containing a bad leaving alcohol
pure profens is evident.
moiety. This slope levels off to À0.15 for (R)-6 and
Table 4 further demonstrates more than an order- (R)-7 [or À0.11 for (S)-6 and (S)-7], indicating that
of-magnitude change of VR/(Et) in (R)-1–(R)-7 and the rate-limiting step has shifted to formation of the
VS/(Et) in (S)-1–(S)-7, indicating that the acylation tetrahedral adduct for the esters containing a good
step must be rate-limiting as (R)-1–(R)-7 [or (S)-1– leaving group. Similar kinetic behaviors have been re-
(S)-7] have the same acyl-enzyme intermediate for ported for the hydrolysis of fast-reacting (S)-2-substi-
proceeding the deacylation step. Since KmR and KmS tuted or (S)-3-substituted carboxylic acid esters but
for 3 or 4 are at least an order-of-magnitude higher not their antipodes.[3d,6g,9] However, only a Brønsted
than the initial enantiomer concentration of 1.5 mM slope of À0.45 for the fast-reacting (R)-1–(R)-3 and
in Table 5, as an approximation one can calculate that of À0.20 for their antipodes (S)-1–(S)-3, as well
k2R/KmR and k2S/KmS from VR/(Et)/(SR) and VS/(Et)/ as À0.40 for (R)-8–(R)-10 and À0.034 for (S)-8–(S)-
(SS), respectively. Figure 1 illustrates the linear struc- 10, are illustrated from the variation of log
ture-reactivity correlations of log(k2R/KmR), log(k2S/ or log(k2S/KmS) with pKa of the leaving azole moiety.
KmS) and log(VR/VS) with pKa of the leaving alcohol The kinetic and mechanistic effects of the leaving
ACHTUGNETRN(NNUG k2R/KmR)
A
R
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
moiety for 4–7. A steep ascent of the Brønsted slope N-acylimidazole groups in non-enzymatic alkaline
À1.39 for (R)-4–(R)-6 [or À1.26 for (S)-4–(S)-6] im- and a-chymotrypsin-catalyzed hydrolysis reactions in
plies the rate-limiting breakdown of the tetrahedral aqueous solutions have been studied, leading to a
Adv. Synth. Catal. 2009, 351, 2333 – 2341
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2337

Pei-Yun Wang et al.
FULL PAPERS
Brønsted slope of À0.28 for the former and that of water content but also solvent characteristics strongly
À0.30 for the latter.[11c,e] Consequently, the transition affect the proton transfer and hence nucleophilic
states for the rate-limiting nucleophilic attack at the attack to the carbonyl carbon atom. In general, the
carbonyl groups in both reactions should resemble high water content is beneficial for giving high k2R
the reactants. This argument can be applied to and k2S that may lead to the same order-of-magnitude
CALB-catalyzed hydrolysis of (R)-1–(R)-3 and (R)-8– of k2R/KmR but not k2S/KmS, and then an optimal enan-
(R)-10 but not (R)-4–(R)-7 in water-saturated MTBE. tioselectivity for water-saturated MTBE. In order to
Therefore, the rate-limiting acylation step for the hy- quantitatively intercorrelate the contributions from
drolysis of (R)-1–(R)-3 or (R)-8–(R)-10 must be either the water content and solvent properties, more studies
À
the formation of tetrahedral adduct without C N of using other (R,S)-azolides as the substrate in or-
À
bond breaking or with moderate C N bond breaking ganic solvents are required.
À
concerted with C O bond formation. By increasing
the number of nitrogens from imidazole to 1,2,4-tri- rated MTBE are further compared and interpreted
The kinetic constants for 3, 4 and 7 in water-satu-
ACHTUNGTRENNUNG
An analysis of the minute increase of Brønsted in comparison with the simple leaving alcohol moiety
slope to À0.20 for (S)-1–(S)-3 [or À0.034 for (S)-8– the bulky 1,2,4-triazole unit is subject to more steric
(S)-10] in comparison with that of their antipodes im- repulsion in the restricted binding pocket. Yet, this
plies that the steric repulsion exerted on the leaving disadvantage can result in the chiral discrimination of
azole moiety in the restricted active pockets may giving k2R/k2S equal to 86 for 3 in contrast to about
À
twist the C N bond of slow-reacting (S)-azolides, and 5.5 for 4 and 7. Moreover, 2,2,2-trifluoroethanol is a
leads to the discrimination of proton transfer between better leaving group than methanol by considering
(R)- and (S)-azolides. Yet, this does not occur for (S)- about 44-fold enhancements of k2R and k2S (Table 5)
4–(S)-7 containing a simple leaving alcohol moiety. for 4 and 7, and hence the shift of rate-limiting break-
Consequently, one obtains excellent improvements on down to formation of the tetrahedral adduct. By fur-
the lipase activity and enantioselectivity for other ther comparing k2R or k2S for 3 and 7, it is evident
(R,S)-N-profenylazoles 11–14 in comparison with that the leaving 1,2,4-triazole group can make the car-
their corresponding methyl esters 15–18. In order to bonyl carbon more electrophilic and susceptible to
further shed insight into effects of solvent, water-con- nucleophilic attack in comparison with the 2,2,2-tri-
tent and substrate structure on the enzymatic cataly- fluoroethanol moiety. Similar kinetic behaviors of
sis, the kinetic constants using Michaelis–Menten ki- giving very large Michaelis constants associated with
netics for the acylation step were estimated and ex- effective nucleophilic attack of catalytic serine oxygen
amined.
atom in the acylation step have been reported for a-
chymotrypsin-catalyzed hydrolysis of N-acylimid-
AHCTUNGTRENNUNG
azoles in aqueous solutions.[11b,e] Therefore, the cur-
Kinetic Analysis
rent approach of using (R,S)-azolides as novel sub-
strates for obtaining excellent improvements on
Figure S1-S3 in Supporting Information illustrates CALB activity and enantioselectivity may apply to
the variation of initial rates with substrate concentra- other serine-type hydrolases that obey the rate-limit-
tions, with which the kinetic constants are estimated ing acylation step when carrying out the hydrolysis or
and represented in Table 5. More than an order-of- alcoholysis in organic solvents.
magnitude higher KmR and KmS for 3 are shown when
the reaction medium of anhydrous MTBE is replaced
with water-saturated MTBE containing 472.7 mM of Conclusions
water.[17] This implies that water molecules adsorbed
on the active site region may impede the substrate af- With CALB-catalyzed hydrolytic resolution of (R,S)-
finity to the binding pockets. Therefore, lower KmR N-profenylazoles in organic solvents as the model
and KmS values are expectable (Table 5) if water-satu- system, (R,S)-N-profenyl-1,2,4-triazoles were exploit-
rated MTBE is further changed to water-saturated ed as best substrates for preparing optically pure pro-
hexane only containing 15 mM of water.[18] Similar ki- fens. By using 3 as the substrate, water-saturated
netic behaviors of obtaining very large Michaelis con- MTBE and 458C were selected as the best reaction
stants in the acylation step have been reported for the conditions for studying effects of varying the substrate
a-chymotrypsin-catalyzed hydrolysis of N-acylimid- structure on improving the enzyme performance. In
A
general high enantioselectivity (VR/VS >81.8) was ob-
An inspection of changing of k2R and k2S for 3 in tainable for the substrates containing a leaving 1,2,4-
different organic solvents indicates that not only triazole but not a pyrazole, imidazole or alcohol
2338
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2333 – 2341

ACHTUNGTRENNUNG(R,S)-Azolides as Novel Substrates for Lipase-Catalyzed Hydrolytic Resolution
FULL PAPERS
6.0 mL benzene. The resultant solution was stirred for 2 h at
room temperature. After being quenched in succession with
0.1M HCl solution (3ꢁ10 mL), 0.1M NaOH solution (3ꢁ
10 mL) and 0.1M NaCl solution (3ꢁ10 mL), the organic
phase was separated, dried over anhydrous MgSO4, filtered
and concentrated under reduced pressure, giving the desired
(R,S)-N-profenylpyrazoles.
To 300 mmol alcohol were added 3.5 mmol (S)- or (R,S)-
2-phenylpropionic acid and 5 mmol sulfuric acid and the
mixture was stirred at 558C for 18 h. After removing the re-
maining methanol under reduced pressure, the residue was
extracted with a mixture consisting of 0.1M NaOH solution
(10 mL) and dichloromethane (10 mL). The organic phase
was separated, dried over anhydrous MgSO4, filtered and
concentrated under reduced pressure, giving the desired (S)-
or (R,S)-2-phenylpropionyl esters.
moiety. The structure-reactivity correlations in the
acylation step for (R,S)-N-2-phenylpropionylazoles
(1–3), (R,S)-2-phenylpropionyl esters (4–7) and (R,S)-
N-naproxenylazoles (8–10) were further compared for
elucidating the rate-limiting formation or breakdown
of the tetrahedral adduct for 4–7, and formation of
À
tetrahedral adduct without C N bond breaking or
À
with moderate C N bond breaking concerted with
C O bond formation for 1–3 and 8–10. A thorough
À
kinetic analysis at various reaction conditions further
shed insights into effects of solvent, water content,
and substrate structure on the enzymatic catalysis.
Moreover by using pH 7 phosphate buffers as the ex-
tract at 48C, 95.0% of (S)-3 on the basis of initial en-
antiomer concentration was recovered from the or-
ganic phase.
Similar procedures were carried out to prepare other
(R,S)-profenyl esters. All the synthesized substrates were
confirmed from the retention time in HPLC analysis and
1H NMR spectra recorded at 400 MHz on Brucker AC-400
spectrometer in DMSO-d6 solution with TMS as an internal
standard.
Experimental Section
Materials
Analysis
Novozym 435 from Candida antartica lipase B was provided
by Novo Nordisk (Bagsvaerd, Denmark). Dimethyl sulfox-
ide-d6 (DMSO-d6) containing 1% v/v tetramethylsilane
Hydrolysis of (R,S)-N-profenylazoles was monitored by
HPLC using a chiral column from Daicel (OB-H, OD-H or
OJ-H; Tokyo, Japan) or Regis ((S,S)-whelk-01; Morton-
Grove, IL) that is capable of separating the internal stan-
dard of 2-nitrotoluene or p-nitrophenol. UV detection at
220 or 270 nm was employed for quantification at the room
temperature. Detailed analytical conditions are represented
in Table S1 (Supporting Information).
1
(TMS) for H NMR analysis was from Cambridge Isotope
Laboratories (Andover, MA). Other chemicals of analytical
grade were commercially available: pyrazole from Acros
(Geel, Belgium); N,N’-carbonyldiimidazole (CDI), (S)-flur-
biprofen [i.e., (S)-2-fluoro-a-methyl-4-biphenylacetic acid],
(S)-ibuprofen [i.e., (S)-2-(4-isobutylphenyl)-propionic acid],
(S)-2-phenylpropionic acid, (S)-ketoprofen [i.e., (S)-2-(3-
benzoylphenyl)-propionic acid], and 2,2,2-trifluoroethanol
from Aldrich (Milwaukee, WI); N,N’-carbonyldi-1,2,4-tri-
Effects of Solvent, Water Content, Temperature, and
Substrate Structure
ACHTUNGTRENNUNGazole (CDT) and (R,S)-2-phenylpropionic acid from Fluka
(Buchs, Switzerland); (R,S)-ketoprofen, (S)-naproxen [i.e.,
(S)-2-(6-methoxy-2-naphthyl)-propionic acid] and (R,S)-nap-
roxen from TCI (Tokyo, Japan); (R,S)-flurbiprofen, (R,S)-
ibuprofen and (R,S)-fenoprofen [i.e., (R,S)-2-(3-phenoxy-
phenyl)-propionic acid] from Sigma (St. Louis, MO); di-
chloromethane, isooctane, isopropyl alcohol, hexane, diiso-
propyl ether, methanol, methyl tert-butyl ether (MTBE) and
triethylamine from Tedia (Fairfield, OH); 2-ethoxyethanol
and 2-methoxyethanol from Merck (Darmstadt, Germany).
Anhydrous MTBE was prepared by adding calcium hydride
from Riedel-de Haen (Seelze, Germany) to MTBE for 24 h.
To 10 mL water-saturated solvent (i.e., isooctane, hexane,
MTBE, IPE or cyclohexanone) containing 3 mM of 3 was
added specific amounts of Novozym 435 (referred as Novo
435). The resultant solution was stirred with a magnetic stir-
rer at 458C. Samples were removed from the reaction
medium at different time intervals for HPLC analysis, with
which the time-course conversion XR or XS, initial rates for
both enantiomers VR and VS, racemate conversion Xt, and
enantiomeric excess for the substrate ees were determined.
Similar experiments were performed in MTBE containing
0%, 20%, 40%, 60% and 80% of saturated water at 458C
for studying effects of water content on the activity and
enantioselectivity. More experiments were carried out in
water-saturated MTBE by varying the temperature from 15
to 558C for investigating temperature effects on the lipase
performance. Other substrates illustrated in Scheme 1 were
also employed as the substrate in water-saturated MTBE for
investigating effects of the acyl or leaving moiety.
In order to compare the enzyme kinetic behaviors at
458C, the hydrolytic reactions with substrate concentrations
varied in the range of 3–200 mM for 3, 2.5–15 mM for 4,
1.25–7.5 mM for (S)-4, 3–50 mM for 7 and 1.5–25 mM for
(S)-7 in water-saturated MTBE, as well as 1.0–30 mM for 3
in water-saturated hexane and 0.25–15 mM for (S)-3 in an-
hydrous MTBE were further carried out to estimate the ini-
tial rate for each enantiomer. Since the lipase has high enan-
General Procedures for Substrates Synthesis
To 5 mL benzene were added 1 mmol (S)- or (R,S)-2-phe-
nylpropionic acid and 1.5 mmol CDT and the mi xture was
stirred at 558C for 2 h. The resultant mixture was filtered
and evaporated under reduced pressure, giving the desired
(R,S)-N-(2-phenylpropionyl)-1,2,4-triazole.
Similar procedures were carried out to prepare other
(R,S)-N-profenyl-1,2,4-triazoles, and then (R,S)-N-profenyl-
ACHTUNGTRENNUNG
thionyl chloride were added dropwise to a mixture consist-
ing of 2.50 mmol pyrazole, 3.25 mmol (R,S)-2-phenylpro-
pionic acid or (R,S)-naproxen, 10 mmol triethylamine and
Adv. Synth. Catal. 2009, 351, 2333 – 2341
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2339

Pei-Yun Wang et al.
FULL PAPERS
tioselectivity for (R)-3 in water-saturated hexane or MTBE,
its concentration was negligible in comparison with that of
(S)-3 in determining the initial rate VS. Therefore, the cata-
lytic constant k2S and Michaelis constant KmS in the acylation
step could be at first estimated from the rate equation with
(S)-3 as the only substrate, and then KmR and k2R from the
rate equation for VR by using 3 as the substrate.[3d] As the
enzyme enantioselectivities for 3 in anhydrous MTBE as
well as 4 and 7 in water-saturated MTBE were low, more ex-
periments of using (S)-3, as well as (S)-4 and (S)-7 as the
substrates were carried out such that one can follow the
same procedures for estimating all kinetic constants.
Bull. 1999, 47, 591–600; d) P. Y. Wang, T. L. Chen,
S. W. Tsai, K. Wolfgang, Biotechnol. Bioeng. 2007, 98,
30–38.
[4] a) R. R. Bott, G. Chan, B. Domingo, G. Ganshaw, C. Y.
Hsia, M. Knapp, C. J. Murray, Biochemistry 2003, 42,
10545–10553; b) A. Case, R. Stein, Biochemistry 2003,
42, 3335–3348; c) T. Ishida, S. Kato, J. Am. Chem. Soc.
2003, 125, 12035–12048.
[5] a) L. D. Sutton, D. M. Quinn, J. Am. Chem. Soc. 1990,
112, 8404–8408; b) K. Nishizawa, Y. Ohgami, N.
Matsuo, H. Kisida, H. Hirohara, J. Chem. Soc. Perkin
Trans. 2 1997, 26, 1293–1298; c) T. Ema, S. Maeno, Y.
Takaya, T. Sakai, M. Utaka, J. Org. Chem. 1996, 61,
8610–8616; d) H. Hirohara, M. Nishizawa, Biosci. Bio-
technol. Biochem. 1998, 62, 1–9.
Extraction of Remaining Substrate (S)-3
To 10 mL water-saturated MTBE containing 100 mM of 3
was added 2 mgmLÀ1 Novozym 435 at 458C for 1.5 h, giving
the racemate conversion Xt =51.5% and enantiomeric
excess for the substrate ees =100% [i.e., 48.55 mM of (S)-3
remained] from HPLC analysis. The lipase was filtered and
the resultant solution was added to 10 mL pH 7 phosphate
buffers (300 mM) at 48C with stirring for 2 h. An HPLC
analysis for the resultant organic phase gave 47.50 mM of
(S)-3 without the traces of acid products, implying that
95.0% of (S)-3 on the basis of initial enantiomer concentra-
tion was recovered.
[6] a) G. DeSantis, J. B. Jones, J. Am. Chem. Soc. 1998,
120, 8582–8586; b) T. Penning, Chem. Rev. 2001, 101,
3027–3046; c) M. T. Reetz, Advances in Catalysis 2006,
49, 1–69; d) B. G. Davis, Curr. Opin. Biotechnol. 2003,
14, 379–386; e) P. Berglund, S. Park, Curr. Org. Chem.
2005, 9, 325–336; f) M. D. Toscano, K. J. Woycechow-
sky, D. Hilvert, Angew. Chem. 2007, 119, 3274–3300;
Angew. Chem. Int. Ed. 2007, 46, 3212–3236; g) P. Y.
Wang, T. L. Chen, S. W. Tsai, Biotechnol. Bioeng. 2008,
101, 460–469.
[7] a) F. Theil, Tetrahedron 2000, 56, 2905–2919; b) P. Ber-
glund, Biomol. Eng. 2001, 18, 13–22; c) U. T. Bornsche-
uer, C. Bessler, R. Srinivas, S. H. Krishna, Trends Bio-
technol. 2002, 20, 433–437; d) E. P. Hudson, R. K.
Eppler, D. S. Clark, Curr. Opin. Biotechnol. 2005, 16,
637–643.
[8] a) K. Faber, S. Riva, Synthesis 1992, 895–910; b) T.
Ema, S. Maeno, Y. Takaya, T. Sakai, M. Utaka, J. Org.
Chem. 1996, 61, 8610–8616; c) T. Miyazawa, M. Shi-
maoka, T. Yamada, Biotechnol. Lett. 1999, 21, 309–
312; d) H. Yang, E. Henke, U. T. Bornscheuer, J. Org.
Chem. 1999, 64, 1709–1712; e) U. Hanefeld, Org.
Biomol. Chem. 2003, 1, 2405–2415; f) R. Chenevert, N.
Pelchat, F. Jacques, Curr. Org. Chem. 2006, 10, 1067–
1094; g) T. Miyazawa, H. Minowa, T. Yamada, Biotech-
nol. Lett. 2006, 28, 295–299; h) L. Zheng, S. Zhang, L.
Zhao, G. Zhu, X. Yang, G. Cao, S. Cao, J. Mol. Cat. B:
Enzym. 2006, 38, 119–125.
Supporting Information
Detailed information on product characterization specifica-
tions (1H NMR spectra and retention time in HPLC analy-
sis) and the variation of initial rates with substrate concen-
trations is available in the Supporting Information.
Acknowledgements
Financial supports from National Science Council (Grant
No. NSC 97-2221-E-182-018-MY3) is gratefully acknowl-
edged.
References
[9] a) S. W. Tsai, C. C. Chen, H. S. Yang, I. S. Ng, T. L.
Chen, BBA Protein and Proteomics 2006, 1764, 1424–
1428; b) P. Y. Wang, S. W. Tsai, J. Mol. Cat. B: Enzym.
2009, 57, 158–163.
[1] a) R. J. Kazlauskas, U. T. Bornscheuer, Biotransforma-
tions I, in: (Eds.: H. J. Rehm, G. Reed, A. Puehler, P.
Stadler), Bioetchnologies series, Vol. 8a, Wiley-VCH,
Weinheim, 1998; b) K. Faber, Biotransformation in Or-
ganic Chemistry, 5th edn., Springer-Verlag, Berlin, 2004;
c) A. Ghanem, Tetrahedron 2007, 63, 1721–1754; d) A.
Kamal, M. A. Azhar, T. Krishnaji, M. S. Malik, S.
Azeeza, Coord. Chem. Rev. 2008, 252, 569–592.
[2] a) G. Dodson, A. Wlodawer, Trends Biochem. Sci.
1998, 23, 347–352; b) L. Hedstrom, Chem. Rev. 2002,
102, 4501–4523; c) U. T. Bornscheuer, R. J. Kazlauskas,
Hydrolases in Organic Synthesis: Regio- and Stereose-
lective Biotransformations, 2nd edn., Wiley-VCH, Wein-
heim, pp 37–275, 2006.
[10] V. E. Anderson, M. W. Ruszczycky, M. E. Harris,
Chem. Rev. 2006, 106, 3236–3251.
[11] a) J. P. Fox, W. P. Jencks, J. Am. Chem. Soc. 1974, 96,
1436–1449; b) R. L. Kogan, J. A. Fee, T. H. Eife, J.
Am. Chem. Soc. 1982, 104, 3569–3576; c) . H. Eife, R.
Natarajan, M. H. Werner, J. Org. Chem. 1987, 52, 740–
746; d) L. P. Lee, R. Bembi, T. H. Eife, J. Org. Chem.
1997, 62, 2827–2876; e) T. H. Eife, M. H. Werner,
Bioorg. Chem. 1998, 26, 119–130; f) H. A. Staab, H.
Bauer, K. M. Schneider, Azolides in Organic Synthesis
and Biochemistry, Wiley-VCH, Weinheim, 1998; g) J.
Elguero, C. Foces-Foces, D. Sanz, R. M. Claramunt,
Azolides: Structural Aspects, in: Advances in Nitrogen
Heterocycles, Vol. 4, pp 295–367, JAI Press, Stamford,
2000.
[3] a) M. Botta, F. Corelli, F. Manetti, A. Tafi, Farmaco
2002, 57, 153–165; b) R. Fujii, Y. Nakagawa, J. Hira-
take, A. Sogabe, K. Sakata, Protein Eng. Des. Sel. 2005,
18, 93–101; c) F. Heffner, T. Norin, Chem. Pharm.
2340
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2333 – 2341

ACHTUNGTRENNUNG(R,S)-Azolides as Novel Substrates for Lipase-Catalyzed Hydrolytic Resolution
FULL PAPERS
[12] a) E. M. Anderson, K. M. Larsson, K. M. Kirk, Bioca-
tal. Biotransform. 1998, 16, 181–204; b) V. Gotor-Fer-
nandez, E. Busto, V. Gotor, Adv. Synth. Catal. 2006,
348, 797–812; c) N. Li, M. H. Zong, D. Ma, Eur. J. Org.
Chem. 2008, 32, 5375–5378;.
Pleiss, M. Fischer, R. D. Schmid, Chem. Phys. Lipids,
1998, 93, 67–80; c) R. T. Otto, H. Scheib, U. T. Born-
scheuer, J. Pleiss, C. Syldatk, R. D. Schmid, J. Mol. Cat.
B: Enzym. 2000, 8, 201–211.
[15] a) J. Ottosson, L. Fransson, J. W. King, K. Hult, Bio-
chim. Biophys Acta 2002, 1594, 325–334; b) L. Yang,
J. S. Dordick, S. Garde, Biophys. J. 2004, 87, 812–821;
c) G. Dodson, C. S. Verma, Cell. Mol. Life Sci. 2006,
63, 207–219; d) P. Trodler, J. Pleiss, BMC Struct. Biol.
2008, 8–9, 1–10.
[16] a) E. Wehtje, P. Adlercreutz, Biotechnol. Bioeng. 1997,
55, 798–806; b) J. M. Mora-Pale, S. Perez-Munguia,
J. C. Gonzalez-Mejia, J. S. Dordick, E. Barzana, Bio-
technol. Bioeng. 2007, 98, 535–542.
[17] J. A. Alkandary, A. S. Aljimaz, M. S. Fandary, M. A.
Fahim, Fluid Phase Equilib. 2001, 187–188, 131–138.
[18] a) A. Maczynski, M. Goral, B. Wisniewska-Goclowaka,
A. Skrzecz, D. Shaw, Monatsh. Chem. 2003, 134, 633–
653; b) H. Y. Lin, E. Lay, W. Y. Wen, H. Dewi, Y. C.
Cheng, S. W. Tsai, J. Phys Org. Chem. 2004, 17, 387–
392.
[13] a) M. Arroyo, J. V. Sinisterra, J. Org. Chem. 1994, 59,
4410–4417; b) R. Morrone, G. Nicolosi, A. Patti, M.
Piattelli, Tetrahedron: Asymmetry 1995, 6, 1773–1778;
c) P. Mertoli, G. Nicolosi, A. Patti, M. Piattelli, Chirali-
ty 1996, 8, 377–380; d) A. Ducret, T. Michael, L.
Robert, Enzyme Microb. Technol. 1998, 22, 212–216;
e) H. Yang, E. Henke, U. T. Bornscheuer, J. Org.
Chem. 1999, 64, 1709–1712; f) A. L. Ong, A. H. Ka-
maruddin, S. Bhatia, W. S. Long, S. T. Lim, R. Kumari,
Enzyme Microb. Technol. 2006, 39, 924–929; g) P. Y.
Wang, T. L. Chen, S. W. Tsai, A. Cipiciani, F. Bellezza,
R. Ruzziconi, J. Mol. Cat. B: Enzymatic 2006, 42, 90–
94; h) Y. Z. Hua, X, Wang, B. C. Chi, Chirality 2007,
19, 245–249;.
[14] a) J. Uppenberg, N. ꢂhrner, M. Norin, K. Hult, G. J.
Kleywegt, S. Patkar, V. Waagen, T. Anthonsen, T.
Alwyn Jones, Biochemistry 1995, 34, 16838–16851; b) J.
Adv. Synth. Catal. 2009, 351, 2333 – 2341
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2341